United States
     Environmental Protection                            September 2011
     A9encY                                   EPA/600/R-10/076B
Integrated Science Assessment for Ozone
   and Related Photochemical Oxidants
     National Center for Environmental Assessment-RTP Division
             Office of Research and Development
            U.S. Environmental Protection Agency
                Research Triangle Park, NC

-------
DISCLAIMER
              This document is the second external review draft for review purposes only and does not
              constitute U.S. Environmental Protection Agency policy. Mention of trade names or
              commercial products does not constitute endorsement or recommendation for use.

-------
TABLE   OF  CONTENTS


      AUTHORS, CONTRIBUTORS, AND REVIEWERS	XXVII

      CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE OZONE NAAQS REVIEW PANEL	XXXI

      ACRONYMS AND ABBREVIATIONS	XXXIII

      PREAMBLE	XLIX
         Process of ISA Development	xlix
         EPA Framework for Causal Determination	 Nil
         Evaluating Evidence for Inferring Causation	liv
              Consideration of evidence from scientific disciplines	Iv
              Application of Framework for Causal Determination	 lix
                           Table I      Aspects to aid in judging causality	Ix
              Determination of Causality	 Ixi
                           Table II     Weight of evidence for causal determination	Ixii
                    Quantitative relationships: Effects on Human Populations	Ixiii
                    Quantitative relationships: Effects on Ecosystems or Public Welfare	Ixv
              Concepts in Evaluating Adversity of Health Effects	Ixv
              Concepts in Evaluating Adversity of Ecological Effects	Ixvi

      PREFACE	LXVIII
              Legislative Requirements for the NAAQS Review	Ixviii
              History of the NAAQS for Ozone	Ixix
                           Table III     Summary of primary and secondary NAAQS promulgated for ozone
                                      during the period 1971-2008	Ixx
              References	Ixxiii

      1   EXECUTIVE SUMMARY                                                                  1-1
1.1
1.2
1.3
1.4
1.5
1.6
Introduction
Scope
Atmospheric Chemistry and Ambient Concentrations
Figure 1-1 Schematic overview of photochemical processes influencing stratospheric
and tropospheric ozone.
Human Exposure
Dosimetry and Modes of Action
Table 1-1 Summary of ozone causal determinations by exposure duration and
health outcome
Integration of Ozone Health Effects
1.6.1 Respiratory Effects
1-1
1-1
1-2
1-3
1-4
1-4
1-5
1-6
1-6
                                      decreases in the proportion of the population affected moving up the
         1.7
pyramid.
1.6.2 Mortality Effects
1.6.3 Emerging Evidence
1.6.4 Populations at Increased Risk
1.6.5 Ozone Concentration-Response Relationship
Integration of Effects on Vegetation and Ecosystems
Table 1-2 Summary of ozone causal determination for welfare effects
1.7.1 Visible Foliar Iniury
1.7.2 Growth, Productivity, Carbon Storage and Agriculture
1.7.3 Water Cycling
1-7
1-8
1-8
1-9
1-10
1-10
1-11
1-11
1-12
1-12
     - Do Not Cite or                                iii                                               2011

-------
1.8
1.9
1.7.4 Below Ground Processes
1.7.5 Community Composition
1.7.6 Ozone Exposure-Response Relationships
The Role of Tropospheric Ozone in Climate Change and UV-B Effects
Figure 1-3 Schematic illustrating the effects of tropospheric O3 on climate.
Table 1-3 Summary of ozone causal determination for climate change and UV-B
effects
Conclusion
2 INTEGRATIVE SUMMARY
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
Policy-Relevant Questions for O3 NAAQS Review
ISA Development and Scope
Atmospheric Chemistry and Ambient Concentrations
2.3.1 Physical and Chemical Processes
Figure 2-1 Schematic overview of photochemical processes influencing stratospheric
and tropospheric ozone.
2.3.2 Atmospheric Modeling of Background Ozone Concentrations
2.3.3 Monitoring
Figure 2-2 GEOS-Chem modeled U.S. policy relevant background seasonal-mean
surface ozone concentrations in spring (left) and summer (right), 2006.
2.3.4 Ambient Concentrations
Human Exposure
Dosimetry and Mode of Action
Integration of Ozone Health Effects
Table 2-1 Summary of evidence from epidemiologic, controlled human exposure,
and animal toxicological studies on the health effects associated with
short- and long-term exposure to ozone
2.6.1 Respiratory Effects
Figure 2-3 Snapshot of evidence for the association of O3 with the continuum of
respiratory effects, including sub-clinical effects (bottom level of the
pyramid) and clinical effects, increasing in severity moving up the
pyramid.
2.6.2 Mortality Effects
2.6.3 Cardiovascular Health Effects
2.6.4 Central Nervous System Effects
2.6.5 Reproductive and Developmental Effects
2.6.6 Cancer and Mutagenicity and Genotoxicity
2.6.7 Policy Relevant Considerations
2.6.7.1 Populations at Increased Risk
2.6.7.2 Lag Structure in Epidemiologic Studies
2.6.7.3 Ozone Concentration-Response Relationship
Integration of Effects on Vegetation and Ecosystems
2.7.1 Visible Foliar Iniury
Figure 2-4 An illustrative diagram of the major pathway through which O3 enters
leaves and the major endpoints that O3 may affect in plants and
ecosystems.
Table 2-2 Summary of ozone causal determinations for vegetation and ecosystem
effects
2.7.2 Growth, Productivity, Carbon Storage and Agriculture
2.7.2.1 Natural Ecosystems
2.7.2.2 Agricultural Crops
2.7.3 Water Cycling
2.7.4 Below-Ground Processes
2.7.5 Community Composition
2.7.6 Policy Relevant Considerations
2.7.6.1 Air Quality Indices
2.7.6.2 Exposure-Response
The Role of Tropospheric Ozone in Climate Change and UV-B Effects
1-13
1-13
1-13
1-14
1-15
1-16
1-16
2-1
2-1
2-4
2-6
2-6
2-7
2-8
2-10
2-10
2-11
2-12
2-14
2-17
2-18
2-20
2-23
2-29
2-30
2-30
2-31
2-31
2-31
2-31
2-33
2-34
2-36
2-37
2-37
2-38
2-39
2-39
2-41
2-42
2-42
2-43
2-44
2-44
2-45
2-46
- Do Not Cite or                                iv                                              2011

-------
          2.8.1   Tropospheric Ozone as a Greenhouse Gas	2-47
                        Figure 2-5   Schematic illustrating the effects of tropospheric O3 on climate including
                                    the relationship between precursor emissions, tropospheric O3
                                    abundance, radiative forcing, climate response, and climate impacts.	2-48
          2.8.2   Tropospheric Ozone and UV-B related effects	2-48
     2.9   Summary of Causal Determinations for Health Effects and Welfare Effects	2-50
                        Table 2-3   Summary of ozone causal determinations by exposure duration and
                                    health outcome	2-50
                        Table 2-4   Summary of ozone causal determination for welfare effects	2-51
                        Table 2-5   Summary of ozone causal determination for climate change and UV-B
                                    effects	2-51
     2.10  References	 2-52

 3   ATMOSPHERIC CHEMISTRY AND AMBIENT CONCENTRATIONS	3-1
     3.1   Introduction	 3-1
     3.2   Physical and Chemical Processes	 3-1
                        Figure 3-1   Schematic overview of photochemical processes influencing stratospheric
                                    and tropospheric ozone.	3-2
          3.2.1   Sources of Precursors Involved in Ozone Formation	3-5
                        Figure 3-2   Estimated  anthropogenic emissions of ozone precursors for 2005. 	3-6
          3.2.2   Gas Phase Reactions Leading to Ozone Formation	3-9
          3.2.3   Multiphase Processes	3-13
                 3.2.3.1  Indoor Air   3-15
          3.2.4   Temperature and Chemical Precursor Relationships	3-15
                        Figure 3-3   Measured  concentrations of O3 and NOZ	3-19
     3.3   Atmospheric Modeling	 3-20
                        Figure 3-4   Sample CMAQ modeling domains.	3-21
                        Figure 3-5   Main components of a comprehensive atmospheric chemistry modeling
                                    system, such as the U.S. EPA's Community Model for Air Quality
                                    (CMAQ) System.	3-22
          3.3.1   Global Scale CTMs	3-26
                        Figure 3-6   Comparison of global CTM predictions of maximum daily 8-h avg ozone
                                    concentrations and multi-model mean with monthly averaged CASTNET
                                    observations in the Intermountain West and Southeast regions of the U.S.	3-27
     3.4   Background Ozone Concentrations	3-28
                        Figure 3-7   Schematic overview of contributions to North American background
                                    concentrations of ozone, i.e., ozone concentrations that would exist in the
                                    absence of anthropogenic emissions from the U.S., Canada, and Mexico. 	3-30
          3.4.1   Contributions from Anthropogenic Emissions outside North America	3-30
                        Figure 3-8   Time series of daily maximum 8-h avg ozone concentrations (ppm)
                                    measured  at Trinidad Head, CA, from April 18, 2002 through December
                                    31, 2009. 	3-33
          3.4.2   Contributions from Natural Sources	3-33
                 3.4.2.1  Contributions from the Stratosphere	3-33
                 3.4.2.2  Contributions from Other Natural Sources	3-35
          3.4.3   Estimating Background Concentrations	3-37
                        Figure 3-9   North American background ozone concentration in surface air for spring
                                    and summer 2006 (top). GEOS-Chem calculated concentrations for the
                                    base case, i.e., including all sources in surface air for the U.S., Canada
                                    and Mexico for spring and summer of 2006 (bottom).	3-39
                        Figure 3-10  North American background ozone concentrations calculated when base
                                    case ozone is > 60 ppb.	3-41
                        Figure 3-11 a Simulated  vs.  observed daily 8-h max ozone concentrations for spring
                                    (March-May) and summer (June-August) 2006 for the ensemble of
                                    CASTNET sites in the intermountain West.	3-42
                        Figure 3-11b Simulated  vs.  observed daily 8-h max ozone concentrations for spring
                                    (March-May) and summer (June-August) 2006 for the ensembles of
                                    CASTNET sites in the Northeast U.S., Great Lakes, and the Southeast
                                    U.S.                                                                 3-43
- Do Not Cite or                                   v                                                    2011

-------
                         Figure 3-12  Frequency distributions of daily 8-h max ozone concentrations in March-
                                      August 2006 for the ensemble of low-altitude (<1.5 km) and high-altitude
                                      CASTNET sites in the U.S. 	3-44
          3.4.4  Summary of Background Results	3-45
     3.5   Monitoring	 3-46
          3.5.1  Routine Monitoring Techniques	3-46
          3.5.2  Precision and Bias	3-49
                         Table 3-1     Summary of ozone monitors meeting 40 CFR Part 58, Appendix A
                                      Precision and Bias Goals	3-50
                         Figure 3-13  Box plots of precision data by year (2005-2009) for all ozone monitors
                                      reporting single-point QC check data to AQS.	3-50
                         Figure 3-14  Box plots of percent-difference data  by year (2005-2009) for all ozone
                                      monitors reporting single-point QC check data to AQS. 	3-51
                 3.5.2.1  Precision from Co-located UV Ozone Monitors in Missouri	3-51
                         Figure 3-15  Box plots of RPD data by year for the co-located ozone monitors at two
                                      sites in Missouri from 2006-2009.	3-52
     3.6
3.5.3
3.5.4
3.5.5
3.5.6
Figure 3-1 6 Box plots of RPD data by year for all U.S. ozone sites reporting single-
point QC check data to AQS from 2005-2009.
Performance Specifications
Table 3-2 Performance specifications for ozone based in 40 CFR Part 53
Monitor Calibration
Other Monitorinq Techniques
3.5.5.1 Portable UV Ozone Monitors
3.5.5.2 NO-based Chemiluminescence Monitors
3.5.5.3 Passive Air Samplinq Devices and Sensors
3.5.5.4 Differential Optical Absorption Spectrometry
3.5.5.5 Satellite Remote Sensinq
Ambient Ozone Network Desiqn
3.5.6.1 Monitor Sitinq Requirements
Fiqure 3-17 U.S. ozone sites reportinq data to AQS in 2010.
Fiqure 3-18 U.S. Rural NCore, CASTNET and NPS POMS ozone sites in 2010.
3.5.6.2 Probe/Inlet Sitinq Requirements
Ambient Concentrations
3.6.1
Measurement Units, Metrics, and Averaqinq Times
3-52
3-53
3-53
3-53
3-55
3-55
3-55
3-56
3-57
3-58
3-59
3-59
3-62
3-63
3-63
3-64
3-65
                                      ozone metrics including 24-h avg, 1-h daily max and 8-h daily max using

          3.6.2  Spatial
                 3.6.2.1
'ariability
Urban-Focus
Figure 3-20
Table 3-3
Figure 3-21
AQS data, 2007-2009.

;ed Variability
Required ozone monitoring time periods (ozone season) identified by
monitorinq site.
Summary of ozone data sets oriqinatinq from AQS
Location of the 457 ozone monitors meeting the year-round data set
completeness criterion for all 3 years between 2007 and 2009.
3-66
3-67
3-67
3-68
3-69
3-70
                         Figure 3-22  Location of the 1,064 ozone monitors meeting the warm-season data set
                                      completeness criteria for all 3 years between 2007 and 2009.	3-70
                         Table 3-4    Nationwide distributions of ozone concentrations (ppb) from the year-
                                      round data set	3-71
                         Table 3-5    Nationwide distributions of ozone concentrations (ppb) from the warm-
                                      season data set	3-72
                         Table 3-6    Seasonally stratified distributions of 8-h daily max ozone concentrations
                                      (ppb) from the year-round data set (2007-2009)	3-73
                         Figure 3-23  Highest monitor (by county) 3-year avg (2007-2009) of the 8-h daily max
                                      ozone concentration based on the year-round data set (top map) with
                                      seasonal stratification (bottom 4 maps). 	3-74
                         Figure 3-24  Highest monitor (by county) 3-year avg (2007-2009) of the 8-h daily max
                                      ozone concentration based on the warm-season data set (top map) with
                                      annual stratification (bottom 3 maps).	3-75
                         Table 3-7    Focus cities used in this and previous assessments	3-78
                         Table 3-8    City-specific distributions of 8-h daily max ozone concentrations (ppb)
                                      from the warm-season data set (2007-2009)	3-79
- Do Not Cite or                                     vi                                                      2011

-------
                         Figure 3-25  Map of the Atlanta CSA including ozone monitor locations, population
                                      gravity centers, urban areas, and major roadways.	3-80
                         Figure 3-26  Map of the Boston CSA including ozone monitor locations, population
                                      gravity centers, urban areas, and major roadways.	3-80
                         Figure 3-27  Map of the Los Angeles CSA including ozone monitor locations,
                                      population gravity centers, urban areas, and major roadways.	3-81
                         Figure 3-28  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Atlanta CSA.	3-82
                         Figure 3-29  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Boston CSA.	3-82
                         Figure 3-30  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Los Angeles CSA.	3-84
                         Figure 3-31   Pair-wise monitor correlations expressed as a histogram (top), contour
                                      matrix (middle) and scatter plot versus distance between monitors
                                      (bottom) for the Atlanta CSA.	3-86
                         Figure 3-32  Pair-wise monitor correlations expressed as a histogram (top), contour
                                      matrix (middle) and scatter plot versus distance between monitors
                                      (bottom) for the Boston CSA.	3-87
                         Figure 3-33  Pair-wise monitor correlations expressed as a histogram (top), contour
                                      matrix (middle) and scatter plot versus distance between monitors
                                      (bottom) for the Los Angeles CSA. 	3-88
                         Figure 3-34  Pair-wise monitor COD expressed as a histogram (top), contour matrix
                                      (middle) and scatter plot versus distance between monitors (bottom) for
                                      the Atlanta CSA.	3-89
                         Figure 3-35  Pair-wise monitor COD expressed as a histogram (top), contour matrix
                                      (middle) and scatter plot versus distance between monitors (bottom) for
                                      the Boston CSA.	3-90
                         Figure 3-36  Pair-wise monitor COD expressed as a histogram (top), contour matrix
                                      (middle) and scatter plot versus distance between monitors (bottom) for
                                      the Los Angeles CSA.	3-91
                         Figure 3-37  Terrain map showing the location of two nearby AQS ozone monitoring
                                      sites (red dots) along the western edge of the Los Angeles CSA. Site AL
                                      is near shore, 3 m  above sea level, while Site AK is in an agricultural
                                      valley surrounded by mountains, 262 m above sea level.	3-93
                         Figure 3-38  Terrain map showing the location of four AQS ozone  monitoring sites (red
                                      dots) located in or near the city limits in the center of the Boston CSA.
                                      Site characteristics range from Site A near downtown at 6 m above sea
                                      level to Site D in a forested area on Blue Hill at 192 m above sea  level.	3-94
                 3.6.2.2  Rural-Focused Variability and Ground-Level Vertical Gradients	3-96
                         Table 3-9    Rural focus areas	3-97
                         Figure 3-39  Rural focus area site information, statistics and box plots for 8-h daily
                                      max ozone from AQS monitors meeting the warm-season data set
                                      inclusion criteria within the rural focus areas. 	3-98
                         Figure 3-40  Terrain map showing the location of five AQS ozone monitoring sites
                                      (green/black stars) in Great Smoky Mountain National Park, NC-TN
                                      (SMNP).	3-99
                         Figure 3-41   Pair-wise monitor correlations (left) and coefficients of divergence (COD,
                                      right) expressed as a histogram (top), contour matrix  (middle) and scatter
                                      plot vs. distance between monitors (bottom) for Great Smoky Mountain
                                      National Park, NC-TN (SMNP).	3-100
                         Figure 3-42  Terrain map showing the location of the AQS ozone monitoring site in
                                      Rocky Mountain National Park, CO (black/green star) and the  Denver
                                      CSA (red dots) along with ozone monitoring sites used in the Brodin et al.
                                      (2010) study (blue circles).	3-101
                         Figure 3-43  Terrain map showing the location of two AQS ozone monitoring sites
                                      (black/green stars) in Sequoia National Park, CA.	3-102
          3.6.3  Temporal Variability	3-103
                 3.6.3.1  Multiyear Trends	3-103
                         Figure 3-44  National 8-h ozone trends, 2001-2008 (average of the annual fourth
                                      highest 8-h daily max concentrations in ppm).	3-104
                 3.6.3.2  Hourly Variations	3-105
- Do Not Cite or                                     vii                                                     2011

-------
                        Figure 3-45  Diel patterns in 1-h avg ozone for Atlanta, Boston and Los Angeles
                                    between 2007 and 2009 using the year-round data set for the cold
                                    month/warm month comparison (left half) and the warm-season data set
                                    for the weekday/weekend comparison (right half). 	3-106
          3.6.4  Associations with Co-pollutants	3-109
                        Figure 3-47  Distribution of Pearson correlation coefficients for comparison of 8-h daily
                                    max ozone from the year-round data set with co-located 24-h avg CO,
                                    SO2, NO2, PM10 and PM2.5from AQS, 2007-2009	3-110
                        Figure 3-48  Distribution of Pearson correlation coefficients for comparison of 8-h daily
                                    max ozone from the warm-season (May-Sept) data set with co-located
                                    24-h avg CO, SO2, NO2, PM10 and PM2.5 from AQS, 2007-2009.	3-111
     3.7   Chapter Summary 	3-112
          3.7.1  Physical and Chemical Processes	3-112
          3.7.2  Atmospheric Modeling	3-113
          3.7.3  Background Concentrations	3-115
          3.7.4  Monitoring	3-116
          3.7.5  Ambient Concentrations	3-117
     3.8   Supplemental Ozone Model Predictions from the Literature	 3-120
          3.8.1  Time Series of GEOS-Chem Model Predictions and Observations at Selected CASTNET Sites	3-120
                        Figure 3-49  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at four CASTNET sites in the Northeast
                                    with GEOS-Chem predictions for the base case and for the North
                                    American background case in 2006.	3-121
                        Figure 3-50  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at four CASTNET sites in the Southeast
                                    with GEOS-Chem predictions for the base case and for the North
                                    American background case in 2006.	3-121
                        Figure 3-51  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at four CASTNET sites in the Upper
                                    Midwest with GEOS-Chem predictions for the base case and for the
                                    North American background case in 2006.	3-122
                        Figure 3-52  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at CASTNET sites in the Intermountain
                                    West with GEOS-Chem predictions for the base case and the North
                                    American background case in 2006.	3-122
                        Figure 3-53  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at CASTNET sites in the Intermountain
                                    West with GEOS-Chem predictions for the base case and the North
                                    American background case in 2006.	3-123
                        Figure 3-54  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at CASTNET sites in the West with GEOS-
                                    Chem predictions for the base case and the North American background
                                    case in 2006.	3-123
                        Figure 3-55  Comparison of time series of measurements of daily maximum 8-hour
                                    average ozone concentrations at monitoring sites in Californiawith GEOS-
                                    Chem predictions for the base case and the North American background
                                    case in 2006.	3-124
                        Figure 3-56  Comparison of daily maximum 8-h average ozone predicted using GEOS-
                                    Chem at 0.5°x0.67° and 2°x2.5° (left figure only) resolution with
                                    measurements at Mount Bachelor, OR (left) and Trinidad Head, CA
                                    (right) from March to August 2006. 	3-124
                        Figure 3-57  Comparison of monthly mean ± 1 standard deviation ozone calculated
                                    GEOS-Chem (in red) with ozonesondes (in black) at Trinidad Head and
                                    Boulder, CO during April and August 2006.	3-125
     3.9   Supplemental Ozone Model Predictions Using the Latest Release of GEOS-Chem	3-125
          3.9.1  Introduction	3-125
          3.9.2  GEOS-Chem  Model Application	3-126
          3.9.3  Model Scenarios	3-127
                        Table 3-10  Summary of GEOS-Chem model scenarios	3-128
          3.9.4  Model Performance Evaluation	3-128
                        Figure 3-58  Frequency distributions of 8-hr daily max ozone concentration from March
                                    -August 2006for low-elevation (<1.5 km; top panel) and high-elevation
                                    (>1.5 km; bottom panel) CASTNET sites.	3-130
- Do Not Cite or                                  viii                                        September 2011

-------
          3.9.5  Model Results	3-131
                         Figure 3-59  Mean 8-hr daily max O3 concentration (ppb) for the Base Case (left panel)
                                     and North American Background scenario (right panel), based on the
                                     2006, 2007 and 2008 simulation period.	3-132
                         Figure 3-60  Mean 8-hr daily max O3 concentration (ppb) for the North American
                                     Background scenario during winter (Dec-Feb, upper right panel), spring
                                     (Mar-May, upper right panel), summer (Jun-Aug, lower left panel), and fall
                                     (Sep-Nov, lower right panel), based on the 2006, 2007 and 2008
                                     simulation period.	3-133
          3.9.6  Model Attributes and Limitations	3-134
          3.9.7  Summary of Modeling Results	3-135
     3.10  Supplemental Figures of Observed Ambient Ozone Concentrations	3-136
          3.10.1  Ozone Monitor Maps for the Urban Focus Cities	3-136
                         Figure 3-61  Map of the Atlanta CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-137
                         Figure 3-62  Map of the Baltimore CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-137
                         Figure 3-63  Map of the Birmingham CSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-138
                         Figure 3-64  Map of the Boston CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-138
                         Figure 3-65  Map of the Chicago CSA including  ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-139
                         Figure 3-66  Map of the Dallas CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-139
                         Figure 3-67  Map of the Denver CSA including ozone  monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-140
                         Figure 3-68  Map of the Detroit CSA including ozone monitor locations,  population
                                     gravity centers, urban areas, and major roadways.	3-140
                         Figure 3-69  Map of the Houston CSA including  ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-141
                         Figure 3-70  Map of the Los Angeles CSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-141
                         Figure 3-71  Map of the Minneapolis CSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-142
                         Figure 3-72  Map of the New York CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-142
                         Figure 3-73  Map of the Philadelphia CSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-143
                         Figure 3-74  Map of the Phoenix CBSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-143
                         Figure 3-75  Map of the Pittsburgh CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-144
                         Figure 3-76  Map of the Salt Lake City CSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-145
                         Figure 3-77  Map of the San Antonio CBSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-145
                         Figure 3-78  Map of the San Francisco CSA including ozone monitor locations,
                                     population gravity centers, urban areas, and major roadways.	3-146
                         Figure 3-79  Map of the Seattle CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-146
                         Figure 3-80  Map of the St. Louis CSA including ozone monitor locations, population
                                     gravity centers, urban areas, and major roadways.	3-147
          3.10.2 Ozone Concentration Box Plots for the Urban Focus Cities	3-147
                         Figure 3-81  Site information, statistics and box  plots for 8-h daily max ozone from
                                     AQS monitors meeting the warm-season data set inclusion criteria within
                                     the Atlanta CSA.	3-148
                         Figure 3-82  Site information, statistics and box  plots for 8-h daily max ozone from
                                     AQS monitors meeting the warm-season data set inclusion criteria within
                                     the Baltimore CSA.	3-148
                         Figure 3-83  Site information, statistics and box  plots for 8-h daily max ozone from
                                     AQS monitors meeting the warm-season data set inclusion criteria within
                                     the Birmingham CSA.	3-149
- Do Not Cite or                                     ix                                                     2011

-------
                          Figure 3-84  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Boston CSA.	3-149
                          Figure 3-85  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Chicago CSA.	3-150
                          Figure 3-86  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Dallas CSA.	3-150
                          Figure 3-87  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Denver CSA. 	3-151
                          Figure 3-88  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Detroit CSA.	3-151
                          Figure 3-89  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Houston CSA. 	3-152
                          Figure 3-90  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Los Angeles CSA.	3-153
                          Figure 3-91  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Minneapolis CSA.	3-154
                          Figure 3-92  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the New York CSA. 	3-154
                          Figure 3-93  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Philadelphia CSA. 	3-155
                          Figure 3-94  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Phoenix CBSA. 	3-155
                          Figure 3-95  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Pittsburgh CSA.	3-156
                          Figure 3-96  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Salt Lake  City CSA.	3-156
                          Figure 3-97  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the San Antonio CBSA.	3-157
                          Figure 3-98  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the San Francisco CSA.	3-157
                          Figure 3-99  Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the Seattle CSA.	3-158
                          Figure 3-100 Site information, statistics and box plots for 8-h daily max ozone from
                                      AQS monitors meeting the warm-season data set inclusion criteria within
                                      the St. Louis CSA.	3-158
           3.10.3 Ozone Concentration Relationships for the Urban Focus Cities	3-159
                          Figure 3-101 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Atlanta CSA.	3-159
                          Figure 3-102 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Baltimore CSA.	3-160
                          Figure 3-103 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Birmingham CSA.	3-161
                          Figure 3-104 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Boston CSA.	3-162
- Do Not Cite or                                     x                                                      2011

-------
                          Figure 3-105 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Chicago CSA.	3-163
                          Figure 3-106 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Dallas CSA.	3-164
                          Figure 3-107 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Denver CSA.	3-165
                          Figure 3-108 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Detroit CSA.	3-166
                          Figure 3-109 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Houston CSA.	3-167
                          Figure 3-110 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Los Angeles CSA.	3-168
                          Figure 3-111 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Minneapolis CSA.	3-169
                          Figure 3-112 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the New York CSA.	3-170
                          Figure 3-113 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Philadelphia CSA.	3-171
                          Figure 3-114 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Phoenix CBSA.	3-172
                          Figure 3-115 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Pittsburgh CSA.	3-173
                          Figure 3-116 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Salt Lake City CSA.	3-174
                          Figure 3-117 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the San Antonio CBSA.	3-175
                          Figure 3-118 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the San Francisco CSA. of R.	3-176
                          Figure 3-119 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the Seattle CSA.	3-177
                          Figure 3-120 Pair-wise monitor correlation coefficients (R) expressed as a histogram
                                      (top), contour matrix (middle) and scatter plot versus distance between
                                      monitors (bottom) for the St. Louis CSA.	3-178
                          Figure 3-121 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                      histogram  (top), contour matrix (middle) and  scatter plot versus distance
                                      between monitors (bottom) for the Atlanta CSA.	3-179
                          Figure 3-122 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                      histogram  (top), contour matrix (middle) and  scatter plot versus distance
                                      between monitors (bottom) for the Baltimore  CSA.	3-180
                          Figure 3-123 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                      histogram  (top), contour matrix (middle) and  scatter plot versus distance
                                      between monitors (bottom) for the Birmingham CSA.	3-181
                          Figure 3-124 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                      histogram  (top), contour matrix (middle) and  scatter plot versus distance
                                      between monitors (bottom) for the Boston CSA.	3-182
                          Figure 3-125 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                      histogram  (top), contour matrix (middle) and  scatter plot versus distance
                                      between monitors (bottom) for the Chicago CSA.	3-183
- Do Not Cite or                                    xi                                                      2011

-------
                         Figure 3-126 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Dallas CSA.	3-184
                         Figure 3-127 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Denver CSA.	3-185
                         Figure 3-128 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Detroit CSA.	3-186
                         Figure 3-129 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Houston CSA.	3-187
                         Figure 3-130 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Los Angeles CSA.	3-188
                         Figure 3-131 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Minneapolis CSA.	3-189
                         Figure 3-132 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the New York CSA.	3-190
                         Figure 3-133 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Philadelphia CSA.	3-191
                         Figure 3-134 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Phoenix CBSA.	3-192
                         Figure 3-135 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Pittsburgh CSA.	3-193
                         Figure 3-136 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Salt Lake City CSA.	3-194
                         Figure 3-137 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the San Antonio CBSA.	3-195
                         Figure 3-138 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the San Francisco  CSA.	3-196
                         Figure 3-139 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the Seattle CSA.	3-197
                         Figure 3-140 Pair-wise monitor coefficient of divergence (COD) expressed as a
                                     histogram (top), contour matrix (middle) and scatter plot versus distance
                                     between monitors (bottom) for the St. Louis  CSA.	3-198
          3.10.4 Hourly Variations in Ozone for the Urban Focus Cities	3-199
                         Figure 3-141 Diel patterns in 1-h avg ozone for select CSAs between 2007 and 2009
                                     using the year-round data set for the cold month/warm month comparison
                                     (left half) and the warm-season data set for the weekday/weekend
                                     comparison (right half). 	3-199
                         Figure 3-142 Diel patterns in 1-h avg ozone for select CSAs between 2007 and 2009
                                     using the year-round data set for the cold month/warm month comparison
                                     (left half) and the warm-season data set for the weekday/weekend
                                     comparison (right half). 	3-200
                         Figure 3-143 Diel patterns in 1-h avg ozone for select CSAs between 2007 and 2009
                                     using the year-round data set for the cold month/warm month comparison
                                     (left half) and the warm-season data set for the weekday/weekend
                                     comparison (right half). 	3-201
                         Figure 3-144 Diel patterns in 1-h avg ozone for select CSAs/CBSAs between 2007 and
                                     2009  using the year-round data set for the cold month/warm month
                                     comparison (left half) and the warm-season  data set for the
                                     weekday/weekend comparison (right half).	3-202
                         Figure 3-145 Diel patterns in 1-h avg ozone for select CSAs/CBSAs between 2007 and
                                     2009  using the year-round data set for the cold month/warm month
                                     comparison (left half) and the warm-season  data set for the
                                     weekday/weekend comparison (right half).	3-203
- Do Not Cite or                                    xii                                                     2011

-------
     3.11  References	3-204

 4   EXPOSURE TO AMBIENT OZONE 	4-1
     4.1   Introduction	 4-1
     4.2   General Exposure Concepts	 4-1
     4.3   Exposure Measurement	 4-4
          4.3.1   Personal Monitoring Techniques	4-4
          4.3.2   Indoor-Outdoor Concentration Relationships	4-5
                        Table 4-1    Relationships of Indoor and Outdoor Ozone Concentration	4-6
          4.3.3   Personal-Ambient Concentration Relationships	4-8
                        Figure 4-1    Variation in hourly personal and ambient concentrations of O3 and PM25
                                     in various microenvironments during daytime hours.	4-9
                        Table 4-2    Correlations between Personal and Ambient Ozone Concentration	4-11
                        Table 4-3    Ratios of Personal to Ambient Ozone Concentration 	4-13
          4.3.4   Co-Exposure to Other Pollutants and Environmental Stressors	4-14
                 4.3.4.1  Personal Exposure to Ozone and Co-pollutants	4-14
                 4.3.4.2 Near-Road Exposure to Ozone and Co-pollutants	4-15
                        Figure 4-2    Correlations between 1 -week concentrations of O3 and copollutants
                                     measured near roadways.	4-16
                 4.3.4.3 Indoor Exposure to Ozone and Co-pollutants	4-16
     4.4   Exposure-Related Metrics	  4-17
          4.4.1   Activity Patterns	4-17
                        Figure 4-3    Distribution of time that NHAPS respondents spent in ten
                                     microenvironments based on smoothed 1-min diary data.  	4-18
          4.4.2   Ozone Averting Behavior	4-19
          4.4.3   Population Proximity to Fixed-Site Ozone Monitors	4-21
                        Table 4-4    Fraction of the 2009 population living within a specified distance of an
                                     ozone monitor in selected U.S. cities	4-23
     4.5   Exposure Modeling	  4-24
                        Table 4-5    Characteristics of exposure modeling approaches	4-24
          4.5.1   Concentration Surface Modeling	4-25
          4.5.2   Residential Air Exchange Rate Modeling	4-27
          4.5.3   Microenvironment-Based Models	4-29
     4.6   Implications for Epidemiologic Studies	4-32
          4.6.1   Nonambient Ozone Exposure 	4-33
          4.6.2   Spatiotemporal Variability	4-33
                 4.6.2.1  Spatial Variability	4-35
                 4.6.2.2 Seasonal ity  4-37
          4.6.3   Exposure to Co-pollutants and Ozone Reaction Products	4-38
          4.6.4   Averting Behavior	4-38
                        Figure 4-4    Adjusted asthma hospital admissions by age on lagged ozone by alert
                                     status,  ages 5-19.	4-40
                        Figure 4-5    Adjusted asthma hospital admissions by age on lagged ozone by alert
                                     status,  ages 20-64	4-40
          4.6.5   Exposure Estimation Methods in Epidemiologic Studies	4-41
     4.7   Summary and Conclusions	  4-41
     4.8   References	  4-44

 5   DOSIMETRY AND MODE OF ACTION	5-1
     5.1   Introduction	 5-1
                        Figure 5-1    Schematic of the O3 exposure and response pathway.	5-2
     5.2   Human and Animal Ozone Dosimetry	 5-2
          5.2.1   Introduction	5-2
                        Figure 5-2    Representation of respiratory tract regions in humans.	5-3
                        Figure 5-3    Structure of lower airways with progression from the large airways to the
                                     alveolus.	5-4
          5.2.2   Ozone Uptake	5-5
- Do Not Cite or                                   xiii                                        September 2011

-------
                          Table 5-1     Human respiratory tract uptake efficiency data	5-6
                  5.2.2.1  Gas Transport Principles	5-6
                  5.2.2.2  Target Sites for Ozone Dose	5-7
                  5.2.2.3  Upper Respiratory Tract Ozone Removal and Dose	5-8
                  5.2.2.4  Lower Respiratory Tract Ozone Uptake and Dose 	5-9
                          Figure 5-4    Total ozone uptake efficiency as a function of breathing frequency at a
                                       constant minute ventilation of 30 L/min.	5-11
                          Figure 5-5    Ozone uptake fraction as a function of volumetric penetration (Vp) in a
                                       representative subject.	5-12
                  5.2.2.5  Mode of Breathing	5-12
                  5.2.2.6  Interindividual Variability in Dose	5-13
                  5.2.2.7  Physical Activity	5-15
                          Figure 5-6    Modeled effect of exercise on tissue dose of the  LRT.	5-16
                  5.2.2.8  Summary    5-16
           5.2.3   Ozone Reactions and Reaction Products 	5-17
                          Figure 5-7    Schematic overview of ozone interaction with PUFA in ELF and lung
                                       cells.	5-19
                  5.2.3.1  Summary    5-24
                          Figure 5-8    Details of the O3 interaction with the airway ELF to form secondary
                                       oxidation products.	5-24
     5.3   Possible Pathways/Modes of Action	5-25
           5.3.1   Introduction	5-25
           5.3.2   Activation of Neural Reflexes	5-26
           5.3.3   Initiation of inflammation	5-29
           5.3.4   Alteration of epithelial barrier function	5-34
           5.3.5   Sensitization of bronchial smooth muscle	5-35
           5.3.6   Modification  of innate/adaptive  immune system responses	5-39
           5.3.7   Airways remodeling	5-43
           5.3.8   Systemic inflammation and oxidative/nitrosative stress	5-44
           5.3.9   Impaired  alveolar-arterial O2 transfer	5-45
           5.3.10  Summary	5-46
                          Figure 5-9    The modes of action/possible pathways underlying the health effects
                                       resulting from inhalation exposure to O3.	5-46
     5.4   Interindividual Variability in Response	  5-50
           5.4.1   Dosimetric Considerations	5-50
           5.4.2   Mechanistic  Considerations	5-52
                  5.4.2.1  Gene-Environment Interactions	5-52
                  5.4.2.2  Pre-existing  Diseases and Conditions	5-56
                  5.4.2.3  Nutritional Status	5-61
                  5.4.2.4  Lifestage     5-62
                  5.4.2.5  Attenuation of Responses	5-64
                  5.4.2.6  Co-Exposures with Particulate Matter	5-66
                  5.4.2.7  Summary    5-67
                          Figure 5-10   Factors which contribute to the interindividual variability in responses
                                       resulting from inhalation exposure to ozone.	5-67
     5.5   Species Homology and Interspecies  Sensitivity	5-67
           5.5.1   Dosimetry	5-68
                          Figure 5-11   Species comparison of antioxidant / protein ratios of: (a) nasal lavage
                                       fluid and, (b) bronchoalveolar lavage fluid. 	5-70
                          Figure 5-12   Humans and animals are similar in the regional pattern of O3 tissue dose
                                       distribution.	5-71
                          Figure 5-13   Oxygen-18 incorporation into different fractions of BALF from humans
                                       and rats exposed to 0.4 and 2.0 ppm 18O3.	5-72
           5.5.2   Homology of Response	5-73
           5.5.3   Summary	5-75
     5.6   Chapter Summary 	  5-76
     5.7   References                                                                                          5-77
- Do Not Cite or                                     xiv                                                       2011

-------
 6   INTEGRATED HEALTH EFFECTS OF SHORT-TERM OZONE EXPOSURE
                                                                                           6-1
6.1
6.2
Introduction
Respiratory Effects
6.2.1 Lunq Function
6.2.1.1 Controlled Human Exposure
Figure 6-1 Cross-study comparison of mean ozone-induced FEVi decrements
following 6.6 hours of exposure to ozone.
Figure 6-2 Frequency distributions of FENA, decrements observed by Schelegle et al.
6-1
6-1
6-3
6-3
6-7

                                     (2009) in young healthy adults (16 F, 15 M) following 6.6-h exposures to
                                     ozone or filtered air.
                 6.2.1.2  Epidemiology6-23
                         Table 6-1
                         Figure 6-3
                         Table 6-2
                         Figure 6-4
                         Table 6-3
                         Figure 6-5
                         Table 6-4
                         Table 6-5
                         Table 6-6
                         Figure 6-6
                         Table 6-7
                         Figure 6-7
                         Table 6-8
                         Table 6-9
                         Table 6-10
                         Figure 6-8
                         Table 6-11
                         Table 6-12
                         Figure 6-9
                    Mean and upper percentile concentrations of ozone in epidemiologic
                    studies examining lung function in populations with increased outdoor
                    exposures	
                    Changes in FEVi (ml) or PEF (ml/sec) in association with ambient ozone
                    exposure in studies of children attending summer camp.	
                    Additional characteristics and quantitative data for studies represented in
                    Figure 6-3	
                    Percent change in FEVi in association with ambient ozone exposures of
                    adults exercising outdoors. 	
                    Additional characteristics and quantitative data for studies represented in
                    Figure 6-4 and results from studies in children exercising outdoors	
                    Percent change in lung function in association with ambient ozone
                    exposures among outdoor workers.	
                    Additional characteristics and quantitative data for studies represented in
                    Figure 6-5	
                    Associations between ambient ozone exposure and lung function
                    decrements in different ranges of ambient ozone concentrations	
                    Mean and upper percentile concentrations of ozone in epidemiologic
                    studies examining lung function in children with asthma	
                    Percent change in FEVi in association with ambient ozone exposures
                    among children with asthma.	
                    Additional characteristics and quantitative data for studies represented in
                    Figure 6-6	
                    Percent change in PEF or FEF2s-75% in association with ambient ozone
                    exposures among children with asthma.	
                    Additional characteristics and quantitative data for studies represented in
                    Figure 6-7	
                    Mean and upper percentile concentrations of ozone in epidemiologic
                    studies examining lung function in adults with respiratory disease	
                                                                                           6-13
                    Mean and upper percentile concentrations of ozone in epidemiologic
                    studies examining lung function in populations not restricted to individuals
                    with asthma	
                    Percent change in lung function in association with ambient ozone
                    exposures in studies not restricted to children with asthma.	
                    Additional characteristics and quantitative data for studies represented in
                    Figure 6-8 and results from other studies in children	
                    Associations between ambient ozone exposure and changes in lung
                    function in studies of adults	
                    Comparison of ozone-associated changes in lung function in single- and
                    copollutant models.	
          6.2.2
          6.2.3
        Table 6-13   Additional characteristics and quantitative data for studies presented in
                    Figure 6-9	
6.2.1.3  Toxicology   6-56
Airway Hyperresponsiveness	
6.2.2.1  Controlled Human Exposures	
6.2.2.2  Toxicology   6-59
Pulmonary Inflammation, Injury and Oxidative Stress	
6.2.3.1  Controlled Human Exposures	
6.2.3.2  Epidemiology6-66
_6-24

_6-26

_6-27

_6-28

_6-29

_6-31

_6-31

_6-32

_6-34

_6-35

_6-36

_6-37

_6-38

_6-44


_6-45

_6-46

_6-47

_6-49

_6-52

_6-53


_6-57
_6-57


_6-61
 6-61
- Do Not Cite or
                                                    xv
                                                                              September 2011

-------
                         Table 6-14   Mean and upper percentile ozone concentrations in studies examining
                                      biological markers of pulmonary inflammation and oxidative stress	6-67
                         Figure 6-10  Percent change in exhaled nitric oxide (eNO) per standardized increment
                                      in ambient ozone exposure in studies of individuals with and without
                                      asthma. 	6-68
                         Table 6-15   Additional characteristics and quantitative data for studies represented in
                                      Figure 6-10	6-69
                         Table 6-16   Associations between short-term ambient ozone exposure and biological
                                      markers of pulmonary inflammation and oxidative stress	6-69
                 6.2.3.3 Toxicology   6-79
                         Table 6-17   Morphometric observations in non-human primates after acute O3
                                      exposure 	6-81
          6.2.4  Respiratory Symptoms and Medication Use	6-84
                 6.2.4.1  Children with Asthma	6-86
                         Table 6-18   Mean and upper percentile ozone concentrations in epidemiologic studies
                                      examining respiratory symptoms, medication use, and activity levels in
                                      children with asthma	6-87
                         Figure 6-11   Associations of ambient ozone exposure with respiratory symptoms in
                                      children with asthma.	6-88
                         Table 6-19   Additional characteristics and quantitative data for studies presented in
                                      Figure 6-11	6-89
                         Figure 6-12  Associations of ambient ozone exposure with asthma medication use.	6-92
                         Table 6-20   Additional characteristics and quantitative data for studies presented in
                                      Figure 6-12	6-92
                 6.2.4.2 Adults with Respiratory Disease 	6-93
                         Table 6-21   Mean and upper percentile ozone concentrations in epidemiologic studies
                                      examining respiratory symptoms and medication  use in adults with
                                      respiratory disease	6-94
                 6.2.4.3 Populations not Restricted to Individuals with Asthma	6-95
                         Table 6-22   Mean and upper percentile ozone concentrations in epidemiologic studies
                                      examining respiratory symptoms in populations not restricted to
                                      individuals with asthma	6-95
                         Figure 6-13  Associations of ambient ozone exposure with respiratory symptoms in
                                      studies not restricted to children with asthma.	6-97
                         Table 6-23   Additional characteristics and quantitative data for studies presented in
                                      Figure 6-13	6-97
                 6.2.4.4 Confounding in Epidemiologic Studies of Respiratory Symptoms and Medication Use	6-98
                         Table 6-24   Associations between short-term ozone exposure and respiratory
                                      symptoms in single- and copollutant models	6-99
                 6.2.4.5 Summary of Epidemiologic Studies of Respiratory Symptoms and Asthma Medication Use _6-100
          6.2.5  Lung Host Defenses	6-101
                 6.2.5.1  Mucociliary  Clearance	6-102
                 6.2.5.2 Alveolobronchiolar Transport Mechanism	6-102
                 6.2.5.3 Alveolar Macrophages	6-103
                 6.2.5.4 Infection and Adaptive Immunity	6-104
          6.2.6  Allergic and Asthma-Related Responses	6-108
          6.2.7  Hospital Admissions, Emergency Department Visits, and Physicians Visits	6-110
                 6.2.7.1  Summary of Findings from 2006 Ozone AQCD	6-110
                         Table 6-25   Mean and upper percentile concentrations of respiratory-related hospital
                                      admission and emergency department visit studies evaluated	6-112
                 6.2.7.2 Hospital Admission Studies	6-113
                         Figure 6-14  Percent increase in respiratory hospital admissions from natural spline
                                      models for a 40 ppb increase in 1-h max ozone concentrations for each
                                      location of the APHENA study.  	6-116
                         Table 6-26   Corresponding effect estimates for Figure 6-14	6-117
                         Figure 6-15  Estimated relative risks (RRs) of ozone-related asthma hospital
                                      admissions allowing for possible nonlinear relationships using natural
                                      splines.	6-123
                 6.2.7.3 Emergency Department Visit Studies 	6-124
                         Figure 6-16  Risk ratio for respiratory ED visits and different ozone exposure metrics in
                                      Atlanta from  1993-2004.                                                 6-125
- Do Not Cite or                                     xvi                                                     2011

-------
                          Figure 6-17  Loess dose-response estimates and twice-standard error estimates from
                                      generalized additive models for associations between 3-day avg ozone
                                      concentrations and ED visits for pediatric asthma.	
                 6.2.7.4
                 6.2.7.5
Figure 6-18   Relative risk of asthma ED visits children and adults for a 30 ppb increase
             in max 8-h avg O3 concentrations in Seattle, WA, 1998-2002.	
Outpatient and Physician Visit Studies	
Summary
Figure 6-19
                         Table 6-27
                         Figure 6-20
            6-132
            Percent increase in respiratory-related hospital admission and ED visits in
            studies that presented all-year and/or seasonal results.	
            Corresponding Effect Estimates for Figure 6-19	
                         Table 6-28
          6.2.8  Respiratory Mortality	
          6.2.9  Summary and Causal Determination
     6.3   Cardiovascular Effects	
          6.3.1  Controlled Human Exposure	
          6.3.2  Epidemiology	
                 6.3.2.1  Arrhythmia   6-145
             Percent increase in respiratory-related hospital admissions and ED visits
             for studies that presented single and copollutant model results.	
             Corresponding effect estimates for Figure 6-20	
                 6.3.2.2
                 6.3.2.3
                 6.3.2.4
                 6.3.2.5
                 6.3.2.6
Table 6-29   Characterization of ozone concentrations (in ppb) from studies of
             arrhythmias	
Heart Rate/Heart Rate Variability	
Table 6-30   Characterization of ozone concentrations (in ppb) from studies of heart
             rate variability	
             6-151
Stroke
Figure 6-21
Biomarkers
Table 6-31
             Odds ratio (95% confidence interval) for stroke by quintiles of ozone
             6-153
             Characterization of ozone concentrations (in ppb) from studies of
             biomarkers
Myocardial Infarction (Ml)
Blood Pressure	
Table 6-32
                                      Characterization of ozone concentrations (in ppb) from studies of blood
                                      pressure	
                 6.3.2.7   Hospital Admissions and Emergency Department Visits	
                          Table 6-33   Characterization of ozone concentrations (in ppb) from studies of HAs
                                      and ED visits
                          Figure 6-22
                         Table 6-34
                          Figure 6-23
                         Table 6-35
                          Figure 6-24
                         Table 6-36
             Odds Ratio (95% Cl) per increment ppb increase in ozone for congestive
             heart failure ED visits or HAs for studies in Figure 6-23	
             Odds Ratio (95% confidence interval) per increment ppb increase in
             ozone for ischemic heart disease, coronary heart disease, myocardial
             infarction, and angina pectoris ED visits or HAs.	
             Odds Ratio (95% Cl) per increment ppb increase in ozone for ischemic
             heart disease, coronary heart disease, myocardial infarction, and angina
             pectoris ED visits or HAs for studies presented in Figure 6-24	
                          Figure 6-25  Odds Ratio (95% confidence interval) per increment ppb increase in
                                      ozone for stroke ED visits or HAs.	
                          Table 6-37
             Odds Ratio (95% Cl) per increment ppb increase in ozone for stroke ED
             visits or HAs for studies presented in Figure 6-25	
                          Figure 6-26  Odds Ratio (95% confidence interval) per increment ppb* increase in
                                      ozone for arrhythmia and dysrhythmia ED visits or HAs.	
                 6.3.2.8
                 6.3.2.9
Table 6-38   Odds Ratio (95% Cl) per increment ppb* increase in ozone for arrhythmia
             and dysrhythmia ED visits or HAs for studies presented in Figure 6-26	
Cardiovascular Mortality	
Summary of Epidemiologic Studies_
                                                                                   6-128
                                                                                  _6-130
                                                                                   6-131
                                                                                  _6-133
                                                                                   6-134
                                                                                  _6-135
                                                                                  _6-136
                                                                                  _6-136
                                                                                  _6-137
                                                                                   6-143
                                                                                  _6-143
                                                                                   6-144
                                                                                  _6-145
                                                                                   6-147
                                                                                                            6-148
                                                                                                            6-153
                                                                                  _6-154
                                                                                  _6-159
                                                                                   6-160
                                                                                  _6-160
                                                                                   6-161
                                                                                   6-162
             Odds ratio (95% Cl) per increment ppb increase in ozone for over all
             cardiovascular ED visits or HAs.	6-166
             Odds ratio (95% Cl) per increment ppb increase in ozone for overall
             cardiovascular ED visits or HAs in studies presented in Figure 6-22 	6-167
             Odds Ratio (95% Cl) per increment ppb increase in ozone for congestive
             heart failure ED visits or HAs. 	6-168
                                                                                                            6-169
                                                                                                            6-170
                                                                                                            6-171
                                                                                   6-172
                                                                                                            6-173
                                                                                   6-173
          6.3.3  Toxicology
                                                                                  _6-174
                                                                                  _6-174
                                                                                  _6-175
                                                                                   6-175
- Do Not Cite or
                                                    XVII
                                                                       September 2011

-------
     6.4
       6.3.3.1  Summary of Findings from Previous Ozone AQCDs	
       6.3.3.2  Recent Cardiovascular Toxicology Studies	
               Table 6-39   Characterization of study details for Section 6.3.3.2	
6.3.4   Summary and Causal Determination	
Central Nervous System Effects	
               Table 6-40   Central Nervous System and Behavioral Effects of Short-term O3
                           Exposure in Rats	
       Neuroendocrine Effects	
          6.4.1                         	
          6.4.2  Summary and Causal Determination	
     6.5   Effects on Other Organ Systems	
          6.5.1   Effects on the Liver and Xenobiotic Metabolism 	
          6.5.2  Effects on Cutaneous and Ocular Tissues	
     6.6   Mortality	
          6.6.1   Summary of Findings from 2006 Ozone AQCD	
          6.6.2  Associations of Mortality and Short-Term Ozone Exposure_
                         Figure 6-27


                         Table 6-41
                         Table 6-42
                 6.6.2.1
                 6.6.2.2
                 6.6.2.3
                 6.6.2.4
                           Range of mean and upper percentile ozone concentrations in previous
                           and recent multicity studies	
               Confounding 6-198
               Table 6-43   Correlations between PM and ozone by season and region	
               Figure 6-28

               Figure 6-29
                                      Scatter plots of ozone mortality risk estimates with versus without
                                      adjustment for PM10 in NMMAPS cities.	
                                      Community-specific ozone-mortality risk estimates for nonaccidental
                                      mortality per 10 ppb increase in same-day 24-h avg summertime ozone
                                      concentrations in single-pollutant models and copollutant models with
                                      sulfate.	
                         Table 6-44
                         Table 6-45
                           Sensitivity of ozone risk estimates per 10 |jg/m3 increase in 24-h avg
                           ozone concentrations at lag 0-1 to alternative methods for adjustment of
                           seasonal trend, for all-cause mortality using Berkey MLE and TLNSE
                           Hierarchical Models	
               Effect Modification	
               Table 6-46
                                      Additional percent change in ozone-related mortality for individual-level
                                      susceptibility factors	
                         Figure 6-31

                         Table 6-47


                         Table 6-48


                         Figure 6-32

                         Figure 6-33
                           Ozone mortality risk estimates and community-specific characteristics,
                           U.S., 1987-2000.	
                           Community-specific Bayesian ozone-mortality risk estimates in 98 U.S.
                           communities.	
               Interaction
               Evaluation of the Ozone-Mortality C-R Relationship and Related Issues	
               Table 6-49   Estimated effect of a 10 ppb increase in 8-h max ozone concentrations on
                                      mortality during the summer months for single-day and distributed lag
                                      models
                         Figure 6-34
                                                                     _6-175
                                                                     _6-177
                                                                     _6-182
                                                                     _6-183
                                                                      6-184
                                                                     _6-189
                                                                     _6-190
                                                                                                 6-192
                                                                                                _6-192
                                                                                                _6-193
                                                                                                 6-193
                                                                                                _6-193
                                                                                                 6-194
                           Summary of mortality risk estimates for short-term ozone exposure and
                           all-cause (nonaccidental) mortality from all-year and summer season
                           analyses.	6-195
                           Corresponding effect estimates for Figure 6-27	6-196
                                                                                                            6-197
                                                                                                            6-199
                                                                      6-200
                                                                                                 6-202
               Figure 6-30  Percent increase in all-cause (nonaccidental) and cause-specific mortality
                           from the APHENA study for single- and copollutant models.	6-204
                           Corresponding Effect Estimates for Figure 6-30	6-205
                                                                     _6-207
                                                                      6-207
                                                                                                 6-209
                                                                      6-211
                           Percent change in all-cause mortality, for all ages, associated with a
                           40ppb increase in 1-h max ozone concentrations at Lag 0-1 at the 25th
                           and 75th percentile of the center-specific distribution of selected effect
                           modifiers	6-212
                           Percentage increase in daily mortality for a 10 ppb increase in 24-h avg
                           ozone concentrations during the previous week by geographic region in
                           the U.S., 1987-2000	6-213
                                                                      6-214
Map of spatially dependent ozone-mortality coefficients for 8-h max
ozone concentrations using summer data.  	6-214
6-215
                                                                      6-216
                                                                                                 6-218
                           Estimated combined smooth distributed lag for 48 U.S. cities during the
                           summer months.                                                       6-219
- Do Not Cite or
                                                    XVIII
                                                                                      September 2011

-------
                         Table 6-50   Estimated percent increase in cause-specific mortality (and 95% CIs) for
                                     a 10-|jg/m  increase in maximum 8-h ozone during June-August, for the
                                     same day (lag 0), the average of the same and previous day (lag 0-1), the
                                     unconstrained distributed lag model for the sum of 0-20 days and the
                                     penalized distributed lag model (lag 0-20)	6-220
                         Figure 6-35  Estimated combined smooth distributed lag in 21 European cities during
                                     the summer (June-August) months.	6-222
                         Table 6-51   Percent excess all-cause mortality per 10 ppb increase in daily 8-h max
                                     ozone on the same day, by season, month, and age groups	6-223
                         Figure 6-36  Estimated combined C-R curve for  ozone and nonaccidental mortality
                                     using the nonlinear (spline) model.  	6-225
                 6.6.2.5  Associations of Cause-Specific Mortality and Short-term Ozone Exposure	6-227
                         Figure 6-37  Percent increase in cause-specific mortality.	6-229
                         Table 6-52   Corresponding effect estimates for  Figure 6-37	6-230
          6.6.3   Summary and Causal Determination	6-231
     6.7   Overall Summary	 6-233
                         Table 6-53   Summary of causal determinations  for short-term exposures to ozone	6-233
     6.8   References	6-234

 7   INTEGRATED HEALTH EFFECTS OF LONG-TERM OZONE EXPOSURE                    7-1
7.1 Introduction
7.2 Respii
7.2.1
7.2.2
7.2.3
7.2.4
7.2.5
7.2.6
7.2.7
7.2.8
•atory Effects
New Onset Asthma
Figure 7-1 Interaction of gene presence and O3 level on the Hazard Ratio (HR) of
new-onset asthma in the 12 Children's Health Study communities.
Figure 7-2 Ozone modifies the effect of TNF G-308A genotype on bronchitic
symptoms among children with asthma in the CHS.
Asthma Hospital Admissions and ED Visits
Figure 7-3 Ozone-asthma concentration-response relationship using the mean
concentration during the entire follow-up period.
Pulmonary Structure and Function
7.2.3.1 Pulmonary Structure and Function: Evidence from Toxicoloqical Studies
Table 7-1 Respiratory effects in nonhuman primates and rodents resulting from
long-term O3 exposure
Pulmonary Inflammation, Iniury, and Oxidative Stress
Allergic Responses
Host Defense
Respiratory Mortality
Summary and Causal Determination
7-1
7-1
7-2
7-5
7-8
7-11
7-13
7-14
7-17
7-22
7-23
7-25
7-27
7-27
7-28
                                     and respiratory health effects	7-29
                         Table 7-3    Studies providing evidence concerning potential confounding by PM for
                                     available endpoints	7-31
     7.3   Cardiovascular Effects	 7-33
          7.3.1   Cardiovascular Disease	7-33
                 7.3.1.1   Cardiovascular Epidemiology	7-33
                 7.3.1.2  Cardiovascular Toxicology	7-34
                         Table 7-4    Characterization of study details for Section 7.3.1.2	7-36
          7.3.2   Cardiac Mortality	7-36
          7.3.3   Summary and Causal Determination	7-36
     7.4   Reproductive and Developmental Effects	7-37
          7.4.1   Effects on Sperm	7-38
          7.4.2   Effects on Reproduction	7-39
          7.4.3   Birth Weight	7-41
                         Figure 7-4    Birthweight deficit by decile of 24-h avg O3 concentration averaged over
                                     the entire pregnancy compared with the decile group with the lowest O3
                                     exposure.	7-42
                         Table 7-5    Brief summary of epidemiologic studies of birth weight	7-44
          7.4.4   Preterm  Birth                                                                              7-45
- Do Not Cite or                                   xix                                                    2011

-------
                        Table 7-6   Brief summary of epidemiologic studies of PTB	7-49
          7.4.5   Fetal Growth	7-50
                        Table 7-7   Brief summary of epidemiologic studies of fetal growth	7-53
          7.4.6   Postnatal growth	7-53
          7.4.7   Birth Defects	7-54
                        Table 7-8   Brief summary of epidemiologic studies of birth defects	7-56
          7.4.8   Developmental Respiratory Effects	7-57
          7.4.9   Developmental Central Nervous System Effects	7-61
                 7.4.9.1  Laterality    7-61
                 7.4.9.2  Brain Morphology and Neurochemical Changes	7-61
                 7.4.9.3  Neurobehavioral Outcomes	7-62
                 7.4.9.4  Sleep Aberrations after Developmental Ozone Exposure	7-63
          7.4.10  Early Life Mortality	7-63
                 7.4.10.1 Stillbirth     7-64
                 7.4.10.2 Infant Mortality, Less than 1 Year	7-64
                 7.4.10.3 Neonatal Mortality, Less than 1 Month	7-65
                 7.4.10.4 Postneonatal Mortality, 1  Month to 1 Year	7-65
                 7.4.10.5 Sudden Infant Death Syndrome	7-67
                        Table 7-9   Brief summary of infant mortality studies	7-68
                        Table 7-10   Summary of Key Reproductive and Developmental Toxicological Studies	7-69
          7.4.11  Summary and Causal Determination	7-70
     7.5   Central Nervous System Effects	7-71
          7.5.1   Effects on the Brain and Behavior	7-71
                        Table 7-11   Central nervous system effects of long-term O3 exposure in rats	7-73
          7.5.2   Summary and Causal Determination	7-74
     7.6   Carcinogenic and Genotoxic Potential of Ozone	7-75
          7.6.1   Introduction	7-75
          7.6.2   Lung Cancer Incidence and Mortality	7-77
          7.6.3   DNA Damage	7-77
          7.6.4   Summary and Causal Determination	7-80
     7.7   Mortality	 7-80
                        Figure 7-5   Adjusted ozone-mortality relative risk estimates (95% Cl) by time period
                                    of analysis per subject-weighted mean O3 concentration in the Cancer
                                    Prevention  Study II by the American Cancer Society.	7-81
                        Table 7-12   Relative risk (and 95% Cl) of death attributable to a 10-ppb change in the
                                    ambient O3 concentration*	7-84
          7.7.1   Summary and Causal Determination	7-85
     7.8   Overall Summary	 7-85
                        Table 7-13   Summary of causal determinations for long-term exposures to ozone	7-85
     7.9   References	 7-86

     POPULATIONS POTENTIALLY AT INCREASED RISK FOR OZONE-RELATED HEALTH
     EFFECTS	8-1
     8.1   Preexisting Disease/Conditions	 8-3
8.2
8.3
8.4
Table 8-1 Prevalence of respiratory diseases, cardiovascular diseases, and
diabetes among adults by aqe and region in the U.S.
8.1.1 Influenza/Infections
8.1.2 Asthma 8-4
8.1 .3 Chronic Obstructive Pulmonary Disease (COPD)
8.1.4 Cardiovascular Disease
8.1.5 Diabetes
8.1.6 Hyperthyroidism
Lifestaqe
8.2.1 Children 8-11
8.2.2 Older Adults
Sex
Genetics
8-4
8-4
8-7
8-8
8-10
8-10
8-10
8-14
8-16
8-18
- Do Not Cite or                                   xx                                        September 2011

-------
8.5
8.6
8.7
8.8
8.9
8.10
8.11
8.12
8.13
Diet
Body Mass Index and Physical Conditioning
Socioeconomic Status
Race/Ethnicity
Smoking
Heightened Exposure
Healthy Responders
Summary
References
8-23
8-25
8-26
8-28
8-29
8-30
8-32
8-32
8-34
 9   ENVIRONMENTAL EFFECTS: OZONE EFFECTS ON VEGETATION AND ECOSYSTEMS  9-1
     9.1   Introduction	 9-1
                         Figure 9-1   An illustrative diagram of the major pathway through which O3 enters
                                    plants and the major endpoints that O3 may affect in plants and
                                    ecosystems.	9-3
     9.2   Experimental Exposure Methodologies	 9-3
          9.2.1   Introduction	9-3
          9.2.2   "Indoor," Controlled Environment, and Greenhouse Chambers	9-4
          9.2.3   Field Chambers	9-4
          9.2.4   Plume and FACE-Type Systems	9-6
          9.2.5   Ambient Gradients	9-7
          9.2.6   Comparative Studies	9-8
     9.3   Mechanisms Governing Vegetation Response to Ozone	  9-10
          9.3.1   Introduction	9-10
          9.3.2   Ozone Uptake into the Leaf	9-12
                 9.3.2.1   Changes in Stomatal  Function	9-13
                         Figure 9-2   The  microarchitecture of a dicot leaf.	9-15
                         Figure 9-3   Possible reactions of ozone within water. 	9-15
                         Figure 9-4   The  Crigee mechanism of ozone attack of a double bond.	9-16
          9.3.3   Cellular to Systemic Responses	9-17
                 9.3.3.1   Ozone Sensing and Signal Transduction	9-17
                 9.3.3.2  Gene and Protein  Expression Changes in Response to Ozone	9-18
                         Figure 9-5   Composite diagram of major themes in the temporal evolution of the
                                    genetic  response to ozone stress.	9-23
                 9.3.3.3  Role of Phytohormones in Plant Response to Ozone	9-24
                         Figure 9-6   The  oxidative cell death cycle.	9-26
          9.3.4   Detoxification	9-26
                 9.3.4.1   Overview of Ozone-Induced Defense Mechanisms	9-26
                 9.3.4.2  Role of Antioxidants in Plant Defense  Responses	9-27
          9.3.5   Effects on Primary and Secondary Metabolism	9-30
                 9.3.5.1   Light and Dark Reactions of Photosynthesis	9-30
                 9.3.5.2  Respiration and Dark  Respiration	9-32
                 9.3.5.3  Secondary Metabolism	9-33
          9.3.6   Summary	9-36
     9.4   Nature of Effects on Vegetation and Ecosystems	  9-38
          9.4.1   Introduction	9-38
                 9.4.1.1   Ecosystem  Scale,  Function, and Structure	9-39
                 9.4.1.2  Ecosystem  Services	9-40
          9.4.2   Visible Foliar Injury and Biomonitoring	9-41
                 9.4.2.1   Biomonitoring9-42
                 9.4.2.2  Summary   9-44
          9.4.3   Growth, productivity and carbon storage in natural ecosystems	9-45
                 9.4.3.1   Plant growth and biomass allocation	9-45
                 9.4.3.2  Summary   9-49
                 9.4.3.3  Reproduction 9-49
- Do Not Cite or                                   xxi                                                    2011

-------
                         Table 9-1    Ozone effects on plant reproductive processes (derived from Table AX9-
                                      22 of the 2006 ozone AQCD)	9-51
                 9.4.3.4 Ecosystem Productivity and Carbon Sequestration	9-51
                         Table 9-2    Comparison of models used to simulate the ecological consequences of
                                      O3 exposure	9-53
                 9.4.3.5 Summary    9-57
                         Table 9-3    Modeled effects of ozone on primary production, C exchange, and
                                      C sequestration	9-58
           9.4.4  Crop yield and quality in agricultural systems	9-58
                 9.4.4.1  Yield        9-59
                 9.4.4.2 Crop Quality 9-64
                 9.4.4.3 Summary    9-64
                         Table 9-4    Summary of recent studies of ozone effects on crops (exclusive of growth
                                      and yield)	9-66
                         Table 9-5    Modeled effects of ozone on crop yield loss at regional and global scales	9-68
           9.4.5  Water Cycling	9-68
                         Figure 9-7   The potential effects of ozone exposure on watering cycling.	9-69
                 9.4.5.1  Summary    9-71
           9.4.6  Below-Ground Processes	9-72
                         Figure 9-8   Conceptual diagram showing where ozone alters C, water and nutrient
                                      flow in a tree-soil system, including transfer between biotic and abiotic
                                      components below ground that influence soil physical and chemical
9.4.6.1
9.4.6.2
9.4.6.3
9.4.6.4
9.4.6.5
9.4.6.6
properties.
Litter Carbon Chemistry, Litter Nutrient and Their Ecosystem Budqets
Table 9-6 The effect of elevated ozone on leaf/litter nutrient concentrations
Decomposer Metabolism and Litter Decomposition
Soil respiration and carbon formation
Table 9-7 The temporal variation of ecosystem responses to ozone exposure at
Aspen FACE site
Nutrient cvclinq
Dissolved Orqanic Carbon and Bioqenic Trace Gases Emission
Summary 9-81
9-73
9-73
9-75
9-75
9-76
9-77
9-79
9-80

           9.4.7  Community composition	9-81
                 9.4.7.1   Forest       9-82
                 9.4.7.2   Grassland and Agricultural Land	9-83
                 9.4.7.3   Microbes    9-84
                 9.4.7.4   Summary    9-85
           9.4.8  Factors that Modify Functional and Growth Response  	9-86
                 9.4.8.1   Genetics    9-87
                 9.4.8.2   Environmental Biological Factors	9-87
                 9.4.8.3   Physical Factors	9-88
                 9.4.8.4   Interactions with other Pollutants	9-89
                          Table 9-8    Response of plants to the interactive effects of elevated ozone exposure
                                      and N enrichment	9-91
           9.4.9  Insects  and Other Wildlife  	9-93
                 9.4.9.1   Insects      9-93
                 9.4.9.2   Wildlife      9-96
                 9.4.9.3   Indirect Effects on Wildlife	9-97
                 9.4.9.4   Summary    9-100
     9.5   Effects-Based Air Quality Exposure Indices and Dose Modeling	9-100
           9.5.1  Introduction	9-100
           9.5.2  Description of Exposure Indices Available in the Literature	9-101
                          Figure 9-9    Diagrammatic representation of several exposure indices	9-102
           9.5.3  Important Components of Exposure Indices	9-107
                 9.5.3.1   Role of Concentration 	9-107
                          Figure 9-10  Trends in May to September 12-h SUM06, peak 1-h ozone concentration
                                      and number of daily exceedances of 95 ppb	9-109
                          Figure 9-11   The number of hourly average concentrations between 50 and 89 ppb for
                                      the period 1980-2000 for the Crestline, San Bernardino County, CA,
                                      monitoring site.	9-110
- Do Not Cite or                                   xxii                                                      2011

-------
9.5.3.2  Diurnal and Seasonal Exposure	
        Figure 9-12   Mean diurnal.	
        Figure 9-13   Maximum 3-month, 12-h W126 plotted against maximum 6-month, 12-h
                    W126.	
Ozone Uptake/Dose Modeling for Vegetation	
Summary	
                                                                                                       9-113
     9.5.4
     9.5.5
9.6  Ozone Exposure-Plant Response Relationships
     9.6.1   Introduction	
     9.6.2
                                                                                                       9-120
                                                                                                       9-120
                 Estimates of Crop Yield Loss and Tree Seedling Biomass Loss in the 1996 and 2006 Ozone AQCDs 9-123
                        Figure 9-14  Quantiles of predicted relative yield loss for 34 NCLAN crop experiments.	9-125
                        Figure 9-15  Quantiles of predicted relative yield loss for 4 crop species in NCLAN
                                    experiments.	9-126
                        Figure 9-16  Quantiles of predicted relative biomass loss for 49 tree species in
                                    NHEERL/WED experiments.	
                        Figure 9-17  Quantiles of predicted relative biomass loss for 4 tree species in
                                    NHEERL/WED experiments.	
                        Table 9-9
                                                                                      9-127
                                                                                      9-128
                    Ozone exposures at which 10 and 20% yield loss is predicted for 50 and
                    75% of crop species	
                                                                                                       9-129
                        Table 9-10
                        Table 9-11
                    Ozone exposures at which 10 and 20% yield loss is predicted for 50 and
                    75% of crop species under drought conditions and adequate moisture	
                    Ozone exposures at which 10 and 20% biomass loss is predicted for 50
                    and 75 %of tree species, based on composite functions for the 50th and
                    75th percentiles of 49 Weibull curves for relative above-ground biomass
                    loss data from 49 studies of 11 tree species grown under well-watered
                    conditions for 1 or 2 year; curves were standardized to 90-day W126	
                                                                                                       9-129
          9.6.3
                                                                                                _9-130
            Validation of 1996 and 2006 Ozone AQCD Models and Methodology Using the 90 day 12-h W126 and
            Current FACE Data                                                                     9-130
                9.6.3.1
        Comparison of NCLAN-Based Prediction and SoyFACE Data	9-132
        Table 9-12   Comparison between change in yield observed in the SoyFACE
                    experiment between elevated and ambient ozone, and change predicted
                    at the same values of ozone by the median composite function for
                    NCLAN	9-133
        Table 9-13   Comparison between yield observed in the SoyFACE experiment and
                    yield predicted at the same values of ozone by the median composite
                    function for NCLAN	9-133
        Figure 9-18   Comparison of yield observed in SoyFACE experiment in a given year
                    with yield predicted by the median composite function based on NCLAN.	9-134
        Figure 9-19   Comparison of composite functions for the quartiles of 7 curves for 7
                    genotypes of soybean grown in the SoyFACE experiment, and for the
                    quartiles of 11 curves for 5 genotypes of soybean grown in the NCLAN
                    project.	9-135
                9.6.3.2  Comparison of NHEERL/WED-Based Prediction of Tree Biomass Response and Aspen FACE
                        Data
                        Table 9-14
                        Table 9-15
                   9-136
                   Comparison between change in above-ground biomass elevated and
                   ambient ozone in Aspen FACE experiment in 6 year, and change
                   predicted at the same values of ozone by the median composite function
                   for NHEERL/WED (two-parameter relative biomass model)	
                                                                                                       9-137
                    Comparison between above-ground biomass observed in Aspen FACE
                    experiment in 6 year and biomass predicted by the median composite
                    function based on NHEERL/WED (three-parameter absolute biomass
                    model with intercept scaled to Aspen FACE data)	
                                                                                                       9-137
                9.6.3.3
        Figure 9-20   Comparison between above-ground biomass observed in Aspen FACE
                    experiment in 6 year and biomass predicted by the median composite
                    function based on NHEERL/WED.	
        Exposure-Response in a Gradient Study	
        Figure 9-21   Above-ground biomass for one genotype of cottonwood grown in seven
                                                                                                 _9-138
                                                                                                  9-139
                                    locations for one season in 3 years._
                9.6.3.4  Meta-analyses of growth and yield studies	
                        Table 9-16   Meta-analyses of growth or yield studies published since 2005
                9.6.3.5  Additional exposure-response data	
                                                                                     _9-140
          9.6.4   Summary
                                                                                     _9-142
                                                                                      9-142
                        Table 9-17   Summary of studies of effects of ozone exposure on growth and yield of
                                    agricultural crops	9-144
- Do Not Cite or
                                                 XXI11
                                                                           September 2011

-------
                        Table 9-18   Summary of studies of effects of ozone exposure on growth of natural
                                    vegetation	9-146
    9.7   References	9-148

 10 THE ROLE OF TROPOSPHERIC OZONE IN CLIMATE CHANGE AND UV-B EFFECTS_10-1
    10.1  Introduction	  10-1
    10.2  Physics of the Earth's Radiation Budget 	  10-1
                        Figure 10-1   Diagram of the factors that determine human exposure to ultraviolet
                                    radiation. 	10-2
    10.3  Effects of Tropospheric Ozone on Climate	  10-3
          Background 	10-3
          10.3.1 Climate Change Evidence and the Influence of Tropospheric Ozone	10-4
                10.3.1.1 Climate Change in the Recent Past	10-4
                10.3.1.2 Projections of Future Climate Change	10-5
                10.3.1.3 Metrics of Potential Climate Change	10-6
                10.3.1.4 Tropospheric Ozone as a Greenhouse Gas	10-7
                        Figure 10-2  Schematic illustrating the effects of tropospheric ozone on climate.	10-8
                        Figure 10-3  Global average radiative forcing (RF) estimates and uncertainty ranges in
                                    2005 for anthropogenic CO2, CH4, ozone and other important agents and
                                    mechanisms.	10-9
          10.3.2 Factors that Influence the Effect of Tropospheric Ozone on Climate	10-9
                10.3.2.1 Trends in the Concentration of Tropospheric Ozone	10-10
                10.3.2.2 The Effect of Surface Albedo on Ozone Radiative Forcing	10-13
                10.3.2.3 The Effect of Vertical Distribution on Ozone Radiative Forcing	10-13
                10.3.2.4 Feedback Factors that Alter the Climate Response to Changes in Ozone Radiative Forcing_10-13
                10.3.2.5 Indirect Effects of Tropospheric Ozone on the Carbon Cycle	10-15
          10.3.3 Competing Effects of Ozone Precursors on Climate	10-15
          10.3.4 Calculating Radiative Forcing and Climate Response to Past Trends in Tropospheric Ozone	10-17
                        Figure 10-4  Ensemble average 1900-2000 surface temperature trends (°C per
                                    century)  in response to tropospheric ozone changes.	10-18
    10.4  UV-B Related  Effects and Tropospheric Ozone	 10-19
          10.4.1 Background 	10-19
          10.4.2 Human Exposure and Susceptibility to Ultraviolet Radiation	10-19
          10.4.3 Human Health Effects due to UV-B Radiation	10-20
          10.4.4 Ecosystem and Materials Damage Effects Due to UV-B Radiation	10-22
          10.4.5 UV-B Related Effects Associated with Changes in Tropospheric Ozone Concentrations	10-24
    10.5  Summary	 10-27
          10.5.1 Summary of the Effects of Tropospheric Ozone on Climate	10-27
          10.5.2 Summary of UV-B Related Effects on Human Health, Ecosystems, and Materials Relating to Changes in
                Tropospheric Ozone Concentrations	10-28
    10.6  References	 10-29

 11 REFERENCES                                                                                 11-1
- Do Not Cite or                                 xxiv                                      September 2011

-------
OZONE  PROJECT  TEAM
Executive Direction
                Dr. John Vandenberg (Director)—National Center for Environmental Assessment-RTP Division,
                Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle
                Park, NC

                Ms. Debra Walsh (Deputy Director)—National Center for Environmental Assessment-RTP Division,
                Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle
                Park, NC

                Dr. Mary Ross (Branch Chief)—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Scientific Staff
                Dr. James Brown (Os Team Leader)—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Christal Bowman—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Barbara Buckley—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Ms. Ye Cao—Oak Ridge Institute for Science and Education, National Center for Environmental
                Assessment, Office of Research and Development, U.S. Environmental Protection Agency,
                Research Triangle Park, NC

                Mr. Allen Davis—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Jean-Jacques Dubois—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Steven J. Dutton—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Jeffrey Herrick—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Erin Mines—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Doug Johns—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Dennis Kotchmar—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Meredith Lassiter—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Lingli Liu— Oak Ridge Institute for Science and Education, National Center for Environmental
                Assessment, Office of Research and Development, U.S. Environmental Protection Agency,
                Research Triangle Park, NC

                Dr. Thomas Long—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Thomas Luben—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Qingyu Meng— Oak Ridge Institute for Science and Education,  National Center for
                Environmental Assessment, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle Park, NC
     - Do Not Cite or                              xxv                                             2011

-------
                Dr. Kristopher Novak—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Dr. Elizabeth Oesterling  Owens—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research  Triangle Park, NC

                Dr. Molini Patel—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Dr. Joseph P. Pinto—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Ms. Joann Rice—Office  of Air Quality Planning and Standards, Office of Air and Radiation,
                U.S. Environmental Protection Agency, Research Triangle Park, NC

                Mr. Jason Sacks—National  Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Dr. Lisa Vinikoor-lmler—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC


Technical Support Staff

                Mr. Kenneth J.  Breito-Senior Environmental Employment Program, National Center for
                Environmental Assessment, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle  Park,  NC

                Mr. Gerald Gurevich—National Center for Environmental Assessment,  Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Mr. Ryan Jones—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Ms. Ellen Lorang—National Center for Environmental Assessment,  Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Mr. J. Sawyer Lucy-Student Services Authority, National Center for Environmental Assessment,
                Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle
                Park, NC

                Ms. Deborah Wales—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Mr. Richard N. Wilson-National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park,  NC

                Ms. Barbara Wright—Senior Environmental  Employment Program, National Center for
                Environmental Assessment, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle  Park,  NC
     - Do Not Cite or                              xxvi                                   September 2011

-------
AUTHORS,  CONTRIBUTORS,  AND  REVIEWERS
Authors
                Dr. James Brown (O3 Team Leader)—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Christal Bowman—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Barbara Buckley—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Maggie Clark—Department of Environmental and Radiological Health Sciences, Colorado State
                University, Fort Collins, CO

                Dr. Jean-Jacques Dubois—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Steven J.  Dutton—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Kelly Gillespie— Donald Danforth Plant Science Center, St. Louis, MO

                Dr. Terry Gordon—Department of Environmental Medicine, New York University School of
                Medicine, Tuxedo, NY

                Dr. Jeffrey Herrick—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Erin Mines—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Kazuhiko  Ito—Department of Environmental Medicine, New York University School of Medicine,
                Tuxedo, NY

                Dr. Doug Johns—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Dennis Kotchmar—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Meredith Lassiter—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Lingli Liu— Oak Ridge Institute for Science and Education, Postdoctoral Research Fellow to
                National Center for Environmental Assessment, Office of Research and Development,
                U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Thomas Long—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Thomas Luben—National Center for Environmental Assessment, Office of Research  and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Loretta J.  Mickley—School of Engineering & Applied Sciences, Harvard University, Cambridge,
                MA

                Dr. Kristopher Novak—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Elizabeth  Oesterling Owens—National Center for Environmental Assessment, Office  of
                Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Molini Patel—National  Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Jennifer Peel—Department of Environmental and Radiological Health Sciences, Colorado State
                University, Fort Collins, CO
     - Do Not Cite or Quote                         xxvii                                             2011

-------
                Dr. Joseph Pinto—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Ms. Joann Rice—on detail to the National Center for Environmental Assessment, Office of
                Research and Development, from the Office of Air Quality Planning and Standards, Office of Air
                and Radiation, U.S. Environmental Protection Agency,  Research Triangle Park, NC

                Mr. Jason Sacks—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. George Thurston—Department of Environmental Medicine, New York University School of
                Medicine, Tuxedo, NY

                Dr. Lisa Vinikoor-lmler—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. CosimaWiese—Department of Biology, Misericordia University, Dallas, PA


Contributors

                Mr. Brian Adams—Oak Ridge Institute for Science and Education, National Center for
                Environmental Assessment, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle Park, NC

                Dr. Halil Cakir—Oak Ridge Institute for Science and Education, National Center for Environmental
                Assessment, Office of Research and Development, U.S. Environmental Protection Agency,
                Research Triangle Park, NC

                Ms. Ye Cao—Oak Ridge Institute for Science and Education, National Center for Environmental
                Assessment, Office of Research and Development, U.S. Environmental Protection Agency,
                Research Triangle Park, NC

                Mr. Allen Davis—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

                Mr. Mark Evangelista—Office of Air Quality Planning and Standards, Office of Air and Radiation,
                U.S. Environmental Protection Agency, Research Triangle Park, NC

                Mr. Jay Haney—ICF International, San Rafael, CA

                Dr. E.  Henry Lee—National Health and Environmental  Effects Research Laboratory, U.S.
                Environmental Protection Agency, Corvallis, OR

                Dr. Qingyu Meng—Oak Ridge Institute for Science and Education, Postdoctoral Research Fellow to
                National Center for Environmental Assessment, Office  of Research and Development, U.S.
                Environmental Protection Agency, Research Triangle Park, NC

                Mr. David Mintz—Office of Air Quality Planning and Standards, Office of Air and Radiation, U.S.
                Environmental Protection Agency, Research Triangle Park, NC

                Mr. Tom Myers—ICF International, San Rafael, CA

                Mr. Mark Schmidt—Office of Air Quality Planning and Standards, Office of Air and Radiation, U.S.
                Environmental Protection Agency, Research Triangle Park, NC

                Dr. Huiquin Wang—School of Engineering and Applied Science, Harvard University, Cambridge,
                MA

                Mr. Benjamin Wells—Office of Air Quality Planning and Standards, Office of Air and Radiation,
                U.S. Environmental Protection Agency, Research Triangle Park, NC

                Dr. Lin Zhang—School of Engineering and Applied Science,  Harvard University, Cambridge, MA


Reviewers

                Dr. Christian Andersen—National Health and Environmental Effects Research Laboratory, U.S.
                Environmental Protection Agency, Corvallis, OR

                Ms. Lea Anderson—Office of General Counsel, U.S. Environmental Protection Agency,
                Washington, D.C.
     - Do Not Cite or Quote                        xxviii                                             2011

-------
           Dr. Susan Anenberg—Office of Air Quality Planning and Standards, U.S. Environmental Protection
           Agency, Washington, D.C.
           Dr. Robert Arnts—National Exposure Research Laboratory, Office of Research and Development,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. John Balmes— Department of Medicine, University of California, San Francisco and School of
           Public Health, University of California, Berkeley, CA
           Dr. Lisa Baxter—National Exposure Research Laboratory, Office of Research and Development,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Souad Benromdhane—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Fitzgerald Booker—USDA-ARS Plant Science Research Unit, Raleigh, NC
           Dr. Michael Breen—National Exposure Research Laboratory, Office of Research  and Development,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Philip Bromberg—School of Medicine, University of North Carolina, Chapel Hill, NC
           Dr. Kent Burkey—USDA-ARS Plant Science Research Unit, Raleigh, NC
           Dr. David DeMarini—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
           Dr. Russ Dickerson—Department of Atmospheric and Oceanic Science, University of Maryland,
           College Park, MD
           Mr. Patrick Dolwick—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Aimen Farraj—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
           Dr. Arlene Fiore—NOAA/Geophysical Dynamics Laboratory, Princeton, NJ
           Dr. Ian Gilmour—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
           Dr. Stephen Graham—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Tara Greaver—National Center for Environmental Assessment, Office of Research and
           Development, U.S. Environmental Protection Agency, Research Triangle  Park, NC
           Dr. Gary Hatch—National Health and  Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
           Dr. Bryan Hubbel—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Kristin Isaacs—National Exposure Research Laboratory, Office of Research and Development,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Scott Jenkins—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Karl Jensen—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
           Dr. Urmila Kodavanti—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
           Dr. Petros Koutrakis—Department of Environmental Health, Harvard School of Public Health,
           Boston, MA
           Mr. John Langstaff—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC
           Dr. Christopher Lau—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
- Do Not Cite or                               xxix                                    September 2011

-------
           Mr. Gary Lear—Office of Administration and Policy, Office of Air and Radiation, U.S. Environmental
           Protection Agency, Washington, DC

           Dr. Morton Lippmann—Nelson Institute of Environmental Medicine, New York University, Tuxedo,
           NY

           Dr. Karen Martin—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC

           Ms. Connie Meacham—National Center for Environmental Assessment, Office of Research and
           Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

           Mr. David Mintz—Office of Air Quality Planning and Standards, Office of Air and Radiation, U.S.
           Environmental Protection Agency,  Research Triangle Park, NC

           Dr. Pradeep Rajan—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC

           Dr. John Rogers—National Health and Environmental Effects Research Laboratory, Office of
           Research and Development, U.S.  Environmental Protection Agency, Research Triangle Park, NC

           Ms. Vicki Sandiford—Office of Air Quality Planning and Standards, Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC

           Ms. Susan Stone—Office of Air Quality Planning and Standards,  Office of Air and Radiation,
           U.S.  Environmental Protection Agency, Research Triangle Park, NC

           Dr. John Vandenberg—National Center for Environmental Assessment, Office of Research and
           Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

           Dr. James G. Wagner—Department of Pathobiology and Diagnostic Investigation, Michigan State
           University, East Lansing, Ml

           Ms. Debra Walsh—National Center for Environmental Assessment, Office  of Research and
           Development, U.S. Environmental Protection Agency, Research Triangle Park, NC

           Dr. Jason West—Department of Environmental Sciences & Engineering, University of North
           Carolina, Chapel Hill, NC
- Do Not Cite or                                xxx                                              2011

-------
CLEAN   AIR  SCIENTIFIC  ADVISORY  COMMITTEE
OZONE  NAAQS  REVIEW  PANEL


Chair of the Environmental Protection Agency's Clean Air Scientific Advisory Committee
               Dr. Jonathan M. Samet*, Department of Preventive Medicine at the Keck School of Medicine, and
               Director of the Institute for Global Health at the University of Southern California, Los Angeles, CA

Chair of the Ozone Review Panel
               Dr. Jonathan M. Samet*, Department of Preventive Medicine at the Keck School of Medicine, and
               Director of the Institute for Global Health at the University of Southern California, Los Angeles, CA

Members
               Dr. George A. Allen*, Northeast States for Coordinated Air Use Management (NESCAUM), Boston,
               MA
               Professor Ed Avol, Department of Preventive Medicine, Keck School of Medicine, University of
               Southern California, Los Angeles, CA
               Dr. John Bailar, The National Academies, Washington, D.C.
               Dr. Michelle Bell, School of Forestry & Environmental  Studies, Yale University, New Haven, CT
               Dr. Joseph Brain*,  Department of Environmental Health, Harvard School of Public Health, Harvard
               University, Boston, MA
               Dr. David Chock, Independent Consultant, Bloomfield Hills,  Ml
               Dr. William Michael Foster, Division of Pulmonary, Allergy, and Critical Care  Medicine, Duke
               University Medical Center, Durham,  NC
               Dr. H. Christopher Frey*, Department of Civil, Construction and Environmental Engineering,
               College of Engineering, North Carolina State University, Raleigh, NC
               Dr. Judith Graham, Independent Consultant, Pittsboro, NC
               Dr. David Grantz, College of Natural and Agricultural Sciences, Air Pollution  Research Center,
               University of California Riverside, Parlier, CA
               Dr. Jack Harkema, Center for Integrated  Toxicology, Michigan State University, East Lansing, Ml
               Dr. Daniel Jacob, Atmospheric Chemistry and Environmental Engineering, Harvard University,
               Cambridge, MA
               Dr. Steven Kleeberger, National Institute of Environmental Health Sciences,  National Institutes of
               Health, Research Triangle Park, NC
               Dr. Frederick J. Miller, Independent Consultant,  Gary,  NC
               Dr. Howard Neufeld,  Department of Biology, Appalachian State University, Boone, NC
               Dr. Armistead (Ted) Russell*, Department of Civil and  Environmental Engineering, Georgia Institute
               of Technology, Atlanta, GA
               Dr. Helen Suh Macintosh*, Environmental Health, NORC at the University of Chicago, and the
               School of Public Health, Harvard University, Boston, MA
               Dr. James Ultman, Department of Chemical Engineering, Pennsylvania State University, University
               Park, PA
               Dr. Sverre Vedal, Department of Environmental  and Occupational Health Sciences, School of
               Public Health and Community  Medicine,  University of Washington, Seattle, WA
               Dr. Kathleen Weathers*, Gary Institute of Ecosystem Studies,  Millbrook, NY
     - Do Not Cite or                              xxxi                                  September 2011

-------
               Dr. Peter Woodbury, Department of Crop and Soil Sciences, Cornell University, Ithaca, NY

               * Members of the statutory Clean Air Scientific Advisory Committee (CASAC) appointed by the EPA
               Administrator
Science Advisory Board Staff

               Dr. Holly Stallworth, Designated Federal Officer, U.S. Environmental Protection Agency, Mail Code
               1400R, 1300 Pennsylvania Avenue, NW, Washington, DC, 20004, Phone: 202-564-2073, Email:
               stallworth.holly@epa.gov
Draft - Do Not Cite or Quote                        xxxii                                   September 2011

-------
ACRONYMS  AND  ABBREVIATIONS
129                 mouse strain (12931/SvlmJ)
a                   alpha, ambient exposure factor
a-ATD              alpha 1-antitrypsin deficiency
a-SMA              alpha-smooth muscle actin
a-tocopherol         alpha-tocopherol
a-TOH              alpha tocopherol
a                   air exchange rate of the
                    microenvironment
A2                  climate scenario in IPCC
AADT               annual average daily traffic
A1 B                 climate scenario in I PCC
ABA                abscisic acid
ABI                 abscisic acid insensitive
A1c                 glycosylated hemoglobin blood
                    test
Ach                 acetylcholine
ACM                (Harvard University) Atmospheric
                    Chemistry Modeling (Group)
ACS                American Cancer Society
ACS-CPSII           ACS Cancer Prevention Study II
ADC                arginine decarboxylase
ADSP               Adirondack State Park, NY
AER                air exchange rate
AH2                 ascorbic acid; ascorbate
AHR                airway(s) hyperresponsiveness,
                    airway(s) hyperreactivity
AhR                 aryl hydrocarbon receptor
AHSMOG           (California Seventh Day) Adventist
                    Heath and Smog (Study)
Al                  alveolar interstitial
AIC(s)              Akaike's information criterion
AIRS                Aerometric Information Retrieval
                    System; Atmospheric Infrared
                    Sounder (instrument)
A/J                  mouse strain
Ala-9Val             genotype associated with
                    Manganese superoxide dismutase
                    (MnSOD)gene
AM                  alveolar macrophage(s)
ANF                 atrial natriureticfactor
AOT20              seasonal sum of the difference
                    between an hourly concentration
                    at the threshold value of 20 ppb,
                    minus the threshold value of 20
                    ppb
AOT30              seasonal sum of the difference
                    between an hourly concentration
                    at the threshold value of 30 ppb,
                    minus the threshold value of 30
                    ppb
AOT40              seasonal sum of the difference
                    between an hourly concentration
                    at the threshold value of 40 ppb,
                    minus the threshold value of 40
                    ppb
AOT60
AOTx

AP
A2p

APEX
APHEA(2)

APHENA

ApoB
ApoE
APX
aq
AQCD
AQI
AQS

AR
AR4

AR5

ARG

ARIC

ARIES

atm
ATP
ATPase

ATS
avg
AVHRR
B
B1
B6
BAL
BALB/c
BALF
bb
seasonal sum of the difference
between an hourly concentration
at the threshold value of 60 ppb,
minus the threshold value of 60
ppb
family of cumulative, cutoff
concentration-based exposure
indices
activated protein
climate scenario in I PCC
(preliminary version of A2)
Air Pollutants Exposure (model)
Air Pollution on Health: a
European Approach (study)
Air Pollution and Health: A
European and North American
Approach
apolipoprotein B
apolipoprotein E
ascorbate peroxidase
aqueous form: (aq)O3
Air Quality Criteria Document
Air Quality Index
(U.S. EPA) Air Quality System
(database)
acoustic rhinometry
Fourth Assessment Report (AR4)
from the IPCC
Fifth Assessment Report (AR5)
from the IPCC
arginase variants (ex., ARG1,
ARG2, ARG1h4)
Atherosclerosis Risk in
Communities
(Atlanta) Aerosol Research and
Inhalation Epidemiology Study
atmosphere
adenosine triphosphate
adenosine triphosphatase;
adenosine triphosphate synthase
American Thoracic Society
average
advanced very high resolution
radiometer
beta, beta coefficient; regression
coefficient; standardized
coefficient; shape parameter; scale
parameter
boron
climate scenario in IPCC
mouse strain (C57BL/6J)
bronchoalveolar lavage
mouse strain
bronchoalveolar lavage fluid
bronchials
     - Do Not Cite or
                                   September 2011

-------
BB
BC
B cells

B6C3F1
BDNF
BEAS-2B
BEIS

BELD

BIPM

BM
BMI
BMP
BP
BPD
bpm
Br
BRFSS

BS
BSA
Bsp, BSP
Bt, BT, bt
BTEX

BW
C

°C

C3
C3


C4


C16:0
C18:1
Ca
Ca
[Ca]
Ca2+
CA
CAA
CALINE4


CAM


CAMP
bronchial airways
black carbon
bone-marrow-derived
lymphocytes; B lymphocytes
mouse strain
brain-derived neurotrophic factor
human bronchial epithelial cell line
Biogenic Emissions Inventory
System
Biogenic Emissions Landcover
Database
International Bureau of Weights
and Measures
basement membrane
body mass index
(3 -type natriuretic peptide
blood pressure
biparietal diameter
breaths per minute
bromine
Behavioral Risk Factor
Surveillance System
black smoke
bovine serum albumin
black smoke particles
Bacillus thuringiensis', bacterium
proteins used in pesticides (or
genetically engineered plants
produce Bt toxin)
family of compounds (benzene,
toluene, ethylbenzene, and xylene)
body weight
carbon; concentration; ([vitamin] C,
ascorbate)
degrees Celsius
carbon-13 isotope
mouse strain (C3H/HEJ)
plants that use only the Calvin
cycle for fixing the carbon dioxide
from the air
plants that use the Hatch-Slack
cycle for fixing the carbon dioxide
from the air
palmitic acid (saturated fatty acid)
unsaturated fatty acid
calcium
ambient concentration
calcium concentration
calcium ion
Canada (ICD-10-CA)
Clean Air Act
California line source dispersion
model for predicting air pollutant
concentrations near roadways
plants that use crassulacean acid
metabolism for fixing the carbon
dioxide from the air
Childhood Asthma Management
Program
CAMx

CAN
CAP(s)
CAR
CASAC

CASTNET

CAT
CB

C57BL/6
C57BL/6J
CBSA
C/C
CCSP
CD
CD-1
CDC

CF
CF2

C-fibers

CFR
CGRP
CH3
CH4
C2H2
C2H4
C3H

C3H6
CHAD

CH3Br
CHs-CHO
CH3CI
CH3-CO
CHD
CHF
C2H5-H
C3H/HeJ
CH3I
CHIP
                                                           CH302'
                                                           CH3OOH
                                                           CHS
                                                           Cl
Comprehensive Air Quality Model,
with extensions
Canada
concentrated ambient particles
centriacinar region
Clean Air Scientific Advisory
Committee
Clean Air Status and Trends
Network
catalase
carbon black; CMAQ mechanisms
(ex., CB04, CB05, CB06)
mouse strain
mouse strain
core-based statistical area
carbon of total carbon
Clara cell secretory protein
cluster of differentiation (various
receptors on T-cells: CD8+, CD44,
etc.); criteria document (see
AQCD)
mouse strain
Centers for Disease Control and
Prevention
charcoal-filtered; carbon filtered air
twice-filtered air (particulate filter
and activated charcoal filter)
afferent, slow, unmylenated nerves
innervating the respiratory system
Code of Federal Regulations
calcitonin gene-related peptide
methyl group
methane
acetylene
ethylene
mouse strain (C3H/HEJ or
C3H/OuJ)
propylene
Consolidated Human Activity
Database
methyl bromide
acetaldehyde
methyl chloride
acetyl radical(s)
coronary heart disease
congestive heart failure
ethane
mouse strain
methyl iodide
Effects of Elevated Carbon Dioxide
and Ozone on Potato Tuber
Quality in the European Multiple
Site Experiment
methyl peroxy (radical)
acetic acid; methyl hydroperoxide
Child Health Study
confidence interval(s)
     - Do Not Cite or
                                                                           September 2011

-------
Cl
cr
CI2
CLE

CLM
CINO2
cm
cm2
CM
CMAQ

CN


CNA
CMS
CO
CO2
COD

Col-0
COP

COPD

COX-2
C-R
CRA
CRP
CS
CSA

csb, Csb

CSF
CST
CSTR
CSV

CT
CTM(s)
cum avg
CUOt


CV, C.V.
cv, c.v.
CVD
airborne O3 concentration at
microenvironmentj
chlorine
chlorine ion
chlorine gas
Current Legislation (climate
scenario in IPCC)
chemiluminescence method
nitryl chloride
centimeter(s)
square centimeters
Clinical Modification (ICD-9-CM)
Community Multi-scale Air Quality
modeling system
constant atmospheric nitrogen
deposition (in PnET-CN
ecosystem model)
continental North America
central nervous system
carbon monoxide; Cardiac output
carbon dioxide
coefficient of divergence;
coefficient of determination
(Arabidopsis ecotype) Columbia-0
Conference of Parties (to the
UNFCCC)
chronic obstructive pulmonary
disease
cyclooxygenase 2 enzyme
concentration-response
Centra di ricerca per la
cerealicoltura (CRA) [The Centre
for Cereal Research] - Unit 5: The
Research Unit for Cropping
Systems in Dry Environments in
Bari, Italy (water-stressed
conditions)
C-reactive protein
corticosteroid
cross-sectional area; combined
statistical area
cockayne syndrome (cb)
gene/protein group A
colony-stimulating factor
central standard time
continuous stirred tank reactor
comma-separated values (a
spreadsheet format)
computer tomography
chemical transport model(s)
cumulative average
The cumulative stomatal uptake of
O3, using a constant O3 uptake
rate threshold (t) of nmol/m2/s
coefficient of variation
cultivar
cardiovascular disease
CXC


CXCR2

CXR
CyS
Cys-LT

cyt
A, 6
AFEV,
AVD

2-D
3-D
DAMPS

DBP
DC(s)
DDM
DEP(s)
df
DGGE

DMA
DHAR
DHBA
DLEM
dm3
DMA
DO AS

DOC
DR

df

DTH
DU
DW
E
EEC
EC
ECE

ECG
ECOPHYS
ED
chemokine family of cytokines,
with highly conserved motif:cys-
xxx-cys (CXC) amino acid group
CXC chemokine receptor 2
(CXCR2)
Chest (x-ray) radiograph(s)
protein cysteines
cysteinyl leukotrienes (LTC4, LTD4,
LTE4)
cytosolic-free
delta, difference; change
change in FEVi
change in dead space volume of
the respiratory tract
two-dimensional
three-dimensional
3-deoxy-D-arabino-heptulosonat-
7-phosphate synthase
diastolic blood pressure
dendritic cell(s)
direct decoupled method
diesel exhaust particle(s)
degrees of freedom
denaturing gradient gel
electrophoresis
dehydroascorbate
dehydroascorbate reductase
2,3-dihydroxybenzoic acid
Dynamic Land Ecosystem Model
cubic decimeter(s)
deoxyribonucleic acid
differential optical absorption
spectroscopy
dissolved organic carbon
type of human leukocyte antigens
(HLA-DR)
Portion of time-period spent in
microenvironment/
delayed-type hypersensitivity
Dobson unit(s)
dry weight
embryonic day (ex., E15,  E16,
etc);  [vitamin]  E
exposure to pollutant of ambient
origin
exhaled breath condensate (fluid)
elemental carbon
endothelin converting enzyme(s)
[i.e.,  ECE-1]
electrocardiogram
physiological process modeling to
predict the response of aspen
forest ecosystems (modeling
growth and environmental stress in
Populus)
emergency department; embryonic
day (ex., EDS, ED20)
     - Do Not Cite or
                                                                          September 2011

-------
EGEA



EGEA2

EHC-93


ELF
EMI

Ena

ENA-78

eNO
eNOS
EN VI SAT
EOTCP

EP
EPA

EPIC


ER
ESA
ET

ET,

ET2


ETS
EU
EUS
0
0PSII-max

f
F
F344
F2a

FA
FACE
FACES

fe
FC
FEF
FEF25.75


FEFx
(The) Epidemiology (study on)
Genetics and Environment of
Asthma, (adults and children with
asthma)
follow-up study on EGEA (adults
with asthma only)
ambient PM reference sample
(urban dust [air particles] collected
in Ottawa Canada)
extracellular lining fluid
(U.S.  EPA) Exposure  Model for
Individuals
exposure to pollutant of
nonambient origin
epithelial cell-derived neutrophil-
activating peptide 78
exhaled nitric oxide
endothelial nitric oxide synthase
(EAS) Earth Observation satellite
European Open Top Chamber
Programme
epithelial cells
U.S.  Environmental Protection
Agency
European Prospective
Investigation into Cancer and
Nutrition
emergency room
European Space Agency
extrathoracic; endothelin (i.e.  ET-
1)
anterior nasal passages within the
extrathoracic (ET) region
oral airway and posterior nasal
passages within the extrathoracic
(ET)  region
environmental tobacco smoke
European Union
eastern U.S.
Phi; calculated efficiency
maximum photochemical effective
quantum yield of PSII
Fraction of the relevant time period
female
Fischer 344 (rat strain)
8-isoprostane (major F2
prostaglandin [8 iso-PGF2a])
filtered air
free-air-CO2 enrichment (system)
Fresno Asthmatic Children's
Environment Study
frequency of breathing
fibrocartilaginous coat
forced expiratory flow
forced expiratory flow between the
times at which 25% and 75%  of
the vital capacity is reached
forced expiratory flow after (x)%
vital capacity (e.g., after 25, 50, or
75%  vital capacity)
FEM
FeNO
FEV,

FHM

FIA

Finf
Finf,/

FLAG

FLRT

l~nose
FPM
FR
FRAP
FRC
FRM
FRT

FstO-i
FURT

FVC
Fv/Fm

FVI
Y
Y-TOH
g, mg, kg, |jg, ng, pg
g
GAM
GCLC


GCLM


G-CSF

GD
GEE
GEOS

GEOS5
GEOS-Chem

GFAP
Federal equivalent method
exhaled nitric oxide fraction
forced expiratory volume in 1
second
(USDA Forest Service) Forest
Health  Monitoring Program
(USDA Forest Service) Forest
Inventory and Analysis Program
infiltration factor
infiltration factor for indoor
environment (i)
Federal land managers' air quality
related values workgroup
fractional uptake efficiency of the
lower respiratory tract (LRT)
fractional uptake efficiency via
nasal absorption
fraction of time spent in outdoor
microenvironments
Forest Pest Management
Federal Register
ferric reducing ability of plasma
functional residual capacity
Federal reference method
fractional uptake efficiency of the
respiratory tract (RT)
flux cut off threshold
fractional uptake efficiency of the
upper respiratory tract (URT)
forced vital capacity
a ratio: a measure of the maximum
efficiency of Photosystem II
fruits and vegetables index
gamma
gamma-tocopherol
gram(s), milligram(s), kilogram(s),
microgram(s), nanogram(s),
picogram(s)
granulocyte; guanosine
gram(s); gaseous form: (g)Os
generalized additive model(s)
conductance through boundary
layer and stomata
(glutathione genetic variant)
glutamate-cysteine ligase catalytic
subunit
(glutathione genetic variant)
glutamate-cysteine ligase modifier
subunit
granulocyte colony-stimulating
factor (receptor)
gestational day
generalized estimating equations
(NASA) Goddard Earth Observing
System model
GEOS version 5
GEOS-Chemistry (tropospheric
model)
glial fibrillary acidic protein
     - Do Not Cite or
                                                                           September 2011

-------
GH
GHG
GLM(s)
GMAO

GM-CSF

GOME

GOMOS
G6P
G6PD

GPP
G-proteins
GPT
GR
GSH
GSO37GSO32"
GSR
GSS
GSSG
GST
GSTM1
GSTP1

GTP
GTPases
GWP
GxE
h
h/day
H; H+; H-

3H
H2
ha
HA
HA(s)
Hb
HbA1c

HC(s)
HCFC(s)
HCHO
H2CO
HCO-
HDM
2HDM
HDMA
3He
growth hormone
greenhouse gas
generalized linear model(s)
(NASA) Global Modeling and
Assimilation Office
granulocyte macrophage colony-
stimulating factor
(ESA) Global Ozone Monitoring
Experiment (spectrometer)
Global Ozone Monitoring by
Occultation of Stars (ESA
ENVISAT spectrometer measuring
long-term trends in O3)
glucose-6-phosphate
glucose-6-phosphate
dehydrogenase
gross primary production
GTPases
gas phase titration
glutathione reductase
glutathione; reduced glutathione
guanine sulfonates
glutathione reductase
glutathione synthetase
glutathione disulfide
glutathione S-transferase
glutathione S-transferase
polymorphism M1 genotypes
(GSTM1-null, -GSTM1-sufficient)
glutathione S-transferase
polymorphism P1 genotypes
guanosine triphosphate
G-proteins/enzymes
global warming potential
gene-environmental interaction
hour(s)
hour(s) per day
atomic hydrogen, hydrogen ion;
hydrogen radical
radiolabeled hydrogen; tritium
molecular hydrogen
hectare
hyaluronic acid
hospital admission(s)
hemoglobin
glycosylated hemoglobin (blood
test)
hydro carbon(s)
hydro chlorofluorocarbon(s)
formaldehyde
formaldehyde
formyl (radical)
house dust mite
second-highest daily maximum
house dust mite allergen
non-radioactive isotope of helium
HeJ

HEPA
HERO


12-HETE
HF

MFCs
Hg
HHP-C9
HIST
HLA
HLA-DR

HMOX
HMOX-1

HNE
HNO2
HNO3
HNO4
HO
HO-
HO-1
HO2-

HO3-
H2O
H202
H30+
HOCH2OOH
HONO
HO2NO2
HOONO
HOX
hPa
HPLC

HPOT
HR
HRrnax
HRP
HRV
HSC
hs-CRP
H2S04
HSP

HSP70
HSS

5-HT
hv
O3-resistant C3H mouse strain
(C3H/HeJ)
high efficiency particle air (filter)
Health and Environmental
Research Online, NCEA Database
System
12-Hydroxyeicosatetraenoic acid
(HRV signal) high-frequency
power
hydrofluoro carbons
mercury
1 -hydroxy-1 -hydro peroxynonane
histamine
human leukocyte antigen
human leukocyte antigen receptor
genes
Heme oxygenase
heme-oxygenase-1
(polymorphism)
4-hydroxynonenal
nitrous acid
nitric acid
pernitric acid
hydroxyl; heme oxygenase
hydroxyl radical
heme oxygenase 1
hydroperoxyl; hydroperoxy radical;
protonated superoxide
protonated ozone radical
water
hydrogen peroxide
hydronium ion
hydroxymethylhydro peroxide
nitrous acid
peroxynitric acid
pernitrous acid
hydrogen radical(s)
hecto pascal
high-pressure liquid
chromatography
13-hydroperoxide linolenic acid
heart rate, hazard ratio
maximum heart rate
horseradish peroxidase
heart rate variability
Houston Ship Channel (Texas)
high-sensitivity C-reactive protein
sulfuric acid
high speed pellet (after centrifuge
spin)
heat shock protein 70
high speed supernatant (after
centrifuge spin)
5-hydroxytryptamine
Energy per photon of
electromagnetic energy at
frequency v
     - Do Not Cite or
                                                                          September 2011

-------
HVAC
Hz
I
IARC
IAS
IBM

1C

ICAM-1
ICARTT


I CAS
ICC
ICD


ICD-9

ICD-10

ICEM

ICNIRP

ICP Forests


ICU
ICVE
IDW
IFN
IFN-Y
ig
igA
igE
IGF-1
igG
igM
IHD
IL

IL-1(3
lie
i.m.
IMPACT


IMPROVE

IN
INF
inh
iNKT

iNOS
heating, ventilation, and air
conditioning
hertz
iodine
International Agency for Research
on Cancer
interalveolar septum
individual-based model or
modeling
inspiratory capacity; intracloud
(lightning flash)
intercellular adhesion molecule 1
International Consortium for
Atmospheric Research on
Transport and Transformation
Inner City Asthma Study
intraclass correlation coefficient
implantable cardioverter
defibrillator(s);  International
Classification of Diseases
International Classification of
Disease 9th revision
International Classification of
Disease 10th revision
Indoor Chemistry and Exposure
Model
International Commission on Non-
Ionizing Radiation Protection
International Cooperative
Programme on Assessment of Air
Pollution Effects on Forests
Intensive Care Unit
ischemic cerebrovascular events
inverse-distance-weighted
interferon (e.g., IFN-Q)
interferon-gamma
immunoglobulin (e.g., IgE)
immunoglobulin A
immunoglobulin E
insulin-like growth factor 1
immunoglobulin G
immunoglobulin M
ischemic heart disease
interleukin (e.g., IL-2, IL-4, IL-6,  IL-
8, etc.)
interleukin-1(3
isoleucine
intramuscular (route)
Interactive Modeling Project for
Atmospheric Chemistry and
Transport
Interagency Monitoring of
Protected Visual Environment
intranasal
interferon
inhalation
invariant (type  I) natural killer T-
cell
inducible nitric oxide synthase
INRA                National agronomical research
                     institute (INRA) in Thiverval-
                     Grignon. France (adequately-
                     watered conditions)
INTRASTAND        a stand-level model designed for
                     hourly, daily and annual integration
                     of forest carbon and water cycle
                     fluxes
I/O                  indoor-outdoor ratio
IOM                 Institute of Medicine
i.p.                  intraperitoneal (route)
IPCC                Intergovernmental Panel on
                     Climate Change
IPCC-A2             Intergovernmental Panel on
                     Climate Change 2nd Assessment
                     Report
IPCC-AR4            Intergovernmental Panel on
                     Climate Change 4th Assessment
                     Report
IPCC-AR5            Intergovernmental Panel on
                     Climate Change 5th Assessment
                     Report
IPCC-TAR            Intergovernmental Panel on
                     Climate Change Third Assessment
                     Report
IPMMI               International  Photolysis Frequency
                     Measurement and Modeling Inter-
                     comparison
IQR                 interquartile range
IR                   infrared
I/R                  ischemia-reperfusion
IRIS                 Integrated Risk Information
                     System
IRP                 Integrated Review Plan for the
                     Ozone National Ambient Air
                     Quality Standards
ISA                 Integrated Science Assessment
ISCCP               International  Satellite Cloud
                     Climatology Project
ISO                 International  Standards
                     Organization
8-iso-PGF            8-isoprostane
IT                   intratracheal
IU                   International  Units
IUGR                intrauterine growth restriction
i.v.                  intravenous (route)
IVF                  in vitro fertilization
j                     Microenvironment
JA                  jasmonic acid
Jmax                maximum rate of electron transport
                     (for regeneration of RuBP)
JNK                 jun N-terminal kinase
JPL                 Jet Propulsion Laboratory
K                    kappa
KB                  kappa B
k                    dissociation rate;  rootshoot
                     allometric coefficient; rate of O3
                     loss in the microenvironment
K                    potassium
K+                   potassium ion
     - Do Not Cite or
                                                                            September 2011

-------
Ka                  intrinsic mass transfer
                    coefficient/parameter
KC                 keratinocyte-derived chemokine
kg                  kilogram
Kg                  mass transfer coefficient for gas
                    phase
kHz                 kilohertz
kJ                  kilojoules
Kl                  mass transfer coefficient for liquid
                    phase
km                 kilometer
KM                 particle optical reflectance
KML                keyhole markup language
KMZ                zipped  KML computer language
KO                 knockout
Kr                  reaction rate constant
KROFEX            Krauzberg Ozone Fumigation
                    Experiment
L, dl_, ml, |jl_         Liter, deciliter,  milliLiter, microLiter
LO                  Lag (e.x., Lag 0, Lag 1, etc.)
LAI                 leaf area index
LBL                 Lawrence Berkeley Laboratory
LBLX               Lawrence Berkeley Laboratory
                    model including airflow from
                    natural ventilation
Lb(s)                pound(s)
LEW                low birth weight
LC50                median lethal concentration
LCL                 lower 95th% confidence limit
LDH                lactate  dehydrogenase
LDL                 low-density lipoprotein ; lower
                    detectable level
LF                   (HRV signal) low-frequency power
LFHFR              low frequency/high frequency
                    (ratio)
LFT                 lower free troposphere
LI                  labeling index
LIDAR               Light Detection and Ranging
                    (remote sensing system)
LIF                 laser-induced fluorescence
LINKAGES           individual-based model of forest
                    succession
LIS                 lateral intercellular space
LLJ                 low-level jet
L/min               liters per minute
Ln                  Natural logarithm
LnRMSSD           natural log of RMSSD; measure of
                    HRV
InSDNN             natural log of the standard
                    deviation of NN intervals in an
                    EKG
LOAEL              lowest observed adverse  effect
                    level
LOD                limit of detection
LOEL               lowest-observed-effect level
LOESS              locally weighted scatterplot
                    smoothing
LOP                lipid ozonation products
LOSU
LOWESS

LOX-1

LPS
LRS
LRT

LSI
LT

LT-a
LTA
LUR
LVEDD

LVEDP

LWC
M
|jeq
M9
|jg/m3
|jm
m, cm, |jm, nm


M
M, mM, |jM, nM, pM

m2
m3
M#

M2
M7
M12
ma
mAOT

MAP

MAPK

MAQSIP

MARAT

MARCO

max
MBL
MCA
MGCP

Mch; MCh
MCM
level of scientific understanding
locally weighted scatter plot
smoother
Lipoxygenase; lectin-like oxidized
low density lipoprotein receptor-1
lipopolysaccharide
lower respiratory symptoms
lower respiratory tract; lower
airways; Long range transport
local standard time
leukotriene (e.g., LTB4 ,  LTC 4,
LTD4 ,  LTE4); local time
lymphotoxin-a
lymphotoxin-alpha
land use regression
left ventricular chamber
dimensions at end diastole
left ventricular end diastolic
pressure
liquid water content
mu, micro
microequivalent
microgram
micrograms per cubic meter
micrometer, micron
meter(s), centimeter(s),
micrometer/[micron](s),
nanometer(s)
male
Molar, milliMolar, microMolar,
nanoMolar, picoMolar
square meters
cubic meters
Month (M1 Monthl; M2 Month2;
M3 MonthS; M4 Month4)
type of muscarinic receptor
7-hour seasonal mean
12-hour seasonal mean of O3
moving average
modified accumulated exposure
over threshold
mitogen-activated protein; mean
arterial pressure
mitogen-activated protein
kinase(s), MAP kinase
Multiscale Air Quality Simulation
Platform (model)
Mid-Atlantic Regional Assessment
Team
Macrophage  receptor with
collagenous structure
maximum
marine boundary layer
minimum cross-sectional area
Mountain Cloud Chemistry
Program
methacholine
master chemical mechanism
     - Do Not Cite or
                                   September 2011

-------
MCP-1
MDA
MDAR
MDI
MDL
MED
MEF50%

MEGAN

MeJA
MENTOR

METs

MFR
Mg
MGDG
mg/m3
MHC
mi
Ml

MIESR

min
MIP
MIP-2

ml
mL/min
MLN
Mm
mm
MM Ml
MM5


MM AD


MMEF
mmHg
MMMD

MMP-2
MMP-3
MMP-9
MMSP
Mn
M/N

MnSOD
mo
MOA(s)
MOBILE
monocyte chemotactic protein 1
malondialdehyde
monodehydroascorbate reductase
Mediterranean diet index
minimum detection level
minimal erythema dose
maximal midexpiratory flow at 50%
of forced vital capacity
model of emissions of gases and
aerosols from nature
methyl jasmonate
Modeling Environment for Total
Risk Studies
metabolic equivalent unit(s) [of
work]
Maximum Feasible Reduction
magnesium
monogalactosyldiacylglycerol
milligrams per cubic meter
major histocompatibility complex
mile(s)
myocardial infarction, "heart
attack"
matrix isolation electron spin
resonance (spectroscopy)
minute; minimum
macrophage inflammatory  protein
macrophage inflammatory  protein-
2
milliliter
milliliter(s) per minute
mediastinal lymph node
megameter
millimeter(s)
Mt.  Mitchell site
National Center for Atmospheric
Research/Penn State Mesoscale
Model (version 5)
mass median aerodynamic
diameter; mass median
aerodynamic density
maximal midexpiratory flow
millimeters of mercury
mean  maximum mixing height
depth
matrix metalloproteinase-2
matrix metalloproteinase-3
metalloproteinase-9
Mount Mitchell State  Park,  NC
manganese
pooled data from mouth and nasal
exposure
Manganese superoxide dismutase
mo nth (s)
mode(s) of Action
(U.S. EPA) mobile vehicle
emission factor model (on-road
vehicles)
MOBILES


MODNR

MONICA


MoOx
MOSES

MOVES



MOZAIC

MOZART

MPAN

MPO
MQL
MRI


mRNA
ms
MS

MSA

MSL
MS/MS
MT
MT, Mt
MT1
MTBE
mtDNA
Mtn
MV
MW
MyD88

n, N
N

15N

N2

Na
NA
NA; N/A
Na+
NAAQS

NAD
vehicle emissions modeling
software version 6; replaced by
MOVES
Missouri Department of Natural
Resources
Monitoring of Trends and
Determinants in Cardiovascular
Disease
molybdenum oxides
Met Office Surface Exchange
Scheme
Motor Vehicle Emission Simulator
(replaced MOBILES; for estimating
emissions from cars, trucks, and
motorcycles
Measurement of Ozone and Water
Vapor by Airbus In-Service Aircraft
Model for Ozone and Related
chemical Tracers
peroxymethacryloyl nitrate;
peroxy-methacrylic nitric anhydride
myeloperoxidase
Minimum quantification limit
magnetic resonance imaging;
Midwest Research Institute;
Meteorological Research Institute
messenger RNA
millisecond(s)
mass spectrometry; Mt.
Moosilauke site
Metropolitan Statistical Area;
methane sulfonic acid
mean sea level
tandem mass spectrometry
million ton(s); metric ton(s)
metallothionein
mitochondria
methyl-tertiary-butyl ether
mitochondrial DNA
mountain
minute volume
molecular weight
myeloid differentiation primary
response gene 88
number; number of observations
nitrogen; North; nasal exposure by
natural breathing
nitrogen-15, stable isotope of
nitrogen
molecular nitrogen; nonreactive
nitrogen
sodium
noradrenaline;  North American
not available; not applicable
sodium ion
National Ambient Air Quality
Standards
nicotinamide adenine nucleotide
     - Do Not Cite or
                                                                          September 2011

-------
NADH               reduced nicotinamide adenine
                    dinucleotide; nicotinamide adenine
                    dinucleotide dehydrogenase
NADP               National Atmospheric Deposition
                    Program
NADPH             reduced nicotinamide adenine
                    dinucleotide phosphate
NADPH-CR          reduced nicotinamide adenine
                    dinucleotide phosphate -
                    cytochrome c reductase
NaE                sodium erythorbate
NAG                N-acetyl-glucosaminidase
Na-K-ATPase        sodium-potassium-dependent
                    adenosine triphosphatase
NAMS               National Ambient Monitoring
                    Stations
NAPAP             National Acid Precipitation
                    Assessment Program
NAPBN             National Air Pollution Background
                    Network
NARE               North Atlantic Regional
                    Experiment
NARSTO            North American Regional Strategy
                    for Tropospheric Ozone
NAS                National Academy of Sciences;
                    Normative Aging Study
NASA               National Aeronautics and Space
                    Administration
NBS                National Bureau of Standards
NBTH               3-methyl-2-benzothiazolinone
                    acetone azine
NCEA               National Center for Environmental
                    Assessment
NCEA-RTP          NCEA Division in Research
                    Triangle Park, NC
NCHS               National Center for Health
                    Statistics
NCICAS             National Cooperative Inner-City
                    Asthma Study
NCLAN             National Crop Loss Assessment
                    Network
NCore               National Core multi-pollutant
                    monitoring  network
NC-R               resistant clones of white clover
NC-S               sensitive clones of white clover
ND; n.d.             not detectable; not detected; no
                    data
2ndHDM             second-highest daily maximum
NDF                neutral detergent fiber
NEE                net ecosystem CO2 exchange
NEI                 National Emissions Inventory
NEM                National Ambient Air Quality
                    Standards Exposure Model
NEP                Net Ecosystem Production
NERL               National Exposure Research
                    Laboratory
NESCAUM          Northeast States for Coordinated
                    Air Use Management
NF                  National Forest; non-filtered air
NF-KB               nuclear factor kappa B
ng
NGF
NH
NH3
NH4+
NH4HSO4
(NH4)2HSO4
NHANES

NHANESIII

NHAPS

NHEERL


NHIS
(NH4)2S04
NIH
NIST

NK
NKT
NL
NLF
NM
NMHC(s)
NMMAPS

NMOC(s)
NMVOCs

NN


NNK

nNOS

NO
•NO

NO2
NO3; NO3-
NO3"
N2O
N205
NOAA

NOAEL
NOS

NOX

NOY
nanogram(s)
nerve growth factor
northern hemisphere
ammonia
ammonium ion
ammonium bisulfate
ammonium sulfate
National Health and Nutrition
Examination Survey
National Health and Nutrition
Examination Survey III
National Human Activity Pattern
Survey
(U.S. EPA) National Health and
Environmental Effects Research
Laboratory
National Health Interview Survey
ammonium sulfate
National Institutes of Health
National Institute of Standards and
Technology
natural killer cells; neurokinin
natural killer T cells
nasal lavage
nasal lavage fluid
National Monument
nonmethane hydrocarbon(s)
National Morbidity, Mortality, and
Air Pollution Study
nonmethane organic compound(s)
nonmethane volatile organic
compounds
normal-to-normal (NN or RR) time
interval between each QRS
complex in the EKG
4-(N-nitrosomethylamino)-1 -(3-
pyridyl)-1-butanone
neuronal nitric oxide synthase
(NOS)
nitric oxide
nitric oxide concentration
(interpunct NO)
nitrogen dioxide
nitrate, nitrate radical
nitrate, nitrate ion
nitrous oxide
dinitrogen pentoxide
National Oceanic and Atmospheric
Administration
no observed adverse effect level
nitric oxide synthase (types, NOS-
1, NOS-2, NOS-3)
nitrogen oxides, oxides of nitrogen
(NO + NO2)
sum of NOX and NOZ; odd
nitrogen species; total oxidized
nitrogen
     - Do Not Cite or
                                  September 2011

-------
NOZ
NP
NPP
NPS

NQO1

NQOIwt

NR
Nr
NRC
Nrf-2

Nrf2-ARE

NS; n.s.

NSAID

NSBR

NSF
NTE
NTN
NTP
NTRMs

NTS

NWR
NWS
NZW
O
18Q

02
02-
02-
102
03
1803
03*
OAQPS

OAR
OEMs
OC
OD
0(1D)
OH, OH-
8-OHdG
OLS
OMI
sum of all inorganic and organic
reaction products of NOX (HONO,
HNO3, HNO4, organic nitrates,
particulate nitrate, nitro-PAHs,
etc.)
National Park
net primary production
National Park Service, U.S.
Department of the Interior
NAD(P)H-quinone oxidoreductase
(genotype)
NAD(P)H-quinone oxidoreductase
wild type (genotype)
not reported
reactive nitrogen
National Research Council
nuclear factor erythroid 2-related
factor 2
NF-E2-related factor 2-antioxidant
response element
nonsignificant; non-smoker;
national seashore; natural spline
non-steroidal anti-inflammatory
agent
nonspecific bronchial
responsiveness
National Science Foundation
nasal  turbinate epithelial (cells)
National Trends Network
National Toxicology Program
NIST Traceable Reference
Materials
nucleus of the solitary tract (in
brainstem)
national wildlife refuge
National Weather Service
New Zealand white (rabbit)
oxygen; horizon forest floor
oxygen-18,  stable isotope of
oxygen
molecular oxygen
superoxide
superoxide  radical
singlet oxygen
ozone
(oxygen-18 labeled) ozone
electronically excited ozone
Office of Air Quality Planning and
Standards
Office of Air and Radiation
observationally based methods
organic carbon
outer diameter; optical density
electronically excited oxygen atom
hydroxyl group, hydroxyl radical
8-hydroxy-2'-deoxyguanosine
ordinary least squares
Ozone Monitoring Instrument
ON
ONOO"
0(3P)
OPE
OPECs

OR
ORD

OSHA

OTC
OuJ

OVA
OX
OxComp

oz
P
P
P450
p53
P90

PACF

PAD

PAF

PAH(s)
PAI-1

PAL
PAMS

PAN
PaO2
PAPA

PAR

Palm
p-ATP
Pb
PEL

PBM

PEN
PBPK

PBS
PC
Ontario
peroxynitrate ion
ground-state oxygen atom
ozone production efficiency
Outdoor Plant Environment
Chambers
odds ratio
Office of Research and
Development
Occupational Safety and Health
Administration
open-top chamber
O3-sensitive C3H mouse strain
(C3H/OuJ)
ovalbumin
odd oxygen species; total oxidants
oxidative capacity of the
atmosphere
ounce(s)
pressure in atmospheres; plants
grown in pots; phosphorus;
penetration fraction of O3 into the
microenvironment; pulmonary
region
probability value
cytochrome P450
cell cycle protein gene
90th percentile of the absolute
difference in concentrations
partial autocorrelation function of
the model residuals
peripheral arterial disease;
pollutant-applied dose
platelet-activating factor;
paroxysmal atrial fibrillation
polycyclic aromatic hydrocarbon(s)
plasminogen activator fibrinogen
inhibitor-1
phenylalanine ammonia lyase
Photochemical Assessment
Monitoring Stations network
peroxyacetyl nitrate
arterial oxygen pressure
Public Health and Air Pollution in
Asia
photosynthetically active radiation;
proximal alveolar region
Pressure in atmospheres
para-acetamidophenol
Lead
planetary boundary layer;
peripheral blood lymphocytes
population-based model or
modeling
C-phenyl N-tert-butyl nitrone
physiologically based
pharmacokinetic (model)
phosphate buffered saline
phosphatidylchloline
     - Do Not Cite or
                                                                           September 2011

-------
PC2o



PC2oFEVi

PC50
PCA
PC-ALF

PCD
PCI
pCNEM


PCO2

pCO2
PCR
PCR-DGGE

PD
PD20
PD10
PDI
PE

PEF
PEFojs

PEFR
PEFT
PEG-CAT
PEG-SOD

PEM(s)
Penh
PEPc
PFD
PFT
pg
PG

6PGD

PGE2
PGF2a
PGHS-2

PGP
               provocative concentration that
               produces a 20% decrease in
               forced expiratory volume in 1
               second
               provovative concentration that
               produces a 20% decrease in
               provocative concentration that
               produces a 50% decrease in
               forced expiratory volume in 1
               second
               principal component analysis
               1 -palmitoyl-2-(9-oxonononoyl)-sn-
               glycero-3-phosphocholine
               programmed cell death
               picryl chloride
               Canadian version of National
               Ambient Air Quality Standards
               Exposure Model
               Average partial pressure of O2 in
               lung capillaries
               partial pressure of carbon dioxide
               polymerase chain reaction
               PCR-denaturing gradient gel
               electrophoresis
               pregnancy day
               provocative dose that produces a
               20% decrease in
               provocative dose that produces a
               20% decrease in
               provocative dose that produces a
               100% increase in sRAW
               provocative dose that produces a
               100% increase in SRaw
               pain on deep inspiration
               post exposure,
               phosphatidylethanolamine
               peak expiratory flow
               peak expiratory flow in 0.75
               second
               peak expiratory flow rate
               time to peak flow
               polyethylene glycol-catalase
               polyethylene glycol-superoxide
               dismutase
               personal exposure monitor(s)
               enhanced pause
               phosphoenolpyruvate carboxylase
               photosynthetic flux density
               pulmonary function test
               picogram(s)
               prostaglandin  (e.g.,  PGE2 ,PGF2);
               phosphatidylglycerol
               6-phosphogluconate
               dehydrogenase
               prostaglandin  E2
               prostaglandin  F2-alpha
               prostaglandin  endoperoxide G/H
               synthase 2
               protein gene product (e.g.,
               PGP9.5)
PGSM

pH


PHA
PI

PIF
PiZZ
PK
pKa
PLFA
PM

PMx
                                                      PM2
Plant Growth Stress Model
relative acidity; Log of the
reciprocal of the hydrogen ion
concentration
phytohemagglutinin A
phosphatidylinositol; probability
interval; posterior interval
peak inspiratory flow
respiratory phenotype
pharmaco kinetics
dissociation constant
phospholipid fatty acid
particulate matter
Particulate matter of a specific size
range not defined for regulatory
use.  Usually X refers to the 50%
cut point, the aerodynamic
diameter at which the sampler
collects 50% of the particles and
rejects 50% of the particles. The
collection efficiency, given by a
penetration curve, increases for
particles with smaller diameters
and decreases for particles with
larger diameters. The definition of
PMx is sometimes abbreviated as
"particles with a  nominal
aerodynamic diameter less than or
equal to X |jm" although X is
usually a 50% cut point.
In general terms, particulate matter
with an aerodynamic diameter less
than or equal to  a nominal 2.5 |jm;
a measurement of fine particles in
regulatory terms, particles with an
upper 50% cut-point of 2.5 |jm
aerodynamic diameter (the 50%
cut point diameter is the diameter
at which the sampler collects 50%
of the particles and rejects 50% of
the particles) and a  penetration
curve as measured  by a reference
method based on Appendix L of 40
CFR Part 50 and designated in
accordance with 40 CFR Part 53,
by an equivalent method
designated in accordance with 40
CFR Part 53, or by an approved
regional method designated in
accordance with Appendix C of 40
CFR Part 58.
- Do Not Cite or
                                                                                               September 2011

-------
PM10
PM10-2.5
PM10,
pSSMAPK

PM-CAMx



PMN(s)
PMT
PND
pNEM

PnET

PNN
In general terms, participate matter
with an aerodynamic diameter less
than or equal to a nominal 10 |jm;
a measurement of thoracic
particles (i.e., that subset of
inhalable particles thought small
enough to penetrate beyond the
larynx into the thoracic region of
the respiratory tract) in regulatory
terms, particles with an upper 50%
cut-point of 10± 0.5 |jm
aerodynamic diameter (the 50%
cut point diameter is the diameter
at which the sampler collects 50%
of the particles and rejects 50% of
the particles) and a penetration
curve as measured by a reference
method based on Appendix J of 40
CFR Part 50 and designated in
accordance with 40 CFR Part 53
or by an equivalent method
designated  in accordance with 40
CFR Part 53.
In general terms, particulate matter
with an aerodynamic diameter less
than or equal to a nominal 10 |jm
and greater than a nominal 2.5
|jm; a measurement of thoracic
coarse particulate matter or the
coarse fraction of PM10 in
regulatory terms, particles with an
upper 50% cut-point of 10 |jm
aerodynamic diameter and a lower
50% cut-point of 2.5 |jm
aerodynamic diameter (the 50%
cut point diameter is the diameter
at which the sampler collects 50%
of the particles and rejects 50% of
the particles) as measured  by a
reference method based on
Appendix O of 40 CFR Part 50 and
designated  in accordance with 40
CFR Part 53 or by an  equivalent
method designated in accordance
with 40 CFR Part 53.
The PMio-2.5 concentration of PMio-
2.5 measured by the 40 CFR Part
50 Appendix O reference method
which consists of currently
operated, co-located low-volume
(16.7Lpm)PM10andPM2.5
reference method samplers.
p38 mitogen-activated protein
kinase(s)
Comprehensive Air Quality Model
with extensions and with
particulate matter chemistry
polymorphonuclear leukocyte(s)
photomultiplier tube
post natal day
probabilistic National Exposure
Model
Photosynthetic EvapoTranspiration
model
proportion of interval differences of
successive  normal-beat intervals
inEKG
PNN50



P02
POC
POD
polyADPR

POMS

ppb
ppb-h
ppbv
pphm
ppm
ppm-h
                                                            ppmv
                                                            PPN

                                                            PPPs
                                                            ppt
                                                            pptv
                                                            PQH2
                                                            PR
                                                            PR-1
                                                            PRB
                                                            preproET-1
                                                            PRYL

                                                            PS
                                                            PS
                                                            PS II
PSA
PSC
PTB
PTR-MS

PU,PUL
PUFA(s)
PV
PVCD

PVD
PVOCs

PWM
PWTES

Pxase
QA
QC
proportion of interval differences of
successive normal-beat intervals
greater than 50 ms in EKG
partial pressure of oxygen
particulate organic carbon
peroxidase
poly(adenosinediphosphate-
ribose)
Portable Ozone Monitoring
Systems
parts per billion
parts per billion per hour
parts per billion by volume
parts per hundred million
parts per million
parts per million hours; weighted
concentration values based on
hourly concentrations: usually
summed over a certain number of
hours, day(s), months, and/or
season.
parts per million by volume
peroxypropionyl nitrate;
peroxypropionic nitric anhydride
power plant plumes
parts per trillion
parts per trillion by volume
plastoquinone
pathogenesis-related (protein)
promoter region 1
policy-relevant background
pre-protein form of ET-1 mRNA
predicted relative yield (biomass)
loss
penalized spline
paradoxical sleep
Photosystem II: enzyme that uses
light to obtain electrons from water
(for photosynthesis).
picryl sulfonic acid
polar stratospheric clouds
preterm birth
proton-transfer-reaction mass
spectroscopy
pulmonary
polyunsaturated fatty acid(s)
potential vorticity
peripheral vascular and
cerebrovascular disease
peripheral vascular disease
photochemical volatile organic
compounds
pokeweed mitogen
(left ventricular) posterior wall
thickness at end systole
peroxidase
Quality Assurance
quality control
     - Do Not Cite or
                                                      XIIV
                                                                           September 2011

-------
QCE
qNP
C|NP
qP
QRS
QT
QTc
r
R, r
r2
R2

R2, r2
RACM

RADM
rALP
RAMS

RANTES
Raw
RB
RBC(s)
rbcL
rbcS
R'CO acyl
R'C(O)-O2
rcdl

RCD3
RCP

RDBMS

Re
REHEX
RER

RF
RGR
RH
RIOPA

RL
RLKs
RMNP

RMR
rMSSD
quasi continuous exercise
non-photochemical quenching
non-photochemical quenching
photochemical quenching
A complex of three distinct
electrocardiogram waves which
represent the beginning of
ventricular contraction
interval measure of the time
interval between the start of the Q
wave and the end of the T wave in
the heart's electrical cycle
corrected QT interval
Pearson correlation coefficient
correlation coefficient
correlation coefficient
multiple regression correlation
coefficient
coefficient of determination
Regional Atmospheric Chemistry
Mechanism
Regional Acid Deposition Model
recombinant antileukoprotease
Regional Atmospheric Modeling
System
regulated upon activation, normal
T cell expressed and secreted
(cells)
airway resistance
respiratory bronchiole
red blood cell(s); erythrocyte(s)
Rubisco large subunit
Rubisco small subunit
acyl carrier protein
acyl peroxy
Arabidopsis mutant radical
induced cell death
rod-cone dysplasia 3
Representative Concentration
Pathways
Relational  Database Management
Systems
Reynolds number
Regional Human Exposure  Model
rough endoplasmic reticulum;
Respiratory exchange ratio
radiative forcing
relative growth rate
relative humidity
Relationship of Indoor, Outdoor,
and Personal Air (study)
total  pulmonary resistance
receptor-like/Pelle kinase group
Rocky Mountain National Park,
Colorado
resting metabolic rate
root mean squared differences
between adjacent normal-to-
normal heartbeat intervals
Rn
RNA
RO2
ROG
ROI

RONO2
ROOM
ROONO2, RO2NO2
ROS
RPD
RR
RRMS
RT
RT
RTLF
RuBisCO; Rubisco

RuBP
a
09

s
S
s.c.
SA
SAB
SAC

SAG21
SAI
S-allele
SAMD

SaO2
SAPALDIA

SAPRC
SAR
SAROAD
SAWgrp
SBNF

SBP
SBUV

SC
Sc
nasal resistance
ribonucleic acid
organic peroxyl; organic peroxy
reactive organic gases
reactive oxygen
intermediate/superoxide anion
organic nitrate
organic peroxides
peroxy nitrate
reactive oxygen species
relative percent difference
normal-to-normal (NN or RR) time
interval between each QRS
complex in the EKG; risk ratio;
relative risk; respiratory rate
relatively remote monitoring sites
respiratory tract
transepithelial resistance
respiratory tract lining fluid
ribulose-1,5-bisphosphate
carboxylase/oxygenase
ribulose bisphosphate
sigma, standard deviation
sigma-g; (geometric standard
deviation)
second
Short; smoker; sulfur; South
subcutaneous (route)
salicylic acid
Science Advisory Board
Staphylococcus aureus Cowan 1
strain
senescence
Systems Applications International
short-allele
S-adenosyl methionine
decarboxylase
oxygen saturation of arterial blood
Study of Air Pollution and Lung
Diseases in Adults
Stratospheric Processes and their
Role in Climate; Statewide Air
Pollution Research  Center,
University of California, Riverside
systemic acquired resistance
Storage and  Retrieval of
Aerometric Data (U.S. EPA
centralized database; superseded
by Aerometric Information
Retrieval System [AIRS])
small airway function group
San  Bernardino National Forest,
California
systolic blood pressure
Solar Backscatter Ultraviolet
Spectrometer
stratum corneum
scandium
     - Do Not Cite or
                                                      XIV
                                                                           September 2011

-------
SCAQS

SCE(s)
SD

SDNN
SE
SEBAS

sec
Sess.
SEM


SENP
SES
SF
SF6
SGA
sRaw
SH
SHEDS

SHEN
slCAM-1

SIDS
SIGMOID

SINIC
SIP
SIPK

SK
SLA
SLAC1

SLAMS

SM
SMD
SME
SMNP


SMOKE

SN

SNAAQS

SNP(s)
SO2
SO42"
SOA
SOC
Southern California Air Quality
Study
sister chromatid exchange(s)
standard deviation; Sprague-
Dawley rat
standard deviation normal-to-
normal (NN or RR) time interval
between each QRS complex in the
EKG
standard error
Social Environment and
Biomarkers of Aging Study
second
session
simultaneously extracted metal;
standard error of the mean;
scanning electron microscopy
Sequoia National  Park, California
socioeconomic status
San Francisco Bay Area
sulfur hexafluoride (tracer gas)
small for gestational age
specific airway conductance
Shenandoah National  Park site
Stochastic Human Exposure and
Dose Simulation
Shenandoah National  Park
soluble intercellular adhesion
molecule
sudden infant death syndrome
sigmoid weighted summed
concentration
Simple Nitrogen Cycle model
State Implementation Plan
salicylic acid (SA) induced protein
kinase
shikimate kinase
specific leaf area
(protein) slow anion channel
associated 1
State and Local Air Monitoring
Stations
smooth muscle
soil moisture deficit
soybean oil methyl ester
Great Smoky Mountain National
Park (North Carolina and
Tennessee)
Spare-Matrix Operator Kernel
Emissions
normalized  slope of the alveolar
plateau
Secondary National Ambient Air
Quality Standards
single-nucleotide polymorphism
sulfur dioxide
sulfate
secondary organic aerosol
soil organic carbon
SOD                superoxide dismutase
SOS                Southern Oxidant Study
SOX                sulfur oxides
SoyFACE            Soybean Free Air gas
                    Concentration Enrichment
                    (Facility)
SP                  surfactant protein (e.g., SPA,
                    SPD); substance P
SP-A                surfactant protein-A
SPF                specific pathogen free
SPMs               special purpose monitors
SP-NK              substance P - neurokinin receptor
                    complex
sRaw,               specific airway resistance
SRBC               sheep red blood cell
SRES               Special Report on Emissions
                    Scenarios
SRM                standard reference method
SRP                standard reference photometers
SSCP               single-strand conformation
                    polymorphism
12931/SvlmJ         mouse strain
STE                stratosphere-troposphere
                    exchange
STEP               Stratospheric-Tropospheric-
                    Exchange Project
STN                speciation trends network
sTNFRI             soluble tumor necrosis factor
                    receptor 1
STP                standard temperature and
                    pressure
STPD               standard temperature and
                    pressure, dry
STRF               Spatio-Temporal Random Field
                    (theory)
subscript i            Index of indoor microenvironments
subscript o           Index of outdoor
                    microenvironments
subscript o,i          Index of outdoor
                    microenvironments adjacent to a
                    given indoor microenvironment /
SUMOO             sum of all hourly average
                    concentrations
SUM06             seasonal sum of all hourly average
                    concentrations 5 0.06 ppm
SUM07             seasonal sum of all hourly average
                    concentrations > 0.07 ppm
SUM08             seasonal sum of all hourly average
                    concentrations > 0.08 ppm
SURE               Sulfate Regional Experiment
                    Program
SVE                supraventricular ectopy
S-W                square-wave
SWS                slow wave sleep
SZA                solar zenith  angle
T                   tau, photochemical lifetime;
                    atmospheric lifetime
t                   t-test statistical value; t statistic
T                   time; duration of exposure
     - Do Not Cite or
                                                     xlvi
                                                                         September 2011

-------
T cell(s)             T lymphocyte(s), thymus-
                    dependent lymphocytes
T1                  first trimester
T2                  second trimester
T3                  triiodothyronine
T3                  third trimester
T4                  thyroxine
TAR                IPCC Third Assessment  Report
TAR WGI            IPCC Third Assessment  Report of
                    Working Group I
TB                  tracheobronchial; terminal
                    bronchioles; tuberculosis
TEA                thiobarbituric acid
TEARS              thiobarbituric acid reactive
                    substances
TC                  total carbon
99mTc                Technetium-99m
T-cells              T-lymphocytes, Thymus-derived
                    lymphocytes
99mTc-DTPA          99mTc-
                    diethylenetriaminepentaacetic acid
Tco                 core temperature
TOLAS              Tunable Diode Laser Absorption
                    Spectrometer
Te                  expiratory time
TEM                transmission electron microscopy;
                    Terrestrial Ecosystem Model
TES                Tropospheric Emission
                    Spectrometer
TexAQS             Texas Air Quality Field Study
Tg                  teragram(s)
TGF                transforming growth factor
TGF (3              transforming growth factor beta
Th                  T helper cell type
Th2                 T helper cell type 2
THC                Total hydrocarbon content
tHcy                total homocysteine
Ti                   inspiratory time
Ti                   titanium
TIA                 transient ischemic attack
TIMP-2              tissue inhibitor of matrix
                    metalloprotease-2
TiO2                titanium dioxide
TLC                total lung capacity
TLNISE             two-level normal independent
                    sampling estimation
Tlr                  toll-like receptor gene
TLR                Toll-like receptor protein  (ex.,
                    TLR2, TLR4)
TMPO              tetramethylphrrolise 1-oxide
TNC                total nonstructural carbohydrate
TNF                tumor necrosis factor (e.g., TNF-a)
TNF-308             tumor necrosis factor genotype
TNF-a              tumor necrosis factor alpha
TNFR               tumor necrosis factor receptor
                                                     TOMS

                                                     TOPSE

                                                     tPA
                                                     TPLIF

                                                     TRAMP

                                                     TREGRO
                                                     TRIFFID

                                                     TRIM

                                                     TRIM.Expo

                                                     TRP

                                                     TSH
                                                     TSP
                                                     TTFMS

                                                     TWA
                                                     TX
                                                     TXB2
                                                     UA
                                                     UAM
                                                     UCL
                                                     UDGT

                                                     UDP
                                                     U.K.
                                                     UNECE

                                                     UNEP

                                                     UNFCCC

                                                     U-O
                                                     U-O2"
                                                     U-03"
                                                     URI
                                                     URS
                                                     URT

                                                     U.S.
                                                     USC; U.S.C.
                                                     USDA
                                                     USFS
                                                     USGCRP

                                                     USGS
                                                     UV
                                                     UV-A
Total Ozone Mapping/Monitoring
Satellite; total ozone mapping
spectrometer
Tropospheric Ozone Production
About the Spring Equinox
tissue plasminogen activator
two-photon laser-induced
fluorescence
TexAQS-ll Radical and Aerosol
Measurement Project
Tree Growth Model
Top-down  Representation of
Interactive Foliage and Flora
Including Dynamics
Total Risk  Integrated Methodology
(model)
Total Risk  Integrated Methodology
Exposure Event (model)
transient receptor potential (ion
channel[s], ex., TRP-A1, TRP-V1,
TRP-M8)
thyroid stimulating hormone
total suspended particles
two-tone frequency-modulated
spectra scopy
time-weighted average
thromboxane (e.g., TXB2)
thromboxane B2
uric acid; Urate
Urban Airshed Model
upper 95th% confidence limit
UDP -galactose-1,2,-diacylglycerol
galactosyltransferase
uridine diphosphate
United Kingdom
United Nations Economic
Commission for Europe
United Nations Environmental
Programme
United Nations Framework
Convention on Climate Change
epioxides formed from uric acid
peroxides formed from uric acid
ozonides formed from  uric acid
upper respiratory infection
upper respiratory symptoms
upper respiratory tract; upper
airways
United States (of America)
U.S. Code
U.S. Department of Agriculture
U.S. Forest Service
U.S. Global Change Research
Program
U.S. Geological Survey
ultraviolet radiation
ultraviolet radiation at wavelengths
of 320 to 400 nm
- Do Not Cite or
                                               XIVII
               September 2011

-------
UV-B

UV-C

UV-DIAL

V
V, mV, |JV
VA
Val
VC
VCAM
Vd

VD

VE

VEGF
VEmax
Vmax
Vmax25%

Vmax50%

Vmax75o/0

VMD
Vn
VO2
VO2max
VOC(s)
VP
VP50%

VPD

VT
VTB
ultraviolet radiation at wavelengths
of 280 to 320 nm
ultraviolet radiation at wavelengths
of 200 to 280 nm
Ultraviolet Differential Absorption
Lidar
vanadium
volt, millivolt, microvolt
alveolar ventilation
valine
vital capacity
vascular cell adhesion molecule
deposition rate, deposition velocity
(cm/s)
volume of the anatomic or
physiological dead space
ventilation rate; minute ventilation;
ventilatory volume
vascular endothelial growth factor
maximum minute ventilation
maximum velocity
maximum expiratory flow at 25%
of the vital capacity
maximum expiratory flow at 50%
of the vital capacity
maximum expiratory flow at 75%
of the vital capacity
volume median diameter
nasal volume
oxygen consumption
maximum volume per time, of
oxygen (maximal oxygen
consumption, maximal oxygen
uptake or aerobic capacity)
volatile organic compound(s)
volumetric penetration
volume at which 50% of an inhaled
bolus is absorbed
vapor pressure deficit; Vehicles
per day; Ventricular premature
depolarization
tidal volume
terminal bronchiole region volume
VTmax              maximum tidal volume
VUA                volume of the upper airways
vWF                von Willebrand factor
W                  width; wilderness; week(s)
W126               cumulative integrated exposure
                    index with a sigmoidal weighting
                    function
W95                cumulative integrated exposure
                    index with a sigmoidal weighting
                    function
WBC                white blood cell
WBGT              wet bulb globe temperature
we                  sigmoidal weighting of hourly O3
                    concentration
WCB                warm conveyor belt
WED                (U.S. EPA  NHEERL) Western
                    Ecology Division
WF, WFM           White Face Mountain site
WHI                 Women's Health Initiative
WHO                World Health Organization
wk(s)                week(s)
W/m2, W m"2         watts per square meter
WMO               World Meteorological Organization
WMO/UNEP         World Meteorological
                    Organization/United Nations
                    Environment Program
WRF                Weather Research and
                    Forecasting model
Ws                 Wassilewskija Arabidopsis ecotype
WS                 wood smoke
WT                 wild type; White Top Mountain site
wt %                percent by weight
WUS                western U.S.
w/v                 weight per volume
Y                   three parameter Weibull model
yr                   year
Z                   Airway generation
ZAPS               Zonal Air Pollution System
ZELIG              a forest succession simulation
                    model
Zn                  zinc
     - Do Not Cite or
                                                                          September 2011

-------
      PREAMBLE
      Process  of ISA Development

 1                   This preamble outlines the general process for developing an Integrated Science
 2                   Assessment (ISA) including the framework for evaluating weight of evidence and
 3                   drawing scientific conclusions and causal judgments. The ISA provides a concise review,
 4                   synthesis, and evaluation of the most policy-relevant science to serve as a scientific
 5                   foundation for the review of the National Ambient Air Quality Standards (NAAQS). The
 6                   general process for NAAQS reviews is described at
 7                   http://www.epa.gov/ttn/naaqs/review.html: information for individual NAAQS reviews is
 8                   available at www .epa. gov/ttn/naaqs. This preamble is a general discussion of the basic
 9                   steps and criteria used in developing an ISA; for each ISA, specific details and
10                   considerations are included in the introductory section for that assessment.

11                   The fundamental process for developing an ISA includes:

12                       •  literature searches;
13                       •  study selection;
14                       •  evaluation and integration of the evidence; and
15                       •  development of scientific conclusions and causal judgments.

16                   An initial step in this process is publication of a call for information in the Federal
17                   Register that invites the public to provide information relevant to the assessment, such as
18                   new publications on health or welfare1  effects of the pollutant, or from atmospheric and
19                   exposure sciences fields. EPA maintains an ongoing literature search process for
20                   identification of relevant scientific studies published since the last review of the NAAQS.
21                   Search strategies  are designed for pollutants and scientific disciplines and iteratively
22                   modified to optimize identification of pertinent publications. Papers are identified for
23                   inclusion in several additional ways:  specialized searches on specific topics; independent
24                   review of tables of contents for journals in which relevant papers may be published;
25                   independent identification of relevant literature by expert scientists; review of citations in
26                   previous assessments and identification by the public and CASAC during the external
27                   review process. Publications considered for inclusion in the ISA are added to the Health
28                   and Environmental Research Online (HERO) database developed by EPA
29                   (http://hero.epa.gov/): the references in the ISA include a hyperlink to the database.
        1 Welfare effects as defined in Clean Air Act section 302(h) [42 U.S.C. 7602(h)] include, but are not limited to, "effects on soils,
      water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility and climate, damage to and deterioration of
      property, and hazards to transportation, as well as effects on economic values and on personal comfort and well-being."
      Draft - Do Not Cite or Quote                        xlix                                  September 2011

-------
 1                   Studies that have undergone scientific peer review and have been published or accepted
 2                   for publication and reports that have undergone review are considered for inclusion in the
 3                   ISA. Analyses conducted by EPA using publicly available data are also considered for
 4                   inclusion in the ISA. All relevant epidemiologic, controlled human exposure,
 5                   toxicological, and ecological and welfare effects studies published since the last review
 6                   are considered, including those related to exposure-response relationships, mode(s) of
 7                   action (MOA), and potentially at-risk populations and lifestages. Studies on atmospheric
 8                   chemistry, environmental fate and transport, dosimetry, toxicokinetics and exposure are
 9                   also considered for inclusion in the document, as well as analyses of air quality and
10                   emissions data. References that were considered for inclusion in a specific ISA can be
11                   found using the HERO website (http://hero.epa.gov).

12                   Each  ISA builds upon the conclusions of previous assessments for the pollutant under
13                   review. EPA focuses on peer reviewed literature published following the completion of
14                   the previous review and on any new interpretations of previous literature, integrating the
15                   results of recent  scientific studies with previous findings. Important older studies may be
16                   discussed in detail to reinforce key concepts and conclusions or for reinterpretation in
17                   light of newer data.  Older studies also are the primary focus in some areas of the
18                   document where research efforts have subsided, or if these older studies remain the
19                   definitive works available in the literature.

20                   Selection of studies for inclusion in the ISA is based on the general scientific quality of
21                   the study, and consideration of the extent to which the study is informative and policy-
22                   relevant. Policy relevant and informative studies include those that provide a basis for or
23                   describe the relationship between the criteria pollutant and effects, including studies that
24                   offer  innovation in method or design and studies that reduce uncertainty on critical issues,
25                   such as analyses of confounding or effect modification by copollutants or other variables,
26                   analyses of concentration-response or dose-response relationships, or analyses related to
27                   time between exposure and response. Emphasis is placed on studies that examine effects
28                   associated with pollutant concentrations  relevant to current population and ecosystem
29                   exposures, and particularly those pertaining to concentrations currently found in ambient
30                   air. Other studies are included if they contain unique data, such as a previously
31                   unreported effect or MOA for an observed effect, or examine multiple concentrations to
32                   elucidate exposure-response relationships. In general, in assessing the scientific quality
33                   and relevance of health and welfare effects studies, the following considerations have
34                   been taken into account when selecting studies for inclusion in the ISA.

35                       •   Are the study populations, subjects, or animal models adequately selected, and
36                          are they sufficiently well defined to allow for meaningful  comparisons
37                          between  study or exposure groups?
      Draft - Do Not Cite or Quote                         I                                   September 2011

-------
 1                       •  Are the statistical analyses appropriate, properly performed, and properly
 2                          interpreted? Are likely covariates adequately controlled or taken into account
 3                          in the study design and statistical analysis?
 4                       •  Are the air quality data, exposure, or dose metrics of adequate quality and
 5                          sufficiently representative of information regarding ambient conditions?
 6                       •  Are the health, ecological or welfare effect measurements meaningful, valid
 7                          and reliable?
 8                       "Do the analytical methods provide adequate sensitivity and precision to
 9                          support conclusions?

10                   Considerations specific to particular disciplines include the following. In selecting
11                   epidemiologic studies, EPA considers whether a given study: (1) presents information on
12                   associations with short- or long-term pollutant exposures at or near ambient conditions;
13                   (2) addresses potential confounding by other pollutants;  (3) assesses potential effect
14                   modifiers; (4) evaluates health endpoints and populations not previously extensively
15                   researched; and (5) evaluates important methodological issues related to interpretation of
16                   the health evidence (e.g., lag or time period between exposure and effects, model
17                   specifications, thresholds, mortality displacement).

18                   Considerations for the selection of research evaluating controlled human exposure or
19                   animal toxicological studies includes a focus on studies conducted using relevant
20                   pollutant exposures. For both types of studies, relevant pollutant exposures are
21                   considered to be those generally within one or two orders of magnitude of ambient
22                   concentrations. Studies in which higher doses were used may also be considered if they
23                   provide information relevant to understanding MOA or mechanisms, as noted below.

24                   Evaluation of controlled human exposure studies focuses on those that approximated
25                   expected human exposure conditions in terms of concentration and duration.  Studies
26                   should include control exposures to filtered air, as appropriate. In the selection of
27                   controlled human exposure studies, emphasis is placed on studies that: (1) investigate
28                   potentially at-risk populations and lifestages such as people with asthma or
29                   cardiovascular diseases, children or older adults; (2) address issues  such as concentration-
30                   response or time-course  of responses; and (3) have sufficient statistical power to assess
31                   findings.

32                   Review of the animal toxicological evidence focuses on studies that approximate
33                   expected human dose conditions, which vary depending on the dosimetry, toxicokinetics
34                   and biological sensitivity of the particular laboratory animal species or strains studied.
35                   Emphasis is placed on studies that: (1) investigate animal models of disease that can
36                   provide information on populations potentially at increased risk of effects; (2) address
      Draft - Do Not Cite or Quote                        li                                  September 2011

-------
 1                   issues such as concentration-response or time-course of responses; and (3) have sufficient
 2                   statistical power to assess findings. Due to resource constraints on exposure duration and
 3                   numbers of animals tested, animal studies typically utilize high-concentration exposures
 4                   to acquire data relating to mechanisms and assure a measurable response. Emphasis is
 5                   placed on studies using doses or concentrations generally within  1-2 orders of magnitude
 6                   of current levels. Studies with higher concentration exposures or doses are considered to
 7                   the extent that they provide useful information to inform our understanding of
 8                   interspecies differences and potential differences between healthy and susceptible human
 9                   populations. Results from in vitro studies may also be included if they provide
10                   mechanistic insight or further support for results demonstrated in vivo.

11                   These criteria provide benchmarks for evaluating various studies and for focusing on the
12                   policy-relevant studies in assessing the body of health, ecological and welfare effects
13                   evidence. As  stated initially, the intent of the ISA is to provide a  concise review,
14                   synthesis, and evaluation of the most policy-relevant science to serve as a scientific
15                   foundation for the review of the NAAQS, not extensive summaries of all health,
16                   ecological and welfare effects studies for a pollutant. Of most relevance for inclusion  of
17                   studies is whether they provide useful qualitative or quantitative  information on
18                   exposure-effect or exposure-response relationships for effects associated with pollutant
19                   exposures at doses or concentrations relevant to ambient conditions that can inform
20                   decisions on whether to retain or revise the standards.

21                   In developing an ISA, EPA reviews and  summarizes the evidence from: studies of
22                   atmospheric sciences and exposure; the health effects evidence from toxicological,
23                   controlled human exposure and epidemiologic studies; and ecological and welfare effects
24                   evidence. In the  process of developing the first draft ISA, EPA may convene a public
25                   workshop in which EPA and non-EPA experts review the  scientific content of
26                   preliminary draft materials to ensure that the ISA is up to date and focused on the most
27                   policy-relevant findings, and to assist EPA with integration of evidence within and across
28                   disciplines.

29                   EPA integrates the evidence from across scientific disciplines or study types  and
30                   characterizes the weight of evidence for relationships between the pollutant and various
31                   outcomes. The integration of evidence on health, and ecological or welfare effects,
32                   involves collaboration between scientists from various disciplines. As an example, an
33                   evaluation of health effects evidence would include the integration of the results from
34                   epidemiologic, controlled human exposure, and toxicological studies, and application of
35                   the causal framework (described below) to draw conclusions. Using the causal
36                   framework described in the following section, EPA scientists consider aspects such as
37                   strength, consistency, coherence, and biological plausibility of the evidence,  and develop
      Draft - Do Not Cite or Quote                        III                                   September 2011

-------
 1                  draft causality determinations on the nature of the relationships. Causality determinations
 2                  often entail an iterative process of review and evaluation of the evidence. Two drafts of
 3                  the ISA are typically released for review by the CASAC and the public, and comments
 4                  received on the characterization of the science as well as the implementation of the causal
 5                  framework are carefully considered in revising and completing the final ISA.
      EPA Framework for Causal Determination

 6                  EPA has developed a consistent and transparent basis to evaluate the causal nature of air
 7                  pollution-related health or welfare effects for use in developing IS As. The framework
 8                  described below establishes uniform language concerning causality and brings more
 9                  specificity to the findings. This standardized language was drawn from sources across the
10                  federal government and wider scientific community, especially the National Academy of
11                  Sciences (NAS) Institute of Medicine (IOM) document, Improving the Presumptive
12                  Disability Decision-Making Process for Veterans (2008). a comprehensive report on
13                  evaluating causality. This framework:

14                      •  describes the kinds of scientific evidence used in establishing a general causal
15                         relationship between exposure and health effects;
16                      •  characterizes the evidence necessary to reach a conclusion about the existence
17                         of a causal relationship;
18                      •  identifies issues and approaches related to uncertainty;  and
19                      •  provides a framework for classifying and characterizing the weight of
20                         evidence in support of a general causal relationship.

21                  Approaches to assessing the separate and combined lines of evidence
22                  (e.g., epidemiologic, controlled human exposure, and animal toxicological studies) have
23                  been formulated by a number of regulatory and science agencies, including the IOM of
24                  the NAS (2008). International Agency for Research on Cancer (2006). EPA Guidelines
25                  for Carcinogen Risk Assessment (2005). and Centers for Disease Control and Prevention
26                  (2004). Causal inference criteria have also been described for ecological effects evidence
27                  (U.S. EPA. 1998; Fox. 1991). These formalized approaches offer guidance for assessing
28                  causality. The frameworks are similar in nature, although adapted to different purposes,
29                  and have proven effective in providing a uniform structure and language for causal
30                  determinations.
      Draft - Do Not Cite or Quote                       Nil                                 September 2011

-------
      Evaluating  Evidence for Inferring Causation

 1                   The 1964 Surgeon General's report defined "cause" as a "significant, effectual
 2                   relationship between an agent and an associated disorder or disease in the host" (HEW):
 3                   more generally, a cause is defined as an agent that brings about an effect or a result. An
 4                   association is the statistical relationship among variables; alone, however, it is
 5                   insufficient proof of a causal relationship between an exposure and a health outcome.
 6                   Unlike an association, a causal claim supports the creation of counterfactual claims, that
 7                   is, a claim about what the world would have been like under different or changed
 8                   circumstances (Samet and Bodurow. 2008).

 9                   Many of the health and environmental outcomes reported in these studies have complex
10                   etiologies. Diseases such as asthma, coronary heart disease (CHD) or cancer are typically
11                   initiated by multiple agents. Outcomes depend on a variety of factors, such as age,
12                   genetic susceptibility, nutritional status, immune competence, and social factors (Samet
13                   and Bodurow. 2008; Gee and Payne-Sturges. 2004). Effects on ecosystems are often also
14                   multifactorial with a complex web of causation. Further, exposure to a combination of
15                   agents could cause  synergistic or antagonistic effects. Thus, the observed risk may
16                   represent the net effect of many actions and counteractions.

17                   In estimating the causal influence of an exposure on health or environmental effects, it is
18                   recognized that scientific findings incorporate uncertainty. "Uncertainty" can be defined
19                   as having limited knowledge to exactly describe an existing state or future outcome,
20                   e.g., the lack of knowledge about the correct value for a specific measure or estimate.
21                   Uncertainty analysis may be qualitative or quantitative in nature. In many cases, the
22                   analysis is qualitative, and can include professional judgment or inferences based on
23                   analogy with similar situations. Quantitative uncertainty analysis may include use of
24                   simple measures (e.g., ranges) and analytical techniques. Quantitative uncertainty
25                   analysis might progress to more complex measures and techniques, if needed for decision
26                   support. Various approaches to evaluating uncertainty include classical statistical
27                   methods, sensitivity analysis, or probabilistic uncertainty analysis, in order of increasing
28                   complexity and data requirements. However, data may not be available for all aspects of
29                   an assessment and those data that are available may be of questionable or unknown
30                   quality. Ultimately, the assessment is based on a number of assumptions with varying
31                   degrees of uncertainty. The ISA generally evaluates uncertainties qualitatively in
32                   assessing the evidence from across studies; in some situations quantitative analysis
33                   approaches, such as meta-regression, may be used.

34                   Publication bias is a source of uncertainty regarding the magnitude of health risk
35                   estimates. It is well understood that studies reporting non-null findings are more likely to
36                   be published than reports of null findings, and publication bias can also result in


      Draft - Do Not Cite or Quote                        liv                                 September 2011

-------
 1                   overestimation of effect estimate sizes (loannidis, 2008). For example, effect estimates
 2                   from single-city epidemiologic studies have been found to be generally larger than those
 3                   from multicity studies (Bell et al.. 2005).

            Consideration of evidence from scientific disciplines
 4                   Moving from association to causation involves the elimination of alternative explanations
 5                   for the association. The ISA focuses on evaluation of the findings from the body of
 6                   evidence, drawing upon the results of all studies determined to meet the criteria described
 7                   previously. Causality determinations are based on the evaluation and synthesis of
 8                   evidence from across scientific disciplines. The relative importance of different types of
 9                   evidence varies by pollutant or assessment, as does the availability of different types of
10                   evidence for causality determination. Three general types of studies inform consideration
11                   of human health effects: controlled human exposure, epidemiologic and toxicological
12                   studies. Evidence on ecological or welfare effects may be drawn from a variety of
13                   experimental approaches (e.g., greenhouse, laboratory, field) and numerous disciplines
14                   (e.g., community ecology, biogeochemistry and paleological/historical reconstructions).

15                   The most direct evidence of a causal relationship between pollutant exposures and human
16                   health effects comes from controlled human exposure studies. Controlled human
17                   exposure studies experimentally evaluate the health effects of administered exposures in
18                   human volunteers under highly controlled laboratory conditions. Also referred to as
19                   human clinical studies, these  experiments allow investigators to expose subjects to known
20                   concentrations of air pollutants under carefully regulated environmental conditions and
21                   activity levels. In some instances, controlled human exposure studies can also be used to
22                   characterize concentration-response relationships at pollutant concentrations relevant to
23                   ambient conditions. Controlled human exposures are typically conducted using a
24                   randomized crossover design, with subjects exposed both to the pollutant and a clean air
25                   control. In this way, subjects  serve as their own controls, effectively controlling for many
26                   potential confounders. However, controlled human exposure studies are limited by a
27                   number of factors, including small sample size and short exposure time. For example,
28                   exposure patterns relevant to understanding real-world exposures, especially long-term
29                   exposures, are generally not practical to replicate in a laboratory setting.  In addition,
30                   although subjects do serve as their own controls, personal exposure to pollutants in the
31                   hours and days preceding the controlled exposures may vary significantly between and
32                   within individuals. Finally, controlled human exposure studies require investigators to
33                   adhere to stringent health criteria for subjects included in the study, and therefore the
34                   results cannot necessarily be generalized to an entire population. Although some
35                   controlled human exposure studies have included health-compromised individuals such
36                   as those with respiratory or cardiovascular disease, these individuals must also be
      Draft - Do Not Cite or Quote                        Iv                                 September 2011

-------
 1                   relatively healthy and may not represent the most sensitive individuals in the population.
 2                   In addition, the study design is limited to exposures and endpoints that are not expected
 3                   to result in severe health outcomes. Thus, not observing an effect in controlled human
 4                   exposure studies does not necessarily mean that a causal relationship does not exist.
 5                   While controlled human exposure studies provide important information on the biological
 6                   plausibility of associations observed in epidemiologic studies, observed effects in these
 7                   studies may underestimate the response in certain populations.

 8                   Epidemiologic studies provide important information on the associations between health
 9                   effects and exposure of human populations to ambient air pollution. In epidemiologic or
10                   observational studies of humans, the investigator generally does not control exposures or
11                   intervene with the study population. Broadly, observational studies can describe
12                   associations between exposures and effects. These studies fall into several categories:
13                   e.g., cross-sectional, prospective cohort, panel and time-series studies. "Natural
14                   experiments" offer the opportunity to investigate changes in health related to a change in
15                   exposure, such as closure of a pollution source.

16                   In evaluating epidemiologic studies, consideration of many study design factors and
17                   issues must be taken into account to properly inform their interpretation. One key
18                   consideration is evaluation of the potential contribution of the pollutant to a health
19                   outcome when it is a component of a complex air pollutant mixture. Reported effect
20                   estimates in epidemiologic studies may reflect: independent effects on health outcomes;
21                   effects of the pollutant acting as an indicator of a copollutant or a complex ambient air
22                   pollution mixture; effects resulting from interactions between that pollutant and
23                   copollutants.

24                   In the evaluation of epidemiologic evidence, one important consideration is potential
25                   confounding. Confounding is "... a confusion of effects. Specifically, the apparent  effect
26                   of the exposure of interest is distorted because the effect of an extraneous factor is
27                   mistaken for or mixed with the actual exposure effect (which may be null)" (Rothman
28                   and Greenland. 1998). One approach to remove spurious associations due to possible
29                   confounders is to control for characteristics that may differ between exposed and
30                   unexposed persons; this is frequently termed "adjustment." Scientific judgment is needed
31                   to evaluate likely sources and extent of confounding, together with consideration of how
32                   well the existing constellation of study designs, results, and analyses address this
33                   potential threat to inferential validity. A  confounder is associated with both the exposure
34                   and the effect; for example, confounding can occur between correlated pollutants that are
3 5                   associated with the same effect.

36                   Several statistical methods are available to detect and control for potential confounders,
37                   with none of them being completely satisfactory. Multivariable regression models
      Draft - Do Not Cite or Quote                        Ivi                                  September 2011

-------
 1                   constitute one tool for estimating the association between exposure and outcome after
 2                   adjusting for characteristics of participants that might confound the results. The use of
 3                   multipollutant regression models has been the prevailing approach for controlling
 4                   potential confounding by copollutants in air pollution health effects studies. Finding the
 5                   likely causal pollutant from multipollutant regression models is made difficult by the
 6                   possibility that one or more air pollutants may be acting as a surrogate for an unmeasured
 7                   or poorly measured pollutant or for a particular mixture of pollutants. In addition, more
 8                   than one pollutant may exert similar health effects, resulting in independently observed
 9                   associations for multiple pollutants. The number and degree of diversity of covariates,  as
10                   well as their relevance to the potential confounders, remain matters of scientific
11                   judgment. Despite these limitations, the use of multipollutant models is still the
12                   prevailing approach employed in most air pollution epidemiologic studies and provides
13                   some insight into the potential for confounding or interaction among pollutants.

14                   Confidence that unmeasured confounders are not producing the findings is increased
15                   when multiple studies are conducted in various settings using different subjects or
16                   exposures, each of which might eliminate another source of confounding from
17                   consideration. For example, multicity studies which use a consistent method to analyze
18                   data from across locations with different levels of covariates can provide insight on
19                   potential confounding in associations. Intervention studies, because of their quasi -
20                   experimental nature, can be particularly useful in characterizing causation.

21                    Another important consideration in the evaluation of epidemiologic evidence is effect
22                   modification, which occurs when the effect differs between subgroups or strata; for
23                   example, effect estimates that vary by age group or potential risk factor. "Effect-measure
24                   modification differs from confounding in several ways. The main difference is  that,
25                   whereas confounding is a bias that the investigator hopes to prevent or remove  from the
26                   effect estimate, effect-measure modification is a property of the effect under study ...  In
27                   epidemiologic analysis one tries to eliminate confounding but one tries to detect and
28                   estimate effect-measure modification" (Rothman and Greenland. 1998). When  a risk
29                   factor is a confounder, it is the true cause of the association observed between the
30                   exposure and the  outcome; when a risk factor is an effect modifier, it changes the
31                   magnitude of the  association between the exposure and the outcome in stratified analyses.
32                   For example, the presence of a preexisting disease or indicator of low socioeconomic
33                   status may be an effect modifier in causing increased risk of effects related to air
34                   pollution exposure. It is often possible to stratify the relationship between health outcome
35                   and exposure by one or more of these potential effect modifiers. For variables that
36                   modify the association, effect estimates in each stratum will be different from one another
37                   and different from the overall estimate, indicating a different exposure-response
38                   relationship  may exist in populations represented by these variables.
      Draft - Do Not Cite or Quote                        Ivii                                  September 2011

-------
 1                   Another key consideration for epidemiologic evidence is exposure measurement error.
 2                   There are several components that contribute to exposure measurement error in
 3                   epidemiologic studies, including the difference between true and measured ambient
 4                   concentrations, the difference between average personal exposure to ambient pollutants
 5                   and ambient concentrations at central monitoring sites, and the use of average population
 6                   exposure rather than individual exposure estimates.

 7                   The third main type of health effects evidence, animal toxicological studies, provides
 8                   information on the pollutant's biological action under controlled and monitored exposure
 9                   circumstances. Taking into account physiological differences of the experimental species
10                   from humans, these studies inform characterization of health effects of concern,
11                   exposure-response relationships and MOAs. Further, animal models can inform
12                   determinations of at-risk or susceptible populations. These studies evaluate the effects of
13                   exposures to a variety of pollutants in a highly controlled laboratory setting and allow
14                   exploration of toxicological pathways or mechanisms by which a pollutant may cause
15                   effects. Understanding the biological mechanisms underlying various health outcomes
16                   can prove crucial in establishing or negating causality. In the absence of human studies
17                   data, extensive, well-conducted animal toxicological studies can support determinations
18                   of causality, if the evidence base indicates that similar responses are expected in humans
19                   under ambient exposure conditions.

20                   Interpretations of animal toxicological studies are affected by limitations associated with
21                   extrapolation between animal and human responses. The differences between humans
22                   and other species have to  be taken into consideration, including metabolism, hormonal
23                   regulation,  breathing pattern, and differences in lung structure and anatomy. Also, in spite
24                   of a high degree  of homology and the existence of a high percentage of orthologous
25                   genes across humans and rodents (particularly mice), extrapolation of molecular
26                   alterations at the gene level is complicated by species-specific differences in
27                   transcriptional regulation. Given these differences, there are uncertainties associated with
28                   quantitative extrapolations of observed pollutant-induced pathophysiological alterations
29                   between laboratory animals and humans, as those alterations are under the control of
30                   widely varying biochemical, endocrine, and neuronal factors.

31                   For ecological effects assessment, both laboratory and field studies (including field
32                   experiments and observational studies) can provide useful data for causality
33                   determination. Because conditions can be controlled in laboratory studies, responses may
34                   be less variable and smaller differences easier to detect. However, the control conditions
35                   may limit the range of responses (e.g., animals may not be able to seek alternative food
36                   sources), so they may not reflect responses that would occur in the natural environment.
37                   In addition, larger-scale processes are difficult to reproduce in the laboratory.
      Draft - Do Not Cite or Quote                       Iviii                                 September 2011

-------
 1                   Field observational studies measure biological changes in uncontrolled situations, and
 2                   describe an association between a disturbance and an ecological effect. Field data can
 3                   provide important information for assessments of multiple stressors or where site-specific
 4                   factors significantly influence exposure. They are also often useful for analyses of larger
 5                   geographic scales and higher levels of biological organization. However, because
 6                   conditions are not controlled, variability is expected to be higher and differences  harder
 7                   to detect. Field surveys are most useful for linking stressors with effects when stressor
 8                   and effect levels are measured concurrently. The presence of confounding factors can
 9                   make it difficult to attribute observed effects to specific stressors.

10                   Intermediate between laboratory and field are studies that use environmental media
11                   collected from the field to examine response in the laboratory, and experiments that are
12                   performed in the natural environment while controlling for some environmental
13                   conditions (i.e. mesocosm studies). This type of study in manipulated natural
14                   environments can be considered a hybrid between a field experiment and laboratory study
15                   since some aspects are performed under controlled conditions but others are not.  They
16                   make it possible to observe community and/or ecosystem dynamics, and provide  strong
17                   evidence for causality when combined with findings of studies that have been made
18                   under more controlled conditions.

            Application of Framework for Causal Determination
19                   In its evaluation of the scientific evidence on health or welfare effects of criteria
20                   pollutants, EPA determines the weight of evidence in support of causation and
21                   characterizes the strength of any resulting causal classification. EPA also evaluates the
22                   quantitative evidence and draws scientific conclusions, to the extent possible, regarding
23                   the concentration-response relationships and the loads to ecosystems, exposure doses or
24                   concentrations, duration and pattern of exposures at which effects are observed.

25                   To aid judgment, various "aspects"2 of causality have been discussed by many
26                   philosophers and scientists. The 1964 Surgeon General's report on tobacco smoking
27                   discussed criteria for the evaluation of epidemiologic studies, focusing  on consistency,
28                   strength,  specificity, temporal relationship, and coherence (HEW. 1964). Sir Austin
29                   Bradford Hill (1965) articulated aspects of causality in epidemiology and public health
30                   that have been widely used (Samet and Bodurow. 2008: IARC. 2006: U.S. EPA.  2005:
31                   HHS. 2004). These aspects (Hill, 1965) have been modified (Table I) for use in causal
        2 The "aspects" described by Hill (1965) have become, in the subsequent literature, more commonly described as "criteria." The
      original term "aspects" is used here to avoid confusion with "criteria" as it is used, with different meaning, in the Clean Air Act.
      Draft - Do Not Cite or Quote                        lix                                  September 2011

-------
     Table I
                  Aspects to aid in judging causality
     Consistency of the
     observed association
                        An inference of causality is strengthened when a pattern of elevated risks is observed
                        across several independent studies. The reproducibility of findings constitutes one of the
                        strongest arguments for causality. If there are discordant results among investigations,
                        possible reasons such as differences in exposure, confounding factors, and the power of
                        the study are considered.
Coherence              An inference of causality from one line of evidence (e.g., epidemiologic, clinical or animal
                        studies) may be strengthened by other lines of evidence that support a cause-and-effect
                        interpretation of the association. Evidence on ecological or welfare effects may be drawn
                        from a variety of experimental approaches (e.g., greenhouse, laboratory, and field) and
                        subdisciplines of ecology (e.g., community ecology, biogeochemistry and
                        paleological/historical reconstructions). The coherence of evidence from various fields
                        greatly adds to the strength of an inference of causality. In addition, there may be
                        coherence in demonstrating effects across multiple study designs or related health
                        endpoints within one scientific line of evidence.
      Biological plausibility.
      Biological gradient
      (exposure-response
      relationship)

      Strength of the observed
      association

      Experimental evidence
     Temporal relationship of
     the observed association
     Specificity of the
     observed association
                        An inference of causality tends to be strengthened by consistency with data from
                        experimental studies or other sources demonstrating plausible biological mechanisms. A
                        proposed mechanistic linking between an effect and exposure to the agent is an important
                        source of support for causality, especially when data establishing the existence and
                        functioning of those mechanistic links are available.
                        A well-characterized exposure-response relationship (e.g., increasing  effects associated
                        with greater exposure) strongly suggests cause and effect, especially when such
                        relationships are also observed for duration of exposure (e.g., increasing effects observed
                        following longer exposure times).
                        The finding of large, precise risks increases confidence that the association is not likely
                        due to chance,  bias, or other factors. However, it is noted that a small  magnitude in an
                        effect estimate may represent a substantial effect in a population.
                        Strong evidence for causality can be provided through "natural experiments" when a
                        change in exposure is found to result in  a change in occurrence or frequency of health or
                        welfare  effects.
                        Evidence of a temporal sequence between the introduction of an agent, and appearance
                        of the effect, constitutes another argument in favor of causality.
                        Evidence linking an exposure to a specific outcome can provide a strong argument for
                        causation. However, it must be recognized that rarely, if ever, does exposure to a pollutant
                        invariably predict the occurrence of an outcome,  and that a given outcome may have
                        multiple causes.
     Analogy                 Structure activity relationships and information on the agent's structural analogs can
                              provide insight into whether an association is causal. Similarly, information on mode of
                              action for a chemical, as one  of many structural analogs, can inform decisions regarding
                              likely causality.
1
2
3
4
5
                determinations specific to health and welfare effects for pollutant exposures (U.S. EPA.

                2009d_).3 Although these aspects provide a framework for assessing the evidence, they do
                not lend themselves to being considered in terms of simple formulas or fixed rules of

                evidence leading to conclusions about causality (Hill. 1965). For example, one cannot
                simply count the number of studies reporting statistically significant results or
       3 The Hill aspects were developed for interpretation of epidemiologic results. They have been modified here for use with a broader
     array of data, i.e., epidemiologic, controlled human exposure, ecological, and animal toxicological studies, as well as in vitro data,
     and to be more consistent with EPA's Guidelines for Carcinogen Risk Assessment.
      Draft - Do Not Cite or Quote
                                                    Ix
September 2011

-------
 1                   statistically nonsignificant results and reach credible conclusions about the relative
 2                   weight of the evidence and the likelihood of causality. Rather, these aspects are taken into
 3                   account with the goal of producing an objective appraisal of the evidence, informed by
 4                   peer and public comment and advice, which includes weighing alternative views on
 5                   controversial issues. In addition, it is important to note that the aspects in Table I cannot
 6                   be used as a strict checklist, but rather to determine the weight of the evidence for
 7                   inferring causality. In particular, not meeting one or more of the principles does not
 8                   automatically preclude a determination of causality [see discussion in (HHS. 2004)].

            Determination  of Causality
 9                   In the ISA, EPA assesses the body of relevant literature, building upon evidence available
10                   during previous NAAQS reviews, to draw conclusions on the causal relationships
11                   between relevant pollutant exposures and health or environmental effects. ISAs use a
12                   five-level hierarchy that classifies the weight of evidence for causation4. In developing
13                   this hierarchy, EPA has drawn on the work of previous evaluations, most prominently the
14                   lOM's Improving the Presumptive Disability Decision-Making Process for Veterans
15                   (Samet and Bodurow. 2008). EPA's Guidelines for Carcinogen Risk Assessment (U.S.
16                   EPA. 2005). and the U.S. Surgeon General's smoking report (HHS. 2004). This weight of
17                   evidence evaluation is based on various lines of evidence from across the health and
18                   environmental effects disciplines. These separate judgments are integrated into a
19                   qualitative statement about the overall weight of the evidence and causality. The five
20                   descriptors for causal determination are described in Table II.

21                   Determination of causality involves the evaluation of evidence for different types of
22                   health, ecological or welfare effects associated with short- and long-term exposure
23                   periods. In making determinations of causality, evidence is evaluated for major outcome
24                   categories and then conclusions are drawn based upon the integration of evidence from
25                   across  disciplines and also across the spectrum of related endpoints. In making causal
26                   judgments, the ISA focuses on major outcome categories (e.g., respiratory effects,
27                   vegetation growth), by evaluating the coherence  of evidence across a spectrum of related
28                   endpoints (e.g., health effects ranging from inflammatory effects to respiratory mortality)
29                   to draw conclusions regarding causality. In discussing the causal determination, EPA
30                   characterizes the evidence on which the judgment is based, including strength of
31                   evidence for individual endpoints within the major outcome category.
        4 It should be noted that the Center for Disease Control (CDC) and IOM frameworks use a four-category hierarchy for the strength
      of the evidence. A five-level hierarchy is used here to be consistent with the EPA Guidelines for Carcinogen Risk Assessment and to
      provide a more nuanced set of categories.
      Draft - Do Not Cite or Quote                        Ixi                                  September 2011

-------
      Table II
  Weight of evidence for causal determination
                                      Health Effects
                                                          Ecological and Welfare Effects
      Causal        Evidence is sufficient to conclude that there is a causal
      relationship    relationship with relevant pollutant exposures (i.e., doses
                    or exposures generally within one to two orders of
                    magnitude of current levels). That is, the pollutant has
                    been shown to result in health effects in studies in which
                    chance, bias, and confounding could be ruled out with
                    reasonable confidence. For example: a) controlled
                    human exposure studies that demonstrate consistent
                    effects; or b) observational studies that cannot be
                    explained by plausible alternatives or are supported by
                    other lines of evidence (e.g., animal studies or mode of
                    action information). Evidence includes replicated and
                    consistent high-quality studies by multiple investigators.
                                                 Evidence is sufficient to conclude that there is a causal
                                                 relationship with relevant pollutant exposures i.e., doses
                                                 or exposures generally within one to two orders of
                                                 magnitude  of current levels). That is, the pollutant has
                                                 been shown to result in effects in studies in which
                                                 chance, bias, and confounding could be ruled out with
                                                 reasonable confidence. Controlled exposure studies
                                                 (laboratory or small- to medium-scale field studies)
                                                 provide the strongest evidence for causality, but the
                                                 scope of inference may be limited. Generally,
                                                 determination is based on multiple studies conducted by
                                                 multiple research groups, and evidence that is considered
                                                 sufficient to infer a causal  relationship is usually obtained
                                                 from the joint consideration of many lines of evidence that
                                                 reinforce each other.
      Likely to be a  Evidence is sufficient to conclude that a causal
      causal        relationship is likely to exist with relevant pollutant
      relationship    exposures, but important uncertainties remain. That is,
                    the pollutant has been shown to result in health effects in
                    studies in which chance and bias can be ruled out with
                    reasonable confidence but potential issues remain. For
                    example: a) observational studies show an association,
                    but copollutant exposures are difficult to address and/or
                    other lines of evidence (controlled human exposure,
                    animal, or mode of action information) are limited or
                    inconsistent; or b)  animal toxicological evidence from
                    multiple studies from different laboratories that
                    demonstrate effects, but limited or no human data are
                    available. Evidence generally includes replicated and
     	high-quality studies by multiple investigators.	
                                                 Evidence is sufficient to conclude that there is a likely
                                                 causal association with relevant pollutant exposures. That
                                                 is, an association has been observed between the
                                                 pollutant and the outcome in studies in which chance,
                                                 bias and confounding are minimized, but uncertainties
                                                 remain. For example, field studies show a relationship,
                                                 but suspected interacting factors cannot be controlled,
                                                 and other lines of evidence are limited or inconsistent.
                                                 Generally, determination is based on multiple studies in
                                                 multiple research groups.
      Suggestive of  Evidence is suggestive of a causal relationship with
      a causal       relevant pollutant exposures, but is limited. For example,
      relationship    (a) at least one high-quality epidemiologic study shows
                    an association with a given health outcome but the
                    results of other studies are inconsistent; or (b) a well-
                    conducted toxicological study, such as those conducted
                    in the National Toxicology Program (NTP), shows effects
     	in animal species.	
                                                 Evidence is suggestive of a causal relationship with
                                                 relevant pollutant exposures, but chance, bias and
                                                 confounding cannot be ruled out. For example, at least
                                                 one high-quality study shows an effect, but the results of
                                                 other studies are inconsistent.
      Inadequate to  Evidence is inadequate to determine that a causal
      infer a causal  relationship exists with relevant pollutant exposures. The
      relationship    available studies are of insufficient quantity, quality,
                    consistency or statistical power to permit a conclusion
     	regarding the presence or absence of an effect.	
                                                The available studies are of insufficient quality,
                                                consistency or statistical power to permit a conclusion
                                                regarding the presence or absence of an effect.
      Not likely to    Evidence is suggestive of no causal relationship with
      be a causal    relevant pollutant exposures. Several adequate studies,
      relationship    covering the full range of levels of exposure that human
                    beings are known to encounter and considering at-risk
                    populations, are mutually consistent in not showing an
     	effect at any level of exposure.	
                                                 Several adequate studies, examining relationships with
                                                 relevant exposures, are consistent in failing to show an
                                                 effect at any level of exposure.
1
2
3
4
5
6
In drawing judgments regarding causality for the criteria air pollutants, the ISA focuses
on evidence of effects in the range of relevant pollutant exposures or doses, and not on
determination of causality at any dose. Emphasis is placed on evidence of effects at doses
(e.g., blood lead concentration) or exposures (e.g., air concentrations) that are relevant to,
or somewhat above, those currently experienced by the population. The extent to which
studies of higher concentrations are considered varies by pollutant and major outcome
      Draft - Do Not Cite or Quote
                                       Ixii
September 2011

-------
 1                   category, but generally includes those with doses or exposures in the range of one to two
 2                   orders of magnitude above current or ambient conditions. Studies that use higher doses or
 3                   exposures may also be considered to the extent that they provide useful information to
 4                   inform our understanding of mode of action, interspecies differences or factors that may
 5                   increase risk of effects for a population. Thus, a causality determination is based on
 6                   weight of evidence evaluation for health, ecological or welfare effects, focusing on the
 7                   evidence from exposures or doses generally ranging from current levels to one or two
 8                   orders of magnitude above current levels.

 9                   In addition, EPA evaluates evidence relevant to understand the quantitative relationships
10                   between pollutant exposures and health, ecological or welfare effects. This includes
11                   evaluation of the form of concentration-response or dose-response relationships and, to
12                   the extent possible, drawing conclusions on the levels at which effects are observed.  The
13                   ISA also draws scientific conclusions regarding important exposure conditions for effects
14                   and populations that may be at greater risk for effects, as described in the following
15                   section.
                     Quantitative relationships:  Effects on Human  Populations

16                   Once a determination is made regarding the causal relationship between the pollutant and
17                   outcome category, important questions regarding quantitative relationships include:
18                       •  What is the concentration-response, exposure-response, or dose-response
19                         relationship in the human population?
20                       •  What is the interrelationship between incidence and severity of effect?
21                       •  What exposure conditions (dose or exposure, duration and pattern) are
22                         important?
23                       •  What populations and lifestages appear to be differentially affected (i.e., more
24                         at risk of experiencing effects)?

25                   To address these questions, the entirety of quantitative evidence is evaluated to
26                   characterize pollutant concentrations and exposure durations at which effects were
27                   observed for exposed populations, including populations and lifestages potentially at
28                   increased risk. To accomplish this, evidence is considered from multiple and diverse
29                   types of studies, and a study or set of studies that best approximates the concentration-
30                   response relationships between health outcomes and the pollutant may be identified.
31                   Controlled human exposure studies provide the most direct and quantifiable exposure-
32                   response data on the human health effects of pollutant exposures. To the extent available,
33                   the ISA evaluates results from across epidemiologic studies that use various methods to
      Draft - Do Not Cite or Quote                       Ixiii                                 September 2011

-------
 1                   characterize the form of relationships between the pollutant and health outcomes and
 2                   draws conclusions on the shape of these relationships. Animal data may also inform
 3                   evaluation of concentration-response relationships, particularly relative to MOAs and
 4                   characteristics of susceptible populations.

 5                   An important consideration in characterizing the public health impacts associated with
 6                   exposure to a pollutant is whether the concentration-response relationship is linear across
 7                   the range of concentrations or if nonlinear relationships exist along any part of this range.
 8                   Of particular interest is the shape of the concentration-response curve at and below the
 9                   level of the current standards. Various sources of variability and uncertainty, such as low
10                   data density in the lower concentration range, possible influence of exposure
11                   measurement error, and variability between individuals in susceptibility to air pollution
12                   health effects, tend to smooth and "linearize" the concentration-response function, and
13                   thus can obscure the  existence of a threshold or nonlinear relationship. Since individual
14                   thresholds vary from person to person due to individual differences such as genetic level
15                   susceptibility or preexisting disease conditions (and even can vary from one time to
16                   another for a given person), it can be difficult to demonstrate that a threshold exists in a
17                   population study. These sources of variability and uncertainty may explain why the
18                   available human data at ambient concentrations for some environmental pollutants
19                   (e.g., particulate matter [PM], O3, lead [Pb], environmental tobacco smoke [ETS],
20                   radiation) do not exhibit thresholds for cancer or noncancer health effects, even though
21                   likely mechanisms include nonlinear processes for some key events. These attributes of
22                   human population dose-response relationships have been extensively discussed  in the
23                   broader epidemiologic  literature (Rothman and Greenland. 1998).

24                   Finally, identification of the population groups or lifestages that may be at greater risk of
25                   health effects from air pollutant exposures contributes to an understanding of the public
26                   health impact of pollutant exposures. In the ISA, the term "at-risk population" is used to
27                   encompass populations variously described as susceptible, vulnerable, or sensitive. "At-
28                   risk populations" is defined here as those populations or lifestages that have a greater
29                   likelihood of experiencing health effects related to exposure to an air pollutant due to a
30                   variety of factors. These factors may be intrinsic, such as genetic or developmental
31                   factors, race, gender, lifestage, or the presence of preexisting diseases, or they may be
32                   extrinsic, such as socioeconomic status (SES), activity pattern and exercise level, reduced
33                   access to health care, low educational attainment, or increased pollutant exposures (e.g.,
34                   near roadways). Epidemiologic studies can help identify populations potentially at
35                   increased risk of effects by evaluating health responses in the study population.  Examples
36                   include testing for interactions or effect modification by factors such as gender, age
37                   group, or health status. Experimental studies using animal models of susceptibility or
      Draft - Do Not Cite or Quote                        Ixiv                                  September 2011

-------
 1                  disease can also inform the extent to which health risks are likely greater in specific
 2                  population groups.
                    Quantitative relationships:  Effects on Ecosystems or
                    Public Welfare

 3                  Key questions for understanding the quantitative relationships between exposure (or
 4                  concentration or deposition) to a pollutant and risk to ecosystems or the public welfare
 5                  include:

 6                      •  What elements of the ecosystem (e.g., types, regions, taxonomic groups,
 7                         populations, functions, etc.) appear to be affected, or are more sensitive to
 8                         effects? Are there differences between locations or materials in welfare effects
 9                         responses, such as impaired visibility or materials damage?
10                      •  Under what exposure conditions (amount deposited or concentration, duration
11                         and pattern) are effects seen?
12                      •  What is the shape of the concentration-response or exposure-response
13                         relationship?

14                  Evaluations of causality generally consider the probability of quantitative changes in
15                  ecological and welfare effects in response to exposure. A challenge to the quantification
16                  of exposure-response relationships for ecological effects is the great regional and local
17                  variability in ecosystems. Thus, exposure-response relationships are often determined for
18                  a specific ecological system and scale, rather than at the national or even regional scale.
19                  Quantitative relationships therefore are available site by site and may differ greatly
20                  between ecosystems.

            Concepts in Evaluating Adversity of Health  Effects
21                  In evaluating health evidence, a number of factors can be considered in delineating
22                  between adverse and nonadverse health effects resulting from exposure to air pollution.
23                  Some health outcomes, such as hospitalization for respiratory or cardiovascular diseases,
24                  are clearly considered adverse. It is more difficult to determine the extent of change that
25                  constitutes adversity in more subtle health measures. These include a wide  variety of
26                  responses, such as alterations in markers of inflammation or oxidative stress, changes in
27                  pulmonary function or heart rate variability, or alterations in neurocognitive function
28                  measures. The challenge is determining the magnitude of change in these measures when
29                  there is no clear point at which a change become adverse; for example, what percentage
30                  change in a lung function measure represents an adverse effect. What constitutes an
      Draft - Do Not Cite or Quote                       Ixv                                September 2011

-------
 1                   adverse health effect may vary between populations. Some changes that may not be
 2                   considered adverse in healthy individuals would be potentially adverse in more
 3                   susceptible individuals.

 4                   For example, the extent to which changes in lung function are adverse has been discussed
 5                   by the American Thoracic Society (ATS) in an official statement titled What Constitutes
 6                   an Adverse Health Effect of Air Pollution? (2000b). An air pollution-induced shift in the
 7                   population distribution of a given risk factor for a health outcome was viewed as adverse,
 8                   even though it may not increase the risk of any one individual to an unacceptable level.
 9                   For example, a population of asthmatics could have a distribution of lung function such
10                   that no identifiable individual has a level associated with significant impairment.
11                   Exposure to air pollution could shift the distribution such that no identifiable individual
12                   experiences clinically relevant effects. This shift toward decreased  lung function,
13                   however, would be considered adverse because individuals within the population would
14                   have diminished reserve function and therefore would be at increased risk to further
15                   environmental insult. The committee also observed that elevations of biomarkers, such as
16                   cell number and types, cytokines and reactive oxygen species, may signal risk for ongoing
17                   injury and clinical effects or may simply indicate transient responses that can provide
18                   insights into mechanisms of injury, thus illustrating the lack of clear boundaries that
19                   separate adverse from nonadverse effects.

20                   It is important to recognize that the more subtle health outcomes may be connected
21                   mechanistically to health events that are clearly adverse. For example, air pollution may
22                   affect markers of transient myocardial ischemia such as ST-segment abnormalities and
23                   onset of exertional angina. These effects may not be apparent to the individual, yet may
24                   still increase the risk of a number of cardiac events, including myocardial infarction and
25                   sudden death. Thus, small changes in physiological measures may not appear to be
26                   clearly adverse when considered alone, but contribute to a coherent and biologically
27                   plausible group of related health outcomes, including responses that are very clearly
28                   adverse.

            Concepts in Evaluating Adversity of  Ecological Effects
29                   Adversity of ecological effects can be understood in terms ranging  in scale from the
30                   cellular level to the individual organism and to the population, community and ecosystem
31                   levels. In the context of ecology,  a population is a group of individuals of the same
32                   species, and a community is an assemblage of populations of different species interacting
33                   with one another that inhabit an area. An ecosystem is the interactive system formed from
34                   all living organisms and their abiotic (physical and  chemical) environment within a given
35                   area (IPCC, 2007a). The boundaries of what could be called an ecosystem are somewhat
      Draft - Do Not Cite or Quote                       Ixvi                                September 2011

-------
 1                   arbitrary, depending on the focus of interest or study. Thus, the extent of an ecosystem
 2                   may range from very small spatial scales to, ultimately, the entire Earth (IPCC. 2007a).

 3                   Effects on an individual organism are generally not considered to be adverse, however if
 4                   effects occur to enough individuals within a population, communities and ecosystems
 5                   may be disrupted. Changes to populations, communities and ecosystems can in turn result
 6                   in an alteration of ecosystem processes. Ecosystem processes are defined as the metabolic
 7                   functions of ecosystems including energy flow, elemental cycling, and the production,
 8                   consumption and decomposition of organic matter (U.S. EPA. 2002). Growth,
 9                   reproduction, and mortality are species-level endpoints that can be clearly linked to
10                   community and ecosystem effects and are considered to be adverse when negatively
11                   affected. Other endpoints such as changes in behavior and physiological stress can
12                   decrease ecological fitness of an organism, but are harder to link unequivocally to effects
13                   at the population, community and ecosystem level. The degree to which pollutant
14                   exposure is considered adverse may also depend on the location and its intended use (i.e.
15                   city park, commercial cropland). Support for consideration of adversity beyond the
16                   species level by making explicit the linkages between stress-related effects at the species
17                   and effects at the ecosystem level is found in A Framework for Assessing and Reporting
18                   on Ecological Condition: an SAB report (U.S. EPA. 2002). Additionally, the National
19                   Acid Precipitation Assessment Program (NAPAP) uses the following working definition
20                   of adverse ecological effects in the preparation of reports to Congress mandated by the
21                   Clean Air Act: "any injury (i.e. loss of chemical or physical quality or viability) to any
22                   ecological or ecosystem component, up to and including at the regional level, over both
23                   long and short terms."

24                   On a broader scale, ecosystem services may provide indicators for ecological impacts.
25                   Ecosystem services are the benefits that people obtain from ecosystems (UNEP. 2003).
26                   According to the Millennium Ecosystem Assessment, ecosystem services include:
27                   "provisioning services such as food and water; regulating services such as regulation of
28                   floods, drought, land degradation, and disease; supporting services such as soil formation
29                   and nutrient cycling; and cultural services such as recreational, spiritual, religious and
30                   other nonmaterial benefits." For example, a more subtle ecological effect of pollution
31                   exposure may result in a clearly adverse impact on ecosystem services if it results in a
32                   population decline in a species that is recreationally or culturally important.
      Draft - Do Not Cite or Quote                        Ixvii                                 September 2011

-------
      PREFACE


            Legislative Requirements for the NAAQS  Review
 1                   Two sections of the Clean Air Act (CAA) govern the establishment and revision of
 2                   the National Ambient Air Quality Standards (NAAQS). Section 108 (42 USC §7408)
 3                   directs the Administrator to identify and list certain air pollutants and then to issue air
 4                   quality criteria for those pollutants. The Administrator is to list those air pollutants
 5                   that in her "judgement; cause or contribute to air pollution which may reasonably be
 6                   anticipated to endanger public health or welfare" and whose "presence... in the
 7                   ambient air results from numerous or diverse mobile or stationary sources" (CAA.
 8                   1990a). Air quality criteria are intended to "accurately reflect the latest scientific
 9                   knowledge useful in indicating the kind and extent of identifiable effects on public
10                   health or welfare which may be expected from the presence of [a] pollutant in
11                   ambient air ... [42 USC §7408(b)].

12                   Section 109 (CAA. 1990b) directs the Administrator to propose and promulgate
13                   "primary" and "secondary" NAAQS for pollutants for which air quality criteria have
14                   been issued. Section 109(b)(l)  defines a primary standard as one "the attainment and
15                   maintenance of which in the j udgment of the Administrator, based on such criteria
16                   and allowing an adequate margin of safety, are requisite to protect the public
17                   health."5 A secondary standard, as defined in section 109(b)(2), must "specify a level
18                   of air quality the attainment and maintenance of which, in the j udgment of the
19                   Administrator, based on such criteria, is required to protect the public welfare from
20                   any known or anticipated adverse effects associated with the presence of [the]
21                   pollutant in the ambient air."6

22                   The requirement that primary standards include  an adequate margin of safety was
23                   intended to address uncertainties associated with inconclusive scientific and technical
24                   information available at the time of standard setting. It was also intended to provide a
25                   reasonable degree  of protection against hazards  that research has not yet identified.
26                   See Lead Industries Association v. EPA, 647 F.2d  1130, 1154 (D.C. Cir 1980), cert.
27                   denied, 449 U.S. 1042 (1980); American Petroleum Institute v. Costle, 665 F.2d
28                   1176, 1186 (D.C. Cir. 1981), cert, denied, 455 U.S. 1034 (1982). Both kinds of
29                   uncertainties are components of the risk associated with pollution at levels below
        5 The legislative history of section 109 indicates that a primary standard is to be set at "the maximum permissible ambient air level
      . . . which will protect the health of any [sensitive] group of the population," and that for this purpose "reference should be made to a
      representative sample of persons comprising the sensitive group rather than to a single person in such a group" [S. Rep. No. 91-
      1196, 91s Cong., 2d Sess. 10 (1970)].
        6 Welfare effects as defined in section 302(h) include, but are not limited to, "effects on soils, water, crops, vegetation, man-made
      materials, animals, wildlife, weather, visibility and climate, damage to and deterioration of property, and hazards to transportation, as
      well as effects on economic values and on personal comfort and well-being" (CAA,  2005).
      Draft - Do Not Cite or Quote                       Ixviii                                September 2011

-------
 1                  those at which human health effects can be said to occur with reasonable scientific
 2                  certainty. Thus, in selecting primary standards that include an adequate margin of
 3                  safety, the Administrator is seeking not only to prevent pollution levels that have
 4                  been demonstrated to be harmful but also to prevent lower pollutant levels that may
 5                  pose an unacceptable risk of harm, even if the risk is not precisely identified as to
 6                  nature or degree.

 7                  In selecting a margin of safety, the EPA considers such factors as the nature and
 8                  severity of the health effects involved, the size of the sensitive population(s) at risk,
 9                  and the  kind and degree of the uncertainties that must be addressed. The selection of
10                  any particular approach to providing an adequate margin of safety is a policy choice
11                  left specifically to the Administrator's judgment. See Lead Industries Association v.
12                  EPA, supra, 647 F.2d at 1161-1162.

13                  In setting standards that are "requisite" to protect public health and  welfare, as
14                  provided in Section 109(b), EPA's task is to establish standards that are neither more
15                  nor less stringent than necessary. In so doing, EPA may not consider the costs of
16                  implementing the standards. [See generally Whitman v. American Trucking
17                  Associations, 531 U.S.  457, 465-472, 475-76.]

18                  Section 109(d)(l) requires that "not later than December 31,  1980, and at 5-year
19                  intervals thereafter, the Administrator shall complete a thorough review of the criteria
20                  published under section 108 and the national ambient air quality standards ... and
21                  shall make such revisions in such criteria and standards and promulgate such new
22                  standards as may be appropriate..." Section 109(d)(2)  requires that  an independent
23                  scientific review committee "shall  complete a review of the criteria ... and the
24                  national primary and secondary  ambient air quality standards ...  and shall
25                  recommend to the Administrator any new . . . standards and revisions of existing
26                  criteria and standards as may be appropriate ..." Since the early 1980s, this
27                  independent review function has been performed by CASAC.

            History of the NAAQS for  Ozone
28                  Tropospheric (ground-level) O3  is the indicator for the mix of photochemical
29                  oxidants (e.g., peroxyacetyl nitrate, hydrogen peroxide) formed from biogenic and
30                  anthropogenic precursor emissions. Naturally occurring O3 in the troposphere can
31                  result from biogenic organic precursors reacting with naturally occurring nitrogen
32                  oxides (NOX) and by stratospheric O3 intrusion into the troposphere. Anthropogenic
33                  precursors of O3, especially NOX, and volatile organic  compounds (VOCs), originate
34                  from a wide variety of stationary and  mobile sources. Ambient O3 concentrations
      Draft - Do Not Cite or Quote                       Ixix                                September 2011

-------
 1
 2

 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
produced by these emissions are directly affected by temperature, solar radiation,
wind speed, and other meteorological factors.

NAAQS are comprised of four basic elements: indicator, averaging time, level, and
form. The indicator defines the pollutant to be measured in the ambient air for the
purpose of determining compliance with the standard. The averaging time defines the
time period over which air quality measurements are to be obtained and averaged or
cumulated, considering evidence of effects associated with various time periods of
exposure. The level of a standard defines the air quality concentration used (i.e., an
ambient concentration  of the indicator pollutant) in determining whether the standard
is achieved. The form of the standard specifies the air quality measurements that are
to be used for compliance purposes (e.g., the annual fourth-highest daily maximum
8-hour concentration, averaged over 3 years), and whether the statistic is to be
averaged across multiple years. These four elements taken together determine the
degree of public health and welfare protection afforded by the NAAQS.
      Table
  Summary of primary and secondary NAAQS promulgated for ozone
  during the period 1971-2008
Final Rule
1971 (36 FR 81 86)
1979 (44 FR 8202)
1 993 (58 FR 13008)
1997 (62 FR 38856)
2008 (73 FR 16483)
Indicator
Total photochemical
oxidants
03
Avg
Time
1-h
1-h
Level
(ppm)
0.08
0.12
EPA decided that revisions to the standards were
03
03
8-h
8-h
0.08
0.075
Form
Not to be exceeded more than 1 hour per year
Attainment is defined when the expected number of days per
calendar year, with maximum hourly average concentration greater
than 0.1 2 ppm, is < 1
not warranted at the time.
Annual fourth-highest daily maximum 8-h concentration averaged
over 3 years
Form of the standards remained unchanged relative to the 1997
standard
15
16
17

18
19
20
21
22

23
24
Table III summarizes the O3 NAAQS that have been promulgated to date. In each
review, the secondary standard has been set to be identical to the primary standard.
These reviews are briefly described below.

EPA first established primary and secondary NAAQS for photochemical oxidants in
1971 . Both primary and secondary standards were set at a level of 0.08 parts per
million (ppm), 1 -h avg, total photochemical  oxidants, not to be exceeded more than
1 hour per year. The standards were based on scientific information contained in the
1970 AQCD.

In 1977, EPA announced the first periodic review of the 1970 AQCD in accordance
with Section 109(d)(l) of the Clean Air Act.  In 1978, EPA published an AQCD.
      Draft - Do Not Cite or Quote
                              Ixx
September 2011

-------
 1                  Based on the 1978 AQCD, EPA published proposed revisions to the original NAAQS
 2                  in 1978 (U.S. EPA. 1978b) and final revisions in 1979 (U.S. EPA. 1979a). The level
 3                  of the primary and secondary standards was revised from 0.08 to 0.12 ppm; the
 4                  indicator was revised from photochemical oxidants to O3; and the form of the
 5                  standards was revised from a deterministic to a statistical form, which defined
 6                  attainment of the standards as occurring when the expected number of days per
 7                  calendar year with maximum hourly average concentration greater  than 0.12 ppm is
 8                  equal to  or less than one.

 9                  In 1982, EPA announced plans to revise the 1978 AQCD (U.S. EPA. 1978a). In 1983,
10                  EPA announced that the second periodic review of the primary and secondary
11                  standards for O3 had been initiated (U.S. EPA. 1983). EPA subsequently published
12                  the 1986 O3 AQCD (U.S. EPA. 1986) and 1989  Staff Paper (U.S. EPA. 1989).
13                  Following publication of the 1986 O3 AQCD, a number of scientific abstracts and
14                  articles were published that appeared to be of sufficient importance concerning
15                  potential health and welfare effects  of O3 to warrant preparation of a Supplement to
16                  the 1986 O3 AQCD (Costa et al.  1992). Under the terms of a court order, on August
17                  10, 1992, EPA published a proposed decision (U.S. EPA. 1992) stating that revisions
18                  to the existing primary and secondary standards were not appropriate at the time
19                  (U.S. EPA. 1992). This notice explained that the proposed decision would complete
20                  EPA's review of information on health and welfare effects of O3 assembled over a
21                  7-year period and contained in the 1986 O3 AQCD (U.S. EPA. 1986) and its
22                  Supplement to the  1986 O3 AQCD (Costa et al.. 1992). The proposal also announced
23                  EPA's intention to proceed as rapidly as possible with the next review of the air
24                  quality criteria  and standards for O3 in light of emerging evidence of health effects
25                  related to 6- to  8-hour O3 exposures. On March 9, 1993, EPA concluded the review
26                  by deciding that revisions to the standards were  not warranted at that time (U.S. EPA.
27                  1993).

28                  In August 1992, EPA announced plans to initiate the third periodic review of the air
29                  quality criteria  and O3 NAAQS (U.S. EPA. 1992). On the basis of the scientific
30                  evidence contained in the 1996 O3 AQCD and the 1996 Staff Paper (U.S. EPA.
31                  1996e). and related technical support documents, linking exposures to ambient O3 to
32                  adverse health and welfare effects at levels allowed by the then existing standards,
33                  EPA proposed to revise the primary and secondary O3 standards on December 13,
34                  1996 (U.S. EPA. 1996d). The EPA proposed to replace the then existing 1-hour
35                  primary  and secondary standards  with 8-h avg O3 standards set at a level of 0.08 ppm
36                  (equivalent to 0.084 ppm using standard rounding conventions). The EPA also
37                  proposed, in the alternative, to establish a new distinct secondary standard using a
38                  biologically based  cumulative seasonal form. The EPA completed the review on July
      Draft - Do Not Cite or Quote                      Ixxi                               September 2011

-------
 1                  18, 1997 by setting the primary standard at a level of 0.08 ppm, based on the annual
 2                  fourth-highest daily maximum 8-h avg concentration, averaged over 3 years, and
 3                  setting the secondary standard identical to the revised primary standard (U.S. EPA.
 4                  1997).

 5                  On May 14, 1999, in response to challenges to EPA's 1997 decision by industry and
 6                  others, the U.S. Court of Appeals for the District of Columbia Circuit (D.C. Cir.)
 7                  remanded the O3 NAAQS to EPA,  finding that Section 109 of the CAA, as
 8                  interpreted by EPA, effected an unconstitutional delegation of legislative authority. In
 9                  addition, the D.C.  Cir. directed that, in responding to the remand, EPA should
10                  consider the potential beneficial health effects of O3 pollution in shielding the public
11                  from the effects of solar ultraviolet (UV) radiation, as well as adverse health effects.
12                  On January 27, 2000, EPA petitioned the U.S. Supreme Court for certiorari on the
13                  constitutional issue (and two other issues) but did not request review of the D.C. Cir.,
14                  ruling regarding the potential beneficial health effects of O3.  On February 27, 2001,
15                  the U.S. Supreme  Court unanimously reversed the judgment of the D.C. Cir. on the
16                  constitutional issue, holding that Section 109 of the CAA does not delegate
17                  legislative power to the EPA in  contravention of the Constitution, and remanded the
18                  case to the D.C. Cir. to consider challenges to the O3 NAAQS that had not been
19                  addressed by that Court's earlier decisions. On March 26,  2002, the D.C. Cir. issued
20                  its final  decision, finding the 1997  O3 NAAQS to be "neither arbitrary nor
21                  capricious," and denied the remaining petitions for review. On November 14, 2001,
22                  in response to the D.C. Cir. remand to consider the potential  beneficial health effects
23                  of O3 pollution in shielding the  public from effects of solar (UV) radiation, EPA
24                  proposed to leave the 1997 8-h  O3 NAAQS unchanged (U.S.  EPA.  2001). After
25                  considering public comment on the proposed decision, EPA published its final
26                  response to this remand on January 6, 2003, reaffirming the 8-h O3 NAAQS set in
27                  1997 (U.S. EPA. 2003). On April 30, 2004, EPA announced the decision to make the
28                  1-h O3 NAAQS no longer applicable to areas 1 year after the effective date of the
29                  designation of those areas for the 8-h NAAQS (2004). For most areas, the date that
30                  the 1-h NAAQS no longer applied was June 15,  2005.

31                  EPA initiated the next periodic review if the air quality criteria and O3 standards in
32                  September 2000 with a call for information (U.S. EPA. 2000). The schedule for
33                  completion of that rulemaking later became governed by a consent decree resolving a
34                  lawsuit filed in March 2003 by  a group of plaintiffs representing national
35                  environmental and public health organizations. Based on the 2006 O3 AQCD (U.S.
36                  EPA. 2006b) published in March 2006, the Staff Paper (U.S. EPA. 2007b) and related
37                  technical support documents, the proposed decision was published in the Federal
38                  Register on July 11, 2007 (U.S. EPA. 2007a). The EPA proposed to revise the level of
      Draft - Do Not Cite or Quote                      Ixxii                               September 2011

-------
 1                   the primary standard to a level within the range of 0.075 to 0.070 ppm. Two options
 2                   were proposed for the secondary standard: (1) replacing the current standard with a
 3                   cumulative, seasonal standard, expressed as an index of the annual sum of weighted
 4                   hourly concentrations cumulated over 12 daylight hours during the consecutive
 5                   3-month period within the O3 season with the maximum index value, set at a level
 6                   within the range of 7 to 21 ppm-h; and (2) setting the secondary standard identical to
 7                   the revised primary standard. The EPA completed the rulemaking with publication of
 8                   a final decision on March 27, 2008 (U.S. EPA. 2008eX revising the level of the
 9                   8-hour primary O3 standard from 0.08 ppm to 0.075 ppm and revising the secondary
10                   standard to be identical to the primary standard.


            References

      ATS. (American Thoracic Society). (2000b). What constitutes an adverse health effect of air pollution? This
             official statement of the American Thoracic Society was adopted by the ATS Board of Directors, July
             1999. Am J Respir Grit Care Med 161: 665-673.
      Bell.  ML: Dominici. F: Samet.  JM. (2005). A meta-analysis of time-series studies of ozone and mortality with
             comparison to the national morbidity, mortality, and air pollution study. Epidemiology 16: 436-445.
      CAA. (Air quality criteria and control techniques, Section 108 of the Clean Air Act). 42. § 7408, (1990a).
      CAA. (National primary  and secondary ambient air quality standards, Section 109 of the Clean Air Act). 42. §
             7409, (1990b).
      CAA. (Definitions. Section 302 of the Clean Air Act). 42. § 7602, (2005).
      Costa. PL: Folinsbee. LJ: Raub. JA: Tilton.  B: Tingev. DT (1992). Summary of selected new information on
             effects of ozone on health and vegetation: Supplement to 1986 air quality criteria for ozone and other
             photochemical  oxidants. (EPA/600/8-88/105F). Research Triangle Park,  NC: U.S. Environmental
             Protection Agency.
      Fox,  GA. (1991). Practical causal inference for ecoepidemiologists. J Toxicol Environ Health A 33: 359-373.
             http://dx.doi. orq/10.1080/15287399109531535.
      Gee. GC: Pavne-Sturges. DC. (2004). Environmental health disparities: A framework integrating psychosocial
             and environmental concepts. Environ Health Perspect 112: 1645-1653.
             http://dx.doi.org/10.1289/ehp.7074.
      HEW. (U.S. Department of Health, Education and Welfare). (1964). Smoking and health: Report of the advisory
             committee to the surgeon general of the public health service. Washington, DC: U.S. Department of
             Health, Education, and Welfare.
      HHS. (U.S. Department of Health and Human Services, Office of the Surgeon  General). (2004). The health
             consequences  of smoking: A report of the Surgeon General.  Washington, DC: U.S. Department of
             Health and Human Services, http://www.suraeonaeneral.gov/librarv/smokinaconseauences/.
      Hill. AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.
      IARC. (International Agency for Research on Cancer). (2006). Preamble to the IARC monographs. Lyon,
             France.  http://monographs.iarc.fr/ENG/Preamble/.
      loannidis. JPA. (2008). Why most discovered true associations are inflated. Epidemiology 19: 640-648.
             http://dx.doi. org/10.1097/EDE.Ob013e31818131 e7.
      IPCC. (Intergovernmental Panel on Climate Change). (2007a). Summary for policymakers.  In: Impacts,
             Adaptation and Vulnerability. Contribution of Working Group  II to the Fourth Assessment Report of the
             Intergovernmental Panel on Climate Change. In Climate Change 2007. Cambridge, UK: Cambridge
             University Press.
      Rothman. KJ:  Greenland. S. (1998). Modern epidemiology.  In (2nd ed.). Philadelphia, PA: Lippincott, Williams,
             SWilkins.
      Draft - Do Not Cite or Quote                        Ixxiii                                 September 2011

-------
Samet. JM: Bodurow. CC. (2008). Improving the presumptive disability decision-making process for veterans. In
        JM Samet; CC Bodurow (Eds.). Washington, DC: National Academies Press.
U.S. EPA. (U.S. Environmental Protection Agency). (1971). National primary and secondary ambient air quality
        standards. Fed Reg 36: 8186-8201.
U.S. EPA. (U.S. Environmental Protection Agency). (1978a).  Air quality criteria for ozone and other
        photochemical oxidants. (EPA/600/8-78/004). Washington, DC.
U.S. EPA. (U.S. Environmental Protection Agency). (1978b).  Photochemical oxidants: Proposed revisions to the
        national ambient air quality standards. Fed Reg 43: 26962-26971.
U.S. EPA. (U.S. Environmental Protection Agency). (1979a).  National primary and secondary ambient air quality
        standards: Revisions to the national ambient air quality standards for photochemical oxidants. Fed Reg
        44: 8202-8237.
U.S. EPA. (U.S. Environmental Protection Agency). (1982). Air quality criteria document for ozone and other
        photochemical oxidants. Fed Reg 47: 11561.
U.S. EPA. (U.S. Environmental Protection Agency). (1983). Review of the national ambient air quality standards
        for ozone. Fed Reg 48: 38009.
U.S. EPA. (U.S. Environmental Protection Agency). (1986). Air quality criteria for ozone and other
        photochemical oxidants. (EPA-600/8-84-020aF - EPA-600/8-84-020eF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1989). Review of the national ambient air quality standards
        for ozone: Assessment of scientific and technical information: OAQPS staff report. (EPA/450/2-92-001).
        Research Triangle Park, NC. http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=2000LOW6.txt.
U.S. EPA. (U.S. Environmental Protection Agency). (1992). National ambient air quality standards for ozone;
        Proposed decision. Fed Reg 57: 35542-35557.
U.S. EPA. (U.S. Environmental Protection Agency). (1993). National ambient air quality standards for ozone -
        Final decision. Fed Reg 58:  13008-13019.
U.S. EPA. (U.S. Environmental Protection Agency). (1996d).  National ambient air quality standards for ozone:
        Proposed decision. Fed Reg 61: 65716-65750.
U.S. EPA. (U.S. Environmental Protection Agency). (1996e).  Review of national ambient air quality standards
        for ozone: Assessment of scientific and technical information: OAQPS staff paper. (EPA/452/R-96/007).
        Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1997). National ambient air quality standards for ozone;
        final rule. Fed Reg 62: 38856-38896.
U.S. EPA. (U.S. Environmental Protection Agency). (1998). Guidelines for ecological risk assessment.
        (EPA/630/R-95/002F). Washington, DC. http://www.epa.gov/raf/publications/guidelines-ecological-risk-
        assessment.htm.
U.S. EPA. (U.S. Environmental Protection Agency). (2000). Air quality criteria for ozone and related
        photochemical oxidants. Fed Reg 65: 57810.
U.S. EPA. (U.S. Environmental Protection Agency). (2001). National ambient air quality standards for ozone:
        Proposed response to remand.  Fed Reg 66: 57268-57292.
U.S. EPA. (U.S. Environmental Protection Agency). (2002). A framework for assessing and reporting on
        ecological condition: An SAB report. Washington, DC.
U.S. EPA. (U.S. Environmental Protection Agency). (2003). National ambient air quality standards for ozone:
        Final response to remand. Fed Reg 68: 614-645.
U.S. EPA. (U.S. Environmental Protection Agency). (2004). Final rule to implement the 8-hour ozone national
        ambient air quality standard-phase 1. Fed Reg 69: 23951-24000.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment.
        (EPA/630/P-03/001F). Washington, DC. http://www.epa.aov/cancerguidelines/.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b).  Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park,  NC: U.S. Environmental
        Protection Agency,  Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2007a).  National ambient air quality standards for ozone.
        Fed Reg 72: 37818-37919.
Draft - Do Not Cite or Quote                       Ixxiv                                  September 2011

-------
U.S. EPA. (U.S. Environmental Protection Agency). (2007b). Review of the national ambient air quality
       standards for ozone: Policy assessment of scientific and technical information: OAQPS staff paper.
       (EPA/452/R-07/003). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2008e). National ambient air quality standards for ozone.
       Fed Reg 73: 16436-16514.
U.S. EPA. (U.S. Environmental Protection Agency). (2009d). Integrated science assessment for particulate
       matter. (EPA/600/R-08/139F). Research Triangle Park, NC.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
UNEP. (United Nations Environment Programme). (2003). Millennium Ecosystem Assessment: Ecosystems and
       human well-being: A framework for assessment. Washington, DC: Island Press.
Draft - Do Not Cite or Quote                        Ixxv                                  September 2011

-------
1      EXECUTIVE  SUMMARY
    1.1    Introduction
               This Integrated Science Assessment (ISA) is a synthesis and evaluation of the most
               policy-relevant science that forms the scientific foundation for the review of the primary
               (health-based) and secondary (welfare-based) national ambient air quality standard
               (NAAQS) for ozone (O3) and related photochemical oxidants. The current primary O3
               standard includes an 8-hour average standard set in 2008 at 75 parts per billion (ppb). The
               secondary standard for O3 is equal to the primary standard. The current primary NAAQS
               protects against respiratory health effects incurred after short-term exposure to O3, while
               the secondary NAAQS protects against damage to vegetation  and ecosystems.
    1.2    Scope

               EPA has developed an extensive and robust process for evaluating the scientific evidence
               and drawing conclusions regarding air pollution-related health and welfare effects. This
               ISA is focused on health and welfare effects resulting from current ambient
               concentrations of O3. This review builds upon the findings of previous assessments, and
               evaluates the relevant results pertaining to the atmospheric science of O3; short- and
               long-term exposure to ambient O3; health effects due to ambient O3 exposure as
               characterized in epidemiologic, controlled human exposure, and toxicological studies;
               and ecological or welfare effects; as well as O3 exposure-response relationships, mode(s)
               of action (MOA), and populations at increased risk for O3-related health effects. In this
               ISA, the conclusions and key findings  from previous reviews provide the foundation for
               the consideration of evidence from recent studies. Conclusions are drawn based on the
               synthesis of evidence from recent studies and building upon the extensive evidence
               presented in previous reviews.

               EPA has developed a consistent and transparent approach to evaluate the causal nature of
               air pollution-related health and environmental effects for use in developing ISAs; the
               framework for causal determinations is described in the Preamble to this document.
               Causality determinations are based on  the evaluation and synthesis of evidence from
               across scientific disciplines; the type of evidence that is most important for such
               determinations will vary by pollutant or assessment. EPA assesses the entire body of
               relevant literature, building upon evidence available during the previous NAAQS
               reviews, to draw conclusions on the causal relationships between relevant pollutant
               exposures and health or welfare effects. EPA also evaluates the quantitative evidence and

-------
           draws scientific conclusions, to the extent possible, regarding the concentration-response
           relationships and the loads to ecosystems, exposure doses or concentrations, duration and
           pattern of exposures at which effects are observed. This ISA uses a five-level hierarchy
           that classifies the weight of evidence for causation, not just association. This weight of
           evidence evaluation is based on various lines of evidence from across the health and
           environmental effects disciplines. These separate judgments are integrated into a
           qualitative statement about the overall weight of the evidence and causality. The causal
           determinations are:

               •  Causal relationship
               •  Likely to be a causal relationship
               •  Suggestive of a causal relationship
               •  Inadequate to infer a causal relationship
               •  Not likely to be a causal relationship
1.3   Atmospheric Chemistry and Ambient Concentrations

           Ozone is naturally present in the stratosphere, where it serves the beneficial role of
           blocking harmful ultraviolet radiation from the Sun and preventing the majority of this
           radiation from reaching the  surface of the Earth. However, in the troposphere, O3 acts as
           a powerful oxidant and can  harm living organisms and materials. Tropospheric O3 is
           present not only in polluted urban air, but throughout the globe.
           Ozone in the troposphere originates from both anthropogenic (i.e., man-made) and
           natural source categories. Ozone attributed to anthropogenic sources is formed in the
           atmosphere by photochemical reactions involving sunlight and precursor pollutants
           including volatile organic compounds, nitrogen oxides, and carbon monoxide. Ozone
           attributed to natural sources is formed through the same photochemical reactions
           involving natural emissions of precursor pollutants from vegetation, microbes, animals,
           biomass burning, lightning, and geogenic sources. A schematic overview of the major
           photochemical cycles influencing O3 in the troposphere and the stratosphere is shown in
           the figure to the right.

-------
Figure 1-1     Schematic overview of photochemical processes influencing
                stratospheric and tropospheric ozone.
              Ozone in rural areas is produced from emissions of O3 precursors emitted directly within
              the rural areas and from emissions in urban areas that are processed during transport.
              Because O3 is produced downwind of urban source areas and O3 tends to persist longer
              in rural than in urban areas as a result of lower chemical scavenging, the result is
              substantial cumulative exposures for humans and vegetation in rural areas, that are often
              higher than cumulative exposures in urban areas.

              On a smaller scale, O3 can be influenced by local meteorological conditions, circulation
              patterns, emissions, and topographic barriers, resulting in heterogeneous concentrations
              across an individual urban area. On a larger scale, O3 persists in the atmosphere long
              enough that it can be transported  from continent to continent and around the globe. The
              degree of influence from intercontinental transport varies greatly by location and time.

              Background concentrations of O3 have been given various definitions in the literature
              over time. In the context of a review of the NAAQS, it is useful to define background O3
              concentrations in a way that distinguishes between concentrations that result from

-------
           precursor emissions that are relatively less directly controllable from those that are
           relatively more directly controllable through U.S. policies. For this document, we have
           focused on the sum of those background concentrations from natural sources everywhere
           in the world and from anthropogenic sources outside the U.S., Canada and Mexico, i.e.,
           North American background. Since North American background is a construct that
           cannot be measured, the range of North American background O3 concentrations is
           estimated using chemistry transport models. Model-predicted annual average North
           American background estimates are typically less than 50 ppb across the country with
           highest concentrations in the Intermountain West during the spring and the Southwest
           during the summer.
1.4   Human Exposure

           Ozone is ubiquitous throughout the environment, originating from both natural and
           anthropogenic sources, although few indoor sources exist. As such, people are routinely
           exposed to O3 as they participate in normal day-to-day activities. A number of factors
           affect the pattern of personal O3 exposure. These include: the variation in O3
           concentrations at various spatial and temporal scales; individual's activity patterns,
           particularly time spent outdoors, which may involve changes in personal behavior to
           avoid known high exposure to O3; and infiltration of ambient O3 into indoor
           microenvironments, which is driven by air exchange rate.

           Several approaches have been used to measure or quantify exposure to ambient O3,
           giving an indication of the impact of some of the factors that affect the pattern of human
           exposure to O3. These approaches  include characterizing the correlation and ratio
           between personal exposure and ambient O3 concentrations, determining the ratio between
           indoor and outdoor concentrations, and using models to estimate exposure to O3 based on
           ambient concentrations. The factors affecting the pattern of personal  exposure, as well as
           the types of approaches used for quantification of exposure, may have implications for
           epidemiologic studies.
1.5   Dosimetry and Modes of Action

           When O3 is inhaled, the amount of O3 that is absorbed is affected by a number of factors
           including the shape and size of the respiratory tract, route of breathing (nose or mouth),
           as well as how quickly and deeply a person is breathing. Another factor involves the
           reaction of O3 with compounds present in the lung lining fluid to produce secondary

-------
                  oxidation products. On a breath-by-breath basis, humans at rest absorb between 80 and

                  95% of inhaled O3. The  site of the greatest O3 dose to the lung tissue is the junction of

                  the conducting airway and the gas exchange region, in the deeper portion of the

                  respiratory tract. Additionally, the primary site of O3  uptake moves deeper into the

                  respiratory tract during exercise when breathing becomes faster and the breathing route

                  begins to move from the nose only to oronasal breathing (i.e., through the nose and

                  mouth).

                  Once O3 has been inhaled, there are several key events in the toxicity pathway of O3 in

                  the respiratory tract that  lead to  O3-induced health effects. These include formation of

                  secondary oxidation products in the lung, activation of neural reflexes, initiation of

                  inflammation, alterations of epithelial barrier function, sensitization of bronchial smooth

                  muscle, modification  of innate and adaptive immunity, and airway remodeling. Another

                  key event, systemic inflammation and vascular oxidative/nitrosative stress, may be

                  critical to the extrapulmonary effects of O3.
 Table 1-1        Summary of ozone causal determinations by exposure duration
                    and health outcome
  Health Outcome
               Conclusions from Previous Review
  Conclusions from
  2011 2nd Draft ISA
 Short-Term Exposure to O3
 Respiratory effects     The overall evidence supports a causal relationship between acute ambient 03        Causal Relationship
	exposures and increased respiratory morbidity outcomes.	
 Cardiovascular effects   The limited evidence is highly suggestive that 03 directly and/or indirectly contributes to
                    cardiovascular-related morbidity, but much remains to be done to more fully
	substantiate the association.	
 Central nervous system  Toxicological studies report that acute exposures to 03 are associated with alterations
 effects              in neurotransmitters, motor activity, short and long term memory, sleep patterns, and
	histological signs of neurodegeneration.	
 Mortality
The evidence is highly suggestive that 03 directly or indirectly contributes to non-
accidental and cardiopulmonary-related mortality.	
Suggestive of a Causal
Relationship

Suggestive of a Causal
Relationship

Likely to be a Causal
Relationship	
 Long-term Exposure to O3
 Respiratory effects     The current evidence is suggestive but inconclusive for respiratory health effects from
	long-term 03 exposure.	
                                                                    Likely to be a Causal
                                                                    Relationship	
 Cardiovascular Effects  No studies from previous review.
                                                                    Suggestive of a Causal
                                                                    Relationship	
 Reproductive and
 developmental effects
Limited evidence for a relationship between air pollution and birth-related health
outcomes, including mortality, premature births, low birth weights, and birth defects,
with little evidence being found for 03 effects.	
Suggestive of a Causal
Relationship
 Central nervous system  Evidence regarding chronic exposure and neurobehavioral effects was not available.
 effects
                                                                    Suggestive of a Causal
                                                                    Relationship	
 Cancer
Little evidence for a relationship between chronic 03 exposure and increased risk of
lung cancer.	
Inadequate to infer a Causal
Relationship	
 Mortality
There is little evidence to suggest a causal relationship between chronic 03 exposure
and increased risk for mortality in humans.
Suggestive of a Causal
Relationship

-------
1.6   Integration of Ozone Health Effects

           This ISA evaluates and integrates the evidence from short-term (i.e., hours, days, weeks)
           or long-term (i.e., months to years) exposure studies across scientific disciplines (i.e.,
           controlled human exposure studies, toxicology, and epidemiology) in interpreting the
           health effects evidence that spans all lifestages, and varies in severity from minor
           subclinical effects to death. The results from the health studies evaluated in combination
           with the evidence from atmospheric chemistry and exposure assessment studies
           contribute to the causal determinations made for the health outcomes discussed in this
           ISA. The conclusions from the previous NAAQS review and the causality determinations
           from this review are summarized in the table below. Additional details are provided here
           for respiratory health effects and mortality, for which there is the strongest evidence of an
           effect from O3,  and for additional health effects for which there is emerging evidence of
           an association with O3; details for all health effects are provided in the ISA.
   1.6.1   Respiratory Effects

           The clearest evidence for health effects associated with exposure to O3 is provided by
           studies of respiratory effects. Collectively, a very large amount of evidence spanning
           several decades supports the causal association between exposure to O3 and a continuum
           of respiratory effects (See figure below). The majority of this evidence is derived from
           studies investigating short-term exposure (i.e., hours to weeks) to O3, although animal
           toxicological studies and recent epidemiologic evidence demonstrate that long-term
           exposure (i.e., months to years) may also be detrimental to the respiratory system.

           The last review concluded that there was clear, consistent evidence of a causal
           relationship between short-term exposure to O3 and respiratory health effects. This causal
           association was substantiated in this ISA by the coherence of effects observed across
           controlled human exposure, epidemiologic, and toxicological studies indicating
           associations of short-term O3 exposures with a range of respiratory health endpoints from
           respiratory tract inflammation to respiratory emergency department (ED) visits and
           hospital admissions (HA). Across disciplines, short-term O3 exposures induced or were
           associated with statistically significant declines in lung function. An equally strong body
           of evidence from controlled human exposure and toxicological studies demonstrated O3-
           induced inflammatory responses, increased epithelial permeability, and airway
           hyperresponsiveness. Toxicological studies provided additional evidence for O3-induced
           impairment of host defenses.  Combined, these findings from experimental studies
           provided support for epidemiologic evidence, in which short-term O3 exposure was

-------
              consistently associated with increases in respiratory symptoms and asthma medication
              use in asthmatic children, respiratory-related hospital admissions, and asthma-related ED
              visits. Although O3 was consistently associated with nonaccidental and cardiopulmonary
              mortality, the contribution of respiratory causes to these findings was uncertain. The
              combined evidence across disciplines supports a causal relationship between short-
              term O3 exposure and respiratory effects.
                                                                                A
                                 Emergency Department Visits
                                   and Hospital Admissions

                                 Doctor visits, school absences
                            Respiratory symptoms, medication use,
                                       asthma attacks
                      Lung function decrements, inflammation and increased
                             permeability, susceptibility to infection
                              Proportion of Population Affected
Figure 1-2     The continuum of respiratory effects, noting increases in severity
                but decreases in the proportion of the population affected moving
                up the pyramid.
              Recent evidence for a relationship between long-term O3 exposure and respiratory
              morbidity comes from a single cohort demonstrating associations between long-term
              measures of O3 exposure and new-onset asthma in children and increased respiratory

-------
        symptom effects in asthmatics. While the evidence may be limited, this multi-community
        cohort demonstrates that asthma risk is affected by interactions between genetic
        variability, environmental O3 exposure, and behavior. Other recent studies provide
        coherent evidence for long-term O3 exposure and respiratory morbidity effects such as
        first asthma hospitalization and respiratory symptoms in asthmatics. Generally, the
        epidemiologic and toxicological evidence provides a compelling case for a relationship
        between long-term exposure to ambient O3  and respiratory morbidity. The evidence for
        effects of short-term exposure to O3 on respiratory endpoints provides coherence and
        biological plausibility for the effects of long-term exposure to O3. Building upon
        evidence from studies of short-term exposure, the more recent epidemiologic evidence,
        combined with toxicological studies in rodents and non-human primates, provides
        biologically plausible evidence that there is likely to be a causal relationship between
        long-term exposure to O3  and respiratory health effects.
1.6.2   Mortality Effects

        The last review concluded that the overall body of evidence was highly suggestive that
        short-term exposure to O3 directly or indirectly contributes to non-accidental and
        cardiopulmonary-related mortality; but that additional research was needed to more fully
        establish underlying mechanisms by which such effects occur. The evaluation of new
        multicity studies that have examined the association between short-term O3 exposure and
        mortality found evidence which supports the conclusions of the last review. These recent
        studies reported consistent positive associations between short-term O3 exposure and
        total (nonaccidental) mortality, with associations being stronger during the warm season.
        They also added support for associations between O3 exposure and cardiovascular
        mortality being similar to or stronger than those between O3 exposure and respiratory
        mortality. Additionally, these new studies examined previously identified areas of
        uncertainty in the O3-mortality relationship, and provide evidence that continues to
        support an association between short-term O3 exposure and mortality. The body of
        evidence indicates that there is likely to  be a causal relationship between  short-term
        exposures to O3 and mortality.
1.6.3   Emerging Evidence

        In the last review, completed in 2006, there were a number of health effects for which an
        insufficient amount of evidence existed to adequately characterize the relationships with
        exposure to O3. However, recent evidence indicates that O3 may impart health effects

-------
        through exposure durations and biological mechanisms not previously considered. This
        includes:

            •  Toxicological studies provide evidence for cardiovascular morbidity, while
              epidemiologic studies provide evidence for cardiovascular mortality, and
              together, this evidence is suggestive of a causal relationship for both
              relevant short- and long-term exposures to O3 and cardiovascular effects.
            •  Recent toxicological studies add to earlier evidence that short- and long-term
              exposures to O3 can produce a range of effects on the central nervous system
              and behavior. The single epidemiologic study conducted showed that long-
              term exposure to O3 affects memory in humans as well. Together the evidence
              from studies of short- and long-term exposure to O3 is suggestive of a causal
              relationship between O3 exposure and adverse central nervous system
              effects.
            •  There is limited though positive toxicological evidence for O3-induced
              developmental effects. Limited epidemiologic evidence exists for an
              association with O3 concentration and decreased sperm concentration and
              associations with reduced birth weight and restricted fetal growth. Overall, the
              evidence is suggestive of a causal relationship between long-term
              exposures to O3 and reproductive and developmental effects.
            •  Several recent studies provide evidence of an association between long-term
              exposure to  O3 and mortality, especially respiratory mortality. Collectively,
              the evidence is suggestive of a causal relationship between long-term O3
              exposures and mortality.
1.6.4   Populations at Increased Risk

        The examination of populations potentially at increased risk for O3 exposure allows for
        the NAAQS to provide an adequate margin of safety for both the general population and
        for sensitive populations. Some studies attempt to identify populations that are at
        increased risk for O3-related health effects; these studies do so by examining groups
        within the study population, such as those with an underlying health condition or genetic
        polymorphism; categories of age, race, or sex; or by developing animal models that
        mimic the conditions associated with an adverse health effect. Such studies have
        identified a multitude of factors that could potentially contribute to whether an individual
        is at increased risk for O3-related health effects. The populations  identified that are most
        at risk for O3-related health effects are individuals with influenza/infection, individuals
        with asthma, and older age groups. Other potential factors, including preexisting

-------
           conditions such as chronic obstructive pulmonary disease and cardiovascular disease,
           young age, sex, and variations in multiple genes (such as GSTM1, GSTP1, HMOX-1,
           NQO1, and TNF-a), appear related to susceptibility, but further evidence is needed.
   1.6.5   Ozone Concentration-Response Relationship

           An important consideration in characterizing the association of O3 with morbidity and
           mortality is the shape of the concentration-response relationship across the O3
           concentration range. In this ISA, studies have been identified that attempt to characterize
           the shape of the O3 concentration-response curve along with possible O3 "thresholds"
           (i.e., O3 levels which must be exceeded in order to elicit a physiological response). These
           studies have indicated a generally linear concentration-response function with no
           indication of a threshold for O3 concentrations greater than 30 or 40 ppb, thus if a
           threshold exists, it is likely at the lower end of the range of ambient O3 concentrations.
1.7   Integration of Effects on Vegetation and Ecosystems

           The ISA presents the most policy-relevant information pertaining to the review of the
           NAAQS for the effects of O3 on vegetation and ecosystems. It integrates key findings
           about plant physiology, biochemistry, whole plant biology, ecosystems and exposure-
           response relationships. The welfare effects of O3 can be observed across spatial scales,
           starting at the cellular and subcellular level, then the whole plant and finally, ecosystem-
           level processes. Ozone effects at small spatial scales, such as the leaf of an individual
           plant, can result in effects at a continuum of larger spatial scales. These effects include
           altered rates of leaf gas exchange, growth and reproduction at the individual plant level
           and can result in changes in ecosystems, such as productivity, C storage, water cycling,
           nutrient cycling, and community composition. The conclusions from the previous
           NAAQS review and the causality determinations from this review are summarized in the
           table below. Further discussion of these conclusions is provided below for visible foliar
           injury, growth, productivity, and carbon storage, reduced yield and quality of agricultural
           crops, water cycling, below-ground processing, community composition, and O3
           exposure-response relationships; discussion for all relevant welfare effects is provided in
           the ISA.

-------
Table 1-2      Summary of ozone causal determination for welfare effects
Vegetation and
Ecosystem Effects
Visible Foliar Injury
Effects on Vegetation
Reduced Vegetation
Growth
Reduced Productivity in
Terrestrial Ecosystems
Reduced Carbon (C)
Sequestration in
Terrestrial Ecosystems
Reduced Yield and
Quality of Agricultural
Crops
Alteration of Terrestrial
Ecosystem Water Cycling
Alteration of Below-
ground Biogeochemical
Cycles
Alteration of Terrestrial
Community Composition
Conclusions from Previous Review
Data published since the 1996 03 AQCD strengthen previous conclusions that there is
strong evidence that current ambient 03 concentrations cause impaired aesthetic quality of
many native plants and trees by increasing foliar injury.
Data published since the 1996 03 AQCD strengthen previous conclusions that there is
strong evidence that current ambient 03 concentrations cause decreased growth and
biomass accumulation in annual, perennial and woody plants, including agronomic crops,
annuals, shrubs, grasses, and trees.
There is evidence that 03 is an important stressor of ecosystems and that the effects of
03 on individual plants and processes are scaled up through the ecosystem, affecting net
primary productivity.
Limited studies from previous review
Data published since the 1996 03 AQCD strengthen previous conclusions that there is
strong evidence that current ambient 03 concentrations cause decreased yield and/or
nutritive quality in a large number of agronomic and forage crops.
Ecosystem water quantity may be affected by 03 exposure at the landscape level.
Ozone-sensitive species have well known responses to 03 exposure, including altered C
allocation to below-ground tissues, and altered rates of leaf and root production, turnover,
and decomposition. These shifts can affect overall C and N loss from the ecosystem in terms
of respired C, and leached aqueous dissolved organic and inorganic C and N.
Ozone may be affecting above- and below -ground community composition through impacts
on both growth and reproduction. Significant changes in plant community composition
resulting directly from 03 exposure have been demonstrated.
Conclusions from
2011 2nd Draft
ISA
Causal Relationship
Causal Relationship
Causal Relationship
Likely to be a Causal
Relationship
Causal Relationship
Likely to be a Causal
Relationship
Causal Relationship
Likely to be a Causal
Relationship
      1.7.1   Visible Foliar Injury

              Visible foliar injury resulting from exposure to O3 has been well characterized and
              documented over several decades on many tree, shrub, herbaceous and crop species.
              Ozone-induced visible foliar injury symptoms on certain plant species are considered
              diagnostic of exposure to O3, as experimental evidence has clearly established a
              consistent association, with greater exposure often resulting in greater and more
              prevalent injury. Additional sensiptive species showing visible foliar injury continue to
              be identified from field surveys and verified in controlled exposure studies. Overall,
              evidence is sufficient to conclude that there is a causal relationship between
              ambient O3 exposure and the occurrence of O3-induced visible foliar injury on
              sensitive vegetation across the U.S.

-------
1.7.2   Growth, Productivity, Carbon Storage and Agriculture

        Ambient O3 concentrations have long been known to cause decreases in photosynthetic
        rates and plant growth. The O3-induced effects at the plant scale may translate to the
        ecosystem scale, and cause changes in productivity and C storage. The effects of O3
        exposure on photosynthesis, growth, biomass allocation, ecosystem production and
        ecosystem C sequestration were reviewed for natural ecosystems, and crop productivity
        and crop quality were reviewed for agricultural ecosystems. There is strong and
        consistent evidence that ambient concentrations of O3 decrease plant photosynthesis and
        growth in numerous plant species across the U.S. Studies conducted during the past four
        decades have also demonstrated unequivocally that O3 alters biomass allocation and plant
        reproduction. Studies at the leaf and plant scales showed that O3 reduced photosynthesis
        and plant growth, providing coherence and biological plausibility for the reported
        decreases in ecosystem productivity. In addition to primary productivity, other indicators
        such as net ecosystem CO2 exchange and C sequestration were often assessed by
        modeling studies. Model simulations consistently found that O3 exposure caused
        negative impacts on those indicators, but the severity of these impacts was influenced by
        multiple interactions of biological and environmental factors. Although O3 generally
        causes negative effects on ecosystem productivity, the magnitude of the response varies
        among plant communities. Overall, evidence is sufficient to conclude that there is a
        causal relationship between O3 exposure and reduced plant growth and
        productivity, and a likely causal relationship between  O3 exposure and reduced
        carbon  sequestration in terrestrial ecosystems.
        The detrimental effect of O3 on crop production has been recognized since the 1960's,
        and current O3 concentrations across the U.S. are high enough to cause yield loss for a
        variety of agricultural crops including, but not limited to, soybean, wheat, potato,
        watermelon, beans, turnip, onion, lettuce, and tomato. Continued increases in O3
        concentration may  further decrease yield in these sensitive crops while also initiating
        yield losses in less sensitive crops. Research has linked increasing O3 concentration to
        decreased photosynthetic rates and  accelerated senescence, which are related to yield.
        Evidence is sufficient to conclude that there  is a causal relationship between O3
        exposure and reduced yield and quality of agricultural crops.
1.7.3   Water Cycling

        Ozone can affect water use in plants and ecosystems through several mechanisms
        including damage to stomatal functioning and loss of leaf area. Possible mechanisms for

-------
        O3 exposure effects on stomatal functioning include the build-up of CO2 in the
        substomatal cavity, impacts on signal transduction pathways and direct O3 impact on
        guard cells. Regardless of the mechanism, O3 exposure has been shown to alter stomatal
        performance, which may affect plant and stand transpiration and therefore may affect
        hydrological cycling. Although the direction of the response differed among studies, the
        evidence is sufficient to conclude that there is likely to be a causal relationship
        between O3 exposure and the alteration of ecosystem water cycling.
1.7.4   Below Ground Processes

        Below-ground processes are tightly linked with above-ground processes. The responses
        of above-ground process to O3 exposure, such as reduced photosynthetic rates, increased
        metabolic cost, and reduced root C allocation, have provided biologically plausible
        mechanisms for the alteration of below-ground processes. These include altered quality
        and quantity of C input to soil, microbial community composition, and C and nutrient
        cycling. The evidence is sufficient to conclude that there is a causal relationship
        between O3 exposure and the alteration of below-ground biogeochemical cycles.
1.7.5   Community Composition

        Ozone exposure changes competitive interactions and leads to loss of O3-sensitive
        species or genotypes. Studies at the plant level found that the severity of O3 damage to
        growth, reproduction and foliar injury varied among species, which provided the
        biological plausibility for the alteration of community composition. For example, there is
        a tendency for O3 exposure to shift the biomass of grass-legume mixtures in favor of
        grass species. Ozone exposure not only altered community composition of plant species,
        but also microorganisms: research since the last review has shown that O3 can also alter
        community composition and diversity of soil microbial communities. Shifts in
        community composition of bacteria and fungi have been observed in both natural and
        agricultural ecosystems, although no general patterns could be identified. The evidence
        is sufficient to conclude that there is likely a causal relationship between O3
        exposure and the alteration of community composition.
1.7.6   Ozone Exposure-Response Relationships

        Previous reviews of the NAAQs have included exposure-response functions for the yield
        of many crop species, and for the biomass accumulation of tree species. They were based

-------
           on large-scale experiments designed to obtain clear exposure-response data, and are
           updated in this ISA by using the W126 metric to quantify exposure. In recent years,
           extensive exposure-response data obtained in more naturalistic settings have become
           available for yield of soybean and growth of aspen. This ISA validates the exposure-
           response median functions based on previous data by comparing their predictions with
           the newer observations. The functions supply very accurate predictions of effects in
           naturalistic settings. Recent meta-analyses of large sets of crop and tree studies do not
           give rise to exposure-response functions, but their results are consistent with the
           functions presented in the ISA. It is important to note that although these median
           functions provide reliable models for groups of species or group of genotypes within a
           species, the original data and recent results consistently show that some species, and
           within species and some genotypes within species are much more severely affected by
           exposure to O3.
1.8   The Role of Tropospheric Ozone  in Climate Change and UV-B
       Effects

           Atmospheric O3 plays an important role in the Earth's energy budget by interacting with
           incoming solar radiation and outgoing infrared radiation. Tropospheric O3 makes up only
           a small portion of the total column of O3, but  it has important incremental effects on the
           overall radiation budget. Therefore, perturbations in tropospheric O3 concentrations can
           have direct effects on climate and indirect effects on health, ecology and welfare by
           shielding the earth's surface from solar ultraviolet (UV) radiation.
           Ozone is an important greenhouse gas, and increases in its abundance in the troposphere
           may contribute to climate change. Models calculate that the global burden of tropospheric
           O3 has doubled since the preindustrial era, while observations indicate that in some
           regions O3 may have increased by factors as great as 4 or 5. These increases are tied to
           the rise in emissions of O3 precursors from human activity, mainly fossil fuel
           consumption and agricultural processes.

-------
                                 Precursor Emissions of
                                 CO, VOCs,CH,,NOx
                                         (Tg/y)
                                     Froposphcric O,
                                      Abundance
                                          (To)
                                          ^  Of
                                    Radiative Forcing
                                   Due to O, Change
                                        (W/rrr)
                                   Climate Response
                                                           G
Figure 1-3     Schematic illustrating the effects of tropospheric O3 on climate.
              Figure 1-3 shows the main steps involved in the influence of tropospheric O3 on climate.
              Emissions of O3 precursors lead to production of tropospheric O3. A change in the
              abundance of tropospheric O3 perturbs the radiative balance of the atmosphere, an effect
              quantified by the radiative forcing (RF) metric. The earth-atmosphere-ocean system
              responds to the radiative forcing with a climate response, typically expressed as a change
              in surface temperature. Finally, the climate response causes downstream climate-related
              health and ecosystem impacts. Feedbacks from both the climate response and
              downstream impacts can, in turn, affect the abundance of tropospheric O3 and O3
              precursors through multiple feedback mechanisms as indicated in the figure.

              UV radiation emitted from the Sun contains sufficient energy when it reaches the Earth to
              have damaging effects on living organisms and materials. Atmospheric O3 plays a crucial
              role in reducing exposure to UV radiation at the Earth's surface. Ozone in the
              stratosphere is responsible for the majority of this shielding, but O3 in the troposphere
              provides supplemental shielding of UV radiation in the mid-wavelength range (UV-B),
              thereby influencing human and ecosystem health.

-------
                The conclusions from the previous NAAQS review and the causality determinations from

                this review relating climate change and UV-B effects are summarized in the table below,

                with details provided in the ISA.
 Table 1-3      Summary of ozone causal determination for climate change and
                  UV-B effects
       Effects
               Conclusions from Previous Review
 Conclusions from
 2011 2nd Draft ISA
 Radiative Forcing
 Climate Change
Climate forcing by 03 at the regional scale may be its most important impact on climate.  Causal Relationship
While more certain estimates of the overall importance of global-scale forcing due to
tropospheric 03 await further advances in monitoring and chemical transport modeling,
the overall body of scientific evidence suggests that high concentrations of 03 on the
regional scale could have a discernable influence on climate, leading to surface
temperature and hydrological cycle changes.	
Likely to be a Causal
Relationship
 UV-B Related Health and   UV-B has not been studied in sufficient detail to allow for a credible health benefits
 Welfare Effects          assessment. In conclusion, the effect of changes in surface-level 03 concentrations on
                     UV-induced health outcomes cannot yet be critically assessed within reasonable
	uncertainty.	
                                                               Inadequate to Determine if
                                                               a Causal Relationship
                                                               Exists
     1.9    Conclusion
                The clearest evidence for human health effects associated with exposure to O3 is

                provided by studies of respiratory effects. Collectively, there is a very large amount of

                evidence spanning several decades in support of a causal association between exposure to

                O3 and a continuum of respiratory effects. The majority of this evidence is derived from

                studies investigating short-term O3 exposure (i.e., hours to weeks), although animal

                toxicological studies and recent epidemiologic evidence demonstrate that long-term

                exposure (i.e., months to years) may also be detrimental to the respiratory system.

                Additionally, consistent positive associations between short-term O3 exposure and total

                (nonaccidental) mortality have helped to resolve previously identified areas of

                uncertainty in the O3-mortality relationship, indicating that there is likely to be a causal

                relationship between short-term exposures to O3 and all-cause mortality. Recent

                evidence is suggestive of a causal relationship between long-term O3 exposures and

                mortality. The evidence for these health effects indicates that the relationship between

                concentration and response is linear within concentrations present in the U.S., with no

                indication of a threshold of O3 concentrations under which no effect would be observed.

                The populations identified as being most at risk for O3-related health effects are

                individuals with influenza/infection, individuals with asthma, and older age groups.

-------
There has been over 40 years of research on the effects of O3 exposure on vegetation and
ecosystems. The best evidence for effects is from controlled exposure studies. These
studies have clearly shown that exposure to O3 is causally linked to visible foliar injury,
decreased photosynthesis, changes in reproduction, and decreased growth. Recently,
studies at larger spatial scales support the results from controlled studies and indicate that
ambient O3 exposures can affect ecosystem productivity, crop yield, water cycling, and
ecosystem community composition. And on a global scale, tropospheric O3  is the third
most important greenhouse gas, playing an important role in climate change.

-------
      2   INTEGRATIVE  SUMMARY

 1                   This Integrated Science Assessment (ISA) forms the scientific foundation for the review
 2                   of the national ambient air quality standards (NAAQS) for ozone (O3). The ISA is a
 3                   concise evaluation and synthesis of the most policy-relevant science, and it
 4                   communicates critical science judgments relevant to the review of the NAAQS for O3.
 5                   The ISA accurately reflects "the latest scientific knowledge useful in indicating the kind
 6                   and extent of identifiable effects on public health which may be expected from the
 7                   presence of [a] pollutant in ambient air" (CAA. 1990a). Key information and judgments
 8                   contained in prior Air Quality Criteria Documents (AQCD) for O3 are incorporated into
 9                   this assessment. Additional details of the pertinent scientific literature published since the
10                   last review, as well as selected older studies of particular interest, are included. This ISA
11                   thus serves to update and revise the evaluation of the scientific evidence available at the
12                   time of the completion of the 2006 O3 AQCD. The current primary O3 standard includes
13                   an 8-hour (h) average (avg) standard  set at 75 parts per billion (ppb). The secondary
14                   standard for O3 is set equal to the primary standard. Further information on the legislative
15                   and historical background for the O3 NAAQS is contained in the Preface to this ISA.

16                   This chapter summarizes and synthesizes the newly available  scientific evidence and is
17                   intended to provide a concise synopsis of the ISA conclusions and findings that best
18                   inform consideration of the policy-relevant questions that frame this assessment
19                   (presented in Section 2.1). It includes:

20                      •  An integration of the evidence on the health effects associated with short- and
21                         long-term exposure to O3, discussion of important uncertainties identified in
22                         the interpretation of the scientific evidence, and an integration of health
23                         evidence from the different scientific disciplines and exposure durations.
24                      "An integration of the evidence on the ecological and welfare effects associated
25                         with exposure to O3,  and discussion of important uncertainties identified in the
26                         interpretation of the scientific evidence.
27                      •  Discussion of policy-relevant considerations,  such as potentially  at-risk
28                         populations and concentration-response relationships.
          2.1    Policy-Relevant Questions for O3 NAAQS Review

29                   The draft Integrated Review Plan for the Ozone National Ambient Air Quality Standards
30                   (IRP) (U.S. EPA. 2009c) identified key policy-relevant questions that provide a
31                   framework for this assessment of the scientific evidence. These questions frame the entire
      Draft - Do Not Cite or Quote                       2-1                                 September 2011

-------
 1                   review of the NAAQS for O3 and thus are informed by both science and policy
 2                   considerations. The ISA organizes, presents, and integrates the scientific evidence which
 3                   is considered along with findings from risk analyses and policy considerations to help the
 4                   U.S. Environmental Protection Agency (EPA) address these questions during the
 5                   NAAQS review. In evaluating the health evidence, the focus of this assessment is on
 6                   scientific evidence that is most relevant to the following questions taken directly from the
 7                   Integrated Review Plan:

 8                      "To what extent has new scientific information become available that alters or
 9                         substantiates our understanding of the health effects associated with various
10                         time periods of exposure to ambient O3, including short-term (1-3 hours),
11                         prolonged (6-8 hours), and chronic (months to years) exposures?
12                      "To what extent has new scientific information become available that alters or
13                         substantiates our understanding of the health effects of O3 on at-risk
14                         populations, including those with potentially increased susceptibility such as
15                         children and disadvantaged populations?
16                      "To what extent has new scientific information become available that alters or
17                         substantiates conclusions from previous reviews regarding the plausibility of
18                         adverse health effects caused by O3 exposure?
19                      "At what levels of O3 exposure are health effects observed? Is there evidence of
20                         effects at exposure levels lower than those previously observed, and what are
21                         the important uncertainties associated with that evidence? What is the nature
22                         of the exposure-response relationships of O3 for the various health effects
23                         evaluated?
24                      •  To what extent has new scientific information become available that alters or
25                         substantiates our understanding of non-O3-exposure factors that might
26                         influence the associations between O3 levels and health effects being
27                         considered (e.g., weather-related factors; behavioral factors such as heating/air
28                         conditioning use; driving patterns; and time-activity patterns)?
29                      •  To what extent do risk and/or exposure analyses suggest that exposures of
30                         concern for O3-related health effects are likely to occur with current ambient
31                         levels of O3 or with levels that just meet the O3 standard? Are these
32                         risks/exposures of sufficient magnitude such that the health effects might
33                         reasonably be judged to be important from a public health perspective? What
34                         are the important uncertainties associated with these risk/exposure estimates?
35                      "To what extent have important uncertainties identified in the last rulemaking
36                         been addressed and/or have new uncertainties emerged?
      Draft - Do Not Cite or Quote                       2-2                                 September 2011

-------
 1                  In evaluating the welfare evidence, the available scientific evidence will focus on key
 2                  policy-relevant issues by addressing a series of questions including the following:

 3                      "To what extent has new scientific information become available that alters or
 4                         substantiates our understanding of the effects on vegetation and other welfare
 5                         effects  following exposures to levels of O3 found in the ambient air?
 6                      "To what extent has new scientific information become available to inform our
 7                         understanding of the nature of the exposures that are associated with such
 8                         effects  in terms of biologically relevant cumulative, seasonal exposure
 9                         indices?
10                      "To what extent has new scientific information become available that alters or
11                         substantiates our understanding of the effects of O3 on sensitive plant species,
12                         ecological receptors, or ecosystem processes?
13                      "To what extent has new scientific information become available that alters or
14                         substantiates our understanding of exposure factors other than O3 that might
15                         influence the associations between O3 levels and welfare effects being
16                         considered (e.g., site specific features such as  elevation, soil moisture level,
17                         presence of co-occurring competitors, pests, pathogens, other pollutant
18                         stressors, weather-related factors)?
19                      "To what extent has new scientific information become available that alters or
20                         substantiates conclusions regarding the occurrence of adverse  welfare effects
21                         at levels of O3 as low as or lower than those observed previously? What is the
22                         nature of the exposure-response relationships of O3 for the various welfare
23                         effects  evaluated?
24                      •  Given recognition in the last review that the significance of O3-induced effects
25                         to the public welfare depends in part on the intended use of the plants or
26                         ecosystems on which those effects occurred, to what extent has new scientific
27                         evidence become available to suggest additional locations where the
28                         vulnerability of sensitive species or ecosystems would have special
29                         significance to the public welfare and should be given increased focus in this
30                         review?
31                      "To what extent do risk and/or exposure analyses suggest that exposures of
32                         concern for O3-related welfare  effects are likely to occur with  current ambient
33                         levels of O3 or with levels that just meet the O3 standard? Are these
34                         risks/exposures of sufficient magnitude such that the welfare effects might
35                         reasonably be judged to be important from a public welfare perspective?  What
36                         are the  important uncertainties  associated with these risk/exposure estimates?
      Draft - Do Not Cite or Quote                       2-3                                 September 2011

-------
 1                      •  To what extent have important uncertainties identified in the last review been
 2                         addressed and/or have new uncertainties emerged?
 3                      "To what extent does newly available information reinforce or call into
 4                         question any of the basic elements of the current O3 standard?
          2.2    ISA Development and Scope

 5                   EPA has a developed a robust, consistent, and transparent process for evaluating the
 6                   scientific evidence and drawing conclusions and causal judgments regarding air
 7                   pollution-related health and environmental effects. The ISA development process
 8                   includes literature search strategies, criteria for selecting and evaluating studies,
 9                   approaches for evaluating weight of the evidence, and a framework for making causality
10                   determinations. The process and causality framework are described in more detail in the
11                   Preamble to the ISA [website]. This section provides a brief overview of the process for
12                   development of this ISA.

13                   EPA initiated the current review of the NAAQS for O3 on September 29, 2008, with a
14                   call for information from the public (U.S. EPA. 2008f). Literature searches were
15                   conducted routinely to identify studies published since the last review, focusing on
16                   studies published from 2005 (close of previous scientific assessment) through July 2011.
17                   References that were considered for inclusion in this ISA can be found using the HERO
18                   website (http://hero.epa.gov/ozone). This site contains HERO links to lists of references
19                   that are cited in the ISA, as well as those that were considered for inclusion, but not cited
20                   in the ISA, with bibliographic information and abstracts.

21                   This review has endeavored to evaluate all relevant data published since the last review
22                   pertaining to the atmospheric science of O3, human exposure to ambient O3,
23                   epidemiologic, controlled human exposure, toxicological, and ecological or welfare
24                   effects studies, including studies related to exposure-response relationships, mode(s) of
25                   action (MOA), and understanding of at-risk or susceptible populations for effects of O3
26                   exposure. Added to the body of research were EPA's analyses of air quality and
27                   emissions data, studies on atmospheric chemistry, transport, and fate of these emissions,
28                   as well as issues related to exposure to O3.
29                   Previous AQCDs (U.S. EPA. 2006b. 1996a. b, 1984. 1978a) have included an extensive
30                   body of evidence on both health and ecological effects of O3 exposure, as well as an
31                   understanding of the atmospheric chemistry of O3 (U.S. EPA. 2006b). In this ISA, the
32                   conclusions and key findings from previous reviews are summarized at the beginning of
33                   each section, to provide the foundation for consideration of evidence from recent studies.
      Draft - Do Not Cite or Quote                       2-4                                 September 2011

-------
 1                   Results of key studies from previous reviews are included in discussions or tables and
 2                   figures, as appropriate, and conclusions are drawn based on the synthesis of evidence
 3                   from recent studies with the extensive literature summarized in previous reviews.

 4                   The Preamble discusses the general framework for conducting the science assessment
 5                   and developing an ISA, including criteria for evaluating studies and developing scientific
 6                   conclusions. For selection of epidemiologic studies in the O3 ISA, particular emphasis is
 7                   placed on those studies most relevant to the review of the NAAQS. Studies conducted in
 8                   the United States (U.S.) or Canada are discussed in more detail than those from other
 9                   geographical regions, and particular emphasis is placed on: (1) recent multicity studies
10                   that employ standardized analysis methods for evaluating effects of O3 and that provide
11                   overall estimates for effects, based on combined analyses of information pooled across
12                   multiple cities; (2) studies that help understand quantitative relationships between
13                   exposure concentrations and effects; (3) new studies that provide  evidence on effects in
14                   susceptible populations; and (4) studies that consider and report O3 as a component of a
15                   complex mixture of air pollutants. In evaluating toxicological and controlled human
16                   exposure studies, emphasis is placed on studies using concentrations or doses  that are
17                   within about an order of magnitude of ambient O3 concentrations. Consideration of issues
18                   important for evaluation of human exposure to ambient O3 include the relationship
19                   between O3 measured at central site monitors and personal exposure to ambient O3
20                   environments, since penetration of O3 into indoor environments may be limited.

21                   This ISA uses a five-level hierarchy that classifies the weight of evidence for causation:

22                       •  Causal relationship
23                       •  Likely to be a causal relationship
24                       •  Suggestive of a causal relationship
25                       •  Inadequate to infer a causal relationship
26                       •  Not likely to be a causal relationship

27                   Beyond judgments regarding causality are questions relevant to quantifying health or
28                   environmental risks based on our understanding of the quantitative relationships between
29                   pollutant exposures and health or welfare effects. Once a determination is made regarding
30                   the causal relationship between the pollutant and outcome category, important questions
31                   regarding quantitative relationships include:

32                       •  What is the concentration-response or dose-response relationship?
33                       •  Under what exposure conditions (dose or concentration, duration and pattern)
34                         are effects observed?
      Draft - Do Not Cite or Quote                        2-5                                  September 2011

-------
 1                      •  What populations appear to be differentially affected i.e., more susceptible to
 2                         effects?
 3                      •  What elements of the ecosystem (e.g., types, regions, taxonomic groups,
 4                         populations, functions, etc.) appear to be affected or are more sensitive to
 5                         effects?

 6                  This chapter summarizes and integrates the newly available scientific evidence that best
 7                  informs consideration of the policy-relevant questions that frame this assessment.
 8                  Section 2.3 discusses the trends in ambient concentrations and sources of O3 and provides
 9                  a brief summary of ambient air quality for short- and long-term exposure  durations.
10                  Section 2.4 presents the evidence regarding personal exposure to ambient O3 in outdoor
11                  and indoor microenvironments, and it discusses the relationship between ambient O3
12                  concentrations and personal exposure to O3 from ambient sources. Section 2.5 provides a
13                  discussion of the dosimetry and mode of action evidence for O3 exposure. Section 2.6
14                  integrates the evidence for studies that examine the health effects associated with short-
15                  and long-term exposure to O3 and discusses important uncertainties identified in the
16                  interpretation of the scientific evidence. Section 2.7 provides a discussion of policy-
17                  relevant considerations, such as potentially at-risk populations, lag structure, and the O3
18                  concentration-response relationship. Section 2.8 integrates the health evidence from the
19                  different scientific disciplines and exposure durations. Finally, Section 2.9 summarizes
20                  the evidence for welfare effects related to O3 exposure, and Section 2.10 reviews the
21                  literature on the influence of tropospheric O3 on climate and exposure to solar ultraviolet
22                  radiation.
          2.3    Atmospheric Chemistry and Ambient Concentrations


            2.3.1    Physical and Chemical Processes

23                   Ozone in the troposphere originates from both anthropogenic (i.e., man-made) and
24                   natural source categories. Ozone attributed to anthropogenic sources is formed in the
25                   atmosphere by photochemical reactions involving sunlight and precursor pollutants
26                   including volatile organic compounds (VOCs), nitrogen oxides (NOX), and carbon
27                   monoxide (CO). Ozone attributed to natural sources is formed through the same
28                   photochemical reactions involving natural emissions of precursor pollutants from
29                   vegetation, microbes, animals, biomass burning, lightning, and geogenic sources. A
30                   schematic overview of the major photochemical cycles influencing O3 in the troposphere
31                   and the stratosphere is shown in Figure 2-1. The processes depicted in this figure are
32                   fairly well understood, and were covered in detail in the previous O3 AQCD. The

      Draft - Do Not Cite or Quote                       2-6                                 September 2011

-------
1
2
3
4
5
formation of O3, other oxidants, and oxidation products from these precursors is a
complex, nonlinear function of many factors including: (1) the intensity and spectral
distribution of sunlight; (2) atmospheric mixing; (3) concentrations of precursors in the
ambient air and the rates of chemical reactions of these precursors; and (4) processing on
cloud and aerosol particles.
     Figure 2-1     Schematic overview of photochemical processes influencing
                    stratospheric and tropospheric ozone.
7
8
Ozone is present not only in polluted urban atmospheres but throughout the troposphere,
even in remote areas of the globe. The same basic processes involving sunlight-driven
reactions of NOX, VOCs and CO contribute to O3 formation throughout the troposphere.
     Draft - Do Not Cite or Quote
                              2-7
September 2011

-------
 1                   These processes also lead to the formation of other photochemical products, such as
 2                   peroxyacetyl nitrate, nitric acid, and sulfuric acid, and to other compounds, such as
 3                   formaldehyde and other carbonyl compounds. In urban areas, NOX, VOCs and CO are all
 4                   important for O3 formation. In nonurban vegetated areas, biogenic VOCs emitted from
 5                   vegetation tend to be the most important precursor to O3 formation. In the remote
 6                   troposphere, methane - structurally the simplest VOC - and CO are the main carbon-
 7                   containing precursors to O3 formation. Throughout the troposphere, O3 is subsequently
 8                   lost through a number of gas phase reactions and deposition to surfaces as  shown in
 9                   Figure 2-1.

10                   Convective processes and turbulence transport O3 and other pollutants both upward and
11                   downward throughout the planetary boundary layer and the free troposphere. In many
12                   areas of the  U.S., O3 and its precursors can be transported over long distances, aided by
13                   vertical mixing. The transport of pollutants downwind of major urban centers is
14                   characterized by the development of urban plumes. Meteorological conditions, small-
15                   scale circulation patterns, localized chemistry, and mountain barriers can influence
16                   mixing on a smaller scale, resulting in frequent heterogeneous O3 concentrations across
17                   an individual urban area.

18                   Furthermore, the mean tropospheric lifetime of O3 is long enough that it can be
19                   transported from continent to continent and latitudinally around the globe.  The degree of
20                   influence from intercontinental transport varies greatly by location and time. For
21                   instance, high elevation  sites are most susceptible to the intercontinental transport of
22                   pollution, particularly during spring. Given the nonlinear chemistry involving O3
23                   formation, the task of isolating the influence of intercontinental transport of O3 and O3
24                   precursors on regional air quality is quite complex and the topic of the next section.
             2.3.2   Atmospheric Modeling of Background Ozone Concentrations

25                   Background concentrations of O3 have been given various definitions in the literature
26                   over time. In the context of a review of the NAAQS, it is useful to define background O3
27                   concentrations in a way that distinguishes between concentrations that result from
28                   precursor emissions that are relatively less directly controllable from those that are
29                   relatively more directly controllable through U.S. policies. North American (NA)
30                   background O3 can include contributions that result from emissions from natural sources
31                   (e-g-, stratospheric intrusion, biogenic methane and more short-lived VOC emissions),
32                   emissions of pollutants that contribute to global concentrations of O3 (e.g., anthropogenic
33                   methane) from countries outside North America. In previous NAAQS reviews, a specific
34                   definition of background concentrations was used and referred to as policy relevant
      Draft - Do Not Cite or Quote                       2-8                                September 2011

-------
 1                  background (PRB). In those previous reviews, PRB concentrations were defined by EPA
 2                  as those concentrations that would occur in the U.S. in the absence of anthropogenic
 3                  emissions in continental North America (CNA), defined here as the U.S., Canada, and
 4                  Mexico. For this document, we have focused on the sum of those background
 5                  concentrations from natural sources everywhere in the world and from anthropogenic
 6                  sources outside CNA. North American background concentrations so defined facilitate
 7                  separation of pollution that can be controlled directly by U.S. regulations or through
 8                  international agreements with neighboring countries from that which would require more
 9                  comprehensive international agreements, such as are being discussed as part of the
10                  United Nations sponsored Convention on Long Range Transboundary Air Pollution Task
11                  Force on Hemispheric Air Pollution. There is no chemical difference between
12                  background O3 and O3 attributable to CNA anthropogenic sources, and background
13                  concentrations can contribute to the risk of health effects. However, to inform policy
14                  considerations regarding the current or potential alternative standards, it is useful to
15                  understand how total O3 concentrations can be attributed to different source.

16                  Since North American background as defined above is a construct that cannot be directly
17                  measured, the range of background O3 concentrations are estimated using chemistry
18                  transport models (CTMs). The 2006 O3 AQCD provided regional  estimates of PRB O3
19                  concentrations based on a coarse resolution (2°x2.5°, or -200 km><200 km) GEOS-Chem
20                  model. For the current assessment, updated results from a finer resolution (0.5°x0.667°,
21                  or ~50 kmx50 km) GEOS-Chem model were used. Base-case model performance
22                  evaluations comparing 2006 predicted to observed mean O3 concentrations from March
23                  to August showed general agreement to within ~5 ppb at most (26 out of 28) sites
24                  investigated. Exceptions included over-prediction of mean O3 during the summer at a site
25                  on the Atlantic coast of Florida and under-prediction of mean O3 year-round at a site in
26                  Yosemite NP. The finer resolution GEOS-Chem model agrees more closely with
27                  observations in the intermountain West than earlier versions.

28                  The GEOS-Chem model-predicted North American O3 seasonal mean concentrations for
29                  spring and summer, 2006 are  shown in Figure 2-2. As can be seen, North American
30                  background concentrations are generally higher in spring than in summer across the U.S.,
31                  with exception in the Southwest where predictions peak in the summer. Highest estimates
32                  are found in the Intermountain West during the spring (less than 47 ppb) and in the
33                  Southwest during the summer (less than 49 ppb). Lowest estimates occur over the East in
34                  the spring (greater than 23 ppb) and over the Northeast in the summer (greater than
35                  15 ppb).
      Draft - Do Not Cite or Quote                      2-9                                September 2011

-------
            2.3.3   Monitoring
 1
 2

 4
 5
 6
 9
10
11
12
13
The federal reference method (FRM) for O3 measurement is based on the detection of
chemiluminescence resulting from the reaction of O3 with ethylene gas. However, almost
all of the state and local air monitoring stations (SLAMS) that reported data to the EPA's
Air Quality System (AQS) database from 2005 to 2009 used the federal equivalence
method (FEM) UV absorption photometer. More than 96% of O3 monitors met precision
and bias goals during this period.

In 2010, there were 1250 SLAMS O3 monitors reporting data to AQS. Ozone is required
to be monitored at SLAMS during the local "ozone season" which varies by state. In
addition, National Core (NCore) is a new multipollutant monitoring network
implemented to meet multiple monitoring objectives and each state is required to operate
at least one NCore site. The NCore network consists of 60 urban and 20 rural sites
nationwide (See Figure 3-16). The densest concentrations of O3 sites are located in
California and the eastern half of the U.S.
                        Spring
                                                Summer
               15       25
       Source: Zhang et al. (In Press)
                         45
55
65
75    [ppbv]
      Figure 2-2     GEOS-Chem modeled U.S. policy relevant background seasonal-
                     mean surface ozone concentrations in spring (left) and summer
                     (right), 2006.
14
15
The Clean Air Status and Trends Network (CASTNET) is a regional monitoring network
established to assess trends in acidic deposition and also provides concentration
      Draft - Do Not Cite or Quote
                             2-10
                            September 2011

-------
 1                   measurements of O3. CASTNET O3 monitors operate year round and are primarily
 2                   located in rural areas; in 2010, there were 80 CASTNET sites reporting O3 data to AQS.
 3                   The National Park Service (NFS) operates 23 CASTNET sites in national parks and other
 4                   Class-I areas, and provided data to AQS from 20 additional Portable Ozone Monitoring
 5                   Systems (POMS) in 2010 (See Figure 3-17). Compared to urban-focused monitors, rural-
 6                   focused monitors are relatively scarce across the U.S.
             2.3.4   Ambient Concentrations

 7                   Ozone is the only photochemical oxidant other than NO2 that is routinely monitored and
 8                   for which a comprehensive database exists. Other photochemical oxidants are typically
 9                   only measured during special field studies. Therefore, the concentration analyses in
10                   Chapter 3 are limited to widely available O3 data obtained directly from AQS for the
11                   period from 2007 to 2009. The median 24-h average, 8-h daily maximum, and 1-h daily
12                   maximum O3 concentrations across all U.S.  sites reporting data to AQS between 2007
13                   and 2009 were 29, 40, and 44 ppb, respectively.

14                   To investigate O3 variability in urban areas across the U.S., 20 combined statistical areas
15                   (CSAs) were selected for closer analysis based on their importance in O3 epidemiology
16                   studies and on their location. Several CSAs had relatively little spatial variability in 8-h
17                   daily maximum O3 concentrations (e.g., inter-monitor correlations ranging from 0.61 to
18                   0.96 in the Atlanta CSA) while other CSAs exhibited considerably more variability in O3
19                   concentrations (e.g.,  inter-monitor correlations ranging from -0.06 to 0.97 in the
20                   Los Angeles CSA). As a result, caution should be observed in using data from the
21                   network of ambient O3 monitors to approximate community-scale exposures.

22                   To investigate O3 variability in rural settings across the U.S., six focus areas were
23                   selected for closer analysis based on the impact of O3 or O3 precursor transport from
24                   upwind urban areas. The selected rural focus area with the largest number of available
25                   AQS monitors was Great Smoky Mountain National Park where the May-September
26                   median 8-h daily maximum O3 concentration ranged from 47 ppb at the lowest elevation
27                   (564 m) site to 60 ppb at the highest elevation (2,021  m) site. Correlations between sites
28                   within each rural focus area ranged from 0.78 to 0.92. Ozone in rural areas is produced
29                   from emissions of O3 precursors emitted directly within the rural areas, from emissions in
30                   urban areas that are processed during transport, and from occasional stratospheric
31                   intrusions. Factors contributing to variations observed within these rural focus areas
32                   include proximity to  local O3 precursor emissions, local scale circulations related to
33                   topography, and possibly stratospheric intrusions as a function of elevation. In addition,
34                   O3 tends to persist longer in rural than in urban areas  as a  result of less chemical
      Draft - Do Not Cite or Quote                       2-11                                 September 2011

-------
 1                   scavenging. This results in a more uniform O3 concentration throughout the day and night
 2                   without the typical nocturnal decrease in O3 concentration observed in urban areas.
 3                   Persistently high O3 concentrations observed at many of the rural sites investigated here
 4                   indicate that cumulative exposures for humans and vegetation in rural areas can be
 5                   substantial and often higher than cumulative exposures in urban areas.

 6                   According to the 2010 National Air Quality Status and Trends report (U.S. EPA, 2010e),
 7                   O3 concentrations have declined over the last decade; with the majority of this decline
 8                   occurring before 2004. A noticeable decrease in O3 between 2003 and 2004 coincides
 9                   with NOX emissions reductions resulting from implementation of the NOX SIP Call rule,
10                   which began in 2003 and was fully implemented in 2004. This rule was designed to
11                   reduce NOX emissions from power plants and other large combustion sources in the
12                   eastern U.S. As noted in the 2006 O3 AQCD, trends in national parks and rural areas are
13                   similar to nearby urban areas, reflecting the regional nature of O3 pollution.

14                   Since O3 is a secondary pollutant, it is not expected to be highly correlated with primary
15                   pollutants such as CO and NOX. Furthermore, O3 formation is strongly influenced by
16                   meteorology, entrainment, and transport of both O3 and O3 precursors, resulting in a
17                   broad range in correlations with other pollutants which can vary substantially with
18                   season. Correlations between 8-h daily maximum O3 and other criteria pollutants exhibit
19                   mostly negative correlations in the winter and mostly positive correlations in the summer.
20                   The median seasonal correlations are modest at best with the highest  positive correlation
21                   at 0.52 for PM2 5 in the summer and the highest negative correlation at -0.38 for PM2 5 in
22                   the winter. As a result, statistical analyses that may be sensitive to correlations between
23                   copollutants need to take seasonality into consideration, especially when O3 is being
24                   investigated.
          2.4   Human Exposure

25                   Ozone is ubiquitous throughout the environment, originating from both natural and
26                   anthropogenic sources. As such, people are routinely exposed to O3 as they participate in
27                   normal day-to-day activities. A number of factors affect the pattern of personal O3
28                   exposure. These include: the variation in O3 concentrations at various spatial and
29                   temporal scales; individuals' activity patterns, particularly time spent outdoors, which
30                   may involve changes in personal behavior to avoid exposure to O3; and infiltration of
31                   ambient O3 into indoor microenvironments, which is driven by air exchange rate.
32                   Similarly, several approaches have been used to measure or quantify exposure to ambient
33                   O3, giving an indication of the impact of these factors. These approaches include
34                   characterizing the correlation and ratio between personal exposure  and ambient O3
      Draft - Do Not Cite or Quote                       2-12                                September 2011

-------
 1                   concentration, determining the ratio between indoor and outdoor concentrations, and
 2                   using models to estimate exposure to O3 based on ambient concentrations. Both the
 3                   factors affecting the pattern of exposure as well as the type of approaches used for
 4                   quantification of exposure may have implications for epidemiologic studies.

 5                   Variations in O3 concentrations occur over multiple spatial and temporal scales. Near
 6                   roadways, O3 concentrations are reduced due to reaction with NO and other species
 7                   (Section 4.3.4.2). Over spatial scales of a few kilometers and away from roads, O3 may
 8                   be somewhat more homogeneous due to its formation as a secondary pollutant, while
 9                   over scales of tens of kilometers, additional atmospheric processing can result in higher
10                   concentrations downwind of an urban area. Although local-scale variability impacts the
11                   magnitude of O3 concentrations, O3 formation rates are influenced by factors that vary
12                   over larger spatial scales, such as temperature (Section 3.2), suggesting that urban
13                   monitors may track one another temporally but miss small-scale variability. This
14                   variation in concentrations changes the pattern of exposure people experience as they
15                   move through different microenvironments and affects the magnitude of exposures in
16                   different locations within an urban area.

17                   Another factor that may influence the pattern of exposure is the tendency for people to
18                   avoid O3 exposure by altering their behavior (e.g., reducing time spent outdoors) on high-
19                   O3 days. Activity pattern has a substantial effect on ambient O3 exposure, with time spent
20                   outdoors contributing to increased exposure (Section 4.4.2). Air quality alerts and public
21                   health recommendations induce reductions in outdoor activity on high-O3 days among
22                   some populations, particularly for children, older adults, and people with respiratory
23                   problems.  Such effects are less pronounced in the general population, possibly due to the
24                   opportunity cost of behavior modification. Preliminary epidemiologic evidence  reports
25                   increased asthma hospital admissions among children and older adults when O3 alert days
26                   were excluded from the analysis of daily hospital admissions and O3 concentrations
27                   (presumably thereby eliminating averting behavior based on high O3 forecasts). The
28                   lower rate of admissions observed when alert days were included in the analysis suggests
29                   that estimates of health effects based on dose-response functions which do not account
30                   for averting behavior may be biased towards the null.

31                   Personal exposure to O3 is moderately correlated with ambient O3 concentration, as
32                   indicated by studies reporting  correlations generally in the range of 0.3-0.8 (Table 4-2).
33                   To the extent that relative changes in central-site monitor concentration are associated
34                   with relative changes in exposure concentration, this indicates that ambient monitor
35                   concentrations are representative of day-to-day changes in average total personal
36                   exposure and in personal exposure to ambient O3. The ratio between personal exposure
37                   and ambient concentration varies widely depending on activity patterns, housing
      Draft - Do Not Cite or Quote                       2-13                                 September 2011

-------
 1                   characteristics, and season. Personal-ambient ratios are typically 0.1-0.3, although
 2                   individuals spending substantial time outdoors (e.g., outdoor workers) have shown much
 3                   higher ratios (0.5-0.9) (Table 4-3). Thus, applying personal-ambient ratios for outdoor
 4                   workers to the general population or susceptible populations spending substantial time
 5                   indoors can result in overestimates of the magnitude of personal exposure for these
 6                   groups. Some studies report much lower personal-ambient correlations, a result
 7                   attributable in part to low air exchange rate and O3 concentrations below the sampler
 8                   detection limit, conditions often encountered during wintertime. Low correlations may
 9                   also occur for individuals or populations spending increased time indoors.  Since there are
10                   relatively few indoor sources of O3, indoor O3 concentrations are often substantially
11                   lower than outdoor concentrations due to reactions of O3 with indoor surfaces and
12                   airborne constituents (Section 4.3.2). The lack of indoor sources also suggests that
13                   fluctuations in ambient O3 may be primarily responsible for changes in personal
14                   exposure, even under low-ventilation, low-concentration conditions.

15                   The factors affecting exposure patterns and quantification of exposure result in
16                   uncertainty which may contribute to exposure measurement error in epidemiologic
17                   studies. Low personal-ambient correlations are a source of exposure error for
18                   epidemiologic studies, tending to obscure the presence of thresholds, bias effect estimates
19                   toward the null, and widen confidence intervals, and this impact may be more
20                   pronounced among populations spending substantial time indoors. The impact of this
21                   exposure error may tend more toward widening confidence intervals than biasing effect
22                   estimates, since epidemiologic studies evaluating the influence of monitor selection
23                   indicate that effect estimates  are similar  across different spatial averaging scales and
24                   monitoring sites.
          2.5   Dosimetry and Mode of Action

25                   Upon inspiration, O3 uptake in the respiratory tract is affected by a number of factors
26                   including respiratory tract morphology, and breathing route, frequency, and volume.
27                   Additionally, physicochemical properties of O3 itself and how it is transported, as well as
28                   the physical and chemical properties of the extracellular lining fluid (ELF) and tissue
29                   layers in the respiratory tract can influence O3 uptake. Experimental studies and models
30                   have suggested that there are differences between the total absorption of O3 from the
31                   inhaled air and the O3 dose reaching the respiratory tract tissues. The total O3 absorption
32                   gradually decreases with distal progression into the respiratory tract. In contrast, the
33                   primary site of O3 delivery to the lung epithelium is believed to be the centriacinar region
34                   or the junction of the conducting airways with the gas exchange region.
      Draft - Do Not Cite or Quote                       2-14                                 September 2011

-------
 1                   Ozone uptake efficiency is sensitive to a number of factors including tidal volume,
 2                   minute volume, breathing frequency, O3 concentration, and exposure time. However, the
 3                   greatest source of variability in uptake efficiency is interindividual variability, primarily
 4                   due to differences in tracheobronchial volume and thus surface area. An increase in tidal
 5                   volume and breathing frequency are both associated with increased physical activity.
 6                   These changes and a switch to oronasal breathing during exercise result in deeper
 7                   penetration of O3 into the lung with a higher absorbed fraction in the upper respiratory
 8                   tract, tracheobronchial, and alveolar airways. For these reasons, increased physical
 9                   activity acts to move the maximum tissue dose of O3 distally in the respiratory tract and
10                   into the alveolar region.

11                   The ELF is a complex mixture of lipids, proteins, and antioxidants that serves as the first
12                   barrier and target for inhaled O3 (see Figure 5-8). Distinct products with diverse reactivity
13                   (i.e., secondary oxidation products), are formed by reactions of O3 with soluble ELF
14                   components or plasma membranes. The thickness of the ELF and that of the mucus layer,
15                   within the ELF, are important determinants of the dose of O3 to the tissues; a thicker ELF
16                   generally results in a lower dose of O3 to the tissues. Additionally, the quenching ability
17                   and the concentrations of antioxidants and  other ELF components are determinants of the
18                   formation of secondary oxidation products. These reactions appear to limit interaction of
19                   O3 with underlying tissues and to prevent penetration of O3 distally into the respiratory
20                   tract.

21                   In addition to contributing to the driving force for O3 uptake, formation of secondary
22                   oxidation products contributes to oxidative stress which may lead to cellular injury and
23                   altered cell signaling in the respiratory tract. Secondary oxidation products initiate
24                   pathways (See Figure  5-9) that provide the mechanistic basis for short- and long-term
25                   health effects described in detail in Chapters 6 and 7. Other key events involved in the
26                   mode of action of O3 in the respiratory tract include the activation of neural reflexes,
27                   initiation of inflammation, alterations of epithelial barrier function, sensitization of
28                   bronchial smooth muscle, modification of innate and adaptive  immunity, and airway
29                   remodeling. Another key event, systemic inflammation and vascular oxidative/nitrosative
30                   stress, may be critical to the extrapulmonary effects of O3.

31                   Secondary oxidation products can transmit signals to respiratory tract cells resulting in
32                   the activation of neural reflexes. Nociceptive sensory nerves mediate the involuntary
33                   truncation of respiration, resulting in decreases in lung function (i.e.,  FVC, FEVi, and
34                   tidal volume), and pain upon deep inspiration. Studies implicate TRPA1 receptors on
35                   bronchial C-fibers in this reflex. Another neural  reflex involves vagal sensory nerves,
36                   which mediate a mild increase in airways obstruction (i.e., bronchoconstriction)
      Draft - Do Not Cite or Quote                       2-15                                 September 2011

-------
 1                   following exposure to O3 via parasympathetic pathways. Substance P release from
 2                   bronchial C-fibers and the SP-NK receptor pathway may also contribute to this response.

 3                   Secondary oxidation products also initiate the inflammatory cascade following exposure
 4                   to O3. Studies have implicated eicosanoids, chemokines and cytokines, vascular
 5                   endothelial adhesion molecules, and tachykinins in mediating this response.
 6                   Inflammation is characterized by airways neutrophilia as well as the influx of other
 7                   inflammatory cell types. Recent studies demonstrate a later phase of inflammation
 8                   characterized by increased numbers of macrophages, which is mediated by hyaluronan.
 9                   Inflammation further contributes to O3-induced oxidative stress.

10                   Alteration of the epithelial barrier function of the respiratory tract also occurs as a result
11                   of O3-induced secondary oxidation product formation. Increased epithelial permeability
12                   may lead to enhanced sensitization of bronchial smooth muscle, resulting in  airways
13                   hyperresponsiveness (AHR). Neurally-mediated sensitization also occurs and is mediated
14                   by cholinergic postganglionic pathways and bronchial C-fiber release of substance P.
15                   Recent studies implicate hyaluronan and toll-like receptor 4 (TLR4) signaling in
16                   bronchial smooth muscle sensitization, while older studies demonstrate roles for
17                   eicosanoids, cytokines, and chemokines.

18                   Evidence is accumulating that exposure to O3 modifies innate and adaptive immunity
19                   through effects on macrophages, monocytes, and dendritic cells. Enhanced antigen
20                   presentation, adjuvant activity, and altered responses to endotoxin have been
21                   demonstrated. TLR4 signaling contributes to some of these responses. Effects on innate
22                   and adaptive immunity may result in both short- and longer-term consequences related to
23                   the exacerbation and/or induction of asthma and to alterations in host defense.

24                   Airway remodeling has been demonstrated following chronic and/or intermittent
25                   exposure to O3 by mechanisms which are not well understood. However, the TGF-(3
26                   signaling pathway has recently been implicated in O3-induced deposition of collagen in
27                   the airways wall. These studies were conducted in adult animal models and their
28                   relevance to effects in humans is unknown.

29                   Evidence is also accumulating that O3 exposure results in systemic inflammation and
30                   vascular oxidative/nitrosative stress. The release of diffusible mediators from the O3-
31                   exposed lung into the circulation may initiate or propagate inflammatory responses in the
32                   vascular or in systemic compartments. This may provide a mechanistic basis for
33                   extrapulmonary effects, such as vascular dysfunction.

34                   Both dosimetric and mechanistic factors contribute to the understanding of inter-
35                   individual variability in response. Inter-individual variability is influenced by variability
36                   in respiratory tract volume and thus surface area, breathing route, certain genetic
      Draft - Do Not Cite or Quote                       2-16                                September 2011

-------
 1                   polymorphisms, pre-existing conditions and disease, nutritional status, lifestages,
 2                   attenuation, and coexposures. In particular, functional genetic polymorphisms of genes
 3                   associated with antioxidant defense have been implicated in O3-mediated health effects.
 4                   Pre-existing asthma, allergic airways disease, and obesity modulate immune and
 5                   inflammatory responses to O3. Older adults exhibit diminished spirometric responses to
 6                   O3 compared with younger adults. Very young individuals may be sensitive to
 7                   developmental effects of O3 since studies in animal models demonstrated altered
 8                   development of lung and other organ systems.

 9                   Some of these factors are also influential in understanding species homology and
10                   sensitivity. Qualitatively, animal models exhibit a similar pattern of tissue dose
11                   distribution for O3 with the largest tissue dose delivered to the centriacinar region.
12                   However, due to anatomical and biochemical respiratory tract differences, the actual O3
13                   dose delivered differs between humans and animal models. Animal data obtained in
14                   resting conditions underestimates the dose to the respiratory tract relative to exercising
15                   humans. Further, it should be noted that, with the exception of airways remodeling, the
16                   mechanistic pathways discussed above have been demonstrated in both animals and
17                   human subjects in response to the inhalation of O3. Even though interspecies differences
18                   limit quantitative comparison between species, the short- and long-term functional
19                   responses of laboratory animals to O3 appear qualitatively homologous to those of the
20                   human making them a useful tool in determining mechanistic and cause-effect
21                   relationships with O3 exposure.
          2.6   Integration of Ozone Health Effects

22                   This section evaluates the evidence from toxicological, controlled human exposure, and
23                   epidemiologic studies that examined the health effects associated with short- and long-
24                   term exposure to O3s and summarizes the main conclusions of this assessment regarding
25                   the health effects of O3 and the concentrations at which those effects are observed. The
26                   conclusions from the previous NAAQS review and the causality determinations from this
27                   review are summarized in Table 2-1. The results from the health studies evaluated in
28                   combination with the evidence from atmospheric chemistry and exposure assessment
29                   studies contribute to the causal determinations made for the health outcomes discussed in
30                   this assessment (See Preamble to this document). In the following sections a discussion
31                   of the causal determinations will be presented by exposure duration (i.e., short-term [i.e.,
32                   hours, days, weeks] or long-term [i.e., months to years] exposure) for the health effects
33                   for which sufficient evidence was available to conclude a causal,  likely to be causal or
34                   suggestive relationship.  This section also integrates the evidence from short- and long-
35                   term exposure studies across scientific disciplines (i.e., controlled human exposure

      Draft - Do Not Cite or Quote                       2-17                                 September 2011

-------
1
2
studies, toxicology, and epidemiology) in interpreting the health effects evidence that
spans from prenatal development to death.
      Table 2-1        Summary of evidence from  epidemiologic, controlled human
                           exposure,  and animal  toxicological studies on the health effects
                           associated with short- and long-term exposure to ozone
         Health Outcome
          Conclusions from 2006 O3 AQCD
     Conclusions from 2011 2nd Draft ISA
      Short-Term Exposure to O3
      Respiratory effects         The overall evidence supports a causal relationship
                              between acute ambient 03 exposures and increased
     	respiratory morbidity outcomes.	
                                                    Causal relationship
       Lung function
     Results from controlled human exposure studies and
     animal toxicological studies provide clear evidence of
     causality for the associations observed between acute
     (£ 24 h) 03 exposure and relatively small, but
     statistically significant declines in lung function
     observed in numerous recent epidemiologic studies.
     Declines in lung function are particularly noted in
     children, asthmatics, and adults who work or exercise
     outdoors.
Recent controlled human exposure studies demonstrate
group mean decreases in FEV, in the range of 2 to 3% with
6.6 h exposures to as low as 60 ppb 03. The collective body
of epidemiologic evidence demonstrates associations
between short-term ambient 03 exposure and decrements in
lung function, particularly in asthmatics, children, and adults
who work or exercise outdoors.
       Airway
         hyperresponsiveness
     Evidence from human clinical and animal toxicological
     studies clearly indicate that acute exposure to 03 can
     induce airway hyperreactivity, thus likely placing atopic
     asthmatics at greater risk for more prolonged bouts of
     breathing difficulties due to airway constriction in
     response to various airborne allergens or other
     triggering stimuli.	
A limited number of studies have observed airway
hyperresponsiveness in rodents and guinea pigs after
exposure to less than 300 ppb 03. As previously reported in
the 2006 03 AQCD, increased airway responsiveness has
been demonstrated at 80 ppb in young, health adults, and at
50 ppb in certain strains of rats, suggesting a genetic
component.	
       Pulmonary inflammation,
         injury and oxidative
         stress
     The extensive human clinical and animal toxicological
     evidence, together with the limited available
     epidemiologic evidence, is clearly indicative of a causal
     role for03 in inflammatory responses in the airways.
Epidemiologic studies provided new evidence for
associations of ambient 03 with mediators of airway
inflammation and oxidative stress and indicate that higher
antioxidant levels may reduce pulmonary inflammation
associated with 03 exposure. Generally, these studies had
mean 8-h max 03 concentrations less than 73 ppb.	
       Respiratory symptoms
        and medication use
     Young healthy adult subjects exposed in clinical
     studies to 03 concentrations a 80 ppb for 6 to 8 h
     during moderate exercise exhibit symptoms of cough
     and pain on deep inspiration. The epidemiologic
     evidence shows significant associations between acute
     exposure to ambient 03 and increases in a wide variety
     of respiratory symptoms (e.g., cough, wheeze,
     production of phlegm, and shortness of breath) and
     medication use in asthmatic children.
The collective body of epidemiologic evidence demonstrates
positive associations between short-term exposure to
ambient 03 and respiratory symptoms (e.g., cough, wheeze,
production of phlegm, and shortness of breath) in asthmatic
children. Generally, these studies had mean 8-h max 03
concentrations less than 69 ppb.
       Lung host defenses       Toxicological studies provided extensive evidence that
                              acute 03 exposures as low as 80 to 500 ppb can cause
                              increases in susceptibility to infectious diseases due to
                              modulation of lung host defenses. A single controlled
                              human exposure study found decrements in the ability
                              of alveolar macrophages to phagocytose
                              microorganisms upon exposure to 80 to 100 ppb 03.
                                                    Recent studies in human subjects demonstrate the
                                                    increased expression of cell surface markers and alterations
                                                    in sputum leukocyte markers related to innate adaptive
                                                    immunity with short-term 03 exposures of 80400 ppb.
                                                    Recent studies demonstrating altered immune responses
                                                    and natural killer cell function build on prior evidence that 03
                                                    can affect multiple aspects of innate and acquired immunity
                                                    with short-term 03 exposures as low as 80 ppb.
      Draft - Do Not Cite or Quote
                                       2-18
                                 September 2011

-------
   Health Outcome
     Conclusions from 2006 O3 AQCD
     Conclusions from 2011 2nd Draft ISA
 Allergic and asthma
   related responses
Previous lexicological evidence indicated that 03
exposure skews immune responses toward an allergic
phenotype, and enhances the development and
severity of asthma-related responses such as AHR.
Recent studies in human subjects demonstrate enhanced
allergic cytokine production in atopic individuals and
asthmatics, increased IgE receptors in atopic asthmatics,
and enhanced markers of innate immunity and antigen
presentation in health subjects or atopic asthmatics with
short-term exposure to 80400 ppb 03, all of which may
enhance allergy and/or asthma. Further evidence for 03-
induced allergic skewing is provided by a few recent studies
in rodents using exposure concentrations as low as
200 ppb.	
  Hospital admissions, ED
   visits, and physician
   visits
Aggregate population time-series studies observed that
ambient 03 concentrations are positively and robustly
associated with respiratory-related hospitalizations and
asthma ED visits during the warm season.
Strong evidence demonstrated associations of ambient 03
with respiratory hospital admissions and ED visits in the
U.S., Europe, and Canada with supporting evidence from
single city studies. Generally, these studies had mean 8-h
max 03 concentrations less than 60 ppb.	
  Respiratory Mortality
Aggregate population time-series studies specifically
examining mortality from respiratory causes were
limited in number and showed inconsistent
associations between acute exposure to ambient 03
exposure and respiratory mortality.
Recent multicity time-series studies and a multicontinent
study consistently demonstrated associations between
ambient 03 and respiratory-related mortality visits across the
U.S., Europe, and Canada with supporting evidence from
single city studies. Generally, these studies had mean 8-h
max 03 concentrations less than 63 ppb.	
 Cardiovascular effects
The limited evidence is highly suggestive that 03
directly and/or indirectly contributes to cardiovascular-
related morbidity, but much remains to be done to
more fully substantiate the association.	
Suggestive of a Causal Relationship
 Central nervous system
 effects
Toxicological studies report that acute exposures to 03
are associated with alterations in neurotransmitters,
motor activity, short- and long-term memory, sleep
patterns, and histological signs of neurodegeneration.
Suggestive of a Causal Relationship
 Mortality
The evidence is highly suggestive that 03 directly or
indirectly contributes to non-accidental and
cardiopulmonary-related mortality.	
Likely to be a Causal Relationship
 Long-term Exposure to Oz
 Respiratory effects          The current evidence is suggestive but inconclusive for
	respiratory health effects from long-term 03 exposure.
                                                   Likely to be a Causal Relationship
  New onset asthma
No studies at this time.
Evidence for a relationship between different genetic
variants (HMOX, GST, ARC) that, in combination with 03
exposure, are related to new onset asthma. These results
were observed when subjects living in areas where the
mean annual 8-h max 03 concentration was 55.2 ppb,
compared to those who lived where it was 38.4 ppb.
 Asthma hospital
   admissions
No studies at this time.
Chronic 03 exposure was related to first childhood asthma
hospital admissions in a positive concentration-response
relationship. Generally, these studies had mean annual 8-h
max 03 concentrations less than 41 ppb.	
  Pulmonary structure and
   function
Epidemiologic studies observed that reduced lung
function growth in children was associated with
seasonal exposure to 03; however, cohort studies of
annual or multiyear 03 exposure observed little clear
evidence for impacts of longer-term, relatively low-level
03 exposure on lung function development in children.
Animal toxicological studies reported chronic 03-
induced structural alterations in  several regions of the
respiratory tract including the centriacinar region.
Morphologic evidence from studies using exposure
regimens that mimic  seasonal exposure patterns report
increased lung injury compared  to conventional chronic
stable exposures.	
Evidence for pulmonary function effects is inconclusive, with
some new epidemiologic studies (mean annual 8-h max 03
concentrations less than 65 ppb). Information from
toxicological studies indicates that long-term maternal
exposure during gestation (100 ppb) or development (500
ppb) can result in irreversible morphological changes in the
lung, which in turn can influence pulmonary function.
  Pulmonary inflammation,
   injury and oxidative
   stress
Extensive human clinical and animal toxicological
evidence, together with limited epidemiologic evidence
available, suggests a causal role for 03 in inflammatory
responses in the airways.
Several epidemiologic studies (mean 8-h max 03
concentrations less than 69 ppb) and toxicology studies (as
low as 500 ppb) add to observations of 03-induced
inflammation and injury.
 Draft - Do Not Cite or Quote
                                     2-19
                                   September 2011

-------
   Health Outcome
                                  Conclusions from 2006 O3 AQCD
    Conclusions from 2011 2nd Draft ISA
 Lung host defenses       Toxicological studies provided evidence that chronic 03
                       exposure as low as 100 ppb can cause increases in
                       susceptibility to infectious diseases due to modulation
                       of lung host defenses, but do not cause greater effects
	on infectivity than short exposures.	
                                                                         Consistent with decrements in host defenses observed in
                                                                         rodents exposed to 100 ppb 03, recent evidence
                                                                         demonstrates a decreased ability to respond to pathogenic
                                                                         signals in infant monkeys exposed to 500 ppb 03.
 Allergic responses        Limited epidemiologic evidence supported an
                       association between ambient 03 and allergic
                       symptoms. Little if any information was available from
                       toxicological studies.
                                                                         Evidence relates positive outcomes of allergic response and
                                                                         03 exposure but with variable strength for the effect
                                                                         estimates; exposure to 03 may increase total IgE in adult
                                                                         asthmatics. Allergic indicators in monkeys were increased by
                                                                         exposure to 03 concentrations of 500 ppb.	
 Respiratory mortality
                             Studies of cardio-pulmonary mortality were insufficient
                             to suggest a causal relationship between chronic 03
                             exposure and increased risk for mortality in humans.
A single study demonstrated that exposure to 03 (long-term
mean 03 less than 104 ppb) elevated the risk of death from
respiratory causes and this effect was robust to the inclusion
of PM2.5.	
Cardiovascular Effects
                             No studies at this time.
Suggestive of a Causal Relationship
Reproductive and
developmental effects
                             Limited evidence for a relationship between air
                             pollution and birth-related health outcomes, including
                             mortality, premature births, low birth weights, and birth
                             defects, with little evidence being found for 03 effects.
Suggestive of a Causal Relationship
Central nervous system
effects
                             Toxicological studies reported that acute exposures to
                             03 are associated with alterations in neurotransmitters,
                             motor activity, short and long term memory, sleep
                             patterns, and histological signs of neurodegeneration.
                             Evidence regarding chronic exposure and
                             neurobehavioral effects was not available.
Suggestive of a Causal Relationship
Cancer
                             Little evidence for a relationship between chronic 03
                             exposure and increased risk of lung cancer.	
Inadequate to infer a Causal Relationship
Mortality
                             There is little evidence to suggest a causal relationship
                             between chronic 03 exposure and increased risk for
                             mortality in humans.	
Suggestive of a Causal Relationship
 1
 2
 3
 4
 5
 6
 7

 8
 9
10
11
12
13
14
15
        2.6.1    Respiratory Effects


                  The clearest evidence for health effects associated with exposure to O3 is provided by

                  studies of respiratory effects. Collectively, there is a vast amount of evidence spanning

                  several decades that supports a causal association between exposure to O3 and a

                  continuum of respiratory effects (Figure 2-3). The majority of this evidence is derived

                  from studies investigating short-term exposure (i.e., hours to weeks) to O3, although

                  animal toxicological studies and recent epidemiologic evidence demonstrate that long-

                  term exposure (i.e., months to years) may also be detrimental to the respiratory system.

                  The 2006 O3 AQCD concluded that there was clear, consistent evidence of a causal

                  relationship between short-term exposure to O3 and respiratory health effects (U.S. EPA.

                  2006b). This causal association was substantiated by the coherence of effects observed

                  across controlled human exposure, epidemiologic, and toxicological studies indicating

                  associations of short-term O3 exposures with a range  of respiratory health endpoints from

                  respiratory tract inflammation to respiratory emergency department (ED) visits and

                  hospital admissions. Across disciplines, short-term O3 exposures induced or were

                  associated with statistically significant declines in lung function.  An equally strong body
 Draft - Do Not Cite or Quote
                                                              2-20
                               September 2011

-------
 1                   of evidence from controlled human exposure and toxicological studies demonstrated O3-
 2                   induced inflammatory responses, increased epithelial permeability, and airway
 3                   hyperresponsiveness (both specific and nonspecific). Toxicological studies provided
 4                   additional evidence for O3-induced impairment of host defenses. Combined, these
 5                   findings from experimental studies provided support for epidemiologic evidence, in
 6                   which short-term O3 exposure was consistently associated with increases in respiratory
 7                   symptoms and asthma medication use in asthmatic children, respiratory-related hospital
 8                   admissions, and asthma-related ED visits. Although O3 was consistently associated with
 9                   non-accidental and cardiopulmonary mortality, the contribution of respiratory causes to
10                   these findings was uncertain. The combined evidence across disciplines supports a causal
11                   relationship between short-term O3 exposure and respiratory effects.

12                   Mechanistic evidence for the effects of O3 on the respiratory system was characterized in
13                   the 1996 O3 AQCD, which identified O3-induced changes in a variety of lung lipid
14                   species whose numerous biologically active metabolites, in turn, can affect host defenses,
15                   lung function, and the immune system. As summarized in Section 2.5 and fully
16                   characterized in Chapter 5, key events in the toxicity pathway of O3 have been identified
17                   in humans and animal models. They include activation of neural reflexes, initiation of
18                   inflammation, alteration of epithelial barrier function, sensitization of bronchial smooth
19                   muscle, modification of innate/adaptive immunity, airway remodeling,  and systemic
20                   inflammation and oxidative/nitrosative stress.

21                   As demonstrated in Figure 2-3, O3 is associated with a continuum of respiratory effects,
22                   including altered development of the respiratory tract. Recent toxicological studies of
23                   long-term exposure to O3 occurring throughout various lifestages, beginning with
24                   prenatal and early life exposures, provide novel evidence for effects on the development
25                   of the respiratory system, including ultrastructural changes in bronchiole development,
26                   effects on the developing immune system, and increased offspring airway hyper-
27                   reactivity (Section 7.4.7. The strongest evidence for O3-induced effects on the developing
28                   lung comes from a series of experiments using infant rhesus monkeys episodically
29                   exposed to 500 ppb O3 for approximately 5 months, starting at one month of age.
30                   Functional changes in the conducting airways of infant rhesus monkeys exposed to either
31                   O3 alone or O3 + antigen were accompanied by a number of cellular and morphological
32                   changes. In addition to these functional and cellular changes, significant structural
33                   changes in the respiratory tract were observed. Importantly, the  O3-induced structural
34                   pathway changes persisted after recovery in filtered air for six months after cessation of
35                   the O3 exposures. Exposure to O3 has also been associated with similar types of
36                   alterations in pulmonary structure, including airway remodeling and pulmonary injury
37                   and increased permeability, in all adult laboratory animal species studied, from rats to
38                   monkeys (U.S. EPA.  1996a).
      Draft - Do Not Cite or Quote                       2-21                                September 2011

-------
                Short- and long-term exposure in
                 various geographic locations
/ Respiratory \ /
/ Mortality \ |
/Short- and long-term y
nergency Department Visits^" exposure in various
\ geographic locations
and Hospital Admissions \
\
new onset asthma / \ 1 :
modified by genetic ^^^N^ \
variants (e.g., HMOX, / New Onset Asthma Short- and long- *
GSTs,ARG) / \^ term exposure y
/ \arnong asthmatic
Asthma Exacerbations >T children
/
/Altered
/Development
/ of Respiratory

Altered
Morphology

Injury or
Increased
Permeability

Pulmonary
Inflammation
& Oxidative

Decreased
Host
Defenses
Decrements
in Pulmonary
Function and
Increased
Respiratory

\
Airways HyperX,
responsiveness \
/ Tract iiress Symptoms
T
Studies with
infant
monkeys,
long-term
exposure, 500
ppb














T
Studies
with
multiple
animal
species,
short- and
long-term
exposure,
200-500
ppb










T
Adults with
light
exercise, 2 h
exposure.
200 ppb

Studies with
multiple
animal
species,
short-term
exposure,
100-500 ppb







1
Short-term
exposure in
children with
and without
asthma

Adults with
exercise, 6.6 h
exposure, 60 ppb

Studies with
rodents, short-
term exposure,
100 ppb

Studies with
multiple animal
species, short-
term exposure,
100-600 ppb
t
Adults
with
exercise.
6.6 h
exposure,
80 ppb

Studies
with
rodents.
short-term
exposure,
80-100
ppb






t
Short-term exposure
among children and
adults with outdoor
activity or exercise

Adults with exercise.
6.6 h exposure, 60
ppb

Healthy adults with
heavy exercise, 1 h,
120 ppb

Adolescents with
exercise, 1 h, 140 ppb

Studies with multiple
animal species, short-
term exposure, 250-
400 ppb
1
Adults with
exercise, 6.6 h
exposure, 80
ppb

Allergic
asthmatics with
light exercise, 3
h exposure, 250
ppb

Studies with
rodents and
monkeys,
short-term
exposure, 50
ppb and higher



Green=Animal Toxicological Studies; Blue=Controlled Human Exposure Studies; Purple=Epidemiologic Studies; AM=Alveolar
Macrophage.

Figure 2-3    Snapshot of evidence for the association of Os with the continuum
               of respiratory effects, including sub-clinical effects (bottom level of
               the pyramid) and clinical effects, increasing in severity moving up
               the pyramid.
Draft - Do Not Cite or Quote
2-22
September 2011

-------
 1                   In addition to effects on the development and structure of the respiratory tract, there is
 2                   extensive evidence for the effects of short-term exposure to O3 on pulmonary
 3                   inflammation and oxidative stress. Previous evidence from controlled human exposure
 4                   studies indicated that O3 causes an inflammatory response in the lungs (U.S. EPA.
 5                   1996a). This inflammatory response to O3 was detected after a single  1-h exposure with
 6                   exercise to O3 concentrations of 300 ppb; the increased levels  of some inflammatory cells
 7                   and mediators persisted for at least 18 hours, lexicological studies provided additional
 8                   evidence for increases in permeability and inflammation in rabbits at levels as low as 100
 9                   ppb O3. Evidence summarized in the 2006 O3 AQCD demonstrated that inflammatory
10                   responses were observed subsequent to 6.6 hours O3 exposure to the lowest tested level
11                   of 80 ppb in healthy human adults, while toxicological studies provided extensive
12                   evidence that short-term (1-3 hours) O3 exposure in the range  of 100-500 ppb could cause
13                   lung inflammatory responses. The limited epidemiologic evidence reviewed in the 2006
14                   O3 AQCD demonstrated an association between short-term ambient O3 exposure and
15                   airway inflammation in children (1-h max O3 of approximately 100 ppb). Recent studies
16                   in animals and in vitro models described inflammatory and injury responses mediated by
17                   toll-like receptors (e.g.,  TLR4, TLR2), receptors for TNF or IL-1, multiple signaling
18                   pathways (e.g., p38, JNK, NFKB, MAPK/AP-1), and oxidative stress (Section 6.2.3.3).
19                   The most recent epidemiologic studies provide additional supporting evidence by
20                   demonstrating associations of ambient O3 with mediators of airway inflammation and
21                   indicating that populations with diminished antioxidant capacity may have increased
22                   susceptibility to pulmonary inflammation and oxidative stress associated with O3
23                   exposure (Sections 6.2.4 and 8.1).

24                   The normal inflammatory response in lung tissue is part of host defense that aids in
25                   removing microorganisms or particles that have reached the distal airways and alveolar
26                   surface. The 1996 O3 AQCD concluded that short-term exposure to elevated
27                   concentrations of O3 resulted in alterations in these host defense mechanisms in the
28                   respiratory system. Specifically, toxicological studies of short-term exposures as low as
29                   100 ppb O3  were shown to decrease the ability of alveolar macrophages to ingest
30                   particles, and short-term exposures as low as 80 ppb for 3 hours prevented mice from
31                   resisting infection with streptococcal bacteria and resulted in infection-related mortality.
32                   Similarly, alveolar macrophages removed from the lungs of human subjects after 6.6
33                   hours of exposure to 80 and  100 ppb O3 had decreased ability  to ingest microorganisms,
34                   indicating some impairment of host defense capability. These  altered host defense
35                   mechanisms can lead to susceptibility to respiratory infections, which are associated with
36                   increased risk of developing asthma when occurring in early life. Despite the strong
37                   toxicological evidence, in the limited body of epidemiologic evidence, O3 exposure has
38                   not been consistently associated with hospital admissions or ED visits for respiratory
39                   infection, pneumonia, or influenza (Sections 6.2.7.2 and 6.2.7.3).

      Draft - Do Not Cite or Quote                       2-23                                 September 2011

-------
 1                   The most commonly observed and strongest evidence for respiratory effects associated
 2                   with short-term exposure to O3 is transient decrements in pulmonary function. Controlled
 3                   human exposure studies characterized in previous NAAQS reviews demonstrated O3-
 4                   induced decrements in pulmonary function, characterized by alterations in lung volumes
 5                   and flow and airway resistance and responsiveness for multihour exposures (up to 8
 6                   hours) to O3 concentrations as low as 80 ppb (U.S. EPA. 1996a). A series of mobile
 7                   laboratory studies of lung function and respiratory symptoms reported pulmonary
 8                   function decrements at mean ambient O3 concentrations of 140 ppb in exercising healthy
 9                   adolescents and increased respiratory symptoms and pulmonary function decrements at
10                   150 ppb in heavily exercising athletes and at 170 ppb in lightly exercising healthy and
11                   asthmatic subjects. Epidemiologic and animal toxicological evidence is coherent with the
12                   results of the controlled human exposure studies, both indicating decrements in lung
13                   function upon O3 exposure. A combined statistical analysis of epidemiologic studies in
14                   children at summer camp demonstrated decrements in FEVi  of 0.50 mL/ppb with
15                   previous hour O3 concentration. For preadolescent children exposed to 120 ppb ambient
16                   O3, this amounted to an average decrement of 2.4-3.0% in FEVi • Key studies  of lung
17                   function measurements (FEVi) taken before and after well-defined outdoor exercise
18                   events in adults yielded exposure-response slopes of 0.40 and 1.35 mL/ppb ambient O3
19                   after exposure for up to 1 hour. Animal toxicological studies reported similar  respiratory
20                   effects in rats at exposures as low as 200 ppb O3 for 3 hours. The 2006 O3 AQCD
21                   characterized the controlled human exposure and animal toxicological studies as
22                   providing clear evidence  of causality for the associations observed between short-term (<
23                   24 hours) O3 exposure and relatively small, but statistically significant declines in lung
24                   function observed in numerous recent epidemiologic studies. Declines in lung function
25                   were particularly noted in children, asthmatics, and adults who work or exercise
26                   outdoors.

27                   Recent controlled human exposure studies examined lower concentration O3 exposures
28                   (40-80 ppb) and demonstrated that FEVi, respiratory symptoms, and inflammatory
29                   responses were affected by O3 exposures of 6.6 hours and in the range of 60 to 80 ppb
30                   (Section 6.2.1.1 and 6.2.3.1).  These studies demonstrated average decreases in FEVi in
31                   the range of 2.8 to 3.6% with O3 exposures 6.6 hours in duration and as low as 60 ppb in
32                   concentration. However,  considerable intersubject variability has been reported with
33                   some subjects experiencing considerably greater decrements than average. Recent
34                   epidemiologic studies provide greater insight into subject factors that may increase
35                   susceptibility for O3-associated respiratory morbidity. It was in these potentially
36                   susceptible populations (e.g.,  individuals with asthma with concurrent respiratory
37                   infection, older adults with AHR or elevated body mass index, or groups with diminished
38                   antioxidant capacity) that O3-associated decreases in lung function were consistently
39                   observed.

      Draft - Do Not Cite or Quote                       2-24                                 September 2011

-------
 1                   In addition to alterations in lung volumes and flow, changes in pulmonary function due to
 2                   exposure to O3 may manifest as respiratory symptoms (e.g., coughing, wheezing,
 3                   shortness of breath). The 1996 O3 AQCD identified an association between respiratory
 4                   symptoms and  increasing ambient O3, particularly among asthmatic children. In the 2006
 5                   O3 AQCD, symptoms of cough and pain on deep inspiration were well documented in
 6                   young healthy adult subjects after exposure of >80 ppb O3 for 6-8 hours during moderate
 7                   exercise. Limited data suggested an increase in respiratory symptoms down to 60 ppb.
 8                   More recently, these effects have been observed at 70 ppb in healthy adults. Controlled
 9                   human exposure studies of healthy adults, have also reported an increased incidence of
10                   cough with O3  exposures as low as 120 ppb and 1-3 hours in duration with very heavy
11                   exercise. The controlled human exposure studies also demonstrated lesser respiratory
12                   symptom responses in children and older adults relative to young healthy adults. Previous
13                   epidemiologic evidence showed significant associations between short-term exposure to
14                   ambient O3 and increases in a wide variety of respiratory symptoms (e.g., cough, wheeze,
15                   production of phlegm, and shortness of breath) in asthmatic children (U.S. EPA. 2006b).
16                   Epidemiologic studies also indicated that short-term O3 exposure is likely associated with
17                   increased asthma medication use in asthmatic children. Similar to what was observed for
18                   pulmonary function, recent epidemiologic studies provided insight into additional subject
19                   factors that may increase susceptibility for O3-associated respiratory symptoms. It was in
20                   these potentially susceptible populations (e.g., asthmatics with diminished antioxidant
21                   capacity and infants with asthmatic mothers) where the recent evidence of O3-associated
22                   increases in respiratory symptoms was the strongest.  Additionally, recent epidemiologic
23                   studies provide evidence for an association between long-term exposure to O3 and
24                   respiratory symptoms (Section 7.2.2).

25                   Ozone exposure has been shown to result in both specific and non-specific airway
26                   hyperresponsiveness. Increased airway responsiveness is an important consequence of
27                   exposure to O3 because its presence represents a change in airway smooth muscle
28                   reactivity and implies that the airways are predisposed to narrowing on inhalation of a
29                   variety of stimuli (e.g., specific allergens, SO2, cold air). Specifically, short-term (2 or
30                   3 hours) exposure to 250 or 400 ppb O3 was found to cause increases in airway
31                   responsiveness in response to allergen challenges among allergic asthmatic subjects who
32                   characteristically already had somewhat increased airway responsiveness at baseline.
33                   Increased non-specific airway responsiveness has been demonstrated in healthy young
34                   adults down to 80 ppb O3 following 6.6 hours of exposure during moderate exercise.
35                   While AHR has not been widely examined in epidemiologic studies, findings for O3-
36                   induced increases in AHR in controlled human exposure (Section 6.2.2.1) and
37                   toxicological (Section 6.2.2.2) studies provide biological plausibility for associations
38                   observed between ambient O3 exposure and increases in respiratory symptoms in subjects
39                   with asthma.

      Draft - Do Not Cite or Quote                       2-25                                September 2011

-------
 1                   In addition to asthma exacerbations, recent epidemiologic evidence has revealed an
 2                   association between long-term exposure to O3 and new onset asthma (Section 7.2.1,
 3                   Table 7-2). The new epidemiologic evidence base consists of studies using a variety of
 4                   designs and analysis methods evaluating the relationship between long-term annual
 5                   measures of exposure to ambient O3 and measures of respiratory morbidity conducted by
 6                   different research groups in different locations. Studies from two California cohorts have
 7                   provided evidence for a relationship between different variants in genes related to
 8                   oxidative or nitrosative stress (e.g., HMOX, GSTs, ARG) that, in combination with O3
 9                   exposure, are related to new onset asthma. This is the first time that evidence has
10                   extended beyond the association of exposure to O3 and asthma exacerbations to suggest
11                   that long-term exposure to O3 may play a role in the development of the disease and
12                   contribute to incident cases of asthma.

13                   When respiratory symptoms, asthma exacerbations,  or other respiratory diseases become
14                   too serious to be cared for at home, they can result in ED visits or hospital admissions.
15                   The frequency of these  types of ED visits and hospital admissions is associated with
16                   short-term changes in ambient O3 concentrations. Summertime daily hospital admissions
17                   for respiratory causes in various locations of eastern North America were consistently
18                   associated with ambient levels of O3 in studies reviewed in the 1996 O3 AQCD. This
19                   association remained even when considering only concentrations below 120 ppb O3. The
20                   2006 O3 AQCD concluded that aggregate population time-series studies demonstrate a
21                   positive and robust association between ambient O3  concentrations and respiratory-
22                   related hospitalizations and asthma ED visits during the warm season. Recent
23                   epidemiologic time-series studies that include additional multicity studies and a
24                   multicontinent study further support that short-term  exposures to ambient O3
25                   concentrations are consistently associated with increases in respiratory hospital
26                   admissions and ED visits specifically during the warm/summer months in multiple
27                   geographic locations and across a range of O3 concentrations (Section 6.2.7). There is
28                   also recent evidence for an association between respiratory hospital admissions and long-
29                   term exposure to O3 (Section 7.2.2).

30                   Finally, O3 exposure may contribute to death from respiratory causes. Recent evidence
31                   from several multicity studies and a multicontinent study demonstrate consistent positive
32                   associations between short-term exposure to ambient O3 concentrations and increases in
33                   respiratory mortality (Section 6.6.2.5). Similarly, a study of long-term exposure to
34                   ambient O3 concentrations also demonstrated an association between O3 and increases in
35                   respiratory mortality (Section 7.7.1). Evidence from these recent mortality studies is
36                   consistent and coherent with the evidence from epidemiologic, controlled human
37                   exposure, and animal toxicological studies for the effects of short- and long-term
38                   exposure to O3 on respiratory effects. Additionally, the evidence for respiratory morbidity
      Draft - Do Not Cite or Quote                       2-26                                 September 2011

-------
 1                   after short- and long-term exposure provides biological plausibility for mortality due to
 2                   respiratory disease.

 3                   In summary, recent studies support or build upon the strong body of evidence presented
 4                   in the 1996 and 2006 O3 AQCDs that short-term O3 exposure is causally associated with
 5                   adverse respiratory health effects. Recent controlled human exposure studies demonstrate
 6                   statistically significant group mean decreases in pulmonary function to exposures as low
 7                   as 60-70 ppb O3 in young, healthy adults. Equally strong evidence demonstrated
 8                   associations of ambient O3 with respiratory hospital admissions and ED visits across the
 9                   U.S., Europe, and Canada. Most effect estimates ranged from a 1.6 to 5.4% increase in
10                   daily all respiratory-related ED visits or hospital admissions in all-year analyses for
11                   standardized increases in ambient O3 concentrations.  Several multeity studies and a
12                   multicontinent study reported associations between short-term exposure to ambient O3
13                   concentrations and increases in respiratory mortality. This evidence is supported by
14                   individual-level epidemiologic studies that provide new evidence for associations of
15                   ambient O3 with mediators of airway inflammation and oxidative stress, and across
16                   endpoints, they indicate that groups with diminished antioxidant capacity or
17                   comorbidities such as atopy,  AHR, or elevated body mass index may have increased
18                   susceptibility to respiratory morbidity associated with O3  exposure. The potential
19                   susceptibility of these populations identified in recent epidemiologic studies are strongly
20                   supported by findings from experimental studies that demonstrated O3-induced decreases
21                   in intracellular antioxidant levels, increases in airway responses with co-exposures to
22                   allergens, and increases in airway responses in animal models of obesity. By
23                   demonstrating O3-induced airway hyperresponsiveness, decreased pulmonary function,
24                   allergic responses, lung injury, impaired host defense, and airway inflammation,
25                   toxicological studies have characterized O3 modes of action and have provided biological
26                   plausibility for epidemiologic associations of ambient O3  exposure with lung function
27                   and respiratory symptoms, hospital admissions, ED visits, and mortality. Together, the
28                   evidence integrated across controlled human exposure,  epidemiologic,  and toxicological
29                   studies and across the spectrum of respiratory health endpoints continues to demonstrate
30                   that there is a causal relationship between short-term  O3 exposure and respiratory
31                   health effects.

32                   The strongest evidence for a relationship between long-term O3 exposure and respiratory
33                   morbidity is contributed by recent studies from a single cohort demonstrating
34                   associations between long-term measures of O3 exposure  and new-onset asthma in
35                   children and increased respiratory symptom effects in asthmatics. While the evidence is
36                   limited, this U.S. multicommunity prospective cohort demonstrates that asthma risk is
37                   affected by interactions among genetic variability, environmental O3 exposure, and
3 8                   behavior. Other recent studies provide coherent evidence  for long-term O3 exposure and
      Draft - Do Not Cite or Quote                        2-27                                 September 2011

-------
 1                   respiratory morbidity effects such as first asthma hospitalization and respiratory
 2                   symptoms in asthmatics. Generally, the epidemiologic and toxicological evidence
 3                   provides a compelling case that supports the hypothesis that a relationship exists between
 4                   long-term exposure to ambient O3 and measures of respiratory morbidity. The evidence
 5                   for short-term exposure to O3 and effects  on respiratory endpoints provides coherence
 6                   and biological plausibility for the effects of long-term exposure to O3. Building upon that
 7                   evidence, the more recent epidemiologic evidence, combined with toxicological studies
 8                   in rodents and non-human primates, provides biologically plausible evidence that there
 9                   is likely to be a causal relationship between long-term exposure to O3 and
10                   respiratory health effects.
             2.6.2   Mortality Effects

11                   The 2006 O3 AQCD concluded that the overall body of evidence was highly suggestive
12                   that short-term exposure to O3 directly or indirectly contributes to non-accidental and
13                   cardiopulmonary-related mortality, but additional research was needed to more fully
14                   establish underlying mechanisms by which such effects occur. The evaluation of new
15                   multicity studies that examined the association between short-term O3 exposure and
16                   mortality found evidence which supports the conclusions of the 2006 O3 AQCD. These
17                   new studies reported consistent positive  associations between short-term O3 exposure and
18                   total (nonaccidental) mortality, with associations being stronger during the warm season,
19                   as well as additional support for associations between O3 exposure and cardiovascular
20                   mortality being similar or larger in magnitude compared to respiratory mortality.
21                   Additionally, these new studies examined previously identified areas of uncertainty in the
22                   O3-mortality relationship. Taken together, the body of evidence indicates that there is
23                   likely to be a causal relationship between short-term exposures to O3 and all-cause
24                   mortality.

25                   The 2006 O3 AQCD concluded that an insufficient amount of evidence existed "to
26                   suggest a causal relationship between chronic O3 exposure and increased risk for
27                   mortality in humans" (U.S. EPA. 2006b). Several additional studies have been conducted
28                   since the last review, an ecologic study that finds no association between mortality and
29                   O3, and a reanalysis of the ACS cohort that specifically points to a relationship between
30                   long-term O3 exposure and an increased risk of respiratory mortality. The findings from
31                   the reanalysis of the ACS study are consistent and coherent with the evidence  from
32                   epidemiologic, controlled human exposure, and animal toxicological studies for the
33                   effects of short- and long-term exposure to O3 on respiratory effects. Additionally, the
34                   evidence for short- and long-term respiratory morbidity provides biological plausibility
      Draft - Do Not Cite or Quote                       2-28                                September 2011

-------
 1                   for mortality due to respiratory disease. Collectively, the evidence is suggestive of a
 2                   causal relationship between long-term O3 exposures and mortality.
             2.6.3   Cardiovascular Health Effects

 3                   In past O3 AQCDs the effects of short- and long-term exposure to O3 on the
 4                   cardiovascular system could not be thoroughly evaluated due to the paucity of
 5                   information available. However, studies investigating O3-induced cardiovascular events
 6                   have advanced in the last two decades. Overall, there is limited, inconsistent evidence for
 7                   cardiovascular morbidity in epidemiologic studies examining both short- and long-term
 8                   exposure to O3. Positive associations between short-term O3 exposure and cardiovascular
 9                   mortality have been consistently reported in multiple epidemiologic studies. Animal
10                   toxicological studies provide more evidence for both short- and long-term O3 exposure
11                   leading to cardiovascular morbidity. The toxicological studies demonstrate O3-induced
12                   cardiovascular effects, specifically enhanced atherosclerosis and ischemia/reperfusion
13                   injury with or without the corresponding development of a systemic oxidative, pro-
14                   inflammatory environment, disrupted NO-induced vascular reactivity, decreased cardiac
15                   function, and increased HRV. Taking into consideration the positive toxicological studies
16                   and evidence for an association between O3 exposure and cardiovascular mortality, the
17                   generally limited body of evidence is suggestive of a causal relationship for both
18                   relevant short-  and long-term exposures to O3 and  cardiovascular effects.
             2.6.4   Central Nervous System Effects

19                   In rodents, O3 exposure has been shown to cause physicochemical changes in the brain
20                   indicative of oxidative  stress and inflammation. Recent toxicological studies add to
21                   earlier evidence that short- and long-term exposures to O3 can produce a range of effects
22                   on the central  nervous system and behavior. Previously observed effects, including
23                   neurodegeneration, alterations in neurotransmitters, short- and long-term memory, and
24                   sleep patterns, have been further supported by recent studies. In instances where
25                   pathology and behavior are both examined, animals exhibit decrements in behaviors tied
26                   to the brain regions or chemicals found to be affected or damaged. The single
27                   epidemiologic study conducted showed that long-term exposure to O3 affects memory in
28                   humans as well. Notably, exposure to O3 levels as low as 250 ppb has resulted in
29                   progressive neurodegeneration and deficits in both short- and long-term memory in
30                   rodents. Additionally, changes in the CNS, including biochemical, cellular, and
31                   behavioral effects, have been observed in animals whose sole exposure occurred in utero,
32                   at levels as a low as 300 ppb. Together the evidence from studies of short- and long-term
      Draft - Do Not Cite or Quote                       2-29                                 September 2011

-------
 1                  exposure to O3 is suggestive of a causal relationship between O3 exposure and
 2                  adverse CMS effects.
            2.6.5   Reproductive and Developmental  Effects

 3                  There is limited though positive toxicological evidence for O3-induced developmental
 4                  effects, including effects on pulmonary structure and function and central nervous system
 5                  effects after developmental exposure to O3.  Limited epidemiologic evidence exists for an
 6                  association with O3 concentration and decreased sperm concentration. A recent
 7                  toxicological study provides limited evidence for a possible biological mechanism
 8                  (histopathology showing impaired spermatogenesis and rescue with antioxidants) for
 9                  such an association. Additionally, though the evidence for an association between O3
10                  concentrations and adverse birth outcomes is generally inconsistent, there are several
11                  influential studies that indicate an association with reduced birth weight and restricted
12                  fetal growth. Overall, the evidence is suggestive of a causal relationship between
13                  long-term exposures to O3 and reproductive and developmental effects.
            2.6.6   Cancer and Mutagenicity and Genotoxicity

14                  The 2006 O3 AQCD reported that evidence did not support ambient O3 as a pulmonary
15                  carcinogen. Since the 2006 O3 AQCD, very few epidemiologic and toxicological studies
16                  have been published that examine O3 as a carcinogen, but collectively, study results
17                  indicate that O3 may contribute to DNA damage. Overall, the evidence is inadequate to
18                  determine if a causal relationship exists between ambient O3 exposures and
19                  cancer.
            2.6.7   Policy Relevant Considerations


                    2.6.7.1    Populations at Increased Risk

20                  Upon evaluating the association between short- and long-term exposure to O3 and various
21                  health outcomes, studies also attempted to identify populations that are at increased risk
22                  for O3-related health effects. These studies did so by conducting stratified epidemiologic
23                  analyses; by examining individuals with an underlying health condition, genetic
24                  polymorphism, or categorized by age, race, or sex in controlled human exposure studies;
25                  or by developing animal models that mimic the pathophysiological conditions associated
26                  with an adverse health effect. These studies identified a multitude of factors that could

      Draft - Do Not Cite or Quote                      2-30                               September 2011

-------
 1                   potentially contribute to whether an individual is at increased risk for O3-related health
 2                   effects. The examination of at risk populations for O3 exposure allows for the NAAQS to
 3                   provide an adequate margin of safety for both the general population and for sensitive
 4                   populations.

 5                   The populations identified in Chapter 8 that are most at risk for O3-related health effects
 6                   are individuals with influenza/infection, individuals with asthma, and younger and older
 7                   age groups. There were a small number of studies on influenza/infection but both
 8                   reported influenza/infection to modify the association between O3 exposure  and
 9                   respiratory effects, with individuals having influenza or an infection being at increased
10                   risk. Asthma as a factor affecting risk was supported by controlled human exposure and
11                   toxicological studies, as well as some evidence from epidemiologic studies. Most studies
12                   comparing age groups reported greater effects of short-term O3 exposure on mortality
13                   among older adults, although studies of other health outcomes had inconsistent findings
14                   regarding whether older adults were at increased risk. Generally, studies of age groups
15                   also reported positive associations for respiratory hospital admissions and ED visits
16                   among children. Biological plausibility for this increased risk is supported by
17                   toxicological and clinical research. Diet and obesity are also both likely factors that affect
18                   risk. Multiple epidemiologic, controlled human exposure, and toxicological studies
19                   reported that diets deficient in vitamins E and C are associated with risk of O3 -related
20                   health effects. Similarly, studies of effect measure modification by body mass index
21                   (BMI) observed greater O3 -related respiratory decrements for individuals who were
22                   obese.

23                   Other potential  factors [preexisting conditions (such as chronic obstructive pulmonary
24                   disease and cardiovascular disease), sex, and multiple genes  (such as GSTM1, GSTP1,
25                   HMOX-1, NQO1, and TNF-a)] provided some evidence of susceptibility, but further
26                   investigation is warranted. In addition, examination of modification of the associations
27                   between O3 exposure and health effects by SES and race were available in a limited
28                   number of studies, and demonstrated  possible increased odds of health effects related to
29                   O3 exposure among those with low SES and black race.

30                   Individuals with increased ambient exposure were examined in a recent study of outdoor
31                   workers, in which no effect modification was observed, and in studies of air conditioning
32                   prevalence, which demonstrated inconsistent findings. However, previous evidence along
33                   with biological  plausibility from toxicological and controlled human studies has shown
34                   individuals exposed to more outdoor air to be at increased risk of O3-related health
35                   effects. Studies of physical conditioning and smoking were conducted but little evidence
36                   was available to determine whether increased risk of O3-related health effects is present
37                   for these factors. The only studies examining effect measure modification by diabetes
      Draft - Do Not Cite or Quote                       2-31                                 September 2011

-------
 1                   examined O3 exposure and cardiovascular outcomes and none reported increased risks for
 2                   individuals with diabetes. Toxicological studies also identified hyperthyroidism to be a
 3                   factor warranting further examination. Future research will provide additional insight into
 4                   whether these factors affect risk of O3-related health effects.
                     2.6.7.2   Lag Structure in Epidemiologic Studies

 5                   Epidemiologic studies have attempted to identify the time-frame in which exposure to O3
 6                   can impart a health effect. Although O3 exposure-response relationships have
 7                   traditionally been examined using air quality data for a defined lag period (e.g., 1 day or
 8                   average of 0-1 days), the relationship can potentially be influenced by a multitude of
 9                   factors, such as the underlying susceptibility of an individual (e.g., age, pre-existing
10                   diseases), which could increase or decrease the lag times observed. Different lag times
11                   have been evaluated for specific health outcomes.

12                   The epidemiologic evidence evaluated in the 2006 O3 AQCD indicated that one of the
13                   remaining uncertainties in characterizing the O3-mortality relationship was identifying
14                   the appropriate lag structure (e.g., single-day lags versus distributed lag model). An
15                   examination of lag times used in the epidemiologic studies evaluated in this assessment
16                   can provide further insight on the relationship between O3 exposure and morbidity and
17                   mortality outcome s.

18                   Collectively, recent epidemiologic studies of lung function, respiratory symptoms, and
19                   biological markers of airway inflammation and oxidative stress examined associations
20                   with single-day ambient O3 exposures (using various averaging times) lagged from 0 to 7
21                   days as well as concentrations averaged over 2 to 19 days. Lags of 0 and 1 day ambient
22                   O3 exposures were associated with decreases in lung function and increases in respiratory
23                   symptoms, airway inflammation, and oxidative stress. Additionally, several studies found
24                   that multiday averages of O3 exposure were associated with these endpoints, indicating
25                   that not only single day, but exposures accumulated over several days led to a respiratory
26                   health effect. In studies of respiratory hospital admissions and ED visits, investigators
27                   either examined the lag structure of associations  by including both single-day and the
28                   average of multiday lags, or selecting lags a priori. Of the studies evaluated, the
29                   collective evidence indicates a rather immediate  response within the first few days of O3
30                   exposure (i.e., for lags days averaged at 0-1, 0-2, and 0-3 days) for hospital admissions
31                   and ED visits for all respiratory outcomes, asthma, and chronic obstructive pulmonary
32                   disease in all-year and seasonal analyses.

33                   The majority of epidemiologic studies that focused on the association between short-term
34                   O3 exposure and mortality (i.e., all-cause, respiratory and cardiovascular) examined the


      Draft - Do Not Cite or Quote                      2-32                                 September 2011

-------
 1                   average of multiday lags with some studies examining single-day lags. Across a range of
 2                   multiday lags (i.e., average of 0-1 to 0-6 days), the studies evaluated consistently
 3                   demonstrate that the O3 effects on mortality occur within a few days of exposure (Figure
 4                   6-28). Additionally, several recent studies have conducted more extensive analysis of lag
 5                   structure to investigate "mortality displacement" (i.e., deaths are occurring in frail
 6                   individuals and exposure is only moving the day of death to a day slightly earlier), which
 7                   also inform upon the lag structure of associations (Section 6.6.2.4).  Collectively, these
 8                   studies suggest that the positive associations between O3 and mortality are observed
 9                   mainly in the first few days after exposure.
                     2.6.7.3    Ozone Concentration-Response Relationship

10                   An important consideration in characterizing the O3-morbidity and mortality association
11                   is whether the C-R relationship is linear across the full concentration range that is
12                   encountered or if there are concentration ranges where there are departures from linearity
13                   (i.e., nonlinearity). In this ISA studies have been identified that attempt to characterize
14                   the shape of the O3 C-R curve along with possible O3 "thresholds" (i.e., O3 levels which
15                   must be exceeded in order to elicit a health response). The controlled human exposure
16                   and epidemiologic studies that examined the shape of the C-R curve and the potential
17                   presence of a threshold have  indicated a generally linear C-R function with no indication
18                   of a threshold for O3 concentrations greater than 30 or 40 ppb, which corresponds with
19                   PRB and the lower bound of O3 concentrations included in the C-R functions.

20                   Controlled human exposure studies have provided strong and quantifiable C-R data on
21                   the human health effects of O3. The magnitude of respiratory effects in these studies is
22                   generally a function of O3 exposure, i.e., the product of concentration (C), minute
23                   ventilation (VE), and exposure duration. Recent studies provide evidence  for a smooth C-
24                   R curve without indication of a threshold in young healthy adults, exposed during
25                   moderate exercise for 6.6 hours to O3  concentrations between 40 and 120 ppb
26                   (Figure 6-1).

27                   Although relatively few epidemiologic studies have examined the O3-health effects C-R
28                   relationship, the C-R relationship has  been examined across multiple health endpoints
29                   and exposure durations. Some studies of populations engaged in outdoor  activity found
30                   that associations between O3  and lung function decrements persisted at lower O3
31                   concentrations (Table 6-5). For example, a study found ambient O3 exposure (10-min to
32                   1-h)  during outdoor exercise  to be associated with decreases in  lung function in analyses
33                   restricted to concentrations less than 51 ppb, though effect estimates were near zero with
34                   O3 concentrations less than 41 ppb. In contrast, a subsequent study found associations
      Draft - Do Not Cite or Quote                       2-33                                 September 2011

-------
 1                   persisted with 1-h max O3 concentrations less than 40 ppb. A study examining the C-R
 2                   relationship between short-term O3 exposure and pediatric asthma ED visits found no
 3                   evidence of a threshold. In both quintile and loess dose-response analyses this study
 4                   found evidence that suggests that there are elevated associations for pediatric asthma ED
 5                   visits with O3 concentrations as low as 30 ppb (Figure 6-11). In an additional study,
 6                   authors used a smooth function while also accounting for the potential confounding
 7                   effects of PM2 5, to examine whether the shape of the C-R curve for short-term exposure
 8                   to O3 and asthma hospital admissions (i.e., both general and ICU for all ages) is linear.
 9                   When comparing the curve to a linear fit, the authors found that the linear fit is a
10                   reasonable approximation of the C-R relationship between O3 and asthma hospital
11                   admissions around and below the current NAAQS (Figure 6-9). Although the  C-R
12                   relationship between short-term O3 exposure and respiratory-related hospital admissions
13                   and ED visits has not been extensively examined, these preliminary examinations
14                   indicate a linear, no threshold relationship between short-term O3 exposure and pediatric
15                   asthma ED visits and asthma hospitalizations.

16                   The O3-health effects C-R relationship was further examined in studies of short-term O3
17                   exposure and mortality. Evaluation of the C-R relationship for short-term exposure to O3
18                   and mortality is difficult due to the evidence from multicity studies indicating highly
19                   heterogeneous O3-mortality associations across regions of the U.S. In addition, there are
20                   numerous issues that may influence the shape of the O3-mortality C-R relationship that
21                   need to be taken into consideration including: multiday effects (distributed lags),
22                   potential adaptation and mortality displacement (i.e., hastening  of death by a short
23                   period). Several recent studies applied a variety of statistical approaches to examine the
24                   shape of the O3-mortality C-R relationship and whether a threshold exists. These studies
25                   did not find any evidence that supports a threshold for the association between short-term
26                   exposure to  O3 and mortality within a range of O3 concentrations observed in the U.S.
27                   Recent evidence also suggests that the shape of the O3-mortality C-R curve remains
28                   linear across the full range of the O3 concentrations. However, studies have also
29                   demonstrated heterogeneity in the O3-mortality relationship across cities (or regions),
30                   which complicates the interpretation of a combined C-R curve and threshold analysis.
31                   Additionally, given the effect modifiers identified in mortality analyses that are also
32                   expected to vary regionally (e.g., temperature, air conditioning prevalence), a national or
33                   combined analysis may not be appropriate to identify whether a threshold exists in the
34                   O3-mortality C-R relationship.

35                   An evaluation of long-term exposure studies identified studies of long-term exposure to
36                   O3 and birth outcomes that have characterized the C-R relationship. Evidence  from the
37                   southern California Children's Health Study identified a C-R relationship of birth weight
3 8                   with 24-h avg O3 concentrations averaged over the entire pregnancy that was clearest
      Draft - Do Not Cite or Quote                       2-34                                 September 2011

-------
 1                   above the 30 ppb level (Figure 7-4). Relative to the lowest decile of 24-h avg O3,
 2                   estimates for the next 5 lowest deciles were approximately -40 g to -50 g, with no clear
 3                   trend and with 95% confidence bounds that included zero. The highest four deciles of O3
 4                   exposure showed an approximately linear decrease in birth weight, and all four 95% CIs
 5                   excluded zero, and ranged from mean decreases of 74 grams to decreases of 148 grams.
 6                   Another study conducted in southern  California reported increased risks for cardiac birth
 7                   defects in a dose-response manner with second-month O3 exposure.

 8                   Collectively, both short- and long-term exposure studies that examined the O3-health
 9                   effects C-R relationship have provided no evidence of a threshold. Additionally, these
10                   studies indicate a linear C-R relationship across the full range of O3 concentrations
11                   observed in the U.S.
          2.7   Integration of Effects on Vegetation and Ecosystems

12                   Chapter 9 presents the most policy-relevant information related to this review of the
13                   NAAQS for the effects of O3 on vegetation and ecosystems. This section integrates the
14                   key findings from the disciplines evaluated in this assessment of the O3 scientific
15                   literature, which includes plant physiology, whole plant biology, ecosystems, and
16                   exposure-response.

17                   Ozone effects at small spatial scales, such as the leaf of an individual plant, can result in
18                   effects at a continuum of larger spatial scales. Figure 2-4 is a simplified illustrative
19                   diagram of the major pathway through which O3 enters leaves and the major endpoints O3
20                   may affect in vegetation and ecosystems. The sections of Chapter 9 are organized
21                   according to increasing spatial scales, starting with the cellular and subcellular level, then
22                   the whole plant and finally, ecosystem-level processes. Ozone enters leaves through
23                   stomata, and can alter stomatal conductance and disrupt CO2 fixation (Section 9.3). These
24                   effects can change rates of leaf gas exchange, growth and reproduction at the individual
25                   plant level and result in changes in ecosystems, such as productivity, C storage, water
26                   cycling, nutrient cycling, and community composition (Section 9.4). The framework for
27                   causal determinations has been applied to the body of scientific evidence to collectively
28                   examine effects attributed to O3 exposure (Table 2-2). The summary below provides brief
29                   integrated summaries of the evidence that supports the causal determinations. The
30                   detailed discussion of the underlying evidence used to formulate each causal
31                   determination can be found in Chapters 9. This summary ends with a short discussion of
32                   policy relevant considerations.
      Draft - Do Not Cite or Quote                       2-35                                 September 2011

-------
       2.7.1   Visible Foliar Injury
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
               Visible foliar injury resulting from exposure to O3 has been well characterized and
               documented over several decades of research on many tree, shrub, herbaceous, and crop
               species (U.S. EPA. 2006b. 1996b. 1984. 1978a) (Section 9.4.2). Ozone-induced visible
               foliar injury symptoms on certain bioindicator plant species are considered diagnostic as
               they have been verified experimentally in exposure-response studies, using exposure
               methodologies such as continuous stirred tank reactors (CSTRs), open-top chambers
               (OTCs), and free-air fumigation. Experimental evidence has clearly established a
               consistent association of visible injury with O3 exposure, with greater exposure often
               resulting in greater and more prevalent injury. Since the 2006 O3 AQCD, several
               multiple-year field surveys of O3-induced visible foliar injury have been conducted at
               National Wildlife Refuges in Maine, Michigan, New Jersey, and South Carolina. New
               sensitive species showing visible foliar injury continue to be identified from field surveys
               and verified in controlled exposure studies.
                      O3 exposure
                  O3 uptake & physiology (Fig 9-2)
                  •Antioxidant metabolism up-regulated
                  •Decreased photosynthesis
                  •Decreased stomatal conductance
                  or sluggish stomatal response
                Effects on leaves
                •Visible leaf injury
                •Altered leaf production
                •Altered leaf chemical composition
                  Plant growth (Fig 9.8)
                  •Decreased biomass accumulation
                  •Altered reproduction
                  •Altered carbon allocation
                  •Altered crop quality
                 Belowground processes (Fig 9.8)
                 •Altered litter production and decomposition
                 •Altered soil carbon and nutrient cycling
                 •Altered soil fauna and microbial communities

                                                            (D
                                                            3
                                                            a.
                                                                     Affected ecosystem services
                                                                     •Decreased productivity
                                                                     •Decreased C sequestration
                                                                     •Altered water cycling (Fig 9-7)
                                                                     •Altered community composition
                                                                     (i.e., plant, insect & microbe)
Figure 2-4     An illustrative diagram of the major pathway through which O3
                 enters leaves and the major endpoints that Os may affect in plants
                 and ecosystems.
Draft - Do Not Cite or Quote
                                                      2-36
September 2011

-------
      Table 2-2       Summary of ozone causal determinations for vegetation and
                       ecosystem effects
          Vegetation and
         Ecosystem Effects
                    Conclusions from 2006 O3 AQCD
  Conclusions from
  2011 2nd Draft ISA
      Visible Foliar Injury Effects on
      Vegetation
        Data published since the 1996 ft AQCD strengthen previous conclusions that
        there is strong evidence that current ambient 03 concentrations cause impaired
        aesthetic quality of many native plants and trees by increasing foliar injury.
Causal Relationship
Reduced Vegetation Growth
Reduced Productivity in
Terrestrial Ecosystems
Reduced Carbon (C)
Sequestration in Terrestrial
Ecosystems
Reduced Yield and Quality of
Agricultural Crops
Alteration of Terrestrial
Ecosystem Water Cycling
Alteration of Below-ground
Biogeochemical Cycles
Alteration of Terrestrial
Community Composition
Data published since the 1996 ft AQCD strengthen previous conclusions that
there is strong evidence that current ambient 03 concentrations cause
decreased growth and biomass accumulation in annual, perennial and woody
plants, including agronomic crops, annuals, shrubs, grasses, and trees.
There is evidence that 03 is an important stressor of ecosystems and that the
effects of 03 on individual plants and processes are scaled up through the
ecosystem, affecting net primary productivity.
Limited studies from previous review
Data published since the 1996 ft AQCD strengthen previous conclusions that
there is strong evidence that current ambient 03 concentrations cause
decreased yield and/or nutritive quality in a large number of agronomic and
forage crops.
Ecosystem water quantity may be affected by 03 exposure at the landscape
level.
Ozone-sensitive species have well known responses to 03 exposure, including
altered C allocation to below-ground tissues, and altered rates of leaf and root
production, turnover, and decomposition. These shifts can affect overall C and N
loss from the ecosystem in terms of respired C, and leached aqueous dissolved
organic and inorganic C and N.
Ozone may be affecting above- and below -ground community composition
through impacts on both growth and reproduction. Significant changes in plant
community composition resulting directly from 03 exposure have been
demonstrated.
Causal Relationship
Causal Relationship
Likely to be a Causal
Relationship
Causal Relationship
Likely to be a Causal
Relationship
Causal Relationship
Likely to be a Causal
Relationship
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
 The use of biological indicators in field surveys to detect phytotoxic levels of O3 is a
longstanding and effective methodology. The USDA Forest Service through the Forest
Health Monitoring (FHM) Program (1990-2001) and currently the Forest Inventory and
Analysis (FIA) Program has been collecting data regarding the incidence and severity of
visible foliar injury on a variety of O3 sensitive plant species throughout the U.S. The
network has provided evidence that O3 concentrations were high enough to induce visible
symptoms on sensitive vegetation. From repeated observations and measurements made
over a number of years, specific geographical patterns of visible O3 injury symptoms can
be identified. In addition, a study assessed the  risk of O3-induced visible foliar injury on
bioindicator plants in 244 national parks in support of the National Park Service's Vital
Signs Monitoring Network. The results of the study demonstrated that the risk of visible
foliar injury was high in 65 parks (27%), moderate in 46 parks (19%), and low in 131
parks (54%). Some of the well4mown parks with a high risk of O3-induced visible foliar
injury include Gettysburg, Valley Forge, Delaware Water Gap, Cape Cod, Fire Island,
Antietam, Harpers  Ferry, Manassas, Wolf Trap Farm Park, Mammoth Cave, Shiloh,
      Draft - Do Not Cite or Quote
                                2-37
      September 2011

-------
 1                  Sleeping Bear Dunes, Great Smoky Mountains, Joshua Tree, Sequoia and Kings Canyon,
 2                  and Yosemite. Overall, evidence is sufficient to conclude that there is a causal
 3                  relationship between ambient O3 exposure and the occurrence of O3-induced
 4                  visible foliar injury on sensitive vegetation across the U.S.
            2.7.2   Growth, Productivity, Carbon Storage and Agriculture

 5                  Ambient O3 concentrations have long been known to cause decreases in photosynthetic
 6                  rates and plant growth. The O3-induced damages at the plant scale may translate to the
 7                  ecosystem scale, and cause changes in productivity and C storage. The effects of O3
 8                  exposure on photosynthesis, growth, biomass allocation, ecosystem production and
 9                  ecosystem C sequestration were reviewed for the natural ecosystems, and crop
10                  productivity and crop quality were reviewed for the agricultural ecosystems.
                    2.7.2.1    Natural Ecosystems

11                  The previous O3 AQCDs concluded that there is strong and consistent evidence that
12                  ambient concentrations of O3 decrease plant photosynthesis and growth in numerous
13                  plant species across the U.S. Studies published since the last review continue to support
14                  that conclusion (Section 9.4.3.1). New studies, based on the Aspen free-air carbon-
15                  dioxide/ozone enrichment (FACE) experiment, found that O3 caused reductions in total
16                  biomass relative to the control in aspen, paper birch, and sugar maple communities
17                  during the first seven years of stand development. Overall, the  studies at the  Aspen FACE
18                  experiment were consistent with the open-top chamber (OTC)  studies that were the
19                  foundation of previous O3 NAAQS reviews. These results strengthen our understanding
20                  of O3 effects on forests and demonstrate the relevance of the knowledge gained from
21                  trees grown in open-top chamber studies.

22                  A set of meta-analyses assessed the effects of O3 on plant photosynthesis and growth
23                  across different species and fumigation methods (such as OTC and FACE). Those studies
24                  reported that current O3 concentrations in the northern hemisphere are decreasing
25                  photosynthesis (~11%) across tree species, and the decreases in photosynthesis are
26                  consistent with cumulative uptake of O3 into the leaf. The current ambient O3
27                  concentrations (~40 ppb) significantly decreased annual total biomass growth of forest
28                  species by an average of 7%, with potentially greater decreases (11-17%) in  areas that
29                  have higher O3 concentrations (Section 9.4.3.1). The meta-analyses further confirmed
30                  that reduction of plant photosynthesis and growth under O3 exposure are coherent across
31                  numerous species  and various experimental techniques.
      Draft - Do Not Cite or Quote                       2-38                                September 2011

-------
 1                   Studies during recent decades have also demonstrated O3 alters biomass allocation and
 2                   plant reproduction (Section 9.4.3). Recent meta-analyses have generally indicated that O3
 3                   reduced C allocated to roots, although the findings of individual studies were mixed.
 4                   Several recent studies published since the 2006 O3 AQCD further demonstrate that O3
 5                   altered reproductive processes, such as timing of flowering, number of flowers, fruits and
 6                   seeds, in herbaceous and woody plant species.  However, a knowledge gap still exists
 7                   pertaining to the exact mechanism of the responses of reproductive processes to O3
 8                   exposure (Section 9.4.3.3).

 9                   Studies at the leaf and plant scales showed that O3 reduced photosynthesis and plant
10                   growth, providing coherence and biological plausibility for the reported decreases in
11                   ecosystem productivity. During the previous NAAQS reviews, there were very few
12                   studies that investigated the effect of O3 exposure  on ecosystem productivity and  C
13                   sequestration. Recent studies from long-term FACE experiments and ecosystem models
14                   provided evidence of the association of O3 exposure and reduced productivity at the
15                   ecosystem scale. Elevated O3 reduced stand biomass at Aspen FACE after 7 years of O3
16                   exposure, and annual volume growth at the Kranzberg Forest in Germany. Results across
17                   different ecosystem models were consistent with the FACE experimental  evidence, which
18                   showed that O3 reduced ecosystem productivity (Section 9.4.3.4). In addition to primary
19                   productivity, other indicators such as net ecosystem CO2 exchange (NEE) and C
20                   sequestration were often assessed by model studies. Model simulations consistently
21                   found that O3 exposure caused negative impacts on those indicators (Section 9.4.3.4,
22                   Table 9-3), but the severity of these impacts  was influenced by multiple interactions of
23                   biological and environmental factors. The suppression of ecosystem C sinks results in
24                   more CO2 accumulation in the  atmosphere. A recent study suggested that the indirect
25                   radiative forcing caused by O3  exposure through lowering ecosystem C sink could have
26                   an even greater impact on global warming than the direct radiative forcing of O3.

27                   Although O3 generally causes negative effects  on ecosystem productivity, the magnitude
28                   of the response varies among plant communities (Section 9.4.3.4). For example, O3 had
29                   little impact on white fir, but greatly reduced growth of ponderosa pine in southern
30                   California. Ozone decreased net primary production (NPP) of most forest types in Mid-
31                   Atlantic region, but had small impacts on spruce-fir forest. Ozone could also affect
32                   regional C budgets through interacting with multiple factors, such as N deposition,
33                   elevated CO2 and land use history. Model simulations suggested that O3 partially  offset
34                   the growth stimulation caused by elevated CO2 and N deposition in both Northeast- and
35                   Mid-Atlantic-region forest ecosystems of the U.S.
36                   Overall, evidence is sufficient to conclude that there is a causal relationship
37                   between O3 exposure and reduced plant growth and productivity, and a likely
      Draft - Do Not Cite or Quote                       2-39                                September 2011

-------
 1                   causal relationship between O3 exposure and reduced carbon sequestration in
 2                   terrestrial ecosystems.
                     2.7.2.2    Agricultural Crops

 3                   The detrimental effect of O3 on crop production has been recognized since the 1960's and
 4                   a large body of research has subsequently stemmed from those initial findings. Previous
 5                   O3 AQCDs have extensively reviewed this body of literature. Current O3 concentrations
 6                   across the U.S. are high enough to cause yield loss for a variety of agricultural crops
 7                   including, but not limited to, soybean, wheat, potato, watermelon, beans, turnip, onion,
 8                   lettuce, and tomato (Section 9.4.4.1). Continued increases in O3 concentration may
 9                   further decrease yield in these sensitive crops. Despite the well-documented yield losses
10                   due to increasing O3 concentration, there is still a knowledge gap pertaining to the exact
11                   mechanism of O3-induced yield loss. Research has linked increasing O3 concentration to
12                   decreased photosynthetic rates and accelerated senescence, which are related to yield.

13                   In addition, new research has highlighted the effects of O3 on crop quality. Increasing O3
14                   concentration decreases nutritive quality of grasses, decreases macro- and micro-nutrient
15                   concentrations in fruits and vegetable crops, and decreases cotton fiber quality. These
16                   areas of research require further investigation to determine the mechanism and dose-
17                   responses (Section 9.4.4.2).

18                   During the previous NAAQS reviews, there were very few studies that estimate O3
19                   impacts on crop yields at large spatial scales. Recent modeling studies found that O3
20                   generally reduced crop yield, but the impacts varied across regions and crop species
21                   (Section 9.4.4.1). For example, the largest O3-induced crop yield losses occurred in high-
22                   production areas exposed to high O3 concentrations, such as the Midwest and the
23                   Mississippi Valley regions of the U.S. Among crop species, the estimated yield loss for
24                   wheat and soybean were higher than rice and maize. Satellite and ground-based O3
25                   measurements have been used to assess yield loss caused by O3 over the continuous tri-
26                   state area of Illinois, Iowa and Wisconsin. The results showed that O3 concentrations
27                   significantly reduced soybean yield, which correlates well with the previous results from
28                   FACE-type experiments and OTC experiments (Section 9.4.4.1).
29                   Evidence is sufficient to conclude that there is a causal relationship between O3
30                   exposure and reduced yield and quality of agricultural  crops.
      Draft - Do Not Cite or Quote                       2-40                                 September 2011

-------
            2.7.3  Water Cycling

 1                  Ozone can affect water use in plants and ecosystems through several mechanisms
 2                  including damage to stomatal functioning and loss of leaf area. Section 9.3.6 reviewed
 3                  possible mechanisms for O3 exposure effects on stomatal functioning including the build-
 4                  up of CO2 in substomatal cavity, impacts on signal transduction pathways and direct O3
 5                  impact on guard cells. Regardless of the mechanism, O3 exposure has been shown to alter
 6                  stomatal performance, which may affect plant and stand transpiration and therefore
 7                  possibly affecting hydrological cycling.

 8                  Although the evidence was from a limited number of field and modeling studies, these
 9                  findings showed an association of O3 exposure and the alteration of water use and cycling
10                  in vegetation and ecosystem level (Section 9.4.5). There is not a clear consensus on the
11                  nature of Ieaf4evel stomatal conductance response to O3 exposure. When measured at
12                  steady-state high light conditions, leaf-level stomatal conductance is often found to be
13                  reduced when exposed to O3. However, measurements of stomatal conductance under
14                  dynamic light and vapor pressure deficit conditions indicate sluggish responses under
15                  elevated O3 exposure which could potentially lead to increased water loss from
16                  vegetation. Field studies suggested that peak hourly O3 exposure increased the rate of
17                  water loss from several tree species, and led to a reduction in the late-season modeled
18                  stream flow in three forested watersheds in eastern Tennessee. Sluggish stomatal
19                  responses during O3 exposure was suggested as a possible mechanism for increased water
20                  loss during peak O3 exposure. Currently, the O3-induced reduction in stomatal aperture is
21                  the biological assumption for most process-based models. Therefore, results of those
22                  models normally found that O3  reduced water loss. For example, one study found that O3
23                  damage and N limitation together reduced evapotranspiration and increase runoff.

24                  Although the direction of the response differed among studies, the evidence is
25                  sufficient to conclude that there is likely  to be a causal relationship between O3
26                  exposure and the alteration of ecosystem water cycling.
            2.7.4   Below-Ground Processes

27                   Below-ground processes are tightly linked with aboveground processes. The responses of
28                   aboveground process to O3 exposure, such as reduced photosynthetic rates, increased
29                   metabolic cost, and reduced root C allocation, have provided biologically plausible
30                   mechanisms for the alteration of below-ground processes. Since the 2006 O3 AQCD,
31                   more evidence has shown that although the responses are often species specific, O3
32                   altered the quality and quantity of C input to soil, microbial community composition, and
33                   C and nutrient cycling.

      Draft - Do Not Cite or Quote                      2-41                                 September 2011

-------
 1                   Results from Aspen FACE and other experimental studies consistently found that O3
 2                   reduced litter production and altered C chemistry, such as soluble sugars, soluble
 3                   phenolics, condensed tannins, lignin, and macro/micro nutrient concentration in litter
 4                   (Section 9.4.6.1). The changes in substrate quality and quantity could alter microbial
 5                   metabolism under elevated O3, and therefore soil C and nutrient cycling. Several studies
 6                   indicated that O3 generally suppressed soil enzyme activities (Section 9.4.6.2). However,
 7                   the impact of O3 on litter decomposition was inconsistent and varied among species, sites
 8                   and exposure length. Similarly, O3 had inconsistent impacts on dynamics of micro and
 9                   macro nutrients (Section 9.4.6.4).

10                   Studies from the Aspen FACE experiment suggested that the response of below-ground
11                   C cycle to O3 exposure, such as litter decomposition, soil respiration and soil C content,
12                   changed over time. For example, in the early part of the experiment (1998-2003), O3 had
13                   no impact on soil respiration but reduced the formation rates of total soil C under
14                   elevated CO2. However, after 10 to  11 years of exposure, O3 was found to increase soil
15                   respiration but have no significant impact on soil C formation under elevated CO2
16                   (Section 9.4.6.3).

17                   The evidence is sufficient to infer that there is a causal relationship between O3
18                   exposure and the alteration of below-ground biogeochemical cycles.
            2.7.5   Community Composition

19                   In the 2006 O3 AQCD, the impact of O3 exposure on species competition and community
20                   composition was assessed. Ozone was found to be one of the dominant factors causing a
21                   significant decline in ponderosa and Jeffrey pine in the San Bernardino Mountains in
22                   southern California. Ozone exposure also tended to shift the grass-legume mixtures in
23                   favor of grass species. Since the 2006 O3 AQCD, more evidence has shown that O3
24                   exposure changed the competitive interactions and led to loss of O3 sensitive species or
25                   genotypes. Studies found that the severity of O3 damage on growth, reproduction and
26                   foliar injury varied among species (Section 9.4.3), which provided the biological
27                   plausibility for the alteration of community composition. Additionally, research since the
28                   last review has shown that O3 can alter community composition and diversity of soil
29                   microbial communities.

30                   The decline of conifer forests under O3 exposure was continually observed in several
31                   regions. Ozone damage was believed to be an important causal factor in the dramatic
32                   decline of sacred fir in the valley of Mexico, as well as cembran pine in southern France
33                   and Carpathian Mountains, although several factors, such as drought, insect outbreak and
34                   forest management, may also contribute to or even be the dominant factors causing the
      Draft - Do Not Cite or Quote                      2-42                                September 2011

-------
 1                  mortality of the conifer trees. Results from the Aspen FACE site indicated that O3 could
 2                  alter community composition of broadleaf forests as well. At the Aspen FACE site, O3
 3                  reduced growth and increased mortality of a sensitive aspen clone, while the O3 tolerant
 4                  clone emerged as the dominant clone in the pure aspen community. In the mixed aspen-
 5                  birch and aspen-maple communities, O3 reduced the competitive capacity of aspen
 6                  compared to birch and maple (Section 9.4.7.1).

 7                  The tendency for O3-exposure to shift the biomass of grass4egume mixtures in favor of
 8                  grass species was reported in the 2006 O3 AQCD and has been generally confirmed by
 9                  recent studies. However, in a high elevation mature/species-rich grass4egume pasture, O3
10                  fumigation showed no significant impact on community composition (Section 9.4.7.2).

11                  Ozone exposure not only altered community composition of plant species, but also
12                  microorganisms. The shift in community composition of bacteria and fungi has been
13                  observed in both natural  and agricultural ecosystems, although no general patterns could
14                  be identified (Section 9.4.7.3).
15                  The evidence is sufficient to conclude that there is likely a causal relationship
16                  between O3 exposure and the alteration of community composition.
            2.7.6   Policy Relevant Considerations


                     2.7.6.1    Air Quality Indices

17                   Exposure indices are metrics that quantify exposure as it relates to measured plant
18                   damage (e.g., reduced growth). They are summary measures of monitored ambient O3
19                   concentrations over time intended to provide a consistent metric for reviewing and
20                   comparing exposure-response  effects obtained from various studies. No new information
21                   is available since 2006 that alters the basic conclusions put forth in the 2006 and 1996 O3
22                   AQCDs. These AQCDs focused on the research used to develop various exposure indices
23                   to help quantify effects on growth and yield in crops, perennials, and trees (primarily
24                   seedlings). The performance of indices was compared through regression analyses of
25                   earlier studies designed to support the estimation of predictive O3 exposure-response
26                   models for growth and/or yield of crops and tree (seedling) species.
27                   Another approach for improving risk assessment of vegetation response to ambient O3 is
28                   based on determining the O3 concentration from the atmosphere that enters the leaf (i.e.,
29                   flux or deposition). Interest has been increasing in recent years, particularly in Europe, in
30                   using mathematically tractable flux models for O3 assessments at  the regional, national,
31                   and European scale. While some efforts have been made in the U.S. to calculate O3 flux

      Draft - Do Not Cite or Quote                       2-43                               September 2011

-------
 1                   into leaves and canopies, little information has been published relating these fluxes to
 2                   effects on vegetation. There is also concern that not all O3 stomatal uptake results in a
 3                   yield reduction, which depends to some degree on the amount of internal detoxification
 4                   occurring with each particular species. Species having high detoxification capacity may
 5                   show little relationship between O3 stomatal uptake and plant response. The lack of data
 6                   in the U.S. and the lack of understanding of detoxification processes have made this
 7                   technique less viable for vulnerability and risk assessments in the U.S.

 8                   The main conclusions from the 1996 and 2006 O3 AQCDs regarding indices based on
 9                   ambient exposure remain valid. These key conclusions can be restated as follows:

10                       •  O3 effects in plants are cumulative;
11                       •  higher O3 concentrations appear to be more important than lower
12                         concentrations in eliciting a response;
13                       •  plant sensitivity to O3 varies with time of day and plant development stage;
14                         and
15                       •  exposure indices that cumulate hourly O3 concentrations and preferentially
16                         weight the higher concentrations have better statistical fits to growth/yield
17                         response data than do the mean and peak indices.

18                   Various weighting functions have been used, including threshold-weighted (e.g.,
19                   SUM06) and continuous sigmoid-weighted (e.g., W126) functions. Based on statistical
20                   goodness-of-fit tests, these cumulative, concentration-weighted indices could not be
21                   differentiated from one another using data from previous exposure studies. Additional
22                   statistical forms for O3  exposure indices are summarized in Section 9.5 of this ISA. The
23                   majority of studies published since the 2006 O3 AQCD do not change earlier conclusions,
24                   including the importance of peak concentrations, and the duration and occurrence of O3
25                   exposures in altering plant growth and yield.

26                   Given the current state  of knowledge and the best available data, exposure indices  that
27                   cumulate and differentially weight the higher hourly average concentrations and also
28                   include the mid-level values  continue to offer the most defensible approach for use in
29                   developing response functions and comparing studies, as well as for defining future
3 0                   indice s for vegetation protection.
                     2.7.6.2    Exposure-Response

31                   None of the information on effects of O3 on vegetation published since the 2006 O3
32                   AQCD has modified the assessment of quantitative exposure-response relationships that
33                   was presented in that document (U.S. EPA. 2006b). This assessment updates the 2006

      Draft - Do Not Cite or Quote                      2-44                                 September 2011

-------
 1                  exposure-response models by computing them using the W126 metric, cumulated over
 2                  90 days. Almost all of the experimental research on the effects of O3 on growth or yield
 3                  of plants published since 2006 used only two levels of exposure. In addition, hourly O3
 4                  concentration data that would allow calculations of exposure using the W126 metric are
 5                  generally unavailable. However, two long-term experiments, one with a crop  species
 6                  (soybean), one with a tree species (aspen), have produced data that are used in Section
 7                  9.6 to validate the exposure-response models presented in the 2006 O3 AQCD, and  the
 8                  methodology used to derive them. EPA compared predictions from the models presented
 9                  in the 2006 O3 AQCD, updated to use the 90 day 12hr W126 metric, with more recent
10                  observations for yield of soybean and biomass growth of trembling aspen. The models
11                  were parameterized using data from the NCLAN and NHEERL-WED projects, which
12                  were conducted in OTCs. The more recent observations were from experiments using
13                  FACE technology, which is intended to provide conditions closer to natural environments
14                  than OTC. Observations from these new experiments were exceptionally close to
15                  predictions from the models. The accuracy of model predictions for two widely different
16                  plant species, grown under very different conditions, provides support for the validity of
17                  the models for crops and trees developed using the same methodology and data for  other
18                  species. However, variability observed among species in the NCLAN and NHEERL-
19                  WED projects indicates that the range of sensitivity between and among species is likely
20                  quite wide.

21                  Results from several meta-analyses have provided approximate values for responses of
22                  yield of soybean, wheat, rice and other crops under broad categories of exposure, relative
23                  to charcoal-filtered air. Additional reports have summarized yield data for six crop
24                  species under various broad comparative exposure categories, and reviewed 263 studies
25                  that reported effects on tree biomass. However, these analyses have proved difficult to
26                  compare with exposure-response models, especially given that exposure was not
27                  expressed on the same W126 scale.
          2.8    The Role of Tropospheric Ozone in Climate Change and  UV-B
                 Effects

28                  Atmospheric O3 plays an important role in the Earth's energy budget by interacting with
29                  incoming solar radiation and outgoing infrared radiation. Tropospheric O3 makes up only
30                  a small portion of the total column of O3, but it has important incremental effects on the
31                  overall radiation budget. Chapter 10 assesses the specific role of tropospheric O3 in the
32                  earth's radiation budget and how perturbations in tropospheric O3 might affect (1) climate
33                  through its role as a greenhouse gas, and (2) health, ecology and welfare through its role
34                  in shielding the earth's surface from solar ultraviolet (UV) radiation.

      Draft - Do Not Cite or Quote                      2-45                               September 2011

-------
            2.8.1   Tropospheric Ozone as a Greenhouse Gas

 1                  Ozone is an important greenhouse gas, and increases in its abundance in the troposphere
 2                  may contribute to climate change according to the 2007 climate assessment by the
 3                  Intergovernmental Panel on Climate Change (IPCC). Models calculate that the global
 4                  burden of tropospheric O3 has doubled since the preindustrial era, while observations
 5                  indicate that in some regions O3 may have increased by factors as great as 4 or 5. These
 6                  increases are tied to the rise in emissions of O3 precursors from human activity, mainly
 7                  fossil fuel consumption and agricultural processes.

 8                  Units shown are those typical for each quantity illustrated. Feedbacks from both the
 9                  climate response and climate impacts can, in turn, affect the abundance of tropospheric
10                  O3 and O3  precursors through multiple feedback mechanisms. Climate impacts are
11                  deemphasized in the figure since these downstream effects are extremely complex and
12                  outside the  scope of this assessment.

13                  Figure 2-5 shows the main steps involved in the influence of tropospheric O3 on climate.
14                  Emissions of O3 precursors including CO, VOCs, CH4, and NOX lead to production of
15                  tropospheric O3. A change in the abundance of tropospheric O3 perturbs the radiative
16                  balance of the atmosphere, an effect quantified by the radiative forcing (RF) metric. The
17                  earth-atmosphere-ocean system responds to the forcing with a climate response, typically
18                  expressed as a change in surface temperature. Finally, the climate response causes
19                  downstream climate-related health and ecosystem impacts,  such as redistribution of
20                  diseases or ecosystem characteristics due to temperature changes. Feedbacks from both
21                  the climate response and downstream impacts can, in turn, affect the abundance of
22                  tropospheric O3 and O3  precursors through multiple feedback mechanisms as indicated in
23                  Figure 2-5. Direct feedbacks are discussed in Section 10.2.3.4 while downstream climate
24                  impacts and their feedbacks are extremely complex and outside the scope of this
25                  assessment.

26                  The impact of the tropospheric O3 change since preindustrial times on climate has been
27                  estimated to be about 25-40% of anthropogenic CO2 impact and about 75% of
28                  anthropogenic CFI4 impact according to the IPCC, ranking it third in importance among
29                  the greenhouse gases. There are large uncertainties in the RF estimate attributed to
30                  tropospheric O3, however, making the impact of tropospheric O3 on climate more
31                  uncertain than the impact of the long-lived greenhouse gases. Despite these uncertainties,
32                  the evidence supports a causal relationship between changes  in tropospheric O3
33                  concentrations and radiative forcing.

34                  RF does not take into account the climate feedbacks that could amplify or dampen the
35                  actual surface temperature response. Quantifying the change in surface temperature
      Draft - Do Not Cite or Quote                       2-46                                September 2011

-------
 1
 2
 3
 4
 5
 6
              requires a complex climate simulation in which all important feedbacks and interactions
              are accounted for. As these processes are not well understood or easily modeled, the
              surface temperature response to a given RF is highly uncertain and can vary greatly
              among models and from region to region within the same model. In light of these
              uncertainties, the evidence supports a likely to be a causal relationship between
              changes in tropospheric O3 concentrations and climate change.
                                         Precursor Emissions of
                                         CO, VOCs, CHr NOX
                                               (Tg/y)

Tropospheric O,
Abundance
(Tg)
^
                                           Radiative Forcing
                                           Due to O, Change
                                               (W/nf)
Climate Response
("C)

 9
10
11
12
13
  Units shown are those typical for each quantity illustrated. Feedbacks from both the climate response and climate impacts can, in
turn, affect the abundance of tropospheric O3 and O3 precursors through multiple feedback mechanisms. Climate impacts are
deemphasized in the figure since these downstream effects are extremely complex and outside the scope of this assessment.

Figure 2-5     Schematic illustrating the effects of tropospheric O3 on climate
                including the relationship between precursor emissions,
                tropospheric O3 abundance, radiative forcing, climate  response,
                and climate impacts. Tropospheric Ozone and UV-B related effects

               UV radiation emitted from the Sun contains  sufficient energy when it reaches the Earth to
               break (photolyze) chemical bonds in molecules, thereby leading to damaging effects on
               living organisms and materials. Atmospheric O3 plays a crucial role in reducing exposure
               to solar UV radiation at the Earth's surface. Ozone in the stratosphere is responsible for
               the majority of this shielding effect, as approximately 90% of total atmospheric O3 is
               located there over mid-latitudes. Ozone in the troposphere provides supplemental
               shielding of radiation in the wavelength band from 280-315 nm, referred to as UV-B
      Draft - Do Not Cite or Quote
                                             2-47
September 2011

-------
 1                   radiation. UV-B radiation has important effects on human health and ecosystems, and is
 2                   associated with materials damage.

 3                   Adverse human health effects associated with solar UV-B radiation exposure include
 4                   erythema,  skin cancer, ocular damage, and immune system suppression. A potential
 5                   human health benefit of increased UV-B exposure involves the UV-induced production
 6                   of vitamin D which may help reduce the risk of metabolic bone disease, type I diabetes,
 7                   mellitus, and rheumatoid arthritis, and may provide beneficial immunomodulatory effects
 8                   on multiple sclerosis, insulin-dependent diabetes mellitus, and rheumatoid arthritis.

 9                   Adverse ecosystem and materials damage effects associated with solar UV-B radiation
10                   exposure include terrestrial and aquatic ecosystem impacts, alteration of biogeochemical
11                   cycles, and degradation of man-made materials. Terrestrial ecosystem effects from
12                   increased UV-B radiation include reduced plant productivity and plant cover, changes in
13                   biodiversity, susceptibility to infection, and increases in natural UV protective responses.
14                   In general, however, these effects are small for moderate  UV-B increases  at mid-
15                   latitudes. Aquatic ecosystem effects from increased UV-B radiation include sensitivity in
16                   growth, immune response, and behavioral patterns of aquatic organisms and the potential
17                   for increased catalysis  and mobility of trace metals. Biogeochemical cycles, particularly
18                   the carbon cycle, can also be influenced by increased UV-B radiation with effects ranging
19                   from UV-induced increases in CO2 uptake through soil respiration to UV-induced release
20                   of CO2 through photodegradation of above-ground plant litter. Changes in solar UV
21                   radiation may also have effects on carbon cycling and CO2 uptake in the oceans as well
22                   as release of dissolved organic matter from sediment and  algae. Finally, materials damage
23                   from increased UV-B radiation includes UV-induced photodegradation of wood and
24                   plastics.

25                   There is a  lack of published studies that critically examine the incremental health or
26                   welfare effects (adverse or beneficial)  attributable specifically to changes  in UV-B
27                   exposure resulting from perturbations in tropospheric O3 concentrations. The effects are
28                   expected to be small and they cannot yet be critically assessed within reasonable
29                   uncertainty. Overall, the evidence is inadequate to determine if a causal relationship
30                   exists between tropospheric O3 and UV-B related health and welfare  effects.
      Draft - Do Not Cite or Quote                       2-48                                September 2011

-------
           2.9    Summary of Causal  Determinations for Health Effects and
                   Welfare Effects
1
2
                       This chapter has provided an overview of the underlying evidence used in making the

                       causal determinations for the health and welfare effects of O3. This review builds upon

                       the conclusions of the previous AQCDs for O3.
 4
 5
 6
 7
 8
 9
10
                      The evaluation of the epidemiologic, toxicological, and controlled human exposure
                      studies published since the completion of the 2006 O3 AQCD have provided additional
                      evidence for O3-related health outcomes. Table 2-3 provides an overview of the causal
                      determinations for all of the health outcomes evaluated. Causal determinations for O3 and
                      welfare effects are included in Table 2-4, while causal determinations for climate change
                      and UV-B effects are in Table 2-5. Detailed discussions of the scientific evidence and
                      rationale for these causal determinations are provided in subsequent chapters of this ISA.
       Table 2-3        Summary of ozone causal determinations by exposure duration
                          and health outcome
        Health Outcome
                                         Conclusions from 2006 O3 AQCD
  Conclusions from
  2011 1st Draft ISA
       Short-Term Exposure to O3
       Respiratory effects      The overall evidence supports a causal relationship between acute ambient 03        Causal Relationship
      	exposures and increased respiratory morbidity outcomes.	
      Cardiovascular effects
                          The limited evidence is highly suggestive that 03 directly and/or indirectly contributes to
                          cardiovascular-related morbidity, but much remains to be done to more fully
                          substantiate the association.
Suggestive of a Causal
Relationship
       Central nervous system  Toxicological studies report that acute exposures to 03 are associated with alterations in
       effects              neurotransmitters, motor activity, short and long term memory, sleep patterns, and
      _ histological signs of neurodegeneration. _
       Mortality
                         The evidence is highly suggestive that 03 directly or indirectly contributes to non-
                         accidental and cardiopulmonary-related mortality.	
Suggestive of a Causal
Relationship

Likely to be a Causal
Relationship _
       Long-term Exposure to Oz
       Respiratory effects      The current evidence is suggestive but inconclusive for respiratory health effects from
      	long-term 03 exposure.	
                                                                                           Likely to be a Causal
                                                                                           Relationship	
       Cardiovascular Effects   No studies from previous review
                                                                                           Suggestive of a Causal
                                                                                           Relationship	
       Reproductive and
       developmental effects
                         Limited evidence for a relationship between air pollution and birth-related health
                         outcomes, including mortality, premature births, low birth weights, and birth defects, with
                         little evidence being found for 03 effects.	
Suggestive of a Causal
Relationship
       Central nervous system  Evidence regarding chronic exposure and neurobehavioral effects was not available.
       effects
                                                                                           Suggestive of a Causal
                                                                                           Relationship	
       Cancer
                         Little evidence for a relationship between chronic 03 exposure and increased risk of
                         lung cancer.	
Inadequate to infer a
Causal Relationship
       Mortality
                         There is little evidence to suggest a causal relationship between chronic 03 exposure
                         and increased risk for mortality in humans.
Suggestive of a Causal
Relationship
       Draft - Do Not Cite or Quote
                                                          2-49
        September 2011

-------
Table 2-4        Summary of ozone causal determination for welfare effects
     Vegetation and
   Ecosystem Effects
                    Conclusions from 2006 O3 AQCD
   Conclusions from
  2011 2nd Draft ISA
Visible Foliar Injury Effects on
Vegetation
     Data published since the 1996 ft AQCD strengthen previous conclusions that
     there is strong evidence that current ambient 03 concentrations cause impaired
     aesthetic quality of many native plants and trees by increasing foliar injury.
Causal Relationship
Reduced Vegetation Growth
     Data published since the 1996 ft AQCD strengthen previous conclusions that
     there is strong evidence that current ambient 03 concentrations cause
     decreased growth and biomass accumulation in annual, perennial and woody
     plants, including agronomic crops, annuals, shrubs, grasses, and trees.	
Causal Relationship
Reduced Productivity in
Terrestrial Ecosystems
     There is evidence that 03 is an important stressor of ecosystems and that the
     effects of 03 on individual plants and processes are scaled up through the
     ecosystem, affecting net primary productivity.	
Causal Relationship
Reduced Carbon (C)
Sequestration in Terrestrial
Ecosystems	
     Limited studies from previous review
Likely to be a Causal
Relationship
Reduced Yield and Quality of
Agricultural Crops
     Data published since the 1996 ft AQCD strengthen previous conclusions that
     there is strong evidence that current ambient 03 concentrations cause
     decreased yield and/or nutritive quality in a large number of agronomic and
     forage crops.	
Causal Relationship
Alteration of Terrestrial
Ecosystem Water Cycling
     Ecosystem water quantity may be affected by 03 exposure at the landscape
     level.
Likely to be a Causal
Relationship	
Alteration of Below-ground
Biogeochemical Cycles
Alteration of Terrestrial
Community Composition
     Ozone-sensitive species have well known responses to 03 exposure, including   Causal Relationship
     altered C allocation to below-ground tissues, and altered rates of leaf and root
     production, turnover, and decomposition. These shifts can affect overall C and N
     loss from the ecosystem in terms of respired C, and leached aqueous dissolved
     organic and inorganic C and N.	
     Ozone may be affecting above- and below -ground community composition
     through impacts on both growth and reproduction. Significant changes in plant
     community composition resulting directly from 03 exposure have been
     demonstrated.
Likely to be a Causal
Relationship
Table 2-5        Summary of ozone causal determination for climate change and
                     UV-B effects
       Effects
                  Conclusions from 2006 O3 AQCD
    Conclusions from
    2011  1st Draft ISA
Radiative Forcing
Climate Change
Climate forcing by 03 at the regional scale may be its most important impact on climate.   Causal Relationship
While more certain estimates of the overall importance of global-scale forcing due to
tropospheric 03 await further advances in monitoring and chemical transport modeling,
the overall body of scientific evidence suggests that high concentrations of 03 on the
regional scale could have a discernable influence on climate, leading to surface
temperature and hydrological cycle changes.	
  Likely to be a Causal
  Relationship
UV-B Related Health and   UV-B has not been studied in sufficient detail to allow for a credible health benefits
Welfare Effects            assessment. In conclusion, the effect of changes in surface-level 03 concentrations on
                        UV-induced health outcomes cannot yet be critically assessed within reasonable
                        uncertainty.
                                                                           Inadequate to Determine if
                                                                           a Causal Relationship
                                                                           Exists
Draft - Do Not Cite or Quote
                                  2-50
        September 2011

-------
    2.10  References

CAA. (Air quality criteria and control techniques, Section 108 of the Clean Air Act). 42. § 7408, (1990a).
U.S. EPA.  (U.S. Environmental Protection Agency). (1978a). Air quality criteria for ozone and other
        photochemical oxidants. (EPA/600/8-78/004). Washington, DC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1984). Air quality criteria for ozone and other
        photochemical oxidants, Vol. 3. (EPA/600/8-84/020A). Research Triangle Park, NC.
        http://nepis.epa. aov/Exe/ZvPURL.cgi?Dockev=2000AVEV.txt.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996a). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/P-93/004AF). Research Triangle Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996b). Air quality criteria for ozone and related
        photochemical oxidants, Vol. II of III. (EPA/600/P-93/004BF). Research Triangle Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA.  (U.S. Environmental Protection Agency). (2008f). Notice of workshop and call for information on
        integrated science assessment for ozone. Fed Reg 73: 56581-56583.
U.S. EPA.  (U.S. Environmental Protection Agency). (2009c). Integrated review plan for the ozone National
        Ambient Air Quality Standards review (external review draft). (EPA 452/D-09-001). Washington, DC.
        http://www.epa.gov/ttnnaaqs/standards/ozone/data/externalreviewdraftO3IRP093009.pdf.
U.S. EPA.  (U.S. Environmental Protection Agency). (2010e). Our nation's air: Status and trends through 2008.
        (EPA-454/R-09-002). Research Triangle Park, NC.
        http://www.epa.gov/airtrends/2010/report/fullreport.pdf.
Zhang. L: Jacob. DJ: Downey. NV: Wood. DA: Blewitt. D: Carouge. CC: Van donkelaar. A: Jones. DBA: Murray.
        LT: Wang. Y (In Press) Improved estimate of the policy-relevant background ozone in the United States
        using the GEOS-Chem global model with 1/2° * 2/3°  horizontal resolution over North America. Atmos
        Environ. http://dx.doi.Org/10.1016/i.atmosenv.2011.07.054.
Draft - Do Not Cite or Quote                        2-51                                  September 2011

-------
      3  ATMOSPHERIC  CHEMISTRY  AND  AMBIENT
          CONCENTRATIONS
          3.1    Introduction

 1                  In the stratosphere, O3 serves the beneficial role of blocking the Sun's harmful ultraviolet
 2                  radiation and preventing the majority of this radiation from reaching the Earth's surface.
 3                  In the troposphere, however, O3 and other photochemical oxidants are air pollutants with
 4                  potentially harmful effects on living organisms. This chapter discusses the atmospheric
 5                  chemistry associated with tropospheric O3 and other related photochemical oxidants and
 6                  provides a detailed description of their surface-level concentrations. The focus of this
 7                  chapter is on O3 since it is the NAAQS indicator for all photochemical oxidants. To the
 8                  extent possible, other photochemical oxidants are discussed, but limited information is
 9                  currently available. Although O3 is involved in reactions in indoor air, the focus in this
10                  chapter will be on chemistry occurring in outdoor, ambient air.

11                  The material in this chapter is organized as follows.  Section 3.2 outlines the physical and
12                  chemical processes involved in O3 formation and removal. Section 3.3 describes the
13                  latest methods used to model global O3  concentrations, and Section 3.4 describes the
14                  application of these methods for estimating background concentrations of O3 that are
15                  useful for risk and policy assessments informing decisions about the NAAQS. Section 3.1
16                  includes a comprehensive description of available O3 monitoring techniques and
17                  monitoring networks, while  Section 3.6 presents information on the spatial and temporal
18                  variability of O3 concentrations across the U.S. and their associations with other
19                  pollutants using available monitoring data. Section 3.7 summarizes the main conclusions
20                  of Chapter 3. Section 3.8 provides supplemental material for atmospheric  model
21                  predictions of background O3 concentrations described in Section 3.4; Section 3.9
22                  contains supplemental material for model predictions of background O3 concentrations
23                  using a more recent version of the atmospheric model described in Section 3.4; and
24                  Section 3.10 contains supplemental figures of observed ambient O3 concentrations.
          3.2    Physical and Chemical Processes

25                  O3 in the troposphere is a secondary pollutant formed by photochemical reactions of
26                  precursor gases and is not directly emitted from specific sources. Ozone and other
27                  oxidants, such as PAN and H2O2 form in polluted areas by atmospheric reactions
      Draft - Do Not Cite or Quote                      3-1                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7

 8
 9
10
11
12
13
involving two main classes of precursor pollutants: VOCs and NOx.1 Carbon monoxide
(CO) is also important for O3 formation in polluted areas and in the remote troposphere.
The formation of O3, other oxidants and oxidation products from these precursors is a
complex, nonlinear function  of many factors including (1) the intensity and spectral
distribution of sunlight; (2) atmospheric mixing; (3) concentrations of precursors in the
ambient air and the rates of chemical reactions of these precursors; and (4) processing on
cloud and aerosol particles.

Ozone is present not only in polluted urban atmospheres, but throughout the troposphere,
even in remote areas of the globe. The same basic processes involving sunlight-driven
reactions of NOX, VOCs and CO contribute to O3 formation throughout the troposphere.
These processes also lead to the formation of other photochemical products, such as
PAN, HNO3, and H2SO4, and to other compounds, such as HCHO and other carbonyl
compounds, and to secondary components of particulate matter.
      Figure 3-1     Schematic overview of photochemical processes influencing
                      stratospheric and tropospheric ozone.
       1 The term VOCs refers to all organic gas-phase compounds in the atmosphere, both biogenic and anthropogenic in origin. This
      definition excludes CO and CO2. NOX, also referred to as nitrogen oxides, is equal to the sum of NO and NO2.
      Draft - Do Not Cite or Quote
                               3-2
September 2011

-------
 1                   The processes responsible for producing summertime O3 episodes are fairly well
 2                   understood, and were covered in detail in the previous O3 AQCD. This section focuses
 3                   on topics that form the basis for discussions in later chapters and for which there is
 4                   substantial new information since the previous O3 AQCD. A schematic overview of the
 5                   major photochemical cycles influencing O3 in the troposphere and the stratosphere is
 6                   given in Figure 3-1.

 7                   Major episodes of high O3 concentrations in the eastern U.S. and in Europe are
 8                   associated with slow moving high pressure systems. High pressure systems during the
 9                   warmer seasons are associated with the sinking of air, resulting in warm, generally
10                   cloudless skies, with light winds. The sinking of air results in the development of stable
11                   conditions near the surface which inhibit or reduce the vertical mixing of O3 precursors.
12                   The combination of inhibited vertical mixing and light winds minimizes the dispersal of
13                   pollutants emitted in urban areas, allowing their concentrations to build up. Photochemi-
14                   cal activity involving these precursors is enhanced because of higher temperatures and
15                   the availability of sunlight during the warmer seasons. In the eastern U.S., concentrations
16                   of O3 and other secondary pollutants are determined by meteorological and chemical
17                   processes extending typically over areas of several hundred thousand square kilometers
18                   (Civerolo et al.. 2003; Rao et al.. 2003). Ozone  episodes are thus best regarded as
19                   regional in nature. The conditions conducive to formation of high O3 can persist for
20                   several days. These conditions have  been described in greater detail in the 1996 and 2006
21                   O 3 AQCDs (U.S. EPA. 2006b. 1996a). The transport of pollutants downwind of maj or
22                   urban centers is characterized by the development of urban plumes. Mountain barriers
23                   limit mixing (as in Los Angeles and Mexico City) and result in a higher frequency and
24                   duration of days with high O3 concentrations. However, orographic lifting over the San
25                   Gabriel Mountains results in O3 transport from Los Angeles to areas hundreds of
26                   kilometers downwind (e.g., in Colorado and Utah) (Langford et al.. 2009). Ozone
27                   concentrations in southern urban areas (such as Houston, TX and Atlanta, GA) tend to
28                   decrease with increasing wind speed. In northern U.S. cities (such as Chicago, IL;
29                   New York, NY; Boston, MA; and Portland, ME), the average O3 concentrations over the
30                   metropolitan areas increase with wind speed, indicating that transport of O3 and its
31                   precursors from upwind areas is important (Schichtel and Husar. 2001; Husar and
32                   Renard. 1998).

33                   Aircraft observations indicate that there can be substantial differences in mixing ratios of
34                   key species between the surface and the overlying atmosphere (Berkowitz and Shaw.
35                   1997; Fehsenfeld et al.. 1996). In particular, mixing ratios of O3 can be higher in the
36                   lower free troposphere (aloft) than in the planetary boundary layer (PEL) during multiday
37                   O3 episodes (Taubman et al.. 2006; Taubman et al.. 2004). Convective processes and
3 8                   turbulence transport O3 and  other pollutants both upward and downward throughout the
      Draft - Do Not Cite or Quote                       3-3                                 September 2011

-------
 1                   planetary boundary layer and the free troposphere. During the day, convection driven by
 2                   heating of the earth's surface results in a deeper PEL with vertically well mixed O3 and
 3                   precursors. As solar heating of the surface decreases going into night, the daytime
 4                   boundary layer collapses leaving behind O3 and its precursors in a residual layer above a
 5                   shallow nighttime boundary layer. Pollutants in the residual layer have now become
 6                   essentially part of the free troposphere, as described in Section AX2.3.2 of the 2006 O3
 7                   AQCD. Winds in the free troposphere tend to be stronger than those closer to the surface
 8                   and so are capable of transporting pollutants over long distances. Thus, O3 and its
 9                   precursors can be transported vertically by convection into the upper part of the mixed
10                   layer on one day, then transported overnight as a layer of elevated mixing ratios, and then
11                   entrained into a growing convective boundary layer downwind and brought back down to
12                   the surface.

13                   High O3 concentrations showing large diurnal variations at the surface in southern New
14                   England were associated with the presence of such layers (Berkowitz et al., 1998). Winds
15                   several hundred meters above the ground can bring pollutants from the west, even though
16                   surface winds are from the southwest during periods of high O3 in the eastern U.S.
17                   (Blumenthal et al.. 1997). These considerations suggest that in many areas of the U.S., O3
18                   and its precursors can be transported over hundreds if not thousands of kilometers.

19                   Nocturnal low level jets (LLJs) are an efficient means for transporting pollutants over
20                   hundreds of kilometers that have been entrained into the residual boundary layer. LLJs
21                   are most prevalent in the central U.S. extending northward from eastern Texas, and along
22                   the Atlantic states extending southwest to northeast. LLJs have also been observed off the
23                   coast of California. Turbulence induced by wind shear associated with LLJs brings
24                   pollutants to the surface and results in secondary O3 maxima during the night and early
25                   morning in many locations (Corsmeier et al.. 1997). Comparison of observations at low-
26                   elevation surface sites with those at nearby high-elevation sites at night can be used to
27                   discern the effects of LLJs. For example, Fischer et al. (2004) found occasions when O3
28                   at the  base of Mt. Washington during the night was much higher than typically observed,
29                   and closer to those observed at the summit of Mt. Washington. They suggested that
30                   mechanically driven turbulence due to wind shear caused O3 from aloft to penetrate the
31                   stable nocturnal inversion thus causing O3 to increase near the base of Mt. Washington.
32                   The high wind speeds causing this mechanically driven turbulence could have resulted
33                   from the development of an  LLJ. Stratospheric intrusions and intercontinental transport
34                   of O3  are also important and are covered in  Section 3.4 in relation to policy relevant
3 5                   background concentrations.
      Draft - Do Not Cite or Quote                        3-4                                 September 2011

-------
            3.2.1    Sources of Precursors Involved in Ozone Formation

 1                   Emissions of O3 precursor compounds (NOX, VOCs, and CO) can be divided into natural
 2                   and anthropogenic source categories. Natural sources can be further divided into biogenic
 3                   from vegetation, microbes, and animals, and abiotic from biomass burning, lightning, and
 4                   geogenic sources. However, the distinction between natural and anthropogenic sources is
 5                   often difficult to make in practice, as human activities directly or indirectly affect
 6                   emissions from what would have been considered natural sources during the preindustrial
 7                   era. Thus, emissions from plants and animals used in agriculture have been referred to as
 8                   anthropogenic or biogenic in different applications. Wildfire emissions can be considered
 9                   natural, except that forest management practices can lead to buildup of fuels on the forest
10                   floor, thereby altering the frequency and severity of forest fires.

11                   Estimates of emissions for NOX, VOCs, and CO (U.S. EPA. 2008a) are shown in
12                   Figure 3-2 to provide a general indication of the relative importance of the different
13                   sources in the U.S. as a whole. The magnitudes of the sources are strongly location and
14                   time dependent and so should not be  used to apportion sources of exposure. Shown in
15                   Figure 3-2 are Tier 1 categories. The miscellaneous category can be quite large compared
16                   to total emissions, especially for CO  and VOCs. The miscellaneous category includes
17                   agriculture and forestry,  wildfires, prescribed burns, and a much more modest
18                   contribution from structural fires.

19                   Anthropogenic NOX emissions are associated with combustion processes. Most
20                   emissions are in the form of NO, which is formed at high combustion temperatures from
21                   atmospheric nitrogen (N2) and oxygen (O2) and from fuel nitrogen  (N). According to the
22                   2005 National Emissions Inventory (U.S. EPA, 2008a). the largest sources of NOX are
23                   on- and off-road mobile  sources and  electric power generation plants. Emissions of NOX
24                   therefore are highest in areas having a high density of power plants  and in urban regions
25                   having high traffic density. Dallman  and Harley (2010) compared NOX emissions
26                   estimates  from the National Emissions Inventory, mobile sector data (U.S. EPA. 2008a)
27                   with an alternative method based on fuel consumption and found reasonable agreement in
28                   total U.S.  anthropogenic emissions between the two techniques  (to within about 5%).
29                   However, emissions from on-road diesel engines in the fuel based inventory constituted
30                   46% of total mobile source NOX compared to 35% in the EPA inventory. As a result,
31                   emissions from on-road diesel engines in the fuel based approach are even larger than
32                   electric power generation as estimated in the 2005 NEI, and on-road diesel engines might
33                   represent the largest single NOX source category. Differences between the two techniques
34                   are largely accounted for by differences in emissions from on-road gasoline engines.
35                   Uncertainties in the fuel  consumption inventory ranged from 3% for on-road gasoline
36                   engines to 20% for marine sources, and in the EPA inventory uncertainties ranged from
      Draft - Do Not Cite or Quote                       3-5                                 September 2011

-------
1                       16% for locomotives to 30% for off-road diesel engines. It should be noted that the on-
2                       road diesel engine emissions estimate by Dallman and Harley (2010) is still within the
3                       uncertainty of the EPA estimate (22%).
                                                          Nimgra Oxides (KQ, |
                                                          ToalEmcam-IHMr
                                          HIGHWAY VEHICLES
                                              OFF-HIGHWAY
                                       FUEL COM3 ELEC. UTIL •
                                       FUEL COMB INDUSTRIAL
                                          FUEL COMB. OTHER ^^B06
                                  OTHER INDUSTRIAL PROCESSES ^BO.44
                               PETROLEUM S RELATED INDUSTRIES CZJ 0.32
                                            MISCELLANEOUS El 026
                                  WASTE DISPOSAL & RECYCLING 0 0.13
                                         METALS PROCESSING 1006
                                CHEMICAL & ALLIED PRODUCT MFG 10.05
                                       STORAGE S TRANSPORT I0.01S
                                         SOLVENT UTILIZATION 10,004
                                                       01234567
                                                                               Emissions (MlflloiH Tof»rr«f)
                                                      VoUik OiganeCompmimlslVOO
                                                         fonlEnis.Mis-l6.mT
                                          HIGHWAY VEHICLES ^^^^E        	
                                              OFF-HIGHWAY ^^^^H
                                       PUEL COMB ELEC. UTIL. 10 04
                                       FUEL COMB INDUSTRIAL B0.12
                                          FUEL COMB OTHER I^^BO 53
                                  OTHER INDUSTRIAL PROCESSES I^^O 41
                               PETROLEUM & RELATED INDUSTRIES '     j 0 51
                                            MISCELLANEOUS •
                                   WASTE DISPOSAL & RECYCLING  	loss
                                         METALS PROCESSING  JO 04
                                CHEMICAL & ALLIED PRODUCT MFG  13021
                                       STORAGE & TRANSPORT  ^=^=
                                         SOLVENT UTILIZATION  J^^^^
                                                                                 3        4
                                                                               EnrtMtona (Million* Ton »!Yr JO
                                                         Carixn Monoxide (CO)
                                                          teal Emlsskni at MT
                                          HIGHWAY VEHICLES ^^^^^^^^^=
                                              OFF-HIGHWAY
                                        FUEL COMB ELEC. UTIL. D0.58
                                       FUEL COMB INDUSTRIAL in 1.04
                                          FUEL COMB. OTHER ^3,02
                                  OTHER INDUSTRIAL PROCESSES iO 48
                                PETROLEUM 4 RELATED INDUSTRIES 1032
                                            MISCELLANEOUS
                                   WASTE DISPOSAL & RECYCLING B1 41
                                         METALS PROCESSING DO 75
                                 CHEMICAL & ALLIED PRODUCT MFG 10 19
                                       STORAGE 8 TRANSPORT I O.I
                                         SOLVENT UTILIZATION O.O02
                                                                              30  3S  40  45
                                                                               Eml»lo<» (Wlllont
        Source: U.S. EPA (2008a)
        NOX (top), VOCs (middle), and CO (bottom) in the U.S. in million metric tons (MT) per year.

      Figure 3-2      Estimated anthropogenic emissions of ozone precursors for 2005.
4                       Major natural sources of NOX in the U.S. include lightning, soils, and wildfires.
5                       Uncertainties in natural NOX emissions are much larger than for anthropogenic NOX

      Draft - Do Not Cite or Quote                            3-6                                       September 2011

-------
 1                   emissions. Fang et al. (2010) estimated lightning generated NOX of ~0.6 MT for July
 2                   2004. This value is -40% of the anthropogenic emissions for the same period, but Fang et
 3                   al. estimated that -98% is formed in the free troposphere and so contributions to the
 4                   surface NOX burden are low because most of this NOX is oxidized to nitrate containing
 5                   species during downward transport into the planetary boundary layer. The remaining 2%
 6                   is formed within the planetary boundary layer. Both nitrifying and denitrifying organisms
 7                   in the soil can produce NOX, mainly in the form of NO. Emission rates depend mainly on
 8                   fertilization amount and soil temperature and moisture. Nationwide, about 60% of the
 9                   total NOX emitted by soils is estimated to occur in the central corn belt of the U.S. Spatial
10                   and temporal variability in soil NOX emissions leads to considerable uncertainty in
11                   emissions estimates. However, these emissions are relatively low, only -0.97 MT/year, or
12                   about 6% of anthropogenic NOX emissions. However, these emissions occur mainly
13                   during summer when O3 is of most concern.

14                   Hundreds of VOCs, containing mainly 2 to -12 carbon (C) atoms, are emitted by
15                   evaporation and combustion processes from a large number of anthropogenic sources.
16                   The two largest anthropogenic  source categories in the U.S. EPA's emissions inventories
17                   are industrial processes and transportation. Emissions of VOCs from highway vehicles
18                   account for roughly two-thirds  of the transportation-related emissions. The accuracy of
19                   VOC emission  estimates is difficult to determine, both for stationary and mobile sources.
20                   Evaporative emissions, which depend on temperature and other environmental factors,
21                   compound the difficulties of assigning accurate emission factors. In assigning VOC
22                   emission estimates to the mobile source category, models are used that incorporate
23                   numerous input parameters (e.g., type of fuel used, type of emission controls, and age of
24                   vehicle), each of which has some degree of uncertainty.

25                   On the U.S. and global scales, emissions of VOCs from vegetation are much larger than
26                   those from anthropogenic sources. Emissions of VOCs from anthropogenic sources in the
27                   2005 NEI were -17 MT/year (wildfires constitute -1/6 of that total and were included in
28                   the 2005 NEI under the anthropogenic category, but see Section 3.4 for how wildfires are
29                   treated for background.), but were  29 MT/year from biogenic sources. Uncertainties in
30                   both biogenic and anthropogenic VOC emission inventories prevent determination of the
31                   relative contributions of these two  categories, at least in many areas. Vegetation emits
32                   significant quantities of VOCs, such as terpenoid compounds (isoprene, 2-methyl-3-
33                   buten-2-ol, monoterpenes), compounds in the hexanal family, alkenes, aldehydes, organic
34                   acids, alcohols, ketones, and alkanes. The major chemicals emitted by plants are isoprene
35                   (40%), other terpenoid and sesqui-terpenoid compounds (25%) and the remainder
36                   consists of assorted oxygenated compounds and hydrocarbons  according to the 2005 NEI.
37                   Coniferous forests represent the largest source on a nationwide basis because of their
38                   extensive land coverage. Most biogenic emissions occur during the summer because of
      Draft - Do Not Cite or Quote                       3-7                                 September 2011

-------
 1                  their dependence on temperature and incident sunlight. Biogenic emissions are also
 2                  higher in southern states than in northern states for these reasons and because of species
 3                  variations. The uncertainty in natural emissions is about 50% for isoprene under midday
 4                  summer conditions and could be as much as a factor often higher for some compounds
 5                  (Guenther et al.. 2000). In EPA's regional modeling efforts, biogenic emissions of VOCs
 6                  are estimated using the BEIS model (U.S. EPA. 201 Ob) with data from the Biogenic
 7                  Emissions Landcover Database (BELD) and annual meteorological data. However, other
 8                  emissions models are used such as MEGAN (Model of Emissions of Gases and Aerosols
 9                  from Nature) (Guenther et al.. 2006). especially in global modeling efforts.

10                  Anthropogenic CO is emitted primarily by incomplete combustion of carbon-containing
11                  fuels. In general, any increase in fuel O2 content, burn temperature, or mixing time in the
12                  combustion zone will tend to decrease production of CO relative to CO2. However, it
13                  should be noted that controls mute the response of CO formation to fuel-oxygen. CO
14                  emissions from large fossil-fueled power plants are typically very low since the boilers at
15                  these plants are tuned for highly efficient combustion with the lowest possible fuel
16                  consumption. Additionally, the CO-to-CO2 ratio in these emissions is shifted toward CO2
17                  by allowing time for the furnace flue gases to mix with air and be oxidized by OH to CO2
18                  in the hot gas stream before the OH concentrations drop as the flue gases cool,.
19                  Nationally, on-road mobile sources constituted about half of total CO emissions in the
20                  2005 NEI. When emissions from non-road vehicles are included, it can be seen  from
21                  Figure 3-2 that all mobile  sources accounted for about three-quarters of total
22                  anthropogenic CO emissions in the U.S.

23                  Analyses by Harley et al. (2005) and Parrish (2006) are consistent with the suggestion in
24                  Pollack et al. (2004) that the EPA MOBILE6 vehicle emissions model (U.S. EPA. 2010d)
25                  overestimates vehicle CO  emissions by a factor of ~2. Field measurements by Bishop and
26                  Stedman (2008) were in accord with Parrish's  (2006) findings that the measured trends of
27                  CO and NOX concentrations from mobile sources in the U.S.  indicated that modeled CO
28                  emission estimates were substantially too high. Hudman et al. (2008) found that the NEI
29                  overestimated anthropogenic CO emissions by 60% for the eastern U.S. during the period
30                  July 1-August 15, 2004 based on comparison of aircraft observations of CO from the
31                  International Consortium for Atmospheric Research on Transport and Transformation
32                  (ICARTT) campaign (Fehsenfeld et al.. 2006)  and results  from a tropospheric chemistry
33                  model (GEOS-Chem). Improvements in emissions technologies not  correctly represented
34                  in MOBILE  emission models have been suggested as one cause for this discrepancy. For
35                  example, Pokharel et al. (2003. 2002) demonstrated substantial decrements in the CO
36                  fraction of tailpipe exhaust in several U.S. cities and Burgard et al. (2006) documented
37                  improvements in emission from heavy-duty on-road diesel engines.  Some of the largest
      Draft - Do Not Cite or Quote                      3-8                                September 2011

-------
 1                  errors in the MOBILE models are addressed in the successor model, MOVES (U.S. EPA.
 2                  2011e).

 3                  Estimates of biogenic CO emissions in the 2005 NEI are made in a manner similar to that
 4                  for VOCs. National biogenic emissions, excluding fires, were estimated to contribute
 5                  ~7% and wildfires added another -16% to the national CO emissions total.
 6                  Photodecomposition of organic matter in oceans, rivers, lakes, and other surface waters,
 7                  and from soil surfaces also releases CO (Goldstein and Galbally. 2007). However, soils
 8                  can act as a CO source  or a sink depending on soil moisture, UV flux reaching the soil
 9                  surface, and soil temperature (Conrad and  Seiler. 1985). Soil uptake of CO is driven by
10                  anaerobic bacteria (Inman et al., 1971). Emissions  of CO from soils appear to occur by
11                  abiotic processes, such as thermo- or photodecomposition of organic matter. In general,
12                  warm and moist conditions found in most soils favor CO uptake, whereas hot and dry
13                  conditions found in deserts and some savannas favor the release of CO (King. 1999).
            3.2.2   Gas Phase Reactions Leading to Ozone Formation

14                  Photochemical processes involved in O3 formation have been extensively reviewed in a
15                  number of books (Jacobson. 2002; Jacob. 1999; Seinfeld and Pandis, 1998; Finlayson-
16                  Pitts and Pitts.  1986) and in the previous O3 AQCDs. The photochemical formation of O3
17                  in the troposphere proceeds through the oxidation of NO to nitrogen dioxide (NO2) by
18                  organic (RO2) or hydro-peroxy (HO2) radicals. The photolysis of NO2 yields NO and a
19                  ground-state oxygen atom, O(3P), which then reacts with molecular oxygen to form O3.
20                  Free radicals oxidizing NO to NO2 are formed during the oxidation of VOCs (Annex
21                  AX2.2.2 in the 2006 O3 AQCD) (U.S. EPA. 2006b).

22                  VOCs important for the photochemical formation of O3 include alkanes, alkenes,
23                  aromatic hydrocarbons, carbonyl compounds (e.g., aldehydes and ketones), alcohols,
24                  organic peroxides, and halogenated organic compounds (e.g., alkyl halides). This array of
25                  compounds encompasses a wide range of chemical properties and lifetimes: isoprene has
26                  an atmospheric lifetime of approximately an hour, whereas methane has an atmospheric
27                  lifetime of about a decade.

28                  In urban areas, compounds representing all classes of VOCs and CO are important for O3
29                  formation. In nonurban vegetated areas, biogenic VOCs emitted from vegetation tend to
30                  be the most important. In the remote troposphere, methane (CH4) and CO are the main
31                  carbon-containing precursors to O3 formation. The oxidation of VOCs is initiated mainly
32                  by reaction with hydroxyl (OH) radicals. The primary source of OH radicals in the
33                  atmosphere is the reaction of electronically excited O atoms, O(:D), with water vapor.
34                  O(:D) is produced by the photolysis of O3 in the Hartley bands. In polluted areas, the


      Draft - Do Not Cite or Quote                       3-9                                September 2011

-------
 1                  photolysis of aldehydes (e.g., HCHO), HONO and H2O2 can also be significant sources
 2                  of OH, or HO2 radicals that can rapidly be converted to OH (Eisele et al.. 1997). O3 can
 3                  oxidize alkenes, as can NO3 radicals. NO3 radicals are most effective at night when they
 4                  are most abundant. In coastal environments and other selected environments, atomic Cl
 5                  and Br radicals can also initiate the oxidation of VOCs (Annex AX2.2.3 in the 2006 O3
 6                  AQCD) (U.S. EPA. 2006b). It should also be emphasized that the reactions of
 7                  oxygenated VOCs are important components of O3 formation (Annex AX2.2.9 in the
 8                  2006 O3  AQCD) (U.S. EPA. 2006b). They may be present in ambient air not only as the
 9                  result of the atmospheric oxidation of hydrocarbons but also by direct emissions. For
10                  example, motor vehicles and some industrial processes emit formaldehyde (Rappengliick
11                  et al.. 2009) and vegetation emits methanol.

12                  There are a large number of oxidized N-containing compounds in the atmosphere
13                  including NO, NO2, NO3, HNO2, HNO3, N2O5, HNO4, PAN and its homologues, other
14                  organic nitrates, such as alkyl nitrates, isoprene nitrates and particulate nitrate.
15                  Collectively these species are referred to as NOY. Oxidized nitrogen compounds are
16                  emitted to the atmosphere mainly as NO which rapidly interconverts with NO2 and so
17                  NO and NO2 are often "lumped" together into their own group or family, which is
18                  referred to as NOX. All the other species mentioned above in the definition of NOY are
19                  products of NOX reactions are referred to as NOZ, such that NOY = NOX + NOZ. The
20                  major reactions involving interconversions of oxidized N species were covered in the
21                  2006 O3  AQCD (Annex AX2.2.4). Mollner et al. (2010) identified pernitrous  acid
22                  (HOONO), an unstable isomer of nitric acid, as a product of the major gas phase reaction
23                  forming HNO3. However, since pernitrous acid is unstable, it is not a significant reservoir
24                  for NOX. This finding stresses the importance of identifying products in addition to
25                  measuring the rate of disappearance of reactants in kinetic studies.

26                  The photochemical cycles by which the oxidation of hydrocarbons leads to O3 production
27                  are best understood by considering the oxidation of methane, structurally the simplest
28                  VOC. The CH4 oxidation cycle serves as a model for the chemistry of the relatively clean
29                  or unpolluted troposphere (although this is a simplification because vegetation releases
30                  large quantities of complex VOCs, such as isoprene, into the atmosphere). In the polluted
31                  atmosphere, the underlying chemical principles are the same, as discussed in the 2006 O3
32                  AQCD (U.S. EPA. 2006b) (Annex AX2.2.5). The conversion of NO to NO2 occurring
33                  with the oxidation of VOCs is accompanied by the production of O3 and the efficient
34                  regeneration of the OH radical, which in turn can react with other VOCs as shown in
35                  Figure 3-1.

36                  The oxidation of alkanes and alkenes in the atmosphere has been treated in depth in the
37                  1996 O3  AQCD and was updated in the 2006 O3 AQCD (Annexes AX2.2.6 and
      Draft - Do Not Cite or Quote                      3-10                                September 2011

-------
 1                   AX2.2.7). In contrast to simple hydrocarbons containing one or two C atoms, detailed
 2                   kinetic information about the gas phase oxidation pathways of many anthropogenic
 3                   hydrocarbons (e.g., aromatic compounds such as benzene and toluene), biogenic
 4                   hydrocarbons (e.g., isoprene, the monoterpenes), and their intermediate oxidation
 5                   products (e.g., epoxides, nitrates, and carbonyl compounds) is lacking. Reaction with OH
 6                   radicals represents the major loss process for alkanes. Reaction with chlorine (Cl) atoms
 7                   is an additional sink for alkanes. Stable products of alkane photooxidation are known to
 8                   include carbonyl compounds, alkyl nitrates, and d-hydroxycarbonyls. Major uncertainties
 9                   in the atmospheric chemistry of the alkanes concern the chemistry of alkyl nitrate
10                   formation; these uncertainties affect the amount of NO-to-NO2 conversion occurring and,
11                   hence, the amounts of O3 formed during photochemical degradation of the alkanes.

12                   The reaction of OH radicals with aldehydes produced during the oxidation of alkanes
13                   forms acyl (R'CO) radicals, and acyl peroxy radicals (R'C(O)-O2) are formed by the
14                   further addition of O2. As an example, the oxidation of ethane (C2H5-H) yields
15                   acetaldehyde (CH3 -CHO). The reaction of CH3 -CHO with OH radicals yields acetyl
16                   radicals (CH3-CO). The acetyl radicals will then participate with O2 in a termolecular
17                   recombination reaction to form acetyl peroxy radicals, which can then react with NO to
18                   form CH3 + CO2 or they can react with NO2 to form PAN. PAN acts as a temporary
19                   reservoir for NO2.  Upon the thermal decomposition of PAN, either locally or elsewhere,
20                   NO2 is released to  participate in the O3 formation process again.

21                   Alkenes react in ambient air with OH, NO3,  and Cl radicals and with O3. All of these
22                   reactions are important atmospheric transformation processes, and all proceed by initial
23                   addition to the carbon double bonds. Major products of alkene photooxidation include
24                   carbonyl compounds. Hydroxynitrates and nitratocarbonyls, and decomposition products
25                   from the energy-rich biradicals formed in alkene-O3 reactions are also produced. Major
26                   uncertainties in the atmospheric chemistry of the alkenes concern the products and
27                   mechanisms of their reactions with O3, especially the yields of free radicals that
28                   participate in O3 formation. Examples  of oxidation mechanisms of complex alkanes and
29                   alkenes can be found in comprehensive texts such as Seinfeld and Pandis (1998). Apart
30                   from the effects of the oxidation of isoprene  on production of free radicals and O3
31                   formation, isoprene nitrates appear to play an important role as NOX reservoirs over the
32                   eastern U.S. (e.g.. Perring et al. 2009). Their decomposition leads to the recycling of
33                   NOX, which can participate in the O3 formation process again as was the case with
34                   decomposition of PAN and the even more unstable pernitrous acid. Although the
35                   photochemistry of isoprene is crucial for understanding ozone formation, major
36                   uncertainties in its  oxidation pathways still exist. Issues concern the lack of regeneration
37                   of OH + HO2 radicals especially in low NOX (<~ 1 ppb) environments. The
38                   isomerization of the isoprene hydroxy-peroxy radicals that are formed after initial OH
      Draft - Do Not Cite or Quote                       3-11                                September 2011

-------
 1                   attack and subsequent reactions could resolve this problem (Peeters and Muller. 2010;
 2                   Peeters et al.. 2009) and result in increases in OH concentrations from 20 to 40% over the
 3                   southeastern U.S. (Archibald et al.. 2011). Hofzumahaus et al. (2009) also found under
 4                   predictions of OH in the Pearl River Delta. They also note that the sequence of reactions
 5                   beginning with OH attack on VOCs introduces enormous complexity which is far from
 6                   being explored.

 7                   The oxidation of aromatic hydrocarbons constitutes an important component of the
 8                   chemistry of O3 formation in urban atmospheres (Annex AX2.2.8 in the 2006 O3 AQCD)
 9                   (U.S. EPA. 2006b). Virtually all of the important aromatic  hydrocarbon precursors
10                   emitted in urban atmospheres are lost through reaction with the hydroxyl radical. Loss
11                   rates for these compounds vary from slow (e.g., benzene) to moderate (e.g., toluene), to
12                   very rapid (e.g., xylene and trimethylbenzene isomers). However, the mechanism for the
13                   oxidation of aromatic hydrocarbons following reaction with OH is poorly understood, as
14                   is evident from the poor mass balance of the reaction products. The mechanism for the
15                   oxidation of toluene has been studied most thoroughly, and there is general agreement on
16                   the initial steps in the mechanism. However, at present there is no promising approach for
17                   resolving the remaining issues concerning the later steps. The oxidation of aromatic
18                   hydrocarbons also leads to particle formation that could remove gas-phase constituents
19                   that participate in O3 formation.

20                   Adequate analytical techniques needed to identify and quantify key intermediate species
21                   are not available for many compounds. In addition, methods to synthesize many of the
22                   suspected intermediate compounds are not available so that laboratory studies of their
23                   reaction kinetics cannot be performed.  Similar considerations apply to the oxidation of
24                   biogenic hydrocarbons besides isoprene. These considerations are important because
25                   oxidants, other than O3, that are formed from the chemistry described above could exert
26                   effects on human health and perhaps also on vegetation (Doyle et al., 2007; Doyle  et al..
27                   2004; Sexton et al.. 2004). Gas phase oxidants include PAN, H2O2, CH3OOH and other
28                   organic hydroperoxides.

29                   Ozone is lost through a number of gas phase reactions and  deposition to  surfaces. The
30                   reaction of O3 with NO to produce NO2, e.g., in urban centers near roads, mainly results
31                   in the recycling of O3 downwind via the recombination of O(3P) with O2 to re-form O3.
32                   By itself, this reaction does not lead to a net loss of O3 unless the NO2 is converted to
33                   stable end products such as HNO3. Ozone reacts with unsaturated hydrocarbons and with
34                   OH and HO2 radicals.

3 5                   Perhaps the most recent field study aimed at obtaining a better understanding of
36                   atmospheric chemical processes was the Second Texas Air Quality Field Study
37                   (TexAQS-II) conducted in Houston in August and September 2006 (Olaguer et al.. 2009).
      Draft - Do Not Cite or Quote                       3-12                                September 2011

-------
 1                   The TexAQS-II Radical and Aerosol Measurement Project (TRAMP) found evidence for
 2                   the importance of short-lived radical sources such as HCHO and HONO in increasing O3
 3                   productivity. During TRAMP, daytime HCHO pulses as large as 32 ppb were observed
 4                   and attributed to industrial activities upwind in the Houston Ship Channel (HSC) and
 5                   HCHO peaks as large as 52 ppb were detected by in-situ surface monitors in the HSC.
 6                   Primary HCHO produced in flares from local refineries and petrochemical facilities could
 7                   increase peak O3 by -30 ppb (Webster et al., 2007). Other findings from TexAQS-II
 8                   included significant concentrations of HONO during the day, with peak concentrations
 9                   approaching 1 ppb at local noon. These concentrations are well in excess of current air
10                   quality model predictions using gas phase mechanisms alone (Sarwar et al.. 2008) and
11                   multiphase processes are needed to account for these observations.  Olaguer et al. (2009)
12                   also noted that using measured HONO  brings modeled O3 concentrations  into much
13                   belter agreement with observations and could result in the production of an additional
14                   10 ppb O3. Large nocturnal vertical gradients indicating a surface or near-surface source
15                   of HONO, and  large concentrations of night-time radicals (-30 ppt HO2) were also found
16                   during TRAMP.
            3.2.3   Multiphase Processes

17                   In addition to reactions occurring in the gas phase, reactions occurring on the surfaces of
18                   or within cloud droplets and airborne particles also occur. Their collective surface area is
19                   huge, implying that collisions with gas phase species occur on very short time scales. In
20                   addition to hydrometeors (e.g., cloud and fog droplets and snow and ice crystals) there
21                   are also potential reactions involving atmospheric particles of varying composition (e.g.,
22                   wet [deliquesced] inorganic particles, mineral dust, carbon chain agglomerates and
23                   organic carbon particles) to consider. Multiphase reactions are involved in the formation
24                   of a number of species such as particulate nitrate, and gas phase HONO that can act to
25                   both increase and reduce the rate of O3 formation in the polluted troposphere. Data
26                   collected in Houston as part of TexAQS-II summarized by Olaguer et al. (2009) indicate
27                   that concentrations of HONO  are much higher than can be explained by gas phase
28                   chemistry and by tailpipe emissions; and that the photolysis of HONO formed in
29                   multiphase reactions in addition to the other sources can help narrow the discrepancy
30                   between observed and predicted production of O3. However, removal of HOX and NOX
31                   onto hydrated particles will reduce the production of O3.

32                   Multi-phase processes have been associated with the release of gaseous halogen
33                   compounds from marine aerosol, mainly in marine and coastal environments. However,
34                   Thornton et al.  (2010) found production rates  of gaseous nitryl chloride near Boulder, CO
35                   from reaction of N2O5 with particulate Cl", similar to those found in coastal and marine
      Draft - Do Not Cite or Quote                       3-13                                September 2011

-------
 1                   environments. C1NO2 readily photolyzes to yield Cl. They also found that substantial
 2                   quantities of N2O5 are recycled through C1NO2 back into NOX instead of forming HNO3
 3                   (a stable reservoir for reactive nitrogen compounds). The oxidation of hydrocarbons by
 4                   Cl radicals released from the marine aerosol could lead to the rapid formation of peroxy
 5                   radicals and higher rates of O3  production in selected coastal environments and in
 6                   continental environments. It should be noted that in addition to production from marine
 7                   aerosol, reactive halogen species are also produced by the oxidation of halogenated
 8                   organic compounds (e.g., CH3C1, CH3Br, and CH3I). The atmospheric chemistry of
 9                   halogens is complex because Cl, Br and I containing species can react among themselves
10                   and with hydrocarbons and other species and could also be important for O3 destruction,
11                   as has been noted for the lower stratosphere  (McElroy et al.. 1986; Yung et al.. 1980).
12                   For example, the reactions of Br and Cl containing radicals deplete O3 in selected
13                   environments such as the Arctic during the spring (Barrie et al.. 1988). the tropical
14                   marine boundary layer (Dickerson et al..  1999). and inland salt flats and salt lakes (Stutz
15                   et al.. 2002). Mahajan et al. (2010) found that I and Br species acting together resulted in
16                   O3 depletion that was much larger than would have been expected if they acted
17                   individually and did not interact with each other (see Section AX2.2.10.3). It should be
18                   stressed that knowledge  of multiphase processes is still evolving and there are still many
19                   questions that remain to  be answered. However, it is becoming clear that multiphase
20                   processes are important for O3  chemistry.

21                   Reactions of O3 with monoterpenes have been shown to produce oxidants in the aerosol
22                   phase, principally as components of ultrafine particles. Docherty et al. (2005) found
23                   evidence for the substantial production of organic hydroperoxides in secondary organic
24                   aerosol (SOA) resulting  from the reaction of monoterpenes with O3. Analysis of the SOA
25                   formed in their environmental chamber indicated that the SOA consisted mainly of
26                   organic hydroperoxides. In particular, they obtained yields of 47% and 85% of organic
27                   peroxides from the oxidation of a- and (3-pinene. The hydroperoxides then react with
28                   aldehydes in particles to form peroxyhemiacetals, which can either rearrange to form
29                   other compounds such as alcohols and acids or revert back to the hydroperoxides. The
30                   aldehydes are also produced in large measure during the ozonolysis of the monoterpenes.
31                   Monoterpenes also react with OH radicals resulting in the production of more
32                   lower-molecular-weight products than in the reaction with monoterpenes and O3. Bonn et
33                   al. (2004) estimated that hydroperoxides lead to 63% of global SOA formation from the
34                   oxidation of terpenes. The oxidation of anthropogenic aromatic hydrocarbons by OH
35                   radicals could also produce organic hydroperoxides in SOA (Johnson et al.. 2004).
36                   Recent measurements show that the abundance of oxidized SOA exceeds that of more
37                   reduced hydrocarbon like organic aerosol in Pittsburgh (Zhang et al.. 2005) and in about
38                   30 other cities across the Northern Hemisphere (Zhang et al.. 2007b). Based on aircraft
39                   and ship-based sampling of organic aerosols over coastal waters downwind of

      Draft - Do Not Cite or Quote                       3-14                                September 2011

-------
 1                  northeastern U.S. cities, de Gouw et al. (2008) reported that 40-70% of measured organic
 2                  mass was water soluble and estimated that approximately 37% of SOA is attributable to
 3                  aromatic precursors, using PM yields estimated for NOx-limited conditions.
 4                  Uncertainties still exist as to the pathways by which the oxidation of isoprene leads to the
 5                  formation of SOA. Noziere et al. (2011) found that a substantial fraction of 2-
 6                  methyltetrols are primary in origin, although these species have been widely viewed
 7                  solely as products of the atmospheric oxidation of isoprene. This finding points to
 8                  lingering uncertainty in reaction pathways in the oxidation of isoprene and in estimates of
 9                  the yield of SOA from isoprene oxidation.

10                  Reactions of O3 on the surfaces of particles, in particular those with humic acid like
11                  composition, are instrumental in the processing of SOA and the release of
12                  low-molecular-weight products such as HCHO (D'Anna et al.. 2009). However, direct
13                  reactions of O3 and atmospheric particles appear to be too slow to represent a major O3
14                  sink in the troposphere (D'Anna et al., 2009).
                    3.2.3.1    Indoor Air

15                  Except when activities such as photocopying or welding are occurring, the major source
16                  of O 3 to indoor air is through infiltration of outdoor air. Reactions involving ambient O3
17                  with NO either from exhaled breath or from gas-fired appliances, surfaces of furnishings
18                  and terpenoid compounds from cleaning products, air fresheners and wood products also
19                  occur in indoor air as was discussed in the previous O3 AQCD. The previous O3 AQCD
20                  also noted that the ozonolysis of terpenoid compounds could be a significant source of
21                  secondary organic aerosol in the ultrafine size fraction. Chen et al. (2011) examined the
22                  formation of secondary organic aerosol from the reaction of O3 that has infiltrated
23                  indoors with terpenoid components of commonly used air fresheners. They focused on
24                  the formation and decay of particle bound reactive oxygen species (ROS) and on their
25                  chemical properties. They found that the ROS content of samples can be decomposed
26                  into fractions that differ in terms of reactivity and volatility, however the overall ROS
27                  content of samples decays and over 90% is lost within a day at room temperature. This
28                  result also suggests loss of ROS during sampling periods longer than a couple of hours.
            3.2.4  Temperature and Chemical Precursor Relationships

29                  As might be expected based on the temperature dependence of many reactions involved
30                  in the production and destruction of O3 and the temperature dependence of emissions
31                  processes such as evaporation of hydrocarbon precursors and the emissions of
      Draft - Do Not Cite or Quote                      3-15                                September 2011

-------
 1                   biogenically important precursors such as isoprene, ambient concentrations of O3 also
 2                   show temperature dependence. Bloomer et al. (2009) determined the sensitivity of O3 to
 3                   temperature at rural sites in the eastern U.S. They found that O3 increased on average at
 4                   rural (CASTNET) sites by ~3.2 ppbv/°C before 2002, and after 2002 by -2.2 ppbv/°C.
 5                   This change in sensitivity was largely the result of reductions in NOX emissions from
 6                   power plants.  These results are in accord with model predictions by Wu et al. (2008a)
 7                   showing that the sensitivity of O3 to temperature decreases with decreases in precursor
 8                   emissions. However, this study was basically confined to the eastern U.S., but results
 9                   from sites downwind of Phoenix, AZ showed basically no sensitivity of O3  to
10                   temperature (R2=0.02) (U.S. EPA. 2006b). However, Wise and Comrie (2005) did find
11                   that meteorological parameters (mixing height and temperature) typically accounts for 40
12                   to 70% of the  variability in O3 in the five  southwestern cities (including Phoenix) they
13                   examined. It is likely that differences in the nature of sites chosen (urban vs. rural)
14                   accounted for this difference and is at least partially responsible for the difference in
15                   results. Jaffe et al. (2008) regressed O3 on temperature at Yellowstone and Rocky
16                   Mountain NP  and found weak associations (R2 = 0.09 and 0.16). They found that
17                   associations with area burned by wildfires are much stronger. These results  demonstrate
18                   that the associations of O3 with temperature are not as clear in the West as they might be
19                   in the East. Other sources as discussed in  Section 3.4 might also be more important in the
20                   West than in the East.

21                   The warmer months of the year are generally regarded as being the most conducive to
22                   higher O3 concentrations. However, Schnell et al. (2009) reported observations of high
23                   O3 concentrations (maximum 1-h avg of 140 ppb; maximum 8-h avg of 120 ppb) in the
24                   Jonah-Pinedale gas fields in Wyoming during winter at temperatures of -17°C. Potential
25                   factors contributing to these anomalously high concentrations include a highly reflective
26                   snow surface, emissions of short-lived radical reservoirs (e.g., HONO and HCHO) and a
27                   very shallow,  stable boundary layer trapping these emissions (Schnell et al.. 2009).
28                   Multiphase processes might also be involved in the production of these short4ived
29                   reservoirs. At a temperature of -17°C, the production of hydroxyl radicals (by the
30                   photolysis of O3 yielding O:D followed by the reaction, O(:D) + H2O, needed to initiate
31                   hydrocarbon oxidation) is severely limited, suggesting that another source office radicals
32                   is needed. Radicals can be produced by the photolysis of molecules such as HONO and
33                   HCHO which photolyze in optically thin regions of the solar spectrum. A similar issue,  in
34                   part due to the under-prediction office radicals, has arisen in the Houston airshed where
35                   chemistry transport models under-predict O3 (Olaguer et al., 2009).

36                   Rather than varying directly with emissions of its precursors, O3 changes in a nonlinear
37                   fashion with the concentrations of its precursors. At the low NOX concentrations found  in
38                   remote continental areas to rural  and suburban areas downwind of urban centers (low-
      Draft - Do Not Cite or Quote                       3-16                                 September 2011

-------
 1                   NOX regime), the net production of O3 typically increases with increasing NOX. At the
 2                   high NOX concentrations found in downtown metropolitan areas, especially near busy
 3                   streets and roads, and in power plant plumes, there is scavenging (titration) of O3 by
 4                   reaction with NO (high-NOx regime). In between these two regimes, there is a transition
 5                   stage in which O3 shows only a weak dependence on NOX concentrations.

 6                   In the low-NOx regime described above, the overall effect of the oxidation of VOCs is to
 7                   generate (or at least not consume) free radicals, and O3 production varies directly with
 8                   NOX. In the high-NOx regime, NO2 scavenges OH radicals which would otherwise
 9                   oxidize VOCs to produce peroxy radicals, which in turn would oxidize NO to NO2. In
10                   this regime, O3 production is limited by the availability of free radicals. The production
11                   of free radicals is in turn limited by the availability of solar UV radiation capable of
12                   photolyzing O3 (in the Hartley bands) or aldehydes and/or by the abundance of VOCs
13                   whose oxidation produce more radicals than they consume. There are a number of ways
14                   to refer to the chemistry in these two chemical regimes. Sometimes the terms VOC-
15                   limited and NOx-limited are used. However, there are difficulties with this usage because
16                   (1) VOC measurements  are not as abundant as they are for nitrogen oxides; (2) rate
17                   coefficients for reaction of individual VOCs with free radicals  vary over an extremely
18                   wide range; and (3) consideration is not given to CO nor to reactions that can produce
19                   free radicals without involving VOCs. The terms NOx-limited and NOx-saturated (Jaegle
20                   et al.. 2001) will be used wherever possible to more adequately describe  these two
21                   regimes. However, the terminology used in original articles will also be used here. In
22                   addition to these two regimes, there is also a "very low NOX regime" in the remote
23                   marine troposphere in which NOX concentrations are less than about 20 ppt. Under these
24                   very low NOX conditions, which are not likely to be found in the U.S, HO2 and CH3O2
25                   radicals react with each other and HO2 radicals undergo self-reaction (to form H2O2),
26                   and OH and HO2 react with O3, leading to net destruction of O3 and inefficient OH
27                   radical regeneration by comparison with much higher NOX concentrations found in
28                   polluted areas. In polluted areas, HO2 and CH3O2 radicals react with NO to convert NO
29                   to NO2, regenerate the OH radical, and, through the photolysis of NO2, produce  O3 as
30                   noted in 2006 O3 AQCD (U.S. EPA. 2006b) (Annex AX2.2.5). There are no sharp
31                   transitions between these regimes. For example, in the "low NOX regime" there  still may
32                   be significant peroxy-peroxy radical reactions depending on the local NOX concentration.
33                   In any case, in all of these NOX regimes, O3 production is also limited by the abundance
34                   of HOX radicals.

3 5                   The chemistry of OH radicals, which are responsible for initiating the oxidation of
36                   hydrocarbons, shows behavior similar to that for O3  with respect to NOX concentrations
37                   (Poppe et al.. 1993; Zimmermann and Poppe. 1993; Hameedetal.. 1979). These
38                   considerations introduce a high degree of uncertainty into attempts to relate changes in
      Draft - Do Not Cite or Quote                       3-17                                September 2011

-------
 1                   O3 concentrations to emissions of precursors. There are no definitive rules governing the
 2                   concentrations of NOX at which the transition from NOx-limited to NOx-saturated
 3                   conditions occurs. The transition between these two regimes is highly spatially and
 4                   temporally dependent and depends also on the nature and abundance of the hydrocarbons
 5                   that are present.

 6                   Trainer et al. (1993) and Olszyna et al. (1994) have shown that O3 and NOY are highly
 7                   correlated in rural areas in the eastern U.S. Trainer et al. (1993) also showed that O3
 8                   concentrations correlate even better with NOZ than with NOY, as may be expected
 9                   because NOZ represents the amount of NOX that has been oxidized, forming O3 in the
10                   process. NOZ is equal to the difference between measured total reactive nitrogen (NOY)
11                   and NOX and represents the summed products of the oxidation of NOX. NOZ is
12                   composed mainly of HNO3, PAN and other organic nitrates, particulate nitrate, and
13                   HNO4. Trainer et al. (1993) also suggested that the slope of the regression line between
14                   O3 and NOZ can be used to estimate the rate of O3 production per NOX oxidized (also
15                   known as the O3 production efficiency [OPE]). Ryerson et al. (2001; 1998) used
16                   measured correlations between O3 and NOZ to identify different rates of O3 production in
17                   plumes from large point sources. A number of studies in the planetary boundary layer
18                   over the continental U.S. have found that the OPE ranges typically from 1 to nearly 10.
19                   However, it may be higher in the upper troposphere and in certain areas, such as the
20                   Houston-Galveston area in Texas. Observations indicate that the OPE depends mainly on
21                   the abundance of NOX and also on availability of solar UV radiation, VOCs and O3
22                   itself.

23                   Various techniques have been proposed to use ambient NOX and VOC measurements to
24                   derive information about the dependence of O3 production on their concentrations. For
25                   example, it has been suggested that O3 formation in individual urban areas could be
26                   understood in terms of measurements of ambient NOX and VOC concentrations during
27                   the early morning (NRC.  1991). In this approach, the ratio of summed (unweighted) VOC
28                   to NOX is used to determine whether conditions were NOx-limited or VOC-limited. This
29                   procedure is inadequate because it omits many factors that are important for O3
30                   production such as the impact of biogenic VOCs (which are typically not present in urban
31                   centers during early morning); important differences in the ability of individual VOCs to
32                   generate free radicals (rather than just total VOC) and other differences in O3 forming
33                   potential for individual VOCs (Carter. 1995): and changes in the VOC to NOX ratio due
34                   to photochemical reactions and deposition as air moves downwind from urban areas
35                   (Milford et al.. 1994).

36                   Photochemical production of O3 generally occurs simultaneously with the production of
37                   various other species such as HNO3, organic nitrates, and other oxidants such as
      Draft - Do Not Cite or Quote                       3-18                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
hydrogen peroxide. The relative rate of production of O3 and other species varies
depending on photochemical conditions, and can be used to provide information about
O3-precursor sensitivity. Sillman (1995) and Sillman and He  (2002) identified several
secondary reaction products that show different correlation patterns for NOx-limited and
NOx-saturated conditions. The most important correlations are for O3 versus NOY, O3
versus NOZ, O3 versus HNO3, and H2O2 versus HNO3. The  correlations between O3 and
NOY, and O3 and NOZ are especially important because measurements of NOY and NOX
are more widely available than for VOCs. Measured O3 versus NOZ (Figure 3-3) shows
distinctly different patterns in different locations. In rural areas and in urban areas such as
Nashville, TN, O3 is highly correlated with NOZ. By contrast, in Los Angeles, CA, O3 is
not as highly correlated with NOZ, and the rate of increase of O3 with NOZ is lower and
the O3 concentrations for a given NOZ value are generally lower. The different O3 versus
NOZ relations in Nashville, TN and Los Angeles, CA reflects the difference between
NOx-limited conditions in Nashville versus an approach to NOx-saturated conditions in
Los Angeles.
                                                            -x-
                                                        X X
                                          10
                                  20
30
                                                 NOZ (ppb)
40
       Source: Adapted with permission of American Geophysical Union (Sillman and He, 2002: Sillman et al., 1998: Trainer et al., 1993)

      Figure 3-3    Measured concentrations of Os and NOz (NOy-NOx) during the
                     afternoon at rural sites in the eastern U.S. (grey circles) and in
                     urban areas and urban plumes associated with Nashville, TN (gray
                     dashes); Paris, France (black diamonds); and Los Angeles, CA (Xs).
16
17
The difference between NOx-limited and NOx-saturated regimes is also reflected in
measurements of H2O2. H2O2 production is highly sensitive to the abundance of free
      Draft - Do Not Cite or Quote
                              3-19
                   September 2011

-------
 1                   radicals and is thus favored in the NOx-limited regime. Measurements in the rural eastern
 2                   U.S. (Jacob etal..  1995). Nashville, TN (Sillman et al.. 1998). and Los Angeles, CA
 3                   (Sakugawa and Kaplan. 1989). show large differences in H2O2 concentrations between
 4                   likely NOx-limited and NOx-saturated locations.

 5                   The applications of indicator species mentioned above are limited to individual urban
 6                   areas either because they are based on point measurements or by the range of the aircraft
 7                   carrying the measurement instruments. Satellites provide a platform for greatly extending
 8                   the range of applicability of the indicator technique and also have the resolution
 9                   necessary to examine urban to rural differences. Duncan et al. (2010) used satellite data
10                   from OMI (Ozone Monitoring Instrument) for HCHO to NO2 column ratios to diagnose
11                   NOx-limited and radical-limited  (NOx-saturated) regimes. HCHO can be used as an
12                   indicator of VOCs as it is a common, short-lived, oxidation product of many VOCs that
13                   is a source of HOX (Sillman. 1995). In adopting the satellite approach, chemistry-
14                   transport models (discussed further in Section 3.3) are used to estimate the fractional
15                   abundance of the indicator species in the planetary boundary layer. Duncan et al. (2010)
16                   found that O3 formation over most of the U.S. became more sensitive to NOX over most
17                   of the U.S. from 2005 to 2007 largely because of decreases in NOX  emissions. They also
18                   found that surface  temperature is correlated with the ratio of HCHO to NO2 especially in
19                   cities in the Southeast where emissions of isoprene (a major source of HCHO) are  high
20                   due to high temperatures in summer.
          3.3   Atmospheric Modeling

21                   Chemistry-transport models (CTMs) have been widely used to compute the interactions
22                   among atmospheric pollutants and their transformation products, and the transport and
23                   deposition of pollutants. They have also been widely used to improve our basic
24                   understanding of atmospheric chemical processes and to develop control strategies. The
25                   spatial scales over which pollutant fields are calculated range from intra-urban to regional
26                   to global. Generally, these models are applied to problems on different spatial scales but
27                   efforts are underway to link across spatial scales for dealing with global scale
28                   environmental issues that affect population health within cities. Many features are
29                   common to all of these models and hence they share many of the same  problems. On the
30                   other hand, there are significant differences in approaches to parameterizing physical and
31                   chemical processes that must be addressed in applying these models across spatial scales.
32                   CTMs solve a set of coupled, non-linear partial differential equations, or continuity
33                   equations, for relevant chemical species.  Jacobson (2005) described the governing partial
34                   differential equations, and the methods that are used to solve them. Because of limitations
      Draft - Do Not Cite or Quote                       3-20                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
imposed by the complexity and spatial-temporal scales of relevant physical and chemical
processes, the CTMs must include parameterizations of these processes, which include
atmospheric transport; the transfer of solar radiation through the atmosphere; chemical
reactions; and removal to the surface by turbulent motions and precipitation.
Development of parameterizations for use in CTMs requires data for three dimensional
wind fields, temperatures, humidity, cloudiness, and solar radiation; emissions data for
primary (i.e., directly emitted from sources) species such as NOX, SO2, NH3, VOCs, and
primary PM; and chemical reactions.
      Figure 3-4     Sample CMAQ modeling domains. 36 km-grid-spacing; outer parent
                     domain in black; 12 km western U.S. (WUS) domain in  red; 12 km
                     eastern U.S. (EUS) domain in blue.
 9
10
11
12
13
14
15
The domains of CTMs extend from a few hundred kilometers on a side to the entire
globe. Most major regional (i.e., sub-continental) scale air-related modeling efforts at
EPA rely on the Community Multi-scale Air Quality modeling system (CMAQ) (Byun
and Schere. 2006; Byun and  Ching. 1999). CMAQ's horizontal domain typically extends
over North America with efforts underway to extend it over the entire Northern
Hemisphere. Note that CTMs can be 'nested' within each other as shown in Figure 3-4
which shows domains for CMAQ (Version 4.6.1); additional details on the model
      Draft - Do Not Cite or Quote
                             3-21
September 2011

-------
 1
 2
 3
 4
 5
 6
 9
10
11
12
13
configuration and application are found in (U.S. EPA. 2009e). The figure shows the outer
domain (36 km horizontal grid spacing) and two 12 km spatial resolution (east and west)
sub-domains. The upper boundary for CMAQ is typically set at about 100 hPa, or at
about 16 km altitude on average, although in some recent applications the upper
boundary has been set at 50 hPa. These domains and grid spacings are quite common and
can also be found in a number of other models.

The main components of a CTM such as EPA's CMAQ are summarized in Figure 3-5.
The capabilities of a number of CTMs designed to study local- and regional-scale air
pollution problems were summarized by Russell and Dennis (2000) and in the 2006 O3
AQCD. Historically, CMAQ has been driven most often by the MM5 mesoscale
meteorological model (Seaman.  2000). though it could be driven by other meteorological
models including the Weather Research Forecasting (WRF) model and the Regional
Atmospheric Modeling System (RAMS) (ATMET. 2011).
Initial/Boundary
Conditions and
Continuous Updates
of Met. Fields
from Observations


>• Meteorological
Model
                                                               Emissions
                                                                Model
                                                              Anthropogenic
                                                              (point, area sources)
                                                                  *
                                                             Biogenic Emissions
                                               Chemistry Transport Model
                                                   Visualization of Output
                                                     Process Analyses
      Figure 3-5     Main components of a comprehensive atmospheric chemistry
                     modeling system, such as the U.S. EPA's Community Model for Air
                     Quality (CMAQ) System.
14
15
Simulations of pollution episodes over regional domains have been performed with a
horizontal resolution down to 1 km; see the application and general survey results
      Draft - Do Not Cite or Quote
                             3-22
September 2011

-------
 1                   reported in Ching et al. (2006). However, simulations at such high resolution require
 2                   better parameterizations of meteorological processes such as boundary layer fluxes, deep
 3                   convection, and clouds (Seaman. 2000). Finer spatial resolution is necessary to resolve
 4                   features such as urban heat island circulation; sea, bay, and land breezes; mountain and
 5                   valley breezes; and the nocturnal low-level jet, all of which can affect pollutant concen-
 6                   trations. Other major air quality systems used for regional scale applications include the
 7                   Comprehensive Air Quality Model with extensions (CAMx) (ENVIRON. 2005) and the
 8                   Weather Research and Forecast model with Chemistry (WRF/Chem) (NOAA. 2010).

 9                   CMAQ and other grid-based or Eulerian air quality models subdivide the modeling
10                   domain into a three-dimensional array of grid cells. The most common approach to
11                   setting up the horizontal domain is to nest a finer grid within a larger domain of coarser
12                   resolution. The use of finer horizontal resolution in CTMs will  necessitate finer-scale
13                   inventories of land use and better knowledge of the exact paths of roads, locations of
14                   factories, and, in general, better methods for locating sources and estimating their
15                   emissions. The vertical resolution of CTMs is variable and usually configured to have
16                   more layers in the PEL and fewer in the free troposphere.

17                   The meteorological fields are produced either by other numerical prediction models such
18                   as those used for weather forecasting (e.g., MM5, WRF), and/or by assimilation of
19                   satellite data. The flow of information shown in Figure 3-5 has most often been
20                   unidirectional in the sense that information flows into the CTM (large box)  from outside;
21                   feedbacks on the meteorological fields and on boundary conditions (i.e., out of the box)
22                   have not been included. However, CTMs now have the capability to consider these
23                   feedbacks as well; see, for example, Binkowski et al. (2007)  and the Weather Research
24                   and Forecast model with Chemistry (WRF/Chem).

25                   Because of the large number of chemical species and reactions  that are involved in the
26                   oxidation of realistic mixtures of anthropogenic and biogenic hydrocarbons, condensed
27                   mechanisms must be used in atmospheric models. These mechanisms can be tested by
28                   comparison with smog chamber data. However, the existing chemical mechanisms often
29                   neglect many important processes such as the formation and subsequent reactions of
30                   long-lived carbonyl compounds, the incorporation of the most recent information about
31                   intermediate compounds, and heterogeneous reactions involving cloud droplets and
32                   aerosol particles.

33                   The initial conditions, or starting concentration fields of all species computed by a model,
34                   and the boundary conditions, or concentrations of species along the horizontal and upper
35                   boundaries of the model domain throughout the simulation, must be specified at the
36                   beginning of the simulation. Both initial and boundary conditions can be estimated from
37                   models or data or, more generally, model + data hybrids. Because data for vertical
      Draft - Do Not Cite or Quote                      3-23                                September 2011

-------
 1                  profiles of most species of interest are very sparse, results of model simulations over
 2                  larger, usually global, domains are often used.

 3                  Chemical kinetics mechanisms representing the important reactions occurring in the
 4                  atmosphere are used in CTMs to estimate the rates of chemical formation and destruction
 5                  of each pollutant simulated as a function of time. The Master Chemical Mechanism (Univ
 6                  of Leeds, 2010) is a comprehensive reaction database providing as near an explicit
 7                  treatment of chemical reactions in the troposphere as is possible. The MCM currently
 8                  includes over 12,600 reactions and 4,500 species. However, mechanisms that are this
 9                  comprehensive are still computationally too demanding to be incorporated into CTMs for
10                  regulatory use. Simpler treatments of tropospheric chemistry have been assembled by
11                  combining chemical species into mechanisms that group together compounds with
12                  similar chemistry. It should be noted that because of different approaches to the lumping
13                  of organic compounds into surrogate groups for computational efficiency, chemical
14                  mechanisms can produce different results under similar conditions. Jimenez et al. (2003)
15                  provided brief descriptions of the features of the main mechanisms in use and compared
16                  concentrations of several key species predicted by seven chemical mechanisms in a box-
17                  model simulation over 24 hours. There are several of these mechanisms (CB04, CB05,
18                  SAPRC) that have been incorporated into CMAQ (Luecken et al.. 2008) and Fuentes et
19                  al. (2007) for RACM2. The CB mechanism is currently undergoing extension (CB06) to
20                  include, among other things, longer lived species to better simulate chemistry in the
21                  remote and upper troposphere. These mechanisms were developed primarily for
22                  homogeneous gas phase reactions and treat multi-phase chemical reactions in a very
23                  cursory manner, if at all. As an example of the effects of their neglect, models such as
24                  CMAQ could have difficulties with capturing the regional nature of O3 episodes, in part
25                  because of uncertainty in the chemical pathways converting NOX to HNO3 and recycling
26                  of NOX (Godowitch et al.. 2008; Hains et al.. 2008). Much of this uncertainty also
27                  involves multi-phase processes as described in Section  3.2.

28                  CMAQ and other CTMs incorporate  processes and interactions of aerosol-phase
29                  chemistry (Zhang and Wexler. 2008: Gavdos et al.. 2007: Binkowski and Roselle. 2003).
30                  There have also been several attempts to study the feedbacks of chemistry on
31                  atmospheric dynamics using meteorological models like MM5  and WRF (Liu et al..
32                  2001: Park etaL 2001: Grell et al.. 2000: LuetaL 1997). This coupling is necessary to
33                  accurately simulate feedbacks from PM (Park et al.. 2001: Lu et al.. 1997) over areas
34                  such as Los Angeles or the Mid-Atlantic region. Photolysis rates in CMAQ can now be
35                  calculated interactively with model produced O3, NO2, and aerosol fields (Binkowski et
36                  al.. 2007).
      Draft - Do Not Cite or Quote                      3-24                                September 2011

-------
 1                   Spatial and temporal characterizations of anthropogenic and biogenic precursor emissions
 2                   must be specified as inputs to a CTM. Emissions inventories have been compiled on grids
 3                   of varying resolution for many hydrocarbons, aldehydes, ketones, CO, NH3, and NOX.
 4                   Preprocessing of emissions data for CMAQ is done by the Spare-Matrix Operator Kernel
 5                   Emissions (SMOKE) system (CEMPD. 2011). For many species, information concerning
 6                   the temporal variability of emissions is lacking, so long-term annual averages are used in
 7                   short-term, episodic simulations. Annual emissions estimates can be modified by the
 8                   emissions model to produce emissions more characteristic of the time of day and season.
 9                   Significant errors in emissions can occur if inappropriate time dependence is used.

10                   Each of the model components described above has associated uncertainties; and the
11                   relative importance of these uncertainties varies with the modeling application. The
12                   largest errors in photochemical modeling are still thought to arise from the
13                   meteorological and emissions inputs to the model (Russell and Dennis. 2000). While the
14                   effects of poorly specified boundary conditions propagate through the model's domain,
15                   the effects of these errors remain undetermined. Because many meteorological processes
16                   occur on spatial scales smaller than the model's vertical or horizontal grid spacing and
17                   thus are not calculated explicitly, parameterizations of these processes must be used.
18                   These parameterizations introduce additional uncertainty.

19                   The performance of CTMs must be evaluated by comparison with field data as part of a
20                   cycle of model evaluations and subsequent improvements (NRC. 2007). However, they
21                   are too demanding of computational time to have the full range of their sensitivities
22                   examined by using Monte Carlo techniques (NRC. 2007). Models of this complexity are
23                   evaluated by comparison with field observations for O3 and other species. Evaluations of
24                   the performance of CMAQ are given in Arnold et al. (2003). Eder and Yu (2005). Appel
25                   et al. (2005). and Fuentes and Raftery (2005). Discrepancies between model predictions
26                   and observations can be used to point out gaps in current understanding of atmospheric
27                   chemistry and to spur improvements in parameterizations of atmospheric chemical and
28                   physical processes. Model evaluation does not merely involve a straightforward
29                   comparison between model predictions and the concentration field of the pollutant of
30                   interest. Such comparisons may not be meaningful because it is difficult to determine if
31                   agreement between model predictions and observations truly represents an accurate
32                   treatment of physical and chemical processes in the CTM or the effects of compensating
33                   errors in complex model routines (in other words, it is important to know if the right
34                   answer is obtained for the right reasons). Each of the model components (emissions
35                   inventories, chemical mechanism, and meteorological driver) should be evaluated
36                   individually as has been done in to large extent in some major field studies such as
37                   TexAQS I and II. In addition to comparisons between concentrations of calculated and
3 8                   measured species, comparisons of correlations between measured primary VOCs and
      Draft - Do Not Cite or Quote                      3-25                               September 2011

-------
 1                  NOX and modeled VOCs and NOX are especially useful for evaluating results from
 2                  chemistry-transport models. Likewise, comparisons of correlations between measured
 3                  species and modeled species can be used to provide information about the chemical state
 4                  of the atmosphere and to evaluate model representations. A CTM that demonstrates the
 5                  accuracy of both its computed VOC and NOX in comparison with ambient
 6                  measurements, and the spatial and temporal relations among the critical secondary
 7                  species associated with O3  has a higher probability of representing O3-precursor
 8                  relations correctly than one that does not.

 9                  The above techniques are sometimes referred to as "static" in the sense that individual
10                  model variables are compared to observations. It is also crucial to understand the
11                  dynamic response to changes in inputs and to compare the model responses to those that
12                  are observed. These tests might involve changes  in some natural forcing or in emissions
13                  from an anthropogenic source. As an example, techniques such as the direct decoupled
14                  method (DDM) (Dunker et al.. 2002; Dunker. 1981) could be used. However, the
15                  observational basis for comparing a model's response is largely unavailable for many
16                  problems of interest, in large part because meteorological conditions are also changing
17                  while the emissions are changing. As a result, methods such as DDM are used mainly to
18                  address the effectiveness of emissions controls.
            3.3.1   Global Scale CTMs

19                  With recognition of the global nature of many air pollution problems, global scale CTMs
20                  have been applied to regional scale pollution problems (NRC. 2009). Global-scale CTMs
21                  are used to address issues associated with global change, to characterize long-range
22                  transport of air pollutants, and to provide boundary conditions for the regional-scale
23                  models. The upper boundaries of global scale CTMs extend anywhere from the
24                  tropopause (~8 km at the poles to -16 km in the tropics) to the mesopause at -80 km, in
25                  order to obtain more realistic boundary conditions for problems involving stratospheric
26                  dynamics and chemistry. The global-scale CTMs consider the same processes shown in
27                  Figure 3-5 for the regional scale models. In addition, many of the same issues that have
28                  arisen for the regional models have also arisen for the global scale models (Emmerson
29                  and Evans. 2009). For example, predictions of HNO3 were found to be too high and
30                  predictions of PAN were found to be too low over the U.S. during summer in the
31                  MOZART model (Fang et al..  2010). Similar findings were obtained in a box model of
32                  upper tropospheric chemistry (Henderson et al., 2010).

33                  The GEOS-Chem model is a community-owned, global scale CTM that has been widely
34                  used to study issues associated with the intra- and inter-hemispheric transport of pollution
      Draft - Do Not Cite or Quote                      3-26                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
and global change (Harvard University. 2010a). Comparisons of the capabilities of
GEOS-Chem and several other models to simulate intra-hemispheric transport of
pollutants are given in a number of articles (Fiore et al., 2009; Reidmiller et al., 2009).
Reidmiller et al. (2009) showed comparisons among 18 global models and their ensemble
average to spatially and monthly averaged observations of O3 at CASTNET sites (see
Figure 3-6). These  results show that the multi-model ensemble agrees much better with
the observations than do most of the individual models. The GEOS-Chem model was run
for two grid spacings, 4°x4.5° and 2°x2.5° with very similar results that lie close to the
ensemble average.  In general, the model ensemble and the two GEOS-Chem simulations
are much closer to the observations in the Intermountain West than in the Southeast. In
particular, there are sizable over-predictions by most of the models in the Southeast
during summer, the time when major O3 episodes occur.
                                  Mountain West Region
                             JFMAMJJASOND
                                            -B- CAMCHEM
                                            -B- ECHAM5
                                                EM5P
                                            -B- FRSGCUCI
                                            -a— GEI\*«a-EC
                                            -B— GEMAQ-vl pO
                                            -B- GEOSChem-v07
                                            -B- GEOSChem-v45
                                            -B- GISS-PUCCINI
                                            -B- GWI
                                             • •  MCA-vSSz
                                            -B- LLNL-IWPACT
                                            -B- MOZARTGFDL
                                                MOZECH
                                            -a— OsloCTM2
                                            -a— TM5-JRC
                                            -A-OBS
                                            —•—Multi-model mean
                                  MAMJJASOND
       Source: Used with permission from Copernicus Publications (Reidmiller et al.. 2009)

      Figure 3-6    Comparison of global CTM predictions of maximum daily 8-h avg
                     ozone concentrations and multi-model mean with monthly
                     averaged CASTNET observations in the Intermountain West and
                     Southeast regions of the U.S.
      Draft - Do Not Cite or Quote
                             3-27
September 2011

-------
 1                   Global models are not alone in overestimating O3 in the Southeast. Godowitch et al.
 2                   (2008). Gilliland et al. (2008) and Nolte et al. (2008) found positive O3 biases in regional
 3                   models over the eastern U.S., as well, which they largely attributed to uncertainties in
 4                   temperature, relative humidity and planetary boundary layer height. Agreement between
 5                   monthly average values is expected to be better than with daily values because of a
 6                   number of factors including the increasing uncertainty of emissions at finer time
 7                   resolution. Kasibhatla and Chameides (2000) found that the accuracy of simulations
 8                   improved as the averaging time of both the simulation and the observations increased.

 9                   Simulations of the effects of long-range transport at particular locations must be able to
10                   link multiple horizontal resolutions from the global to the local scale. Because of
11                   limitations on computational resources, global simulations are not made at the same
12                   horizontal resolutions found in the regional scale models, i.e., down to 1-4 km resolution
13                   on a side. They are typically conducted with a horizontal grid spacing of l°-2° of latitude
14                   and longitude (or roughly 100-200 km at mid-latitudes). Some models such as GEOS-
15                   Chem have the capability to include nested models at a resolution of 0.5°x0.667° (Wang
16                   et al.. 2009a) and efforts are underway to achieve even higher spatial resolution. Another
17                   approach is to nest regional models within GEOS-Chem. Caution must be exercised with
18                   nesting different models because of differences in chemical mechanisms  and numerical
19                   schemes, and in boundary conditions between the outer and inner models. As an example
20                   of these issues, surface O3 concentrations that are too high have been observed in models
21                   in which CMAQ was nested inside of GEOS-Chem. The high O3 results  in large measure
22                   from stratospheric O3 intruding into the CMAQ domain [for one way to address this issue
23                   see Lam (2010)1. In addition, downward mixing of this O3 in CMAQ that is too rapid
24                   might also be involved.  Ozone has large vertical gradients in the upper troposphere that
25                   must be preserved if its  downward transport is to be simulated correctly.  Using a vertical
26                   resolution in CMAQ that is too coarse could be involved, coupled with using fewer layers
27                   in CMAQ than in the driving MM5 or WRF meteorological model. As a  result of the
28                   above factors, O3 gradients are eliminated and O3 is mixed too rapidly in the upper
29                   troposphere. Efforts are also being made to extend the domain of CMAQ over the
30                   Northern Hemisphere. In this approach, the same  numerical schemes are  used for
31                   transporting species and the same chemistry is used throughout all spatial scales. Finer
32                   resolution in models of any scale can only improve scientific understanding to the extent
33                   that the governing processes are accurately described. Consequently, there is a crucial
34                   need for observations at the appropriate scales to evaluate the scientific understanding
35                   represented by the models.
      Draft - Do Not Cite or Quote                       3-28                                 September 2011

-------
          3.4   Background Ozone Concentrations

 1                   Background concentrations of O3 have been given various definitions in the literature
 2                   over time. In the context of a review of the NAAQS, it is useful to define background O3
 3                   concentrations in a way that distinguishes between concentrations that result from
 4                   precursor emissions that are relatively less directly controllable from those that are
 5                   relatively more directly controllable through U.S. policies. North American (NA)
 6                   background O3 can include contributions that result from emissions from natural sources
 7                   (e.g., stratospheric intrusion, biogenic methane and more short-lived VOC emissions),
 8                   emissions of pollutants that contribute to global concentrations of O3 (e.g., anthropogenic
 9                   methane) from countries outside North America. In previous NAAQS reviews, a specific
10                   definition of background concentrations was used and referred to as policy relevant
11                   background (PRB). In those previous reviews, PRB concentrations were defined by EPA
12                   as those concentrations that would occur in the U.S. in the absence of anthropogenic
13                   emissions in continental North America (CNA), defined here as the U.S., Canada, and
14                   Mexico.  For this document, we have focused on the sum of those background
15                   concentrations from natural sources everywhere in the world and from anthropogenic
16                   sources outside CNA. North American background concentrations so defined facilitate
17                   separation of pollution that can be controlled directly by U.S. regulations or through
18                   international agreements with neighboring countries  from that which would require more
19                   comprehensive international agreements, such as are being discussed as part of the
20                   United Nations sponsored Convention on Long Range Transboundary Air Pollution Task
21                   Force on Hemispheric Air Pollution. There is  no chemical difference between
22                   background O3 and O3 attributable to  CNA anthropogenic sources, and  background
23                   concentrations can contribute to the risk of health effects. However, to inform policy
24                   considerations regarding the current or potential alternative standards, it is useful to
25                   understand how total O3 concentrations can be attributed to different sources.

26                   Contributions to  NA background O3 include photochemical reactions involving natural
27                   emissions of VOCs, NOX, and CO as well as  the long-range transport of O3 and its
28                   precursors from outside CNA, and the stratospheric-tropospheric exchange (STE)  of O3.
29                   These sources have the greatest potential for producing the highest background
30                   concentrations, and therefore are discussed in greater detail below. Natural sources of O3
31                   precursors include biogenic emissions, wildfires, and lightning. Biogenic emissions from
32                   agricultural activities in CNA are not considered in the formation of NA background O3.
33                   Sources included in the definition of NA background O3 are  shown schematically in
34                   Figure 3-7. Definitions of background and approaches to derive background
35                   concentrations were reviewed in the 2006 O3  AQCD and in Reid et al. (2008).
      Draft - Do Not Cite or Quote                      3-29                                September 2011

-------
               Outside natural
               influences

       Long-range transport
       of pollution
                                              stratosphere
                                                                      lightning
                                                      "Background" air
Figure 3-7
                                                       t      5
                                                  Fires  Land       Human
                                                         biosphere  activity
                    Schematic overview of contributions to North American
                    background concentrations of ozone, i.e., ozone concentrations
                    that would exist in the absence of anthropogenic emissions from
                    the U.S., Canada, and Mexico.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16

17
18
      3.4.1   Contributions from Anthropogenic Emissions outside North America

             In addition to emissions from North America, emissions from Eurasia have contributed to
             the global burden of O3 in the atmosphere and to the U.S. (NRC. 2009. and references
             therein). Because the mean tropospheric lifetime of O3 is 30-35 days (Hsu and Prather.
             2009). O3 can be transported from continent to continent and around the globe in the
             Northern Hemisphere and O3 produced by U.S. emissions can be recirculated around
             northern mid-latitudes back to the U.S. High elevation sites are most susceptible to the
             intercontinental transport of pollution especially during spring. An O3 concentration of
             ~85 ppb was observed at Mt. Bachelor Observatory, OR (elevation 2,700 m) on April 22,
             2006 with a number of occurrences of O3  >60 ppb from mid-April to mid-May of 2006.
             Calculations using GEOS-Chem, a global-scale, chemistry-transport model, indicate that
             Asia contributed 9 ± 3 ppb to a modeled mean concentration of 53 ± 9 ppb O3 at Mt.
             Bachelor during the same period compared to measured concentrations of 54 ± 10 ppb
             (Zhang et al.. 2008). Zhang et al. (2008) also calculated a contribution of 5 to 7 ppb to
             surface O3 over the western U.S. during that period from Asian anthropogenic emissions.
             They also estimated an increase in NOX emissions of- 44% from Asia from 2001 to
             2006 resulting in an increase of 1-2 ppb in O3 over North America.

             Cooper et al. (2010) analyzed all available O3 measurements in the free troposphere
             above western North America at altitudes  of 3-8 km (above sea level) during April and
Draft - Do Not Cite or Quote
                                               3-30
September 2011

-------
 1                  May of 1995 to 2008 (i.e., times when intercontinental transport is most prominent).
 2                  They derived a trend of+0.63 ± 0.34 ppb/year in median O3 concentrations with
 3                  indication of a similar rate of increase since 1984. Back trajectories that were likely to
 4                  have been strongly and recently influenced by North American emissions were filtered
 5                  out, resulting in a trend of+0.71 ± 0.45 ppb/year. Considering only trajectories with an
 6                  Asian origin resulted in a trend of +0.80 ± 0.34 ppb/year. These results suggest that local
 7                  North American emissions were not responsible for the measured O3 increases. This O3
 8                  could have been produced from natural and anthropogenic precursors in Asia and Europe
 9                  with some contribution from North American emissions that have circled the globe.
10                  Cooper et al. (2010) also found that it is unlikely that the trends in tropospheric O3 are
11                  associated with trends in stratospheric intrusions. Note, however, that these results relate
12                  to O3 trends above ground level and not to surface O3. Jaffe (2011) found associations
13                  between ozonesonde data and the average of 10 CASTNET Sites in the western U.S. with
14                  R2 ranging between 0.048 in October and 0.45 in August for all days on a monthly basis
15                  for which there was an ozonesonde launch. Model results (Zhang et al.. 2008) show that
16                  surface O3 contributions from Asia are  much smaller than those derived in the free
17                  troposphere because of dilution and chemical destruction during downward transport to
18                  the surface. These processes tend to reduce the strength of associations between free
19                  tropospheric and surface O3 especially if air from other sources is sampled by the surface
20                  monitoring sites.

21                  Sampling locations and times at which measurements might be expected to reflect in
22                  large measure North American background O3 contributions include Trinidad Head, CA
23                  at times during spring (Oltmans et al.. 2008; Goldstein et al.. 2004). The monitoring
24                  station at Trinidad Head is on an elevated peninsula extending out from the mainland of
25                  northern California, and so might be  expected at times to intercept air flowing in from the
26                  Pacific Ocean with little or no influence from sources on the mainland. Figure 3-8 shows
27                  the time series of daily maximum 8-h avg O3 concentrations measured at Trinidad Head
28                  from April 18, 2002 through December 31, 2009. The data show pronounced seasonal
29                  variability with spring maxima and summer minima. Springtime concentrations typically
30                  range from 40 to 50 ppb with a number of occurrences >50 ppb. The two highest daily
31                  maxima were 60 and 62 ppb. The data also show much lower concentrations during
32                  summer, with concentrations typically ranging between 20 and 30 ppb. Oltmans et al.
33                  (2008) examined the time series of O3 and back trajectories reaching Trinidad Head.
34                  They found that springtime maxima (April-May) were largely associated with back
35                  trajectories passing over the Pacific Ocean and most likely entraining emissions from
36                  Asia, with minimal interference from local sources. However, Parrish et al. (2009) noted
37                  that only considering trajectories coming from a given direction is not sufficient for
38                  ruling out local continental influences, as sea breeze circulations are complex phenomena
39                  involving vertical mixing and entrainment of long-shore components. They found that

      Draft - Do Not Cite or Quote                       3-31                                 September 2011

-------
 1                   using a wind speed threshold, in addition to a criterion for wind direction, allowed for
 2                   determination of background trajectories not subject to local influence; as judged by
 3                   measurements of chemical tracers such as CO2, MTBE and radon. By applying the two
 4                   criteria for wind speed and direction, they found that Trinidad Head met these criteria
 5                   only 30% of the time during spring. Goldstein et al. (2004) used CO2 as an indicator of
 6                   exchange with the local continental environment and found that O3 concentrations were
 7                   higher by about 2-3 ppb when filtered against local influence indicating higher O3 in air
 8                   arriving from over the Pacific Ocean. At Trinidad Head during spring, O3 is more likely
 9                   to be titrated by local emissions of NOX than to be photochemically produced (Parrish et
10                   al.. 2009). At other times of the year, Trinidad Head is less strongly affected by air
11                   passing over Asia and many trajectories have long residence times over the semi-tropical
12                   and tropical Pacific Ocean, where O3 concentrations are much lower than they are at mid-
13                   latitudes. The use of the Trinidad Head data to derive contributions from background
14                   sources requires the use of screening procedures adopted by Parrish et al. (2009) and the
15                   application of photochemical models to determine the extent either of titration of O3 by
16                   fresh NOX emissions and the extent of local production of O3 from these emissions. As
17                   noted above, anthropogenic emissions from North America also contribute to
18                   hemispheric background and must be filtered out from observations even when it is
19                   thought that air sampled came directly from over the Pacific Ocean and was not
20                   influenced by local pollutant emissions.

21                   Parrish et al. (2009) also examined data obtained at other marine boundary layer sites on
22                   the Pacific Coast. These include Olympic NP, Redwood NP, Point Arena, and Point
23                   Reyes. Using data from these sites, they derived trends in O3 of+0.46 ppb/year (with a
24                   95% confidence interval of 0.13 ppb/year) during spring and +0.34 ppb/year
25                   (0.09 ppb/year) for the annual mean O3 increase in air arriving from over the Pacific
26                   during the past two decades. Although O3 data are available from the Channel Islands,
27                   Parrish et al. (2009) noted that these data are not suitable for determining background
28                   influence because of the likelihood of circulating polluted air from the South Coast Basin.

29                   Cooper et al. (In Press) further examined O3 profiles measured above four coastal sites in
30                   California, including Trinidad Head. Based on comparison with the ozone profiles, they
31                   suggested that Asian pollution, stratospheric intrusions and international shipping made
32                   substantial contributions to lower tropospheric O3 measured at inland California sites.
33                   These contributions tended to increase on a relative basis in going from south north. In
34                   particular, no increases in lower tropospheric O3 in the northern Central Valley, and
35                   increases of 32 to 63% in the LA basin due to local pollution were found. It should be
36                   noted that the extent of photochemical production and loss, involving both anthropogenic
37                   and natural precursors, occurring in descending air still remains to be determined. Cooper
38                   et al. (In Press) also note that very little (8-10%) of the sources noted above and affecting
      Draft - Do Not Cite or Quote                       3-32                                September 2011

-------
1
2
3
the vertical O3 measurements reach the eastern U.S. However, this does not necessarily
mean that the effects of the Asian sources were fully captured in the ozone profiles or that
stratospheric intrusions do not occur over the eastern U.S.
                                           Trinidad Head
                 0.07
                 0.06
                        CO Q
                     coQS->coQS->coQS-
                   Time (Days: April 19, 2002 - December 31, 2009
      Source: Used with permission from Elsevier Ltd. (Oltmans et al.. 2008) and NOAA Climate Monitoring Diagnostics Laboratory for
     data from 2008-2009

     Figure 3-8     Time series of daily maximum 8-h avg ozone concentrations (ppm)
                    measured at Trinidad Head, CA, from April 18, 2002 through
                    December 31, 2009.
           3.4.2   Contributions from Natural Sources
4
5
3.4.2.1   Contributions from the Stratosphere

The basic atmospheric dynamics and thermodynamics of STE were outlined in the 2006
O3 AQCD; as noted there, stratospheric air rich in O3 is transported into the troposphere.
     Draft - Do Not Cite or Quote
                             3-33
September 2011

-------
 1                   Ozone is produced naturally by photochemical reactions in the stratosphere as shown in
 2                   Figure 3-1 in Section 3.2. Some of this O3 is transported downward into the troposphere
 3                   throughout the year, with maximum contributions at mid-latitudes during late winter and
 4                   early spring mainly coming from a process known as tropopause folding. These folds
 5                   occur behind most cold fronts, bringing stratospheric air with them. The tropopause
 6                   should not be interpreted as a material surface through which there is no exchange.
 7                   Rather these folds should be thought of as regions in which mixing of tropospheric and
 8                   stratospheric air is occurring (Shapiro. 1980). This imported  stratospheric air contributes
 9                   to the natural background of O3 in the troposphere, especially in the free troposphere
10                   during winter and spring. STE also occurs during other seasons including summer.

11                   Methods for estimating the contribution of stratospheric intrusions rely on the use of
12                   tracers of stratospheric origin that can be either dynamical or chemical. Thompson et  al.
13                   (2007). based on analysis of ozonesonde data found that roughly 20-25% of tropospheric
14                   O3 over northeastern North America during July-August 2004 was of stratospheric
15                   origin. This O3 can be mixed into the PEL where it can either be destroyed or transported
16                   to the surface. They relied on the combined use of low relative humidity and high
17                   (isentropic) potential vorticity (PV) (> 2 PV units) to identify stratospheric contributions.
18                   PV has been a widely used tracer for stratospheric air; see the 2006 O3 AQCD. Lefohn et
19                   al. (2011) used these and additional criteria to assess stratospheric influence on sites in
20                   the intermountain West and in the Northern Tier. Additional  criteria include
21                   consideration of trajectories originating at altitudes above the 380 K potential
22                   temperature surface with a residence time requirement at these heights. They identified
23                   likely stratospheric influence at the surface sites  on a number of days during spring of
24                   2006 to 2008. However, they noted that their analysis of stratospheric intrusions captures
25                   only the frequency and vertical penetration of the intrusions but does not provide
26                   information about the contribution of the intrusions to the measured O3 concentration.
27                   These results are all generally consistent with what was noted in the 2006 O3 AQCD.
28                   Fischer et al. (2004) analyzed the O3 record during summer at Mount Washington and
29                   identified a stratospheric contribution to 5% of events during the summers of 1998 -2003
30                   when O3 was > 65 ppb; the air was dry and trajectories originated from altitudes where
31                   potential vorticity was elevated (PV > 1 PV unit). However, this analysis did not quantify
32                   the relative contributions of anthropogenic and stratospheric  O3 sources, because as they
33                   note identifying stratospheric influences is complicated by transport over
34                   industrialized/urban source regions. Stratospheric O3 was hypothesized to influence the
35                   summit during conditions also potentially conducive to photochemical O3 production,
36                   which make any relative contribution calculations difficult without additional
37                   measurements of anthropogenic and stratospheric tracers.
      Draft - Do Not Cite or Quote                       3-34                                 September 2011

-------
 1                   Although most research has been conducted on tropopause folding as a source of
 2                   stratosphere to troposphere exchange, this is not the only mechanisms by which
 3                   stratospheric ozone can be brought to lower altitudes. Tang et al. (2011) estimated that
 4                   deep convection capable of penetrating the tropopause can increase the overall downward
 5                   flux of O3 by ~ 20%. This mechanism operates mainly during summer in contrast with
 6                   tropopause folding which is at a maximum from late winter through spring and at lower
 7                   latitudes. Yang et al. (2010) estimated that roughly 20% of free tropospheric O3 above
 8                   coastal California in 2005  and 2006 was stratospheric in origin. Some of this O3  could
 9                   also contribute to O3 at the surface.

10                   It should be noted that there is considerable uncertainty in the magnitude and distribution
11                   of this potentially important source of tropospheric O3. Stratospheric intrusions that reach
12                   the surface are much less frequent than intrusions which penetrate only to the middle and
13                   upper troposphere. However,  O3 transported to the upper and middle troposphere can still
14                   affect surface concentrations through various exchange mechanisms that mix air from the
15                   free troposphere with air in the PEL.

16                   Several instances of STE producing high concentrations of O3  around Denver and
17                   Boulder, CO were analyzed by Langford et al.  (2009) and several likely instances of
18                   STE, including one of the cases analyzed by Langford et al. (2009) were also cited in the
                         '        O                   J     J    O         '	/
19                   2006 O3 AQCD (U.S. EPA. 2006b) (Annex AX23, Section AX3.9). Clear examples of
20                   STE have also been observed  in southern Quebec province by Hocking et al. (2007). in
21                   accord with previous estimates by Wernli et al. (2002) and James et al. (2003). As also
22                   noted in the 2006 O3 AQCD,  the identification of stratospheric O3, let alone the
23                   calculation of its contributions, is highly problematic and requires data for other tracers.
                     3.4.2.2    Contributions from Other Natural Sources

24                   Biomass burning consists of wildfires and the intentional burning of vegetation to clear
25                   new land for agriculture and for population resettlement; to control the growth of
26                   unwanted plants on pasture land; to manage forest resources with prescribed burning; to
27                   dispose of agricultural and domestic waste; and as fuel for cooking, heating, and water
28                   sterilization. Globally, most wildfires may be ignited directly as the result of human
29                   activities, leaving only 10-30% initiated by lightning (Andreae. 1991). However, because
30                   fire management practices suppress natural wildfires, the buildup of fire fuels increases
31                   the susceptibility of forests to more severe but less frequent fires in the future. Thus there
32                   is considerable uncertainty in attributing the fraction of wildfire emissions to human
33                   activities because the emissions from naturally occurring fires that would have been
34                   present in the absence of fire suppression practices are not known. Contributions to NOX,
      Draft - Do Not Cite or Quote                       3-35                                September 2011

-------
 1                  CO and VOCs from wild fires and prescribed fires are considered as precursors to
 2                  background O3 formation.

 3                  Biomass burning also exhibits strong seasonality and interannual variability
 4                  (van der Werf et al.. 2006). with most biomass burned during the local dry season. This is
 5                  true for both prescribed burns and wildfires. Jaffe et al. (2008) examined the effects of
 6                  wildfires on O3 in the western U.S. They found a strong association (R2 = 0.60) between
 7                  O3 measured at various national park and CASTNET sites and area burned within
 8                  surrounding 50*5° and 10°* 10° areas. However, no such association was found when
 9                  considering the surrounding l°x 1° area, reflecting near source consumption of O3 and the
10                  time necessary for photochemical processing of emissions to form O3. Jaffe et al. (2008)
11                  estimate that burning 1 million acres results in an increase of O3 of 2 ppb, on average;
12                  and that O3 increased by 3.5 and 8.8 ppb during mean and maximum fire years. The
13                  unusually warm and dry weather in central Alaska and western Yukon in the summer of
14                  2004, for example, contributed to the burning of 11 million acres there. Subsequent
15                  modeling by Pfister et al. (2005) showed that the CO contribution from these fires in July
16                  2004 was 33.1 (± 5.5) MT that summer, roughly comparable to total U.S. anthropogenic
17                  CO emissions during the same period. These results underscore the importance of
18                  wildfires as a source of important O3 precursors. In addition to emissions from forest
19                  fires in the U.S., emissions from forest fires in other countries can be transported to the
20                  U.S., for example from boreal forest fires in Canada (Mathur. 2008). Siberia (Generoso et
21                  al.. 2007) and tropical forest fires in the Yucatan Peninsula and Central America (Wang
22                  et al.. 2006). These fires have all resulted in notable increases in O3 concentrations in the
23                  U.S.

24                  Estimates of biogenic VOC and CO emissions are made using the BEIS model with data
25                  from the BELD and annual meteorological data. VOC emissions from vegetation were
26                  described in Section 3.2. As noted earlier, NOX is produced by lightning. Kaynak et al.
27                  (2008) found contributions of 2 to 3 ppb background O3 centered mainly over the
28                  southeastern U.S.  during summer. Although total column estimates of lightning-produced
29                  NOX are large compared to anthropogenic NOX during summer, lightning-produced NOX
30                  does not contribute substantially to the NOX burden in the continental boundary layer.
31                  This is because only 2% of NOX production by lightning occurs within the boundary
32                  layer and most occurs in the free troposphere (Fanget al.. 2010). In addition, much of the
33                  NOX produced in the free troposphere is converted to more oxidized N species during
34                  downward transport.
      Draft - Do Not Cite or Quote                      3-36                                September 2011

-------
            3.4.3  Estimating Background Concentrations

 1                  Historically, two approaches to estimating North American background concentrations
 2                  (previously referred to as PRB) have been considered in previous O3 assessments. In the
 3                  1996 and earlier O3 AQCDs, measurements from remote monitoring sites were used. In
 4                  the 2006 O3 AQCD, the use of chemistry-transport models was adopted, because as
 5                  noted in Section 3.9 of the 2006 O3 AQCD (U.S. EPA. 2006b). estimates of background
 6                  concentrations cannot be obtained directly by examining measurements of O3 obtained at
 7                  relatively remote monitoring sites in the U.S. because of the long-range transport from
 8                  anthropogenic source regions within North America. The 2006 O3 AQCD also noted that
 9                  it is impossible to determine sources of O3 without ancillary data that could be used as
10                  tracers of sources or to calculate photochemical production and loss rates. As further
11                  noted by Reid et al. (2008). the use of monitoring data for estimating background
12                  concentrations is essentially limited to the edges of the domain of interest. This is
13                  because background O3 entering from outside North America can only be destroyed over
14                  North America either through chemical reactions or by deposition to the surface. Within
15                  North America, background O3 is only produced by interactions between natural sources
16                  and between North American natural sources and precursors from other continents. The
17                  current definition of North American background implies that only CTMs (see
18                  Section 3.3 for description and associated uncertainties) can be used to estimate the range
19                  of background concentrations. An advantage to using models is that the entire range of
20                  O3 concentrations measured in different environments can be used to evaluate model
21                  performance. In this regard, data from the relatively small number of monitoring sites, at
22                  which large contributions to background are expected, are best used to evaluate model
23                  predictions.

24                  Estimates of North American background concentrations in the 2006 O3 AQCD were
25                  based on output from the GEOS-Chem model (Fiore etal.. 2003). The GEOS-Chem
26                  model estimates indicated that background O3 concentrations in eastern U.S. surface air
27                  are 25 ± 10 ppb (or generally  15-35 ppb) from June through August, based on conditions
28                  for 2001. These values and all subsequent values given for background concentrations
29                  refer to daily 8-h maximum O3 concentrations. Background concentrations decline from
30                  spring to summer. Background O3 concentrations may be higher, especially at high
31                  altitude sites during the spring, due to enhanced contributions from (1) pollution sources
32                  outside North America; and (2) stratospheric O3 exchange. Only one model (GEOS-
33                  Chem. Harvard University. 201 Ob) was documented in the literature for calculating
34                  background O3 concentrations (Fiore etal.. 2003). The simulated monthly mean
35                  concentrations in different quadrants of the U.S. are typically within 5 ppbv of
36                  observations at CASTNET sites, with no significant bias, except in the Southeast in
37                  summer when the model is 8-12 ppbv too high. This bias might be due to excessive

      Draft - Do Not Cite or Quote                       3-37                                September 2011

-------
 1                   background O3 transported in from the Gulf of Mexico and the tropical Atlantic Ocean in
 2                   the model and/or to inaccuracies in emissions inventories within the U.S.

 3                   Although many of the features of the day-to-day variability in O3 at relatively remote
 4                   monitoring sites in the U.S. were simulated reasonably well by Fiore et al. (2003).
 5                   uncertainties in the calculation of the temporal variability of O3 originating from different
 6                   sources on shorter time scales must be recognized. The uncertainties stem in part from an
 7                   underestimate in the seasonal variability in the STE of O3 (Tusco and Logan. 2003). the
 8                   geographical variability of this exchange, and the  variability in the exchange between the
 9                   free troposphere and the PEL in the model. In addition, the relatively coarse spatial
10                   resolution in that version of GEOS-Chem (2°x2.5°) limited the ability to provide separate
11                   estimates for cities located close to each other, and so only regional estimates were
12                   provided for the 2006 O3 AQCD based  on the results of Fiore et al. (2003).

13                   Wang et al. (2009a) recomputed North American  background concentrations for 2001
14                   using GEOS-Chem at higher spatial resolution (l°x 1°) over North America and not only
15                   for afternoon hours but for the  daily maximum 8-h O3 concentration. These GEOS-Chem
16                   calculations represent the latest results documented in the literature. The resulting
17                   background concentrations, 26.3 ± 8.3 ppb for summer, are consistent with those of
18                   26 ± 7 ppb reported by Fiore et al. (2003). suggesting horizontal resolution was not a
19                   significant factor limiting the accuracy of the earlier results. In addition to computing
20                   North American background contributions, Wang et al. (2009a) also computed U.S.
21                   background concentrations (i.e., including anthropogenic contributions from everywhere
22                   outside the U.S., including Canada and Mexico) of 29.6 ± 8.3 ppb with higher
23                   contributions near the Canadian and Mexican borders.

24                   Zhang et al. (In Press) computed North American background, United States background
25                   and natural background (including only contributions from natural sources everywhere in
26                   the world) O3 concentrations using an even finer grid spacing of (0.5°x0.667°) over
27                   North America for 2006 through 2008. For March through August 2006, mean North
28                   American background O3  concentrations of 27 ± 8 ppb at low elevation (< 1,500 m) and
29                   40 ± 7 ppb at high elevation (> 1,500 m) were predicted. These model predicted values
30                   can be compared to the baseline O3 concentrations estimated by Chan and Vet (2010) of
31                   37 ± 9 ppb for the continental eastern U.S., 51 ± 6 ppb for the continental western  U.S.,
32                   44 ± 10 ppb for the  coastal western U.S. from March to May; and 32 ± 2 ppb for the
33                   continental eastern U.S., 25 ± 10 ppb for the continental western U.S. and 39 ± 12  ppb for
34                   the coastal western U.S. from June to August (baseline as defined by Chan and Vet
3 5                   (2010) refers to concentrations at locations that are not likely to be near anthropogenic
36                   sources or to have been affected by anthropogenic emissions  within the past few days).
      Draft - Do Not Cite or Quote                       3-38                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
As noted above, increases in Asian emissions only accounted for an average increase of
between 1 to 2 ppb in background O3 across the U.S. even though Asian emissions have
increased by about 44% from 2001 to 2006. United States background concentrations
(i.e., O3 concentrations based on including Canadian and Mexican emissions as
background contributions) are on average 2 ppb higher than North American background
concentrations, with higher contributions close to the borders. Zhang et al. (In Press) also
investigated the effects of model resolution on the results and found that North American
background concentrations are ~ 4 ppb higher, on average, in the 0.5°x0.667° version
than in the coarser 2°x2.5° version.
                             Spring
                                          Summer
                         15
       Source: Zhang et al. (In Press).
              25
35
45
55
65
75   [ppbv]
      Figure 3-9     North American background ozone concentration in surface air for
                     spring and summer 2006 (top). GEOS-Chem calculated
                     concentrations for the base case, i.e., including all sources in
                     surface air for the U.S., Canada and Mexico for spring and summer
                     of 2006 (bottom).
10
11
12
North American background and base case (calculated including U.S. anthropogenic
sources) O3 concentration in surface air for spring and summer 2006 calculated with
GEOS-Chem by Zhang et al. (In Press) are shown in the upper and lower panels of
     Draft - Do Not Cite or Quote
                             3-39
                                        September 2011

-------
 1                   Figure 3-9. As can be seen from the upper panels, North American background
 2                   concentrations tend to be higher in the West, particularly in the intermountain West and
 3                   in the Southwest than in the East in both spring and summer. North American
 4                   background concentrations tend to be highest in the Southwest during summer, however,
 5                   in large measure due to wildfires. Intercontinental transport and stratospheric intrusions
 6                   are major contributors to the high elevation, intermountain West during spring with
 7                   wildfires becoming more important sources during summer. The base case O3
 8                   concentrations (lower panels) show two broad maxima with highest concentrations
 9                   extending throughout the Southwest, intermountain West and the East in both spring and
10                   summer. These maxima extend over many thousands of kilometers demonstrating that O3
11                   is a regional pollutant. Low-level outflow from the Northeast out over the Atlantic Ocean
12                   and from the Southeast out over the Gulf of Mexico is also apparent.

13                   Lower bounds to North American background  concentrations tend to be higher by several
14                   ppb at high elevations than at low elevations, reflecting the increasing importance of
15                   background sources such as STE and intercontinental transport with altitude. In addition,
16                   background concentrations tend to increase with increasing base model (and measured)
17                   concentrations at higher elevation sites, particularly during spring.

18                   Figure 3-10 shows that when model predicted O3 is > 60  ppb, North American
19                   background concentrations are generally higher in both the higher-elevation West and in
20                   the lower-elevation East compared to their seasonal means. Although results are broadly
21                   consistent with results from earlier coarser resolution versions of GEOS-Chem mentioned
22                   above, there are some differences of note. Concentrations of O3 for both the base case
23                   and the North American background case are higher in the intermountain West than in
24                   earlier versions. Also of note, in many areas in the East, background concentrations tend
25                   to be higher on days when predicted O3 is >60 ppb or at least do not decrease with
26                   increasing O3. This result contrasts somewhat with Fiore et al. (2003) who found that
27                   background concentrations in the East tend to decrease with increasing O3.
      Draft - Do Not Cite or Quote                       3-40                                September 2011

-------
                  Ozone>
                  60 ppbv
       Source: Zhang et al. (In Press).
                                         Spring
                                                      Summer
                                    15     25      35     45     55     65     75   [ppbv]
      Figure 3-10   North American background ozone concentrations calculated when
                      base case ozone is > 60 ppb.
 1
 2
 3
 4
 5
 6
 1
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Figures 3-1 la and b show comparison among observed and base case GEOS-Chem
results and corresponding North American and natural backgrounds in 10 ppb bins as box
plots. Comparisons between GEOS-Chem and measurements at individual CASTNET
sites are shown in Figures 3-49 to 3-55 as supplemental material in Section 3.8. In
general, the modeled mean concentrations agree to within ~ 5 ppb at the majority of sites
(26 out of 28) and the model agrees more closely with observations in the intermountain
West than earlier versions (see Section 3-8 Figures 3-52 to 3-53). Substantial over
predictions are found in Florida but not at other sites in the Southeast (see Figure 3-50 in
Section 3.8). Comparison between results in Wang et al. (2009a) for 2001with data
obtained at the Virgin Islands indicate that the model over-predicts summer mean O3
concentrations there by 10 ppb (28 vs.  18 ppb). The Virgin Islands NP site appears not to
have been affected by U.S. emissions, as was found from the close agreement between
the base case and the PRB case. Wind roses calculated for the Virgin Islands site indicate
that flows affecting this site are predominantly easterly/southeasterly in spring and
summer. The over-predictions at the Virgin Islands site imply that modeled O3 over the
tropical Atlantic Ocean is too  high. As a result, inflow of O3 over Florida and into the
Gulf of Mexico is also likely to be too high as winds are predominantly easterly at these
low latitudes. Similar considerations apply to the results of Zhang  et al. (In Press). The
most likely explanation involves deficits in model chemistry, for example, reactions
involving halogens are not included. It is not yet clear why the model under-predicts
mean O3 at Yosemite (elevation 1,680 m) by ~ 10 ppb (see Figure 3-55 in Section 3.8).
However,  predictions are within a few ppb at an even higher elevation site in California
(Converse Station, elevation 1,837 m) or at the low elevation sites.
      Draft - Do Not Cite or Quote
                               3-41
September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
Figures 3-56 a-b in Section 3.8 show a comparison of GEOS-Chem output with
measurements at Mt. Bachelor, OR from March-August, 2006. In general, mean
concentrations are simulated reasonably well at both coarse and finer grid resolutions
with mean values 2 ppb higher in the finer resolution model. Although the finer
resolution version provides some additional day to day variability, it still does not capture
peak concentrations. Figure 3-57 in Section 3.8 shows a comparison of vertical profiles
(mean ± la) calculated by GEOS-Chem with ozonesondes launched at Trinidad Head
and Boulder, CO. As can be seen from the figure, variability in both model and
measurements increases with altitude, but variability in the model results is much smaller
at high altitudes than seen in the observations. This may be due in large measure to the
inability of grid-point models to capture the fine-scale, layered structure often seen in O3
in the mid and upper troposphere (Rastigejev et al.. 2010; Newell et al.. 1999).
                                       Spring
                                             Summer
                        > 100 rNA background
                            0   20   40   60   80   100      20    40
                                                Observed ozone [ppbv]
                                                                         80   100
       Source: Zhang et al. (In Press).
       Also shown is the 1:1 line and North American background and natural background model statistics for 10-ppbv bins of observed
      ozone concentrations: the minimum, 25th, 50th, 75th percentile, and maximum.

      Figure 3-11 a   Simulated vs. observed daily 8-h max ozone concentrations for
                      spring (March-May) and summer (June-August) 2006 for the
                      ensemble of CASTNET sites in the intermountain West.
      Draft - Do Not Cite or Quote
                              3-42
September 2011

-------
                                  Spring
                                                           Summer
100
 80
 60
 40
 20
 0
100
 80
 60
 40
 20
 0
100
 80
 60
 40
 20
 0
                         Northeast
                         Great Lakes
                        . Southeast
                                                       NA background  .  . -  ;; '•..
                                                       Natural        "
                       0   20   40    60   80   100      20   40   60   80   100
                                          Observed ozone [ppbv]
 Source: Zhang et al. (In Press).
 Also shown is the 1:1 line and North American background and natural background model statistics for 10-ppbvbins of observed
ozone concentrations: the minimum, 25th, 50th, 75th percentile, and  maximum.

Figure 3-11 b  Simulated vs. observed daily 8-h max ozone concentrations  for
                spring (March-May) and summer (June-August) 2006 for the
                ensembles of CASTNET sites in the Northeast U.S., Great Lakes,
                and the Southeast U.S.
1
2
3
4
5
6
1
              The natural background for O3 averages 18 ± 6 ppbv at the low-elevation sites and 27 ± 6
              ppbv at the high-elevation sites. The difference between North American background and
              natural background concentrations reflects contributions from intercontinental pollution
              and anthropogenic methane (given by the difference between values in 2006 and the pre-
              industrial era, or 1,760 ppb and 700 ppb). The difference between the two backgrounds
              averages 9 ppbv at the low-elevations sites and 13 ppbv at sites in the intermountain
              West. The United States background is on average 1-3 ppbv higher than the North
              American background, reflecting anthropogenic sources in Canada and Mexico, with
              little variability except in border regions.
Draft - Do Not Cite or Quote
                                                 3-43
                                                           September 2011

-------
                         0.10
                         0.05
                         0.00
                         0.10
                         0.05
                         0.00
                              Low-altitude sites (< 1.5 km)
                                                              observation
                                                              GEOS-Chem
                                                              NA background  -
                                                              Natural
                                                                   '18
                              High-altitude sites (>1.5 km)
                                                                          0
                                                                          18
                                                                            I
                                   20
                                           40      60
                                           Ozone [ppbv]
                                                           80
                                                                       -Jo
                                                                        100
 Source: Zhang et al. (In Press).
 Model results (red) are compared to observations (black). Also shown are frequency distributions for the North American
background (solid blue) and natural background (dashed green).

Figure 3-12    Frequency distributions of daily 8-h max ozone concentrations in
                March-August 2006 for the ensemble of low-altitude (<1.5 km) and
                high-altitude CASTNET sites in the U.S.
 1
 2
 3
 4
 5
 6
 7

 8
 9
10

11
12
13
              Figure 3-12 shows frequency distributions for measurements at low-altitude and high-
              altitude CASTNET sites, GEOS-Chem results for the base case, North American
              background and the natural background. Most notable is the shift to higher concentrations
              and the narrowing of the concentration distributions for all three simulations and the
              observations in going from low to high altitudes. However, maximum concentrations
              show little if any dependence on altitude, except for the natural background which tends
              to be slightly higher at lower altitudes.

              As noted in Section 3.3, CTMs are subject to uncertainty in model inputs for emissions,
              meteorology, and chemistry. For example, many of the chemical processes described in
              Section 3.2 have not yet been included in GEOS-Chem.

              Another approach to modeling background concentrations involves using a regional CTM
              such as CMAQ or CAMx with boundary conditions taken from a global scale CTM such
              as GEOS-Chem. Mueller and Mallard (201 la), while not calculating North American
Draft - Do Not Cite or Quote
                                                   3-44
September 2011

-------
 1                  background values exactly as defined here, calculated contributions from natural sources
 2                  and inflow from the boundaries to O3 for 2002 using MM5 and CMAQ for the outermost
 3                  domain (36 km resolution) shown in Figure 3-4 with boundary conditions from GEOS-
 4                  Chem. The overall bias based on comparison with AQS monitors for the base case is
 5                  about 3 ppb; the annual mean fractional bias and mean fractional error were 7% and 21%
 6                  for the O3 season across the U.S. Note that Figure 2 in their paper is mislabeled, as it
 7                  should refer to the case with total emissions - not to natural emissions in North America
 8                  only (Mueller and Mallard. 201 Ib). However, boundary conditions are fixed according to
 9                  monthly averages based on an earlier version of GEOS-Chem and do not reflect shorter
10                  term variability or trends in Northern Hemispheric emissions of pollution. In addition,
11                  fluxes of O3 from the stratosphere are not defined. Note that their natural background
12                  includes North American natural background emissions only and influence from
13                  boundary conditions and thus is not a global natural background. Calculated values
14                  including natural emissions from North America and from fluxes through the boundaries
15                  are somewhat larger than given in Zhang et al. (In Press), in large measure because of
16                  much larger contributions from wildfires and lightning. Wildfire contributions reach
17                  values of- 140 ppb in Redwoods National Park and higher elsewhere in the U.S. and in
18                  Quebec. However as noted by Singh et al. (201 Ob) significant enhancements of O3 in
19                  California fire plumes are found only when mixed with urban pollution. Lightning
20                  contributions (ranging up to ~ 30 ppb) are substantially larger than estimated by Kaynak
21                  et al. (2008) (see Section 3.4.2.1). The reasons for much larger contributions from
22                  wildfires and lightning are not clear and need to be investigated further.
            3.4.4  Summary of Background Results

23                  In general, the GEOS-Chem predictions tend to show smaller disagreement with
24                  observations at the high-altitude sites than at the low-altitude sites. Overall agreement
25                  between model results for the base case and measurements is within a few ppb for spring-
26                  summer means in the Northeast (see Figure  3-49 in Section 3.8) and the Southeast (see
27                  Figure 3-50 in Section 3.8), except in and around Florida where the base case over
28                  predicts O3 by 10 ppb at one site, at least. In the Upper Midwest (see Figure 3-51 in
29                  Section 3.8), the model predictions are within 5 ppb of measurements, the same is true for
30                  sites in the intermountain West (see Figures 3-52 and 3-53) and at lower elevations sites
31                  in the West (see Figure 3-54 in Section 3.8) including California (see Figure 3-55 in
32                  Section 3.8). However, the model under predicts O3 by 10 ppb  at the Yosemite site.
33                  These results suggest that the model is capable of calculating March to August mean O3
34                  to within ~ 5 ppb at most (26 out of 28) sites chosen. Currently, there are no simulations
35                  of North American background concentrations available in the  literature apart from those
      Draft - Do Not Cite or Quote                      3-45                                September 2011

-------
 1                  using GEOS-Chem alone. However, as noted in the 2006 O3 AQCD, an ensemble
 2                  approach as is done in many other applications of atmospheric models is to be preferred.

 3                  The GEOS-Chem calculations presented here represent the latest results documented in
 4                  the literature. However, all models undergo continuous updating of inputs,
 5                  parameterizations of physical and chemical processes, and inputs and improvements in
 6                  model resolution. Inputs that might be considered most relevant include emissions
 7                  inventories, chemical reactions and meteorological fields. This leads to uncertainty in
 8                  model predictions in part because there is typically a lag between updated information for
 9                  these above inputs, as outlined in Section 3.2 for chemical processes and emissions and in
10                  Section 3.3 for model construction, and their implementation in  CTMs including GEOS-
11                  Chem. Examples might include updated emissions for year specific shipping, wildfires
12                  and updates to the 2005 NEI; updates to the chemistry of isoprene and multi-phase
13                  processes, including those affecting the abundance of halogens;  and updates to species
14                  such as methane. To the extent that results from an updated model become available, they
15                  will be presented and used to help inform NAAQS setting.

16                  Supplemental material given in Section 3.9 summarizes results of modeling work using
17                  GEOS-Chem that is still in progress. Results for the current definition of North American
18                  background, U.S. background and natural background are given for January 2006 to
19                  December 2008. Major differences from the work of Zhang et al. (In Press) include the
20                  use of a later model version which incorporates updates to the chemistry of isoprene
21                  nitrates and to the generation of lightning NOX. In addition, anthropogenic emissions
22                  were updated for each model year from the NEI 2005 inventory. The complete draft
23                  report is available on-line (U.S. EPA. 201 lc).
          3.5    Monitoring
            3.5.1   Routine Monitoring Techniques

24                  The FRM for O3 measurement is called the Chemiluminescence Method (CLM) and is
25                  based on the detection of chemiluminescence resulting from the reaction of O3 with
26                  ethylene gas. The UV absorption photometric analyzers were approved as FEMs in 1977
27                  and gained rapid acceptance for NAAQS compliance purposes due to ease of operation,
28                  relatively low cost, and reliability. The UV absorption method  is based on the principle
29                  that O3 molecules absorb UV radiation at a wavelength of 254  nm from a mercury lamp.
30                  The concentration of O3 is computed from Beer's law using the radiation absorbed across
31                  a fixed path length, the absorption coefficient, and the measured pressure and temperature
      Draft - Do Not Cite or Quote                       3-46                                September 2011

-------
 1                  in the detection cell. UV absorption photometry is the predominant method for assessing
 2                  compliance with the NAAQS for O3. Almost all of the state and local air monitoring
 3                  stations (SLAMS) that reported data to EPA AQS from 2005 to 2009 used UV absorption
 4                  photometer FEMs. No CLM monitors, approved as FRMs or FEMs, reported O3 data to
 5                  AQS from 2005 to 2009 and only one monitor reported data using a long-path or open
 6                  path Differential Optical Absorption Spectrometer (DOAS) FEM during this period.

 7                  The rationale, history, and calibration of O3 measurements were summarized in the 1996
 8                  O3 AQCD and the 2006 O3 AQCD and focused on the state of ambient O3 measurements
 9                  at that time as well as evaluation of interferences and new developments. This discussion
10                  will continue with the current state of O3 measurements, interferences, and new
11                  developments for the period 2005 to 2010.

12                  UV O3 monitors use mercury lamps as the source of UV radiation and employ an O3
13                  scrubber (typically manganese  dioxide) to generate an ozone-free air flow to serve as a
14                  reference channel for O3 measurements. There are known interferences with UV O3
15                  monitors. The 2006 O3 AQCD reported on the investigation of the effects of water vapor,
16                  aromatic compounds,  ambient particles, mercury vapor and alternative materials in the
17                  instrument's O3 scrubber. The  overall conclusions from the 2006 O3 AQCD review of
18                  the  scientific literature are briefly summarized below.

19                  Kleindienst et al. (1993) found water vapor to have no significant impact and aromatic
20                  compounds to have a minor impact (as much as 3% higher than the FRM extrapolated to
21                  ambient conditions) on UV absorption measurements. UV O3 monitor response evaluated
22                  by chamber testing using cigarette smoke, reported an elimination of the O3 monitor
23                  response to the smoke when a particle filter was used that filtered out particles less than
24                  0.2  (im in diameter (Arshinov et al.. 2002). One study (Leston et al.. 2005) in
25                  Mexico City compared a UV O3 FEM to a CLM FRM. The UV FEM commonly reported
26                  consistently higher O3 than the CLM FRM. The typical difference was 20 ppb with a
27                  range up to 50 ppb. Leston et al. (2005) also presented smog chamber data which
28                  demonstrated that heated metal and heated silver wool scrubbers perform better in the
29                  presence of aromatic hydrocarbon irradiations than manganese dioxide scrubbers when
30                  compared to the FRM. They also suggested the use of humidified calibration gas and
31                  alternative scrubber materials to improve UV O3 measurements. Some O3  monitor
32                  manufacturers now offer heated silver wool scrubbers as an alternative to manganese
33                  dioxide. Another possible solution to the O3  scrubber problem may be the use of a gas
34                  phase scrubber such as NO. A commercial version of this has recently been introduced by
35                  2B Technologies as an option on their model 202 FEM; however, it has not been field
36                  tested or approved for use as an FEM.
      Draft - Do Not Cite or Quote                       3-47                                September 2011

-------
 1                   Review of the recent literature is summarized below. Study of UV monitors by Williams
 2                   et al. (2006) concluded that well maintained monitors showed no significant interferences
 3                   when operated in locations with significant concentrations of potentially interfering
 4                   VOCs including Nashville, Houston, and the Gulf of Maine. Monitors were tested in
 5                   urban and suburban environments, as well as on board a ship in both polluted and clean
 6                   marine air. Comparisons of UV measurements to a non-FRM/FEM NO based CLM
 7                   demonstrated agreement to within 1%. At the Houston location, they did observe a brief
 8                   period on one day for about 30 minutes where the UV measurements exceeded the  CLM
 9                   by about 8 ppb (max). This was attributed to probable instrument malfunction.

10                   Wilson and Birks (2006) investigated water vapor interference in O3 measurements by
11                   four different UV monitors. In extreme cases where a rapid step change in relative
12                   humidity between 0 and 90% was presented, large transitory responses (tens to hundreds
13                   of ppb) were found for all monitors tested. Rapid changes in relative humidity such as
14                   this would not be expected during typical ambient O3 measurements and could only be
15                   expected during  measurement of vertical profiles from balloon or aircraft. The magnitude
16                   of the interference and the direction (positive or negative) was dependent on the
17                   manufacturer and model. Wilson and Birks (2006) also hypothesized that water vapor
18                   interference is caused by physical interactions of water vapor on the detection cell.  The
19                   O3 scrubber was also thought to act as a reservoir for water vapor and either added or
20                   removed water vapor from the air stream, subsequently affecting the detector signal and
21                   producing either a positive or negative response. They demonstrated that the use of a
22                   Nafion permeation membrane just before the O3 detection cell to remove water vapor
23                   eliminated this interference.

24                   Dunlea et al. (2006) evaluated multiple UV O3 monitors with two different O3 scrubber
25                   types (manganese dioxide and heated metal wool) in Mexico City. Large spikes in O3
26                   concentrations were observed while measuring diesel exhaust where large increases in
27                   particle number  density were observed. The interference due to small particles passing
28                   through the Teflon filter and scattering/absorbing light in the detection cell were
29                   estimated to cause at most a 3% increase in measurements in typical ambient air
30                   environments. This estimate pertains to measurements in the immediate vicinity of fresh
31                   diesel emissions and most monitor siting guidelines would not place the monitor close to
32                   such sources, so  actual interferences are expected to be much less than 3%. Dunlea et al.
33                   (2006) also observed no evidence for either a positive or negative interference or
34                   dependence due  to variations in aromatics during their field study.

35                   Li et al. (2006c)  verified early reports of gas phase mercury interference with the UV O3
36                   measurement. They found that 300 ng/m3 of mercury produced an instrument response of
      Draft - Do Not Cite or Quote                       3-48                                September 2011

-------
 1                  about 35 ppb O3. Background concentrations of mercury are around 1-2 ng/m3 and
 2                  expected to produce an O3 response that would be <1 ppb.

 3                  Spicer et al. (2010) examined potential UV O3 monitor interferences by water vapor,
 4                  mercury, aromatic compounds, and reaction products from smog chamber simulations.
 5                  Laboratory tests showed little effect of changing humidity on conventional FEM UV O3
 6                  monitors with manganese dioxide or heated metal wool scrubbers in the absence of other
 7                  interferences. Mercury vapor testing produced an O3 response by the UV monitors that
 8                  was <1 ppb O3 per 1 ppt (about 8 ng/m3) mercury vapor. Interference by aromatic
 9                  compounds at low (3% RH) and high (80% RH) humidity showed some positive
10                  responses that varied by UV monitor and ranged from 0 to 2.2 ppb apparent O3 response,
11                  per ppb  of aromatic compound tested. The authors acknowledged that the aromatic
12                  compounds most likely to interfere are rarely measured in the atmosphere and therefore,
13                  make it  difficult to assess the impact of these compounds during ambient air monitoring.
14                  Comparison of UV and CLM responses to photochemical reaction products in smog
15                  chamber simulations at 74 to 85% RH showed varied responses under low
16                  (0.125 ppmv/0.06 ppmv) to high (0.50 ppmv/0.19 ppmv) hydrocarbon/NOx conditions.
17                  The conventional UV monitors were as much as 2 ppb higher than the CLM under low
18                  hydrocarbon/NOx conditions and 6 ppb higher under the high hydrocarbon/NOx
19                  conditions. Two FEM UV monitors were also co-located at six sites in Houston from
20                  May to October, 2007 with one UV monitor equipped with Nafion permeation
21                  membrane. The average difference between 8-h daily max O3 concentrations using the
22                  UV and the UV with Nafion permeation membrane ranged from -4.0 to 4.1 ppb.
            3.5.2   Precision and Bias

23                  In order to provide decision makers with an assessment of data quality, EPA's Quality
24                  Assurance (QA) group derives estimates of both precision and bias for O3 and the other
25                  gaseous criteria pollutants from the biweekly single point quality control (QC) checks
26                  using calibration gas, performed at each site by the monitoring agency. The single-point
27                  QC checks are typically performed at concentrations around 90 ppb. Annual summary
28                  reports of precision and bias can be obtained for each monitoring site at
29                  http://www.epa.gov/ttn/amtic/qareport.html. The assessment of precision and bias are
30                  based on the percent-difference values, calculated from single-point QC checks. The
31                  percent difference is based on the difference between the pollutant concentration
32                  indicated by monitoring equipment and the known (actual) concentration of the standard
33                  used during the QC check. The monitor precision is estimated from the 90% upper
34                  confidence limit of the coefficient of variation (CV) of relative percent difference (RPD)
35                  values. The bias is estimated from the 95% upper confidence limit on the mean of the


      Draft - Do Not Cite or Quote                       3-49                               September 2011

-------
1
2
3
4
5
absolute values of percent differences. The data quality goal for O3 precision and bias at
the 90 and 95% upper confidence limits is 7% (40 CFR Part 58, Appendix A). Table 3-1
presents a summary of the number of monitors that meet the precision and bias goal of
7% for 2005 to 2009. Greater than 96% of O3 monitors met the precision and bias goal
between 2005 and 2009.
     Table 3-1      Summary of ozone monitors meeting 40 CFR Part 58, Appendix A
                    Precision and Bias Goals
Year
2005
2006
2007
2008
2009
Number of Monitors
879
881
935
955
958
Monitors with Acceptable
Precision (%)
96.5
98.1
98.1
97.1
97.4
Monitors with Acceptable
Bias (%)
96.7
97.6
98.1
96.7
97.5
6
7
Another way to look at the precision (CV) and bias (percent difference) information
using the single-point QC check data from the monitoring network is to present box plots
of the monitors' individual precision and percent-difference data; Figure 3-13 and
Figure 3-14 include this information for O3 monitors operating from 2005 to 2009.
                                                                    25"' percent! I e
                                                                    10s'1 percent! I e
                       2005
                       N=1151
                   2006
                  N=1159
2007
N=1166
2008
N=1178
2009
N=1158
     Figure 3-13   Box plots of precision data by year (2005-2009) for all ozone
                    monitors reporting single-point QC check data to AQS.
     Draft - Do Not Cite or Quote
                             3-50
                               September 2011

-------
                     C
                     0!
                    I
                                                                           Legend
                                                                     9Dll! percent] le
                                                                     75th percent) le
                                                                         Mean
                                                                        Median

                                                                     25th percent) le
                                                                     10lil percent! le
                          2005
                         N=52724
                    2006
                   N=51814
 2007
N=53262
 2008
N=57315
 2009
N=67305
      Figure 3-14    Box plots of percent-difference data by year (2005-2009) for all
                      ozone monitors reporting single-point QC check data to AQS.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
3.5.2.1    Precision from Co-located UV Ozone Monitors in Missouri

The Missouri Department of Natural Resources (MODNR) maintains a network of co-
located UV O3 analyzers. The MODNR provided co-located data from four monitors:
two co-located at the same monitoring site in Kansas City (AQS ID 290370003) and two
co-located at the same monitoring site in St. Louis (AQS ID 291831002). Hourly
observations for the co-located measurements at these two sites between April and
October, 2006-2009 were used to evaluate precision from co-located UV monitors. These
data were then compared with the precision obtained by the biweekly single point QC
checks for all sites reporting single-point QC check data to AQS between 2005 and 2009;
the method normally used for assessing precision. Box plots of the RPD between the
primary and co-located hourly O3 measurements in Missouri are shown in Figure 3-15
and box plots of the RPD between the actual and indicated QC check for all U.S. sites are
shown in Figure 3-16. As mentioned above, the average concentration of the single-point
QC check is 90 ppb, whereas the average ambient O3 concentration measured at the two
sites in Missouri was 34  ppb. The mean RPD for the co-located monitors in Missouri and
the single-point QC check data from all sites were less than 1 percent.
      Draft - Do Not Cite or Quote
                              3-51
                                September 2011

-------
           1
           b

           §  0
           a.

           fo  -i
                                                                90°'percentile

                                                                75lhpercentile

                                                                    Mean

                                                                   Median

                                                                25lh percentjle

                                                                1QU> percentile
                 2006
                N=10017
      2007
     N=10133
          2008
          N=9884
               2009
              N=10211
Figure 3-15    Box plots of RPD data by year for the co-located ozone monitors at
                two sites in Missouri from 2006-2009.
           | -i
           S.
            £  1 i
                                                                 90lli percent! I e

                                                                 75th percent! I e
                                                                    Mean

                                                                   Median

                                                                 25"'percent) I e

                                                                 1Q11'percent) I e
                 I
               I
               I
               _L
                 2005
                N=52724
 2006
N=51814
 2007
N=53262
 2008
N=57315
 2009
N=67305
Figure 3-16    Box plots of RPD data by year for all U.S. ozone sites reporting
                single-point QC check data to AQS from 2005-2009.
Draft - Do Not Cite or Quote
             3-52
                                September 2011

-------
            3.5.3   Performance Specifications

 1                  The performance specifications for evaluating and approving new FEMs in accordance
 2                  with 40 CFR Part 53 are provided in Table 3-2. These specifications were developed and
 3                  originally published in the Federal Register in 1975. Modern, commercially-available
 4                  instruments can now perform much better than the requirements specified below. For
 5                  example, the lower detectable limit (LDL) performance specification is 10 ppb and the
 6                  typical vendor-stated performance for the LDL is now less than 0.60 ppb. The amount of
 7                  allowable interference equivalent for total interference substances is 60 ppb, and the
 8                  current NAAQS for O3 is 75 ppb, with an averaging time of 8 hours. Improvements in
 9                  new measurement technology have occurred since these performance specifications were
10                  originally developed. These specifications should be revised to more accurately reflect
11                  the necessary performance requirements for O3 monitors used to support the current
12                  NAAQS.
      Table 3-2       Performance specifications for ozone based in 40 CFR Part 53
Parameter
Range
Noise
LDL - defined as two times the noise
Specification
0 - 0.5 ppm (500 ppb)
0.005 ppm (5 ppb)
0.01 ppm (10 ppb)
Interference equivalent
Each interfering substance
Total interfering substances
±0.02 ppm (20 ppb)
0.06 ppm (60 ppb)
Zero drift
12h
24 h
±0.02 ppm (20 ppb)
±0.02 ppm (20 ppb)
Span Drift, 24 h
20% of upper range limit
80% of upper range limit
Lag time
Rise time
Fall time
± 20.0%
±5.0%
20min
15min
15min
Precision
20% of upper range limit
80% of upper range limit
0.01 ppm (10 ppb)
0.01 ppm (10 ppb)

            3.5.4   Monitor Calibration

13                  The calibration of O3 monitors was summarized in detail in the 1996 O3 AQCD. The
14                  calibration of O3 monitors is done using an O3 generator and UV photometers. UV
15                  photometry is the prescribed procedure for the calibration of reference methods to


      Draft - Do Not Cite or Quote                      3-53                               September 2011

-------
 1                  measure O3 in the atmosphere. Because O3 is unstable and cannot be stored, the O3
 2                  calibration procedure specifically allows the use of transfer standards for calibrating
 3                  ambient O3 monitors. A transfer standard is calibrated against a standard of high
 4                  authority and traceability and then moved to another location for calibration of O3
 5                  monitors. The EPA and the National Institute of Standards and Technology (NIST) have
 6                  established a network of standard reference photometers (SRPs) that are used to verify
 7                  transfer standards. The International Bureau of Weights and Measures (BIPM) maintain
 8                  one NIST SRP (SRP27) as the World's O3  reference standard. NIST maintains two SRPs
 9                  (SRPO and SRP2) that are used for comparability to ten other SRPs maintained by the
10                  EPA's Regional QA staff.

11                  SRPs have been compared to other reference standards. Tanimoto et al. (2006) compared
12                  NIST SRPS 5, owned by the National Institute for Environmental Studies in Japan, to gas
13                  phase titration (GPT). The SRP was found to be 2% lower than GPT. GPT is no longer
14                  used as a primary or transfer standard in the U.S. Viallon et al. (2006) compared SRP27
15                  built at BIPM to four other NIST SRPs maintained by BIPM (SRP28, SRPS 1, SRP32,
16                  and SRP33). A minimum bias of+0.5% was found for all SRP measurement results, due
17                  to use of the direct cell length measurement for the optical path length; this bias was
18                  accounted for by applying the appropriate correction factor. Study of the bias-corrected
19                  SRPs showed systematic biases and measurement uncertainties for the BIPM SRPs. A
20                  bias of -0.4% in the instrument O3 mole fraction measurement was identified and
21                  attributed to non-uniformity of the gas temperature in the instrument gas cells, which was
22                  compensated by a bias of+0.5% due to an under-evaluation of the UV light path length
23                  in the gas cells. The relative uncertainty of the O3 absorption cross section was 2.1% at
24                  253.65 nm and this was proposed as an internationally accepted consensus value until
25                  sufficient experimental data is available to assign a new value.

26                  In November, 2010, the EPA revised the Technical Assistance Document for Transfer
27                  Standards for Calibration of Air Monitoring Analyzers for Ozone (201 Of) that was first
28                  finalized in 1979 (U.S. EPA.  1979b). The revision removed methods no longer in use and
29                  updated definitions and procedures where appropriate. In the revised document, the
30                  discussion of transfer standards for O3 applies to the family of standards that are used
31                  beyond SRPs or Level 1 standards. To reduce confusion, EPA reduced the number of
32                  common terms that were used in the past such as: primary standard,  local primary
33                  standard, transfer standard, and working standard. Beyond the SRPs, all other standards
34                  are considered transfer standards.
      Draft - Do Not Cite or Quote                      3-54                                September 2011

-------
            3.5.5   Other Monitoring Techniques
                    3.5.5.1    Portable UV Ozone Monitors

 1                  Small, lightweight, and portable UV O3 monitors with low power consumption are
 2                  commercially available. These monitors are based on the same principle of UV
 3                  absorption by O3 at 254 nm. Monitors of this type are typically used for vertical profiling
 4                  using balloons, kites, or light aircraft where space and weight are limited. They have also
 5                  been used for monitoring at remote locations such as National Parks. Burley and Ray
 6                  (2007) compared portable O3 monitor measurements to those from a conventional UV
 7                  monitor in Yosemite National Park. Calibrations of the portable O3 monitors against a
 8                  transfer standard resulted in an overall precision of ± 4 ppb and accuracy of ± 6%. Field
 9                  measurement comparisons between the portable and conventional monitor at Turtleback
10                  Dome showed the portable monitor to be 3.4 ppb lower on average, with daytime
11                  deviation typically on the order of 0-3 ppb. Agreement between the portable and
12                  conventional monitor during daylight hours (9:00 a.m. to 5:00 p.m. PST) resulted in an
13                  R2 of 0.95, slope of 0.95, and intercept of 0.36 ppb. Significant deviations were observed
14                  in the predawn hours where the portable monitor was consistently low. These deviations
15                  were attributed to the  difference in sampling inlet location. The portable monitor was
16                  located at 1.3 m above ground and the conventional monitor was located at 10 m above
17                  ground.  Agreement between the portable and conventional monitors for all hours sampled
18                  resulted in an R2 of 0.88, slope of 1.06, and intercept of -6.8 ppb. Greenberg et al. (2009)
19                  also compared a portable UV O3 monitor to a conventional UV monitor in Mexico City
20                  and obtained good agreement for a 14 day period with an R2 of 0.97, slope of 0.97, and
21                  intercept of 6 ppb. One portable O3 monitor was recently approved as  an FEM (EQOA-
22                  0410-190) on April 27, 2010 (75 FR 22126).
                    3.5.5.2    NO-based Chemiluminescence Monitors

23                  One commercially available NO-based chemiluminescence monitor is currently
24                  undergoing FEM testing (Teledyne Advanced Pollution Instrumentation, Douglassville,
25                  GA). It may also be designated as a second or replacement FRM since the ethene based
26                  FRMs are no longer manufactured. Although this is a relatively new monitor, other NO-
27                  based CLM instruments have been custom built for various field studies since the early
28                  1970s. A commercial version that measured both O3 and NOX was offered in the early
29                  1970s but failed to gain commercial acceptance. Initial testing with SO2, NO2, C12,
30                  C2H2, C2H4 and C3H6 (Stedman et al.. 1972) failed to identify any interferences. In the
31                  intervening years, custom built versions have not been found to have any interference;

      Draft - Do Not Cite or Quote                       3-55                               September 2011

-------
 1                  however, they do experience a slight decrease in response with increasing relative
 2                  humidity (due to quenching of the excited species by the water molecules). The new NO-
 3                  based CLM solves this problem with the use of a Nafion membrane dryer. A custom built
 4                  NO-based CLM similar to the monitor undergoing FEM testing was used by Williams et
 5                  al. (2006) in Houston, TX; Nashville, TN; and aboard ship along the New England coast.
 6                  It was found to be in good agreement with a standard UV based FEM and with a custom
 7                  built DOAS.
                    3.5.5.3   Passive Air Sampling Devices and Sensors

 8                  A passive O3 sampling device depends on the diffusion of O3 in air to a collecting or
 9                  indicating medium. In general, passive samplers are not adequate for compliance
10                  monitoring because of the limitations in averaging time (typically one week or more),
11                  particularly for O3. However, these devices are valuable for personal human exposure
12                  estimates and for obtaining long-term data in rural areas where conventional UV
13                  monitors are not practical or feasible to deploy. The 1996 O3 AQCD provided a detailed
14                  discussion of passive samplers, along with the limitations and uncertainties of the
15                  samplers evaluated and published in the literature from 1989 to 1995. The 2006  O3
16                  AQCD provided a brief update on available passive samplers developed for use  in direct
17                  measurements of personal exposure published through 2004. The 2006 O3 AQCD also
18                  noted the sensitivity of these samplers to wind velocity, badge placement, and
19                  interference by other co-pollutants that may result in measurement error.

20                  Subsequent evaluations of passive diffusion samplers in Europe showed good correlation
21                  when compared to conventional UV O3 monitors, but a tendency for the diffusion
22                  samplers to overestimate the O3 concentration (Gottardini et al.. 2010; Vardoulakis et al..
23                  2009; Buzica et al., 2008). The bias of O3 diffusion tubes were also found to vary with
24                  concentration, season, and exposure duration (Vardoulakis et al.. 2009). Development of
25                  simple,  inexpensive, passive O3 measurement devices that rely on O3 detection papers
26                  and a variety of sensors with increased time resolution (sampling for hours instead of
27                  weeks) and improved sensitivity have been reported (Maruo etal.. 2010; Ebeling et al..
28                  2009; Miwa etal.. 2009; Ohira etal.. 2009; Maruo. 2007; O-Keeffe et al.. 2007; Utembe
29                  et al.. 2006). Limitations for some of these sensors and detection papers include  air flow
30                  dependence and relative humidity interference.
      Draft - Do Not Cite or Quote                      3-56                                September 2011

-------
                    3.5.5.4    Differential Optical Absorption Spectrometry

 1                  Optical remote sensing methods can provide direct, sensitive, and specific measurements
 2                  of O3 over a broad area or open path in contrast with conventional single-point UV
 3                  monitors. The 1996 O3 AQCD provided a brief discussion of DOAS for O3
 4                  measurements and cited references to document the sensitivity (1.5 ppb for a 1-minute
 5                  averaging time), correlation (r = 0.89), and agreement (on the order of 10%) with UV O3
 6                  monitors (Stevens etal.. 1993). The 2006 O3 AQCD provided an update on DOAS where
 7                  a positive interference due to an unidentified absorber was noted (Reisinger. 2000).

 8                  More recent study of the accuracy of UV absorbance monitors by Williams et al. (2006)
 9                  compared UV and DOAS measurements at two urban locations. In order to compare the
10                  open path measurements and UV, the data sets were averaged to 30-minute periods and
11                  only data when the boundary layer was expected to be well mixed (between 10:00 a.m.
12                  and 6:00 p.m. CST) were evaluated. The comparisons showed variations of no more than
13                  ± 7% (based on the slope of the linear least squares regression over a concentration range
14                  from about 20 to 200 ppb) and good correlation (R2 = 0.96 and 0.98). Lee et al. (2008b)
15                  evaluated DOAS and UV O3 measurements in Korea and found the average DOAS
16                  concentration to be 8.6% lower than the UV point measurements with a good correlation
17                  (R2 = 0.94).

18                  DOAS has also been used for the measurement of HNO2 (or HONO). DOAS was
19                  compared to chemical point-measurement methods for HONO. Acker et al. (2006)
20                  obtained good results when comparing wet chemical and DOAS during well mixed
21                  atmospheric conditions (wet chemical = 0.009 + 0.92 x DOAS; r = 0.7).  Kleffmann and
22                  Wiesen (2008) noted that interferences with the HONO wet chemical methods can affect
23                  results from inter-comparison studies if not addressed. In an earlier study, Kleffman et al.
24                  (2006) demonstrated that when the interferences were addressed, excellent agreement
                    v	/                                                  ~         O
25                  with DOAS can be obtained. Stutz et al. (2009) found good agreement (15% or better)
26                  between DOAS and a wet chemical method (Mist Chamber/Ion Chromatography) in
27                  Houston, TX except generally during mid-day when the chemical method showed a
28                  positive bias that may have been related to concentrations of O3. DOAS  remains
29                  attractive due to its sensitivity, speed of response, and ability to simultaneously measure
30                  multiple pollutants; however, further inter-comparisons and interference testing are
31                  recommended.
      Draft - Do Not Cite or Quote                      3-57                               September 2011

-------
                    3.5.5.5   Satellite Remote Sensing

 1                  Satellite observations for O3 are growing as a resource for many purposes, including
 2                  model evaluation, assessing emissions reductions, pollutant transport, and air quality
 3                  management. Satellite remote sensing instruments do not directly measure the
 4                  composition of the atmosphere. Satellite retrievals are conducted using the solar
 5                  backscatter or thermal infrared emission spectra and a variety of algorithms. Most
 6                  satellite measurement systems have been developed for stratospheric measurement of the
 7                  total O3 column. Mathematical techniques have been developed and must be applied to
 8                  derive information from these systems about tropospheric O3 (Tarasick and  Slater. 2008;
 9                  Ziemke etal.. 2006). Direct retrieval of global tropospheric O3 distributions from solar
10                  backscattered UV spectra have been reported from the Ozone Monitoring Instrument
11                  (OMI) and Global Ozone Monitoring Experiment (GOME) (Liu et al.. 2006). Another
12                  satellite measurement system, Tropospheric Emission Spectrometer (TES), produces
13                  global-scale vertical concentration profiles of tropospheric O3 from measurements  of
14                  thermal infrared emissions. TES has been designed specifically to focus on mapping the
15                  global distribution of tropospheric O3 extending from the surface to about 10-15 km
16                  altitude (Beer. 2006).

17                  In order to improve the understanding of the quality and reliability of the data, satellite-
18                  based observations of total column and tropospheric O3 have been validated in several
19                  studies using a variety of techniques, such as aircraft observations, ozonesondes, CTMs,
20                  and ground-based spectroradiometers. Anton et al. (2009) compared satellite data from
21                  two different algorithms (OMI-DOAS and OMI-TOMS) with total column O3 data from
22                  ground-based spectroradiometers at five locations. The satellite total column O3 data
23                  underestimated ground-based measurements by less than 3%. Richards et al. (2008)
24                  compared TES tropospheric O3 profiles using airborne differential absorption lidar
25                  (DIAL) and found TES to have a 7 ppbv positive bias relative to DIAL throughout the
26                  troposphere. Nasser et al. (2008)  compared TES O3 profiles and ozonesonde
27                  coincidences and found a positive bias of 3-10 ppbv for TES. Worden et al. (2007a) also
28                  compared TES with ozonesondes and found TES O3 profiles to be biased high in the
29                  upper troposphere (average bias of 16.8 ppbv for mid-latitudes and 9.8 ppbv for the
30                  tropics) and biased low in the lower troposphere (average bias of -2.6 ppbv for mid-
31                  latitudes and -7.4 ppbv for the tropics). Comparisons of TES and OMI with ozonesondes
32                  by Zhang  et al. (201 Ob) showed a mean positive bias if 5.3 ppbv (10%)  for TES and
33                  2.8 ppbv (5%) for OMI at 500 hPa. In addition, Zhang et al. (201 Ob) used a CTM
34                  (GEOS-Chem) to determine global differences between TES and OMI.  They found
35                  differences between TES and OMI were generally ±10 ppbv except at northern mid-
36                  latitudes in summer and over tropical continents. Satellite observations have also been
      Draft - Do Not Cite or Quote                      3-58                                September 2011

-------
 1                   combined (e.g., OMI and TES) to improve estimates of tropospheric O3 (Worden et al.,
 2                   2QQ7_b).
             3.5.6   Ambient Ozone Network Design
                     3.5.6.1    Monitor Siting Requirements

 3                   To monitor compliance with the NAAQS, state and local monitoring agencies operate O3
 4                   monitoring sites at various locations depending on the area size (population and
 5                   geographic characteristics2) and typical peak concentrations (expressed in percentages
 6                   below, or near the O3 NAAQS). SLAMS make up the ambient air quality monitoring
 7                   sites that are primarily needed for NAAQS comparisons, but may also serve some other
 8                   basic monitoring objectives that include: providing air pollution data to the general public
 9                   in a timely manner; emissions strategy development; and support for air pollution
10                   research. SLAMS include National Core (NCore), Photochemical Assessment
11                   Monitoring Stations (PAMS), and all other State or locally-operated stations except for
12                   the monitors designated as special purpose monitors (SPMs).

13                   The SLAMS minimum monitoring requirements to meet the O3  design criteria are
14                   specified in 40 CFR Part 58, Appendix D. Although NCore and PAMS are a subset of
15                   SLAMS, the monitoring requirements for those networks are separate and discussed
16                   below. The minimum number of O3 monitors required in a Metropolitan Statistical Area
17                   (MSA) ranges from zero for areas with a population of at least 50,000 and under 350,000
18                   with no recent history of an O3 design value3 greater than 85 percent of the NAAQS, to
19                   four for areas  with a population greater than 10 million and an O3 design value greater
20                   than 85 percent of the NAAQS. Within an O3 network, at least one  site for each MSA, or
21                   Combined Statistical Area (CSA) if multiple MSAs are involved, must be designed to
22                   record the maximum concentration for that particular metropolitan area. More than one
23                   maximum concentration site may be necessary in some areas. The spatial scales for O3
24                   sites are neighborhood, urban and regional.

25                       •  Neighborhood scale:  represents concentrations  within some extended area of
26                         the city that has relatively uniform land use with dimensions in the 0.5-4.0 km
27                         range.  The neighborhood and urban scales listed below have the potential to
        2 Geographic characteristics such as complexity of terrain, topography, land use, etc.
        3 A design value is a statistic that describes the air quality status of a given area relative to the level of the NAAQS. Design values
      are typically used to classify nonattainment areas, assess progress towards meeting the NAAQS, and develop control strategies.
      See http://epa.gov/airtrends/values.html (U.S. EPA, 201 Oa) for guidance on how these values are defined.
      Draft - Do Not Cite or Quote                       3-59                                 September 2011

-------
 1                         overlap in applications that concern secondary or homogeneously distributed
 2                         primary air pollutants.
 3                      •  Urban scale: represents concentrations within an area of city-like dimensions,
 4                         on the order of 4-50 km. Within a city, the geographic placement of sources
 5                         may result in there being no single site that can be said to represent air quality
 6                         on an urban scale.
 7                      •  Regional scale: usually defines a rural area of reasonably homogeneous
 8                         geography without large sources, and extends from tens to hundreds of
 9                         kilometers.

10                  Since O3 concentrations decrease significantly in the colder parts of the year in many
11                  areas, O3 is required to be monitored at SLAMS monitoring sites only during the "ozone
12                  season." Table D-3 of 40 CFR Part 58, Appendix D lists the beginning and ending month
13                  of the ozone season for each U.S. state or territory. Most operate O3 monitors only during
14                  the ozone season. Those that operate some or all of their O3 monitors on a year-round
15                  basis include Arizona, California, Hawaii, Louisiana, Nevada, New Mexico, Puerto Rico,
16                  Texas, American Samoa, Guam and the Virgin Islands.

17                  The total number of SLAMS O3 sites needed to support the basic monitoring objectives
18                  includes more sites than the minimum numbers required in 40 CFR Part 58, Appendix D.
19                  In 2010, there were 1250 O3 monitoring sites reporting values to the EPA AQS database
20                  (Figure 3-17). Monitoring site information for EPA's air quality monitoring networks is
21                  available in spreadsheet format (CSV) and keyhole markup language format (KML or
22                  KMZ) that is compatible with Google Earth™ and other software applications on the
23                  AirExplorer website (U.S. EPA. 201 Id). States may operate O3 monitors in non-urban or
24                  rural areas to meet other objectives (e.g., support for research studies of atmospheric
25                  chemistry or ecosystem impacts). These monitors are often identified as SPMs and can be
26                  operated up to 24 months without being considered in NAAQS compliance
27                  determinations. The current monitor and probe siting requirements have an urban focus
28                  and do not address the siting for SPMs or monitors in non-urban, rural areas to support
29                  ecosystem impacts and the secondary standards.

30                  NCore is a new multi-pollutant monitoring network implemented to meet multiple
31                  monitoring objectives. Those objectives include: timely reporting of data to the public
32                  through AirNow (U.S. EPA. 201 la): support for the development of emission reduction
33                  strategies; tracking long-term trends of criteria pollutants and precursors; support to
34                  ongoing reviews of the NAAQS and NAAQS compliance;  model evaluation; support for
35                  scientific research studies; and support for ecosystem assessments. Each state is required
36                  to operate at least one NCore site. The NCore monitoring network began January 1, 2011
37                  at about 80 stations (about 60 urban and 20 rural sites). NCore has leveraged the  use of
      Draft - Do Not Cite or Quote                      3-60                                September 2011

-------
 1                   sites in existing networks; for example, some IMPROVE sites also serve as rural NCore
 2                   sites. In addition to O3, other components including CO, NOX, NOY, SO2, and basic
 3                   meteorology are also measured at NCore sites. The spatial scale for urban NCore stations
 4                   is urban or neighborhood; however, a middle-scale4 site may be acceptable in cases
 5                   where the site can represent many such locations throughout a metropolitan area. Rural
 6                   NCore sites are located at a regional or larger scale, away from any large local emission
 7                   sources so that they represent ambient concentrations over an extensive area. Ozone
 8                   monitors at NCore sites are operated year round.

 9                   PAMS provides more comprehensive data on O3 in areas classified as serious, severe, or
10                   extreme nonattainment for O3. In addition to O3, PAMS provides data for NOX, NOY,
11                   VOCs, carbonyls, and meteorology. The PAMS network design criteria are based on
12                   locations relative to O3 precursor source areas and predominant wind directions
13                   associated with high O3 concentrations. The overall network design is location specific
14                   and geared toward enabling characterization of precursor emission sources in the area, O3
15                   transport, and photochemical processes related to O3 nonattainment. Minimum
16                   monitoring for O3 and its precursors is required annually during the months of June, July,
17                   and August when peak O3 concentrations are expected. In 2006, the EPA reduced the
18                   minimum PAMS monitoring requirements (71 FR 61236). There were a total of 92
19                   PAMS sites reporting values to the AQS data base in 2010.
       4 Middle scale defines an area up to several city blocks in size with dimensions ranging from about 100 to 500 m.
      Draft - Do Not Cite or Quote                       3-61                                 September 2011

-------
                                                   o Urban NCore
                                                   o PA MS
                                                   • Other Sites Reporting to AQS
     0  250  500    1000 Miles    D 55110 220 Mies   0      250     600           1 COO Miles
           Alaska
                                                                                 Puerto Rico &
                                                                                 Virgin Islands
                                                                                 D 25 5D  100 Miles
Figure 3-17    U.S. ozone sites reporting data to AQS in 2010.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13

14
15
16
              The Clean Air Status and Trends Network (CASTNET) is a regional monitoring network
              established to assess trends in acidic deposition due to emission reduction regulations.
              CASTNET also provides concentration measurements of air pollutants involved in acidic
              deposition, such as sulfate and nitrate, in addition to the measurement of O3. CASTNET
              O3 monitors operate year round and are primarily located in rural areas. In 2010, there
              were 80 CASTNET sites located in, or near, rural areas. As part of CASTNET, the
              National Park Service (NFS) operates 23 sites located in national parks and other Class-I
              areas. Ozone data collected at the 23 NFS sites is compliant with the SLAMS QA
              requirements in 40 CFR Part 58, Appendix A. Ozone measurements at the remaining
              CASTNET sites were not collected with the QA requirements for SLAMS outlined in 40
              CFR Part 58, Appendix A, and therefore, these O3 data cannot be used for NAAQS
              compliance purposes. The SLAMS QA requirements and procedures are currently being
              implemented at the remaining sites.

              The NPS also operates a Portable Ozone Monitoring Systems (POMS) network. The
              POMS couples the small, low-power O3 monitor with a data logger, meteorological
              measurements, and solar power in a self contained system for monitoring in remote
Draft - Do Not Cite or Quote
                                                  3-62
September 2011

-------
 1
 2
 3
 4
 5
locations. Typical uses for the POMS data include research projects, survey monitoring,
and assessments of spatial O3 distribution. The portable O3 monitor in use by the NFS
was recently designated as an equivalent method for O3 (75 FR 22126). Seventeen NFS
POMS monitors were operating in 2010 (NPS. 2011). A map of the rural NCore sites,
along with the CASTNET, and the NFS POMS sites are shown in Figure 3-18.
            Alaska
         D  2Sn SDD
                    mn nm   D sun 223 nits
                                            Rural HCore
                                         t>  HPS POMS
                                         *  CASTNET
                                               i JED mm.
                                                                                        D 2SSD 1U1 lilts
      Figure 3-18    U.S. Rural NCore, CASTNET and NPS POMS ozone sites in 2010.
 6
 7
 8
 9
10
11
12
13
3.5.6.2    Probe/Inlet Siting Requirements

Probe and monitoring path siting criteria for ambient air quality monitoring are contained
in 40 CFR Part 58, Appendix E. For O3, the probe must be located between 2 and 15 m
above ground level and be at least 1 m away (both in the horizontal and vertical direc-
tions) from any supporting structure, walls, etc. If it is located on the side of a building, it
must be located on the windward side, relative to prevailing wind direction during the
season of highest potential O3 concentration. Ozone monitors are placed to determine air
quality in larger areas (neighborhood, urban, or regional scales) and therefore, placement
of the monitor probe should not be near local, minor sources of NO, O3-scavenging
      Draft - Do Not Cite or Quote
                              3-63
September 2011

-------
 1                   hydrocarbons, or O3 precursors. The probe or inlet must have unrestricted air flow in an
 2                   arc of at least 180 degrees and be located away from any building or obstacle at a
 3                   distance of at least twice the height of the obstacle. The arc of unrestricted air flow must
 4                   include the predominant wind direction for the season of greatest O3 concentrations.
 5                   Some exceptions can be made for measurements taken in street canyons or sites where
 6                   obstruction by buildings or other structures is unavoidable. The scavenging effect of trees
 7                   on O3 is greater than other pollutants and the probe/inlet must be located at least 10m
 8                   from the tree drip line to minimize interference with normal air flow. When siting O3
 9                   monitors near roadways, it is important to minimize the destructive interfereences from
10                   sources of NO, since NO reacts readily with O3.  For siting neighborhood and urban scale
11                   O3 monitors, guidance on the minimum distance from the edge of the nearest traffic lane
12                   is based on roadway average daily traffic count (40 CFR Part 58, Appendix E, Table E-
13                   1). The minimum distance from roadways is 10 m (average daily traffic count < 1,000)
14                   and  increases to a maximum distance of 250 m (average daily traffic count > 110,000).
          3.6   Ambient Concentrations

15                   This section investigates spatiotemporal variability in ambient O3 concentrations and
16                   associations between O3 and co-pollutants. To set the stage for the rest of the section,
17                   common O3 measurement units, metrics, and averaging times are described and
18                   compared in Section 3.6.1. Spatial variability is covered in Section 3.6.2 and is divided
19                   into urban-focused variability and rural-focused variability. Urban-focused variability is
20                   organized by scale, extending from national-scale down to neighborhood-scale and the
21                   near-road environment. Rural-focused variability is organized by region and includes
22                   observations of ground-level vertical O3 gradients where available. Temporal variability
23                   is covered in Section 0 and is organized by time, extending from multiyear trends down
24                   to hourly (diel) variability. In many instances, spatial and temporal variability are
25                   inseparable (e.g., seasonal dependence to spatial variability), resulting in some overlap
26                   between Sections 3.6.2 and 0. Finally, Section 0  covers associations between O3 and
27                   co-pollutants including CO, SO2, NO2, PM25 and PM10.

28                   As noted in the 2006 O3 AQCD, O3 is the only photochemical oxidant other than
29                   nitrogen dioxide (NO2) that is routinely monitored and for which a comprehensive
30                   database exists. Data for other photochemical oxidants (e.g., PAN, H2O2, etc.) typically
31                   have been obtained only as part of special field studies. Consequently, no data on
32                   nationwide  patterns of occurrence are available for these other oxidants; nor are extensive
33                   data available on the relationships of concentrations and patterns of these oxidants to
34                   those of O3. As a result, this section focuses solely on O3, the NAAQS indicator for
35                   photochemical oxidants. The majority of ambient O3 data reported in this section were

      Draft - Do Not Cite or Quote                       3-64                                 September 2011

-------
 1                   obtained from AQS, EPA's repository for detailed, hourly data that has been subject to
 2                   EPA quality control and assurance procedures (see Section 3.1 for a description of the
 3                   AQS network).
             3.6.1   Measurement Units, Metrics, and Averaging Times

 4                   Several approaches are commonly used for reporting O3 data. In atmospheric sciences
 5                   and epidemiology, O3 is frequently reported as a concentration, expressed as a volume-
 6                   to-volume mixing ratio, commonly measured in ppm or ppb. In human exposure, O3 is
 7                   frequently reported as a cumulative exposure, expressed as a mixing ratio times time
 8                   (e.g., ppm-h). In ecology, cumulative exposure indicators are frequently used that extend
 9                   over longer time periods, such as growing season or year. This section focuses on
10                   ambient concentrations derived primarily from hourly average O3 measurements and
11                   concentrations are reported in ppb wherever possible. Further details on human and
12                   ecological exposure metrics can be found in Chapter 4 and Chapter 9, respectively.

13                   As discussed in Section 3.1, most continuous O3 monitors report hourly average concen-
14                   trations to AQS with a required precision of 10 ppb and LDL of 10 ppb (see Table 3-2).
15                   This data can be used as reported (1-h avg), or further summarized in one of several ways
16                   to focus on important aspects of the data while simultaneously reducing the volume of
17                   information.  Three common daily reporting metrics include: (1) the average of the hourly
18                   observations over a 24-h period (24-h avg); (2) the maximum hourly observation
19                   occurring in a 24-h period (1-h daily max); and (3) the maximum 8-h running average of
20                   the hourly observations occurring in a 24-h period (8-h daily max)5. Throughout this ISA
21                   and the literature, O3 concentrations are reported using different averaging times as
22                   appropriate, making it important to recognize the differences between these metrics.

23                   Nation-wide, year-round  1-h avg O3 data reported to AQS from 2007-2009 was used to
24                   compare these different daily metrics. Correlations between the 24-h avg,  1-h daily max
25                   and 8-h daily max metrics were generated on a site-by-site basis. Figure 3-19 contains
26                   box plots of the distribution in correlations from all sites. The top comparison in
27                   Figure 3-19 is between 8-h daily max and 1-h daily max O3. Not surprisingly, these two
28                   metrics are very highly correlated (median r = 0.97, IQR = 0.96-0.98). There are a couple
29                   outlying sites, with correlations between these two metrics as low as 0.63, but 95% of
30                   sites have correlations above 0.93. The middle comparison in Figure 3-19  is between 8-h
31                   daily max and 24-h avg O3. For these metrics, the distribution in correlations is shifted
32                   down and broadened out  (median r = 0.89, IQR = 0.86-0.92). Finally, the bottom
        5 For O3 regulatory monitoring purposes, the 8-h daily max is calculated by first generating all 8-h running averages and storing
      these averages hourly by the first hour in the 8-h period. The 8-h daily max is then set equal to the maximum of the 24 individual 8-h
      avg occurring in a given day.
      Draft - Do Not Cite or Quote                       3-65                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
comparison in Figure 3-19 is between 1-h daily max and 24-h avg O3. Again, for these
metrics the distribution in correlations is shifted down and broadened out relative to the
other two comparisons (median r = 0.83, IQR = 0.78-0.88). The correlation between the
two daily maximum metrics (1-h daily max and 8-h daily max) are quite high for most
sites, but correlations between the daily maximum metrics and the daily average metric
(24-h avg) are lower. This illustrates the influence of the overnight period on the 24-h avg
O3 concentration. In contrast, the 1-h daily max and 8-h daily max are more indicative of
the daytime, higher O3 periods. The correlation between these metrics, however, can be
very site-specific, as is evident from the broad range in correlations in Figure 3-19 for all
three comparisons. Therefore, understanding which O3 metric is being used in a given
study is very important since they capture different aspects of O3 temporal variability.
                   8-h daily max
                       vst
                   1-h daily max
                   8-h daily max
                       vs"
                     24-h avg
                   1-h daily max
                       vs"
                     24-h avg
                                0.0   0.1    0.2   0.3   0.4   0.5   0.6   0.7   0.8    0.9    1.0
                                                        Correlation
       Shown are the median (red line), mean (green star), inner-quartile range (box), 5th and 95th percentiles (whiskers), and extremes
      (black circles).

      Figure 3-19    Distribution in nation-wide year-round site-level correlations
                       between daily ozone metrics including 24-h avg, 1-h daily max and
                       8-h daily max using AQS data, 2007-2009.
12
13
14
15
16
17
18
The median 1-h daily max, 8-h daily max, and 24-h avg O3 concentrations across all sites
included in the 3-year nation-wide data set were 44, 40, and 29 ppb, respectively.
Representing the upper end of the distribution, the 99th percentiles of these same metrics
across all sites were 94, 80, and 60 ppb, respectively. While the ratio of these metrics will
vary by location, typically the 1-h daily max will be the highest value representing peak
concentrations and the 24-h avg will be considerably lower representing daily average
concentrations incorporating the overnight period. The 8-h daily max typically represents
      Draft - Do Not Cite or Quote
                               3-66
September 2011

-------
 1                   the higher mid-day concentrations and will generally lie somewhere between the other
 2                   two metrics6.
             3.6.2   Spatial Variability
                     3.6.2.1    Urban-Focused Variability

                     National-Scale Variability

 3                   AQS contains a large depository of national O3 data collected to meet the monitoring
 4                   objectives described in Section 3.5.6.1. In many areas, O3 concentrations decrease
 5                   significantly during months with lower temperatures and decreased sunlight. As a result,
 6                   year-round O3 monitoring is only required in certain areas. Table D-3 of 40 CFR Part 58,
 7                   Appendix D lists the beginning and ending month of the ozone season (defined in
 8                   Section 3.5.6.1) by geographic area and Figure 3-20 illustrates these time periods on a
 9                   monitor-by-monitor basis. Monitoring is optional outside the ozone season and many
10                   states elect to operate their monitors year-round or for time periods outside what is
11                   strictly mandated.

12                   Hourly FRM and FEM O3 data reported to AQS for the period 2007 - 2009 were used to
13                   investigate national-scale spatial variability in O3 concentrations. Given the variability in
14                   O3 monitoring time periods available in AQS as a result of the regionally-varying ozone
15                   seasons, the analyses in this section were based on two distinct data sets:

16                       "a year-round data set: data only from monitors reporting year-round;
17                       •   a warm-season data set: data from all monitors reporting May through
18                          September.
       6 The 8-h daily max is not strictly limited to lie between the 1-h daily max and the 24-h avg since the 8-h averaging period used to
      calculate the 8-h daily max can extend into the morning hours of the subsequent day. However, the 8-h daily max typically
      incorporates the middle of the day when O3 concentrations are at their highest, resulting in an 8-h daily max somewhere between
      the 1-h daily max and the 24-h avg calculated for that day.
      Draft - Do Not Cite or Quote                       3-67                                 September 2011

-------
                                          Required Ozone Monitoring Time  Periods
                   Time Period
               • Apr-Sep  • Mar-Nov
               * Apr-Oct  • May-Sep
                 Apr-Nov  • May-Oct
                 Mar-Sep    Jun-Sep
               • Mar-Ocl  • Year round
                                                                                  Puerto Rico
       Source: U.S. EPA (2QQ8d)
     Figure 3-20    Required ozone monitoring time periods (ozone season) identified
                      by monitoring site.
1
2
3
4
5
6
7
The warm-season data set was used to capture the majority of ozone season data while
providing a consistent time-frame for comparison across states. All available monitoring
data including data from year-round monitors was included in the warm-season data set
after removing observations outside the 5-month window. Data were retrieved from AQS
on February 25, 2011 for these two data sets, and all validated data was included
regardless of flags or regional concurrence7. A summary of the two O3 data sets
including the applied completeness criteria is provided in Table 3-3. Figure 3-21 and
Figure 3-22 show the location of the 457 year-round and 1,064 warm-season monitors
meeting the completeness criteria for all three years (2007-2009).
       7 Concentrations that might have been affected by exceptional events (and contribute to a violation of the NAAQS) can be flagged
     in the Air Quality System (AQS) by the reporting organization. Exceptional events are defined as unusual or naturally occurring
     events that can affect air quality but are not reasonably controllable using techniques that tribal, state or local air agencies may
     implement in order to attain and maintain the National Ambient Air Quality Standards (NAAQS). The corresponding EPA Regional
     Office is responsible for reviewing the data and evidence of the event, and deciding whether to concur with the flag. Flagged data
     that has been concurred by the Regional office is typically excluded for regulatory purposes.
     Draft - Do Not Cite or Quote
                                3-68
September 2011

-------
      Table 3-3       Summary of ozone data sets originating from AQS
                                              Year-Round Data Set             Warm-Season Data Set
      Years	2007-2009	2007-2009	
      Months	January - December (12 mo)	May - September (5 mo)	
      Completeness Criteria                   75% of hours in a day	75% of hours in a day	
                                        75% of days in a calendar quarter	75% of days between May - September
      	all 4 quarters per year	
      Number of monitors meeting completeness     618 containing at least one valid year in 2007-   1,267 containing at least one valid year in
      criteria                             2009                              2007-2009
                                        550 containing at least two valid years in 2007-  1,169 containing at least two valid years in
                                        2009	2007-2009	
                                        457 containing all three valid years in 2007-    1,064 containing all three valid years in 2007-
      	2009	2009	

  1                    Tabulated statistics generated from the year-round and warm-season data sets are
 2                    included in Table 3-4 and Table 3-5, respectively. This information was used to compare
 3                    (1) the year-round and warm-season data sets; (2) the O3 distribution variability across
 4                    years (2005-2009); and (3) four different averaging times (1-h avg, 24-h avg, 1-h daily
 5                    max, and 8-h daily max). Summary statistics for 2005 and 2006 were added to these
 6                    tables in order to gain a broader view of year-to-year variability, but the year-round and
 7                    warm-season data sets used for analyses in the rest of this section are limited to 2007-
 8                    2009 as described above and in Table 3-3. The 8-h daily max pooled by site was also
 9                    included in these tables to show the distribution of the annual and 3-year (2007-2009)
10                    site-averages of the 8-h daily max statistic.

11                    The year-round data set includes data from roughly half the number of monitors as the
12                    warm-season data set and a larger fraction of the year-round monitors are located in the
13                    southern half of the U.S. due to extended monitoring requirements in these areas. Despite
14                    these differences, the mean, SD and percentiles of the nation-wide O3 concentrations
15                    were quite similar for the year-round data presented in Table 3-4 and the warm-season
16                    data presented in Table 3-5. In both data sets, there was very little variability  across years
17                    in the central statistics; for example, the median 1-h avg concentrations between 2005
18                    and 2009 ranged from 28 to 29 ppb for the year-round data and from 29 to 30 ppb for the
19                    warm-season data. The 8-h daily max showed similar uniformity in median across the
20                    five years, with concentrations ranging from 39 to 41 ppb for the year-round data and
21                    from 40 to 43 for the warm-season data. The upper percentiles (95th and above) showed a
22                    general downward trend from 2005 to 2009 in both nation-wide data sets. For example,
23                    the 99th percentile of the 8-h daily max observed in the warm-season data dropped from
24                    85 ppb in 2005 to 75 ppb in 2009. Trends in O3 concentrations investigated over a longer
25                    time period are included in Section 3.6.3.1.
      Draft - Do Not Cite or Quote                        3-69                                  September 2011

-------
Figure 3-21   Location of the 457 ozone monitors meeting the year-round data
             set completeness criterion for all 3 years between 2007 and 2009.
Figure 3-22   Location of the 1,064 ozone monitors meeting the warm-season
             data set completeness criteria for all 3 years between 2007 and
             2009.
Draft - Do Not Cite or Quote
3-70
September 2011

-------
Table 3-4 Nationwide distributions
year-round data set
Time Period
N Monitors
NObs
Mean
SD
Min
Of
1
ozone concentrations (ppb) from the
5
10
25
50
75
90
95
98
99
Max
Max Site ID'1
1 -h avg
2005
2006
2007
2008
2009
2007-2009
499
532
522
520
551
599
4,284,219
4,543,205
4,547,280
4,470,065
4,716,962
13,734,307
29
30
29
30
29
29
18
18
18
17
16
17
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
5
5
6
6
6
15
16
16
17
17
17
28
29
29
29
29
29
41
42
41
41
40
40
53
54
52
52
50
51
61
61
60
59
56
58
71
71
68
67
64
67
78
78
75
74
70
73
182
175
237
222
188
237
060710005
060370016
450790021
450210002
720770001
450790021
24-h avg
2005
2006
2007
2008
2009
2007-2009
504
536
531
528
556
611
183,815
194,884
194,873
191,875
202,142
588,890
29
30
29
30
29
29
13
13
12
12
11
12
2
2
2
2
2
2
4
5
5
5
6
5
9
10
11
11
11
11
13
14
14
14
14
14
20
21
20
21
21
21
28
29
29
29
28
29
37
38
37
38
37
37
46
47
45
46
44
45
51
52
50
50
48
49
57
58
56
56
53
55
61
62
60
61
57
60
103
102
96
98
95
98
060719002
061070009
060651016
060710005
060710005
060710005
1-h daily max
2005
2006
2007
2008
2009
2007-2009
504
536
531
528
556
611
183,815
194,884
194,873
191,875
202,142
588,890
48
48
47
47
45
46
18
18
17
17
15
16
2
2
2
2
2
2
11
13
14
14
14
14
21
23
23
23
22
23
26
28
28
27
27
27
35
36
36
35
35
35
46
46
45
45
44
44
58
58
57
56
54
55
71
71
69
67
64
67
80
80
77
76
72
75
91
91
87
87
83
86
100
100
94
96
91
94
182
175
237
222
188
237
060710005
060370016
450790021
450210002
720770001
450790021
8-h daily max
2005
2006
2007
2008
2009
2007-2009
8-h daily max
2005
2006
2007
2008
2009
2007-2009
504
536
528
528
556
608
(pooled by site)
508
538
538
529
558
457
183,279
194,285
194,266
191,283
201,536
587,085

508
538
538
529
558
457
42
42
41
41
40
41

42
42
41
41
40
41
16
16
15
15
14
15

6
6
6
6
6
6
2
2
2
2
2
2

23
12
17
20
20
19
7
9
10
11
11
10

27
28
27
28
26
29
16
18
19
19
18
19

32
31
31
31
30
32
21
23
23
23
23
23

34
34
34
34
33
34
30
31
31
31
30
31

38
38
38
37
36
38
40
41
40
40
39
40

42
43
41
40
39
40
52
52
51
51
49
50

45
46
45
45
44
45
63
63
61
60
57
60

48
50
49
50
48
49
70
70
68
66
63
66

51
52
51
52
50
51
78
79
75
75
71
74

53
54
54
55
53
54
84
85
81
82
77
80

55
55
55
57
54
55
145
142
137
172
128
172

61
61
63
61
60
61
060710005
060710005
060710005
450210002
060712002
450210002

060710005
060719002
060719002
060719002
060719002
060719002
  aAQS Site ID corresponding to the observation in the Max column
Draft - Do Not Cite or Quote
3-71
September 2011

-------
Table 3-5      Nationwide distributions of ozone concentrations (ppb) from the
              warm-season data set
1
2
3
4
5
Time Period
1-h avg
2005
2006
2007
2008
2009
2007-2009
24-h avg
2005
2006
2007
2008
2009
2007-2009
1-h daily max
2005
2006
2007
2008
2009
2007-2009
8-h daily max
2005
2006
2007
2008
2009
2007-2009
8-h daily max
2005
2006
2007
2008
2009
2007-2009
N Monitors N Obs

1,023
1,036
1,021
1,034
1,029
1,103

1,103
1,110
1,100
1,120
1,141
1,197

1,103
1,110
1,100
1,120
1,141
1,197

1,104
1,112
1,097
1,120
1,141
1,194
(pooled by
1,141
1,152
1,164
1,163
1,173
1,064

7,455,018
7,590,796
7,711,463
7,701,597
7,835,074
23,248,134

319,410
324,993
330,197
329,918
335,669
995,784

319,410
324,993
330,197
329,918
335,669
995,784

318,771
324,327
329,482
329,223
334,972
993,677
site)
1,141
1,152
1,164
1,163
1,173
1,064
Mean

30
31
31
31
29
30

30
31
31
31
29
30

50
50
50
48
46
48

44
44
44
43
40
42

45
44
45
43
41
43
SD Min

19 2
18 2
18 2
17 2
16 2
17 2

12 2
12 2
12 2
12 2
11 2
12 2

18 2
17 2
17 2
16 2
15 2
16 2

16 2
16 2
15 2
15 2
13 2
15 2

6 14
6 12
7 17
6 20
5 20
6 19
1 5

2 2
2 2
2 2
2 2
2 2
2 2

5 10
6 12
6 12
6 12
6 12
6 12

12 23
15 25
16 25
16 25
15 23
16 24

9 18
11 20
12 20
12 20
12 19
12 20

28 34
29 34
28 34
29 33
28 32
29 34
10 25 50

5 16 29
6 17 30
6 18 30
7 18 30
7 17 29
7 18 30

14 22 30
15 22 30
16 23 31
16 22 30
15 21 29
16 22 30

28 38 49
29 38 48
30 38 48
29 37 47
28 36 45
29 37 47

23 32 43
25 33 43
25 33 43
25 33 42
24 31 40
24 32 42

36 41 46
37 41 45
36 40 45
36 39 44
35 38 41
36 39 43
75

43
43
43
42
40
42

39
39
39
38
37
38

61
60
60
58
54
58

55
54
54
52
49
52

49
48
50
48
44
47
90

55
55
55
53
50
53

46
47
47
46
44
45

74
72
72
69
64
68

66
64
65
61
57
61

52
51
54
50
47
50
95 98 99

64 73 79
62 71 77
63 71 77
60 68 74
56 63 69
60 68 74

51 57 61
52 58 61
51 57 61
50 56 60
48 53 56
50 55 59

81 91 99
80 90 98
80 88 95
76 86 93
71 80 87
76 85 93

72 79 85
70 78 84
71 78 82
67 74 80
63 69 75
67 75 80

54 56 57
54 58 59
56 58 59
53 56 58
50 53 55
52 55 57
Max Max Site ID

182 060710005
175 060370016
237 450790021
222 450210002
259 311090016
259 311090016

103 060719002
102 061070009
96 060651016
98 060710005
95 060710005
98 060710005

182 060710005
175 060370016
237 450790021
222 450210002
259 311090016
259 311090016

145 060710005
142 060710005
137 060710005
172 450210002
128 060712002
172 450210002

61 040139508
65 060170020
64 471550102
61 060719002
63 060651016
61 060719002
Given the strong diurnal pattern in O3 concentrations, the selection of averaging time has
a substantial effect on the magnitude of concentration reporting. The nation-wide median
1-h avg, 24-h avg, 1-h daily max, and 8-h daily max concentrations for the year-round
data set in 2009 were 29, 28, 44 and 39 ppb, respectively. The median concentrations for
the warm-season data set in 2009 were: 29, 29, 45 and 40 ppb, respectively. The 1-h avg
Draft - Do Not Cite or Quote
3-72
September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
and 24-h avg both include the lowest concentrations typically observed in the overnight
period which lowers their values relative to the daily maximum statistics.

A strong seasonal pattern in O3 concentrations can also be seen in the year-round data.
Table 3-6 shows the 8-h daily max stratified by season, with the seasons defined as:

    •  winter: December-February;
    •  spring: March-May;
    •  summer: June-August; and
    •  fall: September-November.

In addition, warm-season (May-Sept) and cold-season (Oct-Apr) stratifications of the
year-round data set  are included in the table for comparison with the four seasonal
stratifications. Substantial seasonal variability in the 8-h daily max concentration for the
period 2007-2009 was evident with lower concentrations present in fall (median =
36 ppb)  and winter (median = 32 ppb) and higher concentrations in spring (median =
47 ppb)  and summer (median = 46 ppb). The seasonal differences were even more
pronounced in the upper percentiles. For example, the 99th percentile in the 8-h daily
max over the 2007-09 time period ranged from 52 ppb in winter to 90 ppb in summer.
The distribution in 8-h daily max O3  during the warm-season (as defined above) and
during summer were very similar, which is not surprising given their close overlap in
months.  The distribution during the cold-season (as defined above) is shifted toward
higher 8-h daily max O3 concentrations compared with the distribution during winter.
This is a result of including the four transition months (Oct, Nov, Mar and Apr) in the
cold-season when high O3 concentrations can occur. Further investigation of temporal
variability including multiyear trends and diel behavior is included in Section 0.
      Table 3-6      Seasonally stratified distributions of 8-h daily max ozone
                      concentrations (ppb) from the year-round data set (2007-2009)
Time Period N Monitors
NObs
Mean
SD
Min
1
5
10
25
50
75
90
95
98
99
Max
Max Site ID
8-h daily max (2007-2009)
Year-round 608
587,085
41
15
2
10
19
23
31
40
50
60
66
74
80
172
450210002
8-h daily max by season (2007-2009)
Winter (Dec-Feb) 608
Spring (Mar-May) 612
Summer (Jun-Aug) 613
Fall (Sep-Nov) 608
Warm-season (May-Sep) 616
Cold-season (Oct-Apr) 608
143,855
148,409
148,280
146,541
246,233
340,852
31
47
47
37
47
36
10
12
16
13
16
12
2
2
2
2
2
2
6
20
16
10
16
8
14
28
22
17
22
16
18
33
26
21
27
21
25
40
35
28
35
28
32
47
46
36
46
36
38
55
57
45
57
44
43
62
67
54
66
52
46
67
75
61
73
57
49
72
84
68
81
63
52
77
90
75
87
67
172
118
137
116
137
172
450210002
060370016
060710005
060370016
060710005
450210002
      Draft - Do Not Cite or Quote
                              3-73
September 2011

-------
           O3 > 60 ppb
       50 < O3 < 60 ppb
       40 < O3 < 50 ppb
       30 < O3 < 40 ppb
           O3 < 30 ppb
                Winter
                Spring
Figure 3-23   Highest monitor (by county) 3-year avg (2007-2009) of the 8-h daily
             max ozone concentration based on the year-round data set (top
             map) with seasonal stratification (bottom 4 maps).
Draft - Do Not Cite or Quote
3-74
September 2011

-------
          O3 > 60 ppb
       50 < O3 < 60 ppb
       40
-------
 1                   A national picture of AQS O3 concentrations was generated from the year-round and
 2                   warm-season data sets by aggregating the 8-h daily max observations by U.S. county. For
 3                   this purpose, the 8-h daily max concentrations at each site were averaged over one or
 4                   more calendar years and then the highest site in each county was selected for that county.
 5                   Figure 3-23 contains the county-scale 8-h daily max O3 concentrations from the year-
 6                   round data set for 2007-2009 (top map) with seasonal stratification (bottom four maps).
 7                   Figure 3-24 contains the county-scale 8-h daily max O3 concentrations from the warm-
 8                   season data set for 2007-2009 (top map) along with individual maps for each calendar
 9                   year between 2007 and 2009 (bottom three maps). These maps are meant to illustrate the
10                   general national-scale distribution in long-term average 8-h daily max O3 concentrations
11                   and are not representative of O3 concentrations at all locations or times within the
12                   counties shown; considerable spatial variability can exist within a county. This is
13                   particularly important in the West where counties are larger on average than in the East.
14                   These maps are limited by monitor availability, resulting in the majority of U.S. counties
15                   not having available data (the white regions in Figure 3-23 and Figure 3-24).

16                   As shown in the top county-scale map generated from the 2007-2009 year-round data set
17                   in Figure 3-23, the highest 3-year avg 8-h daily max O3 concentrations (> 50 ppb) occur
18                   in counties in central and southern California, Arizona, Colorado and high elevation
19                   counties in Tennessee. The highest year-round average concentration of 61 ppb over this
20                   period comes from Site #060719002 located at an elevation of 1,244 m in Joshua Tree
21                   National Monument, San Bernardino County, CA. The lowest 3-year avg 8-h daily max
22                   O3 concentrations (<30 ppb) occur in Pacific Coast counties in northern California and
23                   Washington as well as in two northeastern counties in Pennsylvania and Massachusetts.
24                   The seasonally-stratified county-scale maps in Figure 3-24 reinforce the strong
25                   seasonality in 8-h daily max O3 concentrations shown in Table 3-6. The highest
26                   wintertime concentrations (> 40 ppb) occur in the West with the highest 3-year
27                   wintertime avg of 46 ppb calculated for Site #080690007 located at an elevation of
28                   2,743 m near Rocky Mountain National Park, Larimer County, CO. In spring and
29                   summer, the concentrations increase considerably across all counties, with the highest
30                   concentrations  (> 60 ppb) occurring during the summer in  15 counties in California, 3
31                   counties in Colorado and 1  county each in Nevada and Arizona. Many counties in rural
32                   Wyoming, Montana, North Dakota, Maine, and along the Gulf Coast peak in the spring
33                   instead of the summer. In the fall, 8-h daily max O3 concentrations drop back down
34                   below their spring and summer concentrations.

35                   The top county-scale map in Figure 3-24 based on the 2007-2009 warm-season data set
36                   looks similar to the corresponding map in Figure 3-23 based on the year-round data set.
37                   The warm-season map, however, incorporates approximately twice as many monitors
38                   across the U.S., providing more spatial coverage. Several counties in Utah, New Mexico,
      Draft - Do Not Cite or Quote                       3-76                                 September 2011

-------
  1                    Indiana, Ohio, Maryland, North Carolina, and Georgia in addition to California, Arizona,
 2                    Colorado and Tennessee identified above have 3-year avg (2007-2009) 8-h daily max O3
 3                    concentrations > 50 ppb based on the warm-season data set. The individual yearly
 4                    average county-maximum 8-h daily max O3 concentrations in the lower half of
 5                    Figure 3-23 show a general decrease in most counties from 2007 to 2009. The number of
 6                    counties containing a monitor reporting an annual average 8-h daily max O3
 7                    concentration above 50 ppb dropped from 230 counties in 2007 to 30 counties in 2009.
 8                    This is consistent with the general decrease across these years shown in Table 3-4 and
 9                    Table 3-5 for the upper percentiles of the  8-h daily max O3 concentration.


                      Urban-Scale Variability

10                    Statistical analysis of the human health effects of airborne pollutants based on aggregate
11                    population time-series data have often relied on ambient concentrations of pollutants
12                    measured at one or more central monitoring sites in a given metropolitan area. The
13                    validity of relying on central monitoring sites is strongly dependent on the spatial
14                    variability in concentrations within a given metropolitan area. To investigate urban-scale
15                    variability, 20 focus cities were  selected for closer analysis of O3 concentration
16                    variability; these cities are listed in Table  3-7 and were selected based on their
17                    importance in O3  epidemiology studies and on their geographic distribution across the
18                    U.S. In order to provide a well-defined boundary around each city, the combined
19                    statistical area (CSA) encompassing each city was used. If the city was not within a CSA,
20                    the smaller core-based statistical area (CBSA) was selected. The CSAs/CBSAs are
21                    defined by the U.S. Census Bureau (2011)8 and have been used to establish analysis
22                    regions around cities in previous ISAs for particulate matter (U.S. EPA. 2009d) and
23                    carbon monoxide (U.S. EPA. 2010c).
        8A CBSA represents a county-based region surrounding an urban center of at least 10,000 people determined using 2000 census
      data and replaces the older Metropolitan Statistical Area (MSA) definition from 1990. The CSA represents an aggregate of adjacent
      CBSAs tied by specific commuting behaviors. The broader CSA definition was used when selecting monitors for the cities listed
      above with the exception of Phoenix and San Antonio, which are not contained within a CSA. Therefore, the smaller CBSA definition
      was used for these metropolitan areas.
      Draft - Do Not Cite or Quote                        3-77                                 September 2011

-------
Table 3-7
Focus City
Atlanta, GA
Baltimore, MD
Birmingham, AL
Boston, MA
Chicago, IL
Dallas, TX
Denver, CO
Detroit, Ml
Houston, TX
Los Angeles, CA
Minneapolis, MN
New York, NY
Philadelphia, PA
Phoenix, AZ
Pittsburgh, PA
Salt Lake City, UT
San Antonio, TX
San Francisco, CA
Seattle, WA
St Louis, MO
Focus cities used in this and previous assessments
Short Name
Atlanta CSA
Baltimore CSA
Birmingham CSA
Boston CSA
Chicago CSA
Dallas CSA
Denver CSA
Detroit CSA
Houston CSA
Los Angeles CSA
Minneapolis CSA
New York CSA
Philadelphia CSA
Phoenix CBSA
Pittsburgh CSA
Salt Lake City CSA
San Antonio CBSA
San Francisco CSA
Seattle CSA
St Louis CSA
Year-Round
CSA/CBSA Name3 O3 Monitoring
Sites'3
Atlanta -Sandy Springs-Gainesville
Washington-Baltimore-northern VA
Birmingham-Hoover-Cullman
Boston-Worcester-Manchester
Chicago-Naperville-Michigan City
Dallas-Fort Worth
Denver-Aurora-Boulder
Detroit-Warren-Flint
Houston-Baytown-Huntsville
Los Angeles-Long Beach-Riverside
Minneapolis-St. Paul-St. Cloud
New York-Newark-Bridgeport
Philadelphia-Camden-Vineland
Phoenix-Mesa-Scottsdale
Pittsburgh-New Castle
Salt Lake City-Ogden-Clearfield
San Antonio
San Jose-San Francisco-Oakland
Seattle-Tacoma-Olympia
St. Louis-St. Charles-Farmington
0
9
1
3
11
19
12
0
21
47
2
20
9
14
2
2
5
25
5
3
Warm-Season |nc|uded prjor
O3 Monitoring iCA0d
Sites0 ISAs
11
19
9
18
15
0
3
9
0
3
6
10
8
17
12
10
0
6
5
13
CO, PM,SOX, NOX
NOX
PM
CO, PM, NOX
PM, NOX

CO, PM
PM
CO, PM, NOX
CO, PM,SOX, NOX

CO, PM,SOX, NOX
PM, NOX
CO, PM
CO, PM



CO, PM
CO, PM,SOX
        "Defined based on 2000 Census data from the U.S. Census Bureau (2011).
      bThe number of sites within each CSA/CBSA with AQS monitors meeting the year-round data set inclusion criteria.
      The number of sites within each CSA/CBSA with AQS monitors meeting the warm-season data set inclusion criteria; the warm-season data set
      includes May - September data from both the warm-season and year-round monitors meeting the warm-season data set inclusion criteria.
      Boundaries for the CO ISA (U.S. EPA. 201 Oc) and PM ISA (U.S. EPA. 2009d) focus cities were based on CSA/CBSA definitions; boundaries for
      the SOX ISA (U.S. EPA. 2008c) and NOX ISA (U.S. EPA. 2008b) focus cities were based on similar metropolitan statistical area (MSA) definitions
      from the 1990 U.S. Census.
 1                     The distribution of the 8-h daily max O3 concentrations from 2007-2009 for each of the
 2                     20 focus cities is included in Table 3-8. These city-specific distributions were extracted
 3                     from the warm-season data set and can be compared to the nationwide warm-season 8-h
 4                     daily max distribution for 2007-2009 in Table 3-5  (and repeated in the first line of
 5                     Table 3-8 for reference). The median 8-h daily max concentration in these focus cities
 6                     was 41 ppb, similar to the nationwide median of 42 ppb. Seattle had the lowest median
 7                     (31 ppb) and Salt Lake City had the  highest median (53 ppb) of the 20 cities investigated.
 8                     The 99th percentile of the 8-h daily max concentration in the focus cities was 84 ppb;
 9                     similar once again  to the nationwide 99th percentile of 80 ppb.  Seattle had the lowest
10                     99th percentile (64 ppb) and Los Angeles had the highest 99th percentile (98 ppb) of the
11                     20 cities investigated. In aggregate, the 20 focus cities selected are similar in distribution
12                     to the nationwide data set, but there is substantial city-to-city variability in the individual
13                     distributions of the 8-h daily max concentrations based on the warm-season data set.
       Draft - Do Not Cite or Quote
3-78
September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
Table 3-8
Time Period
City-specific distributions of 8-h daily max ozone concentrations
(ppb) from the warm-season data set (2007-2009)
N
Monitors
NObs
Mean
SD
Min
1
5 10
25
50 75
90 95
98 99
Max
Max Site
ID
8-h daily max (2007-2009)
Nationwide
1,194
993,677
42
15
2
12
20 24
32
42 52
61 67
75 80
172
450210002
8-h daily max by CSA/CBSA (2007-2009)
Atlanta CSA
Baltimore CSA
Birmingham CSA
Boston CSA
Chicago CSA
Dallas CSA
Denver CSA
Detroit CSA
Houston CSA
Los Angeles CSA
Minneapolis CSA
New York CSA
Philadelphia CSA
Phoenix CBSA
Pittsburgh CSA
Salt Lake City CSA
San Antonio CSA
San Francisco CSA
Seattle CSA
St Louis CSA
All CSAs/CBSAs
listed
11
28
10
21
27
19
15
9
21
49
8
21
14
22
13
12
5
31
5
19
360
7,844
20,999
7,676
12,603
20,764
19,858
12,217
5,016
22,305
49,295
5,315
26,304
12,673
26,129
9,814
5,146
4,701
28,325
6,148
11,569
314,701
47
43
44
41
37
41
44
45
36
47
40
39
41
49
43
51
39
34
31
43
42
16
16
15
14
14
15
15
14
15
18
12
16
17
12
15
14
13
12
12
15
16
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
15
9
14
13
9
11
8
15
8
10
15
6
8
18
12
8
13
8
4
12
9
22 27
18 23
21 25
21 25
15 19
20 24
18 24
23 28
15 19
20 26
21 25
15 20
17 21
27 32
19 24
23 32
20 23
16 20
12 17
19 23
18 22
36
31
34
31
27
31
34
35
25
35
31
28
29
41
32
44
29
26
23
32
31
47 58
43 54
44 54
40 49
37 47
39 50
44 55
44 52
34 46
45 58
40 48
37 47
39 52
50 58
43 53
53 61
37 46
33 41
31 39
43 53
41 52
67 72
64 70
63 68
59 67
57 62
61 67
63 68
62 69
57 64
72 81
54 58
59 68
64 70
65 68
62 68
67 71
56 62
48 55
46 51
61 68
63 69
81 87
78 83
76 83
75 81
69 74
74 79
72 76
77 83
72 78
91 98
63 67
77 83
78 83
72 75
74 78
77 80
67 72
63 68
59 64
76 81
78 84
124
118
108
104
108
121
98
100
110
137
86
123
125
85
100
96
90
110
91
113
137
130890002
240030014
010732006
250270015
170310042
484390075
080590006
260990009
482011034
060710005
270031002
090050005
240150003
040137021
420050001
490353008
480290032
060010007
530330023
295100086
060710005
Maps showing the location of central monitoring sites with O3 monitors reporting to
AQS for each of the 20 focus cities are included as supplemental material in
Section 3.10.1, Figure 3-61 through Figure 3-80; examples for Atlanta, Boston and
Los Angeles are shown in Figure 3-25 through Figure 3-27. The sites are delineated in
the maps as year-round or warm-season based on their inclusion in the year-round data
set and the warm-season data set (the warm-season data set includes May-September data
from both the warm-season monitors and the year-round monitors meeting the warm-
season data inclusion criteria). The maps also include the CSA/CBSA boundary selected
for monitor inclusion, the location of urban areas and water bodies, the major roadway
network, as well as the population gravity center based on the entire CSA/CBSA and the
individual focus city boundaries. Population gravity center is calculated from the average
longitude and latitude values for the input census tract centroids and represents the mean
center of the population in a given area. Census tract centroids are weighted by their
population during this calculation.
      Draft - Do Not Cite or Quote
                               3-79
September 2011

-------
Figure 3-25   Map of the Atlanta CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
                                    ?
Figure 3-26   Map of the Boston CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-80
September 2011

-------
                      Legend
                      Monitor Locations
                      O  Vl&rm-season Monitors
                      •  Year-round Monitors
                      •  City-based Population Cravity Center
                      •  CSA-based Population Gravity Center
                      	 Interstate Highways
                         Major Highways
                         Water Bod.es
                         Urban Areas
                         Los Angeles CSA
                        .
                                                               200 Kilometers
      Figure 3-27   Map of the Los Angeles CSA including ozone monitor locations,
                      population gravity centers, urban areas, and major roadways.
 1                  The Atlanta CSA contains 11 warm-season monitors distributed evenly yet sparsely
 2                  around the city center (Figure 3-25). The population gravity center for the city and the
 3                  larger CSA are only separated by 4 km, indicating that the majority of the population
 4                  lives within or evenly distributed around the city limits. Atlanta is landlocked with a
 5                  radial network of interstate highways leading to the city center. The Boston CSA contains
 6                  3 year-round and 18 warm-season monitors spread evenly throughout the CSA. Boston is
 7                  a harbor city with the Atlantic Ocean to the east, resulting in the city-based population
 8                  gravity center being located 17  km east of the CSA-based population gravity center. The
 9                  Los Angeles CSA contains the  largest number of monitors of the 20 CSA/CBSAs
10                  investigated with 47 year-round and 3 warm-season monitors. These monitors are
11                  primarily concentrated in the Los Angeles urban area with relatively few monitors
12                  extending out to the northern and eastern reaches of the CSA. These unmonitored areas
13                  are very sparsely populated, resulting in only 15 km separating the city-based and the
14                  CSA-based population gravity centers despite the vast area of the Los Angeles CSA.

15                  Other CSAs/CBSAs (see Section 3.10.1) with monitors concentrated within the focus city
16                  limits include Birmingham, Chicago, Denver, Houston, Phoenix, San Antonio, and Salt
17                  Lake  City. The remaining CSAs/CBSAs have monitors distributed more evenly through-
18                  out the CSA/CBSA area. Baltimore is contained within the same CSA as Washington DC
      Draft - Do Not Cite or Quote                      3-81                                September 2011

-------
1
2
3
and suburbs, resulting in a 50-km separation (the largest of the focus cities investigated)
between the city-based population gravity center for Baltimore and the CSA-based
population gravity center for the Washington-Baltimore-Northern Virginia CSA.
                                         Atlanta CSA
             Site tD
            131210055
            130890002
            131350002
            130670003
            132470001
            130970004
            131130001
            131510002
            130770002
            130850001
            132230003
Years N Mean SD Median IQR
07-09 450 53 17 54 22

07-09 459 51 16 52 22
07-09 455 52 15 53 22
07 09 459 51 17 51 22
07-09 455 47 16 47 19
07-09 458 47 13 47 17
07-09 455 50 14 50 21

Key c i Tiif
t, % I 1 |i K %
f 	 l__!__t__i 	 1 	 1
Site
A-

D-
f ™
H
1-
J-
K-

C

i i i


^
.....
'
• 	 r
:----{
i ...

)


-L
C_

—
L
| —


!


|
4
=£

4


3
4
1
5C


L







>

1 I I
] 	 -1
1 ^
1 	 '
J----!


- - - i
...<

1(
3 (ppb)
1111








30 U


-A
D
-D
-F
H
- 1
-.!
-K

>0

    Figure 3-28   Site information, statistics and box plots for 8-h daily max ozone
                   from AQS monitors meeting the warm-season data set inclusion
                   criteria within the Atlanta CSA.
                                          Boston CSA
Site ID
250250042
250250041
250092006
250213003
250171102
250170009
250095005
330111011
250270024
250094004
440071010
250270015
330110020
330150016
330115001
330150014
250051002
440030002
330131007
440090007
330012004


Years
07-09
08-09
07-09
07-09
07-09
07-09
07-09
07-09
09
07-08
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
Ko
N
459
306
459
459
457
439
459
457
153
305
453
458
455
458
459
459
459
458
459
459
459
f-
f« |
CM £
H *
Mean
35
42
44
46
41
41
44
40
38
46
46
47
38
41
46
40
46
44
39
46
39
ro
E
I
SD
13
12
15
15
15
14
14
13
12
14
15
15
12
13
13
12
13
15
12
14
11
lii?
IE

Median IQR Site
33
41
41
44
40
39
42
37
38
43
44
46
36
40
44
39
45
42
37
45
37

*LD
\~
16
17
20
20
20
19
18
18
16
18
21
19
16
16
17
16
19
18
16
19
15

- H
A-
B-
c-
D-
E-
F —
G-
H-
I-
J-
K-
L-
M-
N-
O
P-
Q-
R -
S-
T-
u-
I , , r , i
I I* ..
-L
c
r
i_
rz
r
i. rjj
!• f 1
r
' [^
-, ,
:--.| \
'-{I


^ ' " C

h..Q
*•--•£

0

j»_j- -^
ig '_ — i


If | - - -:

sU "*

3sZl --^

~~f^\ — <

f3 	 <
ZED 	 '
i»^| 	 ^
: ^ I — i
J* | 	 4

: » | 	 i
£~\---i

-A
-B
-C
- D
-E
-F
— G
-H
-i
- j
- K
- L
-M
- N
- 0
- P
-Q
- R
-S
-T
- U
50 100 150
03 (ppb)
    Figure 3-29   Site information, statistics and box plots for 8-h daily max ozone
                   from AQS monitors meeting the warm-season data set inclusion
                   criteria within the Boston CSA.
    Draft - Do Not Cite or Quote
                            3-82
September 2011

-------
 1                   Box plots depicting the distribution of 2007-2009 warm-season 8-h daily max O3 data
 2                   from each individual monitor in the 20 focus cities are included as supplemental material
 3                   in Section 3.10.2, Figure 3-81 through Figure 3-100; examples for Atlanta, Boston and
 4                   Los Angeles are shown in Figure 3-28 through Figure 3-30. The Atlanta CSA has very
 5                   little  spatial variability in 8-h daily max O3 concentrations with median concentrations
 6                   ranging from 47 ppb at Sites I and J located far from the city center to 54 ppb at Site A
 7                   located closest to the city center. The variation in warm-season 8-h daily max concentra-
 8                   tions are also relatively similar across monitors with IQRs ranging from 17 ppb at Site J
 9                   to 23 ppb at Site B. The Boston CSA has more spatial variability in 8-h daily max O3
10                   concentrations than the Atlanta CSA with median concentrations ranging from 33 ppb at
11                   Site A nearest to the city center to 46 ppb at Site L located 84 km west of the city center.
12                   For monitors located within and just adjacent to the Boston city limits (Sites A-D), the O3
13                   concentrations can vary over relatively short distances owing to differing degrees of NOX
14                   titration and influence from the local topography. Like the Atlanta CSA, the variation in
15                   warm-season 8-h daily max concentrations are relatively similar across monitors within
16                   the Boston CSA with IQRs ranging from 15 ppb at Site U to 21 ppb at Site K. The
17                   Los Angeles CSA exhibits the most variability in O3 concentrations between monitors of
18                   all the CSAs/CBSAs investigated. The median 8-h daily max O3 concentration in the
19                   Los Angeles CSA ranged from 20 ppb at Site AM in the south-central extreme of the
20                   CSA to 80 ppb at Site AE near Crestline, CA in the San Bernardino National Forest just
21                   north of San Bernardino, CA. These two sites are at approximately the same longitude
22                   and are separated by only 85 km, but the Crestline site is downwind of the Los Angeles
23                   basin, resulting in substantially higher O3 concentrations. Site AM also contains data for
24                   only  2009, which could explain some of the deviation when comparing this site with
25                   others in the Los Angeles CSA. Sites AM and AE also had the lowest (8 ppb) and highest
26                   (28 ppb) IQR, respectively. The remaining focus cities exhibited spatial variability
27                   ranging from uniform as in the Atlanta CSA to non-uniform as observed in the
28                   Los Angeles CSA (see supplemental figures in Section 3.10.2).
      Draft - Do Not Cite or Quote                       3-83                                September 2011

-------
                                       Los Angeles CSA
Site 10
060371602
060371301
060371302
060371103
060372005
060374002
060595001
060590007
060375005
060371002
060370002
060370113
060370016
060371701
060591003
060371201
060711004
060376012
060650004
060592022
061112002
060658005
060712002
060658001
061110007
060710012
060379033
061110009
060719004
060659001
060710005
060656001
060714003
060714001
060710306
061113001
061111004
061112003
060650009
060650012
060651016
060710001
060655001
060719002
060652002
060651999
060651010
060711234
060650008
060659003


Years
07-09
07-08
09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
08
07-09
07-09
08-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
09
07-09
07-09
07-09
07-09
07-09
07-09
07-08
09
07-09
07,09
07-09
Key
^
N
458
306
152
457
459
459
459
459
459
459
459
459
458
459
459
459
457
457
127
457
455
276
459
440
459
456
452
458
457
453
459
459
459
455
459
453
458
457
153
457
459
455
459
452
448
283
153
453
265
444

«__
Mean
48
36
44
46
54
38
50
48
45
56
57
48
64
61
45
61
66
68
69
52
62
65
68
69
54
67
6?
58
70
68
79
72
73
68
64
44
57
41
22
73
73
61
69
73
62
49
59
59
58
42
J l
__!_
so
13
9
10
12
15
10
12
10
9
14
17
10
18
16
9
14
19
18
18
13
12
15
19
16
10
13
13
11
19
16
19
17
18
14
12
9
11
9
8
15
16
11
14
13
13
17
10
10
10
10
I
o'E
Median IQR Site . . , , p , , , , i , , , ,
47
34
44
45
53
37
49
47
45
55
56
47
63
60
44
60
66
69
65
50
62
64
67
68
54
67
66
58
70
67
80
73
73
68
64
43
57
40
20
71
73
60
68
73
61
50
59
58
57
42
£
— ,^~
17
10
12
14
18
11
14
12
12
19
22
13
23
20
12
19
23
27
23
15
16
18
24
18
12
18
19
14
26
21
28
24
25
21
17
11
14
12
8
22
23
15
21
18
18
22
15
13
14
13

«>
A-
B-
C-
D-
E-
F -
G-
H-
I -
J-
K-
L-
M-
N-
o-
P-
Q -
R-
s-
j .^
u -
V-
w-
X-
Y-
z-
AA-
AB-
AC-
AD-
AE-
AF-
AG-
AH-
Al-
AJ-
AK-
AL-
AM -
AN-
AO-
AP
AQ-
AR-
AS-
AT-
AU-
AV-
AW-
AX-
'r - - -fj^J' 	 ~.
:- - £JT} - - < ;
* - -{~JT]- - it
i- •• * {~~jT] - ! - • i
•" — CU^IU 	 ^
• -~csii" 'j- '
>-CH; 	 '
;• - - Q|» | — - *
-- - ryT- '-> <
};••- \ "^ \ 	 i
'f"~ \ fr \ ...... • -5
1- - {^fj- '>- - -i
:- 	 ^-^I} 	 ;
j~~~y~i - - ;
~n 	

[ ]^_ } 	 £
^ , , , , ^
• J I- 1 	 H
ZEJ
CD
r . .(
dDJ --

cd
ti j -•
> -CJO ^
co^

[ 1> ] - - - J
^ ~l 	 Ji
f 4 1 - - -- i
r i i
llr
op ...
CD
i"~f~] j
cc
re <
'f il*] ' ^
, ^ | — <
' i i 	 i
CD ^
-.[ Jt } - ••:
_o --••

i^^it^i^j
i 3 '
Ctl
CD
'-•rn---^
- A
-B
-c
-D
-E
-F
-G
-H
-1
- J
-K
-L
-M
-N
-o
-P
-Q
-R
-S
- T
- U
-V
- w
-X
- Y
-z
- AA
- AB
-AC
-AD
-AE
- AF
-AG
- AH
- Al
-AJ
- AK
- AL
-AM
-AN
- AO
- AP
-AQ
- AR
-AS
-AT
- AU
- AV
- AW
-AX
0 50 100 150
03 (ppb)
Figure 3-30    Site information, statistics and box plots for 8-h daily max ozone
                from AQS monitors meeting the warm-season data set inclusion
                criteria within the Los Angeles CSA.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
              Pair-wise monitor comparisons were used to further evaluate spatial variability between
              monitors within the 20 focus cities. In the particular case of ground-level O3, central-site
              monitoring has been justified as a regional measure of exposure mainly on the grounds
              that correlations between concentrations at neighboring sites measured over time are
              usually high. In areas with multiple monitoring sites, averages over the monitors have
              often been used to characterize population exposures. However, substantial differences in
              concentrations between monitors can exist even though concentrations measured at the
              monitoring sites are highly correlated, thus leading to the potential for exposure
              misclassification error. Therefore, both the Pearson correlation coefficient and the
              coefficient of divergence (COD) were calculated for each monitor pair within the
Draft - Do Not Cite or Quote
                                                  3-84
September 2011

-------
 1                   CSA/CBSAs using the 8-h daily max O3 data. The correlation provides an indication of
 2                   temporal linear dependence across sites while the COD provides an indication of the
 3                   variability in absolute concentrations across sites. The COD is defined as follows:
                                                                                          Equation 3-1

 4                   where Xy and Xik represent observed concentrations averaged over some measurement
 5                   averaging period /' (hourly, daily, etc.) at sites j and k, and/? is the number of paired
 6                   observations. A COD of 0 indicates there are no differences between concentrations at
 7                   paired sites (spatial homogeneity), while a COD approaching 1 indicates extreme spatial
 8                   heterogeneity. These methods for analysis of spatial variability follow those used in
 9                   previous ISAs for CO, PM, SOX and NOX as well as those used in Pinto et al. (2004) for
10                   PM25.

11                   Histograms and contour matrices of the Pearson correlation coefficient between 8-h daily
12                   max O3 concentrations from each monitor pair are included as supplemental material in
13                   Section 3.10.3, Figure 3-101 through Figure 3-120; examples for Atlanta, Boston and
14                   Los Angeles are shown in Figure 3-31 through Figure 3-33. Likewise, histograms,
15                   contour matrices, and scatter plots of the COD between 8-h daily max O3 concentrations
16                   from each monitor pair are included as supplemental material in Section 3.10.3, Figure 3-
17                   121 through Figure 3-140; examples for Atlanta, Boston and Los Angeles are shown in
18                   Figure 3-34 through Figure 3-36. These figures also contain scatter plots of correlation
19                   and COD as a function of straight-line distance between monitor pairs.

20
      Draft - Do Not Cite or Quote                       3-85                                September 2011

-------
                                            Atlanta CSA
5
           20-
           15-
           10
            5-
            -0.1    0.0     0.1     0.2     0.3
                                              0.4     0.5
                                               Correlation
                               CO     O
            1.0-

            0.9-

            08

            0.7-

            0.6-
         c
         I  0.5

         °  0.4-

            0.3-

            0.2

            0.1-

            00

           -0.1
                                                                            0.75    076
                                                     0.82    088    0.90    0.87   074    075
                                                     0.77    0.73    0.75   0,78   079    0.68
                                             0.90   0.82   0.77    0.81    081
                                                              085    058
                                                  084    076    088    075
                                                                    081    076
                                                              0.86    0,63   070   I H
                                                                    069
                                                                         OB1
               0     50    100    150   200   250    300   350   400    450
                                       Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the correlations.

Figure 3-31    Pair-wise  monitor correlations expressed as a histogram (top),
                 contour matrix (middle) and scatter plot versus distance between
                 monitors (bottom) for the Atlanta CSA.
Draft - Do Not Cite or Quote
                                      3-86
September 2011

-------
                                            Boston CSA
          60-
          40-
           -0.1
—I—
 0.0
0.1
0.2
0.3
0.4     0.5
 Correlation
                                                        10
                                                                  o
                                                                        a  en
                                                     0 83 0 85 079 0 88 0 79 0 90 0 78 081 078 0 81 0 74

                                                     085 0.85 080 0.90 077 0.90 073 0.80 077 074 0.74


                                                     M' 088 081 082 080 083 076 085 077 079 0.72  -D


                                               ^Io81 082 ^B 089 077 0.86 080 062 080 096 065 030
                                                                                    ••

                                                I 082 079 I   I 077 0.84 0.82 050 0.77 0.89 061 0.83  F
1.0 -


0.9-
0.8

0.7-

0.6-
E
1 °5'
"oi
•
3 0.4-

0.3-
0.2-

0.1-
0.0-
-n 1 -

079 088 •• 088 084 089 063 075 088 :) bs 083
.rt-
*** "*^"A*
\ »•* •
,t^..j i,fm
!'V»*l».* \ Io65 070 076 065 073
T ** ** % • * • *

• * • / .
• .• • *
.% .. » 079 06< 083 0.83 068 0.77
• • *
."•.*• 074 0.87 0.80 054 0 69 II 0 56 H
* • • 07lllo67 069 074 066 071
^^H^^^l
* * 07S 059 070 0,84 060 0.82

066 068 0.81 065 077
0.76 0 53 II 0.49
065 0.85 0.60
0.54^1
0.49




-G

-H
1
-J

-K
•L
-M
-N
-O

-P
-Q
- R
-s
-T
-U


                   50    100   150    200   250   300   350    400   450
                                      Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the correlations.

Figure 3-32    Pair-wise monitor correlations expressed as a histogram (top),
                 contour matrix (middle) and scatter plot versus distance between
                 monitors (bottom) for the Boston CSA.
Draft - Do Not Cite or Quote
                             3-87
                                                         September 2011

-------
                                        Los Angeles CSA
150
w
§ 100-
o
0 50




3







1OQ


i7n




	 1
164




147




148




151




144






87
	
                                                                                 29
          -0.1
0.0
0.1
0.2
0.3
0.4     0.5
 Correlation
0.6
0.7
0.8
09
1 0
                                                ,   <
      I
      o
                         ...,',«.,
   Ti
                                                                     3«oa«.-:-. -r» mi--.ru-. nrs •:- iif n
         1.0

         0.9

         0.8

         0.7

         0.6

         0.5
         0.4-
         0.3

         0.2

         0.1

         0.0

        -0.1
                                    B "«iB»>!fcl«««'i«m.T..v.., , ««.OD,«i«uiigii
                                    i".*.... .||jjj|.»:.,:,^..»,jj
                                            . :.r^^moir m i:.j4«]^oH fl»d«(
                 50   100   150   200   250   300   350   400   450
                                    Distance (km)
                                                                  P
                                                                  I
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the correlations.

Figure 3-33    Pair-wise monitor correlations expressed as a histogram (top),
                contour matrix  (middle) and scatter plot versus distance between
                monitors (bottom) for the Los Angeles CSA.
Draft - Do Not Cite or Quote
                           3-88
                                                       September 2011

-------
                                        Atlanta CSA


~c
0
O

30-
25-
20-
15-
10-
5-
33

22









0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55












§
te
>
5
efficient of
0
0












< m
006


0.55-


0.50-


0.45-


0.40-
0.35-


0.30-
0.25-
0.20-


0.15-


0.10-
0.05-
n nn














. ,/Y. *
•\v*. '
'••'

Coefficient of Divergence
OQUJLi-C5I_-5*:
009 009 008 0.09 0.08 0.08 0.11 013 011
010 0.10 008 0,10 009 0.09 0.11 0.13 0.12
010 011 0.11 0.12 011 0.13 0.12 0.13

011 0.07 0.09 010 0.11 010 0.08

011 0,08 005 010 013 012

0.08 010 0.10 0.11 0.07

006 007 012 0.10

010 013 012
013 011

011



-A
-B
-C

-D

- E

-F

-G

-H
-I

- J

-K



           0    50    100   150   200   250   300   350   400   450   500
                                   Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the CODs.

Figure 3-34    Pair-wise monitor COD expressed as a histogram (top), contour
                matrix (middle) and scatter plot versus distance between monitors
                (bottom) for the Atlanta CSA.
Draft - Do Not Cite or Quote
3-89
September 2011

-------
                                         Boston CSA
100-
- 80-
§ 60-
0 40-
20-
2

75
114
18
          0.00   0.05    0.10    0.15    0.20   0.25    0.30    0.35
                                        Coefficient of Divergence
                    0.40
0.45
0.50
0.55
012 014 0.16 013 0.13 015 0.14 0.10 017 017 018 012 0.14 0.18 012 019 017 013 019 014
0.06 0.07 0.10 010 0.07 012 010 006 009 009 012 0.07 0.10 0.07 0.10 0.11 012 010 0.11

007 010 010 007 011 010 0.06 009 010 012 0.08 010 0.08 0.11 0.11 012 0.11 012
010 0.11 007 0.12 008 0.08 0.07 0.09 013 0.11 010 011 010 009 013 010 014
007 0,08 0.08 0,05 0.12 0.11 0.10 0.09 0.12 011 011 015 0.12 010 014 011
0,08 0.08 0,07 012 012 011 0.08 0.12 012 011 015 0.13 0.09 015 010
0.55-

0.50-

0.45-

0.40-

8
§ 0.35-
f
5 0.30-
S 0.25-
u
° 0.20-

0.15-

0.10-
0.05-
n nn
009 0.08 007 0.10 0.09 010 0.09 0.09 0.08 0.12 011 0.10 012 010
0.09 0.12 0,13 0,12 0.06 0.12 012 011 016 013 0.07 016 0.09

0.08 0.09 0.10 0.10 0.09 0.11 012 011 010 012 011
0,11 0.10 013 0.06 0.11 0.07 0.11 011 013 012 012
0.09 015 0.12 0.11 013 0.09 0.08 015 0.08 0.15

014 0.12 0.08 0.12 0.12 0.10 013 012 013

0.12 012 0.10 0.17 014 0.05 017 0.07
012 0.04 013 0.13 011 013 011
0.11 0.12 0.12 0.12 013 011
013 0.13 010 014 010
0.11 0.16 0.06 0.16
014 0.09 0,15
.* **.*
. •»**•••• ° 16 O-06
*•"*.* • *
.'•"•i**:»«lV»: *«•*'••' * °16
,;JtyP • ''
-A
-B

-c
-D
-E
-F
-G
-H

-I
-J
•K

-L

-M
-N
-O
- P
-Q
-R
-s

-T
-u
ff •

           0     50    100    150   200   250    300   350   400    450   500
                                   Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the CODs.

Figure 3-35    Pair-wise monitor COD expressed as a histogram (top), contour
                matrix (middle) and scatter plot versus distance between monitors
                (bottom) for the Boston CSA.
Draft - Do Not Cite or Quote
3-90
        September 2011

-------
                                        Los Angeles CSA
        400-
      .„ 300-
      o 200-
      O
        100-
3
155

417

257
181

108
43 16 6 12 16
  0.00    0.05    0.10   0.15    0.20    0.25    0.30    0.35
                                Coefficient of Divergence
                                                                0.40    0.45
                                                                              0.50
                                                                                     0.55
        0.55-

        0.50-

        0.45-

        0.40-

        0.35-
      ai
      5 0.30-
        0.25-
     <
0.20-

0.15-

0.10-

0.05-
        0.00
                                                                                   -A
                                                                           I
                                                                           I
                                                                           I
                                                                            w
                                                                            AB
                                                                            AC
                                                                            AD
                                                                            i
                                                                                    AU
                                                                                   -AX
            0     50    100   150    200   250   300   350    400   450   500
                                    Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the CODs.

Figure 3-36    Pair-wise monitor COD  expressed as a histogram (top), contour
                matrix (middle) and scatter plot versus distance between monitors
                (bottom) for the Los Angeles CSA.
Draft - Do Not Cite or Quote
                                     3-91
September 2011

-------
 1                   The monitor pairs within the Atlanta CSA (Figure 3-31) were generally well correlated
 2                   with correlations between 8-h daily max O3 concentrations ranging from 0.61 to 0.96.
 3                   The correlations shown in the scatter plot were highest for close monitor pairs and
 4                   dropped off with distance in a near-linear form. At a monitor separation distance of 50
 5                   km or less, the correlations ranged from 0.79 to 0.96. The monitor pairs within the
 6                   Boston CSA (Figure 3-32) were also generally well correlated with correlations ranging
 7                   from 0.49 to 0.96. Again, the correlations shown in the scatter plot were highest for close
 8                   monitor pairs, but there was slightly more  scatter in correlation as a function of distance
 9                   in the Boston CSA compared with the Atlanta CSA. At a monitor separation distance of
10                   50 km or less, the correlations ranged from 0.81 to 0.96. The monitor pairs within the
11                   Los Angeles CSA (Figure 3-33) showed a much broader range in correlations, extending
12                   from -0.06 to 0.97. At a monitor separation distance of 50 km or less, the correlations
13                   shown in the scatter plot ranged from 0.21 to 0.97. The negative and near-zero
14                   correlations were between monitors with a relatively large separation distance (>150 km),
15                   but even some of the closer monitor pairs were not very highly correlated.  For example,
16                   Site AL located at Emma Wood State Beach in Ventura and Site AK situated in an
17                   agricultural valley surrounded by mountains 20 km inland (see map in Figure 3-37) had a
18                   correlation coefficient of only 0.21 over the 2007-2009 warm-season time  period. This
19                   was slightly lower than the correlation between Site AL and Site AX on the Arizona
20                   border, 441 km away (R = 0.28). San Francisco and Seattle (Figure 3-118 and Figure 3-
21                   119 in Section 3.10.3) also showed a broad range in pair-wise correlations, likely
22                   resulting from their similar geography where background  air coming in from the Pacific
23                   Ocean rapidly mixes with urban pollutants such as NOX and VOCs from coastal cities
24                   and is transported downwind into diversified terrain to create spatially and temporally
25                   varying O3 concentrations.
      Draft - Do Not Cite or Quote                        3-92                                 September 2011

-------
      Figure 3-37   Terrain map showing the location of two nearby AQS ozone
                     monitoring sites (red dots) along the western edge of the
                     Los Angeles CSA. Site AL is near shore, 3 m above sea level, while
                     Site AK is in an agricultural valley surrounded by mountains, 262 m
                     above sea level.
 1
 2
 3
 4
 5
 6
 1
 8
 9
10
11
12
13
The COD between 8-h daily max O3 measured at paired monitors in all CSAs/CBSAs
(Figure 3-121 through Figure 3-140 in Section 3.10.3) were generally low, with values
similar to those shown in Figure 3-34 and Figure 3-35 for Atlanta and Boston. This
suggests a generally uniform distribution in the 8-h daily max O3 concentration across
monitors within these cities and is consistent with the uniformity observed in the box
plots (e.g., Figure 3-28, Figure 3-29, and Figure 3-81  through Figure 3-100 in
Section 3.10.2). Los Angeles (Figure 3-30) and San Francisco (Figure 3-138 in
Section 3.10.3), however, had several monitor pairs with COD >0.30 indicating greater
spatial heterogeneity. This is consistent with the variability observed in the box plots for
these two CSAs (Figure 3-30 and Figure 3-98 in  Section  3.10.2). In particular, Site AM
in the Los Angeles CSA had consistently lower concentrations (median = 20 ppb, IQR =
17-25 ppb) relative to other sites in the CSA (Figure 3-27), resulting in high CODs with
other monitors as shown in Figure 3-36. The O3 monitor at Site AM is located on the
      Draft - Do Not Cite or Quote
                             3-93
September 2011

-------
1
2
3
4
5
6
7
Pechanga Tribal Government Building in Temecula, CA, and began collecting data on
June 9, 2008. It is located in a suburban setting and is classified as a general background
monitor. Another close by site (site ID = 060731201) located in the Pala Reservation, 9.5
km south of this one (just outside the boundary of the Los Angeles CSA) reported
similarly low 2009 8-h daily max O3 concentrations (median = 28 ppb, IQR = 23-32 ppb)
between May-June, 2009 (the only warm-season months with available data from this site
between 2007 and 2009).
                                                                               I
     Figure 3-38   Terrain map showing the location of four AQS ozone monitoring
                   sites (red dots) located in or near the city limits in the center of the
                   Boston CSA. Site characteristics range from Site A near downtown
                   at 6 m above sea level to Site D in a forested area on Blue Hill at
                   192 m above sea level.
    Draft - Do Not Cite or Quote
                            3-94
September 2011

-------
 1                   There are instances where sites in an urban area may exhibit substantial differences in
 2                   median concentrations, but still be moderately well correlated in time. For example, Sites
 3                   A and D in Boston (see terrain map in Figure 3-38) have an 11 ppb difference in median
 4                   8-h daily max O3 concentration (COD = 0.16), but a high correlation (R = 0.90). In this
 5                   example, Site A is located in the Boston city limits at an elevation of 6 m while Site D is
 6                   located 13 km to the south in a forested area on Blue Hill, the highest point in Norfolk
 7                   County (elevation = 192 m). The difference in median O3 concentration at these two sites
 8                   can be attributed to differing degrees of NOX titration between the neighborhood scale
 9                   site (Site A) and the regional scale site (Site D) and to the influence of local topography.

10                   Comparison of monitoring data within the selected focus cities has demonstrated
11                   considerable variability between cities in the behavior of the O3 concentration fields.
12                   Median O3 concentrations vary more within certain urban areas than  others. Likewise,
13                   pair-wise monitor statistics (R and COD) are dependent on the urban area under
14                   investigation. These conclusions are consistent with those drawn in the 2006 O3 AQCD
15                   where a subset of these focus cities were investigated using similar statistics. As a result,
16                   caution should be observed in using data from a sparse network of ambient O3 monitors
17                   to approximate community-scale exposures.


                     Neighborhood-Scale Variability and the Near-Road Environment

18                   Ozone is a secondary pollutant formed in the atmosphere from precursor emissions and
19                   therefore is generally more regionally homogeneous than primary pollutants emitted from
20                   stationary or mobile point sources. However, O3 titration from primary NO emissions
21                   does result in substantial localized O3 gradients. This is evident in the near-road
22                   environment where fresh NO emissions from motor vehicles titrate O3 present in the
23                   urban background air, resulting in an O3 gradient down-wind from the roadway. Ozone
24                   titration occurring in street canyons where NO emissions are continuously being
25                   generated is more efficient because of inhibited transport away from the source of NO.

26                   Several studies have reported O3 concentrations that increase with increasing distance
27                   from the roadway, both upwind and downwind of the road. Beckerman et al. (2008)
28                   measured O3 profiles in the vicinity of heavily traveled roadways with Annual Average
29                   Daily Traffic (AADT) >340,000 vehicles in Toronto, Canada. Ozone was observed to
30                   increase with increasing distance from the roadway, both upwind and downwind of the
31                   road. This is consistent with scavenging of O3 in the near-road environment by reaction
32                   with NO to form NO2. Upwind of the road, concentrations were >75% of the maximum
33                   observed value at >100 m from the road; downwind, concentrations were approximately
34                   60% of the maximum within 200-400 m of the road. The O3 concentration adjacent to the
35                   road on the upwind side was approximately 40% of the maximum value observed at the
      Draft - Do Not Cite or Quote                       3-95                                September 2011

-------
 1                   site. Concentrations measured with Ogawa passive samplers over a 1-week period ranged
 2                   from 7.3-19.4 ppb with the mean at the two sites ranging from 13.0-14.7 ppb. In a study
 3                   of patrol cars during trooper work shifts, Riediker et al. (2003) made simultaneous 9-h O3
 4                   measurements inside patrol cars, at the roadside, and at a centrally-located ambient
 5                   monitoring site. The roadside concentrations were approximately 81% of the ambient
 6                   values (mean of 22.8 ppb versus 28.3 ppb). Wind direction relative to the roadway was
 7                   not reported.

 8                   Johnson (1995) measured O3, NO, and CO concentrations at upwind and downwind
 9                   locations near a variety of roadways in Cincinnati, OH. The effects of O3 scavenging by
10                   NO were apparent in the O3 reduction in the interval between 9 m upwind  and 82 m
11                   downwind of the road. A similar effect was observed by Rodes and Holland (1981)
12                   during an earlier study in which outdoor O3 concentrations were monitored downwind of
13                   a freeway in Los Angeles, CA. In this study, O3 concentrations measured near the
14                   roadway were approximately 20% of the concentrations measured simultaneously at
15                   more distant locations judged to be unaffected by the roadway. Minimal separation
16                   distances of the samplers from the roadway to eliminate measurable influence were
17                   estimated to be approximately 400-500 m for NO, NO2, and O3. Similar results have
18                   been observed outside the U.S., e.g., in the city of Daegu, Korea, where the yearly
19                   roadside concentrations of CO and SO2 showed a well-defined decreasing  trend with
20                   distance from the roadway, whereas concentrations  of NO2 and O3 exhibited the reverse
21                   trend (Jo and Park. 2005). During the peak O3 month of May, O3 concentrations in a
22                   residential neighborhood were approximately 40% higher than concentrations at roadside
23                   monitors located 1 m from the edge of multiple-lane freeways.
                     3.6.2.2   Rural-Focused Variability and Ground-Level Vertical
                               Gradients

24                   AQS O3 data for monitors located at several rural monitoring sites (e.g., national parks,
25                   national forests, state parks, etc.) were used to investigate rural-focused O3 concentration
26                   variability in contrast with the urban-focused variability discussed in Section 3.6.2.1.
27                   These rural monitoring sites tend to be less directly affected by dire t anthropogenic
28                   pollution sources than urban sites. However, they can be regularly affected by transport
29                   of O3 or O3 precursors from upwind urban areas, or by local anthropogenic sources
30                   within the rural areas such as emissions from motor vehicles, power generation, biomass
31                   combustion, or oil and gas operations. As a result, monitoring data from these rural
32                   locations are not unaffected by anthropogenic emissions.
      Draft - Do Not Cite or Quote                      3-96                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
Six rural focus areas were selected for their geographic distribution across the U.S. as
well as their unique topography and relevance to the ecological assessment in Chapter 9.
Table 3-9 lists the rural focus areas and provides some cursory site information along
with the number of available AQS monitors reporting year-round and only during the
warm-season. Accompanying box plots depicting the distribution of 2007-2009 warm-
season 8-h daily max O3 data from each individual monitor in the six rural focus areas
are included in Figure 3-39. This analysis was restricted to AQS monitors meeting the
same data completeness criteria outlined in Table 3-3 for a direct comparison with the 20
urban focus areas investigated in Section 3.6.2.1. Given the population-center emphasis
of the AQS network, limited monitoring sites (between one and five) were available for
each rural focus area. Expanded analyses of O3 concentrations measured using the more
rural-focused  CASTNET monitoring network are included in Chapter 9.
Table 3-9
Focus Area
Adirondack State
Park, NY
Mount Mitchell State
Park, NC
Great Smoky
Mountain National
Park, NC-TN
Rocky Mountain
National Park, CO
San Bernardino
National Forest, CA
Sequoia National
Park, CA
Rural focus areas
Year-Round
Short O3
Name Monitoring
Sites3
ADSP 1
MMSP 0
SMNP 2
RMNP 1
SBNFc 1
SENP 2
Warm-Season
Os Monitor
Monitoring Elevation (m)
Sites'3
0 1 ,483
1 1 ,982
3 564-2,021
0 2,743
0 1 ,384
0 560-1 ,890
Site Descriptions
One site on the summit of Whiteface Mountain in
the Adirondack Mountains
One site near the summit of Mount Mitchell
(highest point in the eastern U.S.) in the
Appalachian Mountains
Five different locations within Great Smoky
Mountain National Park in the Appalachian
Mountains
One site in a valley at the foot of Longs Peak in the
Rocky Mountains
One site in Lake Gregory Regional Park (near
Crestline, CA) in the San Bernardino Mountains
Two contrasting sites at different elevations within
Sequoia NP in the Sierra Nevada Mountains
      "Number of AQS monitors meeting the year-round data set inclusion criteria; the year-round data set is limited to these monitors.
      bNumber of AQS monitors meeting the warm-season data set inclusion criteria; the warm-season data set includes May-September data from
      both the warm-season and year-round monitors.
      °Same AQS site as Site AE in the Los Angeles CSA shown in Figure 3-27.
      Draft - Do Not Cite or Quote
                                 3-97
September 2011

-------
                                     Rural Focus Areas
Site ID
360310002
371990004
370870036
470090102
470090101
471550101
471550102
080690007
060710005
061070009
061070006
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
N
445
447
456
459
459
458
457
456
459
416
459
"to
 " J~
E E " "S
•

|--
	 H


Site
A
A
A
B
C
D
E
A
A
A
B
i , i i , , i i , i
r - - "1 fr 1 	 -:
"~--\ 4 [---•
>:..[33-< '
,-an-M
>- QD
••- 1 4 f - •
I 1
I 4 h
-- rn-
0 50 100 1£
03 (ppb)
     Figure 3-39   Rural focus area site information, statistics and box plots for 8-h
                    daily max ozone from AQS monitors meeting the warm-season data
                    set inclusion criteria within the rural focus areas. Includes:
                    Adirondack State Park, NY (ADSP); Mount Mitchell State Park, NC
                    (MMSP); Great Smoky Mountain National Park, NC-TN (SMNP);
                    Rocky Mountain National Park, CO (RMNP); San  Bernardino
                    National Forest, CA(SBNF); and Sequoia National Park, CA(SENP).
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
Eastern Rural Focus Areas

In the East, the distribution in warm-season 8-h daily max O3 concentrations from the
Adirondack State Park (ADSP) site on Whiteface Mountain in Upstate NY (median =
49 ppb) (Figure 3-39) was among the lowest of the rural focus monitors investigated, but
was still higher than concentration distributions measured in the Boston CSA (medians
ranging from 33 to 46 ppb) (Figure 3-30) located 320 km to the southeast. The ADSP
AQS site was included in an analysis for the 2006 O3 AQCD and had the lowest year-
round median hourly O3 concentration of the rural forested sites investigated (including
Yellowstone NP, the Great Smoky Mountains NP, and Shenandoah NP). For the
Appalachian Mountain monitors in Mount Mitchell State Park, NC (MMSP) and Great
Smoky Mountain National Park, NC-TN (SMNP), there was a fair amount of variability
in concentration distribution. Within SMNP, the median warm-season 8-h daily max O3
concentration ranged from 47 ppb at the lowest elevation site (elevation = 564 m; site ID
= 470090102) to 60 ppb at the highest elevation site (elevation = 2,021 m; site ID =
471550102); these sites are shown on the terrain map in Figure 3-40. The warm-season
median 8-h daily max O3 concentration gradient between these two sites located 26.2 km
apart in SMNP was 0.9 ppb per 100 m elevation gain.
     Draft - Do Not Cite or Quote
                            3-98
September 2011

-------
 1                  Data from the five sites within SMNP allowed for further investigation of spatial variabil-
 2                  ity within the park; Figure 3-41 contains histograms, contour plots and scatter plots as a
 3                  function of distance for the pair-wise correlation and COD (defined in Equation 3-1) for
 4                  SMNP. The correlations between the five sites ranged from 0.78 to 0.92 and the CODs
 5                  ranged from 0.04 to 0.16. The plots of correlation and COD as a function of distance be-
 6                  tween SMNP monitor pairs in Figure 3-41 show a large degree of spatial variability be-
 7                  tween monitors over relatively short distances. A host of factors may contribute to these
 8                  variations, including proximity to local O3 precursor emissions, variations in boundary-
 9                  layer influences, meteorology and stratospheric intrusion as a function of elevation, and
10                  differences in wind patterns and transport behavior due to local topography.
                               -    Motmtjtm        NW-Math81      L-Mbr^'i  "

                           N      /'1,«                      """"


                           JJF'     —          @   — '   -
                          ^f'*P. / j£~^.  —^—r*-r_ ,.. ..        *S^*r
                                           .. KQ'd^rr
                       .^-^'Knoxvilte -
                                              ,©
                    [jB
                    5
                                  '
                                   *
                                                              .'@J \i£Z
       The lowest elevation site (Site B) is 564 m above sea level, while the highest elevation site (Site E) is 2,021 m above sea level.

      Figure 3-40   Terrain map showing the location  of five AQS ozone monitoring
                      sites (green/black stars) in Great Smoky Mountain National Park,
                      NC-TN  (SMNP).
      Draft - Do Not Cite or Quote                      3-99                                September 2011

-------
                                      Smokey Mtn NP, NC-TN
      I!
        -0.1  0.0  0.1  0.2  0.}  0.4  O.i  0.6  0.7  0.3  0.9  1.0
                       Correlation
                                                        0.00  0.05  0.10 015 0.20 0.25  0.30  0-35 0.40 0.45 0.50  055
                                                                   Coefficient of Divergence
          0 50 100 150 200 250  300 350 400 450
                   Distance (km)
                                                          0  50 100 150 200 250 300 350 400 450 500
                                                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histograms includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of the correlations and CODs.

Figure 3-41    Pair-wise monitor correlations (left) and coefficients of divergence
                (COD, right) expressed as a histogram (top), contour matrix
                (middle) and scatter plot vs. distance between monitors (bottom)
                for Great Smoky Mountain National Park, NC-TN (SMNP).
 1
 2
 3
 4
 5
 6
 1
 8
 9
10
11
12
13
14
              Western Rural Focus Areas

              The Rocky Mountain National Park (RMNP) site in Colorado at 2,743 m in elevation had
              a warm-season 8-h daily max O3 concentration distribution (median = 56 ppb, IQR =
              11 ppb) (Figure 3-39) that is comparable to the distributions at sites in the Denver CSA
              located 75 km southeast at elevations around 1,600 m (medians ranging from 41 to
              59 ppb, IQRs ranging from 10 to 16 ppb; see Figure 3-72 in Section 3.10.1). In nearby
              Boulder County, CO, a 1-year time-series (Sep 2007 - Aug 2008) of ambient surface-
              level O3 measurements was collected by Brodin et al. (2010) along an elevation gradient
              ranging from 1,608 m to 3,528 m. The 7 sites used in this study are shown in Figure 3-42
              along with the RMNP site and the 15 Denver CSA sites. In fall, winter, and spring, they
              observed a clear monotonic increase in O3 concentration with elevation, with a rate of
              increase in the mean O3 concentration of 1.5 ppb per 100 m elevation gain during winter.
              In summer, the O3 gradient was similar in magnitude over the seven-site transect (1.3 ppb
              per 100 m), but much less monotonic; the majority of the vertical gradient occurred
              between the lowest two sites (4.5 ppb per 100 m) and between the highest two sites
Draft - Do Not Cite or Quote
                                                  3-100
September 2011

-------
1
2
3
4
5
6
(5.5 ppb per 100 m), with the middle five sites all having approximately equal median O3
concentrations. Ozone concentrations at the lowest site in Boulder were influenced by
NO titration as evidenced by traffic-related diel cycles in O3 concentrations, but the
remaining six sites were located at elevation in more rural/remote settings and illustrate a
positive surface-level O3 elevation gradient similar to that seen in SMNP and typical of
areas under less direct influence of boundary layer pollution.
@ :,„,.,,
53
xTbv g§"©,,
' X I/ \ ^^ ^-^ • 1 i Campion
»0;e' -.
**^ i^
*
© f-7
• ,© 3
lynn*. ' M683 *"
©
K^ Longmoni
§i W«S
• +:
£.^
,, ;HI - -
D ' -*»' '*' A
e "O Boul0
"i | \ 1 . • •• ;>
Greasy
^'^1"*® "'
_
— ©
-
['""""f -«»nb»i CS'i
*"»"* LiUi
D.COOO
/
^ /
                                           ,
                                            ©,.
                                            •ar^, ""7
                                 Loui5,,ll.^- fc,_U TMSCr,,k
                     10km
                     j Evergreen
                     .' --   " i ijiniM!
                      M^lnV'  .
                            Tin Cup ^Ken C

                       ^Jk
     Figure 3-42    Terrain map showing the location of the AQS ozone monitoring site
                    in Rocky Mountain National Park, CO (black/green star) and the
                    Denver CSA (red dots) along with ozone monitoring sites used in
                    the Brodin et al. (2010) study (blue circles).  Elevations range from
                    approximately 1,600 m above sea level in Denver and Boulder to
                    3,528 m above sea level  at the highest mountainous site.
7
8
9
The three sites in California-one in San Bernardino National Forest (SBNF) and two in
Sequoia National Park (SENP)-had the highest distribution of 8-h daily max O3
concentrations of the selected rural focus area monitors included in Figure 3-39. The
     Draft - Do Not Cite or Quote
                            3-101
September 2011

-------
1
2
3
4
5
6
7
SBNF site had a warm-season 8-h daily max O3 concentration mean of 80 ppb and a
maximum of 137 ppb measured on July 1, 2007. This site is located in Crestline, CA, 90
km down-wind of Los Angeles in the San Bernardino Mountains. This site was included
in the Los Angeles CSA shown in Figure 3-27 (Site AE) and had the highest median 8-h
daily max O3 concentration of any AQS site in the Los Angeles CSA during this time
period (Figure 3-30). This site was also included in an analysis performed for the 2006
O3 AQCD where similarly high O3 concentrations were observed using 2004 year-round
hourly observations.
                   £p\.«
                   V437
                                                  @>
                                                     Cm>
                                       IT  *
                                        r«
                                                             m
    Figure 3-43   Terrain map showing the location of two AQS ozone monitoring
                   sites (black/green stars) in Sequoia National Park, CA. The lower
                   site (site ID = 061070009) is 560 m above sea level and the higher
                   site (site ID = 061070006) is 1,890 m above sea level.
    Draft - Do Not Cite or Quote
                           3-102
September 2011

-------
 1                   The two sites in SENP are located 9.7 km apart at contrasting elevations as is illustrated
 2                   in the terrain map in Figure 3-43. The correlation in 8-h daily max O3 between these two
 3                   sites was 0.86 and the COD was 0.09, which are within the range in correlations and
 4                   CODs for SMNP (Figure 3-41). The distribution of 8-h daily max O3 concentrations at
 5                   the lower elevation site (elevation = 560 m; site ID = 061070009) is shifted slightly
 6                   higher with a median of 76 ppb compared to the higher elevation site (elevation =
 7                   1,890 m; site ID = 061070006) with a median of 69 ppb. The lower elevation site is
 8                   located at the entrance to the park and is at a low enough elevation to be influenced by
 9                   boundary layer pollution coming  upwind from Fresno and the San Joaquin Valley. The
10                   higher elevation site is in the free troposphere above the planetary boundary layer and is
11                   less influenced by such pollution. This gives rise to a negative average surface-level
12                   elevation gradient of-0.5 ppb per 100 m elevation gain in SENP, illustrating the location-
13                   specific complexities inherent to high-altitude surface-level O3 concentrations.

14                   Since O3 produced from emissions in urban areas is transported to more rural downwind
15                   locations, elevated O3 concentrations can occur at considerable distances from urban
16                   centers. In  addition, major sources of O3  precursors such as highways, power plants,
17                   biomass combustion, and oil and  gas operations are commonly found in rural areas,
18                   adding to the O3 in these areas. Due to lower chemical scavenging in nonurban areas,  O3
19                   tends to persist longer in rural than in urban areas which tends to lead to higher
20                   cumulative exposures in rural areas influenced by anthropogenic precursor emissions.
21                   The persistently high O3  concentrations observed at many of these rural sites investigated
22                   here indicate that cumulative exposures for humans and vegetation in rural areas can be
23                   substantial and often higher than  cumulative exposures in urban areas.
             3.6.3   Temporal Variability
                     3.6.3.1    Multiyear Trends

24                   Nationally, O3 concentrations have declined over the last decade, as shown in Figure 3-
25                   44 from the 2010 National Air Quality Status and Trends report (U.S. EPA. 2010e). The
26                   majority of this decline occurred before 2004 with national average concentrations
27                   remaining relatively flat between 2004 and 2008. The large decreases in 2003 and 2004
28                   coincides with NOX emissions reductions resulting from implementation of the NOX
29                   State Implementation Plan (SIP) Call rule, which began in 2003 and was fully
30                   implemented in 2004. This rule  was designed to reduce NOX emissions from power
31                   plants and other large combustion sources in the eastern U.S. The reduction in NOX and
      Draft - Do Not Cite or Quote                      3-103                               September 2011

-------
 1
 2
O3 during the 2003-2004 timeframe is particularly evident in the eastern U.S. where the
NOX SIP Call was focused (U.S. EPA. 201 Oe).
n 19
U. I *-
— 0.1 -
E
a.
•& 0.08 -
o
'lo 0.06 -
U.
§ 0.04 -
c
o
0 0.02-
n -,

Average 1050 sites
90 percent of sites are below this line.
i
* tau^^H
	 -^___ 	 	 	 	 	 	
Current National Standard (revised 2008)
I
10 percent of sites are below this line.

















                             01     02     03      04     05     06      07
                                           2001 to 2008: 10% decrease
                                                            08
       Source: U.S. EPA (201 Oe)
      Figure 3-44    National 8-h ozone trends, 2001 -2008 (average of the annual fourth
                      highest 8-h daily max concentrations in ppm).
 4
 5
 6
 1
 8
 9
10
11

12
13
14
15
16
17
Weather can have a strong influence on O3 and O3 trends as well. The number of hot, dry
days can significantly alter the number of high-O3 days in any given year, even if O3-
forming emissions do not change. To better evaluate the progress and effectiveness of
emissions reduction programs, EPA uses a statistical model to estimate the influence of
atypical weather on O3 formation (U.S. EPA. 2010e). After adjusting for the influence of
weather, the downward trend in national 8 hours O3 concentrations between 2001 and
2008 increases slightly from an 8% reduction to an 11% reduction. These trends are
region-specific, with lower reductions (3%) in California and higher reductions (15%) in
eastern states over this same time period (U.S. EPA. 2010e).

Sites that showed the greatest reduction in O3 over this period were in or near the
following metropolitan areas:  Anderson, IN;  Chambersburg, PA; Chicago, IL; Cleveland,
OH; Houston, TX; Michigan City, IN; Milwaukee, WI; New York, NY; Racine, WI;
Watertown, NY;  and parts of Los Angeles, CA. Sites that showed an increase  in O3  over
this time period and had measured concentrations above the 2008 O3 standard9 during the
2006-2008 time period were located in or near the following metropolitan areas: Atlanta,
      9 On September 16, 2009, EPA announced it would reconsider the 2008 03 NAAQS, which, at the time, included primary and secondary standards of 0.075 ppm (8-h
      daily max).
      Draft - Do Not Cite or Quote
                              3-104
September 2011

-------
 1                   GA; Baton Rouge, LA; Birmingham, AL; Denver, CO; El Centre, CA; San Diego, CA;
 2                   Seattle, WA; and parts of Los Angeles, CA.

 3                   As noted in the 2006 O3 AQCD, trends in national parks and rural areas are similar to
 4                   nearby urban areas, reflecting the regional nature of O3 pollution. Therefore, caution
 5                   should be exercised in using trends calculated at national parks to infer contributions
 6                   from distant sources either inside or outside of North America because of the influence of
 7                   regional pollution (see Section 3.4 for a discussion of background O3 concentrations and
 8                   international transport). Trends in tropospheric O3 on a global scale have been monitored
 9                   around the world using ozonesondes, remote surface monitors, mountain top monitors,
10                   and satellites. Positive trends in O3 measurements in the free troposphere above western
11                   North America at altitudes of 3-8 km (above sea level) during April and May of 1995 to
12                   2008 were reported by Cooper et al.  (2010) and discussed in Section 3.4.1 as they relate
13                   to intercontinental transport. Note, however, that these results relate to O3 trends above
14                   ground level and not to surface O3. Other observations of global trends in the burden of
15                   tropospheric O3 as they relate to climate change are discussed in Chapter 10,
16                   Section 10.2.3.1.
                     3.6.3.2    Hourly Variations

17                   Ozone concentrations frequently possess a strong degree of diel variability resulting from
18                   daily patterns in temperature, sunlight, and precursor emissions. Other factors, such as the
19                   relative importance of transport versus local photochemical production and loss rates, the
20                   timing for entrainment of air from the nocturnal residual boundary layer, and the diurnal
21                   variability in mixing layer height also play a role in daily O3 patterns. The 2006 O3
22                   AQCD looked at composite urban diel variations from April to October 2000 to 2004 and
23                   found 1-h maxima to occur in mid-afternoon and 1-h minima to occur in early morning.
24                   On a national basis, however, there was a high degree of spread in these times and
25                   caution was raised in extrapolating results from one city to another in determining the
26                   time of day for O3 maxima and minima.

27                   Urban diel variability in O3 concentrations was investigated for the 20 focus cities listed
28                   in Table 3-7 using 1-h avg O3 data from AQS. The year-round data set described in
29                   Table 3-3 was used to compare diel patterns during cold months (October - April) and
30                   warm months (May - September) between 2007 and 2009. The warm-season data set,
31                   also described in Table 3-3, was used to compare weekday and weekend diel patterns.
32                   Figure 3-141 through Figure 3-145 in the supplemental material in Section 3.10.4 show
33                   these patterns for each of the 20 cities; examples for Atlanta, Boston  and Los Angeles are
34                   shown in Figure 3-45.
      Draft - Do Not Cite or Quote                      3-105                               September 2011

-------
                     Cold Months
                                           Warm Months
                                                                   Weekdays
                                                                                         Weekends
o
to
*-•
_2
<
            ISO -
          Q- 100 -
            50 -
             0 -
          0 days. 0 year-round sites
          • • • •  mean
          	  median
          <=> 5* 95»
                    no year-round data
                                       0 days. 0 year-round sites
                                           no year-round data
                                                              327 days, 11 warm-season sites
                                                                                    132 days, 11 warm-season sites
              00:00  06:00  1200  18:00 00:00 00:00  06:00  12:00 18:00  00:00 00:00  06:00  12:00  18:00  00:00 00:00 06:00 12:00  18:00 00:CC
            150 -
o  s
=  a
o  ~,
       &
            100 -
            50 -
             0 -
                 637 days, 3 year-round sites
              mean
              median
              5* 95
                        *
                                       459 days. 3 year-round sites
                                                              327 days, 21 warm-season sites
                                                                                    132 days, 21 warm-season sites
              00:00  06:00  12'00  18:00 00:00 00:00  06:00  12:00 18:00  00:00 00:00 06:00  12:00  (8:00  00:00 00:00 06:00 12:00  18:00 00:OC
       O)
       o
       Itl
       2- S
       40 ppb) in median 1-h O3 concentrations included Atlanta, Birmingham,
              Los Angeles, Phoenix, Pittsburgh, and Salt Lake City (Figure 3-141 through Figure 3-145
      Draft - Do Not Cite or Quote
                                              3-106
September 2011

-------
 1                   in Section 3.10.4). Cities with small daily swings (<25 ppb) in median 1-h O3
 2                   concentrations included Boston, Minneapolis, San Francisco and Seattle (Figure 3-141
 3                   through Figure 3-145 in Section 3.10.4). These results are very similar to those found in
 4                   the 2006 O3 AQCD where many of these same urban areas were investigated. This
 5                   supports the conclusions drawn in the AQCD that diel patterns in O3 have remained
 6                   stable over the last 20 years, with times of occurrence of the daily maxima varying by no
 7                   more than an hour from year to year.

 8                   Using the warm-season data, there was very little difference in the median diel profiles
 9                   for weekdays compared with weekends across all urban areas. This result stresses the
10                   complexity of O3 formation and the importance of meteorology, entrainment, biogenic
11                   precursor emissions, and transport in addition to anthropogenic precursor emissions.
12                   There was, however, a subtle deviation between weekdays and weekends in the lower
13                   percentiles (1st and 5th) of the distribution. The lower end of the distribution tended to be
14                   lower on weekdays relative to weekends. This is consistent with analyses in the 2006 O3
15                   AQCD and is a result of lower traffic volumes on weekends relative to weekdays, leading
16                   to less NO emissions and O3 titration on the weekends.

17                   Seasonal and site-to-site variations in diel patterns within a subset of the urban focus
18                   areas presented here were investigated in the 2006 O3 AQCD. In northern cities, there
19                   was substantial seasonal variability in the diel patterns with higher extreme values in the
20                   O3 distribution during the warm season than during the cold season. In southern cities,
21                   the seasonal differences in extreme O3 concentrations were much smaller, and some of
22                   the highest O3 concentrations in the Houston CSA were found outside of summer. The
23                   general pattern that emerged from investigating site-to-site variability within the urban
24                   areas was that peaks in 1-h avg O3 concentrations are higher and tend to occur later in the
25                   day at downwind sites relative to sites  located in the urban core. Differences between
26                   sites were not only related to the  distance between them, but also depend on the presence
27                   or absence of nearby O3 sources  or sinks.

28                   Rural diel variability in O3 concentrations was investigated for the six rural focus areas
29                   listed in Table 3-9 using  1-h avg  O3 data from AQS. As with the urban analysis, the year-
30                   round data set described in Table 3-3 was used to compare diel patterns during cold
31                   months (October - April) and warm months  (May - September) between 2007 and 2009.
32                   The warm-season data set, also described in Table 3-3, was used to compare weekday
33                   and weekend diel patterns. Figure 3-46 shows the diel patterns for each of the rural areas
34                   investigated.
      Draft - Do Not Cite or Quote                      3-107                                September 2011

-------

rondack SP, Ml
0;lpptj
S : :
5 o.
Cold Months
me _n
	 n t«3ivtn
— 	 _ — _____ 	 _ — — - — — ..



Warm Months
~-^^^^=^=^-^
-

-

Weekdays

— _ — --1 1™1™1 	 ™~ - — „
W*!*k«nd«
__„
— 	 ^ _ 	 ,_ -— ^
          oo 00 O6OO
                           oo oo OQ oo O6 oo l?oo seoo oo rxj no no
                                                                                 IROI> r>ri t"
   O
   a.
   CO _
            O days 0 year-round scless
                                0 days 0 year-round si(<
         0 -
           7    T    I    I    i   r    i    !    t    r    i
          OO OO OG OO  1? OO  IB DO  OO OO OO OO OO GO 1 3 OO S8 OO OO OO Ou CO
                                                          1 ,f uo  . 8 or-  00000000 OG oo 12 oo i o oo oo oc
      ~fL 1C>C'"
    >.   6U -
    5
    O
    g    o  1
    CO
                                                      days. !li warm-sotts
                                                                       1 3.i. days, 'J> wam<-si©ti3>
                           OO OO OO (ISO OB OO 1 2 GO 10 OO OO OO OO 130 O*> OO  12 OO  1W OO  OO OO OO IJQ Ofi OO 1 2 OO 10 Ot> CXJ OC
                                 7 days 1 year1 round s
                                                                       131 day». 1
                        cso  oft Oo oo ctt o. oo 1 r OQ 1 <", no on r j 01 i
                                                                              J Od 10 OO GO of
          OO DO OB !'.H! 1.? (JO  ^K f.ffl  iX> W3 OO i.K') £SH JKS 1? (>Q 1H f..Jl i'K) (X. OO iX5 Of. fX)  1 _'   fK) ffl3 OO (K) OH IX) 1 2 IMS 1H (JO
   .a ~
   O o

   S"
   CO
            629 days 2 year round sites
                                 9 ci-iays. 2 year round s
                                                                       132 d&y&, 2 warm season si.es
                                          16 00 00 00 00 00 0600  12 00  (6 >\  Ou OU C^ L.O
                                                                             1. oc> iaoo oo ex

                                                                             hour
Figure 3-46    Diel patterns in 1-h avg ozone for six rural focus areas between
                2007 and 2009 using the year-round data set for the cold
                month/warm month comparison (left half) and the warm-season
                data set for the weekday/weekend comparison (right half). Mt.
                Mitchell SP,  NC had no year-round monitors available for the cold
                month/warm month comparison.
Draft - Do Not Cite or Quote
3-108
September 2011

-------
 1                   There was considerable variability in the diel patterns observed in the six rural focus
 2                   areas. The eastern sites in ADSP, MMSP, and SMNP all exhibited a generally flat profile
 3                   with very little hourly variability in the median concentration and the upper percentiles.
 4                   In SMNP, there was some diel variability in the lower percentiles, with higher values
 5                   during the daylight hours in the warm season data. This behavior was not present in the
 6                   data coming from the two year-round monitors located at lower elevation sites (Sites C
 7                   and Site D; see map in Figure 3-40), however, possibly resulting  from differing impacts
 8                   from local sources within SMNP. For the western rural areas, there was a clear diel
 9                   pattern to the hourly O3 data with a peak in concentration in the afternoon similar to
10                   those seen in the urban areas in Figure 3-45 and Figure 3-141 through Figure 3-145 in
11                   Section 3.10.4. This was especially obvious at the SBNF site which sits 90 km east of
12                   Los Angeles in the San Bernardino Mountains at an elevation of  1,384 m.  This site was
13                   located here to monitor O3 transported downwind from major urban areas in the South
14                   Coast Air Basin. It had the highest 2007-2009 median 8-h daily max O3 concentration of
15                   any AQS site in the Los Angeles CSA (see Figure 3-30), and is clearly impacted by the
16                   upwind urban plume which has sufficient time and sunlight to form O3 from precursor
17                   emissions  and concentrate the O3 in the shallow boundary layer present at this elevation.

18                   As with the urban analysis, there was very little difference observed in the weekday and
19                   weekend diel profiles using the warm-season data, even down at the lower percentiles in
20                   the distribution. This is consistent with the regional nature of tropospheric O3. Using the
21                   year-round data, there was an upward shift in the distribution going from the cold months
22                   to the warm months, and in some instances the general shape of the distribution changed
23                   considerably as was seen in several urban sites.
             3.6.4   Associations with Co-pollutants

24                   Correlations between O3 and other criteria pollutants are discussed in this section. Since
25                   O3 is a secondary pollutant formed in the atmosphere from precursor emissions, it is not
26                   expected to be highly correlated with primary pollutants such as CO and NOX.
27                   Furthermore, O3 formation is strongly influenced by meteorology, entrainment, and
28                   transport of both O3 and O3 precursors, resulting in a broad range in correlations with
29                   other pollutants which can vary substantially with season.

30                   To investigate correlations with co-pollutants, 8-h daily max O3 from the year-round and
31                   warm-season data sets (Table 3-4 and Table 3-5) were compared with co-located 24-h
32                   avg CO, SO2, NO2, PM2.5 and PM10 obtained from AQS for 2007-2009. Figure 3-47 and
33                   Figure 3-48 contain copollutant box plots of the correlation between co-located monitors
34                   for the year-round data set and the warm-season data set, respectively.
      Draft - Do Not Cite or Quote                       3-109                                September 2011

-------
                                       Year-Round
          co-
          scx-
         NO, -
         PM, - -
               .0   -0.8   -0.6   -0.4   -0.2   0.0    0.2   0.4   0.6    0.8    1.0
                               Correlation with 8-h Daily Max O,
                          Winter
                  Spring
CO
so, .
NO .

PM:,j -

-4T^D— +- •
~-t— nzH-
i I LMMH
^^ j ^*^*
_^ 	 [-^— | 	 (....
,,j i n I L__.
H L_LL_J F
-
-


-
	 l_rnE^_f_.
-4 	 rm- —
• MltMJ 1 Si S !•§
••••H— H *n — f- ••
— "H — L 	 i 	 1 — H —
.0 -O.R -0.fi -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -1.0 -O.X -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 O.S I.I
Summer Fall
CO
SO, -
NO, .
PM,« -
PM.,.
. ..4^.-rr>^
,. ..(_! f l_^_
. _^ — HTH- •
-.-1 — | i __H—
— 4— C- h-H
-
-
-
.
-
>•>!- 1 t 1 i-m.
*"*! 	 i 1* 1 	 	 T
* *****] 	 1 T 	 T** *
. . — 1 — (— 7: — 1 — f—
... .1 — [_J J — (.
•• H — CZIIIZH^
             -1.0 -0.8 -0.6 -0.4 -02 00 02  04 06 OK  1.0  -1.0 -O.R -0.6 -0.4 -0.2 0.0  0.2 0.4 0.6 0.8  I.O
                               Correlation with 8-h Daily Max O,


Figure 3-47   Distribution of Pearson correlation coefficients for comparison of
              8-h daily max ozone from the year-round data set with co-located
              24-h avg CO, SO2, NO2, PM™ and PM2.s from AQS, 2007-2009 (top
              figure) with seasonal stratification (bottom four figures). Shown are
              the median (red line), mean (green star), inner-quartile range (box),
              5th and 95th percentiles (whiskers) and extremes (black circles).
Draft - Do Not Cite or Quote
3-110
September 2011

-------
                                                Warm-Season
               CO-
              so,_
              NO,_
             PM10-|
             PM2.5 -I
                   -1.0   -0,8    -0.6
                   -0.4    -0.2    0.0     0.2    0.4
                    Correlation with 8-h Daily Max O,
0.6
      Figure 3-48   Distribution of Pearson correlation coefficients for comparison of
                     8-h daily max ozone from the warm-season (May-Sept) data set with
                     co-located 24-h avg CO, SO2, NO2, PM10 and PM2.s from AQS, 2007-
                     2009. Shown are the median (red line), mean (green star), inner-
                     quartile range (box), 5th and 95th percentiles (whiskers),  and
                     extremes (black circles).
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
The year-round 8-h daily max O3 data (Figure 3-47) had a very wide range in correlations
with all the 24-h avg co-pollutants. A clearer pattern emerged when the data were
stratified by season (bottom four plots in Figure 3-47) with mostly negative correlations
in the winter and mostly positive correlations in the summer for all co-pollutants. In
summer, the IQR in correlations is positive for all co-pollutants. However, the median
seasonal correlations are still modest at best with the highest positive correlation at 0.52
for PM2 5 in the summer and the highest negative correlation at -0.38 for PM2 5 in the
winter. Spring and fall lie in between with spring having a slightly narrower distribution
than fall for all co-pollutants. The warm-season 8-h daily max O3 data (Figure 3-48)
shows a very similar distribution to the summer stratification of the year-round data due
to their overlap in time periods (May-Sept and Jun-Aug, respectively).

The seasonal fluctuations in correlations present in Figure 3-47 result in part from the
mixture of primary and secondary sources for the co-pollutants. For example, O3 is a
secondary pollutant whereas PM2 5 has both primary and secondary origins and these two
pollutants show the largest summertime/wintertime swing in correlation distributions.
      Draft - Do Not Cite or Quote
                             3-111
           September 2011

-------
 1                  This situation arises because the secondary component to PM2 5 is larger during the
 2                  summer and is formed in conditions conducive to secondary O3 formation. This results in
 3                  positive correlations between O3 and PM2 5 during the summer. During the winter,
 4                  photochemical production of O3 is much smaller than during summer and O3 comes
 5                  mainly from aloft, i.e., the free troposphere (see Section 3.4 for further details). In
 6                  addition, concentrations of PM25 are much lower aloft. On relatively clean days, this can
 7                  lead to high concentrations of O3 and lower concentrations of primary pollutants such as
 8                  PM2 5 or NO. On relatively dirty days with elevated NO and PM2 5, the  intruding O3 is
 9                  readily titrated by NO in the boundary layer. These processes result in negative
10                  correlations between O3 and PM25 during the winter.
          3.7    Chapter Summary

11                  This section contains a summary of the major topics included in this chapter on the
12                  atmospheric chemistry and ambient concentrations of tropospheric O3 and other related
13                  photochemical oxidants. This chapter has built upon information previously reported in
14                  the 2006 O3 AQCD and includes updated material on: (1) physical and chemical
15                  processes of O3 formation and removal; (2) atmospheric modeling; (3) policy relevant
16                  background concentrations; (4) monitoring techniques and networks; and (5) ambient
17                  concentrations.
            3.7.1    Physical and Chemical Processes

18                   Ozone in the troposphere is a secondary pollutant; it is formed by photochemical
19                   reactions of precursor gases and is not directly emitted from specific sources. Ozone and
20                   other oxidants, such as peroxyacetyl nitrate and hydrogen peroxide form in polluted areas
21                   by atmospheric reactions involving two main classes of precursor pollutants: VOCs and
22                   NOX. Carbon monoxide is also important for O3 formation in polluted areas and in the
23                   remote troposphere. The formation of O3, other oxidants, and oxidation products from
24                   these precursors is a complex, nonlinear function of many factors including: (1) the
25                   intensity and spectral distribution of sunlight; (2) atmospheric mixing; (3) concentrations
26                   of precursors in the ambient air and the rates of chemical reactions of these precursors;
27                   and (4) processing on cloud and aerosol particles.

28                   Ozone is present not only in polluted urban atmospheres  but throughout the troposphere,
29                   even in remote areas of the globe. The same basic processes involving sunlight-driven
30                   reactions of NOX, VOCs and CO contribute to O3 formation throughout the troposphere.
31                   These processes also lead to the formation of other photochemical products, such as
      Draft - Do Not Cite or Quote                     3-112                               September 2011

-------
 1                   PAN, nitric acid, and sulfuric acid, and to other compounds, such as formaldehyde and
 2                   other carbonyl compounds. In urban areas, NOX, VOCs and CO are all important for O3
 3                   formation. In nonurban vegetated areas, biogenic VOCs emitted from vegetation tend to
 4                   be the most important precursor to O3 formation. In the remote troposphere, methane -
 5                   structurally the simplest VOC - and CO are the main carbon-containing precursors to O3
 6                   formation. In the troposphere, O3 is subsequently lost through a number of gas phase
 7                   reactions as well as deposition to surfaces.

 8                   Convective processes and small scale turbulence transport O3 and other pollutants both
 9                   upward and downward throughout the planetary boundary layer and the free troposphere.
10                   In many areas of the U.S., O3 and its precursors can be transported over long distances,
11                   aided by vertical mixing. The transport of pollutants  downwind of major urban centers is
12                   characterized by the development of urban plumes. Meteorological conditions, small-
13                   scale circulation patterns, localized chemistry, and mountain barriers can influence
14                   mixing on a smaller scale, resulting in frequent heterogeneous O3 concentrations across
15                   an individual urban area.

16                   Emissions of O3 precursor compounds (NOX, VOCs, and CO) can be divided into
17                   anthropogenic and natural source categories. Natural sources can be further divided into
18                   biogenic from vegetation, microbes, and animals, and abiotic from biomass burning,
19                   lightning, and geogenic sources. However, the distinction between natural sources and
20                   anthropogenic sources is often difficult to make in practice, as human activities affect
21                   directly or indirectly emissions from what would have been considered natural sources
22                   during the preindustrial era. The magnitudes of O3 precursor sources are strongly
23                   location- and time-dependent and so average emission estimates should not be used to
24                   apportion sources of exposure.
            3.7.2   Atmospheric Modeling

25                   CTMs have been widely used to compute the interactions among atmospheric pollutants
26                   and their transformation products, and the transport and deposition of pollutants. They
27                   have also been widely used to improve our basic understanding of atmospheric chemical
28                   processes and to develop control strategies. The domains of CTMs extend from a few
29                   hundred kilometers on a side to the entire globe.
30                   Most major regional (i.e., sub-continental) scale air-related modeling efforts at EPA rely
31                   on the CMAQ modeling system. CMAQ's horizontal domain typically extends over
32                   North America with efforts underway to extend it over the entire Northern Hemisphere.
33                   The upper boundary for CMAQ is typically set at 100 hPa, which is located on average at
34                   about 16-km altitude. CMAQ is most often driven by the MM5 mesoscale meteorological

      Draft - Do Not Cite or Quote                      3-113                               September 2011

-------
 1                  model, though it may be driven by other meteorological models including the WRF
 2                  model and the RAMS. Other major air quality systems used for regional scale
 3                  applications include CAMx and WRF/Chem.

 4                  Fine scale resolution is necessary to resolve features which can affect pollutant
 5                  concentrations such as urban heat island circulation; sea breezes; mountain and valley
 6                  breezes; and the nocturnal low-level jet. Horizontal domains are typically modeled by
 7                  nesting  a finer grid model within a larger domain model of coarser resolution. Caution
 8                  must be exercised in using nested models because certain parameterizations like those for
 9                  convection might be valid on a relatively coarse grid scale but may not be valid on finer
10                  scales and because incompatibilities can occur at the model boundaries. The use of finer
11                  resolution in CTMs will require advanced parameterizations of meteorological processes
12                  such as  boundary layer fluxes, deep convection, and clouds, and necessitate finer-scale
13                  inventories of land use, source locations, and emission inventories.

14                  Because of the large number of chemical species and reactions that are involved in the
15                  oxidation of realistic mixtures of anthropogenic and biogenic hydrocarbons, condensed
16                  mechanisms must be used to simplify atmospheric models. These mechanisms can be
17                  tested by comparison with smog chamber data. However, the existing chemical
18                  mechanisms often neglect many important processes such as the formation and
19                  subsequent reactions of long-lived carbonyl compounds, the  incorporation of the most
20                  recent information about intermediate compounds, and heterogeneous reactions involving
21                  cloud droplets and aerosol particles. As a result, models such as CMAQ have had
22                  difficulties with capturing the regional nature of O3 episodes, in part because of
23                  uncertainty in the chemical pathways converting NOX to HNO3 and recycling of NOX.

24                  The largest errors in photochemical modeling are still thought to arise from the
25                  meteorological and emissions inputs to the model. Algorithms must be used for
26                  simulating meteorological processes that occur on spatial scales smaller than the model's
27                  grid spacing and for calculating the dependence of emissions on meteorology and time.
28                  Significant errors in emissions can occur if inappropriate assumptions are used in these
29                  parameterizations.

3 0                  The performance of CTMs must be evaluated by comparison with field data as part of a
31                  cycle of model evaluations and subsequent improvements. Discrepancies between model
32                  predictions and observations can be used to point out gaps in current understanding of
33                  atmospheric chemistry and to spur improvements in parameterizations of atmospheric
34                  chemical and physical processes.
      Draft - Do Not Cite or Quote                      3-114                                September 2011

-------
            3.7.3   Background Concentrations

 1                   Because the mean tropospheric lifetime of O3 is 30-35 days, O3 can be transported from
 2                   continent to continent and around the globe in the Northern Hemisphere. The degree of
 3                   influence from intercontinental transport varies greatly by location and time. High
 4                   elevation sites are most susceptible to the intercontinental transport of pollution,
 5                   particularly during spring. However, the chemistry involving O3 formation is nonlinear,
 6                   thereby complicating the task of isolating the influence of intercontinental transport of O3
 7                   and O3 precursors on U.S. air quality. Careful consideration of fine-scale spatial and
 8                   temporal variation is necessary to appropriately characterize the impact of
 9                   intercontinental transport on tropospheric O3.

10                   Since North American background (i.e., O3 concentrations that would exist in the absence
11                   of anthropogenic emissions from the U.S. Canada and Mexico) is a construct that cannot
12                   be directly measured, the range of background O3 concentrations are estimated using
13                   chemistry transport models (CTMs). The 2006 O3 AQCD (U.S. EPA. 2006b) provided
14                   regional estimates of North American background O3 concentrations based on a coarse
15                   resolution (2°x2.5°, or -200 kmx200 km) GEOS-Chem model. For the current
16                   assessment, updated results from a finer resolution (0.5°x0.667°, or ~50 kmxSO km)
17                   GEOS-Chem model were used. In general, the GEOS-Chem predictions tend to show
18                   smaller disagreement with observations at the high-altitude sites than at the low-altitude
19                   sites. Overall agreement between model results for the base case and measurements is
20                   within a few ppb for spring-summer means in the Northeast and the Southeast,  except in
21                   and around Florida where the base case over predicts O3 by 10 ppb at one site,  at least. In
22                   the Upper Midwest, the model predictions are within 5 ppb of measurements, the same is
23                   true for sites in the intermountain West and at lower elevations sites in the West including
24                   California. However, the model under predicts O3 by 10 ppb at the Yosemite site. These
25                   results suggest that the model is capable of calculating March to August mean O3 to
26                   within ~ 5 ppb at most (26 out of 28) sites chosen. Currently, there are no simulations of
27                   North American background concentrations available in the literature apart from those
28                   using GEOS-Chem alone. However, as noted in , the 2006 O3 AQCD, an ensemble
29                   approach as is done in many other applications of atmospheric models is to be preferred.

30                   The GEOS-Chem calculations presented here represent the latest results documented in
31                   the literature. However, all models undergo continuous updating of inputs,
32                   parameterizations of physical and chemical processes, and inputs and improvements in
33                   model resolution. Inputs that might be considered most relevant include emissions
34                   inventories, chemical reactions and meteorological fields. This leads to uncertainty in
35                   model predictions in part because there is typically a lag between updated information for
36                   these above inputs. Examples might include updated emissions for year specific shipping,
      Draft - Do Not Cite or Quote                      3-115                               September 2011

-------
 1                  wildfires and updates to the 2005 NEI; updates to the chemistry of isoprene and multi-
 2                  phase processes, including those affecting the abundance of halogens; and updates to
 3                  species such as methane. To the extent that results from an updated model become
 4                  available, they will be presented and used to help inform NAAQS setting.
            3.7.4   Monitoring

 5                  The FRM for O3 measurement is the CLM and is based on the detection of
 6                  chemiluminescence resulting from the reaction of O3 with ethylene gas. Almost all of the
 7                  SLAMS that reported data to AQS from 2005 to 2009 used UV absorption photometer
 8                  FEMs and greater than 96% of O3 monitors met precision and bias goals during this
 9                  period.

10                  State and local monitoring agencies operate O3 monitors at various locations depending
11                  on the area size and typical peak concentrations (expressed in percentages below, or near
12                  the O3 NAAQS). SLAMS make up the ambient air quality monitoring sites that are
13                  primarily needed for NAAQS comparisons and include PAMS, NCore, and all other State
14                  or locally-operated stations except for the monitors designated as SPMs.

15                  In 2010, there were 1250 SLAMS O3 monitors reporting values to the EPA AQS
16                  database. Since O3 levels decrease significantly in the colder parts of the year in many
17                  areas, O3 is required to be monitored at SLAMS monitoring sites only during the "ozone
18                  season" which varies by state. PAMS provides more comprehensive data on O3 in areas
19                  classified as serious, severe, or extreme nonattainment for O 3. There were a total of 119
20                  PAMS reporting values to the EPA AQS database in 2009. NCore is a new
21                  multi-pollutant monitoring network currently being implemented to meet multiple
22                  monitoring objectives. Each state is required to operate at least one NCore site and the
23                  network will consist of about 60 urban and 20 rural sites nationwide.

24                  CASTNET is a regional monitoring network established to assess trends in acidic
25                  deposition and also provides concentration measurements of O3. CASTNET  O3 monitors
26                  operate year round and are primarily located in rural areas. At the beginning of 2010,
27                  there were 80 CASTNET sites located in, or near, rural areas. The NPS also operates a
28                  POMS network. The POMS couples the small, low-power O3 monitor with a data logger,
29                  meteorological measurements, and solar power in a self contained system for monitoring
30                  in remote locations. Twenty NPS POMS reported O3 data to AQS in 2010. A map of the
31                  current and proposed rural NCore sites, along with the CASTNET, and the NPS POMS
32                  sites was shown in Figure 3-18.
      Draft - Do Not Cite or Quote                      3-116                                September 2011

-------
 1                   Satellite observations for O3 are growing as a resource for many purposes, including
 2                   model evaluation, assessing emissions reductions, pollutant transport, and air quality
 3                   management. Satellite remote sensing instruments do not directly measure the
 4                   composition of the atmosphere. Satellite retrievals are conducted using the solar
 5                   backscatter or thermal infrared emission spectra and a variety of algorithms. Most
 6                   satellite measurement systems have been developed for stratospheric measurement of the
 7                   total O3 column. Mathematical techniques have been developed and must be applied to
 8                   derive information from these systems about tropospheric O3.
             3.7.5   Ambient Concentrations

 9                   Ozone is the only photochemical oxidant other than NO2 that is routinely monitored and
10                   for which a comprehensive database exists. Other photochemical oxidants are typically
11                   only measured during special field studies. Therefore, the concentration analyses
12                   contained in this chapter have been limited to widely available  O3 data obtained directly
13                   from AQS for the period from 2007 to 2009.

14                   Most continuous O3 monitors report hourly average concentrations to AQS. This data can
15                   be used as reported (1-h avg), or reported as a daily metric such as: (1) the average of the
16                   hourly observations over a 24-h period (24-h avg); (2) the maximum 8-h running average
17                   of the hourly observations occurring in a 24-h period (8-h daily max), or (3) the
18                   maximum hourly observation occurring in a 24-h period (1-h daily max). The median 24-
19                   h avg, 8-h daily max, and  1-h daily max O3 concentrations across all U.S. sites reporting
20                   data to AQS between 2007 and 2009 were 29, 40, and 44 ppb, respectively. Representing
21                   the upper end of the distribution, the 99th percentiles of these same metrics across all
22                   sites were 60, 80, and 94 ppb, respectively. Correlations between these different
23                   averaging time metrics generated from the same hourly observations in the 3-year nation-
24                   wide data set were shown  in Figure 3-19. The  8-h daily max and  1-h daily max metrics
25                   were highly correlated (median R = 0.97, IQR = 0.96-0.98) while comparisons with the
26                   24-h avg metric were lower (e.g., median R =  0.83, IQR = 0.78-0.88 for comparison
27                   between the 24-h avg and  the 1-h daily max). The ratio and correlation between these
28                   metrics, however, can be very site-specific.

29                   To investigate urban-scale O3 variability, 20 focus cities were selected for closer
30                   analysis; these cities were  selected based on their importance in O3 epidemiologic studies
31                   and on their geographic distribution across the U.S. Several of these cities had relatively
32                   little spatial variability in 8-h daily max O3 concentrations (e.g., inter-monitor
33                   correlations ranging from  0.61 to 0.96 in Atlanta) while other cities exhibited
34                   considerably more variability in O3 concentrations (e.g., inter-monitor correlations
      Draft - Do Not Cite or Quote                      3-117                                September 2011

-------
 1                   ranging from -0.06 to 0.97 for Los Angeles). The negative and near-zero correlations in
 2                   Los Angeles were between monitors with a relatively large separation distance (>150
 3                   km), but even some of the closer monitor pairs were not very highly correlated. Similar to
 4                   the correlation, the coefficient of divergence was found to be highly dependent on the
 5                   urban area under investigation. As a result, caution should be observed in using data from
 6                   a sparse network of ambient O3 monitors to  approximate community-scale exposures.

 7                   To investigate rural-focused O3 variability using AQS data, all monitors located within
 8                   six rural monitoring areas were examined. These rural monitoring sites are impacted by
 9                   transport of O3 or O3  precursors from upwind urban areas, and by local anthropogenic
10                   emissions within the rural areas such as emissions from motor vehicles, power
11                   generation, biomass combustion, or oil and gas operations. As a result, monitoring data
12                   from these rural locations are not unaffected by anthropogenic emissions. The rural area
13                   investigated with the largest number of available AQS monitors was Great Smoky
14                   Mountain National Park in NC and TN where the median warm-season 8-h daily max O3
15                   concentration ranged from 47 ppb at the lowest elevation  site (elevation = 564 m;  site ID
16                   = 470090102) to 60 ppb at the highest elevation site (elevation = 2,021 m; site ID  =
17                   471550102), with correlations between the 5 sites ranging from 0.78 to 0.92 and CODs
18                   ranging from 0.04 to 0.16. A host of factors  may contribute to variations observed at
19                   these rural sites, including proximity to local O3  precursor emissions, variations in
20                   boundary-layer influences, meteorology and stratospheric intrusion as a function of
21                   elevation, and differences in wind patterns and transport behavior due to local
22                   topography. Expanded analyses of O3 concentrations measured using the more rural-
23                   focused CASTNET monitoring network are  included in Chapter 9.

24                   Since O3 produced from emissions in urban  areas is transported to more rural downwind
25                   locations, elevated O3 concentrations can occur at considerable distances from urban
26                   centers. In addition, major sources of O3 precursors such as  highways, power plants,
27                   biomass combustion, and oil and gas operations are commonly found in rural areas,
28                   adding to the O3 in these areas. Due to lower chemical scavenging in nonurban areas, O3
29                   tends to persist longer in rural than in urban  areas which tends to lead to higher
30                   cumulative exposures in rural areas influenced by anthropogenic precursor emissions.
31                   The persistently high O3 concentrations observed at many of these rural sites investigated
32                   here indicate that cumulative exposures for humans and vegetation in rural areas can be
33                   substantial and often higher than cumulative exposures in urban areas.

34                   According to the 2010 National Air Quality  Status  and Trends report (U.S. EPA. 20106).
35                   O3 concentrations have declined steadily over the last decade; with the majority of this
36                   decline occurring before 2004. A noticeable  decrease in O3 between 2003 and 2004
37                   coincides with NOX emissions reductions resulting from implementation of the NOX SIP
      Draft - Do Not Cite or Quote                      3-118                                September 2011

-------
 1                   Call rule, which began in 2003 and was fully implemented in 2004. This rule was
 2                   designed to reduce NOX emissions from power plants and other large combustion sources
 3                   in the eastern U.S. As noted in the 2006 O3 AQCD, trends  in national parks and rural
 4                   areas are similar to nearby urban areas, reflecting the regional nature of O3 pollution.
 5                   However, caution should be exercised in using trends calculated at national parks to infer
 6                   contributions from distant sources either inside or outside of North America because of
 7                   the influence of regional pollution. Global scale observations have, indeed, indicated a
 8                   general rise in O3 by a factor of 2 or more as discussed in Chapter 10, Section  10.2.3.1.

 9                   Urban O3 concentrations show a strong degree of diel variability resulting from daily
10                   patterns in temperature, sunlight, and precursor emissions.  Other factors, such  as the
11                   relative importance of transport versus local photochemical production and loss rates, the
12                   timing for entrainment of air from the nocturnal residual boundary layer, and the diurnal
13                   variability  in mixing layer height also play a role in daily O3 patterns. Urban diel
14                   variations investigated in this assessment show no substantial change in patterns since the
15                   2006 O3 AQCD. The 1-h max concentrations tend to occur in mid-afternoon and 1-h min
16                   concentrations tend to occur in early morning, with more pronounced peaks in the warm
17                   months relative to the cold months. There is city-to-city variability in these times,
18                   however, and caution is raised in extrapolating results from one city to another in
19                   determining the time of day for O3 maxima and minima.

20                   Rural O3 concentrations show a varying degree of diel variability depending on their
21                   location relative to larger urban areas. Three rural areas investigated in the east showed
22                   relatively little diel variability, reflecting the regional nature of O3 in the east. In contrast,
23                   three rural  areas investigated in the west did display diel variability  resulting from their
24                   proximity to fresh urban emissions. These six areas investigated were selected as
25                   illustrative examples and do not represent all rural areas in the U.S.

26                   Since O3 is a secondary pollutant formed in the atmosphere from precursor emissions, it
27                   is not expected to be highly correlated with primary pollutants such as CO and NOX.
28                   Furthermore, O3 formation is strongly influenced by meteorology, entrainment, and
29                   transport of both O3 and O3 precursors, resulting in a broad range in correlations with
30                   other pollutants which can vary substantially with season. In the copollutant analyses
31                   shown in Figure 3-45, the year-round 8-h daily max O3  data exhibited a very wide range
32                   in correlations with all the 24-h avg co-pollutants. A clearer pattern emerged when the
33                   data are stratified by season with mostly negative correlations in the winter and mostly
34                   positive correlations in the summer for all co-pollutants. The median seasonal
35                   correlations are modest at best with the highest positive correlation at 0.52 for  PM2 5 in
36                   the summer and the highest negative correlation at -0.38 for PM2 5 in the winter.
      Draft - Do Not Cite or Quote                       3-119                                 September 2011

-------
 1                 Therefore, statistical analyses that may be sensitive to correlations between co-pollutants
 2                 need to take seasonality into consideration, particularly when O3 is being investigated.
         3.8    Supplemental Ozone Model Predictions from the Literature
           3.8.1   Time Series of GEOS-Chem Model Predictions and Observations at
                   Selected CASTNET Sites

 3                 This section contains comparisons between GEOS-Chem predictions of 8-h daily max O3
 4                 concentrations with observations for 2006 from Zhang et al. (In Press). Further details on
 5                 these predictions can be found in Section 3.4.3. Figures 3-49 through 3-55 show GEOS-
 6                 Chem predictions for the base model (i.e., model including all anthropogenic and natural
 7                 sources; labeled as GEOS-Chem in the figure) and the North American background
 8                 model (i.e., model including natural sources everywhere in the world and anthropogenic
 9                 sources outside the U.S., Canada, and Mexico; labeled as NA background in the figure)
10                 along with measurements obtained from selected CASTNET sites (labeled as
11                 Measurement in the figure). Figures 3-56 a-b show a comparison of GEOS-Chem output
12                 with measurements at Mt. Bachelor, OR from March-August, 2006. Figure 3-57 shows a
13                 comparison of vertical profiles (mean ± 1 standard deviation) calculated by GEOS-Chem
14                 with ozonesondes launched at Trinidad Head and Boulder, CO.
     Draft - Do Not Cite or Quote                    3-120                              September 2011

-------
                      Connecticut Hill, NY (42N, 76W, 501m)
        Acadia NP. ME (44N. 68W, 158m)
                   Huntington Wildlife Forest, NY (43N, 74W, 502m)   Kane Exp. Forest, PA (41N, 78W. 622m)
                     Mar   Apr   May   Jun   Jul  Aug   Mar  Apr  May  Jun  Jul  Aug

 Source: Zhang et al. (In Press).


Figure 3-49   Comparison  of time series of measurements of daily maximum 8-
              hour average ozone concentrations at four CASTNET sites in the
              Northeast with GEOS-Chem predictions for the base case and for
              the North American background case in 2006.
                      Coffeeville, MS (34M, B9W, 134m)
       Sand Mountain, AL (34N, 85W, 352m)
                           Measurement  GEOS-Chem .
                                   NA background -
                     Georgia Station, GA (33N, 84W, 270m)
       Indian River Lagoon, FL (27N, SOW, 2m)
                    Mar   Apr   May  Jun  Jul  Aug    Mar   Apr  May  Jun   Juf   Aug
 Source: Zhang et al. (In Press).
Figure 3-50   Comparison of time series of measurements of daily maximum
              8-hour average ozone concentrations at four CASTNET sites in the
              Southeast with GEOS-Chem predictions for the base case and for
              the North American background case in 2006.
Draft - Do Not Cite or Quote
3-121
September 2011

-------
                       Ann Arbor, Ml (42N, 83W, 267m)
                                  PerMnstown, Wl (45N, SOW, 472m)
                 100

                 80


                 60


                 40


                 20

                  0



                 100

                 80


                 60


                 40


                 20
              Measurement  GEOS-Chem
                     NA hackground -
         Untonville. Ml (43N, 83W, 201m)
                                   Bondville. IL (40N. 88W, 212rn)
                                             49.6 54.8 26.5
                    Mar  Apr  May   Jun   Jul   Aug

 Source: Zhang et al. (In Press).
                                              Mar   Apr   May   Jun   Jul   Aug
Figure 3-51   Comparison of time series of measurements of daily maximum 8-
              hour average ozone concentrations at four CASTNET sites in the
              Upper Midwest with GEOS-Chem predictions for the base case and
              for the North American background case in 2006.
                   Yellowstone, WY (45N, 110W, 2400m)
                                   Centennial, WY (41N, 106W, 3178m)
           1
100

 80

 60

 40

 20
                                  NA background -
                 56,4 50.7 38.4
                    Pinedale, WY (43N, 110W. 2388m)
                                   Rocky Mtn, CO (40N, 106W, 2743m)
                                              58.1 60.0 42.0
                  Mar  Apr   May   Jun   Jul   Aug   Mar   Apr   May  Jun   Jul   Aug
 Source: Zhang et al. (In Press).
Figure 3-52   Comparison of time series of measurements of daily maximum 8-
              hour average ozone concentrations at CASTNET sites in the
              Intermountain West with GEOS-Chem predictions for the base case
              and the North American background case in 2006.
Draft - Do Not Cite or Quote
                          3-122
September 2011

-------
                       Gothic, CO (39N, 107W, 2926m)
                               Great Basin NP, NV (39N. 114W. 2060m)
                100

                 80

                 60


                 40

                 20
                100
           Measurement  GEOS-Chem
                   NA background -
                    56.4 56.1 42,2
                                              56.7 56.3 40.9
                    Mesa Verde NP, CO (37N, 10BW, 2165m)     Canyonlands NP, UT (38N, 110W, 1B09m)
                    Mar  Apr  May   Jun  Jui   Aug   Mar   Apr  May  Jun   Jul   Aug
 Source: Zhang et al. (In Press).
Figure 3-53    Comparison of time series of measurements of daily maximum 8-
               hour average ozone concentrations at CASTNET sites in the
               Intermountain West with GEOS-Chem predictions for the base case
               and the North American background case in 2006.
                    Grand Canyon NP, AZ (36N, 112W, 2073m)
                                Big Bend NP, TX (29N, 103W, 1052m)

100

80

60




20

 0
                            Measurement   GEOS-Chem
                                   NA background -
                    5B.8 S8.9 43,2
                     Petrified Forest, AZ (35N, 110W, 1723m)
                               Mount Rainier NP, WA (47N, 122W, 415m)
                    Mar  Apr  May   Jun  Jul  Aug   Mar   Apr  May  Jun   Jul   Aug
 Source: Zhang et al. (In Press).
Figure 3-54    Comparison of time series of measurements of daily maximum 8-
               hour average ozone concentrations at CASTNET sites in the West
               with GEOS-Chem predictions for the base case and the North
               American background case in 2006.
Draft - Do Not Cite or Quote
                        3-123
September 2011

-------
                     Yosemrte NP, CA (38N, 120W. 1605m)
                   Converse Station, CA (34N, 117W, 1837m:)
                120

                100

                 80

                 60

                 40

                 20
                120

                100

              I  80
              Q.
              7  60

              |  40

                 20

                 0
Measurement  GEOS-Chem
        NA background
                    64.0 54,0 29,0
                                               68.871,438.1
                      Death Valley, CA (37N, 117W, 125m)
                    Trinidad Head, CA (41N, 124W, 107m)
                  Mar   Apr  May  Jun   Jul


 Source: Zhang et al. (In Press).
           Aug
                Mar   Apr  May  Jun   Jul   Aug
Figure 3-55   Comparison of time series of measurements of daily maximum 8-
               hour average ozone concentrations at monitoring sites in
               Californiawith GEOS-Chem predictions for the base case and the
               North American background case in 2006.
            Mt Bachelor (44.0N, 121.7W, 2700m)
                         Trinidad Head (41.05N, -124.15W, 107m)
                                                     observation (41.9 ppbv)'
                                                     GEOS-Chem (38.3 ppbv)
       Mar
                    May    Jun
                      Date
                                Jul
           Aug
                                                 Mar
                                                        Apr
May     Jun
   Date
                                                                           Jul
                                                     Aug
 Source: Zhang et al. (In Press).
Figure 3-56   Comparison of daily maximum 8-h average ozone predicted using
               GEOS-Chem at 0.5°x0.67° and 2°x2.5° (left figure only) resolution
               with measurements at Mount Bachelor, OR (left) and Trinidad Head,
               CA (right) from March to August 2006.
Draft - Do Not Cite or Quote
             3-124
           September 2011

-------
                               12
                               10
                                8
                                6
                                4
                                2
                                0
                               12
                               10
                                8
                                6
                                4
                                2
                                0
                                         April
                                        August
                            n=13
                                               n=30
                                               n=31
                                    30  60  90  120 150  30  60  90  120 15C
                                      Ozone (ppbv)          Ozone (ppbv)
       Source: Zhang et al. (In Press).
       The letter 'n' refers to the number of ozonesonde profiles, and the model was sampled on the same days as the ozonesonde
      launches. As can be seen from the figure, variability in both model and measurements increases with altitude, but variability in the
      model results is much smaller at both sites, at high altitudes than seen in the observations.

      Figure 3-57    Comparison of monthly mean ± 1 standard deviation ozone
                     calculated GEOS-Chem (in red) with ozonesondes (in black) at
                     Trinidad Head and Boulder,  CO during April and August 2006.

         3.9   Supplemental Ozone Model  Predictions Using the Latest
                Release of GEOS-Chem
            3.9.1   Introduction
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
This section summarizes work that is currently underway on O3 modeling over the U.S.
using the latest release of GEOS-Chem (v9-01-01). This release includes several changes
to the emissions inputs supplied to the model. Two of the updates that are likely to affect
the simulated O3 concentrations are 1) a correction for the yield of isoprene nitrates that
increases the lifetime of NOX and may result in greater ozone production, and 2) a
correction for lightning NOX emissions which may also increase O3. A full list of updates
is provided in the release notes for the current version of the model (Harvard University.
201 Ib). For the current analysis, GEOS-Chem was applied using nested grids with
anthropogenic emissions updated for each model year from the 2005 NEI inventory.
Zhang et al. (In Press) recently completed a similar study for North America, using the
same grid configuration, but an earlier version of the GEOS-Chem model with 2005
emissions inputs.
      Draft - Do Not Cite or Quote
                             3-125
September 2011

-------
 1                  This summary includes an overview of the GEOS-Chem model application and
 2                  evaluation methods, an assessment of model performance for the base case simulations,
 3                  modeling results for several background O3 simulations, and a discussion of the attributes
 4                  and limitations of the modeling methods and results. The full report is available online
 5                  (U.S. EPA. 201 Ic).
            3.9.2   GEOS-Chem Model Application

 6                   For the current analysis, GEOS-Chem is being applied using nested grids with 2°x2.5°
 7                   horizontal resolution at the global scale and 0.5°x0.667° horizontal resolution over North
 8                   America (140°-40°W, 10°-70°N). In addition, a coarser resolution grid (4°x5°) is also
 9                   being used for the start-up simulation period—an annual simulation period run to spin up
10                   the model and to ensure a reasonable representation of long-range (global) transport. The
11                   modeling domain includes 47 vertical  layers that increase in thickness with height above
12                   ground; the top of the modeling domain is at approximately 80 km. The GEOS-Chem
13                   model is being applied for the years 2006, 2007 and 2008 with a one-year spin-up period
14                   (2005).

15                   As noted in Section 3.3, the GEOS-Chem model is driven by assimilated meteorological
16                   observations from GEOS and is able to represent long-range transport as well as
17                   stratospheric-tropospheric exchange processes. For this analysis, meteorological inputs
18                   for the four simulation years were provided by the Harvard University Atmospheric
19                   Chemistry Modeling (ACM)  Group (Harvard University, 201 Ic). The version of the
20                   model used for this study utilizes the GEOS-5 data product from NASA's Global
21                   Modeling and Assimilation Office (GMAO). Data used by GEOS-Chem include surface
22                   albedo, parameters defining properties at the surface including moisture content and land
23                   type, various precipitation measures, cloud fraction, heat and radiation fluxes, PEL
24                   thickness, air temperature, tropopause pressure, ground (skin) temperature, U and V wind
25                   components, friction velocity, specific humidity and others.

26                   Emission inputs for 2005 were also provided by Harvard's ACM Group (Harvard
27                   University. 2011 a). They include both anthropogenic and biogenic emissions, and
28                   account for fossil fuel combustion and usage, biomass burning, biofuel burning, and
29                   natural aerosol emissions. Examples of categories of emissions included are aircraft
30                   emissions, shipping emissions, and soil and fertilizer NOX emissions. The emissions also
31                   include estimates for NOX generated by lightning. Various sources of data provide global
32                   coverage with the more reliable and highly resolved emissions data sources taking
33                   precedence. For example, the 2005 NEI inventory is used in order to enhance the
34                   emissions estimates over the United States. Temporal resolution for emissions varies
      Draft - Do Not Cite or Quote                      3-126                                September 2011

-------
 1                  depending on data source from annual to seasonal. Within the model, emissions are
 2                  introduced at hourly intervals. Anthropogenic emissions were projected to each simulated
 3                  year (2006, 2007, and 2008) using scale factors developed for each region of the world
 4                  based on available information. More detailed information on the scale factors for
 5                  anthropogenic emissions and links to additional documents can be found on the Harvard
 6                  University ACM Group website (Harvard University. 201 Id).

 7                  All other inputs for the application of GEOS-Chem were provided by Harvard's ACM
 8                  Group, including Total Ozone Mapping Spectrometer (TOMS) data, surface UV albedo,
 9                  dry deposition coefficients, and land use codes. Note that roughness lengths and terrain
10                  heights are included in the GEOS-5 data. Other files define the chemical mechanism and
11                  provide data for calculating photolysis rates. The leaf area index used for ozone dry
12                  deposition was based on data from the Moderate Resolution Imaging Spectroradiometer
13                  (MODIS). Use of MODIS data is expected to result in less ozone dry deposition and
14                  higher ozone concentrations compared to the use of Advanced Very High Resolution
15                  Radiometer (AVHRR) derived values. Additional operating parameters for GEOS-Chem
16                  are provided in the full report (U.S. EPA. 20 lie).
            3.9.3   Model Scenarios

17                   Table 3-10 summarizes the different model scenarios considered for this analysis. In
18                   addition to the Base Case which modeled the existing atmosphere for 2006, 2007, and
19                   2008 with all natural and anthropogenic emissions turned on, three background air
20                   quality scenarios were considered to explore different impacts on U.S. O3 concentrations.
21                   They included 1) a U.S. Background scenario with all anthropogenic emissions in the
22                   U.S. turned off; 2) a North American Background scenario with all anthropogenic
23                   emissions in North America (U.S., Canada, and Mexico) turned off (equivalent to the
24                   previously used definition introduced in Section 3.4); and 3) a Natural Background
25                   scenario with all anthropogenic emissions across the globe turned off. Methane
26                   concentrations used in the GEOS-Chem model were adjusted to reflect the different
27                   emission scenarios with zonal average concentrations listed in Table 3-2 of the full report
28                   (U.S. EPA. 201 Ic).
      Draft - Do Not Cite or Quote                      3-127                                September 2011

-------
      Table 3-10     Summary of GEOS-Chem model scenarios
        Model Scenario
               Anthropogenic
             Emissions from the
                   U.S.
  Anthropogenic       Anthropogenic
  Emissions from    Emissions from the
Canada and Mexico    Rest of the Globe
Natural Emissions
   Everywhere
Base Case
U.S. Background
N.A. Background8
Natural Background
On
Off
Off
Off
On
On
Off
Off
On
On
On
Off
On
On
On
On
      'North American (N.A.) background is equivalent to the previously used definition
 2
 3
 4
 5
 6
 7
 8
 9
10
11

12
13
14
15
16
17
18
19
20
21
22
3.9.4   Model Performance Evaluation

        Model evaluation was performed on the Base Case for the three simulation years. This
        evaluation focused primarily on the ability of the GEOS-Chem model to replicate
        observed O3 and other pollutant concentrations for the entire U.S., selected subregions of
        the U.S., and at individual sites representing key areas in the U.S. that have been
        identified as important for characterizing background O3 concentrations (In Press). The
        evaluation also included a qualitative assessment of how well the stratospheric-
        tropospheric exchange is being simulated and an examination of the effects of interannual
        variability in meteorology on the Base Case results. A wide range of statistical and
        graphical analyses relating to model performance are presented in the full report (U.S.
        EPA. 20lie). Key findings from the model performance evaluation include:

            •  Model performance for all species was consistent among the three modeled
              years  (2006, 2007, and 2008)
            •  Ozone concentrations  were  overestimated by the GEOS-Chem model for all
              three years and nearly all regions and time periods considered in this analysis.
            •  Overestimation was greater for 24-h avg O3 than for  8-h daily max O3.
            •  Model performance for O3 varied by season and by subregion.
            •  As expected, given the grid resolution, O3 concentrations were better
              represented for the more rural CASTNET sites compared to the  more urban
              AQS sites.
            •  Based on comparison with the CASTNET data, the bias and error statistics for
              8-h daily max O3 suggested reasonably good performance. Overestimation of
      Draft - Do Not Cite or Quote
                                     3-128
                                         September 2011

-------
 1                        O3 during the summer and autumn months was most prevalent in the eastern
 2                        U.S. (Central States, Great Lakes, Southeast, and Northeast subregions). For
 3                        the western subregions, there was less overestimation of O3 and overall good
 4                        agreement with the observations. The best agreement with the observations
 5                        was achieved for the Intermountain West subregion.
 6                      •  Site-specific model performance for 8-h daily max O3 concentration (again for
 7                        the CASTNET sites) indicated that for some sites, especially those in the west,
 8                        the annual variation in O3 concentration is very well replicated. For most of
 9                        the subregions, model performance improves with site elevation.
10                      •  The spatial and temporal variability of NO2, SO2, CO data from the AQS
11                        network are not well characterized. NO2 concentrations are underestimated for
12                        all subregions and time periods. SO2 is underestimated in the Intermountain
13                        West, where observed values are low and overestimated elsewhere, including
14                        in the East where observed values are higher. Simulated CO is only weakly
15                        correlated with the observed concentrations, and concentrations are
16                        underestimated for all regions and time periods. The modeled values exhibit
17                        less seasonal variation than the observed values. These results are perhaps
18                        understandable, especially for NO2 and CO, given the coarse grid resolution
19                        and the probable strong response of AQS monitors to local emissions.
20                      •  Model performance for 24-h avg PM25 is mixed. For many of the regions and
21                        time periods, the bias and error statistics indicate good model performance.
22                        However, the seasonal variation in PM2 5 that is characterized by higher
23                        concentrations during the summer months is not replicated by the model.
24                      •  Overall, model performance for dry deposition is quite good. Dry deposition
25                        of O3 is underestimated by the model, which could contribute to
26                        overestimation of the O3 concentrations.
27                      •  Comparison of simulated and observed O3 profiles for five case-study periods
28                        (a total of 12 days) and several locations within the western U.S. gave mixed
29                        results. For some cases, the simulated vertical profiles showed poor agreement
30                        with the observations. For other cases the results were more promising in that
31                        although the detailed vertical structure of observed profiles was not well
32                        simulated, the model results showed layers of high O3 in the middle
33                        troposphere, consistent with observations.
      Draft - Do Not Cite or Quote                     3-129                              September 2011

-------
           Frequency Distributions of MDA8 of 03 Cone in March-August 2006 - Low Elevation Sites (<1.5 km)
            10   20    30    40    50    60    70    80    90    100   110   120   130   140
     0.1
    0.08 -
    0.06 -
    0.04 -
    0.02 -
           Frequency Distributions of MDA8 of 03 Cone in March-August 2006 - High Elevation Sites (> 1.5 km)
            10   20    30    40    50    60    70    80    90    100   110   120   130   140
Figure 3-58   Frequency distributions of 8-hr daily max ozone concentration from
              March -August 2006 for low-elevation (<1.5 km; top panel) and
              high-elevation (>1.5 km; bottom panel) CASTNET sites. Observed
              values are in black; modeled Base Case values are in red (labeled
              GEOS-Chem),  modeled North American  Background values are in
              blue (labeled PRB), and modeled Natural Background values are in
              Green (labeled NB).
Draft - Do Not Cite or Quote
3-130
September 2011

-------
 1                  Figure 3-58 includes a comparison between the frequency distribution of observed 8-h
 2                  daily max O3 from March - August, 2006 at CASTNET sites with the Base Case model
 3                  results (labeled GEOS-Chem in the figure) at corresponding sites and times. This figure
 4                  illustrates the overestimation of the Base Case model relative to observations at low
 5                  elevation (< 1.5 km) CASTNET sites, and the general agreement between the model and
 6                  observations at high elevation CASTNET sites (> 1.5 km). Further details on the sites
 7                  used in the model performance evaluation and additional model evaluation results
 8                  including site-specific case studies are included in the full report (U.S. EPA. 201 Ic).
            3.9.5   Model Results

 9                   Figure 3-59 displays the mean 8-hr daily max O3 concentration for the Base Case (left
10                   panel) and North American Background scenario (right panel), based on all three
11                   simulation years. For the current atmosphere scenario, mean 8-hr daily max O3
12                   concentrations range from 21.3 to 82.6 ppb within the modeling grid. For the North
13                   American background scenario, mean 8-hr daily max O3 concentration ranges from 17.4
14                   to 42.9 ppb. The annual average North American background 8-hr daily max O3
15                   concentration forthe entire U.S. is estimated to be 31.0 ppb. This value varies
16                   geographically; for the western U.S., the estimated range is 35.5-38.9 ppb and for the
17                   eastern U.S., the estimated range is 27.6-31.2 ppb. The highest estimated North
18                   American Background concentrations do not necessarily occur in the areas with the
19                   highest modeled Base Case concentrations.

20                   The estimated North American Background 8-hr daily max O3 concentrations vary by
21                   season, as illustrated in Figure 3-60. For most areas within the domain, the highest
22                   concentrations tend to occur during the  spring (March-May). Fiigh values also occur
23                   during the summer (June-August) in the western U.S., especially over the more
24                   mountainous regions. For the spring months, the average North American Background 8-
25                   hr daily max O3 concentration is estimated to be 33.2 ppb for the entire U.S.; it ranges
26                   from 37.8-41.7 ppb forthe western U.S. and from 29.3-32.6 ppb forthe eastern U.S. For
27                   the summer months, the average North American Background 8-hr daily max O3
28                   concentration is estimated to be 30.0 ppb forthe entire U.S.; it ranges from 33.9-40.4 ppb
29                   for the western U.S. and from 22.9-34.2 ppb for the eastern U.S.

30                   Figure 3-58 includes the frequency distribution of the 8-h daily max O3 concentrations
31                   from March - August, 2006 at CASTNET sites  modeled in the North American
32                   Background scenario and the Natural Background scenario for direct comparison with
33                   observations and the Base Case model results (labeled GEOS-Chem in the  figure).
      Draft - Do Not Cite or Quote                      3-131                               September 2011

-------
 1
 2
Additional analyses of the North American Background model results including case
studies at selected CASTNET sites are included in the full report (U.S. EPA. 201 lc).
          I KVr • 1 Oionr (ppb>
                    r • |-i | M | Ljiijuiu .LnaViu MdffttTk|?tf*|fctMiM4Mti-i-i4T
                                                       r>:V?:t. I Osone (ppb)
          Current Atmo«phoro - :i Yewr AveraRe
      Figure 3-59    Mean 8-hr daily max Os concentration (ppb) for the Base Case (left
                      panel) and North American Background scenario (right panel),
                      based on the 2006, 2007 and 2008 simulation period.
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
All model scenarios showed considerable spatial and temporal variability across the U.S.
and none can be represented by a single value. For the Base Case scenario, mean 8-h
daily max O3 concentrations ranged from 21.3 to 82.6 ppb within the modeling grid. For
the North American Background scenario, 8-h daily max O3 concentrations ranged from
17.4 to 42.9 ppb. For the U.S. Background scenario, 84i daily max O3 concentrations
ranged from 19.9 to 73.4 ppb. Compared to the North American Background, the
simulated U.S. Background values are higher throughout the domain including over the
continental U.S. This increase is attributable to Canadian, Mexican and offshore
emissions. For the Natural Background scenario, the concentrations are substantially
lower with 84i daily max O3 concentrations from 12.8 to 30.7 ppb. Within the U.S., the
simulated Natural Background concentrations were very low along the northeast corridor
and the highest concentrations were found over Colorado. Additional model results
including a seasonal and geographic analysis for the U.S. Background and Natural
Background scenarios are included in the full report (U.S. EPA. 201 Ic).
      Draft - Do Not Cite or Quote
                              3-132
September 2011

-------
                                            L,
      ti.tu i_li- 1. liltlll-LfJlJtJ LiljlL] LlLi_I^L_ljJ!:PiJto'ii 1 i. , liLl.t ; Hit
    north America Backgrouna - wtnler - 3 Veor
                                             America Background - Spring - 3 Yc«r A
                                          North Atncrlcu Background - toll - II Tfeitr ,
Figure 3-60   Mean 8-hr daily max Os concentration (ppb) for the North American
              Background scenario during winter (Dec-Feb, upper right panel),
              spring (Mar-May, upper right panel), summer (Jun-Aug, lower left
              panel), and fall (Sep-Nov, lower right panel), based on the 2006,
              2007 and 2008 simulation period.
Draft - Do Not Cite or Quote
3-133
September 2011

-------
            3.9.6   Model Attributes and Limitations

 1                  The GEOS-Chem tool allows the estimation of contributions to background O3
 2                  concentrations from a variety of sources and source regions, and the non-linear effects
 3                  associated with the removal of emissions from selected sources and source regions has
 4                  been shown to be small. The following is a summary of the attributes and limitations of
 5                  the GEOS-Chem results included here:

 6                      •  The grid resolution required for a global simulation is not expected to resolve
 7                        regional- and urban-scale O3 production in the U.S. and elsewhere and thus
 8                        may under- or over-estimate the anthropogenic contribution to long-range O3
 9                        transport.
10                      •  The GEOS-Chem chemical mechanism includes a relatively detailed
11                        representation of the reactions and species involved in the production of O3 in
12                        the atmosphere. Details of the  mechanism are presented in Evans et al.
13                        (2003a). The mechanism includes hundreds of reactions and more than 80
14                        species. In a comparison of chemical mechanisms, Emmerson (2009) noted
15                        that the GEOS-Chem mechanism (and the other mechanisms evaluated in their
16                        paper) should be able to represent the atmospheric chemistry in the
17                        troposphere. Nevertheless, all chemical mechanisms suffer from the necessity
18                        to limit the number of reactions and species to a finite set rather than the many
19                        thousands of reactions and species actually taking part in the chemistry of the
20                        troposphere. Perhaps even more important, limitations in grid resolution of the
21                        model can limit the model's  ability to properly represent the relative
22                        proportions of species. This limitation could result in alterations in the
23                        estimates of O3 production rates compared to what would be simulated with
24                        higher grid resolution. Good performance at monitors will not guarantee that
25                        alterations in the chemical mix (e.g., by removing a category of emissions)
26                        will produce the correct response in O3 production.  Hence, the chemical
27                        mechanism and the interaction of the chemical mechanism with grid
28                        resolution must be considered  to be potential sources of uncertainty in the
29                        model results.
30                      •  Emissions estimates for GEOS-Chem are based on data available to the model
31                        developers. These data are more readily available for some parts of the world
32                        (e.g., the U.S. and Europe) than others (e.g., developing nations). The
33                        magnitude and distribution of emissions are therefore sources of uncertainty in
34                        the model runs, and this uncertainty may be difficult to quantify for many
35                        parts of the world. Although the chemical mechanism in GEOS-Chem
36                        includes a wide range of hydrocarbons, the number  of emitted species
      Draft - Do Not Cite or Quote                     3-134                               September 2011

-------
 1                        included in the model is much smaller. The speciation of emissions into
 2                        constituent hydrocarbons is difficult even for domestic U.S. emissions, where
 3                        relatively robust data are available for making these estimates. For parts of the
 4                        world where data are lacking, even greater uncertainty is present. The use of a
 5                        more limited set of emitted species is likely appropriate given that a more
 6                        detailed speciation of hydrocarbons would necessarily involve some guess
 7                        work. Uncertainties in the speciation of hydrocarbons introduce another
 8                        uncertainty into the GEOS-Chem simulation results.
 9                      • Considering the scale and resolution of the modeling domain,
10                        parameterizations of small-scale processes, such as boundary layer ventilation
11                        and downward mixing of free tropospheric and surface air are key sources of
12                        uncertainty in any global model application, including this application of
13                        GEOS-Chem.
14                      • Finally, it is difficult to fully evaluate model performance, given the grid
15                        resolution and the available data. In particular, it is difficult to confirm that the
16                        model reliably simulates the vertical distribution (and transport) of O3 aloft as
17                        well as the various processes, such as vertical mixing within the troposphere
18                        and stratospheric O3 intrusion, that influence background O3 concentrations.
            3.9.7   Summary of Modeling Results

19                  The assessment of model performance for this analysis reveals consistent overestimation
20                  of O3 concentrations throughout the U.S., but especially in the eastern U.S. In the
21                  western U.S. and at high elevations, the model performance was much better with Base
22                  Case model estimates matching well with observations from multiple CASTNET sites.
23                  Three different definitions of background were considered including North American
24                  Background, U.S. Background, and Natural Background. The Base Case and all three
25                  background scenarios showed considerable spatial and temporal variability across the
26                  U.S. The estimated  8-h daily max O3 concentrations ranged from 17.4 to 42.9 ppb for the
27                  North American Background scenario, from  19.9 to 73.4 ppb for the U.S. Background
28                  scenario, and from 12.8 to 30.7 ppb for the Natural Background scenario.
      Draft - Do Not Cite or Quote                      3-135                               September 2011

-------
         3.10  Supplemental Figures of Observed Ambient Ozone
                Concentrations
           3.10.1  Ozone Monitor Maps for the Urban Focus Cities

 1                 This section contains supplemental maps showing the location of O3 monitors reporting
 2                 to AQS for each of the 20 urban focus cities introduced in Section 3.6.2.1. The monitors
 3                 are delineated in the maps as year-round or warm-season based on their inclusion in the
 4                 year-round data set and the warm-season data set discussed in Section 3.6.2.1. The maps
 5                 also include the CSA/CBSA boundary selected for monitor inclusion, the location of
 6                 urban areas and water bodies, the major roadway network, as well as the population
 7                 gravity center based on the entire CSA/CBSA and the individual focus city boundaries.
 8                 Population gravity center is calculated from the average longitude and latitude values for
 9                 the input census tract centroids and represents the mean center of the population in a
10                 given area. Census tract centroids are weighted by their population during this
11                 calculation.
      Draft - Do Not Cite or Quote                    3-136                             September 2011

-------
                                                11  30     aft KJonebn
Figure 3-61   Map of the Atlanta CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Figure 3-62   Map of the Baltimore CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-137
September 2011

-------
                Mwiflw lo

                0 Wfr'i

                * YtV

                • "iff -
                  IntMUita HqtmMf*
                  omwigruni cs*
Figure 3-63   Map of the Birmingham CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Figure 3-64   Map of the Boston CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-138
September 2011

-------
                     i
                         B  !9   49
Figure 3-65
             Map of the Chicago CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Figure 3-66   Map of the Dallas CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
                                    3-139
September 2011

-------
                                  0   20  «
Figure 3-67   Map of the Denver CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Figure 3-68   Map of the Detroit CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-140
September 2011

-------
                                          \
Figure 3-69   Map of the Houston CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
                              0   50   100
Figure 3-70   Map of the Los Angeles CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-141
September 2011

-------
Figure 3-71   Map of the Minneapolis CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
               0   »  X     too
Figure 3-72   Map of the New York CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-142
September 2011

-------
Figure 3-73   Map of the Philadelphia CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
                                              0  IS  3D    BQK
Figure 3-74   Map of the Phoenix CBSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-143
September 2011

-------
                                                     Legend

                                                     Monitor Loc

                                                     O V*«m MMOn Uontori


                                                     • (~Tf hiiol Population Gravity Cantor

                                                  0   12 5  2!i
Figure 3-75   Map of the Pittsburgh CSA including ozone monitor locations,
              population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-144
September 2011

-------
                                   0  30  00
Figure 3-76   Map of the Salt Lake City CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
               legend
Figure 3-77   Map of the San Antonio CBSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-145
September 2011

-------
Figure 3-78   Map of the San Francisco CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
                             o   s
Figure 3-79   Map of the Seattle CSA including ozone monitor locations,
             population gravity centers, urban areas, and major roadways.
Draft - Do Not Cite or Quote
3-146
September 2011

-------
                       0    M    40
     Figure 3-80   Map of the St. Louis CSA including ozone monitor locations,
                    population gravity centers, urban areas, and major roadways.
1
2
3
4
5
6
7
3.10.2  Ozone Concentration Box Plots for the Urban Focus Cities

        This section contains box plots depicting the distribution of 2007-2009 warm-season 8-h
        daily max O3 data from each individual monitor in the 20 urban focus cities introduced in
        Section 3.6.2.1. Monitor information including the AQS site id, the years containing
        qualifying data between 2007 and 2009, and the number of 8-h daily max O3
        observations included in the data set are listed next to the box plot. Statistics including
        the mean, standard deviation (SD), median and inner quartile range (IQR) are also shown
        for each monitor with the site letter corresponding to the sites listed in the figures above.
     Draft - Do Not Cite or Quote
                                    3-147
September 2011

-------
                                    Atlanta CSA
Site ID
131210055
130890002
131350002
1X670003
132470001
1X970004
131130001
131510002
130770002
1X850001
132230003
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-08
07-09
07-09
07-09
07-09
Key
N
450
452
446
459
450
455
306
459
455
458
455
c
^ ra
To 1
r-j E
-I •
Mean
53
52
52
51
51
52
52
51
47
47
50
median

SD
17
18
16
16
18
15
15
17
16
13
14
1
o|E
1
Median IQR
54 22
52 23
52 18
52 22
51 22
53 22
52 20
51 22
47 19
47 17
50 21
h-
I---
to
•H
Site
A-
B-
c-
D-
E-
F-
G-
H-
J """
K-
C
, , , , I , , , , I , , ,
=- 	 I * I 	 •>
i- — 1 — -I _ ^
;--[y^-/

[""*. fl i""'
i f I H
. ' , * . ' H
> ,.;.ypL^
> 50 100 1f
03 (ppb)
Figure 3-81    Site information, statistics and box plots for 8-h daily max ozone
               from AQS monitors meeting the warm-season data set inclusion
               criteria within the Atlanta CSA.
                                    Baltimore CSA
        Site ID
       245100054
       240053001
       240051007
       240330030
       240251001
       240030014
       240130001
       240313001
       110010025
       110010041
       110010043
       240259001
       240338003
       510130020
       510595001
       515100009
       510591005
       240210037
       510590030
       510590018
       511071005
       240090011
       510590005
       240170010
       511530009
       511790001
       510690010
       510610002
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
08-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09

Key
£
N
454
456
459
445
450
459
459
292
453
459
459
458
452
459
459
456
432
458
459
459
456
439
459
456
453
459
459
456

C
Is i
-\ •
Mean
42
51
46
50
54
53
50
49
49
50
51
52
51
51
50
47
52
51
52
51
52
51
49
51
49
48
46
44

c
to
 i *
i- | 	
i :

^ -j i
i- 	 | 	 f
i- 	 : ^
s. , , , - 1 i
> " - - - •! j
i 	 j 	 i
! 	 [J

:- 	 -| |
;- 	 [ i
•, 	 |~j
v 	 ] *
'r---\ $]
$• — i 4 *
h — ( 4 '

~~~t
^j- 	 i
3 " * " * ^
	 }• - • — •(

<| | — -f
— i
	 r 	 •
^j — ^
"] — -i
j. ..^
	 } - -H
» j ...<
ZZ1 ""*

	 | 	 -i
^J 	 <
i] J 	 -i
> |- - - - <*

i__J 	 -E
[ ] 	 ^
[ | 	 4
~} 	 -J
[ \ 	 4

^j 	 t
D — ^
	 -5


-A
-B
-C
-D
	 E
-F
-G
-H
- 1
- J
-K
~» I
-M
-N
-0
-P
-Q
-R
S
_T
-U
-V
- W
-X
-Y
-Z
- AA
- AB

If _ I 0 50 100 150
1?

"sb
h-
\"
"to
03

03 (ppb)
Figure 3-82   Site information, statistics and box plots for 8-h daily max ozone
               from AQS monitors meeting the warm-season data set inclusion
               criteria within the Baltimore CSA.
Draft - Do Not Cite or Quote
3-148
September 2011

-------
                              Birmingham CSA
Site ID
010730023
010731003
010736002
010732006
011170004
010731010
010731005
010735002
010735003
010731009
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
£«
* —
N
450
459
459
455
459
452
459
459
456
457
c
£ TO
« «j

1
SD
15
15
15
16
16
14
16
13
15
15
li
O'S
;
Median IQR
48 21
44 22
49 20
48 23
49 23
46 20
46 21
47 18
48 21
46 21
Tb
r~-
I-
^
*f>
o>
_ _ 4
Site
A-
B-
C-
D-
C **,
F-
G-
H-
I-
J-
C
i i f i J i t i t i 1 i t t i
>---\ f f---H
' 1 	 fl—f^
^HE?.:r
J----I f !----<
I---T » I-----1
-A
-B
-c
-D
~ E
-F
-G
-H
-I
-J
) 50 100 150
03 (ppb)
Figure 3-83   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Birmingham CSA.
                                 Boston CSA
Site ID
250250042
250250041
250092006
250213003
250171102
250170009
250095005
330111011
250270024
250094004
440071010
250270015
3301 10020
330150016
3301 15001
330150014
250051002
440030002
330131007
440090007
330012004


Years
07-09
08-09
07-09
07-09
07-09
07-09
07-09
07-09
09
07-08
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
*« ~"
>• 	
N
459
306
459
459
457
439
459
457
153
305
453
458
455
458
459
459
459
458
459
459
459
c

o> ~ < A- B- c- D E- C _ G- H- I - J- K- L- M - N - o- P - Q- £3 _ s- T- u ,,,,!,::, ^ - - [_j!lj °' ^H )» | •! '' - "C~i* 1 ' i - - (3fjr[] t •r - - \ |» | •! :- - -rntf -i •- - -[ ^ i -( i.-rTpT-----i H-qb--^ >H3«I] < > - • PT{r~> 1 >--ITFl < >--n»n < .-00--^ :- --rTFl ^ •r-n3-----< >-^CD""< >---rti~\ < .--rg^--H . '--i2?r"' -A -B -C - D -E r- P -G -H .- 1 - J -K ^L i-M - N -0 - P -Q -R i- S -T -U s • • • • i 0 50 100 150 03 (ppb) Figure 3-84 Site information, statistics and box plots for 8-h daily max ozone from AQS monitors meeting the warm-season data set inclusion criteria within the Boston CSA. Draft - Do Not Cite or Quote 3-149 September 2011


-------
Chicago CSA
Site ID
170314002
170311003
170310076
170310042
170310072
170310064
170436001
170310001
170314007
170311601
170310032
170317002
170314201
180890030
180892008
170890005
180890022
171110001
181270024
170971002
170971007
550590019
181270026
171971011
180910005
180910010


Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
^> ^
h - 	
N
458
452
458
412
459
459
459
459
458
457
450
450
611
440
451
459
455
458
456
459
459
457
453
458
453
456
rt
E
•
Mean
39
44
44
45
42
41
39
46
39
48
46
43
42
45
45
44
42
43
46
41
46
47
43
42
42
44
c
TJ
£
I
SD
13
13
14
14
12
13
12
14
13
14
13
13
13
15
13
13
13
12
14
13
13
14
13
12
12
13
»§
§8
o;£

Median IQR Site , , . , i , , , , i , , , .
38
43
44
44
42
40
39
46
37
47
45
42
41
44
44
42
41
42
44
39
46
45
42
41
41
43
c
i-
\ - -
18
17
18
17
17
18
16
19
18
19
17
17
17
19
18
16
15
15
17
18
18
19
18
15
15
17

-.?
A-
R —
c-
D-
E-
F-
G-
H "•
| -
J-
K-
L-
M-
N -
0 -
P-
Q-
R -
S -
T-
U-
v-
w-
X-
Y-
z-
i .... j jT | • . - 1
'' ' ' ~\ P 1 	 ^
}----! t ~| 	 <
t - - -Lji j 	 .1
•r - - j ^ [ - - -i
:- - - - K ; - - H
i- 	 [_|,j---i
j,,,£^J^] 	 <
^ " -f E*' 1 	 J-
i I '
|_J> 	 | 1
P"~|» -
cm
FFi -'
i j» i ,
\ j H
{ 1* '
i > \
'"""]*""!
> nn
> 1 !•
\ji_ i - '
r~lr~T


-A
i-B
i- C
i-D
^ E
hF
-G
-H
i- 1
- J
>- K
-L
-M
-0
^ P
-Q
p— R
- S
^ J
^ u
- V
-w
^ x
- Y
•-Z
0 50 100 150
03 (ppb)
Figure 3-85   Site information, statistics and box plots for 8-h daily max ozone
              from AQS monitors meeting the warm-season data set inclusion
              criteria within the Chicago CSA.
                                    Dallas CSA
        Site ID
       481130069
       481130075
       481130087
       484393009
       484393011
       480850005
       481390016
       484391002
       483970001
       481210034
       484390075
       482570005
       481211032
       482510003
       481391044
       482311006
       483670081
       482210001
Years
07-08
07-09
07-09
07-09
07-09

07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
08-09
07-09
07-09
07-09

Key
£
1- 	
N
279
456
429
459
457

456
455
458
449
456
459
459
459
459
306
459
459
459

C
q 1

i_»
Mean
41
48
47
48
46

52
43
46
47
52
52
47
50
47
47
43
48
44

C
nj
TZJ
E

1

SD
14
15
16
16
15
1 e
I O
14
14
16
13
15
16
12
13
15
12
12
14
15

fl
sS
tJ.c
™^ji_
*
Median IQR
38
46
44
46
44
CA
ou
51
42
44
47
50
50
45
49
45
45
42
47
41


"ib

f--

22
21
24
23
22

21
22
23
21
22
24
18
19
22
19
18
22
22


Th
Oi
Site
A-
B-
/-• 	
r\ M
C ™

—
G-
H -
I -
J-
K-
L-
M-
N -
0-
P-
Q-
R-
S

, , , , i , , , , i , , , ,
:- - ^ !•'!--- - 1
:- - - -I |» | 	 i
I----I K [ 	 H
t • - [—— y— [ . . . . . ^
; - - 1 (• ]- - 	 i

\.\ j f /...,
i - - 1 |»; |--- - -!
; - - 1 fc ^ 	 H
i---| y^.__H
k---[~i~)» 	 1 	 <
i- . . {^£^j 	 <
•"""CTtT}"""^
;----[ fr ^ 	 ^
!- - - 1 (^ F 	 ',
r--\ |» |----i
r - -| ^; t-- - H
•- , -| t t 	 4
' " "l_J*i_J" 	 ^
1 	 1 	 1 	 1 f~**"J™ 	 , 	 , 	 , 	 ! 	 1 	 ! 	 , 	 , 	 , 	

-A
-B
- C
-D
-E

-G
-H
- 1
- J
-K
m» 1
-M
-N
-0
-P
-Q
-R
S

0 50 100 150

03 (ppb)
Figure 3-86   Site information, statistics and box plots for 8-h daily max ozone
              from AQS monitors meeting the warm-season data set inclusion
              criteria within the Dallas CSA.
Draft - Do Not Cite or Quote
3-150
September 2011

-------
                                Denver CSA
Site ID
080310025
080310002
080310014
080013001
080590002
080590005
080050002
08059001 1
080350004
080590006
080590013
080130011
080137001
080137002
081230009
Years
08-09
07
07-09
07-09
07-09
07-09
07,09
07-09
07-09
07-09
09
07-09
07
07
07-09
N
299
153
450
441
459
456
306
459
456
457
150
453
152
142
451
Mean
49
39
51
55
54
55
54
56
58
60
50
56
42
56
55
SD
12
10
12
11
12
12
11
12
11
12
9
12
10
11
11
Key
In "
^ 	 \

mean
*

median
I

f
g;E
j
1
s
r-
(--

0)
SD
12
10
12
11
12
12
11
12
11
12
9
12
10
11
11
Median IQR Site
51 16
41 13
52 15
57 13
56 16
56 16
55 14
57 15
58 14
59 15
50 10
56 14
42 12
56 13
56 14
ni
A-
B-
C-
D-
F-
p „
H-
j
1
J-
K-
L-
M-
N-
o-
C
i , , , , i
----- LSLJ-H
,;H:
A:;;HK^
) 50 100 1£
                                                                      A
                                                                      B
                                                                      C
                                                                      D
                                                                      •E
                                                                      F

                                                                     -H
                                                                     _. I
                                                                      J
                                                                      K
                                                                      L
                                                                      M
                                                                      N
                                                                      0
                                                  03 (ppb)
Figure 3-87   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Denver CSA.
                                Detroit CSA
Site ID
261250001
261630019
260991003
261630001
261610008
260990009
260490021
261470005
260492001
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
£
m
N
459
456
452
459
459
459
458
459
455
*8 1
-1 •
Mean
46
47
47
42
45
46
44
43
45
c
1
SD
14
15
15
13
13
15
13
15
14
|
oE

Median IQR Site
46 18
46 19
46 18
41 16
44 17
45 18
44 19
41 19
45 18
e
r-
I —
S
A-
B-
c-
D-
C __
F-
H -
I -
C
iitt
I-----] 	 J
|. 	 .| *
>~L_fe,
".-.a
:---S
r-—\^t
5
I , , , , I

>- — <
[• — -:

— 1- -s
F
-A
-B
-C
-D
_ C
-F
-G
-H
-I
50 100 150
03 (ppb)
Figure 3-88   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Detroit CSA.
Draft - Do Not Cite or Quote
3-151
September 2011

-------
                                     Houston CSA
     Site ID
    482010075
    482010070
    482010066
    482010047
    482010055
    482010416
    482010046
    482011035
    482010051
    482010024
    482011034
    482010062
    480391004
    482010026
    482011039
    482011015
    482010029
    482011050
    483390078
    481671034
    480391016
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
N
430
451
433
451
441
451
454
437
445
455
443
450
430
454
442
428
449
444
455
427
455
Mean
36
34
37
38
40
38
37
35
38
45
39
34
38
41
41
38
45
40
43
37
34
SO Median IQR Site
17   32   25  A-
16   30   24  B-
     32
     35
17
16
18
17
16
17
17
17
16
16
18
16
18
15
16
17
12
17
16
     35
     34
     34
     31
     33
     43
     37
     28
     33
     39
     36
     33
     42
     35
     42
     33
     28
25
22
25
26
23
24
25
24
22
24
26
23
27
21
22
27
16
27
23
Key
_£,
SlO
n (N
!•- 	 |


ra
E
*

c
(0
tJ
E
1

_,c
jyto
§;E
*
'

£
" "8
[ — i

C
D
E
r-

G
H
 I
 1
J ~
K
L
M
N
0
D

Q
R
S
T
U
                                                       50
                                       100
 A
 B
 C
 D
 E
 F
 G
 H
 I
 J
 K
•L
•M
 N
 0
 P
 Q
 R
 S
 T
 U
                                                   150
                                                          03(ppb)
Figure 3-89   Site information, statistics and box plots for 8-h daily max ozone
               from AQS monitors meeting the warm-season data set inclusion
               criteria within the  Houston CSA.
Draft - Do Not Cite or Quote
             3-152
                                               September 2011

-------
                                    Los Angeles CSA
     Site ID
    060371602
    060371301
    060371302
    060371103
    060372005
    060374002
    060595001
    060590007
    060375005
    060371002
    060370002
    060370113
    060370016
    060371701
    060591003
    060371201
    060711004
    060376012
    060650004
    060592022
    061112002
    060658005
    060712002
    060658001
    061110007
    060710012
    060379033
    061110009
    060719004
    060659001
    060710005
    060656001
    060714003
    060714001
    060710306
    061113001
    061111004
    061112003
    060650009
    060650012
    060651016
    060710001
    060655001
    060719002
    060652002
    060651999
    060651010
    060711234
    060650008
    060659003
Years
07-09
07-08
09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
08
07-09
07-09
08-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
09
07-09
07-09
07-09
07-09
07-09
07-09
07-08
09
07-09
07,09
07-09
Key
£
(....
N
458
306
152
457
459
459
459
459
459
459
459
459
458
459
459
459
457
457
127
457
455
276
459
440
459
456
452
458
457
453
459
459
459
455
459
453
458
457
153
457
459
455
459
452
448
283
153
453
265
444
:H?Sb 	 « '
1 ( J4_ L — L, ^
' •>- \ u i


' \ fc i ~"
,, 	 ,J 	 L 	 1 	 j
• .1 	 c 	 1 1
'j . * , I ^
' • ' — I A 	 1 ^
I 	 -r-t- 	 ' "*
'»«•'
"LLi ^ — —
^-rrS=l"""'
^$^'
-A
-B
-C
-D
-E
-F
-G
-H
-I
""* J
-K
-L
-M
-N
-0
™ p
-R
-S
-T
-U
- V
-w
-X
- Y
-z
-AA
-AB
-AC
-AD
-AE
-AF
-AG
-AH
-Al
-AJ
-AK
-AL
-AM
-AN
-AO
- AP
-AQ
-AR
-AS
-AT
-AU
-AV
-AW
-AX
) 50 100 150
03 (ppb)
Figure 3-90    Site information, statistics and box plots for 8-h daily max ozone
                from AQS monitors meeting the warm-season data set inclusion
                criteria within the Los Angeles CSA.
Draft - Do Not Cite or Quote
3-153
September 2011

-------
                               Minneapolis CSA
         Site ID  Years  N Mean SD Median IQR Site
270031002 07-09
271390505 07-09
271636015 07-09
271713201 07-09
270031001 07-09
551091002 07-09
270495302 07-09
271453052 07-09


Key
*"?...
456
459
439
446
455
457
454
453
.c re
« £
CM E
"\ *

41
42
43
42
39
43
44
39
c
ro
E
I

12
11
12
11
12
11
10
11
P
lit


41
42
42
43
38
42
44
39

to
| „ „

16
14
16
16
18
15
14
15

"tft
iite
A-
B -
C-
D-
E -
P _
G-
H-

I , I I , f , J 1 , , , r
:- - - j f |- - - 1
t- | f | <
t - - - |_J_} - - H
j i STi t
"• • " -I ¥ } - - '
; -- {_J_J- - - <
[, . . . | ^ [ - . 4
•r - - -f 	 ^f]-- ^
' ' ' ' I ' ' ' ' 1 ' ' ' '

-A
-B
-C
-D
_ £
- F
_ Q
- H

0 50 100 150
                                                  Oj(ppb)
Figure 3-91   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Minneapolis CSA.
                                 New York CSA
Site ID
360810124
360610135
360050110
360050133
340030006
3401 70006
340130003
360850067
361192004
090010017
361030002
340315001
340250005
34023001 1
340273001
090019003
361030009
340190001
360790005
340210005
090013007
090011123
340290006
360715001
361030004
090090027
360270007
090093002
090050005
361111005
Years
07-09
08-09
07-09
07-09
08-09
07-09
09
07.09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
s
«o
> - . -
N
446
298
457
459
300
442
122
298
444
447
454
445
458
459
456
457
890
455
459
456
457
459
456
459
453
456
456
459
446
459
c
j- CO
fn I
(N E
-\ •
Mean
43
39
40
41
42
45
36
45
46
49
47
45
47
48
48
47
47
50
44
49
49
46
51
45
48
41
43
47
46
41
c
.™
-o
I
SD
15
15
15
14
15
17
14
16
17
15
15
15
15
17
16
16
16
16
14
16
15
17
16
14
14
14
14
15
15
12
_*:
"SS
is
o;E
!
Median IQR Site
41 21
38 19
39 19
39 18
41 20
43 20
36 19
43 23
44 22
47 20
46 20
43 19
45 19
47 22
47 22
44 21
46 20
48 21
42 19
48 22
47 19
43 22
49 20
43 17
46 18
40 17
41 17
45 18
43 19
39 16
*«
h~
!•-
*„
o>
- ^
A-
B-
C -
D-
E-
F -
G-
H-
I -
J-
K-
L-
M-
N -
0-
P-
Q-
R-
S-
T ™
u-
V-
w-
X-
Y-
z-
AA-
AB-
AC^
AD-
C
• ' '^j^,L i i i i 1 i i • i
^--QL_1 	 •'•
1 	 ^1
l.-r-ir-i-..-^
ISI
>r-^Q»»»'*
>.-rpri-...^
r~~~~fiT---~l j
J - -1 l» t 	 <
^--CED — <
-A
- B
-C
- D
- E
- F
-G
- H
- I
-J
- K
- L
-M
_ N
-0
- P
-Q
-R
-S
-T
- U
- V
- w
-X
- Y
-z
-AA
-AB
-AC
-AD
i i , i i . i i i j . i i .
) 50 100 150
03 (ppb)
Figure 3-92   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the New York CSA.
Draft - Do Not Cite or Quote
3-154
September 2011

-------
                                Philadelphia CSA
Site ID
421010014
421010004
340070003
420910013
340150002
421010024
420450002
420170012
100031013
100031010
340071001
420290100
3*110007
100031007
420110006
4201 1001 1




Years
07
07-09
07-08
07-09
07-09
07-09
07-09
08-09
07-09
07-08
07-09
07-09
07-09
07-09
08-09
08-09
Key

to ^
N
153
459
298
454
433
429
458
305
455
304
450
457
458
450
306
306
c

i E
Mean
50
39
51
51
50
51
49
48
48
53
52
50
50
49
44
47
§

e
SD
15
13
17
16
16
16
15
16
15
16
15
15
1ft
lo
14
15
13
14
fl

O'f"
Median 1QR Site
48
38
51
50
50
49
49
47
48
51
52
50
£.1
O I
50
48
43
46


t^
22
17
23
22
21
21
20
22
20
21
20
20
1 O
I 3
21
20
17
19

£

^ 	 {^__*^_l ; 	 I - - - - *
A ^
B-
C-
D-
E-
F -
G-
H-
I -
J-
K -
L -
M ~
N-
O -
P -
Q -
i , , , 1 i , , a 1 , i , ,
..y^j^ "
*• 	 i > ^ 	 <
:- - - - {]^;f^l 	 -i
i 	 Q Infill ----- ^
>•--- -LJfe_J 	 H
;" " • -[ jjt } 	 -'
> - - - { 'f'Z~l " " * ' "*
J. - . .. (,,^1^;;; - - 'i
i---^Q^3 	 i
••"•""" [IltlLlTI 	 ^
>-.OD---<
L... r=y=|^_. '
^ • - • 1 ^. 1 • - • - 1
t---Lj:J-----i



-A
-B
-C
- 0
-E
- F
- G
i- H
- 1
~ J
-K
- L
-N
- O
- P
- Q
0 50 100 150

Of rv_-U\
3 (PPD)
Figure 3-93   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Philadelphia CSA.
                                 Phoenix CBSA
Site ID
040133002
040133003
040139997
040131004
040134005
040134003
040130019
040137020
040137024
040137022
040132001
040137021
040134004
040131010
040137003
040132005
040139704
040135100
040134010
040134008
040139702
040139706
040213001
040213010
040213009
040217001
040139508
040134011
040213003
040218001
040213007



Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
£
h - --
N
455
459
455
459
454
459
459
459
459
457
459
457
459
456
455
459
459
453
457
459
451
448
459
458
459
459
459
459
459
459
459
i- G
s 1
H •
Mean
53
57
56
58
55
55
55
56
56
56
53
59
56
55
52
57
58
55
48
58
53
58
59
45
48
52
57
46
52
59
50
^6
1
1
SD
9
10
10
10
10
9
10
9
9
10
10
9
9
9
8
8
9
10
9
9
9
11
9
9
9
9
8
9
9
9
8
sS
^
5.<
:-{3D--<
j -- -MLJ--- <
'• - • ffVn - ~ •'
i — rf~i--<
^ - - j t j- - -:
•' - - ryi - - <
:---f--ryn--<
>- EU-^
i--{33--<
,-jtH
>-{jO--^
-^'ffl,-'1
!...^.1
:- — OEM
^-CC-H
.'--CE-i
>-OD-i
j-{HD--<
t-.P3.M
>--[J3;-<
^•CE--^
>^5D-<
^{TB-M
>-£!>-
^--C3J--'
,...133..,
f--cp-i
i . i . '-t-* i i i i i i t • t i

-A
-B
-C
-D
-E
u-F
-G
^H
^ 1
- J
-K
^L
-M
-N
^0
i-P
-Q
^R
i-S
-T
-U
i- V
-w
-X
i- Y
-Z
i- AA
^-AB
-AC
i- AD
-AE
0 50 100 150

O3 (ppb)
Figure 3-94   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Phoenix CBSA.
Draft - Do Not Cite or Quote
3-155
September 2011

-------
                              Pittsburgh CSA
Site ID
420030008
420030010
420030067
421290006
420031005
421250005
421255001
421250200
421290008
420070005
420070014
420070002
420050001
420730015
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
N
459
459
449
459
450
459
449
459
456
452
459
452
459
459
c
!M E
-) •
Mean
49
48
49
45
50
48
48
46
47
47
46
49
50
45
median
I
SD
14
13
12
13
15
12
13
12
13
13
13
13
15
13
1
]
Median IQR Site
48 19
47 18
49 16
44 18
50 20
48 17
47 17
46 16
46 17
47 16
46 18
49 16
48 21
44 19
£«
}--
-H
A-
B-
c-
D-
E-
F-
G-
H -
I
1
J-
K-
I _
M-
N-
C
, , , i I , , , « I , , , i

r---\ f 1 ^
^--Pfl-.-M
>---! h 1 	 -i
J 	 2 — I j
i 1j r •
1— -|— fT-)--M
) 50 100 1£
03 (ppb)

-H
A-
B-
c-
D-
E-
F-
G-
H-
1-
J-
K-
L-
C
, l.
I
--LJU~--<
V:.^
V-S
__4S
> 50
[ 	 1
H""1
}--1
J--H
-A
-B
-C
-D
-E
-F
-G
-H
-I
-J
-L
100 150
03 (ppb)
Figure 3-96   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Salt Lake City CSA.
Draft - Do Not Cite or Quote
3-156
September 2011

-------
                              San Antonio CBSA
480290055
480290032
480290052
480290622



08-09
07-09
07-09
08-09

Key
h....j
306
454
456
305

C
1
*
40
42
43
37

m
'•5
1
14
15
13
13

	 «c
"fflte

37
39
41
33

"E
\ - '
20
20
18
20

cn
- ^
A-
B.
c-
D-

(




)

^i~~l 	 -\
![• |- • - • - • H
"T"|n | 	 j
"4 ] 	 -;
• — i ^

50 100 U
3 (PPb)
-A
-B
-C
-D

30

Figure 3-97   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the San Antonio CBSA.
San Francisco CSA
Site ID
060010009
060750005
060010006
060012004
060012001
060811001
060131004
060011001
060130002
060410001
060950006
060010007
060852007
060950004
0601 33001
060850005
060851001
060950005
060131002
060550003
060870006
060870003
060953003
060870007
060970003
060852006
060870004
060850002
060971003
060690002
060690003
Years
08-09
07-09
07-08
08-09
07-09
07-09
07-08
07-09
07-09
07-09
07-08
07-09
07-08
07-09
07-08
07-09
07-09
07-09
07-09
07-09
07-08
07-09
07-09
07-09
07-09
07-09
07-09
07-09
08-09
07-09
07-09
N
306
458
303
306
459
459
306
456
458
458
306
459
306
459
306
456
459
459
459
459
306
456
455
456
459
458
459
459
306
456
457
Mean
29
28
31
25
35
31
31
34
42
29
40
43
34
35
41
36
39
39
47
37
38
31
44
33
31
44
33
44
34
42
54
SD
9
8
9
7
10
9
8
10
13
8
11
14
10
10
10
10
12
11
12
9
9
8
13
8
8
11
8
11
10
10
12
Median IQR
28
27
30
24
33
29
29
33
40
28
39
41
33
34
41
35
37
37
45
35
37
30
43
32
31
43
32
42
33
40
54
12
10
12
10
12
11
12
12
18
10
13
18
13
12
12
13
16
12
15
10
11
10
17
10
10
14
10
15
13
12
16
Site , , , , i , , , , i , , , ,
A-
B-
/"* _
D-
p — ,
C _
/"* ._
H-
I-
J-
K-
E ™
M-
N-
O-
P-
Q-
R-
S-
T-
u -
v-
W-
X-
Y-
z 	
AA-
AB-
AC~
AD-
AE-
>-LBr-^
>-[&•<
•,-E53~<
i-CE-n
f.-.j
• -*.
-H
,..<
^-CTED--<
^-€&-H
»--r7n 	 *
>--dEZl 	 ^
^CE3 — H
:-CT}--H
^rr\—-i
>--cp--M
^-ryn---^
1--G1} 	 ^
:-^TF1 	 H
i'-i i» i----<
:--| ;|i 1 	 H
J--QTJ--1
>--[~f~1 — ••<
-CB-'-i
=---00-.
-J,EE 	 :
>-d2-L
^^tD-M
M-CJU-.H
>-CB--<
j""E3H-- .
-A
-B
— f
-D
-E
— P
— f""
-H
- 1
- J
- K
-L
-M
- N
-O
- P
-Q
— R
-s
-T
- U
- V
- w
-X
- Y
-2
- AA
- AB
-AC
-AD
- AE
Figure 3-98   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the San Francisco CSA.
Draft - Do Not Cite or Quote
3-157
September 2011

-------
                                Seattle CSA
Site ID
530330080
530330010
530330017
530330023
530670005
530531008
530531010
530530012
530570018
530570013
Years
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07,09
07-08
07
Key
£
U"}
N
452
456
432
444
459
443
459
286
279
153
c
c. Ifl
13 I
-\ •
Mean
28
32
36
38
35
35
31
39
26
30
median
1
SD
8
12
12
14
10
12
11
9
8
10
	 
A-
B-
c-
D-
E-
F-
G-
H -
I-
J-
C
, , I , , , , I
1P~
;- - - ] |i | 	 i
_A
-c
-D
-E
-F
-G
-H
-I
-J
> i 1 i | i iii | i i i i
) 50 100 150
03 {ppb)
Figure 3-99   Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the Seattle CSA.
                               St. Louis CSA
Site ID
295100086
295100085
291890004
291890014
171630010
290990019
290990012
291831002
291831004
171193007
171190008
171191009
291890005
170831001
291130003
171170002
Years
07-08
07-09
07-08
07-09
07-09
08-09
07
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
07-09
Key
£
N
302
459
459
765
444
306
153
449
459
458
452
458
755
459
457
457
*« |
oj E
-\ *
Mean
50
46
50
48
43
49
55
51
50
48
48
50
46
46
50
46
median

SD
16
14
15
13
13
12
16
14
13
14
13
14
12
12
13
11
His
i
Median IQR Site
50 19
46 18
51 18
48 16
44 17
49 16
53 19
49 18
49 15
48 17
48 17
49 18
46 16
45 16
49 15
46 14
£
(--
-1
A-
D .^,,
c-
D-
E-
G-
H-
I
i
J-
K-
L-
M™
N-
o-
P-
C
l ,,<<[, i ,
V-.fi
y~~;^fi
:-'--&
f- — i «
^ — i
!• 	 1 4
'---US.

^j — 4
^ \ — i
i] — i
• \---i
i i — '.
^j 	 ^
i f 	 -i
^j - - - H
^j 	 	 .;
> [ 	 i
-A
K»w R
-C
-D
-E
-F
-H
-1
- J
-K
-L
-M
-0
-P
) 50 100 150
O3 (ppb)
Figure 3-100  Site information, statistics and box plots for 8-h daily max ozone
             from AQS monitors meeting the warm-season data set inclusion
             criteria within the St. Louis CSA.
Draft - Do Not Cite or Quote
3-158
September 2011

-------
      3.10.3 Ozone Concentration Relationships for the Urban Focus Cities
1
2
3
4
5
              This section contains histograms and contour matrices of the Pearson correlation
              coefficient (R) and the coefficient of divergence (COD) between 8-h daily max O3
              concentrations from each monitor pair within the 20 urban focus cities discussed in
              Section 3.6.2.1. These figures also contain scatter plots of R and COD as a function of
              straight-line distance between monitor pairs.
                                           Atlanta CSA
                   20-
                   15-
                   5
                    -0.1   0.0    0.1
                   1 0-

                   09-

                   08 -

                   07-

                   05-
                   03-


                   07-


                   01 -


                   00-


                   -01
                                   0.2    0.3    0.4    05    O.S    0.7
                                             Correlation
                                              tu   u_   O
                                     06*  0,67   069   065   069  069  OSS   075   076
                                     086   0
                         50   100  150  200  250   300  350  400  450
                                       Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-101  Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Atlanta CSA.
Draft - Do Not Cite or Quote
                                                 3-159
September 2011

-------
                                       Baltimore CSA


I
o




200-
150-
100-

50-

-c







1 0






2 9
0 0.1 0.2 0.3 0.4 0.5 06 0
Correlation


88



7 0

209





8 0

                                                                             70
                                                                                1.0
                                                                               AB
                 50   100   150   200   250   300   350   400   450
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-102   Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Baltimore CSA.
Draft - Do Not Cite or Quote
3-160
September 2011

-------
                                        Birmingham CSA
          25-
         „ 20-
         § 15'
         O 10-
           5-
            -0.1    0.0     0.1     0.2    0.3
                                            0.4    05
                                            Correlation
                                                        0.6     0.7    0.8    09     1.0
                              CD     O
 1.0-

 0.9-

 0.8-

 0.7-

 0.6

 0.5-

 0.4-

 0.3-

 0.2-

 0.1

 o.o-I

-0.1
                                                                                 -c
                                                                                 -D
                                                                                 -E
                                                                                  F
                                                                                  G
                                                                                 -H
                                                                                  J
                   50    100   150   200   250   300   350  400   450
                                    Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-103   Pair-wise monitor correlation coefficients (R) expressed as  a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the  Birmingham CSA.
Draft - Do Not Cite or Quote
                                 3-161
September 2011

-------
                                               Boston CSA
         _

         g  40-

         °20-
             -0.1
                     00
                            0 1
                                   02
                                          03
                                                 0.4     05

                                                  Correlation
                                                                06
                                                                       0 7
                                                                              08
                                                                                     0.9
                         <   m  o  o  in
                                            O  x  _
                                                                   ZOG.OCEOTI-D
             1.0-


             09-


             08-


             0.7-


             06-
             05 -
         JD
         &

         8   0.4
             03-


             0.2-


             0.1 -


             0.0-


            -0.1

             090 085 0.84 0.88 077 0.83  0.88 079 0.82 078 0.84 076 0,86 069


                 096 085 BROW 084 BH 083 089 079 088 079 090 078


                          082 084 «|jj 085 08S 080 080 077 090 073


                          083 OB9  087 OJB OB8 081 082 080 083 076


                          082 094  :;•-.•  .  oflo.88 077 0.86 0 BO OS2


                             082  OK.' :.!'-: 091 0.92 0/V 0 8J 082 Ut-'j


                             0%  090 079 0.88 ^H 0.88 0.84 0.89 063


                             0.88  0 78 0 75 0.89 ^H 0.74 0.85 0.78 0 55


                                    0.87 ^H 0.88 0.78 0.85 0.78 0 73


                                       0.79 0.77 ^1 0.71 ^1 0 65


                                       084 071 073 0.73 07? 081


                                          0.8C 076 0.88 079 004
U.B/


„
074 076



081 078


080 077


085 077


030 086



0 77 0.89


075 0.88


0.73


0.84 0,84


070 076


   050


083 083
069 0.7'


081 07.


074 0


079 0


065


o ei


065


058


0.75 0.78


065 0.'


060 060
./o


»
                                                                 068 077
                                                                 0.85 060
50    100    150    200    250    300

                    Distance (km)
                                                         350    400    450
  The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.


Figure 3-104   Pair-wise monitor correlation  coefficients (R) expressed as a
                  histogram (top), contour matrix (middle) and scatter plot versus
                  distance  between monitors (bottom) for the Boston  CSA.
Draft - Do Not Cite or Quote
                            3-162
                                September 2011

-------
                                                 Chicago CSA
            150-

          § 100-
          o
             50-
              -01
                      00
                              0.1
                                     02
                                            0.3
                                                    0.4     0.5

                                                     Correlation
                                                                   0.6
             1.0-



             09-



             08-



             0.7-



             06-
          |  05 -
          JD
          
-------
                                             Chicago CSA
         o
        O
150-
100-
50-
-c

5
.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0
Correlation
69

7 0.
                        < on O Q
                                          I _
                                                                                  X > N
                               0 77 Q 86 090 0 88M086 087 0.87 0,830.880801


                               ,• : o sn o SH o a: o K Ojifllo 89 o.se o eo o eel
             J 0 87 083 0.87 0 87 0 80 0.85 0 80 Q 80 0

               JOSS 0 900A1 098 0 86 O.SS 0.940 79

             10.89088 088Ho88 0,87 085 088 0 B1 085 089
                                   1 0 79 0 78 0 75, 0 80 0 76 0 81 080 0 70 074 0 76 080 0 74 0 79 0 78 0 89 0 78 0.73 0 76 0 72 0 7T 0 73

                                       0850860900.83

                                                   o SB o 80 oaoHJo.asHoa&^^Hjosa o.sa oss o m oS4 a an

                                                                     oi

                                                                          3083081 088081 084087

                                                                          |oi>808S0030790770.

                                                                          30.320.81 0840.85078

                                                                       10.83 0.87 0 34 O.S3 0 77 0 B6
                                                                        0.84 0.87 O.M 0.86 077 0.80 0,89

                                                                        081 084081 081 OB40 ?a08l

                                                                        0 82 0 81 0 80 O.B7 0 79 0 84 0.88
                                                                          083 083 Mo 790 eel

                                                                               080071 071 I

                                                                               0 79 0 72 0 80 C

                                                                                79071 075C

                                                                                  0780








tr
g
SS
«.
v.yti •.
t. j* jUr" _T £. V*
* J^%5$$&'. t ,
• *-.'v^?%*J*'..1tw;."'. 0760.860850.810.830.81084085
I" f.' ?<»V7
fc:«V.* • . 0630.820800730.82075071
* ' .* .
'* . ' *
• '
082JHo.S4081
080^0.32
083081
^^^^1
0.84





0 50 100 150 200 250 300 350 400 450
Distance (km)
    A
    B
    C
    D
    E
    F
    G
    H
    I
    J
    K
    L
    M
    N
    O
    P
    Q
    R
    S
    T
    u
    v
    w
    X
    Y
    Z
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-106   Pair-wise monitor correlation coefficients (R) expressed as a
                  histogram (top), contour matrix (middle) and scatter plot versus
                  distance between monitors (bottom) for the Dallas CSA.
Draft - Do Not Cite or Quote
3-164
September 2011

-------
                                            Denver CSA
           50-
        _ 40-
        § 30-
        5 20-
           10
                                     10
            -0.1    0.0     0.1     0.2    03     0.4     0.5     0.6    0.7     08     09     10
                                               Correlation
                        
-------
                                           Detroit CSA
           20-
         E 15-
         §10-
            5-
            -0.1    0.0    0,1     0.2    0.3
                      0.4    0.5
                      Correlation
0.6    0.7     0.8    0.9    1.0
                                      o
            1.0


            0.9-


            0.8-

            07


            0.6
         I  0.5-

         I
            0.3-

            0.2

            0.1 •

            00

           -0.1
••'.*
                                                                                 I - A
                                                                             0.64   - B
                                                                                   c
                                                           -D
                                                           -E
                                                                                  -F
                                                           -G
                                                           -H
                   50   100   150   200   250   300   350   400  450
                                     Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-108  Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance  between monitors (bottom) for the Detroit CSA.
Draft - Do Not Cite or Quote
                     3-166
                       September 2011

-------
                                              Houston CSA
§



80-
60-
40-


~c


2 1°
1 00 0.1 02 03 04 05 0
Correlation



6 0
33


7 0
88


8 0.
                          <  co  O  O
                                         u.  O I  _
 1.0-

 0.9

 0.8

 0.7

 0.6

 05-

 04-

 0.3

 0.2-

 0.1 -

 0.0-

-0.1
                               059 094 098 Q.78

        . 082 0.89 HI 0.87 0.88 HI 0.87 0.79 0,38 0.08 0.76

   089 081 I  Hlogi 074 HI OK 088 083 080 069 065

           : HHIoas o.eaHloe9 o.at 084 os? O.TS

         ^^•83 ^S        B 0 85 0 78 087 DM 0 80

                0.76 068 090 082 084 086 0.76 0.77 0.57

                         0 us 0 90 0 83 0 80 0 6S 0.64

                      0 83 093 0 86 0 76 089 : JC OK!

                      077 088 078 0 73 088 OSI 0«2

                                  Q.B4 0.69 0.66

                                     061 0.81
                               077 DM 063 aea

                                  073 077 0&5

                                     057 0.87

                                        0.46
                                                                               A
                                                                              •B
                                                                               C
                                                                              •D
                                                                               E
                                                                               F
                                                           084 076 ^H 088 I

                                                              ^1 0.83 ^H !

                                                                 077 0.88 076 0

                                                                     090 094 C

                                                                       • °
                                          K
                                          -L
                                          M
                                          N
                                          0
                                          P
                      50
                           100   150
                                       200   250   300
                                         Distance (km)
                                                        350   400   450
  The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-109   Pair-wise monitor correlation coefficients (R) expressed as  a
                  histogram (top), contour matrix (middle) and scatter plot versus
                  distance between monitors (bottom) for the Houston  CSA.
Draft - Do Not Cite or Quote
3-167
                                                                            September 2011

-------
                                      Los Angeles CSA
150 -

§ 100-
o
0 50-



3


63



109

17D




164



147



148



151



144




87

                                                                                29
         -0.1    0.0     0.1     0.2    0.3     0.4     0.5    0.6    0.7     0.8     0.9
                                            Correlation
        0.3-

        0.2-

        0.1 -

        0.0-

       -0.1
 „•.   .
•  •.  --.A.     v:
    •    ••.•••
           0     50    100   150   200   250   300   350   400   450
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-110   Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Los Angeles CSA.
Draft - Do Not Cite or Quote
                    3-168
September 2011

-------
                                        Minneapolis CSA
         15-
       o
       O
          5-
          1.0-

          0.9-

          0.8-

          0.7-

          0.6-
       I  0.5-
       ro
       0)
       Q  0.4 H
          0.3-

          0.2-

          0.1 -


          0.0-

         -0.1
                                                                          18
-0.1    0.0     0.1    0.2
                                     0.3     0.4    0.5    0.6
                                            Correlation
                                        o
                                                       LU
                    0.7     0.8    0.9     1.0
                                                                      O
             0     50    100   150   200    250   300   350   400   450
                                    Distance (km)
                                                                                   -G
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-111   Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Minneapolis CSA.
Draft - Do Not Cite or Quote
3-169
                                                                     September 2011

-------
                                        New York CSA

c
i
o



150-

100-

50 -
-C





1 6 | 20
.1 0.0 0.1 0.2 0.3 0.4 0.5 0
Correlation



Dt

6 0



	


7 0.8

                                                                          0.9
                     < co o Q
                                                                         03 O Q
         -0.1
             0     50    100   150   200   250  300   350   400   450
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-112   Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the New York CSA.
Draft - Do Not Cite or Quote
3-170
September 2011

-------
                                          Philadelphia CSA
         ~ 60"
         I 40-
         °20-
            -0.1
                   00
                          0 1
                                02
                                       03
                                              0.4     05
                                              Correlation
                                                           06
                                                                           O   0.  O
            1.0-

            09-

            08-

            0.7-

            06-
            0.5 -
         JD
         &

         8  0.4
            03-

            0.2-

            0.1 -

            0.0-

           -0.1
                                           H 091      087




                                           •fi 098  094  093
                  077  0.86  086 077  062

                  082  068  086 062  065  082 OS


                  083  068  086 085  089  084 06
:195  093 095  091  080 00!  091  088 081  084  089

           0.68  079 081  082  060 081  062  076 06

           0.90  0-89 082  088  088 081  083  087 05

           097  092 38-1  093  091 0 85  090  086 06

               0.87 084  090  089 082  083  0.87
                   flr.  0 91  091  0 fl-1  0 9t  0 83 OB
                   76  089  088 080  067  0.80 0.8
                                                                   0.78  0.86  0.82 069  0.7
                             078  0,65  0,90
                             -
                                 086  085 086
                                     067 071
                                     ::
                    50    100   150   200   250   300   350   400   450
                                      Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-113   Pair-wise monitor correlation coefficients (R) expressed as  a
                 histogram (top), contour matrix (middle) and scatter plot versus
                 distance between monitors (bottom) for the  Philadelphia CSA.
Draft - Do Not Cite or Quote
    3-171
September 2011

-------
                                       Phoenix CBSA

g 100-
3
o
0 50-
-c

15 , — — —
.1 0.0 0.1 0.2 0.3 0.4 0.5 0
Correlation

125
6 0
14"i

7 0.
                                                                      122
                                                                             27
                                                                         0.9
                                                                                1.0
                  50
                      100   150
                                 200   250   300
                                   Distance (km)
                                                350   400   450
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-114  Pair-wise monitor correlation  coefficients (R) expressed as a
               histogram (top), contour matrix (middle) and scatter plot versus
               distance between monitors (bottom) for the Phoenix CBSA.
Draft - Do Not Cite or Quote
3-172
September 2011

-------
                                         Pittsburgh CSA
          40-
        |30
        <3 20
          10-
           -0.1
                  0.0
                         0.1
                               0.2
                                      0.3
                                             0.4    0.5
                                             Correlation
             0.6
                       <    CD   O    Q   LU
                                                  O   X   _
                                                                            5   z
           1.0-

           0.9

           0.8-

           0.7-

           0.6-
        I  0.5-
        m
        O  fi A -
        O    1
           0.3-

           02

           01 -

           0.0-

          -0.1
                                     93  090   091  091   091   092  OPS   OS:
                                                                                08
                                              •1  080   091   095  087   090  OS6   089  06
001   0.84  0.82  089   087  087   0 SO  082  Ofl
                  0 83  0 85   0 65
    A

    B

    C

    D

    E

    F

    G

    H

   -I

    J

    K

    L

   • M

    N
                   50    100   150   200   250   300   350   400   450
                                     Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-115  Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Pittsburgh CSA.
Draft - Do Not Cite or Quote
3-173
September 2011

-------
                                         Salt Lake City CSA
         25
      € 20-
      8 1
      o 10-
         5-
-0.1     0.0     0.1     0.2
                             0.3
                                              0.4     0.5
                                               Correlation
                             00
                                                         O
          1.0-

          0.9-

          0.8-

          0.7-

          0.6-
      B  0.5-
      1
      O
0.4-

0.3-

0.2

0.1 -\

0.0
         -0.1
        •••
                                             077   083   086   085   075   0.79    077   0.67
                                        091   061    0.91   091    ''
                                             0.78    0.86   0.85   0.84
                                                                          83    081    071
                                                                         0,79    077
                                                                                7   0.64
                                             °"
                                                   0.84   0.84   0.80
                                                                         •
                                                                          76    0.73   0.62
                                                    0.84    0.76   0.77    0.77   0.72
                                                    0.92    081   0.84    0.82   0.73
                                                                               •A

                                                                                B

                                                                                C

                                                                                D

                                                                               -E
                                                    0.88    0.74   0.82    0.80   0.77
                                                                                        -H
                  50    100    150    200   250    300    350   400   450
                                      Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-116   Pair-wise monitor correlation coefficients (R) expressed as  a
                 histogram (top), contour matrix (middle) and scatter plot versus
                 distance between monitors (bottom) for the Salt Lake City CSA.
Draft - Do Not Cite or Quote
                                     3-174
                                                                          September 2011

-------
                                       San Antonio CBSA
         5-
       c
       i 3
       O 2-
         1
         -0.1
                0.0
                       0.1
                             0.2
0.3
0.4     0.5
 Correlation
                                                        0.6
          1.0

          0.9-

          0.8-

          0.7

          0.6 -\
      I  0.5 -\
      8
          03


          0.2-

          0.1 -


          0.0-

         -0.1
                                              -c
                                               E
                 50   100   150   200   250   300   350   400
                                    Distance (km)
                         450
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-117   Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the San Antonio CBSA.
Draft - Do Not Cite or Quote
       3-175
                                   September 2011

-------
                                      San Francisco CSA
80-

40
20-
-c




.1 0.0 0.1 0.2 0

31


3 0

41


4 0

60


5 0
98



6 0
95



7 0
96



8 0
                                           Correlation
                                                                            •
                     < CO O Q
                                                                       CO O O LLJ
                  50    100   150   200   250   300   350   400   450
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-118   Pair-wise monitor correlation coefficients (R) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the San Francisco CSA.
                of R.
Draft - Do Not Cite or Quote
3-176
September 2011

-------
                                           Seattle CSA
8-
s a-
g .
o 4
2-







1
| 	 '

4



•5



Q



7


9







          -0.1    0.0     0.1     0.2
0.3
0.4     0.5
 Correlation
0.6     0.7
           1.0-

           09-

           08-

           0.7-

           0.6

           0.5-\

           0.4

           0.3-

           0.2

           0.1 -

           0.0-

         -0.1
                                    0,59          064    056     0,57    042    0,10    062
                                                 079    080     074    049    032
                                                 076    081     080    066    025    071
                        084    061    026     0.65
                        080    059    038
                              0,58    025     0.62
                                    0.18     059
                                           0.24
                                                -J
                  50    100   150   200   250   300   350   400   450
                                     Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-119   Pair-wise monitor correlation coefficients (R) expressed as  a
                 histogram (top), contour matrix (middle) and scatter plot versus
                 distance between monitors (bottom) for the  Seattle CSA.
Draft - Do Not Cite or Quote
       3-177
                                     September 2011

-------
                                            St. Louis CSA
50-
£ 40-
3 -in.
0 20-
10-


29

60


30
          -0.1
    0.0
          1.0-


          09-


          0.8-


          0.7-


          0.6-




          04-


          0.3-


          0.2-


          0.1 -


          00-


         -0.1
:.-i
                         01
0.2
0.3
                            O   U   Q   01   LL
0.4     0.5
 Correlation

      x
0.6
0.7
0.9
1.0
                                                                 X   -1   2   Z   O   Q.
                                               ^B 088   084


                                                0.98  086   084  0.86  0.86  087
                                                                                      078
                                                                             0.79  0.85  0.76
                                                    090   090  088
                                                                     0 86  0 88  0 85
                                                                            •A

                                                                             B

                                                                             C

                                                                            •D
                    083  080  086  085   088  0.76  076  0,76  0.75


                    0.7B  079  0.77  0.77   0.77  0.87  Q.77  0.79  0.72


                    0.84  0.83  0.85  0.86   0.87  0.89  082  0.84  0.79


                            0.93  0.97   Oi)0  Oil


                            OSS ^H  085  076


                                         0.76  0.86  0.63  0.63


                                         0.76


                                         0 78  0 82  0 BO
             0     50    100   150   200   250   300   350   400   450
                                       Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-120   Pair-wise monitor correlation coefficients (R) expressed as a
                 histogram (top), contour matrix (middle) and scatter plot versus
                 distance between monitors (bottom) for the St. Louis CSA.
Draft - Do Not Cite or Quote
                                 3-178
                                                    September 2011

-------
                                         Atlanta CSA
30-
25-
c 20-
o 15-
0 10-
5-





22


33









         0.00    0.05    0.10   0.15
0.20    025   030    0.35
   Coefficient of Divergence
0.40    0.45   0.50    0.55
                            CO     CJ
006 009 009 008 009 003 008 011 013 Oil
010 010 008 010 0.09 009 Oil 013 012
010 011 011 012 011 013 012 013
0.55-

0.50-

0.45-

0.40-
s
g 0.35-
Q 0.30-
"5
1 0.25-
0
8 0.20-

0.15-

0.10-


0.05-
n nn -

011 007 009 010 011 010 0 OB

011 008 005 010 013 012

OOB 010 010 011 007

006 0.07 012 010
010 013 012

013 Oil

Oil

. ..'V. '
• ***l " •
.(iCft
.«,*..
~A
-B
C

D

-E

-F

-G
-H

-1

•J

•K


* .*
•

                 50   100   150   200   250   300   350   400   450   500
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-121   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Atlanta CSA.
Draft - Do Not Cite or Quote
       3-179
               September 2011

-------
                                        Baltimore CSA
200-
•& 150-
3 100-
50-



7
231





138


2
           0.00    0.05   0.10   0.15    0.20   0.25    0.30    0.35   0.40    0.45    0.50
                                        Coefficient of Divergence
                                      0.55
                     * IM • » •* •• •• i •:
i-o *•> •» *OT »• • » - I- Ml «•*••*«• o 	
P» i« .'!•• iw >v *ar to» MI ** »«• *« *• •» * ••
0(* 0- JM »«• M* AW *W *« «» •" *M *"
.. ..^.. „„..„..„..„ 	 	 .,.
	
1.. ^^f ,r .. ., .. .. ..
,. ,. .. „ .. .. .. «. „ .. ,.
tM ft «* «M «•» «>t
.. g 	 ,. ,. . .
....... „ . .. .
H" ™ ** " *" "

.. .. .„
** IW OM
om
* «
* *fc *A * !• *
* * *jftKf*"*'» *"^ "
&$$$$ •' ' '
2r*^

•A
B
C
c
- b
„ c
p
G
- H
-1
•J
•K
-M
•N
-0
P


-S
•T
•U
V
w
X
•Y
-z
•AA
-AB


                  50    100   150   200   250   300   350   400  450   500
                                    Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-122   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the  Baltimore CSA.
Draft - Do Not Cite or Quote
3-180
September 2011

-------
                                         Birmingham CSA
          30-

        I 2°-
        0 10-
                      36
           0.00    0.05    010
                               0.15    0.20    025   0.30    0.35    0.40    0.45   0.50    0.55
                                         Coefficient of Divergence
055

0.50-

0.45-

0.40-

0.35-
        01
        5 0.30-
          020-

          0.15

          0.10-

          0.05-

          0.00
                                     0.07    0.07    0.07    0.07    0.07     0.07    0.11    0.06
                                     009    0.09    0.09    0.10    0 Oe     0.09    0.11    008
                                           009    009    009    009     003    Oil    009
                                                       008    006     009    012    008
                                                       0.08    006     008    Oil    008
                                                             008     0.07    012    009
                                                                          012    0 07
                                                                          010    0.08
              0     50    100    150   200   250   300   350   400   450   500
                                      Distance (km)
                                                                                     -A
                                                                                     -B
                                                                                      C
                                                                                     -D
                                                                                     -E
                                                                                     -G
                                                                                     -H
                                                                            J
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-123   Pair-wise monitor coefficient of divergence (COD) expressed as a
                 histogram (top), contour matrix (middle) and scatter plot versus
                 distance between monitors (bottom) for the  Birmingham CSA.
Draft - Do Not Cite or Quote
                                   3-181
September 2011

-------
                                         Boston CSA
100-
- 80-
1 60-
0 40-
20-
2

75
114
18
	 1
000   005   0.10    0.15   0.20   0.25    0.30    0.35
                            Coefficient of Divergence
                                                              0.40   0.45   050    055
                           O  Q
                                    u_  O I  _
012 014 018 013 013 015 014 010 017 017 018 012 014 018 012 019 017 013 019 014
006 007 010 010 007 012 010 OOC 009 009 012 007 010 007 010 Oil 012 010 Oil
007 010 010 007 Oil 010 008 009 010 012 008 010 008 Oil Oil 012 0.11 012
010 011 007 012 008 008 007 009 013 Oil 010 Oil 010 009 013 010 014
0.07 0.08 0.08 0.05 012 0.11 0.10 0.09 012 011 0.11 0.15 0.12 0.10 0.14 0.11
008 008 007 012 012 011 008 012 012 0.11 015 013 009 015 010
0.55-
0.50-


0.45-

0,40-
|0.35-
1
Q 0.30 n
•s
I 0.25-
i
8 0.20-
0.15^

0.10-
0.05-
n nn -
009 008 007 010 009 010 009 009 008 012 011 010 012 010
009 012 013 012 008 012 012 Oil 016 013 007 016 009
008 0.09 0.10 010 0.09 0.11 0.12 0.11 0.10 0.12 0.11

Oil 010 013 006 P11 007 Oil Oil 013 012 012
0.09 0.15 012 0.11 0.13 0.09 0.08 0.15 0.09 0.15
014 012 008 012 012 010 013 012 013
0.12 0.12 0.10 0.17 014 0.05 017 0.07
01? 004 013 013 Oil 013 Oil
Oil 012 012 012 013 Oil
013 013 010 014 010
0.11 0.16 0.06 0.16
014 0.09 015
. " • • . • 0 16 0.06
. f»t . •. ;* . » . • •

\-fff'^-
-A
-B
-C
-D
-E
-F
-G
-H
•1

-J
-K
-L
-M
-N
-O
-P
-Q
-R
-s
-T
-U
                  50    100   150  200   250   300
                                    Distance (km)
                                                 350   400   450
                                                                 500
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-124   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Boston CSA.
Draft - Do Not Cite or Quote
                               3-182
September 2011

-------
                                        Chicago CSA
150-
= 100-
i
0 50-





2
159




163








1
             0.00    0.05    0.10    0.15    0.20    0.25    0.30    0.35
                                       Coefficient of Divergence
                                                             0.40
                                                                   0.45
                                                                         0.50
                                                                               0.55
0.090.090.140.11 0090.080 100.100130130 120090140 10011 0.11 011 0.120.11 0130150120 12011 0.12
0.08 0 10 0.07 0.07 0.09 0 07 0.10 009 008 009 009 0 10 0.07 0.07 009 0.07 0.08 009 008 009 0 10 010 009 0.09
0.11 0090080090050 11 0.08 009 0.10 0.11 010007008010009008010010011 0.090 100100.09
010011 014012013012009010013013011 010012010011 013010011 012012011 012
0.08011 0100 10011 0.070.070.09011 0.080.090.08 0.080.08009 009 0.090 10 0 10 0.08009
0.090.090.08011 008009010010008009008009009008010011 010011 010009
0.11 0.08011 013012011 013011 008011 0100120.10013014011 009012011

0.55-

0.50-

0.45-

0.40-
0)
I 0.35-
0)
fc
5 0.30-
•5
I 0.25-
£
3 0.20-

0.15-

0.10-


0.05-
n An
0120 07 009 011 Oil 011 007008010009008012010011 010010010009

0.14013011 012012012011 011 011 013008013013012012013012
010011 013012008008011 009009013009010011 010012010
009011 Oil 0070.09009009007012008009011 0 11 009009
0.11 Oil 0090100100100100100090,10011 0 12010010
0150.10010011 0.09012012010012013012009012
0.100.12011 0 120100130 13013011 0120.12011
0.08008008008011 008010009010008008
0 10 DOS 009 01 1 0.09 010010 003 010 0 10
010 008011 011 012 009011 009009
0.09010008009011 009009010
011 009010008010009006
010010011 012012010
0.04011 012011 0.10
012013012010
0100.100.07
. . " •_ 011010
• '*-v?K.v5^*-*v* •* oo9
:**&$ftii&?* * *':'*'

-A
-B
-C
- D
-E
-F
-G
H

-1
J
-K
-1
-M
-N
-o
-P
-Q
-R
-S
-T
-u
-V
-w
-X
-Y
-z

>*{*•.'


0 50 100 150 200 250 300 350 400 450 500
Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-125   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Chicago CSA.
Draft - Do Not Cite or Quote
3-183
September 2011

-------
                                        Dallas CSA
80-
§60-
0 40-
20-
0.





99
64
1
7
	 1
XI 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
Coefficient of Divergence

Oil 010 010 009 0.14 016 0.08 009 010 016 016 012 015 012 012 0.10 013 Oil
0.07 0.07 0.09 0.08 0.08 O.It 0.08 0.06 009 008 0.09 0.09 010 0.09 0.09 010 013
009 007 009 Oil 00* 008 007 Oil 010 008 012 008 007 010 009 011


0.55-


0.50-
0.45-


0.40-


§ 0.35-
m
5 0.30-
•s

1 0.25-
8 0.20-


0.15-
0.10-
0.05-
n on -













'•"-" \ V
rtfwp^V-* '
• "
0.07 0.07 0.09 O.It 007 007 0.03 0.09 0.10 0.09 010 0.10 0.10 0.09 0.12
010 012 006 006 008 012 Oil 0.09 012 007 0.07 010 009 006
007 013 008 009 007 DM 010 0.09 010 0.11 013 O.QB 014
01S 012 006 006 008 010 006 013 011 012 011 016
0.08 0.10 O.fS 014 0.10 015 0.08 0.07 0.10 Oil 0.08
009 ON 010 0.10 0.12 0.08 0.09 0.10 008 009
010 0.09 006 009 010 008 007 0.09 0.12

007 011 008 013 012 013 010 015
010 009 011 Oil 013 007 Old
Oil 0.10 006 009 009 Oil
013 012 Oil 012 016
0.06 0.12 0.08 0.08
010 008 010
Oil 012

..
0.55
•A
-B
C
-D
E
-F
•G
-H
-1
-J

•K
•L
M
• N
-0
•P
•Q
•R
-s

              0    50   100   150   200   250   300   350   400   450   500
                                   Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-126  Pair-wise monitor coefficient of divergence (COD) expressed as a
               histogram (top), contour matrix (middle) and scatter plot versus
               distance between monitors (bottom) for the Dallas CSA.
Draft - Do Not Cite or Quote
3-184
September 2011

-------
                                         Denver CSA
60-
50-
| 40-
o 30-
0 20-
10-




1
66








10 14 7
	 j 	 \ 	 1 	 , 1
0.00   0.05    010    015
0.20   0.25    030    035
   Coefficient of Divergence
                                                             0.40    045    0.50   055
                          CD   o
                                              CD
008 013 008 008 Oil 009 010 012 010 010 010
0 16 019 019 021 020 022 022 0.25 022 0.10 019 021
Oil 008 009 009 009 010 012 008 010 017 010 010
008 009 008 009 009 010 012 009 017 0.05 010
0.55-
0.50-

0.45-

0,40-
g
|0.35-
1
Q 0.30 n
•s
I 0.25-
^S
0 02°"

0.15-
0.10-
0.05-
n nn -
0.05 007 005 007 007 008 008 017 005 007
006 005 005 007 007 006 019 0.06 007
007 005 008 008 007 018 007 006

0.06 0.06 0.07 0.06 0.20 0.06 0.07
^^1
O.OB 0.07 0.07 020 0.07 0.08
^W
0.09 0.05 0.22 0.08 008
008 008
019 006 007
"•'• . '
_•* » 017 019
* ... '
006
•• •
**f+ »*
-A
-B
•C
-D
-E
-F
-G

-H

•1

-J
-K
•L

-M

-N
-0
"fc^* i

             0    50    100   150   200   250   300   350   400   450   500
                                   Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-127   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Denver CSA.
Draft - Do Not Cite or Quote
       3-185
                                                                   September 2011

-------
                                         Detroit CSA
         20
       I 15
       5 10
          5
                    25
          000   005    0.10   0.15
                                    0.20   0.25    0.30   0.35
                                       Coefficient of Divergence
                                                              040   045    050   055
                                     o
005 005 009 006 007 009 010 012
003 009 007 0.08 009 010 Oil
0.55-
0.50-
0.45-
0.40-
|0.35-
(| 0.30-
•5
I 025-
S 0.20-
015-

010-

0.05-
nnn
0.09 0.07 O.OS 008 009 Oil
0.08 0.10 0.10 Oil 0.13
0.08 0.08 011 0.12
0.09 0.07 0.11
^B
009 009
•
0,1

**•*•* •

• * * * * '
-A
-B
-C
-D
-E
-F

-G

-H

•1




                 50   100   150   200   250   300   350   400   450   500
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-128   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Detroit CSA.
Draft - Do Not Cite or Quote
3-186
September 2011

-------
                                       Houston CSA

0
o

100-
80-
60-
40-
20-
oc

22
114


56


18
	 1
0 0.05 010 015 0.20 0.25 030 035 0.40 045 0.50 0.55


< CO O Q
0.09 011 009











1
I
0
c

-------
                                       Los Angeles CSA
        400
       „ 300-
       c
      O
         100-
3
155

417

257
181

108
43 16 6 12 16
          0.00    0.05   010    0.15   0.20    0.25   0.30    035    040   0.45    0.50   055
                                       Coefficient of Divergence

                 50
100   150   200   250   300   350   400   450   500
             Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-130   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Los Angeles CSA.
Draft - Do Not Cite or Quote
                     3-188
September 2011

-------
                                         Minneapolis CSA
            20-
            15
            10-
             5-
             000   005   010   0.15    020    025   030   035   040    045    050   0.55
                                        Coefficient of Divergence
  0.55-

  0.50

  0.45-

  0.40-

I ° 35 "
5 0.30-
•5
I 0.25-

° 0.20

  0.15-

  0.10-

  0.05

  0.00
                                                      009     007      009      008
                                                      010     0 07      0 06
                                                                     008      009
                                                                     008      006
                                                             009      012
                                                                                 -A
                                                                                  B
                                                                                  C
                                                                                 •D
                                                                     0 07      0 09
                                                                                  F
                                                                                  G
                                                                                  H
                    50    100   150   200   250  300   350   400   450   500
                                     Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-131  Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Minneapolis CSA.
Draft - Do Not Cite or Quote
                                  3-189
September 2011

-------
                                         New York CSA
250 i
_ 200
§ 150-
o 100
50-


8?


276





74
3
             0.00    0.05   0,10   0.15   0.20   0.25   0.30    0.35    0.40    0,45   0.50
                                        Coefficient of Divergence
                                                                                0.55
                                                                        <
                                                                        <
                                                                         CD O Q








0.55-
050-

0.45-

0.40-

I 0.35-

5
5 030-
•5
1 0.25-
o

8 0,20-

0.15-


0.10-
0.05-
nnn

'" *" ' " '" '" " "" '" "" '" '" °" '"












' '
on M'
.V

Oil tit Ml [•«? I'l DM IM

,

.M <>..
.. .-.:..•'" •.-

• *J***^"5¥«^% • > *.'V: • *
• • *» * •» ^^x^S^f^^'f*" '•' '
' * *^Tt?*^*ti ^* ****"•• V*" *•
*;:.^i^.'t- "."' '* : '











j
K
-L
-M
- N
-0
-p
-Q
- R
-S
-T
-u

-w

-X
-Y
-z
-AA
-AB
-AC
-AD


                   50    100   150   200   250  300   350   400   450   500
                                    Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-132   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the  New York CSA.
Draft - Do Not Cite or Quote
3-190
September 2011

-------
                                       Philadelphia CSA
so-
^ 60-
c
o 40 -
O
20-
o.oc







0.55-
0.50-
0.45-

0.40-
8
g 035-
15 0.30-
5
| 0.25-
1
o 02°-
0.15-
0.10-
0.05-
nnn
83
38
2 10 ;
0.05 010 015 020 0.25 0.30 035 040 0.45 0.50
Coefficient of Divergence
•



0.55


-A
-B
-c
- D
-E
-F
-G
- H
-I

-J
-K
L
M
N
-O
-P
-Q

                   50   100   150  200  250  300   350   400   450   500
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-133  Pair-wise monitor coefficient of divergence (COD) expressed as a
               histogram (top), contour matrix (middle) and scatter plot versus
               distance between monitors (bottom) for the Philadelphia CSA.
Draft - Do Not Cite or Quote
3-191
September 2011

-------
           300
           250
        £  200
        8  150
        o  10Q
           50
     1        310
      •      h
                                         Phoenix CBSA
            0.00    005    010   015   020    0.25    030   0.35   0.40    0.45    0.50   055
                                        Coefficient of Divergence
           0.55-

           0.50-

           0.45-

           0.40-

           0.35-
        5  030-
         §  0.25-

         I
           0.20-

           0.15-

           0.10-

           0.05-

           0.00
o

                                                                  ••::,
               v*
               j, • -'.  •.
              & :•*•*•  '
  A
 -B
 -C
 -D
 -E
 -F
 -G
 -H
 - I
  J
 -K
 -L
 -M
 -N
 -O
 -P
  Q
  R
  S
  T
  U
  M
  W
  X
  Y
  Z
  AA
  AB
 -AC
 -AD
  AE
              0    50    100   150   200   250   300   350   400  450   500
                                    Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-134   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the Phoenix CBSA.
Draft - Do Not Cite or Quote
                                   3-192
September 2011

-------
                                       Pittsburgh CSA
84
_60-
g 40-
O
20 1 6
0.00 0.05 0,10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0,50 0.55

Coefficient of Divergence
004 006 007 008 007 0.07 007 006 007 008 0.07 007 009
0.06 0.07 0,08 0.06 0.07 006 006 007 0.07 007 0.07 0.09
009 009 006 007 006 007 007 009 006 008 011
010 007 008 007 006 009 OOB 009 009 009
0.55-
050-
0.45-
0.40-
§ 0.35 -
5 0.30-
•5
1 0.25-
g
° 0.20-
0.15-
0.10-
0.05-
nnn
010 010 010 009 009 010 010 007 012
008 005 006 OOB 009 OOB 009 0.11
007 008 007 008 008 008 010
006 008 008 0.07 0.09 0.10
008 008 0.08 0.08 0,10
005 007 008 008

0.08 008 007
0.08 0 1 0
010
>*&/•
-A
-B
-C
-D
-E
-F
-G
- H
-1
-J

-K
- L
-M
-N
• "V*" *
              0    50   100   150   200   250   300   350   400   450   500
                                   Distance (km)

 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-135  Pair-wise monitor coefficient of divergence (COD) expressed as a
               histogram (top), contour matrix (middle) and scatter plot versus
               distance between monitors (bottom) for the  Pittsburgh CSA.
Draft - Do Not Cite or Quote
3-193
September 2011

-------
                                      Salt Lake City CSA
40
c 30 1

O 20 -j 13
10-^^
0.00 0.05 010 0.15 020 025 030 035
Coefficient of Divergence
tf CQ CJ O 1 1 1 u_ C5 T
i i i i i i t i
004 006 0.05 008 008 006 006
DOS 005 006 004 004 005
007 DOB 006 OOB D OH
0.55-
0.50-
0.45-
0.40-
8
£ 0.35-
O)
5
i5 0.30-
•^
1 0.25-
!£
° 0.20-

0.15-

0.10-
0.05-
n nn
0.07 0.07 0.06 0.06
O.OB 0 OS 0.06
0 07 0.05
006











. ..
£f*f*h '






0.40 0.45 050 055
— -5 ^ — I
0.07 007 0,07 008
007 0.06 007
0 07 0 OB 0 07 0 09
0 07 0.07 0,08
0.07 007 007 007
0.06 0.07 0.07 0.08
005 006 005 006

004 005 005


0 05 0 05 0 06
V ^^_
0,05 005
^B
005



•A
B
-c
-D
-E
-F
-G

H


-I

-J

K

- L



                  50   100   150   200   250   300   350   400  450  500
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-136  Pair-wise monitor coefficient of divergence (COD) expressed as a
               histogram (top), contour matrix (middle) and scatter plot versus
               distance between monitors (bottom) for the Salt Lake City CSA.
Draft - Do Not Cite or Quote
3-194
September 2011

-------
                                       San Antonio CBSA
5-
c 4
3 3-
0 2-
1 -




6




4





0.00   0.05    0.10   0.15    0.20   0.25    0.30    0.35
                             Coefficient of Divergence
                                                              0.40
                                                                     0.45
                                                                           0.50
                                                                                  0.55
          0.55-

          0.50-

          0.45-

          0.40-

          035-

          0.30-

          0.25-

          0.20-

          0.15-

          0.10-

          0.05-

          0.00
                                                                                  A
              0    50    100   150   200   250   300   350   400   450   500
                                    Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-137   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter plot versus
                distance between monitors (bottom) for the San Antonio CBSA.
Draft - Do Not Cite or Quote
                                 3-195
September 2011

-------
                                      San Francisco CSA
150-
100-
 50-
                      98
                            170
                                  95
                                        61
            000   005   010    015    020   025   030   035   040   045    0.50
                                       Coefficient of Divergence
                                                                                055
                      
-------
                                         Seattle CSA
       o
15-
10-
5-
3
I 	
16
IO
7

1
	 1 . . . . .
          0.00    005   010    015    020   025    030   035    040    0.45
                                       Coefficient of Divergence
                               050    0.55
                             CD     O
016 0.19 0.22 0.19 020 0.19 0.23 0.17 018
013 015 015 014 015 021 020 017
0.09 0.12 0.10 0.13 0.13 022 014
0.55-

0.50-
0.45-
0.40-
g 0.35-
1
Q 0.30-
0
1 0.25-
0
fe
o 0.20-
0.15-

010-
0.05-
rinn -

Oil 009 014 016 023 017

008 014 014 020 0 IB
012 014 021 016

019 019 014

0.26 021
• « • p .
• • *
. * . • ote
. . * *
. . *
* . • • .
• . .
*
-A
-B
-c

-D
-E
-F

-G

-H

-I

-J


                  50    100   150   200   250   300   350   400   450   500
                                   Distance (km)
 The colors in the histogram bins correspond to the levels of the contour matrix. The histogram includes the number of monitor
pairs per bin and the contour matrix includes the numeric values of R.

Figure 3-139   Pair-wise monitor coefficient of divergence (COD) expressed as a
                histogram (top), contour matrix (middle) and scatter  plot versus
                distance between monitors (bottom) for the Seattle CSA.
Draft - Do Not Cite or Quote
3-197
September 2011

-------
                                         St. Louis CSA
80-
£ 60-
340-
20-



2
90





27

            0.00   0.05    0.10    0.15   0.20   0.25   0.30    0.35    0.40
                                        Coefficient of Divergence
                                                                    0.45
                                                                          0.50
                                                                                 0.55
                       
-------
       3.10.4 Hourly Variations in Ozone for the Urban Focus Cities
1
2

4
5
               This section contains diel plots of 1-h avg O3 data to supplement the discussion on hourly
               variations in O3 concentrations from Section 3.6.3.2 using data from the 20 urban focus
               cities first introduced in Section 3.6.2.1. Comparisons are made between cold months
               (October-April) and warm months (May-September), using the year-round data set, and
               between weekdays (Mon-Fri) and weekends (Sat-Sun) using the warm-season data set.
                                          warm Month*
                                       0 days. 0 year-round s
                      0000 OB 00  1300 16.00 CO 00 0000 0600 1200 1600  CO 00 0000 0600 1200 18.00 0000 0000 06.00 1200 1600  OOOC
                           Cold Months

                                                          Weekdays
                      2  1W-
                      o
                         i days, 9year-roundsites
                         •• mean
                          rmdlan
                          5" -95"
                                       •V.-ri Ssy. 'i >e:ir--ijLi"3 ^-'
                                                       327 itoy« 28 warm-*MHn sites   132 day*. 26 warm-season *t«s
                      0000 06:00  12:00 16:00 MOO COM 06'OQ 12:00 16'00 WOO 0000 06'00 1200 16-00 0000 OOW 06:00 1*00 19-00  ODOC

                             hauu             ruiur              hour             hour
                           Cold Months
                                          Warm Months
                                                          Weekdays
                        •637 (Jays, 1 ywr-rwind Sit*
                        — main
                        	 median
                        <= 5"-95"
                      0000 0600  12-00 1900 0000 0000 0600 12-00 1800  00000000 0600 1200 1800 00000000 06:00 1200 1Q.OO  OOOC

                             hour             hour              hour             hour
                                          Warm Months
                                                          Weekdays
                 3
                 o
                        S37 days. Sy»ar-rourKtsj1*i
                        — mean
                                       459 days. 3 ytar-tound srtw
                      00.00 06:00  12:00 19.00 00.00 0000 06:00 12.00 16.00 00.00 00.00 06:00 1200 19.00 0000 00.00 06:00 UOQ 19:00  OOOC
                             hour             hour              hour             hour
  No year-round monitors were available for the cold month/warm month comparison in the Atlanta CSA.

Figure 3-141  Diel patterns in 1-h avg ozone for select CSAs between 2007 and
                 2009 using the year-round data set for the cold month/warm month
                 comparison (left half) and  the warm-season data set  for the
                 weekday/weekend comparison (right half).
Draft - Do Not Cite or Quote
                                                     3-199
September 2011

-------
                       Cold Months
                                           Warm Months
                                                                Weekdays
              150 -


              100 -


               so -
637 days. 11 year-round sites
«•« mean
	 median
a 5"-95"
                                       459 days 11 year-round sites
                                                            327 days. 26 warm-season sites
                                                                                132 days, 26 warm-season srles
                 0000  06:00 12:00  ISOO

                         hour


                       Cold Months
               000000:00  06-00 1200 18:00 00:0000:00 0600 12:00  18:00  00:0000:00 0600  1200  18:00 OOOC

                           hour                 hour                  Hour
                                           Warm Months
                                                                Weekdays
          35
150 -


100 -


 so -
               0 -
637 days. 19 year-round sites

— mean
	 median
c^ 5"-95"
^=> 1«-9ST
                                       459 days 19 year-round sites
                                                            327 days, 19 warm-season sites
                                                                                132 days. 19 warm-season sites
                 0000  0600 12:00  1800

                         hour


                       Cold Months
               000000:00  0600 1200 1600 000000:00 0600 1200  18:00  000000:00 06:00  1200  1800 OOOC

                           hour                 hour                  hour
                                           Warm Months
                                                                Weekdays
                                                                                     Weekends
                   637 days. 12 year-round sites
                   — mean
                   	 median
                   <=> 5"-95"
                                       459 days 12 year-round sites     327 days, 15 warm-season sites    132 days, 15 warm-season srtes
                                  000000:00 06:00  1200 18:00 00:0000:00  0600 12:00 18:00  000000:00 0600 1200  1800  OOOC

                                              hour                 hour                 hour
                       Cold Months
                                           Warm Months
                                                                Weekdays
          si««
          O ""£
                   0 days. 0 year-round sites

                   — mean
                   — median
                   c=^ 5"-95"
                      no year-round data
                                       0 days, Q year-round ates
                                          no year-round data
                                                            327 days. 9 warm-season srtes    132 days, 9 warm-season sites
                 0000  0600 12:00  I8OO

                         Hour
               0000 00:00  06-00 1200 1600 00:00 OO'OO 0600 1200  1fl:00  0000 0000 0600  1200  1800 OOOC

                           hour                 hour                  hour
Figure 3-142   Diel patterns in 1-h avg ozone for select CSAs between 2007 and
                   2009 using the year-round data set for the cold month/warm month
                   comparison (left half) and the warm-season data set for the
                   weekday/weekend comparison (right half).  No year-round  monitors
                   were available for the cold month/warm month comparison in the
                   Detroit CSA.
Draft - Do Not Cite or Quote
                                3-200
                                                                             September 2011

-------
                     Cold Months
                                             Warm Months
                                                                      Weekdays
                                                                                              Weekends
     150 -
     § a 10M
o s
   |
   5
|    *M
                537 days. 21 year-round sites
                —  mean
                	  median
                <=> 5*-95*
                = l"-991n
                                        459 days. 21 year-round sues
                                                                 327 days, 21 warm-season siles
                                                                                         132 days. 21 warm-season sites
             00:00 06:00  12:00  18:00  00:00 00:00  06:00  12:00  18:00  0000 00:00 06:00 12:00 1800 00:00 00:00  0600  12:00  18:00  OOOC

                        hour                     hour                      hour                     hour
                     Cold Months
                                             Warm Months
                                                                      Weekdays
                                                                                              Weekends
o
S s
f | 100 H
   6'
                637 days, 47 year-round sites
                	  mean
                	  median
                t=^ 5"-95"
                = i" S3"
                                        459 days, 47 year-round sites
                                                                 327 days. 50 warm-season sites
                                                                                         132 days. 50 warm-season sites
             00:00 06:00  12:00  18:00  00:0000:00  06:00  12:00  18:00  00:0000.00 06:00  1200  18:00 00:0000:00  06:00  12:00  18:00  OO.OC

                        riour                     hour                      hour                     hour
                     Cold Months
                                             Warm Months
                                                                      Weekdays
                                                                                              Weekends
     O
     | | too-
     c
                425 days. 2 year-round sites
                —  mean
                	  median
                                        306 days, 2 year-round sites
                                                                 327 days, 8 warm-season sites
                                                                                         132 days, 8 warm-season sites
             00:00 06:00  12:00  18:00  00:0000:00  06:00  12:00  18-00  00-0000:00 06:00 12:00  18:00 00:0000:00  06:00  12-00  18:00  00:OC

                        hour                     hour                      hour                      hour
                     Cold Months
                                             Warm Months
                                                                      Weekdays
                                                                                              Weekends
     O
     t
     z
           637 days. 20 year-round sites
           • •••  mean
           	  median
           => S"  95"
                                        459 days, 20 year-round sites
                                                                 327 days, 30 warm-season siles
                                                                                         132 days, 30 warm-season sites
             00:00 06'QO  12:00  18:00  00:00 00:00  06:00  12:00  18.00  00.00 00:00 06:00 12:00 18:00 00:00 00:00  06:00  1200  18:00  00:OC
                        hour                     hour                      hour                     hour


Figure 3-143   Diel patterns in 1-h avg ozone for select CSAs between 2007 and
                    2009 using the year-round data set for the cold month/warm month
                    comparison (left  half) and the warm-season data set for the
                    weekday/weekend comparison  (right half).
Draft - Do Not Cite or Quote
                                                 3-201
September 2011

-------
                    Cold Months
                                            Warm Months
                                                                      Weekdays
                                                                                              Weekends
     o
      5*-95*
           = 1--99"
                                        459 days. 9 year-round sites
                                                                327 days, 17 warm-season siles
                                                                                        132 days. 17 warm-season sites
             00:00  06:00 12:00  18:00  00:00 00:00  06:00  12:00  18:00  0000 00:00  06:00  12:00  1800  00:00 00:00  0600  12:00  18:00  OOOC

                        hour                      hour                     hour                     hour
                    Cold Months
                                            Warm Months
                                                                      Weekdays
                                                                                              Weekends
     3
     m
     ll
           637 days. 14 year-round sites
           	  mean
           	  median
           r= 5"-95"
           = i" S3"
                                        459 days, 14 year-round sites
                                                                327 days. 31 warm-season sites
                                                                                        132 days. 31 warm-season sites
             00:00  06:00  12:00  18:00  00:0000:00  06:00  12:00  18:00  00:0000.00  06:00  12:00  18:00  00:0000:00  06:00  12:00  18:00  OO.OC

                        tour                     hour                     hour                     hour
                    Cold Months
                                            Warm Months
                                                                      Weekdays
                                                                                              Weekends
o
I
.Q
                637 days. 2 year-round sites
                —  mean
                	  median
                ^=3 5*-95*
                                        459 days, 2 year-round sites
                                                                327 days, 14 warm-season sites
                                                                                        132 days, 14 warm-season sites
             00:00  06:00  12:00  18:00  00:0000:00  06:00  12:00  18-00  00-0000:00  06:00  12:00  18:00  00:0000:00  06:00  12-00  18:00  00:OC

                        hour                     hour                     hour                     hour
                    Cold Months
                                            Warm Months
                                                                      Weekdays
                                                                                              Weekends
o
>N
o
o
                424 days. 2 year-round sites
                • •••  mean
                	  median
                =>  5  95"
                                        306 days, 2 year-round siles
                                                                327 days, 12 warm-season siles
                                                                                        132 days, 12 warm-season sites
             00:00  06'00  12:00  18:00  00:00 00:00  06:00  12:00  18.00  00.00 00:00  06:00  12:00  18:00  00:00 00:00  06:00  1200  18:00  00:OC
                        hour                     hour                     hour                     hour


Figure 3-144   Diel patterns in  1-h avg  ozone for select CSAs/CBSAs between
                    2007 and 2009 using the year-round data  set for the cold
                    month/warm month  comparison (left half) and the warm-season
                    data set for the weekday/weekend comparison (right half).
Draft - Do Not Cite or Quote
                                                 3-202
September 2011

-------
                    Cold Months
                                            Warm Months
                                                                     Weekdays
                                                                                              Weekends
S
o
o
o
     I
          150 -
           537 days. 5 year-round sites
           —  mean
           	  median
           <=> 5*-95*
           = 1--99"
                                        459 days. 5 year-round sites
                                                                327 days, 5 warm-season srtes
                                                                                        132 days. 5 warm-season srtes
             00:00  06:00 12:00  18:00  00:00 00:00  06:00  12:00  18:00  0000 00:00  06:00  12:00  1800  00:00 00:00  0600  12:00  18:00  OO.OC

                        hour                      hour                     hour                     hour
                    Cold Months
                                            Warm Months
                                                                      Weekdays
                                                                                              Weekends
o
o
|
u
5
     re
     V)
                637 days, 25 year-round sites
                	  mean
                	  median
                t= 5"-95"
                = i" S3"
                                        459 days, 25 year-round sites
                                                                327 days. 31 warm-season sites
                                                                                        132 days. 31 warm-season sites
             00:00  06:00  12:00  18:00  00:0000:00  06:00  12:00  18:00  00:0000:00  06:00  1200  18:00  00:0000:00  06:00  12:00  18:00  00:OC

                        hour                     hour                     hour                     hour
                    Cold Months
                                            Warm Months
                                                                      Weekdays
                                                                                              Weekends
     O  5

     1
     ru
                637 days. 5 year-round sites
                —  mean
                	  median
                ^=3 5*-95*
                                        459 days, 5 year-round srtes
                                                                327 days, 10 warm-season sites
                                                                                        132 days, 10 warm-season sites
             00:00  06:00  12:00  18:00  00:0000:00  06:00  12:00  18-00  00-0000:00  06:00  12:00  18:00  00:0000:00  06:00  12-00  18:00  00:OC

                        hour                     hour                     hour                     hour
                    Cold Months
                                            Warm Months
                                                                     Weekdays
                                                                                              Weekends
     O
     in
     O
           635 days. 3 year-round sites
           • -•-  mean
           	  median
           =  S" 95"
           <=  1*-99™
                    i
                                        459 days, 3 year-round sites
                                                                327 days, 16 warm-season sites
                                                                                        132 days, 16 warm-season sites
             00:00  06'00  12:00  18:00  00:0000:00  06:00  12:00  1800  000000:00  06:00  12:00  18:00  00:0000:00  06:00  1200  18:00  00:OC
                        hour                     hour                     hour                     hour


Figure 3-145   Diel patterns in  1-h  avg  ozone for select CSAs/CBSAs  between
                    2007 and 2009 using the year-round data  set for the cold
                    month/warm month  comparison (left half) and the warm-season
                    data set for the weekday/weekend comparison (right half).
Draft - Do Not Cite or Quote
                                                 3-203
                                                                                            September 2011

-------
    3.11  References

Acker, K: Febo, A: Trick, S: Perrino, C: Bruno, P: Wiesen, P: Moller: Wieprecht, W: Auel, R: Giusto, M: Geyer, A:
        Platt. U: Allegrini. I. (2006). Nitrous acid in the urban area of Rome. Atmos Environ 40: 3123-3133.
        http://dx.doi.0rg/10.1016/i.atmosenv.2006.01.028.
Andreae, MO. (1991). Biomass burning: its history, use, and distribution and its impact on environmental quality
        and global climate. In JS Levine (Ed.), Global Biomass Burning: Atmospheric, Climatic, and Biospheric
        Implications (pp. 1-21). Cambridge, MA: MIT Press.
Anton, M: Lopez, M: Vilaplana, JM: Kroon, M: McPeters, R: Banon,  M: Serrano, A. (2009). Validation of OMI-
        TOMS and OMI-DOAS total ozone column using five Brewer spectroradiometers at the Iberian
        peninsula. J Geophys Res 114: D14307. http://dx.doi.ora/10.1029/2009JD012003.
Appel, KW:  Gilliland, A: Eder, B. (2005). An operational evaluation of the 2005 release of models-3 CMAQ
        version 45. Washington DC: National Oceanic and Atmospheric Administration-Air Resources.
Archibald. AT: Levine. JG: Abraham.  NL: Cooke. MC: Edwards. PM: Heard. DE: Jenkin.  ME: Karunaharan. A:
        Pike, RC: Monks, PS: Shallcross, DE: Telford, PJ: Whallev, LK: Pyle, JA. (2011). Impacts of HO x
        regeneration and recycling in the oxidation of isoprene: Consequences for the composition of past,
        present and future atmospheres. Geophys Res Lett 38: L05804.
        http://dx.doi. org/10.1029/201OGL046520.
Arnold. JR:  Dennis. RL: Tonnesen. GS. (2003). Diagnostic evaluation of numerical air quality models with
        specialized ambient observations: testing the Community Multiscale Air Quality  modeling system
        (CMAQ) at selected SOS 95 ground sites. Atmos Environ 37:  1185-1198.
Arshinov. MY: Belan. BD: Krasnov. OA: Kovalevskii. VK: Pirogov. VA: Plotnikov. AP: Tolmachev. GN: Fofonov.
        AV. (2002). Comparison of ultraviolet and chemiluminescent ozonometers. Atmos Ocean 15: 656-658.
ATMET.  (Atmospheric, Meteorological, and Environmental Technologies). (2011). Atmospheric, meteorological,
        and environmental technologies, from http://atmet.com/
Barrie. LA: Bottenheim. JW: Schnell. RC: Crutzen. PJ: Rasmussen. RA.  (1988). Ozone destruction and
        photochemical reactions at polar sunrise in the lower Arctic atmosphere. Nature 334:  138-141.
Beckerman. B: Jerrett. M: Brook. JR: Verma. DK: Arain. MA: Finkelstein. MM. (2008). Correlation of nitrogen
        dioxide with other traffic pollutants near a major expressway. Atmos Environ 42: 275-290.
Beer, R. (2006). TES on the aura mission: Scientific objectives, measurements, and analysis overview. IEEE
        Trans Geosci Remote Sens 44: 1102-1105. http://dx.doi.orq/10.1109/TGRS.2005.863716.
Berkowitz. CM: Shaw. WJ. (1997). Airborne measurements of boundary layer chemistry during the Southern
        Oxidant Study: Acase study. J Geophys Res 102: 12,795-712,804.
        http://dx.doi. org/10.1029/97JD00417.
Berkowitz. CM: Fast. JD: Sprinston. SR: Larsen. RJ: Spicer. CW: Doskev. PV: Hubbe. JM: Plastridge. R. (1998).
        Formation mechanisms and chemical characteristics of elevated photochemical layers over the
        northeast United States. J Geophys Res 103: 10,631-610,647.
Binkowski. F:  Roselle. S. (2003). Models-3 Community Multiscale Air Quality(CMAQ) model aerosol component
        1. Model description. J Geophys Res 108: 4183. http://dx.doi.org/10.1029/2001JD001409.
Binkowski. FS: Arunachalam. S: Adelman. Z: Pinto. JR (2007). Examining photolysis rates with a prototype
        online photolysis module in CMAQ. J Appl Meteor Climatol  46: 1252-1256.
Bishop, GA: Stedman, PH. (2008). A decade of on-road emissions measurements. Environ Sci Technol 42:
        1651-1656. http://dx.doi.org/10.1021/es702413b.
Bloomer. BJ: Stehr.  JW:  Piety. CA: Salawitch. RJ: Dickerson. RR. (2009). Observed relationships of ozone air
        pollution with temperature and emissions. Geophys Res Lett 36:  L09803.
        http://dx.doi. org/10.1029/2009GL037308.
Blumenthal. PL: Lurmann.  FW: Kumar. N: Dye. TS: Ray. SE: Korc. ME: Londergan.  R: Moore. G. (1997).
        Transport and mixing phenomena related to ozone exceedances in the northeast US (analysis based
        on  NARSTO-northeast data). Santa Rosa, CA: Sonoma Technology.
        http://capita.wustl.edu/otag/reports/otagrept/otagrept.html.
Bonn, B: Von  Kuhlmann, R: Lawrence,  MG. (2004). High contribution of biogenic hydroperoxides to secondary
        organic aerosol formation. Geophys Res Lett 31: L10108. http://dx.doi.org/10.1029/2003GL019172.
Draft - Do Not Cite or Quote                       3-204                                 September 2011

-------
Brodin. M: Helmig. D: Oltmans. S. (2010). Seasonal ozone behavior along an elevation gradient in the Colorado
        Front Range Mountains. Atmos Environ 44: 5305-5315.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.06.033.
Burgard. DA: Bishop. GA: Stedman. DH: Gessner. VH: Daeschlein. C. (2006). Remote sensing of in-use heavy-
        duty diesel trucks. Environ Sci Technol 40: 6938-6942. http://dx.doi.org/10.1021/es060989a.
Burlev. JD: Ray. JD. (2007). Surface ozone in Yosemite National Park. Atmos Environ 41: 6048-6062.
Buzica. D: Gerboles. M: Plaisance. H. (2008). The equivalence of diffusive samplers to reference methods for
        monitoring O3, benzene and NO2 in ambient air. J Environ Monit 10: 1052-1059.
Bvun. D: Schere. KL. (2006). Review of the governing equations, computational algorithms, and other
        components of the models-3 community multiscale air quality (CMAQ) modeling system [Review]. Appl
        MechRev59: 51-77.
Bvun. DW: Ching. JKS. (1999). Science algorithms of the EPA models-3 community multiscale air quality
        (CMAQ) modeling system. (EPA/600-R-99-030). Washington, DC: U.S. Environmental Protection
        Agency. http://www.epa.gov/asmdnerl/CMAQ/CMAQscienceDoc.html.
Carter. WPL. (1995). Computer modeling of environmental chamber studies of  maximum  incremental
        reactivities of volatile organic compounds. Atmos Environ 29: 2513-2527.
CEMPD. (University of North Carolina at Chapel Hill, Center for Environmental  Modeling for Policy
        Development). (2011). SMOKE (Version 2.7) [Computer Program]. Chapel Hill, NC. Retrieved from
        http://www.smoke-model.org/index.cfm
Chan, E: Vet, RJ. (2010). Baseline levels and trends of ground level ozone in Canada and the United States.
        Atmos Chem Phys 10: 8629-8647. http://dx.doi.org/10.5194/acp-10-8629-2010.
Chen. X: Hopke. PK: Carter. WP (2011). Secondary organic aerosol from ozonolysis of biogenic volatile organic
        compounds: Chamber studies of particle and reactive oxygen species  formation. Environ Sci Technol
        45: 276-282. http://dx.doi.org/10.1021/es102166c.
Ching. J: Herwehe. J: Swall. J.  (2006).  On joint deterministic grid modeling and sub-grid variability  conceptual
        framework for model evaluation. Atmos Environ 40: 4935-4945.
Civerolo. KL: Mao. HT:  Rao. ST. (2003). The airshed for ozone and fine particulate pollution in the eastern
        United States.  Pure Appl Geophys 160: 81-105.
Conrad, R: Seller, W. (1985). Influence of temperature, moisture, and organic carbon on the flux of H2 and CO
        between soil and atmosphere: Field  studies in subtropical regions. J Geophys Res 90: 5699-5709.
Cooper. OR: Oltmans. SJ: Johnson. BJ: Brioude. J: Angevine. W: Trainer. M: Parrish. DP: Rverson. TR: Pollack.
        I: Cullis, PD: Ives, MA: Tarasick, DW: Al-Saadi, J: Stajner,  I. (In Press)  Measurement of western U.S.
        baseline ozone from the surface to the tropopause and assessment of downwind impact regions. J
        Geophys Res.
Cooper, OR: Parrish, DP: Stohl, A: Trainer, M: Nedelec,  P: Thouret, V: Gammas, JP: Oltmans, SJ:  Johnson, BJ:
        Tarasick. D: Leblanc. T: McDermid. IS: Jaffe. D: Gao. R: Stith. J: Rverson. T: Aikin. K: Campos. T:
        Weinheimer. A: Averv. MA. (2010). Increasing springtime ozone mixing ratios in the free troposphere
        over western North America. Nature 463: 344-348. http://dx.doi.org/10.1038/nature08708.
Corsmeier. U:  Kalthhoff. N: Kolle. O: Motzian. M: Fiedler. F. (1997). Ozone concentration jump in the stable
        nocturnal boundary layer during a LLJ-event. Atmos Environ 31: 1977-1989.
D'Anna, B: Jammoul, A: George, C: Stemmler, K: Fahrni, S: Ammann, M: Wisthaler, A. (2009). Light-induced
        ozone depletion by humic acid films and submicron aerosol particles. J Geophys Res 114: D12301.
        http://dx.doi.org/10.1029/2008JD011237.
Dallmann, TR: Harley, RA. (2010). Evaluation of mobile source emission trends in the United States. J Geophys
        Res 115: D14305. http://dx.doi.Org/10.1029/201OJD013862.
de Gouw. JA: Brock. CA: Atlas. EL: Bates.  TS: Fehsenfeld. FC: Goldan. PD:  JS. H: Kuster. WC: Lerner. BM:
        Matthew, BM: Middlebrook, AM: Onasch, TB: Peltier,  RE: Quinn, PK: Senff, CJ: Stohl, A: Sullivan,  AP:
        Trainer. M: Warneke. C: Weber. RJ:  Williams. EJ. (2008). Sources of particulate matter in the
        northeastern United  States in summer: 1.  Direct emissions and secondary formation of organic matter
        in urban plumes. J Geophys Res 113: D08301.
Dickerson. RR: Rhoads. KP:  Carsev. TP: Oltmans. SJ: Burrows. JP: Crutzen. PJ. (1999).  Ozone in the  remote
        marine boundary layer: A possible role for halogens. J Geophys Res 104: 21,385-321,395.
Draft - Do Not Cite or Quote                       3-205                                 September 2011

-------
Dochertv. KS: Wu. W: Lim. YB: Ziemann. PJ. (2005). Contributions of organic peroxides to secondary aerosol
        formed from reactions of monoterpenes with O3. Environ Sci Technol 39: 4049-4059.
        http://dx.doi. orq/10.1021/es050228s.
Doyle. M: Sexton.  KG: Jeffries. H: Bridge. K: Jaspers. I. (2004). Effects of 1,3-butadiene, isoprene, and their
        photochemical degradation products on human lung cells. Environ Health Perspect 112: 1488-1495.
Doyle. M: Sexton.  KG: Jeffries. H: Jaspers.  I. (2007). Atmospheric photochemical transformations enhance 1,3-
        butadiene-induced inflammatory responses in human epithelial cells: The role of ozone and other
        photochemical degradation products. Chem Biol Interact 166: 163-169.
        http://dx.doi.0rg/10.1016/i.cbi.2006.05.016.
Duncan. BN: Yoshida. Y:  Olson. JR: Sillman. S: Martin. RV: Lamsal.  L: Hu. Y: Pickering. KE: Retscher. C: Allen.
        DJ. (2010). Application of OMI observations to a space-based indicator of NOx and VOC controls on
        surface ozone formation. Atmos Environ 44: 2213-2223.
        http://dx.doi.org/10.1016/i.atmosenv.2010.03.010.
Dunker, A: Yarwood, G: Ortmann, J: Wilson, G. (2002). The decoupled direct method for sensitivity analysis in a
        three-dimensional air quality model implementation, accuracy, and efficiency. Environ Sci Technol 36:
        2965-2976. http://dx.doi.org/10.1021/es0112691.
Dunker, AM. (1981). Efficient calculation of sensitivity coefficients for complex atmospheric models. Atmos
        Environ 15: 1155-1161. http://dx.doi.orq/10.1016/0004-6981 (81 )90305-X.
Dunlea.  EJ: Herndon. SC: Nelson. DP: Volkamer. RM: Lamb. BK: Allwine. EJ: Grutter.  M: Ramos Villegas. CR:
        Marguez,  C: Blanco, S: Cardenas,  B: Kolb, CE: Molina, LT: Molina,  MJ. (2006). Technical note:
        Evaluation of standard ultraviolet absorption ozone monitors in a polluted urban environment.  Atmos
        Chem Phys Discuss 6: 2241-2279.
Ebeling, D: Patel, V:  Findlay, M: Stetter, J. (2009). Electrochemical ozone sensor and instrument with
        characterization  of the electrode and gas flow effects.  Sens Actuators B 137: 129-133.
        http://dx.doi. orq/10.1016/i.snb.20Q8.10.038.
Eder, B: Yu, S. (2005). A performance evaluation of the 2004 release of Models-3 CMAQ. Atmos Environ 40:
        4811-4824. http://dx.doi.0rg/10.1016/i.atmosenv.2005.08.045.
Eisele. FL: Mount.  GH: Tanner. D: Jefferson. A: Shetter. R: Harder. JW: Williams. EJ. (1997). Understanding the
        production and interconversion of the hydroxyl radical  during the tropospheric OH photochemistry
        experiment. J Geophys Res 102: 6457-6465. http://dx.doi.orq/10.1029/96JD02207.
Emmerson. KM: Evans. MJ. (2009). Comparison of tropospheric gas-phase  chemistry  schemes for use within
        global models. Atmos Chem Phys 9: 1831-1845. http://dx.doi.org/10.5194/acpd-8-19957-2008.
ENVIRON. (ENVIRON Holdings Inc.). (2005). CAMx, from http://www.camx.com/over/
Evans.  M: Fiore. A: Jacob. DJ.  (2003a). The GEOS-CHEM chemical mechanism: Version 5-07-8. Leeds, UK:
        University of Leeds.
Fang. Y: Fiore. AM: Horowitz. LW: Lew II. H: Hu. Y: Russell. AG. (2010). Sensitivity of the NOy budget over the
        United States to  anthropogenic and lightning  NOx in summer. J Geophys Res 115: D18312.
        http://dx.doi. org/10.1029/201OJD014079.
Fehsenfeld. FC: Trainer. M: Parrish. DP: Volz-Thomas. A: Penkett. S. (1996). North Atlantic Regional
        Experiment (NARE) 1993 summer  intensive:  Foreword. J Geophys  Res 101: 28869-28875.
Fehsenfeld, FC: Ancellet, G: Bates, TS: Goldstein, AH: Hardesty, RM: Honrath, R: Law, KS: Lewis, AC: Leaitch,
        R: McKeen. S: Meagher. J: Parrish. DP: Pszennv. AAP: Russell. PB: Schlager. H: Seinfeld. J:  Talbot. R:
        Zbinden. R.  (2006). International consortium for atmospheric research on transport and transformation
        (ICARTT): North America to Europe: Overview of the 2004 summer field study. J Geophys Res 111:
        D23S01.21-D23S01.36. http://dx.doi.org/10.1029/2006JD007829.
Finlavson-Pitts. BJ: Pitts. JN. Jr. (1986). Atmospheric chemistry: Fundamentals and  experimental techniques.
        In. New York, NY: John Wiley & Sons.
Fiore. A: Jacob. DJ: Liu. H: Yantosca.  RM: Fairlie. TD: Li. Q. (2003). Variability in surface ozone background
        over the United States: Implications for air quality policy. J Geophys Res 108:  4787.
        http://dx.doi. org/10.1029/2003JD003855.
Draft - Do Not Cite or Quote                       3-206                                 September 2011

-------
Fiore. A: Dentener. F: Wild. O: Cuvelier. C: Schultz. M: Hess. P: Textor. C: Schulz. M: Dohertv. R: Horowitz. L:
        MacKenzie, I: Sanderson, M: Shindell, D: Stevenson, D: Szopa, S: Van Dingenen, R: Zeng, G:
        Atherton. C: Bergmann. D: Bey. I: Carmichael. G: Collins. W: Duncan. B: Faluvegi. G: Folberth. G:
        Gauss. M: Gong. S: Hauglustaine. D: Hollowav. T: Isaksen. I: Jacob. D: Jonson. J: Kaminski. J:
        Keating, T: Lupu, A: Marmer, E: Montanaro, V: Park, R: Pitari, G: Pringle, K: Pyle, J: Schroeder, S:
        Vivanco. M: Wind.  P: Woicik. G: Wu. S: Zuber. A. (2009). Multimodel estimates of intercontinental
        source-receptor relationships for ozone pollution. J Geophys Res 114: D04301.
        http://dx.doi. org/10.1029/2008JD010816.
Fischer. EV. (2004). Summertime ozone at Mount Washington: Meteorological controls at the highest peak in
        the northeast. J Geophys Res 109: D24303. http://dx.doi.orq/10.1029/2004JD004841.
Fuentes, JD: Wang, D: Bowling, PR: Potosnak, M: Monson, RK: Goliff, WS: Stockwell, WR. (2007). Biogenic
        hydrocarbon chemistry within and above a mixed deciduous forest. J Atmos Chem 56: 165-185.
        http://dx.doi. orq/10.1007/s10874-006-9048-4.
Fuentes, M: Raftery, AE. (2005). Model evaluation and spatial interpolation by Bayesian combination of
        observations with  outputs from numerical models. Biometrics 61: 36-45.
Fusco.  AC: Logan. JA. (2003). Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
        midlatitudes with the GEOS-CHEM model. J Geophys Res 108: 4449.
        http://dx.doi. orq/10.1029/2002JD002742.
Gavdos. TM: Finder. R: Koo. B: Fahev. KM: Yarwood. G: Pandis. SN. (2007). Development and application of a
        three-dimensional aerosol chemical transport model, PMCAMx. Atmos Environ 41: 2594-2611.
Generoso. S: Bey. I: Attie. J. -L: Breon. F. -M. (2007). A satellite- and model-based assessment of the 2003
        Russian fires: Impact on the arctic region. J Geophys Res 112:  5302.
Gilliland, AB: Hogrefe, C: Finder,  RW: Godowitch, JM: Foley, Kl_: Rao, ST. (2008). Dynamic evaluation of
        regional air quality models: Assessing changes in O3 stemming from changes in emissions and
        meteorology. Atmos Environ 42: 5110-5123.
Godowitch, JM: Gilliland, AB: Draxler, RR: Rao, ST. (2008). Modeling assessment of point source NOx emission
        reductions on ozone air quality in the eastern United States. Atmos Environ 42: 87-100.
Goldstein. A: Galballv. I. (2007). Known and unexplored organic constituents in the earth's atmosphere. Environ
        Sci Technol 41: 1514-1521. http://dx.doi.org/10.1021/es072476p.
Goldstein. AH:  Millet. DB: McKay. M: Jaegle. L: Horowitz. L: Cooper. O:  Hudman. R: Jacob. DJ: Oltmans. S:
        Clarke. A. (2004).  Impact of Asian emissions on observations at Trinidad Head, California, during ITCT
        2K2. J Geophys Res 109: D23S17. http://dx.doi.org/10.1029/2003JD004406.
Gottardini. E: Cristofori. A:  Cristofolini. F: Ferretti. M. (2010). Variability of ozone concentration in a montane
        environment, northern Italy. Atmos Environ 44: 147-152.
        http://dx.doi. org/10.1016/i.atmosenv.20Q9.10.017.
Greenberg. JP: Guenther. AB: Turnipseed. A. (2009). Tethered balloon-based soundings of ozone, aerosols,
        and solar radiation near Mexico City during MIRAGE-MEX. Atmos Environ 43: 2672-2677.
        http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.019.
Grell. GA: Emeis. S: Stockwell. WR: Schoenemever. T: Forkel. R: Michalakes. J: Knoche. R: Seidl.  W. (2000).
        Application of a multiscale, coupled MM5/chemistry model to the complex terrain of the VOTALP valley
        campaign. Atmos  Environ 34: 1435-1453.
Guenther. A:  Geron. C: Pierce. T: Lamb. B: Harlev. P: Fall. R. (2000). Natural emissions of non-methane volatile
        organic compounds, carbon monoxide, and oxides of nitrogen from North America. Atmos Environ 34:
        2205-2230. http://dx.doi.org/10.1016/S1352-2310(99)00465-3.
Guenther. A:  Karl. T: Harlev. P: Wiedinmver. C: Palmer. PI: Geron. C. (2006).  Estimates of global terrestrial
        isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature). Atmos
        Chem  Phys 6: 3181-3210. http://dx.doi.org/10.5194/acp-6-3181-2006.
Hains. JC: Taubman. BF: Thompson. AM: Stehr. JW: Marufu. LT: Doddridge. BG: Dickerson. RR. (2008).
        Origins of chemical pollution derived from Mid-Atlantic aircraft profiles using a clustering technique.
        Atmos Environ 42: 1727-1741. http://dx.doi.org/10.1016/i.atmosenv.20Q7.11.052.
Hameed. S: Pinto. JP: Stewart. RW. (1979). Sensitivity of the predicted CO-OH-CH4 perturbation to
        tropospheric NOx concentrations. J Geophys Res 84:  763-768.
Harlev, RA: Marr, LC: Lehner, JK: Giddings, SN. (2005). Changes in motor vehicle emissions on diurnal to
        decadal time scales and effects on atmospheric composition. Environ Sci Technol 39: 5356-5362.
Draft - Do Not Cite or Quote                       3-207                                 September 2011

-------
Harvard University. (201 Oa). GEOS-Chem Model [Computer Program]. Cambridge, MA. Retrieved from
        http://acmg.seas.harvard.edu/geos/
Harvard University. (201 Ob). GEOS-Chem Overview, from
        http://acmg.seas.harvard.edu/geos/geos overview.html
Harvard University. (2011a). Anthropogenic emissions, from http://wiki.seas.harvard.edu/geos-
        chem/index.php/Anthropogenic  emissions
Harvard University. (2011 b). Appendix 9:  GEOS-Chem version history, from
        http://acmg.seas.harvard.edu/geos/doc/man/appendix  9.html
Harvard University. (2011c). Overview of GMAO met data products, from http://wiki.seas.harvard.edu/geos-
        chem/index.php/Overview of  GMAO met data  products
Harvard University. (2011d). Scale factors for anthropogenic emissions, from http://wiki.seas.harvard.edu/geos-
        chem/index.php/Scale  factors for anthropogenic emissions
Henderson. BH: Finder. RW: Crooks. J: Cohen. RC: Hutzell. WT: Sarwar. G: Goliff. WS: Stockwell. WR: Fahr. A:
        Mathur, R: Carlton, AG: Vizuete, W. (2010). Evaluation of simulated photochemical partitioning of
        oxidized nitrogen in the upper troposphere. Atmos Chem  Phys 10: 20125-20165.
        http://dx.doi. org/10.5194/acpd-10-20125-2010.
Hocking, WK: Carey-Smith, T: Tarasick, DW: Argall, PS: Strong, K: Rochon, Y: Zawadzki, I: Taylor, PA.  (2007).
        Detection of stratospheric ozone intrusions by windprofiler radars. Nature 450: 281-284.
        http://dx.doi.org/10.1038/nature06312.
Hofzumahaus, A: Rohrer, F: Keding, l_: Bohn, B: Brauers, T: Chih-Chung, C: Fuchs, H:  Holland, F: Kita, K:
        Kondo. Y: Xin. L: Shengrong. L:  Min. S: Limin. Z:  Wahner. A: Yuanhang. Z. (2009). Amplified trace gas
        removal in the troposphere. Science 324: 1702-1704.  http://dx.doi.org/10.1126/science.1164566.
Hsu, J: Prather, MJ. (2009). Stratospheric variability and tropospheric ozone. J Geophys Res 114: D06102.
        http://dx.doi. org/10.1029/2008JD010942.
Hudman. RC: Murray. LT: Jacob. DJ: Millet. DB: Turguetv. S: Wu. S: Blake. PR: Goldstein. AH: Hollowav. J:
        Sachse, GW. (2008). Biogenic versus anthropogenic sources of CO in the United States. Geophys Res
        Lett 35: L04801.  http://dx.doi.Org/10.1029/2007gl032393.
Husar. RB: Renard. WP (1998) Ozone as a function of local wind speed and direction:  Evidence of local and
        regional transport 91st annual meeting and exhibition of the Air & Waste Management Association. San
        Diego, CA.
Inman. RE:  Inqersoll. RB: Lew. EA.  (1971). Soil: a natural sink for carbon monoxide. Science 172: 1229-1231.
        http://dx.doi.org/10.1126/science.172.3989.1229.
Jacob. DJ: Horowitz. LW: Munger. JW: Heikes. BG: Dickerson. RR: Artz. RS: Keene. WC. (1995). Seasonal
        transition from NOx-to hydrocarbon-limited conditions for ozone production over the eastern United
        States in September. J Geophys Res 100:  9315-9324.
Jacob. DJ. (1999). Introduction to atmospheric chemistry.  In. New Jersey: Princeton University Press.
Jacobson. MZ. (2002). Atmospheric  pollution: history, science,  and regulation. In. New York: Cambridge
        University Press.
Jacobson. MZ. (2005). Fundamentals of atmospheric modeling. ln(2ed.). New York: Cambridge University
        Press.
Jaegle, L: Jacob, DJ: Brune, WH: Wennberg, PO. (2001).  Chemistry of HOx radicals in the upper troposphere.
        Atmos Environ 35: 469-489. http://dx.doi.org/10.1016/51352-2310(00)00376-9.
Jaffe. D: Chand.  D: Hafner. W: Westerling. A: Spracklen. D. (2008). Influence of fires on O-3 concentrations in
        the western US.  Environ Sci Technol 42: 5885-5891. http://dx.doi.org/10.1021/es800084k.
Jaffe. D. (2011). Relationship between surface and  free tropospheric ozone in the Western U.S. Environ Sci
        Technol  45: 432-438. http://dx.doi.org/10.1021/es1028102.
James, P: Stohl,  A: Forster, C: Eckhardt,  S: Seibert, P: Frank, A. (2003). A 15-year climatology of stratosphere-
        troposphere exchange with  a Lagrangian particle dispersion model: 2. Mean climate and seasonal
        variability. J Geophys Res 108: D12. http://dx.doi.org/10.1029/2002JD002639.
Jimenez, Jl_: Jayne, JT: Shi, Q: Kolb, CE: Worsnop, PR: Yourshaw, I: Seinfeld, JH: Flagan, RC: Zhang, X:
        Smith. KA. (2003). Ambient aerosol sampling using the Aerodyne Aerosol Mass Spectrometer. J
        Geophys Res 108: 8425.
Draft - Do Not Cite or Quote                       3-208                                 September 2011

-------
Jo. WK: Park. JH. (2005). Characteristics of roadside air pollution in Korean metropolitan city (Daegu) over last
        5 to 6 years: temporal variations, standard exceedances, and dependence on meteorological
        conditions. Chemosphere 59: 1557-1573. http://dx.doi.Org/10.1016/i.chemosphere.2004.12.021.
Johnson. D: Jenkin. ME: Wirtz. K: Martin-Rivieio. M.  (2004). Simulating the formation of secondary organic
        aerosol from the photooxidation of toluene. Environ Chem 1: 150-165.
Johnson. TR. (1995). Recent advances in the estimation of population exposure to mobile source pollutants. J
        Expo Sci Environ Epidemiol 5: 551-571.
Kasibhatla, P: Chameides, WL. (2000). Seasonal modeling of regional ozone pollution in the eastern United
        States.  Geophys Res Lett 27: 1415-1418. http://dx.doi.orq/10.1029/1999GL011147.
Kavnak. B: Hu. Y: Martin. RV: Russell. AG: Choi. Y: Wang. Y. (2008). The effect of lightning NOx production on
        surface ozone in the continental United States. Atmos Chem Phys 8: 5151-5159.
King. GM. (1999). Characteristics and significance of atmospheric carbon monoxide consumption by soils.
        Chemosphere 1: 53-63.
Kleffmann, J:  Lorzer, JC: Wiesen, P: Kern, C:  Trick, S: Volkamer, R: Rodenas, M: Wirtz,  K. (2006).
        Intercomparison of the DOAS and LOPAP techniques for the detection of nitrous acid (MONO). Atmos
        Environ 40: 3640-3652.
Kleffmann, J: Wiesen, P. (2008). Technical note: Quantification of interferences of wet chemical MONO LOPAP
        measurements under simulated polar conditions. Atmos Chem Phys 8: 6813-6822.
Kleindienst. TE:  Hudgens. EE: Smith. DF: McElrov. FF: Bufalini.  JJ. (1993). Comparison of chemiluminescence
        and ultraviolet ozone monitor responses in the presence of humidity and photochemical pollutants. Air
        Waste 43: 213-222.
Lam. Y: Fu. J. (2010). Corrigendum to "A novel downscaling technique for the linkage of global and regional air
        quality modeling" published in Atmos. Chem. Phys., 9, 9169-9185, 2009. Atmos Chem Phys 10: 4013-
        4031. http://dx.doi.org/10.5194/acp-10-4013-2010.
Langford. AO: Aikin. KC: Eubank. CS: Williams.  EJ. (2009). Stratospheric contribution to high surface ozone in
        Colorado during springtime. Geophys Res Lett 36: L12801. http://dx.doi.org/10.1029/2009gl038367.
Lee. J: Kim. KH: Kim. YJ. (2008b). Application of a long-path differential optical absorption spectrometer (LP-
        DOAS) on the measurements of NO(2), SO(2), O(3),  and HNO(2) in Gwangju, Korea. J Environ
        Manage 86: 750-759.
Lefohn. AS: Wernli. H: Shadwick. D: Limbach. S: Oltmans. SJ: Shapiro. M. (2011). The importance of
        stratospheric-tropospheric transport in affecting surface ozone concentrations in the western and
        northern tier of the  United States. Atmos Environ 45: 4845-4857.
        http://dx.doi.0rg/10.1016/i.atmosenv.2011.06.014.
Leston. AR: Ollinson. WM: Spicer. CW: Satola. J. (2005). Potential interference bias in ozone standard
        compliance monitoring. J Air Waste Manag Assoc 55: 1464-1472.
Li. Y: Lee. SR: Wu. CY (2006c). UV-absorption-based measurements of ozone and mercury: An investigation
        on their mutual interferences. Aerosol Air Qual Res 6: 418-429.
Liu, X:  Chance, K: Sioris, CE: Kurosu, TP: Spurr, RJD: Martin, RV: Fu, T, -M: Logan, JA: Jacob, DJ: Palmer, PI:
        Newchurch. MJ: Megretskaia. IA: Chatfield.  RB. (2006). First directly retrieved global distribution of
        tropospheric column ozone from GOME: Comparison with the GEOS-CHEM  model. J Geophys Res
        111: D02308. http://dx.doi.Org/10.1029/2005JD006564.
Liu. XH: Hegg. DA: Stoelinga. MT (2001). Numerical simulation  of new particle formation over the northwest
        Atlantic using the MM5 mesoscale model coupled with sulfur chemistry. J Geophys Res 106: 9697-
        9715.
Lu. R: Turco. RP: Jacobson. MZ. (1997). An integrated air pollution modeling system for urban and regional
        scales:  1 Structure and performance. J  Geophys Res 102: 6063-6079.
Luecken, DJ:  Phillips, S: Sarwar, G: Jang, C. (2008). Effects of using the CB05 vs. SAPRC99 vs. CB4 chemical
        mechanism on model predictions: Ozone and gas-phase photochemical  precursor concentrations.
        Atmos Environ 42:  5805-5820.
Mahajan, AS: Shaw, M: Oetjen, H: Hornsbv, KE: Carpenter, LJ: Kaleschke, L: Tian-Kunze, X: Lee, JD: Moller,
        SJ: Edwards. P. (2010). Evidence of reactive iodine chemistry in the Arctic boundary layer. J Geophys
        Res 115: D20303. http://dx.doi.Org/10.1029/2009JD013665.
Maruo, YY (2007). Measurement of ambient ozone using newly  developed porous glass sensor. Sens Actuators
        B 126: 485-491. http://dx.doi.Org/10.1016/i.snb.2007.03.041.
Draft - Do Not Cite or Quote                       3-209                                 September 2011

-------
Maruo. YY: Akaoka. K: Nakamura. J. (2010). Development and performance evaluation of ozone detection
        paper using azo dye orange I: Effect of pH. Sens Actuators B 143: 487-493.
        http://dx.doi.0rg/10.1016/i.snb.2009.09.042.
Mathur. R. (2008). Estimating the impact of the 2004 Alaskan forest fires on episodic particulate matter pollution
        over the eastern United States through assimilation of satellite-derived aerosol optical depths in a
        regional air quality model. J Geophys Res 113: D17302. http://dx.doi.orq/10.1029/2007JD009767.
McElrov. MB:  Salawitch. RJ: Wofsv. SC: Logan. JA. (1986). Reductions of Antarctic ozone due to synergistic
        interactions of chlorine and bromine. Nature 321: 759-762.
Milford. JB: Gao. D: Sillman. S: Blossev.  P: Russell. AG. (1994). Total reactive nitrogen (NOy) as an indicator of
        the sensitivity of ozone to reductions in hydrocarbon and NOx emissions. J Geophys Res  99: 3533-
        3542.
Miwa. T: Maruo. YY: Akaoka. K: Kunioka. T: Nakamura. J. (2009). Development of colorimetric ozone detection
        papers with high ultraviolet resistance using ultraviolet absorbers. J Air Waste Manag Assoc 59: 801-
        808.  http://dx.doi.0rg/10.3155/1047-3289.59.7.801.
Mollner. AK: Valluvadasan. S: Feng. L: Sprague. MK: Okumura. M: Milligan. DB: Bloss. WJ: Sander. SP:
        Martien. PT: Harlev. RA. (2010). Rate of gas phase association of hydroxyl radical and nitrogen
        dioxide. Science 330:  646-649. http://dx.doi.org/10.1126/science.1193030.
Mueller. SF: Mallard. JW. (2011a). Contributions of natural emissions to ozone and PM 2.5 as simulated by the
        Community Multiscale Air Quality (CMAQ) model. Environ Sci Technol 45: 4817-4823.
        http://dx.doi. org/10.1021/es103645m.
Mueller. SF: Mallard. JW. (2011 b). Errata in 'Contributions of natural emissions to ozone and PM 2.5 as
        simulated by the Community Multiscale Air Quality (CMAQ) model' [Erratum]. Environ Sci  Technol 45:
        7950. http://dx.doi.org/10.1021/es2027086.
Nassar. R: Logan. JA: Worden. HM: Megretskaia. IA: Bowman. KW: Osterman. GB: Thompson. AM: Tarasick.
        DW: Austin. S: Claude. H: Dubev. MK: Hocking. WK: Johnson. BJ: Joseph. E: Merrill. J: Morris. GA:
        Newchurch, M: Oltmans, SJ: Posnv, F: Schmidlin,  FJ: Vomel, H: Whiteman, DM: Witte, JC. (2008).
        Validation of Tropospheric Emission Spectrometer (TES) nadir ozone profiles using ozonesonde
        measurements. D15S17 (13 pp.). http://dx.doi.org/10.1029/2007id008819.
Newell, RE: Thouret, V: Cho, JYN: Stoller, P: Marenco, A: Smit, HG. (1999). Ubiquity of quasi-horizontal layers
        in the troposphere. Nature 398: 316-319. http://dx.doi.org/10.1038/18642.
NOAA. (National Oceanic and Atmospheric Administration). (2010). The Rapid Update Cycle (RUC), from
        http://ruc.noaa.gov/
Nolte. CG: Gilliland. AM: Hogrefe.  C: Micklev. LJ. (2008). Linking global to regional models to assess future
        climate impacts on surface ozone levels in the United States. J Geophys Res 113: D14307.
        http://dx.doi. org/10.1029/2007JD008497.
Noziere. B: Gonzalez. NJD: Borg-karlson. A. -K: Pel. Y: Redebv. JP: Kreici. R: Dommen. J: Prevot. ASH:
        Anthonsen. T. (2011). Atmospheric chemistry in stereo: A new look at secondary organic aerosols from
        isoprene. Geophys Res Lett 38:  L11807. http://dx.doi.org/10.1029/2011GL047323.
NPS. (U.S. National Park Service). (2011). Portable Ozone Monitoring Systems (POMS). Washington, DC.
        http://www.nature.nps.gov/air/studies/porto3.cfm.
NRC. (National Research  Council). (1991). Rethinking the ozone problem in urban and regional air pollution. In.
        Washington,  DC:  The  National Academies Press.
NRC. (National Research  Council). (2007). Models in environmental regulatory decision making. In.
        Washington,  DC:  National Academies Press.
NRC. (National Research  Council). (2009). Global sources of local pollution: An assessment of long-range
        transport of key air pollutants to  and from the United States. Washington, DC: The National Academies
        Press, http://www.nap.edu/catalog.php7record id=12743.
O-Keeffe. S: Fitzpatrick. C: Lewis. E. (2007). An optical fibre based ultra violet and visible absorption
        spectroscopy system for ozone concentration monitoring. Sens Actuators B 125:  372-378.
        http://dx.doi.0rg/10.1016/i.snb.2007.02.023.
Ohira. SI: Dasgupta. PK: Schug. KA. (2009). Fiber optic sensor for simultaneous determination of atmospheric
        nitrogen dioxide, ozone, and relative humidity. Anal Chem  81: 4183-4191.
        http://dx.doi.org/10.1021/ac801756z.
Draft - Do Not Cite or Quote                       3-210                                  September 2011

-------
Olaguer. EP: Rappengluck. B: Lefer. B: Stutz. J: Dibb. J: Griffin. R: Brune. WH: Shauck. M: Buhr. M: Jeffries. H:
       Vizuete, W: Pinto, JP. (2009). Deciphering the role of radical precursors during the Second Texas Air
       Quality Study. J Air Waste Manag Assoc 59: 1258-1277. http://dx.doi.orq/10.3155/1047-
       3289.59.11.1258.
Olszyna, KJ: Bailey, EM: Simonaitis, R: Meagher, JF. (1994). O3 and NOy relationships at a rural site. J
       Geophys Res 99: 14557-14563.
Oltmans. SJ: Lefohn. AS: Harris. JM: Shadwick. PS. (2008). Background ozone levels of air entering the west
       coast of the US and assessment of longer-term changes. Atmos Environ 42: 6020-6038.
       http://dx.doi.ora/10.1016/i.atmosenv.2008.03.034.
Park. RJ: Stenchikov. GL: Pickering: Dickerson. RR: Allen. DJ: Kondragunta. S. (2001). Regional air pollution
       and its radiative forcing: Studies with a single column chemical and radiation transport model. J
       Geophys Res 106: 28,751-728,770.
Parrish. DP. (2006). Critical evaluation of US on-road vehicle emission inventories. Atmos Environ 40: 2288-
       2300.
Parrish. DP: Millet. DB: Goldstein. AH. (2009).  Increasing ozone in marine boundary layer inflow at the west
       coasts of North America and Europe. Atmos Chem Phys 9: 1303-1323. http://dx.doi.orq/10.5194/acpd-
       8-13847-2008.
Peeters. J: Nguyen. TL: Vereecken. L. (2009). HOx radical regeneration in the oxidation of isoprene.  Phys
       Chem Chem Phys 11: 5935. http://dx.doi.org/10.1039/B908511D.
Peeters, J: Muller, J, -F. (2010). HOx radical regeneration in isoprene oxidation via peroxy radical
       isomerisations. II: Experimental evidence and global impact. Phys Chem Chem Phys 12: 14227-14235.
       http://dx.doi.ora/10.1039/COCP00811G.
Perring, AE: Bertram, TH: Wooldridge, PJ: Fried, A: Heikes, BG: Dibb, J: Crounse, JD: Wennberg, PO: Blake,
       NJ: Blake. PR: Brune. WH: Singh. HB: Cohen. RC. (2009). Airborne observations of total RONO2:  New
       constraints on the yield and lifetime of isoprene nitrates. Atmos Chem  Phys 9: 1451-1463.
Pfister, G: Hess, PG: Emmons, LK:  Lamarque, JF: Wiedinmyer, C: Edwards, DP:  Petron, G: Gille, JC: Sachese,
       GW. (2005). Quantifying  CO emissions from the 2004 Alaskan wildfires using MOPITT CO data.
       Geophys Res Lett 32: L11809.
Pinto, JP: Lefohn, AS: Shadwick,  PS. (2004). Spatial variability of PM2.5  in urban areas in the United  States. J
       Air Waste Manag Assoc 54: 440-449.
Pokharel. SS: Bishop. GA: Stedman. PH. (2002). An on-road motor vehicle emissions inventory for Penver: An
       efficient alternative to modeling. Atmos Environ 36: 5177-5184.
Pokharel. SS: Bishop. GA: Stedman. PH. (2003). Emissions reductions as a result of automobile improvement.
       Environ Sci Technol 37: 5097-5101.
Pollack, AK: Lindhjem, C: Stoeckenius, TE: Iran, C: Mansell, G: Jimenez, M: Wilson, G: Coulter-Burke, S.
       (2004). Final Report: Evaluation of the US EPA MOBILE6 highway vehicle emission factor model.
       (CRC Project E-64). Novato, CA: ENVIRON International Corporation.
Poppe, P: Wallasch, M: Zimmermann, J. (1993). The dependence of the concentration of OH on its precursors
       under moderately polluted conditions: A model study.  J Atmos  Chem 16: 61-78.
Rao. ST: Ku. J. -Y: Berman. S: Zhang. K: Mao. H. (2003). Summertime characteristics of the atmospheric
       boundary layer and relationships to ozone levels over the eastern United States. Pure Appl Geophys
       160: 21-55.
Rappengluck. B: Pasgupta. PK: Leuchner. M: Li. Q: Luke. W. (2009). Formaldehyde and its relation to CO,
       PAN, and SO2 in the Houston-Galveston airshed. Atmos Chem Phys Piscuss 9: 24193-24223.
       http://dx.doi. orq/10.5194/acp-10-2413-2010.
Rastigeiev. Y: Park. R: Brenner. MP: Jacob. PJ. (2010). Resolving intercontinental pollution plumes in global
       models of atmospheric transport. J Geophys Res 115: P02302.
       http://dx.doi.orq/10.1029/2009JP012568.
Reid. N: Yap. P: Bloxam.  R. (2008). The potential role of background ozone on current and emerging  air issues:
       An overview. Air Qual Atmos Health  1: 19-29. http://dx.doi.Org/10.1007/s11869-008-0005-z.
Reidmiller. PR: Fiore. AM: Jaffe. PA: Bergmann. P: Cuvelier. C: Pentener. FJ: Puncan: Bryan. N: Folberth. G:
       Gauss. M: Gong. S: Hess. P: Jonson.  JE:  Keating. T: Lupu. A: Marmer. E: Park.  R: Schultz. MG:
       Shindell, PT: Szopa, S: Vivanco, MG: Wild, O: Zuber, A. (2009). The influence of foreign vs.  North
       American emissions on surface ozone in the US. Atmos Chem Phys 9: 5027-5042.
Draft - Po Not Cite or Quote                       3-211                                  September 2011

-------
Reisinger. AR. (2000). Unidentified interference in DOAS measurements of ozone. Appl Spectres Rev 54: 72-
        79.
Richards. NAD: Osterman. GB: Browell. EV: Hair. JW: Averv. M: Qinbin. L. (2008). Validation of tropospheric
        emission spectrometer ozone profiles with aircraft observations during the intercontinental chemical
        transport experiment-B. J Geophys Res 113: D16S29. http://dx.doi.Org/10.1029/2007id008815.
Riediker. M: Williams. R: Devlin. R: Griggs. T: Bromberg. P. (2003). Exposure to particulate matter, volatile
        organic compounds, and other air pollutants inside patrol cars. Environ Sci Technol 37: 2084-2093.
        http://dx.doi. org/10.1021/esQ26264y.
Rodes. CE: Holland. DM. (1981). Variations of NO, NO2 and O3 concentrations downwind of a Los Angeles
        freeway. Atmos Environ 15: 243-250.
Russell, A: Dennis, R. (2000). NARSTO critical review of photochemical models and  modeling [Review]. Atmos
        Environ 34: 2283-2324. http://dx.doi.orq/10.1016/S1352-2310(99)00468-9.
Rverson. TB: Buhr. MP: Frost. GJ: Goldan. PD: Hollowav. JS: Hubler. G: Jobson. BT: Kuster. WC: McKeen. SA:
        Parrish, DP: Roberts, JM: Sueper, DT: Trainer, M: Williams,  J:  Fehsenfeld, FC. (1998). Emissions
        lifetimes and ozone formation in power plant plumes. J Geophys Res 103: 22569-22583.
        http://dx.doi.org/10.1029/98JD01620.
Rverson, TB: Trainer, M: Hollowav, JS:  Parrish, DP: Huey, LG: Sueper,  DT: Frost, GJ: Donnelly, SG:  Schauffler,
        S: Atlas. EL: Kuster. WC: Goldan. PD: Hubler. G: Meagher. JF: Fehsenfeld.  FC. (2001). Observations
        of ozone formation in power plant plumes and implications for ozone control strategies. Science 292:
        719-723. http://dx.doi.org/10.1126/science. 1058113.
Sakugawa. H: Kaplan.  IR. (1989). H2O2 and O3 in the atmosphere of Los Angeles and its vicinity: Factors
        controlling their formation and their role as oxidants of SO2.  J Geophys Res 94: 12957-12973.
Sarwar,  G: Roselle, SJ: Mathur, R: Appel, W: Dennis, RL: Vogel, B. (2008). A comparison of CMAQ MONO
        predictions with observations from the Northeast Oxidant and Particle Study. Atmos  Environ 42: 5760-
        5770.
Schichtel, BA: Husar, RB. (2001). Eastern North American transport climatology during high- and low-ozone
        days. Atmos Environ 35: 1029-1038. http://dx.doi.orq/10.1016/S1352-2310(00)00370-8.
Schnell. RC: Oltmans. SJ:  Neelv. RR: Endres. MS: Molenar. JV: White. AB. (2009). Rapid photochemical
        production of ozone at high concentrations in a rural site during winter. Nat Geosci 2: 120-122.
        http://dx.doi. orq/10.1038/NGEO415.
Seaman. NL. (2000). Meteorological modeling for air quality assessments. Atmos Environ 34: 2231-2259.
Seinfeld, JH: Pandis, SN. (1998). Atmospheric chemistry and physics: From air pollution to climate change. In.
        New York: John Wiley & Sons.
Sexton.  KG: Jeffries. HE: Jang. M: Kamens.  RM: Doyle. M: Voicu.  I: Jaspers. I. (2004). Photochemical products
        in urban mixtures enhance  inflammatory responses in lung cells. Inhal Toxicol  1: 107-114.
Shapiro. MA. (1980). Turbulent mixing within tropopause folds as a mechanism for the  exchange of chemical
        constituents between the stratosphere and troposphere. J Atmos Sci 37: 994-1004.
Sillman, S. (1995). The use of NOy,  H2O2 and HNO3 as indicators for ozone-NOx-hydrocarbon sensitivity in
        urban locations. J Geophys Res 100: 14175-14188.
Sillman. S: He. D: Pippin. MR: Daum. PH: Imre. DG: Kleinman. LI: Lee. JH: Weinstein-Llovd. J. (1998). Model
        correlations for ozone, reactive nitrogen, and peroxides for Nashville in comparison with
        measurements: implications for O3-NOx-hydrocarbon chemistry. J Geophys Res 103: 22629-22644.
Sillman. S: He. D. -Y. (2002). Some  theoretical results concerning O3-NOx-VOC chemistry and NOx-VOC
        indicators. J Geophys Res  107: 4659. http://dx.doi.org/10.1029/2001JD001123.
Singh. HB: Anderson. BE: Brune. WH: Cai. C: Cohen. RC: Crawford. JH: Cubison. MJ: Czech. EP: Emmons. L:
        Fuelberg.  HE.  (201 Ob). Pollution influences on atmospheric composition and chemistry at high northern
        latitudes: Boreal and California forest fire emissions. Atmos Environ  44: 4553-4564.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.08.026.
Spicer. CW: Joseph. DW: Ollison. WM. (2010). A re-examination of ambient air ozone monitor interferences. J
        Air Waste Manag Assoc 60: 1353-1364. http://dx.doi.org/10.3155/1047-3289.60.11.1353.
Stedman. DH: Dabv. EE: Stuhl. F: Niki. H. (1972). Analysis of ozone and nitric oxide  by a chemiluminescent
        method in laboratory and atmospheric studies of photochemical smog. J Air  Waste Manag Assoc 22:
        260-263.
Draft - Do Not Cite or Quote                       3-212                                 September 2011

-------
Stevens. R: Pinto. J: Mamane. Y: Ondov. J: Abdulraheem. M: AI-Maied. N: Sadek. M: Gofer. W: Ellenson. W:
       Kellogg, R. (1993). Chemical and physical properties of emissions from Kuwaiti oil fires. Water Sci
       Technol 27: 223-233.
Stutz. J: Ackermann. R: Fast. JD: Barrie. L. (2002). Atmospheric reactive chlorine and bromine at the Great Salt
       Lake, Utah. Geophys Res Lett 29: 1380. http://dx.doi.org/10.1029/2002GL014812.
Stutz. J: Oh. HJ: Whitlow. SI: Anderson. C: Dibb. JE: Flvnn. JH: Rappengluck. B: Lefe. B. (2009). Simultaneous
       DOAS and mist-chamber 1C measurements of MONO in Houston, TX. Atmos Environ TBD: TBD.
       http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.003.
Tang. Q: Prather. MJ:  Hsu. J.  (2011). Stratosphere-troposphere exchange ozone flux related to deep
       convection. Geophys Res Lett 38: L03806. http://dx.doi.orq/10.1029/201OGL046039.
Tanimoto, H: Mukai, H: Hashimoto, S: Morris, JE. (2006). Intercomparison of ultraviolet photometry and gas-
       phase titration techniques for ozone reference standards at ambient levels.  J Geophys Res 111:
       D16313. http://dx.doi.orq/10.1029/2005JD006983.
Tarasick, DW: Slater, R. (2008).  Ozone in the troposphere: Measurements, climatology, budget, and trends.
       Atmos Ocean 46: 93-115. http://dx.doi.org/10.3137/ao.460105.
Taubman. BF: Marufu. LT: Piety. CA: Doddridge. BG: Stehr. JW: Dickerson. RR. (2004). Airborne
       characterization of the chemical, optical, and meteorological properties, and origins of a combined
       ozone-haze episode  over the eastern United States. J Atmos Sci 61: 1781-1793.
Taubman. BF: Mains. JC: Thompson. AM: Marufu. LT: Doddridge. BG: Stehr. JW: Piety. CA: Dickerson. RR.
       (2006). Aircraft vertical profiles of trace gas and aerosol pollution over the mid-Atlantic United States:
       Statistics and meteorological cluster analysis. J Geophys Res 111: D10S07.
       http://dx.doi. orq/10.1029/2005JD006196.
Thompson,  AM: Stone, JB: Witte, JC:  Miller, SK: Oltmans, SJ: Kucsera, TL: Ross,  KL: Pickering, KE:  Merrill, JT:
       Forbes. G: Tarasick. DW: Joseph. E: Schmidlin. FJ: McMillan. WW: Warner. J: Hintsa. EJ: Johnson. JE.
       (2007). Intercontinental Chemical Transport Experiment Ozonesonde Network study (IONS) 2004: 2
       Tropospheric ozone budgets  and variability over northeastern North America. J Geophys  Res 112:
       D12S13. http://dx.doi.orq/10.1029/2006JD007670.
Thornton. JA: Kercher. JP: Riedel. TP: Wagner.  NL: Cozic. J: Hollowav. JS: Dube.  WP: Wolfe. GM: Quinn. PK:
       Middlebrook,  AM: Alexander,  B: Brown, SS. (2010). A large atomic chlorine source inferred from mid-
       continental reactive nitrogen chemistry. Nature 464: 271-274. http://dx.doi.org/10.1038/nature08905.
Trainer. M: Parrish. DP: Buhr. MP: Norton. RB: Fehsenfeld. FC: Anlauf. KG: Bottenheim. JW: Tang. YZ: Wiebe.
       HA: Roberts,  JM: Tanner, RL: Newman, L: Bowersox, VC: Meagher, JF: Olszyna, KJ: Rodgers, MO:
       Wang. T: Berresheim. H: Demeriian. KL: Rovchowdhurv. UK. (1993). Correlation of ozone with NOy in
       photochemically aged air. J Geophys Res 98: 2917-2925.
U.S. Census Bureau.  (2011). U.S. Census Bureau, from http://www.census.gov/
U.S. EPA. (U.S. Environmental Protection Agency). (1979b). Transfer standards for the calibration of ambient air
       monitoring analyzers for ozone: Technical assistance document. (EPA-600/4-79-056). Research
       Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1996a). Air quality criteria for  ozone and related
       photochemical oxidants. (EPA/600/P-93/004AF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for  ozone and related
       photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
       Protection Agency, Office of Research and Development.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2008a). 2005 National Emissions Inventory data and
       documentation, from http://www.epa.gov/ttn/chief/net/2005inventorv.html
U.S. EPA. (U.S. Environmental Protection Agency). (2008b). Integrated science assessment for oxides of
       nitrogen: Health criteria. (EPA/600/R-08/071). Research Triangle Park, NC.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645.
U.S. EPA. (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for sulfur oxides:
       Health criteria. (EPA/600/R-08/047F). Research Triangle Park, NC.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=198843.
U.S. EPA. (U.S. Environmental Protection Agency). (2008d). National air quality: Status and trends through
       2007. (EPA/454/R-08/006). Research Triangle Park, NC.
Draft - Do Not Cite or Quote                       3-213                                 September 2011

-------
U.S. EPA. (U.S. Environmental Protection Agency). (2009d). Integrated science assessment for particulate
        matter. (EPA/600/R-08/139F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
U.S. EPA. (U.S. Environmental Protection Agency). (2009e). Risk and exposure assessment for review of the
        secondary National Ambient Air Quality Standards for oxides of nitrogen and oxides of sulfur.
        (EPA/452/R-09/008A). Research Triangle Park, NC.
        http://www.epa.gov/ttnnaags/standards/no2so2sec/cr rea.html.
U.S. EPA. (U.S. Environmental Protection Agency). (201 Oa). Air trends: Design values, from
        http://epa.qov/airtrends/values.html
U.S. EPA. (U.S. Environmental Protection Agency). (2010b). Biogenic Emissions Inventory System (BEIS)
        modeling, from http://www.epa.gov/AMD/biogen.html
U.S. EPA. (U.S. Environmental Protection Agency). (2010c). Integrated science assessment for carbon
        monoxide. (EPA/600/R-09/019F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=218686.
U.S. EPA. (U.S. Environmental Protection Agency). (2010d). MOBILE6 vehicle emission modeling software,
        from http://www.epa.gov/otag/m6.htm
U.S. EPA. (U.S. Environmental Protection Agency). (2010e). Our nation's air: Status and trends through 2008.
        (EPA-454/R-09-002). Research Triangle Park, NC.
        http://www.epa.gov/airtrends/2010/report/fullreport.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (201 Of). Transfer standards for calibration of air monitoring
        analyzers for ozone. (EPA-454/B-10-001).  Research Triangle Park, NC.
        http://www.epa.gov/ttn/amtic/files/ambient/gagc/OzoneTransferStandardGuidance.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2011a). AirNow, from http://www.airnow.gov/
U.S. EPA. (U.S. Environmental Protection Agency). (2011c). Integrated  science assessment for ozone:
        Modeling for policy relevant background concentrations. Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2011d). Map monitoring sites, from
        http://www.epa.gov/airexplorer/monitor kml.htm
U.S. EPA. (U.S. Environmental Protection Agency). (2011e). MOVES (Motor Vehicle Emission Simulator), from
        http://www.epa.gov/otag/models/moves/index.htm
Univ of Leeds.  NCAS.. (University of Leeds, National Centre for Atmospheric Science). (2010). The master
        chemical mechanism, from  http://mcm.leeds.ac.uk/MCM/home.htt
Utembe, SR: Hansford, GM: Sanderson, MG: Freshwater, RA: Pratt, KFE: Williams, DE: Cox, RA: Jones, RL.
        (2006). An ozone monitoring instrument based on the tungsten trioxide (WO3) semiconductor. Sens
        Actuators B 114: 507-512. http://dx.doi.Org/10.1016/i.snb.2005.04.049.
van der Werf, GR: Randerson, JT: Giglio, l_: Collate, GJ: Kasibhatla, PS: Arellano, AF, Jr. (2006). Interannual
        variability in global biomass burning emissions from 1997 to 2004. Atmos Chem Phys 6: 3423-3441.
Vardoulakis. S: Lumbreras. J: Solazzo. E. (2009). Comparative evaluation of nitrogen oxides and ozone passive
        diffusion tubes for exposure studies. Atmos Environ 43: 2509-2517.
        http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.048.
Viallon. J:  Moussav. P: Norris. JE: Guenther. FR: Wielgosz. Rl. (2006). A study of systematic biases and
        measurement uncertainties  in ozone mole  fraction measurements with the NIST Standard Reference
        Photometer. Metrologia 43:  441-450. http://dx.doi.Org/10.1088/0026-1394/43/5/016.
Wang. HQ: Jacob. DJ: Le Sager. P:  Streets. DG: Park. RJ: Gilliland. AB: van Donkelaar. A. (2009a). Surface
        ozone background in the United States: Canadian and Mexican pollution influences. Atmos Environ 43:
        1310-1319. http://dx.doi.0rg/10.1016/i.atmosenv.2008.11.036.
Wang. J: Christopher. SA: Nair. US:  Reid. JS: Prins. EM: Szvkman. J: Hand. JL. (2006). Mesoscale modeling of
        Central American smoke transport to the United States: 1. "Top-down" assessment of emission
        strength and diurnal variation impacts. J Geophys Res 111: D05S17.
        http://dx.doi.org/10.1029/2005JD006416.
Webster, M: Nam, J: Kimura, Y:  Jeffries, H: Vizuete, W: Allen, DT (2007). The effect of variability in industrial
        emissions  on ozone formation in Houston,  Texas. Atmos  Environ 41: 9580-9593.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.052.
Draft - Do Not Cite or Quote                       3-214                                 September 2011

-------
Wernli. H: Bourgui. M. (2002). A Lagrangian "1-year climatology" of (deep) cross-tropopause exchange in the
        extratropical Northern Hemisphere. J Geophys Res 107: 4021.
        http://dx.doi.org/10.1029/2001JD000812.
Williams. EJ: Fehsenfeld. FC: Jobson. BT: Kuster. WC: Goldan. PD: Stutz. J: McClennv. WA. (2006).
        Comparison of ultraviolet absorbance, chemiluminescence, and DOAS instruments for ambient ozone
        monitoring. Environ Sci Technol 40: 5755-5762. http://dx.doi.org/10.1021/es0523542.
Wilson. KL: Birks. JW. (2006). Mechanism and elimination of a water vapor interference in the measurement of
        ozone by UV absorbance. Environ Sci Technol 40: 6361-6367. http://dx.doi.org/10.1021/es052590c.
Wise. EK: Comrie. AC. (2005). Meteorologically adjusted urban air quality trends in the Southwestern United
        States. Atmos Environ 39: 2969-2980. http://dx.doi.Org/10.1016/i.atmosenv.2005.01.024.
Worden, HM: Logan, JA: Worden, JR: Beer, R: Bowman, K: Clough, SA: Eldering, A: Fisher, BM: Gunson, MR:
        Herman. RL: Kulawik. SS: Lampel. MC: Luo. M: Megretskaia. IA: Osterman. GB: Shephard. MW.
        (2007a). Comparisons of Tropospheric Emission Spectrometer (TES) ozone profiles to ozonesondes:
        Methods and initial results. J Geophys Res 112: D03309. http://dx.doi.org/10.1029/2006JD007258.
Worden. J: Liu. X: Bowman. K: Chance. K: Beer. R: Eldering. A: Gunson. M: Worden. H. (2007b). Improved
        tropospheric ozone profile retrievals using OMI and TES radiances. Geophys Res Lett 34: L01809.
        http://dx.doi. org/10.1029/2006GL027806.
Wu. S: Micklev. LJ: Leibensperger. EM: Jacob. DJ: Rind. D: Streets. DG. (2008a).  Effects of 2000-2050 global
        change on ozone air quality in the United States. J Geophys Res 113: D06302.
        http://dx.doi. org/10.1029/2007JD008917.
Yang. Q: Cunnold. DM: Choi. Y: Wang. Y: Nam. J: Wang. HJ: Froidevaux. L: Thompson. AM: Bhartia. PK.
        (2010). A study of tropospheric ozone column enhancements over North America using satellite data
        and a global chemical transport model. J Geophys Res 115: D08302.
        http://dx.doi.org/10.1029/2009JD012616.
Yung. YL: Pinto. JP: Watson. RT:  Sander. SR (1980). Atmospheric bromine and ozone perturbations in the
        lower stratosphere. J Atmos Sci 37: 339-353.
Zhang. K: Wexler. A. (2008).  Modeling urban and regional aerosols: Development  of the UCD Aerosol Module
        and implementation in CMAQ model. Atmos Environ 42: 3166-3178.
Zhang, L: Jacob, DJ: Downey, NV: Wood, DA: Blewitt, D: Carouge, CC: Van donkelaar, A: Jones, DBA: Murray,
        LT: Wang. Y. (In Press) Improved estimate of the policy-relevant background ozone in the United States
        using the GEOS-Chem global model with 1/2° * 2/3° horizontal resolution over North America. Atmos
        Environ. http://dx.doi.Org/10.1016/i.atmosenv.2011.07.054.
Zhang. L: Jacob. DJ: Boersma. KF: Jaffe. DA: Olson. JR: Bowman. KW: Worden. JR: Thompson. AM: Averv.
        MA: Cohen. RC: Dibb. JE: Flock. FM: Fuelberg. HE: Huev. LG: McMillan.  WW: Singh. HB: Weinheimer.
        AJ. (2008). Transpacific transport of ozone pollution and the effect of recent Asian emission increases
        on air quality in North America: An integrated analysis using satellite, aircraft, ozonesonde, and surface
        observations. Atmos Chem Phys 8: 6117-6136.
Zhang, L: Jacob, DJ: Logan, JA: Chance, K: Eldering, A: Bojkov, BR.  (201 Ob). Intercomparison methods for
        satellite measurements of atmospheric composition: Application to tropospheric ozone from TES and
        OMI. Atmos Chem Phys 10: 4725-4739. http://dx.doi.org/10.5194/acpd-10-1417-2010.
Zhang, Q: Jimenez, Jl_: Canagaratna,  MR: Jayne, JT: Worsnop, DR. (2005). Time- and size-resolved chemical
        composition of submicron particles in Pittsburgh: Implications for aerosol sources and processes. J
        Geophys Res 110: 1-19.
Zhang, X: Zhuang, G: Guo, J: Yin, K:  Zhang, P. (2007b). Characterization of aerosol over the Northern South
        China  Sea during two cruises in 2003. Atmos Environ  41: 7821-7836.
Ziemke. JR: Chandra. S: Duncan. BN: Froidevaux. L: Bhartia. PK: Levelt. PF: Waters. JW. (2006). Tropospheric
        ozone determined from Aura OMI and MLS: Evaluation of measurements and  comparison with the
        Global Modeling Initiative's Chemical Transport Model. J Geophys Res 111: D19303.
        http://dx.doi. org/10.1029/2006JD007089.
Zimmermann, J: Poppe, D. (1993). Nonlinear chemical couplings in the tropospheric NOx-HOx gas phase
        chemistry. J Atmos Chem 17: 141-155.
Draft - Do Not Cite or Quote                      3-215                                 September 2011

-------
      4  EXPOSURE  TO  AMBIENT  OZONE
          4.1    Introduction

 1                  The 2006 O3 AQCD evaluated O3 concentrations and exposures in multiple
 2                  microenvironments, discussed methods for estimating personal and population exposure
 3                  via monitoring and modeling, analyzed relationships between personal exposure and
 4                  ambient concentrations, and discussed the implications of using ambient O3
 5                  concentrations as an estimate of exposure in epidemiologic studies. This chapter presents
 6                  new information regarding exposure to ambient O3 in the context of existing relevant
 7                  information summarized in the 2006 O3 AQCD, which in many areas remains definitive.
 8                  A brief summary of findings from the 2006 O3 AQCD is presented at the beginning of
 9                  each section as appropriate.

10                  Section 4.2 presents general exposure concepts describing the relationship between
11                  ambient pollutant concentrations and personal exposure. Section 4.3 describes exposure
12                  measurement techniques and studies that measured personal, ambient, indoor, and
13                  outdoor concentrations of O3 and related pollutants. Section 4.4 presents material on
14                  parameters relevant to exposure estimation, including activity patterns, averting behavior,
15                  and population proximity to ambient monitors. Section 4.5 describes techniques for
16                  modeling local O3 concentrations, air exchange rates, microenvironmental
17                  concentrations, and personal and population exposure. Section 4.6 discusses the
18                  implications of using  ambient O3 concentrations to estimate exposure in epidemiologic
19                  studies, including several factors that contribute to exposure error.
          4.2    General Exposure Concepts

20                  A theoretical model of personal exposure is presented to highlight measurable quantities
21                  and the uncertainties that exist in this framework. An individual's time-integrated total
22                  exposure to O3 can be described based on a compartmentalization of the person's
23                  activities throughout a given time period:
                                              ET =    Cj dt

                                                                                         Equation 4-1

24                  where ET = total (T) exposure over a time-period of interest, C; = airborne O3
25                  concentration at microenvironment/, and dt = portion of the time-period spent in


      Draft - Do Not Cite or Quote                      4-1                                 September 2011

-------
 1                   microenvironment/ Equation 4-1 can be decomposed into a model that accounts for
 2                   exposure to O3, of ambient (Ea) and nonambient (Ena) origin of the form:

                                              P  —  P  -4-  P
                                              CT —  ca i   cna
                                                                                          Equation 4-2
 3                   Ambient O3 is formed through photochemical reactions involving NOX, VOCs, and other
 4                   compounds, as described in Chapter 3. Although nonambient sources of O3 exist, such as
 5                   O3 generators and laser printers, these sources are specific to individuals  and may not
 6                   represent important sources of population exposure. Ozone concentrations generated by
 7                   ambient and nonambient sources are subject to spatial and temporal variability that can
 8                   affect estimates of exposure and influence epidemiologic effect estimates. Exposure
 9                   parameters affecting interpretation of epidemiologic studies are discussed in Section 4.5.

10                   This assessment focuses on the ambient component of exposure because this is more
1 1                   relevant to the NAAQS review. Assuming steady-state outdoor conditions, Ea can be
12                   expressed in terms of the fraction of time spent in various outdoor and indoor
13                   microenvironments (Wallace  et al.. 2006; Wilson et al.. 2000):
                                        ^a    2jf o^o  '  2j J i-' infi^o,i

                                                                                          Equation 4-3

14                   where /= fraction of the relevant time period (equivalent to dt in Equation 4-1), subscript
15                   o = index of outdoor microenvironments, subscript / = index of indoor
16                   microenvironments, subscript o,i = index of outdoor microenvironments adjacent to a
1 7                   given indoor microenvironment /', and F^j = infiltration factor for indoor
1 8                   microenvironment (i). Equation 4-3 is subject to the constraint ~Lf0 + Z/j =  1 to reflect the
19                   total  exposure over a specified time period, and each term on the right hand side of the
20                   equation has a summation because it reflects various microenvironmental exposures.
21                   Here, "indoors" refers to being inside any aspect of the built environment,  e.g., home,
22                   office buildings, enclosed vehicles (automobiles, trains, buses), and/or recreational
23                   facilities (movies, restaurants, bars). "Outdoor" exposure can occur in parks or yards, on
24                   sidewalks, and on bicycles or motorcycles. F^ is a function of the building air exchange
25                   characteristics. Assuming steady state ventilation conditions, the infiltration factor is a
26                   function of the penetration (P) of O3 into the microenvironment, the air exchange  rate (a)
27                   of the microenvironment, and the rate of O3 loss (k) in the microenvironment; .Fmf =
28                   Pal(a+k).

29                   In epidemiologic studies, the central-site ambient concentration, Ca, is often used  in lieu
30                   of outdoor microenvironmental data to represent these exposures based on the availability
      Draft - Do Not Cite or Quote                       4-2                                 September 2011

-------
 1                   of data. Thus it is often assumed that C0 = Ca and that the fraction of time spent outdoors
 2                   can be expressed cumulatively as/0; the indoor terms still retain a summation because
 3                   infiltration differs among different microenvironments. If an epidemiologic study
 4                   employs only Ca, then the assumed model of an individual's exposure to ambient O3,
 5                   first given in Equation 4-3, is re-expressed solely as a function of Ca:
                                         Ea=  (f0+
                                                                                           Equation 4-4
 6                   The spatial variability of outdoor O3 concentrations due to meteorology, varying
 7                   precursor emissions and O3 formation rates; design of the epidemiologic study; and other
 8                   factors determine whether or not Equation 4-4 is a reasonable approximation for
 9                   Equation 4-3. Errors and uncertainties inherent in use of Equation 4-4 in lieu of
10                   Equation 4-3 are described in Section 4.6 with respect to implications for interpreting
11                   epidemiologic studies. Epidemiologic  studies often use concentration measured at a
12                   central site monitor to represent ambient concentration; thus a, the ratio between personal
13                   exposure to ambient O3 and the ambient concentration of O3, is defined as:

                                                       Ea
                                                  a= —
                                                       ua

                                                                                          Equation 4-5
14                   Combination of Equation 4-4 and Equation 4-5 yields:

                                            "= fo+ ZfiFinn

                                                                                          Equation 4-6
15                   where a varies between 0 and 1. If a person's exposure occurs in  a single
16                   microenvironment, the ambient component of a microenvironmental O3 concentration
17                   can be represented as the product of the ambient concentration and Fmf. Wallace et al.
18                   (2006) note that time-activity data and corresponding estimates of F^ for each
19                   microenvironmental exposure are needed to compute an individual's a with accuracy. In
20                   epidemiologic studies, a is assumed to be constant in lieu of time-activity data and
21                   estimates of Finf, which can vary with building and meteorology-related air exchange
22                   characteristics. If local outdoor sources and sinks exist and are significant but not
23                   captured by central site monitors, then the ambient component of the local outdoor
24                   concentration may be estimated using dispersion models, land use regression models,
25                   receptor models, fine scale CTMs or some combination of these techniques. These
26                   techniques are described in Section 4.5.


      Draft - Do Not Cite or Quote                       4-3                                 September 2011

-------
          4.3    Exposure Measurement

 1                  This section describes techniques that have been used to measure microenvironmental
 2                  concentrations of O3 and personal O3 exposures as well as results of studies using those
 3                  techniques. Previous studies from the 2006 O3 AQCD are described along with newer
 4                  studies that evaluate indoor-outdoor concentration relationships, associations between
 5                  personal exposure and ambient monitor concentration, and multipollutant exposure to
 6                  other pollutants in conjunction with O3. Tables are provided to summarize important
 7                  study results.
            4.3.1   Personal Monitoring Techniques

 8                  As described in the 2006 O3 AQCD, passive samplers have been developed and deployed
 9                  to measure personal exposure to O3 (Grosjean and Hisham. 1992; Kanno and
10                  Yanagisawa. 1992). Widely used versions of these samplers utilize a filter coated with
11                  nitrite, which is converted to nitrate by O3 and then quantified by a technique such as ion
12                  chromatography (Koutrakis et al.. 1993). This method has been licensed and marketed by
13                  Ogawa, Inc., Japan (Ogawa and Co. 2007). The cumulative sampling and the detection
14                  limit of the passive badges makes them mainly suitable for monitoring periods of 24
15                  hours or greater, which limits their ability to measure short-term daily fluctuations in
16                  personal O3 exposure. Longer sampling periods give lower detection limits; use of the
17                  badges for a 6-day sampling period yields a detection limit of 1 ppb, while a 24-hour
18                  sampling period gives a detection limit of approximately 5-10 ppb. This can result in a
19                  substantial fraction of daily samples being below the detection limit (Sarnat et al.. 2006b:
20                  Sarnat et al., 2005). which is a limitation of past and current exposure studies.
21                  Development of improved passive samplers capable of shorter-duration monitoring with
22                  lower detection limits would enable more precise characterization of personal exposure in
23                  multiple microenvironments with relatively low participant burden.

24                  The nitrite-nitrate conversion reaction has also been used as the basis for an active
25                  sampler consisting of a nitrite-coated  glass tube through which air is drawn by a pump
26                  operating at 65 mL/min (Geyh et al.,  1999; Geyh et al.. 1997). The reported detection
27                  limit is 10 ppb-h, enabling the quantification of O3 concentrations measured over a few
28                  hours rather than a full day (Geyhetal.. 1999).

29                  A portable active O3 monitor based on the UV photometric technique used for stationary
30                  monitors (Chapter 3)  has recently been approved as a FEM (75 FR 22126). This monitor
31                  includes a Nafion tube in the inlet line to equalize humidity, reducing the effect of
32                  humidity changes in different microenvironments (Wilson and Birks. 2006). Its size and
      Draft - Do Not Cite or Quote                      4-4                                September 2011

-------
 1                   weight (approximately 10x20x30 cm; 2 kg) make it suitable for use in a backpack
 2                   configuration. The monitors are currently used by the U.S. National Park service as
 3                   stationary monitors to measure O3 in several national parks (Chapter 3). Future
 4                   improvements and continued miniaturization of real-time O3 monitors can yield highly
 5                   time-resolved personal measurements to further evaluate O3 exposures in specific
 6                   situations, such as near roadways or while in transit.
            4.3.2   Indoor-Outdoor Concentration Relationships

 7                   Several studies summarized in the 2006 O3 AQCD, along with some newer studies, have
 8                   evaluated the relationship between indoor O3 concentration and the O3 concentration
 9                   immediately outside the indoor microenvironment. These studies show that the indoor
10                   concentration is often substantially lower than the outdoor concentration unless indoor
11                   sources are present. Low indoor O3 concentrations can be explained by reactions of O3
12                   with surfaces and airborne constituents. Studies have shown that O3 is deposited onto
13                   indoor surfaces where reactions produce secondary pollutants such as formaldehyde
14                   (Reiss et al., 1995a; Reiss et al., 1995b). However, the indoor-outdoor relationship is
15                   greatly affected by the air exchange rate; under conditions of high air exchange rate, such
16                   as open windows, the indoor O3 concentration may approach the outdoor concentration.
17                   Table 4-1 summarizes indoor-outdoor (I/O) ratios and correlations reported by older and
18                   more recent studies, with discussion of individual studies in the subsequent text. In
19                   general, I/O ratios range from about 0.1 to 0.4, with some evidence for higher ratios
20                   during the O3 season when concentrations are higher.

21                   O3 concentrations near and below the monitor detection limit cause uncertainty in I/O
22                   ratios, because  small changes in low concentration values cause substantial variation in
23                   resulting ratios. This problem is particularly acute in the non-ozone season when ambient
24                   O3 concentrations are low. Further improvements in characterization of
25                   microenvironmental O3 concentrations and I/O ratios will rely on improved monitoring.
26                   Until new monitoring techniques are available and can be used in the field, past studies
27                   summarized in the 2006 O3 AQCD remain relevant to consider along with more recent
28                   studies in evaluating the relationship between indoor and outdoor O3 concentrations.
      Draft - Do Not Cite or Quote                       4-5                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
Table 4-1 Relationships of Indoor and Outdoor Ozone Concentration
Study
Geyh etal.
(2000)
Avol etal.
Q998b)
Romieu et
al. (1998b)
Lee et al.
(2004a)
Heroux et
al. (2010)
Lopez-
Aparicio et
al. (2011)
Liu etal.
(1995)
Romieu et
al. (1998b)
Blondeau et
al. (2005)
Riediker et
al. (2003)
Location
Upland, Southern
California
Mountain
Communities,
Southern California
Southern California
Mexico City, Mexico
Nashville, TN
Regina,
Saskatchewan,
Canada
Prague, Czech
Republic
Toronto, Canada
Mexico City, Mexico
La Rochelle, France
North Carolina
Years/Season
June - September
1995 and May 1996
October 1995-April
1996
June - September
1995 and May 1996
October 1995-April
1996
February-
December, 1994
Summer
Non-summer
September 1993 -
July 1994
Summer 1994
Summer 2007
July 2009 - March
2010
Winter, 1992
Summer, 1992
Summer, 1992
Summer, 1992
September 1993 -
July 1994
Spring, 2000
August - October
2001
Population
Children
NR
Children
Children
All age groups
NR
All age groups
Children
Children (during
school hours)
Children
Adults
Sample
duration
6 day
24 h
7 or 14 day
1 week
5 day
1 month
1 week
12h
24 h/day,
1 4 days
5 h/day, 10
days
NR
9h
Ratio3 Correlation enJK;ent
0.24 NR Home
"0~15
0.37 0.58 Home
SD: 0.25
0.43 NR
SD: 0.29
0.32 NR
SD:0.21
0.20 NR Home
0.1 5b
Range:
0.01-
1.00
0.1 NR Home
0.13 NR Home
0.10- NR Home
0.30
0.07 NR Home
SD:0.10
0.40
SD: 0.29
0.30
SD: 0.32
SD: 0.54
0.15 NR School
0.30-
0.40
Range: NR School
0.00-
0.45
0.51 NR Vehicle
Others
Air-conditioned
Opening
windows




No heating or
air conditioning


Daytime
Nighttime

Immediately
outside the
schools


      " Mean value unless otherwise indicated
      b Median
      NR = not reported
      SD = standard deviation
Geyh et al. (2000) measured 6-day indoor and outdoor concentrations at 116 homes in
southern California, approximately equally divided between the community of Upland
and several mountain communities. The extended sampling period resulted in a relatively
low detection limit (1 ppb) for the passive samplers used. The Upland homes were nearly
all air-conditioned, while the mountain community homes were ventilated by opening
windows. During the O3 season, the indoor O3 concentration averaged over all homes
was approximately 24% of the overall mean outdoor concentration in Upland
(11.8 versus 48.2 ppb), while in the mountain communities, the indoor concentration was
36% of the outdoor concentration (21.4 versus 60.1 ppb). This is consistent with the
increased air exchange rate expected in homes using window ventilation. In the non-
ozone season, when homes are likely to be more tightly closed to conserve heat, the ratios
      Draft - Do Not Cite or Quote
                                4-6
September 2011

-------
 1                   of indoor to outdoor concentration were 0.15 (3.2 versus 21.1 ppb) and 0.08 (2.8 versus
 2                   35.7 ppb) in Upland and the mountain communities, respectively. Avol et al. (1998b)
 3                   observed a mean I/O ratio of 0.37 for 239 matched 24-h samples collected between
 4                   February and December at homes in the Los Angeles area. The I/O ratio during summer
 5                   was higher than the non-summer I/O ratio (0.43 versus 0.32). The authors also reported a
 6                   correlation of 0.58 between the 24-h avg indoor concentration and the outdoor
 7                   concentration, which was only slightly higher than the correlation between the indoor
 8                   concentration and the concentration at the neighborhood fixed-site monitor (0.49).
 9                   Substantially higher summer I/O ratios were reported in a study in Toronto (Liu et al..
10                   1995). which found summer I/O ratios of 0.30-0.43, in comparison with a winter I/O ratio
11                   of 0.07. Romieu et al. (1998b) reported a mean I/O ratio of 0.20 in 145 homes in
12                   Mexico City for 7- or 14-day cumulative samples, with the highest ratios observed in
13                   homes where windows were usually open during the day and where there was no
14                   carpeting or air filters. Studies conducted in Nashville, TN and Regina, Saskatchewan
15                   reported mean residential I/O ratios of approximately 0.1 (Heroux et al.. 2010; Lee et al..
16                   20Q4a).

17                   Investigators have also measured I/O ratios for non-residential microenvironments,
18                   including schools and vehicles. Romieu et al. (1998b) reported that O3 concentrations
19                   measured during school hours (10-day cumulative sample, 5 h/day) were 30-40% of
20                   concentrations immediately outside the schools, while overall I/O ratios (14-day
21                   cumulative sample, 24 h/day) were approximately 15%. The authors attribute this
22                   discrepancy to increased air exchange during the school day due to opening doors and
23                   windows. Air exchange was also identified as an important factor in the I/O ratios
24                   measured at eight French schools (Blondeau et al.. 2005). In this study, the I/O ratios
25                   based on simultaneous continuous measurements  ranged from 0-0.45, increasing with
26                   decreasing building tightness. A historical library building in Prague, Czech Republic
27                   with no heating  or air conditioning (i.e., natural ventilation) was observed to have ratios
28                   of one-month indoor and outdoor concentrations ranging from 0.10-0.30 during a nine-
29                   month sampling campaign, with the highest ratios reported in Nov-Dec 2009 and the
30                   lowest ratios during Jul-Aug 2009 (Lopez-Aparicio et al.. 2011). Indoor concentrations
31                   were relatively constant (approximately 3-7 ug/m3 or 2-3 ppb), while outdoor
32                   concentrations were lower in the winter (9-10 ug/ m3 or about 5 ppb) than in the summer
33                   (35-45 ug/ m3 or about 20 ppb). This seasonal variation in outdoor concentrations
34                   coupled with homogeneous indoor concentrations, together with increased wintertime air
3 5                   exchange rate due to higher indoor-outdoor temperature differences,  is likely responsible
36                   for the observed seasonal pattern in I/O ratios.

37                   Exposures in near-road, on-road and in-vehicle microenvironments are likely to be highly
3 8                   variable and lower than those in other microenvironments due to reaction of O3 with NO
      Draft - Do Not Cite or Quote                       4-7                                 September 2011

-------
 1                   and other combustion emissions. Depending on wind direction, O3 concentrations near
 2                   the roadway have been found to be 20-80% of ambient concentrations at sites 400 m or
 3                   more distant from roads (Section 3.6.2.1). A study on patrol cars during trooper work
 4                   shifts reported in-vehicle 9-h concentrations that were approximately 51% of
 5                   simultaneously measured roadside concentrations (mean of 11.7 versus 22.4 ppb)
 6                   (Riedikeretal.. 2003).
            4.3.3   Personal-Ambient Concentration Relationships

 7                   Several factors influence the relationship between personal O3 exposure and ambient
 8                   concentration. Due to the lack of indoor O3 sources, along with reduction of ambient O3
 9                   that penetrates into enclosed microenvironments, indoor and in-vehicle O3 concentrations
10                   are highly dependent on air exchange rate and therefore vary widely in different
11                   microenvironments. Ambient O3 varies spatially due to reactions with other atmospheric
12                   species, especially near busy roadways where O3 concentrations are decreased by
13                   reaction with NO (Section  3.6.2.1). This is in contrast with pollutants  such as CO and
14                   NOX, which show appreciably higher concentrations near the  roadway than several
15                   hundred meters away (Karner et al.. 2010). O3 also varies temporally  over multiple
16                   scales, with a generally increasing trend during the daytime hours, and higher O3
17                   concentrations during summer than in winter. An example of this variability is shown in
18                   Figure 4-1, taken from a personal exposure study conducted by Chang et al. (2000).

19                   Hourly personal exposures are seen to vary from a few ppb in some indoor
20                   microenvironments to tens of ppb in vehicle and outdoor microenvironments. The
21                   increase in ambient O3 concentration during the day is apparent from  the outdoor
22                   monitoring data. In comparison, ambient PM2 5 exhibits less temporal variability over the
23                   day than O3,  although personal exposure to PM2 5 also varies by microenvironment.  This
24                   combined spatial and temporal variability for O3 results in varying relationships between
25                   personal exposure and ambient concentration.

26                   Correlations between personal exposure to O3 and corresponding ambient concentrations,
27                   summarized in Table 4-2, exhibit a wide range (generally 0.3-0.8, although both higher
28                   and lower values have been reported), with higher correlations generally observed in
29                   outdoor microenvironments, high building ventilation conditions, and during the summer
30                   season. Low O3  concentrations indoors and during the winter  lead to a high proportion of
31                   personal exposures below the sampler detection limit, which may partially explain the
32                   low correlations observed in some studies under those conditions. Ratios of personal
33                   exposure to ambient concentration, summarized in Table 4-3,  are generally lower in
34                   magnitude (typically 0.1-0.3), and are also variable, with increasing time spent outdoors
      Draft - Do Not Cite or Quote                       4-8                                 September 2011

-------
 1
 2
 3
associated with higher ratios. The next two subsections describe studies that have
reported personal-ambient correlations and slopes for a variety of seasons, locations, and
populations.
   90
   80
   70


ob 50
^.
"a 40
   30
   20
   10
    0
                              O
                                   Personal and Outdoor PM2 s and O.,:
                                     Baltimore, MD, August 12, 1998
                                                             .0»»0""<
                                                         £r
                                                                                  - 50
                                                                                  - 10
                                          10   11   12   13   14   15   16  17
                                                                             iy  20
                             Diking kilt:ncn study'  nxmi to«lih
                                    TV room to mom  chin-
                                               Clock Hour (KST)

       Source: Reprinted with permission of Air and Waste Management Association (Chang et al.. 2000)
       The notation below each clock hour shows the location or activity during that hour.

      Figure 4-1     Variation in hourly personal and ambient concentrations of Os and
                      PM2.5 in various microenvironments during daytime hours.
 4
 5
 6
 7
 8
 9
10
11
12
13
14
O3 concentrations near and below the passive sampler detection limit lead to uncertainty
in personal-ambient correlations and ratios. Correlations are reduced in magnitude by
values below the detection limit because noise obscures the underlying signal in the data,
while ratios tend to fluctuate widely at low concentration since small changes in
measured values cause large relative changes in resulting ratios. As with I/O ratios, this
problem is particularly acute in the non-ozone season when ambient O3 concentrations
are low. Improved characterization of the relationship between personal exposure and
ambient concentration will depend on improved monitoring techniques to accurately
capture low O3 concentrations, preferably at high time resolution to facilitate evaluation
of the effect of activity pattern on exposure. Until new monitoring techniques are
available, past studies summarized in the 2006 O3 AQCD remain relevant to consider
      Draft - Do Not Cite or Quote
                                4-9
September 2011

-------
 1                   along with more recent studies in evaluating personal-ambient concentration
 2                   relationships.

 3                   Personal-Ambient Correlations. Correlations between personal exposure and ambient
 4                   O3 concentrations have been evaluated in several research studies, many of which were
 5                   conducted prior to 2005 and are discussed in the 2006 O3 AQCD. Some studies evaluated
 6                   subject-specific, or longitudinal correlations, which describe multiple daily measurements
 7                   for a single individual. These studies indicate the inter-individual variability of personal-
 8                   ambient correlations. Another type of correlation is a pooled correlation, which combines
 9                   data from multiple individuals over multiple monitoring periods (e.g., days), providing an
10                   overall indicator of the personal-ambient relationship for all study subjects. A third type
11                   of correlation is a community-average correlation, which correlates average exposure
12                   across all study subjects with fixed-site monitor concentrations. Community-average
13                   correlations are particularly informative for interpreting time-series epidemiologic
14                   studies, in which ambient concentrations are used as a surrogate for community-average
15                   exposure. However, few studies report this metric; this represents another opportunity for
16                   improvement of future personal exposure studies. Table 4-2 summarizes studies reporting
17                   personal-ambient correlations, and the studies in the table are discussed in the subsequent
18                   text.

19                   The results  of these studies indicate that personal exposures are moderately well
20                   correlated with ambient concentrations, and that the  ratio of personal exposure to ambient
21                   concentration is higher in outdoor microenvironments and during the summer season. In
22                   situations where a lack of correlation was observed, this may be due in part to a high
23                   proportion of personal measurements below the detection limit. The effect of season is
24                   unclear, with mixed evidence on whether higher correlations are observed during the O3
25                   season. Chang et al.  (2000) measured hourly personal exposures in multiple
26                   microenvironments and found that the pooled correlation between personal exposure and
27                   ambient concentration was highest for outdoor microenvironments (r = 0.68-0.91). In-
28                   vehicle microenvironments showed moderate to high correlations (0.57-0.72).
29                   Correlations in residential indoor microenvironments were  very low (r = 0.05-0.09), with
30                   moderate correlations (0.34-0.46) in other indoor microenvironments such as restaurants
31                   and shopping malls. Liard et al. (1999) evaluated community-average correlations based
32                   on 4-day mean personal O3 exposure measurements for adults and children and found a
33                   relatively high correlation (r = 0.83) with ambient concentrations, even though  31-82% of
34                   the personal measurements were below the detection limit.  Sarnat et al. (2000) studied  a
3 5                   population of older adults in Baltimore and found that longitudinal correlations between
36                   24-h personal exposure and ambient concentration varied by subject and season, with
37                   somewhat higher correlations observed in this study during summer (mean = 0.20) than
38                   in winter (mean = 0.06).  Some evidence was presented that subjects living in well-
      Draft - Do Not Cite or Quote                       4-10                                 September 2011

-------
 1
 2
 3
 4
 5
ventilated indoor environments have higher correlations than those living in poorly
ventilated indoor environments, although exceptions to this were also observed. Ramirez-
Aguilar et al. (2008) measured 48- to 72-h personal exposures of four groups of asthmatic
children aged 6-14 in Mexico City during 1998-2000. A moderate pooled correlation (r =
0.35) was observed between these exposures and corresponding ambient concentrations.
 6
 7
 8
 9
10
11
12
13
14
Table 4-2 Correlations between Personal and Ambient Ozone Concentration
Study Location
Chang et al. Baltimore, MD
(2000)
Liard et al. Paris, France
(1999)
Sarnat et al. Baltimore, MD
(2000)
Linn et al. Southern
(1996) California
Brauerand Vancouver,
Brook (1997) Canada
Ramirez- Mexico City,
Aguilar et al. Mexico
(2008)
Years/Season Population fufs^m
Summer 1998 Older adults 1h
Winter 1999
Summer 1998
Winter 1999
Summer 1998
Winter 1999
Summer 1998
Winter 1999
Summer 1998
Winter 1999
Summer 1996 All age groups 4 day
Summer Older adults 24 h
Winter
All seasons from Children 24 h
1992to1993
Summers 1992 and Health clinic 24 h
1993 workers
Camp counselors 24 h
Farm workers 24 h
December 1 998- Asthmatic 48 h to 72
April 2000 children h
Correlation
0.91
0.77
0.68
0.86
0.72
0.57
0.09
0.05
0.34
0.46
0.83
0.20
0.06
0.61
0.60
0.42
0.64
0.35
Study Type Others
Pooled Outdoor near
roadway

Outdoor away
from road

In vehicle

Indoors-
residence

Indoors-other

Community-
averaged
Longitudinal

Community-
averaged
Pooled 0-25% of time
outdoors
Pooled 7.5-45% of
time outdoors
Pooled 100% of time
outdoors
Pooled
       NR = not reported
Consistent with hourly microenvironment-specific results from the Chang et al. (2000)
study described above, studies have found moderate to high personal-ambient
correlations for individuals spending time outdoors. A moderate pooled correlation of
0.61 was reported between 24-h avg personal and central-site measurements by Linn et
al. (1996) for a population of southern California schoolchildren who spent an average of
101-136 minutes per day outdoors. The authors also report a correlation of 0.70 between
central-site measurements and concentrations outside the children's schools. Although
the average school outdoor concentration (34 ppb) was higher than the  average central-
site concentration (23 ppb) and the average personal exposure concentration was lower
      Draft - Do Not Cite or Quote
                                4-11
September 2011

-------
 1                   (5 ppb) than the central-site value, the similarity between the correlations indicate that
 2                   central-site monitor concentrations can represent personal exposures in addition to
 3                   representing local outdoor concentrations. A study in Vancouver, BC provided another
 4                   illustration of the effect of outdoor microenvironments on personal-ambient relationships
 5                   by comparing three groups spending different amounts of time outdoors: health clinic
 6                   workers (0-25% of time outdoors), camp counselors (7.5-45% of time outdoors), and
 7                   farm workers (100% of time outdoors) (Brauer and Brook. 1997). Health clinic workers
 8                   and camp counselors were monitored 24 h/day, while farm workers were monitored
 9                   during their work shift (6-14 hours). In this study, the pooled correlations between
10                   personal exposure and fixed-site concentration not substantially different among the
11                   groups (r = 0.60, 0.42, and 0.64, respectively). The ratios of personal exposure to fixed-
12                   site monitor concentration increased among the groups with increasing amount of time
13                   spent outdoors (0.35, 0.53, and 0.96, respectively). This indicates that temporal variations
14                   in personal exposure to O3 are driven by variations in ambient concentration, even for
15                   individuals that spend little time outdoors.

16                   Personal-Ambient Ratios. Studies indicate that the ratio between personal O3 exposure
17                   and ambient concentration varies widely, depending on activity patterns, housing
18                   characteristics, and season. Higher personal-ambient ratios are generally observed with
19                   increasing time spent outside, higher air exchange rate, and in seasons other than winter.
20                   Table 4-3 summarizes the results of several such studies  discussed in the 2006 O3 AQCD
21                   together with newer studies showing the same pattern of results.

22                   O'Neill et al. (2003) studied a population of shoe cleaners working outdoors in
23                   Mexico City and presented a regression model indicating a 0.56 ppb increase in 6-h
24                   personal exposure for each 1  ppb increase in ambient concentration. Regression analyses
25                   by Sarnat et al. for 24-h data from mixed populations of children and older adults in
26                   Baltimore (2001) and Boston (2005) found differing results between the two cities, with
27                   Baltimore subjects showing a near-zero slope (0.01) during the summertime while Boston
28                   subjects showed a positive slope of 0.27 ppb personal exposure per  1 ppb ambient
29                   concentration. In both cities, the winter slope was near zero. The low slope observed in
30                   Baltimore may have been due to differences in time spent outdoors, residential
31                   ventilation conditions, or other factors. Xue et al. (2005) measured 6-day personal
32                   exposure of children in southern California and found that the average ratio of personal
33                   exposure to ambient concentration was relatively stable throughout the year at 0.3. These
34                   authors also regressed personal exposures on ambient concentration after adjusting for
35                   time-activity patterns and housing characteristics and found a slope  of 0.54 ppb/ppb, with
36                   the regression R2 value of 0.58. Unadjusted regression slopes were not presented.
      Draft - Do Not Cite or Quote                       4-12                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
Table 4-3
Study
Sarnat et al.
(2001)
Brauerand Brook
(1997)
O'Neill etal.
(2003)
Sarnat et al.
(2005)
Xue et al. (2005)
Sarnat et al.
(2006b)
Ramirez-Aguilar
etal. (2008)
Ratios of Personal to Ambient Ozone Concentration
Location
Baltimore
Vancouver,
Canada
Mexico City,
Mexico
Boston
Southern
California
Steubenville,
OH
Mexico City,
Mexico
Years/Season
Summer 1998
Winter 1999
Summers 1992
and 1993
April -July 1996
Summer
Winter
June 1995 -May
1996
Summer
Fall
Dec.1998-Apr.
2000
Population
Older adults, children, and
individuals with COPD
Health clinic workers
Camp counselors
Farm workers
Shoe cleaners
Older adults and children

Children
Older adults

Asthmatic children
Sample
duration
24 h
24 h


6h
24 h
6 day
24 h
48hto
72 h
Ratio
0.01
0.00
0.35
0.53
0.96
0.56
95% Cl:
0.43-0.69
0.27
95% Cl:
0.18-0.37
0.04
95% Cl:
0.00-0.07
0.3
0.54
0.15
SE: 0.02
0.08
SE: 0.04
0.27
SE: 0.03
0.20
SE: 0.05
0.17
Study
Type
Longitudinal

Pooled
Pooled
Pooled
Longitudinal
Longitudinal

Longitudinal

Longitudinal



Pooled
Others


0-25% of time
outdoors
7.5-45% of time
outdoors
100% of time
outdoors


Ratio
Repression slope
(Rf= 0.58)
High-ventilation
Low-ventilation
High-ventilation
Low-ventilation

       NR = not reported
A few additional studies have been published since the 2006 O3 AQCD comparing
personal exposures with ambient concentrations, and these findings generally confirm the
conclusions of the 2006 O3 AQCD that ventilation conditions, activity pattern, and
season may impact personal-ambient ratios. Sarnat et al. (2006b) measured 24-h personal
exposures for a panel of older adults in Steubenville, OH during summer and fall 2000.
Subjects were classified as high-ventilation or low-ventilation based on whether they
spent time in indoor environments with open windows. Regression of personal exposures
on ambient concentration found a higher slope for high-ventilation subjects compared
with low-ventilation subjects in both summer (0.15 versus 0.08) and fall (0.27 versus
0.20). Suh and Zanobetti (2010) reported an average 24-h personal exposure  of 2.5 ppb as
compared to 24-h ambient concentration of 29 ppb for a group of individuals with either
recent MI or diagnosed COPD in Atlanta. A similar result was observed in Detroit, where
the mean 24-h personal exposure across 137 participants in summer and winter was 2.1
ppb, while the mean ambient concentration on sampling days was 25 ppb (Williams et
al.. 2009b). Although no personal exposures were measured, McConnell et al. (2006)
found that average 24-h home outdoor O3 concentrations were within 6 ppb of O3
concentrations measured at central-site monitors in each of three southern California
communities, with a combined average home outdoor concentration of 33 ppb compared
to the central-site average of 36 ppb. In Mexico City, Ramirez-Aguilar et al.  (2008)
      Draft - Do Not Cite or Quote
                               4-13
September 2011

-------
 1                   regressed 48- to 72-h personal exposures of four groups of asthmatic children aged 6-14
 2                   with ambient concentrations and found slope of 0.17 ppb/ppb after adjustment for
 3                   distance to the fixed-site monitor, time spent outdoors, an interaction term combining
 4                   these two variables, and an interaction term representing neighborhood and study group.
            4.3.4   Co-Exposure to Other Pollutants and Environmental Stressors

 5                   Exposure to ambient O3 occurs in conjunction with exposure to a complex mixture of
 6                   ambient pollutants that varies over space and time. Multipollutant exposure is an
 7                   important consideration in evaluating health effects of O3 since these other pollutants
 8                   have either known or potential health effects that may impact health outcomes due to O3
 9                   The co-occurrence of high O3 concentrations with high heat and humidity may also
10                   contribute to health effects.  This section presents data on relationships between overall
11                   personal O3 exposure and exposure to other ambient pollutants, as well as co-exposure
12                   relationships for near-road O3 exposure.
                     4.3.4.1    Personal Exposure to Ozone and Co-pollutants

13                   Personal exposure to O3 shows variable correlation with personal exposure to other
14                   pollutants, with differences in correlation depending on factors such as instrument
15                   detection limit, season, city-specific characteristics, and spatial variability of the
16                   copollutant. Suh and Zanobetti (2010) reported Spearman rank correlation coefficients
17                   during spring and fall between 24-h avg O3 measurements and co-pollutants of 0.14,
18                   0.00, and -0.03 for PM25, EC, and NO2, respectively. Titration of O3 near roadways is
19                   likely to contribute to the low or slightly negative correlations with the traffic-related
20                   pollutants EC and NO2. The somewhat higher correlation with PM2 5 may reflect the
21                   influence of air exchange rate and time spent outdoors on co-exposures to ambient PM2 5
22                   and O3. Overall, the copollutant correlations are quite small, which may be due to the
23                   very low personal exposures observed in this study (2-3 ppb), likely to be near or below
24                   the detection limit of the passive sampler over a 24-h period. Chang et al. (2000)
25                   measured hourly personal exposures to PM2 5 and O3 in summer and winter in Baltimore,
26                   Maryland. Correlations between PM2 5 and O3 were 0.05  and -0.28 in summer and
27                   winter, respectively.  Results indicate personal O3 exposures were not significantly
28                   associated with personal PM2 5 exposures in either summer or winter. These non-
29                   significant correlations may be attributed in part to the relatively low personal O3
30                   exposures observed in this study; in both summer and winter, the mean personal O3
31                   exposure was below the calculated limit of detection.
      Draft - Do Not Cite or Quote                      4-14                                September 2011

-------
 1                   Studies conducted in Baltimore (Sarnat et al., 2001) and Boston (Sarnat et al., 2005)
 2                   found differing results for the correlation between 24-h avg personal O3 and personal
 3                   PM2 5 exposures, particularly during the winter season. Sarnat et al. (2001) found a
 4                   positive slope when regressing personal exposures of both total PM2 5 (0.21) and PM2 5 of
 5                   ambient origin (0.22) against personal O3 exposures during the summer season, but
 6                   negative slopes (-0.05 and -0.18, respectively) during the winter season. The summertime
 7                   slope for personal PM2 5 exposure versus personal O3 exposure was much higher for
 8                   children (0.37) than for adults (0.07), which may be the result of different activity
 9                   patterns. This team of researchers  also found a positive, although higher, summer slope
10                   between 24-h avg personal O3 and personal PM25 in Boston (0.72) (Sarnat et al.. 2005).
11                   However, the winter slope was positive (1.25) rather than negative, as in Baltimore. In
12                   both cities during both seasons, there was a wide range of subject-specific correlations
13                   between personal O3 and personal PM25 exposures, with some subjects showing
14                   relatively strong positive correlations (>0.75) and others showing strong negative
15                   correlations (<-0.50). The median correlation in both cities was slightly positive in the
16                   summer and near zero (Boston) or slightly negative (Baltimore) in  the winter.  These
17                   results indicate the potential  effects of city-specific characteristics, such as housing stock
18                   and building ventilation patterns, on relationships between O3 and  co-pollutants.
                     4.3.4.2    Near-Road Exposure to Ozone and Co-pollutants

19                   Beckerman et al. (2008) measured both 1-week and continuous concentrations of O3,
20                   NO, NO2, NOX, PM25, PMLO, and several VOCs (the BTEX compounds, MTBE,
21                   hexane, and THC) in the vicinity of heavily traveled (annual average daily traffic
22                   [AADT] >340,000) roadways in Toronto, Canada. Passive samplers were deployed for
23                   one week in August 2004. Ozone concentrations were negatively correlated with all
24                   pollutants, with the exception of VOCs at one of the monitoring sites which were
25                   suspected of being influenced by small area sources. Site specific correlations are given
26                   in Figure 4-2.  Correlations were -0.77 to  -0.85 for NO2, -0.48 to -0.62 for NO, and -0.55
27                   to -0.63 for NOX. Pooled correlations using data from both sites were somewhat lower in
28                   magnitude. PM2 5 and PMi.0 correlations were -0.35 to -0.78 and -0.34 to -0.58,
29                   respectively. At the monitoring site not influenced by small area sources, O3-VOC
30                   correlations ranged from -0.41 to -0.66.
31                   Beckerman et al. (2008) also made on-road measurements of multiple pollutants with a
32                   instrumented vehicle. Concentrations were not reported,  but correlations between O3 and
33                   other pollutants were negative and somewhat greater in magnitude (i.e., more negative)
34                   than the near-road correlations. SO2, CO, and BC were measured in the mobile
35                   laboratory, although not at the roadside, and they all showed negative correlations with
      Draft - Do Not Cite or Quote                       4-15                                September 2011

-------
 1                  O3 when the data were controlled for site. Correlations for continuous concentrations
 2                  between O3 and co-pollutants were somewhat lower than the 1-week correlations, except
 3                  for O3-PM2 5 correlations. Correlations were -0.90, -0.66, -0.77, and -0.89 for NO2, NO,
 4                  NOX, and PMi 0 respectively. The continuous O3-PM2 5 correlation was -0.62, which is
 5                  in the range of the 1-week correlation.
          Copollutant
          NO2
          NO

          NOX
          PM2.5

          PMLo
          voc
              -0.9      -0.8      -0.7     -0.6      -0.5      -0.4      -0.3     -0.2      -0.1       0
                                          Pearson Correlation Coefficient
       Source data from: Beckerman et al. (2008)

      Figure 4-2     Correlations between 1-week concentrations of Os and
                      copollutants measured near roadways.
                    4.3.4.3    Indoor Exposure to Ozone and Co-pollutants

 6                  Ambient O3 that infiltrates indoors reacts with organic compounds and other chemicals to
 7                  form oxidized products, as described in Section 3.2.3 as well as the 2006 O3 AQCD. It is
 8                  anticipated that individuals are exposed to these reaction products, although no evidence
 9                  was identified regarding personal exposures. The reactions are similar to those occurring
10                  in the ambient air, as summarized in Chapter 3. For example, O3 can react with terpenes
11                  and other compounds from cleaning products, air fresheners, and wood products both in
12                  the gas phase and on surfaces to form particulate and gaseous species, such as
13                  formaldehyde (Chen et al.. 2011; Shu and Morrison. 2011; Aoki and Tanabe. 2007; Reiss
14                  et al.. 1995a). Ozone has also been shown to react with material trapped on HVAC filters
15                  and generate airborne products (Beko et al.. 2007; Hyttinen et al.. 2006). Potential
16                  oxygenated reaction products have been found to act as irritants (Anderson et al.. 2007).

      Draft - Do Not Cite or Quote                      4-16                                September 2011

-------
 1                   indicating that these reaction products may have health effects separate from those of O3
 2                   itself Weschler and Shields. 1997). Ozone may also react to form other oxidants, which
 3                   then go on to participate in additional reactions. White et al. (2010) found evidence that
 4                   HONO or other oxidants may have been present during experiments to estimate indoor
 5                   OH concentrations, indicating complex indoor oxidant chemistry. Rates of these reactions
 6                   are dependent on indoor O3 concentration, temperature, and air exchange rate, making
 7                   estimation of exposures to reaction products difficult.
          4.4   Exposure-Related Metrics

 8                   In this section, parameters are discussed that are relevant to the estimation of exposure,
 9                   but are not themselves direct measures of exposure. Time-location-activity patterns,
10                   including behavioral changes to avoid exposure, have a substantial influence on exposure
11                   and dose. Proximity of populations to ambient monitors may influence how well their
12                   exposure is represented by measurements at the monitors, although factors other than
13                   distance play an important role as well.
            4.4.1    Activity Patterns

14                   The activity pattern of individuals is an important determinant of their exposure.
15                   Variation in O3 concentrations among various microenvironments means that the amount
16                   of time spent in each location, as well as the level of activity, will influence an
17                   individual's exposure to ambient O3. The effect of activity pattern on exposure is
18                   explicitly accounted for in Equation 4-3 by the fraction of time spent in different
19                   microenvironments.

20                   Activity patterns vary both among and within individuals, resulting in corresponding
21                   variations in exposure across a population and overtime. Large-scale human activity
22                   databases, such as those developed for the National Human Activity Pattern Survey
23                   (NHAPS) (Klepeis et al.. 2001) or the Consolidated Human Activity Database (CHAD)
24                   (McCurdy et al., 2000), which includes NHAPS data together with other activity study
25                   results, have been designed to characterize exposure patterns among much larger
26                   population subsets than can be  examined during individual panel studies. The complex
27                   human activity patterns across the population (all ages) are illustrated in Figure 4-3
28                   (Klepeis et al., 2001), which is presented to illustrate the diversity of daily activities
29                   among the entire population as well as the proportion of time spent in each microen-
30                   vironment. For example, about 25% of the individuals reported being outdoors or in a
31                   vehicle between 2:00 and 3:00  pm, when daily O3 levels are peaking, although about half
      Draft - Do Not Cite or Quote                       4-17                                September 2011

-------
 1
 2
 3
 4

 5
 6
 7
 8
 9
10
11
12
13
14
15
16
of this time was spent in or near a vehicle, where O3 concentrations are likely to be lower
than ambient concentrations. Different patterns would be anticipated when breaking
down activity patterns only for subgroups such as children or the elderly. Population
exposures can be estimated using O3 concentration data in each microenvironment.

Longitudinal activity pattern information is also an important determinant of exposure, as
different people may exhibit different patterns of time spent outdoors over time due to
age, gender, employment, and lifestyle-dependent factors. These  differences may mani-
fest as higher mean exposures or more frequent high-exposure episodes some individuals.
The extent to which longitudinal variability in individuals contributes to the population
variability in activity and location can be quantified by the ratio of between-person
variance to total variance in time spent in different locations  and activities (the intraclass
correlation coefficient, ICC). Xue et al. (2004) quantified ICC values in time-activity data
collected by Harvard University for 160 children aged 7-12 years in Southern California
(Geyh et al.. 2000). For time spent outdoors, the ICC was approximately 0.15, indicating
that 15% of the variance in outdoor time was due to between-person differences. The ICC
value might be different for other population groups.
               5
               5
              I
               QJ
              a,
 o
 o
                        oooooooooooooooooooooooo
                        OOOOOOOOOOOppppppppppOpO
                     <>i  i—  o-i  rn  T  fi  & i-~ 66 6*1 O  —  r-j  i—  r-j  m  -rr  v> \c i~- co 0s, o  ^H  01

                                                  Time of Day

       Source: Reprinted with permission of Nature Publishing Group (Klepeis et al.. 2001).

      Figure 4-3     Distribution of time that NHAPS respondents spent in ten
                      microenvironments based on smoothed 1-min diary data.
      Draft - Do Not Cite or Quote
                               4-18
September 2011

-------
 1                   The EPA's National Exposure Research Laboratory (NERL) has consolidated the
 2                   majority of the most significant human activity databases into one comprehensive
 3                   database called the Consolidated Human Activity Database (CHAD). The current version
 4                   of CHAD contains data from nineteen human activity pattern studies (including NHAPS),
 5                   which were evaluated to obtain over 33,000 person-days of 24-h human activities in
 6                   CHAD (McCurdy et al.. 2000). The surveys include probability-based recall studies
 7                   conducted by EPA and the California Air Resources Board, as well as real-time diary
 8                   studies conducted in individual U.S. metropolitan areas using both probability-based and
 9                   volunteer subject panels. All ages of both genders are represented in CHAD. The data for
10                   each subject consist of one or more days of sequential activities, in which each activity is
11                   defined by start time, duration, activity type, and microenvironment classification (i.e.,
12                   location). Activities vary from one minute to one hour in duration, with longer activities
13                   being subdivided into clock-hour durations to facilitate exposure modeling. CHAD also
14                   provides information on the level of exertion associated with each activity, which can be
15                   used by exposure models to estimate ventilation rate and pollutant dose.
             4.4.2   Ozone Averting Behavior

16                   Individuals can reduce their exposure to O3 by altering their behaviors, such as reducing
17                   their time outdoors. To protect the public from O3-related health effects, EPA and
18                   organizations such as the American Lung Association recommend that people spend
19                   more time indoors and engage in less strenuous activities on days with relatively high O3
20                   concentrations. To assist individuals concerned about O3 conditions, EPA developed the
21                   Air Quality Index (AQI). This index combines information about O3 (and other pollutant)
22                   concentrations to produce five categories of air-quality, ranging from good to very
23                   unhealthy. Forecasted and actual conditions typically are reported to the public during
24                   high-O3 months through local media outlets, using various versions of this air-quality
25                   categorization scheme. These advisories explicitly state that children in general and
26                   children with asthma in particular are potentially sensitive to air pollution. Parents are
27                   advised to curtail children's outdoor exertion to varying degrees depending on the
28                   predicted pollution levels and whether their children have asthma or other relevant
29                   medical conditions.

30                   Evidence of individual averting behaviors in response to advisories has been found in
31                   several studies, especially for susceptible populations, such as children, older adults, and
32                   asthmatics. Reduced time spent outdoors was reported in an activity diary study in 35
33                   U.S. cities (Mansfield et al.. 2006). which found that asthmatic children who spent at
34                   least some time outdoors reduced their total time spent outdoors by an average of 30 min
35                   on a code red O3 day relative to a code green, yellow, or orange day; however, the


      Draft - Do Not Cite or Quote                       4-19                                September 2011

-------
 1                   authors noted that there was appreciable variation in both the overall amount of time
 2                   spent outdoors and the reduction in outdoor time on high ozone days among asthmatic
 3                   children. Bresnahan et al. (1997) examined survey data collected during 1985-86 from a
 4                   panel of adults in the Los Angeles area, many of whom had compromised respiratory
 5                   function, by an averting behavior model. A regression analysis indicated that individuals
 6                   with smog-related symptoms spent about 12 minutes less time outdoors over a two-day
 7                   period for each 10 ppb increase in O3 concentration above 120 ppb. Considering that the
 8                   average daily maximum O3 concentration at the time was approximately 180 ppb on days
 9                   when the then-current standard (1-h max of 120 ppb) was exceeded, this implies that
10                   those individuals spent about 40 minutes less time outside per day on a typical high O3
11                   day compared to days with O3  concentrations below the standard. However, the behavior
12                   was not specifically linked to exceedances or air quality alerts.

13                   The fraction of individuals who reduce time spent outdoors, or restrict their children's
14                   outdoor activity, has been found to vary based on health status. In the Bresnahan et al.
15                   study (1997). 40 percent of respondents reported staying indoors on days when air quality
16                   was poor. Individuals who reported experiencing smog-related symptoms were more
17                   likely to take the averting actions, although the presence of asthma or other chronic
18                   respiratory conditions did not significantly affect behavior. A study of parents of
19                   asthmatic children (McDermott et al.. 2006) suggests that parents are aware of the hazard
20                   of outdoor air pollution and the official  alerts designed to protect them and their children.
21                   It also suggests that a majority of parents (55%) comply with recommendations of the
22                   alerts to restrict children's outdoor activity, with more parents of asthmatics reporting
23                   awareness and responsiveness to alerts.  However, only 7% of all parents complied with
24                   more than one-third of the advisories issued (McDermott et al.. 2006). Wen et al. (2009)
25                   analyzed data from the 2005 Behavioral Risk Factor Surveillance System (BRFSS) and
26                   indicated that people with lifetime asthma are about twice as likely as people without
27                   asthma to reduce their outdoor activities based on either media alerts of poor air quality
28                   (31% vs. 16%) or individual perception of air quality (26% vs. 12%). Respondents who
29                   had received advice from a health professional to reduce outdoor activity when air quality
30                   is poor were more likely to report a reduction based on media alerts, both for those with
31                   and without asthma. In a study of randomly selected individuals  in Houston, TX and
32                   Portland, OR, Semenza et al. (2008) found that a relatively small fraction of survey
33                   respondents (9.7% in Houston, 10.5% in Portland) changed their behaviors during poor
34                   air quality episodes. This fraction is appreciably  lower than the fraction reported for
35                   people with asthma in the Wen et al. (2009) study, although it is similar to the fraction
36                   reported in that study for those without asthma. Most of the people in the Semenza et al.
37                   (2008) study reported that their behavioral changes were motivated by self-perception of
3 8                   poor air quality rather than an air quality advisory. It should be noted that the McDermott
      Draft - Do Not Cite or Quote                       4-20                                 September 2011

-------
 1                   et al. (2006). Wen et al. (2009). and Semenza et al. (2008) studies evaluated air quality in
 2                   general and therefore are not necessarily specific to O3.

 3                   Commuting behavior does not seem to change based on air quality alerts. A study in the
 4                   Atlanta area showed that advisories can raise awareness among commuters but do not
 5                   necessarily result in a change in an individual's travel behavior (Henry and Gordon.
 6                   2003). This finding is consistent with a survey for 1000 commuters in Denver, Colorado,
 7                   which showed that the  majority (76%) of commuters heard and understood the air quality
 8                   advisories, but did not  alter their commuting behavior (Blanken et al.. 2001).

 9                   Some evidence is available for other behavioral changes in response to air quality alerts.
10                   Approximately 40 percent of the respondents in the Los Angeles  study by Bresnahan et
11                   al. (1997) limited or rearranged leisure activities, and 20 percent increased use of air
12                   conditioners. As with changes in time spent outdoors, individuals who reported
13                   experiencing smog-related symptoms, but not those with asthma or chronic respiratory
14                   conditions, were more  likely to take the averting actions. Other factors influencing
15                   behavioral changes, such as increased likelihood of averting behavior among high school
16                   graduates, are also reported in the study. In a separate Southern California study,
17                   attendance at two outdoor facilities (i.e. a zoo and an observatory) was reduced by 6-13%
18                   on days when smog alerts were announced, with greater decreases observed among
19                   children and older adults (Neidell. 2010. 2009).

20                   The studies discussed in this section indicate that averting behavior is dependent on
21                   several factors, including health status and lifestage. People with  asthma and those
22                   experiencing smog-related symptoms reduce their time spent outdoors and are more
23                   likely to change their behavior than those without respiratory conditions. Children and
24                   older adults appear more likely to change their behavior than the general population.
25                   Commuters, even when aware of air quality advisories, tend not to change their
26                   commuting behavior.
             4.4.3   Population Proximity to Fixed-Site Ozone Monitors

27                   The distribution of O3 monitors across urban areas varies between cities (Section
28                   3.6.2.1), and the population living near each monitor varies as well. Monitoring sites in
29                   rural areas are generally located in national or state parks and forests, and these monitors
30                   may be relevant for exposures of exercising visitors as well as those who live in similar
31                   locations. Rural monitors tend to be less affected than urban monitors by strong and
32                   highly variable anthropogenic sources of species participating in the formation and
33                   destruction of O3 (e.g., onroad mobile sources) and more highly influenced by regional
34                   transport of O3 or O3 precursors (Section 3.6.2.2). This may contribute to less diel


      Draft - Do Not Cite or Quote                       4-21                                 September 2011

-------
 1                   variability in O3 concentration than is observed in urban areas. It is not necessarily true
 2                   that proximity to a monitor determines the degree to which that monitor represents an
 3                   individual's ambient exposure, but proximity is one indicator. One way to calculate
 4                   monitor representativeness is to calculate the fraction of the urban population living
 5                   within a certain radius of a monitor. Table 4-4 presents the fraction of the population in
 6                   selected cities living within 1, 5, 10, and 20 km of an O3 monitor. Values are presented
 7                   for both total population and for those under 18 years of age, a potentially susceptible
 8                   population to the effects of O3. The data indicate that relatively few people live within
 9                   1 km of an O3 monitor, while nearly all  of the population in most cities lives within 20
10                   km of a monitor. Many O3 monitors are sited at "neighborhood scale," intended to
11                   represent an area of the city with dimensions in the 0.5-4 km range (Section 3.5.6.1).
12                   Looking at the results for a 5-km radius, generally 20-30% of the population lives within
13                   this distance from an O3 monitor. Some cities have a greater population in this buffer,
14                   such as Salt Lake City, while others have a lower percentage, such as Minneapolis and
15                   Seattle. Percentages for children are generally similar to the total population, with no
16                   clear trend.

17                   Another approach is to divide the metropolitan area into sectors surrounding each
18                   monitor such that every person in the sector lives closer to that monitor than any other.
19                   This facilitates calculation of the fraction of the city's population represented (according
20                   to proximity) by each monitor. In Atlanta, for example, the population fraction
21                   represented by each of the 11 monitors in the city ranged from  2.9-22%. The two
22                   monitors closest to the city center (sites  A and B on Figure 3-24) accounted for 16% and
23                   8% of the population, respectively. Site  B has two listed monitoring objectives, highest
24                   concentration and population exposure.  The  other monitor in Atlanta with a listed
25                   objective of highest concentration is Site C, which represents the largest fraction of the
26                   population (22%). The eight monitors with a primary monitoring objective of population
27                   exposure  account for 2.9-17% of the population per monitor.
      Draft - Do Not Cite or Quote                       4-22                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
Table 4-4
Fraction of the 2009 population living within a specified distance of
an ozone monitor in selected U.S. cities
Population
City
Atlanta CSA
Baltimore CSA
Birmingham CSA
Boston CSA
Chicago CSA
Dallas CSA
Denver CSA
Detroit CSA
Houston CSA
Los Angeles CSA
Minneapolis CSA
New York CSA
Philadelphia CSA
Phoenix CBSA
Pittsburgh CSA
Salt Lake City CSA
San Antonio CBSA
San Francisco CSA
Seattle CSA
St. Louis CSA
Total
5,901,670
8,421,016
1 ,204,399
7,540,533
9,980,113
6,791,942
3,103,801
5,445,448
5,993,633
18,419,720
3,652,490
22,223,406
6,442,836
4,393,462
2,471,403
1,717,045
2,061,147
7,497,443
4,181,278
2,914,754
<18yr
1,210,932
1,916,106
281 ,983
1,748,918
2,502,454
1 ,530,877
675,380
1,411,875
1 ,387,851
4,668,441
872,497
5,284,875
1 ,568,878
873,084
563,309
460,747
484,473
1,675,711
918,309
720,746
Within 1 km
Total
0.3%
1.3%
1.4%
0.9%
1.5%
0.4%
1.7%
0.8%
1.5%
1.6%
0.3%
1.5%
0.9%
2.0%
1.5%
3.0%
0.5%
2.6%
0.3%
1.3%
<18yr
0.3%
1.1%
1.6%
0.9%
1.5%
0.4%
1.6%
0.9%
1.8%
1.7%
0.3%
1.7%
1.0%
2.4%
1.4%
3.0%
0.5%
2.9%
0.3%
1.5%
Within 5 km
Total
8%
25%
22%
17%
28%
13%
35%
15%
26%
28%
5%
23%
22%
35%
22%
41%
12%
41%
5%
17%
<18yr
9%
24%
24%
16%
29%
13%
36%
17%
28%
29%
4%
23%
24%
41%
21%
38%
12%
40%
5%
18%
Within 10 km
Total
28%
57%
56%
49%
63%
45%
66%
42%
54%
77%
16%
51%
55%
74%
52%
79%
42%
81%
18%
52%
<18yr
29%
55%
59%
47%
65%
44%
68%
44%
57%
79%
16%
50%
56%
79%
50%
79%
43%
81%
16%
53%
Within 20 km
Total
75%
89%
73%
85%
89%
87%
92%
77%
83%
98%
57%
91%
89%
96%
88%
95%
78%
98%
43%
80%
<18yr
77%
89%
74%
85%
91%
87%
93%
78%
84%
98%
56%
91%
89%
97%
88%
95%
80%
98%
39%
82%
               Atlanta population fractions for children (< 1 8 years of age) are similar to those for the
               general population, but other populations show a different pattern of monitor
               representativeness. Older adults (age 65 and up) were somewhat differently distributed
               with respect to the monitors, with most monitors showing a difference of more than a
               percentage point compared to the general population. Based on 2000 population data, the
               fraction of older adults closest to the two city center monitors (A and B) was 4% higher
               and 2% lower, respectively, than the fraction for the population as a whole. Site C
               showed the highest differential, with 21% of the total population but only 15% of the
               older adult population. This indicates the potential for monitors to differentially represent
               potentially susceptible populations.
Draft - Do Not Cite or Quote
                                                     4-23
September 2011

-------
           4.5    Exposure Modeling
 i
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
In the absence of personal exposure measurements, modeling techniques are used to

estimate exposures, particularly for large populations for which individual-level

measurements would be impractical. Model estimates may be used as inputs to

epidemiologic studies or as stand-alone assessments of the level of exposure likely to be

experienced by a population under certain air quality conditions. This section describes

approaches used to improve exposure estimates, including concentration surface

modeling, which calculates local outdoor concentrations over a geographic area; air

exchange rate modeling, which estimates building ventilation based on housing

characteristics and meteorological parameters; and microenvironment-based exposure

modeling, which combines air quality data with demographic information and activity

pattern simulations to estimate time-weighted exposures based on concentrations in

multiple microenvironments. These models each have strengths and limitations, as

summarized in Table 4-5. The remainder of this section provides  more  detail on specific

modeling approaches, as well as results of applying the models.
       Table 4-5       Characteristics of exposure modeling approaches
          Model Type
      Model
Description
Strengths
Limitations
       Concentration Surface Spatial Interpolation
                         (e.g., Inverse
                         Distance Weighting,
                         Kriging)
                   Measured concentrations are
                   interpolated across an area to
                   yield local outdoor concentration
                   estimates
                    High concentration resolution;     Spatial heterogeneity not
                    uses available data; requires low  fully captured; a single
                    to moderate resources for        high-concentration monitor
                    implementation                can skew results; no
                   	location-activity information
                         Chemistry-transport
                         (e.g., CMAQ)
                   Grid-based 03 concentrations are
                   calculated from precursor emis-
                   sions, meteorology, and atmos-
                   pheric chemistry and physics
                    First-principles characterization of  Grid cell resolution;
                    physical and chemical processes  resource-intensive; no
                    influencing 03 formation         location-activity information
                         Land-use regression
                         (LUR)
                   Merges concentration data with
                   local-scale variables such as land
                   use factors to yield local
                   concentration surface
                    High concentration resolution
                   Reactivity and small-scale
                   spatial variability of 03;
                   location-specific, limiting
                   generalizability; no location-
                   activity information	
       Air Exchange Rate
 Mechanistic         Uses database on building leak-
 (LBL, LBLX)         age tests to predict AER based on
                   building characteristics and
                   meteorological variables (including
 	natural ventilation in LBLX)	
                    Physical characterization of
                    driving forces for air exchange
                   Moderate resource
                   requirement; no location-
                   activity information
                         Empirical
                   Predicts AER based on factors
                   such as building age and floor
                    Low input data requirements
                   Cannot account for
                   meteorology; no location-
                   activity information	
       Integrated           Population
       Microenvironmental   (APEX, SHEDS)
       Exposure and Dose
                   Stochastic treatment of air quality
                   data, demographic variables, and
                   activity pattern to generate
                   estimates of microenvironmental
                   concentrations, exposures, and
                   doses
                    Probabilistic estimates of
                    exposure and dose distributions
                    for specific populations;
                    consideration of nonambient
                    sources; small to moderate
                    uncertainty for exercising
                    asthmatic children (APEX)
                   Resource-intensive;
                   evaluation with measured
                   exposures; underestimation
                   of multiple high-exposure
                   events in an individual
                   (APEX)
       Draft - Do Not Cite or Quote
                                    4-24
                                                    September 2011

-------
             4.5.1   Concentration Surface Modeling

 1                   One approach to improve exposure estimates in urban areas involves construction of a
 2                   concentration surface over a geographic area, with the concentration at locations between
 3                   monitors estimated using a model to compensate for missing data. The calculated O3
 4                   concentration surface can then be used to estimate exposures outside residences, schools,
 5                   workplaces, roadways, or other locations of interest. This technique does not estimate
 6                   exposure directly because it does not account for activity patterns or concentrations in
 7                   different microenvironments. There are three main types of approaches: spatial
 8                   interpolation of measured concentrations; statistical models using meteorological
 9                   variables, pollutant concentrations, and other predictors to estimate concentrations at
10                   receptors in the domain; and rigorous first-principle models, such as chemistry-transport
11                   models or dispersion models incorporating O3 chemistry. Some researchers have
12                   developed models that combine these techniques. The models may be applied over urban,
13                   regional, or national spatial scales, and can be used to estimate daily concentrations or
14                   longer-term averages. This discussion will focus on short-term concentrations estimated
15                   across urban areas.

16                   The 2006 O3 AQCD discussed concentration surface models, focusing on chemistry-
17                   transport models as well as geospatial and spatiotemporal interpolation techniques (e.g.,
18                   Christakos and Vyas. 1998a. b; Georgopoulos et al.. 1997). Recent research has
19                   continued to refine and extend the modeling approaches. A few recent papers have
20                   compared different approaches for the same urban area.

21                   Marshall et al. (2008) compared four spatial interpolation techniques for estimation of O3
22                   concentrations in Vancouver, BC. The investigators assigned a daily average O3
23                   concentration to each of the 51,560 postal-code centroids using one of the following
24                   techniques: (1) the concentration from the nearest monitor within 10 km; (2) the average
25                   of all monitors within 10 km; (3) the inverse-distance-weighted (IDW) average  of all
26                   monitors in the area; and (4) the IDW average of the 3 closest monitors within 50 km.
27                   Method 1 (the nearest-monitor approach) and Method 4 (IDW-50 km) had similar mean
28                   and median estimated annual- and monthly-average concentrations, although the 10th-
29                   90th percentile range was smaller for IDW-50. This is consistent with the averaging of
30                   extreme values inherent in IDW methods. The Pearson correlation coefficient between
31                   the two methods was 0.93 for monthly-average concentrations and 0.78 for annual-
32                   average concentrations. Methods 2 and  3 were considered sub-optimal and were excluded
33                   from further analysis. In the case of Method 2, a single downtown high-concentration
34                   monitor skewed the results in the vicinity, partially as a result of the asymmetric layout of
35                   the coastal city of Vancouver. Method 3 was too spatially homogenous because it
36                   assigned most locations a concentration near the regional average, except for locations
      Draft - Do Not Cite or Quote                       4-25                                 September 2011

-------
 1                   immediately adjacent to a monitoring site. CMAQ concentration estimates using a 4
 2                   km*4 km grid were also compared to the interpolation techniques in this study. Mean and
 3                   median concentrations from CMAQ were approximately 50% higher than Method 1 and
 4                   Method 4 estimates for both annual and monthly average concentrations. This may be
 5                   due in part to the CMAQ grid size, which was too coarse to reveal near-roadway
 6                   decrements in O3 concentration due to titration by NO. The IQR for the annual average
 7                   was similar between CMAQ and the  interpolation techniques, but the monthly average
 8                   CMAQ IQR was approximately twice as large, indicating a seasonal effect.

 9                   Bell (2006) compared CMAQ estimates for northern Georgia with nearest-monitor and
10                   spatial interpolation techniques, including IDW and kriging. The area-weighted
11                   concentration estimates from CMAQ indicated areas of spatial heterogeneity that were
12                   not captured by approaches based on the monitoring network. The author concluded that
13                   some techniques, such as spatial interpolation, were not suitable for estimation of
14                   exposure in certain situations, such as for rural areas. Using the concentration from the
15                   nearest monitor resulted in an overestimation of exposure relative to model estimates.

16                   Land use regression (LUR) models have been developed to estimate levels of air
17                   pollutants, predominantly NO2, as a function of several land use  factors, such as land use
18                   designation, traffic counts, home heating usage, point source strength, and  population
19                   density (Ryan and LeMasters. 2007:  Gillilandetal.. 2005: Briggsetal.. 1997). LUR,
20                   initially termed regression mapping (Briggs etal.. 1997). is a regression derived from
21                   monitored concentrations  as a function of data from a combination of the land use
22                   factors. The regression is then used for predicting concentrations at multiple locations
23                   based on the independent variables at those particular locations without monitors. Hoek
24                   et al. (2008) warn of several limitations of LUR, including distinguishing real
25                   associations between pollutants and covariates from those of correlated co-pollutants,
26                   limitations in spatial resolution from  monitor data, applicability of the LUR model under
27                   changing temporal  conditions, and introduction of confounding factors when LUR is used
28                   in epidemiologic studies. These limitations may partially explain the lack of LUR models
29                   that have been developed for O3 at the urban scale.  Brauer et al. (2008) evaluated the use
30                   of LUR and IDW-based spatial-interpolation models in epidemiologic analyses for
31                   several different pollutants in Vancouver, BC and suggested that LUR is appropriate for
32                   directly-emitted pollutants with high  spatial variability, such as NO and BC, while IDW
33                   is appropriate for secondary pollutants such as NO2 and PM2 5 with less spatial
34                   variability. Although O3 is also a secondary pollutant, its reactivity and high small-scale
3 5                   spatial variability near high-traffic roadways indicates this conclusion may not apply for
36                   03.
      Draft - Do Not Cite or Quote                       4-26                                September 2011

-------
 1                   At a much larger spatial scale, EU-wide, Beelen et al. (2009) compared a LUR model for
 2                   O3 with ordinary kriging and universal kriging, which incorporated meteorological,
 3                   topographical, and land use variables to characterize the underlying trend. The LUR
 4                   model performed reasonably well at rural locations (5-km resolution), explaining a higher
 5                   percentage of the variability (R2 = 0.62) than for other pollutants. However, at the urban
 6                   scale (1-km resolution), only one variable was selected into the O3 LUR model (high-
 7                   density residential land use), and the R2 value was very low (0.06). Universal kriging was
 8                   the best method for the large-scale composite EU concentration map, for O3  as well as
 9                   for NO2 and PM10, with an R2 value for O3 of 0.70. The authors noted that these methods
10                   were not designed to capture spatial variation in concentrations that are known to occur
11                   within tens of meters of roadways (Section 3.6.2.1), which could partially explain poor
12                   model performance at the urban scale.

13                   Titration of O3 with NO emitted by motor vehicles tends to reduce O3 concentrations
14                   near roadways. McConnell et al. (2006) developed a regression model to predict
15                   residential O3 concentrations in southern California using estimates of residential NOX
16                   calculated from traffic data with the CALINE4 line source dispersion model. The authors
17                   estimated that local traffic contributes 18% of NOX concentrations measured in the study
18                   communities, with the remainder coming from regional background. Their regression
19                   model indicates that residential NOX reduces residential O3 concentrations by 0.51 ppb
20                   O3 per 1 ppb NOX, and that a 10th-90th percentile increase in local NOX results in a
21                   7.5 ppb  decrease in local O3 concentrations. This intra-urban traffic-related variability in
22                   O3 concentrations suggests that traffic patterns are an important factor in the relationship
23                   between central site monitor and residential O3, and that differences in traffic density
24                   between the  central site monitor and individual homes could result in either an
25                   overestimate or underestimate of residential O3.

26                   A substantial number of researchers have used geostatistical methods and chemistry-
27                   transport models to estimate O3 concentrations at urban, regional, national, and
28                   continental scales, both in the U.S. and in other countries (Section 3.3). In addition to
29                   short-term exposure assessment for epidemiologic studies, such models may also be used
30                   for long-term exposure assessment, O3 forecasts, or evaluating  emission control
31                   strategies. It is difficult to determine the utility of these methods for exposure assessment;
32                   while improved local-scale estimates of outdoor concentrations may contribute to better
33                   assignment of exposures,  information on activity patterns is needed to produce estimates
34                   of personal exposure.
      Draft - Do Not Cite or Quote                       4-27                                 September 2011

-------
            4.5.2   Residential Air Exchange Rate Modeling

 1                  The residential air exchange rate (AER), which is the airflow into and out of a home, is
 2                  an important mechanism for entry of ambient O3. As described in Section 4.3.2, the
 3                  indoor-outdoor relationship is greatly affected by the AER. Since studies show that
 4                  people spend approximately 66% of their time indoors at home (Leech et al.. 2002;
 5                  Klepeis et al.. 2001). the residential AER is a critical parameter for exposure models,
 6                  such as APEX, SHEDS, and EMI (discussed in Section 4.5.3) (U.S. EPA. 201 Ib. 2009b:
 7                  Burke et al.. 2001). Since the appropriate AER measurements may not be available for
 8                  exposure models, mechanistic and empirical (i.e., regression-based) AER models can be
 9                  used for exposure assessments. The input data for the AER models can include building
10                  characteristics (e.g., age, number of stories, wind sheltering), occupant behavior (e.g.,
11                  window opening), climatic region, and meteorology (e.g., local temperature and wind
12                  speed). Mechanistic AER models use these meteorological parameters to account for the
13                  physical driving forces of the airflows due to pressure differences across the building
14                  envelope from wind and indoor-outdoor temperature differences (ASHRAE. 2009).
15                  Empirical AER models do not consider the driving forces from the wind and indoor-
16                  outdoor temperature differences. Instead, a scaling constant can be used based on factors
17                  such as building age and floor area (Chan et al.. 2005b).

18                  Single-zone mechanistic models represent a whole-building as a single, well-mixed
19                  compartment. These AER models, such as the Lawrence Berkeley Laboratory (LBL)
20                  model, can predict residential AER using input data from whole-building pressurization
21                  tests (Sherman and Grimsrud. 1980). or leakage  area models (Breen et al.. 2010; Sherman
22                  and Me Williams. 2007). Recently, the LBL air infiltration model was linked with a
23                  leakage area model using population-level census and residential survey data (Sherman
24                  and Me Williams. 2007) and individual-level questionnaire data (Breen et al.. 2010). The
25                  LBL model, which predicts the AER from air infiltration (i.e., small uncontrollable
26                  openings in the building envelope) was also extended to include airflow from natural
27                  ventilation (LBLX), and evaluated using window opening  data (Breen et al.. 2010). The
28                  AER predictions from the  LBL and LBLX models were compared to daily AER
29                  measurements on seven consecutive days during each season from detached homes in
30                  central North Carolina (Breen etal.. 2010). For the individual model-predicted and
31                  measured AER, the median absolute difference was 43% (0.17 h"1) and 40% (0.17 h"1) for
32                  the LBL and LBLX models, respectively. Given the uncertainty of the AER
33                  measurements (accuracy of 20-25% for occupied homes), these results demonstrate the
34                  feasibility of using these AER models for both air infiltration (e.g., uncontrollable
35                  openings) and natural ventilation (e.g., window opening) to help reduce the AER
36                  uncertainty in exposure models. The capability of AER models could help support the
37                  exposure modeling needs,  as described in Section 4.5.3, which includes the ability to

      Draft - Do Not Cite or Quote                      4-28                               September 2011

-------
 1                  predict indoor concentrations of ambient O3 that may be substantial for conditions of
 2                  high AER such as open windows.
            4.5.3  Microenvironment-Based Models

 3                  Population-based methods, such as the Air Pollution Exposure (APEX) and Stochastic
 4                  Human Exposure and Dose Simulation (SHEDS) integrated microenvironmental
 5                  exposure and dose models, involve stochastic treatment of the model inputs (U.S. EPA.
 6                  2009b: Burke etal.. 2001). These are described in detail in the 2008 NOX ISA (U.S.
 7                  EPA. 2008b). in AX3.6.1. Stochastic models utilize distributions of pollutant-related and
 8                  individual-level variables, such as ambient and local O3 concentration contributions and
 9                  breathing rate respectively, to compute the distribution of individual exposures across the
10                  modeled population. The models also have the capability to estimate received dose
11                  through a dosimetry model. Using distributions of input parameters in the model
12                  framework rather than point estimates  allows the models to incorporate uncertainty and
13                  variability explicitly into exposure estimates (Zidek et al.. 2007). These models estimate
14                  time-weighted exposure for modeled individuals by summing exposure in each
15                  microenvironment visited during the exposure period.

16                  The initial set of input data for population exposure models is ambient air quality data,
17                  which may come from a monitoring network or model estimates. Estimates of
18                  concentrations in a set of microenvironments are generated either by mass balance
19                  methods, which can incorporate AER models (Section 4.5.3), or microenvironmental
20                  factors. Microenvironments modeled include indoor residences; other indoor locations,
21                  such as schools, offices, and public buildings; and vehicles. The sequence of
22                  microenvironments and exertion levels during the exposure period is determined from
23                  characteristics of each modeled individual. The APEX model does this by generating a
24                  profile for each simulated individual by sampling from distributions of demographic
25                  variables such as age, gender, and employment; physiological variables such as height
26                  and weight; and situational variables such as living in a house with a gas stove or air
27                  conditioning. Activity and location (microenvironmental) patterns from a database such
28                  as CHAD are assigned to the simulated individual in a longitudinal manner, using age,
29                  gender, andbiometric characteristics (U.S. EPA. 2009a: Glen etal.. 2008).  Breathing
30                  rates for each individual are calculated for each activity based on predicted energy
31                  expenditures, and the corresponding received intake or blood dose may then be
32                  computed. APEX has an algorithm to estimate O3 dose  and changes in FEVi resulting
33                  from O3 exposure. Summaries of individual- and population-level metrics are produced,
34                  such as maximum exposure or dose, number of individuals exceeding a specified
35                  exposure/dose, and number of person-days at or above benchmark exposure levels. The


      Draft - Do Not Cite or Quote                      4-29                               September 2011

-------
 1                   models also consider the nonambient contribution to total exposure. Nonambient source
 2                   terms are added to the infiltration of ambient pollutants to calculate the total
 3                   concentration in the microenvironment. Output from model runs with and without
 4                   nonambient sources can be compared to estimate the ambient contribution to total
 5                   exposure and dose.

 6                   Georgopoulos et al. (2005) used a version of the SHEDS model as the exposure
 7                   component of a modeling framework known as MENTOR (Modeling Environment for
 8                   Total Risk Studies) in a simulation of O3 exposure in Philadelphia over a 2-week period
 9                   in July  1999. 500 individuals were sampled from CHAD in each of 482 census tracts to
10                   match local  demographic characteristics from U.S. Census data. Outdoor concentrations
11                   over the modeling domain were calculated from interpolation of photochemical modeling
12                   results and fixed-site monitor concentrations. These concentrations were then used as
13                   input data for SHEDS. Median microenvironmental concentrations predicted by SHEDS
14                   for nine simulated microenvironments were strongly correlated with outdoor
15                   concentrations, a result consistent with the lack of indoor O3 sources in the model. A
16                   regression of median microenvironmental concentrations against outdoor concentrations
17                   indicated that the  microenvironmental concentrations were appreciably lower than
18                   outdoor concentrations (regression slope = 0.26). 95th percentile microenvironmental
19                   concentrations were also well correlated with outdoor concentrations and showed a
20                   regression slope of 1.02, although some microenvironmental concentrations were well
21                   below the outdoor values. This suggests that in most cases the high-end concentrations
22                   were associated with outdoor microenvironments. Although the authors did not report
23                   exposure statistics for the population, their dose calculations also indicated that O3 dose
24                   due to time spent  outdoors dominated the upper percentile s of the population dose
25                   distribution. They found that both the 50th and 95th percentile O3 concentrations were
26                   correlated with census-tract level outdoor concentrations estimated by photochemical
27                   modeling combined with spatiotemporal interpolation, and attributed this correlation to
28                   the lack of indoor sources of O3. Relationships between exposure and concentrations at
29                   fixed-site monitors were not reported.

30                   As part of the previous NAAQS review completed in 2008, EPA's Office of Air Quality
31                   Planning and Standards used APEX-O3 to estimate O3 exposures in 12 cities during the
32                   O3 monitoring seasons of 2002-04 and reported the results in the 2007 O3 Staff Paper
33                   (U.S. EPA. 2007b). Exposures  were modeled for the general population, school-age
34                   children (ages 5-18), and asthmatic school-age children. Hourly air quality input data
3 5                   from monitors in each city were adjusted to simulate just meeting various alternative
36                   standards, ranging from 65 to 85 ppb (8-h average), to demonstrate the effect of different
37                   standards on O3 exposure metrics. O3 decay (i.e., reaction) in indoor microenvironments
38                   was modeled, but no indoor O3 sources were included.
      Draft - Do Not Cite or Quote                       4-30                                September 2011

-------
 1                   Results of the model runs indicated that children and asthmatic children had similar
 2                   exposures, with the general population experiencing lower exposure. For example, in
 3                   Boston using 2002 air quality data adjusted to meet the then-current 8-h standard of 0.08
 4                   ppm (fourth-highest maximum averaged over 3 yr), approximately 28% of children or
 5                   asthmatic children were estimated to experience one or more 8-h avg exposures of 70 ppb
 6                   or greater during an 8-h period in which they engaged in moderate exercise. In
 7                   comparison, about 10% of the general population (including children) would experience a
 8                   70 ppb-8h or greater exposure under the same conditions (Exhibit 2 and Figure 4-7 of the
 9                   Staff Paper). A similar pattern was observed in other cities, although the magnitude of
10                   exposure was different. In most cases, exposures were substantially higher in 2002 than
11                   2004, with 2003 exposures in between the estimates for the other two years (Figure 4-8 of
12                   the Staff Paper).

13                   Exposures were quite variable across cities due primarily to differing air quality
14                   distributions that resulted in a differential result from the air quality adjustment
15                   procedure. For example, the same 74 ppb-8h (fourth maximum) alternative standard
16                   scenario for 2002 estimated that 10% of Boston children but very few (<0.5%) of Los
17                   Angeles children experience exposures above 70 ppb-8h while engaged in moderate
18                   exertion. The relationship between the fourth-highest concentrations (the basis for the air
19                   quality adjustment) and the remainder of the air quality distribution is quite different
20                   between the two cities, with the result that more of the upper range of the air quality data
21                   was rolled back in Los Angeles than in Boston. This substantially reduced the occurrence
22                   of modeled high-end exposures.

23                   Simulations indicate that meeting O3 air quality standards would reduce the fraction of
24                   individuals experiencing high-end exposures, as expected. Using unadjusted 2004 air
25                   quality data (the lowest of the three years simulated), the estimate of the fraction of
26                   children experiencing a 60 ppb-8h exposure while engaging in moderate exertion ranged
27                   from 12% (Chicago) to 69% (Los Angeles). Adjusting air quality data to meet fourth-
28                   maximum alternative standards of 85, 75, and 65 ppb reduced that range to 1-26%, 0-
29                   11 %, and 0-1 %, respectively (Exhibit 9 of the Staff Paper).

30                   An analysis has been conducted for the APEX model to evaluate the contribution of
31                   uncertainty in input parameters  and databases to the uncertainty in model outputs
32                   (Langstaff. 2007). The Monte Carlo analysis indicates that the uncertainty in model
33                   exposure estimates  for asthmatic children during moderate exercise is small to moderate,
34                   with 95% confidence intervals of at  most ± 6 percentage points at exposures above 60,
35                   70, and 80 ppb (8-h avg) However, APEX appears to substantially underestimate the
36                   frequency of multiple high-exposure events for a single individual. The two main sources
37                   of uncertainty identified were related to the activity pattern database and the spatial
      Draft - Do Not Cite or Quote                       4-31                                 September 2011

-------
 1                   interpolation of fixed-site monitor concentrations to other locations. One area of potential
 2                   improvement in the activity pattern database is additional information on children's
 3                   activities, including longitudinal patterns. Improved information on spatial variation of
 4                   O3 concentrations, including in near-roadway and indoor microenvironments, would also
 5                   contribute to reduced uncertainty. Another area of need is for improved personal
 6                   exposure monitors with shorter averaging times to capture peak exposures and lower
 7                   detection limits to capture low indoor concentrations. A similar modeling approach is
 8                   currently being developed which is suitable for panel epidemiologic studies or for
 9                   controlled human exposure studies, in which activity pattern data specific to the
10                   individuals in the study can be collected. Time-activity data is combined with
11                   questionnaire data on housing characteristics, presence of indoor or personal sources, and
12                   other information to develop a personalized set of model input parameters for each
13                   individual. This model, the Exposure Model for Individuals, is under development by
14                   EPA's National Exposure Research Laboratory (U.S. EPA. 201 Ib: Zartarian and Schultz.
15                   2010).
          4.6   Implications for Epidemiologic Studies

16                   Exposure measurement error, which refers to the uncertainty associated with using
17                   exposure metrics to represent the actual exposure of an individual or population, can be
18                   an important contributor to variability in epidemiologic study results. Time-series studies
19                   assess the daily health status of a population of thousands or millions of people over the
20                   course of multiple years (i.e., thousands of days) across an urban area by estimating their
21                   daily exposure using a short monitoring interval (hours to days). In these studies, the
22                   community-averaged concentration of an air pollutant measured at central-site monitors
23                   is typically used as a surrogate for individual or population ambient exposure. In
24                   addition, panel studies, which consist of a relatively small sample (typically tens) of
25                   study participants followed over a period of days to months, have been used to examine
26                   the health effects associated with short-term exposure to ambient concentrations of air
27                   pollutants (Delfino et al.. 1996). Panel studies may also apply a microenvironmental
28                   model to represent exposure to an air pollutant. A longitudinal cohort epidemiologic
29                   study, such as the ACS cohort study, typically involves hundreds or thousands of subjects
30                   followed over several years or decades (Jerrett et al., 2009). Concentrations are generally
31                   aggregated over time  and by community to estimate exposures.

32                   Exposure error can under- or over-estimate epidemiologic associations between ambient
33                   pollutant concentrations and health outcomes by biasing effect estimates toward or away
34                   from the null. Exposure misclassification can also tend to obscure the presence of
35                   thresholds for health effects, as demonstrated by a simulation study of nondifferential

      Draft - Do Not Cite or Quote                       4-32                                 September 2011

-------
 1                   exposure misclassification (Brauer et al.. 2002). The importance of exposure
 2                   misclassification varies with study design and is dependent on the spatial and temporal
 3                   aspects of the design. For example, the use of a community-averaged O3 concentration in
 4                   a time-series epidemiologic study may be adequate to represent the day-to-day temporal
 5                   concentration variability used to evaluate health effects, but may not capture differences
 6                   in the magnitude of exposure due to spatial variability. Other factors that could influence
 7                   exposure estimates include nonambient exposure, topography of the natural and built
 8                   environment, meteorology, measurement errors, use of ambient O3 concentration as a
 9                   surrogate  for ambient O3 exposure, and the presence of O3 in a mixture of pollutants. The
10                   following sections will consider various sources of error and how they affect the
11                   interpretation of results from epidemiologic studies of different  designs.
            4.6.1    Nonambient Ozone Exposure

12                   For other criteria pollutants, nonambient sources can be an important contributor to total
13                   personal exposure. There are relatively few indoor sources of O3; as a result, personal O3
14                   exposure is expected to be dominated by ambient O3 in outdoor microenvironments and
15                   in indoor microenvironments with high air exchange rates (e.g., with open windows).
16                   Even in microenvironments where nonambient exposure is substantial, such as in a room
17                   with an O3 generator, this nonambient exposure is unlikely to be temporally correlated
18                   with ambient O3 exposure (Wilson and Suh. 1997), and therefore would not affect
19                   epidemiologic associations between O3 and a health effect (Sheppard et al.. 2005). In
20                   simulations of a nonreactive pollutant, Sheppard et al. (2005) concluded that nonambient
21                   exposure does not influence the health outcome effect estimate if ambient and
22                   nonambient concentrations are independent. Since personal exposure to ambient O3 is
23                   some fraction of the ambient concentration, it should be noted that effect estimates
24                   calculated based on personal exposure rather than ambient concentration will be
25                   increased in proportion to the ratio of ambient concentration to ambient exposure, and
26                   daily fluctuations in this ratio can widen the confidence intervals in the ambient
27                   concentration effect estimate, but uncorrelated nonambient exposure will not bias the
28                   effect estimate.
            4.6.2   Spatiotemporal Variability

29                   Spatial and temporal variability in O3 concentrations can contribute to exposure error in
30                   epidemiologic studies, whether they rely on central-site monitor data or concentration
31                   modeling for exposure assessment. Spatial variability in the magnitude of concentrations
32                   may affect cross-sectional and large-scale cohort studies by undermining the assumption

      Draft - Do Not Cite or Quote                       4-33                                September 2011

-------
 1                   that intra-urban concentration and exposure differences are less important than inter-
 2                   urban differences. This issue may be less important for time-series studies, which rely on
 3                   day-to-day temporal variability in concentrations to evaluate health effects. Low inter-
 4                   monitor correlations contribute to exposure error in time-series studies, including bias
 5                   toward the null and increased confidence intervals.

 6                   The averaging time of the daily exposure metrics used to evaluate daily aggregated health
 7                   data (e.g., 1-h or 8-h daily maximum vs. 24-h avg concentration) may also impact
 8                   epidemiologic results, since different studies report different daily metrics. Correlations
 9                   between  1-h daily max, 8-h daily max, and 24-h avg concentrations for U.S. monitoring
10                   sites are presented in Section 3.6.1 (Figure 3-18 and accompanying text). The two daily
11                   peak values (1-h max and 8-h max) are well correlated, with a median (IQR) correlation
12                   of 0.97 (0.96-0.98). The correlation between the 8-h max and 24-h avg are somewhat less
13                   well correlated with a median (IQR) correlation of 0.89 (0.86-0.92). While this may
14                   complicate quantitative comparisons between epidemiologic studies using different daily
15                   metrics, as well as the interpretation of studies using metrics other than the current 8-h
16                   standard, the high inter-metric correlations suggest it is a relatively small source of
17                   uncertainty in comparing the results of studies using different metrics. This is supported
18                   by a study comparing each of these metrics in a time-series study of respiratory ED visits
19                   (Darrow et al., 201 Ib). which found positive associations for all metrics, with the
20                   strongest association for the 8-h daily max exposure metric (Section 6.7.3.2).

21                   The ratios of 1-h daily max, 8-h daily max, and 24-h avg concentrations to one another
22                   have been found to differ across communities and across time within  individual
23                   communities (Anderson and Bell. 2010). For example, 8:24 hour ratios ranged from  1.23-
24                   1.83, with a median of 1.53. Lower ratios were generally observed in the spring and
25                   summer compared to fall and winter. O3 concentration was identified as the most
26                   important predictor of ozone metric ratios, with higher overall O3 concentrations
27                   associated with lower ratios.  In communities with higher long-term ozone concentrations,
28                   the low 8:24 hour ratio is attributed to high baseline O3, which results in elevated 24-h
29                   average values. Differences in the representativeness of O3 metrics introduces uncertainty
30                   into epidemiologic results and complicates comparison of studies using different metrics.
31                   Preferably, studies will report results using multiple metrics. In cases where this does not
32                   occur, the results of this study can inform the uncertainty associated with using a standard
33                   increment to adjust effect estimates based on different metrics so that they are
34                   comparable (Chapter 6).

35                   A study compared measures of spatial  and temporal variability for 1-h daily max and 24-
36                   h daily avg O3 concentrations in Brazil (Bravo and Bell. 2011). The 1-h daily max value
37                   was found to have higher correlation between monitors (i.e., lower temporal variability)
      Draft - Do Not Cite or Quote                       4-34                                 September 2011

-------
 1                   and lower COD (a measure of spatiotemporal variability which incorporates differences
 2                   in concentration magnitude, with lower values indicating lower variability; see Chapter 3)
 3                   than the 24-h avg value. The range of correlation coefficients and COD values was
 4                   similar between the two metrics, although the variation was lower for the 1-h daily max,
 5                   as indicated by the R2 value for the regression of correlation coefficient on inter-monitor
 6                   distance.

 7                   Long-term exposure epidemiologic studies use concentrations averaged over months,
 8                   years, or decades to evaluate health effects of extended O3 exposure. A study in Canada
 9                   comparing exposure assessment methods for long-term O3 exposure found that the
10                   annual average concentration in the census tract of a subject's residence during 1980 and
11                   1994 was well-correlated (0.76 and 0.83, respectively) with a concentration metric
12                   accounting for movement among census subdivisions during 1980-2002 (Quay et al.,
13                   2011). This may have been due in part to a relatively low rate of movement, with subjects
14                   residing on average for 71% of the 22-year period in the same census subdivision they
15                   were in during 1980. This suggests that an exposure metric based on a single year can
16                   represent exposure over a multi-decade period.
                     4.6.2.1    Spatial Variability

17                   Spatial variability of O3 concentrations is highly dependent on spatial scale; in effect, O3
18                   is a regional pollutant subject to varying degrees of local variability. In the immediate
19                   vicinity of roadways, O3 concentrations are reduced due to reaction with NO and other
20                   species (Section 4.3.4.2); over spatial scales of a few kilometers, O3 may be more
21                   homogeneous due to its formation as a secondary pollutant; over scales of tens of
22                   kilometers, atmospheric processing can result in higher concentrations downwind of an
23                   urban area than in the urban core. Local-scale variations have a large impact on the
24                   relative magnitude of concentrations among urban monitors, while conditions favoring
25                   high or low rates of O3 formation (e.g., temperature) vary over large spatial scales. This
26                   suggests that neighborhood monitors are likely to track one another temporally, but miss
27                   small-scale spatial variability in magnitude. In rural areas, a lower degree  of fluctuation
28                   in O3 precursors such as NO and VOCs is likely to make the diel concentration profile
29                   less variable than in  urban areas, resulting in more sustained ambient levels. Spatial
30                   variability contributes to exposure error if the ambient O3 concentration measured at the
31                   central site monitor is used as an ambient exposure surrogate and differs from the actual
32                   ambient O3 concentration outside a subject's residence and/or worksite (in the absence of
33                   indoor O3 sources). Averaging data from a large number of samplers will  dampen
34                   intersampler variability, and use of multiple monitors over smaller land areas may allow
35                   for more variability to be incorporated into an epidemiologic analysis.


      Draft - Do Not Cite or Quote                       4-35                                 September 2011

-------
 1                   Community exposure may not be well represented when monitors cover large areas with
 2                   several subcommunities having different sources and topographies, such as the
 3                   Los Angeles CSA (Section 3.6.2.1). Ozone monitors in Los Angeles had a much wider
 4                   range of intermonitor correlations (-0.06 to 0.97) than Atlanta (0.61 to 0.96) or Boston
 5                   (0.56 to 0.97) using 2007-2009 data. Although the negative and near-zero correlations in
 6                   Los Angeles were observed for monitors located some distance apart (>150 km), some
 7                   closer monitor pairs had low positive correlations, likely due to changes in topography
 8                   and airflow patterns over short distances. Lower COD values, which indicate less
 9                   variability among monitors in the magnitude of O3 concentrations, were observed in
10                   Atlanta (0.05-0.13) and Boston (0.05-0.19) than Los Angeles (0.05-0.56), although a
11                   single monitor (AM) was responsible for all Los Angeles COD values above 0.40. The
12                   spatial and temporal variability in O3 concentration in 24 MSAs across the U.S. was also
13                   examined in the 2006 O3 AQCD by using Pearson correlation coefficients, values of the
14                   90th percentile of the absolute difference in O3 concentrations, and CODs. No clear
15                   discernible regional differences across the U.S. were found in the ranges of parameters
16                   analyzed.

17                   An analysis of the impact of exposure error due to spatial variability and instrument
18                   imprecision on time-series epidemiologic study results indicated that O3 has relatively
19                   low exposure error compared to other routinely monitored  pollutants, and that the
20                   simulated impact on effect estimates is minor. Goldman et  al.  (2011) computed
21                   population-weighted scaled semivariances and Pearson correlation coefficients for daily
22                   concentration metrics of twelve pollutants measured at multiple central-site monitors in
23                   Atlanta. 8-h daily max O3 exhibited the lowest semivariance and highest correlation of
24                   any of the pollutants. Although this indicates some degree  of urban-scale homogeneity
25                   for O3, the analysis did not account for near-road effects on O3 concentrations.

26                   Studies evaluating the influence of monitor selection on epidemiologic study results  have
27                   found that O3 effect estimates are similar across different spatial averaging scales and
28                   monitoring sites. A study in Italy compared approaches for using fixed-site monitoring
29                   data in a case-crossover epidemiologic study of daily O3 and mortality (Zauli Sajani et
30                   al., 2011). O3 effect estimates were found to be similar whether the nearest monitor was
31                   used, or whether single-city, three-city, or six-city regional averages were used for
32                   exposure assessment. In contrast, effect estimates for PM10 and NO2 increased with
33                   increasing scale of spatial averaging. Confidence intervals  increased with increasing
34                   spatial scale for all pollutants. The authors attributed the  consistency of O3 effect
35                   estimates to the relative spatial homogeneity of O3 over multi-km spatial scales, and
36                   pointed to the high (0.85-0.95) inter-monitor correlations to support this. The use of
37                   background monitors rather than monitors influenced by local sources in this study
38                   suggests that local-scale spatial variation in O3, such as that due to titration by traffic
      Draft - Do Not Cite or Quote                       4-36                                 September 2011

-------
 1                   emissions, was not captured in the analyses. Sarnat et al. (2010) studied the spatial
 2                   variability of O3, along with PM2 5, NO2, and CO, in the Atlanta, GA, metropolitan area
 3                   and evaluated how spatial variability affects interpretation of epidemiologic results, using
 4                   time-series data for circulatory disease ED visits. The authors found that associations
 5                   with ambient 8-h daily maximum O3 concentration were similar among all sites tested,
 6                   including multiple urban sites and a rural site some 38 miles from the city center. This
 7                   result was also observed for 24-h PM2 5 concentrations. In contrast, hourly CO and NO2
 8                   showed different associations for the rural site than the urban sites, although the urban
 9                   site associations were similar to one another for CO. This suggests that the choice of
                                                                            OO
10                   monitor may have little impact on the results of O3 time-series studies, consistent with
11                   the moderate to high inter-monitor correlations observed in Atlanta (Chapter 3).

12                   One potential explanation for this finding from the study by Sarnat et al. (2010) is that
13                   although spatial variability at different scales contributes to a complicated pattern of
14                   variations in the magnitude of O3 concentrations between near-road, urban core, and
15                   urban downwind sites, day-to-day fluctuations in concentrations may be reflected across
16                   multiple urban microenvironments. In addition, time-averaging of O3 and PM25
17                   concentrations may smooth out some of the intra-day spatial variability observed with the
18                   hourly CO and NO2 concentrations. However, some uncertainty in observed effect
19                   estimates due to spatial variability and associated exposure error is expected to remain,
20                   including a potential bias towards the null.
                     4.6.2.2   Seasonality

21                   The relationship between personal exposure and ambient concentration has been found to
22                   vary by season, with at least three factors potentially contributing to this variation:
23                   differences in building ventilation (e.g., air conditioning or heater use versus open
24                   window ventilation), higher O3 concentrations during the O3 season contributing to
25                   increased exposure and improved detection by personal monitors; and changes in activity
26                   pattern resulting in more time spent outside. Evidence has been presented in studies
27                   conducted in several cities regarding the effect of ventilation on personal-ambient and
28                   indoor-outdoor O3 relationships (see Sections 4.3.2 and 4.3.3). More limited evidence is
29                   available regarding the specific effects of O3 detection limits and activity pattern changes
30                   on O3 relationships.
31                   Several studies have found increased summertime correlations or ratios between personal
32                   exposure and ambient concentration (Sarnat et al.. 2005; Sarnat et al.. 2000) or between
33                   indoor and outdoor O3 concentrations (Gevh et al.. 2000; Avol et al..  1998b). However,
34                   others have found higher ratios in fall than in summer (Sarnat et al.. 2006b) or equivalent,
      Draft - Do Not Cite or Quote                       4-37                                 September 2011

-------
 1                   near-zero ratios in winter and summer (Sarnat et al., 2001). possibly because summertime
 2                   use of air conditioners decreases building air exchange rates. It should be noted that O3
 3                   concentrations during winter are generally much lower than summertime concentrations,
 4                   possibly obscuring wintertime relationships due to detection limit issues. Studies
 5                   specifically evaluating the effect of ventilation conditions on O3 relationships have found
 6                   increased correlations or ratios for individuals or buildings experiencing higher air
 7                   exchange rates (Sarnat et al. 2006b; Gevh et al., 2000; Sarnat et al., 2000; Romieu et al..
 8                   1998b).

 9                   Increased correlations or ratios between personal exposure and ambient concentration, or
10                   between indoor and outdoor concentration, are likely to reduce error in exposure
11                   estimates used in epidemiologic studies. This suggests that studies conducted during the
12                   O3 season or in periods when communities are likely to have high air exchange rates
13                   (e.g., during mild weather) may be  less prone to exposure error than studies conducted
14                   only during winter. Year-round studies that include both the O3 and non-O3 seasons may
15                   have an intermediate level of exposure error.
             4.6.3   Exposure to Co-pollutants and Ozone Reaction Products

16                   Although indoor O3 concentrations are usually well below ambient concentrations, the
17                   same reactions that reduce O3 indoors form particulate and gaseous species, including
18                   other oxidants, as summarized in Section 4.3.4.3. Exposures to these reaction products
19                   would therefore be expected to be correlated with ambient O3 concentrations, although
20                   no evidence was identified regarding personal exposures. Such exposure could
21                   potentially contribute to health effects observed in epidemiologic studies.
             4.6.4   Averting  Behavior

22                   As described in Section 4.4.2, several recent studies indicate that some populations alter
23                   their behavior on high ozone days to avoid exposure. Such behavioral responses to
24                   information about forecasted air quality may introduce systematic measurement error in
25                   air pollution exposure, leading to biased estimates of the impact of air pollution on health.
26                   For example, studies have hypothesized that variation in time spent outdoors may be a
27                   driving factor behind the considerable heterogeneity in ozone mortality impacts across
28                   communities (Bell et al., 2004). If averting behavior in fact results in smaller, in
29                   magnitude,  effect estimates, then studies that do not account for averting behavior may
30                   produce effect estimates that are biased towards the null (Section 6.2.7.5).
      Draft - Do Not Cite or Quote                       4-38                                 September 2011

-------
 1                   This is supported by an epidemiologic study that examined the association between
 2                   exposure to ambient ozone concentrations and asthma hospitalizations in Southern
 3                   California during 1989-1997, which indicates that controlling for avoidance behavior
 4                   increases the effect estimate for both children and older adults, but not for adults aged 20-
 5                   64 (Neidell and Kinney. 2010; NeidelL 2009). Figure 4-4 and Figure 4-5, reproduced
 6                   from Neidell (2009). show covariate-adjusted asthma hospital admissions as a function of
 7                   daily maximum 1-h O3 concentration for all days (gray line) and days when no O3 alert
 8                   was issued (black line). Stage 1 smog alerts were  issued by the State of California for
 9                   days when ambient O3 concentrations were forecast to be above 0.20 ppm;  however, the
10                   concentration-response functions are based on measured O3 concentrations. For children
11                   aged 5-19 (Figure 4-4), hospital admissions were  higher on high-O3 days when no alert
12                   was issued, especially on days with O3 concentrations above 0.15 ppm (150 ppb). The
13                   concentration-response curves for all days and days with no alert diverge at measured O3
14                   concentrations between 0.10 and 0.15 ppm because smog alerts begin to be issued more
15                   frequently in this range. This suggests that in the absence of information that would
16                   enable averting behavior, children experience higher ozone exposure and subsequently a
17                   greater number of asthma hospital admissions than on alert days with similar O3
18                   concentrations. The lower rate of admissions observed when alert days were included in
19                   the analysis suggests that averting behavior reduced O3 exposure and asthma hospital
20                   admissions. In both cases, O3 was found to be associated with asthma hospital
21                   admissions, although the strength of the association is underestimated when not
22                   accounting for averting behavior. A similar result was not observed when examining
23                   associations for adults aged 20-64 (Figure 4-5), who had similar rates of hospital
24                   admissions on non-alert days as on all days. The lack of change for adults aged 20-64,
25                   which is primary employment age, may reflect lower response to air quality alerts due to
26                   the increased opportunity cost of behavior change. The finding that air quality
27                   information reduces the daily asthma hospitalization rate in these populations provides
28                   additional support for a link between ozone and health effects.
      Draft - Do Not Cite or Quote                       4-39                                September 2011

-------
                 •2
                 <
                 i  s
                 -s
                 
-------
            4.6.5   Exposure Estimation Methods in Epidemiologic Studies

 1                   The use of O3 measurements from central ambient monitoring sites is the most common
 2                   method for assigning exposure in epidemiologic studies. However, fixed-site
 3                   measurements do not account for the effects of spatial variation in O3 concentration,
 4                   ambient and non-ambient concentration differences, and varying activity patterns on
 5                   personal exposures (Brown et al.. 2009; Chang et al.. 2000; Zeger et al.. 2000). The use
 6                   of fixed-site concentrations results in minimal exposure error when:  (1) O3
 7                   concentrations are uniform across the region; (2) personal activity patterns are similar
 8                   across the population; and (3) housing characteristics, such as air exchange rate and
 9                   indoor reaction rate, are constant over the study area. Since these factors vary by location
10                   and population, there will be errors in the magnitude of total exposure based solely on
11                   ambient monitoring data.

12                   Modeling approaches can also be used to estimate exposures for epidemiologic studies,
13                   as discussed in Section 4.5. Geostatistical spatial interpolation techniques can provide
14                   finer-scale estimates of local concentration over urban areas. A microenvironmental
15                   modeling  approach simulates exposure using empirical distributions of concentrations  in
16                   specific microenvironments together with human activity pattern data.  The main
17                   advantage of the modeling approach is that it can be used to estimate exposures over a
18                   wide range of population and scenarios. A main disadvantage of the  modeling approach
19                   is that the results of modeling exposure assessment must be compared to an independent
20                   set of measured exposure levels  (Klepeis. 1999). In addition, resource-intensive
21                   development of validated and representative model inputs is required, such as human
22                   activity patterns, distributions of air exchange rate, and deposition rate. Therefore,
23                   modeled exposures are used much less frequently in epidemiologic studies.
          4.7    Summary and Conclusions

24                   This section will briefly summarize and synthesize the main points of the chapter, with
25                   particular attention to the relevance of the material for the interpretation of epidemiologic
26                   studies.
27                   Passive badge samplers are the most widely used technique for measuring personal O3
28                   exposure (Section 4.3.1). The detection limit of the badges for a 24-h sampling period is
29                   approximately 5-10 ppb, with lower detection limits at longer sampling durations. In low-
30                   concentration conditions this may result in an appreciable fraction of 24-h samples being
31                   below the detection limit. The use of more sensitive portable active monitors, including
      Draft - Do Not Cite or Quote                      4-41                                September 2011

-------
 1                   some that have recently become available, may help overcome this issue and improve
 2                   personal monitoring in the future.

 3                   Since there are relatively few indoor sources of O3, indoor O3 concentrations are often
 4                   substantially lower than outdoor concentrations due to reactions of O3 with indoor
 5                   surfaces and airborne constituents (Section 4.3.2). Air exchange rate is a key determinant
 6                   of the I/O ratio, which is generally in the range of 0.1-0.4 (Table 4-1), with some
 7                   evidence for higher ratios during the O3 season when concentrations are higher.

 8                   Personal exposure is moderately correlated with ambient O3 concentration, as indicated
 9                   by studies reporting correlations generally in the range of 0.3-0.8 (Table 4-2). Hourly
10                   concentration correlations are more variable than those averaged over 24 hours or longer,
11                   with correlations in outdoor microenvironments (0.7-0.9) much higher than those in
12                   residential indoor (0.1) or other indoor (0.3-0.4) microenvironments.  Some studies report
13                   substantially lower personal-ambient correlations, a result attributable in part to low air
14                   exchange rate and O3 concentrations below the sampler detection limit, conditions often
15                   encountered during wintertime. Low correlations may also occur for individuals or
16                   populations spending increased time indoors.

17                   The ratio between personal exposure and ambient concentration varies widely depending
18                   on activity patterns, housing  characteristics, and season, with higher personal-ambient
19                   ratios generally observed with increasing time spent outside, higher air exchange rate,
20                   and in seasons other than winter (Table 4-3). Personal-ambient ratios are typically  0.1-
21                   0.3, although individuals spending substantial time outdoors (e.g., outdoor workers) may
22                   have much higher ratios (0.5-0.9). Thus, applying personal-ambient ratios for outdoor
23                   workers to the general population or susceptible populations spending substantial time
24                   indoors can result in overestimates of the magnitude of personal exposure for these
25                   groups.

26                   Personal exposure to other pollutants shows variable association with personal exposure
27                   to O3, with differences in copollutant relationships depending on factors such as season,
28                   city-specific characteristics, activity pattern, and spatial variability of the copollutant
29                   (Section 4.3.4). In near-road  and on-road microenvironments, correlations between O3
30                   and traffic-related pollutants  are moderately to strongly negative, with the most strongly
31                   negative correlations observed for NO2 (-0.8 to -0.9). This is consistent with the
32                   chemistry of NO oxidation, in which O3 is consumed to form NO2. The more moderate
33                   negative correlations observed for PM25, PMi.0, and VOC may reflect reduced
34                   concentrations of O3 in polluted environments due to other scavenging reactions. A
35                   similar process occurs indoors, where infiltrated O3 reacts with airborne or surface-
36                   associated materials to form  secondary compounds, such as formaldehyde. Although such
      Draft - Do Not Cite or Quote                       4-42                                 September 2011

-------
 1                   reactions decrease indoor O3 exposure, they result in increasing exposure to other species
 2                   which may themselves have health effects.

 3                   Variations in ambient O3 concentrations occur over multiple spatial and temporal scales.
 4                   Near roadways, O3 concentrations are reduced due to reaction with NO and other species
 5                   (Section 4.3.4.2). Over spatial scales of a few kilometers and away from  roads, O3 may
 6                   be somewhat more homogeneous due to its formation as a secondary pollutant, while
 7                   over scales of tens of kilometers, additional atmospheric processing can result in higher
 8                   concentrations  downwind of an urban area. Although local-scale variability impacts the
 9                   magnitude of O3 concentrations, O3 formation rates are influenced by factors that vary
10                   over larger spatial scales, such as temperature (Section 3.2), suggesting that urban
11                   monitors may track one another temporally but miss small-scale variability in magnitude.
12                   The resulting uncertainty in exposure contributes to exposure measurement error in
13                   epidemiologic  studie s.

14                   Another factor that may influence epidemiologic results is the tendency for people to
15                   avoid O3 exposure by altering their behavior (e.g., reducing time spent outdoors) on high-
16                   O3 days. Activity pattern has a substantial effect on ambient O3 exposure, with time
17                   spent outdoors  contributing to increased exposure  (Section 4.4.2).  Averting behavior has
18                   been predominantly observed among children, older adults, and people with respiratory
19                   problems. Such effects are  less pronounced in the general population, possibly due to the
20                   opportunity cost of behavior modification.  Preliminary epidemiologic evidence reports
21                   increased asthma hospital admissions among children and older adults when O3 alert
22                   days were excluded from the analysis (presumably thereby eliminating averting behavior
23                   based on high O3 forecasts). The lower rate of admissions observed when alert days were
24                   included in the analysis suggests that estimates of health effects based on dose-response
25                   functions which do not account for averting behavior may be biased towards the null.

26                   The range of personal-ambient correlations reported by most studies (0.3-0.8) is similar
27                   to that for NO2 (U.S. EPA. 2008b) and somewhat lower than that for PM2 5 (U.S. EPA.
28                   2009d). To the extent that relative changes in central-site monitor concentration are
29                   associated with relative changes in exposure  concentration, this indicates that ambient
30                   monitor concentrations are  representative of day-to-day changes in average total personal
31                   exposure and in personal exposure to ambient O3.  The  lack of indoor sources of O3, in
32                   contrast to NO2 and PM2 5, is partly responsible for low indoor-outdoor ratios (generally
33                   0.1-0.4) and low personal-ambient ratios (generally 0.1-0.3), although it contributes to
34                   increased personal-ambient correlations. The lack of indoor sources also  suggests that
35                   fluctuations in  ambient O3  may be primarily  responsible for changes in personal
36                   exposure, even under low-ventilation, low-concentration conditions. Nevertheless, low
37                   personal-ambient correlations are a source  of exposure error for epidemiologic studies,
      Draft - Do Not Cite or Quote                       4-43                                September 2011

-------
1                   tending to obscure the presence of thresholds, bias effect estimates toward the null, and
2                   widen confidence intervals, and this impact may be more pronounced among populations
3                   spending substantial time indoors. The impact of this exposure error may tend more
4                   toward widening confidence intervals than biasing effect estimates, since epidemiologic
5                   studies evaluating the influence of monitor selection indicate that effect estimates are
6                   similar across different spatial averaging scales and monitoring sites.
         4.8    References

     Anderson. GB: Bell. ML. (2010). Does one size fit all? The suitability of standard ozone exposure metric
             conversion ratios and implications for epidemiology. J Expo Sci Environ Epidemiol 20: 2-11.
             http://dx.doi. org/10.1038/jes.2008.69.
     Anderson. SE: Wells. JR:  Fedorowicz. A: Butterworth. LF: Meade. BJ: Munson. AE. (2007). Evaluation of the
             contact and respiratory sensitization potential of volatile organic compounds generated  by simulated
             indoor air chemistry. Toxicol Sci 97: 355-363. http://dx.doi.org/10.1093/toxsci/kfm043.
     Aoki. T: Tanabe. S. (2007). Generation of sub-micron particles and secondary pollutants from building materials
             by ozone reaction. Atmos Environ 41: 3139-3150. http://dx.doi.orq/10.1016/i.atmosenv.20Q6.07.053.
     ASHRAE. (American Society of Heating,  Refrigerating and Air-Conditioning Engineers Inc.). (2009). The 2009
             ASHRAE Handbook-Fundamentals. In. Atlanta, GA: American Society of Heating, Refrigerating and
             Air-Conditioning Engineers Inc.
     Avol, EL: Navidi, WC: Colome, SD. (1998b). Modeling ozone levels in and around southern California homes.
             Environ Sci Technol 32: 463-468.
     Beckerman.  B: Jerrett. M: Brook. JR:  Verma. DK: Arain. MA: Finkelstein. MM. (2008). Correlation of nitrogen
             dioxide with other traffic pollutants near a major expressway. Atmos Environ 42: 275-290.
     Beelen. R: Hoek. G:  Pebesma. E: Vienneau. D: de Hoogh. K: Briggs. DJ. (2009). Mapping of background air
             pollution at a fine spatial scale across the European Union. Sci Total Environ 407: 1852-1867.
             http://dx.doi. org/10.1016/i.scitotenv.20Q8.11.048.
     Beko. G:  Clausen. G: Weschler. CJ. (2007). Further studies of oxidation processes on filter surfaces: Evidence
             for oxidation products and the influence of time in service. Atmos Environ 41: 5202-5212.
             http://dx.doi.0rg/10.1016/i.atmosenv.2006.07.063.
     Bell. ML:  McDermott. A: Zeger. SL: Samet. JM: Dominici.  F (2004). Ozone and short-term mortality in 95  US
             urban communities, 1987-2000.  JAM A 292: 2372-2378. http://dx.doi. orq/10.1001/iama.292.19.2372.
     Bell, ML.  (2006). The use  of ambient air quality modeling to estimate individual and population exposure for
             human health research: A case study of ozone in the Northern Georgia region of the United States.
             Environ Int 32: 586-593.
     Blanken,  PD: Dillon, J: Wismann, G. (2001). The impact of an air quality advisory program on voluntary mobile
             source air pollution reduction. Atmos Environ 35: 2417-2421.
     Blondeau. P: lordache. V: Poupard. O: Genin. D: Allard. F. (2005).  Relationship between outdoor and  indoor air
             quality in eight French schools. Indoor Air 15: 2-12.
     Brauer. M: Brook. JR. (1997). Ozone  personal exposures and health effects for selected groups  residing in the
             Fraser Valley. Atmos Environ 31: 2113-2121.
     Brauer, M: Brumm, J: Vedal, S: Petkau, AJ. (2002). Exposure misclassification and threshold concentrations in
             time series analyses of air pollution health effects. Risk Anal 22: 1183-1193.
     Brauer. M: Lencar. C: Tamburic. L: Koehoorn. M: Demers. P: Karr. C. (2008). A cohort study of traffic-related air
             pollution impacts on birth outcomes. Environ Health Perspect 116: 680-686.
     Bravo. MA: Bell. ML. (2011). Spatial heterogeneity of PM10 and O3 in Sao Paulo, Brazil, and implications for
             human health studies. J Air Waste Manag Assoc61: 69-77.
     Draft - Do Not Cite or Quote                        4-44                                  September 2011

-------
Breen. MS: Breen. M: Williams. RW: Schultz. BD. (2010). Predicting residential air exchange rates from
        questionnaires and meteorology: Model evaluation in central North Carolina. Environ Sci Technol 44:
        9349-9356. http://dx.doi.orq/10.1021/es101800k.
Bresnahan. BW: Dickie. M: Gerking. S. (1997). Averting behavior and urban air pollution. Land Econ 73: 34-57.
Briggs, DJ: Collins, S: Elliott, P: Fischer, P:  Kingham, S: Lebret, E: Pryl, K: Van Reeuwijk, H: Smallbone, K: Van
        Per Veen. A.  (1997). Mapping urban air pollution using GIS: A regression-based approach. Int J Geogr
        Inform Sci 11: 699-718.
Brown, K:  Sarnat, J: Suh, H: Coull,  B: Koutrakis, P. (2009). Factors influencing relationships between personal
        and ambient concentrations of gaseous and particulate pollutants. Sci Total Environ 407: 3754-3765.
Burke. JM: Zufall. MJ: Ozkavnak. H. (2001). A population exposure model for particulate matter:  Case study
        results for PM2.5 in Philadelphia, PA. J Expo Sci Environ Epidemiol 11: 470-489.
Chan. WR: Nazaroff. WW: Price. PN: Sohn. MD: Gadqil. AJ. (2005b). Analyzing a database of residential air
        leakage in the United States. Atmos Environ 39: 3445-3455.
Chang, L,  -T: Koutrakis, P: Catalano, PJ: Suh, HH. (2000). Hourly personal exposures to fine particles and
        gaseous pollutants—Results from Baltimore, Maryland. J Air Waste Manag Assoc 50: 1223-1235.
Chen. X: Hopke. PK: Carter. WP (2011). Secondary organic aerosol from ozonolysis of biogenic volatile organic
        compounds: Chamber studies of particle and reactive oxygen species formation. Environ Sci  Technol
        45: 276-282.  http://dx.doi.org/10.1021/es102166c.
Christakos. G: Vvas. VM. (1998a). A composite  space/time approach to studying ozone distribution over eastern
        United States. Atmos Environ 32: 2845-2857. http://dx.doi.Org/10.1016/S1352-2310(98)00407-5.
Christakos. G: Vvas. VM. (1998b). A novel method for studying population health impacts of spatiotemporal
        ozone distribution. Soc Sci Med 47: 1051-1066.
Darrow, LA: Klein, M:  Sarnat, JA: Mulholland, JA: Strickland, MJ: Sarnat, SE: Russell, AG: Tolbert, PE. (2011 b).
        The use of alternative pollutant metrics in time-series studies of ambient air pollution and respiratory
        emergency department visits. J Expo Sci Environ Epidemiol 21: 10-19.
        http://dx.doi. org/10.1038/jes.2009.49.
Delfino. RJ: Coate. BD: Zeiger. RS: Seltzer. JM: Street.  DH: Koutrakis. P.  (1996). Daily asthma severity in
        relation to personal ozone  exposure and outdoor fungal spores. Am J Respir Crit Care Med 154: 633-
        641.
Georgopoulos. PG: Purushothaman. V: Chiou. R. (1997). Comparative evaluation of methods for estimating
        potential human exposure  to ozone: Photochemical modeling and ambient monitoring. J Expo Sci
        Environ Epidemiol 7:  191-215.
Georgopoulos. PG: Wang. S. -W: Vvas. VM: Sun. Q: Burke. J: Vedantham. R:  McCurdv. T: Ozkavnak. H.
        (2005). A source-to-dose assessment of population exposures to fine  PM and ozone in Philadelphia,
        PA, during a summer 1999 episode. J Expo Sci Environ Epidemiol 15:  439-457.
Gevh. AS:  Wolfson. JM: Koutrakis.  P: Mulik. JD: Avol. EL. (1997). Development and evaluation of a small active
        ozone sampler. Environ Sci Technol 31: 2326-2330.
Gevh, AS:  Roberts, PT: Lurmann, FW:  Schoell,  BM: Avol, EL. (1999). Initial field evaluation of the Harvard
        active ozone sampler for personal  ozone monitoring. J Expo Sci Environ Epidemiol 9: 143-149.
Gevh. AS:  Xue. J: Ozkavnak. H: Spengler. JD. (2000). The Harvard Southern California chronic ozone exposure
        study: Assessing ozone exposure of grade-school-age children in two southern California communities.
        Environ Health Perspect 108: 265-270.
Gilliland. F: Avol. E: Kinnev. P:  Jerrett. M: Dvonch. T: Lurmann.  F: Buckley. T: Brevsse. P: Keeler. G: de Villiers.
        T: McConnell, R. (2005). Air pollution exposure assessment for epidemiologic studies of pregnant
        women and children:  Lessons learned from the Centers for Children's Environmental Health and
        Disease Prevention Research.  Environ Health Perspect 113: 1447-1454.
Glen,  G: Smith, L: Isaacs, K: Mccurdv, T: Langstaff, J. (2008). A new method of longitudinal diary assembly for
        human exposure modeling. J Expo Sci  Environ Epidemiol 18:  299-311.
        http://dx.doi.org/10.1038/si.ies.7500595.
Goldman,  GT: Mulholland, JA:  Russell, AG: Strickland, MJ: Klein, M: Waller, LA: Tolbert, PE. (2011). Impact of
        exposure measurement error in air pollution epidemiology: Effect of error type in time-series studies.
        Environ Health Global Access Sci Source 10: 61. http://dx.doi.org/10.1186/1476-069X-10-61.
Grosjean,  D: Hisham, MWM. (1992). A passive sampler for atmospheric ozone. J Air Waste Manag Assoc 42:
        169-173.
Draft - Do Not Cite or Quote                        4-45                                 September 2011

-------
Guav. M: Stieb. DM: Smith-Doiron. M. (2011). Assessment of long-term exposure to air pollution in a longitudinal
        national health survey. J Expo Sci Environ Epidemiol 21: 337-342.
        http://dx.doi. orq/10.1038/ies.2Q10.37.
Henry. GT: Gordon. CS. (2003). Driving less for better air: Impacts of a public information campaign. J Policy
        Anal Manage 22: 45-63. http://dx.doi.Org/10.1002/pam. 10095.
Heroux. ME: Clark. N: Van Rvswvk. K: Mallick. R: Gilbert. NL: Harrison. I: Rispler. K: Wang. D:
        Anastassopoulos. A: Guav. M: MacNeill. M: Wheeler. AJ. (2010). Predictors of indoor air concentrations
        in smoking and non-smoking residences. Int J Environ Res Public Health 7: 3080-3099.
        http://dx.doi.orq/10.3390/iierph7083080.
Hoek. G: Beelen. R: de Hoogh. K: Vienneau.  D: Gulliver. J: Fischer. P: Briggs. D. (2008). A review of land-use
        regression models to assess spatial variation of outdoor air pollution [Review]. Atmos Environ 42: 7561-
        7578.
Hvttinen. M:  Pasanen.  P: Kalliokoski.  P. (2006). Removal of ozone on  clean, dusty and sooty supply air filters.
        Atmos Environ 40: 315-325.
Jerrett. M: Burnett. RT: Pope. CA. Ill:  Ito. K: Thurston. G: Krewski. D: Shi. Y: Calle. E: Thun. M. (2009). Long-
        term ozone exposure and mortality. N Engl J Med 360: 1085-1095.
        http://dx.doi.org/10.1056/NEJMoa0803894.
Kanno. S: Yanagisawa. Y. (1992). Passive ozone/oxidant sampler with coulometric determination using
        iodine/nylon-6 charge-transfer complex. Environ Sci Technol 26: 744-749.
        http://dx.doi.org/10.1021/es00028a012.
Karner. AA: Eisinger. PS: Niemeier. DA. (2010). Near-roadway air quality: Synthesizing the findings from real-
        world data. Environ Sci Technol 44: 5334-5344. http://dx.doi.orq/10.1021/es1 OOOOSx.
Klepeis, NE. (1999). An introduction to the indirect exposure assessment approach: Modeling human exposure
        using microenvironmental measurements and the recent National Human Activity Pattern Survey.
        Environ Health Perspect 107: 365-374.
Klepeis, NE: Nelson, WC: Ott, WR: Robinson, JP: Tsang, AM: Switzer, P: Behar, JV: Hern, SC: Engelmann,
        WH. (2001). The National Human Activity Pattern Survey (NHAPS): A resource for assessing exposure
        to environmental pollutants. J Expo Sci Environ Epidemiol 11: 231-252.
Koutrakis, P: Wolfson,  JM: Bunyaviroch, A: Froehlich, SE: Hirano, K: Mulik, JD. (1993). Measurement of
        ambient ozone using a nitrite-coated filter. Anal Chem 65: 209-214.
Langstaff. JE. (2007). Analysis of uncertainty in ozone population exposure modeling [technical memorandum].
        Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
        Standards.
Lee. K: Parkhurst. WJ: Xue. J: Ozkavnak. H: Neuberg. D: Spengler. JD. (2004a). Outdoor/indoor/personal
        ozone exposures of children  in Nashville, Tennessee. J Air Waste Manag Assoc 54: 352-359.
Leech. JA: Nelson. WC: Burnett.  RT: Aaron. S: Raizenne. ME. (2002). It's about time: A comparison of
        Canadian and American time-activity patterns. J Expo Anal Environ Epidemiol 12: 427-432.
        http://dx.doi.org/10.1038/si.iea.7500244.
Liard. R: Zureik. M: Le  Moullec. Y: Soussan. D: Glorian.  M: Grimfeld. A: Neukirch. F. (1999). Use of personal
        passive samplers for measurement of NO2, NO, and O3 levels in  panel studies. Environ Res 81: 339-
        348.
Linn. WS: Shamoo. DA: Anderson. KR:  Peng. R. -C: Avol. EL:  Hackney. JD: Gong. H. Jr. (1996). Short-term  air
        pollution exposures and  responses in Los Angeles area schoolchildren. J Expo Sci Environ Epidemiol
        6: 449-472.
Liu. L. -JS: Koutrakis. P:  Leech. J: Broder. I. (1995). Assessment of ozone exposures in the greater metropolitan
        Toronto area. J Air Waste Manag Assoc 45: 223-234.
Lopez-Aparicio, S: Smolfk, J: Maskova, L: Souckova, M: Gr0ntoft, T: Ondrackova, L: Stankiewicz, J. (2011).
        Relationship of indoor and outdoor air pollutants in a naturally ventilated historical building envelope.
        Build Environ 46: 1460-1468. http://dx.doi.Org/10.1016/i.buildenv.2011.01.013.
Mansfield, CA: Johnson,  FR: Van Houtven, GL. (2006). The missing piece: Valuing averting behavior for
        children's ozone exposures. Resource Energ Econ 28: 215-228.
        http://dx.doi.0rg/10.1016/i.reseneeco.2006.02.002.
Marshall, JD: Netherv,  E: Brauer, M. (2008). Within-urban variability in ambient air pollution: Comparison of
        estimation methods. Atmos Environ 42: 1359-1369. http://dx.doi.Org/10.1016/i.atmosenv.2007.08.012.
Draft - Do Not Cite or Quote                        4-46                                  September 2011

-------
McConnell. R: Berhane. K: Yao. L: Lurmann. FW: Avol. E: Peters. JM. (2006). Predicting residential ozone
        deficits from nearby traffic. Sci Total Environ 363: 166-174.
McCurdv. T: Glen. G: Smith. L: Lakkadi. Y. (2000). The National Exposure Research Laboratory's consolidated
        human activity database. J Expo Sci Environ Epidemiol  10: 566-578.
McDermott, M: Srivastava, R: Croskell, S. (2006). Awareness of and compliance with air pollution advisories: A
        comparison of parents of asthmatics with other parents.  J Asthma 43: 235-239.
        http://dx.doi.org/10.1080/02770900600567114.
Neidell,  M. (2009). Information, avoidance behavior, and health: The effect of ozone on asthma hospitalizations.
        Journal of Human Resources 44: 450-478.
Neidell.  M. (2010). Air quality warnings and outdoor activities: Evidence from Southern California using a
        regression discontinuity design. J Epidemiol Community Health 64: 921-926.
        http://dx.doi.orq/10.1136/iech.2008.081489.
Neidell.  M: Kinnev. PL. (2010). Estimates of the association between ozone and asthma hospitalizations that
        account for behavioral responses to air quality information. Environ Sci Pol 13: 97-103.
        http://dx.doi.0rg/10.1016/i.envsci.2009.12.006.
O'Neill. MS: Ramirez-Aguilar. M:  Meneses-Gonzalez. F: Hernandez-Avila. M: Gevh. AS:  Sienra-Monge. JJ:
        Romieu, I. (2003). Ozone exposure among Mexico City outdoor workers. J Air Waste Manag Assoc 53:
        339-346.
Ogawa:  Co. (Ogawa & Company). (2007). Ambient air passive sampler for NO-NO2, NOx, SO2, O3, NH3.
        Pompano Beach, FL: Ogawa & Company USA,  Inc. http://www.ogawausa.com/passive.html.
Ramfrez-Aguilar.  M: Barraza-Villarreal. A:  Moreno-Macfas. H: Winer. AM: Cicero-Fernandez. P: Velez-Marguez.
        MG: Cortez-Lugo. M: Sienra-Monge. JJ: Romieu. I. (2008). Assessment of personal exposure to ozone
        in asthmatic children residing in Mexico City. Salud Publica Mex 50: 67-75.
Reiss. R: Ryan. PB: Tibbetts. SJ: Koutrakis. P. (1995a).  Measurement of organic acids, aldehydes, and ketones
        in residential environments and their relation to ozone. J Air Waste  Manag Assoc 45: 811-822.
Reiss, R: Ryan, PB: Koutrakis, P: Tibbetts, SJ. (1995b). Ozone reactive chemistry on interior latex paint.
        Environ Sci Technol 29: 1906-1912.
Riediker. M: Williams. R:  Devlin. R: Griggs. T: Bromberg. P. (2003). Exposure to particulate matter, volatile
        organic compounds, and other air pollutants inside patrol cars. Environ Sci Technol 37: 2084-2093.
        http://dx.doi. org/10.1021/es026264v.
Romieu. I: Lugo. MC: Colome. S: Garcia. AM: Avila. MH: Gevh. A: Velasco.  SR: Rendon. EP (1998b).
        Evaluation of indoor ozone concentration and predictors of indoor-outdoor ratio in Mexico City. J Air
        Waste Manag Assoc 48: 327-335.
Ryan. PH: LeMasters. GK. (2007). A review of land-use regression models for characterizing intraurban air
        pollution exposure  [Review].  Inhal Toxicol 19: 127.
Sarnat. JA: Koutrakis. P:  Suh. HH. (2000). Assessing the relationship between personal particulate and
        gaseous exposures of senior citizens living in Baltimore, MD. J Air Waste Manag Assoc 50: 1184-1198.
Sarnat, JA: Schwartz, J: Catalano, PJ: Suh, HH.  (2001).  Gaseous pollutants in particulate matter epidemiology:
        Confounders or surrogates?  Environ Health Perspect 109: 1053-1061.
Sarnat. JA: Brown. KW: Schwartz. J:  Coull. BA: Koutrakis. P. (2005). Ambient gas concentrations and personal
        particulate matter exposures: implications for studying the health effects of particles. Epidemiology 16:
        385-395.
Sarnat. SE: Coull. BA: Schwartz. J: Gold. PR: Suh. HH. (2006b). Factors affecting the association between
        ambient concentrations and personal exposures to particles and gases. Environ Health Perspect 114:
        649-654.
Sarnat. SE: Klein.  M: Sarnat. JA: Flanders. WD: Waller.  LA: Mulholland. JA: Russell. AG: Tolbert. PE. (2010). An
        examination of exposure measurement error from air pollutant spatial variability in time-series  studies.
        J Expo Sci Environ Epidemiol 20: 135-146. http://dx.doi.org/10.1038/ies.20Q9.10.
Semenza. JC: Wilson. DJ: Parra. J: Bontempo. BD: Hart. M: Sailor. DJ: George. LA. (2008). Public perception
        and behavior change in relationship to hot weather and air pollution. Environ Res 107: 401-411.
        http://dx.doi.0rg/10.1016/i.envres.2008.03.005.
Sheppard. L: Slaughter. JC:  Schildcrout. J: L-JS. L: Lumlev. T. (2005). Exposure and measurement
        contributions to estimates of  acute air pollution effects. J Expo Sci Environ Epidemiol 15: 366-376.
Draft - Do Not Cite or Quote                        4-47                                  September 2011

-------
Sherman. M: McWilliams. J. (2007). Air leakage of U.S. homes: Model prediction. (LBNL-62078). Berkeley, CA:
        Lawrence Berkeley National Laboratory, http://epb.lbl.gov/publications/lbnl-62078.pdf.
Sherman. MH: Grimsrud. DT. (1980). Infiltration-pressurization correlation: Simplified physical modeling. In
        ASHRAE Transactions (pp. 778-807). Denver, CO: Lawrence Berkeley Laboratory.
Shu, S: Morrison, GC. (2011). Surface reaction rate and probability of ozone and alpha-terpineol on glass,
        polyvinyl chloride, and latex paint surfaces. Environ Sci Technol 45: 4285-4292.
        http://dx.doi. orq/10.1021/es200194e.
Suh, HH: Zanobetti, A. (2010). Exposure error masks the relationship between traffic-related air pollution and
        heart rate variability [Erratum]. J Occup Environ Med 52:  1138.
        http://dx.doi. orq/10.1097/JOM.Ob013e3181 fd2632.
U.S. EPA. (U.S. Environmental Protection Agency). (2007b). Review of the national ambient air quality
        standards for ozone: Policy assessment of scientific and  technical information: OAQPS staff paper.
        (EPA/452/R-07/003). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2008b). Integrated science assessment for oxides of
        nitrogen: Health criteria. (EPA/600/R-08/071). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645.
U.S. EPA. (U.S. Environmental Protection Agency). (2009a). Consolidated Human Activity Database, from
        http://www.epa.gov/chadnet1/
U.S. EPA. (U.S. Environmental Protection Agency). (2009b). Human exposure modeling: Air pollutants exposure
        model (APEX/TRIM.Expo Inhalation), from http://www.epa.gov/ttn/fera/human apex.html
U.S. EPA. (U.S. Environmental Protection Agency). (2009d). Integrated science assessment for particulate
        matter. (EPA/600/R-08/139F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
U.S. EPA. (U.S. Environmental Protection Agency). (2011 b). Exposure model  for individuals, from
        http://www.epa.gov/heasd/products/emi/emi.html
Wallace, L: Williams, R: Suggs, J: Jones, P. (2006). Estimating contributions of outdoor fine particles to indoor
        concentrations and personal exposures: Effects of household characteristics  and  personal activities.
        (EPA/600/R-06/023). Research Triangle Park, NC:  U.S. Environmental Protection Agency.
Wen, XJ: Balluz, L: Mokdad, A. (2009). Association between media alerts of air quality index and change of
        outdoor activity among adult asthma in six states,  BRFSS, 2005. J Community Health 34: 40-46.
        http://dx.doi. orq/10.1007/s10900-008-9126-4.
Weschler, CJ: Shields, HC. (1997). Potential reactions among indoor pollutants. Atmos Environ 31: 3487-3495.
White. IR: Martin. D: Munoz. MP: Petersson. FK: Henshaw. SJ: Nickless. G: Lloyd-Jones.  GC: Clemitshaw. KG:
        Shallcross. DE. (2010). Use of reactive tracers to determine ambient  OH radical concentrations:
        Application within the indoor environment. Environ Sci Technol 44: 6269-6274.
        http://dx.doi.org/10.1021/es901699a.
Williams. R: Rea. A: Vette. A: Croghan. C: Whitaker. D: Stevens. C:  McDow. S: Fortmann.  R: Sheldon. L:
        Wilson, H: Thornburg, J: Phillips, M: Lawless, P: Rodes,  C: Daughtrev, H. (2009b). The design and field
        implementation of the Detroit exposure and aerosol research study. J Expo Sci  Environ Epidemiol 19:
        643-659. http://dx.doi.orq/10.1038/ies.2008.61.
Wilson, KL: Birks, JW. (2006). Mechanism and elimination  of a water vapor interference in the measurement of
        ozone by UV absorbance. Environ Sci Technol 40: 6361-6367. http://dx.doi.orq/10.1021/es052590c.
Wilson. WE: Suh. HH. (1997). Fine particles and coarse particles: Concentration relationships relevant to
        epidemiologic studies. J Air Waste Manag Assoc 47: 1238-1249.
Wilson. WE: Mage. DT: Grant. LD. (2000).  Estimating separately  personal exposure to ambient and  nonambient
        particulate matter for epidemiology and risk assessment: Why and how. J Air Waste Manag  Assoc 50:
        1167-1183.
Xue. J: McCurdv. T: Spengler. J: Ozkavnak. H. (2004). Understanding variability in time spent in selected
        locations for 7-12-year old children. J Expo Anal Environ  Epidemiol 14: 222-233.
        http://dx.doi. org/10.1038/sj.jea.7500319.
Xue. J: Liu. SV: Ozkavnak. H: Spengler. JD. (2005).  Parameter evaluation and model validation of ozone
        exposure assessment using Harvard Southern California Chronic Ozone Exposure Study data. J Air
        Waste Manag Assoc 55: 1508-1515.
Draft - Do Not Cite or Quote                        4-48                                  September 2011

-------
     Zartarian. VG: Schultz. BD. (2010). The EPA's human exposure research program for assessing cumulative risk
             in communities. J Expo Sci Environ Epidemiol 20: 351-358. http://dx.doi.org/10.1038/ies.2009.20.
     Zauli Saiani. S: Hanninen. O: Marchesi. S: Lauriola. P.  (2011). Comparison of different exposure settings in a
             case-crossover study on air pollution and daily mortality: Counterintuitive results. J Expo Sci Environ
             Epidemiol 21: 385-394. http://dx.doi.org/10.1038/jes.2010.27.
     Zeger. SL: Thomas. D: Dominici. F: Samet. JM: Schwartz. J: Dockerv. D: Cohen. A. (2000). Exposure
             measurement error in time-series studies of air pollution: Concepts and consequences.  Environ Health
             Perspect 108: 419-426.
1    Zidek. JV: Shaddick. G: Meloche. J: Chatfield.  C: White. R. (2007).  A framework for predicting
2    personal exposures to environmental hazards. Environ Ecol Stat 14: 411-431
     Draft - Do Not Cite or Quote                        4-49                                 September 2011

-------
      5   DOSIMETRY  AND  MODE  OF  ACTION
          5.1    Introduction

 1                   This chapter has two main purposes. The first is to describe the principles which underlie
 2                   the dosimetry of O3 and to discuss factors which influence it. The second is to describe
 3                   the modes of action leading to the health effects that will be presented in Chapters 6 and
 4                   7. This chapter is not intended to be a comprehensive overview, but rather, it updates the
 5                   basic concepts derived from O3 literature presented in previous documents (U.S. EPA.
 6                   2006b. 1996a) and introduces the recent relevant literature.

 7                   In Section 5.2, particular attention is given to dosimetric factors influencing individual
 8                   risk of developing effects from O3 exposure. As there have been few O3 dosimetry
 9                   studies published since the last AQCD, the reader is referred to previous documents (U.S.
10                   EPA, 2006b. 1996a) for more detailed discussion of the past literature. Evaluation of the
11                   progress in the interpretation of past dosimetry studies, as well as studies published since
12                   2005, in the areas of uptake, reactions, and models for O3 dosimetry, is discussed.

13                   Section 5.3 highlights findings of studies published since the 2006 O3 AQCD, which
14                   provide insight into the biological pathways by which O3 exerts its actions. Since
15                   common mechanisms lead to health effects from both short- and long-term exposure to
16                   O3, these pathways are discussed in Chapter 5 rather than in later chapters. The relevant
17                   sections of Chapters 6 and 7 are indicated. Older studies which represent the current state
18                   of the science are also discussed. Studies conducted at more environmentally-relevant
19                   concentrations of O3 are of greater interest, since mechanisms responsible for effects at
20                   low O3 concentrations may not be identical to those occurring at high O3 concentrations.
21                   The topics of dosimetry and mode of action are bridged by reactions of O3 with
22                   components of the extracellular lining fluid (ELF), which play a role in both O3 uptake
23                   and biological responses (Figure 5-1).

24                   In addition, this chapter discusses interindividual variability in responses, and issues
25                   related to species comparison of doses and responses (Sections 5.4 and 5.5). These topics
26                   are included  in this chapter because they are influenced by both dosimetric and
27                   mechanistic considerations.
      Draft - Do Not Cite or Quote                        5-1                                 September 2011

-------
                                               Net03
                                                dose
        03 exposure
           Inhaled
           03dose
Modes of Action
                                                                                Health Effects
                                          Tissue 03 dose and
                                          product formation
      Figure 5-1     Schematic of the Os exposure and response pathway.  Os
                     concentrations can be reported as the exposure concentration,
                     inhaled dose, the net dose, or the local tissue dose. The net dose
                     refers to the total absorption of Os and is the sum of all the tissue
                     compartmental doses. Chapter 5 discusses the concepts of dose
                     and modes of action that result in the health effects discussed in
                     Chapters 6 and 7.
         5.2    Human and Animal Ozone Dosimetry
           5.2.1   Introduction
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13

14
15
16
17
Dosimetry refers to the measurement or estimation of the quantity of or rate at which a
chemical and/or its reaction products are absorbed and retained at target sites. Dose refers
to the amount of O3 crossing an exposure surface to enter a target area. In the literature,
surrogates of dose of reactive gases, such as O3, can range in refinement from their
concentration in the ambient exposure atmosphere to the "effective" dose of the chemical
or its reaction products that actively participate in toxic reactions (Dahl. 1990). However,
ambient concentrations are not a true measure of dose. Ideally, the units for the
expression of the dose of O3 might range from the quantity of gas inhaled as the product
of gas concentration x minute  ventilation x time (units of ppm x L x h), to the quantity of
gas retained by the whole body, to the concentration of gas molecules that have been
absorbed or reacted with the tissue (moles/g tissue weight). In modeling studies, the dose
rate is often expressed as a flux per unit of surface area of a region of respiratory
epithelium.

Ozone is a highly reactive, though poorly water soluble, gas at physiological temperature.
The latter feature is believed to be the reason why it is able to penetrate into targets in the
lower respiratory tract (LRT).  Figure 5-2 presents the basic structure of the human
respiratory tract (RT). The lung can be divided into three major regions: the extrathoracic
     Draft - Do Not Cite or Quote
                             5-2
                     September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
(ET) region or upper respiratory tract (URT, from the nose/mouth to larynx); the
tracheobronchial (TB) tree (from trachea to the terminal bronchioles); and the alveolar or
pulmonary region (from the respiratory bronchioles to the terminal alveolar sacs). The
latter two regions comprise the LRT. Although the structure varies, the illustrated
anatomic regions are common to all mammalian species with the exception of the
respiratory bronchioles. Respiratory bronchioles, the transition region between ciliated
and fully alveolated airways, are found in humans, dogs, ferrets, cats, and monkeys.
Respiratory bronchioles are  absent in rats and mice and abbreviated in hamsters, guinea
pigs, sheep, and pigs. The branching structure of the ciliated bronchi and bronchioles also
differs between species from being a rather symmetric and dichotomous branching
network of airways in humans to a more monopodial branching network in other
mammals.
                            Extrathoracic
                             Region
                                   Pha
                 rynxJ '
 Posterior
 Nasal Passage
Nasal Part
 Oral Part
                          Tracheobronchial
                             Region
                                                                       Bronchiolar Region
                                                              Bronchioles
                                                                Terminal Bronchioles
                                                              Respiratory Bronchioles
                                                                       Alveolar Interstitial
                                                            Alveolar Duct +
                                                            Alveoli
       Source: Based on ICRP (1994)

      Figure 5-2     Representation of respiratory tract regions in  humans. Structures
                      are anterior nasal  passages, ETi; oral airway and posterior nasal
                      passages, ET2; bronchial airways, BB; bronchioles, bb; and
                      alveolar interstitial, Al.
      Draft - Do Not Cite or Quote
                                5-3
                                               September 2011

-------
                  b.
                          Air
                         Lquic
                          nun
                         Tissue
                                       Tissue
                                                        Air
                                                        Air
                  C.
  Transport Factors
   Gas Phase
    Convection
    Diffusion
    Dispersion

   Liquid Phase
    Solubility
    Diffusion
    Chemical Reaction
    Convection
 Source: Panel (a) reprinted with permission from McGraw-Hill (Weibel. 1980)

Figure 5-3    Structure of lower airways with progression from the large airways
              to the alveolus, (a) Illustrates basic airway anatomy. Structures are
              epithelial cells, EP; basement membrane, BM; smooth muscle cells,
              SM; and fibrocartilaginous coat, FC.  (b) Illustrates the relative
              amounts of liquid, tissue, and blood  with distal progression. In the
              bronchi there is a thick surface lining over a relatively thick layer of
              tissues. With distal progress, the lining diminishes allowing
              increased access of compounds crossing the air-liquid  interface to
              the tissues and the blood, (c) Presents the factors acting in the gas
              and liquid phases of Oz transport.
Draft - Do Not Cite or Quote
5-4
September 2011

-------
 1                   Figure 5-3 illustrates the structure of the LRT with progression from the large airways in
 2                   the TB region to the alveolus in the alveolar region. The fact that O3 is so chemically
 3                   reactive has suggested to some that its effective dose at the target sites exists in the form
 4                   of oxidation products such as aldehydes and peroxides (see Section 5.2.3). Reaction
 5                   products are formed when O3 interacts with components of the ELF such as lipids and
 6                   antioxidants. The ELF varies throughout the length of the RT with the bronchial tree lined
 7                   with a thin film of mucus and the alveolar region lined with a thinner layer of surfactant.
 8                   Ozone toxicity is observed to some extent in the nasal cavity, however further toxicity
 9                   exists in the LRT where the thinness of the ELF layer allows O3  to react directly with
10                   cells protruding from the ELF (Figure 5-3b). Ozone uptake relates directly to these ELF
11                   substrate reactions and is termed "reactive absorption." Thus, the uptake of O3 is related
12                   to both the concentration of O3 as well as the availability of substrates within the ELF.

13                   Chemical reactions are not the only processes controlling the uptake of O3 from the
14                   airstream into compartments of the RT (Figure 5-3c). Ozone uptake is affected by
15                   complex interactions between a number of major factors including RT morphology,
16                   breathing route, frequency, and volume, physicochemical properties of the gas, physical
17                   processes of gas transport, as well as the physical and chemical properties of the ELF and
18                   tissue layers. The role of these processes varies throughout the length of the RT and as O3
19                   moves from the gas to liquid compartments of the RT.

20                   Two types of measurements have been used to arrive at the O3 dose to target sites during
21                   breathing: (1) measurement of removal of O3 from the air stream (termed "uptake"); and
22                   (2) measurement of chemical reactions in tissues or with biomolecules known to be
23                   present in tissues (termed "reactants"). The results of the above measurements have been
24                   incorporated into mathematical models for the purpose of explaining, predicting, and
25                   extrapolating O3 dose in different exposure scenarios. Few new studies have investigated
26                   the uptake of O3 in the RT since the last O3 assessment (U.S. EPA. 2006b). The studies
27                   that have  been conducted generally agree with the results presented in the past and do not
28                   change the dosimetry conclusions of the last document.
             5.2.2   Ozone Uptake

29                   Past AQCDs provide information on the majority of literature relevant to understanding
30                   the state of the science in O3 dosimetry. One method of quantifying O3 dosimetry is to
31                   measure the amount of O3 removed from the air stream during breathing (termed
32                   "uptake"). The O3 in the breath that is removed during the breathing period is termed
33                   "uptake efficiency" or fractional absorption. Uptake studies have utilized bolus and
34                   continuous O3 breathing techniques as well as modeling to investigate uptake efficiency
      Draft - Do Not Cite or Quote                       5-5                                 September 2011

-------
1 and distribution of O3 uptake between the upper and lower respiratory tract. A number of
2 the studies that have measured the fractional uptake of O3 in the human RT (FRT), URT
3 (FURT), and LRT (FLRT) are presented in Table 5-1 .
4 Table 5-1 Human respiratory tract uptake efficiency data
Reference
Mouth/Nose'1
Inspiratory Flow
(mL/s)
VT (mL|
fB (bpm)b
FRT FURT
FLRT
CONTINUOUS EXPOSURE
Gerrityetal. (1988)



Gerrityetal. (1994)°

Gerrityetal. Q995)
Wiesteretal. (1996c)

Santiago et al. (2001)

Rigasetal. (2000)
M
N
M/N
M/N
M
M
Mouthpiece
M
N
N
N
Face mask
509
456
350
634
1,360
1,360
330
539
514
50
250
480
832
754
832
778
1,650
1,239
825
631
642


1,100
18
18
12
24
25
35
12
16
16


27.6
0.40
0.36
0.41
0.38
0.81 0.37
0.78 0.41
0.91 0.27
0.76
0.73
0.80d
0.33
0.86
0.91
0.91
0.93
0.89
0.43
0.36
0.95





BOLUS EXPOSURE
Hu et al. Q992)
Ultmanetal. (1994)

Ultmanetal. (2004)

Nodelman and Ultman (1999)



Mouthpiece
Mouthpiece
Mouthpiece
M
M
Nasal Cannula
Nasal Cannula
Mouthpiece
Mouthpiece
250
250
250
490
517
150
1,000
150
1,000

500e
500
450e
574
500
500
500
500

15
15
32.7
27
18
120
18
120
0.96 0.46
0.30
0.47
0.87
0.91
0.90
0.45
0.80
0.25







0.95
0.90
        aM = mouth exposure during spontaneous breathing; N = nasal exposure during spontaneous breathing; M/N = pooled data from mouth and
       nasal exposure; mouthpiece = exposure by mouthpiece; FRT = total RT uptake; FUra- = upper RT uptake; FLRT = lower RT uptake.
        bfB is either measured or is computed from flows and VT.
        Total RT uptake reported by Gerrity et al. (1988) and Gerrity et al. (1994) did not include the contribution from URT uptake efficiency during
       expiration. The data include an expiratory URT contribution, assuming it equals inspiratory URT uptake efficiency.
        dFURT from Santiago et al. (2001) represents nasal absorption (Fnose).
        eVT is computed from flow and fB.
 5
 6
 7
 8
 9
10
5.2.2.1     Gas Transport Principles

Transport of O3 in the gas phase is governed by bulk flow or convection, effective axial
dispersion, and loss to airway walls (Figure 5-3c) (Miller, 1995). The relative importance
of these gas phase transport mechanisms varies among RT regions for a given level of
ventilation in any species. For example, bulk airflow is the predominant mechanism for
gas transport in the URT and bronchi, while diffusion is the major transport mechanism in
the alveolar region of the lung.
       Draft - Do Not Cite or Quote
                                    5-6
September 2011

-------
 1                   Gas transport in the TB region occurs by a combination of bulk flow and mixing
 2                   (Ultman. 1985). Mixing can occur by diffusion processes associated with the molecular
 3                   nature of the gas or by dispersion processes which depend on local velocity patterns. The
 4                   complexity of the airway structure and surface affects the bulk airflow patterns so that not
 5                   all nasal and lung surfaces receive the same O3 exposure or dose (Miller and Kimbell.
 6                   1995). The principal influence on mixing in the TB region comes from the axial velocity
 7                   profile and diffusion. When passing through a bifurcation a velocity profile is altered; the
 8                   inspiratory profile is sharper than the more flattened expiratory profile (Schroter and
 9                   Sudlow. 1969). The longitudinal dispersion of inspired air depends on the flow regime,
10                   laminar flow (i.e., streamlined) or turbulent flow (i.e., possessing random velocity
11                   fluctuations), and is influenced by Taylor dispersion forces. In humans, turbulent flow
12                   regime persists only a few generations into the RT Turbulence generation also varies by
13                   species and flow rates. For example, airflow is nonturbulent in the rat nose at any
14                   physiologic flow rate but may be highly turbulent in the human nose during exercise
15                   (Miller. 1995).

16                   Conversely, the principal mechanism of gas mixing in the lung periphery is molecular
17                   diffusion (Engel. 1985). While moving into more distal areas  of the RT, the cross-
18                   sectional area of the airways rapidly increases and linear velocities decrease,  leading to a
19                   greater role of molecular diffusion of gases. Gas molecules close to the alveolocapillary
20                   membrane have almost zero convective velocity with respect  to the membrane. Overall,
21                   the diffusion of O3 into the ELF where chemical reactions occur drives alveolar gas
22                   uptake.
                     5.2.2.2    Target Sites for Ozone Dose

23                   A primary uptake site of O3 delivery to the lung epithelium is believed to be the
24                   centriacinar region (CAR). The CAR refers to the zone at the junction of the TB airways
25                   and the gas exchange region. This area is also termed the proximal alveolar region (PAR)
26                   and is defined as the first generation distal to the terminal bronchioles. Contained within
27                   the CAR, the respiratory bronchioles were confirmed as the site receiving the greatest O3
28                   dose (18O mass/lung weight) in resting O3 exposed rhesus monkeys, when not
29                   considering the nose (Plopper et al.. 1998). Furthermore, the greatest cellular injury
30                   occurred in the vicinity of the respiratory bronchioles and was dependent on the delivered
31                   O3 dose to these tissues (see also Section 5.4.1). However, 18O label was detected to a
32                   lesser extent in other regions of the TB airway tree, showing that O3 is delivered to these
33                   compartments as well, although in a smaller dose. Earlier models predicted that the net
34                   O3 dose (total absorption, O3 flux to air-liquid interface) gradually decreased with distal
3 5                   progression from the trachea to the end of the TB region and then rapidly decreased in the

      Draft - Do Not Cite or Quote                        5-7                                 September 2011

-------
 1                   alveolar region (Miller et al., 1985). However, the tissue O3 dose (O3 flux to liquid-tissue
 2                   interface) was low in the trachea, increased to a maximum in the terminal bronchioles
 3                   and the CAR, and then rapidly decreased in the alveolar region. Despite the exclusion of
 4                   the URT and O3 reactions with ELF constituents after the 16th generation, the model
 5                   predicted experimental results showing that the CAR received the greatest O3 tissue dose
 6                   (Miller et al.. 1985).

 7                   Inhomogeneity in the RT structure may affect the dose delivered to this target site.
 8                   Models have predicted that the farther the PAR is from the trachea, the less the O3 tissue
 9                   dose to the region. Ultman and Anjilvel (1990) and Overton et al. (1989) predicted
10                   approximately a 50 to 300% greater PAR dose for the shortest path relative to the longest
11                   path in humans and rats, respectively. In addition, Mercer et al. (1991) found that both
12                   path distance and ventilatory unit size affected dose. The variation of O3 dose among
13                   anatomically equivalent ventilatory units was predicted to vary as much as six-fold, as a
14                   function of path length from the trachea. This could have implications in regional damage
15                   to the LRT, such that even though the average LRT dose may be at a level that would be
16                   considered insignificant, local regions of the RT may receive significantly higher than
17                   average doses and therefore be at greater risk of effects.
                     5.2.2.3    Upper Respiratory Tract Ozone Removal and Dose

18                   The URT provides a defense against O3 entering the lungs by removing half of the
19                   inhaled O3 from the airstream. In both animals and humans, about 50% of the absorbed
20                   O3 was removed in the head (nose, mouth, and pharynx), about 7% in the larynx/trachea,
21                   and about 43% in the lungs (Huetal.. 1992; Hatch etal.. 1989; Miller et al.. 1979). The
22                   fraction of O3 taken up was inversely related to flow rate and weakly related to inlet O3
23                   concentration (Yokoyama and Frank.  1972). The limiting factors in nasal O3 uptake were
24                   simultaneous diffusion and chemical reaction of O3 in the nasal ELF layer (Santiago et
25                   al.. 2001). The ELF layer in the nose is thicker than in the rest of the RT, and
26                   mathematical estimates predicted that O3 penetrates less than the thickness of the ELF
27                   layer; reaction products are likely the  agents damaging the nasal tissue and not O3 itself.
28                   It was hypothesized that the nasal nonlinear reaction kinetics could result from the
29                   depleting substrates in the nasal ELF becoming the limiting factor of the reaction
30                   (Santiago etal..2001).

31                   Uptake efficiencies have been measured for various segments of the URT (Table 5-1).
32                   Gerrity et al. (1995) reported unidirectional uptake efficiencies of O3  inhaled from a
33                   mouthpiece; of 17.6% from the mouth to vocal cords, 9.5% from the vocal cords to the
34                   upper trachea (totaling 27.1%), 8.4% from the upper trachea to the main bifurcation
      Draft - Do Not Cite or Quote                       5-8                                 September 2011

-------
 1                  carina (totaling 35.5%), and essentially zero between the carina and the bronchus
 2                  intermedius (totaling 32.5%). These values are lower than those calculated by Hu et al.
 3                  (1992) that reported uptake efficiencies of 21, 36, 44, and 46% between the mouth and
 4                  the vocal cords, the upper trachea, the main bifurcation carina, and the bronchus
 5                  intermedius, respectively. The lower efficiencies seen in Gerrity et al. (1995) may have
 6                  resulted from the mouthpiece scrubbing O3 from the breath during inhalation.

 7                  Past studies investigating nasal uptake of O3 have shown that the nose partially protects
 8                  the LRT from damage from inspired O3 (Santiago et al.. 2001; Gerrity et al..  1988).
 9                  Sawyer et al. (2007) further investigated nasal uptake of O3 in healthy adults during
10                  exercise. Fractional O3 uptake, acoustic rhinometry (AR), and nasal NO measurements
11                  were taken on ten adults (8 women, 2 men) exposed to 200 ppb O3 before and after
12                  moderate exercise at two flow rates (10 and 20 L/min).  The percent nasal uptake of O3
13                  was -50% greater at 10 L/min compared to 20 L/min both pre- and post-exercise.
14                  However, the inhaled O3 delivered dose to the LRT (i.e., flow rate  x [O3 ppm] x nasal O3
15                  penetration) was 1.6-fold greater at the higher flow than at the lower flow (2.5 compared
16                  to 0.9 ppm-L/min). Prior exercise did not affect O3 uptake at  either flow rate, but did
17                  significantly increase nasal volume (Vn) and AR measurements of nasal cross-sectional
18                  area (minimum cross-sectional area (MCA) which corresponds to the nasal valve, CSA2
19                  which corresponds to the anterior edge of the nasal turbinates, and  CSA3 which
20                  corresponds to the posterior edge of the nasal turbinates) (p < 0.05). Conversely, exercise
21                  decreased nasal resistance (Rn) (p < 0.01) and NO production (nonsignificant, p > 0.05).
22                  The change in Vn and CSA2:MCA ratio was correlated with the percent change in nasal
23                  uptake, however the overall effect was small and sensitive to elimination of outliers and
24                  gender segregation.

25                  Overall, the URT removes half of the inhaled O3 by reactions in the nasal ELF. The exact
26                  uptake efficiency will change due to variations in flow rate and inhaled concentration.
                    5.2.2.4   Lower Respiratory Tract Ozone Uptake and Dose

27                  Total O3 uptake in the entire RT in rats and guinea pigs ranges from 40-54% efficient
28                  (Hatch etal.  1989; Wiesteretal.. 1988; Wiester et al.. 1987). while in humans at rest it
29                  ranges from 80-95% efficient (Huetal..  1992). Approximately 43% of inhaled O3 is
30                  absorbed in the LRT of both humans and animals. Models predicted that the net O3 dose
31                  decreases distally from the trachea toward the end of the TB region and then rapidly
32                  decreases in the alveolar region (Miller etal.. 1985). However, these models predicted
33                  low tissue O3 dose in the trachea and large bronchi.  As injury has been seen in these
34                  areas, net dose may be a better predictor of local toxic tissue dose.
      Draft - Do Not Cite or Quote                       5-9                                September 2011

-------
 1                   Uptake efficiency depends on a number of variables, including O3 exposure
 2                   concentration, exposure time, and breathing pattern. For breaths of similar waveforms,
 3                   respiratory patterns are uniquely described by breathing frequency (fB) and tidal volume
 4                   (Vj); by minute volume (MV = fB x VT) and fB; or by MV and VT. Simulations from the
 5                   Overton et al. (1996) single-path anatomical respiratory tract model, where the upper and
 6                   lower respiratory tracts were modeled but uptake by the URT was not considered,
 7                   predicted that fractional uptake and  PAR O3 dose increased with VT when fB was held
 8                   constant. Likewise, experimental studies found that O3 uptake was positively correlated
 9                   with changes in VT (Ultman et al., 2004; Gerrity et al.. 1988). Also, O3 exposure led to a
10                   reflex mediated increase in fB and reduction in VT, hypothesized to be protective by
11                   decreasing the dose delivered to the lung at a particular MV (Gerrity et al.. 1994). Nasal
12                   O3 uptake was inversely proportional to flow rate (Santiago et al.. 2001). so that an
13                   increase in MV will increase O3 delivery to the lower airways. At a fixed MV, increasing
14                   VT (corresponding to decreasing fB) drove O3 deeper into the lungs and increased total
15                   respiratory uptake efficiency (Figure 5-4) (Ultman et al., 2004; Wiester et al., 1996c;
16                   Gerrity et al.. 1988). Modeling also  predicted a decrease in fractional uptake with
17                   increased fB when VT was held constant, but an increase in PAR dose with increased fB
18                   (Overton et al.. 1996). Similarly, increased fB (80 - 160 bpm) and shallow breathing in
19                   rats decreased midlevel tracheal 18O content and an increased 18O content in the mainstem
20                   bronchi (Alfaro et al.. 2004). This dependence may be a result of frequency-induced
21                   alterations in contact time that affects the first-order absorption rate for O3 (Postlethwait
22                   etal.. 1994). Also, an association of O3 uptake efficiency was found with MV and
23                   exposure time.

24                   Increasing flow leads to deeper penetration of O3 into the lung, such that a smaller
25                   fraction of O3 is absorbed in the URT and uptake shifts to the TB airways and respiratory
26                   airspaces (Nodelman and Ultman. 1999; Hu et al.. 1994; Ultman et al.. 1994). Hu et al.
27                   (1994) and Ultman et al. (1994) found that O3 absorption increased with volumetric
28                   penetration (Vp) of a bolus of O3 into the RT (Figure 5-5). Ozone uptake efficiency and
29                   Vp were not affected by bolus O3 concentration (Kabel et al.. 1994; Hu et al.. 1992).
30                   indicating that O3 uptake is a linear absorption process, where the diffusion and chemical
31                   reaction rates of O3 are proportional to the O3 concentration. This relationship was also
32                   true for nasal cavity uptake (Santiago et al..  2001). Rigas et al.  (2000) found a weak but
33                   significant negative dependence of O3 concentration on uptake efficiency in exercising
34                   individuals; however, only due to large changes in O3 concentration suggesting that O3
35                   uptake is likely still essentially linear with respect to O3 concentration. This study also
36                   found that exposure time had a small but significant influence on uptake efficiency;
37                   however, this negative dependence may be an artifact of progressive depletion of reactive
38                   substrates from the ELF.
      Draft - Do Not Cite or Quote                       5-10                                September 2011

-------
                               1.0 -,
        g  0.9 -
        S
        I
        a
        &
        
-------
 1
 2
with Re > 100. Additional hot spots were found during expiration on the parent branch
wall downstream of the branching region.
1.0-1

2
"• 0.5-
f
"
o.o -
v






JA VP60% V
• 4
• ^f
.*•"•

:: ;-
i
.*
D t • * *
-.%v





                               20   40   60   80   100  120 140 160  180 200
                                        Penetration Volume (ml_)

       Source: Adapted with permission of Health Effects Institute (Ultman et al.. 2004)

     Figure 5-5    Ozone uptake fraction as a function of volumetric penetration (Vp)
                    in a representative subject. Each point represents the O3 uptake of
                    a bolus inspired through a mouthpiece by the subject. The
                    volumes, VUA and VD, are the volume of the upper airways and
                    anatomical dead space, respectively, and VP50% is the Vp at which
                    50% of the inspired bolus was absorbed. In 47 healthy subjects,
                    Ultman et al. (2004) found that VP50% was well correlated with VD
                    and better correlated with the volume of the conducting airways,
                    i.e., VD minus VUA.
 4
 5
 6
 1
Overall O3 inhalation uptake in humans is over 80% efficient, but the exact efficiency
that determines how much O3 is available at longitudinally distributed compartments in
the lung is sensitive to changes in VT, fB, and to a minor extent, exposure time.
Decreased fB at a fixed penetration volume will shift the O3 uptake from the upper
airways to the central airways and respiratory airspaces.
 9
10
5.2.2.5   Mode of Breathing

Ozone uptake and distribution is sensitive to the mode of breathing. Variability in TB
airways volume had a weaker influence on O3 absorption during nasal breathing
compared to oral breathing. This could be a result of O3 scrubbing in the nasal
     Draft - Do Not Cite or Quote
                            5-12
September 2011

-------
 1                   passageways that are bypassed by oral breathing. Studies by Ultman and colleagues using
 2                   bolus inhalation demonstrated that O3 uptake fraction was greater during nasal breathing
 3                   than during oral breathing at each Vp (e.g. 0.90 during nasal breathing and 0.80 during
 4                   oral breathing at 150 mL/s and 0.45 during nasal breathing and 0.25 during oral breathing
 5                   at 1,000 mL/s) (Nodelman and Ultman. 1999; Kabeletal.. 1994; Ultman et al.. 1994).
 6                   Therefore, oral breathing results in deeper penetration of O3 into the RT with a higher
 7                   absorbed fraction in the URT, TB, and alveolar airways (NodeIman and Ultman. 1999).
 8                   Similar results were obtained from O3 uptake studies in dogs (Yokoyama and Frank.
 9                   1972). Earlier human studies suggesedt that oral or oronasal breathing results in a higher
10                   O3 uptake  efficiency than nasal breathing ("Wiester et al.. 1996c: Gerrity et al.. 1988):
11                   however the difference observed between inspired O3 taken up during oral versus nasal
12                   breathing may not be biologically significant. These human studies measured total RT
13                   absorption after continuous O3 exposure using a pharyngeal sampling tube, which may
14                   decrease sensitivity and lead to measurement errors. Overall, the mode of breathing may
15                   have little effect of the RT uptake efficiency, but does play an important role in the
16                   distribution of O3 deposited in the distal airways.
                     5.2.2.6    Interindividual Variability in Dose

17                   Similarly exposed individuals vary in the amount of actual dose delivered to the LRT
18                   (Santiago etal., 2001; Rigas et al., 2000; BushetaL 1996). Interindividual variability
19                   accounted for between 10-50% of the absolute variability in O3 uptake measurements
20                   (Santiago et al.. 2001; Rigas et al.. 2000). When concentration, time, and MV were held
21                   constant, fractional absorption ranged from 0.80 to 0.91 (Rigas et al.. 2000). It has been
22                   hypothesized that interindividual variation in O3  induced response such as FEVi is the
23                   result of interindividual variation in delivered dose or regional O3 uptake among exposed
24                   individuals.

25                   Recent studies have reiterated the importance of intersubject variation in O3 uptake. The
26                   intersubject variability in nasal O3 uptake determined by Sawyer et al. (2007) ranged
27                   from 26.8 to 65.4% (pre- and post-exercise). A second study investigating the use of the
28                   CO2 expirogram to quantify pulmonary responses to O3 found that intersubject
29                   variability accounted for 50% of the overall variance in the study (Taylor et al.. 2006).

30                   Variability in local dose may be attributed to differences in the pulmonary physiology,
31                   anatomy, and biochemistry. Since the TB airways remove the majority of inhaled O3
32                   before it reaches the gas exchange region, the volume and surface area of the upper
33                   airways will influence O3 uptake. Models predicted that fractional O3 uptake and PAR
34                   dose (flux of O3 to the PAR surfaces divided by exposure concentration) increase with
      Draft - Do Not Cite or Quote                       5-13                                 September 2011

-------
 1                   decreasing TB volume and decreasing TB region expansion. On the contrary, alveolar
 2                   expansion had minimal effect on uptake efficiency as relatively little O3 reaches the
 3                   peripheral lung (Bushet al.. 2001; Overton et al.,  1996). Ozone uptake was virtually
 4                   complete by the time O3 reaches the alveolar spaces of the lung (Postlethwait et al..
 5                   1994). Experimental studies have found that differences in TB volumes may account for
 6                   75% of the variation in absorption between subjects (Hitman et al.. 2004). In support of
 7                   this concept, regression analysis showed that O3 absorption was positively correlated
 8                   with anatomical dead space (VD) and TB volume  (i.e., VD minus VURT), but not total lung
 9                   capacity (TLC), forced vital capacity (FVC), or functional residual capacity (FRC)
10                   (Ulttnan et al.. 2004: Bushetal.. 1996: Huetal.. 1994: Postlethwait et al.. 1994).
11                   Variability in VD was correlated more with the variability in the TB volume than the URT
12                   volume.  Similarly, uptake was correlated with changes in individual bronchial cross-
13                   sectional area, indicating that changes in cross-sectional area available for gas diffusion
14                   are related to overall O3 retention (Reeser et al.. 2005: Ultman et al.. 2004). These studies
15                   provide support to the pulmonary physiology, especially the TB volume and surface area,
16                   playing a key role  in variability of O3 uptake between individuals.

17                   When absorption data were normalized to Vp/VD, variability attributed to gender
18                   differences were not distinguishable (Bush etal.. 1996). However, variability due to age
19                   has been predicted. Overton and Graham (1989) predicted that the total quantity of O3
20                   absorbed per minute increased with age from birth to adulthood. This model predicted
21                   that the LRT distribution of absorbed O3 and the CAR O3 tissue dose were not sensitive
22                   to age during quiet breathing. However, during heavy exercise or work O3 uptake was
23                   dependent on age.  A physiologically based pharmacokinetic model simulating O3 uptake
24                   predicted that regional extraction of O3 was relatively insensitive to age, but extraction
25                   per unit surface area was two- to eightfold higher  in infants compared to adults, due to
26                   the fact that children under age 5 have much a much smaller airway surface area in the
27                   extrathoracic (nasal) and alveolar regions (Sarangapani et al.. 2003).

28                   Smoking history, with its known increase in mucus production, was not found to
29                   significantly affect the fractional uptake of a bolus dose of O3 in apparently healthy
30                   smokers with limited smoking history (Bates etal.. 2009). Despite similar internal O3
31                   dose distribution, the smokers exhibited greater pulmonary responses to O3 bolus
32                   exposures, measured as FEVi decrements and increases in the normalized slope of the
33                   alveolar plateau (SN). This was contrary to previous studies conducted in smokers with a
34                   greater smoking history that found decreased O3 induced decrements in FEVi in smokers
35                   during continuous  O3 exposure (Frampton et al.. 1997b: Emmons and Foster. 1991).
      Draft - Do Not Cite or Quote                       5-14                                 September 2011

-------
                     5.2.2.7    Physical Activity

 1                   Exercise increases the overall exposure of the lung to inhaled contaminants due, in most
 2                   part, to the increased intake of air. As exercise increases from a low to moderate level, VT
 3                   increases. This increase in VT is achieved by encroaching upon both the inspiratory and
 4                   expiratory reserve volumes of the lung (Dempsey et al., 1990). After VT reaches about
 5                   50% of the vital capacity, generally during heavy exercise, further increases in ventilation
 6                   are achieved by increasing fB. Ventilatory demands of heavy exercise require airway flow
 7                   rates that often exceed 10 times resting levels and VT that approach 5 times resting levels
 8                   (Dempsev et al.. 2008).

 9                   This increase in VT and flow associated with exercise in humans shifts the O3 dose
10                   further into the periphery of the RT causing a disproportionate increase in distal lung
11                   dose. In addition to increasing the bulk transport of O3  into the lung, exercise also leads
12                   to a switch from nasal to oronasal breathing. Higher ventilatory demand necessitates a
13                   lower-resistance path through the mouth. Modeling heavy exercise by increasing
14                   ventilatory parameters from normal respiration levels predicted a 10-fold increase in total
15                   mass uptake  of O3 (Miller etal.. 1985). This model also predicted that as exercise and
16                   ventilatory demand increased the maximum tissue dose moved distally into the RT
17                   (Figure 5-6). By increasing flow to what is common in moderate exercise (respiratory
18                   flow = 750 -1,000 mL/s compared to 250 mL/s at rest), the URT absorbed a smaller
19                   fraction of the O3 (-0.50 at rest to 0.10 at exercise); however, the trachea and more distal
20                   TB airways received higher doses during exercise than rest (0.65 absorbed in the lower
21                   TB airways,  and 0.25 absorbed in the alveolar zone with exercise compared to 0.5 in the
22                   TB with almost no O3 reaching the alveolar zone at rest) (Hu etal.. 1994). The same shift
23                   in the O3 dose distribution more distally in the lung occurred in other studies mimicking
24                   the effects of exercise (Nodehnan and Ultman, 1999). Also, LRT uptake efficiency was
25                   sensitive to age only under exercise conditions (Overton and Graham. 1989). The total
26                   quantity of O3 absorbed per minute was predicted to  increase with age during heavy work
27                   or exercise. A recent study by Sawyer et al. (2007) showed that doubling minute
28                   ventilation led to only a 1.6-fold higher delivered dose  rate of O3 to the lung. Past models
29                   have predicted the increase in uptake during exercise is distributed unevenly in the RT
30                   compartments and regions. Tissue and mucus layer dose in the TB region increased ~1.4-
31                   fold during heavy exercise compared to resting conditions, whereas the alveolar region
32                   surfactant and tissue uptake increased by factors of 5.2 and 13.6, respectively (Miller et
33                   al.. 1985).
      Draft - Do Not Cite or Quote                       5-15                                September 2011

-------
                                       4    8    12   16   20
                                       AIRWAY GENERATION |Z)
                                          -TB-
 Source: Reprinted with permission. (Miller et al.. 1985)

Figure 5-6     Modeled effect of exercise on tissue dose of the LRT. Curve 1: VT =
                500 ml_; fB = 15 breaths/min. Curve 2: VT = 1,000 ml_; fB = 15
                breaths/min. Curve 3: VT = 1,750 ml_; fB = 20.3 breaths/min. Curve 4:
                VT = 2,250 ml_; fB = 30 breaths/min. TB = tracheobronchial region; P
                = pulmonary region.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
              5.2.2.8   Summary

              In summary, O3 uptake is affected by complex interactions between a number of factors
              including RT morphology, breathing route, frequency, and volume, physicochemical
              properties of the gas, physical processes of gas transport, as well as the physical and
              chemical properties of the ELF and tissue layers. The role of these processes varies
              throughout the length of the RT and as O3 moves from the gas into liquid compartments
              of the RT. The primary uptake site of O3 delivery to the lung epithelium is believed to be
              the CAR, however inhomogeneity in the RT structure may affect the dose delivered to
              this target site with larger path lengths leading to smaller locally delivered doses. Recent
              studies have provided evidence for hot spots of O3 flux around bifurcations in airways.
              Experimental studies and models have suggested that the net O3 dose gradually decreases
              distally from the trachea toward the end of the TB region and then rapidly decreases in
              the alveolar region. However, the tissue O3 dose is low in the trachea, increases to a
Draft - Do Not Cite or Quote
                                                 5-16
September 2011

-------
 1                   maximum in the terminal bronchioles and the CAR, and then rapidly decreases distally
 2                   into the alveolar region.

 3                   O3 uptake efficiency is sensitive to a number of factors. Fractional absorption will
 4                   decrease with increased flow and increase proportional to VT, so that at a fixed MV,
 5                   increasing VT (or decreasing fB) drives O3 deeper into the lungs and increases total
 6                   respiratory uptake efficiency. Individual total airway O3 uptake efficiency is also
 7                   sensitive to large changes in O3 concentration, exposure time, and MV. Major sources of
 8                   variability in absorption of O3 include O3 concentration, exposure time, fB, MV, and VT,
 9                   but the interindividual variation is the greatest source of variability uptake efficiency. The
10                   majority of this interindividual variability is due to differences in TB volume and surface
11                   area.

12                   An increase in VT and fB are both associated with increased physical activity.  These
13                   changes and a switch to oronasal breathing during exercise results in deeper penetration
14                   of O3 into the lung with a higher absorbed fraction in the ET, TB, and alveolar airways.
15                   For these reasons, increased physical activity acts to move the maximum tissue dose of
16                   O3 distally into the RT and into the alveolar region.
             5.2.3   Ozone Reactions and Reaction Products

17                   Ozone dose can be examined by the chemical reactions or the products of these reactions
18                   that result from O3 exposure. Since O3 is chemically reactive with a wide spectrum of
19                   biomolecules, it is not feasible to delineate its many reaction products. Measurements of
20                   reaction product formation have included either the loss of a specific molecule and
21                   appearance of plausible products, or the addition of O3 -derived oxygen to biomolecules
22                   through the use of oxygen-18 labeling. In vitro exposure of ELF showed that O3
23                   disappearance from the gas phase depends on the characteristics of the ELF substrates
24                   (Postlethwait et al.. 1998: Huetal.. 1994).

25                   For O3  to gain access to the underlying cellular compartments, O3 must dissolve at the
26                   air-liquid interface of the airway surface and travel through the ELF layer. The ELF is
27                   comprised of the airway surface lining that includes the periciliary sol layer and
28                   overlying mucus gel layer, and the alveolar surface lining that includes the subphase of
29                   liquid and vesicular surfactant and the continuous surfactant monolayer (Bastacky et al..
30                   1995). There is a progressive decrease in ELF thickness and increase in interfacial
31                   surface with progression from the large airways to the alveolus (Mercer et al.. 1992).
32                   Some cells, such as macrophages, may protrude into the gas phase, allowing for direct
33                   contact between O3 and cell membranes. The progressive thinning of the ELF while
34                   moving further down the RT decreases the radial distance O3 must travel to reach the


      Draft - Do Not Cite or Quote                       5-17                                 September 2011

-------
 1                   cellular tissue layer. A computational fluid dynamics model was able to predict
 2                   experimentally measured O3 uptake, but only with nasal mucus layer thickness
 3                   considered (Cohen-Hubal et al.. 1996). reaffirming the importance of the resistance
 4                   imparted by the ELF layer in dose and lesion patterns in the nasal passage.

 5                   Taking into account the high reactivity and low water solubility of O3, calculations
 6                   suggest that O3 will not penetrate ELF layers greater than 0.1 (im without being
 7                   transformed to other more long-lived reactive species, thus initiating a reaction cascade
 8                   (Pryor. 1992). However, the surfactant layer in the pulmonary region becomes ultrathin,
 9                   possibly allowing for direct interaction of O3 with the underlying epithelial cells. One
10                   study measured pulmonary liquid lining thickness over relatively flat portions of the
11                   alveolar wall to be 0.14 (im, to be 0.89 (im at the alveolar wall junctions, and 0.09 (im
12                   over the protruding features (Bastacky et al.. 1995). Still, the ELF should be considered
13                   an important target for O3 and the resulting secondary oxidation products should be
14                   considered key mediators of toxicity in the airways (role of reaction products in O3
15                   induced toxicity is discussed in Section 5.3). Model calculations of the nasal cavity based
16                   on diffusion equations and reaction rates of O3 with model substrates predict an O3
17                   penetration distance (0.5 (im) less than the thickness of the mucus layer (10 (im)
18                   (Santiago et al.. 2001). Experimental support for this concept comes from several studies
19                   which measured the total oxygen-addition product of O3 reactions in the airways through
20                   the use of oxygen-18 labeled O3. Fiigh concentrations of O3 reaction products were found
21                   in the bronchoalveolar lavage (BAL) mucus and surfactant providing evidence that O3
22                   reacts at the air-liquid  interface. Thus, O3 may cause injury by direct reaction with
23                   constituents of the lining layer, with cells protruding from it and in some cases with cells
24                   underlying the lining fluid. The reaction cascade resulting from the interaction of O3 with
25                   ELF substrates acts to carry the oxidative burden deeper into the tissues.

26                   Ozone may interact with many of the components in the ELF including phospholipids,
27                   neutral lipids, free fatty acids, proteins, and low molecular weight antioxidants (Perez-
28                   Gil 2008; Uppuetal.. 1995). It was estimated that 88% of the O3 that does not come in
29                   contact with antioxidants will react with unsaturated fatty acids in the ELF including
30                   those contained within phospholipids or neutral lipids (Uppu et al.. 1995). Ozone reacts
31                   with the double bond of lipids such as unsaturated fatty acids, a large component of ELF,
32                   to form stable and less reactive ozonide, aldehyde, and hydroperoxide reaction products
33                   via chemical reactions such as the Criegee ozonolysis mechanism (Figure 5-7) (Pryor et
34                   al., 1991). Lipid ozonation products, such as the aldehydes hexanal, heptanal, and
35                   nonanal, have been recovered after O3 exposure in human BAL fluid (BALF), rat BALF,
36                   isolated rat lung, and in vitro systems (Frampton et al., 1999; Postlethwait et al., 1998;
37                   Pryor et al.. 1996). Nonanal has been suggested as a relatively specific biomarker for O3
38                   exposure since the monounsaturated fatty acid parent compound, oleic acid, does not
      Draft - Do Not Cite or Quote                       5-18                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
undergo autoxidation (Pryor et al.. 1996). Adducts of the aldehyde 4-hydroxynonenal
were found in human alveolar macrophages after O3 exposure (Hamilton et al.. 1998).
Polyunsaturated fatty acid (PUFA) reactions are limited by the availability of O3 since
lipids are so abundant in the ELF. Yields of O3-induced aldehydes were increased by the
decrease in other substrates such as ascorbic acid (AH2) (Postlethwait et al.,  1998). Free
radicals are also generated during O3-mediated oxidation reactions with PUFA (Pryor.
1994). These reactions are reduced by the presence of the lipid-soluble free radical
scavenger a-tocopherol (a-TOH) (Prvor. 1994: Fujitaetal.. 1987: Prvor. 1976). PUFA
reactions may not generate sufficient bioactive materials to account for acute cell injury,
however only modest amounts of products may be necessary to induce cytotoxicity
(Postlethwait and Ultman. 2001: Postlethwait et al.. 1998).
A

RHC =
PUFA
either in
the 	 >
absence
ofH2O


CH +


RHC'
\

03 	 »
ozone

CH—
1
- RHC —
1

CH— 	 > RHC = O — O + RHC = O
trioxolane carbonyl oxide
or in the
presence
ofH2O
Criegee ozonide
/OH
	 > RHC 	 > RHC = (
XOOH aldehyde
hydroxyhydroperoxy cpd.
aldehyde

3 + H2O2
hydrogen
peroxide
                                                                            Source: U.S. EPA (2QQ6b)
      Figure 5-7     Schematic overview of ozone interaction with PUFA in ELF and
                      lung cells. It should be noted that not all secondary reaction
                      products are shown.
12
13
14
15
16
17

18
19
20
21
Cholesterol is the most abundant neutral lipid in human ELF. Reaction of cholesterol with
O3 results in biologically active cholesterol products such as the oxysterols, (3-epoxide
and 6-oxo-3,5-diol (Murphy and Johnson. 2008: Pulfer etal.. 2005: Pulfer and Murphy.
2004). Product yields will depend on ozonolysis conditions, however cholesterol
ozonolysis products were formed in similar abundance to phospholipid-derived
ozonolysis products in rat ELF (Pulfer and Murphy. 2004).

The ELF also contains proteins present in blood plasma as well as proteins secreted by
surface epithelial cells. Ozone reactions with proteins have been studied by their in vitro
reactions as well as reactions of their constituent amino acids (the most reactive of which
are cysteine, histidine, methionine, tyrosine, and tryptophan). Ozone has been shown to
      Draft - Do Not Cite or Quote
                               5-19
September 2011

-------
 1                   preferentially react with biomolecules in the following order: thiosulfate > ascorbate >
 2                   cysteine ~ methionine > glutathione (Kanofsky and Sima. 1995). Rate constants for the
 3                   reaction of amino acids with O3 vary between investigations due to differing reaction
 4                   conditions and assumptions; however aliphatic amino acids consistently were very slow
 5                   to react with O3 (e.g., alanine: 25-100 moles/L/sec) (Kanofsky and Sima. 1995;
 6                   Ignatenko and Cherenkevich. 1985; Pryor et al.. 1984; Hoigne and Bader. 1983). Uppu et
 7                   al. (1995) predicted that 12% of inhaled O3 that does not react with antioxidants will
 8                   react with proteins in the ELF, whereas 88% will react with PUFAs.

 9                   Reactions of ozone with low molecular weight antioxidants have been extensively
10                   studied. The consumption of antioxidants such as uric acid (UA), ascorbate (AH2), and
11                   reduced glutathione (GSH) by O3 was linear with time and positively correlated with
12                   initial substrate concentration and chamber O3 concentration (Mudway and Kelly. 1998;
13                   Mudway et al.. 1996). Endogenous antioxidants are present in relatively high
14                   concentrations in the ELF of the human TB airways and display high intrinsic reactivities
15                   toward O3, but do not possess equal O3  reactivity. In individual and in limited composite
16                   mixtures, UA was the most reactive antioxidant tested, followed by AH2 (Mudway and
17                   Kelly.  1998). GSH was consistently less reactive than UA or AH2 (Mudway and Kelly.
18                   1998; Mudway et al.. 1996; Kanofsky and Sima. 1995). To quantify these reactions,
19                   Kermani et al. (2006) recently evaluated the interfacial exposure of aqueous solutions of
20                   UA, AH2, and GSH (50-200 (iM) with O3 (1-5 ppm). Similar to the results of Mudway
21                   and Kelly (1998). this study found the hierarchy in reactivity between O3 and these
22                   antioxidants to be UA>AH2»GSH. UA and AH2 shared a 1:1 stoichiometry with O3,
23                   whereas 2.5 moles of GSH were consumed per mole of O3. Using these stoichiometries,
24                   reaction rate constants were derived (S.SxlO4]^"1 sec"1, S.Sx^M"1 sec"1, and 57.5 M"
25                   ° 75/sec [20.9 M"1 sec"1] for the reaction of O3 with UA, AH2, and GSH, respectively).
26                   These values are similar to those derived from data presented in Mudway and Kelly
27                   (1998). Other studies reported reactive rate constants that are two to three orders of
28                   magnitude larger, however these studies used higher concentrations of O3 and
29                   antioxidants under less physiologically relevant experimental conditions (Kanofsky and
30                   Sima. 1995: Giamalva et al.. 1985: Pryor etal.. 1984).

31                   A series of studies used new techniques  to investigate the reaction products resulting
32                   from initial  air-liquid interface interactions of O3 with ELF components (e.g.,
33                   antioxidants and proteins) in ~1 millisecond (Enami et al.. 2009a. b, c, 2008a. b).
34                   Solutions of aqueous UA, AH2, GSH, a-TOH, and protein cysteines (CyS) were sprayed
35                   as microdroplets in O3/N2  mixtures at atmospheric pressure and analyzed by electrospray
36                   mass spectrometry. These recent studies demonstrated different reactivity toward AH2,
37                   UA, and GSH by O3 when the large surface to volume ratio of microdroplets promote an
38                   interfacial reaction compared to previous studies using bulk liquid phase bioreactors, thus
      Draft - Do Not Cite or Quote                       5-20                                September 2011

-------
 1                   supporting the relevance of reactions between gas phase O3 and antioxidants found in the
 2                   ELF.

 3                   As was seen in previous studies (Kermani et al.. 2006; Kanofsky and Sima. 1995). the
 4                   hierarchy of reactivity of these ELF components with O3 was determined to be AH2 ~
 5                   UA > CyS > GSH. There was some variance between the reaction rates and product
 6                   formation of UA, AH2, and GSH with O3 as investigated by Enami et al. versus O3
 7                   reacting with bulk liquid phase bioreactors as described previously. UA was more
 8                   reactive than AH2 toward O3 in previous studies, but in reactions with O3 with
 9                   microdroplets, these antioxidants had equivalent reactivity (Enami et al.. 2008b). As O3 is
10                   a kinetically slow one-electron acceptor but very reactive O-atom donor, products of the
11                   interaction of O3 with UA, AH2, GSH, CyS, and a-TOH result from addition of n O-
12                   atoms (« = 1-4). These products included epoxides (e.g., U-O"), peroxides (e.g. U-O2"),
13                   and ozonides (e.g., U-O3"). For instance, GSH was oxidized to sulfonates (GSO3 YGSO32
14                   ), not glutathione disulfide (GSSG) by O3 (Enami et al., 2009b). However, it is possible
15                   that other oxidative species are oxidizing GSH in vivo, since sulfonates are not detected
16                   in O3 exposed ELF whereas GSSG is. This is also supported by the fact  that O3 is much
17                   less reactive with GSH than other antioxidants, such that < 3% of O3 will be scavenged
18                   by GSH when in equimolar amounts with AH2 (Enami et al.. 2009b).

19                   Ozonolysis product yields and formation were affected by pH. Acidified conditions (pH ~
20                   3-4), such as those that may result from acidic particulate exposure or pathological
21                   conditions like asthma (pH ~ 6), decreased the scavenging ability of UA and GSH for O3;
22                   such that at low pH, the scavenging of O3 must be taken over by other antioxidants, such
23                   as AH2 (Enami et al.. 2009b. 2008b). Also, under acidic conditions (pH ~ 5), the
24                   ozonolysis products of AH2 shifted from the innocuous dehydroascorbic acid to the more
25                   persistent products, AH2 ozonide and threonic acid (Enami et al.. 2008a). It is possible
26                   that the acidification of the ELF by acidic copollutant exposure will increase the toxicity
27                   of O3 by preventing some antioxidant reactions and shifting the reaction products to more
28                   persistent compounds.

29                   In a red blood cell (RBC) based system, AH2 augmented the in vitro uptake of O3 by six
30                   fold, as computed by the mass balance across the exposure chamber (Ballinger et al..
31                   2005). However, estimated in vitro O3 uptake was not proportional to the production of
32                   O3-derived aldehydes  from exposing O3 to RBC membranes (Ballinger  et al.. 2005). In
33                   addition, O3 induced cell membrane oxidation which required interactions with AH2 and
34                   GSH, but not UA or the vitamin E analog Trolox. Further, aqueous phase reactions
35                   between O3 and bovine serum albumin did not result in membrane oxidation (Ballinger et
36                   al.. 2005). The presence of UA or bovine serum albumin protected against lipid and
37                   protein oxidation resulting from the reaction of O3  and AH2 (Ballinger et al.. 2005). This
      Draft - Do Not Cite or Quote                      5-21                                 September 2011

-------
 1                   study provided evidence that antioxidants may paradoxically facilitate O3-mediated
 2                   damage. This apparent contradiction should be viewed in terms of the concentration-
 3                   dependent role of the ELF antioxidants. Reactions between O3 and antioxidant species
 4                   exhibited a biphasic concentration response, with oxidation of protein and lipid occurring
 5                   at lower, but not higher, concentrations of antioxidant. In this way, endogenous reactants
 6                   led to the formation of secondary oxidation products which were injurious and also led to
 7                   quenching reactions which were protective. Moreover, the formation of secondary
 8                   oxidation products mediated by some antioxidants was opposed by quenching reactions
 9                   involving other antioxidants.

10                   Alterations in ELF composition can result in alterations in O3 uptake. Bolus O3 uptake in
11                   human subjects can be decreased by previous continuous O3 exposure (120-360 ppb),
12                   possibly due to depletion of compounds able to react with O3 (Rigas etal.. 1997;
13                   Asplund et al.. 1996). Conversely, O3 (360 ppb) bolus uptake was increased with prior
14                   NO2 (360-720 ppb) or SO2  (360 ppb) exposure (Rigas etal.. 1997). It was hypothesized
15                   that this increased fractional absorption of O3 could be due to increased production of
16                   reactive substrates in the ELF due to oxidant-induced airway inflammation.

17                   Besides AH2, GSH and UA, the ELF contains numerous antioxidant substances that
18                   appear to be an important cellular defense against O3  including a-TOH, albumin,
19                   ceruloplasmin, lactoferrin, mucins, and transferrin (Mudway et al.. 2006; Freed et al..
20                   1999). The level and type of antioxidant present in ELF varies  between species, regions
21                   of the RT, and can be altered by O3 exposure. Mechanisms underlying the regional
22                   variability are  not well-understood. It is thought that both plasma ultrafiltrate and locally
23                   secreted substances contribute to the antioxidant content of the ELF (Mudway et al..
24                   2006; Freed etal..  1999). In the case of UA, the major source appears to be the plasma
25                   (Peden etal.. 1995). Repletion of UA in nasal lavage fluid was demonstrated during
26                   sequential nasal lavage in human subjects (Mudway et al.. 1999a). When these subjects
27                   were exposed to 200 ppb O3 for 2 hours while exercising, nasal lavage fluid UA was
28                   significantly decreased while plasma UA levels were  significantly increased (Mudway et
29                   al.. 1999a). The finding that UA, but not AH2 or GSH, was depleted in nasal lavage fluid
30                   indicated that UA was the predominant antioxidant with respect to O3 reactivity in the
31                   nasal cavity (Mudway et al.. 1999a). In addition, concentrations of UA were increased by
32                   cholinergic stimulation of the  airways in exercising human subjects exposed to 400 ppb
33                   O3 for 2 hours, which suggested that increased mucosal gland secretions were an
34                   important source (Peden etal.. 1995). Using the O3-specific antioxidant capacity assay on
35                   human nasal lavage samples, Rutkowski et al. (2011)  concluded that about 30% of the
36                   antioxidant capacity of the nasal liquid lining layer was attributed to UA activity. This
37                   assay predicted that more than 50% of the subject-to-subject differences in antioxidant
38                   capacity were driven by differences in UA concentration. However, day-to-day within-
      Draft - Do Not Cite or Quote                       5-22                                September 2011

-------
 1                   subject variations in measured antioxidant capacity were not related to the corresponding
 2                   variations in UA concentration in the nasal lavage fluid. Efforts to identify the
 3                   predominant antioxidant(s) in other RT regions besides the nasal cavity have failed to
 4                   yield definitive results. However, in human BALF samples, the mean consumption of
 5                   AH2 was greater than UA (Mudway et al.. 1996).

 6                   Regulation of AH2, GSH and a-TOH concentrations within the ELF is less clear than that
 7                   of UA (Mudway et al.. 2006). In a sequential nasal lavage study in humans, wash-out of
 8                   AH2 and GSH occurred, indicating the absence of rapidly acting repletion mechanisms
 9                   (Mudway et al.. 1999a). Other studies demonstrated increases in BALF GSH and
10                   decreases in BALF and plasma AH2 levels several hours following O3 exposure (200 ppb
11                   for 2 h, while exercising) (Mudway et al.. 2001; Blomberg et al..  1999; Mudway et al..
12                   1999b). Furthermore, high levels of dehydroascorbate, the oxidized form of AH2, have
13                   been reported in human ELF (Mudway et al.. 2006).  Other investigators have
14                   demonstrated cellular uptake of oxidized AH2 by several cell types leading to
15                   intracellular reduction and export of reduced AH2 (Welch etal.. 1995). Studies with rats
16                   exposed to 0.4-1.1 ppm O3 for 1-6 hours have shown consumption of AH2 that correlates
17                   with O3  exposure (Gunnison and Hatch. 1999; Gunnison et al.. 1996; Vincent et al..
18                   1996a).

19                   ELF exists as a complex mixture, thus it is important to look at O3 reactivity in substrate
20                   mixtures. Individual antioxidant consumption rates decreased as the substrate mixture
21                   complexity increased (e.g., antioxidant mixtures and albumin addition) (Mudway and
22                   Kelly. 1998). However, O3 reactions with AH2 predominated over the reaction with
23                   lipids, when exposed to substrate solution mixtures (Postlethwait et al.. 1998). It was
24                   suggested that O3 may react with other substrates once AH2 concentrations within the
25                   reaction  plane fall sufficiently. Additionally, once AH2 was consumed, the absorption
26                   efficiency diminished, allowing inhaled  O3 to be distributed to more distal airways
27                   (Postlethwait et al.. 1998). Multiple studies have concluded O3 is more reactive with AH2
28                   and UA than with the weakly reacting GSH (or cysteine or methionine) or with amino
29                   acid residues and protein thiols (Kanofsky and Sima. 1995; Cross  et al.. 1992).

30                   In addition to reactions with components of the ELF, O3 may react with plasma
31                   membranes of cells which reside in  the RT. Eicosanoids are an important class of
32                   secondary oxidation products which may be formed rapidly by this mechanism.
33                   Eicosanoids are metabolites of arachidonic acid, a 20-carbon PUFA, which is released
34                   from membrane phospholipids by phospholipase A2-mediated catalysis. Activation of
35                   phospholipase A2 occurs by several cell signaling pathways and may be triggered by O3-
3 6                   mediated lipid peroxidation of cellular membranes (Rashba-Step et al.. 1997).
37                   Additionally, cellular phospholipases A2, C and D may be activated by lipid ozonation
      Draft - Do Not Cite or Quote                      5-23                                September 2011

-------
 1                   products (Kafoury et al., 1998). While the conversion of arachidonic acid to
 2                   prostaglandins, leukotrienes and other eicosanoid products is generally catalyzed by
 3                   cyclooxygenases and lipoxygenases, non-enzymatic reactions also occur during oxidative
 4                   stress leading to the generation of a wide variety of eicosanoids and reactive oxygen
 5                   species. Further, the release of arachidonic acid from phospholipids is accompanied by
 6                   the formation of lysophospholipids which are precursors for platelet activating factors
 7                   (PAFs). Thus, formation of eicosanoids, reactive oxygen species and PAFs accompanies
 8                   Os-mediated lipid peroxidation.
                     5.2.3.1    Summary

 9                   The ELF is a complex mixture of lipids, proteins, and antioxidants that serve as the first
10                   barrier and target for inhaled O3 (Figure 5-8). The thickness of the lining fluid and mucus
11                   layer is an important determinant of the dose of O3 to the tissues. The antioxidant
12                   substances present in the ELF appear in most cases to limit interaction of O3 with
      Draft - Do Not Cite or Quote                       5-24                                September 2011

-------
                                                     Ozone
Mucus Layer
Mudns
Phospholipids
Liquid Layer Antioxidants
Uric Acid, Ascofbate,
Giutathione, a-To-copherol
ELF Macromolecules
Surfactant components
e.g. proteins, phospholipids/
cholesterol, CCSP,
Albumin, Hyaluronan, SP-A
Cellular Macromolecules
Plasma membrane proteins
and phosphotipids
Free fatty acids and
carbohydrates
                     Mechanisms for Antioxidant
                     Repletion
                     • Secretion by epithelial ceHs
                     • Transport from plasma
                     • Reduction of oxidized ascorbate
                   V	J
                                              Secondary Oxidation Products
                              Oxidized proteins
                                Aldehydes
                          Ozonized cholesterol species
                               Lipid Peroxides
                             Eicosanoids and PAF
                            Hyaluronan Fragments
                                                   Cellular injury
                                                  Cellular signaling
Mechanisms for Reaction
Product Removal
* Quenching reactions by ELF
 ant ioxidant sand proteins
* Non-enzymstic reactions with
 cellular aotioxidaots
• Metabolism by cellular GST/WQOl
• Receptor-mediated uptake by
 macrophsges           J
        Contents of this figure not discussed in Section 5.2 will be discussed in Section 5.3. Clara cell secretory protein, CCSP; Surfactant
      Protein-A, SP-A; Platelet activating factor, PAF.

      Figure 5-8     Details of the Os  interaction with the airway ELF to form secondary
                        oxidation products. Ozone will react with components of the ELF to
                        produce reaction products that  may lead to cellular injury and cell
                        signaling as discussed in Section 5.3.
 1
 2
 3
 4
 5
 6
 1
 8
 9
10
11
underlying tissues and to prevent penetration of O3 deeper into the lung. The formation of
secondary oxidation products is likely related to the concentration of antioxidants present
and the quenching ability of the lining fluid. Mechanisms are present to replenish the
antioxidant substrate pools as well as to remove secondary reaction products from tissue
interactions. Important differences exist in the reaction rates for O3 and these ELF
biomolecules and the reactivity of the resulting products. Overall, studies suggest that UA
and AH2 are more reactive with O3 than GSH, proteins,  or lipids. In addition to contri-
buting to the driving force for O3 uptake, formation of secondary oxidation products may
lead to increased cellular injury and cell signaling (discussed in Section 5.3). Studies
indicate that the antioxidants might be participating in reactions where the resulting
secondary oxidation products might penetrate into the tissue layer and cause injury.
      Draft - Do Not Cite or Quote
                                 5-25
                   September 2011

-------
          5.3    Possible Pathways/Modes of Action
            5.3.1    Introduction

 1                   Mode of action refers to a sequence of key events and processes which result in a given
 2                   toxic effect (U.S. EPA. 2005). Elucidation of mechanisms provides a more detailed
 3                   understanding of these key events and processes (U.S. EPA. 2005). Moreover, toxicity
 4                   pathways describe the processes by which perturbation of normal biological processes
 5                   produce changes sufficient to lead to cell injury and subsequent events such as adverse
 6                   health effects (U.S. EPA. 2009f). The purpose of this section of Chapter 5 is to describe
 7                   the key events and toxicity pathways which contribute to health effects resulting from
 8                   short-term and long-term exposures to O3. The extensive research carried out over
 9                   several decades in humans and in laboratory animals has yielded numerous studies on
10                   mechanisms by which O3 exerts its effects. This section will discuss some of the
11                   representative studies with particular emphasis on studies published since the 2006 O3
12                   AQCD and on studies in humans which inform biological mechanisms underlying
13                   responses to O3.

14                   It is well-appreciated that secondary oxidation products, which are formed as a result of
15                   O3 exposure, initiate numerous responses at the cellular, tissue and whole organ level of
16                   the respiratory system. These responses include the activation of neural reflexes,
17                   initiation of inflammation, alteration of epithelial barrier function, sensitization of
18                   bronchial smooth muscle, modification of innate/adaptive immunity and airways
19                   remodeling, as will be discussed below. Exposure to O3 also may result in effects on
20                   other organ systems such as the cardiovascular, central nervous, hepatic and reproductive
21                   systems. It is unlikely  that lipid ozonides and other secondary oxidation products, which
22                   are bioactive  and cytotoxic in the respiratory system, gain access to the vascular space
23                   (Chuang et al.. 2009).  However the inhalation of O3 may result in systemic oxidative
24                   stress. The following subsections describe the  current understanding of potential
25                   pathways and modes of action responsible for the pulmonary and extrapulmonary effects
26                   of O3 exposure.
            5.3.2  Activation of Neural Reflexes

27                  Acute O3 exposure results in reversible effects on lung function parameters through
28                  activation of neural reflexes. The involvement of bronchial C-fibers, a type of nociceptive
29                  sensory nerve, has been demonstrated in dogs exposed through an endotracheal tube to 2-
30                  3 ppm O3 for 20-70 minutes (Coleridge et al.. 1993; Schelegle et al.. 1993). This vagal
      Draft - Do Not Cite or Quote                      5-26                                September 2011

-------
 1                  afferent pathway was found to be responsible for O3-mediated rapid shallow breathing
 2                  and other changes in respiratory mechanics in O3-exposed dogs (Schelegle etal.. 1993).
 3                  Ozone also triggers neural reflexes which stimulate the autonomic nervous system and
 4                  alter electrophysiologic responses of the heart. For example, bradycardia, altered HRV
 5                  and arrhythmia have been demonstrated in rodents exposed to 0.1-0.6 ppm O3 (Hamade
 6                  and Tankersley. 2009; Watkinson et al.. 2001; Aritoetal.. 1990). Another effect is
 7                  hypothermia, which in rodents occurred subsequent to the activation of neural reflexes
 8                  involoving the parasympathetic nervous system (Watkinson et al.. 2001). Vagal afferent
 9                  pathways originating in the respiratory tract may also be responsible for O3-mediated
10                  activation of nucleus tractus solitarius neurons which resulted in neuronal activation in
11                  stress-responsive regions of the central nervous system (CNS) (rats, 0.5-2.0 ppm O3 for
12                  1.5-120hours) (Gackiere etal.. 2011).

13                  Recent studies in animals provide new information regarding the  effects of O3 on reflex
14                  responses mediated by bronchopulmonary C-fibers. In ex vivo mouse lungs, O3 exposure
15                  selectively activated a subset of C-fiber receptors which are TRPA1 ion channels (Taylor-
16                  Clark and Undem.  2010). TRPA1 ion channels are members of the TRP family of ion
17                  channels, which are known to mediate the responses of sensory neurons to inflammatory
18                  mediators (Caceres et al.. 2009). In addition to TRPA1 ion channels possibly playing a
19                  key role in O3-induced decrements in pulmonary function, they may mediate allergic
20                  asthma (Caceres et al.. 2009). Activation of TRPA1 ion channels following O3 exposure
21                  is likely initiated by secondary oxidation products such as aldehydes and prostaglandins
22                  (Taylor-Clark and Undem. 2010) through covalent modification of cysteine and lysine
23                  residues (Trevisani et al.. 2007). Ozonation of unsaturated fatty acids in the ELF was
24                  found to result in the generation of aldehydes (Frampton et al.. 1999) such as
25                  4-hydroxynonenal  and 4-oxononenal (Taylor-Clark et al.. 2008; Trevisani et al.. 2007). 4-
26                  oxononenal is a stronger electrophile than 4-hydroxynonenal and exhibits greater potency
27                  towards the TRPA1 channels (Taylor-Clark et al.. 2008). (Trevisani  etal.. 2007). In
28                  addition, PGE2 is known to sensitize TRPA1 channels (Bang et al.. 2007).

29                  In exercising humans, the response to  O3 (500 ppb for 2 h) was characterized by
30                  substernal discomfort, especially on deep inspiration, accompanied by involuntary
31                  truncation of inspiration (Hazucha et al.. 1989). This led to decreased inspiratory capacity
32                  and to decreased forced vital capacity  (FVC) and forced expiratory volume in one second
33                  (FEVi), as measured by spirometry. These changes, which occurred during O3  exposure,
34                  were accompanied by decreased VT and increased respiratory frequency in human
35                  subjects. Spirometric changes in FEVi and FVC were not due to changes in respiratory
36                  muscle strength (Hazucha et al.. 1989). In addition, parasympathetic involvement in the
37                  O3-mediated decreases in lung volume was minimal (Mudway and Kelly. 2000). since
3 8                  changes in FVC or symptoms were not modified by treatment with bronchodilators such
      Draft - Do Not Cite or Quote                      5-27                                September 2011

-------
 1                   as atropine in exercising human subjects exposed to 400 ppb O3 for 0.5 hour (Beckett et
 2                   al.. 1985). However, the loss of vital capacity was reversible with intravenous
 3                   administration of the rapid-acting opioid agonist, sufentanyl, in exercising human
 4                   subjects exposed to 420 ppb O3 for 2 hours, which indicated the involvement of opioid
 5                   receptor-containing nerve fibers and/or more central neurons (Passannante et al., 1998).
 6                   The effects of sufentanyl may be attributed to blocking C-fiber stimulation by O3 since
 7                   activation of opioid receptors downregulated C-fiber function (Belvisi et al.. 1992). Thus,
 8                   nociceptive sensory nerves, presumably bronchial C-fibers, are responsible for O3-
 9                   mediated responses in humans (Passannante et al., 1998). This vagal afferent pathway  is
10                   responsible for pain-related symptoms and inhibition of maximal inspiration in humans
11                   (Hazucha et al.. 1989).

12                   There is some evidence that eicosanoids (see Section 5.3.3) play a role in the neural
13                   reflex since cyclooxygenase inhibition with indomethacin (Alexis et al.. 2000; Schelegle
14                   et al.. 1987) or ibuprofen, which also blocks some lipoxygenase activity (Hazucha et al.,
15                   1996). before exposure to O3 significantly blunted the spirometric responses. These
16                   studies involved exposures of 1-2 hours to 350-400 ppb O3 in exercising human subjects.
17                   In the latter study, ibuprofen treatment resulted in measurable decreases in BALF levels
18                   of PGE2 and TXB2 at 1-hour postexposure (Hazucha et al.. 1996). Although an earlier
19                   study demonstrated that PGE2 stimulated bronchial C-fibers (Coleridge et al.. 1993;
20                   Coleridge etal.. 1976) and suggested that PGE2 mediated O3-induced decreases in
21                   pulmonary function, no correlation was observed between the degree of ibuprofen-
22                   induced inhibition of BALF PGE2 levels and blunting of the spirometric response to O3
23                   (Hazucha et al.. 1996). These results point to the involvement of a lipoxygenase product.
24                   Further, as noted above, PGE2 may play a role  in the neural reflex by sensitizing TRPA1
25                   channels. A recent study in exercising human subjects exposed for 1 hour to 350 ppb O3
26                   also provided evidence that arachidonic acid metabolites, as well as oxidative stress,
27                   contribute to human responsiveness to O3  (Alfaro et al.. 2007).

28                   In addition to the spirometric changes, mild airways obstruction occurred in exercising
29                   humans during O3 exposure (500 ppb for 2 hours) (Hazucha et al.. 1989). This pulmonary
30                   function decrement is generally measured  as specific airway resistance (sRaw) which is
31                   the product of airway resistance and thoracic gas volume. In several studies involving
32                   exercising human subjects exposed for 1-4 hours to 200-300 ppb O3, changes in sRaw
33                   correlated with changes in inflammatory and injury endpoints measured 18-hours
34                   postexposure, but did not follow the same  time course or change to the same degree as
35                   spirometric changes (i.e. FEVi, FVC) measured during exposure (Balmes etal.. 1996;
36                   Ariset al.. 1993; Schelegle et al.. 1991). In addition, a small but persistent increase in
37                   airway resistance associated with narrowing of small peripheral airways (measured as
38                   changes in isovolumetric FEF25_75) was demonstrated in O3-exposed human subjects (350
      Draft - Do Not Cite or Quote                       5-28                                September 2011

-------
 1                   ppb for 130 minutes with exercise) (Weinmann et al.. 1995a; Weinmann et al.. 1995b). A
 2                   similar study (400 ppb O3 for 2 hours in exercising human subjects) found decreases in
 3                   FEF 25.75 concomitant with increases in residual volume, which is suggestive of small
 4                   airways dysfunction (Kreit et al.. 1989). In separate studies, a statistically significant
 5                   increase in residual volume (500 ppb for 2 hours) (Hazucha et al.. 1989) and a
 6                   statistically significant decrease in FEF25-75 (160 ppb for 7.6 hours) (Horstman et al..
 7                   1995) were observed following O3 exposure in exercising human subjects, providing
 8                   further support for an O3-induced effect on small airways.

 9                   Mechanisms underlying this rapid increase in airway resistance following O3 exposure
10                   are incompletely understood. Pretreatment with atropine decreased baseline sRaw and
11                   prevented O3-induced increases in sRaw in exercising human subjects (400 ppb for 0.5
12                   hours) (Beckett et al.. 1985). indicating the involvement of muscarinic cholinergic
13                   receptors of the parasympathetic nervous system. Interestingly, atropine pretreatment
14                   partially blocked the decrease in FEVi, but had no  effect on the decrease in FVC,
15                   breathing rate, tidal volume or respiratory symptoms (Beckett et al.. 1985). Using a (3-
16                   adrenergic agonist, it was shown that smooth muscle contraction, not increased airway
17                   mucus secretion, was responsible for O3-induced increases in airway resistance (Beckett
18                   etal.. 1985). Thus, pulmonary function decrements measured as FEVi may reflect both
19                   restrictive (such as decreased inspiratory capacity)  and obstructive (such as
20                   bronchoconstriction) type changes in airway responses. This is consistent with
21                   McDonnell et al. (1983) who observed a relatively  strong correlation between sRaw and
22                   FEVi (r=-0.31, p=0.001) and a far weaker correlation between sRaw and FVC (r=-0.16,
23                   p=0.10) in exercising human subjects exposed for 2.5 hours to 120-400 ppb O3.

24                   Furthermore, tachykinins may contribute to O3-mediated increases in airway resistance.
25                   In addition to stimulating CNS reflexes, bronchopulmonary C-fibers mediate local axon
26                   responses by releasing neuropeptides such as substance P (SP), neurokinin (NK) A and
27                   calcitonin gene-related peptide (CGRP). Tachykinins bind to NK receptors resulting in
28                   responses such as bronchoconstriction. Recent studies in animals demonstrated that NK-1
29                   receptor blockade had no effect on O3-stimulated physiologic responses such as VT and
30                   fB in rats over the 8 hour exposure to 1 ppm O3 (Oslund et al., 2008). However, SP and
31                   NK receptors contributed to vagally-mediated bronchoconstriction in guinea pigs 3  days
32                   after a single 4-hour exposure to 2 ppm O3 (Verhein et al., 2011). In one human study in
33                   which bronchial biopsies were performed  and studied by immunohistochemistry, SP was
34                   substantially diminished in submucosal sensory nerves 6 hours following O3 exposure
35                   (200 ppb for 2 hours with exercise) (Krishna et al..  1997). A statistically significant
36                   correlation was observed between loss of SP immunoreactivity from neurons in the
37                   bronchial mucosa and changes in FEVi measured 1-hour postexposure (Krishna et al..
38                   1997). Another study found that SP was increased in lavage fluid of human subjects
      Draft - Do Not Cite or Quote                       5-29                                 September 2011

-------
 1                   immediately after O3 challenge (250 ppb for 1 hour with exercise) (Hazbun et al.. 1993).
 2                   These results provide evidence that the increased airway resistance observed following
 3                   O3 exposure is due to vagally-mediated responses and possibly by local axon reflex
 4                   responses through bronchopulmonary C-fiber-mediated release of SP.
            5.3.3   Initiation of inflammation

 5                   As described previously (5.2.3), O3 reacts with components of the ELF and cellular
 6                   membranes resulting in the generation of secondary oxidation products. Higher
 7                   concentrations of these products may directly injure respiratory tract epithelium. Lower
 8                   concentrations may initiate cellular responses including cytokine generation, adhesion
 9                   molecule expression and modification of tight junctions leading to inflammation and
10                   increased permeability across airway epithelium (Section 5.3.4) (Dahl et al.. 2007;
11                   Mudway and Kelly. 2000). Subsequent airways remodeling may also occur (Section
12                   5.3.7) (Mudwav and Kelly. 2000).

13                   An important hallmark of acute O3 exposure in humans and animals is neutrophilic
14                   airways inflammation. Although neutrophil influx into nasal airways has been
15                   demonstrated in exercising human subjects (400 ppb O3, 2 hours) (Graham and Koren.
16                   1990). most studies of neutrophil influx have focused on the lower airways (Hazucha et
17                   al.. 1996; Aris et al.. 1993). The time course of this response in the lower airways and its
18                   resolution was slower than that of the decrements in pulmonary  function in exercising
19                   human subjects exposed for 2 hours to 500 ppb O3 (Hazucha et al., 1996). In several
20                   studies, airways neutrophilia was observable within 1-2 hours, peaked at 4-6 hours and
21                   was returning to baseline levels at 24 hours following exposure of 1-2 hours to 300-400
22                   ppb O3 in exercising humans (Devlin et al.. 1991; Schelegle et al.. 1991). Since the influx
23                   and persistence of neutrophils in airways  following O3 exposure correlated with the
24                   temporal profile of epithelial injury (guinea pigs, 0.26-1 ppm O3 72 hours) (Hu et al..
25                   1982). neutrophils were probably injurious. However, neutrophils have also been shown
26                   to contribute to repair of O3-injured epithelium in rats exposed for 8 hours to 1 ppm O3
27                   possibly by removing necrotic epithelial cells (Mudway and Kelly. 2000; Vesely et al..
28                   1999). Nonetheless, the degree  of airways inflammation due to O3 is thought to have
29                   more important long-term consequences than the more quickly resolving changes in
30                   pulmonary function since airways inflammation is often accompanied by tissue injury
31                   (Balmes et al.. 1996).

32                   Ozone exposure results in alterations in other airways inflammatory cells besides
33                   neutrophils,  including lymphocytes, macrophages, monocytes and mast cells. Influx of
34                   some of these cells accounts for the later (i.e. 18-20 hours) phase of inflammation
      Draft - Do Not Cite or Quote                      5-30                                September 2011

-------
 1                  following O3 exposure. Numbers of lymphocytes and total cells in BALF were decreased
 2                  early after O3 exposure in exercising humans exposed for 2 hours to 200 ppb O3, which
 3                  preceded the neutrophil influx (Mudway and Kelly. 2000; Blomberg et al..  1999; Krishna
 4                  et al.. 1997). The decrease in total cells was thought to reflect decreases in macrophages,
 5                  although it was not clear whether the cells were necrotic or whether membrane adhesive
 6                  properties were altered making them more difficult to obtain by lavage (Mudway and
 7                  Kelly. 2000; Blomberg et al.. 1999; Mudwav et al.. 1999b; Frampton et al.. 1997a;
 8                  Pearson and Bhalla. 1997). A recent study in exercising human subjects exposed for 6.6
 9                  hours to 80 ppb O3 demonstrated an increase in numbers of sputum monocytes and
10                  dendritic-like cells with increased expression of innate immune surface proteins and
11                  antigen presentation markers (Peden. 2011; Alexis et al.. 2010) (see Section 6.2.3.1). An
12                  increase in submucosal mast cells was observed 1.5 hours after a 2 hour-exposure to 200
13                  ppb O3 (Blomberg et al., 1999) and an increase in BAL mast cell number was observed
14                  18 hours after a 4-hour exposure to 220 ppb O3 exposure in exercising human subjects
15                  (Frampton et al.. 1997a). Mast cells may play an important role in mediating neutrophil
16                  influx since they are an important source of several pro-inflammatory cytokines and since
17                  their influx preceded that of neutrophils in exercising human subjects exposed for 2 hours
18                  to 200 ppb O3 (Stenfors et al.. 2002; Blomberg et al.. 1999). Further, a study using mast
19                  cell-deficient mice demonstrated decreased neutrophilic inflammation in response to O3
20                  (1.75 ppm, 3 hours) compared with wild type mice (Kleeberger et al., 1993). Influx of
21                  these inflammatory cell types in the lung is indicative of O3-mediated activation of innate
22                  immunity as will be discussed in Section 5.3.6.

23                  Much is known about the cellular  and molecular signals involved in inflammatory
24                  responses to O3 exposure (U.S. EPA. 2006b). Eicosanoids are one class of secondary
25                  oxidation products which may be formed rapidly following O3 exposure and which may
26                  mediate inflammation. In addition, secondary reaction products may stimulate
27                  macrophages to produce cytokines such as IL-1, IL-6 and TNF-a which in turn activate
28                  IL-8 production by epithelial cells. Although IL-8 has been proposed to play a role in
29                  neutrophil chemotaxis, measurements of IL-8 in BALF from humans exposed to O3
30                  found increases that were too late to account for this effect (Mudway and Kelly. 2000).
31                  The time-course profiles of PGE2  and IL-6 responses suggest that they may play a role in
32                  neutrophil chemotaxis in humans (Mudway and Kelly. 2000). However, pretreatment
33                  with ibuprofen attenuated O3-induced increases in BALF PGE2 levels, but  had no effect
34                  on neutrophilia in exercising human subjects exposed for 2 hour to 400 ppb O3 (Hazucha
35                  etal.. 1996).

36                  One set of studies in humans focused on the earliest phase of airways inflammation (1-2
37                  hours following exposure). Exercising subjects were exposed to 200 ppb O3 for 2 hours
38                  and bronchial biopsy tissues were obtained 1.5 and 6 hours after exposure (Bosson et al..
      Draft - Do Not Cite or Quote                      5-31                                September 2011

-------
 1                   2009: Bosson et al.. 2003; Stenfors et al.. 2002; Blomberg et al.. 1999). Results
 2                   demonstrated upregulation of vascular endothelial adhesion molecules P-selectin and
 3                   ICAM-1 at both 1.5 and 6 hours (Stenfors et al.. 2002; Blomberg et al.. 1999).
 4                   Submucosal mast cell numbers were increased at 1.5 hours in the biopsy samples without
 5                   an accompanying increase in neutrophil number (Blomberg et al.. 1999). Pronounced
 6                   neutrophil infiltration was observed at 6 hours in the bronchial mucosa (Stenfors et al..
 7                   2002). Surprisingly, suppression of the NF-KB and AP-1 pathways at 1.5 hours and a lack
 8                   of increased IL-8 at 1.5 or 6 hours in bronchial epithelium was observed (Bosson et al..
 9                   2009). The authors suggested that vascular endothelial adhesion molecules, rather than
10                   redox sensitive transcription factors, play key roles in early neutrophil recruitment in
11                   response to O3.

12                   Increases in markers of inflammation occurred to a comparable degree in exercising
13                   human subjects with mild (least sensitive) and more remarkable (more sensitive)
14                   spirometric responses to O3 (200 ppb, 4 hours) (Balmes et al.. 1996). Two other studies
15                   using similar protocols (200 ppb for 4 hours and 300 ppb for 1 hour) found that acute
16                   spirometric changes were not positively correlated with cellular and biochemical
17                   indicators of inflammation (Aris et al.. 1993; Schelegle et al.. 1991). However
18                   inflammation was correlated with changes in sRaw (Balmes et al.. 1996). In another
19                   study, pretreatment with ibuprofen had no effect on neutrophilia although it blunted the
20                   spirometric response in exercising human subjects exposed for 2 hours to 400 ppb O3
21                   (Hazucha et al.. 1996). Taken together, results from these studies indicate different
22                   mechanisms underlying the spirometric and inflammatory responses to O3.

23                   A common mechanism underlying both inflammation and impaired pulmonary function
24                   was suggested by Krishna et al. (1997). This study, conducted in exercising humans
25                   exposed to 200 ppb O3 for 2 hours, demonstrated a correlation between loss of SP
26                   immunoreactivity from neurons in the bronchial mucosa and numbers of neutrophils and
27                   epithelial cells (shed epithelial cells are an index of injury) in the BALF 64iours
28                   postexposure. Furthermore, the loss of SP immunoreactivity was correlated with the
29                   observed changes in FEVi. Another study found that SP was increased in lavage fluid of
30                   exercising human subjects immediately after O3 challenge (250 ppb, 1 hour) (Hazbun et
31                   al.. 1993). SP is a neuropeptide released by sensory nerves which mediates neurogenic
32                   edema and bronchoconstriction (Krishna et al.. 1997). Taken together, these findings
33                   suggest that O3-mediated stimulation of sensory nerves which leads to activation of
34                   central and local axon reflexes is s a common effector pathway leading to impaired
35                   pulmonary function and inflammation.

36                   Studies in animal models have confirmed many of these findings and provided evidence
37                   for additional mechanisms involved in O3-induced inflammation. A study in mice (2 ppm
      Draft - Do Not Cite or Quote                      5-32                                September 2011

-------
 1                  O3, 3 hours) demonstrated that PAF may be important in neutrophil chemotaxis
 2                  (Longphre etal.. 1999). while ICAM-1 and macrophage inflammatory protein-2 (MIP-2),
 3                  the rodent IL-8 homologue, have been implicated in a rat model (1 ppm O3, 3 hours)
 4                  (Bhalla and Gupta. 2000). Key roles for CXCR2, a receptor for keratinocyte-derived
 5                  chemokine (KC) and MIP-2, and for IL-6 in O3-mediated neutrophil influx were
 6                  demonstrated in mice (1 ppm O3, 3 hours) (Johnston et al.. 2005a: Johnston et al..
 7                  2005b). Activation of JNK and p38 pathways and cathepsin-S were also found to be
 8                  important in this response (3 ppm O3, 3 hours) (Williams et al.. 2009a: Williams et al..
 9                  2008b; Williams et al.. 2007a). Matrix metalloproteinase-9 (MMP-9) protected against
10                  O3-induced airways inflammation and injury in mice (0.3 ppm O3, 6-72 hours) (Yoon et
11                  al.. 2007). Interleukin-10 (IL-10) was also found to be protective since IL-10 deficient
12                  mice responded to O3 exposure (0.3 ppm, 24-72 hours) with enhanced numbers of BAL
13                  neutrophils, enhanced NF-KB activation and MIP-2 levels compared with IL-10 sufficient
14                  mice (Backus etal.. 2010).

15                   In addition, lung epithelial cells may release ATP in response to O3 exposure (Ahmad et
16                  al.. 2005). ATP and its metabolites (catalyzed by ecto-enzymes) can bind to cellular
17                  purinergic receptors resulting in activation of cell signaling pathways (Picher et al..
18                  2004). One such metabolite, adenine, is capable of undergoing oxidation leading to the
19                  formation of UA which, if present in high concentrations, could activate inflammasomes
20                  and result in caspase 1 activation and the maturation and secretion of IL-1(3 and IL-18
21                  (Dostert et al.. 2008). A recent study in exercising human subjects exposed for 2 hours to
22                  400 ppb O3 demonstrated a correlation between ATP metabolites  and inflammatory
23                  markers (Esther et al.. 2011). which provides some support for this mechanism.

24                  Several recent studies have focused on the role of toll-like receptor (TLR) and its related
25                  adaptor protein MyD88 in mediating O3-induced neutrophilia. While Hollingsworth et al.
26                  (2004) demonstrated airways neutrophilia which was TLR4-independent following acute
27                  (2 ppm, 3 hours) and subchronic (0.3 ppm,  72 hours) O3 exposure in a mouse model,
28                  Williams et al. (2007b) found that MyD88 was important in mediating O3-induced
29                  neutrophilia in mice (3 ppm, 3 hours), with TLR4 and TLR2 contributing to the speed of
30                  the response. Moreover, MyD88, TLR2 and TLR4 contributed to inflammatory gene
31                  expression in this model and O3 upregulated MyD88, TLR4 and TLR4 gene expression
32                  (Williams et al.. 2007a)

33                  Hyaluronan was found to mediate a later phase (24 hours) of O3-induced inflammation in
34                  mice (Garantziotis et al.. 2010; Garantziotis et al.. 2009). Hyaluronan is an extracellular
3 5                  matrix component which is normally found in the ELF as a large polymer. Exposure to
36                  2 ppm O3 for 3 hours resulted in elevated levels of soluble low molecular weight
37                  hyaluronan in the BALF 24-hours postexposure (Garantziotis et al.. 2010; Garantziotis et
      Draft - Do Not Cite or Quote                      5-33                                September 2011

-------
 1                  al.. 2009). Ozone may have caused the depolymerization of hyaluronan to soluble
 2                  fragments which are known to be endogenous ligands of the CD44 receptor and TLR4 in
 3                  the macrophage (Jiang etal. 2005). Binding of hyaluronan fragments to the CD44
 4                  receptor activates hyaluronan clearance, while binding to TLR4 results in signaling
 5                  through MyD88 to produce chemokines that stimulate the influx of inflammatory cells
 6                  (Jiang et al.. 2005). Activation of NF-KB occurred in both airway epithelia and alveolar
 7                  macrophages 24-hours postexposure to O3. Increases in BALF pro-inflammatory factors
 8                  KC, IL-1|3, MCP-1, TNF-a and IL-6 observed 24 hours following O3 exposure were
 9                  found to be partially dependent on TLR4 (Garantziotis et al., 2010) while increases in
10                  BAL inflammatory cells, which consisted mainly of macrophages, were dependent on
11                  CD44 (Garantziotis et al.. 2009). BAL inflammatory cells number and injury markers
12                  following O3 exposure were similar in wild-type and TLR4-deficient animals
13                  (Garantziotis et al.. 2010).

14                  Since exposure to O3 leads to airways inflammation characterized by neutrophilia, and
15                  since neutrophil-derived oxidants often scavenge ELF antioxidants, concentrations of
16                  ELF antioxidants have been examined during airways neutrophilia (Long etal.. 2001;
17                  Gunnison and Hatch. 1999; Mudway et al.. 1999b). In exercising humans exposed to 200
18                  ppb O3 for 2 hours, UA, GSH and a-TOH levels remained unchanged in BALF 6-hours
19                  postexposure while AH2 was decreased significantly in both BALF and plasma (Mudway
20                  et al.. 1999b). A second study involving the same protocol reported a loss of AH2 from
21                  bronchial wash fluid and BALF, representing proximal and distal airway ELF
22                  respectively, as well as an increase in oxidized GSH in both compartments (Mudway et
23                  al.. 2001). No change was observed in ELF UA levels in response to O3 (Mudway et al..
24                  2001). Further, O3 exposure (0.8 ppm, 4 hours) in female rats resulted in a 50% decrease
25                  in BALF AH2 immediately postexposure (Gunnison and Hatch. 1999). These studies
26                  suggested a role for AH2 and GSH in protecting against the oxidative stress associated
27                  with inflammation.
            5.3.4   Alteration of epithelial barrier function

28                  Following O3 exposure, injury and inflammation can lead to altered airway barrier
29                  function. Histologic analysis has demonstrated damage to tight junctions between
30                  epithelial cells, suggesting an increase in epithelial permeability. In addition, the presence
31                  of shed epithelial cells in the BALF and increased epithelial permeability, which is
32                  measured as the flux of small solutes, have been observed and are indicative  of epithelial
33                  injury. Increases in vascular permeability, as measured by BALF protein and albumin,
34                  have also been demonstrated (Costa et al.. 1985; Hu et al.. 1982).
      Draft - Do Not Cite or Quote                       5-34                                September 2011

-------
 1                   An early study in sheep measured changes in airway permeability as the flux of inhaled
 2                   radiolabeled histamine into the plasma (Abraham et al.. 1984). Exposure of sheep to 0.5
 3                   ppm O3 for 2 hours via an endotracheal tube resulted in an increased rate of histamine
 4                   appearance in the plasma at 1 day postexposure. Subsequently, numerous studies have
 5                   measured epithelial permeability as the flux of the small solute 99mTcDTPA which was
 6                   introduced into the air spaces in different regions of the respiratory tract. Increased
 7                   pulmonary epithelial permeability, measured as the clearance of 99mTc-DTPA, was
 8                   demonstrated in humans 1-2 hours following  a 2-hour exposure to 400 ppb O3 while
 9                   exercising moderately (Kehrl et al., 1987). Another study in human subjects found
10                   increased epithelial permeability 19-hours postexposure to 240 ppb O3 for 130 minutes
11                   while exercising (Foster and Stetkiewicz.  1996). Increased bronchial permeability was
12                   also observed in dogs 1-day postexposure (0.4 ppm O3 by endotracheal tube for 6 hours)
13                   and did not resolve for several days (Foster and Freed. 1999).

14                   A role for tachykinins in mediating airway epithelial injury and decreased barrier function
15                   has been suggested. Nishiyama et al. (1998) demonstrated that capsaicin, which depletes
16                   nerve fibers of substance P, blocked the O3-induced increase in permeability of guinea
17                   pig tracheal mucosa (0.5-3 ppm O3, 0.5 hours). Pretreatment with propranolol or atropine
18                   failed to inhibit this response, suggesting that adrenergic and cholinergic pathways were
19                   not involved. In another study, tachykinins working through NK-1 and CGRP receptors
20                   were found to contribute to airway epithelial injury in O3-exposed rats (1 ppm, 8  hours)
21                   (Oslund et al.. 2009. 2008).

22                   Kleeberger et al. (2000) evaluated genetic susceptibility to O3-induced altered barrier
23                   function in  recombinant inbred strains of mice. Lung hyperpermeability, measured as
24                   BALF protein, was  evaluated 72 hours after exposure to 0.3 ppm O3 and found to be
25                   associated with a functioning TLR4 gene. This study concluded that Tlr4 was a strong
26                   candidate gene for susceptibility to hyperpermeability in response to O3 (Kleeberger et
27                   al.. 2000). A subsequent study by these same investigators found that Tlr4 modulated
28                   Nos2 mRNA levels and suggested that the gene product of Nos2, iNOS, plays an
29                   important role in O2-induced lung hyperpermeability (0.3 ppm, 72 hours) (Kleeberger et
30                   al., 2001). More recently, HSP70 was identified as part of the TLR4 signaling pathway
31                   (0.3 ppm, 6-72 hours) (Bauer et al.. 2011).

32                   Antioxidants have been shown to confer resistance to O3-induced injury. In a  recent
33                   study,  lung  hyperpermeability in response to O3 (0.3 ppm, 48 hours) was unexpectedly
34                   reduced in mice deficient in the  glutamate-cysteine ligase modifier subunit gene
35                   compared with sufficient mice (Johansson et al., 2010). Since the lungs of these mice
36                   exhibited 70% glutathione depletion, protection against O3-induced injury was
37                   unexpected (Johansson et al.. 2010). However it was found that several other antioxidant
      Draft - Do Not Cite or Quote                       5-35                                 September 2011

-------
 1                   defenses, including metallothionein, were upregulated in response to O3 to a greater
 2                   degree in the glutathione-deficient mice compared with sufficient mice (Johansson et al..
 3                   2010). The authors suggested that resistance to O3-induced lung injury was due to
 4                   compensatory augmentation of antioxidant defenses (Johansson et al.. 2010). Antioxidant
 5                   effects have also been attributed to Clara cell secretory protein (CCSP) and surfactant
 6                   protein A (SP-A). CCSP was found to modulate the susceptibility of airway epithelium to
 7                   injury in mice exposed to O3 (0.2 or 1 ppm for 8 hours) by an unknown mechanism
 8                   (Plopper et al.. 2006). SP-A protected against O3-induced airways inflammation and
 9                   injury in mice (2 ppm, 3 hours), possibly by acting as a sacrificial substrate (Hague et al..
10                   2007).

11                   Increased epithelial permeability has been proposed to play a role in allergic  sensitization
12                   (Matsumura. 1970). in activation of neural reflexes and in stimulation of smooth muscle
13                   receptors (Dimeo et al.. 1981). Abraham et al. (1984) reported a correlation between
14                   airway permeability and airways hyperresponsiveness (AHR) in O3-exposed sheep.
15                   However a recent study in human subjects exposed to 220 ppb O3 for 135 minutes while
16                   exercising did not find a relationship between O3-induced changes in airway permeability
17                   and AHR (Oue et al.).
             5.3.5   Sensitization of bronchial smooth muscle

18                   Bronchial reactivity is generally determined in terms of a response to a challenge agent.
19                   Non-specific bronchial reactivity in humans is assessed by measuring the effect of
20                   inhaling increasing concentrations of a bronchoconstrictive drug on lung mechanics
21                   (sRaw or FEVi). Methacholine is most commonly employed but histamine and other
22                   agents are also used. Specific bronchial reactivity is assessed by measuring effects in
23                   response to an inhaled allergen in individuals (or animals) already sensitized to that
24                   allergen. An increase in sRaw in response to non-specific or specific challenge agents
25                   indicates AHR.

26                   In addition to causing mild airway obstruction as discussed above, acute O3 exposure
27                   results in reversible increases in bronchial reactivity by mechanisms which are not well
28                   understood. In one study, bronchial reactivity of healthy subjects was significantly
29                   increased 19-hours postexposure to O3 (120-240 ppb O3 for 2 hours with intermittent
30                   exercise) (Foster et al.. 2000). These effects may be more significant in human subjects
31                   with already compromised airways (Section 5.4.2.2).

32                   Ozone may sensitize bronchial smooth muscle to stimulation through a direct effect on
33                   smooth muscle or through effects on the sensory nerves in the epithelium or on the  motor
34                   nerves innervating the smooth muscle (O'Bvrne et al.. 1984; O'Byrne etal.. 1983;


      Draft - Do Not Cite or Quote                       5-36                                 September 2011

-------
 1                   Holtzman et al., 1979). It is also recognized that increased bronchial reactivity can be
 2                   both a rapidly occurring and a persistent response to O3 (Foster and Freed. 1999).
 3                   Tachykinins and secondary oxidation products of O3 have been proposed as mediators of
 4                   the early response and inflammation-derived products have been proposed as mediators
 5                   of the later response (Foster and Freed, 1999).

 6                   Ozone-induced increases in epithelial permeability, which could improve access of
 7                   agonist to smooth muscle receptors, may be one mechanism of sensitization through a
 8                   direct effect on bronchial smooth muscle (Holtzman et al., 1979). As noted above, a
 9                   correlation between airway permeability and AHR has been reported in O3-exposed
10                   sheep (Abraham et al.,  1984) but not in O3-exposed human subjects (Oue et al.).

11                   Neurally-mediated sensitization has been demonstrated. In human subjects exposed for 2
12                   hours to 600 ppb O3 while exercising, pretreatment with atropine inhibited O3-induced
13                   AHR, suggesting the involvement of cholinergic postganglionic pathways (Holtzman et
14                   al.. 1979). Animal studies have demonstrated that O3-induced AHR involved vagally-
15                   mediated responses (rabbits, 0.2 ppm O3, 72 hours) (Freed et al.. 1996) and local axon
16                   reflex responses through bronchopulmonary C-fiber-mediated release of SP (guinea pigs,
17                   0.8 ppm O3, 2 hours) (Joadetal.. 1996). Further, pretreatment with capsaicin to deplete
18                   nerve fibers of SP blocked O3-mediated AHR (guinea pigs, 1-2 ppm O3, 2-2.25 hours)
19                   (Tepper et al.. 1993). Other investigators demonstrated that SP released from airway
20                   nociceptive neurons in  ferrets contributed to O3-induced AHR (2 ppm O3, 3 hours) (Wu
21                   et al.. 2008b: Wu et al.. 2003).

22                   Some evidence suggests the involvement of arachidonic acid metabolites and neutrophils
23                   in mediating O3-induced AHR (Seltzer et al., 1986; Fabbrietal., 1985). Increased BAL
24                   neutrophils and cyclooxygenase products were found in one study demonstrating AHR in
25                   exercising humans (600 ppb for 2 hours) immediately postexposure to (Seltzer et al.,
26                   1986). Another study found that ibuprofen pretreatment had no effect on AHR in
27                   exercising humans following exposure to 400 ppb O3 for 2 hours, although spirometric
28                   responses were blunted (Hazucha et al.. 1996). This study indicated that the arachidonic
29                   acid metabolites whose generation was blocked by ibuprofen, (i.e. prostaglandins,
30                   thromboxanes and some leukotrienes) did not play a role in AHR. Experiments  in dogs
31                   exposed for 2 hours to 2.1 ppm O3 demonstrated a close correlation between O3-induced
32                   AHR and airways neutrophilic inflammation measured in tissue biopsies (Holtzman et
33                   al., 1983). Furthermore, the increased AHR observed in dogs following O3 exposure  (3
34                   ppm, 2 hours) was inhibited by neutrophil depletion (O'Byrne et al.. 1983) and by pre-
35                   treatment with inhibitors of arachidonic acid metabolism. In one of these studies,
36                   indomethacin pre-treatment did not prevent airways neutrophilia in response to O3 (3
37                   ppm, 2 hours) providing evidence that the subset of arachidonic acid metabolites whose
      Draft - Do Not Cite or Quote                       5-37                                September 2011

-------
 1                  generation was inhibitable by the cyclooxygenase inhibitor indomethacin (i.e.,
 2                  prostaglandins and thromboxanes) was not responsible for neutrophil influx (O'Byrne et
 3                  al.. 1984). Taken together, these findings suggest that arachidonic acid metabolites, but
 4                  probably not prostaglandins or thromboxanes, may be involved in the AHR response
 5                  following O3 exposure in dogs. Studies probing the role of neutrophils in mediating the
 6                  AHR response have provided inconsistent results (Al-Hegelan et al. 2011).

 7                  Evidence for cytokine and chemokine involvement in the AHR response to O3 has been
 8                  described. Some studies have suggested a role for TNF-a (mice, 0.5 and 2 ppm O3, 3
 9                  hours) (Cho et al.. 2001; Shore et al.. 2001) and IL-1 (mice and ferrets, 2 ppm O3, 3
10                  hours) (Wu et al.. 2008b; Park et al.. 2004). The latter study found that SP expression in
11                  airway neurons was upregulated by IL-1 which was released in response to O3. Other
12                  studies in mice have demonstrated a key role for CXCR2, the chemokine receptor for the
13                  neutrophil chemokines KC and MIP-2, but not for IL-6 in O3-mediated AHR ( 1  ppm O3,
14                  3 hours) (Johnston et al.. 2005a; Johnston et al.. 2005b). In contrast, CXCR2 and IL-6
15                  were both required for neutrophil influx in  this model (Johnston et al.. 2005a: Johnston et
16                  al.. 2005b). as discussed above. Williams et al. (2008a) demonstrated that the Th2
17                  cytokine IL-13 contributed to AHR, as well as to airways neutrophilia, in mice (3 ppm
18                  O3, 3 hours).

19                  Other studies have focused on the role of TLR4. Hollingsworth et al. (2004) measured
20                  AHR, as well as airways neutrophilia, in mice 6 and 24 hours following acute (2 ppm O3
21                  for 3 hours) and subchronic (0.3 ppm for 3  days) exposure to O3. TLR4 is a key
22                  component of the innate immune system and is responsible for the immediate
23                  inflammatory response seen following challenge with endotoxin and other pathogen-
24                  associated substances. In this study, a functioning TLR4 was required for the full AHR
25                  response following O3 exposure but not for airways neutrophilia (Hollingsworth et al..
26                  2004). These findings are complemented by an older study demonstrating that O3 effects
27                  on lung hyperpermeability required a functioning TLR4 (mice, 0.3 ppm O3, 72 hours)
28                  (Kleeberger et al.. 2000). Williams et al. (2007b) found that TLR2, TLR4 and the TLR
29                  adaptor protein MyD88 contributed to AHR in mice (3 ppm O3, 3 hours). Ozone was also
30                  found to upregulate MyD88, TLR4 and TLR4 gene expression in this model (Williams et
31                  al.. 2007b).

32                  A newly recognized mechanistic basis for O3-induced AHR is  provided by studies
33                  focusing on the role of hyaluronan following O3 exposure in mice (Garantziotis et al..
34                  2010; Garantziotis et al.. 2009). Hyaluronan is an extracellular matrix component which
35                  is normally found in the ELF as a large polymer. Briefly, TLR4 and CD44 were found to
36                  mediate AHR in response to O3 and hyaluronan. Exposure to 2 ppm O3 for 3 hours
3 7                  resulted in enhanced AHR and elevated levels of soluble low molecular weight
      Draft - Do Not Cite or Quote                      5-38                                September 2011

-------
 1                   hyaluronan in the BALF 24-hours postexposure (Garantziotis et al., 2010; Garantziotis et
 2                   al.. 2009). Ozone may have caused the depolymerization of hyaluronan to soluble
 3                   fragments which are known to be endogenous ligands of the CD44 receptor and TLR4 in
 4                   the macrophage (Jiang et al.. 2005). In the two recent studies, O3-induced AHR was
 5                   attenuated in CD44 and TLR4-deficient mice (Garantziotis et al., 2010; Garantziotis et
 6                   al.. 2009). Hyaluronan fragment-mediated stimulation of AHR was found to require
 7                   functioning CD44 receptor and TLR4 (Garantziotis et al., 2010; Garantziotis et al.. 2009).
 8                   In contrast, high-molecular-weight hyaluronan blocked AHR in response to O3
 9                   (Garantziotis et al.. 2009). In another study high-molecular-weight hyaluronan enhanced
10                   repair of epithelial injury (Jiang et al.. 2005). These studies provide a link between innate
11                   immunity and the development of AHR following O3 exposure, and indicate a role for
12                   TLR4 in increasing airways responsiveness. While TLR4-dependent responses usually
13                   involve activation of NF-KB and the upregulation of proinflammatory factors, the precise
14                   mechanisms leading to AHR are unknown (Al-Hegelan et al.. 2011).

15                   In guinea pigs, AHR was found to be mediated by different pathways at 1- and 3-days
16                   postexposure to a single dose of O3 (2 ppm for 4 hours) (Verhein etal., 2011; Yost et al.,
17                   2005). At 1 day, AHR was due to activation of airway parasympathetic nerves rather than
18                   to a direct effect on smooth muscle (Yostet al.. 2005). This effect occurred as a result of
19                   O3-stimulated release of major basic protein from eosinophils (Yost et al., 2005). Major
20                   basic protein is known to block inhibitory M2 muscarinic receptors which normally
21                   dampen acetylcholine release from parasympathetic nerves (Yostet al., 2005). The
22                   resulting increase in acetylcholine release caused an increase in smooth muscle
23                   contraction following O3 exposure (Yostetal., 2005). Eosinophils played a different role
24                   3-days postexposure to O3 in guinea pigs (Yost et al.. 2005). Ozone-mediated influx of
25                   eosinophils into lung airways resulted in a different population of cells present 3-days
26                   postexposure compared to those present at  1 day (Yost et al.. 2005). At this time point,
27                   eosinophil-derived major basic protein increased smooth muscle responsiveness to
28                   acetylcholine which also contributed to AHR (Yost et al.. 2005). However, the major
29                   effect of eosinophils was to protect against vagal hyperreactivity (Yost et al., 2005). The
30                   authors suggested that these beneficial effects were due to the production of nerve growth
31                   factor (Yost et al., 2005). Further work by these investigators demonstrated a key role for
32                   IL-1|3 in mediating AHR 3-days postexposure to O3 (Verhein et al.. 2011). In this study,
33                   IL-1|3 increased nerve growth factor and SP which acted through the NK1 receptor to
34                   cause vagally-mediated bronchoconstriction (Verhein et al.. 2011). The mechanism by
35                   which SP caused acetylcholine release from parasympathetic nerves following O3
36                   exposure was not determined (Verhein et al.. 2011). Taken together, the above study
37                   results indicate that mechanisms involved in O3-mediated AHR can vary over time
38                   postexposure and that eosinophils and SP can play a role. Results of this animal model
      Draft - Do Not Cite or Quote                       5-39                                 September 2011

-------
 1                   may provide some insight into allergic airways disease in humans which is characterized
 2                   by eosinophilia (Section 5.4.2.2).
            5.3.6   Modification of innate/adaptive immune system responses

 3                   Host defense depends on effective barrier function and on innate immunity and adaptive
 4                   immunity (Al-Hegelan et al.. 2011). Ozone's effect on barrier function in the airways was
 5                   discussed above (Section 5.3.4). This section focuses on the mechanisms by which O3
 6                   impacts innate and adaptive immunity. Both tissue damage  and foreign pathogens are
 7                   triggers for the activation of the innate immune system. This results in the influx of
 8                   inflammatory cells such as neutrophils, mast cells, basophils, eosinophils, monocytes and
 9                   dendritic cells and the generation of cytokines such as TNF-a, IL-1, IL-6, KC and IL-17 .
10                   Further, innate immunity encompasses the actions of complement and collectins and the
11                   phagocytic functions of macrophages, neutrophils and dendritic cells. Airway epithelium
12                   also contributes to innate immune responses. Innate immunity is highly dependent on cell
13                   signaling networks involving TLR4. Adaptive immunity provides immunologic memory
14                   through the actions of B and T cells. Important links between the two systems are
15                   provided by dendritic cells and antigen presentation. Recent studies demonstrate that
16                   exposure to O3 modifies cells and processes which  are required for innate immunity,
17                   contributes to innate-adaptive immune system interaction and primes pulmonary immune
18                   responses to endotoxin.

19                   Ozone exposure of human subjects resulted in recruitment of activated innate immune
20                   cells to the airways. Healthy individuals were exposed to 80 ppb O3 for 6.6 hours with
21                   intermittent exercise and airways inflammation was characterized in induced sputum 18-
22                   hours postexposure  (Alexis et al.. 2010). Previous studies demonstrated that induced
23                   sputum contains liquid and cellular constituents of the ELF  from central conducting
24                   airways (Alexis et al.. 200 Ib) and also identified these airways as a site of preferential O3
25                   absorption during exercise (Hu et al.. 1994). Ozone exposure resulted in increased
26                   numbers of neutrophils, airway monocytes and dendritic-like cells in sputum (Alexis et
27                   al.. 2010). In addition, increased expression of cell surface markers characteristic of
28                   innate immunity and antigen presentation (i.e. CD-14 and HLA-DR) was demonstrated
29                   on airway monocytes (Alexis et al.. 2010). Enhanced antigen presentation contributes to
30                   exaggerated T cell responses and promotes Th2 inflammation and an allergic phenotype
31                   (Lay et al.. 2007). Upregulation of pro-inflammatory cytokines was also demonstrated in
32                   sputum of O3-exposed subjects (Alexis etal.. 2010). One of these cytokines, IL-12p70,
33                   correlated with numbers of dendritic-like cells in the sputum, and is an indicator of
34                   dendritic cell activation (Alexis et al.. 2010). These authors have previously reported that
35                   exposure of exercising human subjects to 400 ppb O3 for 2  hours resulted in activation of


      Draft - Do Not Cite or Quote                       5-40                                 September 2011

-------
 1                   monocytes and macrophages (Lav et al.. 2007). which could play a role in exacerbating
 2                   existing asthma by activating allergen-specific memory T cells. The current study
 3                   confirms these findings and extends them by suggesting a potential mechanism whereby
 4                   O 3-activated dendritic cells could stimulate naive T-cells to promote the development of
 5                   asthma (Alexis et al.. 2010). A companion study by these same investigators (described in
 6                   detail in Section 5.4.2.1) provides evidence of dendritic cell activation, measured as
 7                   increased expression of HLA-DR, in a subset of the human subjects (GSTM1 null)
 8                   exposed to 400 ppb O3 for 2 hours with intermittent exercise (Alexis et al.. 2009). Since
 9                   dendritic cells are a link between innate and adaptive immunity, these studies provide
10                   evidence for an O3-mediated interaction between the innate and adaptive immune
11                   systems.

12                   Another recent study linked O3-mediated activation of the innate immune system to the
13                   development of non-specific AHR in a mouse model (Pichavant et al.. 2008). Repeated
14                   exposure to 1 ppm O3 for 3 hours (3 days over a 5 day period) induced non-specific AHR
15                   measured 24 hours following the last exposure (Pichavant et al.. 2008). This response
16                   was found to require NKT cells, which are effector lymphocytes of innate immunity, as
17                   well as IL-17 and airways neutrophilia (Pichavant et al.. 2008). Since glycolipids such as
18                   galactosyl ceramide are ligands for the invariant CD 1 receptor on NKT cells and serve as
19                   endogenous activators of NKT cells, a role for O3-oxidized lipids in activating NKT cells
20                   was proposed (Pichavant et al.. 2008). The authors contrasted this innate immunity
21                   pathway with that of allergen-provoked specific AHR which involves adaptive immunity,
22                   the cytokines IL-4, IL -13, IL-17, and airways eosinophilia (Tichavant et al.. 2008).
23                   Interestingly, NKT cells were required for both the specific AHR provoked by allergen
24                   and the non-specific AHR provoked by O3 (Tichavant et al.. 2008). Different cytokine
25                   profiles of the NKT cells from allergen and O3-exposed mice was proposed to account
26                   for the different pathways (Tichavant et al.. 2008). More recently, NKT cells have been
27                   found to function in both innate and adaptive immunity (Vivier et al.. 2011).

28                   An interaction between allergen and O3 in the induction of nonspecific AHR was shown
29                   in another animal study (Larsen et al.. 2010). Mice were sensitized with the aerosolized
30                   allergen OVA on 10 consecutive days followed by exposure to O3 (0.1-0.5 ppm for 3
31                   hours) (Larsen etal.. 2010). While allergen sensitization alone did not alter airways
32                   responsiveness to a nonspecific challenge, O3 exposure of sensitized mice resulted in
33                   nonspecific AHR at 6- and 24-hours postexposure (Larsen et al.. 2010). The  effects of O3
34                   on AHR were independent of airways eosinophilia and neutrophilia (Larsen et al.. 2010).
3 5                   However, OVA pretreatment led to goblet cell metaplasia which was enhanced by O3
36                   exposure (Larsen et al.. 2010). It should be noted that OVA sensitization using only
37                   aerosolized antigen in this study is less common than the usual procedure for OVA
3 8                   sensitization achieved by one or more initial systemic injections of OVA and adjuvant
      Draft - Do Not Cite or Quote                       5-41                                 September 2011

-------
 1                  followed by repeated inhalation exposure to OVA. This study also points to an interaction
 2                  between innate and adaptive immune systems in the development of the AHR response.

 3                  Furthermore, O3 was found to act as an adjuvant for allergic sensitization (Hollingsworth
 4                  etal.. 2010). Oropharyngeal aspiration of OVA on day 0 and day 6 failed to lead to
 5                  allergic sensitization unless mice were first exposed to 1 ppm O3 for 2 hours
 6                  (Hollingsworth et al.. 2010). The O3-mediated response involved Th2 (IL-4, IL-5 and IL-
 7                  9) and Thl7 cytokines (IL-17) and was dependent on a functioning TLR4 (Hollingsworth
 8                  etal.. 2010). Ozone exposure also activated OVA-bearing dendritic cells in the thoracic
 9                  lymph nodes, as measured by the presence of the CD86 surface marker, which suggests
10                  naive T cell stimulation and the involvement of Th2 pathways (Hollingsworth et al..
11                  2010). Thus the adjuvant effects of O3 may be due to activation of both innate and
12                  adaptive immunity.

13                  Priming of the innate immune system by O3 was reported by Hollingsworth et al. (2007).
14                  In this study, exposure of mice to 2 ppm O3  for 3 hours led to nonspecific AHR at 24-
15                  and 48-hours postexposure, an effect which  subsided by 72 hours (Hollingsworth et al.,
16                  2007). However, in mice treated with aerosolized endotoxin immediately following O3
17                  exposure, AHR was greatly enhanced at 48-and 72-hours postexposure (Hollingsworth et
18                  al., 2007). In addition,  O3 pre-exposure was found to reduce the number of inflammatory
19                  cells in the BALF, to increase cytokine production and total protein in the BALF and to
20                  increase systemic IL-6 following exposure to endotoxin (Hollingsworth et al., 2007).
21                  Furthermore, O3 stimulated the  apoptosis of alveolar macrophages 24-hours
22                  postexposure, an effect which was greatly enhanced by endotoxin treatment. Apoptosis of
23                  circulating blood monocytes was also observed in response to the combined exposures
24                  (Hollingsworth et al., 2007). Ozone pre-exposure enhanced the response of lung
25                  macrophages to endotoxin (Hollingsworth et al.. 2007). Taken together, these findings
26                  demonstrated that O3 exposure increased innate immune responsiveness to endotoxin.
27                  The authors attributed these effects to the increased surface expression of TLR4 and
28                  increased signaling in macrophages observed in the study (Hollingsworth et al., 2007). It
29                  was proposed that the resulting decrease in airway inflammatory cells could account for
30                  O3-mediated decreased clearance of bacterial pathogens observed in numerous animal
31                  models (Hollingsworth et al.. 2007).

32                  More recently, these authors demonstrated that hyaluronan contributed to the O3-primed
33                  response to endotoxin (Li etal., 2010). In this study, exposure of mice to 1 ppm O3 for 3
34                  hours resulted in enhanced responses to endotoxin, which was mimicked by intratracheal
35                  instillation of hyaluronan fragments (Li etal.. 2010).  Hyaluronan, like O3, was also found
36                  to induce TLR4 receptor peripheralization in the macrophage membrane (Li etal.. 2010;
37                  Hollingsworth et al., 2007). an effect which  is associated with enhanced responses to
      Draft - Do Not Cite or Quote                      5-42                                September 2011

-------
 1                   endotoxin. This study and previous ones by the same investigators showed elevation of
 2                   BALF hyaluronan in response to O3 exposure (Garantziotis et al.. 2010; Li etal.. 2010;
 3                   Garantziotis et al.. 2009). providing evidence that the effects of O3 on innate immunity
 4                   are at least in part mediated by hyaluronan fragments. The authors note that excessive
 5                   TLR4 signaling can lead to lung injury and suggest that O3 may be responsible for an
 6                   exaggerated innate immune response which may underlie lung injury and decreased host
 7                   defense (Li etal.. 2010).

 8                   Activation or upregulation of the immune system has not been reported in all studies.
 9                   Impaired antigen-specific immunity was demonstrated following subacute O3 exposure
10                   (0.6 ppm, 10 h/day for 15 days) in mice (Feng et al.. 2006). Specifically,  O3 exposure
11                   altered the lymphocyte subset and cytokine profile and impacted thymocyte early
12                   development leading to immune dysfunction. Further, recent studies demonstrated SP-A
13                   oxidation in mice exposed for 3-6 hours to 2 ppm O3. SP-A is an important innate
14                   immune protein which plays a number of roles in host defense including  acting as
15                   opsonin for the recognition of some pathogens (Hague etal.. 2009). These investigations
16                   found that O 3 -mediated carbonylation of SP-A was associated with impaired macrophage
17                   phagocytosis in vitro (Mikerov et al.. 2008b). Furthermore, O3 exposure  (2 ppm for 3
18                   hours) in mice was found to increase susceptibility to pneumonia infection in mice
19                   through an impairment of SP-A dependent phagocytosis (Mikerov et al..  2008a; Mikerov
20                   etal.. 2008c).

21                   Taken together, results of recent studies provide evidence that O3 alters host
22                   immunologic response and leads to immune system dysfunction through  its effects on
23                   innate and adaptive immunity.
            5.3.7   Airways remodeling

24                   As noted above, the degree of airways inflammation due to O3 may have important long-
25                   term consequences since airways inflammation is often accompanied by tissue injury
26                   (Bahnes et al.. 1996). The nasal airways, conducting airways and distal airways (i.e.
27                   respiratory bronchioles or centriacinar region depending on the species) have all been
28                   identified as sites of O3-mediated injury and inflammation (Mudway and Kelly. 2000). At
29                   all levels of the respiratory tract, loss of sensitive epithelial cells, degranulation of
30                   secretory cells, proliferation of resistant epithelial cells and neutrophilic influx have been
31                   observed as a result of O3 exposure (Mudway and Kelly. 2000; Cho  etal.. 1999). An
32                   important study (Plopper et al.. 1998) conducted in adult rhesus monkeys (0.4 and
33                   1.0 ppm O3 for 2 hours) found that 1 ppm O3 resulted in the greatest epithelial injury in
34                   the respiratory bronchioles immediately postexposure although injury was observed at all
      Draft - Do Not Cite or Quote                       5-43                                September 2011

-------
 1                   of the RT sites studied except for the lung parenchyma. Exposure to 0.4 ppm O3 resulted
 2                   in epithelial injury only in the respiratory bronchioles.

 3                   Persistent inflammation and injury, observed in animal models of chronic and intermittent
 4                   exposure to O3, are associated with airways remodeling, including mucous cell
 5                   metaplasia of nasal transitional epithelium (Harkema et al.. 1999; Hotchkiss et al.. 1991)
 6                   and bronchiolar metaplasia of alveolar ducts (Mudway and Kelly. 2000). Fibrotic changes
 7                   such as deposition of collagen in the  airways and sustained lung function decrements
 8                   especially in small airways have also been demonstrated as a response to chronic O3
 9                   exposure (Mudway and Kelly. 2000; Chang et al.. 1992). These effects, described in
10                   detail in Section 7.2.3.1, have been demonstrated in rats exposed to levels of O3 as low as
11                   0.25 ppm. Mechanisms responsible for the resolution of inflammation and the repair of
12                   injury remain to  be clarified and there is only  a limited understanding of the biological
13                   processes underlying long-term morphological changes. However, a recent study in mice
14                   demonstrated a key role for the TGF-|3 signaling pathway in the deposition of collagen in
15                   the airways wall following chronic intermittent exposure to 0.5 ppm O3 (Katre et al..
16                   2011).

17                   It should be noted that repeated exposure to O3 results in attenuation of some O3 -
18                   induced responses, including those associated with the activation of neural reflexes (e.g.
19                   decrements in pulmonary function), as discussed in Section 5.3.2. However, numerous
20                   studies  demonstrate that some markers of injury and inflammation remain increased
21                   during multi-day exposures to O3.  Mechanisms responsible for attenuation, or the lack
22                   thereof, are incompletely understood.
            5.3.8   Systemic inflammation and oxidative/nitrosative stress

23                   Extrapulmonary effects of O3 have been noted for decades (U.S. EPA. 2006b). It has
24                   been proposed that lipid oxidation products resulting from reaction of O3 with lipids
25                   and/or cellular membranes in the ELF are responsible for systemic effects, however it is
26                   not known whether they gain access to the vascular space (Chuang et al.. 2009).
27                   Alternatively, extrapulmonary release of diffusible mediators may initiate or propagate
28                   inflammatory responses in the vascular or in systemic compartments (Cole and Freeman.
29                   2009). A role for O3 in modulating endothelin, a potent vasoconstrictor, has also been
30                   proposed. Studies in rats found that exposure to 0.4 and 0.8 ppm O3 induced endothelin
31                   system genes in the lung and increased circulating levels of endothelin (Thomson et al..
32                   2006; Thomson et al.. 2005). Systemic oxidative stress is suggested by studies in humans
33                   which reported associations between O3 exposure and levels of plasma 8-isoprostanes
      Draft - Do Not Cite or Quote                       5-44                                September 2011

-------
 1                   and the presence of peripheral blood lymphocyte micronuclei (Chen et al.. 2007; Chen et
 2                   al.. 2006a).

 3                   Ozone-induced perturbations of the cardiovascular system were recently investigated in
 4                   young mice and monkeys (Chuang et al., 2009) and in rats (Kodavanti et al.. 2011;
 5                   Perepu et al.. 2010) (see Sections 6.3.3.2 and 7.3.1.2). These are the first studies to
 6                   suggest that systemic oxidative stress and inflammation play a mechanistic role in O3-
 7                   induced effects on the systemic vascular and heart. Exposure to 0.5 ppm O3 for 5 days
 8                   resulted in oxidative/nitrosative stress, vascular dysfunction and mitochondrial DNA
 9                   damage in the aorta (Chuang et al.. 2009). Chronic exposure to 0.8 ppm O3 resulted in an
10                   enhancement of inflammation and lipid peroxidation in the heart following an ischemia-
11                   reperfusion challenge (Perepu etal.. 2010).  In addition, chronic intermittent exposure to
12                   0.4 ppm O3 increased aortic levels of mRNA for biomarkers of oxidative stress,
13                   thrombosis, vasoconstriction and proteolysis and aortic lectin-like oxidized4ow density
14                   lipoprotein receptor-l(LOX-l) mRNA and protein levels (Kodavanti et al., 2011). The
15                   latter study suggests a role for circulating oxidized lipids in mediating the effects of O3.

16                   Systemic inflammation and oxidative/nitrosative stress may similarly affect other organ
17                   systems as well as the plasma compartment. Circulating cytokines have the potential to
18                   enter the brain through diffusion and active  transport and to contribute to
19                   neuroinflammation, neurotoxicity, cerbrovascular damage and a break-down of the  blood
20                   brain barrier (Block and Calderon-Garciduenas. 2009) (see Sections 6.4 and 7.5). They
21                   can also activate  neuronal afferents (Block and Calderon-Garciduenas. 2009). Vagal
22                   afferent pathways originating in the respiratory tract may also be responsible for O3-
23                   mediated activation of nucleus tractus solitarius neurons which resulted in neuronal
24                   activation in stress-responsive regions of the CNS in rats (0.5 or 2 ppm O3 for 1.5-120
25                   hours) (Gackiere et al.. 2011). Recent studies have demonstrated O3-induced brain lipid
26                   peroxidation, cytokine production in the brain and upregulated expression  of VEGF in
27                   rats (0.5 ppm O3, 3 hours or 0.25-0.5 ppm O3, 4  h/day, 15-60 days) (Guevara-Guzman et
28                   al.. 2009; Araneda  et al., 2008; Perevra-Munoz et al., 2006). Further, O3-induced
29                   oxidative stress resulted in increased plasma lipid peroxides (0.25 ppm, 4h/day, 15-60
30                   days) (Santiago-Lopez et al., 2010). which was correlated with damage to  specific brain
31                   regions (Pereyra-Munoz et al.. 2006).

32                   Oxidative stress is one mechanism by which testicular and sperm function is disrupted
33                   (see  Section  7.4.1). Oxidative stress may inhibit testicular steroidogenesis  leading to
34                   decreased testosterone levels (Diemer et al.. 2003). It may decrease sperm quality by lipid
3 5                   peroxidation of sperm plasma membrane which leads to impaired sperm mobility
36                   (Agarwal et al.. 2003). Further, it may damage DNA in the  sperm nucleus leading to
37                   apoptosis and a decline in sperm counts  (Agarwal et al.. 2003). Since oxidative stress is a
      Draft - Do Not Cite or Quote                       5-45                                 September 2011

-------
 1                   key event underlying many of the health effects of O3, it is possible that sperm quality
 2                   and quantity may be impacted by this mechanism (Sokol et al.. 2006).

 3                   A role for plasma antioxidants in modulating O3-induced respiratory effects was
 4                   suggested by a recent study (Aibo etal.. 2010). In this study, pretreatment of rats with a
 5                   high dose of acetaminophen resulted in increased levels of plasma cytokines and the
 6                   influx of inflammatory cells into the lung following O3 exposure (0.25-0.5 ppm, 6 hours)
 7                   (Aiboet al.. 2010). These effects were not observed in response to O3 alone.
 8                   Furthermore, acetaminophen-induced liver injury was exacerbated by O3 exposure. A
 9                   greater increase in hepatic neutrophil accumulation and greater alteration in gene
10                   expression profiles was observed in mice exposed to O3 and acetaminophen compared
11                   with either exposure alone (Aibo et al.. 2010). Although not measured in this study,
12                   glutathione depletion in the liver is known to occur in acetaminophen toxicity. Since liver
13                   glutathione is the source of plasma glutathione, acetaminophen treatment may have
14                   lowered plasma glutathione levels and altered the redox balance in the vascular
15                   compartment. These findings indicate interdependence between respiratory tract, plasma
16                   and liver responses to O3, possibly related to glutathione status.
            5.3.9   Impaired alveolar-arterial 02 transfer

17                   O3 may impair alveolar-arterial oxygen transfer and reduce the supply of arterial oxygen
18                   to the myocardium. This may have a greater impact in individuals with compromised
19                   cardiopulmonary systems. Gong et al. (1998) provided evidence of a small decrease in
20                   arterial oxygen saturation in human subjects exposed for 3 hours to 300 ppb O3 while
21                   exercising. In addition, Delaunois et al. (1998) demonstrated pulmonary vasoconstriction
22                   in O3-exposed rabbits (0.4 ppm, 4 hours). Although of interest, the contribution of this
23                   pathway to O3-induced cardiovascular effects remains uncertain.
            5.3.10  Summary

24                   This section summarizes the modes of action and toxicity pathways resulting from O3
25                   inhalation (Figure 5-9). These pathways provide a mechanistic basis for the health effects
26                   which are described in detail in Chapters 6 and 7. Three distinct short-term responses
27                   have been well-characterized in humans challenged with O3: decreased pulmonary
28                   function, airways inflammation, and increased bronchial reactivity. In addition, O3
29                   exposure exacerbates, and possibly also causes, asthma and allergic airways disease in
30                   humans. Animal studies have demonstrated airways remodeling and fibrosis in response
31                   to chronic and intermittent O3 exposures and a wide range of other responses. While the
      Draft - Do Not Cite or Quote                       5-46                                September 2011

-------
 1
 2
respiratory tract is the primary target tissue, cardiovascular and other organ effects occur
following short- and long-term exposures of animals to O3.
                                      Ozone + Respiratory Tract
                                  Formation of secondary oxidation products
           Activation
           of neural
           reflexes
   Initiation of
   inflammation
Sensitization
of bronchial
smooth muscle
                  Systemic inflammation and
                  oxidative/nitrosative stress
                     Extrapulmonary Effects
                                       Decrements in pulmonary function
                                       Pulmonary inflammation/oxidative stress
                                       Increases in airways permeability
                                       Airways hyperresponsiveness
                                       Exacerbation/induction of asthma
                                       Decreased  host defenses
                                       Epithelial metaplasia and fibrotic airways
                                       Altered lung development
      Figure 5-9     The modes of action/possible pathways underlying the health
                       effects resulting from inhalation exposure to O3.
 4
 5
 6
 7
10
11
12
13
14
The initial key event in the toxicity pathway of O3 is the formation of secondary
oxidation products in the respiratory tract. This involves direct reactions with components
of the ELF and/or plasma membranes of cells residing in the respiratory tract. The
resulting secondary oxidation products transmit signals to the epithelium, nociceptive
sensory nerve fibers and, if present, dendritic cells, mast cells and eosinophils. Thus, O3
effects are mediated by components of ELF and by the multiple cell types found in the
respiratory tract. Further, oxidative stress is an implicit part of this initial key event.

Another key event in the toxicity pathway of O3 is the activation of neural reflexes which
lead to decrements in pulmonary function (see Section 6.2.1). Evidence is accumulating
that secondary oxidation products are responsible for this effect. Eicosanoids have been
implicated  in humans, while both eicosanoids and aldehydes are effective in animal
models. Different receptors on bronchial C-fibers have been shown to mediate separate
      Draft - Do Not Cite or Quote
                                5-47
                                   September 2011

-------
 1                   effects of O3 on pulmonary function. Nociceptive sensory nerves are involved in the
 2                   involuntary truncation of respiration which results in decreases in FVC, FEVi, tidal
 3                   volume and pain upon deep inspiration. Opioids block these responses while atropine has
 4                   only a minimal effect. New evidence in an animal model suggests that TRPA1 receptors
 5                   on bronchial C-fibers mediate this pathway. Ozone exposure also results in activation of
 6                   vagal sensory nerves and a mild increase in airway obstruction measured as increased
 7                   sRaw. Atropine and (3-adrenergic agonists greatly inhibit this response in humans
 8                   indicating that the airway obstruction is due to bronchoconstriction. Other studies in
 9                   humans implicated SP release from bronchial C-fibers resulting in airway narrowing due
10                   to either neurogenic edema or bronchoconstriction. New evidence in an animal model
11                   suggests that the SP-NK receptor pathway caused bronchoconstriction following O3
12                   exposure.

13                   Initiation of inflammation  is also a key event in the toxicity pathway of O3.  Secondary
14                   oxidation products, as well as chemokines and cytokines elaborated by airway epithelial
15                   cells and macrophages, have been implicated in the initiation of inflammation. Vascular
16                   endothelial adhesion molecules may also play a role. Work from several laboratories in
17                   using human subjects and animal models suggest that O3 triggers the release of
18                   tachykinins such as  SP from airway sensory nerves which could contribute to
19                   downstream effects  including inflammation (see  Sections 6.2.3 and 7.2.4). Airways
20                   neutrophilia has been demonstrated in BALF, mucosal biopsy and induced sputum
21                   samples. Influx of mast cells, monocytes and macrophages also occur. Inflammation
22                   further contributes to O3-mediated oxidative stress. Recent investigations show that O3
23                   exposure leads to the generation of hyaluronan fragments from high molecular weight
24                   polymers of hyaluronan normally found in the ELF in mice. Hyaluronan activates TLR4
25                   and CD44-dependent signaling pathways in macrophages, and results in an increased
26                   number of macrophages in the BALF. Activation of these pathways occurs later than the
27                   acute neutrophilic response suggesting that they may contribute to longer-term effects of
28                   O3. The mechanisms involved in clearing O3-provoked inflammation remain to be
29                   clarified. It should be noted that inflammation, as measured by airways neutrophilia, is
30                   not correlated with decrements in pulmonary function as measured by spirometry.

31                   A fourth key event in the toxicity pathway of O3  is alteration of epithelial barrier
32                   function. Increased permeability occurs as a result of damage to tight junctions between
33                   epithelial cells subsequent to O3-induced injury and inflammation. It may play a role in
34                   allergic sensitization and in AHR (see Sections 6.2.2, 6.2.6, and 7.2.5). Tachykinins
35                   mediate this response while antioxidants confer protection. Genetic susceptibility has
36                   been associated with a functioning TLR4 gene and with iNOS.
      Draft - Do Not Cite or Quote                       5-48                                September 2011

-------
 1                   A fifth key event in the toxicity pathway of O3 is the sensitization of bronchial smooth
 2                   muscle.

 3                   Increased bronchial reactivity can be both a rapidly occurring and a persistent response.
 4                   The mechanisms responsible for early and later AHR are not well-understood (see
 5                   Section 6.2.2). One proposed mechanism of sensitization, O3-induced increases in
 6                   epithelial permeability, would improve access of agonist to smooth muscle receptors. The
 7                   evidence for this mechanism is not consistent. Another proposed mechanism, for which
 8                   there is greater evidence, is neurally-mediated sensitization. In humans exposed to O3,
 9                   atropine blocked the early AHR response indicating the involvement of cholinergic
10                   postganglionic pathways. Animal studies demonstrated that O3-induced AHR involved
11                   vagally-mediated responses and local axon reflex responses through bronchopulmonary
12                   C-fiber-mediated release of SP. Later phases of increased bronchial reactivity may
13                   involve the induction of IL-1(3 which in turn upregulates SP production. In guinea pigs,
14                   eosinophil-derived major basic protein contributed to the stimulation of cholinergic
15                   postganglionic pathways. A novel role for hyaluronan in mediating the later phase effects
16                   O 3 -induced AHR has recently been demonstrated. Hyaluronan fragments stimulated AHR
17                   in a TLR4- and CD44 receptor-dependent manner. Tachykinins and secondary oxidation
18                   products of O3 have been proposed as mediators of the early response and inflammation-
19                   derived products have been proposed as mediators of the later response. Inhibition of
20                   arachidonic acid metabolism was ineffective in blocking O3-induced AHR in humans
21                   while in animal models mixed results were found. Other cytokines and chemokines have
22                   been implicated in the AHR response to O3 in animal models.

23                   A sixth key event in the toxicity pathway of O3 is the modification of innate/adaptive
24                   immunity. While the majority of evidence for this key event comes from animal studies,
25                   there are several studies suggesting that this pathway may also be relevant in humans. O3
26                   exposure of human subjects resulted in recruitment of activated innate immune cells to
27                   the airways. This included macrophages and monocytes with increased expression of cell
28                   surface markers characteristic of innate immunity and antigen presentation, the latter of
29                   which could contribute to exaggerated T cell responses and the promotion of an allergic
30                   phenotype. Evidence of dendritic cell activation was observed in GSTM1 null human
31                   subjects exposed to O3, suggesting O3-mediated interaction between the innate and
32                   immune systems. Animal studies  further linked O3-mediated activation of the innate
33                   immune system to the development of nonspecific AHR, demonstrated an interaction
34                   between allergen and O3 in the induction of nonspecific AHR, and found that O3 acted as
35                   an adjuvant for allergic sensitization through the activation of both innate and adaptive
36                   immunity. Priming of the innate immune system by O3 was reported in mice. This
37                   resulted in an exaggerated response to endotoxin which included enhanced TLR4
38                   signaling in macrophages. Ozone-mediated impairment of the function of SP-A, an innate
      Draft - Do Not Cite or Quote                       5-49                                September 2011

-------
 1                  immune protein, has also been demonstrated. Taken together these studies provide
 2                  evidence that O3 can alter host immunologic response and lead to immune system
 3                  dysfunction. These mechanisms may underlie the exacerbation and induction of asthma
 4                  (see Sections 6.2.6 and 7.2.1), as well as decreases in host defense (see Sections 6.2.5 and
 5                  7.2.6).

 6                  Another key event in the toxicity pathway of O3 is airways remodeling. Persistent
 7                  inflammation and injury, which are observed in animal models of chronic and
 8                  intermittent exposure to O3, are associated with morphologic changes such as mucous
 9                  cell metaplasia of nasal epithelium, bronchiolar metaplasia of alveolar ducts and fibrotic
10                  changes in small airways (see Section 7.2.3). Mechanisms responsible for these responses
11                  are not well-understood. However a recent study in mice demonstrated a key role for the
12                  TGF-(3 signaling pathway in the deposition of collagen in the airway wall following
13                  chronic intermittent exposure to O3.

14                  Systemic inflammation and vascular oxidative/nitrosative stress are also key events in the
15                  toxicity pathway of O3. Extrapulmonary effects of O3 occur in numerous organ systems,
16                  including the cardiovascular, central nervous, reproductive and hepatic systems (see
17                  Sections 6.3 to 6.5 and 7.3 to 7.5). It has been proposed that lipid oxidation products
18                  resulting from reaction of O3 with lipids and/or cellular membranes in the ELF are
19                  responsible for systemic responses, however it is not known whether they gain access to
20                  the vascular space. Alternatively, release of diffusible mediators from the lung into the
21                  circulation may initiate or propagate inflammatory responses in the vascular or in
22                  systemic compartments. Systemic oxidative stress is suggested by  studies in humans
23                  which reported associations between O3 exposure and levels of plasma 8-isoprostanes
24                  and the presence of peripheral blood lymphocyte micronuclei.
          5.4   Interindividual Variability in Response

25                  Responses to O3 exposure are variable within the population and the basis for this
26                  variability is not clear (Mudway and Kelly. 2000). Both dosimetric and mechanistic
27                  factors are likely to contribute to this variability and are discussed below.
            5.4.1    Dosimetric Considerations

28                   Two studies have investigated the correlation of O3 uptake with the pulmonary function
29                  responses to O3 exposure (Reeser et al., 2005; Gerrityetal.. 1994). These studies found
30                  that the large subject-to-subject variability in %AFEVi response to O3 does not appear to
      Draft - Do Not Cite or Quote                      5-50                                September 2011

-------
 1                   have a dosimetric explanation. Reeser et al. (2005) found no significant relationship
 2                   between %AFEVi and fractional absorption of O3 using the bolus method. Contrary to
 3                   previous findings, the percent change in dead space volume of the respiratory tract
 4                   (%AVD) did not correlate with O3 uptake, possibly due to the contraction of dead space
 5                   caused by airway closure. Gerrity et al. (1994) found that intersubject variability in FEVi
 6                   and airway resistance was not related to differences in the O3 dose delivered to the lower
 7                   airways, whereas minute ventilation was predictive of FEVi decrement. No study has yet
 8                   demonstrated that subjects show a consistent pattern of O3 retention when re-exposed
 9                   over weeks of time, as has been shown to be the case for the FEVi response, or that
10                   within-subject variation in FEVi response is related to fluctuations in O3 uptake.

11                   A delay in onset of O3-induced pulmonary function responses has been noted in
12                   numerous studies. Recently the delay was characterized in terms of changes in fB
13                   (Schelegle et al.. 2007). In humans exposed for 1-2 hours to 120-350 ppb O3 while
14                   exercising, no change in fB was observed until a certain cumulative inhaled dose of O3
15                   had been reached. Subsequently, the magnitude of the change in fB was correlated with
16                   the inhaled dose rate (Schelegle et al.. 2007). These investigators proposed that initial
17                   reactions of O3 with ELF resulted in a time-dependent depletion of ELF antioxidants, and
18                   that activation of neural reflexes occurred only after the antioxidant defenses were
19                   overwhelmed (Schelegle et al.. 2007).

20                   Other studies investigated the relationship between O3 dose and cellular injury. In two
21                   studies, the initial cellular injury was found to correlate with the site-specific O3 dose.
22                   Contained within the CAR, the respiratory bronchioles were confirmed as the site
23                   receiving the greatest O3 dose (18O mass/lung weight) and sustained the greatest initial
24                   cellular injury in O3  (0.4 and 1.0 ppm for 2 hours) exposed resting rhesus monkeys
25                   (Plopper et al.. 1998). The respiratory bronchioles, having the highest concentration of
26                   local O3 dose, were also the site of significant GSH reduction. In addition, a study in
27                   isolated perfused rat lungs found greater injury in conducting airways downstream of
28                   bifurcations where local doses of O3 were higher (Postlethwait et al.. 2000).

29                   Further, the degree of inflammation in rats has been correlated with 18O-labeled O3 dose
30                   markers in the lower lung. In female rats exposed to 0.8 ppm O3 for 4 hours, BAL
31                   neutrophil number and 18O reaction product were directly proportional (Gunnison and
32                   Hatch. 1999). Kari et al. (1997) observed that a 3-week caloric restriction (75%) in rats
33                   abrogated the toxicity of O3  (2 ppm, 2 hours), measured as BALF increases in protein,
34                   fibronectin and neutrophils, which was seen in normally fed rats. Accompanying this
35                   resistance to O3 toxicity was a reduction (30%) in the accumulation  of 18O reaction
36                   product in the lungs. These investigations also demonstrated an inverse relationship
37                   between AH2 levels  and O3  dose and provided evidence for AH2 playing a protective
      Draft - Do Not Cite or Quote                       5-51                                 September 2011

-------
 1                   role following O3 exposure in these studies. Pregnant and lactating rats had lower AH2
 2                   content in BALF and exhibited a greater increase in accumulation of 18O reaction
 3                   products compared with pre-pregnant rats in response to O3 exposure (Gunnison and
 4                   Hatch. 1999). In the calorie restricted model, a 30% higher basal BALF AH2
 5                   concentration and a rapid accumulation of AH2 into the lungs to levels 60% above
 6                   normal occurred as result of O3 exposure (Kari etal.. 1997). However, this relationship
 7                   between AH2 levels and O3 dose did not hold up in every study. Aging rats (9 and 24
 8                   months old) had 49% and 64% lower AH2 in lung tissue compared with month-old rats
 9                   but the aging-induced AH2 loss did not increase the accumulation of 18O reaction
10                   products following O3 exposure (0.4-0.8 ppm, 2-6 hours) (Vincent et al.. 1996a).

11                   Interindividual variability in the neutrophilic response has been noted in human subjects
12                   (Holz etal.. 1999; Devlin etal.. 1991; Schelegle et al.. 1991). One study demonstrated a
13                   threefold difference in airways neutrophilia, measured as percent of total cells in
14                   proximal BALF, among human subjects exposed to 300 ppb O3 for 1 hour while
15                   exercising (Schelegle et al.. 1991). Another study reported a 20-fold  difference in BAL
16                   neutrophils following exposure to 80-100 ppb O3 for 6.6 hours in exercising human
17                   subjects (Devlin et al.. 1991). Reproducibility of intra-individual responses to 1-hour
18                   exposure to 250 ppb O3, measured as sputum neutrophilia, was demonstrated by Holz
19                   (1999). Few studies have examined the dose- or concentration-responsiveness of airways
20                   neutrophilia in O3-exposed humans (Holz et al.. 1999; Devlin etal..  1991). No
21                   concentration-responsiveness was observed in healthy human subjects exposed for 1 hour
22                   to 125-250 ppb O3 and a statistically significant increase in sputum neutrophilia was
23                   observed only at the higher dose (Holz etal.. 1999). However, concentration-dependent
24                   and statistically significant increases in BAL neutrophils and the inflammatory mediator
25                   IL-6 were reported following exposure to 80 and 100 ppb O3 for 6.6 hours in exercising
26                   humans (Devlin et al.. 1991). Additional evidence is provided by a meta-analysis of the
27                   O3 dose-inflammatory response in controlled human exposure studies involving  exposure
28                   to 80-600 ppb O3 for 60-396 minutes (Mudwav and Kelly. 2004b). Results demonstrated
29                   a linear relationship between inhaled O3 dose (determined as the product of
30                   concentration, ventilation and time) and BAL neutrophils at 0-6 hours and 18-24 hours
31                   following O3 exposure  (Mudway and Kelly. 2004b).

32                   Collectively these studies demonstrate a correlation between dose and response for some
33                   O3-induced effects and suggest a role for ELF antioxidants in modulating the dose to
34                   tissue. The lack of correlation between O3-induced effects and calculated O3 dose may be
35                   a result of interindividual differences in TB volume.
      Draft - Do Not Cite or Quote                      5-52                                September 2011

-------
            5.4.2   Mechanistic Considerations

 1                   There was a large range of pulmonary function responses to O3 among healthy young
 2                   adults exposed for 4 hours to 200 ppb O3 or for 1.5 hours to 420 ppb O3 while exercising
 3                   (Hazucha et al.. 2003; Balmes et al.. 1996). Since individual responses were relatively
 4                   consistent across time, it was thought that responsiveness reflected an intrinsic
 5                   characteristic of the subject (Mudway and Kelly. 2000). Older adults were generally not
 6                   responsive to O3 (Hazucha et al.. 2003). while obese young women may have been more
 7                   responsive than lean young women (420 ppb, 1.5 hours, while exercising) (Bennett et al..
 8                   2007). The lack of spirometric responsiveness was not attributable to the presence of
 9                   endogenous endorphins, which could potentially block C-fiber stimulation by O3, as
10                   demonstrated in a study involving intravenous administration of naloxone immediately
11                   following the O3 exposure (420 ppb, 2 hours, while exercising) to weak responders
12                   (Passannante et al.. 1998). Inflammation and other responses to O3 were also
13                   characterized by a large degree of interindividual variability. Currently, the mechanisms
14                   underlying this variability are not known. It has been proposed that some of the variation
15                   in responses may be genetically determined (Yang et al.. 2005a). The role  of gene-
16                   environment interactions, pre-existing diseases and conditions, nutritional status,
17                   lifestage, attenuation, and co-exposures in modulating responses to O3 are discussed
18                   below.
                     5.4.2.1   Gene-Environment Interactions

19                   The significant interindividual variation in responses to O3 infers that genetic background
20                   is an important determinant of susceptibility to O3 (Cho and Kleeberger. 2007;
21                   Kleeberger et al.. 1997) (see also Section 8.4). Strains of mice which are prone or
22                   resistant to O3-induced effects have been used to systematically identify candidate
23                   susceptibility genes. Genome wide linkage analyses (also known as positional cloning)
24                   demonstrated quantitative trait loci for O3-induced lung inflammation and
25                   hyperpermeability on chromosome 17 (Kleeberger et al.. 1997) and chromosome 4
26                   (Kleeberger et al.. 2000). respectively, using these recombinant inbred strains of mice and
27                   exposures to 0.3 ppm O3 for up to 72 hours. More specifically, these studies found that
28                   Tnf, whose protein product is the inflammatory cytokine TNF-a, and Tlr4, whose protein
29                   product is TLR4, were candidate susceptibility genes (Kleeberger et al.. 2000; Kleeberger
30                   etal.. 1997). Other studies, which used targeted deletion, identified genes encoding iNOS
31                   and heat shock proteins as TLR4 effector genes (Bauer etal.. 2011; Kleeberger et al..
32                   2001) and found that IL-10 protects against O3-induced pulmonary inflammation
33                   (Backus etal.. 2010). Investigations in inbred mouse strains found that differences in
34                   expression of certain proteins, such as CCSP (1.8 ppm O3 for 3 hours) (Broeckaert et al..

      Draft - Do Not Cite or Quote                       5-53                                September 2011

-------
 1                   2003) and MARCO (0.3 ppm O3 for up to 48 hours) (Dahl et al.. 2007). were responsible
 2                   for phenotypic characteristics, such as epithelial permeability and scavenging of oxidized
 3                   lipids, respectively, which confer sensitivity to O3.

 4                   Genetic polymorphisms have received increasing attention as modulators of O3-mediated
 5                   effects. Functionally relevant polymorphisms in candidate susceptibility genes have been
 6                   studied at the individual and population level in humans, and also in animal models.
 7                   Genes whose protein products are involved in antioxidant defense/oxidative stress and
 8                   xenobiotic metabolism, such as glutathione-S-transferase Ml (GSTM1) and
 9                   NADPH:quinone oxidoreductase 1 (NQO1), have also been a major focuses of these
10                   efforts. This is because oxidative stress resulting from O3 exposure is thought to
11                   contribute to the pathogenesis of asthma, and because xenobiotic metabolism detoxifies
12                   secondary oxidation products formed by O3 which contribute to oxidative stress (Islam et
13                   al.. 2008). TNF-a is of interest since it is linked to a candidate O3 susceptibility gene and
14                   since it plays a key role in initiating airways inflammation (Li et al.. 2006d).
15                   Polymorphisms of genes coding for GSTM1, NQO1 and TNF-a have been associated
16                   with altered susceptibility to O3-mediated effects (Li et al.. 2006d; Yang et al.. 2005a;
17                   Romieu et al.. 2004a: Corradi et al.. 2002; Bergamaschi et al.. 2001). Additional studies
18                   have focused on functional variants in other genes involved in antioxidant defense such
19                   as catalase (CAT), myeloperoxidase, heme oxygenase (HMOX-1) and manganese
20                   superoxide dismutase (MnSOD) (Wenten et al.. 2009; Islam et al.. 2008). These studies
21                   are discussed below.

22                   GSTM1 is a phase II antioxidant enzyme which is transcriptionally regulated by NF-E2-
23                   related factor 2-antioxidant response element (Nrf2-ARE) pathway. A large proportion
24                   (40-50%) of the general public (across ethnic populations) has the GSTMl-null genotype,
25                   which has been linked to an increased risk of health effects due to exposure to air
26                   pollutants (London. 2007). A role for GSTs in metabolizing electrophiles such as 4-
27                   hydroxynonenal, which is a secondary oxidation product formed following O3 exposure,
28                   has been demonstrated (Awasthi et al.. 2004). A recent study found that the GSTM1
29                   genotype modulated the time course of the neutrophilic inflammatory response following
30                   acute O3 exposure (400 ppb for 2 hours with intermittent exercise) in healthy adults
31                   (Alexis et al.. 2009). In GSTMl-null and -sufficient subjects, O3-induced sputum
32                   neutrophilia was similar at 4 hours. However, neutrophilia resolved by 24 hours in
33                   sufficient subjects but not in GSTMl-null subjects. In contrast, no differences in 24 hour
34                   sputum neutrophilia were observed between GSTMl-null and -sufficient human subjects
35                   exposed to 60 ppb O3 for 2 hours with intermittent exercise (KimetaL 2011). It is not
36                   known whether the effect seen at the higher exposure level (Alexis et al.. 2009) was due
37                   to the persistence of pro-inflammatory stimuli, impaired production of downregulators or
38                   impaired neutrophil apoptosis and clearance. However, a subsequent in vitro study by
      Draft - Do Not Cite or Quote                       5-54                                September 2011

-------
 1                  these same investigators found that GSTM1 deficiency in airway epithelial cells
 2                  enhanced IL-8 production in response to 0.4 ppm O3 for 4 hours (Wu etal.. 2011).
 3                  Furthermore, NF-KB activation was required for O3-induced IL-8 production (Wu et al..
 4                  2011). Since IL-8 is a potent neutrophil activator and chemotaxin, this study provides
 5                  additional evidence for the role of GSTM1 as a modulator of inflammatory responses due
 6                  to O3 exposure.

 7                   In addition, O3 exposure increased the expression of the surface marker CD 14 in airway
 8                  neutrophils of GSTMl-null subjects compared with sufficient subjects (Alexis et al..
 9                  2009). Furthermore, differences in airway macrophages were noted between the GSTM1-
10                  sufficient and -null subjects. Nnumbers of airway macrophages were decreased at 4 and
11                  24 hours following O3 exposure in GSTM1-sufficient subjects (Alexis et al.. 2009).
12                  Airway macrophages in GSTMl-null subjects were greater in number and found to have
13                  greater oxidative burst and phagocytic capability than those of sufficient subjects. Airway
14                  macrophages and dendritic cells from GSTMl-null subjects exposed to O3 expressed
15                  higher levels of the  surface marker HLA-DR, suggesting activation of the innate  immune
16                  system (Alexis et al., 2009). These differences in inflammatory responses between the
17                  GSTMl-null and -sufficient subjects may provide biological plausibility for the
18                  differences in O3-mediated effects reported in controlled human exposure studies
19                  (Corradi et al.. 2002; Bergamaschi etal.. 2001). It should also be noted that GSTM1
20                  genotype did not affect the acute pulmonary function (i.e. spirometric) response to O3
21                  which provides additional evidence for separate mechanisms underlying O3's effects on
22                  pulmonary function and inflammation in adults (Alexis et al.. 2009). However, GSTM1-
23                  null asthmatic children were previously found to be more at risk of O3-induced effects on
24                  pulmonary function than GSTM 1-sufficient asthmatic children (Romieu et al.. 2004a).

25                  Another enzyme involved in the metabolism of secondary oxidation products is NQO1.
26                  NQO1 catalyzes the 2-electron reduction by NADPH of quinones to hydroquinones.
27                  Depending on the substrate, it is capable of both protective detoxification reactions and
28                  redox cycling reactions resulting in the generation of reactive oxygen species. A recent
29                  study using NQO 1 -null mice demonstrated that NQO 1 contributes to O 3 -induced
30                  oxidative stress, AHR and inflammation following a 3-hour exposure to 1 ppm O3
31                  rVbynow et al.. 2009). These experimental results may provide biological plausibility for
32                  the increased biomarkers of oxidative stress and increased pulmonary function
33                  decrements observed in O3-exposed individuals bearing both the wild-type NQO1 gene
3 4                  and the null GSTM 1 gene (Corradi et al.. 2002; Bergamaschi etal.. 2001).

35                  Besides enzymes, other mechanisms participate in the removal of secondary oxidation
36                  products formed as a result of O3  inhalation. One involves scavenging of oxidized lipids
37                  via the macrophage receptor with collagenous structure (MARCO) expressed on  the cell
      Draft - Do Not Cite or Quote                      5-55                                September 2011

-------
 1                  surface of alveolar macrophages. A recent study demonstrated increased gene expression
 2                  of MARCO in the lungs of an O3-resistant C3H mouse strain (HeJ) but not in an O3-
 3                  sensitive, genetically similar strain (OuJ) (Dahl et al.. 2007). Upregulation of MARCO
 4                  occurred in mice exposed to 0.3 ppm O3 for 24-48 hours; inhalation exposure for 6 hours
 5                  at this concentration was insufficient for this response. Animals lacking the MARCO
 6                  receptor exhibited greater inflammation and injury, as measured by BAL neutrophils,
 7                  protein and isoprostanes, following exposure to 0.3 ppm O3 (Dahl et al.. 2007). MARCO
 8                  also protected against the inflammatory effects of oxidized surfactant lipids (Dahl et al..
 9                  2007). Scavenging of oxidized lipids may limit O3-induced injury since ozonized
10                  cholesterol species formed in the ELF (mice, 0.5-3 ppm O3, 3 hours) (Pulfer et al.. 2005;
11                  Pulfer and Murphy. 2004) stimulated apoptosis and cytotoxicity in vitro (Gao et al..
12                  2009b: Sathishkumar et al.. 2009; Sathishkumar et al.. 2007a: Sathishkumar et al..
13                  2007b).

14                  Two studies reported relationships between TNF promoter variants and O3-induced
15                  effects in humans. In one study, O3-induced change in lung function was significantly
16                  lower in adult subjects with TNF promoter variants -308A/A and -308G/A compared with
17                  adult subjects with the variant -308G/G (Yang et al.. 2005a). This response was
18                  modulated by a specific polymorphism of LTA (Yang et al.. 2005a). a previously
19                  identified candidate  susceptibility gene whose protein product is lymphotoxin-a
20                  (Kleeberger et al.. 1997). In the second study, an association between the TNF promoter
21                  variant -308G/G and decreased risk of asthma and lifetime wheezing in children was
22                  found (Li et al.. 2006d). The protective effect on wheezing was modulated by ambient O3
23                  levels and by GSTM1 and GSTP1 polymorphisms. The authors suggested that the
24                  TNF-308 G/G genotype may have a protective role in the development of childhood
25                  asthma (Li et al.. 2006d).

26                  Similarly,  a promoter variant of the gene HMOX-1, consisting of a smaller number of
27                  (GT)n repeats, was associated with a reduced risk for new-onset asthma in non-Hispanic
28                  white children (Islam et al.. 2008). The number of (GT)n repeats in this promoter has
29                  been shown to be inversely related to the inducibility of HMOX-1. A modulatory effect
30                  of O3 was demonstrated since the beneficial effects of this polymorphism were seen only
31                  in children living in  low O3 communities (Islam et al.. 2008). This study also identified
32                  an association between a polymorphism of the CAT gene and increased risk of new-onset
33                  asthma in Hispanic children; however no modulation by O3 was seen (Islam et al.. 2008).
34                  No association was observed in this  study between a MnSOD polymorphism and asthma
35                  (Mam et al.. 2008).
      Draft - Do Not Cite or Quote                      5-56                                September 2011

-------
 1                   Studies to date indicate that some variability in individual responsiveness to O3 may be
 2                   accounted for by functional genetic polymorphisms. Further, the effects of gene-
 3                   environment interactions may be different in children and adults.
                     5.4.2.2   Pre-existing Diseases and Conditions

 4                   Pre-existing diseases and conditions can alter the response to O3 exposure. For example,
 5                   responsiveness to O3, as measured by spirometry, is decreased in smokers and individuals
 6                   with COPD (U.S. EPA. 2006b). Asthma and allergic diseases are of major importance in
 7                   this discussion. In individuals with asthma, there is increased responsiveness to
 8                   bronchoconstrictor challenge. This results from a combination of structural and
 9                   physiological factors including increased airway inner-wall thickness, smooth muscle
10                   responsiveness and mucus secretion. Although inflammation is likely to contribute, its
11                   relationship to AHR is not clear (U.S. EPA. 2006b). However, some asthmatics have
12                   higher baseline levels of neutrophils, lymphocytes, eosinophils and mast cells in
13                   bronchial washes and bronchial biopsy tissue (Stenfors et al.. 2002). It has been proposed
14                   that enhanced sensitivity to O3 is conferred by the presence of greater numbers of
15                   resident airway inflammatory cells in disease states such as asthma (Mudway and Kelly.
16                   2000).

17                   In order to determine whether asthmatics exhibit greater responses to O3, several older
18                   studies compared pulmonary function in asthmatic and non-asthmatic subjects following
19                   O3 exposure. Some also probed mechanisms which could account for enhanced
20                   sensitivity. While the majority focused on measurements of FEVi and FVC and found no
21                   differences between the two groups following exposures of 2-4 hours to 125-250 ppb O3
22                   or to a 30-minute exposure to 120-180 ppb O3 by mouthpiece while exercising (Stenfors
23                   et al.. 2002; Mudwavetal.. 2001; Holz et al.. 1999; Scannell et al.. 1996; Koenig et al..
24                   1987; Linn et al.. 1978). there were notable exceptions. In one study, greater airway
25                   obstruction in asthmatics compared with non-asthmatic subjects was observed
26                   immediately following a 2-hour exposure to 400 ppb O3 with  intermittent exercise (Kreit
27                   etal.. 1989). These changes were measured as statistically significant greater decreases in
28                   FEVi and in FEF 25.75 (but not in FVC) in the absence of a bronchoconstrictor challenge
29                   (Kreit et al.. 1989). These results suggest that this group of asthmatics responded to
30                   O3-exposure with a greater degree of vagally-mediated bronchoconstriction compared
31                   with the non-asthmatics. A second study demonstrated a statistically significant greater
32                   decrease in FEVi and in FEVi/FVC (but not in FVC) in asthmatics compared with non-
33                   asthmatics exposed to 160 ppb O3 for 7.6 hours  with light exercise (Horstman et  al..
34                   1995). These responses were accompanied by wheezing and inhaler use in the asthmatics
35                   (Horstman et al.. 1995). Aerosol bolus dispersion measurements demonstrated a


      Draft - Do Not Cite or Quote                      5-57                               September 2011

-------
 1                   statistically significant greater change in asthmatics compared with non-asthmatics,
 2                   which was suggestive of O3-induced small airway dysfunction (Horstman et al.. 1995).
 3                   Furthermore, a statistically significant correlation was observed between the degree of
 4                   baseline airway status and the FEVi response to O3 in the asthmatic subjects (Horstman
 5                   et al., 1995). A third study found similar decreases in FVC and FEVi in both asthmatics
 6                   and non-asthmatics exposed to 400 ppb O3 for 2 hours with mild exercise (Alexis et al..
 7                   2000). However, a statistically significant decrease in FEF75, a measure of small airway
 8                   function, was observed in asthmatics but not in non-asthmatics (Alexis et al.. 2000).
 9                   Taken together, these latter studies indicate that while the magnitude of restrictive type
10                   spirometric decline was similar in asthmatics and non-asthmatics, that obstructive type
11                   changes (i.e. bronchoconstriction) were greater in asthmatics. Further, asthmatics
12                   exhibited greater sensitivity to O3 in terms of small airways function.

13                   Since asthma exacerbations occur in response to allergens and/or other triggers, some
14                   studies have focused on O3-induced changes in AHR following a bronchoconstrictor
15                   challenge. No difference in sensitivity to methacholine bronchoprovocation was observed
16                   between asthmatics and non-asthmatics exposed to 400 ppb O3 for 2 hours with moderate
17                   exercise (Kreit et al.. 1989). However, increased bronchial reactivity to inhaled allergens
18                   was demonstrated in mild allergic asthmatics exposed to 160 ppb for 7.6 hours, 250 ppb
19                   for 3 hours and 120 ppb for 1 hour while exercising (Kehrl et al.. 1999; Torres et al..
20                   1996; Molfmo et al.. 1991) and in allergen-sensitized guinea pigs following O3 exposure
21                   (1 ppm, 1 hour) (Sun et al.. 1997). Similar, but modest, responses were reported for
22                   individuals with allergic rhinitis (Torres et al.. 1996). Further, the contractile response of
23                   isolated airways from human donor lung tissue, which were sensitized and challenged
24                   with allergen, was increased by pre-exposure to  1 ppm O3 for 20 (Rouxet al.. 1999).
25                   These studies provide support for O3-mediated enhancement of responses to allergens in
26                   allergic subjects.

27                   In terms of airways neutrophilia, larger responses were observed in asthmatics compared
28                   to non-asthmatics subjects exposed to O3 in some (Balmes etal.. 1997; Scannell et al..
29                   1996; Bashaetal.. 1994) but not all (Mudwavet al.. 2001) of the older studies. While
30                   each of these studies involved exposure of exercising human subjects to 200 ppb O3, the
31                   duration of exposure was longer (i.e. 4-6 hours) in the former studies than in the latter
32                   study (2 hours). Further, statistically significantly increases in myeloperoxidase levels (an
33                   indicator of neutrophil activation) in bronchial washes was observed in mild asthmatics
34                   compared with non-asthmatics, despite no difference in O3-stimulated neutrophil influx
35                   between the  2 groups following exposure to 200 ppb O3 for 2 hours with mild exercise
36                   (Stenfors et al.. 2002). A more recent study found that atopic asthmatic subjects exhibited
37                   an enhanced inflammatory  response to O3 (400 ppb, 4 hours, with exercise) (Hernandez
38                   etal.. 2010). This response was characterized by greater numbers of neutrophils, higher
      Draft - Do Not Cite or Quote                       5-58                                September 2011

-------
 1                   levels of IL-6, IL-8 and IL-1(3 and greater macrophage cell-surface expression of TLR4
 2                   and IgE receptors in induced sputum compared with healthy subjects. This study also
 3                   reported a greater increase in hyaluronan in atopic subjects and atopic asthmatics
 4                   compared with healthy subjects following O3 exposure. Animal studies have previously
 5                   reported that hyaluronic acid activates TLR4 signaling and results in AHR (see Section
 6                   5.3.5). Furthermore, levels of IL-10, a potent anti-inflammatory cytokine, were greatly
 7                   reduced in atopic asthmatics compared to healthy subjects. These results provide
 8                   evidence that innate immune and adaptive responses are different in asthmatics and
 9                   healthy subjects exposed to O3.

10                   Eosinophils may be an important modulator of responses to O3 in asthma and allergic
11                   airways disease. Eosinophils and associated proteins are thought to affect muscarinic
12                   cholinergic receptors which are involved in vagally-mediated bronchoconstriction
13                   (Mudway and Kelly. 2000). Studies described in Section 5.3.5 which demonstrated a key
14                   role of eosinophils in O3-mediated AHR may be relevant to human allergic airways
15                   disease which is characterized by airways eosinophilia (Yost et al.. 2005). Furthermore,
16                   O3 exposure sometimes results in airways eosinophilia in allergic subjects or animal
17                   models. For example, eosinophilia of the nasal and other airways was observed in
18                   individuals with pre-existing allergic disease following O3 inhalation (270 and 400 ppb
19                   O3, 2 hours, with exercise) (Vagaggini et al.. 2002; Peden etal.. 1995). Further, O3
20                   exposure (0.5 ppm, 8 hours/day for 1-3 days) increased allergic responses, such as
21                   eosinophilia and augmented intraepithelial mucosubstances, in the nasal airways of
22                   ovalbumin (OVA)-sensitized rats (Wagner et al.. 2002). In contrast, Stenfors (2002) found
23                   no stimulation of eosinophil influx measured in bronchial washes and BALF of mild
24                   asthmatics following exposure to a lower concentration (200 ppb, 2 hours, with exercise)
25                   ofO3.

26                   The role of mast cells in O3-mediated asthma exacerbations has been investigated. Mast
27                   cells are thought to play a key role in O3-induced airways inflammation, since airways
28                   neutrophilia was decreased in mast cell-deficient mice exposed to  O3 (Kleeberger et al..
29                   1993). However, another study found that mast cells were not involved in the
30                   development of increased bronchial reactivity in O3-exposed mice (Noviski et al.. 1999).
31                   Nonetheless, mast cells release a wide variety  of important inflammatory mediators
32                   which may lead to asthma exacerbations (Stenfors et al.. 2002). A large increase in mast
33                   cell number in bronchial  submucosa was observed in non-asthmatics and a significant
34                   decrease in mast cell number in bronchial epithelium was observed in mild asthmatics 6
35                   hours following exposure to 200 ppb O3 for 2  hours during mild exercise (Stenfors et al..
36                   2002). While these results point to an O3-mediated flux in bronchial  mast cell
37                   populations which differed between the non-asthmatics and mild asthmatics,
      Draft - Do Not Cite or Quote                       5-59                                 September 2011

-------
 1                   interpretation of these findings is difficult. Furthermore, mast cell number did not change
 2                   in airway lavages in either group in response to O3 (Stenfors et al.. 2002)

 3                   Cytokine profiles in the airways have been investigated as an indicator of O3 sensitivity.
 4                   Differences in epithelial cytokine expression were observed in bronchial biopsy samples
 5                   in non-asthmatic and asthmatic subjects both at baseline and 6-hours postexposure to 200
 6                   ppb O3 for 2 hours (Bosson et al., 2003). The asthmatic subjects had a higher baseline
 7                   expression of IL-4 and IL-5 compared to non-asthmatics. In addition, expression of IL-5,
 8                   IL-8, GM-CSF, and ENA-78 in asthmatics was increased significantly following O3
 9                   exposure compared to non-asthmatics (Bosson et al.. 2003). Some of these (IL-4, IL-5
10                   and GM-CSF) are Th2-related cytokines or neutrophil chemoattractants, and play a role
11                   in IgE production, airways eosinophilia and suppression of Thl-cytokine production
12                   (Bosson et al., 2003). These findings suggest a link between adaptive immunity and
13                   enhanced responses of asthmatics to O3.

14                   A further consideration is the compromised status of ELF antioxidants in disease states
15                   such as asthma (Mudway and Kelly. 2000). This could possibly be due to ongoing
16                   inflammation which causes antioxidant depletion or to abnormal  antioxidant transport or
17                   synthesis (Mudway and Kelly. 2000). For example, basal levels of AH2 were
18                   significantly lower and basal levels of oxidized GSH and UA were significantly higher in
19                   bronchial wash fluid and BALF of mild asthmatics compared with healthy control
20                   subjects (Mudway et al.. 2001). Differences in ELF antioxidant content have also been
21                   noted between species. These observations have led to the suggestion that the amount and
22                   composition of ELF antioxidants, the capacity to replenish antioxidants in the ELF or the
23                   balance between beneficial and injurious interactions between antioxidants and O3 may
24                   contribute to O3 sensitivity, which varies between individuals and species (Mudway et
25                   al.. 2006: Mudwav and Kelly. 2000: Mudwav et al.. 1999a). The complexity of these
26                   interactions was demonstrated by a study in which a 2-hour exposure to 200 ppb O3,
27                   while exercising,  resulted in similar increases in airway neutrophils and decreases in
28                   pulmonary function in both mild asthmatics and healthy controls, despite differences in
29                   ELF antioxidant concentrations prior to O3 exposure (Mudway et al.. 2001). Further, the
30                   O3-induced increase in oxidized GSH and decrease in AH2 observed in ELF of healthy
31                   controls was not observed in mild asthmatics (Mudway et al.. 2001). While the authors
32                   concluded that basal AH2 and oxidized GSH concentrations were not predictive of
33                   responsiveness to O3, they also suggested that the increased basal UA concentrations in
34                   the mild asthmatics may have played a protective role (Mudwav et al., 2001). Thus
3 5                   compensatory mechanisms resulting in enhanced total antioxidant capacity may play a
36                   role in modulating responses to O3.
      Draft - Do Not Cite or Quote                      5-60                                September 2011

-------
 1                   Collectively these older and more recent studies provide insight into mechanisms which
 2                   may contribute to enhanced responses of asthmatic and atopic individuals following O3
 3                   exposure. Greater airways inflammation and/or greater bronchial reactivity have been
 4                   demonstrated in asthmatics compared to non-asthmatics. This pre-existing inflammation
 5                   and altered baseline bronchial reactivity may contribute to the enhanced
 6                   bronchoconstriction seen in asthmatics exposed to O3.  Furthermore, O3-induced
 7                   inflammation may contribute to O3-mediated AHR. An enhanced neutrophilic response
 8                   has been demonstrated in some asthmatics. A recent study in humans provided evidence
 9                   for differences in innate immune responses related to TLR4 signaling between asthmatics
10                   and healthy subjects. Animal studies have demonstrated a role for eosinophil-derived
11                   proteins in mediating the effects of O3. Since airways eosinophilia occurs in both allergic
12                   humans and allergic animal models, this pathway may  underlie the exacerbation of
13                   allergic asthma by O3. In addition, differences have been noted in epithelial cytokine
14                   expression in bronchial biopsy samples of healthy and  asthmatic subjects. A Th2
15                   phenotype, indicative of adaptive immune system activation and enhanced allergic
16                   responses, was observed before O3 exposure and was increased by O3 exposure in
17                   asthmatics. These findings support links between innate and adaptive immunity and
18                   sensitivity to O3-mediated effects in asthmatics and allergic airways disease.

19                   In addition to asthma and allergic diseases, obesity may alter responses  to O3. While O3
20                   is a trigger for asthma, obesity is a known risk factor for asthma (Shore. 2007). The
21                   relationship between obesity and asthma is not well understood but recent investigations
22                   have focused on alterations in endocrine function of adipose tissue in obesity. It is
23                   thought that the increases in serum levels of factors produced by adipocytes (i.e.
24                   adipokines) such as cytokines, chemokines, soluble cytokine  receptors and energy
25                   regulating hormones, may contribute to the relationship between obesity and  asthma.
26                   Some of these same mechanisms may be relevant to insulin resistant states such as
27                   metabolic syndrome.

28                   In a reanalysis of the data of Hazucha  (2003). increasing body mass index in young
29                   women was associated with increased O3 responsiveness, as measured by spirometry
30                   following a 2-hour exposure to 500 ppb O3 while exercising (Bennett et al., 2007). In
31                   several mouse models of obesity, airways were found to be innately more
32                   hyperresponsive and responded more vigorously to acute O3 exposure than lean controls
33                   (Shore. 2007). Pulmonary inflammatory and injury in response to O3 were also enhanced
34                   (Shore. 2007). It was postulated that oxidative stress resulting from obesity-related
35                   hyperglycemia could account  for these effects (Shore. 2007). However,  responses to O3
36                   in the different mouse models are somewhat variable and depend on whether exposures
37                   are acute or subacute. For example, diet-induced obesity augmented inflammation and
3 8                   injury, as measured by BALF  markers, and enhanced AHR in mice exposed acutely to O3
      Draft - Do Not Cite or Quote                       5-61                                September 2011

-------
 1                   (2 ppm, 3 hours) (Johnston et al.. 2008). In contrast, the inflammatory response following
 2                   sub-acute exposure to O3 was dampened by obesity in a different mouse model (0.3 ppm,
 3                   72 hours) (Shore et al.. 2009).
                     5.4.2.3   Nutritional Status

 4                   Many investigations have focused on antioxidant deficiency and supplementation as
 5                   modulators of O3-mediated effects. One study in mice found that vitamin A deficiency
 6                   enhanced lung injury induced by exposure to 0.3 ppm O3 for 72 hours (Paquette et al..
 7                   1996). Ascorbate deficiency was shown to increase the effects of acute (0.5-1 ppm for 4
 8                   hours), but not subacute (0.2-0.8 ppm for 7 days), O3 exposure in guinea pigs (Kodavanti
 9                   et al.. 1995; Slade et al.. 1989). Supplementation with AH2 and a-TOH was protective in
10                   healthy adults who were on an AH2-deficient diet and exposed to 400 ppb O3 for 2 hours
11                   while exercising (Samet etal.. 2001). In this study, the protective effect consisted of a
12                   smaller reduction in FEVi following O3 exposure (Samet et al.. 2001). However the
13                   inflammatory response (influx of neutrophils and levels of IL-6) measured in BALF 1
14                   hour after O3 exposure was not different between supplemented and non-supplemented
15                   subj ects (Samet etal.. 2001). Other investigators found that AH2 and a-TOH
16                   supplementation failed to ameliorate the pulmonary function decrements or airways
17                   neutrophilia observed in humans exposed to 200 ppb O3 for 2 hours (Mudway et al..
18                   2006). It was suggested that supplementation may be ineffective in the absence of
19                   antioxidant deficiency (Mudway et al.. 2006).

20                   In asthmatic  adults, these same dietary antioxidants reduced O3-induced bronchial
21                   hyperresponsiveness  (120 ppb, 45 min, with exercise) (Trenga et al.. 2001). Furthermore,
22                   supplementation with AH2 and a-tocopherol protected against pulmonary function
23                   decrements and nasal inflammatory responses which were associated with high levels of
24                   ambient O3 in asthmatic children living in Mexico City (Sienra-Monge et al.. 2004;
25                   Romieu et al.. 2002). Similarly, supplementation with ascorbate, a-tocopherol and
26                   (3-carotene improved  pulmonary function in Mexico City street workers (Romieu et al..
27                   1998a).

28                   Protective effects of supplementation with a-tocopherol alone have not been observed in
29                   humans experimentally exposed to O3 (Mudway and Kelly. 2000). Alpha-TOH
30                   supplementation also failed to protect against O3-induced effects in animal models of
31                   allergic rhinosinusitis and lower airways allergic inflammation (rats, 1 ppm O3 for 2
32                   days) (Wagner et al..  2007). However, protection in these same animal models was
33                   reported using y-TOH supplementation (Wagner et al.. 2009; Wagner et al.. 2007). Other
34                   investigators  found that a-TOH deficiency led to an increase in liver lipid peroxidation
      Draft - Do Not Cite or Quote                      5-62                                September 2011

-------
 1                   (rats, 0.3 ppm 3 hours/day for 7 months) (Sato et al., 1980) and a drop in liver a-TOH
 2                   levels following O3 exposure (mice, 0.5 ppm, 6 hours/day for 3 days) (Vasu etal.. 2010).
 3                   A recent study used a-TOH transfer protein null mice as a model of a-TOH deficiency
 4                   and demonstrated an altered adaptive response of the lung genome to O3 exposure (Vasu
 5                   etal., 2010). Taken together, these studies provide evidence that the tocopherol system
 6                   modulates Os-induced responses.
                     5.4.2.4    Lifestage

 7                   Responses to O3 are modulated by factors associated with lifestage. On one end of the
 8                   lifestage spectrum is aging. The spirometric response to O3 appears to be lost in humans
 9                   as they age, as was demonstrated in two studies involving exposures of exercising human
10                   subjects to 420-450 ppb O3 for 1.5-2 hours (Hazucha et al., 2003; Drechsler-Parks.
11                   1995). In mice, physiological responses to O3 (600 ppb, 2 hours) were diminished with
12                   age (Hamade et al., 2010). Mechanisms accounting for this effect have not been well-
13                   studied but could include altered number and sensitivity of receptors or altered signaling
14                   pathways involved in neural reflexes.

15                   On the other side of the lifestage spectrum is pre/postnatal development. Critical
16                   windows of development during the pre/postnatal period are associated with an enhanced
17                   sensitivity to environmental toxicants. Adverse birth outcomes and developmental
18                   disorders may occur as a result.

19                   Adverse birth outcomes may result from stressors which  impact transplacental oxygen
20                   and nutrient transport by a variety of mechanisms including oxidative stress, placental
21                   inflammation and placental vascular dysfunction (Kannan et al. 2006). These
22                   mechanisms may be linked since oxidative/ nitrosative stress is reported to cause vascular
23                   dysfunction in the placenta (Myatt et al.. 2000). As described in Section 7.4, systemic
24                   inflammation and oxidative/nitrosative stress and modification of innate and adaptive
25                   immunity are key events underlying the health effects of O3 and as such they may
26                   contribute to adverse birth outcomes. An animal toxicology study showing that exposure
27                   to 2 ppm O3 led to anorexia (Kavlock et al., 1979) (see Section 7.4.2) in exposed rat
28                   dams provide an additional mechanism by which O3 exposure could lead to diminished
29                   transplacental nutrient transport. Disturbances of the pituitary-adrenocortico-placental
30                   system (Ritz et al.. 2000) may also impact normal intrauterine growth and development.
31                   Further, restricted fetal growth may result from pro-inflammatory cytokines which limit
32                   trophoblast invasion during the early stages of pregnancy (Hansen et al.. 2008). Direct
33                   effects on maternal health, such as susceptibility to infection, and on fetal health, such as
34                   DNA damage, have also been proposed as mechanisms underlying adverse birth
      Draft - Do Not Cite or Quote                       5-63                                 September 2011

-------
 1                   outcomes (Ritz et al., 2000). In addition to restricted fetal growth, preterm birth may
 2                   contribute to adverse birth outcomes. Preterm birth may result from the development of
 3                   premature contractions and/or premature rupture of membranes as well as from disrupted
 4                   implantation and placentation which results in suboptimal placental function (Darrow et
 5                   al., 2009; Ritz et al., 2000). Genetic mutations are thought to be an important cause of
 6                   placental abnormalities in the first trimester, while vascular alterations may be the main
 7                   cause of placental abnormalities in later trimesters (Jalaludin et al.. 2007). Ozone-
 8                   mediated systemic inflammation and oxidative stress/nitrosative stress may possibly be
 9                   related to these effects although there is no firm evidence.

10                   Enhanced sensitivity to environmental toxicants during critical windows of development
11                   may also result in developmental disorders. For example, normal migration and
12                   differentiation of neural crest cells are important for heart development and are
13                   particularly sensitive to toxic insults (Ritz et al.. 2002). Further, immune dysregulation
14                   and related pathologies are known to be associated with pre/postnatal environmental
15                   exposures (Dietert et al.. 2010). Ozone exposure is associated with developmental effects
16                   in several organ systems. These include neurobehavioral changes which could reflect
17                   O3 's effects on CNS plasticity or the hypothalamic-pituitary axis (Auten and Foster. In
18                   Press) (see Section 7.4.9).

19                   The majority of developmental effects due to O3 have been described for the respiratory
20                   system (see Section 7.2.3 and 7.4.8). Since its growth and development take place during
21                   both the prenatal and early postnatal periods, both prenatal and postnatal exposures to O3
22                   have been studied. Maternal exposure to 0.4-1.2 ppm O3 during gestation resulted in
23                   developmental health effects in the RT of mice (Sharkhuu et al.. 2011). Recent studies
24                   involving postnatal exposure to O3 have focused on differences between developing and
25                   adult animals in antioxidant defenses, respiratory physiology and sensitivity to cellular
26                   injury (Auten and Foster. In Press). In particular, one  set of studies in infant rhesus
27                   monkeys exposed to 0.5  ppm O3 intermittently over 5 months has identified numerous
28                   O3-mediated perturbations in the developing lung and immune system (Plopperetal..
29                   2007). These investigations were prompted by the dramatic rise in the incidence of
30                   childhood asthma and focused on the possible role of O3 and allergens in promoting
31                   remodeling of the epithelial-mesenchymal trophic unit during postnatal development of
32                   the tracheobronchial airway wall. These and other studies have focused on mechanisms,
33                   such as lung structural changes, antigen sensitization, interaction with nitric oxide
34                   signaling, altered airway afferent innervation and loss of alveolar repair capacity, by
35                   which early O3 exposure could lead to  asthma pathogenesis or exacerbations in later life
36                   (Auten and Foster. In Press). Further, a recent study demonstrated that maternal exposure
37                   to particulate matter (PM) resulted in augmented lung inflammation, airway epithelial
3 8                   mucous metaplasia and AHR in young  mice exposed  chronically and intermittently to 1
      Draft - Do Not Cite or Quote                       5-64                                 September 2011

-------
 1                   ppm O3 (Auten et al.. 2009). Early life exposure to O3 has also been found to modulate
 2                   pulmonary and systemic innate immunity later in life in the infant rhesus monkey model
 3                   (Maniar-Hew et al.. 2011).
                     5.4.2.5   Attenuation of Responses

 4                   In responsive individuals, a striking degree of response attenuation occurred following
 5                   repeated daily exposures to O3. Generally, the young O3 responder was no longer
 6                   responsive on the fourth or fifth day of consecutive daily O3 exposure (200-500 ppb O3
 7                   for 2-4 hours) and required days to weeks of non-exposure in order for the subject to
 8                   regain O3 responsiveness (Christian et al.. 1998; Devlin et al.. 1997; Linn et al.. 1982a;
 9                   Horvath et al.. 1981; Hackney et al.. 1977). This phenomena has been reported for both
10                   lung function and symptoms such as upper airway irritation, nonproductive cough,
11                   substernal discomfort and pain upon deep inspiration (Linnet al.. 1982a; Horvath et al..
12                   1981; Hackney et al.. 1977). Repeated daily exposures also led to an attenuation of the
13                   sRaw response in exercising human subjects exposed for 4 hours to 200 ppb O3
14                   (Christian et al.. 1998) and to a dampened AHR response compared with a single day
15                   exposure in exercising human subjects exposed for 2 hours to 400 ppb O3 (Dimeo et al..
16                   1981). However, one group reported persistent small airway dysfunction despite
17                   attenuation of the FEVi response on the third day of consecutive O3 exposure (250 ppb,
18                   2 hours, with exercise) (Frank et al.. 2001).

19                   Studies in rodents also indicated an attenuation of the physiologic response measured by
20                   breathing patterns and tidal volume following five consecutive days of exposure to 0.35-1
21                   ppm O3 for 2.25 hours (Tepper et al..  1989). Attenuation of O3-induced bradycardic
22                   responses, which also result from activation of neural reflexes, has been reported in
23                   rodents (0.5-0.6 ppm O3, 2-6 h/dy, 3-5 days (Hamade and Tankersley. 2009; Watkinson et
24                   al.. 2001).

25                   Multi-day exposure to O3 has been found to decrease some markers of inflammation
26                   compared with a single day exposure (Christian et al.. 1998; Devlin et al.. 1997). For
27                   example, in human subjects exposed for 4 hours to 200 ppb O3 during moderate exercise,
28                   decreased numbers of BAL neutrophils and decreased levels of BALF fibronectin and IL-
29                   6 were observed after 4 days of consecutive exposure compared with responses after 1
30                   day (Christian et al.. 1998). Results indicated an attenuation of the inflammatory response
31                   in both proximal airways and distal lung. However markers of injury, such as lactate
32                   dehydrogenase (LDH) and protein in the BALF, were not attenuated in this study
33                   (Christian et al.. 1998). Other investigators found that repeated O3 exposure (200 ppb O3
34                   for 4 hours on 4 consecutive days with intermittent exercise) resulted in increased
      Draft - Do Not Cite or Quote                      5-65                                September 2011

-------
 1                   numbers of neutrophils in bronchial mucosal biopsies despite decreased BAL
 2                   neutrophilia (Torres et al.. 2000). Other markers of inflammation, including BALF protein
 3                   and IL-6 remained elevated following the multi-day exposure (Torres et al.. 2000).

 4                   In rats, the increases in BALF levels of proteins, fibronectin, IL-6 and inflammatory cells
 5                   observed after one day of exposure to 0.4 ppm O3 for 12 hours were no longer observed
 6                   after 5 consecutive days of exposure (Van Bree et al., 2002). A separate study in rats
 7                   exposed to 0.35-1 ppm O3 for 2.25 hours for 5 consecutive days demonstrated a lack of
 8                   attenuation of the increase in BALF protein, persistence of macrophages in the
 9                   centriacinar region and histological evidence of progressive tissue injury (Tepper et al..
10                   1989). Findings that injury, measured by BALF markers or by histopathology, persist in
11                   the absence of BAL neutrophila or pulmonary function decrements suggested that
12                   repeated exposure to O3 may have serious long-term consequences such as airway
13                   remodeling. In particular, the small airways were identified as a site where cumulative
14                   injury may occur (Frank et al., 2001).

15                   Some studies examined the recovery of responses which were attenuated by repeated O3
16                   exposure. In a study of humans undergoing heavy intermittent exercise who were
17                   exposed for 2 hours to 400 ppb O3 for  five consecutive days (Devlin et al.. 1997).
18                   recovery of the inflammatory responses which were diminished by repeated exposure
19                   required 10-20 days following the exposure (Devlin et al.. 1997). In an animal study
20                   conducted in parallel (Van Bree et al.. 2002). full susceptibility to O3 challenge following
21                   exposure to O3 for five consecutive days required 15-20 days recovery.

22                   Several mechanisms have been postulated to explain the attenuation of responses
23                   observed in human subjects and animal models following repeated exposure to O3. First,
24                   the upregulation of antioxidant defenses (or conversely, a decrease in critical O3-reactive
25                   substrates) may protect against O3-mediated adverse effects. Increases in antioxidant
26                   content of the BALF have been demonstrated in rats exposed to 0.25 and 0.5 ppm O3 for
27                   several hours on consecutive days (Devlin et al.. 1997; Wiester et al.. 1996a; Tepper et
28                   al.. 1989). Second, IL-6 was demonstrated to be  an important mediator of attenuation in
29                   rats exposed to 0.5 ppm for 4 hours on two consecutive days (McKinney et al.. 1998).
30                   Third, a protective role for increases in mucus producing cells and mucus concentrations
31                   in the airways has been proposed (Devlin et al.. 1997). Fourth, epithelial hyperplasia or
32                   metaplasia may decrease susceptibility to subsequent O3 challenge (Carey et al.. 2007;
33                   Harkema et al.. 1987a; Harkema et al.. 1987b). These morphologic changes have been
34                   observed in nasal and lower airways in monkeys exposed chronically to 0.15-0.5 ppm O3.
35                   Although there is some evidence to support these possibilities, there is no consensus on
36                   mechanisms underlying response attenuation. Recent studies demonstrating that O3
      Draft - Do Not Cite or Quote                       5-66                                 September 2011

-------
 1                  activates TRP receptors suggest that modulation of TRP receptor number or sensitivity by
 2                  repeated O3 exposures may also contribute to the attenuation of responses.
                    5.4.2.6   Co-Exposures with Particulate Matter

 3                  Numerous studies have investigated the effects of co-exposure to O3 and PM because of
 4                  the prevalence of these pollutants in ambient air. Results are highly variable and depend
 5                  on whether exposures are simultaneous or sequential, the type of PM employed and the
 6                  endpoint examined. Additive and interactive effects have been demonstrated. For
 7                  example, simultaneous exposure to O3 (120 ppb for 2 hours at rest) and concentrated
 8                  ambient particles (CAPs) in human subjects resulted in a diminished systemic IL-6
 9                  response compared with  exposure to CAPs alone (Urch et al.. 2010). However, exposure
10                  to O3 alone did not alter  blood IL-6 levels (Urchet al., 2010). The authors provided
11                  evidence that O3 mediated a switch to shallow breathing which may have accounted for
12                  the observed antagonism (Urchet al.. 2010). Further, simultaneous exposure to O3 (114
13                  ppb for 2 hours at rest) and CAPs but not exposure to either alone, resulted in increased
14                  diastolic blood pressure in human subjects (Fakhri et al., 2009). Mechanisms underlying
15                  this potentiation of response were not explored. In some strains of mice, pre-exposure to
16                  O3 (0.5 ppm for 2 hours) modulated the effects of carbon black PM on heart rate, HRV
17                  and breathing patterns (Hamade and Tankersley. 2009). Another recent study in mice
18                  demonstrated that treatment with carbon nanotubes followed 12 hours later by O3
19                  exposure (0.5 ppm for 3  hours) resulted in a dampening of some of the pulmonary effects
20                  of carbon nanotubes measured as markers of inflammation and injury in the BALF (Han
21                  et al.. 2008). Lastly, Harkema et  al. (2005) found that epithelial and inflammatory
22                  responses in the airways  of rats were enhanced by co-exposure to O3 (0.5 ppm for 3 days)
23                  and LPS (used as a model of biogenic PM) or to O3 (1 ppm for 2 days) and OVA (used  as
24                  a model of an aeroallergen). Many of the demonstrated responses were more-than-
25                  additive. Overall, these findings are hard to interpret but demonstrate the complexity of
26                  responses following combined exposure to PM and O3.
                    5.4.2.7   Summary

27                  Collectively, these older and more recent studies provide evidence for mechanisms which
28                  may underlie the variability in responsiveness seen among individuals (Figure 5-10).
29                  Certain functional genetic polymorphisms, pre-existing conditions and diseases,
30                  nutritional status, lifestage and co-exposures contribute to altered risk of O3-induced
31                  effects. Attenuation of responses may also be important, but it is incompletely
32                  understood, both in terms of the pathways involved and the resulting consequences.

      Draft - Do Not Cite or Quote                      5-67                                September 2011

-------
              Dosimetric factors

              Nutritional status
              Life stage
              Attenuation factors
              Co-exposures
                                        Ozone + Respiratory Tract
                                               Gene-environment interactions
                                               Pre-existing diseases/conditions
                                                    COPD/smoking status
                                                 Asthma/allergic airways disease
                                                 Obesity/metabolic syndrome
                    \
                                  Formation of secondary oxidation products
            Activation
            of neural
            reflexes
Initiation of
inflammation
                      Y7
               Systemic inflammation and
               oxidative/nitrosative stress
Sensitization
of bronchial
smooth muscle
                                                           Respiratory System Effects
                Extrapulmonary Effects
                      Obesity/
                    Metabolic Stress
                      Lifestage
                                                           Attenuation
                                                             factors
      Figure 5-10   Factors which contribute to the interindividual variability in
                      responses resulting from inhalation exposure to ozone.
          5.5    Species Homology and Interspecies Sensitivity

 1                  The previous O3 AQCDs discussed the suitability of animal models for comparison with
 2                  human O3 exposure and concluded that the acute and chronic functional responses of
 3                  laboratory animals to O3 appear qualitatively homologous to human responses. Thus,
 4                  animal studies can provide important data in determining cause-effect relationships
 5                  between exposure and health outcome that would be impossible to collect in human
 6                  studies. Still, care must be taken when comparing quantitative dose-response
 7                  relationships in animal models to  humans due to obvious interspecies differences. This
 8                  section will describe basic concepts in species homology concerning both dose and
 9                  response to O3 exposure. This will not be a quantitative extrapolation of doses where O3
10                  effects have been observed. Overall, there have been few new publications examining
11                  interspecies differences in dosimetry and response to O3 since the last AQCD. These
12                  studies do not overtly change the  conclusions discussed in the previous document.
      Draft - Do Not Cite or Quote
                           5-68
                                  September 2011

-------
             5.5.1   Dosimetry

 1                   As discussed in Section 5.2.1, O3 uptake depends on complex interactions between RT
 2                   morphology, breathing route, rate, and depth, physicochemical properties of the gas,
 3                   physical processes of gas transport, as well as the physical and chemical properties of the
 4                   ELF and tissue layers. Understanding differences in these variables between humans and
 5                   experimental animals is important to interpreting delivered doses in animal and human
 6                   toxicology studies.

 7                   Physiological and anatomical differences exist between experimental species. The
 8                   structure of the URT is vastly different between rodents and humans and scales according
 9                   to body mass. The difference in the cross-sectional shape and size of the nasal passages
10                   affects bulk airflow patterns such that major airflow streams are created. The nasal
11                   epithelium is lined by squamous, respiratory, or olfactory cells, depending on location.
12                   The differences in airflow patterns in the URT mean that not all nasal surfaces and cell
13                   types receive the  same exposure to inhaled O3 leading to differences in local absorption
14                   and potential for site-specific tissue damage. The morphology of the LRT also varies
15                   within and among species. Rats and mice do not possess respiratory bronchioles;
16                   however, these structures are present in humans, dogs, ferrets, cats, and monkeys.
17                   Respiratory bronchioles are abbreviated in hamsters, guinea pigs, sheep, and pigs. The
18                   branching structure of the ciliated bronchi and bronchioles also differs between species
19                   from being a rather symmetric and dichotomous branching network of airways in humans
20                   and primates to a more monopodial branching network in other mammals. In addition,
21                   rodents have fewer terminal bronchioles due to a smaller lung size  compared to humans
22                   or canines (TvIcBride. 1992). The cellular composition in the pulmonary region is similar
23                   across mammalian species; at least 95% of the alveolar epithelial tissue is composed of
24                   Type I cells. However, significant differences exist between species in the number and
25                   type of cells in the TB airways. Differences also exist in breathing  route and rate.
26                   Primates are oronasal breathers, while rodents are obligate nasal breathers. Past studies of
27                   the effect of body size on resting oxygen consumption also suggest that rodents inhale
28                   more volume of air per lung mass than primates. These distinctions as well as differences
29                   in nasal structure between primates and rodents could affect the amount of O3 uptake.

30                   As O3 absorption and activity relies on ELF antioxidant substances as described in
31                   Section 5.2.3, variability in antioxidant concentrations and metabolism between species
32                   may affect dose and O3-induced health outcomes. The thickness of the ELF in the TB
33                   airways varies among species. Mercer et al. (1992) found that the human ELF thickness
34                   in bronchi and bronchioles was 6.9 and 1.8 (im, respectively, compared to 2.6 and 1.9 (im
3 5                   for the same locations in the rat. Guinea pigs and mice have a lower basal activity of
36                   GSH transferase and GSH peroxidase, and lower a-TOH levels in the lung compared to
      Draft - Do Not Cite or Quote                       5-69                                 September 2011

-------
 1                   rats (Ichinose et al., 1988; Sagai et al., 1987). Nasal lavage fluid analysis shows that
 2                   humans have a higher proportion of their nasal antioxidants as UA and low levels of AH2
 3                   whereas mice, rats, or guinea pigs have high levels of AH2 and undetectable levels of UA
 4                   (Figure 5-1 la). GSH is not detected in the nasal lavage of most of these species, but is
 5                   present in monkey nasal lavage. Guinea pigs and rats have a higher antioxidant to protein
 6                   ratio in nasal lavage and BALF than humans (Hatch. 1992). The BALF profile differs
 7                   from the nasal lavage (Figure 5-1 Ib). Humans have a higher proportion of GSH and less
 8                   AH2 making up their BALF content compared to the guinea pigs and rats (Slade et al..
 9                   1993; Hatch. 1992). Similar to the nose, rats have the highest antioxidant to protein mass
10                   ratio found in BALF (Slade etal.. 1993). Antioxidant defenses also vary with age
11                   (Servais et al.. 2005) and exposure history (Duanetal.. 1996). Duan et al. (1996;  1993)
12                   reported that differences in antioxidant levels between species and lung regions did not
13                   appear to be the primary factor in O3 induced tissue injury. However, a close association
14                   between site-specific  O3 dose, the degree of epithelial injury,  and reduced glutathione
15                   depletion was later revealed in monkeys (Plopperetal.. 1998).

16                   Humans and animals are similar in the pattern of regional O3  dose distribution. As
17                   discussed for humans in Section 5.2.2, O3 flux to the air-liquid interface of the ELF
18                   slowly decreases distally in the TB region and then rapidly decreases distally in the
19                   alveolar region (Miller et al.. 1985). Modeled tissue dose in the human RT, representing
20                   O3 flux to the liquid-tissue interface, is very low in the trachea, increases to a maximum
21                   in the CAR,  and then rapidly decreases distally in the alveolar region (Figure 5-12).
22                   Similar patterns of O3 tissue dose profiles normalized to inhaled O3 concentration were
23                   predicted for rat, guinea pig, and rabbit (Miller etal.. 1988; Overton etal.. 1987) (Figure
24                   5-12a). Overton et al. (1987) modeled rat and guinea pig  O3 dose distribution and found
25                   that after comparing two different morphometrically based anatomical models for each
26                   species, considerable  difference in predicted percent RT and alveolar region uptakes were
27                   observed. This was due to the variability between the two anatomical models in airway
28                   path distance to the first alveolated duct. As a result, the overall dose  profile was similar
29                   between species however the O3 uptake efficiency varied due to RT size and path length
30                   (Section 5.2.2). A similar pattern of O3 dose distribution was measured in monkeys
31                   exposed to 0.4 and 1.0 ppm 18O3 (Plopper et al..  1998) (Figure 5-12b). Less 18O was
32                   measured in the trachea, proximal bronchus, and distal bronchus than was observed in the
33                   respiratory bronchioles. Again indicating the highest concentration of O3 tissue dose to
34                   be localized to the CAR, which are the respiratory bronchioles in nonhuman primates. In
35                   addition, the lowest 18O detected in the RT was in the parenchyma (i.e. alveolar region),
36                   mimicking the rapid decrease in tissue O3 dose predicted by models for the alveolar
37                   regions of humans and other animals.
      Draft - Do Not Cite or Quote                       5-70                                September 2011

-------
b.
                Rat
           Guinea pig
              Human




1


1
a





D Ascorbic acid
• Uric acid
D Glutathione

                Rat
            Guinea pig
               Human
                            200      400      600      800
                             Antioxidant/ Protein, nanomoles / gram
                  1000
           n
                     0        50       100       150      200      250
                              Antioxidant/ Protein, nanomoles / gram

 Source: Adapted with permission from CRC Press, Inc. (Sladeet al.. 1993: Hatch. 1992)

Figure 5-11    Species comparison of antioxidant / protein ratios of: (a) nasal
                lavage fluid and, (b) bronchoalveolar lavage fluid.
Draft - Do Not Cite or Quote
5-71
September 2011

-------
           a. ^
              E 1C)
              "~«
              o
              n
              E
              (0
              c
              E
              co
              o
              Q
              0)
              =1
              co
              to
10'
  10
           b.
              05  50


              1
                 40
                 30
20
              •51
              OS
              c

              c
              o
              o
              £  10
              c
              0)
              O)
              I  •
                           Human
                              Rat ----
                         Guinea Pig
                             f (bpm)

                            15.0
                            66.0
                            60.9
                        Rabbit
                                 _.^.. 13.20  38.8
              (No absorption in the URT)

              	TB	
I—   Zone  0
   Order  0
Generation  0
Generation  0
                                      6
                                      10
                                      10
                         5
                         7
                         12
                         12
14
14
 6 7
 91011
151617
161718
                                                  **
12
19
20
13
21
22
 8 Rabbit
14 Guinea Pig
23 Rat
23 Human
     0.4ppmO3
     1.0ppmO3
    TRACHEA   PROXIMAL    DISTAL   RESPIRATORY PARENCHYMA
            BRONCHUS   BRONCHUS  BRONCHIOLE
 Source: Panel (a) U.S. EPAQ996a) (b) Plopper et al. (1998)


Figure 5-12   Humans and animals are similar in the regional pattern of Os tissue
              dose distribution. Panel (a) presents the predicted tissue dose of
              Os (as ug of Os per cm2 of segment surface area per min,
              standardized to a trachea! Os value of 1 ug/m3) for various regions
              of the rabbit, guinea pig, rat, and human RT. TB = tracheobronchial
              region, A = alveolar region. Panel (b) presents a comparison of
              excess 18O in the five regions of the TB airways of rhesus monkeys
              exposed to Os for 2h. *p<0.05 comparing the same Os
              concentration  across regions. **p<0.05 comparing different Os
              concentrations in the same region.
Draft - Do Not Cite or Quote
                       5-72
                         September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
Humans and animal models are similar in the pattern of regional O3 dose, but absolute
values differ. Hatch et al. (1994) reported that exercising humans exposed to oxygen-18
labeled O3 (400 ppb) accumulated 4-5 times higher concentrations of O3 reaction product
in BAL cells, surfactant and protein fractions compared to resting rats similarly exposed
(0.4 ppm) (Figure 5-13). It was necessary to expose resting rats to 2 ppm O3 to achieve
the same BALF accumulation of 18O reaction product that was observed in humans
exposed to 400 ppb with intermittent heavy exercise (MV ~60 L/min). The concentration
of 18O reaction product in BALF paralleled the accumulation of BALF protein and
cellular effects of the O3 exposure observed such that these responses to 2.0 ppm O3 were
similar to those of the 400 ppb O3 in exercising humans. This suggests that animal data
obtained in resting conditions would underestimate the dose to the RT and presumably
the resultant risk of effect for humans.
                         60
                         50
                         40
                      9  30
                      8
                      UJ  20
                         10
       Source: Hatch et al. (1994)
                                  BAL Cells
                                  BAL HSP
                                  BAL HSS
                                  Lavaged Lung
                            Exercising Human
                            (0.4 ppm, 2 hours)
                            Resting F-344 Rat
                            (0.4 ppm, 2 hours)
Resting F-344 Rat
(2.0 ppm, 2 hours)
      Figure 5-13    Oxygen-18 incorporation into different fractions of BALF from
                     humans and rats exposed to 0.4 and 2.0 ppm 18O3.The excess 18O
                     in each fraction is expressed relative to the dry weight of that
                     fraction. Fractions assayed include cells, high speed pellet (HSP),
                     high speed supernatant (HSS), and lavaged lung homogenates.
      Draft - Do Not Cite or Quote
                             5-73
                 September 2011

-------
 1                   Recently, a quantitative comparison of O3 transport in the airways of rats, dogs, and
 2                   humans was conducted using a three-compartment airways model, based on upper and
 3                   lower airway casts and mathematical calculation for alveolar parameters (Tsujino etal..
 4                   2005). This model examined how interspecies anatomical and physiological differences
 5                   affect intra-airway O3 concentrations and the amount of gas absorbed. The model was
 6                   designed as cylindrical tubes with constant volume and one-dimensional gas movement
 7                   and no airway branching patterns. Peak, real-time, and mean O3 concentrations were
 8                   higher in the upper and lower airways of humans compared to rats and dogs, but lowest
 9                   in the alveoli of humans. The amount of O3 absorbed was lowest in humans when
10                   normalized by body weight. The intra-airway concentration decreased distally in all
11                   species. Sensitivity analysis demonstrated that VT, fB, and upper and lower airways
12                   surface area had a significant impact on model results. The model is limited in that it did
13                   not account for chemical reactions in the ELF or consider gas diffusion as a driving force
14                   for O3 transport. Also, the model was run at a respiratory rate of 16/min simulating a
15                   resting individual, however exercise may cause a further deviation from animal models as
16                   was seen in Hatch et al. (1994).

17                   Overall,  animal models exhibit qualitatively similar patterns of O3 net and tissue dose
18                   distribution with the largest tissue dose delivered to the CAR. However, due to
19                   anatomical and biochemical RT differences the absolute values of O3 dose delivered
20                   differs. Past results suggest that animal data obtained in resting conditions would
21                   underestimate the dose to the RT and presumably the resultant risk of effect for humans,
22                   especially for humans during exercise.
            5.5.2   Homology of Response

23                   Risk of heath effects from O3 varies between and within species, as well as between
24                   endpoints. Rodents appear to have a slightly higher tachypneic response to O3 and are
25                   less sensitive to changes in pulmonary function test than humans (U.S. EPA. 1996a).
26                   However, rats experience attenuation of pulmonary function and tachypneic ventilatory
27                   responses, similar to humans (Wiester et al.. 1996a). Hatch et al. (1986) reported that
28                   guinea pigs were the most responsive to O3-induced inflammatory cell and protein influx.
29                   Rabbits were the least responsive and rats, hamsters, and mice were intermediate
30                   responders. Further analysis of this study by Miller et al. (1988) found that the protein
31                   levels in guinea pigs increased more rapidly with predicted pulmonary tissue dose than in
32                   rats and rabbits. Alveolar macrophages isolated from guinea pigs and humans mounted
33                   similar qualitative and quantitative cytokine responses to in vitro O3 (0.1-1.0 ppm for 60
34                   minutes) exposure (Arsalane et al.. 1995).
      Draft - Do Not Cite or Quote                      5-74                                September 2011

-------
 1                   Also, because of their higher body surface to volume ratio, rodents can rapidly lower
 2                   body temperature during exposure leading to lowered O3 dose and toxicity CWatkinson et
 3                   al.. 2003; Iwasaki et al., 1998; SladeetaL 1997). In addition to lowering the O3 dose to
 4                   the lungs, this hypothermic response may cause: (1) lower metabolic rate, (2) altered
 5                   enzyme kinetics, and (3) altered membrane function. The thermoregulatory mechanisms
 6                   also may affect disruption of heart rate which may lead to: (1) decreased cardiac output,
 7                   (2) lowered blood pressure, and (3) decreased tissue perfusion (Watkinson et al., 2003).
 8                   These responses have not been observed in humans except at very high exposures, thus
 9                   further complicating extrapolation of effects from animals to humans.

10                   Recently, the three-dimensional detail of the nasal passages of immature Rhesus macaque
11                   monkeys was analyzed for developing predictive dosimetry models and exposure-dose-
12                   response relationships (Carey et al., 2007). In doing so the authors reported that the
13                   relative amounts of the five epithelial cell types in the nasal airways of monkeys remains
14                   consistent between infancy and adulthood (comparing to (Gross et al., 1987; Gross et al.,
15                   1982).  Ozone exposures (0.5 ppm, 8 h/day under acute [5 days] and episodic conditions
16                   [5 replicates of the acute paradigm spaced a week apart]) confirmed that the ciliated
17                   respiratory and transitional epithelium were the most sensitive cell types in the nasal
18                   cavity to O3 exposure, showing 50-80% decreases in epithelial thickness and epithelial
19                   cell volume. The character and location of nasal lesions resulting from O3 exposure were
20                   similar between adult and infant monkeys similarly exposed. However, infant monkeys
21                   did not undergo nasal airway epithelial remodeling or adaptation that occurs in adult
22                   animals and they may develop persistent necrotizing rhinitis following episodic longer-
23                   term exposures.

24                   To further understand the genetic basis for age-dependent differential response to O3,
25                   adult (15 week old) and neonatal (15-16 day old) mice from 8 genetically diverse strains
26                   were examined for O3-induced (0.8  ppm for 5 hours) pulmonary injury and lung
27                   inflammation (Vancza et al.. 2009). Ozone exposure increased polymorphonuclear
28                   leukocytes (PMN) influx in all strains of neonatal mice tested, but significantly greater
29                   PMNs occurred in neonatal compared to adult mice for only some sensitive strains,
30                   suggesting a genetic background effect. This  strain difference was not due to differences
31                   in delivered dose of O3 to the lung, evidenced by 18O lung enrichment. The sensitivity of
32                   strains  for O3-induced increases in BALF protein and PMNs was different for different
33                   strains  of mice suggesting that genetic factors contributed to heightened responses.
34                   Interestingly, adult mice accumulated more than twice the levels of 18O reaction product
35                   of O3 than corresponding strain neonates. Thus,  it appeared that the infant mice showed a
36                   two- to threefold higher response than the adults when expressed relative to the
37                   accumulated O3 reaction product in their lungs. The apparent decrease in delivered O3
      Draft - Do Not Cite or Quote                       5-75                                September 2011

-------
 1                   dose in neonates could be a result of a more rapid loss of body temperature in infant
 2                   rodents incident to maternal separation and chamber air flow.

 3                   Further, O3-induced injury and inflammation responses are variable between species. For
 4                   example, Dormans et al. (1999) found that rats, mice, and guinea pigs all exhibited
 5                   O3-induced (0.2 - 0.4 ppm for 3-56 days) inflammation; however, guinea pigs were the
 6                   most sensitive with respect to alveolar macrophage elicitation and pulmonary cell density
 7                   in the centriacinar region. Mice were the most sensitive to bronchiolar epithelial
 8                   hypertrophy and biochemical changes (e.g. LDH, glutathione reductase, glucose-6-
 9                   phosphate dehydrogenase activity), and had the slowest recovery from O3 exposure. All
10                   species displayed increased collagen in the ductal septa and large lamellar bodies in Type
11                   II pneumocytes at the longest exposure and highest concentration; whereas this response
12                   occurred in the rat and  guinea pig at lower O3 levels (0.2 ppm) as well. Overall, the
13                   authors rated mice as most sensitive, followed by guinea pigs, then rats (Dormans et al..
14                   1999). Rats were also less sensitive to epithelial necrosis and inflammatory responses
15                   from O3 (1.0 ppm for 8 hours) than monkeys and ferrets, which manifested a similar
16                   response (Sterner-Kock et al.. 2000). These data suggest that ferrets may be a good
17                   animal model for O3-induced airway effects due to the similarities in pulmonary structure
18                   between primates and ferrets. However, this study provided no dose metric and, it is
19                   possible that some of these differences may be attributable to disparate total inhaled dose
20                   or local organ dose.
             5.5.3   Summary

21                   In summary, for all species there are limitations that must be considered when attempting
22                   to extrapolate to human O3 exposures. Rats required 4-5 times higher exposure to O3 to
23                   achieve comparable increases in BALF protein and PMNs to exercising humans. New
24                   studies have shown that varied O3 response in different mouse strains was not due to
25                   differences in delivered dose of O3 to the lung but more likely genetic sensitivity, and that
26                   infant mice show greater toxicity relative to their smaller lung dose than adults. Even
27                   though interspecies differences limit quantitative comparison between species, the acute
28                   and chronic functional responses of laboratory animals to O3 appear qualitatively
29                   homologous to those of the human making them a useful tool  in determining mechanistic
30                   and cause-effect relationships with O3 exposure.
      Draft - Do Not Cite or Quote                       5-76                                 September 2011

-------
          5.6   Chapter Summary

 1                   Ozone is a highly reactive gas and a powerful oxidant with a short half-life. Both O3
 2                   uptake and responses are dependent upon the formation of secondary reaction products in
 3                   the ELF; however more complex interactions occur. Uptake in humans at rest is 80-95%
 4                   efficient and it is influenced by a number of factors including RT morphology, breathing
 5                   route, frequency, and volume, physicochemical properties of the gas, physical processes
 6                   of gas transport, as well as the physical and chemical properties of the ELF and tissue
 7                   layers. The primary uptake site of O3 delivery to the lung epithelium is believed to be the
 8                   CAR, however changes in a number of factors (e.g. physical activity) can alter the
 9                   distribution of O3 uptake in the RT. Ozone uptake is chemical reaction-dependent and the
10                   substances present in the ELF appear in most cases to limit interaction of O3 with
11                   underlying tissues and to prevent penetration of O3 distally into the RT. Still, reactions of
12                   O3 with soluble  ELF components or plasma membranes result in distinct products, some
13                   of which are highly reactive and can injure and/or transmit signals to RT cells.

14                   Thus, in addition to contributing to the driving force for O3 uptake, formation of
15                   secondary oxidation products initiates pathways that provide the mechanistic basis for
16                   health effects which are described in detail in Chapters 6 and 7 and which involve the RT
17                   as well  as extrapulmonary systems. These pathways include activation of neural reflexes,
18                   initiation of inflammation, alteration of epithelial barrier function, sensitization of
19                   bronchial smooth muscle, modification of innate and adaptive immunity, airways
20                   remodeling, and systemic inflammation and oxidative/nitrosative stress. With the
21                   exception of airways remodeling, these pathways have been demonstrated in both
22                   animals and human subjects in response to the inhalation of O3.

23                   Both dosimetric and mechanistic factors contribute to the understanding of
24                   interindividual variability in responses to O3. Interindividual variability is influenced by
25                   variability in RT volume and thus surface area, certain genetic polymorphisms, pre-
26                   existing conditions and disease, nutritional status, lifestages, attenuation, and co-
27                   exposures. Some of these factors are also influential in understanding species homology
28                   and sensitivity. Qualitatively, animal models exhibit similar patterns of O3 net and tissue
29                   dose distribution with the largest tissue dose delivered to the CAR. However, due to
30                   anatomical and biochemical RT differences, the absolute value of delivered O3 dose
31                   differs, with animal data obtained in resting conditions underestimating the dose to the
32                   RT and presumably the resultant risk of effect for humans, especially humans during
33                   exercise. Even though interspecies differences limit quantitative comparison between
34                   species, the acute and chronic functional responses of laboratory animals to O3 appear
3 5                   qualitatively homologous to those of the human making them a useful tool in determining
36                   mechanistic and cause-effect relationships with O3 exposure.
      Draft - Do Not Cite or Quote                       5-77                                 September 2011

-------
    5.7    References

Abraham, WM: Delehunt, JC: Yerger, l_: Marchette, B: Oliver W, J, r. (1984). Changes in airway permeability and
        responsiveness after exposure to ozone. Environ Res 34: 110-119.
Agarwal. A: Saleh. RA: Bedaiwv. MA. (2003). Role of reactive oxygen species in the pathophysiology of human
        reproduction. Fertil Steril 79: 829-843.
Ahmad. S: Ahmad. A: McConville. G: Schneider.  BK: Allen. CB: Manzer. R: Mason. RJ: White. CW. (2005). Lung
        epithelial cells release ATP during ozone exposure: Signaling for cell survival. Free Radic Biol  Med 39:
        213-226. http://dx.doi.0rg/10.1016/i.freeradbiomed.2005.03.009.
Aibo. PI: Birmingham. NP: Lewandowski. R: Maddox. JF: Roth. RA: Ganev. PE: Wagner. JG: Harkema. JR.
        (2010). Acute exposure to ozone exacerbates acetaminophen-induced liver injury in mice. Toxicol Sci
        115: 267-285.  http://dx.doi.org/10.1093/toxsci/kfq034.
AI-Hegelan. M: Tighe. RM: Castillo. C: Hollingsworth. JW. (2011). Ambient ozone and pulmonary innate
        immunity. Immunol Res 49:  173-191. http://dx.doi.orq/10.1007/s12026-010-8180-z.
Alexis,  N: Urch, B: Tarlo, S: Corey, P: Pengelly, D: O'Byrne, P: Silverman, F. (2000). Cyclooxygenase
        metabolites play a different  role in ozone-induced pulmonary function decline in asthmatics compared
        to normals. Inhal Toxicol 12: 1205-1224.
Alexis,  N: Soukup, J: Nierkens, S: Becker, S. (2001 b). Association between airway hyperreactivity and  bronchial
        macrophage dysfunction in  individuals with mild asthma. Am J Physiol Lung Cell Mol Physiol 280:
        L369-L375.
Alexis,  NE: Zhou, H: Lay, JC: Harris, B: Hernandez, ML:  Lu, TS: Bromberg, PA: Diaz-Sanchez,  D: Devlin, RB:
        Kleeberger.  SR: Peden.  DB. (2009). The glutathione-S-transferase Mu 1 null genotype modulates
        ozone-induced airway inflammation in human subjects. J Allergy Clin Immunol 124: 1222-1228.
        http://dx.doi.0rg/10.1016/i.iaci.2009.07.036.
Alexis.  NE: Lav. JC: Hazucha. M: Harris. B: Hernandez. ML: Bromberg. PA: Kehrl. H: Diaz-Sanchez. D: Kim. C:
        Devlin. RB: Peden.  DB. (2010). Low-level ozone exposure induces airways inflammation and modifies
        cell surface  phenotypes in healthy humans. Inhal Toxicol 22: 593-600.
        http://dx.doi. org/10.3109/08958371003596587.
Alfaro.  MF: Putney. L: Tarkington. BK: Hatch. GE: Hyde.  DM: Schelegle. ES. (2004). Effect of rapid shallow
        breathing on the distribution of 180-labeled ozone reaction product in the respiratory tract of the rat.
        Inhal Toxicol 16: 77-85.
Alfaro.  MF: Walbv. WF: Adams. WC: Schelegle. ES. (2007). Breath condensate levels of 8-isoprostane and
        leukotriene B4 after ozone inhalation are greater in sensitive versus nonsensitive subjects.  Exp Lung
        Res 33: 115-133.  http://dx.doi.org/779284696 [pii]10.1080/01902140701364367.
Araneda. S: Commin. L: Atlagich. M: Kitahama. K: Parraguez. VH: Peguignot. JM: Dalmaz. Y. (2008). VEGF
        overexpression in the astroglial cells of rat brainstem following ozone exposure. Neurotoxicology 29:
        920-927. http://dx.doi.0rg/10.1016/i.neuro.2008.09.006.
Aris. RM: Christian. D:  Hearne. PQ:  Kerr. K:  Finkbeiner. WE: Balmes. JR. (1993). Ozone-induced airway
        inflammation in human subjects as determined by airway lavage and biopsy. Am J Respir Crit Care
        Med 148: 1363-1372.
Arito. H: Uchivama. I: Arakawa. H: Yokovama. E. (1990). Ozone-induced bradycardia and arrhythmia and their
        relation to sleep-wakefulness in rats. Toxicol Lett 52:  169-178. http://dx.doi.org/10.1016/0378-
        4274(90)90151-6.
Arsalane. K:  Gosset. P: Vanhee.  D: Voisin. C: Hamid. Q: Tonnel. A. -B: Wallaert. B. (1995). Ozone stimulates
        synthesis of inflammatory cytokines by alveolar macrophages in vitro. Am J Respir Cell Mol Biol 13: 60-
        68.
Asplund. PT: Ben-Jebria. A:  Rigas. ML: Ultman. JS. (1996). Longitudinal distribution of ozone absorption in the
        lung: Effect of continuous inhalation exposure. Arch Environ Occup Health 51:  431-438.
Auten.  RL: Foster. WM. (In Press) Biochemical effects of ozone on asthma during postnatal development.
        Biochim Biophys Acta. http://dx.doi.Org/10.1016/i.bbagen.2011.01.008.
Auten,  RL: Potts, EN: Mason, SN: Fischer, B: Huang, Y:  Foster, WM. (2009). Maternal exposure to particulate
        matter increases postnatal ozone-induced airway hyperreactivity in juvenile mice. Am J Respir Crit
        Care Med 180: 1218-1226.  http://dx.doi.org/10.1164/rccm.200901-0116OC.
Draft - Do Not Cite or Quote                        5-78                                  September 2011

-------
Awasthi. YC: Yang. Y: Tiwari. NK: Patrick. B: Sharma. A: Li. J: Awasthi. S. (2004). Regulation of 4-
        hydroxynonenal-mediated signaling by gluathione S-transferases. Free Radio Biol Med 37: 607-619.
        http://dx.doi.0rg/10.1016/i.freeradbiomed.2004.05.033.
Backus. GS: Howden. R: Fostel. J: Bauer. AK: Cho. HY: Marzec. J: Peden. DB: Kleeberger. SR. (2010).
        Protective role of interleukin-10 in ozone-induced pulmonary inflammation. Environ Health Perspect
        118: 1721-1727. http://dx.doi.org/10.1289/ehp.1002182.
Ballinger. CA: Cueto. R: Sguadrito. G: Coffin. JF: Velsor. LW: Prvor. WA:  Postlethwait. EM. (2005). Antioxidant-
        mediated augmentation of ozone-induced membrane oxidation.  Free Radic Biol Med 38: 515-526.
        http://dx.doi. orq/10.1016/i.freeradbiomed.20Q4.11.009.
Balmes. JR: Chen. LL: Scannell. C: Tager. I: Christian. D: Hearne. PQ: Kelly. T: Aris. RM. (1996).  Ozone-
        induced decrements in FEV1 and FVC do not correlate with measures of inflammation. Am J Respir
        Grit Care Med 153: 904-909.
Balmes. JR: Aris. RM: Chen. LL: Scannell. C: Tager. IB: Finkbeiner. W: Christian. D: Kelly. T: Hearne. PQ:
        Ferrando, R: Welch, B. (1997). Effects of ozone on normal and potentially sensitive human subjects.
        Part I: Airway inflammation and responsiveness to ozone in normal and asthmatic subjects. Boston,
        MA: Health Effects Institute.
Bang, S: Kim, KY: Yoo, S: Kim, YG: Hwang, SW. (2007). Transient receptor potential A1  mediates
        acetaldehyde-evoked pain sensation. Eur J Neurosci 26: 2516-2523. http://dx.doi.orq/10.1111/1.1460-
        9568.2007.05882.x.
Basha, MA: Gross, KB: Gwizdala, CJ: Haidar, AH: Popovich, J, Jr. (1994). Bronchoalveolar lavage neutrophilia
        in asthmatic and healthy volunteers after controlled exposure to ozone and filtered purified air.  Chest
        106: 1757-1765.
Bastackv, J: Lee, CY: Goerke, J: Koushafar, H: Yager,  D: Kenaga, L: Speed, TP: Chen, Y: Clements, JA.
        (1995). Alveolar lining  layer is thin and continuous: Low-temperature scanning electron microscopy of
        rat lung.  J Appl Physiol 79: 1615-1628.
Bates, ML:  Brenza, TM: Ben-Jebria, A: Bascom, R: Ultman, JS. (2009). Longitudinal distribution of ozone
        absorption in the lung: Comparison of cigarette smokers and nonsmokers. Toxicol Appl Pharmacol 236:
        270-275.
Bauer, AK: Rondini, EA: Hummel, KA: Degraff, LM: Walker, C: Jedlicka, AE: Kleeberger,  SR. (2011).
        Identification of candidate genes downstream of TLR4 signaling after ozone exposure in  mice:  A role
        for heat shock protein  70. Environ Health Perspect 119:  1091-1097.
        http://dx.doi. orq/10.1289/ehp. 1003326.
Beckett. WS: McDonnell. WF: Horstman. DH: House. DE. (1985). Role of the parasympathetic nervous system
        in acute lung response to ozone. J Appl Physiol 59: 1879-1885.
Belvisi, MG: Stretton, CD: Verleden, GM: Ledingham, SJ: Yacoub, MH: Barnes, PJ. (1992). Inhibition of
        cholinergic neurotransmission in human airways by opioids. J Appl Physiol 72: 1096-1100.
Bennett. WD:  Hazucha. MJ: Folinsbee. LJ: Bromberg.  PA: Kissling. GE: London. SJ. (2007). Acute pulmonary
        function response to ozone in young adults as a function of body mass index. Inhal Toxicol 19: 1147-
        1154. http://dx.doi.org/10.1080/08958370701665475.
Bergamaschi. E: De Palma. G: Mozzoni. P: Vanni. S: Vettori. MV: Broeckaert. F:  Bernard. A:  Mutti. A. (2001).
        Polymorphism of quinone-metabolizing enzymes and susceptibility to ozone-induced acute effects.  Am
        J Respir Grit Care  Med 163: 1426-1431.
Bhalla. DK: Gupta. SK. (2000). Lung injury, inflammation,  and inflammatory stimuli in rats exposed to ozone. J
        Toxicol Environ Health 59: 211-228.
Block. ML: Calderon-Garciduenas. L. (2009). Air pollution: Mechanisms of neuroinflammation and CMS disease.
        Trends Neurosci 32: 506-516. http://dx.doi.Org/10.1016/i.tins.2009.05.009.
Blomberg, A: Mudwav, IS: Nordenhall, C: Hedenstrom, H: Kelly, FJ: Frew, AJ: Holgate, ST: Sandstrom,  T.
        (1999). Ozone-induced lung function decrements do not correlate with early airway inflammatory or
        antioxidant responses.  Eur Respir J 13: 1418-1428.
Bosson, J: Stenfors, N: Bucht, A: Helleday, R: Pourazar, J: Holgate, ST: Kelly, FJ: Sandstrom, T: Wilson, S:
        Frew. AJ: Blomberg. A. (2003). Ozone-induced bronchial epithelial cytokine expression differs between
        healthy and asthmatic subjects. Clin Exp Allergy 33: 777-782.
Draft - Do Not Cite or Quote                        5-79                                 September 2011

-------
Bosson. J: Blomberg. A: Pourazar. J: Mudwav. IS: Frew. AJ: Kelly. FJ: Sandstrom. T. (2009). Early suppression
        of NFkappaB and IL-8 in bronchial epithelium after ozone exposure in healthy human subjects. Inhal
        Toxicol 21: 913-919. http://dx.doi.orq/10.1080/08958370802657389.
Broeckaert. F: Clippe. A: Wattiez. R: Falmagne. P: Bernard. A. (2003). Lung hyperpermeability, Clara-cell
        secretory potein (CC16), and susceptibility to ozone of five inbred strains of mice. Inhal Toxicol 15:
        1209-1230.
Bush. ML: Asplund. PT: Miles. KA: Ben-Jebria. A: Ultman. JS. (1996). Longitudinal distribution of O3 absorption
        in the lung: gender differences and intersubject variability. J Appl Physiol 81: 1651-1657.
Bush. ML: Zhang. W: Ben-Jebria. A: Ultman. JS. (2001). Longitudinal distribution of ozone and chlorine in the
        human respiratory tract: Simulation of nasal and oral breathing with the single-path diffusion model.
        Toxicol Appl Pharmacol 173: 137-145. http://dx.doi.org/10.1006/taap.20Q1.9182.
Caceres. Al: Brackmann. M: Ella. MD: Bessac. BF: del Camino. D: D'Amours. M: Witek. JS: Fancier. CM:
        Chong. JA: Havward. NJ: Homer. RJ: Cohn.  L: Huang. X: Moran.  MM: Jordt. SE. (2009). A sensory
        neuronal ion channel essential for airway inflammation and hyperreactivity  in asthma. PNAS 106: 9099-
        9104. http://dx.doi.ora/10.1073/pnas.0900591106.
Carey. SA: Minard. KR: Trease.  LL: Wagner. JG: Garcia. GJ: Ballinger. CA: Kimbell. JS:  Plopper. CG: Corlev.
        RA: Postlethwait,  EM: Harkema, JR:  Einstein, DR. (2007). Three-dimensional mapping of ozone-
        induced injury in the nasal airways of monkeys using magnetic resonance imaging and morphometric
        techniques. Toxicol Pathol 35: 27-40. http://dx.doi.orq/10.1080/01926230601072343.
Chang, L, -Y: Huang, Y: Stockstill, BL: Graham, JA: Grose, EC: Menache, MG: Miller, FJ: Costa, PL: Crapo, JD.
        (1992). Epithelial  injury and interstitial fibrosis in the proximal alveolar regions of rats chronically
        exposed to a simulated pattern of urban ambient ozone. Toxicol Appl Pharmacol 115: 241-252.
        http://dx.doi.org/10.1016/0041-008X(92)90329-Q.
Chen. C: Ariomandi. M: Qin. H: Balmes. J: Tager. I: Holland. N. (2006a).  Cytogenetic damage in buccal epithelia
        and peripheral lymphocytes of young healthy individuals exposed to ozone. Mutagenesis 21: 131-137.
        http://dx.doi. org/10.1093/mutage/gel007.
Chen. C: Ariomandi. M: Balmes. J: Tager. I: N. H. (2007). Effects of chronic and acute ozone exposure  on lipid
        peroxidation and antioxidant capacity in healthy young adults. Environ Health Perspect 115: 1732-
        1737. http://dx.doi.org/10.1289/ehp.10294.
Cho. H. -Y:  Zhang. L. -Y: Kleeberger. SR. (2001). Ozone-induced lung inflammation and hyperreactivity are
        mediated via tumor necrosis factor-"alpha" receptors. Am J Physiol 280: L537-L546.
Cho, HY: Hotchkiss, JA: Harkema, JR. (1999). Inflammatory and epithelial responses during the development of
        ozone-induced mucous cell metaplasia in the nasal epithelium of rats. Toxicol Sci 51: 135-145.
Cho. HY: Kleeberger. SR.  (2007). Genetic mechanisms of susceptibility to oxidative lung injury in mice. Free
        Radic Biol Med 42: 433-445. http://dx.doi.org/10.1016/i.freeradbiomed.20Q6.11.021.
Christian. PL: Chen. LL: Scannell. CH: Ferrando. RE: Welch. BS: Balmes. JR. (1998). Ozone-induced
        inflammation is attenuated with multiday exposure. Am J Respir Crit Care Med 158: 532-537.
Chuang, GC: Yang, Z:  Westbrook, DG: Pompilius, M: Ballinger, CA: White, RC: Krzywanski, DM: Postlethwait,
        EM: Ballinger. SW. (2009). Pulmonary ozone exposure induces vascular dysfunction, mitochondrial
        damage, and atherogenesis. Am J Physiol Lung Cell Mol Physiol 297: L209-L216.
        http://dx.doi. org/10.1152/ajplung.00102.2009.
Cohen-Hubal. EA: Kimbell. JS: Fedkiw. PS. (1996). Incorporation of nasal-lining mass-transfer resistance into a
        CFD model for prediction of ozone dosimetry in the upper respiratory tract. Inhal Toxicol 8: 831-857.
Cole,  MP: Freeman, BA. (2009). Promotion of cardiovascular disease by exposure to the air pollutant ozone.
        Am J Physiol Lung Cell Mol Physiol 297: L209-L216.
Coleridge. HM: Coleridge.  JCG: Ginzel. KH: Baker. DG: Banzett. RB: Morrison. MA. (1976). Stimulation of
        "irritant1 receptors and afferent C-fibers in the lungs by prostaglandins. Nature 264: 451-453.
Coleridge. JCG: Coleridge. HM: Schelegle. ES: Green. JF (1993). Acute inhalation  of ozone stimulates
        bronchial C-fibers and rapidly adapting receptors in dogs. J Appl Physiol 74: 2345-2352.
Corradi, M: Alinovi, R: Goldoni, M: Vettori, M:  Folesani, G: Mozzoni,  P: Cavazzini, S: Bergamaschi, E: Rossi, L:
        Mutti. A.  (2002). Biomarkers of oxidative stress after controlled human exposure to ozone. Toxicol Lett
        134: 219-225.
Costa, PL: Schafrank,  SN: Wehner, RW: Jellett, E. (1985). Alveolar permeability to protein in rats differentially
        susceptible to ozone. J Appl Toxicol 5: 182-186. http://dx.doi.org/10.1002/iat.25500503Q9.
Draft - Do Not Cite or Quote                        5-80                                  September 2011

-------
Cross. CE: Motchnik. PA: Bruener. BA: Jones. DA: Kaur. H: Ames. BN: Halliwell. B. (1992). Oxidative damage
        to plasma constituents by ozone. FEBS Lett 298: 269-272. http://dx.doi.org/10.1016/0014-
        5793(92)80074-Q.
Dahl. AR. (1990). Dose concepts for inhaled vapors and gases. Toxicol Appl Pharmacol 103: 185-197.
Dahl, M: Bauer, AK: Arredouani, M: Soininen, R: Tryggvason, K: Kleeberger, SR: Kobzik, L. (2007). Protection
        against inhaled oxidants through scavenging of oxidized lipids by macrophage receptors MARCO and
        SR-AI/II. J Clin  Invest 117: 757-764. http://dx.doi.org/10.1172/JCI29968.
Darrow,  LA: Klein, M: Flanders, WD: Waller, LA: Correa, A:  Marcus, M: Mulholland, JA: Russell, AG: Tolbert, PE.
        (2009). Ambient air pollution and preterm birth: A time-series analysis. Epidemiology 20: 689-698.
Delaunois. A: Segura. P: Montano. LM: Vargas. MH: Ansav. M: Gustin. P. (1998). Comparison of ozone-induced
        effects on lung mechanics and hemodynamics in the rabbit. Toxicol Appl Pharmacol 150: 58-67.
Dempsev. J. . A.: Johnson. B. . P.: Saupe. K. . W. (1990). Adaptations and limitations in the pulmonary system
        during exercise. Chest 97: 81S-87S.
Dempsev, J, . A.: McKenzie,  D, . C.: Haverkamp,  H, . C.: Eldridge, M, . W. (2008).  Update in the understanding
        of respiratory limitations to exercise performance in fit, active adults. Chest 134:  613-622.
        http://dx.doi. orq/10.1378/chest. 07-2730.
Devlin, RB: McDonnell, WF:  Mann, R: Becker, S: House, DE: Schreinemachers, D: Koren, HS. (1991).
        Exposure of humans to ambient levels of ozone for 6.6 hours causes cellular and biochemical changes
        in the lung. Am  J Respir Cell Mol Biol 4: 72-81.
Devlin, RB: Folinsbee, LJ:  Biscardi, F: Hatch, G: Becker, S: Madden, MC: Robbins, M:  Koren, HS. (1997).
        Inflammation and cell damage induced by repeated exposure of humans to ozone. Inhal Toxicol 9: 211-
        235.
Diemer,  T: Allen, JA: Hales, KH: Hales, DB. (2003). Reactive oxygen disrupts mitochondria in MA-10 tumor
        Leydig cells and inhibits steroidogenic acute regulatory (StAR) protein and steroidogenesis.
        Endocrinology 144: 2882-2891. http://dx.doi.org/10.1210/en.2002-0090.
Dietert, RR: DeWitt, JC: Germolec, PR: Zelikoff, JT (2010). Breaking patterns of environmentally influenced
        disease for health risk reduction: Immune perspectives. Environ Health Perspect 118:  1091-1099.
        http://dx.doi. orq/10.1289/ehp. 1001971.
Dimeo, MJ: Glenn, MG:  Holtzman, MJ: Sheller, JR: Nadel, JA: Boushev, HA. (1981). Threshold concentration of
        ozone causing an increase in bronchial reactivity in humans and adaptation with repeated exposures.
        Am Rev Respir Dis 124: 245-248.
Dormans, JAM, A: Van Bree, L: Boere, AJF: Marra, M:  Rombout,  PJA. (1999). Interspecies differences in time
        course of pulmonary toxicity following repeated exposure to ozone. Inhal Toxicol 11: 309-329.
Poster!  C: Petrilli. V: Van Bruggen. R: Steele. C: Mossman. BT: Tschopp. J. (2008). Innate immune activation
        through NalpS inflammasome sensing of asbestos and silica. Science 320: 674-677.
Drechsler-Parks. DM. (1995). The dose-response relationship in older men exposed to ozone.  Exp Gerontol 30:
        65-75.
Duan, X: Buckpitt, AR: Plopper, CG. (1993). Variation in antioxidant enzyme activities in anatomic
        subcompartments within rat and rhesus monkey lung. Toxicol Appl Pharmacol 123: 73-82.
Duan. X: Buckpitt. AR: Pinkerton. KE: Ji. C: Plopper. CG. (1996).  Ozone-induced alterations in glutathione in
        lung subcompartments of rats and monkeys. Am J Respir Cell Mol Biol 14: 70-75.
Emmons. K: Foster. WM. (1991). Smoking cessation and acute airway response to ozone. Arch Environ Occup
        Health 46: 288-295. http://dx.doi.org/10.1080/00039896.1991.9934389.
Enami, S: Hoffmann, MR: Colussi, AJ. (2008a). Acidity enhances the formation of a persistent ozonide at
        aqueous ascorbate/ozone gas interfaces. PNAS 105: 7365-7369.
        http://dx.doi.org/10.1073/pnas.0710791105.
Enami, S: Hoffmann, MR: Colussi, AJ. (2008b). Ozonolysis of uric acid at the air/water interface. J Phys Chem
        B 112: 4153-4156. http://dx.doi.org/10.1021/ip712010k.
Enami. S: Hoffmann. MR: Colussi. AJ. (2009a). How phenol and alpha-tocopherol  react with ambient ozone at
        gas/liquid interfaces. J Phys Chem A 113: 7002-7010. http://dx.doi.org/10.1021/ip901712k.
Enami. S: Hoffmann. MR: Colussi. AJ. (2009b). Ozone oxidizes glutathione to a sulfonic acid. Chem Res Toxicol
        22: 35-40. http://dx.doi.org/10.1021/tx800298i10.1021/tx800298i.
Draft - Do Not Cite or Quote                        5-81                                  September 2011

-------
Enami. S: Hoffmann. MR: Colussi. AJ. (2009c). Simultaneous detection of cysteine sulfenate, sulfinate, and
        sulfonate during cysteine interfacial ozonolysis. J Phys Chem B 113: 9356-9358.
        http://dx.doi.org/10.1021/ip904316n.
Engel. LA. (1985). Intraregional gas mixing and distribution. In Gas Mixing and Distribution in the Lung (pp. 287-
        358). New York: Marcel Dekker.
Esther. CR: Peden. DB: Alexis. NE: Hernandez. ML. (2011). Airway purinergic responses in healthy, atopic
        nonasthmatic, and atopic asthmatic subjects exposed to ozone. Inhal Toxicol 23: 324-330.
        http://dx.doi.org/10.3109/08958378.2011.572096.
Fabbri. LM: Aizawa. H: O'Bvrne. PM: Bethel. RA: Walters. EH: Holtzman. MJ: Nadel. JA. (1985). An anti-
        inflammatory drug (BW755C) inhibits airway hyperresponsiveness induced by ozone in dogs. J Allergy
        Clin Immunol 76: 162-166. http://dx.doi.org/10.1016/0091-6749(85)90695-5.
Fakhri. AA:  Ilic. LM: Wellenius. GA: Urch. B: Silverman. F: Gold. PR:  Mittleman. MA. (2009). Autonomic effects
        of controlled fine particulate exposure in young healthy adults: Effect modification  by ozone. Environ
        Health Perspect 117: 1287-1292. http://dx.doi.org/10.1289/ehp.0900541.
Feng. R: He. W: Ochi. H: Castranova. V.  (2006).  Ozone exposure impairs antigen-specific  immunity but
        activates IL-7-induced proliferation of CD4-CD8-thymocytes in BALB/c mice. J Toxicol Environ  Health
        A69: 1511-1526. http://dx.doi.org/10.1080/15287390500468696.
Foster. WM: Stetkiewicz. PT. (1996). Regional clearance of solute from the respiratory epithelia: 18-20 h
        post exposure to ozone. J Appl Physiol 81: 1143-1149.
Foster, WM: Freed, AN. (1999). Regional clearance of solute from peripheral airway epithelia: Recovery after
        sublobar exposure to ozone. J Appl Physiol 86: 641-646.
Foster. WM: Brown. RH: Macri. K: Mitchell. CS. (2000). Bronchial  reactivity of healthy subjects: 18-20 h
        post exposure to ozone. J Appl Physiol 89: 1804-1810.
Frampton. MW: Morrow. PE: Torres. A: Voter. KZ: Whitin. JC: Cox. C: Speers. DM: Tsai. Y:  Utell. MJ. (1997a).
        Effects of ozone on normal and potentially sensitive human subjects Part II: airway inflammation and
        responsiveness to ozone in nonsmokers and smokers. Boston, MA: Health Effects Institute.
Frampton. MW: Morrow. PE: Torres. A: Cox. C: Voter. KZ: Utell. MJ: Gibb. FR: Speers. DM. (1997b). Ozone
        responsiveness in smokers and  nonsmokers. Am J Respir Crit Care Med 155: 116-121.
Frampton, MW: Pryor, WA: Cueto, R: Cox, C: Morrow, PE:  Utell, MJ. (1999). Ozone exposure increases
        aldehydes in epithelial lining fluid in human lung. Am J  Respir Crit Care Med 159:  1134-1137.
Frank. R: Liu. MC: Spannhake. EW: Mlvnarek. S: Macri. K: Weinmann. GG. (2001).  Repetitive ozone exposure
        of young adults: Evidence of persistent small airway dysfunction. Am J Respir Crit Care Med 164:
        1253-1260.
Freed. AN: Chou. CL: Fuller. SD: Croxton. TL. (1996).  Ozone-induced vagal reflex modulates airways reactivity
        in rabbits. Respir Physiol Neurobiol 105: 95-102.
Freed. AN: Cueto. R: Prvor. WA. (1999). Antioxidant transport modulates peripheral  airway reactivity and
        inflammation during ozone exposure. J Appl Physiol 87: 1595-1603.
Fujita, M: Sasayama, S: Ohno, A:  Nakajima, H: Asanoi, H. (1987). Importance of angina  for development of
        collateral circulation. Heart 57: 139-143.
Gackiere. F: Saliba. L: Baude. A: Bosler.  O: Strube. C. (2011). Ozone inhalation activates stress-responsive
        regions of the CMS. J  Neurochem 117: 961-972. http://dx.doi.Org/10.1111/i.1471-4159.2011.07267.x.
Gao. X:  Raghavamenon. AC: D'Auvergne. O: Uppu. RM. (2009b). Cholesterol secoaldehyde induces apoptosis
        in J774 macrophages  via mitochondrial pathway but not involving reactive oxygen species as
        mediators. Biochem Biophys Res Commun 389: 382-387. http://dx.doi.Org/10.1016/i.bbrc.2009.09.005.
Garantziotis. S: Li. Z: Potts. EN: Kimata.  K: Zhuo. L: Morgan. PL:  Savani. RC: Noble. PW:  Foster. WM:
        Schwartz. DA: Hollingsworth. JW. (2009). Hyaluronan mediates ozone-induced airway
        hyperresponsiveness in mice. J  Biol Chem 284: 11309-11317.
        http://dx.doi. org/10.1074/ibc. M802400200.
Garantziotis. S: Li. Z: Potts. EN: Lindsev. JY: Stober. VP: Polosukhin. VV: Blackwell. TS: Schwartz. DA:  Foster.
        WM:  Hollingsworth, JW. (2010).  TLR4 is necessary for hyaluronan-mediated airway
        hyperresponsiveness after ozone inhalation. Am J Respir Crit Care Med 181: 666-675.
        http://dx.doi.org/10.1164/rccm.200903-0381OC.
Gerritv, TR: Weaver, RA: Berntsen, J: House, DE: O'Neil, JJ. (1988). Extrathoracic and intrathoracic removal of
        O3 in tidal-breathing humans. J Appl Physiol 65: 393-400.
Draft - Do Not Cite or Quote                        5-82                                 September 2011

-------
Gerritv. TR: McDonnell. WF: House. DE. (1994). The relationship between delivered ozone dose and functional
        responses in humans. Toxicol Appl Pharmacol 124: 275-283.
Gerritv. TR: Biscardi. F: Strong. A: Garlington. AR: Brown. JS: Bromberg. PA. (1995). Bronchoscopic
        determination of ozone uptake in humans. J Appl Physiol 79: 852-860.
Giamalva, D: Church, DF: Pryor, WA. (1985). A comparison of the rates of ozonation of biological antioxidants
        and oleate and linoleate esters. Biochem Biophys Res Commun 133: 773-779.
Gong. H. Jr: Wong. R: Sarma. RJ: Linn. WS: Sullivan. ED: Shamoo.  DA: Anderson. KR:  Prasad. SB. (1998).
        Cardiovascular effects of ozone exposure in human volunteers. Am J Respir Crit Care Med 158: 538-
        546.
Graham. DE: Koren.  HS. (1990). Biomarkers of inflammation in ozone-exposed humans: Comparison of the
        nasal and bronchoalveolar lavage. Am J Respir Crit Care Med 142: 152-156.
Gross.  EA: Swenberg. JA: Fields. S: Popp. JA. (1982). Comparative  morphometry of the nasal cavity in rats and
        mice. J Anat 135: 83-88.
Gross,  EA: Starr, TB: Randall, HW: Morgan, KT. (1987). Morphometric analysis of the  primate nasal cavity
        [Abstract]. Toxicologist 7: 193.
Guevara-Guzman. R: Arriaga. V: Kendrick. KM: Bernal. C: Vega.  X: Mercado-Gomez.  OF: Rivas-Arancibia. S.
        (2009). Estradiol prevents ozone-induced increases in brain  lipid peroxidation and impaired social
        recognition memory in female rats. Neuroscience 159: 940-950.
        http://dx.doi.0rg/10.1016/i.neuroscience.2009.01.047.
Gunnison, AF: Hatch, GE: Crissman, K: Bowers, A. (1996). Comparative sensitivity of lactating and virgin
        female rats to ozone-induced pulmonary inflammation. Inhal Toxicol 8: 607-623.
Gunnison. AF: Hatch. GE. (1999). OS-induced inflammation in prepregnant, pregnant, and lactating rats
        correlates with O3 dose estimated by 18O. Am J Physiol 276: L332-L340.
Hackney. JD: Linn. WS: Mohler. JG: Collier. CR. (1977). Adaptation to short-term respiratory effects of ozone in
        men exposed repeatedly. J Appl Physiol 43:  82-85.
Hamade, AK: Tankersley, CG. (2009). Interstrain variation in cardiac and respiratory adaptation to repeated
        ozone and particulate matter exposures. Am J Physiol Regul Integr Comp Physiol 296: R1202-R1215.
        http://dx.doi.ora/10.1152/aipregu.90808.2008.
Hamade, AK: Misra, V: Rabold, R: Tankersley, CG. (2010). Age-related changes in cardiac and respiratory
        adaptation to acute ozone and carbon black exposures:  Interstrain variation in mice. Inhal Toxicol 22:
        84-94. http://dx.doi.org/10.3109/08958378.2010.503974.
Hamilton, RF: Li, l_: Eschenbacher, Wl_: Szweda, l_: Holian, A. (1998). Potential involvement of 4-
        hydroxynonenal in the response of human lung cells to ozone. Am J Physiol 274: L8-L16.
Han. SG: Andrews. R: Gairola. CG: Bhalla. DK. (2008). Acute pulmonary effects of combined exposure to
        carbon nanotubes and ozone in mice. Inhal Toxicol 20: 391-398.
        http://dx.doi. orq/10.1080/08958370801904014.
Hansen. CA: Barnett. AG: Pritchard. G. (2008). The effect of ambient air pollution during early pregnancy on
        fetal ultrasonic measurements during mid-pregnancy. Environ Health Perspect 116: 362-369.
Hague. R: Umstead.  TM: Ponnuru. P: Quo. X: Hawgood. S: Phelps. PS: Floros.  J. (2007). Role of surfactant
        protein-A(SP-A) in lung injury in response to acute ozone exposure of SP-A deficient mice. Toxicol
        Appl Pharmacol 220:  72-82. http://dx.doi.org/10.1016/j.taap.20Q6.12.017.
Hague. R: Umstead.  TM: Freeman. WM: Floros. J: Phelps. PS. (2009). The impact of surfactant protein-Aon
        ozone-induced changes in the mouse bronchoalveolar lavage proteome. Proteome Science 7: 12.
        http://dx.doi.org/10.1186/1477-5956-7-12.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Dungworth. PL (1987a). Effects of an ambient level of
        ozone on  primate nasal epithelial  mucosubstances: quantitative histochemistry.  Am J Pathol 127: 90-
        96.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Wilson. DW: Dungworth. PL. (1987b). Response of the
        macaque  nasal epithelium to ambient levels of ozone: A  morphologic and morphometric study of the
        transitional and respiratory epithelium. Am J Pathol 128:  29-44.
Harkema. JR: Hotchkiss. JA: Barr. EB: Bennett. CB: Gallup. M: Lee. JK: Basbaum.  C. (1999).  Long-lasting
        effects of chronic ozone exposure on rat nasal epithelium. Am J Respir Cell Mol Biol 20: 517-529.
Draft - Do Not Cite or Quote                       5-83                                  September 2011

-------
Harkema. JR: Wagner. JG. (2005). Epithelial and inflammatory responses in the airways of laboratory rats
        coexposed to ozone and biogenic substances: Enhancement of toxicant-induced airway injury. Exp
        Toxicol Pathol 57: 129-141. http://dx.doi.Org/10.1016/i.etp.2005.05.013.
Hatch. GE: Slade. R: Stead. AG: Graham.  JA. (1986). Species comparison of acute inhalation toxicity of ozone
        and  phosgene. J Toxicol Environ Health 19: 43-53. http://dx.doi.org/10.1080/15287398609530905.
Hatch. GE: Wiester. MJ: Overton. JH. Jr: Aissa. M. (1989). Respiratory tract dosimetry of [18]O-labeled ozone in
        rats: Implications for a rat-human extrapolation of ozone dose. In Atmospheric ozone research and its
        policy implications (Vol. 35, pp. 553-560). Nijmegend, the Netherlands: Elsevier Science Publishers
        B.V.
Hatch. GE. (1992). Comparative biochemistry of airway lining fluid. In RA Parent (Ed.), Comparative biology of
        the normal lung (Vol.  1, pp. 617-632). Boca Raton, FL: CRC Press, Inc.
Hatch. GE: Slade. R: Harris. LP: McDonnell. WF: Devlin. RB: Koren. HS:  Costa. PL: McKee. J. (1994). Ozone
        dose and effect in humans and rats: A comparison using oxygen-18 labeling and bronchoalveolar
        lavage. Am J Respir Grit Care Med 150: 676-683.
Hazbun. ME: Hamilton. R: Holian. A: Eschenbacher. WL. (1993). Ozone-induced increases in substance P and
        8-epi-prostaglandin F2 alpha in the airways of human subjects. Am J Respir Cell Mol Biol 9:  568-572.
        http://dx.doi.Org/10.1165/aircmb/9.5.568.
Hazucha. MJ: Bates. DV: Bromberg. PA. (1989). Mechanism of action of ozone on the human lung. J Appl
        Physiol67: 1535-1541.
Hazucha, MJ: Madden, M: Pape, G: Becker, S: Devlin, R: Koren, HS: Kehrl, H:  Bromberg,  PA. (1996). Effects of
        cyclo-oxygenase inhibition on ozone-induced respiratory inflammation  and lung function changes. Eur J
        Appl Physiol 73:  17-27.
Hazucha, MJ: Folinsbee,  LJ:  Bromberg, PA. (2003). Distribution and reproducibility of spirometric response to
        ozone by gender and age. J Appl Physiol 95: 1917-1925.
Hernandez. ML:  Lav. JC:  Harris. B: Esther. CR: Brickev. WJ: Bromberg. PA: Diaz-Sanchez. D: Devlin. RB:
        Kleeberger, SR: Alexis, NE: Peden, DB. (2010). Atopic asthmatic subjects but not atopic subjects
        without asthma have enhanced inflammatory response to ozone. J Allergy Clin Immunol 126: 537-544.
        http://dx.doi.0rg/10.1016/i.iaci.2010.06.043.
Hoigne,  J: Bader, H. (1983). Rate constants of reactions of ozone with organic and inorganic compounds in
        water- II: Dissociating organic compounds. Water Res 17:  185-194. http://dx.doi.org/10.1016/0043-
        1354(83)90099-4.
Hollingsworth, JW: Maruoka, S: Li, Z: Potts, EN: Brass, DM: Garantziotis, S: Fong, A: Foster, WM:  Schwartz,
        DA.  (2007). Ambient ozone primes pulmonary innate immunity in mice. J Immunol 179: 4367-4375.
Hollingsworth. JW: Free. ME:  Li. Z: Andrews. LN: Nakano. H: Cook. DN. (2010). Ozone activates pulmonary
        dendritic cells and promotes allergic sensitization through a Toll-like receptor 4-dependent mechanism.
        J Allergy Clin Immunol 125: 1167-1170.
Hollingsworth. JW. II: Cook. DN: Brass. DM: Walker. JKL: Morgan.  PL: Foster. WM: Schwartz. DA. (2004). The
        role  of Toll-like receptor 4 in environmental airway injury in mice. Am J  Respir Crit Care Med 170: 126-
        132.
Holtzman. MJ: Cunningham.  JH: Sheller. JR: Irsigler. GB:  Nadel. JA: Boushev. HA. (1979). Effect of ozone on
        bronchial reactivity in atopic and nonatopic subjects. Am Rev Respir Dis 120: 1059-1067.
Holtzman. MJ: Fabbri. LM: O'Bvrne. PM: Gold. BD: Aizawa. H: Walters. EH: Alpert. SE: Nadel. JA.  (1983).
        Importance of airway inflammation for hyperresponsiveness induced by ozone. Am Rev Respir Dis 127:
        686-690.
Holz. O: Jorres. RA: Timm. P: Mucke. M: Richter. K: Koschvk. S: Magnussen. H. (1999). Ozone-induced airway
        inflammatory changes differ between individuals and are reproducible. Am J Respir Crit Care Med 159:
        776-784.
Horstman. DH: Ball. BA: Brown. J: Gerritv.  T: Folinsbee. LJ. (1995). Comparison of pulmonary responses of
        asthmatic and nonasthmatic subjects performing  light exercise while exposed to a low level of ozone.
        Toxicol Ind Health 11: 369-385.
Horvath. SM: Gliner. JA: Folinsbee. LJ. (1981). Adaptation to ozone: Duration of effect. Am Rev Respir Dis 123:
        496-499.
Hotchkiss, JA: Harkema, JR: Henderson, RF. (1991). Effect of cumulative ozone exposure on ozone-induced
        nasal epithelial hyperplasia and secretory metaplasia in rats. Exp Lung Res 15: 589-600.
Draft - Do Not Cite or Quote                       5-84                                  September 2011

-------
Hu. PC: Miller. FJ: Daniels. MJ: Hatch. G. (1982). Protein accumulation in lung lavage fluid following ozone
        exposure. Environ Res 29: 377-388. http://dx.doi.Org/10.1016/0013-9351(82)90039-1.
Hu. S. -C: Ben-Jebria. A: Ultman. JS. (1994). Longitudinal distribution of ozone absorption in the lung: Effects of
        respiratory flow. J Appl Physiol 77: 574-583.
Hu, SC: Ben-Jebria, A: Ultman, JS. (1992). Longitudinal distribution of ozone absorption in the lung: Quiet
        respiration in healthy subjects. J Appl Physiol 73: 1655-1667.
Ichinose. T: Arakawa. K: Shimoio. N: Sagai. M. (1988). Biochemical effects of combined gases of nitrogen
        dioxide and ozone: II Species differences in lipid peroxides and antioxidative protective enzymes in the
        lungs. Toxicol Lett 42: 167-176.
ICRP (International Commission on Radiological Protection). (1994). Human respiratory tract model for
        radiological protection: A report of a task group of the International Commission on Radiological
        Protection. Ann ICRP 24: 1-482.
Ignatenko. AV: Cherenkevich. SN. (1985). Reactivity of amino-acids and proteins in reactions with ozone. Kinet
        Catal26: 1145-1148.
Islam. T: McConnell. R:  Gauderman. WJ: Avol. E: Peters. JM: Gilliland. FD. (2008). Ozone, oxidant defense
        genes and risk  of asthma during adolescence. Am J Respir Crit Care Med 177: 388-395.
        http://dx.doi.org/10.1164/rccm.200706-863OC.
Iwasaki. T: Takahashi. M: Saito. H: Arito. H. (1998). Adaptation of extrapulmonary responses to ozone exposure
        in conscious rats. Ind Health 36: 57-60.
Jalaludin, B: Mannes, T: Morgan, G: Lincoln, D: Sheppeard, V: Corbett, S. (2007). Impact of ambient air
        pollution on gestational age is modified by season in Sydney, Australia. Environ Health 6: 16.
Jiang. D: Liang. J: Fan. J: Yu. S: Chen. S: Luo. Y: Prestwich. GD: Mascarenhas. MM: Garg. HG: Quinn. DA:
        Homer, RJ: Goldstein, PR: Bucala,  R: Lee, PJ: Medzhitov, R:  Noble, PW. (2005). Regulation of lung
        injury and repair by Toll-like receptors and hyaluronan. Nat Med 11: 1173-1179.
        http://dx.doi. orq/10.1038/nm1315.
Joad, JP: Kott, KS:  Brie, JM. (1996). The local C-fiber contribution to ozone-induced effects on the isolated
        guinea pig lung. Toxicol Appl Pharmacol 141: 561-567.
Johansson. E: Wesselkamper. SC: Shertzer. HG: Leikauf. GD: Dalton. TP: Chen. Y. (2010). Glutathione
        deficient C57BL/6J mice are not sensitized to ozone-induced lung injury. Biochem Biophys Res
        Commun 396: 407-412. http://dx.doi.orq/10.1016/i.bbrc.2010.04.105.
Johnston. RA: Mizgerd.  JP: Shore. SA. (2005a).  CXCR2 is essential for maximal neutrophil recruitment and
        methacholine responsiveness after ozone exposure. Am J Physiol Lung Cell Mol Physiol 288:  L61-L67.
        http://dx.doi.orq/10.1152/aipluna.00101.2004 00101.2004.
Johnston. RA: Schwartzman. IN: Flvnt. L: Shore. SA. (2005b). Role of interleukin-6 in murine airway responses
        to ozone. Am J  Physiol Lung Cell Mol Physiol 288: L390-L397.
        http://dx.doi.ora/10.1152/aipluna.00007.2004.
Johnston. RA: Theman.  TA: Lu. FL: Terry. RD: Williams. ES: Shore. SA. (2008). Diet-induced obesity causes
        innate airway hyperresponsiveness to methacholine and enhances ozone-induced pulmonary
        inflammation. J Appl  Psychol 104: 1727-1735. http://dx.doi.ora/10.1152/iapplphvsiol.00075.2008.
Jorres. R: Nowak. D: Magnussen. H: Speckin. P: Koschvk. S. (1996). The effect of ozone exposure on allergen
        responsiveness in subjects with asthma or rhinitis. Am J Respir Crit Care Med 153: 56-64.
Jorres. RA: Holz. O: Zachgo.  W: Timm. P: Koschvk. S: Mullen B: Grimminger. F: Seeger.  W: Kelly. FJ: Punster.
        C: Frischer. T: Lubec. G: Waschewski. M:  Niendorf. A: Magnussen. H. (2000).  The effect of repeated
        ozone exposures on inflammatory markers in bronchoalveolar lavage fluid and mucosal biopsies. Am J
        Respir Crit Care Med 161: 1855-1861.
Kabel. JR: Ben-Jebria. A: Ultman. JS. (1994). Longitudinal distribution of ozone absorption in the lung:
        Comparison of  nasal and oral quiet breathing. J Appl Physiol 77: 2584-2592.
Kafourv. RM:  Prvor. WA: Squadrito. GL: Salgo. MG: Zou. X: Friedman.  M. (1998). Lipid ozonation products
        activate phospholipases A2, C, and D. Toxicol Appl Pharmacol 150: 338-349.
Kannan, S: Misra, DP: Dvonch, T: Krishnakumar, A. (2006). Exposures to airborne particulate matter and
        adverse perinatal outcomes: A biologically plausible mechanistic framework for exploring potential
        effect modification by nutrition. Environ Health Perspect 114: 1636-1642.
Kanofskv, JR: Sima, PP. (1995). Reactive absorption of ozone by aqueous biomolecule solutions: Implications
        for the role of sulfhydryl compounds as targets for ozone. Arch Biochem Biophys 316: 52-62.
Draft - Do Not Cite or Quote                        5-85                                  September 2011

-------
Karl. F: Hatch. G: Slade. R: Crissman. K: Simeonova. PP: Luster. M. (1997). Dietary restriction mitigates ozone-
        induced lung inflammation in rats: A role for endogenous antioxidants. Am J Respir Cell  Mol Biol 17:
        740-747.
Katre. A: Ballinger. C: Akhter. H: Fanucchi. M: Kim. DK: Postlethwait. E: Liu. RM. (2011). Increased transforming
        growth factor beta 1 expression mediates ozone-induced airway fibrosis in mice. Inhal Toxicol 23: 486-
        494. http://dx.doi.org/10.3109/08958378.2011.584919.
Kavlock. R: Daston. G: Grabowski. CT. (1979). Studies on the developmental toxicity of ozone. I. Prenatal
        effects. Toxicol Appl Pharmacol 48: 19-28. http://dx.doi.org/10.1016/S0041-008X(79)80004-6.
Kehrl. HR: Vincent. LM: Kowalskv. RJ: Horstman. DH: O'Neil. JJ: McCartney. WH: Bromberg. PA. (1987).
        Ozone exposure increases respiratory epithelial permeability in humans. Am Rev Respir Dis 135: 1124-
        1128.
Kehrl. HR: Peden. DB: Ball. BA: Folinsbee. LJ: Horstman. DH. (1999). Increased specific airway reactivity of
        persons with mild allergic asthma after 7.6 hours of exposure to 0.16 ppm ozone. J Allergy Clin
        Immunol 104: 1198-1204.
Kermani. S: Ben-Jebria. A: Ultman. JS. (2006). Kinetics of ozone reaction with uric acid,  ascorbic acid, and
        glutathione at physiologically relevant conditions. Arch Biochem Biophys 451: 8-16.
        http://dx.doi.0rg/10.1016/i.abb.2006.04.015.
Kim. CS: Alexis. NE: Rappold. AG: Kehrl. H: Hazucha. MJ:  Lav. JC: Schmitt. MT: Case. M: Devlin. RB: Peden.
        DB: Diaz-Sanchez. D. (2011). Lung function and inflammatory responses in healthy young adults
        exposed to 0.06 ppm ozone for 6.6 hours. Am J  Respir Grit Care  Med 183: 1215-1221.
        http://dx.doi.org/10.1164/rccm.201011-1813OC.
Kleeberger. SR: Seiden. JE: Levitt. RC: Zhang. L. -Y. (1993). Mast cells modulate acute ozone-induced
        inflammation of the murine lung. Am J Respir Grit Care Med 148: 1284-1291.
Kleeberger. SR: Levitt. RC: Zhang. L. -Y: Longphre. M: Harkema. J: Jedlicka. A: Eleff. SM: DiSilvestre. D:
        Holrovd.  KJ. (1997). Linkage analysis of susceptibility to ozone-induced lung inflammation in inbred
        mice. Nat Genet 17: 475-478.
Kleeberger. SR: Reddv. S: Zhang. L. -Y: Jedlicka. AE. (2000). Genetic susceptibility to ozone-induced lung
        hyperpermeability: Role of toll-like receptor 4. Am J Respir Cell Mol Biol 22: 620-627.
Kleeberger, SR: Reddv, SP: Zhang, L, -Y: Cho, H, -Y: Jedlicka, AE. (2001). Toll-like receptor 4 mediates ozone-
        induced murine lung hyperpermeability via inducible nitric oxide synthase. Am J Physiol 280: L326-
        L333.
Kodavanti, UP: Costa, PL: Dreher, KL: Crissman, K: Hatch, GE. (1995). Ozone-induced tissue injury and
        changes in antioxidant homeostasis in normal and ascorbate-deficient guinea pigs. Biochem
        Pharmacol 50: 243-251. http://dx.doi.org/10.1016/0006-2952(95)00122-G.
Kodavanti, UP: Thomas, R: Ledbetter, AD: Schladweiler,  MC: Shannahan, JH: Wallenborn, JG: Lund, AK:
        Campen. MJ:  Butler. EO:  Gottipolu. RR:  Nvska. A: Richards. JE: Andrews. D: Jaskot.  RH: McKee. J:
        Kotha. SR: Patel. RB:  Parianandi. NL. (2011). Vascular and cardiac impairments in rats  Inhaling ozone
        and diesel exhaust particles. Environ Health Perspect 119: 312-318.
        http://dx.doi. org/10.1289/ehp. 1002386.
Koenig. JQ: Covert. PS: Marshall. SG: Van Belle. G: Pierson. WE. (1987). The effects of ozone and nitrogen
        dioxide on pulmonary function in healthy and in asthmatic adolescents. Am J Respir Crit Care Med 136:
        1152-1157.
Kreit. JW: Gross.  KB: Moore. TB: Lorenzen. TJ: D'Arcv. J: Eschenbacher.  WL. (1989). Ozone-induced changes
        in pulmonary function and bronchial responsiveness in asthmatics. J Appl Physiol 66:  217-222.
Krishna. MT: Springall. D:  Meng. Q. -H: Withers.  N: Macleod. D:  Biscione. G: Frew. A: Polak. J: Holgate.  S.
        (1997). Effects of ozone on epithelium and sensory nerves in the bronchial mucosa of healthy humans.
        Am J Respir Crit Care Med 156: 943-950.
Larsen. ST: Matsubara. S: McConville. G: Poulsen. SS: Gelfand. EW. (2010). Ozone increases airway
        hyperreactivity and mucus hyperproduction in mice previously exposed to allergen. J Toxicol Environ
        Health A 73: 738-747.  http://dx.doi.Org/10.1080/15287391003614034.
Lav. JC: Alexis. NE: Kleeberger. SR: Roubev. RA: Harris. BD: Bromberg. PA: Hazucha. MJ: Devlin. RB: Peden.
        DB. (2007).  Ozone enhances markers of innate  immunity and antigen presentation on airway
        monocytes in healthy individuals. J Allergy Clin Immunol 120: 719-722.
        http://dx.doi.0rg/10.1016/i.iaci.2007.05.005.
Draft - Do Not Cite or Quote                        5-86                                  September 2011

-------
Li. YF: Gauderman. WJ: Avol. E: Dubeau. L: Gilliland. FD. (2006d). Associations of tumor necrosis factor G-
        308Awith childhood asthma and wheezing. Am J Respir Grit Care Med 173: 970-976.
        http://dx.doi.org/10.1164/rccm.200508-1256OC.
Li. Z: Potts. EN: Piantadosi. CA: Foster. WM: Hollingsworth. JW. (2010). Hyaluronan fragments contribute to the
        ozone-primed immune response to lipopolysaccharide. J Immunol 185: 6891-6898.
        http://dx.doi. orq/10.4049/iimmunol. 1000283.
Linn. WS: Buckley. RD: Spier. CE: Blessev. RL: Jones. MP: Fischer. DA: Hackney. JD. (1978). Health effects of
        ozone exposure in asthmatics. Am Rev Respir Dis 117: 835-843.
Linn. WS: Medwav. DA: Anzar. UT: Valencia. LM: Spier. CE: FS-D.  T: Fischer. DA: Hackney. JD. (1982a).
        Persistence of adaptation to ozone in volunteers exposed  repeatedly for six weeks. Am Rev Respir Dis
        125: 491-495.
London. SJ. (2007). Gene-air pollution interactions in asthma. Proc Am Thorac Soc 4: 217-220.
        http://dx.doi.orq/10.1513/pats.200701-031AW.
Long, NC: Suh, J: Morrow, JD: Schiestl, RH: Krishna Murthv, GG: Brain, JD: Frei, B. (2001). Ozone causes lipid
        peroxidation but little antioxidant depletion in exercising and nonexercising hamsters. J Appl Physiol 91:
        1694-1700.
Longphre, M: Zhang, L, -Y: Harkema, JR: Kleeberger, SR. (1999). Ozone-induced pulmonary inflammation and
        epithelial proliferation are partially mediated by PAR J Appl Physiol 86: 341-349.
Maniar-Hew. K: Postlethwait.  EM:  Fanucchi. MV: Ballinger. CA: Evans. MJ: Harkema. JR: Carey. SA:
        McDonald, RJ: Bartolucci, AA: Miller, LA. (2011). Postnatal episodic ozone results in persistent
        attenuation of pulmonary  and peripheral blood responses to LPS  challenge. Am J Physiol Lung Cell
        Mol Physiol 300: L462-L471. http://dx.doi.ora/10.1152/aiplung.00254.2010.
Matsumura, Y. (1970). The effects of ozone, nitrogen dioxide,  and sulfur dioxide on the experimentally induced
        allergic respiratory disorder in guinea pigs: II. The effects of ozone on the absorption and the retention
        of antigen in the lung. Am Rev Respir Dis 102: 438-443.
McBride, JT (1992). Architecture of the tracheobronchial tree. In RA Parent (Ed.), Comparative biology of the
        normal lung (pp. 49-61). Boca Raton, FL: CRC Press.
McDonnell. WF: Horstman. DH: Hazucha. MJ: Seal. E. Jr: Haak. ED: Salaam. SA: House. DE. (1983).
        Pulmonary effects of ozone exposure during exercise: Dose-response characteristics. J Appl Physiol
        54: 1345-1352.
McKinnev. WJ: Jaskot. RH: Richards. JH: Costa. PL: Dreher.  KL. (1998). Cytokine mediation of ozone-induced
        pulmonary adaptation. Am J Respir Cell Mol Biol 18:  696-705.
Mercer. RR: Aniilvel. S: Miller. FJ:  Crapo. JD. (1991). Inhomogeneity of ventilatory unit volume and its effects on
        reactive gas uptake. J Appl Physiol 70: 2193-2205.
Mercer, RR: Russell, ML: Crapo, JD. (1992). Mucous lining layers in human and rat  airways [Abstract]. Am Rev
        Respir Dis 145: A355.
Mikerov. AN: Hague. R: Gan. X: Quo. X: Phelps. PS: Floros. J. (2008a). Ablation of SP-A has a negative impact
        on the susceptibility of mice to Klebsiella pneumoniae infection after ozone  exposure: Sex differences.
        Respir Res 9: 77. http://dx.doi.org/10.1186/1465-9921-9-77.
Mikerov. AN: Umstead. TM: Gan. X: Huang. W: Quo. X: Wang. G: Phelps. PS: Floros. J. (2008b). Impact of
        ozone exposure on the phagocytic activity of human surfactant protein A (SP-A) and SP-A variants. Am
        J Physiol Lung Cell Mol Physiol 294: L121-L130. http://dx.doi.org/10.1152/aiplung.00288.2007.
Mikerov. AN: Gan. X: Umstead. TM: Miller. L: Chinchilli. VM: Phelps. PS: Floros. J. (2008c). Sex differences in
        the impact of ozone on survival and alveolar macrophage function of mice after Klebsiella pneumoniae
        infection. Respir Res 9: 24. http://dx.doi.org/10.1186/1465-9921-9-24.
Miller. FJ: McNeal. CA: Kirtz. JM: Gardner.  DE: Coffin. PL: Menzel. DB. (1979). Nasopharyngeal removal of
        ozone in rabbits and guinea pigs. Toxicology 14: 273-281.  http://dx.doi.Org/10.1016/0300-
        483X(79)90009-X.
Miller. FJ: Overton. JH. Jr: Jaskot.  RH: Menzel. DB. (1985). A model of the regional uptake of gaseous
        pollutants in the  lung: I. The sensitivity of the uptake of ozone in the human lung to lower respiratory
        tract secretions and exercise. Toxicol Appl Pharmacol 79: 11-27. http://dx.doi.org/10.1016/0041-
        008X(85)90364-3.
Draft - Do Not Cite or Quote                        5-87                                  September 2011

-------
Miller. FJ: Overton. JH: Gerritv. TR: Graham. RC. (1988). Interspecies dosimetry of reactive gases. In U Mohr; D
        Dungworth;  R McClellan; G Kimmerle; W Stober; J Lewkowski (Eds.), Inhalation toxicology:  The design
        and interpretation of inhalation studies and their use in risk assessment (pp. 139-155). New  York,  NY:
        Springer-Verlag.
Miller, FJ: Kimbell, JS. (1995). Regional dosimetry of inhaled reactive gases. In RO McClellan; RF Henderson
        (Eds.), Concepts in inhalation toxicology (2nd ed., pp. 257-287). Washington, DC: Taylor & Francis.
Miller. FJ. (1995). Uptake and fate of ozone in the respiratory tract. Toxicol Lett 82-83: 277-285.
Molfino, NA: Wright,  SC: Katz, I: Tarlo, S: Silverman, F: McClean, PA: Szalai, JP: Raizenne, M: Slutskv, AS:
        Zamel. N. (1991). Effect of low concentrations of ozone on inhaled allergen responses in asthmatic
        subjects. Lancet 338: 199-203.
Mudway, IS: Housley, D: Eccles,  R: Richards, RJ: Datta, AK: Tetley, TD: Kelly, FJ. (1996). Differential depletion
        of human respiratory tract antioxidants in response to ozone challenge. Free Radic Res 25:  499-513.
Mudwav. IS: Kelly. FJ. (1998). Modeling the interactions of ozone with pulmonary epithelial lining fluid
        antioxidants. Toxicol Appl Pharmacol 148: 91-100.
Mudwav. IS: Blomberg. A: Frew. AJ:  Holgate. ST: Sandstrom. T: Kelly. FJ. (1999a). Antioxidant consumption
        and repletion kinetics in  nasal lavage fluid following exposure of healthy human volunteers to ozone.
        EurRespirJ 13: 1429-1438.
Mudwav. IS: Krishna. MT:  Frew. AJ:  MacLeod. D: Sandstrom. T: Holgate. ST: Kelly. FJ. (1999b). Compromised
        concentrations of ascorbate in fluid lining the respiratory tract in human subjects after exposure to
        ozone. Occup Environ Med 56: 473-481.
Mudwav. IS: Kelly. FJ. (2000). Ozone and the lung: A sensitive issue. Mol Aspects Med 21: 1-48.
Mudwav. IS: Stenfors. N: Blomberg. A: Helledav.  R: Punster. C: Marklund. SL: Frew. AJ: Sandstrom.  T: Kelly.
        FJ. (2001). Differences in basal airway antioxidant concentrations are not predictive of individual
        responsiveness to ozone: A comparison of healthy and mild asthmatic subjects. Free Radic  Biol Med
        31: 962-974.
Mudwav, IS: Kelly, FJ. (2004b). An investigation of inhaled ozone dose and the magnitude of airway
        inflammation in healthy adults: Online data supplement. Am J Respir Crit Care Med  169: 1089-1095.
        http://dx.doi.org/10.1164/rccm.200309-1325PP.
Mudwav, IS: Behndig, AF: Helledav,  R: Pourazar, J: Frew, AJ: Kelly, FJ: Blomberg, A. (2006). Vitamin
        supplementation does not protect against symptoms in ozone-responsive subjects. Free Radic Biol
        Med 40:  1702-1712.
Murphy, RC: Johnson, KM. (2008). Cholesterol, reactive oxygen species, and the formation of biologically  active
        mediators. J Biol Chem 283: 15521-15525.  http://dx.doi.orq/10.1074/ibc.R700049200.
Mvatt. L: Kossenians. W: Sahav. R: Eis. A: Brockman.  D. (2000). Oxidative stress causes vascular dysfunction
        in the placenta. J Matern Fetal Med 9: 79-82. http://dx.doi.org/10.1002/(SICI)1520-
        6661(200001/02)9:1<79::AID-MFM16>3.0.CO:2-O.
Nishivama. H: Ikeda. H: Kaneko. T: Fu. L: Kudo.  M: Ito. T: Okubo. T. (1998). Neuropeptides mediate  the ozone-
        induced increase in the permeability of the tracheal mucosa in guinea pigs. Am J Physiol 275: L231-
        L238.
Nodelman. V: Ultman. JS. (1999). Longitudinal distribution of chlorine absorption in human airways: A
        comparison  to ozone absorption. J Appl Physiol 87: 2073-2080.
Noviski. N: Brewer. JP: Skornik. WA: Galli. SJ: Drazen. JM: Martin. TR. (1999). Mast cell activation is not
        required  for  induction of airway hyperresponsiveness by ozone in mice. J Appl Physiol 86: 202-210.
O'Byrne, P: Walters,  E: Gold, B: Aizawa,  H: Fabbri, L: Alpert,  S: Nadel, J: Holtzman, M. (1983). Neutrophil
        depletion inhibits airway hyperresponsiveness induced by ozone exposure. Am Rev Respir Dis 130:
        214-219.
O'Byrne, PM: Walters, EH: Aizawa, H: Fabbri, LM: Holtzman, MJ: Nadel, JA. (1984).  Indomethacin inhibits the
        airway hyperresponsiveness but not the neutrophil influx induced by ozone in dogs.  Am Rev Respir Dis
        130: 220-224.
Oslund, KL: Hyde, DM: Putney, LF: Alfaro, MF: Walbv, WF: Tyler, NK: Schelegle, ES. (2008). Activation of
        neurokinin-1 receptors during ozone inhalation contributes to epithelial injury and repair. Am J Respir
        Cell Mol  Biol 39: 279-288. http://dx.doi.ora/10.1165/rcmb.2008-0009OC.
Draft - Do Not Cite or Quote                        5-88                                  September 2011

-------
Oslund. KL: Hyde. DM: Putney. LF: Alfaro. MF: Walbv. WF: Tyler. NK: Schelegle. ES. (2009). Activation of
        calcitonin gene-related peptide receptor during ozone inhalation contributes to airway epithelial injury
        and repair. Toxicol Pathol 37: 805-813. http://dx.doi.org/10.1177/0192623309345691.
Overton. JH: Graham. RC: Miller. FJ. (1987). A model of the regional uptake of gaseous pollutants in the lung: II.
        The sensitivity of ozone uptake in laboratory animal lungs to anatomical and ventilatory parameters.
        Toxicol Appl Pharmacol 88: 418-432. http://dx.doi.org/10.1016/0041-008X(87)90216-X.
Overton. JH: Graham. RC. (1989).  Predictions of ozone absorption in human lungs from newborn to adult.
        Health Phys 1: 29-36.
Overton. JH: Graham. RC: Menache. MG: Mercer. RR: Miller. FJ.  (1996). Influence of tracheobronchial region
        expansion and volume on  reactive gas uptake and interspecies dose extrapolations. Inhal Toxicol 8:
        723-745.
Paquette. NC: Zhang. L. -Y:  Ellis. WA: Scott. AL: Kleeberger. SR.  (1996). Vitamin A deficiency enhances ozone-
        induced lung injury.  Am J Physiol 270: L475-L482.
Park, JW: Taube, C: Swasey, C: Kodama, T: Joetham, A: Balhorn, A: Takeda,  K: Miyahara, N: Allen, CB:
        Dakhama. A: Kim. SH: Dinarello. CA: Gelfand. EW. (2004). lnterleukin-1 receptor antagonist attenuates
        airway hyperresponsiveness following exposure to ozone. Am J Respir Cell Mol Biol 30: 830-836.
        http://dx.doi.org/10.1165/rcmb.2003-0373OC.
Passannante. AN: Hazucha. MJ:  Bromberg. PA:  Seal.  E: Folinsbee. L: Koch. G. (1998). Nociceptive
        mechanisms modulate ozone-induced human lung function decrements. J Appl Physiol 85: 1863-1870.
Pearson, AC: Bhalla, DK. (1997). Effects of ozone on macrophage adhesion in vitro and epithelial and
        inflammatory responses in vivo: The role of cytokines. J Toxicol Environ Health 50: 143-157.
Peden. DB: Setzer. RW. Jr: Devlin. RB.  (1995). Ozone exposure has both a priming effect on allergen-induced
        responses and an intrinsic inflammatory action in the nasal airways of perennially allergic asthmatics.
        Am J Respir Grit Care Med 151: 1336-1345.
Peden. DB. (2011). The role of oxidative stress and innate immunity in O(3) and endotoxin-induced human
        allergic airway disease. Immunol Rev 242: 91-105. http://dx.doi.Org/10.1111/i.1600-065X.2011.01035.x.
Perepu. RS:  Garcia. C: Postal. D: Sethi. R. (2010). Enhanced death signaling in ozone-exposed ischemic-
        reperfused hearts. Mol Cell Biochem 336: 55-64. http://dx.doi.orq/10.1007/s11010-009-0265-4.
Pereyra-Munoz, N: Rugerio-Vargas, C: Angoa-Perez, M: Borgonio-Perez, G: Rivas-Arancibia, S. (2006).
        Oxidative damage in substantia nigra and striatum of rats chronically  exposed  to ozone. J Chem
        Neuroanat31: 114-123. http://dx.doi.Org/10.1016/i.ichemneu.2005.09.006.
Perez-Gil, J. (2008). Structure of pulmonary surfactant membranes and films:  The  role  of proteins and lipid-
        protein interactions. Biochim Biophys Acta 1778:  1676-1695.
        http://dx.doi.0rg/10.1016/i.bbamem.2008.05.003.
Pichavant, M: Goya, S: Meyer, EH: Johnston, RA: Kim, HY: Matangkasombut, P: Zhu,  M: Iwakura, Y: Savage,
        PB:  DeKruvff. RH: Shore. SA: Umetsu. DT (2008). Ozone exposure in a mouse model  induces airway
        hyperreactivity that requires the presence of natural killer T cells and  IL-17. J Exp Med 205: 385-393.
        http://dx.doi. org/10.1084/jem.20071507.
Picher. M: Burch. LH:  Boucher. RC. (2004). Metabolism of P2 receptor agonists in  human airways: Implications
        for mucociliary clearance and cystic fibrosis. J Biol Chem 279: 20234-20241.
        http://dx.doi. org/10.1074/jbc. M400305200.
Plopper. CG: Hatch. GE: Wong. V:  Duan. X: Weir. AJ: Tarkington.  BK: Devlin.  RB:  Becker. S: Buckpitt. AR.
        (1998). Relationship of inhaled ozone concentration to acute tracheobronchial  epithelial injury, site-
        specific ozone dose and glutathione depletion in rhesus monkeys. Am J Respir Cell Mol Biol 19: 387-
        399.
Plopper. CG: Mango. GW: Hatch. GE: Wong. VJ: Toskala. E: Reynolds. SD: Tarkington. BK: Stripp. BR. (2006).
        Elevation of susceptibility to ozone-induced acute tracheobronchial injury  in transgenic  mice deficient in
        Clara cell secretory  protein. Toxicol Appl Pharmacol 213:  74-85.
        http://dx.doi.0rg/10.1016/i.taap.2005.09.003.
Plopper, CG: Smiley-Jewell,  SM:  Miller,  LA: Fanucchi,  MV: Evans, MJ: Buckpitt, AR: Avdalovic, M: Gershwin,
        LJ: Joad.  JP: Kaiekar. R: Larson. S: Pinkerton. KE: Van Winkle. LS: Schelegle. ES: Pieczarka. EM: Wu.
        R: Hyde. DM. (2007). Asthma/allergic airways disease: Does postnatal exposure to environmental
        toxicants promote airway pathobiology? Toxicol Pathol 35: 97-110.
        http://dx.doi. org/10.1080/01926230601132030.
Draft - Do Not Cite or Quote                        5-89                                  September 2011

-------
Postlethwait. EM: Langford. SD: Bidani. A. (1994). Determinants of inhaled ozone absorption in isolated rat
        lungs. Toxicol Appl Pharmacol 125:  77-89.
Postlethwait. EM: Cueto. R: Velsor. LW: Prvor. WA. (1998). OS-induced formation of bioactive lipids: Estimated
        surface concentrations and lining layer effects. Am J Physiol 274: L1006-L1016.
Postlethwait, EM: Joad, JP: Hyde, DM: Schelegle, ES: Brie, JM: Weir, AJ: Putney, LF: Wong, VJ: Velsor, LW:
        Plopper. CG. (2000). Three-dimensional mapping of ozone-induced acute cytotoxicity in
        tracheobronchial airways of isolated perfused rat lung. Am J Respir Cell Mol Biol 22: 191-199.
Postlethwait, EM: Ultman, JS. (2001). Airspace surface chemistry mediates OS-induced lung injury. Hum Ecol
        Risk Assess 7:  1145-1159. http://dx.doi. orq/10.1080/20018091094907.
Prvor. WA. (1976). Free radical reactions in  biology: Initiation of lipid autoxidation by ozone and nitrogen
        dioxide. Environ Health Perspect 16: 180-181.
Prvor. WA: Giamalva. DH: Church. PR (1984). Kinetics of ozonation. 2. Amino acids and model compounds in
        water and comparisons to rates in nonpolar solvents. J Am Chem Soc 106: 7094-7100.
Pryor, WA: Das, B: Church, PR (1991). The  ozonation of unsaturated fatty acids:  Aldehydes and hydrogen
        peroxide as products and possible mediators of ozone toxicity. Chem Res Toxicol 4: 341-348.
Prvor. WA. (1992). How far does ozone penetrate into the pulmonary air/tissue boundary before it reacts? Free
        Radic Biol Med 12: 83-88. http://dx.doi.org/10.1016/0891 -5849(92)90060-T.
Prvor. WA. (1994). Mechanisms of radical formation from  reactions of ozone with  target molecules in the lung.
        Free Radic Biol Med 17: 451-465.
Pryor, WA: Bermudez, E: Cueto,  R: Squadrito, GL. (1996). Detection of aldehydes in bronchoalveolar lavage of
        rats exposed to ozone. Toxicol Sci 34: 148-156.
Pulfer. MK:  Murphy. RC. (2004). Formation of biologically  active oxysterols during ozonolysis of cholesterol
        present in lung surfactant. J Biol Chem 279: 26331-26338.
Pulfer. MK: Taube. C: Gelfand. E:  Murphy. RC. (2005). Ozone exposure in vivo and formation of biologically
        active oxysterols in the lung. J Pharmacol Exp Ther 312: 256-264.
Que, LG: Stiles, JV: Sundy, JS: Foster, WM. (2011). Pulmonary function, bronchial reactivity, and epithelial
        permeability are response phenotypes to ozone and develop differentially in healthy humans. J Appl
        Physiol 111: 679-687. http://dx.doi.ora/10.1152/iapplphvsiol.00337.2011.
Rashba-Step, J: Tatoyan, A:  Duncan, R: Ann, D: Pushpa-Rehka, TR: Sevanian, A. (1997).  Phospholipid
        peroxidation induces cytosolic phospholipase A2  activity: Membrane effects versus enzyme
        phosphorylation. Arch Biochem Biophys 343: 44-54. http://dx.doi.orq/10.1006/abbi. 1997.0134.
Reeser,  WH: Lee, GM: Taylor, A: Wang, l_: Arnold, SF: Ultman, JS: Ben-Jebria, A. (2005). Uptake of ozone in
        human lungs and its relationship to  local physiological response. Inhal Toxicol 17: 699-707.
        http://dx.doi. orq/10.1080/08958370500224433.
Rigas, ML: Ben-Jebria, A: Ultman, JS. (1997). Longitudinal distribution of ozone absorption in the lung: Effects
        of nitrogen dioxide, sulfur dioxide, and ozone exposures. Arch Environ Occup Health 52:  173-178.
Rigas. ML: Catlin. SN: Ben-Jebria. A: Ultman. JS. (2000).  Ozone uptake in the intact human respiratory tract:
        Relationship between inhaled dose  and actual dose. J Appl Physiol 88: 2015-2022.
Ritz. B: Yu.  F: Chapa. G: Fruin. S. (2000). Effect of air pollution on preterm birth among children born in
        Southern California between 1989 and 1993. Epidemiology 11: 502-511.
Ritz, B: Yu,  F: Fruin, S: Chapa, G: Shaw, GM: Harris, JA. (2002). Ambient air pollution and risk of birth defects in
        Southern California. Am J Epidemiol 155: 17-25.
Romieu. I: Meneses. F:  Ramirez. M: Ruiz. S: Padilla. RP:  Sienra. JJ: Gerber. M: Grievink.  L: Dekker. R: Walda.
        I: Brunekreef, B. (1998a). Antioxidant supplementation and respiratory functions among workers
        exposed to high levels of ozone. Am J Respir Grit Care Med 158: 226-232.
Romieu. I: Sienra-Monge. JJ: Ramirez-Aguilar. M: Tellez-Roio. MM: Moreno-Macias. H: Reyes-Ruiz. Nl: Del
        Rio-Navarro, BE: Ruiz-Navarro, MX: Hatch, G: Slade, R: Hernandez-Avila, M. (2002). Antioxidant
        supplementation and lung functions among children with asthma exposed to high levels of air
        pollutants. Am J Respir Grit Care Med 166: 703-709.
Romieu, I: Sienra-Monge, JJ: Ramirez-Aguilar, M: Moreno-Macias, H: Reyes-Ruiz, Nl: Estela del Rio-Navarro,
        B: Hernandez-Avila. M: London. SJ. (2004a). Genetic polymorphism of GSTM1 and antioxidant
        supplementation influence lung function in relation to ozone  exposure in  asthmatic children  in Mexico
        City. Thorax 59: 8-10.
Draft - Do Not Cite or Quote                        5-90                                  September 2011

-------
Roux. E: Hvvelin. J. -M: Savineau. J. -P: Marthan. R. (1999). Human isolated airway contraction: Interaction
       between air pollutants and passive sensitization. Am J Respir Crit Care Med 160: 439-445.
Rutkowski. JM: Santiag. LY: Ben-Jebria. A: Ultman. JS. (2011). Comparison of ozone-specific (OZAC) and
       oxygen radical (ORAC) antioxidant capacity assays for use with nasal lavage fluid. Toxicol In Vitro 25:
       1406-1413.  http://dx.doi.0rg/10.1016/i.tiv.2011.04.008.
Sagai. M: Arakawa. K:  Ichinose. T: Shimoio. N. (1987). Biochemical effects on combined gases of nitrogen
       dioxide and  ozone: I. Species differences of lipid peroxides and phospholipids in lungs. Toxicology 46:
       251-265. http://dx.doi.0rg/10.1016/0300-483X(87)90207-1.
Samet. JM: Hatch. GE: Horstman. D:  Steck-Scott. S: Arab. L: Bromberg. PA: Levine.  M: McDonnell. WF: Devlin.
       RB. (2001).  Effect of antioxidant supplementation on ozone-induced lung injury in human subjects. Am
       J Respir Crit Care Med 164: 819-825.
Santiago-Lopez. D: Bautista-Martfnez. JA: Reyes-Hernandez. Cl: Aguilar-Martfnez. M: Rivas-Arancibia. S.
       (2010). Oxidative stress, progressive damage in the substantia nigra and plasma dopamine oxidation,
       in rats chronically exposed to ozone. Toxicol Lett 197: 193-200.
       http://dx.doi. orq/10.1016/i.toxlet.2Q10.05.020.
Santiago. LY:  Hann.  MC:  Ben-Jebria.  A: Ultman. JS. (2001). Ozone absorption in the human nose during
       unidirectional airflow. J Appl Physiol 91: 725-732.
Sarangapani.  R: Gentry. PR: Covington. TR: Teeguarden.  JG: Clewell  HJ. III. (2003).  Evaluation of the potential
       impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of vapors.
       Inhal  Toxicol 15:  987-1016.
Sathishkumar. K: XI. X: Martin. R: Uppu. RM. (2007a).  Cholesterol secoaldehyde, an ozonation product of
       cholesterol,  induces amyloid  aggregation and  apoptosis in murine GT1-7 hypothalamic neurons. J
       Alzheimers Dis 11: 261-274.
Sathishkumar. K: Murthv. SN: Uppu. RM. (2007b). Cytotoxic effects of oxysterols produced during ozonolysis of
       cholesterol in murine GT1-7 hypothalamic neurons. Free Radic Res 41: 82-88.
       http://dx.doi. org/10.1080/10715760600950566.
Sathishkumar. K: Gao. X: Raghavamenon. AC: Parinandi. N: Prvor. WA: Uppu. RM. (2009). Cholesterol
       secoaldehyde induces apoptosis in  H9c2 cardiomyoblasts through reactive oxygen species involving
       mitochondrial and death receptor pathways. Free Radic Biol Med 47: 548-558.
       http://dx.doi.0rg/10.1016/i.freeradbiomed.2009.05.020.
Sato. S: Shimura. S: Hirose. T: Maeda. S:  Kawakami. M: Takishima. T: Kimura. S. (1980). Effects of long-term
       ozone exposure  and dietary vitamin E in rats. Tohoku J Exp Med 130: 117-128.
Sawyer. K: Brown. J: HazuchaM: Bennett. WD. (2007). The effect of exercise on  nasal uptake of ozone in
       healthy human adults. J Appl Physiol 102: 1380-1386.
       http://dx.doi.org/10.1152/iapplphysiol.00269.2006.
Scannell.  C: Chen. L: Aris. RM: Tager. I: Christian. D: Ferrando. R: Welch. B: Kelly. T: Balmes. JR. (1996).
       Greater ozone-induced inflammatory responses in subjects with asthma. Am J Respir Crit Care Med
       154: 24-29.
Schelegle. ES: Adams. WC: Siefkin. AD. (1987). Indomethacin pretreatment reduces ozone-induced pulmonary
       function decrements in human subjects. Am Rev Respir Dis 136: 1350-1354.
Schelegle, ES: Siefkin, AD:  McDonald, RJ. (1991). Time course of ozone-induced neutrophilia in normal
       humans. Am J Respir Crit Care Med 143: 1353-1358.
Schelegle. ES: Carl.  ML:  Coleridge. HM: Coleridge. JCG: Green. JF. (1993). Contribution of vagal afferents to
       respiratory reflexes evoked by acute inhalation of ozone in dogs. J Appl Physiol 74: 2338-2344.
Schelegle. ES: Walbv. WF: Adams. WC. (2007). Time course of ozone-induced changes in breathing pattern in
       healthy exercising humans. J Appl Physiol 102: 688-697.
       http://dx.doi.org/10.1152/iapplphysiol.00141.2006.
Schroter. RC:  Sudlow.  MF (1969). Flow patterns in models of the human bronchial airways. Respir Physiol
       Neurobiol 7: 341-355.
Seltzer, J: Bigby, BG: Stulbarg, M:  Holtzman, MJ:  Nadel, JA: Ueki, IF: Leikauf, GD: Goetzl, EJ: Boushey, HA.
       (1986). OS-induced change in bronchial reactivity to methacholine and airway inflammation in humans.
       J Appl Physiol 60: 1321-1326.
Draft - Do Not Cite or Quote                        5-91                                  September 2011

-------
Servais. S: Boussouar. A: Molnar. A: Douki. T: Peguignot. JM: Favier. R. (2005). Age-related sensitivity to lung
        oxidative stress during ozone exposure. Free Radio Res 39: 305-316.
        http://dx.doi. orq/10.1080/10715760400011098.
Sharkhuu. T: Doerfler. PL: Copeland. C: Luebke. RW: Gilmour.  Ml. (2011). Effect of maternal exposure to ozone
        on reproductive outcome and immune,  inflammatory, and allergic responses in the offspring. J
        Immunotoxicol 8: 183-194. http://dx.doi.org/10.3109/1547691X.2011.568978.
Shore. SA: Schwartzman. IN: Le Blanc. B: Krishna Murthv.  GG: Doerschuk. CM. (2001). Tumor necrosis factor
        receptor 2 contributes to ozone-induced airway hyperresponsiveness in mice. Am J Respir Crit Care
        Med 164: 602-607.
Shore. SA. (2007). Obesity and  asthma: lessons from animal models. J Appl Physiol 102: 516-528.
Shore, SA: Lang, JE: Kasahara, PI: Lu, Fl_: Verbout, NG: Si, H: Williams, ES: Terry, RD: Lee, A: Johnston, RA.
        (2009). Pulmonary responses to subacute ozone exposure in obese vs. lean mice. J Appl Physiol 107:
        1445-1452. http://dx.doi.org/10.1152/iapplphvsiol.00456.2009.
Sienra-Monge, JJ: Ramirez-Aguilar, M: Moreno-Macias, H:  Reyes-Ruiz,  Ml: Del Rio-Navarro, BE: Ruiz-Navarro,
        MX: Hatch. G: Crissman. K: Slade. R: Devlin. RB:  Romieu. I. (2004). Antioxidant supplementation and
        nasal inflammatory responses among young asthmatics exposed to high levels of ozone. Clin Exp
        Immunol 138: 317-322. http://dx.doi.Org/10.1111/i.1365-2249.2004.02606.x.
Slade. R: Highfill. JW: Hatch. GE. (1989). Effects of depletion of ascorbic acid  or nonprotein sulfhydryls on the
        acute inhalation toxicity of nitrogen dioxide,  ozone, and phosgene. Inhal Toxicol 1: 261-271.
Slade, R: Crissman, K: Norwood, J: Hatch, G. (1993). Comparison of antioxidant substances in bronchoalveolar
        lavage cells and fluid from humans, guinea  pigs, and rats. Exp Lung Res 19: 469-484.
Slade. R: Watkinson. WP: Hatch. GE. (1997). Mouse strain differences in ozone dosimetry and body
        temperature changes. Am J Physiol 272: L73-L77.
Sokol. RZ: Kraft. P: Fowler. IM: Mamet.  R: Kim.  E: Berhane. KT. (2006).  Exposure to environmental ozone alters
        semen quality. Environ  Health Perspect 114: 360-365.
Stenfors, N: Pourazar, J: Blomberg, A: Krishna,  MT:  Mudwav, I: Helleday, R: Kelly, FJ: Frew, AJ: Sandstrom, T.
        (2002). Effect of ozone  on bronchial mucosal inflammation in asthmatic and healthy subjects. Respir
        Med 96: 352-358.
Sterner-Kock, A: Kock, M: Braun, R: Hyde, DM. (2000). Ozone-induced epithelial  injury in the ferret is similar to
        nonhuman primates. Am J Respir Crit Care  Med 162: 1152-1156.
Sun. J: Koto. H: Chung. KF. (1997). Interaction of ozone and allergen challenges on bronchial responsiveness
        and inflammation in sensitised guinea pigs.  Int Arch Allergy Immunol 112: 191-195.
Taylor-Clark. TE: McAlexander. MA: Nassenstein. C: Sheardown. SA: Wilson.  S: Thornton. J: Carr. MJ: Undem.
        BJ. (2008). Relative contributions of TRPA1 and TRPV1 channels in the activation of vagal
        bronchopulmonary C-fibres by the endogenous autacoid 4-oxononenal. J Physiol 586: 3447-3459.
        http://dx.doi. orq/10.1113/|phvsiol.20Q8.153585.
Taylor-Clark. TE: Undem. BJ. (2010).  Ozone activates airway nerves via the selective stimulation of TRPA1 ion
        channels. J Physiol 588: 423-433. http://dx.doi.org/10.1113/iphysiol.20Q9.183301.
Taylor. AB: Lee. GM: Nellore. K: Ben-Jebria. A: Ultman. JS. (2006). Changes in the carbon dioxide expirogram
        in response to ozone exposure. Toxicol Appl Pharmacol 213: 1-9.
        http://dx.doi.0rg/10.1016/i.taap.2005.09.009.
Taylor. AB: Borhan. A: Ultman. JS. (2007). Three-dimensional simulations of reactive gas uptake in single
        airway bifurcations. Ann Biomed Eng 35: 235-249. http://dx.doi.org/10.1007/s10439-006-9195-4.
Tepper, JS:  Costa, PL: Lehmann, JR: Weber, MF: Hatch, GE. (1989). Unattenuated structural and biochemical
        alterations in the rat lung during functional adaptation to ozone.  Am J  Respir Crit Care Med 140: 493-
        501.
Tepper, JS:  Costa, PL: Fitzgerald, S: Doerfler, PL: Bromberg, PA.  (1993). Role of tachykinins in ozone-induced
        acute lung injury in guinea pigs. J Appl  Physiol 75: 1404-1411.
Thomson. E: Kumarathasan. P:  Goegan. P: Aubin. RA: Vincent. R. (2005). Differential regulation of the lung
        endothelin system  by urban particulate matter and ozone. Toxicol Sci  88: 103-113.
Thomson. E: Kumarathasan. P:  Vincent. R. (2006). Pulmonary expression of preproET-1 and preproET-3
        mRNAs is altered reciprocally in rats after inhalation of air pollutants. Exp Biol Med 231: 979-984.
Trenga, CA: Koenig, JQ: Williams, PV. (2001). Dietary antioxidants and ozone-induced bronchial
        hyperresponsiveness in adults with asthma. Arch Environ Occup Health 56:  242-249.
Draft - Do Not Cite or Quote                        5-92                                  September 2011

-------
Trevisani. M: Siemens. J: Materazzi. S: Bautista. DM: Nassini. R: Campi. B: Imamachi. N: Andre. E: Patacchini.
        R: Cottrell, GS: Gatti, R: Basbaum, Al: Bunnett, NW: Julius, D: Geppetti, P. (2007). 4-Hydroxynonenal,
        an endogenous aldehyde, causes pain and neurogenic inflammation through activation of the irritant
        receptor TRPA1. PNAS 104: 13519-13524. http://dx.doi.orq/10.1073/pnas.07Q5923104.
Tsujino,  I: Kawakami, Y: Kaneko, A. (2005). Comparative simulation of gas transport in airway models of rat,
        dog, and human. Inhal Toxicol 17: 475-485. http://dx.doi.ora/10.1080/08958370590964476.
U.S. EPA. (U.S. Environmental Protection Agency). (1996a). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/P-93/004AF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment.
        (EPA/630/P-03/001F). Washington, DC. http://www.epa.gov/cancerguidelines/.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2009f). The U.S. Environmental Protection Agency's
        strategic plan for evaluating the toxicity of chemicals. (EPA/1 OO/K-09/001). Washington, DC.
        http://www.epa.gov/osa/spc/toxicitytesting/docs/toxtest  strategy 032309.pdf.
Ultman.  JS. (1985). Gas transport in the conducting airways. In LA Engel; M Paiva (Eds.), Gas mixing and
        distribution in the lung (pp. 63-136). New York, NY: Marcel Dekker.
Ultman,  JS: Anjilvel,  S. (1990). Monte Carlo simulation of ozone uptake in an asymmetric lung model. In DJ
        Schneck; CL Lucas (Eds.), Biofluid mechanics 3: Proceedings of the third Mid-Atlantic Conference on
        Biofluid Mechanics; October; Blacksburg, VA(pp. 45-52). New York, NY: New York University Press.
Ultman,  JS: Ben-Jebria, A; Hu, S,  -C. (1994). Noninvasive determination of respiratory ozone absorption: The
        bolus-response method. (HEI Research Report 69). Cambridge, MA: Health Effects Institute.
Ultman.  JS: Ben-Jebria. A: Arnold. SF. (2004). Uptake distribution of ozone in human lungs: Intersubject
        variability in physiologic response. (HEI  Research  Report 125). Boston, MA: Health Effects Institute.
        http://pubs.healtheffects.org/view.php?id=70.
Uppu. RM: Cueto. R: Sguadrito. GL: Prvor. WA. (1995). What does ozone react with at the air/lung interface?
        Model studies using human red blood cell membranes. Arch Biochem Biophys 319: 257-266.
Urch. B: Speck. M: Corey.  P: Wasserstein. D: Manno. M: Lukic. KZ: Brook. JR:  Liu. L: Coull. B: Schwartz. J:
        Gold. PR: Silverman. F. (2010). Concentrated ambient fine particles and not ozone induce a systemic
        interleukin-6 response in  humans. Inhal Toxicol 22: 210-218.
        http://dx.doi. orq/10.3109/08958370903173666.
Vagaggini. B: Taccola. M: Clanchetti. S: Carnevali. S: Bartoli.  ML: Bacci. E: Dente. FL: Pi Franco. A: Giannini. D:
        Paggiaro, PL. (2002). Ozone exposure increases eosinophilic airway response induced by previous
        allergen challenge. Am J  Respir Grit Care Med 166: 1073-1077.
Van Bree. L: Dormans. JAM. A: Koren. HS: Devlin. RB: Rombout. PJA. (2002).  Attenuation and recovery of
        pulmonary injury in rats following  short-term, repeated daily exposure to ozone. Inhal Toxicol  14: 883-
        900.
Vancza. EM: Galdanes. K: Gunnison. A: Hatch.  G: Gordon. T (2009). Age, strain, and gender as factors for
        increased sensitivity of the mouse lung to inhaled ozone. Toxicol Sci 107: 535-543.
        http://dx.doi. orq/10.1093/toxsci/kfn253.
Vasu. VT: Oommen. S: Lim. Y: Valacchi. G: Hobson. B: Eiserich. JP: Leonard. SW: Traber.  MG: Cross. CE:
        Gohil,  K. (2010). Modulation of ozone-sensitive genes in alpha-tocopherol transfer protein null mice.
        Inhal Toxicol 22: 1-16. http://dx.doi.org/10.3109/08958370902838145.
Verhein. KG: Hazari. MS: Moulton. BC: Jacobv.  IW: Jacobv. DB: Fryer. AD. (2011). Three days after a single
        exposure to ozone the mechanism of airway hyperreactivity is dependent upon substance P and nerve
        growth factor. Am J Physiol Lung  Cell Mol Physiol  300: L176-L184.
        http://dx.doi.org/10.1152/aiplung.00060.2010.
Vesely, KR; Schelegle, ES; Stovall, MY; Harkema, JR; Green, JF; Hyde, DM. (1999). Breathing pattern
        response and epithelial labeling in ozone-induced  airway injury in neutrophil-depleted rats. Am J Respir
        Cell Mol Biol 20: 699-709.
Draft - Do Not Cite or Quote                        5-93                                  September 2011

-------
Vincent. R: Vu. D: Hatch. G: Poon. R: Dreher. K: Guenette. J: Biarnason. S: Potvin. M: Norwood. J: McMullen.
        E_, (1996a). Sensitivity of lungs of aging Fischer 344 rats to ozone: Assessment by bronchoalveolar
        lavage. Am J Physiol 271: L555-L565.
Vivien E: Raulet. DH:  Moretta. A: Caligiuri. MA: Zitvoqel. L: Lanier. LL: Yokovama. WM: Ugolini. S. (2011).
        Innate or adaptive immunity? The example of natural killer cells. Science 331: 44-49.
        http://dx.doi.org/10.1126/science.1198687.
Vovnow. JA:  Fischer. BM: Zheng. S:  Potts. EN: Grover. AR: Jaiswal. AK: Ghio. AJ:  Foster. WM. (2009).
        NAD(P)H quinone oxidoreductase 1 is essential for ozone-induced oxidative stress in mice and
        humans. Am J Respir Cell Mol Biol 41: 107-113. http://dx.doi.ora/10.1165/rcmb.2008-0381OC.
Wagner. JG:  Hotchkiss. JA: Harkema. JR. (2002).  Enhancement of nasal inflammatory and epithelial responses
        after ozone and allergen coexposure in brown Norway rats. Toxicol Sci 67: 284-294.
Wagner. JG:  Jiang. Q: Harkema. JR: Illek. B: Patel. DP: Ames. BN: Peden.  DB. (2007). Ozone enhancement of
        lower airway allergic  inflammation is prevented by gamma-tocopherol. Free Radic Biol Med 43: 1176-
        1188. http://dx.doi.0rg/10.1016/i.freeradbiomed.2007.07.013.
Wagner. JG:  Harkema. JR: Jiang. Q: Illek. B: Ames. BN: Peden. DB. (2009). Gamma-tocopherol attenuates
        ozone-induced exacerbation of allergic rhinosinusitis in rats. Toxicol Pathol 37: 481-491.
        http://dx.doi. org/10.1177/0192623309335630.
Watkinson. WP: Campen. MJ: Nolan. JP: Costa. PL. (2001). Cardiovascular and systemic responses to inhaled
        pollutants in rodents: Effects of ozone and particulate matter. Environ Health Perspect 109: 539-546.
Watkinson, WP: Campen, MJ: Wichers, LB: Nolan, JP: Costa, PL. (2003). Cardiac and thermoregulatory
        responses to inhaled pollutants in healthy and compromised rodents: Modulation via interaction with
        environmental factors. Environ Res 92: 35-47.
Weibel, ER. (1980). Pesign and structure of the human lung. In AP Fishman (Ed.), Assessment of pulmonary
        function. New York, NY: McGraw-Hill.
Weinmann. GG: Weidenbach-Gerbase. M: Foster. WM: Zacur. H: Frank.  R. (1995a). Evidence for ozone-
        induced small-airway dysfunction: Lack of menstrual-cycle and gender effects. Am J Respir Crit Care
        Med 152: 988-996.
Weinmann. GG: Liu. MC: Proud. P: Weidenbach-Gerbase. M: Hubbard. W: Frank. R. (1995b). Ozone exposure
        in humans: Inflammatory, small and peripheral airway responses. Am J Respir Crit Care Med 152:
        1175-1182.
Welch. RW: Wang. Y: Grossman. A.  Jr: Park. JB:  Kirk. KL: Levine. M. (1995). Accumulation of vitamin C
        (ascorbate) and its oxidized metabolite dehydroascorbic acid occurs by separate mechanisms. J Biol
        Chem 270: 12584-12592. http://dx.doi.org/10.1074/ibc.270.21.12584.
Wenten. M: Gauderman. WJ:  Berhane. K: Lin. PC: Peters. J: Gilliland.  FP. (2009). Functional variants in the
        catalase and myeloperoxidase genes, ambient air pollution,  and respiratory-related school absences:
        An example of epistasis in gene-environment interactions. Am J Epidemiol 170: 1494-1501.
        http://dx.doi.org/10.1093/aie/kwp310.
Wiester, MJ:  Williams, TB: King, ME: Menache, MG: Miller, FJ. (1987). Ozone uptake in awake Sprague-Pawley
        rats. Toxicol Appl Pharmacol 89:  429-437. http://dx.doi. org/10.1016/0041-008X(87)90162-1.
Wiester. MJ:  Tepper. JS: King. ME: Menache. MG: Costa.  PL. (1988). Comparative study of ozone (O3) uptake
        in three strains of rats and in the guinea pig. Toxicol Appl Pharmacol 96: 140-146.
Wiester. MJ:  Tepper. JS: Winsett. PW: Crissman. KM: Richards. JH: Costa. PL. (1996a). Adaptation to ozone in
        rats  and its association with ascorbic acid in the lung. Toxicol Sci 31: 56-64.
Wiester, MJ:  Stevens, MA: Menache, MG: McKee, JL, Jr: Gerritv, TR. (1996c). Ozone uptake in healthy adult
        males during quiet breathing. Toxicol Sci 29: 102-109.
Williams. AS: Issa. R:  Leung.  SY: Puneeta. N: Gregory. P: Ferguson. P: Brvdon. L: Bennett.  I: Adcock. M:
        Chung, KF (2007a). Attenuation  of ozone-induced airway inflammation and hyper-responsiveness by
        c-Jun NH2 terminal kinase inhibitor SP600125. J Pharmacol Exp Ther 322: 351-359.
        http://dx.doi.org/10.1124/ipet.107.121624.
Williams, AS: Leung, SY: Nath, P: Khorasani, NM: Bhavsar, P: Issa, R: Mitchell, JA: Adcock,  IM: Chung, KF.
        (2007b). Role of TLR2, TLR4, and MyP88 in murine ozone-induced airway hyperresponsiveness and
        neutrophilia. J Appl Physiol 103:  1189-1195. http://dx.doi.org/10.1152/iapplphvsiol.00172.2007.
Praft - Po Not Cite or Quote                        5-94                                  September 2011

-------
Williams. AS: Nath. P: Leung. SY: Khorasani. N: McKenzie. ANJ: Adcock. IM: Chung. KF. (2008a). Modulation
        of ozone-induced airway hyperresponsiveness and inflammation by interleukin-13. Eur Respir J 32:
        571-578. http://dx.doi.org/10.1183/09031936.00121607.
Williams. AS: Issa. R: Durham. A: Leung. SY: Kapoun. A: Medicherla. S: Higgins. LS: Adcock. IM: Chung. KF.
        (2008b). Role of p38 mitogen-activated protein kinase in ozone-induced airway hyperresponsiveness
        and inflammation. Eur J Pharmacol 600: 117-122. http://dx.doi.Org/10.1016/i.eiphar.2008.09.031.
Williams. AS: Evnott. PR: Leung. SY: Nath. P: Jupp. R: De Sanctis. GT: Resnick. R: Adcock.  IM: Chung. KF.
        (2009a). Role of cathepsin S in ozone-induced airway hyperresponsiveness and inflammation. Pulm
        Pharmacol Ther 22: 27-32. http://dx.doi.Org/10.1016/i.pupt.2008.11.002.
Wu. W: Doreswamv. V: Diaz-Sanchez. D: Samet. JM: Kesic. M: Dailev.  L: Zhang. W: Jaspers. I: Peden. DB.
        (2011). GSTM1 modulation of IL-8 expression in human bronchial epithelial cells exposed to ozone.
        Free Radic Biol Med 51: 522-529.  http://dx.doi.Org/10.1016/i.freeradbiomed.2011.05.006.
Wu. ZX: Satterfield. BE: Dev. RD. (2003). Substance P released from intrinsic airway neurons contributes to
        ozone-enhanced airway hyperresponsiveness in ferret trachea. J Appl Physiol 95:  742-750.
Wu. ZX: Barker. JS: Batchelor. TP:  Dev. RD. (2008b). Interleukin (IL)-1 regulates ozone-enhanced tracheal
        smooth muscle responsiveness by increasing substance P (SP) production in intrinsic airway neurons
        of ferret. Respir Physiol Neurobiol  164: 300-311. http://dx.doi.Org/10.1016/i.resp.2008.07.019.
Yang. IA: Holz. O: Jorres. RA: Magnussen. H: Barton.  SJ: Rodriguez. S: Cakebread. JA: Hollowav. JW: Holgate.
        ST. (2005a). Association of tumor necrosis factor alpha polymorphisms and ozone-induced change in
        lung function. Am J  Respir Grit Care Med 171: 171-176.
Yokovama. E: Frank. R. (1972). Respiratory uptake of ozone in dogs. Arch Environ Occup Health 25: 132-138.
Yoon. HK: Cho. HY: Kleeberger. SR. (2007). Protective role of matrix metalloproteinase-9 in ozone-induced
        airway inflammation. Environ Health Perspect 115: 1557-1563. http://dx.doi.org/10.1289/ehp.10289.
Yost. BL: Gleich. GJ: Jacobv. DB: Fryer. AD. (2005). The changing role of eosinophils in long-term
        hyperreactivity following a single ozone exposure. Am J Physiol Lung Cell Mol Physiol 289: L627-L635.
        http://dx.doi.org/10.1152/aiplung.00377.2004.
Draft - Do Not Cite or Quote                        5-95                                 September 2011

-------
      6   INTEGRATED  HEALTH  EFFECTS  OF  SHORT-
           TERM  OZONE  EXPOSURE
          6.1    Introduction

 1                   This chapter reviews, summarizes, and integrates the evidence for various health
 2                   outcomes associated with short-term (i.e., hours, days, or weeks) exposures to O3.
 3                   Numerous controlled human exposure, epidemiologic, and toxicological studies have
 4                   permitted evaluation of the relationships of short-term O3 exposure with a range of
 5                   endpoints related to respiratory effects (Section 6.2), cardiovascular effects (Section 6.3),
 6                   and mortality (Sections 6.2, 6.3, and 6.6). A smaller number of studies are available to
 7                   assess the effects of O3 on other physiological systems such as the central nervous system
 8                   (Section 6.4), liver and metabolism (Section 6.5.1), and cutaneous and ocular tissues
 9                   (Section 6.5.2).

10                   Evidence forthe major health effect categories (e.g., respiratory, cardiovascular,
11                   mortality) is described in individual sections that include a brief summary of conclusions
12                   from the 2006 O3 AQCD and an evaluation of recent evidence that is intended to build
13                   upon evidence from previous reviews. Within each section, results are organized by
14                   health endpoint (e.g., lung function, pulmonary inflammation) then by specific scientific
15                   discipline (e.g., controlled human exposure, epidemiology, and toxicology). Each major
16                   section (e.g., respiratory, cardiovascular, mortality) concludes with an integrated
17                   summary of the findings and a conclusion regarding causality. Based upon the framework
18                   described in the Preamble to this ISA, a determination of causality is made for a broad
19                   health effect category, such as  respiratory effects, with coherence and plausibility being
20                   based on the evidence available across disciplines and also across the suite of related
21                   health endpoints, including cause-specific mortality.
          6.2   Respiratory Effects

22                   Based on evidence integrated across human controlled exposure, epidemiologic, and
23                   toxicological studies, the 2006 O3 AQCD concluded that there was clear, consistent
24                   evidence of a causal relationship between short-term O3 exposure and respiratory effects
25                   (U.S. EPA. 2006b). Contributing to this conclusion were consistent and coherent
26                   observations across scientific disciplines of associations of short-term O3 exposures with
27                   pulmonary function decrements and increases in lung  inflammation, lung permeability,
28                   and airway hyperresponsiveness. Collectively, these findings provided biological
      Draft - Do Not Cite or Quote                       6-1                                 September 2011

-------
 1                   plausibility for associations in epidemiologic studies of short-term ambient O3 exposure
 2                   with respiratory symptoms and respiratory-related hospitalizations and emergency
 3                   department (ED) visits.

 4                   Controlled human exposure studies have provided strong and quantifiable exposure-
 5                   response data on the human health effects of O3. The most salient observations from
 6                   studies reviewed in the 1996 and 2006 O3 AQCDs were that:  (1) young healthy adults
 7                   exposed to O3 concentrations > 80 ppb develop significant reversible, transient
 8                   decrements in pulmonary function if minute ventilation (VE) or duration of exposure is
 9                   increased sufficiently; (2) relative to young adults, children experience similar
10                   spirometric responses but lesser symptoms from O3 exposure; (3) relative to young
11                   adults, O3-induced spirometric responses are decreased in older individuals; (4) there is a
12                   large degree of intersubject variability in physiologic and symptomatic responses to O3
13                   but responses tend to be reproducible within a given individual over a period of
14                   several months; (5) subjects exposed repeatedly to O3 for several days experience an
15                   attenuation of spirometric and symptomatic responses on successive exposures, that is
16                   lost after about a week without exposure; and (6) acute O3 exposure initiates an
17                   inflammatory response that may persist for at least 18 to 24 hours postexposure.

18                   Substantial evidence for biologically plausible O3-induced respiratory morbidity has been
19                   derived from the coherence between toxicological and controlled human exposure studies
20                   examining parallel endpoints. For example, O3-induced decrements in lung function have
21                   also been observed in animals, and as in humans, tolerance or attenuation has been
22                   demonstrated in animal models. Both humans and rodents exhibit increased airway
23                   hyperresponsiveness. This is an important consequence of exposure to ambient O3,
24                   because the airways are then predisposed to narrowing upon inhalation of a variety of
25                   ambient stimuli. Additionally, airway hyperresponsiveness tends to resolve more slowly
26                   and appears less subject to attenuation. Increased permeability and inflammation have
27                   been observed in the airways of humans and animals alike after O3 exposure, although
28                   these processes are not necessarily associated with immediate changes in lung function or
29                   hyperresponsiveness. Furthermore, the potential relationship between repetitive bouts of
30                   acute inflammation and the development of chronic respiratory disease is unknown.
31                   Another feature of O3 exposure-related respiratory morbidity is impaired host defense
32                   and reduced resistance to lung infection, which has been strongly supported by
33                   toxicological evidence and to a limited extent by human data. Recurrent respiratory
34                   infection in early life is associated with increased incidence of asthma in humans.

35                   In epidemiologic studies, short-term O3-related respiratory morbidity has been assessed
36                   most frequently using lung function. Several studies of healthy children attending camps
37                   as well as studies of outdoor workers, groups exercising outdoors, and children with
      Draft - Do Not Cite or Quote                        6-2                                 September 2011

-------
 1                   asthma support O3 effects on lung function decrements at ambient levels (U.S. EPA.
 2                   2006b. 1996a). In addition to lung function, ambient O3 exposure has been associated
 3                   with increases in respiratory symptoms (e.g., cough, wheeze, shortness of breath),
 4                   especially in large U.S. panel studies of children with asthma (Gent et al.. 2003;
 5                   Mortimer et al., 2000). The evidence across disciplines for O3 effects on a range of
 6                   respiratory endpoints collectively provides support for epidemiologic studies that have
 7                   demonstrated consistent positive associations between O3 exposure and respiratory
 8                   hospital admissions and ED visits, specifically during the summer or warm months. In
 9                   contrast with other respiratory health endpoints, the association between short-term O3
10                   exposure and respiratory mortality is less clearly indicated. Although O3 has been
11                   consistently associated with nonaccidental and cardiopulmonary mortality, the
12                   contribution of respiratory causes to these findings has been uncertain as the few studies
13                   that have examined mortality specifically from respiratory causes have reported
14                   inconsistent associations with ambient O3 exposures.

15                   As discussed throughout this section, consistent with the strong body of evidence
16                   presented in the 2006 O3 AQCD, recent studies continue to support associations between
17                   short-term O3 exposure and respiratory effects, in particular, lung function decrements in
18                   controlled human exposure studies, airway inflammatory responses in toxicological
19                   studies, and respiratory-related hospitalizations and ED  visits. Recent epidemiologic
20                   studies contribute new evidence on at-risk populations and of associations of ambient O3
21                   exposures with biological markers of airway inflammation and oxidative stress, which is
22                   consistent with the extensive evidence from human controlled exposure and toxicological
23                   studies. Furthermore, extending the potential range of well-established O3-associated
24                   respiratory effects, new multicity studies and a multicontinent study demonstrate
25                   associations between short-term ambient O3 exposure and respiratory-related mortality.
             6.2.1   Lung Function
                     6.2.1.1    Controlled Human Exposure

26                   This section focuses on studies examining O3 effects on lung function and respiratory
27                   symptoms in volunteers exposed, for periods of up to 8 hours to O3 concentrations
28                   ranging from 40 to 500 ppb, while at rest or during exercise of varying intensity.
29                   Responses to acute O3 exposures in the range of ambient concentrations include
30                   decreased inspiratory capacity; mild bronchoconstriction; rapid, shallow breathing
31                   patterns during exercise; and symptoms of cough and pain on deep inspiration (PDI).
32                   Reflex inhibition of inspiration results in a decrease in forced vital capacity (FVC) and
      Draft - Do Not Cite or Quote                        6-3                                 September 2011

-------
 1                  total lung capacity (TLC) and, in combination with mild bronchoconstriction, contributes
 2                  to a decrease in the forced expiratory volume in 1 second (FEVi).

 3                  In studies that have exposed subjects during exercise, the majority of shorter duration
 4                  (< 4-hour exposures) studies utilized an intermittent exercise protocol in which subjects
 5                  rotated between 15-minute periods of exercise and rest. A limited number of 1- to 2-hour
 6                  studies, mainly focusing on exercise performance, have utilized a continuous exercise
 7                  regime. A quasi continuous exercise protocol is common to prolonged exposure studies
 8                  where subjects complete 50-minute periods of exercise followed by 10-minute rest
 9                  periods.

10                  The majority of controlled human exposure studies have been conducted within
11                  chambers, although a smaller number of studies used a facemask to expose subjects to
12                  O3. Little effort has been made herein to differentiate between facemask and chamber
13                  exposures as FEVi and respiratory symptom responses appear minimally affected by
14                  these exposure modalities. Similar responses between facemask and chamber exposures
15                  have been reported for exposures to 80 and 120 ppb O3 (6.6 h, moderate quasi continuous
16                  exercise, 40 L/min) and 300 ppb O3 (2 h, heavy intermittent exercise, 70 L/min) (Adams.
17                  2003a, b, 2002).
18                  The majority of controlled human exposure studies investigating the effects O3  are of a
19                  randomized, controlled, crossover design in which subjects were exposed, without
20                  knowledge of the exposure condition and in random order to clean filtered air (FA; the
21                  control) and, depending on the study, to one or more O3 concentrations. The FA control
22                  exposure provides an unbiased estimate of the effects of the experimental procedures on
23                  the outcome(s) of interest. Comparison of responses following this FA exposure to those
24                  following an O3 exposure allows for estimation of the effects of O3 itself on an outcome
25                  measurement while controlling for independent effects of the experimental procedures.
26                  As individuals may experience small changes in various health endpoints from exercise,
27                  diurnal variation, or other effects in addition to those of O3 during the course of an
28                  exposure, the term "O3-induced" is used herein to designate effects that have been
29                  corrected or adjusted for such extraneous responses as measured during FA exposures.

30                  Spirometry, viz., FEVi, is a common health endpoint used  to assess effects of O3 on
31                  respiratory health in controlled human exposure studies. In considering 6.6 hour
32                  exposures to FA, group mean FEVi changes have ranged from -0.7% (McDonnell et al.,
33                  1991) to 2.7% (Adams. 2006a). On average, across ten 6.6-hour exposure studies, there
34                  has been a  1.0% (n=279) increase in FEVi  (Kim etal.. 2011; Schelegle et al.. 2009;
35                  Adams. 2006a. 2003a. 2002: Adams and Ollison.  1997: Folinsbee et al.. 1994:
36                  McDonnell et al.. 1991; Horstman  et al.. 1990; Folinsbee et al.. 1988). Regardless of the
37                  reason for small changes in FEVi over the  course of FA exposures, whether biologically
      Draft - Do Not Cite or Quote                       6-4                                September 2011

-------
 1                   based or a systematic effect of the experimental procedures, the use of FA responses as a
 2                   control for the assessment of responses following O3 exposure in randomized exposure
 3                   studies serves to eliminate alternative explanations other than those of O3 itself in causing
 4                   the measured responses.

 5                   Considering FEVi responses in young healthy adults, an O3-induced change in FEVi is
 6                   typically the difference between the decrement observed with O3 exposure and the
 7                   improvement observed with FA exposure. Noting that some healthy individuals
 8                   experience small improvements while others have small decrements in FEVi following
 9                   FA exposure, investigators have used the randomized, crossover design with each subject
10                   having their own control exposure to FA to discern relatively small effects with certainty
11                   since alternative explanations for these effects are controlled for by the nature of the
12                   experimental design. The utility of FA control exposures becomes more apparent when
13                   considering individuals with respiratory disease. The occurrence of exercise-induced
14                   bronchospasm is well recognized to in patients with asthma and COPD and may be
15                   experienced during both FA and O3 exposures. Absent correction for FA responses,
16                   exercise-induced changes in FEVi could be mistaken for responses due to O3. This
17                   biological phenomenon serves as an example to emphasize the need for a proper control
18                   exposure in assessing the effects of O3 as well as the role of this control in eliminating the
19                   influence of other factors on the outcomes of interest.


                     Pulmonary Function Effects of Ozone Exposure in Healthy Subjects

                        Acute Exposure of Healthy Subjects
20                   The majority of controlled human exposure studies have investigated the effects of
21                   exposure to O3 in young healthy nonsmoking adults (18-35 years of age). These studies
22                   typically use fixed concentrations of O3 under carefully  regulated environmental
23                   conditions and subject activity levels. The magnitude of respiratory effects (decrements
24                   in spirometry and symptomatic response) in these individuals is a function of O3
25                   concentration (C), minute ventilation (VE), and exposure duration (time). Any physical
26                   activity will increase minute ventilation and therefore the dose of inhaled O3. Dose of
27                   inhaled O3 to the lower airways is also increased due to  a shift from nasal to oronasal
28                   breathing with a consequential decrease in O3 scrubbing by the upper airways. Thus, the
29                   intensity of physiological response following an acute exposure will be strongly
30                   associated with minute ventilation.

31                   The product of C x VE  x time, although actually a measure of exposure, is commonly
32                   used as a surrogate for O3 dose to the respiratory tract in controlled human exposure
33                   studies. The delivery of O3 to the lower respiratory tract varies as a function of breathing
34                   conditions (route and pattern). And, the dose of O3 to the lower respiratory tract can vary

      Draft - Do Not Cite or Quote                       6-5                                 September 2011

-------
 1                   between similarly exposed individuals. In support of the use of the product (C * VE *
 2                   time) as a surrogate for O3 dose, differences in FEVi responses among young healthy
 3                   adults (32 M, 28 F) exposed to O3 (250 ppb, 30 L/min, 2 h) do not appear to be explained
 4                   by intersubject differences in the fraction of inhaled O3 retained in the lung (Ultman et
 5                   al.. 2004). Using the product of C * VE * time as a surrogate for O3 dose is also useful in
 6                   distinguishing between the well defined and characterized exposure of subjects in
 7                   controlled human exposure studies as opposed to the use of ambient O3 concentration to
 8                   characterize exposure in epidemiologic studies.

 9                   For healthy young adults exposed at rest for 2 hours, 500 ppb is the lowest O3
10                   concentration reported to produce a statistically significant O3-induced group mean FEVi
11                   decrement of 6.4% (n=10) (Folinsbee et al.. 1978) to 6.7% (n=13) (Horvath et al.. 1979).
12                   Airway resistance was not clearly affected during at-rest exposure to these
13                   O3 concentrations. When exposed to 200 ppb  for 2.25 h during intermittent periods  of rest
14                   and brisk walking, young healthy subjects (83 M, 55 F) show a statistically significant
15                   group mean FEVi decrement of 8.8% following O3 exposure (Que etal.). For exposures
16                   of 1-2 hours to >  120 ppb O3, statistically significant symptomatic responses and effects
17                   on FEVi are observed when VE is sufficiently increased by exercise (McDonnell et  al..
18                   1999). For instance, 5% of young healthy adults exposed to 400 ppb for 2 h during rest
19                   experienced pain on deep inspiration.  Respiratory symptoms were not observed at lower
20                   exposure concentrations (120-300 ppb) or with only 1 h of exposure. However, when
21                   exposed to 120 ppb for 2 h during moderate intermittent exercise, 9% of individuals
22                   experienced pain on deep inspiration,  5% experienced cough, and 4% experienced
23                   shortness of breath. With very heavy continuous exercise (VE = 89 L/min), an O3-induced
24                   group mean decrement of 9.7% in FEVi has been reported for healthy young adults
25                   exposed for 1 hour to  120 ppb O3 (Gong et al., 1986). Symptoms are present and
26                   decrements in forced expiratory volumes and  flows occur at 160-240 ppb O3 following 1
27                   hour of continuous heavy exercise (VE « 55 to 90 L/min (Gong et al., 1986; Avol et al..
28                   1984; Folinsbee et al.. 1984; Adams and Schelegle. 1983) and following 2 hours of
29                   intermittent heavy exercise (VE « 65-68 L/min) (Linn etal.. 1986; Kulleetal..  1985;
30                   McDonnell et al.. 1983). With heavy intermittent exercise (15-min intervals of rest and
31                   exercise [VE = 68 L/min]), symptoms of breathing discomfort and a group mean O3-
32                   induced decrement of 3.4% in FEVi occurred in young healthy adults exposed for 2
33                   hours to 120 ppb  O3 (McDonnell et al.. 1983).'
        1 In total, subjects were exposed to O3 for 2.5 hours. Intermittent exercise periods, however, were only conducted for the first 2
      hours of exposure and FENA, was determined 5 minutes after the exercise was completed.
      Draft - Do Not Cite or Quote                       6-6                                September 2011

-------
                        20 -
•D 5^
« **
o c

5 E
•- £
i «

§'

0>
  LLJ
                        15 -
                        10 -
                         5 -
    * Adams (2006)

    A Adams (2003)

    * Adams (2002)

    0 Folinsbee etal. (1988)

    n Horstman etal. (1990)

    O McDonnell etal. (1991)
   	McDonnell etal. (2007)
                          0.02      0.04      0.06      0.08      0.1

                                                Ozone  (ppm)
                                             0.12
                                                       0.14
  Source: Brown et al. (2008)
                   O C
                   2 Ol
                   it o
                     u.  2
»Adams(2006)

^Adams(2003)

XAdams(2002)

XAdams(1998)

DHorstman etal. (1990)

O Kim etal. (2011)

C McDonnell etal. (1991)

ASchelegleetal. (2009)
                                                                            (t)
                                                               A (t)   (m)$AA (m)
                                                                          A (t)
                                                                          A (t,m)
                                                   (m)xA (t)
                                    X(m)
                         0.03
                                   0.04
                                            0.05
                                                     0.06
                                                               0.07
                                                                        0.08
                                                                                  0.09
               B
                                                 Ozone (ppm)
  Studies appearing in the figure legends are: Adams (2006a, 2003a. 2002. 1998). Folinsbee et al. (1988). Horstman et al. (1990).
Kim et al. (2011), McDonnell et al. (2007: 1991), and Schelegle et al. (2009).
  Top, panel A: all studies exposed subjects to a constant (square-wave) concentration in a chamber, except Adams (1998) where a
facemask was used. The McDonnell et al. (2007) curve illustrates the predicted FENA, decrement at 6.6 hours as a function of ozone
concentration for a 23-year old (the average age of subjects that participated in the illustrated studies). Note that this curve was not
"fitted" to the plotted data. Error bars (where available) are the standard error of responses. Bottom, panel B: all studies used
constant (square-wave) exposures in a chamber unless designated as triangular (t) and/or facemask (m) exposures.


Figure 6-1      Cross-study comparison of mean ozone-induced FEVi decrements
                  following 6.6 hours of exposure to  ozone.  During each hour of the
                  exposures,  subjects were engaged in  moderate quasi continuous
                  exercise (40 L/min) for 50 minutes and rest for 10 minutes.  Following the
                  third hour, subjects had an additional  35-minute rest period for lunch. The
                  data at 0.06, 0.08 and 0.12 ppm have been offset for illustrative purposes.
Draft - Do Not Cite or Quote
                     6-7
                                                                    September 2011

-------
 1                   For prolonged (6.6 hours) exposures relative to shorter exposures, significant pulmonary
 2                   function responses and symptoms have been observed at lower O3 concentrations and at a
 3                   moderate level of exercise (VE = 40 L/min). The results from studies using 6.6 hours of
 4                   constant or square-wave (S-W) exposures to between 40 and 120 ppb are illustrated in
 5                   Figure 6-l(A). Figure 6-l(B) focuses on the range from 40 to 80 ppb and includes
 6                   triangular exposure protocols as well as facemask exposures. Exposure to 40 ppb for 6.6
 7                   hours produces small, statistically insignificant changes in FEVi that are relatively
 8                   similar to responses from FA exposure (Adams. 2002). Volunteers exposed to 60 ppb O3
 9                   experience group mean O3-induced FEVi decrements of about 3% (Kimetal.. 2011;
10                   Brown et al., 2008) (Adams. 2006a)'; those exposed to 80 ppb have group mean
11                   decrements which range from 6 to 8% (Adams. 2006a. 2003a: McDonnell et al.. 1991;
12                   Horstman et al., 1990); at 100 ppb, group mean decrements range from 8 to  14%
13                   (McDonnell et al.. 1991; Horstman et al.. 1990): and at 120 ppb, group mean decrements
14                   of 13 to  16% are observed (Adams. 2002; Horstman et al.. 1990; Folinsbee et al.. 1988).
15                   As illustrated in Figure 6-1, there is a smooth dose-response curve without evidence of a
16                   threshold for exposures between  40 and 120 ppb O3. Taken together, these data indicate
17                   that mean FEVi is clearly decreased by 6.6-h exposures to 60 ppb O3 and higher
18                   concentrations in subjects performing moderate exercise.

19                   As opposed to constant or S-W concentration patterns used in the studies described
20                   above, many studies conducted at the levels of 40-80 ppb have used variable O3
21                   concentration patterns. It has been suggested that a triangular (variable concentration)
22                   exposure profile can potentially lead to higher FEVi responses than S-W profiles despite
23                   having at the same average O3  concentration over the exposure period. Hazucha et al.
24                   (1992) were the first to investigate the effects of variable versus constant concentration
25                   exposures on responsiveness to O3. In their study, volunteers were randomly exposed to a
26                   triangular concentration profile (averaging 120 ppb over the 8-h exposure) that increased
27                   linearly from 0-240 ppb for the first 4 hours of the 8-h exposure, then decreased linearly
28                   from 240 to 0 ppb over the next 4 hours of the 8-h exposure, and to an S-W exposure of
29                   120 ppb O3 for 8 hours. While  the total inhaled O3 doses at 4 hours and 8 hours for the S-
30                   W and the triangular concentration profile were almost identical, the FEVi response was
31                   dissimilar. For the S-W exposure, FEVi declined ~5% by the fifth hour and then
32                   remained at that level. With the triangular O3 profile, there was minimal FEVi response
33                   over the first 3 hours followed by a rapid decrease in FEVi (-10.3%) over the next 3
         1 Adams (2006a) did not find effects on FE\A at 60 ppb to be statistically significant. In an analysis of the Adams (2006a) data,
         even after removal of potential outliers, Brown et al. (2008) found the average effect on FENA, at 60 ppb to be small, but highly
                            statistically significant (p < 0.002) using several common statistical tests.
      Draft - Do Not Cite or Quote                        6-8                                 September 2011

-------
 1                  hours. During the seventh and eighth hours, mean FEVi decrements improved to -6.3%
 2                  as the O3 concentration decreased from 120 to 0 ppb (mean = 60 ppb). These findings
 3                  illustrate that the severity of symptoms and the magnitude of spirometric responses are
 4                  time-dependent functions of O3 delivery rate with periods of both effect development and
 5                  recovery during the course of an exposure.

 6                  Subsequently, others have also demonstrated that variable concentration exposures can
 7                  elicit greater FEVi and symptomatic responses than do S-W exposures (Adams. 2006a. b,
 8                  2003a). Adams (2006b) reproduced the findings of Hazucha et al. (1992) at 120 ppb.
 9                  However, Adams (2006a, 2003a) found that responses from an 80 ppb O3 (average)
10                  triangular exposure did not differ significantly from those observed in the 80 ppb O3 S-W
11                  exposure at 6.6 hours. Nevertheless, FEVi and symptoms were significantly different
12                  from pre-exposure at 4.6 hours (when the O3 concentration was 150 ppb) in the triangular
13                  exposure, but not until 6.6 hours in the S-W exposure. At the lower O3 concentration of
14                  60 ppb, no temporal pattern differences in FEVi responses between  S-W and triangular
15                  exposure profiles could be discerned (Adams. 2006a). However, total symptom scores
16                  were significantly increased for the 60 ppb triangular (but not the S-W) exposure
17                  following 5.6 and 6.6 hours of exposure. At 80 ppb, respiratory symptoms tended to
18                  increase more rapidly during the triangular than S-W exposure protocol, but then
19                  decreased during the last hour of exposure to be less for the triangular than the S-W
20                  exposure at 6.6 h. Both total symptom scores and pain on deep inspiration were
21                  significantly increased following exposures to 80 ppb relative to all other exposure
22                  protocols, i.e., FA, 40, and 60 ppb exposures. Following the 6.6-hour exposures,
23                  respiratory symptoms at 80 ppb were rougly 2-3 times greater than observed at 60 ppb.
24                  At 40 ppb, triangular and S-W patterns produced spirometric and subjective  symptom
25                  responses similar to FA exposure (Adams. 2006a. 2002).

26                  For exposures of 60 ppb and greater, these studies (Adams. 2006a. b, 2003a; Hazucha et
27                  al.. 1992) demonstrate that during triangular exposure protocols, volunteers exposed
28                  during moderate exercise (VE = 40 L/min) may develop greater spirometric and/or
29                  symptomatic responses during and following peak O3 concentrations as compared to
30                  responses over the same time  interval of S-W exposures. This observation is not
31                  unexpected since the inhaled dose rate during peaks of the triangular protocols
32                  approached twice that of the S-W protocols, e.g., 150 ppb versus 80 ppb peak
33                  concentration. At time intervals toward the end of an exposure, O3 delivery rates for the
34                  triangular protocols were less than those of S-W. At these later time intervals, there is
35                  some recovery of responses during triangular exposure protocols, whereas there is a
36                  continued development of or a plateau of responses in the S-W exposure protocols. Thus,
37                  responses during triangular protocols relative to S-W protocols may be expected to
38                  diverge and be greater following peak exposures and then converge toward the end of an
      Draft - Do Not Cite or Quote                      6-9                                September 2011

-------
 1                   exposure. The ensuing discussion on exposures between 40 and 80 ppb will focus on
 2                   postexposure effects where the influence of triangular and S-W concentration patterns are
 3                   minimal, i.e., FEVi pre-to-post effects are similar (although not identical) between
 4                   triangular and S-W protocols having equivalent average exposure concentrations.

 5                   Schelegle et al. (2009) recently investigated the effects of 6.6 hours variable O3 exposure
 6                   protocols at mean concentrations of 60, 70, 80, and 87 ppb on respiratory symptoms and
 7                   pulmonary function in young healthy adults (16 F, 15 M; 21.4 ± 0.6  years) exposed
 8                   during moderate quasi continuous exercise (VE = 40 L/min). The mean FEVi (±standard
 9                   error) decrements at 6.6 hours (end of exposure relative to pre-exposure) were -0.80 ±
10                   0.90%, 2.72 ± 1.48%, 5.34 ± 1.42%, 7.02 ± 1.60%, and 11.42 ± 2.20% for exposure to
11                   FA, 60, 70, 80, and 87 ppb O3, respectively. Statistically significant  decrements in FEVi
12                   and increases in total subjective symptom scores  (p < 0.05) were found following
13                   exposure to mean concentrations of 70, 80, and 87 ppb O3 relative to FA. Statistically
14                   significant effects were not found at 60 ppb. One of the expressed purposes of the
15                   Schelegle et al. (2009) study was to determine the minimal mean O3 concentration that
16                   produces a statistically significant decrement in FEVi and symptoms in healthy
17                   individuals completing 6.6-h exposure protocols. At 70 ppb, Schelegle et al. (2009)
18                   observed a statistically significant O3-induced of 6.1%. At 60 ppb, an O3-induced 3.5%
19                   FEVi decrement was not found to be statistically significant. However, this effect is
20                   similar in magnitude to the 2.9% FEVi decrement at 60 ppb observed by Adams (2006a)
21                   that was found to be statistically significant by Brown et al. (2008).

22                   More recently, Kim et al. (2011) investigated the effects of a 6.6-h exposure to 60 ppb O3
23                   during moderate quasi continuous exercise (VE = 40 L/min) on pulmonary function and
24                   respiratory symptoms in young healthy adults (32 F, 27 M; 25.0 ± 0.5  year) that were
25                   roughly half GSTM1-null and half GSTM1-positive. Sputum neutrophil levels were also
26                   measured in a subset of the subjects (13 F, 11 M). The mean FEVi (±standard error)
27                   decrements at 6.6 hours (end of exposure relative to pre-exposure) were significantly
28                   different (p = 0.008) between the FA (0.002 ± 0.46%) and O3 (1.76 ± 0.50%) exposures.
29                   The inflammatory response following O3 exposure was also significantly (p<0.001)
30                   increased relative to the FA exposure. Respiratory symptoms were not affected by O3
31                   exposure. There was also no significant effect of GSTM1 genotype on FEVi or
32                   inflammatory responses.

33                   Consideration of the minimal O3 concentration producing statistically significant effects
34                   on FEVi following 6.6-h exposures warrants additional discussion. As discussed above,
35                   numerous studies have demonstrated statistically significant O3-induced group mean
36                   FEVi decrements of 6-8% at 80 ppb. Schelegle et al. (2009) have now reported
37                   statistically significant O3-induced group mean FEVi decrement of 6%, as well as
      Draft - Do Not Cite or Quote                       6-10                                September 2011

-------
 1                   respiratory symptoms, at 70 ppb. At 60 ppb, there is information available from 4
 2                   separate studies (Adams. 1998)1 (Kim etal.. 2011: Schelegle et al.. 2009: Adams. 2006a).
 3                   The group mean O3-induced FEVi decrements observed in these studies were 3.6% by
 4                   Adams (1998)2, 2.8% (triangular exposure) and 2.9% (S-W exposure) by Adams (2006a).
 5                   3.5% by Schelegle et al. (2009). and 1.8% by Kim et al. (2011). Based on data from these
 6                   four studies, at 60 ppb, the weighted-average group mean O3-induced FEVi decrement
 7                   (i.e., adjusted for FA responses) is 2.7% (n=150) (Kim etal.. 2011: Schelegle et al.. 2009:
 8                   Adams. 2006a. 1998). Although not consistently statistically significant, these group
 9                   mean changes in FEVi at 60 ppb are consistent between studies, i.e., none observed an
10                   average improvement in lung function following a 6.6-h exposure to 60 ppb O3. Indeed,
11                   as was illustrated in Figure 6-1, the FEVi responses at 60  ppb fall on a smooth dose-
12                   response curve for exposures between 40 and 120 ppb O3. Furthermore, in a re-analysis
13                   of the 60 ppb S-W data from Adams (2006a). Brown et al. (2008) found the mean effects
14                   on FEVi to be highly statistically significant (p<0.002) using several common statistical
15                   tests even after removal of 3 potential outliers. The time-course and magnitude of FEVi
16                   responses at 40 ppb  resemble those occurring during FA exposures (Adams. 2006a.
17                   2002). Taken together, the available evidence shows that detectable effects of O3 on
18                   group mean FEVi persist down to 60 ppb, but not 40 ppb  in young healthy adults
19                   exposed for 6.6 hours during moderate exercise.

20                   In addition to overt effects of O3 exposure on the large airways indicated by spirometric
21                   responses, O3 exposure also affects the function of the small airways and parenchymal
22                   lung. Foster et al.  (1997: 1993) examined the  effect of O3  on ventilation distribution. In
23                   healthy adult males  (n=6; 26.7 ± 7 years old) exposed to O3 (330 ppb with light
24                   intermittent exercise for 2 h), there was a significant reduction in ventilation to the lower
25                   lung (31% of lung volume) and significant increases in ventilation to the upper- and
26                   middle-lung regions (Foster et al.. 1993). In a subsequent study of healthy males (n=15;
27                   25.4 ± 2 years old) exposed to O3 (350 ppb with moderate intermittent exercise for 2.2 h),
28                   O3 exposure caused  a delayed gas washout (Foster etal.. 1997). The pronounced slow
29                   phase of gas washout following O3 exposure represented a 24% decrease in the washout
30                   rate. A day following O3 exposure, 50% of the subjects  still had (or developed) a delayed
31                   washout relative to the pre- O3 maneuver. These studies suggest a prolonged O3 effect on
32                   the small airways  and ventilation distribution in healthy young individuals.
        1 The American Petroleum Institute has declined to provide a copy of this report to EPA.
        2 This information is from page 133 of Adams (2006a). This decrement may be increased due to a target VE of 23 L/min/m2 BSA
      relative to other studies with which it is listed having the target VE of 20 L/min/m2 BSA. It should also be noted that subjects were
      exposed via a facemask in this study. However, Adams (2003a, b, 2002) found very similar FE\A responses between facemask and
      chamber exposures.
      Draft - Do Not Cite or Quote                        6-11                                 September 2011

-------
 1                  There is a rapid recovery of O3-induced spirometric responses and symptoms; 40 to 65%
 2                  recovery appears to occur within about 2 hours following exposure (Tolinsbee and
 3                  Hazucha. 1989). For example, following a 2-h exposure to 400 ppb O3 with intermittent
 4                  exercise, Nightingale et al. (2000) observed a 13.5% mean decrement in FEVi. By 3
 5                  hours postexposure, however, only a 2.7% FEVi decrement persisted. Partial recovery
 6                  also occurs following cessation of exercise despite continued exposure to O3 (Tolinsbee
 7                  etal.. 1977) and at low O3 concentrations during exposure (Hazucha etal.. 1992). A
 8                  slower recovery phase, especially after exposure to higher O3 concentrations, may take at
 9                  least 24 hours to complete (Folinsbee and Hazucha, 2000; Folinsbee et al., 1993).
10                  Repeated daily exposure studies at higher concentrations typically show that FEVi
11                  response to O3 is enhanced on the second day of exposure. This enhanced response
12                  suggests a residual effect of the previous exposure, about 22 hours earlier, even though
13                  the pre-exposure spirometry may be the same as on the previous day. The absence of the
14                  enhanced response with repeated exposure at lower O3 concentrations may be the result
15                  of a more complete recovery or less damage to pulmonary tissues (Folinsbee etal..  1994).

                        Intersubject Variability in Response of Healthy Subjects
16                  Consideration of group mean changes is important in discerning if observed effects are
17                  due to O3 exposure rather than chance alone. Inter-individual variability in responses is,
18                  however, considerable and pertinent to assessing the fraction of the population that might
19                  actually be affected during an O3 exposure. Hackney et al. (1975) first recognized a wide
20                  range in the sensitivity of subjects to O3. The range in the subjects' ages (29 to 49 years)
21                  and smoking status (0 to 50 pack years) in the Hackney et al. (1975) study are now
22                  understood to affect the spirometric and symptomatic responses to O3. Subsequently,
23                  DeLucia and Adams (1977) examined responses to O3 in  six healthy non-smokers and
24                  found that two exhibited notably greater sensitivity to O3. Since that time, numerous
25                  studies have documented considerable variability in responsiveness to O3 even in subjects
26                  recruited to assure homogeneity in factors recognized or presumed to affect responses.

27                  An individual's FEVi response to a 2-h O3 exposure is generally reproducible over
28                  several months and presumably reflects the intrinsic responsiveness of the individual to
29                  O3  (Hazucha et al., 2003; McDonnell et al.,  1985a). The frequency distribution of
30                  individual FEVi responses following these relatively short exposures becomes skewed  as
31                  the group mean response increases, with some individuals experiencing large reductions
32                  in FEVi (Weinmann et al.. 1995c: Kulle et al.. 1985). For 2-h exposures with intermittent
33                  exercise causing a predicted average FEVi decrement of 10%, individual decrements
34                  ranged from approximately 0 to 40% in white males aged 18-36 years (McDonnell et al..
35                  1997). For an average FEVi decrement of 13%, Ultman et al. (2004) reported FEVi
36                  responses ranging from a 4% improvement to a 56% decrement in young healthy adults
      Draft - Do Not Cite or Quote                      6-12                                September 2011

-------
 1
 2
(32 M, 28 F) exposed for 1 hour to 250 ppb O3. One-third of the subjects had FEVi
decrements of >15%, and 7% of the subjects had decrements of >40%.
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
M 35 |
O 3Q .
5" OT
3
CO 20-
0 15-
M 10-
o
Oi 5 '
Q- n.







n
, — .


. 	 .











Oppb
0%


i — i




-


•
.







nl~l






. 	 .





60 ppb
16%



^ n













n
__



. 	 .





70 ppb
i—i




19%



Hn n













nn






80 ppb
I — i




29%



In] In
                                             10   20   30    -10   0
                                                FEV, Decrement (%)
                                                                                 • 10   0   10   20   30
       Source: Adapted with permission of American Thoracic Society (Schelegle et al., 2009)
       During each hour of the exposures, subjects were engaged in moderate quasi continuous exercise (40 L/min) for 50 minutes and
      rest for 10 minutes. Following the third hour, subjects had an additional 35 minute rest period for lunch. Subjects were exposed to a
      triangular ozone concentration profile having the average ozone concentration provided in each panel. As average ozone
      concentration increased, the distribution of responses became asymmetric with a few individuals exhibiting large FEVi decrements.
      The percentage indicated in each panel  is the portion of subjects having a FEVi decrement in excess of 10%.

      Figure 6-2     Frequency distributions  of FEVi decrements observed  by Schelegle
                       et al. (2009) in young healthy adults (16 F, 15 M) following  6.6-h
                       exposures to ozone or filtered air.
Consistent with the 1- to 2-h studies, the distribution of individual responses following
6.6-h exposure studies becomes skewed with increasing exposure concentration and
magnitude of the group mean FEVi response (McDonnell. 1996). Figure 6-2 illustrates
frequency distributions of individual FEVi responses observed in 31 young healthy adults
following 6.6-h exposures between 0 and 80 ppb. Schelegle et al. (2009) found >10%
FEVi decrements in 16, 19, 29, and 42% of individuals exposed for 6.6 hours to 60, 70,
80, and 87 ppb, respectively. Just as there  are differences in mean decrements between
studies having similar exposure scenarios  (Figure 6-1 at 80 and 120 ppb), there are also
differences in the proportion of individuals affected with >10% FEVi decrements. At
80 ppb, the proportion affected with >10% FEVi decrements was 17% (n=30) by Adams
(2006a)'. 26% (n=60) by  McDonnell (1996). and 29%(n=31) by Schelegle et al. (2009).
At 60 ppb, the proportion with >10% FEVi decrements was 20% (n=30) by Adams
(1998)2. 3% (n=30) by Adams (2006a)5, 16% (n=31) by Schelegle et al. (2009). and 5%
(n=59) by Kim et al. (2011). Based on these studies, the weighted average proportion of
       1 Not assessed by Adams (2006a), the proportion was provided in Figure 8-1B of U.S. EPA (2006b).
       2 This information is from page 761 of Adams (2002).
      Draft - Do Not Cite or Quote
                                6-13
September 2011

-------
 1                   individuals with >10% FEVi decrements is 10% following exposure to 60 ppb. Due to
 2                   limited data within the published papers, these proportions were not corrected for
 3                   responses to FA exposure where lung function typically improves in healthy adults. For
 4                   example, uncorrected versus O3-induced (i.e., adjusted for response during FA exposure)
 5                   proportions of individuals having >10% FEVi decrements in the Adams (2006a)1 study
 6                   were, respectively, 3% versus 7% at 60 ppb and 17% versus 23% at 80 ppb. Thus,
 7                   uncorrected proportions underestimate the actual fraction of healthy individuals affected.

 8                   Given considerable inter-individual variability in responses, the interpretation of
 9                   biologically small group mean decrements requires careful consideration. Following
10                   prolonged 6.6-h exposures to an average level of 60 ppb O3, data available from four
11                   studies yield a weighted-average group mean O3-induced FEVi decrement (i.e., adjusted
12                   for FA responses) of 2.7% (n=150) (Kim etal.. 2011;  Schelegle et al.. 2009; Adams.
13                   2006a. 1998). The data from these studies also yield a weighted-average proportion
14                   (uncorrected for FA responses) of subjects with >10% FEVi decrements of 10% (n=150)
15                   (Kim etal.. 2011: Schelegle et al.. 2009: Adams. 2006a. 1998). In an individual with
16                   relatively "normal" lung function, recognizing technical and biological variability in
17                   measurements, confidence can be given that within-day changes in FEVi of > 5% are
18                   clinically meaningful (Pellegrino et al.. 2005: ATS. 1991). Here focus is given to
19                   individuals with >10% decrements in FEVi since some individuals in the Schelegle et al.
20                   (2009) study experienced 5-10% FEVi decrements following exposure to FA. A  10%
21                   FEVi decrement is also generally accepted as an abnormal response and as reasonable
22                   criterion for assessing exercise-induced bronchoconstriction (Dryden et al.. 2010: ATS.
23                   2000a). The data are not available in the published papers to determine the O3-induced
24                   proportion for either the Adams (1998) or Schelegle et al. (2009) studies. As already
25                   stated, however, this uncorrected proportion likely underestimates that actual proportion
26                   of healthy individuals experiencing O3-induced FEVi decrements in excess of 10%.
27                   Therefore, by considering uncorrected responses and those individuals having >10%
28                   decrements, 10% is an underestimate of the proportion of healthy individuals that are
29                   likely to experience clinically meaningful changes in lung function following exposure
30                   for 6.6 hours to 60 ppb O3 during moderate exercise. Of the studies conducted at 60 ppb,
31                   only Kim et al. (2011) reported FEVi decrements at 60 ppb to be statistically significant.
32                   Although,  Brown et al. (2008) found those from Adams (2006a) to be highly statistically
33                   significant. Though group mean decrements are biologically small and generally do not
34                   attain statistical significance, a considerable fraction of exposed individuals experience
35                   clinically meaningful decrements in lung function.
        1 Not assessed by Adams (2006a), uncorrected and OS-induced proportions are from Figures 8-1B and 8-2, respectively, of the
      2006O3AQCD(2006b).
      Draft - Do Not Cite or Quote                       6-14                                 September 2011

-------
                     Responses in Individuals with Pre-Existing Disease

 1                   Individuals with respiratory disease are of primary concern in evaluating the health
 2                   effects of O3 because a given change in function is likely to have more impact on a
 3                   person with preexisting function impairment and reduced reserve.

 4                   Possibly due to the age of subjects studied, patients with COPD performing light to
 5                   moderate exercise do not generally experience statistically significant pulmonary
 6                   function decrements following 1- and 2-h exposures to < 300 ppb O3 (Kehrl et al.. 1985;
 7                   Linn et al.. 1983; Linn et al.. 1982b: Solic et al.. 1982). Following a 4-hour exposure to
 8                   240 ppb O3 during exercise, Gong et al. (1997b) found an O3-induced FEVi decrement of
 9                   8% in COPD patients which was not statistically different from the decrement of 3% in
10                   healthy subjects. Demonstrating the need for control exposures and presumably due to
11                   exercise, four of the patients in the Gong et al. (1997b) study had FEVi decrements of
12                   >14% following both the FA and O3 exposures. Although the clinical significance is
13                   uncertain, small transient decreases in arterial blood oxygen saturation have also been
14                   observed in some of these studies.

15                   Based on studies reviewed in the 1996 and 2006 O3 AQCDs, asthmatic subjects appear to
16                   be at least as sensitive to acute effects of O3 as healthy nonasthmatic subjects. Horstman
17                   et al. (1995) found the O3-induced FEVi decrement in mild-to-moderate asthmatics to be
18                   significantly larger than in healthy subjects (19% versus  10%, respectively) exposed to
19                   160 ppb O3 during exercise for 7.6-h exposure. In asthmatics, a significant positive
20                   correlation between O3-induced spirometric responses and baseline lung function was
21                   observed, i.e., responses increased with severity of disease. Such differences in
22                   pulmonary function between asthmatics and healthy individuals were not found in shorter
23                   duration studies. Alexis et al. (2000) and Torres et al. (1996) reported a tendency for
24                   slightly greater FEVi decrements in asthmatics than healthy subjects. Several studies
25                   reported similar responses between asthmatics and healthy individuals (Scannell et al..
26                   1996; Hiltermann et al.. 1995; Bashaetal..  1994). The lack of differences in the
27                   Hiltermann et al. (1995) and Basha et al. (1994) studies was not surprising, however,
28                   given extremely small sample sizes and corresponding lack of statistical power. One
29                   study reported a tendency for asthmatics to have smaller O3-induced FEVi decrements
30                   than healthy subjects (3% versus 8%, respectively) when exposed to 200 ppb O3 for 2
31                   hours during exercise (Mudway et al.. 2001). However, the asthmatics in that study also
32                   tended to be older than the healthy subjects, which could partially explain their lesser
33                   response since FEVi responses to O3 diminish with age.

34                   Some, but not all, studies have also reported that asthmatics have a somewhat
35                   exaggerated airway inflammatory response to acute O3 exposure relative to healthy
36                   control  subjects (Holz et al.. 2002; Peden. 2001; Newson et al.. 2000; Hiltermann et al..
      Draft - Do Not Cite or Quote                       6-15                                September 2011

-------
 1                   1999: Michelson et al.. 1999; Vagaggini et al.. 1999; Hiltermann et al.. 1997; Peden et
 2                   al.. 1997: Scannell et al.. 1996: Peden et al.. 1995: Bashaetal.. 1994: McBride et al..
 3                   1994). For example, at 18 hours post-O3 exposure (200 ppb, 4 hours with exercise) and
 4                   corrected for FA responses, Scannell et al. (1996) found significantly increased
 5                   neutrophils in 18 asthmatics (12%) compared to 20 healthy subjects (4.5%). This
 6                   difference in inflammatory response was observed despite no group differences in
 7                   spirometric responses to O3.

 8                   Vagaggini et al. (2010) exposed mild-to-moderate asthmatics (n=23; 33 ± 11 years) to
 9                   300 ppb O3 for 2 hours with moderate exercise. Although the group mean O3-induced
10                   FEVi decrement was only 4%, eight subjects were categorized as "responders" with
11                   >10% FEVi decrements. There were no baseline differences between responders and
12                   nonresponders. At 6 hours post O3 exposure, sputum neutrophils were significantly
13                   increased by 15% relative to FA in responders. The neutrophil increase in responders was
14                   also significantly greater than the 0.2% increase in nonresponders. Across all subjects,
15                   there was a significant (r=0.61, p = 0.015) correlation between changes in FEVi and
16                   changes in sputum neutrophils. Prior studies have reported that inflammatory responses
17                   do not appear to be correlated with lung function responses in either asthmatic or healthy
18                   subjects (Holzetal.. 1999: Balmes et al.. 1997: Balmes et al.. 1996: Devlin et al.. 1991).
19                   Interestingly, the nonresponders in the Vagaggini et al. (2010) study experienced a
20                   significant O3-induced 11.3% increase in sputum eosinophils, while responders had an
21                   nonsignificant 2.6% decrease. Six of the subjects were NQO1 wild type and GSTM1 null,
22                   but this genotype was not found to be associated with the  changes in lung function or
23                   inflammatory responses to O3.

24                   A few recent studies have evaluated the effects of corticosteroid usage on the response of
25                   asthmatics to O3. Vagaggini et al. (2007) evaluated whether corticosteroid usage would
26                   prevent O3-induced lung function decrements and inflammatory responses in a group of
27                   subjects with mild persistent asthma (n=9; 25 ± 7 years). In this study, asthmatics were
28                   randomly exposed on four occasions to 270 ppb O3 or FA for 2 hours with moderate
29                   exercise. Exposures were preceded by four days of treatment with prednisone or placebo.
30                   Pretreatment with corticosteroids prevented an inflammatory response in  induced sputum
31                   at 6 hours postexposure. FEVi responses were, however, not prevented by corticosteroid
32                   treatment and were roughly equivalent to those observed following placebo. Vagaggini et
33                   al. (2001) also found budesonide to decrease airway neutrophil influx in asthmatics
34                   following O3 exposure. In contrast, inhalation of corticosteroid budesonide failed to
35                   prevent or attenuate O3-induced responses in healthy subjects as assessed by
36                   measurements of lung function, bronchial reactivity and airway inflammation
37                   (Nightingale et al.. 2000). High doses of inhaled fluticasone and oral prednisolone have
      Draft - Do Not Cite or Quote                       6-16                                 September 2011

-------
 1                   each been reported to reduce inflammatory responses to O3 in healthy individuals (Holz
 2                   etal.. 2005).

 3                   More recently, Stenfors et al. (2010) exposed persistent asthmatics (n=13; aged 33 years)
 4                   receiving chronic inhaled corticosteroid therapy to 200 ppb O3 for 2 hours with moderate
 5                   exercise. An average O3-induced FEVi decrement of 8.4% was observed, whereas, only a
 6                   3.0% FEVi decrement is predicted for similarly exposed age-matched healthy controls
 7                   (McDonnell et al.. 2007). At 18 hours postexposure, there was a significant O3-induced
 8                   increase in bronchioalveolar lavage (BAL) neutrophils, but not eosinophils. Bronchial
 9                   biopsy also showed a significant O3-induced increase in mast cells. This study suggests
10                   that the protective effect of acute corticosteroid therapy against inflammatory responses
11                   to O3 in asthmatics demonstrated by Vagaggini et al. (2007) may be lost with continued
12                   treatment regime s.


                     Factors Modifying Responsiveness to Ozone

13                   Physical activity increases VE and therefore the dose of inhaled O3. Consequently, the
14                   intensity of physiological response during and following an acute O3 exposure will be
15                   strongly associated with minute ventilation. Apart from inhaled O3 dose and related
16                   environmental factors (e.g., repeated daily exposures), individual-level  factors, such as
17                   health status, age, gender, ethnicity, race, smoking habit, diet, and socioeconomic status
18                   (SES) have been considered as potential modulators of a physiologic response to such
19                   exposures.

20                   Children, adolescents, and young adults (<18 years of age) appear, on average, to have
21                   nearly equivalent spirometric responses to O3, but have greater responses than middle-
22                   aged and older adults when exposed to comparable O3 doses (U.S. EPA, 1996a).
23                   Symptomatic responses to O3 exposure, however, appear to increase with age until early
24                   adulthood and then gradually decrease with increasing age (U.S. EPA. 1996a). For
25                   example, healthy children (aged 8-11 y) exposed to  120 ppb O3 (2.5 h; heavy intermittent
26                   exercise) experienced similar spirometric responses but lesser symptoms than similarly
27                   exposed young healthy adults (McDonnell et al.. 1985b). For subjects aged 18-36 years,
28                   McDonnell et al. (1999) reported that symptom responses from O3 exposure also
29                   decrease with increasing age. Diminished symptomatic responses in children and the
30                   elderly  might put these groups at increased risk for continued O3 exposure, i.e., a lack of
31                   symptoms may result in their not avoiding or ceasing exposure. Once lung growth and
32                   development reaches the peak (18-20 years of age in females and early  twenties in
33                   males), pulmonary function, which is at its maximum as well, begins to decline
34                   progressively with age as does O3 sensitivity.
      Draft - Do Not Cite or Quote                       6-17                                September 2011

-------
 1                   In healthy individuals, the fastest rate of decline in O3 responsiveness appears between
 2                   the ages of 18 and 35 years (Passannante et al.. 1998; Seal et al.. 1996). more so for
 3                   females then males (Hazucha et al., 2003). During the middle age period (35-55 years),
 4                   O3 sensitivity continues to decline but at a much lower rate. Beyond this age (>55 years),
 5                   acute O3 exposure elicits minimal spirometric changes. Whether the same age-dependent
 6                   pattern of O3 sensitivity decline also holds for nonspirometric pulmonary function,
 7                   airway reactivity or inflammatory endpoints has not been determined. Although there is
 8                   considerable evidence that spirometric and symptomatic responses to O3 exposure
 9                   decrease with age beyond young adulthood, this evidence comes from cross-sectional
10                   analyses and has not been confirmed by longitudinal studies of the same individuals.

11                   Several studies have suggested that physiological  differences between sexes may
12                   predispose females to a greater susceptibility to O3. In females, lower plasma and nasal
13                   lavage fluid (NLF) levels of uric acid (the most prevalent antioxidant), the initial defense
14                   mechanism of O3 neutralization in airway surface  liquid, may be a contributing factor
15                   (Housley et al.. 1996). Consequently, reduced absorption of O3 in the upper airways may
16                   promote its deeper penetration. Dosimetric measurements have shown that the absorption
17                   distribution of O3 is independent of gender when absorption is normalized to anatomical
18                   dead space (Bushetal.. 1996). Thus, a gender-related differential removal of O3 by uric
19                   acid seems to be minimal. In general, the physiologic response of young healthy females
20                   to O3 exposure appears comparable to the response of young males (Hazucha et al..
21                   2003). Several studies have investigated the effects of the menstrual cycle on responses to
22                   O3 in healthy young women. In a study of 9 women exposed during exercise to 300 ppb
23                   O3 for an hour, Fox et al.  (1993) found lung function responses to O3 significantly
24                   enhanced during the follicular phase relative to the luteal phase. However, Weinmann et
25                   al. (1995a) found no difference in responses between the follicular and luteal phases as
26                   well as no significant differences between 12 males and  12 females exposed during
27                   exercise to 350 ppb O3 for 2.15 h. Seal et al. (1996) also  reported no effect of menstrual
28                   cycle phase in their analysis of responses of 150 women  (n=25 per exposure group; 0,
29                   120, 240, 300, and 400 ppb O3). Seal et al. (1996) conceded that the methods used by Fox
30                   et al. (1993) more  precisely defined menstrual  cycle phase.

31                   Only two controlled human exposure studies have assessed differences in lung function
32                   responses between races.  Seal et al. (1993) compared lung function responses of whites
33                   (93 M, 94 F) and blacks (undefined ancestry; 92 M, 93 F) exposed to a range of O3
34                   concentrations (0-400 ppb). The main effects of gender-race group and O3 concentration
35                   were statistically significant (both at p < 0.001), although the interaction between gender-
36                   race group and O3 concentration was not significant (p = 0.13). These findings indicate
37                   some overall difference between the gender-race groups that is independent of O3
38                   concentration, i.e., the concentration-response curves for the four gender-race groups are
      Draft - Do Not Cite or Quote                       6-18                                 September 2011

-------
 1                   parallel. In a multiple comparison procedure on data collapsed across all O3
 2                   concentrations for each gender-race group, both black men and black women had
 3                   significantly larger decrements in FEVi than did white men. The authors noted that the
 4                   O3 dose per unit of lung tissue would be greater in blacks and females than whites and
 5                   males, respectively. That this difference in tissue dose might have affected responses to
 6                   O3 cannot be ruled out. The college students recruited for the Seal et al. (1993) study are
 7                   probably from better educated and SES advantaged families, thus reducing potential
 8                   influence of these variables on results. In a follow-up analysis, Seal et al. (1996) reported
 9                   that, of three SES categories, individuals in the middle SES category showed greater
10                   concentration-dependent decline in percent-predicted FEVi (4-5% at 400 ppb O3) than
11                   low and high SES groups. The authors did not have an "immediately clear" explanation
12                   for this finding.

13                   More recently, Que et al.  assessed pulmonary responses in blacks of African American
14                   ancestry (22 M, 24 F) and Caucasians (55 M, 28 F) exposed to 220 ppb O3 for 2.25 h
15                   (alternating 15 min periods of rest and brisk treadmill walking). On average, the black
16                   males experienced a 16.8% decrement in FEVi following O3 exposure which was
17                   significantly larger than mean FEVi  decrements of 6.2, 7.9, and 8.3% in black females
18                   and Caucasian males and Caucasian females, respectively.  In the study by Seal et al.
19                   (1993). there was potential that the increased FEVi decrements in blacks relative to
20                   whites were due to increased O3 tissue doses since exercise rates were normalized to
21                   BSA. Differences in O3 tissue doses  between the races should not have occurred in the
22                   Que et al. study, however, since exercise rates were normalized to lung volume (viz., 6-8
23                   times FVC). Thus, the increased mean FEVi decrement in black males is not likely
24                   attributable to systematically larger O3 tissue doses in blacks relative to whites.

25                   Smokers are less responsive to O3 than nonsmokers. Spirometric and plethysmographic
26                   pulmonary function decline, nonspecific airway hyperreactivity, and inflammatory
27                   response of smokers to O3 were  all weaker than data reported for nonsmokers. Although
28                   all of these responses are intrinsically related, the functional association between them, as
29                   in nonsmokers, has been weak. Similarly, the time course of development and recovery
30                   of these effects as well their reproducibility was not different from nonsmokers. Chronic
31                   airway inflammation with desensitization of bronchial nerve endings and an increased
32                   production of mucus may plausibly explain the reduced responses to O3 in smokers
33                   relative to nonsmokers (Frampton et al.. 1997b: Torres et al.. 1997).

34                   The first line of defense against  oxidative stress is antioxidants-rich ELF which
35                   scavenges free radicals and limit lipid peroxidation. Exposure to O3 depletes the
36                   antioxidant level in nasal ELF probably due to scrubbing of O3 (Mudway et al.. 1999a).
37                   however, the concentration and the activity of antioxidant enzymes either in ELF or
      Draft - Do Not Cite or Quote                       6-19                                 September 2011

-------
 1                   plasma do not appear to be related to O3 responsiveness (Samet et al., 2001; Avissar et
 2                   al.. 2000; Blomberg et al.. 1999). Carefully controlled studies of dietary antioxidant
 3                   supplementation have demonstrated some protective effects of a-tocopherol and
 4                   ascorbate on spirometric lung function from O3 but not on the intensity of subjective
 5                   symptoms and inflammatory response including cell recruitment, activation and a release
 6                   of mediators (Samet etal. 2001; Trengaetal.. 2001). Dietary antioxidants have also been
 7                   reported to attenuate  O3-induced bronchial hyperresponsiveness in asthmatics (Trenga et
 8                   al.. 2001).

 9                   A number of studies(e.g., Romieu et al.. 2004a: David et al.. 2003; Corradi et al.. 2002;
10                   Bergamaschi et al.. 2001) have reported that genetic polymorphisms of antioxidant
11                   enzymes may  modulate pulmonary function and inflammatory response to O3 challenge.
12                   It appears that healthy carriers of NQO1 wild type in combination with GSTM1 null
13                   genotype are more responsive to O3. Adults with GSTM1 null only genotype did not
14                   show O3 hyperresponsiveness. In contrast, asthmatic children with GSTM1 null genotype
15                   (Romieu et al.. 2004a) were reported to be more responsive to O3. However, in a
16                   controlled exposure of mild-to-moderate asthmatics (n=23; 33 ± 11 years) to 300 ppb O3
17                   for 2 hours with moderate exercise, Vagaggini et al. (2010) found that six of the subjects
18                   had a NQO \wt and GSTM1 mill, but this genotype was not associated with the changes
19                   in lung function or inflammatory responses to O3.

20                   Kim et al. (2011) also recently reported that GSTM1 genotype was not predictive of
21                   FEVi responses in young healthy adults (32 F, 27 M; 25.0 ± 0.5 year) that were roughly
22                   half GSTM1-null and half GSTM1-sufficient. Sputum neutrophil levels, measured in a
23                   subset of the subjects (13 F, 11 M), were also not significantly associated with GSTM1
24                   genotype.

25                   In a study of healthy  volunteers with GSTM1 sufficient (n=19; 24 ± 3) and GSTM1 null
26                   (n=16; 25 ± 5) genotypes exposed to 400 ppb O3 for 2 hours with exercise, Alexis et al.
27                   (2009) found that inflammatory responses but not lung function responses to O3 were
28                   dependent on genotype. At 4 hours post O3 exposure, both GSTM1 genotype groups had
29                   significant increases in sputum neutrophils with a tendency for a greater increase in
30                   GSTM1  sufficient than nulls. At 24 h postexposure, sputum neutrophils had returned to
31                   baseline  levels in the GSTM1 sufficient individuals. In the GSTM1 null subjects,
32                   however, sputum neutrophil levels increased from 4 h to 24 h and were significantly
33                   greater than both baseline levels and levels at 24 h in the GSTM1 sufficient  individuals.
34                   Since there was no FA control in the Alexis et al. (2009) study, effects of the exposure
3 5                   other than O3 itself cannot be ruled out. In general, the findings between studies are
36                   inconsistent. Additional studies that include control exposures are needed to clarify the
37                   influence of genetic polymorphisms on O3 responsiveness.
      Draft - Do Not Cite or Quote                      6-20                                September 2011

-------
 1                  In a retrospective analysis of data from 541 healthy, nonsmoking, white males between
 2                  the ages of 18-35 years from 15 studies conducted at the U.S. EPA Human Studies
 3                  Facility in Chapel Hill, North Carolina, McDonnell et al. (2010) found that increased
 4                  body mass index (BMI) was associated with enhanced FEVi responses. The BMI effect
 5                  was of the same  order of magnitude but in the opposite direction of the age effect where
 6                  by FEVi responses diminish with increasing age. In a similar retrospective analysis,
 7                  Bennett et al. (2007) found enhanced FEVi decrements following O3 exposure with
 8                  increasing BMI in a group of 75 healthy, nonsmoking, women (age 24 ± 4 years; BMI
 9                  range 15.7 to 33.4), but not 122 healthy, nonsmoking, men (age 25 ± 4 years; BMI range
10                  19.1 to 32.9). In the women, greater O3-induced FEVi decrements were seen in
11                  overweight (BMI >25) than in normal weight (BMI from 18.5 to 25), and in normal
12                  weight than in underweight (BMI <18.5) (P trend < 0.022). Together, these results
13                  indicate that higher BMI may be a risk factor for pulmonary effects associated with O3
14                  exposure.


                    Repeated Ozone Exposure Effects

15                  Based on studies reviewed in previous O3 AQCDs, several conclusions can be drawn
16                  about repeated Ito  2 h O3 exposures. Repeated exposures to O3 causes enhanced (i.e.,
17                  greater decrements) FVC and FEVi responses on the second day of exposure. The
18                  enhanced response appears to depend to some extent on the magnitude of the initial
19                  response (Horvath  et al.. 1981). Small responses to the first O3 exposure are less likely to
20                  result in an enhanced response on the second day of O3 exposure (Folinsbee et al.. 1994).
21                  With continued daily exposures (i.e., beyond the second day) there is a substantial (or
22                  even total) attenuation of pulmonary function responses, typically on the third to
23                  fifth days of repeated O3 exposure. This attenuation of responses is lost in 1  week (Kulle
24                  etal., 1982; Linn et al., 1982a) or perhaps 2 weeks (Horvath et al.. 1981) without O3
25                  exposure. In temporal conjunction with pulmonary function changes, symptoms induced
26                  by O3 (e.g., cough, pain on deep inspiration, and chest discomfort), are also increased on
27                  the second exposure day and attenuated with repeated O3 exposure thereafter (Folinsbee
28                  etal.. 1998; Foxcroft and Adams. 1986; Linnetal..  1982a: Folinsbee et al..  1980). In
29                  longer-duration (4-6.6 hours), lower-concentration studies that do not cause an enhanced
30                  second-day response, the attenuation of response to O3 appears to proceed more rapidly
31                  (Folinsbee et al.. 1994).

32                  Consistent with other investigators, Frank et al. (2001) found FVC and FEVi decrements
33                  to be significantly attenuated following four consecutive days of exposure to O3 (250
34                  ppb, 2 h). However, the effects of O3 on the small airways (assessed by a combined index
35                  of isovolumetric FEF25_75, Vmax50 and Vmax75) showed a persistent functional reduction
36                  from Day 2 through Day 4. Notably, in contrast to FVC and FEVi which exhibited a

      Draft - Do Not Cite or Quote                       6-21                               September 2011

-------
 1                   recovery of function between days, there was a persistent effect of O3 on small airways
 2                   function such that the baseline function on Day 2 through Day 4 was depressed relative to
 3                   Day 1. Frank et al. (2001) also found neutrophil (PMN) numbers in BAL remained
 4                   significantly higher following O3 (24 h after last O3 exposure) compared to FA.
 5                   Inflammatory markers from bronchioalveolar lavage fluid (BALF) following 4
 6                   consecutive days of both 2-h (Devlin et al.. 1997) and 4-h (Torres et al.. 2000; Christian et
 7                   al.. 1998) exposures have indicated ongoing cellular damage irrespective of the
 8                   attenuation of some cellular inflammatory responses of the airways, lung function and
 9                   symptoms response. These data suggest that the persistent small airways dysfunction
10                   assessed by Frank et al. (2001)  is likely induced by both neurogenic and inflammatory
11                   mediators, since the density of bronchial C-fibers is much lower in the small than large
12                   airways.


                     Summary of Controlled Human Exposure Studies on Lung Function

13                   Responses in humans exposed to ambient O3 concentrations include: decreased
14                   inspiratory capacity; mild bronchoconstriction; rapid, shallow breathing pattern during
15                   exercise; and symptoms of cough and pain on deep inspiration (U.S. EPA. 2006b. 1996a).
16                   Discussed in  subsequent Sections 6.2.2.1 and 6.2.3.1, exposure to O3 also results in
17                   airway hyperresponsiveness, pulmonary inflammation, immune system activation, and
18                   epithelial injury  (Que etal.; Mudway and Kelly. 2004a). Reflex inhibition of inspiration
19                   results in a decrease in forced vital capacity and, in combination with mild
20                   bronchoconstriction, contributes to a decrease in the FEVi. Healthy young adults exposed
21                   to O3 concentrations > 60 ppb develop statistically significant reversible, transient
22                   decrements in lung function if minute ventilation or duration of exposure is increased
23                   sufficiently. With repeated O3 exposures over several days, FEVi and symptom responses
24                   become attenuated in both healthy individuals and asthmatics, but this tolerance is lost
25                   after about a week without exposure (Gong etal.. 1997a: Folinsbee et al.. 1994; Kulle et
26                   al.. 1982). In  contrast to the attention of FEVi responses, there appear to be persistent O3
27                   effects on small  airways function as well as ongoing cellular damage during repeated
28                   exposures.

29                   There is a large degree of intersubject variability in lung function decrements
30                   (McDonnell.  1996). However, these lung function responses tend to be reproducible
31                   within a given individual over a period of several months indicating differences in the
32                   intrinsic responsiveness of individuals (Hazucha et al., 2003; McDonnell et al., 1985a). In
33                   healthy young adults, O3-induced decrements in FEVi do not appear to depend on gender
34                   (Hazucha et al.. 2003). body surface area or height (McDonnell  et al.. 1997). lung size or
35                   baseline FVC (Messineo and Adams.  1990). There  is limited evidence that blacks may
36                   experience greater O3-induced decrements in FEVi than age-matched whites (Que etal.;

      Draft - Do Not Cite or Quote                      6-22                                September 2011

-------
 1                   Seal et al., 1993). Healthy children experience similar spirometric responses but lesser
 2                   symptoms from O3 exposure relative to young adults (McDonnell et al.. 1985b). On
 3                   average, spirometric and symptom responses to O3 exposure appear to decline with
 4                   increasing age beyond about 18 years of age (McDonnell et al.. 1999; Seal et al.. 1996).
 5                   There is a tendency for slightly increased spirometric responses in mild asthmatics and
 6                   allergic rhinitics relative to healthy young adults (Torres etal.. 1996). Spirometric
 7                   responses in asthmatics appear to be affected by baseline lung function, i.e., responses
 8                   increase with disease severity (Horstman et al.. 1995).

 9                          Available information on recovery of lung function following O3 exposure
10                   indicates that an initial phase of recovery in healthy individuals proceeds relatively
11                   rapidly, with acute spirometric and symptom responses resolving within about 2 to 4 h
12                   (Folinsbee and Hazucha. 1989). Small residual lung function effects are almost
13                   completely resolved within 24 h. One day following O3 exposure, persisting effects on
14                   the small airways assessed by decrements in FEF25_75 and altered ventilation distribution
15                   have been reported (Frank etal.. 2001; Foster etal.. 1997).
                     6.2.1.2    Epidemiology

16                   The O3-induced lung function decrements consistently demonstrated in controlled human
17                   exposure studies (Section 6.2.1.1) provide biological plausibility for the epidemiologic
18                   evidence presented in the 1996 and 2006 O3 AQCDs, in which short-term ambient O3
19                   exposure was consistently associated with lung function decrements in diverse
20                   populations (U.S. EPA. 2006b. 1996a). Coherence between the two disciplines was found
21                   not only for effects observed in groups with higher expected personal O3 exposures and
22                   higher exertion levels, including children attending summer camps and adults exercising
23                   or working outdoors, but also for effects observed in children and individuals with pre-
24                   existing respiratory disease such as asthma (U.S. EPA. 2006b. 1996a). Recent
25                   epidemiologic studies focused more on children with asthma rather than on groups with
26                   increased outdoor exposures or other healthy populations. Whereas a majority of recent
27                   studies conducted in children with asthma indicated decreases in lung function in
28                   association with increases in ambient O3 exposure, recent studies in adults with asthma
29                   and individuals without asthma found both O3-associated decreases and increases in lung
30                   function. Recent studies also provided additional data to assess whether particular lags of
31                   O3 exposure were more strongly associated with decrements in lung function; whether O3
32                   associations were confounded by copollutant exposures; and whether risk was affected by
33                   factors such as corticosteroid (CS) use, genetic polymorphisms, elevated BMI, and diet.
      Draft - Do Not Cite or Quote                       6-23                                September 2011

-------
Table 6-1
Study
Korricketal.
(1998)
Thurston et al.
(1997)
Spektor et al.
(1988b)
Spektor et al.
(1988a)
Spektor and
Lippmann (1991)
Berryetal.
(1991)
Neas et al.
(1999)
Girardot et al.
(2006)
Selwyn etal.
(1985)
Thaller etal.
(2008)
Higginsetal.
Avol etal. (1990)
Burnett etal.
(1990)
Raizenne etal.
(1989)
Bra uer etal.
(1996)
Castillejosetal.
(1995)
Romieu et al.
(1998a)
Nickmilderet al.
(2007)
Brunekreef etal.
(1994)
Hoeketal.
(1993)
Braun-Fahrlander
etal. (1994)
Hoppe et al.
(1995): Hoppe et
al. (2003)
Chan etal.
(2005)
Mean and upper percentile concentrations of ozone in
epidemiologic studies examining lung function in populations with
increased outdoor exposures
Location
Mt. Washington,
NH
Connecticut
River Valley, CT
Tuxedo, NY
Fairview Lake,
NJ
Fairview Lake,
NJ
Hamilton, NJ
Philadelphia, PA
Great Smoky
Mountain NP, TN
Houston, TX
Galveston, TX
San Bernardino,
CA
Idyllwild, CA
Lake
Couchiching,
Ontario, CA
Lake Erie,
Ontario, CA
British Columbia,
Canada
Mexico City,
Mexico
Mexico City,
Mexico
Southern
Belgium
Netherlands
Wageningen,
Netherlands
Southern
Switzerland
Munich,
Germany
Taichung City,
Taiwan
Years/Season
1991,1992
Warm season
1991-1993
Warm season
1985
Warm season
1984
Warm season
1988
Warm season
July 1988
1993
Warm season
2002-2003
Warm season
1981
Warm season
2002-2004
Warm season
1987
Warm season
1988
Warm season
1983
Warm season
1986
Warm season
1993
Warm season
June 1990-
October1991
March-August
1996
2002
Warm season
1981
Warm season
1989
Warm season
1989
Warm season
1992
Warm season
2001
Cold season
Os Averaging Time
Hike-time avg
(2-1 2 h)
1-hmax
1 -h avg
1-havga
1-havga
1-h max
12-havg
(9:00 a.m.9:00 p.m.)
Hike-time avg
(2-9 h)
15-min max
1-hmax
1-havga
1-havga
1-havga
1-havga
1-h max
1-hmax
Work shift avg (6-1 2 h)
1-h max
8-h max
Exercise-time avg (10-
145min)
1-hmax
30-min avg
30-min max (1:00p.m.-
4:00 p.m.)
8-h avg
(9:00 a.m.-5:00 p.m.)
Mean/Median
Concentration
(ppb)
40
83.6
NR
53
69
NR
57.5 (Camp 1)
55.9 (Camp 2)
48. 1b
47
35 (median)
123
94
59
71
40
179
67.3
NR
42.8°
NR
NR
High 03 days: 65.9
Control 03 days: 27.2
35.6
Upper Percentile
Concentrations (ppb)
Max: 74
Max: 160
Max: 124
Max (1-h max): 113
Max (1-h max): 137
Max: 204
Max (Camp 1): 106
Max: 74.2b
Max: 135
Max: 118
Max: 245
Max: 161
Max: 95
Max (1-h max): 143
Max: 84
Max: 365
95th: 105.8
Max (across 6 camps): 24.5-112.7°
Max (across 6 camps): 18.9-81.1°
Max: 99.5°
Max: 122°
Max: 80°
Max (high 03 days): 86
Max: 65.1
Max = Maximum; NR = not reported
a1-h avg, preceding lung function measurement.
blndividual-level exposure estimates were derived based on time-activity diary data.
'Concentrations were converted from ug/m3 to ppb using the conversion factor of 0.51 assuming standard temperature (25°C) and pressure (1
atm).
Draft - Do Not Cite or Quote
6-24
September 2011

-------
                     Populations with Increased Outdoor Exposures

 1                   Few epidemiologic studies characterizing acute O3-related respiratory morbidity have
 2                   accounted for time spent outdoors, which may be an important determinant of
 3                   interindividual variability in personal O3 exposure. Among epidemiologic studies, studies
 4                   of individuals engaged in outdoor recreation, exercise, or work are more comparable to
 5                   controlled exposure studies because of improved estimates O3 exposures, measurement of
 6                   lung function before and after discrete periods of outdoor activity, and examination of O3
 7                   effects during exertion when the dose of O3 reaching the lungs may be higher because  of
 8                   higher ventilation and inhalation of larger volumes of air. Characteristics and ambient  O3
 9                   concentration data from epidemiologic studies of populations with increased outdoor
10                   exposures are presented in Table 6-1. Similar to findings from controlled human
11                   exposure studies, the collective body of epidemiologic evidence clearly demonstrates
12                   decrements in lung function in association with O3 exposures during periods of outdoor
13                   activity or exercise of varying intensity and duration (15 minutes to 12 hours) (Figures 6-
14                   3 to 6-5 and Tables 6-2 to 6-4).

                         Children Attending Summer Camps
15                   Studies of children attending summer camps, most of which were discussed in the 1996
16                   O3 AQCD, have provided important understanding of the impact of ambient O3 exposure
17                   on respiratory effects in young, healthy children. These studies were noted for their on-
18                   site measurement of ambient O3 and daily assessment of lung function by trained staff
19                   over 1- to 2-week periods (Thurston et al.. 1997; Berry etal.. 1991; Spektor and
20                   Lippmann. 1991; Avoletal. 1990; Burnett et al.. 1990; Higgins et al.. 1990; Raizenne et
21                   al.. 1989: Spektor et al.. 1988a: Raizenne etal.. 1987V

22                   In groups mostly comprising healthy children (ages 7-17 years), decrements in FEVi
23                   were found to be associated consistently with ambient  O3 exposures averaged over the
24                   1-8 hours preceding lung function measurement (Figure 6-3  and Table 6-2). Kinney et al.
25                   (1996) corroborated this association in a reanalysis combining 5367 lung function
26                   measurements collected from 616 healthy children from six  studies (Spektor and
27                   Lippmann. 1991; Avoletal.. 1990; Burnett et al.. 1990; Higgins et al.. 1990; Spektor et
28                   al.. 1988a; Raizenne et al.. 1987). Based on uniform statistical methods, a 40-ppb
29                   increase in concurrent-hour O3 exposure was associated with a -20 ml (95% CI: -25, -14)
30                   change in afternoon FEVi: (Kinney et al.. 1996). In these studies conducted in locations
        1 To facilitate comparisons among epidemiologic studies, for all health endpoints in Chapter 6, effect estimates are presented in
      terms of a standard increment in ambient O3 concentration, one for each of the three commonly examined O3 averaging times (1-h
      max, 8-h max, and 24-h average). These standard increments are 40 ppb, 30 ppb, and 20 ppb for 1-h max, 8-h max, and 24-h avg
      O3, respectively, and are based on annual mean to 95th percentile differences that are representative of measurements from
      nationwide O3 monitors in U.S. Metropolitan Statistical Areas as described in detail in Section 7.1.3.2 of the 2006 O3AQCD (U.S.
      EPA, 2006b).
      Draft - Do Not Cite or Quote                       6-25                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
across the Northeast U.S. and Canada and California (Table 6-1) with varying pollutant
mix, a wide range in effect estimates was found. Study-specific effect estimates ranged
between a 0.76 and 48 ml decrease or a 0.3% to 2.2% decrease in study mean FE\V

Associations between ambient O3 exposure and peak expiratory flow (PEF) in camp
studies were more variable than were those with FEVi, as indicated by the wider range in
effect estimates and wider 95% CIs (Figure 6-3 and Table 6-2). Nonetheless, most effect
estimates indicated  decreases in PEF in association with ambient O3 exposure. The
largest effect (mean 2.8% decline per 40-ppb increase in  1-h max O3) was estimated in a
group of campers with asthma (Thurston et al.. 1997). In  this study, O3 also was
associated with increases in chest symptoms and bronchodilator use, suggesting that the
observed decreases  in PEF may have been indicative of clinically significant effects.
Study
FEV, (mil
Spektoretal. (1988a)
Spektorand Lippmann
(1991)
Raizenne et al. (1987)
Burnett et al. (1990)
Higginsetal. (1990)
Avoletal. (1990)
Kinneyetal. (1996)
Berry etal. (1991)
Population
Camperswithout asthma <
Camperswithout asthma 	 <
Camperswithout asthma -•-
Camperswithout asthma — • —
Camperswithout asthma — •—
Pooled estimate •••
Camperswithout asthma



                                                -160    -120    -80     -40      0      40      80
                                                  Change in FEV1 (ml) per standardized increment in O3 (95% Cl)
              PEF (ml/sec)
              Spektoretal. (1988a)
              Raizenne et al. (1987)
              Burnett etal. (1990)
              Higginsetal. (1990)
              Avoletal. (1990)
              Kinneyetal. (1996)
              Berry etal. (1991)
              Neasetal. (1999)
              Thurston et al. (1997)
         Camperswithout asthma
         Camperswithout asthma
         Camperswithout asthma
         Camperswithout asthma
         Camperswithout asthma
         Pooled estimate
         Camperswithout asthma
         Camperswithout asthma
         Campers with asthma
•*—•-
                                                -160    -120    -80     -40      0      40      80
                                                  Change in PEF (ml/sec) per standardized increment in O3 (95% Cl)

        Effect estimates are from single-pollutant models and are standardized to a 40-ppb increase for 1-h avg or 1-h max ozone
      exposures and a 30-ppb increase for 12-h avg ozone exposures.

      Figure 6-3     Changes in FEVi (ml) or PEF (ml/sec) in association with ambient
                       ozone exposure in studies of children attending summer camp.
      Draft - Do Not Cite or Quote
                                6-26
                                           September 2011

-------
       Table 6-2       Additional characteristics and quantitative data for studies
                          represented in  Figure 6-3
       Study
Location
                     Population
Standardized percent    Standardized effect
change (95% Cl)a        estimate (95% Cl)a
       FEV,
                                                                     (ml)
       Spektor et al. (1988a)    Fairview Lake, NJ
                     Campers without asthma
                                              -0.93 (-1.5,-0.35)
                       -20.0 (-32.5, -7.5)
Fairview Lake, NJ
                     Campers without asthma
                                                                        -2.2 (-3.1, -1.3)
                       -51. 6 (-72.8,-30.4)
       Raizenne et al. (1989)      ke Erie,
                     Campers without asthma
                                              -0.48 (-0.80,-0.16)
                       -11.6 (-19.4,-3.8)
       Burnett etal. (1990
Lake Couchiching,
Ontario, Canada
                     Campers without asthma
-0.32 (-1.8,1.2)
                                                                                        -7.6 (-42.1,26.9)
       Higgins et al. (1990)     San Bernardino, CA
                     Campers without asthma
                                             -1.6 (-2.4,-0.87)
                       -33.6 (-49.3,-17.9)
       Avol etal. (1991
Pine Springs, CA
                     Campers without asthma
-0.58 (-1.1,-0.12)
                                                                                        -12.8 (-23.0,-2.6)
       Kinney et al. (1996)      Pooled analysis
                     Campers without asthma
                                             -0.90 (-1.2,-0.65)
                       -20.0 (-25.5,-14.5)
       Berry etal. (1991)
Hamilton, NJ
                     Campers without asthma
Data not available
                                                                                        32.8 (6.9, 58.7)
       PEF
                                                                     (ml/sec)
       Spektor et al. (1988a)    Lake Fairview, NJ
                     Campers without asthma
                                              -1.8 (-3.3,-0.40)
                       -80.0 (-142.7,-17.3)
       Raizenne etal. (1989
Lake Erie, Ontario,
Canada
                     Campers without asthma
-0.07 (-0.56, 0.41)
                                                                                        -4.0 (-30.7, 22.7)
Burnett etal. (1990)
                           Lake Couchiching,
                           Ontario, Canada
                     Campers without asthma
                                              -1.9 (-3.8, -0.05)
                       -106.4 (-209.9,-2.9)
       Higgins et al. (1990)     San Bernardino, CA
                     Campers without asthma
                                             -0.87 (-2.1,-0.34)
                       -44.0 (-105,-17.2)
       Avol etal. (1991)
Pine Springs, CA
                     Campers without asthma
1.9(0.71,3.1)
                                                                                        86.8(31.9,142)
       Kinnevetal. (1996'
Pooled analysis
                     Campers without asthma
0.31 (-0.88,1.5)
                                                                                        6.8 (-19.1, 32.7)
       Berry etal. (1991)
Hamilton, NJ
                     Campers without asthma
Data not available
                                                                                        -40.4 (-132.1,51.3)
       Neasetal. (1999'
Philadelphia, PA
                     Campers without asthma
-0.58 (-1.5, 0.33)
                                                                                        -27.5 (-70.8, 15.8)
       Thurston et al. (1997)    CT River Valley, CT
                     Campers with asthma
                                              -2.8 (-4.9, -0.59)
                       -146.7 (-261.7,-31.7)
        aAII effect estimates are standardized to a 40-ppb increase in 1 -h avg or 1 -h max 03, except that from Neas et al. (1999). which is standardized
       to a 30-ppb increase in 12-h avg (9:00 a.m.-9:00 p.m.) 03.

 1                      As has been observed in controlled human exposure studies, FEVi  and PEF responses to
 2                      ambient O3 exposure  varied among individual campers. Based on separate regression
 3                      analyses of data from individual subjects, O3 exposure was associated with a wide range
 4                      of changes in lung function across subjects (Berry etal.. 1991; Fliggins et al.. 1990;
 5                      Spektor et al.. 1988a). For example, in the study of children attending camp in Fairview
 6                      Lake, NJ, 36% of subjects had statistically significant O3-associated decreases in FEVi,
 7                      and the upper decile of response was a 6.3% decrease in FEVi per a 40-pbb increase in 1-
 8                      h avg O3 (Spektor et al..  1988a).

 9                      In contrast with these previous studies, a recent cross-sectional study of children
10                      attending six different summer camps in Belgium did not find an association between
11                      ambient O3 exposure  and lung function. The ambient O3 concentrations in this recent
12                      study was in the range of those in previous studies (Table 6-1); however, this recent study
       Draft - Do Not Cite or Quote
                                 6-27
                                                                           September 2011

-------
 1
 2
 3
 4
differed from previous studies in that each subject was examined only on one day, and
investigators performed between-camp comparisons rather than within-subject
comparisons. Camps with higher daily 1-h max O3 concentrations did not consistently
have larger decreases in mean intraday FEVi or FEVi/FVC (Nickmilder et al.. 2007).
 5
 6
 7
 8
 9
10
11
12
13
Populations Exercising Outdoors

Similar to camp studies, studies of individuals exercising outdoors were noted for the
serial examination of subjects over days with a wide range in ambient O3 concentrations
and onsite assessment of O3 exposures during discrete periods of outdoor exercise. These
studies collectively show that mean O3 exposures ranging from 40 to 66 ppb during
exercise of variable duration and intensity are associated with small (< 1 to 4% per
standardized increment in (V) decreases in lung function in adults (Figure 6-4 and Table
6-3). Similar observations were made in children exercising outdoors (Table 6-3). For
both adults and children, evidence was provided largely by older studies that were
reviewed in the 1996 and 2006 O3 AQCDs.
Study
Korricketal. (1998)
Girardotetal. (2006)
Hoppeetal. (2003)
Spektoretal. (1988b)
Brunekreefetal. (1994)
Population
Adults hiking
Adults hiking
Adults exercising
Adults exercising
Adults exercising
Exercise Duration








                                                          -4
                                                  -2
-1
                                                          Percentchangein FEV1 per standardized
                                                                incrementin O3 (95% Cl)
      Figure 6-4     Percent change in FEVi in association with ambient ozone
                     exposures of adults exercising outdoors. Studies generally are
                     organized in order of decreasing exercise duration. Effect
                     estimates are from single-pollutant models and are standardized to
                     a 40-ppb increase for ozone exposures averaged over 15 minutes
                     to 1  hour and a 30-ppb increase for ozone exposures averaged over
                     3 to 8 hours.
       1 Effect estimates were standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max, and 24-h avg O3.
      Draft - Do Not Cite or Quote
                             6-28
       September 2011

-------
Table 6-3
Study
Additional characteristics and quantitative data for studies
represented in Figure 6-4 and results from studies in children
exercising outdoors
Location
Population
Exercise duration
Os Averaging
Time
Parameter
Standardized percent
change (95% Cl)a
Studies of adults
Korricketal. Q998)
Girardot et al.
(2006)
Hoppe et al. (2003)
Selwynetal.
M QR^b
> laoai
Spektor et al.
(1988a)b
Brunekreef etal.
(1994)
Mt. Washington,
NH
Great Smoky Mt,
TN
Munich, Germany
Houston, TX
Tuxedo, NY
Netherlands
Studies of children not included in
Braun-Fahrlander
etal. (1994)
Castillejosetal.
(1995)
Hoek etal. (1993)
Switzerland
Mexico City,
Mexico
Wageningen,
Netherlands
Adult day hikers
Adult day hikers
Adults exercising
Adults exercising
Adults exercising
Adults exercising
Figure 6-4
Children
exercising
Children
exercising
Children
exercising
2-1 2 h
2-9 h
2h
NR
15-55min
10min-1 h

10 min
15min (2 periods)
1 h
Hike duration
Hike duration
30-min max (1 :00
p.m. -4:00 p.m.)
15-min max
30-min avg
Exercise duration

30-min avg
1 -h avg
1 -h avg
FEV,
FEV,
FEV,
PEF
FEV,
FEV,
FEV,

PEF
FEV,
PEF
-1.5 (-2.8, -0.24)
0.72 (-0.46, 1 .90)
-1.3 (-2.6, 0.13)
-2.8 (-5.9, 0.44)
-16 ml (-31.1, -0.87)°
-1.31 (-2.0, -0.65)
-0.82 (-1.6, -0.02)

-3.8 (-6.9, -0.96)
-0.48 (-0.72, -0.24)
-2.2 (-4.9, 0.55)
         NR= Not reported.
         "Effect estimates are standardized to a 40-ppb increase for 03 exposures averaged over 15 min to 1 h and a 30-ppb increase for 03 exposures
        averaged over 3 to 8 h.
         bResults not included in the figure because data were not provided to calculate percent change in lung function.
         The 95% Cl was constructed using a standard error that was estimated from the p-value
 1                    Two studies of adult day-hikers of similar design and ambient O3 concentrations
 2                    produced contrasting results (Girardot et al.. 2006; Korricketal.. 1998). These studies
 3                    mostly comprised white, healthy adults and examined changes in lung function associated
 4                    with O3 exposures during multihour (2-12 h) periods of outdoor exercise. Although
 5                    analyses of day-hikers were based on a one-time assessment of lung function, they
 6                    included much larger sample sizes compared with panel studies of individuals exercising
 7                    outdoors. Among 530 hikers on Mt. Washington, NH, Korrick et al. (1998) reported
 8                    posthike declines in FEVi and FVC of approximately 0.7-1.5% per a 30-ppb increase in
 9                    2- to 12-h avg O3. In contrast, among 354 hikers in Great Smoky Mountains National
10                    Park, TN, Girardot et al. (2006) more recently found that O3 exposure was associated
11                    with posthike increases in many of the same lung function indices. Several differences in
12                    study characteristics were used by Girardot et al. (2006) to explain discrepant results,
13                    including their use of a larger number of less-well trained technicians, shorter mean
14                    duration of hike (5 hours versus 8 hours), and older mean age of their subjects.

15                    As was observed in camp studies, the magnitudes of O3-associated decreases in lung
16                    function varied among individual subjects. Korrick et al. (1998) found larger O3-
      Draft - Do Not Cite or Quote
6-29
September 2011

-------
 1                  associated decreases in FEVi among hikers who were male, had history of asthma or
 2                  wheeze, were never smokers, and hiked greater than 8 hours. Additionally, O3 was
 3                  associated with an increased odds of a greater than 10% decline in FEF25_75o/0 among
 4                  hikers (OR: 2.3 [95% CI:  1.2, 6.7] per 30-ppb increase in 2- to  12-h avg O3) (Korrick et
 5                  al.. 1998). Likewise, Hoppe et al. (2003) found that on days with 30-min max (1:00 p.m.-
 6                  4:00 p.m.) ambient O3 concentrations above 50 ppb, 14% of athletes had at least a 20%
 7                  decrease in lung function or 10% increase in airway resistance.


                    Outdoor Workers

 8                  The 2006 O3 AQCD indicated that ambient O3 exposure was associated consistently with
 9                  decrements in lung function among outdoor workers (U.S. EPA. 2006b). and recent
10                  studies produced similar findings (Thaller et al., 2008; Chan and Wu. 2005) (Figure 6-5
11                  and Table 6-4). Although most of these studies assessed O3 exposures using central site
12                  measurements, they were noteworthy for the long periods of time spent outdoors (6-14
13                  hours across studies). Further, associations between O3 exposure and lung function
14                  decrements were found for time periods during which ambient O3 concentrations did not
15                  exceed 80 ppb (Table 6-1) (Chan and Wu. 2005: Braueretal.. 1996: Hoppe etal.. 1995).
16                  In particular, Many studies of outdoor workers found that in addition to same-day
17                  exposures, O3 exposures lagged 1 or 2 days (Chan and Wu. 2005: Braueretal.. 1996)  or
18                  exposures averaged over 2 days (Romieu et al., 1998a) were associated with equal or
19                  larger decrements in lung  function (Figure  6-5 and Table 6-4).

20                  Similar to other populations with increased outdoor exposure, the magnitudes of O3-
21                  associated lung function decrements in outdoor workers were small. Per standardized
22                  increment in O3 concentration1, decreases in lung function ranged between less than 1%
23                  and 3.6%. The magnitude of decrease was not found to depend strongly on duration of
24                  outdoor work or ambient O3 concentration. The largest decrease (6.4%  per 40-ppb
25                  increase in 1-h max O3) was observed among berry pickers in British Columbia who were
26                  exposed to relatively low ambient O3 concentrations (work shift mean:  26.0 ppb [SD:
27                  H-8]) but had longer periods of outdoor work (8-14 hours) (Braueretal.. 1996) (Figure
28                  6-5 and Table 6-4). However, a much smaller O3-associated decrease in FEVi was found
29                  among street workers in Mexico City who were exposed to higher O3 concentrations
30                  (work shift mean: 67.3 ppb [SD: 24]) during a similar duration of outdoor work. Among
31                  studies of outdoor workers, the smallest magnitude of decrease (0.4% decrease (95% CI:
32                  -0.8, 0) in afternoon FEVi/FVC per 40-ppb increase in 1-h max O3) was observed among
33                  lifeguards in Galveston, TX (Thaller et al.. 2008) whose outdoor work periods were
34                  shorter than those of the berry pickers but who were exposed to a similar range of
       1 Effect estimates were standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max, and 24-h avg O3.
      Draft - Do Not Cite or Quote                      6-30                                September 2011

-------
 Study           Population


 Thaller et al. (2008)   Lifeguards
Parameter  O3 Lag   Subgroup


FVC      0
FEV^FVC
 Brauer et al. (1996)  Berry pickers    FEV,     0
                                  1
 Hoppe et al. (1995)  Forestry workers  FEV,
 Romieu et al. (1998) Streetworkers   FEV,
        0      Placebo
               Antioxidant supplement
        0-1 avg  Placebo
               Antioxidant supplement
                                                      -7
                                                                          -2   -1
                                                  Percentchange in lung function per standardized increment
                                                                  in 03 (95% Cl)

Figure 6-5     Percent change in lung function in association with ambient ozone
                exposures among outdoor workers.  Studies generally are
                organized in order of increasing mean ambient ozone
                concentration. Effect estimates are from single-pollutant models
                and are standardized to a 40-ppb increase for 30-min or  1-h  max
                ozone exposures.
Table 6-4     Additional characteristics and quantitative data for studies
               represented in Figure 6-5
Study
Thaller et
al. (2008)
Braueretal.
(1996)
Hoppe et al.
(1995)
Romieu et
al. (1998a)
Chanetal.
(2005)"
Location
Galveston, TX
British
Columbia,
Canada
Munich,
Germany
Mexico City,
Mexico
Taichung City,
Taiwan
Population
Lifeguards
Berry pickers
Forestry
workers
Male street
workers
Mail carriers
Parameter
FVC
FEV,/FVC
FEV,
FEV,
FEV,
PEF
Duration of
outdoor work
6-8 h
8-14 h
NR
Mean (SD): 9 h
(1)
8h
O3 Averaging
Time
1-h max
1-h max
30-min max (1 :00
p.m. -4:00 p.m.)
1-h max
8-h avg (9:00 a.m.-
5:00 p.m.)
03
Lag
0
0
1
0
0
0-1
avg
0
1
Subgroup



Placebo
Antioxidant
Placebo
Antioxidant

Standardized percent
change (95% Cl)a
0.24 (-0.28, 0.72)
-0.40 (-0.80, 0)
-5.4 (-6.5, -4.3)
-6.4 (-8.0, -4.7)
-1.4 (-3.0, 0.16)
-2.1 (-3.3, -0.85)
-0.52 (-2.0, 0.97)
-3.4 (-6.0, -0.78)
-1.2 (-4 .2, 1.8)
-1 .0 (-1 .3, -0.66)
-1.1 (-1.5, -0.78)
          NR= Not reported.
          'Effect estimates are standardized to a 40-ppb increase for 30-min or 1-h max 03 and a 30-ppb increase for 8-h max 03.
          aPEF results not included in figure.
Draft - Do Not Cite or Quote
                 6-31
September 2011

-------
 1
 2
 3
 4
ambient O3 concentrations. Few studies provided information on ventilation rate or pulse
rate, thus it was difficult to ascertain whether differences in the magnitudes of O3-
associated decreases in lung function were related to differences in workers' levels of
exertion.
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
Associations at lower ozone concentrations

Studies of populations engaged in outdoor activity examined and found that associations
between O3 and lung function decrements persisted at lower O3 concentrations
(Table 6-5). Among adults exercising outdoors, Spektor et al. (1988b) found that
associations persisted in analyses restricted to 30-min max ambient O3 concentrations
less than 80 ppb, and for most lung function parameters, effect estimates were similar to
those obtained for the full range of O3 concentrations (Table 6-5). In a study of children
attending summer camp, similar  effects were estimated for the full range of 1-h avg O3
concentrations and those less than 60 ppb (Spektor etal.. 1988a). Brunekreef et al. (1994)
found ambient O3 exposure (10-min to 1-h) during outdoor exercise to be associated with
decreases in lung function in analyses restricted to concentrations less than 61 (Table 6-5)
and 51 ppb. However, effect estimates were near zero with O3 concentrations less than
41 ppb (Brunekreef et al., 1994). In contrast, Brauer et al. (1996) found associations
persisted with 1-h max O3 concentrations less than 40 ppb.
Table 6-5
Study
Brunekreef etal. (1994)
Spektor et al. (1988b)
Spektor et al. (1988a)
Korrick etal. Q998)
Associations between ambient ozone exposure and lung function
decrements in different ranges of ambient ozone concentrations
Location Population
Netherlands Adults exercising
Tuxedo, NY Adults exercising
Fairview Lake, NJ Campers without
asthma
Mt. Washington, Adult day hikers
NH
Parameter
% change
FEV,
FEV, (ml)
% change
FEV,
% change
FEV,
03
Averaging
Time
10-mto1-h
LagO
30-min avg
LagO
1 -h avg
LagO
Hike duration
(2-12 h)
LagO
03
Concentration
Range
Full range
03 < 61 ppb
Full range
03 < 80 ppb
Full range
03 < 80 ppb
03 < 60 ppb
Full range
03 > 40 ppb
Standardized percent
change (95% Cl)a
-0.82 (-1 .6, -0.02)
-2.1 (-4.5, 0.32)
-54 (-84, -27)b
-52(-101,-3.4)b
-2.7 (-3.3, -2.0)
-1.4 (-2.5, -0.34)
-2.2 (-3.7, -0.80)
-1.5 (-2.8, -0.24)
-2.6 (-4.9, -0.32)
        "Effect estimates are standardized to a 40-ppb increase for 03 exposures averaged over 10 min to 1 h and a 30-ppb increase for 03 exposures
      averaged over 2 to12 h.
        bData were not provided to calculate percent change.
18
19
20
Korrick et al. (1998) examined associations with hike-time average O3 exposures (2-12 h)
and found effect estimates that were more negative in analyses restricted to O3
concentrations greater than 40 ppb. Based on the results from a nonparametric model in
      Draft - Do Not Cite or Quote
                                6-32
September 2011

-------
 1                   Korrick et al. (1998). it appeared that the association between O3 exposure and lung
 2                   function decrements in this population was limited to 2- to 12-h avg O3 exposures above
 3                   40 ppb.


                     Children with Asthma

 4                   Associations between ambient O3 exposures and lung function decrements in children
 5                   with asthma have been examined in epidemiologic studies conducted across diverse
 6                   geographical locations and a range of ambient O3 concentrations (Table 6-6). Whereas
 7                   studies of populations with increased outdoor exposures monitored O3 exposures at the
 8                   site of subjects' outdoor activities and used trained staff to measure lung function, studies
 9                   of children with asthma relied more heavily on O3 measured at central monitoring sites
10                   and lung function measured by subjects. However, studies of children with asthma have
11                   provided more information on factors that may confer increased susceptibility to the
12                   respiratory effects of O3 exposure, confounding by copollutant exposure or meteorology,
13                   and the potential clinical significance of O3-associated changes in lung function with the
14                   concurrent assessment of respiratory symptoms.

15                   Collectively, the large body of evidence, which includes large U.S. multicity studies and
16                   several smaller studies conducted in the U.S., Mexico City, and Europe, demonstrates
17                   that increases in ambient O3 exposure (various averaging times and lags) are associated
18                   with decrements in FEVi (Figure 6-6 and Table 6-7) and PEF (Figure 6-7 and Table 6-8)
19                   in children with asthma. In addition to examining a single lung function measurement per
20                   day, several studies examined associations of O3 exposure with measures of lung function
21                   variability. Although different definitions of variability were used, studies  consistently
22                   found that O3-associated changes in lung function variability were indicative of poorer
23                   lung function, whether characterized as a decrease from the individual's mean lung
24                   function over the study period (Jalaludin et al.. 2000). a decrease in lung function over
25                   the course of the day (Lewis et al.. 2005). or a decrease in the lowest daily measurement
26                   (Just etal.. 2002).

27                   Studies of children with asthma that were restricted to winter months provided little
28                   evidence of an association between  various single- and multi-day lags of ambient O3
29                   exposure and changes in lung function; several studies reported O3-associated increases
30                   in lung function (Dales et al.. 2009; Liu et al.. 2009a: Rabinovitch et al.. 2004). In colder
31                   months, ambient O3 concentrations are low and in many locations, children remain
32                   primarily indoors. Thus, it is less likely that effects will be demonstrated for O3. As noted
33                   in previous AQCDs for lung function and other endpoints such as respiratory hospital
34                   admissions, ED visits, and mortality, associations with O3 generally are greater in the
35                   warm season.
      Draft - Do Not Cite or Quote                       6-33                                 September 2011

-------
Table 6-6      Mean and upper percentile concentrations of ozone in
                epidemiologic studies examining lung function in children with
                asthma
Study
Mortimer etal. (2002)
Mortimer etal. (2000)
O'Connor etal. (2008)
Thurston et al. (1997)
Lewis etal. (2005)
Rabinovitchetal.
(2004)
Delfino et al. (2004)
Dales etal. (2009)
Liuetal. (2009a)
Romieu et al. (1996)
Romieu et al. (1997)
Romieu etal. (2002):
Romieu etal. (2004a):
Romieu et al. (2006)
Barraza-Villarreal et al.
(2008): Romieu etal.
(2009)
Hernandez-Cadena et
al. (2009)
Gielen etal. (1997)
Just etal. (2002)
Hoppe et al. (2003)
Wiwatanadate and
Trakultivakorn (2010)
Jalaludin et al. (2000)
Location
Bronx, East Harlem, NY; Baltimore,
MD; Washington, DC; Detroit, Ml,
Cleveland, OH; Chicago,!!.; St.
Louis, MO (NCICAS)
Boston, MA; Bronx, Manhattan NY;
Chicago, IL; Dallas, TX, Seattle,
WA; Tucson, AZ(ICAS)
Connecticut River Valley, CT
Detroit, Ml
Denver, CO
Alpine, CA
Windsor, ON, Canada
Northern Mexico City, Mexico
Southern Mexico City, Mexico
Mexico City, Mexico
Mexico City, Mexico
Mexico City, Mexico
Amsterdam, Netherlands
Paris, France
Munich, Germany
Chiang Mai, Thailand
Sydney, Australia
Years/Season
1993
Warm season
1998-2001
All-year
1991-1993
Warm season
2001-2002
All-year
1999-2002
Cold season
September-
October 1999
April-June 2000
2005
Cold season
April-July 1991
November 1991-
February 1992
April-July 1991
November 1991 -
February 1992
1998-2000
All-year
2003-2005
All-year
2005
Warm season
1995
Warm season
April-June 1996
1992-1995
Warm season
August 2005-June
2006
February-
December 1994
O3Averging
Time
8-h avg
(10:00a.m.-
6:00 p.m.)
24-h avg
1-h max
8-h max
1-h max
8-h max
24-h avg
1-h max
1-h max
1-h max
8-h max
1 -h max
8-h max
1 -h max
24-h avg
1-h max
8-h max
24-h avg
30-min max
(1:00 p.m. -4:00
p.m.)
24-h avg
15-h avg (6:00
a.m.-9:00p.m.)
Mean/Median
Concentration
(PPb)
48
NR
83.6a
Eastside: 40.4a
Westside:41.4a
28.2
62.9
14.1
27.2
190
196
69
102
31.6
86.5
26.3
74.5
34.2
30.0
High 03 days: 66.9
Control 03 days: 32.5
17.5
12
Upper Percentile
Concentrations
(PPb)
NR
NR
Max: 160
Overall max: 92.0a
Max 70.0
90th: 83.9, Max: 105.9
75th: 17.8
75th: 32.8
Max: 370
Max: 390
Max: 184
Max: 309
Max (8-h): 86.3
75th: 35.3; Max: 62.8
75th: 92.5; Max: 165.0
Max: 56.5
Max: 61 .7
Max: 91 (high 03 days)
39 (control 03 days)
90th: 26.82
Max: 34.65
Max: 43
 NCICAS = National Cooperative Inner-City Asthma Study, NR = Not Reported, ICAS = Inner City Asthma Study, Max = Maximum.
'Measured at sites established by investigators.
Draft - Do Not Cite or Quote
6-34
September 2011

-------
 Study

 Liu et al. (2009)
 Lewis et al. (2005)
 Hoppe et al. (2003)
O3Lag

0
2
Subgroup
        CS user
        With URI
1        Without asthma
        With asthma
 Barraza-Villarreal et al. (2008)  0-4 avg
 Romieu et al. (2002)
 Romieu et al. (2006)
        Without asthma
        With asthma

        Placebo
        Antioxidant
        Placebo, moderate/severe asthma
        Antioxidant, moderate/severe asthma

        GSTP1 lie/lie Ile/Val
        GSTP1 Val/Val
                                             -10    -8    -6-4-20     2     4
                                              Percent change in FEVi per standardized increment in O3 (95% Cl)

Figure 6-6     Percent change in FEVi in association with ambient ozone
               exposures among children with asthma.  Results generally are
               presented in order of increasing mean ambient ozone
               concentration. CS = Corticosteroid, URI = Upper respiratory
               infection. Effect estimates are from single-pollutant models and are
               standardized to a 40-ppb increase for 30-min or 1-h max ozone
               exposures, a 30-ppb increase for 8-h max or 8-h avg ozone
               exposures, and a 20-ppb increase for 24-h avg ozone exposures.
Draft - Do Not Cite or Quote
                      6-35
                                               September 2011

-------
Table 6-7       Additional characteristics and quantitative data for studies
                   represented in Figure 6-6
Study
Liu et al. (2009a)
Lewis etal. (2005)
Hoppe et al. (2003)
Barraza-Villarreal
etal. (2008)
Romieu et al.
(2002)
Romieu et al.
(2006)
Location/ Population
Windsor, ON, Canada
Children with asthma
Detroit, Ml
Children with asthma
Munich, Germany
Children
Mexico City, Mexico
Children
Mexico City, Mexico
Children with asthma
Mexico City, Mexico
Children with asthma
03
Averaging Os Lag
Time
24-h avg 0
8-h max 2
30-min max 1
(1:00p.m.-
4:00p.m.)
8-h max 0-4 avg
1-hmax 1
1-hmax 1
Parameter
FEV,
Lowest daily FEV,
Afternoon FEV,
Afternoon FVC
FEV,
FEV,
FEV,
Subgroup

CS user
With URI
Without asthma
With asthma
Without asthma
With asthma
Without asthma
With asthma
Placebo
Antioxidant
Placebo, moderate/severe
asthma
Antioxidant,
moderate/severe
asthma
GSTP1 lie/lie or Ile/Val
GSTP1 ValA/al
Standardized
percent change
(95% Clf
-0.89 (-3.5, 1 .8)
-8.0 (-13.5, -2.1)
-5.4 (-11. 3, 1.0)
0.93 (-0.80, 2.7)
-0.56 (-4.6, 3.7)
-0.09 (-1 .7, 1 .6)
-3.5 (-5.9, -1 .0)
-1 .5 (-4.7, 1 .7)b
-0.1 2 (-2.0, 1.8)"
-0.21 (-0.78, 0.36)b
0.05 (-0.59, 0.69)b
-1.1 (-2.0, -0.19)b
-0.04 (-0.92, 0.83)b
-0.51 (-1.1,0.05)
0.50 (-0.25, 1.3)
Studies not included in Figure 6-6b
Dales etal. (2009)
Rabinovitch etal.
(2004)
O'Connor etal.
(2008)
Windsor, ON, Canada
Children with asthma
Denver, CO
Children with asthma
7 U.S. communities
Children with asthma
1-hmax 0
1 -h max 0-2 avg
24-h avg 1 -5 avg
Evening %
predicted FEV,
Morning FEV, (ml)
Change in %
predicted FEV,



-0.47 (-1 .9, 0.95)
53 (-2.4, 108)
-0.41 (-1.0,0.21)
  CS = corticosteroid, URI = Upper respiratory infection.
  "Effect estimates are standardized to a 40-ppb increase for 30-min or
24-h avg 03.
  °Results not presented in Figure 6-6 because a different form of FEV,
provided to calculate percent change in lung function.
 1-h max 03, a 30-ppb increase for 8-h max 03, and a 20-ppb increase for

 with a different scale was examined or because sufficient data were not
Draft - Do Not Cite or Quote
6-36
September 2011

-------
Study            Parameter

Gielenetal. (1997)    PEF

Mortimeretal. (2002)   PEF
Mortimeretal. (2000)   PEF
Thurstonetal. (1997)   PEF

Romieuetal. (2004)    FEF26.76%




Romieu etal. (1996)    Evening PEF


Romieu etal. (1997)    Evening PEF
O3Lag  Subgroup

2

1-5avg  All subjects
      Normal BW
      LowBW
      No asthma medication
      CSuser

0

1     Placebo, GSTM1 null
      Placebo, GSTM1 positive
      Antioxidant, GSTM1 null
      Antioxidant, GSTM1 positive
                                              -10
                                                         -6
                                                               -4
                                                Percent change in lung function parameter per standardized
                                                           increment in O3 (95% Cl)
Figure 6-7     Percent change in PEF or FEF25-75% in association with ambient
               ozone exposures among children with asthma. Results generally
               are presented in order of increasing mean ambient ozone
               concentration. BW = birth weight, CS = Corticosteroid. Effect
               estimates are from single pollutant models and are standardized to
               a 40-ppb increase for 1-h max ozone exposures and  a 30-ppb
               increase for 8-h max or 8-h avg ozone exposures.
Draft - Do Not Cite or Quote
                6-37
September 2011

-------
Table 6-8
Study
Gielenetal. (1997)
Mortimer etal. (2002)
Mortimer etal. (2000)
Thurston et al. (1997)
Romieu et al. (2004a)
Romieu et al. (1996)
Romieu et al. (1997)
Studies not included
Jalaludin et al. (2000)
Wiwatanadate and
Trakultivakorn (2010)
O'Connor etal.
(2008)
Additional characteristics and quantitative data for studies
represented in Figure 6-7
Location/ Population
Amsterdam, Netherlands
Children w/asthma
8 U.S. communities
Children w/asthma
8 U.S. communities
Children w/asthma
CT River Valley, CT
Children w/asthma
Mexico City, Mexico
Children w/asthma
Northern Mexico City, Mexico
Children w/asthma
Southern Mexico City, Mexico
Children w/asthma
in Figure 6-7b
Sydney, Australia
Children w/asthma
Chiang Mai, Thailand
Children w/asthma
7 U.S. communities
Children w/asthma
03
Averaging
Time
8-h max
8-h avg
(10:00a.m.-
6:00p.m.)
8-h avg
(10:00a.m.-
6:00p.m.)
1 -h avg
1-h max
1-hmax
1-h max

24-h avg
24-h avg
24-h avg
03
Lag
2
1-5
avg
1-5
avg
0
1
0
2
0
2

0
0
5
1-5
avg
Parameter Subgroup
PEF
PEF All subjects
PEF Normal BW
LowBW
No medication
CS user
I ntraday change
PEF
FEF2™ Placebo, GSTM1 null
Placebo, GSTM1 positive
Antioxidant, GSTM1 null
Antioxidant, GSTM1 positive
Evening PEF
Evening PEF

% variability Wheeze, no asthma
PEF Asthma, no AHR
Asthma, with AHR
Daily avg PEF
(L/min)
Change in %
predicted PEF
Standardized
percent change
(95% Cl)a
-1.3 (-2.6, -0.10)
-1.2 (-2.1, -0.26)
-0.60 (-1 .6, 0.39)
-3.6 (-5.2, -2.0)
-1.1 (-3.0,0.84)
-1.2 (-2.5, 0.11)
-2.8 (-4.9, -0.59)
-2.3 (-4.2, -0.44)
-0.48 (-1.7, 0.74)
-0.1 6 (-1.8, 1.6)
0.24 (-1.3, 1.8)
-0.1 7 (-0.79, 0.46)
-0.55 (-1.3, 0.1 9)
-0.52 (-1.0, -0.007)
-0.06 (-0.70, 0.58)

3.8 (0.25, 7.38)°
-0.71 (-2.6, 1 .2)°
-5.2 (-8.3, -2.2)°
1 .0 (-1 .6, 3.6)
-2.6 (-5.2, 0)
-0.22 (-0.86, 0.43)
       BW = birth weight, CS = corticosteroid, AHR = Airway hyperresponsiveness.
       "Effect estimates are standardized to a 40-ppb increase for 1-h max 03, a 30-ppb increase for 8-h max or 8-h avg 03, and a 20-ppb increase for
     24-h avg 03.
       bResults are not presented in Figure 6-7 because a different form of PEF with a different scale was examined or because sufficient data were not
     provided to calculate percent change in lung function.
       0 Outcome defined as the percent deviation from individual mean PEF during the study period. Group-stratified effect estimates were provided
     only for models that included PM10 and N02.
1                     The most geographically representative data were provided by the large, multi-U.S. city
2                     National Cooperative Inner City Asthma Study (NCICAS) (Mortimer et al., 2002;
3                     Mortimer et al.. 2000) and Inner-City Asthma Study (ICAS) (O'Connor et al.. 2008V
4                     Although the two studies differed in the cities, seasons, racial  distribution of subjects, and
5                     lung function indices  examined, results were fairly similar. In ICAS, which included
6                     children with asthma and atopy (i.e., allergic sensitization) and year-round examinations
7                     of lung  function, a 20-ppb  increase in the lag 1-5 average of 24-h avg O3 was associated
8                     with a 0.41-point decrease in percent predicted FEVj (95% CI: -1.0, 0.21) and a 0.22-
9                     point decrease in percent predicted PEF (95% CI: -0.86, 0.43).
      Draft - Do Not Cite or Quote
6-38
September 2011

-------
 1                   Lag 1-5 avg O3 (8-h avg, 10:00 a.m.-6:00 p.m.) also was associated with declines in PEF
 2                   in NCICAS, which included different U.S. cities, summer-only measurements, larger
 3                   proportions of Black and Hispanic children, and fewer subjects with atopy (79%)
 4                   (Mortimer et al.. 2002). NCICAS additionally identified groups potentially at increased
 5                   risk of O3-associated decrements in PEF. Larger effects were estimated in males, children
 6                   of Hispanic ethnicity, children living in crowded housing, and as indicated in Figure  6-7
 7                   and Table 6-8, children with low birth weight (Mortimer et al., 2000). Somewhat
 8                   paradoxically, O3 was associated with a larger decrease in PEF among subjects taking
 9                   cromolyn, medication typically used to treat asthma due to allergy, but a smaller decrease
10                   among subjects with positive atopy (as determined by skin prick test). Similar to
11                   observations from studies of populations with increased outdoor exposures, Mortimer et
12                   al. (2002) found that associations persisted at lower ambient O3 concentrations. At
13                   concentrations below 80 ppb, a 30-ppb increase in lag  1-5 of 8-h avg O3 was associated
14                   with a 1.4% decrease (95% CI: -2.6, -0.21) in PEF, which was similar to the effect
15                   estimated for the full range of O3 concentrations (Figure 6-7 and Table 6-8). In a study of
16                   children with asthma in the Netherlands, Gielen et al. (1997) estimated similar effects for
17                   the full range of 8-h max O3 concentrations and concentrations below 51 ppb.

18                   The results from studies of children with asthma indicated that factors in addition to
19                   asthma influenced associations between ambient O3 exposure and changes in lung
20                   function. In comparisons between children with and without asthma, Hoppe et al. (2003)
21                   and Jalaludin et al. (2000) generally found larger O3-associated lung function decrements
22                   in children with asthma; whereas Raizenne et al. (1987) did not consistently demonstrate
23                   differences between campers with and without asthma. In their study of children in
24                   Mexico City, Barraza-Villarreal et al. (2008) estimated larger O3-associated decreases in
25                   children without asthma; however, 72% of these children had atopy. These findings
26                   indicated that in addition to asthma, atopy, a condition also characterized by airway
27                   inflammation and  similar respiratory symptoms, may increase the risk for O3-associated
28                   respiratory effects.

29                   As indicated in Figures 6-6 and 6-7 and Tables 6-7 and 6-8, in most studies of children
30                   with asthma, standardized increments in ambient O3 exposure1 were associated with
31                   decreases in lung function that ranged from less than 1% to 2%. Larger magnitudes of
32                   decreases (3-8% per standardized increments in O3) were found in children with asthma
33                   who also were using CS, had a concurrent upper respiratory infection (URI), were
34                   GSTM1 null, had low birth weight, or had increased outdoor exposure (Romieu  et al.,
35                   2006; Lewis etal.. 2005; Romieu et al.. 2004a: Jalaludin et al.. 2000) than among
36                   children with asthma overall (Barraza-Villarreal et al..  2008; Lewis etal.. 2005;  Delfino
        1 Effect estimates were standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max, and 24-h avg O3.
      Draft - Do Not Cite or Quote                       6-39                                September 2011

-------
 1                  et al.. 2004; Romieu et al.. 2002). For example, Jalaludin et al. (2000) estimated a -5.2%
 2                  deviation from mean FEVi per a 20-ppb increase in 24-h avg O3 among children with
 3                  asthma and airway hyperresponsiveness and a much smaller -0.71% deviation among
 4                  children with asthma without airway hyperresponsiveness. In a group of 86 children with
 5                  asthma in Detroit, MI, Lewis et al. (2005) also reported that associations between O3
 6                  exposure and lung function decrements were confined largely to children with asthma
 7                  who used CS or had a concurrent URL These two groups were observed to have the
 8                  largest O3-associated decrements in lung function among all studies of children with
 9                  asthma. A 30-ppb increase in 8-h max ambient O3 exposure was associated with a 8.0%
10                  decrease in the mean of lowest daily FEVi among CS users and a 5.4% decrease among
11                  subjects reporting concurrent URI (Lewis et al.. 2005) (Figure 6-6 and Table 6-7).

12                  Heterogeneity in lung  function responses to ambient O3 exposure also has been
13                  demonstrated as inter-individual variability in the magnitude of O3-associated changes in
14                  lung function. Mortimer et al. (2002) found that for a 30-ppb increase in lag 1-5 avg of 8-
15                  h avg O3, there was a 30% (95% CI: 4, 61) increase in the incidence of a greater than
16                  10% decline in PEF. Likewise, Hoppe et al. (2003) found that while the percentages of
17                  change in individual lung function parameters were variable and small, 47% of children
18                  with asthma in their study experienced greater than  10% decline in FEVi, FVC, or PEF
19                  or 20% increase in airway resistance on days with 30-min (1:00 p.m.-4:00 p.m.) max
20                  ambient O3 concentrations greater than 50 ppb relative to days with less than 40 ppb O3.
21                  In addition to finding groups of children with asthma with increased sensitivity to O3
22                  exposure, epidemiologic studies have indicated that the decreases in lung function
23                  observed in association with increases in ambient O3 exposure may be clinically
24                  significant  by finding that the same or similar lag of O3 exposure was associated with
25                  decrements in lung function and increases in concurrently assessed respiratory symptoms
26                  (Just et al.. 2002: Mortimer etal.. 2002: Gielenetal.. 1997: Romieu etal..  1997:
27                  Thurston et al.. 1997; Romieu et al.. 1996) (see Figure 6-12 and Table 6-19 for symptom
28                  results).


                    Effect modification by corticosteroid use

29                  In controlled human exposure studies, CS treatment of subjects with asthma generally has
30                  not prevented O3-induced FEVi decrements (Section 6.2.1.1). In epidemiologic studies
31                  reviewed in the 2006 O3 AQCD, evidence was equivocal, as use of inhaled CS showed
32                  both protective (Delfino et al.. 2002; Mortimer et al.. 2000) and exacerbating (Gent et al..
33                  2003) effects on respiratory endpoints. Among recent studies, evidence for effect
34                  modification of lung function responses by CS use also was mixed. In Lewis et al.
35                  (2005). analyses of interactions between O3 exposure and CS use indicated stronger
36                  associations among CS users than among  CS nonusers (quantitative results not reported

      Draft - Do Not Cite or Quote                      6-40                                 September 2011

-------
 1                   for CS nonusers). Among the 11 (12.8%) CS users, a 30-ppb increase in lag 2 of 8-h max
 2                   O3 was associated with an 8.0% decrease (95% CI: -13.5, -2.1) in lowest daily FEVi and
 3                   a 6.7% increase (95% CI: 0.60, 13.2) in diurnal FEVi variability (indicating a decrease
 4                   from morning to evening). Other lags (1 or 3-5 avg) or averaging times (24-h avg) of
 5                   exposure were estimated to have less impact. In contrast to Lewis et al. (2005),
 6                   Hernandez-Cadena et al. (2009) observed greater O3-related decrements in FEVi among
 7                   the 60 CS nonusers than among the 25  CS users. In two winter-only studies,
 8                   consideration of CS use did not largely influence associations between ambient O3 and
 9                   lung function parameters (Liu et al.. 2009a; Rabinovitch et al.. 2004).

10                   Although studies varied in populations and season examined, the inconsistency in effect
11                   modification by CS use may  be explained, at least in part, by differences in the severity
12                   of asthma among CS users and the definition of CS use. Hernandez-Cadena et al. (2009)
13                   did not define CS use; however, the group of CS nonusers included both children with
14                   intermittent and persistent asthma. In Lewis et al. (2005). most children with moderate to
15                   severe asthma (91%) were included in the group of CS users (use for at least 50% of
16                   study days); however, these subjects had a higher percent predicted FEVi • Liu et al.
17                   (2009a) did not provide information on asthma severity; however, they defined CS use
18                   more stringently as daily use. Differences in asthma severity and definition of CS use
19                   may explain why both CS use and nonuse could serve as indicators of severe or
20                   uncontrolled asthma across studies. Additionally, investigators did not assess adherence
21                   to reported CS regimen, and  misclassification of CS use may bias findings.


                     Effect modification by  antioxidant capacity

22                   Ozone is a powerful oxidant  whose secondary oxidation products are recognized to
23                   initiate the key modes of action, including the activation of neural reflexes that mediate
24                   decreases in lung function (Section 5.3.2). Additionally, O3 exposure of humans and
25                   animals induces changes in the levels of antioxidants in the ELF (Section 5.3.3). These
26                   observations support the biological  plausibility for diminished antioxidant capacity
27                   increasing the risk of O3-associated respiratory effects and  augmented antioxidant
28                   capacity decreasing risk. Controlled human exposure studies have demonstrated
29                   protective effects of a-tocopherol (vitamin E) and  ascorbate (vitamin C) supplementation
30                   on O3-induced lung function  decrements  (Section 6.2.1.1),  and epidemiologic studies of
31                   children with asthma conducted in Mexico City have had similar findings. Particularly
32                   among children with moderate to severe asthma, ambient O3 exposure was associated
33                   with a smaller decrease in FEVi in the group supplemented with vitamin C and E as
34                   compared with the placebo group (Romieu et  al.. 2002) (Figure 6-6 and Table 6-7).
35                   Similarly, Romieu et al. (2009) observed protective effect for diets high in vitamins C
36                   and E as well as omega-3 fatty acids. Subjects were assigned to a fruits and vegetables

      Draft - Do Not Cite or Quote                      6-41                                 September 2011

-------
 1                   index (FVI) that characterized consumption of vitamins C and E and a Mediterranean diet
 2                   index (MDI) that additionally represented the intake of omega-3 fatty acids, which have
 3                   anti-inflammatory effects. At lag 0-4 avg of 8-h max O3 concentrations > 38 ppb, a 1-unit
 4                   increase in FVI was associated with a 137 ml (95% CI: 8, 266) increase in FEVi. This
 5                   protective effect of FVI was diminished at O3 concentrations < 25 ppb (65 ml [95% CI: -
 6                   70, 200] increase in FEVi per 1-unit increase in FVI).  Similar results were obtained for
 7                   MDI.

 8                   Antioxidant capacity also can be characterized by variants in genes encoding xenobiotic
 9                   metabolizing enzymes with different enzymatic activities. Ambient O3 exposure has been
10                   associated with greater decreases in lung function among children with asthma with the
11                   GSTM1 null genotype, which is associated with lack of oxidant metaboilizing activity
12                   (Romieu et al.. 2004a). The difference in response between GSTM1 null and positive
13                   subjects was minimal in children supplemented with antioxidant vitamins (Figure 6-7 and
14                   Table 6-8). Although these findings are biologically plausible given the well-
15                   characterized evidence for O3 effects mediated by secondary oxidation products, it is
16                   important to note that a larger body of controlled human exposure studies has not
17                   consistently found larger O3-induced lung function decrements in GSTM1 null subjects
18                   (Section 6.2.1.1). Effect modification by the GSTP1 variant is less clear. Romieu et al.
19                   (2006) observed larger O3-associated decreases  in FEVi in children with asthma with the
20                   GSTP1 lie/lie or Ile/Val variant, both of which are associated with normal oxidative
21                   metabolism activity (Figure 6-6 and Table 6-7). Also unexpectedly,  O3  exposure was
22                   associated with an increase in FEVi among children with the GSTP1 Val/Val variant,
23                   which is associated with reduced oxidative  metabolism. Rather than reflecting effect
24                   modification by the GSTP1 variant, these results may reflect effect modification by
25                   asthma severity, as 77% of subjects with the GSTP1 lie/lie genotype had moderate to
26                   severe asthma. Supporting evidence is provided by an  earlier analysis of the same cohort,
27                   in which the effect of antioxidant supplementation was demonstrated more strongly in the
28                   smaller group of children with moderate to  severe asthma than among all subjects with
29                   asthma (Romieu et al.,  2002).


                     Adults with  Respiratory Disease

30                   Relative to studies in children with asthma, studies of adults with asthma or COPD have
31                   been limited in number. Characteristics and ambient O3 concentration data from these
32                   studies are presented in Table 6-9. Studies that included both children and adults with
33                   asthma did not consistently demonstrate associations between ambient O3 exposure and
34                   decrements in lung function (Ross et al., 2002; Delfino et al., 1997). Ross et al. (2002)
35                   found that a 20-ppb increase in lag 0 of 24-h avg O3 was associated with a 2.6 L/min
36                   decrease (95% CI: -4.3, -0.90) in evening PEF among  subjects ages  5-49 years. This

      Draft - Do Not Cite or Quote                       6-42                                September 2011

-------
 1                   decrement may have been indicative of a clinically significant effect, as lag 0 O3
 2                   exposure also was associated with an increase in symptom score. In another panel study,
 3                   neither ambient nor personal O3 12-h avg exposure was reported to be associated with a
 4                   decrease in lung function among subjects ages 9-46 years (Delfino et al.. 1997).

 5                   Comparisons of adults with and without asthma did not conclusively demonstrate that
 6                   adults with asthma were at increased risk of O3-associated respiratory effects. In the
 7                   recent panel  study of 16- to 27-year-old lifeguards in Galveston, TX, a larger O3-
 8                   associated decrement in FEVi/FVC was found among the 16 lifeguards with asthma (-
 9                   1.6% [95% CI: -2.8, -0.4] per 40 ppb increase in 1-h max O3) than among the 126
10                   lifeguards without asthma (-0.40% [95% CI:  -0.80, 0] per 40 ppb increase in 1-h max O3)
11                   (Brooks. 2010). In the  studies of day-hikers, Korrick et al. (1998) found that the O3-
12                   associated lung function decrements observed among all hikers were driven by
13                   associations  observed in hikers with history of asthma or wheeze (-4.4% [95% CI: -7.5, -
14                   1.2] in FEVi per 30-ppb increase in 2-9 hr avg O3). In contrast, Girardot et al. (2006) did
15                   not find ambient O3 exposure to be consistently associated with decrements in lung
16                   function in subjects with or without respiratory disease history. In another cross-sectional
17                   study of 38 adults with asthma and 13 adults without asthma, atopy was observed to be a
18                   stronger susceptibility  factor than was asthma (Khatri etal. 2009). Investigators reported
19                   a larger decrease in percent predicted FEVi/FVC per 30-ppb increase in lag 2 of 8-h max
20                   O3 among the 38 subjects with atopy (with or without asthma) (-12 points [95% CI: -21, -
21                   3]) than among subjects with asthma (-4.7 points [95% CI: -11, 2.3]). Additionally,
22                   among adults with asthma, O3 was  associated with an increase in FEVi • Based on
23                   correlations observed between decreases in lung function and decreases in quality of life
24                   scores, investigators inferred the O3-associated decreases in lung function to be clinically
25                   significant. They suggested that atopy may influence responses to ambient O3 exposure
26                   because during the summer, high ambient O3 concentrations may increase the
27                   allergenicity of pollens.

28                   O3 was not found to have a strong effect on the lung function of adults with asthma in
29                   panel studies conducted in Europe and Asia during low ambient O3 periods
30                   (Wiwatanadate and Liwsrisakun. 2011; Lagorio et al., 2006; Park et al., 2005a). including
31                   one study conducted in Korea during a period of dust storms (Park etal.. 2005a). In these
32                   studies that examined multiple lags of O3 exposure, O3 generally was associated with
33                   increases in lung function.

34                   Controlled human exposure studies demonstrate robust O3-induced spirometric responses
35                   in children and young adults but diminished, statistically nonsignificant responses in
36                   older adults, both healthy and with COPD (Section 6.2.1.1). Similarly, in a recent
37                   epidemiologic study that followed 94 adults with COPD (ages 40-83 years) daily over a
      Draft - Do Not Cite or Quote                       6-43                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
2-year period, an increase in ambient O3 exposure was not associated consistently with
decreases PEF, FEVi, and FVC (Peacock et al.. 2011). For example, in an analysis
restricted to the summer of 1996, a 30-ppb increase in 8-h max O3 was associated with a
1.7 L/min decrease (95% CI: -3.1, -0.39) in PEF. However, during the summer of 1997,
O3 was found to  have little effect on PEF (-0.21 L/min [95% CI:  -2.4, 2.0] per 30-ppb
increase in  8-h max O3). Further, in this study, an increase in ambient O3 exposure was
associated with a lower odds of a large PEF decrement (OR for a greater than 20% drop
from an individual's median value: 0.89 [95% CI: 0.72,  1.10] per 30-ppb increase in lag 1
of 8-h max O3) and was not consistently associated with increases in respiratory
symptoms (Peacock et al.. 2011). Ozone exposure also was not consistently associated
with decreases in lung function in a smaller panel study of 11 adults with COPD (mean
age 67 years) (Lagorio et al.. 2006). Together, these finding do not provide strong
evidence that increases in O3 exposure are associated with lung function decrements in
adults with COPD.
Table 6-9
Study
Korricket al.
(1998)
Khatrietal. (2009)
Ross et al. (2002)
Thaller etal.
(2008)
Delfino et al.
(1997)
Lagorio etal.
(2006)
Peacock etal.
(2011)
Wiwatanadate et
al. (2011)
Park etal. (2005a)
Mean and upper percentile concentrations of ozone in
epidemiologic studies examining lung function in adults with
respiratory disease
Location
Mt.
Washington,
NH
Atlanta, GA
East Moline, IL
Galveston, TX
Alpine, CA
Rome, Italy
London,
England
Chiang Mai,
Thailand
Incheon, Korea
Years/Season
1991,1992
Warm season
2003, 2005, 2006
Warm season
April-October 1994
2002-2004
Warm season
1994
Warm season
1999
Spring and winter
1995-1997
All-year
August 2005 -
June 2006
March-June 2002
O3 Averaging
Time
Hike-time avg
(2-1 2 h)
8-h max
8-h avg
1-hmax
12-havg
personal
(8:00 a.m.-8:00
p.m.)
24-h avg
8-h max
24-h avg
24-h avg
Mean/Median Upper Percentile
Concentration (ppb) Concentrations (ppb)
40
59 (median)8
41.5
35 (median)
18
Spring: 36.2"
Winter: 8.0b
15.5
17.5
Dust event days: 23.6
Control days: 25.1
Max: 74
75tn: 73a
Max: 78.3
Max: 118
90th: 38
Max: 80
Overall max: 48.6"
Autumn/Winter Max: 32
Spring/Summer Max: 74
90th: 26.82
Max: 34.65
NR
        NR = Not reported, Max = Maximum.
      'Individual-level exposure estimates were derived based on time spent in the vicinity of various 03 monitors.
      bConcentrations converted from ug/m3 to ppb using the conversion factor of 0.51 assuming standard temperature (25°C) and pressure (1 atm).
15
16
Populations Not Restricted to Individuals with Asthma

Several studies have examined associations between ambient O3 exposure and lung
function in children; however, a limited number of studies have examined other
      Draft - Do Not Cite or Quote
                                6-44
September 2011

-------
 1
 2
 3
populations not restricted to individuals with asthma or other healthy populations.
Characteristics and ambient O3 concentration data from studies not restricted to
individuals with asthma are presented in Table 6-10.
Table 6-10
Study
Alexeefetal. (2007)
Alexeefetal. (2008)
Naeheretal. (1999)
Avoletal. (1998a)
Linn et al. (1996)
Gold et al. (1999)
Scarlett etal. (1996)
Ward etal. (2002)
Ulmer etal. (1997)
Hoppe et al. (2003)
Neuberger etal. (2004)
Steinvil etal. (2009)
Chen etal. (1999)
Son et al. (2010)
Mean and upper percentile concentrations of ozone in
epidemiologic studies examining lung function in populations not
restricted to individuals with asthma
Location
Greater Boston, MA
Vinton, VA
6 southern CA
communities
Rubidoux, Upland,
Torre nee, CA
Mexico City, Mexico
Surrey, England
Birmingham and
Sandwell, England
Freudenstadt and
Villingen, Germany
Munich, Germany
Vienna, Austria
Tel Aviv, Israel
3 Taiwan
communities
Ulsan, Korea
Years/Season
1995-2005
All-year
1995-1996
Warm season
1994
Spring and summer
1992-1993,1993-
1994
Fall and spring
1991
Winter, spring, fall
1994
Warm season
1997
Winter and summer
1994
March-October
1992-1995
Warm season
June-October 1999,
January-April 2000
2002-2007
All-year
1995-1996
May-January
2003-2007
All-year
Metric
24-h avg
8-h max
24-h avg personal
24-h avg
24-h avg
8-h max
24-h avg
30-min max
30-min max (1 :00 p.m.-
4:00 p.m.)
NR
8-h avg
(10:00 a.m. -6:00 p.m.)
1-hmax
8-h max
Mean/Median
Concentration (ppb)
24.4a
34.87
NR
23
52.0
50. 7b
Winter median: 13.0
Summer median: 22.0
Freudenstadt median: 50.6
Villingen median: 32.1
High 03 days: 70.4
Control 03 days: 29.8
NR
41.1
NR
35.86
Upper Percentile
Concentrations (ppb)
NR
Max: 56.63
NR
Max: 53
Max: 103
Max: 128b
Winter Max: 33
Summer Max: 41
Freudenstadt 95th:
Villingen 95th: 70.1







89.7
Max (high 03 days): 99
Max (control 03 days): 39
NR
75th: 48.7
Max: 72.8
Max: 110.3b
Max: 59.53




        NR = Not Reported, Max = Maximum.
        'Measured at central monitoring sites established by investigators. Concentations were averaged across all monitors.
        bMeasured at subjects' schools where lung function measurements were performed.

                          Children
 4                    The 2006 O3 AQCD identified children as a potentially at-risk population based on
 5                    consistent evidence of association between ambient O3 exposure and decrements in
 6                    and PEF (U.S. EPA. 2006b) (Figure 6-8 and Table 6-11). No new studies in children
 7                    without asthma are available to compare with previous findings. Hoppe et al.  (2003) O3
 8                    exposure to be associated with decreses in healthy children in Munich, Germany (Figure
 9                    6-8 and Table 6-11). In another panel study of healthy children in Vienna, Austria, O3
10                    was not associated with decrements in total lung capacity (Neuberger et al.. 2004). Most
      Draft - Do Not Cite or Quote
                                 6-45
September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
of the studies in children did not exclusively examine healthy children. However, several
studies of children that included small proportions (4- 10%) of children with history of
respiratory disease or symptoms and found associations between O3 exposure and
decrements in lung function (Chen et al.. 1999; Ulmeretal.. 1997; Scarlett et al.. 1996).
Based on interactions between O3 exposure and asthma/wheeze history, Avol et al.
(1998a) and Ward et al. (2002) did not find lung function responses to ambient O3
exposure to differ between children with history of asthma or wheeze and healthy
children. Combined, these lines of evidence indicate that the associations observed
between ambient O3 exposure and decreases in lung function in children are not driven by
effects in children with asthma or respiratory symptoms, and that healthy children also
may represent a population at increased risk of O3-associated respiratory effects.
                         Parameter
Study

Linnetal. (1996)


Hoppeetal. (2003)


Scarlettetal. (1996)

Chenetal.(1999)     FEV1
                        O3 Lag
                         Intraday change F£V!   0
                         Intraday change FVC   0
                         FEV.,
                         FVC
                        0
                                             1
       Avoletal. (1998)a     Intraday change FEV.,   O(personal)
                         Intraday change FVC
                                                         -10
                                                  -6
                                                                              -2
0
                                                     Percentchangein lung function parameter per standardized
                                                                    increment in O3 (95% Cl)

        Results generally are presented in order of increasing mean ambient ozone concentration.
        aThe 95% Cl was constructed using a standard error that was estimated from the p-value. Effect estimates are from single-
      pollutant models and are standardized to a 40-, 30-, and 20-ppb increase for a 1-h (or 30-min) max, 8-h max, and 24-h avg ozone
      exposures, respectively.

      Figure 6-8     Percent change in lung function in association with ambient ozone
                       exposures in studies not restricted to children with asthma.
      Draft - Do Not Cite or Quote
                                6-46
                                                                                  September 2011

-------
Table 6-11
Study
Linn et al. (1996)
Hoppe et al. (2003)
Scarlett etal. (1996)
Chen etal. (1999)
Avoletal. (1998a)
Studies of children
Ulmer etal. (1997)
Ward etal. (2002)
Gold et al. (1999)
Additional characteristics and quantitative data for studies
represented in Figure 6-8 and results from other studies in children
Location/
Population
3 southern CA
communities
Children
Munich, Germany
Children
Surrey, England
Children
3 Taiwan communities
Children
3 southern CA
communities
Children
O3Lag
0
0
1
1
0 (personal)
Os Averaging
Time
1 -h avg
30-min max (1 :00
p.m.-4:00 p.m.)
8-h max
1-hmax
24-h avg
Parameter
I ntraday change FEVi
Intraday change FVC
FEV,
FVC
FEV,
FEV,
Intraday change FEV,
Intraday change FVC
Effect Estimate (95% Cl)a
-0.56 (-0.99, -0.1 2)
-0.21 (-0.62, 0.20)
-1 .4 (-4.3, 1 .4)
-2.5 (-4.9, -0.10)
-0.04 (-0.32, 0.23)
-1.5 (-2.8, -0.12)
-1 .4 (-3.8, 0.90)b
-2.0 (-4.0, 0.01)b
not included in Figure 6-8°
Freudenstadt and
Villingen, Germany
Children
Birmingham and
Sandwell, England
Children
Mexico City, Mexico
Children
1
0
0-6 avg
1
1-10 avg
1/2-hmax
24-h avg
24-h avg
FEV, (ml)
PEF (L/min)
Intraday change PEF (%
change)
-5.9 (-10.4,1.3)"
-3.2 (-8.3, 2.0)d
-11.1 (-22.0, -0.18)d
-0.54 (-1.1, 0.05)
        aEffect estimates are standardized to a 40-, 30-, and 20-ppb increase for 1-h (or 30-min) max, 8-h max, and 24-h avg O3,
      respectively.
      bThe 95% Cl was constructed using a standard error that was estimated from the p-value.
        °Results are not presented in Figure 6-8 because sufficient data were not provided to calculate percent change in lung function or
      PEF was analyzed.dEffect estimates are from analyses restricted to summer months.
 1                    Among the studies of children, the magnitudes of decrease in lung function per
 2                    standardized increment in ambient O3 exposure1 ranged from less than 1 to 4%, a range
 3                    similar to that estimated in  children with asthma.  However, in contrast with studies of
 4                    children with asthma, studies of children in the general population did not consistently
 5                    find that O3-associated decreases in lung function were accompanied by increases in
 6                    respiratory symptoms. Gold et al. (1999) found that lag  1 of O3 exposure was associated
 7                    with both decreases in PEF and increases in phlegm; however, the increase in phlegm
 8                    was associated with O3 exposure lagged one day whereas the PEF decrement was driven
 9                    by exposures lagged 2 to 4  days. Ozone was weakly associated with cough and shortness
10                    of breath among children in England (Ward et al.. 2002). and O3 was associated with a
11                    decrease in respiratory symptom score among children in California (Linn etal.. 1996).
12                    These findings indicate that while the magnitudes of O3-associated decrease in lung
13                    function may be similar in children with and without asthma, because of the higher
14                    overall lung function in healthy children, the decrements may not be large enough to be
15                    clinically significant in healthy children.
        1 Effect estimates were standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max, and 24-h avg O3.
      Draft - Do Not Cite or Quote
6-47
September 2011

-------
                        Adults
 1                   In the small body of studies conducted in adults, O3 has been associated with decrements
 2                   in lung function in both healthy adults and those with comorbid factors (Table 6-12). In a
 3                   cohort of mostly healthy women, ages 19-43 years, followed for one summer season,
 4                   Naeher et al. (1999) observed associations between 8-h max ambient O3 exposure and
 5                   decreases in PEF. In a large cross-sectional study of 2,380 healthy adults (75th percentile
 6                   of age:  52 years) in Tel Aviv, Israel, across several lags of exposure (single day lags 0-7
 7                   and 0-6 avg), O3 was associated mostly with increases in FEVi, FVC, and FEVi/FVC
 8                   (Steinvil et al.. 2009). Another large cross-sectional study was conducted in 2,102
 9                   children and adults (mean age: 45 years) living near a petrochemical plant in Ulsan,
10                   Korea (Son et al.. 2010). Multiple O3 exposure metrics, including concentrations
11                   averaged across 13 city monitors, concentrations from the nearest monitor, inverse
12                   distance-weighted concentrations, and estimates from kriging, were associated with
13                   decrements in lung function; however, no particular metric consistently showed a larger
14                   effect across the various lags of O3 exposure examined. Lag 0-2 avg of 8-h max O3
15                   exposure was associated with the largest decrements in percent predicted FEVi (1.4-point
16                   decrease [95% CI: -2.7, -0.08] per 30-ppb increase in the 8-h max of lag 0-2 avg O3
17                   averaged across all monitors). Although the health status of subjects was not reported, the
18                   mean percent predicted FEVi in the study population was 82.85%, indicating a large
19                   proportion of subjects with underlying airway obstruction. Results from this study were
20                   not adjusted for meteorological factors and thus, confounding cannot be ruled out.

21                   As described in  Section 6.2.1.1, controlled human exposure studies have not consistently
22                   found O3-induced decreases in lung function in older adults. In an earlier study of adults
23                   ages  69-95 years, Hoppe et al. (2003) did not find ambient O3 exposure-associated
24                   decreases in lung function. However, recently, the Normative Aging Study found that
25                   ambient O3 exposure was associated with decrements in FEVi and FVC in a group of
26                   older men (Alexeeff et al.. 2008). This study in the Greater Boston area conducted
27                   spirometry once every 3 years for 10 years in 900 older men (mean [SD] age = 68.9 [7.2]
28                   years), most of whom were white and healthy. Among all subjects, several lags of 24-h
29                   avg O3 exposure (1- to 7-day avg) were associated with decreases in  FEVi (Alexeeff et
30                   al.. 2008). Additionally, larger effects were estimated in adults  with elevated BMI (> 30),
31                   airway hyperresponsiveness, and reduced activity in antioxidant enzymes (i.e., GSTP1
32                   He/Val or Val/Val variant) (Alexeeff etal.. 2008: Alexeeff etal..  2007) (Table 6-12).
33                   Larger O3-related decrements in FEVi and FVC also were observed in subjects with long
34                   GT dinucleotide repeats in the promoter region of the antioxidant enzyme heme
35                   oxygenase-1 (Alexeeff et al.. 2008). which has been associated with reduced inducibility
36                   (Hiltermann et al.. 1998). The largest O3-related percentages of decrease in lung function
37                   were observed in the group of men with airway hyperresponsiveness and elevated BMI  (-
      Draft - Do Not Cite or Quote                       6-48                                 September 2011

-------
 1
 2
 3
 4
 5
5.3% FEVj [95% CI: -8.3, -2.4] per 20-ppb increase in lag 0-1 avg of 24-h avg O3). In
this cohort, O3 also was associated with decreases in lung function in adults without
airway hyperresponsiveness and BMI < 30, indicating the effects of O3 on lung function
in older adults extends to healthy older adults. However, importantly, the findings may be
generalizable only to older white men.
Table 6-12 Associations between ambient ozone exposure and changes in
lung function in studies of adults
Study
Son et al.
(2010)
Steinviletal.
(2009)
Naeheretal.
(1999)
Hoppe et al.
(2003)
Alexeeffetal.
(2008)
Alexeefetal.
(2007)
Location/ Population
Ulsan, Korea
Children and adults, ages 7-
97 yr
Tel Aviv, Israel
Healthy adults, mean age 43
yr, 75llr%-ile: 52 yr
Vinton, VA
Healthy women, ages 19-43
yr
Munich, Germany
Older adults, ages 69-95 yr
Greater Boston, MA
Older adults, mean (SD) age:
68.8 yr (7.3)
Greater Boston, MA
Older adults, mean (SD) age:
68.8 yr (7.3)
O3 Lag
0-2 avg
0
0-6 avg
0
0-4 avg
0
1
0-1 avg
0-1 avg
?3 •
Averaging
Time
8-h max
8-h avg
(10:00a.m-
6:00 p.m.)
24-h avg
30-min max
(1:00p.m.-
4:00p.m.)
24-h avg
24-h avg
Parameter
Change in %
predicted FEV,
FEV, (ml)
Evening PEF
(L/min)
% change in
evening FEV,
% change in FEVi
% change in FEVi
Os Assessment
Method/Subgroup
All monitor avg
Nearest monitor
IDW
Kriging



GSTP1 lie/lie
GSTP1 NeA/al ValA/al
BMI < 30
BMI > 30
NoAHR
AHR
BMI > 30 and AHR
Effect Estimate
(95% Cl)a
-1.4 (-2.7, -0.08)
-0.76 (-1 .8, 0.25)
-1.1 (-2.2,0.05)
-1.4 (-2.6, -0.11)
40 (0, 80)
94(33,156)
-0.06 (-0.11,0)
-5.1 (-8.7, -1.5)
0.75 (-2. 1,3.7)
1 .2 (-1 .3, 3.6)
-1.0 (-2.2, 0.1 9)
-2.3 (-3.5, -1.0)
-1.5 (-2.5, -0.52)
-3.5 (-5.1, -1.9)
-1.7 (-2.7, -0.73)
-4.0 (-6.2, -1.8)
-5.3 (-8.2, -2.3)
       IDW = Inverse distance weighting, BMI = Body mass index, AHR = airway hyperresponsiveness.
      'Effect estimates are standardized to a 40-ppb increase for 30-min max 03, 30-ppb increase for 8-h max or 8-h avg 03, and 20-ppb increase for 24-
      h avg 03.
 6
 1
 8
 9
10
11
12
13
14
15
16
17
18
Confounding in epidemiologic studies of lung function

The 1996 O3 AQCD noted uncertainty regarding confounding by temperature and pollen
(U.S. EPA. 1996a); however, studies collectively do not provide strong evidence of
confounding by these factors. Most studies, whether they involved year-round or
summer-only examinations, included temperature in statistical analyses and found
associations between O3 exposure and decreases in lung function.  Across studies,
temperature has shown inconsistent associations with lung function, even among studies
conducted in the summer and in the same geographic region. For example, in studies of
children attending summer camps conducted in the Northeast U.S., temperature was
associated with an increase (Berry et al.. 1991) (Thurston et al.. 1997) and decrease
(Raizenne etal.,  1987) in lung function. In the reanalysis of six camp studies,
investigators did not include temperature in models because temperature within the
normal ambient range had not been shown to affect O3-induced lung function responses
in controlled human exposure studies (Kinney  et al.. 1996). In two summer camp studies
      Draft - Do Not Cite or Quote
                                6-49
September 2011

-------
 1                   conducted in the Northeast U.S., O3 was associated with decreases in lung function in
 2                   models without and with temperature (Thurston et al.. 1997; Spektor et al.. 1988a). In
 3                   both studies, temperature and O3 were measured on site of the camps. Spektor et al.
 4                   (1988a) estimated similar effects in a model with and without a temperature-humidity
 5                   index, and Thurston et al. (1997) found that compared with a univariate model,  O3 was
 6                   associated with a nearly 2-fold greater decrease in PEF when temperature was added to
 7                   the model.

 8                   Although evaluated in fewer studies, the evidence does not  indicate that associations
 9                   between ambient O3 exposure and lung function are confounded by pollen. Some camp
10                   studies found that pollen independently was not associated with lung function decrements
11                   (Thurston et al.. 1997; Avol et al.. 1990). A few studies of children with asthma with
12                   follow-up over multiple seasons found O3 to be associated with decrements in lung
13                   function in models that adjusted for pollen counts (Just et al.. 2002; Ross et al..  2002;
14                   Jalaludin et al.. 2000; Gielen et al., 1997). In these studies, large percentages of subjects
15                   had positive atopy (22-98%), with some studies examining  large percentages of subjects
16                   specifically with pollen allergy(Ross et al.. 2002; Gielen et  al.. 1997).

17                   A relatively larger number of studies provided information  on potential confounding by
18                   copollutants such as PM25, PM10, NO2, or SO2. In most cases, investigators indicated that
19                   associations between  O3  exposure and lung function were not driven by copollutant
20                   confounding; however, studies varied in how they considered confounding. Studies of
21                   subjects exercising outdoors indicated that ambient concentrations of copollutants such as
22                   NO2, sulfur dioxide, or acid aerosol were low and thus, not  likely to  confound the
23                   observed O3 effects (Hoppe et al.. 2003; Brunekreef etal.. 1994; Hoeketal.. 1993). In
24                   other studies of children with increased outdoor exposures,  O3 was consistently
25                   associated with decreases in lung function, whereas other pollutants  such as PM2 5,
26                   sulfate, and acid aerosol individually showed variable associations across studies
27                   (Thurston et al.. 1997; Castilleios et al..  1995; Berry etal.. 1991; Avol etal.. 1990;
28                   Spektor et al.. 1988a).

29                   Among studies that conducted copollutant modeling, associations between O3 exposure
30                   and lung function decrements were observed to be robust (Figure 6-9 and Table 6-13). In
31                   copollutant models, O3 effect estimates generally fell within the 95% CI of the single-
32                   pollutant model effect estimates. Whereas some studies used the same averaging time for
33                   all pollutants (Lewis etal.. 2005; Jalaludin et al..  2000). most examined 1-h max or 8-h
34                   max O3 exposures and 24-h avg copollutant exposures (Son et al.. 2010; Chen et al..
35                   1999; Romieuetal.. 1997; Romieuetal.. 1996). In a Philadelphia-area summer camp
36                   study, Neas et al. (1999)  was among the few studies to find that the effect of O3 was
      Draft - Do Not Cite or Quote                       6-50                                September 2011

-------
 1                   attenuated in a copollutant model. In a copollutant model with 24-h avg sulfate, the 12-h
 2                   avg O3 effect estimate was attenuated to near zero (Figure 6-9 and Table 6-13).

 3                   In studies with copollutant modeling, ambient O3 concentrations showed a wide range of
 4                   correlations with concentrations of copollutants (r=-0.31 to 0.74). Among children with
 5                   asthma in Sydney, Australia, Jalaludin et al. (2000) found low correlations of 24-h avg O3
 6                   with 24-h avg PM10 (r = 0.13) and NO2 (r = -0.31), and in two-pollutant models, PM10
 7                   and NO2 continued to be associated with increases in PEF, and O3 continued to be
 8                   associated with decreases in PEF. In a study of children with asthma in Detroit, MI, 24-h
 9                   avg O3 was moderately correlated with 24-h avg PM2 5 (Pearson r= 0.57) and 24-h avg
10                   PM10 (Pearson r=0.59) (Lewis et al., 2005). Inclusion of PM10 or PM25 in models resulted
11                   in larger changes in O3 effect estimates than those observed in other studies. As
12                   illustrated in Figure 6-9  and Table 6-13, the magnitude of change was not consistent
13                   between the two subgroups. Among subjects with a concurrent URI, O3-associated
14                   decreases in lowest daily FEVi were robust to the inclusion of PM10 or PM2 5. Among CS
15                   users, O3 was associated a much larger decrease  in FEVi when PMi0 was included in the
16                   model (Lewis et al.. 2005).

17                   Studies conducted in Mexico City found small changes in O3-associated lung function
18                   decrements in copollutant models, although different averaging times were used for
19                   different pollutants (Romieu et al.. 1997; Romieuetal.. 1996) (Figure 6-9 and Table 6-
20                   13). In these studies, O3 was moderately correlated with co-pollutants such as NO2 and
21                   PMio (range of Pearson r = 0.38-0.58). Studies conducted in Asia also found that
22                   associations between O3 and lung function were  robust to the inclusion of weakly- to
23                   moderately-correlated copollutants (Son et al.. 2010; Chenet al..  1999). Copollutant
24                   effect estimates generally were attenuated, indicating that O3 may confound the results of
25                   copollutants.

26                   In a summer camp study conducted in Connecticut, Thurston et al. (1997) found ambient
27                   concentrations of 1-h max O3 and 12-h avg sulfate to be highly correlated (r = 0.74),
28                   making it more difficult to separate their independent effects. With sulfate in the model,  a
29                   larger decrease in PEF was estimated for O3; however, the 95% CI was much wider
30                   (Figure 6-9 and Table 6-13). Investigators found that the association between sulfate and
31                   PEF was driven by one day when the ambient concentrations of both pollutants were at
32                   their peak. With the removeal of this influential day, the sulfate effect was attenuated,
33                   whereas O3 effects remained robust (Thurston et al.. 1997). Among children with asthma
34                   in Thailand, the O3-associated decrease in PEF was robust to the adjustment of SO2;
35                   however, different lags were examined for O3 (lag 5) and SO2 (lag 4) (Wiwatanadate and
36                   Trakultivakorn. 2010). Some studies did not provide quantitative  results but reported that
37                   O3 effects on lung function decrements remained statistically significant in models that
      Draft - Do Not Cite or Quote                       6-51                                 September 2011

-------
1
2
included copollutants such as PM10, NO2, sulfate, nitrate, or ammonium (Romieu et al.
1998a: Braueretal..  1996: Linnetal.. 1996: Spektor et al.. 1988b).
           Study

           PEF

           Jalaludin etal. (2000)
                          Population
                                          O3 exposure data O3 with copollutant
                                          24-h avg, Lag 0   None
                                          24-havg, Lag 0   24-h avg PM10
                                          24-h avg, Lag 0   24-h avg NO2
           Neasetal. (1999)=    Children attendingcamp 12-havg,Lag1   None
                                          1 2-h avg, Lag 1   24-h avg su Ifate
           Thurston eta,. (1997)
                                          ^ «9 °   None
                                          1-h max, Lag 0   12-h avg sulfate-4-
           Romieu etal. (1996)   Children with asthma   1-h max, Lag 0   None
                                          1-h max, Lag 0   24-h avg PM2 5

           Romieu etal. (1997)   Children with asthma   1-h max, Lag 0   None
                                          1-h max, Lag 0   24-h avg PM10
tant
-O
-O
C

— •-



)

-
                                                                -15   -13   -11    -9    -7    -5    -3-11     3
                                                                Percent change in PEF per standardized increase in O3 (95% Cl)
FEV,
Lewis etal. (2005)
Chen etal. (1999)
Children with asthrna 24_h avg] Lag 2
24-h avg, Lag 2
Children with asthma
withURI 24-h avg, Lag 2
24-h avg, Lag 2
24-h avg, Lag 2
Children 1-h max, Lag 1
1-h max, Lag 1



id.g . ,„

'4ha"i'rM"
24-h avg NO2 	 0 	



                                                                -15   -13   -11    -9    -7    -5    -3-11     3
                                                                Percent change in FEV, per standardized increase in O3 (95% Cl)

        Results are presented for PEF then FEV! and then in order of increasing mean ambient ozone concentration. "Information was not
      available to calculate 95% Cl of the copollutant model. CS = corticosteroid, URI = Upper respiratory infection. Effect estimates are
      standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 12-h avg, and 24-h avg ozone, respectively.  Black circles represent
      ozone effect estimates from single pollutant models, and open circles represent ozone effect estimates from copollutant models.
      Figure 6-9      Comparison of ozone-associated changes in lung function in
                         single- and copollutant models.
      Draft - Do Not Cite or Quote
                                     6-52
September 2011

-------
     Table 6-13     Additional characteristics and quantitative data for studies
                      presented in Figure 6-9
Study
Location/
Population
Os Exposure
Data
Parameter
Os-associated
Percent Change in
Single-Pollutant
Model (95% Cl)a
Os-associated Percent
Change in Copollutant
Model (95% Cl)a
PEF
Jalaludin et al.
(2000)
Neas et al. (1999)
Thurston et al.
(1997)
Romieu et al.
(1996)
Romieu et al.
(1997)
Sydney, Australia
Children with asthma or
wheeze
Philadelphia, PA
Children attending summer
camp
CT River Valley
Children with asthma attending
summer camp
Mexico City, Mexico
Children with asthma
Mexico City, Mexico
Children with asthma
24-h avg
LagO
12-havg
Lag1
1-hmax
LagO
1-hmax
LagO
1-h max
LagO
Intraday change
PEF
Morning PEF
Intraday change
PEF
Evening PEF
Evening PEF
-0.57 (-1.1, -0.06)
-0.94 (-2.0, 0.08)
-2.8 (-4.9, -0.59)
-0.55 (-1.3, 0.1 9)
-0.52 (-1.0, -0.01)
with 24-h avg PM10,
-0.57 (-1.1, -0.06)
with 24-h avg N02
-0.55 (-0.1, -0.04)
with 24-h avg sulfate
-0.02b
with 12-havg sulfate
-11. 8 (-31 .6, 8.1)
with 24-h avg PM2.5
-0.24 (-1.2, 0.68)
with 24-h avg PM10
-0.79 (-1.4, -0.16)
FEV,
Lewis etal.
(2005)
Chen etal. (1999)
Detroit, Ml
Children with asthma using
CS
Children with asthma with
URI
3 Taiwan communities
Children
24-h avg
Lag 2
1-hmax
Lag1
Lowest daily
FEV,
FEV,
0.29 (-4.2, 5.0)
-6.0 (-11. 2, -0.41)
-1.5 (-2.8, -0.12)
with 24-h avg PM2.5
-0.1 8 (-11. 0,11.9)
with 24-h avg PM10
-13.4 (-17.8, -8.8)
with 24-h avg PM2.5
-5.5 (-10.3, -0.42)
with 24-h avg PM10
-7.1 (-11.3, -2.8)
with 24-h avg N02
-2.0 (-3.5, 0.42)
Results not included In Figure 6-9
Wiwatanadate
and
Trakultivakorn
(2010)
Son et al. (2010)
Chiang Mai, Thailand
Children with asthma
Ulsan, Korea
Children and adults
24-h avg
Lag5
8-h max
Lag 0-2 avg
Evening PEF
(L/min)
Change in %
predicted FEVi
-2.6 (-5.2, 0)
-1.4 (-2.6, -0.11)
with Lag 4 S02
-3.2 (-6.2, -0.2)
with PM10
-1.8 (-3.4, -0.25)
                                        (kriging)
       CS = Corticosteroid, URI = Upper respiratory infection.
       aResults represent percent changes in lung function parameter per the following standardized increase in ambient O3
     concentration: 40 ppb for 1-h max O3, 30 ppb for 8-h max or 12-h avg O3, and 20 ppb for 24-h avg O3.
1                   Several studies examined multi-pollutant models that most often included O3, NO2, and
2                   either PM2 5 or PM10. Ozone exposure was associated with similar or larger magnitudes of
3                   decrease lung function in multi-pollutant models (O'Connor et al.. 2008; Thaller et al..
4                   2008; Chan and Wu. 2005;  Romieu et al.. 2002: Korrick et al.. 1998: Higgins et al..
5                   1990): however, the independent effects of O3 exposure are more difficult to assess in
6                   relation to incremental changes in more than one copollutant.
     Draft - Do Not Cite or Quote
6-53
September 2011

-------
                     Summary of Epidemiologic Studies of Lung Function

 1                   The cumulative body of epidemiologic evidence strongly supports associations between
 2                   ambient O3 exposure and decrements in lung function in children, particularly, those with
 3                   asthma. While little new research is available, previous AQCDs have presented
 4                   epidemiologic evidence of heightened effects in children and adults exercising or
 5                   working outdoors during periods of relatively low ambient O3 concentrations (Table 6-1).
 6                   These epidemiologic results are well-supported by observations from controlled human
 7                   exposure studies  in which exposures to lower O3 concentrations induce lung function
 8                   decrements when combined with exercise as compared with exposures during rest.

 9                   Recent epidemiologic investigation continued to focus on children with asthma, and most
10                   recent results in this  population indicated associations between O3 exposure and
11                   decrements in lung function (Figures 6-6 anf 6-7 and Tables 6-7 and 6-8). Based on a
12                   small number of within-study comparisons of groups with and without asthma, larger
13                   effects were not conclusively estimated for groups with asthma. It is important to note
14                   that most of these studies were not designed to assess between-group differences, and in
15                   some studies, the high prevalence of atopy may have contributed to larger associations in
16                   subjects without asthma (Khatri et al.. 2009; Barraza-Villarreal et al.. 2008). A large body
17                   of previous studies demonstrated associations in children. Whereas the 2006 O3 AQCD
18                   reported weak evidence, a new study indicates that O3 exposure may be associated with
19                   decrements in lung function in older adults.

20                   Across the diverse populations examined in epidemiologic studies, ambient O3 exposure
21                   was associated with  1-8% decreases in lung function per standardized increment in O3
22                   concentration1. Larger decreases (3-8%) usually were observed in children with asthma
23                   or older adults with CS use, concurrent URI, airway hyperresponsiveness, or reduced
24                   activity of antioxidant enzymes. These results indicate that common comorbid and
25                   genetic factors may increase the risk of O3-associated respiratory effects. High dietary
26                   antioxidant intake was found to attenuate O3-associated lung function decrements.  Each
27                   of these potential susceptibility or protective factors has been examined in one to two
28                   populations, and  further investigation in diverse populations is warranted. Heterogeneity
29                   in response also was demonstrated by observations that increases in ambient O3 exposure
30                   were associated with increased incidence of a greater than 10% decline in lung function
31                   in children with asthma (Hoppe et al.. 2003; Mortimer et al.. 2002). In considering the
32                   clinical significance  of more subtle health outcomes such as lung function changes, it is
33                   important to note that a small shift in the population mean likely will have a
34                   disproportionate effect in the extreme ends of the distribution of lung function where
35                   these small magnitudes  of decrease lead to clinically-significant airway resistance or
        1 Effect estimates were standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max, and 24-h avg O3.
      Draft - Do Not Cite or Quote                       6-54                                September 2011

-------
 1                   obstruction and where individuals likely have concurrent symptoms. Several
 2                   epidemiologic studies have demonstrated the clinical significance of O3-associated lung
 3                   function decrements,  primarily in individuals with asthma, by finding concomitant
 4                   increases in respiratory symptoms (Khatri et al.. 2009; Just et al.. 2002; Mortimer et al..
 5                   2002; Ross et al.. 2002; Gielenetal.. 1997; Romieuetal.. 1997; Thurston et al.. 1997;
 6                   Romieuetal.. 1996).

 7                   Collectively, epidemiologic studies have examined and found decreases in lung function
 8                   in association with single-day O3 concentrations lagged from 0 to 7 days as well
 9                   concentrations averaged over 2-10 days. A large body of evidence indicates decreases in
10                   lung function in association with O3 exposures over the duration of outdoor activity,
11                   same-day, or previous-day O3 exposures (Sonet al.. 2010; Alexeeff et al.. 2008; Lewis  et
12                   al.. 2005; Ross et al..  2002; Jalaludin et al.. 2000; Chen et al.. 1999; Romieuetal.. 1997;
13                   Braueretal. 1996: Romieuetal.. 1996: Spektor et al.. 1988b). Fewer studies find
14                   associations with longer lags of ambient O3 exposures (5-7 days) (Wiwatanadate and
15                   Trakultivakorn. 2010: Hernandez-Cadena et al.. 2009: Steinvil et al.. 2009). However,
16                   associations with multiday averages of exposure (Sonet al.. 2010: Liu et al.. 2009a:
17                   Barraza-Villarreal et al.. 2008: O'Connor et al.. 2008: Alexeeff et al.. 2007: Mortimer et
18                   al.. 2002: Ward et al.. 2002: Gold et al.. 1999: Naeher et al.. 1999: Neas et al.. 1999)
19                   indicate that exposures accumulated over several days may be important. For single- and
20                   multi-day O3 exposures, associations with lung function decrements were observed for  1-
21                   h max, 8-h max, and 24-h avg O3, without a clear indication that the strength of evidence
22                   varied among the averaging times. Within studies, O3 exposure for various lag periods
23                   were associated with lung function decrements, possibly indicating that multiple modes
24                   of action  may be involved in the responses. Activation of bronchial C-fibers (Section
25                   5.3.2) may lead to decreases in lung function as an immediate response to O3  exposure,
26                   and increased airway  hyperresponsiveness resulting from sensitization of airways
27                   (Section 5.3.5) may mediate lung function responses associated with the lagged or
28                   multiday  O3 exposures (Peden. 2011).

29                   Several studies found that associations with lung function decrements persisted at lower
30                   ambient O3 concentrations. For exposures averaged up to 1 hour during outdoor activity,
31                   multiple studies in individuals engaged in outdoor activities found associations with O3
32                   concentrations limited to those below 80 ppb (Spektor etal., 1988a: Spektor et al..
33                   1988b). 60 ppb (Brunekreef et al.. 1994: Spektor et al.. 1988a). and 50 ppb (Brunekreef et
34                   al.. 1994). Among outdoor workers, Brauer et al. (1996) found a robust association with
35                   daily 1-h  max O3 concentrations below 40 ppb. For 8-h average O3 exposures,
36                   associations with lung function decrements in children with asthma were found to persist
37                   at concentrations less than 80 ppb in a U.S. multicity study (for lag 1-5 avg) (Mortimer et
      Draft - Do Not Cite or Quote                       6-55                                September 2011

-------
 1                   al.. 2002) and less than 51 ppb in a study conducted in the Netherlands (for lag 2) (Gielen
 2                   etal.. 1997).

 3                   Several studies of lung function evaluated confounding by meterological factors and
 4                   copollutant exposures. Most O3 effect estimates remained robust in models that adjusted
 5                   for temperature, humidity, and co-pollutants such as PM2s, PMio, NO2, or SO2. Although
 6                   examined in relatively few epidemiologic studies, O3 was associated with decreases in
 7                   lung function in models that included pollen or acid aerosols. The consistency of
 8                   association in the collective body of evidence with and without adjustment for
 9                   copollutant exposures and meterological factors combined with evidence from controlled
10                   human exposure studies for the direct effect of O3 exposure provide substantial evidence
11                   for the independent effects of ambient O3 exposure on lung function decrements.
                     6.2.1.3    Toxicology

12                   The 2006 O3 AQCD concluded that pulmonary function decrements occur in a number of
13                   species with acute exposures (< 1 week), ranging from 0.25 to 0.4 ppm O3 (U.S. EPA.
14                   2006b). Early work has demonstrated that during acute exposure of ~0.2 ppm O3 in rats,
15                   the most commonly observed alterations are increased frequency of breathing and
16                   decreased tidal volume (i.e., rapid, shallow breathing). Decreased lung volumes are
17                   observed in rats with acute exposures to 0.5 ppm O3. At concentrations of >1 ppm,
18                   breathing mechanics (compliance and resistance) are also affected. Exposures of 6 h/day
19                   for 5 days create a pattern of attenuation of pulmonary function decrements in both rats
20                   and humans without concurrent attenuation of lung injury and morphological changes,
21                   indicating that the attenuation did not result in protection against all the effects  of O3
22                   (Wiester et al.. 1996b). A number of studies examining the effects of O3 on pulmonary
23                   function in rats, mice, and dogs are described in Table 6-13 on p. 6-91 of the 1996 O3
24                   AQCD and Table AX5-11 on p. AX5-34 of the 2006 O3 AQCD (U.S. EPA.  2006b.
25                   1996a). Recent lung imaging studies using hyperpolarized 3He provide evidence of
26                   ventilation abnormalities in rats following exposure to 0.5 ppm O3 (Cremillieux et al..
27                   2008). Rats  were exposed to 0.5 ppm O3 for 2 or 6 days, either continuously (22 h/day) or
28                   alternating (12 h/day). Dynamic imaging of lung filling (2 mL/s) revealed delayed and
29                   incomplete filling of lung segments and lobes. Abnormalities were mainly found in the
30                   upper regions of the lungs and proposed due to the spatial distribution of O3 exposure
31                   within the lung. Although the small number of animals used in the study (n = 3  to
32                   7/group) makes definitive conclusions difficult, the authors suggest that the delayed
33                   filling of lung lobes or segments is likely a result of an increase in airway resistance
34                   brought about by narrowing of the peripheral small airways.
      Draft - Do Not Cite or Quote                       6-56                                September 2011

-------
            6.2.2   Airway Hyperresponsiveness

 1                   Airway hyperresponsiveness refers to a condition in which the conducting airways
 2                   undergo enhanced bronchoconstriction in response to a variety of stimuli. Airway
 3                   responsiveness is typically quantified by measuring changes in pulmonary function (e.g.,
 4                   FEVi or specific airway resistance [sRaw]) following the inhalation of an aerosolized
 5                   specific (allergen) or nonspecific (e.g.,  methacholine) bronchoconstricting agent or
 6                   another stimulus such as exercise or cold air. Asthmatics are generally more sensitive to
 7                   bronchoconstricting agents than nonasthmatics, and the use of an airway challenge to
 8                   inhaled bronchoconstricting agents is a diagnostic test in asthma. Standards for airway
 9                   responsiveness testing have been developed for the clinical laboratory (ATS. 2000aX
10                   although variation in methodology for administering the bronchoconstricting agent may
11                   affect the results (Cockcroft et al..  2005). There is a wide range of airway responsiveness
12                   in nonasthmatic people, and responsiveness is influenced by wide range of factors,
13                   including cigarette smoke, pollutant exposures, respiratory infections, occupational
14                   exposures, and respiratory irritants. Airways hyperresponsiveness in response to O3
15                   exposure has not been examined widely in epidemiologic studies; such evidence is
16                   derived primarily from controlled human exposure and toxicological studies.
                     6.2.2.1   Controlled Human Exposures

17                   Beyond its direct effect on lung function, O3 exposure causes an increase in airway
18                   responsiveness in human subjects as indicated by a reduction in the concentration of
19                   specific (e.g., ragweed) and non-specific (e.g., methacholine) agents required to produce
20                   a given reduction in FEVi or increase in sRaw. Increased airway responsiveness is an
21                   important consequence of exposure to ambient O3, because the airways are then
22                   predisposed to narrowing upon inhalation of a variety of ambient stimuli including
23                   specific allergens, SO2, and cold air.

24                   Increases in airway responsiveness have been reported for exposures to 80 ppb O3 and
25                   above. Horstman et al. (1990) evaluated airway responsiveness to methacholine in young
26                   healthy adults (22 M) exposed to 80, 100, and 120 ppb O3 (6.6 h, quasi continuous
27                   moderate exercise, 39 L/min). Dose-dependent decreases of 33, 47, and 55% in the
28                   cumulative dose of methacholine required to produce a 100% increase in sRaw after
29                   exposure to O3 at 80, 100, and 120 ppb, respectively, were reported. Molfino et al. (1991)
30                   reported increased allergen-specific airway responsiveness in mild asthmatics exposed to
31                   120 ppb O3 (1 h resting exposure). Due to safety concerns, however, the exposures in the
32                   Molfino et al. (1991) study were not randomized with FA conducted first and O3
33                   exposure second. Attempts to reproduce the findings of Molfino et al. (1991) using a
      Draft - Do Not Cite or Quote                      6-57                                September 2011

-------
 1                   randomized exposure design have not found statistically significant changes in airway
 2                   responsiveness at such low levels of O3 exposure. At a considerably higher exposure to
 3                   250 ppb O3 (3 h, light-to-moderate intermittent exercise, 30 L/min), Torres et al. (1996)
 4                   found significant increases in specific and non-specific airway responsiveness of mild
 5                   asthmatics 3 h following O3 exposure. Kehrl et al. (1999) found increased reactivity to
 6                   house dust mite antigen in mild atopic asthmatics 16-18 h after exposure to 160 ppb O3
 7                   (7.6 h, light quasi continuous exercise, 25 L/min). Holz et al. (2002) demonstrating that
 8                   repeated daily exposure to lower concentrations of 125 ppb O3 (3 h for four consecutive
 9                   days; intermittent exercise, 30 L/min) causes an increased response to allergen challenge
10                   at 20 h postexposure in allergic airway disease.

11                   O3 exposure of asthmatic subjects, who characteristically have increased airway
12                   responsiveness at baseline relative to healthy controls (by nearly two orders of
13                   magnitude), can cause further increases in responsiveness (Kreit et al.. 1989). Similar
14                   relative changes in airway responsiveness are seen in asthmatics and healthy control
15                   subject exposed to O3 despite their markedly different baseline airway responsiveness.
16                   Several studies (Kehrl et al.. 1999; Torres et al..  1996; Molfino et al.. 1991) have
17                   suggested an increase in specific (i.e., allergen-induced) airway reactivity. An important
18                   aspect of increased airway responsiveness after O3 exposure is that this may represent a
19                   plausible link between ambient O3 exposure and increased respiratory symptoms in
20                   asthmatics, and  increased hospital admissions and ED visits for asthma.

21                   Changes in airway responsiveness after O3 exposure appear to resolve more slowly than
22                   changes in FEVi or respiratory symptoms (Folinsbee and Hazucha, 2000). Studies
23                   suggest that O3-induced increases in airway responsiveness usually resolve 18 to 24 h
24                   after exposure, but may persist in some individuals for longer periods (Folinsbee and
25                   Hazucha. 1989). Furthermore, in studies of repeated exposure to O3, changes in airway
26                   responsiveness tend to be somewhat less susceptible to attenuation with consecutive
27                   exposures than changes in FEVi (Gong et al.. 1997a: Folinsbee et al..  1994; Kulle et al..
28                   1982; Dimeo etal., 1981). Increases in airway responsiveness do not appear to be
29                   strongly associated with decrements in lung function or increases in symptoms (Aris et
30                   al., 1995). Recently, Que et al.  assessed methacholine responsiveness in healthy young
31                   adults (83M, 55 F) at one day after exposure to 220 ppb O3 and FA for 2.25 h (alternating
32                   15 min periods of rest and brisk treadmill walking). Increases in airways responsiveness
33                   at 1 day post-O3 exposure were not correlated with FEVi responses immediately
34                   following the O3 exposure nor with changes in epithelial permeability assessed 1 day
35                   post-O3 exposure.
      Draft - Do Not Cite or Quote                       6-58                                September 2011

-------
                     6.2.2.2    Toxicology

 1                   In addition to human subjects, a number of species, including nonhuman primates, dogs,
 2                   cats, rabbits, and rodents, have been used to examine the effect of O3 exposure on airway
 3                   hyperresponsiveness. With a few exceptions, commonly used animal models have been
 4                   guinea pigs, rats, or mice acutely exposed to O3 concentrations of 1 to 3 ppm to induce
 5                   airway hyperresponsiveness. These animal models are helpful for determining underlying
 6                   mechanisms of general airway hyperresponsiveness and are relevant for understanding
 7                   airway responses in humans. Although 1-3 ppm may seem like a high exposure
 8                   concentration, based on 18O3 (oxygen-18-labeled ozone) in the BALF of humans and rats,
 9                   an exposure of 0.4 ppm O3 in exercising humans appears roughly equivalent to an
10                   exposure of 2 ppm in resting rats (Hatch etal.. 1994).

11                   A limited number of studies have observed airway hyperresponsiveness in rodents and
12                   guinea pigs after exposure to less than 0.3 ppm O3. As previously reported in the 2006 O3
13                   AQCD, one study demonstrated that a very low concentration of O3 (0.05 ppm for 4 h)
14                   induced airway hyperresponsiveness in some of the nine strains of rats tested (Depuydt et
15                   al., 1999). This effect occurred at a concentration of O3  that was much lower than has
16                   been reported to induce airway hyperresponsiveness in any other species. Similar to
17                   ozone's effects on  other endpoints, these observations suggest a genetic component plays
18                   an important role in O3-induced  airway hyperresponsiveness in this species and warrants
19                   verification in other species. More recently, Chhabra and colleagues (2010)  demonstrated
20                   that exposure of ovalbumin (OVA)-sensitized guinea pigs to 0.12 ppm for 2 h/day for 4
21                   weeks produced specific airway hyperresponsiveness to an inhaled OVA challenge.
22                   Interestingly, in this study , dietary supplementation of the guinea pigs with vitamins C
23                   and E  ameliorated  a portion of the airway hyperresponsiveness as well as indices of
24                   inflammation and oxidative stress. Larsen and colleagues did an O3 concentration-
25                   response study in mice sensitized by 10 daily inhalation treatments with an OVA aerosol
26                   (Larsen et al.. 2010). Although airway responsiveness to methacholine was increased in
27                   non-sensitized animals exposed to a single 3-h exposure to 0.5, but not 0.1 or 0.25, ppm
28                   O3, airway hyperresponsiveness was observed after exposure to 0.1 and 0.25 ppm O3 in
29                   OVA-sensitized mice. Shore and colleagues (Johnston et al.. 2005b) have also
30                   demonstrated O3-induced airway hyperresponsiveness in mice after exposure to 0.3 ppm
31                   O3 for 3 hours. Mice that were exposed to the same concentration of O3 for 72 hours
32                   showed no evidence of airway hyperresponsiveness, indicating attenuation of this effect.
33                   Thus, recent toxicological studies have demonstrated that O3-induced airway
34                   hyperresponsiveness occurs in guinea pigs and mice after either acute or repeated
35                   exposure to relevant concentrations of O3.
      Draft - Do Not Cite or Quote                       6-59                                September 2011

-------
 1                   The mechanisms by which O3 enhances the airway responsiveness to either specific (e.g.,
 2                   OVA) or non-specific (e.g., methacholine) bronchoprovocation are not clear, but appear
 3                   to be associated with complex cellular and biochemical changes in the conducting
 4                   airways. Considerable research effort has been directed towards exploring the causes of
 5                   O3-induced airway hyperresponsiveness, but the majority of such studies have been
 6                   conducted at high  concentrations of O3. It is clear that inflammation  plays a key role in
 7                   O3-induced airway hyperresponsiveness, although the precise mediators and cells that are
 8                   involved have not been identified at relevant concentrations of O3. Because inflammation
 9                   is likely to play a role in O3-induced airway hyperresponsiveness, the mechanism for this
10                   response may be multifactorial, involving the presence of cytokines, prostanoids, or
11                   neuropeptides; activation of macrophages, eosinophils, or mast cells; and epithelial
12                   damage that increases direct access of mediators to the smooth muscle or receptors in the
13                   airways that are responsible for reflex bronchoconstriction. Johnston et al. (2005b)
14                   demonstrated that airway hyperresponsiveness occurred in both wild type and IL-6
15                   knockout mice exposed to 0.3 ppm O3 despite reduction in markers of lung injury and
16                   inflammation in O3-exposed IL-6 knockout mice. This same group of investigators has
17                   demonstrated the involvement of natural killer T cells, obesity, CXCR2, leptin, and IL-17
18                   in O3-induced airway hyperresponsiveness at exposure concentrations of 1-3 ppm O3
19                   (Garantziotis et al.. 2010; Voynow et al.. 2009; Pichavant et al.. 2008; Williams et al..
20                   2007b: Lu et al.. 2006; Johnston et al.. 2005a: Shore et al.. 2003). A  recent study
21                   demonstrated a role for mindin, an extracellular matrix protein, in the AHR response
22                   resulting from acute exposure to 1 ppm O3 (Frush et al..  In Press). Thus, a number of
23                   potential mediators and cells may play a role in O3-induced airway hyperresponsiveness;
24                   mechanistic studies are discussed in greater detail in Chapter 5.

25                   In order to evaluate the ability of O3 to enhance specific and non-specific airway
26                   responsiveness, it  is important to take into  account the phenomenon  of attenuation in
27                   ozone's effects. Several studies have clearly demonstrated that some effects caused by
28                   acute exposure are absent after repeated exposures to O3. The ability of the pulmonary
29                   system to adapt to repeated insults to  O3 is complex, however, and experimental findings
30                   for attenuation to O3-induced airway hyperresponsiveness are inconsistent. As described
31                   above, airway hyperresponsiveness was observed in mice after a 3-h exposure but not in
32                   mice exposed continuously for 72 hours to 0.3 ppm (Johnston et al..  2005b). However,
33                   the Chhabra study demonstrated O3-induced airway hyperresponsiveness in guinea pigs
34                   exposed for 2 h/day for 10 days (Chhabra et al.. 2010). Besides the obvious species
35                   disparity, these studies differ in that the mice were exposed continuously for 72 hours,
36                   whereas the guinea pigs were exposed intermittently over 10 days, suggesting that
37                   attenuation might be lost with periods of rest in between O3 exposures.
      Draft - Do Not Cite or Quote                       6-60                                September 2011

-------
            6.2.3   Pulmonary Inflammation, Injury and Oxidative Stress

 1                   In addition to physiological pulmonary responses, respiratory symptoms, and airway
 2                   hyperresponsiveness, O3 exposure has been shown to result in increased epithelial
 3                   permeability and respiratory tract inflammation. In general, inflammation can be
 4                   considered as the host response to injury and the induction of inflammation as evidence
 5                   that injury has occurred. Inflammation induced by exposure of humans to O3 can have
 6                   several potential outcomes: (1) inflammation induced by a single exposure (or several
 7                   exposures over the course of a summer) can resolve entirely; (2) continued acute
 8                   inflammation can evolve into a chronic inflammatory state; (3) continued inflammation
 9                   can alter the structure and function of other pulmonary tissue, leading to diseases  such as
10                   fibrosis; (4) inflammation can alter the body's host defense response to inhaled
11                   microorganisms, particularly in potentially susceptible populations such as the very
12                   young and old; and (5) inflammation can alter the lung's response to other agents such as
13                   allergens or toxins. Except for outcome (1), the possible chronic responses have only
14                   been directly observed in animals exposed to O3. It is also possible that the profile of
15                   response can be altered in persons with preexisting pulmonary disease (e.g. asthma,
16                   COPD) or smokers. Oxidative stress has been  shown to play a key role in initiating and
17                   sustaining O3-induced inflammation. Secondary oxidation products formed as a result of
18                   reactions between O3 and components of the ELF can increase the expression of
19                   cytokines, chemokines, and adhesion molecules and enhance airway epithelium
20                   permeability (Sections 5.3.3. and 5.3.4.).
                    6.2.3.1    Controlled Human Exposures

21                  As reported in studies reviewed in the 1996 and 2006 O3 AQCDs, acute O3 exposure
22                  initiates an acute inflammatory response throughout the respiratory tract which has been
23                  observed to persist for at least 18-24 hours postexposure. A meta-analysis of 21 studies
24                  (Mudway and Kelly. 2004a) showed that neutrophils (PMN) influx in healthy subjects
25                  was linearly associated (p<0.01) with total O3 dose (i.e., the product of O3 concentration,
26                  exposure duration, and VE). As with FEVi responses to O3, within individual
27                  inflammatory responses to O3 are  generally reproducible and correlated between repeat
28                  exposures (Holz et al.. 1999). Some individuals also appear to be intrinsically more
29                  susceptible to increased inflammatory responses to O3 exposure (Holz et al.. 2005).
30                  The presence of PMNs in the lung has long been accepted as a hallmark of inflammation
31                  and is an important indicator that  O3 causes inflammation in the lungs. Neutrophilic
32                  inflammation of tissues indicates activation of the innate immune system and requires a
33                  complex series of events which are normally followed by processes that clear the
      Draft - Do Not Cite or Quote                      6-61                                September 2011

-------
 1                   evidence of acute inflammation. Inflammatory effects have been assessed in vivo by
 2                   lavage (proximal airway and bronchoalveolar), bronchial biopsy, and more recently,
 3                   induced sputum. A single acute exposure (1-4 hours) of humans to moderate
 4                   concentrations of O3 (0.2-0.6 ppm) while exercising at moderate to heavy intensities
 5                   results in a number of cellular and biochemical changes in the lung, including an
 6                   inflammatory response characterized by increased numbers of PMNs, increased
 7                   permeability of the epithelial lining of the respiratory tract, cell damage, and production
 8                   of proinflammatory cytokines and prostaglandins (U.S. EPA. 2006b). These changes also
 9                   occur in humans exposed to 80 and 100 ppb O3 for 6-8 hours (Alexis et al., 2010; Peden
10                   et al.. 1997; Devlin et al.. 1991). Soluble mediators of inflammation such as the cytokines
11                   (e.g., IL-6, IL-8) and arachidonic acid metabolites (e.g., prostaglandin [PG]E2, PGF2a,
12                   thromboxane, and leukotrienes [LTs] such as LTB4) have been measured in the BALF of
13                   humans exposed to O3. In addition to their role in inflammation, many of these
14                   compounds have bronchoconstrictive properties and may be involved in increased airway
15                   responsiveness following O3 exposure. The possible relationship between repetitive bouts
16                   of acute inflammation in humans caused by O3 and the development of chronic
17                   respiratory disease is unknown.

18                   Studies reviewed in the 2006 O3 AQCD reported that inflammatory responses do not
19                   appear to be correlated with lung function responses in either asthmatic or healthy
20                   subjects. In healthy adults (14 M, 6 F) and asthmatic (12 M, 6 F) volunteers exposed to
21                   200 ppb O3 (4 h with moderate quasi continuous exercise, VE = 44 L/min), percent PMN
22                   and total protein in BAL fluids were significantly increased in the asthmatics relative to
23                   the healthy controls. Spirometric measures of lung function were significantly decreased
24                   following the O3 exposure in both groups, but were not significantly different between
25                   the asthmatic and healthy subjects. Effects of O3 on PMN and total protein were not
26                   correlated with changes in FEVi or FVC (Balmes et al.. 1997: Balmes et al.. 1996V
27                   Devlin  et al. (1991)  exposed healthy adults (18 M) to 80 and 100 ppb O3 (6.6 h  with
28                   moderate quasi continuous exercise, 40 L/min). In BAL fluid collected 18 h after
29                   exposure to 100 ppb O3, significant increases in PMNs, protein, PGE2, fibronectin, IL-6,
30                   lactate dehydrogenase, and a-1 antitrypsin compared to FA. Similar but smaller increases
31                   in all mediators were found after exposure to 80 ppb O3 except for protein and
32                   fibronectin. Changes in BAL markers were not correlated with changes in FEVi. Holz et
33                   al. (1999) examined inflammatory responses in healthy (n=21) and asthmatic (n=15)
34                   subjects exposed to  125 and 250 ppb O3 (3 h, light intermittent exercise, 26 L/min).
35                   Significantly increased percent PMN in sputum due to O3 exposure was observed in both
36                   asthmatics and healthy subjects following the 250 ppb exposure. At the lower, 125 ppb
37                   exposure, only the asthmatic group experienced statistically significantly increases in the
38                   percent PMN. Significant decrements in FEVi were only found following exposure to
39                   250 ppb; these changes in FEVi did not differ significantly between the asthmatic and

      Draft - Do Not Cite or Quote                       6-62                                September 2011

-------
 1                   healthy groups, nor were changes in FEVi correlated with changes in PMN levels. In
 2                   contrast to these earlier findings, Vagaggini et al. (2010) recently reported a significant
 3                   (r=0.61, p=0.015) correlation between changes in FEVi and changes in sputum
 4                   neutrophils in mild-to-moderate asthmatics (n=23;  33 ± 11 years) exposed to 300 ppb O3
 5                   for 2 hours with moderate exercise.

 6                   The time course of the inflammatory response to O3 in humans has not been fully
 7                   characterized. Different markers exhibit peak responses at different times. Studies in
 8                   which lavages were performed 1 hour after O3 exposure (1 h at 0.4 ppm or 4 h at
 9                   0.2 ppm) have demonstrated that the inflammatory responses are quickly initiated (Torres
10                   etal., 1997; Devlin et al., 1996; Schelegle et al., 1991). Inflammatory mediators and
11                   cytokines such as IL-8, IL-6, and PGE2 are greater at 1 h than at 18 h post-O3  exposure
12                   (Torres et al.. 1997; Devlin et al., 1996). However, IL-8 still remained elevated at 18 h
13                   post-O3 exposure (4 h at 0.2 ppm O3 versus FA) in healthy subjects (Balmes et al.. 1996).
14                   Schelegle et al. (1991) found increased PMNs in the "proximal airway" lavage at 1, 6,
15                   and 24 hours after O3 exposure (4 h at 0.2 ppm O3), with a peak response at 6  hours.
16                   However, at  18-24 hours after O3 exposure, PMNs remain elevated relative to 1  hour
17                   postexposure (Torres et al.. 1997; Schelegle et al..  1991).

18                   Alexis et al. (2010) recently reported that a 6.6-hour exposure with moderate exercise to
19                   80 ppb O3  caused increased sputum neutrophil levels at 18 hours postexposure in young
20                   healthy adults (n=15; 24 ± 1 years).  In a prior study, Alexis et al. (2009) found genotype
21                   effects on inflammatory responses but not lung function responses to a 2 h-exposure to
22                   400 ppb O3. At 4 h post O3 exposure, both GSTM1 genotypes had significant  increases in
23                   sputum neutrophils with a tendency  for a greater increase in GSTM1-sufficient than null
24                   individuals. At 24 h postexposure, neutrophils had  returned to baseline levels  in the
25                   GSTM1-sufficient individuals. In the GSTMl-null subjects, however, neutrophil levels
26                   increased further from 4 h to 24 h and were significantly greater than both baseline levels
27                   and 24 h levels in GSTM1-sufficient individuals. Alexis et al. (2009) found that GSTM1-
28                   sufficient individuals (n=19; 24 ± 3 years) had a decrease in macrophage levels at 4-
29                   -24 hours postexposure to 400 ppb O3 for 2 h with  exercise. These studies also provide
30                   evidence for activation of innate immunity and antigen presentation, as discussed in
31                   Section 5.3.6. Effects of the exposure apart from O3 cannot be ruled out in the Alexis et
32                   al. (2010; 2009) studies, however, since no FA exposure was conducted.

33                   Kim et al.  (2011) has more recently  shown a significant (p < 0.001) increase in sputum
34                   neutrophil  levels following a 6.6-hour exposure to  60 ppb O3 relative to FA in young
35                   healthy adults (13 F, 11 M; 25.0 ± 0.5 years). There was no significant effect of GSTM1
36                   genotype (half GSTMl-null) on the  inflammatory responses observed in these
      Draft - Do Not Cite or Quote                       6-63                                 September 2011

-------
 1                   individuals. Previously, inflammatory responses had only been evaluated down to a level
 2                   of80ppbO3.

 3                   Inflammatory responses to O3 exposure have also been studied in asthmatic subjects
 4                   (Pedenetal.. 1997; Scannell et al.. 1996; Bashaetal.. 1994). In these studies, asthmatics
 5                   showed significantly more neutrophils in BALF (18 hours postexposure) than did
 6                   similarly exposed healthy individuals. In one of these studies (Peden et al.. 1997). which
 7                   included only allergic asthmatics who tested positive for Dematophagoides farinae
 8                   antigen, there was an eosinophilic inflammation (twofold increase), as well as
 9                   neutrophilic inflammation (threefold increase). In a study of subjects with intermittent
10                   asthma exposed to 0.4 ppm O3 for 2 hours, increases in eosinophil cationic protein,
11                   neutrophil elastase and IL-8 were found to be significantly increased 16 hours
12                   postexposure and comparable in induced sputum and BALF (Hiltermann et al.. 1999).
13                   Scannell et al. (1996) also reported that IL-8 tends to be higher in the BALF of
14                   asthmatics compared to nonasthmatics following O3 exposure, suggesting  a possible
15                   mediator for the significantly increased neutrophilic inflammation in those subjects.
16                   Bosson et al. (2003) found significantly greater epithelial expression of IL-5, IL-8,
17                   granulocyte-macrophage colony-stimulating factor and epithelial cell-derived neutrophil -
18                   activating peptide-78 in asthmatics compared to healthy subjects following exposure to
19                   0.2 ppm O3 for 2 h. In contrast, Stenfors et al. (2002) did not detect a difference in the O3-
20                   induced increases in neutrophil numbers between 15 mild asthmatic and 15 healthy
21                   subjects by bronchial wash at the 6 h postexposure time point. However, the asthmatics
22                   were on average 5 years older than the healthy subjects in this study, and it is not yet
23                   known how age affects inflammatory responses. It is also possible that the time course of
24                   neutrophil influx differs between healthy and asthmatic individuals. Differences between
25                   asthmatics and healthy subjects in ozone-mediated activation of innate and adaptive
26                   immune responses have been observed  in two studies (Hernandez et al.. 2010;  Bosson et
27                   al., 2003). as discussed in Sections 6.2.5.4  and 5.4.2.2.

28                   Vagaggini et al. (2002) investigated the effect of prior allergen challenge on responses in
29                   mild asthmatics exposed for 2 h to  0.27 ppm O3 or filtered air. At 6 h postexposure,
30                   eosinophil numbers in induced sputum were found to be significantly greater after O3
31                   than after air exposures. Studies such as this suggest that the time course of eosinophil
32                   and neutrophil influx following O3 exposure can occur at levels detectable within the
33                   airway lumen by as early as 6 h. They also suggest that the previous or concurrent
34                   activation of proinflammatory pathways within the airway epithelium may enhance the
35                   inflammatory effects of O3. For example, in an in vitro study of primary human nasal
36                   epithelial cells and BEAS-2B cell line, cytokine production induced by rhinovirus
37                   infection was enhanced synergistically by concurrent exposure to O3 at 0.2 ppm for 3
38                   hours (Spannhake et al.. 2002).
      Draft - Do Not Cite or Quote                       6-64                                 September 2011

-------
 1                   Markers from BALF following both 2 hours (Devlin et al., 1997) and 4 hours (Torres et
 2                   al.. 2000; Christian et al.. 1998) repeated O3 exposures (up to 5 days) indicate that there is
 3                   ongoing cellular damage irrespective of the attenuation of some cellular inflammatory
 4                   responses of the airways, pulmonary function, and symptom responses. Devlin et al.
 5                   (1997) found that several indicators of inflammation (e.g., PMN, IL-6, PGE2, fibronectin)
 6                   were attenuated after 5 days of exposure (i.e., values were not different from FA).
 7                   However, other markers (LDH, IL-8, total protein, epithelial cells) did not show
 8                   attenuation, suggesting that tissue damage probably continues to occur during repeated
 9                   exposure. Some cellular responses did not return to baseline  levels for more than 10-20
10                   days following O3 exposure. Christian et al. (1998) showed decreased numbers of
11                   neutrophils and a decrease in IL-6 levels in healthy adults after 4 days of exposure versus
12                   the single exposure to 0.2 ppm O3 for 4 h. Torres et al. (2000) also found both functional
13                   and BALF cellular responses to O3 were abolished at 24 hours postexposure following
14                   the fourth exposure day. However, levels of total protein, IL-6, IL-8, reduced glutathione
15                   and ortho-tyrosine was still increased significantly. In addition, visual scores
16                   (bronchoscopy) for bronchitis and erythema and the numbers of neutrophils in bronchial
17                   mucosal biopsies were increased. Results indicate that, despite an attention of some
18                   markers of inflammation in BALF and pulmonary function decrements, inflammation
19                   within the airways persists following repeated exposure to O3. The continued presence of
20                   cellular injury markers indicates a persistent effect that may not necessarily be recognized
21                   due to the attenuation of spirometric and symptom responses.

22                   A number of studies show that O3 exposures increase epithelial cell permeability through
23                   direct (technetium-99m labeled diethylene triamine pentaacetic acid, 99mTc-DTPA,
24                   clearance) and indirect (e.g., increased BALF albumin, protein) techniques. Kehrl et al.
25                   (1987) showed increased 99mTc-DTPA clearance in healthy young adults at 75 minutes
26                   postexposure to 0.4 ppm O3 for 2 hours. Foster and Stetkiewicz (1996) have shown that
27                   increased 99mTc-DTPA clearance persists for at least 18-20 hours post-O3 exposure (130
28                   minutes to average O3 concentration of 0.24 ppm), and the effect is greater at the lung
29                   apices than at the base. Increased BALF protein, suggesting O3-induced changes in
30                   epithelial permeability, have also been reported at 1 hour and 18 hours postexposure
31                   (Devlin et al.. 1997; Balmes etal..  1996). Meta-analysis of results from 21  publications
32                   (Mudway and Kelly. 2004a). showed that increased BALF protein is associated with total
33                   inhaled O3 dose (i.e., the product of O3 concentration, exposure duration, and VE).

34                   It has been postulated that changes in permeability associated with acute inflammation
35                   may provide increased access of inhaled antigens, particles, and other inhaled substances
36                   deposited on lung surfaces to the smooth muscle, interstitial cells, and the blood. Hence,
37                   increases in epithelial permeability following O3 exposure might lead to increases in
38                   airway responsiveness to specific and nonspecific agents. Que et al.  investigated this
      Draft - Do Not Cite or Quote                       6-65                                September 2011

-------
 1                   hypothesis in healthy young adults (83M, 55 F) exposed to 220 ppb O3 for 2.25 h
 2                   (alternating 15 min periods of rest and brisk treadmill walking). As has been observed by
 3                   others for FEVi responses, within individual changes in permeability were correlated
 4                   between sequential O3 exposures. This indicates differences in susceptibility to epithelial
 5                   damage from O3 exposure among individuals. Increases in epithelial permeability at 1
 6                   day post-O3 exposure were not correlated with FEVi responses immediately following
 7                   the O3 exposure nor with changes in airway responsiveness to methacholine in assessed 1
 8                   day post-O3 exposure. The authors concluded that changes in FEVi, permeability, and
 9                   airway responsiveness following O3  exposure were relatively constant over time in young
10                   healthy adults; although, these responses appear to be mediated by differing physiologic
11                   pathways.
                     6.2.3.2    Epidemiology

12                   In the 2006 O3 AQCD, epidemiologic evidence of changes in pulmonary inflammation in
13                   association with short-term ambient O3 exposure (30-min or 1-h max) was limited to
14                   observations of increases in nasal lavage levels of inflammatory cell counts, eosinophilic
15                   cationic protein, and myeloperoxidases (U.S. EPA. 2006b). As a result of the
16                   development of less invasive methods to collect exhaled breath samples repeatedly from
17                   individuals in the field, the number of studies assessing ambient O3-related changes in
18                   lower airway inflammation and oxidative stress in recent years has increased
19                   dramatically. Although most of the biomarkers examined in these studies were not
20                   specific to the lung, most studies collected exhaled breath, exhaled breath condensate
21                   (EEC), nasal lavage fluid, or induced sputum with the aim of monitoring inflammatory
22                   responses in airways, as opposed to monitoring systemic responses in blood. These recent
23                   studies form a larger base to establish coherence with findings from human experimental
24                   and animal toxicological studies that have measured similar or related endpoints and
25                   provide further biological plausibility for associations of ambient O3 exposure with
26                   respiratory symptoms and lung function decrements.  These biological markers also allow
27                   the assessment of short-term O3-related acute respiratory effects in populations that are
28                   less likely to experience respiratory symptoms, including healthy populations and groups
29                   with increased outdoor exposures.
      Draft - Do Not Cite or Quote                       6-66                                 September 2011

-------
Table 6-14      Mean and upper percentile ozone concentrations in studies
                   examining biological markers of pulmonary inflammation and
                   oxidative stress
Study
Qian et al.
(2009)
Khatrietal.
(2009)
Ferdinands etal.
(2008)
Adamkiewicz et
al. (2007)
Berhane et al.
(2011)
Delfino et al.
(201 Oa)
Liu et al. (2009a)
Sienra-Monge et
al. (2004)
Barraza-
Villarreal et al.
(2008)
Romieu et al.
(2008)
Location
Boston, MA; New York, NY;
Denver, CO; Philadelphia, PA;
San Francisco, CA; Madison, Wl
(SOCS)
Atlanta, GA
Atlanta, GA
Steubenville, OH
13 Southern California
Communities
Los Angeles, CA
Windsor, ON, Canada
Mexico City, Mexico
Mexico City, Mexico
Mexico City, Mexico
Years
1997-1999
All-year
2003, 2005,
2006
Warm season
2004
Warm season
2000
Cold season
September
2004- June
2005
2005-2007
All-year
2005
Cold season
1999-2000
All-year
2003-2005
All-year
2004
All-year
03
Averaging
Time
8-h max
8-h max
1-h max
24-h avg
1-havgb
8-h avg
(10:00 a.m. -
6:00 p.m.)
24-h avg
24-h avg
8-h max
8-h max
8-h max
Mean/Median
Concentration (ppb)
33.6
59a
61 (median)
15.3
19.8
NR
Warm season: 33.3
Cool season: 20.6
13.0
66.2
31.6
31.1
Upper Percentile
Concentrations (ppb)
75th: 44.4, Max
Max: 73a
75th: 67
75'": 20.2, Max:
75th: 27.5, Max:
NR
Max: 76.4
Max: 44.9
95": 26.5
Max: 142.5
Max (8-h max):
75'": 38.3
Max: 60.7
:91.5


32.2
61.6




86.3

Nickmilder et al.  southern Belgium
(2007)
 2002         1-h max      NR
 Warm season  8-h max      NR
Max (across 6 camps):
24.5-112.7°
Max (across 6 camps):
18.9-81.1°
Chimenti et al.   Sicily, Italy
 NR          8-h avg       Fall: 32.7 (pre-race), 35.1
 All-year       (7:00 a.m.-    (race)0
             3:00 p.m.)     Winter: 37.0 (pre-race),
                         30.8 (race)0
                         Summer: 51.2 (pre-race),
	46.1 (race)0	
NR
  Max = Maximum, NR = Not Reported.
  'Individual-level exposure estimates were derived based on time spent in the vicinity of various 03 monitors.
  bAverage 03 oncentration in the 1 h preceding eNO collection.
  ""Concentrations converted from ug/m3 to ppb using the conversion factor of 0.51 assuming standard temperature (25°C) and pressure (1 atm).
Draft - Do Not Cite or Quote
             6-67
       September 2011

-------
 Study
Population
O3Lag
Subgroup
 Individuals with asthma
 Liu et al. (2009)         Children with asthma 0
 Barraza-Villarreal et al.
       (2008)
Children
 Berhane et al. (2011)    Children
                 0           Without asthma
                             With asthma

                 0-22 cum avg  Without asthma
                             With asthma
                             Without allergy
                             With allergy
 Qian et al. (2009)
 Khatri et al. (2009)
Children and adults
   with asthma
                     Adults with asthma   2
 Older adults
 Adamkiewicz et al. (2004) Olderadults
 Delfino etal.(2010)
                     Olderadults
                 0-4 avg
            Cool season
            Warm season
                                                         -20    -10
                                                                            10     20     30    40    50
                                                          Percent change in eNO per standardized increment in 03
                                                                             (95% Cl)
  Results are presented first for children with asthma followed by results for adults with asthma and older adults. Effect estimates
are from single-pollutant models and are standardized to a 30-ppb increase for 8-h max or 8-h avg ozone exposures and a 20-ppb
increase for 24-h avg ozone exposures.
Figure 6-10    Percent change in exhaled nitric oxide (eNO) per standardized
                  increment in ambient ozone exposure in studies of individuals with
                  and without asthma.
Draft - Do Not Cite or Quote
                            6-68
                                                  September 2011

-------
Table 6-15     Additional characteristics and quantitative data for studies
                  represented in Figure 6-10
Study
Location/
Population
Os Lag Os Averaging Time
Subgroup
Standardized
Percent Change
(95% Cl)a
Studies in individuals with asthma
Liu et al. (2009a)
Barraza-Villarreal et al. (2008)
Berhaneetal. (2011)
Qianetal. (2009)
Khatri et al. (2009)
Windsor, ON, Canada
Children with asthma
Mexico City, Mexico
Children
12 Southern California communities
Children
6 U.S. communities
Children and adults with asthma
Atlanta, GA
Adults with asthma
0 24-h avg
0 8-h max
0-22 cum avg 8-h avg
(10:00a.m.-6:00p.m.)
0 8-h max
2 8-h max

Without asthma
With asthma
Without asthma
With asthma
Without allergy
With allergy


-17.0 (-30.3, -1.1)
13.5(11.2, 15.8)
6.2 (6.0, 6.5)
30.1 (10.6, 53.2)
26.0 (-1.4, 60.9)
25.5 (5.3, 49.6)
32.1 (12.0, 55.9)
-1.2 (-1.7, -0.64)
36.6(1.2,71.9)
Studies in older adults
Adamkiewicz et al. (2007)
Delfinoetal. (2010a)
Steubenville, Ohio
Older adults
Los Angeles, CA
Older adults
0 24-h avg
0-4 avg 24-h avg

Cool season
Warm season
-5.7 (-25.9, 14.5)
23.6 (7.3, 39.9)
-0.58 (-13.4, 12.3)
  'Effect estimates are standardized to a 30-ppb increase for 8-h max or 8-h avg 03 and a 20-ppb increase for 24-h avg 03.
Table 6-16     Associations between short-term ambient ozone exposure and
                  biological markers of pulmonary inflammation and oxidative stress
Study
Liu et al. (2009a)
Romieu et al. (2008)
Barraza-Villarreal etal.
(2008)
Sienra-Monge et al.
(2004)
Khatri etal. (2009)
Ferdinands etal.
(2008)
Location/
Population
Windsor, ON, Canada
Children with asthma
Mexico City, Mexico
Children with asthma
Mexico City, Mexico
Children
Mexico City, Mexico
Children with asthma
Atlanta, GA
Adults with asthma
Atlanta, GA
Children exercising
outdoors
Os Os Averaging
Lag Time
0 24-h avg
0 8-h max
0 8-h max
0-2 avg 8-h max
2 8-h max
0 1-hmax
Biological Marker
EEC 8-isoprostane (%
change)
EEC TEARS (%
change)
EEC MDA"
Nasal lavage IL-8
(pg/ml)
Nasal lavage IL-8"
Nasal lavage IL-6b
Nasal lavage Uric acidb
Nasal lavage GSxb
Blood eosinophils (%
change)
EBCpH
Subgroup


Without asthma
With asthma
Without asthma
With asthma
Placebo
Antioxidant
Placebo
Antioxidant
Placebo
Antioxidant
Placebo
Antioxidant


Effect Estimate
(95% Cl)a
10.2 (-9.2, 33.5)
7.2 (-18.3, 40.7)
1 .3 (1 .0, 1 .7)
1 .6 (1 .4, 1 .8)
1 .6 (1 .4, 1 .9)
-0.10 (-0.27, 0.08)°
-0.10 (-0.20, 0.01)°
1.4 1.0,2.0)
1 .0 0.70, 1 .5)
1.5 1.2,2.0)
1 .0 (0.76, 1 .4)
0.88(0.70,1.1)
1.1 (0.84,1.5)
0.90 (0.82, 0.99)
0.91 (0.83, 0.98)
2.4 (0.62, 4.2)
2.5 (-0.20, 5.1)c
  EEC = exhaled breath condensate, TEARS = thiobarbituric acid reactive substances, MDA = malondialdehyde, IL-8 = interleukin 8, IL-6 =
interleukin 6, Antioxidant=group supplemented with vitamins C and E, GSx = glutathione.
  'Effect estimates are standardized to a 40-, 30- and 20-ppb increase for 1-h max, 8-h max, and 24-h avg 03, respectively
  bModels analyzed log-transformed biological markers. Therefore, effect estimates represent the ratio of the geometric means of biological
markers for a standardized increase in 03 exposure. An estimate less than 1 indicates a decrease in pulmonary inflammation or oxidative stress for
an increase in 03 exposure, and an estimate greater than 1 indicates an increase in pulmonary inflammation or oxidative stress for an increase in
03 exposure.
  °Negative and positive effect estimates indicate increases and decreases in pulmonary inflammation, respectively.
Draft - Do Not Cite or Quote
6-69
September 2011

-------
 1                   Despite the strengths of biomarker studies, it is important to note that research in this
 2                   field continues to develop, and several uncertainties are recognized that may limit the
 3                   interpretations of associations between ambient O3 exposure and changes in biomarker
 4                   levels. Current areas of development include examination of the clinical relevance of the
 5                   observed magnitudes of changes in biological markers of pulmonary inflammation
 6                   (Murugan et al.. 2009; Duramad et al.. 2007). characterization of the time course of
 7                   changes between biomarker levels and other endpoints of respiratory morbidity,
 8                   development of standardized methodologies for collection, improvement of the
 9                   sensitivity and specificity of assay methods, and characterization of subject factors  (e.g.,
10                   asthma severity and recent medication use) that contribute to inter-individual variability.
11                   These sources of uncertainty may  contribute to differences in findings among studies.

12                   In recent epidemiologic studies, the biomarker most frequently measured was exhaled
13                   nitric oxide (eNO), likely related to its ease of collection in the field and automated
14                   measurement. Other biological media analyzed included EEC, induced sputum, and nasal
15                   lavage fluid, all of which are hypothesized to contain aerosolized particles and/or cells
16                   from fluid lining the lower and upper airways (Balbi et al., 2007; Howarth et al., 2005;
17                   Hunt. 2002). These fluids contain  cytokines, cells, and markers of oxidative stress that
18                   mediate inflammatory responses. In particular, several of the cytokines, cells, and
19                   markers of oxidative stress examined in epidemiologic studies also have been examined
20                   in controlled human exposure and toxicological studies.Table 6-14 presents the
21                   characteristics and ambient O3 concentration data from recent studies assessing
22                   associations between O3 exposure and  biological markers of pulmonary inflammation and
23                   oxidative stress. Many recent studies reported positive associations between short-term
24                   ambient O3 exposure and increases in pulmonary inflammation and oxidative stress, in
25                   particular, studies of children with asthma conducted in Mexico City (Figure 6-10 and
26                   Tables 6-15 and 6-16).


                     Populations with Asthma

                         Exhaled Nitric Oxide
27                   Nitric oxide or eNO has not been examined in controlled human exposure  or
28                   toxicological studies of O3 exposure. However, several lines of evidence support its
29                   analysis as an indicator of pulmonary inflammation in epidemiologic studies. Inducible
30                   nitric oxide synthase can  be activated by pro-inflammatory cytokines, and NO can be
31                   produced by cells such as neutrophils,  eosinophils, and epithelial cells in the lung during
32                   an inflammatory response (Barnes and Liew. 1995). Additional support is  provided by
33                   observations of higher eNO in individuals with asthma, and increases in the levels during
34                   acute exacerbations (Jones et al.. 2001; Kharitonov and Barnes. 2000).
      Draft - Do Not Cite or Quote                       6-70                                 September 2011

-------
 1                   As indicated in Figure 6-10 and Table 6-15, several studies found that short-term ambient
 2                   O3 exposure (8-h max or avg) was associated with increases in eNO in children with
 3                   asthma. Liu et al. (2009a) (described in Section 6.2.1.2) reported O3-associated decreases
 4                   in eNO; however, this study was restricted to winter months. In this study, results for
 5                   EEC levels of TEARS and 8-isoprostane as well as lung function did not provide strong
 6                   evidence of O3 effects on airway oxidative stress.

 7                   The two studies that compared children with and without asthma did not find larger O3-
 8                   associated increases in eNO in children with asthma (Figure 6-10 and  Table 6-15).
 9                   Among children in Southern California, Berhane et al. (2011) examined a 0-22 day
10                   cumulative average of 8-h avg (10:00 a.m.-6:00 p.m.) O3 and estimated similar effects for
11                   children with and without asthma and children with and without allergy. Among children
12                   in Mexico City, Barraza-Villarreal et al. (2008) examined lag 0 of 8-h max O3 and
13                   estimated larger effects for children without asthma.

14                   In the two studies that included adults with asthma, ambient O3 exposure was associated
15                   with both decreases and increases in eNO (Khatri et al.. 2009; Qian et al.. 2009). In the
16                   multicity salmeterol ((3-2 adrenergic agonist) trial (Boston, MA; New York, NY; Denver,
17                   CO; Philadelphia, PA; San Francisco, CA; and Madison, WI), eNO was collected every
18                   2-4 weeks over a 16-week period from 119 subjects with persistent asthma, 87% of
19                   whom were 20-65 years of age (Qian et al.. 2009). Among all subjects, lag 0 of 8-h max
20                   O3 was associated with a decrease in eNO as were exposures lagged 1  to 3  days and
21                   averaged over 5 days. Results did not vary among the salmeterol, CS,  and placebo
22                   groups, indicating that the counterintuitive findings for O3 were not simply due to the
23                   reduction of inflammatory responses by medication use. The authors suggested that at
24                   higher O3 exposures, O3 may rapidly react with NO in airways to form reactive nitrogen
25                   species such as peroxynitrite. However, in the other study of adults with asthma, ambient
26                   concentrations of 8-h max O3 were higher, and a positive association was found with
27                   eNO (Khatri et al.. 2009). In this study conducted during a summer season in Atlanta,
28                   GA, a 30-ppb increase in lag 2 of 8-h max O3 was associated with a 36.6% increase in
29                   eNO (95% CI:  1.2, 71.9). These findings should be interpreted with caution as they were
30                   based on a single eNO measurement per subject and were not adjusted for  any
31                   meteorological factors.

                        Other biological markers of pulmonary inflammation and oxidative stress
32                   As indicated in Table 6-16, studies have found associations between short-term ambient
33                   O3 exposure and changes in the levels of proinflammatory cytokines and cells, indicators
34                   of oxidative stress, and antioxidants. Importantly, any particular endpoint was examined
35                   only in one to two studies, and the evidence in individuals with asthma is derived
36                   primarily from studies conducted in Mexico City (Romieu et al.. 2009; Barraza-Villarreal
      Draft - Do Not Cite or Quote                       6-71                                 September 2011

-------
 1                   et al.. 2008; Romieu et al., 2008; Sienra-Monge et al., 2004). Despite the limited
 2                   evidence, the epidemiologic observations are well-supported by controlled human
 3                   exposure and toxicological studies that have measured analogous endpoints.

 4                   Several of the modes of action of O3 are mediated by secondary oxidation products
 5                   produced in the airways by O3 (Section 5.3.3). Reactive oxygen species (ROS) are
 6                   involved in the regulation of inflammation via regulation of the expression of cytokines
 7                   and activity of inflammatory cells in airways (Heidenfelder et al.. 2009). In controlled
 8                   human exposure and toxicological studies, prostaglandins have been frequently measured
 9                   to indicate O3-induced increases in oxidative stress (Sections 5.3.3 and 6.2.3.1).
10                   Prostaglandins are produced by the peroxidation of arachidonic acid in cell membranes
11                   (Morrow et al.. 1990). Romieu et al. (2008) analyzed biweekly samples of EEC
12                   malondialdehyde (MDA), a thiobarbituric acid reactive substance, which like
13                   prostaglandins, is derived from oxidative  degradation of lipids (Janero.  1990). The ratio
14                   of the geometric means of MDA was 1.3 (1.0, 1.7) per a 30-ppb increase in lag 0 of 8-h
15                   max O3. Similar results were reported for lags 1 and 0-1 average exposures. An important
16                   limitation of the study was that 25% of EEC samples had nondetectable levels of MDA.
17                   Thus, the random assignment of concentrations between 0 and 4.1 nmol may have
18                   contributed random measurement error to the estimated O3 effects. Because MDA
19                   represents less than 1% of lipid peroxides and is present at low concentrations, its
20                   reliability as a marker of oxidative stress in vivo has been questioned. However, Romieu
21                   et al. (2008) pointed to their observations of statistically significant associations of EEC
22                   MDA levels with nasal lavage IL-8 levels to support its analysis as a biologically-
23                   relevant indicator of pulmonary inflammation.

24                   Uric acid and glutathione are ROS scavengers that are present in the airway ELF. While
25                   the roles of these markers in the inflammatory cascade of asthma are not well
26                   characterized, they are observed to be consumed in response to short-term O3 exposure in
27                   controlled human exposure and animal studies (Section 5.3.3). Results from an
28                   epidemiologic study also indicate that ambient O3 exposure may stimulate an antioxidant
29                   response. In a panel study with three measurements of nasal lavage at 3-week intervals,
30                   Sienra-Monge et al. (2004) found O3-associated decreases in nasal lavage levels of uric
31                   acid and glutathione in children with asthma not supplemented with antioxidant vitamins
32                   (Table 6-16). The magnitude of association was similar for O3 exposures lagged 2 or 3
33                   days and averaged over 3 days.

34                   Both controlled human exposure and toxicological studies find O3-induced increases in
35                   the cytokines IL-6 and IL-8 (Sections 5.3.3, 6.2.3.1, and 6.3.3.3), which are involved in
36                   initiating an influx of neutrophils, a hallmark of inflammation induced by short-term O3
37                   exposure. Recent epidemiologic studies produced similar findings. Barraza-Villarreal
      Draft - Do Not Cite or Quote                       6-72                                September 2011

-------
 1                   et al. (2008) observed that a 30-ppb increase in lag 0 of 8-h max O3 was associated with a
 2                   1.61 pg/ml increase (95% CI: 1.4, 1.8) in IL-8. In another study of children with asthma
 3                   in Mexico City, Sienra-Monge et al. (2004) found that lags 2, 3, and 0-2 avg of 8-h max
 4                   O3 were associated with increases in nasal lavage levels of IL-6 and IL-8 (placebo group),
 5                   with the largest effects estimated for lag 0-2 average exposure (Table 6-16).

 6                   Neutrophil influx has been a prominent characterisitc of O3-induced inflammation;
 7                   however, controlled human exposure studies also have found O3-induced increases in
 8                   eosinophils in adults with asthma (Section 6.2.3.1). Eosinophils are believed to be the
 9                   main effector cells that initiate and sustain inflammation in asthma and allergy (Schmekel
10                   et al., 2001). Consistent with these findings, in a cross-sectional study of adults with
11                   asthma in Atlanta, GA, a 30-ppb increase in lag 0 of 8-h max O3 was associated with a
12                   2.4% increase (0.62, 4.2) in blood eosinophils (Khatri et al.. 2009). These results were
13                   not adjusted for meteorological factors.

14                   The pH of EEC also was analyzed as an indicator of pulmonary inflammation. EEC pH is
15                   thought to reflect the proton-buffering capacity of ammonium in airways. It has been
16                   widely used in the clinical assessment of asthma, is consistently lower in subjects with
17                   asthma, decreases upon acute asthma exacerbation (on the order of 2 units), and is
18                   negatively correlated with airway levels of proinflammatory cytokines (Carpagnano et
19                   al..  2005: Kostikas et al.. 2002: Hunt et al.. 2000). In addition to finding O3-associated
20                   increases in eNO and nasal lavage IL-8, Barraza-Villarreal et al. (2008) found small O3-
21                   associated decreases in EEC pH (Table 6-16).

22                   The prominent influences of ROS and antioxidants in mediating the effects of O3 provide
23                   biological plausibility for the effect modification by antioxidant supplementation. The
24                   modulation of O3-associated lung function by antioxidant capacity has been described in
25                   controlled human exposure and epidemiologic studies (Sections 6.2.1.1 and 6.2.1.2).
26                   Epidemiologic studies also found that higher levels of dietary or supplemented
27                   antioxidants attenuated inflammation and oxidative stress. Sienra-Monge et al. (2004)
28                   conducted a 12 week-trial with daily vitamin C and E supplements. In the antioxidant
29                   group, the ratios of the geometric means of nasal lavage IL-6 and IL-8 per 30-ppb
30                   increases in lag 0-2 avg of 8-h max O3 were 1.0, reflecting no increases with increases in
31                   O3 exposure (Table 6-16). Effect modification by antioxidant supplementation was not
32                   consistent for uric acid and glutathione (Table 6-16). Ozone was associated with
33                   increases in uric acid in the antioxidant group and decreases in the placebo group across
34                   O3 lags of exposure. Associations with glutathione were similar in both groups.
3 5                   Therefore, the results do not clearly delineate the interactions among inhaled O3,
36                   endogenous antioxidants, and dietary supplementations of antioxidants. In another cohort
37                   of children with asthma in Mexico City, a diet high in fruits and vegetables was found to
      Draft - Do Not Cite or Quote                       6-73                                September 2011

-------
 1                   protect against O3-related increases in nasal lavage IL-8 (Romieu et al., 2009). At high
 2                   ambient O3 levels (> 38 ppb, 8-h max), a 1-unit increase in FVI was associated with a
 3                   0.219 pg/ml decrease (95% CI: -0.38, -0.05) in IL-8. The protective effect was
 4                   diminished by about 49% at O3 levels of 25 ppb or lower. Results from these two studies
 5                   indicate that augmenting the circulating levels of antioxidants, through diet or vitamin
 6                   supplements, may reduce nasal inflammation associated with high ambient O3 exposures.

                        Clinical significance of ozone-associated changes in pulmonary
                        inflammation and oxidative stress in children with asthma
 7                   While the results of epidemiologic studies in children with asthma were consistent with
 8                   the known modes of action of O3 in consuming antioxidants and inducing oxidative stress
 9                   and pulmonary inflammation (Section 5.3.3), the clinical significance of these changes
10                   has not been well-characterized. The levels of several of the biological markers such as
11                   eNO, EEC pH, and MDA have been shown to differ between subjects with and without
12                   asthma and change acutely during an acute asthma exacerbation (Corradi et al., 2003;
13                   Hunt et al. 2000): however, the magnitudes of change for these conditions are not well-
14                   defined. Several studies conducted in individuals with asthma found large O3-associated
15                   increases in eNO; effect estimates ranged between a 6 and 36% increase per standardized
16                   increment in ambient O3 concentration1 (Figure 6-10 and Table 6-15). Standardized
17                   increments in ambient O3 exposure were associated with smaller (1-2%) increases in
18                   interleukins or indicators of oxidative stress  (Khatri et al., 2009; Barraza-Villarreal et al..
19                   2008) (Romieu et al.. 2008; Sienra-Monge et al.. 2004).

20                   Some studies permitted the evaluation of the potential clinical relevance of these changes
21                   in eNO through the concurrent assessment of respiratory symptoms. Among children
22                   with asthma in Mexico City, O3 exposure was associated with increases in eNO and nasal
23                   lavage IL-8 and concurrently assessed cough but not wheeze (Barraza-Villarreal et al.,
24                   2008). Among adults with asthma in Atlanta, O3 was associated with increases in eNO,
25                   blood eosinophils, and a decrease in quality of life score, which incorporates indices for
26                   symptoms, mood, and activity limitations (Khatri  et al.. 2009). These findings suggest
27                   that the more subtle O3-associated increases in biological markers of airway
28                   inflammation may be sufficient to result in respiratory symptoms or activity limitations.


                     Children  without Asthma

29                   Recent studies found that short-term O3 exposure  (8-h max or avg) was associated with
30                   indicators of airway inflammation in children without asthma (Berhane et al., 2011;
31                   Barraza-Villarreal et al.. 2008) (Figure 6-10 and Tables 6-15 and 6-16). In the panel
        1 Effect estimates were standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max, and 24-h avg O3.
      Draft - Do Not Cite or Quote                       6-74                                September 2011

-------
 1                   study of children in Mexico City, O3 exposure was associated with a larger increase in
 2                   eNO in the children without asthma than with asthma (13.5% versus 6.2% increase per
 3                   30-ppb increase in lag 0 of 8-h max O3). Ozone was associated with similar magnitudes
 4                   of changes in IL-8 and EEC pH in children with and without asthma. A distinguishing
 5                   feature of this study was that most of the children without asthma were atopic (72%) as
 6                   indicated by positive skin prick tests, which may have contributed to the similar effects of
 7                   O3 exposure observed in children with and without asthma.

 8                   However, the Southern California Children's Health Study estimated similar effects for
 9                   8-h avg (10:00 a.m.-6:00 p.m.) ambient O3 exposure on eNO in children with and without
10                   respiratory allergy (Berhane et al., 2011). Results from this large study (n = 2240
11                   children)  provided evidence that ambient O3 exposure increases airway inflammation in
12                   healthy children. In comparison with other studies, this analysis from the Children's
13                   Health Study provided detailed information on differences in association among various
14                   lags of 8-h avg (10:00 a.m.-6:00 p.m.) O3 exposure. Consistent with other studies
15                   examining pulmonary inflammation and oxidative stress, Berhane et al. (2011) found that
16                   relatively shorter lags of exposure, including  1 to  5 days, were associated with increases
17                   in eNO. However, in an examination of several types of lag-based models, including
18                   unconstrained lag models, polynomial distributed lag models, spline-based distributed lag
19                   models, and cumulative lag models, investigators found that a 23-day cumulative lag
20                   model best fit the data. Among the studies evaluated in the current assessment, Berhane
21                   et al. (2011) was unique in evaluating and finding larger effects for cumulative average
22                   O3 exposures over multiple weeks (e.g., 13-30 days). O3 exposures averaged over the
23                   several hours preceding eNO collection were not significantly associated with eNO. The
24                   mechanism for the effects  of O3 peaking with a 23-day cumulative lag of exposure is not
25                   known.


                     Populations with Increased Outdoor Exposures

26                   In a limited number of available studies, ambient O3 exposure was not consistently
27                   associated with pulmonary inflammation in populations with increased outdoor
28                   exposures. Important limitations of these studies include small numbers of subjects and
29                   repeated measurements. In a cross-sectional study of children at camps in south Belgium,
30                   although O3 was not associated with lung function, an association was found for eNO
31                   (Nickmilder et al.. 2007). Children at camps with lag 0 1-h max O3 concentrations above
32                   85.2 ppb had greater increases in intraday eNO  compared with children at camps with O3
33                   concentrations below 51 ppb. A benchmark dose analysis indicated that the threshold for
34                   an O3-induced increase of 4.3 ppb eNO (their definition of increased pulmonary
35                   inflammation) was 68.6 ppb for 1-h max O3 and 56.3 ppb for 8-hr max O3. While these
36                   results provide additional evidence for O3-associated increases in airway inflammation in

      Draft - Do Not Cite or Quote                       6-75                                September 2011

-------
 1                   healthy children, they should be interpreted with caution since they were not adjusted for
 2                   any potential confounding factors.

 3                   Recent studies examined associations of O3 exposure with biological markers of airway
 4                   inflammation in populations exercising outdoors. In a panel study of 16 adolescent long-
 5                   distance runners in Atlanta, GA, lags 0, 1, and 2 of 1-h max O3 were associated with
 6                   increases in EEC pH, indicating O3-associated decreases in pulmonary inflammation
 7                   (Ferdinands et al.. 2008). Among 9 adult male runners in Sicily, Italy examined 3 days
 8                   before and 20 hours after 3 races in fall, winter, and summer, weekly average O3
 9                   concentrations (8-h avg, 7:00 a.m.-3:00 p.m.) were positively correlated with apoptosis of
10                   airway cells (Spearman's r = 0.76, p < 0.0005) and bronchial epithelial cell differential
11                   counts (Spearman's r = 0.467, p < 0.05) but not with neutrophil or macrophage cell
12                   counts or levels of the proinflammatory cytokines TNF-a and IL-8 (Chimenti et al..
13                   2009). Although this study provides evidence for some new endpoints, the implications
14                   of the findings are limited since they were not based on a rigorous statistical analysis.


                     Older Adults

15                   Two panel studies examining O3-associated changes in eNO in older adults produced
16                   contrasting findings (Figure  6-10 and Table 6-15). Both studies were similar in that
17                   outdoor  O3 was monitored by investigators in the vicinity of subjects' residences, and
18                   cool season-specific results were presented. However, several differences were
19                   noteworthy, including geographic location, inclusion of healthy subjects, and lags of O3
20                   exposure examined. Delfino et al. (2010a) followed 60 elderly subjects with coronary
21                   artery disease in the Los Angeles, CA area for two 6-week periods, one in the warm
22                   season and one in the cool season, although the exact months were not specified.
23                   Multiday averages of O3 (3-  to 9-day) were associated with increases in eNO, with effect
24                   estimates increasing with increasing number of averaging days. In contrast with most
25                   other studies, a strong positive effect was estimated for the cooler season (4.06 ppb [95%
26                   CI:  1.25, 6.87]) increase in eNO per 20-ppb increase in lag 0-4 of 24-h avg O3), whereas
27                   no association was observed for the warm season (-0.01 ppb change in eNO [95% CI: -
28                   2.31, 2.11]). Despite these unusual findings for the cool season, they were similar to
29                   findings from another study  of Los Angeles area adults with asthma that found O3 effects
30                   (i.e., decrease in indoor activity) during the fall season (Eiswerth et al.. 2005).

31                   Adamkiewicz et al. (2004) did not find a positive association between O3 exposure and
32                   eNO in a group of older adults (ages 54-91 years) comprising healthy subjects and those
33                   with asthma or COPD. The study was conducted in Steubenville, OH between September
34                   and December, and as was observed in most other studies conducted during winter
35                   months,  O3 (concurrent 1 hour and 24 hours preceding eNO collection) was associated
      Draft - Do Not Cite or Quote                       6-76                                September 2011

-------
 1                  with decreases in eNO, indicating a decrease in pulmonary inflammation (Figure 6-10
 2                  and Table 6-16).
                    Confounding in Epidemiologic Studies of Pulmonary Inflammation and
                    Oxidative Stress

 3                  Except where noted in the preceding text, most epidemiologic studies of pulmonary
 4                  inflammation and oxidative stress accounted for the potential for confounding by
 5                  meteorological factors. Ambient O3 exposure was associated with pulmonary
 6                  inflammation or oxidative stress in models that adjusted for temperature and/or humidity
 7                  (Delfinoetal..2010a: Barraza-Villarreal et al.. 2008; Romieu et al.. 2008). Most studies
 8                  conducted over multiple seasons adjusted for season or time trend. Sienra-Monge et al.
 9                  (2004) and Berhane et al. (2011) did not adjust for temperature in their final results after
10                  finding that the inclusion of temperature did not change results.

11                  Although information is limited to a small number of studies conducted in Mexico City,
12                  the evidence does not indicate the confounding of O3 associations by PM2 5 or PM10
13                  exposure. In these studies, which analyzed 8-h averages for both O3 and PM and reported
14                  moderate correlations between pollutants (r=0.46-0.54), robust associations were found
15                  for O3 (Barraza-Villarreal et al.. 2008: Romieu et al.. 2008: Sienra-Monge et al.. 2004).
16                  Only Romieu et al. (2008) provided quantitative results. Lag 0 of 8-h max O3 was
17                  associated with the same magnitude of increase in MDA with and without lag 0 of 8-h
18                  max PM25 in the model (ratio of geometric means per 30-ppb increase: 1.3 [95% CI: 1.0,
19                  1-7]). In the copollutant model, the effect estimate for PM2s was cut in half.


                    Summary of Epidemiologic Studies of Pulmonary Inflammation and
                    Oxidative Stress

20                  Many recent epidemiologic studies reported positive associations between short-term
21                  ambient O3 exposure  and increases in pulmonary inflammation and oxidative stress,
22                  particularly, studies of children with asthma in Mexico City. By also finding that O3-
23                  associated increases in pulmonary inflammation were attenuated with higher antioxidant
24                  intake, these studies, as a whole, provided evidence that inhaled O3 may be an important
25                  source of ROS in airways and/or may increase airway inflammation via oxidative stress-
26                  mediated mechanisms. Studies also indicated that ambient O3 exposure may increase
27                  airway inflammation  in healthy children (Berhane et al.. 2011: Nickmilder et al.. 2007).
28                  The limited available evidence in subjects exercising outdoors and older adults was
29                  inconclusive. Temperature and humidity were not found to confound O3 associations, and
30                  in the few studies that evaluated copollutant models, O3 effect estimates were robust to
      Draft - Do Not Cite or Quote                      6-77                                September 2011

-------
 1                   inclusion of PM7 s or PMin (Barraza-Villarreal et al., 2008; Romieu et al., 2008; Sienra-
 2                   Mongeetal.. 2004).

 3                   Most studies examined associations with daily 8-h max or daytime 8-h avg O3 exposures,
 4                   although associations were observed for 1-h max (Nickmilder et al., 2007) and 24-h avg
 5                   O3 exposures (Delfino et al.. 2010a). Collectively, studies examined associations with
 6                   single-day O3 exposures lagged from 0 to 5 days, and exposures averaged over 2 to 9
 7                   days. Lag 0 of 8-h max O3 exposure was most frequently examined and consistently
 8                   associated with increased airway inflammation and oxidative stress. However, in the few
 9                   studies that examined multiple lags of exposure, multiday cumulative O3 exposures,
10                   primarily based on 8-h max or 8-h avg, were associated with greater increases in airway
11                   inflammation and oxidative stress (Berhane et al.. 2011; Delfino et al..  2010a: Sienra-
12                   Monge etal., 2004). These findings for longer lags of exposure are supported by
13                   controlled human exposure studies that similarly have found that indicators of airway
14                   inflammation remain elevated following exposures to O3 repeated over multiple days
15                   (Section 6.2.3.1).

16                   Several epidemiologic studies simultaneously examined associations of ambient O3
17                   exposure with biological markers of airway inflammation and oxidative stress, lung
18                   function, and respiratory symptoms. In most cases, the results differed between the
19                   various biomarkers  and lung function. Whether evaluated at the same or different lags of
20                   O3 exposure, associations generally were stronger for biological markers of airway
21                   inflammation than for lung function (Barraza-Villarreal et al.. 2008; Nickmilder et al..
22                   2007). Controlled human exposure  studies also have demonstrated a lack of correlation
23                   between inflammatory and spirometric responses induced by O3 exposure. Studies have
24                   suggested that O3-related respiratory morbidity may occur via multiple mechanisms with
25                   varying time courses of action, and the examination of a limited number of O3 exposure
26                   lags in these aforementioned studies may explain some  of the  inconsistencies in
27                   associations of O3 with different respiratory health endpoints.

28                   The clinical significance of changes in biological markers of airway inflammation and
29                   oxidative stress are  not well-characterized.  However, the simultaneous examination of
30                   associations of O3 with respiratory symptoms has permitted the assessment of the clinical
31                   significance of the changes observed in biomarkers. In subjects with asthma, ambient O3
32                   exposure was associated with increases in eNO and IL-6 that were accompanied by a
33                   concomitant increase in cough (Barraza-Villarreal et al., 2008) and increases in eNO and
34                   blood eosinophils that were accompanied by a decrease in quality of life score (Khatri et
35                   al., 2009). These findings support clinically-important increases in O3-associated airway
36                   inflammation in individuals with asthma. Similar data are limited to assess the clinical
      Draft - Do Not Cite or Quote                       6-78                                 September 2011

-------
 1                   significance of changes in other biological markers of airway inflammation and oxidative
 2                   stress and in other populations.
                     6.2.3.3    Toxicology

 3                   The 2006 O3 AQCD states that the "extensive human clinical and animal toxicological
 4                   evidence, together with the limited available epidemiologic evidence, is clearly indicative
 5                   of a causal role for O3 in inflammatory responses in the airways" (U.S. EPA. 2006b).
 6                   Airway ciliated epithelial cells and Type 1 cells are the most O3-sensitive cells and are
 7                   initial targets of O3. These cells are damaged by O3 and produce a number of
 8                   proinflammatory mediators (e.g., interleukins [IL-6, IL-8], PGE2) capable of initiating a
 9                   cascade of events leading to PMN influx into the lung, activation of alveolar
10                   macrophages, inflammation, and increased permeability across the epithelial barrier. One
11                   critical aspect of inflammation is the potential for metaplasia and alterations in
12                   pulmonary morphology. Studies have observed increased thickness of the alveolar septa,
13                   presumably due to increased cellularity after acute exposure to O3. Epithelial hyperplasia
14                   starts early in exposure and increases in magnitude for several weeks, after which it
15                   plateaus until exposure ceases. When exposure persists for a month and longer, excess
16                   collagen and interstitial fibrosis are observed. This response, discussed in Chapter 7,
17                   continues to increase in magnitude throughout exposure and can even continue to
18                   increase after exposure ends (Last et al.. 1984). Previously published toxicological
19                   studies of the ability of O3 to cause inflammation, injury, and morphological changes are
20                   described in Table 6-5 on p. 6-25 and Tables 6-10 and 6-11 beginning on p. 6-61 of the
21                   1996 O3 AQCD, and Tables AX5-8 and AX5-9, beginning on p. AX5-17 of the 2006 O3
22                   AQCD. Numerous recent in vitro and in vivo studies add to this very large body of
23                   evidence for O3-induced inflammation and injury, and provide new information regarding
24                   the underlying mechanisms (Bauer et al.. 2011; Aibo et al.. 2010; Farraj et al.. 2010;
25                   Garantziotis et al.. 2010; Hicks etal.. 2010b: Castagna et al.. 2009;  Damera et al.. 2009;
26                   Oslund et al.. 2009: Vancza et al.. 2009: Vovnow et al.. 2009: Fakhrzadeh et al.. 2008:
27                   Han et al.. 2008: Inoue et al.. 2008: Oslund et al.. 2008: Carey et al.. 2007: Cho et al..
28                   2007: Dahl et al.. 2007: Johnston et al.. 2007: Kooter et al.. 2007: Wagner etal.. 2007:
29                   Wang et al.. 2007: Yoon et al.. 2007: Huffman et al.. 2006: Johnston et al.. 2006: Kenyon
30                   et al.. 2006: Manzer et al.. 2006: Plopper et al.. 2006: Jang et al.. 2005: Janic et al.. 2005:
31                   Johnston et al.. 2005a: Johnston et al.. 2005b: Oyarzun etal.. 2005: Servais etal.. 2005:
32                   Frush et al.. In Press).

33                   A number of species, including dogs, rabbits, guinea pigs, rats, and mice have been used
34                   as models to study the pulmonary effects of O3, but the similarity of non-human primates
35                   to humans makes them an attractive model in which to study the pulmonary response to


      Draft - Do Not Cite or Quote                       6-79                                 September 2011

-------
 1                   O3. As reviewed in the 1996 and 2006 O3 AQCDs, several pulmonary effects, including
 2                   inflammation, changes in morphometry, and airway hyperresponsiveness, have been
 3                   observed in macaque and rhesus monkeys after acute exposure to O3 (Table 6-17 presents
 4                   a highlight of these studies). Increases in inflammatory cells were observed after a single
 5                   8-hr exposure of adult rhesus monkeys to 1  ppm O3 (Hyde et al., 1992). Inflammation
 6                   was linked to morphometric changes, such as increases in necrotic cells, smooth muscle,
 7                   fibroblasts, and nonciliated bronchiolar cells, which were observed in the trachea,
 8                   bronchi, or respiratory bronchioles. Effects  have also been observed after short-term
 9                   repeated exposure to O3 at concentrations that are more relevant to ambient O3 levels.
10                   Morphometry changes in the lung, nose, and vocal cords were observed after exposure to
11                   0.15 ppm O3 for 8-h/day for 6 days (Harkema et al..  1993; Dimitriadis. 1992; Harkema et
12                   al.. 1987a). Since 2006, however, only one  study has been published regarding acute
13                   exposure of non-human primates to O3 (a number of recent chronic studies in non-human
14                   primates are described in Chapter 7). In this study, a single 6-h exposure of adult male
15                   cynomolgus monkeys to 1 ppm O3 induced  significant increases in inflammatory and
16                   injury markers, including BAL neutrophils, total protein, alkaline phosphatase, IL-6, IL-
17                   8, and G-CSF (Hicks et al., 2010b). Gene expression analysis confirmed the increases in
18                   the pro-inflammatory cytokine IL-8, which  had been previously described in O3 exposed
19                   rhesus monkeys (Chang et al.. 1998). The anti-inflammatory cytokine IL-10 was also
20                   elevated, but the fold changes in IL-10 and  G-CSF were relatively low and highly
21                   variable. The single exposure also caused necrosis and sloughing of the epithelial lining
22                   of the most distal portions of the terminal bronchioles and the respiratory bronchioles.
23                   Bronchiolitis, alveolitis, parenchymal and centriacinar inflammation were also observed.
24                   A second exposure protocol (two exposures with a 2-week inter-exposure interval)
25                   resulted in similar inflammatory responses,  with the  exception of total protein and
26                   alkaline phosphatase levels which were attenuated, indicating that attenuation of some
27                   but not  all lavage parameters occurred upon repeated exposure of non-human primates to
28                   O3 (Hicks et al.. 201 Ob). This variability in  adaptation is similar to the findings of earlier
29                   reports  in rodents (Wiester et al.. 1996b) and non-human primates (Tyler etal.. 1988).
30                   Table 6-17 describes morphometric studies  conducted in non-human primates exposed
31                   toO3.
      Draft - Do Not Cite or Quote                      6-80                                September 2011

-------
      Table 6-17     Morphometric observations in non-human primates after acute O3
                      exposure
Reference
Harkema et al.
(1993)
Harkema et al.
(1987a; 1987b)
Dimitriadis (1992)
Leonard et al. (1991)
Chang et al. (1998)
Hyde et al. (1992)
Hicks et al. (201 Oa)
Os concentration
0.15
0.15
0.3
0.25
0.96
0.96
1.0
Exposure
duration
8 h/day for 6
days
8 h/day for 6
days
8 h/day for 7
days
8h
8h
6h
Species, Sex, Age
Macaca radiata
Macaca radiata, M, F
2-6 years old
Macaca radiata
Rhesus, M
Rhesus, M
2-8.5 years old
Cynomolgus, M
5-7 kg
Observation
Several fold increase in thickness of surface
epithelium in respiratory bronchioles
Ciliated cell necrosis, shortened cilia, and
increased mucous cells in the respiratory
epithelium of nose after 0.1 5 ppm; changes in
nonciliated cells, intraepithelial leukocytes, and
mucous cells in the transitional epithelium
The 03 exposure level is not clear - the abstract
states 0.64 ppm, but the text mentions only 0.25
ppm. Morphometric changes in vocal cord mucosa:
disruption and hyperplasia of stratified squamous
epithelium; epithelial and connective tissue
thickness increased
Increase in IL-8 in airway epithelium correlated with
PMN influx
Increased PMNs; morphometric changes in
trachea, conducting airways, respiratory
bronchioles
Increase in PMNs and IL-8 in lavage fluid
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
Confirmation of pulmonary changes observed in non-human primates, at near ambient
(^concentrations, has been done in a large number of studies in guinea pigs and rodents
(see 1996 and 2006 O3 AQCDs) (U.S. EPA. 2006b. 1996a). Mechanistic studies
completed more recently have extended these findings. Exposure of adult BALB/c mice
to 0.1 ppm O3 for 4 hours increased BAL levels of keratinocyte chemoattractant (KC; IL-
8 homologue) (~ sixfold), IL-6 (~12-fold), and TNF-a (~ twofold) (Damera et al.. 2010).
Additionally, O3 increased BAL neutrophils by 21% without changes in  other cell types.
A trend of increased neutrophils with increased O3 concentration (0.12-2 ppm) was
observed in BALB/c mice exposed for 3 hours (Jang et al.. 2005). Although alterations in
the epithelium of the airways were not evident in 129J mice after 4 hours of exposure to
0.2 ppm O3 (Plopper et al.. 2006). detachment of the bronchiolar epithelium was
observed in SD rats after 5 days or 60 days of exposure to 0.25 ppm O3 (Oyarzun et al..
2005). Subacute (65 hours) exposure to 0.3 ppm O3 induced pulmonary inflammation,
cytokine induction, and enhanced vascular permeability in wild type mice of a mixed
background (129/Ola and C57BL/6) and these effects were exacerbated in
metallothionein I/II knockout mice (Inoue et al.. 2008). Three hours or 72 hours of
exposure to 0.3 ppm O3 resulted in similar levels of IL-6 expression in the lungs of
C57BL/6 mice (Johnston et al.. 2005b). along with increases in BAL protein, sTNFRl,
and sTNFR2. Increased neutrophils were observed only after the 72-h exposure, and
neither exposure resulted in detectable levels of IL-6 or KC protein. Levels of BAL
      Draft - Do Not Cite or Quote
                              6-81
September 2011

-------
 1                   protein, sTNFRl, and sTNFR2 were higher in the 72-h exposure group than in the 3-h
 2                   exposure group. In another study, the same subacute (72 hours) exposure protocol elicited
 3                   increases in BALF protein, IP-10, sTNFRl, macrophages, neutrophils, and IL-6, IL-la,
 4                   and IL-1(3 expression (Johnston et al.. 2007). Yoon et al. (2007) exposed C57BL/6J mice
 5                   continuously to 0.3 ppm O3 for 6, 24, 48, or 72 hours, and observed elevated levels of
 6                   KC, MIP-2, metalloproteinases, and inflammatory cells in the lungs at various time
 7                   points. A similar exposure protocol using C3FI/HeJ and C3FI/OuJ mice demonstrated
 8                   elevations in protein, PMNs, and KC, which were predominantly TLR 4 pathway
 9                   dependent based on their prominence in the TLR 4 sufficient C3FI/OuJ strain (Bauer et
10                   al.. 2011). C3H/OuJ mice also had elevated levels of the heat-shock protein HSP70, and
11                   further experiments in HSP70 deficient mice indicated a role for this particular pathway
12                   in O3-related injury, discussed in more detail in Chapter 5.

13                   As reviewed in the 2006 O3 AQCD, the time course for changes in BAL depends on the
14                   parameters being studied. Similarly, after exposing adult C57BL mice to 0.5 ppm O3 for
15                   3 hours, Han et al. (2008) observed early (5 hours postexposure) increases in BAL TNF-a
16                   and IL-lp, which diminished by 24 hours postexposure. Total BAL protein was elevated
17                   at 24 hours, but there were only minimal or negligible changes in LDH, total cells, or
18                   PMNs. Ozone increased BAL mucin levels (with statistical significance by 24 hours
19                   postexposure), and significantly elevated surfactant protein D at both time points. Prior
20                   intratracheal (IT) exposure to multiwall carbon nanotubes  enhanced most of these effects,
21                   but the majority of responses to the combined exposure were not greater than those to
22                   nanotubes alone. Ozone exposure did not induce markers of oxidative stress in lung
23                   tissue, BAL, or serum. Consistent with this study, Aibo et al. (2010) did not detect
24                   changes in BAL inflammatory cell numbers in the same mouse strain after a 6-h exposure
25                   to 0.25 or 0.5 ppm. The majority of inflammatory cytokines (pulmonary or circulating)
26                   were not significantly changed (as assessed 9 hours post O3 exposure).

27                   Animal toxicology studies have also examined susceptibility factors and the findings
28                   complement research in both controlled human exposure and epidemiologic studies. In a
29                   study examining age, strain, and gender as factors for susceptibility to O3 in mice,
30                   increased BAL neutrophils were observed in all 8 strains of neonates and adults but
31                   statistical significance was found in only 4 strains of neonates and 2 strains of adults at
32                   24 hours after exposure to 0.8 ppm  O3 for 5 hours (Vancza et al., 2009). Lung injury, as
33                   measured by BAL protein, was significantly increased in 5 and 8 strains of neonates and
34                   adults, respectively. Interestingly, the observed age-dependent differences in response to
35                   O3 occurred in only certain strains. For example, the fold-increase in neutrophils was
36                   significantly higher, in neonates compared to adults, in the SJL and C3H/HeJ strains and
37                   lower in BALB/c mice. Measurement of 18O determined that the observed strain- and
38                   age-dependent differences were not due to absorbed O3 dose. Subanalysis of the adult
      Draft - Do Not Cite or Quote                       6-82                                September 2011

-------
 1                   mice demonstrated that gender also played a small, but statistically significant, role in the
 2                   effect of O3 on BAL neutrophils and protein. These findings suggest that the response to
 3                   O3, in mice, may consist of a complex interaction of age, gender, and genetic factors.

 4                   A study assessing NQO1 as a susceptibility factor was conducted by Voynow et al.
 5                   (2009). Specific effects of this gene on O3 responses are discussed in Chapter 8;  only
 6                   ozone's effects in wild type C57BL/6 mice are described here. Exposure to 1 ppm for
 7                   3 hours increased BAL total cells, neutrophils, and KC; these responses were greatest at
 8                   24 hours postexposure. F2-isoprostane (8-isoprostane), a marker of oxidative stress, was
 9                   also elevated by O3, peaking at 48 hours postexposure.

10                   Atopic asthma appears to be a risk factor for more severe O3 induced airway
11                   inflammation in humans (Balmes et al., 1997; Scannell et al., 1996). and allergic animal
12                   models are often used to investigate the effects of O3 on this susceptible population.
13                   Farraj et al. (2010) exposed allergen-sensitized adult male BALB/c mice to 0.5 ppm O3
14                   for 5 hours once per week for 4 weeks. Ovalbumin-sensitized mice exposed to O3 had
15                   significantly increased BAL eosinophils by 85% and neutrophils by 103% relative to
16                   OVA sensitized mice exposed to air, but these changes were not evident upon
17                   histopathological evaluation of the lung, and no O3 induced lesions were evident in the
18                   nasal passages. Ozone  increased BAL levels of N-acetyl-glucosaminidase (NAG; a
19                   marker of injury) and protein. DEP co-exposure (2.0 mg/m3, nose only) inhibited these
20                   responses. These pro-inflammatory effects in an allergic mouse model have also been
21                   observed in rats. Wagner et al. (2007) exposed the relatively O3-resistant Brown Norway
22                   rat strain to 1 ppm O3 after sensitizing and challenging with OVA. Rats were exposed for
23                   2 days, and airway inflammation was assessed one day later. Filtered air for controls
24                   contained less than 0.02 ppm O3. Histopathology  indicated O3 induced site-specific lung
25                   lesions in the centriacinar regions,  characterized by wall thickening partly due to
26                   inflammatory cells influx. BAL neutrophils were  elevated by O3 in allergic rats,  and
27                   modestly increased in non-allergic animals (not significant). A slight (but not significant)
28                   increase in macrophages was observed, but eosinophil numbers were not affected by O3.
29                   Soluble mediators of inflammation (Cys-LT, MCP-1,  and IL-6) were elevated by O3 in
30                   allergic animals but not non-allergic rats. Treatment with yT, which neutralizes oxidized
31                   lipid radicals and protects lipids and proteins from nitrosative damage, did not alter the
32                   morphologic character or severity of the centriacinar lesions caused by O3, nor did it
33                   reduce neutrophil influx. It did, however, significantly reduce O3-induced soluble
34                   inflammatory mediators in allergic rats. The effects of O3 in animal models of allergic
35                   asthma are discussed in section 6.2.6.

36                   In summary, a large number of toxicology studies have demonstrated that acute exposure
37                   to O3 produces injury and inflammation in the mammalian lung, supporting the
      Draft - Do Not Cite or Quote                       6-83                                September 2011

-------
 1                   observations in controlled human exposure studies (Section 6.2.3.1). These acute
 2                   changes, both in inflammation and morphology, provide a modicum of evidence for long
 3                   term sequelae of exposure to O3. Related alterations resulting from long term exposure,
 4                   such as fibrotic changes, are discussed in Chapter 7.


                     Mechanisms of Injury

 5                   Since O3 has been well established as a causative agent of airway inflammation and
 6                   injury, which may contribute to functional changes observed in human subjects, the
 7                   majority of recent research has focused on the underlying mechanisms. A brief
 8                   description of some of the recent contributions to this area of research is provided here;
 9                   more detailed descriptions of the mechanisms behind O3-mediated injury and
10                   inflammation can be found in the mode of action chapter (Chapter 5). There are several
11                   signaling pathways responsive to changes in oxidation status, which tend to be influenced
12                   at different levels in different host backgrounds. The molecular mechanisms of TNF
13                   receptor-mediated lung injury induced by O3 and associated signaling pathways (NF-KB,
14                   MAPK/AP-1) have been examined (Fakhrzadeh et al.. 2008; Cho et al.. 2007). along with
15                   the changes in gene expression which characterize O3-induced stress and inflammation
16                   (Wang et al.. 2007). Other contributors to injury and inflammation include the IL-1 and
17                   neurokinin receptors (Oslund et al.. 2008; Johnston et al.. 2007).  calcitonin gene-related
18                   peptide receptor activation (Oslund et al.. 2009). CXCR2,  a receptor for neutrophil
19                   chemokines (Johnston et al.. 2005a). mindin, an extracellular matrix protein (Frush et al..
20                   In Press), and NQO1 (Voynow et al.. 2009). an enzyme involved in oxidative stress.
21                   Studies indicate a role for oxidative stress in mediating inflammation (Wagner et al..
22                   2007; Jang et al.. 2005). Protective roles have been identified for nitric oxide synthase
23                   (Kenyon et al..  2006). metallothionein (Inoue et al.. 2008). matrix metalloproteinases
24                   (Yoon et al.. 2007). Clara cell secretory protein (Plopper et al.. 2006). and the recognition
25                   of oxidized lipids by alveolar macrophages (Dahl et al.. 2007).
            6.2.4   Respiratory Symptoms and Medication Use

26                   Controlled human exposure and toxicological studies have described the modes of action
27                   through which short-term O3 exposure may lead to increases in respiratory symptoms by
28                   demonstrating O3-induced increases in airway hyperresponsiveness, bronchoconstriction
29                   (Section 6.2.2), and pulmonary inflammation (Sections 6.2.3.1 and 6.2.3.3). While
30                   epidemiologic studies have not widely examined associations between ambient O3
31                   exposure and airway hyperresponsiveness, they have found O3-associated increases in
32                   pulmonary inflammation and oxidative stress (Section 6.3.2.2). In addition to decreases
      Draft - Do Not Cite or Quote                      6-84                                September 2011

-------
 1                   in lung function, controlled human exposure studies clearly demonstrate increases in
 2                   subjective respiratory symptoms including cough, pain on deep inspiration, and shortness
 3                   of breath (described in detail in Section 6.2.1.1). Similar to lung function responses, these
 4                   respiratory symptoms increase with exposure concentration, activity level of the exposed
 5                   individual, and duration of exposure (McDonnell et al.,  1999). Increases in subjective
 6                   respiratory symptoms have been reported following 5.6 and 6.6 h of exposure to 60 ppb
 7                   O3. However, the severity of respiratory symptoms following 6.6 h of exposure to 80 ppb
 8                   O3 during moderate exercise is roughly 2-3 times greater than that at 60 ppb O3 (Adams.
 9                   2006a). These findings integrated across disciplines provide biological plausibility for
10                   epidemiologic associations between increases in short-term ambient O3 exposure and
11                   increases in respiratory symptoms.

12                   In epidemiologic studies, respiratory symptom data typically are collected by having
13                   subjects or their parents record symptoms such as wheeze, cough, and shortness of breath
14                   and medication use in a diary without direct supervision by study staff.  Several
15                   limitations of symptom reports are we 11-recognized: recall error if not recorded daily,
16                   differences among subjects in the interpretation of symptoms, biased reporting between
17                   participants with and without asthma, and occurrence in a smaller percentage of the
18                   population compared with changes in lung function and biological markers of pulmonary
19                   inflammation. Nonetheless, symptom diaries remain a convenient and useful tool to
20                   collect individual-level data from a large number of subjects and allow the modeling of
21                   associations between daily changes in O3 exposure and daily changes in respiratory
22                   morbidity. Importantly, most of the limitations described above are sources of random
23                   measurement error that can bias effect estimates to the null or increase the uncertainty
24                   around effect estimates. Furthermore, because respiratory symptoms are associated with
25                   limitations in activity and function and are the primary reason for using medication and
26                   seeking medical care, they provide an assessment of the clinical and public health
27                   significance of ambient O3 exposure.

28                   Most studies have been conducted in individuals with asthma, and as was concluded in
29                   previous O3 AQCD, the collective body of epidemiologic evidence strongly supports
30                   associations between increases in short-term ambient O3 exposure and increases in
31                   respiratory symptoms in children with asthma (U.S. EPA. 2006b.  1996a) (Figure 6-11
32                   and Table 6-19). Evidence also indicates that O3 exposure likely is associated with
33                   increased use of asthma medication (Figure 6-12  and Table 6-20). Studies also find O3
34                   exposure to be associated with respiratory symptoms in adults with asthma. The effects of
35                   O3 exposure on respiratory symptoms in healthy populations are not as clearly indicated
36                   (Figure 6-13 and Table 6-23)
      Draft - Do Not Cite or Quote                       6-85                                September 2011

-------
                    6.2.4.1    Children with Asthma

                    Respiratory Symptoms

 1                  Table 6-18 presents the characteristics and ambient O3 concentration data from studies
 2                  assessing associations of short-term O3 exposure with respiratory symptoms and
 3                  medication use in children with asthma. The strong evidence for associations between
 4                  ambient O3 exposure and respiratory symptoms among children with asthma is derived
 5                  mostly from several single-region or single-city studies (Figure 6-11 and Table 6-19).
 6                  Most studies of children with asthma examined 1-h max, 8-h max, or 8-h average O3
 7                  exposures. In U.S. multicity studies, O3 was associated with both increases and decreases
 8                  in respiratory symptoms among children with asthma (O'Connor et al.. 2008; Schildcrout
 9                  et al.. 2006; Mortimer etal.. 2002). In the NCICAS cohort (described in Section 6.2.1.2),
10                  a 30-ppb increase in lag 1-4 avg of 8-h avg  (10:00 a.m.-6:00 p.m.) O3 was associated with
11                  an increase in morning asthma symptoms with an OR (95% CI) of 1.35 (95% CI: 1.04,
12                  1.69) (Mortimer etal.. 2002). This association did not change (OR: 1.37 [95% CI: 1.02,
13                  1-84]) in an analysis restricted to O3 concentrations below 80 ppb. Odds ratios for lags 2
14                  and 4 of O3 exposure were similar in magntiude. In the ICAS cohort (described in Section
15                  6.2.1.2), associations of 19-day avg of 24-h avg O3 with wheeze and nighttime asthma
16                  were positive and negative, respectively (O'Connor et al.. 2008). NCICAS was conducted
17                  during the warm  season, and symptom data were collected daily (Mortimer et al.. 2002:
18                  Mortimer  et al.. 2000). whereas in ICAS, every 2 months, parents reported the number of
19                  days with  respiratory symptoms over the previous  2 weeks (O'Connor et al.. 2008).
20                  Because of the two-week symptom reporting period, ICAS investigators were precluded
21                  from examining associations with single-day and shorter-duration O3 exposure periods.

22                  Evidence of O3-associated respiratory symptoms also was weak in another recent U.S.
23                  multicity study (with cities in common with NCICAS and ICAS,  Table 6-18) of 990
24                  children with asthma (Schildcrout et al.. 2006). As part of the Childhood Asthma
25                  Management Program, symptom data were  collected daily,  and analyses were restricted
26                  to peak O3 periods between May and September. In meta-analyses that combined city-
27                  specific estimates, a 40-ppb increase in lag 0 of 1-h max O3 was associated with any
28                  asthma symptom with an OR (95% CI) of 1.08 (0.89, 1.31). Odds ratios for lags 1 and 2
29                  and the 3-day sum of O3 were near 1.0. In this study, data were available from an average
30                  of 12 subjects per day per city, and fewer data were collected in summer months.
31                  Because O3 analyses were restricted to summer months, the fewer number of
32                  observations reduced the power to detect associations for O3 relative to other pollutants,
33                  which were analyzed using year-round data.
      Draft - Do Not Cite or Quote                       6-86                                September 2011

-------
Table 6-18    Mean and upper percentile ozone concentrations in epidemiologic
               studies examining respiratory symptoms, medication use, and
               activity levels in children with asthma
Study
Mortimer etal.
(2000)
Mortimer etal.
(2002)
O'Connor etal.
(2008)
Schildcroutetal.
(2006)
Gent etal.
(2003)
Thurston et al.
(1997)
Rabinovitch et
al. (2004)
Mann et al.
(2010)
Ostro et al.
(2001)
Delfino et al.
(2003)
Romieu et al.
(1996)
Romieu et al.
(1997)
Romieu et al.
(2006)
Escamilla-Nunez
etal. (2008)
Gielen etal.
(1997)
Just etal. (2002)
Jalaludin et al.
(2004)
Location
Bronx, East Harlem, NY;
Baltimore, MD; Washington, DC;
Detroit, Ml, Cleveland, OH;
Chicago, IL; St. Louis, MO
(NCI CAS)
Boston, MA; Bronx, Manhattan NY;
Chicago, IL; Dallas, TX, Seattle,
WA; Tucson, AZ
(ICAS)
Albuquerque, NM; Baltimore, MD;
Boston, MA; Denver, CO; San
Diego, CA; Seattle, WA; St. Louis,
MO; Toronto, ON, Canada (CAMP)
CT, southern MA
Connecticut River Valley, CT
Denver, CO
Fresno/Clovia, California
Los Angeles, CA
Los Angeles, CA
northern Mexico City, Mexico
southern Mexico City, Mexico
Mexico City, Mexico
Mexico City, Mexico
Amsterdam, Netherlands
Paris, France
Sydney, Australia
Years/Season
1993
Warm season
1998-2001
All-year
1994-1995
Warm season
2001
April-September
1991-1993
Warm season
1999-2002
Cold season
2000-2005
All-year
1993
August-October
1999-2000
Cold season
April-July 1991
November 1991 -
February 1992
April-July 1991
November 1991 -
February 1992
1998-2000
All-year
2003-2005
All-year
1995
Warm season
1996
April-June
1994
All-year
03
Averaging
Time
8-h avg
(10:00 a.m.-
6:00 p.m.)
24-h avg
1-h max
1-hmax
8-h rolling avg
1-hmax
1-hmax
8-h max
1-h max
1-h max
8-h max
1-h max
1-hmax
1-hmax
8-h max
1-hmax
8-h max
8-h max
24-h avg
15-h avg (6:00
a.m.-9:00
p.m.)
Mean/Median
Concentration (ppb)
48
NR
Range in medians
across cities: 43.0-65.8
58.6
51.3
83.6
28.2
49.4 (median)
Los Angeles: 59.5
Pasadena: 95.8
25.4
17.1
190
196
102
69
86.5
31.6
34.2
30.0
12
Upper Percentile
Concentrations (ppb)
NR
NR
Range in 90th across
cities: 61 .5-94.7
Max: 125.5
Max: 99.6
Max: 160
Max: 70.0
75th: 69.5, Max: 120.0
Max: 130
Max: 220
90th: 38.0, Max: 52
90th: 26.1, Max: 37
Max: 370
Max: 390
Max: 309
Max: 184
Max: 86.3 (8-h max)
Max: 56.5
Max: 61 .7
Max: 43
NCICAS = National Cooperative Inner-City Asthma Study, NR = Not Reported, ICAS = Inner City Asthma Study, NR = Not Reported, CAMP =
Childhood Asthma Management Program, Max = Maximum.
Draft - Do Not Cite or Quote
6-87
September 2011

-------
 Study              Symptom

 Aggregate of symptoms

 Rabinovitchetal. (2004)  Daytime symptoms

 Delfinoetal. (2003)     Bothersome symptoms

 Schildcroutetal. (2006)  Asthma symptoms
Gielenetal. (1997)
LRS
URS
 Mortimeret al. (2002)    Morning symptoms
 Mortimeret al. (2000)
 Romieuetal. (1996)

 Romieuetal. (1997)

 Individual symptoms

 Jalaludinetal. (2004)

 O'Connoretal. (2008)

 Ostroetal.(2001)
 Escamilla-Nunez et al.
      (2008)

 Mann etal. (2010)


 Thurstonetal.(1997)

 Romieuetal. (2006)
LRS

LRS



Wheeze

Wheeze/cough

Wheeze

Cough

Wheeze


Chest symptoms

Difficulty breathing
                O3 Lag   Subgroup



                0-2 avg

                0

                0

                0
                 1-4 avg   All subjects
                        No medication
                        Cromolyn use
                                         Beta-agonist/xanthine use
                                         Steroid use        —
                                         Without allergy
                                         With allergy
0

0



2

1-19 avg

3

0

0
       All
       Fungi allergic
0-5 avg  GSTM1 positive
       GSTM1 null
       GSTP1 lie/lie Ile/Val
       GSTP1 Val/Val
                                                   0.5        1        1.5

                                                        Odds ratio (95% Cl)
                                                                                     2.5
Figure 6-11    Associations of ambient ozone exposure with respiratory
               symptoms in children with asthma. Results are presented first for
               aggregate indices of symptoms then for individual symptoms.
               Within each category, results generally are organized in order of
               increasing mean ambient O3 concentration. LRS = lower respiratory
               symptoms, URS = upper respiratory symptoms. Effect estimates
               are from single-pollutant models and are standardized to a 40-, 30-,
               and 20-ppb increase for 1-h max, 8-h max or 8-h avg, and 15-h avg
               or 24-h avg ozone exposures, respectively.
Draft - Do Not Cite or Quote
                        6-88
                                         September 2011

-------
     Table 6-19     Additional characteristics and quantitative data for studies
                      presented in Figure 6-11
Study
Studies examining
Location/ Population
aggregates of symptoms
Rabinovitch et al. (2004) Denver, CO
Children with asthma
Delfino et al. (2003)
Schildcroutetal. (2006
Gielenetal. (1997)
Mortimer etal. (2002)
Mortimer etal. (2QQO)
Romieu et al. (1996)
Romieu et al. (1997)
Studies examining
Jalaludin et al. (2004)
O'Connor etal. (2008)
Ostro et al. (2001)
Escamilla-Nunez et al.
(2008)
Los Angeles, CA
Children with asthma
I) 8 U.S. communities
Children with asthma
Amsterdam, Netherlands
Children with asthma
8 U.S. communities
Children with asthma
northern Mexico City,
Mexico
Children with asthma
southern Mexico City,
Mexico
Children with asthma
individual symptoms
Sydney, Australia
Children with asthma
7 U.S. communities
Children with asthma
Los Angeles, CA
Children with asthma
Mexico City, Mexico
Children with asthma
03
Lag

0-2
avg
0
0
0
1-4
avg
0
0

2
1-19
avg
3
0
03
Averagin
gTime

1 -h max
1 -h max
1 -h max
8-h max
8-h avg
(10:00 a.m.-
6:00 p.m.)
1 -h max
1 -h max

1 5-h avg
(6:00 a.m.-
9:00 p.m.)
24-h avg
1 -h max
1 -h max
Symptom Subgroup

Daytime
symptoms
Bothersome
symptoms
Asthma symptoms
LRS
URS
Morning All subjects
symptoms No medication use
Cromolyn use
p-agonist/xanthine use
Steroid use
Without allergy
With allergy
LRS
LRS

Wheeze
Wheeze/cough
Wheeze
Wheeze
Odds Ratio
(95% Cl)a

1.34(1.01,1.77)
1 .09 (0.39, 3.03)
1.08(0.89,1.31)
1.04(0.75,1.45)
1.16(1.02,1.32)
1.35(1.04,1.74)
1.08(0.62,1.87
2.13(1.12,4.04)
1.39(0.98,1.98)
1.17(0.79,1.72)
1.59(1.00,2.52)
1.35(0.92,1.96)
1.07(1.02,1.12)
1.09(1.04,1.14)

1.21 (0.92,1.59)
1.02(0.86,1.21)
0.94(0.88,1.00)
1.08(1.03,1.14)
Mann et al. (2010)
Thurston et al. (1997)
Romieu et al. (2006)
Fresno/Clovia, California
Children with asthma
CT River Valley, CT
Children with asthma
Mexico City, Mexico
Children with asthma
0
0
0-5
avg
8-h max
1 -h max
1-h max
Wheeze
Chest symptoms
Difficulty breathing
All
Fungi allergic

GSTM1 sufficient
GSTM1 null
GSTP1 lie/lie NeA/al
GSTP1 ValA/al
1 .00 (0.84,
1 .06 (0.84,
1.28(1.10,
1.10(0.98,
1.17(1.02,
1 .06 (0.94,
1.30(1.10,
1.19)
1.34)
1.50)
1.24)
1.33)
1.20)
1.53)
      LRS = Lower respiratory symptoms, URS = Upper respiratory symptoms.
     "Effect estimates are standardized to a 40, 30, and 20 ppb increase for 1-h max, 8-h max or 8-h avg, and 15-h avg or 24-h avg 03, respectively.
4
5
Several longitudinal studies conducted in multiple cohorts of children with asthma in
Mexico City, Mexico examined 1-h max O3 exposures and found associations with
increases in respiratory symptoms (Escamilla-Nunez et al.. 2008; Romieu et al.. 2006;
Romieu etal.. 1997; Romieu etal.. 1996). Recent studies expanded on earlier evidence
     Draft - Do Not Cite or Quote
                               6-89
September 2011

-------
 1                   by indicating associations with multiday averages of O3 exposure. Both Romieu et al.
 2                   (1996) and Romieu et al. (1997) found that among single-day 1-h max O3 exposures
 3                   lagged 0 to 2 days, lag 0 had the greatest estimated effect on respiratory symptoms.
 4                   Romieu et al. (2006) and Escamilla-Nunez et al. (2008) found that the magnitudes of
 5                   association of ambient 1-h max O3 exposure with respiratory symptoms and medication
 6                   use increased with increasing number of days over which O3 exposure was averaged.

 7                   Studies of children with asthma also identified factors that may contribute to
 8                   heterogeneity in symptom responses to ambient O3 exposure. Multiple studies, all of
 9                   which examined 8-h avg (10:00 a.m.-6:00 p.m.) or 8-h max O3 exposures, found larger
10                   associations among subjects taking asthma medication; however, the medications varied
11                   among studies. Consistent with findings for lung function, in the NCICAS multicity
12                   cohort,  larger associations for morning symptoms were observed in children taking
13                   cromolyn (used to treat asthma with allergy) or beta-agonists/xanthines than in children
14                   taking no medication. Odds ratios did not differ as much between children taking steroids
15                   and children taking no medication (Figure 6-11 and Table 6-19) (Mortimer et al.. 2000).
16                   In a cohort of children with asthma in Southern New England, O3 exposures were
17                   associated with larger increases in chest tightness among children taking maintenance
18                   medication (i.e., steroids, cromolyn, or leukotriene inhibitors).

19                   Most studies of children with asthma reported that a majority of subjects (52 to 100%)
20                   were atopic as determined by a positive skin prick test to any examined allergen;
21                   however, results did not conclusively indicate that children with asthma and atopy were
22                   more susceptible to the effects of O3 exposure. In the multicity NCICAS cohort,
23                   Mortimer et al. (2000) found that O3 was associated with a similar incidence of asthma
24                   symptoms among the 79% of subj ects with atopy and the 21 % of subj ects without atopy
25                   (Figure 6-11 and Table 6-19). Odds ratios did not differ by residential levels of allergens.
26                   In a recent study of children with asthma in Fresno, CA, most associations of single- and
27                   multiday lags of 8-h max O3 exposure (0-14 days) with wheeze were near or below 1.0
28                   (Mann et al.. 2010). The estimated effects did not differ in fungi allergic subjects, A
29                   larger association was found for cat allergic subjects; however, this finding was limited to
30                   O3 exposure lagged 14 days. In this study, many subjects were allergic to multiple
31                   allergens; however, associations were not compared between subjects with any versus no
32                   allergic sensitization.

33                   Although Romieu et al. (2006) did not observe differences  in associations between O3
34                   and lung function by GST genetic polymorphisms (Section 6.2.1.2), they did observe
35                   effect modification for respiratory symptoms. Compared with GSTM1 positive subjects
36                   and GSTP1  lie/lie or Ile/Val subjects, larger effects were estimated for GSTM1 null
37                   subjects and for GSTP1  Val/Val  subjects, respectively (Figure 6-11 and Table 6-19).
      Draft - Do Not Cite or Quote                       6-90                                 September 2011

-------
 1                   Ozone had the greatest estimated effect on difficulty breathing in children with asthma
 2                   who were both GSTM1 null and GSTP1 Val/Val (OR:  1.49 [95% CI: 1.14,  1.93] per 30-
 3                   ppb increase in lag 0-5 avg of 8-h max O3). In the same cohort of children, antioxidant
 4                   supplementation reduced O3-associated increases in airway inflammation (Sienra-Monge
 5                   et al.. 2004). These results add to the body of epidemiologic evidence that antioxidant
 6                   capacity influences risk of O3-related respiratory morbidity. As was discussed in Section
 7                   6.2.1.2, compared with the GSTM1 genotype, evidence for effect modification by GSTP1
 8                   genetic polymorphisms is less certain. Romieu et al. (2006) found that the GSTP1
 9                   Val/Val variant was associated with a lesser O3-associated decrement in lung function but
10                   greater risk of respiratory symptoms. Whereas some studies have reported greater risk of
11                   asthma among GSTP1 lie/lie or Ile/Val subjects (Mapp et al., 2002; Hemmingsen et al..
12                   2001). others have reported greater risk among GSTP1  Val/Val subjects (Tamer et al..
13                   2004). In Romieu et al. (2006). GSTP1 lie/lie was associated with greater severity of
14                   asthma, and Lee  et al. (2004b) also reported greater risk of air pollution-associated
15                   asthma among GSTP1 lie/lie children in the Southern California Children's Health
16                   Study.


                     Asthma Medication Use

17                   The 2006 O3 AQCD concluded that ambient O3 likely was associated with increased
18                   asthma medication use based on the positive associations found in several studies of
19                   children with asthma (Figure 6-12  and Table 6-20). Among the few newly available
20                   studies on asthma medication use,  evidence generally supported the previous conclusion
21                   (Escamilla-Nunez et al.. 2008; Romieu et al.. 2006). Most of these studies examined lags
22                   0 or 1 of 1-h max O3 exposures; however, Romieu et al. (2006) found that lag 0-5 avg of
23                   1-h max O3 was associated with a larger increase in bronchodilator use than were lags 1
24                   or 0-1 avg. As compared with respiratory symptoms, effects on medication use were
25                   estimated with greater uncertainty  as indicated by the wide 95% CIs. The wide 95% CIs
26                   have been attributed to a smaller number of study subjects reporting medication use and
27                   the low frequency of use over the study period. However, within most studies, findings
28                   were similar for respiratory symptoms and asthma medication use. Among recent studies,
29                   Romieu et al. (2006) and Escamilla-Nunez et al. (2008) observed O3-associated increases
30                   in both respiratory symptoms and bronchodilator use. Schildcrout et al. (2006) did not
31                   observe O3-associated increases in either respiratory symptoms or rescue  inhaler use. In
32                   contrast, Romieu et al. (1996) and  Rabinovitch et al. (2004) observed that O3 was
33                   positively associated with daytime respiratory symptoms but not with bronchodilator use.
      Draft - Do Not Cite or Quote                      6-91                                 September 2011

-------
 Study               Outcome

 Schildcroutetal. (2006)  Rescueinhaleruse

 Thurstonetal. (1997)   Beta-agonist use


 Ostro et al. (2001)      Extra medication use


 Romieuetal. (1996)

 Romieuetal. (1997)
 Romieuetal. (2006)
 Gielenetal. (1997)
Bronchodilatoruse

Bronchodilatoruse

Bronchodilatoruse
Lag     Subgroup

0


0


1


0


0


0-5 avg   GSTP1 lie/lie Ile/Val
        GSTP1 Val/Val
                    Bronchodilatoruse
 Jalaludinetal. (2004)   Beta-agonist use/no steroid  1
                    CSuse
                                                       0            0.5            1             1.5
                                                                           Odds ratio (95% Cl)


  CS = corticosteroid. Results are presented in increasing order of ambient ozone concentration. Effect estimates are from single-
pollutant models and are standardized to a 40- and 30-ppb increase for 1-h max and 8-h max ozone, respectively


Figure 6-12    Associations of ambient ozone exposure with asthma medication
                   use.
Table 6-20
Study
Schildcroutetal.
(2006)
Thurston et al.
(1997)
Ostro et al. (2001)
Romieu et al. (1996)
Romieu et al. (1997)
Romieu et al. (2006)
Gielenetal. (1997)
Jalaludin et al.
(2004)
Additional characteristics and quantitative data for studies
presented in Figure 6-12
Location/
Population
8 U.S. communities
Children with asthma
CT River Valley, CT
Asthmatic campers
Los Angeles, CA
Children with asthma
northern Mexico City, Mexico
Children with asthma
southern Mexico City, Mexico
Children with asthma
Mexico City, Mexico
Children with asthma
Amsterdam, Netherlands
Children with asthma
Sydney, Australia
Children with asthma
O3Lag
0
0
1
0
0
0-5 avg
0
1
03
Averaging
Time
1-h max
1-h max
1-h max
1-h max
1-h max
1-h max
8-h max
1-h max
Medication Subgroup
Rescue inhaler use
Beta-agonist use
Extra medication use
Bronchodilator use
Bronchodilator use
Bronchodilator use GSTP1 lie/lie
Ile/Val
GSTP1 ValA/al
Bronchodilator use
Beta-agonist use/no
steroid
ICS use
Odds Ratio
(95% Cl)a
1.01 (0.89,1.15)
1.17(0.96,1.44)
1.10(1.03,1.19)
0.97(0.93,1.01)
1.02(1.00,1.05)
0.96 (0.90, 1 .02)
1.10(1.02,1.19)
1.10(0.78,1.55)
1 .08 (0.89, 1 .32)
1.08(0.96,1.21)
  CS= Corticosteroid.
"Effect estimates are standardized to a 40- and 30-ppb increase for 1-h max and 8-h max 03, respectively.
Draft - Do Not Cite or Quote
                                 6-92
                                                    September 2011

-------
                     Changes in Activity

 1                   While investigation has been limited, evidence has not consistently indicated associations
 2                   between O3 exposure and diminished activity level in children with asthma (O'Connor et
 3                   al.. 2008; Delfino et al.. 2003). These studies have examined a range of O3 averaging
 4                   times and lags of exposure. In the large ICAS cohort, O'Connor et al. (O'Connor et al..
 5                   2008) found that a 20-ppb increase in lag 1-19 avg of 24-h O3 ambeint was associated
 6                   with a 10% lower odds (95% CI: -26, 10) of slow play. In a small (n = 22) panel study
 7                   conducted in children with asthma in Los Angeles CA, Delfino et al. (2003) found that a
 8                   40-ppb increase in lag 0 of 1-h max O3 was associated with an increase in symptoms that
 9                   interfered with daily activity with an OR (95% CI) of 7.41 (1.18, 43.2). Several studies
10                   reported increases  in school absenteeism in children with asthma in association with long
11                   lags of O3 exposure (14-day and 30-day distributed lags or 19-day avg) (O'Connor et al..
12                   2008; Gilliland et al.. 2001; Chen etal.. 2000). Whereas  Chen et al. (2000) and O'Connor
13                   et al. (2008) examined absences for any reason, Gilliland et al. (2001) found associations
14                   with absences for respiratory illnesses. Despite this evidence, several limitations have
15                   been noted, including the uncertain biological relevance  of long lag periods of O3
16                   exposure and the potential for residual seasonal confounding when examining long lag
17                   periods of exposure. In analyses of single-day lags, Gilliland et al. (2001) found that 8-h
18                   avg (10:00 a.m.-6:00 p.m.) O3 exposure was associated with increases in respiratory-
19                   related absences from lag day 1 to lag day 5, indicating an effect of exposures with
20                   shorter lag periods.
                     6.2.4.2    Adults with Respiratory Disease

21                   Characteristics and ambient O3 concentration data from studies of adults with respiratory
22                   disease are presented in Table 6-21. In this relatively small body of literature, several
23                   studies found ambient O3 exposure (1-h max or 8-h max) to be associated with increases
24                   respiratory symptoms and decreases in activity levels in adults with asthma (Khatri et al..
25                   2009: Feo Brito et al.. 2007: Eiswerth et al.. 2005: Ross et al.. 2002). In a recent panel
26                   study of adults with COPD, investigators found lag 1 of 8-h max O3 to be associated with
27                   increased odds of dyspnea and sputum changes but decreased odds of nasal discharge,
28                   wheeze, or upper respiratory symptoms (Peacock et al.. 2011).

29                   In a panel study of children and adults with asthma, lag 1-3 avg of 8-h max O3 exposure
30                   was associated with increases in morning and evening symptom scores and frequency of
31                   asthma medication use (Ross et al.. 2002). During one pollen season (May-June 2000 or
32                   2001), Feo Brito et al. (2007) specifically followed a group of 137 adults who had asthma
33                   and pollen allergy in central Spain. In the industrial Puertollano, a 40-ppb increase in lag
      Draft - Do Not Cite or Quote                       6-93                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
3 of 1-h max O3 was associated with a 14.3% increase (95% CI: 3.6, 26.0) in the number
of subjects reporting respiratory symptoms, adjusting only for time trend. There was a
much weaker association in the less industrialized Ciudad Real with lower ambient air
pollution concentrations and a narrower range of ambient O3 concentrations (2.3% [95%
CI: -14, 21%] per 40-ppb increase in lag 4 of 1-h max O3). Park et al. (2005a) followed
adults with asthma in Korea during a period that included dust storms and found that a
20-ppb increase in lag 0 of 24-h avg O3 was associated with an increased odds of night
symptoms (OR: 1.11 [95% CI: 0.96,  1.29]) but not cough (OR: 1.00 [95% CI: 0.94,
1.06]) or rescue inhaler use (OR: 0.99 [95% CI: 0.94,  1.05]).
Table 6-21
Study
Khatrietal.
(2009)
Ross et al.
(2002)
Eiswerth et al.
(2005)
Peacock etal.
(2Q11)
Feo Brito etal.
(2007)
Wiwatanadate et
al. (2011)
Park etal.
(2005a)
Mean and upper percentile ozone concentrations in epidemiologic
studies examining respiratory symptoms and medication use in
adults with respiratory disease
Location
Atlanta, GA
East Moline, IL
Glendora, CA
London, England
Ciudad Real and
Puertollano, Spain
Chiang Mai, Thailand
Incheon, Korea
Years/Season
2003, 2005, 2006
Warm season
April-October 1994
1983
Cold season
1995-1997
All-year
2000-2001
Warm season
August 2005-June
2006
March-June 2002
« °'-
Averaging
Time
8-h max
8-h avg
1-h max
8-h max
1-h max
24-h avg
24-h avg
Mean/Median
Concentration (ppb)
59a
41.5
NR
15.5
65.9 (Ciudad Real)"
56.8 (Puertollano)"
17.5
Dust event days: 23.6
Control days: 25.1
Upper Percentile
Concentrations (ppb)
Max: 73a
Max: 78.3
NR
Autumn/Winter Max: 32
Spring/Summer Max: 74
Max: 101.5" (Ciudad Real);
70.5b (Puertollano)
90th: 26.82
Max: 34.65
NR
        NR = Not Reported, Max = Maximum.
      'Individual-level exposure estimates were derived based on time spent in the vicinity of various 03 monitors.
      "Concentrations converted from ug/m3 to ppb using the conversion factor of 0.51 assuming standard temperature (25°C) and pressure (1 atm).
10                    Studies also indicated that ambient O3 exposure may result in decreases in activity levels
11                    in adults with asthma. Notably, although conducted over single seasons, these studies did
12                    not consider confounding by meteorological factors. In a cross-sectional summer study in
13                    Atlanta, GA (described in Section 6.2.1.2), Khatri et al. (2009) observed that a 30-ppb
14                    increase in lag 2 of 8-h max O3 was associated with a 0.69-point decrease (95% CI:  -1.28,
15                    -0.11) in the Juniper quality of life score, which incorporates indices for symptoms,
16                    mood, and activity limitations (7-point scale). In a fall study conducted in the Los
17                    Angeles, CA area, Eiswerth et al. (2005) examined the activities of 64 individuals with
18                    asthma (age  16 years and older). A 40-ppb increase in 1-h max O3 was associated with a
19                    0.24% (95% CI: 0.08, 0.40%) lower probability of participation in indoor activities. The
20                    association with outdoor activities was positive but not statistically significant. Although
      Draft - Do Not Cite or Quote
                                6-94
September 2011

-------
 1
 2
 3
 4
 5
 6
the authors acknowledged that their findings were unexpected and may have been
influenced by lack of control for potential confounders, they interpreted the decrease in
indoor activities as rest replacing chores. In contrast, in a panel study of individuals with
asthma (ages 13-78 years) in Thailand, O3 exposure was associated with a lower odds of
symptoms that interfered with activities (OR: 0.74 [95% CI: 0.57, 0.96] per 20-ppb
increase in lag 4 of 24-h avg O3) (Wiwatanadate and Liwsrisakun. 2011).
 9
10
11
12
6.2.4.3   Populations not Restricted to Individuals with Asthma

Characteristics and ambient O3 concentration data from studies of populations not
restricted to individuals with asthma are presented in Table 6-22. In contrast with
findings for lung function (Section 6.2.1.2), epidemiologic studies do not provide
consistent evidence of associations between short-term ambient O3 exposure and
increases in respiratory symptoms in children without asthma (Figure 6-13 and
Table 6-23).
Table 6-22 Mean and upper percentile ozone concentrations in epidemiologic
studies examining respiratory symptoms in populations not
restricted to individuals with asthma
Study
Apteetal.
(2008)
Neas et al.
(1995)
Triche et al.
(2006)
Linn et al. (1996)
Gold et al.
(1999)
Wardetal.
(2002)
Hoekand
Brunekreef
(1995)
Moon et al.
(2009)
Rodriguez etal.
(2007)
Location
Multiple U.S. cities (NR)
Uniontown, PA
Southwestern VA
Rubidoux, Upland,
Torrence, CA
Mexico City, Mexico
Birmingham and Sandwell,
England
Deurne and Enkhuizen,
Netherlands
4 cities, South Korea
Perth, Australia
Years/Season
1994-1998
Winter or summer
June-August 1990
1995-1996
Warm season
1992-1993,1993-1994
Fall and spring
1991
Winter, spring, fall
1997
Winter and summer
1989
March-July
April-May 2003
1996-2003
All-year
Os Averaging
Time
Workday avg
(8:00 a.m. -
5:00 p.m.)
24-h avg
12-h avg (8:00
a.m.-8:00p.m.)
1-h max
8-h max
24-h avg
24-h avg
24-h avg
24-h avg
1-h max
8-h avg (10:00
a.m.-6:00p.m.)
1-h max
24-h avg
Mean/Median
Concentration (ppb)
34.2a
25.5a
37.2
60.8
54.5
35.2
23
52.0
Winter median: 13.0
Summer median: 22.0
Deurne: 57
Enkhuizen: 59
NR
33
28
Upper Percentile
Concentrations (ppb)
Max: 86.2a
Max: 67.3a
Max: 44.9
75th: 70.0, Max: 95.0
75th: 64.1, Max: 87.6
75th: 40.6, Max: 56.6
Max: 53
Max: 103
Winter Max: 33
Summer Max: 41
Max: 107
Max: 114
NR
Max: 95
Max: 74
       NR = Not Reported, Max = Maximum.
      'Concentrations converted from ug/m3 to ppb using the conversion factor of 0.51 assuming standard temperature (25°C) and pressure (1 atm).
      Draft - Do Not Cite or Quote
                                6-95
September 2011

-------
 1                   Among healthy children in Uniontown, PA, Neas et al. (1995) found a stronger
 2                   association between O3 exposure and evening cough using ambient concentrations
 3                   weighted by time spent outdoors (OR: 2.20 [95% CI: 1.02, 4.75] per 30-ppb increase in
 4                   lag 0 of 12-h avg [8:00 a.m.-8:00 p.m.]) than using unweighted concentrations (OR: 1.36
 5                   [95% CI: 0.86, 2.13] per 30-ppb increase in lag 0 of 12-h avg [8:00 a.m.-8:00 p.m.]).
 6                   Several other panel studies of school-aged children, in which asthma prevalence ranged
 7                   between 0 to 50%, reported null or negative associations between various averaging
 8                   times and lags of ambient O3 exposure and respiratory symptoms (Moon et al.. 2009;
 9                   Rodriguez et al.. 2007; Ward et al.. 2002; Linn et al.. 1996; Hoek and Brunekreef. 1995).
10                   For example, a large study of 696 children in four regions in South Korea, Moon et al.
11                   (2009) observed that among all subjects, ORs of lag 0 8-h avg O3 with most respiratory
12                   symptoms were close to 1.0. In city-specific analyses, O3 exposure was only consistently
13                   associated with increases in URS (runny nose or sneezing), with the largest magnitude of
14                   association observed in Jeju island (OR: 1.08 [95% CI: 0.96, 1.21] per a 30-ppb increase
15                   in lag 0 8-h avg O3). Consistent with other studies conducted in Mexico City, Gold et al.
16                   (1999) reported a positive association between lag 1 of 24-h avg O3 exposure and phlegm
17                   in children; however, investigators acknowledged being unable to  distinguish between
18                   the effects of the highly-correlated O3 and PM10 (r = 0.75).

19                   In a recent study, O3 exposure was associated with increased odds of respiratory
20                   symptoms in a group of infants who have mothers with asthma (Triche et al.. 2006).
21                   Triche et al. (2006) followed 691 infants in southwestern VA Yfor 83 days between June
22                   and August of 1995 and/or 1996 and found that a 20-ppb increase in lag 0 of 24-h avg O3
23                   was associated with odds ratios (95% CI) of 2.34 (1.02, 5.37) for wheeze and of 3.63
24                   (1-81, 7.28) for difficulty breathing among the 61 infants who had mothers with asthma.
25                   Investigators estimated smaller magnitudes of association for 1-h and 8-h max O3
26                   exposures. Smaller, statistically nonsignificant associations also were found in analyses
27                   that included all subjects (Figure 6-13 and Table 6-23). While these results  suggested that
28                   children with mothers with asthma may be at greater risk of O3-related respiratory
29                   morbidity, the authors acknowledged that mothers with asthma may be more likely to
30                   report symptoms in their children and that transient wheeze, which is common in infants,
31                   may not predict respiratory morbidity later in life. In a study of children with parental
32                   history of asthma with follow-up to an older age (5 years), ambient O3 exposure was not
33                   associated with increases in respiratory symptoms (Rodriguez et al.. 2007).
      Draft - Do Not Cite or Quote                       6-96                                 September 2011

-------
 Study
O3 Lag   Symptom
                Subgroup
 Ward et al. (2002)    0-6 avg  Wheeze
                           Shortness of breath
 Hoekand Brunekreef
                   0
      (1995)


 Moon etal. (2009)    0


 Neasetal. (1995)    0

 Tricheetal. (2006)    0
 Gold etal. (1999)
0
        Cough
        URS

        URS
        Evening cough
        Wheeze
Phlegm
                All subjects
                Jeju Island
                All
                With asthmatic mothers
                                                              0123
                                                                      Odds ratio (95% Cl)

  LRS = lower respiratory symptoms, URS = Upper respiratory symptoms. Effect estimates are from single-pollutant models and are
standardized to a 40-, 30-, and 20-ppb increase for 1-h max, 8-h max or 12-h avg, and 24-h avg ozone exposures, respectively.

Figure 6-13    Associations of ambient ozone exposure with respiratory
                  symptoms in studies not restricted to children with asthma.
Table 6-23
Study
Ward etal. (2002)
Hoekand
Brunekreef (1995)
Moon et al. (2009)
Neas et al. (1995)
Triche et al. (2006)
Gold et al. (1999)
Additional characteristics and
presented in Figure 6-13
Location/
Population
Birmingham and
Sandwell, England
Children
Enkhuizen,
Netherlands
Children
4 cities, South Korea
Children
Uniontown, PA
Healthy children
southwestern VA
Infants
Mexico City, Mexico
Children
0 Laa °3 Averaging
°3 Lag Time
0-6 avg 24-h avg
0 1-h max
0 24-h avg
0 12-h avg (8:00
a.m.-8:00 p.m.)
0 8-h max
1 24-h avg
quantitative data for studies
Symptom
Wheeze
Shortness of breath
Cough
URS
URS
Evening cough
Wheeze
Phlegm
Subgroup «*«•
0.78 (0.22, 2.79)
1 .80 (0.64, 5.06)
0.95(0.71,1.25)
1.15(1.00,1.33)
All subjects 0.96(0.90,1.03)
Jeju Island 1.08(0.96,1.21)
2.20(1.02,4.75)"
All subjects 1 .60 (0.85, 3.00)
Maternal asthma 2.34 (1 .02, 5.37)
1.04(1.00,1.07)
  LRS = Lower respiratory symptoms, URS = Upper respiratory symptoms
  'Effect estimates are standardized to a 40-, 30-, and 20-ppb increase for 1 -h max, 8-h max or 12-h avg, and 24-h avg 03, respectively.
b03 exposures were weighted by the proportion of time spent outdoors.
Draft - Do Not Cite or Quote
                              6-97
                                                             September 2011

-------
 1                   A recent cross-sectional study examined 4,200 adult workers from 100 office buildings
 2                   across the U.S. and found that a range of ambient O3 exposure metrics, including the 24-
 3                   h, workday (8:00 a.m.-5:00 p.m.), and late workday (3:00 p.m.-6:00 p.m.) averages, were
 4                   associated with increases in building-related URS (nasal congestion or sore throat) and
 5                   LRS (wheeze, shortness of breath, or chest tightness) (Apte et al., 2008). Investigators
 6                   suggested that the findings may have been attributable to formaldehyde and organic acids
 7                   produced from O3-initiated reactions within buildings; however, additional data on indoor
 8                   levels of volatile organic compounds, indoor O3, and infiltration rates is warranted to
 9                   characterize whether the observed associations were attributable to the formation of these
10                   secondary species by ambient O3 penetrating indoors
                     6.2.4.4    Confounding in Epidemiologic Studies of Respiratory
                                Symptoms and Medication Use

11                   Epidemiologic studies did not indicate that associaitons between short-term O3 exposure
12                   and respiratory symptoms were confounded by meteorological factors. Except where
13                   specified in the text, associations between ambient O3 exposure and respiratory
14                   symptoms or medication use were found after adjusting for temperature in models. Some
15                   studies additionally included humidity in models (Triche et al., 2006; Ross et al., 2002) or
16                   found no independent association with respiratory symptoms (Thurston et al..  1997).

17                   Several studies that examined populations with a high prevalence of atopy found O3-
18                   associated increases in respiratory symptoms and asthma medication use in copollutant
19                   models that included daily pollen counts (Just et al.. 2002; Ross et al.. 2002; Gielen et al..
20                   1997). Gielen et al. (1997) and Ross et al.  (2002) specifically reported a high prevalence
21                   of grass pollen allergy in their study populations (52% and 38%, respectively). Ross et al.
22                   (2002) found similar associations of O3 with morning symptoms and asthma medication
23                   use in a single-pollutant model (e.g., 0.21-point [95% CI: 0.12, 0.30] increase in
24                   symptom score per 30-ppb increase in lag  1-3 avg of 8-h max O3) and in a copollutant
25                   model with daily pollen counts (e.g., 0.20-point [95% CI: 0.11, 0.29]  increase  in
26                   symptom score per 30-ppb increase in lag  1-3 avg of 8-h max O3). Feo Brito et al. (2007)
27                   specifically followed a group of adults in central Spain, all of whom had both asthma and
28                   pollen allergy. In one city, O3 was associated with an increase in the number of subjects
29                   reporting symptoms. A smaller, statistically nonsignificant effect estimate was obtained
30                   for pollen.  Conversely, in another city, pollen was associated with an increased incidence
31                   of respiratory symptoms, whereas O3 was not. While copollutant modeling was not
32                   conducted, in both locations, O3 and pollen concentrations were weakly correlated,
33                   indicating that the findings for O3 were not likely confounded by pollen. Rather, the
      Draft - Do Not Cite or Quote                       6-98                                September 2011

-------
 1
 2
results suggested that O3 and pollen may have independent effects that vary between
locations, depending on the mix of airborne pollutants.
Table 6-24
Study
Mortimer etal. (2002)
Thurston et al. (1997)
Romieu et al. (1996)
Romieu et al. (1997)
Associations between short-term ozone exposure and
symptoms in single- and copollutant models
Location/
Population
Bronx, East Harlem,
NY; Baltimore, MD;
Washington, DC;
Detroit, Ml, Cleveland,
OH; Chicago, IL; St.
Louis, MO (NCICAS)
Children with asthma
CT River Valley
Children with asthma
attending summer
camp
Mexico City, Mexico
Children with asthma
Mexico City, Mexico
Children with asthma
O3 Exposure Data Symptom
8-h avg (10:00 a.m.- Morning symptoms
6:00 p.m.)
Lag 1-4 avg
1 -h max Chest symptoms
LagO
1-hmax LRS
LagO
1-hmax LRS
LagO
Os-associated OR
in Single-Pollutant
Model (95% Cl)a
8 cities with S02 data
1.35(1.04,1.69)
7 elites with N02 data
1.25(0.94,1.67)
3 cities with PMiodata
1.21 (0.61,2.41)
1.21 (1.12,1.31)"
1.07(1.02,1.12)
1.09(1.04,1.14)
respiratory
Os-associated OR
in Copollutant
Model (95% Cl)a
with lag 1-2 avg, 3-h
avg S02
1.23(0.94,1.61)
with lag 1-6 avg, 24-h
avg N02
1.14(0.85,1.55)
with lag 1-2 avg, 24-h
avg PM10
1 .08 (0.49, 2.39)
with lag 0, 12-havg
sulfate
1.19(1.06,1.35)"
with lag 0, 24-h avg
PM2.5
1.06(1.02,1.10)
with lag 0, 24-h avg
PM10
1.09(1.01,1.19)
        LRS = Lower respiratory symptoms.
        "Effect estimates are standardized to a 40- and 30-ppb increase for 1-h max and 8-h avg O3, respectively.
        "Temperature not included in models.
 3                   Robust associations between O3 exposure and respiratory symptoms also were observed
 4                   in copollutant models that included PM2 5, PM10, sulfate, SO2, or NO2 (Table 6-24).
 5                   Information on confounding in asthma medication use associations was more limited.
 6                   The association between O3 and bronchodilator use did not change in Gent et al. (2003)
 7                   after adjusting for PM2 5 but decreased in magnitude in Thurston et al. (1997) after
 8                   adjusting for  12-h avg sulfate. For respiratory symptoms and medication use, copollutant
 9                   associations remained robust after adjusting for O3. Notably, studies examined different
10                   averaging times for O3 (1-h max or 8-h avg) and co-pollutants (3-h to 24-h avg) and
11                   reported a range of correlations with co-pollutants. Two studies conducted concurrently
12                   in two regions of Mexico City examined lag 0 exposures of 1-h max O3 and 24-h avg
13                   PM10 or PM25 and found robust associations with respiratory symptoms for both O3 and
14                   co-pollutants  (Romieu et al.. 1997; Romieu et al.. 1996). Romieu et al. (1997) reported a
15                   moderate correlation  between 1-h max O3 and 24-h avg PM10 (r = 0.47). Thurston et al.
16                   (1997) and  Gent et al. (2003) found 1-h max O3 concentrations to be highly correlated
17                   with 12-h avg sulfate (r=0.74) and 24-h avg PM25 (r=0.77), respectively, thus the
18                   copollutant results  should be interpreted with caution. The association between O3
19                   exposure and respiratory symptoms observed in NCICAS was robust in two-pollutant
      Draft - Do Not Cite or Quote
                                6-99
September 2011

-------
 1                  models with SO2, NO2, and or PM10; however, the interpretation is complicated because
 2                  of the different averaging times and lags of exposure examined for O3 and co-pollutants
 3                  (Mortimer et al., 2002) (Table 6-24). Also difficult are interpretations of the robust
 4                  associations observed between ambient O3 exposure and respiratory symptoms after
 5                  adjusting for multiple pollutants (i.e., PM2 5 plus NO2 or PM10.2 5) (Escamilla-Nunez et al.
 6                  2008: Tricheetal.. 2006).
                     6.2.4.5   Summary of Epidemiologic Studies of Respiratory
                               Symptoms and Asthma Medication Use

 7                   With a majority of investigation focused on individuals with asthma, the collective
 8                   epidemiologic evidence clearly demonstrates that short-term ambient O3 exposure is
 9                   associated with increases in respiratory symptoms and asthma medication use in children
10                   with asthma. In a smaller body of literature, several studies find associations in adults
11                   with asthma. In comparison, evidence has not consistently indicated that short-term O3
12                   exposure is associated with reduced activity levels in children or adults with asthma.
13                   Although O3 exposure has been associated with school absenteeism among children with
14                   asthma, only Gilliland et al. (2001) examined absences specifically for respiratory causes
15                   and found associations with O3 exposure lag periods shorter than 14 days. Epidemiologic
16                   studies do not provide consistent evidence of association between short-term ambient O3
17                   exposure and respiratory symptoms in children without asthma.

18                   Collectively, epidemiologic studies most frequently examined 1-h max and 8-h max or
19                   avg O3 exposures, and the few studies that examined both averaging times found similar
20                   magnitudes of associations with respriatory symptoms (Triche et al., 2006;  Delfino et al.,
21                   2003; Gent et al.. 2003). Several studies found increases in respiratory symptoms with O3
22                   exposures averaged over 12 to 24 hours (Triche et al., 2006; Jalaludin et al., 2004; Gold
23                   et al.. 1999; Neas et al.. 1999). Epidemiologic studies  examined associations of
24                   respiratory symptoms with single-day O3 concentrations lagged from 0 to 5 days as well
25                   concentrations averaged over 2 to 19 days. While O3 exposures lagged 0 or 1 days were
26                   consistently associated with respiratory symptoms, several studies that examined a range
27                   of exposure lags found larger effect estimates for multiday averages (3-  to 6-days) of O3
28                   exposure (Escamilla-Nunez et al.. 2008; Romieu et al.. 2006; Rabinovitch et al.. 2004;
29                   Just et al., 2002;  Mortimer et al., 2002; Ross et al.. 2002). These epidemiologic findings
30                   are in contrast with those from controlled human exposure studies that find attenuated
31                   symptom responses with O3 exposures repeated over several days (Section  6.2.1.1). The
32                   epidemiologic findings for lagged O3 exposures or those accumulated over several days
33                   are well-supported by the action of O3 to sensitize bronchial smooth muscle to
34                   hyperreactivity, thus  acting as a primer for subsequent exposure to antigens such as
      Draft - Do Not Cite or Quote                      6-100                                September 2011

-------
 1                   allergens (Section 5.3.5). In several of the studies of individuals with asthma, the
 2                   prevalence of atopy was high (50-100%), and sensitization of airways provides a
 3                   biologically plausible mode of action by which lagged or multiday average O3  exposures
 4                   are associated with increases in respiratory symptoms in these studies of individuals with
 5                   asthma.

 6                   Epidemiologic evidence did not indicate that associations between short-term O3
 7                   exposure and respiratory symptoms were confounded by temperature or pollen. In the
 8                   limited analysis of confounding by co-pollutants (primarily PM), robust associations with
 9                   respiratory symptoms were observed for O3; however, disentangling the independent
10                   effects of O3 exposure in many studies is complicated due to the high correlations
11                   observed between O3 and PM, different averaging times and lags of exposure examined
12                   for co-pollutants, and the multiple co-pollutants included in models. Nonetheless, the
13                   consistency of association among individuals with asthma with and without adjustment
14                   for copollutant exposures combined with evidence from controlled human exposure
15                   studies for the direct effect of O3 exposure provide substantial evidence for the
16                   independent effects of ambient O3 exposure on increases in respiratory symptoms
             6.2.5   Lung Host Defenses

17                   The mammalian respiratory tract has a number of closely integrated defense mechanisms
18                   that, when functioning normally, provide protection from the adverse effects of a wide
19                   variety of inhaled particles and microbes. For simplicity, these interrelated defenses can
20                   be divided into two major parts: (1) nonspecific (transport, phagocytosis, and bactericidal
21                   activity) and (2) specific (immunologic) defense mechanisms. A variety of sensitive and
22                   reliable methods have been used to assess the effects of O3 on these components of the
23                   lung's defense system to provide a better understanding of the health effects associated
24                   with the inhalation of this pollutant. The previous O3 AQCD states that animal
25                   toxicological studies provide extensive evidence that acute O3 exposures as low as 0.08 to
26                   0.5 ppm can cause increases in susceptibility to infectious diseases due to modulation of
27                   lung host defenses.  Tables 6-6 through 6-9, beginning on p. 6-41 of the 1996 O3 AQCD
28                   (U.S. EPA. 1996a). and Table AX5-7, beginning on p. AX5-8 of the 2006 O3 AQCD
29                   (U.S. EPA. 2006b). present studies on the effects of O3 on host defense mechanisms. This
30                   section discusses the various components of host defenses, such as the mucociliary
31                   escalator, the phagocytic, bactericidal, and regulatory role of the alveolar macrophages
32                   (AMs), the adaptive immune system, and integrated mechanisms that are studied by
33                   investigating the host's response to experimental pulmonary infections.
      Draft - Do Not Cite or Quote                      6-101                                September 2011

-------
                     6.2.5.1    Mucociliary Clearance

 1                   The mucociliary system is one of the lung's primary defense mechanisms. It protects the
 2                   conducting airways by trapping and quickly removing material that has been deposited or
 3                   is being cleared from the alveolar region by migrating alveolar macrophages. Ciliary
 4                   movement directs particles trapped on the overlying mucous layer toward the pharynx,
 5                   where the mucus is swallowed or expectorated.

 6                   The effectiveness of mucociliary clearance can be determined by measuring such
 7                   biological activities as the rate of transport of deposited particles; the frequency of ciliary
 8                   beating; structural integrity of the ciliated cells; and the size, number, and distribution of
 9                   mucus-secreting cells. Once this defense mechanism has been altered, a buildup of both
10                   viable and nonviable inhaled substances can occur on the epithelium and may jeopardize
11                   the health of the host, depending on the nature of the uncleared substance. Impaired
12                   mucociliary clearance can result in an unwanted accumulation of cellular secretions,
13                   increased infections, chronic bronchitis, and complications associated with chronic
14                   obstructive pulmonary disease. A number of previous  studies with various animal species
15                   have examined the effect of O3 exposure on mucociliary clearance and reported
16                   morphological damage to the cells of the tracheobronchial tree from acute and sub-
17                   chronic exposure to O3 0.2 ppm and higher. The cilia were either completely absent or
18                   had become noticeably shorter or blunt. After placing these animals in a clean-air
19                   environment, the structurally damaged cilia regenerated and appeared normal (U.S. EPA.
20                   1986). Based on such morphological observations, related effects such as ciliostasis,
21                   increased mucus secretions, and a slowing of mucociliary transport rates might be
22                   expected. However, no measurable changes in ciliary beating activity have been reported
23                   due to O3 exposure alone. Essentially no data are available on the effects of prolonged
24                   exposure to O3 on ciliary functional activity or on mucociliary transport rates measured in
25                   the intact animal. In general, functional studies of mucociliary transport have observed a
26                   delay in particle clearance soon after acute exposure. Decreased clearance is more
27                   evident at higher doses (1 ppm), and there is some evidence of tolerance/adaptation for
28                   these effects (U.S. EPA, 1986). However, no recent studies have evaluated the effects of
29                   O3 on mucociliary clearance.
                     6.2.5.2    Alveolobronchiolar Transport Mechanism

30                   In addition to the transport of particles deposited on the mucous surface layer of the
31                   conducting airways, particles deposited in the deep lung may be removed either up the
32                   respiratory tract or through interstitial pathways to the lymphatic system. The pivotal
33                   mechanism of alveolobronchiolar transport involves the movement of AMs with
      Draft - Do Not Cite or Quote                      6-102                                September 2011

-------
 1                   phagocytized particles to the bottom of the mucociliary escalator. Failure of the AMs to
 2                   phagocytize and sequester the deposited particles from the vulnerable respiratory
 3                   membrane can lead to particle entry into the interstitial spaces. Once lodged in the
 4                   interstitium, particle removal is more difficult and, depending on the toxic or infectious
 5                   nature of the particle, its interstitial location may allow the particle to set up a focus for
 6                   pathologic processes. Although some studies show reduced early (tracheobronchial)
 7                   clearance after O3 exposure, late (alveolar) clearance of deposited material is  accelerated,
 8                   presumably due to macrophage influx (which in itself can be damaging due to proteases
 9                   and oxidative reactions in these cells). In an important older study investigating the
10                   effects of longer term O3 exposure on alveolobronchiolar clearance, rats were exposed to
11                   an urban pattern of O3 (continuous 0.06 ppm, 7 days/week with a slow rise to a peak of
12                   0.25 ppm and subsequent decrease to 0.06 ppm over a 9 h period for 5 days/week) for
13                   6 weeks and were exposed 3 days later to chrysotile asbestos, which can  cause pulmonary
14                   fibrosis and neoplasia (Pinkerton et al. 1989). After 30 days, the lungs of the O3-exposed
15                   animals had twice the number and mass of asbestos fibers as the air-exposed  rats. New
16                   evaluations of O3 effects on alveolar clearance have not been performed.
                     6.2.5.3    Alveolar Macrophages

17                   Within the gaseous exchange region of the lung, the first line of defense against
18                   microorganisms and nonviable particles that reach the alveolar surface is the AM. This
19                   resident phagocyte is responsible for a variety of activities, including the detoxification
20                   and removal of inhaled particles, maintenance of pulmonary sterility via destruction of
21                   microorganisms, and interaction with lymphocytes for immunologic protection. Under
22                   normal conditions, AMs seek out particles deposited on the alveolar surface and ingest
23                   them, thereby sequestering the particles from the vulnerable respiratory membrane. To
24                   adequately fulfill their defense function, the AMs must maintain active mobility, a high
25                   degree  of phagocytic activity, and an optimally functioning biochemical and enzyme
26                   system for bactericidal activity and degradation of ingested material. As discussed in
27                   previous AQCDs,  short periods of O3 exposure can cause a reduction in the number of
28                   free AMs available for pulmonary defense, and these AMs are more fragile, less
29                   phagocytic, and have decreased lysosomal enzyme activities required for killing
30                   pathogens. For example, in results from earlier work in rabbits, a 2-h exposure to 0.1 ppm
31                   O3 inhibited phagocytosis and a 3-h exposure to 0.25 ppm decreased lysosomal enzyme
32                   activities (Driscoll et al..  1987; Hurst et al.. 1970). Similarly, AMs from rats exposed to
33                   0.1 ppm O3 for 1 or 3 weeks exhibited reduced hydrogen peroxide production (Cohen et
34                   al.. 2002). A controlled human exposure study reported  decrements in the ability of
3 5                   alveolar macrophages to phagocytize yeast following exposure of healthy volunteers to
      Draft - Do Not Cite or Quote                       6-103                                September 2011

-------
 1                   80 to 100 ppb O3 for 6.6-h during moderate exercise (Devlin et al., 1991). Although the
 2                   percentage of phagocytosis-capable macrophages was unchanged by O3 exposure, the
 3                   number of yeast engulfed was reduced when phagocytosis was complement-dependent.
 4                   However, there was no difference in the ability of macrophages to produce superoxide
 5                   anion after O3 exposure. These results are consistent with those from another controlled
 6                   human exposure study in which no changes in the level of lysosomal enzymes or
 7                   superoxide anion production were observed in macrophages lavaged from healthy human
 8                   subjects exposed to 400 ppb O3 for 2 h with heavy intermittent exercise (Koren et al..
 9                   1989). More recently, Lay et al. (2007) observed no difference in phagocytic activity or
10                   oxidative burst capacity in macrophages or monocytes from sputum or blood collected
11                   from healthy volunteers after a 2-hour exposure to 400 ppb O3 with moderate intermittent
12                   exercise. However, another study found that oxidative burst and phagocytic activity in
13                   macrophages increased in GSTM1 null subjects compared to GSTM1 positive subjects,
14                   who had relatively unchanged macrophage function parameters after an O3 exposure
15                   identical to that of Lay et al. described above (Alexis etal.. 2009). Collectively, these
16                   studies demonstrate that O3 can affect multiple steps or aspects required for proper
17                   macrophage function, but any concentration-response relationship appears complex and
18                   genotype may be a consideration. A few other recent studies have evaluated ozone's
19                   effects on macrophage function, but these are of questionable relevance due to the use of
20                   in vitro exposure systems  and amphibian animal models (Mikerov et al., 2008b; Dohm et
21                   al.. 2005: Klestadt et al.. 2005).
                    6.2.5.4   Infection and Adaptive Immunity

                    General Effects on the Immune System
22                  The effects of O3 on the immune system are complex and dependent on the exposure
23                  regimen and the observation period. According to toxicological studies it appears that the
24                  T-cell-dependent functions of the immune system are more affected than B-cell-
25                  dependent functions (U.S. EPA. 2006b). Generally, there is an early immunosuppressive
26                  effect that subsides with continued O3 exposure, resulting in either a return to normal
27                  responses or an enhancement of immune responses. However, this is not always the case
28                  as Aranyi (1983) showed decreased T-cell mitogen reactions in mice after  subchronic
29                  (90-day) exposure to 0.1 ppm O3. Earlier studies report changes in cell populations in
30                  lymphatic tissues (U.S. EPA. 2006b). A more recent study in mice demonstrated that
31                  numbers of certain T cell subsets in the spleen were reduced after exposure to 0.6 ppm O3
32                  (lOh/day x 15d) (Feng et al.. 2006).
      Draft - Do Not Cite or Quote                      6-104                                September 2011

-------
 1                   The inflammatory effects of O3 involve the innate immune system, and as such can affect
 2                   adaptive (or acquired) immunity via alterations in antigen presentation and costimulation
 3                   by innate immune cells such as macrophages and dendritic cells. Several recent
 4                   controlled human exposure studies demonstrate increased expression of molecules
 5                   involved in antigen presentation or costimulation. Lay et al. (2007) collected sputum
 6                   monocytes from healthy volunteers exposed to 400 ppb O3 for 2 h with moderate
 7                   intermittent exercise and detected increases in HLA-DR, used to present antigen to T
 8                   cells, and CD86, a costimulatory marker necessary for T cell activation. Upregulation of
 9                   HLA-DR was also observed by Alexis et al. (2009) in sputum dendritic cells and
10                   macrophages from GSTM1 null subjects exposed to 400 ppb O3 for 2 h with moderate
11                   intermittent exercise. On airway monocytes from healthy volunteers 24 hours after
12                   exposure to 80 ppb O3 for 6.6 h with moderate intermittent exercise, HLA-DR, CD86,
13                   and CD 14 (a molecule involved in bacterial endotoxin reacitivity) were increased,
14                   whereas CD80, a costimulatory molecule of more heterogeneous function, was decreased
15                   (Alexis et al.. 2010). Patterns of expression on macrophages were similar, except that
16                   HLA-DR was found to be significantly decreased after O3 exposure and CD86 was not
17                   significantly altered. An increase in IL-12p70, a macrophage and dendritic cell product
18                   that activates T cells, was correlated with increased numbers of dendritic cells. It should
19                   be noted that these results are reported as comparisons to baseline as there was no clean
20                   air control (Alexis et al.. 2010; Alexis et al.. 2009). Another controlled human exposure
21                   study reported no increase in IL-12p70  in sputum from healthy, atopic, or atopic
22                   asthmatic subjects following a  2-hour exposure to 400 ppb O3 with intermittent moderate
23                   exercise (Hernandez etal.. 2010). Levels of HLA-DR, CD14 and CD86 were not
24                   increased on macrophages collected from any of these subjects. It is difficult to compare
25                   these results to those of Lay et  al. (2007) and Alexis et al. (2010) due to differences in O3
26                   concentration, cell type examined, and timing of postexposure analysis.

27                   Although no controlled human exposure studies have examined the effects of O3 on the
28                   ability to mount antigen-specific  responses, upregulation of markers associated with
29                   innate immune activation and antigen presentation could potentially enhance adaptive
30                   immunity and increase immunologic responses to antigen. While this may bolster
31                   defenses against infection, it also may enhance allergic responses (Section 6.2.6).

32                   In animal models, O3 has been  found to alter responses to antigenic stimulation. For
33                   example,  antibody responses to a T-cell-dependent antigen were suppressed after a
34                   56-day exposure of mice to 0.8 ppm O3, and a 14-day exposure to 0.5 ppm O3 decreased
35                   the antiviral antibody response following influenza virus infection (Jakab and Hmieleski.
36                   1988); the latter impairment may pave the way for lowered resistance to reinfection. The
37                   immune response is highly influenced by the temporal relationship between O3 exposure
38                   and antigenic stimulation. When  O3 exposure preceded Listeria infection, there were no
      Draft - Do Not Cite or Quote                      6-105                                September 2011

-------
 1                   effects on delayed-type hypersensitivity or splenic lymphoproliferative responses;
 2                   however, when O3 exposure occurred during or after Listeria infection was initiated,
 3                   these immune responses were suppressed (Van Loveren et al., 1988). In another study, a
 4                   reduction in mitogen activated T-cell proliferation was observed after exposure to
 5                   0.6 ppm for 15 days, and could be ameliorated by antioxidant supplementation. Antigen-
 6                   specific proliferation decreased by 60%,  indicating attenuation of the acquired immunity
 7                   needed for subsequent memory responses (Feng etal.. 2006). O3 exposure also skewed
 8                   the ex-vivo cytokine responses elicited by non-specific stimulation toward inflammation,
 9                   decreasing IL-2 and increasing IFN-y. Modest decreases in immune function assessed in
10                   the offspring of O3-exposed dams (mice) were observed by Sharkhuu et al. (2011). The
11                   ability to mount delayed-type hypersensitivity responses was significantly suppressed in
12                   42 day-old offspring when dams were exposed to 0.8 or 1.2 ppm O3, but not 0.4 ppm,
13                   from gestational day 9-18. Humoral responses to immunization with sheep red blood
14                   cells were unaffected, as were other immune parameters  such as splenic populations of
15                   CD45+ T cells, iNKT cells, and levels of IFN-y, IL-4, and IL-17 in the BALF. Generally,
16                   continuous exposure to O3 impairs immune responses for the first several days of
17                   exposure, followed by an adaptation to O3 that allows a return of normal immune
18                   responses. Most species show little effect of O3 exposures prior to immunization, but
19                   show a suppression of responses to antigen in O3 exposures post-immunization.


                     Microbial Infection

                         Bacterial infection
20                   A relatively large body of evidence shows that O3 increases susceptibility to bacterial
21                   infections.  The majority of studies in this area were conducted before the 1996  O3 AQCD
22                   was published and many are included in Table 6-9 on p. 6-53 of that document. Known
23                   contributing factors are impaired mucociliary streaming, altered chemotaxis/motility,
24                   defective phagocytosis of bacteria, decreased production of lysosomal enzymes or
25                   superoxide radicals by alveolar macrophages, and decreased IFN^y levels. In animal
26                   models of bacterial infection,  exposure to 0.08 ppm O3 increases streptococcus-induced
27                   mortality, regardless of whether O3 exposure precedes or follows infection (Miller et al..
28                   1978; Coffin and Gardner. 1972;  Coffin et al., 1967). Increases in mortality are due to the
29                   infectious agent, thereby reflecting functional impairment of host defenses. Exercise and
30                   copollutants can enhance ozone's effects in infectivity models. Although both mice and
31                   rats exhibit impaired bactericidal macrophage activity after O3 exposure, mortality due to
32                   infection is only observed in mice. Additionally,  although mice and humans share many
33                   host defense mechanisms, there is little compelling evidence from epidemiologic studies
34                   (Section 6.2.7.3).
      Draft - Do Not Cite or Quote                      6-106                                September 2011

-------
                        Viral infection
 1                   Only a few studies, described in previous AQCDs, have examined the effects of O3
 2                   exposure on the outcome of viral respiratory infection (see Table 6-9 on p. 6-53 of the
 3                   1996 O3 AQCD. Some studies show increased mortality, while others show diminished
 4                   severity and increased survival time. There is little to no evidence from studies of animals
 5                   or humans to suggest that O3 increases the incidence of respiratory viral infection in
 6                   humans. In human volunteers infected with rhinovirus prior to O3 exposure (0.3 ppm for
 7                   5 consecutive days), no effect on viral titers, IFN-y production, or blood lymphocyte
 8                   proliferative responses to viral antigen was observed (Henderson et al.. 1988). In vitro
 9                   cell culture studies of human bronchial epithelial cells indicate O3-induced exacerbation
10                   of human rhinovirus infection  (Spannhake et al.. 2002). but this is of limited relevance.
11                   Newer studies on the interactions of O3 and  viral infections have  not been published.
12                   Natural killer (NK) cells, which destroy virally infected cells and tumors in the lung,
13                   appear to be inhibited by higher concentrations of O3 and either unaffected or stimulated
14                   at lower concentrations. Several studies show decreases in NK cell activity following
15                   acute exposures ranging from 0.8 to 1 ppm (Gilmour and Jakab. 1991; Van Loveren et
16                   al.. 1990; Burleson et al.. 1989). However, Van Loveren et al. (1990) showed that a
17                   1-week exposure to 0.2 or 0.4 ppm O3 increased NK cell activity, and an urban pattern of
18                   exposure (base of 0.06 ppm with peaks of 0.25 ppm) had no effect on NK cell activity
19                   after 1, 3, 13,  52, or 78 weeks of exposure (Selgrade et al.. 1990). A more recent study
20                   demonstrated a 35% reduction in NK cell activity after exposure  of mice to 0.6 ppm O3
21                   (lOh/day x 15d) (Feng et al.. 2006). The defective IL-2 production demonstrated in this
22                   study may impair NK cell activation. Alternatively, NK cell surface charge may be
23                   altered by ROS, decreasing their adherence to target cells (Nakamura and Matsunaga.
24                   1998).

                        Summary: Infections
25                   Taken as a whole, the data clearly indicate that an acute O3 exposure impairs the host
26                   defense capability of both humans and animals, primarily by depressing alveolar
27                   macrophage function and perhaps also by decreasing mucociliary clearance of inhaled
28                   particles and microorganisms.  This suggests that humans exposed to O3 could be
29                   predisposed to bacterial infections in the lower respiratory tract. The seriousness of such
30                   infections may depend on how quickly bacteria develop virulence factors and how
31                   rapidly PMNs are mobilized to compensate for the deficit in alveolar macrophage
32                   function. To date, a limited number of epidemiologic studies have examined associations
33                   between O3 exposure and HA/ED for respiratory infection, pneumonia, or influenza.
34                   Results have been mixed, and in some cases conflicting (see Sections 6.2.7.2 and
35                   6.2.7.3). With the exception of influenza, it is difficult to ascertain whether cases of
36                   respiratory infection or pneumonia are of viral or bacterial etiology. A study that

      Draft - Do Not Cite or Quote                      6-107                                September 2011

-------
 1                   examined the association between O3 exposure and respiratory hospital admissions in
 2                   response to an increase in influenza intensity did observe an increase in respiratory
 3                   hospital admissions (Wong et al.. 2009). but information from toxicological studies of O3
 4                   and viral infections is ambiguous.
             6.2.6   Allergic and Asthma-Related Responses

 5                   Effects resulting from combined exposures to O3 and allergens have been studied in a
 6                   variety of animal species, generally as models of experimental asthma. Pulmonary
 7                   function and airways hyperresponsiveness in animal models of asthma are discussed in
 8                   Sections 6.2.1.7 and 6.2.2.2. Previous evidence indicates that O3 exposure skews immune
 9                   responses toward an allergic phenotype. For example, Gershwin et al. (1981) reported
10                   that O3 (0.8 and 0.5 ppm for 4 days) exposure caused a 34-fold increase in the number of
11                   IgE (allergic antibody)-containing cells in the lungs of mice. In general, the number of
12                   IgE-containing cells correlated positively with levels of anaphylactic sensitivity. In
13                   humans, allergic rhinoconjunctivitis symptoms are associated with increases in ambient
14                   O3 concentrations (Riediker et al., 2001).  Recent controlled human exposure studies have
15                   observed O3-induced changes indicating allergic skewing. Airway eosinophils, which
16                   participate in allergic disease and inflammation, were observed to increase in atopic,
17                   mildly asthmatic volunteers 18 h following a 7.6-hour exposure to 160 ppb O3 with light
18                   intermittent exercise (Peden et al.,  1997).  No increase in airway eosinophils was observed
19                   4 h after exposure of healthy, atopic, or atopic asthmatic subjects to 400 ppb O3 for 2 h
20                   with moderate intermittent exercise (Hernandez et al.. 2010). However, atopic subjects
21                   did exhibit increased IL-5, a cytokine involved in eosinophil recruitment and activation,
22                   suggesting that perhaps these two studies  observed the same effect at different time
23                   points. Several epidemiologic studies discussed in Section 7.2.5 describe an association
24                   between eosinophils and long-term O3 exposure, consistent with chronic exposure studies
25                   in non-human primates. Hernandez et al. (2010) also observed increased expression of
26                   high and low affinity IgE receptors on sputum macrophages from atopic  asthmatics,
27                   which may enhance IgE-dependent inflammation. Sputum levels of IL-4 and IL-13, both
28                   pro-allergic cytokines that aid in the production of IgE,were unaltered in any group. The
29                   lack of increase in IL-4 levels in sputum reported by Hernandez et al., along with
30                   increased IL-5, is consistent with results from Bosson et al. (2003). in which IL-5 (but not
31                   IL-4 levels) increased in  bronchial  epithelial biopsy specimens following exposure of
32                   mild atopic asthmatics to 200 ppb O3 for 2 h with moderate intermittent exercise. IL-5
33                   was not elevated in specimens obtained from healthy (non-asthmatic) O3-exposed
34                   subjects. Collectively, findings from these studies suggest that O3 can induce or enhance
35                   certain components of allergic inflammation in atopic and atopic asthmatic individuals.
      Draft - Do Not Cite or Quote                       6-108                                September 2011

-------
 1                   Ozone enhances inflammatory and allergic responses to allergen challenge in sensitized
 2                   animals. Short-term exposure (2 days) to 1 ppm O3 exacerbated allergic rhinitis and lower
 3                   airway allergic inflammation in Brown Norway rats, a rat strain that is comparatively less
 4                   sensitive to O3 than other rats or humans (Wagner et al.. 2009; Wagner et al.. 2007).
 5                   OVA-sensitized rats were intranasally challenged with OVA on days  1 and 2, and
 6                   exposed to 0 or 1 ppm O3 (8 h/day) on days 4 and 5. Analysis at day 6 indicated that O3
 7                   exposure enhanced intraepithelial mucosubstances in the nose and airways, induced cys-
 8                   LTs, MCP-1, and IL-6 production in BALF, and upregulated expression of the
 9                   proallergic cytokines IL-5 and IL-13. These changes were not evident in non-allergic
10                   controls. All of these responses were blunted by gamma-tocopherol (yT; vitamin E)
11                   therapy. yT neutralizes oxidized lipid radicals, and protects lipids and proteins from
12                   nitrosative damage from NO-derived metabolites. Farraj et al. (2010)  exposed allergen-
13                   sensitized adult male BALB/c mice to 0.5 ppm O3 for 5 hours once per week for 4 weeks.
14                   Ozone exposure and O3/DEP (2.0 mg/m3) co-exposure of OVA-sensitized mice elicited
15                   significantly greater serum IgE levels than in DEP-exposed OVA-sensitized  mice (98%
16                   and 89% increases, respectively). Ozone slightly enhanced levels of BAL IL-5, but
17                   despite increases in IgE, caused a significant decrease in BAL IL-4 levels. IL-10, IL-13,
18                   and IFN-y levels were unaffected. Lung resistance and elastance were unaffected in
19                   allergen sensitized mice exposed solely to 0.5 ppm O3 once a week for 4 weeks (Farraj et
20                   al.. 2010). However, co-exposure to O3 and diesel exhaust particles increased lung
21                   resistance.

22                   In addition to exacerbating existing allergic responses, O3 can also act as an adjuvant to
23                   produce sensitization in the respiratory tract. In a model of murine asthma, using OVA
24                   free of detectable endotoxin, inclusion of 1 ppm O3 during the initial exposures to OVA
25                   (2 h, days 1 and 6) enhanced the inflammatory and allergic responses to subsequent
26                   allergen challenge (Hollingsworth et al.. 2010). Compared to air exposed animals, O3
27                   exposed mice exhibited significantly higher levels of total cells, macrophages,
28                   eosinophils, and PMNs in BALF, and increased total serum IgE. Pro-allergic cytokines
29                   IL-4, and IL-5 were also significantly elevated, along with pleiotropic Th2 cytokine IL-9
30                   (associated  with bronchial hyperresponsiveness) and pro-inflammatory IL-17, produced
31                   by activated T cells. Based on lower inflammatory, IgE, and cytokine responses in Toll-
32                   like receptor 4 deficient mice, the effects of O3 seem to be dependent  on TLR 4 signaling,
33                   as are a number of other biological responses to O3 according to studies by Hollingsworth
34                   et al. (2004). Kleeberger et al.QOOO) and Garanziotis et al. (2010). The involvement of
35                   TLR 4, along with its  endogenous ligand, hyaluronan, in O3-induced responses described
36                   in these studies has been corroborated by a controlled human exposure study by
37                   Hernandez et al. (2010). who found increased TLR 4 expression and elevated levels of
38                   hyaluronic acid in atopic and atopic asthmatic volunteers exposed to 400 ppb O3. This
39                   pathway is discussed in more detail in Chapter 5. Examination of dendritic cells (DCs)

      Draft - Do Not Cite or Quote                      6-109                                September 2011

-------
 1                  from the draining thoracic lymph nodes indicated that O3 did not enhance the migration
 2                  of DCs from the lungs to the lymph nodes, nor did it alter the expression of functional
 3                  DC markers such as CD40, MHC class II, or CD83. However, O3 did increase expression
 4                  of CD86, which is generally associated with Th2 responses and is detected at higher
 5                  levels on DCs from allergic asthmatics compared to those from healthy donors (Chen et
 6                  al..  20061)). Increased CD86 has also been observed on airway cells collected from
 7                  human subjects following exposure to O3 in studies by Lay et al. (2007) and Alexis et al.
 8                  (2009). but not Hernandez et al. (2010) (study details described in Section 6.2.5.4).

 9                  Ozone exposure during gestation has modest effects on allergy and asthma related
10                  endpoints in adult offspring. When dams were exposed to 1.2 ppm O3 (but not 0.8 ppm)
11                  from gestational day 9-18, some allergic and inflammatory  responses to OVA
12                  sensitization and challenge were reduced compared to air exposed controls. This included
13                  IgE levels and eosinophils, and was only true of mice that were immunized early in life
14                  (PND 3) as opposed to later (PND 42), perhaps due to the proximity of O3 and antigen
15                  exposure. The effects of gestational O3 exposure on immune function have not been
16                  widely studied, and although reductions in allergic endpoints are not generally observed
17                  in association with O3, other parameters of immune function were found to be reduced, so
18                  a more global immunosuppression may underlie these effects.

19                  In addition to ozone's pro-allergic effects, it could also make airborne allergens more
20                  allergenic. When combined with NO2, O3 has been shown to enhance nitration of
21                  common protein allergens, which may increase their allergenicity (Franze et  al.. 2005).
            6.2.7   Hospital Admissions, Emergency Department Visits, and Physicians
                    Visits
                    6.2.7.1    Summary of Findings from 2006 Ozone AQCD

22                  The 2006 O3 AQCD evaluated numerous respiratory ED visits and hospital admissions
23                  studies, which consisted primarily of time-series studies conducted in the U.S., Canada,
24                  Europe, South America, Australia and Asia. Upon collectively evaluating the scientific
25                  evidence, the 2006 O3 AQCD concluded that "the overall evidence supports a causal
26                  relationship between acute ambient O3 exposures and increased respiratory morbidity
27                  resulting in increased ED visits and [hospital admissions] during the warm season" (U.S.
28                  EPA. 2006b). This conclusion is "strongly supported by the human clinical, animal
29                  toxicologicfal], and epidemiologic evidence for [O3-induced] lung function decrements,
30                  increased respiratory symptoms, airway inflammation, and airway hyperreactivity" (U.S.
31                  EPA. 2006b).

      Draft - Do Not Cite or Quote                      6-110                                September 2011

-------
 1                   Since the completion of the 2006 O3 AQCD, relatively fewer studies conducted in the
 2                   U.S., Canada, and Europe have examined the association between short-term exposure to
 3                   ambient O3 and respiratory hospital admissions and ED visits with a growing number of
 4                   studies having been conducted in Asia. This section focuses primarily on multicity
 5                   studies because they examine the effect of O3 on respiratory-related hospital admissions
 6                   and ED visits over a large geographic area using a consistent statistical methodology.
 7                   Single-city studies that encompass a large number of hospital admissions or ED visits, or
 8                   included a long study-duration were also evaluated because these studies have more
 9                   power to detect whether an association exists between short-term O3 exposure and
10                   respiratory hospital admissions and ED visits compared to smaller single-city studies.
11                   Additional single-city studies were also evaluated within this section, if they were
12                   conducted in locations not represented by the larger single-city and multicity studies, or
13                   examined population-specific characteristics not included in the larger studies that may
14                   modify the association between short-term O3 exposure and respiratory-related hospital
15                   admissions or ED visits. The remaining single-city studies identified were not evaluated
16                   in this section due to factors such as inadequate  study design or insufficient sample size.

17                   It should be mentioned that when examining the association between short-term O3
18                   exposure and respiratory health effects that require medical attention, it is important to
19                   distinguish between hospital admissions and ED visits. This is because it is likely that a
20                   small percentage  of respiratory ED visits will be admitted to the hospital; therefore,
21                   respiratory ED visits may represent potentially less serious, but more common  outcomes.
22                   As a result, in the following sections respiratory hospital admission  and ED visit  studies
23                   are evaluated individually. Additionally, within  each section, results are presented as
24                   either a collection of respiratory diagnoses or as individual diseases  (e.g., asthma, COPD,
25                   pneumonia and other respiratory infections) in order to evaluate the  potential effect of
26                   short-term O3 exposure on each respiratory-related outcome. The ICD codes (i.e., ICD-9
27                   or ICD-10) that encompass each of these endpoints are presented in Table 6-25 along
28                   with the air quality characteristics of the city, or across all cities, included in each study
29                   evaluated in this section.
      Draft - Do Not Cite or Quote                      6-111                                 September 2011

-------
Table 6-25     Mean and upper percentile concentrations of respiratory-related
              hospital admission and emergency department visit studies
              evaluated
Study
Katsouyanni et
al. (2009)"'°
Cakmaketal.
(2006b)
Biggeri et al.
/9nn^°
\£.\J\J\Jf
Dales etal.
(2006)
Lin et al. (2008a)
Wong etal.
(2009)°
Medina-Ramon
etal. (2006)h
Yang etal.
(2005b)
Zanobetti and
Schwartz
(2006)"
Silverman and
lto(2010)b
Stieb et al.
(2009)
Tolbertetal.
(2007)
Darrow et al.
(201 1b)
Villeneuveetal.
(2007)b
Ito et al. (2007b)
Location
90 U.S. cities
(NMMAPS)d
32 European
cities (APHEA)d
12 Canadian
cities
10 Canadian
cities
4 Italian cities'
11 Canadian
cities
11 New York
regions
Hong Kong
36 U.S. cities
Vancouver,
Canada
Boston, MA
New York, NY
7 Canadian
cities
Atlanta, GA
Atlanta, GA
Alberta, CAN
New York, NY
Type of Visit (ICD9/10)
Hospital Admissions:
NMMAPS: All respiratory (460-
519)
APHEA:AII respiratory (460-519)
12 Canadian cities: All
respiratory (460-51 9)e
Hospital Admissions:
All respiratory (466, 480-486,
490,491,492,493,494,496)
Hospital Admissions:
All respiratory (460-51 9)
Hospital Admissions:
Respiratory disorders (486,
768.9, 769, 770.8, 786, 799.0,
799.1)
Hospital Admissions:
Respiratory diseases (466, 490-
493, 496)
Hospital Admissions:
All respiratory (460-519)
Hospital Admissions:
COPD (490-496, excluding 493)
Pneumonia (480-487)
Hospital Admissions:
COPD (490-492, 494, 496)
Hospital Admissions:
Pneumonia (480-487)
Hospital Admissions:
Asthma (493)
Emergency Department Visits:
Asthma (493)
COPD (490-492, 494-496)
Respiratory infection (464, 466,
480-487)
Emergency Department Visits:
All respiratory (460-465, 460.0,
466.1,466.11,466.19,477,480-
486,491,492,493,496,786.07,
786.09)
Emergency Department Visits:
All respiratory (460-466, 477,
480-486,491,492,493,496,
786.09)
Emergency Department Visits:
Asthma (493)
Emergency Department Visits:
Asthma (493)
Averaging
Time
1-hmax
24-h avg
8-h max
24-h avg
8-h maxs
8-h maxs
8-h max
24-h avg
24-h avg
8-h max
24-h avg
8-h max
8-h max
1-hmax
24-h avg
Commute
Day-time
Night-time
8-h max
8-h max
Mean
Concentration (ppb)a
NMMAPS:
50th: 34.9-60.0
APHEA:
50th: 11. 0-38.1
12 Canadian cities:
50th: 6.7-8.3
17.4
Warm season
(May-September): 5.7-60.0
17.0
44.1
18.8
Warm
(May-September): 45.8
Cool
(October-April): 27.6
All year: 14.1
Winter
(January-March): 13.2
Spring
(April-June): 19.4
Summer
(July-September): 13.8
Fall
(October-December): 10.0
22.4
Warm
(April-August):41.0
18.4
Warm: 53.0
Warm
(March-October):
8-h max: 53
1-hmax: 62
24-h avg: 30
Commute: 35'
Day-time: 45' .
Night-time: 14'
Summer
(April-September): 38.0
Winter
(October-March): 24.3
All year: 30.4
Warm
(April-September): 42.7
Cold
(October-March): 18.0
Upper Percentile
Concentrations (ppb)a
NMMAPS:
75th: 46.8-68.8
APHEA:
75th: 15.3-49.4
12 Canadian cities:
75th: 8.9-12.4
Max: 38.0-79.0
95th: 86. 1-90.0
Max: 107.5-115.1
95th: 24.9-46.0
75th: 54.0
Max: 21 7.0
75th: 25.9
Max: 100.3
NR
Max: 38.6
75th: 31.0
95th: 47.6
75th: 53
90th: 68
75th: 19.3-28.6
75th: 67.0
90th: 82.1
Max: 147.5
8-h 24-h avg:
max: 75th: 37
75th: 67 Max: 81
Max: Commute:
148 75th: 45
1-h Max: 106
max:
75th: 76
Max:
180
Summer:
75th: 46.0
Winter:
75th: 31 .5
All year:
95th: 68.0
Warm months:
95th: 77.0
Cold months:
95th: 33.0












Day-
time:
75th: 58
Max:
123
Night-
time:
75th: 22
Max: 64


Draft - Do Not Cite or Quote
6-112
September 2011

-------
Study
Strickland etal.
(2010)
Mar and Koenig
etal. (2009)
Arbexetal.
(2009)
Orazzoetal.
(2009)°
Burra et al.
(2009)
Villeneuve etal.
(2006b)
Sinclair etal.
(2010)'
Location
Atlanta, GA
Seattle, WA
Sao Paulo,
Brazil
6 Italian cities
Toronto,
Canada
Toronto,
Canada
Atlanta, GA
Type of Visit (ICD9/10)
Emergency Department Visits:
Asthma (493)
Wheeze (786.07 after 10/1/98,
786.09 before 10/1/98)
Emergency Department Visits:
Asthma (493-493.9)
Emergency Department Visits:
COPD (J40-44)
Emergency Department Visits:
Wheezing
Physician Visits:
Asthma (493)
Physician Visits:
Allergic rhinitis (177)
Physician Visits:
Asthma
Upper respiratory infection
Lower respiratory infection
Averaging
Time
8-h max
1-h max
8-h max
1-h max
8-h max"
1-h max
8-h max
8-h max
Mean
Concentration (ppb)a
All year: 45.41
Warm
(May-October): 55.2J
Cold
(November-April): 34.5J
Warm (May-October):
1-h max: 38.6
8-h max: 32.2
48.8
Summer
(April-September):
21.1-44.3
Winter
(October-March): 11.5-27.9
33.3
30.0
Total Study Period:
All-year: 44.0
25 mo Period:
All-year: 47.9
Warm: 61 .2
Cold: 27.8
28 mo Period:
All-year: 40.7
Warm: 51 .8
Cold: 26.0
Upper Percentile
Concentrations (ppb)a
NR
75th:
1-h max: 45.5
8-h max: 39.2
75th: 61.0
Max: 143.8
NR
95th: 66
Max: 121
Max: 98.7
NR
  aSome studies did not present an overall value for the mean, middle and/or upper percentiles of the 03 distribution; as a result, the range of the
mean, middle, and/or upper percentiles across all of the cities included in the study are presented.
  bStudy only presented median concentrations.
  °Study presented concentrations as ug/m3 Concentration was converted to ppb using the conversion factor of 0.51 assuming standard
temperature (25°C) and pressure (1 atm).
  dA subset of the European and U.S. cities included in the mortality analyses were used in the hospital admissions analyses: 8 of the 32 European
cities and 14 of 90 U.S. cities.
  eHospital admission data was coded using three classifications (ICD-10-CA, ICD-9, and ICD-9-CM). Attempts were made by the original
investigators to convert diagnosis from ICD-10-CA back to ICD-9.
  'Only 4 of the 8 cities included in the study collected 03 data.
  903 measured from 10:00 a.m. to 6:00 p.m.
  hOnly 35 of the 36 cities included in the analysis had 03 data.
  'Commute (7:00 a.m. to 10:00 a.m., 4:00 p.m. to 7:00 p.m.); Day-time (8:00 a.m. to 7:00 p.m.); Night-time (12:00 a.m. to 6:00 a.m.).
  'Means represent population-weighted 03 concentrations.
  k03 measured from 8:00 a.m. to 4:00 p.m.
  'This study did  not report the ICD codes used for the conditions examined. The 25-month period represents August 1998-August 2000, and the
28-month period  represents September 2000-December 2002. This study defined the warm months as April - October and the cold months as
November-March.
                  6.2.7.2    Hospital Admission Studies
1
2
3
4
5
                  Respiratory Diseases

                  The association between exposure to an air pollutant, such as O3, and daily respiratory-
                  related hospital admissions has primarily been examined using all respiratory-related
                  hospital admissions within the range of ICD-9 codes 460-519. Newly identified studies
                  attempt to further examine the effect of O3 exposure on respiratory-related hospital
                  admissions through a multicity design that examines O3 effects across countries using a
Draft - Do Not Cite or Quote
                                                             6-113
September 2011

-------
 1                   standardized methodology; multicity studies that examine effects within one country; and
 2                   multi- and single-city studies that attempt to examine potential modifiers of the O3-
 3                   respiratory-related hospital admission relationship.

 4                   The Air Pollution and Health: A European and North American Approach (APHENA)
 5                   study combined data from existing multicity study databases from Canada, Europe
 6                   (APHEA2) (Katsouvanni et al.. 2001). and the U.S. (NMMAPS) (Samet et al.. 2000) in
 7                   order to "develop more reliable estimates of the potential acute effects of air pollution on
 8                   human health [and] provide a common basis for [the] comparison of risks across
 9                   geographic areas" (Katsouyanni et al.. 2009). In an attempt to address both of these
10                   issues, the investigators conducted extensive sensitivity analyses to evaluate the
11                   robustness of the results to different model specifications (e.g., penalized splines [PS]
12                   versus  natural splines [NS]) and the extent of smoothing to control for seasonal and
13                   temporal trends. The trend analyses consisted of subjecting the models to varying extent
14                   of smoothing selected either a priori (e.g., 3 df/year, 8 df/year, and 12 df/year) or by
15                   using the absolute sum  of the residuals of the partial autocorrelation function (PACF).
16                   However, the investigators did not identify the model they deemed to  be the most
17                   appropriate for comparing the results across study locations. As a result, when discussing
18                   the results across the three study locations below, the 8 df/year results are presented  for
19                   both the PS and NS models because: (1) 8 df/year is most consistent with the extent  of
20                   temporal adjustment used in previous and recent large multicity studies in the U.S. (e.g.,
21                   NMMAPS); (2) the risk estimates for 8 df/year and 12 df/year are comparable for all
22                   three locations; (3) the models that used the PACF method did not report the actual
23                   degrees of freedom chosen; and (4) the 3 df/year and the PACF method resulted in
24                   negative O3 risk estimates, which is inconsistent with the results obtained using more
25                   aggressive seasonal adjustments. Additionally, when comparing results across studies in
26                   figures, only the results from one of the spline models (e.g., NS) are presented because it
27                   has been previously demonstrated that alternative spline models result in relatively
28                   similar effect estimates (HEI. 2003). However, it should be noted that the underlying data
29                   and model specifications  could result in varying degrees of bias and precision in effect
30                   estimates with different spline models (Ostro et al.. 2006).

31                   Katsouyanni et al. (2009) examined respiratory hospital admissions for people aged
32                   65 years and older using 1-h max O3 data. The extent of hospital admission and O3 data
33                   varied  across the 3 datasets: Canadian dataset included 12 cities with data for 3 years
34                   (1993-1996) per city; European dataset included 8 cities with each city having data for
35                   between 2 and 8 years from 1988-1997; and U.S. dataset included 14 cities with each city
36                   having data for between 4 and 10 years from 1985-1994 and  7 cities having only summer
37                   O3 data. The investigators used a three-stage hierarchical model to account for within-
3 8                   city, within region, and between region variability. Results were presented individually
      Draft - Do Not Cite or Quote                       6-114                                September 2011

-------
 1                   for each region (Figure 6-14; Table 6-26). Ozone and PM10 concentrations were weakly
 2                   correlated in all locations in the summer (r=0.27-0.40), but not in the winter.

 3                   In the Canadian cities, using all-year data, a 40 ppb increase in 1-h max O3
 4                   concentrations at lag 0-1 was associated with an increase in respiratory hospital
 5                   admissions of 8.9% (95% CI:  0.79, 16.8%) in a PS model and 8.1% (95% CI: 0.24,
 6                   16.8%) in a NS model (Katsouyanni et al., 2009). The results were somewhat sensitive to
 7                   the lag day selected, reduced when using a single-day lag (e.g., lag 1) (PS: 6.0%; NS:
 8                   5.5%) and increased when using a distributed lag model (PS: 18.6%; NS: 20.4%). When
 9                   adjusting for PMi0, the magnitude of the effect estimate was slightly larger in the NS
10                   model (5.1% [95% CI: -6.6, 18.6%]) compared to the PS model (3.1% [95% CI: -8.3,
11                   15.9%]); however, the copollutant analysis was only conducted using a 1-day lag. The
12                   large confidence intervals for both models could be attributed to the reduction in days
13                   included in the copollutant analyses as a result of the every-6th-day PM sampling
14                   schedule. When restricting the analysis to the summer months, stronger associations were
15                   observed between O3 and respiratory hospital admissions across the lags examined,
16                   ranging from -22 to 37% (the study does not specify whether these effect estimates are
17                   from a NS or PS model).  Because O3 concentrations across the cities included in the
18                   Canadian dataset (Katsouyanni et al. (2009) are low (median concentrations ranging from
19                   6.7-8.3 ppb  [Table 6-25]), the standardized increment of 40 ppb for a 1-h max increase in
20                   O3 concentrations does not accurately reflect  the observed risk of O3-related respiratory
21                   hospital admissions. Although this increment adequately characterizes the distribution of
22                   1-h max O3  concentrations across the U.S. and European datasets, it misrepresents the
23                   observed O3 concentrations in the Canadian dataset. As a result in summary figures, for
24                   comparability, effect estimates from the Canadian dataset are presented for both a 5.1 ppb
25                   increase in 1-h max O3 concentrations (i.e., an approximate interquartile range  [IQR]
26                   increase in O3 concentrations across the Canadian cities) as well as the standardized
27                   increment used throughout the ISA.

28                   In Europe, weaker but positive associations were also observed in year round analyses;
29                   2.9% (95% CI: 0.63, 5.0%) in the PS model and 1.6% (95% CI: -1.7, 4.2%) in the NS
30                   model at lag 0-1 for a 40 ppb increase in 1-h max O3 concentrations (Katsouyanni et al..
31                   2009). Additionally, at lag 1, associations between O3 and respiratory hospital admissions
32                   were also reduced, but in contrast to the lag 0-1 analysis, greater effects were observed in
33                   the NS model (2.9% [95% CI:  1.0, 4.9%]) compared to the PS model (1.5% [95% CI: -
34                   2.2, 5.4]). Unlike the Canadian analysis, a distributed lag model provided limited
35                   evidence of an association between O3 and respiratory hospital admissions. To  compare
36                   with the Canadian results, when adjusting for PM10 at lag 1, effect estimates were
37                   increased in the PS model (2.5% [95% CI: 0.39-4.8%]) and remained robust in the NS
38                   model (2.4% [95% CI: 0.08, 4.6%]). However, the European analysis also examined the
      Draft - Do Not Cite or Quote                      6-115                                September 2011

-------
1
2
3
4
5
6
7
effect of adjusting for PM10 at lag 0-1 and found results were attenuated in both models
(PS: 0.8% [95% CI: -2.3, 4.0%]; NS: 0.8% [95% CI: -1.8, 3.6%]). Unlike the Canadian
and U.S. datasets, the European dataset consisted of daily PM data. The investigators did
not observe stronger associations in the summer-only analyses for the European cities at
lag 0-1 (PS: 0.4% [95% CI: -3.2, 4.0%]; NS: 0.2% [95% CI: -3.3, 3.9%]), but did observe
some evidence for larger effects during the summer, an -2.5% increase, at lag 1 in both
models (the study does not present the extent of temporal smoothing used for these
models).
        Location       Lag

        U.S.            1
                       1
                      0-1
                      0-1
                     DL(0-2)
                      0-1
                       1

        Canada         1
                      la
                       1
                      la
                      0-1
                      0-la
                     DL(0-2)
                    DL(0-2)a
                       1
                      la
                      0-1
                      0-la
                     DL(0-2)
                    DL(0-2)a

        Europe         1
                       1
                      0-1
                      0-1
                     DL(0-2)
                       1
                      0-1
                                                       % Increase

      Black circles = all-year results; open circles = all-year results in copollutant model with PM10; and red circles = summer only
     results. For Canada, lag days with an "a" next to them represent the risk estimates standardized to an approximate IQR of 5.1 ppb
     for a 1-h max increase in ozone concentrations.


     Figure 6-14   Percent increase in respiratory hospital admissions from natural
                     spline models for a 40 ppb increase in  1-h max ozone
                     concentrations for each location of the APHENA study.












10 -5 C
— •— All-Year
-0 	
— • 	 Summer
• All Ynir
*-
Q
o-
• k

• ^
• ^
— •— Ail-Year
-• —
O 	

t

) 5 10 15 20 25 30 35 40
     Draft - Do Not Cite or Quote
                              6-116
September 2011

-------
      Table 6-26     Corresponding effect estimates for Figure 6-14
Location Season Lag3
Copollutant % Increase (95% Cl)°
U.S.
All-year 1
1
0-1
0-1
DL(0-2)
Summer 0-1
1
Canada All-year 1
1a
1
1a
0-1
0-1 a
DL(0-2)
DL(0-2)a
Summer 1
1a
0-1
0-1 a
DL(0-2)
DL(0-2)a
Europe All-year 1
1
0-1
0-1
DL(0-2)
Summer 1
0-1
2.62 (0.63, 4.64)
PM10 2. 14 (-0.08, 4.40)
2.38 (0.00, 4.89)
PM10 1.42 (-1.33, 4.23)
3.34 (0.02-6.78)
2. 14 (-0.63, 4.97)
2.78 (-0.02, 5.71)
5.54 (-0.94, 12.4)
0.69 (-0.12, 1.50)a
PM10 5.13 (-6.62, 18.6)
PM10 0.64 (-0.87, 2.20)a
8.12(0.24, 16.8)
1 .00 (0.03, 2.00)a
20.4 (4.07, 40.2)
2.4(0.51,4.40)3
21.4(15.0,29.0)
2.50(1.80,3.30)3
32.0(18.6,47.7)
3.60(2.20,5.10)3
37.1 (11.5,67.5)
4.1 (1.40,6.80)3
2.94(1.02,4.89)
PM10 2.38 (0.08, 4.64)
1.58 (-1.71, 4.15)
PM10 0.87 (-1.79, 3.58)
0.79 (-4.46, 6.37)
2.46 (-0.63, 5.54)
0.24 (-3.32, 3.91)
       aFor Canada, lag dsys with an "a" next to them represent the risk estimstes stsndsrdized to an approximate IQR of 5.1 ppb for a 1 -h max
      increase in 03 concentrations.
       bUnless noted, risk estimates standardized to 40 ppb for a 1 -h max increase in 03 concentrations.
 1                   For the U.S. in year round analyses, the investigators reported a 1.4% (95% CI: -0.9,
 2                   3.9%) increase in the PS model and 2.4% (95% CI: 0.0, 4.9%) increase in the NS model
 3                   in respiratory hospital admissions at lag 0-1 for a 40 ppb increase in 1-h max O3
 4                   concentrations with similar results for both models at lag 1 (Katsouyanni et al., 2009).
 5                   The distributed lag model provided results similar to those observed in the European
 6                   dataset with the PS model (1.1% [95% CI: -3.0, 5.3%]), but larger effects in the NS
 7                   model (3.3% [95% CI: 0.02, 6.8%]), which is consistent with the Canadian results. When
 8                   adjusting for PM10 using the U.S. data (i.e., every-6th-day PM data), results were
 9                   attenuated at lag 0-1 (PS: 0.6% [95% CI: -2.0, 3.3%]; NS: 1.4% [95% CI: -1.3, 4.2%])
10                   which is consistent with the results presented for the  European dataset. However, at lag 1,
11                   U.S. risk estimates remained robust to the inclusion of PMi0 in copollutant models as was
12                   observed in the Canadian and European datasets. Compared to the all-year analyses, the
13                   investigators did not observe stronger associations in the summer-only analysis at either
14                   lag  0-1 (-2.2%) or lag 1 (-2.8%) in both the PS and NS models (the study does not
15                   present the extent of temporal smoothing used for these models).

16                   Several additional multicity studies examined respiratory disease hospital admissions in
17                   Canada and Europe. Cakmak et al. (2006b) evaluated the association between ambient O3
      Draft - Do Not Cite or Quote
6-117
September 2011

-------
 1                   concentrations and respiratory hospital admissions for all ages in 10 Canadian cities from
 2                   April 1993 to March 2000. The primary objective of this study was to examine the
 3                   potential modification of the effect of ambient air pollution on daily respiratory hospital
 4                   admissions by education and income using a time-series analysis conducted at the city-
 5                   level. The authors calculated a pooled estimate across cities for each pollutant using a
 6                   random effects model by first selecting the lag day with the strongest association from the
 7                   city-specific models. For O3, the mean lag day across cities that provided the strongest
 8                   association and for which the pooled effect estimate was calculated was 1.2 days. In this
 9                   study, all-year O3 concentrations were used in the analysis, and additional seasonal
10                   analyses were not conducted. Cakmak et al. (2006b) reported a 4.4% increase (95% CI:
11                   2.2, 6.5%) in respiratory hospital admissions for a 20 ppb increase in 24-h average O3
12                   concentrations. The investigators only examined the potential effect of confounding by
13                   other pollutants through the use of a multipollutant model (i.e., two or more additional
14                   pollutants included in the model), which is difficult to interpret due to the potential
15                   multicollinearity between pollutants. Cakmak et al.  (2006b) also conducted an extensive
16                   analysis of potential modifiers, specifically sex, educational attainment, and family
17                   income, on the association between air pollution and respiratory hospital admissions.
18                   When stratifying by sex, the increase in respiratory hospital admissions due to short-term
19                   O3 exposure were similar in males (5.2% [95% CI: 3.0, 7.3%]) and females (4.2% [95%
20                   CI: 1.8, 6.6%]). In addition, the examination of effect modification by income found no
21                   consistent trend across the quartiles of family income. However, there was evidence that
22                   individuals with an education level less than the 9th grade were disproportionately
23                   affected by O3 exposure (4.6% [95% CI: 1.8, 7.5%]) compared to individuals that
24                   completed grades 9-13 (1.7% [95% CI: -1.9,  5.3%]), some university or trade school
25                   (1.4% [95% CI: -2.0, 5.1%]), or have a university diploma (0.66% [95% CI:  -3.3, 4.7%]).
26                   The association between O3 and individuals with an education level less than the 9th
27                   grade was the strongest association across all of the pollutants examined.

28                   A multicity study conducted in Europe by Biggeri et al.  (2005) examined the association
29                   between short-term O3 exposure and respiratory hospital admissions for all ages in four
30                   Italian cities from 1990 to 1999. In this study, O3 was only measured during the warm
31                   season (May-September). The authors examined associations between daily respiratory
32                   hospital admissions and short-term O3 exposure at the city-level using a time-series
33                   analysis. Pooled estimates were calculated by combining city-specific estimates using
34                   fixed and random effects models. The investigators found no evidence of an association
35                   between O3 exposure and respiratory hospital admissions in the warm season in both the
36                   random (0.1% [95% CI: -5.2, 5.7%]; distributed lag 0-3) and fixed effects (0.1% [95%
37                   CI: -5.2, 5.7%]; distributed lag 0-3) models for a 30 ppb increase in 8-h max O3
38                   concentrations.
      Draft - Do Not Cite or Quote                       6-118                                September 2011

-------
 1                   Additional studies examined associations between short-term O3 exposure and respiratory
 2                   hospital admissions specifically in children. In a multicity study conducted in Canada,
 3                   Dales et al. (2006) examined the association between all-year ambient O3 concentrations
 4                   and neonatal (ages 0-27 days) respiratory hospital admissions in 11 Canadian cities from
 5                   1986 to 2000. The investigators used a statistical analysis approach similar to Cakmak et
 6                   al. (2006b) (i.e., time-series analysis to examine city-specific associations, and then a
 7                   random effects model to pool estimates across cities). The authors reported that for O3
 8                   the mean lag day across cities that provided the strongest association was 2 days. The
 9                   authors reported a 5.4% (95% CI: 2.9, 8.0%) increase in neonatal respiratory hospital
10                   admissions for a 20 ppb increase in 24-h avg O3 concentrations at lag-2 days.  The results
11                   from Dales et al. (2006) provide support for the associations observed in a smaller scale
12                   study that examined O3 exposure and pediatric respiratory hospital admissions in
13                   New York state (Lin et al.. 2008a). Lin et al. (2008a) observed a positive association
14                   between O3 and pediatric (i.e., <18 years) respiratory admissions  at lag 2 (results not
15                   presented quantitatively) in a two-stage Bayesian hierarchical model analysis of 11
16                   geographic regions of New York from  1991 to 2001.

17                   Overall, the evidence from epidemiologic studies continues to support an association
18                   between short-term O3 exposure and respiratory-related hospital admissions, but it
19                   remains unclear whether certain factors (individual- or population-level) modify this
20                   association. Wong et al. (2009) examined the potential modification of the relationship
21                   between ambient O3 (along with NO2, SO2, and PM10) and respiratory hospital
22                   admissions by influenza intensity in Hong Kong for the period  1996 - 2002. Influenza
23                   intensity was defined as a continuous variable using the proportion of weekly specimens
24                   positive for influenza A or B instead of defining influenza epidemics. This approach was
25                   used to avoid any potential bias associated with the unpredictable seasonality of influenza
26                   in Hong Kong (Wong et al.. 2009). In models that examined the baseline effect (i.e.,
27                   without taking into consideration influenza intensity) of short-term O3 exposure, the
28                   authors found a 3.6% (95% CI: 1.9, 5.3%) and 3.2% (95% CI: 1.0, 5.4%) increase in
29                   respiratory hospital admissions at lag 0-1 for a 30 ppb increase in 8-h max O3
30                   concentrations for the all age and > 65 age groups, respectively. When examining
31                   influenza intensity, Wong et al. (2009) reported that the association between short-term
32                   exposure to O3 and respiratory hospital admissions was stronger with higher levels of
33                   influenza intensity: additional increase in respiratory hospital admissions above baseline
34                   of 1.4% (95% CI: 0.24, 2.6%) for all age groups and 2.4% (95% CI: 0.94, 3.8%) for those
35                   65 and older when influenza activity increased from 0% to 10%. No difference in effects
36                   was observed when stratifying by sex.
      Draft - Do Not Cite or Quote                       6-119                                September 2011

-------
                    Cause-Specific Respiratory Outcomes

 1                  In the 2006 O3 AQCD a limited number of studies were identified that examined the
 2                  effect of short-term O3 exposure on cause-specific respiratory hospital admissions. The
 3                  limited evidence "reported positive O3 associations with... asthma and COPD,
 4                  especially... during the summer or warm season" (U.S. EPA. 2006b). Of the studies
 5                  evaluated since the completion of the 2006 O3 AQCD, more have focused on identifying
 6                  whether O3 exposure is associated with specific respiratory-related hospital admissions,
 7                  including COPD, pneumonia, and asthma, but the overall body of evidence remains
 8                  small.

                        Chronic Obstructive Pulmonary Disease
 9                  Medina-Ramon et al. (2006) examined the association between short-term exposure to
10                  ambient O3 and PM10 concentrations and Medicare hospital admissions among
11                  individuals > 65  years  of age for COPD in 35 cities in the U.S. for the years 1986-1999.
12                  The cities included in this analysis were selected because they monitored PM10 on a daily
13                  basis. In this study, city-specific results were obtained using a monthly time-stratified
14                  case-crossover analysis. A meta-analysis was then conducted using random effects
15                  models to combine the city-specific results. All cities measured O3 from May through
16                  September, while only 16 of the cities had year-round measurements. The authors
17                  reported a 1.6% increase (95% CI:  0.48, 2.9%) in COPD admissions for lag 0-1 in the
18                  warm season for a 30 ppb increase in 8-h max O3 concentrations. When examining
19                  single-day lags, stronger associations were observed for lag 1 (2.9%  [95% CI: 1.8, 4.0%])
20                  compared to lag  0 (-1.5% [95% CI: -2.7, -0.24%]). The authors found no evidence of
21                  associations in cool season (-1.9%  [95% CI: -3.6, -0.06%]; lag 0-1) or year round (0.24%
22                  [95% CI: -0.78, 1.2%]; lag 0-1) analyses. In a copollutant model using warm season data,
23                  the association between O3 and COPD hospital admissions was robust to the  inclusion of
24                  PM10 in the model (results not presented quantitatively). The authors conducted
25                  additional analyses to examine potential modification of the warm season estimates for
26                  O3 and COPD admissions by several city-level characteristics: percentage living in
27                  poverty, emphysema mortality rate (as an indication of smoking), daily summer apparent
28                  temperature, and percentage of households using central air conditioning. Of the city-
29                  level characteristics examined, stronger associations were only reported for cities with a
3 0                  larger variability in daily apparent summer temperature.

31                  In a single-city study conducted in Vancouver from 1994-1998, a location with low
32                  ambient O3 concentrations (Table 6-25), Yang et al. (2005b) examined the association
33                  between O3 and COPD. Ozone was moderately inversely correlated with CO (r=-0.56),
34                  NO2 (r=-0.32), and SO2 (r=-0.34), and weakly inversely correlated with PM10 (r=-0.09),
3 5                  suggesting that the observed O3 effect is likely not only due to a positive correlation with


      Draft - Do Not Cite or Quote                     6-120                               September 2011

-------
 1                   other pollutants. Yang et al. (2005b) examined 1- to 7-day (e.g., (0-6 days) lagged
 2                   moving averages and observed an 8.8% (95% CI: -12.5, 32.6%) increase in COPD
 3                   admissions for lag 0-3 per 20 ppb increase in 24-h avg O3 concentrations. In two-
 4                   pollutant models at lag 0-3, O3 effect estimates were robust to the inclusion of NO2, SO2,
 5                   and PM10 in the model, but were increased slightly when adding CO (Figure 6-20; Table
 6                   6-28).

                        Pneumonia
 7                   In addition to COPD, Medina-Ramon et al. (2006) examined the association between
 8                   short-term exposure to ambient O3 and PMi0 concentrations and Medicare hospital
 9                   admissions among individuals > 65 years of age for pneumonia (ICD-9: 480-487). The
10                   authors reported an increase in pneumonia hospital admissions in the warm season (2.5%
11                   [95% CI: 1.6, 3.5%] for a 30 ppb increase in 8-h max  O3 concentrations; lag 0-1). Similar
12                   to the results observed for COPD hospital admissions, pneumonia hospital admissions
13                   associations were stronger at lag 1 (2.6%  [95% CI: 1.8, 3.4%]) compared to lag 0 (0.06%
14                   [95% CI: -0.72, 0.78%]), and no evidence of an association was observed in the cool
15                   season or year round. In two-pollutant models, the association between O3 exposure and
16                   pneumonia hospital admissions was robust to the inclusion of PMi0 (results not presented
17                   quantitatively).  The authors also examined potential effect modification of the warm
18                   season estimates for O3-related pneumonia hospital admissions, as was done for COPD,
19                   by several city-level characteristics. Stronger associations were reported in cities with a
20                   lower percentage of central air conditioning use. Across the cities examined, the
21                   percentage of households having central air conditioning ranged from 6 to 93%. The
22                   authors found no evidence of effect modification of the O3-pneumonia hospital admission
23                   relationship when examining the other city-level characteristics.

24                   Results from a single-city study conducted in Boston did not support the results presented
25                   by Medina-Ramon et al. (2006). Zanobetti and Schwartz (2006) examined the association
26                   of O3 and pneumonia Medicare hospital admissions for the period 1995-1999. Ozone was
27                   weakly positively correlated with PM2 5 (r=0.20) and weakly inversely correlated with
28                   black carbon, NO2, and CO (-0.25, -0.14,  and -0.30, respectively). In an all-year analysis,
29                   the investigators reported a 3.8% (95% CI: -7.9, -0.1%) decrease in pneumonia
30                   admissions for a 20 ppb increase in 24-h average O3 concentrations at lag 0 and a 6.0%
31                   (95% CI: -11.1, -1.4%) decrease for the average of lags 0 and 1. It should be noted that
32                   the mean daily counts of pneumonia admissions was low for this study, ~14 admissions
33                   per day compared to -271 admissions per day for Medina-Ramon et al.  (2006). However,
34                   in analyses with other pollutants Zanobetti and Schwartz (2006) did observe positive
3 5                   associations with pneumonia hospital admissions, indicating that the low number of daily
      Draft - Do Not Cite or Quote                      6-121                                September 2011

-------
 1                   hospital admission counts probably did not influence the O3-pneumonia hospital
 2                   admissions association in this study. .

                        Asthma
 3                   There are relatively fewer studies that examined the association between short-term
 4                   exposure to O3 and asthma hospital admissions, presumably due to the limited power
 5                   given the  relative rarity of asthma hospital admissions compared to ED or physician
 6                   visits. A study from New York City examined the association of 8-h max O3
 7                   concentrations with severe acute asthma admissions (i.e., those admitted to the Intensive
 8                   Care Unit [ICU]) during the warm season in the years  1999 through 2006 (Silverman and
 9                   Ito, 2010). In this study, O3 was moderately correlated with PM10 (r=0.59). When
10                   stratifying by age, the investigators reported positive associations with ICU asthma
11                   admissions for the 6-to 18-year age group (26.8% [95% CI: 1.4, 58.2%] for a 30 ppb
12                   increase in maximum 8-h avg O3 concentrations at lag 0-1), but little evidence of
13                   associations for the other age groups examined (<6 years, 19-49, 50+, and all ages).
14                   However, positive associations were observed for each age-stratified group  and all ages
15                   for non-ICU asthma admissions, but again the strongest association was reported for the
16                   6- to 18-years age group (28.2% [95% CI: 15.3, 41.5%]; lag 0-1). In two-pollutant
17                   models, O3 effect estimates for both non-ICU and ICU hospital admissions remained
18                   robust to adjustment for PM2 5. In an additional analysis, using a smooth function, the
19                   authors examined whether the shape of the  C-R curve for O3 and asthma hospital
20                   admissions (i.e., both general and ICU for all ages) is linear. To account for the potential
21                   confounding effects of PM25,  Silverman and Ito (2010) also included a smooth function
22                   of PM2 5 lag 0-1. When comparing the curve to a linear fit line the authors found that the
23                   linear fit is a reasonable approximation of the concentration-response relationship
24                   between O3 and asthma hospital admissions around and below the level of the current
25                   NAAQS (Figure 6-15).
      Draft - Do Not Cite or Quote                      6-122                                September 2011

-------
                                                  Ozone: All Ages
                            cc
                            cc
                               o>
                               o
                                        20
 I
40
    l
   60
Ozone
80
100
       Source: Used with permission from American Academy of Allergy, Asthma & Immunology (Silverman and Ito. 2010).
       The average of 0 day and 1 day lagged 8-h ozone was used in a two-pollutant model with PM2 5 lag 0-1, adjusting for temporal
      trends, day of the week, and immediate and delayed weather effects. The solid lines are smoothed fit data, with long broken lines
      indicating 95% confidence bands. The density of lines at the bottom of the figure indicates sample size.

      Figure 6-15    Estimated relative risks (RRs) of ozone-related asthma hospital
                       admissions allowing for possible nonlinear relationships using
                       natural splines.Averting  Behavior
 1                   The studies discussed above have found consistent positive associations between short-
 2                   term O3 exposure and respiratory-related hospital admissions, however, the strength of
 3                   these associations may be underestimated due to the studies not accounting for averting
 4                   behavior. As discussed in Section 4.6.4, recent studies by Neidell (2009) and Neidell and
 5                   Kinney (2010) conducted in Souther California demonstrate that controlling for
 6                   avoidance behavior increases O3 effect estimates for respiratory hospital admissions,
 7                   specifically for children and older adults. These studies show that on days where no
 8                   public alert was issued warning of high O3 concentrations there was an increase in asthma
 9                   hospital admissions. Although only a few epidemiologic studies have examined averting
10                   behavior and these studies are limited to asthma hospital admissions, they do provide
11                   preliminary evidence indicating that epidemiologic studies may underestimate
      Draft - Do Not Cite or Quote
   6-123
                                   September 2011

-------
 1                   associations between O3 exposure and health effects by not accounting for behaviorial
 2                   modification when public health alerts are issued.
                     6.2.7.3    Emergency Department Visit Studies

 3                   Overall, relatively fewer studies have examined the association between short-term O3
 4                   exposure and respiratory-related ED visits, compared to hospital admissions. In the 2006
 5                   O3 AQCD, positive, but inconsistent, associations were observed between O3 and
 6                   respiratory-related ED visits with effects generally occurring during the warm season.
 7                   Since the completion of the previous AQCD, larger studies have been conducted, in
 8                   terms of sample size, study duration, and in some cases multiple cities, to examine the
 9                   association between O3 and ED visits for all respiratory diseases, COPD, and asthma.

                        Respiratory Disease
10                   A large single-city study conducted in Atlanta, by Tolbert et al. (2007). and subsequently
11                   reanalyzed by Darrow et al. (20 lib), provides evidence for an association between short-
12                   term  exposures to ambient O3 concentrations and respiratory ED visits. Tolbert et al.
13                   (2007) examined the association between air pollution, both gaseous pollutants and PM
14                   and its components, and respiratory disease ED visits in all ages from 1993 to 2004. The
15                   correlations between O3 and the other pollutants examined ranged from 0.2 for CO and
16                   SO2 to 0.5-0.6 for the  PM measures. Using an a priori average of lags 0-2 for each air
17                   pollutant examined, the authors reported a 3.9% (95% CI: 2.7, 5.2%) increase in
18                   respiratory ED visits for a 30 ppb increase in 8-h max O3 concentrations during the warm
19                   season [defined as March-October in Darrow et al. (20lib)]. In copollutant models, the
20                   O3 associations with respiratory ED visits remained robust with CO, NO2, and PM10
21                   (results not presented  quantitatively).

22                   Darrow et al. (20 lib) examined the same data as Tolbert et al. (2007). but explored
23                   whether differences exist in the association between O3 exposure and respiratory-related
24                   ED visits depending on the exposure metric used (i.e., 8-h max, 1-h max, 24-h average,
25                   commuting period [7:00 a.m. to 10:00 a.m.; 4:00 p.m. to 7:00 p.m.], day-time [8:00 a.m.
26                   to 7:00 p.m.] and night-time [12:00 a.m. to 6:00 a.m.]). To examine the association
27                   between the various O3 exposure metrics and respiratory ED visits, the authors used a
28                   time-stratified case-crossover approach, selecting control days as those days within the
29                   same calendar month and maximum temperature as the case day. Darrow et al. (20 lib)
30                   found at lag 1, the results were somewhat variable across exposure metrics. The strongest
31                   associations with respiratory ED visits were found when using the 8-h max, 1-h max, and
32                   day-time exposure metrics with weaker associations using the 24-h avg and commuting
      Draft - Do Not Cite or Quote                      6-124                                September 2011

-------
 1
 2
period exposure metrics; a negative association was observed when using the night-time
exposure metric (Figure 6-16).
                                  1.03 T

                              0
                              y  1.02
                           Ss
                           & 8-   1.01
                           je o
                           * -C
                              9
                              a.  0^g
                            Partial
                            Spearman r.
                        1    0.95  0.93   0.83  078  0.04
• XBiujq-g
>,
a
E
v-
commute -
4)
^-
CM
<->
_g>
"E
       Source: Used with permission from Nature Publishing Group (Darrow et al.. 2011 b).

      Figure 6-16    Risk ratio for respiratory ED visits and different ozone exposure
                      metrics in Atlanta from 1993-2004.
 4
 5
 6
 1
 8
 9
10
11
12
13
In an additional study conducted in 6 Italian cities, Orazzo et al. (2009) examined
respiratory ED visits for ages 0-2 years in 6 Italian cities from 1996 to 2000. However,
instead of identifying respiratory ED visits using the traditional approach of selecting
ICD codes as was done by Tolbert et al. (2007) and Darrow et al. (20lib). Orazzo et al.
(2009) used data on wheeze extracted from medical records as an indicator of lower
respiratory disease. This study examined daily counts of wheeze in relation to air
pollution using a time-stratified case-crossover approach in which control days were
matched on day of week in the  same month and year as the case day. The authors found
no evidence of an association between 8-h max O3 concentrations and respiratory ED
visits in children aged 0-2 years in models that examined both single-day lags and
moving averages  of lags from 0-6 days in year-round and seasonal analyses (i.e., warm
      Draft - Do Not Cite or Quote
                              6-125
September 2011

-------
 1                   and cool seasons). In all-year analyses, the percent increase in total wheeze ranged from -
 2                   1.4% to -3.3% for a 0-1 to 0-6 day lag, respectively.

                         COPD
 3                   Stieb et al. (2009) also examined the association between short-term O3 exposure and
 4                   COPD ED visits in 7 Canadian cities. Across cities, in an all-year analysis, O3 was found
 5                   to be positively associated with COPD ED visits (4.0% [95% CI: -0.54, 8.6%] at lag 2 for
 6                   a 20 ppb increase in 24-h avg O3 concentrations). In seasonal analyses, larger effects
 7                   were observed between O3 and COPD ED visits during the warm season (i.e., April-
 8                   September) 6.8% [95% CI: 0.11, 13.9%] (lag day not specified); with no associations
 9                   observed in the winter season. Stieb et al. (2009) also examined associations between
10                   respiratory-related ED visits, including COPD, and air pollution at sub-daily time scales
11                   (i.e., 3-h avg of ED visits versus 3-h avg pollutant concentrations) and found no evidence
12                   of consistent associations between any pollutant and any respiratory outcome.

13                   In a single-city study, Arbex et al. (2009) examined the association between COPD and
14                   several ambient air pollutants, including O3, in Sao Paulo, Brazil for the years  2001-2003
15                   for individuals over the age of 40. Associations between O3 exposure and COPD ED
16                   visits were examined in both single-day lag (0-6 days) and polynomial distributed lag
17                   models (0-6 days). In all-year analyses, O3 was not found to be associated with an
18                   increase in COPD ED visits (results not presented quantitatively). The authors also
19                   conducted stratified analyses to examine the potential modification of the air pollutant-
20                   COPD ED visits relationship by age (e.g., 40-64, >64) and sex. In these analyses O3 was
21                   found to have an increase in COPD ED visits for women, but not for men or either of the
22                   age groups examined.

                         Asthma
23                   In a study of 7 Canadian cities, Stieb et al. (2009) also examined the association between
24                   exposure to air pollution (i.e.,  CO, NO2, O3, SO2, PM10, PM2.5, and O3) and asthma ED
25                   visits. Associations between short-term O3 exposure and asthma ED visits were examined
26                   at the city level and then pooled using either fixed or random effects models depending
27                   on whether heterogeneity among effect estimates was found to be statistically significant.
28                   Across cities, in an all-year analysis, the authors found that short-term O3 exposure was
29                   associated with a positive increase (3.5% [95% CI: 0.33, 6.8%] at lag 2 for a 20 ppb
30                   increase in 24-h avg O3 concentrations) in asthma ED visits. The authors did not present
31                   the results from seasonal analyses for asthma, but state that no associations were
32                   observed between any pollutant and respiratory ED visits in the winter season. As stated
33                   previously, in analyses of 3-h avg O3 concentrations, the authors observed no evidence of
34                   consistent associations between any pollutant and any respiratory outcome, including
      Draft - Do Not Cite or Quote                       6-126                                September 2011

-------
 1                   asthma. A single-city study conducted in Alberta, Canada (Villeneuve et al., 2007) from
 2                   1992-2002 among individuals two years of age and older provides additional support for
 3                   the findings from Stieb et al. (2009). but also attempts to identify those lifestages (i.e., 2-
 4                   4, 5-14, 15-44, 45-64, 65-74, or 75+) most susceptible to O3-induced asthma ED visits. In
 5                   a time-referent case-crossover analysis, Villeneuve et al. found an increase in asthma ED
 6                   visits in an all-year analysis across all ages (12.0% [95% CI: 6.8, 17.2] for a 30 ppb
 7                   increase in max 8-h avg O3 concentrations at lag 0-2) with associations being stronger
 8                   during the warmer months (19.0% [95% CI: 11.9, 28.1]). When stratifying by age, the
 9                   strongest associations were observed in the warm season for individuals 5-14 (28.1%
10                   [95% CI: 11.9, 45.1]; lag 0-2)  and 15-44 (19.0% [95% CI: 8.5, 31.8]; lag 0-2). These
11                   associations were not found to be confounded by the inclusion of aeroallergens in age-
12                   specific models.

13                   Several additional single-city studies have also provided evidence of an association
14                   between asthma ED visits and ambient O3 concentrations. Ito et al. (2007b) examined the
15                   association between short-term exposure to air pollution and asthma ED visits for all ages
16                   in New York City from 1999 to 2002. Ito et al. (2007b) used three different weather
17                   models with varying extent of smoothing to account for temporal relationships and
18                   multicollinearity among pollutants and meteorological variables  (i.e., temperature and
19                   dew point) to examine the effect of model selection on the air pollutant-asthma ED visit
20                   relationship. When examining O3, the authors reported a positive association with asthma
21                   ED visits, during the warm season across the models (ranging from  8.6 to 16.9%) and an
22                   inverse association in the cool season (ranging from -23.4 to -25.1%), at lag 0-1 for a 30
23                   ppb increase in 8-h max O3 concentrations. Using a simplified version of the weather
24                   model used in NMMAPS analyses (i.e., terms for same-day temperature and 1-3 day
25                   average temperature), Ito et al. (2007b) found that O3 effects were not substantially
26                   changed in copollutant models with PM2 5, NO2, SO2, and CO during the warm season
27                   (Figure 6-19; Table 6-27).
      Draft - Do Not Cite or Quote                       6-127                                 September 2011

-------
                                       Ozone Warm Season
                                     40     50      60     70
                                        Concentration (ppb)
                                                                        80
  Source: Used with permission from American Thoracic Society (Strickland et al.. 2010).
  The reference for the rate ratio is the estimated rate at the 5th percentile of the pollutant concentration. Estimates are presented
for the 5th percentile through the 95th percentile of pollutant concentrations due to instability in the dose-response estimates at the
distribution tails.

Figure 6-17    Loess dose-response estimates and twice-standard error estimates
                 from generalized additive models for associations between 3-day
                 avg ozone concentrations and ED visits for pediatric asthma.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
               Strickland et al. (2010) examined the association between O3 exposure and pediatric
               asthma ED visits (ages 5-17 years) in Atlanta between 1993 and 2004 using the same air
               quality data as Darrow et al. (20lib) and Tolbert et al. (2007). However, unlike Darrow
               et al. (20lib) and Tolbert et al. (2007). which used single centrally located monitors or an
               average of monitors, respectively, Strickland et al. (2010) used population-weighting to
               combine daily pollutant concentrations across monitors. In this study, the authors
               developed a statistical model using hospital-specific time-series data that is essentially
               equivalent to a time-stratified case-crossover analysis (i.e., using interaction terms
               between year, month, and day-of-week to mimic the approach of selecting referent days
               within the same month and year as the case day). The authors observed a 6.4% (95% CI:
               3.2, 9.6%) increase in ED visits for a 30 ppb increase in 8-h max O3 concentrations at lag
               0-2 in an all-year analysis. In seasonal analyses, stronger associations were observed
               during the warm season (i.e., May-October) (8.4% [95% CI: 4.4, 12.7%]; lag 0-2) than
               the cold season (4.5% [95% CI: -0.82, 10.0%]; lag 0-2). Strickland et al. (2011)
               confirmed these findings  in an additional analysis using the same dataset, and found that
               the metric used to assign  exposure (i.e., centrally located monitor, unweighted average
Draft - Do Not Cite or Quote
                                                   6-128
September 2011

-------
 1                   across monitors, and population-weighted average across monitors) did not influence
 2                   pediatric asthma ED visit risk estimates for spatially homogeneous pollutants such as O3.

 3                   In copollutant analyses, Strickland et al. (2010) found that O3 effect estimates were not
 4                   substantially changed when controlling for other pollutants (CO, NO2, PM2.5 elemental
 5                   carbon, PM2 5 sulfate) (results not presented quantitatively). The authors also examined
 6                   the C-R relationship between O3 exposure and pediatric asthma ED visits and found that
 7                   both quintile and loess dose-response analyses (Figure 6-17) suggest that there  are
 8                   elevated associations with O3 at concentrations as low as 30 ppb. These dose-response
 9                   analyses do not provide evidence of a threshold level.

10                   In a single-city study conducted on the West coast, Mar and Koenig (2009) examined the
11                   association between O3 exposure and asthma ED visits (ICD-9 codes_ 493-493.9) for
12                   children (< 18) and adults (> 18) in Seattle, WA from 1998 to 2002. Of the total number
13                   of visits over the study duration, 64% of visits in the age group < 18 comprised boys, and
14                   70% of visits  in the > 18 age group comprised females. Mar and Koenig (2009)
15                   conducted a time-series analysis using both 1-h max and max 8-h avg O3 concentrations.
16                   Although a similar pattern of associations was observed using both metrics, only those
17                   results using the max 8-h avg O3 metric are discussed here since they are more  applicable
18                   to the current O3 NAAQS. Mar and Koenig (2009) presented results for single  day lags of
19                   0 to 5 days, but found consistent positive associations across individual lag days which
20                   supports the findings from the studies discussed above that examined multi-day
21                   exposures. For children, consistent positive associations were observed across all lags,
22                   ranging from  a 19.1-36.8% increase in asthma ED visits for a 30 ppb increase in 8-h max
23                   O3 concentrations with the strongest associations observed at lag 0 (33.1% [95% CI: 3.0,
24                   68.5]) and lag 3 (36.8% [95% CI: 6.1, 77.2]) (Figure x). O3 was also found to be
25                   positively associated with asthma ED visits for adults at all lags, ranging from  9.3-26.0%,
26                   except at lag 0 (Figure 6-18). The slightly different lag times for children and adults
27                   suggest that children may be more immediately responsive to O3 exposures than adults
28                   (Mar and Koenig. 2009).
      Draft - Do Not Cite or Quote                       6-129                                 September 2011

-------
                        LagO
                        Lagl
                        Lag 2
                        Lag3
                        Lag 4
                        LagS
                                                                     Black = Children
                                                                     Red = Adults
                             0.80       1.00
                                                1.20       1.40
                                                 Relative Risk
                                                                    1.60
                                                                              1.80
      Adapted from Mar and Koenig (2009).

     Figure 6-18   Relative risk of asthma ED visits children and adults for a 30 ppb
                     increase in max 8-h avg O3 concentrations in Seattle, WA, 1998-
                     2002.
1
2
3
4
5
6
7
    Respiratory Infection
Although an increasing number of studies have examined the association between O3
exposure and cause-specific respiratory ED visits this trend has not included an extensive
examination of the association between O3 exposure and respiratory infection ED visits.
Stieb et al. (2009) also examined the association between short-term O3 exposure and
respiratory infection ED visits in 7 Canadian cities. In an all-year analysis, there was no
evidence of an association between O3 exposure  and respiratory infection ED visits at all
lags examined (i.e., 0,  1, and 2). Across cities, respiratory infections comprised the single
largest diagnostic category, approximately 32%, of all the ED visits examined, which
also included myocardial infarction, heart failure, dysrhythmia, asthma, and COPD.
     Draft - Do Not Cite or Quote
                              6-130
September 2011

-------
                     6.2.7.4    Outpatient and Physician Visit Studies

 1                   Several studies have examined the association between ambient O3 concentrations and
 2                   physician or outpatient (non-hospital, non-ED) visits for acute conditions in various
 3                   geographic locations. Burra et al. (2009) examined asthma physician visits among
 4                   patients aged 1-17 and 18-64 years in Toronto, Canada from 1992 to 2001. The authors
 5                   found little or no evidence of an association between asthma physician visits and O3;
 6                   however, seasonal analyses were not conducted. It should be noted that in this study,
 7                   most of the relative risks for O3 were less than one and statistically significant, perhaps
 8                   indicating an inverse correlation with another pollutant or an artifact of the strong
 9                   seasonality of asthma visits. Villeneuve et al. (2006b) also focused on physician visits to
10                   examine the effect of short-term O3 exposure on allergic rhinitis among individuals aged
11                   65 or older in Toronto from 1995 to 2000. The authors did not observe any evidence of
12                   an association between allergic rhinitis physician visits and ambient O3 concentrations in
13                   single-day lag models in an all-year analysis (results not presented quantitatively).

14                   In a study conducted in Atlanta, Sinclair et al. (2010) examined the association of acute
15                   asthma and respiratory infection (e.g., upper respiratory infections and lower respiratory
16                   infections) outpatient visits from a managed care organization with ambient O3
17                   concentrations as well as multiple PM size fractions and species from August 1998
18                   through December 2002. The authors separated the analysis into two time periods (the
19                   first 25 months of the study period and the second 28 months of the study period), in
20                   order to compare the air pollutant concentrations and relationships between air pollutants
21                   and acute respiratory visits for the 25-month time-period examined in  Sinclair et al.
22                   (2004) to an additional 28-month time-period of available data from the Atlanta Aerosol
23                   Research Inhalation Epidemiology Study (ARIES). The authors found little evidence of
24                   an association between O3 and asthma visits, for both children and adults, or respiratory
25                   infection visits in all-year analyses and seasonal analyses. For example, a slightly
26                   elevated relative risk (RR) for childhood asthma visits  was observed during the 25-month
27                   period in the cold season (RR: 1.12 [95% CI: 0.86, 1.41]; lag 0-2 for a 30 ppb increase in
28                   8-h max O3), but not in the warm season (RR: 0.97 [95% CI: 0.86, 1.10];  lag 0-2). During
29                   the 28-month period at lag 0-2, a slightly larger positive effect was observed during the
30                   warm season (RR: 1.06 [95% CI: 0.97, 1.17]), compared to the cold season (RR: 1.03
31                   [95% CI: 0.87, 1.21]). Overall, these results contradict those from Strickland et al. (2010)
32                   discussed above. Although the mean number of asthma visits and O3 concentrations in
33                   Sinclair et al. (2010) and Strickland et al. (2010) are similar the difference in results
34                   between the two studies could be attributed to the severity of O3-induced  asthma
35                   exacerbations (i.e., more severe symptoms requiring a  visit to a hospital) and behavior,
36                   such as delaying a visit to the doctor for less severe symptoms.
      Draft - Do Not Cite or Quote                       6-131                                 September 2011

-------
                     6.2.7.5    Summary

 1                   The results of the recent studies evaluated largely support the conclusion of the 2006 O3
 2                   AQCD. While fewer studies were published overall since the previous review, several
 3                   multicity studies (e.g., Cakmak et al.. 2006b: Dales et al.. 2006) and a multi-continent
 4                   study (Katsouvanni et al., 2009) provide supporting evidence for an association between
 5                   short-term O3 exposure and an increase in respiratory-related hospital admissions and ED
 6                   visits. Collectively, in the studies evaluated, both single-city and multicity, there is
 7                   continued evidence for increases in both hospital admissions and ED visits when
 8                   examining all respiratory outcomes combined. Additionally, new studies support an
 9                   association between short-term O3 exposure and asthma (Strickland et al.. 2010; Stieb et
10                   al.. 2009) and COPD (Stieb et al.. 2009; Medina-Ramon et al.. 2006) hospital admissions
11                   and ED visits, with more limited evidence for pneumonia hospital admissions and ED
12                   visits (Medina-Ramon et al., 2006; Zanobetti and Schwartz. 2006). In seasonal analyses,
13                   stronger associations were observed in the warm season or summer months compared to
14                   the cold season, particularly for asthma (Strickland et al.. 2010; Ito et al.. 2007b) and
15                   COPD (Medina-Ramon et al.. 2006)  (Figure 6-19; Table 6-27), which is consistent with
16                   the conclusions of the 2006 O3 AQCD. There is also continued evidence that children are
17                   particularly susceptible to O3-induced respiratory effects (Silverman and Ito. 2010;
18                   Strickland et al.. 2010; Mar and Koenig. 2009; Villeneuve et al.. 2007; Dales et al..
19                   2006).  Although the collective evidence indicates a consistent positive association
20                   between O3 exposure and respiratory-related hospital admissions and ED visits, the
21                   magnitude of these associations may be underestimated due to behavioral modification in
22                   response to forecasted air quality (Neidell and Kinney. 2010; Neidell. 2009)
23                   (Section 4.6.4).

24                   Additional studies that focused on respiratory-related outpatient or physician visits found
25                   no evidence of an association with short-term O3 exposure, but this could be attributed to
26                   the severity of O3-induced respiratory effects requiring more immediate treatment or
27                   behavioral factors that result in delayed visits to a physician.

28                   The studies that examined the potential confounding effects of copollutants found that O3
29                   effect estimates remained relatively robust upon the inclusion of PM and gaseous
30                   pollutants in two-pollutant models (Figure 6-19; Table 6-27), including (Strickland et al..
31                   2010; Tolbert et al.. 2007; Medina-Ramon et al.. 2006). which did not present results
32                   quantitatively. These findings are consistent with the studies evaluated in the 2006 O3
33                   AQCD (U.S. EPA. 2006b) (Figure 7-12, p.  7-80) which found O3 respiratory hospital
34                   admissions risk estimates remained robust to the inclusion of PM in copollutant models.
      Draft - Do Not Cite or Quote                       6-132                                 September 2011

-------
 Study
                               Visit Type
                                       Age
Wong etal. (2009; 196722]
Cakmak etal. (2006; 93272)
Dales etal. (2006; 90744]
Orazzoetal.(2009 202800)a
Katsouyanni et al. 2009; 199899
Katsouyanni et al. 2009; 199899
Katsouyanni et al. 2009; 199899
Katsouyanni et al. 2009; 199899
Darrow etal. (2009; 202800]
Tol be rt etal. (2007,090316)
Biggeri etal. (2005 87395]c
Katsouyanni et al. 2009; 199899
Katsouyanni et al. 2009; 199899
Katsouyanni et al. 2009; 199899
Katsouyanni et al. 2009; 199899
Stiebetal. (2009: 195858]
Villeneuve et al. (2007; 195859]
Strickland etal. (2010; 624878]
Silverman and Ito 2010; 386252]c
Ito etal. (2007; 156594]
Villeneuve et al. (2007; 195859)
Hong Kong
10 Canadian cities
11 Canadian cities
6 Italian cities
APHENA-Europe
APHENA-U.S.
APHENA-Canada
b APHENA-Canada
Atlanta
Atlanta
8 Italian cities
APHENA-Europe
APHENA-U.S.
APHENA-Canada
b APHENA-Canada
7 Canadian Cities
Alberta, CAN
Atlanta
New York
New York
Alberta. CAN
Mar and Koenig [2009; 594410] Seattle,' WA
Strickland et al. (2010; 624878] Atlanta
Silverman and Ito (2010; 386252)d New York
Mar and Koenig (2009; 594410]
Ito etal. (2007; 156594]
Villeneuve et al. (2007; 195859)
Strickland etal. (2010; 624878]
Stiebetal. (2009; 195858]
Medina-Ramon eta. (2006; 8772
Yang etal. (2006; 90184)
Stiebetal. (2009; 195858]e
Medina-Ramon et a. (2006; 8772
Medina-Ramon eta. (2006; 8772
Seattle, WA
New York
Alberta, CAN
Atlanta
7 Canadian Cities
] 36 U.S. cities
Vancouver
7 Canadian Cities
36 U.S. cities
36 U.S. cities
Zanobetti and Schwartz (2006; 90195] Boston
Medina-Ramon eta . 2006; 8772
Medina-Ramon eta. 2006; 8772
Medina-Ramon eta . 2006; 8772
36 U.S. cities
36 U.S. cities
36 U.S. cities
HA
HA
HA
ED
HA
HA
HA
HA
ED
ED
HA
HA
HA
HA
HA
ED
ED
ED
HA
ED
ED
ED
ED
HA
ED
ED
ED
ED
ED
HA
HA
ED
HA
HA
HA
HA
HA
HA
All
All
0-27 days
0-2
65+
65+
65+
65+
All
All
All
65+
65+
65+
65+
All
>2
Children
All
All
>2
18+
Children
6-18
<18
All
>2
Children
All
65+
65+
All
65+
65+
65+
65+
65+
65+
0-1
1.2
2
0-6
0-1
0-1
DLjO-2]
DLJO-2]
1
0-2
0-3
0-1
0-1
DLjO-2]
DLJO-2]
2
0-2
0-2
0-1
0-1
0-2
2
0-2
0-1
0
0-1
0-2
0-2
2
0-1
0-3
NR
0-1
0-1
0-1
0-1
0-1
0-1
                                                     Respiratory
                                                     Asthma
                                                     25  -20  -15   -10   -5   0   5   10   15  20   25   30  35   40
  3 Wheeze used as indicator of lower respiratory disease.
  b APHENA-Canada results standardized to approximate IQR of 5.1 ppb for 1 h max O3 concentrations.
  0 Study included 8 cities; but of those 8, only 4 had O3 data.
  d non-ICU effect estimates.
  e The study did not specify the lag day of the summer season estimate.
  Effect estimates are for a 20 ppb increase in 24 hours; 30 ppb increase in 8-h max; and 40 increase in 1-h max ozone
concentrations. HA=hospital admission; ED=emergency department. Black=AII-year analysis; Red=Summer only analysis;
Blue=Winter only analysis.


Figure 6-19    Percent increase in respiratory-related hospital admission and  ED
                  visits in studies that presented all-year and/or seasonal results.
Draft - Do Not Cite or Quote
6-133
September 2011

-------
Table 6-27 Corresponding Effect Estimates for Figure 6-19
Study
ED Visit or
Hospital
Admission
Location
Age
Lag
Avg Time
% Increase
(95% Cl)
Respiratory
All-year
Wona et al. (2009)
Cakmaketal. (2006b)
Dales et al. (2006)
Orazzoetal. (2011 b)a
Katsouyanni et al. (2009)
Hospital Admission
Hospital Admission
Hospital Admission
ED Visit
Hospital Admission
Hong Kong
10 Canadian cities
11 Canadian cities
6 Italian cities
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
All
All
0-27 days
0-2
65+
65+
65+
65+
0-1
1.2
2
0-6
0-1
0-1
DL(0-2)
DL(0-2)b
8-h max
24-h avg
24-h avg
8-h max
1-hmax
1-hmax
1-hmax
1-hmax
3.58(1.90,5.29)
4.38(2.19,6.46)
5.41 (2.88, 7.96)
-3.34 (-11. 2, 5.28)
1.58 (-1.71, 4.15)
2.38 (0.00, 4.89)
20.4 (4.07, 40.2)
2.4(0.51,4.40)
Warm
Darrowetal. (2011 b)
Tolbert et al. (2007)
Bidden et al. (2005)°
Katsouyanni et al. (2009)
ED Visit
ED Visit
Hospital Admission
Hospital Admission
Atlanta
Atlanta
8 Italian cities
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
All
All
All
65+
65+
65+
65+
1
0-2
0-3
0-1
0-1
DL(0-2)
DL(0-2)b
8-h max
8-h max
8-h max
1-h max
1-h max
1-h max
1-h max
2.08(1.25,2.91)
3.90 (2.70, 5.20)
0.06 (-5.24, 5.66)
0.24 (-3.32, 3.91)
2.14 (-0.63, 4.97)
37.1(11.5,67.5)
4.1(1.40,6.80)
Asthma
All-year
Stiebetal. (2009)
Villeneuve et al. (2007)
Strickland etal. (2010)
ED Visit
ED Visit
ED Visit
7 Canadian cities
Alberta, CAN
Atlanta
All
>2
Children
2
0-2
0-2
24-h avg
8-h max
8-h max
3.48 (0.33, 6.76)
11.9(6.8,17.2)
6.38(3.19,9.57)
Warm
Silverman and Ito (2010)"
Itoetal. (2007b)
Villeneuve et al. (2007)
Mar and Koenid (2009)
Strickland etal. (2010)
Silverman and Ito (2010)"
Mar and Koenid (2009)
Hospital Admission
ED Visit
ED Visit
ED Visit
ED Visit
Hospital Admission
ED Visit
New York
New York
Alberta, CAN
Seattle, WA
Atlanta
New York
Seattle, WA
All
All
>2
18+
Children
6-18
<18
0-1
0-1
0-2
2
0-2
0-1
0
8-h max
8-h max
8-h max
8-h max
8-h max
8-h max
8-h max
12.5(8.27,16.7)
16.9(10.9,23.4)
19.0(11.9,28.1)
19.1(3.00,40.5)
8.43(4.42,12.7)
28.2(15.3,41.5)
33.1(3.00,68.5)
Cold
Itoetal. (2007b)
Villeneuve et al. (2007)
Strickland etal. (2010)
ED Visit
ED Visit
ED Visit
New York
Alberta, CAN
Atlanta
All
>2
Children
0-1
0-2
0-2
8-h max
8-h max
8-h max
-23.4 (-27.3, -19.3)
8.50(0.00,17.2)
4.52 (-0.82, 10.1)
COPD
All-year
Stiebetal. (2009)
Medina-Ramon et al. (2006)
Yand et al. (2005b)
ED Visit
Hospital Admission
Hospital Admission
7 Canadian cities
36 U.S. cities
Vancouver
All
65+
65+
2
0-1
0-3
24-h avg
8-h max
24-h avg
4.03 (-0.54, 8.62)
0.24 (-0.78, 1.21)
8.80 (-12.5, 32.6)
Warm
Stiebetal. (2009)e
Medina-Ramon et al. (2006)
ED Visit
Hospital Admission
7 Canadian cities
36 U.S. cities
All
65+
NR
0-1
24-h avg
8-h max
6.76(0.11,13.9)
1.63(0.48,2.85)
Cold
Medina-Ramon et al. (2006)
Hospital Admission
36 U.S. cities
65+
0-1
8-h max
-1.85 (-3.60, -0.06)
Draft - Do Not Cite or Quote
6-134
September 2011

-------
Study
ED Visit or
Hospital
Admission
Location
Age Lag Avg Time
% Increase
(95% Cl)
Pneumonia
All-year
Zanobetti and Schwartz (2006)
Medina-Ramon et al. (2006)
Hospital Admission
Hospital Admission
Boston
36 U.S. cities
65+ 0-1 24-h avg
65+ 0-1 8-h max
-5.96 (-11.1, -1.36)
1.81 (-0.72, 4.52)
Warm
Medina-Ramon et al. (2006)
Hospital Admission
36 U.S. cities
65+ 0-1 8-h max
2.49(1.57,3.47)
Cold
Medina-Ramon et al. (2006)
Hospital Admission
36 U.S. cities
65+ 0-1 8-h max
-4. 88 (-6. 59, -3.14)
  "Wheeze used as indicator of lower respiratory disease.

  bAPHENA-Canada results standardized to approximate IQRof 5.1 ppbfor 1-h max 03 concentrations.

  °Study included 8 cities, but of those 8 only 4 had 03 data.

  dNon-ICU effect estimates.

  The study did not specify the lag day of the summer season estimate.
Study Location Age Lag Copollutant
Respiratory
Katsouyanni et al. (2009; 199899)a APHENA-U.S. 65+ 1
PM10
APHENA-Europe
PM10
c
pwiin
c PM10 —
COPD
fengetal. (2006;90184)a .an^uv.i C. I 0 „ <
Al 1-Year
«-
-•—


'
J°' *

Asthma
Itoetal. (2007;156594)b New York All 0-1
CO
NO2
SO2
PM2.5
Summer




-10 -50 5 10 15 20 25 30
% Increase
  Effect estimates are for a 20 ppb increase in 24 hours; 30 ppb increase in 8-h max; and 40 ppb increase in 1-h max ozone
concentrations. An "a" represents studies that examined hospital admissions, "b" represents a study that examined ED visits, and "c"
represents risk estimates from APHENA-Canada standardized to an approximate IQR of 5.1 ppb for a 1-h max increase in ozone
concentrations. Black = results from single-pollutant models; Red = results from copollutant models with PM10or PM2.s; Yellow =
results from copollutant models with CO; Blue = results from copollutant models with NO2;  Green = results from copollutant models
with SO2.


Figure 6-20     Percent increase in respiratory-related hospital admissions and  ED
                   visits for studies that presented single and  copollutant model

                   results.
Draft - Do Not Cite or Quote
6-135
September 2011

-------
Table 6-28 Corresponding effect
Study3 Location Visit Type Age
estimates for Figure 6-20
Lag Copollutant

% Increase (95% Cl)
All-year
Respiratory
Katsouyanni et al. (2009) APHENA-U.S.
APHENA-Europe
APHENA-Canada
HA 65+ 1
PM,0

PM,0


PM,0
PM,0
2.62 (0.63, 4.64)
2.14 (-0.08, 4.40)
2.94(1.02,4.89)
2.38 (0.08, 4.64)
5.54 (-0.94, 12.4)
0.69 (-0.12, 1.50)b
5. 13 (-6.62, 18.6)
0.64 (-0.87, 2.20)b
COPD
Yang et al. (2005b) Vancouver
HA 65+ 0-3
CO
N02
S02
PM,0
8.80 (-12.5, 32.6)
22.8 (-2.14, 50.7)
11.1 (-10.4, 37.6)
13.4 (-8.40, 40.2)
11.1 (-8.40, 37.6)
Summer
Asthma
Itoetal. (2007b) New York
ED All 0-1
CO
N02
S02
PM2,
16.9(10.9,23.4)
18.1(12.1,24.5)
10.2(4.29,16.4)
13.1(7.16,19.5)
12.7(6.37,19.3)
        "Averaging times: Katsouyanni et al. (2009) = 1-h max; Yang et al. (2005b) = 24-h avg; and Ito et al. (2007b) = 8-h max.
        bRisk estimates standardized to an approximate IQRof 5.1 ppbfora 1-h max increase in 03 concentrations.

 1                   Additionally, a preliminary examination of the C-R relationship found no evidence of a
 2                   threshold between short-term O3 exposure and pediatric asthma ED visits (Silverman and
 3                   Ito. 2010; Strickland et al.. 2010). Overall, the new body of evidence supports an
 4                   association between short-term O3 exposure and respiratory-related hospital admissions
 5                   and ED visits, with additional  evidence  for stronger associations during the warm season
 6                   for specific respiratory outcomes such as asthma and COPD.
             6.2.8   Respiratory Mortality

 7                   The epidemiologic, controlled human exposure, and toxicological studies discussed
 8                   within this section (Section 6.2) provides evidence for multiple respiratory effects in
 9                   response to short-term O3 exposure. Additionally, the evidence from experimental studies
10                   indicates multiple potential pathways of O3-induced respiratory effects, which support the
11                   continuum of respiratory effects that could potentially result in respiratory-related
12                   mortality. The 2006 O3 AQCD found inconsistent evidence for an association between
13                   short-term O3 exposure and respiratory mortality (U.S. EPA. 2006b). Although some
14                   studies reported a strong positive association between O3 exposure and respiratory
      Draft - Do Not Cite or Quote                       6-136                                 September 2011

-------
 1                   mortality, additional studies reported a small association or no association. Recent
 2                   multicity studies found consistent positive associations between short-term O3 exposure
 3                   and respiratory mortality, specifically during the summer months.

 4                   The APHENA study, described earlier in Section 6.2.7.2, (Katsouyanni et al.. 2009) also
 5                   examined associations between short-term O3 exposure and mortality and found
 6                   consistent positive associations for respiratory mortality in all-year analyses with stronger
 7                   associations in analyses restricted to the summer season. Additional multicity studies
 8                   from the U.S. (Zanobetti and Schwartz. 2008b). Europe (Samoli et al.. 2009). Italy
 9                   (Stafoggia et al.. 2010). and Asia (Wong et al.. 2010) that conducted summer season
10                   and/or all-year analyses provide additional support for an association between short-term
11                   O3 exposure and respiratory mortality (Figure 6-37).

12                   Of the studies evaluated, only the APHENA study (Katsouyanni et al.. 2009) and the
13                   Italian multicity study (Stafoggia et al.. 2010) conducted an analysis of the potential for
14                   copollutant confounding of the O3-respiratory mortality relationship. In the APHENA
15                   study, in the European dataset, when focusing on the natural spline model with 8 df/year
16                   (as discussed in Section 6.2.7.2) and lag 1 results (as discussed in Section 6.6.2.1),
17                   respiratory mortality risk estimates were robust to the inclusion of PM10 in copollutant
18                   models in all-year analyses with O3 respiratory mortality risk estimates increasing in the
19                   Canadian and U.S. datasets. In summer season analyses, respiratory O3 mortality risk
20                   estimates were robust in the U.S. dataset and attenuated in the European dataset.
21                   Similarly, in the Italian multicity study (Stafoggia et al.. 2010). which was limited to the
22                   summer season, respiratory mortality risk estimates were attenuated in copollutant
23                   models with PMi0. Based on the APHENA and Italian multicity results, O3 respiratory
24                   mortality risk estimates appear to be moderately to  substantially sensitive (e.g., increased
25                   or attenuated) to inclusion of PMi0. However, in the APHENA study, the mostly every-
26                   6th-day sampling schedule for PM10 in the Canadian and U.S. datasets greatly reduced
27                   their sample size and limits the interpretation of these results.
             6.2.9   Summary and Causal Determination

28                   The 2006 O3 AQCD concluded that there was clear, consistent evidence of a causal
29                   relationship between short-term O3 exposure and respiratory effects (U.S. EPA. 2006b).
30                   This conclusion was substantiated by evidence from controlled human exposure and
31                   toxicological studies indicating a range of respiratory effects in response to short-term O3
32                   exposure, including pulmonary function decrements, respiratory symptoms, lung
33                   inflammation, increased lung permeability, and airway hyperresponsiveness.
34                   Toxicological studies provided additional evidence for O3-induced impairment of host
      Draft - Do Not Cite or Quote                       6-137                                 September 2011

-------
 1                   defenses. Combined, these findings from experimental studies provided support for
 2                   epidemiologic evidence, in which short-term O3 exposure was consistently associated
 3                   with decreases in lung function in populations with increased outdoor exposures, children
 4                   with asthma, and healthy children; increases in respiratory symptoms and asthma
 5                   medication use in children with asthma; and increases in respiratory-related hospital
 6                   admissions and asthma-related ED visits. Short-term O3 exposure also was consistently
 7                   associated with all-cause and cardiopulmonary mortality; however, the contribution of
 8                   respiratory causes to these findings was uncertain.

 9                   Building on the large body of evidence presented in the 2006 O3 AQCD, recent studies
10                   support associations between short-term O3 exposure and respiratory effects.  Controlled
11                   human exposure studies continue to provide the strongest evidence for lung function
12                   decrements in young healthy adults over a range of O3 concentrations. Studies previously
13                   reported mean O3-induced FEVi decrements of 6-8% at 80 ppb O3 (Adams. 2006a.
14                   2003a; McDonnell et al.. 1991; Horstman et al.. 1990). and new evidence additionally
15                   indicates mean FEVi decrements of 6% at 70 ppb O3 (Schelegle et al. 2009)  and 2-3% at
16                   60 ppb O3 (Kim etal.. 2011; Brown et al.. 2008; Adams. 2006a) (Section 6.2.1.1). In
17                   healthy young adults, O3-induced decrements in FEVi do not appear to depend on gender
18                   (Hazucha et al.. 2003). body surface area or height (McDonnell et al.. 1997).  lung size or
19                   baseline FVC (Messineo and Adams.  1990). There is limited evidence that blacks may
20                   experience greater O3-induced decrements in FEVi than do age-matched whites (Que et
21                   al.; Seal etal.. 1993). Healthy children experience similar spirometric responses but
22                   lesser symptoms from O3 exposure relative to young adults (McDonnell et al.. 1985b).
23                   On average, spirometric and symptom responses to O3 exposure appear to decline with
24                   increasing age beyond about 18 years of age (McDonnell et al.. 1999; Seal et al.. 1996).
25                   There is also a tendency for slightly increased spirometric responses in mild asthmatics
26                   and allergic rhinitics relative to healthy young adults (Torres etal.. 1996). Spirometric
27                   responses in asthmatics appear to be affected by baseline lung function, i.e., responses
28                   increase with disease severity (Horstman et al.. 1995).

29                   Available information from controlled human exposure studies on recovery from O3
30                   exposure indicates that an initial phase of recovery in healthy individuals proceeds
31                   relatively rapidly, with acute spirometric and symptom responses resolving within about
32                   2 to 4 h (Folinsbee and Hazucha. 1989). Small residual lung function effects are almost
33                   completely resolved within 24 h. Effects of O3 on the small airways persisting a day
34                   following exposure, assessed by persistent decrement in FEF25_75 and altered ventilation
35                   distribution, may be due in part to inflammation (Frank etal.. 2001; Foster etal.. 1997).
36                   In more responsive  individuals, this recovery in lung function takes  longer (as much as
37                   48 hours) to return to baseline. Some cellular responses may not return to baseline levels
38                   in humans for more than 10-20 days following O3 exposure (Devlin et al.. 1997). Airway
      Draft - Do Not Cite or Quote                      6-138                               September 2011

-------
 1                   hyperresponsiveness and increased epithelial permeability are also observed as late as 24
 2                   h postexposure (Que etal.).

 3                   With repeated O3 exposures over several days, spirometric and symptom responses
 4                   become attenuated in both healthy individuals and asthmatics, but this tolerance is lost
 5                   after about a week without exposure (Gong et al.. 1997a: Folinsbee et al.. 1994; Kulle et
 6                   al., 1982). Airway responsiveness also appears to be somewhat attenuated with repeated
 7                   O3 exposures in healthy individuals, but becomes increased in individuals with
 8                   preexisting allergic airway disease (Gong et al.. 1997a; Folinsbee et al.. 1994). Some
 9                   indicators of pulmonary inflammation are attenuated with repeated O3 exposures.
10                   However, other markers such as epithelial integrity and damage do not show attenuation,
11                   suggesting continued tissue damage during repeated O3 exposure (Devlin et al.. 1997).

12                   Collectively, epidemiologic evidence  supports observations from controlled human
13                   exposure studies of O3-induced decrements in lung function (Section 6.2.1.2). A notable
14                   difference among newer studies was the relatively limited investigation of the effect of
15                   ambient O3 exposure on lung function in populations engaged in outdoor recreation,
16                   exercise, or work, which contributed to the  weight of evidence in previous AQCDs. As in
17                   previous AQCDs, recent epidemiologic investigation focused on and most consistently
18                   demonstrated associations between increases in ambient O3 exposure and decreases in
19                   lung function in children with asthma. Across the diverse populations examined in
20                   epidemiologic studies, ambient O3 exposure was associated with 1-8%  decreases in mean
21                   lung function per standardized increment in O3 concentration1. Larger decreases (3-8%)
22                   were observed in children with asthma with increased outdoor exposures, CS use, or
23                   concurrent URI and older adults with  airway hyperresponsiveness, elevated BMI, or
24                   GSTP1 Val/Val genotype, indicating the existence of groups within the population with
25                   potentially increased sensitivity to O3  exposure. Further, several epidemiologic studies
26                   found that O3-associated decreases in  lung function were associated with concomitant
27                   increases in respiratory symptoms. Biological plausibility for O3-associated decrements
28                   in lung function in controlled human exposure, epidemiologic, and animal studies is
29                   provided by the well-documented effects of O3 activating bronchial C-fibers (Section
30                   5.3.2).

31                   Across disciplines, studies have examined factors that may potentially increase an
32                   individual's susceptibility to O3-induced decrements in lung function. In the controlled
33                   human exposure studies, there is a large degree of intersubject variability in lung function
34                   decrements, symptomatic responses, pulmonary inflammation, airway
35                   hyperresponsiveness, and altered epithelial  permeability in healthy adults exposed to  O3
        1 Effect estimates were standardized to a 40-ppb increase for 1-h max O3, a 30-ppb increase for 8-h max O3, and a 20-ppb
      increase for 24-h avg O3.
      Draft - Do Not Cite or Quote                       6-139                                September 2011

-------
 1                   (Que et al.; Holz et al.. 2005; McDonnell. 1996). The magnitude of pulmonary
 2                   inflammation, airway hyperresponsiveness, and increases in epithelial permeability do
 3                   not appear to be correlated, nor are these responses to O3 correlated with changes in lung
 4                   function, suggesting that different mechanisms may be responsible for these processes
 5                   (Que et al.; Balmes et al., 1997; Balmes et al., 1996; Aris et al., 1995). However, these
 6                   responses tend to be reproducible within a given individual over a period of several
 7                   months indicating differences in the intrinsic responsiveness of individuals (Holz et al..
 8                   2005: Hazuchaetal.. 2003: Holzetal.. 1999: McDonnell et al.. 1985a). Numerous
 9                   reasons for differences in the susceptibility of individuals to O3 exposure have been
10                   reported in the literature. Dosimetric and mechanistic considerations are discussed in
11                   Section 5.4. Evidence in all three  disciplines suggests a role for antioxidant defenses in
12                   modulating respiratory responses  to O3. The biological plausibility of these  findings is
13                   provided by the well-characterized evidence for O3 exposure leading to the formation of
14                   secondary oxidation products, which subsequently activate neural reflexes that mediate
15                   lung function decrements (Section 5.3.2). Secondary oxidation products also initiate
16                   pulmonary inflammation (Sections 5.3.3). Epidemiologic studies additionally have found
17                   that atopy (Khatri et al., 2009). concurrent respiratory infection (Lewis et al.. 2005).
18                   AHR, and elevated BMI (Alexeeff et al.. 2007) may modify respiratory responses to O3
19                   exposure (Section 6.2.1.2). Retrospective analyses of controlled human exposure studies
20                   of data pooled across 15 controlled human exposure studies also show larger O3-induced
21                   FEVi  decrements in adults with higher BMI (McDonnell et al.. 2010:  Bennett et al..
22                   2007).

23                   Additional respiratory effects induced by short-term O3 exposures in controlled human
24                   exposure studies of healthy, young adults include increases in respiratory symptoms with
25                   O3 concentrations <80 ppb (Schelegle et al.. 2009: Adams.  2006a) (Section 6.2.1.1).
26                   Similarly, epidemiologic studies collectively demonstrate that increases in short-term
27                   ambient O3 exposure are associated with increases in respiratory symptoms and  asthma
28                   medication use among subjects with  asthma (Section 6.2.4.1).  Among recent
29                   epidemiologic studies, the strongest evidence of O3-associated respiratory symptoms was
30                   found in populations with multiple potential susceptibility factors, specifically,
31                   individuals with asthma and atopy (Khatri et al.. 2009: Escamilla-Nunez et al.. 2008: Feo
32                   Brito et al.. 2007) and children with asthma with diminished antioxidant enzyme activity
33                   (Romieu et al.. 2006).

34                   Recent controlled human exposure studies (Section 6.2.3.1) and toxicological studies
35                   (Section 6.2.3.3) also continue to  demonstrate lung injury and  inflammatory responses
36                   upon O3 exposure. Evidence from more than a hundred toxicological studies clearly
37                   indicates that O3 induces damage  and inflammation in the lung, and studies continue to
      Draft - Do Not Cite or Quote                       6-140                                 September 2011

-------
 1                   elucidate the mechanistic pathways involved (Section 5.3). Though inflammation may
 2                   resolve, continued inflammation may alter theJanetrita2010

 3                    structure and function of pulmonary tissues. New controlled human studies support
 4                   previous findings for pulmonary inflammation at 60 ppb O3, the lowest concentration
 5                   evaluated . Building on the extensive experimental evidence, epidemiologic studies
 6                   provide new evidence for ambient O3-associated increases in pulmonary inflammation in
 7                   individuals with asthma. These associations were observed primarily for 8-h max or 8-h
 8                   average O3 exposures but for both same-day and multiday average exposures. Multiple
 9                   studies examined and found increases in eNO (Berhane et al.. 2011; Khatri et al.. 2009;
10                   Barraza-Villarreal et al., 2008). The clinical significance of these findings was supported
11                   by observations of concomitant O3-associated increases in respiratory symptoms (Khatri
12                   et al., 2009; Barraza-Villarreal et al.. 2008). A smaller number of studies examined and
13                   found associations with cytokines such as IL-6 or IL-8 in nasal  lavage samples (Barraza-
14                   Villarreal et al.. 2008; Sienra-Monge et al., 2004) inflammatory cells in blood (e.g.,
15                   eosinophils) (Khatri et al.. 2009). decreased levels of antioxidants (Sienra-Monge et al..
16                   2004). and increased levels of indicators of oxidative stress (Romieu et al.. 2008)
17                   (Section 6.2.3.2).

18                   Modification of innate and adaptive immunity is emerging as a  mechanistic pathway
19                   underlying the effects of ozone on asthma and allergic airways  disease (Section  5.3.6).
20                   While the majority of evidence comes from animal studies, results from controlled
21                   human exposure studies suggest that these pathways may be relevant to humans and may
22                   lead to the induction and exacerbation of asthma (Alexis et al.. 2010; Hernandez et al..
23                   2010; Alexis et al.. 2009; Bosson et al.. 2003). Further, differences between asthmatics
24                   and healthy controls in ozone-mediated innate and adaptive immune responses have been
25                   noted (Section 5.4.2.2).

26                   The subclinical and overt respiratory effects observed across disciplines collectively
27                   provide support for epidemiologic studies that demonstrate consistently positive
28                   associations of short-term O3 exposure with respiratory-related  hospital admissions and
29                   ED visits (Section 6.2.7). Consistent with evidence presented in the 2006 O3 AQCD, new
30                   multicity studies and a multicontinent study (i.e., APHENA) (Katsouyanni et al.. 2009)
31                   found risk estimates ranging from an approximate 1.6 to 5.4% increase in all respiratory-
32                   related hospital admissions and ED visits in all-year analyses for standardized increases
33                   in ambient O3 concentrations1. Positive associations persisted in analyses restricted to the
34                   summer season, but  the magnitude varied depending on the study location (Figure 6-19).
35                   Compared with studies reviewed in the 2006 O3 AQCD, a larger number of recent studies
        1 Effect estimates were standardized to a 20-ppb increase for 24-h avg O3, a 30-ppb increase for 8-h max O3, and a 40-ppb
      increase for 1-h max O3.
      Draft - Do Not Cite or Quote                       6-141                                 September 2011

-------
 1                   examined hospital admissions and ED visits for specific respiratory outcomes. Although
 2                   still limited in number, both single- and multicity studies found consistent, positive
 3                   associations between short-term O3 exposures and asthma and COPD hospital admissions
 4                   and ED visits, with more limited evidence for pneumonia. Consistent with the
 5                   conclusions  of the 2006 O3 AQCD, in studies that conducted seasonal analyses, risk
 6                   estimates were elevated in the warm season compared to cold season or all-season
 7                   analyses, specifically for asthma and COPD. Although recent studies did not include
 8                   detailed age-stratified results, the increased risk of asthma hospital admissions
 9                   (Silverman and Ito, 2010; Strickland et al., 2010; Dales et al., 2006) observed for children
10                   strengthens the conclusion from the 2006 O3 AQCD that children are particularly
11                   susceptible to O3-induced respiratory effects (U.S. EPA. 2006b). Although the
12                   concentration-response relationship has not been extensively examined, preliminary
13                   examinations found no evidence of a threshold between short-term O3 exposure and
14                   asthma hospital admissions and pediatric asthma ED visits (Silverman and Ito. 2010;
15                   Strickland etaL 2010).

16                   New evidence extends the potential range of well-established O3-associated respiratory
17                   effects by demonstrating associations between short-term ambient O3 exposure and
18                   respiratory-related mortality. In all-year analyses, a multicontinent (APHENA) and
19                   multicity (PAPA) study found consistent, positive associations with respiratory mortality
20                   for all ages but less consistent evidence in analyses restricted to ages 75+. Further,
21                   multicity studies in the U.S. and Europe that conducted seasonal analyses found stronger
22                   associations during the summer season (Section 6.2.8).

23                   Several studies of respiratory morbidity and mortality evaluated the potential
24                   confounding effects of copollutants, in particular, PM10, PM2 5, or NO2. In most cases,
25                   effect estimates remained robust to the inclusion of copollutants; however, in several
26                   studies, changes were observed in the magnitude of the O3 association. In studies of lung
27                   function and respiratory symptoms, larger effects frequently were estimated for O3 when
28                   copollutants were added to models. Ozone effect estimates for respiratory-related hospital
29                   admissions and ED visits remained relatively robust upon the inclusion of PM and
30                   gaseous pollutants in two-pollutant models (Strickland et al., 2010; Tolbert et al., 2007;
31                   Medina-Ramon et al. 2006). Although copollutant confounding was not extensively
32                   examined in mortality studies, the O3-respiratory mortality relationship was moderately
33                   to substantially sensitive (e.g., increased or attenuated) to the inclusion of PMi0 in
34                   copollutant models (Stafoggia et al., 2010; Katsouvanni et al.. 2009). However,
35                   interpretation of these results requires caution due to the limited PM datasets used in
36                   these studies. Together, these findings across respiratory endpoints provide support for
37                   the independent effects of short-term  ambient O3 exposures.
      Draft - Do Not Cite or Quote                       6-142                                September 2011

-------
 1                   In summary, new studies evaluated since the completion of the 2006 O3 AQCD support
 2                   and expand upon the strong body of evidence that indicated a causal relationship between
 3                   short-term O3 exposure and respiratory health effects. New controlled human exposure
 4                   studies continue to demonstrate O3-induced decreases in FEVi and pulmonary
 5                   inflammation at concentrations as low as 60 ppb. New epidemiologic studies provide
 6                   evidence for associations of ambient O3 exposure with biological markers of pulmonary
 7                   inflammation and oxidative stress. Toxicological studies have continued to support the
 8                   biological plausibility for the O3-induced respiratory effects observed in the controlled
 9                   human exposure and epidemiologic studies. Additionally, recent epidemiologic studies
10                   further confirm that respiratory morbidity and mortality associations are stronger during
11                   the warm/summer months and remain relatively robust after adjustment for copollutants.
12                   However, despite the consistency of association between short-term O3 exposure and
13                   respiratory effects, new evidence suggests that the magnitude of association may be
14                   underestimated due to behavioral modification in response to forecasted air quality
15                   (Section 4.6.4). Collectively, the new evidence integrated across toxicological, controlled
16                   human exposure, and epidemiologic studies, in conjunction with that reviewed in
17                   previous AQCDs, is sufficient to conclude that there is a causal relationship between
18                   short-term O3 exposure and respiratory health effects.
          6.3    Cardiovascular Effects
            6.3.1    Controlled Human Exposure

19                   O3 reacts rapidly on contact with respiratory system tissue and is not absorbed or
20                   transported to extrapulmonary sites to any significant degree as such. Controlled human
21                   exposure studies discussed in the previous AQCDs failed to demonstrate any consistent
22                   extrapulmonary effects. Some controlled human exposure studies have attempted to
23                   identify specific markers of exposure to O3 in blood. Foster et al. (1996) found a
24                   reduction in the serum levels of the free radical scavenger a-tocopherol after O3 exposure.
25                   Liu et al. (1999; 1997) used a salicylate metabolite, 2,3, dehydroxybenzoic acid (DHBA),
26                   to indicate increased levels of hydroxyl radical which hydroxylates salicylate to DHBA.
27                   Increased DHBA levels after exposure to 120 and 400 ppb suggest that O3 increases
28                   production of hydroxyl radical. The levels of DHBA were correlated with changes in
29                   spirometry.
30                   Gong et  al. (1998) observed a small, statistically significant O3-induced increase in the
31                   alveolar-to-arterial PO2 gradient in both healthy (n = 6) and hypertensive (n =  10) adult
32                   males (aged 41-78 years) exposed for 3 hours with exercise to 300 ppb O3. The
      Draft - Do Not Cite or Quote                      6-143                                September 2011

-------
 1                   mechanism for the decrease in arterial oxygen tension in the Gong et al. (1998) study
                                                         JO                   O      V	/    J
 2                   could be due to an O3-induced ventilation-perfusion mismatch. Gong et al. (1998)
 3                   suggested that by impairing alveolar-arterial oxygen transfer, the O3 exposure could
 4                   potentially lead to adverse cardiac events by decreasing oxygen supply to the
 5                   myocardium. The subjects in the Gong et al. (1998) study had sufficient functional
 6                   reserve so as to not experience significant ECG changes or myocardial ischemia and/or
 7                   injury. In studies evaluating the exercise performance of healthy adults, no significant
 8                   effect of O3 on arterial O2 saturation has been observed (Schelegle and Adams. 1986).

 9                   More recently, Fakhri et al. (2009) evaluated changes in HRV among adult volunteers
10                   (n=50; 27 ± 7 years) during 2-h exposures to PM2 5 CAPs (127±62 ug/m3) and  O3
11                   (114±7 ppb), alone and in combination. High frequency HRV was increased following
12                   CAPs-only (p=0.046) and O3-only (p=0.051) exposures, but not in combination. The
13                   standard deviation of NN intervals and the square root of the mean squared differences of
14                   successive NN intervals also showed marginally significant (0.05
-------
                     6.3.2.1    Arrhythmia
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
                     In the 2006 O3 AQCD, conflicting results were observed when examining the effect of O3
                     on arrhythmias (Dockery et al.. 2005; Rich et al., 2005). A study by Dockery et al. (2005)
                     reported no association between O3 levels and ventricular arrhythmias among patients
                     with implantable cardioverter defibrillators (ICD) living in Boston, MA, although when
                     O3 was categorized into quintiles, there was weak evidence of an association with
                     increasing O3 concentration (median O3 concentration: 22.9 ppb). Rich et al. (2005)
                     performed a re-analysis of this cohort using a case-crossover design and detected a
                     positive association between  O3 exposure and ventricular arrhythmias. Recent studies
                     were conducted in various locations and each used a different cardiac episode to define
                     an arrhythmic event and a different time period of exposure, which may help explain
                     observed differences across studies. Ozone levels for each new study are reported in
                     Table 6-29.
13
14
15
16
17
18
19
20
21
Table 6-29 Characterization of ozone concentrations (in ppb) from studies of
arrhythmias
'Reference Location
Anderson et al. (2010) London, England
Metzger et al. (2007) Atlanta, GA
Rich et al. (2006a) St. Louis, MO
Rich et al. (2006b) Boston, MA
Sarnat et al. (2006a) Steubenville, OH
Averaging Time
8-h max
8-h max
Summer only
24-h
1-h
24-h
24-h
Summer and Fall only
5 days
Mean Concentration
(Standard Deviation)
8.08
53.9 (23)
21*
22.2*
22.6*
21.8(12.6)
22.2(9.1)
Upper Range
of Concentration
75th: 11. 5
Max: 148
75th: 31
75th: 33
Max: 119.5
75th: 30.9
Max: 77.5
75th: 28.5
Max: 74.8
75th: 29.1
Max: 44
        'Median presented (information on mean not given).
                     Multiple studies examined O3-related effects on individuals with ICDs. One study of 518
                     ICD patients who had at least 1 tachyarrythmia within a 10-year period (totaling 6287
                     tachyarrhythmic event-days; 1993-2002) was conducted in Atlanta, Georgia (Metzger et
                     al.. 2007). Tachyarrhythmic events were defined as any ventricular tachyarrhythmic
                     event, any ventricular tachyarrhythmic event that resulted in electrical therapy, and any
                     ventricular tachyarrhythmic event that resulted in defibrillation. In the primary analysis,
                     no evidence of an association was observed for a 30 ppb increase in 8-h max O3
                     concentrations and tachyarrhythmic events (OR: 1.00 [95% CI: 0.92,  1.08]; lag 0).
                     Season-specific as well as several sensitivity analyses (including the use of an
      Draft - Do Not Cite or Quote
                                                    6-145
September 2011

-------
 1                   unconstrained distributed lag model [lags 0-6]) were conducted resulting in similar null
 2                   associations. A strength of this study is that it incorporated a large sample size over a
 3                   long time period.

 4                   In a case-crossover analysis, a population of ICD patients in Boston, previously examined
 5                   by (Rich et al.. 2005) was used to assess the association between air pollution and
 6                   paroxysmal atrial fibrillation (PAF) episodes (Rich et al.. 2006b). In addition to
 7                   ventricular arrhythmias, ICD devices may also detect supraventricular arrhythmias, of
 8                   which atrial fibrillation is the most common. Although atrial fibrillation is generally not
 9                   considered lethal, it has been associated with increased premature mortality as well as
10                   hospitalization and stroke. Ninety-one electrophysiologist-confirmed episodes of PAF
11                   were ascertained among 29 patients. An association (OR: 3.86 [95% CI:  1.44, 10.28] per
12                   40 ppb increase in 1-h max O3 concentrations) was observed between increases in O3
13                   during the concurrent hour and PAF episodes (lag 0-h). The estimated OR for the 24-h
14                   moving average concentration was elevated (OR: 1.81 [95% CI: 0.86,  3.83] per 20 ppb),
15                   but weaker than the estimate for the shorter exposure window. The association between
16                   PAF and O3 in the concurrent hour during the cold months was comparable to that during
17                   the warm months. In addition, no evidence of a deviation from linearity between O3
18                   concentration and the log odds of PAF was observed. Authors report that the  difference
19                   between O3 exposure and observed effect between this study (PAF and 1-h O3) and their
20                   previous study (ventricular arrhythmias and 24-h moving average O3) (Rich et al.. 2005)
21                   suggest a more rapid response to air pollution for PAF (Rich et al., 2006b).

22                   In an additional study, Rich et al. (2006a) employed  a case-crossover design to examine
23                   the association between air pollution and 139 confirmed ventricular arrhythmias among
24                   56 ICD patients in St Louis, Missouri. The authors observed a positive association with
25                   O3 (OR: 1.17 [95% CI: 0.58, 2.38] per 20 ppb increase in 24-h moving avg O3
26                   concentrations [lags 0-23 hours]). Although the authors concluded these results were
27                   similar to their results from Boston (Rich et al.. 2005). they postulated that the pollutants
28                   responsible for the increased risk in ventricular arrhythmias are different (O3  and PM2 5 in
29                   Boston and sulfur dioxide in St Louis).

30                   Anderson et al. (2010) used a case-crossover framework to assess air pollution and
31                   activation of ICDs among patients from all 9 ICD clinics in the London National Health
32                   Service hospitals. "Activation" was defined as tachycardias for which the defibrillator
33                   delivered treatment. Investigators modeled associations using unconstrained distributed
34                   lags from 0 to 5 days. The sample consisted of 705 patients with 5,462 activation days
35                   (O3 information was for 543 patients and 4,092 activation days). Estimates  for O3 were
36                   consistently positive, although weak (OR: 1.09 [95% CI: 0.76, 1.55] per 30 ppb for 0-
      Draft - Do Not Cite or Quote                       6-146                                September 2011

-------
 1                   1 day lag; OR: 1.04 [95% CI: 0.60, 1.81] per 30 ppb for 0-5 day lag) (Anderson et al.
 2                   2010).

 3                   In contrast to arrhythmia studies conducted among ICD patients, Sarnat et al. (2006a)
 4                   recruited non-smoking adults (age range: 54-90 years) to participate in a study of air
 5                   pollution and arrhythmias conducted over two  12-week periods during summer and fall
 6                   of 2000 in a region characterized by industrial pollution (Steubenville, Ohio). Continuous
 7                   ECG data acquired on a weekly basis over a 30-minute sampling period were used to
 8                   assess ectopy, defined as extra cardiac depolarizations within the atria (supraventricular
 9                   ectopy, SVE) or the ventricles (ventricular ectopy, VE). Increases in the 5-day moving
10                   average (days 1-5) of O3 were associated with an increased odds of SVE (OR: 2.17 [95%
11                   CI: 0.93, 5.07] per 20 ppb increase in 24-h avg O3 concentrations). A weaker association
12                   was observed for VE (OR: 1.62 [95% CI: 0.54, 4.90] per 20 ppb increase in 24-h avg O3
13                   concentrations). The results of the effect of 5-day O3 on SVE were robust to the inclusion
14                   of SO42" in the model [OR: 1.62 (95% CI: 0.54, 4.90)]. The authors indicate that the
15                   strong associations observed at the 5-day moving averages, as compared to shorter time
16                   periods, suggests a relatively long-acting mechanistic pathways, such as inflammation,
17                   may have promoted the ectopic beats in this population (Sarnat et al.. 2006a).

18                   Although many studies report positive associations, collectively, studies of arrhythmias
19                   report inconsistent results. This may be due to variation in study populations, length and
20                   season of averaging time, and outcome under study.  Future studies are expected to
21                   provide additional evidence for the various outcomes and exposure periods.
                     6.3.2.2    Heart Rate/Heart Rate Variability

22                   In the 2006 O3 AQCD, two large population-based studies of air pollution and HRV were
23                   summarized (Park et al.. 2005b: Liao et al.. 2004a). In addition, the biological
24                   mechanisms and potential importance of HRV were discussed. Briefly, the study of acute
25                   adverse effects of air pollution on cardiac autonomic control is based on the hypothesis
26                   that increased air pollution levels may stimulate the autonomic nervous system and lead
27                   to an imbalance of cardiac autonomic control characterized by sympathetic activation
28                   unopposed by parasympathetic control (U.S. EPA. 2006b). Examples of HRV indices
29                   include the standard deviation of normal-to-normal intervals (SDNN), the square root of
30                   the mean of the sum of the squares of differences between adjacent NN intervals (r-
31                   MSSD), high-frequency power (HF), low-frequency power (LF), and the LF/HF ratio.
32                   Liao et al. (2004a) examined the association between air pollution and cardiac autonomic
33                   control in the fourth cohort examination (1996-1998) of the U.S.-based Atherosclerosis
34                   Risk in Communities Study. A decrease in log-transformed HF was associated with an
      Draft - Do Not Cite or Quote                      6-147                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
increase in O3 concentration among white study participants. Park et al. (2005b)
examined the effects of air pollution on indices of HRV in a population-based study
among men from the Normative Aging Study in Boston, Massachusetts. Several
associations were observed with O3 and HRV outcomes; a reduction in LF was associated
with increased O3 concentration, which was robust to inclusion of PM25. The associations
with all HRV indices and O3 were stronger among those with ischemic heart disease and
hypertension. In addition to these population-based studies included in the 2006 O3
AQCD was a study by Schwartz et al. (2005). who conducted a panel study to assess the
relationship between exposure to summertime air pollution and HRV.  A weak association
of O3 during the hour immediately preceding the health measures was  observed with r-
MSSD among a study population that consisted of mostly older female participants. In
summary, these studies suggest that short-term exposures to O3 are predictors of
decreased HRV and that the relationship may be stronger among certain subgroups. The
generally consistent (although weak) associations between pollutants and reduced cardiac
autonomic control were observed at relatively low pollution concentrations typically
recent studies of O3 and
studies are presented in
HRV and are described below. The O3 concentrations for these
Table 6-30.
Table 6-30 Characterization of ozone concentrations (in ppb) from studies of
heart rate variability
Reference Location
Baja et al. (2010) Boston, MA
Chan et al. (2005a) Taipei, Taiwan
Chuang et al. (2007a) Taipei, Taiwan
Chuang et al. (2007b) Taipei, Taiwan
Park etal. (2007) Boston, MA
Park etal. (2008) Boston, MA
Ruidavets et al. (2005a) Toulouse, France
Wheeler et al. (2006) Atlanta, GA
Wu et al. (2010) Taipei, Taiwan
Zanobetti et al. (2010) Boston, MA
Averaging Time
Olag
10-hlag
1-h
24-h
48-h
72-h
1-h
24-h
24-h
8-h
4-h
24-h
Working period
0.5-h
2-h
3-D
5-D
Mean Concentration
(Standard Deviation)
23(16)
21 (15)
21.9(15.4)
28.4(12.1)
33.3 (8.9)
33.8(7.1)
35.1
Range of 17.0-29.1
23.4(13)
38.3(14.8)
18.5
29.4
24.9(14.0)
20.7*
20.5*
21.9*
22.8*
Upper Range of
Concentration

Max: 114.9
Max: 49.3
Max: 47.8
Max: 48.3
Max: 192.0


75th: 46.9
Max: 80.3
75th: 22.5
Max: 59.2
75th: 30.33
75th: 30.08
75th: 28.33
75th: 29.28
      'Median presented (information on mean not given).
      Draft - Do Not Cite or Quote
                               6-148
September 2011

-------
 1                   Several follow-up examinations of HRV were conducted among the participants of the
 2                   Normative Aging Study in Boston. A trajectory cluster analysis was used to assess
 3                   whether pollution originating from different locations had varying relationships with
 4                   HRV (Park et al.. 2007). Subjects who were examined on days when air parcels
 5                   originated in the west had the strongest associations with O3; however, the O3
 6                   concentration in this cluster was low (24-h avg, 17.0 ppb) compared to the other clusters
 7                   (24-h avg of 21.3-29.1 ppb). LF and SDNN decreased with increases in the 4-h moving
 8                   average of O3 from the west (LF decreased by 51.2% [95% CI:  1.6, 75.9%] and SDNN
 9                   decreased by 28.2% [95% CI: -0.5, 48.7%] per 30 ppb increase in 4-h avg O3
10                   concentrations) (Parketal.. 2007). The Boston air mass originating in the west traveled
11                   over Illinois, Indiana, and Ohio; states typically characterized by coal-burning power
12                   plants. Due to the low O3 concentrations observed in the west cluster, the authors
13                   hypothesize that O3 on those days could be capturing the effects of other, secondary
14                   and/or transported pollutants from the coal belt or that the relationship between ambient
15                   O3 and personal exposure to O3 is stronger during that period (supported by a
16                   comparatively low apparent temperature which could indicate a likelihood to keep
17                   windows open and reduced air conditioning use) (Park et al.. 2007). An additional
18                   follow-up evaluation using the Normative Aging Study examined the potential for effect
19                   modification by chronic lead exposure on the relationship between air pollution and HRV
20                   (Park et al.. 2008). Authors observed graded reductions in HF and LF of HRV in relation
21                   to O3 (and sulfate) across increasing quartiles of tibia and patella lead (HF: %change 32.3
22                   [95% CI: -32.5, 159.3] for the first quartile of tibia Pb and -59.1 [95% CI: -77.3, -26.1]
23                   for the fourth quartile of tibia Pb per 30 ppb increase in 4-h avg O3 concentrations; LF:
24                   %change 8.0 [95% CI: -36.9, 84.9] for the first quartile of tibia Pb and -59.3 [95% CI: -
25                   74.6, -34.8] for the fourth quartile of tibia Pb per 30 ppb increase in 4-h avg O3
26                   concentrations). In addition, O3 associations were similar when education and cumulative
27                   traffic-adjusted bone lead levels were used in analyses. Authors indicate the possibility
28                   that O3 (which has low indoor concentrations) was acting as a proxy for sulfate
29                   (correlation coefficient for O3 and sulfate = 0.57). Investigators of a more recent follow-
30                   up to the Normative Aging Study hypothesized that the relationships between short-term
31                   air pollution exposures and ventricular repolarization, as measured by changes in the
32                   heart-rate corrected QT interval (QTc), would be modified by participant characteristics
33                   (e.g., obesity, diabetes, smoking history) and genetic susceptibility to oxidative stress
34                   (Bajaetal.. 2010). No evidence of an association between O3 (using a quadratic
35                   constrained distributed lag model and hourly exposure lag models over a 10-h time
36                   window preceding the visit) and QTc was reported (change in mean QTc -0.74 [95% CI:
37                   -3.73, 2.25]); therefore, potential effect modification of personal and genetic
38                   characteristics with O3 was not  assessed (BajaetaL 2010). Collectively, the results from
39                   studies that examined the Normative Aging Study cohort found an association between
      Draft - Do Not Cite or Quote                       6-149                                 September 2011

-------
 1                   increases in short-term exposures to O3 and decreases in HRV (Park et al., 2008; Park et
 2                   al.. 2007; Park et al.. 2005b) although not consistently in all of the studies (Baja et al..
 3                   2010). Further, observed relationships appear to be stronger among those with ischemic
 4                   heart disease, hypertension, and elevated bone lead levels, as well as when air masses
 5                   arrive from the west (the coal belt). However, it is not clear if O3 is acting as a proxy for
 6                   other, secondary particle pollutants (such as sulfate) (2008; 2007; Parketal.. 2005b). In
 7                   addition, since the Normative Aging Study participants were older, predominately white
 8                   men, results may not be generalizable to women, younger individuals, or those of
 9                   different racial/ethnic groups (Bajaet al., 2010).

10                   Additional studies of populations not limited to the Normative Aging Study have also
11                   examined associations between O3 exposure and HRV. A panel study among  18
12                   individuals with COPD and 12 individuals with recent myocardial infarction (MI) was
13                   conducted  in Atlanta, Georgia (Wheeler et al.. 2006). HRV was assessed for each
14                   participant on 7 days in fall 1999 and/or spring 2000. The mean 4-h O3 concentration
15                   (time period immediately preceding the HRV measures) was 18.5 ppb; however, O3
16                   concentrations differed substantially within study sites (8.0 - 33.8 ppb). Ozone
17                   concentrations were not associated with HRV (SDNN) among all subjects (percent
18                   change of 2.36% [95% CI: -10.8%, 17.5%] per 30 ppb 4-h O3 increase) or when stratified
19                   by disease  type (COPD, recent MI, and baseline FEVO (Wheeler et al.. 2006).

20                   HRV and air pollution was assessed in a panel study among 46 predominately white male
21                   patients (study population: 80.4% male, 93.5% white) aged 43-75 years in Boston,
22                   Massachusetts, with coronary artery disease (Zanobetti et al., 2010). Up to four home
23                   visits were made to  assess HRV over the year following the index event. Pollution lags
24                   used in analyses ranged between 30 minutes to a few hours and up to 5 days prior to the
25                   HRV assessments. Decreases in r-MSSD were reported for all averaging times of O3
26                   (percent change of-5.18% [95% CI:  -7.89, -2.30] per 20 ppb of 5-day moving average of
27                   O3 concentration), but no evidence of an association between O3 and HF was observed
28                   (quantitative results not provided). In two-pollutant models with O3 and either PM2 5 or
29                   BC, O3 associations remained robust.

30                   A few studies were  conducted outside of the U.S. to assess the relationship between air
31                   pollution concentrations and heart rate and HRV (Wu et al., 2010; Chuang et al., 2007b;
32                   Chuang et  al.. 2007a; Chan et al.. 2005a; Ruidavets et al.. 2005a). No associations were
33                   reported between O3 and HRV among CHD patients and patients with one or more major
34                   CHD risk factors residing in Taipei, Taiwan (Chan et al.. 2005a). Another study in
3 5                   Taipei, Taiwan examined mail carriers and reported O3 levels measured using personal
36                   monitors. No association was observed between O3 and the measures of HRV (percent
37                   change for SDNN: 0.57 [95% CI: -21.27, 28.46], r-MSSD:  -7.10 [95% CI: -24.24, 13.92],
      Draft - Do Not Cite or Quote                      6-150                                September 2011

-------
 1                   HF: -1.92 [95% CI: -23.68, 26.02], LF: -4.82 [95% CI: -25.34, 21.35] per 40 ppb O3)
 2                   ("Wuetal.. 2010). In addition, no consistent relationships were identified between O3 and
 3                   resting heart rate among middle-aged (35-64 years) participants residing in Toulouse,
 4                   France (Ruidavets et al.. 2005a). A negative trend was reported for the 3-day cumulative
 5                   (lag days 1-3) concentration of O3 with heart rate (p for trend = 0.02); however, the
 6                   individual odds ratios comparing quintiles of exposure showed no association (OR for O3
 7                   ofO.93 [95% CI: 0.86, 1.01] for the highest quintile of resting heart rate compared to the
 8                   lowest). When stratified by current smoking status, non-smokers had a decreased trend
 9                   with increased 3-day cumulative O3 concentrations but none of the quintiles for heart rate
10                   were statistically significant. A panel study  was conducted in Taiwan to assess the
11                   relationship between air pollutants and inflammation, oxidative stress, blood coagulation,
12                   and autonomic dysfunction (Chuang et al.. 2007b: Chuang et al.. 2007a). Participants
13                   were apparently healthy college students (aged 18-25 year) who were living in a
14                   university dormitory in metropolitan Taipei. Health endpoints were measured three times
15                   from April to June in 2004 or 2005. Ozone was assessed in statistical models using the
16                   average  of the 24, 48, and 72 hours before the  hour of each blood sampling. Decreases in
17                   HRV (measured as SDNN, r-MSSD, LF, and HF) were associated with increases in O3
18                   concentrations in single-pollutant models (percent change for SDNN: -13.45 [95% CI: -
19                   16.26, -10.60], r-MSSD -13.76 [95% CI: -21.62, -5.44], LF -9.16 [95% CI: -13.29, -
20                   4.95], HF -10.76 [95% CI: -18.88, -2.32] per 20 ppb  3-day avg O3 concentrations) and
21                   remained associated with 3-day O3 concentrations in two-pollutant models with sulfate.
22                   Another study in Taiwan recruited individuals with coronary heart disease or at risk for
23                   cardiovascular disease from outpatient clinics  (Chuang et al.. 2007b). Mean O3
24                   concentrations were 35.1 ppb (SD 27.5 ppb) during the study period (two weeks in
25                   February). No association was observed between O3  concentration and HRV measures
26                   (SDNN, r-MSSD, LF, HF) (numerical results not provided in publication).

27                   Overall, studies of O3 concentration and HRV report inconsistent results. Multiple studies
28                   in Boston observed positive associations but the authors of many of these studies
29                   postulated that O3 was possibly acting as a proxy for other pollutants. The majority of
30                   other studies, both in the U.S. and internationally, report null findings. The
31                   inconsistencies observed are further complicated by the different HRV measures and
32                   averaging times used by the studies.
                     6.3.2.3    Stroke

33                   The 2006 O3 AQCD did not identify any studies that examined the association between
34                   short-term O3 exposure and stroke. However, recent studies have attempted to examine
35                   this relationship. Lisabeth et al. (2008) used a time-series approach to assess the

      Draft - Do Not Cite or Quote                       6-151                                September 2011

-------
 1                   relationship between daily counts of ischemic stroke and transient ischemic attack (TIA)
 2                   with O3 concentrations in a southeast Texas community among residents 45 years and
 3                   older (2001-2005; median age of cases, 72 years). The median O3 (hourly average per 24-
 4                   h time-period) concentration was 25.6 ppb (IQR 18.1-33.8). The associations between
 5                   same-day (RR: 1.03 [95% CI: 0.96, 1.10] per 20 ppb increase in 24-h avg O3
 6                   concentrations) and previous-day (RR: 1.05 [95% CI: 0.99, 1.12] per 20 ppb increase in
 7                   24-h avg O3 concentrations) O3 concentrations and stroke/TIA risk were positive.
 8                   Associations were robust to adjustment for PM2 5. The effect of season on the relationship
 9                   was not assessed.

10                   A case-crossover design was used in a study conducted in Dijon, France between March
11                   1994 and December 2004, among those 40 years of age and older who presented with
12                   first-ever stroke (Henrotin et al., 2007). The mean O3 concentration, calculated over 8-h
13                   daytime periods, was 14.95 ppb (IQR:  6-22 ppb). No association was observed between
14                   O3 concentration at 0, 1, 2, or 3 days lag and hemorrhagic stroke. However, an
15                   association between ischemic stroke occurrence and O3 concentrations with a 1-day lag
16                   was observed (OR: 1.54 [95% CI: 1.14, 2.09] per 30 ppb increase in 8-h max O3
17                   concentrations). The effect of O3 persisted in two-pollutant models with PMi0, SO2, NO2,
18                   or CO. This association was stronger among men (OR: 2.12 [95% CI: 1.36, 3.30] per 30
19                   ppb increase in 8-h max O3 concentrations) than among women (OR: 1.17 [95%CI: 0.77,
20                   1.78] per 30 ppb increase in 8-h max O3 concentrations) in single pollutant models. When
21                   stroke was examined by subtype among men, an association was observed for ischemic
22                   strokes of large arteries and for transient ischemic attacks but not for cardioembolic or
23                   lacunar ischemic strokes. The subtype analysis was not performed for women.
24                   Additionally, for men a linear exposure-response was observed when O3 was assessed
25                   based on quintiles (p for trend = 0.01) (Figure 6-21). A potential limitation of this study
26                   is that 67.4% of the participating men were smokers compared to 9.3% of the women.

27                   Another study, performed  in Dijon, France, examined the association between O3
28                   concentration and incidence of fatal and non-fatal ischemic cerebrovascular events
29                   (ICVE) (Henrotin etal.. 2010V Mean 8-h O3 concentration was 19.1 ppb (SD 12.2 ppb).
30                   A positive association was observed between recurrent ICVE and O3 concentration with a
31                   3-day lag (OR: 1.92 [95% CI 1.17, 3.12]), butnotfor other lags (0, 1, 2, 4) or cumulative
32                   days (0-1, 0-2, 1-2, 1-3). Although some ORs for incident ICVEs were elevated, none
33                   were statistically significant. Results for associations using the maximum daily 1-h O3
34                   concentrations were similar to the 8-h results but slightly attenuated. ORs were similar in
35                   two pollutant models (data not given). In stratified analyses, the association between 1-
36                   day lagged O3 concentration and incident and recurrent ICVE was greater among those
37                   with multiple other preexisting vascular conditions. Increased associations with ICVE
38                   were also observed for individuals with diabetes or hypertension.
      Draft - Do Not Cite or Quote                      6-152                               September 2011

-------
                     3.5
                      3 -
                     2.5 -
                   o

                   I  2
                     1.5 -
                     0.5 -
                               0-8       9-20      21-32      33-48
                                          O3 concentration (ppb)
 Source: Henrotin et al. (2007).

Figure 6-21    Odds ratio (95% confidence interval) for stroke by quintiles of
                ozone
1
2
3
4
5
6
              6.3.2.4   Biomarkers

              An increasing number of studies have examined the relationship between air pollution
              and biomarkers in an attempt to elucidate the biological mechanisms linking air pollution
              and cardiovascular disease. A wide range of markers assessed as well as different types
              of study designs and locations chosen make comparisons across studies difficult. Table 6-
              31 provides an overview of the O3 concentrations reported in each of the studies
              evaluated.
Draft - Do Not Cite or Quote
                                                  6-153
September 2011

-------
Table 6-31 Characterization of ozone concentrations (in ppb) from studies of
biomarkers
Reference
Baccarelli et al. (2007)
Chen etal. (2007)
Chuanaetal. (2010)
Chuang et al. (2007a)
Goldberg et al. (2008)
Liao et al. (2005)
Rudez etal. (2009)
Steinvil etal. (2008)
Thompson etal. (2010)
Wellenius et al. (2007)
Location
Lombardia, Italy
Los Angeles and
San Francisco, CA
Taiwan
Taipei, Taiwan


Montreal, Quebec
3 U.S. counties
Rotterdam, the Netherlands
Tel -Aviv, Israel
Toronto, Ontario
Boston, MA
Averaging
Time
1-h
8-h/2 wk
8-h/1 mo

24-h
48-h
72-h
24-h
8-h
24-h
0.5-h
1-h/1yr
1-h/24-h
Mean Concentration
(Standard Deviation)
18.3*
30.8*
28.3*
26.83 (9.7)
28.4(12.1)
33.3 (8.9)
33.8(7.1)
NS
40 (20)
22*
29.2 (9.7)
21.94(15.78)
25.1 (12.9)
Upper Range of
Concentration
75th: 35.1
Max: 202.3
Max: 47.9
Max: 43.1
Max: 62.1
Max: 49.3
Max: 47.8
Max: 48.3


75th: 31 .5
Max: 90
75th: 36


      'Median presented (information on mean not given).
 1
 2
 3
 4
 5

 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
Hemostasis and coagulation markers

Multiple studies used various markers to examine if associations were present between
O3 concentrations and hemostatis and coagulation. Some of the markers included in these
studies were as follows: fibrinogen, von Willebrand factor (vWF), plasminogen activator
fibrinogen inhibitor-1 (PAI-1), tissue-type plasminogen activator (tPA), platelet
aggregation, and thrombin generation.

A population-based study in the United States was conducted to assess the relationship
between short-term exposure to air pollution and markers of blood coagulation using the
Atherosclerosis Risk in Communities (ARIC) study cohort (Liao etal.. 2005). Significant
curvilinear associations were observed for O3 (1 day prior to blood draw) and fibrinogen
and vWF (quantitative results not provided for regression models although adjusted
means [SE] of vWF were given as  118% [0.79%] for O3 concentrations <40 ppb, 117%
[0.86%] for O3 concentrations 40-70 ppb, and 124% [1.97%]  for O3  concentrations of 70
ppb). The association between O3 and fibrinogen was more pronounced among those
with a history of cardiovascular disease (CVD) and was statistically significant among
only this subgroup of the population. The curvilinear relationship between exposure and
outcome suggested stronger relationships at higher concentrations of O3 which could
indicate threshold effects. The authors note that the most pronounced associations
occurred when the pollutants were 2-3 standard deviations above the mean. The results
      Draft - Do Not Cite or Quote
                               6-154
September 2011

-------
 1                   from this relatively large-scale cross-sectional study suggest weak associations with O3
 2                   and fibrinogen (among those with a history of CVD) and vWF.

 3                   A retrospective repeated measures analysis was performed in Toronto, Canada among
 4                   adults aged 18-40 years (n=45) between the years of 1999 and 2006 (Thompson et al.,
 5                   2010). Single pollutant models were used with moving averages up to 7 days. No
 6                   evidence of an association was observed for O3 and fibrinogen.

 7                   A repeated measures study was conducted in 40 healthy individuals living or working in
 8                   the city center of Rotterdam, the Netherlands to assess the relationship between air
 9                   pollution and markers of hemostatis and coagulation (platelet aggregation, thrombin
10                   generation, and fibrinogen) (Rudez et al.. 2009). Each participant provided between 11
11                   and 13 blood samples throughout a 1-year period (498 samples on 197 days). Examined
12                   lags ranged from 6 hours to 3 days prior to blood sampling. No consistent evidence of an
13                   association was observed between O3 and any of the biomarkers (percent change of max
14                   platelet aggregation: -6.87 [95% CI: -21.46, 7.70] per 20 ppb 4-day average O3; percent
15                   change of endogenous thrombin potential: 0.95 [95% CI: -3.05, 4.95] per 20 ppb 4-day
16                   avg O3; percent change of fibrinogen: -0.57 [95% CI: -3.05, 2.00] per 20 ppb lag 1-day
17                   O3). Some associations with O3 were in the opposite direction to that hypothesized  which
18                   may be explained by the negative  correlation between O3 and the other pollutants
19                   (correlation coefficients ranged from -0.4 to -0.6). The statistically significant inverse
20                   effects observed with O3 in single-pollutant models were no longer apparent when PM10
21                   was included in the models (Rudez et al.. 2009).

22                   A panel study in Taiwan measured health endpoints using blood samples from healthy
23                   individuals (n=76) at three times from April to June in 2004 or 2005 (Chuang et al..
24                   2007a). Increases in fibrinogen and PAI-1 were associated with increases in O3
25                   concentrations in single-pollutant  models (percent change in fibrinogen: 11.76 [95% CI:
26                   4.03, 19.71] per 20 ppb 3-day avg O3; percent change in PAI-1: 6.08 [95% CI: 38.91,
27                   84.27] per 20 ppb 3-day avg O3). These associations were also observed at 1 and 2 day
28                   averaging times. Associations between PAI-1 and 3-day O3 concentrations remained
29                   robust in two-pollutant models with sulfate. No association was seen between O3 and
30                   tPA, a fibrinolytic factor (percent change 16.15 [95% CI: -4.62, 38.34] per 20 ppb 3-day
31                   avg03).

32                   A study in Israel examined the association between pollutant concentrations and
33                   fibrinogen among 3659 apparently healthy individuals (Steinvil et al.. 2008). In single
34                   pollutant models, O3 was associated with an increase in fibrinogen at a 4-day lag among
35                   men and a same-day O3 concentration among women but results for other lags (0 through
36                   7 days) were mixed (some positive, some negative; none statistically significant).
      Draft - Do Not Cite or Quote                      6-155                                September 2011

-------
                     Inflammatory markers

 1                   Air pollution and inflammatory markers (C-reactive protein [CRP], white blood cell
 2                   [WBC] count, albumin, and Interleukin-6 [IL-6]) were also examined in several studies.

 3                   The ARIC study cohort, which included men and women aged 45-64 years old at the start
 4                   of the study, was utilized to assess the association between O3 concentrations and makers
 5                   of inflammation (Liao et al., 2005). No association was observed between O3
 6                   concentrations and albumin or WBC count.

 7                   Thompson et al. (2010) assessed ambient air pollution exposures and IL-6. This
 8                   retrospective repeated measures analysis was conducted among 45 adults (18-40 years of
 9                   age) in Toronto, Canada between the years of 1999 and 2006. Single pollutant models
10                   were used to analyze the repeated-measures data using moving averages up to 7 days. A
11                   positive association was observed between IL-6 and O3 with the strongest effects
12                   observed for the 4-day moving average of O3 (quantitative results not provided). No
13                   association was seen for shorter averaging times (<1 day). When examined by season
14                   using 2-day moving averages, the association between O3 and IL-6 was positive during
15                   only the spring and summer.

16                   In Rotterdam, the Netherlands, a repeated measures study of healthy individuals living or
17                   working in the city center reported no association between O3 concentration and CRP
18                   (Rudez et al.. 2009). Each of the 40 participants provided between 11 and 13 blood
19                   samples throughout a 1-year period (498 samples on 197 days). No consistent evidence of
20                   an association was observed between O3 and CRP (percent change: -0.48 [95% CI: -
21                   14.05, 13.10] per 20 ppb lag 1-day O3). Additionally, no association was observed with 2
22                   or 3 day lags.

23                   The relationship between pollutant concentrations and one-time measures of
24                   inflammatory biomarkers was assessed in sex-stratified analyses among 3659 apparently
25                   healthy individuals in Tel Aviv, Israel (Steinvil et al..  2008). No evidence of an
26                   association was observed between O3 and CRP or WBC for men and women.

27                   A panel study of healthy individuals  (n=76) was conducted in Taiwan to assess the
28                   relationship between air pollutants and inflammation (Chuang et al.. 2007a). Health
29                   endpoints were measured three times from April to June in 2004 or 2005. Ozone effects
30                   were assessed in statistical models using the average of the 24 hours (1 day), 48 hours
31                   (2 days), and 72 hours (3 days) before the hour of each blood sampling. Increases in CRP
32                   were associated with increases in O3  concentrations in single-pollutant models (percent
33                   change in CRP: 244.38 [95% CI: 4.54, 585.15] per 20 ppb 3-day avg O3).  The association
34                   was also observed using a 2-day averaging time, but no association was present with a 1-
3 5                   day averaging time.


      Draft - Do Not Cite or Quote                      6-156                                September 2011

-------
                     Oxidative stress markers

 1                   A few studies have reported on the relationships between O3 concentration and oxidative
 2                   stress markers. The association between O3 exposure and markers of lipid peroxidation
 3                   and antioxidant capacity was examined among 120 nonsmoking healthy college students,
 4                   aged 18-22 years, from the University of California, Berkeley (February-June 2002)
 5                   (Chen et al.. 2007). By design, students were chosen that had experienced different
 6                   geographic concentrations of O3 over their lifetimes and during recent summer vacation
 7                   in either greater Los Angeles (LA) or the San Francisco Bay Area (SF). Long-term
 8                   (based on lifetime  residential history) and shorter-term (based on the moving averages of
 9                   8-h max concentrations 1-30 days prior to the day of blood collection) O3 exposures were
10                   estimated (lifetime exposure results presented in the chronic exposure section). A marker
11                   of lipid peroxidation, 8-isoprostane (8-iso-PGF), was assessed. This marker is formed
12                   continuously under normal physiological conditions but has been found at elevated
13                   concentrations in response to environmental exposures. A marker of overall antioxidant
14                   capacity, ferric reducing ability of plasma (FRAP), was also measured. Substantial
15                   overlap in the more recent O3 exposure estimates (8-h moving averages) was observed
16                   between the two geographic areas sampled. Levels of 8-iso-PGF were associated with
17                   2-week ((3 = 0.035 [pg/mL]/8-h ppb O3, p = 0.007) and 1-month ((3 = 0.031 [pg/mL]/8-
18                   h ppb O3, p = 0.006) estimated O3 exposure levels. No evidence of association was
19                   observed between  O3 and FRAP. A chamber study performed among a subset of study
20                   participants supported the primary study results. The concentrations of 8-iso-PGF
21                   increased immediately after the 4-h controlled O3 exposure ended (p = 0.10). However,
22                   levels returned to near baseline by 18 hours without further exposure. The authors note
23                   that O3 was highly correlated with PMi0-2.5 and NO2 in this study population;  however,
24                   inclusion of these pollutants in the O3 models did not substantially  change the magnitude
25                   of the associations with O3.

26                   Using blood samples collected between April and June of 2004 or 2005 in Taiwan, the
27                   association between O3 concentrations  and a marker of oxidative stress was studied
28                   among healthy individuals (n=76) (Chuang et al.. 2007a).  Increases in 8-hydroxy-2'-
29                   deoxyguanosine (8-OHdG) were associated with increases in O3 concentrations in single-
30                   pollutant models (percent change in 8-OHdG: 2.46 [95% CI:  1.01,  3.92] per 20 ppb  1-day
31                   avg O3). The association did not persist with 2- or 3-day averaging times.


                     Markers of overall cardiovascular health

32                   Multiple studies used markers that assess overall cardiovascular well-being. Wellenius et
33                   al. (2007) examined B-type  natriuretic peptide (BNP), a marker of heart failure, in a
34                   repeated-measures study conducted in Boston among 28 patients with congestive  heart
      Draft - Do Not Cite or Quote                      6-157                               September 2011

-------
 1                   failure and impaired systolic function. The authors found no evidence of an association
 2                   between BNP and short-term O3 exposures at lags 0-3 days (quantitative results not
 3                   provided). BNP was chosen because it is directly associated with cardiac hemodynamics
 4                   and symptom severity among those with heart failure and is, therefore, considered a
 5                   marker of functional status. However, the authors conclude that the use of BNP may not
 6                   be useful in studies  of the health effects of ambient air pollutants due to the large amount
 7                   of within-person variability in BNP levels observed in this population.

 8                   The relationship between air pollution and oxygen saturation and pulse rate, markers of
 9                   physiological well-being, was examined in a 2-month panel study among 31 congestive
10                   heart failure patients (aged 50-85 years) in Montreal, Canada from July 2002 to October
11                   2003 (Goldberg et al.. 2008). All participants had limited physical functioning
12                   (New York Heart Association Classification > II) and an ejection fraction (the fraction of
13                   blood pumped out of the heart per beat) less than or equal to 35% (normal is above 55%).
14                   Daily mean O3 concentrations were calculated based on hourly measures at  10 monitoring
15                   stations. There was  an inverse association between O3 (lag-0) and oxygen saturation
16                   when adjustment was made for temporal trends. In the models incorporating personal
17                   covariates and weather factors, the association remained but was not statistically
18                   significant. The associations of O3 with a lag of 1 day or a 3-day mean were not
19                   statistically significant. No evidence of association was observed between O3 exposure
20                   and pulse rate.

21                   Total homocysteine (tHcy) is an independent risk factor for vascular disease and
22                   measurement of this marker after oral methionine load is used to identify individuals with
23                   mild impairment of homocysteine metabolism. The effects of air pollution on fasting and
24                   postmethionine-load tHcy levels were assessed among 1,213 apparently healthy
25                   individuals from Lombardia, Italy from January 1995 to September 2005 (Baccarelli et
26                   al.. 2007). An increase  in the 24-h O3 concentrations was associated with an increase in
27                   fasting tHcy (percent change 6.25 [95% CI: 0.84, 11.91] per 20 ppb O3) but no
28                   association was observed with postmethionine-load tHcy (percent change 3.36 [95% CI: -
29                   1.30, 8.39] per 20 ppb O3). In addition, no evidence of association was observed between
30                   7-day O3 concentrations and tHcy (percent change for fasting tHcy 4.16  [95% CI: -1.76,
31                   10.42] and percent change for postmethionine-load tHcy -0.65 [95% CI: -5.66, 4.71] per
32                   20 ppb O3). No evidence of effect modification by smoking was observed.


                     Blood  lipids and glucose metabolism markers

33                   Chuang et al.  (2010) conducted a population-based cross-sectional analysis of data
34                   collected on 7,778 participants during the Taiwanese Survey on Prevalence of
35                   Hyperglycemia, Hyperlipidemia, and Hypertension in 2001. Apolipoprotein B (ApoB),
      Draft - Do Not Cite or Quote                      6-158                                September 2011

-------
 1                   the primary apolipoprotein among low-density lipoproteins, was associated with 3-day
 2                   avg O3 at the p < 0.10 level. The 5-day mean O3 concentration was associated with an
 3                   increase in triglycerides at p < 0.10. In addition, the 1-, 3-, and 5-day mean O3
 4                   concentrations were associated with increased HbAlc levels (a marker used to monitor
 5                   the degree of control of glucose metabolism) at the p < 0.05 level. The 5-day mean O3
 6                   was associated with increased fasting glucose levels (p < 0.10). No association was
 7                   observed between O3 concentration and ApoAl. Copollutant models were not assessed.
                     6.3.2.5    Myocardial Infarction (Ml)

 8                   The 2006 O3 AQCD did not report consistent results indicating an association between
 9                   short-term O3 exposure and MI. One study reported a positive association between
10                   current day O3 concentration and acute MI, especially among the oldest age group (55- to
11                   64-year olds) (Ruidavets et al.. 2005b). No association was observed in a case-crossover
12                   study of O3 during the hours  surrounding the event and MI (Peters etal.. 2001). Since the
13                   2006 O3 AQCD, a few new epidemiologic studies have examined the association between
14                   O3 exposure and MI (Henrotin et al., 2010; Rich et al.. 2010). as well as one study
15                   published on arterial stiffness CWuet al.. 2010) and one study published on ST-segment
16                   depression (Delfino et al., 2011).

17                   One of the studies conducted in the U.S. examined hospital admissions for first MI and
18                   reported no association with  O3 concentrations (Rich et al.. 2010). More details on this
19                   study are reported in the section on hospital admissions. Another study, performed in
20                   Dijon, France, examined the  association between O3 concentration and incident and
21                   recurrent MI (Henrotin et al.. 2010). The mean 8-h O3 concentration was 19.1 ppb (SD
22                   12.2 ppb).  Odds ratios for the association between cumulative O3 concentrations and
23                   recurrent Mis were elevated  but none of the results were statistically significant (OR:
24                   1.71 [95% CI: 0.91, 3.20] per 20 ppb for cumulative 1-3 day O3 exposure). No
25                   association was observed for incident Mis. In analyses stratified by vascular risk factors,
26                   positive associations were observed between 1-day lagged O3 concentrations and Mis
27                   (incident and recurrent combined) among those who reported having
28                   hypercholesterolaemia (OR:  1.52 [95% CI: 1.08, 2.15] per 20 ppb O3) and a slight inverse
29                   association was observed among those who reported not having hypercholesterolaemia
30                   (OR: 0.69  [95% CI: 0.50, 0.94] per 20  ppb O3). In other stratified analyses combining
31                   different vascular factors, only those containing individuals with hypercholesterolaemia
32                   demonstrated a positive association; none were inverse associations.

33                   Wu et al. (2010) examined mail carriers aged 25-46 years and measured exposure to O3
34                   through personal monitors [mean O3 24.9 (SD  14.0) ppb]. Ozone exposure was positively
      Draft - Do Not Cite or Quote                      6-159                               September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
associated with arterial stiffness (percent change 11.24% [95% CI: 3.67, 19.62] per 40
ppb O3) and was robust to adjustment for ultrafme PM.

A study performed in the Los Angeles basin reported on the association between O3
exposure and ST-segment depression, a measure representing cardiac ischemia (Delfino
et al.. 2011). Study participants were nonsmokers, at least 65 years old, had a history of
coronary artery disease, and were living in a retirement community. Study periods
included five consecutive days in both July to mid-October and mid-October to February.
Mean 24-h O3 concentrations were 27.1 ppb (SD 11.5 ppb). No association was observed
between O3 concentrations and ST-segment depression of at least 1.0 mm during any of
the exposure periods (i.e., 1-h, 8-h, 1-day, 2-day avg, 3-day avg, 4-day avg).
11
12
13
14
6.3.2.6    Blood Pressure

In the 2006 O3 AQCD, no epidemiologic studies examined O3-related effects on blood
pressure (BP). Recent studies have been conducted to evaluate this relationship and
overall the findings are inconsistent. The O3 concentrations for these studies are listed in
Table 6-32.
      Table 6-32     Characterization of ozone concentrations (in ppb) from studies of
                      blood pressure
Reference
Choi etal. (2007)
Delfino et al. (201 Ob)
Zanobetti et al. (2004)
Chuang etal. (2010)
Location
Incheon, South Korea
Los Angeles,
California
Boston,
Massachusetts
Taiwan
Averaging Time
8-h
(warm season)
8-h
(cold season)
24-h
1-h
5-days

Mean Concentration
(Standard Deviation)
26.6(11.8)
17.5(7.3)
27.1 (11.5)
20
24
26.83 (9.7)
Upper Range of
Concentration
75th: 34.8
Max: 62.4
75th: 22.9
Max: 33.9
Max: 60.7


Max: 62.1
15
16
17
18
19
20
21
Zanobetti et al. (2004) examined the relationship between air pollutants and BP from
May 1999 to January 2001 for 631 repeat visits among 62 Boston residents with CVD. In
single-pollutant models, higher resting diastolic blood pressure (DBP) was associated
with the 5-day (0-4 days) averages of O3 (RR: 1.03 [95% CI: 1.00, 1.05] per 20 ppb
increase in 24-h O3 concentrations). However, this effect was no longer apparent when
PM2 5 was included in the model (data were not presented) (Zanobetti et al.. 2004).
Delfino et al. (201 Ob) examined 64 subjects 65 years and older with coronary artery
      Draft - Do Not Cite or Quote
                              6-160
September 2011

-------
 1                  disease, no tobacco smoke exposure, and living in retirement communities in the
 2                  Los Angeles air basin with hourly (up to 14-h/day) ambulatory BP monitoring for 5 days
 3                  during a warm period (July-mid-October) and 5 days during a cool period (mid-October-
 4                  February). Investigators assessed lags of 1, 4, and 8 hours, 1 day, and up to 9 days before
 5                  each BP measure; no evidence of association was observed for O3 exposures (change in
 6                  BP associated with a 20 ppb change in 24-h O3 was 0.67 [95% CI: -1.16, 2.51 for systolic
 7                  BP [SBP] and -0.25 [95% CI: -1.25, 0.75] for DBP) (Delfino et al.. 2010b). Choi et al.
 8                  (2007) conducted a cross-sectional study to investigate the relationship between air
 9                  pollutants and BP among 10,459 participants of the Inha University Hospital health
10                  examination from 2001 to 2003. These individuals had no medical history of
11                  cardiovascular disease or hypertension. O3 exposure was associated with an increase in
12                  SBP for 1-day lag in the warm season and similar effect estimates were observed during
13                  the cold season but were not statistically significant (quantitative results not provided).
14                  Associations between O3 and DBP were present in the cold season but not the warm
15                  season (quantitative results not provided). The interaction term between O3 and season
16                  was statistically significant. Chuang et al. (2010)  conducted a similar type of study
17                  among 7,578 participants of the Taiwanese Survey on Prevalence of Hyperglycemia,
18                  Hyperlipidemia, and Hypertension in 2001. Investigators examined 1-, 3-, and 5-day avg
19                  O3 concentrations. An increase in DBP was associated with the 3-day mean O3
20                  concentration (change in BP for a 20 ppb increase in O3 was 0.61 [95% CI:  0.07, 1.14])
21                  (Chuang et al.. 2010). Associations were not observed for other days or with SBP.
                     6.3.2.7   Hospital Admissions and Emergency Department Visits

22                   Upon evaluating the collective evidence for O3-related cardiovascular hospital admissions
23                   (HAs) and emergency department (ED) visits, the 2006 O3 AQCD concluded that "a few
24                   studies observed positive O3 associations, largely in the warm season. Overall, however,
25                   the currently available evidence is inconclusive regarding any association between
26                   ambient O3 exposure on cardiovascular hospitalizations" (U.S. EPA. 2006b). Table 6-33
27                   below provides information on the O3 concentrations reported in each of the recent HA
28                   and ED visit studies evaluated.
29                   Multiple recent studies of O3 exposure and cardiovascular HAs and ED visits have been
30                   conducted in the U.S. and Canada. Peel et al. (2007) used a case-crossover framework
31                   (using a time-stratified approach matching on day of the week in the calendar month of
32                   the event) to assess the relationship between air pollutants and cardiovascular disease ED
33                   visits among those with and without secondary comorbid conditions (hypertension,
34                   diabetes, chronic obstructive pulmonary disease  [COPD], congestive heart failure [CHF],
      Draft - Do Not Cite or Quote                      6-161                                September 2011

-------
Table 6-33 Characterization of ozone concentrations (in ppb) from studies of
HAs and ED visits
Study
Azevedoetal. (2011)
Ballesteretal. (2006)
Bell et al. (2008)
Buadong et al. (2009)
Cakmaketal. (2006a)
Chan etal. (2006)
Halonenetal. (2009)
Hosseinpoor et al. (2005)
Lanki etal. (2006)
Larrieuetal. (2007)
Lee et al. (2003b)
Lee et al. (2007)
Middleton et al. (2008)
Peel et al. (2007)
Rich etal. (2010)
Stieb et al. (2009)
Symons et al. (2006)
Tolbert etal. (2007)
Villeneuveetal. (2006a)
Von Klot etal. (2005)
Wellenius et al. (2005)
Yang (2008)
Zanobetti and Schwartz (2006)
Location
Portugal
Multicity, Spain
Taipei, Taiwan
Bangkok, Thailand
Multicity, Canada
Taipei, Taiwan
Helsinki, Finland
Tehran, Iran
Multicity, Europe
Multicity France
Seoul, Korea
Kaohsiung, Taiwan
Nicosia, Cyprus
Atlanta, GA
New Jersey
Multicity, Canada
Baltimore, MD
Atlanta, GA
Edmonton, Canada
Multicity, Europe
Allegheny County, PA
Taipei, Taiwan
Boston, MA
Averaging
Time
1-h
8-h
warm season
24-h
1-h
1-h max
1-h max
8-h max
warm season
8-h max
8-h max
warm season
8-h max
warm season
1-h max
24-h
8-h max
8-h max
warm season
24-h
24-h
8-h
warm season
8-h max
warm season
24-h
24-h
warm season
24-h
cold season
8h max
warm season
24-h
24-h
24-h
Mean Concentration
(Standard Deviation)
NR
24.2 - 44.3
21.4
14.4(3.2)
17.4
50.9 (26.4)
35.7*
4.9 (4.8)
31.7-57.2*
34.2-53.1
36.0(18.6)
26.5
28.7 - 54.9
55.6 (23.8)
NR
18.4
31.0(20.0)
53.0
17(9.1)
21.8(8)
12.2(7.4)
16.4-28.0
24.3(12.2)
21.0
22.4*
Upper Range of
Concentration


Max: 53.4
Max: 41 .9

Max: 150.3
75th: 42.1
Max: 79.6
75th: 7.2
Max: 99.0


75th: 44.9
75th: 35.5
Max: 83.0




Max: 120.0
75th: 67.0
Max: 147.5
75th: 23.5
75th: 27.0
75th: 17.0

75th: 32.0
75th: 26.3
Max: 62.8
75th: 31 .0
'Median presented (information on mean not given). NR: Not reported
Draft - Do Not Cite or Quote
6-162
September 2011

-------
 1                   and dysrhythmia). Data on over 4 million ED visits from 31 hospitals were collected
 2                   from January 1993 to August 2000. Ozone was monitored from March to October. This
 3                   study was a re-analysis of a time series study conducted to assess the main effects of air
 4                   pollutants on cardiovascular ED visits in Atlanta (Tolbert et al.. 2007; Metzger et al..
 5                   2004). In the initial study, no evidence of associations was observed between O3 and all
 6                   CVD visits or visits for CVD subgroups, such as dysrhythmia, CHF, ischemic heart
 7                   disease (HD), and peripheral vascular and cerebrovascular disease. The relative risk for
 8                   all CVD visits was 1.01 (95% CI: 0.99, 1.02) for a 20 ppb increase in the 3-day moving
 9                   avg (lags 0-2 days) of 8-h O3 (Metzger et al., 2004). Similar to the initial investigation
10                   using a time-series analysis, no evidence of an association was observed for the O3 3-day
11                   moving average and CVD visits among the entire population using the case-crossover
12                   design (Peel et al.. 2007). However, the relationship between O3 and peripheral and
13                   cerebrovascular disease visits was stronger among patients with comorbid COPD (OR:
14                   1.19 [95% CI: 1.03-1.36] per 20 ppb, lag 0-2 days) as compared to patients without
15                   COPD (OR:  1.01 [95% CI: 0.97-1.04] per 20 ppb, lag 0-2 days). The same research
16                   group expanded upon the number of Atlanta hospitals providing ED visit data (41
17                   hospitals) as well as the length of the study period (1993-2004) (Tolbert et al.. 2007).
18                   Again, models assessing the health effects of O3 utilized data collected from March
19                   through October. Similar to the results presented by Metzger et al. (2004) and Peel et al.
20                   (2007) among the entire study population, no evidence of associations was observed for
21                   O3 and CVD visits (Tolbert et al.. 2007).

22                   A study of HAs for MI was performed using  a statewide registry from New Jersey
23                   between January 2004 and December 2006 (Rich et al.. 2010). Using a case-crossover
24                   design, the association between the previous 24 hr O3 concentration and transmural
25                   infarction (n=l,003) was examined. No association was observed (OR: 0.94 [95% CI:
26                   0.79, 1.13] per 20 ppb) and this did not change with the inclusion of PM25 in the model.

27                   Cakmak et al, (2006b) investigated the relationship between gaseous air pollutants and
28                   cardiac hospitalizations in 10 large Canadian cities using a time-series approach. A total
29                   of 316,234 hospital discharge records for primary diagnosis of congestive heart failure,
30                   ischemic heart disease, or dysrhythmia were obtained from April 1993 through March
31                   2000. Correlations between pollutants varied substantially across cities, which could
32                   partially explain discrepancies in effect estimates observed across the cities. In addition,
33                   pollutant lags differed across cities; the average lag for O3 was 2.9 days. The pooled
34                   effect estimate for a 20 ppb increase in the daily 1-h max O3 concentration and the
35                   percent change in hospitalizations among all  10 cities was 2.3 (95% CI: 0.11, 4.50) in an
36                   all-year analysis. The authors reported no evidence of effect modification by gender,
37                   neighborhood-level education, or neighborhood-level income. A similar multicity time-
38                   series study was conducted using  nearly 400,000 ED visits to  14 hospitals in seven
      Draft - Do Not Cite or Quote                       6-163                                September 2011

-------
 1                   Canadian cities from 1992 to 2003 (Stieb et al., 2009). Primary analyses considered daily
 2                   O3 single day lags of 0-2 days; in addition, sub-daily lags of 3-h avg concentrations up to
 3                   12 hours before presentation to the ED were considered. Seasonal variation was assessed
 4                   by stratifying analyses by warm and cold seasons. No evidence of effect of O3 on CVD
 5                   ED visits was observed. One negative, statistically significant association was reported
 6                   between a 1-day lag of O3 and visits for angina/myocardial infarction. Ozone was
 7                   negatively correlated with many of the other pollutants, particularly during the cold
 8                   season.

 9                   The effect of air pollution on daily ED visits for ischemic stroke (n=10,881 visits) in
10                   Edmonton, Canada was  assessed from April 1992 through March 2002 (Szyszkowicz.
11                   2008). A 26.4% (95% CI: 3.16-54.5) increase in stroke ED visits was associated with a
12                   20 ppb increase in O3 at lag 1  among men aged 20-64 years in the warm season. No
13                   associations were present among women or among men age 65 and older. In addition, no
14                   associations were observed for the cold season or for other lags (lag 0 or lag 2).  A similar
15                   investigation over the same time period in Edmonton, Canada, assessed the relationship
16                   between air pollutants and ED visits for stroke (ischemic stroke, hemorrhagic stroke, and
17                   transient ischemic attack) among those 65 years of age and older using a case-crossover
18                   framework (Villeneuve et al.. 2006a). Two-pollutant models were assessed. No  evidence
19                   of association was reported for O3 and stroke hospitalization (Villeneuve et al.. 2006a).

20                   Additional studies reported no evidence of an association between O3 concentrations and
21                   ED visits, hospitalizations, or symptoms leading to hospitalization (Symons et al.. 2006;
22                   Zanobetti and Schwartz. 2006; Wellenius et al., 2005). Symons et al. (2006) used a case-
23                   crossover framework to  assess the relationship between air pollutants and the onset of
24                   symptoms (dyspnea) severe enough to lead to hospitalization (through the  ED) for
25                   congestive heart failure. The study was conducted from April to December of 2002 in
26                   Baltimore, Maryland. Exposures were assigned using 3 index times: 8-h and 24-h periods
27                   prior to symptom onset and date of hospital admission. No evidence of association was
28                   reported for O3 concentrations. Although seasonal variation was not assessed, the time
29                   frame for the study did not involve an entire year (April to  December). Wellenius et al.
30                   (2005) investigated the association between air pollutants and congestive heart failure
31                   hospitalization among Medicare beneficiaries in Pittsburgh, Pennsylvania from  1987 to
32                   1999 utilizing a case-crossover framework. A total of 55,019 admissions from the
33                   emergency room with a  primary discharge diagnosis of CHF were collected. No evidence
34                   of an association was reported for O3 and CHF hospitalization (Wellenius et al., 2005).
35                   Finally, Zanobetti and Schwartz (2006) assessed the relationship between air pollutants
36                   and hospital admissions  through the  ED for myocardial infarction and pneumonia among
37                   patients aged 65 and older residing in the greater Boston area (1995-1999) using a case-
38                   crossover framework with control days in the same month matched on temperature.
      Draft - Do Not Cite or Quote                      6-164                                September 2011

-------
 1                   Pollution exposures were assigned for the same day and for the mean of the exposure the
 2                   day of and the day before the admission. Ozone was not associated with MI admissions in
 3                   all-year and seasonal analyses.

 4                   Several recent studies have examined the relationship between air pollution and CVD
 5                   hospital admissions and/or emergency department visits in Asia. Of note, some areas of
 6                   Asia have a more tropical climate than the U.S. and do not experience similar seasonal
 7                   changes. In Taiwan, fairly consistent positive associations have been reported for O3 and
 8                   congestive heart failure hospital admissions (for single- and copollutant models) in Taipei
 9                   on warm days (Yang. 2008) and in Kaohsiung (Lee et al.. 2007): cerebrovascular disease
10                   ED visits (for lag 0 single- and two-pollutant models but not other lags or 3-pollutant
11                   models) in Taipei (Chan et al.. 2006); and arrhythmia ED visits in Taipei among those
12                   without comorbid conditions (Chiu et al.. 2009: Lee et al.. 2008a) and in Taipei on warm
13                   days among those with and without comorbid conditions (Lee et al.. 2008a: Jansson et al..
14                   2001). However, one study in Taiwan did not shown an association. Bell  et al. (2008)
15                   reported no evidence of an O3 association with hospital admissions for ischemic heart
16                   disease or cerebrovascular disease. Three studies based in Asia but outside Taiwan were
17                   performed. First, a Hong Kong-based investigation (Wong et al.. 2009) reported no
18                   consistent evidence of a modifying effect of influenza on  the relationship between O3 and
19                   CVD admissions. Second, among elderly populations in Thailand, O3 was associated with
20                   CVD visits, but this association was not detected among younger age groups (15-64)
21                   (Buadong et al.. 2009). Third, a study performed in Seoul, Korea reported a positive
22                   association between O3 levels and HAs  for ischemic heart disease; the association was
23                   slightly greater among those over 64 years of age (Lee etal.. 2003b).

24                   Positive effects of O3 on CVD hospital admissions and/or ED  visits have  been reported in
25                   other areas of the world as well (Azevedo et al.. 2011: Linares and Diaz. 2010: Middleton
26                   et al..  2008: Turner et al.. 2007: Yallop  et al.. 2007: Ballester et al.. 2006: De Pablo et al..
27                   2006:  Von Klot et al.. 2005). although not consistently as some studies reported no
28                   association (Oudin et al.. 2010: Halonen et al.. 2009: Larrieu et al.. 2007: Barnett et al..
29                   2006:  Hinwood et al.. 2006: Lanki et al.. 2006: Hosseinpoor et al.. 2005:  Simpson et al..
30                   2005).

31                   A couple of studies (U.S. and Australia) have examined cardiac arrests where emergency
32                   services attempted treatment/resuscitation. No evidence of an  association between O3 and
33                   out-of-hospital cardiac arrest was observed (Dennekamp et al.. 2010: Silverman et al..
34                   2010).

3 5                   An increasing number of air pollution studies have investigated the relationship between
36                   O3 concentrations and CVD hospital admissions and/or ED visits. As summarized in the
37                   2006 O3 AQCD, some, especially those reporting results stratified by season (or
      Draft - Do Not Cite or Quote                      6-165                                September 2011

-------
1
2
3
4
5
6
7
temperature) or comorbid conditions have reported positive associations. However, even
studies performing these stratified analyses are not consistent and the overall evidence
remains inconclusive regarding the effects of O3 on CVD HAs and ED visits. These HA
and ED visit studies are summarized in Figures 6-22 through 6-26, which depict the
associations for studies in which numerical associations were presented for an overall
study population. Tables 6-34 through 6-38 provide the numerical results displayed in the
figures.
Reference
Buadongetal. (2009)
Middleton et al. (2008)
Fungetal. (2005)
Ballesteretal. (2001)
Petroeschevskyetal. (2001)
Linn et al. (2000)
Atkinson etal. (1999)
Wongetal. (1999a)
Wong etal. (1999b)
Prescottetal. (1998)
Polonieckietal. (1997)
Halonen et al. (2009)
Larrieu et al. (2007)
Peel etal. (2007)
Ballesteretal. (2006)
Chang et al. (2005)
Yang et al. (2004)
Wongetal. (1999b)
Chang et al. (2005)
Yang et al. (2004)
Wongetal. (1999a)
Wongetal. (1999b)
Cakmak et al. (2006)
Ballesteretal. (2001)
Morgan etal. (1998)
Larrieu et al. (2007)
Ballesteretal. (2006)
von Klot et al. (2005)
Bell et al. (2008)
Chan et al. (2006)
Ballesteretal. (2001)
Wongetal. (1999a)
Wongetal. (1999b)
Polonieckietal. (1997)
Peel etal. (2007)
Wongetal. (1999b)
Wongetal. (1999b)
Location
Bangkok, Thailand
Nicosia, Cyprus

Brisbane, Australia 	 • 	
Los Angeles, CA •
London, England
Hong Kong
Hong Kong
London, England 0
8 French cities -t
Atlanta, GA -4
14 Spanish cities
Taipei, Taiwan
Kaohsiung, Taiwan
Taipei, Taiwan
Kaohsiung, Taiwan 	
Hong Kong
Hong Kong
10 Canadian cities
Sydney, Australia
8 French cities —
14 Spanish cities
5 European cities
Taipei, Taiwan

Hong Kong 9
London, England 0
Atlanta, GA —
Hong Kong 	
•




»-
K

— o 	

-•-
•-

•

-•-
-o 	
Cardiovascular
disease



Cardiac disease
Cerebrovascular
disease
                                    0.7
                        0.8      0.9
1       1.1
1.2
1.3      1.4      1.5
       Note: Increase in O3 standardized to 20 ppb for 24-h avg period, 30 ppb for 8-h avg period, and 40 ppb for 1-h avg period. Ozone
     concentrations in ppb. Seasons depicted by colors - black: all year; red: warm season; light blue: cold season. Age groups of study
     populations were not specified or were adults with the exception of Fung et al. (2005). Wong et al. (1999b). and Prescott et al.
     (1998). which included only individuals aged 65+.

     Figure 6-22    Odds  ratio (95% Cl) per increment ppb increase in ozone for over
                      all cardiovascular ED visits or HAs.
     Draft - Do Not Cite or Quote
                                6-166
                             September 2011

-------
Table 6-34      Odds ratio (95% Cl) per increment ppb increase in  ozone for overall
                   cardiovascular  ED visits or HAs in studies presented in Figure 6-22
Study
Atkinson et al. (2006a)
Ballesteretal. (2006)
Ballesteretal. (2006)
Bell et al. (2008)
Buadong et al. (2009)
Cakmaketal. (2006a)
Chan etal. (2006)
Chang et al. (2005)
Funa etal. (2006a)
Halonen et al. (2009)
Larrieu et al. (2007)
Linn et al. (2006a)
Middleton etal. (2008)
Morgan et al. (2008)
Peel et al. (2007)
Petroeschevskv etal. (2001)
Poloniecki et al. (2006a)
Prescott etal. (1998)
Von Klot etal. (2005)
Wong et al. (1999b)
Wong et al. (1999a)
Yang et al. (2005)
Location
London, England
Multicity, Spain
Valencia, Spain
Taipei, Taiwan
Bangkok, Thailand
Multicity, Canada
Taipei, Taiwan
Taipei, Taiwan
Wndsor, Canada
Helsinki, Finland
Multicity France
Los Angeles, California
Nicosia, Cyprus
Sydney, Australia
Atlanta, GA
Brisbane, Australia
London, England
Edinburgh, Scotland
Multicity, Europe
Hong Kong
Hong Kong
Kaohsiung, Taiwan
Outcome
Cardiovascular disease
Cardiovascular disease
Cardiac disease
Cardiovascular disease
Cardiac disease
Cerebrovascular disease
Cerebrovascular disease
Cardiovascular disease
Cardiac disease
Cerebrovascular disease
Cardiovascular disease
Cardiovascular disease
Cardiovascular disease
Cardiac disease
Cardiovascular disease
Cardiovascular disease
Cardiac disease
Cardiovascular disease
Cerebrovascular disease
Cardiovascular disease
Cardiovascular disease
Cerebrovascular disease
Cardiovascular disease
Cardiac disease
Cardiovascular disease
Cerebrovascular disease
Cardiovascular disease
Cerebrovascular disease
Cardiovascular disease
Averaging Time
8-h
8-h warm season
8-h warm season
8-h
8-h
8-h
24-h
1-h
1-h max
1-h max
24-h warm season
24-h cold season
1-h
8-h max warm season
8-h max warm season
24-h
8-h max
1-h max
8-h warm season
8-h warm season
8-h
8-h
8-h
24-h
8-h max warm season
24-h
24-h cold season
24-h
24-h
24-h warm season
24-h cold season
24-h
24-h warm season
24-h cold season
24-h warm season
24-h cold season
Standardized Estimate (95% Cl)
1.03(1.00,1.05)
1.04(1.02,1.06)
1.04(1.01,1.07)
0.94(0.84,1.06)
0.88(0.75,1.03)
0.86(0.72,1.04)
0.94(0.87,1.02)
1.01 (1.00, 1.02)
1.02(1.00,1.04)
1.02(1.01,1.03)
1.42(1.33,1.50)
1.15(1.04,1.27)
1.02(0.92,1.13)
1.05(0.96,1.14)
1.01 (0.98, 1.04)
0.99(0.98,1.00)
1.09(1.00,1.18)
1.02(0.99,1.05)
1.00(0.98,1.02)
1.02(0.98,1.05)
0.96(0.92,1.01)
0.97(0.93,1.01)
0.98(0.95,1.02)
0.89(0.78,1.00)
1.11 (1.00, 1.22)
1.08(1.03,1.13)
1.15(1.04,1.26)
0.95(0.90,1.01)
1.02(1.03,1.06)
1.01 (0.96, 1.06)
1.06(1.02,1.11)
0.99(0.95,1.04)
0.98(0.90,1.08)
1.02(0.96,1.10)
1.33(1.26,1.40)
1.05(0.96,1.15)
  Note: Increase in 03 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h averaging period.
Ozone concentrations in ppb. Age groups of study populations were not specified or were adults with the exception of Fung et al. (2006a), Wong et
al. (1999a), and Prescott et al. (1998). which included only individuals aged 65+.
  Warm season defined as: March-October (Peel etal.. 2007). May-October (Ballesteretal.. 2005: Wong etal.. 1999a). May-September (Halonen
etal.. 2009). April-September (Larrieu etal.. 2007: Von Klot et al.. 2005). > 20°C (Chang etal.. 2005) and > 25°C (Yang etal.. 2004). Cold season
defined as: November-April (Wong etal.. 1999a). <20°C (Chang etal.. 2005) and <25°C (Yang etal.. 2004). December-March (Wong etal.. 1999b)
Draft - Do Not Cite or Quote
6-167
September 2011

-------
  Reference

  Stiebetal. (2009)

  Welleniusetal. (2005)

  Wongetal. (1999a)

  Wongetal. (1999b)

  Poloniecki et al. (1997)

  Yang (2008)

  Peel et al. (2007)

  Lee et al. (2007)

  Symonset al. (2006)

  Wongetal. (1999b)

  Yang (2008)

  Lee et al. (2007)

  Wongetal. (1999b)
Location

7 Canadian cities

Allegheny county, PA

Hong Kong

Hong Kong        —

London, England

Taipei, Taiwan

Atlanta, GA

Kaohsiung, Taiwan

Baltimore, MD    —

Hong Kong

Taipei, Taiwan

Kaohsiung, Taiwan

Hong Kong
                                        0.4
                      0.6
0.8
1
1.2
1.4
1.6
1.8
  Note: Increase in O3 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h
averaging period. Ozone concentrations in ppb. Seasons depicted by colors: black: all year; red: warm season; light blue: cold
season. Outcomes were all congestive heart failure, with the exception of Symons et al. (2006). which examined onset of congestive
heart failure symptoms leading to a heart attack. Age groups of study populations were not specified or were adults with the
exception of Wellenius et al. (2005) and Wong et al. (1999a). which included only individuals aged 65+.


Figure 6-23   Odds  Ratio (95% Cl) per  increment ppb increase in ozone for

                  congestive heart failure ED visits or HAs.
Draft - Do Not Cite or Quote
                         6-168
                                   September 2011

-------
Table 6-35      Odds Ratio (95% Cl) per increment ppb increase in ozone for
                   congestive heart failure ED visits or HAs for studies in Figure 6-23
Study
Lee et al. (2007)
Peel et al. (2007)
Poloniecki
etal. (1997)
Stieb et al. (2009)
Symons et al. (2006)
Wellenius et al. (2005)
Wong etal. (1999a)
Yang (2008)
Wonaetal. (1999b)
Location
Kaohsiung, Taiwan
Atlanta, GA
London, England
Multicity, Canada
Baltimore, MD
Allegheny county, PA
Hong Kong
Taipei, Taiwan
Hong Kong
Outcome
congestive heart failure
congestive heart failure
congestive heart failure
congestive heart failure
congestive heart failure
onset of congestive heart failure
symptoms leading to heart attack
congestive heart failure
congestive heart failure
congestive heart failure
congestive heart failure
congestive heart failure
Averaging Time
24-h warm season
24-h cold season
8-h warm season
8-h
24-h
8-h warm season
24-h
24-h
24-h warm season
24-hcold season
24-h warm season
24-h cold season
24-h
Standardized
Estimate (95% Cl)
1.25(1.15,1.36)
1.24(1.09,1.41)
0.96(0.93,1.00)
0.99(0.95,1.03)
1.03(0.98,1.07)
0.83(0.49,1.41)
0.98(0.96,1.01)
1.11 (1.04,1.80)
1.09(0.96,1.23)
1.16(1.06,1.27)
1.39(1.27,1.51)
0.61 (0.52, 0.73)
1.25(1.11,1.41)
  Note: Increase in 03 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h averaging
period. Ozone concentrations in ppb. Outcomes were all congestive heart failure, with the exception of Symons et al. (2006). which
examined onset of congestive heart failure symptoms leading to a heart attack. Age groups of study populations were not specified or were
adults with the exception of Wellenius et al. (2005) and Wong et al. (1999a). which included only individuals aged 65+.
  Warm season defined as: March-October (Peel etal.. 2007). April-November (Svmons etal.. 2006). May-October (Wonaetal.. 1999a)
> 20°C (Yang. 2008). and >25°C (Lee etal.. 2007). Cold season defined as: November-April (Wong etal.. 1999a). <20°C (Yang. 2008). and
<25°C (Lee etal..2007).
Draft - Do Not Cite or Quote
6-169
September 2011

-------
  Reference

  Buadong et al. (2009)
  Bell et al. (2008)
  Lee et al. (2003)
  Atkinson etal. (1999)
  Wongetal. (1999a)
  Wong etal. (1999b)
  Larrieu et al. (2007)
  Peel et al. (2007)
  Lee et al. (2003)
  Wongetal. (1999b)
  Wongetal. (1999b)
Location

Bangkok, Thailand
Taipei, Taiwan
Seoul, Korea
London, England
Hong Kong
Hong Kong
8 French cities
Atlanta, GA
Seoul, Korea
Hong Kong
Hong Kong
  Halonen et al. (2009)     Helsinki, Finland
  Rich etal. (2010)
  Buadong et al. (2009)
  Stiebetal. (2009)
  Zanobetti et al. (2006)
  Poloniecki et al. (1997)
  Lanki et al. (2006)
  von Klot et al. (2005)

  Hosseinpoor et al. (2005)
  Poloniecki et al. (1997)
  von Klot et al. (2005)
New Jersey
Bangkok, Thailand
7 Canadian cities
Boston, MA
London, England
5 European cities
5 European cities

Tehran, Iran
London, England
5 European cities
                      0.5
               0.7
0.9
                                Ischemia heart disease
                                                              Coronary heart disease

                                                              Myocardial infarction
                                Angina pectoris
1.1
1.3
1.5
  Note: Increase in O3 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h
averaging period. Ozone concentrations in ppb. Seasons depicted by colors: black: all year; red: warm season; light blue: cold
season. Age groups of study populations were not specified or were adults with the exception of Wong et al. (1999a) and Atkinson et
al. (2006a), which included only individuals aged 65+.


Figure 6-24    Odds Ratio (95% confidence interval) per increment  ppb increase in
                  ozone for ischemic heart disease, coronary heart disease,
                  myocardial infarction,  and angina pectoris ED visits  or HAs.
Draft - Do Not Cite or Quote
                          6-170
                                    September 2011

-------
Table 6-36      Odds Ratio (95% Cl) per increment ppb increase in ozone for
                  ischemic heart disease, coronary heart disease,  myocardial
                  infarction, and angina pectoris ED visits or HAs for studies
                  presented in Figure 6-24
Study
Atkinson et al. (1999)
Bell et al. (2008)
Buadong et al. (2009)
Halonenetal. (2009)
Hosseinpoor et al. (2005)
Lankietal. (2006)
Larrieuetal. (2007)
Lee et al. (2003b)
Peel et al. (2007)
Polonieckietal. (1997)
Rich et al. (Rich etal., 2010)
Stieb et al. (2009)
Von Klot etal. (2005)
Wong etal. (2009)
Wong etal. (2008)
Zanobetti and Schwartz (2006)
Location
London, England
Taipei, Taiwan
Bangkok, Thailand
Helsinki, Finland
Tehran, Iran
Multicity, Europe
Multicity France
Seoul, Korea
Atlanta, GA
London, England
New Jersey
Multicity, Canada
Multicity, Europe
Hong Kong
Hong Kong
Boston, MA
Outcome
Ischemic heart disease
Ischemic heart disease
Ischemic heart disease
Myocardial infarction
Coronary heart disease
Angina
Myocardial infarction
Ischemic heart disease
Ischemic heart disease
Ischemic heart disease
Ischemic heart disease
Myocardial infarction
Angina
Myocardial infarction
Myocardial infarction
Myocardial infarction
Angina
Ischemic heart disease
Ischemic heart disease
Myocardial infarction
Averaging Time
8-h
24-h
1-h
1-h
8-h max warm season
8-h max
8-h max warm season
8-h max warm season
1-h max
1-h max warm season
8-h warm season
8-h
8-h
24-h
2-h
8-h max warm season
8-h max warm season
24-h
24-h warm season
24-h cold season
24-h
24-h
Standardized
Estimate (95% Cl)
0.97(0.94,1.01)
1.01(0.91,1.12)
1.00(0.98,1.02)
0.97(0.94,1.01)
0.99 (0.79, 1 .25)
0.80 (0.70, 0.92)
0.96(0.92,1.01)
1 .02 (0.98, 1 .07)
1.07(1.02,1.13)
1.07(1.00,1.17)
1 .00 (0.97, 1 .03)
0.98(0.94,1.02)
0.98 (0.94, 1 .03)
0.94(0.79,1.13)
1 .00 (0.96, 1 .04)
1.00(0.83,1.21)
1.19(1.05,1.35)
1.01(0.94,1.06)
1.02(0.94,1.11)
1 .02 (0.95, 1 .09)
1 .03 (0.98, 1 .08)
0.98(0.92,1.03)
  Note: Increase in 03 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h averaging period.
Ozone concentrations in ppb. Age groups of study populations were not specified or were adults with the exception of Wong et al. (1999a) and
Atkinson et al. (2006a). which included only individuals aged 65+.
  Warm season defined as: March-October (Peel etal.. 2007). June-August (Lee etal.. 2003b). May-September (Halonenetal.. 2009). May-
October (Buadong et al.. 2009). and April-September (Larrieu etal.. 2007: Lanki et al.. 2006: Von Klot etal.. 2005). Cold season defined as:
November-April (Buadong etal.. 2009)
Draft - Do Not Cite or Quote
6-171
September 2011

-------
  Reference

  Chan et al. (2006)
  Halonen et al. (2009)
  Larrieu et al. (2007)

  Chan et al. (2006)
  Villeneuve et al. (2006)
  Villeneuve et al. (2006)
  Villeneuve et al. (2006)

  Chan et al. (2006)
  Villeneuve et al. (2006)
  Villeneuve et al. (2006)
  Villeneuve et al. (2006)

  Villeneuve et al. (2006)
  Villeneuve et al. (2006)
  Villeneuve et al. (2006)
Location

Taipei, Taiwan
Helsinki, Finland
8 French cities

Taipei, Taiwan
Edmonton, Canada
Edmonton, Canada
Edmonton, Canada

Taipei, Taiwan
Edmonton, Canada
Edmonton, Canada
Edmonton, Canada

Edmonton, Canada
Edmonton, Canada
Edmonton, Canada
                                All
                                Ischemia
                                Hemorrhagic
     -O-
                                Transient
                                ischemic
               -O
                            0.5
                  0.7
0.9
1.1
1.3
1.5
  Note: Increase in O3 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h
averaging period. Ozone concentrations in ppb. Seasons depicted by colors: black: all year; red: warm season; light blue: cold
season. Age groups of study populations were  not specified or were adults with the exception of Villeneuve et al. (2006a), which
included only individuals aged 65+, and Chan et al. (2006). which included only individuals aged 50+.

Figure 6-25    Odds Ratio  (95% confidence  interval) per increment ppb increase in
                  ozone for stroke ED visits or HAs.
Draft - Do Not Cite or Quote
                          6-172
                                   September 2011

-------
Table 6-37     Odds Ratio (95% Cl) per increment ppb increase in ozone for stroke
                  ED visits or HAs for studies presented in Figure 6-25
Study
Chan etal. (2006)
Halonenetal. (2009)
Larrieu etal. (2007)
Villeneuveetal. (2006a)
Location
Taipei, Taiwan
Helsinki, Finland
Multicity, France
Edmonton,
Canada
Outcome
All/non-specified stoke
Ischemia stroke
Hemorrhagic stroke
All/non-specified stoke
All/non-specified stoke
Ischemic stroke
Ischemic stroke
Ischemic stroke
Hemorrhagic stroke
Hemorrhagic stroke
Hemorrhagic stroke
Transient ischemic stroke
Transient ischemic stroke
Transient ischemic stroke
Averaging Time
1-hmax
1-h max
1-h max
8-h max warm season
8-h max warm season
24-h
24-h warm season
24-h cold season
24-h
24-h warm season
24-h cold season
24-h
24-h warm season
24-h cold season
Standardized Estimate
(95% Cl)
1.01(0.99,1.03)
1.03(0.99,1.07)
0.99(0.92,1.06)
1.08(0.83,1.41)
0.98 (0.93 , 1 .02)
1.00(0.88,1.13)
1.09(0.91,1.32)
0.98(0.80,1.18)
1 .02 (0.87, 1 .20)
1.12(0.88,1.43)
0.97 (0.76, 1 .22)
0.98(0.87,1.10)
0.85(0.70,1.01)
1.11 (0.93,1.32)
  Note: Increase in 03 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h averaging period.
Ozone concentrations in ppb. Age groups of study populations were not specified or were adults with the exception ofVilleneuve etal. (2006a).
which included only individuals aged 65+, and Chan et al. (2006). which included only individuals aged 50+.
  Warm season defined as: May-September (Halonenetal., 2009), and April-September (Larrieu etal., 2007: Villeneuve etal., 2006a). Cold
season defined as: October-March (Villeneuve et al.. 2006a).
      Reference

      Stieb et al. (2009)
      Peel et al. (2007)
                  Location

                  7 Canadian cities
                  Atlanta, GA
      Buadong et al. (2009)  Bangkok, Thailand
                        Hong Kong
                        London, England
Wong etal. (1999b)
Polonieckiet al.
    (1997)
Halonen et al. (2009)  Helsinki, Finland
     Wong etal. (1999b)
     Wong etal. (1999b)
                  Hong Kong
                  Hong Kong
                          0.7
                               0.8
0.9
                                    Dysrhythmia
                                                                                   Arrhythmia
1.1        1.2        1.3        1.4
  Note: Increase in O3 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h
averaging period. Ozone concentrations in ppb. Seasons depicted by colors: black: all year; red: warm season; light blue: cold
season. Age groups of study populations were not specified or were adults with the exception of Wong et al. (1999a). which included
only individuals aged 65+.

Figure 6-26    Odds Ratio (95% confidence interval) per increment ppb* increase
                  in ozone for arrhythmia  and dysrhythmia ED visits or HAs.
Draft - Do Not Cite or Quote
                                            6-173
                                           September 2011

-------
      Table 6-38     Odds Ratio (95% Cl) per increment ppb* increase in ozone for
                       arrhythmia and dysrhythmia ED visits or HAs for studies presented
                       in Figure 6-26
               Study               Location         Outcome     Averaging Time
Buadonq et al. (2009)
Halonenetal. (2009)
Peel et al. (2007)
Polonieckietal. (2009)
Stieb et al. (2009)
Wong etal. (2009)
Bangkok, Thailand
Helsinki, Finland
Atlanta, GA
London, England
Multicity, Canada
Hong Kong
Arrhythmia
Arrhythmia
Dysrhythmia
Arrhythmia
Dysrhythmia
Arrhythmia
1-h
8-h max warm season
8-h warm season
8-h
24-h
24-h
24-h warm season
24-h cold season
0.99 (0.95,
1 .04 (0.80,
1 .01 (0.98,
1 .02 (0.96,
1 .02 (0.95,
1 .06 (0.99,
1.10(0.96,
1.11 (1.01,
1.04)
1.35)
1.04)
1.07)
1.09)
1.12)
1.26)
1.23)
        Note: Increase in 03 standardized to 20 ppb for 24-h averaging period, 30 ppb for 8-h averaging period, and 40 ppb for 1-h averaging period.
      Ozone concentrations in ppb. Age groups of study populations were not specified or were adults with the exception of (Wong etal., 1999a), which
      included only individuals aged 65+. Warm season defined as: March-October (Peel etal..2007). May-October (Wong etal.. 1999a) and May-
      September (Halonenetal..2009). Cold season defined as: November-April (Wong etal.. 1999a).
                     6.3.2.8    Cardiovascular Mortality

 1                   As discussed within this section (Section 6.3), epidemiologic studies provide inconsistent
 2                   evidence of an association between short-term O3 exposure and cardiovascular effects.
 3                   However, toxicological studies have demonstrated O3-induced cardiovascular effects,
 4                   specifically enhanced atherosclerosis and ischemia, which could lead to death. The 2006
 5                   O3 AQCD

 6                   provided evidence, primarily from single-city studies, of consistent positive associations
 7                   between short-term O3 exposure and cardiovascular mortality. Recent multicity studies
 8                   conducted in the U.S., Canada, and Europe further confirm the association between short-
 9                   term O3 exposure and cardiovascular mortality.

10                   As discussed in Section 6.2.7.2, the APHENA study (Katsouyanni et al.. 2009) also
11                   examined associations between short-term O3 exposure and mortality and found
12                   consistent positive associations for cardiovascular mortality in all-year analyses with
13                   associations persisting in analyses restricted to the summer season. Additional multicity
14                   studies from the U.S.  (Zanobetti and Schwartz. 2008b). Europe (Samoli et al.. 2009).
15                   Italy (Stafoggia et al., 2010). and Asia (Wong et al.. 2010) that conducted summer season
16                   and/or all-year analyses provide additional support for an association between short-term
17                   O3 exposure and cardiovascular mortality (Figure 6-37).

18                   Of the studies evaluated,  only the APHENA study (Katsouyanni et al.. 2009) and the
19                   Italian multicity study (Stafoggia et al.. 2010) conducted an analysis of the potential for
20                   copollutant confounding of the O3-cardiovascular mortality relationship. In the European


      Draft - Do Not Cite or Quote                       6-174                                 September 2011

-------
 1                   dataset, when focusing on the natural spline model with 8 df/year (Section 6.2.7.2) and
 2                   lag 1 results in order to compare results across study locations (Section 6.6.2.1),
 3                   cardiovascular mortality risk estimates were robust to the inclusion of PM10 in
 4                   copollutant models in all-year analyses with more variability in the Canadian and U.S.
 5                   datasets (i.e., cardiovascular O3 mortality risk estimates were reduced or increased in
 6                   copollutant models). In summer season analyses, cardiovascular O3 mortality risk
 7                   estimates were robust in the European dataset and attenuated but remained positive in the
 8                   U.S. dataset. Similarly, in the Italian multicity study (Stafoggia et al.. 2010). which was
 9                   limited to the summer season, cardiovascular mortality risk estimates were robust to the
10                   inclusion of PMi0 in copollutant models. Based on the APHENA and Italian multicity
11                   results, O3 cardiovascular mortality risk estimates appear to be robust to inclusion of
12                   PMio in copollutant models. However, in the U.S. and Canadian datasets there was
13                   evidence that O3 cardiovascular mortality risk estimates are moderately to substantially
14                   sensitive (e.g., increased or attenuated) to PMi0. The mostly every-6th-day sampling
15                   schedule for PM10 in the Canadian and U.S. datasets greatly reduced their sample size
16                   and limits the interpretation of these results.
                     6.3.2.9    Summary of Epidemiologic Studies

17                   Overall, the available body of evidence examining the relationship between short-term
18                   exposures to O3 and cardiovascular morbidity is inconsistent. Differences in exposure
19                   metrics and windows of exposure, a wide variety of biomarkers considered, and a lack of
20                   consistency among definitions used for specific cardiovascular disease endpoints (e.g.
21                   arrhythmias, HRV) make comparisons across studies difficult. In addition, several
22                   investigators reporting associations between O3 and cardiovascular morbidity postulate
23                   that O3 may be acting as a proxy for sulfate; differences reported across multicity  studies
24                   and across studies conducted in specific cities/regions point to the importance of
25                   considering multipollutant relationships that vary across geographic regions. Additionally
26                   mortality studies indicate a consistent positive association between O3 and cardiovascular
27                   mortality.
             6.3.3   Toxicology
                     6.3.3.1    Summary of Findings from Previous Ozone AQCDs

28                   In the previous O3 AQCDs (U.S. EPA. 2006b. 1996a) experimental animal studies have
29                   reported relatively few cardiovascular system alterations after exposure to O3 and other
      Draft - Do Not Cite or Quote                      6-175                                September 2011

-------
 1                   photochemical oxidants. The limited amount of research directed at examining O3-
 2                   induced cardiovascular effects has primarily found alterations in heart rate (HR) and BP
 3                   after O3 exposure. A group of studies (Arito et al.. 1992; Arito etal.. 1990; Uchiyama and
 4                   Yokovama. 1989; Yokovama et al.. 1989; Uchivama et al.. 1986) report Q3 (0.1-1.0 ppm)
 5                   exposure in rats decreased core temperature (Tco), HR, and mean arterial pressure
 6                   (MAP). However, these cardiovascular responses to O3 could be attenuated by increased
 7                   ambient temperatures, exhibited adaptation, and were the result of the rodent
 8                   hypothermic response (Watkinson et al.. 2003; Watkinson et al.. 1993). This hypothermic
 9                   response could be an attempt to minimize the irritant effects of O3 inhalation, serving as a
10                   physiological and behavioral defense mechanism (Twasaki etal.. 1998; Arito etal.. 1997).
11                   As humans do not appear to exhibit decreased HR, MAP, and Tco with routine
12                   environmental exposures to O3, caution must be used in extrapolating the results of these
13                   animal studies to  humans (Section 6.3.1).

14                   Other studies have shown that O3 can increase BP in animal models. Rats exposed to
15                   0.6 ppm O3 for 33 days had increased systolic pressure and HR (Revis etal.. 1981).
16                   Increased BP triggers the release of atrial natriuretic factor (ANF), which has been found
17                   in increased levels in the heart, lungs, and circulation of O3 exposed (0.5 ppm) rats
18                   (Vesely et al.. 1994a. b, c). High concentration O3 exposure (1.0 ppm) has also been
19                   found to lead to heart and lung edema (Friedman et al.. 1983). which could be the result
20                   of increased ANF levels. Thus, O3 may increase blood pressure and HR, leading to
21                   increased ANF and tissue edema.

22                   The toxicological studies that have examined the effect of O3 on the cardiovascular
23                   system demonstrate O3-induced responses, but it remains unclear if the mechanism is
24                   through a reflex response or due to O3 reaction products, which have been sparsely
25                   studied. Oxysterols derived from cholesterol ozonation, such as (3-epoxide and 5(3,6(3-
26                   epoxycholesterol  (and its metabolite cholestan-6-oxo-3,5-diol), have been implicated in
27                   inflammation associated with cardiovascular disease (Pulfer et al.. 2005; Pulfer and
28                   Murphy. 2004). Two other cholesterol ozonolysis products, atheronal-A and -B (e.g.
29                   cholesterol secoaldehyde), have been found in human atherosclerotic plaques and shown
30                   in vitro to induce  foam cell formation and induce cardiomyocyte apoptosis and necrosis
31                   (Sathishkumar et  al.. 2005; Wentworth et al.. 2003); however, these products have not
32                   been found in the lung compartment or systemically after O3 exposure. The ability to
33                   form these cholesterol ozonation products in the circulation in the absence of O3 exposure
34                   complicates their implication in O3 induced cardiovascular disease.

3 5                   Although it has been proposed that O3 reaction products released after the interaction of
36                   O3 with ELF constituents (See Section 5.1.2 on O3 interaction with ELF) are responsible
37                   for systemic effects,  it is not known whether they gain access to the vascular space.
      Draft - Do Not Cite or Quote                      6-176                                September 2011

-------
 1                   Alternatively, extrapulmonary release of diffusible mediators, such as cytokines or
 2                   endothelins, may initiate or propagate inflammatory responses in the vascular or systemic
 3                   compartments (Cole and Freeman. 2009) (Section 5.1.9.1). Ozone reacts within the lung
 4                   to amplify ROS production, induce pulmonary inflammation, and activate inflammatory
 5                   cells, resulting in a cascading proinflammatory state and extrapulmonary release of
 6                   diffusible mediators that could lead to cardiovascular injury.
                     6.3.3.2    Recent Cardiovascular Toxicology Studies

 7                   According to recent short-term O3 exposure animal toxicology studies, O3 plays a role in
 8                   inducing vascular oxidative stress and proinflammatory mediators, altering HR and HRV,
 9                   and regulating the pulmonary endothelin system (study details are provided in Table 6-
10                   39). A number of these effects were variable between strains examined, suggesting a
11                   genetic component to development of O3 induced cardiovascular effects. Further, new
12                   studies provide evidence that extended O3 exposure enhances susceptibility to ischemia-
13                   reperfusion (I/R) injury and atherosclerotic lesion development. Still, few studies have
14                   investigated the role of O3 reaction products in these processes, but more evidence is
15                   provided for elevated inflammatory and reduction-oxidation (redox) cascades known to
16                   initiate these cardiovascular pathologies.

17                   A recent study in young mice and rhesus monkeys examined the effects of short-term O3
18                   exposure on a number of cardiovascular endpoints (Chuang et al.. 2009). Mice exposed to
19                   O3 for 5 days had increased HR as well as mean and diastolic blood pressure. Increased
20                   blood pressure could be explained by the inhibition in endothelial-dependent
21                   (acetylcholine) vasorelaxation from decreased bioavailability of aortic nitric oxide (-NO).
22                   Ozone caused a decrease in aortic NOX (nitrite and nitrate levels) and a decrease in total,
23                   but not phosphorylated, endothelial nitric oxide synthase (eNOS). Ozone also increased
24                   vascular oxidative stress in the form of increased aortic and lung lipid peroxidation (F2-
25                   isoprostane), increased aortic protein nitration (3-nitrotyrosine), decreased aortic
26                   superoxide dismutase (SOD2)  protein and activity, and decreased aortic aconitase
27                   activity, indicating  specific inactivation by O2~ and ONOO". Mitochondrial DNA
28                   (mtDNA) damage was also used as a measure of oxidative and nitrative stress in mice
29                   and infant rhesus monkeys exposed to O3. Chuang et al. (2009) observed that MtDNA
30                   damage accumulated in the lung and aorta of mice after 1 and 5 days of O3 exposure and
31                   in the proximal and distal aorta of O3 treated nonhuman primates. Additionally,
32                   genetically hyperlipidemic mice  exposed to O3 for 8 weeks had increased aortic
33                   atherosclerotic lesion area (Section 7.3.1), which may be associated with the short-term
34                   exposure changes discussed. Overall, this study suggests that O3 initiates an oxidative
35                   environment by increasing O2~ production, which leads to mtDNA damage and -NO


      Draft - Do Not Cite or Quote                       6-177                                September 2011

-------
 1                   consumption, known to perturb endothelial function (Chuang et al., 2009). Endothelial
 2                   dysfunction is characteristic of early and advanced atherosclerosis and coincides with
 3                   impaired vasodilation and blood pressure regulation.

 4                   Vascular occlusion resulting from atherosclerosis can block blood flow causing ischemia.
 5                   The restoration of blood flow in the vessel or reperfusion can cause injury to the tissue
 6                   from subsequent inflammation and oxidative damage. Perepu et al. (2010) observed that
 7                   O3 exposure enhanced the sensitivity to myocardial I/R injury in rats while increasing
 8                   oxidative stress levels and pro-inflammatory mediators and decreasing production of anti-
 9                   inflammatory proteins. Ozone was also found to decrease the left ventricular developed
10                   pressure, rate of change of pressure development, and rate of change of pressure decay
11                   while increasing left ventricular end diastolic pressure in isolated perfused hearts. In this
12                   ex vivo heart model, O3 induced oxidative stress by decreasing SOD enzyme activity and
13                   increasing malondialdehyde levels. Ozone also elicited a proinflammatory state which
14                   was evident by an increase in TNF-a and a decrease in the anti-inflammatory cytokine
15                   IL-10. Perepu et al. (2010) concluded that O3 exposure may result in a greater I/R injury.


                     Heart Rate and Heart  Rate Variability

16                   Strain differences in HR and HRV have been observed in response to a 2-h O3
17                   pretreatment followed by exposure to carbon black (CB) in mice (C3H/HeJ [HeJ],
18                   C57BL/6J [B6], and C3H/HeOuJ [OuJ]) (Hamade and Tankerslev. 2009: Hamade et al..
19                   2008). These mice strains were chosen from prior studies on lung inflammatory and
20                   hyperpermeability responses to be susceptible (B6 and OuJ) and resistant (HeJ) to O3-
21                   induced health effects (Kleeberger et al.. 2000). HR decreased during O3 pre-exposure for
22                   all strains, but recovered during the CB exposure (Hamade et al..  2008). This is contrary
23                   to the tachycardia that was reported in 6-week-old B6 mice treated on 1 or 5  days with
24                   O3, as described above (Chuang et al.. 2009). Percent change in HRV parameters, SDNN
25                   (indicating total HRV) and rMSSD (indicating beat-to-beat HRV), were increased in both
26                   C3H mice strains, but not B6  mice, during O3 pre-exposure and recovered during CB
27                   exposure when compared to the filtered air group. The two C3H strains differ by a
28                   mutation in the Toll-like receptor 4 (TLR4) gene, but these effects did not seem to be
29                   related to this mutation since similar responses were observed. Hamade et al. (2008)
30                   speculate that the B6 and C3H strains differ in mechanisms of HR response after O3
31                   exposure between withdrawal of sympathetic tone and increase of parasympathetic tone;
32                   however, no direct evidence for this conclusion was reported. The strain differences
33                   observed in HR and HRV suggest that genetic variability affects cardiac responses after
34                   acute air pollutant exposures.
      Draft - Do Not Cite or Quote                      6-178                                September 2011

-------
 1                   Hamade and Tankersley (2009) continued this investigation of gene-environment
 2                   interactions on cardiopulmonary adaptation of O3 and CB induced changes in HR and
 3                   HRV using the previously described (Hamade et al., 2008) daily exposure scheme for 3
 4                   consecutive days. By comparing day-1 interim values it is possible to observe that O3
 5                   exposure increased SDNN and rMSSD, but decreased HR in all strains. Measures of HR
 6                   and HRV in B6 and HeJ mice recovered to levels consistent with filtered air treated mice
 7                   by day 3; however, these responses in OuJ mice remained suppressed. B6 mice had no
 8                   change in respiratory rate (RR) after O3 treatment, whereas HeJ mice on days 1 and 2 had
 9                   increased RR and OuJ mice on days 2 and 3 exhibited increased RR. VT did not change
10                   with treatment among the strains. Overall, B6 mice were mildly responsive with rapid
11                   adaptation, whereas C3 mice were highly responsive with adaptation only in HeJ mice
12                   with regards to changes in cardiac and respiratory responses. HR and HRV parameters
13                   were not equally correlated with VT and RR between the three mice strains, which
14                   suggest that strains vary in the integration of the cardiac and respiratory systems. These
15                   complex interactions could help explain variability in interindividual susceptibility to
16                   adverse health effects of air pollution.

17                   Hamade et al.  (2010) expanded their investigation to explore the variation of these  strain
18                   dependent cardiopulmonary responses with age. As was observed previously, all
19                   experimental mouse strains (B6, HeJ, and OuJ) exhibited decreased HR and increased
20                   HRV after O3 exposure.  Younger O3-exposed mice had a significantly lower HR
21                   compared to older exposed mice, indicating an attenuation of the bradycardic effect of O3
22                   with age. Younger mice  also had a greater increase in rMSSD in HeJ and OuJ strains and
23                   SDNN in HeJ  mice. Conversely, B6 mice had a slightly greater increase in  SDNN in
24                   aged mice compared to the young mice. No change was observed in the magnitude of the
25                   O3 induced increase of SDNN in OuJ mice or rMSSD in B6 mice. The B6 and HeJ mice
26                   genetically vary in respect to the nuclear factor erythroid 2-related factor 2  (Nrf-2). The
27                   authors propose that the  genetic differences between the mice strains could be altering the
28                   formation of ROS, which tends to increase with age, thus modulating the changes in
29                   cardiopulmonary physiology after O3 exposure.

30                   Strain and age differences in HR and heart function were further investigated in B6 and
31                   12981/SvlmJ (129) mice in response to a sequential O3 and filtered air or CB exposure
32                   (Tankersley et al., 2010). Young 129 mice showed  a decrease in HR after O3 or O3  and
33                   CB exposure. This bradycardia was not observed in B6 or older animals in  this study,
34                   suggesting a possible alteration or adaptation  of the autonomic nervous system activity
35                   with age. However, these authors did previously report bradycardia in similarly aged
36                   young B6 mice (Hamade et al.. 2010; Hamade and  Tankersley. 2009; Hamade et al..
37                   2008). Ozone exposure in 129 mice also resulted in an increase in left ventricular
3 8                   chamber dimensions at end diastole (LVEDD) in young and old mice and a decrease in
      Draft - Do Not Cite or Quote                      6-179                                September 2011

-------
 1                   left ventricular posterior wall thickness at end systole (PWTES) in older mice. The
 2                   increase in LVEDD caused a decrease in fractional shortening, which can be used as a
 3                   rough indicator of left ventricular function. Regression analysis revealed a significant
 4                   interaction between age and strain on HR and PWTES, which implies that aging affects
 5                   the HR and function in response to O3 differently between mouse strains.


                     Effects on Cardiovascular-Related Proteins

 6                   Increased BP, changes in HRV, and increased atherosclerosis may be related to increases
 7                   in the vasoconstrictor peptide, endothelin-1 (amino acids  1-21, ET-l[i_2i]). Regulation of
 8                   the pulmonary endothelin system can be affected in rats by inhalation of PM (0, 5,
 9                   50 mg/m3, EHC-93) and O3 (Thomson et al..  2006; Thomson etal.. 2005). Exposure to
10                   either O3 (0.8 ppm) or PM increased plasma ET-l^i], ET-3[1.2i], and the  ET-1 precursor
11                   peptide, bigET-1. Increases in circulating  ET-1 [1-21] could  be a result of a transient
12                   increase in the gene expression of lung preproET-1 and endothelin converting enzyme-1
13                   (ECE-1) immediately following inhalation of O3 or PM. These latter gene expression
14                   changes (e.g. preproET-1 and ECE-1) were additive with  co-exposure to O3 and PM.
15                   Conversely, preproET-3 decreased immediately after O3 exposure, suggesting the
16                   increase in ET-3[!_21] was not through de novo production. A recent study also found
17                   increased ET-1 gene expression in the aorta of O3 exposed rats (Kodavanti et al.. 2011).
18                   These rats also exhibited an increase in ETBR after O3 exposure; however, they did not
19                   demonstrate increased biomarkers for vascular inflammation, thrombosis, or oxidation.

20                   O3 can oxidize protein functional groups and  disturb the affected protein. For example,
21                   the soluble plasma protein fibrinogen is oxidized by O3 (0.01-0.03 ppm) in vitro, creating
22                   fibrinogen and fibrin aggregates, characteristically similar to  defective fibrinogen
23                   (Rosenfeld et al., 2009; Rozenfeld et al., 2008). In these studies, oxidized fibrinogen
24                   retained the ability to form fibrin gels that are involved in coagulation, however the
25                   aggregation time increased and the gels were  rougher than normal with thicker fibers.
26                   Oxidized fibrinogen also developed the ability to self assemble creating fibrinogen
27                   aggregates that may play a role in thrombosis. Since O3 does  not readily translocate past
28                   the ELF and pulmonary epithelium and fibrinogen is primarily a plasma protein, it is
29                   uncertain if O3 would have the opportunity to react with plasma fibrinogen. However,
30                   fibrinogen can be released from the basolateral face of pulmonary epithelial cells during
31                   inflammation, where the deposition of fibrinogen could lead to lung injury (Lawrence
32                   and Simpson-Haidaris. 2004).
      Draft - Do Not Cite or Quote                      6-180                                September 2011

-------
                    Studies on Ozone Reaction Products

 1                  Although recent toxicological studies have demonstrated O3-induced effects on the
 2                  cardiovascular system, as concluded in previous O3 AQCDs, it remains unclear if the
 3                  mechanism is through a reflex response or the result of effects from O3 reaction products
 4                  (U.S. EPA. 2006b. 1996a). A new study that examined O3 reaction byproducts has shown
 5                  that cholesterol secoaldehyde (e.g., atheronal A) induces apoptosis in vitro in mouse
 6                  macrophages (Gao et al., 2009b) and cardiomyocytes (Sathishkumar et al., 2009).
 7                  Additionally, atheronal-A and -B has been found to induce in vitro macrophage and
 8                  endothelial cell proinflammatory events involved in the initiation of atherosclerosis
 9                  (Takeuchi et al.. 2006). These O3 reaction products when complexed with low density
10                  lipoprotein upregulate scavenger receptor class A and induce  dose-dependent
11                  macrophage chemotaxis. Atheronal-A increases expression of the adhesion molecule, E-
12                  selectin, in endothelial cells, while atheronal-B induces monocyte differentiation. These
13                  events contribute to both monocyte recruitment and foam cell formation in
14                  atherosclerotic vessels. It is unknown whether these O3 reaction products gain access to
15                  the vascular space from the lungs. Alternative explanations include the extrapulmonary
16                  release of diffusible mediators that may initiate or propagate inflammatory responses in
17                  the vascular or systemic compartments.
      Draft - Do Not Cite or Quote                      6-181                                September 2011

-------
      Table 6-39      Characterization of study details for Section 6.3.3.2a
             Study
          Model
                                            O3 (ppm)     Exposure Duration
                                     Effects
      Chuang et al. (2009)
   Mice; C57BI/6; M;
   6 weeks
                                           0.5
1 or 5 days, 8-h/day
                          Monkey; rhesus Macaca
                          mulatta;M; Infant (180
                          days old)
                         0.5
                                                        5 days, 8-h/day
 Increased HR and blood pressure. Initiated an
 oxidative environment by increasing vascular
 02~ production, which lead to mtDNA damage
. and -NO consumption, known to perturb
 endothelial function.
Perepuetal. (2010)
                          Rat; Sprague-Dawley; 50-   0.8
                          75 g
                                      28 days, 8-h/day         Enhanced the sensitivity to myocardial I/R
                                                            injury while increasing oxidative stress and
                                                            pro-inflammatory mediators and decreasing
                                     	production of anti-inflammatory proteins.
      Hamade et al. (2008)    Mice; C57BI/6J, C3H/HeJ,  0.6           2-h
                          and C3H/HeOuJ; M;       (subsequent   followed by 3 h of CB
                                                 CB exposure,
                                         	536 ug/m3)	
   18-20 weeks
                                                                              Decreased HR. Strain differences observed in
                                                                              HRV suggest that genetic variability affects
                                                                              cardiac responses.
Hamade and Tankersley Mice; C57BI/6J, C3H/HeJ,  0.6           3 days, 2-h/day
(2009)               and_C3H/HeOuJ; M;       (subsequent    followed by 3-h of CB
                                           CB exposure,
                                           536 ug/m3)
                          18-20 weeks
                                                            Strains varied in integration of the cardiac and
                                                            respiratory systems, implications in
                                                            interindividual variability. B6 mice were mildly
                                                            responsive with rapid adaptation, whereas C3
                                                            mice were highly responsive with adaptation
                                                            only in HeJ mice with regards to changes in
                                                            cardiac and respiratory responses.	
      Hamade et al. (2010)    Mice; C57BI/6J, C3H/HeJ,  0.6           2-h
                          andC3H/HeOuJ;M;       (subsequent    followed by 3-h of CB
                          5 or 12 mo old            CB exposure,
                                                 536 ug/m3)
                                                            Aged mice exhibited attenuated changes in
                                                            cardiopulmonary physiology after 03 exposure.
                                                            Genetic differences between mice strains
                                                            could be altering formation of ROS, which
                                                            tends to increase with age, thus modulating 03
                                                            induced effects.
      Tankersley et al. (2010)  Mice; C57BI/6J,
                          129S1/SvlmJ;M/F;
                          5 or 18 mo old
                         0.6
                                                        2-h
                         (subsequent   followed by 3-h of CB
                         CB exposure,
                         556 ug/m3)	
                      Significant interaction between age and strain
                      on HR and PWTES, which implies that aging
                      affects the HR and function in response to 03
                      differently between mouse strains.
      Thomson et al. (2005)   Rat; Fischer-344; M; 200-   0.4 or 0.8
                          250 g
                                      4-h
                                                                              Activation of the vasoconstricting ET system.
                                                                              Increased plasma ET-1 through higher
                                                                              production and slower clearance.
Thomson et al. (2006)
Kodavanti et al. (2011)
Rat; Fischer-344; M; 200- 0.8
250 g
Rat; Wistar; M; 0.5 or 1.0
10-1 2 weeks
4-h
2 days, 5-h/day
Increased plasma ET-3 not due to de novo
synthesis, unlike ET-1.
No changes to aortic genes of thrombosis,
inflammation or proteolysis, except ET-1 and
ETBR(LOppm).
        * Results from previous studies are presented in Table AX5-14 of the 2006 03 AQCD and Table 6-23 of the 1996 03 AQCD.
1

2

3

4

5

6

7
Summary of Toxicological Studies


Overall, animal  studies suggest that O3 exposure may disrupt both the -NO and

endothelin systems, which can result in an increase in HR, HRV, and ANF, as is

observed after O3 exposure. Conversely, studies in rodents also exhibit O3 induced

bradycardia, but it is uncertain if this decrease in HR is also observed in humans.

Additionally, O3 may increase oxidative stress and vascular inflammation promoting the

progression of atherosclerosis and leading to increased susceptibility to I/R injury. As O3

reacts quickly with the ELF and does not translocate to the heart and large vessels,

studies suggest that the cardiovascular effects exhibited could be caused by reaction
      Draft - Do Not Cite or Quote
                                     6-182
                                                                                                  September 2011

-------
 1                   byproducts of O3 exposure. However, direct evidence of translocation of O3 reaction
 2                   products to the cardiovascular system has not been demonstrated in vivo. Alternatively,
 3                   extrapulmonary release of diffusible mediators, such as cytokines or endothelins, may
 4                   initiate or propagate inflammatory responses in the vascular or systemic compartments
 5                   leading to the reported cardiovascular pathologies. Further discussion of the modes of
 6                   action that may lead to cardiovascular effects can be found in Section 5.3.8.
             6.3.4   Summary and Causal Determination

 7                   In past O3 AQCDs the effects of O3 to the cardiovascular system did not receive much
 8                   attention due to the paucity of information available. However, in recent years,
 9                   investigation of O3-induced cardiovascular events has advanced. In general, compared
10                   with the epidemiologic evidence, the toxicological evidence is more supportive of O3-
11                   induced cardiovascular effects. Epidemiologic evidence does not consistently
12                   demonstrate a positive relationship between short-term O3 exposure and cardiovascular-
13                   related morbidity. However, most epidemiologic studies have not extensively
14                   investigated the cardiovascular effects of O3 exposure in susceptible populations, which
15                   may further support the toxicological findings. Although the epidemiologic evidence of
16                   cardiovascular morbidity is limited, single-city studies reviewed in the 2006 O3 AQCD,
17                   recent multicity studies, and the multicontinent APHENA study provide evidence of
18                   consistently positive associations between short-term O3 exposure and cardiovascular
19                   mortality. However, in contrast with respiratory effects, there is weak coherence between
20                   associations for cardiovascular morbidity and mortality. Further, there is no apparent
21                   biological mechanism to explain the association observed for short-term O3 exposure
22                   with cardiovascular mortality.

23                   Animal toxicological studies provide evidence for O3-induced cardiovascular effects,
24                   specifically enhanced I/R injury, disrupted NO-induced vascular reactivity, decreased
25                   cardiac function, and increased HRV. The observed increase in HRV is supported by a
26                   recent controlled human exposure study that also finds increased high frequency HRV,
27                   but not altered blood pressure, following O3 exposure. Toxicological studies investigating
28                   the role of O3 in heart rate regulation are mixed with both bradycardie and tachycardic
29                   responses observed. These changes in cardiac function provide evidence for O3-induced
30                   alterations in the  autonomic nervous system leading to cardiovascular complications.
31                   Epidemiologic studies showing positive association between O3 and arrhythmias confirm
32                   the development  of autonomic dysfunction following O3 exposure. It is still uncertain
33                   how O3 inhalation may cause systemic toxicity; however the cardiovascular effects of O3
34                   found in animals correspond to the development and maintenance of an extrapulmonary
3 5                   oxidative, proinflammatory environment.


      Draft - Do Not Cite or Quote                      6-183                                 September 2011

-------
 1                   In conclusion, animal toxicological studies provide stronger evidence for O3 exposure
 2                   leading to cardiovascular morbidity than do epidemiologic studies, among which there is
 3                   a lack of coherence among endpoints. Based on the relatively strong body of
 4                   toxicological evidence, and the consistent evidence of an association between O3 and
 5                   cardiovascular mortality, but weak coherence and biological plausibility for O3-induced
 6                   cardiovascular morbidity, the generally limited body of evidence is suggestive of a
 7                   causal  relationship  between relevant short-term exposures to O3 and
 8                   cardiovascular effects.
          6.4   Central  Nervous System Effects

 9                   The 2006 O3 AQCD included toxicological evidence that acute exposures to O3 are
10                   associated with alterations in neurotransmitters, motor activity, short and long term
11                   memory, and sleep patterns. Additionally, histological signs of neurodegeneration have
12                   been observed. Reports of headache, dizziness, and irritation of the nose with O3
13                   exposure are common complaints in humans, and some behavioral changes in animals
14                   may be related to these symptoms rather than indicative of neurotoxicity. Peterson and
15                   Andrews (1963) and Tepper et al. (1983) showed that mice would alter their behavior to
16                   avoid O3 exposure. Murphy et al. (1964) and Tepper et al. (1982) showed that running -
17                   wheel behavior was suppressed, and Tepper et al. (1985)  subsequently demonstrated the
18                   effects of a 6-h exposure to O3 on the suppression of running-wheel behavior in rats and
19                   mice, with the lowest effective concentration being about 0.12 ppm O3 in the rat and
20                   about 0.2 ppm in the mouse. The suppression of active behavior by 6 h of exposure to
21                   0.12 ppm O3 has recently been confirmed by Martrette et al. (2011) in juvenile female
22                   rats, and the suppression of three different active behavior parameters was found to
23                   become more pronounced after 15 days of exposure. A table of studies examining the
24                   effects of O3 on behavior can be found on p 6-128 of the  1996 O3 AQCD. Generally
25                   speaking, transient changes in behavior in rodent models  appear to be dependent on a
26                   complex interaction of factors such  as (1) the type of behavior being measured, with
27                   some behaviors increased and others suppressed; (2) the factors motivating that behavior
28                   (differences in reinforcement); and (3) the sensitivity of the particular behavior (e.g.,
29                   active behaviors are more affected than more sedentary behaviors). Many behavioral
30                   changes are likely to result from avoidance of irritation, but more recent studies indicate
31                   that O3 also directly affects the CNS.

32                   Research in the area of O3-induced neurotoxicity has notably increased over the past few
33                   years, with the majority of the evidence coming from toxicological studies that examined
34                   the association between O3 exposure, neuropathology, and neurobehavioral effects, and
35                   more limited evidence from epidemiologic studies.  In an epidemiologic study conducted
      Draft - Do Not Cite or Quote                      6-184                                September 2011

-------
 1                   by Chen and Schwartz (2009), data from the NHANES III cohort was utilized to study
 2                   the relationship between long-term O3 exposure (mean annual O3 concentration of
 3                   26.5 ppb) and neurobehavioral effects among adults aged 20-59 years. The authors
 4                   observed an association between annual exposure to O3 and tests measuring coding
 5                   ability and attention/short-term memory. Each 10-ppb increase in annual O3 levels
 6                   corresponded to an aging-related cognitive performance decline of 3.5 years for coding
 7                   ability and 5.3 years for attention/short-term memory. These associations persisted in
 8                   both crude and adjusted models. There was no association between annual O3
 9                   concentrations and reaction time tests. The authors conclude that overall there is a
10                   positive association between O3 exposure and reduced performance on neurobehavioral
11                   tests. Although Chen and Schwartz (2009) is a long-term exposure study, it is included in
12                   this section because it is the first epidemiologic study to demonstrate that exposure to
13                   ambient O3 is associated with decrements in neurocognitive tests related to memory and
14                   attention in humans. This epidemiologic evidence of an effect on the CNS due to
15                   exposure to ambient concentrations of O3 is coherent with animal studies demonstrating
16                   that exposure to O3 can produce a variety of CNS effects including behavioral deficits,
17                   morphological changes, and oxidative stress in the brains of rodents. In these rodent
18                   studies, interestingly, CNS effects were reported at O3 concentrations that were  generally
19                   lower than those concentrations commonly observed to produce pulmonary or cardiac
20                   effects  in rats.

21                   A number of new studies demonstrate various perturbations in neurologic function or
22                   histology, including changes similar to those observed with Parkinson's and Alzheimer's
23                   disease pathologies occurring in similar regions of the brain (Table 6-40). Many of these
24                   include exposure durations ranging from short-term to long-term, and as such are
25                   discussed here and in Chapter 7 with emphasis on the effects resulting from exposure
26                   durations relevant to the respective chapter. Several studies assess short- and long-term
27                   memory acquisition via passive avoidance behavioral testing and find decrements in test
28                   performance after O3 exposure, consistent with the aforementioned observation made in
29                   humans by Chen and Schwartz (2009). Impairment of long-term memory has been
30                   previously described in rats exposed to 0.2 ppm O3 for 4 h (Rivas-Arancibia et al.. 1998)
31                   and in other studies of 4-hour exposures at concentrations of 0.7 to 1 ppm (Dorado-
32                   Martinez etal.. 2001: Rivas-Arancibia et al.. 2000: Avila-Costa et al.. 1999). More
33                   recently, statistically significant decreases in both short and long-term memory were
34                   observed in rats after 15 days of exposure to 0.25 ppm O3 (Rivas-Arancibia et al.. 2010).

35                   The central nervous system is very sensitive to oxidative stress, due in part to its high
36                   content of polyunsaturated fatty acids, high rate of oxygen consumption, and low
37                   antioxidant enzyme capacity. Oxidative stress has been identified as one of the
3 8                   pathophysiological mechanisms underlying neurodegenerative disorders such as
      Draft - Do Not Cite or Quote                      6-185                                September 2011

-------
 1                   Parkinson's and Alzheimer's disease, among others (Simonian and Coyle. 1996). It is
 2                   also believed to play a role in altering hippocampal function, which causes cognitive
 3                   deficits with aging (Vanguilder and Freeman. 2011). A particularly common finding in
 4                   studies of O3-exposed rats is lipid peroxidation in the brain, especially in the
 5                   hippocampus, which is important for higher cognitive function including contextual
 6                   memory acquisition. Performance in passive avoidance learning tests is impaired when
 7                   the hippocampus is injured, and the observed behavioral effects are well correlated with
 8                   histological and biochemical changes in the hippocampus, including reduction in spine
 9                   density in the pyramidal neurons (Avila-Costa et al.. 1999), lipoperoxidation (Rivas-
10                   Arancibia et al. 2010;  Dorado-Martinez et al.. 2001). progressive neurodegeneration, and
11                   activated and phagocytic microglia (Rivas-Arancibia et al.. 2010). The hippocampus is
12                   also one of the main regions affected by age-related neurodegenerative diseases,
13                   including Alzheimer's disease, and it may be more sensitive to oxidative damage in aged
14                   rats. In a study of young (47 days) and aged (900 days) rats exposed to 1 ppm O3 for 4 h,
15                   O3-induced lipid peroxidation occurred to a greater extent in the striatum of young rats,
16                   whereas it was highest in the hippocampus in aged rats (Rivas-Arancibia et al.. 2000).
17                   Martinez-Canabal et al. (2008) showed exposure of rats to 0.25 ppm, 4h/day, for 7, 15, or
18                   30 days increased lipoperoxides in the hippocampus. This effect was observed at day 7
19                   and continued to increase with time, indicating cumulative oxidative damage. O3-induced
20                   changes in lipid peroxidation, neuronal death, and COX-2 positive cells in the
21                   hippocampus could be significantly inhibited by daily treatment with growth hormone
22                   (GH), which declines with age in most species. The protective effect of GH  on -induced
23                   oxidative stress was greatest at  15 days of exposure and was non-significant at day 30.
24                   Consistent with these findings, lipid peroxidation in the hippocampus of rats was
25                   observed to increase significantly after a 30-day exposure to 0.25 ppm , but  not after a
26                   single 4-h exposure to the same concentration (Mokoena et al.. 2010). However, 4 hours
27                   of exposure was sufficient to cause significant increases in lipid peroxidation when the
28                   concentration was increased to 0.7 ppm, and another study observed lipid peroxidation
29                   after a 4-h exposure to 0.4 ppm (Dorado-Martinez et al.. 2001).

30                   Other commonly affected areas of the brain include the striatum, substantia nigra,
31                   cerebellum, olfactory bulb, and frontal/prefrontal cortex.  The  striatum and substantia
32                   nigra are particularly sensitive to oxidative stress because the  metabolism of dopamine,
33                   central to their function, is an oxidative process perturbed by redox imbalance. Oxidative
34                   stress  has been implicated in the premature death of substantia nigra dopamine neurons in
35                   Parkinson's disease. Angoa-Perez et al. (2006)  have  shown progressive lipoperoxidation
36                   in the  substantia nigra and a decrease in nigral dopamine  neurons in ovariectomized
37                   female rats exposed to 0.25 ppm O3, 4h/day, for 7, 15, or 30 days. Estradiol, an
38                   antioxidant, attenuated O3-induced oxidative stress and nigral neuronal death, and the
39                   authors note that in humans, estrogen therapy can ameliorate symptoms of Parkinson's

      Draft - Do Not Cite or Quote                       6-186                                September 2011

-------
 1                   disease, which is more prevalent in men. Progressive oxidative stress has also been
 2                   observed in the striatum and substantia nigra of rats after 15 and 30 days of exposure to
 3                   0.25 ppm O3 for 4 h/day, along with a loss of dopaminergic neurons from the substantia
 4                   nigra (Pereyra-Munoz et al.. 2006). Decreases in motor activity were also observed at 15
 5                   and 30 days of exposure, consistent with other reports (Martrette et al., 2011; Dorado-
 6                   Martinez et al.. 2001). Using a similar O3 exposure protocol, Santiago-Lopez and
 7                   colleagues (2010) also observed a progressive loss of dopaminergic neurons within the
 8                   substantia nigra, accompanied by alterations in the morphology of remaining cells and an
 9                   increase in p53 levels and nuclear translocation.

10                   The olfactory bulb also undergoes oxidative damage in O3 exposed animals, in some
11                   cases altering olfactory-dependent behavior. Lipid peroxidation was observed in the
12                   olfactory bulbs of ovariectomized female rats exposed to 0.25 ppm O3 (4 h/day) for 30 or
13                   60 days (Guevara-Guzman et al.. 2009). O3 also induced decrements in a selective
14                   olfactory recognition memory test, and the authors note that early  deficits in odor
15                   perception and memory are components of human neurodegenerative diseases. The
16                   decrements in olfactory memory were not due to damaged olfactory perception based on
17                   other tests. However, deficits in olfactory perception emerged with longer exposures
18                   (discussed in Chapter 7). As with the study by Angoa-Perez et al.  (2006) described
19                   above, a protective effect for estradiol was demonstrated for both  lipid peroxidation and
20                   olfactory memory defects. The role of oxidative stress in memory deficits and associated
21                   morphological changes has also been demonstrated via attenuation by other antioxidants
22                   as well, such as a-tocopherol (Guerrero et al.. 1999) and taurine (Rivas-Arancibia et al..
23                   2000).

24                   It is unclear how persistent these effects might be. One study of acute exposure, using
25                   1 ppm O3 for 4 hours, observed morphological changes in the olfactory bulb of rats at
26                   2 hours, and 1 and 10 days, but not 15 days, after exposure (Colin-Barenque et al., 2005).
27                   Other acute studies also report changes in the CNS. Lipid peroxidation was observed in
28                   multiple regions of the brain after a 1- to 9-h exposure to 1 ppm O3 (Escalante-Membrillo
29                   et al..  2005). Ozone has also been shown to alter gene expression of endothelin-1
30                   (pituitary) and inducible nitric oxide synthase (cerebral hemisphere) after a single 4-h
31                   exposure to 0.8 ppm O3, indicating potential cerebrovascular effects. This concentration-
32                   dependent effect was not observed at 0.4 ppm O3 (Thomson et al.. 2007). Vascular
33                   endothelial growth factor was upregulated in astroglial cells in the central respiratory
34                   areas of the brain of rats exposed to 0.5 ppm O3 for 3 hours (Araneda et al., 2008). The
35                   persistence of CNS changes after a single exposure was also examined and the increase in
36                   vascular endothelial growth factor was present after a short (3 hours) recovery period.
37                   Thus,  there is evidence that O3-induced CNS effects are both concentration- and time-
3 8                   dependent.
      Draft - Do Not Cite or Quote                       6-187                                September 2011

-------
 1                   Because O3 can produce a disruption of the sleep-wake cycle (U.S. EPA. 2006b). Alfaro-
 2                   Rodriguez et al. (2005) examined whether acetylcholine in a region of the brain involved
 3                   in sleep regulation was altered by O3. After a 24-h exposure to 0.5 ppm O3, the
 4                   acetylcholine concentration in the medial preoptic area was decreased by 58% and
 5                   strongly correlated with a disruption in paradoxical sleep.  Such behavioral-biochemical
 6                   effects of O3 are confirmed by a number of studies which have demonstrated
 7                   morphological and biochemical changes in rats.

 8                   CNS effects have also been demonstrated in newborn and  adult rats whose only exposure
 9                   to O3 occurred in utero. Several neurotransmitters were assessed in male offspring of
10                   dams exposed to 1 ppm O3 during the entire pregnancy (Gonzalez-Pina et al., 2008). The
11                   data showed that catecholamine neurotransmitters were affected to a greater degree than
12                   indole-amine neurotransmitters in the cerebellum. CNS changes, including behavioral,
13                   cellular, and biochemical effects, have also been observed after in utero exposure to
14                   0.5 ppm O3 for 12 h/day from gestational days 5-20 (Boussouar et al., 2009). Tyrosine
15                   hydroxylase labeling in the nucleus tractus solatarius was increased after in utero
16                   exposure to O3 whereas Fos protein labeling did not change.  When these offspring were
17                   challenged by immobilization stress, neuroplasticity pathways, which were activated in
18                   air-exposed offspring, were inhibited in O3-exposed offspring. Although an O3 exposure
19                   concentration-response was not studied in these two in utero studies, it has been
20                   examined in one study. Santucci et al. (2006) investigated behavioral effects and gene
21                   expression after in utero exposure of mice to as little as 0.3 ppm O3. Increased
22                   defensive/submissive behavior and reduced social investigation were observed in both the
23                   0.3 and 0.6 ppm O3 groups. Changes in gene expression of brain-derived neurotrophic
24                   factor (BDNF, increased in striatum) and nerve growth factor (NGF, decreased in
25                   hippocampus) accompanied these behavioral changes. Thus, these three studies
26                   demonstrate that CNS effects can occur as a result of in utero exposure to O3, and
27                   although the mode of action of these effects is not known, it has been suggested that
28                   circulating lipid peroxidation products may play a role (Boussouar et al.. 2009).
29                   Importantly, these CNS effects occurred in rodent models  after in utero only exposure to
3 0                   relevant concentrations of O3.
      Draft - Do Not Cite or Quote                       6-188                                September 2011

-------
Table 6-40      Central Nervous  System and Behavioral Effects of Short-term O3
                    Exposure in Rats
Study
Martretteetal. (2011)
Angoa-Perez et al. (2006)
Guevara-Guzman et al. (2009)
Martinez-Canabal et al. (2008)
Pereyra-Munoz et al. (2006)
Rivas-Arancibia et al. (2010)
Santiago-Lopez et al. (2010)
Thomson et al. (2007)
Model , °3 .
(ppm)
Rat; Wistar; F; 0.12
Weight: 152g;
7 weeks old
Rat; Wistar; F; Weight: 0.25
300g; ovariectomized
Rat; Wistar; F; 264g; 0.25
ovariectomized
Rat; Wistar; M; 0.25
Weight: 300g
Rat; Wistar; M; 250- 0.25
300g
Rat; Wistar; M; 250- 0.25
300g
Rat; Wistar; M; 250- 0.25
300g
Rat; Fischer-344; M; 0.4; 0.8
200-250g
Exposure
Duration
1-15days,6h/day
7 to 60 days, 4-
h/day, 5 days/wk
30 and 60 days,
4h/day
7 to 30 days, 4-h/day
15 and 30 days, 4-h/
day
15 to 90 days, 4-h/
day
15, 30, and 60 days,
4-h/day
4-h; assays at 0 and
24 h post exposure
Effects
Significant decrease in rearing, locomotor activity,
and jumping activity at day 1 , with a further
decrease in these activities by day 15.
Progressive lipid peroxidation and loss of tyrosine
hydrolase-immunopositive neurons in the
substantia nigra starting at 7 days.
Estradiol treatment protected against lipid
peroxidation and decreases in estrogen receptors
and dopamine p-hydroxylase in olfactory bulbs
along with deficits in olfactory recognition
memory.
Growth hormone inhibited 03-induced increases
in lipoperoxidation and COX-2 positive cells in the
hippocampus.
Decreased motor activity, increased lipid
peroxidation, altered morphology, and loss of
dopamine neurons in substantia nigra and
striatum, increased expression of DARPP-32,
iNOS, and SOD.
Ozone produced significant increases in lipid
peroxidation in the hippocampus, and altered the
number of p53 positive immunoreactive cells,
activated and phagocytic microglia cells, GFAP
immunoreactive cells, and doublecortine cells,
and short- and long-term memory-retention
latency.
Progressive loss of dopamine reactivity in the
substantia nigra, along with morphological
changes. Increased p53 levels and nuclear
translocation.
At 0.8 ppm, 03 produced rapid perturbations in
the ET-NO pathway gene expression in the brain.
Ozone induced a small but significant time- and
concentration-dependent increase in prepro-
endothelin-1 mRNA levels in the cerebral
hemisphere and pituitary, whereas TNFa and
iNOS mRNA levels were decreased at 0 hrs and
unchanged or increased, respectively, at 24 h.
Alfaro-Rodriguez and Gonzalez-
Pina (2005)
                            Rat; Wistar; M; 292g   0.5
                              24-h
                            During the light phase, 03 caused a significant
                            decrease in paradoxical sleep accompanied by a
                            significant decrease in Ach levels in the
                            hypothalamic medial preoptic area. The same
                            effects occurred during the dark phase exposure
                            to 03 in addition to a significant increase in slow-
                            wave sleep and decrease in wakefulness.
Araneda et al. (2008)
Boussouar et al. (2009)
Rats; Sprague- 0.5
Dawley; M; 280-320g
Rat; Sprague-Dawley; 0.5
M; adult offspring of
prenatally exposed
dams; 403-41 4g
3-h (measurements
taken at 0 h and 3 h
after exposure)
From embryonic day
E5toE20for12-
h/day; immobilization
stress
Ozone upregulated VEGF in astroglial cells
located in the respiratory center of the brain.
VEGF co-located with IL-6 and TNF in cells near
blood vessel walls, and blood vessel area was
markedly increased.
Prenatal 03 exposure had a long term impact on
the nucleus tractus solitarius of adult rats, as
revealed during immobilization stress.
Soulage et al. (2004)
Rat; Sprague-Dawley;
M;Approx. 7 weeks
old
0.7
                                                         5-h
Ozone produced differential effects on peripheral
and central components of the sympatho-adrenal
system. While catecholamine biosynthesis was
increased in portions of the brain, the
catecholamine turnover rate was significantly
increased in the heart and cerebral cortex and
inhibited in the lung and striatum.
Draft - Do Not Cite or Quote
                           6-189
                                                   September 2011

-------
Study
Guzman et al. (2006: 2005)
Colin-Barenque et al. (2005)
Escalante-Membrillo etal. (2005)
Gonzalez-Pina et al. (2008)
Model
Rat; Wistar; M; 21
days old; well-
nourished and
malnourished groups
Rats; Wistar; M; 250-
300g
Rats; Wistar; M; 280-
320g
Rat; Wistar; M;
Os Exposure
(ppm) Duration
0.75 15 successive days
for 4-h/day
1 .0 4-h; assays at 2-h,
24-h, 10 days, and
15 days after
exposure
1.0 1-,3-,6-,or9-h
1 12-h/day,21daysof
gestation; assays at
0, 5,& 10 days
postnatal
Effects
A significant decrease in body weight was
observed in both well nourished (WN) and
malnourished (MN) rats after 03 exposure.
Localized ATPase, TEARS, and GSH levels
changed in response to ozone in certain brain
areas and the ozone-induced changes were
dependent on nutritional condition.
A significant loss of dendritic spines in granule
cells of the olfactory bulb occurred at 2 hrs to 10
days after exposure. Cytological and
ultrastructural changes returned towards normal
morphology by 15 days.
Significant increases in TEARS occurred in
hypothalamus, cortex, striatum, midbrain,
thalamus, and pons. Partial but significant
recovery was observed by 3 h after the 9 h
exposure.
Prenatal 03 exposure produced significant
decreases in cerebellar monoamine but not
indolamine. content at 0 and 5 days after birth
with a partial recovery by 10 d. 5-hydroxy-indole-
acetic acid levels were significantly increased at
10 days.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
      6.4.1    Neuroendocrine Effects

               According to the 2006 O3 AQCD, early studies suggested an interaction of O3 with the
               pituitary-thyroid-adrenal axis, because thyroidectomy, hypophysectomy, and
               adrenalectomy protected against the lethal effects of O3. Concentrations of 0.7-1.0 ppm
               O3 for a 1-day exposure in male rats caused changes in the parathyroid, thymic atrophy,
               decreased serum levels of thyroid hormones and protein binding, and increased prolactin.
               Increased toxicity to O3 was reported in hyperthyroid rats and T3 supplementation was
               shown to increase metabolic rate  and pulmonary injury in the lungs of O3-treated animals.
               The mechanisms by which O3 affects neuroendocrine function are not well understood,
               but previous work suggests that high ambient levels of O3 can produce marked neural
               disturbances in structures involved in the integration of chemosensory inputs, arousal,
               and motor control, effects that may be responsible for some of the behavioral effects seen
               with O3 exposure. A more recent study exposing immature female rats to 0.12 ppm O3
               demonstrated significantly increased serum levels of the thyroid hormone free T3 after 15
               days of exposure, whereas free T4 was unchanged (Martrette et al., 2011). These results
               are in contrast to those previously presented whereby 1 ppm O3 for 1 day significantly
               decreased T3 and T4 (demons and Garcia. 1980), although comparisons are made
               difficult by highly disparate exposure regimens along with sex differences. Martrette et
               al.(2011) also demonstrated significantly increased corticosterone levels after 15 days of
               exposure, suggesting a  stress related response.
Draft - Do Not Cite or Quote
                                                   6-190
September 2011

-------
            6.4.2   Summary and Causal Determination

 1                   In rodents, O3 exposure has been shown to cause physicochemical changes in the brain
 2                   indicative of oxidative stress and inflammation. Newer toxicological studies add to earlier
 3                   evidence that acute exposures to O3 can produce a range of effects on the central nervous
 4                   system and behavior. Previously observed effects, including neurodegeneration,
 5                   alterations in neurotransmitters, short and long term memory, and sleep patterns, have
 6                   been further supported by recent studies. In instances where  pathology and behavior are
 7                   both examined, animals exhibit decrements in behaviors tied to the brain regions or
 8                   chemicals found to be affected or damaged. For example, damage in the hippocampus,
 9                   which is important for memory acquisition, was correlated with impaired performance in
10                   tests designed to assess memory. Thus the brain is functionally affected by O3 exposure.
11                   The single epidemiologic study conducted showed an association between O3 exposure
12                   and memory deficits in humans as well, albeit on a long-term exposure basis. Notably,
13                   exposure to O3 levels as low as 0.25 ppm for 7 days has resulted in progressive
14                   neurodegeneration and deficits in both short and long-term memory in rodents.
15                   Examination of changes in the brain at lower exposure concentrations or at 0.25 ppm for
16                   shorter durations has not been reported,  but 0.12 ppm O3 has been shown to alter
17                   behavior. It is possible that some behavioral changes may reflect avoidance of irritation
18                   as opposed to  functional changes in brain morphology or chemistry, but in many cases
19                   functional changes are related to oxidative stress and damage. In some instances, changes
20                   were dependent on the nutritional status of the  rats (high versus low protein diet). For
21                   example, O3 produced an increase in glutathione in the brains of rats fed the high protein
22                   diet but decreases in glutathione in rats fed low protein chow (Calderon Guzman et al.,
23                   2006). The hippocampus, one of the main regions affected by age-related
24                   neurodegenerative diseases, appears to be more sensitive to oxidative damage in aged rats
25                   (Rivas-Arancibia et al., 2000). and growth hormone, which declines with age in most
26                   species, may be protective (Martinez-Canabal and Angora-Perez. 2008). Developing
27                   animals may also be sensitive, as changes in the CNS, including biochemical, cellular,
28                   and behavioral effects, have been observed in juvenile and adult animals whose sole
29                   exposure occurred in utero, at levels as a low as 0.3 ppm. A  number of studies
30                   demonstrate ozone-induced changes that are also observed in human neurodegenerative
31                   disorders such as Alzheimer's and Parkinson's disease, including signs of oxidative
32                   stress, loss of neurons/neuronal death, reductions in dopamine levels, increased COX-2
33                   expression, and increases in activated microglia in important regions of the brain
34                   (hippocampus, substantianigra).

35                   Although evidence from epidemiologic and controlled human exposure studies  is lacking,
36                   the toxicological evidence  for ozone's impact on the brain and behavior is strong, and at
      Draft - Do Not Cite or Quote                      6-191                                September 2011

-------
 1                  least is suggestive of a causal relationship between O3 exposure and effects on the
 2                  central nervous system.
          6.5    Effects on Other Organ Systems
            6.5.1   Effects on the Liver and Xenobiotic Metabolism

 3                  Early investigations of the effects of O3 on the liver centered on xenobiotic metabolism,
 4                  and the prolongation of drug-induced sleeping time, which was observed at 0.1 ppm O3
 5                  (Graham et al., 1981). In some species, only adults and especially females were affected.
 6                  In rats, high (1.0-2.0 ppm for 3 hours) acute O3 exposures caused increased production of
 7                  NO by hepatocytes and enhanced protein synthesis (Laskin et al., 1996; Laskin et al.,
 8                  1994). Except for the earlier work on xenobiotic metabolism, the responses occurred only
 9                  after very high acute O3 exposures.  One study, conducted at 1 ppm O3 exposure, has been
10                  identified (Last et al.. 2005) in which alterations in gene expression underlying O3-
11                  induced cachexia and downregulation of xenobiotic metabolism were examined. A
12                  number of the down-regulated genes are known to be interferon (IFN) dependent,
13                  suggesting a role for circulating IFN. A more recent study by Aibo et al. (2010)
14                  demonstrates exacerbation of acetaminophen-induced liver injury in mice after a single
15                  6-h exposure to 0.25 or 0.5 ppm O3. Data indicate that O3 may worsen drug-induced liver
16                  injury by inhibiting hepatic repair. The O3-associated effects shown in the liver are
17                  thought to be mediated by inflammatory cytokines or other cytotoxic mediators released
18                  by activated macrophages or other cells in the lungs (Laskin and Laskin. 2001; Laskin et
19                  al., 1998; Vincent et al.. 1996b). Recently, increased peroxidated lipids were detected in
20                  the plasma of O3  exposed animals (Santiago-Lopez et al.. 2010).

21                  In summary, mediators generated by O3 exposure may cause effects on the liver in
22                  laboratory rodents.  Ozone exposures as low as 0.1 ppm have been shown to affect drug-
23                  induced sleeping time, and exposure to 0.25 ppm can exacerbate liver injury induced by a
24                  common analgesic. However, very few studies at relevant concentrations have been
25                  conducted, and no data from controlled human exposure  or epidemiologic studies are
26                  currently available. Therefore the collective evidence is inadequate to determine if a
27                  causal relationship exists between short-term O3 exposure and effects on the liver
28                  and metabolism.
      Draft - Do Not Cite or Quote                      6-192                               September 2011

-------
             6.5.2   Effects on Cutaneous and Ocular Tissues

 1                   In addition to the lungs, the skin is highly exposed to O3 and contains O3 reactive targets
 2                   (polyunsaturated fatty acids) that can produce lipid peroxides. The 2006 O3 AQCD
 3                   reported that although there is evidence of oxidative stress at near ambient O3
 4                   concentrations, skin and eyes are only affected at high concentrations (greater than
 5                   1-5 ppm). Ozone exposure (0.8 ppm for 7 days) induces oxidative stress in the skin of
 6                   hairless mice, along with proinflammatory cytokines (Valacchi et al., 2009). A recent
 7                   study demonstrated that 0.25 ppm O3 differentially alters expression of
 8                   metalloproteinases in the skin of young and aged mice, indicating age-related
 9                   susceptibility to oxidative stress (Fortino et al.. 2007). In young mice, healing of skin
10                   wounds is not significantly affected by O3 exposure (Lim et al.. 2006). However,
11                   exposure to 0.5 ppm O3 for 6 h/day significantly delays wound closure in aged mice. As
12                   with effects on the liver described above, the effects of O3 on the skin and eyes have not
13                   been widely studied, and information from controlled human exposure or epidemiologic
14                   studies is not currently available. Therefore the collective evidence is inadequate to
15                   determine if a causal  relationship exists between short-term O3 exposure and
16                   effects on cutaneous and ocular tissues.
          6.6   Mortality
             6.6.1   Summary of Findings from 2006 Ozone AQCD

17                   The 2006 O3 AQCD reviewed a large number of time-series studies consisting of single-
18                   and multicity studies, and meta-analyses. In the large U.S. multicity studies that
19                   examined all-year data, summary effect estimates corresponding to single-day lags
20                   ranged from a 0.5-1% increase in all-cause (nonaccidental) mortality per the standardized
21                   unit increase1 in O3. The association between short-term O3 exposure and mortality was
22                   substantiated by a collection of meta-analyses and international multicity studies. The
23                   studies evaluated found some evidence for heterogeneity in O3 mortality risk estimates
24                   across cities and studies. Studies that conducted seasonal analyses, although more limited
25                   in number, reported larger O3 mortality risk estimates during the warm or summer
26                   season. Overall, the 2006 O3 AQCD identified robust associations between various
27                   measures of daily ambient O3 concentrations and all-cause mortality, with additional
28                   evidence for associations with cardiovascular mortality, which could not be readily
29                   explained by confounding due to time, weather, or copollutants. However, it was noted
        1 In the 2006 O3 AQCD and throughout this document to compare across studies that used the same exposure metric, effect
      estimates were standardized to 40 ppb for 1-h maximum, 30 ppb for 8-h maximum, and 20 ppb for 24-h average O3 concentrations.
      Draft - Do Not Cite or Quote                      6-193                                September 2011

-------
 1                   that multiple uncertainties remain regarding the O3-mortality relationship including: the
 2                   extent of residual confounding by copollutants; factors that modify the O3-mortality
 3                   association; the appropriate lag structure for identifying O3-mortality effects (e.g., single-
 4                   day lags versus distributed lag model); the shape of the O3-mortality C-R function and
 5                   whether a threshold exists; and the identification of susceptible populations. Collectively,
 6                   the 2006 O3 AQCD concluded that "the overall body of evidence is highly suggestive that
 7                   O3 directly or indirectly contributes to non-accidental and cardiopulmonary-related
 8                   mortality."
            6.6.2   Associations of Mortality and Short-Term Ozone Exposure

 9                   The recent literature that examined the association between short-term O3 exposure and
10                   mortality further confirmed the associations reported in the 2006 O3 AQCD. New
11                   multicontinent and multicity studies reported consistent positive associations between
12                   short-term O3 exposure and all-cause mortality in all-year analyses, with additional
13                   evidence for larger mortality risk estimates during the warm or summer months (Figure
14                   6-27; Table 6-41). These associations were reported across a range of ambient O3
15                   concentrations that were in some cases quite low (Table 6-42).
      Draft - Do Not Cite or Quote                      6-194                                September 2011

-------
 Study

 Gryparisetal. (2004;57276)
 Bell etal. (2007; 93256)
 Schwartz (2005; 57333)
 Bell and Dominici (2008; 193828)
 Bell etal. (2004; 94417)a
 Levy et al. (2005; 74347)a
 Katsouyanni et al. (2009; 199899)
 Bell etal. (2005; 74345 )a
 Ito etal. (2005; 74346)a
 Wonget al. (2010; 732535)
 Katsouyanni et al. (2009; 199899)
 Cakmak etal. (2011, 699135)
 Katsouyanni et al. (2009; 199899
 Katsouyanni et al. (2009; 199899 b

 Samolietal.(2009;195855)
 Bell etal. (2004; 94417)a
 Schwartz (2005; 57333)
 Zanobetti and Schwartz (2008; 195755
 Zanobetti and Schwartz (2008; 101596
 Franklin and Schwartz (2008; 156448)
 Gryparisetal. (2004;57276)
 Medina-Ramon and Schwartz (2008)
 Katsouyanni et al. (2009; 199899)
 Bell etal. (2005; 74345 )a
 Katsouyanni et al. (2009; 199899
 Katsouyanni et al. (2009; 199899 b
 Levy etal. (2005; 74 347)a
 Ito etal. (2005; 74346)a
 Katsouyanni et al. (2009; 199899)
 Stafoggia et al. (2010; 625034)
    Location

 APHEA2 (23 cities)
98 U.S. communities
   14 U.S. cities
98 U.S. communities
95 U.S. communities
 U.S. and Non-U.S.
  APHENA-Europe
 U.S. and Non-U.S.
 U.S. and Non-U.S.
  PAPA (4 cities
   APHENA-y.S.
  7 Chilean cities
  APHENA-Canada
  APHENA-Canada

 21 European cities
95 U.S. communities

   48 U.S. cities
   48 U.S. cities
18 U.S. communities
 APHEA2 (21 cities)
   48 U.S. cities
  APHENA-Europe
 U.S. and Non-U.S.
  APHENA-Canada
  APHENA-Canada
 U.S. and Non-U.S.
 U.S. and Non-U.S.
   APHENA-U.S.
  10 Italian cities
  Lag

  0-1
  0-1

  0°6
  0-6

DL(0-2)
 0-1
DL 0-2
All-Year
DL
DL
   0-6
DL 0-2
   0-2
 0-1
 0Q6

  0
 0-3
  0
 0-1
 0-2
DL(0-2)

DLjO-2)
DL(0-2)
DL(0-2)
DL(0-5)
                                                           Summer
                                                                                                                        11
                                                                                       % Increase
  Effect estimates are for a 40 ppb increase in 1-h max, 30 ppb increase in 8-h max, and 20 ppb increase in 24-h avg ozone
concentrations. An "a" represent multicity studies and meta-analyses from the 2006 ozone AQCD. Bell et al. (2005). Ito et al. (2005).
and Levy et al. (2005) used a range of lag days in the meta-analysis: Lag 0, 1, 2, or average 0-1 or 1-2; single-day  lags from 0 to 3;
and lag 0 and 1-2; respectively. A"b" represents risk estimates from APHENA-Canada standardized to an approximate IQR of 5.1
ppb for a 1-h max increase in ozone concentrations (see explanation in Section 6.2.7.2).



Figure  6-27     Summary of mortality risk estimates for short-term ozone exposure

                     and all-cause (nonaccidental) mortality from  all-year and  summer

                     season analyses.
Draft - Do Not Cite or Quote
                           6-195
                                                    September 2011

-------
Table 6-41      Corresponding effect estimates for Figure 6-27
Study
Location
Lag
Avg Time
% Increase (95% Cl)
All-year
Gryparis et al. (2004)
Bell et al. (2007)
Schwartz (2005a)
Bell and Dominici (2008)
Bell et al. (2004)a
Lew et al. (2005)a
Katsouyanni et al. (2009)
Bell et al. (2005)a
Ito et al. (2005)a
Wonaetal. (2010)
Katsouyanni et al. (2009)
Cakmaketal. (2011)
Katsouyanni et al. (2009)
Katsouyanni et al. (2009)b
APHEA2 (23 cities)
98 U.S. communities
14 U.S. cities
98 U.S. communities
95 U.S. communities
U.S. and Non-U.S.
APHENA-Europe
U.S. and Non-U.S.
U.S. and Non-U.S.
PAPA (4 cities)
APHENA-U.S.
7 Chilean cities
APHENA-Canada
APHENA-Canada
0-1
0-1
0
0-6
0-6
—
DL(0-2)
—
—
0-1
DL(0-2)
DL(0-6)
DL(0-2)
DL(0-2)
1-hmax
24-h avg
1-hmax
24-h avg
24-h avg
24-h avg
1-hmax
24-h avg
24-h avg
8-h avg
1-hmax
8-h max
1-h max
1-h max
0.24 (-0.86, 1.98)
0.64 (0.34, 0.92)
0.76(0.13,1.40)
1.04(0.56,1.55)
1.04(0.54,1.55)
1.64(1.25,2.03)
1 .66 (0.47, 2.94)
1.75(1.10,2.37)
2.20 (0.80, 3.60)
2.26(1.36,3.16)
3.02(1.10,4.89)
3.35(1.07,5.75)
5.87(1.82,9.81)
0.73(0.23,1.20)
Summer
Samolietal. (2009)
Bell et al. (2004)a
Schwartz (2005a)
Zanobetti and Schwartz (2008a)
Zanobetti and Schwartz (2008b)
Franklin and Schwartz (2008)
Gryparis et al. (2004)
Medina-Ramon and Schwartz (2008)
Katsouyanni et al. (2009)
Bell et al. (2005)a
Katsouyanni et al. (2009)
Katsouyanni et al. (2009)
Lew et al. (2005)a
Ito et al. (2005)a
Katsouyanni et al. (2009)
Stafoggiaetal. (2010)
21 European cities
95 U.S. communities
14 U.S. cities
48 U.S. cities
48 U.S. cities
18 U.S. communities
APHEA2 (21 cities)
48 U.S. cities
APHENA-Europe
U.S. and Non-U.S.
APHENA-Canada
APHENA-Canada
U.S. and Non-U.S.
U.S. and Non-U.S.
APHENA-U.S.
10 Italian cities
0-1
0-6
0
0
0-3
0
0-1
0-2
DL(0-2)
—
DL(0-2)
DL(0-2)
—
—
DL(0-2)
DL(0-5)
8-h max
24-h avg
1-h max
8-h max
8-h max
24-h avg
8-h max
8-h max
1-hmax
24-h avg
1-hmax
1-hmax
24-h avg
24-h avg
1-hmax
8-h max
0.66(0.24,1.05)
0.78(0.26,1.30)
1.00(0.30,1.80)
1.51 (1.14,1.87)
1 .60 (0.84, 2.33)
1 .79 (0.90, 2.68)
1 .80 (0.99, 3.06)
1.96(1.14,2.82)
2.38(0.87,3.91)
3.02(1.45,4.63)
3.34(1.26,5.38)
0.42 (0.16, 0.67)
3.38 (2.27, 4.42)
3.50(2.10,4.90)
3.83(1.90,5.79)
9.15(5.41,13.0)
  aMulticity studies and meta-analyses from the 2006 03 AQCD. Bell et al. (2005)'. Ito et al. (2005)'. and Levy et al. (2005)' used a range of lag
days in the mete-analysis: Lag 0,1, 2, or average 0-1 or 1-2; Single-day lags from 0-3; and Lag 0 and 1-2; respectively.
  bRisk estimates from APHENA-Canada standardized to an approximate IQRof 5.1 ppbfora 1-h max increase in 03 concentrations (see
explanation in Section 6.2.7.2).
Draft - Do Not Cite or Quote
6-196
September 2011

-------
Table 6-42      Range of mean and upper percentile ozone concentrations in
                   previous and recent multicity studies
1
2
Study
Gryparis et al.
(2004)"
Schwartz (2005a)b
Bell et al. (2004)
Bell et al. (2007)
Bell and Dominici
(2008)
Franklin and
Schwartz (2008)
Katsouyanni et al.
(2009)"'e
Medina-Ramon
and Schwartz
(2008)b
Samolietal.
(2009)b
Stafoggiaetal.
(2010)
Cakmaket al.
(2011)
Wong et al. (2010)
Zanobetti and
Schwartz (2008b)
Zanobetti and
Schwartz (2008a)
Location
Years
23 European 1990-1997
cities (APHEA2)
14 U.S. cities 1986-1993
95 U.S. 1987-2000
communities
(NMMAPS)
98 U.S. 1987-2000
communities
(NMMAPS)
98 U.S. 1987-2000
communities (All year and
(NMMAPS) May-September)
18 U.S. 2000-2005
communities (May-September)
NMMAPS 1987-1996 (Canada and
12 Canadian U.S.) varied by city for
cities Europe
(APHEA2)
48 U.S. cities 1989-2000
(May-September)
21 European 1990-1997
cities (APHEA2) (June-August)
10 Italian cities 2001-2005
(April-September)
7 Chilean cities 1997-2007
PAPA (4 cities) 1999-2003 (Bangkok)
1996-2002 (Hong Kong)
2001-2004
(Shanghai)
2001 -2004 (Wuhan)
48 U.S. cities 1989-2000
(June-August)
48 U.S. cities"
1989-2000
Wnter: December-February)
Spring: March-May)
Summer: June-August)
Autumn: September-
\lovem ber)
Averaging
Time
1-h max
8-h max
1-h max
24-h avg
24-h avg
24-h avg
24-h avg
1-h max
8-h max
8-h max
8-h max
8-h max
8-h avg
8-h max
8-h max
Mean
Concentration (ppb)
Summer:
1-h max: 44-1 17
8-h max: 30-99
Winter:
1-h max: 11 -57
8-h max: 8-49
35.1-60
26.0
26.0°
All year: 26.8
May-September: 30.0
21.4-48.7
U.S.: 13.3-38.4
Canada: 6.7-8.4
Europe:1 8.3-41 .9
16.1-58.8
20.0-62.8
41.2-58.9
59.0-87.6
18.7-43.7
15.1-62.8
Wnter: 16.5
Spring: 41. 6
Summer: 47.8
Autumn: 33.5
Upper Percentile
Concentrations (ppb)
Summer:
1-h max: 62-1 73
8-h max: 57-1 54
Winter:
1-h max: 40-88
8-h max: 25-78
25th: 26.5-52
75th: 46.3-69
NR
NR
Maximum:
All year: 37.3
May-September: 47.2
NR
75th:
U.S.: 21. 0-52.0
Canada: 8.7-1 2.5
Europe: 24.0-65.8
NR
75th: 27.2-74.8
75th: 47.0-71 .6
NR
75th: 38.4 -60.4
Max: 92.1 -131. 8
Max: 34.3-146.2
75th: 19.8-75.9
Max:
Wnter: 40.6
Spring: 91. 4
Summer: 103.0
Autumn: 91.2
  a03 concentrations were converted to ppb if the study presented them as ug/m3 by using the conversion factor of 0.51 assuming standard
temperature (25° C) and pressure (1 atm).
  bStudy only reported median 03 concentrations.
  °Cities with less than 75% observations in a season excluded. As a result, 29 cities examined in winter, 32 in spring, 33 in autumn, and all 48 in
the summer.
  dBell et al. (2007)did not report mean 03 concentrations, however, it used a similar dataset as Bell et al. (2004) which consisted of 95 U.S.
communities for 1987-2000. For comparison purposes the 24-h avg 03 concentrations for the 95 communities from Bell et al. (2004) are reported
here.
  eStudy did not present air quality data for the summer months.
  CV=coefficient of variation


                In addition to examining the relationship between short-term O3 exposure and all-cause

                mortality, recent studies attempted to address the uncertainties that remained upon the

                completion of the 2006 O3 AQCD. As a result, given the  robust associations between
Draft - Do Not Cite or Quote
                                                        6-197
September 2011

-------
 1                   short-term O3 exposure and mortality presented across studies in the 2006 O3 AQCD and
 2                   supported in the new multicity studies, the following sections primarily focus on the
 3                   examination of previously identified uncertainties in the O3-mortality relationship,
 4                   specifically: O3 associations with cause-specific mortality, confounding, lag structure
 5                   (e-g-, multiday effects and mortality displacement), effect modification (i.e., sources of
 6                   heterogeneity in risk estimates across cities); and the O3-mortality C-R relationship.
 7                   Focusing specifically on these uncertainties allows for a more detailed characterization of
 8                   the relationship between short-term O3 exposure and mortality.
                     6.6.2.1    Confounding

 9                   Recent epidemiologic studies examined potential confounders of the O3-mortality
10                   relationship. These studies specifically focused on whether PM and its constituents or
11                   seasonal trends confounded the association between short-term O3 exposure and
12                   mortality.


                     Confounding  by PM and PM Constituents

13                   An important question in the evaluation of the association between short-term O3
14                   exposure and mortality is whether the relationship is confounded by particulate matter,
15                   particularly the PM chemical components that are found in the "summer haze" mixture
16                   which also contains O3. However, because of the temporal correlation among these PM
17                   components and O3, and their possible interactions, the interpretation of results from
18                   multipollutant models that attempt to disentangle the health effects associated with each
19                   pollutant is challenging.

20                   The potential confounding effects of PM10 and PM2 5 on the O3-mortality relationship
21                   were examined by Bell et al. (2007) using data on 98 U.S. urban communities for the
22                   years 1987-2000 from the National Morbidity, Mortality, and Air Pollution Study
23                   (NMMAPS). In this analysis the authors included PM as a covariate in time-series
24                   models, and also examined O3.mortality associations on days when O3 concentrations were
25                   below a specified value. This analysis was limited by the small fraction of days when
26                   both PM and O3 data were available, due to the every-3rd- or 6th-day sampling schedule
27                   for the PM indices, and the limited amount of city-specific data for PM2 5 because it was
28                   only collected in most cities since 1999. As a result, of the 91 communities with PM25
29                   data, only 9.2% of days in the study period had data for both O3 and PM2 5, resulting in
30                   the use of only 62 communities in the PM2 5 analysis. An examination of the correlation
31                   between PM (PMi0 and PM2 5) and O3 across various strata of daily PMi0 and PM25
32                   concentrations found that neither PM size fraction was highly correlated with O3 across


      Draft - Do Not Cite or Quote                      6-198                               September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 9
10
11
12
13
14
15
16
17
any of the strata examined. These results were also observed when using 8-h max and 1-h
max O3 exposure metrics. National and community-specific effect estimates of the
association between short-term O3 exposure and mortality were robust to inclusion of
PMio or PM25 in time-series models through the range of O3 concentrations (i.e.,
<10 ppb, 10-20, 20-40, 40-60, 60-80, and >80 ppb). For example, the percent increases in
nonaccidental deaths per 10 ppb increase 24-h avg O3 concentrations at lag 0-1 day were
0.22% (95% CI: -0.22, 0.65) without PM25 and 0.21% (95% CI: -0.22, 0.64) with PM25
in 62 communities.

Although no strong correlations between PM and O3 were reported by Bell et al. (2007)
the patterns observed suggest regional differences in their correlation. (Table 6-43). Both
PMio and PM2 5 show positive correlations with O3 in the Industrial Midwest, Northeast,
Urban Midwest, and Southeast, especially in the summer months, presumably, because of
the summer peaking sulfate. However, the mostly negative or weak correlations between
PM and O3 in the summer in the Southwest, Northwest, and southern California could be
due to winter-peaking nitrate. Thus, the potential confounding effect of PM on the
O3-mortality relationship could be influenced by the relative contribution of sulfate and
nitrate, which varies regionally and seasonally.
Table 6-43 Correlations between PM and ozone by season and region

No. of Communities
Winter
Spring
Summer
Fall
Yearly
PM10
Industrial Midwest
Northeast
Urban Midwest
Southwest
Northwest
southern California
Southeast
U.S.
19
15
6
9
11
7
25
93
0.37
0.34
0.24
0.00
-0.17
0.19
0.33
0.23
0.44
0.44
0.25
0.02
-0.20
0.08
0.35
0.26
0.44
0.36
0.22
-0.02
-0.13
0.12
0.31
0.24
0.39
0.44
0.26
0.10
-0.11
0.19
0.31
0.26
0.41
0.40
0.24
0.03
-0.16
0.14
0.32
0.25
PM2.5
Industrial Midwest
Northeast
Urban Midwest
Southwest
Northwest
southern California
Southeast
U.S.
19
13
4
9
11
7
26
90
0.18
0.05
0.22
-0.15
-0.32
-0.25
0.38
0.09
0.39
0.26
0.31
-0.08
-0.34
-0.22
0.47
0.21
0.43
0.16
0.15
-0.17
-0.39
-0.25
0.30
0.12
0.44
0.43
0.32
-0.15
-0.24
-0.15
0.37
0.22
0.36
0.25
0.20
-0.14
-0.31
-0.23
0.39
0.16
Source: Bell et al. (2007).
      Draft - Do Not Cite or Quote
                               6-199
September 2011

-------
                                                   Raw Estimates
                                15-

                                10-
                              o
                              !  5-
                              .c
                              i
                               1.5-

                               1.0-
                              3
                              I 0.5-

                              ; 0.0 -

                              -0.5-
                                 -1.0
                     -0.5
                                                0           5
                                                   Without PM10
                                                 Posterior Estimates
                                                                      10
0.0      0.5
 Without PM10
1.0
                                                                           1.5
       Source: Reprinted with permission of Informa UK Ltd (Smith et al.. 2009b).
       The diagonal line indicates 1:1 ratio.

      Figure 6-28   Scatter plots of ozone mortality risk estimates with versus without
                      adjustment for PMio in NMMAPS cities.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
In an attempt to reassess a number of issues associated with the O3-mortality relationship,
including confounding, Smith et al. (2009b) re-analyzed the publicly available NMMAPS
database for the years 1987-2000. The authors conducted a number of analyses using
constrained distributed lag models and the average of 0- and 1-day lags. In addition,
Smith et al. (2009b) examined the effect of different averaging times (24-h, 8-h, and 1-h
max) on O3-mortality regression coefficients, and whether PM10 confounded the
Os-mortality relationship. The authors reported that, in most cases, O3 mortality risk
estimates were reduced by between 22% and 33% in copollutant models with PM10. This
is further highlighted in Figure 6-28, which shows scatter plots of O3_mortality risk
estimates with adjustment for PM10 versus without adjustment for PM10. Smith et al.
(2009b) point out that a larger fraction (89 out of 93) of the posterior estimates lie below
the diagonal line (i.e., estimates are smaller with PM10 adjustment) compared to the raw
estimates (56 out of 93). This observation could be attributed to both sets of posterior
estimates being calculated by "shrinking towards the mean." However, the most
      Draft - Do Not Cite or Quote
                               6-200
                                     September 2011

-------
 1                   prominent feature of these plots is that the variation of O3.mortality risk estimates across
 2                   cities is much larger than the impact of PMi0 adjustment on the O3.mortality relationship.

 3                   Franklin and Schwartz (2008) examined the sensitivity of O3 mortality risk estimates to
 4                   the inclusion of PM25 or PM chemical components associated with secondary aerosols
 5                   (e-g-, sulfate [SO42~], organic carbon [OC], and nitrate [NO3-]) in copollutant models.
 6                   This analysis consisted of between 3 and 6 years of data from May through September
 7                   2000-2005 from 18 U.S. communities. The association between O3 and non-accidental
 8                   mortality was examined in single-pollutant models and after adjustment for PM2 5,
 9                   sulfate, organic carbon, or nitrate concentrations. The single-city effect estimates were
10                   combined into an overall estimate using a random-effects model. In the single-pollutant
11                   model, the authors found a 0.89% (95% CI: 0.45, 1.33%) increase in nonaccidental
12                   mortality with a 10 ppb increase in same-day 24-h summertime O3 concentrations across
13                   the 18 U.S. communities. Adjustment for PM25 mass, which was available for 84% of the
14                   days, decreased the O3.mortality risk estimate only slightly (from 0.88% to 0.79%), but the
15                   inclusion of sulfate in the model reduced the risk estimate by 31% (from 0.85% to
16                   0.58%). However, sulfate data were only available for 18% of the  days. Therefore, a
17                   limitation of this study is the limited amount of data for PM2 5 chemical components due
18                   to the every-3rd-day or every-6th-day sampling schedule. For example, when using a
19                   subset of days when organic carbon measurements were available  (i.e., 17% of the
20                   available days), O3 mortality risk estimates were reduced to 0.51% (95% CI: -0.36 to
21                   1.36) in a single-pollutant model.

22                   Consistent with the studies previously discussed, the  results from Franklin and Schwartz
23                   (2008) also demonstrate that the interpretation of the  potential confounding effects of
24                   copollutants on O3 mortality risk estimates is not straightforward. As presented in Figure
25                   6-29, the regional and city-to-city variations in O3 mortality risk estimates appear greater
26                   than the impact of adjusting for copollutants. In addition, in some cases, a negative O3
27                   mortality risk estimate becomes even more negative with the inclusion of sulfate (e.g.,
28                   Seattle) in a copollutant model, or a null O3 mortality risk estimate becomes negative
29                   when sulfate is included (e.g., Dallas and Detroit). Thus, the reduction in the overall O3
30                   mortality risk estimate (i.e., across cities) needs to be assessed in the context of the
31                   heterogeneity in the single-city estimates.
      Draft - Do Not Cite or Quote                       6-201                                 September 2011

-------


Seattle





tl Paso
Dallas
Houston
Beaumont
Kansas City
St. Louis
Detroit

Pittsburgh
Buffalo

Rochester
Philadelphia
Boston


1 	 fa 	 ±H





h-D-^
^C
1 	 '
rts
1 — ^^





1 	 ,
D Ozone with sulfate I
x Ozone alone |


	 1


x n i

r^
j^

=ii
« — i

__i
r-i *. . I



                                           -5         0          5
                                          Percent increase in mortality
                                          with 10 ppb increase in ozone

       Source: Reprinted from Franklin and Schwartz (2008).

      Figure 6-29   Community-specific ozone-mortality risk estimates for
                     nonaccidental mortality per 10 ppb increase in same-day 24-h avg
                     summertime ozone concentrations in single-pollutant models and
                     copollutant models with sulfate.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
In the APHENA study, the investigators from the U.S. (NMMAPS), Canadian, and
European (APHEA2) multicity studies collaborated and conducted a joint analysis of
PMio and O3 using each of these datasets (Katsouyanni et al.. 2009). For mortality, each
dataset consisted of a different number of cities and years of air quality data: U.S.
encompassed 90 cities with daily O3 data from 1987-1996 of which 36 cities had summer
only O3 measurements; Europe included 23 cities with 3-7 years of daily O3 data during
1990-1997; and Canada consisted of 12 cities with daily O3 data from 1987 to 1996. As
discussed in Section 6.2.7.2, the APHENA study conducted extensive sensitivity
analyses, of which the 8 df/year results for both the penalized spline (PS) and natural
spline (NS) models are presented in the text for comparison purposes, but only the NS
results are presented in figures because alternative spline models have previously been
shown to result in similar effect estimates (HEI. 2003). Additionally, for the  Canadian
results, figures contain risk estimates standardized to both a 40 ppb increment for 1-h
      Draft - Do Not Cite or Quote
                             6-202
September 2011

-------
 1                   max O3 concentrations, consistent with the rest of the ISA, but also the approximate IQR
 2                   across the Canadian cities as discussed previously (Section 6.2.7.2).

 3                   In the three datasets, the authors found generally positive associations between short-term
 4                   O3 exposure and all-cause, cardiovascular, and respiratory mortality. The estimated
 5                   excess risks for O3 were larger for the  Canadian cities than for the U.S. and European
 6                   cities. When examining the potential confounding effects of PM10 on O3 mortality risk
 7                   estimates, the sensitivity of the estimates varied across the data sets and age groups. In
 8                   the Canadian dataset, adjusting for PM10 modestly reduced O3 risk estimates for all-cause
 9                   mortality for all ages in the PS (4.5% [95% CI: 2.2, 6.7%]) and NS (4.2% [95% CI: 1.9,
10                   6.5%]) models to 3.8% (95% CI: -1.4, 9.8%) and 3.2% (95%  CI: -2.2, 9.0%),
11                   respectively, at lag 1 for a 40 ppb increase in 1-h max O3 concentrations (Figure 6-30;
12                   Table 6-44). However, adjusting for PM10 reduced O3 mortality risk estimates in the >
13                   75-year age group, but increased the risk estimates in the <75-year age group. For
14                   cardiovascular and respiratory mortality more variable results were observed with O3 risk
15                   estimates being reduced and increased, respectively, in copollutant models with PMi0
16                   (Figure 6-30; Table 6-44). Unlike the European and U.S. datasets, the Canadian dataset
17                   only conducted copollutant analyses at lag 1; as a result, to provide a comparison across
18                   study locations only the lag 1 results are presented for the European and U.S. datasets in
19                   this section.

20                   In the European data, O3 risk estimates were robust when adjusting for PM10 in the year-
21                   round data for all-cause, cardiovascular and respiratory mortality. When restricting  the
22                   analysis to the summer months moderate reductions were observed in O3 risk estimates
23                   for all-cause mortality with more pronounced reductions in respiratory mortality. In the
24                   U.S. data, adjusting for PM10 moderately reduced O3 risk estimates for all-cause mortality
25                   in a year-round analysis at lag 1 (e.g., both the PS and NS models were reduced from
26                   0.18%to 0.13%) (Figure 6-30; Table 6-44). Similarto the European data, when
27                   restricting the analysis to the summer months, adjusting for PMi0 moderately reduced O3
28                   mortality risk estimates in the U.S. However, when examining cause-specific mortality
29                   risk estimates, consistent with the results from the Canadian dataset, which employed a
30                   similar PM sampling strategy (i.e., every-6th-day sampling), O3 risk estimates for
31                   cardiovascular and respiratory mortality were more variable; reduced or increased in
32                   all-year and summer analyses. Overall, the estimated O3 risks appeared to be moderately
33                   to substantially sensitive to inclusion of PMi0 in copollutant models. Despite the multicity
34                   approach, the mostly every-6th-day sampling schedule  for PM10 in the Canadian and U.S.
35                   datasets greatly reduced the sample size and limits the interpretation of these results.
      Draft - Do Not Cite or Quote                      6-203                                September 2011

-------
           Location
           APHENA-U.S.
           APHENA-Canada
                    a
                    a
                    a
                    a
           APHENA-Europe
Ages
 All

>75
<75
>75
<75
 All

 All

>75

<75

 All

 All

>75
<75
>75
<75
 All
                                        All-Cause
                                         Cardiovascular
                                         Respiratory
                                        All-Cause
Cardiovascular
                                         Respiratory
                 -!r-0
                                                      All-Year
                                                      Summer
                                                      All-Year

                                                      Summer

                                                      All-Year
                                                      Summer
                                                      All-Year
	
All-Cause

Cardiovascular
—
_
Respiratory —
—
	
-10 -5 (
w v
•9- All-Year
-+- Summer
-•— All-Year
3=
^£-
+ Summer
^-0 	
1 A 	
^=
!-• 	 All-Year
L-O 	
^
	 ^ 	 Summer
r-0 	
3 5 10 15 20 25 30
% Increase
  Effect estimates are for a 40 ppb increase in 1 -h max O3 concentrations at lag 1. All estimates are for the 8 df/year model with
natural splines. Circles represent all-year analysis results while diamonds represent summer season analysis results. Open circles
and diamonds represent copollutant models with  PM10. Black = all-cause mortality;  red = cardiovascular mortality; and blue =
respiratory mortality. An "a" represents risk estimates from APHENA-Canada standardized to an approximate IQR of 5.1 ppb for a
1-h max increase in O3 concentrations (see explanation in Section 6.2.7.2).

Figure 6-30     Percent increase in all-cause (nonaccidental) and cause-specific
                   mortality from  the APHENA study for  single- and copollutant
                   models.
Draft - Do Not Cite or Quote
                    6-204
                                                  September 2011

-------
Table 6-44 Corresponding
Location Mortality
APHENA-U.S. All-Cause



Cardiovascular







Respiratory



APHENA-Canada All-Cause



Cardiovascular







Respiratory



APHENA-Europe All-Cause



Cardiovascular







Respiratory



Effect
Ages
All



>75

<75

>75

<75

All



All



>75



<75



All



All



>75

<75

>75

<75

All



Estimates
Season
All-year

Summer

All-year



Summer



All-year

Summer

All-year















All-year

Summer

All-year



Summer



All-year

Summer

for Figure 6-30
Copollutant

PM10

PM10

PM10

PM10

PM10

PM10

PM10

PM10


PM10
PM10


PM10
PM10


PM10
PM10


PM10
PM10

PM10

PM10

PM10

PM10

PM10

PM10

PM10

PM10

% Increase (95% Cl)
1.42(0.08,2.78)
1.02 (-1.40, 3.50)
4.31 (2.22, 6.45)
1.90 (-0.78, 4.64)
1.10 (-1.33, 3.67)
0.47 (-4.61, 5.79)
-0.1 6 (-3.02, 2.86)
1.34 (-3.63, 6.61)
3.58 (0.87, 6.37)
-1.1 7 (-6.18, 4.07)
3.18(0.31,6.12)
1 .26 (-4.46, 7.28)
2.46 (-1.87, 6.86)
3.50 (-4.23, 11.8)
6.04(1.18,11.1)
7.03 (-3.48, 18.5)
4.15(1.90,6.45)
0.52 (0.24, 0.80)a
3.18 (-2.18, 8.96)
0.40 (-0.28, 1.1 0)a
5.62 (0.95, 10.7)
0.70(0.12,1.30)3
1.90 (-9.03, 14.1)
0.24 (-1.20,1.70)3
1.10 (-4.08, 6.61)
0.1 4 (-0.53, 0.82)3
-2.64 (-14.7, 11.5)
-0.34 (-2.00, 1.40)3
0.87 (-6.40, 8.96)
0.11 (-0.84,1.10)3
22.3 (-12.6, 71.3)
2.60 (-1.70, 7.10)3
1.02(0.39,1.66)
1.26(0.47,1.98)
2.06(1.10,2.94)
1.26(0.16,2.30)
1.10 (-0.47, 2.70)
1.1 8 (-0.55, 2.94)
1 .34 (-0.24, 2.94)
1.74 (-0.31, 3.75)
2.54 (0.39, 4.80)
1.58 (-0.70, 3.99)
1.66 (-0.70, 4.15)
1.66 (-1.02, 4.40)
1.42 (-1.02, 3.83)
1.42 (-1.02, 3.83)
4.31 (1.66,7.11)
1.18 (-1.79, 4.31)
1
2
  "Risk estimates from APHENA-Canada standardized to an approximate IQR of 5.1 ppb for a 1-h max increase in O3
concentrations (see explanation in Section 6.2.7.2).


               Stafoggia et al. (2010) examined the potential confounding effects of PMi0 on the

               O3-mortality relationship in individuals 35 years of age and older in 10 Italian cities from
     Draft - Do Not Cite or Quote
                                                6-205
September 2011

-------
 1                   2001 to 2005. In a time-stratified case-crossover analysis, using data for the summer
 2                   months (i.e., April-September), the authors examined O3-mortality associations across
 3                   each city, and then obtained a pooled estimate through a random-effects meta-analysis.
 4                   Stafoggia et al. (2010) found a strong association with nonaccidental mortality (9.2%
 5                   [95% CI: 5.4, 13.0%] for a 30 ppb increase in 8-h max O3 concentrations) in an
 6                   unconstrained distributed lag model (lag 0-5) that persisted in copollutant models with
 7                   PM10 (9.2% [95% CI: 5.4, 13.7%]). Additionally, when examining cause-specific
 8                   mortality, the authors found positive associations between short-term O3 exposure and
 9                   cardiovascular (14.3% [95% CI: 6.7, 22.4%]), cerebrovascular (8.5% [95% CI: 0.1,
10                   16.3%]), and respiratory (17.6% [95% CI: 1.8, 35.6%]) mortality in single-pollutant
11                   models. In copollutant models,  O3-mortality effect estimates for cardiovascular and
12                   cerebrovascular mortality were  robust to the inclusion of PMi0 (9.2% [95% CI: 5.4,
13                   13.7%]) and 7.3% [95% CI: -1.2, 16.3%], respectively), and attenuated, but remained
14                   positive, for respiratory mortality (9.2% [95% CI: -6.9, 28.8%]). Of note, the correlations
15                   between O3 and PM10 across cities were found to be generally low, ranging from (-0.03 to
16                   0.49). The authors do not specify the sampling strategy used for PMi0 in this analysis.


                     Confounding by Seasonal  Trend

17                   The APHENA study (Katsouyanni et al.. 2009). mentioned above, also conducted
18                   extensive sensitivity analyses to identify the appropriate: smoothing method and basis
19                   functions to estimate smooth functions  of time in city-specific models; and degrees of
20                   freedom to be used in smooth functions of time, to adjust for seasonal trends. Because O3
21                   peaks in the summer and mortality peaks in the winter, not adjusting or not sufficiently
22                   adjusting for the seasonal trend would result in an apparent negative association between
23                   the O3 and mortality time-series. Katsouyanni et al. (2009) examined the effect of the
24                   extent of smoothing for seasonal trends by using models with 3 df/year, 8 df/year (the
25                   choice  for their main model), 12 df/year, and df/year selected using the sum of absolute
26                   values of partial autocorrelation function of the model residuals (PACF) (i.e., choosing
27                   the degrees of freedom that minimizes positive and negative autocorrelations in the
28                   residuals). Table 6-45 presents the results of the degrees of freedom analysis using
29                   alternative methods to calculate a combined estimate: the Berkey et al. (1998) meta-
30                   regression and the two-level normal independent sampling estimation (TLNISE)
31                   hierarchical method. The results show that the methods used to combine single-city
32                   estimates did not influence the overall results, and that neither 3 df/year nor choosing the
3 3                   df/year by minimizing the sum of absolute values of PACF of regression residuals was
34                   sufficient to adjust for the seasonal negative relationship between O3 and mortality.
35                   However, it should be noted, the majority of studies in the literature that examined the
36                   mortality effects of short-term O3 exposure, particularly the multicity studies, used 7 or
      Draft - Do Not Cite or Quote                       6-206                                September 2011

-------
 1                   8 df/year to adjust for seasonal trends, and in both methods a positive association was
 2                   observed between O3 exposure and mortality.
      Table 6-45     Sensitivity of ozone risk estimates per 10 |jg/m3 increase in
                      24-h avg ozone concentrations at lag 0-1 to alternative methods for
                      adjustment of seasonal trend, for all-cause mortality using Berkey
                      MLE and TLNSE Hierarchical Models
Seasonality Control
3 df/year
8 df/year
12 df/year
PACF
Berkey
-0.54 (-0.88, 0.20)
0.30(0.11,0.50)
0.34(0.15,0.53)
-0.62 (-1 .01 , -0.22)
TLNISE
-0.55 (-0.88, -0.22)
0.31 (0.09, 0.52)
0.33(0.12,0.54)
-0.62 (-0.98, -0.27)
       Source: Reprinted with permission of Health Effects Institute (2009).
                    6.6.2.2   Effect Modification

 3                  There have been several multicity studies that examined potential effect modifiers, or
 4                  time-in variant factors, that may modify O3 mortality risk estimates. These effect
 5                  modifiers can be categorized into either individual-level or community-level
 6                  characteristics, which are traditionally, examined in second stage regression models. The
 7                  results from these analyses also inform upon whether certain populations are susceptible
 8                  to O3-related health effects (Chapter 8). In addition to potentially modifying the
 9                  association between short-term O3 exposure and mortality, both individual-level and
10                  community-level characteristics may also contribute to the apparent geographic pattern of
11                  spatial heterogeneity in O3 mortality risk estimates. As a result, the geographic pattern of
12                  O3 mortality risk estimates is also evaluated in this section.


                    Individual-Level Characteristics

13                  Medina-Ramon and Schwartz (2008) conducted a case-only study in 48 U.S. cities to
14                  identify populations potentially susceptible to O3-related mortality for the period
15                  1989-2000 (May through September of each year [i.e., warm season]). A case-only
16                  design predicts the occurrence of time-invariant characteristics among cases as a function
17                  of the exposure level (Armstrong. 2003). For each potential effect modifier (time-
18                  invariant individual-level characteristics), city-specific logistic regression models were
19                  fitted, and the estimates were pooled across  all cities. Furthermore, the authors examined
20                  potential differences in individual effect modifiers according to several city
21                  characteristics (e.g., mean O3 level, mean temperature, households with central air
      Draft - Do Not Cite or Quote                      6-207                               September 2011

-------
 1                   conditioning, and population density) in a meta-regression. Across cities the authors
 2                   found a 1.96% (95% CI: 1.14-2.82%) increase in mortality at lag 0-2 for a 30 ppb
 3                   increase in 8-h max O3 concentrations. Additionally, Medina-Ramon and Schwartz
 4                   (2008) examined a number of individual-level characteristics (e.g., age, race) and chronic
 5                   conditions (e.g., secondary causes of death) as effect modifiers of the association between
 6                   short-term O3 exposure and mortality. The authors found that older adults (i.e., > 65),
 7                   women >60 years of age, black race, and secondary atrial fibrillation showed the greatest
 8                   additional percent change in O3-related mortality (Table 6-46). In addition, when
 9                   examining city-level characteristics, the authors found that older adults, black race, and
10                   secondary atrial fibrillation had a larger effect on O3 mortality risk estimates in cities with
11                   lower O3 levels. Of note, a similar case-only study (Schwartz. 2005b) examined  potential
12                   effect modifiers of the association between temperature and mortality, which would be
13                   expected to find results consistent with the Medina-Ramon and Schwartz (2008) study
14                   due to the high correlation between temperature and O3. However, when stratifying days
15                   by temperature Schwartz (2005b) found strong evidence that diabetes modified the
16                   temperature-mortality association on hot days, which was not as evident when examining
17                   the O3-mortality association in Medina-Ramon and Schwartz (2008). This difference
18                   could be due to the study design and populations included in both studies, a multeity
19                   study including all ages (Medina-Ramon and  Schwartz. 2008) compared to a single-city
20                   study of individuals > 65 years of age (Schwartz. 2005b). However, when examining
21                   results stratified by race, nonwhites were found to have higher mortality risks  on both hot
22                   and cold days, which provide some support for the additional risk found for black race in
23                   Medina-Ramon and Schwartz (2008).

24                   Individual-level factors that may result in susceptibility to O3-related mortality were also
25                   examined by Stafoggia et al. (2010). As discussed above, using a time-stratified  case-
26                   crossover analysis, the authors found an association between short-term O3 exposure and
27                   nonaccidental mortality in an unconstrained distributed lag model in 10 Italian cities
28                   (9.2% [95% CI: 5.4, 13.0%; lag 0-5 for a 30 ppb increase in 8-h max O3 concentrations).
29                   Stafoggia et al. (2010) conducted additional analyses to examine whether age, sex,
30                   income level, location of death, and underlying chronic conditions increased the risk of
31                   O3-related mortality, but data were only available for nine of the cities for these analyses.
32                   Of the individual-level factors examined, the authors found the strongest evidence for
33                   increased risk of O3-related mortality in individuals > 85  years of age (22.4%  [95% CI:
34                   15.0, 30.2%]), women (13.7% [95% CI: 8.5, 19.7%]), and out-of-hospital deaths (13.0%
35                   [95% CI: 6.0, 20.4%]). When focusing specifically on out-of hospital deaths and the
36                   subset of individuals with chronic conditions, Stafoggia et al. (2010) found the strongest
37                   association for individuals with diabetes, which is consistent with the potentially
38                   increased susceptibility of diabetics on hot days observed in Schwartz (2005b).
      Draft - Do Not Cite or Quote                       6-208                                September 2011

-------
Table 6-46 Additional percent change in ozone-related mortality for individual-
level susceptibility factors

Percentage
(95% Cl)
Socio-demographic characteristics
Age 65 yr or older
Women
Women <60 yr old"
Women > 60 yr old"
Black race
Low education
1.10
0.58
-0.09
0.60
0.53
-0.29
0.44,1.77
0.18,0.98
-0.76, 0.58
0.25, 0.96
0.19,0.87
-0.81,0.23
Chronic conditions (listed as secondary cause)
Respiratory system diseases
Asthma
COPD
1.35
0.01
-0.31,3.03
-0.49, 0.52
Circulatory system diseases
Atherosclerosis
Atherosclerotic CVD
Atherosclerotic heart disease
Congestive heart disease
Atrial fibrillation
Stroke
-0.72
0.74
-0.38
-0.04
1.66
0.17
-1.89,0.45
-0.86, 2.37
-1.70,0.96
-0.39, 0.30
0.03, 3.32
-0.28, 0.62
Other diseases
Diabetes
Inflammatory diseases
0.19
0.18
-0.46, 0.84
-1 .09, 1 .46
        'These estimates represent the additional percent change in mortality for persons who had the characteristic being examined compared to
      persons who did not have the characteristic, when the mean 03 level of the previous 3 days increased 10 ppb. These values were not standardized
      because they do not represent the actual effect estimate for the characteristic being evaluated, but instead, the difference between effect estimates
      for persons with versus without the condition.
      ""Compared with males in the same age group.
        Source: Reprinted with permission from Lippincott Williams & Wilkins, Medina-Ramon and Schwartz (2008).
 1                    Additionally, Cakmak et al. (2011) examined the effect of individual-level characteristics
 2                    that may modify the O3-mortality relationship in 7 Chilean cities. In a time-series analysis
 3                    using a constrained distributed lag of 0-6 days, Cakmak et al. (2011) found evidence for
 4                    larger O3 mortality effects in individuals > 75 years of age compared to younger ages,
 5                    which is similar to Medina-Ramon and Schwartz (2008) and Stafoggia et al.  (2010).
 6                    Unlike the  studies discussed above O3-mortality risk estimates were found to be slightly
 7                    larger in males (3.71% [95% CI: 0.79, 6.66] for a 40 ppb increase in max 8-h avg O3
 8                    concentrations), but were not significantly different than those observed for females
 9                    (3.00% [95% CI: 0.43, 5.68]). The  major focus of Cakmak et al. (2011) is the
10                    examination of the influence of SES indicators (i.e., educational attainment, income level,
11                    and employment  status)  on the O3-mortality relationship. The authors found the largest
12                    risk estimates in the lowest SES categories for each of the indicators examined this
13                    includes: primary school not completed when examining educational attainment; the
14                    lowest quartile of income level; and unemployed individuals when comparing
15                    employment status.

16                    Overall, uncertainties exist in the interpretation of the potential effect modifiers,
17                    identified in Medina-Ramon and Schwartz (2008). Stafoggia et al. (2010).  and Cakmak et
      Draft - Do Not Cite or Quote                        6-209                                  September 2011

-------
 1                   al. (2011) of the O3-mortality relationship due to the expected heterogeneity in O3 mortal-
 2                   ity risk estimates across cities as highlighted in Smith et al. (2009b) (Figure 6-28) and
 3                   Franklin and Schwartz (2008) (Figure 6-29). For example, it is difficult to determine the
 4                   relative importance of a susceptibility factor that results in an additional percent increase
 5                   in mortality in a multicity analysis when analyses of the individual cities within the study
 6                   did not indicate associations between O3 and mortality. In addition, it is likely that
 7                   individual-level susceptibility factors identified in Medina-Ramon and Schwartz (2008).
 8                   Stafoggia et al. (2010). and Cakmak et al. (2011) only modify the O3-mortality relation-
 9                   ship. The factors identified span pollutants as is evident by older adults (i.e., > 65) often
10                   being identified as an effect modifier of PM mortality risk estimates (U.S. EPA. 2009d).


                     Community-level Characteristics

11                   Several studies also examined city-level (i.e., ecological) variables to explain city-to-city
12                   variation in estimated O3 mortality risk  estimates. Bell and Dominici (2008) investigated
13                   whether community-level characteristics, such as race, income, education, urbanization,
14                   transportation use, PM and O3 levels, number of O3 monitors, weather, and air
15                   conditioning use could explain the heterogeneity in O3-mortality risk estimates across
16                   cities. The authors analyzed 98 U.S. urban communities from NMMAPS for the period
17                   1987-2000. In the all-year regression model that included no community-level variables,
18                   a 20 ppb increase in 24-h avg O3 concentrations during the previous week was associated
19                   with a 1.04% (95% CI: 0.56, 1.55) increase in mortality. Bell and Dominci (2008) found
20                   that higher O3.mortality effect estimates were associated with higher: percent
21                   unemployment, fraction of the population Black/African-American, percent of the
22                   population that take public transportation to work; and with lower: temperatures and
23                   percent of households with central air conditioning (Figure 6-31). The modification of
24                   O3-mortality risk estimates reported for city-specific temperature and prevalence of
25                   central air conditioning in this analysis confirm the result from the meta-analyses
26                   reviewed in the 2006 O3 AQCD.

27                   The APHENA project (Katsouvanni et al.. 2009) examined potential effect modification
28                   of O3 risk estimates in the  Canadian, European, and U.S. data sets using a consistent set
29                   of city-specific variables. Table 6-47 presents the results from all age analyses for all-
30                   cause mortality using all-year O3 data for the average of lag 0-1 day. While there are
31                   several significant effect modifiers in the U.S. data, the results are mostly inconsistent
32                   with the results from the Canadian and European data sets. The positive effect
33                   modification by percentage unemployed and the negative effect modification by mean
34                   temperature (i.e., a surrogate for air conditioning rate) are consistent with the results
35                   reported by Bell and Dominici (2008) discussed above. However, the lack of consistency
36                   across the data sets, even between the Canadian and U.S. data, makes it difficult to

      Draft - Do Not Cite or Quote                       6-210                                September 2011

-------
1
2
3









interpret the results. Some of these associations may be due to coincidental correlations
with other unmeasured factors that vary regionally (e.g., mean SO2 tend to be higher in
the eastern U.S.).
fl 4-
I g
i§
*.l 2-
fi
I

fl
|s.2J
I* «i
:: "
i§
' *.->, 00° O 0 fc 2-
• sf-£> jQ £2)_£_— -® — *" c S
_^L-22&a~~£>/^rf~°'° °» "6 !
u. " i j i"^ o Co^f a ti y n
O° ?t»o«,to^ &-E
, o^ '  « S -o
°O O oj 0-
o* ° iSjJ
0
o
6 »
* « o» O ° n
o o e 1° 0 W0 O °___— — —
t> lO^xO— °" — ^^ —
"^^>i*rfc * ° ° °
fs^t*T!t> *X«« ».
•7*fv& ^g * o
•#
CO
00 °°
"1 .• °-| .
                           3   4   5   6   7   B
                          Percentage of population unemployed
                10  20  30  40  50
                  Percentage of population
                  Black/African American
                   tl  *
                   ;.e  2 •
                                                      0 -
                                                  e  -2
                          50  55  60   65   70  75
                             Long-term temperature (°F)
                  10   20   30   40   50
                 Percentage of population taking
                  public transportaton to work
                                  r§  4
                                   i
                                  is
                                   S  2-
                                  It
                                       0    20    40    60    80
                                       Percentage of households with central AC
 Source: Reprinted with permission of Johns Hopkins Bloomberg School of Public Health, Bell and Dominici (2008).

Figure 6-31    Ozone mortality risk estimates and community-specific
                characteristics, U.S., 1987-2000.  The size of each circle
                corresponds to the inverse of the standard error of the
                community's maximum likelihood estimate. Risk estimates are for a
                10 ppb increase in 24-h avg ozone concentrations during the
                previous week.
Draft - Do Not Cite or Quote
6-211
September 2011

-------
Table 6-47 Percent change in all-cause mortality, for all ages, associated with
a 40ppb increase in 1-h max ozone concentrations at Lag 0-1 at the
25th and 75th percentile of the center-specific distribution of
selected effect modifiers
Canada
Effect
Modifier
N02CV
Mean S02
03CV
Mean
N02/PM10
Mean
Temperature
% > 75 yr
Age
standardized
Mortality
%
Unemployed
25th Percentile
Estimate
(95% Cl)
3.10
(1 .90, 4.40)
2.22
(0.71 , 3.83)
2.86
(0.79, 5.05)
3.91
(2.54, 5.29)
2.86
(0.95, 4.72)
2.22
(0.79, 3.58)
2.62
(0.79, 4.48)
2.78
(1.42,4.07)
75th
Percentile t
Estimate Value
(95% Cl)
3.99 1 .33
(2.38, 5.62)
4.72 2.16
(2.94, 6.61)
3.50 0.60
(2.14,4.89)
2.54 -1.58
(0.95,4.15)
3.50 0.83
(2.22, 4.89)
4.23 2.68
(3.02, 5.54)
4.07 1.14
(2.22, 5.87)
3.75 1 .88
(2.54, 4.89)
Europe
25th Percentile
Estimate
(95% Cl)
1.66
(0.71,2.62)
1.58
(0.47, 2.62)
2.62
(1 .50, 3.75)
1.74
(0.87, 2.70)
1.58
(0.39, 2.86)
1.50
(0.55, 2.46)
1.10
(-0.16,2.38)
1.42
(-0.47, 3.34)
75th
Percentile t
Estimate Value
(95% Cl)
1 .34 -0.49
(-0.08,
2.78)
1.66 0.16
(0.39, 2.86)
1.10 -2.65
(0.24, 1 .98)
1 .50 -0.43
(0.47, 2.62)
1 .58 -0.04
(0.31,2.78)
1 .82 0.52
(0.55,3.10)
1 .98 1 .07
(0.79, 3.26)
1 .34 -0.07
(-0.47,
3.18)
U
25th Percentile
Estimate
(95% Cl)
1.26
(0.47, 1 .98)
0.47
(-0.47, 1 .42)
0.16
(-0.70,1.10)
-0.08
(-1 .02, 0.95)
2.14
(1.34,2.94)
1.02
(0.24, 1 .90)
0.00
(-0.94, 0.87)
0.16
(-0.78,1.18)
.S.
75th
Percentile
Estimate
(95% Cl)
0.08
(-0.78,
0.95)
1.98
(1.10,2.94)
1.50
(0.71,2.22)
1.26
(0.47, 2.06)
0.00
(-0.78,
0.79)
1.02
(0.31,1.74)
1.58
(0.87, 2.38)
1.50
(0.71,2.30)

t
Value
-2.87
2.79
2.68
2.64
-4.40
-0.02
3.81
2.45
 Source: Adapted with permission of Health Effects Institute, Katsouyanni et al. (2009).
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
               Regional Pattern of Ozone-Mortality Risk Estimates

               In addition to examining whether individual- and community-level factors modify the
               O3.mortality association, studies also examined whether these associations varied
               regionally within the U.S. Bell and Dominici (2008). in the study discussed above, also
               noted that O3-mortality risk estimates were higher in the Northeast (1.44% [95% Cl: 0.78,
               2.10%]) and Industrial Midwest (0.73% [95% Cl:  0.11, 1.35%]), while null associations
               were observed in the Southwest and Urban Midwest (Table 6-48). The regional
               heterogeneity in O3-mortality risk estimates was further reflected by Bell and Dominici
               (2008) in a map of community-specific Bayesian O3-mortality risk estimates (Figure 6-
               32). It is worth noting that in the analysis of PMi0 using the same data set, Peng et al.
               (2005) also found that both the Northeast and Industrial Midwest showed particularly
Draft - Do Not Cite or Quote
                                                    6-212
September 2011

-------
 1
 2
 3
 4
 5

 6
 7
 8
 9
10
11
12
13
14
15
elevated effects, especially during the summer months. As mentioned above, although no
evidence for confounding of O3 mortality risk estimates by PMi0 was observed, Bell et al.
(2007) did find regional differences in the correlation between O3 and PM10. Thus, the
heterogeneity in O3 mortality risk estimates may need to be examined as a function of the
correlation between PM and O3.

Smith et al. (2009b), as discussed earlier, also examined the regional difference in O3
mortality risk estimates across the same seven regions and similarly found evidence for
regional heterogeneity. In addition, Smith et al. (2009b) constructed spatial maps of the
risk estimates by an extension of a hierarchical model that allows for spatial auto-
correlation among the city-specific random effects. Figure 6-31 presents the spatial map
of O3 mortality coefficients from the Smith et al. (2009b) analysis that used 8-h max O3
concentrations during the summer. The results from the Bell and Dominici (2008)
analysis (Figure 6-32) shows much stronger apparent heterogeneity in O3-mortality risk
estimates across cities than the smoothed map from Smith et al. (2009b) (Figure 6-33),
but both maps generally show larger risk estimates in the eastern region of the U.S.
      Table 6-48     Percentage increase in daily mortality for a 10 ppb increase in 24-h
                      avg ozone concentrations during the previous week by geographic
                      region in the U.S., 1987-2000

No. of Communities
Regional Estimate
95% PI*
Regional results
Industrial Midwest
Northeast
Northwest
southern California
Southeast
Southwest
Urban Midwest
20
16
12
7
26
9
7
0.73
1.44
0.08
0.21
0.38
-0.06
-0.05
0.11,1.35
0.78,2.10
-0.92, 1 .09
-0.46, 0.88
-0.07, 0.85
-0.92, 0.81
-1.28,1.19
National results
All continental communities
All communities
97
98
0.51
0.52
0.27, 076
0.28, 0.77
      Source: Used with permission from Johns Hopkins Bloomberg School of Public Health, Bell and Dominici (2008).
      Draft - Do Not Cite or Quote
                              6-213
September 2011

-------
 Source: Reprinted with permission of Johns Hopkins Bloomberg School of Public Health (Bell and Dominici, 2008).

Figure 6-32    Community-specific Bayesian ozone-mortality risk estimates in 98
               U.S. communities.
                                       8H: summer
                                                                        - 1.0

                                                                          0.8

                                                                        - 0.6

                                                                          0.4

                                                                        -0.2

                                                                        -0.0
 Source: Reprinted with permission of Informa UK Ltd. (Smith et al.. 2009b).

Figure 6-33    Map of spatially dependent ozone-mortality coefficients for 8-h max
               ozone concentrations using summer data.
Draft - Do Not Cite or Quote
6-214
September 2011

-------
                     6.6.2.3    Interaction

 1                   The terms effect modification and interaction are often used interchangeably, but
 2                   theoretically they represent different concepts. Although interactions can lead to either
 3                   antagonistic or synergistic effects, most studies attempt to identify potential factors that
 4                   interact synergistically with O3 to increase the risk of mortality. Within this section,
 5                   interactive effects are defined as time-varying covariates, such as temperature and
 6                   copollutants that are included in 1st stage time-series regression models. To date, only a
 7                   few time-series studies have investigated the potential interaction between O3 exposure
 8                   and copollutants or weather variables. This can be attributed to the moderate to high
 9                   correlation between O3 and these covariates, which makes such investigations
10                   methodologically challenging.

11                   Ren et al.  (2008) examined the possible synergistic effect between O3 and temperature on
12                   mortality in the 60 largest eastern U.S. communities from the NMMAPS data during the
13                   warm months (i.e., April to October) from 1987-2000. This analysis was restricted to the
14                   eastern areas of the U.S. (i.e., Northeast, Industrial Midwest and Southeast) because a
15                   previous study which focused specifically on the eastern U.S. found that
16                   temperature-mortality patterns differ between the northeast and southeast regions
17                   possibly due to climatic differences (Curriero et al.. 2002). To examine possible
18                   geographic differences in the interaction between temperature and O3, Ren et al. (2008)
19                   further divided the NMMAPS regions into the Northeast, which included the Northeast
20                   and Industrial Midwest regions (34 cities), and the Southeast, which included the
21                   Southeast region (26 cities). The potential synergistic effects between O3 and temperature
22                   were examined using two different models. Model 1 included an interaction term in a
23                   Generalized Additive Model (GAM) for O3 and maximum temperature (3-day avg values
24                   were used for both terms) to examine the bivariate response surface and the pattern of
25                   interaction between the two variables in each community. Model 2 consisted of a
26                   Generalized Linear Model  (GLM) that used interaction terms to stratify by "low,"
27                   "moderate," and "high" temperature days using the first and third quartiles of temperature
28                   as cut-offs to examine the percent increase in mortality in each community. Furthermore,
29                   a two-stage Bayesian hierarchical model was used to estimate the overall percent increase
30                   in all-cause mortality associated with short-term O3 exposure across temperature levels
31                   and each region using model 2. The same covariates were used in both model 1 and 2.
32                   The bivariate response surfaces from model  1 suggest possible interactive effects
33                   between O3 and temperature although the interpretation of these results is not
34                   straightforward due to the high correlation between these terms. The apparent interaction
35                   between temperature and O3 as evaluated in model 2 varied across geographic regions. In
36                   the northeast region, a 20 ppb increase in 24-h avg O3 concentrations at lag 0-2 was
37                   associated with an increase of 4.49% (95% posterior interval [PI]: 2.39, 6.36%), 6.21%

      Draft - Do Not Cite or Quote                       6-215                                September 2011

-------
 1                   (95% PI: 4.47, 7.66%) and 12.8% (95% PI: 9.77, 15.7%) in mortality at low, moderate
 2                   and high temperature levels, respectively. The corresponding percent increases in
 3                   mortality in the southeast region were 2.27% (95% PI: -2.23, 6.46%) for low temperature,
 4                   3.02% (95% PI: 0.44, 5.70%) for moderate temperature, and 2.60% (95% PI: -0.66,
 5                   6.01%) for high temperature.

 6                   When examining the relationship between temperature and O3-related mortality, the
 7                   results reported by Ren et al. (2008) (i.e., higher O3-mortality risks on days with higher
 8                   temperatures) may appear to contradict the results of Bell and Dominici (2008) described
 9                   earlier (i.e., communities with higher temperature have lower O3-mortality risk
10                   estimates). However, the observed difference in results can be attributed to the
11                   interpretation of effect modification in a second-stage regression which uses long-term
12                   average temperatures, as was performed by Bell and Dominici (2008). compared to a
13                   first-stage  regression that examines the interaction between daily temperature and O3-
14                   related mortality. In this  case, the second-stage regression results from Bell and Dominici
15                   (2008) indicate that a city with lower temperatures, on average, tend to show a stronger
16                   O3 mortality effect, whereas, in the first-stage regression performed by Ren et al. (2008).
17                   the days with higher temperature tend to show a larger O3-mortality effect. This observed
18                   difference may in part reflect the higher air conditioning use in communities with higher
19                   long-term  average temperatures. Therefore, the findings from Ren et al. (2008) indicating
20                   generally lower O3 risk estimates in the southeast region where the average temperature is
21                   higher than in the northeast region is consistent with the regional results reported by Bell
22                   and Dominici (2008). As demonstrated by the results from both Ren et al. (2008) and
23                   Bell and Dominici (2008) caution is required when interpreting results from studies that
24                   examined  interactive effects using  two different approaches because potential effect
25                   modification as suggested in a second-stage regression generally does not provide
26                   evidence for a short-term interaction examined in a first-stage regression. Overall, further
27                   examination of the potential interactive (synergistic) effects of O3 and covariates in time-
28                   series regression models is required to more clearly understand the factors that may
29                   influence O3 mortality risk estimates.
                     6.6.2.4    Evaluation of the Ozone-Mortality C-R Relationship and
                                Related Issues

30                   Evaluation of the O3-mortality concentration-response relationship is not straightforward
31                   because the evidence from multicity studies (using log-linear models) suggests that
32                   O3-mortality associations are highly heterogeneous across regions. In addition, there are
33                   numerous issues that may influence the shape of the  O3-mortality concentration-response
34                   relationship that warrant examination including: multi-day effects (distributed lags),
      Draft - Do Not Cite or Quote                       6-216                                September 2011

-------
 1                   potential adaptation, mortality displacement (i.e., hastening of death by a short period),
 2                   and the exposure metric used to compute risks (e.g., 1-h daily max versus 24-h avg). The
 3                   following section presents the recent studies identified that conducted an initial
 4                   examination of these issues.
                     Multiday Effects, Mortality Displacement, and Adaptation

 5                   The pattern of positive lagged associations followed by negative associations in a
 6                   distributed lag model may be considered an indication of "mortality displacement" (i.e.,
 7                   deaths are occurring in frail individuals and exposure is only moving the day of death to a
 8                   day slightly earlier). Zanobetti and Schwartz (2008b) examined this issue in 48 U.S. cities
 9                   during the warm season (i.e., June-August) for the years  1989-2000. In an initial analysis,
10                   the authors applied a GLM to examine same-day O3-mortality effects, and in the model
11                   included an unconstrained distributed lag for apparent temperature to take into account
12                   the effect of temperature on the day death occurred and the previous 7 days. To  examine
13                   mortality displacement Zanobetti and Schwartz (2008b) refit models using two
14                   approaches: an unconstrained and a smooth distributed lag each with 21-day lags for O3.
15                   In this study, all-cause mortality as well as cause-specific mortality (i.e., cardiovascular,
16                   respiratory, and stroke) were examined for evidence of mortality displacement. The
17                   authors found a 0.96% (95% CI: 0.60, 1.30%) increase in all-cause mortality across all 48
18                   cities for a 30 ppb increase in 8-h max O3 concentrations at lag 0 whereas the combined
19                   estimate of the unconstrained distributed lag model (lag 0-20)  was 1.54% (95%  CI: 0.15,
20                   2.91%). Similarly, when examining the cause-specific mortality results (Table 6-49),
21                   larger risk estimates were observed for the distributed lag model compared to the lag
22                   0 day estimates. However, for stroke a slightly larger effect was observed at lags 4-20
23                   compared to lags 0-3 suggesting a larger window for O3-induced stroke mortality. This is
24                   further supported by the sum of lags 0 through 20 days showing the greatest  effect.
25                   Overall, these results suggest that estimating the mortality risk using a single day of O3
26                   exposure may underestimate the public health impact, but the extent of multi-day effects
27                   appear to be limited to a few days. This is further supported by the shape of the combined
28                   smooth distributed lag (Figure 6-34). It should be noted that the proportion of total
29                   variation in the effect estimates due to the between-cities heterogeneity, as measured by
30                   I2 statistic, was relatively low (4% for the lag 0 estimates and 21% for the distributed
31                   lag), but 21 out of the 48 cities exhibited null or negative estimates. As a result,  the
32                   estimated  shape of the distributed lag  cannot be interpreted as  a general form of lag
33                   structure of associations applicable to all the cities included in this analysis.
      Draft - Do Not Cite or Quote                       6-217                                 September 2011

-------
Table 6-49 Estimated effect of a 10 ppb increase in 8-h max
concentrations on mortality during the summer
day and distributed lag models

% (Percentage)
ozone
months for single-
95% Cl
Total mortality
LagO
Sum lags 0-20
Sum lags 0-3
Sum lags 4-20
0.32
0.51
0.53
-0.02
0.20, 0.43
0.05, 0.96
0.28, 0.77
-0.35, 0.31
Cardiovascular mortality
LagO
Sum lags 0-20
Sum lags 0-3
Sum lags 4-20
0.47
0.49
0.80
-0.23
0.30, 0.64
-0.01 , 1 .00
0.48,1.13
-0.67, 0.22
Respiratory mortality
LagO
Sum lags 0-20
Sum lags 0-3
Sum lags 4-20
0.54
0.61
0.83
-0.24
0.26, 0.81
-0.41 , 1 .65
0.38,1.28
-1.08,0.60
Stroke
LagO
Sum lags 0-20
Sum lags 0-3
Sum lags 4-20
0.37
2.20
0.92
1.26
0.01,0.74
0.76, 3.67
0.26,1.59
0.05, 2.49
       Source: Reprinted with permission from American Thoracic Society, Zanobetti and Schwartz (2008b).
 1                                  Samoli et al. (2009) also investigated the temporal pattern of mortality
 2                   effects in response to short-term exposure to O3 in 21 European cities that were included
 3                   in the APHEA2 project. Using a method similar to Zanobetti and Schwartz (2008b), the
 4                   authors applied unconstrained distributed lag models with lags up to 21 days in each city
 5                   during the summer months (i.e., June through August) to examine the effect of O3 on all-
 6                   cause, cardiovascular, and respiratory mortality. They also applied a generalized additive
 7                   distributed lag model to obtain smoothed distributed lag coefficients. However, unlike
 8                   Zanobetti and Schwartz (2008b). Samoli et al. (2009) controlled for temperature using a
 9                   linear term for humidity and an unconstrained distributed lag model of temperature at
10                   lags 0-3 days. The choice of 0- through 3-day lags of temperature was based on a
11                   previous European multicity study (Baccini et al.. 2008). which suggested that summer
12                   temperature effects last only a few days. Upon combining the individual city estimates
13                   across cities in a second stage regression, Samoli et al. (2009) found that the estimated
14                   effects on respiratory mortality were extended for a period of two weeks. However, for
15                   all-cause and cardiovascular mortality, the 21-day distributed lag models yielded null or
16                   (non-significant) negative estimates (Table 6-50). Figure 6-35 shows the distributed lag
17                   coefficients for all-cause mortality, which exhibit a declining trend and negative
      Draft - Do Not Cite or Quote                       6-218                                September 2011

-------
1
2
3
4
5
coefficients beyond 5-day lags. The authors' interpretation of these results was that
"using single-day exposures may have overestimated the effects on all-cause and
cardiovascular mortality, but underestimated the effects on respiratory mortality." Thus,
the results in part suggest evidence of mortality displacement for all-cause and
cardiovascular mortality.
                           <"*
                           o'
                        0)
                        0>
                        b
                           o
                           o
                           CN|
                           d '
                                                     10
                                                                  15
                                                     Day Lag
                                                                              20
       Source: Reprinted with permission of American Thoracic Society (Zanobetti and Schwartz. 2008b).
       The triangles represent the percent increase in all-cause mortality for a 10 ppb increase in 8-h max ozone concentrations at each
     lag while the shaded areas are the 95% point-wise confidence intervals.

     Figure 6-34    Estimated combined smooth distributed lag for 48 U.S. cities
                      during the summer months.
     Draft - Do Not Cite or Quote
                                6-219
September 2011

-------
      Table 6-50    Estimated percent increase in cause-specific mortality (and 95%
                     CIs) for a 10-ug/m3 increase in maximum 8-h ozone during
                     June-August, for the same day (lag 0), the average of the same and
                     previous day (lag 0-1), the unconstrained distributed lag model for
                     the sum of 0-20 days and the penalized distributed lag model
                     (lag 0-20)


Fixed effects
% (95% Cl)
Random effects
% (95% Cl)
Total mortality
LagO
Average lags 0-1
Sum lags 0-20, unconstrained
Sum lags 0-20, penalized
0.28(0.11,0.45)
0.24(0.15,0.34)
0.01 (-0.40, 0.41)
0.01 (-0.41 , 0.42)
0.28 (0.07, 0.48)
0.22 (0.08, 0.35)
-0.54 (-1.28, 0.20)
-0.56 (-1.30, 0.1 9)
Cardiovascular mortality
LagO
Average lags 0-1
Sum lags 0-20, unconstrained
Sum lags 0-20, penalized
0.43(0.18,0.69)
0.33(0.19,0.48)
-0.33 (-0.93, 0.29)
-0.32 (-0.92, 0.28)
0.37 (0.05, 0.69)
0.25 (0.03, 0.47)
-0.62 (-1.47, 0.24)
-0.57 (-1.39, 0.26)
Respiratory mortality
LagO
Average lags 0-1
Sum lags 0-20, unconstrained
Sum lags 0-20, penalized
0.36 (-0.21, 0.94)
0.40(0.11,0.70)
3.35(1.90,4.83)
3.66 (2.25, 5.08)
0.36 (-0.21 , 0.94)
0.40(0.11,0.70)
3.35(1.90,4.83)
3.66 (2.25, 5.08)
       Source: Used with permission from BMJ Group (Samoli etal., 2009).
 1                 Although the APHENA project (Katsouyanni et al.. 2009) did not specifically investigate
 2                 mortality displacement and therefore did not consider longer lags (e.g., lag > 3 days), the
 3                 study did present O3 risk estimates for lag 0-1, lag 1, and a distributed lag model of 0-
 4                 2 days in the Canadian, European, and U.S. datasets. Katsouyanni et al. (2009) found that
 5                 the results vary somewhat across the regions, but, in general, there was no indication that
 6                 the distributed lag model with up to a 2-day lag yielded meaningfully larger O3 mortality
 7                 risk estimates than the lag 0-1 and lag 1 results. For example, for all-cause mortality,
 8                 using the model with natural splines and 8 df/year to adjust for seasonal trends, a reported
 9                 percent excess risk for mortality for a 40 ppb increase in  1-h max O3 concentrations for
10                 lag 0-1, lag 1, and the distributed lag model (lag 0-2) was 2.70% (95% Cl: 1.02, 4.40%),
11                 1.42% (95% Cl: 0.08, 2.78%), and 3.02% (95% Cl: 1.10, 4.89%), respectively. Thus, the
12                 observed associations appear to occur over a short time period, (i.e., a few days).
13                 Similarly, the Public Health and Air Pollution in Asia (PAPA) study (Wong etal.. 2010)
14                 also examined multiple lag days (i.e., lag 0, lag 0-1, and lag 0-4), and although it did not
15                 specifically examine mortality displacement it does provide additional evidence
16                 regarding the timing of mortality effects proceeding O3 exposure. In a combined analysis
17                 using data from all four cities examined (Bangkok, Hong Kong, Shanghai, and Wuhan),
18                 excess risk estimates at lag 0-4 were larger than those at lag 0 or lag 0-1 in both fixed and

      Draft - Do Not Cite or Quote                     6-220                              September 2011

-------
 1                   random effect models (results not presented quantitatively). The larger risk estimates at
 2                   lag 0-4 can primarily be attributed to the strong associations observed in Bangkok and
 3                   Shanghai. However, it is worth noting that Bangkok differs from the three Chinese cities
 4                   included in this analysis in that it has a tropical climate and does not exhibit seasonal
 5                   patterns of mortality. As a result, Wong et al. (2010) examined the O3-mortality
 6                   associations at lag 0-1 in only the three Chinese cities and found that risk estimates were
 7                   slightly reduced from 2.26% (95% CI: 1.36,  3.16) in the 4 city analysis to 1.84% (0.77,
 8                   2.86) in the 3 city analysis for a 30 ppb increase in 8-h max O3 concentrations. Overall,
 9                   the PAPA study further supports the observation of the APHENA study that associations
10                   between O3 and mortality occur over a relatively short-time period, but also indicates that
11                   it may be difficult to interpret O3-mortality associations across cities with different
12                   climates and mortality patterns.

13                   When comparing the studies that explicitly examined the potential for mortality
14                   displacement in the O3-mortality relationship, the results from Samoli et al. (2009), which
15                   provide evidence that suggests mortality displacement, are not consistent with those
16                   reported by Zanobetti and Schwartz (2008b). However, the shapes of the estimated
17                   smooth distributed lag associations are similar (Figure 6-34 and Figure 6-35). A closer
18                   examination of these figures shows that in the European data beyond a lag of 5 days the
19                   estimates remain negative whereas in the U.S. data the results remain near zero for the
20                   corresponding lags. These observed difference could be due the differences in the model
21                   specification between the 2 studies, specifically the use of: an unconstrained distributed
22                   lag model for apparent temperature up to 7 previous days (Zanobetti and Schwartz.
23                   2008b) versus a linear term for humidity and an unconstrained distributed lag model of
24                   temperature up to 3 previous days (Samoli et al.. 2009): and natural cubic splines with
25                   2 df per season (Zanobetti and Schwartz. 2008b) versus dummy variables per month per
26                   year to adjust for season (Samoli et al.. 2009). It is important to note, that these
27                   differences in model specification may have  also influenced the city-to-city variation in
28                   risk estimates observed in these two studies (i.e., homogenous estimates across cities in
29                   Zanobetti and Schwartz (2008b) and heterogeneous estimates across cities in Samoli et
30                   al. (2009). Overall, the evidence of mortality displacement remains unclear, but Samoli et
31                   al. (2009). Zanobetti and Schwartz (2008b). and Katsouyanni et al. (2009) all suggest that
32                   the positive associations between O3 and mortality are observed mainly in the first
33                   few days after exposure.
      Draft - Do Not Cite or Quote                       6-221                                 September 2011

-------
                                                                          20
 Source: Reprinted with permission of BMJ Group (Samoli et al.. 2009).
 The triangles represent the percent increase in all-cause mortality for a 10 |jg/m3 increase in 8-h max ozone concentrations at
each lag; the shaded area represents the 95% CIs.

Figure 6-35   Estimated combined smooth distributed lag in 21 European cities
                during the summer (June-August) months.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
              Adaptation

              Controlled human exposure studies have demonstrated an adaptive response to O3
              exposure for respiratory effects, such as lung function decrements, but this issue has not
              been examined in the epidemiologic investigation of mortality effects of O3. Zanobetti
              and Schwartz (2008a) examined if there was evidence of an adaptive response in the
              O3-mortality relationship in 48 U.S. cities from 1989 to 2000 (i.e., the same data analyzed
              in Zanobetti and Schwartz (2008b). The authors examined all-cause mortality using a
              case-crossover design to estimate the same-day (i.e., lag 0) effect of O3, matched on
              referent days from every-3rd-day in the same month and year as the case. Zanobetti and
              Schwartz (2008a) examined O3-mortality associations by: season, month in the summer
              season (i.e., May through  September), and age categories in the summer season (Table 6-
              51). The estimated O3 mortality risk estimate at lag 0 was found to be highest in the
              summer (1.51%  [95% CI: 1.14, 1.87%]; lag 0 for a 30 ppb increase in 8-h max O3
              concentrations),  and, within the warm months, the association was highest in July (1.96%
Draft - Do Not Cite or Quote
                                                  6-222
September 2011

-------
 1
 2
 3
 4
[95% CI: 1.42, 2.48%]; lag O).1 Upon further examination of the summer months, the
authors also observed diminished effects in August (0.84% [95% CI: 0.33, 1.39%]; lag
0). Based on these results, the authors concluded that the mortality effects of O3 appear
diminished later in the O3 season.
Table 6-51
Percent excess all-cause mortality per 10 ppb increase in daily 8-h
max ozone on the same day, by season, month, and age groups
% 95% CI
BY SEASON
Winter
Spring
Summer
Fall
-0.13
0.35
0.50
0.05
-0.56, 0.29
0.16,0.54
0.38, 0.62
-0.14,0.24
BY MONTH
May
June
July
August
September
0.48
0.46
0.65
0.28
-0.09
0.28, 0.68
0.24, 0.68
0.47, 0.82
0.11,0.46
-0.35,0.16
BY AGE GROUP
0-20
21-30
31-40
41-50
51-60
61-70
71-80
80
0.08
0.10
0.07
0.08
0.54
0.38
0.50
0.29
-0.42, 0.57
-0.67, 0.87
-0.38, 0.52
-0.27, 0.43
0.19,0.89
0.16, 0.61
0.32, 0.67
0.13,0.44
        Source: Reprinted with permission from BioMed Central Ltd. (Zanobetti and Schwartz. 2008a).
 5                   To further evaluate the potential adaptive response observed in Zanobetti and Schwartz
 6                   (2008a) the distribution of the O3 concentrations across the 48 U.S. cities during July and
 7                   August was examined. Both July and August were found to have comparable means of
 8                   48.6 and 47.9 ppb with a reported maximum value of 97.9 and 96.0 ppb, respectively.
 9                   Thus, the observed reduction in O3-related mortality effect estimates in August (0.84%)
10                   compared to July (1.96%) appears to support the existence of an adaptive response.
11                   However, unlike an individual's adaptive response to decrements in lung function from
12                   short-term O3 exposure,  an examination of mortality prevents a direct observation of
13                   adaptation. leather, for mortality the adaptation hypothesis is tested with a tacit
14                   assumption that, whatever the mechanism for O3-induced mortality, the risk of death

        1 These values have been standardized to the increment used throughout the ISA for max 8-h avg increase in O3 concentrations of
      30 ppb. These values differ from those presented in Table 6-47 from Zanobetti and Schwartz (2008a) because the authors
      presented values for a 10 ppb increase in max 8-h  avg O3 concentrations.
      Draft - Do Not Cite or Quote
                                6-223
September 2011

-------
 1                   from short-term O3 exposure is reduced over the course of the summer months through
 2                   repeated exposures. This idea would translate to a smaller population that would die from
 3                   O3 exposure towards the end of summer. This may complicate the interpretation of the
 4                   distributed lag coefficients with long lag periods because the decreased coefficients may
 5                   reflect diminished effects of the late summer, rather than diminished effects that are
 6                   constant across the summer. These inter-twined issues need to be investigated together in
 7                   future research.


                     Ozone-Mortality Concentration-Response Relationship and Threshold
                     Analyses

 8                   Several of the recent studies evaluated have applied a variety of statistical approaches to
 9                   examine the shape of the O3-mortality C-R relationship and whether a threshold exists.
10                   The approach used by Bell et al. (2006) consisted of applying four statistical models to
11                   the NMMAPS data, which included 98 U.S. communities for the period 1987-2000.
12                   These models included: a linear analysis (i.e., any change in O3 concentration can be
13                   associated with mortality) (Model 1); a subset analysis (i.e., examining O3-mortality
14                   relationship below a specific concentration, ranging from 5 to 60 ppb) (Model 2); a
15                   threshold analysis (i.e., assuming that an association between O3 and mortality is
16                   observed above a specific concentration and not below it, using the threshold values set at
17                   an increment of 5 ppb between 0 to 60 ppb and evaluating a presence of a local minima in
18                   AICs computed at each increment) (Model 3); and nonlinear models using natural cubic
19                   splines with boundary knots placed at 0 and 80 ppb, and interior knots placed at 20 and
20                   40 ppb (Model 4). A two-stage Bayesian hierarchical model was used to examine these
21                   models and O3-mortality risk estimates at the city-level in the first stage analysis and
22                   aggregate estimates across cities in the 2nd stage analysis using the average of 0- and
23                   1-day lagged 24-h avg O3 concentrations. The results from all of these models suggest
24                   that if a threshold exists it does so well below the current O3 NAAQS. When restricting
25                   the analysis to all days when the current 8-h standard (i.e., 84 ppb daily 8-h max) is met
26                   in each community, Bell et al. (2006) found there was still a 0.60% (95% PI: 0.30,
27                   0.90%) increase in mortality per 20 ppb increase in 24-h avg O3 concentrations at lag 0-1.
28                   Figure 6-36 shows the combined C-R curve obtained using the nonlinear model (Model
29                   4). Although these results suggest the lack of threshold in the O3.mortality relationship, it
30                   is difficult to interpret such a curve because it does not take into consideration the
31                   heterogeneity in O3-mortality risk estimates across cities.
      Draft - Do Not Cite or Quote                      6-224                                September 2011

-------
                         .—
                         £•
                         o
                         E
                         o
                         to
                         re
                         O)
                         01
                         0-
       Source: Bell et al. (2006).
                                         Centra I estimate
                                         95% posterior interval
                                0           20         40          60          80
                                Average of same and previous days' 0, (ppb)
      Figure 6-36   Estimated combined C-R curve for ozone and nonaccidental
                      mortality using the nonlinear (spline) model.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
Using the same NMMAPS dataset as Bell et al. (2006). Smith et al. (20091)) further
examined the O3-mortality C-R relationship. Similar to Bell et al. (2006). Smith et al.
(2009b) conduct a subset analysis, but instead of restricting the analysis to days with O3
concentrations below a cutoff the authors only include days above a defined cutoff in the
analysis. The results of this "reversed subset" approach are in line with those reported by
Bell et al. (2006); consistent positive associations at all cutoff points up to a defined
concentration where the total number of days with O3 concentrations above a value are so
limited that the variability around the central estimate is increased. In the Smith et al.
(2009b) analysis this observation was initially observed at 45 ppb, with the largest
variability at 60 ppb; however, unlike Bell et al. (2006) where 73% of days are excluded
when subsetting the data to less than 20 ppb, the authors do not detail the number of days
of data included in the subset analyses at higher concentrations. In addition to the subset
analysis, Smith et al. (2009b) examined the shape of the C-R curve using a piecewise
linear approach with cutpoints  at 40 ppb, 60 ppb, and 80 ppb. Smith et al. (2009b) found
that the shape of the C-R curve is similar to that reported by Bell et al. (2006) (Figure
6-36), but argue that slopes of the (3 for each piece of the curve are highly variable with
      Draft - Do Not Cite or Quote
                              6-225
September 2011

-------
 1                   the largest variation in the 60-80 ppb range. However, the larger variability around the (3
 2                   between 60-80 would be expected due to the small number of days with O3
 3                   concentrations within that range in an all-year analysis. This result is consistent with that
 4                   observed by Bell et al. (2006). which is presented in Figure 6-36.

 5                   The APHENA project (Katsouyanni et al.. 2009) also analyzed the Canadian and
 6                   European datasets (the U.S. data were analyzed for PM10 only) for evidence of a
 7                   threshold, using the threshold analysis method (Model 3) applied in Bell et al.'s (2006)
 8                   study described above. There was no evidence of a threshold in the Canadian data (i.e.,
 9                   the pattern of AIC values for each increment of a potential threshold value varied across
10                   cities, most of which showed no local minima). Likewise, the threshold analysis
11                   conducted using the European data also showed no evidence of a threshold.

12                   The PAPA study, did not examine whether a threshold exists in the O3-mortality C-R
13                   relationship, but instead the shape of the C-R curve individually for each city (Bangkok,
14                   Hong Kong, Shanghai, and Wuhan) (Wong et al.. 2010). Using a natural spline smoother
15                   with 3df for the O3 term, Wong et al. (2010) examined whether non-linearity was present
16                   by testing the change in deviance between this smoothed, non-linear, model and an
17                   unsmoothed, linear, model with 1 df. For each of the cities, both across the full range of
18                   the O3 distribution and specifically within the range of the 25th to 75th percentile of each
19                   city's O3 concentrations (i.e., a range of 9.7 ppb to 60.4 ppb across the cities) there was
20                   no evidence of a non-linear relationship in the O3-mortality C-R curve. It should be  noted
21                   that the range of the 25th to 75th percentiles in all of the cities, except Wuhan, was lower
22                   than that observed in the U.S. using all-year data where the range from the 25th to 75th
23                   percentiles is 30 ppb to 50 ppb  (Table 3-6).

24                   Additional threshold analyses were conducted using NMMAPS data, by Xia and Tong
25                   (2006) and Stylianou and Nicolich (2009). Both studies used a new statistical approach
26                   developed by Xia and Tong (2006) to examine thresholds in the O3 mortality C-R
27                   relationship. The approach consisted of an extended GAM model, which accounted for
28                   the cumulative and nonlinear effects of air pollution using a weighted cumulative sum for
29                   each pollutant, with the weights (non-increasing further into the past) derived by a
30                   restricted minimization method. The authors did not use the term distributed lag model,
31                   but their model has the form of distributed lag model, except that it allows for nonlinear
32                   functional forms. Using NMMAPS data for 1987-1994 for 3 U.S. cities (Chicago,
33                   Pittsburgh, and El Paso), Xia and Tong (2006) found that the extent of cumulative effects
34                   of O3 on mortality were relatively short. While the authors also note that there was
35                   evidence of a threshold effect around 24-h avg concentrations of 25 ppb, the threshold
36                   values estimated in the analysis were sometimes in the range where data density was low.
37                   Thus, this threshold analysis needs to be replicated in a larger number of cities to confirm
      Draft - Do Not Cite or Quote                       6-226                                September 2011

-------
 1                   this observation. It should be noted that the model used in this analysis did not include a
 2                   smooth function of days to adjust for unmeasured temporal confounders, and instead
 3                   adjusted for season using a temperature term. As a result, these results need to be viewed
 4                   with caution because some potential temporal confounders (e.g., influenza) do not always
 5                   follow seasonal patterns of temperature.

 6                   Stylianou and Nicolich (2009) examined the existence of thresholds following an
 7                   approach similar to Xia and Tong (2006) for all-cause, cardiovascular, and respiratory
 8                   mortality using data from NMMAPS for nine major U.S. cities (i.e., Baltimore, Chicago,
 9                   Dallas/Fort Worth, Los Angeles, Miami, New York, Philadelphia, Pittsburgh, and
10                   Seattle) for the years 1987-2000. The authors found that PM10 and O3 were the two
11                   important predictors of mortality. Stylianou and Nicolich (2009) found that the estimated
12                   O3-mortality risks varied across the nine cities with the models exhibiting apparent
13                   thresholds, in the 10-45 ppb range for O3. However, given the city-to-city variation in
14                   risk estimates, combining the city-specific estimates into an overall estimate complicates
15                   the interpretation of a threshold. Unlike the Xia and Tong (2006) analysis, Stylianou and
16                   Nicolich (2009) included a smooth function of time to adjust for seasonal/temporal
17                   confounding, which could explain the difference in results between the two studies.

18                   In conclusion, the evaluation of the O3-mortality C-R relationship did not find any
19                   evidence that supports a threshold in the relationship between short-term exposure to O3
20                   and mortality within the range of O3 concentrations observed in the U.S. Additionally,
21                   recent evidence suggests that the shape of the O3-mortality C-R curve remains linear
22                   across the full range of O3 concentrations. However, the studies evaluated demonstrated
23                   that the heterogeneity in the O3-mortality relationship across cities (or regions)
24                   complicates the interpretation of a combined C-R curve and threshold analysis. Given the
25                   effect modifiers identified in the mortality analyses that are also expected to vary
26                   regionally (e.g., temperature, air conditioning prevalence), a national or combined
27                   analysis may not be appropriate to identify whether a threshold exists in the  O3-mortality
28                   C-R relationship. Overall, the studies evaluated support a linear O3-mortality C-R
29                   relationship and continue to support the conclusions from the 2006 O3 AQCD, which
30                   stated that "if a population threshold level exists in O3 health effects, it is likely near the
31                   lower limit of ambient O3 concentrations in the United States" (U.S. EPA. 2006b).
                     6.6.2.5    Associations of Cause-Specific Mortality and Short-term
                                Ozone Exposure

32                   In the 2006 O3 AQCD, an evaluation of studies that examined cause-specific mortality
33                   found consistent positive associations between short-term O3 exposure and
      Draft - Do Not Cite or Quote                      6-227                                September 2011

-------
 1                   cardiovascular mortality, with less consistent evidence for associations with respiratory
 2                   mortality. The majority of the evidence for associations between O3 exposure and cause-
 3                   specific mortality were from single-city studies, which had small daily mortality counts
 4                   and subsequently limited statistical power to detect associations.

 5                   New multicity studies evaluated in this review build upon and confirm the associations
 6                   between short-term O3 exposure and cause-specific mortality identified in the 2006 O3
 7                   AQCD (U.S. EPA. 2006b) (Figure 6-37; Table 6-52). In APHENA, a multicontinent
 8                   study that consisted of the NMMAPS, APHEA2 and Canadian multicity datasets,
 9                   consistent positive associations were reported for both cardiovascular and respiratory
10                   mortality in all-year analyses when focusing on the natural spline model with 8 df/year
11                   (Section 6.6.2.1). The associations between O3 exposure and cardiovascular and
12                   respiratory  mortality in all-year analyses were  further supported by the multicity PAPA
13                   study (Wong et al.. 2010). Cardiovascular mortality associations persisted in analyses
14                   restricted to the summer season with evidence  for stronger respiratory mortality
15                   associations compared to those observed in all-year analyses (Figure 6-37; Table 6-52).
16                   Additional  multicity studies from the U.S. (Zanobetti and Schwartz. 2008b) and Europe
17                   (Stafoggia et al.. 2010; Samoli et al.. 2009) that conducted summer season analyses also
18                   found strong associations between O3 exposure and cardiovascular and respiratory
19                   mortality.

20                   Of the studies evaluated, only the APHENA study (Katsouyanni et al..  2009) and an
21                   Italian multicity study (Stafoggia et al.. 2010) conducted an analysis of the potential for
22                   copollutant confounding of the O3 cause-specific mortality relationship. When focusing
23                   on the natural spline model with 8 df/year and  lag 1 results (as discussed in Section
24                   6.6.2.1), the APHENA study found that O3 cause-specific mortality risk estimates were
25                   fairly robust to the inclusion of PMi0 in copollutant models in the European dataset with
26                   more variability in the U.S. and Canadian  datasets (i.e., copollutant risk estimates
27                   increased and decreased for respiratory and cardiovascular mortality). In summer season
28                   analyses cardiovascular O3 mortality risk estimates were robust in the European dataset
29                   and attenuated but remained positive in the U.S. datasets; whereas, respiratory O3
30                   mortality risk estimates were attenuated in the  European dataset and robust in the U.S.
31                   dataset.  The authors did not examine copollutant models during the summer season in the
32                   Canadian dataset (Figure 6-30; Table 6-49). Interpretation of these results requires
33                   caution; however, due to the different PM  sampling schedules employed in each of these
34                   study locations (i.e., primarily every-6th day in the U.S. and Canadian datasets and every-
35                   day in the European dataset). The results of the summer season analyses from the
36                   APHENA study (Katsouyanni et al.. 2009) are consistent with those from a study of 10
37                   Italian cities during the summer months (Stafoggia et al.. 2010). Stafoggia et al. (2010)
38                   found that cardiovascular (14.3% [95% CI: 6.7, 22.4%]) and cerebrovascular (8.5%  [95%
      Draft - Do Not Cite or Quote                       6-228                                September 2011

-------
1
2
3
4
CI: 0.06, 16.3%]) mortality O3 effect estimates were robust to the inclusion of PM10 in
copollutant models (14.3% [95% CI: 6.7, 23.1%] and 7.3% [95% CI: -1.2, 16.3],
respectively), while respiratory mortality O3 effects estimates (17.6% [95% CI: 1.8,
35.5%]) were attenuated, but remained positive (9.2% [95% CI: -6.9, 28.8%]).
Study
Bell etal. (2005; 74345)a
Wong etal. (2010; 732535)
Katsouyanni etal. (2009; 199899)
Gryparis etal. (2004; 57276)a
Samolietal. (2009; 195855)
Zanobettiand Schwartz (2008; 101596)
Stafoggia etal. (2010; 625034)
Katsouyanni etal. (2009; 199899)
Bell etal. (2005; 74345)a
Wong etal. (2010; 732535)
Katsouyanni etal. (2009; 199899)
Gryparis etal. (2004; 57276)a
Zanobettiand Schwartz (2008; 101596)
Katsouyanni etal. (2009; 199899)
Samolietal. (2009; 195855)
Stafoggia etal. (2010; 625034)
Katsouyanni etal. (2009; 199899)
Location
U.S.andnon-U.S.
PAPA (4 cities)
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
21 European cities
21 European cities
48U.S. cities
10 Italian cities
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
U.S.andnon-U.S.
PAPA (4 cities)
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
21 European cities
48U.S. cities
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
21 European cities
10 Italian cities
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
Ages
All
275
<75
All
235
275
<75
All
275
All
235
275
Lag i
NR Cardiovascular ] -•- All-Year

DL(0-2)'b ]-•— ~

DL(0-2)b <-•—
DL(0-2) — | 	 • 	
0-1 i — 0 — Summer
o-i i— O—
o-3 ! -O-
DL(0-2) -j 	 O 	
DL(0-2)b O
DL(0-2) | 	 O 	
DL(0-2)b O
NR Respiratory — •- 	 All-Year
0-1 	 	 	

DL(0-2)b — 	


DL(0-2)b 	 	 	
0-1 | 	 O 	 Summer
0-3 — r —

DL(0-2)b ] — O —
o-i ! — c>—


DL(0-2)b ! 	 O 	
                                                        -10     -5
                                                                     0     5     10     15
                                                                               % Increase
       Effect estimates are for a 20 ppb increase in 24-h avg; 30 in 8-h max; and 40ppb increase in 1-h max ozone concentrations. Red
     = cardiovascular; blue = respiratory; closed circles = all-year analysis; and open circles = summer-only analysis. An "a" represents
     studies from the 2006 ozone AQCD. A "b" represents risk estimates from APHENA-Canada standardized to an approximate IQR of
     5.1 ppb for a 1-h max increase in ozone concentrations (Section 6.2.7.2).


     Figure 6-37    Percent increase in cause-specific mortality.
     Draft - Do Not Cite or Quote
                                 6-229
September 2011

-------
     Table 6-52     Corresponding effect estimates for Figure 6-37
Study
Location
Ages
Lag
Avg Time
"/.Increase (95% Cl)
Cardiovascular
All-year
Bell et al. (2005)a
Wong etal. (2010)
Katsouyanni et al. (2009)
U.S.andnon-U.S.
PAPA (4 cities)
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
All

>75



<75



NR
0-1
DL(0-2)
DL(0-2)
DL(0-2)D
DL(0-2)
DL(0-2)
DL(0-2)
DL(0-2)b
DL(0-2)
24-h avg
8-h max
1-h max







2.23(1.36,3.08)
2.20 (0.06, 4.37)
2.30 (-1.33, 6.04)
8.96(0.75,18.6)
1.1 (0.10,2.20)
2.06 (-0.24, 4.31)
3.83 (-0.1 6, 7.95)
7.03 (-2.71, 17.7)
0.87 (-0.35, 2. 10)
1.98 (-1.09, 5.1 3)
Summer
Gryparis et al. (2004)a
Samoli etal. (2009)
Zanobetti and Schwartz (2008b)
Stafoggiaetal. (2010)
Katsouyanni et al. (2009)
21 European cities
21 European cities
48 U.S. cities
10 Italian cities
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
All


>35
>75



<75



0-1
0-1
0-3
DL(0-5)
DL(0-2)
DL(0-2)
DL(0-2)b
DL(0-2)
DL(0-2)
DL(0-2)
DL(0-2)D
DL(0-2)
8-h max
8-h max
8-h max
8-h max
1-h max







2.7(1.29,4.32)
1.48(0.18,2.80)
2.42(1.45,3.43)
14.3(6.65,22.4)
3.1 8 (-0.47, 6.95)
1 .50 (-2.79, 5.95)
0.1 9 (-0.36, 0.74)
3.67 (0.95, 6.53)
6.78(2.70,11.0)
-1.02 (-4.23, 2.30)
-0.1 3 (-0.55, 0.29)
2.22 (-1.48, 6.04)
Respiratory
All-years
Bell et al. (2005)a
Wong etal. (2010)
Katsouyanni et al. (2009)
U.S.andnon-U.S.
PAPA (4 cities)
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
All





>75



NR
0-1
DL(0-2)
DL(0-2)
DL(0-2)b
DL(0-2)
DL(0-2)
DL(0-2)
DL(0-2)D
DL(0-2)
24-h avg
8-h max
1-h max







0.94 (-1.02, 2.96)
2.02 (-0.41, 4.49)
2.54 (-3.32, 8.79)
1.02 (-11. 9, 15.9)
0.1 3 (-1.60, 1.90)
1.82 (-2.1 8, 6.04)
1.10 (-6.48, 9.21)
-4.61 (-19.3,13.3)
-0.60 (-2.70, 1 .60)
1.10 (-3.48, 5.95)
Summer
Gryparis et al. (2004)a
Zanobetti and Schwartz (2008b)
Katsouyanni et al. (2009)
Samoli etal. (2009)
Stafoggiaetal. (2010)
Katsouyanni et al. (2009)
21 European cities
48 U.S. cities
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
21 European cities
10 Italian cities
APHENA-U.S.
APHENA-Canada
APHENA-Canada
APHENA-Europe
All






>35
>75



0-1
0-3
DL(0-2)
DL(0-2)
DL(0-2)D
DL(0-2)
0-1
DL(0-5)
DL(0-2)
DL(0-2)
DL(0-2)b
DL(0-2)
8-h max
8-h max
1-h max



8-h max
8-h max
1-h max



6.75(4.38,9.10)
2.51 (1.14,3.89)
4.40 (-2. 10, 11.3)
26.1(13.3,41.2)
3.00(1.60,4.50)
3.83 (-1.33, 9.21)
2.38(0.65,4.19)
17.6(1.78,35.5)
4.07 (-4.23, 13.0)
19.5(2.22,40.2)
2.30 (0.28, 4.40)
2.46 (-3.40, 8.62)
       'Studies from the 2006 03 AQCD.
       bRisk estimates from APHENA-Canada standardized to an approximate IQR of 5.1 ppb for a 1 -h max increase in 03 concentrations (Section
     6.2.7.2).
1                   Collectively, the results from the new multicity studies provide evidence of associations
2                   between short-term O3 exposure and cardiovascular and respiratory mortality with
3                   additional evidence indicating these associations persist, and in the case of respiratory
4                   mortality are strengthened, in the summer season. Although copollutant analyses of
5                   cause-specific mortality are limited, the APHENA study found that O3 cause-specific
     Draft - Do Not Cite or Quote
6-230
September 2011

-------
 1                   mortality risk estimates were fairly robust to the inclusion of PM10 in copollutant models
 2                   in the European dataset, which is supported by the results from Stafoggia et al. (2010).
 3                   Additionally, APHENA found that O3 cause-specific mortality risk estimates were
 4                   moderately to substantially sensitive (e.g., increased or attenuated) to inclusion of PMi0
 5                   in the U.S. and Canadian datasets. However, the mostly every-6th-day sampling schedule
 6                   for PMio in the U.S. and Canadian datasets greatly reduced their sample size and limits
 7                   the interpretation of these results.
             6.6.3   Summary and Causal Determination

 8                   The evaluation of new multicity studies that examined the association between short-term
 9                   O3 exposure and mortality found evidence which supports the conclusions of the 2006 O3
10                   AQCD. These new studies reported consistent positive associations between short-term
11                   O3 exposure and all-cause (nonaccidental) mortality, with associations being stronger
12                   during the warm season, as well as additional support for associations between O3
13                   exposure and cardiovascular and respiratory mortality.

14                   New studies further examined potential confounders (e.g., copollutants and seasonality)
15                   of the O3-mortality relationship. Because the PM-O3 correlation varies across regions,
16                   due to the difference in PM chemical constituents, interpretation of the combined effect
17                   of PM on the relationship between O3 and mortality is not straightforward. Unlike
18                   previous studies that were limited to primarily examining the confounding effects of
19                   PM10, the new studies expanded their analyses to include multiple PM indices (e.g., PM10,
20                   PM2 5, and PM components). An examination of copollutant models found evidence that
21                   associations between O3 and all-cause mortality were robust to the inclusion of PM10 or
22                   PM2 5 (Stafoggia etal.. 2010: Katsouvanni et al.. 2009: Bell et al.. 2007). while other
23                   studies found evidence for a modest reduction (-20-30%) when examining PM10 (Smith
24                   etal.. 2009b). Additional evidence suggests potential sensitivity (e.g., increases and
25                   attenuation) of O3 mortality risk estimates to copollutants by age group or cause-specific
26                   mortality (e.g., respiratory and cardiovascular) (Stafoggia et al.. 2010: Katsouvanni et al..
27                   2009). An examination of PM components, specifically sulfate, found evidence for
28                   reductions in O3-mortality risk estimates in copollutant models  (Franklin and Schwartz.
29                   2008). Overall, across studies, the potential impact of PM indices on O3-mortality risk
30                   estimates tended to be much smaller than the variation in O3-mortality risk estimates
31                   across cities suggesting that O3 effects are independent of the relationship between PM
32                   and mortality. Although some studies suggest that O3-mortality risk estimates may be
33                   confounded by PM or its chemical components the interpretation of these results requires
34                   caution due to the limited PM datasets used as a result of the every-3rd- and 6th-day PM
35                   sampling schedule. When examining the potential for seasonal confounding of the


      Draft - Do Not Cite or Quote                       6-231                                September 2011

-------
 1                   O3-mortality relationship it was observed that the extent of smoothing or the methods
 2                   used for adjustment can influence O3 risk estimates because of the opposing seasonal
 3                   trends of O3 and mortality when not instituting enough degrees of freedom to control for
 4                   temporal/seasonal trends (Katsouyanni et al.. 2009).

 5                   The multicity studies evaluated in this review also examined the regional heterogeneity
 6                   observed in O3-mortality risk estimates. These studies provide evidence which suggests
 7                   generally higher O3-mortality risk estimates in northeastern U.S. cities with some regions
 8                   showing no associations between O3 exposure and mortality (e.g., Southwest, Urban
 9                   Midwest) (Smith et al.. 2009b: Bell and Dominici. 2008). Multicity studies that examined
10                   individual- and community-level characteristics identified characteristics that may
11                   explain the observed regional heterogeneity in O3-mortality risk estimates as well as
12                   characteristics of populations potentially susceptible to O3-related health effects. An
13                   examination of community-level characteristics found an increase in the O3-mortality risk
14                   estimates in cities with higher unemployment, percentage of the population
15                   Black/African-American, percentage of the working population that uses public
16                   transportation, lower temperatures, and lower prevalence  of central  air conditioning
17                   (Medina-Ramon and Schwartz. 2008). Additionally, a potential interactive, or synergistic,
18                   effect on the O3-mortality relationship was observed when examining differences in the
19                   O3-mortality association across temperature levels (Renetal.. 2008). An examination of
20                   individual-level characteristics found evidence that older age, female sex, Black race,
21                   having atrial fibrillation, SES indicators (i.e., educational attainment, income level, and
22                   employment status), and out-of hospital deaths, specifically in those individuals with
23                   diabetes, are modify O3-mortality associations (Cakmak et al.. 2011; Stafoggia et al..
24                   2010; Medina-Ramon and Schwartz. 2008). and may increase susceptibility to O3-related
25                   mortality. Overall, additional research is warranted to further confirm whether these
26                   characteristics, individually or in combination, can explain the observed regional
27                   heterogeneity.

28                   Additional  studies were evaluated that examined factors, such as multi-day effects,
29                   mortality displacement, adaptation, and whether a threshold exists in the O3-mortality
30                   relationship, which may influence the shape of the O3-mortality C-R curve. An
31                   examination of multiday effects in a U.S. and European multicity study found conflicting
32                   evidence for mortality displacement, but both studies suggest that the positive
33                   associations between O3 and mortality are observed mainly in the first few days after
34                   exposure (Samoli et al.. 2009; Zanobetti and Schwartz. 2008b). A U.S. multicity study
35                   found evidence of an adaptive response to O3 exposure, with the highest risk estimates
36                   earlier in the O3 season (i.e., July) and diminished effects later (i.e.,  August) (Zanobetti
37                   and Schwartz. 2008a). However, the evidence of adaptive effects has an implication for
38                   the interpretation of multi-day effects, and requires further analysis. Analyses that
      Draft - Do Not Cite or Quote                       6-232                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 9
10
11
12
13
14
15
16
specifically focused on the O3-mortality C-R relationship supported a linear O3-mortality
relationship and found no evidence of a threshold within the range of O3 concentrations
in the U.S., but did observe evidence for potential differences in the C-R relationship
across cities (Katsouyanni et al.. 2009; Stylianou and Nicolich. 2009; Bell et al.. 2006).
Collectively, these studies support the conclusions of the 2006 O3 AQCD that "if a
population threshold level exists in O3 health effects, it is likely near the lower limit of
ambient O3 concentrations in the U.S."

In conclusion, the new epidemiologic studies build upon and confirm the associations
reported in the 2006 O3 AQCD. Additionally, these new studies have provided additional
information regarding key uncertainties previously identified including the potential
confounding effects of copollutants and seasonal trend,  individual- and community-level
factors that may lead to increased risk of O3-induced mortality and the heterogeneity in
O3-mortality risk estimates, and continued evidence  of a linear no-threshold C-R
relationship. Although some uncertainties still remain, the collective body of evidence is
sufficient to conclude that there is likely to be a causal relationship between short-
term O3 exposure and mortality.
17
18
19
         6.7    Overall Summary
The evidence reviewed in this chapter describes the recent findings regarding the health
effects of short-term exposure to ambient O3 concentrations. Table 6-53 provides an
overview of the causal determinations for each of the health categories evaluated.
      Table 6-53     Summary of causal determinations for short-term exposures to
                      ozone
Health Category
Respiratory Effects
Cardiovascular Effects
Central Nervous System Effects
Effects on Liver and Xenobiotic Metabolism
Effects on Cutaneous and Ocular Tissues
Mortality
Causal Determination
Causal relationship
Suggestive of a causal relationship
Suggestive of a causal relationship
Inadequate to infer a causal relationship
Inadequate to infer a causal relationship
Likely to be a causal relationship
      Draft - Do Not Cite or Quote
                              6-233
September 2011

-------
    6.8    References

Adamkiewicz, G: Ebelt, S: Syring, M: Slater, J: Speizer, FE: Schwartz, J: Suh, H: Gold, DR. (2004). Association
        between air pollution exposure and exhaled nitric oxide in an elderly population. Thorax 59: 204-209.
        http://dx.doi.org/10.1136/thorax.2003.006445.
Adams, WC: Schelegle, ES.  (1983). Ozone and high ventilation effects on pulmonary function and endurance
        performance. J Appl Physiol 55: 805-812.
Adams. WC: Ollison. WM. (1997). Effects of prolonged simulated ambient ozone dosing patterns on human
        pulmonary function and symptomatology. In Presented at: 90th annual meeting of the Air & Waste
        Management Association; June; Toronto, Ontario, Canada Pittsburgh, PA: Air & Waste Management
        Association; paper no 97-MP902.
Adams, WC. (1998). Dose-response effect of varied equivalent minute  ventilation rates on pulmonary function
        responses during exposure to ozone. Washington, DC: American Petroleum Institute.
Adams. WC. (2002). Comparison of chamber and face-mask 6.6-hour exposures to ozone on pulmonary
        function and symptoms responses. Inhal Toxicol 14: 745-764.
Adams. WC. (2003a). Comparison of chamber and  face mask 6.6-hour exposure to 0.08 ppm ozone via square-
        wave and triangular profiles on pulmonary responses.  Inhal Toxicol 15: 265-281.
Adams, WC. (2003b). Relation of pulmonary responses induced by 66-h exposures to 008 ppm ozone and 2-h
        exposures to 030 ppm ozone via chamber and face-mask inhalation. Inhal Toxicol 15: 745-759.
Adams. WC. (2006a). Comparison of chamber 6.6-h exposures to 0.04-0.08 PPM ozone via square-wave and
        triangular  profiles on pulmonary responses. Inhal Toxicol 18: 127-136.
        http://dx.doi. org/10.1080/08958370500306107.
Adams. WC. (2006b). Human pulmonary responses with 30-minute time intervals of exercise and rest when
        exposed for 8 hours to 0.12 ppm ozone via square-wave and acute triangular profiles. Inhal Toxicol 18:
        413-422. http://dx.doi.org/10.1080/08958370600563599.
Aibo. PI: Birmingham. NP: Lewandowski. R: Maddox. JF: Roth. RA: Ganev. PE: Wagner. JG: Harkema. JR.
        (2010). Acute exposure to ozone exacerbates acetaminophen-induced liver injury in mice. Toxicol Sci
        115: 267-285. http://dx.doi.org/10.1093/toxsci/kfq034.
Alexeeff. SE: Litoniua. AA: Suh.  H: Sparrow. D: Vokonas. PS: Schwartz. J. (2007). Ozone exposure and lung
        function: Effect modified by obesity and airways hyperresponsiveness in the VA Normative Aging Study.
        Chest 132: 1890-1897.  http://dx.doi.org/10.1378/chest.07-1126.
Alexeeff. SE: Litoniua. AA: Wright. RO: Baccarelli. A: Suh. H: Sparrow.  D: Vokonas. PS: Schwartz. J. (2008).
        Ozone exposure, antioxidant genes, and lung function in an elderly cohort: VA Normative Aging Study.
        Occup Environ Med 65: 736-742. http://dx.doi.ora/10.1136/oem.2007.035253.
Alexis. N: Urch. B:  Tarlo. S: Corey. P: Pengellv. D: O'Bvrne. P:  Silverman. F. (2000). Cyclooxygenase
        metabolites play a different role in ozone-induced pulmonary function decline in asthmatics compared
        to normals. Inhal Toxicol 12: 1205-1224.
Alexis. NE: Zhou. H: Lav. JC: Harris. B: Hernandez. ML: Lu. TS: Bromberg. PA: Diaz-Sanchez. D: Devlin. RB:
        Kleeberger,  SR; Peden, DB. (2009). The glutathione-S-transferase Mu 1 null genotype modulates
        ozone-induced airway inflammation in human subjects. J Allergy Clin Immunol 124: 1222-1228.
        http://dx.doi.0rg/10.1016/i.iaci.2009.07.036.
Alexis, NE; Lay, JC; Hazucha, M; Harris, B; Hernandez, ML; Bromberg, PA; Kehrl, H; Diaz-Sanchez, D; Kim, C;
        Devlin. RB: Peden. DB. (2010). Low-level ozone exposure induces airways inflammation and modifies
        cell surface  phenotypes in healthy humans. Inhal Toxicol 22: 593-600.
        http://dx.doi. org/10.3109/08958371003596587.
Alfaro-Rodrfguez. A: Gonzalez-Pina. R. (2005). Ozone-induced paradoxical sleep decrease is related to
        diminished acetylcholine levels in the medial preoptic area in rats. Chem Biol Interact 151: 151-158.
        http://dx.doi.org/S0009-2797(04)00162-0 fpiil 10.1016/j.cbi.2004.10.001.
Anderson. HR: Armstrong. B: Haiat. S: Harrison. R:  Monk. V: Poloniecki. J: Timmis. A: Wilkinson. P. (2010). Air
        pollution and activation of implantable cardioverter defibrillators in London. Epidemiology 21: 405-413.
        http://dx.doi.org/10.1097/EDE.Ob013e3181d61600.
Angoa-Perez. M: Jiang. H: Rodrfguez. Al: Lemini. C: Levine. RA: Rivas-Arancibia. S. (2006). Estrogen
        counteracts ozone-induced oxidative stress and nigral neuronal death. Neuroreport 17: 629-633.
Draft - Do Not Cite or Quote                       6-234                                 September 2011

-------
Apte. MG: Buchanan. IS: Mendell. MJ. (2008). Outdoor ozone and building-related symptoms in the BASE
        study. Indoor Air 18: 156-170. http://dx.doi.Org/10.1111/i.1600-0668.2008.00521.x.
Araneda. S: Commin. L: Atlagich. M: Kitahama. K: Parraguez. VH: Pequignot. JM: Dalmaz. Y. (2008). VEGF
        overexpression in the astroglial cells of rat brainstem following ozone exposure. Neurotoxicology 29:
        920-927. http://dx.doi.0rg/10.1016/i.neuro.2008.09.006.
Aranvi. C: Vana. SC: Thomas. PT: Bradof. JN: Renters. JD: Graham. JA: Miller.  FJ. (1983). Effects of
        subchronic exposure to a mixture of O3,  SO2, and (NH4)2SO4 on host defenses of mice. J Toxicol
        Environ Health 12: 55-71.
Arbex. AM: de Souza Conceicao. GM: Perez Cendon. S: Arbex. FF: Lopes. AC: Providello Movses. E: Santiago.
        SL: Saldiva. PHN: Pereira. LAA: Ferreira Braga. AL. (2009). Urban air pollution and COPD-related
        emergency room visits. J Epidemiol Community Health 966: 777-783.
        http://dx.doi.org/10.1136/iech.2008.078360.
Aris. RM: Tager. I: Christian. D: Kelly.  T: Balmes.  JR. (1995). Methacholine responsiveness is not associated
        with OS-induced decreases in FEV1. Chest 107: 621-628.
Arito. H: Uchivama. I: Arakawa. H: Yokovama. E.  (1990). Ozone-induced bradycardia and arrhythmia and their
        relation to sleep-wakefulness in rats. Toxicol Lett 52:  169-178. http://dx.doi.org/10.1016/0378-
        4274(90)90151-6.
Arito. H: Uchivama. I: Yokovama. E. (1992). Acute effects of ozone on EEC activity, sleep-wakefulness and
        heart rate in rats. Ind Health 30: 23-34.
Arito, H: Takahashi, M: Iwasaki, T: Uchivama, I. (1997). Age-related changes in ventilatory and heart rate
        responses to acute ozone exposure in the conscious rat. Ind Health 35: 78-86.
Armstrong. BG. (2003). Fixed factors  that modify the effects of time-varying factors: Applying the case-only
        approach. Epidemiology 14: 467-472.
Atkinson. RW: Bremner. SA: Anderson. HR:  Strachan.  DP:  Bland. JM: Ponce de Leon. A. (1999). Short-term
        associations between emergency hospital admissions for respiratory and cardiovascular disease and
        outdoor air pollution in London. Arch Environ Occup Health 54: 398-411.
ATS. (American Thoracic Society). (1991). Lung function testing: Selection of reference values and
        interpretative strategies. Am J Respir Grit Care Med 144: 1202-1218.
ATS. (American Thoracic Society). (2000a). Guidelines for methacholine and exercise challenge testing-1999.
        Am J Respir Grit Care Med 161: 309-329.
Avila-Costa. MR: Colin-Barengue. L: Fortoul. Tl: Machado-Salas. JP: Espinosa-Villanueva. J: Rugerio-Vargas.
        C: Rivas-Arancibia,  S. (1999). Memory deterioration in an oxidative stress model and its correlation
        with cytological changes on rat hippocampus CA1. Neurosci Lett 270: 107-109.
Avissar. NE: Reed. CK: Cox. C:  Frampton. MW: Finkelstein. JN. (2000). Ozone, but not nitrogen dioxide,
        exposure decreases glutathione peroxidases in epithelial lining fluid of human lung. Am J Respir Crit
        Care Med 162: 1342-1347.
Avol. EL: Linn. WS: Venet. TG: Shamoo. DA: Hackney. JD. (1984). Comparative respiratory effects of ozone
        and ambient oxidant pollution exposure during heavy exercise. J Air Waste Manag Assoc 34: 804-809.
Avol. EL: Trim. SC: Little. DE: Spier. CE: Smith. MN: Peng.  RC:  Linn. WS: Hackney. JD: Gross.  KB: D'Arcv. JB:
        Gibbons. D: Higgins. ITT. (1990). Ozone exposure and lung function in children attending a southern
        California summer camp. In Proceedings of the 83rd  A&WMAAnnual Meeting (Vol. 8, pp. 90-150.153).
        Pittsburgh, PA: Air & Waste Management Association.
Avol. EL: Trim. SC: Little. DE: Spier. CE: Smith. MN: Peng.  RC:  Linn. WS: Hackney. JD. (1991). Ozone
        exposure and lung function: A southern California summer camp study. In RL Berglund; DR Lawson; DJ
        McKee (Eds.), Tropospheric ozone and the environment: Papers from an international conference;
        March 1990; Los Angeles,  CA(pp. 90-99). Los Angeles, CA: Air & Waste  Management Association.
Avol, EL; Navidi, WC; Rappaport, EB; Peters, JM. (1998a). Acute effects of ambient ozone on asthmatic,
        wheezy, and healthy children. (82). Topsfield, MA:  Health Effects Institute; Flagship Press.
Azevedo. JM: Goncalves. FL: de Fatima Andrade. M. (2011).  Long-range ozone transport and its impact on
        respiratory and cardiovascular health in the north of Portugal. Int J Biometeorol 55:  187-202.
        http://dx.doi. org/10.1007/s00484-010-0324-2.
Baccarelli. A: Zanobetti. A: Martinelli.  I: Grillo. P: Hou. L: Lanzani. G: Mannucci.  PM: Bertazzi. PA: Schwartz. J.
        (2007). Air pollution, smoking, and plasma homocysteine. Environ Health Perspect 115: 176-181.
Draft - Do Not Cite or Quote                       6-235                                 September 2011

-------
Baccini. M: Bigger!. A: Accetta. G: Kosatskv. T: Katsouvanni. K: Analitis. A: Anderson. HR: Bisanti. L: D'lppoliti.
        D: Danova, J: Forsberg, B: Medina, S: Paldy, A: Rabczenko, D: Schindler, C: Michelozzi, P. (2008).
        Heat effects on mortality in 15 European cities. Epidemiology 19: 711-719.
        http://dx.doi. org/10.1097/EDE.Ob013e318176bfcd.
Baja, ES: Schwartz, JD: Wellenius, GA: Coull, BA:  Zanobetti, A: Vokonas, PS: Suh, HH. (2010). Traffic-related
        air pollution and QT interval:  Modification by diabetes, obesity, and oxidative stress gene
        polymorphisms in the Normative Aging Study. Environ Health Perspect 118: 840-846.
        http://dx.doi.org/10.1289/ehp.0901396.
Balbi. B: Pignatti. P: Corradi. M: Baiardi. P: Bianchi. L: Brunetti. G: Radaeli. A: Moscato. G: Mutti. A: Spanevello.
        A: Malerba. M. (2007). Bronchoalveolar lavage, sputum and exhaled clinically relevant inflammatory
        markers: Values in healthy adults. Eur Respir J 30: 769-781.
        http://dx.doi.ora/10.1183/09031936.00112306.
Ballester. F: Saez. M: Daponte. A: Ordonez. JM: Taracido. M: Cambra. K: Arribas. F: Bellido. JB: Guillen. JJ:
        Aguinaga,  I: Canada, A: Lopez, E: Iniguez, C: Rodriguez, P: Perez-Hoyos, S: Barcelo, MA: Ocana, R:
        Aranguez.  E. (2005). [The EMECAS Project: Spanish multicentre study on short-term health effects of
        air pollution]. Rev Esp Salud Publica 79: 229-242.
Ballester, F: Rodriguez, P: Iniguez, C: Saez, M: Daponte, A:  Galan, I: Taracido, M: Arribas, F: Bellido, J: Cirarda,
        FB: Canada. A: Guillen. JJ: Guillen-Grima. F: Lopez. E: Perez-Hovos. S: Lertxundi. A: Toro. S. (2006).
        Air pollution and cardiovascular admisisons association in Spain: Results within the EMECAS project. J
        Epidemiol Community Health 60: 328-336.
Balmes. JR: Chen.  LL: Scannell. C: Tager. I: Christian. D: Hearne. PQ: Kelly. T: Aris. RM.  (1996). Ozone-
        induced decrements in FEV1 and FVC do not correlate with measures of inflammation. Am J Respir
        Grit Care Med 153: 904-909.
Balmes. JR: Aris. RM: Chen. LL: Scannell. C: Tager. IB: Finkbeiner. W: Christian. D: Kelly. T: Hearne. PQ:
        Ferrando. R: Welch. B. (1997). Effects of ozone on normal and potentially sensitive human subjects.
        Part I: Airway inflammation and responsiveness to ozone in normal and asthmatic subjects. Boston,
        MA: Health Effects Institute.
Barnes. PJ: Liew. FY. (1995). Nitric oxide and asthmatic inflammation. Immunol Today 16: 128-130.
        http://dx.doi.org/10.1016/0167-5699(95)80128-6.
Barnett. AG: Williams. GM: Schwartz. J: Best. TL: Neller. AH: Petroeschevskv. AL: Simpson. RW. (2006). The
        effects of air pollution on hospitalizations for cardiovascular disease in elderly people in Australian and
        New Zealand cities. Environ  Health Perspect 114: 1018-1023.
Barraza-Villarreal. A: Sunver. J:  Hernandez-Cadena. L: Escamilla-Nunez. MC: Sienra-Monge. JJ: Ramirez-
        Aguilar. M: Cortez-Lugo. M: Holguin. F:  Diaz-Sanchez. D: Olin. AC: Romieu. I. (2008). Air pollution,
        airway inflammation, and lung function in a cohort study of Mexico City schoolchildren. Environ Health
        Perspect 116: 832-838. http://dx.doi.org/10.1289/ehp. 10926.
Basha. MA: Gross.  KB: Gwizdala. CJ: Haidar. AH:  Popovich. J. Jr. (1994). Bronchoalveolar lavage neutrophilia
        in asthmatic and healthy volunteers after controlled  exposure to ozone and filtered purified air. Chest
        106: 1757-1765.
Bauer. AK: Rondini. EA: Hummel. KA: Degraff. LM: Walker. C: Jedlicka. AE: Kleeberger. SR. (2011).
        Identification of candidate genes downstream of TLR4 signaling after ozone exposure in mice: A role
        for heat shock protein 70. Environ Health Perspect 119: 1091-1097.
        http://dx.doi. org/10.1289/ehp. 1003326.
Bell, ML: McDermott, A: Zeger, Sl_: Samet, JM: Dominici, F. (2004). Ozone and short-term mortality in 95 US
        urban communities, 1987-2000. JAM A 292: 2372-2378. http://dx.doi. org/10.1001/iama.292.19.2372.
Bell. ML: Dominici.  F: Samet. JM. (2005). A meta-analysis of time-series studies of ozone and  mortality with
        comparison to the national morbidity, mortality, and air pollution study. Epidemiology 16: 436-445.
Bell. ML: Peng. RD: Dominici. F. (2006). The exposure-response curve for ozone and risk of mortality and the
        adequacy of current ozone regulations.  Environ Health Perspect 114: 532-536.
Bell, ML: Kim, JY: Dominici, F. (2007). Potential confounding of particulate matter on the short-term association
        between ozone and mortality in multisite time-series studies. Environ Health Perspect 115: 1591-1595.
        http://dx.doi. org/10.1289/ehp. 10108.
Draft - Do Not Cite or Quote                        6-236                                 September 2011

-------
Bell.  ML: Dominici. F. (2008). Effect modification by community characteristics on the short-term effects of
       ozone exposure and mortality in 98 US communities. Am J Epidemiol 167: 986-997.
       http://dx.doi. orq/10.1093/aie/kwm396.
Bell.  ML: Levy. JK: Lin. Z. (2008). The effect of sandstorms and air pollution on cause-specific hospital
       admissions  in Taipei, Taiwan. Occup Environ Med 65: 104-111.
       http://dx.doi. orq/10.1136/oem.2006.031500.
Bennett. WD: Hazucha. MJ: Folinsbee. LJ: Bromberg. PA: Kissling.  GE:  London. SJ. (2007). Acute pulmonary
       function response to ozone in young adults as a function of body mass index. Inhal Toxicol 19: 1147-
       1154. http://dx.doi.ora/10.1080/08958370701665475.
Beraamaschi. E: De  Palma. G: Mozzoni. P: Vanni. S: Vettori. MV: Broeckaert. F: Bernard. A: Mutti. A. (2001).
       Polymorphism of quinone-metabolizing enzymes and susceptibility to ozone-induced acute effects. Am
       J Respir Grit Care  Med 163:  1426-1431.
Berhane.  K: Zhang. Y: Linn. WS: Rappaport. EB: Bastain. TM: Salam. MT: Islam. T: Lurmann. F: Gilliland. FD.
       (2011). The  effect of ambient air pollution on exhaled nitric  oxide in the Children's Health Study. Eur
       Respir J 37: 1029-1036. http://dx.doi.ora/10.1183/09031936.00081410.
Berkev. CS: Hoaalin. DC: Antczak-Bouckoms. A: Mosteller. F: Colditz. GA. (1998). Meta-analysis of multiple
       outcomes by regression with random effects. Stat Med 17:  2537-2550.
       http://dx.doi. orq/10.1002/(SICI)1097-0258(19981130)17:22<2537::AID-SIM953>3.0.CO:2-C.
Berry. M: Liov. PJ:  Gelperin. K: Buckler. G: Klotz. J. (1991). Accumulated exposure to ozone and measurement
       of health effects in children and counselors at two summer camps. Environ Res 54:  135-150.
Bigger!. A: Baccini. M: Bellini. P: Terracini. B. (2005). Meta-analysis of the Italian studies of short-term effects of
       air pollution  (MISA), 1990-1999. Int J Occup Environ Health 11:  107-122.
Blomberg, A: Mudwav,  IS: Nordenhall, C: Hedenstrom, H: Kelly, FJ: Frew, AJ: Holgate, ST: Sandstrom, T.
       (1999). Ozone-induced lung  function decrements do not correlate with early airway inflammatory or
       antioxidant responses. Eur Respir J 13: 1418-1428.
Bosson, J: Stenfors,  N: Bucht, A: Helleday, R: Pourazar, J: Holgate, ST:  Kelly, FJ:  Sandstrom, T: Wilson, S:
       Frew. AJ: Blomberg. A. (2003). Ozone-induced bronchial epithelial cytokine expression differs between
       healthy and asthmatic subjects. Clin Exp Allergy 33: 777-782.
Boussouar, A: Araneda, S: Hamelin, C: Soulage, C: Kitahama, K:  Dalmaz, Y. (2009). Prenatal ozone exposure
       abolishes stress activation of Fos and tyrosine hydroxylase in the nucleus tractus solitarius of adult rat.
       Neurosci Lett 452: 75-78.
Brauer, M: Blair, J: Vedal, S. (1996). Effect of ambient ozone exposure on lung function in farm workers. Am J
       Respir Grit Care Med 154: 981-987.
Braun-Fahrlander.  C. h: Kunzli. N: Domeniqhetti. G: Carell. CF: Ackermann-Liebrich. U. (1994). Acute effects of
       ambient ozone on  respiratory function of Swiss schoolchildren after a 10-minute heavy exercise.
       Pediatr Pulmonol 17: 169-177.  http://dx.doi.orq/10.1002/ppul. 1950170306.
Brooks. EG. (2010).  Correspondence from Dr. Brooks - Re: Clarifications in 2008 J Occup Environ Med article
Brown, JS: Bateson, TF: McDonnell, WF. (2008). Effects of exposure to 0.06 ppm  ozone on FEV1 in humans: A
       secondary analysis of existing data. Environ Health Perspect 116: 1023-1026.
       http://dx.doi.org/10.1289/ehp.11396.
Brunekreef, B: Hoek, G: Breugelmans, O: Leentvaar,  M. (1994). Respiratory effects of low-level photochemical
       air pollution  in amateur cyclists. Am J Respir Crit Care Med 150: 962-966.
Buadong. D: Jinsart. W: Funatagawa. I: Karita.  K: Yano. E. (2009). Association between PM10 and O3 levels
       and hospital visits for cardiovascular diseases in Bangkok,  Thailand. J Epidemiol 19: 182-188.
       http://dx.doi.org/10.2188/iea.JE20080047.
Burleson. GR: Keves. LL: Stutzman. JD. (1989). Immunosuppression of pulmonary natural killer activity by
       exposure to ozone. Immunopharmacol Immunotoxicol 11: 715-735.
       http://dx.doi.org/10.3109/08923978909005397.
Burnett. R: Raizenne. M: Krewski. D.  (1990). Acute health effects of transported air pollution: A study of children
       attending a residential summer camp. Can J Stat 18: 367-373. http://dx.doi.org/10.2307/3315843.
Burra. TA:  Moineddin. R: Agha. MM: Glazier. RH. (2009). Social disadvantage, air pollution, and asthma
       physician visits in Toronto, Canada. Environ Res 109: 567-574.
       http://dx.doi.0rg/10.1016/i.envres.2009.03.004.
Draft - Do Not Cite or Quote                        6-237                                  September 2011

-------
Bush. ML: Asplund. PT: Miles. KA: Ben-Jebria. A: Ultman. JS. (1996). Longitudinal distribution of O3 absorption
        in the lung: gender differences and intersubject variability. J Appl Physiol 81: 1651-1657.
Cakmak. S: Dales. RE: Judek. S. (2006a). Do gender, education, and income modify the effect of air pollution
        gases on cardiac disease? J Occup Environ Med 48: 89-94.
        http://dx.doi.org/10.1097/01.iom.0000184878.11956.4b.
Cakmak. S: Dales. RE: Judek. S. (2006b). Respiratory health effects of air pollution gases: Modification by
        education and income. Arch Environ Occup Health 61: 5-10.
Cakmak, S: Dales, RE: Angelica Rubio, M: Blanco Vidal, C. (2011). The risk of dying on days of higher air
        pollution among the socially disadvantaged elderly. Environ Res 111: 388-393.
        http://dx.doi.0rg/10.1016/i.envres.2011.01.003.
Calderon Guzman, D: Barragan Mejia, G: Hernandez Garcia, E: Juarez Olguin, H. (2006). Effect of nutritional
        status and ozone exposure on some biomarkers of oxidative stress in rat brain regions. Nutr Cancer
        55: 195-200. http://dx.doi.org/10.1207/s15327914nc5502 11.
Calderon Guzman, D: Hernandez Islas, Jl_: Mejfa, GB: Santamarfa del Angel, D: Hernandez Garcfa, E: Juarez
        Olgufn. H. (2005). Effect of nutritional status and ozone exposure on Na+/K+ ATPpase and lipid
        peroxidation in rat brain. Proc West Pharmacol Soc 48: 118-121.
Carey, SA: Minard, KR: Trease, l_l_: Wagner, JG: Garcia, GJ:  Ballinger, CA: Kimbell, JS: Plopper, CG: Corlev,
        RA: Postlethwait. EM: Harkema. JR: Einstein. DR. (2007). Three-dimensional mapping of ozone-
        induced injury in the nasal  airways of monkeys using magnetic resonance imaging and morphometric
        techniques.  Toxicol Pathol  35: 27-40. http://dx.doi.org/10.1080/01926230601072343.
Carpagnano. GE: Foschino Barbaro. MP: Cagnazzo. M:  Pi Gioia. G:  Giliberti. T: Pi Matteo. C: Resta. O. (2005).
        Use of exhaled breath condensate in the study of airway inflammation after hypertonic saline solution
        challenge. Chest 128: 3159-3166. http://dx.doi.Org/10.1378/chest.128.5.3159.
Castagna, R; Davis,  PA; Vasu, VT;  Soucek, K; Cross, CE; Greci, L; Valacchi, G. (2009). Nitroxide radical
        TEMPO reduces ozone-induced chemokine IL-8 production in lung epithelial cells. Toxicol In Vitro 23:
        365-370. http://dx.doi.0rg/10.1016/i.tiv.2008.12.016.
Castilleios. M: Gold.  PR: Damokosh. Al: Serrano. P: Allen.  G:  McDonnell. WF: Dockerv. D: Velasco. SR:
        Hernandez.  M: Haves. C. (1995). Acute effects of ozone on the pulmonary function of exercising
        schoolchildren from Mexico City. Am J Respir Grit Care Med 152: 1501-1507.
Chan. C. -C: Chuang. K. -J: Su. T. -C:  Lin. L. -Y. (2005a). Association between nitrogen dioxide and heart rate
        variability in a susceptible population. Eur J Cardiovasc Prev Rehabil 12: 580-586.
Chan, C, -C: Wu, T, -H. (2005). Effects of ambient ozone exposure on mail carriers' peak expiratory flow rates.
        Environ Health Perspect 113:  735-738.
Chan. C. -C: Chuang. K. -J: Chien.  L. -C: Chen. W. -J: Chang. W. -T.  (2006). Urban air  pollution and emergency
        admissions  for cerebrovascular diseases in Taipei, Taiwan. Eur Heart J 27: 1238-1244.
Chang. C. -C: Tsai. S. -S: Ho. S. -C: Yang. C. -Y. (2005). Air pollution and hospital admissions for cardiovascular
        disease in Taipei, Taiwan. Environ Res 98: 114-119.
Chang, MM, -J: Wu,  R: Plopper, CG: Hyde, DM. (1998).  IL-8 is one of the major chemokines produced by
        monkey airway epithelium after ozone-induced injury. Am J Physiol 275: L524-L532.
Chen. C: Ariomandi.  M: Balmes. J: Tager. I: N. H. (2007). Effects of chronic and acute ozone exposure on lipid
        peroxidation and antioxidant capacity in healthy young adults. Environ Health Perspect 115: 1732-
        1737.  http://dx.doi.org/10.1289/ehp.10294.
Chen. J. -C: Schwartz. J. (2009). Neurobehavioral effects of ambient air pollution on cognitive performance in
        US adults. Neurotoxicology 30: 231-239. http://dx.doi.Org/10.1016/i.neuro.2008.12.011.
Chen. L: Jennison. BL: Yang. W: Omave. ST. (2000). Elementary school absenteeism and air pollution. Inhal
        Toxicol 12: 997-1016. http://dx.doi.org/10.1080/08958370050164626.
Chen, P, -C: Lai, Y, -M: Chan, C, -C: Hwang, J, -S: Yang, C, -Y: Wang, J, -D.  (1999). Short-term effect of ozone
        on the pulmonary function  of children in  primary school. Environ Health Perspect 107:  921-925.
        http://dx.doi.org/10.1289/ehp.99107921.
Chen, XQ: Yang, J: Hu, SP: Nie, HX: Mao, GY: Chen, HB. (2006b). Increased expression of CD86 and reduced
        production of IL-12 and IL-10  by monocyte-derived dendritic cells from allergic asthmatics and their
        effects on Th1- and Th2-type cytokine balance.  Respiration 73: 34-40.
        http://dx.doi.org/10.1159/000087457.
Draft - Do Not Cite or Quote                       6-238                                  September 2011

-------
Chhabra. SK: Yasir. A: Chaudhrv. K: Shah. B. (2010). Effect of ozone on response to ovalbumin & its modulation
        by vitamins C & E in sensitized guinea pigs. Indian J Med Res 132: 87-93.
Chimenti. L:  Morici. G: Paterno. A: Bonanno. A: Vultaggio. M: Bellia. V: Bonsignore. MR. (2009). Environmental
        conditions, air pollutants, and airway cells in runners: A longitudinal field study. J Sports Sci 27: 925-
        935. http://dx.doi.0rg/10.1080/02640410902946493.
Chiu. HF: Cheng. MH: Yang. CY. (2009). Air pollution and hospital admissions for pneumonia in a subtropical
        city: Taipei, Taiwan. Inhal Toxicol 21: 32-37.
Cho, HY: Morgan, PL: Bauer, AK: Kleeberger, SR. (2007). Signal transduction pathways of tumor necrosis
        factor—mediated lung injury induced by ozone in mice. Am J Respir Crit Care Med 175: 829-839.
        http://dx.doi.ora/10.1164/rccm.200509-1527OC.
Choi, JH: Xu, QS: Park,  SY: Kim, JH: Hwang,  SS: Lee,  KH: Lee, HJ: Hong, YC. (2007). Seasonal variation of
        effect of air pollution on blood pressure. J Epidemiol Community Health 61: 314-318.
Christian. PL: Chen. LL: Scannell. CH: Ferrando.  RE: Welch. BS: Balmes. JR. (1998). Ozone-induced
        inflammation is  attenuated with multiday exposure. Am J Respir Crit Care Med 158: 532-537.
Chuang. GC: Yang. Z: Westbrook. DG: Pompilius. M: Ballinger. CA: White. RC: Krzvwanski.  DM: Postlethwait.
        EM: Ballinger. SW. (2009). Pulmonary ozone exposure induces vascular dysfunction, mitochondrial
        damage, and atherogenesis. Am J Physiol Lung Cell Mol Physiol 297: L209-L216.
        http://dx.doi. orq/10.1152/ajplung.00102.2009.
Chuang. K. -J: Chan.  C. -C: Su. T. -C: Lee. C.  -T: Tang. C. -S. (2007a). The effect of urban air pollution on
        inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults. Am J Respir
        Crit  Care Med 176: 370-376.
Chuang. K. -J: Yan. Y -H: Cheng. T. -J. (2010). Effect of air pollution on blood pressure, blood lipids, and blood
        sugar: A population-based approach.  J Occup Environ Med 52: 258-262.
        http://dx.doi. orq/10.1097/JOM.Ob013e3181 ceff7a.
Chuang. KJ:  Chan. CC:  Su. TC: Lin. LY:  Lee. CT  (2007b). Associations between particulate  sulfate and organic
        carbon exposures and heart rate variability in patients with or at risk for cardiovascular diseases.  J
        Occup Environ  Med 49: 610-617.
demons. GK: Garcia. JF. (1980). Changes in thyroid function after short-term ozone exposure in rats. J Environ
        Pathol Toxicol Oncol 4: 359-369.
Cockcroft. DW: Davis. BE: Todd. DC: Smvcniuk. AJ. (2005). Methacholine challenge: Comparison of two
        methods. Chest 127: 839-844.
Coffin, PL: Blommer,  EJ: Gardner, DE: Holzman, R. (1967). Effect of air pollution on alteration of susceptibility
        to pulmonary infection. Cincinnati, OH: U.S. Department of Health, Education, and Welfare.
Coffin. PL: Gardner. DE. (1972). Interaction of biological agents and chemical air pollutants.  Ann Occup Hyg 15:
        219-234.
Cohen. MD:  Sisco. M: Baker. K: Li. Y: Lawrence. D: Van Loveren. H: Zelikoff. JT: Schlesinger. RB. (2002).
        Effects of inhaled ozone on pulmonary immune cells critical to antibacterial responses in situ. Inhal
        Toxicol 14: 599-619.
Cole. MP: Freeman. BA. (2009). Promotion of cardiovascular disease by exposure to the air pollutant ozone.
        Am  J Physiol Lung Cell Mol Physiol 297:  L209-L216.
Colfn-Barenque, L: Dorado-Martinez, C:  Rivas-Arancibia, S: Avila-Costa, MR: Fortoul, Tl. (2005). Morphological
        recovery of the  granule cells from the olfactory bulb after the cessation of acute ozone exposure.  Int J
        Neurosci 115: 411-421. http://dx.doi.orq/10.1080/00207450590521028.
Corradi, M: Alinovi, R: Goldoni, M: Vettori, M: Folesani,  G: Mozzoni, P: Cavazzini, S: Bergamaschi, E: Rossi, L:
        Mutti. A.  (2002). Biomarkers of oxidative stress after controlled human exposure to ozone. Toxicol Lett
        134: 219-225.
Corradi, M: Folesani,  G: Andreoli, R:  Manini, P: Bodini, A: Piacentini, G:  Carraro, S: Zanconato, S: Baraldi, E.
        (2003). Aldehydes  and glutathione in exhaled breath condensate of children with asthma exacerbation.
        Am  J Respir  Crit Care Med 167: 395-399. http://dx.doi.ora/10.1164/rccm.200206-507OC.
Cremillieux, Y: Servais,  S: Berthezene, Y: Dupuich, D: Boussouar, A: Stupar, V: Peguignot, JM. (2008). Effects
        of ozone exposure in rat lungs investigated with hyperpolarized 3He MRI. J Magn Reson  Imaging 27:
        771-776. http://dx.doi.org/10.1002/imri.21216.
Curriero, FC: Heiner,  KS: Samet, JM: Zeger, SL: Strug,  L: Patz, JA. (2002). Temperature and mortality in 11
        cities of the eastern United States. Am J Epidemiol 155: 80-87.  http://dx.doi.orq/10.1093/aie/155.1.80.
Draft - Do Not Cite or Quote                        6-239                                  September 2011

-------
Dahl. M: Bauer. AK: Arredouani. M: Soininen. R: Tryggvason. K: Kleeberger. SR: Kobzik. L. (2007). Protection
        against inhaled oxidants through scavenging of oxidized lipids by macrophage receptors MARCO and
        SR-AI/II. J Clin Invest 117: 757-764. http://dx.doi.org/10.1172/JCI29968.
Dales. R: Chen. L: Frescura. AM: Liu. L: Villeneuve. PJ. (2009). Acute effects of outdoor air pollution on forced
        expiratory volume in  1 s: A panel study of schoolchildren with asthma. Eur Respir J 34: 316-323.
        http://dx.doi. orq/10.1183/09031936.00138908.
Dales. RE: Cakmak. S: Doiron. MS.  (2006). Gaseous air pollutants and hospitalization for respiratory disease in
        the neonatal period. Environ Health Perspect 114: 1751-1754. http://dx.doi.org/10.1289/ehp.9044.
Damera, G; Zhao, H; Wang, M; Smith, M; Kirby, C; Jester, WF; Lawson, JA; Panettieri, RA, Jr. (2009). Ozone
        modulates IL-6 secretion in  human airway epithelial and smooth muscle cells. Am J Physiol Lung Cell
        Mol Physiol 296: L674-L683. http://dx.doi.org/10.1152/aiplung.90585.2008.
Damera. G: Jester William. F: Jiang. M: Zhao. H: Fogle Homer. W: Mittelman. M: Haczku. A: Murphy. E:  Parikh.
        I: Panettieri  Reynold. A. (2010). Inhibition of myristoylated alanine-rich C kinase substrate (MARCKS)
        protein inhibits ozone-induced airway neutrophilia and inflammation. Exp Lung Res 36: 75-84.
        http://dx.doi.org/10.3109/01902140903131200.
Darrow.  LA: Klein. M: Sarnat.  JA: Mulholland. JA: Strickland. MJ: Sarnat. SE: Russell. AG: Tolbert. PE. (2011 b).
        The use of alternative pollutant metrics in time-series studies of ambient air pollution and respiratory
        emergency department visits. J Expo Sci Environ Epidemiol 21:  10-19.
        http://dx.doi. orq/10.1038/ies.2009.49.
David, GL: Romieu, I: Sienra-Monge, JJ: Collins, WJ: Ramirez-Aguilar, M: Del Rio-Navarro, BE: Reyes-Ruiz, Ml:
        Morris. RW: Marzec.  JM: London. SJ. (2003). Nicotinamide adenine dinucleotide (phosphate)
        reduced:quinone oxidoreductase and glutathione s-transferase ml polymorphism and childhood
        asthma.  Am J  Respir Grit Care Med 168: 1199-1204.
De Pablo. F: Lopez. A: Soriano. LR:  Tomas. C: Diego. L: Gonzalez. M: Barrueco. M. (2006). Relationships of
        daily mortality and hospital admissions to air pollution in Castilla-Leon, Spain. Atmosfera 19: 23-39.
Delfino,  RJ: Zeiger, RS: Seltzer, JM: Street, DH: Matteucci,  RM: Anderson, PR: Koutrakis, P. (1997). The effect
        of outdoor fungal spore concentrations on daily asthma severity. Environ Health Perspect 105: 622-
        635.
Delfino,  RJ: Zeiger, RS: Seltzer, JM: Street, DH: McLaren, CE. (2002). Association of asthma symptoms with
        peak particulate air pollution and effect modification by anti-inflammatory medication use. Environ
        Health Perspect 110: A607-A617.
Delfino,  RJ: Gone, H: Linn, WS: Pellizzari, ED: Hu, Y. (2003). Asthma symptoms in Hispanic children and daily
        ambient exposures to toxic and criteria air pollutants. Environ Health Perspect 111: 647-656.
        http://dx.doi.org/10.1289/ehp.5992.
Delfino,  RJ: Quintana,  PJE: Floro, J: Gastanaga, VM: Samimi,  BS: Kleinman, MT: Liu, LJS: Bufalino, C:  Wu,
        CF: McLaren.  CE. (2004). Association of FEV1 in asthmatic children with personal and
        microenvironmental exposure to airborne particulate matter. Environ Health Perspect 112: 932-941.
Delfino,  RJ: Staimer, N: Tjoa,  T: Arhami, M: Polidori, A: Gillen, PL: George, SC: Shafer, MM: Schauer, JJ:
        Sioutas.  C. (201 Oa). Associations of primary and secondary organic aerosols with airway and aystemic
        inflammation in an elderly panel cohort. Epidemiology 21: 892-902.
        http://dx.doi. org/10.1097/EDE.Ob013e3181 f20e6c.
Delfino.  RJ: Tioa. T: Gillen. PL: Staimer. N: Polidori. A:  Arhami. M: Jamner. L: Sioutas. C: Longhurst. J. (201 Ob).
        Traffic-related  air pollution and blood pressure in elderly subjects with coronary artery disease.
        Epidemiology  21: 396-404.  http://dx.doi.org/10.1097/EDE.Ob013e3181d5e19b.
Delfino.  RJ: Gillen. PL: Tioa. T: Staimer. N: Polidori. A:  Arhami. M: Sioutas. C: Longhurst. J. (2011).
        Electrocardiographic ST-segment depression and exposure to traffic-related aerosols in elderly
        subjects with coronary artery disease. Environ Health Perspect 119: 196-202.
        http://dx.doi.org/10.1289/ehp.1002372.
DeLucia. AJ: Adams. WC. (1977). Effects of O3 inhalation during exercise on pulmonary function and blood
        biochemistry. J Appl Physiol 43: 75-81.
Dennekamp. M: Akram. M: Abramson. MJ: Tonkin. A: Sim.  MR: Fridman.  M: Erbas. B. (2010).  Outdoor air
        pollution as  a trigger  for out-of-hospital cardiac arrests. Epidemiology 21: 494-500.
        http://dx.doi. org/10.1097/EDE.Ob013e3181 e093db.
Draft - Do Not Cite or Quote                        6-240                                 September 2011

-------
Depuvdt. P: Joos. GF: Pauwels. RA. (1999). Ambient ozone concentrations induce airway hyperresponsiveness
        in some rat strains. Eur Respir J 14: 125-131.
Devlin. RB: McDonnell. WF: Mann. R: Becker. S: House. DE: Schreinemachers. D: Koren. HS. (1991).
        Exposure of humans to ambient levels of ozone for 6.6 hours causes cellular and biochemical changes
        in the lung. Am J Respir Cell Mol Biol 4: 72-81.
Devlin. RB: McDonnell. WF: Becker. S: Madden.  MC: McGee. MP: Perez. R:  Hatch. G: House. DE: Koren. HS.
        (1996). Time-dependent changes of inflammatory mediators in the lungs of humans exposed to 0.4
        ppm ozone for 2 hr: A comparison of mediators found in bronchoalveolar lavage fluid  1 and 18 hr after
        exposure. Toxicol Appl Pharmacol 138: 176-185.
Devlin. RB: Folinsbee. LJ: Biscardi. F: Hatch. G: Becker. S: Madden. MC: Robbins. M: Koren.  HS. (1997).
        Inflammation and cell damage induced by repeated exposure of humans to ozone.  Inhal Toxicol 9: 211-
        235.
Dimeo. MJ: Glenn. MG: Holtzman. MJ: Sheller. JR: Nadel. JA: Boushev. HA.  (1981). Threshold concentration of
        ozone causing an increase in bronchial reactivity in humans and adaptation with repeated exposures.
        Am Rev Respir Dis 124: 245-248.
Dimitriadis. VK. (1992). Carbohydrate cytochemistry of bonnet monkey  (Macaca radiata) nasal epithelium:
        Response to ambient levels of ozone.  Histol Histopathol 7: 479-488.
Dockerv. DW: Luttmann-Gibson. H: Rich. DQ: Link. MS:  Mittleman. MA: Gold. PR: Koutrakis. P:  Schwartz. JD:
        Verrier. RL. (2005). Association of air pollution with increased incidence of ventricular tachyarrhythmias
        recorded by implanted cardioverter defibrillators. Environ Health Perspect 113: 670-674.
Dohm. MR: Mautz. WJ: Andrade. JA: Gellert. KS: Salas-Ferguson. LJ: Nicolaisen. N: Fuiie. N. (2005). Effects of
        ozone exposure on nonspecific phagocytic capacity of pulmonary macrophages from an amphibian,
        Bufo marinus. Environ Toxicol Chem 24: 205-210.
Dorado-Martinez. C: Parades-Carbaial. C: Mascher. D: Borgonio-Perez. G: Rivas-Arancibia. S. (2001). Effects
        of different ozone doses on memory, motor activity and lipid peroxidation levels, in rats.  Int J Neurosci
        108: 149-161.
Driscoll. KE: Vollmuth. TA: Schlesinger. RB. (1987). Acute and subchronic ozone inhalation in  the rabbit:
        Response of alveolar macrophages. J Toxicol Environ Health 21: 27-43.
        http://dx.doi. org/10.1080/15287398709531 OOP.
Drvden. DM: Spooner. CH: Stickland. MK: Vandermeer. B: Tiosvold. L: Bialv.  L:  Wong. K: Rowe.  BH.  (2010).
        Exercise-induced bronchoconstriction and asthma.  In Evidence Report/Technology Assessment.
        (AHRQ Publication No. 10-E001). Rockville,  MD: Agency for Healthcare Research and Quality.
Duramad. P: Tager. IB: Holland. NT.  (2007). Cytokines and other immunological biomarkers in children's
        environmental health studies. Toxicol Lett 172: 48-59.
Eiswerth, ME: Shaw, WD: Yen, ST. (2005). Impacts of ozone on the activities  of asthmatics:  Revisiting the data.
        J Environ Manage 77: 56-63. http://dx.doi.Org/10.1016/i.ienvman.2005.02.010.
Escalante-Membrillo. C: Gonzalez-Maciel. A: Revnoso-Robles. R: Gonzalez-Pina. R. (2005). Brain thiobarbituric
        acid-reactive substances  in rats  after short periods of ozone exposure.  Environ Res 99:  68-71.
        http://dx.doi.0rg/10.1016/i.envres.2005.02.006.
Escamilla-Nunez. MC: Barraza-Villarreal. A: Hernandez-Cadena. L: Moreno-Macias. H: Ramirez-Aguilar. M:
        Sienra-Monge, JJ: Cortez-Lugo, M: Texcalac, Jl_: del Rio-Navarro, B: Romieu,  I. (2008). Traffic-related
        air pollution and respiratory symptoms among asthmatic children, resident in Mexico City: The EVA
        cohort study. Respir Res 9:  74. http://dx.doi.org/10.1186/1465-9921-9-74.
Fakhri, AA: Ilic, LM: Wellenius, GA: Urch, B: Silverman, F: Gold, PR:  Mittleman, MA. (2009). Autonomic effects
        of controlled fine particulate exposure in young healthy adults:  Effect modification by ozone.  Environ
        Health Perspect 117: 1287-1292. http://dx.doi.org/10.1289/ehp.0900541.
Fakhrzadeh, L: Laskin, JD: Laskin, PL. (2008). Regulation of caveolin-1 expression, nitric oxide production and
        tissue injury by tumor necrosis factor-alpha following ozone inhalation.  Toxicol Appl Pharmacol 227:
        380-389. http://dx.doi.0rg/10.1016/i.taap.2007.11.012.
Farraj, AK: Boykin, E: Ledbetter, A: Andrews, D: Gavett, SH. (2010). Increased  lung resistance after diesel
        particulate and ozone co-exposure not associated with enhanced lung  inflammation in allergic mice.
        Inhal Toxicol 22: 33-41. http://dx.doi.org/10.3109/08958370902862434.
Draft - Do Not Cite or Quote                       6-241                                 September 2011

-------
Feng. R: He. W: Ochi. H: Castranova. V. (2006). Ozone exposure impairs antigen-specific immunity but
       activates IL-7-induced proliferation of CD4-CD8-thymocytes in BALB/c mice. J Toxicol Environ Health
       A 69: 1511-1526. http://dx.doi.orq/10.1080/15287390500468696.
Feo Brito. F: Mur Gimeno. P: Martinez. C: Tobias. A: Suarez. L: Guerra. F: Boria. JM: Alonso. AM. (2007). Air
       pollution and seasonal asthma during the pollen season: A cohort study in Puertollano and Ciudad Real
       (Spain). Allergy 62: 1152-1157.
Ferdinands. JM: Crawford. CA: Greenwald. R: Van Sickle. D: Hunter. E: league. WG. (2008).  Breath
       acidification in adolescent runners exposed to atmospheric pollution: A prospective, repeated measures
       observational study. Environ Health Global Access Sci Source 7: 11. http://dx.doi.org/10.1186/1476-
       069X-7-10.
Folinsbee, LJ: Silverman, F: Shephard, RJ. (1977).  Decrease of maximum work performance following ozone
       exposure. J Appl Physiol 42: 531-536.
Folinsbee. LJ: Drinkwater. BL: Bedi. JF: Horvath. SM.  (1978). The influence of exercise on the pulmonary
       function changes due to exposure to low concentrations of ozone. In LJ Folinsbee; JA Wagner; JF
       Borgia; BL Drinkwater; JAGIiner; JF Bedi (Eds.), Environmental stress: individual human adaptations
       (pp. 125-145). New York, NY: Academic Press.
Folinsbee, LJ; Bedi, JF; Horvath, SM. (1980). Respiratory responses in humans repeatedly exposed to low
       concentrations of ozone. Am Rev Respir Dis 121: 431-439.
Folinsbee. LJ: Bedi. JF: Horvath. SM. (1984). Pulmonary function changes after 1 h continuous heavy exercise
       in 0.21 ppm ozone. J Appl Physiol 57: 984-988.
Folinsbee. LJ: McDonnell. WF: Horstman. PH. (1988). Pulmonary function and symptom  responses after 6.6-
       hour exposure to 0.12 ppm ozone with moderate exercise. J Air Waste Manag Assoc  38: 28-35.
Folinsbee, LJ; Hazucha, MJ. (1989). Persistence of ozone-induced changes in lung function and airway
       responsiveness. In Atmospheric ozone research and its policy implications (pp. 483-492). Amsterdam,
       The Netherlands: Elsevier.
Folinsbee, LJ; Devlin, RB; Abdul-Salaam, S; Koren, HS. (1993). Repeated severe ozone exposure causes
       depressed baseline spirometry [Abstract]. Am Rev Respir Dis 147: A638.
Folinsbee. LJ: Horstman. DH: Kehrl. HR: Harder. S: Abdul-Salaam. S: Ives. PJ. (1994). Respiratory responses
       to repeated prolonged exposure to 0.12 ppm ozone. Am J Respir Crit Care Med 149:  98-105.
Folinsbee. LJ: Devlin. RB: Robbins. MK: Biscardi. FH: Abdul-Salaam. S: Koren. HS. (1998). Repeated exposure
       of humans to ozone: Pulmonary function and  symptom responses. Research Triangle Park, NC: U.S.
       Environmental Protection Agency.
Folinsbee. LJ: Hazucha. MJ. (2000). Time course of response to ozone exposure in healthy adult females. Inhal
       Toxicol 12: 151-167.
Fortino, V; Maioli, E; Torricelli, C; Davis, P; Valacchi, G. (2007). Cutaneous MMPs are differently modulated by
       environmental stressors in old and young mice. Toxicol Lett 173: 73-79.
       http://dx.doi.orq/10.1016/i.toxlet.2007.06.004.
Foster, WM;  Silver, JA; Groth, ML. (1993).  Exposure to ozone alters regional function and particle dosimetry in
       the human lung. J Appl Physiol 75: 1938-1945.
Foster. WM:  Stetkiewicz. PT (1996). Regional clearance of solute from the respiratory epithelia:  18-20 h
       post exposure to ozone. J Appl Physiol 81: 1143-1149.
Foster. WM: Weinmann. GG: Menkes. E: Maori.  K. (1997). Acute exposure of humans to ozone impairs small
       airway function. Ann Occup Hyg 1: 659-666.
Fox,  SD; Adams, WC; Brookes, KA; Lasley, BL.  (1993). Enhanced response to ozone exposure during the
       follicular phase of the menstrual cycle. Environ Health Perspect 101: 242-244.
Foxcroft. WJ: Adams. WC. (1986). Effects of ozone exposure on four consecutive days on work performance
       and VO2max.  J Appl Physiol 61: 960-966.
Frampton. MW: Morrow. PE: Torres. A: Cox. C: Voter. KZ: Utell. MJ: Gibb.  FR: Speers. DM. (1997b). Ozone
       responsiveness in smokers and nonsmokers. Am J Respir Crit Care Med 155: 116-121.
Frank, R;  Liu, MC; Spannhake, EW; Mlynarek, S; Maori, K; Weinmann, GG. (2001). Repetitive ozone exposure
       of young adults: Evidence of persistent small airway dysfunction. Am J Respir Crit Care Med 164:
       1253-1260.
Franklin, M; Schwartz,  J. (2008). The impact of secondary particles on the association between ambient ozone
       and mortality. Environ Health Perspect 116: 453-458. http://dx.doi.org/10.1289/ehp. 10777.
Draft - Do Not Cite or Quote                       6-242                                 September 2011

-------
Franze. T: Weller. MG: Niessner. R: Poschl. U. (2005). Protein nitration by polluted air. Environ Sci Technol 39:
        1673-1678.
Friedman. M: Gallo. JM: Nichols. HP: Bromberg. PA. (1983). Changes in inert gas rebreathing parameters after
        ozone exposure in dogs. Am Rev Respir Dis 128: 851-856.
Frush, S: Li, Z: Potts,  EN: Du, W: Eu, JP: Garantziotis, S: He, YW: Foster, WM: Hollingsworth, JW. (In Press)
        The role of the extracellular matrix protein mindin in airway response to environmental airways injury.
        Environ Health Perspect. http://dx.doi.orq/10.1289/ehp. 1003339.
Fung, KY: Luginaah, I: Gorev, KM: Webster, G. (2005). Air pollution and daily hospital admissions for
        cardiovascular diseases in Windsor, Ontario. Can J Public Health 96: 29-33.
Gao. X:  Raghavamenon. AC: D'Auvergne. O:  Uppu. RM. (2009b).  Cholesterol secoaldehyde induces apoptosis
        in J774 macrophages via mitochondrial pathway but not involving reactive oxygen species as
        mediators. Biochem Biophys Res Commun 389: 382-387. http://dx.doi.Org/10.1016/i.bbrc.2009.09.005.
Garantziotis. S: Li. Z:  Potts. EN: Lindsev. JY: Stober. VP: Polosukhin. VV: Blackwell. TS: Schwartz. DA: Foster.
        WM: Hollingsworth, JW. (2010). TLR4 is necessary for hyaluronan-mediated airway
        hyperresponsiveness after ozone inhalation. Am J Respir Crit Care Med 181: 666-675.
        http://dx.doi.ora/10.1164/rccm.200903-0381OC.
Gent, JF: Triche, EW: Holford, TR: Belanger, K: Bracken, MB: Beckett, WS: Leaderer, BP  (2003). Association of
        low-level ozone and fine particles with respiratory symptoms in  children with asthma. JAMA 290: 1859-
        1867. http://dx.doi.org/10.1001/iama.290.14.1859.
Gershwin, LJ:  Osebold, JW: Zee, YC. (1981).  Immunoglobulin E-containing cells in mouse lung following
        allergen inhalation and  ozone exposure. International Arch Allergy Appl Immunol 65: 266-277.
Gielen. MH: Van Per Zee. SC: Van Wiinen. JH: Van Steen. CJ: Brunekreef. B. (1997). Acute effects of summer
        air pollution on respiratory health of asthmatic children. Am J Respir Crit Care Med 155:  2105-2108.
Gilliland.  FD: Berhane. K: Rappaport. EB: Thomas. DC: Avol. E: Gauderman. WJ: London. SJ: Margolis. HG:
        McConnell. R: Islam. KT: Peters. JM. (2001). The effects of ambient air pollution on school
        absenteeism due to respiratory illnesses. Epidemiology 12:  43-54.
Gilmour. Ml: Jakab. GJ. (1991).  Modulation of immune function in mice exposed to 08 ppm ozone. Inhal Toxicol
        3: 293-308.
Girardot, SP: Ryan, PB: Smith, SM: Davis, WT: Hamilton, CB: Obenour, RA: Renfro, JR: Tromatore, KA: Reed,
        GD. (2006).  Ozone and PM2.5 exposure and acute pulmonary health effects: A study of hikers in the
        Great Smoky Mountains National Park. Environ Health Perspect 113: 612-617.
        http://dx.doi.org/10.1289/ehp.8637.
Gold. PR: Damokosh. Al: III. PC: Dockerv. DW: McDonnell. WF: Serrano.  P: Retama. A: Castilleios. M. (1999).
        Particulate and ozone pollutant effects on the respiratory function of children in southwest Mexico City.
        Epidemiology 10:  8-16.
Goldberg. MS: Giannetti. N: Burnett. RT: Mayo. NE:  Valois. MF:  Brophv. JM. (2008). A panel study in congestive
        heart failure to estimate the short-term effects from personal factors and environmental conditions  on
        oxygen saturation and pulse rate. Occup Environ Med 65: 659-666.
        http://dx.doi.org/10.1136/oem.2007.034934.
Gong. H. Jr: Bradley. PW: Simmons. MS: Tashkin. DP. (1986). Impaired exercise performance and pulmonary
        function in elite cyclists during low-level ozone exposure in a hot environment. Am J Respir Crit Care
        Med 134: 726-733.
Gong. H. Jr: McManus. MS: Linn. WS. (1997a). Attenuated response to  repeated daily ozone exposures in
        asthmatic subjects. Arch Environ Occup Health 52: 34-41.
Gong. H. Jr: Shamoo. DA: Anderson. KR: Linn. WS. (1997b). Responses of older men with and without chronic
        obstructive pulmonary disease to prolonged ozone exposure. Arch Environ Occup Health 52: 18-25.
Gong, H, Jr: Wong, R: Sarma, RJ: Linn,  WS: Sullivan, ED: Shamoo,  DA: Anderson, KR: Prasad,  SB. (1998).
        Cardiovascular effects of ozone exposure in human volunteers. Am J Respir Crit Care Med 158: 538-
        546.
Gonzalez-Pina, R: Escalante-Membrillo, C: Alfaro-Rodriguez, A: Gonzalez-Maciel, A. (2008). Prenatal exposure
        to ozone disrupts cerebellar monoamine contents in newborn rats. Neurochem Res 33: 912-918.
        http://dx.doi. orq/10.1007/s11064-007-9534-3.
Draft - Do Not Cite or Quote                       6-243                                 September 2011

-------
Graham. JA: Menzel. DB: Miller. FJ: Illing. JW: Gardner. DE. (1981). Influence of ozone on pentobarbital-
       induced sleeping time in mice, rats, and hamsters. Toxicol Appl Pharmacol 61: 64-73.
       http://dx.doi.org/10.1016/0041-008X(81)90008-9.
Grunewald. J: Eklund. A. (2007). Role of CD4+ T cells in sarcoidosis. Proc Am Thorac Soc 4: 461-464.
Gryparis, A: Forsberg, B: Katsouyanni, K: Analitis, A: Touloumi, G: Schwartz, J: Samoli, E: Medina, S: Anderson,
       HR: Niciu. EM: Wichmann. HE: Khz. B: Kosnik. M: Skorkovskv. J: Vonk. JM: Dortbudak. Z. (2004).
       Acute effects of ozone on mortality from the "Air pollution and health: A European approach" project.
       Am J Respir Grit Care Med 170: 1080-1087.  http://dx.doi.org/10.1164/rccm.200403-333OC.
Guerrero. AL: Dorado-Martinez. C:  Rodriguez. A: Pedroza-Rios. K: Borgonio-Perez. G: Rivas-Arancibia. S.
       (1999). Effects of vitamin E on ozone-induced memory deficits and lipid peroxidation in rats.
       Neuroreport 10: 1689-1692.
Guevara-Guzman. R: Arriaga. V: Kendrick. KM: Bernal. C: Vega. X: Mercado-Gomez. OF: Rivas-Arancibia.  S.
       (2009). Estradiol prevents ozone-induced increases in brain lipid peroxidation and impaired social
       recognition memory in female rats. Neuroscience 159: 940-950.
       http://dx.doi.0rg/10.1016/i.neuroscience.2009.01.047.
Hackney. JD:  Linn. WS:  Mohler. JG: Pedersen. EE: Breisacher. P: Russo. A. (1975). Experimental studies on
       human health effects of air pollutants: II. Four-hour exposure to ozone alone and in combination with
       other pollutant gases. Arch Environ Occup Health 30: 379-384.
Halonen. Jl: Lanki. T: Tiittanen. P: Niemi. JV: Loh. M: J. P. (2009). Ozone and cause-specific cardiorespiratory
       morbidity and mortality. J Epidemiol Community Health 64: 814-820.
       http://dx.doi. org/10.1136/iech.2009.087106.
Hamade. AK:  Rabold. R: Tankerslev. CG. (2008). Adverse cardiovascular effects with acute particulate matter
       and ozone exposures: Interstrain variation in mice. Environ Health Perspect 116: 1033-1039.
       http://dx.doi. org/10.1289/ehp. 10689.
Hamade. AK:  Tankerslev. CG. (2009). Interstrain variation in cardiac and respiratory adaptation to repeated
       ozone and particulate matter exposures. Am J Physiol Regul Integr Comp Physiol 296: R1202-R1215.
       http://dx.doi.org/10.1152/aipregu.90808.2008.
Hamade. AK:  Misra.  V: Rabold. R: Tankerslev. CG. (2010). Age-related changes in cardiac and respiratory
       adaptation to acute ozone  and carbon black exposures:  Interstrain variation in mice. Inhal Toxicol 22:
       84-94. http://dx.doi.org/10.3109/08958378.2010.503974.
Han. SG: Andrews. R: Gairola. CG: Bhalla. DK. (2008). Acute pulmonary effects of combined exposure to
       carbon nanotubes  and ozone in mice. Inhal Toxicol 20: 391-398.
       http://dx.doi. org/10.1080/08958370801904014.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Dungworth. PL (1987a). Effects of an ambient level of
       ozone on primate nasal epithelial mucosubstances: quantitative histochemistry. Am J Pathol 127: 90-
       96.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Wilson. DW: Dungworth. PL. (1987b). Response of the
       macaque nasal epithelium to ambient levels of ozone: A morphologic and morphometric study of the
       transitional and respiratory epithelium. Am J Pathol 128: 29-44.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Wilson. DW: Dungworth. PL. (1993).  Response of
       macaque bronchiolar epithelium to ambient concentrations of ozone. Am J Pathol 143: 857-866.
Hatch. GE: Slade.  R: Harris. LP: McPonnell. WF: Pevlin. RB: Koren. HS: Costa. PL: McKee. J. (1994). Ozone
       dose and effect in humans and rats: A comparison using oxygen-18 labeling and bronchoalveolar
       lavage. Am J Respir Grit Care Med 150: 676-683.
Hazucha. MJ: Folinsbee. LJ: Seal. E. Jr. (1992). Effects of steady-state and variable ozone concentration
       profiles on pulmonary function. Am J Respir Crit Care Med 146:  1487-1493.
Hazucha, MJ: Folinsbee, LJ: Bromberg, PA. (2003). Pistribution and reproducibility of spirometric response  to
       ozone by gender and age.  J Appl Physiol 95: 1917-1925.
HEI. (Health Effects Institute). (2003). Revised analyses of time-series studies of air pollution and health:
       Revised analyses of the National Morbidity, Mortality, and Air Pollution Study (NMMAPS),  Part II.
       Cambridge,  MA. http://pubs.healtheffects.org/view.php?id=4.
Praft - Po Not Cite or Quote                       6-244                                  September 2011

-------
Heidenfelder. BL: Reif. DM: Harkema. JR: Cohen Hubal. EA: Hudgens. EE: Bramble. LA: Wagner. JG:
        Morishita, M: Keeler, GJ: Edwards, SW: Gallagher, JE. (2009). Comparative microarray analysis and
        pulmonary changes in brown Norway rats exposed to ovalbumin and concentrated air particulates.
        ToxicolSci 108: 207-221.
Hemmingsen, A: Fryer, AA: Hepple, M: Strange, RC: Spiteri, MA. (2001). Simultaneous identification of GSTP1
        Ne105->Val105 and Ala114->Val114 substitutions using an amplification refractory mutation system
        polymerase chain reaction assay: Studies in patients with asthma. Respir Res 2: 255-260.
        http://dx.doi. org/10.1186/rr64.
Henderson. FW: Dubovi. EJ: Harder. S: Seal. E. Jr: Graham. D. (1988). Experimental rhinovirus infection in
        human volunteers exposed to ozone. Am J Respir Crit Care Med 137: 1124-1128.
Henrotin, JB: Besancenot, JP: Bejot, Y: Giroud, M. (2007). Short-term effects of ozone air pollution on
        ischaemic stroke occurrence: A case-crossover analysis from a 10-year population-based study in
        Dijon,  France. Occup Environ  Med 64: 439-445.
Henrotin, JB: Zeller, M: Lorgis, L: Cottin, Y: Giroud, M:  Bejot, Y. (2010). Evidence of the role of short-term
        exposure to ozone on ischaemic cerebral and cardiac events: The Dijon Vascular Project (DIVA).  Heart
        96: 1990-1996. http://dx.doi.orq/10.1136/hrt.2Q10.200337.
Hernandez-Cadena, L: Holguin, F: Barraza-Villarreal, A: Del Rfo-Navarro, BE: Sienra-Monge, JJ: Romieu, I.
        (2009). Increased levels of outdoor air pollutants are associated with reduced bronchodilation in
        children with asthma. Chest 136: 1529-1536. http://dx.doi.org/10.1378/chest.08-1463.
Hernandez, ML: Lay, JC: Harris, B: Esther, CR:  Brickev, WJ: Bromberg, PA: Diaz-Sanchez,  D: Devlin, RB:
        Kleeberger. SR: Alexis. NE: Peden. DB. (2010). Atopic asthmatic subjects but not atopic subjects
        without asthma have enhanced inflammatory response to ozone. J Allergy Clin Immunol 126: 537-544.
        http://dx.doi.0rg/10.1016/i.iaci.2010.06.043.
Hicks. A: Kourteva.  G: Hilton. H: Li. H:  Lin. T: Liao. W:  Li. Y: Wei. X: March. T: Benson. J: Renzetti. L. (201 Oa).
        Cellular and molecular characterization of ozone-induced pulmonary inflammation in the Cynomolgus
        monkey.  Inflammation 33: 144-156. http://dx.doi.org/10.1007/s10753-009-9168-5.
Hicks. A: Goodnow. R. Jr: Cavallo. G: Tannu. SA: Ventre.  JD: Lavelle.  D: Lora. JM: Satiawatcharaphong. J:
        Brovarnev. M: Dabbagh. K: Tare. NS: Oh. H: Lamb. M: Sidduri. A: Dominigue. R: Qiao. Q: Lou. JP:
        Gillespie, P: Fotouhi, N: Kowalczyk, A: Kurylko, G: Hamid, R:  Wright, MB: Pamidimukkala, A: Egan, T:
        Gubler. U: Hoffman. AF: Wei. X: Li. YL: O'Neil. J: Marcano. R: Pozzani. K: Molinaro. T: Santiago. J:
        Singer. L: Hargaden. M: Moore. D: Catala. AR: Chao. LC: Benson. J: March. T: Venkat. R: Mancebo. H:
        Renzetti, LM. (201 Ob). Effects of LTB4 receptor antagonism on pulmonary inflammation in rodents and
        non-human primates. Prostaglandins Other Lipid Mediat 92: 33-43.
        http://dx.doi. org/10.1016/i. prostaglandins.2010.02.003.
Higgins, ITT: D'Arcy, JB: Gibbons, PI: Avol, EL:  Gross, KB. (1990). Effect of exposures to ambient ozone on
        ventilatory lung function in children. Am J Respir  Crit Care Med 141: 1136-1146.
Hiltermann. JTN:  Lapperre. TS: Van Bree. L: Steerenberg. PA: Brahim. JJ: Sont. JK: Sterk. PJ: Hiemstra. PS:
        Stolk, J. (1999). Ozone-induced inflammation  assessed in sputum and bronchial lavage fluid from
        asthmatics: A new noninvasive tool in epidemiologic studies on air pollution and asthma. Free Radic
        Biol Med 27: 1448-1454.
Hiltermann, TJN:  Stolk, J: Hiemstra, PS: Fokkens, PHB: Rombout,  PJA: Sont, JK: Sterk, PJ: Dijkman, JH.
        (1995). Effect of ozone exposure on maximal airway narrowing in non-asthmatic and asthmatic
        subjects. Clin Sci (Lond) 89: 619-624.
Hiltermann, TJN:  de Bruijne, CR: Stolk, J: Zwinderman, AH:  FThM, S:  Roemer, W: Steerenberg, PA:  Fischer,
        PH: van Bree. L: Hiemstra. PS. (1997). Effects of photochemical air pollution and allergen exposure on
        upper respiratory tract inflammation in asthmatics. Am J Respir Crit Care Med 156: 1765-1772.
Hiltermann, TJN:  Peters, EA: Alberts, B: Kwikkers, K: Borggreven, PA: Hiemstra, PS: Dijkman, JH: van Bree,
        LA: Stolk. J. (1998). Ozone-induced airway  hyperresponsiveness in patients with asthma: Role of
        neutrophil-derived serine proteinases. Free Radic Biol Med 24: 952-958.
Hinwood, Al_: De  Klerk, N: Rodriguez, C: Jacobv, P:  Runnion, T: Rye, P: Landau, L: Murray, F: Feldwick, M:
        Spickett. J. (2006). The relationship between changes in daily air pollution and hospitalizations in Perth,
        Australia 1992-1998: A case-crossover study.  Int J Environ  Health Res 16: 27-46.
        http://dx.doi.org/10.1080/09603120500397680.
Draft - Do Not Cite or Quote                       6-245                                  September 2011

-------
Hoek. G: Brunekreef. B: Kosterink. P: Van den Berg. R: Hofschreuder. P. (1993). Effect of ambient ozone on
        peak expiratory flow of exercising children in the Netherlands. Arch Environ Occup Health 48: 27-32.
        http://dx.doi. orq/10.1080/00039896.1993.9938390.
Hoek. G: Brunekreef. B. (1995). Effect of photochemical air pollution on acute respiratory symptoms in children.
        Am J Respir Grit Care Med 151: 27-32.
Hollingsworth. JW: Free.  ME:  Li. Z: Andrews. LN: Nakano. H: Cook. DN. (2010). Ozone activates pulmonary
        dendritic cells and promotes allergic sensitization through a Toll-like receptor 4-dependent mechanism.
        J Allergy Clin Immunol 125: 1167-1170.
Hollingsworth. JW. II: Cook. DN: Brass.  DM: Walker. JKL: Morgan.  PL: Foster. WM: Schwartz. DA. (2004). The
        role of Toll-like receptor 4 in environmental airway injury in mice. Am J Respir Crit Care Med 170: 126-
        132.
Holz. O: Jorres. RA: Timm. P: Mucke. M: Richter. K: Koschvk. S: Magnussen. H. (1999). Ozone-induced airway
        inflammatory changes differ between individuals and are reproducible. Am J Respir Crit Care Med 159:
        776-784.
Holz. O: Mucke. M: Paasch. K: Bohme.  S: Timm. P: Richter. K: Magnussen. H: Jorres. RA. (2002). Repeated
        ozone exposures enhance bronchial allergen responses in subjects with rhinitis or asthma.  Clin Exp
        Allergy 32: 681-689.
Holz. O: Tal-Singer. R: Kanniess. F: Simpson. KJ: Gibson. A: Vessev.  RSJ: Janicki. S: Magnussen. H: Jorres.
        RA: Richter. K. (2005). Validation of the human ozone challenge model as a tool for assessing anti-
        inflammatory drugs in early development. J Clin Pharmacol 45: 498-503.
Hoppe. P: Praml. G: Rabe. G: Lindner. J: Fruhmann. G: Kessel. R.  (1995). Environmental ozone field study on
        pulmonary and subjective responses of assumed risk groups. Environ Res 71: 109-121.
Hoppe, P: Peters, A:  Rabe, G: Praml,  G: Lindner, J: Jakobi, G: Fruhmann, G: Nowak, D. (2003).  Environmental
        ozone effects in different population subgroups. Int J Hyg Environ Health 206: 505-516.
        http://dx.doi.orq/10.1078/1438-4639-00250.
Horstman, DH: Folinsbee, LJ: Ives,  PJ: Abdul-Salaam, S: McDonnell,  WF. (1990). Ozone concentration and
        pulmonary response  relationships for 6.6-hour exposures with five hours of moderate exercise to 0.08,
        0.10, and 0.12 ppm. Am J Respir Crit Care Med 142: 1158-1163.
Horstman, DH: Ball, BA: Brown, J: Gerritv, T: Folinsbee, LJ.  (1995). Comparison of pulmonary responses of
        asthmatic and nonasthmatic subjects performing light exercise while exposed to a low level of ozone.
        Toxicol Ind Health 11: 369-385.
Horvath, SM: Gliner, JA: Matsen-Twisdale, JA. (1979). Pulmonary function and maximum exercise responses
        following acute ozone exposure. Aviat Space Environ Med 50: 901-905.
Horvath. SM: Gliner. JA: Folinsbee. LJ. (1981). Adaptation to ozone: Duration of effect. Am Rev Respir Dis 123:
        496-499.
Hosseinpoor. AR: Forouzanfar. MH: Yunesian.  M: Asghari. F: Naieni. KH: Farhood. D. (2005). Air pollution and
        hospitalization due to angina pectoris in Tehran, Iran: A time-series study. Environ Res 99: 126-131.
Houslev, DG: Eccles, R: Richards, RJ. (1996).  Gender difference in the concentration of the antioxidant uric
        acid in human nasal lavage. Acta Otolaryngol  116:  751-754.
Howarth. PH: Persson. CG: Meltzer. EO: Jacobson. MR: Durham. SR: Silkoff. PE. (2005). Objective monitoring
        of nasal airway inflammation  in  rhinitis. J Allergy Clin Immunol 115: S414-S441.
Huffman. LJ: Beighlev. CM:  Frazer. DG:  McKinnev. WG: Porter.  DW. (2006). Increased susceptibility of the
        lungs of hyperthyroid rats to oxidant injury: Specificity of effects. Toxicology 225: 119-127.
        http://dx.doi.0rg/10.1016/i.tox.2006.05.008.
Hunt. J. (2002). Exhaled breath condensate: An evolving tool for noninvasive evaluation of lung disease. J
        Allergy Clin Immunol 110: 28-34. http://dx.doi.orq/10.1067/mai.2002.124966.
Hunt, JF: Fang, K: Malik,  R: Snyder, A: Malhotra, N: Platts-Mills, TAE: Gaston, B. (2000). Endogenous airway
        acidification: Implications for asthma pathophysiology. Am  J Respir Crit Care Med 161: 694-699.
Hurst. DJ: Gardner. DE: Coffin. PL.  (1970). Effect of ozone on acid  hydrolases of the pulmonary  alveolar
        macrophage. J Reticuloendothel Soc 8: 288-300.
Hyde. DM: Hubbard.  WC: Wong. V: Wu. R: Pinkerton.  K: Plopper. CG. (1992). Ozone-induced acute
        tracheobronchial epithelial injury: relationship to granulocyte emigration in the lung. Am J Respir Cell
        Mol Biol 6: 481-497.
Draft - Do Not Cite or Quote                       6-246                                 September 2011

-------
Inoue. K: Takano. H: Kaewamatawong. T: Shimada. A: Suzuki. J: Yanagisawa. R: Tasaka. S: Ishizaka. A: Satoh.
        J\/L (2008).  Role of metallothionein in lung inflammation induced by ozone exposure in mice. Free Radic
        Biol Med 45: 1714-1722. http://dx.doi.Org/10.1016/i.freeradbiomed.2008.09.008.
Ito. K: De Leon. SF: Lippmann. M. (2005). Associations between ozone and daily mortality, analysis and meta-
        analysis. Epidemiology 16: 446-457.
Ito. K: Thurston. GD: Silverman. RA. (2007b). Characterization of PM2.5, gaseous pollutants, and
        meteorological interactions in the context of time-series health effects models. J Expo Sci Environ
        Epidemiol  17: S45-S60.
Iwasaki. T: Takahashi. M: Saito. H: Arito. H. (1998). Adaptation of extrapulmonary responses to ozone exposure
        in conscious rats. Ind Health 36: 57-60.
Jakab, GJ: Hmieleski, RR. (1988). Reduction of influenza virus pathogenesis  by exposure to 0.5  ppm ozone. J
        Toxicol Environ Health 23: 455-472. http://dx.doi.orq/10.1080/15287398809531128.
Jalaludin. BB: Chev. T: O'Toole. Bl: Smith. WT: Capon. AG: Leeder. SR. (2000). Acute effects of low levels of
        ambient ozone on  peak expiratory flow rate in a cohort of Australian children. Int J Epidemiol 29: 549-
        557. http://dx.doi.Org/10.1093/iie/29.3.549.
Jalaludin. BB: O'Toole. Bl: Leeder. SR. (2004). Acute effects of urban ambient air pollution on respiratory
        symptoms, asthma medication use, and doctor visits for asthma in a  cohort of Australian children.
        Environ Res 95:  32-42. http://dx.doi.orq/10.1016/S0013-9351 (03)00038-0.
Janero.  DR. (1990). Malondialdehyde and thiobarbituric acid-reactivity as diagnostic indices of lipid peroxidation
        and peroxidative tissue injury. Free Radic Biol Med 9: 515-540.  http://dx.doi.Org/10.1016/0891-
        5849(90)90131-2.
Jang. AS: Choi. IS:  Yang. SY: Kim. YG: Lee. JH: Park. SW: Park. CS. (2005).  Antioxidant responsiveness in
        BALB/c mice exposed to ozone. Respiration 72: 79-84. http://dx.doi.org/10.1159/000083405.
Janic, B; Umstead,  TM; Phelps, DS; Floras, J. (2005). Modulatory effects of ozone on THP-1 cells in response
        to SP-A stimulation. Am J Physiol Lung Cell Mol Physiol 288: L317-L325.
        http://dx.doi.org/10.1152/aiplung.00125.2004.
Jansson. M: Bergstrom. A. -K: Drakare. S: Blomgvist. P. (2001). Nutrient limitation of bacterioplankton and
        phytoplankton in humic lakes in northern Sweden. Freshw Biol 14: 76-85.
Johnston, C; Holm, B; Gelein, R; Finkelstein, J. (2006). Postnatal lung development: Immediate-early gene
        responses post ozone and LPS exposure. Inhal Toxicol 18: 875-883.
        http://dx.doi. orq/10.1080/08958370600822466.
Johnston, RA:  Mizgerd, JP: Shore, SA. (2005a).  CXCR2 is essential for maximal neutrophil recruitment and
        methacholine responsiveness after ozone exposure. Am J Physiol Lung Cell  Mol Physiol 288: L61-L67.
        http://dx.doi.orq/10.1152/ajplung.OQ101.2004 00101.2004.
Johnston, RA:  Schwartzman, IN: Flynt, L: Shore, SA. (2005b). Role of interleukin-6 in murine airway responses
        to ozone. Am J Physiol Lung Cell Mol Physiol 288:  L390-L397.
        http://dx.doi.org/10.1152/aiplung.00007.2004.
Johnston, RA:  Mizgerd, JP: Flynt, L: Quinton, LJ: Williams, ES: Shore, SA. (2007). Type I interleukin-1 receptor
        is required for pulmonary responses to subacute ozone exposure in mice. Am J Respir Cell Mol Biol 37:
        477-484. http://dx.doi.org/10.1165/rcmb.2006-0315OC.
Jones, SL: Kittelson, J: Cowan, JO: Flannery, EM: Hancox,  RJ: McLachlan, CR: Taylor, DR. (2001). The
        predictive value of exhaled nitric oxide measurements in assessing changes  in asthma control. Am J
        Respir Grit Care Med 164:  738-743.
Jorres, R: Nowak, D: Magnussen, H: Speckin, P: Koschyk, S.  (1996). The effect of ozone exposure on allergen
        responsiveness in  subjects with asthma or rhinitis. Am J Respir Crit Care Med 153: 56-64.
Jorres. RA: Holz. O: Zachgo. W: Timm. P: Koschvk. S: Mullen B: Grimminger. F: Seeger. W: Kelly. FJ: Punster.
        C:  Frischer, T: Lubec, G: Waschewski, M: Niendorf, A: Magnussen, H. (2000). The effect of repeated
        ozone exposures on inflammatory markers in bronchoalveolar lavage fluid and mucosal  biopsies. Am J
        Respir Crit Care Med 161:  1855-1861.
Just, J: Segala, C: Sahraoui, F: Priol, G: Grimfeld, A: Neukirch, F. (2002). Short-term health effects of particulate
        and photochemical air pollution in asthmatic children. Eur Respir J 20: 899-906.
        http://dx.doi. orq/10.1183/09031936.02.00236902.
Draft - Do Not Cite or Quote                        6-247                                  September 2011

-------
Katsouvanni. K: Touloumi. G: Samoli. E: Gryparis. A: Le Tertre. A: Monopolis. Y: Rossi. G: Zmirou. D: Ballester.
        F: Boumghar, A: Anderson, HR: Wojtyniak, B: Paldy, A: Braunstein, R: Pekkanen, J: Schindler, C:
        Schwartz. J. (2001). Confounding and effect modification in the short-term effects of ambient particles
        on total mortality:  Results from 29 European cities within the APHEA2 project. Epidemiology 12: 521-
        531.
Katsouvanni. K: Samet. JM: Anderson. HR: Atkinson. R: Le Tertre. A: Medina. S: Samoli. E: Touloumi. G:
        Burnett. RT: Krewski. D: Ramsay. T: Dominici. F: Peng. RD: Schwartz. J:  Zanobetti. A. (2009). Air
        pollution and health: A European and North American approach (APHENA). (Research Report 142).
        Boston, MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=327.
Kehrl. HR: Hazucha. MJ: Solic. JJ: Bromberg. PA. (1985). Responses of subjects with chronic obstructive
        pulmonary disease after exposures to 0.3 ppm ozone. Am Rev Respir Dis 131: 719-724.
Kehrl. HR: Vincent. LM: Kowalskv. RJ: Horstman. DH: O'Neil. JJ:  McCartney. WH: Bromberg. PA. (1987).
        Ozone exposure increases respiratory epithelial permeability in humans. Am Rev Respir Dis 135: 1124-
        1128.
Kehrl. HR: Peden.  DB: Ball. BA: Folinsbee. LJ: Horstman. DH.  (1999).  Increased specific airway reactivity of
        persons with mild allergic asthma after 7.6  hours of exposure to 0.16 ppm ozone. J Allergy Clin
        Immunol 104: 1198-1204.
Kenvon. NJ: Last.  MS: Eiserich. JP: Morrissev. BM: Temple. LM: Last. JA. (2006).  Differentiation of the roles of
        NO from airway epithelium and inflammatory cells in ozone-induced lung  inflammation. Toxicol Appl
        Pharmacol 215: 250-259. http://dx.doi.Org/10.1016/i.taap.2006.03.005.
Kharitonov. SA: Barnes. PJ. (2000). Clinical aspects of exhaled nitric oxide. Eur Respir J 16: 781-792.
Khatri. SB: Holguin. FC: Ryan.  PB: Mannino. D: Erzurum. SC: league. WG. (2009). Association of ambient
        ozone exposure with airway inflammation and allergy in adults with asthma. J Asthma 46: 777-785.
        http://dx.doi. org/10.1080/02770900902779284.
Kim. CS: Alexis. NE:  Rappold. AG: Kehrl. H: Hazucha. MJ: Lav. JC: Schmitt. MT: Case. M: Devlin. RB:  Peden.
        DB: Diaz-Sanchez, D.  (2011). Lung function and inflammatory responses  in healthy young adults
        exposed to 0.06 ppm ozone for 6.6 hours. Am J Respir Grit Care Med 183: 1215-1221.
        http://dx.doi.ora/10.1164/rccm.201011-1813OC.
Kinney, PL: Thurston, GD: Raizenne, M. (1996). The effects of ambient ozone on lung function in children: A
        reanalysis of six summer camp studies. Environ Health Perspect 104: 170-174.
Kleeberger. SR: Reddv. S: Zhang. L. -Y: Jedlicka. AE. (2000). Genetic susceptibility to ozone-induced lung
        hyperpermeability: Role of toll-like receptor 4. Am J Respir Cell Mol Biol 22: 620-627.
Klestadt. D: Laval-Gillv. P: Foucaud. L: Falla. J. (2005). Influences of ozone exposure upon macrophage
        responsivity to N-formyl-methionyl-leucyl-phenylalanine: Mobility and metabolic changes. Toxicol In
        Vitro 19: 199-206. http://dx.doi.Org/10.1016/i.tiv.2004.08.004.
Kodavanti. UP: Thomas. R: Ledbetter. AD: Schladweiler. MC: Shannahan. JH: Wallenborn. JG: Lund. AK:
        Campen.  MJ: Butler. EO: Gottipolu. RR: Nvska. A:  Richards. JE: Andrews. D: Jaskot. RH: McKee. J:
        Kotha, SR: Patel,  RB:  Parianandi, NL. (2011). Vascular and cardiac impairments in rats Inhaling ozone
        and diesel exhaust particles. Environ Health Perspect 119: 312-318.
        http://dx.doi. org/10.1289/ehp. 1002386.
Kooter, IM: Pennings, Jl_: Fokkens,  PH:  Leseman, PL: Boere, AJ: Gerlofs-Nijland,  ME:  Cassee, FR: Schalk, JA:
        Orzechowski. TJ:  Schaap. MM: Breit. TM: Dormans. JA: van Oostrom. CT: de Vries. A: van Steeg. H.
        (2007). Ozone induces clear cellular and molecular responses in the mouse lung independently of the
        transcription-coupled repair status. JAppI Physiol 102: 1185-1192.
        http://dx.doi.ora/10.1152/iapplphvsiol.00796.2006.
Koren. HS: Devlin. RB: Graham. DE: Mann. R: McGee. MP: Horstman. DH: Kozumbo. WJ: Becker. S: House.
        DE: McDonnell, WF: Bromberg, PA.  (1989). Ozone-induced inflammation  in the lower airways  of
        human subjects. Am J  Respir Grit Care Med 139: 407-415.
Korrick.  SA: Neas. LM: Dockerv. DW: Gold. PR: Allen. GA: Hill. LB: Kimball. KD: Rosner. BA: Speizer. FE.
        (1998). Effects of ozone and other pollutants on the pulmonary function of adult hikers. Environ Health
        Perspect 106: 93-99. http://dx.doi.org/10.1289/ehp.9810693.
Kostikas. K: Papatheodorou. G: Ganas.  K: Psathakis. K: Panagou. P: Loukides. S. (2002). pH in expired breath
        condensate  of patients with inflammatory airway diseases. Am J Respir Grit Care Med 165: 1364-1370.
Draft - Do Not Cite or Quote                       6-248                                 September 2011

-------
Kreit. JW: Gross. KB: Moore. TB: Lorenzen. TJ: D'Arcv. J: Eschenbacher. WL. (1989). Ozone-induced changes
        in pulmonary function and bronchial responsiveness in asthmatics. J Appl Physiol 66: 217-222.
Kulle. TJ: Sauder. LR: Kerr. HP: Farrell. BP: Bermel. MS: Smith. DM. (1982).  Duration of pulmonary function
        adaptation to ozone in humans. Am Ind Hyg Assoc J 43: 832-837.
Kulle, TJ: Sauder, LR: Hebel, JR: Chatham, MD. (1985). Ozone response relationships in healthy nonsmokers.
        Am Rev Respir Dis 132: 36-41.
Lagorio. S: Forastiere. F: Pistelli. R: lavarone. I: Michelozzi. P: Fano. V: Marconi. A: Ziemacki. G: Ostro.  BD.
        (2006). Air pollution and lung function among susceptible adult subjects: A panel study. Environ  Health
        5: 11. http://dx.doi.org/10.1186/1476-069X-5-11.
Lanki. T: Pekkanen. J: Aalto. P: Elosua. R: Berglind. N: D'lppoliti. D: Kulmala. M: Nvberg.  F: Peters. A: Picciotto.
        S: Salomaa, V: Sunyer, J:  Tiittanen, P: Von Klot, S: Forastiere, F. (2006). Associations of traffic-related
        air pollutants with hospitalisation for first acute myocardial infarction: The HEAPSS study. Occup
        Environ Med 63: 844-851.
Larrieu,  S: Jusot, JF: Blanchard, M: Prouvost, H: Declercq, C: Fabre, P: Pascal, l_: Le Tertre, A: Wagner, V:
        Riviere. S: Chardon. B: Borelli. D:  Cassadou. S: Eilstein. D: Lefranc. A. (2007). Short term effects of air
        pollution on hospitalizations for cardiovascular diseases in eight French cities: The PSAS program. Sci
        Total Environ 387: 105-112.
Larsen.  ST: Matsubara. S: McConville. G: Poulsen. SS: Gelfand. EW. (2010). Ozone  increases airway
        hyperreactivity and mucus hyperproduction in mice previously exposed to allergen. J Toxicol Environ
        Health A 73: 738-747. http://dx.doi.Org/10.1080/15287391003614034.
Laskin. PL: Pendino. KJ: Punjabi. CJ: del Valle. MR: Laskin. JD. (1994). Pulmonary and hepatic effects of
        inhaled ozone in rats. Environ Health Perspect 10:  61-64.
Laskin, PL: Heck, DE: Laskin, JD. (1998). Role of inflammatory cytokines and nitric oxide in hepatic and
        pulmonary toxicity. Toxicol Lett 102-103: 289-293.
Laskin. PL: Laskin. JD. (2001). Role of macrophages and inflammatory mediators in chemically induced toxicity.
        Toxicology 160: 111-118.
Laskin. JD: Heck. DE: Laskin.  PL. (1996). Nitric oxide production in the lung and liver following inhalation of the
        pulmonary irritant ozone. Adv Exp Med Biol 387: 141-146.
Last, JA: Reiser, KM: Tyler, WS: Rucker, RB. (1984). Long-term consequences of exposure to ozone. I. Lung
        collagen content. Toxicol Appl Pharmacol 72: 111-118.
Last. JA: Gohil.  K: Mathrani. VC: Kenvon.  NJ.  (2005). Systemic responses to inhaled  ozone in mice: cachexia
        and  down-regulation of liver xenobiotic metabolizing genes. Toxicol Appl Pharmacol 208: 117-126.
        http://dx.doi.0rg/10.1016/i.taap.2005.02.001.
Lawrence. SO: Simpson-Haidaris. PJ. (2004). Regulated de novo biosynthesis of fibrinogen in extrahepatic
        epithelial cells in response to inflammation. Thromb Haemostasis 92: 234-243.
        http://dx.doi.org/10.1160/TH04-01-0024.
Lav. JC: Alexis.  NE: Kleeberger. SR: Roubev. RA: Harris. BD: Bromberg. PA: Hazucha. MJ: Devlin. RB: Peden.
        DB.  (2007). Ozone enhances markers of innate immunity and antigen presentation on airway
        monocytes in healthy individuals. J Allergy Clin Immunol 120: 719-722.
        http://dx.doi.0rg/10.1016/i.iaci.2007.05.005.
Lee, IM: Tsai, SS: Ho, CK: Chiu, HF: Yang, CY. (2007). Air pollution and hospital admissions for congestive
        heart failure in a tropical city: Kaohsiung, Taiwan. Inhal Toxicol 19:  899-904. http://dx.doi.org/781182105
        [pii]10.1080/08958370701479406.
Lee, IM: Tsai, SS: Ho, CK: Chiu, HF: Wu, TN: Yang, CY. (2008a). Air pollution and hospital admissions for
        congestive heart failure: Are there potentially sensitive groups? Environ Res  108: 348-353.
        http://dx.doi.0rg/10.1016/i.envres.2008.07.024.
Lee, JT: Kim, H: Cho, YS: Hong, YC: Ha, EH:  Park, H. (2003b). Air pollution and hospital admissions for
        ischemic heart diseases among individuals 64+ years of age  residing in Seoul, Korea. Arch Environ
        Health  58: 617-623.
Lee, Y -L:  Lin, Y, -C: Lee, Y, -C: Wang, J, -Y: Hsiue, T, -R: Quo, YL. (2004b). Glutathione S-transferase P1 gene
        polymorphism and air pollution as interactive risk factors for childhood asthma. Clin Exp Allergy 34:
        1707-1713.
Leonard, RJ: Charpied, GL: Faddis, B. (1991). Effects of ambient inhaled ozone on vocal fold mucosa in Bonnet
        monkeys. J Voice 5: 304-309. http://dx.doi.org/10.1016/50892-1997(05)80060-8.
Draft - Do Not Cite or Quote                        6-249                                 September 2011

-------
Levy. Jl: Chemervnski. SM: Sarnat. JA. (2005). Ozone exposure and mortality, an empiric Bayes
        metaregression analysis. Epidemiology 16: 458-468.
Lewis. TC: Robins. TG: Dvonch. JT: Keeler. GJ: Yip. FY: Mentz. GB: Lin. X: Parker. EA: Israel. BA: Gonzalez. L:
        Hill. Y. (2005). Air pollution-associated changes in lung function among asthmatic children in Detroit.
        Environ Health Perspect 113: 1068-1075.
Liao. D:  Duan. Y: Whitsel. EA: Zheng. Z. -J: Heiss. G: Chinchilli. VM: Lin. H. -M. (2004a). Association of higher
        levels of ambient criteria pollutants with impaired cardiac autonomic control: a population-based study.
        Am J Epidemiol 159: 768-777.
Liao. D:  Heiss. G: Chinchilli. VM: Duan. Y: Folsom. AR: Lin. HM: Salomaa. V.  (2005). Association  of criteria
        pollutants with plasma hemostatic/inflammatory markers: A population-based study. J Expo Sci Environ
        Epidemiol 15: 319-328.
Lim. Y: Phung. AD: Corbacho. AM: Aung. HH: Maioli. E: Reznick. AZ: Cross. CE: Davis. PA: Valacchi. G. (2006).
        Modulation of cutaneous wound healing by ozone: Differences between young and aged mice. Toxicol
        Lett 160: 127-134. http://dx.doi.Org/10.1016/i.toxlet.2005.06.013.
Lin. S: Bell. EM: Liu. W: Walker. RJ: Kim. NK: Hwang. SA. (2008a). Ambient ozone concentration and hospital
        admissions due to childhood respiratory diseases in New York State, 1991-2001. Environ Res 108: 42-
        47. http://dx.doi.0rg/10.1016/i.envres.2008.06.007.
Linares. C:  Diaz.  J. (2010). Short-term effect of concentrations of fine particulate matter on hospital admissions
        due to cardiovascular and respiratory causes among the over-75 age group in Madrid, Spain. Public
        Health 124: 28-36. http://dx.doi.Org/10.1016/i.puhe.2009.11.007.
Linn. WS: Medwav. DA: Anzar. UT:  Valencia. LM: Spier. CE: FS-D. T: Fischer. DA: Hackney. JD. (1982a).
        Persistence of adaptation to ozone in volunteers exposed repeatedly for six weeks. Am Rev Respir Dis
        125: 491-495.
Linn. WS: Fischer. DA: Medwav. DA: Anzar. UT: Spier. CE: Valencia. LM: Venet. TG: Hackney. JD. (1982b).
        Short-term respiratory effects of 0.12 ppm ozone exposure in volunteers with chronic obstructive
        pulmonary disease. Am Rev Respir Dis 125: 658-663.
Linn. WS: Shamoo. DA: Venet. TG: Spier.  CE: Valencia.  LM: Anzar. UT: Hackney. JD. (1983). Response to
        ozone in volunteers with chronic obstructive pulmonary disease. Arch Environ Occup Health 38: 278-
        283.
Linn. WS: Avol. EL: Shamoo. DA: Spier. CE: Valencia.  LM: Venet. TG:  Fischer. DA: Hackney. JD.  (1986). A
        dose-response study of healthy, heavily exercising men exposed to ozone at concentrations near the
        ambient  air quality standard. Toxicol Ind  Health 2: 99-112.
Linn. WS: Shamoo. DA: Anderson.  KR: Peng. R.  -C: Avol. EL:  Hackney. JD: Gong. H. Jr. (1996).  Short-term air
        pollution exposures and responses in Los Angeles area schoolchildren. J Expo Sci Environ Epidemiol
        6: 449-472.
Lisabeth. LD:  Escobar. JD: Dvonch. JT: Sanchez. BN:  Maiersik. JJ: Brown. PL:  Smith. MA: Morgenstern. LB.
        (2008). Ambient air pollution and risk for ischemic stroke and transient ischemic attack. Ann Neurol 64:
        53-59. http://dx.doi.org/10.1002/ana.21403.
Liu. L: Leech.  JA: Urch. RB: Silverman. FS. (1997). In vivo salicylate hyroxylation: A potential biomarker for
        assessing acute  ozone exposure  and effects in humans. Am J Respir Crit Care Med 156: 1405-1412.
Liu, L: Leech,  JA: Urch, RB: Poon, R: Zimmerman, B: Kubay, JM: Silverman,  FS. (1999). A comparison of
        biomarkers of ozone exposure in  human plasma, nasal lavage, and sputum. Inhal Toxicol 11: 657-674.
Liu. L: Poon. R: Chen. L:  Frescura.  AM: Montuschi. P: Ciabattoni. G: Wheeler. A: Dales. R. (2009a). Acute
        effects of air pollution on pulmonary function, airway inflammation, and oxidative stress in asthmatic
        children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehp11813.
Lu. FL: Johnston. RA: Flvnt. L: Theman. TA: Terry. RD: Schwartzman. IN: Lee. A: Shore. SA. (2006). Increased
        pulmonary responses to acute ozone exposure in obese db/db mice. Am J Physiol Lung Cell Mol
        Physiol 290: L856-L865.  http://dx.doi.org/10.1152/aiplung.00386.2005.
Mann. JK: Balmes. JR: Bruckner. TA: Mortimer. KM: Margolis.  HG: Pratt.  B: Hammond. SK: Lurmann. F: Tager.
        IB. (2010). Short-term effects of air pollution on wheeze in asthmatic  children in  Fresno,  California.
        Environ Health Perspect 118: 1497-1502. http://dx.doi.org/10.1289/ehp.0901292.
Manzer, R;  Wang, J; Nishina, K; McConville, G; Mason, RJ. (2006). Alveolar epithelial cells secrete chemokines
        in response to IL-1beta and lipopolysaccharide but not to ozone. Am  J Respir Cell Mol Biol 34: 158-
        166. http://dx.doi.org/10.1165/rcmb.2005-0205OC.
Draft - Do Not Cite or Quote                       6-250                                 September 2011

-------
Mapp. CE: Fryer. AA: De Marzo. N: Pozzato. V: Padoan. M: Boschetto. P: Strange. RC: Hemmingsen. A:
        Spiteri,  MA. (2002). Glutathione S-transferase GSTP1 is a susceptibility gene for occupational asthma
        induced by isocyanates. J Allergy Clin Immunol 109: 867-872.
        http://dx.doi. orq/10.1067/mai.20Q2.123234.
Mar, TF: Koenig, JQ. (2009). Relationship between visits to emergency departments for asthma and ozone
        exposure in greater Seattle, Washington. Ann Allergy Asthma Immunol 103: 474-479.
Martfnez-Canabal. A: Angora-Perez. M. (2008). Effect of growth hormone on cyclooxygenase-2 expression in
        the hippocampus of rats chronically exposed to ozone.  Int J Neurosci 118: 455-469.
        http://dx.doi. orq/10.1080/00207450701593160.
Martrette. JM: Thornton. SN: Trabalon. M. (2011). Prolonged ozone exposure effects behaviour, hormones and
        respiratory muscles in young  female rats.  Physiol  Behav 103: 302-307.
        http://dx.doi.0rg/10.1016/i.phvsbeh.2011.02.024.
McBride. DE: Koenig. JQ: Luchtel. PL: Williams. PV: Henderson. WR. Jr. (1994).  Inflammatory effects of ozone
        in the upper airways of subjects with asthma. Am J Respir Crit Care Med 149:  1192-1197.
McDonnell. WF: Horstman. DH: Hazucha. MJ: Seal. E. Jr:  Haak. ED: Salaam. SA: House. DE. (1983).
        Pulmonary effects of ozone exposure during exercise: Dose-response characteristics. J Appl Physiol
        54: 1345-1352.
McDonnell. WF: Kehrl.  HR: Abdul-Salaam. S: Ives. PJ: Folinsbee. LJ: Devlin. RB: O'Neil. JJ: Horstman. DH.
        (1991).  Respiratory response of humans exposed to low levels of ozone  for 66 hours. Arch Environ
        Occup Health 46: 145-150.
McDonnell. WF.  (1996). Individual variability in human lung function responses to ozone exposure. Environ
        Toxicol  Pharmacol 2: 171-175.
McDonnell, WF: Stewart, PW: Andreoni, S: Seal, E, Jr: Kehrl,  HR: Horstman, DH: Folinsbee, LJ: Smith,  MV.
        (1997).  Prediction of ozone-induced FEV1 changes: Effects of concentration, duration, and ventilation.
        Am J Respir Crit Care Med 156: 715-722.
McDonnell, WF: Stewart, PW: Smith, MV: Pan, WK: Pan, J. (1999). Ozone-induced respiratory symptoms:
        Exposure-response models and association with lung function. Eur Respir J 14: 845-853.
McDonnell. WF: Stewart. PW: Smith. MV. (2007). The temporal dynamics of ozone-induced FEV1 changes in
        humans: An exposure-response model. Inhal Toxicol  19: 483-494.
McDonnell. WF: Stewart. PW: Smith. MV. (2010). Prediction of ozone-induced lung function responses in
        humans. Inhal Toxicol 22:  160-168. http://dx.doi.ora/10.3109/08958370903089557.
McDonnell, WF,  III: Horstman, DH: Abdul-Salaam,  S: House, DE. (1985a). Reproducibility of individual
        responses to ozone exposure. Am Rev Respir Dis 131: 36-40.
McDonnell. WF.  Ill: Chapman. RS: Leigh. MW: Strope. GL: Collier. AM. (1985b). Respiratory responses of
        vigorously exercising children to 012 ppm ozone exposure. Am Rev Respir Dis 132: 875-879.
Medina-Ramon. M: Zanobetti. A: Schwartz. J. (2006). The  effect of ozone and PM10 on hospital admissions for
        pneumonia and chronic obstructive pulmonary disease: A national multicity study. Am J Epidemiol 163:
        579-588. http://dx.doi.0rg/10.1093/aie/kwjQ78.
Medina-Ramon. M: Schwartz. J. (2008). Who is more vulnerable to die from ozone air pollution? Epidemiology
        19: 672-679.
Messineo, TD: Adams,  WC. (1990). Ozone inhalation effects in females varying widely  in lung size:  Comparison
        with males. J Appl Physiol 69: 96-103.
Metzger. KB: Tolbert. PE: Klein. M: Peel. JL: Flanders. WD: Todd. KH: Mulholland. JA: Ryan. PB: Frumkin. H.
        (2004).  Ambient air pollution and cardiovascular emergency department visits. Epidemiology  15: 46-56.
Metzger. KB: Klein. M: Flanders. WD:  Peel. JL: Mulholland. JA: Langberg. JJ: Tolbert. PE. (2007). Ambient air
        pollution and cardiac arrhythmias in patients with implantable defibrillators. Epidemiology 18:  585-592.
        http://dx.doi. org/10.1097/EDE.Ob013e318124ffOe.
Michelson. PH: Dailev.  L: Devlin. RB: Peden. DB. (1999). Ozone effects on the immediate-phase response to
        allergen in the nasal airways of allergic asthmatic  subjects. Otolaryngol Head Neck Surg 120: 225-232.
Middleton, N: Yiallouros, P: Kleanthous, S: Kolokotroni, O:  Schwartz, J: Dockery,  DW: Demokritou, P:  Koutrakis,
        £. (2008). A 10-year time-series analysis of respiratory and cardiovascular morbidity in Nicosia, Cyprus:
        The effect of short-term changes in air pollution and dust storms. Environ Health 7: 39.
Draft - Do Not Cite or Quote                       6-251                                  September 2011

-------
Mikerov. AN: Umstead. TM: Gan. X: Huang. W: Quo. X: Wang. G: Phelps. PS: Floros. J. (2008b). Impact of
        ozone exposure on the phagocytic activity of human surfactant protein A(SP-A) and SP-A variants. Am
        J Physiol Lung Cell Mol Physiol 294: L121-L130. http://dx.doi.org/10.1152/aipluna.00288.2007.
Miller. FJ: Illina. JW: Gardner. DE. (1978). Effect of urban ozone levels on laboratory-induced respiratory
        infections. Toxicol Lett 2: 163-169.
Mokoena. ML:  Harvey. BH: Oliver. DW: Brink. CB. (2010). Ozone modulates the effects of imipramine on
        immobility in the forced swim test, and nonspecific parameters of hippocampal oxidative stress in the
        rat. Metab Brain Dis 25: 125-133. http://dx.doi.org/10.1007/s11011 -010-9189-7.
Molfino. NA: Wright. SC: Katz. I: Tarlo. S: Silverman. F: McClean. PA: Szalai. JP: Raizenne. M: Slutskv. AS:
        Zamel. N. (1991). Effect of low concentrations of ozone on inhaled allergen responses in asthmatic
        subjects. Lancet 338: 199-203.
Moon. JS: Kim. YS: Kim. JH: Son. BS: Kim. PS: Yang. W. (2009). Respiratory health effects among
        schoolchildren and their relationship to air pollutants in Korea. Int J Environ Health Res 19: 31-48.
        http://dx.doi.org/10.1080/09603120802272201.
Morrow. JD: Hill. KE: Burk. RF: Nammour. TM: Badr. KF:  Roberts. LJ. 2nd. (1990). A series of prostaglandin F2-
        like compounds are produced in vivo in humans by a non-cyclooxygenase,  free radical-catalyzed
        mechanism. PNAS 87: 9383-9387.
Mortimer. KM: Tager. IB: Dockerv.  DW: Neas. LM: Redline.  S. (2000). The effect of ozone on inner-city children
        with asthma:  Identification of susceptible subgroups. Am J Respir Crit Care Med 162: 1838-1845.
Mortimer, KM: Neas, LM: Dockerv, DW:  Redline, S: Tager, IB. (2002). The effect of air pollution on inner-city
        children with asthma. Eur Respir J 19: 699-705. http://dx.doi.org/10.1183/09031936.02.00247102.
Mudwav. IS: Blomberg. A:  Frew. AJ: Holgate. ST: Sandstrom. T: Kelly. FJ. (1999a). Antioxidant consumption
        and repletion kinetics in nasal lavage fluid following exposure of healthy human volunteers to ozone.
        Eur Respir J 13: 1429-1438.
Mudwav. IS: Stenfors. N: Blomberg. A: Helledav. R: Punster. C: Marklund. SL: Frew. AJ: Sandstrom. T:  Kelly.
        FJ. (2001). Differences in basal airway antioxidant concentrations are not predictive of individual
        responsiveness to ozone: A comparison of healthy and mild asthmatic subjects. Free Radic Biol Med
        31: 962-974.
Mudwav, IS: Kelly, FJ. (2004a). An investigation of inhaled ozone dose and the magnitude of airway
        inflammation  in healthy adults. Am J Respir Crit Care Med 169: 1089-1095.
Murphy. SD: Ulrich. CE: Frankowitz. SH: Xintaras. C.  (1964). Altered function in animals inhaling low
        concentrations of ozone and  nitrogen dioxide. Am Ind  Hyg Assoc J 25: 246-253.
Murugan. A: Prvs-Picard. C: Calhoun. WJ. (2009). Biomarkers in  asthma. Curr Opin Pulm Med 15: 12-18.
        http://dx.doi. org/10.1097/MCP.Ob013e32831 de235.
Naeher, LP: Holford, TR: Beckett, WS: Belanger, K: Triche,  EW: Bracken, MB: Leaderer, BP (1999). Healthy
        women's PEF variations with ambient summer concentrations of PM10, PM2.5, SO42-, H+, and O3.
        Am J Respir Crit Care Med 160: 117-125.
Nakamura, K: Matsunaga, K. (1998). Susceptibility of natural killer (NK) cells to reactive oxygen species (ROS)
        and their restoration by the mimics of superoxide dismutase (SOD). Cancer Biother Radiopharm 13:
        275-290.
Neas, LM: Dockerv, DW: Koutrakis, P: Tollerud, DJ: Speizer, FE. (1995).  The association of ambient air pollution
        with twice daily peak expiratory flow rate measurements in children. Am J Epidemiol 141: 111-122.
Neas. LM: Dockerv. DW: Koutrakis. P: Speizer. FE. (1999). Fine particles and peak flow in children: Acidity
        versus mass. Epidemiology 10: 550-553.
Neidell.  M. (2009). Information, avoidance behavior, and health: The effect of ozone on asthma hospitalizations.
        Journal of Human Resources 44: 450-478.
Neidell,  M: Kinnev, PL. (2010). Estimates of the association between ozone and asthma hospitalizations that
        account for behavioral responses to air quality information. Environ  Sci Pol  13: 97-103.
        http://dx.doi.0rg/10.1016/i.envsci.2009.12.006.
Neuberger,  M:  Schimek, MG:  Horak, F, Jr: Moshammer, H:  Kundi, M:  Frischer, T: Gomiscek, B: Puxbaum, H:
        Hauck. H: AUPHEP-Team. (2004). Acute  effects of particulate matter on respiratory diseases,
        symptoms and functions:  Epidemiological results of the Austrian Projects on Health Effects of
        Particulate Matter (AUPHEP). Atmos Environ 38: 3971-3981.
Draft - Do Not Cite or Quote                       6-252                                 September 2011

-------
Newson. EJ: Krishna. MT: Lau. LCK: Howarth. PH: Holgate. ST: Frew. AJ. (2000). Effects of short-term
        exposure to 0.2 ppm ozone on biomarkers of inflammation in sputum, exhaled nitric oxide, and lung
        function in subjects with mild atopic asthma. J Occup Environ Med 42: 270-277.
Nickmilder. M: De Burbure. C: Svlviane. C: Xavier. D: Alfred. B: Alain. D. (2007). Increase of exhaled nitric oxide
        in children exposed to low levels of ambient ozone. J Toxicol Environ Health A 70: 270-274.
Nightingale. JA: Rogers. DF:  Chung. KF: Barnes. PJ. (2000). No effect of inhaled budesonide on the response
        to inhaled ozone in normal subjects. Am J Respir Crit Care Med 161: 479-486.
O'Connor, GT:  Neas, l_: Vaughn, B: Kattan, M: Mitchell, H: Grain, EF: III, ER: Gruchalla, R: Morgan, W: Stout, J:
        Adams. GK: Lippmann.  M. (2008). Acute respiratory health effects of air pollution on children with
        asthma in US  inner cities. J Allergy Clin Immunol 121: 1133-1139.
        http://dx.doi.0rg/10.1016/i.iaci.2008.02.020.
Orazzo. F: Nespoli. L: Ito.  K: Tassinari. D: Giardina.  D: Funis. M: Cecchi. A: Trapani. C: Forgeschi. G: Vignini. M:
        Nosetti. L: Pigna.  S:  Zanobetti. A. (2009). Air pollution, aeroallergens, and emergency room visits for
        acute respiratory diseases and gastroenteric disorders among young children in six Italian cities.
        Environ Health Perspect 117: 1780-1785. http://dx.doi.org/10.1289/ehp.0900599.
Oslund. KL: Hyde. DM:  Putney. LF: Alfaro.  MF: Walbv. WF: Tyler. NK: Schelegle. ES. (2008). Activation of
        neurokinin-1 receptors during ozone inhalation contributes to epithelial injury and repair. Am J Respir
        Cell Mol Biol 39: 279-288.  http://dx.doi.org/10.1165/rcmb.2008-0009OC.
Oslund. KL: Hyde. DM:  Putney. LF: Alfaro.  MF: Walbv. WF: Tyler. NK: Schelegle. ES. (2009). Activation of
        calcitonin gene-related peptide receptor during ozone inhalation contributes to airway epithelial injury
        and repair. Toxicol Pathol 37: 805-813. http://dx.doi.org/10.1177/0192623309345691.
Ostro. B: Lipsett. M: Mann. J: Braxton-Owens. H: White. M. (2001). Air pollution and exacerbation of asthma in
        African-American children  in Los Angeles. Epidemiology 12: 200-208.
Ostro. B: Broadwin. R:  Green. S: Feng. WY: Lipsett. M. (2006). Fine particulate air pollution and mortality in
        nine California  counties: Results from CALFINE. Environ Health Perspect 114: 29-33.
Oudin, A: Stromberg, U: Jakobsson, K: Stroh, E: Bjork, J. (2010). Estimation of short-term effects of air pollution
        on stroke hospital admissions in southern Sweden. Neuroepidemiology 34:  131-142.
        http://dx.doi.org/10.1159/000274807.
Oyarzun, M: Dussaubat, N: Gonzalez, S. (2005). Effect of 0.25 ppm ozone exposure on pulmonary damage
        induced by bleomycin. Biol Res 38: 353-358.
Park. JW: Lim. YH: Kvung. SY: An. CH: Lee. SP: Jeong. SH: Ju. S. -Y. (2005a). Effects of ambient particulate
        matter on peak expiratory flow rates and respiratory symptoms of asthmatics during Asian dust periods
        in Korea.  Respirology 10: 470-476. http://dx.doi.Org/10.1111/i.1440-1843.2005.00728.x.
Park. SK: O'Neill.  MS: Vokonas.  PS: Sparrow. D: Schwartz. J. (2005b). Effects of air pollution on heart rate
        variability: The VA Normative Aging Study. Environ Health Perspect 113: 304-309.
Park. SK: O'Neill.  MS: Stunder. BJB: Vokonas. PS: Sparrow. D: Koutrakis. P: Schwartz. J. (2007). Source
        location of air pollution and cardiac autonomic function: Trajectory cluster analysis for exposure
        assessment. J  Expo  Sci Environ Epidemiol 17: 488-497.
Park. SK: O'Neill.  MS: Vokonas.  PS: Sparrow. D: Wright. RO:  Coull. B: Nie. H:  Hu. H: Schwartz. J. (2008). Air
        pollution and heart rate variability: Effect modification by chronic lead exposure.  Epidemiology 19: 111-
        120. http://dx.doi.0rg/10.1097/EDE.Ob013e31815c408a.
Passannante. AN: Hazucha. MJ: Bromberg. PA:  Seal. E: Folinsbee. L: Koch. G. (1998). Nociceptive
        mechanisms modulate ozone-induced human lung function decrements. J Appl Physiol 85: 1863-1870.
Peacock, Jl_: Anderson, HR: Bremner, SA: Marston, L: Seemungal, TA: Strachan, DP: Wedzicha, JA. (2011).
        Outdoor air pollution  and respiratory health in patients with COPD. Thorax 66: 591-596.
        http://dx.doi. orq/10.1136/thx.2Q10.155358.
Peden, DB: Setzer, RW, Jr: Devlin, RB. (1995).  Ozone exposure has both  a priming effect on allergen-induced
        responses and an intrinsic inflammatory action in the nasal airways of perennially allergic asthmatics.
        Am J Respir Crit Care Med 151: 1336-1345.
Peden, DB: Boehlecke, B: Horstman, D: Devlin, R. (1997). Prolonged acute exposure to 0.16 ppm ozone
        induces eosinophilic  airway inflammation in asthmatic subjects with allergies. J Allergy Clin Immunol
        100: 802-808.
Peden, DB. (2001). Air pollution in asthma: Effect of pollutants on airway inflammation. Ann Allergy Asthma
        Immunol 3: 12-17.
Draft - Do Not Cite or Quote                        6-253                                 September 2011

-------
Peden. DB. (2011). The role of oxidative stress and innate immunity in O(3) and endotoxin-induced human
        allergic airway disease. Immunol Rev 242: 91-105. http://dx.doi.Org/10.1111/i.1600-065X.2011.01035.x.
Peel. JL: Metzger. KB: Klein. M: Flanders. WD: Mulholland. JA: Tolbert. PE. (2007). Ambient air pollution and
        cardiovascular emergency department visits in potentially sensitive groups. Am J Epidemiol 165: 625-
        633.
Pellegrino. R: Viegi. G: Brusasco. V: Crapo. RO: Burgos. F: Casaburi. R: Coates. A: van der Grinten. CP:
        Gustafsson. P:  Hankinson. J: Jensen. R: Johnson. DC: Maclntvre. N: McKay. R: Miller. MR: Navaias.
        D: Pedersen, OF: Wanger, J. (2005). Interpretative strategies for lung function tests.  Eur Respir J 26:
        948-968. http://dx.doi.org/10.1183/09031936.05.00035205.
Peng. RD: Dominici. F: Pastor-Barriuso. R: Zeger. SL: Samet. JM. (2005). Seasonal analyses of air pollution
        and mortality in 100 US cities. Am J Epidemiol 161: 585-594.
Perepu. RS: Garcia. C: Postal. D: Sethi. R. (2010). Enhanced death signaling in ozone-exposed ischemic-
        reperfused hearts. Mol Cell Biochem 336: 55-64. http://dx.doi.orq/10.1007/s11010-009-0265-4.
Pereyra-Munoz, N: Rugerio-Vargas, C: Angoa-Perez, M: Borgonio-Perez, G: Rivas-Arancibia, S. (2006).
        Oxidative damage in substantia nigra and striatum of rats chronically exposed to ozone. J Chem
        Neuroanat31: 114-123.  http://dx.doi.Org/10.1016/i.ichemneu.2005.09.006.
Peters, A: Dockery, DW: Muller, JE: Mittleman,  MA. (2001). Increased particulate air pollution and the triggering
        of myocardial infarction.  Circulation 103: 2810-2815.
Peterson. DC: Andrews. HL. (1963). The role of ozone in radiation avoidance in the mouse. Radiat Res 19: 331-
        336.
Petroeschevskv. A: Simpson. RW: Thalib. L: Rutherford. S. (2001). Associations between outdoor air pollution
        and hospital admissions in Brisbane, Australia. Arch Environ Occup Health 56: 37-52.
Pichavant, M: Goya, S: Meyer, EH: Johnston, RA: Kim, HY: Matangkasombut, P: Zhu, M: Iwakura, Y: Savage,
        PB: DeKruvff. RH: Shore. SA: Umetsu. DT (2008). Ozone exposure in a mouse model induces airway
        hyperreactivity that requires the presence of natural killer T cells and IL-17. J Exp Med 205: 385-393.
        http://dx.doi. org/10.1084/jem.20071507.
Pinkerton. KE: Brodv. AR:  Miller.  FJ: Crapo. JD. (1989). Exposure to low levels of ozone results in enhanced
        pulmonary retention of inhaled asbestos fibers. Am J Respir Crit Care Med 140: 1075-1081.
Plopper, CG:  Mango,  GW: Hatch, GE: Wong, VJ: Toskala, E: Reynolds, SD: Tarkington, BK: Stripp, BR. (2006).
        Elevation of susceptibility to  ozone-induced acute tracheobronchial injury in transgenic mice deficient in
        Clara cell secretory protein. Toxicol Appl Pharmacol 213: 74-85.
        http://dx.doi.0rg/10.1016/i.taap.2005.09.003.
Poloniecki. JD: Atkinson. RW: Ponce de Leon. A: Anderson. HR. (1997). Daily time series for cardiovascular
        hospital admissions and previous day's air pollution in London, UK. Occup Environ Med 54: 535-540.
Prescott, GJ:  Cohen,  GR:  Elton, RA:  Fowkes, FGR: Agius, RM. (1998). Urban air pollution and cardiopulmonary
        ill health: A 145 year time series study. Occup Environ Med 55: 697-704.
Pulfer.  MK:  Murphy. RC. (2004). Formation of biologically active oxysterols during ozonolysis of cholesterol
        present in lung surfactant. J  Biol Chem 279: 26331-26338.
Pulfer.  MK: Taube. C: Gelfand. E:  Murphy. RC. (2005). Ozone exposure in vivo and formation of biologically
        active oxysterols in the lung. J  Pharmacol Exp Ther 312: 256-264.
Qian, Z: Lin, H, -M: Chinchilli, VM: Lehman, EB: Duan, Y: Craig, TJ: Wilson, WE: Liao, D: Lazarus, SC:
        Bascom. R. (2009). Interaction of ambient air pollution with asthma medication on exhaled nitric oxide
        among asthmatics. Arch Environ Occup Health 64: 168-176.
        http://dx.doi. org/10.1080/19338240903240616.
Que. LG: Stiles. JV: Sundv. JS: Foster. WM. (2011). Pulmonary function, bronchial reactivity, and epithelial
        permeability are response phenotypes to ozone and develop differentially in healthy  humans. J Appl
        Physiol 111: 679-687. http://dx.doi.org/10.1152/iapplphysiol.00337.2011.
Rabinovitch. N: Zhang. LN: Murphy. JR: Vedal.  S: Dutton. SJ: Gelfand. EW. (2004). Effects of wintertime
        ambient air pollutants on asthma exacerbations in urban minority children with moderate to severe
        disease. J Allergy Clin Immunol 114: 1131-1137. http://dx.doi.Org/10.1016/i.iaci.2004.08.026.
Raizenne. M: Stern. B: Burnett. R: Spengler. J.  (1987). Acute respiratory function and transported air pollutants:
        Observational studies (paper no. 87-32.6). In Proceedings of the 80th Annual Meeting of the Air
        Pollution Control Association (pp. 18). New York, NY: Air Pollution Control Association.
Draft - Do Not Cite or Quote                        6-254                                 September 2011

-------
Raizenne. ME: Burnett. RT: Stern. B: Franklin. CA: Spengler. JD. (1989). Acute lung function responses to
        ambient acid aerosol exposures in children. Environ Health Perspect 79: 179-185.
Ren. C: Williams. GM: Mengersen. K: Morawska. L: long. S. (2008). Does temperature modify short-term
        effects of ozone on total mortality in 60 large eastern US communities? An assessment using the
        NMMAPS data. Environ Int 34: 451-458.
Revis. NW: Major. T: Dalbev. WE. (1981). Cardiovascular effects of ozone  and cadmium inhalation in the rat. In
        Proceedings of the research planning workshop on health effects  of oxidants. (EPA-600/9-81-001).
        Raleigh, NC: U.S. Environmental Protection Agency.
Rich. DQ: Schwartz. J: Mittleman. MA: Link. M: Luttmann-Gibson. H: Catalano. PJ: Speizer. FE: Dockerv.  DW.
        (2005). Association of short-term ambient air pollution concentrations and ventricular arrhythmias. Am J
        Epidemiol 161: 1123-1132.
Rich. DQ: Kim. MH: Turner. JR: Mittleman. MA: Schwartz. J:  Catalano. PJ: Dockerv. DW. (2006a). Association
        of ventricular arrhythmias detected by implantable cardioverter defibrillator and ambient air pollutants in
        the St Louis, Missouri metropolitan area. Occup Environ Med 63: 591-596.
Rich. DQ: Mittleman. MA: Link. MS:  Schwartz. J: Luttmann-Gibson. H: Catalano. PJ: Speizer. FE:  Gold. PR:
        Dockerv. DW. (2006b). Increased risk of paroxysmal atrial fibrillation episodes associated with acute
        increases in ambient air pollution. Environ Health Perspect 114: 120-123.
Rich. DQ: Kipen. HM: Zhang. J: Kamat. L: Wilson. AC: Kostis. JB. (2010). Triggering oftransmural infarctions,
        but not nontransmural infarctions, by ambient fine particles. Environ Health Perspect 118: 1229-1234.
        http://dx.doi.org/10.1289/ehp.0901624.
Riediker. M: Monn. C: Koller. T: Stahel. WA: Wuthrich. B. (2001). Air pollutants enhance rhinoconjunctivitis
        symptoms in pollen-allergic individuals. Ann Allergy Asthma Immunol 87: 311-318.
Rivas-Arancibia, S: Vazquez-Sandoval, R: Gonzalez-Kladiano, D: Schneider-Rivas, S: Lechuga-Guerrero, A.
        (1998). Effects of ozone exposure in rats on memory and levels of brain and pulmonary superoxide
        dismutase. Environ Res 76: 33-39.
Rivas-Arancibia, S: Dorado-Martinez, C: Borgonio-Perez, G:  Hiriart-Urdanivia, M: Verdugo-Diaz, L: Duran-
        Vazguez. A: Colin-Barangue. L: Avila-Costa.  MR. (2000). Effects oftaurine on ozone-induced memory
        deficits and  lipid peroxidation levels in brains of young, mature,  and old rats. Environ Res 82: 7-17.
        http://dx.doi. org/10.1006/enrs. 1999.3996.
Rivas-Arancibia. S: Guevara-Guzman. R: Lopez-Vidal. Y: Rodrfguez-Martfnez.  E: Gomes. MZ: Angoa-Perez.  M:
        Raisman-Vozari. R.  (2010). Oxidative stress caused by  ozone exposure induces loss of brain repair in
        the hippocampus of adult rats. Toxicol Sci  113: 187-197. http://dx.doi.org/10.1093/toxsci/kfp252.
Rodriguez. C: Tonkin. R: Hevworth. J: Kusel.  M: De Klerk. N: Sly. PD:  Franklin.  P: Runnion. T: Blocklev. A:
        Landau. L: Hinwood. AL. (2007). The relationship between outdoor air quality and respiratory
        symptoms in young  children. Int J Environ Health Res 17: 351-360.
        http://dx.doi.org/10.1080/09603120701628669.
Romieu. I: Meneses. F: Ruiz. S: Sienra. JJ: Huerta. J: White. MC: Etzel. RA. (1996). Effects of air  pollution on
        the respiratory health of asthmatic children living in Mexico City. Am J Respir Crit Care Med 154: 300-
        307.
Romieu. I: Meneses. F: Ruiz. S: Huerta. J: Sienra. JJ: White. M:  Etzel. R: Hernandez. M.  (1997). Effects of
        intermittent ozone exposure on peak expiratory flow and respiratory symptoms among asthmatic
        children in Mexico City. Arch Environ Occup Health 52: 368-376.
Romieu. I: Meneses. F: Ramirez. M: Ruiz. S:  Padilla.  RP: Sienra. JJ: Gerber. M: Grievink. L: Dekker. R: Walda.
        I: Brunekreef, B. (1998a). Antioxidant supplementation and respiratory functions  among workers
        exposed to high levels of ozone. Am J Respir Crit Care  Med 158: 226-232.
Romieu. I: Sienra-Monge. JJ: Ramirez-Aguilar. M: Tellez-Roio. MM: Moreno-Macias. H: Reyes-Ruiz. Nl: Del
        Rio-Navarro, BE: Ruiz-Navarro, MX: Hatch, G: Slade, R: Hernandez-Avila, M. (2002). Antioxidant
        supplementation and lung functions among children with asthma exposed to high levels of air
        pollutants. Am J Respir Crit Care Med 166: 703-709.
Romieu, I: Sienra-Monge, JJ: Ramirez-Aguilar, M: Moreno-Macias, H: Reyes-Ruiz,  Nl: Estela del Rio-Navarro,
        B: Hernandez-Avila. M: London. SJ. (2004a). Genetic polymorphism of GSTM1 and antioxidant
        supplementation influence lung function in relation to ozone exposure in asthmatic children in  Mexico
        City. Thorax 59: 8-10.
Draft - Do Not Cite or Quote                        6-255                                  September 2011

-------
Romieu. I: Ramirez-Aguilar. M: Sienra-Monqe. JJ: Moreno-Macias. H: Del Rio-Navarro. BE: David. G: Marzec.
        J: Hernandez-Avila, M: London, S. (2006). GSTM1 and GSTP1 and respiratory health in asthmatic
        children exposed to ozone. Eur Respir J 28: 953-959. http://dx.doi.org/10.1183/09031936.06.00114905.
Romieu. I: Barraza-Villarreal. A: Escamilla-Nunez. C: Almstrand. AC: Diaz-Sanchez. D: Sly. PD: Olin. AC.
        (2008). Exhaled breath malondialdehyde as a marker of effect of exposure to air pollution in children
        with asthma. J Allergy Clin Immunol 121: 903-909. http://dx.doi.Org/10.1016/i.iaci.2007.12.004.
Romieu. I: Barraza-Villarreal. A: Escamilla-Nunez. C: Texcalac-Sangrador. JL: Hernandez-Cadena. L: Dfaz-
        Sanchez, D: De Batlle, J: Del Rio-Navarro, BE. (2009). Dietary intake, lung function and airway
        inflammation in Mexico City school children exposed to air pollutants. Respir Res 10: 122.
Rosenfeld. MA: Leonova. VB: Konstantinova. ML: Razumovskii. SD. (2009). Self-assembly of fibrin monomers
        and fibrinogen aggregation during ozone oxidation. Biochemistry (Mosc) 74: 41-46.
        http://dx.doi. org/10.1134/S0006297909010064.
Ross. MA: Perskv. VW:  Scheff. PA: Chung. J: Curtis. L: Ramakrishnan. V: Wadden. RA: Hrvhorczuk. DO.
        (2002). Effect of ozone and aeroallergens on the respiratory health of asthmatics. Arch Environ Occup
        Health 57: 568-578.
Rozenfeld. MA: Leonova. VB: Konstantinova. ML: Razumovskii. SD: Makarov. VA:  Nevedrova. OE:
        Belozerskaja, GG. (2008). Disturbance of functional properties of fibrinogen under ozone oxidation.
        Dokl Biochem  Biophys 422: 315-318.  http://dx.doi.org/10.1134/S1607672908050165.
Rudez. G: Janssen. NA: Kilinc. E: Leebeek. FW: Gerlofs-Niiland. ME: Spronk. HM: ten Gate. H: Cassee. FR: de
        Maat, MR (2009). Effects of ambient air pollution on hemostasis and inflammation. Environ Health
        Perspect 117: 995-1001.
Ruidavets. J. -B: Cassadou. S: Cournot. M: Bataille. V: Mevbeck. M: Ferrieres. J. (2005a). Increased resting
        heart rate with pollutants in a population based study. J Epidemiol Community Health 59: 685-693.
Ruidavets. J. -B: Cournot. M: Cassadou. S: Giroux. M: Mevbeck. M: Ferrieres. J. (2005b). Ozone air pollution is
        associated with acute  myocardial infarction. Circulation 111: 563-569.
Samet, JM: Zeger, Sl_:  Dominici, F: Curriero, F: Coursac, I: Dockery, DW: Schwartz, J: Zanobetti, A. (2000). The
        national morbidity, mortality, and air pollution study. Part II:  Morbidity, mortality, and air pollution in the
        United States. Cambridge, MA: Health Effects Institute.
Samet, JM: Hatch, GE:  Horstman, D: Steck-Scott, S: Arab, L: Bromberg, PA: Levine, M: McDonnell, WF: Devlin,
        RB. (2001). Effect of antioxidant supplementation on ozone-induced lung injury in human subjects. Am
        J Respir Grit Care Med 164: 819-825.
Samoli, E: Zanobetti, A: Schwartz, J: Atkinson, R: Le Tertre, A: Schindler, C: Perez, L:  Cadum, E: Pekkanen,  J:
        Paldv. A: Touloumi. G: Katsouvanni. K. (2009). The temporal pattern of mortality responses to ambient
        ozone in the APHEA project. J Epidemiol Community Health 63: 960-966.
        http://dx.doi.org/10.1136/iech.2008.084012.
Santiago-Lopez. D: Bautista-Martfnez. JA: Reyes-Hernandez.  Cl: Aguilar-Martfnez. M: Rivas-Arancibia. S.
        (2010). Oxidative stress, progressive damage in the substantia nigra and plasma dopamine oxidation,
        in rats chronically exposed to ozone. Toxicol Lett 197: 193-200.
        http://dx.doi. org/10.1016/i.toxlet.2Q10.05.020.
Santucci. D: Sorace. A:  Francia. N: Aloe. L: Alleva. E. (2006). Prolonged prenatal exposure to low-level ozone
        affects aggressive behaviour as well as NGF and BDNF levels in the central nervous system of CD-1
        mice. Behav Brain Res 166: 124-130. http://dx.doi.Org/10.1016/i.bbr.2005.07.032.
Sarnat. SE: Suh.  HH: Coull. BA: Schwartz. J: Stone. PH: Gold. DR.  (2006a). Ambient particulate air pollution
        and cardiac arrhythmia in a panel of older adults in Steubenville, Ohio. Occup Environ Med 63: 700-
        706.
Sathishkumar. K: Hague. M:  Perumal. TE: Francis. J: Uppu. RM. (2005). A major ozonation product of
        cholesterol, 3beta-hydroxy-5-oxo-5,6-secocholestan-6-al, induces apoptosis in H9c2 cardiomyoblasts.
        FEBS Lett 579: 6444-6450.
Sathishkumar. K: Gao.  X: Raghavamenon. AC: Parinandi. N: Prvor.  WA: Uppu. RM. (2009). Cholesterol
        secoaldehyde  induces apoptosis in H9c2 cardiomyoblasts through  reactive oxygen species involving
        mitochondrial and death receptor pathways. Free Radic Biol Med 47: 548-558.
        http://dx.doi.0rg/10.1016/i.freeradbiomed.2009.05.020.
Draft - Do Not Cite or Quote                       6-256                                  September 2011

-------
Scannell. C: Chen. L: Aris. RM: Tager. I: Christian. D: Ferrando. R: Welch. B: Kelly. T: Balmes. JR. (1996).
        Greater ozone-induced inflammatory responses in subjects with asthma. Am J Respir Crit Care Med
        154: 24-29.
Scarlett. JF: Abbott. KJ: Peacock. JL: Strachan. DP: Anderson. HR. (1996). Acute effects of summer air
        pollution on  respiratory function in primary school children in southern England. Thorax 51: 1109-1114.
Schelegle. ES: Adams. WC. (1986). Reduced exercise time in competitive simulations consequent to low level
        ozone exposure. Med Sci Sports  Exerc 18: 408-414.
Schelegle, ES: Siefkin, AD: McDonald, RJ. (1991).  Time course of ozone-induced neutrophilia in normal
        humans. Am J Respir Crit Care Med 143: 1353-1358.
Schelegle. ES: Morales. CA: Walbv. WF: Marion. S: Allen. RP. (2009). 6.6-hour inhalation of ozone
        concentrations from 60 to 87 parts per billion in healthy humans. Am J Respir Crit Care Med 180: 265-
        272. http://dx.doi.ora/10.1164/rccm.200809-1484OC.
Schildcrout. JS: Sheppard.  L: Lumlev. T: Slaughter. JC: Koenig. JQ: Shapiro. GG. (2006). Ambient air pollution
        and asthma exacerbations in children: An eight-city analysis. Am J Epidemiol 164: 505-517.
        http://dx.doi. orq/10.1093/aie/kwi225.
Schmekel. B: Ahlner. J: Malmstrom. M: Venae. P. (2001). Eosinophil cationic protein (ECP) in saliva: A new
        marker of disease activity in bronchial asthma. Respir Med 98: 670-675.
        http://dx.doi. orq/10.1053/rmed.20Q1.1123.
Schwartz. J. (2005a). How sensitive is the association between ozone and daily deaths to control for
        temperature? Am J Respir Crit Care Med 171: 627-631.
Schwartz. J: Litoniua. A: Suh. H: Verrier. M: Zanobetti. A: Svring. M:  Nearing. B: Verrier. R: Stone. P:
        MacCallum. G: Speizer. FE: Gold. DR. (2005). Traffic related pollution and heart rate variability in a
        panel of elderly subjects. Thorax  60: 455-461.
Schwartz. J. (2005b). Who is sensitive to extremes of temperature? A case-only analysis. Epidemiology 16: 67-
        72. http://dx.doi.ora/10.1097/01.ede.0000147114.25957.71.
Seal, E, Jr: McDonnell, WF: House, DE: Salaam, SA: Dewitt, PJ: Butler, SO: Green, J: Raggio, L. (1993). The
        pulmonary response of white and black adults to six concentrations of ozone. Am J Respir Crit Care
        Med 147: 804-810.
Seal, E, Jr: McDonnell, WF: House, DE. (1996). Effects of age, socioeconomic status, and menstrual cycle on
        pulmonary response to ozone. Arch Environ Occup Health 51: 132-137.
Belgrade. MK: Daniels. MJ: Grose.  EC. (1990). Acute, subchronic, and chronic exposure to a simulated urban
        profile of ozone: Effects on extrapulmonary natural killer cell activity and  lymphocyte mitogenic
        responses. Inhal Toxicol 2: 375-389.
Selwvn. BJ: Stock. TH: Hardy. RJ: Chan. FA: Jenkins. DE: Kotchmar. DJ: Chapman. RS. (1985). Health effects
        of ambient ozone exposure in vigorously exercising adults. In Evaluation of the scientific basis for
        ozone/oxidants standards:  Proceedings of an apca international specialty conference (pp. 281-296).
        Houston, TX: Air Pollution Control Association.
Servais, S: Boussouar, A: Molnar, A: Douki, T:  Peguignot, JM:  Favier, R. (2005). Age-related sensitivity to lung
        oxidative stress during ozone exposure.  Free Radic Res 39: 305-316.
        http://dx.doi. orq/10.1080/10715760400011098.
Sharkhuu, T: Doerfler, PL: Copeland, C: Luebke, RW: Gilmour, Ml. (2011). Effect  of maternal exposure to ozone
        on reproductive outcome and immune, inflammatory, and allergic responses in the offspring. J
        Immunotoxicol 8: 183-194. http://dx.doi.org/10.3109/1547691X.2011.568978.
Shore,  SA: Rivera-Sanchez, YM: Schwartzman, IN: Johnston, RA. (2003). Responses to ozone are increased in
        obese mice. J Appl Physiol 95: 938-945.
Sienra-Monge. JJ: Ramirez-Aguilar. M: Moreno-Macias. H: Reyes-Ruiz. Nl:  Del Rio-Navarro. BE: Ruiz-Navarro.
        MX:  Hatch, G: Crissman, K: Slade, R: Devlin, RB: Romieu,  I. (2004). Antioxidant supplementation and
        nasal inflammatory responses among young asthmatics exposed to high levels of ozone. Clin Exp
        Immunol 138: 317-322. http://dx.doi.orq/10.1111/i.1365-2249.2004.02606.x.
Silverman, RA: Ito, K. (2010). Age-related association of fine particles and ozone  with severe acute asthma in
        New York City. J Allergy Clin Immunol 125:  367-373.e365. http://dx.doi.Org/10.1016/i.iaci.2009.10.061.
Silverman. RA: Ito. K: Freese. J: Kaufman. BJ: De Claro. D: Braun. J: Prezant. DJ. (2010). Association of
        ambient  fine particles with  out-of-hospital cardiac arrests in New York City. Am J Epidemiol 172: 917-
        923. http://dx.doi.orq/10.1093/aie/kwg217.
Draft - Do Not Cite or Quote                       6-257                                  September 2011

-------
Simonian. NA: Covle. JT. (1996). Oxidative stress in neurodegenerative diseases. Annu Rev Pharmacol Toxicol
        36: 83-106. http://dx.doi.org/10.1146/annurev.pa.36.040196.000503.
Simpson. R: Williams. G: Petroeschevskv. A: Best. T: Morgan. G: Denison. L: Hinwood. A: Neville. G. (2005).
        The short-term effects of air pollution on hospital admissions in four Australian cities. Aust N Z J Public
        Health 29: 213-221.
Sinclair. AH: Tolsma. D. (2004). Associations and lags between air pollution and acute respiratory visits in an
        ambulatory care setting: 25-month results from the aerosol research and inhalation epidemiological
        study. J Air Waste Manag Assoc 54: 1212-1218.
Sinclair. AH: Edgerton. ES: Wvzga. R: Tolsma. D. (2010). A two-time-period comparison of the effects of
        ambient air pollution on outpatient visits for acute respiratory illnesses. J Air Waste Manag Assoc 60:
        163-175.  http://dx.doi.0rg/10.3155/1047-3289.60.2.163.
Smith.  RL: Xu. B: Switzer. P. (2009b). Reassessing the relationship between ozone and short-term mortality in
        U.S.  urban communities. Inhal Toxicol 21: 37-61. http://dx.doi.org/10.1080/08958370903161612.
Solic, JJ: Hazucha, MJ: Bromberg, PA. (1982). The acute effects of 0.2 ppm ozone in patients with chronic
        obstructive pulmonary disease. Am Rev Respir Dis 125: 664-669.
Son. JY: Bell.  ML:  Lee. JT. (2010).  Individual exposure to air pollution and lung function in Korea: Spatial
        analysis using multiple exposure approaches. Environ Res 110: 739-749.
        http://dx.doi.0rg/10.1016/i.envres.2010.08.003.
Soulage. C: Perrin. D: Cottet-Emard. J. -M: Peguignot. J: Dalmaz. Y: Peguignot. J. -M. (2004). Central and
        peripheral changes in catecholamine biosynthesis and turnover in rats after a short period of ozone
        exposure. Neurochem Int45: 979-986.
Spannhake. EW: Reddv. SPM: Jacobv. DB:  Yu. X.  -Y: Saatian. B: Tian. J. (2002). Synergism between rhinovirus
        infection and oxidant pollutant exposure enhances airway epithelial cell cytokine production.  Environ
        Health Perspect 110: 665-670.
Spektor. DM:  Lippmann. M: Liov. PJ: Thurston. GD: Citak. K: James. DJ: Bock. N: Speizer.  FE: Haves. C.
        (1988a). Effects of ambient ozone on respiratory function in active, normal children. Am  Rev Respir Dis
        137:  313-320.
Spektor. DM:  Lippmann. M: Thurston. GD: Liov. PJ: Stecko. J: O'Connor. G: Garshick. E: Speizer. FE: Haves.
        C± (1988b). Effects of ambient ozone on respiratory function in healthy adults exercising outdoors. Am
        Rev Respir Dis 138: 821-828.
Spektor. DM:  Lippmann. M. (1991). Health effects of ambient ozone on healthy children at a summer camp. In
        RL Berglund; DR Lawson; DJ McKee (Eds.), Tropospheric Ozone and the Environment: Papers from
        an International Conference; March 1990; Los Angeles, CA(pp. 83-89). Pittsburgh, PA: Air & Waste
        Management Association.
Stafoggia, M;  Forastiere, F; Faustini, A; Biggeri, A; Bisanti, L; Cadum, E; Cernigliaro, A; Mallone,  S; Pandolfi, P;
        Serinelli. M: Tessari. R: Vigotti. MA: Perucci. CA. (2010). Susceptibility factors to ozone-related
        mortality:  A population-based case-crossover analysis. Am J Respir Crit Care Med 182:  376-384.
        http://dx.doi.org/10.1164/rccm.200908-1269OC.
Steinvil. A: Kordova-Biezuner. L:  Shapira. I:  Berliner. S: Rogowski.  O. (2008). Short-term exposure to air
        pollution and inflammation-sensitive biomarkers. Environ Res 106: 51-61.
Steinvil, A; Fireman, E; Kordova-Biezuner, L; Cohen, M; Shapira, I;  Berliner, S; Rogowski, O. (2009).
        Environmental air pollution has decremental effects on pulmonary function test parameters up to one
        week after exposure. Am J Med  Sci 338: 273-279. http://dx.doi.org/10.1097/MAJ.ObOl 3e3181 adb3ed.
Stenfors, N; Pourazar, J; Blomberg, A; Krishna, MT; Mudway,  I; Helleday, R; Kelly, FJ; Frew, AJ;  Sandstrom, T.
        (2002).  Effect of ozone on bronchial mucosal inflammation in asthmatic and healthy subjects. Respir
        Med  96: 352-358.
Stenfors, N; Bosson, J;  Helleday, R; Behndig, AF; Pourazar, J; Torngvist, H;  Kelly, FJ; Frew, AJ; Sandstrom, T;
        Mudwav.  IS: Blomber. A. (2010). Ozone exposure enhances mast-cell inflammation in asthmatic
        airways despite inhaled corticosteroid therapy. Inhal Toxicol 22: 133-139.
        http://dx.doi.org/10.3109/08958370903005736.
Stieb. DM: Szvszkowicz. M: Rowe. BH: Leech. JA. (2009). Air pollution and emergency department visits for
        cardiac and respiratory conditions: A multi-city time-series analysis. Environ Health Global Access Sci
        Source 8: 25. http://dx.doi.org/10.1186/1476-069X-8-25.
Draft - Do Not Cite or Quote                        6-258                                  September 2011

-------
Strickland. MJ: Darrow. LA: Klein. M: Flanders. WD: Sarnat. JA: Waller. LA: Sarnat. SE:  Mulholland. JA: Tolbert.
        PE. (2010). Short-term associations between ambient air pollutants and pediatric asthma emergency
        department visits. Am J Respir Grit Care Med 182: 307-316. http://dx.doi.org/10.1164/rccm.200908-
        1201OC.
Strickland, MJ: Darrow, LA: Mulholland, JA: Klein, M: Flanders, WD: Winquist, A: Tolbert, PE. (2011).
        Implications of different approaches for characterizing ambient air pollutant concentrations within the
        urban airshed for time-series studies and health benefits analyses. Environ Health Global Access Sci
        Source 10: 36. http://dx.doi.org/10.1186/1476-069X-10-36.
Stvlianou. M: Nicolich. MJ. (2009). Cumulative effects and threshold levels in air pollution mortality: Data
        analysis of nine large US cities using the NMMAPS dataset. Environ Pollut 157: 2216-2223.
        http://dx.doi.0rg/10.1016/i.envpol.2009.04.011.
Svmons. JM: Wang. L: Guallar. E: Howell. E: Dominici. F: Schwab. M: Ange. BA: Samet. J: Ondov. J: Harrison.
        D: Gevh. A. (2006). A case-crossover study of fine particulate matter air pollution and onset of
        congestive heart failure symptom exacerbation leading to  hospitalization. Am J Epidemiol 164: 421-
        433.
Szvszkowicz. M. (2008). Ambient air pollution and daily emergency department visits for ischemic stroke in
        Edmonton, Canada.  Int J Occup Med Environ Health 21: 295-300.  http://dx.doi.org/10.2478/v10001-
        008-0029-5.
Takeuchi. C: Galve. R: Nieva. J: Witter. DP: Wentworth. AD: Troseth. RP: Lerner. RA: Wentworth P. J. r. (2006).
        Proatherogenic effects of the cholesterol ozonolysis products, atheronal-Aand atheronal-B.
        Biochemistry 45: 7162-7170. http://dx.doi.orq/10.1021/bi0604330.
Tamer. L: Calikoglu. M: Ates.  NA: Yildirim. H: Ercan. B: Saritas. E: Unlu. A: Atik. U.  (2004). Glutathione-S-
        transferase gene polymorphisms (GSTT1, GSTM1, GSTP1) as increased risk factors for asthma.
        Respirology 9: 493-498.
Tankerslev. CG: Peng. RD: Bedga. D: Gabrielson. K: Champion. HC. (2010). Variation in echocardiographic and
        cardiac hemodynamic effects  of PM and ozone inhalation  exposure in strains related to Nppa and Npr1
        gene knock-out mice. Inhal Toxicol 22: 695-707. http://dx.doi.orq/10.3109/08958378.2010.487549.
Tepper. JL: Weiss. B:  Cox. C. (1982). Microanalysis of ozone depression of motor activity. Toxicol Appl
        Pharmacol 64: 317-326.
Tepper. JL: Weiss. B:  Wood. RW. (1983). Behavioral indices of ozone exposure. In.
Tepper. JS: Weiss. B: Wood.  RW. (1985). Alterations in behavior produced by inhaled ozone or ammonia.
        Toxicol Sci 5: 1110-1118.
Thaller. El: Petronella. SA: Hochman. D:  Howard. S: Chhikara. RS: Brooks. EG. (2008).  Moderate increases in
        ambient PM2.5 and ozone are associated with lung function decreases in beach lifeguards. J Occup
        Environ Med 50: 202-211.  http://dx.doi.org/10.1097/JOM.Ob013e31816386b4.
Thompson. AM: Zanobetti. A: Silverman.  F: Schwartz. J:  Coull. B: Urch. B: Speck.  M: Brook. JR: Manno. M:
        Gold. DR. (2010). Baseline Repeated Measures from Controlled Human Exposure Studies:
        Associations between Ambient Air Pollution Exposure and the Systemic Inflammatory Biomarkers IL-6
        and Fibrinogen. Environ Health Perspect 118: 120-124.  http://dx.doi.org/10.1289/ehp.0900550.
Thomson. E: Kumarathasan.  P: Goegan. P: Aubin.  RA: Vincent.  R. (2005). Differential regulation of the lung
        endothelin system by urban particulate matter and ozone.  Toxicol Sci 88: 103-113.
Thomson. E: Kumarathasan.  P: Vincent.  R. (2006). Pulmonary expression of preproET-1 and preproET-3
        mRNAs is altered reciprocally in rats after inhalation of air  pollutants.  Exp  Biol Med 231: 979-984.
Thomson, EM: Kumarathasan, P: Calderon-Garciduenas, L: Vincent, R.  (2007). Air pollution alters brain and
        pituitary endothelin-1 and inducible nitric oxide synthase gene expression. Environ Res 105: 224-233.
        http://dx.doi.0rg/10.1016/i.envres.2007.06.005.
Thurston, GD: Lippmann, M: Scott, MB: Fine, JM. (1997). Summertime haze air pollution and children with
        asthma. Am J Respir Grit Care Med 155: 654-660.
Tolbert. PE: Klein.  M:  Peel. JL: Sarnat. SE: Sarnat. JA. (2007). Multipollutant modeling issues in a study of
        ambient air quality and emergency department visits in Atlanta. J Expo Sci Environ Epidemiol 17: S29-
        S35. http://dx.doi.org/10.1038/si.ies.7500625.
Torres. A: Utell.  MJ: Morow. PE: Voter. KZ: Whitin. JC: Cox. C: Loonev. RJ:  Speers. DM: Tsai. Y: Frampton. MW.
        (1997). Airway inflammation in smokers and nonsmokers with varying responsiveness to ozone. Am J
        Respir Grit Care Med 156: 728-736.
Draft - Do Not Cite or Quote                       6-259                                  September 2011

-------
Trenga. CA: Koenig. JQ: Williams. PV. (2001). Dietary antioxidants and ozone-induced bronchial
        hyperresponsiveness in adults with asthma. Arch Environ Occup Health 56: 242-249.
Triche. EW: Gent. JF: Holford. TR: Belanger. K: Bracken. MB: Beckett. WS: Naeher. L: McSharrv. J. -E:
        Leaderer. BP. (2006). Low-level ozone exposure and respiratory symptoms in infants. Environ Health
        Perspect 114: 911-916. http://dx.doi.org/10.1289/ehp.8559.
Turner. RM: Muscatello. DJ: Zheng.  W: Willmore. A: Arendts. G. (2007). An outbreak of cardiovascular
        syndromes requiring urgent medical treatment and its association with environmental factors: an
        ecological study. Environ Health 6: 37. http://dx.doi.org/10.1186/1476-069X-6-37.
Tyler. WS: Tyler. NK: Last. JA: Gillespie. MJ:  Barstow. TJ. (1988). Comparison of daily and seasonal exposures
        of young monkeys to ozone. Toxicology 50:  131-144.
U.S. EPA. (U.S. Environmental Protection Agency). (1986). Air quality criteria for ozone and other
        photochemical oxidants. (EPA-600/8-84-020aF - EPA-600/8-84-020eF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1996a). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/P-93/004AF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2009d). Integrated science assessment for particulate
        matter. (EPA/600/R-08/139F). Research Triangle Park,  NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
Uchivama. I: Simomura. Y: Yokovama. E. (1986). Effects of acute exposure to ozone on  heart rate and blood
        pressure of the conscious rat. Environ Res 41: 529-537.
Uchivama. I: Yokovama. E. (1989). Effects of short- and long-term exposure to ozone on heart rate and blood
        pressure of emphysematous rats. Environ Res 48: 76-86.
Ulmer,  C: Kopp, M: Ihorst, G: Frischer, T:  Forster, J: Kuehr, J. (1997). Effects of ambient  ozone exposures
        during the spring and summer of 1994 on pulmonary function of schoolchildren.  Pediatr Pulmonol 23:
        344-353.  http://dx.doi.org/10.1002/(SICI)1099-0496(199705)23:5<344::AID-PPUL6>3.0.CO:2-K.
Ultman, JS: Ben-Jebria, A: Arnold, SF. (2004). Uptake distribution of ozone in human lungs: Intersubject
        variability in physiologic response. (HEI Research Report 125). Boston, MA: Health Effects Institute.
        http://pubs.healtheffects.org/view.php?id=70.
Urch, B: Silverman, F: Corey, P: Brook, JR: Lukic,  KZ: Rajagopalan,  S: Brook, RD. (2005). Acute blood pressure
        responses in healthy adults during controlled air pollution exposures. Environ Health Perspect 113:
        1052-1055.
Vagaggini, B: Carnevali, S: Macchioni, P:  Taccola, M: Fornai, E: Bacci, E: Bartoli,  ML: Cianchetti, S: Dente, Fl_:
        Pi Franco. A: Giannini. D: PaggiaroPL. (1999). Airway inflammatory response to ozone in subjects with
        different asthma severity. Eur Respir J 13: 274-280.
Vagaggini, B: Taccola, M: Conti, I: Carnevali, S: Cianchetti, S: Bartoli, ML: Bacci, E: Dente, Fl_: Pi Franco, A:
        Giannini.  D: Paggiaro. PL. (2001). Budesonide reduces neutrophilic but not functional airway response
        to ozone in mild asthmatics. Am J Respir Grit Care Med 164: 2172-2176.
Vagaggini, B: Taccola, M: Cianchetti, S: Carnevali, S: Bartoli, ML: Bacci, E: Dente, Fl_: Pi Franco, A: Giannini, D:
        Paggiaro. PL. (2002). Ozone exposure increases eosinophilic airway response induced by previous
        allergen challenge. Am J Respir Grit Care Med 166: 1073-1077.
Vagaggini, B: Cianchetti, S: Bartoli,  M: Ricci,  M: Bacci, E: Dente, FL: Pi Franco, A: Paggiaro, P. (2007).
        Prednisone blunts airway neutrophilic inflammatory response due to ozone  exposure in asthmatic
        subjects.  Respiration 74: 61-58. http://dx.doi.org/10.1159/000096078.
Vagaggini, B: Bartoli, MLE: Cianchetti, S:  Costa, F: Bacci, E: Dente,  FL: Pi Franco, A:  Malagrino, L: Paggiaro, P.
        (2010). Increase in markers of airway inflammation after ozone exposure can be observed also in
        stable treated asthmatics with minimal functional response to ozone. Respir Res 11:5.
        http://dx.doi.org/10.1186/1465-9921-11-5.
Valacchi. G: Pecorelli. A: Mencarelli. M: Maioli. E: Davis. PA. (2009).  Beta-carotene prevents ozone-induced
        proinflammatory markers in murine skin. Toxicol Ind Health 25: 241-247.
        http://dx.doi. org/10.1177/0748233709103030.
Draft - Do Not Cite or Quote                        6-260                                 September 2011

-------
Van Loveren. H: Krainc. El: Rombout. PJ: Blommaert. FA: Vos. JG. (1990). Effects of ozone,
        hexachlorobenzene, and bis(tri-n-butyltin)oxide on natural killer activity in the rat lung. Toxicol Appl
        Pharmacol 102: 21-33.
Van Loveren. H: Rombout. PJA: Wagenaar. SS: Walvoort. HC: Vos. JG. (1988). Effects of ozone on the defense
        to a respiratory Listeria monocytogenes infection in the rat: Suppression of macrophage function and
        cellular immunity  and aggravation of histopathology in lung and liver during infection. Toxicol Appl
        Pharmacol 94: 374-393.
Vancza, EM: Galdanes, K: Gunnison, A: Hatch, G: Gordon, T.  (2009). Age, strain, and gender as factors for
        increased sensitivity of the mouse lung to inhaled ozone. Toxicol Sci 107: 535-543.
        http://dx.doi. orq/10.1093/toxsci/kfn253.
Vanguilder, HP: Freeman, WM. (2011). The hippocampal  neuroproteome with aging and cognitive decline: Past
        progress and future directions. Front Aging Neurosci 3: 8. http://dx.doi.ora/10.3389/fnaai.2011.00008.
Veselv. PL: Giordano. AT:  Raska-Emerv. P: Montgomery.  MR. (1994a). Increase in atrial natriuretic factor in the
        lungs, heart, and  circulatory system owing to ozone. Chest 105:  1551-1554.
Veselv. PL: Giordano. AT:  Raska-Emerv. P: Montgomery.  MR. (1994b). Ozone increases amino- and  carboxy-
        terminal atrial  natriuretic factor prohormone peptides in lung, heart, and circulation. J Biochem  Mol
        Toxicol 9: 107-112.
Veselv. PL: Giordano. AT:  Raska-Emerv. P: Montgomery.  MR. (1994c). Ozone increases atrial natriuretic
        peptides in heart, lung and circulation of aged vs adult animals. Gerontology  40: 227-236.
        http://dx.doi. org/10.1159/000213590.
Villeneuve. PJ: Chen. L: Stieb. D: Rowe. BH. (2006a). Associations between outdoor  air pollution and
        emergency department visits for stroke in Edmonton, Canada. Eur J Epidemiol 21: 689-700.
Villeneuve, PJ: Doiron, M, -S: Stieb, D: Dales, R: Burnett, RT: Dugandzic, R. (2006b). Is outdoor air pollution
        associated with physician visits for allergic rhinitis among the elderly in Toronto, Canada? Allergy 61:
        750-758. http://dx.doi.0rg/10.1111/i.1398-9995.2006.01070.x.
Villeneuve, PJ: Chen, L: Rowe, BH: Coates, F. (2007). Outdoor air pollution and emergency department visits
        for asthma among children and adults: A case-crossover study in northern Alberta, Canada. Environ
        Health Global  Access Sci Source 6: 40. http://dx.doi.org/10.1186/1476-069X-6-40.
Vincent, R: Janzen, EG: Chen, G: Kumarathasan,  P: Haire, PL: Guenette, J: Chen, JZ: Bray, TM. (1996b). Spin
        trapping study in the lungs and liver of F344 rats  after exposure to ozone. Free Radic Res 25: 475-488.
Von Klot. S: Peters. A:  Aalto.  P: Bellander. T: Berglind. N:  D'lppoliti. D: Elosua. R: Hormann. A: Kulmala. M:
        Lanki, T: Lowel, H: Pekkanen, J: Picciotto, S: Sunyer, J: Forastiere, F: Group, HEoPoSSS. (2005).
        Ambient air pollution is associated with increased risk of hospital cardiac readmissions of myocardial
        infarction survivors in five European cities. Circulation 112: 3073-3079.
        http://dx.doi.org/10.1161/CIRCULATIONAHA.105.548743.
Vovnow. JA: Fischer. BM:  Zheng.  S: Potts. EN: Grover. AR: Jaiswal. AK: Ghio. AJ: Foster.  WM. (2009).
        NAD(P)H quinone oxidoreductase 1 is essential for ozone-induced oxidative  stress in mice and
        humans. Am J Respir Cell Mol Biol 41: 107-113. http://dx.doi.org/10.1165/rcmb.2008-0381OC.
Wagner. JG: Jiang. Q:  Harkema. JR: Illek. B: Patel. DP: Ames. BN: Peden. DB. (2007). Ozone enhancement of
        lower airway allergic inflammation is prevented by gamma-tocopherol. Free Radic Biol Med 43: 1176-
        1188. http://dx.doi.0rg/10.1016/i.freeradbiomed.2007.07.013.
Wagner. JG: Harkema. JR: Jiang. Q: Illek. B: Ames. BN: Peden. DB. (2009). Gamma-tocopherol attenuates
        ozone-induced exacerbation of allergic rhinosinusitis in rats. Toxicol Pathol 37: 481-491.
        http://dx.doi. org/10.1177/0192623309335630.
Wang. DJ: Zhou. WD: Dai. XJ: Yan. Y. (2007). Study on effect and mechanism of sodium ferulate in preventing
        and treating ozone induced lung injury in mice. Chin J Integr Med 13: 211-214.
        http://dx.doi.org/10.1007/s11655-007-0211-9.
Ward. DJ: Roberts. KT: Jones. N: Harrison. RM: Avres. JG: Hussain. S: Walters. S. (2002). Effects of daily
        variation in outdoor particulates and ambient acid species in normal and asthmatic children. Thorax 57:
        489-502. http://dx.doi.0rg/10.1136/thorax.57.6.489.
Watkinson. WP: Aileru. AA: Dowd. SM: Doerfler. PL: Tepper. JS: Costa. PL. (1993). Acute  effects of ozone on
        heart rate and body temperature in the unanesthetized, unrestrained  rat maintained at different ambient
        temperatures. Inhal Toxicol 5: 129-147.
Draft - Do Not Cite or Quote                        6-261                                  September 2011

-------
Watkinson. WP: Campen. MJ: Wichers. LB: Nolan. JP: Costa. PL. (2003). Cardiac and thermoregulatory
        responses to inhaled pollutants in healthy and compromised rodents: Modulation via interaction with
        environmental factors. Environ Res 92: 35-47.
Weinmann. GG: Weidenbach-Gerbase. M: Foster. WM: Zacur. H: Frank. R. (1995a). Evidence for ozone-
        induced small-airway dysfunction: Lack of menstrual-cycle and gender effects. Am J Respir Crit Care
        Med 152: 988-996.
Weinmann. GG: Bowes. SM: Gerbase. MW: Kimball. AW: Frank. R.  (1995c).  Response to acute ozone
        exposure in healthy men. Results of a screening procedure. Am J Respir Crit  Care Med 151: 33-40.
Wellenius. GA: Bateson. TF: Mittleman. MA: Schwartz. J. (2005). Particulate  air pollution and the rate of
        hospitalization for congestive heart failure among medicare beneficiaries in Pittsburgh, Pennsylvania.
        Am J Epidemiol 161: 1030-1036.
Wellenius. GA: Yeh. GY: Coull. BA: Suh.  HH:  Phillips. RS: Mittleman. MA. (2007). Effects of ambient air
        pollution on functional status in patients with chronic congestive heart failure:  A repeated-measures
        study.  Environ Health 6: 1-7.
Wentworth. P. Jr: Nieva. J: Takeuchi. C: Galve. R: Wentworth. AD: Dillev. RB: DeLaria. GA: Saven. A: Babior.
        BM: Janda. KD: Eschenmoser. A: Lerner. RA. (2003). Evidence for ozone formation in human
        atherosclerotic arteries. Science 302: 1053-1056.
Wheeler. A: Zanobetti. A: Gold. PR: Schwartz. J:  Stone. P: Suh.  HH. (2006). The relationship between ambient
        air pollution and heart rate variability differs for individuals with heart and pulmonary disease. Environ
        Health Perspect 114: 560-566.
Wiester. MJ: Watkinson. WP: Costa. PL:  Crissman. KM: Richards. JH: Winsett. DW: Highfill. JW. (1996b).
        Ozone toxicity in the rat. III. Effect of changes in ambient temperature on pulmonary parameters. J Appl
        Physiol 81: 1691-1700.
Williams. AS: Leung. SY: Nath. P: Khorasani. NM: Bhavsar.  P: Issa. R: Mitchell. JA: Adcock. IM:  Chung. KF.
        (2007b). Role of TLR2,  TLR4, and  MyD88 in murine ozone-induced airway hyperresponsiveness and
        neutrophilia. J Appl Physiol 103: 1189-1195. http://dx.doi.org/10.1152/iapplphysiol.00172.2007.
Wiwatanadate. P:  Trakultivakorn. M. (2010). Air pollution-related peak expiratory flow rates among asthmatic
        children in Chiang Mai,  Thailand. Inhal Toxicol 22: 301-308.
        http://dx.doi.org/10.3109/08958370903300327.
Wiwatanadate. P:  Liwsrisakun. C. (2011). Acute effects of air pollution on peak expiratory flow rates and
        symptoms among asthmatic patients in Chiang Mai, Thailand. Int J Hyg Environ Health 214: 251-257.
        http://dx.doi.0rg/10.1016/i.iiheh.2011.03.003.
Wong. C. -M: Ma.  S: AJ. H: Lam. T -H. (1999a). Does ozone have any effect on daily hospital admissions for
        circulatory diseases? J  Epidemiol Community Health 53: 580-581.
Wong, CM: Yang,  L: Thach, TQ: Chau, PY:  Chan, KP: Thomas, GN: Lam, TH: Wong, TW: Hedley, AJ: Peiris,
        JS. (2009). Modification by influenza on  health effects of air pollution in Hong  Kong. Environ Health
        Perspect  117: 248-253. http://dx.doi.org/10.1289/ehp.11605.
Wong, CM: Vichit-Vadakan, N: Vajanapoom, N: Ostro,  B: Thach, TQ: Chau, PY: Chan, EK: Chung, RY: Ou, CQ:
        Yang.  L: Peiris. JS: Thomas. GN: Lam. TH: Wong. TW: Hedlev. AJ: Kan. H: Chen. B: Zhao. N:  London.
        SJ: Song. G: Chen. G: Zhang. Y: Jiang.  L: Qian. Z:  He. Q: Lin. HM:  Kong. L: Zhou. D: Liang. S: Zhu. Z:
        Liao, D: Liu, W: Bentlev, CM: Dan,  J: Wang, B: Yang, N: Xu, S: Gong, J: Wei,  H: Sun, H: Qin, Z. (2010).
        Part 5. Public health and air pollution in Asia (PAPA): A combined analysis of four studies of air pollution
        and mortality. In Public  Health and Air Pollution in Asia (PAPA): Coordinated Studies of Short-Term
        Exposure to Air Pollution and Daily Mortality in Four Cities (Vol. 154). Boston,  MA: Health Effects
        Institute. http://pubs.healtheffects.org/view.php?id=348.
Wong. TW: Lau. TS: Yu. TS: Neller. A: Wong. SL: Tarn. W: Pang. SW. (1999b). Air pollution and hospital
        admissions for respiratory and cardiovascular diseases  in Hong Kong. Occup Environ Med 56: 679-
        683.
Wu. CF: Kuo. 1C: Su. TC: Li. YR: Lin. LY: Chan. CC: Hsu. SC. (2010). Effects of personal exposure to
        particulate matter and ozone on arterial stiffness and heart  rate variability  in healthy adults. Am J
        Epidemiol 171: 1299-1309. http://dx.doi.org/10.1093/aie/kwg060.
Xia. Y: Tong. H. (2006). Cumulative effects of air  pollution on public  health. Stat Med 25: 3548-3559.
        http://dx.doi. org/10.1002/sim.2446.
Draft - Do Not Cite or Quote                       6-262                                 September 2011

-------
Yallop. D: Duncan. ER: Morris. E: Fuller. GW: Thomas. N: Walters. J: Dick.  MC: Height. SE: Thein. SL: Rees.
        DC. (2007). The associations between air quality and the number of hospital admissions for acute pain
        and sickle-cell disease in an urban environment. Br J Haematol 136: 844-848.
        http://dx.doi.0rg/10.1111/i.1365-2141.2007.06493.x.
Yang, C, -Y: Chen, Y, -S: Yang, C, -H: Ho, S, -C. (2004). Relationship between ambient air pollution and hospital
        admissions for cardiovascular diseases in Kaohsiung, Taiwan. J Toxicol Environ Health A 67: 483-493.
Yang. CY. (2008). Air pollution and hospital admissions for congestive heart failure in a subtropical city: Taipei,
        Taiwan. J Toxicol Environ Health A 71: 1085-1090.
Yang. Q: Chen. Y: Krewski. D: Burnett. RT: Shi. Y: McGrail. KM. (2005b). Effect of short-term exposure to low
        levels of gaseous pollutants on chronic obstructive pulmonary disease hospitalizations. Environ Res
        99: 99-105. http://dx.doi.Org/10.1016/i.envres.2004.09.014.
Yokovama.  E: Uchivama. I: Arito. H. (1989). Extrapulmonary effects of low level ozone exposure.  In T
        Schneider; SD Lee; GJR Wolters; LD Grant (Eds.), Atmospheric ozone research and its policy
        implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The
        Netherlands (Vol. 35, pp. 301-309). Nijmegen, The Netherlands: Elsevier.
Yoon. HK: Cho. HY: Kleeberger. SR. (2007). Protective role of matrix metalloproteinase-9 in ozone-induced
        airway inflammation. Environ Health Perspect 115: 1557-1563. http://dx.doi.org/10.1289/ehp.10289.
Zanobetti. A: Canner. MJ: Stone. PH: Schwartz. J: Sher. D: Eagan-Bengston. E: Gates. KA: Hartley. LH: Suh. H:
        Gold. DR. (2004). Ambient pollution and  blood pressure in cardiac rehabilitation patients. Circulation
        110: 2184-2189. http://dx.doi.org/10.1161/01.cir.0000143831.33243.d8.
Zanobetti. A: Schwartz. J. (2006). Air pollution and emergency admissions in Boston, MA. J Epidemiol
        Community Health 60: 890-895.
Zanobetti, A; Schwartz, J. (2008a). Is there adaptation in the ozone mortality relationship: A multi-city case-
        crossover analysis. Environ Health 7: 22. http://dx.doi.org/10.1186/1476-069X-7-22.
Zanobetti. A: Schwartz. J. (2008b). Mortality displacement in the association of ozone with  mortality: An analysis
        of 48 cities in the United States. Am J Respir Grit Care Med 177: 184-189.
        http://dx.doi.org/10.1164/rccm.200706-823OC.
Zanobetti, A; Gold, PR; Stone, PH; Suh, HH; Schwartz, J;  Coull, BA; Speizer, FE. (2010). Reduction in heart
      rate variability with traffic and air pollution in patients with coronary artery disease. Environ Health
      Perspect 118:  324-330.
Draft - Do Not Cite or Quote                        6-263                                  September 2011

-------
           INTEGRATED  HEALTH  EFFECTS  OF  LONG-
           TERM  OZONE  EXPOSURE
          7.1    Introduction

 1                   This chapter reviews, summarizes, and integrates the evidence on relationships between
 2                   health effects and long-term exposures to O3. Both epidemiologic and toxicological
 3                   studies provide a basis for examining long-term O3 exposure health effects for respiratory
 4                   effects, cardiovascular effects, reproductive and developmental effects, central nervous
 5                   system effects, cancer outcomes, and mortality. Long-term exposure has been defined as
 6                   a duration of approximately 30 days (1 month) or longer.

 7                   Conclusions from the 2006 O3 AQCD are summarized briefly at the beginning of each
 8                   section, and the evaluation of evidence from recent studies builds upon what was
 9                   available during the previous review. For each health outcome (e.g., respiratory disease,
10                   lung function), results are summarized for studies from the specific scientific discipline,
11                   i.e., epidemiologic and toxicological studies. The major sections  (i.e. respiratory,
12                   cardiovascular, mortality, reproductive/developmental, cancer) conclude with summaries
13                   of the evidence for the various health outcomes within that category and integration of
14                   the findings that lead to conclusions regarding causality based upon the framework
15                   described in Chapter 1. Determination of causality is made for the overall health effect
16                   category, such as respiratory effects, with coherence and plausibility being based on
17                   evidence from across disciplines and also across the suite of related health outcomes,
18                   including cause-specific mortality.
          7.2    Respiratory Effects

19                   Studies reviewed in the 2006 O3 AQCD examined evidence for relationships between
20                   long-term O3 exposure (several months to yearly) and effects on respiratory health
21                   outcomes including declines in lung function, increases in inflammation, and
22                   development of asthma in children and adults. Animal toxicology data provided a clearer
23                   picture indicating that long-term O3 exposure may have lasting effects. Chronic exposure
24                   studies in animals have reported biochemical and morphological changes suggestive of
25                   irreversible long-term O3 impacts on the lung. In contrast to supportive evidence from
26                   chronic animal studies, the epidemiologic studies on longer-term (annual) lung function
27                   declines, inflammation, and new asthma development remained inconclusive.
      Draft - Do Not Cite or Quote                       7-1                                 September 2011

-------
 1                   Several studies (Horak et al., 2002; Frischer et al., 1999) collectively indicated that O3
 2                   exposure over several summer months was associated with smaller increases in lung
 3                   function growth in children. For longer time periods (annual), the definitive analysis in
 4                   the Child Health Study (CHS) reported by Gauderman et al. (2004) provided little
 5                   evidence that such long-term exposure to ambient O3 was associated with significant
 6                   deficits in the growth rate of lung function in children in contrast to the effects observed
 7                   with other pollutants such as acid vapor, NO2, and PM2 5. Asthmatic children with
 8                   GSTM1 null genotype were found to be more susceptible to the impact of O3 exposure
 9                   (over a 12 week study period) on small airways function in Mexico (Romieu et al..
10                   2004a). Limited epidemiologic research examined the relationship between long-term O3
11                   exposures and inflammation. Evidence of inflammation and allergic responses consistent
12                   with known effects of O3 exposure (30 day mean) such as increased eosinophil levels
13                   were observed in an Austrian study (Frischer et al., 2001). The cross-sectional surveys
14                   available for the 2006 O3 AQCD detected no associations between long-term O3
15                   exposures and asthma prevalence, asthma-related symptoms or allergy to common
16                   aeroallergens in children after controlling for covariates.

17                   New evidence presented below reports consistent associations between long-term O3
18                   exposure and new-onset asthma related to genotype in U.S. cohorts in multi-community
19                   studies. Related studies report coherent relationships between respiratory symptoms
20                   among asthmatics and long-term O3 exposure. Short-term exposure to O3 is associated
21                   with increases in respiratory symptoms and asthma medication use, as summarized in
22                   Section 6.2.4.2. A new line of evidence reports a positive exposure response relationship
23                   between first asthma hospitalization and long-term O3 exposure. Results from recent
24                   studies examining pulmonary function, inflammation, and allergic responses are also
25                   presented.
             7.2.1    New Onset Asthma

26                   Risk for new-onset asthma is related in part to genetic susceptibility, behavioral factors
27                   and environmental exposure (Gilliland et al..  1999). Complex chronic diseases, such as
28                   asthma, are partially the result of a sequence of biochemical reactions involving
29                   exposures to various environmental agents metabolized by a number of different genes
30                   (Conti et al., 2003). Understanding the relation between genetic polymorphisms and
31                   environmental exposure can help identify high-risk subgroups in the population and
32                   provide better insight into pathway mechanisms for these complex diseases. Oxidative
33                   stress likely underlies these mechanistic hypotheses (Gilliland et al.. 1999). Susceptibility
34                   genes act through modification of disease risk associated with environmental factors.
35                   Epidemiologic investigation of hypotheses of possible mechanisms involving the gene-

      Draft - Do Not Cite or Quote                       7-2                                 September 2011

-------
 1                   environmental (GxE) interaction involves statistical analysis of these interactions for the
 2                   risk of new-onset asthma in children being influenced by exposure to air pollution
 3                   (Gauderman. 2002. 2001; Gilliland et al.. 1999).

 4                   Evidence for the potential importance of genetic susceptibility and behavioral factors on
 5                   new onset asthma are provided by several recent studies (Himes et al.. 2009; Islam et al..
 6                   2008; Li et al.. 2008; Hanene et al.. 2007; Ercan et al.. 2006; Li et al.. 2006a: Tamer et
 7                   al.. 2004; Gilliland et al.. 2002). Evidence for a gene-pollution interaction in the
 8                   pathogenesis of asthma are supported by recent study findings (Islam et al.. 2009; Islam
 9                   et al.. 2008: Orvszczvn et al.. 2007: Lee et al.. 2004b: Gilliland et al.. 2002).

10                   Evidence for associations between long-term exposure to O3 and new-onset asthma is
11                   provided by new studies from the CHS. Initiated in the early 1990's, the CHS was
12                   originally designed to examine whether long-term exposure to ambient pollutants was
13                   related to chronic respiratory outcomes in children in 12 communities in southern
14                   California (Peters et al.. 1999a: Peters et al.. 1999b). About 10 years later, the CHS
15                   inaugurated a series of genetic studies (Gilliland et al.. 1999) nested within the CHS
16                   cohort by obtaining biological samples from the study subjects (buccal  cells). These new
17                   studies examined the relationship between health outcomes, genetic susceptibility,
18                   behavioral factors and environmental exposure.

19                   First, the hypothesis that the functional polymorphisms of HMOX-1 [(GT)n repeat], CAT
20                   (-262C > T -844C > TO, and MNSOD (Ala-9Val) are associated with new-onset asthma
21                   was evaluated, and then whether the effects of these variants varied by exposure to O3
22                   (Islam et al.. 2008). HMOX1 [heme oxygenase (decycling) 1] is a human gene that
23                   encodes  for the enzyme heme oxygenase. Heme oxygenase 1 (HO-1) is an enzyme that
24                   catalyzes the metabolism of heme. The heme iron serves as a source or sink of electrons
25                   during electron transfer or redox chemistry, so the presence of the HMOX1 gene, and
26                   therefore the generation of heme oxygenase, protects against oxidative stress in the body.
27                   The authors observed that functional promoter variants in CAT and HMOX-1 showed
28                   ethnicity-specific associations with new-onset asthma and that oxidant gene protection
29                   was restricted to children living in low-O3 communities.

30                   The subjects were drawn from the CHS cohort. Children with a history of asthma or
31                   wheeze were excluded from this analysis. Analyses were restricted to children of
32                   Hispanic (n = 576) or non-Hispanic white ethnicity (n = 1,125). New-onset asthma was
33                   classified as having no prior history of asthma at study entry with subsequent report of
34                   physician-diagnosed asthma at follow-up with the date of onset assigned to be the
35                   midpoint of the interval between the interview date when asthma diagnosis was first
36                   reported and the previous interview date. As a sensitivity analysis, the asthma definition
37                   was restricted to those new-onset asthma cases who also used an inhaler (n = 121). They
      Draft - Do Not Cite or Quote                       7-3                                 September 2011

-------
 1                   calculated long-term mean pollutant levels (1994 - 2003) to assign exposure to children
 2                   in each community for use in the statistical analysis. The effect of ambient air pollution
 3                   on the relationship between genetic polymorphism and new-onset asthma was assessed
 4                   using models where the community specific average air pollution levels were fitted as a
 5                   continuous variable together with the appropriate interaction terms for genes and air
 6                   pollutants (Berhane et al.. 2004). Cox proportional hazard regression models were fitted
 7                   to the data. A stratified analysis for the two independent fourth-grade cohorts of the study
 8                   population recruited in 1993 and 1996 was conducted to assess whether the results could
 9                   be replicated in independent groups of children.

10                    Over the follow-up period, 160 new cases of asthma were diagnosed (Islam et al., 2008).
11                   The evidence indicated that the effect of variation in the HMOX-1 gene on risk of new-
12                   onset asthma differed by ambient O3 level. An interaction P value was reported of 0.003
13                   from the hierarchical two stage Cox proportional hazard model fitting the community-
14                   specific O3 and PM10 levels (continuous) and controlling for random effect of the
15                   communities. Average O3 levels showed low correlation with the other monitored
16                   pollutants. The interaction indicated a greater effect (association) of community O3 level
17                   among children with the gene than with children without the gene. Alleles with 23 or
18                   fewer (GT)n repeats are categorized as short (S). The S-allele variant of this protective
19                   enzyme is more readily induced than those with more numerous repeats. The largest
20                   protective effect of the (GT)n repeat polymorphism of HMOX-1 was observed for
21                   children who were S-allele carriers and resided in low-O3 communities with Hazard
22                   Ratio (HR) of 0.44 (95% CI:  0.23, 0.83). The ratio of HR of S-allele carriers who resided
23                   in high O3 communities (HR=0.88; [95% CI: 0.33, 2.34]) was twofold greater than in
24                   those who resided in the low-O3 communities (HR=0.44). The non-parallelism of the two
25                   lines in An interaction p-value of 0.003 was obtained from the hierarchical two stage Cox
26                   proportional hazard model fitting the community specific O3 and controlling for random
27                   effect of the communities. The interaction indicates there is  a greater effect (association)
28                   of community O3 level on children with the gene than with children without the gene.
29                   The HRs are off-set as opposed to overlapping in the figure to allow clearer presentation
30                   of the results.

31                   Figure 7-1 illustrates the interaction: Children with the S-allele have protection against
32                   the onset of asthma; however, in high- O3 communities, this protection is attenuated. The
33                   results from sensitivity analyses on the two fourth-grade cohorts, and the inhaler
34                   definition for asthma were both consistent with the main results. An analysis related to
35                   children's participation in sports or time spent outdoors produced the same outcome. No
36                   significant interactions were observed between PM10 or other pollutants and the HMOX -
37                   1 gene; quantitative results were not presented. A potential concern for not adjusting for
3 8                   multiple testing was considered by the authors as not a factor in this analysis because the
      Draft - Do Not Cite or Quote                        7-4                                 September 2011

-------
 1
 2
 3
selection of the genes was based on a priori hypotheses defined by a well-studied
biological pathway. Thus in this cohort in southern California, Islam et al. (2008) related
new-onset asthma to O3 exposure in genetically susceptible children.
                                   Interaction of Gene presence and Ozone Level on the
                                   Hazard Ratio of New Onset Asthma (P-value of 0.003)
                              2.5 -
                               2 -
                              1.5 -
                               1  -
                              0.5 -
                               0 J
                                                                         (2.43)
                                                 Children with no S-Allele
                                                       0.94
                    •rfO.83)                	 	 „ — -•  **

                 .-. Ill 	 ^^   r^UIMr^n ,.,^U O All^l^
                                                  Children with S-Allele
                                                                              (2.34)
                                                                                 0.88
                   (0.28)
                Low
              (38.4 ppb)
                                                                         (0.36) (0.33)
                                                Community Mean Ozone Level
  High
(55.2 ppb)
                                        (Confidence limits based on comparison with reference group)
        Source: Developed by EPA with data from Islam et al. (2008) (used by permission of American Thoracic Society).
        An interaction p-value of 0.003 was obtained from the hierarchical two stage Cox proportional hazard model fitting the community
      specific O3 and controlling for random effect of the communities. The interaction indicates there is a greater effect (association) of
      community O3 level on children with the gene than with children without the gene. The HRs are off-set as opposed to overlapping in
      the figure to allow clearer presentation of the results.

      Figure  7-1      Interaction of gene presence and Os level  on the Hazard Ratio (HR)
                        of new-onset asthma in the 12 Children's Health Study
                        communities.
 4
 5
 6
 7
 8
 9
10
11
12
13
Related to the findings in Islam et al. (2008) discussed above, Islam et al. (2009)
examined whether GSTP1, GSTM1, exercise and O3 exposure have interrelated effects
on the pathogenesis of asthma. A modifying role of air pollution on the association
between IlelOSVal and asthma in a cohort of children had been observed (Lee et al..
2004b), but the study did not examine O3 specifically or consider exercise. A primary
conclusion that the authors (Islam et al.. 2009) reported was that the common functional
variants of GSTP1 and GSTM1 null genotypes modulate the risk of new onset asthma
during adolescence. Children who had the GSTM1  null genotype were at 1.6-fold (95%
CI: 1.2, 2.2) increased risk of developing new onset asthma compared with those without
the null genotype. Further, the CHS investigators examined the complex interrelationship
      Draft - Do Not Cite or Quote
                                 7-5
                  September 2011

-------
 1                   of antioxidant defenses with asthma risk with increasing doses of O3, resulting from
 2                   increasing ventilation associated with vigorous exercise characterized by the number of
 3                   team sports played. In an earlier analysis, McConnell et al. (2002) had reported that the
 4                   risk of new onset asthma was associated with outdoor exercise, especially in high O3
 5                   communities but did not consider genetic variants. In this new study, Islam et al. (2009)
 6                   find a six fold increased risk of asthma (HR=6.15, [95% CI:  2.2,  7.4]) for children who
 7                   were homozygous  for He 105, participated in three or more team sports and lived in
 8                   high-O3 communities, demonstrating the potential importance of a combination of
 9                   genetic variability, O3 exposure and behavior on asthma risk.

10                   Epidemiologic evidence of associations of arginase variants  with asthma are limited (Li
11                   etal.. 2006a). Asthmatic subjects have higher arginase activity than non-asthmatic
12                   subjects (Morris et al.. 2004). NO is a mediator of nitrosative stress synthesized from L-
13                   arginine by nitric oxide synthases. In the CHS, Salam et al. (2009) examined whether
14                   arginase variants (ARG1 and ARG2 genes) were associated  with asthma and whether
15                   atopy and exposures to smoking and air pollution influence the associations. The
16                   modifying effect of O3 and atopy on the association between haplotypes and asthma were
17                   evaluated using likelihood ratio tests with appropriate interaction terms. They found that
18                   both ARG1 and ARG2 genetic loci were associated with childhood-onset asthma. The
19                   effect of the ARG1 haplotype varied by the child's history of atopy and ambient O3.
20                   Among atopic children living in high O3 communities, those carrying the ARG1
21                   haplotype had reduced asthma risk (Odds Ratio [OR] per ARGlh4 haplotype copy =
22                   0.12; [95% CI: 0.04, 0.43]; P heterogeneity across atopy/O3  categories = 0.008).

23                   Further, the CHS presents  results examining the relationship of new onset asthma with
24                   traffic-related pollution near homes and schools (McConnell et al.. 2010). Asthma risk
25                   increased with modeled traffic-related pollution exposure from roadways near homes and
26                   near schools. The HR was  0.76 (95% CI: 0.38, 1.54) across the range of ambient O3
27                   exposure in the communities. With adjustment for school and residential non-freeway
28                   traffic-related exposure, the estimated HR for O3 was 1.01 (95% CI: 0.49, 2.11). Gene
29                   variants were not evaluated in this study.

30                   Some cross-sectional studies reviewed in the 2006 O3 AQCD observed positive
31                   relationships between chronic exposure to O3 and prevalence of asthma and asthmatic
32                   symptoms in school children (Ramadour et al.. 2000; Wang  et al.. 1999) while others
33                   (Kuo et al.. 2002; Charpin et al.. 1999) did not. Recent studies provide additional
34                   evidence.

35                   In a cross-sectional nationwide study of 32,672 Taiwanese school children, Hwang et al.
36                   (2005) assessed the effects of air pollutants on the risk of asthma. The study population
37                   was recruited from elementary and middle schools within 1 km of air monitoring stations.
      Draft - Do Not Cite or Quote                       7-6                                September 2011

-------
 1                   The risk of asthma was related to O3 in the one-pollutant model. The addition of other
 2                   pollutants (NOX, CO2, SO2, and PMi0), in two-pollutant and three-pollutant models,
 3                   increased the O3 risk estimates. The prevalence of childhood asthma was assessed in
 4                   Portugal by contrasting the risk of asthma between a high O3 rural area and an area with
 5                   low O3 levels (Sousaet al., 2011; Sousa et al., 2009; Sousaet al., 2008). The locations
 6                   were selected to provide a difference in O3 levels without the confounding effects of
 7                   other pollutants. Both evaluation for asthma symptoms and FEVi suggested that O3
 8                   increased asthma prevalence. Clark et al. (2010) investigated the effect of exposure to
 9                   ambient air pollution in utero and during the first year of life on risk of subsequent
10                   incidence asthma diagnosis up to 3-4 years of age in a population-based nested case-
11                   control study for all children born in southwestern British Columbia in 1999 and 2000
12                   (n=37,401; including 3,482 [9.3%] with asthma). Air pollution exposure for each subject
13                   was estimated based on their residential address history using regulatory monitoring data,
14                   land use regression modeling, and proximity to stationary pollutant sources. Daily values
15                   from the three closest monitors within 50 km were used to calculate exposures. Traffic-
16                   related pollutants were associated with the highest risk. Ozone was inversely correlated
17                   with the primary traffic-related pollutants (r = -0.7 to -0.9). The low reliability of asthma
18                   diagnosis in infants makes this study difficult to interpret (Martinez et al.. 1995). In a
19                   cross-sectional analysis, Akinbami et al. (2010) examined the association between
20                   chronic exposure to outdoor pollutants (12-month avg levels by county) and asthma
21                   outcomes in a national sample of children ages 3-17 years living in U.S. metropolitan
22                   areas (National Health Interview Survey, N = 34,073). A 5-ppb increase in estimated 8-h
23                   max O3 concentration (annual average) yielded a positive association for both  currently
24                   having asthma and for having at least 1  asthma attack in the previous year; while the
25                   adjusted odds ratios for other pollutants were not statistically significant. Models in
26                   which pollutant value ranges were divided into quartiles produced comparable results.
27                   Multi-pollutant models (SO2 and PM) produced similar results. The median value for
28                   12-month avg O3 levels was 39.5 ppb and the IQR was 35.9-43.7 ppb.  The adjusted odds
29                   for current asthma for the highest quartile (49.9-59.5 ppb) of estimated O3 exposure was
30                   1.56 (95% CI: 1.15,2.10) with a positive dose-response relationship apparent from the
31                   lowest quartile to the highest. Thus, this cross-sectional analysis and Hwang et al. (2005)
32                   provides further evidence relating O3 exposure and the risk of asthma.

33                   The occurrence of bronchitic symptoms among children with asthma was investigated in
34                   the CHS examining the role of gene-environment interactions and long-term O3
35                   exposure. Lee et al. (2009b) studied associations of TNF-308 genotype with bronchitis
36                   symptoms among asthmatic children and investigated whether associations vary with
37                   ambient O3 exposure since increased airway TNF may be related to inflammation.
38                   Asthmatic children with the GG genotype had a lower prevalence of bronchitic symptoms
39                   compared with children carrying at least one A-allele (e.g., GA or AA). Low-versus high-

      Draft - Do Not Cite or Quote                       7-7                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
O3 strata were defined as less than or greater than 50- ppb O3 avg. Asthmatic children
with TNF-308 GG genotype had a significantly reduced risk of bronchitic symptoms with
low-O3 exposure (OR=0.53; [95% CI: 0.31, 0.91]). The risk was not reduced in children
living in high-O3 communities (OR=1.42; [95% CI: 0.75, 2.70]). The difference in
genotypic effects between low- and high-O3 environments was statistically significant
among asthmatics (P for interaction = 0.01), but insignificant among non-asthmatic
children. Using indicator variables for each category based on genotype and O3 exposure,
Lee et al. (2009b) calculated the effect of TNF-308 GG genotype on the occurrence of
bronchitic symptoms among children with asthma. Figure 7-2 presents adjusted O3
community-specific beta-coefficients plotted against ambient O3 concentration, using
weights proportional to the inverse variance. They further report that they found no
substantial differences in the effect of the GG genotype in asthmatic children in relation
to exposure to PM10, PM2 5, NO2, acid vapor or second-hand smoke exposure. These
results suggest a role of gene-environment interactions such as long-term O3 exposure on
the occurrence of bronchitic symptoms among children with asthma.
                   Q.
                   2?
                   o
                   O o
                   O S.
                   03 E
                   o >,
                   CO W
                   £
                   to
                         20         30         40         50         60         70
                               Average ozone from 10 a.m. to 6 p.m. in communities (ppb)
       Source: Reprinted with permission of John Wiley & Sons (Lee et al.. 2009b).

      Figure 7-2     Ozone modifies the effect of TNF G-308A genotype on bronchitic
                     symptoms among children with asthma in the CHS.  Using
                     indicator variables for each category based on genotype and O3
                     exposure, betas were calculated of TNF-308 GG genotype on the
                     occurrence of bronchitic symptoms among children with asthma.
      Draft - Do Not Cite or Quote
                              7-8
September 2011

-------
 1                   The French Epidemiology study on Genetics and Environment of Asthma (EGEA)
 2                   investigated the relationship between ambient air pollution and asthma severity in a
 3                   cohort in five French cities (Paris, Lyon, Marseille, Montpellier, and Grenoble) (Rage et
 4                   al.. 2009a). In this cross-sectional study, asthma severity over the past 12 months was
 5                   assessed among 328 adult asthmatics using two methods: (1) a four-class severity score
 6                   that integrated clinical events and type of treatment; and (2)  a five-level asthma score
 7                   based only on symptoms. Two measures of exposure were also assessed: (1 [first
 8                   method]) closest monitor data from 1991 to 1995 where atotal of 93%ofthe subjects
 9                   lived within 10 km of a monitor, but where 70% of the O3 concentrations were back-
10                   extrapolated values; and (2 [second method]) a validated spatial model that used
11                   geostatistical interpolations and then assigned air pollutants to the geocoded residential
12                   addresses of all participants and individually assigned exposure to ambient air pollution
13                   estimates. Higher asthma severity scores were  significantly related to both the 8-h avg O3
14                   during April-September and the number of days with 8-h O3 avgs above 55 ppb. Both
15                   exposure assessment methods and severity score methods resulted in very similar
16                   findings. Effect estimates of O3 were similar in three-pollutant models. No PM data were
17                   available. Since these estimates were not sensitive to the inclusion of ambient NO2 in the
18                   three-pollutant models, the authors viewed the  findings not to be explained by particles
19                   which usually have substantial correlations between PM and NO2. Effect estimates for
20                   O3 in three-pollutant models including O3, SO2, and NO2 yielded OR for O3-days of
21                   2.74 (95% CI: 1.68, 4.48) per IQR days of 10-28 (+18) ppb. The effect estimates for SO2
22                   and NO2 in the three-pollutant model were 1.33 (95% CI: 0.85, 2.11) and 0.94 (95% CI:
23                   0.68, 1.29) respectively. Taking into account duration of residence did not change the
24                   result. This study suggests that a higher asthma severity score is related to long-term O3
25                   exposure.

26                   An EGEA follow-up study (Jacquemin et al.. In Press), examines the relationship
27                   between asthma and O3, NO2, and PM10. New aspects considered include:  1)
28                   examination of three domains of asthma control (symptoms, exacerbations, and lung
29                   function); 2) levels of asthma control (controlled, partially controlled, and uncontrolled
30                   asthma); and 3) PMi0 and multi-pollutant analysis. In this cross-sectional analysis,
31                   EGEA2 studied 481 adult subjects with current asthma from 2003 to 2007. The IQRs
32                   were 11 (41-52) (ig/m3 for annual O3 and  13 (25-38) (ig/m3 for summer (April-
33                   September) O3. The association between asthma control and air pollutants was expressed
34                   by ORs (reported for one IQR of the pollutant), derived from multinomial logistic
35                   regression. For each factor, the simultaneous assessment of the risk for uncontrolled
36                   asthma and for partly controlled asthma was compared with controlled asthma using a
37                   composite of the three domains. In crude and adjusted models, O3-sum and PM10 were
3 8                   positively associated with partly controlled and uncontrolled asthma, with a clear gradient
      Draft - Do Not Cite or Quote                       7-9                                September 2011

-------
 1                   from controlled, partly controlled (OR=1.53, 95%CI: 1.01, 2.33) and uncontrolled
 2                   (OR=2.14, 95% CI:  1.34, 3.43) (from the multinomial logistic regression).

 3                   Separately, they used a composite asthma control classification that used the ordinal
 4                   logistic regression for risk comparing controlled to partly controlled asthma and
 5                   comparing partly controlled to uncontrolled asthma.  For these two pollutants, the ORs
 6                   assessed using the ordinal logistic regression were significant (ORs were 1.69 (95% CI:
 7                   1.22, 2.34) and 1.35 (95% CI: 1.13, 1.64) for O3-sum and PM10, respectively). For two
 8                   pollutant models using the ordinal logistic regression, the adjusted ORs for O3-sum and
 9                   PMio included simultaneously in a unique model were 1.50 (95% CI: 1.07, 2.11) for O3-
10                   sum and 1.28 (95% CI: 1.06, 1.55) for PM10, respectively. This result suggests that the
11                   effects of both pollutants are independent.

12                   The analysis of the associations between air pollution for all asthma subjects and each
13                   one of the three asthma control domains showed the  following:  1) for lung function
14                   defined dichotomously as % predicted FEVj value < or > =80 (OR=1.35, 95%CI: 0.80,
15                   2.28 for adjusted O3-sum); 2) for symptoms defined as asthma attacks or dyspnoea or
16                   woken by asthma attack or shortness of breath in the past three months (OR=1.59,
17                   95%CI: 1.10, 2.30 for adjusted O3-sum); and for exacerbations defined at least one
18                   hospitalizations or ER visits in the last year or oral corticosteroids in the past three
19                   months (OR=1.58, 95%CI: 0.97, 2.59 for adjusted O3-sum). Since the estimates for both
20                   pollutants were more stable and significant when using the integrated measure of asthma
21                   control, this indicates that the results are not driven by one domain. These results support
22                   an effect of long-term exposure to O3 on asthma control in adulthood in subjects with
23                   pre-existing asthma.

24                   The interrelationships between variants in catalase (CAT) and myeloperoxidase (MPO)
25                   genes, ambient pollutants, and acute respiratory illness were investigated in a national
26                   U.S. cohort (Wenten et al.. 2009). Health information, air pollution, and incident
27                   respiratory-related school absences were ascertained in January-June 1996 for 1,136
28                   Hispanic and non-Hispanic white U.S. elementary schoolchildren as part of the
29                   prospective Air Pollution and Absence  Study, a population based cohort study conducted
30                   as part of the CHS. A related earlier study (Gilliland et al.. 2001). which was discussed in
31                   the 2006 O3 AQCD, examined the effects of ambient air pollution on school absenteeism
32                   due to respiratory illness without a genetic aspect to the study. In a new study Wenten et
33                   al. (2009) hypothesized that variation in the level or function of these enzymes would
34                   modulate respiratory illness risk, especially under high levels of oxidative stress. The
35                   joint effect of these two genes on respiratory illness was examined. Risk of respiratory-
36                   related school absences was elevated for children with the CAT (G/G) and MPO (G/A or
37                   A/A) genes (relative risk = 1.35, [95% CI: 1.03, 1.77]; P-interaction = 0.005).  To assess
      Draft - Do Not Cite or Quote                       7-10                                 September 2011

-------
 1                   effects of long-term average levels of O3 on acute effects, communities were divided into
 2                   high and low exposure groups by median levels (46.9 ppb O3). The epistatic effect of
 3                   CAT and MPO variants was evident in communities exhibiting high ambient O3 levels
 4                   (P-interaction = 0.03). The association of respiratory-illness absences with functional
 5                   variants in CAT and MPO that differ by air pollution levels illustrates the need to
 6                   consider genetic epistasis in assessing gene-environment interactions. In high O3
 7                   communities, CAT/MPO genotypes that resulted in decreased oxidative stress were
 8                   associated with a decreased risk of respiratory related school absences compared with the
 9                   CAT/MPO wild-type genotype (Relative Risk [RR] = 0.42, [95% CI: 0.20, 0.89]).
             7.2.2   Asthma Hospital Admissions and ED Visits

10                   The studies on O3-related hospital discharges and emergency department (ED) visits for
11                   asthma and respiratory disease that were available in the 2006 O3 AQCD mainly looked
12                   at the daily time metric. Collectively the short-term O3 studies presented earlier in
13                   Section 6.2.7.5 indicate that there is evidence for increases in both hospital admissions
14                   and ED visits related to both all respiratory outcomes and asthma with stronger
15                   associations in the warm months. New studies evaluated long-term O3 exposure metrics
16                   providing a new line of evidence that suggests a positive exposure-response relationship
17                   between first asthma hospital admission and long-term O3 exposure.

18                   An ecologic study (Moore et al.. 2008) evaluated time trends  in associations between
19                   declining warm-season O3 concentrations and hospitalization  for asthma in children in
20                   California's South Coast Air Basin who ranged in age from birth to 19 years. Quarterly
21                   average concentrations from 195 spatial grids, 10* 10 km, were used. Ozone was the only
22                   pollutant associated with increased hospital admissions over the study period. A linear
23                   relation was observed for asthma hospital discharges (Moore et al.. 2008). A matched
24                   case-control study (Karr et al.. 2007) was conducted of infant  bronchiolitis (ICD 9, code
25                   466.1) hospitalization and two measures of long-term pollutant exposure (the month prior
26                   to hospitalization and the lifetime average) for O3 in the South Coast Air Basin of
27                   southern California among 18,595 infants born between 1995  and 2000. Ozone was
28                   associated with reduced risk in the single-pollutant model, but this relation did not persist
29                   in multi-pollutant models (CO, NO2 and PM2 5).

30                   In a cross-sectional study, Meng et al. (2010) examined associations between air
31                   pollution and asthma morbidity in the San Joaquin Valley in California by using the 2001
32                   California Health Interview Survey data from subjects ages 1 to 65+ who reported
33                   physician-diagnosed asthma (n = 1,502). Subjects were assigned annual average
34                   concentrations for O3 based on residential ZIP code and the closet air monitoring station
      Draft - Do Not Cite or Quote                       7-11                                 September 2011

-------
 1                   within 8 km but did not have data on duration of residence. Multi-pollutant models for O3
 2                   and PM did not differ substantially from single-pollutant estimates, indicating that
 3                   pollutant multi-collinearity is not a problem in these analyses. The authors reported
 4                   increased asthma-related ED visits or hospitalizations for O3 (OR=1.49; [95% CI: 1.05,
 5                   2.11] per 10 ppb) for all ages. Positive associations were obtained for symptoms but 95%
 6                   CIs included null values. Associations for symptoms for adults (ages 18 +) were observed
 7                   (OR=1.40; [95% CI: 1.02, 1.91] per 10 ppb).

 8                   Associations between air pollution and poorly controlled asthma among adults in
 9                   Los Angeles and San Diego Counties, were investigated using the California Health
10                   Interview Survey data collected between November 2000 and September 2001 (Meng et
11                   al.. 2007). Each respondent was assigned an annual average concentration measured at
12                   the nearest station within 5 miles of the residential cross-street intersection. Poorly
13                   controlled asthma was defined as having daily or weekly asthma symptoms or at least one
14                   ED visit or hospitalization because of asthma during the past 12 months. This cross-
15                   sectional study reports an OR of 3.34 (95% CI: 1.01, 11.09) for poorly controlled asthma
16                   when comparing those 65 years of age and older above the 90th percentile (28.7 ppb)
17                   level to those below that level. Multi-pollutant (PM) analysis produced similar results.

18                   Evidence associating long-term O3 exposure to first asthma hospital admission in a
19                   concentration-response relationship is provided in  a retrospective cohort study (Lin et al..
20                   2008b). This study investigated the association between chronic exposure to O3 and
21                   childhood asthma admissions (defined as a principal diagnosis of ICD9, code 493) by
22                   following a birth cohort of 1,204,396 eligible births born in New York State during 1995-
23                   1999 to first asthma admission or until 31 December 2000. There were 10,429 (0.87%)
24                   children admitted to the hospital for asthma between 1 and 6 years of age. The asthma
25                   hospitalization rate in New York State in 1993 was 2.87 per 1,000 (Linetal.. 1999).
26                   Three annual indicators (all 8-h max from 10:00 a.m. to 6:00 p.m.) were used to define
27                   chronic O3 exposure: (1) mean concentration during the follow-up period (41.06 ppb); (2)
28                   mean concentration during the O3 season (50.62 ppb); and (3) proportion of follow-up
29                   days with O3 levels >70 ppb. In this study the authors aimed to predict the risk of having
30                   asthma admissions in a birth cohort, but the time to the first admission in children that is
31                   usually analyzed in survival models was not their primary interest. The effects of
32                   co-pollutants were assessed and controlled for using the Air Quality Index (AQI).
33                   Interaction terms were used to assess potential effect modifications. A positive
34                   association between chronic exposure to O3 and childhood asthma hospital admissions
35                   was observed indicating that children exposed to high O3 levels over time are more likely
36                   to develop asthma severe enough to be admitted to the hospital. The various factors were
37                   examined and differences were found for younger  children (1-2 years), poor
38                   neighborhoods, Medicaid/self-paid births, geographic region and others. As shown in
      Draft - Do Not Cite or Quote                       7-12                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 9
10
11
12
13
14
15
16
Adjusted for child's sex, age, birth weight, and gestational age; maternal race, ethnicity, age, education,
insurance, and smoking status during pregnancy; and regional poverty level and temperature. ORs by low,
medium, and high exposure are shown for New York City (NYC: low [37.3 ppb],
medium [37.3 -38.11] ppb, high [38.11 + ppb])  and other New York State regions (Other
NYS regions: low [42.58 ppb], medium [42.58-45.06 ppb], high [45.06+ ppb]) for first
asthma hospital admission.

Figure 7-3, positive concentration-response relationships were observed. Asthma
admissions were significantly associated with increased O3 levels for all chronic
exposure indicators (ORs, 1.16-1.68). When estimating the O3 effect using the
exceedance proportion, an increase was observed (OR=1.68; [95% CI:  1.64,  1.73]) in
hospital admissions with an IQR (2.51%) increase in O3. A proportional hazards model
for the New York City data was run as a sensitivity analysis and it yielded similar results
between asthma admissions and chronic exposure to O3 (Cox model: HR = 1.14, [95%
CI: 1.124, 1.155] is similar to logistic model results: OR= 1.16 (95% CI: 1.15, 1.17)
(Lin. 2010). Thus, this study provides  evidence associating long-term O3 exposure to first
asthma hospital admission in a concentration-response relationship.
3.0
2.5
« 2.0
^p
o^
8 1.5
g 1.0
0.5
n
i i Low exposure 0-33%
i i Medium exposure 34-66%
•zzi High exposure ^ 67%
1
(1.29
1.00
(ref)

43
-1.58)
T
1
1.6E
1.52-1

1
.80) (1.4E
1.00
(ref)


(1
.64
-1.82)
T
1
2.06
07 9 n

7)
                                    New York City
                                          Other NYS regions
                                 Regions
       Adjusted for child's sex, age, birth weight, and gestational age; maternal race, ethnicity, age, education, insurance, and smoking
      status during pregnancy; and regional poverty level and temperature. ORs by low, medium, and high exposure are shown for
      New York City (NYC: low [37.3 ppb], medium [37.3 - 38.11] ppb, high [38.11 + ppb]) and other New York State regions (Other NYS
      regions: low [42.58 ppb], medium [42.58-45.06 ppb], high [45.06+ ppb]) for first asthma hospital admission.

      Figure 7-3     Ozone-asthma concentration-response relationship using the mean
                       concentration during the entire follow-up period.
      Draft - Do Not Cite or Quote
                                7-13
September 2011

-------
             7.2.3   Pulmonary Structure and Function

 1                   The definitive 8-year follow-up analysis of the first cohort of the CHS, which is
 2                   discussed in Section 7.2 (Gauderman et al., 2004). provided little evidence that long-term
 3                   exposure to ambient O3 was associated with significant deficits in the growth rate of lung
 4                   function in children. A later CHS study (Islam et al., 2007) examined relationships
 5                   between air pollution, lung function, and new onset asthma and reported no substantial
 6                   differences in the effect of O3 on lung function. Ozone concentrations from the least to
 7                   most polluted communities (mean annual average of 8-h avg O3) ranged from 30 to
 8                   65 ppb, as compared to the ranges observed for the other pollutants, which had four- to
 9                   eightfold differences in concentrations. In a more recent CHS study, Breton et al. (2011)
10                   hypothesized that genetic variation in genes on the glutathione metabolic pathway may
11                   influence the association between ambient air pollutant exposures and lung function
12                   growth in children. They investigated whether genetic variation in glutathione genes
13                   GSS, GSR, GCLC, and GCLM was associated with lung function growth in healthy
14                   children using data collected on 2,106 children over an  8-year time-period as part of the
15                   Children's Health  Study. Breton et al. (2011) found that variation in the GSS locus was
16                   associated with differences in susceptibility of children for lung function growth deficits
17                   associated with NO2, PMi0, PM25, elemental carbon, organic carbon, and O3. The
18                   negative effects of air pollutants were largely observed within participants who had a
19                   particular GSS haplotype.  The effects ranged from -124.2 to -149.1 mL for FEVi, -92.9
20                   to -126.7 mL for FVC and -193.9 to -277.9 mL/s for MMEF for all pollutants except O3,
21                   for which some positive associations were reported:  25.9 mL for FEVi; 0.1 mL for FVC,
22                   and 166.5 mL/s for MMEF. Ozone was associated with larger decreases in lung function
23                   in children without this haplotype, when compared to the other pollutants with values of -
24                   76.6 mL for FEVj, -17.2 mL for FVC, and -200.3 mL/s for MMEF, but only the
25                   association with MMEF was statistically significant.

26                   As discussed in the 2006 O3 AQCD, a study of freshman students at the University of
27                   California, Berkeley reported that lifetime exposure to O3 was associated with decreased
28                   measures of small airways (<2 mm) function (FEF75 and FEF25_75) (Tager et al.. 2005).
29                   There was an interaction with the FEF25-75/FVC ratio, a measure of intrinsic airway size.
30                   Subjects with a large ratio were less likely to have decreases in FEF75 and FEF25_75 for a
31                   given estimated lifetime exposure to O3. Kinney and Lippmann (2000) examined 72
32                   nonsmoking adults (mean  age 20 years) from the second-year class of students at the U.S.
33                   Military Academy in West Point, NY, and reported results that appear to be consistent
34                   with a decline in lung function that may in part be due to O3 exposures over a period of
35                   several summer months. Ilhorst et al. (2004) examined 2,153 children with a median age
36                   of 7.6 years and reported pulmonary function results which indicated that significantly
37                   lower FVC and FEVi increases were associated with higher O3 exposures over the

      Draft - Do Not Cite or Quote                       7-14                                 September 2011

-------
 1                   medium-term of several summer months, but not over several months in the winter.
 2                   Semi-annual mean O3 concentrations ranged from 22 to 54 ppb during the summer
 3                   months and 4 to 36 ppb during the winter months. However, over the longer-term
 4                   3.5-year period Ilhorst et al. (2004) found no associations between increases in lung
 5                   function and mean summer months O3 levels for FVC and FEVi, in contrast to the
 6                   significant medium-term effects. Frischer et al. (1999) showed results similar to the
 7                   Ilhorst et al. (2004) study.

 8                   Mortimer et al. (2008a, b) examined the association of prenatal and lifetime exposures to
 9                   air pollutants with pulmonary function and allergen sensitization in a subset of asthmatic
10                   children (ages 6-11) included in the Fresno Asthmatic Children's Environment Study
11                   (FACES). Monthly means of pollutant levels for the years 1989-2000 were created and
12                   averaged separately across several important developmental time-periods, including: the
13                   entire pregnancy, each trimester, the first 3 years of life, the first 6 years of life, and the
14                   entire lifetime. In the first analysis (Mortimer et al.. 2008a). negative effects on
15                   pulmonary function were found for exposure to PMi0, NO2, and CO during key neonatal
16                   and early life developmental periods. The authors did not find a negative effect of
17                   exposure to O3 within this cohort. In the second analysis (Mortimer et al.. 2008b).
18                   sensitization to at least one allergen was associated, in general, with higher levels of CO
19                   and PM10 during the entire pregnancy and second trimester, and higher PM10 during the
20                   first 2 years of life. Lower exposure to O3  during the entire pregnancy or second trimester
21                   was associated with an increased risk of allergen sensitization. Although the pollutant
22                   metrics across time periods were correlated, the strongest associations with the outcomes
23                   were observed for prenatal exposures. Though it may be difficult to disentangle the effect
24                   of prenatal and postnatal exposures, the models from this group of studies suggest that
25                   each time period of exposure may contribute independently to different dimensions of
26                   school-aged children's pulmonary function. For 4 of the 8 pulmonary-function measures
27                   (FVC, FEVi, PEF, FEF25.75), prenatal exposures were more influential on pulmonary
28                   function than early-lifetime metrics, while, in contrast, the ratio of measures (FEVi/FVC
29                   and FEF25_75/FVC) were most influenced by postnatal exposures. When lifetime metrics
30                   were considered alone, or in combination with the prenatal  metrics, the lifetime measures
31                   were not associated with any of the outcomes. This suggests that the timing of the O3
32                   exposure may be more important than the overall dose, and prenatal exposures are not
33                   just markers for lifetime or current exposures.

34                   Latzin et al. (2009) examined whether prenatal exposure to air pollution was associated
35                   with lung function changes in the newborn. Tidal breathing, lung volume, ventilation
36                   inhomogeneity and eNO were measured in 241 unsedated, sleeping neonates (age =
37                   5 weeks). Consistent with the previous studies, no association was found for prenatal
38                   exposure to O3 and lung function.
      Draft - Do Not Cite or Quote                       7-15                                September 2011

-------
 1                   In a cross-sectional study of adults, Qian et al. (2005) examined the association oflong-
 2                   term exposure to O3 and PMi0 with pulmonary function from data of 10,240 middle-aged
 3                   subjects who participated in the Atherosclerosis Risk in Communities (ARIC) study in
 4                   four U.S. communities. A surrogate for long-term O3 exposure from daily data was
 5                   determined at the individual level. Ozone was significantly and negatively associated
 6                   with measures of pulmonary function.

 7                   To determine the extent to which long-term exposure to outdoor air pollution accelerates
 8                   adult decline in lung function, Forbes et al. (2009b) studied the association between
 9                   chronic exposure to outdoor air pollution and lung function in approximately 42,000
10                   adults aged 16 and older who were representatively sampled cross-sectionally from
11                   participants in the Health Survey for England (1995, 1996, 1997, and 2001). FEVj was
12                   not associated with O3 concentrations. In contrast to the results for PM10, NO2, and SO2
13                   combining the results of all the survey years showed that a 5-ppb difference in O3 was
14                   counter-intuitively associated with a higher FEVi by 22 mL.

15                   In a prospective cohort study consisting of school-age, non-asthmatic children in
16                   Mexico City (n = 3,170) who were 8 years of age at the beginning of the study, Rojas-
17                   Martinez et al. (2007) evaluated the association between long-term exposure to  O3, PM10
18                   and NO2 and lung function growth every 6 months from April 1996 through May 1999.
19                   Exposure data were provided by 10 air quality monitor stations located within 2 km of
20                   each child's school. Over the  study period, 8-h O3 concentrations ranged from 60 ppb
21                   (SD, ±25) in the northeast area of Mexico City to 90 ppb (SD, ±34) in the southwest, with
22                   an overall mean of 69.8 ppb. In multi-pollutant models, an IQR increase in mean O3
23                   concentration of 11.3 ppb was associated with an annual deficit in FEVi of 12 mL in
24                   girls and 4 mL in boys. Single-pollutant models showed an association between ambient
25                   pollutants (O3, PMi0 and NO2) and deficits in lung function growth. While the estimates
26                   from copollutant models were not substantially different than single pollutant models,
27                   independent effects for pollutants could not be estimated accurately because the traffic -
28                   related pollutants were correlated. To reduce exposure misclassification,
29                   microenvironmental and personal exposure assessments were conducted in a randomly
30                   selected subsample of 60 children using passive O3 samplers. Personal O3 concentrations
31                   were correlated (p < 0.05) with the measurements obtained from the fixed-site air
32                   monitoring stations.

33                   In the 2006 O3 AQCD, few studies had investigated the effect of chronic O3 exposure on
34                   pulmonary function. The strongest evidence was for medium-term effects of extended O3
35                   exposures over several summer months on lung function in children, i.e., reduced lung
36                   function growth being associated with higher ambient O3 levels. Longer-term studies,
37                   investigating the association of chronic O3 exposure on lung function such as the CHS,
      Draft - Do Not Cite or Quote                       7-16                                September 2011

-------
 1                  were inconclusive. Short-term O3 exposure studies presented in Section 6.2.1.2 provide a
 2                  cumulative body of epidemiologic evidence that strongly supports associations between
 3                  ambient O3 exposure and decrements in lung function among children. For new studies
 4                  of long-term O3 exposure relationship to pulmonary function, one study, where O3 and
 5                  other pollutant levels were higher (90 ppb at high end of the range) than those in the
 6                  CHS, observes a relationship between O3 concentration and pulmonary function declines
 7                  in school-aged children. Two studies of adult cohorts provide mixed results where long-
 8                  term exposures were at the high end of the range with levels of 49.5 ppb in one study and
 9                  27 ppb IQR in the other. Thus there is little new evidence to build upon the very limited
10                  studies from the 2006 O3 AQCD.
                    7.2.3.1    Pulmonary Structure and Function: Evidence from
                               lexicological Studies

11                  As reviewed in the 1996 and 2006 O3 AQCDs, there are both qualitative and quantitative
12                  uncertainties in the extrapolation of data generated by rodent toxicology studies to the
13                  understanding of health effects observed in humans, as documented by epidemiologic and
14                  controlled exposure studies. Chief among these data extrapolation issues are the
15                  differences between rodent and human respiratory physiology, cellular makeup,
16                  dosimetry, and morphometry (see Chapter 5). However, important evidence is available
17                  from O3-inhalation studies performed in nonhuman primates whose respiratory system
18                  most closely resembles that of the human. A long series of studies have used nonhuman
19                  primates to examine the effect of O3 alone or in combination with an inhaled allergen,
20                  house dust mite antigen, on morphology and lung function. These studies, by Plopper and
21                  colleagues, have demonstrated changes in pulmonary function and airway morphology in
22                  adult and infant nonhuman primates repeatedly exposed to environmentally relevant
23                  concentrations of O3 (Joad et al.. 2008; Carey et al.. 2007; Plopper et al.. 2007; Fanucchi
24                  et al.. 2006; Joad et al.. 2006; Evans et al.. 2004; Larson et al.. 2004; Tran et al.. 2004;
25                  Evans et al.. 2003b; Schelegle etal.. 2003: Fanucchi et al.. 2000; Hydeetal.. 1989;
26                  Harkema et al..  1987a; Harkema et al.. 1987b; Fuiinaka et al.. 1985). The findings of
27                  these nonhuman primate studies have also been  observed in rodent studies discussed at
28                  the end of this section and included in Table 7-1.

29                  Since publication of the 1996 and 2006 O3 AQCDs, the initial observations in adult
30                  nonhuman primates have been expanded in a series of experiments using infant rhesus
31                  monkeys repeatedly exposed to 0.5 ppm O3 starting at 1 month of age (Plopper et al.,
32                  2007). Many of the observations found in adult monkeys have also been noted in infant
33                  rhesus monkeys, although a direct comparison of the degree of effects between adult and
34                  infant monkeys  has not been reported.  In terms of pulmonary function changes, after
      Draft - Do Not Cite or Quote                       7-17                                September 2011

-------
 1                   several episodic exposures of infant monkeys to O3 (each cycle: 5 days of 0.5 ppm O3 at
 2                   8 h/day, followed by 9 days of filtered air exposure), they observed more than a doubling
 3                   in the baseline airway resistance, which was accompanied by a small increase in airway
 4                   responsiveness to inhaled histamine (Schelegle et al.. 2003). although neither
 5                   measurement was statistically different from filtered air control values. Exposure of
 6                   animals to inhaled house dust mite antigen alone also produced small but not statistically
 7                   significant changes in baseline airway resistance and airway responsiveness, whereas the
 8                   combined exposure to both (O3 + antigen) produced statistically significant and greater
 9                   than additive changes in both functional measurements. This nonhuman primate  evidence
10                   of an O 3-induced change in airway responsiveness supports the biologic plausibility of
11                   long-term exposure to O3 contributing to the effects of asthma in children. To understand
12                   which conducting airways and inflammatory mechanisms are involved in O3-induced
13                   airway hyperresponsiveness in the infant rhesus monkey, a follow-up study examined
14                   airway responsiveness ex vivo in lung slices (Joad et al.. 2006). Using  video microscopy
15                   to morphometrically evaluate the response of bronchi and respiratory bronchioles to
16                   methacholine, (a bronchoconstricting agent commonly used to evaluate airway
17                   responsiveness in asthmatics), the investigators observed differential effects for the two
18                   airway sizes. While episodic exposure to O3 alone (0.5 ppm) had little effect on ex vivo
19                   airway responsiveness in bronchi and respiratory bronchioles, exposure to dust mite
20                   antigen alone produced airway hyperresponsiveness in the large bronchi, whereas O3 +
21                   antigen produced significant increases in airway hyperresponsiveness only in the
22                   respiratory bronchioles. These results suggest that ozone's effect on airway
23                   responsiveness occurs predominantly in the smaller bronchioles, where dosimetric
24                   models indicate the dose would be higher.

25                   The functional changes in the conducting airways of infant rhesus monkeys exposed to
26                   either O3 alone or O3 + antigen were accompanied by a number of cellular and
27                   morphological changes, including a significant fourfold increase in eosinophils, (a cell
28                   type important in allergic asthma), in the bronchoalveolar lavage of infant monkeys
29                   exposed to O3 alone. Thus, these studies demonstrate both functional and cellular
30                   changes in the lung of infant monkeys after cyclic exposure to 0.5 ppm O3. This
31                   concentration, while higher than those used in controlled human exposure studies,
32                   provides relevant information to understanding the adverse effects of ambient O3
33                   exposure on the respiratory tract of humans. No concentration-response data, however,
34                   are available from these nonhuman primate studies.

35                   In addition to these functional and cellular changes, significant structural changes in the
36                   respiratory tract have been observed in infant rhesus monkeys exposed to O3. During
37                   normal respiratory tract development, conducting airways increase in diameter and length
38                   in the infant rhesus monkey. Exposure to O3 alone (5 days of 0.5 ppm  O3 at 8 h/day,
      Draft - Do Not Cite or Quote                       7-18                                September 2011

-------
 1                   followed by 9 days of filtered air exposures for 11 cycles), however, markedly affected
 2                   the growth pattern of distal conducting airways (Fanucchi et al.. 2006). Whereas the first
 3                   alveolar outpocketing occurred at airway generation 13 or 14 in filtered air-control infant
 4                   monkeys, the most proximal alveolarized airways occurred at an average of 10 airway
 5                   generations in O3-exposed monkeys. Similarly, the diameter and length of the terminal
 6                   and respiratory bronchioles were significantly decreased in O3-exposed monkeys.
 7                   Importantly, the O3-induced structural pathway changes persisted after recovery in
 8                   filtered air for 6 months after cessation of the O3 exposures. These structural effects were
 9                   accompanied by significant increases in mucus goblet cell mass, alterations in smooth
10                   muscle orientation in the respiratory bronchioles, epithelial nerve fiber distribution, and
11                   basement membrane zone morphometry. These latter effects are significant because of
12                   their potential contribution to airway obstruction and airway hyperresponsiveness which
13                   are central features of asthma.

14                   As noted above, a significant increase in airway responsiveness to inhaled histamine
15                   occurred in infant rhesus monkeys exposed to O3 + house dust mite antigen, but not to O3
16                   alone (Schelegle et al.. 2003). To study the underlying mechanisms of this airway
17                   hyperresponsiveness, these investigators evaluated the effect of exposure to O3 alone and
18                   in combination with (+) antigen on non-specific airway responsiveness to methacholine
19                   at different airway generations. After exposure to filtered air, O3, antigen, or O3 +
20                   antigen, the bronchi and respiratory bronchioles of 6-month-old monkeys were
21                   challenged ex vivo with methacholine. Exposure to O3 alone had no significant effect on
22                   airway responsiveness to methacholine in either airway, whereas O3 + antigen produced
23                   a significant increase in airway responsiveness in the respiratory bronchioles but not the
24                   larger bronchi.

25                   Because many cellular and biochemical factors are known to contribute to allergic
26                   asthma, the effect of exposure to O3 alone or O3 + antigen on immune system parameters
27                   was also examined in infant rhesus monkeys. Mast cells, which contribute to asthma via
28                   the release of potent proteases, were elevated in animals exposed to antigen alone but O3
29                   alone had little effect on mast cell numbers and the response of animals exposed to O3 +
30                   antigen was not different from that of animals exposed to antigen alone; thus suggesting
31                   that mast cells played little role in the interaction between O3 and antigen in this model of
32                   allergic asthma (Van Winkle et al.. 2010). Increases in CD4+ and CD8+ lymphocytes
33                   were observed at 6 months of age in the blood and bronchoalveolar lavage fluid of infant
34                   rhesus monkeys exposed to O3 + antigen but not in monkeys exposed to either agent
35                   alone (Miller et al.. 2009). Activated lymphocytes (i.e., CD25+ cells) were
36                   morphometrically evaluated in the airway mucosa and significantly increased in infant
37                   monkeys exposed to antigen alone or O3 + antigen. Although O3 alone had no effect on
3 8                   CD25+ cells, it did alter the anatomic distribution of CD25+ cells within the airways.
      Draft - Do Not Cite or Quote                       7-19                                September 2011

-------
 1                   Ozone had only a small effect on these sets of immune cells and did not produce a strong
 2                   interaction with an inhaled allergen in this nonhuman primate model.

 3                   In addition to alterations in the immune system, nervous system interactions with
 4                   epithelial cells are thought to play a contributing role to airway hyperresponsiveness. As
 5                   noted in the 2006 O3 AQCD, exposure of infant rhesus monkeys altered the normal
 6                   development of neural innervation in the epithelium of the conducting airways (Larson et
 7                   al.. 2004). Whereas, a significant reduction in airway innervation occurred after exposure
 8                   to O3 alone, a significantly greater reduction occurred in monkeys exposed to O3 +
 9                   antigen. This reduction in overall airway innervation was accompanied, however, by an
10                   increase in the abundance of protein gene product 9.5, a nonspecific neural marker.
11                   Significant increases in protein gene product 9.5 were still observed in O3 alone- and O3
12                   + antigen-exposed infant monkeys after a 6-month recovery protocol (Kajekar et al.,
13                   2007). Thus, in addition to structural, immune, and inflammatory effects, exposure to O3
14                   produces alterations in airway innervation which may contribute to O3-induced
15                   exacerbation of asthma.

16                   A number of studies in both  nonhuman primates and rodents demonstrate that O3
17                   exposure can increase collagen synthesis and deposition, inducing fibrotic-like changes  in
18                   the lung (Lastetal.. 1994; Chang etal.. 1992; Moffatt et al..  1987; Reiser et al.. 1987;
19                   Lastetal.. 1984). Increased collagen content is often associated with elevated abnormal
20                   cross links that appear to be irreversible (Reiser et al.. 1987). Generally changes in
21                   collagen content have been observed in rats exposed to 0.5 ppm O3 or higher, although
22                   extracellular matrix thickening has been observed in the lungs of rats exposed to an urban
23                   pattern of O3 with daily peaks of 0.25 ppm for 38 weeks (Chang etal..  1992; Chang et
24                   al., 1991). A more recent study using an urban pattern of exposure to 0.5 ppm O3
25                   demonstrated that O3-induced collagen deposition in mice is dependent on the activity of
26                   TGF-(3 (Katre etal.. 2011). Sex differences have been observed with respect to increased
27                   centriacinar collagen deposition and crosslinking, which was observed in female but not
28                   male rats exposed to 0.5 and 1.0 ppm O3 for 20 months (Lastetal.. 1994). Few other
29                   long-term exposure morphological studies have presented sex differences and most only
30                   evaluated males. It is unclear what the long-term effects of these structural changes may
31                   be. A number of studies indicate that structural changes in the respiratory system are
32                   persistent or irreversible. For example, O3-induced hyperplasia was still evident in the
33                   nasal epithelia of rats  13 weeks after recovery from 0.5 ppm O3 exposure (Harkema et
34                   al., 1999). In a study of episodic exposure to 0.25 ppm O3, Chang et al. (1992) observed
35                   no reversal of basement membrane thickening in rat lungs up to 17 weeks post-exposure.
36                   Episodic exposure (0.25 ppm O3, every other month) of monkeys induced equivalent
37                   morphological changes compared to continuously exposed animals, even though they
      Draft - Do Not Cite or Quote                       7-20                                 September 2011

-------
1                   were exposed for half the time and evaluation occurred a month after exposure ceased as
2                   opposed to immediately (Tyler etal.. 1988).
     Draft - Do Not Cite or Quote                       7-21                                 September 2011

-------
Table 7-1      Respiratory effects in nonhuman primates and rodents resulting
              from long-term Oa exposure
Study
Catalanoetal. (1995a;
1995b); Chang et al.
(1995): Harkema et al.
(1997a:1997b:1994):
Last etal. (1994);
Pinkertonetal. (1995);
Plopperetal.(1994);
Stockstill et al. (1995);
Pinkertonetal. (1998)
Herbert etal. (1996)
Chang et al. (1991)
Chang et al. (1992)
Barry etal. (1985. 1983)
Tyler etal. (1988)
Harkema et al. (1999)
Van Bree etal. (2002)
Katre etal. (2011)
Model
Rat, male and
female, Fischer
F344, 6-8 weeks
old
Mice, male and
female, B6C3F1,
6-7 weeks old,
Rat, male, F344, 6
weeks old
Rat, male, F344, 6
weeks old
Rat, male, 1 day
old or 6 weeks old
Monkey; male,
Macaca
fascicularis, 7 mo
old
Rat, male, Fischer
F344/N HSD, 10-
14 weeks old
Rat, male, Wistar,
7 weeks old, n =
5/group
Mice; male,
C57BL/6,
6-8 week sold
O3 (ppm)
0.12
0.5
1.0
0.12
0.50
1.0
Continuous: 0.1 2
or 0.25
Episodic/urban:
baseline 0.06;
peak 0.25
baseline 0.06;
peak 0.25
0.1 2 (adults only)
0.25
0.25
0.25
0.5
0.4
0.5
Exposure Duration
6 h/day, 5 days/week for
20 months
6 h/day, 5 days/week for
24 and 30 months
Continuous: 12 h/day for
6 weeks
Simulated urban pattern;
slow rise to peak 9 h/day,
5 days/week, 13 weeks
Slow rise to peak
9 h/day, 5 days/week,
13 and 78 weeks
Recovery in filtered
air for 6 or 17 weeks
12 h/day for 6 weeks
8 h/day, 7 days/week,
Daily for 18 mo or
episodically every other
mo for 18mos
Episodic group
evaluated 1 mo post
exposure
8 h/day, 7 days/week for
13 weeks
23.5 h/day for 1,3,7,
28,or 56 days
8 h/day, [5 days/week
03, and 2 days filtered
air] for 5 or 10 cycles
Effects
Effects similar to (or a model of) early fibrotic human
disease were greater in the periacinar region than in
terminal bronchioles. Thickened alveolar septa
observed in rats exposed to 0.12 ppm 03. Other
effects (e.g., mucous cell metaplasia in the nose and
mild fibrotic response in the parenchyma, increased
collagen in CAR of females) observed at 0.5 to 1 .0
ppm. Some morphometric changes such as epithelial
thickening and bronchiolarization occurred after 2 or 3
months of exposure to 1 .0 ppm.
Similar to the response of rats in the same study (see
rat above). Effects were seen in the nose and
centriacinar region of the lung at 0.5 and 1 .0 ppm.
Increased Type 1 and 2 epithelial volume assessed by
TEM. Linear relationship observed between increases
in Type 1 epithelial cell volume and concentration x
time product. Degree of injury not related to pattern of
exposure (continuous or episodic).
Progressive epithelial hyperplasia, fibroblast
proliferation, and interstitial matrix accumulation
observed using TEM. Interstitial matrix thickening due
to deposition of basement membrane and collagen
fibers. Partial recovery of interstitial matrix during
follow-up periods in air; but no resolution of basement
membrane thickening.
Lung and alveolar development not significantly
affected. Increased Type 1 and 2 epithelial cells and
AM in CAR alveoli, thickened Type 1 cells with smaller
volume and less surface coverage as assessed by
TEM (adults and juveniles). In adults, smaller but
statistically significant similar changes at 0.12 ppm,
suggesting linear concentration-response relationship.
No statistically significant age-related effects
observed.
Increased collagen content, chest wall compliance,
and inspiratory capacity in episodic group only.
Respiratory bronchiolitis in both groups. Episodically
exposed group incurred greater alterations in
physiology and biochemistry and equivalent changes
in morphometry even though exposed for half the time
as the daily exposure group.
Mucous cell hyperplasia in nasal epithelium after
exposure to 0.25 and 0.5 ppm 03; still evident after 13
weeks recovery from 0.5 ppm 03 exposure.
Acute inflammatory response in BALF reached a
maximum at day 1 and resolved within 6 days during
exposure. Centriacinar region inflammatory responses
throughout 0 3 exposure with increased collagen and
bronchiolization still present after a recovery period.
Sustained elevation in TGF-p and PAI-1 in lung (5 or
10 cycles); elevated a-SMA and increased collagen
deposition in airway walls (after 10 cycles). Collagen
increase shown to depend on TGF-p.
Draft - Do Not Cite or Quote
7-22
September 2011

-------
             Study
                              Model
     (ppm)     Exposure Duration
                                    Effects
       Schelegle et al. (2003):  Monkey; Rhesus,
                         30 days old*
                                        0.5
              8 h/day for 5 days, every Goblet cell metaplasia, increased AHR, and increased
              5 days for a total of 11   markers of allergic asthma (e.g., eosinophilia) were
              episodes             observed, suggesting that episodic exposure to 03
                                 alters postnatal morphogenesis and epithelial
                                 differentiation and enhances the allergic effects of
                                 house dust mite allergen in the lungs of infant
             	primates.	
       Larson etal. (2004'
                         Monkey;Macaca
                         mulatta, 30 days
                         old*
0.5
11 episodes of 5 days
each, 8 h/day followed
by 9 days of recovery
03 or 03 + house dust mite antigen caused changes
in density and number of airway epithelial nerves in
small conducting airways. Suggests episodic 03 alters
pattern of neural innervation in epithelial compartment
of developing lungs.	
       Plopper et al. (2007)
                          Monkey; Rhesus,  0.5
                         30 days old*
              5 months of episodic    Non-significant increases airway resistance and airway
              exposure; 5 days 03    responsiveness with 03 or inhaled allergen alone.
              followed by 9 days of    Allergen + 03 produced additive changes in both
              filtered air, 8h/day.	measures.	
       Fanucchietal. (2006'
                         Monkey; male
                         Rhesus,30 days
                         old
0.5
5 months of episodic    Cellular changes and significant structural changes in
exposure; 5 days 03    the distal respiratory tract in infant rhesus monkeys
followed by 9 days of    exposed to 03 postnatally.
filtered air, 8h/day.	
       Reiser etal. (1987)
                         Monkey; male and 0.61
                         female
                         Cynomolgus 6-7
                         mo old
              8 h/day for 1 year      Increased lung collagen content associated with
                                 elevated abnormal cross links that were irreversibly
                                 deposited.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
       * sex not reported
                       Collectively, evidence from animal studies strongly suggests that chronic O3 exposure is

                       capable of damaging the distal airways and proximal alveoli, resulting in lung tissue

                       remodeling and leading to apparent irreversible changes. Potentially, persistent

                       inflammation and interstitial remodeling play an important role in the progression and

                       development of chronic lung disease. Further discussion of the modes of action that lead

                       to O3-induced morphological changes can be found in Section 5.3.7. The findings

                       reported in chronic animal studies offer insight into potential biological mechanisms for

                       the suggested association between seasonal O3 exposure and reduced lung function

                       development in children as observed in epidemiologic studies (see Section 7.2.3).

                       Discussion of mechanisms involved in lifestage susceptibility and developmental effects

                       can be found in Section 5.4.2.4.
12
13
14
15
16
17
18
              7.2.4   Pulmonary Inflammation,  Injury,  and Oxidative Stress


                       The 2006 O3 AQCD stated that the extensive human clinical and animal toxicological
                       evidence, together with the limited epidemiologic evidence available,  suggests a causal
                       role for O3 in inflammatory responses in the airways. Short-term exposure epidemiologic
                       studies discussed earlier in Section 6.2.3.2 show consistent associations of O3 exposure
                       and increased airway inflammation and oxidative stress. Further discussion of the
                       mechanisms underlying inflammation and oxidative stress responses can be found in
                       Section 5.3.3. Though the majority of recent studies focus on short-term exposures,
       Draft - Do Not Cite or Quote
                                                          7-23
                                                          September 2011

-------
 1                   several epidemiologic and toxicology studies of long-term exposure add to observations
 2                   of O 3 -induced inflammation and inj ury.

 3                   Inflammatory markers and peak expiratory pulmonary function were examined in 37
 4                   allergic children with physician-diagnosed mild persistent asthma in a highly polluted
 5                   urban area in Italy and then again 7 days after relocation to a rural location with
 6                   significantly lower pollutant levels (Renzetti et al., 2009). The authors observed a
 7                   fourfold decrease in nasal eosinophils and a statistically significant decrease in fractional
 8                   exhaled nitric oxide along with an improvement in lower airway function. Several
 9                   pollutants were examined, including PMi0, NO2,  and O3, though pollutant-specific
10                   results were not presented. These results are consistent with studies showing that traffic -
11                   related exposures are  associated with increased airway inflammation and reduced lung
12                   function in children with asthma and contribute to the notion that this negative influence
13                   may be rapidly reversible. Exhaled NO (eNO) has been shown to be a useful biomarker
14                   for airway inflammation in large population-based studies (Linn et al., 2009). Thus, while
15                   the time scale of 7 days between examinations for eNO needs to be evaluated for
16                   appropriateness, the results suggest that inflammatory responses are reduced when O3
17                   levels are decreased.

18                   Chest radiographs (CXR) of 249 children in Mexico City who were chronically exposed
19                   to O3 and PM25 were analyzed by Calderon-Garciduenas et al. (2006). They reported an
20                   association between chronic exposures to O3 and other pollutants and a significant
21                   increase in abnormal CXR's and lung CTs suggestive of a bronchiolar, peribronchiolar,
22                   and/or alveolar duct inflammatory process, in clinically healthy children with no risk
23                   factors for lung disease. These CXR and CT results should be viewed with caution
24                   because it is  difficult to attribute effects to O3 exposure.

25                   In a cross-sectional study, Wood et al. (2009) examined the association of outdoor air
26                   pollution with respiratory phenotype  (PiZZ type)  in alpha 1-Antitrypsin deficiency (a-
27                   ATD) from the U.K. a-ATD registry. In total, 304 PiZZ subjects underwent full lung
28                   function testing and quantitative high-resolution computed tomography to identify the
29                   presence and severity of COPD - emphysema. Mean annual air pollution data for 2006
30                   was matched to the location of patients' houses and used in regression models to identify
31                   phenotypic associations with pollution controlling for covariates. Relative trends in  O3
32                   levels were assessed to validate use of a single year's data to indicate long-term exposure
33                   and validation; data showed good correlations between modeled and measured data
34                   (Stedman and Kent. 2008). Regression models showed that estimated higher exposure to
35                   O3 exposure was associated with worse gas transfer and more severe emphysema, albeit
36                   accounting for only a small proportion of the lung function variability. This suggests that
37                   a gene-specific group demonstrates a long-term O3  exposure effect.
      Draft - Do Not Cite or Quote                       7-24                                September 2011

-------
 1                   The similarities of nonhuman primates to humans make them attractive models in which
 2                   to study the effects of O3 on the respiratory tract. The nasal mucous membranes, which
 3                   protect the more distal regions of the respiratory tract, are susceptible to injury from O3.
 4                   Carey et al. (2007) conducted a study of O3 exposure in infant rhesus macaques, whose
 5                   nasal airways closely resemble that of humans. Monkeys were exposed either acutely for
 6                   5 days (8 h/day) to 0.5 ppm O3, or episodically for several biweekly cycles alternating
 7                   5 days of 0.5 ppm O3 with 9 days of filtered air (0 ppm O3), designed to mimic human
 8                   exposure (70 days total). All monkeys acutely exposed to O3 had moderate to marked
 9                   necrotizing rhinitis, with focal regions of epitheliar exfoliation, numerous infiltrating
10                   neutrophils, and some eosinophils. The distribution, character, and severity of lesions in
11                   episodically exposed monkeys were similar to that of acutely exposed animals. Neither
12                   group exhibited mucous cell metaplasia proximal to the lesions, a protective adaptation
13                   observed in adult monkeys exposed continuously to 0.3 ppm O3 in another study
14                   (Harkema et al.. 1987a). Adult monkeys also exhibit attenuation of inflammatory
15                   responses with continued daily exposure (Harkema et al.. 1987a). but inflammation did
16                   not resolve  over time in young episodically exposed monkeys(Carey et al.. 2011).
17                   Inflammation in conducting airways has also been observed in rats chronically exposed to
18                   O3. Using an agar-based technique to fill the  alveoli so that only the rat bronchi are
19                   lavaged, a 90-day exposure of rats to 0.8 ppm O3  (8 h/day) elicited significantly elevated
20                   pro-inflammatory eicosanoids PGE2 and 12-HETE in the conducting airway compared to
21                   filtered air-exposed rats (Schmelzer et al.. 2006).
             7.2.5   Allergic Responses

22                   The association of air pollutants with childhood respiratory allergies was examined in the
23                   U.S. using the 1999-2005 National Health Interview Survey of approximately 70,000
24                   children, and ambient air pollution data from the U.S. EPA, with monitors within 20
25                   miles of each child's residential block (Parker et al., 2009). The authors examined the
26                   associations between the reporting of respiratory allergy or hay fever and medium-term
27                   exposure to O3 over several summer months, controlling for demographic and geographic
28                   factors. Increased respiratory allergy/hay fever was associated with increased O3 levels
29                   (adjusted OR per 10 ppb =  1.20; [95% CI: 1.15,1.26]). These associations persisted after
30                   stratification by urban-rural status, inclusion of multiple pollutants (O3, SO2, NO2, PM),
31                   and definition of exposure by differing exposure radii; smaller samples within 5 miles of
32                   monitors were remarkably similar to the primary results. No associations between the
33                   other pollutants and the reporting of respiratory allergy/hay fever were apparent.
34                   Ramadour et al. (2000) reported no relationship between O3 levels and rhinitis symptoms
35                   and hay fever. Hwang et al. (2006) report the prevalence of allergic rhinitis (adjusted OR
      Draft - Do Not Cite or Quote                       7-25                                September 2011

-------
 1                   per 10 ppb = 1.05; [95% CI: 0.98, 1.12]) in a large cross-sectional study in Taiwan. In a
 2                   large cross-sectional study in France, Penard-Morand et al. (2005) reported a positive
 3                   relationship between lifetime allergic rhinitis and O3 exposure in a two-pollutant model
 4                   with NO2. These studies related positive outcomes of allergic response and O3 exposure
 5                   but with variable strength for the effect estimates. A toxicological study reported that
 6                   five weeks of continuous exposure to 0.4 ppm O3 (but not 0.1 or 0.2 ppm O3) augmented
 7                   sneezing and nasal secretions in a guinea pig model of nasal allergy (lijima and
 8                   Kobayashi. 2004). Nasal eosinophils, which participate in allergic disease and
 9                   inflammation, and allergic antibody levels in serum were also elevated by exposure to
10                   concentrations as low as 0.2 ppm (lijima and Kobayashi. 2004).

11                   Nasal eosinophils were observed to decrease by fourfold in 37 atopic, mildly asthmatic
12                   children 7 days after relocation from a highly polluted urban area in Italy to a rural
13                   location with significantly lower pollutant levels (Renzetti et al.. 2009). Inflammatory
14                   and allergic effects of O3 exposure (30 day mean) such as increased eosinophil levels
15                   were observed in children in an Austrian study (Frischer et al.. 2001). Episodic exposure
16                   of infant rhesus monkeys to 0.5 ppm O3 for 5 months appears to significantly increase the
17                   number and proportion of eosinophils in the blood and airways (lavage) [protocol
18                   described above in 7.2.3.1 for Fanucchi et al. (2006)] (Maniar-Hew et al.. 2011). These
19                   changes were not evident at 1 year of age (6 months after O3 exposure ceased). Increased
20                   eosinophils levels have also been observed after acute or prolonged exposures to O3 in
21                   adult bonnet and rhesus monkeys (Hyde et al..  1992; Eustis et al.. 1981).

22                   Total IgE levels were related to air pollution levels in 369 adult  asthmatics in five French
23                   centers using generalized estimated equations (GEE) as part of the EGEA study described
24                   earlier (Rage et al.. 2009b). Geostatistical models were performed on 4x4 km grids to
25                   assess individual outdoor air pollution exposure that was assigned to subject's home
26                   address. Ozone concentrations were positively related to total IgE levels and an increase
27                   of 5 ppb of O3 resulted in an increase of 20.4% (95% CI: 3.0, 40.7) in total IgE levels.
28                   Nearly 75% of the subjects were atopic. In two-pollutant models including O3  and NO2,
29                   the O3 effect estimate was decreased by 25% while the NO2 effect estimate was decreased
30                   by 57%. Associations were not sensitive to adjustment for covariates or the season of IgE
31                   measurements. These cross-sectional results suggest that exposure to O3 may increase
32                   total IgE in adult asthmatics.

33                   Although very few toxicological studies of long-term exposure examining allergy are
34                   available, short-term exposure studies in rodents and nonhuman primates demonstrate
35                   allergic skewing of immune responses and enhanced IgE production. Due to the
36                   persistent nature of these responses, the short-term toxicological evidence lends
      Draft - Do Not Cite or Quote                       7-26                                September 2011

-------
 1                   biological plausibility to the limited epidemiologic findings of an association between
 2                   long-term O3 exposure and allergic outcomes.
            7.2.6   Host Defense

 3                   Short-term exposures to O3 have been shown to cause decreases in host defenses against
 4                   infectious lung disease in animal models. However, acute O3-induced suppression of
 5                   alveolar phagocytosis  and immune functions observed in animals appears to be transient
 6                   and attenuated with continuous or repeated exposures. Chronic exposures (weeks,
 7                   months) of 0.1 ppm do not cause greater effects on infectivity than short exposures, due
 8                   to defense parameters  becoming reestablished with prolonged exposures, although
 9                   chronic exposure has been shown to slow alveolar clearance. No detrimental effects were
10                   seen with a 120-day exposure to 0.5 ppm O3 on acute lung injury from influenza virus
11                   administered immediately before O3 exposure started. However, O3 was shown to
12                   increase the severity of postinfluenzal alveolitis and lung parenchymal changes (Jakab
13                   and Bassett. 1990). Little new evidence has become available to address the effects of
14                   long-term exposure on host defense mechanisms. However, a recent study by Maniar-
15                   Hew et al. (2011) demonstrated that the immune system of infant rhesus monkeys
16                   episodically exposed to 0.5 ppm O3 for 5 months1 appeared to be altered in ways that
17                   could diminish host defenses. Reduced numbers of circulating leukocytes were observed,
18                   particularly polymorphonuclear leukocytes (PMNs) and lymphocytes, which were
19                   decreased in the  blood and airways  (bronchoalveolar lavage). These changes did not
20                   persist at 1 year of age (6 months postexposure); rather, increased numbers of monocytes
21                   were observed at that time point. Challenge with LPS, a bacterial ligand that activates
22                   monocytes and other innate immune cells, elicited lower responses in O3-exposed
23                   animals even though the relevant reactive cell population was increased. This was
24                   observed in both an in vivo inhalation challenge and an ex vivo challenge of peripheral
25                   blood mononuclear cells. Thus a decreased ability to respond to pathogenic signals was
26                   observed six months after O3 exposure ceased, in both the lungs and periphery.
            7.2.7   Respiratory Mortality

27                   A limited number of epidemiologic studies have assessed the relationship between long-
28                   term exposure to O3 and mortality. The 2006 O3 AQCD concluded that an insufficient
29                   amount of evidence existed "to suggest a causal relationship between chronic O3
30                   exposure and increased risk for mortality in humans" (U.S. EPA. 2006b). Though total
        1 Exposure protocol is described above in Section 7.2.3.1 for Fanucchi et al. (2006)
      Draft - Do Not Cite or Quote                      7-27                                September 2011

-------
 1                   and cardio-pulmonary mortality were considered in these studies, respiratory mortality
 2                   was not specifically considered. In the most recent follow-up analysis of the ACS cohort
 3                   (Jerrett et al.. 2009). cardiopulmonary deaths were subdivided into respiratory and
 4                   cardiovascular, separately, as opposed to combined in the Pope et al. (2002) work. A 10-
 5                   ppb increment in exposure to O3 elevated the risk of death from respiratory causes and
 6                   this effect was robust to the inclusion of PM2 5. The association between increased O3
 7                   concentrations and increased risk of death from respiratory causes was insensitive to the
 8                   use of a random-effects survival model allowing for spatial clustering within the
 9                   metropolitan area and state of residence, and to adjustment for several ecologic variables
10                   considered individually. Additionally, a recent study (Zanobetti and Schwartz. In Press)
11                   observed an association between long-term exposure to O3 and elevated risk of mortality
12                   among Medicare enrollees that had previously experienced an emergency hospital
13                   admission due to COPD.
             7.2.8   Summary and Causal Determination

14                   The epidemiologic studies reviewed in the 2006 O3 AQCD detected no associations
15                   between long-term (annual) O3 exposures and asthma-related symptoms, asthma
16                   prevalence, or allergy to common aeroallergens among children after controlling for
17                   covariates. Little evidence was available to relate long-term exposure to current ambient
18                   O3 concentrations to deficits in the growth rate of lung function in children. Additionally,
19                   limited evidence was available evaluating the relationship between long-term O3 levels
20                   and pulmonary inflammation and other endpoints. From toxicological studies, it appeared
21                   that O3-induced inflammation tapered off during long-term exposures, but that
22                   hyperplastic and fibrotic changes remained elevated and in some cases even worsened
23                   after a postexposure period in clean air. Episodic exposures were also known to cause
24                   more severe pulmonary morphologic changes than continuous exposure (U.S. EPA.
25                   2006b).

26                   The new epidemiologic evidence base consists of studies using a variety of designs and
27                   analysis methods evaluating the relationship between long-term annual measures of
28                   exposure to ambient  O3 and measures of respiratory morbidity conducted by different
29                   research groups in different locations. See Table 7-2 for O3 concentrations associated
30                   with selected studies. The positive results from various designs and locations support an
31                   association between long-term O3 concentrations and respiratory morbidity.

32                   New studies examined the relationship between long-term O3 exposure and new onset
33                   asthma in children. Studies have provided evidence for a relationship between different
34                   genetic variants (HMOX, GST, ARG) that, in combination with O3 exposure, are related
      Draft - Do Not Cite or Quote                       7-28                                September 2011

-------
 1
 2
 3
 4
 5
 6
to new onset asthma (Islam et al., 2009; Salam et al., 2009; Islam etaL 2008). These
studies involve two separate cohorts in 12 California communities of the CHS. These
prospective cohort studies represent strong evidence because they are methodologically
rigorous epidemiology studies. The stratified analysis for the two independent fourth-
grade cohorts of the study population recruited in 1993 and 1996 yielded consistent
results and provided replication in independent groups of children.
      Table 7-2      Summary of selected key new studies examining annual ozone
                      exposure and respiratory health effects
Study; Health Effect; Location
Akinbami et al. (2010): current asthma
United States
Hwang et al. (2005): prevalence of asthma
Taiwan
Islam et al.(2008): new-onset asthma;
CHS
Islam et al. (2009): new-onset asthma; CHS
Salam et al. (2009); childhood onset asthma; CHS
Lin et al. (2008b): first asthma hospital admission;
New York State - 10 regions
Moore et al. (2008): asthma
hospital admissions; South Coast Basin
Meng et al. (2010):
asthma ED visits or hospitalizations;
San Joaquin Valley, CA
Lee et al. (2009b):
bronchitic symptoms in asthmatic children; CHS
Rage et al. (2009b):
asthma severity; five French cities
Jacquemin et al. (In Press):
asthma control in adults; five French cities
Wentenetal. (2009):
respiratory school absence, U.S.
Annual Mean Os Concentration (ppb)
12 month median 39.8
Mean 23. 14
55.2 high vs. 38.4 low communities
10:00 a.m. to 6:00 p.m.
55.2 high vs. 38.4 low communities
10:00 a.m. to 6:00 p.m.
O3 greater than or less than 50 ppb
Range of mean O3 concentrations over the
10 New York Regions 37.51 to 47.78
Median 87.8 ppb
Median 30.3 ppb
Above and below 50 ppb
Mean 30 ppb
Median 46.9 ppb;
Median 46.9 ppb; 10:00 a.m. - 6:00 p.m.
O3 Range
(PPb)
Percent! les
IQR35.9to
43.7
Range 18.65 to
31.17
See left
See left
See left
See left
Range 28.6 to
199.9
25-75% range
27.1 to 34.0
See left
25th-75th
21-36
25th-75th
41-52
Min-Max
27.6-65.3
 9
10
11
12
13
14
15
Studies using a cross-sectional design provide support for a relationship between long-
term O3 exposure and health effects in asthmatics. A long-term O3 exposure study relates
bronchitic symptoms to TNF-308 genotype asthmatic children with ambient O3 exposure
in the CHS (Lee et al., 2009b). A study in five French cities reports effects on asthma
severity related to long-term O3 exposure (Rage etal.. 2009a). A follow-up study of this
cohort (Jacquemin et al.. In Press) supports an effect of long-term O3-sum exposure on
asthma control in adulthood in subjects with pre-existing asthma. Akinbami et al. (2010)
and Hwang et al. (2005) provides further evidence relating O3 exposures and the risk of
asthma.  For the respiratory health of a cohort based on the general U.S. population, risk
      Draft - Do Not Cite or Quote
                               7-29
September 2011

-------
 1                   of respiratory-related school absences was elevated for children with the CAT and MPO
 2                   variant genes related to communities with high ambient O3 levels (Wenten et al.. 2009).

 3                   Chronic O3 exposure was related to first childhood asthma hospital admissions in a
 4                   positive concentration-response relationship in a New York State birth cohort (Lin et al..
 5                   2008b). A separate hospitalization cross-sectional study in San Joaquin Valley in
 6                   California reports similar findings (Meng et al., 2010). Another study relates asthma
 7                   hospital admissions to quarterly average O3 in the  South Coast Air Basin of California
 8                   (Moore et al.. 2008).

 9                   Information from toxicological studies indicates that long term exposure to O3 during
10                   gestation or development can result in irreversible  morphological changes in the lung,
11                   which in turn can influence pulmonary function. Studies by Plopper and colleagues have
12                   demonstrated changes in pulmonary function and airway morphology in adult and infant
13                   nonhuman primates repeatedly exposed to environmentally relevant concentrations of O3
14                   (Tanucchi et al.. 2006: Joad et al.. 2006: Schelegle  et al.. 2003: Harkema et al.. 1987b).
15                   This nonhuman primate evidence of an  O3-induced change in airway responsiveness
16                   supports the biologic plausibility of long term exposure to O3 contributing to the adverse
17                   effects of asthma in children. Results from epidemiologic studies examining long-term
18                   O3 exposure and pulmonary function effects are inconclusive with some new studies
19                   relating effects at higher exposure levels. The results from the CHS described in the 2006
20                   O3 AQCD remain the definitive line of evidence. Other cross-sectional studies provide
21                   mixed results.
      Draft - Do Not Cite or Quote                       7-30                                 September 2011

-------
1
2
3
4

5
6
Table 7-3 Studies providing evidence concerning potential confounding by
PM for available endpoints
Study
Exposure
Endpomt
Hwang et al. (2005)
	 10ppbO3
Asthma risk in children
Jacquemin et al. (In
presS) IQR 25-38 ppb O3
, , • , „ summer
Asthma control in adults
Lin et al. (2008b)
Asthma admissions in IQR 2.5%
children
Akinbami et al. (2010)
Asthma prevalence in IQR 35.9-43.7 ppb
children
Lee et al. (2009b)
Bronchitic symptoms High O3 >50 ppb
asthmatics
Rage et al. (2009a)
Asthma severity in IQR 28.5-33.9 ppb
adults
Meng et al. (2007)
	 1 ppm
Asthma control
Meng et al. (201 0)
Asthma ED visits, 10 ppb
Hospitalization
Karretal. (2007)
Bronchiolitis 10 ppb
Hospitalization
Rojas-Martinez et al.
(2007)
i-i-w , i % ™ r -t 11. 3 ppb IQR
FEN/! (mL) Deficit ^
Girls
Parker et al. (2009)
10 ppb
Respiratory allergy
Single
Pollutant O3
1.138
(1.001, 1.293
1.69
(1 .22, 2.34)
1.16
(1.15, 1.17)
1.56
(1.15,2.10)
1.42
(0.75, 2.70)
2.53
(1 .69, 3.79)
1.70
(0.91,3.18)
1.49
(1.05,2.11)
0.92
(0.88, 0.96)
-24
(-30, -19)
1.24
(1.15, 1.34)
Single
Pollutant PM
0.934
(0.909, 0.960)
1.33
(1 .06, 1 .67)
NA
PM2.5
1.43
(0.98,2.10)
NA
NA
PM10 2.06
(1.17,3.61)
women
PM10
1.29
(0.99, 1 .69)
1.09
(1.04, 1.14)
PM10
IQR
36.4 ug/m3
-29(-36, -21)
1.23
(1 .04, 1 .46)
O3 with PM
PM10
1.253(1.089,1.442)
PM10
1.50(1.07,2.11)
Air Quality Index
1.24(1.23, 1.25)
Adjusted for
S02,PM2.5,PM10
1.86(1.02-3.40)
Adjusted for PM25,
PM10
1.36(0.91-2.02)
No substantial
differences
PM10, PM2.5
No PM data
Three pollutant (O3,
NO2, SO2)
2.74(1.68,4.48)
Did not differ
Did not differ
PM2.5
1.02(0.94, 1.10)
-17 (-23, -12)
Multi-pollutant
1.18(1.09,1.27)
PM with O3
0.925
(0.899, 0.952)
1.28
(1 .06, 1 .55)
NA
PM2.5
1.24(0.70-2.21)
PM2.5
1.26(0.80-1.98)
NA
NA
NA
NA
1.09
(1.03, 1.15)
-24 (-31 ,
-16)
1.29(1.07,1.56)
       The highest quartile is shown for all results.
       NA= not available
Several studies (see Table 7-3) provide results from studies that adjusted for potential
confounders, presenting results for both O3 and PM (single and multipollutant models) as
well as other pollutants where PM effects were not provided. As shown in the table, O3
associations are generally robust to adjustment for potential confounding by PM.

The 2006 O3 AQCD concluded that the extensive human clinical and animal
toxicological evidence, together with the limited epidemiologic evidence available,
     Draft - Do Not Cite or Quote
                                7-31
September 2011

-------
 1                   suggests a causal role for short-term O3 exposure in inflammatory responses in the
 2                   airways. Though the majority of recent studies focus on short-term exposures, several
 3                   epidemiologic and toxicological studies of long-term exposure add to observations of O3 -
 4                   induced inflammation and injury. Toxicological studies in rodents and nonhuman
 5                   primates indicate that chronic O3 exposure causes structural changes in the respiratory
 6                   tract, and simulated seasonal exposure studies suggest that such exposures might have
 7                   cumulative impacts. The strongest epidemiologic evidence for a relationship between
 8                   long-term O3 exposure and respiratory morbidity is provided by new studies that
 9                   demonstrate associations between long-term measures of O3 exposure and new-onset
10                   asthma in children and increased respiratory symptom effects in asthmatics. While there
11                   are currently a limited number of studies in this data base, the U.S. multi-community
12                   prospective cohort studies are methodologically rigorous epidemiologic studies. Asthma
13                   risk is related to complex relationships between genetic variability, environmental O3
14                   exposure, and behavior. The genes, evaluated in these studies, are both key candidates in
15                   the oxidative stress pathway and have been shown to play an important role in asthma
16                   development. Reduced risk for asthma development is reported in some studies in
17                   children living in low- O3  communities. Mean O3 concentrations in the studies (10:00
18                   a.m. to 6:00 p.m.) ranged from 28.6 to 45.5 ppb in low O3 communities
19                   (mean = 38.4 ppb) and from 46.5 to 64.9 ppb in high O3 communities (mean = 55.2 ppb).
20                   These CHS multi-community studies form a foundation for the evidence base in which
21                   findings for several genes  indicate the breath of the evidence across different gene
22                   variants. The several other studies with different  designs, analysis, locations and
23                   researchers provide a cumulative collective body of evidence informing these
24                   relationships. The other studies in the new data base provide coherent evidence for long-
25                   term O3 exposure and respiratory morbidity effects such as first asthma hospitalization
26                   and respiratory symptoms in asthmatics. Studies  considering other pollutants provide data
27                   suggesting that the effects related to O3 are independent from potential effects of the
28                   other pollutants. Some studies provide evidence for a positive concentration-response
29                   relationship. Short-term studies provide supportive evidence with increases in respiratory
30                   symptoms and asthma medication use, hospital admissions and ED visits for all
31                   respiratory outcomes and asthma, and decrements in lung function in children. The above
32                   discussion of the recent epidemiologic and toxicological data base provides a compelling
33                   case to support the hypothesis that a relationship  exists between long-term exposure to
34                   ambient O3 and measures  of respiratory morbidity. The 2006 O3 AQCD concluded the
35                   evidence was suggestive but inconclusive at that  time. The new epidemiological data
36                   base, combined with toxicological studies in rodents and nonhuman primates,
37                   provides biologically plausible evidence that there is likely to be causal
38                   relationship between long-term exposure to O3 and respiratory morbidity.
      Draft - Do Not Cite or Quote                       7-32                                 September 2011

-------
          7.3    Cardiovascular Effects
            7.3.1   Cardiovascular Disease
                    7.3.1.1    Cardiovascular Epidemiology

 1                  Long-term exposure to O3 and its effects on cardiovascular morbidity were not
 2                  considered in the 2006 O3 AQCD. However, recent studies have assessed the chronic
 3                  effects of O3 exposure on cardiovascular morbidity (Chuang etal.. 2011; Forbes et al.
 4                  2009a; Chen etal.. 2007). The association between O3 exposure and markers of lipid
 5                  peroxidation and antioxidant capacity was examined among 120 nonsmoking healthy
 6                  college students, aged 18-22 years, from the University of California, Berkeley (Feb-Jun
 7                  2002) (Chen et al.. 2007). By design, students were chosen from geographic areas so they
 8                  had experienced different levels of O3 over their lifetimes and during recent summer
 9                  vacation in either greater Los Angeles (LA) or the San Francisco Bay Area (SF). A
10                  marker of lipid peroxidation, 8-isoprostane (8-iso-PGF) in plasma, was assessed. This
11                  marker is formed continuously under normal physiological conditions but has been found
12                  at elevated concentrations in response to environmental exposures. A marker of overall
13                  antioxidant capacity, ferric reducing ability of plasma (FRAP), was also measured. The
14                  lifetime O3 exposure estimates (estimated monthly average) did not show much overlap
15                  between the two geographic areas [median (range): LA, 42.9 ppb (28.5-65.3); SF,  26.9
16                  ppb (17.6-33.5)]. Estimated lifetime O3 exposure was related to 8-iso-PGF  [(3 = 0.025
17                  (pg/mL)/8-h ppb O3, p = 0.0007]. For the 17-ppb cumulative lifetime O3 exposure
18                  difference between LA and SF participants, there was a 17.41-pg/mL (95% CI: 15.43,
19                  19.39) increase in 8-iso-PGF. No evidence of association was observed between lifetime
20                  O3 exposure and FRAP [(3 = -2.21 (pg/mL)/8-h ppb O3, p = 0.45]. The authors note that
21                  O3 was highly correlated with PM10_2 5 and NO2 in this study population; however, their
22                  inclusion in the  O3 models did not substantially modify the magnitude of the associations
23                  with O3. Because the lifetime exposure results were supported by shorter-term exposure
24                  results from analyses considering O3 concentrations up to 30 days prior to sampling, the
25                  authors conclude that persistent exposure to O3 can lead to  sustained oxidative stress and
26                  increased lipid peroxidation. However, because there was not much overlap in lifetime
27                  O3 exposure estimates between LA and SF, it is possible that the risk estimates involving
28                  the lifetime O3 exposures could be confounded by unmeasured factors related to other
29                  differences between  the two cities.

30                  Forbes et al. (2009a) used the annual average exposures to assess the relationship
31                  between chronic ambient air pollution and levels of fibrinogen and C-reactive protein
      Draft - Do Not Cite or Quote                      7-33                                September 2011

-------
 1                   (CRP) in a cross-sectional study conducted in England. Data were collected from the
 2                   Health Survey of England for 1994, 1998, and 2003. The sampling strategy was designed
 3                   to obtain a representative sample of the English population; however, due to small group
 4                   sizes, only data from white ethnic groups were analyzed. For analyses, the annual
 5                   concentrations of O3 were averaged for the year of data collection and the previous year
 6                   with the exception of 1994 (because pollutant data were not available for 1993). Median
 7                   O3 concentrations were 26.7 ppb, 25.4 ppb, and 28 ppb for 1994, 1998, and 2003,
 8                   respectively. Year specific adjusted effect estimates were created and combined in a
 9                   meta-analysis. No evidence of association was observed for O3 and levels of fibrinogen
10                   or CRP (e.g., the combined estimates for the percent change in fibrinogen and CRP for a
11                   10 ppb increase  in O3 were -0.28 [95% CI: -2.43, 1.92] and -3.05 [95% CI: -16.10,
12                   12.02], respectively).

13                   A study was performed in Taiwan to examine the association between long-term O3
14                   concentrations and blood pressure and blood markers using the Social Environment and
15                   Biomarkers of Aging  Study (SEBAS) (Chuang et al.. 2011). Individuals included in the
16                   study were 54 years of age and older. The mean annual O3 concentration during the study
17                   period was 22.95 ppb (SD 6.76 ppb). Positive associations were observed between O3
18                   concentrations and both systolic and diastolic blood pressure [changes in systolic and
19                   diastolic blood pressure were 21.51mmHg (95% CI: 16.90, 26.13) and 20.56 mmHg
20                   (95% CI: 18.14, 22.97) per 8.95 ppb increase in O3, respectively). Increased O3
21                   concentrations were also associated with increased levels of total cholesterol, fasting
22                   glucose, hemoglobin Ale, and neutrophils. No associations were observed between O3
23                   concentrations and triglyceride and IL-6 levels. The observed associations were reduced
24                   when other pollutants were added to the models. Further research will be important for
25                   understanding the effects, if any, of chronic O3 exposure on cardiovascular morbidity
26                   risk.
                     7.3.1.2    Cardiovascular Toxicology

27                   Three new studies have investigated the cardiovascular effects of long-term exposure to
28                   O3 in animal models (See Table 7-4 for study details). In addition to the short-term
29                   exposure effects described in Section 6.3.3, a recent study found that O3 exposure in
30                   genetically hyperlipidemic mice enhanced aortic atherosclerotic lesion area compared to
31                   air exposed controls (Chuang et al.. 2009). Chuang et al. (2009) not only provided
32                   evidence for increased atherogenesis in susceptible mice, but also reported an elevated
33                   vascular inflammatory and redox state in wild-type mice and infant primates
34                   (Section 6.3.3.2). This study is compelling in that it identifies biochemical and cellular
3 5                   events responsible for transducing the airway epithelial reactions  of O3 into

      Draft - Do Not Cite or Quote                       7-34                                 September 2011

-------
 1                   proinflammatory responses that are apparent in the extrapulmonary vasculature (Cole and
 2                   Freeman. 2009).

 3                   Another recent study provides further evidence for increased vascular inflammation and
 4                   oxidation and long term effects in the extrapulmonary space. Rats episodically exposed to
 5                   O3 for 16 weeks presented marked increases in gene expression of biomarkers of
 6                   oxidative stress, thrombosis, vasoconstriction, and proteolysis (Kodavanti et al., 2011).
 7                   Ozone exposure upregulated aortic mRNA expression of heme oxygenase-1 (HO-1),
 8                   tissue plasminogen activator (tPA), plasminogen activator inhibitor-1 (PAI-1), von
 9                   Willebrand factor (vWf), thrombomodulin, endothelial nitric oxide  synthase (eNOS),
10                   endothelin-1 (ET-1), matrix metalloprotease-2 (MMP-2), matrix metalloprotease-3
11                   (MMP-3), and tissue inhibitor of matrix metalloprotease-2 (TIMP-2). In addition, O3
12                   exposure depleted some cardiac mitochondrial phospholipid fatty acids (C16:0 and
13                   CIS: 1), which may be the result of oxidative modifications. The authors speculate that
14                   oxidatively modified lipids and proteins produced in the lung and heart promote vascular
15                   pathology through activation of lectin-like oxidized-low density lipoprotein receptor-1
16                   (LOX-1). Activated LOX-1 induces expression of a number of the biomarkers induced by
17                   O3 exposure and is considered pro-atherogenic. Both LOX-1 mRNA and protein were
18                   increased in mouse aorta after O3  exposure. This study provides a possible pathway and
19                   further support to the observed O3 induced atherosclerosis.

20                   Vascular occlusion resulting from atherosclerosis can block blood flow through vessels
21                   causing ischemia. The restoration of blood flow or reperfusion can cause injury to the
22                   tissue from subsequent inflammation and oxidative damage. Ozone exposure enhanced
23                   the sensitivity to myocardial ischemia-reperfusion (I/R) injury in rats while increasing
24                   oxidative stress levels and pro-inflammatory mediators and decreasing production of anti-
25                   inflammatory proteins (Perepu et al.. 2010). Both long- and short-term O3 exposure
26                   decreased the left ventricular developed pressure, rate of change of pressure
27                   development, and rate of change of pressure decay and increased left ventricular end
28                   diastolic pressure in isolated perfused hearts (Section 6.3.3.2 for short-term exposure
29                   discussion). In this ex vivo heart model, O3 induced oxidative stress by decreasing SOD
30                   enzyme activity and  increasing malondialdehyde levels.  Ozone also elicited a
31                   proinflammatory state evident by  an increase in TNF-a and a decrease in the anti-
32                   inflammatory cytokine IL-10. The authors conclude that O3 exposure will result in a
33                   greater I/R injury.
      Draft - Do Not Cite or Quote                       7-35                                September 2011

-------
Table 7-4
Study
Chuang et al. (2009)
Kodavanti et al.
(2011)
Perepu et al. (201 0)
Characterization of study details for Section 7.3.1.2
Model
Mice; ApoE-/-; M;
6 weeks
Rat; Wistar; M;
10-1 2 weeks
Rat; Sprague-Dawley;
Weight: 50-75 g
_ . . Exposure
°3  Duration
n ,- 8 wks, 5 days/week,
u'° 8 h/day
n . 16 wks, 1 day/week,
u'4 5 h/day
0.8 56 days, 8 h/day
Effects
Enhanced aortic atherosclerotic lesion
area compared to air controls.
Increased vascular inflammation and
oxidative stress, possibly through
activation of LOX-1 signaling.
Enhanced the sensitivity to myocardial
I/R injury while increasing oxidative
stress and pro-inflammatory mediators
and decreasing production of anti-
inflammatory proteins.
        No previous studies investigated cardiovascular effects from long-term exposure to O3.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
             7.3.2   Cardiac Mortality

                     A limited number of epidemiologic studies have assessed the relationship between long-
                     term exposure to O3 and mortality. The 2006 O3 AQCD concluded that an insufficient
                     amount of evidence existed "to suggest a causal relationship between chronic O3
                     exposure and increased risk for mortality in humans" (U.S. EPA. 2006b). Though total
                     and cardio-pulmonary mortality were considered in these studies, cardiovascular
                     mortality was not specifically considered. In the most recent follow-up analysis of the
                     ACS cohort (Jerrett et al.. 2009). cardiopulmonary deaths were subdivided into
                     respiratory and cardiovascular, separately, as opposed to combined in the Pope et al.
                     (2002) work. A 10-ppb increment in exposure to O3 elevated the risk of death from the
                     cardiopulmonary, cardiovascular, and ischemic heart disease. Inclusion of PM25 as a
                     copollutant attenuated the association with exposure to O3 for all of the cardiovascular
                     endpoints to become null. Additionally, a recent study (Zanobetti and Schwartz. In Press)
                     observed an association between long-term exposure to O3 and elevated risk of mortality
                     among Medicare enrollees that had previously experienced an emergency hospital
                     admission due to congestive heart failure (CHF) or myocardial infarction (MI).
16
17
18
19
20
21
            7.3.3  Summary and Causal Determination

                    Previous AQCDs did not address the cardiovascular effects of long-term O3 exposure due
                    to limited data availability. The evidence remains limited; however the emerging data is
                    supportive of a role for O3 in chronic cardiovascular diseases. Few epidemiologic studies
                    have investigated cardiovascular morbidity after long-term O3 exposure, and the majority
                    only assessed cardiovascular disease related biomarkers. A study on O3 and
                    cardiovascular mortality reported no association after adjustment for PM2 5 levels.
      Draft - Do Not Cite or Quote
                                                    7-36
September 2011

-------
 1                  Further epidemiologic studies on cardiovascular morbidity and mortality after long-term
 2                  exposure have not been published.

 3                  lexicological evidence on long-term O3 exposure is also limited but three strong
 4                  toxicological studies have been published since the previous AQCD. These studies
 5                  provide evidence for O3 enhanced atherosclerosis and I/R injury, corresponding with
 6                  development of a systemic oxidative, proinflammatory environment. Further discussion
 7                  of the mechanisms that may lead to cardiovascular effects can be found in Section 5.3.8.
 8                  Although questions exist for how O3 inhalation causes systemic effects, a recent study
 9                  proposes a mechanism for development of vascular pathology that involves activation of
10                  LOX-1 by O3 oxidized lipids and proteins. This activation may also be responsible for O3
11                  induced changes in genes involved in proteolysis, thrombosis, and vasoconstriction.
12                  Taking into consideration the findings of toxicological studies, and the emerging
13                  evidence from epidemiologic studies, the generally limited body of evidence is
14                  suggestive of a causal relationship between long-term exposures to O3 and
15                  cardiovascular effects.
          7.4    Reproductive and  Developmental Effects

16                  Although the body of literature is growing, the research focusing on adverse birth
17                  outcomes is small. Among these studies, various measures of birth weight and fetal
18                  growth, such as low birth weight (LEW), small for gestational age (SGA), and
19                  intrauterine growth restriction (IUGR), and preterm birth (<37-week gestation; [PTB])
20                  have received more attention in air pollution research, while congenital malformations
21                  are less studied. There are also new studies on reproductive and developmental effects.

22                  Infants and fetal development processes may be particularly susceptible to O3-induced
23                  health effects, and although the physical mechanisms are not fully understood, several
24                  hypotheses have been proposed; these include: oxidative stress, systemic inflammation,
25                  vascular dysfunction and impaired immune function (Section 5.3). Study of these
26                  outcomes can be difficult given the need for detailed exposure data and potential
27                  residential movement of mothers during pregnancy. Air pollution epidemiologic studies
28                  reviewed in the 2006 O3 AQCD examined impacts on birth-related endpoints, including
29                  intrauterine, perinatal, postneonatal, and infant deaths; premature births; intrauterine
30                  growth retardation; very low birth weight (weight < 1,5 00 grams) and low birth weight
31                  (weight <2,500 grams); and birth defects. However, in the limited number of studies that
32                  investigated O3, no associations were found between O3 and birth outcomes, with the
33                  possible exception of birth defects.
      Draft - Do Not Cite or Quote                      7-37                                September 2011

-------
 1                   Several recent articles have reviewed methodological issues relating to the study of
 2                   outdoor air pollution and adverse birth outcomes (Chen et al.. 2010a: Woodruff etal..
 3                   2009; Ritz and Wilhelm. 2008; Slamaet al.. 2008). Some of the key challenges to
 4                   interpretation of these study results include the difficulty in assessing exposure as most
 5                   studies use existing monitoring networks to estimate individual exposure to ambient air
 6                   pollution; the inability to control for potential confounders such as other risk factors that
 7                   affect birth outcomes (e.g., smoking); evaluating the exposure window (e.g., trimester) of
 8                   importance; and limited evidence on the physiological mechanism of these effects (Ritz
 9                   and Wilhelm. 2008; Slama et al.. 2008). Recently, an international collaboration was
10                   formed to better understand the relationships between air pollution and adverse birth
11                   outcomes and to examine some of these methodological issues through standardized
12                   parallel analyses in datasets from different countries (Woodruff et al.. 2010). Initial
13                   results from this collaboration have examined PM and birth weight (Parker et al., 2011);
14                   work on O3 has not yet been performed. Although early animal studies (Kavlock et al..
15                   1980) found that exposure to O3  in the late gestation of pregnancy in rats led to some
16                   abnormal reproductive performances for neonates, to date  human studies have reported
17                   inconsistent results for the association of ambient O3 on birth outcomes.
             7.4.1   Effects on Sperm

18                   A limited amount of research has been conducted to examine the association between air
19                   pollution and male reproductive outcomes, specifically semen quality. To date, the
20                   epidemiologic studies have considered various exposure durations before semen
21                   collection that encompass either the entire period of spermatogenesis (i.e., 90 days) or
22                   key periods of sperm development that correspond to epididymal storage, development of
23                   sperm motility, and spermatogenesis. In an analysis conducted as part of the Teplice
24                   Program, 18-year-old men residing in the heavily polluted district of Teplice in the Czech
25                   Republic were found to be at greater risk of having abnormalities in sperm morphology
26                   and chromatin integrity than men of similar age residing in Prachatice, a less polluted
27                   district (Selevan et al.. 2000; Sram etal.. 1999). A follow-up longitudinal study
28                   conducted on a subset of the same men from Teplice revealed associations between total
29                   episodic air pollution and  abnormalities in sperm chromatin  (Rubes et al.. 2005). A
30                   limitation of these studies is that they did not identify specific pollutants and their
31                   concentrations.

32                   More recent epidemiologic studies conducted in the U.S. have also reported associations
33                   between ambient air pollution and sperm quality for individual air pollutants, including
34                   O3 and PM25. In a repeated measures study in Los Angeles, CA, Sokol et al. (2006)
35                   reported a reduction in average  sperm concentration during three exposure windows (0-9,
      Draft - Do Not Cite or Quote                       7-38                                September 2011

-------
 1                   10-14, and 70-90 days before semen collection) associated with high ambient levels of
 2                   O3 in healthy sperm donors. This effect persisted under a joint additive model for O3,
 3                   CO, NO2 and PM10. The authors did not detect a reduction in sperm count. Hansen et al.
 4                   (2010) investigated the effect of exposure to O3 and PM2 5 (using the same exposure
 5                   windows used by Sokol et al. (2006) on sperm quality in three southeastern counties
 6                   (Wake County, NC; Shelby County, TN; Galveston County, TX). Outcomes included
 7                   sperm concentration and count, morphology, DNA integrity and chromatin maturity.
 8                   Overall, the authors found both protective and adverse effects, although some results
 9                   suggested adverse effects on sperm concentration, count and morphology.

10                   The biological mechanisms linking ambient air pollution to decreased sperm quality have
11                   yet to be determined, though O3-induced oxidative stress, inflammatory reactions, and
12                   the induction of the formation of circulating toxic species have been suggested as
13                   possible mechanisms (see Section 5. 3.8). Decremental effects on testicular morphology
14                   have been demonstrated in toxicological studies with histological evidence of O3-induced
15                   depletion of germ cells in testicular tissue and decreased seminiferous tubule epithelial
16                   layer. Jedlinska-Krakowska et al. (2006) demonstrated histopathological evidence of
17                   impaired spermatogenesis (round spermatids/ spermatocytes, giant spermatid cells, and
18                   focal epithelial desquamation with denudation to the basement membrane). The exposure
19                   protocol used five month old adult rats exposed to O3 as adults (0.5 ppm, 5 h/day for
20                   50 days). This degeneration could be rescued by vitamin E administration, indicating an
21                   antioxidant effect. Vitamin C administration had no effect at low doses of ascorbic acid
22                   and exacerbated the O3-dependent damage at high doses, as would be expected as
23                   vitamin C can be a radical generator instead of an antioxidant at higher doses. In
24                   summary, this study provided toxicological evidence of impaired spermatogenesis with
25                   O3 exposure that was rescued with certain antioxidant supplementation.

26                   Overall, there is limited epidemiologic evidence for an association with O3 concentration
27                   and decreased sperm concentration. A recent toxicological study provides limited
28                   evidence for a possible biological mechanism (histopathology showing impaired
29                   spermatogenesis) for such an association.
             7.4.2   Effects on Reproduction

30                   Evidence suggests that exposure to air pollutants during pregnancy is associated with
31                   adverse birth outcomes, which has been attributed to the increased susceptibility of the
32                   fetus due to physiologic immaturity. Gametes (i.e., ova and sperm) may be even more
33                   susceptible, especially outside of the human body, as occurs with assisted reproduction.
34                   Smokers require twice the number of in vitro fertilization (IVF) attempts to conceive as
      Draft - Do Not Cite or Quote                       7-39                                 September 2011

-------
 1                   non-smokers (Feichtinger et al., 1997). suggesting that a preconception exposure can be
 2                   harmful to pregnancy. A recent study used an established national-scale, log-normal
 3                   kriging method to spatially estimate daily mean concentrations of criteria pollutants at
 4                   addresses of women undergoing their first IVF cycle and at their IVF labs from 2000 to
 5                   2007 in the northeastern U.S. (Legro et al., 2010). Increasing O3 concentration at the
 6                   patient's address was significantly associated with an increased chance of live birth
 7                   during ovulation induction  (OR=1.13, [95% CI: 1.05, 1.22] per 10 ppb increase), but with
 8                   decreased odds of live birth when exposed from embryo transfer to live birth (OR=0.79,
 9                   [95% CI: 0.69, 0.90] per 10 ppb increase). After controlling for NO2 in a copollutant
10                   model, however, O3 was no longer significantly associated with IVF failure. The results
11                   of this study suggest that exposure to O3 during ovulation was beneficial (perhaps due to
12                   early conditioning to O3), whereas later exposure to O3 (e.g., during gestation) was
13                   detrimental, and reduced the likelihood of a live birth.

14                   In most toxicological studies, reproductive success appears to be unaffected by O3
15                   exposure. Nonetheless, one study has reported that 25% of the BALB/c mouse dams in
16                   the highest O3 exposure group (1.2 ppm, GD9-18) did not complete a successful
17                   pregnancy, a significant reduction (Sharkhuu et al.. 2011). Ozone administration
18                   (continuous 0.4, 0.8 or 1.2 ppm O3) to CD-I  mouse dams during the majority of
19                   pregnancy (PD7-17, which excludes the pre-implantation period), led to no adverse
20                   effects on reproductive success (proportion of successful  pregnancies, litter size, sex
21                   ratio, frequency of still birth, or neonatal mortality) (Bignami et al.. 1994). There was a
22                   nearly statistically significant increase in pregnancy duration (0.8 and 1.2 ppm O3).
23                   Initially, dam body weight  (0.8 and  1.2 ppm), water consumption (0.4, 0.8 and 1.2 ppm
24                   O3) and food consumption  (0.4, 0.8 and 1.2 ppm) during pregnancy were decreased with
25                   O3 exposure but these deficits dissipated a week or two after the initial exposure
26                   (Bignami et al..  1994). The anorexigenic effect of O3 exposure on the pregnant dam
27                   appears to dissipate with time; the dams seem to adapt to the O3 exposure. In males, data
28                   exist showing morphological evidence of altered spermatogenesis in O3 exposed animals
29                   (Jedlinska-Krakowska et al.. 2006).  Some evidence suggests that O3 may affect
30                   reproductive success when  combined with other chemicals. Kavlock et al. (1979) showed
31                   that O3 acted synergistically with sodium salicylate to increase the rate of pup resorptions
32                   after midgestational exposure (1.0 ppm O3, GD9-12). At  low doses of O3 exposure,
33                   toxicological studies show  reproductive effects to include a transient anorexigenic effect
34                   of O3 on gestational weight gain, and a synergistic effect of O3 on salicylate-induced pup
35                   resorptions; other fecundity, pregnancy and gestation related outcomes appear unaffected
36                   by O3 exposure.
      Draft - Do Not Cite or Quote                       7-40                                September 2011

-------
 1                   Collectively, there is very little epidemiologic evidence for the effect of O3 on
 2                   reproductive success, and the reproductive success in rats appears to be unaffected in
 3                   toxicological studies of O3 exposure.
             7.4.3   Birth Weight

 4                   With birth weight routinely collected in vital statistics and being a powerful predictor of
 5                   infant mortality, it is the most studied outcome within air pollution-birth outcome
 6                   research. Air pollution researchers have analyzed birth weight as a continuous variable
 7                   and/or as a dichotomized variable in the form of LEW (<2,500 g [5 Ibs, 8 oz]).

 8                   Birth weight is primarily determined by gestational age and intrauterine growth, but also
 9                   depends on maternal, placental and fetal factors as well as on environmental influences.
10                   In both developed and developing countries, LEW is the most important predictor for
11                   neonatal mortality and is a significant determinant of postneonatal mortality and
12                   morbidity. Studies report that infants who are smallest at birth have a higher incidence of
13                   diseases and disabilities, which continue into adulthood (Hack and Fanaroff. 1999).

14                   The strongest evidence for an effect of O3 on birth weight comes from the Children's
15                   Health Study conducted in southern California. In this study, Salam et al. (2005) report
16                   that maternal exposure to 24-h avg O3 concentrations averaged over the entire pregnancy
17                   was associated with reduced birth weight (39.3 g decrease [95% CI: -55.8, -22.8] in birth
18                   weight per 10 ppb and 8-h avg (19.2-g decrease [95% CI: -27.7, -10.7] in birth weight per
19                   10 ppb). This effect was stronger for concentrations averaged over the second  and third
20                   trimesters. PM10, NO2 and CO concentrations averaged over the entire pregnancy were
21                   not statistically significantly associated with birth weight, although CO concentrations in
22                   the first trimester and PM10 concentrations in the third trimester were associated with a
23                   decrease in birth weight. Additionally, the authors observed a concentration-response
24                   relationship  of birth weight with 24-h avg O3 concentrations averaged over the entire
25                   pregnancy that was clearest above the 30-ppb level (see Figure 7-4). Relative to the
26                   lowest decile of 24-h avg O3, estimates for the next 5 lowest deciles were approximately
27                   -40 g to -50  g, with no clear trend and with 95%  confidence bounds that included zero.
28                   The highest  four deciles of O3 exposure showed  an approximately linear decrease in birth
29                   weight, and  all four 95% CIs excluded zero, and  ranged from mean decreases of
30                   74 grams to  decreases of 148 grams.
      Draft - Do Not Cite or Quote                       7-41                                 September 2011

-------
                       50

                        0

                      -50
                 '5  -100
                     -150

                     -200

                     -250

c



3
C




(




3 f




> c


_,

)
c



•)
c



)
[



)
                                                                0
                                     20           30
                                           24-hr 03{ppb)
                                                                   40
50
 Source: Salam et al. (2005)
  Deficits are plotted against the decile-group-specific median O3 exposure. Error bars represent 95% CIs. Indicator variables for
each decile of O3 exposure (except the least-exposed group) were included in a mixed model.

Figure 7-4     Birthweight deficit by decile of 24-h avg O3 concentration averaged
                 over the entire pregnancy compared with the decile group with the
                 lowest Oz exposure.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
               Several additional studies conducted in the U.S. and Canada also investigated the
               association between ambient O3 concentrations and birth weight and report some weak
               evidence for an association. Morello-Frosch et al. (2010) estimated ambient O3
               concentrations throughout pregnancy and for each trimester in the neighborhoods of
               women who delivered term singleton births between 1996 and 2006 in California. A 10-
               ppb increase in O3 averaged across the entire pregnancy was associated with a 5.7-g
               decrease (95% CI: -6.6, -4.9) in birth weight when exposures were calculated using
               monitors within 10 km of the maternal address at date of birth. When the distance from
               the monitor was restricted to 3 km, the decrease in birth weight associated with a 10-ppb
               increase in O3 increased to 8.9 g (95% CI: -10.6, -7.1). These results persisted in
               copollutant models and in models that stratified by trimester of exposure, SES, and race.
               Darrow et al. (2011 a) did not observe an association with birth weight and O3
               concentrations during two exposure periods of interest (i.e., the first month and last
               trimester), but did find an association with reduced birth weight when examining the
               cumulative air pollution concentration during the entire pregnancy period. Additionally,
               they observed effect modification by race and ethnicity, such that associations between
               birth weight and third-trimester O3 concentrations were significantly stronger in
Draft - Do Not Cite or Quote
                                                    7-42
        September 2011

-------
 1                   Hispanics and non-Hispanic African Americans than in non-Hispanic whites. Chen et al.
 2                   (2002) used 8-h avg O3 concentrations to create exposure variables based on average
 3                   maternal exposure for each trimester. Ozone was not found to be related to birth weight
 4                   in single-pollutant models, though the O3 effect during the third trimester was borderline
 5                   statistically significant in a copollutant model with PM10.

 6                   Several studies found no association between ambient O3 concentrations and birth
 7                   weight. Wilhelm and  Ritz (2005) extended previous analyses of term LEW (Ritz et al..
 8                   2000; Ritz and Yu. 1999) to include the period 1994-2000. The authors examined varying
 9                   residential distances from monitoring stations to see if the distance affected risk
10                   estimation, exploring  the possibility that effect attenuation may result from local pollutant
11                   heterogeneity inadequately captured by ambient monitors. As in their previous studies,
12                   the authors observed associations  between elevated concentrations of CO and PM10 both
13                   early and late in pregnancy and risk of term LEW. After adjusting for CO and/or PMi0
14                   the authors did not observe associations between O3 and term LEW in any of their
15                   models.  Brauer et al. (2008) evaluated the impacts of air pollution (CO, NO2, NO, O3,
16                   SO2, PM2 5, PM10) on birth weight for the period  1999-2002 using spatiotemporal
17                   residential exposure metrics by month of pregnancy in Vancouver, BC. Quantitative
18                   results were not presented for the  association between O3 and LEW, though the authors
19                   observed associations that were largely protective. Dugandzic et al. (2006)  examined the
20                   association between LEW and ambient levels of air pollutants by trimester of exposure
21                   among a cohort of term singleton  births from 1988-2000. Though there was some
22                   indication of an  association with SO2 and PMi0, there were no effects for O3.

23                   Similarly, studies conducted in Australia, Latin America, and Asia report limited
24                   evidence for an association between ambient O3 and measures of birth weight. In Sydney,
25                   Australia, Mannes et al. (2005) found that O3 concentrations in the second trimester of
26                   pregnancy had small adverse effects on birth weight (7.5-g decrease; [95 % CI:  -13.8,
27                   1.2] per  10 ppb), although this effect disappeared when the analysis was limited to births
28                   with a maternal address within 5 km of a monitoring station (87.7-g increase; [95% CI:
29                   10.5, 164.9] per 10 ppb). Hansen et al. (2007) reported that trimester and monthly
30                   specific  exposures to all pollutants were  not statistically significantly associated with a
31                   reduction in birth weight in Brisbane, Australia. In Sao Paulo, Brazil, Gouveia et al.
32                   (2004) found that O3 exhibited a small inverse  relation with birth weight over the third
33                   trimester (6.0-g decrease;  [95% CI: -30.8, 18.8] per 10 ppb). Lin et al. (2004b) reported a
34                   positive, though not statistically significant, exposure-response relationship for  O3 during
35                   the entire pregnancy in a Taiwanese study. In a study performed in Korea, Ha et al.
36                   (2001) reported no O3 effect during the first trimester of pregnancy, but they found that
37                   during the third trimester of pregnancy O3 was associated with LEW (RR=1.05 [95% CI:
38                   1.02, 1.08] per 10 ppb).
      Draft - Do Not Cite or Quote                       7-43                                September 2011

-------
1

2

3

4
Table 7-5
Study
Salametal. (2005)
Morello-Froschetal.
(2010)
Darrow et al. (2011 a)
Chen etal. (2002)
Wilhelm and Ritz (2005)
Braueretal. (2008)
Dugandzicetal. (2006)
Mannesetal. (2005)
Hansen et al. (2007)
Gouveiaetal. (2004)
Lin et al. (2004b)
Ha et al. (2001)
Brief summary of epidemiologic studies of birth weight
Location
Sample Size
California, U.S.
(n=3,901)
California, U.S.
(n=3,545,177)
Atlanta, GA
(N=406,627)
Northern Nevada, US
(n=36,305)
Los Angeles County, CA
(n=136,134)
Vancouver, BC, Canada
(n=70,249)
Nova Scotia, Canada
(n=74,284)
Sydney, Australia
(n=138,056)
Brisbane, Australia
(n=26,617)
Sao Paulo, Brazil
(n=1 79,460)
Kaohsiung and Taipei,
Taiwan
(n=92,288)
Seoul, Korea
(n=276,763)
Mean Os (ppb)
24-h avg:
27.3
8h:
50.6
24-h avg:
23.5
8-h max:
44.8
8-h:
27.2
1-h:
21.1-22.2
24-h avg:
14
24-h avg:
21
1-h max:
31.6
8 h max:
26.7
1-h max:
31.5
24-h avg: 15.86-
47.78
8-h avg: 22.4-23.3°
Exposure assessment
ZIP code level
Nearest Monitor
(within 10,5,3km)
Population-weighted spatial
average
County level
Varying distances from monitor
Nearest Monitor
(within 10km)
Inverse Distance Weighting (IDW)
Nearest Monitor
(within 25 km)
City-wide avg and
<5 km from monitor
City-wide avg
City-wide avg
Nearest monitor
(within 3 km)
City-wide avg
Effect Estimate3
(95% Cl)
Entire pregnancy:
-39.3 g (-55.8, -22.8)
T1 : -6.1 g (-16.8, 4.8)
T2: -20.0 g (-31 .7, -8.4)
T3: -20.7 g (-32.1, -9.3)
Entire pregnancy: -5.7 g (-6.6, -
4.9)
T1:-2.1g(-2.9, -1.4)
T2: -2.3 g (-3.1, -1.5)
T3:-1.3g(-2.1,-0.6)
Entire pregnancy:
-1 2.3 g (-17.8, -6.8)
First 28 days:
-0.5 g (-3.0, 2.1)
T3: -0.9g (-4.5, 2.8)
Entire pregnancy:
20.9 g (6.3, 35.5)
T1: 23.4 g (-35.6, 82.4)
T2:-1 9.4 g (-77.0, 38.2)
T3: 7.7 g (-50.9, 66.3)
T1:NR
T3:NR
6 weeks before birth: NR
Entire pregnancy: NR
First 30 days of pregnancy: NR
Last 30 days of pregnancy: NR
T1:NR
T3:NR
T1: 0.97 (0.81, 1.1 8)°
T2: 1.06 (0.87, 1.27)d
T3:1.01 (0.83-1 .24)d
T1 : -0.9 g (-6.6, 4.8)
T2: -7.5 g (-13.8, 1.2)
T3: -4.5 g (-10.8, 1.8)
Last 30 days:
-1.1 g (-5.6, 3.4)
T1 : 2.8 g (-10.5, 16.0)
T2: 4.4 g (-11. 4, 20.1)
T3: 11. 3 g (-4.4, 27.1)
T1 : -3.2 g (-25.6, 19)
T2: -0.2 g (-23.8, 23.4)
T3: -6.0 g (-30.8, -18.8)
Entire pregnancy:
1.13(0.92,1.38)°
T1: 1.02 (0.85, 1.22)°
T2: 0.93 (0.78, 1.12)°
T3: 1 .05 (0.87, 1 .26)°
T1 : 0.87 (0.81 , 0.94)°
T3: 1.05 (1.02, 1.08)°
"Change in birthweight per 10 ppb change in 03
"Median
°0dds ratios of LEW; Highest quartile of exposure compared to lowest quartile of exposure
dRelative risk of LEW per 10 ppb change in 03
T1 = First Trimester, T2 = Second Trimester, T3 = Third Trimester
NR: No quantitative results reported

               Table 7-5 provides a brief overview of the epidemiologic studies of birth weight. In

               summary, only the Children's Health Study conducted in southern California (Salam et

               al..  2005) provides strong evidence for an effect of ambient O3 on birth weight. The study

               by Morello-Frosch et al. (2010). also conducted in California, provides support for the
      Draft - Do Not Cite or Quote
                                                 7-44
September 2011

-------
 1                   results of the Children's Health Study. Additional studies conducted in the U.S., Canada,
 2                   Australia, Latin America, and Asia provide limited and inconsistent evidence to support
 3                   the effect reported in the Children's Health Study. The toxicological literature on the
 4                   effect of O3 on birth weight is sparse. In some studies, the reporting of birth weight may
 5                   be avoided because birth weight can be confounded by decreased litter size resulting
 6                   from an increased rate of pup resorption (aborted pups) in O3 exposed dams. In one
 7                   toxicological study by Haro and Paz (1993). no differences in litter size were observed
 8                   and decreased birth weight in pups from dams who were exposed to Ippm O3  during
 9                   pregnancy was reported. A second animal toxicology study recapitulated these finding
10                   with pregnant BALB/c mice that exposed to O3 (1.2 ppm, GD9-18) producing pups with
11                   significantly decreased birth weights (Sharkhuu et al.. 2011).
             7.4.4   Preterm Birth

12                   Preterm birth (PTB) is a syndrome (Romero et al., 2006) that is characterized by multiple
13                   etiologies. It is therefore unusual to be able to identify an exact cause for each PTB. In
14                   addition, PTB is not an adverse outcome in itself, but an important determinant of health
15                   status (i.e., neonatal morbidity and mortality). Although some overlap exists for common
16                   risk factors, different etiologic entities related to distinct risk factor profiles and leading
17                   to different neonatal and postneonatal complications are attributed to PTB and measures
18                   of fetal growth. Although both restricted fetal growth and PTB can result in LEW,
19                   prematurity does not have to result in LEW or growth restricted babies.

20                   A major issue in studying environmental exposures and PTB is selecting the relevant
21                   exposure period, since the biological mechanisms leading to PTB and the critical periods
22                   of vulnerability are poorly understood (Bobak. 2000). Exposures proximate to the birth
23                   may be most relevant if exposure causes an acute effect. However, exposure occurring in
24                   early gestation might affect placentation, with results observable later in pregnancy, or
25                   cumulative exposure during pregnancy may be the most important determinant. The
26                   studies reviewed have  dealt with this issue in different ways. Many have considered
27                   several exposure metrics based on different periods of exposure.  Often the time periods
28                   used are the first month (or first trimester) of pregnancy and the last month (or 6 weeks)
29                   prior to delivery. Using a time interval prior to delivery introduces an additional problem
30                   since cases and controls are not in the same stage of development when they are
31                   compared. For example, a preterm infant delivered at 36 weeks is a 32-week fetus
32                   4 weeks prior to birth,  while an infant born at term (40 weeks)  is a 36-week fetus 4 weeks
33                   prior to birth.
      Draft - Do Not Cite or Quote                       7-45                                September 2011

-------
 1                   Recently, investigators have examined the association of PTB with both short-term (i.e.,
 2                   hours, days, or weeks) and long-term (i.e., months or years) exposure periods. Time-
 3                   series studies have been used to examine the association between air pollution
 4                   concentrations during the days immediately preceding birth. An advantage of these time-
 5                   series studies is that this approach can remove the influence of covariates that vary across
 6                   individuals over a short period of time. Retrospective cohort and case-control studies
 7                   have been used to examine long-term exposure periods, often averaging air pollution
 8                   concentrations over months or trimesters of pregnancy.

 9                   Reported studies fail to show consistency in pollutants and periods during pregnancy
10                   when an effect occurs. For example, while some studies  find the strongest effects
11                   associated with exposures early in pregnancy, others report effects when the exposure is
12                   limited to the second or third trimester. However, the effect of air pollutant exposure
13                   during pregnancy on PTB has a biological basis. There is an expanding list of possible
14                   mechanisms that may explain the association between O3 exposure and PTB (see
15                   Section 5. 4.2.4).

16                   Many studies of PTB compare exposure in quartiles, using the lowest quartile as the
17                   reference (or control) group. No studies use a truly unexposed control group. If exposure
18                   in the lowest quartile confers risk, than it may be difficult to demonstrate additional risk
19                   associated with a higher quartile. Thus negative studies must be interpreted with caution.

20                   Preterm birth occurs both naturally (idiopathic preterm), and as a result of medical
21                   intervention (iatrogenicpreterm). Ritz et al. (2007; 2000) excluded all births by Cesarean
22                   section to limit their studies to idiopathic preterm. No other studies attempted to
23                   distinguish the type of PTB, although air pollution exposure maybe associated with only
24                   one type. This is a source of potential effect misclassification.

25                   Generally, studies of air pollution-birth outcome conducted in North America and the
26                   United Kingdom have not identified an association between PTB and maternal exposure
27                   to O3. Most recently, Darrow et al. (2009) used vital record data to construct a
28                   retrospective cohort of 476,489 births occurring between 1994 and 2004 in 5 central
29                   counties of metropolitan Atlanta. Using a time-series approach, the authors examined
30                   aggregated daily counts of PTB in relation to ambient levels of CO, NO2, SO2, O3, PM10,
31                   PM2 5 and speciated PM measurements. This  study investigated 3 gestational windows of
32                   exposure: the first month of gestation, the final week of gestation, and the final 6 weeks
33                   of gestation. The authors did not observe associations of PTB with O3.

34                   A number of U.S. studies were conducted in southern California, and report somewhat
35                   inconsistent results. Ritz et al. (2000) evaluated the effect of air pollution (CO, NO2, O3,
36                   PMio) exposure during pregnancy on the occurrence of PTB in a cohort of 97,518
      Draft - Do Not Cite or Quote                       7-46                                 September 2011

-------
 1                   neonates born in southern California between 1989 and 1993. The authors use both short-
 2                   and long-term exposure windows, averaging pollutant measures taken at the closest air-
 3                   monitoring station over distinct periods, such as 1, 2, 4, 6, 8, 12, and 26 weeks before
 4                   birth and the whole pregnancy period. Additionally, they calculated average exposures
 5                   for the first and second months of pregnancy. The authors found no consistent effects for
 6                   O3 over any of the pregnancy periods in single or multi-pollutant models. Wilhelm and
 7                   Ritz (2005) extended previous analyses of PTB (Ritz et al., 2000; Ritz and Yu. 1999) in
 8                   California to include 1994-2000. The authors examined varying residential distances
 9                   from monitoring stations to see if the distance affected risk estimation, because effect
10                   attenuation may result from local pollutant heterogeneity inadequately captured by
11                   ambient monitors. The authors analyzed the association between O3 exposure during
12                   varying periods of pregnancy and PTB, finding a positive association between O3 levels
13                   in both the first trimester of pregnancy (RR=1.23 [95% CI: 1.06, 1.42]  per 10 ppb
14                   increase in 24-h avg O3) and the first month of pregnancy (results for first trimester
15                   exposure were similar, but slightly smaller, quantitative results not presented) in models
16                   containing all pollutants. No association was observed between O3 in the 6 weeks before
17                   birth and preterm delivery. Finally, Ritz et al. (2007) conducted a case-control survey
18                   nested within a birth cohort and assessed the extent to which residual confounding and
19                   exposure misclassification impacted air pollution effect estimates. The authors calculated
20                   mean exposure levels for three gestational periods: the entire pregnancy, the first
21                   trimester, and the last 6 weeks before delivery. Though positive associations were
22                   observed for CO and PM2 5, no consistent patterns of increase in the odds of PTB for O3
23                   or NO2 were observed.

24                   One study conducted in Canada evaluated the impacts of air pollution (including CO,
25                   NO2, NO, O3, SO2, PM25, and PM10) on PTBs (1999-2002) using spatiotemporal
26                   residential exposure metrics by month of pregnancy in Vancouver, BC (Brauer et al..
27                   2008). The authors did not observe consistent associations with any of the pregnancy
28                   average  exposure metrics except for PM25 for PTB. The O3 associations were largely
29                   protective, and no quantitative results were presented for O3. Additionally, Lee et al.
30                   (2008c)  used time-series techniques to investigate the short-term associations of O3 and
31                   PTB in London, England. In addition to exposure on the day of birth, cumulative
32                   exposure up to 1 week before birth was investigated. The risk of PTB did not increase
33                   with exposure to the levels of ambient air pollution experienced by this population.

34                   Conversely, studies conducted in Australia and China provide evidence for an association
35                   between ambient O3 and PTB. Hansen et al. (2006) reported that exposure to O3 during
36                   the first trimester was associated with an increased risk of PTB (OR=1.38, [95% CI:
37                   1.14, 1.69] per 10 ppb increase). Although the test for trend was significant due to the
38                   strong effect in the highest quartile, there was not an obvious exposure-response pattern
      Draft - Do Not Cite or Quote                       7-47                                September 2011

-------
 1                   across the quartiles of O3 during the first trimester. The effect estimate was diminished
 2                   and lost statistical significance when PMi0 was included in the model (OR=1.23, [95%
 3                   CI: 0.97, 1.59] per 10 ppb increase). Maternal exposure to O3 during the 90 days prior to
 4                   birth showed a weak, positive association with PTB (OR=1.09, [95% CI: 0.85, 1.39] per
 5                   10 ppb increase). Jalaludin et al. (2007) found that O3 levels in the month and
 6                   three months preceding birth had a statistically significant association with PTB. Ozone
 7                   levels in the first trimester of pregnancy were associated with increased risks for PTBs
 8                   (OR=1.15 [95% CI:  1.05,  1.24] per 10 ppb increase  in 1-h max O3 concentration), and
 9                   remained a significant predictor of PTB in copollutant models (ORs between 1.07 and
10                   1.10). ORs increased for first month of pregnancy when restricted to within 5 km of a
11                   monitoring station (OR=1.60, [95%  CI: 1.27, 2.03]), but did not show a cumulative effect
12                   for first 3 months of pregnancy (OR=0.81, [95% CI: 0.67, 0.98]). Jiang et al. (2007)
13                   examined the  acute effect of air pollution on PTB, including risk in relation to levels of
14                   pollutants for  a single day exposure window with lags from 0 to  6 days before birth. An
15                   increase of 10 ppb of the 8-week avg of O3 corresponded to 9.47 % (95% CI: 0.70,
16                   18.7%) increase in PTBs. Increases in  PTB were also observed for PMi0, SO2, and NO2.
17                   The authors did not observe any significant acute effect of outdoor air pollution on PTB
18                   among the 1-day acute time windows examined in the week before birth.

19                   Little data is available from toxicological studies; one study reported a nearly statistically
20                   significant increase in pregnancy duration in mice when exposed to 0.8 or 1.2 ppm O3.
21                   This phenomenon was most likely due to the anorexigenic effect of relatively high O3
22                   concentrations (Bignami et al.. 1994).
      Draft - Do Not Cite or Quote                       7-48                                 September 2011

-------
1
2
3
4
Table 7-6
Study
Darrow et al. (2009)
Ritzetal. (2000)
Wilhelm and Ritz (2005)
Ritzetal. (2007)
Braueretal. (2008)
Lee et al. (2008c)
Hansen et al. (2006)
Jalaludin et al. (2007)
Jiang etal. (2007)
Brief summary of epidemiologic studies of PTB
Location
Sample Size
Atlanta, GA
(n=476,489)
California, US
(n=97,158)
Los Angeles, CA
(n=1 06,483)
Los Angeles, CA
(n=58,316)
Vancouver, BC,
Canada
(n=70,249)
London, UK
Brisbane, Australia
(n=28,200)
Sydney, Australia
(n=1 23,840)
Shanghai, China
(n=3,346 preterm
births)
Mean O3
(PPb)
8-h max: 44.1
8 h: 36.9
1 h: 21. 1-22.2
24-h avg: 22.5
24-h avg:
14
24-h avg: NR
8-h max:
26.7
1-h max:
30.9
8-h avg:
32.7
Exposure
assessment
Population-weighted
spatial averages
Nearest Monitor (within
4 miles)
<2 mi of monitor
Varying distances to
monitor
Nearest monitor to ZIP
code
Nearest Monitor (within
10km)
Inverse Distance
Weighting (IDW)
1 monitor
City-wide avg
City-wide avg and <5
km from monitor
City-wide avg
Effect Estimate3 (95% Cl)
First month: 0.98 (0.97, 1.00)
Last week: 0.99 (0.98, 1 .00)
Last 6 weeks: 1.00 (0.98, 1.02)
First month: NR
Last 6 weeks: NR
First month: 1.23 (1.06, 1.42)
T1:NR
12:1.38(1.14,1.66)
Last 6 weeks: NR
Entire pregnancy: NR
11:0.93(0.82,1.06)
Last 6 weeks: NR
Entire pregnancy: NR
First 30 days of pregnancy: NR
Last 30 days of pregnancy: NR
T1:NR
T3:NR
Lag 0:1. 00 (1.00, 1.01)
11:1.39(1.15,1.70)
T3: 1 .09 (0.88, 1 .39)
First month: 1.604 (1.268,2.030)"
T1 : 0.807 (0.668, 0.976)b
T3: 1.011 (0.910,1.124)"
Last month: 0.984 (0.906, 1 .069)"
4 wks before birth: 1 .06 (1 .00, 1 .12)
6 wks before birth: 1 .06 (0.99, 1 .13)
8 wks before birth: 1 .09 (1 .01 , 1 .19)
                                                                        LO: NR (results presented in figure)
                                                                        L1: NR (results presented in figure)
                                                                        L2: NR (results presented in figure)
                                                                        L3: NR (results presented in figure)
                                                                        L4: NR (results presented in figure)
                                                                        L5: NR (results presented in figure)
                                                                        L6: NR (results presented in figure)
"Relative risk of PTB per 10 ppb change in 03.
"Relative risk of PTB per 1 ppb change in 03.
T1 = First Trimester, T2 = Second Trimester, T3 = Third Trimester
LO = Lag 0, L1= Lag 1, L2 = Lag 2, L3 = Lag 3, L4 = Lag 4, L5 = Lag 5, L6 = Lag 6
NR: No quantitative results reported

                Table 7-6 provides a brief overview of the epidemiologic studies of PTB. In summary,
                the evidence is consistent when examining shorter-term, late-pregnancy exposure to O3
                and reports no association with PTB. However when long-term exposure to O3 early in
                pregnancy is examined the results are inconsistent. Studies conducted in the U.S.,
      Draft - Do Not Cite or Quote
                                                    7-49
September 2011

-------
 1                   Canada, and England find no association with O3 and PTB, while studies conducted in
 2                   Australia and China report an O3 effect on PTB.
             7.4.5   Fetal Growth

 3                   Low birth weight has often been used as an outcome measure because it is easily
 4                   available and accurately recorded on birth certificates. However, LEW may result from
 5                   either short gestation, or inadequate growth in utero. Most of the studies investigating air
 6                   pollution exposure and LEW limited their analyses to term infants to focus on inadequate
 7                   growth. A number of studies were identified that specifically addressed growth restriction
 8                   in utero by identifying infants who failed to meet specific growth standards. Usually
 9                   these infants had birth weight less than the 10th percentile for gestational age, using an
10                   external standard. Many of these studies have been previously discussed, since they also
11                   examined other reproductive outcomes (i.e., LEW or PTB).

12                   Fetal growth is influenced by maternal, placental, and fetal factors. The biological
13                   mechanisms by which air pollutants may influence the developing fetus remain largely
14                   unknown. Several mechanisms have been proposed, and are the same as those
15                   hypothesized for birth weight (see Section 5. 4.2.4). Additionally, in animal toxicology
16                   studies, O3 causes transient anorexia in exposed pregnant dams. This may be one of
17                   many possible contributors to O3-dependent decreased fetal growth.

18                   A limitation of environmental studies that use birth weight as a proxy measure of fetal
19                   growth is that patterns of fetal growth during pregnancy cannot be assessed. This is
20                   particularly important when investigating pollutant exposures during early pregnancy as
21                   birth weight is recorded many months after the exposure period. The insult of air
22                   pollution may have a transient effect on fetal growth, where growth is hindered at one
23                   point in time but catches up at a later point. For example, maternal smoking during
24                   pregnancy can alter the growth rate of individual body segments of the  fetus at variable
25                   developmental  stages, as the fetus experiences selective growth restriction and
26                   augmentation (Lampl and Jeanty. 2003).

27                   The terms small-for-gestational-age (SGA), which is defined as a birth  weight <10th
28                   percentile for gestational age (and often sex and/or race), and intrauterine growth
29                   retardation (IUGR) are often used interchangeably. However, this definition of SGA does
30                   have limitations. For example, using it for IUGR may overestimate the  percentage of
31                   "growth-restricted" neonates as it is unlikely that 10% of neonates have growth
32                   restriction (Wollmann. 1998). On the other hand, when the 10th percentile is based on the
33                   distribution of live births at a population level, the percentage of SGA among PTB is
34                   most likely underestimated (Hutcheon and Platt 2008). Nevertheless, SGA represents a


      Draft - Do Not Cite or Quote                       7-50                                 September 2011

-------
 1                  statistical description of a small neonate, whereas the term IUGR is reserved for those
 2                  with clinical evidence of abnormal growth. Thus all IUGR neonates will be SGA, but not
 3                  all SGA neonates with be IUGR (Wollmann. 1998). In the following section the terms
 4                  SGA and IUGR are referred to as each cited study used the terms.

 5                  Over the past decade a number of studies examined various metrics of fetal growth
 6                  restriction. Salam et al.  (2005) assessed the effect of increasing O3 concentrations on
 7                  IUGR in a population of infants born in California from 1975-1987 as part of the
 8                  Children's Health Study. The authors reported that maternal O3 exposures averaged over
 9                  the entire pregnancy and during the third trimester were associated with increased risk of
10                  IUGR. A 10-ppb difference in 24-h maternal O3 exposure during the third trimester
11                  increased the risk of IUGR by 11% (95% CI: 0, 20%). Brauer et al. (2008) evaluated the
12                  impacts of air pollution (CO, NO2, NO, O3, SO2, PM2 5, PM10) on SGA (1999-2002)
13                  using spatiotemporal residential exposure metrics by month of pregnancy in Vancouver,
14                  BC. The O3 associations were largely protective (OR= 0.87, [95% CI: 0.81, 0.93] for a
15                  10 ppb increase in inverse distance weighted SGA), and no additional quantitative results
16                  were presented for O3. Liu et al. (2007b) examined the  association between IUGR among
17                  singleton term live births and SO2, NO2, CO, O3, and PM2 5 in 3 Canadian cities for the
18                  period 1985-2000. No increase in the risk of IUGR in relation to exposure to O3 averaged
19                  over each month and trimester of pregnancy was noted.

20                  Three studies conducted in Australia provide evidence for an association between
21                  ambient O3 and fetal growth restriction. Hansen et al. (2007) examined SGA among
22                  singleton, full-term births in Brisbane, Australia in relation to ambient air pollution (bsp,
23                  PMio, NO2, O3) during pregnancy. They also examined head circumference and crown-
24                  heel length in a subsample of term neonates. Trimester  specific exposures to all pollutants
25                  were not statistically significantly associated with a reduction in head circumference or
26                  an increased risk of SGA. When monthly-specific exposures were examined, the authors
27                  observed an increased risk of SGA associated with exposure to O3 during month 4
28                  (OR=1.11 [95% CI: 1.00, 1.24]  per 10 ppb increase). In a subsequent study, Hansen et al.
29                  (2008) examined the possible associations between fetal ultrasonic measurements and
30                  ambient air pollution (PM10, O3, NO2, SO2) during early pregnancy. This study had two
31                  strengths: (1) fetal growth was assessed during pregnancy as opposed to at birth; and (2)
32                  there was little delay between exposures and fetal growth measurements, which reduces
33                  potential confounding and uses exposures that are concurrent with the observed growth
34                  pattern of the fetus. Fetal ultrasound biometric measurements were recorded for biparietal
35                  diameter (BPD), femur length, abdominal circumference, and head circumference. To
36                  further improve exposure assessment, the authors restricted the samples to include only
37                  scans from women for whom the centroid of their postcode was within  14 km of an air
38                  pollution monitoring site. Ozone during days 31-60 was associated with decreases in all
      Draft - Do Not Cite or Quote                      7-51                                September 2011

-------
 1                   of the fetal growth measurements, and a 1.78 mm reduction in abdomen circumference
 2                   per 10 ppb increase in O3 concentration, though this effect did not persist in copollutant
 3                   models. The change in ultrasound measurements associated with O3 during days 31-60 of
 4                   gestation indicated that increasing O3 concentration decreased the magnitude of
 5                   ultrasound measurements for women living within 2 km of the monitoring site. The
 6                   relationship decreased toward the null  as the distance from the monitoring sites increased.
 7                   When assessing effect modification due to SES, there was some evidence of effect
 8                   modification for most of the associations, with the effects of air pollution stronger in the
 9                   highest SES quartile. In the third study, Marines et al. (2005) estimated the effects of
10                   pollutant (PMio, PM2 5, NO2, CO and  O3) exposure in the first, second and third
11                   trimesters of pregnancy and risk of SGA in Sydney, Australia. Citywide average air
12                   pollutant concentrations in the last month, third trimester, and first trimester of pregnancy
13                   had no effect on SGA. Concentrations of O3 in the second trimester of pregnancy had
14                   small but adverse effects on SGA (OR=1.10 [95% CI: 1.00, 1.14] per 10 ppb  increment).
15                   This  effect disappeared when the analysis was limited to births with a maternal address
16                   within 5 km of a monitoring station (OR=1.00 [95% CI: 0.60, 1.79] per 10 ppb
17                   increment).

18                   Very little information from toxicological studies is available to address effects on fetal
19                   growth. However, there is evidence to  suggest that prenatal exposure to O3 can affect
20                   postnatal growth. A few studies reported that mice or rats exposed developmentally
21                   (gestationally ± lactationally) to O3 had deficits in body weight gain in the postpartum
22                   period (Bignami et al.. 1994: Haro and Paz. 1993: Kavlock et al.. 1980).

23                   Table 7-7 provides a brief overview of the epidemiologic studies of fetal growth
24                   restriction. In summary, the evidence is inconsistent when examining exposure to O3 and
25                   fetal  growth restriction. Similar to PTB, studies conducted in Australia have reported an
26                   effect of O3 on fetal growth, whereas studies  conducted in other areas have not found
27                   such an effect. This may be due to the  restriction of births to those within 2-14 km of a
28                   monitoring  station,  as was done in the  Australian studies.
      Draft - Do Not Cite or Quote                       7-52                                 September 2011

-------
Table 7-7
Study
Salametal.
(2005)


Braueretal.
(2008)

Liu et al. (2007b)



Hansen et al.
(2007)

Hansen et al.
(2008)


Mannes etal.
(2005)


Brief summary of epidemiologic studies of fetal growth
Location
(Sample Size)
California, U.S.
(n=3901)


Vancouver, BC, Canada
(n=70,249)

Calgary, Edmonton, and
Montreal, Canada
(n= 16,430)


Brisbane, Australia
(n=26,617)

Brisbane, Australia
(n=15,623)


Sydney, Australia
(n=138,056)


Mean O3 (ppb)
24-h avg:
27.3
8h:
50.6
24-h avg:
14

24-h avg:
16.5
1-h max:
31.2

8-h max:
26.7

8-h avg:
24.8


1-h max:
31.6


Exposure
assessment
ZIP code level


Nearest Monitor (within
10km)
Inverse Distance
Weighting (IDW)

Census Subdivision avg



City-wide avg

Within 2 km of monitor


City-wide avg and
<5 km from monitor


Effect Estimate3 (95% Cl)
Entire pregnancy: 1 .16 (1 .00, 1 .32)
11:1.00(0.94,1.11)
12:1.06(1.00,1.12)
73:1.11 (1.00,1.17)
Entire pregnancy: NR
First 30 days of pregnancy: NR
Last 30 days of pregnancy: NR
71:NR
73: NR
Entire pregnancy: NR (results presented in
figure)
71 : NR (results presented in figure)
72: NR (results presented in figure)
73: NR (results presented in figure)
71:1.01 (0.89,1.15)
72:1.00(0.86,1.17)
73: 0.83 (0.71 , 0.97)
M1: -0.32 (-1.56,0.91)"
M2: -0.58 (-1.97,0.80)"
M3: 0.26 (-1.07, 1.59)"
M4:0.11 (-0.98,1.21)"
71 : 0.90 (0.48, 1 .34)
72:1.00(0.60,1.79)
73:1.10(0.66,1.97)
Last 30 days of pregnancy: 1 .10 (0.74,
1.79)
'Relative risk of fetal growth restriction per 10 ppb change in 03.
"Mean change in fetal ultrasonic measure of head circumference recorded between 13 and 26 weeks gestation for a 10-ppb increase in maternal
exposure to 03 during early pregnancy
71 = First 7rimester, 72 = Second 7rimester, 73 = 7hird 7rimester
M1 = Month 1, M2 = Month 2, M3 = Month 3, M4 = Month 4
NR: No quantitative results reported
1
2
3
4
5
6
       7.4.6   Postnatal growth

                Time-pregnant BALB/c mice were exposed to O3 (0, 0.4, 0.8, or 1.2 ppm) GD9-18 with
                parturition at GD20-21 (Sharkhuu et al.. 2011). As the offspring aged, postnatal litter
                body weight continued to be significantly decreased in the highest dose (1.2 ppm) O3
                group at PND3 and PND7. When the pups were weighed separately by sex at PND42, the
                males with the highest dose of O3 exposure (1.2 ppm, GD9-18) had significant
                decrements in body weight (Sharkhuu et al.. 2011).
Draft - Do Not Cite or Quote
                                                      7-53
September 2011

-------
             7.4.7   Birth Defects

 1                   Despite the growing body of literature evaluating the association between ambient air
 2                   pollution and various adverse birth outcomes, relatively few studies have investigated the
 3                   effect of temporal variations in ambient air pollution on birth defects. Heart defects and
 4                   oral clefts have been the focus of the majority of these recent studies, given the higher
 5                   prevalence than other birth defects and associated mortality. Mechanistically, air
 6                   pollutants could be involved in the etiology of birth defects via a number of key events
 7                   (see Section 5. 4.2.4).

 8                   Several studies have been conducted examining the relationship between O3 exposure
 9                   during pregnancy and birth defects and reported  a positive association with cardiac
10                   defects. The earliest of these studies was conducted in southern California (Ritz et al..
11                   2002). This study evaluated the effect of air pollution on the occurrence of cardiac birth
12                   defects in neonates and fetuses delivered in southern California in 1987-1993. Maternal
13                   exposure estimates were based on  data from the fixed site closest to the mother's ZIP
14                   code area. When using a case-control design where cases were matched to 10 randomly
15                   selected controls, results showed increased risks  for aortic  artery and valve defects
16                   (OR=1.56 [95% CI: 1.16, 2.09] per 10 ppb O3), pulmonary artery and valve anomalies
17                   (OR=1.34 [95% CI: 0.96, 1.87] per 10 ppb O3), and conotruncal defects (OR=1.36 [95%
18                   CI: 0.91, 2.03] per 10 ppb O3) in a dose-response manner with second-month O3
19                   exposure. A study conducted in Texas (Gilboaetal.. 2005) looked at a similar period of
20                   exposure but reported no association with most of the birth defects studied (O3
21                   concentration was studied using quartiles with the lowest representing <18 ppb and the
22                   highest representing > 31 ppb). The authors found slightly elevated odds ratios for
23                   pulmonary artery and valve defects. They also detected an inverse association between
24                   O3 exposure and isolated ventricular septal defects. Overall, this study provided some
25                   weak evidence that air pollution increases the risk of cardiac defects. Hansen et al. (2009)
26                   investigated the possible association between ambient air pollution and the risk of cardiac
27                   defects. When analyzing all births with exposure estimates for O3 from the nearest
28                   monitor there was no indication for an association with cardiac defects. There was also
29                   no adverse association when restricting the analyses to only include births where the
30                   mother resided within  12 km of a monitoring station. However, among births within 6 km
31                   of a monitor, a 10 ppb increase in  O3  was associated with an increased risk of pulmonary
32                   artery and valve defects (OR=8.76 [95% CI: 1.80, 56.55]). As indicated by the very wide
33                   credible intervals, there were very few cases in the sensitivity analyses for births within 6
34                   km of a monitor, and this effect could be a result of type I errors. Dadvand et al. (2011)
3 5                   investigated the association between maternal exposure to  ambient air pollution and the
36                   occurrence of cardiac birth defects in  England. Similar to Hansen et al. (2009), they
37                   found no associations with maternal exposure to O3 except for when the analysis was

      Draft - Do Not Cite or Quote                       7-54                                  September 2011

-------
 1                   limited to those subjects residing within a 16 km distance of a monitoring station (OR for
 2                   malformations of pulmonary and tricuspid valves=1.64 [95% CI: 1.04, 2.60] per 10 ppb
 3                   increase in O3).

 4                   Despite the association between O3 and cardiac defects observed in the above studies, a
 5                   recent study did not observe an increased risk of cardiac birth defects associated with
 6                   ambient O3 concentrations. The study, conducted in Atlanta, GA, examined O3 exposure
 7                   during the third through seventh week of pregnancy and reported no association with risk
 8                   of cardiovascular malformations (mean long-term average of 8-h O3 concentrations
 9                   excluding November through February ranged by 5-year groups from 39.8 to 43.3 ppb)
10                   (Strickland etaL 2009).

11                   Several of these studies have also examined the relationship between O3 exposure during
12                   pregnancy and oral cleft defects. The study by Ritz et al. (Ritz et al.. 2002) evaluated the
13                   effect of air pollution on the occurrence of orofacial birth defects and did not observe
14                   strong associations between ambient O3 concentration and orofacial defects. They did
15                   report an OR of 1.13 (95% CI: 0.90, 1.40) per 10 ppb during the second trimester for cleft
16                   lip with or without cleft palate. Similarly, Gilboa et al. (Gilboa et al.. 2005) reported and
17                   OR of 1.09 (95% CI: 0.70, 1.69) for oral cleft defects when the fourth quartile was
18                   contrasted with the first quartile of exposure during 3-8 weeks of pregnancy. Hansen et
19                   al. (2009) reported no indication for an association with cleft defects. Hwang and Jaakola
20                   (2008)  conducted a population-based case-control study to investigate exposure to
21                   ambient air pollution and the risk of cleft lip with or without cleft palate in Taiwan. The
22                   risk of cleft lip with or without cleft palate was increased in relation to O3 levels in the
23                   first gestational month (OR= 1.17 [95% CI:  1.01, 1.36] per 10 ppb) and second gestational
24                   month (OR=1.22 [95% CI: 1.03, 1.46] per 10 ppb), but was not related to any of the other
25                   pollutants. In three-pollutant models, the effect estimates for O3 exposure were stable for
26                   the four different combinations of pollutants and were all statistically significant.
27                   Marshall et al. (2010) compared estimated exposure to ambient pollutants during early
28                   pregnancy among mothers of children with oral cleft defects to that among mothers of
29                   controls. The authors observed no consistent elevated associations between any of the air
30                   pollutants examined and cleft malformations, though there was a weak association
31                   between cases of cleft palate only and increasing O3 concentrations. This association
32                   increased when cases and controls were limited to those with residences within 10 km of
33                   the closest O3 monitor (OR=2.2 [95% CI: 1.0, 4.9], comparing highest quartile  [>33 ppb]
34                   to lowest quartile [<15 ppb]).

35                   A limited number of toxicological studies have examined birth defects in animals
36                   exposed gestationally to O3. Kavlock et al. (1979) exposed pregnant rats to O3  for precise
37                   periods during organogenesis. No significant teratogenic effects were found in rats
      Draft - Do Not Cite or Quote                       7-55                                September 2011

-------
 1
 2
 3
 4

 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
exposed 8 hr/day to concentrations of O3 varying from 0.44 to 1.97 ppm during early
(days 6-9), mid (days 9-12), or late (days 17 to 20) gestation, or the entire period of
organogenesis (days 6-15). Earlier research found eyelid malformation following
gestational and postnatal exposure to 0.2 ppm O3 (Veninga. 1967).

Table 7-8 provides a brief overview  of the epidemiologic studies of birth defects. These
studies have focused on cardiac and  oral cleft defects, and the results from these studies
are not entirely consistent. This inconsistency could be due to the absence of true
associations between O3 and risks of cardiovascular malformations and oral cleft defects;
it could also be due to differences in populations, pollution levels, outcome definitions, or
analytical approaches. The lack of consistency of associations between O3 and
cardiovascular malformations or oral cleft defects might be due to issues relating to
statistical power or measurement error. A recent meta-analysis of air pollution and
congenital anomalies concluded that there was no statistically significant increase in risk
of congenital anomalies and O3 (Vrijheid et al., 2011). These authors note that
heterogeneity in the results of these studies may be due to inherent differences in study
location, study design, and/or analytic methods, and comment that these studies have not
employed some recent advances in exposure assessment used in other areas of air
pollution research that may help refine or reduce this heterogeneity.
Table 7-8
Study
Ritzetal. (2002)
Gilboaetal. (2005)
Hwang and Jaakola
Strickland etal. (2009)
Hansen et al. (2009)
Marshall et al. (2010)
Dadvand et al. (2011)
Brief summary of epidemiologic studies of birth defects
ExlCmineedS
Cardiac and Cleft
Defects
Cardiac and Cleft
Defects
Oral Cleft Defects
Cardiac Defects
Cardiac and Cleft
Defects
Oral Cleft Defects
Cardiac Defects
(Samp^Size)
Southern California
(n=3,549 cases;
10,649 controls)
7 Counties in TX
(n=5,338 cases;
4,580 controls)
Taiwan
(n=653 cases;
6,530 controls)
Atlanta, GA
(n=3,338 cases)
Brisbane, Australia
(n=150,308 births)
New Jersey
(n=71 7 cases;
12,925 controls)
Northeast England
(n=2, 140 cases;
14,256 controls)
Mean 03 (p
24-h avg:
NR
24-h avg:
NR
24-h avg:
27.31
8-h max:
39.8-43.3
8-h max:
25.8
24-h avg:
25
24-h avg:
18.8
.. Exposure
p ' Assessment
Nearest Monitor
(within 10 mi)
Nearest Monitor
Inverse Distance
Weighting (IDW)
Weighted City-wide
avg
Nearest Monitor
Nearest Monitor
(within 40 km)
Nearest Monitor
Exposure Window
Month 1,2,3
Trimester 2,3
3-mo period prior to
conception
Weeks 3-8 of gestation
Months 1,2,3
Weeks 3-7 of gestation
Weeks 3-8 of gestation
Weeks 5-1 0 of gestation
Weeks 3-8 of
gestation' 1
      Draft - Do Not Cite or Quote
                                7-56
September 2011

-------
            7.4.8   Developmental Respiratory Effects

 1                   The issue of prenatal exposure has assumed increasing importance since ambient air
 2                   pollution exposures of pregnant women have been shown to lead to adverse pregnancy
 3                   outcomes, as well as to respiratory morbidity and mortality in the first year of life.
 4                   Growth and development of the respiratory system take place mainly during the prenatal
 5                   and early postnatal periods. This early developmental phase is thought to be very
 6                   important in determining long-term lung growth. Studies have recently examined this
 7                   emerging issue. Several studies were included in Sections 7.2.1 and 7.2.3, and are
 8                   included here because they reported both prenatal and post-natal  exposure periods.

 9                   Mortimer et al. (2008a, b) examined the association of prenatal and lifetime exposures to
10                   air pollutants with pulmonary function and allergen sensitization in a subset of asthmatic
11                   children (ages 6-11) included in the Fresno Asthmatic Children's Environment Study
12                   (FACES). Monthly means of pollutant levels for the years 1989-2000 were created and
13                   averaged separately across several important developmental time-periods, including the
14                   entire pregnancy, each trimester, the first 3 years of life, the first  6 years of life, and the
15                   entire lifetime. The 8-h avg O3 concentrations were approximately 50 ppb for each of the
16                   exposure metrics (estimated from figure). In the first analysis (Mortimer et al.. 2008a).
17                   negative effects on pulmonary function were found for exposure to PM10, NO2, and CO
18                   during key neonatal and early life developmental periods. The authors did not find a
19                   negative effect of exposure to O3 among this cohort. In the  second analysis (Mortimer et
20                   al.. 2008b). sensitization to at least one allergen was associated, in general, with higher
21                   levels of CO and PM10 during the entire pregnancy and second trimester and higher PM10
22                   during the first 2 years of life. Lower exposure to O3 during the entire pregnancy or
23                   second trimester was associated with an increased risk of allergen sensitization. Although
24                   the pollutant metrics across time periods are  correlated, the strongest associations with
25                   the outcomes were observed for prenatal exposures. Though it may be difficult to
26                   disentangle the effect of prenatal and postnatal exposures, the models from this group of
27                   studies suggest that each time period of exposure may contribute independently to
28                   different dimensions of school-aged children's pulmonary function. For 4 of the 8
29                   pulmonary-function measures (FVC, FEVi,  PEF, FEF25-75), prenatal exposures were
30                   more influential on pulmonary function than early-lifetime  metrics, while, in contrast, the
31                   ratio of measures (FEVi/FVC and FEF25_75/FVC) were most influenced by postnatal
32                   exposures. When lifetime metrics were considered alone, or in combination with the
33                   prenatal metrics, the lifetime measures were  not associated  with any of the outcomes,
34                   suggesting the timing of the exposure may be more important than the overall dose and
35                   prenatal exposures are not just markers for lifetime or current exposures.
      Draft - Do Not Cite or Quote                       7-57                                September 2011

-------
 1                   Clark et al. (2010) investigated the effect of exposure to ambient air pollution in utero
 2                   and during the first year of life on risk of subsequent asthma diagnosis (incident asthma
 3                   diagnosis up to age 3-4) in a population-based nested case-control study. Air pollution
 4                   exposure for each subject based on their residential address history was estimated using
 5                   regulatory monitoring data, land use regression modeling, and proximity to stationary
 6                   pollution sources. An average exposure was calculated for the duration of pregnancy
 7                   (-15 ppb)  and the first year of life (-14 ppb). In contrast to the Mortimer et al. studies
 8                   (2008a. b), the effect estimates for first-year exposure were generally larger than for in
 9                   utero exposures. However, similar to the Mortimer et al. studies, the observed
10                   associations with O3 were largely protective. Because of the relatively high correlation
11                   between in utero and first-year exposures for many pollutants, it was difficult to discern
12                   the relative importance of the individual exposure periods.

13                   Latzin et al. (2009) examined whether prenatal exposure to air pollution was  associated
14                   with lung function changes in the newborn. Tidal breathing, lung volume, ventilation
15                   inhomogeneity and eNO were measured in 241 unsedated, sleeping neonates (age=
16                   5 weeks). The median of the 24-h avg O3 concentrations averaged across the post-natal
17                   period was -44 ppb. Consistent with the previous  studies, no  association was found for
18                   prenatal exposure to O3 and lung function.

19                   The new toxicological literature  since the 2006 O3 AQCD, covering respiratory changes
20                   related to developmental O3 exposure, reports ultrastructural  changes in bronchiole
21                   development, alterations in placental and pup cytokines, and increased pup airway hyper-
22                   reactivity.  These studies are detailed below. Older studies are discussed where new
23                   information is not available.

24                   Fetal rat lung bronchiole development is triphasic, comprised of the glandular phase
25                   (measured at GDIS), the canalicular phase (GD20), and the saccular phase (GD21). The
26                   ultrastructural lung development in fetuses of pregnant rats exposed to 1-ppm O3 (12
27                   h/day, out to either GDIS, GD20 or GD21) was examined by electron microscopy during
28                   these three phases.  In the glandular phase,  bronchiolar columnar epithelial cells in  fetuses
29                   of dams exposed to O3 had cytoplasmic damage and swollen  mitochondria. Bronchial
30                   epithelium at the canalicular phase in O3 exposed pups had delayed maturation in
31                   differentiation, i.e., glycogen abundance in secretory cells had not diminished as it should
32                   with this phase of development. Congruent with this finding,  delayed maturation of
33                   tracheal epithelium following  early neonatal O3 exposure  (1 ppm, 4-5 h/day for
34                   first week  of life) in lambs has been previously reported (Mariassy et al.. 1990; Mariassy
35                   et al., 1989). Also at the canalicular phase, atypical cells were seen in the bronchiolar
36                   lumen of O3 exposed rat fetuses. Finally, in the saccular phase, mitochondrial
37                   degradation was present in the non-ciliated bronchiolar cells of rats exposed in utero to
      Draft - Do Not Cite or Quote                       7-58                                September 2011

-------
 1                   O3. In conclusion, O3 exposure of pregnant rats produced ultra-structural damage to near-
 2                   term fetal bronchiolar epithelium (Lopez et al.. 2008).

 3                   Exposure of laboratory animals to multiple airborne pollutants can differentially affect
 4                   pup physiology. One study showed that exposure of C57BL/6 mouse dams to 0.48 mg
 5                   PM intratracheally twice weekly for 3 weeks during pregnancy augmented O3-induced
 6                   airway hyper-reactivity in juvenile offspring. Maternal PM exposure also significantly
 7                   increased placental cytokines above vehicle-instilled controls. Pup postnatal O3 exposure
 8                   (1 ppm 3 h/day, every other day, thrice weekly for 4 weeks) induced significantly
 9                   increased cytokine levels (IL-1(3,  TNF-a, KC, and IL-6) in whole lung versus postnatal
10                   air exposed groups; this was further exacerbated with gestational PM exposure (Auten et
11                   al.. 2009).

12                   A series of experiments using infant rhesus monkeys repeatedly exposed to 0.5 ppm O3
13                   starting at one-month of age have examined the effect of O3 alone or in combination with
14                   an inhaled allergen on morphology and lung function (Plopper et al.. 2007). Exposure to
15                   O3 alone or allergen alone produced small but not statistically significant changes in
16                   baseline airway resistance and airway responsiveness, but the combined exposure to both
17                   O3 + antigen produced  statistically significant and greater than additive changes in both
18                   functional measurements. Additionally, cellular changes and significant structural
19                   changes in the respiratory tract have been observed in infant rhesus monkeys exposed to
20                   O3 (Fanucchi et al., 2006). A more detailed description of these studies can be found in
21                   Section 7.2.3 (Pulmonary Structure and Function), with mechanistic information found in
22                   Section 5.4.2.4.

23                   Lung immunological response in O3 exposed pups was followed by analyzing BAL and
24                   lung tissue. Sprague Dawley (SD) pups were exposed to a single 3h exposure of air or O3
25                   (0.6 ppm) on PND 13 (Han  etal.. 2011). Bronchoalveolar lavage (BAL) was performed
26                   10 hours after the end of O3 exposure. BALF polymorphonuclear leukocytes (PMNs) and
27                   total BALF protein were significantly elevated in O3 exposed pups. Lung tissue from O3
28                   exposed pups had significant elevations of manganese superoxide dismutase (SOD)
29                   protein and significant decrements of extra-cellular SOD protein.

30                   Various immunological outcomes were followed in offspring after their pregnant dams
31                   (BALB/c mice) were exposed gestationally to O3 (0, 0.4, 0.8, or 1.2 ppm, GD9-18)
32                   (Sharkhuu et al., 2011). Delayed type hypersensitivity (DTH) was initiated with initial
33                   BSA injection at 6 weeks of age and then challenge 7 days later. The normal edematous
34                   response of the exposed footpad (thickness after BSA injection) was recorded as an
35                   indicator of DTH. In female offspring, normal footpad swelling with BSA injection that
36                   was seen in air exposed animals was significantly attenuated with O3 exposure (0.8 and
37                   1.2 ppm O3), implying immune suppression of O3 exposure specifically in DTH.
      Draft - Do Not Cite or Quote                      7-59                                September 2011

-------
 1                  Humoral immunity was measured with the sheep red blood cell (SRBC) response.
 2                  Animals received primary immunization with SRBC and then blood was drawn for
 3                  SRBC IgM measurement. A SRBC booster was given 2 weeks later with blood collected
 4                  5 days after booster for IgG measurement. Maternal O3 exposure had no effect on
 5                  humoral immunity in the offspring as measured by IgG and IgM titers after SRBC
 6                  primary and booster immunizations (Sharkhuu et al.. 2011).

 7                  Toxicity assessment and allergen sensitization was also assessed in these O3 exposed
 8                  offspring. At PND42, animals were euthanized for analysis of immune and inflammatory
 9                  markers (immune proteins, inflammatory cells, T cell populations in the spleen). A subset
10                  of the animals was intra-nasally instilled or sensitized with ovalbumin on either PND2
11                  and 3 or PND42 and 43. All animals were challenged with OVA on PND54, 55, and 56.
12                  One day after final OVA challenge, lung function, lung inflammation and immune
13                  response were determined. Offspring of O3 exposed dams that were initially sensitized at
14                  PDN3 (early) or PND42 (late) were tested to determine the level of allergic sensitization
15                  or asthma-like inflammation after OVA challenge.  Female offspring sensitized early in
16                  life developed significant eosinophilia (1.2 ppm O3) and elevated  serum OVA-specific
17                  IgE (1.2 ppm O3), which is a marker of airway allergic inflammation. The females that
18                  were sensitized  early also had significant decrements in BALF total cells, macrophages,
19                  and lymphocytes (1.2 ppm O3). Offspring that were sensitized later (PND42) in life did
20                  not develop the  aforementioned changes in BALF,  but these animals did develop modest,
21                  albeit significant neutropenia (0.8 and 1.2 ppm O3) (Sharkhuu et al.. 2011).

22                  BALF cytology in non-sensitized animals was followed. BALF of offspring born to dams
23                  exposed to  O3 was relatively unaffected (cytokines, inflammatory cell numbers/types) as
24                  were splenic T cell subpopulations. LDH was significantly elevated in BALF of females
25                  whose mothers were exposed to 1.2 ppm during pregnancy (Sharkhuu et al.. 2011). In
26                  summary, the females born to mothers exposed to O3 developed modest
27                  immunocompromise. Males were unaffected (Sharkhuu et al.. 2011).

28                  Overall, animal  toxicological studies have reported ultrastructural changes in bronchiole
29                  development, alterations in placental and pup cytokines, and increased pup airway hyper-
30                  reactivity related to exposure to O3 during the developmental period. Epidemiologic
31                  studies have found no association between prenatal exposure to O3 and growth and
32                  development of the respiratory system. Fetal origins of disease have received a lot of
33                  attention recently, thus additional research to further explore the inconsistencies between
34                  these two lines of evidence is warranted.
      Draft - Do Not Cite or Quote                      7-60                                September 2011

-------
            7.4.9   Developmental Central Nervous System Effects

 1                   The following sections describe the results of toxicological studies of O3 and
 2                   developmental central nervous system effects. No epidemiologic studies of this
 3                   association have been published.
                     7.4.9.1    Laterality

 4                   Two reports of laterality changes in mice developmentally exposed to O3 have been
 5                   reported in the literature. Mice developmentally exposed to 0.6 ppm O3 (6 days before
 6                   breeding to weaning at PND21) showed a turning preference (left turns) distinct from air
 7                   exposed controls (clockwise turns) (Dell'Omo et al.. 1995); in previous studies this
 8                   behavior in mice has been found to correlate with specific structural asymmetries of the
 9                   hippocampal mossy fiber projections (Schopke etal.. 1991). The 2006 O3 AQCD
10                   evidence for the effect of O3 on laterality or handedness demonstrated that rats exposed
11                   to O3 during fetal and neonatal life showed limited, gender-specific changes in
12                   handedness after exposure to the intermediate dose of O3  (only seen in female mice
13                   exposed to 0.6 ppm O3, and not in males at 0.6 ppm or in either sex of 0.3 or 0.9 ppm O3
14                   with exposure from 6 days before breeding to PND26) (Petruzzi et al.. 1999).
                     7.4.9.2   Brain Morphology and Neurochemical Changes

15                   The nucleus tractus solitarius (NTS), a medullary area of respiratory control, of adult
16                   animals exposed prenatally to 0.5 ppm O3 (12h/day, ED5-ED20) had significantly less
17                   tyrosine hydroxylase staining versus control (Boussouar et al.. 2009). Tyrosine
18                   hydroxylase is the rate-limiting enzyme for dopamine synthesis and serves as a precursor
19                   for catecholamine synthesis; thus, decreased staining is used as a marker of dopaminergic
20                   or catecholaminergic cell or activity loss in these regions and thus functions in neuronal
21                   plasticity. After physical restraint stress, control animals respond at the histological level
22                   with Fos activation, a marker of neuronal activity, and tyrosine hydroxylase activation in
23                   the NTS, a response which is absent or attenuated in adult animals exposed prenatally to
24                   0.5 ppm O3 (Boussouar et al.. 2009) when compared to control air exposed animals who
25                   also were restrained. The O3-exposed offspring in this study were cross-fostered to
26                   control air exposed dams to avoid O3-dependent dam related neonatal effects on
27                   offspring outcomes (i.e., dam behavioral or lactational contributions to pup outcomes)
28                   (Boussouar etal.. 2009).
      Draft - Do Not Cite or Quote                      7-61                                 September 2011

-------
 1                  Developmental exposure to 0.3 or 0.6 ppm O3 prior to mating pair formation through
 2                  GDI? induced significant increased levels of BDNF in the striatum of adult (PND140)
 3                  O3 exposed offspring as compared to control air exposed animals; these O3-exposed
 4                  animals also had significantly decreased level of NGF in the hippocampus versus control
 5                  (Santucci etal., 2006).

 6                  Changes in the pup cerebellum with prenatal 1 ppm O3 exposure include altered
 7                  morphology (Romero-Velazquez et al.. 2002; Rivas-Manzano and Paz. 1999). decreased
 8                  total area (Romero-Velazquez et al.. 2002). decreased number of Purkinje cells (Romero-
 9                  Velazquez et al.. 2002). and altered monoamine neurotransmitter content with the
10                  catecholamine system affected and the indoleamine system unaffected by O3 (Gonzalez-
11                  Pina et al.. 2008).
                    7.4.9.3   Neurobehavioral Outcomes

12                  O3 administration to dams during pregnancy with or without early neonatal exposure has
13                  been shown to contribute to multiple neurobehavioral outcomes in offspring that are
14                  described in further detail below.

15                  O3 administration (0.4, 0.8 or 1.2 ppm O3) during the majority of pregnancy (PD7-17) of
16                  CD-I mice did not affect pup behavioral outcomes including early behavioral ultrasonic
17                  vocalizations and more permanent later measurements (PND60 or 61) including pup
18                  activity, habituation and exploration and d-amphetamine-induced hyperactivity (Bignami
19                  et al.. 1994); these pups were all cross-fostered or reared on non- O3 exposed dams.

20                  Testing for aggressive behavior in mice continuously exposed to O3 (0.3 or 0.6 ppm from
21                  30 days prior to mating to GDI7) revealed that mice had significantly increased
22                  defensive/ submissive behavior (increased freezing posturing on the first day only of a
23                  multiple-day exam) versus air exposed controls (Santucci et al.. 2006). Similar to this and
24                  as reported in previous AQCDs, continuous exposure of adult animals to O3 induced
25                  significant increases in fear behavior and decreased aggression as measured by
26                  significantly decreased freezing behavior (Petruzzi et al..  1995).

27                  Developmentally exposed animals also had significantly decreased amount of time spent
28                  nose sniffing other mice (Santucci et al.. 2006): this social behavior deficit, decreased
29                  sniffing time, was not found in an earlier study with similar exposures (Petruzzi et al..
30                  1995). but sniffing of specific body areas was measured in Santucci et al. (2006) and total
31                  number of sniffs of the entire body was measured in Petruzzi et al. (1995). The two
32                  toxicology studies exploring social behavior (sniffing) employ different study designs
33                  and find opposite effects in animals exposed to O3
      Draft - Do Not Cite or Quote                      7-62                                September 2011

-------
                     7.4.9.4    Sleep Aberrations after Developmental Ozone Exposure

 1                   The effect of gestational O3 exposure (1 ppm O3, 12h/day, during dark period) on sleep
 2                   patterns in rat offspring was followed using 24 h polysomnographic recordings at 30, 60
 3                   and 90 days of age (Haro and Paz. 1993). Ozone-exposed pups manifested with inverted
 4                   sleep-wake patterns or circadian rhythm phase-shift. Rat vigilance was characterized in
 5                   wakefulness, slow wave sleep (SWS), and paradoxical sleep (PS) using previously
 6                   characterized criteria. The O3 exposed offspring spent longer time in the wakefulness
 7                   state during the light period, more time in SWS during the period of darkness, and
 8                   showed significant decrements in PS. Chronic O3 inhalation significantly decreased the
 9                   duration of PS during both the light and dark periods (Haro and Paz. 1993). These effects
10                   were consistent at all time periods measured (30, 60 and 90 days of age). These sleep
11                   effects reported after developmental exposures expand upon the existing literature on
12                   sleep aberrations in adult animals exposed to O3 [rodents: (Paz and Huitron-Resendiz.
13                   1996; Aritoetal.. 1992); and cats: (Paz and Bazan-Perkins. 1992)]. A role for inhibition
14                   of cyclooxygenase-2 and the interleukins and prostaglandins in the O3-dependent sleep
15                   changes potentially exists with evidence from a publication on indomethacin pretreatment
16                   attenuating O3-induced sleep aberrations in adult male animals (Rubio and Paz. 2003).
            7.4.10  Early Life Mortality

17                   Infants may be particularly susceptible to the adverse effects of air pollution. Within the
18                   first year of life, infants develop rapidly; therefore their susceptibility may change within
19                   weeks or months. During the neonatal and post-neonatal periods, the developing lung is
20                   highly susceptible to environmental toxicants. The lung is not well developed at birth,
21                   with 80% of alveoli being formed postnatally. An important question regarding the
22                   association between O3 and infant mortality is the critical window of exposure during
23                   development for which infants are susceptible. Several age intervals have been explored:
24                   neonatal (<1 month); postneonatal (1 month to 1 year); and an overall interval for infants
25                   that includes both the neonatal and postneonatal periods (<1 year). Within these various
26                   age categories, multiple causes of deaths have been investigated, particularly total deaths
27                   and respiratory-related deaths. The studies reflect a variety of study designs, exposure
28                   periods, regions, and adjustment for confounders. As discussed below, a handful of
29                   studies have examined the effect of ambient air pollution on neonatal and postneonatal
30                   mortality, with the former the least studied. These studies varied somewhat with regard to
31                   the outcomes and exposure periods examined and study designs employed.

32                   A major issue  in studying environmental exposures and infant mortality is selecting the
33                   relevant exposure period, since the biological mechanisms leading to death and the
      Draft - Do Not Cite or Quote                       7-63                                September 2011

-------
 1                   critical periods of exposure are poorly understood. Both short-term (days to weeks) and
 2                   long-term (months to years) exposure studies are included in this section and are
 3                   characterized accordingly in the text and tables. All studies of infant mortality are
 4                   included in the Reproductive and Developmental Effects section, as opposed to the
 5                   sections devoted to all- and cause-specific mortality, because infant development
 6                   processes, much like fetal development processes, may be particularly susceptible to O3-
 7                   induced health effects. Exposures proximate to the death may be most relevant if
 8                   exposure causes an acute effect. However, exposure occurring in early life might affect
 9                   critical growth and development, with results observable later in the first year of life, or
10                   cumulative exposure during the first year of life may be the most important determinant.
11                   The studies reviewed below have dealt with this issue in different ways. Many have
12                   considered  several exposure metrics based on different periods of exposure.
                     7.4.10.1   Stillbirth

13                   Pereira et al. (1998) investigated the association among daily counts of intrauterine
14                   mortality (over 28 weeks of gestation) and air pollutant concentrations in Sao Paulo,
15                   Brazil from 1991 through 1992. The association was strong for NO2, but lesser for SO2
16                   and CO. These associations exhibited a short lag time, less than 5 days. No significant
17                   association was detected between short-term O3 exposure and intrauterine mortality.
                     7.4.10.2  Infant Mortality, Less than 1 Year

18                   Ritz et al. (2006) linked birth and death certificates for infants who died between 1989
19                   and 2000 to evaluate the influence of outdoor air pollution on infant death in the South
20                   Coast Air Basin of California. The authors examined short- and long-term exposure
21                   periods 2 weeks, 1 month, 2 months, and 6 months before each case subject's death and
22                   reported no association between ambient levels of O3 and infant mortality. Similarly,
23                   Diaz et al. (2004) analyzed the effects of extreme temperatures and short-term exposure
24                   to air pollutants on daily mortality in children less than 1 year of age in Madrid, Spain,
25                   from 1986 to 1997 and observed no statistically significant association between mortality
26                   and O3 concentrations. Hajat et al. (2007) analyzed time-series data of daily infant
27                   mortality counts in 10 major cities in the  UK to quantify any associations with short-term
28                   changes in air pollution. When the results from the 10 cities were combined there was no
29                   relationship between O3  and infant mortality, even after restricting the analysis to just the
30                   summer months.
      Draft - Do Not Cite or Quote                       7-64                                 September 2011

-------
 1                  Conversely, a time-series study of infant mortality conducted in the southwestern part of
 2                  Mexico City in the years 1993-1995 found that infant mortality was associated with
 3                  short-term exposure to NO2 and O3 3-5 days before death, but not as consistently as with
 4                  PM. A 10-ppb increase in 24-h avg O3 was associated with a 2.78% increase (95% CI:
 5                  0.29, 5.26%) in infant mortality (lag 3) (Loomis et al., 1999). This increase was
 6                  attenuated, although still positive when evaluated in a two-pollutant model with PM2 5.
 7                  One-hour max concentrations of O3 exceeded prevailing Mexican and international
 8                  standards nearly every day.
                    7.4.10.3  Neonatal Mortality, Less than 1  Month

 9                  Several studies have evaluated ambient O3 concentrations and neonatal mortality and
10                  observed no association. Ritz et al. (2006) linked birth and death certificates for infants
11                  who died between 1989 and 2000 to evaluate the influence of outdoor air pollution on
12                  infant death in the South Coast Air Basin of California. The authors examined short- and
13                  long-term exposure periods 2 weeks, 1 month, 2 months, and 6 months before each case
14                  subject's death and reported no association between ambient levels of O3 and neonatal
15                  mortality. Hajat et al. (2007) analyzed time-series data of daily infant mortality counts in
16                  10 major cities in the UK to quantify any associations with short-term changes in air
17                  pollution. When the results from the 10 cities were combined there was no relationship
18                  between O3 and neonatal mortality, even after restricting the analysis to just the summer
19                  months. Lin et al. (2004a) assessed the impact of short-term changes in air pollutants on
20                  the number of daily neonatal deaths in Sao Paulo, Brazil. The authors observed no
21                  association between ambient levels of O3 and neonatal mortality.
                    7.4.10.4  Postneonatal Mortality,  1  Month to 1 Year

22                  A number of studies focused on the postneonatal period when examining the effects of
23                  O3 on infant mortality. Ritz et al. (2006) linked birth and death certificates for infants
24                  who died between 1989 and 2000 to evaluate the influence of outdoor air pollution on
25                  infant death in the South Coast Air Basin of California. The authors examined short- and
26                  long-term exposure periods 2 weeks, 1 month, 2 months, and 6 months before each case
27                  subject's death and reported no association between ambient levels of O3 and
28                  postneonatal mortality. Woodruff et al. (2008) evaluated the county-level relationship
29                  between cause-specific postneonatal infant mortality and long-term early-life exposure
30                  (first 2 months of life) to air pollutants across the U.S. Similarly, they found no
31                  association between O3 exposure and deaths from respiratory causes. In the U.K., Hajat
32                  et al. (2007) analyzed time-series data of daily infant mortality counts in 10 major cities

      Draft - Do Not Cite or Quote                      7-65                                September 2011

-------
 1                   to quantify any associations with short-term changes in air pollution. When the results
 2                   from the 10 cities were combined there was no relationship between O3 and postneonatal
 3                   mortality, even after restricting the analysis to just the summer months. In Ciudad Juarez,
 4                   Mexico, Romieu et al. (2004b) examined the daily number of deaths between 1997 and
 5                   2001, estimating the modifying effect of SES on the risk of postneonatal mortality.
 6                   Ambient O3  concentrations were not related to infant mortality overall, or in any of the
 7                   SES groups.  In a follow-up study, Carbajal-Arroyo (2011) evaluated the relationship of
 8                   1-h daily max O3 levels with postneonatal infant mortality in the Mexico City
 9                   Metropolitan Area between 1997 and 2005. Generally, short-term exposure to O3 was not
10                   significantly related to infant mortality. However, upon estimating the modifying effect
11                   of SES on the risk of postneonatal mortality, the authors found that O3 was statistically
12                   significantly related to respiratory mortality among those with low SES. In a separate
13                   analysis, the  effect of PM10 was evaluated with O3  level quartiles. PM10 alone was related
14                   to a significant increase in all-cause mortality. The magnitude of this effect remained the
15                   same when only the days when O3 was in the lowest quartile were included in the
16                   analyses. However, when only the days when O3 was in the highest quartile were
17                   included in the analyses, the magnitude of the PM10 effect increased dramatically
18                   (OR=1.06 [95% CI: 0.909, 1.241] for PM10 on days with O3 in lowest quartile; OR=1.26
19                   [95% CI: 1.08, 1.47] for PM10 on days with O3 in the highest quartile. These results
20                   suggest that while O3 alone may not have an effect on infant mortality, it may serve to
21                   potentiate the observed effect of PM10 on infant mortality.

22                   Tsai et al. (2006) used a case-crossover analysis to examine the relationship between
23                   short-term exposure to air pollution and postneonatal mortality in Kaohsiung, Taiwan
24                   during the period 1994-2000. The risk of postneonatal deaths was 1.023 (95% CI: 0.564,
25                   1.858) per 10-ppb increase in 24-h avg O3. The confidence interval for this effect
26                   estimate is very wide, likely due to the small number of infants that died each day,
27                   making it difficult to interpret this result. Several other studies conducted in Asia did not
28                   find any association between O3 concentrations and infant mortality in the postneonatal
29                   period. Ha et al. (2003) conducted a daily time-series study in Seoul, Korea to evaluate
30                   the  effect of short-term changes in ambient 8-h O3 concentrations on postneonatal
31                   mortality. Son et al. (2008) examined the relationship between air pollution and
32                   postneonatal mortality from all causes among firstborn infants in Seoul, Korea during
33                   1999-2003. Yang et al. (2006) used a case-crossover analysis to examine the relationship
34                   between air pollution exposure and postneonatal mortality in Taipei, Taiwan for the
35                   period 1994-2000. The authors observed no associations between ambient levels of O3
36                   and postneonatal mortality.
      Draft - Do Not Cite or Quote                       7-66                                September 2011

-------
                     7.4.10.5  Sudden Infant Death Syndrome

 1                   The strongest evidence for an association between ambient O3 concentrations and SIDS
 2                   comes from a study that evaluated the county-level relationship between SIDS and long-
 3                   term early-life exposure (first 2 months of life) to air pollutants across the U.S.(Woodruff
 4                   et al.. 2008). The authors observed a 1.20 (95% CI: 1.09, 1.32) odds ratio for a 10-ppb
 5                   increase in O3 and deaths from SIDS. There was a monotonic increase in odds of SIDS
 6                   for each quartile of O3 exposure compared with the lowest quartile (highest quartile
 7                   OR=1.51; [95% CI: 1.17, 1.96]). In a multi-pollutant model including PM10 or PM25, CO
 8                   and SO2, the OR for SIDS and O3 was not substantially lower than that found in the
 9                   single-pollutant model. When examined by season, the relationship between SIDS deaths
10                   and O3 was generally consistent across seasons with a slight increase for those babies
11                   born in the summer. When stratified by birth weight, the OR for LEW babies was 1.27
12                   (95% CI: 0.95, 1.69) per 10-ppb increase in O3 and the OR for normal weight babies was
13                   1.16 (95% CI: 1.01, 1.32) per 10-ppb increase in O3.

14                   Conversely, two additional studies reported no association between ambient levels of O3
15                   and SIDS.  Ritz et al. (2006) linked birth and  death certificates for infants who died
16                   between 1989 and 2000 to evaluate the influence of outdoor air pollution on infant death
17                   in the South Coast Air Basin of California. The authors examined short- and long-term
18                   exposure periods 2 weeks, 1 month, 2 months, and 6 months before each case subject's
19                   death and reported no association between ambient levels of O3 and SIDS. Dales et al.
20                   (2004) used time-series analyses to compare the daily mortality rates for SIDS and short-
21                   term air pollution concentrations in 12 Canadian cities during the period of 1984-1999.
22                   Increased daily rates of SIDS were associated with previous day increases in the levels of
23                   SO2, NO2, and CO, but not O3 or PM2 5.

24                   Table 7-9 provides a brief overview of the epidemiologic studies of infant mortality.
25                   These studies have focused on short-term exposure windows (e.g., 1-3 days) and long-
26                   term exposure windows (e.g., up to 6 months). Collectively, they provide no evidence for
27                   an association between ambient O3  concentrations and infant mortality.
      Draft - Do Not Cite or Quote                      7-67                                September 2011

-------
Table 7-9       Brief summary of infant mortality studies
Study
Pereiraetal. (1998)
Diaz etal. (2004)
Loomisetal. (1999)
Ritz etal. (2006)
Haiat etal. (2007)
Lin et al. (2004a)
Ha et al. (2003)
Romieu et al. (2004b)
Carbajal-Arroyo et al.
(2011)
Son et al. (2008)
Tsai et al. (2006)
Woodruff etal. (2008)
Yana etal. (2006)
Dales etal. (2004)
Location
Sao Paulo, Brazil
Madrid, Spain
Mexico City, Mexico
Southern California
10 Cities in the UK
Sao Paulo, Brazil
Seoul, South Korea
Ciudad Juarez,
Mexico
Mexico City, Mexico
Seoul, South Korea
Kaohsiung, Taiwan
Nationwide, US
Taipei, Taiwan
12 Canadian cities
Mean Os (ppb)
1-h max: 33.8
24-havg:11.4
24-h avg: 44.1
1-h max: 163.5
24-h avg: 21 .9-22.1
24-h avg: 20.5-42.6
24-h avg: 38.06
8-havg:21.2
8-h avg: 43.43-55.1 2
1-h max: 103.0
8-ha avg: 25.61
24-h avg: 23.60
24-h avg: 26.6
24-h avg: 18.14
24-h: 31 .77
Exposure
Assessment
Citywide avg
Citywide avg
1 monitor
Nearest Monitor
Citywide avg
Citywide avg
Citywide avg
Citywide avg
Citywide avg
Citywide avg
Citywide avg
County wide avg
Citywide avg
Citywide avg
Effect Estimate3 (95% Cl):
10-2:1.00(0.99,1.01)
NR
LO: 0.99 (0.97, 1 .02)
11:0.99(0.96,1.01)
L2: 1 .00 (0.98, 1 .03)
L3: 1 .03 (1 .00, 1 .05)
L4: 1 .01 (0.98, 1 .03)
L5: 1 .02 (0.99, 1 .04)
10-2:1.02(0.99,1.05)
2 wk before death: 1 .03 (0.93, 1 .14)
1 mo before death: NR
2 mo before death: 0.93 (0.89, 0.97)
6 mo before death: NR
10-2:1.00(0.96,1.06)
10:1.00(0.99,1.01)
LO: 0.93 (0.90, 0.96)
L1: 0.96 (0.90, 1.03)
L2: 0.97 (0.91 , 1 .04)
LO-1 cum: 0.96 (0.89, 1.04)
LO-2 cum: 0.94 (0.87, 1.02)
LO: 1 .00 (0.99, 1 .00)
L1: 0.99 (0.99, 0.99)
L2: 0.99 (0.99, 1.00)
LO-2: 0.99 (0.99, 1.00)
L(NR): 0.984 (0.976, 0.992)b
LO-2 cum: 1.02 (0.56, 1.86)
First 2 mo of life: 1.04 (0.98, 1.10)
LO-2 cum:1. 00 (0.62, 1.61)
LO:NR
L1:NR
L2:NR
L3:NR
L4:NR
L5:NR
Multiday lags of 2-6 days: NR
  "Relative risk of infant mortality per 10 ppb change in 03
  bNo increment provided
  LO = Lag 0, L1= Lag 1, L2 = Lag 2, L3 = Lag 3, L4 = Lag 4, L5 = Lag 5, L6 = Lag 6
  NR: No quantitative results reported
Draft - Do Not Cite or Quote
7-68
September 2011

-------
Table 7-10    Summary of Key Reproductive and Developmental Toxicological
             Studies
Study
Sharkhuu et
al. (2011)
Bignami et al.
(1994)
Haro and Paz
(1993)
Lopez et al.
(2008)
Auten et al.
(2009)
Plopper et al.
(2007)
Fanucchi et
al. (2006)
Dell'Omo et
al. (1995)
Santucci et al.
(2006)
Hanetal.
(2011)
Campos-
Bedolla et al.
(2002)
Kavlocketal.
(1980)
Jedlinska-
Krakowska et
al. (2006)
Model
Pregnant mice;
BALB/c; F; GD9-
18; effects in
offspring
Pregnant CD-1
dams(PD7-17)
Rat dams,
Exposure over
the entirety of
pregnancy;
Rats; Pregnant
dams;GD1-
GD18, GD20, or
GD21.
C57BL/6 mouse
pups
Infant rhesus
monkeys
Infant male
Rhesus
monkeys, post-
natal exposure
CD-1 Mouse
dams and pups
CD-1 Mouse
dams
Rat; Sprague
Dawley, M & F;
PND13
Pregnant Rats;
Sprague Dawley
(GD5, 10 or 18)
CD-1 mice;
(pregnancy day
7-17)
5 month old male
Wistar Hannover
rats
03
(ppm)
0.4, 0.8,
or 1.2
0.4, 0.8
or 1.2
1.0
1.0
1.0
0.5
0.5
0.6
0.3 or
0.6
0.6
3.0
0.4, 0.8
and 1.2
3.0
Exposure Duration
Continuously for 10
consecutive days
Continuous
12h/day during dark cycle
(12h/day, out to either
GD18, GD20orGD21)
3 h/day, every other day,
thrice weekly for 4 weeks
Postnatal, PND30-6month
of age, 5 months of cyclic
exposure, 5days03
followed by 9 days of
filtered air, 8h/day.
5 months of episodic
exposure, age 1 month-
age 6 months, 5 days 03
followed by 9 days of
filtered air, 8h/day.
6 days before breeding to
weaning at PND21
Dam exposure prior to
mating through GD17.
3 h, BALF examined 10h
after 0 3 exposure
1 h on one day of
gestation, uteri collected
16-1 8 h later
Continuous, pregnancy
day 7-1 7
0.5 ppm, 5h/dayfor50
days
Effects
Dams: Decreased number of dams reaching parturition. Offspring: 1-
Decreased birth weights. 2-Decreased rate of postnatal growth (body
weight). 3-lmpaired delayed type hypersensitivity.4-No effect on humoral
immunity. 5-Significantly affected allergic airway inflammation markers
(eosinophilia, IgE) in female offspring sensitized early in life. 6-BALF LDH
significantly elevated in female offspring.
Reproductive success was not affected by 03 exposure (PD7-17,
proportion of successful pregnancies, litter size, ex ratio, frequency of still
birth, or neonatal mortality). Ozone acted as a transient anorexigen in
pregnant dams.
Decreased birth weight and postnatal body weight of offspring out to PND
90. Ozone-exposed pups manifested with inverted sleep-wake patterns or
circadian rhythm phase-shift.
03 induced delayed maturation of near term rodent bronchioles, with
ultra-structural damage to bronchiolar epithelium.
Postnatal 03 exposure significantly increased lung inflammatory cytokine
levels; this was further exacerbated with gestational PM exposure.
Non-significant increases airway resistance and airway responsiveness
with 03 or inhaled allergen alone. Allergen + 03 produced additive
changes in both measures.
Cellular changes and significant structural changes in the distal
respiratory tract in infant rhesus monkeys exposed to 03 postnatally.
Laterality changes in offspring: Ozone exposed pups showed a turning
preference (left turns) distinct from air exposed controls (clockwise turns)
as adults.
Developmental 03 caused increased defensive/submissive behavior in
offspring. 03 exposed offspring also had significant elevations of striatal
BDNF and hippocampal NGF v. air exposed controls.
BALF polymorphonuclear leukocytes and total BALF protein were
significantly elevated in 03 exposed pups. Lung tissue from 03 exposed
pups had significant elevations of manganese superoxide dismutase
(SOD) protein and significant decrements of extra-cellular SOD protein.
Ozone inhalation modifies the contractile response of the pregnant
uterus. The 03 exposed pregnant uteri had significant increases in the
maximum response to acetyl choline stimulation at GD5 and 10; they also
had a significant increase in maximal response to oxytocin at GD 5.
03 induced decrements in postnatal body weight gain. When 03 was co-
administered with sodium salicylate, 03 synergistically increased the rate
of pup resorption (1 .0 ppm GD9-12).
Histopathological evidence of impaired spermatogenesis (round
spermatids/ 21 spermatocytes, giant spermatid cells, and focal epithelial
desquamation with denudation to the 22 basement membrane). Vitamin E
exposure concomitant with 03 protected against pathological changes
but Vitamin C did not.
Draft - Do Not Cite or Quote
7-69
September 2011

-------
             7.4.11  Summary and Causal Determination

 1                   The 2006 O3 AQCD concluded that the limited number of studies that investigated O3
 2                   demonstrated no associations between O3 and birth outcomes, with the possible
 3                   exception of birth defects. The current review included an expanded body of evidence on
 4                   the associations between O3 and reproductive and developmental effects. Recent
 5                   epidemiologic and toxicological studies provide evidence for an effect of prenatal
 6                   exposure to O3 on pulmonary structure and function, including lung function changes in
 7                   the newborn, incident asthma, ultrastructural changes in bronchiole development,
 8                   alterations in placental and pup cytokines, and increased pup airway hyper-reactivity.
 9                   Also, there is limited toxicological evidence for an effect of prenatal and early life
10                   exposure on central nervous system effects, including laterality, brain morphology,
11                   neurobehavioral abnormalities, and sleep aberration. Recent epidemiologic studies have
12                   begun to explore the effects of O3 on sperm quality, and provide limited evidence for
13                   decrements in sperm concentration, while there is limited toxicological evidence for
14                   testicular degeneration associated with O3.

15                   While the collective evidence for many of the birth outcomes examined is generally
16                   inconsistent (including birth defects), there are several well-designed, well-conducted
17                   studies that indicate an association between O3  and adverse outcomes. For example, as
18                   part of the southern California Children's Health Study, Salam et al. (2005) observed a
19                   concentration-response relationship of decreasing birth weight with increasing O3
20                   concentrations averaged over the entire pregnancy that was clearest above the 30-ppb
21                   level (see Figure 7-4). Simiarly,Hansen et al.  (2008) utilized fetal ultrasonic
22                   measurements and found a change in ultrasound measurements associated with O3 during
23                   days 31-60 of gestation indicated that increasing O3 concentration decreased an
24                   ultrasound measurement for women living within 2 km of the monitoring site.

25                   There is no evidence that prenatal or early life O3 concentrations are associated with
26                   infant mortality. Collectively, there is limited though positive toxicological evidence for
27                   O3-induced developmental effects, including  effects on pulmonary structure and function
28                   and central nervous system effects. Limited epidemiologic evidence for an effect on
29                   prenatal O3 exposure on respiratory development provides coherence  with the effects
30                   observed in toxicological studies. There is also limited epidemiologic evidence for an
31                   association with O3 concentration and decreased sperm concentration. A recent
32                   toxicological study provides limited evidence for a possible biological mechanism
33                   (histopathology showing impaired spermatogenesis) for such an association.
34                   Additionally, though the evidence for an association between O3 concentrations and
35                   adverse birth outcomes is generally inconsistent, there are several influential studies that
36                   indicate an association with reduced birth weight and restricted fetal growth. Taking into
      Draft - Do Not Cite or Quote                        7-70                                 September 2011

-------
 1                   consideration the positive evidence for developmental and reproductive outcomes from
 2                   toxicological and epidemiological studies, and the few influential birth outcome studies,
 3                   the evidence is suggestive of a causal relationship between long-term exposures to O3
 4                   and reproductive and developmental effects.
          7.5    Central  Nervous System  Effects
            7.5.1    Effects on the Brain and Behavior

 5                   The 2006 O3 AQCD included toxicological evidence that acute exposures to O3 are
 6                   associated with alterations in neurotransmitters, motor activity, short and long term
 7                   memory, and sleep patterns. Additionally, histological signs of neurodegeneration have
 8                   been observed. Reports of headache, dizziness, and irritation of the nose with O3
 9                   exposure are common complaints in humans, and some behavioral changes in animals
10                   may be related to these symptoms rather than indicative of neurotoxicity. Research in the
11                   area of O3-induced neurotoxicity has notably increased over the past few years, and new
12                   studies examining the effects of long-term exposure have demonstrated progressive
13                   damage in various regions of the brains of rodents in conjunction with altered behavior.
14                   Evidence from epidemiologic studies has been more limited. A recently published
15                   epidemiologic study examined the association between O3 exposure and neurobehavioral
16                   effects. Chen et al. (2009) utilized data from the NHANES III cohort to study the
17                   relationship between O3 levels (mean annual O3 concentration 26.5 ppb) and
18                   neurobehavioral effects among adults aged 20-59 years. The authors observed an
19                   association between annual exposure to O3 and tests measuring coding ability (symbol-
20                   digit substitution test) and attention/short-term memory (serial-digit learning test). Each
21                   10-ppb increase in annual O3 levels corresponded to an aging-related cognitive
22                   performance decline of 3.5 yr for coding ability and 5.3 years for attention/short-term
23                   memory. These associations persisted in both crude and adjusted models. There was no
24                   association between O3 levels and reaction time tests. The authors concluded that overall,
25                   there is an association between long-term O3 exposure and reduced performance on
26                   neurobehavioral tests.

27                   A number of new toxicological studies demonstrate various perturbations in neurologic
28                   function or histology with long-term exposure to O3, including changes similar to those
29                   observed in neurodegenerative disorders such as Parkinson's and Alzheimer's disease
30                   pathologies in relevant regions of the brain (Table 7-11). The central nervous system is
31                   very sensitive to oxidative stress, due in part to its high content of polyunsaturated fatty
32                   acids, high rate of oxygen consumption, and low antioxidant enzyme capacity. Oxidative
      Draft - Do Not Cite or Quote                       7-71                                September 2011

-------
 1                   stress has been identified as one of the pathophysiological mechanisms underlying
 2                   neurodegenerative disease (Simonian and Coyle. 1996). and it is believed to play a role in
 3                   altering hippocampal function, which causes cognitive deficits with aging (Vanguilder
 4                   and Freeman. 2011). A particularly common finding in studies of O3-exposed rats is lipid
 5                   peroxidation in the brain, especially in the hippocampus, which is important for higher
 6                   cognitive function including contextual memory acquisition. Performance in passive
 7                   avoidance learning tests is impaired when the hippocampus is injured. For example, in a
 8                   subchronic study, exposure of rats to 0.25 ppm O3 (4 h/day) for 15-90 days caused a
 9                   complex array of responses, including a time-dependent increase in lipid peroxidation
10                   products and immunohistochemical changes in the hippocampus that were correlated
11                   with decrements in passive avoidance behavioral tests (Rivas-Arancibia et al., 2010).
12                   Changes included increased numbers of activated microglia, a sign of inflammation, and
13                   progressive neurodegeneration. Notably, continued exposure tends to bring about
14                   progressive, cumulative damage, as shown by this study (Rivas-Arancibia et al.. 2010)
15                   and others (Santiago-Lopez et al., 2010; Guevara-Guzman et al., 2009; Angoa-Perez et
16                   al.. 2006). The effects of O3 on passive avoidance test performance were particularly
17                   evident at 90 days for both short- and long-term memory. The greatest extent of cell loss
18                   was also observed at this time point, whereas lipid peroxidation did not increase much
19                   beyond 60 days of exposure.

20                   The substantia nigra is another region of the brain affected by O3, and seems particularly
21                   sensitive to oxidative stress because the metabolism of dopamine, central to its function,
22                   is an oxidative process perturbed by redox imbalance. Oxidative stress has been
23                   implicated in the  premature death of substantia nigra dopamine neurons in Parkinson's
24                   disease. Progressive damage has been found in the substantia nigra of male rats after 15,
25                   30, and 60 days of exposure to 0.25 ppm O3 for 4 h/day. Santiago-Lopez and colleagues
26                   (2010) observed a reduction dopaminergic neurons within the substantia nigra over time,
27                   with a complete loss of normal morphology in the remaining cells and virtually no
28                   dopamine immunoreactivity at 60 days. This was accompanied by an increase in p53
29                   levels and nuclear translocation, a process associated with programmed cell death.
30                   Similarly, Angoa-Perez et al.  (2006) have shown progressive lipoperoxidation in the
31                   substantia nigra and a decrease in nigral neurons in ovariectomized  female rats exposed
32                   to 0.25 ppm O3, 4h/day, for 7-60 days. Lipid peroxidation effectively doubled between
33                   the 30 and 60 day time points. Total nigral cell number was also diminished to the
34                   greatest extent at 60 days, and cell loss was particularly evident in the tyrosine
35                   hydroxylase positive cell population (90%), indicating a selective loss of dopamine
36                   neurons or a loss  of dopamine pathway functionality.

37                   The olfactory bulb also undergoes oxidative damage in O3-exposed animals, in some
3 8                   cases altering olfactory-dependent behavior. Lipid peroxidation was observed in the
      Draft - Do Not Cite or Quote                       7-72                                 September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 9
10
11
12
13
14
15
16
olfactory bulbs of ovariectomized female rats exposed to 0.25 ppm O3 (4 h/day) for 30 or

60 days (Guevara-Guzman et al.. 2009). O3 also induced decrements in a selective

olfactory recognition memory test, which were significantly greater at 60 days compared

to 30 days, and the authors note that early deficits in odor perception and memory are

components of human neurodegenerative diseases. The decrements in olfactory memory

did not appear to be due to damaged olfactory perception based on other tests early on,

but by 60 days deficits in olfactory perception had emerged.

Memory deficits and associated morphological changes can be attenuated by

administration of a-tocopherol (Guerrero etal.. 1999). taurine (Rivas-Arancibia et al..

2000). and estradiol (Guevara-Guzman et al.. 2009; Angoa-Perez et al.. 2006). all of

which have antioxidant properties. In the study by Angoa-Perez et al. (2006) described

above, estradiol seemed particularly effective at protecting against lipid peroxidation and

nigral cell loss at 60  days compared to shorter exposure durations. The same was true for

amelioration of decrements in olfactory recognition memory (Guevara-Guzman et al..

2009). although protection against lipid peroxidation was  similar for the 30 and 60 day

exposures.
Table 7-11
Study
Angoa-Perez et al.
(2006)
Central nervous system effects of long-term Os exposure in rats
Model
Rat; Wistar; F;
Weight: 300 g;
Ovariectomized
O3 (ppm) Exposure Duration
0.25 7 to 60 days, 4 h/day,
5 days/wk
Effects
Long-term estradiol treatment protected against
03-induced oxidative damage to nigral
dopamine neurons, lipid peroxidation, and loss
of tyrosine hydrolase-immunopositive cells.
      Guevara-Guzman et Rat; Wistar; F;
      al. (2009)         Weight: 264 g;
                      Ovariectomized
                  0.25
            30 and 60 days, 4h/day
                     Long-term estradiol treatment protected against
                     03-induced oxidative stress and decreases in a
                     and p estrogen receptors and dopamine p-
                     hydroxlyase in olfactory bulb, and deficits in
                     olfactory social recognition memory and
                     chocolate recognition.	
      Rivas-Arancibia et
      al. (2010)
Rat; Wistar; M;
Weight: 250-300 g
0.25
15 to 90 days, 4h/day
Ozone produced significant increases in lipid
peroxidation in the hippocampus, and altered
the number of p53 positive immunoreactive
cells, activated and phagocytic microglia, GFAP
immunoreactive cells, double cortine cells, and
short- and long-term memory-retention latency
Santiago-Lopez et
al. (2010)
Santucci et al.
(2006)
Rat; Wistar; M; 0.25
Weight: 250-300 g
Mice;CD-1;M; 0.3; 0.6
18 weeks old
15, 30, and 60 days,
4 h/day
Females continuously
exposed from 30 days prior
to breeding untilGD17
Progressive loss of dopamine reactivity in the
substantia nigra, along with morphological
changes. Increased p53 levels and nuclear
translocation.
Upon behavioral challenge with another male,
there was a significant increase in defensive
and freezing postures and decrease in the
frequency of nose-sniffing. These behavioral
changes were accompanied by a significant
increase in BDNF in the striatum and a
decrease of NGF in the hippocampus.
      Draft - Do Not Cite or Quote
                                  7-73
                                                       September 2011

-------
 1                   CNS effects have also been demonstrated in adult mice whose only exposure to O3
 2                   occurred while in utero, a period particularly critical for brain development. Santucci et
 3                   al. (2006) investigated behavioral effects and gene expression after in utero exposure of
 4                   mice to 0.3 or 0.6 ppm O3. Exposure began 30 days prior to mating and continued
 5                   throughout gestation. Testing of adult animals demonstrated increased
 6                   defensive/submissive behavior and reduced social investigation were observed in both the
 7                   0.3 and 0.6 ppm O3 groups. Changes in gene expression of brain-derived neurotrophic
 8                   factor (BDNF, increased in striatum) and nerve growth factor (NGF, decreased in
 9                   hippocampus) accompanied these behavioral changes. BDNF and NGF are involved in
10                   neuronal organization and the growth, maintenance, and survival of neurons during early
11                   development and in adulthood. This study and two others using short-term exposures
12                   demonstrate that CNS effects can occur as a result of in utero exposure to O3, and
13                   although the mode of action of these effects is not known, it has been suggested that
14                   circulating lipid  peroxidation products may play a role (Boussouar et al.. 2009).
15                   Importantly, these CNS effects occurred in rodent models after in utero only exposure to
16                   (semi-) relevant  concentrations of O 3.
            7.5.2   Summary and Causal Determination

17                   The 2006 O3 AQCD included toxicological evidence that acute exposures to O3 are
18                   associated with alterations in neurotransmitters, motor activity, short and long term
19                   memory, and sleep patterns. Additionally, histological signs of neurodegeneration have
20                   been observed. However, evidence regarding chronic exposure and neurobehavioral
21                   effects was not available. Recent research in the area of O3-induced neurotoxicity has
22                   included several long-term exposure studies. Notably, the first epidemiologic study to
23                   examine the relationship between O3 exposure and neurobehavioral effects observed an
24                   association between annual O3 levels and an aging-related cognitive performance decline
25                   in tests measuring coding ability and attention/short-term memory. This observation is
26                   supported by studies in rodents which demonstrate progressive oxidative stress and
27                   damage in the brain and associated decrements in behavioral tests, including those
28                   measuring memory, after subchronic exposure to 0.25 ppm O3. Additionally,
29                   neurobehavioral changes are evident in animals whose only exposure to O3 occurred in
30                   utero. Collectively, the limited epidemiologic and  toxicological evidence is coherent and
31                   suggestive of a causal relationship between O3 exposure and CNS effects.
      Draft - Do Not Cite or Quote                      7-74                                September 2011

-------
          7.6   Carcinogenic and Genotoxic Potential of Ozone
             7.6.1   Introduction

 1                   The radiomimetic and clastogenic qualities of O3, combined with its ability to stimulate
 2                   proliferation of cells in the respiratory tract, have suggested that O3 could act as a
 3                   carcinogen. However, toxicological studies of tumorigenesis in the rodent lung have
 4                   yielded mixed and often confusing results, and the epidemiologic evidence is equally
 5                   conflicted. The 2006 O3 AQCD concluded that,  "the weight of evidence from recent
 6                   animal toxicological studies and a very limited number of epidemiologic studies do not
 7                   support ambient O3 as a pulmonary carcinogen" 2(U.S. EPA, 2006b).

 8                   Multiple epidemiologic studies reported in the 2006 O3 AQCD examined the association
 9                   between O3 exposure and cancer. The largest of these studies, by Pope et al. (2002).
10                   included 500,000 adults from the American Cancer Society's (ACS) Cancer Prevention II
11                   study. In this study, no association was observed between O3 and lung cancer mortality.
12                   The Adventist Health Study of Smog (AHSMOG) also examined the association between
13                   O3 and lung cancer mortality (Abbey etal..  1999). There was a positive association
14                   between O3 levels and lung cancer mortality among men. No association was reported for
15                   women. Another study using the AHSMOG cohort assessed the risk of incident lung
16                   cancer (Beeson et al., 1998). Among males, an association with incidence of lung cancer
17                   was observed with increasing O3 concentrations. When stratified by smoking status, the
18                   association persisted among never smokers but was null for former smokers. No
19                   association was detected for females. The Six Cities Study examined various air
20                   pollutants and mortality but did not specifically explore the association between O3
21                   concentrations and lung cancer mortality due to low variability in O3 levels across the
22                   cities (Dockery et al., 1993). An ecologic study performed in Sao Paulo City, Brazil
23                   examined the correlations between O3 levels in four of the city districts and incident
24                   cancer of the larynx and lung reported in 1997 (Pereira et al., 2005). A correlation
25                   between the  average number of days O3 levels exceeded air quality standards from 1981
26                   to 1990 and cancer incidence was present for larynx cancer but not for lung cancer.

27                   Early toxicological research demonstrated lung adenoma3 acceleration in mice with daily
28                   exposure to 1 ppm over 15 months (Stokinger. 1962). Later work demonstrated a
29                   significant increase in lung tumor numbers in one strain of mouse (A/J) but not another
        2 The toxicological evidence is presented in detail in Table 6-18 on p. 6-116 of the 1996 O3 AQCD and Table AX5-13 on p.AX5-43
      of the 2006 O3 AQCD.
        3 NOTE: Although adenomas are benign, over time they may progress to become malignant, at which point they are called
      adenocarcinomas. Adenocarcinoma is the predominant lung cancer subtype in most countries, and is the only lung cancer found in
      nonsmokers. From  page 8-33 of the 1970 O3 AQCD: "No true lung cancers have been reported, however, from experimental
      exposures to either O3 alone or any other combination or ingredient of photochemical oxidants."
      Draft - Do Not Cite or Quote                       7-75                                September 2011

-------
 1                  after exposure to 0.3-0.8 ppm O3 (Lastetal.. 1987; Hassett et al.. 1985). The A/J mouse
 2                  strain is known to have a high incidence of spontaneous adenomas, and further studies
 3                  using this strain found a statistically significant increase in lung tumor incidence after a
 4                  9-month exposure to 0.5 ppm and incidence and multiplicity after a 5 month exposure to
 5                  0.12 ppm with a 4-month recovery period (Witschi et al.. 1999). However, these findings
 6                  were discounted by the study authors due to the lack of a clear dose response, and results
 7                  from the Hassett et al. 1985 and Last et al.  1987 studies were retrospectively deemed
 8                  spurious based on what appeared to be unusually low spontaneous tumor incidences in
 9                  the control groups (Witschi. 1991). A study of carcinogenicity of O3  by the National
10                  Toxicology Program (NTP. 1994) reported increased incidences of alveolar/bronchiolar
11                  adenoma or carcinoma (combined) in female B6C3FJ mice exposed  over 2 years to
12                  1.0 ppm O3, but not 0.12 or .5 ppm. No effect was detected in male mice. For a lifetime
13                  exposure to 0.5 or 1.0 ppm O3,  an increase in the number of female mice with adenomas
14                  (but not carcinomas or total neoplasms) was found. The number of total neoplasms was
15                  also unaffected in male mice, but there was a marginally increased incidence of
16                  carcinoma in males exposed to  0.5 and  1.0 ppm. Thus there was equivocal evidence of
17                  carcinogenic activity in male mice and some evidence of carcinogenic activity of O3 in
18                  females. Some semblance of a dose-response relationship was also evident in  this study.
19                  Experimental details of the NTP study are available in Table 6-19 on p. 6-121 of the 1996
20                  O3 AQCD.

21                  In Fischer-344/N rats (50 of each sex per group), neither a 2-year nor lifetime exposure to
22                  O3 ranging from 0.12 to 1.0 ppm was found to be carcinogenic (Boorman et al.. 1994).
23                  However, a marginally significant carcinogenic effect of 0.2 ppm O3 was reported in a
24                  study of male Sprague-Dawley rats exposed for 6 months (n = 50) (Monchaux et al..
25                  1996). These two studies also examined co-carcinogenicity of O3 with NNK4 (Boorman
26                  etal.. 1994) or a relatively high dose of radon (Monchaux et al.. 1996). finding no
27                  enhancement of NNK related tumors and a slight non-significant increase in tumor
28                  incidence after combined exposure with radon, respectively. Another study exploring co-
29                  carcinogenicity was conducted  in hamsters. Not only was there no enhancement of
30                  chemically induced tumors in the peripheral lung or nasal cavity, but results suggested
31                  that O3 could potentially delay  or inhibit tumor development (Witschi et al.. 1993). Thus
32                  there is no concrete evidence that O3 can act as a co-carcinogen.

33                  Immune surveillance is an important defense against cancer, and it should be noted that
34                  natural killer (NK) cells, which destroy tumor cells in the lung, appear to be inhibited by
35                  higher doses of O3 and either unaffected or stimulated at lower doses (Section 6.2.5.4,
       4 4-(N-nitrosomethylamino)-1 -(3-pyridyl)-1 -butanone
      Draft - Do Not Cite or Quote                      7-76                                September 2011

-------
 1                  Infection and Adaptive Immunity). This aspect of tumorigenesis adds yet another layer of
 2                  complexity which may be reflected by conflicting results across studies.

 3                  The following sections will examine epidemiologic studies of cancer incidence and
 4                  mortality and toxicological studies that have been published since the 2006 O3 AQCD.
 5                  An epidemiologic study has been published with cancer as the outcome; most
 6                  epidemiologic studies examine markers of exposure or susceptibility.
            7.6.2   Lung Cancer Incidence and Mortality

 7                  A recent re-analysis of the full ACS CPSII cohort by the Health Effects Institute is the
 8                  only epidemiologic study that has explored the association between O3 and cancer
 9                  mortality since the last O3 AQCD. Krewski et al. (2009) conducted an extended follow-
10                  up of the cohort (1982-2000). Mean O3 levels [obtained from the Aerometric Information
11                  Retrieval System (AIRS) for 1980] were 22.91 ppb for the full year and 30.15 ppb for the
12                  summer months (April-September). No association was reported between lung cancer
13                  mortality and O3 (HR=1.00 [95% CI:  0.96-1.04] per 10 ppb O3). Additionally, no
14                  association was observed when O3 was restricted to the summer months.  There was also
15                  no association present in a sub-analysis of the cohort examining the relationship between
16                  O3 and lung cancer mortality in the Los Angeles area.

17                  Since the 2006 O3 AQCD, two toxicological studies have examined potential
18                  carcinogenicity of O3 (Kim and Cho.  2009a. b). Looking across both studies, which used
19                  the same mouse strain as the National Toxicology Program study described above (NTP.
20                  1994). 0.5 ppm O3 alone or in conjunction with chemical tumor inducers did not enhance
21                  lung tumor incidence in males or females. However, a 10% incidence of oviductal
22                  carcinoma was observed in mice exposed to 0.5 ppm O3 for 16 weeks. The implications
23                  of this observation are unclear, particularly in light  of the lack of statistical information
24                  reported. Additionally, there is no mention of oviductal carcinoma after 32 weeks of
25                  exposure, and no oviductal carcinoma was observed after one year of exposure. The NTP
26                  study did not report any increase in tumors at extrapulmonary sites.
            7.6.3   DMA Damage

27                  The potential for genotoxic effects relating to O3 exposure was predicted from the
28                  radiomimetic properties of O3. The decomposition of O3 in water produces OH and HO2
29                  radicals, the same species that are generally considered to be the biologically active
30                  products of ionizing radiation. Ozone has been observed to cause degradation of DNA in
31                  a number of different models and bacterial strains. The toxic effects of O3 have been

      Draft - Do Not Cite or Quote                       7-77                                September 2011

-------
 1                   generally assumed to be confined to the tissues directly in contact with the gas, such as
 2                   the respiratory epithelium. Due to the highly reactive nature of O3, little systemic
 3                   absorption is predicted. Zelac et al. (1971a. b), however, reported a significant increase in
 4                   chromosome aberrations in peripheral blood lymphocytes from Chinese hamsters
 5                   exposed to 0.2 ppm for 5 hours. Other in vivo exposure studies found increased DNA
 6                   strand breaks in respiratory cells from guinea pigs (Ferng et al.. 1997) and mice
 7                   (Bornholdt et al., 2002) but only with exposure to higher doses of O3 (1 ppm for 72 hours
 8                   and 1 or 2 ppm for 90 minutes, respectively). In other studies there were no observations
 9                   of chromosomal aberrations in germ cells, but mutagenic effects have been seen in
10                   offspring of mice exposed to 0.2 ppm during gestation (blepharophimosis or dysplasia of
11                   the eyelids). The overall evidence for mutagenic activity from in vitro studies is positive,
12                   and in the National Toxicology Program report described above, O3 was found to be
13                   mutagenic in Salmonella, with and without S9 metabolic activation. No new
14                   toxicological studies of DNA damage have become available since the 2006 O3 AQCD.

15                   A number of epidemiologic studies looked at the association between O3 and DNA and
16                   cellular level damages. These changes may be relevant to mechanisms leading to cancers
17                   development and serve as early indicators of elevated risk of mutagenicity.

18                   Two studies performed in California examined cytogenetic damage in relation to O3
19                   exposures. Huen et al. (2006) examined cytogenetic damage among African American
20                   children and their mothers in Oakland,  CA. Increased O3 (mean monthly 8-h O3
21                   concentrations ranged from about 30 ppb in April to 14 ppb in November) was associated
22                   with increased cytogenetic damage (micronuclei frequency among lymphocytes and
23                   buccal cells) even after adjustment for household/personal smoking  status and distance-
24                   weighted traffic density. Chen et al. (2006a) recruited college students at the University
25                   or California, Berkeley who reported never smoking and compared their levels of
26                   cytogenetic damage (micronuclei frequency from buccal cells) in the spring and fall.
27                   Cytogenetic damage was greater in the fall, which the authors attributed to the increase in
28                   O3 over the summer. However, O3 levels over 2, 7, 10, 14, or 30 days (concentrations not
29                   given) before collection of buccal cells did not correlate with cytogenetic damage.
30                   Estimated lifetime  O3 exposure was also not correlated with cytogenetic damage.
31                   Additionally, the authors exposed a subset of the students (n=15) to  200 ppb O3 for
32                   4 hours while the students exercised intermittently. Ozone was found to be associated
33                   with an increase in cytogenetic damage in degenerated cells but not in normal cells 9-
34                   10 days after exposure. Increased cytogenetic damage was also noted in peripheral blood
35                   lymphocytes collected 18 hours after exposure.

36                   A study performed in Mexico recruited 55 male workers working indoors (n=27) or
37                   outdoors (n=28) in Mexico City or Puebla, Mexico in order to study the relationship
      Draft - Do Not Cite or Quote                       7-78                                September 2011

-------
 1                  between O3 and DNA damage (detected from peripheral blood samples using the Comet
 2                  assay) (Tovalin et al.. 2006). The median estimated daily O3 concentrations were
 3                  estimated to be 28.5 ppb for outdoor workers and 5.1 ppb for indoor workers in
 4                  Mexico City and 36.1 ppb for outdoor workers and 19.5 ppb for indoor workers in
 5                  Puebla. Overall, a positive correlation between O3 levels and DNA damage was
 6                  observed. However, when examining the relationship by city and workplace, only DNA
 7                  damage in outdoor workers in Mexico City remained correlated with O3 levels.

 8                  Three studies examining the relationship between O3 and DNA-level damage have been
 9                  performed in Europe. The largest of these was the GenAir case-control study, which was
10                  nested within the European Prospective Investigation into Cancer and Nutrition (EPIC)
11                  study, and included individuals recruited between 1993 and 1998 from ten European
12                  countries. Only non-smokers (must not have smoked for at least 10 years prior to
13                  enrollment) were enrolled in the study. The researchers examined DNA adduct levels
14                  (DNA bonded to cancer-causing chemicals) and their relationship with O3 concentrations
15                  (concentrations not given) (Peluso  M Hainaut et al.. 2005). A positive association was
16                  seen between DNA adduct levels and O3 concentrations from 1990-1994 but not O3
17                  levels from 1995-1999. In adjusted analyses with DNA adduct levels dichotomized as
18                  high and low (detectable versus non-detectable), the OR was 1.97 (95% CI: 1.08, 3.58)
19                  when comparing the upper tertile of O3 concentration to the lower two tertiles. Two other
20                  European studies were conducted in Florence, Italy. The most recent of these enrolled
21                  individuals from the EPIC study into a separate study between March and September of
22                  1999 (Palli et al.. 2009). The purpose of the study was to examine oxidative DNA
23                  damage (determined by Comet assay using blood lymphocytes) in association with
24                  varying periods of O3 exposure. The researchers observed that longer periods of high O3
25                  exposure (concentrations not given) were more strongly correlated with oxidative DNA
26                  damage than shorter exposures (i.e., the rho [p-value] was 0.26 [0.03] for 0-10 days and
27                  0.35 [0.002] for 0-90 days).  This correlation was stronger among men compared to
28                  women. The correlations for all time periods had p-values <0.05 for ex- and never-
29                  smokers. For current smokers, the correlation was only observed among time periods <
30                  25 days. When adjusted for age, gender, smoking history, traffic pollution exposure,
31                  period of blood draw, and area of residence, the association between O3 levels and
32                  oxidative DNA damage was positive for O3 levels 0-60 days, 0-75 days, and 0-90 days
33                  prior to blood draw. Positive, statistically significant associations were not observed
34                  among shorter time periods. The other study performed in Florence recruited healthy
35                  volunteers who reported being non-smokers or light smokers (Giovannelli et al.. 2006).
36                  The estimated O3 levels during the study ranged from approximately 4-40 ppb for 3-day
37                  avgs, 5-35 ppb for  7-day avgs, and 7.5-32.5 ppb for 30-day avgs. Ozone concentrations
38                  were correlated with DNA strand breaks (measured from blood lymphocytes) over longer
39                  exposure periods (p-value: 0.002 at 30 days, p-value: 0.04 at 7 days; p-value: 0.17 at

      Draft - Do Not Cite or Quote                      7-79                                September 2011

-------
 1                   3 days). This association was robust to control for temperature, solar radiation, gender,
 2                   and age. No association was seen between O3 concentrations and measures of oxidative
 3                   DNA damage at 3, 7, or 30 days.
            7.6.4   Summary and Causal Determination

 4                   The 2006 O3 AQCD reported that evidence did not support ambient O3 as a pulmonary
 5                   carcinogen. Since the 2006 O3 AQCD, very few epidemiologic and toxicological studies
 6                   have been published that examine O3 as a carcinogen, but collectively, study results
 7                   indicate that O3 may contribute to DNA damage. Overall, the evidence is inadequate to
 8                   determine if a causal relationship exists between ambient O3 exposures and
 9                   cancer.
          7.7    Mortality

10                  A limited number of epidemiologic studies have assessed the relationship between long-
11                  term exposure to O3 and mortality in adults. The 2006 O3 AQCD concluded that an
12                  insufficient amount of evidence existed "to suggest a causal relationship between chronic
13                  O3 exposure and increased risk for mortality in humans" (U.S. EPA. 2006b). In addition
14                  to the infant mortality studies discussed in Section 7.4.9, additional studies have been
15                  conducted among adults since the last review; an ecologic study that finds no association
16                  between mortality and O3, several reanalyses of the ACS cohort, one of which
17                  specifically points to a relationship between long-term O3 exposure and an increased risk
18                  of respiratory mortality, and a study of four cohorts of persons with potentially
19                  predisposing conditions. These studies supplement the evidence from long-term cohort
20                  studies characterized in previous reviews of O3 and are summarized here briefly.

21                  In the Harvard Six Cities Study (Dockery et al.. 1993), adjusted mortality rate ratios were
22                  examined in relation to long-term mean O3 concentrations in six cities: Topeka, KS; St.
23                  Louis, MO; Portage, WI; Harriman, TN; Steubenville, OH; and Watertown, MA. Mean
24                  O3 concentrations from 1977 to 1985 ranged from 19.7 ppb in Watertown to 28.0 ppb in
25                  Portage. Long-term mean O3 concentrations were not found to  be associated with
26                  mortality in the six cities. However, the authors noted that "The small differences in O3
27                  levels among the (six) cities limited the power of the study to detect associations between
28                  mortality and O3 levels." In addition, while total and cardio-pulmonary mortality were
29                  considered in this study, respiratory mortality  was not specifically considered.

30                  In a subsequent large prospective cohort study of approximately 500,000 U.S. adults,
31                  Pope et al.  (2002) examined the effects of long-term exposure to air pollutants on

      Draft - Do Not Cite or Quote                      7-80                               September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
1 1
                    mortality (American Cancer Society, Cancer Prevention Study II). All-cause,
                    cardiopulmonary, lung cancer and other mortality risk estimates for long-term O3
                    exposure are shown in Figure 7-5. While consistently positive associations were not
                    observed between O3 and mortality (effect estimates labeled A in Figure 7-5) , the
                    mortality risk estimates were larger in magnitude when analyses considered more
                    accurate exposure metrics, increasing when the entire period was considered (effect
                    estimates labeled B in Figure 7-5) and becoming marginally significant when the
                    exposure estimate was restricted to the summer months (July to September; effect
                    estimates labeled C in Figure 7-5), especially when considering cardiopulmonary deaths.
                    In contrast, consistent positive and significant effects of PM2 5 were observed for both
                    lung cancer and cardio-pulmonary mortality.
All Cause Cardiopulmonary
Mortality Mortality
^ ' ;• [ ] [ — 1
^ > i i i „ . a
§ r J
i ' . , i o ° | ? 1
1 (
ABC A B C
., . „ . „ „ .. Number of
Years of Data Collection .. . ... .
Metropolitan Areas
A 1980-1981 134
B 1982-1998 119
C 1982-1998 (July -Sept) 134
I ,ung Cancer
Mortality
I , I
-I ' o
1 °
A B C
Number of Participants
(in thousands)
559
525
557
All Other Ci
Mortalit
I
I f\
\J
A B
1-h Max Os Mean
47.9(11.0)
45.5 (7.3)
59.7(12.8)
a uses
V
I
0
C
(SD)



       Source: Reprinted with permission of American Medical Association, Pope et al. (2002).

      Figure 7-5     Adjusted ozone-mortality relative risk estimates (95% Cl) by time
                      period of analysis per subject-weighted mean Os  concentration in
                      the Cancer Prevention Study II by the American Cancer Society.
12
13
14
15
16
                    A study by Abbey et al. (1999) examined the effects of long-term air pollution exposure,
                    including O3, on all-cause (n = 1,575), cardiopulmonary (n = 1,029), nonmalignant
                    respiratory (n = 410), and lung cancer (n = 30) mortality in the long-term prospective
                    Adventist Health  Study of Smog (AHSMOG) of 6,338 nonsmoking, non-Hispanic white
                    individuals living in California. A particular strength of this study was the extensive
      Draft - Do Not Cite or Quote
                                                  7-81
September 2011

-------
 1                   effort devoted to assessing long-term air pollution exposures, including interpolation to
 2                   residential and work locations from monitoring sites over time and space. No associations
 3                   with long-term O3 exposure were observed for all cause, cardiopulmonary, and
 4                   nonmalignant respiratory mortality. In a follow-up, Chen et al. (2005) utilized data from
 5                   the AHSMOG study and reported no evidence of associations between long-term O3
 6                   exposure (mean O3 concentration 26.2 ppb) and fatal coronary heart disease. Thus, no
 7                   association of chronic O3 exposure with mortality outcomes has been detected in this
 8                   study.

 9                   Lipfert et al. (2003. 2000) reported positive effects on all-cause mortality for peak O3
10                   exposures (95th percentile  levels) in the U.S. Veterans Cohort study of approximately
11                   50,000 middle-aged men recruited with a diagnosis of hypertension. The actual analysis
12                   involved smaller subcohorts based on exposure and mortality follow-up periods. Four
13                   separate exposure periods were associated with three mortality follow-up periods. For
14                   concurrent exposure periods, peak O3 was positively associated with all-cause mortality,
15                   with a 9.4% (95% CI: 0.4,  18.4) excess risk per mean 95th percentile O3 less estimated
16                   background level (not stated). "Peak" refers, in this case, to the  95th percentile of
17                   the hourly measurements, averaged by year and county. In a further analysis, Lipfert et al.
18                   (2003) reported the strongest positive association for concurrent exposure to peak O3  for
19                   the subset of subjects with low diastolic blood pressure during the 1982 to 1988 period.
20                   Two more recent studies of this cohort focused specifically on traffic density (Lipfert et
21                   al.. 2006a; 2006b). Lipfert (2006b) concluded that: "Traffic density is seen to be a
22                   significant and robust predictor of survival in this cohort, more so than ambient air
23                   quality, with the possible exception of O3," reporting a significant O3 effect even with
24                   traffic density included in the model: RR=1.080 per 40 ppb peak O3 (95% CI: 1.019,
25                   1.146). However, in Lipfert (2006a), which considers only the EPA Speciation Trends
26                   Network (STN) sites, O3 drops to non-significant predictor of total mortality for this
27                   cohort. The authors acknowledge that: "Peak O3  has been important in analyses of this
28                   cohort for previous periods, but in the STN data set, this variable has limited range and
29                   somewhat lower values and its small coefficient of variation results in a relatively large
30                   standard error." The restriction to subjects near STN sites likely reduced the power of this
31                   analysis, though the size  of the remaining subjects considered was not reported in this
32                   paper. In addition, these various Veterans Cohort studies considered only total mortality,
33                   and did not consider mortality on a by-cause basis.

34                   An ecological study in Brisbane, Australia used a geospatial approach to analyze the
35                   association of long-term exposure to gaseous air  pollution with cardio-respiratory
36                   mortality, in the period 1996-2004 (Wang et al.. 2009c). A generalized estimating
37                   equations model  was  employed to  investigate the impact of NO2, O3 and SO2, but PM
38                   was not addressed. The results indicated that long-term exposure to O3 was not
      Draft - Do Not Cite or Quote                       7-82                                 September 2011

-------
 1                   associated with cardio-respiratory mortality, but the fact that this study considered only
 2                   one city, and that the range of O3 exposure across that city (23.7-35.6 ppb) was low and
 3                   slight in variation in comparison to the range of other pollutants across the city, limited
 4                   study power. In addition, confounding factors (e.g., smoking) could not be addressed at
 5                   the individual level in this ecological study. Respiratory mortality was not evaluated
 6                   separately.

 7                   A recent study by Zanobetti and Schwartz (In Press) examined whether year-to-year
 8                   variations in 8-h mean daily O3 concentrations for the summer (May-September) around
 9                   their city-specific long-term trend were associated with year-to-year variations in
10                   mortality around its long-term trend. This association was examined among Medicare
11                   participants with potentially predisposing conditions, including COPD, diabetes, CHF,
12                   and MI, defined as patients discharged alive after an emergency admission for one of
13                   these four conditions. The analyses was repeated in 105 cities using available data from
14                   1985 through 2006, and the results were combined using methods previously employed
15                   by these authors (Zanobetti et al.. 2008; Zanobetti and Schwartz. 2007). This study
16                   design eliminated potential confounding by factors that vary across city, which is a
17                   common concern in most air pollution cohort studies, and also avoided both  confounding
18                   by cross-sectional factors that vary by city and the short-term factors that confound daily
19                   time-series studies, but are not present in annual analyses. The average 8-h mean daily
20                   summer O3 concentrations ranged from 15.6 ppb (Honolulu, HI) to 71.4 ppb
21                   (Bakersfield, CA) for the 105 cities. The  authors  observed associations between yearly
22                   fluctuations in summer O3 concentrations and mortality in each of the four cohorts; the
23                   hazard ratios (per 10 ppb increment) were 1.12 (95% CI: 1.06, 1.17) for the CHF cohort,
24                   1.19 (95% CI 1.12, 1.25) for the MI cohort, 1.14  (95% CI: 1.10, 1.21) forthe diabetes
25                   cohort, and 1.14 (95% CI: 1.08,  1.19)  for the COPD cohort. A key advantage to this  study
26                   is that fluctuations from summer to summer in O3 concentrations around long-term level
27                   and trend in a specific  city are unlikely to be correlated with most other predicators of
28                   mortality risk, except for temperature, which was controlled for in  the regression. Key
29                   limitations of the study were the inability to control for PM2 5, since it was not reliably
30                   measured in these cities until 1999, and the inability to separate specific causes of death
31                   (e.g., respiratory, cardiovascular), since Medicare does not provide the underlying cause
32                   of death.

33                   In the most recent follow-up analyses  of the ACS cohort (Jerrett et al.. 2009; Smith et al..
34                   2009a), the effects of long-term exposure to O3 were evaluated alone, as well as in
35                   copollutant models with PM2 5 and components of PM2 5. Jerrett et al. (2009) utilized the
36                   ACS cohort with data from 1977 through 2000 (mean O3 concentration ranged from 33.3
37                   to  104.0 ppb) and subdivided cardiopulmonary deaths into respiratory and cardiovascular,
38                   separately, as opposed to combined into one category, as was done by Pope et al. (2002).
      Draft - Do Not Cite or Quote                       7-83                                 September 2011

-------
 1                   Increases in exposure to O3 were associated with an elevated risk of death from
 2                   cardiopulmonary, cardiovascular, ischemic heart disease, and respiratory causes.
 3                   Inclusion of PM2 5 concentrations measured in 1999-2000 as a copollutant attenuated the
 4                   association with O3 for all end points except death from respiratory causes, for which a
 5                   significant association persisted (Table 7-12). The association between increased O3
 6                   concentrations and increased risk of death from respiratory causes was insensitive to the
 7                   use of a random-effects survival model allowing for spatial clustering within the
 8                   metropolitan area and state of residence, and adjustment for several ecologic variables
 9                   considered individually. Subgroup analyses showed that temperature and region of
10                   country,  but not sex, age at enrollment, body-mass index, education, or PM25
11                   concentration, modified the effects of O3 on the risk of death from respiratory causes
12                   (i.e., risks were higher at higher temperature, and in the Southeast, Southwest, and Upper
13                   Midwest). Ozone threshold analyses indicated that the threshold model was not a better
14                   fit to the data (p > 0.05) than a linear representation of the overall O3-mortality
15                   association.  Overall, this new analysis indicates that long-term exposure to PM2 5
16                   increases risk of cardiac death, while long-term exposure to O3 is specifically associated
17                   with an increased risk of respiratory death, and suggests that combining cardiovascular
18                   and respiratory causes of mortality into one category for analysis may obscure any effect
19                   that O3 may have on respiratory-related causes of mortality.
      Table 7-12     Relative risk (and 95% Cl) of death attributable to a 10-ppb change
                       in the ambient Os concentration*
Cause of Death
Any Cause
Cardiopulmonary
Respiratory
Cardiovascular
Ischemic Heart Disease
O3 (96 MSAs)
1 .001 (0.996, 1 .007)
1.014(1.007,1.022)
1.029(1.010,1.048)
1.011 (1.003,1.023)
1.015(1.003,1.026)
O3 (86 MSAs)
1.001 (0.996,1.007)
1.016(1.008,1.024)
1 .027 (1 .007, 1 .046)
1.014(1.005,1.023)
1.017(1.006,1.029)
O3 +PM2.5 (86 MSAs)
0.989 (0.981 , 0.996)
0.992(0.982,1.003)
1.040(1.013,1.067)
0.983 (0.971 , 0.994)
0.973 (0.958, 0.988)
      * Ozone concentrations were measured from April to September during the years from 1977 to 2000, with follow-up from 1982 to 2000; changes in
      the concentration of PM2.5 of 10 ug per cubic meter were recorded for members of the cohort in 1999 and 2000.
      Source: Reprinted with permission of Massachusetts Medical Society (Jerrettetal.. 2009)
20                   In a similar analysis, Smith et al. (2009a) used data from 66 MSAs in the ACS cohort to
21                   examine the association of O3 concentrations during the warm season and all-cause and
22                   cardiopulmonary mortality. Mortality effects were estimated in single pollutant and
23                   copollutant models, adjusting for two PM2 5 constituents, sulfate and EC. When all-cause
24                   mortality was investigated, there was a 0.8% (95% CI: -0.31, 1.9) increase associated
25                   with a 10 ppb increase in O3 concentration. This association was diminished when sulfate
26                   or EC were included in the model. There was a 2.48% (95% CI: 0.74, 4.3) increase in
27                   cardiopulmonary mortality associated with a 10 ppb increase in O3 concentration. The
      Draft - Do Not Cite or Quote                       7-84                                 September 2011

-------
 1                   cardiopulmonary association was robust to adjustment for sulfate, and diminished, though
 2                   still positive, after adjustment for EC (1.63% increase; 95% CI: -0.41, 3.7). Smith et al.
 3                   (2009a) did not specifically separate out cardiovascular and respiratory causes of death
 4                   from the cardiopulmonary category, as was done by Jerrett et al. (2009).
            7.7.1    Summary and Causal Determination

 5                   The 2006 O3 AQCD concluded that an insufficient amount of evidence existed "to
 6                   suggest a causal relationship between chronic O3 exposure and increased risk for
 7                   mortality in humans" (U.S. EPA. 2006b). Several additional studies have been conducted
 8                   since the last review, including an ecologic study that finds no association between
 9                   mortality and O3 (Wang et al.. 2009c).  a study of four cohorts of Medicare enrollees with
10                   potentially predisposing conditions that observes associations between O3 and mortality
11                   among each of the cohorts (Zanobetti and Schwartz. In Press), and reanalyses of the ACS
12                   cohort that provide weak evidence for an association with cardiopulmonary mortality
13                   (Smith et al.. 2009a) and specifically point to a relationship between long-term O3
14                   exposure and an increased risk of respiratory mortality (Jerrett et al.. 2009). The findings
15                   from the Jerrett et al. (2009) study are consistent and coherent with the evidence from
16                   epidemiologic, controlled human exposure, and animal toxicological studies for the
17                   effects of short- and long-term exposure to O3 on respiratory effects. Additionally, the
18                   evidence for short- and long-term respiratory morbidity provides biological plausibility
19                   for mortality due to respiratory disease. Collectively, the evidence is suggestive of a
20                   causal relationship between long-term O3 exposures and mortality.
          7.8    Overall Summary

21                   The evidence reviewed in this chapter describes the recent findings regarding the health
22                   effects of long-term exposure to ambient O3 concentrations. Table 7-13 provides an
23                   overview of the causal determinations for each of the health categories evaluated.
      Draft - Do Not Cite or Quote                      7-85                                September 2011

-------
Table 7-13      Summary of causal determinations for long-term exposures to
                  ozone
Health Category
   Causal Determination
Respiratory Effects
   Likely to be a causal relationship
Cardiovascular Effects
   Suggestive of a causal relationship
Reproductive and Developmental Effects
   Suggestive of a causal relationship
Central Nervous System Effects
   Suggestive of a causal relationship
Carcinogenicity and Genotoxicity
   Inadequate to infer a causal relationship
Mortality
   Suggestive of a causal relationship
    7.9    References

Abbey, DE: Nishino, N: McDonnell, WF: Burchette, RJ: Knutsen, SF: Beeson, Wl_: Yang, JX. (1999). Long-term
        inhalable particles and other air pollutants related to mortality  in nonsmokers. Am J Respir Crit Care
        Med 159: 373-382.
Akinbami, LJ: Lynch, CD: Parker, JD: Woodruff, TJ. (2010). The association between childhood asthma
        prevalence and  monitored air pollutants in metropolitan areas, United States, 2001-2004. Environ Res
        110: 294-301. http://dx.doi.Org/10.1016/i.envres.2010.01.001.
Angoa-Perez, M: Jiang, H: Rodrfguez, Al: Lemini, C: Levine, RA: Rivas-Arancibia, S. (2006). Estrogen
        counteracts ozone-induced oxidative stress and nigral neuronal death. Neuroreport 17: 629-633.
Arito.  H: Uchivama.  I: Yokovama. E. (1992). Acute effects of ozone on  EEC activity, sleep-wakefulness and
        heart rate in rats. Ind Health 30: 23-34.
Auten. RL: Potts. EN: Mason. SN: Fischer. B: Huang. Y: Foster. WM. (2009). Maternal exposure to particulate
        matter increases postnatal ozone-induced airway hyperreactivity in juvenile mice. Am J Respir Crit
        Care Med 180: 1218-1226. http://dx.doi.org/10.1164/rccm.200901-0116OC.
Barry. BE: Miller. FJ: Crapo. JD. (1983). Alveolar epithelial injury caused by inhalation of 025 ppm of ozone. In
        74th Meeting of the Air Pollution Control Association (Vol. NJ). Pittsburgh, PA: Air Pollution Control
        Association.
Barry. BE: Miller. FJ: Crapo. JD. (1985). Effects of inhalation of 0.12 and 0.25 parts per million ozone on the
        proximal alveolar region of juvenile and adult rats. Lab Invest  53: 692-704.
Beeson, WL: Abbey, DE: Knutsen, SF. (1998). Long-term concentrations of ambient air pollutants and incident
        lung cancer in California adults: Results from the AHSMOG study. Environ Health Perspect  106: 813-
        823.
Berhane, K:  Gauderman, WJ: Stram, DO: Thomas,  DC. (2004). Statistical issues in studies of the long-term
        effects of air pollution: The Southern California Children's Health Study. Stat Sci 19: 414-449.
        http://dx.doi.org/10.1214/088342304000000413.
Bignami, G:  Musi, B: Dell'Omo, G: Laviola, G: Alleva, E. (1994). Limited effects of ozone exposure during
        pregnancy on physical and neurobehavioral development of CD-1 mice. Toxicol Appl Pharmacol 129:
        264-271.  http://dx.doi.org/10.1006/taap. 1994.1251.
Bobak, M. (2000). Outdoor air pollution, low birth weight, and prematurity. Environ Health Perspect 108: 173-
        176.
Boorman. GA: Hailev. R: Grumbein. S: Chou. BJ: Herbert. RA: Goehl.  T: Mellick. PW: Rovcroft. JH: Haseman.
        JK:  Sills,  R. (1994). Toxicology and carcinogenesis studies of  ozone and ozone 4-(N-
        nitrosomethylamino)-1-(3-pyridyl)-1-butanone  in Fischer-344/N rats. Toxicol Pathol 22: 545-554.
Bornholdt. J: Dvbdahl. M: Vogel. U: Hansen. M: Loft. S: Wallin. H. (2002). Inhalation of ozone induces DMA
        strand breaks and inflammation in mice. DMA  Repair 520: 63-71.
Draft - Do Not Cite or Quote
7-86
September 2011

-------
Boussouar. A: Araneda. S: Hamelin. C: Soulage. C: Kitahama. K: Dalmaz. Y. (2009). Prenatal ozone exposure
        abolishes stress activation of Fos and tyrosine hydroxylase in the nucleus tractus solitarius of adult rat.
        Neurosci Lett 452: 75-78.
Brauer. M: Lencar. C: Tamburic. L: Koehoorn. M: Demers. P: Karr. C. (2008). A cohort study of traffic-related air
        pollution impacts on birth outcomes. Environ Health Perspect 116: 680-686.
Breton. CV: Salam. MT: Vora. H: Gauderman. WJ: Gilliland. FD. (2011). Genetic variation in the glutathione
        synthesis pathway, air pollution, and children's lung function growth. Am J Respir Crit Care Med 183:
        243-248. http://dx.doi.org/10.1164/rccm.201006-0849OC.
Calderon-Garciduenas. L: Mora-Tiscareno. A: Fordham. LA: Chung. CJ: Valencia-Salazar. G: Flores-Gomez. S:
        Solt. AC: Gomez-del Campo. A: Jardon-Torres. R: Henriguez-Roldan. C: Hazucha. MJ: Reed. W.
        (2006). Lung radiology and pulmonary function of children chronically exposed to air pollution. Environ
        Health Perspect 114: 1432-1437.
Campos-Bedolla. P: Vargas. MH: Montano. LM. (2002). Effect of acute ozone exposure on pregnant rat uterus
        contractile responses. Reprod Toxicol 16: 269-273.
Carbaial-Arrovo. L: Miranda-Soberanis. V: Medina-Ramon. M: Roias-Bracho. L: Tzintzun. G: Solfs-Gutierrez. P:
        Mendez-Ramfrez. I: Hurtado-Dfaz. M: Schwartz. J: Romieu. I. (2011). Effect of PM10 and O3 on infant
        mortality among residents  in the Mexico City Metropolitan Area: A case-crossover analysis, 1997-2005.
        J Epidemiol Community Health 65: 715-721. http://dx.doi.orq/10.1136/iech.2009.101212.
Carey. SA: Minard. KR: Trease. LL: Wagner. JG: Garcia. GJ: Ballinger. CA: Kimbell. JS: Plopper. CG: Corlev.
        RA: Postlethwait, EM: Harkema, JR: Einstein, DR. (2007). Three-dimensional mapping of ozone-
        induced injury in the nasal  airways of monkeys using magnetic resonance imaging and morphometric
        techniques. Toxicol Pathol  35: 27-40. http://dx.doi.org/10.1080/01926230601072343.
Carey, SA: Ballinger, CA: Plopper, CG: McDonald, RJ: Bartolucci, AA: Postlethwait, EM: Harkema, JR. (2011).
        Persistent rhinitis and epithelial remodeling induced by cyclic ozone exposure in the nasal airways of
        infant monkeys. Am J  Physiol Lung Cell Mol Physiol 300: L242-L254.
        http://dx.doi.org/10.1152/aiplung.00177.2010.
Catalano. PJ:  Rogus. J: Ryan. LM.  (1995a). Consequences of prolonged inhalation of ozone on F344/N rats:
        collaborative studies Part X: robust composite scores based on median polish analysis.
Catalano, PJ:  L-YL, C:  Harkema, JR: Kaden, DA: Last, JA: Mellick, PW: Parks, WC: Pinkerton, KE:
        Radhakrishnamurthv. B: Ryan. LM: Szarek. JL. (1995b). Consequences of prolonged inhalation of
        ozone on F344/N rats: Collaborative studies Part XI: Integrative summary. Cambridge, MA: Health
        Effects Institute.
Chang. L: Miller. FJ: Ultman. J: Huang. Y: Stockstill. BL: Grose. E: Graham. JA: Ospital. JJ: Crapo. JD. (1991).
        Alveolar epithelial cell injuries by subchronic exposure to low concentrations of ozone correlate with
        cumulative exposure. Toxicol Appl Pharmacol 109: 219-234. http://dx.doi.org/10.1016/0041-
        008X(91)90170-J.
Chang. L. -Y: Huang. Y: Stockstill. BL: Graham. JA: Grose. EC: Menache. MG: Miller.  FJ: Costa. PL: Crapo. JD.
        (1992). Epithelial injury and interstitial fibrosis in the proximal alveolar regions of rats chronically
        exposed to a simulated pattern of urban ambient ozone. Toxicol Appl Pharmacol 115: 241-252.
        http://dx.doi. org/10.1016/0041-008X(92)90329-Q.
Chang, L, -Y: Stockstill, BL: Menache,  MG: Mercer, RR: Crapo, JD.  (1995). Consequences of prolonged
        inhalation of ozone on F344/N rats: Collaborative studies - Part VIII: morphometric analysis of structural
        alterations in alveolar regions.
Charpin, D: Pascal, L: Birnbaum, J: Armengaud, A: Sambuc, R: Lanteaume, A: Vervloet, D. (1999). Gaseous air
        pollution and atopy. Clin Exp Allergy 29: 1474-1480.
Chen. C: Ariomandi. M: Qin. H: Balmes. J: Tager. I: Holland. N. (2006a). Cytogenetic damage in buccal epithelia
        and peripheral  lymphocytes of young healthy individuals exposed to ozone. Mutagenesis 21: 131-137.
        http://dx.doi. org/10.1093/mutage/gel007.
Chen. C: Ariomandi. M: Balmes. J:  Tager. I: N.  H. (2007). Effects of chronic and acute ozone exposure on lipid
        peroxidation and antioxidant capacity in healthy young adults. Environ Health Perspect 115: 1732-
        1737. http://dx.doi.org/10.1289/ehp.10294.
Chen. J. -C: Schwartz.  J. (2009). Neurobehavioral effects of ambient air pollution on cognitive performance in
        US adults. Neurotoxicology 30: 231-239. http://dx.doi.Org/10.1016/i.neuro.2008.12.011.
Draft - Do Not Cite or Quote                        7-87                                  September 2011

-------
Chen. L: Yang. W: Jennison. BL: Goodrich. A: Omave. ST. (2002). Air pollution and birth weight in northern
        Nevada, 1991-1999. InhalToxicol 14: 141-157.
Chen. L: Bell. EM: Caton. AR: Druschel. CM: Lin. S. (201 Oa). Residential mobility during pregnancy and the
        potential for ambient air pollution exposure misclassification. Environ Res 110: 162-168.
        http://dx.doi.0rg/10.1016/i.envres.2009.11.001.
Chen. LH: Knutsen. SF: Shavlik. D: Beeson. WL: Petersen.  F: Ghamsarv. M: Abbey.  D. (2005). The association
        between fatal coronary heart disease and  ambient particulate air pollution: Are females at greater risk?
        Environ Health Perspect 113: 1723-1729.
Chuang. GC: Yang. Z: Westbrook. DG: Pompilius.  M: Ballinger. CA: White. RC: Krzvwanski. DM:  Postlethwait.
        EM: Ballinger. SW. (2009). Pulmonary ozone exposure induces vascular dysfunction, mitochondrial
        damage, and atherogenesis. Am J Physiol Lung Cell Mol Physiol 297: L209-L216.
        http://dx.doi. orq/10.1152/ajplung.00102.2009.
Chuang. KJ: Yan. YH: Chiu. SY:  Cheng. TJ. (2011). Long-term air pollution exposure and risk factors for
        cardiovascular diseases among the elderly in Taiwan. Occup Environ Med 68: 64-68.
        http://dx.doi.ora/10.1136/oem.2009.052704.
Clark. NA: Demers. PA: Karr. CJ: Koehoorn. M:  Lencar. C: Tamburic. L: Brauer. M. (2010). Effect  of early life
        exposure to air pollution on development of childhood asthma. Environ Health Perspect  118: 284-290.
        http://dx.doi.org/10.1289/ehp.0900916.
Cole. MP: Freeman. BA. (2009). Promotion of cardiovascular disease by exposure to the air pollutant ozone.
        Am J Physiol Lung Cell  Mol Physiol 297: L209-L216.
Conti. DV: Cortessis. V: Molitor. J: Thomas. DC. (2003). Bayesian modeling of complex metabolic pathways.
        Hum Hered 56: 83-93. http://dx.doi.ora/10.1159/000073736.
Dadvand, P: Rankin, J: Rushton, S: Pless-Mulloli, T (2011).  Ambient air pollution  and congenital  heart disease:
        A register-based study. Environ Res 111: 435-441. http://dx.doi.Org/10.1016/i.envres.2011.01.022.
Dales. R: Burnett. RT: Smith-Doiron. M: Stieb. DM: Brook. JR. (2004). Air pollution and sudden infant death
        syndrome. Pediatrics 113: 628-631.
Darrow.  LA:  Klein. M: Flanders. WD: Waller. LA: Correa. A: Marcus. M:  Mulholland. JA: Russell. AG: Tolbert. PE.
        (2009). Ambient air pollution and preterm birth: A time-series analysis. Epidemiology 20:  689-698.
Darrow,  LA:  Klein, M: Strickland, MJ: Mulholland, JA: Tolbert, PE. (2011a). Ambient air pollution and birth weight
        in full-term infants  in Atlanta, 1994-2004.  Environ Health Perspect 119: 731-737.
        http://dx.doi. orq/10.1289/ehp. 1002785.
Dell'Omo, G: Wolfer, D: Alleva, E: Lipp, H, -P. (1995). Developmental exposure to ozone induces  subtle
        changes in swimming navigation of adult mice. Toxicol Lett 81: 91-99.
Diaz. J: Linares. C: Garcia-Herrera. R: Lopez. C: Trigo. R. (2004). Impact of temperature and air  pollution on the
        mortality of children in Madrid. J Occup Environ  Med 46: 768-774.
Dockerv.  DW: Pope. CA. Ill: Xu. X: Spengler. JD: Ware. JH:  Fay. ME: Ferris.  BG.  Jr: Speizer. FE. (1993). An
        association between air pollution and mortality in six US cities. N Engl J Med 329: 1753-1759.
Dugandzic, R: Dodds, L: Stieb, D: Smith-Doiron, M. (2006).  The association between low level exposures to
        ambient air pollution and term low birth weight: A retrospective cohort study. Environ Health 5:  3.
        http://dx.doi.org/10.1186/1476-069X-5-3.
Ercan, H: Birben, E: Dizdar, EA:  Keskin, O: Karaaslan, C: Soyer, OU: Put, R: Sackesen, C: Besler, T: Kalayci,
        Q_ (2006). Oxidative stress and genetic and epidemiologic determinants  of oxidant injury in childhood
        asthma. J Allergy Clin Immunol 118: 1097-1104. http://dx.doi.Org/10.1016/i.iaci.2006.08.012.
Eustis, SL: Schwartz, LW: Kosch, PC: Dungworth,  PL. (1981). Chronic  bronchiolitis in nonhuman primates  after
        prolonged ozone exposure. Am J Pathol 105: 121-137.
Evans.  MJ: Fanucchi. MV: Baker. GL: Van Winkle.  LS: Pantle. LM: Nishio. SJ: Schelegle. ES: Gershwhin. LJ:
        Miller, LA: Hyde, DM: Sannes, PL: Plopper, CG. (2003b). Atypical development of the tracheal
        basement membrane zone of infant rhesus monkeys exposed  to ozone and allergen. Am J Physiol
        285: L931-L939. http://dx.doi.Org/10.1152/ajplung.00175.2003.
Evans,  MJ: Fanucchi, MV: Baker, GL: Van Winkle,  LS: Pantle, LM: Nishio, SJ: Schelegle, ES: Gershwin,  LJ:
        Miller. LA: Hyde. DM: Plopper. CG. (2004). The remodelled tracheal basement membrane zone of
        infant rhesus monkeys  after 6 months of recovery. Clin Exp Allergy 34: 1131-1136.
        http://dx.doi.0rg/10.1111/i.1365-2222.2004.02004.x CEA2004.
Draft - Do Not Cite or Quote                        7-88                                  September 2011

-------
Fanucchi. MV: Wong. VJ: Hinds. D: Tarkington. BK: Van Winkle. LS: Evans. MJ: Plopper. CG. (2000). Repeated
        episodes of exposure to ozone alters postnatal development of distal conducting airways in infant
        rhesus monkeys [Abstract]. Am J Respir Grit Care Med 161: A615.
Fanucchi. MV: Plopper. CG: Evans. MJ: Hyde. DM: Van Winkle. LS: Gershwin. LJ: Schelegle. ES. (2006).
        Cyclic exposure to ozone alters distal airway development in infant rhesus monkeys. Am J Physiol
        Lung Cell Mol Physiol 291: L644-L650. http://dx.doi.org/10.1152/aipluna.00027.2006.
Feichtinaer. W: Papalambrou. K: Poehl. M: Krischker. U: Neumann. K. (1997). Smoking and in vitro fertilization:
        A meta-analysis. J Assist Reprod Genet  14: 596-599. http://dx.doi.Org/10.1023/A: 1022584802711.
Ferng. S. -F: Castro. CE:  Afifi. AA: Bermudez. E:  Mustafa. MG. (1997). Ozone-induced DNAstrand breaks in
        guinea pig tracheobronchial epithelial cells. J Toxicol Environ Health 51: 353-367.
Forbes,  LJ:  Patel, MD: Rudnicka, AR: Cook,  DG:  Bush, T: Stedman, JR: Whincup, PH: Strachan, DP: Anderson,
        RH. (2009a). Chronic exposure to outdoor air  pollution and markers of systemic inflammation.
        Epidemiology 20: 245-253. http://dx.doi.orq/10.1097/EDE.Ob013e318190ea3f.
Forbes,  LJL: Kapetanakis, V: Rudnicka, AR:  Cook,  DG: Bush, T: Stedman, JR: Whincup, PH: Strachan, DP:
        Anderson. HR. (2009b). Chronic exposure to outdoor air pollution and lung function in adults. Thorax
        64: 657-663.
Frischer, T:  Studnicka, M: Gartner, C: Tauber, E: Horak, F: Veiter, A: Spengler, J: Kuhr, J: Urbanek, R. (1999).
        Lung function growth and ambient ozone: A three-year population study in school children. Am J Respir
        Grit Care Med 160: 390-396.
Frischer, T:  Studnicka, M: Halmerbauer, G: Horak, F: Gartner, C: Tauber, E: Koller, DY. (2001). Ambient ozone
        exposure is associated with eosinophil activation  in  healthy children. Clin Exp Allergy 31: 1213-1219.
Fuiinaka. LE: Hyde. DM: Plopper. CG: Tyler.  WS: Dungworth. PL: Lollini. LO. (1985). Respiratory bronchiolitis
        following long-term ozone exposure in bonnet monkeys: A morphometric study. Exp Lung Res 8:  167-
        190.
Gauderman. WJ. (2001).  Sample size requirements for matched case-control studies of gene-environment
        interaction. Stat Med 21: 35-50. http://dx.doi.org/10.1002/sim.973.
Gauderman. WJ. (2002).  Sample size requirements for association studies of gene-gene interaction. Am J
        Epidemiol 155: 478-484.
Gauderman, WJ: Avol, E: Gilliland, F: Vora, H: Thomas, D: Berhane, K: McConnell, R: Kuenzli, N: Lurmann, F:
        Rappaport. E: Margolis. H: Bates. D: Peters. J. (2004). The effect of air pollution on lung development
        from 10 to 18 years of age. N Engl J Med 351: 1057-1067.
Gilboa, SM: Mendola, P: Olshan, AF:  Langlois, PH: Savitz, DA: Loomis, D: Herring, AH: Fixler, DE. (2005).
        Relation between ambient air quality and selected birth defects, seven county study, Texas, 1997-2000.
        Am J Epidemiol 162: 238-252.
Gilliland, FD: McConnell,  R: Peters, J: Gong Jr, H. (1999). Atheoretical basis for investigating ambient air
        pollution and children's respiratory health. Environ Health Perspect 107: 403-407.
Gilliland. FD: Berhane. K: Rappaport. EB: Thomas. DC: Avol. E: Gauderman. WJ: London. SJ: Margolis.  HG:
        McConnell,  R: Islam, KT: Peters, JM. (2001). The effects of ambient air pollution on school
        absenteeism due to respiratory illnesses. Epidemiology 12: 43-54.
Gilliland. FD: Rappaport.  EB: Berhane. K: Islam. T: Dubeau. L: Gauderman. WJ: McConnell. R. (2002). Effects
        of glutathione S-Transferase P1, M1, and T1 on acute respiratory illness in school children. Am J
        Respir Grit Care  Med 166: 346-351.
Giovannelli. L: Pitozzi. V:  Moretti. S: Boddi. V: Dolara. P. (2006). Seasonal variations of DMA damage in human
        lymphocytes: Correlation with different environmental variables. Mutat Res-Fundam Mol Mech Mutagen
        593: 143-152. http://dx.doi.Org/10.1016/i.mrfmmm.2005.07.002.
Gonzalez-Pina. R: Escalante-Membrillo. C: Alfaro-Rodriguez. A: Gonzalez-Maciel. A. (2008). Prenatal exposure
        to ozone disrupts cerebellar monoamine contents in newborn rats. Neurochem  Res 33: 912-918.
        http://dx.doi. org/10.1007/s11064-007-9534-3.
Gouveia. N: Bremner. SA: Novaes. HMD. (2004). Association between ambient air pollution and  birth weight in
        Sao Paulo, Brazil. J Epidemiol Community Health 58: 11-17.
Guerrero. AL: Dorado-Martinez. C: Rodriguez. A:  Pedroza-Rios. K: Borgonio-Perez. G: Rivas-Arancibia. S.
        (1999).  Effects of vitamin E on ozone-induced memory deficits and lipid peroxidation in rats.
        Neuroreport 10: 1689-1692.
Draft - Do Not Cite or Quote                        7-89                                  September 2011

-------
Guevara-Guzman. R: Arriaga. V: Kendrick. KM: Bernal. C: Vega. X: Mercado-Gomez. OF: Rivas-Arancibia. S.
        (2009). Estradiol prevents ozone-induced increases in brain lipid peroxidation and impaired social
        recognition memory in female rats. Neuroscience 159: 940-950.
        http://dx.doi.0rg/10.1016/i.neuroscience.2009.01.047.
Ha, E, -H: Hong, Y, -C: Lee, B, -E: Woo, B, -H: Schwartz, J:  Christiani, DC. (2001). Is air pollution a risk factor
        for low birth weight in Seoul? Epidemiology 12: 643-648.
Ha. E. -H: Lee. J. -T: Kim. H: Hong. Y. -C: Lee. (2003). Infant susceptibility of mortality to air pollution in Seoul,
        South Korea. Pediatrics 111: 284-290.
Hack. M: Fanaroff. AA. (1999). Outcomes of children of extremely low birth weight and gestational age in the
        1990s. Early Hum Dev  53: 193-218. http://dx.doi.org/10.1016/50378-3782(98)00052-8.
Hajat, S: Armstrong, B: Wilkinson, P: Busby, A: Polk, H. (2007). Outdoor air pollution and infant mortality:
        Analysis of daily time-series data in 10 English cities.  J Epidemiol Community Health 61: 719-722.
        http://dx.doi.orq/10.1136/iech.2006.053942.
Han, SG: Bhoopalan, V: Akinbiyi, T:  Gairola, CG: Bhalla, DK. (2011). In utero tobacco smoke exposure alters
        pulmonary responses of newborn rats to ozone. J Toxicol Environ Health A 74: 668-677.
        http://dx.doi. org/10.1080/15287394.2011.539133.
Hanene, C: Jihene, l_: Jame, A:  Kamel, H: Agnes, H. (2007). Association of GST genes polymorphisms with
        asthma in Tunisian children. Mediators Inflamm 19564: 6.  http://dx.doi.org/10.1155/2007/19564.
Hansen. C: Neller. A: Williams. G: Simpson. R. (2006). Maternal exposure to low levels of ambient air pollution
        and preterm birth in Brisbane, Australia. BJOG 113: 935-941.
Hansen. C: Neller. A: Williams. G: Simpson. R. (2007). Low  levels of ambient air pollution during pregnancy and
        fetal growth among term neonates in Brisbane, Australia. Environ Res 103: 383-389.
Hansen, C: Luben, TJ: Sacks, JD: Olshan, A: Jeffav, S:  Strader, L:  Perreault, SD. (2010). The effect of ambient
        air pollution on sperm quality. Environ Health Perspect 118: 203-209.
        http://dx.doi.org/10.1289/ehp.0901022.
Hansen, CA: Barnett, AG: Pritchard, G. (2008). The effect of ambient air pollution during early pregnancy on
        fetal ultrasonic measurements during mid-pregnancy.  Environ Health Perspect 116: 362-369.
Hansen. CA: Barnett. AG: Jalaludin. B: Morgan. G. (2009). Ambient air pollution  and birth defects in Brisbane,
        Australia. PLoS ONE 4: e5408. http://dx.doi.org/10.1371/iournal.pone.0005408.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Dungworth. PL (1987a). Effects of an ambient level of
        ozone on primate nasal epithelial mucosubstances: quantitative histochemistry. Am J Pathol 127: 90-
        96.
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Wilson.  DW: Dungworth. PL. (1987b). Response of the
        macaque nasal epithelium to ambient levels of ozone: A morphologic and morphometric study of the
        transitional and respiratory epithelium. Am J Pathol  128: 29-44.
Harkema. JR: Morgan. KT: Gross. EA: Catalano. PJ: Griffith. WC. (1994). Consequences of prolonged
        inhalation of ozone on F344/N rats: Collaborative studies Part VII: Effects on  the nasal mucociliary
        apparatus (Vol. 65). Cambridge, MA: Health Effects Institute.
Harkema. JR: Catalano. PJ: Hotchkiss. JA. (1997a). Consequences of prolonged inhalation of ozone on F344/N
        rats: Collaborative studies: Part XII Atrophy of bone in nasal turbinates.  Cambridge, MA: Health Effects
        Institute.
Harkema. JR: Hotchkiss. JA: Griffith. WC. (1997b). Mucous  cell metaplasia in rat nasal epithelium after a 20-
        month exposure to ozone: A morphometric study of epithelial differentiation. Am J Respir Cell Mol Biol
        16: 521-530.
Harkema. JR: Hotchkiss. JA: Barr. EB: Bennett. CB: Gallup. M: Lee. JK:  Basbaum. C. (1999). Long-lasting
        effects of chronic ozone exposure on rat nasal  epithelium. Am J Respir Cell Mol Biol 20: 517-529.
Haro, R: Paz, C. (1993). Effects of ozone exposure during pregnancy on ontogeny of sleep in rats.  Neurosci
        Lett 164: 67-70.  http://dx.doi.Org/10.1016/0304-3940(93)90859-J.
Hassett. C: Mustafa. MG: Coulson. WF: Elashoff. RM. (1985).  Murine lung carcinogenesis following exposure to
        ambient ozone concentrations. J Natl Cancer Inst 75:  771-777.
Herbert. RA: Hailev. JR: Grumbein. S: Chou. BJ: Sills. RC: Haseman. JK: Goehl. T: Miller. RA: Rovcroft. JH:
        Boorman. GA. (1996). Two-year and lifetime toxicity and carcinogenicity studies of ozone in B6C3F1
        mice. Toxicol Pathol 24: 539-548.
Draft - Do Not Cite or Quote                        7-90                                  September 2011

-------
Himes. BE: Hunninghake. GM: Baurlev. JW: Rafaels. NM: Sleiman. P: Strachan. DP: Wilk. JB: Willis-Owen.
        SAG: Klanderman, B: Lasky-Su, J: Lazarus, R: Murphy, AJ: Soto-Quiros, ME: Avila, l_: Beaty, T:
        Mathias. RA: Ruczinski. I: Barnes. KG: Celedon. JC: Cookson. WOC: Gauderman. WJ: Gilliland. FD:
        Hakonarson. H: Lange. C: Moffatt. MF: O'Connor. GT: Rabv. BA: Silverman. EK: Weiss. ST.  (2009).
        Genome-wide Association Analysis Identifies PDE4D as an Asthma-Susceptibility  Gene. Am J Hum
        Genet 84: 581-593. http://dx.doi.Org/10.1016/i.aiha.2009.04.006.
Horak. F. Jr: Studnicka. M: Gartner. C: Spengler. JD: Tauber.  E: Urbanek. R: Veiter. A: Frischer. T. (2002).
        Particulate matter and lung function growth in children: a 3-yr follow-up study in Austrian
        schoolchildren. EurRespirJ  19: 838-845.
Huen. K: Gunn. L: Duramad. P:  Jena. M: Scalf. R: Holland. N. (2006). Application of a geographic information
        system to explore associations between air pollution and micronucleus frequencies in African American
        children and adults. Environ Mol Mutagen 47: 236-246. http://dx.doi.orq/10.1002/em.2Q193.
Hutcheon. JA: Platt. RW.  (2008). The missing data  problem in birth weight percentiles and  thresholds for "small-
        for-gestational-age". Am J Epidemiol 167: 786-792. http://dx.doi.Org/10.1093/aje/kwm327.
Hwang. B. -F: Lee. Y. -L: Lin. Y.  -C: Jaakkola. JJK: Guo. YL. (2005). Traffic related air pollution as a determinant
        of asthma among Taiwanese school children. Thorax 60: 467-473.
Hwang, B, -F: Jaakkola, JJK:  Lee, Y, -L: Lin, Y, -C: Y-LL, G. (2006). Relation between air pollution and allergic
        rhinitis in Taiwanese schoolchildren. Respir Res 7: 23.
Hwang. BF: Jaakkola. JJ. (2008). Ozone and other air pollutants and the risk of oral clefts.  Environ Health
        Perspect 116: 1411-1415.
Hyde. DM: Plopper. CG: Harkema. JR: St George. JA: Tyler. WS: Dungworth. PL. (1989). Ozone-induced
        structural changes in monkey respiratory system. In T Schneider; SD Lee; GJR Wolters; LD  Grant
        (Eds.), Atmospheric ozone research and its policy implications: Proceedings of the 3rd US-Dutch
        International Symposium, Nijmegen, the Netherlands May 9-13, 1988 (pp. 523-532). Nijmegen, the
        Netherlands: Elsevier Science Publishers.
Hyde, DM; Hubbard, WC; Wong, V; Wu,  R; Pinkerton, K; Plopper, CG. (1992). Ozone-induced acute
        tracheobronchial epithelial injury: relationship to granulocyte emigration in the lung. Am J Respir Cell
        Mol Biol 6: 481-497.
Ihorst, G; Frischer, T; Horak, F;  Schumacher, M; Kopp, M; Forster, J;  Mattes, J; Kuehr, J. (2004). Long- and
        medium-term ozone effects on lung growth including a broad spectrum of exposure. Eur Respir J 23:
        292-299.
lijima, MK; Kobayashi,  T. (2004). Nasal allergy-like symptoms aggravated by ozone exposure in a
        concentration-dependent manner in guinea pigs. Toxicology  199: 73-83.
        http://dx.doi.0rg/10.1016/i.tox.2004.01.008.
Islam, T; Gauderman, WJ; Berhane, K; McConnell, R: Avol, E; Peters, JM; Gilliland, FD. (2007). The
        relationship between air pollution, lung function and asthma in adolescents. Thorax 62: 957-963.
Islam. T: McConnell. R: Gauderman. WJ: Avol. E: Peters. JM: Gilliland.  FD. (2008). Ozone, oxidant defense
        genes and risk of asthma during adolescence. Am J  Respir Crit Care Med 177: 388-395.
        http://dx.doi.org/10.1164/rccm.200706-863OC.
Islam. T: Berhane. K: McConnell. R: Gauderman. WJ: Avol. E: Peters. JM: Gilliland. FD. (2009). Glutathione-S-
        transferase (GST) P1, GSTM1, exercise, ozone and asthma incidence in school children. Thorax 64:
        197-202. http://dx.doi.org/10.1136/thx.2008.099366.
Jacguemin. B: Kauffmann. F: Pin. I: Le Moual. N: Bousguet. J: Gormand.  F:  Just. J: Nadif.  R: Pison. C: Vervloet.
        D;  Kunzli, N; Siroux, V.  (In Press) Air pollution and asthma control in the epidemiological study on the
        genetics and environment of asthma. J Epidemiol Community Health.
        http://dx.doi. orq/10.1136/iech.2Q10.130229.
Jakab, GJ; Bassett, DJP  (1990). Influenza virus infection, ozone exposure, and fibrogenesis. Am J Respir Crit
        Care Med 141: 1307-1315.
Jalaludin. B: Mannes. T: Morgan. G: Lincoln. D: Sheppeard. V: Corbett. S. (2007). Impact of ambient air
        pollution on gestational age is modified by season in Sydney, Australia. Environ Health 6: 16.
Jedlinska-Krakowska. M:  Bomba. G: Jakubowski. K: Rotkiewicz. T: Jana. B: Penkowskii. A. (2006). Impact of
        oxidative stress and supplementation with vitamins E and C on testes morphology in rats. J Reprod
        Dev 52: 203-209.
Draft - Do Not Cite or Quote                        7-91                                  September 2011

-------
Jerrett. M: Burnett. RT: Pope. CA. Ill: Ito. K: Thurston. G: Krewski. D: Shi. Y: Calle. E: Thun. M. (2009). Long-
        term ozone exposure and mortality. N Engl J Med 360: 1085-1095.
        http://dx.doi.org/10.1056/NEJMoa0803894.
Jiang. LL: Zhang. YH: Song. GX: Chen. GH: Chen. BH: Zhao. NQ: Kan. HP. (2007). A time series analysis of
        outdoor air pollution and preterm birth in Shanghai, China. Biomed Environ Sci 20: 426-431.
Joad. JP: Kott. KS: Brie. JM: Peake. JL: Plopper. CG: Schelegle.  ES: Gershwin. LJ: Pinkerton. KE. (2006).
        Structural and functional localization of airway effects from episodic exposure of infant monkeys to
        allergen and/or ozone. Toxicol Appl Pharmacol 214: 237-243.
        http://dx.doi.0rg/10.1016/i.taap.2005.12.012.
Joad. JP: Kott. KS: Brie. JM: Schelegle. ES: Gershwin. LJ: Plopper. CG: Peake. JL: Pinkerton. KE. (2008). The
        effects of inhaled corticosteroids on intrinsic responsiveness and histology of airways from infant
        monkeys exposed to house dust mite allergen and ozone. Toxicol Appl Pharmacol 226: 153-160.
        http://dx.doi.0rg/10.1016/i.taap.2007.09.005.
Kajekar, R: Pieczarka, EM: Smiley-Jewell, SM: Schelegle,  ES: Fanucchi, MV: Plopper, CG. (2007). Early
        postnatal exposure to allergen and ozone leads to hyperinnervation of the pulmonary epithelium.
        Respir Physiol Neurobiol 155: 55-63. http://dx.doi.Org/10.1016/i.resp.2006.03.002.
Karr, C: Lumley, T: Schreuder, A: Davis, R: Larson, T: Ritz, B:  Kaufman, J. (2007). Effects of subchronic and
        chronic exposure to ambient air pollutants on infant bronchiolitis. Am J Epidemiol 165: 553-560.
Katre. A: Ballinger. C: Akhter. H: Fanucchi. M: Kim. DK: Postlethwait. E: Liu.  RM.  (2011). Increased transforming
        growth factor beta 1 expression mediates ozone-induced airway fibrosis in mice. Inhal Toxicol 23: 486-
        494. http://dx.doi.org/10.3109/08958378.2011.584919.
Kavlock. R: Daston. G: Grabowski. CT (1979). Studies on  the developmental toxicity of ozone. I. Prenatal
        effects. Toxicol Appl Pharmacol 48: 19-28. http://dx.doi.org/10.1016/S0041-008X(79)80004-6.
Kavlock. RJ: Meyer. E: Grabowski. CT. (1980). Studies on the developmental toxicity of ozone: Postnatal
        effects. Toxicol Lett 5: 3-9. http://dx.doi.org/10.1016/0378-4274(80)90141-1.
Kim, MY: Cho, MY. (2009a). Toxicity and carcinogenicity of ozone in combination with 4-(N-methyl-N-
        nitrosamino)-1-(3-pyridyl)-1-butanone and dibutyl phthalate in  B6C3F1 mice for 16 and 32 weeks.
        Biomed Environ Sci 22: 216-222.
Kim, MY: Cho, MY. (2009b). Tumorigenesis in B6C3F1 mice exposed to ozone in combination with 4-(N-methyl-
        N-nitrosamino)-1-(3-pyridyl)-1-butanone and dietary dibutyl phthalate.  Toxicol Ind Health 25: 189-195.
        http://dx.doi. org/10.1177/0748233709106185.
Kinney,  PL: Lippmann, M. (2000). Respiratory effects of seasonal exposures to ozone and particles. Arch
        Environ Occup Health 55: 210-216.
Kodavanti. UP: Thomas.  R: Ledbetter. AD: Schladweiler. MC: Shannahan.  JH:  Wallenborn. JG: Lund. AK:
        Campen, MJ:  Butler, EO: Gottipolu, RR: Nvska, A: Richards, JE: Andrews, D: Jaskot, RH:  McKee, J:
        Kotha. SR: Patel. RB: Parianandi. NL. (2011). Vascular and cardiac impairments in rats Inhaling ozone
        and diesel exhaust particles. Environ Health Perspect 119: 312-318.
        http://dx.doi. org/10.1289/ehp. 1002386.
Krewski. D: Jerrett. M: Burnett. RT: Ma. R: Hughes. E: Shi. Y:  Turner. MC:  3rd. PA: Thurston. G: Calle. EE:
        Thun.  MJ. (2009). Extended follow-up and spatial  analysis of the American Cancer Society study
        linking particulate air pollution and mortality. (Report Nr. 140). Cambridge,  MA:  Health Effects Institute.
Kuo. HW: Lai. JS: Lee. MC: Tai. RC. (2002). Respiratory effects of air pollutants among asthmatics in central
        Taiwan. Arch Environ Occup Health 57: 194-200.
Lampl, M: Jeantv, P. (2003). Timing is everything:  A reconsideration of fetal growth velocity patterns identifies
        the  importance of individual and sex differences. Am J Hum Biol 15: 667-680.
        http://dx.doi.org/10.1002/aihb.10204.
Larson,  SD:  Schelegle, ES: Walbv, WF: Gershwin, LJ: Fanuccihi, MV: Evans, MJ: Joad, JP: Tarkington, BK:
        Hyde. DM: Plopper. CG. (2004). Postnatal remodeling of the neural components of the epithelial-
        mesenchymal trophic unit in the proximal airways  of infant rhesus monkeys exposed to ozone and
        allergen. Toxicol Appl Pharmacol 194: 211-220.
Last. JA: Reiser. KM: Tyler. WS: Rucker. RB. (1984).  Long-term consequences of exposure to ozone. I.  Lung
        collagen content. Toxicol Appl Pharmacol 72: 111-118.
Last, JA: Warren, PL: Pecguet-Goad, E: Witschi, H. (1987). Modification by ozone of lung tumor development in
        mice. J Natl Cancer Inst 78: 149-154.
Draft - Do Not Cite or Quote                        7-92                                  September 2011

-------
Last. JA: Gelzleichter. TR: Harkema. J: Hawk. S. (1994). Consequences of prolonged inhalation of ozone on
       Fischer-344/N rats: Collaborative studies.  Part I: Content and cross-linking of lung collagen.
Latzin. P: Roosli. M: Huss. A: Kuehni. CE: Frev. U. (2009). Air pollution during pregnancy and lung function in
       newborns: A birth cohort study. Eur Respir J 33: 594-603.
Lee, SJ: Hajat, S: Steer, PJ: Filippi, V. (2008c). A time-series analysis of any short-term effects of meteorological
       and air pollution factors on preterm births  in London, UK. Environ Res 106: 185-194.
Lee. Y. -L:  Lin. Y. -C: Lee. Y. -C: Wang. J. -Y: Hsiue. T. -R: Quo. YL. (2004b). Glutathione S-transferase P1 gene
       polymorphism and air pollution as interactive risk factors for childhood asthma. Clin Exp Allergy 34:
       1707-1713.
Lee. YL: McConnell. R: Berhane. K: Gilliland. FD. (2009b). Ambient ozone modifies the effect of tumor necrosis
       factor G-308Aon  bronchitic symptoms among children with asthma. Allergy 64: 1342-1348.
       http://dx.doi. orq/10.1111/i.1398-9995.2009.02014.x.
Legro. RS: Sauer. MV: Mottla. GL: Richter. KS: Li.  X: Dodson. WC: Liao. D. (2010). Effect of air quality on
       assisted human reproduction. Hum Reprod 25: 1317-1324. http://dx.doi.org/10.1093/humrep/deq021.
Li. H:  Romieu. I: Sienra-Monge. JJ: Ramirez-Aguilar. M: Estela del Rio-Navarro. B: Kistner. EO: Giessing. HK:
       Irma del Carmen. LS: Chiu. GY: London. SJ. (2006a). Genetic polymorphisms in arginase I and II and
       childhood asthma and atopy. J Allergy Clin Immunol 117: 119-126.
Li. YF: Gauderman. WJ: Conti. DV: Lin. PC: Avol. E: Gilliland. FD.  (2008).  Glutathione S-Transferase P1,
       Maternal Smoking, and Asthma in Children: A Haplotype-Based Analysis. Environ Health  Perspect 116:
       409-415.
Lin. CA: Pereira. LAA: Nishioka. DC: Conceicao. QMS: Graga. ALF: Saldiva. PHN. (2004a). Air pollution and
       neonatal deaths in Sao Paulo, Brazil. Braz J Med  Biol Res 37: 765-770.
Lin, CM: Li, C, -Y: Yang, G, -Y: Mao, IF. (2004b). Association between maternal exposure to elevated ambient
       sulfur dioxide during  pregnancy and term low birth weight. Environ Res 96: 41-50.
Lin. S: Fitzgerald. E: Hwang.  SA: Munsie. JP: Stark. A. (1999). Asthma hospitalization rates and socioeconomic
       status in New York State (1987-1993). J Asthma 36: 239-251.
Lin. S: Liu. X: Le. LH: Hwang. SA. (2008b). Chronic exposure to ambient ozone and asthma hospital
       admissions among children. Environ Health Perspect 116:  1725-1730.
       http://dx.doi.org/10.1289/ehp.11184.
Lin. S. (2010). E-mail correspondence from Shao Lin to Dennis Kotchmar dated December 20, 2010
Linn. WS: Rappaport. EB:  Berhane. KT: Bastain. TM: Avol. EL: Gilliland. FD. (2009). Exhaled nitric oxide in a
       population-based study of southern California schoolchildren. Respir Res 10: 28.
Lipfert. FW: Perry. HM. Jr: Miller. JP: Batv. JD: Wvzga. RE: Carmodv. SE. (2000). The Washington University-
       EPRI veterans' cohort mortality study: Preliminary results. Inhal Toxicol 4: 41-73.
Lipfert, FW: Perry, HM, Jr: Miller, JP: Batv, JD: Wyzga, RE: Carmodv, SE. (2003). Air pollution, blood pressure,
       and their long-term associations with mortality. Inhal Toxicol 15: 493-512.
Lipfert. FW: Batv. JD: Miller. JP: Wvzga. RE.  (2006a). PM2.5 constituents and related air quality variables as
       predictors of survival in a cohort of U.S. military veterans.  Inhal Toxicol  18: 645-657.
Lipfert. FW: Wvzga. RE: Batv. JD: Miller. JP. (2006b). Traffic density as a surrogate measure of environmental
       exposures in studies of air pollution health effects: Long-term mortality in a cohort of US veterans.
       Atmos Environ 40: 154-169.
Liu. S: Krewski.  D: Shi. Y:  Chen. Y: Burnett. R. (2007b). Association between maternal exposure to ambient air
       pollutants during pregnancy and fetal growth restriction. J Expo Sci Environ Epidemiol 17: 426-432.
Loomis, D: Castillejos, M: Gold, PR: McDonnell, W: Borja-Aburto, VH. (1999). Air pollution and infant mortality in
       Mexico City. Epidemiology 10: 118-123.
Lopez. I: Sanchez.  I: Bizarro.  P: Acevedo. S: Ustarroz. M: Fortoul.  T. (2008). Ultrastructural alterations during
       embryonic rats' lung  development caused by ozone. J Electron Microsc (Tokyo) 57: 19-23.
       http://dx.doi. orq/10.1093/imicro/dfm033.
Maniar-Hew. K:  Postlethwait.  EM: Fanucchi. MV: Ballinger. CA: Evans. MJ: Harkema. JR: Carey. SA:
       McDonald, RJ:  Bartolucci, AA: Miller, LA. (2011).  Postnatal episodic ozone results in persistent
       attenuation of pulmonary and peripheral blood responses to LPS challenge. Am  J Physiol Lung Cell
       Mol Physiol 300: L462-L471. http://dx.doi.ora/10.1152/aiplung.00254.2010.
Mannes, T: Jalaludin, B: Morgan, G: Lincoln, D: Sheppeard, V: Corbett, S. (2005). Impact of ambient air
       pollution on birth weight in Sydney, Australia. Occup Environ Med 62: 524-530.
Draft - Do Not Cite or Quote                        7-93                                  September 2011

-------
Mariassv. AT: Sielczak. MW: McCrav. MN: Abraham. WM: Wanner. A. (1989). Effects of ozone on lamb tracheal
        mucosa: Quantitative glycoconjugate histochemistry. Am J Pathol 135: 871-879.
Mariassv. AT: Abraham. WM:  Phipps. RJ: Sielczak. MW: Wanner. A. (1990). Effect of ozone on the postnatal
        development of lamb mucociliary apparatus. J Appl Physiol 68: 2504-2510.
Marshall, E: Harris, G: Wartenberg, D. (2010). Oral cleft defects and  maternal exposure to ambient air pollutants
        in New Jersey. Birth Defects Res AClin Mol Teratol 88: 205-215. http://dx.doi.orq/10.1002/bdra.20650.
Martinez. FD: Wright. AL: Taussig. LM: Holberg. CJ: Halonen.  M:  Morgan. WJ: Associates. GHM. (1995).
        Asthma and wheezing in the first six years of life. N Engl J Med 332: 133-138.
McConnell. R: Berhane. K: Gilliland. F: London. SJ: Islam. T: Gauderman. WJ: Avol. E: Margolis. HG: Peters.
        JM. (2002). Asthma in exercising children exposed to ozone: A cohort study. Lancet 359: 386-391.
McConnell, R: Islam, T: Shankardass, K: Jerrett, M: Lurmann, F: Gilliland,  F: Gauderman, J: Avol, E: Kuenzli, N:
        Yao. L: Peters. J: Berhane. K. (2010). Childhood incident asthma and traffic-related air pollution at
        home and school. Environ Health Perspect 118: 1021-1026. http://dx.doi.org/10.1289/ehp.0901232.
Meng, YY: Wilhelm, M: Rull, RP: English, P: Ritz, B. (2007). Traffic and outdoor air pollution levels near
        residences and poorly controlled asthma in adults. Ann Allergy Asthma Immunol 98: 455-463.
Mena. YY: Rull. RP: Wilhelm. M: Lombardi. C: Balmes. J: Ritz. B. (2010). Outdoor air pollution and uncontrolled
        asthma in the San Joaquin Valley, California.  J Epidemiol Community Health 64: 142-147.
        http://dx.doi.orq/10.1136/iech.2008.083576.
Miller. LA: Gerriets. JE: Tyler. NK: Abel. K: Schelegle.  ES: Plopper. CG: Hyde. DM. (2009). Ozone and allergen
        exposure during postnatal development alters the frequency and airway distribution of CD25+ cells in
        infant rhesus monkeys. Toxicol Appl  Pharmacol  236: 39-48.
        http://dx.doi. orq/10.1016/i.taap.20Q8.12.031.
Moffatt,  RK: Hyde, DM: Plopper,  CG:  Tyler, WS: Putney,  LF.  (1987). Ozone-induced adaptive and reactive
        cellular changes in respiratory bronchioles of Bonnet  monkeys. Exp Lung Res 12: 57-74.
Monchaux. G: Morlier. JP: Morin. M: Rochefort.  P: Maximilien. R:  Tredaniel. J. (1996). Co-carcinogenic effects
        in rats of combined exposure to radon and ozone. Environ Int 221: S909-S915.
Moore. K: Neugebauer. R: Lurmann.  F: Hall. J: Braier. V: Alcorn. S: Tager.  I. (2008). Ambient ozone
        concentrations cause increased hospitalizations for asthma  in  children: An 18-year study in Southern
        California. Environ Health Perspect 116: 1063-1070. http://dx.doi.org/10.1289/ehp.10497.
Morello-Frosch. R: Jesdale. BM:  Sadd. JL: Pastor.  M.  (2010). Ambient air pollution exposure and full-term birth
        weight in California. Environ  Health 9: 44. http://dx.doi.org/10.1186/1476-069X-9-44.
Morris, CR: Poljakovic, M: Lavrisha,  L: Machado, L: Kuypers, FA: Morris, SM, Jr. (2004).  Decreased arginine
        bioavailability and increased serum arginase activity in asthma. Am J Respir Crit Care Med 170: 148-
        153. http://dx.doi.org/10.1164/rccm.200309-1304OC.
Mortimer, K:  Neugebauer, R:  Lurmann, F: Alcorn, S: Balmes, J: Tager, I. (2008a). Air pollution and pulmonary
        function in asthmatic children: Effects of prenatal and lifetime exposures. Epidemiology 19: 550-557.
        http://dx.doi. orq/10.1097/EDE.Ob013e31816a9dcb.
Mortimer, K:  Neugebauer, R:  Lurmann, F: Alcorn, S: Balmes, J: Tager, I. (2008b). Early-lifetime exposure to air
        pollution and allergic sensitization in children with asthma. J Asthma 45: 874-881.
        http://dx.doi. orq/10.1080/02770900802195722.
NTP (National Toxicology Program). (1994). Toxicology and carcinogenesis: Studies of ozone (CAS No 10028-
        15-6) and ozone/NNK (CAS  No 10028-15-6/64091-91-4) in F344/N rats and B6C3F1 mice. (Technical
        Report No. 440).  Research Triangle  Park,  NC. http://ntp.niehs.nih.gov/index.cfm?obiectid=070AOEBD-
        081E-B501-E38F640803C3542C.
Orvszczvn. MP: Bouzigon. E: Maccario. J: Siroux. V:  Nadif. R: Wright. A: Kauffmann. F. (2007).
        Interrelationships of quantitative asthma-related phenotypes in the epidemiological study on the
        genetics and environment of asthma, bronchial hyperresponsiveness, and atopy. J Allergy Clin
        Immunol 119: 57-63.
Palli. D: Sera. F: Giovannelli.  L: Masala. G: Grechi. D: Bendinelli. B: Caini. S: Dolara. P: Saieva. C. (2009).
        Environmental ozone exposure and oxidative DN A damage  in  adult residents of Florence, Italy. Environ
        Pollut 157: 1521-1525. http://dx.doi.org/S0269-7491 (08)00472-7 [pii110.1016/i.envpol.2008.09.011.
Parker. JD: Akinbami. LJ: Woodruff. TJ. (2009).  Air pollution and childhood respiratory allergies in the United
        States. Environ Health Perspect 117: 140-147. http://dx.doi.org/10.1289/ehp.11497.
Draft - Do Not Cite or Quote                        7-94                                  September 2011

-------
Parker. JD: Rich. DQ: Glinianaia. SV: Leem. JH: Wallenberg. D: Bell. ML: Bonzini. M: Brauer. M: Darrow. L:
        Gehring, U: Gouveia, N: Grillo, P: Ha, E: van den Hooven, EH: Jalaludin, B: Jesdale, BM: Lepeule, J:
        Morello-Frosch. R: Morgan. GG: Slama. R: Pierik. FH:  Pesatori. AC: Sathvanaravana. S: Seo. J:
        Strickland. M: Tamburic. L: Woodruff. TJ. (2011). The international collaboration on air pollution and
        pregnancy outcomes:  Initial results. Environ Health Perspect 119: 1023-1028.
        http://dx.doi.org/10.1289/ehp.1002725.
Paz. C: Bazan-Perkins. B. (1992). Sleep-wake disorganization in cats exposed to ozone. Neurosci Lett 140:
        270-272.
Paz. C: Huitron-Resendiz. S. (1996). The effects of ozone exposure on the sleep-wake cycle and serotonin
        contents in the pons of the rat. Neurosci Lett 204: 49-52.
Peluso M Hainaut, P: Airoldi, L: Autrup, H: Dunning, A: Garte, S: Gormally, E: Malaveille, C: Matullo, G:
        Munniaa. A: Riboli. E:  investigators. VPE. (2005). Methodology of laboratory measurements in
        prospective studies on gene-environment interactions:  The experience of GenAir. DMA Repair 574: 92-
        104.
Penard-Morand. C: Charpin. D: Raherison. C: Kopferschmitt. C: Caillaud. D: Lavaud. F: Annesi-Maesano. I.
        (2005). Long-term exposure to background air pollution related to respiratory and allergic health in
        schoolchildren. Clin Exp Allergy 35: 1279-1287.
Pereira. FAC: De Assuncao. JV: Saldiva. PHN:  Pereira. LAA: Mirra. AP: Braga. ALF. (2005).  Influence of air
        pollution on the incidence of respiratory tract neoplasm. J Air Waste Manag Assoc 55: 83-87.
Pereira, LAA: Loomis, D:  Conceicao, QMS: Braga,  ALF: Areas,  RM: Kishi, HS:  Singer, JM: Bohm, GM: Saldiva,
        PHN. (1998). Association between air pollution and intrauterine mortality in Sao Paulo, Brazil. Environ
        Health Perspect  106: 325-329.
Perepu, RS: Garcia, C: Postal, D: Sethi, R. (2010). Enhanced death signaling in ozone-exposed ischemic-
        reperfused hearts. Mol Cell Biochem 336:  55-64. http://dx.doi.org/10.1007/s11010-009-0265-4.
Peters. JM: Avol. E: Navidi. W: London. SJ: Gauderman. WJ: Lurmann. F: Linn. WS: Margolis. H: Rappaport. E:
        Gong, H, Jr: Thomas,  DC. (1999a). A study of twelve southern  California communities with differing
        levels and types  of air pollution I Prevalence of respiratory morbidity. Am J  Respir Crit Care Med 159:
        760-767.
Peters, JM: Avol, E: Gauderman, WJ: Linn, WS: Navidi, W: London, SJ: Margolis, H: Rappaport, E: Vora, H:
        Gong. H. Jr: Thomas.  DC. (1999b). A study of twelve southern  California communities with differing
        levels and types  of air pollution II Effects on pulmonary function. Am J  Respir Crit Care Med 159: 768-
        775.
Petruzzi. S: Fiore. M: Dell'Omo. G: Bignami. G:  Alleva. E. (1995). Medium and  long-term behavioral effects in
        mice of extended gestational exposure to ozone. Neurotoxicol  Teratol 17: 463-470.
Petruzzi, S: De Acetis, L:  Chiarotti, F: Sorace, A: Alleva, E. (1999).  Limited changes in handedness and
        morphine reactivity in CD-1 mice after pre- and postnatal ozone exposure.  Acta Neurobiol Exp (Wars)
        59: 115-122.
Pinkerton,  KE: Menache,  MG: Plopper, CG. (1995). Consequences of prolonged inhalation of ozone on F344/N
        rats: Collaborative studies Part IX Changes in the tracheobronchial epithelium, pulmonary acinus, and
        lung antioxidant enzyme activity.
Pinkerton,  KE: Weller, BL: Menache, MG: Plopper,  CG. (1998).  Consequences of prolonged inhalation of ozone
        on F344/N rats: Collaborative studies Part XIII A comparison of changes in the tracheobronchial
        epithelium and pulmonary acinus in male rats at 3 and  20  months.
Plopper, CG: Chu, F, -P: Haselton, CJ: Peake, J: Wu, J: Pinkerton, KE.  (1994). Dose-dependent tolerance to
        ozone: I tracheobronchial epithelial reorganization in rats after  20 months' exposure. Am J Pathol 144:
        404-420.
Plopper, CG: Smiley-Jewell, SM: Miller, LA: Fanucchi, MV: Evans,  MJ:  Buckpitt, AR: Avdalovic, M: Gershwin,
        LJ: Joad. JP: Kaiekar. R: Larson. S: Pinkerton. KE: Van Winkle. LS: Schelegle. ES:  Pieczarka. EM: Wu.
        R: Hyde. DM. (2007). Asthma/allergic airways disease: Does postnatal exposure to  environmental
        toxicants promote airway pathobiology? Toxicol Pathol  35: 97-110.
        http://dx.doi. org/10.1080/01926230601132030.
Pope CA.  Ill: Burnett. RT:  Thun. MJ: Calle.  EE:  Krewski. D:  Ito.  K: Thurston. GD. (2002). Lung cancer,
        cardiopulmonary  mortality, and long-term exposure to fine particulate air pollution. JAMA 287: 1132-
        1141.
Draft - Do Not Cite or Quote                        7-95                                  September 2011

-------
Qian. Z: Liao. D: Lin. H. -M: Whitsel. EA: Rose. KM: Duan. Y. (2005). Lung function and long-term exposure to
        air pollutants in middle-aged American adults. Arch Environ Occup Health 60: 156-163.
Rage. E: Siroux. V: Kunzli. N: Pin. I: Kauffmann. F. (2009a). Air pollution and asthma severity in adults. Occup
        Environ Med 66: 182-188. http://dx.doi.ora/10.1136/oem.2007.038349.
Rage, E: Jacquemin, B: Nadif, R: Oryszczyn, MP: Siroux, V: Aguilera, I: Kauffmann, F: Kunzli, N. (2009b). Total
        serum IgE levels are associated with ambient ozone concentration in asthmatic adults. Allergy 64: 40-
        46.
Ramadour, M: Burel, C: Lanteaume, A: Vervloet, D: Charpin, D: Brisse, F: Dutau, H. (2000). Prevalence of
        asthma and rhinitis in relation to long-term exposure to gaseous air pollutants. Allergy 55: 1163-1169.
Reiser. KM: Tyler. WS: Hennessv. SM: Dominguez. JJ:  Last. JA. (1987). Long-term consequences of exposure
        to ozone:  II. Structural alterations in lung collagen of monkeys. Toxicol Appl Pharmacol 89: 314-322.
        http://dx.doi. org/10.1016/0041 -008X(87)90151-7.
Renzetti. G: Silvestre. G: D'Amario. C: Bottini. E: Gloria-Bottini. F: Bottini. N: Auais. A: Perez. MK: Piedimonte.
        G^ (2009).  Less air pollution leads to rapid reduction of airway inflammation and improved airway
        function in asthmatic children. Pediatrics 123: 1051-1058. http://dx.doi.org/10.1542/peds.2008-1153.
Ritz. B: Yu. F. (1999). The effect of ambient carbon monoxide on  low birth weight among children born in
        southern California between 1989 and 1993. Environ Health Perspect 107:  17-25.
Ritz. B: Yu. F: Chapa. G: Fruin. S. (2000). Effect of air pollution on preterm birth among children born in
        Southern California between 1989 and 1993. Epidemiology 11: 502-511.
Ritz, B: Yu, F: Fruin, S: Chapa, G: Shaw, GM: Harris, JA. (2002).  Ambient air pollution and risk of birth defects in
        Southern California. Am J Epidemiol 155: 17-25.
Ritz. B: Wilhelm. M: Zhao. Y. (2006). Air pollution and infant death in southern California, 1989-2000. Pediatrics
        118: 493-502.
Ritz. B: Wilhelm. M: Hoggatt. KJ: Ghosh. JK. (2007). Ambient air  pollution and  preterm birth in the environment
        and pregnancy outcomes study at the University of California, Los Angeles. Am J Epidemiol 166: 1045-
        1052.
Ritz. B: Wilhelm. M. (2008). Ambient air pollution and adverse birth outcomes:  Methodologic issues in an
        emerging field. Basic Appl Ecol 102: 182-190.
Rivas-Arancibia, S: Dorado-Martinez, C: Borgonio-Perez, G: Hiriart-Urdanivia,  M: Verdugo-Diaz, L: Duran-
        Vazguez. A: Colin-Barangue. L: Avila-Costa. MR. (2000). Effects oftaurine on ozone-induced memory
        deficits and lipid peroxidation  levels in brains of young, mature, and old rats. Environ Res 82: 7-17.
        http://dx.doi. org/10.1006/enrs. 1999.3996.
Rivas-Arancibia. S: Guevara-Guzman. R:  Lopez-Vidal. Y: Rodrfguez-Martfnez. E: Gomes.  MZ: Angoa-Perez. M:
        Raisman-Vozari. R. (2010). Oxidative stress caused by ozone exposure induces loss of brain repair in
        the hippocampus of adult rats. Toxicol Sci 113: 187-197.  http://dx.doi.org/10.1093/toxsci/kfp252.
Rivas-Manzano. P: Paz. C. (1999). Cerebellar morphological alterations in rats induced by prenatal ozone
        exposure.  Neurosci Lett 276: 37-40.
Rojas-Martinez, R:  Perez-Padilla,  R: Olaiz-Fernandez, G: Mendoza-Alvarado, L: Moreno-Macias,  H: Fortoul, T:
        McDonnell. W: Loomis. D: Romieu. I. (2007). Lung function growth in children with long-term exposure
        to air pollutants in Mexico City. Am J Respir Grit Care Med 176: 377-384.
Romero-Velazquez, RM: Alfaro-Rodriguez, A: Gonzalez-Pina, R:  Gonzalez-Maciel, A. (2002). Effect of ozone
        prenatal exposure on postnatal development of cerebellum. Proc West Pharmacol Soc45: 65-67.
Romero. R: Espinoza. J: Kusanovic. JP: Gotsch. F: Hassan. S: Erez. O: Chaiworapongsa. T: Mazor.  M. (2006).
        The preterm parturition syndrome. BJOG 113:  17-42. http://dx.doi.org/10.1111/j. 1471-
        0528.2006.01120.x.
Romieu. I: Sienra-Monge. JJ: Ramirez-Aguilar.  M: Moreno-Macias. H: Reyes-Ruiz.  Nl: Estela del Rio-Navarro.
        B: Hernandez-Avila, M: London, SJ. (2004a). Genetic polymorphism of GSTM1 and antioxidant
        supplementation influence lung function in relation to ozone exposure in asthmatic children in Mexico
        City. Thorax 59: 8-10.
Romieu, I: Ramirez-Aguilar, M: Moreno-Macias, H: Barraza-Villarreal, A: Miller, P: Hernandez-Cadena, L:
        Carbaial-Arrovo. LA: Hernandez-Avila. M. (2004b). Infant mortality and air pollution: Modifying effect by
        social class. J Occup Environ Hyg 46: 1210-1216.
Draft - Do Not Cite or Quote                        7-96                                  September 2011

-------
Rubes. J: Selevan. SG: Evenson. DP: Zudova. D: Vozdova. M: Zudova. Z: Robbins. WA: Perreault. SD. (2005).
        Episodic air pollution is associated with increased DMA fragmentation in human sperm without other
        changes in semen quality. Hum Reprod 20: 2776-2783.
Rubio. C: Paz. C. (2003). Indomethacin reverts sleep disorders produced by ozone exposure in rats. Toxicology
        191: 89-96.  http://dx.doi.Org/10.1016/S0300-483X(03)00245-2.
Salam. MT: Millstein. J: Li. Y. -F:  Lurmann. FW: Margolis.  HG: Gilliland. FD. (2005). Birth outcomes and prenatal
        exposure to ozone, carbon monoxide, and particulate matter: Results from the Children's Health Study.
        Environ Health Perspect 113: 1638-1644.
Salam. MT: Islam. T: Gauderman. WJ: Gilliland. FD. (2009). Roles of arginase variants, atopy, and  ozone in
        childhood asthma. J Allergy Clin Immunol 123: 596-602. http://dx.doi.Org/10.1016/i.iaci.2008.12.020.
Santiago-Lopez, D: Bautista-Martfnez, JA: Reyes-Hernandez, Cl: Aguilar-Martfnez, M:  Rivas-Arancibia, S.
        (2010). Oxidative stress, progressive damage in the substantia nigra and plasma dopamine oxidation,
        in rats chronically exposed to ozone. Toxicol Lett 197: 193-200.
        http://dx.doi. org/10.1016/j.toxlet.2Q10.05.020.
Santucci. D: Sorace. A: Francia.  N: Aloe. L: Alleva.  E. (2006). Prolonged prenatal exposure to low-level ozone
        affects aggressive behaviour as well as NGF and BDNF levels in the central nervous system of CD-1
        mice. Behav Brain Res 166: 124-130. http://dx.doi.Org/10.1016/i.bbr.2005.07.032.
Schelegle. ES: Miller. LA: Gershwin. LJ: Fanucchi.  MV: Van Winkle. LS: Gerriets. JE: Walbv. WF: Mitchell. V:
        Tarkington. BK: Wong. VJ: Baker. GL: Pantle. LM: Joad. JP: Pinkerton. KE: Wu. R: Evans.  MJ: Hyde.
        DM: Plopper, CG. (2003). Repeated episodes of  ozone inhalation amplifies the effects of allergen
        sensitization and inhalation on airway immune and structural development in Rhesus monkeys. Toxicol
        Appl Pharmacol 191: 74-85.
Schmelzer, KR: Wheelock, AM: Dettmer, K: Morin, D: Hammock, BD. (2006). The role of inflammatory
        mediators in the synergistic toxicity of ozone and 1-nitronaphthalene in rat airways. Environ Health
        Perspect  114:  1354-1360.
Schopke, R: Wolfer,  DP: Lipp, HP: Leisinger-Trigona, MC. (1991). Swimming navigation and structural
        variations of the infrapyramidal mossy fibers in the hippocampus of the mouse. Hippocampus 1: 315-
        328. http://dx.doi.org/10.1002/hipo.450Q10322.
Selevan, SG: Borkovec, L:  Slott,  Vl_: Zudova, Z: Rubes, J: Evenson, DP: Perreault, SD. (2000).  Semen quality
        and reproductive health  of young Czech men exposed to seasonal air pollution. Environ Health
        Perspect  108:  887-894.
Sharkhuu, T: Doerfler,  PL: Copeland, C: Luebke, RW: Gilmour, Ml. (2011). Effect of maternal exposure to ozone
        on reproductive outcome and immune, inflammatory, and allergic responses in the offspring. J
        Immunotoxicol 8:  183-194. http://dx.doi.org/10.3109/1547691X.2011.568978.
Simonian,  NA:  Coyle, JT (1996). Oxidative stress in neurodegenerative diseases. Annu Rev Pharmacol Toxicol
        36: 83-106.  http://dx.doi.org/10.1146/annurev.pa.36.040196.000503.
Slama. R:  Darrow. L: Parker. J: Woodruff. TJ: Strickland. M: Nieuwenhuiisen. M: Glinianaia. S: Hoggatt. KJ:
        Kannan,  S:  Hurley, F: Kalinka, J: Sram, R: Brauer, M: Wilhelm, M: Heinrich, J:  Ritz, B. (2008). Meeting
        report: Atmospheric pollution and human reproduction. Environ Health Perspect 116: 791-798.
Smith. KR: Jerrett. M: Anderson.  HR: Burnett. RT: Stone.  V: Derwent.  R: Atkinson. RW: Cohen. A: Shonkoff. SB:
        Krewski, D:  Pope,  CA, III: Thun, MJ: Thurston, G. (2009a). Public health benefits of strategies to
        reduce greenhouse-gas emissions: Health implications of short-lived greenhouse pollutants. Lancet
        374: 2091-2103. http://dx.doi.org/10.1016/s0140-6736(09)61716-5.
Sokol, RZ: Kraft, P: Fowler, IM: Mamet, R: Kim, E: Berhane, KT (2006). Exposure to environmental ozone alters
        semen quality. Environ Health Perspect 114: 360-365.
Son. JY: Cho. YS: Lee. JT. (2008). Effects of air pollution  on postneonatal infant mortality among firstborn
        infants in  Seoul, Korea: Case-crossover and time-series analyses. Arch Environ Occup Health 63:  10S-
        US.
Sousa. SI: Ferraz. C: Alvim-Ferraz. MC: Martins. FG: Vaz. LG: Pereira. MC. (2011). Spirometric tests to assess
        the prevalence of childhood asthma at Portuguese rural areas: Influence of exposure to high ozone
        levels. Environ Int 37: 474-478. http://dx.doi.Org/10.1016/i.envint.2010.11.014.
Sousa. SIV: Pereira. MMC: Martins. FG: Alvim-Ferraz. CM. (2008). Identification of regions with high ozone
        concentrations aiming the impact assessment on childhood asthma. Hum Ecol Risk Assess 14: 610-
        622. http://dx.doi.org/10.1080/10807030802074147.
Draft - Do Not Cite or Quote                        7-97                                  September 2011

-------
Sousa. SIV: Alvim-Ferraz. MCM: Martins. FG: Pereira. MC. (2009). Ozone exposure and its influence on the
        worsening of childhood asthma. Allergy 64: 1046-1055.  http://dx.doi.org/10.1111/1.1398-
        9995.2009.01946.x.
Sram. RJ: Binkova. B: Rossner. P: Rubes. J: Topinka. J: Deimek. J. (1999). Adverse reproductive outcomes
        from exposure to environmental mutagens. Mutat Res-Fundam Mol Mech Mutagen 428: 203-215.
        http://dx.doi. orq/10.1016/S1383-5742(99)00048-4.
Stedman. JR: Kent. AJ. (2008). An analysis of the spatial patterns of human health related surface ozone
        metrics across the UK in 1995, 2003 and 2005. Atmos Environ 42: 1702-1716.
Stockstill. BL: Chang. L. -Y: Menache. MG: Mellick. PW: Mercer. RR: Crapo. JD. (1995). Bronchiolarized
        metaplasia and interstitial fibrosis in rat lungs chronically exposed to high ambient levels of ozone.
        Toxicol Appl Pharmacol  134:  251-263.
Stokinger. HE. (1962). Effects of air pollution in animals. In AC Stern (Ed.), Air pollution (Vol. 1, pp. 282-334).
        New York, NY: Academic Press.
Strickland, MJ: Klein, M: Correa, A: Reller, MD: Mahle, WT: Riehle-Colarusso, TJ: Botto, LD: Flanders, WD:
        Mulholland. JA: Siffel. C: Marcus. M: Tolbert. PE. (2009). Ambient air pollution and cardiovascular
        malformations in Atlanta, Georgia, 1986-2003.  Am J  Epidemiol 169: 1004-1014.
Tager, IB: Balmes, J: Lurmann, F: Ngo, L: Alcorn, S: Kunzli, N. (2005). Chronic exposure to ambient ozone and
        lung function in young adults. Epidemiology 16: 751-759.
        http://dx.doi.ora/10.1097/01.ede.0000183166.68809.bO.
Tamer, L: Calikoglu, M: Ates, NA: Yildirim,  H: Ercan, B:  Saritas, E: Unlu, A: Atik, U. (2004). Glutathione-S-
        transferase gene polymorphisms (GSTT1, GSTM1, GSTP1) as increased risk factors for asthma.
        Respirology 9: 493-498.
Tovalin, H: Valverde, M: Morandi, MT: Blanco, S: Whitehead, L: Rojas, E. (2006). DNAdamage in outdoor
        workers occupationally exposed to environmental air pollutants. Occup Environ Med 63: 230-236.
Iran. MU: Weir. AJ: Fanucchi. MV: Rodriguez. AE: Pantle. LM: Smiley-Jewell.  SM: Van Winkle. LS: Evans. MJ:
        Miller, LA: Schelegle, ES: Gershwin, LJ: Hyde,  DM: Plopper, CG. (2004). Smooth muscle hypertrophy
        in distal airways of sensitized infant rhesus monkeys exposed to house dust mite allergen. Clin Exp
        Allergy 34: 1627-1633.  http://dx.doi.orq/10.1111/i.1365-2222.2004.02057.x.
Tsai, S, -S: Chen, C, -C: Hsieh,  H, -J: Chang, C, -C: Yang, C, -Y. (2006). Air pollution and postneonatal mortality
        in a tropical city: Kaohsiung,  Taiwan. Inhal Toxicol 18: 185-189.
Tyler. WS: Tyler. NK: Last. JA: Gillespie. MJ: Barstow. TJ. (1988). Comparison of daily and seasonal exposures
        of young monkeys to ozone.  Toxicology 50: 131-144.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
Van Bree. L: Dormans. JAM. A:  Koren. HS: Devlin. RB:  Rombout. PJA. (2002). Attenuation and recovery of
        pulmonary injury in rats following short-term, repeated daily exposure to ozone. Inhal Toxicol 14: 883-
        900.
Van Winkle. LS: Baker. GL: Chan. JK: Schelegle. ES: Plopper. CG. (2010). Airway mast cells in a rhesus model
        of childhood allergic airways  disease. Toxicol Sci 116: 313-322. http://dx.doi.org/10.1093/toxsci/kfq119.
Vanquilder. HP: Freeman. WM.  (2011). The hippocampal neuroproteome with  aging and cognitive  decline: Past
        progress and future directions. Front Aging Neurosci 3: 8. http://dx.doi.org/10.3389/fnagi.2011.00008.
Veninga, TS. (1967). Toxicity of ozone in comparison with ionizing radiation. Strahlentherapie  134: 469-477.
Vriiheid. M: Martinez. D: Manzanares. S: Dadvand. P: Schembari. A: Rankin. J: Nieuwenhuiisen. M. (2011).
        Ambient air pollution and risk of congenital anomalies: A systematic review and meta-analysis. Environ
        Health Perspect 119: 598-606. http://dx.doi.org/10.1289/ehp. 1002946.
Wang. T. -N: Ko. Y. -C: Chao. Y. -Y: Huang. C. -C: Lin. R. -S.  (1999). Association between indoor and outdoor
        air pollution and adolescent asthma from 1995 to 1996 in Taiwan. Environ Res 81: 239-247.
Wang, XY: Hu, W: Tong, S. (2009c). Long-term exposure to gaseous air pollutants and cardio-respiratory
        mortality in Brisbane, Australia. Geospat Health 3: 257-263.
Draft - Do Not Cite or Quote                        7-98                                  September 2011

-------
Wenten. M: Gauderman. WJ: Berhane. K: Lin. PC: Peters. J: Gilliland. FD. (2009). Functional variants in the
        catalase and myeloperoxidase genes, ambient air pollution, and respiratory-related school absences:
        An example of epistasis in gene-environment interactions. Am J Epidemiol 170: 1494-1501.
        http://dx.doi.org/10.1093/aie/kwp310.
Wilhelm, M: Ritz, B. (2005). Local variations in CO and particulate air pollution and adverse birth outcomes in
        Los Angeles County, California, USA. Environ  Health Perspect  113: 1212-1221.
Witschi. H. (1991). Effects of oxygen and ozone on mouse lung tumorigenesis. Exp Lung Res 17: 473-483.
Witschi, H: Wilson, DW: Plopper, CG. (1993). Modulation of N-nitrosodiethylamine-induced hamster lung tumors
        by ozone. Toxicology 77: 193-202.
Witschi. H: Espiritu. I: Pinkerton. KE: Murphy. K:  Maronpot. RR. (1999).  Ozone carcinogenesis revisited. Toxicol
        Sci 52: 162-167.
Wollmann. HA. (1998). Intrauterine growth restriction: Definition and etiology. Horm Res 49: 1-6.
Wood. AM: Harrison. RM: Semple. S: Avres. JG: Stocklev. RA. (2009). Outdoor air pollution is associated with
        disease severity in a1-antitrypsin deficiency. Eur Respir J 34: 346-353.
Woodruff. TJ: Darrow. LA: Parker. JD. (2008). Air pollution and postneonatal infant mortality in the United
        States, 1999-2002. Environ Health Perspect 116: 110-115.
Woodruff, TJ: Parker, JD: Darrow, LA: Slama, R: Bell, ML: Choi, H: Glinianaia, S: Hoggatt, KJ: Karr, CJ: Lobdell,
        DT: Wilhelm. M. (2009). Methodological issues in studies of air pollution and reproductive health.
        Environ Res 109: 311-320.
Woodruff, TJ: Parker, JD: Adams, K: Bell, ML: Gehring, U: Glinianaia, S: Ha, EH: Jalaludin, B: Slama, R. (2010).
        International Collaboration on Air Pollution and Pregnancy Outcomes (ICAPPO). Int J Environ Res
        Public Health 7: 2638-2652. http://dx.doi.org/10.3390/iierph7062638.
Yang, C, -Y: Hsieh, H, -J: Tsai, S, -S: Wu, T, -N: Chiu, H, -F. (2006).  Correlation between air pollution and
        postneonatal mortality in a subtropical city: Taipei, Taiwan. J Toxicol Environ Health A 69: 2033-2040.
        http://dx.doi. orq/10.1080/15287390600746181.
Zanobetti, A: Schwartz, J. (In Press) Ozone and survival in four cohorts with potentially  predisposing diseases.
        Am J Respir Grit Care Med. http://dx.doi.org/10.1164/rccm.201102-0227OC.
Zanobetti. A: Schwartz. J. (2007). Particulate air pollution, progression, and survival after myocardial infarction.
        Environ Health Perspect 115:  769-775.
Zanobetti. A: Bind. MAC: Schwartz. J. (2008). Particulate air pollution and survival in a COPD cohort. Environ
        Health Perspect 7: 48.
Zelac, RE: Cromrov, HL: Bolch, WE, Jr: Dunavant, BG: Bevis, HA. (1971a). Inhaled ozone as a mutagen: I
        chromosome aberrations induced in Chinese hamster lymphocytes. Environ Res 4: 262-282.
Zelac, RE: Cromrov, HL: Bolch, WE, Jr: Dunavant, BG: Bevis, HA. (1971b). Inhaled ozone as a mutagen: II
      effect on the frequency of chromosome aberrations observed in irradiated Chinese hamsters. Environ
      Res 4: 325-342.
Draft - Do Not Cite or Quote                        7-99                                  September 2011

-------
      8   POPULATIONS  POTENTIALLY AT  INCREASED
           RISK  FOR  OZONE-RELATED  HEALTH  EFFECTS

 1                   Interindividual variation in human responses to air pollution exposure suggests that some
 2                   groups are at increased risk for detrimental effects in response to ambient exposure to an
 3                   air pollutant. The NAAQS are intended to provide an adequate margin of safety for both
 4                   the population as a whole and those individuals potentially at increased risk for health
 5                   effects in response to ambient air pollution (Preface to this ISA). To facilitate the
 6                   identification of populations at greater risk for O3-related health effects, studies have
 7                   evaluated factors that may contribute to the susceptibility and/or vulnerability of an
 8                   individual to O3. The definitions of susceptibility and vulnerability have been found to
 9                   vary across studies, but in most instances "susceptibility" refers to biological or intrinsic
10                   factors (e.g., lifestage, sex) while "vulnerability" refers to  nonbiological or extrinsic
11                   factors (e.g., socioeconomic status [SES]) (U.S. EPA. 2010c. 2009d). Additionally, in
12                   some cases, the terms "at-risk" and "sensitive" populations have been used to encompass
13                   these concepts more generally. Previous IS As and reviews (Sacks etal.. 2011; U.S. EPA.
14                   2010c. 2009d) have used an all encompassing definition for "susceptible population" to
15                   focus on identifying the populations at greater risk for O3-induced heath effects and
16                   circumvent the need to distinguish between susceptible and vulnerable factors. In this
17                   chapter, "at-risk" groups are defined as those with characteristics that increase the risk of
18                   O3-related health effects in a population. These characteristics include various factors,
19                   such as genetic background, race, sex, lifestage, diet, preexisting disease, SES, and
20                   characteristics that may modify exposure to O3 (e.g., time  spent outdoors).

21                   Individuals, and ultimately populations,  could experience increased risk for O3-induced
22                   health effects due to:

23                       •   Intrinsically increased risk: This describes individuals at greater risk due to a
24                          biological mechanism;
25                       •   Extrinsically increased risk: This describes individuals at greater risk due to  an
26                          external, non-biological factor;  and
27                       •   Increased dose:  This describes individuals that have a greater dose at a given
28                          concentration due to breathing patterns or other factors

29                   In addition, some individuals might be placed at risk of experiencing a greater exposure
30                   by being exposed at higher concentrations. For example, individuals in  lower SES groups
31                   might be exposed to higher O3 concentrations due to less availability/use of home air
32                   conditioners (i.e., more  open windows on high O3 days).
      Draft - Do Not Cite or Quote                        8-1                                 September 2011

-------
 1                   Examples of potential factors intrinsically related to increased risk through biological
 2                   mechanisms are genetic background and sex, while extrinsic factors, such as SES, may
 3                   also increase the risk of O3-related health effects. However, some factors that may lead to
 4                   increased risk of O3-related health effects have both intrinsic and extrinsic components.
 5                   For example, SES may affect access to medical care, which could then affect the
 6                   presence of preexisting diseases and conditions often considered to be intrinsic factors.
 7                   Additionally, children tend to spend more time outdoors than adults, which increases
 8                   their dose of O3, but they also have intrinsic differences compared to adults based on
 9                   lung growth and development.

10                   The emphasis of this chapter is on identifying and understanding the characteristics that
11                   potentially increase the risk of O3-related health effects, regardless of whether the
12                   increased risk at a given concentration is due to intrinsic factors, extrinsic factors, or
13                   increased dose. The following sections examine factors that may modify the association
14                   between O3 and health effects, but does not categorize them as intrinsic factors, extrinsic
15                   factors, or increased dose, due to the convoluted and often connected pathways between
16                   factors. However, the different role of intrinsic risk, extrinsic risk, and increased dose are
17                   discussed as appropriate throughout the chapter.

18                   Epidemiologic studies often conduct stratified analyses to identify the  presence or
19                   absence of effect measure modification to indicate whether O3 differentially affects
20                   certain populations. This allows researchers to examine the effects of O3 exposure within
21                   each group under study. A thorough evaluation of potential effect measure modifiers may
22                   help identify populations that are more at-risk to health effects associated with O3
23                   exposure. Toxicological and controlled human exposure studies can provide support and
24                   biological plausibility for factors that may lead to increased risk for O3-related health
25                   effects through the study of animal models of disease or individuals with underlying
26                   disease or genetic polymorphisms that allow for comparisons between subgroups. The
27                   results from these studies, combined with those results obtained through stratified
28                   analyses in epidemiologic studies, comprise the overall weight of evidence for the
29                   increased risk of specific populations to O3-related health effects.

30                   This chapter discusses the epidemiologic, controlled human exposure, and toxicological
31                   studies evaluated in Chapters 5, 6, and 7 that provide information on potential at-risk
32                   populations. The epidemiologic studies included in this chapter consist of only those
33                   studies that presented stratified results (e.g., males versus females or <65 years of age
34                   versus > 65 years of age).  This approach allowed for a comparison between populations
35                   exposed to similar O3 concentrations and within the same study design. Numerous
36                   studies that focus on only one potentially at-risk population are described in previous
37                   chapters, but these studies are not discussed in detail in this chapter because of the lack of
      Draft - Do Not Cite or Quote                       8-2                                 September 2011

-------
 1                   an adequate comparison group within the study. Controlled human exposure studies that
 2                   consisted of individuals with an underlying disease or genetic polymorphism, or studies
 3                   that categorized the study population by age, race, etc., and toxicological studies that
 4                   used animal models of disease were also evaluated for coherence and biological
 5                   plausibility.

 6                   Factors examined that may lead to increased risk of O3-related health effects based on the
 7                   overall evidence integrated across disciplines are described in greater detail in the
 8                   following sections.
          8.1    Preexisting  Disease/Conditions

 9                   Individuals with certain preexisting diseases are likely to constitute an at-risk population.
10                   Previous O3 AQCDs concluded that some people with preexisting pulmonary disease,
11                   especially asthma, are among those at increased risk from O3 exposure. Extensive
12                   toxicological evidence was available indicating that altered physiological, morphological
13                   and biochemical states typical of respiratory diseases, such as asthma, COPD, and
14                   chronic bronchitis, may render people sensitive to additional oxidative burden induced by
15                   O3 exposure. In addition, a number of epidemiologic studies found that some individuals
16                   with respiratory diseases are at increased risk of O3-related effects. Little evidence,
17                   however, was available on the potential for increased risk for people with other
18                   preexisting conditions, such as cardiovascular diseases.

19                   Recent studies that examined whether preexisting diseases and conditions lead to
20                   increased risk of O3-induced health effects were identified and are summarized below.
21                   Table 8-1 displays the prevalence rates of some of these conditions categorized by age
22                   and region among adults in the U.S. population; data for children, when available, are
23                   presented within sections. Substantial proportions of the U.S. population are affected by
24                   these conditions and therefore may represent a potentially large at-risk population. While
25                   these diseases and conditions are intrinsic to individuals, the pathways to their
26                   development may have intrinsic or extrinsic origins.
      Draft - Do Not Cite or Quote                        8-3                                 September 2011

-------
Table 8-1 Prevalence of respiratory diseases, cardiovascular diseases, and
diabetes among adults by age and region in the U.S.
Adults

Chronic Disease/Condition
N (in thousands)

Age
18-44

45-64

65-74

75+
Region
Northeast

Midwest

South

West
Respiratory Diseases
Asthma8
28,260
13.5
12.0
12.0
10.0
12.8
13.4
11.2
13.9
COPD
Chronic Bronchitis
Emphysema
9,832
3,789
3.2
0.2
5.5
2.0
5.9
5.7
5.3
5.0
3.4
1.2
4.8
1.9
5.2
1.9
2.9
1.3
Cardiovascular Diseases
All Heart Disease
Coronary Heart Disease
Hypertension
Diabetes
26,628
14,428
56,159
18,651
4.6
1.1
8.7
2.3
12.3
6.7
32.5
12.1
26.7
16.9
54.4
20.4
39.2
26.7
61.1
17.3
11.3
5.7
22.9
4.5
12.7
6.5
24.1
7.6
12.2
7.3
27.1
9.0
9.9
4.9
20.6
7.7
'Asthma prevalence is reported for "ever had asthma"
Source: Statistics for adults: Pleis et al. (2009)
             8.1.1   Influenza/Infections

 1                   Recent studies have indicated that underlying infections may increase the risk of
 2                   individuals to O3-related health effects, although there are only a limited number of
 3                   studies. A study of hospitalizations in Hong Kong reported that increased levels of
 4                   influenza intensity resulted in increased excess risk of respiratory disease hospitalizations
 5                   related to O3 exposure (Wong et al.. 2009). In addition, a study of lung function in
 6                   asthmatic children reported decreases in lung function with increased short-term O3
 7                   exposure for those with upper respiratory infections but not for those without infections
 8                   (Lewis et al.. 2005). Toxicological studies provide biological plausibility for the increase
 9                   in Os-induced health effects observed in epidemiologic studies that examined infections.
10                   Toxicological studies demonstrated that 0.08 ppm O3 increased streptococcus-induced
11                   mortality, regardless of whether O3 exposure  precedes or follows infection (Miller et al..
12                   1978; Coffin and Gardner. 1972; Coffin et al., 1967). Ozone exposure likely impairs host
13                   defenses, which may increase mortality due to an infectious agent. However, there is little
14                   toxicological evidence that infection or influenza itself renders an individual at greater
15                   risk of an O 3 -induced health effect.
             8.1.2   Asthma

16                   Previous O3 AQCDs identified individuals with asthma as a population at risk for O3-
17                   related health effects. Within the U.S., approximately 12% of adults have reported ever
      Draft - Do Not Cite or Quote                        8-4                                  September 2011

-------
 1                   having asthma (Pleis et al., 2009). The prevalence of asthma is approximately 7.2%.
 2                   16.2%, and 16.6% among U.S. children aged 0-4, 5-11, and 12-17, respectively (Bloom
 3                   et al.. 2008).

 4                   Multiple epidemiologic studies included within this ISA have evaluated the potential for
 5                   increased risk of O3-related health effects among individuals with asthma. A study of
 6                   lifeguards in Texas reported decreased lung function with short-term O3 exposure among
 7                   both individuals with and without asthma, however, the decrease was greater among
 8                   those with asthma (Thaller et al., 2008). A Mexican study of children ages 6-14 detected
 9                   an association between short-term O3 and wheeze, cough, and bronchodilator use among
10                   asthmatics but not non-asthmatics, although this may have been the result of a small
11                   non-asthmatic population (Escamilla-Nunez et al.. 2008). A study of the modification of
12                   the effect of greater O3 associated decreases in short-term O3 exposure on lung function
13                   by airway hyperresponsiveness (AHR) (a condition common among asthmatics) reported
14                   greater O3-associated decreases in lung function in elderly individuals with AHR,
15                   especially among those who were obese (Alexeeff et al.. 2007). However, no evidence
16                   for increased risk was found in a study performed among children in Mexico City that
17                   examined the effect of short-term O3 exposure on respiratory health (Barraza-Villarreal et
18                   al.. 2008). In this study, a positive association was reported for airway inflammation
19                   among asthmatic children, but the observed association was similar in magnitude to that
20                   of non-asthmatics. Similarly, a study of children in California reported an association
21                   between O3 concentration and exhaled nitric oxide fraction (FeNO) that persisted both
22                   among children with and without asthma as well as those with and without respiratory
23                   allergy (Berhane etal., 2011). Finally, some studies have reported null results for both
24                   individuals with and without asthma. Khatri et al. (2009) found no association between
25                   short-term O3 exposure and altered lung function for either asthmatic or non-asthmatic
26                   adults, but did note a decrease in lung function among individuals with allergies.

27                   Additional evidence for difference in effects among asthmatics has been observed in
28                   studies that examined the association between O3 exposure and altered lung function by
29                   asthma medication use. A study of children with asthma living in Detroit reported a
30                   greater association between short-term O3 and lung function for corticosteroid users
31                   compared with noncorticosteroid users (Lewis et al.. 2005). Conversely, another study
32                   found decreased lung function among noncorticosteroid users compared to users,
33                   although in this study,  a large proportion of non-users were considered to be persistent
34                   asthmatics (Hernandez-Cadena et al.. 2009). Lung function was not related to short-term
35                   O3 exposure among corticosteroid users and non-users in a study taking place during the
36                   winter months  in Canada (Liu et al., 2009a). Additionally, a study of airway
37                   inflammation reported a counterintuitive inverse association with O3 of similar
38                   magnitude for all groups of corticosteroid users and non-users (Qian et al.. 2009).
      Draft - Do Not Cite or Quote                        8-5                                 September 2011

-------
 1                   Controlled human exposure studies that have examined the effects of O3 on both
 2                   individuals with asthma and healthy controls are limited. Based on studies reviewed in
 3                   the 1996 and 2006 O3 AQCDs, subjects with asthma appeared to be more sensitive to
 4                   acute effects of O3 in terms of FEVi and inflammatory responses than healthy
 5                   non-asthmatic subjects. For instance, Horstman et al. (1995) observed that
 6                   mild-to-moderate asthmatics, on average, experienced double the O3-induced FEVi
 7                   decrement of healthy subjects (19% versus 10%, respectively, p = 0.04). Moreover, a
 8                   statistically significant positive correlation between FEVi  responses to O3 exposure and
 9                   baseline lung function was observed in individuals with asthma, i.e., responses increased
10                   with severity of disease. Minimal evidence exists suggesting that individuals with asthma
11                   have smaller O3-induced FEVi decrements than healthy subjects (3% versus 8%,
12                   respectively) (Mudway et al.. 2001). However, the asthmatics in that study  also tended to
13                   be older than the healthy subjects, which could partially explain their lesser response
14                   since FEVi  responses to O3 exposure diminish with age. Individuals with asthma also
15                   had significantly more neutrophils in the BALF (18 hours  postexposure) than similarly
16                   exposed healthy individuals (Pedenetal.. 1997; Scannell et al.. 1996; Bashaetal.. 1994).
17                   Furthermore, one newer study examined the effects of O3 on both individuals with atopic
18                   asthma and healthy controls (Hernandez et al.. 2010). Greater numbers of neutrophils,
19                   higher levels of cytokines and hyaluronan, and greater expression of macrophage
20                   cell-surface markers were observed in induced sputum of atopic asthmatics compared
21                   with healthy controls. Differences in O3-induced epithelial cytokine expression were
22                   noted in bronchial biopsy samples from asthmatics and healthy controls (Bosson et al..
23                   2003). Cell-surface marker and cytokine expression results, and the presence of
24                   hyaluronan, are consistent with O3 having greater effects on innate and adaptive
25                   immunity in these asthmatic individuals (see Section 5.4.2.2). In addition, older studies
26                   have demonstrated that O3 exposure leads to increased bronchial reactivity to inhaled
27                   allergens in mild allergic asthmatics (Kehrl et al.. 1999; Torres et  al.. 1996)  and to the
28                   influx of eosinophils in individuals with pre-existing allergic disease (Vagaggini et al..
29                   2002; Pedenetal.. 1995). Taken together, these results point to several mechanistic
30                   pathways which could account for the enhanced sensitivity to O3  in subjects with asthma
31                   (see Section 5.4.2.2).

32                   Toxicological studies provide biological plausibility for greater effects of O3 among
33                   those with asthma or AHR. In animal toxicological studies, an asthmatic phenotype is
34                   modeled by allergic sensitization of the respiratory tract. Many of the studies that provide
35                   evidence that O3 exposure is an inducer of AHR and remodeling utilize these types of
36                   animal models. For example,  a series of experiments in infant rhesus monkeys have
37                   shown these effects, but only  in monkeys  sensitized to house dust mite allergen (Fanucchi
38                   et al.. 2006: Joad et al.. 2006: Schelegle et al.. 2003). Similarly, Funabashi et al. (2004)
39                   demonstrated adverse changes in pulmonary function in mice  exposed to O3, and Wagner

      Draft - Do Not Cite or Quote                       8-6                                 September 2011

-------
 1                   et al. (2007) demonstrated enhanced inflammatory responses in rats exposed to O3, but
 2                   only in animals sensitized to allergen. In general, it is the combined effects of O3 and
 3                   allergic sensitization which result in measurable effects on pulmonary function. In a
 4                   bleomycin induced pulmonary fibrosis model, exposure to 250 ppb O3 for 5 days
 5                   increased pulmonary inflammation and fibrosis, along with the frequency of
 6                   bronchopneumonia in rats. Thus, short-term exposure to O3 may enhance damage in a
 7                   previously injured lung (Oyarzun et al., 2005).

 8                   In the 2006 O3 AQCD, the potential for individuals with asthma to have greater risk of
 9                   O3-related health effects was supported by a number of controlled human exposure
10                   studies, evidence from toxicological studies, and a limited number of epidemiologic
11                   studies. Overall, in the recent epidemiologic literature some, but not all, studies report
12                   greater risk of health effects among individuals with asthma. Studies examining effect
13                   measure modification of the relationship between  short-term O3 exposure and altered
14                   lung function by corticosteroid use provided limited evidence of O3-related health
15                   effects. Inconsistent findings observed in epidemiologic studies may be due to the
16                   differences in O3 concentration across the studies. Additionally, recent studies of
17                   behavioral responses have found that studies do not take into account individual
18                   behavioral adaptations to forecasted air pollution levels (such as avoidance and reduced
19                   time outdoors), which may underestimate the  observed associations in studies that
20                   examined the effect of O3 exposure on respiratory health (Neidell and Kinney. 2010).
21                   This could explain some inconsistency observed among recent epidemiologic studies.
22                   The evidence from controlled human exposure studies provides support for increased
23                   detriments in FEVi and greater inflammatory responses to O3 in individuals with asthma
24                   than in healthy individuals without a history of asthma.  The collective evidence for
25                   increased risk of O3-related health effects among individuals with asthma from controlled
26                   human exposure studies is supported by recent toxicological studies which provide
27                   biological plausibility for heightened risk of asthmatics to respiratory effects due to O3
28                   exposure.
             8.1.3   Chronic Obstructive Pulmonary Disease (COPD)

29                   Although not extensively examined in the literature, initial evidence suggests that
30                   preexisting COPD may modify the association between short-term O3 exposure and
31                   cardiovascular-related health effects. In the U.S. over 4% of adults report having chronic
32                   bronchitis and almost 2% report having emphysema, both of which are classified as
33                   COPD (Pleis et al.. 2009).
      Draft - Do Not Cite or Quote                       8-7                                September 2011

-------
 1                   In a recent study, Peel et al. (2007) found that individuals with COPD were at increased
 2                   risk of cardiovascular ED visits in response to short-term O3 exposure compared to
 3                   healthy individuals in Atlanta, GA. The authors reported that short-term O3 exposure was
 4                   associated with higher odds of an ED visit for peripheral and cerebrovascular disease
 5                   among individuals with COPD compared to individuals without COPD. However,
 6                   preexisting COPD did not increase the odds of hospitalization for all CVD outcomes (i.e.
 7                   IHD, dysrhythmia, or congestive heart failure). In an additional study performed in
 8                   Taiwan, both individuals with and without COPD had higher odds of congestive heart
 9                   failure associated with O3 exposure on warm days (Lee et al.. 2008a). An additional
10                   study also found no association between O3  exposure and lung function regardless of
11                   whether the study participant had COPD or other health issues (asthma or IHD) (Lagorio
12                   et al.. 2006).

13                   Recent epidemiologic evidence indicates that persons with COPD may have increased
14                   O3-related cardiovascular effects, but little information is available for other O3-related
15                   health effects among individuals with COPD.
             8.1.4   Cardiovascular Disease

16                   Cardiovascular disease (CVD) has become increasingly prevalent in the U.S., with about
17                   12% of adults reporting a diagnosis of heart disease (Table 8-1). A high prevalence of
18                   other cardiovascular-related conditions has also been observed, such as hypertension
19                   which is prevalent among approximately 24% of adults. In the 2006 O3 AQCD, little
20                   evidence was available regarding preexisting CVD as a susceptibility factor. Recent
21                   epidemiologic studies have examined cardiovascular-related diseases as modifiers of the
22                   O3-outcome associations; however, no recent evidence is available from controlled
23                   human exposure studies or toxicological studies.

24                   Peel et al. (2007) compared the associations between short-term O3 exposure and
25                   cardiovascular ED visits in Atlanta, GA among multiple comorbid conditions. The
26                   authors found no evidence of increased risk of cardiovascular ED visits in individuals
27                   previously diagnosed with dysrhythmia, congestive heart failure, or hypertension
28                   compared to healthy individuals. Similarly, a study in France examined the association
29                   between O3 concentrations and ischemic cerebrovascular events (ICVE) and myocardial
30                   infarction (MI) and the influence of multiple vascular risk factors on any observed
31                   associations (Henrotin et al.. 2010). The association between O3 exposure and ICVE was
32                   elevated for individuals with multiple risk factors, specifically individuals with diabetes
33                   or hypertension. For the association between O3 and MI, increased odds were apparent
34                   only for those with hypercholesterolaemia. In a study conducted in Taiwan, a positive
      Draft - Do Not Cite or Quote                       8-8                                 September 2011

-------
 1                   association was observed for O3 on warm days and congestive heart failure hospital
 2                   admissions (HAs), but the association did not differ between individuals with/without
 3                   hypertension or with/without dysrhythmia (Lee et al.. 2008a). Another study in Taiwan
 4                   reported that the association between O3 levels and ED visits for arrhythmias were
 5                   greater on warm days among those with congestive heart failure compared to those
 6                   without congestive heart failure; however, the estimate and 95% CIs for those without
 7                   congestive heart failure is completely contained within the 95% CI of those with
 8                   congestive heart failure (Chiu and Yang. 2009).

 9                   Although not studied extensively, a study has examined the increased risk of O3-related
10                   changes in blood markers for individuals with CVD. There was a greater association
11                   between O3 exposure and some, but not all, blood inflammatory markers among
12                   individuals with a history of CVD. Liao et al. (2005) found that fibrinogen was positively
13                   associated with short-term O3 exposure but this association was present only among
14                   individuals with a history of CVD. No association was observed among those without a
15                   history of CVD. However, for another biomarker (vWF), CVD status did not modify the
16                   positive association with short-term O3 exposure (Liao et al.. 2005).

17                   Mortality studies provide some evidence for a potential increase in O3-induced mortality
18                   in individuals with preexisting atrial fibrillation and atherosclerosis. In a study of 48 U.S.
19                   cities, increased risk of mortality with short-term O3 exposure was observed only among
20                   individuals with secondary atrial fibrillation (Medina-Ramon and Schwartz. 2008). No
21                   association was observed for short-term O3 exposure and mortality in a study of
22                   individuals with diabetes with or without CVD prior to death; however, there was some
23                   evidence of increased risk of mortality during the warm season if individuals had diabetes
24                   and atherosclerosis compared to only having diabetes (Goldberg et al..  2006).

25                   Finally, although not extensively examined, a study explored whether a preexisting CVD
26                   increased the risk of an O3-induced respiratory effect. Lagorio et al. (2006) examined the
27                   effect of O3 exposure on lung function among participants with a variety of preexisting
28                   diseases, including IHD. No association was observed regardless of whether the
29                   participant had IHD.

30                   Overall, most short-term exposure studies did not report increased O3-related health
31                   effects for individuals with preexisting CVD, with the possible exception of O3 exposure
32                   and mortality. Future research among those with CVD compared to those without will
33                   increase the understanding of potential increased risk of O3-related health effects among
34                   this group.
      Draft - Do Not Cite or Quote                        8-9                                 September 2011

-------
             8.1.5   Diabetes

 1                   Recent literature has not extensively examined whether individuals with diabetes (about
 2                   8% of U.S. adults) are potentially at increased risk of O3-related health effects. In a study
 3                   of short-term O3 exposure and cardiovascular ED visits in Atlanta, GA, no association
 4                   was observed for individuals with or without diabetes (Peel et al., 2007). A similar study
 5                   conducted in Taiwan reported a positive association between O3 exposure on warm days
 6                   and HAs for congestive heart failure; however, no modification of the association by
 7                   diabetes was observed (Lee et al.. 2008a). Finally, in a study of O3 exposure and ED
 8                   visits for arrhythmia in Taiwan, there was no evidence of effect measure modification by
 9                   diabetes on warm or cool days (Chiu and Yang. 2009).
             8.1.6   Hyperthyroidism

10                   Hyperthyroidism has been identified in toxicological studies as a potential factor that may
11                   lead to increased risk of O3-related health effects but has not yet been explored in
12                   epidemiologic or controlled human exposure studies. Lung damage and inflammation due
13                   to oxidative stress may be modulated by thyroid hormones. Compared to controls,
14                   hyperthyroid rats exhibited elevated levels of BAL neutrophils and albumin after a 4-hour
15                   exposure to O3, indicating O3-induced inflammation and damage. Hyperthyroidism did
16                   not affect production of reactive oxygen or nitrogen species, but BAL phospholipids were
17                   increased, indicating greater activation of Type II cells and surfactant protein production
18                   compared to normal rats (Huffman et al.. 2006). Thus, this study provides some
19                   underlying evidence which suggests that individuals with hyperthyroidism may represent
20                   an at-risk population.
          8.2    Lifestage

21                   The 1996 and 2006 O3 AQCDs identified children, especially those with asthma, and
22                   older adults as at-risk populations. These previous AQCDs reported clinical evidence that
23                   children have greater spirometric responses to O3 than middle-aged and older adults
24                   (U.S. EPA.  1996a). Similar results were observed for symptomatic responses and O3
25                   exposure. Among older adults, most studies reported in the 2006 O3 AQCD reported
26                   greater effects of short-term O3 exposure and mortality compared to other age groups.
27                   New evidence, summarized below, further supports these findings.
      Draft - Do Not Cite or Quote                       8-10                                September 2011

-------
            8.2.1    Children

 1                   The 2000 Census reported that 28.6% of the U.S. population was under 20 years of age,
 2                   with 14.1% under the age of 10 (SSDAN CensusScope. 2010a). Children are considered
 3                   to be more at risk for O3-related health effects compared to adults because they spend
 4                   more time outside and are more highly active, especially during the summer when O3
 5                   concentrations are the highest (U.S. EPA. 2006b). Moreover, children's respiratory
 6                   systems are undergoing lung growth until about 18-20 years  of age and are therefore
 7                   thought to be intrinsically more at risk for O3-induced damage (U.S. EPA. 2006b).

 8                   The 1996 O3 AQCD, reported clinical evidence that children, adolescents, and young
 9                   adults (<18 years of age) appear, on average, to have nearly equivalent spirometric
10                   responses to O3 exposure, but have greater responses than middle-aged and older adults
11                   (U.S. EPA. 1996a). Sycalmptomatic responses (e.g., cough, shortness of breath, pain on
12                   deep inspiration) to O3 exposure,  however, appear to increase with age until early
13                   adulthood and then gradually decrease with increasing age (U.S. EPA. 1996a). For
14                   subjects aged 18-36 years, McDonnell et al. (1999)  reported that symptom responses
15                   from O3 exposure also decrease with increasing age. Complete lung growth and
16                   development is not achieved until 18-20 years of age in women  and the early 20s for
17                   men; pulmonary function is at its  maximum during this time  as well. Additionally, PBPK
18                   modeling reported regional extraction of O3 to be higher in infants compared to adults.
19                   This is thought to be due to the smaller nasal and pulmonary regions' surface area in
20                   children under the age of 5 years compared to the total airway surface area observed in
21                   adults (Sarangapani et al., 2003).

22                   Recent  epidemiologic studies have been performed  examining different age groups and
23                   their susceptibility to O3-related respiratory HAs and emergency department (ED) visits.
24                   A study in Cyprus of short-term O3 concentrations and respiratory HA detected possible
25                   effect measure modification by age with a larger association  among individuals <
26                   15 years of age compared with those > 15 years of age. However, this difference was
27                   only apparent with a 2-day lag (Middleton et al., 2008). Similarly, a Canadian study of
28                   asthma-ED visits reported the strongest O3-related associations among 5- to 14-year olds
29                   compared to the other age groups (ages examined 0-75+) (Villeneuve et al., 2007).
30                   Greater O3-associated change in asthma-related ED visits were also reported among
31                   children (<15 years) as compared to adults (15 to 64 years) in a study from Finland
32                   (Halonen et al.. 2009). A study of New York City HAs demonstrated an increase in the
33                   association between O3 exposure  and asthma-related HAs for 6- to 18-year olds
34                   compared to those < 6 years old and those > 18 years old (Silverman and Ito. 2010).
35                   When examining long-term O3 exposure and asthma HA among children, associations
36                   were determined to be larger among children 1 to 2  years old compared to children 2 to 6
      Draft - Do Not Cite or Quote                      8-11                                September 2011

-------
 1                   years old (Lin et al., 2008b). A few studies reported positive associations among both
 2                   children and adults and no modification of the effect by age. A study performed in Hong
 3                   Kong examined O3 exposure and asthma-related HAs for ages 0 to!4, 15 to 65, and >65
 4                   (Ko et al.. 2007). The researchers reported that the association was greater among the 0 to
 5                   14 and 14 to 65 age groups compared to the >65 age group. Another study looking at
 6                   asthma-related ED visits and O3 exposure in Maine reported positive associations for all
 7                   age groups (ages 2 to 65) (Paulu and Smith. 2008). Effects of O3 exposure on asthma
 8                   hospitalizations among both children and adults (<18 and >18 years old) were
 9                   demonstrated in a study in Washington, but only children (<18 years of age) had
10                   statistically significant results at lag day 0, which the authors wrote, "suggests that
11                   children are more immediately responsive to adverse effects of O3 exposure" (Mar and
12                   Koenig. 2009).

13                   The evidence observed in epidemiologic studies is supported by recent toxicological
14                   studies which observed O3-induced health effects in immature animals. Early life
15                   exposures of multiple species of laboratory animals, including infant monkeys,  resulted
16                   in changes in conducting airways at the cellular, functional, ultra-structural, and
17                   morphological  levels. Carey et al. (2007) conducted a study of O3 exposure in infant
18                   rhesus macaques, whose nasal airways closely resemble that of humans. Monkeys were
19                   exposed either  acutely for 5 days to 0.5 ppm O3, or episodically for 5 biweekly cycles
20                   alternating 5 days of 0.5 ppm O3 with 9 days of filtered air, designed to mimic human
21                   exposure (70 days total). All monkeys acutely exposed to O3 had moderate to marked
22                   necrotizing rhinitis, with focal regions of epithelial exfoliation, numerous  infiltrating
23                   neutrophils, and some eosinophils. The distribution, character, and severity of lesions in
24                   episodically exposed monkeys were similar to that of acutely exposed animals.  Neither
25                   group exhibited mucous cell metaplasia proximal to the lesions, a protective adaptation
26                   observed in adult monkeys exposed continuously to 0.3 ppm O3 in another study
27                   (Harkema et al., 1987a). Functional (increased airway resistance and responsiveness with
28                   antigen + O3 co-exposure) and cellular changes in conducting airways (increased
29                   numbers of inflammatory eosinophils) also manifested among the infant monkeys
30                   (Plopper et al..  2007). In addition, the lung structure of the conducting airways was
31                   significantly stunted or altered versus control animals and this aberrant development was
32                   persistent 6 months postexposure (Fanucchi et al.. 2006).

33                   Similarly, rat fetuses exposed to O3 in utero had significant ultrastructural changes in
34                   bronchiolar epithelium when examined near the end of gestation (Lopez et al., 2008). In
35                   addition, exposure of mice to mixtures of air pollutants early in development affected pup
36                   lung cytokine levels (TNF, IL-1, KC, IL-6, and MCP-1) (Auten et al.. 2009). In utero
37                   exposure of animals to PM augmented O3-induced airway hyper-reactivity in these pups
38                   as juveniles.
      Draft - Do Not Cite or Quote                       8-12                                September 2011

-------
 1                   Age may affect the inflammatory response to O3 exposure. In comparing neonatal mice
 2                   to adult mice, increased bronchoalveolar lavage (BAL) neutrophils were observed in four
 3                   strains of neonates 24 hours after exposure to 0.8 ppm O3 for 5 hours (Vancza et al.,
 4                   2009). Three of these strains also exhibited increased BAL protein, although the two
 5                   endpoints were not necessarily consistently correlated in a given strain. In some strains,
 6                   however, adults were more sensitive, indicating a strain-age interaction. Toxicological
 7                   studies reported that the difference in effects among younger lifestage may be due to
 8                   age-related changes in endogenous antioxidants and sensitivity to oxidative stress. A
 9                   recent study demonstrated that 0.25 ppm O3  exposure differentially alters expression of
10                   metalloproteinases in the skin of young (8 weeks old) and aged (18 months old) mice,
11                   indicating age-related susceptibility to oxidative stress (Fortino et al., 2007). Valacchi et
12                   al. (2007) found that aged mice had more vitamin E in their plasma but less in their lungs
13                   compared to young mice, which may affect their pulmonary antioxidant defenses. Servais
14                   et al. (2005) found higher levels of oxidative damage  indicators in immature (3 weeks
15                   old) and aged (20 months old) rats compared to adult rats, which were relatively resistant
16                   to an intermittent 7-day exposure to 0.5 ppm O3. Immature rats exhibited a higher
17                   ventilation rate, which may have increased exposure.  Additionally, a series of
18                   toxicological studies reported an association between O3 exposure and bradycardia that
19                   was present among young mice but not among older mice (Hamade et al.. 2010;
20                   Tankersley et al.. 2010; Hamade and Tankersley. 2009; Hamade et al., 2008). Regression
21                   analysis revealed a significant interaction between age and strain on heart rate, which
22                   implies that aging may affect heart rate differently between mouse strains (Tankersley et
23                   al.. 2010). The authors proposed that the genetic differences between the mice strains
24                   could be altering the formation of ROS, which tends to increase with age, thus
25                   modulating the changes in cardiopulmonary physiology after O3 exposure.

26                   The previous and current human clinical and toxicological studies reported evidence of
27                   increased risk from O3 exposure for younger ages, which provides coherence and
28                   biological plausibility to the epidemiologic studies on children. Recent studies of
29                   respiratory HAs and ED visits observed inconsistent findings for associations among
30                   children and young adults, although generally studies reported positive associations
31                   among both children and adults or just among children. For other outcomes, there were
32                   also inconsistent findings regarding increased risk of O3-related health effects. The
33                   interpretation of these studies is limited by the lack of consistency in comparison age
34                   groups and outcomes examined.
      Draft - Do Not Cite or Quote                       8-13                                 September 2011

-------
             8.2.2   Older Adults

 1                   Older adults may be at greater risk of health effects associated with O3 exposure through
 2                   a variety of intrinsic pathways. The gradual decline in physiological processes that occur
 3                   with aging may lead to increased risk of O3-related health effects (U.S. EPA. 2006a).
 4                   Older adults may also differ in amounts of exposure because diminished symptomatic
 5                   responses may allow the elderly to withstand increased continued O3 exposure. In
 6                   addition, older adults, in general, have a higher prevalence of preexisting diseases
 7                   compared to younger age groups and this may also lead to increased susceptibility to
 8                   O3-related health effects (see Table 8-1 that gives preexisting rates by age).With the
 9                   number of older Americans increasing in upcoming years (estimated to increase from
10                   12.4% of the U.S. population to 19.7% between 2000 to 2030, which is approximately 35
11                   million and 71.5 million individuals, respectively) this group represents a large
12                   population potentially at risk of O3 -related health effects (SSDAN CensusScope, 2010a;
13                   U.S. Census Bureau. 2010).

14                   Multiple epidemiologic studies of O3 exposure and HAs were stratified by age groups. A
15                   positive association was reported between O3 levels and respiratory HAs for adults >65
16                   years old but not for those adults aged  15 to 64 years (Halonen et al.. 2009). In the same
17                   study, no association was observed between O3 concentration and respiratory mortality
18                   among those >65 years old or those 15 to 64 years old; however, an inverse association
19                   between O3 concentration and cardiovascular mortality was present among individuals >
20                   65 years old but not among individuals < 65 years old. This inverse association among
21                   those >65 years old persisted when examining HAs for coronary heart disease. A study of
22                   CVD-related hospital visits in Bangkok, Thailand reported an increase in percent change
23                   for hospital visits with previous day and cumulative 2-day O3 levels among those >
24                   65 years old, whereas no association was present for individuals less than 65 years of age
25                   (Buadong et al.. 2009). No association was observed for current day or cumulative 3-day
26                   averages in any age group. A study examining O3 and HAs for CVD-related health
27                   effects reported no association for individuals aged 15 to 64 or individuals aged > 65
28                   years, although one lag-time did show an inverse effect for coronary heart disease among
29                   elderly that was not present among 15- to 64-year olds (Halonen et al.. 2009). No
30                   modification by age (40 to 64 year olds versus >64 year olds) was observed in a study
31                   from Brazil examining O3 levels and COPD ED visits (Arbex et al.. 2009).

32                   The majority of recent studies reported greater effects  of short-term O3 exposure and
33                   mortality among older adults, which is consistent with the findings of the 2006 O3
34                   AQCD. A study conducted in 48 cities across the U.S. reported larger effects among
35                   adults >65 years old compared to those < 65 years (Medina-Ramon and Schwartz. 2008).
36                   Further investigation of this study population revealed no association between O3
      Draft - Do Not Cite or Quote                       8-14                                 September 2011

-------
 1                   exposure and mortality until age 50 and a reduced effect after age 80 (Zanobetti and
 2                   Schwartz. 2008a). A study of 7 urban centers in Chile reported similar results, with
 3                   greater effects in adults >65 years old, however the effects were smaller among those
 4                   >85 years old compared to those in the 75 to 84 years old age range (Cakmak et al..
 5                   2007). More recently, a study conducted in the same area reported similar associations
 6                   between O3 exposure and mortality in adults aged < 64 years old and 65 to 74 years old,
 7                   but the risk was increased among older age groups (Cakmak et al., 2011). A study
 8                   performed in China reported greater effects in populations >45 years old (compared to 5
 9                   to 44 year olds), with statistically significant effects present only among those >65 years
10                   old (Kanetal.. 2008). An Italian study reported higher risk of all-cause mortality
11                   associated with increased O3 concentrations among individuals >85 year old as compared
12                   to those 35 to 84 years old. Those 65 to 74 and 75 to 84 years old did not show a greater
13                   increase in risk compared to those aged 35 to 64 years (Stafoggia et al.. 2010). The Air
14                   Pollution and Health: A European and North American Approach (APHENA) project
15                   examined the association between O3 exposure and mortality for those <75 and >
16                   75 years of age. In Canada, the associations for all-cause and cardiovascular mortality
17                   were greater among those >75 years old in the summer-only and all-year analyses. Age
18                   groups were not compared in the analysis for respiratory mortality in Canada. In the U.S.,
19                   the association for all-cause mortality was slightly greater for those <75 years of age
20                   compared to those >75 years old in summer-only analyses. No consistent pattern was
21                   observed for CVD mortality. In Europe, slightly larger associations for all-cause
22                   mortality were observed  in those <75 years old in all-year and summer-only analyses.
23                   Larger associations were reported among those <75years for CVD mortality in all-year
24                   analyses, but the reverse  was true for summer-only analyses (Katsouyanni et al.. 2009).

25                   Biological plausibility for increased risk among older adults is provided by clinical and
26                   toxicological studies. Respiratory symptom responses to O3 exposure appears to increase
27                   with age until early adulthood and then gradually decrease with increasing age (U.S.
28                   EPA. 1996a). which may put them at increased risk by withstanding continued O3
29                   exposure. Regarding cardiac outcomes, biological plausibility is provided by a
30                   toxicological study. O3 exposure resulted in an increase in left ventricular chamber
31                   dimensions at end diastole (LVEDD) in young and old mice, whereas decreases in left
32                   ventricular posterior wall thickness at end systole (PWTES) were only observed among
33                   older mice (Tankersley et al., 2010). Other toxicological studies also indicate increased
34                   susceptibility in older animals for some endpoints. The hippocampus, one of the main
35                   regions affected by age-related neurodegenerative diseases, may be more sensitive to
36                   oxidative damage in aged rats. In a study of young (47 days) and aged (900 days) rats
37                   exposed to 1 ppm O3 for 4 hours, O3-induced lipid peroxidation occurred to a greater
38                   extent in the striatum of young rats, whereas it was highest in the hippocampus in aged
39                   rats (Rivas-Arancibia et al.. 2000). In young mice, healing of skin wounds is not

      Draft - Do Not Cite or Quote                       8-15                                September 2011

-------
 1                   significantly affected by O3 exposure (Lim et al., 2006). However, exposure to 0.5 ppm
 2                   O3 for 6 h/day significantly delays wound closure in aged mice.

 3                   Although some outcomes reported mixed findings regarding an increase in risk for older
 4                   adults, recent studies of O3 exposure and mortality reported associations present for older
 5                   adults. This is consistent with the results reported in the 2006 O3 AQCD.
          8.3    Sex

 6                   The distribution of males and females in the U.S. is similar. In 2000, 49.1% of the U.S.
 7                   population was male and 50.9% were female. The distribution did vary by age with a
 8                   greater prevalence of females >65 years old compared to males (SSDAN CensusScope.
 9                   2010a). The 2006 O3 AQCD did not report evidence of differences between the sexes in
10                   health responses to O3 exposure. Recent epidemiologic studies have evaluated the effects
11                   of short-term and long-term exposure to O3 on multiple health endpoints stratified by sex
12                   and overall, the results are inconsistent.

13                   A study in Maine that examined short-term O3 concentrations and asthma ED visits
14                   detected greater effects among males ages 2 to!4 years and among females ages 15 to 34
15                   years compared to males and females in the same age groups  (no difference was detected
16                   for males and females aged 35 to  64) (Paulu and Smith. 2008). A Canadian study
17                   reported no associations between  short-term O3 and respiratory infection HAs for either
18                   boys or girls under the age of 15 (Lin et al., 2005). whereas another Canadian study
19                   reported a slightly higher but non-statistically significant increase in respiratory HA for
20                   males (mean ages 47.6 to 69.0 years) (Cakmak et al., 2006b).  A recent study from Hong
21                   Kong examining individuals of all ages reported no effect measure modification by sex
22                   for overall respiratory disease HAs, but did detect a greater excess risk of HAs for COPD
23                   among females compared to males (Wong et al.. 2009). Similarly a study in Brazil found
24                   higher effect estimates for COPD ED visits among females compared to males (Arbex et
25                   al.. 2009). Higher levels of respiratory HA with greater O3 concentrations was also
26                   observed for females in a study of individuals living in Cyprus (Middleton et al., 2008).
27                   A study of lung function unrelated to HA and ED visits was conducted among lifeguards
28                   in Texas and reported decreased lung function with increased O3 exposure among
29                   females but not males (Thaller et al.. 2008). This study included individuals aged 16 to 27
30                   years, and the majority of participants were male. A New York study found no effect
31                   measure modification of the association between long-term O3 exposure and asthma HA
32                   among males and females between 1 and 6 years old (Lin et al., 2008b).

33                   In addition to examining the potential modification of O3 associations with respiratory
34                   outcomes by sex,  studies also examined cardiovascular-related outcomes specifically

      Draft - Do Not Cite or Quote                      8-16                                September 2011

-------
 1                   HAs and ED visits. All of these studies reported no effect modification by sex with some
 2                   studies reporting null associations for both males and females (Wong et al.. 2009;
 3                   Middleton et al., 2008; Villeneuve et al.. 2006a) and one study reporting a positive
 4                   associations for both sexes (Cakmak et al.. 2006a). A French study examining the
 5                   associations between O3 concentrations and risk of ischemic strokes (not limited to ED
 6                   visits or HAs) reported no association for either males or females with lags of 0, 2, or
 7                   3 days (Henrotin et al., 2007). A positive association was reported for males with a lag of
 8                   1 day, but this association was null for females. The authors noted that men in the study
 9                   had much higher rates of current and former smoking than women (67.4% versus 9.3%).

10                   A biomarker study investigating the effects of O3 concentrations on high-sensitivity
11                   C-reactive protein (hs-CRP), fibrinogen, and white blood cell (WBC) count, reported
12                   observations for various lag times ranging from 0 to 7 days (Steinvil et al.. 2008). Most
13                   of the associations were null for males and females although one association between O3
14                   and fibrinogen was positive for  males and null for females (lag day 4); however, this
15                   positive association was null or negative when other pollutants were included in the
16                   model. Only one study examining correlations between O3 levels and oxidative DNA
17                   damage examined results stratified by sex. In this study Palli et al. (2009) reported
18                   stronger correlations for males than females, both during short-term exposure (less than
19                   30 days) and long-term exposure (0-90 days). However, the authors commented that this
20                   difference could have been partially explained by different distributions of exposure to
21                   traffic pollution at work.

22                   A few studies have examined the association between short-term O3 concentrations and
23                   mortality stratified by sex and in contrast with studies of other endpoints, were more
24                   consistent in reporting elevated  risks among females. These studies, conducted in the
25                   U.S. (Medina-Ramon and Schwartz. 2008). Italy (Stafoggia et al.. 2010). and Asia (Kan
26                   et al.. 2008). reported larger effect estimates in females compared to males. In the U.S.
27                   study, the elevated risk of mortality among females was greater specifically among those
28                   >60 years old (Medina-Ramon and Schwartz. 2008). However, a recent study in Chile
29                   reported similar associations between O3 exposure and mortality among both men and
30                   women (Cakmak et al.. 2011). One long-term O3 exposure study of respiratory mortality
31                   stratified their results by sex and reported relative risks of 1.01 (95 % CI: 0.99, 1.04) for
32                   males and 1.04 (95% CIs 1.03,  1.07) for females (Jerrett et al.. 2009).

33                   Experimental research provided a further understanding of the underlying mechanisms
34                   that may explain a possible differential risk in O3-related health  effects among males and
35                   females. Several studies have suggested that physiological differences between sexes
36                   may predispose females to a greater susceptibility to O3. In females, lower plasma and
37                   nasal lavage fluid (NLF) levels  of uric acid (most prevalent antioxidant), the initial
      Draft - Do Not Cite or Quote                       8-17                                 September 2011

-------
 1                   defense mechanism of O3 neutralization, may be a contributing factor (Housley et al.,
 2                   1996). Consequently, reduced absorption of O3 in the upper airways of females may
 3                   promote its deeper penetration. Dosimetric measurements have shown that the absorption
 4                   distribution of O3 is independent of sex when absorption is normalized to anatomical
 5                   dead space (Bush et al.,  1996). Thus, a differential removal of O3 by uric acid seems to
 6                   be minimal. In general, the physiologic response of young healthy females to O3
 7                   exposure appears comparable to the response of young males (Hazucha et al., 2003). A
 8                   few studies have examined changes in O3 responses during various menstrual cycle
 9                   phases. Lung function response to O3 was enhanced during the follicular phase of the
10                   menstrual cycle compared to the luteal phase in a small study of women (Fox et al..
11                   1993). However, Seal et al. (1996) later reported no effect of menstrual cycle phase in
12                   their analysis of responses from 150 women, but conceded that the methods used by Fox
13                   et al. (1993) more precisely defined the menstrual cycle phase. Another study also
14                   reported no difference in responses among females during the  follicular and luteal phases
15                   of their cycle (Weinmann et al.. 1995a). Additionally, in this study the responses in
16                   women were comparable to those reported for men in the  study. In a toxicological study,
17                   small differences in effects by sex were seen in adult mice with respect to pulmonary
18                   inflammation and injury after a 5-h exposure to 0.8 ppm O3, and although adult females
19                   were generally more susceptible, these differences were strain-dependent, with some
20                   strains exhibiting greater susceptibility in males (Vancza et al., 2009). The most obvious
21                   sex difference was apparent in lactating females, which incurred the greatest lung injury
22                   or inflammation among several of the strains.

23                   Overall, results have varied, with recent evidence for increased risk for O3-related health
24                   effects present for females in some studies and males in other studies. Most studies
25                   examining the associations O3 and mortality report females to be at greater risk than
26                   males. Little evidence is available regarding a difference between the sexes for other
27                   outcomes. Inconsistent findings were reported on whether effect measure modification
28                   exists by sex for respiratory and cardiovascular HAs and ED visits.
          8.4   Genetics

29                   Multiple studies that examined the effect of short- and long-term O3 exposure on
30                   respiratory function have focused on whether various gene profiles modify the effect of
31                   O3 on various health effects. A study of wheeze in infants reported larger associations
32                   between short-term O3 exposure and wheeze and difficulty breathing in infants whose
33                   mothers have asthma compared to infants of mothers without asthma, illustrating the
34                   potential for genetics to play a role in O3-related health effects (Triche et al.. 2006).
      Draft - Do Not Cite or Quote                       8-18                                 September 2011

-------
 1                  Multiple genes, including glutathione S-transferase Mu 1 (GSTM1) and tumor necrosis
 2                  factor-a (TNF-a) were evaluated in the 2006 O3 AQCD and found to have a "potential
 3                  role... in the innate susceptibility to O3." Studies performed since the 2006 O3 AQCD
 4                  have continued to examine the roles of GSTM1 and TNF-a on O3-related health effects
 5                  and have also examined other gene variants that may increase the risk of O3-related
 6                  health effects. Due to small sample sizes, many controlled human exposure studies are
 7                  limited in their ability to test genes with low frequency and therefore, some genes
 8                  important for O3-related health effects may not have been examined.

 9                  Epidemiologic studies that examined the effects of short-term exposure to O3 on lung
10                  function included analyses of potential gene-environment interactions. Romieu et al.
11                  (2006) reported an association between O3 and respiratory symptoms that were larger
12                  among children with GSTM1 null or glutathione S-transferase P 1 (GSTP1) Val/Val
13                  genotypes. However, results suggested that O3-associated decreases in lung function may
14                  be greater among children with GSTP1 lie/lie or Ile/Val compared to GSTP1 Val/Val.
15                  Alexeef et al. (2008) reported greater decreases in lung function among GSTP1 Val/Val
16                  adults than those with other genotypes. In addition, they detected greater decreases in
17                  lung function for adults with long GT dinucleotide repeats in heme-oxygenase-1
18                  (HMOX1) promoters.

19                  Several controlled human exposure studies have reported that genetic polymorphism of
20                  antioxidant enzymes may modulate pulmonary function and inflammatory response to O3
21                  challenge. It appears that healthy carriers of NAD(P)H quinone oxidoreductase 1 (NQO1)
22                  wild type (wt) in combination with GSTM1 null genotype had greater decreases in lung
23                  function parameters with exposure to O3 (Bergamaschi et al.. 2001). Adults with GSTM1
24                  null only genotype did not show the same response to O3. In contrast, asthmatic children
25                  with GSTM1 null genotype (Romieu et al.. 2004a) were reported to have greater
26                  decreases in lung function in relation to O3 exposure. In a similar study, Vagaggini et al.
27                  (2010) exposed mild-to-moderate asthmatics to O3 during moderate exercise. In subjects
28                  with NQO1 wt and GSTM1 null, there was no evidence of changes in lung function or
29                  inflammatory responses to O3. Kim et al. (2011) also recently conducted a study among
30                  young adults, about half of whom were GSTMl-null and half of whom were
31                  GSTM1-sufficient. They detected no difference in the  FEVi responses to O3 exposure by
32                  GSTM1 genotype.

33                  In a study of healthy volunteers with GSTM1 sufficient (n=19; 24 ± 3) and GSTM1 null
34                  (n=16; 25 ± 5) genotypes exposed to 400 ppb O3 for 2 hours with exercise, Alexis et al.
35                  (2009) found genotype effects  on inflammatory responses but not lung function responses
36                  to O3. At 4 hours post O3 exposure, individuals with either GSTM1 genotype had
37                  significant increases in sputum neutrophils with a tendency for a greater increase in
      Draft - Do Not Cite or Quote                      8-19                                September 2011

-------
 1                  GSTM1 sufficient than GSTM1 nulls. At 24 hours postexposure, neutrophils had
 2                  returned to baseline levels in the GSTM1 sufficient individuals. In the GSTM1 null
 3                  subjects, neutrophil levels increased from 4 to 24 hours and were significantly greater
 4                  than both baseline levels and levels at 24 hours in the GSTM1 sufficient individuals. In
 5                  addition, O3 exposure increased the expression of the surface marker CD 14 in airway
 6                  neutrophils of GSTM1 null subjects compared with GSTM1 sufficient subjects. CD 14
 7                  and TLR4 are co-receptors for endotoxin, and signaling through this innate immune
 8                  pathway has been shown to be important for a number of biological responses to O3
 9                  exposure in toxicological studies (Garantziotis et al., 2010; Hollingsworth et al., 2010;
10                  Hollingsworth et al.. 2004; Kleeberger et al.. 2000). Alexis et al. (2009) also
11                  demonstrated decreased numbers of airway macrophages at 4 and 24 hours following O3
12                  exposure in GSTM1 sufficient subjects. Airway macrophages in GSTM1 null subjects
13                  were greater in number and found to have greater oxidative burst and phagocytic
14                  capability than those of GSTM1 sufficient subjects. Airway macrophages and dendritic
15                  cells from GSTM1  null subjects exposed to O3 expressed higher levels of the surface
16                  marker HLA-DR, again suggesting activation of the innate immune system. Since there
17                  was no FA control in the Alexis et al. (2009) study, effects of the exposure other than O3
18                  cannot be ruled out. In general, the findings between these  studies are inconsistent and
19                  additional, better-controlled studies are needed to clarify the influence of genetic
20                  polymorphisms on  O3 responsiveness in humans.

21                  Several epidemiologic studies of long-term O3 exposure examined interactions with
22                  different gene variants, including GSTP1, HMOX1, and TNF-a using data from the
23                  Children's Health Study. A study among children reported a three-way interaction effect
24                  between He 105 homozygotes of GSTP1, O3 exposure, and playing more than two team
25                  sports, and new onset of asthma (Islam et al.. 2009). Additionally, Islam et al. found that
26                  non-Hispanic white children with less than 23 repeats in the HMOX1 gene had decreased
27                  risk of new-onset asthma (Islam et al.. 2008). ARG1 and ARG2 (encoded by arginases)
28                  modification were examined for the  association between genotypes and new-onset
29                  asthma (Salam et al.,  2009). Reduced asthma risk was observed among atopic children
30                  living in high O3 concentration areas and having the ARG1 haplotypes. There was no
31                  difference in risk for children with ARG2 haplotypes. A decreased risk of bronchitic
32                  symptoms was observed among asthmatic children in low O3 concentration areas with
33                  TNF-a variant G-308A (TNF-308GG genotype), a variant that may alter gene expression.
34                  There was no decrease in risk for children with this TNF-a variant and living in areas
35                  with high O3  concentrations. Additionally, this modification for high and low levels of
36                  O3 was not present among non-asthmatic children (Lee et al.. 2009b). Wenten et al.
37                  (2009) observed increased risk of respiratory-related school absences among children
38                  with variants of catalase (CAT) and myeloperoxidase (MPO) genes, especially when the
39                  children were living in high O3 concentration areas.

      Draft - Do Not Cite or Quote                       8-20                               September 2011

-------
 1                   In general, toxicological studies have reported differences in cardiac and respiratory
 2                   effects after O3 exposure among different mouse strains, which alludes to differential risk
 3                   among individuals due to genetic variability (Tankersley et al.. 2010; Chuang et al.. 2009;
 4                   Hamade and Tankersley. 2009; Hamade et al.. 2008). Thus strains of mice which are
 5                   prone to or resistant to O3-induced effects have been used to systematically identify
 6                   candidate genes that may increase risk of O3-related health effects. Genome wide linkage
 7                   analyses have identified quantitative trait loci for O3-induced lung inflammation and
 8                   hyperpermeability on chromosome 17 (Kleeberger et al.. 1997) and chromosome 4
 9                   (Kleeberger et al., 2000). respectively, using recombinant inbred strains of mice. More
10                   specifically, these studies found that Tnf (protein product is the inflammatory cytokine
11                   TNF-a) and Tlr4 (protein product is TLR4, involved in endotoxin responses) were
12                   candidate susceptibility genes (Kleeberger et al.. 2000; Kleeberger et al.. 1997). The TNF
13                   receptors 1  and 2 have also been found to play a role in injury, inflammation, and airway
14                   hyperreactivity in studies of O3-exposed knockout mice (Cho et al.. 2001). In addition to
15                   Tlr4, other innate immune pattern recognition signaling pathway genes, including Tlr2
16                   and Myd88, appear to be important in responses to O3, as demonstrated by Williams et
17                   al. (2007b). A role for the inflammatory cytokine IL-6 has been demonstrated in
18                   gene-deficient mice with respect to inflammation and injury, but not AHR (Johnston et
19                   al.. 2005b; Yu et al.. 2002). Mice deficient in IL-10, an anti-inflammatory cytokine,
20                   demonstrated increased pulmonary inflammation in response to O3 exposure (Backus et
21                   al.. 2010). Thus genes related to innate immune  signaling and pro- and anti-inflammatory
22                   genes are important for O3-induced responses.

23                   Altered O3 responses between mouse strains could be due to genetic variability in
24                   nuclear factor erythroid 2-related factor 2 (Nrf-2), suggesting a role for genetic
25                   differences in altering the formation of ROS  (Hamade et al., 2010;  Cho and Kleeberger.
26                   2007). Additionally, some studies have reported O3-related effects to vary by Inf-1 and
27                   Inf-2 quantitative trait loci (Tankersley and Kleeberger. 1994) and a gene coding for
28                   Clara cell secretory protein (CCSP) (Broeckaert et al.. 2003; Wattiez et al.. 2003). Other
29                   investigations in inbred mouse strains found that differences in expression of certain
30                   proteins, such as CCSP (Broeckaert et al.. 2003) and MARCO (Dahl et al.. 2007). are
31                   responsible for phenotypic characteristics, such as epithelial permeability and scavenging
32                   of oxidized lipids, respectively, which confer sensitivity to  O3.

33                   Nitric oxide (NO), derived from activated macrophages, is produced upon exposure to O3
34                   and is thought to participate in lung damage.  Mice deficient in the gene for inducible
35                   nitric oxide synthase (NOS2/NOSIMNOS) are partially protected against lung injury
36                   (Kleeberger et al.. 2001). and it appears that O3-induced iNOS expression is tied to the
37                   TLR4 pathway described above. Similarly, iNOS deficient  mice do not produce reactive
38                   nitrogen intermediates after O3 exposure, in contrast to their wild-type counterparts, and
      Draft - Do Not Cite or Quote                       8-21                                 September 2011

-------
 1                   also produce less PGE2 comparatively (Fakhrzadeh et al., 2002). These gene-deficient
 2                   mice were protected from O3-induced lung injury and inflammation. In contrast, another
 3                   study using a similar exposure concentration but longer duration of exposure found that
 4                   iNOS deficient mice were more susceptible to O3-induced lung damage (Kenyon et al..
 5                   2002). Therefore it is unclear whether inducible nitric oxide synthase plays a protective
 6                   role or mediates damage.

 7                   Voynow et al. (2009) have shown that NQO1 deficient mice, like their human
 8                   counterparts, are resistant to O3-induced AHR and inflammation. NQO1 catalyzes the
 9                   reduction of quinones to hydroquinones, and is capable of both protective detoxification
10                   reactions and redox cycling reactions resulting in the generation of reactive oxygen
11                   species. Reduced production of inflammatory mediators and cells and blunted AHR were
12                   observed in NQO1 null mice after exposure to 1 ppm O3 for 3 hours. These results
13                   correlated with those from in vitro experiments in which human bronchial epithelial cells
14                   treated with an NQO1 inhibitor exhibited reduced inflammatory responses to exposure to
15                   0.4 ppm O3 for 5 hours. This study may provide biological plausibility for the increased
16                   biomarkers of oxidative stress and increased pulmonary function decrements observed in
17                   O3-exposed individuals bearing both the wild-type NQO1 gene and the null GSTM1 gene
18                   (Corradi et al.. 2002: Bergamaschi et al.. 2001).

19                   The role of TNF-a signaling in O3-induced responses has been previously established
20                   through depletion experiments, but a more recent toxicological study investigated the
21                   effects of combined O3 and PM exposure in transgenic TNF overexpressing mice.
22                   Kumarathasan et al. (2005) found that subtle effects of these pollutants were difficult to
23                   identify in the midst of the severe pathological changes caused by constitutive TNF-a
24                   overexpression. However, there was evidence that TNF transgenic mice were more
25                   susceptible to O3/PM-induced oxidative stress, and they exhibited elevation of a serum
26                   creatine kinase after pollutant exposure, which may suggest potential systemic or cardiac
27                   related effects. Differential  susceptibility to O3 among inbred strains of animals does  not
28                   seem to be dose dependent since absorption of 18O in various strains of mice did not
29                   correlate with resistance or sensitivity (Vancza et al.. 2009).

30                   Defects in DNA repair mechanisms may also confer increased risk of O3-related health
31                   effects. Cockayne syndrome, a rare autosomal recessive disorder in humans, is
32                   characterized by UV sensitivity abnormalities, neurological abnormalities, and premature
33                   aging. The same genetic defect in mice  (Csb~A) makes them sensitive to oxidative
34                   stressors, including O3. Kooter et al. (2007) demonstrated that Csb"7" mice produced
35                   significantly more TNF-a after exposure to 0.8 ppm O3 than their wild-type counterparts.
36                   However, there were no significant differences in other markers of inflammation or lung
      Draft - Do Not Cite or Quote                       8-22                                 September 2011

-------
 1                   injury between the two strains of mice. Further discussion of candidate genes in the
 2                   context of their respective signaling pathways can be found in Chapter 5.

 3                   Overall, multiple genes, such as GSTM1, GSTP1, HMOX-1, NQO1, and TNF-a, appear
 4                   to potentially be involved in populations being more at-risk than others to the effects of
 5                   O3 exposure on health. Future studies of these and other genes in human populations will
 6                   be important for determining the role of each genotype and  its effect on risk. For NQO1
 7                   and TNF-a, biological plausibility is provided by toxicological studies. Additionally,
 8                   studies of rodents have identified a number of other genes that may affect O 3 -related
 9                   health outcomes, but testing of these genes has not been performed in humans due to
10                   power limitations.
          8.5   Diet

11                   Diet was not examined as a factor affecting risk in previous O3 AQCDs, but recent
12                   studies have examined modification of the association between O3 and health effects by
13                   dietary factors. Because O3 mediates some of its toxic effects through oxidative stress,
14                   the antioxidant status of an individual is an important factor that may contribute to
15                   increased risk of O3-related health effects. Supplementation with vitamin E has been
16                   investigated in a number of studies as a means of inhibiting O3-mediated damage.

17                   Epidemiologic studies have examined effect measure modification by diet and found
18                   evidence that certain dietary components are related to the effect O3  has on respiratory
19                   outcomes. The most recent study examined fruit/vegetable intake and Mediterranean diet
20                   (Romieu et al., 2009). Increases in these food patterns, which have been noted for their
21                   high vitamins C and E and omega-3 fatty acid content, protected against O3-related
22                   decreases in lung function among children living in Mexico City. Another study
23                   examined supplementation of the diets of asthmatic children in Mexico with vitamins C
24                   and E (Sienra-Monge et al., 2004). Associations were detected between short-term O3
25                   exposure and nasal airway inflammation among children in the placebo group but not in
26                   those receiving the supplementation. The authors concluded that "vitamin C and E
27                   supplementation above the minimum dietary requirement in asthmatic children with a
28                   low intake of vitamin E might provide some protection against the nasal acute
29                   inflammatory response to ozone."

30                   The epidemiologic evidence is supported by controlled human exposure studies, which
31                   have shown that the first line of defense against oxidative stress is antioxidants-rich
32                   extracellular lining fluid (ELF) which scavenge free radicals and limit lipid peroxidation.
33                   Exposure to O3 depletes the  antioxidant level in nasal ELF probably due to scrubbing of
34                   O3 (Mudway et al.. 1999a): however, the concentration and the activity of antioxidant

      Draft - Do Not Cite or Quote                       8-23                                 September 2011

-------
 1                   enzymes either in ELF or plasma do not appear to be related to O3 responsiveness (Samet
 2                   etal.. 2001: Avissaretal.. 2000: Blomberg et al.. 1999). Carefully controlled studies of
 3                   dietary antioxidant supplementation have demonstrated some protective effects of
 4                   a-tocopherol (a form of vitamin E) and ascorbate (vitamin C) on spirometric measures of
 5                   lung function after O3 exposure but not on the intensity of subjective symptoms and
 6                   inflammatory response including cell recruitment, activation and a release of mediators
 7                   (Samet etal.. 2001: Trengaet al., 2001). Dietary antioxidants have also afforded partial
 8                   protection to asthmatics by attenuating postexposure bronchial hyperresponsiveness
 9                   (Trengaetal.. 2001).

10                   Toxicological studies provide evidence of biological plausibility to the epidemiologic and
11                   controlled human exposure studies. Wagner et al. (2009: 2007) have shown reductions in
12                   O 3-exacerbated nasal allergy responses in rats with y-tocopherol treatment (a form of
13                   vitamin E). O3-induced inflammation and mucus production were also inhibited by
14                   y-tocopherol. Inconsistent results were observed in toxicological studies of vitamin C
15                   deficiency and O3-induced responses. Guinea pigs deficient in vitamin C displayed only
16                   minimal injury and inflammation after exposure to O3 (Kodavanti et al., 1995). A recent
17                   study in mice demonstrated a protective effect of p-carotene in the skin, where it limited
18                   the production of proinflammatory markers and indicators of oxidative stress induced by
19                   O3 exposure (Valacchi et al., 2009). Deficiency of vitamin A, which has a role in
20                   regulating the maintenance and repair of the epithelial layer, particularly in the lung,
21                   appears to enhance the risk of O3-induced lung injury (Paquette et al.. 1996).
22                   Differentially susceptible strains that were fed a vitamin A sufficient diet were observed
23                   to have different tissue concentrations of the vitamin, potentially contributing to their
24                   respective differences in O3-related outcomes. In addition to the studies of antioxidants,
25                   one toxicological study examined protein deficiency. Protein deficiency alters the levels
26                   of enzymes and chemicals in the brain involved with redox status; exposure to 0.75 ppm
27                   O3 has been shown to differentially affect Na+/K+ ATPase, glutathione, and lipid
28                   peroxidation, depending on the nutritional status of the animal, but the significance of
29                   these changes is unclear (Calderon Guzman et al., 2006).  There may be a protective
30                   effect of overall dietary restriction with respect to lung injury, possibly related to
31                   increased vitamin C in the lung surface fluid (Kari etal..  1997).

32                   Epidemiologic studies find that individuals with diets deficient in vitamins E and C are at
33                   risk for O3  -related health effects. This is supported  by controlled human exposure and
34                   toxicological  studies.
      Draft - Do Not Cite or Quote                       8-24                                 September 2011

-------
          8.6    Body Mass Index and Physical Conditioning

 1                  Obesity, defined as a BMI of 30 kg/m2 or greater, is an issue of increasing importance in
 2                  the U.S., with self-reported rates of 26.7% in 2009, up from 19.8% in 2000 (Sherry et al..
 3                  2010). A few studies have been performed examining the association between BMI and
 4                  O 3-related changes in lung function. An epidemiologic study reported decreased lung
 5                  function with increased short-term O3 exposure for both obese and non-obese subjects;
 6                  however, the magnitude of the reduction in lung function was greater for those subjects
 7                  who were obese (Alexeeff et al., 2007). Further decrements in lung function were noted
 8                  for obese individuals with AHR. Controlled human exposure studies have also detected
 9                  differential effects of O3 exposure on lung function for individuals with varying BMIs. In
10                  a retrospective analysis of data from 541 healthy, nonsmoking, white males between the
11                  ages of 18-35 years from 15 studies conducted at the U.S. EPA Human Studies Facility in
12                  Chapel Hill, North Carolina, McDonnell et al. (2010) found that increased body mass
13                  index (BMI) was found to be associated with enhanced FEVi responses. The BMI effect
14                  was of the same order of magnitude but in the opposite direction of the age effect
15                  whereby FEVi responses diminish with increasing age. In a similar analysis, Bennett et
16                  al. (2007) found enhanced FEVi decrements following O3 exposure with increasing BMI
17                  in a group of healthy, nonsmoking, women (BMI range 15.7 to 33.4), but not among
18                  healthy, nonsmoking men (BMI range  19.1 to 32.9). In the  women, greater O3-induced
19                  FEVi decrements were seen in individuals that were overweight/obese (BMI >25)
20                  compared normal weight (BMI from 18.5 to 25), and in normal weight compared to
21                  underweight (BMI <18.5). Even disregarding the five underweight women, a greater O3
22                  response in the overweight/obese category (BMI >25) was  observed compared with the
23                  normal weight group (BMI from 18.5 to 24.9).

24                  Studies in genetically and dietarily obese mice have shown enhanced pulmonary
25                  inflammation and injury with acute O3exposure, but responses to longer exposures at a
26                  lower concentration appear to differ. A recent study found that obese mice are actually
27                  resistant to O3-induced pulmonary injury and inflammation and reduced lung compliance
28                  following exposure to 0.3 ppm O3 for 72 hours, regardless of whether obesity was
29                  genetic- or diet-induced (Shore et al.. 2009).

30                  In addition to studies of obesity, physical conditioning affects BMI and may also affect
31                  the risk of O3-related health effects. The 2008 Summary of Health Statistics for U.S.
32                  Adults from the CDC reported the prevalence of regular leisure-time physical activity as
33                  slightly above 30% for adults >18 years of age in the U.S. (Pleis et al.. 2009). Forty-nine
34                  percent of individuals >65 years old reported no leisure-time physical activity. A study of
3 5                  effect measure modification by exercise habits ten years prior to death observed excess
36                  risk of mortality with increasing O3 concentrations among individuals that never
      Draft - Do Not Cite or Quote                       8-25                                September 2011

-------
 1                   exercised compared to individuals that exercised at least once a month for both adults
 2                   >30 years of age and adults >65 years of age (Wong et al.. 2007). No recent studies
 3                   examining modification of O3 -related health effects by current physical activity were
 4                   identified.

 5                   Multiple epidemiologic and human clinical studies have reported increased O3-related
 6                   respiratory health effects among obese individuals. Future research of the effect
 7                   modification of the relationship between O3 and other health-related outcomes besides
 8                   respiratory health effects by BMI and studies examining the role of physical conditioning
 9                   will advance understanding of obesity as a factor potentially increasing an individual's
10                   risk.
          8.7   Socioeconomic Status

11                   SES is often represented by personal or neighborhood SES, educational attainment,
12                   health insurance status, and other such factors. SES is indicative of such things as access
13                   to healthcare, quality of housing, and pollution gradient. Based on the 2000 Census data,
14                   12.4% of Americans live in poverty (poverty threshold for family of four was $17,463)
15                   (SSDAN CensusScope. 2010c).

16                   Multiple epidemiologic studies have reported individuals of low SES to have increased
17                   risk for the effects of short-term O3 exposure on respiratory HAs and ED visits. A study
18                   performed in Korea examined the association between O3 concentrations and asthma HA
19                   and reported larger effect estimates in areas of moderate and low SES compared with
20                   areas of high SES (SES was based on average regional insurance rates) (Lee et al.. 2006).
21                   A Canadian study reported inverse effects of O3 on respiratory HA and ED visits
22                   regardless of SES, measured by average census tract household income (Burra et al..
23                   2009). In addition, a study conducted across 10 cities in Canada found the largest
24                   association between O3 exposure and respiratory HA was among those with an
25                   educational level less than grade 9, but no consistent trend in the effect was seen across
26                   quartiles of income (Cakmak et al.. 2006b). In New York State, larger associations
27                   between long-term O3 exposure and asthma HA were observed among children of
28                   mothers who did not graduate from high school, whose births were covered by
29                   Medicaid/self-paid, or who were living in poor neighborhoods compared to children
30                   whose mothers graduated from high school, whose births were covered by other
31                   insurance, or who were not living in poor neighborhoods, respectively (Lin et al.. 2008b).

32                   The examination of the potential effects of SES on O3-related  cardiovascular health
33                   effects is relatively limited. A study conducted in Canada reported the association
34                   between short-term O3 and ED visits for cardiac disease by quartiles of

      Draft - Do Not Cite or Quote                       8-26                                 September 2011

-------
 1                   neighborhood-level education and income. No effect measure modification was apparent
 2                   for either measure of SES (Cakmak et al.. 2006a).

 3                   Several studies were conducted that examined the modification of the relationship
 4                   between short-term O3 concentrations and mortality by SES. A U.S. multicity study
 5                   reported that communities with a higher proportion of the population unemployed had
 6                   higher mortality effect estimates (Bell and Dominici. 2008). A study in seven urban
 7                   centers in Chile reported on modification of the association between O3 exposure and
 8                   mortality using multiple SES markers (Cakmak et al.. 2011). Increased risk was observed
 9                   among the categories of low SES for  all measures (personal educational attainment,
10                   personal occupation, community income level). Additionally, the APHENA study, which
11                   examined the association between O3 and mortality by percentage unemployed, reported
12                   a higher percent change in mortality with increased percent unemployed but this varied
13                   across the regions included in the study (U.S., Canada, Europe) (Katsouyanni et al..
14                   2009). A Chinese study reported that  the greatest effects between O3 concentrations and
15                   mortality at lag day 0 were among individuals living  in areas of high social deprivation
16                   (i.e. low SES), but this association was not consistent across lag days (at other lag times,
17                   the middle social deprivation index category had the  greatest association) (Wong et al..
18                   2008). However, another study in Asia comparing low to high educational attainment
19                   populations reported no evidence of greater mortality effects (total,  CVD, or respiratory)
20                   (Kan et al.. 2008). Additionally, a study in Italy reported no difference in risk of mortality
21                   among census-block level derived income levels (Stafoggia et al.. 2010). A study of
22                   infant mortality in Mexico reported no association between O3 concentrations and infant
23                   mortality among any of the three levels of SES determined using  a socioeconomic index
24                   based on residential areas (Romieu et al.. 2004b). Another study in Mexico reported a
25                   positive  association between O3 levels at lag 0 and respiratory-related infant mortality in
26                   only the low SES group (determined based on education, income, and household
27                   conditions across residential areas), but no association was observed in any of the  SES
28                   groups with other lags  (Carbajal-Arroyo et al.. 2011).

29                   Studies of O3 concentrations and reproductive outcomes have also examined associations
30                   by SES levels. A study in California reported greater decreases in birth weight associated
31                   with full pregnancy O3 concentration for those with neighborhood poverty levels of at
32                   least 7% compared with those in neighborhoods with less than 7% poverty (Morello-
33                   Frosch et al.. 2010). However, no dose response was  apparent and those with
34                   neighborhood poverty levels of 7-21% had greater decreases observed for the association
3 5                   than those living in areas with poverty rates of at least 22%. An Australian study reported
36                   an inverse association between O3 exposure during days 31-60 of gestation and
37                   abdominal circumference during gestation (Hansen et al.. 2008). The interaction with
3 8                   SES (area-level measured socioeconomic disadvantage) was examined and although the
      Draft - Do Not Cite or Quote                       8-27                                 September 2011

-------
 1                   inverse association remained statistically significant in only the highest SES quartile,
 2                   there were large confidence interval overlaps among estimates for each quartile so no
 3                   difference in the association for the quartiles was apparent.

 4                   Evidence from a controlled human exposure study that examined O3 effects on lung
 5                   function does not provide support for greater O3-related health effects in individuals of
 6                   lower SES. In a follow-up study (Seal et al., 1993) on modification by race, Seal et al.
 7                   (1996) reported that, of three SES categories, individuals in the middle SES category
 8                   showed greater concentration-dependent decline in percent-predicted FEVi (4-5% at
 9                   400 ppb O3) than in low and high SES groups. The authors did not have an "immediately
10                   clear" explanation for this finding and controlled human exposure studies are typically
11                   not designed to answer questions about SES.

12                   Overall, most studies of individuals have reported that individuals with low SES and
13                   those living in neighborhoods with low SES are more at risk for O3-related health effects
14                   resulting in higher odds of respiratory HAs and ED visits. Inconsistent results have been
15                   observed in the few studies examining effect modification of associations between O3
16                   exposure and mortality and reproductive outcomes.
          8.8    Race/Ethnicity

17                   Based on the 2000 Census, 69.1% of the U.S. population comprises non-Hispanic whites.
18                   Approximately 12.1% of people reported their race/ethnicity as non-Hispanic black and
19                   12.6% reported being Hispanic (SSDAN CensusScope. 201 Ob).
20                   Two studies examined the associations between short-term O3 concentrations and
21                   mortality and reported higher effect estimates among blacks (Medina-Ramon and
22                   Schwartz. 2008) and among communities with larger proportions of blacks (Bell and
23                   Dominici. 2008). Another study examined long-term exposure to O3 concentrations and
24                   asthma HAs among children in New York State. These authors reported no statistically
25                   significant difference in the odds of asthma HA for blacks compared to other races but
26                   did detect higher odds for Hispanics  compared to non-Hispanics (Lin et al.. 2008b).
27                   Additionally, recent epidemiologic studies have stratified by race when examining the
28                   association between O3 concentration and birth outcomes. A study conducted in Atlanta,
29                   GA reported decreases in birth weight with increased third trimester O3 concentrations
30                   among Hispanics but not among non-Hispanic whites (Darrow et al., 201 la). An inverse
31                   association was also present for non-Hispanic blacks but was not statistically significant.
32                   A California study reported that the greatest decrease in birth weight associated with full
33                   pregnancy O3 concentration was among non-Hispanic whites (Morello-Frosch et al..
      Draft - Do Not Cite or Quote                       8-28                                September 2011

-------
 1                   2010). The inverse association was also apparent, although not as strong, for
 2                   non-Hispanic blacks. Increased birth weight was associated with higher O3 exposure
 3                   among Hispanics and among non-Hispanic Asians and Pacific Islanders but neither of
 4                   these results were statistically significant.

 5                   Similar to the epidemiologic studies, a controlled human exposure study suggested
 6                   differences in lung function responses by race (Seal et al., 1993). The independent effects
 7                   of sex-race group and O3 concentration on lung function were positive, but the
 8                   interaction between sex-race group and O3 concentration was not statistically significant.
 9                   The findings indicated some overall difference between the sex-race groups that was
10                   independent of O3 concentration (the concentration-response curves for the four sex-race
11                   groups are parallel). In a multiple comparison procedure on data collapsed across all O3
12                   concentrations for each sex-race group, both black men and black women had larger
13                   decrements in FEVi than did white men. The authors noted that the O3 dose per unit of
14                   lung tissue would be greater in blacks and females than whites and males, respectively.
15                   That this difference in tissue dose might have affected responses to O3 cannot be ruled
16                   out. The college students recruited for the Seal et al. (1993) study were probably from
17                   belter educated and more SES advantaged families, thus reducing potential for these
18                   variables to be confounding factors. Que et al. also examined pulmonary responses to O3
19                   exposure in blacks of African American ancestry and in whites. On average, the black
20                   males experienced the greatest decrements in FEVi following O3 exposure. This
21                   decrease was larger than the decrement observed among black females, white males, and
22                   white females.

23                   Overall, the results of recent studies suggest that there may be race-related increase  in
24                   risk of O3-related health effects for some outcomes, although the overall understanding of
25                   potential effect measure modification by race is limited by the small number of studies.
26                   Additionally, these results may be confounded by other factors, such as SES.
          8.9   Smoking

27                   Previous O3 AQCDs have concluded that smoking does not increase the risk of
28                   O3-related health effects; in fact, in controlled human exposure studies, smokers have
29                   been found to be at less risk of O3-related health effects than non-smokers. Data from
30                   recent interviews conducted as part of the 2008 National Health Interview Survey (NHIS)
31                   (Pleis et al.. 2009) have shown the rate of smoking among adults >18 years old to be
32                   approximately 20% in the U.S. Approximately 21% of individuals surveyed were
33                   identified as former smokers.
      Draft - Do Not Cite or Quote                       8-29                                 September 2011

-------
 1                  Baccarelli et al. (2007) performed a study of O3 concentrations and plasma homocysteine
 2                  levels (a risk factor for vascular disease). They found no interaction of smoking (smokers
 3                  versus non-smokers) for the associations between O3 concentrations and plasma
 4                  homocysteine levels. Another study examined the association between O3 and resting
 5                  heart rate and also reported no interaction with smoking status (current smokers versus
 6                  current non-smokers) (Ruidavets et al.. 2005a).

 7                  A study examining correlations between O3 levels and oxidative DNA damage examined
 8                  results stratified by current versus never and former smokers (Palli et al.. 2009). Ozone
 9                  was positively associated with DNA damage for short-term and long-term exposures
10                  among never/former smokers. For current smokers, short-term O3 concentrations were
11                  inversely associated with DNA damage; however, the number of current smokers in the
12                  study was small (n= 12).

13                  The findings of Palli et al. (2009) were consistent with those from controlled human
14                  exposure studies that have confirmed that smokers are less responsive to O3 exposure
15                  than non-smokers. Spirometric and plethysmographic pulmonary function decline,
16                  nonspecific AHR, and inflammatory responses of smokers to O3 exposure were all
17                  weaker than those reported for non-smokers. Similarly, the time course of development
18                  and recovery from these effects, as well as their reproducibility, was not different from
19                  non-smokers. Chronic airway inflammation with desensitization of bronchial nerve
20                  endings and an increased production of mucus may plausibly explain the
21                  pseudo-protective effect of smoking (Frampton et al.. 1997b: Torres et al.. 1997).

22                  These findings for smoking are consistent with previous AQCD conclusions. An
23                  epidemiologic study of O3-associated DNA damage reported smokers to be less at risk
24                  for O3-related health effects. However, both epidemiologic studies of short-term
25                  exposure and CVD outcomes found no effect measure modification by smoking.
          8.10  Heightened Exposure

26                  Studies included in the 2006 O3 AQCD reported that individuals who participate in
27                  outdoor activities or work to be a population at increased risk based on consistently
28                  reported associations between O3 exposure and respiratory health outcomes in these
29                  groups (U.S. EPA. 2006b). Outdoor workers are  exposed to ambient O3 concentrations
30                  outside for a greater period of time than individuals who spend their days indoors.
31                  Additionally, an increase in dose to the lower airways is possible during exercise due to
32                  both increases in the amount of air breathed (i.e., minute ventilation) and a shift from
33                  nasal to oronasal breathing (Sawyer et al., 2007; Nodelman and Ultman,  1999; Hu et al.,
34                  1994). For further discussion of the association between FEVi responses to O3 exposure

      Draft - Do Not Cite or Quote                       8-30                               September 2011

-------
 1                   and minute ventilation, refer to Section 6.2.3.1 of the 2006 O3 AQCD. A recent study has
 2                   explored the potential effect measure modification of O3 exposure and DNA damage by
 3                   indoor/outdoor workplace (Tovalin et al., 2006). In a study of indoor and outdoor
 4                   workers in Mexico, individuals who worked outdoors in Mexico City had a slight
 5                   association between O3 exposure and DNA damage (measured by comet tail length
 6                   assay), whereas no association was observed for indoor workers in Mexico City. Workers
 7                   in another Mexican city, Puebla, demonstrated no association between O3 levels and
 8                   DNA damage, regardless of whether they worked indoors or outdoors.

 9                   Air conditioning use is an important component of O3 exposure, as use of central air
10                   conditioning will limit exposure to O3 by blocking the penetration of O3 into the indoor
11                   environment (further information can be found in Section 4.4 of this ISA). Air
12                   conditioning use is a difficult effect measure modifier to examine in epidemiologic
13                   studies. Air conditioning use is often measured based on regional prevalence and may not
14                   reflect individua!4evel use. More generally, air conditioning prevalence is associated
15                   with temperature of a region; those areas with higher temperatures have a greater
16                   prevalence of households with air conditioning. Despite these limitations, a few studies
17                   have examined effect measure modification by prevalence of air conditioning use in an
18                   area. Studies examining multiple cities across the U.S. have assessed whether
19                   associations between O3 concentrations and HA and mortality varied among areas with
20                   high and low prevalence of air conditioning. Medina-Ramon et al. (2006) conducted a
21                   study during the warm season and observed a greater association between O3 levels  and
22                   pneumonia HA among areas with a lower proportion of households having central air
23                   conditioning compared to areas with a larger proportion of households with air
24                   conditioning. The same trend of increased association for areas with a lower prevalence
25                   of central air conditioning was noted in a study of O3 concentrations and mortality (Bell
26                   and Dominici. 2008). Conversely, Medina-Ramon and Schwartz (2008) found that
27                   among individuals with atrial fibrillation, a lower risk of mortality was observed for areas
28                   with a lower prevalence of central air conditioning.

29                   Previous work has shown that increased dose of O3 concentrations from outdoor work
30                   leads to increased risk of O3-related health effects among individuals who participate in
31                   outdoor activities or work, although there is no evidence of modification by outdoor
32                   activity in this recent study. Lower prevalence of air conditioning also appears to affect
33                   risk of O3-related health effects, but this is not true of all studies. Overall, increased
34                   exposure to outdoor air does appear to confer additional risk and individuals with greater
35                   exposure to outdoor air may experience more O3-related health effects.
      Draft - Do Not Cite or Quote                       8-31                                 September 2011

-------
          8.11  Healthy Responders

 1                  Within the general population, there is evidence for variability in responses to O3
 2                  exposure, with some healthy individuals demonstrating greater O3-related health effects
 3                  compared to other healthy individuals in controlled human exposure studies. These
 4                  individuals do not fit in any of the at-risk populations discussed in this chapter; however,
 5                  studies have found that they have greater responses to O3 exposure than would be
 6                  expected, indicating a unique population that needs to be considered.

 7                  Controlled human exposure studies have demonstrated a large degree  of intersubject
 8                  variability in lung function decrements, symptomatic responses, pulmonary
 9                  inflammation, AHR, and altered epithelial permeability in healthy adults exposed to O3
10                  (Que et al.; Holz et al., 2005; McDonnell. 1996). The magnitude of increases in
11                  pulmonary inflammation, AHR, and epithelial permeability, in response to O3 exposure,
12                  do not appear to be correlated, nor are these responses correlated with changes in lung
13                  function (Que et al.: Balmesetal.. 1997; Balmesetal.. 1996; Arisetal.. 1995). However,
14                  these responses to O3 exposure  in healthy individuals tend to be reproducible within a
15                  given individual over a period of several months indicating differences in the intrinsic
16                  responsiveness of individuals (Holz et al.. 2005; Hazucha et al.. 2003; Holz et al.. 1999;
17                  McDonnell et al.. 1985a). It should be noted that even when group mean responses are
18                  small and seem physiologically insignificant, some intrinsically more  responsive
19                  individuals experience distinctly larger effects under the same exposure conditions. For
20                  example, small group mean changes (e.g., <5%) in FEVi have been observed in healthy
21                  young adults at levels as low as 120 ppb O3 for 1 to 3 hour exposure periods. However,
22                  some individuals within a study may experience FEVi decrements in excess of 15%
23                  under these conditions, even with group mean decrements of less than 5%. Therefore,
24                  within the general population, a proportion of otherwise healthy individuals, who do not
25                  have characteristics  discussed above that increase risk, may be at increased risk of
26                  O 3 -induced health effects.
          8.12  Summary

27                  In this section, epidemiologic, controlled human exposure, and toxicological studies have
28                  been evaluated that contribute information on potential at-risk populations. Overall, this
29                  review provides evidence that various factors may lead to increased risk of O3-related
30                  health effects.
31                  The populations identified in this section that are most at risk for O3-related health effects
32                  are individuals with influenza/infection, individuals with asthma, and younger and older
      Draft - Do Not Cite or Quote                      8-32                                September 2011

-------
 1                   age groups. There were a small number of studies on influenza/infection but both
 2                   reported influenza/infection to modify the association between O3 exposure and
 3                   respiratory effects, with individuals having influenza or an infection being at increased
 4                   risk. Asthma as a factor affecting risk was supported by controlled human exposure and
 5                   toxicological studies, as well as some evidence from epidemiologic studies. Most studies
 6                   comparing age groups reported greater effects of short-term O3 exposure on mortality
 7                   among older adults, although studies of other health outcomes had inconsistent findings
 8                   regarding whether older adults were at increased risk. Generally, studies of age groups
 9                   also reported positive associations for respiratory HAs and ED visits among children.
10                   Biological plausibility for this increased risk is supported by toxicological and clinical
11                   research. Diet and obesity are also both likely factors affecting risk. Multiple
12                   epidemiologic, controlled human exposure, and toxicological studies reported that diets
13                   deficient in vitamins E and C are associated with risk of O3-related health effects.
14                   Similarly, studies of effect measure modification by BMI observed greater O3-related
15                   respiratory decrements for individuals who were obese.

16                   Other potential factors [preexisting conditions (such as COPD and CVD), sex, and
17                   multiple genes (such as GSJM1, GSTP1, HMOX-1, NQO1, and TNF-a)} provided some
18                   evidence of increased risk, but further evidence is needed. In addition, examination of
19                   modification of the associations between O3 exposure and health effects by SES and race
20                   were available in a limited number of studies, and demonstrated possible increased odds
21                   of health effects related to O3  exposure among those with low SES and  black race.

22                   Individuals with increased outdoor exposure were examined in a recent  study of outdoor
23                   workers, in which no effect modification was observed, and studies of air conditioning
24                   prevalence, which demonstrated inconsistent  findings. However, previous evidence along
25                   with biological plausibility from toxicological and controlled human studies has shown
26                   individuals exposed to more outdoor air to be at increased risk of O3-related health
27                   effects. Studies  of physical conditioning and smoking were conducted but little evidence
28                   was available to determine whether increased risk of O3-related health effects is present
29                   for these factors. The only studies examining  effect measure modification by diabetes
30                   examined O3 exposure and cardiovascular outcomes and none reported increased risks for
31                   individuals with diabetes. Toxicological studies also identified hyperthyroidism to be a
32                   factor warranting further examination. Future research will provide additional insight into
33                   whether these factors affect risk of O3-related health effects.
      Draft - Do Not Cite or Quote                       8-33                                 September 2011

-------
    8.13  References

Alexeeff, SE: Litonjua, AA: Suh, H: Sparrow, D: Vokonas, PS: Schwartz, J. (2007). Ozone exposure and lung
       function: Effect modified by obesity and airways hyperresponsiveness in the VA Normative Aging Study.
       Chest 132: 1890-1897. http://dx.doi.org/10.1378/chest.07-1126.
Alexeeff, SE: Litonjua, AA: Wright, RO: Baccarelli, A: Suh, H: Sparrow, D: Vokonas, PS: Schwartz, J. (2008).
       Ozone exposure, antioxidant genes, and lung function in an elderly cohort: VA Normative Aging Study.
       Occup Environ Med 65: 736-742. http://dx.doi.org/10.1136/oem.2007.035253.
Alexis, NE: Zhou, H: Lay, JC: Harris, B: Hernandez, ML: Lu, TS: Bromberg, PA: Diaz-Sanchez, D: Devlin,  RB:
       Kleeberger. SR: Peden. DB. (2009). The glutathione-S-transferase Mu 1 null genotype modulates
       ozone-induced airway inflammation in human subjects. J Allergy Clin Immunol 124: 1222-1228.
       http://dx.doi.0rg/10.1016/i.iaci.2009.07.036.
Arbex. AM: de  Souza Conceicao. GM: Perez Cendon. S: Arbex. FF: Lopes. AC: Providello Movses. E: Santiago.
       SL: Saldiva. PHN: Pereira. LAA: Ferreira Braga. AL. (2009). Urban air pollution and COPD-related
       emergency room visits. J Epidemiol Community Health 966: 777-783.
       http://dx.doi.org/10.1136/iech.2008.078360.
Aris. RM: Tager. I: Christian.  D: Kelly. T: Balmes. JR. (1995). Methacholine responsiveness is not associated
       with OS-induced decreases in FEV1. Chest 107: 621-628.
Auten. RL: Potts. EN: Mason. SN: Fischer.  B: Huang.  Y: Foster. WM. (2009). Maternal exposure  to particulate
       matter increases postnatal ozone-induced airway hyperreactivity in juvenile mice. Am J Respir Crit
       Care Med 180: 1218-1226. http://dx.doi.org/10.1164/rccm.200901-0116OC.
Avissar. NE:  Reed.  CK: Cox. C: Frampton.  MW: Finkelstein. JN. (2000). Ozone, but not nitrogen  dioxide,
       exposure decreases glutathione peroxidases in epithelial lining fluid of human lung. Am J Respir Crit
       Care Med 162: 1342-1347.
Baccarelli. A: Zanobetti. A: Martinelli. I: Grillo. P: Hou.  L: Lanzani. G: Mannucci. PM: Bertazzi. PA: Schwartz. J.
       (2007). Air pollution, smoking, and plasma homocysteine. Environ Health Perspect 115:  176-181.
Backus, GS: Howden,  R: Fostel, J: Bauer, AK: Cho, HY: Marzec, J: Peden, DB: Kleeberger, SR.  (2010).
       Protective role of interleukin-10 in ozone-induced pulmonary inflammation. Environ Health  Perspect
       118: 1721-1727. http://dx.doi.org/10.1289/ehp.1002182.
Balmes, JR: Chen,  l_l_: Scannell, C: Tager, I: Christian, D: Hearne, PQ: Kelly, T: Aris, RM. (1996). Ozone-
       induced decrements in FEV1 and FVC do not correlate with measures of inflammation. Am J Respir
       Crit Care Med 153: 904-909.
Balmes, JR: Aris, RM: Chen, l_l_: Scannell, C: Tager, IB:  Finkbeiner, W: Christian, D: Kelly, T: Hearne, PQ:
       Ferrando. R: Welch. B. (1997). Effects of ozone on normal and potentially sensitive human subjects.
       Part I: Airway inflammation and responsiveness to ozone in normal and asthmatic subjects. Boston,
       MA: Health Effects Institute.
Barraza-Villarreal. A: Sunver. J: Hernandez-Cadena. L: Escamilla-Nunez. MC: Sienra-Monge. JJ: Ramirez-
       Aguilar. M: Cortez-Lugo. M: Holguin. F: Diaz-Sanchez.  D: Olin. AC: Romieu. I. (2008). Air pollution,
       airway inflammation, and lung function in a cohort study of Mexico City schoolchildren. Environ Health
       Perspect 116:  832-838. http://dx.doi.orq/10.1289/ehp. 10926.
Basha. MA: Gross.  KB: Gwizdala. CJ: Haidar. AH: Popovich. J. Jr. (1994). Bronchoalveolar lavage neutrophilia
       in asthmatic and healthy volunteers after controlled exposure to ozone and filtered purified air. Chest
       106: 1757-1765.
Bell. ML: Dominici. F. (2008). Effect modification by community characteristics on the short-term effects of
       ozone exposure and mortality in 98 US communities. Am J Epidemiol 167: 986-997.
       http://dx.doi. orq/10.1093/aie/kwm396.
Bennett. WD: Hazucha. MJ:  Folinsbee. LJ:  Bromberg. PA: Kissling. GE: London. SJ. (2007). Acute  pulmonary
       function response to ozone in young adults as a function of body mass index. Inhal Toxicol 19: 1147-
       1154.  http://dx.doi.org/10.1080/08958370701665475.
Bergamaschi. E: De Palma. G: Mozzoni.  P: Vanni. S: Vettori. MV: Broeckaert. F: Bernard. A: Mutti. A. (2001).
       Polymorphism of quinone-metabolizing enzymes and susceptibility to ozone-induced acute effects. Am
       J Respir Crit Care Med 163: 1426-1431.
Draft - Do Not Cite or Quote                        8-34                                  September 2011

-------
Berhane. K: Zhang. Y: Linn. WS: Rappaport. EB: Bastain. TM: Salam. MT: Islam. T: Lurmann. F: Gilliland. FD.
        (2011). The effect of ambient air pollution on exhaled nitric oxide in the Children's Health Study. Eur
        Respir J 37: 1029-1036. http://dx.doi.org/10.1183/09031936.00081410.
Blomberg. A: Mudwav. IS: Nordenhall.  C: Hedenstrom. H: Kelly. FJ:  Frew. AJ: Holgate. ST: Sandstrom. T.
        (1999). Ozone-induced lung function decrements do not correlate with early airway inflammatory or
        antioxidant responses. Eur Respir J 13: 1418-1428.
Bloom. B:  Cohen. RA: Freeman. G. (2008). Summary health  statistics for U.S. children: National Health
        Interview Survey, 2008. Washington, DC: National Center for Health Statistics.
Bosson. J: Stenfors. N: Bucht. A: Helledav. R: Pourazar. J: Holgate. ST: Kelly. FJ: Sandstrom. T: Wilson. S:
        Frew. AJ: Blomberg. A. (2003). Ozone-induced bronchial epithelial cytokine expression differs between
        healthy and asthmatic subjects. Clin Exp Allergy 33:  777-782.
Broeckaert. F:  Clippe. A: Wattiez. R: Falmagne. P: Bernard. A. (2003). Lung hyperpermeability, Clara-cell
        secretory potein (CC16), and susceptibility to ozone  of five  inbred strains of mice. Inhal Toxicol 15:
        1209-1230.
Buadong.  D: Jinsart. W: Funatagawa. I: Karita. K: Yano. E. (2009). Association between PM10 and O3 levels
        and hospital visits for cardiovascular diseases in Bangkok, Thailand. J Epidemiol 19: 182-188.
        http://dx.doi.org/10.2188/iea.JE20080047.
Burra. TA: Moineddin. R: Agha. MM: Glazier. RH. (2009). Social disadvantage, air pollution, and asthma
        physician visits in Toronto, Canada. Environ Res 109: 567-574.
        http://dx.doi.0rg/10.1016/i.envres.2009.03.004.
Bush. ML: Asplund. PT: Miles. KA:  Ben-Jebria. A: Ultman. JS. (1996). Longitudinal distribution of O3 absorption
        in the  lung: gender differences and intersubject variability. J Appl Physiol 81: 1651-1657.
Cakmak, S: Dales, RE: Judek, S. (2006a).  Do gender, education,  and income modify the effect of air pollution
        gases on cardiac disease? J Occup Environ Med 48: 89-94.
        http://dx.doi.org/10.1097/01.iom.0000184878.11956.4b.
Cakmak, S: Dales, RE: Judek, S. (2006b).  Respiratory health effects of air pollution gases: Modification by
        education and income. Arch Environ Occup Health 61: 5-10.
Cakmak. S: Dales. RE: Vidal. CB. (2007). Air pollution and mortality in Chile: Susceptibility among the elderly.
        Environ Health Perspect 115: 524-527.
Cakmak. S: Dales. RE: Angelica Rubio. M: Blanco Vidal. C. (2011). The risk of dying on days of higher air
        pollution among the socially disadvantaged elderly. Environ Res 111: 388-393.
        http://dx.doi.0rg/10.1016/i.envres.2011.01.003.
Calderon Guzman.  D: Barragan Meiia. G: Hernandez Garcia. E: Juarez Olguin. H. (2006). Effect of nutritional
        status and  ozone exposure on some biomarkers of oxidative stress in rat brain regions. Nutr Cancer
        55: 195-200.  http://dx.doi.org/10.1207/s15327914nc5502  11.
Carbaial-Arrovo. L:  Miranda-Soberanis. V:  Medina-Ramon. M: Roias-Bracho. L: Tzintzun. G: Solfs-Gutierrez. P:
        Mendez-Ramfrez. I: Hurtado-Dfaz. M: Schwartz. J: Romieu. I. (2011). Effect of PM10 and O3 on infant
        mortality among residents in the Mexico City Metropolitan Area: A case-crossover analysis, 1997-2005.
        J  Epidemiol Community Health 65: 715-721. http://dx.doi.org/10.1136/iech.2009.101212.
Carey. SA: Minard.  KR: Trease. LL: Wagner. JG: Garcia. GJ:  Ballinger. CA: Kimbell. JS: Plopper. CG: Corlev.
        RA: Postlethwait, EM: Harkema, JR: Einstein, DR. (2007). Three-dimensional mapping of ozone-
        induced injury in the nasal airways of monkeys using magnetic resonance imaging and morphometric
        techniques. Toxicol Pathol 35: 27-40. http://dx.doi.org/10.1080/01926230601072343.
Chiu, HF: Yang, CY. (2009). Air pollution and emergency room visits for arrhythmias: Are  there potentially
        sensitive groups? J Toxicol Environ Health A 72: 817-823.
        http://dx.doi. org/10.1080/15287390902800405.
Cho, H,  -Y: Zhang,  L,  -Y: Kleeberger, SR. (2001). Ozone-induced lung inflammation and  hyperreactivity are
        mediated via tumor necrosis factor-"alpha" receptors. Am J Physiol 280: L537-L546.
Cho. HY: Kleeberger.  SR. (2007). Genetic  mechanisms of susceptibility to oxidative lung  injury in mice. Free
        Radic  Biol  Med 42: 433-445. http://dx.doi.org/10.1016/i.freeradbiomed.20Q6.11.021.
Chuang. GC: Yang. Z: Westbrook. DG:  Pompilius. M: Ballinger. CA: White. RC: Krzvwanski.  DM: Postlethwait.
        EM: Ballinger. SW. (2009). Pulmonary ozone exposure induces vascular dysfunction, mitochondrial
        damage, and atherogenesis. Am J Physiol Lung Cell Mol Physiol 297: L209-L216.
        http://dx.doi. org/10.1152/ajplung.00102.2009.
Draft - Do Not Cite or Quote                        8-35                                  September 2011

-------
Coffin. PL: Blommer. EJ: Gardner. DE: Holzman. R. (1967). Effect of air pollution on alteration of susceptibility
        to pulmonary infection. Cincinnati, OH: U.S. Department of Health, Education, and Welfare.
Coffin. PL: Gardner. DE. (1972). Interaction of biological agents and chemical air pollutants. Ann Occup Hyg 15:
        219-234.
Corradi, M: Alinovi, R: Goldoni, M: Vettori, M: Folesani, G: Mozzoni, P: Cavazzini, S: Bergamaschi,  E: Rossi, l_:
        Mutti. A. (2002). Biomarkers of oxidative stress after controlled human exposure to ozone. Toxicol Lett
        134: 219-225.
Dahl, M: Bauer, AK: Arredouani, M: Soininen, R: Tryggvason, K: Kleeberger, SR: Kobzik, L. (2007).  Protection
        against inhaled oxidants through scavenging of oxidized lipids by macrophage receptors MARCO and
        SR-AI/II. J Clin Invest 117: 757-764. http://dx.doi.org/10.1172/JCI29968.
Darrow,  LA: Klein, M: Strickland, MJ: Mulholland, JA: Tolbert, PE. (2011a). Ambient air pollution and birth weight
        in full-term infants in Atlanta, 1994-2004. Environ Health Perspect 119: 731-737.
        http://dx.doi. orq/10.1289/ehp. 1002785.
Escamilla-Nunez,  MC:  Barraza-Villarreal, A: Hernandez-Cadena, L: Moreno-Macias, H: Ramirez-Aguilar, M:
        Sienra-Monge. JJ: Cortez-Lugo. M: Texcalac. JL: del Rio-Navarro. B: Romieu. I. (2008). Traffic-related
        air pollution and respiratory symptoms among asthmatic children, resident in Mexico City: The EVA
        cohort study. Respir Res 9: 74. http://dx.doi.org/10.1186/1465-9921-9-74.
Fakhrzadeh. L: Laskin. JD: Laskin. PL. (2002). Deficiency in inducible nitric oxide synthase protects mice from
        ozone-induced lung inflammation and tissue injury. Am J Respir Cell Mol  Biol 26: 413-419.
Fanucchi, MV:  Plopper, CG: Evans, MJ:  Hyde, DM: Van Winkle, LS: Gershwin, LJ: Schelegle, ES. (2006).
        Cyclic exposure to ozone alters distal airway development in infant rhesus monkeys. Am J  Physiol
        Lung Cell Mol Physiol 291:  L644-L650.  http://dx.doi.org/10.1152/aiplung.00027.2006.
Fortino, V: Maioli,  E: Torricelli,  C: Davis, P: Valacchi, G. (2007). Cutaneous MMPs are differently modulated by
        environmental stressors in old and young mice. Toxicol Lett 173: 73-79.
        http://dx.doi.0rg/10.1016/i.toxlet.2007.06.004.
Fox, SD: Adams, WC: Brookes, KA: Lasley, BL.  (1993). Enhanced response to ozone exposure during the
        follicular phase of the menstrual cycle. Environ Health Perspect 101: 242-244.
Frampton. MW: Morrow. PE: Torres. A: Cox. C: Voter. KZ: Utell. MJ: Gibb. FR: Speers. DM. (1997b). Ozone
        responsiveness in smokers and nonsmokers. Am J Respir Crit Care Med 155: 116-121.
Funabashi. H: Shima. M: Kuwaki. T: Hiroshima.  K: Kurivama. T (2004). Effects of repeated ozone exposure on
        pulmonary function and bronchial responsiveness in mice sensitized with ovalbumin. Toxicology 204:
        75-83. http://dx.doi.0rg/10.1016/i.tox.2004.06.047.
Garantziotis. S: Li. Z: Potts. EN: Lindsev. JY: Stober. VP: Polosukhin. VV: Blackwell. TS: Schwartz.  DA: Foster.
        WM: Hollingsworth. JW. (2010). TLR4 is necessary for hyaluronan-mediated airway
        hyperresponsiveness after ozone inhalation. Am J Respir Crit Care Med 181: 666-675.
        http://dx.doi.org/10.1164/rccm.200903-0381OC.
Goldberg.  MS:  Burnett.  RT: Yale. JF: Valois. MF: Brook. JR. (2006). Associations between ambient  air pollution
        and daily  mortality among persons with diabetes and cardiovascular disease. Environ Res  100: 255-
        267.
Halonen. Jl: Lanki. T: Tiittanen. P: Niemi. JV: Loh. M: J. P. (2009). Ozone and cause-specific cardiorespiratory
        morbidity  and  mortality. J Epidemiol Community  Health 64: 814-820.
        http://dx.doi. org/10.1136/iech.2009.087106.
Hamade. AK: Rabold. R: Tankerslev. CG. (2008). Adverse cardiovascular effects with acute particulate matter
        and ozone exposures: Interstrain variation in mice. Environ Health Perspect 116: 1033-1039.
        http://dx.doi. org/10.1289/ehp. 10689.
Hamade. AK: Tankerslev. CG.  (2009). Interstrain variation in cardiac and respiratory adaptation to repeated
        ozone and particulate matter exposures. Am J Physiol Regul Integr Comp Physiol 296: R1202-R1215.
        http://dx.doi.org/10.1152/aipregu.90808.2008.
Hamade. AK: Misra. V: Rabold. R: Tankerslev. CG. (2010). Age-related changes in cardiac and respiratory
        adaptation to acute ozone and carbon black exposures: Interstrain variation in mice. Inhal Toxicol 22:
        84-94. http://dx.doi.org/10.3109/08958378.2010.503974.
Hansen. CA: Barnett. AG: Pritchard. G. (2008). The effect of ambient  air pollution during early pregnancy on
        fetal ultrasonic measurements during mid-pregnancy. Environ Health Perspect 116: 362-369.
Draft - Do Not Cite or Quote                         8-36                                   September 2011

-------
Harkema. JR: Plopper. CG: Hyde. DM: StGeorge. JA: Dungworth. PL (1987a). Effects of an ambient level of
        ozone on primate nasal epithelial mucosubstances: quantitative histochemistry. Am J Pathol 127: 90-
        96.
Hazucha. MJ: Folinsbee. LJ: Bromberg. PA. (2003). Distribution and reproducibility of spirometric response to
        ozone by gender and age. J Appl Physiol 95: 1917-1925.
Henrotin. JB: Besancenot. JP: Beiot. Y: Giroud. M. (2007). Short-term effects of ozone  air pollution on
        ischaemic stroke occurrence: A case-crossover analysis from a 10-year population-based study in
        Dijon,  France. Occup Environ Med 64: 439-445.
Henrotin. JB: Zeller. M: Lorgis.  L: Cottin. Y: Giroud. M: Beiot. Y. (2010). Evidence of the role of short-term
        exposure to ozone on  ischaemic cerebral and cardiac events: The Dijon Vascular Project (DIVA). Heart
        96: 1990-1996. http://dx.doi.org/10.1136/hrt.2Q10.200337.
Hernandez-Cadena. L: Holguin. F: Barraza-Villarreal. A: Del Rfo-Navarro. BE: Sienra-Monge. JJ: Romieu. I.
        (2009). Increased levels of outdoor air pollutants are associated with reduced  bronchodilation in
        children with asthma. Chest 136:  1529-1536. http://dx.doi.org/10.1378/chest.08-1463.
Hernandez. ML: Lav. JC: Harris. B: Esther. CR: Brickev. WJ: Bromberg. PA: Diaz-Sanchez.  D: Devlin. RB:
        Kleeberger. SR: Alexis. NE: Peden. DB.  (2010). Atopic asthmatic subjects but  not atopic subjects
        without asthma have enhanced inflammatory response to ozone. J Allergy Clin Immunol 126: 537-544.
        http://dx.doi.0rg/10.1016/i.iaci.2010.06.043.
Hollingsworth. JW: Free. ME: Li. Z: Andrews. LN: Nakano. H: Cook. DN. (2010). Ozone activates pulmonary
        dendritic cells and promotes allergic sensitization through a Toll-like receptor 4-dependent mechanism.
        J Allergy Clin Immunol 125: 1167-1170.
Hollingsworth. JW. II: Cook. DN: Brass. DM: Walker. JKL: Morgan. PL: Foster. WM: Schwartz. DA. (2004). The
        role of Toll-like receptor 4 in environmental airway injury in mice. Am J Respir Crit Care Med 170: 126-
        132.
Holz. O: Jorres. RA: Timm. P: Mucke. M: Richter. K: Koschvk. S: Magnussen.  H. (1999). Ozone-induced airway
        inflammatory changes differ between individuals and are reproducible. Am J Respir Crit Care Med  159:
        776-784.
Holz. O: Tal-Singer. R: Kanniess. F: Simpson. KJ: Gibson. A: Vessev. RSJ: Janicki. S: Magnussen. H: Jorres.
        RA: Richter, K.  (2005). Validation of the  human ozone challenge model as a tool for assessing anti-
        inflammatory drugs in early development. J Clin Pharmacol 45: 498-503.
Horstman. DH: Ball. BA:  Brown. J: Gerritv. T: Folinsbee. LJ. (1995). Comparison of pulmonary responses of
        asthmatic and nonasthmatic subjects performing light exercise while exposed  to a low level of ozone.
        Toxicol Ind Health 11: 369-385.
Houslev. DG: Eccles. R: Richards. RJ. (1996). Gender difference in the concentration of the antioxidant  uric
        acid in human nasal lavage. Acta Otolaryngol 116: 751-754.
Hu. S. -C: Ben-Jebria. A: Ultman. JS. (1994). Longitudinal distribution of ozone absorption in the lung: Effects of
        respiratory flow. J Appl Physiol 77: 574-583.
Huffman, LJ: Beighley, CM: Frazer, DG: McKinney, WG: Porter, DW. (2006). Increased susceptibility of the
        lungs of hyperthyroid rats to oxidant injury: Specificity of effects. Toxicology 225: 119-127.
        http://dx.doi.0rg/10.1016/i.tox.2006.05.008.
Islam, T: McConnell, R:  Gauderman, WJ: Avol, E: Peters, JM: Gilliland, FD. (2008). Ozone,  oxidant defense
        genes and risk  of asthma during adolescence.  Am J Respir Crit Care Med 177: 388-395.
        http://dx.doi.org/10.1164/rccm.200706-863OC.
Islam, T: Berhane, K: McConnell, R: Gauderman, WJ: Avol, E: Peters, JM: Gilliland, FD. (2009). Glutathione-S-
        transferase (GST) P1, GSTM1, exercise, ozone and asthma incidence in school children. Thorax 64:
        197-202.  http://dx.doi.org/10.1136/thx.2008.099366.
Jerrett, M: Burnett, RT: Pope, CA, III:  Ito, K: Thurston, G: Krewski, D: Shi, Y: Calle, E: Thun, M. (2009). Long-
        term ozone exposure and mortality. N Engl J Med 360: 1085-1095.
        http://dx.doi.org/10.1056/NEJMoa0803894.
Joad, JP: Kott, KS: Brie, JM: Peake, JL: Plopper, CG: Schelegle, ES: Gershwin, LJ: Pinkerton, KE. (2006).
        Structural and functional localization of airway  effects from episodic exposure  of infant monkeys to
        allergen and/or ozone. Toxicol Appl Pharmacol 214: 237-243.
        http://dx.doi.0rg/10.1016/i.taap.2005.12.012.
Draft - Do Not Cite or Quote                        8-37                                  September 2011

-------
Johnston. RA: Schwartzman. IN: Flvnt. L: Shore. SA. (2005b). Role of interleukin-6 in murine airway responses
        to ozone. Am J Physiol Lung Cell Mol Physiol 288: L390-L397.
        http://dx.doi.org/10.1152/aipluna.00007.2004.
Jorres. R: Nowak. D: Magnussen. H: Speckin. P: Koschvk. S. (1996). The effect of ozone exposure on allergen
        responsiveness in subjects with asthma or rhinitis. Am J Respir Crit Care Med  153: 56-64.
Kan. H: London. SJ: Chen. G: Zhang. Y: Song. G: Zhao. N: Jiang. L: Chen. B. (2008). Season, sex, age, and
        education as modifiers of the effects of outdoor air pollution on daily  mortality in Shanghai, China: The
        Public Health and Air Pollution in Asia (PAPA) Study. Environ Health  Perspect 116:  1183-1188.
Karl. F: Hatch. G: Slade. R: Crissman. K: Simeonova.  PP: Luster. M. (1997).  Dietary  restriction mitigates ozone-
        induced lung inflammation in rats: A role for endogenous antioxidants. Am J  Respir Cell Mol Biol 17:
        740-747.
Katsouvanni. K: Samet. JM: Anderson. HR: Atkinson. R:  Le Tertre. A: Medina. S: Samoli. E:  Touloumi. G:
        Burnett. RT: Krewski. D: Ramsay. T: Dominici. F: Peng.  RD: Schwartz. J: Zanobetti. A. (2009). Air
        pollution and health: A European and North American approach (APHENA).  (Research Report 142).
        Boston, MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=327.
Kehrl. HR: Peden. DB: Ball. BA: Folinsbee. LJ: Horstman. PH. (1999). Increased specific airway reactivity of
        persons with mild allergic asthma after 7.6 hours of exposure to 0.16 ppm ozone. J Allergy Clin
        Immunol 104: 1198-1204.
Kenvon. NJ: Van  Per Vliet. A: Schock. BC: Okamoto. T: McGrew. GM: Last. JA. (2002). Susceptibility to ozone-
        induced acute lung injury in iNOS-deficient mice. Am J Physiol 282: L540-L545.
Khatri. SB: Holguin. FC: Ryan. PB: Mannino. D: Erzurum. SC: Teague. WG. (2009). Association of ambient
        ozone exposure with airway inflammation and allergy in  adults with asthma.  J Asthma 46: 777-785.
        http://dx.doi. org/10.1080/02770900902779284.
Kim. CS: Alexis. NE: Rappold. AG: Kehrl. H: Hazucha. MJ: Lav. JC: Schmitt.  MT: Case. M: Devlin. RB: Peden.
        DB: Diaz-Sanchez. D. (2011). Lung function and inflammatory responses in  healthy young adults
        exposed to 0.06 ppm ozone for 6.6  hours. Am J Respir Crit Care Med 183: 1215-1221.
        http://dx.doi.ora/10.1164/rccm.201011-1813OC.
Kleeberaer. SR: Levitt. RC: Zhang. L. -Y: Longphre.  M: Harkema. J: Jedlicka. A: Eleff. SM: DiSilvestre. D:
        Holrovd, KJ. (1997).  Linkage analysis of susceptibility to ozone-induced lung inflammation in inbred
        mice. Nat Genet 17: 475-478.
Kleeberaer. SR: Reddv. S: Zhang. L. -Y: Jedlicka. AE.  (2000). Genetic susceptibility to ozone-induced lung
        hyperpermeability: Role of toll-like receptor 4. Am J Respir Cell Mol Biol 22: 620-627.
Kleeberaer. SR: Reddv. SP: Zhang. L. -Y:  Cho. H. -Y: Jedlicka. AE. (2001). Toll-like receptor 4 mediates ozone-
        induced murine lung hyperpermeability via inducible nitric oxide synthase. Am J Physiol 280: L326-
        L333.
Ko. FWS: Tarn. W: Wong. TW: Lai. CKW. (2007). Effects of air pollution on asthma hospitalization rates in
        different age groups in  Hong Kong.  Clin Exp Allergy 37:  1312-1319.
Kodavanti, UP: Costa, PL: Dreher, KL: Crissman, K: Hatch, GE. (1995). Ozone-induced tissue injury and
        changes in antioxidant homeostasis in normal and ascorbate-deficient guinea pigs. Biochem
        Pharmacol 50: 243-251. http://dx.doi.org/10.1016/0006-2952(95)00122-G.
Kooter, IM: Pennings, JL: Fokkens, PH: Leseman, PL: Boere, AJ: Gerlofs-Nijland, ME: Cassee, FR: Schalk, JA:
        Orzechowski. TJ: Schaap. MM: Breit. TM: Dormans. JA: van Oostrom. CT: de Vries. A: van Steeg. H.
        (2007). Ozone induces clear cellular and molecular responses in the mouse  lung independently of the
        transcription-coupled repair status. JAppI Physiol 102: 1185-1192.
        http://dx.doi.ora/10.1152/iapplphvsiol.00796.2006.
Kumarathasan. P: Blais. E: Goegan. P: Yagminas. A: Guenette. J: Adamson.  IY: Crapo. JD: Mason. RJ:
        Vincent, R. (2005). 90-day repeated inhalation exposure of surfactant Protein-C/tumor necrosis factor-
        alpha, (SP-C/TNF-alpha) transgenic mice to air  pollutants. Int J Toxicol 24: 59-67.
Lagorio. S: Forastiere. F: Pistelli. R: lavarone. I: Michelozzi. P: Fano. V: Marconi. A: Ziemacki. G: Ostro. BD.
        (2006). Air pollution and lung function among susceptible adult subjects: A panel study. Environ Health
        5: 11. http://dx.doi.org/10.1186/1476-069X-5-11.
Lee. IM: Tsai. SS: Ho. CK: Chiu. HF: Wu. TN: Yang.  CY. (2008a). Air pollution and hospital admissions for
        congestive heart failure: Are there potentially sensitive groups? Environ Res 108: 348-353.
        http://dx.doi.0rg/10.1016/i.envres.2008.07.024.
Draft - Do Not Cite or Quote                        8-38                                  September 2011

-------
Lee. JT: Son. JY: Kim. H: Kim. SY. (2006). Effect of air pollution on asthma-related hospital admissions for
        children by socioeconomic status associated with area of residence. Arch Environ Occup Health 61:
        123-130.
Lee. YL: McConnell. R: Berhane. K:  Gilliland. FD. (2009b). Ambient ozone modifies the effect of tumor necrosis
        factor G-308Aon bronchitic symptoms among children with asthma. Allergy 64: 1342-1348.
        http://dx.doi. orq/10.1111/i.1398-9995.2009.02014.x.
Lewis. TC: Robins. TG: Dvonch. JT: Keeler.  GJ: Yip. FY: Mentz. GB: Lin. X: Parker. EA: Israel. BA: Gonzalez. L:
        Hill, Y. (2005). Air pollution-associated changes in lung function among asthmatic children in Detroit.
        Environ Health Perspect 113:  1068-1075.
Liao. D:  Heiss. G:  Chinchilli. VM: Duan. Y: Folsom. AR: Lin. HM: Salomaa. V. (2005). Association of criteria
        pollutants with plasma hemostatic/inflammatory markers: A population-based study. J Expo Sci Environ
        Epidemiol 15: 319-328.
Lim. Y: Phung. AD: Corbacho. AM: Aung. HH: Maioli. E: Reznick. AZ: Cross. CE:  Davis. PA: Valacchi. G. (2006).
        Modulation of cutaneous wound healing by ozone: Differences between young and aged  mice. Toxicol
        Lett 160: 127-134.  http://dx.doi.ora/10.1016/i.toxlet.2005.06.013.
Lin. M: Stieb. DM:  Chen. Y.  (2005). Coarse particulate matter and hospitalization for respiratory infections in
        children younger than 15 years in Toronto: A case-crossover analysis.  Pediatrics 116: 235-240.
Lin. S: Liu. X: Le. LH: Hwang. SA. (2008b). Chronic exposure to ambient ozone and asthma hospital
        admissions among children. Environ Health Perspect 116: 1725-1730.
        http://dx.doi.org/10.1289/ehp.11184.
Liu. L: Poon. R: Chen. L:  Frescura. AM: Montuschi. P: Ciabattoni. G: Wheeler. A:  Dales. R. (2009a). Acute
        effects of  air  pollution on pulmonary function, airway inflammation,  and oxidative stress in asthmatic
        children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehp11813.
Lopez. I: Sanchez. I: Bizarro. P: Acevedo. S: Ustarroz. M: Fortoul. T. (2008). Ultrastructural alterations during
        embryonic rats' lung development caused by ozone. J Electron Microsc (Tokyo) 57: 19-23.
        http://dx.doi. org/10.1093/imicro/dfm033.
Mar. TF: Koenig. JQ. (2009). Relationship between visits to emergency departments for asthma and ozone
        exposure  in greater Seattle, Washington. Ann Allergy Asthma Immunol 103: 474-479.
McDonnell, WF. (1996). Individual variability in human lung function responses to ozone exposure. Environ
        Toxicol Pharmacol  2: 171-175.
McDonnell. WF: Stewart. PW: Smith. MV: Pan.  WK: Pan. J. (1999). Ozone-induced respiratory symptoms:
        Exposure-response models and association with lung function. Eur Respir J 14: 845-853.
McDonnell. WF: Stewart. PW: Smith. MV. (2010). Prediction of ozone-induced  lung function responses in
        humans. Inhal Toxicol 22: 160-168.  http://dx.doi.org/10.3109/08958370903089557.
McDonnell, WF, III: Horstman, DH: Abdul-Salaam, S: House,  DE.  (1985a). Reproducibility of individual
        responses to ozone exposure. Am Rev Respir Dis 131: 36-40.
Medina-Ramon. M: Zanobetti. A: Schwartz. J. (2006). The effect of ozone and PM10 on hospital admissions for
        pneumonia and chronic obstructive  pulmonary disease: A national  multicity study. Am J Epidemiol 163:
        579-588.  http://dx.doi.org/10.1093/aie/kwi078.
Medina-Ramon. M: Schwartz. J. (2008). Who is more vulnerable to die from ozone air pollution? Epidemiology
        19: 672-679.
Middleton. N: Yiallouros. P:  Kleanthous. S: Kolokotroni. O: Schwartz. J: Dockerv.  DW: Demokritou. P: Koutrakis.
        P, (2008). A 10-year time-series analysis of respiratory and cardiovascular morbidity in Nicosia, Cyprus:
        The effect of short-term changes in  air pollution and dust storms. Environ Health 7: 39.
Miller. FJ: Illing. JW: Gardner. DE. (1978). Effect of urban ozone levels on laboratory-induced respiratory
        infections. Toxicol Lett 2:  163-169.
Morello-Frosch, R: Jesdale, BM: Sadd, JL: Pastor, M. (2010). Ambient air pollution exposure and full-term birth
        weight in California. Environ Health  9: 44. http://dx.doi.org/10.1186/1476-069X-9-44.
Mudwav. IS: Blomberg. A: Frew. AJ:  Holgate. ST: Sandstrom. T: Kelly. FJ. (1999a). Antioxidant consumption
        and repletion kinetics in nasal lavage fluid following exposure of healthy human volunteers to ozone.
        Eur Respir J  13:  1429-1438.
Draft - Do Not Cite or Quote                        8-39                                  September 2011

-------
Mudwav. IS: Stenfors. N: Blomberg. A: Helledav. R: Punster. C: Marklund. SL: Frew. AJ: Sandstrom. T: Kelly.
        FJ. (2001). Differences in basal airway antioxidant concentrations are not predictive of individual
        responsiveness to ozone: A comparison of healthy and mild asthmatic subjects. Free Radic Biol Med
        31: 962-974.
Neidell,  M: Kinnev, PL. (2010). Estimates of the association between ozone and asthma hospitalizations that
        account for behavioral responses to air quality information. Environ Sci Pol 13: 97-103.
        http://dx.doi.0rg/10.1016/i.envsci.2009.12.006.
Nodelman, V: Ultman, JS. (1999). Longitudinal distribution of chlorine absorption in human airways: A
        comparison to ozone absorption. J Appl Physiol 87:  2073-2080.
Ovarzun. M: Dussaubat. N: Gonzalez. S. (2005). Effect of 0.25 ppm ozone exposure on pulmonary damage
        induced by bleomycin. Biol Res 38: 353-358.
Palli. D:  Sera. F: Giovannelli. L: Masala. G: Grechi. D: Bendinelli. B: Caini. S: Dolara. P: Saieva. C. (2009).
        Environmental ozone exposure and oxidative DMA damage in adult residents of Florence, Italy. Environ
        Pollut 157: 1521-1525. http://dx.doi.org/S0269-7491 (08)00472-7 fpii110.1016/i.envpol.2008.09.011.
Paquette. NC: Zhang. L. -Y: Ellis. WA: Scott. AL: Kleeberger. SR. (1996). Vitamin A deficiency enhances ozone-
        induced lung injury. Am J Physiol 270: L475-L482.
Paulu, C: Smith, AE. (2008). Tracking associations between ambient ozone and asthma-related emergency
        department visits using case-crossover analysis. J Public Health Manag Pract 14: 581-591.
Peden. DB: Setzer. RW. Jr: Devlin. RB. (1995). Ozone exposure has both a priming effect on allergen-induced
        responses and an intrinsic inflammatory action in the nasal airways of perennially allergic asthmatics.
        Am J Respir Grit Care Med 151: 1336-1345.
Peden. DB: Boehlecke. B: Horstman. D: Devlin. R. (1997). Prolonged acute exposure to 0.16 ppm ozone
        induces eosinophilic airway inflammation in asthmatic subjects with allergies. J Allergy Clin Immunol
        100: 802-808.
Peel. JL: Metzger. KB: Klein. M: Flanders. WD: Mulholland. JA: Tolbert. PE. (2007). Ambient air pollution and
        cardiovascular emergency department visits in potentially sensitive groups. Am J  Epidemiol 165: 625-
        633.
Pleis. JR: Lucas. JW: Ward. BW. (2009). Summary health statistics for U.S. adults: National Health Interview
        Survey, 2008. In Vital and  Health Statistics, 10 (Vol.  242). (DHHS 2010-1570). Hyattsville, MD: National
        Center  for Health Statistics.
Plopper. CG: Smiley-Jewell. SM: Miller. LA:  Fanucchi. MV: Evans. MJ: Buckpitt. AR: Avdalovic. M: Gershwin.
        LJ: Joad, JP: Kajekar, R: Larson, S: Pinkerton, KE: Van Winkle, LS: Schelegle, ES: Pieczarka, EM: Wu,
        R: Hyde. DM. (2007). Asthma/allergic airways disease: Does postnatal exposure to environmental
        toxicants promote airway pathobiology? Toxicol Pathol 35: 97-110.
        http://dx.doi. org/10.1080/01926230601132030.
Qian. Z: Lin. H. -M: Chinchilli. VM: Lehman.  EB: Duan. Y: Craig. TJ: Wilson. WE: Liao. D:  Lazarus. SC:
        Bascom. R. (2009). Interaction of ambient air pollution with asthma medication on exhaled nitric oxide
        among  asthmatics. Arch Environ Occup Health 64: 168-176.
        http://dx.doi. org/10.1080/19338240903240616.
Que. LG: Stiles. JV: Sundv. JS: Foster. WM. (2011). Pulmonary function,  bronchial reactivity, and epithelial
        permeability are response phenotypes to ozone and develop differentially in healthy humans. J Appl
        Physiol 111: 679-687.  http://dx.doi.ora/10.1152/iapplphvsiol.00337.2011.
Rivas-Arancibia. S: Dorado-Martinez. C: Borgonio-Perez. G: Hiriart-Urdanivia. M: Verdugo-Diaz. L: Duran-
        Vazquez, A: Colin-Baranque, L: Avila-Costa, MR. (2000). Effects oftaurine on ozone-induced memory
        deficits and  lipid peroxidation  levels in brains of young, mature, and old rats.  Environ Res 82: 7-17.
        http://dx.doi. org/10.1006/enrs. 1999.3996.
Romieu, I: Sienra-Monge, JJ: Ramirez-Aguilar, M:  Moreno-Macias, H: Reyes-Ruiz, Ml: Estela del Rio-Navarro,
        B: Hernandez-Avila. M: London. SJ. (2004a). Genetic polymorphism of GSTM1 and antioxidant
        supplementation influence lung function in relation to ozone exposure in asthmatic children in Mexico
        City. Thorax 59: 8-10.
Romieu. I: Ramirez-Aguilar. M: Moreno-Macias. H: Barraza-Villarreal. A: Miller. P: Hernandez-Cadena. L:
        Carbaial-Arrovo. LA: Hernandez-Avila. M.  (2004b). Infant mortality and air pollution: Modifying effect by
        social class. J Occup Environ Hyg 46: 1210-1216.
Draft - Do Not Cite or Quote                        8-40                                  September 2011

-------
Romieu. I: Ramirez-Aguilar. M: Sienra-Monqe. JJ: Moreno-Macias. H: Del Rio-Navarro. BE: David. G: Marzec.
        J: Hernandez-Avila, M: London, S. (2006). GSTM1 and GSTP1 and respiratory health in asthmatic
        children exposed to ozone. Eur Respir J 28: 953-959. http://dx.doi.org/10.1183/09031936.06.00114905.
Romieu. I: Barraza-Villarreal. A: Escamilla-Nunez. C: Texcalac-Sangrador. JL: Hernandez-Cadena. L: Dfaz-
        Sanchez, D: De Batlle, J: Del Rio-Navarro, BE. (2009). Dietary intake, lung function and airway
        inflammation in Mexico City school children exposed to air pollutants. Respir Res 10: 122.
Ruidavets. J. -B: Cassadou. S: Cournot. M: Bataille. V: Mevbeck. M: Ferrieres. J. (2005a). Increased resting
        heart rate with pollutants in a population based study. J Epidemiol Community Health 59: 685-693.
Sacks. JD: Stanek. LW: Luben. TJ: Johns. DO: Buckley. BJ: Brown. JS: Ross. M. (2011). Particulate matter
        induced health effects: Who's susceptible? Environ Health Perspect 119: 446-454.
        http://dx.doi. org/10.1289/ehp. 1002255.
Salam. MT: Islam. T: Gauderman. WJ: Gilliland. FD. (2009). Roles of arginase variants, atopy, and ozone in
        childhood asthma.  J Allergy Clin Immunol 123: 596-602. http://dx.doi.Org/10.1016/i.iaci.2008.12.020.
Samet, JM: Hatch, GE: Horstman, D: Steck-Scott, S: Arab, L: Bromberg, PA: Levine, M: McDonnell, WF: Devlin,
        RB. (2001). Effect of antioxidant supplementation on ozone-induced lung injury in human subjects. Am
        J Respir Grit Care Med 164: 819-825.
Sarangapani,  R: Gentry, PR: Covington, TR: Teeguarden,  JG: Clewell HJ, III. (2003). Evaluation of the potential
        impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of vapors.
        InhalToxicol 15: 987-1016.
Sawyer, K: Brown, J: HazuchaM:  Bennett, WD. (2007). The effect of exercise on nasal uptake of ozone in
        healthy human adults. J Appl Physiol 102: 1380-1386.
        http://dx.doi.org/10.1152/iapplphvsiol.00269.2006.
Scannell, C: Chen,  L: Aris, RM: Tager, I: Christian, D: Ferrando, R: Welch, B: Kelly,  T: Balmes, JR. (1996).
        Greater ozone-induced inflammatory responses in subjects with asthma. Am J Respir Crit Care Med
        154: 24-29.
Schelegle, ES: Miller, LA: Gershwin, LJ: Fanucchi, MV: Van Winkle, LS: Gerriets, JE: Walby, WF:  Mitchell, V:
        Tarkington. BK: Wong. VJ: Baker. GL: Pantle. LM: Joad. JP: Pinkerton.  KE: Wu. R: Evans. MJ: Hyde.
        DM: Plopper. CG. (2003). Repeated episodes of ozone inhalation amplifies the effects of allergen
        sensitization and inhalation on airway immune and structural development in Rhesus monkeys. Toxicol
        Appl Pharmacol 191: 74-85.
Seal. E.  Jr:  McDonnell. WF: House. DE: Salaam. SA:  Dewitt. PJ: Butler. SO: Green. J: Raggio. L. (1993). The
        pulmonary response of white and black adults to six concentrations of ozone. Am J Respir Crit Care
        Med 147: 804-810.
Seal. E.  Jr:  McDonnell. WF: House. DE. (1996). Effects of age, socioeconomic status,  and menstrual cycle on
        pulmonary response to ozone. Arch Environ Occup Health 51: 132-137.
Servais. S: Boussouar.  A: Molnar. A: Douki. T: Peguignot.  JM: Favier. R. (2005). Age-related sensitivity to lung
        oxidative stress during ozone exposure. Free Radic Res 39: 305-316.
        http://dx.doi. org/10.1080/10715760400011098.
Sherry. B: Blanck. HM:  Galuska. DA: Pan. L: Dietz. WH: Balluz. L. (2010). Vital signs: State-specific obesity
        prevalence among adults - United States, 2009. MMWR Recomm Rep  59: 951-955.
Shore, SA: Lang, JE: Kasahara, PI: Lu, Fl_: Verbout, NG: Si, H: Williams, ES: Terry, RD: Lee, A: Johnston, RA.
        (2009). Pulmonary responses to subacute ozone  exposure in obese vs. lean mice. J Appl Physiol 107:
        1445-1452. http://dx.doi.org/10.1152/iapplphvsiol.00456.2009.
Sienra-Monge, JJ: Ramirez-Aguilar, M: Moreno-Macias, H: Reyes-Ruiz, Ml: Del  Rio-Navarro, BE:  Ruiz-Navarro,
        MX: Hatch. G:  Crissman.  K: Slade. R: Devlin. RB: Romieu. I. (2004). Antioxidant supplementation and
        nasal inflammatory responses among young asthmatics exposed to high levels of ozone. Clin Exp
        Immunol 138:  317-322. http://dx.doi.Org/10.1111/i.1365-2249.2004.02606.x.
Silverman. RA: Ito.  K. (2010). Age-related association of fine particles and ozone with severe acute asthma in
        New York City. J Allergy Clin Immunol 125: 367-373.e365. http://dx.doi.Org/10.1016/i.iaci.2009.10.061.
SSDAN  CensusScope. (Social Science Data Analysis Network, CensusScope).  (201 Oa). United States: Age
        distribution [Data Set]. Ann Arbor, Michigan: Social Science Data Analysis Network.  Retrieved from
        http://www.censusscope.org/us/chart age.html
Draft - Do Not Cite or Quote                        8-41                                  September 2011

-------
SSDAN CensusScope. (Social Science Data Analysis Network, CensusScope). (201 Ob). United States:
        Population by race [Data Set]. Ann Arbor, Michigan. Retrieved from
        http://www.censusscope.org/us/chart  race.html
SSDAN CensusScope. (Social Science Data Analysis Network, CensusScope). (2010c). United States: Poverty
        by age [Data Set]. Ann Arbor, Michigan. Retrieved from
        http://www.censusscope.org/us/chart  povertv.html
Stafoggia. M: Forastiere. F: Faustini. A:  Bigger!. A: Bisanti. L: Cadum. E: Cernigliaro. A: Mallone. S: Pandolfi. P:
        Serinelli, M: Tessari, R: Vigotti,  MA: Perucci, CA. (2010). Susceptibility factors to ozone-related
        mortality: A population-based case-crossover analysis. Am J Respir Crit Care Med 182: 376-384.
        http://dx.doi.org/10.1164/rccm.200908-1269OC.
Steinvil, A: Kordova-Biezuner, L:  Shapira, I: Berliner, S:  Rogowski, O. (2008). Short-term exposure to air
        pollution and inflammation-sensitive biomarkers. Environ Res 106: 51-61.
Tankerslev. CG: Kleeberger. SR. (1994). Ozone-induced inflammation and altered ventilation in genetically
        susceptible mice: A comparison of acute and subacute exposures. Toxicol Lett 72: 279-289.
Tankerslev. CG: Peng. RD:  Bedga.  D: Gabrielson. K: Champion.  HC. (2010). Variation in echocardiographic and
        cardiac hemodynamic effects of PM and ozone inhalation exposure in strains related to Nppa and Npr1
        gene knock-out mice.  Inhal Toxicol 22: 695-707. http://dx.doi.org/10.3109/08958378.2010.487549.
Thaller. El: Petronella. SA: Hochman. D: Howard. S: Chhikara. RS: Brooks. EG. (2008). Moderate increases in
        ambient PM2.5 and ozone are associated with  lung function decreases in beach lifeguards. J Occup
        Environ Med 50: 202-211.  http://dx.doi.org/10.1097/JOM.Ob013e31816386b4.
Torres. A: Utell. MJ: Morow. PE: Voter. KZ: Whitin. JC: Cox. C: Loonev.  RJ: Speers. DM: Tsai. Y: Frampton. MW.
        (1997). Airway inflammation in smokers and nonsmokers with varying responsiveness to ozone. Am J
        Respir Crit Care Med 156:  728-736.
Tovalin. H: Valverde. M:  Morandi. MT: Blanco.  S: Whitehead. L: Roias. E. (2006).  DNAdamage in outdoor
        workers occupationally exposed to environmental air pollutants. Occup Environ Med 63: 230-236.
Trenga, CA:  Koenig, JQ: Williams, PV. (2001).  Dietary antioxidants and ozone-induced bronchial
        hyperresponsiveness in adults with asthma. Arch Environ Occup Health 56: 242-249.
Triche. EW: Gent. JF: Holford. TR:  Belanger. K: Bracken. MB: Beckett. WS: Naeher. L: McSharrv. J. -E:
        Leaderer, BP (2006). Low-level ozone exposure and respiratory symptoms in infants. Environ Health
        Perspect 114: 911-916. http://dx.doi.org/10.1289/ehp.8559.
U.S. Census Bureau. (2010). U.S. population projections [Data Set]. Retrieved from
        http://www.census.gov/population/www/proiections/proiectionsagesex.html
U.S. EPA. (U.S. Environmental Protection Agency). (1996a). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/P-93/004AF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2006a). Aging and toxic response: Issues relevant to  risk
        assessment. (EPA/600/P-03/004A). Washington, DC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=156648.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2009d). Integrated science assessment for particulate
        matter. (EPA/600/R-08/139F).  Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
U.S. EPA. (U.S. Environmental Protection Agency). (2010c). Integrated science assessment for carbon
        monoxide.  (EPA/600/R-09/019F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=218686.
Vagaggini. B: Taccola. M: Clanchetti. S:  Carnevali. S: Bartoli. ML: Bacci. E: Dente. FL: Pi Franco. A: Giannini. D:
        Paggiaro. PL. (2002). Ozone exposure increases eosinophilic airway response induced by previous
        allergen challenge. Am J Respir Crit Care Med  166: 1073-1077.
Vagaggini. B: Bartoli.  MLE:  Cianchetti. S: Costa. F: Bacci. E: Dente. FL: Pi Franco. A:  Malagrino. L: Paggiaro. P.
        (2010). Increase in markers of airway  inflammation after ozone exposure can be observed also in
        stable treated asthmatics with minimal functional response to ozone. Respir Res 11:5.
        http://dx.doi.org/10.1186/1465-9921-11-5.
Draft - Do Not Cite or Quote                        8-42                                  September 2011

-------
Valacchi. G: Vasu. VT: Yokohama. W: Corbacho. AM: Phung. A: Lim. Y: Aung. HH: Cross. CE: Davis. PA.
        (2007). Lung vitamin E transport processes are affected by both age and environmental oxidants in
        mice. Toxicol Appl Pharmacol 222: 227-234. http://dx.doi.Org/10.1016/i.taap.2007.04.010.
Valacchi. G: Pecorelli. A: Mencarelli. M:  Maioli. E: Davis.  PA. (2009). Beta-carotene prevents ozone-induced
        proinflammatory markers in murine skin. Toxicol Ind Health 25: 241-247.
        http://dx.doi. orq/10.1177/0748233709103030.
Vancza. EM: Galdanes.  K: Gunnison. A: Hatch. G: Gordon. T. (2009). Age, strain, and gender as factors for
        increased sensitivity of the mouse lung to inhaled ozone. Toxicol Sci 107: 535-543.
        http://dx.doi. orq/10.1093/toxsci/kfn253.
Villeneuve. PJ: Chen. L: Stieb. D: Rowe. BH.  (2006a). Associations between outdoor air pollution and
        emergency department visits for stroke in Edmonton, Canada. Eur J Epidemiol21: 689-700.
Villeneuve. PJ: Chen. L: Rowe. BH: Coates. F. (2007). Outdoor air pollution and emergency department visits
        for asthma among children and adults: A case-crossover study in northern Alberta, Canada. Environ
        Health Global Access Sci Source 6: 40.  http://dx.doi.org/10.1186/1476-069X-6-40.
Vovnow. JA:  Fischer. BM: Zheng. S: Potts. EN: Grover. AR: Jaiswal. AK: Ghio. AJ: Foster. WM. (2009).
        NAD(P)H quinone oxidoreductase 1  is essential for ozone-induced oxidative stress in mice and
        humans. Am J Respir Cell  Mol Biol 41: 107-113. http://dx.doi.org/10.1165/rcmb.2008-0381OC.
Wagner. JG:  Jiang. Q: Harkema. JR: Illek. B:  Patel. DP: Ames. BN: Peden. DB. (2007). Ozone enhancement of
        lower airway allergic inflammation is  prevented by gamma-tocopherol. Free Radic Biol Med 43: 1176-
        1188. http://dx.doi.0rg/10.1016/i.freeradbiomed.2007.07.013.
Wagner. JG:  Harkema. JR: Jiang. Q: Illek. B:  Ames. BN:  Peden. DB. (2009). Gamma-tocopherol attenuates
        ozone-induced  exacerbation of allergic rhinosinusitis in rats. Toxicol Pathol 37: 481-491.
        http://dx.doi. org/10.1177/0192623309335630.
Wattiez. R: Noel-Georis. I: Cruvt. C: Broeckaert.  F:  Bernard. A: Falmagne. P. (2003). Susceptibility to oxidative
        stress: proteomic analysis  of bronchoalveolar lavage from ozone-sensitive and ozone-resistant strains
        of mice. Proteomics 3: 658-665. http://dx.doi.org/10.1002/pmic.20030Q417.
Weinmann. GG: Weidenbach-Gerbase.  M: Foster. WM: Zacur. H: Frank. R. (1995a). Evidence for ozone-
        induced small-airway dysfunction: Lack of  menstrual-cycle and gender effects. Am J Respir Crit Care
        Med 152: 988-996.
Wenten. M: Gauderman. WJ: Berhane. K: Lin. PC:  Peters. J: Gilliland. FD. (2009). Functional variants in the
        catalase and myeloperoxidase genes, ambient air pollution, and respiratory-related school absences:
        An example of epistasis in  gene-environment interactions. Am J Epidemiol 170: 1494-1501.
        http://dx.doi.org/10.1093/aie/kwp310.
Williams. AS: Leung. SY: Nath. P: Khorasani.  NM: Bhavsar.  P: Issa. R: Mitchell. JA: Adcock. IM: Chung. KF.
        (2007b). Role of TLR2, TLR4, and MyD88  in murine ozone-induced airway hyperresponsiveness and
        neutrophilia. J Appl Physiol 103: 1189-1195. http://dx.doi.org/10.1152/iapplphvsiol.00172.2007.
Wong. C. -M: Ou. C. -Q: Thach. T. -Q: Chau. Y. -K:  Chan. K. -P: Ho. S. -Y: Chung. RY: Lam. T. -H: Hedlev. AJ.
        (2007). Does regular exercise protect against air pollution-associated mortality? Prev Med 44: 386-392.
Wong. CM: Ou. CQ: Chan. KP: Chau. YK: Thach. TQ: Yang. L: Chung. RY: Thomas. GN: Peiris. JS: Wong. TW:
        Hedlev. AJ: Lam. TH. (2008). The effects of air pollution  on mortality in socially deprived urban areas in
        Hong Kong, China. Environ Health Perspect 116: 1189-1194.
Wong. CM: Yang. L: Thach. TQ: Chau. PY: Chan. KP: Thomas. GN: Lam. TH: Wong. TW:  Hedlev. AJ: Peiris.
        JS. (2009). Modification by influenza  on  health effects of air pollution in Hong Kong. Environ Health
        Perspect 117: 248-253. http://dx.doi.org/10.1289/ehp.11605.
Yu. M: Zheng. X: Witschi. H: Pinkerton. KE. (2002). The role of interleukin-6 in pulmonary inflammation and
        injury induced by exposure to environmental air  pollutants. Toxicol Sci 68: 488-497.
Zanobetti, A: Schwartz, J. (2008a).  Is there adaptation in the ozone mortality relationship: A multi-city case-
       crossover analysis. Environ Health  7: 22.  http://dx.doi.org/10.1186/1476-069X-7-22.
Draft - Do Not Cite or Quote                        8-43                                  September 2011

-------
      9  ENVIRONMENTAL  EFFECTS:  OZONE   EFFECTS
          ON  VEGETATION  AND  ECOSYSTEMS
          9.1    Introduction

 1                  This chapter synthesizes and evaluates the relevant science to help form the scientific
 2                  foundation for the review of a vegetation- and ecologically-based secondary NAAQS for
 3                  O3. The secondary NAAQS are based on welfare effects. The Clean Air Act (CAA)
 4                  definition of welfare effects includes, but is not limited to, effects on soils, water,
 5                  wildlife, vegetation, visibility, weather, and climate, as well as effects on materials,
 6                  economic values, and personal comfort and well-being. The effects of O3  as a greenhouse
 7                  gas and its direct effects on climate are discussed in Chapter 10 of this document.

 8                  The intent of the ISA, according to the CAA, is to "accurately reflect the latest scientific
 9                  knowledge expected from the presence of [a] pollutant in ambient air" (42 U.S.C.7408
10                  and 42 U.S.C.7409 (1999). This chapter of the ISA includes scientific research from
11                  biogeochemistry, soil science, plant physiology, and ecology conducted at multiple scales
12                  (e-g-, organ, organism, population, community, ecosystem). Key information  and
13                  judgments formerly found in the AQCDs regarding O3 effects on vegetation and
14                  ecosystems are found in this chapter. This chapter of the O3 ISA serves to update and
15                  revise Chapter 9 and AX9 of the 2006 O3 AQCD (U.S. EPA. 2006b).

16                  Numerous studies of the effects of O3 on vegetation and ecosystems were reviewed in the
17                  2006 O3 AQCD. That document concluded that the effects of ambient O3  on  vegetation
18                  and ecosystems appear to be widespread across the U.S., and  experimental studies
19                  demonstrated plausible mechanisms for these effects. Ozone effect studies published
20                  from 2005 to July 2011 are reviewed in this document in the context of the previous O3
21                  AQCDs. From 2005 to 2011,  some areas have had very little new research published and
22                  the reader is referred back to sections of the 2006 O3 AQCD for a more comprehensive
23                  discussion of those subjects. This chapter is focused on studies of vegetation and
24                  ecosystems that occur in the U.S. and that report endpoints or processes most relevant to
25                  the review of the secondary standard. Many studies have been published about vegetation
26                  and ecosystems outside of the U.S. and North America, largely in Europe  and Asia. This
27                  document includes discussion of studies of vegetation and ecosystems outside of North
28                  America only if those studies contribute to the general understanding of O3 effects across
29                  species and ecosystems. For example, studies outside North America are discussed that
30                  consider physiological and biochemical processes that contribute to the understanding of
31                  effects of O3 across species. Also, ecosystem studies outside of North America that
      Draft - Do Not Cite or Quote                      9-1                                September 2011

-------
 1                   contribute to the understanding of O3 effects on general ecosystem processes are
 2                   discussed in the chapter.

 3                   Sections of this chapter first discuss exposure methods, followed by effects on vegetation
 4                   and ecosystems at various spatial scales and ends with policy-relevant discussions of
 5                   exposure indices and exposure-response. Figure 9-1 is a simplified illustrative diagram of
 6                   the major pathway through which O3 enters plants and the major endpoints O3 may
 7                   affect. First, Section 9.2 presents a brief overview of various methodologies that have
 8                   been, and continue to be, central to quantifying O3 effects on vegetation (AX9.1 of the
 9                   2006 O3 AQCD for more detailed discussion) (U.S. EPA. 2006b). Sections 9.3 through
10                   9.4 begin with a discussion of effects at the cellular and subcellular level followed by
11                   consideration of the O3 effects on plant and ecosystem processes (Figure 9-1). In Section
12                   9.3, research is reviewed from the molecular to the biochemical and physiological levels
13                   in impacted plants, offering insight into the mode of action of O3. Section 9.4 provides a
14                   review of the effects of O3 exposure on major endpoints at the whole plant scale
15                   including growth, reproduction, visible foliar injury and leaf gas exchange in woody and
16                   herbaceous plants in the U.S., as well as a brief discussion of O3 effects on agricultural
17                   crop yield and quality. Section 9.4 also integrates the effects of O3 on individual plants in
18                   a discussion of available research for assessing the effect of O3 on ecosystems, along
19                   with available studies that could inform assessments of various ecosystem services (See
20                   section 9.4.1.2). The development of indices of O3 exposure and dose modeling is
21                   discussed in Section 9.5. Finally, exposure-response relationships for a number of tree
22                   species, native vegetation, and crop species and cultivars are reviewed, tabulated, and
23                   compared in Section 9.6 to form the basis for an assessment of the potential risk to
24                   vegetation  from current ambient levels of O3.
      Draft - Do Not Cite or Quote                       9-2                                 September 2011

-------
                    03 exposure
               03 uptake & physiology (Fig 9-2)
               •Antioxidant metabolism up-regulated
               •Decreased photosynthesis
               •Decreased stomatal conductance
               or sluggish stomatal response
              Effects on leaves
              Visible leaf injury
              Altered leaf production
              Altered leaf chemical composition
                Plant growth (Fig 9.8)
                •Decreased biomass accumulation
                •Altered reproduction
                •Altered carbon allocation
                •Altered crop quality
                                              Affected ecosystem services
                                              •Decreased productivity
                                              •Decreased C sequestration
                                              •Altered water cycling (Fig 9-7)
                                              •Altered community composition
                                              (i.e., plant, insects microbe)
              Belowground processes (Fig 9.8)
              •Altered litter production and decomposition
              •Altered soil carbon and nutrient cycling
              •Altered soil fauna and microbial communities
     Figure 9-1     An illustrative diagram of the major pathway through which Oz
                    enters plants and the major endpoints that Os may affect in plants
                    and ecosystems.
        9.2   Experimental Exposure Methodologies
           9.2.1   Introduction
1
2
3
4
5
6
7
A variety of methods for studying plant response to O3 exposures have been developed

over the last several decades. Methodological advancements since 2006 have not

fundamentally altered our understanding of O3 effects on plants or ecosystems. The

majority of methodologies currently used have been discussed in detail in the 1996 O3

AQCD and 2006 O3 AQCD. This section will serve as a short overview of the

methodologies and the reader is referred to the previous O3 AQCDs for more in-depth

discussion.
     Draft - Do Not Cite or Quote
                              9-3
September 2011

-------
            9.2.2   "Indoor," Controlled Environment, and Greenhouse Chambers

 1                  The earliest experimental investigations of the effects of O3 on plants utilized simple
 2                  glass or plastic-covered chambers, often located within greenhouses, into which a flow of
 3                  O3-enriched air or oxygen could be passed to provide the exposure. The types, shapes,
 4                  styles, materials of construction, and locations of these chambers have been numerous.
 5                  Hogsett et al. (1987a) have summarized the construction and performance of more
 6                  elaborate and better instrumented chambers since the 1960s, including those installed in
 7                  greenhouses (with or without some control of temperature and light intensity).

 8                  One greenhouse chamber approach that continues to yield useful information on the
 9                  relationships of O3 uptake to both physiological and growth effects employs continuous
10                  stirred tank reactors (CSTRs) first described by Heck et al. (1978). Although originally
11                  developed to permit mass-balance studies of O3 flux to plants, their use has more recently
12                  widened to include short-term physiological and growth studies of O3 * CO2 interactions
13                  (Loats and Rebbeck. 1999; Reinert et al.. 1997; Raoetal.. 1995; Reinert and Ho. 1995;
14                  Heagle etal.. 1994a). and validation of visible foliar injury on a variety of plant species
15                  (Kline et al., 2009; Orendovici et al., 2003).  In many cases, supplementary lighting and
16                  temperature control of the surrounding structure have been used to control or modify the
17                  environmental  conditions (Heagle et al., 1994a).

18                  Many investigations have utilized commercially available controlled environment
19                  chambers and walk-in rooms adapted to permit the introduction of a flow of O3 into the
20                  controlled air-volume. Such chambers continue to find use in genetic screening and in
21                  physiological and biochemical studies aimed primarily at improving our understanding of
22                  modes of action. For example, some of the studies of the O3 responses of common
23                  plantain (Plantago major) populations have been conducted in controlled environment
24                  chambers (Whitfield et al.. 1996: Reiling and Davison. 1994).

25                  More recently, some researchers have been interested in attempting to investigate direct
26                  O3 effects on reproductive processes, separate from the effects on vegetative processes
27                  (Black etal.. 2010). For this purpose, controlled exposure systems have been employed
28                  to expose the reproductive structures of annual plants to gaseous pollutants independently
29                  of the vegetative component (Black etal.. 2010; Stewart et al.. 1996).
            9.2.3  Field Chambers

30                  In general, field chamber studies are dominated by the use of various versions of the open
31                  top chamber (OTC) design, first described by Heagle et al. (1973) and Mandl et al.
32                  (1973). The OTC method continues to be a widely used technique in the U.S. and  Europe

      Draft - Do Not Cite or Quote                       9-4                                September 2011

-------
 1                   for exposing plants to varying levels of O3. Most of the new information confirms earlier
 2                   conclusions and provides additional support for OTC use in assessing plant species and in
 3                   developing exposure-response relationships. Chambers are generally ~3 m in diameter
 4                   with 2.5-m-high walls. Hogsett et al. (1987b) described in detail many of the various
 5                   modifications to the original OTC designs that appeared subsequently, e.g., the use of
 6                   larger chambers for exposing small trees (Kats etal..  1985) or grapevines (Mandl et al..
 7                   1989). the addition of a conical baffle at the top to improve ventilation (Kats etal.. 1976).
 8                   a frustum at the top to reduce ambient air incursions,  and a plastic rain-cap to exclude
 9                   precipitation (Hogsett et al.. 1985).  All versions of OTCs included the discharge of air via
10                   ports in annular ducting or interiorly perforated double-layered walls at the base of the
11                   chambers to provide turbulent mixing and the upward mass flow of air.

12                   Chambered systems, including OTCs, have several advantages. For instance, they can
13                   provide a range of treatment levels  including charcoal-filtered (CF), clean-air control, and
14                   several above ambient concentrations for O3 experiments. Depending on experimental
15                   intent, a replicated, clean-air control treatment is an essential component in many
16                   experimental designs. The OTC can provide a consistent, definable exposure because of
17                   the constant wind speed and delivery systems. Statistically robust concentration-response
18                   (C-R) functions can be developed using such systems for evaluating the implications of
19                   various alternative air quality scenarios on vegetation response. Nonetheless, there are
20                   several characteristics of the OTC design and operation that can lead to exposures that
21                   might differ from those experienced by plants in the field. First, the OTC plants are
22                   subjected to constant air flow turbulence, which, by lowering the boundary layer
23                   resistance to diffusion, may result in increased uptake. This may lead to an
24                   overestimation of effects relative to areas with less turbulence (Krupaet al.. 1995; Legge
25                   etal.. 1995). However, other research has found that OTC's may slightly change vapor
26                   pressure deficit (VPD) in a way that may decrease the uptake of O3 into leaves (Piikki et
27                   al.. 2008b). As with all methods that expose vegetation to modified O3 concentrations in
28                   chambers, OTCs create internal environments that differ from ambient air. This so-called
29                   "chamber effect" refers to the modification of microclimatic variables, including reduced
30                   and uneven light intensity, uneven rainfall, constant wind speed, reduced dew formation,
31                   and increased air temperatures (Fuhrer. 1994; Manning and Krupa. 1992). However, in at
32                   least one case where canopy resistance was quantified in OTCs and in the field, it was
33                   determined that gaseous pollutant exposure to crops in OTCs was similar to that which
34                   would have occurred at the same concentration in the field (Unsworth et al.. 1984a. b).
35                   Because of the standardized methodology and protocols used in National Crop  Loss
36                   Assessment Network (NCLAN) and other programs, the database can be assumed to be
37                   internally consistent.
      Draft - Do Not Cite or Quote                       9-5                                 September 2011

-------
 1                  While it is clear that OTCs can alter some aspects of the microenvironment and plant
 2                  growth, it is important to establish whether or not these differences affect the relative
 3                  response of a plant to O3. As noted in the  1996 O3 AQCD, evidence from a number of
 4                  comparative studies of OTCs and other exposure systems suggested that responses were
 5                  essentially the same regardless of exposure system used and chamber effects did not
 6                  significantly affect response. For example, a study of chamber effects examined the
 7                  responses of tolerant and sensitive white clover clones (Trifolium repens) to ambient O3
 8                  in greenhouse, open top, and ambient plots (Heagle etal..  1996). The response found in
 9                  OTCs was the same as in ambient plots.

10                  Another type of field chamber called a "terracosm" has been developed and used in
11                  recent studies (Lee et al.. 2009a). Concern over the need to establish realistic plant-litter-
12                  soil relationships as a prerequisite to studies of the effects of O3 and CO2 enrichment on
13                  ponderosa pine (Pinusponderosa) seedlings led Tingey et al. (1996) to develop closed,
14                  partially environmentally controlled, sun-lit chambers ("terracosms") incorporating 1-m-
15                  deep lysimeters containing forest soil in which the appropriate horizon structure was
16                  retained.

17                  Other researchers have recently published studies using another type of out-door chamber
18                  called recirculating Outdoor Plant Environment Chambers (OPECs) (Flowers et al..
19                  2007). These closed chambers are approximately 2.44 mx 1.52 m with a growth volume
20                  of approximately 3.7 m3 in each chamber. These chambers admit 90% of full sunlight and
21                  control temperature, humidity and vapor pressure (Tiscus etal.. 1999).
            9.2.4  Plume and FACE-Type Systems

22                  Plume systems are chamberless exposure facilities in which the atmosphere surrounding
23                  plants in the field is modified by the injection of pollutant gas into the air above or
24                  around them from multiple orifices spaced to permit diffusion and turbulence, so as to
25                  establish relatively homogeneous conditions as the individual plumes disperse and mix
26                  with the ambient air. They can only be used to increase the O3 levels in the ambient air.

27                  The most common plume system used in the U.S. is a modification of the free-air carbon-
28                  dioxide/ozone enrichment (FACE) system (Hendrey et al., 1999; Hendrey and Kimball.
29                  1994). Although originally designed to provide chamberless field facilities for studying
30                  the CO2 effects of climate change, FACE systems have been adapted to include the
31                  dispensing of O3 (Karnosky et al.. 1999). This method has been employed in Illinois
32                  (SoyFACE) to study soybeans (Morgan et al., 2004; Rogers et al., 2004) and in
33                  Wisconsin (Aspen FACE) to study trembling aspen (Populus tremuloides), birch (Betula
34                  papyriferd) and maple (Acer saccharum) (Karnosky et al.. 1999). Volk et al. (2003) also


      Draft - Do Not Cite or Quote                        9-6                                September 2011

-------
 1                   described a similar system for exposing grasslands that uses 7-m diameter plots. FACE
 2                   systems discharge the pollutant gas (O3 and/or CO2) through orifices spaced along an
 3                   annular ring (or torus) or at different heights on a ring of vertical pipes. Computer-
 4                   controlled feedback from the monitoring of gas concentration regulates the feed rate of
 5                   enriched air to the dispersion pipes. Feedback of wind speed and direction information
 6                   ensures that the discharges only occur upwind of the treatment plots, and that discharge is
 7                   restricted or closed down during periods of low wind speed or calm conditions. The
 8                   diameter of the arrays and their height (25-30 m) in some FACE systems requires large
 9                   throughputs of enriched air per plot, particularly in forest tree systems. The cost of the
10                   throughputs tends to limit the number of enrichment treatments, although Hendrey et al.
11                   (1999) argued that the cost on an enriched volume basis is comparable to that of chamber
12                   systems.

13                   Although plume systems make virtually none of the modifications to the physical
14                   environment that are inevitable with chambers, their successful use depends on selecting
15                   the appropriate numbers, sizes, and orientations of the discharge orifices to avoid "hot-
16                   spots" resulting from the direct impingement of jets of pollutant-enriched air on plant
17                   foliage (Werner and Fabian. 2002). Because mixing is unassisted and completely
18                   dependent on wind turbulence and diffusion, local gradients are inevitable especially in
19                   large-scale systems. FACE systems have provisions for shutting down under low wind
20                   speed or calm conditions and for an experimental area that is usually defined within a
21                   generous border in order to strive for homogeneity of the exposure concentrations within
22                   the treatment area. They are also dependent upon continuous computer-controlled
23                   feedback of the O3 concentrations in the mixed treated air and of the meteorological
24                   conditions. Plume and FACE systems also are unable to reduce O3 levels below ambient
25                   in areas where O3 concentrations are phytotoxic.
             9.2.5   Ambient Gradients

26                   Ambient O3 gradients that occur in the U.S. hold potential for the examination of plant
27                   responses over multiple levels of exposure. However, few such gradients can be found
28                   that meet the rigorous statistical requirements for comparable site characteristics such as
29                   soil type, temperature, rainfall, radiation, and aspect (Manning and Krupa. 1992):
30                   although with small plants, soil variability can be avoided by the use of plants in large
31                   pots. The use of soil monoliths transported to various locations along natural O3 gradients
32                   is another possible approach to overcome differences in soils; however, this approach is
33                   also limited to small plants.
      Draft - Do Not Cite or Quote                       9-7                                 September 2011

-------
 1                   Studies in the 1970s used the natural gradients occurring in southern California to assess
 2                   yield losses of alfalfa and tomato (Oshima et al.. 1977; Oshima et al.. 1976). A transect
 3                   study of the impact of O3 on the growth of white clover and barley in the U.K. was
 4                   confounded by differences in the concurrent gradients of SO2 and NO2 pollution
 5                   (Ashmore etal.. 1988). Studies of forest tree  species in national parks in the eastern U.S.
 6                   (Winner et al.. 1989) revealed increasing gradients of O3 and visible foliar injury with
 7                   increased elevation.

 8                   Several studies have used the San Bernardino Mountains Gradient Study in southern
 9                   California to study the effects of O3 and N deposition on forests dominated by ponderosa
10                   and Jeffrey pine (Jones and Paine. 2006; Arbaugh et al.. 2003; Grulke. 1999; Miller and
11                   Elderman. 1977). However, it is difficult to separate the effects of N and O3 in some
12                   instances in these studies (Arbaugh et al.,  2003). An O3 gradient in Wisconsin has been
13                   used to study foliar injury in a series of trembling aspen clones (Populus tremuloides)
14                   differing in O3 sensitivity (Mankovska et  al., 2005; Karnosky et al.. 1999).

15                   More recently, studies have been published that have used natural gradients to study a
16                   variety of endpoints and species. For example, Gregg et al. (2003) studied cottonwood
17                   saplings grown in an urban to rural gradient of O3 in the New York City area. The
18                   secondary nature of the reactions of O3  formation and NOX titration reactions within the
19                   city center resulted in significantly higher cumulative O3 exposures in the rural sites. The
20                   results of this gradient study were similar to those of a parallel OTC study. Also, the U.S.
21                   forest service Forest Inventory and Analysis (FIA) program uses large-scale O3 exposure
22                   patterns across the continental U.S. to study occurrences of foliar injury due to O3
23                   exposure  (Smith et al.. 2003) (Section 9.4.2). Finally, McLaughlin et al. (2007a; 2007b)
24                   used spatial and temporal O3 gradients to  study forest growth and water use  in the
25                   southern Appalachians. These studies found varying O3 exposures between years and
26                   between sites.
             9.2.6   Comparative Studies

27                   All experimental approaches used to expose plants to O3 have strengths and weaknesses.
28                   One potential weakness of laboratory, greenhouse, or field chamber studies is the
29                   potential effect of the chamber on micrometeorology. In contrast, plume, FACE and
30                   gradient systems are limited by the very small number of possible exposure levels
31                   (almost always no more than two), small replication and an inability to reduce O3 levels
32                   below ambient. In general, experiments that aim at characterizing the effect of a single
33                   variable, e.g., exposure to O3, must not only manipulate the levels of that variable, but
34                   also control potentially interacting variables and confounders, or else account for them.
      Draft - Do Not Cite or Quote                        9-8                                 September 2011

-------
 1                  However, while increasing control of environmental variables makes it easier to discern
 2                  the effect of the variable of interest, it must be balanced with the ability to extend
 3                  conclusions to natural, non-experimental settings. More naturalistic exposure systems, on
 4                  the other hand, let interacting factors vary freely, resulting in greater unexplainable
 5                  variability. The various exposure methodologies used with O3 vary in the balance each
 6                  strikes between control of environmental inputs, closeness to the natural environment,
 7                  noisiness, and ability to make general inferences.

 8                  Studies have examined the comparability of results obtained though the various exposure
 9                  methodologies. As noted in the 1996 O3 AQCD, evidence from the comparative studies
10                  of OTCs and from closed chamber and O3-exclusion exposure systems on the growth of
11                  alfalfa (Medicago sativd) by Olszyk et al. (1986) suggested that, since significant
12                  differences were found for fewer than 10% of the growth parameters measured, the
13                  responses were, in general, essentially the same regardless of exposure system used, and
14                  chamber effects did not significantly affect response. In 1988, Heagle et al. (1988)
15                  concluded: "Although chamber effects on yield are common, there are no results showing
16                  that this will result in a changed yield response to O3." A study of the effects of an
17                  enclosure examined the responses of tolerant and sensitive white clover clones (Trifolium
18                  repens) to ambient O3 in a greenhouse, open-top chamber, and ambient (no chamber)
19                  plots (Heagle etal.. 1996). For individual harvests, greenhouse O3 exposure reduced the
20                  forage weight of the sensitive clone 7 to 23% more than in OTCs. However, the response
21                  in OTCs was the same as in ambient plots. Several studies have shown very similar
22                  response of yield to O3 for plants grown in pots or in the ground, suggesting that even
23                  such a significant change in environment does not alter the proportional response to O3,
24                  providing that the plants are well watered (Heagle et al.. 1983; Heagle. 1979).

25                  A few recent studies have compared results of O3 experiments between OTCs, FACE
26                  experiments, and gradient studies. For example, a series of studies undertaken at Aspen
27                  FACE (Isebrands et al.. 2001;  Isebrands et al.. 2000) showed that O3 symptom
28                  expression was generally similar in OTCs, FACE, and  ambient O3 gradient sites, and
29                  supported the previously observed variation among trembling aspen clones using OTCs
30                  (Mankovska et al.,  2005; Karnosky et al., 1999). In the SoyFACE experiment in Illinois,
31                  soybean (Pioneer 93B15 cultivar) yield loss data from a two-year study was published
32                  (Morgan et al.. 2006). This cultivar is a recent selection and, like most modern cultivars,
33                  has been selected under an already high current O3 exposure. It was found to have
34                  average sensitivity to O3 compared to 22 other cultivars tested at SoyFACE. In this
35                  experiment, ambient hourly O3 concentrations were increased by approximately 20% and
36                  measured yields were decreased by 15% in 2002 as a result of the increased O3 exposure
37                  (Morgan et al.. 2006). To compare these results to chamber studies, Morgan et al. (2006)
3 8                  calculated the expected yield loss from a linear relationship constructed from chamber
      Draft - Do Not Cite or Quote                       9-9                                 September 2011

-------
 1                  data using seven-hour seasonal averages (Ashmore. 2002). They calculated an 8%
 2                  expected yield loss from the 2002 O3 exposure using that linear relationship. In another
 3                  study, Gregg et al. (2006. 2003) found similar O3 effects on cottonwood sapling biomass
 4                  growth along an ambient O3 gradient in the New York City area and a parallel OTC
 5                  study.

 6                  Finally, EPA conducted comparisons of exposure-response model predictions based on
 7                  OTC studies, and more recent FACE observations. These comparisons include yield of
 8                  annual crops, and biomass growth of trees. They are presented in section 9.6.3 of this
 9                  document.
          9.3    Mechanisms Governing Vegetation Response to Ozone
            9.3.1   Introduction

10                  This section focuses on the effects of O3 stress on plants and their responses to that stress
11                  on the molecular, biochemical and physiological levels. First, the pathway of O3 uptake
12                  into the leaf and the initial chemical reactions occurring in the substomatal cavity and
13                  apoplast will be described (Section 9.3.2); additionally, direct effects of O3 on the
14                  stomatal apparatus will be discussed. Once O3 has entered the substomatal cavity and
15                  apoplast, it is thought that the cell must be able to sense the presence of O3  or its
16                  breakdown products in order to initiate the rapid changes in signaling pathways and gene
17                  expression that have been measured in O3-treated plants. While it remains unclear exactly
18                  how O3 and/or its breakdown products are sensed in the apoplast, much progress has been
19                  made in examining several different mechanisms that may contribute both to sensing the
20                  presence of O3 and its breakdown products, and also initiating a signal transduction
21                  cascade, which will be described in Section 9.3.3.1. The next section focuses on changes
22                  in gene and protein expression measured in plants exposed to O3, with particular
23                  emphasis on results from transcriptome (all RNA molecules produced in a cell) and
24                  proteome (all proteins produced in a cell) analyses (Section 9.3.3.2). Subsequently, the
25                  role of phytohormones such as salicylic acid (SA), ethylene (ET), jasmonic acid (JA), and
26                  abscisic acid (ABA) and their interactions in both signal transduction processes and in
27                  determining plant response to O3 is discussed in Section 9.3.3.3. After O3 uptake and
28                  sensing, some plants can respond to the oxidative stress with detoxification to minimize
29                  damage. These mechanisms of detoxification, with particular emphasis on antioxidant
30                  enzymes and metabolites, are reviewed in Section 9.3.4. The next section focuses on
31                  changes in primary and secondary metabolism in plants exposed to O3, looking at
32                  photosynthesis, respiration and several secondary metabolites, some of which may also
      Draft - Do Not Cite or Quote                      9-10                                September 2011

-------
 1                   act as antioxidants and protect the plant from oxidative stress (Section 9.3.5). For many
 2                   of these topics, information from the 2006 O3 AQCD has been summarized, as this
 3                   information is still valid and supported by more recent findings. For other topics, such as
 4                   genomics and proteomics, which have arisen due to the availability of new technologies,
 5                   the information is based solely on new publications with no reference to the 2006 O3
 6                   AQCD.

 7                   As Section 9.3 focuses on mechanisms underlying effects of O3 on plants and their
 8                   response to it, the conditions that are used to study these mechanisms do not always
 9                   reflect conditions that a plant may be exposed to in an agricultural setting or natural
10                   ecosystem. The goal of many of these studies is to generate an O3 effect in a relatively
11                   short period of time and not always to simulate ambient O3 exposures. Therefore, plants
12                   are often exposed to unrealistically high O3  concentrations for several hours or days
13                   (acute exposure), and only in some cases to  ambient or slightly elevated O3
14                   concentrations for longer time periods (chronic exposure). Additionally, the plant species
15                   utilized in these studies are often not agriculturally important or commonly found as part
16                   of natural ecosystems. Model organisms such as Arabidopsis thaliana are used frequently
17                   as they are easy to work with, and mutants or transgenic plants are easy to develop or
18                   have already been developed. Furthermore,  the Arabidopsis genome has been sequenced,
19                   and much is known about the molecular basis of many biochemical and cellular
20                   processes.

21                   Many of the studies described in this section focus on changes in the expression  of genes
22                   in O3-treated plants. Changes in gene expression (i.e., either up- or down-regulation of
23                   gene expression) do not always translate into changes in protein quantity and/or  activity,
24                   as there are many levels of post-transcriptional and post-translational modifications
25                   which impact protein quantity and activity. Many studies do not evaluate whether the
26                   observed changes in gene expression lead to changes at the protein level and, therefore, it
27                   is not always clear how relevant the changes in gene expression are in determining plant
28                   response to O3. However, with the advent of proteomics, some very recent studies  have
29                   evaluated changes in protein expression for  large numbers of proteins in O3 treated
30                   plants,  and the findings from these studies support the previous results regarding changes
31                   in gene expression studies as a result of O3 exposure. The next step in the process is to
32                   determine  the implications of the measured  changes occurring at the cellular level to
33                   whole plants and ecosystems, which is an important topic of study which has not been
34                   widely addressed.

35                   The most significant new body of research since the 2006 O3 AQCD is on the
36                   understanding of molecular mechanisms underlying how plants are affected by O3; a
37                   significant number of recent studies reviewed here focus on changes in gene expression
      Draft - Do Not Cite or Quote                       9-11                                 September 2011

-------
 1                   in plants exposed to elevated O3. Conclusions from the 2006 O3 AQCD have been
 2                   supported by these new studies, and the advent of new technologies has allowed for a
 3                   more comprehensive understanding of the mechanisms governing how plants are affected
 4                   byO3.

 5                   In summary, these new studies have increased knowledge of the molecular, biochemical
 6                   and cellular mechanisms occurring in plants in response to O3 by often using artificial
 7                   exposure conditions and model organisms. This information adds to the understanding of
 8                   the basic biology of how plants are affected by oxidative stress in the absence of any
 9                   other potential stressors. The results of these studies provide important insights, even
10                   though they may not always directly translate into effects observed in other plants under
11                   more realistic exposure conditions.
            9.3.2   Ozone Uptake into the Leaf

12                   Appendix AX9.2.3 of the 2006 O3 AQCD clearly described the process by which O3
13                   enters plant leaves through open stomata (U.S. EPA. 2006b). This information continues
14                   to be valid and is only summarized here.

15                   Stomata provide the principal pathway for O3 to enter and affect plants (Massman and
16                   Grantz. 1995; Fuentes et al.,  1992; Reich. 1987; Leuning et al., 1979). Ozone moves into
17                   the leaf interior by diffusing through open stomata, and environmental conditions which
18                   promote high rates of gas exchange will favor the uptake of the pollutant by the leaf.
19                   Factors that may limit uptake include boundary layer resistance and the  size of the
20                   stomatal aperture (Figure 9-2) (U.S. EPA. 2006b). Once inside the substomatal cavity, O3
21                   is thought to  rapidly react with the aqueous apoplast to form breakdown products known
22                   as reactive oxygen species (ROS), such as hydrogen peroxide (H2O2), superoxide (O2 ),
23                   hydroxyl radicals (HO) and peroxy radicals (HO2) (Figure 9-3). Hydrogen peroxide is
24                   not only a toxic breakdown product of O3, but has been shown to function as a signaling
25                   molecule, which is activated in response to both biotic and abiotic stressors. The role of
26                   H2O2 in signaling was described in detail in the 2006 O3 AQCD. Additional organic
27                   molecules present in the apoplast or cell wall, such as those containing double bonds or
28                   sulfhydryls that are sensitive to oxidation, could also be converted to oxygenated
29                   molecules after interacting with O3 (Figure 9-4). These reactions are not only pH
30                   dependent, but are also influenced by the presence of other molecules in the apoplast
31                   (U.S. EPA. 2006b). The 2006 O3 AQCD provided a comprehensive summary of what is
32                   known about the possible interactions of O3 with other biomolecules (U.S. EPA. 2006b).
33                   It is in the apoplast that initial detoxification reactions by antioxidant metabolites and
34                   enzymes take place, and these initial reactions are critical to reduce concentrations of the
      Draft - Do Not Cite or Quote                      9-12                                September 2011

-------
 1                   oxidative breakdown products of O3; these reactions are described in more detail in
 2                   Section 9.3.4 of this document.
                     9.3.2.1    Changes in Stomatal Function

 3                   The effects of O3 exposure on stomatal conductance have been reviewed in detail in
 4                   previous O3 AQCDs. Although the nature of these effects depends upon many different
 5                   factors, including the plant species, concentration and duration of the O3 exposure, and
 6                   prevailing meteorological conditions, stomatal conductance is often negatively affected
 7                   by plant exposure to O3 CWittig etal.. 2007). Decreases in conductance have been shown
 8                   to result from declines in photosynthetic carboxylation capacity, leading to a buildup of
 9                   CO2 in the substomatal cavity and subsequent stomatal closure (Wittig et al.. 2007).
10                   However, results from the use of Arabidopsis mutants and new technologies, which allow
11                   for analysis of guard cell function in whole plants rather than in isolated guard cells or
12                   epidermal peels, suggest that O3 may also have a direct impact on stomatal guard cells,
13                   leading to alterations in stomatal conductance. The use of a new simultaneous O3
14                   exposure/gas exchange device has demonstrated that exposure of Arabidopsis ecotypes
15                   Col-0 and Ler to 150 ppb O3 resulted in a 60-70% decline in stomatal conductance within
16                   9-12 minutes of beginning the exposure. Twenty to thirty minutes later, stomatal
17                   conductance had returned to its initial value, even with continuing exposure to O3,
18                   indicating a rapid direct effect of O3  on stomatal function (Kollist et al., 2007). This
19                   transient decrease in stomatal conductance was not observed in the abscisic acid
20                   insensitive (ABI2) Arabidopsis mutant. As the ABE protein is thought to regulate the
21                   signal transduction process involved in stomatal response downstream of ROS
22                   production, the authors suggest that the transient decrease in stomatal conductance in the
23                   Col-0 and Ler ecotypes results from the biological action of ROS in transducing signals,
24                   rather than direct physical damage to guard cells by ROS (Kollist et al.. 2007). This rapid
25                   transient decrease in stomatal conductance was also not observed when exposing the
26                   Arabidopsis mutant slacl (slow anion channel-associated 1) to 200 ppb O3 (Vahisalu et
27                   al.. 2008). The SLAC1 protein was shown to be essential for guard cell slow anion
28                   channel functioning and for stomatal closure in response to O3. Based on additional
29                   studies using a variety of Arabidopsis mutants impaired in various aspects of stomatal
30                   function, Vahisalu et al. (2010) suggest that the presence of ROS in the guard cell
31                   apoplast (formed either by O3 breakdown or through ROS production from NADPH
32                   oxidase activity) leads to the activation of a signaling pathway in the  guard cells, which
33                   includes SLAC1, and results in stomatal closure.

34                   A review by McAinsh et al. (2002) discusses the role of calcium as a part of the signal
35                   transduction pathway involved in regulating stomatal responses to pollutant stress. A


      Draft - Do Not Cite or Quote                       9-13                                September 2011

-------
 1                   number of studies in this review provide some evidence that exposure to O3 increases the
 2                   cytosolic free calcium concentration ([Ca2+]cyt) in guard cells, which may result in an
 3                   inhibition of the plasma membrane inward-rectifying K+ channels in guard cells, which
 4                   allow for the K+ uptake needed for stomatal opening (McAinsh et al.. 2002; Torsethaugen
 5                   et al., 1999). This would compromise the ability of the stomata to respond to various
 6                   stimuli, including light, CO2 concentration and  drought. Pei et al. (2000) reported that the
 7                   presence of H2O2 activated Ca2+ -permeable channels, which mediate increases in
 8                   [Ca2+]cyt in guard cell plasma membranes of Arabidopsis. They also determined that
 9                   abscisic acid (ABA) induced H2O2 production in guard cells, leading to ABA-induced
10                   stomatal closure via activation of the membrane Ca2+ channels. Therefore, it is possible
11                   that H2O2, a byproduct of O3 breakdown in the  apoplast, could disrupt the  Ca2+-ABA
12                   signaling pathway that is involved in regulating stomatal responses (McAinsh et al..
13                   2002). The studies described here provide some evidence to suggest that O3 and its
14                   breakdown products can directly affect stomatal functioning by impacting the signal
15                   transduction pathways which regulate guard cells. Stomatal sluggishness has been
16                   described as a delay in stomatal response to changing environmental conditions in
17                   sensitive species exposed to higher concentrations and/or longer-term O3 exposures
18                   (Paoletti and Grulke. 2010. 2005; McAinsh et al.. 2002). It is possible that the signaling
19                   pathways described above could be involved in  mediating this  stomatal sluggishness in
20                   some plant species under certain O3 exposure conditions (Paoletti and Grulke. 2005;
21                   McAinsh et al.. 2002).
      Draft - Do Not Cite or Quote                       9-14                                September 2011

-------
                                 Light
                Cuticle
           Epidermis

             Pallisade
             Mesophyll

              Spongy
             Mesophyll

           Epidermis
                Cuticle
                                           1?
mrnrn
Vascular
System
                    C0=[C02]-
Figure 9-2   The microarchitecture of a dicot leaf. While details among species
            vary, the general overview remains the same. Light that drives
            photosynthesis generally falls upon the upper (adaxial) leaf
            surface. Carbon dioxide and ozone enter through the stomata on
            the lower (abaxial) leaf surface, while water vapor exits through the
            stomata (transpiration).
                       a.
                                      Stiperoxide
                       Hydroxyt
                       Radical
                       b.
                          H2°2
                                     HO-    H2O2
                                          Pe/oxy^
                                          Radical
Figure 9-3   Possible reactions of ozone within water, (a) Ozone reacts at the
            double bonds to form carbonyl groups, (b) Under certain
            circumstances, peroxides are generated.
Draft - Do Not Cite or Quote
             9-15
      September 2011

-------
        a.
2.
O
OH-
                                    Crigee
                                    Mechanism
                         H2C = CH2
                         H2C = CH2
0   O
 \   I
                                              OH
                                               \
                                                      H
                                                     HC=O

                                                      0
                                                      ii
                                                     HC-OH
NO
                         H2C=CH2
                                              ONO2
        b.
 Source: Adapted from Mudd (1996).
                                          CH(OH)CH O2H
                                          CH(OH)CH 02H
                                       OH
                                         \
                            0=C         CH(OH)CH02H

                                 CHO , CHO
                              Further Oxidation
Figure 9-4    The Crigee mechanism of ozone attack of a double bond,  (a) The
             typical Crigee mechanism is shown in which several reactions
             paths from the initial product is shown, (b) Typical reaction of
             ascorbic acid with ozone.
Draft - Do Not Cite or Quote
                     9-16
                                                        September 2011

-------
             9.3.3   Cellular to Systemic Responses
                     9.3.3.1    Ozone Sensing and Signal Transduction

 1                   New technologies allowing for large-scale analysis of oxidative stress-induced changes in
 2                   gene expression have facilitated the study of signal transduction processes associated
 3                   with the perception and integration of responses to the stress. Many of these studies have
 4                   been conducted using Arabidopsis or tobacco plants, for which a variety of mutants are
 5                   available and/or which can be easily genetically modified to generate either loss-of-
 6                   function or over-expressing genotypes. Several comprehensive review articles provide an
 7                   overview of what is known of O3-induced signal transduction processes and how they
 8                   may help to explain differential sensitivity of plants to the pollutant (Ludwikow and
 9                   Sadowski.  2008; Baier et al.. 2005; Kangasjarvi et al.. 2005). Additionally, analysis of
10                   several studies of transcriptome changes has also allowed for the compilation of these
11                   data to determine an initial time-course for O3-induced activation of various signaling
12                   compounds (Kangasjarvi et al.. 2005).

13                   A number of different mechanisms for plant sensing of O3 have been proposed; however,
14                   there is still much that is not known about this process. Some of the earliest events that
15                   occur in plants exposed to O3 have been described in the guard cells of stomata. Reactive
16                   oxygen species were observed in the chloroplasts of guard cells in the O3 tolerant Col-0
17                   Arabidopsis thaliana ecotype plants within 5 minutes of plant exposure to 350 ppb O3
18                   (Joo et al..  2005). Reactive oxygen species from the breakdown of O3 in the apoplast are
19                   believed to activate GTPases (G-proteins), which, in turn, activate several intracellular
20                   sources of ROS, including ROS derived from the chloroplasts. G-proteins are also
21                   believed to play a role in activating membrane-bound NADPH oxidases to produce ROS
22                   and, as a result, propagate the oxidative burst to neighboring cells (Joo et al.. 2005).
23                   Therefore,  G-proteins are recognized as important molecules involved in plant responses
24                   to O3 and may play a role in perceiving ROS from the breakdown of O3 in the apoplast
25                   (Kangasjarvi et al..  2005; Booker etal.,  2004b).

26                   A change in the redox state of the plant and the oxidation of sensitive molecules in itself
27                   may represent a means of perception and signaling of oxidative stress in plants.
28                   Disulfide-thiol conversions in proteins and the redox state of the glutathione pool may be
29                   important components of redox sensing and  signal transduction (Foyer and Noctor.
30                   2005a. b).

31                   Calcium (Ca2+) has also been implicated in the transduction of signals to the nucleus in
32                   response to oxidative stress. The influx of Ca2+ from the apoplast  into the cell occurs
33                   early during plant exposure to O3, and it is thought to play a role in  regulating the activity
      Draft - Do Not Cite or Quote                       9-17                                 September 2011

-------
 1                   of protein kinases, which are discussed below (Baier et al.. 2005; Hamel et al.. 2005).
 2                   Calcium channel blockers inhibited O3-induced activation of protein kinases in tobacco
 3                   suspension cells exposed to 500 ppb O3 for 10 minutes, indicating that the opening of
 4                   Ca2+ channels is an important upstream signaling event or that the as yet unknown
 5                   upstream process has a requirement for Ca2+ (Samuel et al.. 2000).

 6                   Further transmission of information regarding the presence of ROS to the nucleus t
 7                   involves mitogen-activated protein kinases (MAPK), which phosphorylate proteins and
 8                   activate various cellular responses (Hamel et al.. 2005). Mitogen-activated protein
 9                   kinases are induced in several different plant species in response to O3 exposure,
10                   including tobacco (Samuel et al.. 2005), Arabidopsis (Ludwikow et al.. 2004), the shrub
11                   Phillyrea latifolia (Paolacci et al.. 2007) and poplar (Hamel et al.. 2005). Disruption of
12                   these signal transduction pathways by over-expressing or suppressing MAP kinase
13                   activity in different Arabidopsis and tobacco lines resulted in increased plant sensitivity
14                   to O3 (Miles et al., 2005; Samuel and Ellis, 2002). Additionally, greater O3 tolerance of
15                   several Arabidopsis ecotypes was correlated with greater up-regulation of MAP kinase
16                   signaling pathways upon O3 exposure than in more sensitive Arabidopsis ecotypes (Li et
17                   al.. 2006b; Mahalingam et al.. 2006; Overmyer et al.. 2005). indicating that determination
18                   of plant sensitivity and plant response to O3  may, in part, be determined not only by
19                   whether these pathways are turned on, but also by the magnitude of the signals moving
20                   through these communication channels.

21                   In conclusion, experimental evidence suggests that there are likely several different
22                   mechanisms by which the plant senses the presence of O3 or its breakdown products.
23                   These mechanisms may vary by species or developmental stage of the plant, or may  co-
24                   exist and be activated by different exposure conditions. Calcium and protein kinases are
25                   likely involved in relaying information about the presence of the stressor to the nucleus
26                   and other cellular compartments as a first step in determining whether and how the plant
27                   will respond to the stress.
                     9.3.3.2    Gene and Protein Expression Changes in Response to
                                Ozone

28                   The advent of DNA microarray technology has allowed for the study of gene expression
29                   in cells on a large scale. Rather than assessing changes in gene expression of individual
30                   genes, DNA microarrays facilitate the evaluation of entire transcriptomes, providing a
31                   comprehensive picture of simultaneous alterations in gene expression. In addition, these
32                   studies have provided more insight into the complex interactions between molecules, how
33                   those interactions lead to the communication of information in the cell (or between
      Draft - Do Not Cite or Quote                       9-18                                 September 2011

-------
 1                   neighboring cells), and which role these interactions play in determining tolerance or
 2                   sensitivity and how a plant may respond to stresses such as O3 (Ludwikow and
 3                   Sadowski. 2008). Transcriptome analysis of O3-treated plants has been performed in
 4                   several species, including Arabidopsis thaliana (Li et al.. 2006b: Tosti et al.. 2006;
 5                   Heidenreich et al., 2005;  Mahalingam et al., 2005; Tamaoki et al., 2003). pepper
 6                   (Capsicum annuum) (Lee and Yun. 2006). clover (Medicago truncatuld) (Puckette et al..
 7                   2008). Phillyrea latifolla (Paolacci et al.. 2007). poplar (Street et al.. 2011). and European
 8                   beech (Fagus sylvatica) (Olbrich etal.. 2010: Olbrich et al.. 2009; Olbrich et al.. 2005).
 9                   In some cases, researchers compared transcriptomes of two or more cultivars, ecotypes or
10                   mutants that differed in their sensitivity to O3 (Puckette et al.. 2008; Rizzo et al.. 2007;
11                   Lee and Yun. 2006; Li et al.. 2006b; Tamaoki et al.. 2003). Species, O3  exposure
12                   conditions (concentration, duration of exposure) and sampling times varied significantly
13                   in these studies. However, functional classification of the genes  that were either up- or
14                   down-regulated by plant  exposure to O3 exhibited common trends. Genes involved in
15                   plant defense, signaling and those associated with the synthesis of plant hormones and
16                   secondary metabolism were generally up-regulated, while those related to photosynthesis
17                   and general metabolism were typically down-regulated in O3-treated plants (Puckette et
18                   al.. 2008; Lee and Yun. 2006; Li et al.. 2006b; Tosti et al.. 2006; Olbrich et al.. 2005;
19                   Tamaoki etal.. 2003).

20                   Analysis of the transcriptome has been used to evaluate differences in gene expression
21                   between O3 sensitive and tolerant plants. In pepper, 67% of the  180 genes studied that
22                   were affected by O3  were differentially regulated in the sensitive and tolerant cultivars.
23                   At both 0 hours and 48 hours after a 3-day exposure at 150 ppb, O3 responsive genes
24                   were either up- or down-regulated more markedly in the sensitive than in the tolerant
25                   cultivar (Lee and Yun. 2006). Transcriptome analysis also revealed differences in timing
26                   and magnitude of changes in gene expression between sensitive  and tolerant clovers.
27                   Acute exposure (300 ppb O3 for 6 hours) led to the production of an oxidative burst in
28                   both clovers (Puckette et al.. 2008). However, the sensitive Jemalong cultivar exhibited a
29                   sustained ROS burst and  a concomitant down-regulation of defense response genes at
30                   12 hours after the onset of exposure, while the tolerant JE 154 accession showed much
31                   more rapid and large-scale transcriptome changes than the Jemalong cultivar (Puckette et
32                   al.. 2008).

33                   Arabidopsis ecotypes WS and  Col-0 were exposed to 1.2 x ambient O3 concentrations for
34                   8-12 days at the Soy FACE site (Li et al.. 2006b).  The sensitive WS ecotype showed a far
35                   greater number of changes in gene expression in response to this low-level O3 exposure
36                   than the tolerant Col-0 ecotype. In a different study, exposure of the WS ecotype to
37                   300 ppb O3 for 6 hours showed a rapid induction  of genes leading to cell death, such as
      Draft - Do Not Cite or Quote                       9-19                                September 2011

-------
 1                   proteases, and down-regulation or inactivation of cell signaling genes, demonstrating an
 2                   ineffective defense response in this O3 sensitive ecotype (Mahalingam et al.. 2006).

 3                   The temporal response of plants to O3 exposure was evaluated in the Arabidopsis Col-0
 4                   ecotype during a 6-h exposure at 350 ppb O3 and for 6 hours after the exposure was
 5                   completed. Results of this study, shown in Figure 9-5, indicate that genes associated with
 6                   signal transduction and regulation of transcription were in the class of early up-regulated
 7                   genes, while genes associated with redox homeostasis and defense/stress response were
 8                   in the class of late up-regulated genes (Mahalingam et al.. 2005).

 9                   A few studies have been conducted to evaluate  transcriptome changes in response to
10                   longer term chronic O3 exposures in woody plant species. Longer term exposures
11                   resulted in the up-regulation of genes associated with secondary metabolites, including
12                   isoprenoids, polyamines and phenylpropanoids in 2-year-old seedlings of the
13                   Mediterranean shrub Phillyrea latifolia exposed to 110 ppb O3 for 90 days (Paolacci et
14                   al.. 2007). In 3-year-old European beech saplings exposed to O3 for 20 months, with
15                   monthly average twice ambient O3 concentrations ranging from 11 to 80 ppb, O3-induced
16                   changes in gene transcription were similar to those observed for herbaceous species
17                   (Olbrich et al.. 2009). Genes encoding proteins associated with plant stress response,
18                   including ethylene biosynthesis, pathogenesis-related proteins and enzymes detoxifying
19                   ROS, were up-regulated. Some genes associated with primary metabolism, cell structure,
20                   cell division and cell growth were reduced (Olbrich et al.. 2009). In a similar study using
21                   adult European beech trees, it was determined that the magnitude of the transcriptional
22                   changes described above was far greater in the  saplings than in the adult trees exposed to
23                   the same O3 concentrations for the same time period (Olbrich etal.. 2010).

24                   The results from transcriptome studies described above have been substantiated by results
25                   from proteome analysis in rice, poplar, European beech, wheat, and soybean. Exposure of
26                   soybean to 120 ppb O3 for 12-h/day for 3 days  in growth  chambers resulted in decreases
27                   in the quantity of proteins associated with photosynthesis, while proteins involved with
28                   antioxidant defense and carbon metabolism increased (Ahsan etal.. 2010). Young poplar
29                   plants exposed to 120 ppb O3 in a growth chamber for 35 days also showed significant
30                   changes in proteins involved in carbon metabolism (Bohler et al.. 2007). Declines in
31                   enzymes associated with carbon fixation, the Calvin cycle and photosystem II were
32                   measured, while ascorbate peroxidase and enzymes associated with glucose catabolism
33                   increased in abundance. In another study to determine the impacts of O3 on both
34                   developing and fully expanded poplar leaves, young poplars were exposed to  120 ppb O3
35                   for 13-h per day for up to 28 days (Bohler et al.. 2010). Impacts  on protein quantity only
36                   occurred after the plants had been exposed to O3 for 14 days, and at this point in time,
37                   several Calvin cycle enzymes were reduced  in quantity, while the effects on the light
      Draft - Do Not Cite or Quote                       9-20                                September 2011

-------
 1                   reactions appeared later, at 21 days after beginning treatment. Some of the antioxidant
 2                   enzymes increased in abundance with O3 treatment, while others (ascorbate peroxidase)
 3                   did not.  In relationship to leaf expansion, it was shown that O3 did not affect protein
 4                   quantity until leaves had reached full expansion, after about 7 days (Bohler et al.. 2010).

 5                   Two-week-old rice seedlings exposed to varying levels of O3 (4, 40, 80, 120 ppb) in a
 6                   growth chamber for 9 days showed reductions in quantities of proteins associated with
 7                   photosynthesis and energy metabolism, and increases in some antioxidant and defense
 8                   related proteins (Fenget al., 2008a). A subsequent study of O3-treated rice seedlings
 9                   (exposed to 200 ppb O3 for 24-h) focusing on the integration of transcriptomics and
10                   proteomics, supported and further enhanced these results (Cho et al., 2008). The authors
11                   found that of the 22,000 genes analyzed from the rice genome, 1,535 were differentially
12                   regulated by O3. Those differentially regulated genes were functionally categorized as
13                   transcription factors, MAPK cascades, those encoding for enzymes involved in the
14                   synthesis of jasmonic acid (JA),ethylene (ET), shikimate, tryptophan and lignin, and
15                   those involved in glycolysis, the citric acid cycle, oxidative respiration and
16                   photosynthesis. The authors determined that the proteome and metabolome (all  small
17                   molecule metabolites in a cell) analysis supported the results of the transcriptome
18                   changes described above (Cho et al.. 2008). This type of study, which ties together results
19                   from changes in gene expression, protein quantity and activity, and metabolite levels,
20                   provides the most complete picture of the molecular and biochemical changes occurring
21                   in plants exposed to a stressor such as O3.

22                   Sarkar et al. (2010) compared proteome s of two cultivars of wheat grown in OTCs at
23                   several O3 concentrations, including filtered air, ambient O3 (mean concentration
24                   47 ppb), ambient +10 ppb and ambient + 20 ppb for 5-h/day for 50 days. Declines in the
25                   rate of photosynthesis and stomatal conductance were related to decreases in proteins
26                   involved in carbon fixation and electron transport and increased proteolysis of
27                   photosynthetic proteins such as the large subunit of ribulose-l,6-bisphosphate
28                   carboxylase/oxygenase (Rubisco). Enzymes that take part in energy metabolism, such as
29                   ATP synthesis, were also down-regulated, while defense/stress related proteins were up-
30                   regulated in O3 treated plants. In comparing the two wheat cultivars, Sarkar et al. (2010)
31                   found that while the qualitative changes in protein expression between the two cultivars
32                   were similar, the magnitude of these changes differed between the sensitive and tolerant
33                   wheat cultivars. Greater foliar injury and a smaller decline in stomatal conductance was
34                   observed in the sensitive cultivar as compared to the more tolerant cultivar, along with
3 5                   greater losses in photosynthetic enzymes and higher quantities of antioxidant enzymes.
36                   Results from a three year exposure of European beech saplings to elevated O3 (AOT 40
37                   value was  52.6 ul 1-1-h for 2006 when trees were sampled) supported the results from the
38                   short-term exposure studies described above (Kerner et al., 2011). The O3 treatment of
      Draft - Do Not Cite or Quote                       9-21                                 September 2011

-------
1                   the saplings resulted in reductions in enzymes associated with the Calvin cycle, which
2                   could lead to reduced carbon fixation. Enzymes associated with carbon
3                   metabolism/catabolism were increased, and quantities of starch and sucrose were reduced
4                   in response to the O3 treatment in these trees, indicating a potential impact of O3 on
5                   overall carbon metabolism in long-term exposure conditions (Kerner et al., 2011).
     Draft - Do Not Cite or Quote                        9-22                                 September 2011

-------
                           (A)
Signaling
Transcription
                                               Redox homeostasis
                                               Dcfcnsc/slrcss response
                  PR proteins
                   t
                                                                                      12 hr
                                                                                      12 hr
                                                      Photosynthesis
       Source: Used with permission from Springer (Mahalingam et al.. 2005).
       (A) Temporal profile of the oxidative stress response to ozone. The biphasic ozone-induced oxidative burst is represented in black,
     with the ROS control measurements shown as a broken line. Average transcript profiles are shown for early up-regulated genes
     (yellow, peaks at 0.5-1 hours), and the 3 hours (blue), 4.5 hours (red) and 9-12 hours (green) late up-regulated genes and for the
     down-regulated genes coding for photosynthesis proteins (brown). (B) Diagrammatic representation of redox regulation of the
     oxidative stress response.

     Figure 9-5     Composite diagram of major themes in the temporal evolution of
1                      the  genetic response  to  OZOne Stress. All of these studies describe common
2                    trends for changes in gene and protein expression which occur in a variety of plant
3                    species exposed to O3. While  genes associated with carbon assimilation and general
4                    metabolism are typically down-regulated, genes associated with signaling, catabolism,
5                    and defense are up-regulated.  The magnitude of these changes in gene and protein
6                    expression appears to be related to plant species, age and their sensitivity or tolerance to
7                    03.
     Draft - Do Not Cite or Quote
9-23
                                                            September 2011

-------
                    9.3.3.3   Role of Phytohormones in Plant Response to Ozone

 1                  Many studies of O3 effects on plants have analyzed the importance of plant hormones
 2                  such as SA, ET and JA in determining plant response to O3; some of the roles of these
 3                  hormones were described in the 2006 O3 AQCD. Transcriptome analysis and the use of a
 4                  variety of mutants have allowed for further elucidation of the complex interactions
 5                  between SA, ET, JA and the role of abscisic acid (ABA) in mediating plant response to
 6                  O3 (Ludwikow and Sadowski. 2008). In addition to their roles in signaling pathways,
 7                  phytohormones also appear to regulate, and be regulated by, the MAPK signaling
 8                  cascades described previously. Most evidence suggests that while ET and SA are needed
 9                  to develop O3-induced leaf lesions, JA acts antagonistically to SA and ET to limit the
10                  lesions (Figure 9-6) (Kangasjarvi et al.. 2005).

11                  The rapid production of ET in O3 treated plants has been described in many plant species
12                  and has been further characterized through the use of a variety of mutants that either
13                  over-produce or are insensitive to ET.  Production of stress ET in O3-treated plants, which
14                  is thought to be part of a wounding response, was found to be correlated to the degree of
15                  injury development in leaves (U.S. EPA. 2006b). More recent studies have supported
16                  these conclusions and have also focused on the interactions occurring between several
17                  oxidative-stress induced phytohormones. Yoshida et al. (2009) determined that ET likely
18                  amplifies the oxidative signal generated by ROS, thereby promoting lesion formation. By
19                  analyzing the O3-induced transcriptome of several Arabidopsis mutants of the Col-0
20                  ecotype, Tamaoki et al. (2003) determined that at 12 hours after initiating the O3
21                  exposure (200 ppb for 12 hours), the ET and JA signaling pathways were the main
22                  pathways used to activate plant defense responses, with a lesser role for SA. The authors
23                  also demonstrated that low levels of ET production could stimulate the expression of
24                  defense genes, rather than promoting cell death which occurs when ET production is
25                  high. Tosti et al. (2006) supported these findings by showing that plant exposure to O3
26                  not only results in activation of the biosynthetic pathways of ET, JA and SA, but also
27                  increases the expression of genes related to the signal transduction pathways of these
28                  phytohormones in O3-treated Arabidopsis plants (300 ppb O3 for 6 hours). Conversely, in
29                  the O3 sensitive Ws  ecotype, its sensitivity may, in part, be due to intrinsically high ET
30                  levels leading to SA accumulation, and the high ET and SA may act to repress JA-
31                  associated genes, which would serve to inhibit the spread of lesions (Mahalingam et al..
32                  2006). Ogawa et al. (2005) found that  increases in SA in O3-treated plants leads to the
33                  formation of leaf lesions in tobacco plants exposed to 200 ppb O3 for 6 hours.
34                  Furthermore, in transgenic tobacco plants with reduced levels of ET production in
35                  response to O3 exposure, several genes encoding for enzymes in the biosynthetic pathway
36                  of SA were suppressed, suggesting that SA levels are, in part, controlled by ET in the
37                  presence of O3.

      Draft - Do Not Cite or Quote                       9-24                               September 2011

-------
 1                  Exposure of the Arabidopsis mutant rcdl to acute doses of O3 (250 ppb O3 for 8-h/day
 2                  for 3 days) resulted in programmed cell death (PCD) and the formation of leaf lesions
 3                  (Overmyer et al., 2000). They determined that the observed induction of ET synthesis
 4                  promotes cell death, and that ET perception and signaling are required for the
 5                  accumulation of superoxide, which leads to cell death and propagation of lesions. .
 6                  Jasmonic acid, conversely, contains the spread of leaf lesions (Overmyer et al.. 2000).
 7                  Transcriptome analysis of several Arabidopsis mutants, which are insensitive to SA, ET
 8                  and JA, exposed to 12-h of 200 ppb O3 showed that approximately 78 of the up-regulated
 9                  genes measured in this study were controlled by ET and JA signaling pathways, while SA
10                  signaling pathways were suggested to antagonize ET and JA pathways (Tamaoki et al..
11                  2003). In a subsequent transcriptome study on the Col-0 ecotype exposed to 150 ppb O3
12                  for 48-h, JA and ET synthesis was down-regulated, while SA was up-regulated in O3-
13                  treated plants. In cotton plants exposed to a range of O3 concentrations (0-120 ppb) and
14                  methyl jasmonate (MeJA), Grantz et  al. (2010a) determined that exogenous applications
15                  of MeJA did not protect plants from chronic O 3 exposure.

16                  Abscisic acid has been investigated for its role in regulating stomatal aperture and also
17                  for its  contribution to signaling pathways in the plant. The role of ABA and the
18                  interaction between ABA and H2O2 in O3-induced stomatal closure was described in the
19                  2006 O3 AQCD. More recently, it was determined that synthesis of ABA was induced in
20                  O3-treated Arabidopsis plants (250-350 ppb O3 for 6 hours), with a more pronounced
21                  induction in the O3 sensitive rcd3 mutant as compared to the wildtype Col-0 (Overmyer
22                  et al.. 2008). The rcd3 mutant also exhibited a lack of O3-induced stomatal closure, and
23                  the RCD3 protein has been shown to be required for slow anion channels (Overmyer et
24                  al.. 2008) (see Section 9.3.4.1). Ludwikow et al. (2009) used Arabidopsis ABIltd
25                  mutants, in which a key negative regulator of ABA action (abscisic acid insensitive 1
26                  protein phosphatase 2C) has been knocked out, to examine O3 responsive genes in this
27                  mutant compared to the Arabidopsis  Col-0. Results of this study indicate a role for ABU
28                  in negatively regulating the synthesis of both ABA and ET in O3-treated plants (350 ppb
29                  O3 for 9 hours). Additionally, ABU may stimulate JA-related gene expression, providing
30                  evidence for an antagonistic interaction between ABA and JA signaling pathways
31                  (Ludwikow etal.. 2009).

32                  Nitric  oxide (NO) has also been shown to play a role in regulating gene expression in
33                  plants  in response to O3 exposure. However, little is known to date about NO and its role
34                  in the complex interactions of molecules in response to O3.  Exposure of tobacco to O3
35                  (150 ppb for 5 hours) stimulated NO  and NO-dependent ET production, while NO
36                  production itself did not depend on the presence of ET (Ederli etal.. 2006). Analysis of
37                  O3-treated Arabidopsis indicated the  possibility of a dual role for NO in the initiation of
38                  cell death and later lesion containment (Ahlfors et al., 2009).
      Draft - Do Not Cite or Quote                      9-25                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
While much work remains to be done to better elucidate how plants sense O3, what
determines their sensitivity to the pollutant and how they might respond to it, it is clear
that the mechanism for O3 sensing and signal transduction is very complex. Many of the
phytohormones and other signaling molecules thought to be involved in these processes
are interactive and depend upon a variety of other factors, which could be either internal
or external to the plant. This results in a highly dynamic and complex system, capable of
resulting in a spectrum of plant sensitivity to oxidative stress and generating a variety of
plant responses to that stress.
                                                     ozone
                                                      Cell
                                                     death
        Source: Used with permission from Blackwell Publishing Ltd. (Kangasiarvi et al.. 2005).
        Ozone-derived radicals induce endogenous ROS production (1) which results in salicylic acid (SA) accumulation and programmed
      cell death; (2) Cell death triggers ethylene (ET) production, which is required for the continuing ROS production responsible for the
      propagation of cell death; (3) Jasmonates counteract the progression of the cycle by antagonizing the cell death promoting function
      of SA and ET; (4) Abscisic acid (ABA) antagonizes ET function in many situations and might also have this role in ozone-induced
      cell death; (5) Mutually antagonistic interactions between ET, SAand jasmonic acid (JA) are indicated with red bars.

      Figure 9-6      The oxidative cell  death cycle. Detoxification
 9
10
11
12
13
14
15
9.3.4.1     Overview of Ozone-Induced Defense Mechanisms

Plants are exposed to an oxidizing environment on a continual basis, and many reactions
that are part of the basic metabolic processes, such as photosynthesis and respiration,
generate ROS. As a result, there is an extensive and complex mechanism in place to
detoxify these oxidizing radicals, including both enzymes and metabolites, which are
located in several locations in the cell and also  in the apoplast of the cell. As O3 enters
the leaf through open stomata, the first point of contact of O3 with the plant is likely in
the apoplast, where it breaks down to form oxidizing radicals such as H2O2, O2, HO- and
      Draft - Do Not Cite or Quote
                                 9-26
September 2011

-------
 1                   HO 2. Another source of oxidizing radicals is an oxidative burst, generated by a
 2                   membrane-bound NADPH oxidase enzyme, which is recognized as an integral
 3                   component of the plant's defense system against pathogens (Schraudner et al., 1998).
 4                   Antioxidant metabolites and enzymes located in the apoplast are thought to form a first
 5                   line of defense by detoxifying O3 and/or the ROS that are formed as breakdown products
 6                   of O3 (Section 9.3.2.). However, even with the presence of several antioxidants,
 7                   including ascorbate, the redox buffering capacity of the apoplast is far less than that of
 8                   the cytoplasm, as it lacks the regeneration systems necessary to retain a reduced pool of
 9                   antioxidants (Foyer and Noctor. 2005b).

10                   Redox homeostasis is regulated by the presence of a pool of antioxidants, which are
11                   typically found in a reduced state and detoxify ROS produced by oxidases or electron
12                   transport components. As ROS increase due to environmental stress such as O3, it is
13                   unclear whether the antioxidant pool can maintain its reduced state (Foyer and Noctor.
14                   2005b). As such, not only the quantity and types of antioxidant enzymes and metabolites
15                   present, but also the cellular ability to regenerate those antioxidants are important
16                   considerations in mechanisms of plant tolerance to oxidative  stress (Dizengremel et al.,
17                   2008). Molecules such as glutathione (GSH), thioredoxins and NADPH play very
18                   important roles in this regeneration process; additionally,  it has been hypothesized that
19                   alterations in carbon metabolism would be necessary to supply the needed reducing
20                   power for antioxidant regeneration (Dizengremel et al.. 2008).
                     9.3.4.2    Role of Antioxidants in Plant Defense Responses

21                   Ascorbate has been the focus of many different studies as an antioxidant metabolite that
22                   protects plants from exposure to O3. It is found in several cellular locations, including the
23                   chloroplast, the cytosol and the apoplast (Noctor and Foyer. 1998). Ascorbate is
24                   synthesized in the cell and transported to the apoplast. Apoplastic ascorbate can be
25                   oxidized to dehydroascorbate (DHA) with exposure to O3 and is then transported back to
26                   the cytoplasm. Here, DHA is reduced to ascorbate by the enzyme dehydroascorbate
27                   reductase (DHAR) and reduced GSH, which is part of the ascorbate-glutathione cycle
28                   (Noctor and Foyer. 1998). Many studies have focused on evaluating whether ascorbate is
29                   the primary determining factor in differential sensitivity  of plants to O3. An evaluation of
30                   several species of wildflowers in Great Smoky Mountains National Park showed a
31                   correlation between higher quantities of reduced apoplastic ascorbate and lower levels of
32                   foliar injury from O3 exposure in the field in tall milkweed plants (Asclepsias exaltata L.)
33                   (Burkey et al.. 2006; Souza et al.. 2006). Cheng et al. (2007) exposed two soybean
34                   cultivars to elevated O3 (77 ppb) and filtered air for 7-h/day for 6 days. The differences in
35                   sensitivity between the two cultivars could not be explained by differential O3 uptake or

      Draft - Do Not Cite or Quote                       9-27                                 September 2011

-------
 1                   by the fraction of reduced ascorbate present in the apoplast. However, total antioxidant
 2                   capacity of the apoplast was twofold higher in the tolerant Essex cultivar as compared to
 3                   the sensitive Forrest cultivar, indicating that there may be other compounds in the leaf
 4                   apoplast that scavenge ROS. D'Haese et al. (2005) exposed the NC-S (sensitive) and NC-
 5                   R (resistant) clones of white clover (Trifolium repens) to 60 ppb O3 for 7-h/day for
 6                   5 days in environmental chambers. Surprisingly, the NC-S clone had a higher constitutive
 7                   concentration of apoplastic ascorbate with a higher redox status than the NC-R clone.
 8                   However, the redox status of symplastic GSH was higher in NC-R, even though the
 9                   concentration of GSH was not higher than in NC-S. In addition, total symplastic
10                   antioxidative capacity was not a determining factor in differential sensitivity between
11                   these two clones. Severino et al. (2007) also examined the role of antioxidants in the
12                   differential sensitivity of the two white clover clones by growing them in the field for a
13                   growing season and then exposing them to  elevated O3 (100 ppb for 8-h/day for 10 days)
14                   in OTC at the end of the field season. The NC-R clone had greater quantities of total
15                   ascorbate and total antioxidants than the NC-S clone  at the end of the experiment. In snap
16                   bean, plants of the O3 tolerant Provider cultivar had greater total ascorbate and more
17                   ascorbate in the apoplast than the sensitive  SI56 cultivar after exposure to 71 ppb O3 for
18                   10 days in OTC (Burkey et al.. 2003). While most of the apoplastic ascorbate was in the
19                   oxidized form, the ratio of reduced ascorbate to total  ascorbate was higher in Provider
20                   than S156, indicating that Provider is better able to maintain this ratio to maximize plant
21                   protection from oxidative stress. Exposure  of two wheat varieties to ambient (7-h average
22                   44 ppb O3) and elevated (7-h average 56 ppb O3) for 60 days in open-air field conditions
23                   showed higher concentrations of reduced ascorbate in the apoplast in the tolerant Y16
24                   variety than the more sensitive Y2 variety,  however no varietal differences were seen in
25                   the decrease in reduced ascorbate quantity in response to O3 exposure (Feng etal.. 2010).
26                   There is much evidence that supports an important role for ascorbate, particularly
27                   apoplastic ascorbate, in protecting plants from oxidative stressors such as O3; however, it
28                   is also clear that there is much variation in the importance of ascorbate for different plant
29                   species  and differing exposure conditions. Additionally, the work of several authors
30                   suggests that there may be other compounds in the apoplast which have the capacity to
31                   act as antioxidants.

32                   While the quantities of antioxidant metabolites such as ascorbate are an important
33                   indicator of plant tolerance to O3, the ability of the plant to recycle oxidized ascorbate
34                   efficiently also plays a  large role in determining the plant's ability to effectively protect
35                   itself from sustained exposure to oxidative  stress. Tobacco plants over-expressing DHAR
36                   were better protected from exposure to either chronic (100 ppb O3 4-h/day for 30 days) or
37                   acute (200 ppb O3 for 2 hours) conditions than control plants and those with reduced
3 8                   expression of DHAR (Chen and Gallie. 2005). The DHAR over-expressing plants
39                   exhibited an increase in guard cell ascorbic acid, leading to a decrease in stomatal

      Draft - Do Not Cite or Quote                       9-28                                 September 2011

-------
 1                   responsiveness to O3 and an increase in stomatal conductance and O3 uptake. Despite
 2                   this, the presence of higher levels of ascorbic acid led to a lower oxidative load and a
 3                   higher level of photosynthetic activity in the DHAR over-expressing plants (Chen and
 4                   Gallie. 2005). A subsequent study with tobacco plants over-expressing DHAR confirmed
 5                   some of these results. Levels of ascorbic acid were higher in the transgenic tobacco
 6                   plants, and they exhibited greater tolerance to O3 exposure (200 ppb O3) as demonstrated
 7                   by higher photosynthetic rates in the transgenic plants as compared to the control plants
 8                   (Eltayeb et al.. 2006). Over-expression of monodehydroascorbate reductase (MDAR) in
 9                   tobacco plants also showed enhanced stress tolerance in response to O3 exposure
10                   (200 ppb O3), with higher rates of photosynthesis and higher levels of reduced ascorbic
11                   acid as compared to controls (Eltaveb et al., 2007). Results of these studies demonstrate
12                   the importance of ascorbic acid as a detoxification mechanism in some plant species, and
13                   also emphasize that the recycling of oxidized ascorbate to maintain a reduced pool of
14                   ascorbate is a factor in determining plant tolerance to oxidative stress.

15                   The roles of other antioxidant metabolites and enzymes, including GSH,  catalase (CAT),
16                   and superoxide dismutase (SOD), were comprehensively reviewed in the 2006 O3
17                   AQCD. Additional studies have supported the findings reported in that document.
18                   Superoxide dismutase (SOD) and peroxidase (POD) activities were measured in both the
19                   tolerant Bel B and sensitive Bel W3 tobacco cultivars exposed to ambient O3
20                   concentrations for 2 weeks 3 times throughout a growing season (Borowiak et al.. 2009).
21                   In this study, SOD and POD activity, including that of several different isoforms,
22                   increased in both the sensitive and tolerant tobacco cultivars with exposure to O3,
23                   however the isoenzyme composition for POD differed between the sensitive and tolerant
24                   tobacco cultivars (Borowiak et al.. 2009)  Tulip poplar (Liriodendron tulipifera) trees
25                   exposed to increasing O3 concentrations (from 100 to 300 ppb O3 during a 2-week
26                   period) showed increases in activities of SOD, ascorbate peroxidase (APX), glutathione
27                   reductase (GR), MDAR, DHAR, CAT and POD in the 2-week period, although
28                   individual enzyme activities increased at different times during the 2-week period (Ryang
29                   et al.. 2009).

30                   Longer, chronic O3  exposures in trees revealed increases in SOD and APX activity in
31                   Quercus mongolica after 45 days of plant exposure to 80 ppb O3, which were followed
32                   by declines in the activities and quantities of these enzymes after 75 days of exposure
33                   (Yanetal.. 2010). Similarly, activities of SOD, APX, DHAR, MDAR, and GR increased
34                   in Gingko biloba trees during the first 50  days of exposure to 80 ppb O3, followed by
35                   decreases in activity below control values after 50 days of exposure (He etal.. 2006).
36                   Soybean plants exposed to 70 or 100 ppb O3 for 4-h/day over the course  of a growing
37                   season showed elevated POD activity and a decrease in CAT activity at 40 and 60 days
38                   after germination (Singh et al., 2010a).
      Draft - Do Not Cite or Quote                       9-29                                September 2011

-------
 1                  Antioxidant enzymes and metabolites have been shown to play an important role in
 2                  determining plant tolerance to O3 and mediating plant responses to O3. However, there is
 3                  also some evidence to suggest that the direct reaction of ascorbate with O3 could lead to
 4                  the formation of secondary toxicants, such as peroxy compounds, which may act upon
 5                  signal transduction pathways and modulate plant response to O3 (Sandermann, 2008).
 6                  Therefore, the role of ascorbate and other antioxidants and their interaction with other
 7                  plant responses to O3, such as the activation of signal transduction pathways, is likely far
 8                  more complex than is currently understood.
            9.3.5   Effects on Primary and Secondary Metabolism
                     9.3.5.1    Light and Dark Reactions of Photosynthesis

 9                   Declines in the rate of photosynthesis and stomatal conductance in O3-treated plants have
10                   been documented for many different plant species (Booker et al., 2009; U.S. EPA. 2006b)
11                   (Wittig et al.. 2007). The 2006 O3 AQCD outlined what is known about the effects of O3
12                   on carbon assimilation, and the more recent scientific literature confirms these findings.
13                   While several measures of the light reactions of photosynthesis are sensitive to exposure
14                   to O3 (see below), photosynthetic carbon assimilation is generally considered to be more
15                   affected by pollutant exposure, resulting in an overall decline in photosynthesis (Guidi
16                   and Degl'lnnocenti. 2008; Heath. 2008; Fiscus et al.. 2005). Loss of carbon assimilation
17                   capacity has been shown to result primarily from declines in the quantity of Rubisco
18                   (Singh et al., 2009; Calatayud et al.. 2007a). Experimental evidence suggests that both
19                   decreases in Rubisco synthesis and enhanced degradation of the protein contribute to the
20                   measured reduction in its quantity (U.S. EPA, 2006b). Reduced carbon assimilation has
21                   been linked to reductions in biomass and yield (Wang et al.. 2009b; He et al..  2007;
22                   Novak et al.. 2007; Gregg et al., 2006; Keutgen et al., 2005). Recent studies evaluating
23                   O3 induced changes in the transcriptome and proteome of several different species
24                   confirm these findings. Levels of mRNA for the small subunit of Rubisco (rbcS) declined
25                   in European beech saplings exposed to 300 ppb O3 for 8-h/day for up to 26 days (Olbrich
26                   et al., 2005). Similar declines in rbcS mRNA were also measured in the beech saplings in
27                   a free air exposure system over a course of two growing seasons (Olbrich et al..  2009).
28                   Proteomics studies have also confirmed the effects of O3 on proteins involved in carbon
29                   assimilation. Reductions in quantities of the small and large subunit (rbcL) of Rubisco
30                   and Rubisco activase were measured in soybean plants exposed to 120 ppb O3 for 3 days
31                   in growth chambers (Ahsan et al.. 2010). Exposure of young poplar trees to 120 ppb O3
32                   for 35 days in exposure chambers resulted in reductions of Rubisco, Rubisco activase,
33                   and up to 24 isoforms of Calvin cycle enzymes, most of which play a role in regenerating

      Draft - Do Not Cite or Quote                      9-30                                September 2011

-------
 1                   the CO2 acceptor molecule, ribulose-1.5-bisphosphate (Bohler et al., 2007). Reductions
 2                   in protein quantity of both the small and large subunit of Rubisco were seen in wheat
 3                   plants exposed to ambient (average concentration 47.3 ppb O3) and elevated O3 (ambient
 4                   + 10 or 20 ppb O3) in open-top chambers for 5-h/day for 50 days (Sarkar et al.. 2010).
 5                   Lettuce plants exposed to 100 ppb O3 in growth chambers for 8-h/day for 3 weeks also
 6                   showed reductions in transcript and protein levels of the small and large subunits of
 7                   Rubisco and Rubisco activase (Goumenaki et al.. 2010). The reductions in carbon
 8                   assimilation have been associated with declines in both the mRNA of the small and large
 9                   subunits of Rubisco, and with reductions in Rubisco activase mRNA and protein.
10                   Additionally, the reduction in Rubisco quantity has  also been associated with the O3-
11                   induced oxidative modification of the enzyme, which is evidenced by the increases in
12                   carbonyl groups on the protein after plant exposure to O3.

13                   In addition to impacts on carbon assimilation, the deleterious effects of O3 on the
14                   photosynthetic light reactions have received more attention in recent years. Chlorophyll
15                   fluorescence provides a useful measure of changes to the photosynthetic process from
16                   exposure to oxidative stress. Decreases in the Fv/Fm ratio (a measure of the maximum
17                   efficiency of Photosystem II) in dark adapted leaves indicate a decline in the efficiency of
18                   the PSII photosystems and a concomitant increase in non-photochemical quenching
19                   (Guidi and Degl'lnnocenti. 2008; Scebbaetal.. 2006). Changes in these parameters have
20                   been correlated to differential sensitivity of plants to the pollutant. In a study to evaluate
21                   the response of 4 maple species to O3 (exposed to an 8-h avg of 51 ppb for ambient and
22                   79 ppb for elevated treatment in OTC), the 2 species which were most sensitive based on
23                   visible injury and declines in CO2 assimilation also  showed the greatest decreases in
24                   Fv/Fm in symptomatic leaves. In asymptomatic leaves, CO2 assimilation decreased
25                   significantly but there was no significant decline in  Fv/Fm (Calatavud et al., 2007a). Degl
26                   'Innocenti et al. (2007) measured significant decreases in Fv/Fm in young and
27                   symptomatic leaves of a resistant tomato genotype (line 93.1033/1) in response to O3
28                   exposure (150 ppb O3 for 3 hours in a growth chamber), but only minor decreases in
29                   asymptomatic leaves with no associated changes in  net photosynthetic rate. In the O3
30                   sensitive tomato cultivar Cuor Di Bue, the Fv/Fm ratio  did not change, while the
31                   photosynthetic rate declined significantly in asymptomatic leaves (Degl'Innocenti et al..
32                   2007). In two soybean cultivars, Fv/Fm also declined significantly with plant exposure to
33                   O3 (Singh et al., 2009).  It appears that in asymptomatic leaves, photoinhibition, as
34                   indicated by a decrease  in Fv/Fm, is not the main reason for a decline in photosynthesis.

35                   An evaluation of photosynthetic parameters of two white clover (Trifolium repens cv.
36                   Regal) clones that differ in their O3 sensitivity revealed that O3 (40-110 ppb O3 for 7-
37                   h/day  for 5 days) increased the coefficient of non-photochemical quenching (QM>) in both
38                   the resistant (NC-R) and sensitive (NC-S) clones, however q^p was significantly lower
      Draft - Do Not Cite or Quote                       9-31                                 September 2011

-------
 1                   for the sensitive clone (Crous et al., 2006). Sensitive Acer clones had a lower coefficient
 2                   of non-photochemical quenching, while exposure to O3 increased q^p in both sensitive
 3                   and tolerant clones (Calatavud et al., 2007a). While exposure to O3 also increased q^p in
 4                   tomato, there were no differences in the coefficient of photochemical quenching between
 5                   cultivars thought to be differentially sensitive to O3. (Degl'Innocenti et al., 2007). Higher
 6                   qM. as a result of exposure to O3 indicates a reduction in the proportion of absorbed light
 7                   energy being used to drive photochemistry. A lower coefficient of non-photochemical
 8                   quenching in O3 sensitive plants could indicate increased vulnerability to ROS generated
 9                   during exposure to oxidative stress (Crous et al.. 2006).

10                   Most of the research on O3 effects on photosynthesis has focused on C3 (Calvin cycle)
11                   plants because C4 (Hatch-Slack) plants have lower stomatal conductance and are,
12                   therefore, thought to be less sensitive to O3 stress. However, a few studies have been
13                   conducted to evaluate the effects of O3 on C4 photosynthesis. In older maize leaves,
14                   Leitao et al. (2007b; 2007c) found that the activity, quantity and transcript levels of both
15                   Rubisco and phosphoenolpyruvate carboxylase (PEPc) decreased as a function of rising
16                   O3 concentration. In younger maize leaves, the quantity, activity, and transcript levels of
17                   the carboxylases were either increased or unaffected in plants exposed to 40 ppb O3 for
18                   7- h/day for 28-33 days, but decreased at  80 ppb (Leitao et al.. 2007a: Leitao et al..
19                   2007b).
                     9.3.5.2    Respiration and Dark Respiration

20                   While much research emphasis regarding O3 effects on plants has focused on the
21                   negative impacts on carbon assimilation, other studies have measured impacts on
22                   catabolic pathways such as shoot respiration and photorespiration. Generally, shoot
23                   respiration has been found to increase in plants exposed to O3. Bean plants exposed to
24                   ambient (average 12-h mean 43 ppb) and twice ambient (average 12-h mean 80 ppb) O3
25                   showed increases in respiration. When mathematically partitioned, the maintenance
26                   coefficient of respiration was significantly increased in O3 treated plants, while the
27                   growth coefficient of respiration was not affected (Amthor. 1988). Loblolly pines were
28                   exposed to ambient (12-h  daily mean was 45 ppb) and twice ambient (12 hours daily
29                   mean was 86 ppb) O3 for  12-h/day for approximately seven months per year for 3 and
30                   4 years. While photosynthetic activity declined with the age of the needles and increasing
31                   O3 concentration, enzymes associated with respiration showed higher levels of activity
32                   with increasing O3 concentration (Dizengremel et al., 1994). In their review on the role of
33                   metabolic changes in plant redox status after O3 exposure, Dizengremel  et al. (2009)
34                   summarized multiple studies in which several different tree species were exposed to O3
35                   concentrations ranging from ambient to 200 ppb O3 for at least several weeks. In all

      Draft - Do Not Cite or Quote                       9-32                                September 2011

-------
 1                   cases, the activity of enzymes, including phosphofructokinase, pyruvate kinase and
 2                   fumarase, which are part of several catabolic pathways, were increased in O3 treated
 3                   plants.

 4                   Photorespiration is a light-stimulated process which consumes O2 and releases CO2.
 5                   While it has been regarded as a wasteful process, more recent evidence suggests that it
 6                   may play a role in photoprotection during photosynthesis (Bagard et al., 2008). The few
 7                   studies that have been conducted on O3 effects on photorespiration suggest that rates of
 8                   photorespiration decline concomitantly with rates of photosynthesis. Soybean plants were
 9                   exposed to ambient (daily averages 43-58 ppb) and 1.5 ambient O3 (daily averages 63-
10                   83 ppb) O3 in OTCs for 12-h/day for 4 months. Rates of photosynthesis and
11                   photorespiration and photorespiratory enzyme activity declined only at the end of the
12                   growing season and did not appear to be very sensitive to O3 exposure (Booker et al.,
13                   1997). Young hybrid poplars exposed to 120 ppb O3 for 13-h/day for 35 days in
14                   phytotron chambers showed that effects on photorespiration and photosynthesis were
15                   dependent upon the developmental stage of the leaf. While young leaves were  not
16                   impacted, reductions in photosynthesis and photorespiration were measured in fully
17                   expanded leaves (Bagard et al.. 2008).
                     9.3.5.3    Secondary Metabolism

18                   Transcriptome analysis of Arabidopsis plants has revealed modulation of several genes
19                   involved in plant secondary metabolism (Ludwikow and Sadowski, 2008). Phenylalanine
20                   ammonia lyase (PAL) has been the focus of many studies involving plant exposure to O3
21                   due to its importance in linking the phenylpropanoid pathway of plant secondary
22                   metabolism to primary metabolism in the form of the shikimate pathway. Genes encoding
23                   several enzymes of the phenylpropanoid pathway and lignin biosynthesis were up-
24                   regulated in transcriptome analysis of Arabidopsis plants (Col-0) exposed to 350 ppb O3
25                   for 6 hours, while 2 genes involved in flavonoid biosynthesis were down-regulated
26                   (Ludwikow et al., 2004). Exposure of Arabidopsis (Col-0) to lower O3 concentrations
27                   (150 ppb for 8-h/day for 2 days) resulted in the induction of 11 transcripts involved in
28                   flavonoid synthesis. In their exposure of 2-year-old Mediterranean shrub Phillyrea
29                   latifolia to 110 ppb O3 for 90 days, Paolacci et al. (2007) identified four clones that were
30                   up-regulated and corresponded to genes involved in the synthesis of secondary
31                   metabolites, such as isoprenoids, polyamines and phenylpropanoids. Up-regulation of
32                   genes involved in isoprene synthesis was also observed mMedicago trunculata exposed
33                   to 300 ppb O3 for 6 hours, while genes encoding enzymes of the flavonoid synthesis
34                   pathway were either up- or down-regulated (Puckette et al., 2008). Exposure of red clover
35                   to 1.5 x ambient O3 (average concentrations of 32.4 ppb) for up to 9 weeks in an open
      Draft - Do Not Cite or Quote                       9-33                                September 2011

-------
 1                   field exposure system resulted in increases in leaf total phenolic content. However, the
 2                   types of phenolics that were increased in response to O3 exposure differed depending
 3                   upon the developmental stage of the plant. While almost all of the 31 different phenolic
 4                   compounds measured increased in quantity initially during the exposure, after 3 weeks
 5                   the quantity of isoflavones decreased while other phenolics increased (Saviranta et al.,
 6                   2010). Exposure of beech saplings to ambient and 2 x ambient O3 concentrations over 2
 7                   growing seasons resulted in the induction of several enzymes which contribute to lignin
 8                   formation, while enzymes involved in flavonoid biosynthesis were down-regulated
 9                   (Olbrich et al., 2009). Exposure of tobacco Bel W3 to 160 ppb O3 for 5 hours showed up-
10                   regulation of almost all genes encoding for enzymes which are part of the prechorismate
11                   pathway (Janzik et al., 2005). Isoprenoids can serve as antioxidant compounds in plants
12                   exposed to oxidative stress (Paolacci et al.. 2007).

13                   The prechorismate pathway is the pathway leading to the formation of chorismate, a
14                   precursor to the formation of the aromatic amino acids tryptophan, tyrosine and
15                   phenylalanine. These amino acids are precursors for the formation of many secondary
16                   aromatic compounds, and, therefore, the prechorismate pathway represents a branch-
17                   point in the regulation of metabolites into either primary or secondary metabolism (Janzik
18                   et al.. 2005). Exposure of the O3 sensitive Bel W3 tobacco cultivar at 160 ppb for 5 hours
19                   showed an increase in transcript levels of most of the genes encoding enzymes of the
20                   prechorismate pathway. However, shikimate kinase (SK) did not show any change in
21                   transcript levels and only one of three isoforms of DAHPS (3-deoxy-D-arabino-
22                   heptulosonat-7-phosphate synthase), the first enzyme in this pathway, was induced by O3
23                   exposure (Janzik et al.,  2005). Differential induction of DAHPS isoforms was also
24                   observed in European beech after 40 days of exposure to 150-190 ppb O3. At this time
25                   point in the beech experiment, transcript levels of shikimate pathway enzymes, including
26                   SK, were generally strongly induced after an only weak initial induction after the first
27                   40 days of exposure. Both soluble and cell-wall bound phenolic metabolites showed only
28                   minimal increases in response to O3 for the duration of the exposure period (Alonso et
29                   al., 2007). Total leaf phenolics decreased with leafage in Populus nigra exposed to
30                   80 ppb O3 for 12-h/day for 14 days. Ozone increased the concentration of total leaf
31                   phenolics in newly expanded leaves, with the most significant increases occurring in
32                   compounds such as quercitin glycoside, which has a high antioxidant capacity (Fares et
33                   al., 201 Ob). While several phenylpropanoid pathway enzymes were induced in two poplar
34                   clones exposed to 60 ppb O3  for 5-h/day for  15 days, the degree of induction differed
35                   between the two clones. In the tolerant 1-214 clone, PAL activity increased nine fold in
36                   O3-treated plants as compared to controls, while there was no significant difference in
37                   PAL activity in the sensitive  Eridano clone (Di Baccio et al.. 2008).
      Draft - Do Not Cite or Quote                      9-34                                September 2011

-------
 1                   Polyamines such as putrescine, spermidine and spermine play a variety of roles in plants
 2                   and have been implicated in plant defense responses to both abiotic and biotic stresses.
 3                   They exist in both a free form and conjugated to hydroxycinnamic acids. Investigations
 4                   on the role of polyamines have found that levels of putrescine increase in response to
 5                   oxidative stress. This increase stems largely from the increase in the activity of arginine
 6                   decarboxylase (ADC), a key enzyme in the synthesis of putrescine (Groppa and
 7                   Benavides. 2008). Langebartels et al. (1991) described differences in putrescine
 8                   accumulation in O3-treated tobacco plants exposed to several O3 concentrations, ranging
 9                   from 0-400 ppb for 5-7 hours. A large and rapid increase in putrescine occurred in the
10                   tolerant Bel B cultivar and only a small increase in the sensitive Bel W3 cultivar, which
11                   occurred only after the formation of necrotic leaf lesions. Van Buuren et al. (2002)
12                   further examined the role of polyamines in these two tobacco cultivars during an acute
13                   (130 ppb O3 for 7-h in a growth chamber) exposure. They found that while free
14                   putrescine accumulated in undamaged tissue of both cultivars, conjugated putrescine
15                   predominantly accumulated in tissues undergoing cell death after plant exposure to O3
16                   (van Buuren et al.. 2002). The authors suggest that while free putrescine may not play a
17                   role in conferring tolerance in the Bel B cultivar, conjugated putrescine may play a role in
18                   O3-induced programmed cell death in Bel W3 plants.

19                   Isoprene is emitted by some plant species and represents the predominant biogenic source
20                   of hydrocarbon emissions in the atmosphere (Guenther et al.. 2006). In the atmosphere,
21                   the oxidation of isoprene by hydroxyl radicals can enhance O3 formation in the presence
22                   of NOX, thereby impacting the O3 concentration that plants are exposed to. While
23                   isoprene emission varies widely between species, it has been proposed to stabilize
24                   membranes and provide those plant species that produce it with a mechanism of
25                   thermotolerance (Sharkey et al.. 2008). It has also been suggested that isoprene may act
26                   as an antioxidant compound to scavenge O3 (Loreto and Velikova. 2001). Recent studies
27                   using a variety of plant species have shown conflicting results in trying to understand the
28                   effects of O3 on isoprene emission. Exposure to acute doses of O3 (300 ppb for 3-h) in
29                   detached leaves ofPhmgmites australis resulted in stimulation of isoprene emissions
30                   (Velikova et al.. 2005). Similar increases in isoprene emissions were measured in
31                   Populus nigra after exposure to 100 ppb O3 for 5 days continuously (Fares etal., 2008).
32                   Isoprene emission in attached leaves of Populus alba, which were exposed to 150 ppb O3
33                   for 11-h/day for 30 days inside cuvettes, was inhibited, while isoprene emission and
34                   transcript levels of isoprene synthase mRNA were increased in the leaves exposed to
35                   ambient O3 (40 ppb), which were located above the leaves enclosed in the exposure
36                   cuvettes (Fares et al.. 2006). Exposure of 2 genotypes of hybrid poplar to 120 ppb O3 for
37                   6-h/day for 8 days resulted  in a significant reduction in isoprene emission in the O3-
38                   sensitive but not the tolerant genotype (Ryan et al.. 2009).  Similarly, O3 treatment
39                   (80 ppb 12-h/day for 14 days) of Populus nigra showed that isoprene emission was

      Draft - Do Not Cite or Quote                       9-35                                September 2011

-------
 1                   reduced in the treated plants relative to the control plants (Fares et al., 201 Ob). Based on
 2                   results of this and other studies, Fares et al. (201 Ob) concluded that the isoprenoid
 3                   pathway may be induced in plants exposed to acute O3 doses, while at lower doses
 4                   isoprene emission may be inhibited. Vickers et al. (2009) developed transgenic tobacco
 5                   plants with the isoprene synthase gene from Populus alba and exposed them to 120 ppb
 6                   O3 for 6-h/day for 2 days. They determined that the wildtype plants showed significantly
 7                   more O3 damage, including the development of leaf lesions and a decline in
 8                   photosynthetic rates, than the transgenic, isoprene-emitting plants. Transgenic plants also
 9                   accumulated less H2O2 and had lower levels of lipid peroxidation following exposure to
10                   O3 than the wildtype plants (Vickers et al.. 2009). These results indicate that isoprene
11                   may have a protective role for plants exposed to oxidative stress.
             9.3.6   Summary

12                   The results of recent studies on the effects of O3 stress on plants support and strengthen
13                   those reported in the 2006 O3 AQCD. The most significant new body of evidence since
14                   the 2006 O3 AQCD comes from research on molecular mechanisms of the biochemical
15                   and physiological changes observed in many plant species in response to O3 exposure.
16                   Recent studies have employed new techniques, such as those used in evaluating
17                   transcriptomes and proteomes to perform very comprehensive analyses of changes in
18                   gene transcription and protein expression in plants exposed to  O3. These newer molecular
19                   studies not only provide very important information regarding the many mechanisms of
20                   plant responses to O3, they also allow for the analysis of interactions between various
21                   biochemical pathways which are induced in response to O3. However, many of these
22                   studies have been conducted in artificial conditions with model plants, which are
23                   typically exposed to very high, short doses of O3.  Therefore, additional work remains to
24                   elucidate whether these plant responses are transferable to other plant species exposed to
25                   more realistic ambient conditions.

26                   Ozone is taken up into leaves through open stomata. Once inside the substomatal cavity,
27                   O3 is thought to rapidly react with the aqueous layer surrounding the cell (apoplast) to
28                   form breakdown products such as hydrogen peroxide (H2O2),  superoxide (O2"), hydroxyl
29                   radicals (HO) and peroxy radicals (HO2). Plants could be sensing the presence of O3
30                   and/or its breakdown products in a variety different ways, depending upon the plant
31                   species and the exposure parameters. Experimental evidence suggests that mitogen-
32                   activated protein kinases and calcium are important components of the signal
33                   transduction pathways, which communicate signals to the nucleus and lead to changes in
34                   gene expression in response to O3. It is probable that there are multiple sensing
3 5                   mechanisms and signal transduction pathways, and their activation may depend upon the
      Draft - Do Not Cite or Quote                       9-36                                September 2011

-------
 1                   plant species, its developmental stage and/or O3 exposure conditions. Initiation of signal
 2                   transduction pathways in O3 treated plants has also been observed in stomatal guard cells.
 3                   Reductions in stomatal conductance in have been described for many plant species
 4                   exposed to O3, and new experimental evidence suggests that this reduction may be due
 5                   not only to a decrease in carboxylation efficiency, but also to a direct impact of O3 on
 6                   stomatal guard cell function, leading to a changes in stomatal conductance.

 7                    Alterations in gene transcription that have been observed in O3-treated plants are now
 8                   evaluated more  comprehensively using DNA microarray studies, which measure changes
 9                   in the entire transcriptome rather than measuring the transcript levels of individual genes.
10                   These studies have demonstrated very consistent trends, even though O3 exposure
11                   conditions (concentration, duration of exposure), plant species and sampling times vary
12                   significantly. Genes involved in plant defense, signaling, and those associated with the
13                   synthesis of plant hormones and secondary metabolism are generally up-regulated in
14                   plants exposed to O3, while those related to photosynthesis and general metabolism are
15                   typically down-regulated. Proteome studies support these results by demonstrating
16                   concomitant increases or decreases in the proteins encoded by these genes. Transcriptome
17                   analysis has also illuminated the complex interactions that exist between several  different
18                   phytohormones and how they modulate plant sensitivity and response to O3.
19                   Experimental evidence suggests that while  ethylene and salicylic acid are needed to
20                   develop O3-induced leaf lesions, jasmonic acid acts antagonistically to ethylene and
21                   salicylic acid to limit the spread of the lesions. Abscisic acid, in addition to its role in
22                   regulating stomatal aperture, may also act antagonistically to the jasmonic acid signaling
23                   pathway. Changes in the quantity and activity of these phytohormones  and the
24                   interactions between them reveal some of the complexity of plant responses to an
25                   oxidative stressor such as O3.

26                   Another critical area of interest is to better understand and quantify the capacity of the
27                   plant to detoxify oxygen radicals using antioxidant metabolites, such as ascorbate and
28                   glutathione, and the enzymes that regenerate them. Ascorbate  remains an important focus
29                   of research, and, due to its location in the apoplast in addition to other cellular
30                   compartments, it is regarded as a first line of defense against oxygen radicals formed in
31                   the apoplast. Most studies demonstrate that antioxidant metabolites and enzymes increase
32                   in quantity and activity in plants exposed to O3, indicating that they play an important
33                   role in protecting plants from oxidative stress. However, attempts to quantify the
34                   detoxification capacity of plants have remained unsuccessful, as high quantities of
3 5                   antioxidant metabolites and enzymes do not always translate into greater protection of the
36                   plant. Considerable variation exists between plant species, different developmental
37                   stages, and the environmental and  O3 exposure conditions which plants are exposed to.
      Draft - Do Not Cite or Quote                        9-37                                 September 2011

-------
 1                   As indicated earlier, the described alterations in transcript levels of genes correlate with
 2                   observed changes quantity and activity of the enzymes and metabolites involved in
 3                   primary and secondary metabolism. In addition to the generalized up-regulation of the
 4                   antioxidant defense system, photosynthesis typically declines in O3 treated plants.
 5                   Declines in C fixation due to reductions in quantity and activity of Rubisco were
 6                   extensively described in the 2006 O3 AQCD. More recent studies support these results
 7                   and indicate that declines in Rubisco activity may also result from reductions in Rubisco
 8                   activase enzyme quantity. Other studies, which have focused on the light reactions of
 9                   photosynthesis, demonstrate that plant exposure to O3 results in declines in electron
10                   transport efficiency and a decreased capacity to quench oxidizing radicals. Therefore, the
11                   overall declines in photosynthesis observed in O3 -treated plants likely result from
12                   combined impacts on stomatal conductance, carbon fixation and the light reactions.
13                   While photosynthesis generally declines in plants exposed to O3, catabolic pathways such
14                   as respiration have been shown to increase. It has been hypothesized that increased
15                   respiration may result from greater energy needs for defense and repair. Secondary
16                   metabolism is generally up-regulated in a variety of species exposed to O3 as a part of a
17                   generalized plant defense mechanism. Some secondary metabolites, such as flavonoids
18                   and polyamines, are of particular interest as they are known to have antioxidant
19                   properties. The combination of decreases in C assimilation and increases in catabolism
20                   and the production of secondary metabolites would negatively impact plants by
21                   decreasing the energy available for growth and reproduction.
          9.4   Nature of Effects on Vegetation and Ecosystems
            9.4.1    Introduction

22                   Ambient O3 concentrations have long been known to cause visible symptoms, decreases
23                   in photosynthetic rates, decreases in growth and yield of plants as well as many other
24                   effects on ecosystems (U.S. EPA. 2006b. 1996c. 1986. 1978a). Numerous studies have
25                   related O3 exposure to plant responses, with most effort focused on the yield of crops and
26                   the growth of tree seedlings. Many experiments exposed individual plants grown in pots
27                   or soil under controlled conditions to known concentrations of O3 for a segment of
28                   daylight hours for some portion of the plant's life span. Information in this section also
29                   goes beyond individual plant scale responses to consider effects at the broader ecosystem
30                   scale, including effects related to ecosystem services.
31                   This section will focus mainly on studies published since the release of the 2006 O3
32                   AQCD. However, because much O3 research  was conducted prior to the 2006 O3 AQCD,
      Draft - Do Not Cite or Quote                       9-38                                September 2011

-------
 1                   the present discussion of vegetation and ecosystem response to O3 exposure is largely
 2                   based on the conclusions of the 1978, 1986, 1996, and 2006 O3 AQCDs.
                     9.4.1.1    Ecosystem Scale, Function, and Structure

 3                   Information presented in this section was collected at multiple spatial scales, ranging
 4                   from the physiology of a given species to population, community, and ecosystem
 5                   investigations. An ecological population is a group of individuals of the same species and
 6                   a community is an assemblage of populations of different species interacting with one
 7                   another that inhabit an area. For this assessment, "ecosystem" is defined as the interactive
 8                   system formed from all living organisms and their abiotic (physical and chemical)
 9                   environment within a given area (IPCC. 2007a). The boundaries of what could be called
10                   an ecosystem are somewhat arbitrary, depending on the focus of interest or study. Thus,
11                   the extent of an ecosystem may range from very small spatial scales to, ultimately, the
12                   entire Earth (IPCC. 2007a). All ecosystems, regardless of size or complexity, have
13                   interactions and physical exchanges between biota and abiotic factors, this includes both
14                   structural (e.g., soil type and food web trophic levels) and functional (e.g., energy flow,
15                   decomposition, nitrification) attributes.

16                   Ecosystems are most often defined by their structure based on the number and type of
17                   species present. Structure may refer to a variety of measurements including the species
18                   richness, abundance, community composition and biodiversity as well as landscape
19                   attributes. Competition among and within species and their tolerance to environmental
20                   stressors are key elements of survivorship. When environmental conditions are shifted,
21                   for example, by the presence of anthropogenic air pollution, these competitive
22                   relationships may change and tolerance to stress may be exceeded. Ecosystems may also
23                   be defined on a functional basis. "Function" refers to the suite of processes and
24                   interactions among the ecosystem components and their environment that involve
25                   nutrient and energy flow as well as other attributes including water dynamics and the  flux
26                   of trace gases. Plant processes including photosynthesis, respiration, C allocation,
27                   nutrient uptake and evaporation, are directly related to functions of energy flow and C,
28                   nutrient and water cycling. The energy accumulated and stored by vegetation (via
29                   photosynthetic C capture) is available to other organisms. Energy moves from one
30                   organism to another through food webs, until it is ultimately released as heat. Nutrients
31                   and water can be  recycled. Air pollution alters the function of ecosystems when elemental
32                   cycles or the energy flow are altered. This alteration can also be manifested in changes in
33                   the biotic composition of ecosystems.
      Draft - Do Not Cite or Quote                       9-39                                 September 2011

-------
 1                   There are at least three levels of ecosystem response to pollutants: (1) the individual
 2                   organism and its environment; (2) the population and its environment; and (3) the
 3                   biological community composed of many species and their environment (Billings. 1978).
 4                   Individual organisms within a population vary in their ability to withstand the stress of
 5                   environmental change. The response of individual organisms within a population is based
 6                   on their genetic constitution, stage of growth at time of exposure to stress, and the
 7                   microhabitat in which they are growing (Levine and Pinto. 1998). The stress range within
 8                   which organisms can exist and function determines the ability of the population to
 9                   survive. Those best able to cope with environmental stressors survive and reproduce.
10                   Competition among different species results in succession (community change over time)
11                   and, ultimately, sensitive species may be progressively replaced and communities shift to
12                   favor those species that may have the capability to tolerate stressors such as O3  (Rapport
13                   and Whitford. 1999; Guderian.  1985).
                     9.4.1.2    Ecosystem Services

14                   Ecosystem structure and function may be translated into ecosystem services. Ecosystem
15                   services are the benefits people obtain from ecosystems (UNEP. 2003). Ecosystems
16                   provide many goods and services that are of vital importance for the functioning of the
17                   biosphere and provide the basis for the delivery of tangible benefits to human society.
18                   Hassan et al. (2005) define these benefits to include supporting, provisioning, regulating,
19                   and cultural services:

20                      •  Supporting services are necessary for the production of all other ecosystem
21                         services. Some examples include biomass production, production of
22                         atmospheric O2, soil formation and retention, nutrient cycling, water cycling,
23                         and provisioning of habitat. Biodiversity is  a supporting service that is
24                         increasingly recognized to sustain many of the goods and services that humans
25                         enjoy from ecosystems. These provide a basis for three higher-level categories
26                         of services.
27                      •  Provisioning services, such as products (Gitay et al.. 2001). i.e., food
28                         (including game, roots, seeds, nuts and other fruit,  spices, fodder), water, fiber
29                         (including wood, textiles), and medicinal and cosmetic products (such as
30                         aromatic plants, pigments).
31                      •  Regulating services that are of paramount importance for human society such
32                         as (1) C sequestration, (2) climate and water regulation, (3) protection from
33                         natural hazards such as floods, avalanches,  or rock-fall, (4) water and air
34                         purification, and (5) disease and pest regulation.
      Draft - Do Not Cite or Quote                       9-40                                September 2011

-------
 1                      •  Cultural services that satisfy human spiritual and aesthetic appreciation of
 2                         ecosystems and their components including recreational and other nonmaterial
 3                         benefits.

 4                   In the sections that follow, available information on individual, population and
 5                   community response to O3 will be discussed. Effects of O3 on productivity and
 6                   C sequestration, water cycling, below-ground processes, competition and biodiversity,
 7                   and insects and wildlife are considered below and in the context of ecosystem services
 8                   where appropriate.
             9.4.2   Visible Foliar Injury and Biomonitoring

 9                   Visible foliar injury resulting from exposure to O3 has been well characterized and
10                   documented over several decades on many tree, shrub, herbaceous, and crop species
11                   (U.S. EPA. 2006b. 1996b. 1984. 1978a). Visible foliar injury symptoms are considered
12                   diagnostic as they have been verified experimentally in exposure-response studies, using
13                   exposure methodologies such as CSTRs, OTCs, and free-air fumigation (see Section 9.2
14                   for more detail on exposure methodologies). Several pictorial atlases and guides have
15                   been published, providing details on diagnosis and identification of O3-induced visible
16                   foliar injury on many plant species throughout North America (Flagler. 1998; NAPAP.
17                   1987) and  Europe (Innes etal.. 2001; Sanchez et al.. 2001). Typical visible injury
18                   symptoms  on broad-leaved plants include: stippling, flecking,  surface bleaching, bifacial
19                   necrosis, pigmentation (e.g., bronzing), chlorosis, and/or premature senescence. Typical
20                   visible injury symptoms for conifers include: chlorotic banding, tip burn, flecking,
21                   chlorotic mottling, and/or premature senescence of needles. Although common patterns
22                   of injury develop within a species, these foliar lesions can vary considerably between and
23                   within taxonomic groups. Furthermore, the degree and extent of visible foliar injury
24                   development varies from year to year and site to site (Orendovici-Best et al.. 2008;
25                   Chappelka et al.. 2007; Smith et al.. 2003). even among co-members of a population
26                   exposed to similar O3 levels, due to the influence of co-occurring environmental and
27                   genetic factors. Nevertheless, Chappelka et al. (2007) reported that the average incidence
28                   of O3-induced foliar injury was 73% on milkweed in the Great Smoky Mountains
29                   National Park in the years 1992-1996.

30                   Although the majority of O3-induced visible foliar injury occurrence has been observed
31                   on seedlings and small plants, many studies have reported visible injury of mature
32                   coniferous trees, primarily in the western U.S. (Arbaugh et al.. 1998) and to mature
33                   deciduous  trees in eastern North America (Schaub et al.. 2005; Vollenweider et al..  2003;
      Draft - Do Not Cite or Quote                       9-41                                 September 2011

-------
 1                   Chappelka et al.. 1999a: Chappelka et al.. 1999b: Somerset al.. 1998; Hildebrand et al.
 2                   1996V

 3                   It is important to note that visible foliar injury occurs only when sensitive plants are
 4                   exposed to elevated O3 concentrations in a predisposing environment. A major modifying
 5                   factor for O3-induced visible foliar injury is the amount of soil moisture available to a
 6                   plant during the year that the visible foliar injury is being assessed. This is because  lack
 7                   of soil moisture generally decreases stomatal conductance of plants and, therefore, limits
 8                   the amount of O3 entering the leaf that can cause injury (Matvssek et al.. 2006; Panek.
 9                   2004: Grulke et al.. 2003a: Panek and Goldstein. 2001: Temple et al.. 1992: Temple et
10                   al., 1988). Consequently, many studies have shown that dry periods in local areas tend to
11                   decrease the incidence and severity  of O3-induced visible foliar injury;  therefore, the
12                   incidence of visible foliar injury is not always higher in years and areas with higher O3,
13                   especially with co-occurring drought (Smith et al.. 2003). Other factors such as leafage
14                   influence the severity of symptom expression with older leaves showing greater injury
15                   severity as a result of greater seasonal exposure (Zhang et al.. 2010a).

16                   Although visible injury is a valuable indicator of the presence of phytotoxic
17                   concentrations of O3 in ambient air, it is not always a reliable indicator of other negative
18                   effects on vegetation. The significance of O3 injury at the leaf and whole plant levels
19                   depends on how much of the total leaf area of the plant has been affected, as well as the
20                   plant's age, size, developmental stage, and degree of functional redundancy among the
21                   existing leaf area. Previous O3 AQCDs have noted the difficulty in relating visible foliar
22                   injury symptoms to other vegetation effects such as individual plant growth,  stand
23                   growth,  or ecosystem characteristics (U.S. EPA. 2006b. 1996b). As a result, it is not
24                   presently possible to determine, with consistency across species and environments,  what
25                   degree of injury at the leaf level has significance to the vigor of the whole plant.
26                   However, in some cases, visible foliar symptoms have been correlated with decreased
27                   vegetative growth (Somers et al.. 1998: Karnosky et al.. 1996: Peterson etal.. 1987:
28                   BenoitetaL 1982) and with impaired reproductive function (Chappelka.  2002: Black et
29                   al.. 2000). Conversely, the lack of visible injury does not always indicate a lack of
30                   phytotoxic concentrations of O3 or a lack of non-visible O3 effects (Gregg et al., 2006.
31                   2003).
                     9.4.2.1    Biomonitoring

32                   The use of biological indicators to detect phytotoxic levels of O3 is a longstanding and
33                   effective methodology (Chappelka and Samuelson. 1998: Manning and Krupa. 1992). A
34                   plant bioindicator can be defined as a vascular or nonvascular plant exhibiting a typical
      Draft - Do Not Cite or Quote                       9-42                                 September 2011

-------
 1                   and verifiable response when exposed to a plant stress such as an air pollutant (Manning.
 2                   2003). To be considered a good indicator species, plants must (1) exhibit a distinct,
 3                   verified response; (2) have few or no confounding disease or pest problems; and (3)
 4                   exhibit genetic stability (U.S. EPA. 2006b). Such sensitive plants can be used to detect
 5                   the presence of a specific air pollutant such as O3 in the ambient air at a specific location
 6                   or region and, as a result of the magnitude of their response, provide unique information
 7                   regarding specific ambient air quality. Bioindicators can be either introduced sentinels,
 8                   such as the widely used tobacco (Nicotiana tabacuni) variety  Bel W3 (Calatayud et al..
 9                   2007b: Laffrav et al.. 2007; Nali et al.. 2007; Gombert et al.. 2006; Kostka-Rick and
10                   Hahn. 2005; Heggestad. 1991) or detectors, which are sensitive native plant species
11                   (Chappelka et al.. 2007; Souza et al.. 2006). The approach is especially useful in areas
12                   where O3 monitors are not operated (Manning. 2003). For example, in remote wilderness
13                   areas where instrument monitoring is generally not available, the use of bioindicator
14                   surveys in conjunction with the use of passive samplers (Krupa et al.. 2001) may be a
15                   useful methodology (Manning. 2003). However, it requires expertise in recognizing those
16                   signs and symptoms uniquely attributable to exposure to O3 as well as in their
17                   quantitative assessment.

18                   Since the 2006 O3 AQCD, new sensitive plant species have been identified from field
19                   surveys and verified in controlled exposure studies (Kline et al.. 2009; Kline et al.. 2008).
20                   Several multiple-year field surveys have also been conducted at National Wildlife
21                   Refuges in Maine, Michigan, New Jersey, and South Carolina (Davis. 2009. 2007a. b;
22                   Davis and Orendovici. 2006).

23                   The USDA Forest Service through the Forest Health Monitoring Program (FHM) (1990 -
24                   2001) and currently the Forest Inventory and Analysis (FIA) Program has been collecting
25                   data regarding the incidence and severity of visible foliar injury on a variety of O3
26                   sensitive  plant species throughout the U.S. (Coulston et al.. 2003; Smith et al.. 2003). The
27                   plots where these data are taken are known as  biosites. These biosites are located
28                   throughout the country and analysis of visible foliar injury within these  sites follows a set
29                   of established protocols. For more details, see http://www.nrs.fs.fed.us/fia/topics/ozone/
30                   (USDA. 2011). The network has provided evidence of O3 concentrations high enough to
31                   induce visible symptoms on sensitive vegetation. From repeated observations and
32                   measurements made over a number of years, specific patterns of areas experiencing
33                   visible  O3 injury symptoms can be identified.  Coulston et al. (2003) used information
34                   gathered  over a 6-year period (1994-1999) from the network to identify several species
35                   that were sensitive to O3 over a regional scale including sweetgum (Liquidambar
36                   styraciflud), loblolly pine (Pinus taedd), and black cherry (Prunus serotind). In a study of
37                   the west coast of the U.S, Campbell et al. (2007) reported O3  injury in 25-37% of biosites
38                   in California forested ecosystems from 2000-2005.
      Draft - Do Not Cite or Quote                       9-43                                September 2011

-------
 1                   A study by Kohut (2007) assessed the risk of O3-induced visible foliar injury on
 2                   bioindicator plants (NPS. 2006) in 244 national parks in support of the National Park
 3                   Service's Vital Signs Monitoring Network (NFS. 2007). The risk assessment was based
 4                   on a simple model relating response to the interaction of species, level of O3 exposure,
 5                   and exposure environment. Kohut (2007) concluded that the risk of visible foliar injury
 6                   was high in 65 parks (27%), moderate in 46 parks (19%), and low in 131 parks (54%).
 7                   Some of the well-known parks with a high risk of O3-induced visible foliar injury include
 8                   Gettysburg, Valley Forge, Delaware Water Gap, Cape Cod, Fire Island, Antietam,
 9                   Harpers Ferry, Manassas, Wolf Trap Farm Park, Mammoth Cave, Shiloh, Sleeping Bear
10                   Dunes, Great Smoky Mountains,  Joshua Tree, Sequoia and Kings Canyon, and Yosemite.

11                   Lichens have also long been used as biomonitors of air pollution effects on forest health
12                   (Nash. 2008).  It has been suspected, based on field surveys in the San Bernardino
13                   Mountains surrounding the Los Angeles air basin, that declines in lichen diversity and
14                   abundance were correlated with measured O3 gradients (Gul et al., 2011). Several recent
15                   studies in North America (Geiser and Neitlich. 2007; Gombert et al.. 2006; Jovan and
16                   McCune. 2006) and Europe (Nali et al.. 2007; Gombert et al., 2006) have used lichens as
17                   biomonitors of atmospheric deposition (e.g., N and S) and O3 exposure. Nali et al. (2007)
18                   found that epiphytic lichen biodiversity was not related to O3 geographical distribution.
19                   In addition, a recent study by Riddell et al. (2010) found that lichen species, Ramalina
20                   menziesii, showed no decline in physiological response to low and moderate
21                   concentrations of O3 and may not be a good indicator for O3 pollution. Mosses have also
22                   been used as biomonitors of air pollution; however, there remains a knowledge gap in the
23                   understanding of the effects of ozone on mosses as there has been very little information
24                   available on this topic in recent years.
                     9.4.2.2    Summary

25                   Visible foliar injury resulting from exposure to O3 has been well characterized and
26                   documented over several decades of research on many tree, shrub, herbaceous, and crop
27                   species (U.S. EPA. 2006b. 1996b. 1984. 1978a). Ozone-induced visible foliar injury
28                   symptoms on certain bioindicator plant species are considered diagnostic as they have
29                   been verified experimentally in exposure-response studies, using exposure methodologies
30                   such as continuous stirred tank reactors (CSTRs), OTCs, and free-air fumigation.
31                   Experimental evidence has clearly established a consistent association of visible injury
32                   with O3 exposure, with greater exposure often resulting in greater and more prevalent
33                   injury. Since the 2006 O3 AQCD, several multi-year field surveys of O3-induced visible
34                   foliar injury have been conducted at National Wildlife Refuges in Maine, Michigan, New
      Draft - Do Not Cite or Quote                       9-44                                September 2011

-------
 1                  Jersey, and South Carolina. New sensitive species showing visible foliar injury continue
 2                  to be identified from field surveys and verified in controlled exposure studies.

 3                  The use of biological indicators in field surveys to detect phytotoxic levels of O3 is a
 4                  longstanding and effective methodology. The USDA Forest Service through the Forest
 5                  Health Monitoring (FHM) Program (1990-2001) and currently the Forest Inventory and
 6                  Analysis (FIA) Program has been collecting data regarding the incidence and severity of
 7                  visible foliar injury on a variety of O3 sensitive plant species throughout the U.S. The
 8                  network has provided evidence that O3 concentrations were high enough to induce visible
 9                  symptoms on sensitive vegetation. From repeated observations and measurements made
10                  over a number of years, specific patterns of areas experiencing visible O3 injury
11                  symptoms can be identified. As noted in the preceding section, a study of 244 national
12                  parks indicated that the risk of visible foliar injury was high in 65 parks (27%), moderate
13                  in 46 parks (19%), and low in 131 parks (54%).
14                  Evidence is sufficient to conclude that there is a causal relationship between
15                  ambient O3 exposure and the occurrence of O3-induced visible foliar injury on
16                  sensitive vegetation across the U.S.
            9.4.3  Growth, productivity and carbon storage in natural ecosystems

17                  Ambient O3 concentrations have long been known to cause decreases in photosynthetic
18                  rates, decreases in growth, and decreases in yield (U.S. EPA. 2006b. 1996c. 1986.
19                  1978a). The O3-induced damages at the plant scale may translate to the ecosystem scale,
20                  and cause changes in productivity and C storage. This section focuses on the responses of
21                  C cycling to seasonal or multi-year exposures to O3 from the plant to ecosystem scale.
22                  Quantitative responses include changes in plant growth, plant biomass allocation,
23                  ecosystem production and ecosystem C sequestration. Because of the available
24                  information, most of discussion at the plant scale focuses on the response of individual
25                  plants, especially tree seedlings and crops, with limited discussion of mixtures of
26                  herbaceous species. Changes at the ecosystem scale are difficult to evaluate directly due
27                  to the complexity and the large spatial and temporal scale. The discussion of ecosystem
28                  effects focuses on the new studies at the large-scale FACE experiments and on ecological
29                  model simulations.
                    9.4.3.1    Plant growth and biomass allocation

30                  The previous O3 AQCDs concluded that there is strong evidence that exposure to O3
31                  decreases photosynthesis and growth in numerous plant species (U.S. EPA. 2006b.
      Draft - Do Not Cite or Quote                      9-45                                September 2011

-------
 1                   1996b. 1984. 1978a). Studies published since the last review support those conclusions
 2                   and are summarized below.

 3                   In general, research conducted over several decades has indicated that exposure to O3
 4                   alters stomatal conductance and reduces photosynthesis in a wide variety of plant species.
 5                   In a review of more than 55 studies, Wittig et al. (2007) reported that current O3
 6                   concentrations in the northern hemisphere are decreasing stomatal conductance (13%)
 7                   and photosynthesis (11%) across tree species. It was also found that younger trees (<4
 8                   year) were affected less by O3 than older trees. Further, the authors also found that
 9                   decreases in photosynthesis are consistent with the cumulative uptake of O3 into the leaf.
10                   In contrast, several studies reported that O3 exposure may result in loss of stomatal
11                   control, incomplete stomatal closure at night and a decoupling of photosynthesis and
12                   stomatal conductance, which may have implications for whole- plant water use (Section
13                   9.4.5).

14                   In a recently published meta-analysis, Wittig et al.  (2009) quantitatively compiled peer
15                   reviewed studies from the past 40 years on the effect of current and future O3 exposures
16                   on the physiology and growth of forest species. Wittig et al. (2009) reported that current
17                   ambient O3 concentrations as reported in those studies (-40 ppb) significantly decreased
18                   annual total biomass growth (7%) across 263 studies. However, this effect could be
19                   greater (11 to  17%) in areas that have higher O3 concentrations and as background O3
20                   increases in the future (Wittig et al.. 2009). This meta-analysis demonstrates the
21                   coherence of O3 effects across numerous studies and species that used a variety of
22                   experimental techniques, and these results support the conclusion of the previous AQCD.

23                   In two companion papers, McLaughlin et al. (2007a; 2007b) investigated the effects of
24                   ambient O3 on tree growth and hydrology at forest sites in the southern Appalachian
25                   Mountains. The  authors reported that the cumulative effects of ambient levels of O3
26                   decreased seasonal stem growth by 30-50% for most trees species in a high O3 year in
27                   comparison to a low O3 year (McLaughlin et al., 2007a). The authors also report that
28                   high ambient O3 concentrations can disrupt whole-tree water use and in turn reduce late-
29                   season streamflow (McLaughlin et al.. 2007b); see Section 9.4.5 for more on water
30                   cycling.

31                   Since the 2006 O3 AQCD, several new studies based on the Aspen FACE "free air" O3
32                   and CO2 exposure experiment in a forest in Wisconsin were published (Darbah et al..
33                   2008: Riikonen et al.. 2008: Darbah et al.. 2007: Kubiske et al.. 2007: Kubiske et al..
34                   2006; King et al.. 2005). Over the first seven years of stand development, Kubiske et al.
35                   (2006) observed that elevated O3 decreased tree heights, diameters, and main stem
36                   volumes in the aspen community by  11, 16, and 20%, respectively. In addition, Kubiske
37                   et al. (2007) reported that elevated O3 may change  the intra- and inter-species
      Draft - Do Not Cite or Quote                       9-46                                 September 2011

-------
 1                   competition. For example, O3 treatments increased the rate of conversion from a mixed
 2                   aspen-birch community to a birch dominated community. In a comparison presented in
 3                   Section 9.6.3 of this document, EPA found that effects on biomass accumulation in aspen
 4                   during the first seven years closely agreed with the exposure-response function based on
 5                   data from earlier OTC experiments.

 6                   Several studies at the Aspen FACE site also  considered other growth-related effects of
 7                   elevated O3. Darbah et al. (2008; 2007) reported that O3 treatments decreased paper birch
 8                   seed weight and seed germination and that this would likely lead to a negative impact of
 9                   regeneration for that species. Riikonen et al.  (2008) found that elevated O3  decreased the
10                   amount of starch in birch buds by 16%, and reduced aspen bud size, which may have
11                   been related to the observed delay in spring leaf development. The results suggest that
12                   elevated O3 concentrations have the potential to alter C metabolism of overwintering
13                   buds, which may have carry-over effects in the subsequent growing season (Riikonen et
14                   al.. 2008).

15                   Effects on growth of understory vegetation were also investigated at Aspen FACE.
16                   Bandeff et al. (2006) found that the effects of elevated CO2 and O3 on understory species
17                   composition, total and individual species biomass, N content, and 15N recovery were a
18                   result of overstory community responses to those treatments; however, there were no
19                   apparent direct treatment effects due to high  variability of the data. Total understory
20                   biomass increased with increasing light and was greatest under the open canopy of the
21                   aspen/maple community, as well as the more open canopy of the elevated O3 treatments
22                   (Bandeff et al.. 2006). Similarly,  data from a study by Awmack et al. (2007) suggest that
23                   elevated CO2 and O3 may have indirect growth effects on red (Trifolium pratense) and
24                   white (Trifolium repens) clover in the understory via overstory community effects;
25                   however, no direct effects of elevated O3 were observed.

26                   Overall, the studies at the Aspen  FACE experiment are consistent with many of the OTC
27                   studies that were evaluated in previous O3 AQCDs. These results strengthen our
28                   understanding of O3 effects on forests and demonstrate the relevance of the knowledge
29                   gained from trees grown in open-top chamber studies.

30                   For some annual species, particularly crops, the endpoint for an assessment of the risk of
31                   O3 exposure is yield or growth, e.g., production of grain. For plants grown  in mixtures
32                   such as hayfields, and natural or semi-natural grasslands (including native nonagricultural
33                   species), endpoints other than production of biomass may be important. Such endpoints
34                   include biodiversity or species composition,  and effects may result from competitive
35                   interactions among plants in mixed-species communities. Most of the available data on
36                   non-crop herbaceous species are for grasslands, with many of the recent studies
      Draft - Do Not Cite or Quote                       9-47                                September 2011

-------
 1                   conducted in Europe. See Section 9.4.7 for a review of the recent literature on O3 effects
 2                   on competition and biodiversity in grasslands.


                     Root Growth

 3                   Although O3 does not penetrate soil, it could alter root development by decreasing
 4                   C assimilation via photosynthesis leading to less C allocation to the roots (Andersen.
 5                   2003). The response  of root development to O3 exposure depends on available
 6                   photosynthate within the plant and could vary over time. Many biotic and abiotic factors,
 7                   such as community dynamics and drought stress, have been found to alter root
 8                   development under elevated O3. An earlier study at the Aspen FACE experiment found
 9                   that elevated O3 reduced coarse root and fine roots biomass in young stands of paper
10                   birch and trembling aspen (King etal. 2001). However, this reduction disappeared
11                   several years later. Ozone significantly increased fine-root (<1.0 mm) in the aspen
12                   community (Pregitzer et al., 2008). This increase in fine root production was due to
13                   changes in community composition, such as better survival of the O3-tolerant aspen
14                   genotype, birch, and  maple, rather than changes in C allocation at the individual tree level
15                   (Pregitzer et al.. 2008; Zak et al.. 2007). In an adult European beech/Norway spruce
16                   forest in Germany, drought was found to nullify the O3-driven stimulation of fine root
17                   growth. Ozone  stimulated fine-root production of beech during the humid year, but had
18                   no significant impact on fine root production in the dry year (Matyssek et al., 2010;
19                   Nikolova et al.. 2010).

20                   Using a non-destructive method, Vollsnes et al. (2010) studied the in vivo root
21                   development of subterranean clover (Trifolium subterraneuni) before, during and after
22                   short-term O3 exposure. It was found that O3  reduced root tip formation, root elongation,
23                   the total root length,  and the ratios between below- and above-ground growth within
24                   one week after exposure. Those effects persisted for up to three weeks; however, biomass
25                   and biomass ratios were not significantly altered at the harvest five weeks after exposure.

26                   Several recent meta-analyses have generally indicated that O3 reduced C allocated to
27                   roots. In one meta-analysis, Grantz et al.  (2006) estimated the effect of O3 on the
28                   root: shoot allometric coefficient (k), the ratio between the relative growth rate of the root
29                   and shoot. The results showed that O3 reduced the rootshoot allometric coefficient by
30                   5.6%, and the largest decline of the rootshoot allometric coefficient was observed in
31                   slow-growing plants. In another meta-analysis including 263 publications, Wittig et al.
32                   (2009)  found that current O3 exposure had no significant impacts on  root biomass and
33                   rootshoot ratio when compared to pre-industrial O3 exposure. However, if O3
34                   concentrations rose to 81-101 ppb (projected O3 levels in 2100), both root biomass and
35                   root:shoot ratio were found to significantly decrease. Gymnosperms and angiosperms
      Draft - Do Not Cite or Quote                       9-48                                 September 2011

-------
 1                  differed in their responses, with gymnosperms being less sensitive to elevated O3. In two
 2                  other meta-analyses, Wang et al. (2010) found elevated O3 reduced biomass allocation to
 3                  roots by 8.3% at ambient CO2 and 6.0% at elevated CO2, and Morgan et al. (2003) found
 4                  O3 reduced root dry weight of soybean. While there is clear evidence that O3 reduced C
 5                  allocation to roots, results of recent individual studies have been mixed, showing negative
 6                  (Jones et al.. 2010). non-significant (Andersen et al.. 2010; Phillips et al.. 2009) and
 7                  positive effects (Pregitzer et al.. 2008; Grebenc and Kraigher. 2007) on root biomass and
 8                  root: shoot ratio.
                    9.4.3.2   Summary

 9                  The previous O3 AQCDs concluded that there is strong and consistent evidence that
10                  ambient concentrations of O3 decrease photosynthesis and growth in numerous plant
11                  species across the U.S. Studies published since the last review continue to support that
12                  conclusion.

13                  The meta-analysis by Wittig et al.(2007) and (2009) demonstrates the coherence of O3
14                  effects on plant photosynthesis and growth across numerous studies and species using a
15                  variety of experimental techniques. Since the 2006 O3 AQCD, several studies were
16                  published based on the Aspen FACE experiment using "free air," O3, and CO2 exposures
17                  in a forest in Wisconsin.  Overall, the studies at the Aspen FACE experiment were
18                  consistent with many of the open-top chamber (OTC) studies that were the foundation of
19                  previous O3 NAAQS reviews. These results strengthen our understanding of O3  effects
20                  on forests and demonstrate the relevance of the knowledge gained from trees grown in
21                  open-top chamber studies.

22                  In recent studies, O3 was shown to have either negative, non-significant, or positive
23                  effects on root biomass and root: shoot ratio. While the findings of individual studies were
24                  mixed, recent meta-analyses have generally indicated that O3 reduced C allocated to roots
25                  (Wittig et al.. 2009; Grantz et al.. 2006).

26                  Evidence is sufficient to conclude that there is a causal relationship between O3
27                  exposure and reduced growth of woody and herbaceous vegetation.
                    9.4.3.3   Reproduction

28                  Studies during recent decades have demonstrated O3 effects on various stages of plant
29                  reproduction. The impacts of O3 on reproductive development, as reviewed by Black et
30                  al. (2000). can occur by influencing (1) age at which flowering occurs, particularly in
      Draft - Do Not Cite or Quote                      9-49                               September 2011

-------
 1                   long-lived trees that often have long juvenile periods of early growth without flower and
 2                   seed production; (2) flower bud initiation and development; (3) pollen germination and
 3                   pollen tube growth; (4) seed, fruit, or cone yields; and (5) seed quality (Table 9-1) (U.S.
 4                   EPA. 2006b). Several recent studies since the 2006 O3 AQCD further demonstrate the
 5                   effects of O3 on reproductive processes in herbaceous and woody plant species. Although
 6                   there have been documented effects of ozone on reproductive processes, a knowledge gap
 7                   still exists pertaining to the exact mechanism of these responses.

 8                   Ramo et al. (2007) exposed several meadow species to elevated O3 (40-50 ppb) and CO2
 9                   (+100 ppm), both individually and combined, over three growing seasons in ground-
10                   planted mesocosms, using OTCs. Elevated O3 delayed flowering of Campanula
11                   rotundifolia and Vicia cracca. Ozone also reduced the overall number of produced
12                   flowers and decreased fresh weight of individual Fragaria vesca berries.

13                   Black et al. (2007) exposed Brassica campestris to 70 ppb for two days during late
14                   vegetative growth or ten days during most of the vegetative phase. The two-day exposure
15                   had no effect on growth or reproductive characteristics, while the 10 day exposure
16                   reduced vegetative growth and reproductive site number on the terminal raceme,
17                   emphasizing the importance of exposure duration and timing. Mature  seed number and
18                   weight per pod were  unaffected due to reduced seed abortion, suggesting that, although
19                   O3 affected reproductive processes, indeterminate species such as B. campestris possess
20                   enough compensatory flexibility to avoid reduced seed production (Black et al.. 2007).

21                   In the determinate species, Plantago major, Black et al. (2010) found  that O3 may have
22                   direct effects on reproductive development in populations of differing sensitivity. Only
23                   the first flowering spike was exposed to 120 ppb O3 for 7 hours per day on 9 successive
24                   days (corresponding to flower development) while the leaves and second spike were
25                   exposed to charcoal-filtered air. Exposure of the first spike to O3 affected seed number
26                   per capsule on both spikes even though spike two was not exposed. The combined seed
27                   weight of spikes one  and two was increased by 19% in the two resistant populations,
28                   suggesting an overcompensation for injury; whereas, a decrease of 21% was observed in
29                   the most sensitive population (Black etal.. 2010). The question remains as to whether
30                   these effects are true  direct ozone-induced effects or compensatory responses.

31                   A study by Darbah et al. (2008; 2007) of paper birch (Betula papyrifera) trees at the
32                   Aspen FACE site in Rhinelander, WI investigated the effects of elevated O3 and/or CO2
33                   on reproductive fitness. Elevated O3 increased flowering, but decreased seed weight and
34                   germination success rate of seeds from the exposed trees. These results suggest that O3
35                   can dramatically affect flowering, seed production, and seed quality of paper birch,
36                   ultimately affecting its reproductive fitness (Darbah et al., 2008; Darbah et al..  2007).
      Draft - Do Not Cite or Quote                      9-50                                September 2011

-------
      Table 9-1      Ozone effects on plant reproductive processes (derived from Table
                      AX9-22 of the 2006 ozone AQCD)
             Species
             Condition Measures
               References
      Apocynun
      androsaemifolium
        Flowering time
Bergweiler and Manning (1999)
      Buddleia davidii
        Flowering time
Findley et al. (1997)
      Rubus cuneifolius
        Pollen germination
Chappelka (2002)
      Plantago major
        Pollen tube elongation
Stewart (1998)
      Fragaria * ananassa
        Fruit yield
Drogoudi and Ashmore (2001): Drogoudi and
Ash mo re (2000)
      Plantago major
        Seed yield
Lyons and Barnes (1998): Pearson et al. (1996):
Reiling and Davison (1992): Whitfield et al. (1997)
      Understory herbs
        Seed yield
Harward and Treshow (1975)
       Source: Adapted from 2006 O3 AQCD
 1
 2
 3
 4
 5
 6
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
9.4.3.4    Ecosystem Productivity and Carbon Sequestration

During the previous NAAQS review, there were limited studies that investigated the
effect of O3 exposure on ecosystem productivity and C sequestration. Recent studies
from long-term FACE experiments provide more evidence of the association of O3
exposure and changes in productivity at the ecosystem scale. In addition to experimental
studies, model studies also assessed the impact of O3 exposure on productivity and
C sequestration from stand to global scales.

Two types of models are most often used to study the ecological consequences of O3
exposure: (1) single plant growth models such as TREGRO and PnET-II (Hogsett et al.,
2008; Martin etal.. 2001; Ollinger et al.. 1997b). and (2) process-based ecosystem
models such as PnET-CN, Dynamic Land Ecosystem Model (DLEM), Terrestrial
Ecosystem Model (TEM), or Met Office Surface Exchange Scheme - Top-down
Representation of Interactive Foliage and Flora Including Dynamics (MOSES-TRIFFID)
(Telzer et al.. 2009; Ren et al.. 2007a: Sitch et al.. 2007; Ollinger et al.. 2002) (Table 9-2).
In these models, carbon uptake is simulated through photosynthesis (TREGRO, PnET -
II, PnET- CN, DLEM and MOSES-TRIFFID) or gross primary production (TEM).
Photosynthesis rate at leaf level is modeled by a function of stomatal conductance  and
other parameters in TREGRO, PnET -II, PnET- CN, DLEM and MOSES-TRIFFID.
Photosynthesis at canopy level is calculated by summing either photosynthesis of
different leaf types (TREGRO, DLEM, and MOSES-TRIFFID) or photosynthesis of
different canopy layers (PnET -II, PnET- CN). The detrimental effect of O3  on plant
growth is often simulated by multiplying photosynthesis rate by a coefficient that is
dependent on stomatal conductance and cumulative O3 uptake  (Table 9-2). Different
      Draft - Do Not Cite or Quote
                              9-51
                           September 2011

-------
 1                  plant functional groups (PFTs, such as deciduous trees, coniferous trees or crops) show
 2                  different responses to O3 exposure. PnET-II, PnET-CN, TEM, DLEM and MOSES-
 3                  TRIFFID estimate this difference by modifying net photosynthesis with coefficients that
 4                  represent the O3 induced fractional reduction of photosynthesis for each functional group.
 5                  The coefficients used in PnET-II, PnET-CN, TEM, DLEM are derived from the functions
 6                  of O3 exposure (AOT40) versus photosynthesis reduction from Reich (1987) and
 7                  Tjoelker et al. (1995). The coefficients used in MOSES-TRIFFID are derived from the
 8                  O3 dose-photosynthesis response function from Pleijel et al. (2004a) and Karlsson et al.
 9                  (2004). where O3 dose is estimated by a metric named CUOt (cumulative stomatal uptake
10                  of O3). The O3 threshold of CUOt is 1.6 nmol/m2/s for woody PFT and 5 nmol/m2/s for
11                  grass PFT, and is different from AOT40, which has an O3 threshold level of 40 ppb for
12                  all PFTs. Experimental and model studies  on ecosystem productivity and C sequestration
13                  at the forest stand scale as well as regional and global scales are reviewed in the
14                  following section.
      Draft - Do Not Cite or Quote                       9-52                                September 2011

-------
 Table 9-2        Comparison of models used to simulate the ecological
                     consequences of Os exposure
  Model
   Model
  feature
           Carbon uptake
            Ozone effect
Reference
 TREGRO
Hourly or
daily step,
single plant
model
simulating
vegetation
growth
process
Leaf: leaf photosynthesis is a function of
stomatal conductance, mesophyll
conductance and the gradient of CO2 from
atmosphere to the mesophyll cells
Canopy: Leaf is divided into different ages.
The canopy photosynthesis rate is the sum
the photosynthesis of all foliage groups
The effect of O3 on photosynthesis is      Hogsett et al.
simulated by reducing mesophyll          (2008):
conductance, and increasing respiration.   Weinstein et al.
The degree of O3 damage is determined   (2005): Tingey
by ambient O3 exposure, and the         et al. (2004)
threshold O3 concentration below which
O3 does not affect mesophyll
conductance and respiration
 PnET-ll    PnET-ll:      Leaf: Maximum photosynthesis rate is
 and PnET  monthly time- determined by a function of foliar N
 -CN       step, single   concentration, and stomatal conductance is
           plant model   determined by a function of the actual rate
           pnET_CN'   of the photosynthesis.
           monthly time- Canopy: canopy is divided into multiple,
           step,         even-mass layers and photosynthesis is
           ecosystem    simulated by a multilayered canopy
           mode        submodel
                                                    The effect of O3 on photosynthesis is
                                                    simulated by an equation of stomatal
                                                    conductance and O3 dose (AOT40). The
                                                    model assumes that photosynthesis and
                                                    stomatal conductance remain coupled
                                                    under O3 exposure, with a reduction in
                                                    photosynthesis for a given month causing
                                                    a proportion reduction in stomatal
                                                    conductance.
                                                                             Ollinger et al.
                                                                             (2002: 1997b):
                                                                             Pan et al.
                                                                             (2009)
 TEM      monthly time- Ecosystem: TEM is run at a 0.5*0.5 degree
           step,         resolution. Each grid cell is classified by
           ecosystem    vegetation type and soil texture, and
           mode        vegetations and detritus are assumed to
                        distribute homogeneously within grid cells.
                        Carbon flows into ecosystem via gross
                        primary production, which is a function of
                        maximum rate of assimilation,
                        photosynthetically active radiation, the leaf
                        area relative to the maximum annual leaf
                        area, mean monthly air temperate, and
	nitrogen availability.	
                                                    The direct O3 reduction on GPP is
                                                    simulated by multiplying GPP by f(O3)t,
                                                    where f(O3)t is determined by
                                                    evapotranspiration, mean stomatal
                                                    conductance, ambient AOT40, and
                                                    empirically O3 response coefficient
                                                    derived from previous publications.
                                                                             Felzer et al.
                                                                             (2005: 2004)
 DLEM     daily time-    Leaf: photosynthesis is a function of 6
           step         parameters: photosynthetic photon flux
           ecosystem    density, stomatal conductance, daytime
           model        temperature, the atmospheric CO2
                        concentration, the leaf N content and the
                        length of daytime.
                        Canopy: Photosynthetic rates for sunlit leaf
                        and shaded leaf scale up to the canopy
                        level by multiplying the estimated leaf area
                        index
                        Ecosystem: GPP is the sum of gross C
	fixation of different plant function groups
                                                    The detrimental effect of O3 is simulated
                                                    by multiplying the rate of photosynthesis
                                                    by O3eff, where O3eff is a function of
                                                    stomatal conductance, ambient AOT40,
                                                    and O3 sensitive coefficient. Ozone's
                                                    indirect effect on stomatal conductance is
                                                    also simulated, with a reduction in
                                                    photosynthesis for a given month causing
                                                    a reduction in stomatal conductance, and
                                                    therefore canopy conductance.
                                                                             Ren et al.
                                                                             (2007a:
                                                                             2007b): Zhang
                                                                             et al. (2007a)
 MOSES-   30 minutes   Leaf: photosynthesis is a function of
 TRIFFID   time-step,     environmental and leaf parameters and
           dynamic      stomatal conductance;  Stomatal
           global        conductance is a function of the
           vegetation    concentration of CO2 and H2O in air at the
           model        leaf surface and the current rate of
                        photosynthesis of the leaf
                        Canopy: Photosynthetic rates scale up to
                        the canopy level by multiplying a function of
                        leaf area index and PAR extinction
                        coefficient
                        Ecosystem: GPP is the sum of gross C
	fixation of different plant function groups
                                                    The effect of O3 is simulated by
                                                    multiplying the rate of photosynthesis by
                                                    F, where F depends upon stomatal
                                                    conductance, O3 exposure, a critical
                                                    threshold for O3 damage, and O3
                                                    sensitive coefficient (functional type
                                                    dependent)
                                                                             Sitch et al.
                                                                             (2007)
 Draft - Do Not Cite or Quote
                                            9-53
                                                                          September 2011

-------
                    Local Scale

 1                  The above- and below-ground biomass and net primary production (NPP) were measured
 2                  at the Aspen FACE site after 7 years of O3 exposure. Elevated O3 caused 23, 13 and 14%
 3                  reductions in total biomass relative to the control in the aspen, aspen-birch and aspen-
 4                  maple communities, respectively (King et al., 2005). At the Kranzberg Forest FACE
 5                  experiment in Germany, O3 reduced annual volume growth by 9.5 m3/ha in a mixed
 6                  mature stand of Norway spruce and European beech (Pretzsch et al., 2010). At the
 7                  grassland FACE experiment at Alp Flix,  Switzerland, O3 reduced the seasonal mean rates
 8                  of ecosystem respiration and GPP by 8%, but had no significant impacts on aboveground
 9                  dry matter productivity or growing season net ecosystem production (NEP) (Yolk et al..
10                  2011). Ozone also altered C accumulation and turnover in soil, as discussed in Section
11                  9.4.6.

12                  Changes in forest stand productivity under elevated O3 were assessed by several model
13                  studies. TREGRO (Table 9-2) has been widely used to simulate the effects of O3 on the
14                  growth of several species in different regions in the U.S. Hogsett et al. (2008) used
15                  TREGRO to evaluate the effectiveness of various forms and levels of air quality
16                  standards for protecting tree growth in the San Bernardino Mountains of California. They
17                  found that O3 exposures at the Crestline  site resulted in a mean 20.9% biomass reduction
18                  from  1980 to 1985 and 10.3% biomass reduction from 1995 to 2000, compared to the
19                  "background" O3 concentrations (O3  concentration in Crook County, Oregon). The
20                  level of vegetation protection projected was different depending on the air quality
21                  scenarios under consideration. Specifically, when air quality was simulated to just meet
22                  the California 8-h average maximum of 70 ppb and the maximum 3 months 12-h SUM06
23                  of 25  ppm-h, annual growth reductions were limited to 1% or less, while air quality that
24                  just met a previous NAAQS (the second  highest 1-h max [125 ppb]) resulted in 6-7%
25                  annual reduction in growth, resulting in the least protection relative to background O3
26                  (Hogsett etal.. 2008).

27                  ZELIG is a forest succession gap model, and has been used to evaluate the dynamics of
28                  natural stand succession. Combining TREGRO with ZELIG, Weinstein et al. (2005)
29                  simulated the effects of different O3 levels ( 0.5, 1.5,  1.75, and 2 times ambient) on the
30                  growth and competitive interactions of white fir and ponderosa pine at three sites in
31                  California: Lassen National Park, Yosemite National Park, and Crestline. Their results
32                  suggested that O3 had little impact on white fir, but greatly reduced the growth of
33                  ponderosa pine. If current O3 concentrations continue over the next century, ambient O3
34                  exposure (SUM06 of 110 ppm-h) at Crestline was predicted to decrease  individual tree
35                  C budget by 10% and decrease ponderosa pine abundance by 16%. Effects at Lassen
      Draft - Do Not Cite or Quote                      9-54                                September 2011

-------
 1                  National Park and Yosemite National Park sites were found to be smaller because of
 2                  lower O3 exposure levels (Weinstein et al.. 2005).

 3                  The effects of O3 on stand productivity and dynamics were also studied by other tree
 4                  growth or stand models, such as ECOPHYS, INTRASTAND and LINKAGES.
 5                  ECOPHYS is a functional-structural tree growth model. The model used the linear
 6                  relationship between the maximum capacity of carboxylation and O3 dose to predict the
 7                  relative  effect of O3 on leaf photosynthesis (Martin et al.. 2001). Simulations with
 8                  ECOPHYS found that O3 decreased stem dry matter production, stem diameter and leaf
 9                  dry matter production, induced earlier leaf abscission, and inhibited root growth (Martin
10                  et al., 2001). INTRASTAND is an hourly time step model for forest stand carbon and
11                  water budgets. LINKAGES is a monthly time step model simulating forest growth and
12                  community dynamics. Linking INTRASTAND with LINKAGES, Hanson et al. (2005)
13                  found that a simulated increase in O3 concentration in 2100 (a mean 20-ppb increase over
14                  the current O3 concentration) yields a 35% loss of net ecosystem C exchange (NEE) with
15                  respect to the current conditions (174 g C/m2/year).


                    Regional and Global Scales

16                  Since the publication of the 2006 O3 AQCD, there is additional evidence suggesting that
17                  O3 exposure alters ecosystem productivity and biogeochemical cycling at the regional
18                  and continental scale. Most of those studies were conducted by using process-based
19                  ecosystem models (Table 9-2) and are briefly reviewed in the following sections.

20                  Ollinger et al. (1997a) simulated the effect of O3 on hardwood forest productivity of 64
21                  hardwood sites in the northeastern U.S. with PnET-II (Table 9-2). Their simulations
22                  indicated that O3 caused a 3-16% reduction in NPP from 1987 to 1992 (Table 9-3). The
23                  interactive effects of O3, N deposition, elevated CO2 and land use history on C dynamics
24                  were estimated by PnET-CN (Table 9-2) (Ollinger etal.. 2002). The results indicated that
25                  O3 offset the increase in net C exchange caused by elevated CO2 and N deposition by
26                  13% (25.0 g C/m2/year) under agriculture site history, and 23% (33.6 g C/m2/year) under
27                  timber harvest site history.  PnET-CN was also used to assess changes in C sequestration
28                  of U.S. Mid-Atlantic temperate forest. Pan et al. (2009) designed a factorial modeling
29                  experiment to separate the effects of changes in atmospheric composition, historical
30                  climatic variability and land-disturbances on the C cycle. They found that O3 acted as a
31                  negative factor, partially offsetting the growth stimulation caused by elevated CO2 and N
32                  deposition in U.S. Mid-Atlantic temperate forest. Ozone decreased NPP of most forest
33                  types by 7-8%. Among all the forest types, spruce-fir forest was most resistant to O3
34                  damage, and NPP decreased by only 1 % (Pan et al.. 2009).
      Draft - Do Not Cite or Quote                       9-55                                September 2011

-------
 1                   Felzer et al. (2004) developed TEM 4.3 (Table 9-2) to simulate the effects of O3 on plant
 2                   growth and estimated effects of O3 on NPP and C sequestration of deciduous trees,
 3                   conifers and crops in the conterminous U.S. The results indicated that O3 reduced NPP
 4                   and C sequestration in the U.S. (Table 9-3) with the largest decreases (over 13% in some
 5                   locations) in NPP occurring in the Midwest agricultural lands during the mid-summer.
 6                   TEM was also used to evaluate the magnitude of O3 damage at the global scale (Table 9-
 7                   3) (Felzer etal., 2005). Simulations forthe period 1860 to 1995  show that the largest
 8                   reductions in NPP and net C exchange occurred in the mid western U.S., eastern Europe,
 9                   and eastern China (Felzer et al., 2005). DLEM (Table 9-2) was developed to simulate the
10                   detrimental effect of O3 on ecosystems, and has been used to examine the O3 damage on
11                   NPP and C sequestration in Great Smoky Mountains National Park (Zhang et al., 2007a).
12                   grassland ecosystems and terrestrial ecosystems in China (Ren et al.. 2007a: Ren et al..
13                   2007b). Results of those simulations are listed in  Table 9-3.

14                   Instead of using AOT40 as their O3 exposure metric as PnET, TEM and DLEM did,
15                   Sitch et al. (2007) incorporated a  different O3 metric named CUOt (cumulative stomatal
16                   uptake of O3), derived from Pleijel et al. (2004a). into the MOSES-TRIFFID coupled
17                   model (Table 9-2). In the CUOt metric, the fractional reduction of plant production is
18                   dependent on O3 uptake by stomata over a critical threshold for damage with this
19                   threshold level varying by plant functional type. Consistent with previous studies, their
20                   model simulation indicated that O3 reduced global gross primary production (GPP),
21                   C exchange rate and C sequestration (Table 9-3).  The largest reductions in GPP and  land-
22                   C storage were projected over North America, Europe, China and India. In the model,
23                   reduced ecosystem C uptake due to O3 damage, results in additional CO2 accumulation
24                   in the atmosphere and an indirect radiative forcing of climate change. Their simulations
25                   indicated that the indirect radiative forcing caused by O3  (0.62-1.09 W/m2) could have
26                   even greater impact on global warming than the direct radiative  forcing of O3
27                   (0.89 W/m2) (Sitch et al.. 2007).

28                   Results from the various model studies presented in Table 9-3 are difficult to compare
29                   because of the various spatial and temporal scales used in these studies. However, all the
30                   studies showed that O3 exposure decreased ecosystem productivity and C sequestration.
31                   These results are consistent and coherent with experimental results from the  leaf, plant
32                   and ecosystem scales (Sitch et al., 2007; Felzer etal.. 2005).  Many of the models use the
33                   same underlying function to simulate the effect of O3 exposure to C uptake. For example
34                   the functions of O3 exposure (AOT40) versus photosynthesis reduction for PnET-II,
35                   PnET-CN, TEM, DLEM were all from Reich (1987) and Tjoelker et al. (1995).
36                   Therefore, it is not surprising that the results are similar. While these models can be
37                   improved and more evaluation with experimental data can be done, these models
      Draft - Do Not Cite or Quote                      9-56                                September 2011

-------
 1                  represent the state of the science for estimating the effect of O3 exposure on productivity
 2                  and C sequestration.
                    9.4.3.5   Summary

 3                  During the previous NAAQS reviews, there were very few studies that investigated the
 4                  effect of O3 exposure on ecosystem productivity and C sequestration. Recent studies
 5                  from long-term FACE experiments, such as Aspen FACE, SoyFACE and the Kranzberg
 6                  Forest (Germany), provided evidence of the association of O3 exposure and reduced
 7                  productivity at the ecosystem level. Studies at the leaf and plant scales showed that O3
 8                  reduced photosynthesis and plant growth, which provided coherence and biological
 9                  plausibility for the decrease in ecosystem productivity.  Results across different ecosystem
10                  models, such as TREGRO, PnET, TEM and DLEM, were consistent with the FACE
11                  experimental evidence, which showed that O3 reduced ecosystem productivity.

12                  Although O3  generally causes negative effects on plant growth, the magnitude of the
13                  response varies among plant communities. For example, O3 had little impact on white fir,
14                  but greatly reduced growth of ponderosa pine in southern California (Weinstein et al..
15                  2005). Ozone decreased net primary production (NPP) of most forest types in Mid-
16                  Atlantic region, but had small impacts on spruce-fir forest (Pan et al.. 2009).

17                  In addition to plant growth, other indicators that are typically estimated by model studies
18                  include net ecosystem CO2 exchange (NEE), C sequestration, and crop yield. Model
19                  simulations consistently found that O3 exposure caused negative impacts on those
20                  indicators, but the severity of these impacts was influenced by multiple interactions of
21                  biological and environmental factors. The suppression of ecosystem C sinks results in
22                  more CO2 accumulation in the atmosphere. Globally, the indirect radiative forcing caused
23                  by O3 exposure through lowering ecosystem C sink could have an even greater impact on
24                  global warming than the direct radiative forcing of O3 (Sitch etal.. 2007). Ozone could
25                  also affect regional C budgets through interacting with multiple factors, such as N
26                  deposition, elevated CO2 and land use history. Model simulations suggested that O3
27                  partially offset the growth stimulation caused by elevated CO2 and N deposition in both
28                  Northeast- and Mid-Atlantic-region forest ecosystems of the U.S. (Pan et al.. 2009;
29                  Ollinger etal.. 2002).

30                  The evidence is sufficient to infer that there is a causal relationship between O3
31                  exposure and reduced productivity, and a likely causal relationship between O3
32                  exposure and reduced carbon sequestration  in terrestrial ecosystems.
      Draft - Do Not Cite or Quote                      9-57                                September 2011

-------
     Table 9-3      Modeled effects of ozone on primary production, C exchange, and
                    C sequestration
Scale
GPP Global
Global
U.S.
U.S.
NPP
Northeastern
U.S.
U.S. Mid-
Atlantic
China
Global
C exchange
Global
Global
U.S.
GSM National
Park
C sequestration
China
China
China
Model
MOSES-
TRIFFID
TEM
TEM
TEM
PnET
PnET
DLEM
TEM
MOSES-
TRIFFID
MOSES-
TRIFFID
TEM
DLEM
DLEM
DLEM
DLEM
Index
CUOta
AOT40
AOT40
AOT40
AOT40
AOT40
AOT40
AOT40
cuot
cuot
AOT40
AOT40
AOT40
AOT40
AOT40
Ozone Impacts
Decreased by 14-23% over the period 1901-2100
Decreased by 0.8% without agricultural
management and a decrease of 2.9% with optimal
agricultural management
Reduced by 2.3% without optimal N fertilization and
7.2% with optimal N fertilization from 1983-1993
Reduced by 2.6-6.8% during the late 1980s-early
1990s.
A reduction of 3-1 6% from 1 987-1 992
Decreased NPP of most forest types by 7-8%
Reduced NPP of grassland in China by 8.5 Tg C
from 1960s to 1990s
Reduced net C exchange (1950-1995) by 0.1 Pg
C/yr without agricultural management and 0.3 Pg
C/yr with optimal agricultural management
Decreased global mean land-atmosphere C fluxes
by 1 .3 Pg C/yr and 1 .7 Pg C/yr for the 'high' and
'low' plant O3 sensitivity models, respectively
Reduced land-C storage accumulation by between
143 Pg C/yr and 263 Pg C/yr from 1900-2100
Reduced C sequestration by 1 8-38 Tg C/yr from
1950to1995
Decreased the ecosystem C storage of deciduous
forests by 2.5% and pine forest by 1 .4% from 1 971
to 2001
Reduced total C storage by 0.06% in 1960s and
1 .6% in 1990s in China's terrestrial ecosystems
O3 exposure reduced the net C sink of China's
terrestrial ecosystem by 7% from 1961 to 2005
Ozone induced net carbon exchange reduction
ranged from 0.4-43.1% , depending on different
forest type
Reference
Sitch et al.
(2007)
Felzer et al.
(2005)
Felzer et al.
(2005)
Felzer et al.
(2004)
Ollinger et al.
(1997a)
Pan et al.
(2009)
Ren et al.
(2007b)
Felzer et al.
(2005)
Sitch et al.
(2007)
Sitch et al.
(2007)
Felzer et al.
(2004)
Zhang et al.
(2007a)
Ren et al.
(2007a)
Tian et al.
(2011)
Ren et al.
(2011)
     aCUOt is defined as the cumulative stomatal uptake of Os, using a constant Os-uptake rate threshold oft nmol/m /s.
      dPg equals 1 x 1015 grams.
2
O
4
5
6
9.4.4   Crop yield and quality in agricultural systems

        The detrimental effect of O3 on crop production has been recognized since the 1960s and
        a large body of research has stemmed from that recognition. Previous O3 AQCDs have
        extensively reviewed this body of literature. Table 9-4 summarizes recent experimental
        studies of O3 effects on agricultural crops, exclusive of growth and yield. Growth and
        yield results are summarized in Table 9-17.
     Draft - Do Not Cite or Quote
                                     9-58
September 2011

-------
 1                  The actual concentration and duration threshold for O3 damage varies from species to
 2                  species and sometimes even among genotypes of the same species (Guidi et al.. 2009;
 3                  Sawada and Kohno. 2009; Biswas et al.. 2008; Arivaphanphitak et al.. 2005; Dalstein and
 4                  Vas. 2005; Keutgen et al.. 2005). A number of comprehensive reviews and meta-analyses
 5                  have recently been published discussing both the current understanding of the
 6                  quantitative  effects of O3  concentration on a variety of crop species and the potential
 7                  focus areas for biotechnological improvement to a future growing environment that will
 8                  include higher O3 concentrations (Bender and Weigel. 2011; Booker et al.. 2009;
 9                  Van Dingenen et al.. 2009; Ainsworth. 2008; Feng et al.. 2008b; Haves et al.. 2007; Mills
10                  et al.. 2007b; Grantz et al.. 2006; Morgan etal.. 2003). Since the 2006 O3 AOCDOJ.S.
11                  EPA. 2006b). exposure-response indices for a variety of crops have been suggested
12                  (Mills et al.. 2007b) and many reports have investigated the effects of O3 concentration
13                  on seed or fruit quality to extend the knowledge base beyond yield quantity. This section
14                  will outline the key findings from these papers as well as highlight some of the recent
15                  research addressing the endpoints such as yields and crop quality.

16                  This section will also highlight recent literature that focuses on O3 damage to crops as
17                  influenced by other environmental factors. Genetic variability is not the only factor that
18                  determines crop response to O3 damage. Ozone concentration throughout a growing-
19                  season is not homogeneous and other environmental conditions such as elevated CO2
20                  concentrations, drought, cold or nutrient availability may alleviate or exacerbate the
21                  oxidative stress response to a given O3 concentration.
                     9.4.4.1    Yield

22                   It is well known that yield is negatively impacted in many crop species in response to
23                   high O3 concentration. However, the concentrations at which damage is observed vary
24                   from species to species. Numerous analyses of experiments conducted in OTCs and with
25                   naturally occurring gradients demonstrate that the effects of O3 exposure also vary
26                   depending on the growth stage of the plant; plants grown for seed or grain are often most
27                   sensitive to exposure during the seed or grain-filling period (Sojaetal.. 2000; Pleijel et
28                   al.. 1998; Younglove et al.. 1994; Leeetal.. 1988a). AX9.5.4.1 of the 2006 O3 AQCD
29                   summarized many previous studies on crop yield.

                     Field studies and meta-analyses
30                   The effect of O3 exposure on U.S. crops remains an important area of research and
31                   several studies have been published on this topic since the 2006 O3 AQCD (U.S. EPA.
32                   2006b) (Table 9-4 and 9-17). For example, one study with cotton in a crop-weed

      Draft - Do Not Cite or Quote                      9-59                               September 2011

-------
 1                   interaction study (Grantz and Shrestha. 2006) utilizing OTCs suggests that 12-hour
 2                   average O3 concentrations of 79.9 ppb decreased cotton biomass by 25% and 12-hour
 3                   average O3 concentration of 122.7 ppb decreased cotton biomass by 75% compared to
 4                   charcoal filtered control (12-h avg: 12.8 ppb). Further, this study suggests that the weed,
 5                   yellow nutsedge, was less sensitive to increasing O3 concentration, which would increase
 6                   weed competition (Grantz and Shrestha. 2006). In a study of peanuts in North Carolina,
 7                   near ambient and elevated exposures of O3 reduced photosynthesis and yield compared to
 8                   very low O3 conditions (Booker et al.. 2007; Burkey et al.. 2007). In another study,
 9                   Grantz and Vu (2009) reported that sugarcane biomass growth significantly declined
10                   under O3 exposure.

11                   The average yield loss reported across a number of meta-analytic studies have been
12                   published recently for soybean (Morgan et al.. 2003). wheat (Feng et al.. 2008b). rice
13                   (Ainsworth. 2008). semi-natural vegetation (Hayes et al.. 2007). potato,  bean and barley
14                   (Feng and Kobayashi. 2009). Meta-analysis allows for the objective development of a
15                   quantitative consensus of the effects of a treatment across a wide body of literature.
16                   Further, this technique allows for a compilation of data across a range of O3 fumigation
17                   techniques, durations and concentrations in order to assemble the existing literature in a
18                   meaningful manner.

19                   Morgan et al. (2003) reported an average seed yield loss for soybean of 24% compared to
20                   charcoal filtered air across all O3 concentrations used in the 53  compiled studies. The
21                   decrease in seed yield appeared to be the product of nearly equal decreases (7-12%) in
22                   seed weight, seed number and pod number. As would be expected, the lowest O3
23                   concentration (30-59 ppb) resulted in the smallest yield losses, approximately 8%, while
24                   the highest O3 concentration (80-120 ppb  ) resulted in the largest yield losses,
25                   approximately 35% (Morgan et al.. 2003). Further, the oil/protein ratio within the
26                   soybean seed was altered due to growth at elevated O3 concentrations, with a decrease in
27                   oil content. The studies included in this meta-analysis all used enclosed fumigation
28                   systems or growth chambers which may have altered the coupling of the atmosphere to
29                   the lower plant canopy (McLeod and Long. 1999), although the results of Morgan et al.
30                   (2006). Betzelberger et al. (2010). and the comparisons presented in Section 9.6.3
31                   strongly suggest that decreases in yield between ambient and elevated exposures are not
32                   affected by exposure method. Utilizing the Soybean Free Air gas Concentration
33                   Enrichment Facility (SoyFACE; www.soyface.illinois.edu). Morgan et al. (2006) report a
34                   20% seed yield loss due to a 23% increase in average daytime O3 concentration
35                   (56-69 ppb) within a single soybean cultivar across two growing seasons in Illinois,
36                   which lies within the range  predicted by the meta-analysis. A further breakdown of the
37                   effects of current O3 concentrations (AOT40 of 4.7 ppm-h) on bean seed quality
38                   (Phaseolus vulgaris) has identified that growth at current O3 concentrations compared to
      Draft - Do Not Cite or Quote                       9-60                                 September 2011

-------
 1                   charcoal-filtered air raised total lipids, total crude protein and dietary fiber content (Iriti et
 2                   al.. 2009). An increase in total phenolics was also observed, however the individual
 3                   phenolics compounds responded differently, with significant decreases in anthocyanin
 4                   content. The seeds from ambient O3 exposed plants also displayed increased total
 5                   antioxidant capacity compared to charcoal-filtered air controls (Iriti et al.. 2009).
 6                   Betzelberger et al. (2010) has recently utilized the  SoyFACE facility to compare the
 7                   impact of elevated O3 concentrations across 10 soybean cultivars to investigate
 8                   intraspecific variability of the O3 response to find physiological or biochemical markers
 9                   for eventual O3 tolerance breeding efforts (Betzelberger et al.. 2010). They report an
10                   average 17% decrease in yield across all 10 cultivars across two growing seasons due to a
11                   doubling of ambient O3 concentrations, with the individual cultivar responses ranging
12                   from -7% to -36%.  The exposure-response functions derived for these 10 current
13                   cultivars were similar to the response functions derived from the NCLAN studies
14                   conducted in the 1980s (Heagle. 1989) suggesting there has not been  any selection for
15                   increased tolerance to O3 in more recent cultivars. More complete comparisons between
16                   yield predictions based on data from cultivars used in NCLAN studies, and yield data for
17                   modern cultivars from SoyFACE are reported in Section 9.6.3 of this document. They
18                   confirm that the response of soybean yield to O3 exposure has not changed in current
19                   cultivars.

20                   A meta-analysis has also been performed on studies investigating the effects of O3
21                   concentrations on wheat (Feng et al., 2008b). Across 23 studies included,  elevated O3
22                   concentrations (ranging from a 7-h daily average of 31-200 ppb) decreased grain yield by
23                   29%. Winter wheat and spring wheat did not differ in their responses; however the
24                   response in both varieties to increasing O3 concentrations resulted in  successively larger
25                   decreases in yield, from a 20% decrease  in 42 ppb to 60% in 153 ppb O3.  These yield
26                   losses were mainly caused by a combination of decreases in individual grain weight (-
27                   18%), ear number per plant (-16%), and  grain number per ear (-11%). Further, the grain
28                   starch concentration decreased by 8% and the grain protein yield decreased by 18% due
29                   to growth at elevated O3 concentrations as well. However, increases in grain calcium and
30                   potassium levels were reported (Fenget  al.. 2008b).

31                   A recent meta-analysis found that growth at elevated O3 concentrations negatively
32                   impacts nearly every aspect of rice performance as well (Ainsworth. 2008). While rice is
33                   not a major crop in the U.S., it provides a staple food for over half of the global
34                   population (IRRI. 2002) and the effects of rising O3 concentrations on rice yields merits
35                   consideration. On average, rice yields decreased 14% in 62 ppb O3 compared to charcoal-
36                   filtered air. This yield loss was largely driven by a 20% decrease in grain number
37                   (Ainsworth. 2008).
      Draft - Do Not Cite or Quote                       9-61                                 September 2011

-------
 1                   Feng and Kobayashi (2009) have recently compiled yield data for six major crop species,
 2                   potato, barley, wheat, rice, bean and soybean and grouped the O3 treatments used in those
 3                   studies into three categories: baseline O3 concentrations (<26 ppb), current ambient 7- or
 4                   12-h daily O3 concentrations (31-50 ppb), and future ambient 7- or 12-h daily O3
 5                   concentrations (51-75 ppb). Using these categories, they have effectively characterized
 6                   the effects of current O3 concentrations and the effects of future O3 concentrations
 7                   compared to baseline O3 concentrations. At current O3 concentrations, which ranged
 8                   from 41-49 ppb in the studies included, soybean (-7.7%), bean (-19.0%), barley (-8.9%),
 9                   wheat (-9.7%), rice (-17.5%) and potato (-5.3%) all had yield losses compared to the
10                   baseline O3 concentrations (<26 ppb). At future O3 concentrations, averaging 63 ppb,
11                   soybean (-21.6%), bean (-41.4%), barley (-14%), wheat (-28%), rice (-17.5%) and potato
12                   (-11.9%) all had significantly larger yield losses compared to the losses at current O3
13                   concentrations (<26 ppb) (Feng and Kobayashi, 2009).

14                   A review of OTC studies has determined the AOT40 critical level that causes a 5% yield
15                   reduction across a variety of agricultural and horticultural species (Mills et al.. 2007b).
16                   The authors classify the species studied into three groups: sensitive, moderate and
17                   tolerant. The sensitive crops, including watermelon, beans, cotton, wheat, turnip, onion,
18                   soybean, lettuce, and tomato, respond with a 5% reduction in yield under a 3-month
19                   AOT40 of 6 ppm-h. Watermelon was the most sensitive with a critical level of
20                   1.6 ppm-h. The  moderately sensitive crops, including sugar beet, oilseed rape, potato,
21                   tobacco, rice, maize, grape and broccoli, responded with a 5% reduction in yield between
22                   8.6 and 20 ppm-h. The crops classified as tolerant, including strawberry, plum and barley,
23                   responded with  a 5% yield reduction between 62-83.3 ppm-h (Mills et al.. 2007b).

24                   Feng and Kobayashi (2009) compared their exposure-response results to those published
25                   by Mills et al. (2007b) and found the ranges of yield loss to be similar for soybean,  rice
26                   and bean. However, Feng and Kobayasi (2009) reported smaller yield losses for potato
27                   and wheat and larger yield losses for barley compared to the dose-response functions
28                   published by Mills et al. (2007b), which they attributed to their more lenient criteria for
29                   literature inclusion.

30                   While the studies investigating the impact of various O3 concentrations on yield are
31                   important and aid in determining the vulnerability of various crops to a variety of O3
32                   concentrations, there is still uncertainty as to how these crops respond under field
33                   conditions with interacting environmental factors such as temperature, soil moisture, CO2
34                   concentration, and soil fertility (Booker et al.. 2009). Further, there appears to be a
35                   distinct developmental and genotype dependent influence on plant sensitivity to O3 that
36                   has yet to be fully investigated across O3 concentrations in a field setting. The potentially
37                   mitigating effect of breeding selection for O3 resistance has received very little attention
      Draft - Do Not Cite or Quote                       9-62                                 September 2011

-------
 1                   in the published scientific literature. Anecdotal reports suggest that such selection may
 2                   have occurred in recent decades for some crops in areas of the country with high ambient
 3                   exposures. However, the only published literature available is on soybean and these
 4                   studies indicate that sensitivity has not changed in cultivars of soybean between the
 5                   1980s and the 2000s (Betzelberger et al.. 2010). This conclusion for soybeans is
 6                   confirmed by comparisons presented in Section 9.6.3 of this document.


                     Yield loss at regional and global scales

 7                   Because O3 is heterogeneous in both time and space and O3 monitoring stations are
 8                   predominantly near urban areas, the impacts of O3 on current crop yields at large spatial
 9                   scales are difficult to estimate. Fishman et al. (2010) have used satellite observations to
10                   estimate O3 concentrations in the contiguous tri-state area of Iowa, Illinois and Indiana
11                   and have combined that information with other measured environmental variables to
12                   model the historical impact of O3 concentrations on soybean yield across the 2002-2006
13                   growing seasons. When soybean yield across Iowa, Indiana and Illinois was modeled as a
14                   function of seasonal temperature, soil moisture and O3 concentrations, O3 had the largest
15                   contribution to the variability in yield for the southern-most latitudes included in the
16                   dataset. Fishman et al. (2010) determined that O3 concentrations significantly reduced
17                   soybean yield by 0.38 to 1.63% for every additional ppb of exposure across the 5 years.
18                   This value is consistent with previous chamber studies (Heagle. 1989) and results from
19                   SoyFACE (Morgan et al.. 2006). Satellite estimates of tropospheric O3 concentrations
20                   exist globally (Fishman et al., 2008). therefore utilizing this historical modeling approach
21                   is feasible across a wider geographical area, longer time-span and perhaps for more crop
22                   species.

23                   The detrimental effects of O3 on crop production at regional or global scales were also
24                   assessed by several model studies. Two large scale field studies were conducted in the
25                   U.S. (NCLAN) and in Europe (European Open Top Chamber Programme, EOTCP) to
26                   assess the impact of O3 on crop production. Ozone exposure-response regression models
27                   derived from the two programs have been widely used to estimate  crop yield loss
28                   (Avnerv et al.. 201 la. b; Van Dingenen et al.. 2009: Tong and Mauzerall. 2008: Wang
29                   and Mauzerall. 2004). Those studies found that O3 generally reduced crop yield and that
30                   different crops showed different sensitivity to O3 pollution (Table  9-5). Ozone was
31                   calculated to induce a possible 45-82 million metric tons loss for wheat globally.
32                   Production losses for rice, maize and soybean were on the order of 17-23 million metric
33                   tons globally (Van Dingenen et al.. 2009). The largest yield losses occur in high-
34                   production areas exposed to high O3 concentrations, such the Midwest and the
35                   Mississippi Valley regions in the U.S., Europe, China and India (Van Dingenen et al..
36                   2009: Tong etal.. 2007).

      Draft - Do Not Cite or Quote                       9-63                                 September 2011

-------
                    9.4.4.2   Crop Quality

 1                  In general, it appears that increasing O3 concentrations above current ambient
 2                  concentrations can cause species-dependent biomass losses, decreases in root biomass
 3                  and nutritive quality, accelerated senescence and shifts in biodiversity. A study conducted
 4                  with highbush blackberry has demonstrated decreased nutritive quality with increasing
 5                  O3 concentration despite no change in biomass between charcoal-filtered control,
 6                  ambient O3 and 2 x ambient O3 exposures (Ditchkoff et al.. 2009). A study conducted
 7                  with sedge using control (30 ppb), low (55 ppb), medium (80 ppb) and high (105 ppb) O3
 8                  treatments has demonstrated decreased root biomass and accelerated senescence in the
 9                  medium and high O3 treatments (Jones et al.. 2010). Alfalfa showed no biomass changes
10                  across two years of double ambient O3 concentrations (AOT40 of 13.9 ppm-h) using
11                  FACE fumigation (Maggio et al.. 2009). However a modeling study has demonstrated
12                  that 84% of the variability in the relative feed value in high-yielding alfalfa was due to
13                  the variability in mean O3 concentration from  1998-2002 (Lin et al.. 2007). Further, in a
14                  managed grassland FACE system, the reduction in total biomass harvest over five years
15                  decreased twice as fast in the elevated treatment (AOT40 of 13-59 ppm-h)  compared to
16                  ambient (AOT40 of 1-20.7 ppm-h). Compared with the ambient control, loss in annual
17                  dry matter yield was 23% after 5 year. Further, functional groups were differentially
18                  affected, with legumes showing the strongest negative response (Volketal.. 2006).
19                  However, a later study by Stampfli and Fuhrer (2010) at the same site suggested that
20                  Volk et al.(2006) was likely overestimated the effects of O3 on yield reduction because
21                  the overlapping effects of species dynamics caused by heterogeneous initial conditions
22                  and a change in management were not considered in Volk et al. (2006). An OTC study
23                  conducted with Trifolium subterraneum exposed to filtered ( <15 ppb), ambient, and
24                  40 ppb above ambient O3 demonstrates decreases in biomass in the highest O3  treatment
25                  as well as 10-20% decreased nutritive quality which was mainly attributed to accelerated
26                  senescence (Sanz  et al.. 2005). A study conducted with Eastern gamagrass  and big
27                  bluestem in OTCs suggested that big bluestem is not sensitive to O3, but gamagrass
28                  displayed decreased nutritive quality in the 2 x ambient O3 treatment, due to higher
29                  lignin content and decreased N, (Lewis etal. 2006).
                    9.4.4.3   Summary

30                  The detrimental effect of O3 on crop production has been recognized since the 1960's
31                  and a large body of research has subsequently stemmed from those initial findings.
32                  Previous O3 AQCDs have extensively reviewed this body of literature (U.S. EPA,
33                  2006b). Current O3 concentrations across the U.S. are high enough to cause yield loss for
34                  a variety of agricultural crops including, but not limited to, soybean, wheat, potato,

      Draft - Do Not Cite or Quote                      9-64                                September 2011

-------
 1                   watermelon, beans, turnip, onion, lettuce, and tomato. Continued increases in O3
 2                   concentration may further decrease yield in these sensitive crops. Despite the well-
 3                   documented yield losses due to increasing O3 concentration, there is still a knowledge
 4                   gap pertaining to the exact mechanisms of O3-induced yield loss. Research has linked
 5                   increasing O3 concentration to decreased photosynthetic rates and accelerated
 6                   senescence, which are related to yield.

 7                   New research is beginning to consider the mechanism of damage caused by prolonged,
 8                   lower O3 concentration (so-called chronic exposure) compared to short, very high O3
 9                   concentration (so-called acute exposure). Both types of O3 exposure cause damage to
10                   agricultural crops, but through very different mechanisms. Historically, most research on
11                   the mechanism of O3 damage used acute exposure studies. During the last decade, it has
12                   become clear that the cellular and biochemical processes involved in the  response to
13                   acute O3 exposure are not involved in response to chronic O3 exposure, even though both
14                   cause yield loss in agriculturally important crops.

15                   In addition, new research has highlighted the effects of O3 on crop quality. Increasing O3
16                   concentration decreases nutritive quality of grasses, decreases macro- and micro-nutrient
17                   concentrations in fruits and vegetable crops, and decreases cotton fiber quality. These
18                   areas of research require further investigation to determine mechanisms and exposure-
19                   response relationships.

20                   During the previous NAAQS reviews, there were very few studies that estimated O3
21                   impacts on crop yields at large spatial scales. Recent modeling studies found that O3
22                   generally reduced crop yield, but the impacts varied across regions and crop species. For
23                   example, the largest O3-induced crop yield losses occurred in high-production areas
24                   exposed to high O3 concentrations, such the Midwest and the Mississippi Valley regions
25                   of the U.S. (Van Dingenen et al.. 2009). Among crop species, the estimated yield loss for
26                   wheat and soybean  were higher than for rice and maize (Van Dingenen et al.. 2009).
27                   Using satellite air-column observations with direct air-sampling O3 data, Fishman et al.
28                   (2010) modeled the yield-loss due to O3 over the continuous tri-state area of Illinois,
29                   Iowa and Wisconsin. They determined that O3 concentrations significantly reduced
30                   soybean yield, which further reinforces previous results from FACE-type experiments
31                   and OTC experiments.
32                   Evidence  is sufficient to conclude that there is a causal relationship between O3
33                   exposure and reduced yield and quality of agricultural crops.
      Draft - Do Not Cite or Quote                       9-65                                September 2011

-------
Table 9-4
Species
Facility
Location
Alfalfa (Medicago
sativa cv. Beaver)
Growth chambers
Bean
(Phaseolus vulgaris \.
cv Borlotto)
OTC, ground-planted
Curno, Italy
Big Blue Stem
(Andropogon gerardif)
OTC
Alabama, U.S.
Brassica napus
Growth chambers
Belgium
Brassica napus
cv. Westar
Growth chambers
Finland
Eastern Gamagrass
(Tripsacum
dactyloides)
OTC
Alabama, U.S.
Lettuce
(Lactuca sativa)
OTC
Carcaixent
Experimental Station,
Spain
Peanut
(Arachis hypogaea)
OTC
Raleigh, NC; U.S.
Poa pratensis
OTC
Braunschweig,
Germany
Potato
(Solarium tuberosum
cv. Bintje
OTC
Sweden & Finland
Potato
(Solanum tuberosum
cv. Indira)
Climate chambers
Germany
Soybean
OTC
Italy
Summary of recent studies of ozone effects on crops (exclusive of
growth and yield)
Exposure
Duration
1,2 or
4 days
4 months
4 months
4 days
17-26 days
4 months
30 days
Syr
3yr;
4-5 wk
in the
spring
2yr
8wk
Syr
Ozone Exposure9
(Additional treatment)
3, 5 or - h/day
85ppb
(Exposure duration)
Seasonal AOT40:
CF = 0.5ppm-h;
Ambient = 4.6 ppm-h
(N/A)
12-havg:
CF=14ppb;
Ambient = 29 ppb;
Elevated = 71 ppb
(N/A)
CF&176ppb
for 4 h/day
(N/A)
8-h avg:
CF&100ppb
(Bt/non-Bt;
herbivory)
12-havg:
CF=14ppb;
Ambient = 29 ppb;
Elevated = 71 ppb
(N/A)
12-h mean:
CF= 10.2 ppb;
NF = 30.1 ppb;
NF+03 = 62.7 ppb
(4 cultivars)
12-havg:
CF = 22 ppb;
Ambient = 46 ppb;
Elevated = 75 ppb
(C02:375ppm;
548 ppm; 730 ppm)
8-h avg:
CF+25=21.7ppb;
NF+50=73.1 ppb
(Competition)
CF=10ppb;
Ambient = 25 ppb);
Ambient(+) = (36 ppb);
Ambient(++) = (47 ppb)
(N/A)
CF=10ppb;
Ambient = 50 ppb;
2xAmbient= 100 ppb
(CO 2 : 400 ppm &
700 ppm)
AOT40:
CF = 0 ppm-h;
Ambient = 3.4 ppm-h;
Elevated = 9.0 ppm-h
(Well-watered &
water-stressed)
Variable(s) measured
Relative feed value
Seed lipid,
Protein content
Fiber content
Relative feed value
Glucosinolates
VOC emissions
Relative feed value
Lipid peroxidation;
Root length
Harvest biomass
Relative feed value
[K],[Ca],[Mg],[P],[N]perdry
weight of tubers *dose-response
regression, report significant
positive or negative slope with
increasing [03]
Pathogen infestation using %
necrosis
Daily
evapotranspiration
percent change from
(percent change from
ambient)
n.s.
"high variability among
treatment groups (N/A)
+28.5 (N/A)
+7.88 (N/A)
+14.54 (N/A)
n.s. (n.s.)
-41 (N/A)
-30.7 (N/A);
-34 (N/A)
-17 (-12)
+77 (+38)
-22 (-14)
-40 (-10)
N/A (n.s.; -8)
[N] [P] [Ca] n.s.;
[K]&[Mg]sig +
(N/A)
+52 (n.s.)
-28 (-14)
Reference
Muntifering etal.
(2006)
Iriti etal. (2009)
Lewis etal. (2006)
Gielen etal. (2006)
Himanen et al.
(2009b)
Lewis et al. (2006)
Calatayud et. al.
(2002)
Booker etal.
(2007)
Bender etal.
(2006)
Piikki et al. (2007)
Plessl etal. (2007)
Jaude et al. (2008)
Draft - Do Not Cite or Quote
9-66
September 2011

-------
Species
Facility
Location
Soybean
(Glycine max
cv. 93B15)
SoyFACE
Urbana, IL; U.S.
Soybean
(Glycine max
CV.93B15)
SoyFACE
Urbana, IL; U.S.
Soybean
(Glycine max
cv. Essex)
OTC, ground-planted
Raleigh, NC; U.S.
Soybean
(Glycine max
cv. Essex)
OTCs, 21 L pots
Raleigh, NC; U.S.
Soybean
(Glycine max)
lOcultivars)
SoyFACE
Urbana, IL; U.S.
Spring Wheat
(Triticum aestivum
cv. Minaret; Satu;
Drabant; Dragon)
OTCs
Belgium, Finland,
& Sweden
Strawberry
(Fragaria x ananassa
Duch. Cv. Korona
& Elsanta)
Growth chambers
Bonn, Germany
Sweet Potato
Growth Chambers
Bonn, Germany
Tomato
(Lycopersicon
esculentum)
OTC
Valencia, Spain
Trifolium repens &
Trifolium pretense
Aspen FACE
Rhinelander,WI;U.S.
Exposure
Duration
Syr
May-Oct
4 months
2yr
2x3
months
2yr
7yr
2 months
4wk
133 days
3 months
Ozone Exposure9
(Additional treatment)
AOT40:
Ambient = 5-22 ppm-h;
Elevated = 20-43 ppm-h
(C02:550ppm;
environmental
variability)
8-h avg:
Ambient = 38.5 ppb;
Elevated = 52 ppb
(Herbivory)
12-havg:
CF = 21 ppb;
1.5xAmbient = 74 ppb
(C02:370ppm&
714ppm)
12-havg:
CF=18ppb);
elevated - 11 ppb)
(CO 2 : 367 & 71 8)
8-h avg (ppb):
Ambient = 46.3 & 37.9;
Elevated = 82.5 & 61 .3
(Cultivar comparisons)
Seasonal AOT40s
ranged from
Oto16ppm-h
(N/A)
8-h avg:
CF = 0 ppb;
Elevated = 78 ppb
(N/A)
8-h avg:
CF = 0 ppb;
Ambient < 40 ppb;
Elevated = 255 ppb
(N/A)
8- mean:
CF= 16.3 ppb;
NF = 30.1 ppb;
NF(+) = 83.2 ppb
(Various cultivars;
early & late harvest)
3-mo daylight avg:
Ambient = 34.8 ppb;
1.2xAmbient = 42.23
ppb
(C02;560ppm)
Variable(s) measured
Photosynthesis in new leaves,
Herbivory
defense-related
genes
Post-harvest residue
Water-use efficiency
Total antioxidant capacity
Seed protein content;
1 ,000-seed weight regressed
across all experiments
Total leaf area
Tuber weight
Brix degree
Lignin;
Dry-matter
digestibility
percent change from
(percent change from
ambient)
N/A (n.s.)
N/A (N/A)
N/A (-15.46)
n.s. (N/A)
N/A (+19)
N/A (Significant negative
correlation)
N/A (Significant negative
correlation)
-16 (N/A)
-14 (-11. 5)
-7.2 (-3.6)
N/A (+19.3)
N/A (-4.2)
Reference
Bernacchietal.
(2006)
Casteel et al.
(2008)
Booker etal.
(2005)
Booker etal.
(Booker etal..
2004a)
Betzelberger et al.
(2010)
Piikkietal. (2008a)
Keutgen et al.
(2005)
Keutgen et al.
(2008)
Calvo, et al. (2005)
Muntifering et al.
(2006)
aOzone exposure in ppb unless otherwise noted.
bCF = Carbon-filtered air.
NF = Non-filtered air.
Draft - Do Not Cite or Quote
9-67
September 2011

-------
Table 9-5      Modeled effects of ozone on crop yield loss at regional and global
                scales
Scale
Global
Global
U.S.
U.S.
East
Asia
Index
M7a;M12b;
AOT40
M12b;AOT40
M7;M12;AOT40
SUM06
M7;M12
Ozone Impacts
Reduced by 7.3% to 12.3% for wheat, 5.4% to 15.6% for soybean, 2.8% to 3.7% for rice,
and 2.4% to 4.1% for maize in year 2000.
Ovinduced global yield reductions ranged from 8.5-1 4% for soybean, 3.9-15% for wheat,
and 2.2-5.5% for maize in year 2000. Global crop production losses totaled 79-121 million
metric tons, worth $11-18 billion annually (USD2000).
Reduced by 4.1 % to 4.4% for wheat, 7.1 % to 1 7.7% for soybean, 2.6% to 3.2% for rice,
and 2.2% to 3.6% for maize in year 2000.
Caused a loss of 53.8 million to 438 million bushels in soybean production, which account
for 1 .7-14.2% of total U.S. soybean production in 2005
Reduced the yield of wheat, rice and corn by 1-9% and soybean by 23-27% in China,
Japan and South Korea in 1990
Reference
Van Dingenenetal.
(2009)
Avneryetal. (2011 a)
Van Dingenen et al.
(2009)
long et al. (2007)
Wang and Mauzerall
(2004)
aM7 is defined as 7-h mean 03 concentration (ppb).
bM12 is defined as 12-h mean 03 concentration (ppb).
1
2
3
4
5
6
7
      9.4.5  Water Cycling

              Ozone can affect water use in plants and ecosystems through several mechanisms
              including damage to stomatal functioning and loss of leaf area. Section 9.3.2 reviewed
              possible mechanisms for effects of O3 exposure on stomatal functioning including build-
              up of CO 2 in substomatal cavity, impacts on signal transduction pathways, and direct O3
              impact on guard cells. Regardless of the mechanism, O3 exposure has been shown to alter
              stomatal performance, which may affect plant and stand transpiration and therefore could
              affect hydrological cycling (Figure 9-7).
Draft - Do Not Cite or Quote
                                                  9-68
September 2011

-------
                        O3 exposure
                           Decrease stomatal
                           conductance or
                           sluggish stomatal
                          ^response
                                                                     Altered canopy
                                                                     water loss
Figure 9-7     The potential effects of ozone exposure on watering cycling.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
              In the literature, there is not a clear consensus on the nature of leaf-level stomatal
              conductance response to O3 exposure. At the leaf level, O3 exposure is known to result in
              stomatal patchiness (Paoletti and Grulke. 2005; Omasa etal.. 1987; Ellenson and
              Amundson. 1982). i.e., the heterogeneous aperture of stomata on the leaf surface, and, as
              a result, the collective response of groups of stomata on leaves and canopies determines
              larger-scale responses to O3. When measured at steady-state high light conditions, leaf-
              level stomatal conductance is often found to be reduced when exposed to O3. For
              example, a meta-analysis of 55 studies found that O3 reduced stomatal conductance by
              11% (Wittig etal.. 2007). However, these steady-state measurements were generally
              taken at saturating light conditions and steady-state vapor pressure deficit (VPD).
              Saturating light and steady-state VPD conditions are not common in the field since many
              parts of the plant canopy are shaded throughout the day. When studied under varying
              environmental conditions, many studies  have reported incomplete stomatal closure with
              elevated O3 exposure during the day (Mills et al.. 2009; Grulke et al.. 2007b; Matyssek et
              al.. 1995; Wieser and Havranek. 1995) or at night (Grulke et al.. 2004). This may be due
              to sluggish stomatal response. Sluggish stomatal response, defined as a delay in stomatal
              response to changing environmental factors relative to controls (Paoletti and Grulke.
              2010) has also been documented by several researchers (Grulke et al.. 2007c; Matyssek et
Draft - Do Not Cite or Quote
                                                   9-69
September 2011

-------
 1                   al.. 1995; Pearson and Mansfield. 1993; Wallin and Skarbv. 1992; Lee et al.. 1990;
 2                   Skarbyetal.. 1987; Keller and Hasler. 1984; Reich and Lassoie. 1984). Sluggish stomatal
 3                   response associated with O3 exposure suggests an uncoupling of the normally tight
 4                   relationship between carbon assimilation and stomatal conductance as measured under
 5                   steady-state conditions (Gregg et al.. 2006; Paoletti and Grulke. 2005). Several tree and
 6                   ecosystem models, such as TREGRO, PnET and DLEM, rely on this tight relationship to
 7                   simulate water and carbon dynamics. The O3-induced impairment of stomatal control
 8                   may be more pronounced for plants growing under water stress (Wilkinson and Davies.
 9                   2010; Grulke et al.. 2007a: Paoletti and Grulke. 2005; Bonn et al.. 2004; Kellomaki and
10                   Wang.  1997; Tjoelker et al.. 1995; Reich and Lassoie. 1984). Since leaf-level stomatal
11                   regulation is usually assessed in a steady state rather than as a dynamic response to
12                   changing environmental conditions, steady state measurements cannot detect sluggish
13                   stomatal response. Because of sluggish stomatal responses, water loss from plants may be
14                   greater under dynamic environmental conditions over days and months.

15                   In addition to the impacts on stomatal performance, O3-induced physiological changes,
16                   such  as reduced leaf area index and accelerated leaf senescence could alter water use
17                   efficiency. It is well established from chamber and field studies that O3 exposure is
18                   correlated with lower foliar retention (Karnosky et al.. 2003; Topaetal.. 2001; Pell et al..
19                   1999; Grulke and Lee.  1997; Karnosky et al.. 1996; Miller et al.. 1972; Miller et al..
20                   1963). However, Lee et al. (2009a) did not find changes in needle area of ponderosa pine
21                   and reported that greater canopy conductance followed by water stress under elevated O3
22                   may have been caused by stomatal dysfunction. At the Aspen FACE experiment, stand-
23                   level water use, as indicated by sap flux per unit ground area, was not significantly
24                   affected by elevated O3 despite a 22% decrease in leaf area index and 20% decrease in
25                   basal area (Uddling et al.. 2008). The lack of negative effect of elevated O3 on stand
26                   water use may be due to the substantially increased whole plant hydraulic conductance
27                   per unit leaf area under elevated O3, as indicated by the sap  flux per unit total leaf area
28                   (kl) (Uddling et al.. 2009). The increased kl may be caused by the sluggish of stomatal
29                   response. In pure aspen stands, the stomatal closure response to increasing vapor pressure
30                   deficit was less sensitive and mid-day leaf water potential was lower under elevated O3,
31                   suggesting O3 impaired stomatal control over transpiration (Uddling et al.. 2009). Other
32                   potential factors contributing to the unchanged stand-level water use included the higher
33                   proportion of sun leaves, and similar or even increased fine root biomass under elevated
34                   O3 (Uddling et al.. 2008). Elevated O3 could also affect evapotranspiration by altering
35                   tree crown interception of precipitation. Ozone has been shown to change branch
36                   architectural parameters, and the effects were species-dependent at the Aspen FACE
37                   experiment (Rheaet al.. 2010). The authors found that there was a significant correlation
38                   between canopy architecture parameters and stem flow for birch but not aspen.
      Draft - Do Not Cite or Quote                       9-70                                 September 2011

-------
 1                   It is difficult to scale up physiology measurements from leaves to ecosystems. Thus, the
 2                   current understanding of how stomatal response at leaf scale is integrated at the scale of
 3                   whole forest canopies, and therefore how it influences tree and forest stand water use is
 4                   limited. Field studies by McLaughlin et al. (2007a; 2007b) provided valuable insight into
 5                   the possible consequences of stomatal sluggishness for ecosystem water cycling.
 6                   McLaughlin et al. (2007a; 2007b) indicated that O3 increased water use in a mixed
 7                   deciduous forest in eastern Tennessee. McLaughlin et al. (2007a; 2007b) found that O3,
 8                   with daily maximum levels ranging from 69.2 to 82.9 ppb, reduced stem growth by 30-
 9                   50% in the high-O3  year 2002. The decrease in growth rate was caused in part by
10                   amplification of diurnal cycles of water loss and recovery. Peak hourly O3 exposure
11                   increased the rate of water loss through transpiration as indicated by the increased stem
12                   sap flow. The authors suggested that a potential mechanism for the  increased sap flow
13                   could be altered stomatal  regulation from O3 exposure, but this was inferred through sap
14                   flow measurements  and was not directly measured. The increased canopy water loss
15                   resulted in higher water uptake by the trees as reflected in the reduced soil moisture in the
16                   rooting zone. The change in tree water use led to further impacts on the hydrological
17                   cycle at the landscape level. Increased water use under high O3 exposure was reported to
18                   reduce late-season modeled streamflow in three forested watersheds in eastern  Tennessee
19                   (McLaughlin et al.. 2007b).

20                   Felzer et al. (2009) used TEM-Hydro to assess the interactions of O3, climate, elevated
21                   CO2 and N limitation on the hydrological cycle in the eastern U.S. They found that
22                   elevated CO2 decreased evapotranspiration by 2-4%  and increased runoff by 3-7%, as
23                   compared to the effects of climate alone. When O3 damage and N limitation were
24                   included, evapotranspiration was reduced by an additional 4-7% and runoff was increased
25                   by an additional 6-11% (Felzer et al.. 2009). Based upon simulation with INTRAST and
26                   LINKAGES, Hanson et al. (2005) found that increasing O3 concentration by 20 ppb
27                   above the current ambient level yields a modest 3% reduction in water use. Those
28                   ecological models were generally built on the assumption that O3 induces stomatal
29                   closure and have not incorporated possible stomatal sluggishness due to O3 exposure.
30                   Because of this assumption, results of those models normally found that O3 reduced
31                   water use.
                     9.4.5.1   Summary

32                   Although the evidence was from a limited number of field and modeling studies, findings
33                   showed an association between O3 exposure and alteration of water use and cycling in
34                   vegetation and at the ecosystem level. There is not a clear consensus on the nature of
35                   leaf-level stomatal conductance response to O3 exposure. When measured under steady-

      Draft - Do Not Cite or Quote                       9-71                                September 2011

-------
 1                   state high light conditions, leaf-level stomatal conductance is often found to be reduced
 2                   when plants are exposed to O3. However, measurements of stomatal conductance under
 3                   dynamic light and VPD conditions indicate sluggish responses under elevated O3
 4                   exposure, which could potentially lead to increased water loss from vegetation. Field
 5                   studies conducted by McLaughlin et al. (2007a; 2007b) suggested that peak hourly O3
 6                   exposure increased the rate of water loss from several tree species, and led to a reduction
 7                   in the late-season modeled stream flow in three forested watersheds in eastern Tennessee.
 8                   Sluggish stomatal responses during O3 exposure was suggested as a possible mechanism
 9                   for increased water loss during peak O3 exposure. Currently, the O3-induced reduction in
10                   stomatal aperture is the biological assumption for most process-based models. Because of
11                   this assumption, results of those models normally found that O3 reduced water loss. For
12                   example, Felzer (2009) found that O3 damage and N limitation together reduced
13                   evapotranspiration and increased runoff.

14                   Although the direction of the response differed among studies, the evidence is
15                   sufficient to conclude that there is likely to be a causal relationship between O3
16                   exposure and the alteration of ecosystem water cycling.
             9.4.6   Below-Ground Processes

17                   Above-ground and below-ground processes are tightly interconnected. Because roots and
18                   soil organisms are not exposed directly to O3, below-ground processes are affected by O3
19                   through alterations in the quality and quantity of C supply from photosynthates and
20                   litterfall (Andersen. 2003). Ozone can decrease leaf C uptake by reducing photosynthesis
21                   (Section 9.3). Ozone can also increase metabolic costs by stimulating the production of
22                   chemical compounds for defense and repair processes, and by increasing the synthesis of
23                   antioxidants to neutralize free radicals (see Section 9.3), both of which increase the
24                   consumption of carbon for above-ground processes. Therefore,  O3 could significantly
25                   reduce the amount of C available for allocation to  below-ground by decreasing C uptake
26                   while increasing C consumption of above-ground processes (Andersen. 2003).

27                   Since the 2006 O3 AQCD, there is additional evidence for O3 effects on below-ground
28                   processes. Ozone has been found to alter root growth, soil food  web structure,
29                   decomposer activities, C turnover, water cycling and nutrient flow (Figure 9-8). Ozone
30                   effects on root development and root biomass production and soil food web structure are
31                   reviewed in sections 9.4.3.1 and 9.4.9.2, respectively. The focus in this section is on the
32                   response of litter input, decomposer activities,  soil respiration, soil C formation and
33                   nutrient cycling.
      Draft - Do Not Cite or Quote                       9-72                                September 2011

-------
                                      CO2, H2O                   CO2, H2O

                                             Altered stomatal function
                                                                       Litter production
                                                                       and chemistry
                                                                                CO, release
                          '  Altered species competition
                        K   "-1
                                                          Soil foodweb
                                                            •Bacteria
                                                            •Fungi
                                                      Micro & marco invertebrates
                              Soil physical &
                            chemical properties
      Source: Modified from Andersen (2003)
      Arrows denote C flux pathways that are affected by ozone. Dashed lines indicate where the impact of ozone is suspected but
     unknown.


     Figure 9-8    Conceptual diagram showing where ozone alters C, water and
                    nutrient flow in a tree-soil system, including transfer between biotic
                    and abiotic components below ground that  influence soil physical
                    and chemical properties.
I
2
9.4.6.1   Litter Carbon Chemistry, Litter Nutrient and Their
          Ecosystem Budgets

Consistent with previous findings, recent studies show that, although the responses are
often species-dependent, O3 tends to alter litter chemistry (U.S. EPA. 2006b).Alterations
     Draft - Do Not Cite or Quote
                             9-73
September 2011

-------
 1                   in chemical parameters, such as changes in C chemistry and nutrient concentrations, were
 2                   observed in both leaf and root litter (9-7).

 3                   At the Aspen FACE site, several studies investigated litter chemistry changes (Parsons et
 4                   al.. 2008; Johnson and Pregitzer. 2007; Chapman et al.. 2005; Liu et al.. 2005). In both
 5                   aspen and birch leaf litter, elevated O3 increased the concentrations of soluble sugars,
 6                   soluble phenolics and condensed tannins (Parsons et al.. 2008; Liu et al.. 2005).
 7                   Compared to other treatments, aspen litter under elevated O3 had the highest fiber
 8                   concentration, with the lowest concentration associated with the birch litter under the
 9                   same conditions (Parsons et al.. 2008). Chapman et al. (2005) measured chemical
10                   changes in fine root litter and found that elevated O3 decreased lignin  concentration. O3-
11                   induced chemistry changes were also reported from other experimental sites. Results
12                   from an OTC study in Finland suggested that elevated O3 increased the concentration of
13                   acid-soluble lignin, but had no significant impact on other chemicals such as total sugars,
14                   hemicelluloses, cellulose or total lignin in the litter of silver birch (Kasurinen et al..
15                   2006). Results from the free air canopy O3 exposure experiment at Kranzberg Forest
16                   showed that O3  increased starch concentrations but had no impact on cellulose and lignin
17                   in beech and spruce leaf litter (Aneja et al.. 2007). The effect of O3 on three antioxidants
18                   (ascorbate, glutathione and ot-tocopherol) in fine roots of beech was also assessed at
19                   Kranzberg Forest. The results indicated that  O3 had no significant effect on ot-tocopherol
20                   and ascorbate concentrations,  but decreased glutathione concentrations in fine roots
21                   (Haberer et al., 2008).  In addition to  changing C chemistry, O3 also altered nutrient
22                   concentrations in green leaves and litter (Table 9-6).

23                   The combined effects of O3 on biomass productivity and chemistry changes may alter
24                   C chemicals and nutrient contents at the canopy or ecosystem level. For example,
25                   although O3 had different impacts on their concentrations, annual fluxes of C chemicals
26                   (soluble sugar, soluble phenolics, condensed tannins, lipid and hemicelluloses), macro
27                   nutrients (N, P,  K and  S) and micro nutrients (Mg, B, Cu and Zn) to soil were all reduced
28                   due to lower litter biomass productivity at Aspen FACE (Liu et al., 2007a; Liu et al..
29                   2005). At the Kranzberg Forest, N content of spruce canopy in a mixed culture and Ca2+
30                   content of beech canopy in a monoculture increased due to elevated O3 increased leaf
31                   concentrations of those nutrients although leaf production was not significantly altered by
32                   O3 (Rodenkirchen et al.. 2009).
      Draft - Do Not Cite or Quote                       9-74                                 September 2011

-------
Table 9-6 The effect of elevated ozone on leaf/litter nutrient concentrations
Study Site
Suonenjoki Research
Station, Finland
Aspen FACE
Aspen FACE
Kranzberg Forest, Germany
Kranzberg Forest, Germany
Salerno, Italy
Kuopio University Research
Garden, Finland
Species
Silver birch
Aspen and birch
Birch
Beech and spruce
Beech and spruce
Holm oak
Red Clover
Ozone
Concentration
Ambient: 10-60ppb
Elevated: 2xambient
Ambient: 50-60 ppb
Elevated: 1.5xambient
Ambient: 50-60 ppb
Elevated: 1.5xambient
Ambient: 9-41 ppb
Elevated: 2xambient
Ambient: 9-41 ppb
Elevated: 2xambient
Non-filtered OTC: 29 ppb
Filtered OTC:17ppb
Ambient: 25.7 ppb
Elevated: 1.5xambient
Response
Decreased the concentration of P, Mn, Zn
and B in leaf litter
Decreased the concentrations of P, S, Ca
and Zn, but had no impact on the
concentrations of N, K, Mg, Mn, B and Cu
in leaf litter.
Increase N concentration in birch litter
Increased N concentration in beach leaf,
but not in spruce needle
1) Had no significant effects on spruce
needle chemistry; 2) increased Ca
concentration in beech leaves in
monoculture, but had no impacts on other
nutrients
Ozone had no significant impacts on litter
C, N, lignin and cellulose concentrations
increased the total phenolic content of
leaves and had minor effects on the
concentrations of individual phenolic
compounds
Reference
Kasurinen et al.
(2006)
Liu et al. (2007a)
Parsons et al. (2008)
Kozovits et al. (2005)
Rodenkirchen etal.
(2009)
Baldantoni etal.
(2011)
Saviranta et
al.(2010)
 1
 2
 3
 4
 5
 6
 9
10
11
12
13
14
15
16
17
9.4.6.2    Decomposer Metabolism and Litter Decomposition

The above- and below-ground physiological changes caused by O3 exposure cascade
through the ecosystem and affect soil food webs. In the 2006 O3 AQCD, there were very
few studies on the effect of O3 on the structure and function of soil food webs, except
two studies conducted by Larson et al. (2002) and Phillips et al. (2002). Since the last O3
AQCD, new studies have provided more information on how O3 affects the metabolism
of soil microbes and soil fauna.

Chung et al.(2.006) found that the activity of the cellulose-degrading enzyme 1,4-p-
glucosidase was reduced by 25% under elevated O3 at Aspen FACE. The decrease in
cellulose-degrading enzymatic activity was associated with the lower cellulose
availability under elevated O3 (Chung et al.. 2006). However, a later study at the same
site, which was conducted in the  10th year of the experiment, found that O3 had no
impact on cellulolytic activity in soil (Edwards and Zak. 2011). In a lysimeter study of
beech trees (Fagus sylvaticd) in Germany, soil enzyme activity was found to be
suppressed by O3 exposure (Esperschutz et al.. 2009; Pritsch et al.. 2009). Except for
xylosidase, enzyme activities involved in plant cell wall degradation (cellobiohydrolase,
beta-glucosidase and glucuronidase) were decreased in rhizosphere soil samples under
elevated O3 (2 x ambient level) (Pritsch etal.. 2009). Similarly, Chen et al. (2009) found
      Draft - Do Not Cite or Quote
                               9-75
September 2011

-------
 1                   O3 exposure, with a 3-month AOT40 of 21.4-44.1 ppm-h, decreased the microbial
 2                   metabolic capability in the rhizosphere and bulk soil of wheat, although the observed
 3                   reduction in bulk soil was not significant.

 4                   Ozone-induced change in soil organisms' activities could affect litter decomposition
 5                   rates. Results of recent studies indicated that O3 slightly reduced or have no impacts on
 6                   litter decomposition (Liu et al., 2009b; Parsons et al., 2008; Kasurinen et al., 2006)
 7                   (Baldantoni et al.. 2011). The responses varied among species, sites and exposure length.
 8                   Parsons et al. (2008) collected litter from aspen and birch seedlings at Aspen FACE site,
 9                   and conducted a 23-month field litter incubation starting  in 1999. They found that
10                   elevated O3 had different impacts on the decomposition of aspen and birch litter.
11                   Elevated O3 was found to reduce aspen litter decomposition. However, O3 accelerated
12                   birch litter decomposition under ambient CO2, but reduced it under elevated CO2
13                   (Parsons et al.. 2008). Liu et al. (2009b) conducted another litter decomposition study at
14                   Aspen FACE from 2003 to 2006, when stand leaf area index (LAI) reached its maximum.
15                   During the 93 5-day field incubation, elevated O3 was shown to reduce litter mass loss in
16                   the first year, but not in the second year. They suggested  that higher initial tannin and
17                   phenolic concentrations under elevated O3 reduced microbial activity in the first year
18                   (Liu et al.. 2009b). In an OTC experiment, Kasurinen et al. (2006) collected silver birch
19                   leaf litter from three consecutive growing seasons and conducted three separate litter-bag
20                   incubation experiments.  Litter decomposition was not affected by O3 exposure in the first
21                   two incubations, but a slower decomposition rate was found in the third incubation. Their
22                   principle component analysis indicated that the litter chemistry changes caused by O3
23                   (decreased Mn, P, B and increased C:N) might be partially responsible for the decreased
24                   mass loss of their third incubation. In another OTC experiment, Baldatoni et al. (2011)
25                   found that O3 significantly reduced leaf litter decomposition of Quercus ilex L, although
26                   litter C, N, lignin and cellulose concentrations were not altered by O3 exposure.
                     9.4.6.3    Soil respiration and carbon formation

27                   Ozone could reduce the availability of photosynthates for export to roots, and increase
28                   root mortality and turnover rates. Ozone has also been shown to reduce above-ground
29                   litter productivity and alter litter chemistry, which would affect the quality and quantity
30                   of the C supply to soil organisms (Section 9.4.6.1). The complex interactions among
31                   those changes make it difficult to predict the response of soil C cycling under elevated
32                   O3. The 2006 O3 AQCD concluded that O3 had no consistent impact on soil respiration
33                   (U.S. EPA. 2006b). Ozone could increase  or decrease soil respiration, depending on the
34                   approach and timing of the measurements. Ozone may also alter soil C formation.
35                   However, very few experiments directly measured changes in soil organic matter content

      Draft - Do Not Cite or Quote                       9-76                                September 2011

-------
 1
 2
 3
 4
 5
 6
under O3 fumigation (U.S. EPA. 2006b). Recent studies on soil respiration and soil
C content also found mixed responses. Most importantly, recent results from long-term
fumigation experiments, such as the Aspen FACE experiment, suggest that ecosystem
response to O3 exposure can change overtime. Observations made during the late
exposure years can be inconsistent with those during the early years, highlighting the
need for caution when assessing O3 effects based on short-term studies (Table 9-7).
      Table 9-7      The temporal variation of ecosystem responses to ozone exposure
                      at Aspen FACE site
Endpoint
Litter decomposition
Fine root production
Soil respiration
Soil C formation
Period of
Measurement
1999-2001
2003-2006
1999
2002, 2005
1998-1999
2003-2007
1998-2001
2004-2008
Response
03 reduced aspen litter decomposition. However, 03 accelerated
birch litter decomposition under ambient C02, but reduced it under
elevated CO 2
03 reduced litter mass loss in the first year, but not in the second
year.
03 had no significant impact on fine root biomass
03 increased fine root biomass
Soil respiration under +C02+03 treatment was lower than that
under +C02 treatment
Soil respiration under +C02+03 treatment was 5-25% higher than
under elevated C02 treatment.
03 reduced the formation rates of total soil C by 51% and acid-
insoluble soil C by 48%
No significant effect of 03 on the new C formed under elevated
CO 2
Reference
Parsons et al. (2008)
Liu et al. (2009b)
Kina et al. (2001)
Pregitzeretal. (2008)
King etal. (2001)
Pregitzeretal. (2006) (2008)
Loya et al. (2003)
Talhelm etal. (2009)
 9
10
11
12
13
14
15
16
17
18
Soil Respiration

Ozone has shown inconsistent impacts on soil respiration. A sun-lit controlled-
environment chamber study found that O3 had no significant effects on soil respiration,
fine root biomass or any of the soil organisms in a reconstructed ponderosa pine/soil-litter
system (Tingey et al.. 2006). In an adult European beech/Norway spruce forest at
Kranzberg Forest, the free air O3 fumigation (AOT40 of 10.2-117 ppm-h) increased soil
respiration under both beech and spruce during a humid year (Nikolova et al.. 2010) . The
increased soil respiration under beech has been accompanied by the increase in fine root
biomass and ectomycorrhizal fungi diversity and turnover (Grebenc and Kraigher. 2007).
The stimulating effect on soil respiration disappeared under spruce in a dry year, which
was associated with a decrease in fine root production in spruce under drought. This
finding suggested that drought was a more dominant stress than O3 for spruce  (Nikolova
etal.. 2010). Andersen et al.  (2010) labeled the canopies of European beech and Norway
      Draft - Do Not Cite or Quote
                              9-77
September 2011

-------
 1                   spruce with CO2 depleted in 13C at the same site. They did not observe any significant
 2                   changes in soil respiration for either species.

 3                   The nearly 10 year long studies at Aspen FACE indicated that the response of soil
 4                   respiration to O3 interacted with CO2 exposure and varied temporally (Table 9-7)
 5                   (Pregitzer et al. 2008; Pregitzeretal.. 2006; King etal.. 2001). Ozone treatment alone
 6                   generally had the lowest mean soil respiration rates, although those differences between
 7                   control and elevated O3 were usually not significant. However, soil respiration rates were
 8                   different with O3 alone and when acting in combination with elevated CO2. In the first
 9                   five years (1998-2002), soil respiration under +CO2+O3 treatment was similar to that
10                   under control and lower than that under +CO2 treatment (Pregitzer et al., 2006; King et
11                   al.. 2001). Since 2003,  +CO2+O3 treatment started to show the greatest impact on soil
12                   respiration. Compared to elevated CO2, soil respiration rate under +CO2+O3 treatment
13                   was 15-25% higher from 2003-2004, and 5-10% higher from 2005-2007 (Pregitzer et al..
14                   2008; Pregitzer et al., 2006). Soil respiration was highly correlated with the biomass of
15                   roots with diameters of <2 mm and <1 mm, across plant community and atmospheric
16                   treatments. The authors suggested that the increase in soil respiration rate may be due to
17                   +CO2+O3 increased fine  root (<1.0 mm) biomass production (Pregitzer et al.. 2008).

18                   Changes in leaf chemistry and productivity due to O3 exposure have been shown to affect
19                   herbivore growth and abundance (See Section 9.4.9.1). Canopy insects could affect soil
20                   carbon and nutrient cycling through frass deposition, or altering chemistry and quantity
21                   of litter input to the forest floor. A study at the Aspen FACE found that although elevated
22                   O3 affected the chemistry of frass and greenfall, these changes had small impact on
23                   microbial respiration and no effect on nitrogen leaching (Hillstrom et al.. 2010a).
24                   However, respiratory carbon loss and nitrate immobilization were nearly double  in
25                   microcosms receiving herbivore inputs than those receiving no herbivore inputs
26                   (Hillstrom  et al.. 2010a).


                     Soil Carbon Formation

27                   Ozone-induced reductions in plant growth can result in reduced C input to soil and
28                   therefore soil C content (Andersen. 2003). The simulations of most ecosystem models
29                   support this prediction  (Ren et al., 2007a; Zhang et al.. 2007a; Felzer et al.. 2004).
30                   However, very few studies have directly measured soil C dynamics under elevated O3.
31                   After the first four years of fumigation (from 1998 to 2001) at the Aspen FACE site,
32                   Loya et al. (2003) found that forest stands exposed to both elevated O3 and CO2
33                   accumulated 51% less total soil C, and 48% less acid-insoluble soil C compared  to stands
34                   exposed only to elevated  CO2. Soil organic carbon (SOC) was continuously monitored at
3 5                   the Aspen FACE site, and the later data showed that the initial reduction in new
      Draft - Do Not Cite or Quote                       9-78                                September 2011

-------
 1                   C formation (soil C derived from plant litter since the start of the experiment) by O3
 2                   under elevated CO2 is only a temporary effect (Table 9-7) (Talhelm et al.. 2009). The
 3                   amount of new soil C in the elevated CO2 and the combined elevated CO2 and O3
 4                   treatments has converged since 2002. There was no significant effect of O3 on the new C
 5                   formed under elevated CO2 over the last four years of the study (2004-2008). Talhelm
 6                   et al. (2009) suggested the observed reduction in the early years of the experiment might
 7                   be driven by a suppression of C allocated to fine root biomass. During the early exposure
 8                   years, O3 had no significant impact on fine root production (King et al.. 2001). However,
 9                   the effect of O3 on fine root biomass was observed later in the experiment. Ozone
10                   increased fine root production and the highest fine root biomass was observed under the
11                   combined elevated CO2 and O3 treatment in the late exposure years (Table 9-7)
12                   (Pregitzer et al.. 2006). This increase in fine root production was due to changes in
13                   community composition, such as better survival of O3-tolerant aspen genotype, birch and
14                   maple, rather than changes in C allocation at the individual tree level (Pregitzer et al..
15                   2008; Zak et al.. 2007).
                     9.4.6.4    Nutrient cycling

16                   Ozone can affect nutrient cycling by changing nutrient release from litter, nutrient uptake
17                   by plants, and soil microbial activity. Nitrogen is the limiting nutrient for most temperate
18                   ecosystems, and several studies examined N dynamics under elevated O3. Nutrient
19                   mineralization from decomposing organic matter is important for sustaining ecosystem
20                   production. Holmes et al. (2006) found that elevated O3 decreased gross N mineralization
21                   at the Aspen FACE site, indicating that O3 may reduce N availability. Other N cycling
22                   processes, such as NH4+ immobilization, gross nitrification, microbial biomass N and soil
23                   organic N, were not affected by elevated O3 (Holmes et al.. 2006). Similarly, Kanerva
24                   et al. (2006) found total N, NO3-, microbial biomass N, potential nitrification and
25                   denitrification in their meadow mesocosms were not affected by elevated O3 (40-50 ppb).
26                   Ozone was found to decreased soil mineral N content at SoyFACE, which was likely
27                   caused by a reduction in plant material input and increased denitrification (Pujol Pereira
28                   et al.. 2011). Ozone also showed small impact on other micro and macro nutrients. Liu
29                   et al. (2007a) assessed N, P, K, S, Ca, Mg, Mn, B,  Zn and  Cu release dynamics at Aspen
30                   FACE, and they found that O3 had no effects on most nutrients, except to decrease N and
31                   Ca release from litter. These studies reviewed above suggested that soil N cycling
32                   processes were not affected or slightly reduced by  O3 exposure. However, in a lysimeter
33                   study with young beech trees Stoelken et al. (2010) found  that elevated O3 stimulated N
34                   release from litter which was largely attributed to an enhanced mobilization of inert
35                   nitrogen fraction.
      Draft - Do Not Cite or Quote                      9-79                                September 2011

-------
 1                  Using the Simple Nitrogen Cycle model (SINIC), Hong et al. (2006) evaluated the
 2                  impacts of O3 exposure on soil N dynamics and streamflow nitrate flux. The detrimental
 3                  effect of O3 on plant growth was found to reduce plant uptake of N and therefore increase
 4                  nitrate leaching. Their model simulation indicated that ambient O3 exposure increased the
 5                  mean annual stream flow nitrate export by 12% (0.042 g N/m2/year) at the Hubbard
 6                  Brook Experimental Watershed from 1964-1994 (Hong et al.. 2006).
                    9.4.6.5    Dissolved Organic Carbon and Biogenic Trace Gases
                               Emission

 7                  The O3-induced changes in plant growth, C and N fluxes to soil and microbial
 8                  metabolism can alter other biogeochemical cycling processes, such as soil dissolved
 9                  organic carbon (DOC) turnover and trace gases emission.

10                  Jones et al. (2009) collected fen cores from two peatlands in North Wales, UK and
11                  exposed them to one of four levels of O3 (AOT40 of 0, 3.69, 5.87 and 13.80 ppm-h for
12                  41 days). They found the concentration of porewater DOC in fen cores was significantly
13                  decreased by increased O3 exposure.  A reduction of the low molecular weight fraction of
14                  DOC was concurrent with the observed decrease in DOC concentration. Their results
15                  suggested that O3 damage to overlying vegetation may decrease utilizable C flux to soil.
16                  Microbes, therefore, have to use labile C in the soil to maintain their metabolism, which,
17                  the authors hypothesized, leads to a decreased DOC concentration with a shift of the
18                  DOC composition to more aromatic, higher molecular weight organic compounds.

19                  Several studies since the 2006 O3 AQCD have examined the impacts of O3 on nitrous
20                  oxide (N2O) and methane (CH4) emission. Kanerva et al.  (2007) measured the fluxes of
21                  N2O and  CH4 in meadow mesocosms, which were exposed to elevated CO2 and O3 in
22                  OTCs in south-western Finland.  They found that the daily N2O fluxes were decreased in
23                  the NF+O3 (non-filtered air + elevated O3, 40-50 ppb) after three seasons of exposure.
24                  Elevated  O3 alone or combined with CO2 did not have any significant effect on the daily
25                  fluxes of CH4 (Kanerva  et al.. 2007). In another study conducted in central Finland, the
26                  4 year open air O3 fumigation (AOT40 of 20.8-35.5 ppm-h for growing season) slightly
27                  increased potential CH4  oxidation by 15% in the peatland microcosms, but did not affect
28                  the rate of potential CH4 production or net CH4 emissions, which is the net result of the
29                  potential  CH4 production and oxidation (Morsky et al.. 2008). However, several studies
30                  found that O3 could significantly reduce CH4 emission. Toet et al. (2011) exposed
31                  peatland mesocosms to O3 in OTCs for two years, and found that CH4 emissions were
32                  significantly reduced by about 25% during midsummer periods of both years. In an OTC
33                  study of rice paddy, Zheng et al. (2011) found that the daily mean CH4 emissions were
      Draft - Do Not Cite or Quote                      9-80                               September 2011

-------
 1                  significantly lower under elevated O3 treatments than those in charcoal-filtered air and
 2                  nonfiltered air treatments. They found that the seasonal mean CH4 emissions were
 3                  negatively related with AOT40, but positively related to the relative rice yield,
 4                  aboveground biomass and underground biomass.
                    9.4.6.6   Summary

 5                  Since the 2006 O3 AQCD, more evidence has shown that although the responses are
 6                  often site specific, O3 altered the quality and quantity of litter input to soil, microbial
 7                  community composition, and C and nutrient cycling. Biogeochemical cycling of below-
 8                  ground processes is driven by C input from plants. Studies at the leaf and plant level have
 9                  provided biologically plausible mechanisms, such as reduced photosynthetic rates,
10                  increased metabolic cost, and reduced root C allocation for the association of O3
11                  exposure and the alteration of below-ground processes.

12                  Results from Aspen FACE and other experimental studies consistently found that O3
13                  reduced litter production and altered C chemistry, such as soluble sugars, soluble
14                  phenolics, condensed tannins, lignin, and macro/micro nutrient concentration in litter
15                  (Parsons et al., 2008; Kasurinen et al., 2006; Liu et al., 2005). The changes in substrate
16                  quality and quantity could alter microbial metabolism under elevated O3, and therefore
17                  soil C and nutrient cycling. Several studies indicated that O3 suppressed soil enzyme
18                  activities (Pritsch et al.. 2009; Chung et al.. 2006). However, the impact of O3 on litter
19                  decomposition was inconsistent and varied among species, sites and exposure length.
20                  Similarly, O3 had inconsistent impacts on dynamics of micro and macro nutrients.

21                  Studies from the Aspen FACE experiment suggested that the response of below-ground
22                  C cycle to O3 exposure, such as litter decomposition, soil respiration and soil C content,
23                  changed over time. For example, in the early part of the experiment (1998-2003), O3 had
24                  no impact on soil respiration but reduced the formation rates of total soil C under
25                  elevated CO2. However, after 10-11 yr of exposure, O3 was found to increase soil
26                  respiration but have no significant impact on soil C formation under elevated CO2.
27                  The evidence is sufficient to infer that there is a causal relationship between O3
28                  exposure and the alteration of below-ground  biogeochemical cycles.
            9.4.7  Community composition

29                  The effects of O3 on species competition (AX9.3.3.4) and community composition
30                  (AX9.6.4) were summarized in the 2006 O3 AQCD. Plant species differ in their
      Draft - Do Not Cite or Quote                      9-81                                September 2011

-------
 1                   sensitivity to O3. Fast growing plants with high stomatal conductance and high specific
 2                   leaf area (SLA) were more likely to be sensitive to O3 exposure. Further, different
 3                   genotypes of a given species also vary in their sensitivity. This differential sensitivity
 4                   could change the competitive interactions that lead to loss in O3 sensitive species or
 5                   genotypes. In addition, O3 exposure has been found to alter reproductive processes in
 6                   plants (See Section 9.4.3.3). Changes in reproductive success could lead to changes in
 7                   species composition. However, since ecosystem-level responses result from the
 8                   interaction of organisms with one another and with their physical environment, it takes
 9                   longer for a change to develop to a level of prominence  at which it can be identified and
10                   measured. A shift in community composition in forest and grassland ecosystems noted in
11                   the 2006 O3 AQCD has continued to be observed from experimental and gradient studies.
12                   Additionally, research since the last review has shown that O3 can alter community
13                   composition and diversity of soil microbial communities.
                     9.4.7.1    Forest

14                   In the San Bernardino Mountains in southern California, O3 pollution caused a
15                   significant decline in  ponderosa pine (Pinus ponderosa ) and Jeffrey pine (Pinus jeffreyi}
16                   (U.S. EPA. 2006b). Pine trees in the young mature age class group exhibited higher
17                   mortality rates compared with mature trees at a site with severe O3 visible foliar injury.
18                   The vulnerability of young mature pines was most likely caused by the fact that trees in
19                   this age class were emerging into the canopy, where higher O3 concentrations were
20                   encountered (McBride and Laven. 1999). Because of the loss of O3-sensitive pines,
21                   mixed forests of ponderosa pine, Jeffery Pine and white fir (Abies concolor) shifted to
22                   predominantly white  fir (Miller, 1973). Ozone may have indirectly caused the decline in
23                   understory diversity in coniferous forests in the San Bernardino Mountains through an
24                   increase in pine litterfall. This increase in litterfall from O3 exposure  results in an
25                   understory layer that may prohibit the establishment of native herbs, but not exotic annual
26                   Galium aparine (Allen et al.. 2007).

27                   Ozone  damage to conifer forests has also been observed in several other regions. In the
28                   Valley of Mexico, a widespread mortality of sacred fir (Abies religiosd) was observed in
29                   the heavily polluted area of the  Desierto de los Leones National Park in the early 1980s
30                   (de Lourdes de Bauer and Hernandez-Tejeda, 2007; Fenn et al.. 2002). Ozone damage
31                   was widely believed to be an important causal factor in the dramatic decline of sacred fir.
32                   In alpine regions of southern France and the Carpathians Mountains,  O3  was also
33                   considered as the major cause of the observed decline in cembran pine (Pinus
34                   cembra)(Wieser et al., 2006). However, many environmental  factors  such as light,
35                   temperature, nutrient  and soil moisture, and climate extremes such as unusual dry and


      Draft - Do Not Cite or Quote                      9-82                                September 2011

-------
 1                  wet periods could interact with O3 and alter the response of forest to O3 exposure. For
 2                  those pollution gradient studies, several confounding factors, such as drought, insect
 3                  outbreak and forest management, may also contribute to or even be the dominant factors
 4                  causing the mortality of trees (de Lourdes de Bauer and Hernandez-Tejeda. 2007; Wieser
 5                  et al.. 2006).

 6                  New evidence from long-term free O3 fumigation experiments provided additional
 7                  support for the potential impacts of O3 on species competition and community
 8                  composition changes in forest ecosystems. At the Aspen FACE site, community
 9                  composition at both the genetic and species levels was altered after seven years of
10                  fumigation with O3 (Kubiske et al., 2007). In the pure aspen community, O3 fumigation
11                  reduced growth and increased mortality of sensitive  clone 259, while the O3 tolerant
12                  clone 8L emerged as the dominant clone. Growth of clone 8L was even greater under
13                  elevated O3 compared to controls, probably due to O3 alleviated competitive pressure on
14                  clone 8L by reducing growth of other clones. In the mixed aspen-birch and aspen-maple
15                  communities, O3 reduced the competitive capacity of aspen compared to birch and maple
16                  (Kubiske et al.. 2007). In a phytotron study, O3 fumigation reduced growth of beech but
17                  not spruce in mixed culture, suggesting a higher susceptibility of beech to O3 under
18                  interspecific competition (Kozovits et al.. 2005).
                    9.4.7.2   Grassland and Agricultural Land

19                  The response of managed pasture, often cultivated as a mixture of grasses and clover, to
20                  O3 pollution has been studied for many years. The tendency for O3-exposure to shift the
21                  biomass of grass-legume mixtures in favor of grass species, reported in the previous O3
22                  AQCD has been generally confirmed by recent studies. In a mesocosm study, Trifolium
23                  repens and Loliumperenne mixtures were exposed to an episodic rural O3 regime within
24                  solardomes for 12 weeks. T. repens showed significant changes in biomass but notZ.
25                  perenne, and the proportion of T. repens decreased in O3-exposed mixtures compared to
26                  the control (Haves et al., 2009). The changes  in community composition of grass-legume-
27                  forb mixtures were also observed at the Le Mouret FACE experiment, Switzerland.
28                  During the 5-year O3 fumigation (AOT40 of  13.3-59.5 ppm-h), the dominance of
29                  legumes in fumigated plots declined more quickly than those in the control plots (Yolk et
30                  al., 2006). However, Stampfli and Fuhrer (2010) re-analyzed the species and soil data and
31                  suggested that Volk et al. (2006)  overestimated the O3 effect. Stampfli and Fuhrer (2010)
32                  found that the difference in the species dynamics between control and O3 treatment was
33                  more caused by heterogeneous initial conditions than O3 exposure. Several studies also
34                  suggested the mature/species-rich ecosystems were more resilient to O3 exposure. At
35                  another FACE experiment, located at Alp  Flix, Switzerland, O3 fumigation (AOT40 of
      Draft - Do Not Cite or Quote                      9-83                                September 2011

-------
 1                   15.2-64.9 ppm-h) showed no significant impact on community composition of this
 2                   species-rich pasture (Bassin et al.. 2007b). Although most studies demonstrated an
 3                   increase in grass:forb ration with O3 exposure (Haves et al., 2009; U.S. EPA. 2006b). a
 4                   study on a simulated upland grassland community O3 reduce grass:forb ratio (Felicity et
 5                   al., 2010). which may be due to grass species in this community, such as Anthoxanthum
 6                   odoratum, was more sensitive to O3 than other most studied grass species such as L.
 7                   perenne (Haves et al., 2009). Pfleeger et al. (2010) collected seed bank soil from an
 8                   agricultural field and examined how the plant community responded over several
 9                   generations to elevated O3 exposures. Sixty plant species from 22 families emerged in the
10                   chambers over their four year study. Overall, they found that O3 appeared to have small
11                   impacts on seed germination and only a minor effect on species richness of pioneer plant
12                   communities.

13                   Several review papers have discussed the physiological and ecological characteristics of
14                   O3-sensitive herbaceous plants. Hayes et al. (2007) assessed species traits associated with
15                   O3 sensitivity by the changes in biomass caused by O3 exposure. Plants of the therophyte
16                   (e.g., annual) life form were particularly sensitive to O3. Species with higher mature leaf
17                   N concentration tended to be more sensitive than those with lower leaf N concentration.
18                   Plants growing under high oxidative stress environments, such as high light or high
19                   saline, were more sensitive to O3. Using the same dataset from Hayes et al. (2007). Mills
20                   et al. (2007a) identified the O3 sensitive communities. They found that the largest number
21                   of these O3 sensitive communities were associated with grassland ecosystems. Among
22                   grassland ecosystems, alpine grassland, sub-alpine grassland, woodland fringe, and dry
23                   grassland were identified as the most  sensitive communities.
                     9.4.7.3    Microbes

24                   Several methods have been used to study microbial composition changes associated with
25                   elevated O3. Phospholipid fatty acid (PLFA) analysis is widely used to determine
26                   whether O3 elicits an overall effect on microbial community composition. However,
27                   since PLFA markers cover a broad range of different fungi, resolution of this method
28                   may be not fine enough to detect small changes in the composition of fungal
29                   communities. Methods, such as microscopic analyses and polymerase chain reaction-
30                   denaturing gradient gel electrophoresis (PCR-DGGE), have better resolution to
31                   specifically analyze the fungal community composition. The resolution differences
32                   among those methods needs to be considered when assessing the O3 impact on microbial
33                   community composition.
      Draft - Do Not Cite or Quote                       9-84                                September 2011

-------
 1                  Kanerva et al. (2008) found that elevated O3 (40-50 ppb) decreased total, bacterial,
 2                  actinobacterial and fungal PLFA biomass values as well as fungal:bacterial PLFA
 3                  biomass ratio in their meadow mesocosms in south-western Finland. The relative
 4                  proportions of individual PLFAs between the control and elevated O3 treatments were
 5                  significantly different, suggesting that O3 modified the structure of the microbial
 6                  community. Morsky et al. (2008) exposed boreal peatland microcosms to elevated O3,
 7                  with growing season AOT40 of 20.8-35.3 ppm-h, in an open-air O3 exposure field in
 8                  Central Finland. They also found that microbial composition was altered after three
 9                  growing seasons with O3 fumigation, as measured by PLFA. Ozone tended to increase
10                  the presence of Gram-positive bacteria and the biomass of fungi in the peatland
11                  microcosms. Ozone also resulted in higher microbial biomass, which co-occurred with
12                  the increases in concentrations of organic acids and leaf density of sedges (Morsky et al..
13                  2008). In a lysimeter experiment in Germany, O3 was found to alter the PLFA profiles in
14                  the upper 0-20 cm rhizosphere soil of European beech. Elevated O3 reduced bacterial
15                  abundance but had no detectable effect on fungal abundance (Pritsch et al.. 2009). Using
16                  microscopic analyses, Kasurinen et al. (2005) found that elevated O3, with 5 or 6 months
17                  of AOT40 of 20.6-30.9 ppm-h, decreased the proportions of black and liver-brown
18                  mycorrhizas and increased that of light brown/orange mycorrhizas. In an herbaceous
19                  plant study, SSCP (single-strand conformation polymorphism) profiles indicated that O3
20                  stress (about 75 ppb) had a very small effect on the structural diversity of the bacterial
21                  community in rhizospheres (Dohrmann and Tebbe. 2005). At the Aspen FACE site, O3
22                  had no significant effect on fungal relative abundance, as indicated by PLFA profile.
23                  However, elevated O3  altered fungal community composition, according to the
24                  identification of 39 fungal taxonomic units from soil using polymerase chain reaction-
25                  denaturing gradient gel electrophoresis (PCR-DGGE) (Chung et al.. 2006). In another
26                  study at Aspen FACE, phylogenetic analysis suggested that O3 exposure altered
27                  agaricomycete community. The ectomycorrhizal communities developing under elevated
28                  O3 had higher proportions of Cortinarius and Inocybe species, and lower proportions of
29                  Laccaria and Tomentella (Edwards and Zak. 2011). Ozone was found to change
30                  microbial community composition in an agricultural system. Chen et al. (201 Ob) found
31                  elevated O3 (100-150 ppb) had significant effects on soil microbial composition
32                  expressed as PLFA percentage in a rice paddy in China.
                    9.4.7.4   Summary

33                  In the 2006 O3 AQCD, the impact of O3 exposure on species competition and community
34                  composition was assessed. Ozone was found to cause a significant decline in ponderosa
3 5                  and Jeffrey pine in the San Bernardino Mountains in southern California. Ozone exposure
      Draft - Do Not Cite or Quote                       9-85                                September 2011

-------
 1                  also tended to shift the grass-legume mixtures in favor of grass species (U.S. EPA.
 2                  20061)). Since the 2006 O3 AQCD, more evidence has shown that O3 exposure changed
 3                  the competitive interactions and could lead to loss of O3 sensitive species or genotypes.
 4                  Studies at plant level found that the severity of O3 damage on growth, reproduction and
 5                  foliar injury varied among species, which provided the biological plausibility for the
 6                  alteration of community composition. Additionally, research since the last review has
 7                  shown that O3 can alter community composition and diversity of soil microbial
 8                  communities.

 9                  The decline of conifer forests under O3 exposure was continually observed in several
10                  regions. Ozone damage was believed to be an important causal factor in the dramatic
11                  decline of sacred fir in the valley of Mexico (de Lourdes de Bauer and Hernandez-
12                  Tejeda. 2007). as well as cembran pine in southern France and  Carpathian Mountains
13                  ("Wieser et al.. 2006). Results from the Aspen FACE site indicated that O3 could alter
14                  community composition of broadleaf forests as well. At the Aspen FACE site, O3
15                  reduced growth and increased mortality of a sensitive aspen clone, while the O3 tolerant
16                  clone emerged as the dominant clone in the pure aspen community. In the mixed aspen-
17                  birch and aspen-maple communities, O3 reduced the competitive capacity of aspen
18                  compared to birch and maple (Kubiske et al.. 2007).

19                  The tendency for O3-exposure to shift the biomass of grass-legume mixtures in favor of
20                  grass species, was reported in the 2006 O3 AQCD and has been generally confirmed by
21                  recent studies. However, in a high elevation mature/species-rich grass-legume pasture, O3
22                  fumigation showed no significant impact on community composition (Bassin et al..
23                  2007b).

24                  Ozone exposure not only altered community composition of plant species, but also
25                  microorganisms. The  shift in community composition of bacteria and fungi has been
26                  observed in both natural and agricultural ecosystems, although  no general patterns could
27                  be identified OCanerva et al.. 2008; Morsky et al.. 2008; Kasurinen et al.. 2005).
28                  The evidence is sufficient to conclude that there  is likely a causal  relationship
29                  between O3 exposure and the alteration of community composition.
            9.4.8   Factors that Modify Functional and Growth Response

30                  Many biotic and abiotic factors, including insects, pathogens, root microbes and fungi,
31                  temperature, water and nutrient availability, and other air pollutants, as well as elevated
32                  CO2, influence or alter plant response to O3. These modifying factors were
33                  comprehensively reviewed in AX9.3 of the 2006 O3 AQCD and thus, this section serves
      Draft - Do Not Cite or Quote                       9-86                                September 2011

-------
 1                   mainly as a brief summary of the previous findings. A limited number of new studies
 2                   published since the 2006 O3 AQCD add to our understanding of the role of these
 3                   interactions in modifying O3-induced plant responses. Many of these modifying factors
 4                   and interactions are integrated into discussions elsewhere in this chapter and the reader is
 5                   directed to those sections.
                     9.4.8.1    Genetics

 6                   It is well known that species vary greatly in their responsiveness to O3. Even within a
 7                   given species, individual genotypes or populations can also vary significantly with
 8                   respect to O3 sensitivity (U.S. EPA. 2006b). Therefore, caution should be taken when
 9                   considering a species' degree of sensitivity to O3. Plant response to O3 is determined by
10                   genes that are directly related to oxidant stress and to an unknown number of genes that
11                   are not specifically related to oxidants, but instead control leaf and cell wall thickness,
12                   stomatal conductance, and the internal architecture of the air spaces. It is rarely the case
13                   that single genes are responsible for O3 tolerance. Studies using molecular biological
14                   tools and transgenic plants have positively verified the role of various genes and gene
15                   products in O3 tolerance and are continuing to increase the understanding of O3 toxicity
16                   and differences in O3 sensitivity. See  Section 9.3.3.2 of this document for a discussion of
17                   recent studies related to gene expression changes in response to O3.
                     9.4.8.2    Environmental Biological Factors

18                   As stated in the 2006 O3 AQCD, the biological factors within the plant's environment
19                   that may influence its response to O3 encompass insects and other animal pests, diseases,
20                   weeds, and other competing plant species. Ozone may influence the severity of a disease
21                   or infestation by a pest or weed, either by direct effects on the causal species, or
22                   indirectly by affecting the host, or both. In addition, the interaction between O3, a plant,
23                   and a pest, pathogen, or weed may influence the response of the target host species to O3
24                   (U.S. EPA. 2006b). Several recent studies on the effects of O3 on insects via their
25                   interactions with plants are discussed in Section 9.4.9.1. In addition, O3 has also been
26                   shown to alter soil fauna communities (Section 9.4.9.2).

27                   In contrast to detrimental biological interactions, there are mutually beneficial
28                   relationships or symbioses involving higher plants and bacteria or fungi. These include
29                   (1) the nitrogen-fixing species Rhizobium and Frankia that nodulate the roots of legumes
30                   and alder and (2) the mycorrhizae that infect the roots of many crop and tree species, all
      Draft - Do Not Cite or Quote                       9-87                                 September 2011

-------
 1                   of which may be affected by exposure of the host plants to O3. Some discussion of
 2                   mycorrhizae can be found in Section 9.4.6.

 3                   In addition to the interactions involving animal pests, O3 also has indirect effects on
 4                   higher herbivorous animals, e.g., livestock, due to O3-induced changes in feed quality.
 5                   Recent studies on the effects of O3 on nutritive quality of plants are discussed in Sections
 6                   9.4.4.2.

 7                   Intra- and interspecific competition are also important factors in determining vegetation
 8                   response to O3. Plant competition involves the ability of individual plants to acquire the
 9                   environmental resources needed for growth and development: light, water, nutrients, and
10                   space. Intraspecific competition involves individuals of the same species, typically in
11                   monoculture crop situations, while interspecific competition refers to the interference
12                   exerted by individuals of different species on each other when they are in a mixed
13                   culture. This topic was previously reviewed in AX9.3.3.4 of the 2006 O3 AQCD. Recent
14                   studies on competition and its implications for community composition are discussed in
15                   Section 9.4.7.
                     9.4.8.3    Physical Factors

16                   Physical or abiotic factors play a large role in modifying plant response to O3, and have
17                   been extensively discussed in previous O3 AQCDs. This section summarizes those
18                   findings as well as recent studies published since the last review.

19                   Although some studies have indicated that O3 impact significantly increases with
20                   increased ambient temperature (Ball et al.. 2000; Mills et al.. 2000). other studies have
21                   indicated that temperature has little effect (Balls et al.. 1996; Fredericksen et al.. 1996). A
22                   recent study by Riikonen et al. (2009) at the Ruohoniemi open air exposure field in
23                   Kuopio, Finland found that the effects of temperature and O3  on total leaf area and
24                   photosynthesis of Betulapendula were counteractive. Elevated O3 reduced the saplings'
25                   ability to utilize the warmer growth environment by increasing the stomatal limitation for
26                   photosynthesis and by reducing the redox state of ascorbate in the apoplast in the
27                   combination treatment as compared to temperature alone (Riikonen et al.. 2009).

28                   Temperature affects the rates of all physiological processes based on enzyme catalysis
29                   and diffusion; each process and overall growth (the integral of all  processes) has a
30                   distinct optimal temperature range. It is important to note that a plant's response to
31                   changes in temperature will depend on whether it is growing near its optimum
32                   temperature for growth or near its maximum temperature (Rowland-Bamford. 2000).
33                   However, temperature is very likely an important variable affecting plant O3  response in
      Draft - Do Not Cite or Quote                       9-88                                September 2011

-------
 1                   the presence of the elevated CO2 levels contributing to global climate change. In contrast,
 2                   some evidence suggests that O3 exposure sensitizes plants to low temperature stress
 3                   (Colls and Unsworth. 1992) and, also, that O3 decreases below-ground carbohydrate
 4                   reserves, which may lead to responses in perennial species ranging from rapid demise to
 5                   impaired growth in subsequent seasons (i.e., carry-over effects) (Andersen et al., 1997).

 6                   Light, a component of the plant's physical environment, is an essential "resource" of
 7                   energy content that drives photosynthesis and C assimilation. It has been suggested that
 8                   increased light intensity may increase the O3 sensitivity of light-tolerant species while
 9                   decreasing that of shade-tolerant species, but this appears to be an oversimplification with
10                   many exceptions. Several studies suggest that the interaction between O3 sensitivity and
11                   light environment is complicated by the developmental stage as well as the light
12                   environment of individual leaves in the canopy (Kitao et al., 2009; Topaet al., 2001;
13                   Chappelkaand Samuelson.  1998).

14                   Although the relative humidity of the ambient air has generally been found to increase the
15                   effects of O3 by increasing stomatal conductance (thereby increasing O3 flux into the
16                   leaves), abundant evidence also indicates that the ready availability of soil  moisture
17                   results in greater O3 sensitivity (Mills. 2002). The partial "protection" against the effects
18                   of O3 afforded by drought has been observed in field experiments (Low et  al., 2006) and
19                   modeled in computer simulations (Broadmeadow and Jackson. 2000). Conversely,
20                   drought may exacerbate the effects of O3 on plants (Pollastrini et al., 2010; Grulke et al.,
21                   2003b). There is also some evidence that O3 can predispose plants to drought stress
22                   (Maier-Maercker. 1998). Hence, the nature of the response is largely species-specific and
23                   will depend to some extent upon the sequence in which the stressors occur.
                     9.4.8.4    Interactions with other Pollutants

                     Ozone-Nitrogen Interactions
24                   Elevated O3 exposure and N deposition often co-occur. However, the interactions of O3
25                   exposure and N deposition on vegetation are complex and less well understood compared
26                   to their independent effects. Consistent with the conclusion of the 2006 O3 AQCD, the
27                   limited number of studies published since the last review indicated that the interactive
28                   effects of N and O3 varied among species and ecosystems (Table 9-8). To better
29                   understand these interactions in ecosystems across the U.S., more information is needed
30                   considering combined O3 exposure and N deposition related effects.
31                   Nitrogen deposition could stimulate relative growth rate (RGR), and lead to increased
32                   stomatal conductance. Therefore, plants might become more susceptible to O3 exposure.

      Draft - Do Not Cite or Quote                       9-89                                September 2011

-------
 1                   Alternatively, N deposition may increase the availability of photosynthates for use in
 2                   detoxification and plants could become more tolerant to O3 (Bassin et al.. 2007a). Only a
 3                   few recent studies have investigated the interactive effects of O3 and N in the U.S. Grulke
 4                   et al. (2005) measured stomatal conductance of California black oak (Quercus kelloggii)
 5                   at a long-term N-enrichment site located in the San Bernardino Mountains, which is
 6                   accompanied by high O3 exposure (80 ppb, 24-h avg. over a six month growing season).
 7                   The authors found that N amendment led to poor stomatal control in full sun in
 8                   midsummer of the average precipitation years, but enhanced stomatal control in shade
 9                   leaves of California black oak. In an OTC  study, Handley and  Grulke (2008) found that
10                   O3 lowered photosynthetic ability and water-use efficiency, and increased leaf chlorosis
11                   and necrosis of California black oak. Nitrogen fertilization tended to reduce plant
12                   sensitivity to O3 exposure; however, the interaction was not statistically significant.

13                   Studies conducted outside the U.S. are also summarized in Table 9-8. Generally, the
14                   responses were species specific. The  O3-induced reduction in photosynthetic rate and
15                   biomass loss were greater in the relatively  high N treatment for watermelon (Citrillus
16                   tenants) (Calatayud et al., 2006) and Japanese beech (Fagus crenata) seedlings
17                   (Yamaguchi et al.. 2007). However, there was no significant interactive effect of O3 and
18                   N on biomass production for Quercus serrata seedlings (Watanabe et al.. 2007). young
19                   Norway spruce (Picea abies) trees (Thomas et al.. 2005). and young European beech
20                   (Fagus sylvatica) trees (Thomas et al.. 2006).
      Draft - Do Not Cite or Quote                       9-90                                September 2011

-------
    Table 9-8     Response of plants to the interactive effects of elevated ozone
                   exposure and N enrichment
Site
San
Bernardino
Mountains,
U.S.
San
Bernardino
Mountains,
U.S.
Switzerland
Switzerland
Switzerland
Switzerland
Switzerland
Switzerland
Spain
Spain
Japan
Japan
Species
California black
oak (Quercus
kelloggii)
California black
oak (Quercus
kelloggii)
Spruce trees
(P/cea abies)
Beech trees
(Fagus sylvatica)
Alpine pasture
Alpine pasture
Alpine pasture
Alpine pasture
Watermelon
(Citrillus tenants)
Trifolium striatum
Japanese beech
seedlings (Fagus
crenafa)
Quercus serrata
seedlings
Ozone exposure
SOppb
0,75, and 1 SOppb
Filtered (19.4-28.1 ppb);
ambient (37.6-47.4 ppb)
Filtered (19.4-28.1 ppb);
ambient (37.6-47.4 ppb)
Ambient (AOT40 of 11.1-
12.6ppm-h); 1.2 ambient
(AOT40 of 15.2 -29.5 ppm-
h) and 1 .6 ambient (28.4-
64.9 ppm-h)
Ambient (AOT40 of 11.1-
12.6 ppm-h); 1.2 ambient
(AOT40 of 15.2-29.5 ppm-
h) and 1 .6 ambient (28.4-
64.9 ppm-h)
Ambient (AOT40 of 11.1-
12.6 ppm-h); 1.2 ambient
(AOT40 of 15.2-29.5 ppm-
h) and 1 .6 ambient (28.4-
64.9 ppm-h)
Ambient (AOT40 of 11.1-
12.6 ppm-h); 1.2 ambient
(AOT40 of 15.2-29.5 ppm-
h) and 1 .6 ambient (28.4-
64.9 ppm-h)
03free (AOT40 of 0 ppm-
h), ambient (AOT40 of 5.1-
6.3 ppm-h) and elevated
03(AOT40 of 32.5-35.6
ppm-h)
Filtered (24-h avg. of 8-22
ppb); ambient (29-34 ppb),
elevated 0 3 (35-56 ppb)
Filtered (24-h avg. of 10.3-
13.2 ppb); ambient (42.0-
43.3 ppb), 1.5 ambient
(62.6-63.9 ppb) and 2.0
ambient (82.7-84.7 ppb)
Filtered (24-h avg. of 10.3-
13.2 ppb); ambient (42.0-
43.3 ppb), 1.5 ambient
(62.6-63.9 ppb) and 2.0
ambient (82.7-84.7 ppb)
N addition
0, and 50 kg N/
ha/yr
0, and 50 kg N/
ha/yr
0, 20, 40 and 80
kg N/ ha/yr
0, 20, 40 and
80 kg N/ ha/yr
0,5, 10' 25, 50
kg N/ ha/yr
0,5,10,25,50
kg N/ha/yr
0,5, 10' 25, 50
kg N/ ha/yr
0,5, 10' 25, 50
kg N/ ha/yr
140, 280, and
436 kg N/ ha/yr
10, 30, and 60
kg N/ ha/yr
0, 20 and 50 kg
N/ ha/yr
0, 20 and 50 kg
N/ ha/yr
Responses
N-amended trees had lower late
summer C gain and greater foliar
chlorosis in the drought year, and poor
stomatal control and lower leaf water
use efficiency and in midsummer of the
average precipitation year.
N fertilization tended to reduce plant
sensitivity to 03 exposure; however
the interaction was not statistically
significant.
Higher N levels alleviated the negative
impact of 03 on root starch concentrations
N addition amplified the negative
effects of 0 3 on leaf area and shoot
elongation.
The positive effects of N addition on
canopy greenness were counteracted
by accelerated leaf senescence in the
highest 03 treatment.
Only a small number of species
showed significant 03 and N
interactive effects on leaf chlorophyll
concentration, leaf weight and change
in 180, and the patterns were not
consistent.
The positive effects of N addition on
canopy greenness were counteracted
by accelerated leaf senescence in the
highest 03 treatment.
Highest N addition resulted in carbon
loss, but there was no interaction
between 03 and N treatments.
High N concentration enhanced the
detrimental effects of 03 on
Chlorophyll a fluorescence
parameters, lipid peroxidation, and the
total yield.
03 reduced total aerial biomass. N
fertilization counterbalanced 03-
induced effects only when plants were
exposed to moderate 03 levels
(ambient) but not under elevated 03
concentrations.
The 03-induced reduction in net
photosynthesis and whole-plant dry
mass were greater in the relatively
high N treatment than that in the low N
treatment.
No significant interactive effects of 03
and N load on the growth and net
photosynthetic rate were detected.
References
Grulke et al. (2005)
Handley andGruIke
(2008)
Thomas et al. (2005)
Thomas et al. (2006)
Bassin et al. (2007b)
Bassin et al. (2009)
Bassin et al. (2007b)
Volketal. (2011)
Calatayud et al. (2006)
Sanz et al. (2007)
Yamaguchi etal.
(2007)
Watanabe etal. (2007)
1
2
Ozone-Carbon Dioxide Interactions

Several decades of research has shown that exposure to elevated CO2 increases
photosynthetic rates (Bernacchi et al.. 2006; Bernacchi et al.. 2005; Tissue et al.. 1999;
    Draft - Do Not Cite or Quote
                            9-91
September 2011

-------
 1                   Tissue et al.. 1997; Will and Ceulemans. 1997). decreases stomatal conductance
 2                   (Ainsworth and Rogers. 2007; Paoletti et al.. 2007; Bernacchi et al.. 2006; Leakey et al..
 3                   2006; Medlyn et al., 2001) and generally increases the growth of plants(McCarthy et al..
 4                   2009; Norby et al.. 2005). This is in contrast to the decrease in photosynthesis and growth
 5                   in many plants that are exposed to elevated O3. The interactive effects on vegetation have
 6                   been the subject of research in the past two decades due to the implications on
 7                   productivity and water use of ecosystems. This area of research was discussed in detail in
 8                   AX9.3.8.1 of the 2006 O3 AQCD and the conclusions made then are still relevant (U.S.
 9                   EPA. 2006b).

10                   The bulk of the available evidence shows that, under the various experimental conditions
11                   used (which almost exclusively employed abrupt or "step" increases in CO2
12                   concentration, as discussed below), increased CO2 levels (ambient + 200 to 400 ppm)
13                   may protect plants from the adverse effects of O3 on growth. This protection may be
14                   afforded in part by CO2 acting together with O3 in inducing stomatal closure, thereby
15                   reducing O3 uptake, and in part by CO2 reducing the negative effects of O3 on Rubisco
16                   and its activity in CO2-fixation. Although both CO2-induced and O3-induced decreases in
17                   stomatal conductance have been observed primarily in short-term studies, recent data
18                   show a long-term and sustained reduction in stomatal conductance under elevated CO2
19                   for a number of species (Ainsworth and Long. 2005; Ellsworth et al.. 2004; Gunderson et
20                   al.. 2002). Instances of increased stomatal conductance have also been observed in
21                   response to O3 exposure, suggesting partial stomatal dysfunction after extended periods
22                   of exposure (Paoletti and Grulke. 2010; Grulke et al.. 2007a; Maier-Maercker. 1998).

23                   Important caveats must be raised with regard to the findings presented in published
24                   research. The first caveat concerns the distinctly different natures of the exposures to O3
25                   and CO2 experienced by plants in the field. Changes in the ambient concentrations of
26                   these gases have very different dynamics. In the context of climate change, CO2 levels
27                   increase relatively slowly (globally 2 ppm/year) and may change little over several
28                   seasons of growth. On the other hand, O3 presents a fluctuating stressor with
29                   considerable hour-to-hour, day-to-day and regional variability (Polle and Pell. 1999).
30                   Almost all of the evidence presented comes from experimentation involving plants
31                   subjected to an abrupt step increase to a higher, steady CO2  concentration. In contrast, the
32                   O3 exposure  concentrations usually varied from day to day.  Luo and  Reynolds (1999).
33                   Hui et al. (2002). and Luo (2001) noted the difficulties in predicting the likely effects of a
34                   gradual CO2 increase from experiments involving a step increase  or those using a range
35                   of CO2 concentrations. It is also important to note that the levels of elevated  CO2 in
36                   many of the studies will not be experienced in the field for 30 or 40 years, but elevated
37                   levels of O3 can occur presently in several areas of the U.S.  Therefore, the CO2  * O3
38                   interaction studies may be less relevant for current ambient conditions.
      Draft - Do Not Cite or Quote                       9-92                                September 2011

-------
 1                   Another caveat concerns the interactions of O3 and CO2 with other climatic variables,
 2                   such as temperature and precipitation. In light of the key role played by temperature in
 3                   regulating physiological processes and modifying plant response to increased CO2 levels
 4                   (Morison and Lawlor. 1999; Long. 1991) and the knowledge that relatively modest
 5                   increases in temperature may lead to dramatic consequences in terms of plant
 6                   development (Lawlor. 1998). it is important to consider that studying CO2 and O3
 7                   interactions alone may not create a complete understanding of effects on plants under
 8                   future climate change.
            9.4.9   Insects and Other Wildlife
                     9.4.9.1    Insects

 9                   Insects may respond indirectly to changes to plants (i.e., increased reactive oxygen
10                   species, altered phytochemistry, altered nutrient content) that occur under elevated O3
11                   conditions, or O3 can have a direct effect on insect performance (Menendez et al.. 2009).
12                   Effects of O3 on insects occur at the species level (i.e., growth, survival, reproduction,
13                   development, feeding behavior) and at the population and community-level (i.e.,
14                   population growth rate, community composition).  In general, effects of O3 on insects are
15                   highly context- and species-specific (Lindroth. 2010; Bidart-Bouzat and Imeh-Nathaniel.
16                   2008). Furthermore, plant responses to O3 exposure and herbivore attack have been
17                   demonstrated to share signaling pathways, complicating characterization of these
18                   stressors (Lindroth. 2010; Menendez et al., 2010. 2009). Although both species-level and
19                   population and community-level responses to elevated O3 are observed in field and
20                   laboratory studies discussed below, there is  no consensus on how insects respond to
21                   feeding on O3-exposed plants.


                     Species-Level Responses

22                   In considering insect growth, survival and reproduction in elevated O3 conditions, several
23                   studies have indicated an effect while others have found no correlation. The performance
24                   of five herbivore species (three moths and two weevils) was assessed in an OTC
25                   experiment at 2 x ambient concentration (Peltonen et al.. 2010). Growth of larvae of the
26                   Autumnal moth, Epirrita autumna,  was significantly decreased in the O3 treatment while
27                   no effects were observed in the other species. In an aphid oviposition preference study
28                   using birch buds grown in a three year OTC experiment, O3 had neither a stimulatory or
29                   deterring effect on egg-laying (Peltonen et al.. 2006). Furthermore, changes in birch bud
30                   phenolic compounds associated with the doubled ambient concentrations of O3 did not

      Draft - Do Not Cite or Quote                      9-93                                September 2011

-------
 1                   correlate with changes in aphid oviposition (Peltonen et al., 2006). Reproduction in
 2                   Popilliajaponica, that were fed soybeans and grown under elevated O3 appeared to be
 3                   unaffected (O'Neill et al., 2008). In a meta-analysis of effects of elevated O3 on 22
 4                   species of trees and 10 species of insects, the rates of survival, reproduction and food
 5                   consumption were typically unaffected while development times were reduced and pupal
 6                   masses were increased (Valkama et al.. 2007).

 7                   At the Aspen FACE site insect performance under elevated (50-60 ppb) O3 conditions
 8                   (approximately 1.5 x background ambient levels of 30-40 ppb O3) have been considered
 9                   for several species. Cumulative fecundity of aphids (Cepegillettea betulaefoliae), that
10                   were reared on  O3-exposed paper birch (Betula papyrifera) trees, was lower than aphids
11                   from control plots (Awmack et al.. 2004). No  effects on growth, development, adult
12                   weight, embryo number and birth weight of newborn nymphs were observed. In a study
13                   conducted using three aspen genotypes, performance of the aspen beetle (Chrysomela
14                   crochi) decreased across all parameters measured (development time, adult mass and
15                   survivorship) under elevated O3 (Vigue and Lindroth. 2010). There was an increase in the
16                   development time of male and female aspen beetle larvae although the percentages varied
17                   across genotypes. Decreased beetle adult mass and survivorship was observed across all
18                   genotypes under elevated O3 conditions. Another study from the Aspen FACE site, did
19                   not find any significant effects of elevated O3  on performance (longevity, fecundity,
20                   abundance) of the invasive weevil  (Polydrusus senceus) (Hillstrom et al.. 201 Ob).

21                   Since the 2006  O3  AQCD, several studies  have considered the effect of elevated O3 on
22                   feeding behavior of insects. In a feeding preference study, the common leaf weevil
23                   (Phyllobius pyri) consumed significantly more leaf discs from one aspen clone when
24                   compared to a second clone under  ambient air conditions (Freiwald et al., 2008). In a
25                   moderately elevated O3  environment (1.5 x ambient), this preference for a certain aspen
26                   clone was less evident, however, leaves from O3-exposed trees were significantly
27                   preferred to leaves grown under ambient conditions. Soybeans grown under enriched O3
28                   had significantly less loss of leaf tissue to herbivory in August compared to earlier in the
29                   growing season (July) when herbivory was not affected (Hamilton et al.. 2005). Other
30                   plant-herbivore interactions have shown no effects of elevated O3 on feeding. Feeding
31                   behavior of Japanese beetles (P. japonicd) appeared to be unchanged when beetles were
32                   fed soybean leaves grown under elevated O3 conditions (O'Neill et al.. 2008). At the
33                   Aspen FACE site, feeding by the invasive  weevil (Polydrusus senceus), as measured by
34                   leaf area consumption, was not significantly different between foliage that was grown
35                   under elevated O3 versus ambient conditions (Hillstrom et al.. 2010b).
      Draft - Do Not Cite or Quote                      9-94                                September 2011

-------
                    Population-Level and Community-Level Responses

 1                  Recent data on insects provide evidence of population-level and community-level
 2                  responses to O3. Elevated levels of O3 can affect plant phytochemistry and nutrient
 3                  content which in turn can alter population density and structure of the associated
 4                  herbivorous insect communities and impact ecosystem processes (Cornelissen. 2011;
 5                  Lindroth. 2010). In 72-hour exposures to elevated O3, mean relative growth rate of the
 6                  aphid Diuraphis noxia increased with ozone concentration suggesting that more rapid
 7                  population growth may occur when atmospheric O3 is elevated (Summers et al., 1994,
 8                  735955). In a long-term study of elevated O3 on herbivore performance at the Aspen
 9                  FACE site, individual performance and population-level effects of the aphid
10                  C. betulaefoliae were assessed. Elevated O3  levels had a strong positive effect on the
11                  population growth rates of the aphids;  although effects were not detected by measuring
12                  growth, development, adult weight, embryo number or birth weight of newborn nymphs
13                  (Awmack et al.. 2004). Conversely, a lower rate of population growth was observed in
14                  aphids previously exposed to O3 in an OTC (Menendez et al., 2010). No direct effects of
15                  O3 were observed; however, nymphs born from adults exposed to and feeding on O3
16                  exposed plants were less capable of infesting new plants when compared to nymphs in
17                  the control plots (Menendez et al.. 2010). Elevated O3 reduced total arthropod abundance
18                  by 17% at Aspen FACE, largely as a result of the negative effects on parasitoids,
19                  although phloem-feeding insects may benefit (Hillstrom and Lindroth. 2008). Herbivore
20                  communities affected by O3 and N were  sampled along an air pollution gradient in the
21                  Los Angeles basin (Jones and Paine. 2006). Abundance, diversity, and richness of
22                  herbivores were not affected. However, a shift in community structure, from phloem-
23                  feeding to chewing dominated communities, was observed along the gradient. No
24                  consistent effect of elevated O3 on herbivory or insect population size was detected at
25                  SoyFACE (O'Neill etal.. 2010: Dermodv et al.. 2008).

26                  Evidence of modification of insect populations and communities in response to  elevated
27                  O3 includes genotypic and phenotypic changes. In a study conducted at the Aspen FACE
28                  site, elevated O3 altered the genotype frequencies of the pea aphid (Acyrthosiphon pi sum)
29                  grown on red clover (Trifoliumpratense) over multiple generations (Mondor et al..
30                  2005). Aphid color was used to distinguish between the two genotypes. Ozone increased
31                  the genotypic frequencies of pink-morph:green-morph aphids from 2:1 to 9:1, and
32                  depressed wing-induction responses more strongly in the pink than the green genotype
33                  (Mondor et al.. 2005). Growth and development of individual green and pink aphids
34                  reared as a single genotype or mixed genotypes were unaffected by elevated O3 (Mondor
35                  et al.. 2010). However, growth of pea aphid populations is not readily predictable using
36                  individual growth rates.
      Draft - Do Not Cite or Quote                      9-95                                September 2011

-------
                     9.4.9.2   Wildlife

                     Herpetofauna

 1                   Since the 2006 O3 AQCD, direct effects of O3 exposure including physiological changes
 2                   and alterations of ecologically important behaviors such as feeding and thermoregulation
 3                   have been observed in wildlife. These studies have been conducted in limited laboratory
 4                   exposures, and the levels of O3 treatment (e.g. 0.2-0.8 ppm) were often unrealistically
 5                   higher than the ambient levels. Amphibians may be especially vulnerable to airborne
 6                   oxidants due to the significant gas exchange that occurs across the skin (Andrews et al..
 7                   2008; Dohm et al., 2008). Exposure to 0.2 ppm to 0.8 ppm O3 for 4 hours resulted in a
 8                   decrease of oxygen consumption and depressed lung ventilation in the California tree
 9                   frog Pseudacris cadaverina (Mautz and Dohm. 2004). Following a single 4-h exposure to
10                   O3, reduced pulmonary macrophage phagocytosis was observed at 1  and 24 hours post
11                   exposure in the marine toad (Bufo marinus) indicating an effect on immune system
12                   function (Dohm et al.. 2005). There was no difference in macrophage function at
13                   48 hours post exposure in exposed and control individuals.

14                   Behavioral effects of O3 observed in amphibians include responses to minimize the
15                   surface area of the body exposed to the air and a decrease in feeding  rates (Dohm et al..
16                   2008; Mautz and Dohm. 2004). The adoption of a low-profile "water conservation
17                   posture" during O3 exposure was observed in experiments with the California tree frog
18                   (Mautz and Dohm. 2004). Marine toads, Bufo marinus, exposed to 0.06 (iL/L O3 for
19                   4 hours ate significantly fewer mealworms at 1 hour and 48 hours post exposure than
20                   control toads (Dohm et al.. 2008). In the same study, escape/exploratory behavior as
21                   measured by total distance moved was not adversely affected in the O3-exposed
22                   individuals as compared to the controls (Dohmet al.. 2008).

23                   Water balance and thermal preference in herpetofauna are altered with elevated O3.
24                   Marine toads exposed to 0.8 ppm O3 for 4 hours exhibited behavioral hypothermia when
25                   temperature selection in the toads was assessed at I, 24 and 48 hours post exposure
26                   (Dohm et al.. 2001). Ozone-exposed individuals lost almost 5g more body mass on
27                   average than controls due to evaporative water loss. At 24 hours after exposure, the
28                   individuals that had lost significant body mass selected lower body temperatures(Dohm
29                   et al.. 2001). Behavioral hypothermia was also observed in reptiles following 4-h
30                   exposures to 0.6 ppm O3.  Exposure of the Western Fence Lizard (Sceloporus
31                   occidentalis) at 25°C induced behavioral hypothermia that recovered to control
32                   temperatures by 24 hours  (Mautz and Dohm. 2004). The behavioral hypothermic
33                   response persisted in lizards exposed to O3 at 35°C at 24 hours post exposure resulting in
34                   a mean body temperature  of 3.3°C over controls.
      Draft - Do Not Cite or Quote                      9-96                                September 2011

-------
                     Soil Fauna Communities

 1                   Ozone has also been shown to alter soil fauna communities (Meehan et al.. 2010;
 2                   Kasurinen et al., 2007; Loranger et al.. 2004). Abundance of Acari (mites and ticks)
 3                   decreased by 47% under elevated O3 at Aspen FACE site, probably due to the higher
 4                   secondary metabolites and lower N concentrations in litter and foliage under elevated O3
 5                   (Loranger et al.. 2004). In another study from the Aspen FACE site, leaf litter collected
 6                   from aspen grown under elevated O3 conditions were higher in fiber and lignin
 7                   concentrations than trees grown under ambient conditions. These chemical characteristics
 8                   of the leaves were associated with increased springtail population growth following
 9                   10 weeks in a laboratory microcosm (Meehan et al.. 2010). Consumption rates of
10                   earthworms fed on leaf litter for 6 weeks from trees grown under elevated O3 conditions
11                   and ambient air did not vary significantly between treatments (Meehan et al.. 2010). In
12                   another study on juvenile earthworms Lumbricus terrestris, individual growth was
13                   reduced when worms were fed high-O3 birch litter from trees exposed for three years to
14                   elevated O3 in an OTC system (Kasurinen et al.. 2007). In the same study no significant
15                   growth or mortality effects were observed in isopods.
                     9.4.9.3    Indirect Effects on Wildlife

16                   In addition to the direct effects of O3 exposure on physiological and behavioral endpoints
17                   observed in the laboratory, there are indirect effects to wildlife. These effects include
18                   changes in biomass and nutritive quality of O3-exposed plants (reviewed in Section 9.4.4)
19                   that are consumed by wildlife. Reduced digestibility of O3-exposed plants may alter
20                   dietary intake and foraging strategies in herbivores. In a study using native highbush
21                   blackberry (Rubus argutus) relative feed value of the plants decreased in bushes exposed
22                   to double ambient concentrations of O3 (Ditchkoff et al.. 2009). Indirect effects of
23                   elevated O3 on wildlife include changes in chemical signaling important in ecological
24                   interactions reviewed below.


                     Chemical  Signaling in Ecological Interactions

25                   Ozone has been shown to degrade or alter biogenic VOC signals important to ecological
26                   interactions including; (1) attraction of pollinators and seed dispersers; (2) defense
27                   against herbivory; and (3) predator-prey interactions (Pinto et al.. 2010; McFrederick et
28                   al.. 2009; Yuan et al.. 2009; Pinto et al.. 2007a; Pinto et al.. 2007b). Each signal released
29                   by emitters has an atmospheric lifetime and a unique chemical signature comprised of
30                   different ratios of individual hydrocarbons that is susceptible to atmospheric oxidants
31                   such as O3 (Yuan et al.. 2009; Wright et al.. 2005). Under elevated O3 conditions, these


      Draft - Do Not Cite or Quote                       9-97                                September 2011

-------
 1                   olfactory cues may travel shorter distances before losing their specificity (McFrederick et
 2                   al.. 2009; McFrederick et al.. 2008). Additional non-phytogenic VOC-mediated
 3                   interrelationships with the potential to be modified by O3 include territorial marking,
 4                   pheromones for attraction of mates and various social interactions including scent trails,
 5                   nestmate recognition and signals involved in aggregation behaviors (McFrederick et al..
 6                   2009). For example, the alcohols, ketones and aldehydes comprising sex pheromones in
 7                   moths could be especially vulnerable to degradation by O3, since some males travel >100
 8                   m to find mates (Carde and Haynes. 2004). In general, effects of O3 on scent-mediated
 9                   ecological interactions are highly context- and species-specific (Lindroth. 2010; Bidart-
10                   Bouzat and Imeh-Nathaniel. 2008).


                     Pollination and Seed Dispersal

11                   Phytogenic VOC's attract pollinators and seed dispersers to flowers and fruits (Dudareva
12                   et al., 2006; Theis and Raguso. 2005). These floral scent trails in plant-insect interactions
13                   may be destroyed or transformed by O3 (McFrederick et al.. 2008). Using a Lagrangian
14                   model, the rate of destruction of phytogenic VOC's was estimated in air parcels at
15                   increasing distance from a source in response to increased regional levels of O3, hydroxyl
16                   and nitrate radicals (McFrederick et al., 2008). Based on the model, the ability of
17                   pollinators to locate highly reactive VOCs from emitting flowers may have decreased
18                   from kilometers during pre-industrial times to  <200 m at current ambient conditions
19                   (McFrederick et al.. 2008). Scents that travel shorter distances (0-10 m) are less
20                   susceptible to air  pollutants, while highly reactive scents that travel longer distances (10
21                   to 100's of meters), are at a higher risk for degradation (McFrederick et al.. 2009). For
22                   example, male euglossine bees can detect bait stations from a distance of at least one
23                   kilometer (Dobson. 1994).


                     Defense Against Herbivory

24                   Ozone can alter the chemical signature of VOCs emitted by plants and these VOCs are
25                   subsequently detected by herbivores (Blande et al.. 2010; Iriti and Faoro. 2009; Pinto et
26                   al.. 2007a; Vuorinen et al.. 2004; Jackson et al.. 1999; Cannon. 1990). These
27                   modifications can make the plant either more attractive or repellant to phytophagous
28                   insects (Pinto etal.. 2010). For example, under elevated O3, the host plant preference by
29                   forest tent caterpillars increased for birch compared to aspen  (Agrell et al., 2005). Ozone-
30                   induced emissions from red spruce needles were found to repel spruce budworm larvae
31                   (Cannon. 1990). Transcriptional profiles of field grown soybean (Glycine max) grown in
32                   elevated O3 conditions were altered due to herbivory by Japanese beetles. The herbivory
      Draft - Do Not Cite or Quote                       9-98                                 September 2011

-------
 1                   resulted in a higher number of transcripts in the leaves of O3-exposed plants and up-
 2                   regulation of antioxidant metabolism associated with plant defense (Casteel et al.. 2008).

 3                   Ozone may modify signals involved in plant-to-plant interactions and plant defense
 4                   against pathogens (Blandeetal., 2010; Pinto etal., 2010; McFrederick et al., 2009; Yuan
 5                   et al.. 2009). In a recent study with lima beans, 80 ppb O3 degraded several herbivore-
 6                   induced VOCs, reducing the distance over which plant-to-plant signaling occurred
 7                   (Blandeetal.,2010).


                     Predator-Prey  Interactions

 8                   Elevated O3 conditions are associated with disruption of pheromone-mediated
 9                   interactions at higher trophic  levels (e.g., predators and parasitoids of herbivores). In a
10                   study from the Aspen FACE site, predator escape behaviors of the aphid (Chatophorus
11                   stevensis) were enhanced on O3-fumigated aspen trees although the mechanism of this
12                   response remains unknown (Tvlondor et al.. 2004). The predatory mite Phytoseiulus
13                   persimilis can distinguish between the VOC signature of ozonated lima bean plants and
14                   ozonated lima bean plants simultaneously damaged by T. urticae (Vuorinen et al.. 2004)
15                   however, other tritrophic interactions have shown no effect (Pinto et al.. 2007b).

16                   There are few studies that consider host location behaviors of parasites under elevated
17                   O3. In closed chambers fumigated with O3, the searching efficiency and proportion of the
18                   host larval fruit flies parasitized by Asobara tabida, declined when compared to filtered
19                   air controls (Gate etal.. 1995). The host location behavior and  rate of parasitism of the
20                   wasp (Coesiaplutellae) on Plutella xylostella-mfested potted cabbage plants was tested
21                   under ambient and doubled O3 conditions in an open-air fumigation system (Pinto  et al..
22                   2008). The number of wasps found in the field and the percentages of parasitized larvae
23                   were not significantly different from controls under elevated O3.

24                   Elevated O3 has the potential to perturb specialized food-web communication in
25                   transgenic crops. In insect-resistant oilseed rape Brassica napus grown under 100 ppb O3
26                   in a growth chamber, reduced feeding damage by Putella xylostella led to deceased
27                   attraction of the endoparasitoid (Costesia vestalis), however this tritrophic interaction
28                   was influenced by the degree of herbivore feeding (Himanen et al.. 2009a; Himanen et
29                   al.. 2009b). Under chronic O3-exposure, the insect resistance trait BT crylAc in
30                   transgenic B. napus was higher than the control (Himanen et al., 2009c).  There was a
31                   negative relative  growth rate of the Bt target herbivore, P. xylostella, in all O3 treatments.
      Draft - Do Not Cite or Quote                       9-99                                September 2011

-------
                    9.4.9.4   Summary

 1                  New information on O3 effects on insects and other wildlife is limited to a few species
 2                  and there is no consensus on how these organisms respond to elevated O3 Studies
 3                  published since the last review show impacts of elevated O3 on both species-level
 4                  responses (reproduction, growth, feeding behavior) and community and ecosystem-level
 5                  responses (population growth, abundance, shift in community structure) in some insects
 6                  and soil fauna. Changes in ecologically important behaviors such as feeding and
 7                  thermoregulation have recently been observed with O3 exposure in amphibians and
 8                  reptiles, however, these responses occur at concentrations of O3 much higher that
 9                  ambient levels.

10                   New information available since the last review considers the effects  of O3 on chemical
11                  signaling in insect and wildlife interactions. Specifically, studies on O3 effects on
12                  pollination and seed dispersal, defenses against herbivory and predator-prey interactions
13                  all consider the ability of O3 to alter the chemical signature of VOCs emitted during these
14                  pheromone-mediated events. The effects of O3 on chemical signaling  between plants,
15                  herbivores and pollinators as well as interactions between multiple trophic levels  is an
16                  emerging area of study that may result in further elucidation of O3 effects at the species,
17                  community and ecosystem-level.
          9.5    Effects-Based Air Quality Exposure Indices and Dose
                 Modeling
            9.5.1    Introduction

18                   Exposure indices are metrics that quantify exposure as it relates to measured plant
19                   damage (e.g., reduced growth). They are summary measures of monitored ambient O3
20                   concentrations over time, intended to provide a consistent metric for reviewing and
21                   comparing exposure-response  effects obtained from various studies. Such indices may
22                   also provide a basis for developing a biologically-relevant air quality standard for
23                   protecting vegetation and ecosystems. Effects on plant growth and/or yield have been a
24                   major focus of the characterization of O3 impacts on plants for purposes of the air quality
25                   standard setting process (U.S.  EPA. 2007b. 1996e. 1986). The relationship of O3 and
26                   plant responses can be characterized quantitatively as "dose-response" or "exposure-
27                   response." The distinction is in how the pollutant concentration is expressed: "dose" is
28                   the pollutant concentration absorbed by the leaf over some time period, and is very
29                   difficult to measure directly, whereas "exposure" is the ambient air  concentration
      Draft - Do Not Cite or Quote                      9-100                                September 2011

-------
 1                   measured near the plant over some time period, and summarized for that period using an
 2                   index. Exposure indices have been most useful in considering the form of secondary O3
 3                   NAAQS, in large part because they only require ambient air quality data rather than more
 4                   complex indirect calculations of dose to the plant. The attributes of exposure indices that
 5                   are most relevant to plant damage are the weighting of O3 concentrations and the daily
 6                   and seasonal time-periods. Several different types of exposure indices are discussed in
 7                   Section 9.5.2.

 8                   Form a theoretical perspective, a measure of plant O3 uptake or dose from ambient air
 9                   (either rate of uptake or cumulative seasonal uptake) might be a belter predictor of O3
10                   damage to plants than an exposure index  and may be useful in improving risk assessment.
11                   An uptake estimate would have to integrate all those environmental factors that influence
12                   stomatal conductance, including but not limited to temperature, humidity, and soil water
13                   status  (Section 9.5.4). Therefore, uptake values are generally obtained with simulation
14                   models that require knowledge of species- and site-specific values for the variables
15                   mentioned. However, a limitation of modeling dose is that environmental variables are
16                   poorly characterized. In addition, it has also been recognized that O3 detoxification
17                   processes and the temporal dynamics of detoxification must be taken into account in dose
18                   modeling (Heath et al.. 2009) (Section 9.5.4).  Because of this, research has focused
19                   historically on predictors of O3 damage to plants based only on exposure as a summary
20                   measure of monitored ambient pollutant concentration over some integral of time, rather
21                   than dose (U.S. EPA.  1996c: Costa et al.. 1992; Leeetal.. 1988b: U.S. EPA. 1986;
22                   Lefohn and  Benedict. 1982: O'Gara. 1922).
            9.5.2   Description of Exposure Indices Available in the Literature

23                   Mathematical approaches for summarizing ambient air quality information in biologically
24                   meaningful forms for O3 vegetation effects assessment purposes have been explored for
25                   more than 80 years (U.S. EPA. 1996b: O'Gara. 1922). In the context of national standards
26                   that protect for "known or anticipated" effects on many plant species in a variety of
27                   habitats, exposure indices provide a numerical summary of very large numbers of
28                   ambient observations of concentration over extended periods. Like any summary statistic,
29                   exposure indices retain information on some, but not all, characteristics of the original
30                   observations. Several indices have been developed to attempt to incorporate some of the
31                   biological, environmental, and exposure factors that influence the magnitude of the
32                   biological response and contribute to observed variability (Hogsett et al.. 1988). In the
33                   1996 O3 AQCD, the  exposure indices were arranged into five categories; (1) One event,
34                   (2) Mean, (3) Cumulative, (4) Concentration weighted, and (5) Multicomponent, and
35                   were discussed in detail (Lee etal..  1989). Figure 9-9 illustrates how several of the

      Draft - Do Not Cite or Quote                      9-101                                September 2011

-------
1

2
3

4
5

6
indices weight concentration and accumulate exposure. For example, the SUM06 index

(panel a) is a threshold-based approach wherein concentrations below 0.06 ppm are given

a weight of zero and concentrations above 0.06 ppm are given a weight of 1.0 that is

summed usually over 3 to 6 months . The Sigmoid approach (panel b), which is similar to

the W126 index, is a non-threshold approach wherein all concentrations are given a

weight that increases from zero to 1.0 with increasing concentration and summed.
          0.15
Ppm 2ndHDM->
M-7 = 0.05 ppm
) 2 4 6 8
Day


1
                                                                           0.10
                                                                           0.05
                                                                           0.00
       Source: Used with permission from Air and Waste Management Association (Tingevet al.. 1991)
       (a) SUM06: the upper graphic illustrates an episodic exposure profile; the shaded area under some of the peaks illustrates the
     concentrations greater than or equal to 0.06 ppm that are accumulated in the index. The insert shows the concentration weighting (0
     to 1) function. The lower portion of the graphic illustrates how concentration is accumulated over the exposure period, (b) SIGMOID:
     the upper graphic illustrates an episodic exposure profile; the variable shaded area under the peaks illustrates the concentration-
     dependent weights that are accumulated in the index. The insert shows the sigmoid concentration weighting function. This is similar
     to the W126 function. The lower portion of the graphic illustrates how concentration is accumulated over the exposure period, (c)
     second HDM and M-7: the upper graphic illustrates an episodic exposure profile. The lower portion of the graphic illustrates that the
     second HDM considers only a single exposure peak, while the M-7 (average of 7-h daily means) applies a constant exposure value
     over the exposure period.


     Figure 9-9     Diagrammatic representation of several exposure indices

                       illustrating how they weight concentration and accumulate
                       exposure.
                     Various factors with known or suspected bearing on the exposure-response relationship,

                     including concentration, time of day, respite time, frequency of peak occurrence, plant
     Draft - Do Not Cite or Quote
                                 9-102
September 2011

-------
 1                   phenology, predisposition, etc., have been weighted with various functions in a large set
 2                   of indices. The resulting indices were evaluated by ranking them according to the
 3                   goodness-of-fit of a regression model of growth or yield response (Lee etal. 1989). The
 4                   statistical evaluations for each of these indices were completed using growth or yield
 5                   response data from many earlier exposure studies (e.g., NCLAN). This retrospective
 6                   approach was necessary because there were no studies specifically designed  to test the
 7                   goodness of fit of the various indices. The goodness of fit of a set of linear and nonlinear
 8                   models for exposure-response was ranked as various proposed indices were used in turn
 9                   to quantify exposure. This approach provided evidence for the best indices. The results of
10                   retrospective analyses are described below.

11                   Most of the early retrospective studies reporting regression approaches used  data from the
12                   NCLAN program or data from Corvallis, Oregon or California (Costa etal..  1992; Lee et
13                   al.. 1988b: Lefohn et al.. 1988: Musselman et al.. 1988: Lee et al..  1987: U.S. EPA.
14                   1986). These studies were previously reviewed by the EPA (U.S. EPA, 1996c; Costa et
15                   al.. 1992) and were in general agreement that the best fit to the  data resulted  from using
16                   cumulative concentration-weighted exposure indices (e.g. W126, SUM06). Lee et al.
17                   (1987) suggested that exposure indices that included all the 24-h data performed better
18                   than those that  used only 7 hours of data; this was consistent with the conclusions of
19                   Heagle et al. (1987) that plants receiving exposures for an  additional 5-h/day showed
20                   10% greater yield loss than those exposed for 7-h/day. In an analysis using the National
21                   Crop Loss Assessment Network (NCLAN) data, Lee et al. (1988) found several indices
22                   which only cumulated and weighted higher concentrations (e.g., W126, SUM06, SUM08,
23                   and AOT40) performed very well. Amongst this group no  index had consistently better
24                   fits than the other indices across all studies and species (Heagle et al.. 1994b; Lefohn et
25                   al.. 1988; Musselman et al..  1988). Lee et al. (1988) found that adding phenology
26                   weighting to the index somewhat improved the performance of the indices. The "best"
27                   exposure index was a phenologically weighted cumulative index, with sigmoid weighting
28                   on concentration and a gamma weighting function as a surrogate for plant growth stage.
29                   This index provided the best statistical fit when used in the models under consideration,
30                   but it required data on species and site conditions, making  specification of weighting
31                   functions difficult for general use.

32                   Other factors, including predisposition time (Hogsett et al.. 1988; McCool et al..  1988)
33                   and crop development stage (Tingey et al.. 2002; Heagle et al.. 1991) contributed to
34                   variation in the biological response and suggested the need for weighting O3
35                   concentrations  to account for predisposition time and phenology. However, the roles of
36                   predisposition and phenology in plant response vary considerably with species and
37                   environmental  conditions; therefore, specification of a weighting function for general use
38                   in characterizing plant exposure has not been possible.
      Draft - Do Not Cite or Quote                      9-103                                September 2011

-------
 1                   European scientists took a similar approach in developing indices describing growth and
 2                   yield loss in crops and tree seedlings, using OTCs with modified ambient exposures, but
 3                   many fewer species and study locations were employed in the European studies. There is
 4                   evidence from some European studies that a lower (Pleijel et al.. 1997) or higher (Finnan
 5                   et al., 1997; Finnan et al.,  1996) cutoff value in indices with a threshold may provide a
 6                   better statistical fit to the experimental data. Finnan et al. (1997) used seven exposure
 7                   studies of spring wheat to confirm that cumulative exposure indices emphasizing higher
 8                   O3 concentrations were best related to plant response and that cumulative exposure
 9                   indices using weighting functions, including cutoff concentrations, allometric and
10                   sigmoidal, provided a better fit  and that the ranking of these indices differed depending
11                   on the exposure-response model used. Weighting those concentrations associated with
12                   sunshine hours in an attempt to incorporate an element of plant uptake did not improve
13                   the index performance (Finnan  et al.. 1997). A more recent study using data from several
14                   European studies of Norway spruce, analyzed the relationship between relative biomass
15                   accumulation and  several cumulative, weighted indices, including the AOT40 (area over
16                   a threshold of 40ppb) and the SUM06 (Skarby et al.. 2004). All the indices performed
17                   relatively well in regressing biomass and exposure index, with the AOT20 and AOT30
18                   doing slightly better than others (r2 = 0.46-0.47). In another comparative study of four
19                   independent data sets of potato  yield and different cumulative uptake indices with
20                   different cutoff values, a similarly narrow range of r2 was observed (r2 = 0.3-0.4) (Pleijel
21                   et al.. 2004b).

22                   In Europe, the cutoff concentration-weighted index AOT40 was selected in developing
23                   exposure-response relationships based on OTC studies of a limited number of crops and
24                   trees (Grunhage and Jager. 2003). The United Nations Economic Commission for Europe
25                   (UNECE.  1988) adopted the critical levels approach for assessment of O3 risk to
26                   vegetation across Europe. As used by the UNECE, the critical levels are not like the air
27                   quality regulatory  standards used in the U.S., but rather function as planning targets for
28                   reductions in pollutant emissions to protect ecological resources. Critical levels for O3 are
29                   intended to prevent long-term deleterious effects on the most sensitive plant species
30                   under the most sensitive environmental conditions, but not intended to quantify O3
31                   effects. A critical level was defined as "the concentration of pollutant in the atmosphere
32                   above which direct adverse effects on receptors, such as plants, ecosystems, or materials
33                   may occur according to present knowledge" (UNECE. 1988). The nature of the "adverse
34                   effects" was not specified in the original definition, which provided for different levels
35                   for different types of harmful effect (e.g., visible injury or loss of crop yield). There are
36                   also different critical levels for  crops, forests, and semi-natural vegetation. The caveat,
37                   "according to present knowledge" is important because critical levels are not rigid; they
38                   are revised periodically as new  scientific information becomes available. For example,
39                   the original critical level for O3 specified concentrations for three averaging times, but

      Draft - Do Not Cite or Quote                       9-104                                September 2011

-------
 1                   further research and debate led to the current critical level being stated as the cumulative
 2                   exposure (concentration x hours) over a cutoff concentration of 40 ppb (AOT40) (Tuhrer
 3                   etal.. 1997).

 4                   More recently in Europe, a decision was made to work towards a flux-based approach
 5                   (see section 9.5.4) for the critical levels ("Level II"), with the goal of modeling O3 flux-
 6                   effect relationships for three vegetation types: crops, forests, and semi-natural vegetation
 7                   (Grunhage and Jager. 2003). Progress has been made in modeling flux (U.S. EPA. 2006b)
 8                   and the Mapping Manual is being revised (Ashmore et al.. 2004a. b; Grennfelt. 2004;
 9                   Karlsson et al.. 2003). The revisions may include a flux-based approach for three crops:
10                   wheat, potatoes, and cotton. However, because of a lack of flux-response data, a
11                   cumulative, cutoff concentration-based (AOTx) exposure index will remain in use for the
12                   near future for most crops and for forests and semi-natural herbaceous vegetation
13                   (Ashmore et al.. 2004b)

14                   In both the U.S. and Europe, the adequacy of these numerical summaries of exposure in
15                   relating biomass and yield changes have, for the most part, all been evaluated using data
16                   from studies not necessarily designed to compare one index to another (Skarby et al..
17                   2004; Lee etal.. 1989; Lefohnetal.. 1988). Very few studies in the U.S. have addressed
18                   this issue since the 2006 O3 AQCD. McLaughlin et al. (2007a) reported that the
19                   cumulative exposure index of AOT60 related well to reductions in growth rates at forest
20                   sites in the southern Appalachian Mountains. However, the authors did not report an
21                   analysis to compare multiple indices. Overall, given the available data from previous O3
22                   AQCDs and the few recent studies, the cumulative, concentration-weighted indices
23                   perform better than the  peak or mean indices. It is still not possible, however, to
24                   distinguish the differences in performance among the  cumulative, concentration-weighted
25                   indices.

26                   The main conclusions from the 1996 and 2006 O3  AQCDs regarding an index based on
27                   ambient exposure are still valid.  No information has come forth since the 2006 O3 AQCD
28                   to alter those conclusions significantly. These key conclusions can be restated as follows:

29                      •  O3 effects in plants are cumulative;
30                      •  higher O3 concentrations appear to be more important than lower
31                         concentrations in eliciting a response;
32                      •  plant sensitivity to O3 varies with time of day and plant development stage;
33                         and
34                      •  exposure indices that accumulate the O3 hourly concentrations and
35                         preferentially weight the higher concentrations have better statistical fits to
36                         growth/yield response than do the mean and peak indices.
      Draft - Do Not Cite or Quote                      9-105                                September 2011

-------
 1                  Following the 2006 criteria review process (U.S. EPA. 2006b). the EPA proposed an
 2                  alternative form of the secondary NAAQS for O3 using a cumulative, concentration-
 3                  weighted exposure index to protect vegetation from damage (72 FR37818). The EPA
 4                  considered two specific concentration-weighted indices: the cutoff concentration
 5                  weighted SUM06 and the sigmoid-weighted W126 exposure index (U.S. EPA. 2007b).
 6                  These two indices performed equally well in predicting the exposure-response
 7                  relationships observed in the crop and tree seedlings studies (Lee etal.. 1989). At a
 8                  workshop convened to consider the science supporting these indices (Heck and Cowling.
 9                  1997) there was a consensus that these cumulative concentration-weighted indices being
10                  considered were equally capable of predicting plant response. Below are the definitions
11                  of the two cumulative index forms considered in the previous staff paper review (U.S.
12                  EPA. 2007b):

13                      •  SUM06:  Sum of all hourly O3 concentrations greater than or equal to
14                         0.06 ppm observed during a specified daily and seasonal time window (Figure
15                         9-9a).
16                      •  W126: Sigmoidally weighted sum of all hourly O3 concentrations observed
17                         during a specified daily and seasonal time window (Similar to  Figure 9-9b).
18                         The sigmoidal weighting of hourly O3  concentration is given in the equation
19                         below, where C is the hourly O3 concentration in ppm:
                                                       1
                                         W  =
                                           c       4403e~126C
                                                                                         Equation 9-1
20                  The SUM06 and W126 indices have a variety of relevant time windows that may be
21                  applied and are discussed in Section 9.5.3.

22                  It should be noted that there are some important differences between the SUM06 and
23                  W126. When considering the response of vegetation to ozone exposures represented by
24                  the threshold (e.g., SUM06) and non-threshold (e.g., W126) indices, the W126 metric
25                  does not have a cut-off in the weighting scheme as does SUM06 and thus it includes
26                  consideration of potentially damaging exposures below 60 ppb. The W126 metric also
27                  adds increasing weight to hourly concentrations from about 40 ppb to about 100 ppb.
28                  This is unlike cut-off metrics such as the SUM06 where all concentrations above 60 ppb
29                  are treated equally. This is an important feature of the W126 since as hourly
30                  concentrations become higher, they become increasingly likely to overwhelm plant
31                  defenses  and are known to be more detrimental to vegetation (See Section 9.5.3.1).
      Draft - Do Not Cite or Quote                     9-106                               September 2011

-------
            9.5.3   Important Components of Exposure Indices

 1                   In the previous O3 AQCDs it was established that higher hourly concentrations have
 2                   greater effects on vegetation than lower concentrations (U.S. EPA. 2006b. 1996c).
 3                   Further, it was determined that the diurnal and seasonal duration of exposure is important
 4                   for plant response. Weighting of hourly concentrations and the diurnal and seasonal time
 5                   window of exposure are the most important variables in a cumulative exposure index and
 6                   will be discussed below. However, these variables must be taken in the context of plant
 7                   phenology, diurnal conductance rates, plant canopy structure, and detoxification
 8                   mechanisms of vegetation as well as the climate and meteorology, all of which are
 9                   determinants of plant response. These more specific factors will be discussed in the
10                   uptake and dose  modeling section 9.5.4.
                     9.5.3.1    Role of Concentration

11                   The significant role of peak O3 concentrations was established based on several
12                   experimental studies  (U.S. EPA. 1996c). Several studies (Oksanen and Holopainen.
13                   2001; Yun and Laurence. 1999; Nussbaum et al., 1995) have added support for the
14                   important role that peak concentrations, as well as the pattern of occurrence, plays in
15                   plant response to O3. Oksanen and Holopainen (2001) found that the peak concentrations
16                   and the shape of the O3 exposure (i.e., duration of the event) were important determinants
17                   of foliar injury in European white birch saplings, but growth reductions were found to be
18                   more related to total cumulative exposure. Based on air quality data from 10 U.S. cities,
19                   three 4-week exposure treatments having the same SUM06 value were constructed by
20                   Yun and Laurence (1999). The authors used different exposure regimes to explore effects
21                   of treatments with variable versus uniform peak occurrence during the exposure period.
22                   The authors reported  that the variable peak exposures were important in causing injury,
23                   and that the different exposure treatments, although having the same SUM06, resulted in
24                   very different patterns of foliar injury. Nussbaum et al. (1995) also found peak
25                   concentrations and the pattern of occurrence to be  critical in determining the measured
26                   response. The authors recommended that to describe the effect on total forage yield, peak
27                   concentrations >0.11  ppm must be emphasized by using an AOT with higher threshold
28                   concentrations.

29                   A greater role for peak concentrations in effects on plant growth might be inferred based
30                   on air quality analyses for the southern California area (Tingey et al.. 2004; Lee et al..
31                   2003a). In the late 1960s and 1970s, extremely high O3 concentrations had impacted the
32                   San Bernardino National Forest. However, over the past 20+ years, significant reductions
33                   in O3 exposure have occurred (Bvtnerowicz et al., 2008; Lee et al., 2003a; Lefohn and
      Draft - Do Not Cite or Quote                      9-107                                September 2011

-------
 1                   Shadwick. 2000; Davidson. 1993). An illustration of this improvement in air quality is
 2                   shown by the 37-year history of O3 air quality at the Crestline site in the San Bernardino
 3                   Mountains (Figure 9-10) (Lee et al., 2003a). Ozone exposure increased from 1963 to
 4                   1979 concurrent with increased population and vehicular miles, followed by a decline to
 5                   the present mirroring decreases in precursor emissions. The pattern in exposure was
 6                   evident in various exposure indices including the cumulative concentration weighted
 7                   (SUM06), as well as maximum peak event (1-h peak), and the number of days having
 8                   hourly averaged O3 concentrations greater than or equal to 95 ppb. The number of days
 9                   having hourly averaged O3 concentrations greater than or equal to 95 ppb declined
10                   significantly from 163 days in 1978 to 103 days in 1997. The changes in ambient O3 air
11                   quality for the Crestline site were reflected in the changes in frequency and magnitude of
12                   the peak hourly concentration and the duration of exposure (Figure 9-10). Considering
13                   the role of exposure patterns in determining response, the seasonal and diurnal patterns in
14                   hourly O3 concentration did not vary appreciably from year to year over the 37-year
15                   period (Lee  et al., 2003a).

16                   The potential importance of exposure to peak concentrations comes both from results of
17                   measures of tree conditions on established plots and from results of model simulations.
18                   Across a broad area of the San Bernardino National Forest, the Forest Pest Management
19                   (FPM) method of injury assessment indicated an improvement in crown condition from
20                   1974 to 1988;  and the area of improvement in injury assessment is coincident with an
21                   improvement in O3 air quality (Miller and Rechel. 1999). A  more  recent analysis of
22                   forest changes in the San Bernardino National Forest, using an expanded network of
23                   monitoring sites, has verified significant changes in growth,  mortality rates, basal area,
24                   and species  composition throughout the area since 1974 (Arbaugh et al.. 2003). A model
25                   simulation of ponderosa pine growth over the 40-year period in the San Bernardino
26                   National Forest showed a significant impact of O3 exposure  on tree growth and indicates
27                   improved growth with reduced O3 concentrations. This area has also experienced
28                   elevated N deposition and based on a number of environmental  indicators, it appears that
29                   this area is experiencing N saturation (Fenn et al., 1996). To account for this potential
30                   interaction, the model simulations were conducted under conditions of unlimited soil N.
31                   The actual interactions are not known. The improvement in growth over the years was
32                   attributed to improved air quality, but no distinction was made regarding the relative role
33                   of mid-range and higher hourly concentrations, only that improved growth tracked
34                   decreasing SUM06, maximum peak concentration, and number of days of hourly O3
35                   >95 ppb (Tingev et al.. 2004). A summary of air quality data from 1980 to 2000 for the
36                   San Bernardino National Forest area of the number of "mid-range" hourly concentrations
37                   indicated no dramatic changes over this 20-year period, ranging from about 1,500 to
38                   2,000 hours per year (Figure 9-11). There was a slow increase in the number of mid-
39                   range concentrations from  1980 to 1986, which corresponds to the period after

      Draft - Do Not Cite or Quote                      9-108                                September 2011

-------
1
2
3
4
5

6

7
implementation of the air quality standard. Another sharper increase was observed in the
late 1990s. This pattern of occurrence of mid-range hourly concentrations suggests a
lesser role for these concentration ranges compared to the higher values in either of the
ground-level tree injury observations of the model simulation of growth over the 40-year
period.

e >•
.9|
E o
LU ii

E £T
Gj o

It
3 Q.
w S
JC
11
*-s
T—
£*
S ft
o "°


o
i-n •

o -
h*
CM "
M .

I"
8;

i-
o
0 -
0 .


• ROG .— 	 -• *^-^2\
O NOx 0__ 	 	 ° \!°~°O

• •
. •
• \ * •. • • A • • •*• / v« •
• • / * * * » *
•
oo o0
oo°w°N o7 X°0/ •*•••*•*•••>.•..
o0^ V °° VV--
o °
'*r»* Ox^^^'N* •
O" ' */O OO #•/ *
*-.\
Q
_ / XQ O — .* \ — _^' *Q O * ^ ^' * • • * * * *
o' o o' ' • *\.
" • •














                              1965   1970   1975   1980    1985    1990    1995   2000
                                                  Year

      Source: Used with permission from Elsevier Science Ltd. (Lee et a I.. 2003a).
      Annual ROG and NOX emissions data for San Bernardino County were obtained from Alexis et al. (2001 a) and the California Air
     Resource Board's emission inventory available at http://www.arb.ca.gov/aqd/aqdpage.htm (Cal/EPA, 2010).

     Figure 9-10   Trends in May to September 12-h SUM06, peak 1-h ozone
                    concentration and number of daily exceedances of 95 ppb for the
                    Crestline site in 1963 to 1999 in relation to trends in mean daily
                    maximum  temperature for Crestline and daily reactive organic
                    gases (ROG) and oxides of nitrogen (NOx) for San Bernardino
                    County.
     Draft - Do Not Cite or Quote
                             9-109
September 2011

-------
                                       Crestline, San Bernardino, CA
                                        Number of Hours 50 - 89 ppb
                                                 060710005
       §
       o
          2500
          2000
          1500
       0 1000
       .Q
       E
       3
           500
                                                                                  i
                                                                                         CN
                                                    Year
      Figure 9-11    The number of hourly average concentrations between 50 and 89
                     ppb for the period 1980-2000 for the Crestline, San Bernardino
                     County, CA, monitoring site.
                   9.5.3.2   Diurnal and Seasonal Exposure
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
Diurnal Exposure

The diurnal patterns of maximal leaf/needle conductance and occurrence of higher
ambient concentrations can help determine which hours during the day over a season
should be included in an exposure index. Stomatal conductance is species and phenology
dependent and is linked to both diurnal and seasonal meteorological activity as well as to
soil/site conditions (e.g., VPD, soil moisture). Daily patterns of leaf/needle conductance
are often highest in midmorning, whereas higher ambient O3 concentrations generally
occurred in early to late afternoon when stomata were often partially closed and
conductances were lower. Total O3 flux depends on atmospheric and boundary layer
resistances, both of which exhibit variability throughout the day. Experimental studies
with tree species demonstrated the decoupling of ambient O3 exposure, peak occurrence,
     Draft - Do Not Cite or Quote
                            9-110
September 2011

-------
 1                   and gas exchange, particularly in areas of drought (Panek. 2004). Several studies have
 2                   suggested that ponderosa pine trees in the southern and northern Sierra Nevada
 3                   Mountains may not be as susceptible to high O3 concentrations as to lower
 4                   concentrations, due to reduced needle conductance and O3 uptake during the period when
 5                   the highest concentrations occur (Panek et al., 2002; Panek and Goldstein, 2001; Bauer et
 6                   al.. 2000; Arbaugh et al.. 1998). Panek et al. (2002) compared direct O3 flux
 7                   measurements into a canopy of ponderosa pine and demonstrated a lack of correlation of
 8                   daily patterns of conductance and O3  occurrence, especially in the late season drought
 9                   period; the authors concluded that a consideration of climate or season was essential,
10                   especially considering the role of soil moisture and conductance/uptake. In contrast,
11                   Grulke et al. (2002) reported high conductance when O3 concentrations were high in the
12                   same species, but under different growing site conditions. The longer-term biological
13                   responses reported by Miller and Rechel (1999) for ponderosa pine in the same region,
14                   and the general reduction in recent years in ambient O3 concentrations, suggest that
15                   stomatal conductance alone may not be a sufficient indicator of potential vegetation
16                   injury or damage. Another consideration for the effect of O3 uptake is the diurnal pattern
17                   of detoxification capacity of the plant. The detoxification capacity may not follow the
18                   same pattern as stomatal conductance (Heath et al.. 2009).

19                   The use of a 12-h (8:00 a.m. to 8:00 p.m.) daylight period for a W126 cumulating
20                   exposure was based primarily on evidence that the conditions for uptake of O3 into the
21                   plant occur mainly during the daytime hours. In general, plants have the highest stomatal
22                   conductance during the daytime and in many areas atmospheric turbulent mixing is
23                   greatest during the day as well (Uddling etal.. 2010; U.S. EPA. 2006b). However,
24                   notable exceptions to maximum daytime conductance are cacti and other plants with
25                   crassulacean acid metabolism (CAM photosynthesis) which only open their stomata at
26                   night. This section will focus on plants with C3 and C4 photosynthesis, which generally
27                   have  maximum stomatal conductance during the daytime.

28                   Recent reviews of the literature  reported that a large number of species had varying
29                   degrees of nocturnal stomatal conductance (Caird et al.. 2007; Dawson et al.. 2007;
30                   Musselman and Minnick. 2000). The  reason for night-time water loss through stomata  is
31                   not well understood and is an area of active research (e.g. (Christman et al.. 2009;
32                   Howard et al.. 2009) Night-time stomata opening may be enhanced by O3  damage that
33                   could result in loss of stomatal control, and less complete closure of stomata, than under
34                   low O3 conditions (Caird et al.,  2007; Grulke et al., 2007b). In general, the rate of
3 5                   stomatal conductance at night is much lower than during the day (Caird et al.. 2007).
36                   Atmospheric turbulence at night is also often low, which results in stable boundary layers
37                   and unfavorable conditions for O3 uptake into vegetation (Finkelstein et al.. 2000).
38                   Nevertheless, nocturnal turbulence does intermittently occur and may result in
      Draft - Do Not Cite or Quote                      9-111                                September 2011

-------
 1                   nonnegligible O3 flux into the plants. In addition, plants might be more susceptible to O3
 2                   exposure at night than during the daytime, because of potentially lower plant defenses
 3                   (Heath et al., 2009; Loreto and Fares. 2007; Musselman et al., 2006; Musselman and
 4                   Minnick. 2000). For significant nocturnal stomatal flux and O3 effects to occur, specific
 5                   conditions must exist. A susceptible plant with nocturnal stomatal conductance and low
 6                   defenses must be growing in an area with relatively high night-time O3 concentrations
 7                   and appreciable nocturnal atmospheric turbulence. It is unclear how many areas there are
 8                   in the U.S. where these conditions occur. It may be possible that these conditions exist in
 9                   mountainous areas of southern California, front-range of Colorado (Turnipseed et al.,
10                   2009) and the Great Smoky Mountains of North Carolina and Tennessee. Tobiessen et al.
11                   (1982) found that shade intolerant tree species showed opening of stomata in the dark and
12                   did not find this in shade tolerant species. This may indicate shade intolerant trees may be
13                   more likely to be susceptible to O3 exposure at night. More information is needed in
14                   locations with high night-time O3 to assess the local O3 patterns, micrometeorology and
15                   responses of potentially vulnerable plant species.

16                   Several field studies have attempted to quantify night-time O3 uptake with a variety of
17                   methods. However, many of these studies have not linked the night-time flux to measured
18                   effects on plants. Grulke et al. (2004) showed that the stomatal conductance at night for
19                   ponderosa pine in the San Bernardino National Forest (CA) ranged from one tenth to one
20                   fourth that of maximum daytime  stomatal conductance. In June, at a high-elevation site, it
21                   was calculated that 11% of the total daily O3 uptake of pole-sized trees occurred at night.
22                   In late summer, however, O3 uptake at night was negligible. However, this study did not
23                   consider the turbulent conditions at night. Finklestein et al. (2000) investigated O3
24                   deposition velocity to forest canopies at three different sites. The authors found the total
25                   flux (stomatal and non-stomatal)  to the canopy to be very low during night-time hours as
26                   compared to day-time hours. However, the authors did note that higher nocturnal
27                   deposition velocities at conifer sites may be due to some degree of stomatal opening at
28                   night (Finkelstein et al.. 2000). Work by Mereu et al. (2009) in Italy on Mediterranean
29                   species indicated that nocturnal uptake was from 10 to 18% of total daily uptake during a
30                   weak drought and up to 24% as the drought became more pronounced. The proportion of
31                   night-time uptake was greater during the drought due to decreases in daytime stomatal
32                   conductance (Mereu et al.. 2009). In a study conducted in California, Fares et al. (Fares et
33                   al., 2011) reported that calculated mean percentages of nocturnal uptake were 5%, 12.5%,
34                   6.9% of total O3 uptake for lemon, mandarin, and orange, respectively. In another recent
35                   study at  the Aspen FACE site in Wisconsin, calculated leaf-level stomatal O3 flux was
36                   near zero from the night-time hours of 8:00 p.m. to 5:00 a.m. (Uddling et al.. 2010). This
37                   was likely due to low horizontal wind speed (>1 m/s) and low O3 concentrations
38                   (<25 ppb) during those same night-time hours (Figure 9-12).
      Draft - Do Not Cite or Quote                      9-112                                September 2011

-------
        E
         j-:
        O
        "o
           0.15
           o.io
           0.05
           0.00
         *T  6
          n
         •••7
          E 5
          o
          g
          en
          u.
          •O
          CO
                	 2004
                	2005  Sf
                    5      10      15
                          Time of day
                                                      50
                                              £ 40
                                              o
                                              I 30
                                              .1
                                              1 20
                                                    s
                                                   cf
                                                         5     10     15
                                                              Time of day
                                                                                   20
                                              T   6
                                              W
                                              •]|
                                              E  5
                                              o
                                              I  «
                                               .
                                              LL
                                              •0  2
                                              CO
              5      10      15     20
                    Time of day

Source: Used with permission from Elsevier Ltd (Uddling et al.. 2010).
                                                               5      ID      15     20
                                                                    Time of day
     Figure 9-12    Mean diurnal, (a) conductance through boundary layer and
                    stomata (gbs), (b) Ozone concentration, and leaf-level stomatal
                    ozone flux without flux cut-off threshold (FstOi) in control plots
                    from mid-June through August in (c) 2004 and (d) 2005 in the Aspen
                    FACE experiment. Subscripts "max" and "min" refer to stomatal
                    fluxes calculated neglecting and accounting for potential non-
                    stomatal ozone flux, respectively.
1
2
3
4
5
6
1
            A few studies have tested the biological effects of night-time O3 exposure on vegetation
            in controlled chambers. Biomass of ponderosa pine seedlings was significantly reduced
            when seedlings were exposed to either daytime or nighttime episodic profiles (Lee and
            Hogsett 1999). However, the biomass reductions were much greater with daytime peak
            concentrations than with nighttime peak concentrations. Similarly, birch cuttings grown
            in field chambers that were exposed to O3 at night only, daytime only, and 24 hours
            showed similar reductions in biomass in night only and day only treatments. Birch
            seedling showed greater reductions in growth in 24-h exposures than those exposed to O3
            at night or day only (Matyssek et al.. 1995). Field mustard (Brassica rapd) plants
     Draft - Do Not Cite or Quote
                                        9-113
September 2011

-------
 1                  exposed to O3 during the day or night showed little significant difference in the amounts
 2                  of injury or reduced growth response to O3 treatment, although the stomatal conductance
 3                  was 70-80% lower at night (Winner et al., 1989). These studies show that effects can be
 4                  seen with night-time exposures to O3 but when atmospheric conditions are stable at night,
 5                  it is uncertain how these exposures may affect plants and trees with complex canopies in
 6                  the field.


                    Seasonal Exposure

 7                  Vegetation across the U.S. has widely varying periods of physiological activity during the
 8                  year due to variability in climate and phenology. In order for a particular plant to be
 9                  vulnerable to O3 pollution, it must have foliage and be physiologically active. Annual
10                  crops are typically grown for periods of two to three months. In contrast, perennial
11                  species may be photosynthetically active longer (up to 12 months each year for some
12                  species) depending on the species and where it is grown. In general, the period of
13                  maximum physiological activity and thus, potential O3 uptake for vegetation coincides
14                  with some or all of the intra-annual period defined as the O3 season, which varies on a
15                  state-by-state basis (Figure 3-19). This is because the high  temperature and high light
16                  conditions that typically promote the formation of tropospheric O3 also promote
17                  physiological activity in vegetation. There are very limited exceptions to this pattern
18                  where O3 can form in the winter in areas in the western U.S. with intense natural gas
19                  exploration (Pinto. 2009). but this is typically when plants  are dormant and there is little
20                  chance of O3 uptake. The selection of any single window of time for a national standard
21                  to consider hourly O3 concentrations represents a compromise, given the significant
22                  variability in growth patterns and lengths of growing season among the wide range of
23                  vegetation species that may experience adverse effects associated with O3 exposure.

24                  Various intra-annual averaging and accumulation time periods have been considered for
25                  the protection of vegetation. The 2010 proposal for secondary O3 standard (75 FR 2938,
26                  p. 3003) proposed to use the maximum consecutive 3-month period within the O3 season.
27                  The U.S. Forest Service and federal land managers have used a 24-h W126 accumulated
28                  for 6 months from April through September (2000). However, some monitors in the U.S.
29                  are operational for as little as four months and would not have enough data for a 6-month
30                  seasonal window. The exposure period in the vast majority of O3 exposure studies
31                  conducted in the U.S. has been much shorter than 6 months. Most of the crop studies
32                  done through NCLAN had exposures less than three months with an average of 77 days.
33                  Open-top chamber studies of tree seedlings, compiled by the EPA, had an average
34                  exposure of just over three months or 99 days. In more recent FACE experiments,
35                  Soy FACE exposed soybeans for an average of approximately 120 days per year and the
36                  Aspen FACE experiment exposed trees to an average of approximately 145 days per year

      Draft - Do Not Cite or Quote                      9-114                               September 2011

-------
1
2
3
of elevated O3, which included the entire growing season at those particular sites. Despite
the possibility that plants may be exposed to ambient O3 longer than 3 months in some
locations, there is a lack of exposure experiments conducted for longer than 3 months.
    B
                                           30      40

                                         Highest 3 month W126
                                           30     40      50

                                         Highest 3 month W126
    Figure 9-13   Maximum 3-month, 12-h W126 plotted against maximum 6-month,
                  12-h W126. Data are from the AQS and CASTNET monitors for the
                  years 2008 and 2009. (A) W126, 3 month versus 6 month, 2008
                  (Pearson correlation = 0.99); (B) W126, 3 month versus 6 month,
                  2009 (Pearson correlation = 0.99).
    Draft - Do Not Cite or Quote
                          9-115
September 2011

-------
 1                   In an analysis of the 3- and 6-month maximum W126 values calculated for over 1,200
 2                   AQS (Air Quality System) and CASTNET (Clean Air Status and Trend Network) EPA
 3                   monitoring sites for the years 2008-2009, it was found that these 2 accumulation periods
 4                   resulted in highly correlated metrics (Figure 9-13). The two accumulation periods were
 5                   centered on the yearly maximum for each monitoring site, and it is possible that this
 6                   correlation would be weaker if the two periods were not temporally aligned. In the U.S.,
 7                   W126 cumulated over 3 months, and W126 cumulated over 6 months are proxies of one
 8                   another, as long as the period in which daily W126 is accumulated corresponds to the
 9                   seasonal maximum. Therefore, it is expected that either statistic will predict vegetation
10                   response equally well. In other words, the strength of the correlation between maximum
11                   3-month W126 and maximum 6-month W126 is such that there is no material difference
12                   in their predictive value for vegetation response.
            9.5.4   Ozone Uptake/Dose Modeling for Vegetation

13                   Another approach for improving risk assessment of vegetation response to ambient O3 is
14                   based on estimating the O3 concentration from the atmosphere that enters the leaf (i.e.,
15                   flux or deposition). Interest has been increasing in recent years, particularly in Europe, in
16                   using mathematically tractable flux models for O3 assessments at the regional, national,
17                   and European scale (Matyssek et al.. 2008; Paoletti and Manning. 2007; M and M. 2004;
18                   Emberson et al., 2000b; Emberson et al., 2000a). Some researchers have claimed that
19                   using flux models can be used to better predict vegetation responses to O3 than exposure-
20                   based approaches (Matyssek et al., 2008). However, other research has suggested that
21                   flux models do not predict vegetation responses to O3 better than exposure-based models,
22                   such as AOT40 (Gonzalez-Fernandez et al., 2010). While some efforts have been made in
23                   the U.S. to calculate O3 flux into leaves and canopies (Fares etal.. 2010a: Turnipseed et
24                   al.. 2009: Uddling et al.. 2009: Bergweiler et al.. 2008: Hogg et al.. 2007: Grulke et al..
25                   2004: GrantzetaL 1997: GrantzetaL 1995), little information has been published
26                   relating these fluxes to effects on vegetation. The lack of flux data in the U.S. and the
27                   lack of understanding of detoxification processes have made this technique less viable for
28                   vulnerability and risk assessments in the U.S.

29                   Flux calculations are data intensive and must be carefully implemented. Reducing
30                   uncertainties in flux estimates for areas with diverse surface or terrain conditions to
31                   within ±50% requires "very careful application of dry deposition models, some model
32                   development, and support by experimental observations" (Wesely and Hicks. 2000). As
33                   an example, the annual average deposition velocity of O3 among three nearby sites in
34                   similar vegetation was found to vary by ±10%, presumably due to terrain (Brook et al.,
35                   1997). Moreover, the authors stated that the actual variation was even greater, because
      Draft - Do Not Cite or Quote                      9-116                                September 2011

-------
 1                   stomatal uptake was unrealistically assumed to be the same among all sites, and flux is
 2                   strongly influenced by stomatal conductance (Brook et al.. 1997; Massman and Grantz.
 3                   1995; Fuentes et al., 1992; Reich. 1987; Leuning et al., 1979). This uptake-based
 4                   approach to quantify the vegetation impact of O3 requires inclusion of those factors that
 5                   control the diurnal and seasonal O3 flux to vegetation (e.g., climate patterns, species
 6                   and/or vegetation-type factors and site-specific factors). The models have to distinguish
 7                   between stomatal and non-stomatal components of O3 deposition to adequately estimate
 8                   actual concentration reaching the target tissue of a plant to elicit a response (Uddling et
 9                   al., 2009). Determining this O3 uptake via canopy and stomatal conductance relies on
10                   models to predict flux and ultimately the "effective" flux (Grunhage et al.. 2004;
11                   Massman. 2004; Massman et al., 2000). "Effective flux" has been defined as the balance
12                   between O3 flux and detoxification processes (Heath et al.. 2009; Musselman and
13                   Massman. 1999; Grunhage and Haenel, 1997; Dammgen et al., 1993). The time-
14                   integrated "effective flux" is termed "effective dose." The uptake mechanisms and the
15                   resistances in this process, including stomatal conductance and biochemical defense
16                   mechanisms, are discussed below. The flux-based index is the goal for the "Level II"
17                   critical level for assessment of O3 risk to vegetation and ecosystems across Europe
18                   (Ashmoreetal..2004a).

19                   An important consideration in both O3 exposure and uptake is how the O3 concentration
20                   at the top of low vegetation such as, crops and tree seedlings may be lower than the
21                   height at which the measurement is taken. Ambient monitor inlets in the U.S. are
22                   typically at heights of 3 to 5 meters. During daytime hours, the vertical O3 gradient can
23                   be relatively small because turbulent mixing maintains the downward flux of O3. For
24                   example, Horvath et al. (1995) calculated a 7% decrease in O3 going from a height of 4
25                   meters down to 0.5 meters above the surface during unstable (or turbulent) conditions in
26                   a study over low vegetation in Hungary [see section AX3.3.2. of the 2006 O3 AQCD
27                   (U.S. EPA. 2006b)1. There have been several studies indicating decreased O3
28                   concentrations under tree canopies (Kolb et al.. 1997; Samuelson and Kelly. 1997; Joss
29                   andGraber. 1996; Fredericksen et al.. 1995; Lorenzini and Nali.  1995; Enders. 1992;
30                   FontanetaL 1992; Neufeld et al.. 1992). In contrast, for forests,  measured data may
31                   underestimate O3 concentration at the top of the canopy. The difference between
32                   measurement height and canopy height is a function of several factors, the intensity of
33                   turbulent mixing in the surface layer and other meteorological factors, canopy height and
34                   total deposition to the canopy. Some researchers have used deposition models to estimate
35                   O3 concentration at canopy-top height based on concentrations at measurement height
36                   (Emberson et al.. 2000a). However, deposition models usually require meteorological
37                   data inputs that are not always available or well characterized across large spatial scales.
      Draft - Do Not Cite or Quote                      9-117                               September 2011

-------
 1                   Soil moisture is a critical factor in controlling O3 uptake through its effect on plant water
 2                   status and stomatal conductance. In an attempt to relate uptake, soil moisture, and
 3                   ambient air quality to identify areas of potential risk, available O3 monitoring data for
 4                   1983 to  1990 were used along with literature-based seedling exposure-response data from
 5                   regions within the southern Appalachian Mountains that might have experienced O3
 6                   exposures sufficient to inhibit growth (Lefohn et al.. 1997). In a small number of areas
 7                   within the region,  O3  exposures and soil moisture availability were sufficient to possibly
 8                   cause growth reductions in some O3 sensitive species (e.g., black cherry). The
 9                   conclusions were limited, however, because of the uncertainty in interpolating O3
10                   exposures in many of the areas and because the hydrologic index used might not reflect
11                   actual water stress.

12                   The non-stomatal component of plant defenses are the most difficult to quantify, but
13                   some studies are available (Heath et al.. 2009; Barnes et al.. 2002; Plochl et al.. 2000;
14                   Chen et  al.. 1998; Massman and Grantz. 1995). Massman et al. (2000) developed a
15                   conceptual model of a dose-based index to determine how plant injury response to O3
16                   relates to the traditional exposure-based parameters. The index used time-varying -
17                   weighted fluxes to account for the fact that flux was not necessarily correlated with plant
18                   injury or damage. The model applied only to plant foliar injury and suggested that
19                   application of flux-based models for determining plant damage (yield or biomass) would
20                   require a better understanding and quantification of the relationship between injury and
21                   damage.
             9.5.5   Summary

22                   Exposure indices are metrics that quantify exposure as it relates to measured plant
23                   damage (i.e., reduced growth). They are summary measures of monitored ambient O3
24                   concentrations over time intended to provide a consistent metric for reviewing and
25                   comparing exposure-response effects obtained from various studies. No new information
26                   is available since 2006 that alters the basic conclusions put forth in the 2006 and 1996 O3
27                   AQCDs. These AQCDs focused on the research used to develop various exposure indices
28                   to help quantify effects on growth and yield in crops, perennials, and trees (primarily
29                   seedlings). The performance of indices was  compared through regression analyses of
30                   earlier studies designed to support the estimation of predictive O3 exposure-response
31                   models for growth and/or yield of crops and tree (seedling) species.
32                   Another approach for improving risk assessment of vegetation response to ambient O3 is
33                   based on determining the O3 concentration from the atmosphere that enters the leaf (i.e.,
34                   flux or deposition). Interest has been increasing in recent years, particularly in Europe, in
      Draft - Do Not Cite or Quote                       9-118                                September 2011

-------
 1                   using mathematically tractable flux models for O3 assessments at the regional, national,
 2                   and European scale (Matyssek et al.. 2008; Paoletti and Manning. 2007; M and M. 2004;
 3                   Emberson et al., 2000b; Emberson et al., 2000a). While some efforts have been made in
 4                   the U.S. to calculate O3 flux into leaves and canopies (Turnipseed et al.. 2009; Uddling et
 5                   al.. 2009; Bergweiler et al.. 2008; Hogg et al.. 2007; Grulke et al.. 2004; Grantz et al..
 6                   1997; Grantz etal.. 1995). little information has been published relating these fluxes to
 7                   effects on vegetation. There is also concern that not all O3 stomatal uptake results in a
 8                   yield reduction, which depends to some degree on the amount of internal detoxification
 9                   occurring with each particular species. Those species having high amounts of
10                   detoxification potential may, in fact, show little relationship between O3 stomatal uptake
11                   and plant response (Musselman and Massman. 1999). The lack of data in the U.S. and the
12                   lack of understanding of detoxification processes have made this technique less viable for
13                   vulnerability and risk assessments in the U.S.

14                   The main conclusions from the 1996 and 2006 O3 AQCDs regarding indices based on
15                   ambient exposure are still valid. These key conclusions can be restated as follows:

16                       •  O3 effects in plants are cumulative;
17                       •  higher O3 concentrations appear to be more important than lower
18                         concentrations in eliciting a response;
19                       •  plant sensitivity to O3 varies with time of day and plant development stage;
20                         and
21                       •  exposure indices that cumulate hourly O3 concentrations and preferentially
22                         weight the higher concentrations have better statistical fits to growth/yield
23                         response data than do the mean and peak indices.

24                   Various weighting functions have been used, including threshold-weighted (e.g.,
25                   SUM06) and continuous sigmoid-weighted (e.g., W126) functions. Based on statistical
26                   goodness-of-fit tests, these cumulative, concentration-weighted indices could not be
27                   differentiated from one another using data from previous exposure studies. Additional
28                   statistical forms for O3 exposure indices have been discussed in Lee et al.  (1988b). The
29                   majority of studies published since the 2006 O3 AQCD do not change earlier
30                   conclusions, including the importance  of peak concentrations, and the duration and
31                   occurrence of O3 exposures in altering plant growth and yield.

32                   Given the current state of knowledge and the best available data, exposure indices that
33                   cumulate and differentially weight the  higher hourly average concentrations and  also
34                   include the mid-level values continue to offer the most defensible approach for use in
35                   developing response functions and comparing studies, as well as for defining future
3 6                   indice s for vegetation protection.
      Draft - Do Not Cite or Quote                       9-119                                September 2011

-------
          9.6   Ozone Exposure-Plant  Response Relationships
             9.6.1   Introduction

 1                   The adequate characterization of the effects of O3 on plants for the purpose of setting air
 2                   quality standards is contingent not only on the choice of the index used (i.e. SUM06,
 3                   W126) to summarize O3 concentrations (Section 9.5), but also on quantifying the
 4                   response of the plant variables of interest at specific values of the selected index. The
 5                   many factors that determine the response of plants to O3 exposure have been discussed in
 6                   previous sections. They include species, genotype and other genetic characteristics
 7                   (Section 9.3), biochemical and physiological status (Section 9.3), previous and current
 8                   exposure to other stressors (Section 9.4.8), and characteristics of the exposure itself
 9                   (Section 9.5). Establishing a secondary air quality standard requires the capability to
10                   generalize those observations, in order to obtain predictions that are reliable enough
11                   under a broad variety of conditions, taking into account these factors. This section
12                   reviews results that have related specific quantitative observations of O3 exposure with
13                   quantitative observations of plant responses, and the predictions of responses that have
14                   been derived from those observations through empirical models.

15                   For four decades, exposure to O3 at ambient concentrations found in many areas of the
16                   U.S. has been known to cause detrimental effects in plants (U.S. EPA. 2006b. 1996b.
17                   1984. 1978a). Results published after the 2006 O3 AQCD continue to support this
18                   finding, and the following sections deal with the quantitative characterizations of the
19                   relationship, and what new insights may have appeared since 2006. Detrimental effects
20                   on plants include visible injury, decreases in the rate of photosynthesis, reduced growth,
21                   and reduced yield of marketable plant parts. Most published exposure-response data have
22                   reported O3 effects  on the yield of crops and the growth of tree seedlings, and those two
23                   variables have been the focus of the characterization of ecological impacts of O3 for the
24                   purpose of setting secondary air quality standards. In order to  support quantitative
25                   modeling of exposure-response  relationships, data should preferably include more than
26                   three levels of exposure, and some  control of potential confounding or interacting factors
27                   should  be present in order to model the relationship with sufficient accuracy. Letting
28                   potential confounders,  such as other stressors, vary freely when generating O3 exposure-
29                   response data might improve the 'realism' of the data, but it also greatly increases the
30                   amount of data necessary to extract a clear quantitative description of the relationship.
31                   Conversely however, experimental settings should not be so exhaustively restrictive as to
32                   make generalization outside of them problematic. During the last four decades, many of
33                   the studies of the effects of O3 on growth and yield of plants have not included enough
34                   levels of O3 to parameterize more than the simplest linear model. The majority of these


      Draft - Do Not Cite or Quote                      9-120                                September 2011

-------
 1                   studies have only contrasted two levels, ambient and elevated, or sometimes three by
 2                   adding carbon filtration in OTC studies, with little or no consideration of quantitatively
 3                   relating specific values of exposure to specific values of growth or yield. This is not to
 4                   say that studies that did not include more than two or three levels of O3 exposure, or
 5                   studies that were conducted in uncontrolled environments, do not provide exposure-
 6                   response information that is highly relevant to reviewing air quality standards. In fact,
 7                   they can be essential in verifying the agreement between predictions obtained through the
 8                   empirical models derived from experiments such as NCLAN, and observations. The
 9                   consensus of model predictions and observations from a variety of studies conducted in
10                   other locations, at other times, and using different exposure methods, greatly increases
11                   confidence in the reliability of both. Furthermore, if they are considered in the aggregate,
12                   studies with few levels of exposure or high unaccounted variability can provide
13                   additional independent estimates of decrements in plant growth and yield, at least within
14                   a few broad categories of exposure.

15                   Extensive exposure-response information on a wide variety of plant species has been
16                   produced by two long-term projects that were designed with the explicit aim of obtaining
17                   quantitative characterizations of the response of such an assortment of crop plants and
18                   tree seedlings to O3 under North American conditions: the NCLAN project for crops, and
19                   the EPA National Health and Environmental Effects Research Laboratory, Western
20                   Ecology Division tree seedling project (NHEERL/WED). The NCLAN project was
21                   initiated by the EPA in 1980 primarily to improve estimates of yield loss under  field
22                   conditions and to estimate the magnitude of crop losses caused by O3 throughout the U.S.
23                   (HecketaL 1991; Hecketal..  1982). The cultural conditions used in the NCLAN studies
24                   approximated typical agronomic practices, and the primary objectives were: (1) to define
25                   relationships between yields of major agricultural crops and O3 exposure as required to
26                   provide data necessary for economic assessments and development of O3 NAAQS; (2) to
27                   assess the national economic consequences resulting from O3 exposure of major
28                   agricultural crops; and (3) to advance understanding of cause-and-effect relationships that
29                   determine crop responses to pollutant exposures.

30                   NCLAN experiments yielded 54 exposure-response curves  for 12 crop species,  some of
31                   which were represented by multiple cultivars at several of 6 locations throughout the U.S.
32                   The NHEERL/WED project was initiated by EPA in 1988 with the same objectives for
33                   tree species, and yielded 49 exposure-responses curves for multiple genotypes of 11 tree
34                   species grown for up to three years in Oregon, Michigan, and the Great Smoky
35                   Mountains National Park. Both projects used OTCs to expose plants to three to  five
36                   levels of O3. Eight of the 54 crop datasets were from plants grown under a combination
37                   of O3 exposure and experimental drought conditions. Figure 9-14 through 9-17
3 8                   summarize some of the NCLAN and NHEERL/WED results.
      Draft - Do Not Cite or Quote                      9-121                                September 2011

-------
 1                   It should be noted that data from FACE experiments might also be used for modeling
 2                   exposure-response. They only use two levels of O3 (ambient concentration at the site and
 3                   a multiple of it), but given that the value of both levels of exposure changes every year,
 4                   and that they are typically run for many consecutive years, aggregating data over time
 5                   produces twice as many levels of O3 as there are years. As described in Section 9.2.4,
 6                   FACE experiments seek to impose fewer constraints on the growth environment then
 7                   OTCs. As a consequence, FACE studies have to contend with larger variability,
 8                   especially year-to-year variability, but the difference in experimental conditions between
 9                   the two methodologies makes comparisons between their results especially useful.

10                   Growth and yield of at least one crop (soybean) has been investigated in yearly
11                   experiments since 2001 at a FACE facility in Illinois (University of Illinois. 2010;
12                   Morgan et al.. 2006). however almost all analyses of SoyFACE published so far have
13                   been based on subsets of one or two years, and have only contrasted ambient versus
14                   elevated O3 as categorical variables. They have not modeled the response of growth and
15                   yield to O3 exposure continuously over the range of exposure values that have occurred
16                   over time. The only exception is a study by Betzelberger et al. (2010). who used a linear
17                   regression model on data pooled over 2 years. Likewise, trees of three species (trembling
18                   aspen, paper birch, and sugar maple) were grown between 1998 and 2009 in a FACE
19                   experiment located in Rhinelander, Wisconsin (Pregitzer et al.. 2008; Dickson et al..
20                   2000). The Aspen FACE experiment has provided extensive data on responses of trees
21                   beyond the seedling stage under long-term exposure, and also on ecosystem-level
22                   responses (Section 9.4), but the only attempt to use those data in a continuous model of
23                   the response of tree growth to O3 exposure (Percy et al.. 2007) suffered severe
24                   methodological problems, some of which are discussed in Section 9.6.3. Finally, one
25                   experiment was able to exploit a naturally occurring gradient of O3 concentrations to fit a
26                   linear regression model to the growth of cottonwood (Gregg et al.. 2006. 2003). Factors
27                   such as genotype, soil type and soil moisture were under experimental control, and the
28                   authors were able to partition out the effects of potential confounders such as
29                   temperature, atmospheric N deposition, and ambient CO2.

30                   A serious difficulty in assessing results of exposure-response research is the multiplicity
31                   of O3 metrics that have been used in reporting. As described in Section 9.5, metrics that
32                   entail either weighting or thresholding of hourly values cannot be converted into one
33                   another, or into unweighted metrics such as hourly average. When computing O3
34                   exposure using weighted or thresholded metrics, the computation of each metric has to
3 5                   start with the original hourly data. Comparisons of exposure-response models can only be
36                   made between studies that used the same metric, and the  value of exposure at which a
37                   given plant response is expected using one metric of exposure cannot be exactly
38                   converted to another metric. Determining the exposure value at which an effect would be
      Draft - Do Not Cite or Quote                      9-122                                September 2011

-------
 1                  observed in a different metric can only be accomplished by first computing the
 2                  experimental exposures in this metric from the hourly data, then estimating (fitting)
 3                  model coefficients again. This problem is irremediable, although useful comparisons
 4                  might be made using categorical exposures such as 'current ambient exposure' or '2050
 5                  projected exposure', which can serve as a common reference for quantitative values
 6                  expressed in various metrics. Studies that contained growth or yield exposure-response
 7                  data at few levels of exposure, and/or using metrics other than W126 are summarized in
 8                  Tables 9-18 and 9-19.
            9.6.2   Estimates of Crop Yield Loss and Tree Seedling Biomass Loss in the
                    1996 and 2006 Ozone AQCDs

 9                  The 1996 and 2006 O3 AQCDs relied extensively on analyses of NCLAN and
10                  NHEERL/WED by Lee et al. (1994: 1989. 1988b. 1987). Hogsett et al. (1997). Lee and
11                  Hogsett (1999). Heck et al. (1984). Rawlings and Cure (1985). Lesser et al. (1990). and
12                  Gumpertz and Rawlings (1992). Those analyses concluded that a three-parameter
13                  Weibull model-
                                                                                        Equation 9-2

14                  is the most appropriate model for the response of absolute yield and growth to O3
15                  exposure, because of the interpretability of its parameters, its flexibility (given the small
16                  number of parameters), and its tractability for estimation. In addition, removing the
17                  intercept a results in a model of relative yield (yield relative to [yield at exposure=0])
18                  without any further reparameterization. Formulating the model in terms of relative yield
19                  or relative yield loss (yield loss=[l - relative yield]) is essential in comparing exposure-
20                  response across species, genotypes, or experiments for which absolute values of the
21                  response may vary greatly. In the  1996 and 2006 O3 AQCDs, the two-parameter model
22                  of relative yield was used in deriving common models for multiple species, multiple
23                  genotypes within species, and multiple locations.

24                  Given the disparate species, genotypes, and locations that were included in the NCLAN
25                  and NHEERL/WED projects, and in the absence of plausible distributional assumptions
26                  with respect to those variables, a three step process using robust methods was used to
27                  obtain parameter estimates that could be generalized. The models that were derived for
28                  each species or group of species were referred to as median composite functions. In the
29                  first step, the three parameters of the Weibull model were computed for absolute yield or

      Draft - Do Not Cite or Quote                      9-123                               September 2011

-------
 1                   biomass data from each NCLAN and NHEERL/WED experiment (54 crop datasets and
 2                   49 tree seedling datasets), using nonlinear regression. When data were only available for
 3                   three levels of exposure because of experimental problems, the shape parameter (3 was
 4                   constrained to 1, reducing the model to an exponential decay model. In the second step, a
 5                   was dropped, and predicted values of relative yield or biomass were then computed for
 6                   12-hr W126 exposures between 0 and 60 ppm-h. At each of these W126 exposure values,
 7                   the 25th, 50th, and 75th percentiles of the response were identified among the predicted
 8                   curves of relative response. For example, for the 34 NCLAN studies of 12 crop species
 9                   grown under non-droughted conditions for a complete cropping cycle (Figure 9-14), the 3
10                   quartiles of the response were identified at every integer value of W126 between 0 and
11                   60. The third step fitted a two-parameter Weibull model to those percentiles, yielding the
12                   median composite function for the relative yield or biomass response to O3 exposure for
13                   each grouping of interest (e.g.,  all crops, all trees, all datasets  for one species), as well as
14                   composite functions for the other quartiles. In the 1996 and 2006 O3 AQCDs this
15                   modeling of crop yield loss and tree seedling biomass loss was conducted using the
16                   SUM06 metric for exposure. This section updates those results by using the 12-hr W126
17                   as proposed in 2007 (72 FR 37818) and 2010 (75 FR 2938, p. 3003). Figures 9-14
18                   through 9-17 present quantiles  of predicted relative yield or biomass loss at seven values
19                   of the 12-h W126 for some representative groupings of NCLAN and NHEERL/WED
20                   results. Tables 9-10 through 9-12 give the 90-day 12-h W126 O3 exposure values at
21                   which 10 and 20% yield or biomass losses are predicted in 50 and 75% of crop or tree
22                   species using the composite functions.
      Draft - Do Not Cite or Quote                      9-124                                September 2011

-------
100 i
90 •
80 •
g 70 •
w
8 60 •
1 50 •
1 40 •
S. 30 •
20 •
10 •
n .

34 crop datasets





jj


H







•--
4




•

j 90thPctile
T




^*
\




^


?-*



^

\




M i
75thPctile


50thPctile
25thPctile
10thPctile
                              10      20      30      40      50


                                         12 hrW126 (ppm-hr)
                                                                 60
  Source of Weibull parameters: Lee and Hogsett (1996).
  Quantiles of the predicted relative yield loss at 7 values of 12-hour W126 for 34 Weibull curves estimated using nonlinear
regression on data from 34 studies of 12 crop species grown under well-watered conditions for the full duration of 1 cropping cycle.



Figure 9-14    Quantiles of predicted relative yield loss for 34 NCLAN  crop

                  experiments.
Draft - Do Not Cite or Quote
9-125
September 2011

-------
        11 Soybean datasets
                                                10    20   30    40   50    60
  100 -

  90 -

  80 -

  70 -

  60 -

  50 -

  40 -

  30 -

  20 -

  10 -
5 Cotton datasets
             20    30   40    50

                12hrW126 (ppm-hr
100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

 0
                                        2 Com datasets
                                             20   30    40   50

                                                12hrW126 (ppm-hr)
 Source of Weibull parameters: Lee and Hogsett (1996).

Figure 9-15   Quantiles of predicted relative yield loss for 4 crop species in
              NCLAN experiments. Quantiles of the predicted relative yield loss
              at 7 values of 12-h W126 for Weibull curves estimated using
              nonlinear regression for 4 species grown under well-watered
              conditions for the full duration of 1 cropping cycle. The number of
              studies available for each species is indicated on each plot.
Draft - Do Not Cite or Quote
                                9-126
                              September 2011

-------
          in
          in
          o
          _i
          in
          in
          ra
          o
          In
          
-------
   100

   90

   80

 I ™

 I 6°
 1 50
 o
 to 40
 s
 te 30
 Q-
   20

   10

    0
14 Aspendatasets
                       ^-  SCPPctile
100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

 0
                                        11 Ponderosapine datasets
         10   20    30   40    50
  100 -

   90 -

   80 -
   50-
 O
 ™  40-

 I  30 -
 Q_
   20 -

   10 -

   0
7 Douglas firdatasets
             20    30   40    50

              90 day 12 hr W126 (ppm-hr)
100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -
                                        STulip poplardatasets
                                        10    20   30    40   50

                                              90 day 12 hr W126 (ppm-hr)
 Source of Weibull parameters: Lee and Hogsett (1996).

Figure 9-17    Quantiles of predicted relative biomass loss for 4 tree species in
               NHEERL/WED experiments. Quantiles of the predicted relative
               above-ground biomass loss at 7 exposure values of 12-h W126 for
               Weibull curves estimated using nonlinear regression on data for 4
               tree species grown under well-watered conditions for 1 or 2 year.
               Curves were standardized to 90-day W126.  The number of studies
               available for each species is indicated on each plot.
Draft - Do Not Cite or Quote
                                9-128
                              September 2011

-------
 Table 9-9      Ozone exposures at which 10 and 20% yield loss is predicted for 50
                and 75% of crop species, based on composite functions for the
                50th and 75th percentiles of 34 Weibull curves for relative yield loss
                data from 34 non-droughted NCLAN studies of 12 crop species;
                curves were standardized to 90-day W126

                              90-day 12-h W126 for 10% yield loss     90-day 12-h W126 for 20% yield loss
	(Ppm-h)	(ppm-h)	
 Model for the 50th Percentile of 34 curves	
 Relative yield=exp(-(W126/104.82)**1.424)	22	37	
 Model for the 75th Percentile of 34 curves	
 Relative yield=exp(-(W126/78.12)**1.415)	16	27	
  Source of parameters for the 34 curves: Lee and Hogsett (1996)
 Table 9-10     Ozone exposures at which 10 and 20% yield loss is predicted for 50
                and 75% of crop species under drought conditions and adequate
                moisture, based on composite functions for the 50th and 75th
                percentiles of 16 Weibull curves for relative yield loss data from 8
                NCLAN studies that paired droughted and watered conditions for
                the same genotype; curves were standardized to 90-day W126
                                             90day12-hW126for10%  90 day 12-h W126 for 20%
	yield loss (ppm-h)	yield loss (ppm-h)
 Model for the 50th Percentile of 2x8 curves	
 Watered	Relative yield=exp(-(W126/132.86)**1.170)	19	37	
 Droughted     Relative yield=exp(-(W126/179.84)**1.713)	48	75	
 Model for the 75th Percentile of 2x8 curves	
 Watered	Relative yield=exp(-(W126/90.43)**1.310)	16	29	
 Droughted     Relative yield=exp(-(W126/105.16)"1.833)	31	46	

 Source of parameters for the 16 curves: Lee and Hogsett (1996)
 Draft - Do Not Cite or Quote                    9-129                             September 2011

-------
      Table 9-11     Ozone exposures at which 10 and 20% biomass loss is predicted
                     for 50 and 75 %of tree species, based on composite functions for
                     the 50th and 75th percentiles of 49 Weibull curves for relative
                     above-ground biomass loss data from 49 studies of 11 tree species
                     grown under well-watered conditions for 1  or 2 year; curves were
                     standardized to 90-day W126
                                      90 day 12 h W126 for 10% yield     90 day 12 h W126 for 20% yield
     	loss (ppm-h)	loss (ppm-h)	
      Model for the 50th Percentile of 49 curves
      Relative yield=exp(-(W126/131.57)**! .242)             21                             39
      Model for the 75th Percentile of 49 curves
      Relative yield=exp(-(W126/65.49)**! .500)               15                             24
       Source of parameters for the 49 curves: Lee and Hogsett (1996)
            9.6.3  Validation of 1996 and 2006 Ozone AQCD Models and Methodology
                   Using the 90 day 12-h W126 and Current FACE Data

 1                 Since the completion of the NCLAN and NHEERL/WED projects, almost no studies
 2                 have been published that could provide a basis for estimates of exposure-response that
 3                 can be compared to those of the 1996 and 2006 O3 AQCDs. Most experiments,
 4                 regardless of exposure methodology, include only two levels of exposure. In addition,
 5                 very few studies have included measurements of exposure using the W126 metric, or the
 6                 hourly O3 concentration data that would allow computing exposure using the W126. Two
 7                 FACE projects, however, were conducted over multiple years, and by adding to the
 8                 number of exposure levels over time, may support independent model estimation and
 9                 prediction using the same model and the same robust process as summarized in Section
10                 9.6.2. Hourly O3 data were available from both FACE projects.

11                 The SoyFACE project is situated near Champaign, IL, and comprises 32 octagonal rings
12                 (20m-diameter), 4 of which in a given year are exposed to ambient conditions, and 4 of
13                 which are exposed to elevated O3 as a fixed proportion of the instantaneous ambient
14                 concentration (Betzelberger et al., 2010; University of Illinois. 2010; Morgan et al., 2006;
15                 Morgan et al.. 2004). Since 2002, yield data have been collected for up to 8 genotypes of
16                 soybean grown in subplots within each ring. The Aspen FACE project is situated in
17                 Rhinelander, WI, and comprises 12 rings (30m-diameter), 3 of which are exposed to
18                 ambient conditions, and 3 of which are exposed to O3 as a fixed proportion of the
19                 instantaneous ambient concentration (Pregitzer et al., 2008; Karnosky et al., 2005;
20                 Dickson et al.. 2000). In the summer of 1997, half the area of each ring was planted with
21                 small (five to seven leaf sized)  clonally propagated plants of five genotypes of trembling
      Draft - Do Not Cite or Quote                     9-130                              September 2011

-------
 1                   aspen, which were left to grow in those environments until 2009. Biomass data are
 2                   currently available for the years 1997-2005 (King et al.. 2005). Ozone exposure in these
 3                   two FACE projects can be viewed as a categorical variable with two levels: ambient, and
 4                   elevated. However, this overlooks the facts that yearly ambient and elevated exposure
 5                   both vary with every year, and that the proportionality between them also changes. This
 6                   change has two sources: first, the dispensing of O3 into the elevated exposure rings varies
 7                   from the proportionality set point to some extent, and for SoyFACE, the set point
 8                   changed between years.  Second, the proportionality does not propagate predictably from
 9                   the hourly data to the yearly value when using threshold or concentration-weighted
10                   cumulative metrics (such as AOT40, SUM06 or W126). Hourly average elevated
11                   exposures that are, for example, a constant  1.5 times greater than ambient do not result in
12                   AOT40, SUM06 or W126 values that are some constant multiple of the ambient values of
13                   those indices. The greater the fraction of elevated hourly values that are above the
14                   threshold or heavily weighted, compared to the fraction of hourly ambient values that are,
15                   the greater the difference between ambient and elevated yearly exposure, as measured
16                   using weighted cumulative indices. When elevated exposure is a multiple of ambient
17                   hourly intervals, the number of hours for which elevated exposure meets the threshold for
18                   inclusion can vary widely, even though the  hourly mean for  the year retains the
19                   proportionality. As a consequence, the number of exposure levels in multi-year
20                   experiments is twice the number of years. In the case of SoyFACE for the period between
21                   2002 and 2008, ambient exposure in the highest year was approximately equal to elevated
22                   exposure in the lowest year, with 14 levels of O3 exposure evenly distributed from lowest
23                   to highest. The particular conditions of the Aspen FACE experiment resulted in 12
24                   exposure levels between 1998 and 2003, but they were not as evenly distributed between
25                   minimum and maximum over the 6-year period.

26                   There are necessary differences in the modeling of exposure-response in annual plants
27                   such as soybean, and in perennial plants such as aspen trees, when exposure takes place
28                   over multiple years. In annual plants, responses recorded at the end of the life cycle, i.e.,
29                   yearly, are  analyzed in relationship to that year's exposure. Yield of soybeans is affected
30                   by exposure during the year the crop was growing, and a new crop is planted every year.
31                   Thus an exposure-response relationship  can be modeled from yearly responses matched
32                   to yearly exposures, with those exposure-response data points having been generated in
33                   separate years. For perennial organisms, which are not harvested yearly and continue to
34                   grow from  year to year, such pairing of exposure and response cannot be done without
3 5                   accounting for time. Not only does the size  of the organism at the beginning of each year
36                   of exposure increase, but size is also dependent on the exposure from previous years.
37                   Therefore the relationship of response and exposure must be analyzed either one year at a
38                   time, or by standardizing the response as a yearly increment relative to size at the
39                   beginning of each year. Furthermore, the relevant measurement of exposure is

      Draft - Do Not Cite or Quote                      9-131                                September 2011

-------
 1                  cumulative, or cumulative yearly average exposure, starting in the year exposure was
 2                  initiated, up to the end of the year of interest. When analyzing the growth of trees over
 3                  several years, it would be evidently incorrect to pair the exposure level in every discrete
 4                  year with absolute size of the trees that year, and posit a direct relationship between them,
 5                  without taking increasing age into consideration. In the Aspen FACE experiment, for
 6                  example, one could not establish an exposure-response relationship by matching
 7                  12 yearly exposures and 12 yearly tree sizes, while disregarding age as if size did not also
 8                  depend on it. This is the basis of the 2007 study of Aspen FACE data by Percy et al.
 9                  (2007). which compares the size of trees of various  ages as if they were all the same age,
10                  and was therefore not informative.
                    9.6.3.1    Comparison of NCLAN-Based Prediction and SoyFACE
                               Data

11                  For this ISA, EPA conducted a comparison between yield of soybean as predicted by the
12                  composite function three-step process (Section 9.6.2) using NCLAN data, and
13                  observations of yield in SoyFACE. The median composite function for relative yield was
14                  derived for the 11 NCLAN soybean Weibull functions for non-droughted studies, and
15                  comparisons between the predictions of the median composite and SoyFACE
16                  observations were conducted as follows.

17                  For the years 2007 and 2008, SoyFACE yield data were available for 7 and 6 genotypes,
18                  respectively. The EPA used those data to compare the relative change in yield observed
19                  in SoyFACE in a given year between ambient O3 and elevated O3, versus the relative
20                  change in yield predicted by the NCLAN-based median composite function between
21                  those same two values  of O3 exposure. The two parameter median composite function for
22                  relative yield of soybean based on NCLAN data was used to predict yield response at the
23                  two values of exposure observed in SoyFACE in each year, and the change between yield
24                  under ambient  and elevated was compared to the change observed in SoyFACE for the
25                  relevant year (Table 9-12). This approach results in a direct comparison of predicted
26                  versus observed change in yield. Because the value of relative response between any two
27                  values of O3 exposure is independent of the intercept a, this comparison does not require
28                  prediction of the absolute values of the responses.

29                  Since comparisons of absolute values might be of interest, the predictive functions were
30                  also scaled to the observed data: SoyFACE data were used to compute an intercept a
31                  while the shape and scale parameters (|3 and r\) were held at their value in the NCLAN
32                  predictive model. This method gives a comparison of prediction and observation that
33                  takes all the observed information into account to provide the best possible estimate of
      Draft - Do Not Cite or Quote                      9-132                                September 2011

-------
 1
 2
 3
 4
 5
 6
 7
 9
10
11
    the intercept, and thus the best possible scaling (Table 9-13 and Figure 9-18). For the
    comparison of NCLAN and SoyFACE, this validation was possible for 2007 and 2008,
    where data for 7 and 6 soybean genotypes, respectively, were available. The median
    composite function for relative yield was derived for the 11 NCLAN soybean Weibull
    functions for nondroughted studies, and the values of median yield under ambient
    exposure at SoyFACE in 2007 and 2008 were used to obtain an estimate of the intercept
    a for the NCLAN median function in each of the two years.

    Table 9-12 presents the results of ambient/elevated relative yield comparisons between
    the NCLAN-derived predictions and SoyFACE observations. Table 9-13 and figure 9-18
    present the results of comparisons between NCLAN-derived predictions and SoyFACE
    observations of yield, with the predictive function scaled to provide absolute yield values.
     Table 9-12    Comparison between change in yield observed in the SoyFACE
                    experiment between elevated and ambient ozone, and change
                    predicted at the same values of ozone by the median composite
                    function for NCLAN (two-parameter relative yield model)
Year
90-day 12-hW1 26 (ppm-h)
observed at SoyFACE
Ambient Elevated
2007
2008
4.39 46.23
3.23 28.79
Yield in
Elevated O3 Relative to Ambient O3 (%)
Predicted by NCLAN Observed at SoyFACE
75 76
85 88
     Table 9-13    Comparison between yield observed in the SoyFACE experiment
                    and yield predicted at the same values of ozone by the median
                    composite function for NCLAN (three-parameter absolute yield
                    model with intercept scaled to SoyFACE data)
       Year
  90-day 12-h W126 (ppm-h)
   observed at SoyFACE
            Yield predicted by NCLAN (g/m2)
                            Yield observed at SoyFACE
                            	(g/m2)	
              Ambient
            Elevated
            Ambient
             Elevated
             Ambient
             Elevated
     2007
4.39
46.23
309.2
230.6
305.2
230.6
     2008
3.23
28.79
350.3
298.2
344.8
304.4
     Draft - Do Not Cite or Quote
                                9-133
                                                    September 2011

-------
          400 -

          350 -

          300 -

          250 -

          200 -

          150 -

          100 -

           50 -

           0 -
                  2007, 7 genotypes
  400

  350

  300

_ 250
^E
3 200 -
1
*~ 150 -

  100 -

  50 -

   0
                                                                  2008, 6 genotypes
10    20    30    40   50

      90day12hrW126 (ppm-hr)
                                                                     20    30    40    50

                                                                       90 day 12 hr W126 (ppm-hr)
       Source of data: Betzelberger et al. (2010): Morgan et al. (2006): Lee and Hogsett (1996).
       Note: Black dots are the median of 7 or 6 soybean genotypes in SoyFACE (2007, 2008); bars are IQR for genotypes; dashed line
      is median composite model for 11 studies in NCLAN.

      Figure 9-18   Comparison of yield observed in SoyFACE experiment in a given
                      year with yield predicted by the median composite function based
                      on NCLAN.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
   Finally, a composite function for the 25th, 50th, and 75th percentiles was developed from
   SoyFACE annual yield data, and compared to the NCLAN-based function. The process
   described in Section 9.6.2 was applied to SoyFACE data for individual genotypes,
   aggregated over the years during which each was grown; one genotype from 2003 to
   2007, and six genotypes in 2007 and 2008. First, the three parameter Weibull model
   described in Section 9.6.2 was estimated using nonlinear regression on exposure-yield
   data for each genotype separately, over the years for which data were available, totaling
   seven curves. The 25th, 50th, and 75th percentiles of the predicted values for the two
   parameter relative yield curves were then identified at every integer of W126 between 0
   and 60, and a two-parameter Weibull model estimated by regression for the three
   quartiles. The comparison between these composite functions for the quartiles of relative
   yield loss in SoyFACE and the corresponding composite functions for NCLAN is
   presented in Figure 9-19.
      Draft - Do Not Cite or Quote
                                  9-134
                               September 2011

-------
                  100

                   90 •

                   80 •

               g  70 •
               (/)
               §  60 •

               1  50 •

               |  40 -

               I  30

                   20

                   10

                    0 -I
                                                               FACE75thPctile
                                                               FACE 25th Pctile
                                                                        FACE median
                       10      20      30      40      50
                                  90 day 12hrW126 (ppm-hr)
                                                             60
                                                                     70
 Source of data: Betzelberger et al. (2010); Morgan et al. (2006): Lee and Hogsett (1996).

Figure 9-19    Comparison of composite functions for the quartiles of 7 curves for
                7 genotypes of soybean grown in the SoyFACE experiment, and for
                the quartiles of 11 curves for 5 genotypes of soybean grown in the
                NCLAN project.
 1
 2
 3
 4
 5
 6
 1
 8
 9
10
11
12
13
14
15
16
              As seen in Tables 9-13 and 9-14, and in Figure 9-18, the agreement between predictions
              based on NCLAN data and SoyFACE observations was notably close in single-year
              comparisons. Together with the very high agreement between median composite models
              for NCLAN and SoyFACE (Figure 9-19), it provides very strong mutual confirmation of
              those two projects' results with respect to the response of yield of soybeans to O3
              exposure. It is readily apparent from these results that the methodology described in
              Section 9.6.2 for obtaining predictions of yield or yield loss from NCLAN data is
              strongly validated by SoyFACE results. As described in Section 9.2, the exposure
              technologies used in the two projects were in sharp contrast, specifically with respect to
              the balance each achieved between control of potential interacting factors or confounders,
              and fidelity to natural conditions. The comparisons that EPA conducted therefore
              demonstrate that the methodology used in developing the composite functions is resistant
              to the influence of nuisance  variables and that predictions are reliable. They may also
              suggest that the aspects in which the two exposure technologies differ have less influence
              on exposure-response than initially supposed. These results are also in agreement with
              comparative studies reviewed in 9.2.6.
Draft - Do Not Cite or Quote
                                                 9-135
September 2011

-------
                    9.6.3.2   Comparison of NHEERL/WED-Based Prediction of Tree
                              Biomass Response and Aspen FACE Data

 1                  EPA also conducted two comparisons between prediction of above-ground biomass loss
 2                  based on NHEERL/WED results and observations from Aspen FACE. The median
 3                  composite function was developed from NHEERL/WED data for 11 studies that used
 4                  wild-type seedlings of aspen as well as four clonally propagated genotypes. All plants
 5                  were grown in OTCs for one growing season before being destructively harvested. Aspen
 6                  FACE data were from clonally propagated trees of five genotypes grown from 1998 to
 7                  2003, with above-ground biomass calculated using allometric  equations derived from
 8                  data for trees harvested destructively in 2000 and 2002 (King et al.. 2005).

 9                  The two parameter median composite function for relative biomass was used to predict
10                  biomass response under the observed elevated exposure, relative to its value under
11                  observed ambient exposure, for each separate year of Aspen FACE. EPA first compared
12                  Aspen FACE observations of the change in biomass between ambient and elevated
13                  exposure with the corresponding prediction at the same values of exposure. Comparisons
14                  between observed and predicted absolute biomass values were then conducted for each
15                  year by scaling the predictive function to yearly Aspen FACE data as described for
16                  soybean data in Section 9.6.3.1. In all cases, yearly 90 day 12-hour W126 values for
17                  Aspen FACE were computed as the  cumulative average from the year of planting up to
18                  the year of interest. A comparison of composite functions between NHEERL/WED and
19                  Aspen FACE, similar to the one performed for NCLAN and SoyFACE, was not possible:
20                  as discussed in the introduction to Section 9.6, the pairing of 12 exposure values from
21                  separate years and 12 values of biomass cannot be the basis for a model of exposure-
22                  response, because the trees continued growing for the six-year period of exposure.
23                  Because the same trees were used for the entire duration, and continued to grow, data
24                  could not be aggregated over years. Table 9-14 presents the results of ambient/elevated
25                  relative biomass comparisons between the NHEERL/WED-derived predictions and
26                  Aspen FACE observations. Table 9-15 and Figure 9-20 present the results of the
27                  comparison between NHEERL/WED-derived predictions and Aspen  FACE observations
28                  for absolute biomass, using Aspen FACE data to scale the NHEERL/WED-derived
29                  composite function.
      Draft - Do Not Cite or Quote                      9-136                              September 2011

-------
Table 9-14    Comparison between change in above-ground biomass elevated
              and ambient ozone in Aspen FACE experiment in 6 year, and
              change predicted at the same values of ozone by the median
              composite function for NHEERL/WED (two-parameter relative
              biomass model)
   Year
        90-day 12-h W126 (ppm-h)
Cumulative Average observed at Aspen FACE
                                Above-Ground Biomass in
                            Elevated O$ relative To Ambient O$ (%)

1998
1999
2000
2001
2002
2003
Ambient
3.19
2.61
2.43
2.55
2.51
2.86
Elevated
30.08
33.85
30.16
31.00
30.27
29.12
Predicted by NHEERL/WED
74
70
74
73
74
75
Observed at Aspen FACE
75
70
71
71
69
71
Table 9-15    Comparison between above-ground biomass observed in Aspen
              FACE experiment in 6 year and biomass predicted by the median
              composite function based on NHEERL/WED (three-parameter
              absolute biomass model with intercept scaled to Aspen FACE data)
            90day12-hW126(ppm-h)
   Year    Cumulative Average observed
  	at Aspen FACE	
                           Biomass Predicted by
                           NHEERL/WED (g/m"
                                       Biomass Observed, at Aspen
                                            FACE (g/n?)
          Ambient
           Elevated
           Ambient
             Elevated
             Ambient
             Elevated
1998
3.19
30.08
276.0
203.2
274.7
204.9
1999
2.61
33.85
958.7
668.3
955.3
673.3
2000
2.43
30.16
1382.4
1022.8
1400.3
998.6
2001
2.55
31.00
1607.0
1173.7
1620.7
1154.9
2002
2.51
30.27
2079.0
1532.1
2125.9
1468.41
2003
2.86
29.12
2640.1
1981.2
2695.2
1907.8
Draft - Do Not Cite or Quote
                             9-137
                                                 September 2011

-------
                   3000 n
                   2500 -
                   2000 -
     •*-...,
                   1000 -
                    500 -
     -*--.            *^
1500 -       - - .. „
        * ^ •» ^     ^ ••» ^
                 ^ •* ^    ^ ^
                                              " f
                                              "it
                                                    2003
                                                    2002
                                 2001
                                 2000
                                                    1999
                                                    1998
                               10      20      30      40      50      60

                               90 day 12 hr W126 (yearly cumulative average, ppm-hr)
                                                            70
       Source of data: King et al. (2005). Lee and Hogsett (1996).
       Note: Black dots are aspen biomass/m2 for 3 FACE rings filled with an assemblage of 5 clonal genotypes of aspen at Aspen
      FACE; bars are SE for 3 rings; dashed line is median composite model for 4 clonal genotypes and wild-type seedlings in 11
      NHEERL/WED 1-year OTC studies.

      Figure  9-20    Comparison between above-ground biomass observed in Aspen
                      FACE experiment in 6 year and biomass predicted by the median
                      composite function based on NHEERL/WED.
 1
 2
 3
 4
 5
 6
 7
 8

 9
10
11
12
13
14
 As in the comparisons between NCLAN and SoyFACE, the agreement between
 predictions based on NHEERL/WED data and Aspen FACE observations was very close.
 The results of the two projects strongly reinforce each other with respect to the response
 of aspen biomass to O3 exposure. The methodology used for obtaining the median
 composite function is shown to be capable of deriving a predictive model despite
 potential confounders, and despite the added measurement error that is expected from
 calculating biomass using allometric equations. In addition, the function based on
 one year of growth was shown to be applicable to subsequent years.

 The results of experiments that used different exposure methodologies, different
 genotypes, locations, and durations converged to the same values of response to O3
 exposure for each of two very dissimilar plant species, and predictions based on the
 earlier experiments were validated by the data from current ones. However, in these
 comparisons, the process used in establishing predictive functions involved aggregating
 data over variables such as time,  locations, and genotypes, and the use of a robust statistic
      Draft - Do Not Cite or Quote
                              9-138
September 2011

-------
 1                  (quartiles) for that aggregation. The validating data, from SoyFACE and Aspen FACE,
 2                  were in turn aggregated over the same variables. The accuracy of predictions is not
 3                  expected to be conserved for individual values of those variables over which aggregation
 4                  occurred. For example, the predicted values for soybean, based on data for five
 5                  genotypes, are not expected to be valid for each genotype separately. As shown in the
 6                  validation, however, aggregation that occurred over different values of the same variable
 7                  did not affect accuracy: composite functions based on one set of genotypes were
 8                  predictive for another set, as long as medians were used for both sets. A study of
 9                  cottonwood (Populus deltoides) conducted using a naturally occurring gradient of O3
10                  exposure (Gregg et al.. 2006. 2003) may provide an illustration of the response of an
11                  individual species whose response is far from the median response for an aggregation of
12                  species.
                    9.6.3.3   Exposure-Response in a Gradient Study

13                  Gregg et al. (2003) grew saplings of one clonally propagated genotype of cottonwood
14                  (Populus deltoides) in seven locations within New York City and in the surrounding
15                  region between July and September in 1992, 1993 and 1994, and harvested them 72 days
16                  after planting. Owing to regional gradients of atmospheric O3 concentration, the
17                  experiment yielded eight levels of exposure (Figure 9-21), and the authors were able to
18                  rule out environmental variables  other than O3 to account for the large differences in
19                  biomass observed after one season of growth. The deficit in growth increased
20                  substantially faster with increasing O3 exposure than has been observed in aspen, another
21                  species of the same genus (Populus tremuloides, Section 9.6.3.2). Using a three
22                  parameter Weibull model (Figure 9-21), the biomass of cottonwood at a W126 exposure
23                  of 15 ppm-h, relative to biomass  at 5 ppm-h, is estimated to be 0.18 (18% of growth at
24                  5 ppm-h). The relative biomass of trembling aspen within the same 5-15 ppm-h range of
25                  exposure is estimated to be 0.92, using the median composite model for aspen whose
26                  very close agreement with Aspen FACE data was shown in Section 9.6.3.2. Using a
27                  median  composite function for all deciduous trees in the NHEERL/WED project (6
28                  species in 21 studies) also gives predictions that are very distant from the cottonwood
29                  response observed in this experiment. For all deciduous tree species in NHEERL/WED,
30                  biomass at a W126 exposure of 15 ppm-h, relative to biomass at 5 ppm-h, was estimated
31                  to be 0.87.
      Draft - Do Not Cite or Quote                     9-139                               September 2011

-------
                           10      20      30      40      50

                                    72 day 12 hr W126 (ppm-hr)
                                                               60
                                                                       70
 Source: Modified with permission from Nature Publishing Group (Gregg et al.. 2003).

Figure 9-21    Above-ground  biomass for one genotype of cottonwood grown in
                seven locations for one season in 3 years.  Line represents the
                three-parameter Weibull model.
 1
 2
 3
 4
 5
 6
 7

 8
 9
10
11
12
13
14
15
16
17
              These cottonwood data confirm that, as should be expected, some individual tree species
              are substantially more sensitive than the median of NHEERL/WED (Figure 9-16). As
              shown in Section 9.6.2, the median models available for trembling aspen and soybean
              have verifiable predictive ability for those particular species. This suggests that the
              corresponding NCLAN- and NHEERL/WED-based models for multiple crop and tree
              species can provide reliable estimates of losses for similar assortments of species.
              However, their predictive ability would likely be poor for individual species not tested.

              An alternative hypothesis for the difference between the response of cottonwood in this
              experiment and deciduous tree species in NHEERL/WED, or the difference between the
              response of cottonwood and aspen in NHEERL/WED and Aspen FACE, could be the
              presence of confounding factors in the environments where the experiment was
              conducted. However, variability in temperature, moisture, soil fertility, and atmospheric
              deposition of N were all ruled out by Gregg et al. (2003) as contributing to the  observed
              response to O3. In addition, this hypothesis would imply that the unrecognized
              confounder(s)  were either absent from both OTC and FACE studies, or had the same
              value in both. This is not impossible, but the hypothesis that cottonwood is very sensitive
              to O3 exposure is more parsimonious, and sufficient.
Draft - Do Not Cite or Quote
                                                 9-140
September 2011

-------
                    9.6.3.4   Meta-analyses of growth and yield studies
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
                    Since the 2006 O3 AQCD, five studies have used meta-analytic methods to integrate
                    results from experimental studies of crops or tree species relevant to the U.S. It is
                    possible to obtain exposure-response data for growth and yield from those meta-analyses,
                    but because all of them provided summary measurements of O3 exposure as hourly
                    averages of various lengths of exposures, comparisons with exposure-response results
                    where exposure is expressed as W126 are problematic. Table 9-16 summarizes the
                    characteristics of the five meta-analyses. They all included studies conducted in the U.S.
                    and other locations worldwide, and all of them expressed responses as comparative
                    change between levels of exposure to O3,  with carbon filtered air (CF) among those
                    levels. Using hourly average concentration to summarize exposure, CF rarely equates
                    absence of O3, although it almost always near zero when exposure is summarized as
                    W126, SUM06, or AOT40.
Table 9-1 6
Study
Ainsworth (2008)
Fena et al. (2008b)
Feng and
Kobavashi (2009)
Grantz et al. (2006)
Wittig et al. (2009)
Meta-analyses of growth or yield studies published since
Number of articles
included
12
53
All crops together : 81
16
All responses:263
Articles that included
biomass:unreported
Years pf
1980-2007
1980-2007
1980-2007
1992-2004
1970-2006
Crop, species or
genera
rice
wheat
Potato, barley, wheat, rice,
bean, soybean
34 herbaceous dicots
21 herbaceous monocots
5 tree species
4 gymnosperm tree genera
11 angiosperm tree genera
Response
Yield
Yield
Yield
Relative
Growth Rate
Total
biomass
Number
OfC-3
levels
2
5
3
2
4
2005
Duration of
exposure
unreported
> 10 days
> 10 days
2-24 weeks
> 7 days
13
14
15
16
17
18
19
20
21
22
23
24
                    The only effect of O3 exposure on yield of rice reported in Ainsworth (2008) was a
                    decrease of 14% with exposure increasing from CF to 62 ppb average concentration.
                    Feng et al. (2008b) were able to separate exposure of wheat into four classes with average
                    concentrations of 42, 69, 97, and 153 ppb, in data where O3 was the only treatment. Mean
                    responses relative to CF were yield decreases of 17, 25, 49, and 61% respectively. Feng
                    et al. (2008b) observed that wheat yield losses were smaller under conditions of drought,
                    and that Spring wheat and Winter wheat appeared similarly affected. However, mean
                    exposure in studies of Winter wheat was substantially higher than in studies of Spring
                    wheat (86 versus 64 ppb), which suggests that the yield of Spring wheat was in fact more
                    severely affected, since yield was approximately the same, even though Spring wheat was
                    exposed to lower concentrations. Exposures of the six crops considered in Feng and
                    Kobayashi (2009) were classified into two ranges, each compared to CF air. In the lower
      Draft - Do Not Cite or Quote
                                                  9-141
September 2011

-------
 1                  range of exposure (41-49 ppb), potato studies had the highest average exposure (45 ppb)
 2                  and wheat and rice the lowest (41 ppb). In the higher range (51-75 ppb), wheat studies
 3                  had the highest average exposure (65 ppb), and potato, barley and rice the lowest (63
 4                  ppb). In other words, across the studies included, all crops were exposed to very similar
 5                  levels of O3. At approximately 42 ppb, the yield of potato, barley, wheat, rice, bean, and
 6                  soybean declined by 5.3, 8.9, 9.7, 17.5, 19, and 7.7% respectively, relative to CF air. At
 7                  approximately 64 ppb O3, declines were 11.9,  12.5, 21.1, 37.5, 41.4, and 21.6%. Grantz
 8                  et al. (2006) reported Relative Growth Rate (RGR) rather than growth, and did not report
 9                  O3 exposure values in a way that would allow  calculation of mean exposure for each of
10                  the three categories of plants for which RGR changes are reported. All studies used only
11                  two levels of exposure, with CF air as the lower one, and most used elevated exposure in
12                  the range of 40 to 70 ppb. Decline in RGR was 8.2% for the 34 herbaceous dicots, 4.5%
13                  for the 21 herbaceous monocots, and 17.9% for the 5 tree species. Finally, Wittig et al.
14                  (2009) divided the studies analyzed into three classes of comparisons: CF versus ambient,
15                  CF versus elevated, and ambient versus elevated, but reported comparisons between three
16                  average levels of exposure besides CF: 40 ppb, 64 ppb, and 97 ppb. Corresponding
17                  decreases in total biomass relative to CF were 7, 17, and 17%.

18                  These meta-analyses provide very strong confirmation of EPA's conclusions from
19                  previous O3 AQCDs: compared to lower levels of ambient O3, current levels in many
20                  locations are having a substantial detrimental effect on the growth and yield of a wide
21                  variety of crops and natural vegetation. They also confirm strongly that decreases in
22                  growth and yield continue at exposure levels higher than current ambient levels.
23                  However, direct comparisons with the predictions of exposure-response models that use
24                  concentration-weighted cumulative metrics are difficult.
                    9.6.3.5   Additional exposure-response data

25                  The studies summarized in Tables 9-18 and 9-19 contain growth or yield exposure-
26                  response data at too few levels of exposure for exposure-response models, and/or used
27                  metrics other than W126. These tables update Tables AX9-16 through AX9-19 of the
28                  2006 O3 AQCD.
            9.6.4  Summary

29                  None of the information on effects of O3 on vegetation published since the 2006 O3
30                  AQCD has modified the assessment of quantitative exposure-response relationships that
31                  was presented in that document. This assessment updates the 2006 exposure-response
      Draft - Do Not Cite or Quote                     9-142                               September 2011

-------
 1                  models by computing them using the W126 metric, cumulated over 90 days. Almost all
 2                  of the experimental research on the effects of O3 on growth or yield of plants published
 3                  since 2006 used only two levels of exposure. In addition, hourly O3 concentration data
 4                  that would allow calculations of exposure using the W126 metric are generally
 5                  unavailable. However, two long-term experiments, one with a crop species (soybean),
 6                  one with a tree species (aspen), have produced data that can be used to validate the
 7                  exposure-response models presented  in the 2006 O3 AQCD, and methodology used to
 8                  derive them.

 9                  Quantitative characterization of exposure-response in the 2006 O3 AQCD was based on
10                  experimental data generated for that purpose by the National Crop Loss Assessment
11                  Network (NCLAN) and EPA National Health and Environmental Effects Research
12                  Laboratory, Western Ecology Division (NHEERL-WED) projects, using OTCs to expose
13                  crops and trees seedling to O3. In recent years, yield and growth results for two of the
14                  species that had provided extensive exposure-response information in those projects have
15                  become available from studies that used FACE technology, which is intended to provide
16                  conditions much closer to natural environments (Pregitzer et al.. 2008; Morgan et al..
17                  2006; Morgan et al.. 2004; Dickson et al.. 2000). The robust methods that were used
18                  previously with exposure measured as SUM06 were applied to the NCLAN and
19                  NHEERL-WED data with exposure measured as W126, in order to derive single-species
20                  median models for soybean and aspen from studies involving different genotypes, years,
21                  and locations. The resulting models were used to predict the change in yield of soybean
22                  and biomass of aspen between the two levels of exposure reported in current FACE
23                  experiments. Results from these new  experiments were exceptionally close to predictions
24                  from the models. The accuracy of model predictions for two widely different plant
25                  species provides support for the validity of the corresponding multiple-species models for
26                  crops and trees in the NCLAN and NHEERL-WED projects. However, variability among
27                  species in those projects  indicates that the range of sensitivity is likely quite wide. This
28                  was confirmed by a recent experiment with cottonwood in a naturally occurring gradient
29                  of exposure (Gregg et al., 2006), which established the occurrence of species with
30                  responses substantially more severe under currently existing conditions than are predicted
31                  by the median model for multiple species.

32                  Results from several meta-analyses have provided approximate values for responses of
33                  yield of soybean, wheat,  rice and other crops under broad categories of exposure, relative
34                  to charcoal-filtered air (Ainsworth. 2008;  Feng et al.. 2008b; Morgan et al..  2003).
35                  Likewise, Feng and Kobayashi (2009) have summarized yield data for six crop species
36                  under various broad comparative exposure categories, while Wittig et al. (2009) reviewed
37                  263 studies that reported effects on tree biomass. However, these analyses have proved
      Draft - Do Not Cite or Quote                      9-143                               September 2011

-------
1
2
difficult to compare with exposure-response models, especially given that exposure was
not expressed in the same W126 metric.
Table 9-1 7
Species
Facility
Location
Alfalfa (Medicago
sativa)
OTC; 0.27m3 pots
Federico, Italy
Bean (Phaseolus
vulgaris I . cv
Borlotto)
OTC; ground-
planted
Curno, Italy
Big Blue Stem
(Andropogon
gerardii)
OTC
Alabama
Brassica napus cv.
Westar
Growth chambers
Finland
Corn (Zea mays cv.
Chambord)
OTC
France
Cotton cv. Pima
OTC; 9-L pots
France
Eastern Gamagrass
(Tripsacum
dactybides)
OTC
Alabama
Grapevine (Vitis
vinivera)
OTC
Austria
Mustard (Brassica
campestris)
Chambers;
7.5-cm pots
Oilseed Rape
(Brassica napus)
OTC
Yangtze Delta,
China
Peanut (Arachis
hypogaea)
OTC
Raleigh, NC
Poa pratensis
OTC
Braunschweig,
Germany
Summary of studies of effects of ozone exposure on growth and
yield of agricultural crops
Exposure
Duration
2 yr, 2005,
2006
3 months,
2006
4 months,
2003
17-26 days
33 days
8wk
4 months,
2003
3 yr, May-Oct
10 days
39 days
Syr
2000-2002:
4-5 wk in the
Spring
O3 Exposure
(Additional Treatment)
AOT40: CF 0 ppm-h
1 3.9 ppm-h (2005), 10.1 ppm-h
(2006)
(NaCI: 0.29, 0.65, 0.83,
1 .06 deciSiemens/meter)
Seasonal AOT40:
CF (0.5 ppm-h);
ambient (4.6 ppm-h)
(N/A)
12-havg:
CF(14ppb),
Ambient (29 ppb),
Elevated (71 ppb)
(N/A)
8-h avg:
CF(Oppb), 100 ppb
(Bt/non-Bt; herbivory)
AOT40 ppm-h: 1.1; 1.3; 4.9;
7.2; 9.3; 12.8
(N/A)
1 2-h avg: 12.8 ±0.6; 79.9 ±
6.3; 122.7 ±9.7
(N/A)
12-havg:
CF (14ppb),
Ambient (29 ppb),
Elevated (71 ppb)
(N/A)
AOT40 ppm-h:
CF (0),
Ambient (7-20),
Elevated. 1 (20-30), Elevated.
2 (38-48)
CF&
67.8 ppb for 7 h
(N/A)
Daily avg: 100 ppb, one with
diurnal variation and one with
constant concentration
(N/A)
12-havg:
CF (22 ppb),
Ambient (46 ppb),
Elevated (75ppb)
(C02:375ppm;548ppm;
730 ppm)
8-h avg:
CF+25(21.7),
NF+50(73.1)
(Competition)
Response
Measured
Total shoot yield
# Seeds per plant;
100-seed weight
Final harvest
biomass;
RVF
Shoot biomass
Total above-ground
biomass
Above-ground
biomass
Final harvest
biomass;
RVF
Total fruit yield/
Sugar yield
Seeds/plant
Biomass and pods
per plant
Yield (seed weight,
g/m)
Total biomass (g
DW/pot)
percent change
from CF
(percent change
from ambient)
n.s. (N/A)
-33 (N/A)
n.s. (N/A)
n.s. (n.s.)
-7 (-7)
-30.70 (N/A)
N/A (Highest
treatment caused -
26% change)
-76 (n.s.)
+68 (+42);
-17 (-12)
-20 to -80 in different
yr
(-20 to -90 in different
yr)
n.s. (N/A)
Diurnal variability
reduced both
biomass and pod
number more than
constant fumigation
(N/A)
-33 (-8)
N/A (n.s.)
Reference
Maggio et al.
(2009)
Gerosa et al.
(2009)
Lewis et al.
(2006)
Himanen et al.
(2009b)
Leitao etal.
(2007c)
Grantz and
Shrestha
(2006)
Lewis et al.
(2006)
Sola et al.
(2004)
Black et al.
(2007)
Wang et al.
(2008)
Burkey et al.
(2007)
Bender etal.
(2006)
     Draft - Do Not Cite or Quote
                               9-144
September 2011

-------
Species
Facility
Location
Potato (Solarium
tuberosum)
OTC; CHIP
6 northern European
locations
Rice (Oryza saf/Va)
OTC
Raleigh, NC
Rice (Oryza saf/Va)
20 Asian cultivars
OTC
Gunma Prefecture,
Japan
Seminatural grass
FACE
Le Mouret,
Switzerland
Soybean
OTC; CRA
Bari, Italy
Soybean (Glycine
maxcv. 93B15)
SoyFACE
Urbana, IL
Soybean (Glycine
max cv. Essex)
Chambers; 21 L
Raleigh, NC
Soybean (Glycine
max cv. Essex)
OTCs;21-Lpots
Raleigh, NC
Soybean (Glycine
max cv. Tracaja)
Chambers; pots
Brazil
Soybean (Glycine
max) 10 cultivars
SoyFACE
Urbana, IL
Spring Wheat
(Triticum aestivum
cv. Minaret; Satu;
Drabant; Dragon)
OTCs
Belgium, Finland, &
Sweden
Strawberry (Fragaria
xananassa Duch.
Cv Korona &
Elsanta)
Growth chambers
Bonn, Germany
Sugarbeet (Befa
vulgaris cv. Patriot)
OTC
Belgium
Sugarcane
(Saccharum spp)
CSTR
San Joaquin Valley,
CA
Exposure
Duration
1988,1999.
Emergence to
harvest
1997-1998,
June-
September
2008 growing
season
Syr
2003-2005
growing
seasons
2002, 2003
growing
seasons
2x3 months
2x3 months
20 days
2007 & 2008
1990-2006
2 months
2003, 2004;
5 months
2007;
11-13wk.
Os Exposure
(Additional Treatment)
AOT40:CF (0);
Ambient (0.27-5.19); NF
(0.002-2.93)
NF+ (3.10-24.78
(N/A)
12-hmean ppb:
CF (27.5),
Elevated (74.8)
(CO,
Daily avg (ppb):
CF (2),
O.Sxambient (23);
1 xambient (28);
1.5xambient(42);
2xambient (57)
(Cultivar comparisons)
Seasonal AOT40: Ambient
(0.1-7.2ppm-h);
Elevated. (1.8-24.1 ppm-h)
(N/A)
Seasonal AOT40 ppm-h: CF
(0),
Ambient (3.4), High (9.0)
(Drought)
8-h avg:
Ambient (62 & 50 ppb),
Elevated (75 & 63 ppb)
(N/A)
12-havg:CF(28),
Elevated (79),
Elevated flux (11 2)
(C02: 365 & 700)
12-havg:CF(18);
Elevated (72)
(C02:367&718)
12-havg:CF&30ppb
(N/A)
8-h avg: Ambient (46.3 & 37.9),
Elevated (82.5 & 61 .3)
(Cultivar comparisons)
Seasonal AOT40s ranged from
0 to 16 ppm-h
(N/A)
8-h avg: CF (0 ppb) &
Elevated (78 ppb)
(N/A)
8-h avg: Ambient (36 ppb);
Elevated (62 ppb)
(N/A)
12-havg:CF(4ppb);
Ambient (58);
Elevated (147)
(N/A)
Response
Measured
Tuber yield averaged
across 5 field-sites;
Tuber starch content
regressed against
[03]reportsig.
± slope with
increasing [03]
Total biomass;
Seed yield
Yield
Relative annual yield
Yield
Yield
Seed mass per plant
Seed mass per plant
Biomass
Yield
Seed protein
content;
1 ,000-seed weight
regressed across all
experiments
Fruit yield
(weight/plant)
Sugar yield
Total biomass
(g/plant)
percent change
from CF
[percent change
from ambient)
N/A (-27 % -+27%,
most comparisons
n.s.) Linear
regression slope =
-0.0098)
-25(N/A)
-13 to 20 (N/A)
From n.s. to -30
across all cultivars
N/A (2xfaster
decrease in yield/yr)
-46 (-9)
N/A
(-15 in 2002;
-25 in 2003)
-30 (N/A)
-34 (N/A)
-18 (N/A)
N/A (-17.20)
N/A (significant
negative correlation)
N/A (sig negative
correlation)
-16 (N/A)
N/A (-9)
-40 (-30)
Reference
Vandermeiren
et al. (2005)
Reid, etal.
(2008)
Sawada and
Kohno (2009)
Volketal.
(2006)
Bou Jaude
et al. (2008)
Morgan et al.
(2006)
Booker and
Fiscus (2005)
Booker etal.
(2004a)
Bulbovas et al.
(2007)
Betzelberger
et al. (2010)
Piikki et al.
(2008a)
Keutgenetal.
(2005)
De
Temmerman
et al. (2007)
Grantzand Vu
(2009)
Draft - Do Not Cite or Quote
9-145
September 2011

-------
Species
Facility
Location
Sweet Potato
Growth chambers
Bonn, Germany
Tomato
(Lycopersicon
esculentum)
OTC
Valencia, Spain
Trifolium
Subterraneum
OTC; 2.5-L pots
Madrid, Spain
Watermelon
(Citrullus lanatus)
OTC
Valencia, Spain
Yellow Nutsedge
OTC; 9-L pots
Exposure
Duration
4wk
133 days in
1998
29 days
2000, 2001 .
90 days
8wk
Os Exposure
(Additional Treatment)
8-h avg: CF (0 ppb),
Ambient (<40 ppb) Elevated
(255 ppb)
(N/A)
8-h mean ppb:
CF 16.3, NF 30.1,
NF+ 83.2
(Various cultivars; early & late
harvest)
12-havg:CF(<7.9±6.3);
Ambient (34.4±1 0.8);
Elevated (56.4±22.3)
(N: 5, 15 & 30 kg/ha)
AOT40: CF (0 ppm-h)
Ambient (5.7 ppm-h), Elevated
(34.1 ppm-h)
(N:0, 14.0 & 29.6 g/pot)
12-h avg: 12.8 ±0.6;
79.9 ±6.3; 122.7 ±9.7
(N/A)
Response
Measured
Tuber weight
Yield
Above-ground
biomass
total fruit yield (kg)
above-ground
biomass
percent change
from CF
[percent change
from ambient)
-14 (-11. 5)
n.s (n.s.)
-45 (-35)
n.s. (54)
n.s. (n.s.)
Reference
Keutgen et al.
(2008)
Calvo et al.
(2005)
Sanz et al.
(2005)
Calatayud
et al. (2006)
Grantz and
Shrestha
(2006)
 In studies where variables other than 03 were included in the experimental design, response to 03 is only provided for the control level of those
 variables.
Table 9-18    Summary of studies of effects of ozone exposure on growth of
                natural vegetation
Species
Facility
Location
Yellow nutsedge (Cyperus
esculentus)
CSTR
Parlier, CA
35 herbaceous species
OTC
Corvallis, OR
Highbush blackberry (Rubus
argutus)
OTC
Auburn, AL
Horseweed (Conyza
canadensis)
CSTR
San Joaquin Valley, CA
Red Oak (Quercus rubrum)
Forest sites
Look Rock & Twin Creeks
Forests, TN
Exposure
Duration
53 days in
2008
1999-2002,
May-August
2004,
May-August
2005, 2 runs,
28 days each
(July-Aug,
Sept)
2001-2003,
April-October
O3 Exposure
(Additional
Treatment)
12-h mean ppb:
CF (4); CF+ (60);
CF2+(115)
4-yr avg; yearly
W1 26 ppm-h:
CF (0),
CF+ (21),
CF 2+ (49.5)
12-h mean ppb:
CF(21.7),
Ambient (32.3),
Elevated (73.3)
W126ppm-hr:
CF(0),
CF+(11),
CF 2+ (30)
(Glyphosate
resistance)
AOT60:
2001 (11.5),
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
Response
Measured
Above-ground
biomass; tubers
(g/plant)
Total community
above-ground biomass
(35 species) after 4
years
Vegetative regrowth
after pruning
Total biomass (g/plant)
Annual circumference
increment (change
relative to 2001 in year
2002;2003)
Response
ns;CF(4.1)CF+(3.9)
CF2+(2.7)
CF (459 g/m2), CF+
(457 g/m3, CF2+
(398 g/m2)
CF (75.1 g/plant),
Ambient (76.4
g/plant),
Elevated (73.1
g/plant)
Glyphosate sensitive:
CF (0.354)
CF+ (0.197)
CF2+(0.106)
Glyphosate resistant:
CF(0.510)
CF+(0.313)
CF2+(0.143)
-42.8%; +1%
Reference
Grantz etal. (201 Ob)
Pfleegeretal. (2010)
Ditchkoff etal. (2009)
Grantz etal. (2008)
McLaughlin et al.
(2007a)
Draft - Do Not Cite or Quote
9-146
September 2011

-------
Species
Facility
Location


Pine species
Forest sites
Look Rock Forest, TN




Hickory species
Forest sites
Look Rock Forest, TN




Chestnut Oak (Quercus
prints)
Forest sites
Look Rock Forest, TN




Black Cherry (Prunus rigida)
Forest sites
Twin Creeks Forest, TN




Shortleaf pine (Pinus
echinata)
Forest sites
Twin Creeks Forest, TN




Hemlock (Tsuga
canadensis)
Forest sites
Twin Creeks Forest, TN




Red Maple (Acerrubrum)
Forest sites
Twin Creeks Forest, TN



Yellow Poplar (Liriodendron
tulipifera)
Forest sites
Look Rock, Oak Ridge, &
Twin Creeks Forest, TN

Exposure
Duration


2001-2003,
April-October




2001-2003,
April-October




2001-2003,
April-October




2002-2003,
April-October





2002-2003,
April-October






2002-2003,
April-October





2002-2003,
April-October



2002-2003,
April-October

Os Exposure
(Additional
Treatment)
AOT60:
2001 (11.5),
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2001 (11.5),
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2001 (11.5),
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
AOT60:
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
&!= ^sponse


Annual circumference
increment (change 62 5o/ . 2 9%
relative to 2001 in year '
2002;2003)




Annual circumference
increment (change 1 4% . ono,
relative to 2001 in year " ' * *' JU *
2002;2003)




Annual circumference
increment (change 440/ . .rr0,
relative to 2001 in year ^ *' °° *
2002;2003)




Annual circumference
increment (change 7j-0/
relative to 2003 in year ~'°*
2002)




Annual circumference
increment (change icao/
relative to 2003 in year •|b'b*
2002)




Annual circumference
increment (change 2i 9%
relative to 2003 in year " '
2002)




Annual circumference
increment (change r0 coi
relative to 2003 in year "OS'D*
2002)



Annual circumference
increment (change /ICQO/ icoco/
relative to 2001 in -40.y*, -IO.^OA
years 2002; 2003)

Reference


McLaughlin et al.
(2007a)




McLaughlin et al.
(2007a)




McLaughlin etal.
(2007a)




McLaughlin etal.
(2007a)





McLaughlin etal.
(2007a)






McLaughlin etal.
(2007a)





McLaughlin etal.
(2007a)



McLaughlin etal.
(2007a)

Draft - Do Not Cite or Quote
9-147
September 2011

-------
Species
Facility
Location
Sugar Maple (Acer
saccharum)
Forest sites
Twin Creeks Forest, TN
Trembling aspen (Populus
tremuloides), 5 genotypes
Aspen FACE
Rhinelander, Wl
Hybrid Poplar (Populus
trichocarpa x Populus
deltoides)
OTC
Seattle, WA
Exposure
Duration
2002-2003,
April-October
1998-2004,
May-October
2003,
3 months
Os Exposure
(Additional
Treatment)
AOT60:
2002 (24.0),
2003(11.7)
(Observational
study with
multiple
environmental
variables)
Cumulative avg
90-day 12-h
W126.
Ambient 3.1
ppm-h Elevated:
27.2 ppm-h
(Competition with
birch, maple)
Daily mean
(ug/g):
CF(<9),
Elevated (85-128)
MeaS ResP°"se
Annual circumference
increment (change Ro H0/
relative to 2003 in year ~M-ot>
2002)
main stem volume Ambient: 6.22 dm3.
after 7 years Elevated: 4.73 dm
Total biomass ^^etevated:
Reference
Mclaughlin etal.
(2007a)
Kubiske et al. (2006)
Woo and Hinckley
(2005)
  In studies where variables other than 03 were included in the experimental design, response to 03 is only provided for the control level of those
 variables.
    9.7    References

Aqrell. J: Kopper. BJ: McDonald. EP: Lindroth. RL (2005). CO2 and O3 effects on host plant preferences of the
        forest tent caterpillar (Malacosoma disstria). Global Change Biol 11: 588-599.
        http://dx.doi.0rg/10.1111/i.1365-2486.2005.00924.x.
Ahlfors.  R: Brosche. M: Kollist. H: Kangasiarvi. J. (2009). Nitric oxide modulates ozone-induced cell death,
        hormone biosynthesis and gene expression in Arabidopsis thaliana. Plant J 58: 1-12.
        http://dx.doi.Org/10.1111/i.1365-313X.2008.03756.x.
Ahsan. N: Nanio. Y:  Sawada. H: Kohno. Y: Komatsu. S. (2010). Ozone stress-induced proteomic changes in
        leaf total soluble and chloroplast proteins of soybean reveal that carbon allocation is involved in
        adaptation in the early developmental stage. Proteomics 10: 2605-2619.
        http://dx.doi. orq/10.1002/pmic.2Q1000180.
Ainsworth, EA: Long, SP (2005). What have we learned from 15 years of free-air CO2 enrichment (FACE)? A
        meta-analytic review of the responses of photosynthesis, canopy properties and plant production to
        rising CO2 [Review]. New Phytol 165: 351-371.
Ainsworth, EA: Rogers, A. (2007). The response  of photosynthesis and stomatal conductance to rising [CO2]:
        Mechanisms and environmental interactions [Review]. Plant Cell Environ 30: 258-270.
        http://dx.doi.0rg/10.1111/i.1365-3040.2007.01641.x.
Ainsworth, EA. (2008). Rice production in a changing climate: a meta-analysis of responses to elevated carbon
        dioxide and elevated ozone concentration. Global Change Biol 14:  1642-1650.
        http://dx.doi.0rg/10.1111/i.1365-2486.2008.01594.x.
Alexis, A: Garcia, A:  Nystrom, M: Rosenkranz, K. (2001 a). The 2001 California almanac of emissions and air
        quality. Sacremento, CA: California Air Resources Board.
        http://www.arb.ca.gov/agd/almanac/almanac01/almanac01.htm.
Allen, EB: Temple, PJ: Bvtnerowicz, A: Arbaugh,  MJ: Sirulnik, AG: Rao, LE.  (2007). Patterns of understory
        diversity in mixed coniferous forests of southern California impacted by air pollution.
        ScientificWorldJournal  7: 247-263. http://dx.doi.org/10.1100/tsw.2007.72.
Alonso,  R: Bermejo, V: Sanz, J: Vails, B: Elvira, S: Gimeno, BS. (2007). Stomatal conductance of semi-natural
        Mediterranean grasslands: Implications for the development of ozone critical levels. Environ  Pollut 146:
        692-698. http://dx.doi.0rg/10.1016/i.envpol.2006.06.009.
Draft - Do Not Cite or Quote
9-148
September 2011

-------
Amthor. JS. (1988). Growth and maintenance respiration in leaves of bean (Phaseolus vulgaris L) exposed to
        ozone in open-top chambers in the field. New Phytol 110: 319-325. http://dx.doi.org/10.1111/1.1469-
        8137.1988.tb00268.x.
Andersen. CP: Wilson. R: Plocher. M: Hogsett. WE. (1997). Carry-over effects of ozone on root growth and
        carbohydrate concentrations of ponderosa pine seedlings. Tree Physiol 17: 805-811.
Andersen. CP. (2003). Source-sink balance and  carbon allocation below ground in plants exposed to ozone.
        New Phytol 157: 213-228.
Andersen, CP: Ritter, W: Gregg, J: Matyssek, R: Grams, TEE. (2010). Below-ground carbon allocation in
        mature beech and spruce trees following long-term, experimentally enhanced O3 exposure in Southern
        Germany. Environ Pollut  158: 2604-2609. http://dx.doi.Org/10.1016/i.envpol.2010.05.008.
Andrews, KM: Gibbons, JW: Jochimsen, DM. (2008). Ecological effects of roads on amphibians and reptiles: A
        literature review. In JC Mitchell; REJ Brown; B Bartholomew (Eds.),  Urban Herpetology (Vol. 3, pp. 121-
        143). Salt Lake City: Society for the Study of Amphibians and Reptiles.
Aneja,  MK; Sharma, S; Fleischmann, F; Stich, S; Heller, W; Bahnweg, G; Munch,  JC; Schloter, M. (2007).
        Influence of ozone on litter guality and its subseguent effects on the initial structure of colonizing
        microbial communities. Microb Ecol 54:  151-160. http://dx.doi.org/10.1007/s00248-006-9183-0.
Arbaugh, M;  Bytnerowicz, A; Grulke, N; Fenn, M; Poth, M; Temple, P; Miller,  P. (2003). Photochemical smog
        effects in mixed conifer forests along a natural gradient of ozone and nitrogen deposition in the San
        Bernardino Mountains. Environ Int29: 401-406. http://dx.doi.org/10.1016/50160-4120(02)00176-9.
Arbaugh, MJ; Miller,  PR; Carroll, JJ; Takemoto, BL; Proctor, T (1998). Relationships of ozone exposure to  pine
        injury in the Sierra Nevada and San Bernardino Mountains of California, USA. Environ Pollut 101: 291-
        301.  http://dx.doi.org/10.1016/S0269-7491(98)00027-X.
Ariyaphanphitak, W; Chidthaisong, A; Sarobol, E; Bashkin, VN; Towprayoon, S. (2005). Effects of elevated
        ozone concentrations on  Thai Jasmine rice cultivars (Oryza sativa L.). Water Air Soil Pollut 167: 179-
        200.  http://dx.doi.org/10.1007/s11270-005-8650-4.
Ashmore,  M;  Emberson, L; Karlsson, PE; Pleijel, H. (2004a). Introduction for ozone deposition special issue.
        Atmos Environ  38: 2211-2212.
Ashmore.  M:  Emberson. L: Karlsson. PE: Pleiiel. H. (2004b). New directions: A new generation of ozone critical
        levels for the protection of vegetation in Europe (correspondence). Atmos Environ 38: 2213-2214.
Ashmore.  MR: Bell. JNB: Mimmack. A. (1988). Crop growth along a gradient of ambient air pollution. Environ
        Pollut 53: 99-121. http://dx.doi.org/10.1016/0269-7491 (88)90028-0.
Ashmore,  MR. (2002). Effects of oxidants at the whole plant and community  level. In JNB Bell; M Treshow
        (Eds.), Air pollution and plant life (pp. 89-118). London: Wiley.
Avnerv. S: Mauzerall. PL: Liu. J: Horowitz. LW. (2011a). Global crop yield reductions due to surface ozone
        exposure: 1. Year 2000 crop production losses and economic damage. Atmos Environ 45: 2284-2296.
        http://dx.doi. org/10.1016/i.atmosenv.2Q10.11.045.
Avnerv. S: Mauzerall. PL: Liu. J: Horowitz. LW. (2011b). Global crop yield reductions due to surface ozone
        exposure: 2. Year 2030 potential  crop production losses and economic damage under two scenarios of
        O3 pollution. Atmos Environ 45: 2297-2309. http://dx.doi.Org/10.1016/i.atmosenv.2011.01.002.
Awmack. CS: Harrington. R: Lindroth. RL. (2004). Aphid individual performance may not predict population
        responses to elevated CO2 or O3.  Global Change Biol Biol.10: 1414-1423.
Awmack. CS: Mondor. EB: Lindroth. RL.  (2007).  Forest understory clover populations in enriched CO2 and O-3
        atmospheres: Interspecific, intraspecific, and indirect effects. Environ Exp Bot 59: 340-346.
        http://dx.doi.0rg/10.1016/i.envexpbot.2006.04.003.
Bagard. M: Le Thiec. D: Delacote. E: Hasenfratz-Sauder. MP: Banvov. J: Gerard.  J: Dizengremel. P: Jolivet. Y
        (2008). Ozone-induced changes  in photosynthesis and photorespiration of hybrid poplar in relation to
        the developmental stage  of the leaves. Physiol Plant 134: 559-574.  http://dx.doi.org/10.1111/j. 1399-
        3054.2008.01160.x.
Baler. M: Kandlbinder. A: Golldack. D: Dietz. K. (2005). Oxidative stress and ozone: Perception; signalling  and
        response. Plant Cell Environ 28:  1012-1020. http://dx.doi.Org/10.1111/i.1365-3040.2005.01326.x.
Baldantoni. D: Fagnano. M: Alfani. A. (2011). Tropospheric ozone effects on chemical composition and
        decomposition  rate of Quercus ilex L. leaves. Sci Total Environ 409: 979-984.
        http://dx.doi. org/10.1016/i.scitotenv.2Q10.11.022.
Draft - Do Not Cite or Quote                        9-149                                  September 2011

-------
Ball. GR: Palmer-Brown. D: Fuhrer. J: Skarbv. L: Gimeno. BS: Mills. G. (2000). Identification of non-linear
        influences on the seasonal ozone dose-response of sensitive and resistant clover clones using artificial
        neural networks. Ecol Modell 129: 153-168.
Balls. GR: Palmer-Brown. D: Sanders. GE. (1996). Investigating microclimatic influences on ozone injury in
        clover (Trifolium subterraneum) using artificial neural networks. New Phytol 132: 271-280.
        http://dx.doi.0rg/10.1111/i.1469-8137.1996.tb01846.x.
Bandeff. JM: Pregitzer. KS: Lova. WM: Holmes. WE:  Zak. DR. (2006). Overstory community composition and
        elevated atmospheric CO2 and O3 modify understory biomass production and nitrogen acquisition.
        Plant Soil 282: 251-259. http://dx.doi.ora/10.1007/s11104-005-5930-0.
Barnes. J: Zheng. Y: Lyons. T. (2002). Plant resistance to ozone: The role of ascorbate. In K Omasa; H Saji; S
        Youssefian; N Kondo (Eds.), Air pollution and plant biotechnology - Prospects for phytomonitoring and
        phytoremediation (pp. 235-252).  Tokyo: Springer-Verlag.
Bassin. S: Volk. M: Fuhrer. J.  (2007a). Factors affecting the ozone sensitivity of temperate European
        grasslands: An overview. Environ Pollut 146: 678-691. http://dx.doi.Org/10.1016/i.envpol.2006.06.010.
Bassin. S: Volk. M: Suter.  M: Buchmann. N: Fuhrer. J. (2007b). Nitrogen deposition but not ozone affects
        productivity and community composition of subalpine grassland after 3 yr of treatment. New Phytol 175:
        523-534.  http://dx.doi.0rg/10.1111/i.1469-8137.2007.02140.x.
Bassin. S: Werner. RA: Sorgel. K: Volk.  M: Buchmann. N: Fuhrer. J. (2009). Effects of combined ozone and
        nitrogen deposition on the in situ  properties of eleven key plant species of a subalpine pasture.
        Oecologia 158: 747-756. http://dx.doi.org/10.1007/s00442-008-1191-y.
Bauer.  MR: Hultman. NE:  Panek. JA: Goldstein. AH.  (2000). Ozone deposition to a ponderosa pine plantation in
        the Sierra Nevada Mountains (CA): A comparison of two different climatic years. J Geophys Res 105:
        22,123-122,136. http://dx.doi.org/10.1029/2000JD900168.
Bender. J: Muntifering. RB: Lin. JC: Weigel. HJ. (2006). Growth and nutritive quality of Poa pratensis as
        influenced by ozone and competition. Environ Pollut 142:  109-115.
        http://dx.doi.0rg/10.1016/i.envpol.2005.09.012.
Bender. J: Weigel. H. -J. (2011). Changes in atmospheric chemistry and crop health: A review [Review]. Agron
        Sustain Dev 31: 81-89. http://dx.doi.org/10.1051/agro/2010013.
Benoit, LF: Skellv, JM:  Moore, LD: Dochinger, LS. (1982). Radial growth reductions of Pinus strobus L
        correlated with foliar ozone sensitivity as an  indicator of ozone-induced losses in eastern forests. Can J
        For Res 12: 673-678. http://dx.doi.org/10.1139/x82-101.
Bergweiler, C: Manning, WJ: Chevone,  Bl. (2008). Seasonal and diurnal gas exchange differences in ozone-
        sensitive common milkweed (Asclepias syriaca L.) in relation to ozone uptake. Environ Pollut 152: 403-
        415. http://dx.doi.0rg/10.1016/i.envpol.2007.06.019.
Bergweiler, CJ: Manning, WJ. (1999). Inhibition of flowering and reproductive success in spreading dogbane
        (Apocynum androsaemifolium) by exposure  to ambient ozone. Environ Pollut 105: 333-339.
        http://dx.doi. org/10.1016/S0269-7491(99)00044-5.
Bernacchi, CJ: Morgan, PB: Ort, PR: Long, SP (2005). The growth of soybean under free air CO2 enrichment
        (FACE) stimulates photosynthesis while decreasing in vivo Rubisco capacity. Planta 220: 434-446.
        http://dx.doi. org/10.1007/s00425-004-1320-8.
Bernacchi, CJ: Leaky, ADB: Heady, LE:  Morgan,  PB:  Dohleman, FG: McGrath, JM: Gillespie, KM: Wittig, VE:
        Rogers. A: Long.  SP: Ort. DR. (2006). Hourly and seasonal variation in photosynthesis and stomatal
        conductance of soybean grown at future CO2 and ozone concentrations for 3 years under fully open-air
        field conditions. Plant Cell Environ 29: 2077-2090. http://dx.doi.Org/10.1111/i.1365-3040.2006.01581.x.
Betzelberger. AM: Gillespie. KM: McGrath. JM: Koester. RP: Nelson. RL: Ainsworth. EA. (2010). Effects of
        chronic elevated ozone concentration on antioxidant capacity,  photosynthesis and seed yield of 10
        soybean cultivars. Plant Cell Environ 33:  1569-1581. http://dx.doi.Org/10.1111/i.1365-
        3040.2010.02165.x.
Bidart-Bouzat.  MG: Imeh-Nathaniel. A. (2008). Global change effects on plant chemical defenses  against insect
        herbivores. J Integr Plant Biol 50: 1339-1354. http://dx.doi.Org/10.1111/i.1744-7909.2008.00751.x.
Billings. WD. (1978). Plants and the ecosystem. In. Belmont, CA: Wadsworth Publishing Company, Inc.
Biswas. DK: Xu. H: Li. YG: Sun. JZ: Wang. XZ: Han.  XG:  Jiang. GM. (2008). Genotypic differences in leaf
        biochemical, physiological and  growth responses to ozone in 20 winter wheat cultivars released over
        the past 60 years. Global Change Biol 14: 46-59. http://dx.doi.Org/10.1111/i.1365-2486.2007.01477.x.
Draft - Do Not Cite or Quote                        9-150                                 September 2011

-------
Black. VJ: Black. CR: Roberts. JA: Stewart. CA. (2000).  Impact of ozone on the reproductive development of
        plants. New Phytol 147: 421-447.
Black. VJ: Stewart. CA: Roberts. JA: Black. CR. (2007).  Ozone affects gas exchange, growth and reproductive
        development in Brassica campestris (Wisconsin Fast Plants). New Phytol 176: 150-163.
        http://dx.doi.0rg/10.1111/i.1469-8137.2007.02163.x.
Black. VJ: Stewart. CA: Roberts. JA: Black. CR. (2010).  Direct effects of ozone on reproductive development in
        Plantago major L. populations differing in sensitivity. Environ Exp  Bot 69: 121-128.
        http://dx.doi.0rg/10.1016/i.envexpbot.2010.04.006.
Blande. JD: Holopainen. JK: Li. T. (2010). Air pollution impedes plant-to-plant communication by volatiles. Ecol
        Lett 13: 1172-1181.  http://dx.doi.Org/10.1111/i.1461-0248.2010.01510.x.
Bohler, S: Bagard, M: Oufir, M: Planchon, S: Hoffmann, L: Jolivet, Y: Hausman, JF: Dizengremel, P: Renaut, J.
        (2007). A DICE analysis of developing poplar leaves subjected  to ozone reveals major changes in
        carbon metabolism. Proteomics 7: 1584-1599. http://dx.doi.org/10.1002/pmic.200600822.
Bohler, S: Sergeant, K: Lefevre, I: Jolivet, Y: Hoffmann, L: Renaut, J: Dizengremel,  P: Hausman, JF. (2010).
        Differential impact of chronic ozone exposure on expanding and fully expanded poplar leaves. Tree
        Physiol 30: 1415-1432. http://dx.doi.org/10.1093/treephvs/tpg082.
Bonn, B:  Von Kuhlmann, R: Lawrence, MG. (2004). High contribution of biogenic hydroperoxides to secondary
        organic aerosol formation.  Geophys  Res Lett 31: L10108. http://dx.doi.orq/10.1029/2003GL019172.
Booker. F: Muntifering. R: McGrath. M: Burkev. K: Decoteau. D:  Fiscus.  E: Manning. W: Krupa. S: Chappelka.
        A: Grantz,  D. (2009). The ozone component of global change: Potential effects on agricultural and
        horticultural plant yield, product quality and interactions with invasive species. J Integr Plant Biol 51:
        337-351. http://dx.doi.0rg/10.1111/i.1744-7909.2008.00805.x.
Booker, FL: Reid, CD: Brunschon-Harti, S:  Fiscus, EL: Miller, JE. (1997). Photosynthesis and photorespiration in
        soybean [Glycine max (L) Merr] chronically exposed to elevated carbon dioxide and ozone. J  Exp Bot
        48: 1843-1852.
Booker, FL: Fiscus, EL: Miller, JE. (2004a). Combined effects of elevated atmospheric carbon dioxide  and
        ozone on soybean whole-plant water use. Environ  Manage 33:  S355-S362.
        http://dx.doi.org/10.1007/s00267-003-9144-z.
Booker, FL: Burkev, KO: Overmyer, K: Jones, AM. (2004b).  Differential responses of G-protein Arabidopsis
        thaliana  mutants to ozone. New Phytol 162: 633-641.
Booker. FL: Prior. SA: Torbert. HA: Fiscus.  EL: Purslev. WA: Hu. S. (2005). Decomposition of soybean grown
        under elevated concentrations of CO2 and O3. Global Change  Biol 11:  685-698.
        http://dx.doi.0rg/10.1111/i.1365-2486.2005.00939.x.
Booker. FL: Fiscus. EL. (2005). The role of ozone flux and antioxidants in the suppression of ozone injury by
        elevated CO2 in soybean. J Exp Bot 56: 2139-2151. http://dx.doi.Org/10.1093/jxb/eri214.
Booker. FL: Burkev. KO: Purslev. WA: Heagle. AS. (2007). Elevated carbon dioxide and ozone effects  on
        peanut: I. Gas-exchange, biomass, and leaf chemistry. Crop Sci 47: 1475-1487.
        http://dx.doi.org/10.2135/cropsci2006.08.0537.
Borowiak. K: Rucinska-Sobkowiak.  R: Rvmer. K: Gwozdz. EA: Zbierska. J. (2009). Biochemical markers of
        tropospheric ozone: Experimentation with test-plants. Polish Journal of Ecology 57: 3-14.
Bou Jaoude, M: Katerji, N: Mastrorilli, M: Rana, G. (2008). Analysis of the  ozone effect on soybean in the
        Mediterranean region II. The consequences on growth,  yield and  water use efficiency. Eur J Agron 28:
        519-525. http://dx.doi.0rg/10.1016/i.eia.2007.09.001.
Bou Jaoude, M: Katerji, N: Mastrorilli, M: Rana, G. (2008). Analysis of the  effect of ozone on soybean  in the
        Mediterranean region I: The consequences on crop-water status.  Eur J Agron 28: 508-518.
        http://dx.doi.0rg/10.1016/i.eia.2007.09.002.
Broadmeadow, MSJ: Jackson, SB.  (2000).  Growth responses of Quercus  petraea, Fraxinus excelsior  and Pinus
        sylvestris to elevated carbon dioxide, ozone and water supply. New Phytol 146: 437-451.
        http://dx.doi. orq/10.1046/1.1469-8137.2000.00665.x.
Brook, JR: DiGiovanni, F: Cakmak, S: Meyers, TP (1997). Estimation of dry deposition velocity using inferential
        models and site-specific meteorology—uncertainty due to siting  of meteorological towers. Atmos
        Environ 31: 3911-3919.
Draft - Do Not Cite or Quote                        9-151                                  September 2011

-------
Bulbovas. P: de Souza. SR: de Moraes. RM: Luizao. F: Artaxo. P. (2007). Soybean Tracaja' seedlings exposed
        to ozone under controlled conditions. Pesqui Agropecu Bras 42: 641-646.
        http://dx.doi. orq/10.1590/S0100-204X2007000500005.
Burkev.  KO: Eason. G: Fiscus. EL. (2003). Factors that affect leaf extracellular ascorbic acid content and redox
        status. Physiol Plant 117: 51-57.  http://dx.doi.Org/10.1034/1.1399-3054.2003.1170106.x.
Burkev.  KO: Neufeld.  HS: Souza. L:  Chappelka. AH:  Davison. AW. (2006). Seasonal profiles of leaf ascorbic
        acid content and redox state in ozone-sensitive wildflowers. Environ Pollut 143: 427-434.
        http://dx.doi.0rg/10.1016/i.envpol.2005.12.009.
Burkev.  KO: Booker. FL: Purslev. WA: Heagle. AS. (2007). Elevated carbon dioxide and ozone effects on
        peanut: II. Seed yield and quality. Crop Sci 47: 1488-1497.
        http://dx.doi.org/10.2135/cropsci2006.08.0538.
Bvtnerowicz. A: Arbaugh. M: Schilling. S:  Fraczek. W: Alexander. D. (2008). Ozone distribution and phytotoxic
        potential in mixed conifer forests  of the San  Bernardino Mountains, Southern California. Environ Pollut
        155: 398-408. http://dx.doi.Org/10.1016/i.envpol.2008.01.046.
Caird. MA: Richards. JH: Donovan. LA. (2007). Nighttime stomatal conductance and transpiration in C-3 and C-
        4 plants. Plant Physiol 143: 4-10. http://dx.doi.orq/10.1104/pp. 106.092940.
Cal/EPA. (California Environmental Protection Agency). (2010). Air quality data branch main page, from
        http://www.arb.ca.gov/agd/agdpage.htm
Calatavud. A: Alvarado. JW: Barreno. E. (2002). Similar effects of ozone on four cultivars of lettuce in open top
        chambers during winter. Photosynthetica 40: 195-200. http://dx.doi.Org/10.1023/A: 1021333305592.
Calatavud. A: Pomares. F: Barreno.  E. (2006).  Interactions between nitrogen fertilization and ozone in
        watermelon cultivar Reina de Corazones in open-top chambers. Effects on chlorophyll alpha
        fluorescence, lipid peroxidation, and yield. Photosynthetica 44:  93-101.
        http://dx.doi.org/10.1007/s11099-005-0163-2.
Calatavud. V: Cervero. J: Sanz. MJ. (2007a). Foliar, physiologial and growth responses of four maple species
        exposed to ozone. Water Air Soil  Pollut 185: 239-254.  http://dx.doi.org/10.1007/s11270-007-9446-5.
Calatavud. V: Sanz. MJ: Calvo. E: Cervero. J: Ansel. W: Klumpp. A. (2007b). Ozone biomonitoring with Bel-W3
        tobacco plants in the city of Valencia (Spain). Water Air Soil Pollut 183: 283-291.
        http://dx.doi.org/10.1007/s11270-007-9376-2.
Campbell. SJ: Wanek. R: Coulston. JW. (2007). Ozone injury in west coast forests: 6 years of monitoring -
        Introduction.  Portland, OR: U.S. Department of Agriculture.
Cannon, WN. (1990). Olfactory response  of eastern spruce budworm larvae to red spruce needles exposed to
        acid rain and elevated levels of ozone. J  Chem Ecol 16: 3255-3261.
        http://dx.doi. orq/10.1007/BF00982096.
Garde, RT: Haynes, KF. (2004). Stucture  of the pheromone communication channel in moths. In Advances in
        insect chemical ecology (pp. 283-332). Cambridge: Cambridge University Press.
Casteel. CL: O'Neill. BF: Zavala. JA: Bilgin. DP: Berenbaum. MR: DeLucia. EH. (2008). Transcriptional profiling
        reveals elevated CO2 and elevated O-3 alter resistance of soybean (Glycine max) to Japanese beetles
        (Popillia japonica). Plant Cell Environ 31: 419-434. http://dx.doi.orq/10.1111/i. 1365-3040.2008.01782.x.
Chapman. JA: King. JS:  Pregitzer. KS: Zak. DR. (2005). Effects of elevated concentrations of atmospheric CO2
        and tropospheric O-3 on decomposition of fine roots. Tree Physiol 25: 1501-1510.
Chappelka. A: Skellv.  J: Somers. G:  Renfro. J: Hildebrand. E. (1999a). Mature black cherry used as a
        bioindicator of ozone injury. Water Air Soil Pollut 116: 261-266.
Chappelka, A: Somers, G: Renfro, J. (1999b). Visible ozone injury on forest trees in Great Smoky Mountains
        National Park, USA. Water Air Soil Pollut 116: 255-260.
Chappelka. AH:  Samuelson. LJ. (1998). Ambient ozone effects on forest trees of the eastern  United States:  A
        review [Review]. New Phytol 139: 91-108. http://dx.doi.Org/10.1046/j. 1469-8137.1998.00166.x.
Chappelka. AH.  (2002). Reproductive development of blackberry (Rubus cuneifolius) as influenced by ozone.
        New Phytol 155: 249-255. http://dx.doi.orq/10.1046/1.1469-8137.2002.00464.x.
Chappelka, AH:  Somers, GL: Renfro, JR.  (2007).  Temporal patterns of foliar ozone symptoms on tall milkweed
        (Asclepias exaltata L) in Great Smoky Mountains  National Park. Environ Pollut 149: 358-365.
        http://dx.doi.0rg/10.1016/i.envpol.2007.05.015.
Chen, CW: Tsai, WT:  Lucier, AA. (1998). A model  of air-tree-soil system  for ozone impact analysis. Ecol Modell
        111:207-222.
Draft - Do Not Cite or Quote                        9-152                                 September 2011

-------
Chen. Z: Gallie. DR. (2005). Increasing tolerance to ozone by elevating foliar ascorbic acid confers greater
        protection against ozone than increasing avoidance. Plant Physiol 138: 1673-1689.
        http://dx.doi. orq/10.1104/pp. 105.062000.
Chen. Z: Wang. XK: Feng. ZZ: Xiao. Q: Duan. XN.  (2009). Impact of elevated O-3 on soil microbial community
        function under wheat crop. Water Air Soil Pollut 198: 189-198. http://dx.doi.org/10.1007/s11270-008-
        9838-1.
Chen. Z: Wang. XK: Yao. FF: Zheng. FX: Feng. ZZ. (201 Ob). Elevated ozone changed soil microbial community
        in a rice paddy. Soil Sci Soc Am J 74: 829-837. http://dx.doi.org/10.2136/sssai2009.0258.
Cheng. FY: Burkev. KO: Robinson. JM: Booker. FL. (2007). Leaf extracellular ascorbate in relation to O-3
        tolerance of two soybean cultivars. Environ Pollut 150: 355-362.
        http://dx.doi.0rg/10.1016/i.envpol.2007.01.022.
Cho. K: Shibato. J: Agrawal. GK: Jung. YH: Kubo. A: Jwa. NS: Tamogami. S: Satoh. K: Kikuchi. S: Higashi. T:
        Kimura. S:  Sail. H: Tanaka. Y: Iwahashi. H: Masuo. Y: Rakwal. R. (2008). Integrated transcriptomics,
        proteomics, and metabolomics analyses to survey ozone responses in the leaves of rice seedling. J
        Proteome Res 7:  2980-2998.  http://dx.doi.org/10.1021/pr800128g.
Christman.  MA: Donovan. LA: Richards. JH. (2009). Magnitude of nighttime transpiration does not affect plant
        growth or nutrition in well-watered Arabidopsis. Physiol Plant 136: 264-273.
        http://dx.doi.0rg/10.1111/i.1399-3054.2009.01216.x.
Chung. HG: Zak. PR: Lilleskov. EA. (2006). Fungal community composition and metabolism under elevated
        CO2 and O-3. Oecologia 147: 143-154. http://dx.doi.org/10.1007/s00442-005-0249-3.
Colls. JJ: Unsworth. MH. (1992). Air pollution interactions with natural stressors. In JR Barker;  DT Tingey (Eds.),
        Air pollution effects on biodiversity. New York, NY: Van Nostrand Reinhold.
Cornelissen, T. (2011). Climate change and its effects on terrestrial insects and herbivory patterns. Neotrop
        Entomol 40: 155-163. http://dx.doi.org/10.1590/51519-566X2011000200001.
Costa. PL: Folinsbee. LJ:  Raub. JA: Tilton. B: Tingev. DT. (1992). Summary of selected new information on
        effects of ozone on health and vegetation: Supplement to 1986 air quality criteria for ozone and other
        photochemical oxidants. (EPA/600/8-88/105F). Research Triangle Park,  NC: U.S. Environmental
        Protection Agency.
Coulston, JW: Smith, GC: Smith, WD.  (2003).  Regional assessment of ozone sensitive tree species using
        bioindicator plants. Environ Monit Assess 83: 113-127.
Crous. KY:  Vandermeiren. K: Ceulemans. R. (2006). Physiological responses to cumulative ozone uptake in two
        white clover (Trifolium repens L. cv. Regal) clones with different ozone sensitivity. Environ Exp Bot 58:
        169-179. http://dx.doi.0rg/10.1016/i.envexpbot.2005.07.007.
D'Haese. D: Vandermeiren. K: Asard. H: Horemans. N. (2005). Other factors than apoplastic ascorbate
        contribute to the differential ozone tolerance of two clones of Trifolium repens L.  Plant Cell Environ 28:
        623-632. http://dx.doi.0rg/10.1111/i.1365-3040.2005.01308.x.
Dalstein. L: Vas. N.  (2005). Ozone concentrations and ozone-induced symptoms on coastal and alpine
        mediterranean pines in southern France. Water Air Soil Pollut 160:  181-195.
Dammgen.  U: Grunhage. L: Haenel. H. -D: Jager. H. -J. (1993). Climate and stress in ecotoxicology: A coherent
        system of definitions and terms. J Appl Bot Food Qual 67: 157-162.
Darbah, JNT: Kubiske, ME: Neilson, N: Oksanen, E: Vaapavuori,  E: Karnoskv, DF. (2007). Impacts of elevated
        atmospheric CO2 and O3 on paper birch (Betula papyrifera): Reproductive fitness.
        ScientificWorldJournal 7: 240-246. http://dx.doi.org/10.110O/tsw.2007.42.
Darbah, JNT: Kubiske, ME: Nelson, N: Oksanen, E: Vapaavuori, E:  Kamoskv, DF. (2008). Effects of decadal
        exposure to interacting elevated CO2 and/or O-3 on paper birch (Betula papyrifera) reproduction.
        Environ Pollut 155: 446-452. http://dx.doi.Org/10.1016/i.envpol.2008.01.033.
Davidson, A. (1993). Update of ozone trends in California's South Coast Air Basin. J Air Waste Manag Assoc 43:
        226-227.
Davis. DP: Orendovici. T. (2006). Incidence of ozone symptoms on vegetation within a National Wildlife Refuge
        in New Jersey, USA. Environ Pollut 143: 555-564. http://dx.doi.org/10.1016/|.envpol.2005.10.051.
Davis. DP. (2007a).  Ozone-induced symptoms on vegetation within the Moosehorn National Wildlife Refuge in
        Maine. Northeast Nat 14: 403-414. http://dx.doi.org/10.1656/1092-
        6194(2007)14f403:OSOVWT12.0.CO:2.
Draft - Do Not Cite or Quote                        9-153                                  September 2011

-------
Davis. DP. (2007b). Ozone injury to plants within the Seney National Wildlife Refuge in northern Michigan.
        Northeast Nat 14: 415-424.
Davis. DP. (2009). Ozone-induced stipple on plants in the Cape Remain National Wildlife Refuge, South
        Carolina. Southeastern Naturalist 8: 471-478.
Dawson, TE: Burgess, SS: Tu, KP: Oliveira,  RS: Santiago, LS: Fisher, JB: Simonin,  KA: Ambrose, AR. (2007).
        Nighttime transpiration in woody plants from contrasting ecosystems. Tree Physiol 27: 561-575.
        http://dx.doi. orq/10.1093/treephvs/27.4.561.
De Temmerman, l_:  Legrand, G: Vandermeiren, K. (2007). Effects of ozone on sugar beet grown in open-top
        chambers.  Eur J Agron 26: 1-9. http://dx.doi.Org/10.1016/i.eia.2006.08.001.
de Lourdes de Bauer. M: Hernandez-Teieda. T. (2007). A review of ozone-induced effects on the forests of
        central Mexico [Review]. Environ Pollut 147: 446-453. http://dx.doi.Org/10.1016/i.envpol.2006.12.020.
Pegl'lnnocenti. E: Guidi. L: Soldatini. GF. (2007).  Effects of elevated ozone on chlorophyll a fluorescence in
        symptomatic and asymptomatic leaves of two tomato genotypes. Biol Plantarum 51: 313-321.
        http://dx.doi. org/10.1007/s10535-007-0061 -5.
Dermodv. O: O'Neill. BF: Zangerl. AR: Berenbaum. MR: DeLucia. EH. (2008). Effects of elevated CO2 and O3
        on leaf damage and insect abundance in a soybean agroecosystem. Arthropod-Plant Inte 2: 125-135.
Pi Baccio, D: Castagna, A: Paoletti, E: Sebastiani, L: Ranieri, A. (2008). Could the differences in O3 sensitivity
        between two poplar clones be related to a difference in antioxidant defense and secondary metabolic
        response to O3 influx? Tree Physiol 28: 1761-1772.
Dickson, RE: Lewin, KF: Isebrands, JG: Coleman, MD: Heilman, WE: Riemenschneider, DE: Sober, J: Host,
        GE:  Zak. PR:  Hendrev. GR: Pregitzer. KS: Karnoskv. PR (2000). Forest Atmosphere Carbon Transfer
        and  Storage (FACTS-II) the Aspen Free-Air CO2  and O3 Enrichment (FACE) project: An overview.
        (General Technical Report NC-214). St. Paul, MN: U.S. Pept. of Agriculture, Forest Service.
        http://nrs.fs.fed.us/pubs/278.
Pitchkoff. SS: Lewis. JS: Lin. JC: Muntifering. RB: Chappelka. AH. (2009). Nutritive quality of highbush
        blackberry  (Rubus argutus) exposed to tropospheric ozone. Rangeland  Ecol Manag 62: 364-370.
Pizengremel. P: Sasek. T: Brown.  K: Richardson. C. (1994). Ozone-induced changes in primary carbon
        metabolism enzymes of loblolly pine needles. J Plant Physiol 144: 300-306.
Pizengremel, P: Le Thiec, P: Bagard, M: Jolivet, Y. (2008). Ozone risk assessment for plants: Central role of
        metabolism-dependent changes in reducing power. Environ Pollut 156: 11-15.
        http://dx.doi.0rg/10.1016/i.envpol.2007.12.024.
Pizengremel, P: Le Thiec, P: Hasenfratz-Sauder, MP: Vaultier, MN: Bagard, M: Jolivet, Y. (2009). Metabolic-
        dependent changes in plant cell redox power after ozone exposure. Plant Biol (Stuttg) 11: 35-42.
        http://dx.doi.0rg/10.1111/i.1438-8677.2009.00261.x.
Pobson, HEM. (1994). Floral volatiles in insect biology. In  EA Bernays (Ed.), Insect-plant interactions: Vol 5 (pp.
        47-82). Boca Raton, FL: CRC Press.
Pohm. MR: Mautz. WJ: Looby. PG: Gellert. KS: Andrade. JA. (2001). Effects of ozone on evaporative water loss
        and  thermoregulatory  behavior of marine toads (Bufo marinus).  Environ Res 86: 274-286.
Pohm. MR: Mautz. WJ: Andrade. JA: Gellert. KS: Salas-Ferguson. LJ: Nicolaisen. N: Fuiie.  N. (2005). Effects of
        ozone exposure on nonspecific phagocytic capacity of pulmonary macrophages from an amphibian,
        Bufo marinus. Environ Toxicol Chem 24: 205-210.
Pohm. MR: Mautz. WJ: Poratt. RE: Stevens. JR.  (2008). Ozone exposure affects feeding and locomotor
        behavior of adult Bufo marinus. Environ Toxicol Chem 27: 1209-1216. http://dx.doi.org/10.1897/07-
        388.1.
Pohrmann. AB: Tebbe. CC. (2005). Effect of elevated tropospheric ozone on the structure of bacterial
        communities inhabiting the rhizosphere of herbaceous plants native to Germany. Appl Environ
        Microbiol 71: 7750-7758. http://dx.doi.org/10.1128/AEM.71.12.7750-7758.2005.
Progoudi. PP: Ashmore. MR. (2000). Poes elevated ozone have differing effects in flowering and deblossomed
        strawberry? New Phytol 147: 561-569. http://dx.doi.org/10.1046/i. 1469-8137.2000.00718.x.
Progoudi, PP: Ashmore, M. (2001). 14C-allocation of flowering and deblossomed strawberry in  response to
        elevated ozone. New Phytol 152: 455-461. http://dx.doi.Org/10.1046/i.0028-646X.2001.00270.x.
Pudareva. N: Negre. F: Nagegowda. PA: Orlova. I. (2006). Plant volatiles: Recent advances and future
        perspectives [Review]. Grit Rev Plant Sci 25: 417-440.
Praft - Po Not Cite or Quote                       9-154                                 September 2011

-------
Ederli. L: Morettini. R: Borgogni. A: Wasternack. C: Miersch. O: Reale. L: Ferranti. F: Tosti. N: Pasgualini. S.
        (2006). Interaction between nitric oxide and ethylene in the induction of alternative oxidase in ozone-
        treated tobacco plants. Plant Physiol 142: 595-608. http://dx.doi.orq/10.1104/pp. 106.085472.
Edwards. IP: Zak. DR. (2011). Fungal community composition and function after long-term exposure of northern
        forests to elevated atmospheric CO2 and tropospheric O3. Global Change Biol 17: 2184-2195.
        http://dx.doi. org/10.1111/i.1365-2486.2010.02376.x.
Ellenson. JL: Amundson. RG. (1982). Delayed light imaging for the early detection of plant stress. Science 215:
        1104-1106. http://dx.doi.0rg/10.1126/science.215.4536.1104.
Ellsworth. PS:  Reich. PB:  Naumburg. ES: Koch. GW: Kubiske. ME: Smith. SD.  (2004). Photosynthesis,
        carboxylation and leaf nitrogen responses of 16 species to elevated pCO2 across four free-air CO2
        enrichment experiments in forest, grassland and desert. Global Change Biol 10: 2121-2138.
        http://dx.doi. orq/10.1111/i.1365-2486.2004.00867.x.
Eltaveb. AE: Kawano. N: Badawi.  GH: Kaminaka. H: Sanekata. T:  Morishima. I: Shibahara. T:  Inanaga. S:
        Tanaka, K. (2006). Enhanced tolerance to ozone and drought stresses in transgenic tobacco
        overexpressing dehydroascorbate reductase in cytosol. Physiol Plant 127: 57-65.
        http://dx.doi.ora/10.1111/i.1399-3054.2005.00624.x.
Eltayeb, AE: Kawano, N: Badawi,  GH: Kaminaka, H: Sanekata, T:  Shibahara, T: Inanaga, S: Tanaka, K.  (2007).
        Overexpression of monodehydroascorbate reductase in transgenic tobacco confers enhanced
        tolerance to ozone, salt and polyethylene glycol stresses. Planta 225: 1255-1264.
        http://dx.doi. org/10.1007/s00425-006-0417-7.
Emberson.  L: Ashmore. MR: Cambridge. HM: Simpson. D: Tuovinen. J. -P. (2000a). Modelling stomatal  ozone
        flux across Europe. Environ Pollut 109: 403-413. http://dx.doi.orq/10.1016/S0269-7491 (00)00043-9.
Emberson,  LD: Wieser, G: Ashmore, MR. (2000b). Modelling of stomatal conductance and ozone flux of Norway
        spruce: Comparison with field data. Environ Pollut  109: 393-402.  http://dx.doi.orq/10.1016/S0269-
        7491(00)00042-7.
Enders,  G.  (1992). Deposition of ozone to a mature spruce  forest:  Measurements and comparison to models.
        Environ Pollut 75: 61-67.  http://dx.doi.orq/10.1016/0269-7491 (92)90057-H.
Esperschutz. J: Pritsch. K: Gattinger. A: Welzl. G: Haesler. F:  Buegger. F:  Winkler. JB: Munch. JC: Schloter. M.
        (2009). Influence  of chronic ozone stress on carbon translocation pattern into rhizosphere microbial
        communities of beech trees (Fagus sylvatica L.) during a growing season.  Plant Soil 323: 85-95.
        http://dx.doi.ora/10.1007/s11104-009-0090-2.
Fares, S: Barta, C: Brilli, F: Centritto, M:  Ederli,  L: Ferranti,  F:  Pasqualini, S: Reale, L: Tricoli, D: Loreto,  F.
        (2006). Impact of high ozone on isoprene emission, photosynthesis and histology  of developing
        Populus alba leaves directly or indirectly exposed to the pollutant. Physiol Plant 128: 456-465.
        http://dx.doi.0rg/10.1111/i.1399-3054.2006.00750.x.
Fares. S: Loreto. F: Kleist. E: Wildt. J.  (2008). Stomatal uptake and stomatal deposition of  ozone in isoprene
        and monoterpene emitting plants. Plant Biol (Stuttg) 10: 44-54. http://dx.doi.orq/10.1055/S-2007-
        965257.
Fares. S: McKay. M: Holzinger. R: Goldstein. AH. (201 Oa). Ozone fluxes in a Pinus ponderosa ecosystem are
        dominated by non-stomatal processes:  Evidence from long-term continuous measurements. Agr Forest
        Meteorol 150: 420-431. http://dx.doi.Org/10.1016/i.agrformet.2010.01.007.
Fares. S: Oksanen. E: Lannenpaa. M: Julkunen-Tiitto. R:  Loreto. F. (201 Ob). Volatile emissions and phenolic
        compound concentrations along a vertical profile of Populus  nigra leaves exposed to realistic ozone
        concentrations. Photosynth Res 104: 61-74. http://dx.doi.org/10.1007/s11120-010-9549-5.
Fares. S: Gentner.  PR:  Park. JH: Ormeno. E: Karlik. J:  Goldstein. AH. (2011). Biogenic emissions from Citrus
        species in California. Atmos Environ 45: 4557-4568. http://dx.doi.Org/10.1016/i.atmosenv.2011.05.066.
Felicity,  H: Gina, M: Laurence, J: Mike, A. (2010). Does a simulated upland grassland community respond to
        increasing background, peak or accumulated exposure of ozone? Atmos Environ 44: 4155-4164.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.07.037.
Felzer, B: Kicklighter, D: Melillo, J: Wang, C: Xhuang, Q: Prinn, R.  (2004).  Effects of ozone on net primary
        production and carbon sequestration in the conterminous United States using a biogeochemistry
        model. Tellus B Chem Phys Meteorol 56: 230-248.  http://dx.doi.Org/10.1111/i.1600-0889.2004.00097.x.
Draft - Do Not Cite or Quote                        9-155                                  September 2011

-------
Felzer. B: Reillv. J: Melillo. J: Kicklighter. D: Sarofim. M: Wang. C: Prinn. R: Zhuang. Q. (2005). Future effects of
        ozone on carbon sequestration and climate change policy using a global biogeochemical model. Clim
        Change 73: 345-373. http://dx.doi.orq/10.1007/s10584-005-6776-4.
Felzer. BS: Cronin. TW: Melillo. JM: Kicklighter. DW: Schlosser. CA. (2009). Importance of carbon-nitrogen
        interactions and ozone on ecosystem hydrology during the 21st century. J  Geophys Res 114: G01020.
        http://dx.doi. org/G0102010.1029/2008ia000826.
Feng. YW: Komatsu. S: Furukawa. T: Koshiba. T: Kohno. Y. (2008a). Proteome analysis of proteins responsive
        to ambient and elevated ozone in rice seedlings. Agric Ecosyst Environ 125: 255-265.
        http://dx.doi.0rg/10.1016/i.aqee.2008.01.018.
Feng. Z: Pang. J: Nouchi. I: Kobavashi. K: Yamakawa. T: Zhu. J. (2010). Apoplastic ascorbate contributes to the
        differential ozone sensitivity in two varieties of winter wheat under fully open-air field conditions.
        Environ Pollut 158: 3539-3545. http://dx.doi.Org/10.1016/i.envpol.2010.08.019.
Feng. ZZ: Kobavashi. K: Ainsworth. EA. (2008b). Impact of elevated ozone concentration on growth,  physiology,
        and yield of wheat (Triticum aestivum L): A meta-analysis. Global Change Biol  14: 2696-2708.
        http://dx.doi.0rg/10.1111/i.1365-2486.2008.01673.x.
Feng. ZZ: Kobavashi. K. (2009). Assessing the impacts of current and future concentrations of surface ozone on
        crop yield with meta-analysis. Atmos Environ 43: 1510-1519.
        http://dx.doi.0rg/10.1016/i.atmosenv.2008.11.033.
Fenn. ME: Poth. MA: Johnson. DW. (1996). Evidence for nitrogen saturation in the  San Bernardino Mountains in
        southern California. For Ecol Manage 82: 211-230. http://dx.doi.org/10.1016/0378-1127(95)03668-7.
Fenn. ME: de Bauer. LI: Hernandez-Teieda. T. (2002). Summary of air pollution impacts on forests in  the Mexico
        City air basin. In Urban air pollution and forests (pp. 337-355). New York, NY: Springer-Verlag.
Findley, DA:  Keever, GJ: Chappelka, AH: Eakes, DJ: Gillian, DJ. (1997). Differential responses of buddleia
        (Buddleia davidii Franch) to ozone. Environ Pollut 98: 105-111.
Finkelstein. PL: Ellestad. TG: Clarke. JF: Meyers. TP: Schwede. DB: Hebert. EO: Neal. JA. (2000). Ozone and
        sulfur dioxide dry deposition to forests: Observations and model evaluation. J Geophys Res  105:
        15365-15377.
Finnan. JM: Jones.  MB: Burke. Jl. (1996). A time-concentration study on the effects of ozone on spring wheat
        (Triticum aestivum L):  2. A comparison of indices. Agric Ecosyst Environ 57: 169-177.
        http://dx.doi. orq/10.1016/0167-8809(95)01004-1.
Finnan. JM: Burke. JL: Jones. MB. (1997). An evaluation of indices that describe the impact of ozone on the
        yield of spring wheat (Triticum aestivum L). Atmos Environ 31: 2685-2693.
        http://dx.doi. orq/10.1016/51352-2310(97)00105-2.
Fiscus. EL: Philbeck. R: Britt. AM: Booker. FL. (1999). Growth of Arabidopsis flavonoid mutants under solar
        radiation and UV filters. Environ Exp Bot41: 231-245. http://dx.doi.org/10.1016/50098-8472(99)00011-
        8.
Fiscus. EL: Booker. FL: Burkev. KO. (2005). Crop responses to ozone: Uptake, modes of action,  carbon
        assimilation and partitioning. Plant Cell Environ 28: 997-1011.
Fishman. J: Bowman. KW: Burrows. JP: Richter. A: Chance. KV: Edwards. DP: Martin. RV: Morris. GA: Pierce.
        RB: Ziemke. JR: Al-Saadi. JA: Creilson. JK: Schaack. TK: Thompson. AM.  (2008). Remote sensing of
        tropospheric pollution from space. Bull Am Meteorol Soc 89: 805-821.
        http://dx.doi. orq/10.1175/2008BAMS2526.1.
Fishman. J: Creilson. JK: Parker. PA: Ainsworth. EA: Vining. GG: Szarka. J: Booker. FL: Xu. XJ. (2010). An
        investigation of widespread ozone damage to the soybean crop in the upper Midwest determined from
        ground-based and satellite measurements. Atmos Environ 44: 2248-2256.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.01.015.
FLAG. (Federal Land Manager's Air Quality Related Values Workgroup). (2000). Phase  I report. Lakewood,  CO:
        U.S. Forest Service.
Flagler. RB. (1998).  Recognition of air pollution injury to vegetation: A pictorial atlas. In (2nd ed.). Pittsburgh, PA:
        Air & Waste Management Association.
Flowers. MD: Fiscus. EL: Burkev.  KO: Booker. FL: Dubois. JJB. (2007). Photosynthesis, chlorophyll
        fluorescence, and yield of snap bean (Phaseolus vulgaris L.) genotypes differing in sensitivity to ozone.
        Environ Exp Bot 61: 190-198. http://dx.doi.Org/10.1016/i.envexpbot.2007.05.009.
Draft - Do Not Cite or Quote                        9-156                                  September 2011

-------
Fontan. JA: Minga. A: Lopez. A: Druilhet. A. (1992). Vertical ozone profiles in a pine forest. Atmos Environ 26:
        863-869.  http://dx.doi.org/10.1016/0960-1686(92)90245-G.
Foyer. CH: Noctor. G. (2005a). Oxidant and antioxidant signalling in plants: A re-evaluation of the concept of
        oxidative  stress in a physiological context. Plant Cell Environ 28: 1056-1071.
Foyer, CH: Noctor, G. (2005b). Redox homeostasis and antioxidant signaling: A metabolic interface between
        stress perception and physiological responses. Plant Cell 17: 1866-1875.
        http://dx.doi. orq/10.1105/tpc. 105.033589.
Fredericksen, TS:  Joyce, BJ: Skellv, JM: Steiner, KG: Kolb, TE: Kouterick, KB: Savage, JE: Snvder, KR. (1995).
        Physiology, morphology, and ozone uptake of leaves of black cherry seedlings, saplings, and canopy
        trees. Environ Pollut 89: 273-283. http://dx.doi.orq/10.1016/0269-7491 (94)00077-Q.
Fredericksen, TS:  Kolb, TE:  Skellv, JM: Steiner, KG: Joyce,  BJ: Savage, JE. (1996).  Light environment alters
        ozone uptake per net photosynthetic rate in black cherry trees. Tree Physiol 16:  485-490.
Freiwald. V: Haikio. E: Julkunen-Tiitto. R:  Holopainen. JK: Oksanen. E. (2008). Elevated  ozone modifies the
        feeding behaviour of the common leaf weevil on hybrid aspen through shifts in developmental,
        chemical, and structural properties of leaves. Entomol Exp Appl 128: 66-72.
        http://dx.doi.ora/10.1111/i.1570-7458.2008.00677.x.
Fuentes, JD: Gillespie, TJ: den Hartog, G: Neumann, HH. (1992). Ozone deposition  onto a deciduous forest
        during dry and wet conditions. Agr Forest Meteorol 62: 1-18.  http://dx.doi.orq/10.1016/0168-
        1923(92)90002-1.
Fuhrer, J. (1994).  Effects of  ozone on managed pasture: 1. Effects of open-top chambers on microclimate,
        ozone flux, and plant growth. Environ Pollut 86: 297-305.
Fuhrer. J: Skarbv.  L: Ashmore. MR. (1997). Critical levels for ozone effects on vegetation in Europe. Environ
        Pollut 97: 91-106. http://dx.doi.org/10.1016/30269-7491(97)00067-5.
Gate. IM: McNeill. S: Ashmore. MR. (1995). Effects of air pollution on the searching behaviour of an insect
        parasitoid. Water Air Soil Pollut 85: 1425-1430.  http://dx.doi.orq/10.1007/BF00477181.
Geiser, LH: Neitlich, PN. (2007).  Pollution and climate gradients in western Oregon and Washington indicated
        by epiphytic macrolichens. Environ Pollut 145:  203-218. http://dx.doi.Org/10.1016/i.envpol.2006.03.024.
Gerosa. G: Marzuoli. R: Rossini.  M: Panigada. C: Meroni. M: Colombo. R: Faoro. F:  Iriti.  M. (2009). A flux-based
        assessment of the effects of ozone on foliar injury, photosynthesis, and yield of bean (Phaseolus
        vulgaris L. cv. Borlotto Nano Lingua di Fuoco) in open-top chambers. Environ Pollut 157: 1727-1736.
        http://dx.doi.0rg/10.1016/i.envpol.2008.06.028.
Gielen, B: Vandermeiren,  K: Horemans, N: D'Haese,  D:  Serneels,  R: Valcke, R. (2006). Chlorophyll a
        fluorescence imaging of ozone-stressed Brassica napus L. plants differing in glucosinolate
        concentrations. Plant Biol (Stuttg) 8: 698-705. http://dx.doi.org/10.1055/S-2006-924150.
Gitay, H:  Brown, S: Easterling, W: Jallow, B. (2001). Ecosystems and their goods and services. In Climate
        change 2001: Impacts, adaptation and vulnerability: Contribution of Working Group II to the third
        assessment report of the Intergovernmental  Panel on Climate Change (pp. 237-342). Cambridge,
        United Kingdom:  Cambridge University Press.
Gombert. S: Asta.  J:  Seaward. MRP. (2006). Lichens and tobacco plants as complementary biomonitors of air
        pollution in the Grenoble area (Isere, southeast France). Ecol Indicat 6: 429-443.
        http://dx.doi.0rg/10.1016/i.ecolind.2005.06.001.
Gonzalez-Fernandez. I: Kaminska. A: Dodmani. M: Goumenaki. E: Quarrie. S:  Barnes. JD. (2010). Establishing
        ozone flux-response relationships for winter wheat: Analysis of uncertainties based on data for UK and
        Polish genotypes. Atmos Environ 44: 621-630. http://dx.doi.Org/10.1016/i.atmosenv.2009.11.021.
Goumenaki. E: Tavbi. T: Borland. A: Barnes. J. (2010).  Mechanisms underlying the impacts of ozone on
        photosynthetic performance. Environ Exp  Bot 69: 259-266.
        http://dx.doi.0rg/10.1016/i.envexpbot.2010.04.011.
Grantz. DA: Zhang. XJ: Massman. WJ: Den Hartog. G:  Neumann. HH: Pederson. JR.  (1995). Effects of
        stomatal conductance and surface wetness on ozone deposition in field-grown grape. Atmos  Environ
        29: 3189-3198. http://dx.doi.org/10.1016/1352-2310(95)00129-M.
Grantz. DA: Zhang. XJ: Massman. W: Delanv. A: Pederson. R. (1997). Ozone deposition to a cotton
        (Gossypium hirsutum L) field: Stomatal and surface wetness effects during the California Ozone
        Deposition experiment. Agr Forest Meteorol 85: 19-31. http://dx.doi.org/10.1016/S0168-
        1923(96)02396-9.
Draft - Do Not Cite or Quote                        9-157                                 September 2011

-------
Grantz. DA: Gunn. S: Vu. HB. (2006). O3 impacts on plant development: A meta-analysis of root/shoot
        allocation and growth. Plant Cell Environ 29: 1193-1209. http://dx.doi.org/10.1111/1.1365-
        3040.2006.01521.x.
Grantz. DA: Shrestha. A. (2006). Tropospheric ozone and interspecific competition between yellow nutsedge
        and pima cotton. Crop Sci 46: 1879-1889. http://dx.doi.org/10.2135/cropsci2005.06.0167.
Grantz. DA: Shrestha. A: Vu. HB. (2008). Early vigor and ozone response  in horseweed (Conyza canadensis)
        biotypes differing in glyphosate resistance. Weed Sci 56: 224-230. http://dx.doi.org/10.1614/ws-07-
        130.1.
Grantz. DA: Vu. HB. (2009). O3 sensitivity in a potential C4 bioenergy crop: Sugarcane in California. Crop Sci
        49: 643-650.
Grantz, DA: Vu, HB: Aguilar, C: Rea, MA. (201 Oa). No interaction between methyl jasmonate and ozone in Pima
        cotton: Growth and allocation respond independently to both. Plant Cell Environ 33: 717-728.
        http://dx.doi.0rg/10.1111/i.1365-3040.2009.02096.x.
Grantz, DA: Shrestha, A: Vu, HB. (201 Ob). Ozone impacts on assimilation  and allocation to reproductive sinks in
        the vegetatively propagated C-4 weed, yellow nutsedge. Crop Sci 50: 246-252.
        http://dx.doi.org/10.2135/cropsci2009.03.0127.
Grebenc, T: Kraigher, H. (2007). Changes in the community of ectomycorrhizal fungi and increased fine root
        number under adult beech trees chronically fumigated with double ambient ozone concentration. Plant
        Biol (Stuttg) 9: 279-287. http://dx.doi.org/10.1055/s-2006-924489.
Gregg, JW: Jones, CG: Dawson, TE.  (2003). Urbanization effects on tree growth in the vicinity of New York City
        [Letter/Response]. Nature 424: 183-187. http://dx.doi.org/10.1038/nature01728.
Gregg. JW: Jones. CG: Dawson. TE.  (2006). Physiological and developmental effects  of O3 on cottonwood
        growth in urban and rural sites. Ecol Appl 16:  2368-2381. http://dx.doi.org/10.1890/1051-
        0761(2006)016r2368:PADEOO12.0.CO:2.
Grennfelt. P. (2004). New directions: Recent research  findings  may change ozone control policies. Atmos
        Environ 38: 2215-2216.
Groppa. MD: Benavides. MP (2008).  Polyamines and abiotic stress: Recent advances. Amino Acids 34: 35-45.
        http://dx.doi.org/10.1007/s00726-007-0501-8.
Grulke, N: Neufeld, H: Davison, A: Roberts,  M: Chappelka, A. (2007a). Stomatal behavior of ozone-sensitive
        and -insensitive coneflowers (Rudbeckia laciniata van digitata) in  Great Smoky Mountains National
        Park. New Phytol 173: 100-109. http://dx.doi.Org/10.1111/i.1469-8137.2006.01872.x.
Grulke, NE: Lee, EH. (1997). Assessing visible ozone-induced injury in ponderosa pine. Can J For Res 27:
        1658-1668.
Grulke. NE. (1999). Physiological responses of ponderosa pine to gradients of environmental stressors. In PR
        Miller; JR McBride (Eds.), Oxidant air pollution impacts in the montane  forests of Southern California.
        New York, NY: Springer-Verlag.
Grulke. NE: Preisler. HK: Rose. C:  Kirsch. J: Balduman. L.  (2002). O3 uptake and drought stress effects on
        carbon acguisition of ponderosa pine in  natural stands. New Phytol 154: 621-631.
        http://dx.doi. org/10.1046/1.1469-8137.2002.00403.x.
Grulke. NE: Johnson. R: Esperanza. A: Jones. D: Nguyen. T: Posch. S: Tausz. M. (2003a). Canopy
        transpiration of Jeffrey pine in mesic and xeric microsites: O3 uptake and injury  response. Trees Struct
        Funct 17: 292-298.
Grulke. NE: Johnson. R: Monschein. S: Nikolova. P: Tausz. M. (2003b). Variation in morphological and
        biochemical O3 injury attributes of mature Jeffrey pine within canopies and between microsites. Tree
        Physiol 23: 923-929.
Grulke. NE: Alonso. R: Nguyen. T: Cascio. C: Dobrowolski.  W.  (2004). Stomata  open at night in pole-sized and
        mature ponderosa pine: Implications for O3 exposure  metrics. Tree Physiol 24: 1001-1010.
Grulke. NE: Dobrowolski. W: Mingus. P:  Fenn. ME.  (2005).  California black oak response to nitrogen
        amendment at a high O3, nitrogen-saturated site. Environ Pollut 137: 536-545.
        http://dx.doi.0rg/10.1016/i.envpol.2005.01.039.
Grulke. NE: Paoletti. E: Heath. RL. (2007b). Chronic vs. short-term acute O3 exposure effects on nocturnal
        transpiration in two Californian oaks. ScientificWorldJournal 7: 134-140.
        http://dx.doi.org/10.1100/tsw.2007.33.
Draft - Do Not Cite or Quote                        9-158                                 September 2011

-------
Grulke. NE: Paoletti. E: Heath. RL. (2007c). Comparison of calculated and measured foliar O3 flux in crop and
        forest species. Environ Pollut 146: 640-647. http://dx.doi.Org/10.1016/i.envpol.2006.04.014.
Grunhage. L:  Haenel. H. -D. (1997). PLATIN (PLant-ATmosphere-INteraction) I: A model of plant-atmosphere
        interaction for estimating absorbed doses of gaseous air pollutants. Environ Pollut 98: 37-50.
        http://dx.doi.org/10.1016/S0269-7491(97)00114-0.
Grunhage. L:  Jager. H. -J. (2003). From critical levels to critical loads for ozone: a discussion of a new
        experimental and modelling approach for establishing flux-response relationships for agricultural crops
        and native plant species. Environ Pollut 125: 99-110.
Grunhage. L:  Krupa. SV: Legge. AH: Jager. H. -J. (2004). Ambient flux-based critical values of ozone for
        protecting vegetation: Differing spatial scales and uncertainties in risk assessment.  Atmos Environ 38:
        2433-2437. http://dx.doi.Org/10.1016/i.atmosenv.2003.12.039.
Guderian. R. (1985). Air pollution by photochemical oxidants: Formation, transport, control, and effects  on
        plants. In. New York: Springer-Verlag.
Guenther, A: Karl, T: Harlev, P: Wiedinmyer, C: Palmer, PI: Geron, C. (2006). Estimates of global terrestrial
        isoprene  emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature). Atmos
        Chem Phys 6: 3181-3210. http://dx.doi.org/10.5194/acp-6-3181-2006.
Guidi, L: Degl'lnnocenti, E.  (2008). Ozone effects on high light-induced photoinhibition in Phaseolus vulgaris.
        Plant Sci  174: 590-596. http://dx.doi.Org/10.1016/i.plantsci.2008.03.003.
Guidi. L: Degl'lnnocenti. E:  Martinelli. F: Piras. M. (2009). Ozone effects on carbon metabolism in sensitive and
        insensitive Phaseolus cultivars. Environ Exp Bot 66: 117-125.
        http://dx.doi.0rg/10.1016/i.envexpbot.2008.12.005.
GDI. H: Gaga. EO: Dogeroglu. T: Ozden. O: Avvaz. O: Ozel.  S: Gungor. G. (2011). Respiratory health symptoms
        among students exposed to different levels of air pollution in a Turkish city. Int J Environ Res Public
        Health 8:  1110-1125. http://dx.doi.org/10.3390/iierph8041110.
Gumpertz. ML: Rawlings. JO. (1992). Nonlinear regression with variance components: Modeling effects of
        ozone on crop yield. Crop Sci 32: 219-224.
Gunderson. CA: Sholtis. JD: Wullschleger. SD: Tissue.  DT: Hanson.  PJ: Norbv. RJ. (2002). Environmental and
        stomatal control of photosynthetic enhancement in the canopy of a sweetgum (Liquidambar styraciflua
        L) plantation during 3 years of CO2 enrichment. Plant Cell Environ 25: 379-393.
        http://dx.doi.0rg/10.1046/i.0016-8025.2001.00816.x.
Haberer.  K: Herbinger. K: Alexou. M: Rennenberg. H: Tausz. M. (2008). Effects of drought and canopy ozone
        exposure on antioxidants in fine roots of mature European beech (Fagus sylvatica). Tree Physiol 28:
        713-719.  http://dx.doi.0rg/10.1093/treephvs/28.5.713.
Hamel. LP:  Miles.  GP:  Samuel.  MA: Ellis. BE: Seguin. A: Beaudoin. N. (2005). Activation of stress-responsive
        mitogen-activated  protein kinase pathways in hybrid poplar (Populus trichocarpa x Populus deltoides).
        Tree  Physiol 25: 277-288. http://dx.doi.org/10.1093/treephvs/25.3.277.
Hamilton. JG: Dermodv. O: Aldea. M: Zangerl. AR: Rogers. A: Berenbaum. MR: DeLucia. EH. (2005).
        Anthropogenic changes in tropospheric composition increase susceptibility of soybean to insect
        herbivory. Environ  Entomol 34: 479-485.
Handlev. T: Grulke. NE. (2008). Interactive effects of O3 exposure on California black  oak (Quercus kelloggii
        Newb.) seedlings with and without N amendment. Environ Pollut 156: 53-60.
        http://dx.doi.0rg/10.1016/i.envpol.2008.01.002.
Hanson. PJ: Wullschleger.  SD: Norbv. RJ: Tschaplinski. TJ: Gunderson. CA. (2005). Importance of changing
        CO2, temperature, precipitation, and ozone on carbon and water cycles of an upland-oak forest:
        incorporating experimental results into model simulations. Global Change Biol 11: 1402-1423.
        http://dx.doi.0rg/10.1111/i.1365-2486.2005.00991.x.
Harward, M: Treshow,  M. (1975). Impact of ozone on the growth and  reproduction of understorey plants in the
        Aspen zone of western USA. Environ Conserv 2: 17-23.
        http://dx.doi.org/10.1017/S0376892900000564.
Hassan, R:  Scholes, R: Ash, N. (2005). Ecosystems and human well-being: Current state and trends. In
        Millennium ecosystem assessment series (Vol.  1). Washington, DC:  Island Press.
Haves. F: Jones. MLM: Mills. G: Ashmore. M. (2007). Meta-analysis  of the relative sensitivity of semi-natural
        vegetation species to ozone. Environ Pollut 146: 754-762.
        http://dx.doi.0rg/10.1016/i.envpol.2006.06.011.
Draft - Do Not Cite or Quote                        9-159                                  September 2011

-------
Haves. F: Mills. G: Ashmore. M. (2009). Effects of ozone on inter- and intra-species competition and
        photosynthesis in mesocosms of Lolium perenne and Trifolium repens. Environ Pollut 157: 208-214.
        http://dx.doi.0rg/10.1016/i.envpol.2008.07.002.
He. XY: Ruan. YN: Chen. W: Lu. T. (2006).  Responses of anti-oxidative system in leaves of Ginkgo biloba to
        elevated ozone concentration in urban area. Botanical Studies 47: 409-416.
He. XY: Fu. SL: Chen. W: Zhao. TH: Xu. S: Tuba. Z. (2007). Changes in effects of ozone exposure on growth,
        photosynthesis, and respiration of Ginkgo biloba in Shenyang urban area. Photosynthetica 45: 555-
        561. http://dx.doi.org/10.1007/s11099-007-0095-0.
Heagle. AS: Body. DE: Heck. WW. (1973). An open-top field chamber to assess the impact of air pollution on
        plants. J Environ Qual 2: 365-368.
Heagle, AS. (1979). Effects of growth media, fertiliser rate and hour and season of exposure on sensitivity of
        four soybean cultivars to ozone. Environ Pollut 18: 313-322.
Heagle. AS: Letchworth. MB: Mitchell. CA. (1983). Effects of growth medium and fertilizer rate on the yield
        response of soybeans exposed to chronic doses of ozone. Phytopathology 73: 134-139.
        http://dx.doi. orq/10.1094/Phvto-73-134.
Heagle. AS: Heck. WW: Lesser. VM: Rawlings. JO. (1987). Effects of daily ozone exposure duration and
        concentration fluctuation on yield of tobacco. Phytopathology 77: 856-862.
        http://dx.doi. orq/10.1094/Phvto-77-856.
Heagle. AS: Kress. LW: Temple. PJ: Kohut. RJ: Miller. JE: Heggestad. HE. (1988). Factors influencing ozone
        dose-yield response relationships in open-top field chamber studies. In WW Heck; OC Taylor; DT
        Tingey (Eds.), Assessment of crop loss from air  pollutants: Proceedings of an international conference
        (pp. 141-179). Raleigh, NC: Elsevier Applied Science.
Heagle, AS. (1989). Ozone and crop yield*. Annu Rev Phytopathol 27: 397-423.
        http://dx.doi.org/10.1146/annurev.py.27.090189.002145.
Heagle. AS: Miller. JE: Rawlings. JO: Vozzo. SF. (1991).  Effect of growth stage on soybean response to chronic
        ozone exposure. J Environ Qual 20: 562-570.
        http://dx.doi.org/10.2134/ieg1991.00472425002000030010x.
Heagle. AS: Brandenburg. RL: Burns. JC: Miller. JE. (1994a). Ozone and carbon dioxide effects on spider mites
        in white clover and peanut. J Environ Qual 23: 1168-1176.
        http://dx.doi.org/10.2134/ieg1994.00472425002300060006x.
Heagle. AS: Miller. JE: Sherrill. DE. (1994b). A white clover system to estimate effects  of tropospheric ozone on
        plants. J Environ Qual 23: 613-621. http://dx.doi.org/10.2134/jeg1994.00472425002300030030x.
Heagle. AS: Reinert. RA: Miller. JE. (1996). Response of white clover to ozone in  different environments. J
        Environ Qual 25: 273-278. http://dx.doi.org/10.2134/ieg1996.0047242500250002001 Ox.
Heath, RL. (2008). Modification of the biochemical pathways of plants induced by ozone: What are the varied
        routes to change?  Environ Pollut 155: 453-463.  http://dx.doi.Org/10.1016/i.envpol.2008.03.010.
Heath. RL: Lefohn. AS: Musselman. RC. (2009). Temporal processes that contribute to nonlinearity in
        vegetation responses to ozone exposure and dose. Atmos Environ 43: 2919-2928.
        http://dx.doi.0rg/10.1016/i.atmosenv.2009.03.011.
Heck. WW: Philbeck.  RB: Dunning. JA. (1978). A continuous stirred tank reactor (CSTR) system for exposing
        plants to gaseous air contaminants: Principles, specifications, construction, and operation. Washington,
        DC: U.S. Government Printing Office.
Heck. WW: Taylor. OC: Adams. R: Bingham. G: Miller. J:  Preston. E: Weinstein. L. (1982). Assessment of crop
        loss from ozone. J Air Pollut Control Assoc 32: 353-361.
Heck. WW: Cure. WW: Rawlings. JO: Zaragoza.  LJ: Heagle. AS: Heggestad. HE: Kohut. RJ: Kress. LW:
        Temple. PJ. (1984). Assessing impacts of ozone on agricultural crops: II.  Crop yield functions and
        alternative exposure statistics. J Air Pollut Control Assoc 34:  810-817.
Heck. WW: Heagle. AS: Miller. JE: Rawlings. JO. (1991). A national program (NCLAN) to assess the impact of
        ozone on agricultural resources. In RL Berglund; DR Lawson; DJ McKee (Eds.), Tropospheric ozone
        and the environment: papers from an  international conference; March 1990; Los Angeles,  CA(pp. 225-
        254).  Pittsburgh, PA: Air & Waste Management Association.
Heck. WW: Cowling. EB. (1997). The need  for a long term cumulative secondary ozone standard - An
        ecological perspective. EM January: 23-33.
Draft - Do Not Cite or Quote                       9-160                                  September 2011

-------
Heggestad. HE. (1991). Origin of Bel-W3, Bel-C and Bel-B tobacco varieties and their use as indicators of
        ozone. Environ Pollut 74: 263-291. http://dx.doi.org/10.1016/0269-7491 (91)90076-9.
Heidenreich. B: Haberer. G: Mayer. K: Sandermann. H: Ernst. D. (2005). CDNA array-analysis of mercury- and
        ozone-induced genes in Arabidopsis thaliana. Acta Physiologiae Plantarum 27: 45-51.
        http://dx.doi. org/10.1007/s11738-005-0035-1.
Hendrev. GR: Kimball. BA. (1994). The FACE program. Agr Forest Meteorol 70: 3-14.
        http://dx.doi. orq/10.1016/0168-1923(94)90044-2.
Hendrev, GR: Ellsworth, PS: Lewin, KF: Nagy, J. (1999). A free-air enrichment system for exposing tall forest
        vegetation to elevated atmospheric CO2. Global Change Biol 5: 293-309.
        http://dx.doi. orq/10.1046/1.1365-2486.1999.00228.x.
Hildebrand, E: Skellv, JM: Fredericksen, TS. (1996). Foliar response of ozone-sensitive hardwood tree species
        from 1991 to 1993 in the Shenandoah  National Park, Virginia. Can J For Res 26: 658-669.
Hillstrom. M: Meehan. TD: Kelly. K: Lindroth.  RL. (2010a). Soil carbon and nitrogen mineralization following
        deposition of insect frass and greenfall from forests under elevated CO2 and O3. Plant Soil 336: 75-85.
        http://dx.doi.ora/10.1007/s11104-010-0449-4.
Hillstrom. ML: Lindroth. RL. (2008). Elevated  atmospheric carbon dioxide and ozone alter forest insect
        abundance and community composition.  Insect Conservation and Diversity 1: 233-241.
        http://dx.doi.0rg/10.1111/i.1752-4598.2008.00031.x.
Hillstrom. ML: Vigue. LM: Covle. PR: Raffa. KF: Lindroth. RL. (201 Ob). Performance of the invasive weevil
        Polydrusus sericeus is influenced by atmospheric  CO2 and host species. Agr Forest Entomol 12: 285-
        292.  http://dx.doi.0rg/10.1111/i.1461-9563.2010.00474.x.
Himanen. SJ: Nerg. AM: Holopainen. JK. (2009a). Degree  of herbivore feeding damage as an important
        contributor to multitrophic plant-parasitoid signaling under climate change.  4: 249-251.
Himanen. SJ: Nerg. AM: Nissinen. A: Pinto. DM: Stewart. CN: Poppy. GM: Holopainen. JK. (2009b). Effects of
        elevated carbon dioxide and ozone on volatile terpenoid emissions and multitrophic communication of
        transgenic insecticidal oilseed rape (Brassica  napus). New Phytol 181: 174-186.
        http://dx.doi.0rg/10.1111/i.1469-8137.2008.02646.x.
Himanen. SJ: Nerg. AM: Nissinen. A: Stewart. CN: Poppy.  GM: Holopainen. JK. (2009c). Elevated atmospheric
        ozone increases concentration of insecticidal  Bacillus thuringiensis (Bt) Cry 1 Ac protein in Bt Brassica
        napus and reduces feeding of a Bt target herbivore on the non-transgenic parent. Environ Pollut 157:
        181-185. http://dx.doi.0rg/10.1016/i.envpol.2008.07.006.
Hogg, A: Uddling, J: Ellsworth, D: Carroll, MA: Presslev, S: Lamb, B: Vogel, C. (2007). Stomatal and non-
        stomatal fluxes of ozone to a northern  mixed hardwood forest. Tellus B Chem Phys  Meteorol 59: 514-
        525.  http://dx.doi.0rg/10.1111/i.1600-0889.2007.00269.x.
Hogsett, WE: Tingey,  DT: Holman, SR. (1985). A programmable exposure  control system for determination of
        the effects of pollutant exposure regimes on plant  growth. Atmos Environ 19: 1135-1145.
        http://dx.doi. org/10.1016/0004-6981(85)90198-2.
Hogsett, WE: Olszyk, D: Ormrod, DP: Taylor,  GE,  Jr: Tingey, DT. (1987a). Air pollution exposure systems and
        experimental protocols: Volume 2: Description of facilities. (EPA/600/3-87/037b). Corvallis, OR: U.S.
        Environmental Protection Agency, Environmental Research Laboratory.
        http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=30000KQH.txt.
Hogsett. WE: Olszvk. D: Ormrod. DP: Taylor.  GE.  Jr: Tingev. DT. (1987b). Air pollution exposure systems and
        experimental protocols: Volume I: A review  and evaluation of performance. (EPA/600/3-87/037a).
        Corvallis, OR: U.S. Environmental Protection Agency.
Hogsett. WE: Tingev.  DT: Lee. EH. (1988). Ozone exposure indices: Concepts for development and evaluation
        of their use.  In Assessment of crop loss from air pollutants: Proceedings of an international conference
        (pp. 107-138). New York:  Elsevier Applied Science.
Hogsett. WE: Weber. JE: Tingev. D: Herstrom. A:  Lee.  EH:  Laurence. JA. (1997). Environmental auditing: An
        approach for characterizing tropospheric  ozone risk to forests. J Environ Manage 21: 105-120.
        http://dx.doi. org/10.1007/s002679900010.
Hogsett. WE: Tingev.  DT: Lee. EH: Beedlow.  PA: Andersen. CP (2008). An approach for evaluating the
        effectiveness of various ozone Air Quality Standards for protecting trees. Environ Manage 41: 937-948.
        http://dx.doi. org/10.1007/s00267-007-9057-3.
Draft - Do Not Cite or Quote                        9-161                                  September 2011

-------
Holmes. WE: Zak. PR: Pregitzer. KS: King. JS. (2006). Elevated CO2 and O3 alter soil nitrogen transformations
        beneath trembling aspen, paper birch, and sugar maple. Ecosystems 9: 1354-1363.
        http://dx.doi.org/10.1007/s10021-006-0163-5.
Hong. B: Weinstein. D: Swanev. D. (2006). Assessment of ozone effects on nitrate export from Hubbard Brook
        Watershed 6. Environ Pollut 141: 8-21. http://dx.doi.Org/10.1016/i.envpol.2005.08.030.
Horvath. L: Nagy. Z: Weidinger. T: Artz. R: Luke. WT: Valigura. R: Pinto. JP: Womack. J.  (1995). Measurement
        effluxes of trace gases (O3, NOX, SO2, CO2, HNO3), particulate sulfate and nitrate, water vapour
        over short vegetation by gradient and eddy correlation techniques in Hungary. Ann Geophys 13: C490.
Howard. AR: van lersel. MW: Richards. JH: Donovan. LA. (2009). Night-time transpiration can decrease
        hydraulic redistribution. Plant Cell Environ 32:  1060-1070. http://dx.doi.Org/10.1111/i. 1365-
        3040.2009.01988.x.
Hui. D: Sims. DA: Johnson. DW:  Chang. W: Luo. Y. (2002). Effects of gradual versus step increases in carbon
        dioxide on Plantago photosynthesis and growth in a microcosm study. Environ Exp Bot 47: 51-66.
Innes, Jl_:  Skellv, JM: Schaub, M. (2001). Ozone and broadleaved species: A guide to the identification of
        ozone-induced foliar injury.  In. Bern, Switzerland: Paul Haupt Publishers.
IPCC. (Intergovernmental Panel on Climate Change). (2007a). Summary for policymakers. In: Impacts,
        Adaptation and Vulnerability. Contribution  of Working Group II to the Fourth  Assessment Report of the
        Intergovernmental Panel on Climate Change. In Climate Change 2007. Cambridge, UK: Cambridge
        University Press.
Iriti, M: Faoro, F.  (2009). Chemical diversity and defence metabolism: How plants cope with pathogens and
        ozone pollution. International Journal of Molecular Sciences 10: 3371-3399.
        http://dx.doi.org/10.3390/iims10083371.
Iriti, M: Pi  Maro, A: Bernasconi, S: Burlini, N: Simonetti, P:  Picchi, V:  Panigada, C: Gerosa,  G: Parente, A:
        Faoro. F. (2009). Nutritional traits of bean  (Phaseolus vulgaris) seeds from plants chronically exposed
        to ozone pollution. J Agric Food Chem 57: 201-208. http://dx.doi.org/10.1021/if802819m.
IRRI. (International Rice Research Institute). (2002). Annual Report.  Los Banos, Laguna  in the Philippines.
        http://irri.org/.
Isebrands. JG: Dickson. RE: Rebbeck. J: Karnoskv. DF. (2000). Interacting effects of multiple stresses on
        growth and physiological processes in northern forest trees. In RA Mickler; RABirsdey; J Horn (Eds.),
        Responses of northern US forests to environmental change (pp. 149-180). New York, NY: Springer-
        Verlag.
Isebrands, JG: McDonald, EP: Kruger, E: Hendrev, G: Percy, K: Pregitzer, K: Sober,  J: Karnoskv, DF. (2001).
        Growth responses of Populus tremuloides clones to interacting carbon dioxide and tropospheric ozone.
        Environ  Pollut 115: 359-371.
Jackson, DM: Heagle, AS: Eckel, RVW. (1999). Ovipositional response of tobacco hornworm moths
        (Lepidoptera: Sphyngidae) to tobacco plants grown under elevated levels of ozone. Environ Entomol
        28: 566-571.
Janzik, I: Preiskowski, S: Kneifel, H. (2005). Ozone has dramatic effects on the regulation of the prechorismate
        pathway in tobacco (Nicotiana tabacum L. cv. Bel W3). Planta 223: 20-27.
        http://dx.doi. org/10.1007/s00425-005-0060-8.
Johnson, RM: Pregitzer, KS. (2007). Concentration of sugars, phenolic acids, and amino acids in forest soils
        exposed to elevated atmospheric CO2 and O3. Soil Biol Biochem 39: 3159-3166.
        http://dx.doi.0rg/10.1016/i.soilbio.2007.07.010.
Jones, ME: Paine, TD. (2006). Detecting changes  in insect herbivore communities along a pollution gradient.
        Environ  Pollut 143: 377-387. http://dx.doi.Org/10.1016/i.envpol.2005.12.013.
Jones. MLM: Hodges. G: Mills. G. (2010). Nitrogen mediates above-ground effects of ozone but not below-
        ground effects in a rhizomatous sedge. Environ Pollut 158: 559-565.
        http://dx.doi.0rg/10.1016/i.envpol.2009.08.002.
Jones. TG: Freeman. C: Lloyd. A: Mills. G. (2009).  Impacts of elevated atmospheric ozone on peatland below-
        ground doc characteristics.  Ecol Eng 35: 971-977. http://dx.doi.Org/10.1016/i.ecoleng.2008.08.009.
Joo. JH: Wang. SY:  Chen. JG: Jones. AM: Fedoroff. NV. (2005). Different signaling and cell death roles of
        heterotrimeric G protein  alpha and beta subunits in the arabidopsis oxidative stress response to ozone.
        Plant Cell 17: 957-970. http://dx.doi.org/10.1105/tpc.104.029603.
Draft - Do Not Cite or Quote                        9-162                                 September 2011

-------
Joss. U: Graber. WK. (1996). Profiles and simulated exchange of H2O, O3, NO2 between the atmosphere and
        the HartX Scots pine plantation. Theor Appl Climatol 53: 157-172.
Jovan.  S: McCune. B. (2006). Using epiphytic macrolichen communities for biomonitoring ammonia in forests of
        the greater Sierra Nevada, California. Water Air Soil Pollut 170: 69-93.
Kanerva, T: Palojarvi, A: Ramo, K: Ojanpera, K: Esala, M: Manninen, S. (2006). A 3-year exposure to CO2 and
        O3 induced minor changes in soil N cycling in a meadow ecosystem. Plant Soil 286: 61-73.
        http://dx.doi.ora/10.1007/s11104-006-9026-2.
Kanerva, T: Regina, K: Ramo, K:  Ojanpera, K: Manninen, S. (2007). Fluxes of N2O, CH4 and CO2 in a meadow
        ecosystem exposed to elevated ozone and carbon dioxide for three years. Environ Pollut 145: 818-828.
        http://dx.doi.0rg/10.1016/i.envpol.2006.03.055.
Kanerva, T: Palojarvi, A: Ramo, K: Manninen, S. (2008). Changes in soil microbial community structure under
        elevated tropospheric O3 and CO2. Soil Biol  Biochem 40: 2502-2510.
        http://dx.doi.ora/10.1016/i.soilbio.2008.06.007.
Kangasjarvi, J: Jaspers, P: Kollist, H. (2005). Signalling and cell death in ozone-exposed plants. Plant Cell
        Environ 28: 1021-1036.
Karlsson. PE:  Sellden.  G: Plaiiel.  H. (2003). Establishing ozone critical levels II UNECE workshop report.
        Gothenburg, Sweden: IVL Swedish Environmental  Institute.
Karlsson. PE:  Uddling.  J: Braun. S: Broadmeadow. M: Elvira. S: Gimeno. BS:  Le Thiec. D: Okansen. E:
        Vandermeiren. K: Wilkinson. M: Emberson. L. (2004).  New critical levels for ozone effects on young
        trees  based on AOT40 and simulated cumulative leaf uptake of ozone. Atmos Environ 38: 2283-2294.
Karnoskv. DF: Gagnon. ZE: Dickson. RE: Coleman. MD:  Lee. EH: Isebrands.  JG. (1996). Changes in growth,
        leaf abscission, biomass associated with seasonal tropospheric ozone exposures of Populus
        tremuloides clones and seedlings. Can J For Res 26: 23-37.
Karnoskv. DF: Mankovska. B: Percy. K: Dickson. RE:  Podila. GK: Sober. J: Noormets. A: Hendrev. G: Coleman.
        MD: Kubiske.  M: Pregitzer. KS: Isebrands. JG. (1999). Effects of tropospheric ozone on trembling
        aspen and interaction with CO2: Results from an OS-gradient and a FACE experiment. Water Air Soil
        Pollut 116: 311-322.
Karnoskv. DF: Zak. PR: Pregitzer. KS: Awmack. CS: Bockheim. JG: Dickson.  RE: Hendrev. GR: Host. GE: King.
        JS: Kopper, BJ: Kruger, EL: Kubiske, ME: Lindroth, Rl_: Mattson, WJ: McDonald, EP: Noormets, A:
        Oksanen. E: Parsons. WFJ: Percy. KE: Podila. GK: Riemenschneider. DE: Sharma. P: Thakur. R:
        Sober. A: Sober. J: Jones. WS: Anttonen.  S: Vapaavuori. E: Mankovska. B: Heilman. W:  Isebrands. JG.
        (2003). Tropospheric O3 moderates responses of temperate hardwood forests to elevated CO2: A
        synthesis of molecular to ecosystem results from the Aspen FACE project. Funct Ecol 17: 289-304.
Karnoskv. DF: Pregitzer. KS: Zak. PR: Kubiske. ME: Hendrev. GR: Weinstein. D: Nosal. M: Percy. KE. (2005).
        Scaling ozone responses of forest trees to the ecosystem level in a changing climate.  Plant  Cell
        Environ 28: 965-981. http://dx.doi.Org/10.1111/i.1365-3040.2005.01362.x.
Kasurinen. A:  Keinanen. MM: Kaipainen. S: Nilsson. LO: Vapaavuori. E: Kontro. MH: Holopainen. T. (2005).
        Below-ground  responses of silver birch trees exposed to elevated CO2 and O3 levels during three
        growing seasons. Global Change Biol 11: 1167-1179. http://dx.doi.Org/10.1111/i. 1365-
        2486.2005.00970.x.
Kasurinen, A:  Riikonen, J: Oksanen, E: Vapaavuori, E: Holopainen, T.  (2006).  Chemical composition and
        decomposition of silver birch leaf litter produced under elevated CO2 and O3. Plant Soil 282: 261-280.
        http://dx.doi.org/10.1007/s11104-005-6026-6.
Kasurinen, A:  Peltonen, PA: Julkunen-Tiitto, R: Vapaavuori,  E: Nuutinen, V:  Holopainen, T: Holopainen, JK.
        (2007). Effects of elevated CO2 and O3 on leaf litter phenolics and subsequent performance of litter-
        feeding soil macrofauna. Plant Soil 292: 25-43. http://dx.doi.org/10.1007/s11104-007-9199-3.
Kats, G: Thompson, CR: Kubv, WC. (1976). Improved ventilation of open top greenhouses. J Air Pollut Control
        Assoc26: 1089-1090.
Kats. G: Olszvk. DM: Thompson.  CR. (1985). Open top experimental chambers for trees. J Air Waste Manag
        Assoc35: 1298-1301.
Keller. T: Hasler.  R. (1984). The influence of a fall fumigation with ozone on the stomatal behavior of spruce and
        fir. Oecologia 64: 284-286. http://dx.doi.org/10.1007/BF00376884.
Draft - Do Not Cite or Quote                       9-163                                 September 2011

-------
Kellomaki. S: Wang. K. -Y. (1997). Effects of elevated O3 and CO2 concentrations on photosynthesis and
        stomatal conductance in Scots pine. Plant Cell Environ 20: 995-1006. http://dx.doi.org/10.1111/1.1365-
        3040.1997.tb00676.x.
Kerner. R: Winkler. J: Dupuv. J: Jurgensen. M: Lindermavr. C: Ernst. D: Muller-starck. G. (2011). Changes in the
        proteome of juvenile European beech following three years exposure to free-air elevated ozone. iForest
        4: 69-76. http://dx.doi.org/10.3832/ifor0570-004.
Keutgen. AJ: Noga. G: Pawelzik. E. (2005). Cultivar-specific impairment of strawberry growth, photosynthesis,
        carbohydrate and nitrogen accumulation by ozone. Environ Exp Bot 53: 271-280.
        http://dx.doi.0rg/10.1016/i.envexpbot.2004.04.003.
Keutgen. N: Keutgen. AJ: Janssens. MJJ. (2008). Sweet potato [Ipomoea batatas (L.) Lam.] cultivated as tuber
        or leafy vegetable supplier as affected by elevated tropospheric ozone. J Agric Food Chem 56: 6686-
        6690. http://dx.doi.org/10.1021/if8006272.
King. JS: Pregitzer. KS: Zak. PR: Sober. J: Isebrands. JG:  Dickson. RE: Hendrev. GR: Karnoskv. DF.  (2001).
        Fine-root biomass and fluxes of soil carbon in young stands of paper birch and trembling aspen as
        affected by elevated atmospheric CO2 and tropospheric O3. Oecologia 128: 237-250.
King. JS: Kubiske. ME: Pregitzer. KS: Hendrev. GR: McDonald. EP: Giardina. CP:  Quinn. VS: Karnoskv. DF.
        (2005). Tropospheric O3 compromises net primary production in young stands of trembling aspen,
        paper birch  and sugar maple in response to elevated atmospheric CO2. New Phytol 168: 623-635.
        http://dx.doi.0rg/10.1111/i.1469-8137.2005.01557.x.
Kitao, M: Low, M: Heerdt, C: Grams, TEE: Haberle, KH: Matyssek, R.  (2009). Effects of chronic elevated ozone
        exposure on gas exchange responses of adult beech trees (Fagus sylvatica) as related to the within-
        canopy light gradient.  Environ Pollut 157: 537-544. http://dx.doi.Org/10.1016/i.envpol.2008.09.016.
Kline, LJ: Davis, DP: Skellv, JM: Savage, JE: Ferdinand, J. (2008). Ozone sensitivity of 28 plant selections
        exposed to ozone under controlled conditions. Northeast Nat 15: 57-66. http://dx.doi.org/10.1656/1092-
        6194(2008)15r57:OSOPSE12.0.CO:2.
Kline, LJ: Davis, DP: Skellv, JM: Decoteau, DR. (2009). Variation in ozone sensitivity within Indian hemp and
        common milkweed selections from the Midwest. Northeast Nat 16:  307-313.
        http://dx.doi.org/10.1656/045.016.0210.
Kohut, R. (2007). Assessing the risk of foliar injury from ozone on vegetation in parks  in the US National Park
        Service's Vital Signs Network. Environ Pollut 149:  348-357.
        http://dx.doi.0rg/10.1016/i.envpol.2007.04.022.
Kolb, TE: Fredericksen, TS: Steiner, KC: Skellv, JM. (1997). Issues in  scaling tree size and age responses to
        ozone: A review [Review]. Environ Pollut 98: 195-208. http://dx.doi.org/10.1016/S0269-7491 (97)00132-
        2.
Kollist, T: Moldau, H: Rasulov,  B: Oja, V: Ramma, H: Huve, K: Jaspers, P: Kangasjarvi, J: Kollist, H. (2007). A
        novel device detects a rapid ozone-induced transient stomatal closure in intact Arabidopsis and its
        absence in abi2 mutant. Physiol Plant 129: 796-803. http://dx.doi.Org/10.1111/i. 1399-
        3054.2006.00851.x.
Kostka-Rick. R:  Hahn. HU. (2005). Biomonitoring using tobacco Bel W3 provides supplemental information for
        risk assessment of vegetation injury due to ozone. Gefahrstoffe Reinhaltung Der Luft 65: 485-491.
Kozovits, AR:  Matvssek, R: Blaschke, H: Gottlein, A: Grams, TEE. (2005). Competition increasingly dominates
        the responsiveness of juvenile beech and spruce to elevated  CO2 and/or  O3 concentrations throughout
        two subseguent growing seasons. Global Change Biol 11: 1387-1401. http://dx.doi.Org/10.1111/i. 1365-
        2486.2005.00993.x.
Krupa. SV: Grunhage. L: Jager. H. -J: Nosal.  M: Manning. WJ: Legge. AH: Hanewald. K. (1995). Ambient ozone
        (O3) and adverse crop response: A unified view of cause and effect. Environ  Pollut 87: 119-126.
        http://dx.doi.org/10.1016/S0269-7491(99)80014-1.
Krupa. SV: Nosal. M: Peterson. PL. (2001). Use of passive ozone O3  samplers in vegetation effects
        assessment. Environ Pollut 112: 303-309.
Kubiske, ME: Quinn, VS: Heilman, WE: McDonald, EP: Marguardt, PE: Teclaw, RM: Friend, Al_: Karnoskev, DF.
        (2006). Interannual climatic variation mediates elevated CO2  and O3 effects on forest growth. Global
        Change Biol 12: 1054-1068.  http://dx.doi.Org/10.1111/i.1365-2486.2006.01152.x.
Draft - Do Not Cite or Quote                        9-164                                  September 2011

-------
Kubiske. ME: Quinn. VS: Marquardt. PE: Karnoskv. PR (2007). Effects of elevated atmospheric CO2 and/or O3
       on intra- and interspecific competitive ability of aspen. Plant Biol (Stuttg) 9: 342-355.
       http://dx.doi. orq/10.1055/S-2006-924760.
Laffrav. X:  Rose.  C: Garrec. JP. (2007). Estimation of ozone concentration in a valley of the alps mountains
       based on bel-w3 tobacco leaf injury. Water Air Soil Pollut 186: 29-42. http://dx.doi.org/10.1007/s11270-
       007-9460-7.
Langebartels. C:  Kerner. K: Leonard!. S: Schraudner. M: Trost. M: Heller. W: Sandermann. H. Jr. (1991).
       Biochemical plant responses to ozone: I. Differential induction of polyamine and ethylene biosynthesis
       in tobacco. J Plant Physiol 95: 882-889. http://dx.doi.Org/10.1104/pp.95.3.882.
Larson. JL: Zak. PR: Sinsabaugh. RL. (2002). Extracellular enzyme activity beneath temperate trees growing
       under elevated carbon dioxide and ozone. Soil Sci Soc Am J 66: 1848-1856.
Lawlor. DW. (1998). Plant responses to global change: Temperature and drought stress. In LJ De Kok; I Stulen
       (Eds.), Responses of plant metabolism to air pollution and global change. Leiden, The Netherlands:
       Backhuys Publishers.
Leakey. ADB: Bernacchi. CJ: Ort. PR:  Long. SP (2006). Long-term growth of soybean at elevated CO2 does
       not cause acclimation of stomatal conductance under fully open-air conditions. Plant Cell Environ 29:
       1794-1800. http://dx.doi.0rg/10.1111/i.1365-3040.2006.01556.x.
Lee. EH: Tingev.  DT: Hogsett. WE. (1987). Selection of the best exposure-response model using various 7-hour
       ozone exposure statistics. Research Triangle Park, NC: U.S. Environmental Protection Agency.
Lee, EH: Tingev,  DT: Hogsett, WE. (1988a). Evaluation of ozone-exposure indices for relating exposure to plant
       production and for estimating  agricultural losses. (EPA/600/3-88/039). Washington, DC: U.S.
       Environmental Protection Agency.
Lee, EH: Tingev,  DT: Hogsett, WE. (1988b). Evaluation of ozone exposure indices in exposure-response
       modeling. Environ  Pollut 53: 43-62. http://dx.doi.org/10.1016/0269-7491 (88)90024-3.
Lee. EH: Tingev.  DT: Hogsett. WE. (1989). Interrelation of experimental exposure and ambient air quality data
       for comparison of ozone exposure indices and estimating agricultural losses. (EPA/600/3-89/047).
       Corvallis, OR: U.S. Environmental Protection Agency.
Lee. EH: Hogsett. WE: Tingev. DT. (1994). Attainment and effects issues regarding alternative secondary ozone
       air quality standards. J Environ Qual 23: 1129-1140.
       http://dx.doi.org/10.2134/ieg1994.00472425002300060002x.
Lee. EH: Hogsett. WE. (1996). Methodology for calculating inputs for ozone secondary standard benefits
       anaylsis: Part II. Research Triangle Park, NC: U.S.  Environmental Protection Agency.
Lee. EH: Hogsett. WE. (1999). Role of concentrations and time of day in developing ozone exposure indices for
       a secondary standard.  J Air Waste Manag Assoc 49: 669-681.
Lee, EH: Tingev,  DT: Hogsett, WE: Laurence, JA. (2003a).  History of tropospheric ozone for the San Bernardino
       Mountains of southern  California, 1963-1999. Atmos Environ 37: 2705-2717.
       http://dx.doi. orq/10.1016/51352-2310(03)00203-6.
Lee, EH: Tingev,  DT: Waschmann, RS: Phillips, PL: Olszyk, DM: Johnson, MG: Hogsett, WE. (2009a).
       Seasonal and long-term effects of CO2 and O-3 on water loss in ponderosa pine and their interaction
       with climate and soil moisture. Tree Physiol 29: 1381-1393. http://dx.doi.org/10.1093/treephvs/tpp071.
Lee, S: Yun, SC.  (2006). The ozone stress transcriptome of pepper  (Capsicum annuum L.). Molecules and Cells
       21: 197-205.
Lee. WS:  Chevone. Bl: Seller. JR. (1990). Growth and gas exchange of loblolly pine seedlings as influenced by
       drought and air pollutants. Water Air Soil Pollut 51:  105-116. http://dx.doi.org/10.1007/BF00211508.
Lefohn. AS: Benedict. HM.  (1982). Development of a mathematical  index that describes ozone concentration,
       frequency and duration. Atmos Environ 16: 2529-2532. http://dx.doi.org/10.1016/0004-6981 (82)90145-
       7.
Lefohn. AS: Laurence. JA: Kohut.  RJ. (1988). A comparison of indices that describe the relationship between
       exposure to ozone and reduction in the yield of agricultural crops. Atmos Environ 22: 1229-1240.
       http://dx.doi.org/10.1016/0004-6981(88)90353-8.
Lefohn. AS: Jackson. W: Shadwick. PS: Knudsen. HP. (1997). Effect of surface ozone exposures on vegetation
       grown in the southern Appalachian Mountains: Identification of possible areas of concern. Atmos
       Environ 31: 1695-1708. http://dx.doi.Org/10.1016/S1352-2310(96)00258-0.
Draft - Do Not Cite or Quote                        9-165                                  September 2011

-------
Lefohn. AS: Shadwick. PS. (2000). Differences in trending estimates in the United States using several ozone
        metrics. In Proceedings of the 93rd Air & Waste Management Association Annual Conference and
        Exhibition (pp. AS 1d-645). Pittsburgh, PA: Air & Waste Management Association.
Legge. AH: Grunhage. L: Nosal. M: Jager. H. -J: Krupa. SV.  (1995). Ambient ozone and adverse crop response:
        An evaluation of North American and European data as they relate to exposure indices and critical
        levels. J Appl Bot Food Qual 69: 192-205.
Leitao. L: Delacote. E: Dizengremel. P: Le Thiec. D: Biollev.  JP (2007a). Assessment of the impact of increasing
        concentrations of ozone on photosynthetic components of maize (Zea mays L.), a C-4 plant. Environ
        Pollut 146: 5-8. http://dx.doi.Org/10.1016/i.envpol.2006.05.019.
Leitao. L: Maoret. JJ:  Biollev. JP. (2007b). Changes in PEP carboxylase, rubisco and rubisco activase mRNA
        levels from maize (Zea mays) exposed to a chronic ozone stress. Biol Res 40: 137-153.
        http://dx.doi.ora/10.4067/S0716-97602007000200005.
Leitao. L: Bethenod. O: Biollev. JP. (2007c). The impact of ozone on juvenile maize (Zea mays L.) plant
        photosynthesis: Effects on vegetative biomass, pigmentation, and carboxylases (PEPc and Rubisco).
        Plant Biol (Stuttg) 9: 478-488. http://dx.doi.orq/10.1055/S-2007-964942.
Lesser. VM: Rawlings. JO: Spruill. SE: Somerville. MC.  (1990). Ozone effects on  agricultural crops: Statistical
        methodologies and estimated dose-response relationships. Crop Sci 30: 148-155.
Leuning. R: Unsworth. MH: Neumann. HN: King. KM. (1979). Ozone fluxes to tobacco and soil under field
        conditions. Atmos Environ 13: 1155-1163. http://dx.doi.org/10.1016/0004-6981 (79)90039-8.
Levine, JS: Pinto, JP.  (1998). The production of CO by biomass burning. In MAK Khalil; JP Pinto; MJ Shearer
        (Eds.), Atmospheric carbon monoxide and its environmental effects: Proceedings of the international
        conference; December 1997; Portland, Oregon (pp. 251-256). Portland, OR:  U.S. Environmental
        Protection Agency, Office of Research and Development.
Lewis. JS: Ditchkoff. SS: Lin. JC: Muntifering. RB: Chappelka. AH. (2006). Nutritive quality of big  bluestem
        (Andropogon gerardii) and eastern gamagrass (Tripsacum dactyloides) exposed to tropospheric ozone.
        Rangeland Ecol Manag 59: 267-274.
Li. PH: Mane. SP:  Sioson. AA: Robinet. CV: Heath. LS:  Bohnert. HJ: Grene.  R. (2006b). Effects of chronic
        ozone exposure on gene expression in Arabidopsis thaliana ecotypes and in  Thellungielia halophila.
        Plant Cell Environ 29: 854-868. http://dx.doi.Org/10.1111/i.1365-3040.2005.01465.x.
Lin. JC: Nosal. M:  Muntifering. RB: Krupa. SV. (2007). Alfalfa nutritive quality for ruminant livestock as
        influenced by ambient air quality in west-central Alberta. Environ Pollut 149: 99-103.
        http://dx.doi.0rg/10.1016/i.envpol.2006.12.009.
Lindroth. RL. (2010). Impacts of elevated atmospheric CO2 and O3 on forests: Phytochemistry, trophic
        interactions, and ecosystem  dynamics. J Chem Ecol 36: 21-Feb. http://dx.doi.org/10.1007/s10886-009-
        9731-4.
Liu. L: King. J: Giardina. C. (2005). Effects of elevated concentrations of atmospheric CO2 and tropospheric O3
        on leaf litter production and chemistry in trembling aspen and paper birch communities. Tree Physiol
        25: 1511-1522.
Liu. L: King. JS: Giardina. CP (2007a). Effects of elevated atmospheric CO2 and  tropospheric O3 on nutrient
        dynamics: Decomposition of leaf litter in termbling aspen and paper birch communities. Plant Soil 299:
        65-82.
Liu. LL: King. JS: Giardina. CP:  Booker. FL. (2009b).  The influence of chemistry, production and community
        composition on leaf litter decomposition under elevated atmospheric CO2 and tropospheric O-3 in a
        northern hardwood ecosystem. Ecosystems  12: 401-416.  http://dx.doi.org/10.1007/s10021-009-9231-y.
Loats. KV: Rebbeck. J. (1999). Interactive effects of ozone and elevated carbon dioxide on the growth and
        physiology of black cherry, green ash, and yellow poplar seedlings. Environ Pollut 106: 237-248.
        http://dx.doi.org/10.1016/S0269-7491(99)00069-X.
Long. SP. (1991). Modification of the response of photosynthetic productivity to rising temperature by
        atmospheric CO2 concentrations: Has its importance been underestimated?  Plant Cell Environ 14:
        729-739. http://dx.doi.0rg/10.1111/i.1365-3040.1991.tb01439.x.
Loranger. Gl: Pregitzer. KS: King. JS. (2004). Elevated CO2 and O3t concentrations differentially affect selected
        groups of the fauna in temperate forest soils. Soil Biol Biochem 36: 1521-1524.
Lorenzini, G; Nali,  C. (1995). Analysis of vertical ozone and nitrogen oxides profiles in a Prunus cerasifera
        canopy. Int J  Biometeorol 39: 1-4. http://dx.doi.orq/10.1007/BF01320885.
Draft - Do Not Cite or Quote                        9-166                                  September 2011

-------
Loreto. F: Velikova. V. (2001). Isoprene produced by leaves protects the photosynthetic apparatus against
        ozone damage, quenches ozone products, and reduces lipid peroxidation of cellular membranes. Plant
        Physiol 127: 1781-1787. http://dx.doi.org/10.1104/pp.010497.
Loreto. F: Fares. S. (2007). Is ozone flux inside leaves only a damage indicator? Clues from volatile isoprenoid
        studies.  Plant Physiol 143: 1096-1100. http://dx.doi.Org/10.1104/pp. 106.091892.
Low. M: Herbinger. K: Nunn. AJ: Haberle. KH: Leuchner. M: Heerdt. C: Werner. H: Wipfler. P: Pretzsch. H:
        Tausz. M: Matvssek. R. (2006). Extraordinary drought of 2003 overrules ozone impact on adult beech
        trees (Fagus sylvatica). Trees Struct Funct 20: 539-548. http://dx.doi.org/10.1007/s00468-006-0069-z.
Lova. WM: Pregitzer. KS: Karberg. NJ: King. JS: Giardina. CP (2003). Reduction of soil carbon formation by
        tropospheric ozone under elevated carbon dioxide. Nature 425: 705-707.
Ludwikow, A: Gallois, P: Sadowski, J. (2004). Ozone-induced oxidative stress response in Arabidopsis:
        Transcription profiling by microarray approach. Cell  Mol Biol Lett 9: 829-842.
Ludwikow. A: Sadowski. J. (2008). Gene networks in plant ozone stress response and tolerance. J Integr Plant
        Biol 50: 1256-1267. http://dx.doi.Org/10.1111/i.1744-7909.2008.00738.x.
Ludwikow. A: Kierzek. D: Gallois. P: Zeef. L: Sadowski. J. (2009). Gene expression profiling of ozone-treated
        Arabidopsis abiltd insertional mutant: Protein phosphatase  2C ABU  modulates biosynthesis ratio of
        ABA and ethylene. Planta 230: 1003-1017. http://dx.doi.org/10.1007/s00425-009-1001 -8.
Luo. Y: Reynolds. JF (1999). Validity of extrapolating field CO2 experiments to predict carbon sequestration in
        natural ecosystems. Ecology 80: 1568-1583.
Luo, Y. (2001). Transient ecosystem response to free-air CO2 enrichment (FACE): Experimental evidence and
        methods of analysis. New Phytol 152: 3-8.
Lyons.  TM: Barnes. JD.  (1998). Influence of plant age on ozone resistance in Plantago major. New Phytol 138:
        83-89. http://dx.doi.0rg/10.1046/i.1469-8137.1998.00879.x.
M. I: M. (International Cooperative Programme on Modelling and Mapping). (2004). Mapping critical levels for
        vegetation. In Manual on  methodologies  and criteria for modelling and mapping critical loads and
        levels, and air pollution effects, risks and trends.
Maggio. A: Chiaranda. FQ:  Cefariello. R: Fagnano. M. (2009). Responses to ozone pollution of alfalfa exposed
        to increasing salinity levels. Environ Pollut 157: 1445-1452.
        http://dx.doi.0rg/10.1016/i.envpol.2008.09.013.
Mahalingam. R: Shah. N: Scrvmgeour. A: Fedoroff. N. (2005). Temporal evolution of the Arabidopsis oxidative
        stress response. Plant Mol Biol 57: 709-730. http://dx.doi.org/10.1007/s11103-005-2860-4.
Mahalingam, R: Jambunathan, N: Gunjan, SK: Faustin, E: Weng, H:  Ayoubi, P. (2006). Analysis of oxidative
        signalling induced  by ozone in Arabidopsis thaliana. Plant Cell Environ 29: 1357-1371.
        http://dx.doi.0rg/10.1111/i.1365-3040.2006.01516.x.
Maier-Maercker,  U. (1998). Predisposition of trees to drought stress  by ozone. Tree Physiol 19: 71-78.
Mandl. RH: Weinstein. LH:  McCune. DC:  Kevenv.  M. (1973). A cylindrical, open-top chamber for the exposure of
        plants to air pollutants in the field. J Environ Qual 2: 371-376.
Mandl, RH: Laurence, JA: Kohut, RJ. (1989). Development and testing of open-top chambers for exposing
        large, perennial plants to  air pollutants. J Environ Qual 18: 534-540.
        http://dx.doi. org/10.2134/ieg1989.00472425001800040026x.
Mankovska, B: Percy, KE: Karnoskv, DF. (2005). Impacts of greenhouse gases on epicuticular waxes of
        Populus tremuloides Michx.: Results from an open-air exposure and a natural O3 gradient. Environ
        Pollut 137: 580-586. http://dx.doi.Org/10.1016/i.envpol.2005.01.043.
Manning, WJ: Krupa, SV. (1992). Experimental methodology for studying the effects of ozone on crops and
        trees. In AS Lefohn (Ed.), Surface level ozone exposures and their effects on vegetation (pp. 93-156).
        Chelsea, Ml: Lewis Publishers.
Manning, WJ. (2003). Detecting plant effects is necessary to give biological significance to ambient ozone
        monitoring data and predictive ozone standards. Environ Pollut 126: 375-379.
Martin. MJ: Host. GE: Lenz. KE: Isebrands. JG. (2001). Simulating the growth response of aspen to elevated
        ozone: A mechanistic approach to scaling a leaf-level model of ozone effects on photosynthesis to a
        complex canopy architecture. Environ Pollut 115: 425-436.
Massman. WJ: Grantz. DA. (1995). Estimating canopy conductance  to ozone uptake from observations of
        evapotranspiration at the canopy scale and at the leaf scale. Global Change Biol 1: 183-198.
        http://dx.doi.0rg/10.1111/i.1365-2486.1995.tb00020.x.
Draft - Do Not Cite or Quote                        9-167                                 September 2011

-------
Massman. WJ: Musselman. RC: Lefohn. AS. (2000). A conceptual ozone dose-response model to develop a
        standard to protect vegetation. Atmos Environ 34: 745-759. http://dx.doi.org/10.1016/S1352-
        2310(99)00395-7.
Massman. WJ. (2004). Toward an ozone standard to protect vegetation based on effective dose: A review of
        deposition resistances and a possible metric [Review]. Atmos Environ 38: 2323-2337.
Matvssek. R: Gunthardt-Goerg. MS: Maurer. S: Keller. T. (1995).  Nighttime exposure to ozone reduces whole-
        plant production in Betula pendula. Tree Physiol 15: 159-165.
Matvssek, R: Le Thiec, D:  Low, M: Dizengremel, P: Nunn, AJ: Haberle, KH. (2006). Interactions between
        drought and O3 stress in forest trees. Plant Biol (Stuttg)  8: 11-17. http://dx.doi.org/10.1055/s-2005-
        873025.
Matvssek, R: Sandermann, H: Wieser, G: Booker, F: Cieslik, S: Musselman, R: Ernst, D. (2008). The challenge
        of making ozone risk assessment for forest trees more mechanistic. Environ Pollut 156: 567-582.
        http://dx.doi.0rg/10.1016/i.envpol.2008.04.017.
Matvssek, R: Wieser, G: Ceulemans, R: Rennenberg, H: Pretzsch, H: Haberer, K: Low, M: Nunn, AJ: Werner, H:
        Wipfler. P: Obwald. W: Nikolova. P:  Hanke. DE: Kraigher. H: Tausz. M: Bahnweg. G: Kitao. M: Dieler. J:
        Sandermann. H: Herbinger. K: Grebenc. T: Blumenrother. M: Deckmvn.  G: Grams. TEE: Heerdt. C:
        Leuchner, M: Fabian, P:  Haberle,  KH. (2010). Enhanced ozone strongly reduces carbon sink strength
        of adult beech (Fagus sylvatica): Resume from the free-air fumigation study at Kranzberg Forest.
        Environ Pollut 158: 2527-2532. http://dx.doi.Org/10.1016/i.envpol.2010.05.009.
Mautz, WJ: Dohm, MR. (2004). Respiratory and behavioral effects of ozone on a lizard and a frog.  Comp
        Biochem Physiol A Mol Integr Physiol 139: 371-377.
McAinsh. MR: Evans. NH: Montgomery. LT: North. KA. (2002). Calcium signalling in stomatal responses to
        pollutants. NewPhytol 153: 441-447.
McBride. JR: Laven. RD. (1999).  Impact of oxidant air pollutants on forest succession in the mixed  conifer
        forests of the San Bernardino Mountains. In PR Miller; JR McBride (Eds.), Oxidant air pollution impacts
        in the montane forests of southern California: A case study of the San Bernardino Mountains (pp. 338-
        352). New York, NY: Springer-Verlag.
McCarthy. HR: Oren. R: Johnsen. KH: Gallet-Budvnek. A: Pritchard. SG: Cook. CW: LaDeau. SL: Jackson. RB:
        Finzi, AC. (2009). Re-assessment of plant carbon dynamics at the Duke free-air CO2 enrichment site:
        Interactions of atmospheric [CO2] with nitrogen and water availability over stand development. New
        Phytol 185: 514-528. http://dx.doi.Org/10.1111/i.1469-8137.2009.03078.x.
McCool, PM: Musselman,  RC: Younglove, T: Teso, RR.  (1988). Response of kidney bean to sequential ozone
        exposures. Environ Exp Bot 28: 307-313.
McFrederick. QS: Kathilankal. JC: Fuentes. JD. (2008).  Air pollution modifies floral scent trails. Atmos Environ
        42: 2336-2348. http://dx.doi.Org/10.1016/i.atmosenv.2007.12.033.
McFrederick. QS: Fuentes. JD: Roulston. T:  Kathilankal. JC: Lerdau. M. (2009). Effects of air pollution on
        biogenic volatiles and ecological interactions. Oecologia 160: 411-420.
        http://dx.doi. org/10.1007/s00442-009-1318-9.
McLaughlin. SB: Nosal. M: Wullschleger. SD: Sun. G. (2007a). Interactive effects of ozone and climate on tree
        growth and water use in a southern  Appalachian forest in the USA. New Phytol 174: 109-124.
        http://dx.doi.0rg/10.1111/i.1469-8137.2007.02018.x.
McLaughlin. SB: Wullschleger. SD:  Sun. G: Nosal. M. (2007b). Interactive effects of ozone and climate on water
        use, soil moisture content and streamflow in a southern Appalachian forest in the USA. New Phytol
        174: 125-136. http://dx.doi.Org/10.1111/i.1469-8137.2007.01970.x.
McLeod. AR: Long. SP (1999). Free-air carbon dioxide enrichment (FACE) in global change research: A review.
        Adv Ecol Res 28:  1-56. http://dx.doi.org/10.1016/50065-2504(08)60028-8.
Medlyn,  BE: Barton, CVM: Broadmeadow, MSJ: Ceulemans, R: DeAngelis, P: Forstreuter, M: Freeman, M:
        Jackson. SB: Kellomaki.  S:  Laitat. E: Rev. A: Roberntz. P: Sigurdsson. BD: Strassemever.  J: Wang. K:
        Curtis. PS: Jarvis. PG. (2001). Stomatal conductance of forest species after long-term exposure to
        elevated CO2 concentration: A synthesis. New Phytol 149: 247-264. http://dx.doi.Org/10.1046/j. 1469-
        8137.2001.00028.x.
Meehan. TD: Crosslev. MS: Lindroth. RL. (2010). Impacts of elevated CO2 and O3 on  aspen leaf litter chemistry
        and earthworm and springtail productivity. Soil Biol Biochem 42: 1132-1137.
        http://dx.doi.0rg/10.1016/i.soilbio.2010.03.019.
Draft - Do Not Cite or Quote                       9-168                                 September 2011

-------
Menendez. Al: Romero. AM: Folcia. AM:  Martinez-Ghersa. MA. (2009). Getting the interactions right: Will higher
        O3 levels interfere with induced  defenses to aphid feeding? Basic Appl Ecol 10: 255-264.
        http://dx.doi.0rg/10.1016/i.baae.2008.03.010.
Menendez. Al: Romero. AM: Folcia. AM:  Martinez-Ghersa. MA. (2010). Aphid and episodic O3 injury in arugula
        plants (Eruca sativa Mill) grown  in open-top field chambers. Agric Ecosyst Environ 135: 10-14.
        http://dx.doi.ora/10.1016/i.agee.2009.08.005.
Mereu. S: Gerosa. G: Finco. A: Fusaro. L: Muvs. B: Manes. F. (2009). Improved sapflow methodology reveals
        considerable night-time ozone uptake by Mediterranean species. Biogeosciences 6: 3151-3162.
Miles. GP: Samuel. MA: Zhang. YL: Ellis. BE. (2005). RNA interference-based (RNAi) suppression of AtMPK6,
        an Arabidopsis mitogen-activated protein kinase, results in hypersensitivity to ozone and misregulation
        of AtMPKS. Environ Pollut 138: 230-237. http://dx.doi.Org/10.1016/i.envpol.2005.04.017.
Miller. PL. (1973). Oxidant-induced community change in a mixed conifer forest. In JA Naegele (Ed.), Air
        pollution damage to vegetation (pp. 101-117). Washington, DC: American Chemical Society.
Miller, PR: Parmeter, JR, Jr: Taylor, OC: Cardiff, EA. (1963). Ozone injury to the foliage of Pinus ponderosa.
        Phytopathology 53: 1072-1076.
Miller. PR: McCutchan.  MH: Ryan. BC. (1972). Influence of climate and topography on oxidant air pollution
        concentrations that damage conifer forests in southern California. Mitt Forstl Bundesversuchsanst Wien
        97: 585-607.
Miller. PR: Elderman. MJ. (1977). Photochemical oxidant air pollutant effects on a mixed conifer forest
        ecosystem: A progress report, 1976. Corvallis, Oregon: U.S. Environmental Protection Agency.
Miller. PR: Rechel. J. (1999). Temporal changes in crown condition indices, needle litterfall, and collateral
        needle injuries of Ponderosa and Jeffrey pines.  In PR Miller; JR McBride (Eds.),  Oxidant air pollution
        impacts in the Montane forests of southern California: A case study of the San Bernardino Mountains
        (pp. 164-178). New York,  NY:  Springer.
Mills. G: Ball. G: Haves. F: Fuhrer. J: Skarbv. L: Gimeno. B: De Temmerman. L. (2000). Development of a multi-
        factor model for predicting the effects of ambient ozone on the biomass of white  clover. Environ Pollut
        109: 533-542. http://dx.doi.orq/10.1016/S0269-7491 (00)00057-9.
Mills. G. (2002). Modification of plant response by environmental conditions.  In JNB Bell;  M Treshow (Eds.), Air
        pollution and plant life (2nd ed.,  pp. 343-358). Chichester, United Kingdom: John Wiley & Sons.
Mills. G: Haves. F: Jones. MLM: Cinderbv. S. (2007a). Identifying ozone-sensitive communities of (semi-)natural
        vegetation suitable for mapping  exceedance of critical levels. Environ Pollut 146: 736-743.
        http://dx.doi.0rg/10.1016/i.envpol.2006.04.005.
Mills. G: Buse. A: Gimeno. B: Bermeio. V: Holland.  M: Emberson. L: Pleiiel. H. (2007b). A synthesis of AOT40-
        based response functions and critical levels of ozone for agricultural and horticultural crops. Atmos
        Environ 41: 2630-2643. http://dx.doi.Org/10.1016/i.atmosenv.2006.11.016.
Mills. G: Haves. F: Wilkinson. S: Davies. WJ. (2009). Chronic exposure to increasing background ozone impairs
        stomatal functioning in grassland species.  Global Change Biol 15: 1522-1533.
        http://dx.doi.0rg/10.1111/i.1365-2486.2008.01798.x.
Mondor. EB: Tremblav. MN: Awmack. CS: Lindroth. RL. (2004). Divergent pheromone-mediated insect
        behaviour under global atmospheric change. Global Change Biol 10: 1820-1824.
Mondor, EB; Tremblav, MN; Awmack, CS; Lindroth, RL. (2005). Altered genotypic and phenotypic frequencies of
        aphid populations under enriched CO2 and O3 atmospheres. Global Change Biol 11: 1990-1996.
        http://dx.doi.0rg/10.1111/i.1365-2486.2005.01054.x.
Mondor, EB; Awmack, CS; Lindroth, RL.  (2010).  Individual growth rates do not predict aphid  population
        densities under altered atmospheric conditions.  Agr Forest Entomol  12: 293-299.
        http://dx.doi.0rg/10.1111/i.1461-9563.2010.00478.x.
Morgan, PB; Ainsworth, EA; Long, SP (2003). How does elevated ozone impact soybean? A meta-analysis of
        photosynthesis, growth and yield. Plant Cell Environ 26: 1317-1328.
Morgan. PB: Bernacchi. CJ: Ort. PR: Long. SP.  (2004). An in vivo analysis of the  effect of season-long open-air
        elevation of ozone to anticipated 2050 levels on photosynthesis in soybean. J Plant Physiol 135: 2348-
        2357.
Draft - Do Not Cite or Quote                        9-169                                  September 2011

-------
Morgan. PB: Mies. TA: Bollero. GA: Nelson. RL: Long. SP. (2006). Season-long elevation of ozone
        concentration to projected 2050 levels under fully open-air conditions substantially decreases the
        growth and production of soybean. New Phytol 170: 333-343. http://dx.doi.orq/10.1111/1.1469-
        8137.2006.01679.x.
Morison, JIL: Lawlor, DW. (1999). Interactions between increasing CO2 concentration and temperature on plant
        growth. Plant Cell Environ 22: 659-682. http://dx.doi.orq/10.1046/1.1365-3040.1999.00443.x.
Morskv. SK: Haapala. JK: Rinnan. R: Tiiva. P: Saarnio. S: Silvola. J:  Holopainen. T: Martikainen. PJ. (2008).
        Long-term ozone effects on vegetation, microbial community and methane dynamics of boreal peatland
        microcosms in open-field conditions. Global Change Biol 14: 1891-1903.
        http://dx.doi.0rg/10.1111/i.1365-2486.2008.01615.x.
Mudd, JB. (1996). Biochemical basis for the toxicity of ozone. In M Yunus; M Iqbal (Eds.), Plant response to air
        pollution (pp. 267-283).  New York, NY: John Wiley & Sons.
Muntifering. RB: Chappelka. AH: Lin. JC: Karnoskv. DF: Somers. GL. (2006). Chemical  composition and
        digestibility of Trifolium exposed to elevated ozone and carbon dioxide in a free-air (FACE) fumigation
        system. Funct Ecol 20: 269-275. http://dx.doi.Org/10.1111/i. 1365-2435.2006.01093.x.
Musselman. RC:  McCool. PM: Younglove. T. (1988). Selecting ozone exposure statistics for determining crop
        yield loss from air pollutants. Environ Pollut 53: 63-78.  http://dx.doi.org/10.1016/0269-7491 (88)90025-5.
Musselman. RC:  Massman. WJ. (1999). Ozone flux to vegetation and its relationship to plant response and
        ambient air quality standards. Atmos Environ 33: 65-73.
Musselman, RC:  Minnick, TJ. (2000). Nocturnal stomatal conductance and ambient air quality standards for
        ozone. Atmos Environ 34: 719-733. http://dx.doi.orq/10.1016/S1352-2310(99)00355-6.
Musselman. RC:  Lefohn. AS: Massman. WJ: Heath.  RL. (2006). A critical review and analysis of the use of
        exposure- and flux-based ozone indices for predicting vegetation effects [Review]. Atmos Environ 40:
        1869-1888. http://dx.doi.orq/10.1016/i.atmosenv.20Q5.10.064.
Nali. C:  Balducci. E: Frati. L: Paoli. L: Loppi. S: Lorenzini. G. (2007).  Integrated biomonitoring of air quality with
        plants and lichens: A case study on ambient ozone from central Italy. Chemosphere 67: 2169-2176.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.12.036.
NAPAP (National Acid Precipitation Assessment Program). (1987). Diagnosing injury to Eastern forest trees: A
        manual for identifying damage caused by air pollution, pathogens, insects, and abiotic stresses. In.
        University Park,  PA: Pennsylvania State University.
Nash. TH. III. (2008). Lichen sensitivity to air pollution. In Lichen Biology (pp. 299-314). Cambridge, United
        Kingdom: Cambridge University Press.
Neufeld. HS: Renfro. JR: Hacker. WD: Silsbee. D. (1992). Ozone in Great Smoky Mountains National Park:
        Dynamics and effects on plants. In RL Berglund (Ed.), Tropospheric ozone and the environment II:
        Effects, modeling and control (pp.  594-617). Atlanta, GA: Air & Waste  Management Association.
Nikolova. PS: Andersen.  CP:  Blaschke. H:  Matvssek. R: Haberle. KH. (2010). Belowground  effects of enhanced
        tropospheric ozone and drought in a beech/spruce forest (Fagus sylvatica L./Picea abies [L.] Karst).
        Environ Pollut 158: 1071-1078. http://dx.doi.Org/10.1016/i.envpol.2009.07.036.
Noctor. G: Foyer. CH. (1998). Ascorbate and glutathione:  Keeping active oxygen under control.  49: 249-279.
Norbv. RJ: DeLucia. EH:  Gielen. B: Calfapietra. C:  Giardina. CP: King. JS: Ledford. J: McCarthy. HR: Moore.
        DJP: Ceulemans, R: De Angelis, P: Finzi, AC: Karnoskv, DF: Kubiske, ME: Lukac,  M: Pregitzer, KS:
        Scarascia-Mugnozza. GE: Schlesinger. WH: Oren. R. (2005). Forest response to elevated CO2 is
        conserved across a broad range of productivity. PNAS 102:  18052-18056.
        http://dx.doi. org/10.1073/pnas.0509478102.
Novak. K: Cherubini. P: Saurer.  M: Fuhrer.  J: Skellv. JM: Krauchi. N:  Schaub.  M.  (2007). Ozone air pollution
        effects on tree-ring growth, delta(13)C, visible foliar injury and leaf gas exchange in three ozone-
        sensitive woody plant species. Tree Physiol 27: 941-949.
NPS. (U.S. National Park Service). (2006). Ozone bioindicators. Washington,  DC.
        http://www.nature.nps.gov/air/Pubs/bioindicators/index.cfm.
NPS. (U.S. National Park Service). (2007). Ozone effects studies. Washington, DC: U.S. Department of the
        Interior, National Park Service. http://www.nature.nps.gov/air/studies/ecoOzone.cfm.
Nussbaum. S:  Geissmann. M: Fuhrer. J. (1995). Ozone exposure-response relationships for mixtures of
        perennial ryegrass and white clover depend on ozone exposure patterns. Atmos Environ 29: 989-995.
        http://dx.doi. org/10.1016/1352-2310(94)00368-U.
Draft - Do Not Cite or Quote                        9-170                                 September 2011

-------
O'Gara. PJ. (1922). Sulfur dioxide and fume problems and their solution. J Ind Eng Chem 14: 744-745.
O'Neill, BF: Zangerl, AR: Delucia, EH:  Berenbaum, MR. (2008). Longevity and fecundity of Japanese beetle
        (Popillia japonica) on foliage grown under elevated carbon dioxide. Environ Entomol 37: 601-607.
O'Neill. BF: Zanqerl. AR: Dermodv. O:  Bilqin. DP: Casteel. CL:  Zavala. JA: DeLucia. EH:  Berenbaum.  MR.
        (2010). Impact of elevated levels of atmospheric CO2  and herbivory on flavonoids of soybean (Glycine
        max Linnaeus). J Chem Ecol 36: 35-45. http://dx.doi.orq/10.1007/s10886-009-9727-0.
Ogawa. D: Nakaiima. N: Sano. T: Tamaoki. M: Aono. M:  Kubo.  A: Kanna. M: loki. M: Kamada. H: Sail. H.
        (2005). Salicylic acid accumulation under O-3 exposure is regulated by ethylene in tobacco plants.
        Plant Cell Physiol 46: 1062-1072. http://dx.doi.orq/10.1093/pcp/pci118.
Oksanen. E: Holopainen. T. (2001). Responses of two birch (Betula pendula Roth) clones to different ozone
        profiles with similar AOT40 exposure. Atmos Environ 35: 5245-5254.  http://dx.doi.Org/10.1016/S1352-
        2310(01)00346-6.
Olbrich. M: Betz. G: Gerstner. E: Langebartels. C: Sandermann. H: Ernst. D. (2005). Transcriptome analysis of
        ozone-responsive genes in leaves of European  beech  (Fagus sylvatica L.). Plant Biol (Stuttg) 7: 670-
        676. http://dx.doi.ora/10.1055/s-2005-873001.
Olbrich. M: Gerstner. E: Welzl. G: Winkler. JB: Ernst. D. (2009). Transcript responses in leaves of ozone-treated
        beech saplings seasons at an outdoor free  air model fumigation site over two growing seasons. Plant
        Soil 323: 61-74. http://dx.doi.orq/10.1007/S11104-009-0129-4.
Olbrich. M: Gerstner. E: Bahnweg. G: Haberle. KH: Matvssek. R: Welzl. G: Heller. W: Ernst.  D. (2010).
        Transcriptional signatures in leaves of adult European  beech trees (Fagus sylvatica L.) in an
        experimentally enhanced free air ozone setting. Environ Pollut 158: 977-982.
        http://dx.doi.0rg/10.1016/i.envpol.2009.08.001.
Ollinger, SV: Aber, JD: Reich, PB. (1997a). Simulating ozone effects on forest productivity: Interactions among
        leaf- and stand-level processes. Ecol Appl 123: 351-358.
Ollinger. SV: Aber. JD: Reich. PB. (1997b). Simulating ozone effects on forest productivity: Interactions among
        leaf-, canopy-, and stand-level processes. Ecol Appl 7: 1237-1251.
Ollinger. SV: Aber. JD: Reich. PB: Freuder. RJ. (2002). Interactive effects of nitrogen deposition, tropospheric
        ozone, elevated CO2 and land use history on the carbon dynamics of northern hardwood forests.
        Global Change Biol 8: 545-562. http://dx.doi.org/10.1046/1.1365-2486.2002.00482.X.
Olszvk. DM: Kats.  G: Dawson. PJ: Bvtnerowicz. A: Wolf. J: Thompson. CR. (1986). Characteristics of air
        exclusion systems vs chambers for field air pollution studies.  J Environ  Qual 15:  326-334.
Omasa, K: Shimazaki, Kl: Aiga, I: Larcher, W: Onoe, M. (1987). Image analysis  of chlorophyll fluorescence
        transients for diagnosing the photosynthetic system of attached leaves. Plant Physiol 84: 748-752.
        http://dx.doi.0rg/10.1104/pp.84.3.748.
Orendovici-Best, T: Skellv, JM: Davis,  DP: Ferdinand, JA: Savage, JE: Stevenson,  RE. (2008). Ozone uptake
        (flux) as it relates to ozone-induced foliar symptoms of Prunus serotina and Populus maximowizii x
        trichocarpa. Environ Pollut 151: 79-92. http://dx.doi.Org/10.1016/i.envpol.2007.03.003.
Orendovici, T: Skellv, JM: Ferdinand, JA: Savage, JE: Sanz, M, -J: Smith, GC. (2003). Response of native
        plants of northeastern United  States and southern Spain to ozone exposures; determining
        exposure/response relationships. Environ Pollut 125: 31-40.
Oshima, RJ: Poe,  MP: Braegelmann, PK: Baldwin, DW: Van Way, V. (1976). Ozone dosage-crop loss  function
        for alfalfa: A standardized method for assessing crop losses from air pollutants. J Air Pollut Control
        Assoc26: 861-865.
Oshima, RJ: Braegelmann, PK: Baldwin, DW: Van Way, V: Taylor, OC. (1977). Reduction of tomato fruit size
        and yield by ozone. J Am Soc Hortic Sci 102: 289-293.
Overmver. K: Tuominen. H: Kettunen.  R: Betz. C: Langebartels. C: Sandermann. H. Jr: Kangasiarvi. J. (2000).
        Ozone-sensitive Arabidopsis rcdl mutant reveals opposite roles for ethylene and jasmonate signaling
        pathways in regulating superoxide-dependent cell death. Plant Cell 12:  1849-1862.
        http://dx.doi. org/10.1105/tpc. 12.10.1849.
Overmver, K: Brosche, M: Pellinen, R:  Kuittinen, T: Tuominen,  H: Ahlfors, R: Keinanen, M: Saarma, M: Scheel,
        D: Kangasiarvi. J. (2005). Ozone-induced programmed cell death in the Arabidopsis radical-induced
        cell deathl mutant. Plant Physiol 137: 1092-1104. http://dx.doi.org/10.1104/pp. 104.055681.
Draft - Do Not Cite or Quote                        9-171                                  September 2011

-------
Overmver. K: Kollist. H: Tuominen. H: Betz. C: Langebartels. C: Wingsle. G: Kangasiarvi. S: Brader. G:
        Mullineaux, P: Kangasjarvi, J. (2008). Complex phenotypic profiles leading to ozone sensitivity in
        Arabidopsisthaliana mutants. Plant Cell Environ 31: 1237-1249. http://dx.doi.Org/10.1111/i. 1365-
        3040.2008.01837.x.
Pan, YD: Birdsev, R: Horn, J: McCullough, K. (2009). Separating effects of changes in atmospheric composition,
        climate and land-use on carbon sequestration of US Mid-Atlantic temperate forests. For Ecol Manage
        259: 151-164. http://dx.doi.Org/10.1016/i.foreco.2009.09.049.
Panek, J: Kurpius, MR:  Goldstein, AH. (2002). An evaluation of ozone exposure metrics for a seasonally
        drought-stressed ponderosa pine ecosystem.  Environ Pollut 117: 93-100.
        http://dx.doi. org/10.1016/S0269-7491 (01 )00155-5.
Panek, JA: Goldstein, AH. (2001). Responses of stomatal conductance to drought in ponderosa pine:
        Implications for carbon and ozone uptake. Tree Physiol 21: 337-344.
Panek. JA. (2004). Ozone uptake, water loss and carbon exchange dynamics in annually drought-stressed
        Pinus ponderosa forests: Measured trends and parameters for uptake modeling. Tree Physiol 24: 277-
        290.
Paolacci. AR: Miraldi. C: Tanzarella. OA: Badiani. M: Porceddu. E: Mali. C:  Lorenzini. G: Ciaffi.  M. (2007). Gene
        expression induced by chronic ozone in the Mediterranean shrub Phillyrea latifolia: Analysis by cDNA-
        AFLP. Tree Physiol 27: 1541-1550. http://dx.doi.org/10.1093/treephvs/27.11.1541.
Paoletti. E: Grulke. NE. (2005). Does living in elevated CO2 ameliorate tree response to ozone? A review on
        stomatal responses [Review]. Environ Pollut 137: 483-493.
        http://dx.doi.0rg/10.1016/i.envpol.2005.01.035.
Paoletti. E: Seufert. G: Delia Rocca. G: Thomsen. H. (2007). Photosynthetic responses to elevated CO2 and  O-
        3 in Quercus ilex leaves at a natural CO2 spring. Environ Pollut 147: 516-524.
        http://dx.doi.0rg/10.1016/i.envpol.2006.08.039.
Paoletti. E: Manning. WJ.  (2007). Toward a biologically significant and usable standard for ozone that will also
        protect plants. Environ Pollut 150: 85-95. http://dx.doi.Org/10.1016/i.envpol.2007.06.037.
Paoletti. E: Grulke. NE. (2010). Ozone exposure and stomatal  sluggishness in different plant physiognomic
        classes. Environ Pollut 158:  2664-2671. http://dx.doi.Org/10.1016/i.envpol.2010.04.024.
Parsons, WFJ: Bockheim, JG: Lindroth, RL. (2008). Independent, interactive, and species-specific responses of
        leaf litter decomposition to elevated CO2 and  O3 in a  northern hardwood forest. Ecosystems 11: 505-
        519.
Pearson, M:  Mansfield,  TA. (1993). Interacting effects of ozone and water stress on the stomatal resistance of
        beech (Fagus sylvatica L). New Phytol  123: 351-358.  http://dx.doi.Org/10.1111/i. 1469-
        8137.1993.tb03745.x.
Pearson, S:  Davison, AW: Reiling, K: Ashenden, T: Ollerenshaw, JH.  (1996). The effects of different ozone
        exposures on three contrasting populations of Plantago  major. New Phytol 132: 493-502.
        http://dx.doi.0rg/10.1111/i.1469-8137.1996.tb01869.x.
Pel, Z, -M: Murata, Y: Benning, G: Thomine, S: Klusener, B: Allen, GJ: Grill, E: Schroeder, Jl. (2000). Calcium
        channels activated by hydrogen peroxide mediate  abscisic acid signalling in guard cells. Nature 406:
        731-734.  http://dx.doi.org/10.1038/35021067.
Pell, EJ: Sinn, JP: Brendley, BW: Samuelson, L: Vinten-Johansen, C: Tien, M: Skillman, J. (1999). Differential
        response of four tree species to ozone-induced acceleration of foliar senescence. Plant Cell Environ
        22:  779-790. http://dx.doi.org/10.1046/i. 1365-3040.1999.00449.x.
Peltonen, PA: Julkunen-Tiitto, R: Vapaavuori, E: Holopainen, JK. (2006). Effects of elevated carbon dioxide and
        ozone on aphid oviposition preference and birch bud exudate phenolics. Global Change Biol 12: 1670-
        1679. http://dx.doi.0rg/10.1111/i.1365-2486.2006.01226.x.
Peltonen, PA: Vapaavuori, E: Heinonen, J: Julkunen-Tiitto,  R: Holopainen, JK. (2010). Do elevated atmospheric
        CO2 and O-3 affect food quality and performance of folivorous insects on silver birch? Global Change
        Biol 16: 918-935. http://dx.doi.Org/10.1111/i.1365-2486.2009.02073.x.
Percy,  KE: Nosal,  M: Heilman, W: Dann, T: Sober, J: Legge, AH: Karnoskv,  DF. (2007). New exposure-based
        metric approach for evaluating O3 risk to North American aspen forests. Environ Pollut 147: 554-566.
        http://dx.doi. org/10.1016/i.envpol.20Q6.10.009.
Draft - Do Not Cite or Quote                        9-172                                  September 2011

-------
Peterson. PL: Arbaugh. MJ: Wakefield. VA: Miller. PR. (1987). Evidence of growth reduction in ozone-injured
        Jeffrey pine (Pinus jeffreyi Grev and Balf) in Sequoia and Kings Canyon National Parks. J Air Waste
        Manag Assoc 37: 906-912.
Pfleeger. TG: Plocher. M: Bichel. P. (2010). Response of pioneer plant communities to elevated ozone
        exposure. Agric Ecosyst Environ 138: 116-126.  http://dx.doi.Org/10.1016/i.agee.2010.04.009.
Phillips. PL: Johnson. MG:  Tingev. DT: Storm. MJ. (2009). Elevated CO2 and O3 effects on fine-root
        survivorship in ponderosa pine mesocosms. Oecologia 160: 827-837. http://dx.doi.org/10.1007/s00442-
        009-1339-4.
Phillips. RL: Zak. PR: Holmes. WE: White. DC. (2002). Microbial community composition and function beneath
        temperate trees exposed to elevated atmospheric carbon dioxide and ozone. Oecologia 131: 236-244.
Piikki, K: Vorne, V: Ojanpera, K: Pleijel, H. (2007). Impact of elevated O-3 and CO2 exposure on potato
        (Solanum tuberosum L. cv. Bintje) tuber macronutrients (N, P, K, Mg, Ca). Agric Ecosyst Environ 118:
        55-64. http://dx.doi.0rg/10.1016/i.agee.2006.04.012.
Piikki, K: De Temmerman, L: Ojanpera, K: Danielsson, H: Pleijel, H. (2008a). The grain quality of spring wheat
        (Triticum aestivum L.) in relation to elevated ozone uptake and carbon dioxide exposure. Eur J Agron
        28: 245-254. http://dx.doi.Org/10.1016/i.eia.2007.07.004.
Piikki, K: De Temmerman, L: Hogv, P:  Pleijel, H. (2008b). The open-top chamber impact on vapour pressure
        deficit and its consequences for stomatal ozone uptake. Atmos Environ 42: 6513-6522.
        http://dx.doi.0rg/10.1016/i.atmosenv.2008.04.014.
Pinto, DM: Blande, JD: Nykanen,  R: Dong, WX: Nerg, AM: Holopainen, JK. (2007a). Ozone degrades common
        herbivore-induced plant volatiles: Does this affect herbivore prey location by predators and parasitoids?
        J Chem Ecol 33: 683-694. http://dx.doi.org/10.1007/s10886-007-9255-8.
Pinto, DM: Nerg, AM: Holopainen, JK.  (2007b). The role of ozone-reactive compounds, terpenes,  and green leaf
        volatiles (GLVs), in the orientation of Cotesia plutellae. J Chem Ecol 33: 2218-2228.
        http://dx.doi. org/10.1007/s10886-007-9376-0.
Pinto, DM: Himanen, SJ: Nissinen, A: Nerg, AM: Holopainen, JK. (2008).  Host location behavior of Cotesia
        plutellae Kurdjumov (Hymenoptera:  Braconidae) in ambient and  moderately  elevated ozone in field
        conditions. Environ Pollut 156: 227-231. http://dx.doi.Org/10.1016/i.envpol.2007.12.009.
Pinto, DM: Blande, JD: Souza, SR: Nerg, AM: Holopainen, JK. (2010). Plant volatile organic compounds (VOCs)
        in ozone (O-3) polluted atmospheres: The ecological effects. J Chem  Ecol 36: 22-34.
        http://dx.doi. org/10.1007/s10886-009-9732-3.
Pinto, J. (2009). Wyoming winter smog. Nat Geosci 2: 88-90. http://dx.doi.org/10.1038/ngeo430.
Pleiiel. H: Oianpera.  K: Mortensen. L. (1997). Effects of tropospheric ozone on the yield and grain protein
        content of spring wheat (Triticum aestivum  L) in the nordic countries. Acta Agric Scand B Soil Plant Sci
        47: 20-25. http://dx.doi.org/10.1080/09064719709362434.
Pleiiel. H: Danielsson. H: Gelang. J: Slid. E: Sellden. G.  (1998).  Growth stage dependence of the grain yield
        response to ozone in spring wheat (Triticum aestivum L). Agric Ecosyst Environ 70: 61-68.
        http://dx.doi.org/10.1016/S0167-8809(97)00167-9.
Pleiiel. H: Danielsson. H: Oianpera. K: De Temmerman.  L: Hogv. P: Badiani. M: Karlsson. PE. (2004a).
        Relationships between ozone exposure and yield loss in European wheat and potato—a comparison of
        concentration- and flux-based exposure indices. Atmos Environ 38: 2259-2269.
Pleiiel. H: h.  D: Oianpera. K: De Temmerman. L: Hogv. P. (2004b). Relationships between ozone exposure and
        yield loss in wheat and potato - Suggestions of critical levels for ozone effects on crops. Atmos Environ
        38: 2259-2269. http://dx.doi.Org/10.1016/i.atmosenv.2003.09.076.
Plessl. M: Elstner. EF: Rennenberg. H: Habermever. J: Heiser.  I. (2007). Influence of elevated CO2 and ozone
        concentrations on  late blight resistance and growth of potato plants. Environ Exp Bot 60: 447-457.
        http://dx.doi.0rg/10.1016/i.envexpbot.2007.01.003.
Plochl. M: Lyons. T:  Ollerenshaw. J: Barnes.  J. (2000). Simulating ozone detoxification in the leaf apoplast
        through the  direct reaction with ascorbate. Planta 210: 454-467.  http://dx.doi.org/10.1007/PL00008153.
Pollastrini, M: Desotgiu, R:  Cascio, C:  Bussotti, F: Cherubini, P:  Saurer, M: Gerosa, G: Marzuoli, R. (2010).
        Growth and physiological responses to ozone and mild drought stress of tree species with different
        ecological requirements. Trees Struct Funct 24: 695-704. http://dx.doi.org/10.1007/s00468-010-0439-4.
Polle, A: Pell, EJ. (1999). Role of carbon dioxide in modifying the plant response to ozone. In Y Luo; HA Mooney
        (Eds.), Carbon dioxide and environmental stress (pp. 193-213). San Diego, CA: Academic Press.
Draft - Do Not Cite or Quote                        9-173                                  September 2011

-------
Pregitzer. K: Lova. W: Kubiske. M: Zak. D. (2006). Soil respiration in northern forests exposed to elevated
        atmospheric carbon dioxide and ozone. Oecologia 148: 503-516. http://dx.doi.Org/10.1007/s00442-006-
        0381-8.
Pregitzer. KS: Burton. AJ: King. JS: Zak. DR. (2008). Soil respiration, root biomass, and root turnover following
        long-term exposure of northern forests to elevated atmospheric Co-2 and tropospheric O-3. New Phytol
        180: 153-161. http://dx.doi.Org/10.1111/i.1469-8137.2008.02564.x.
Pretzsch. H: Dieler. J: Matvssek. R: Wipfler. P. (2010). Tree and stand growth of mature Norway spruce and
        European beech under long-term ozone fumigation. Environ Pollut 158: 1061-1070.
        http://dx.doi.0rg/10.1016/i.envpol.2009.07.035.
Pritsch. K: Esperschuetz. J: Haesler. F: Raidl. S: Winkler. B: Schloter. M. (2009). Structure and activities of
        ectomycorrhizal and microbial communities in the rhizosphere of Fagus sylvatica under ozone and
        pathogen stress in a lysimeter study. Plant Soil 323: 97-109. http://dx.doi.org/10.1007/s11104-009-
        9972-6.
Puckette, MC: Tang, YH: Mahalingam, R. (2008). Transcriptomic changes induced by acute ozone in resistant
        and sensitive Medicago truncatula accessions. BMC Plant Biol 8: 46. http://dx.doi.org/10.1186/1471-
        2229-8-46.
Pujol Pereira, El: Chung, H: Scow, K: Sadowskv,  MJ: van Kessel, C: Six, J. (2011). Soil nitrogen transformations
        under elevated atmospheric CO2 and O3 during the soybean growing season. Environ Pollut 159: 401-
        407. http://dx.doi.0rg/10.1016/i.envpol.2010.10.033.
Ramo, K: Kanerva, T: Ojanpera, K: Manninen, S. (2007). Growth onset, senescence, and reproductive
        development of meadow species in mesocosms exposed to elevated O3 and CO2. Environ Pollut 145:
        850-860. http://dx.doi.0rg/10.1016/i.envpol.2006.03.054.
Rao, MV: Hale, BA: Ormrod, DP. (1995). Amelioration of ozone-induced oxidative damage in wheat plants
        grown under high carbon dioxide: Role of antioxidant enzymes. J Plant Physiol 109: 421-432.
Rapport. DJ: Whitford. WG. (1999). How ecosystems respond to stress: Common  properties of arid and  aquatic
        systems. Bioscience 49: 193-203.
Rawlings. JO: Cure. WW. (1985). The Weibull function as a dose-response model to describe ozone effects on
        crop yields. Crop Sci 25: 807-814.
Reich, PB: Lassoie, JP  (1984). Effects of low level O3 exposure on leaf diffusive conductance and water-use
        efficiency in hybrid poplar. Plant Cell Environ  7:  661-668. http://dx.doi.org/10.1111/1365-
        304Q.ep11571645.
Reich, PB. (1987). Quantifying plant response to ozone: A unifying theory. Tree Physiol 3: 63-91.
        http://dx.doi. org/10.1093/treephvs/3.1.63.
Reid. CD: Fiscus. EL. (2008). Ozone and density affect the response of biomass and seed yield to elevated
        CO2 in rice. Global Change Biol 14: 60-76. http://dx.doi.Org/10.1111/i.1365-2486.2007.01472.x.
Reiling. K: Davison. AW. (1992). Effects of a short ozone exposure given at different stages in the development
        of Plantago major L. New Phytol 121: 643-647.  http://dx.doi.Org/10.1111/i.1469-8137.1992.tb01135.x.
Reiling, K: Davison, AW. (1994). Effects of exposure to ozone at different stages in the development of Plantago
        major L on chlorophyll fluorescence and gas exchange. New Phytol 128: 509-514.
        http://dx.doi.0rg/10.1111/i.1469-8137.1994.tb02998.x.
Reinert,  RA: Ho, MC. (1995). Vegetative growth of soybean as affected by elevated carbon dioxide and ozone.
        Environ Pollut 89: 89-96. http://dx.doi.org/10.1016/0269-7491 (94)00039-G.
Reinert.  RA: Eason. G:  Barton. J. (1997). Growth and fruiting of tomato as influenced by elevated carbon
        dioxide and ozone. New Phytol 137: 411-420. http://dx.doi.org/10.1046/j. 1469-8137.1997.00846.x.
Ren. W: Tian. HQ: Liu. ML: Zhang. C: Chen. GS: Pan. SF: Felzer. B: Xu. XF. (2007a). Effects of tropospheric
        ozone pollution on net primary productivity and carbon storage in terrestrial ecosystems of China. J
        Geophys Res 112: D22S09. http://dx.doi.org/10.1029/2007id008521.
Ren. W: Tian. H: Chen.  G: Liu. M: Zhang. C: Chappelka. AH: Pan. S. (2007b). Influence of ozone pollution and
        climate variability on net primary productivity and carbon storage in China's grassland ecosystems from
        1961 to 2000.  Environ Pollut 149: 327-335. http://dx.doi.Org/10.1016/i.envpol.2007.05.029.
Ren. W: Tian. H: Tao. B: Chappelka. A: Sun. G: Lu. C: Liu. M: Chen. G: Xu. X. (2011). Impacts of tropospheric
        ozone and climate change on net primary productivity and net carbon exchange of China's forest
        ecosystems. Global Ecology and Biogeography 20: 391-406. http://dx.doi.org/10.1111/i.1466-
        8238.2010.00606.x.
Draft - Do Not Cite or Quote                        9-174                                 September 2011

-------
Rhea. L: King. J: Kubiske. M: Saliendra. N: Teclaw. R. (2010). Effects of elevated atmospheric CO2 and
       tropospheric O3 on tree branch growth and implications for hydrologic budgeting. Environ Pollut 158:
       1079-1087. http://dx.doi.0rg/10.1016/i.envpol.2009.08.038.
Riddell. J: Padgett. PE: Nash. TH. III. (2010). Responses of the lichen Ramalina menziesii Tayl. to ozone
       fumigations. In TH Nash, III (Ed.), Biology of lichens: Symbiosis, ecology, environmental monitoring,
       systematics, cyber applications (Vol. 105, pp. 113-123). Stuttgart: J. Cramer in der Gebruder
       Borntraeger Verlagsbuchhandlung.
Riikonen, J: Kets, K: Darbah, J: Oksanen, E: Sober, A: Vapaavuori, E: Kubiske,  ME: Nelson, N: Karnoskv, PR
       (2008). Carbon gain  and bud physiology in  Populus tremuloides and Betula papyrifera grown under
       long-term exposure to elevated concentrations of CO2 and O3. Tree Physiol 28: 243-254.
       http://dx.doi. org/10.1093/treephys/28.2.243.
Riikonen. J: Maenpaa. M: Alavillamo. M: Silfver. T: Oksanen. E. (2009).  Interactive effect of elevated
       temperature and O3 on antioxidant capacity and gas exchange in Betula  pendula saplings. Planta 230:
       419-427. http://dx.doi.org/10.1007/s00425-009-0957-8.
Rizzo. M: Bernard!. R: Salvini. M: Mali. C: Lorenzini. G: Durante. M. (2007). Identification of differentially
       expressed genes induced by ozone stress in sensitive and tolerant poplar hybrids. J Plant Physiol 164:
       945-949. http://dx.doi.0rg/10.1016/i.iplph.2006.07.012.
Rodenkirchen.  H: Gottlein. A: Kozovits. AR:  Matvssek. R: Grams. TEE. (2009). Nutrient contents and
       efficiencies of beech and spruce saplings as influenced by competition  and O3/CO2 regime. European
       Journal of Forest Research 128: 117-128. http://dx.doi.org/10.1007/s10342-008-0221-y.
Rogers. A: Allen. DJ: Davev. PA: Morgan. PB: Ainsworth. EA: Bernacchi. CJ: Cornic. G: Dermodv. OC:
       Dohleman. FG: Heaton. EA: Mahonev. J: Zhu. X. -G: Delucia. EH: Ort.  PR: Long. SP (2004). Leaf
       photosynthesis and carbohydrate dynamics of soybean grown throughout their life-cycle under free-air
       carbon dioxide enrichment. Plant Cell Environ 27: 449-458.
Rowland-Bamford. AJ. (2000). Plant responses to changing carbon dioxide and temperature. In SN Singh (Ed.),
       Trace gas emissions and plants (pp. 63-74). Dordecht, The Netherlands:  Kluwer Academic Publishers.
Ryan. A: Coiocariu. C: Possell.  M: Davies. WJ: Hewitt. CN. (2009). Defining hybrid poplar (Populus deltoides x
       Populus trichocarpa) tolerance to ozone: Identifying key parameters. Plant Cell Environ 32: 31-45.
       http://dx.doi.0rg/10.1111/i.1365-3040.2008.01897.x.
Rvang. SZ: Woo. SY: Kwon. SY: Kim.  SH: Lee. SH: Kim. KN: Lee. DK. (2009). Changes of net photosynthesis,
       antioxidant enzyme activities, and antioxidant contents of Liriodendron  tulipifera under elevated ozone.
       Photosynthetica 47:  19-25.  http://dx.doi.org/10.1007/s11099-009-0005-8.
Samuel. MA: Miles. GP: Ellis. BE. (2000). Ozone treatment rapidly activates MAP kinase signalling in plants.
       Plant J 22: 367-376.  http://dx.doi.Org/10.1046/i. 1365-313x.2000.00741 .x.
Samuel, MA: Ellis, BE. (2002). Double jeopardy: Both overexpression and suppression of a redox-activated
       plant mitogen-activated protein kinase render tobacco plants ozone sensitive. Plant Cell 14: 2059-
       2069. http://dx.doi.org/10.1105/tpc.OQ2337.
Samuel, MA: Walia, A: Mansfield, SD:  Ellis,  BE. (2005). Overexpression of SIPK in tobacco enhances ozone-
       induced ethylene formation and blocks ozone-induced SA accumulation. J Exp Bot 56: 2195-2201.
       http://dx.doi. org/10.1093/ixb/eri219.
Samuelson, LJ: Kelly, JM. (1997). Ozone uptake in Prunus serotina, Acer rubrum and Quercus rubra forest
       trees of different sizes. New Phytol 136: 255-264. http://dx.doi.org/10.1046/i. 1469-8137.1997.00734.x.
Sanchez. MJS: Pena. GS: Lorente. VC: Gallego. TM: Albert. JC. (2001).  La contaminacion atmosferica en los
       bosques: Gufa para la  identificacfon de danos visibles causados por Ozono. In (Vol. 6). Madrid, Spain:
       Ministerio de Medio Ambiente.
Sandermann. H. (2008).  Ecotoxicology of ozone: Bioactivation  of extracellular ascorbate. Biochem Biophys  Res
       Commun 366: 271-274. http://dx.doi.org/10.1016/j.bbrc.2007.12.018.
Sanz. J: Muntifering. RB: Bermeio. V: Gimeno. BS: Elvira. S. (2005). Ozone and increased nitrogen supply
       effects on the yield and nutritive quality of Trifolium subterraneum. Atmos Environ 39: 5899-5907.
       http://dx.doi.0rg/10.1016/i.atmosenv.2005.06.022.
Sanz. J: Bermeio. V: Gimeno. BS: Elvira. S: Alonso. R. (2007).  Ozone sensitivity of the Mediterranean terophyte
       trifolium striatum is modulated by soil nitrogen  content. Atmos Environ 41: 8952-8962.
       http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.016.
Draft - Do Not Cite or Quote                        9-175                                 September 2011

-------
Sarkar. A: Rakwal. R: Aqrawal. SB: Shibato. J: Ogawa. Y: Yoshida. Y: Aqrawal. GK: Aqrawal. M. (2010).
        Investigating the impact of elevated levels of ozone on tropical wheat using integrated phenotypical,
        physiological, biochemical, and proteomics approaches. J Proteome Res 9: 4565-4584.
        http://dx.doi. orq/10.1021/PM 002824.
Saviranta, NMM: Julkunen-Tiitto, R: Oksanen, E: Karjalainen, RO. (2010). Leaf phenolic compounds in red
        clover (Trifolium pratense L.) induced by exposure to moderately elevated ozone. Environ Pollut 158:
        440-446. http://dx.doi.0rg/10.1016/i.envpol.2009.08.029.
Sawada, H: Kohno, Y. (2009). Differential ozone sensitivity of rice cultivars as indicated by visible injury and
        grain yield. Plant Biol (Stuttg) 11: 70-75. http://dx.doi.Org/10.1111/i. 1438-8677.2009.00233.x.
Scebba. F: Giuntini. D: Castagna. A: Soldatini. G: Ranieri. A. (2006). Analysing the impact of ozone on
        biochemical and physiological variables in plant species belonging to natural ecosystems. Environ Exp
        Bot 57: 89-97. http://dx.doi.Org/10.1016/i.envexpbot.2005.04.005.
Schaub. M: Skellv. JM: Zhang. JW: Ferdinand. JA: Savage. JE:  Stevenson. RE: Davis.  DP: Steiner. KG. (2005).
        Physiological and foliar symptom response in the crowns of Prunus serotina, Fraxinus americana and
        Acer rubrum canopy trees to ambient ozone under forest conditions. Environ Pollut 133: 553-567.
        http://dx.doi.0rg/10.1016/i.envpol.2004.06.012.
Schraudner,  M: Moeder, W: Wiese, C: Van Camp, W: Inze,  D: Langebartels, C: Sandermann,  H, Jr. (1998).
        Ozone-induced oxidative burst in the ozone biomonitor plant, tobacco Bel W3.  Plant J 16: 235-245.
        http://dx.doi. orq/10.1046/1.1365-313x. 1998.00294.x.
Severino, JF: Stich, K: Soja, G. (2007). Ozone stress and antioxidant substances in Trifolium repens and
        Centaureajacea leaves. Environ Pollut 146: 707-714. http://dx.doi.Org/10.1016/i.envpol.2006.04.006.
Sharkev. TD: Wiberlev. AE: Donohue. AR. (2008). Isoprene emission from plants: Why and how. Ann Bot 101: 5-
        18. http://dx.doi.0rg/10.1093/aob/mcm240.
Singh. E: Tiwari. S: Agrawal. M. (2009). Effects of elevated ozone on photosynthesis and stomatal conductance
        of two soybean varieties: A case study to assess impacts of one component of predicted global climate
        change. Plant Biol (Stuttg) 11: 101-108. http://dx.doi.org/10.1111/1.1438-8677.2009.00263.x.
Singh. E: Tiwari. S: Agrawal. M. (201 Oa). Variability in antioxidant and metabolite levels, growth and yield of two
        soybean varieties: An assessment of anticipated yield losses under projected elevation of ozone. Agric
        Ecosyst Environ 135: 168-177. http://dx.doi.Org/10.1016/i.agee.2009.09.004.
Sitch. S: Cox. PM: Collins. WJ: Huntingford. C. (2007). Indirect radiative forcing of climate change through
        ozone effects on the land-carbon sink. Nature  448: 791-794. http://dx.doi.org/10.1038/nature06059.
Skarbv, L: Troeng, E:  Bostrom, C, -A. (1987). Ozone uptake and effects on transpiration, net photosynthesis,
        and  dark respiration in Scots pine. Forest  Sci 33: 801-808.
Skarbv. L: Ottosson. S: Karlsson. PE: Wallina. G: Sellden. G: Medina. EL: Pleiiel. H. (2004). Growth of Norway
        spruce (Picea abies) in relation to different ozone exposure indices: a synthesis. Atmos Environ 38:
        2225-2236.
Smith. G: Coulston. J: Jepsen. E: Prichard. T. (2003). A national ozone biomonitoring program: Results from
        field surveys of ozone sensitive plants in northeastern forests (1994-2000). Environ Monit Assess 87:
        271-291.
Soia. G: Barnes. JD: Posch. M: Vandermeiren. K: Pleiiel. H: Mills. G. (2000). Phenological weighting of ozone
        exposures in the calculation of critical levels for wheat,  bean and plantain. Environ Pollut 109: 517-524.
        http://dx.doi. org/10.1016/S0269-7491(00)00055-5.
Soia. G: Reichenauer. TG: Eid. M: Soia. A. -M: Schaber. R: Gangl. H. (2004). Long-term ozone exposure and
        ozone uptake of grapevines in open-top chambers. Atmos Environ 38: 2313-2321.
Somers. GL: Chappelka. AH: Rosseau. P: Renfro.  JR.  (1998). Empirical evidence of growth decline related to
        visible ozone injury. For Ecol Manage 104: 129-137. http://dx.doi.org/10.1016/50378-1127(97)00252-1.
Souza, L: Neufeld, HS: Chappelka, AH: Burkey, KO:  Davison, AW. (2006). Seasonal development of ozone-
        induced foliar injury on tall milkweed (Asclepias exaltata) in Great Smoky Mountains National Park.
        Environ Pollut 141: 175-183. http://dx.doi.Org/10.1016/i.envpol.2005.07.022.
Stampfli, A: Fuhrer, J. (2010). Spatial heterogeneity confounded ozone-exposure experiment in semi-natural
        grassland. Oecologia 162: 515-522. http://dx.doi.org/10.1007/s00442-009-1462-2.
Stewart. CA: Black. VJ: Black.  CR:  Roberts. JA. (1996). Direct effects of ozone on the reproductive
        development of Brassica species. J Plant  Physiol 148: 172-178.
Draft - Do Not Cite or Quote                        9-176                                  September 2011

-------
Stewart. CA. (1998) Impact of ozone on the reproductive biology of Brassica campestris L and Plantago major
        L. Loughborough University of Technology, England. Retrieved from
        http://ethos.bl.uk/OrderDetails.do?did=1&uin=uk.bl. ethos. 299673
Stoelken. G: Pritsch. K: Simon. J: Mueller. CW: Grams. TEE: Esperschuetz. J: Gavler. S: Buegger. F:
        Brueggemann, N: Meier, R: Zeller, B: Winkler, JB: Rennenberg, H. (2010). Enhanced ozone exposure
        of European beech (Fagus sylvatica) stimulates nitrogen mobilization from leaf litter and nitrogen
        accumulation in the soil. Plant Biosystems 144: 537-546.
        http://dx.doi. org/10.1080/11263500903429346.
Street. NR:  James. TM: James. T: Mikael. B: Jaakko. K: Mark. B: Taylor. G. (2011). The physiological,
        transcriptional and genetic responses of an ozone-sensitive and an ozone tolerant poplar and selected
        extremes of their F2 progeny.  Environ Pollut 159: 45-54.  http://dx.doi.Org/10.1016/i.envpol.2010.09.027.
Talhelm. AF: Pregitzer. KS: Zak.  DR. (2009). Species-specific responses to atmospheric carbon dioxide and
        tropospheric ozone mediate changes  in soil carbon.  Ecol Lett 12: 1219-1228.
        http://dx.doi.0rg/10.1111/i.1461-0248.2009.01380.x.
Tamaoki. M: Nakaiima. N: Kubo. A: Aono. M: Matsuvama. T: Sail. H. (2003). Transcriptome analysis of OS-
        exposed Arabidopsis reveals that multiple signal pathways act mutually antagonistically to induce gene
        expression. Plant Mol Biol 53: 443-456. http://dx.doi.Org/10.1023/B:PLAN.0000019064.55734.52.
Temple. PJ: Kupper. RS:  Lennox. RW: Rohr. K. (1988). Injury and yield responses of differentially irrigated
        cotton to ozone.  Agron  J 80: 751-755. http://dx.doi.org/10.2134/agroni1988.00021962008000050011x.
Temple, PJ: Riechers, GH: Miller, PR. (1992). Foliar injury responses of ponderosa pine seedlings to ozone, wet
        and dry acidic deposition, and drought. Environ Exp  Bot 32: 101-113. http://dx.doi.org/10.1016/0098-
        8472(92)90035-Z.
Theis, N: Raguso, RA. (2005). The effect of pollination on floral fragrance in thistles. J Chem Ecol 31: 2581-
        2600.
Thomas. VFD:  Braun. S:  Fluckiger. W. (2005).  Effects of simultaneous ozone exposure and nitrogen loads on
        carbohydrate concentrations, biomass, and growth of young spruce  trees (Picea abies). Environ Pollut
        137: 507-516. http://dx.doi.Org/10.1016/i.envpol.2005.02.002.
Thomas. VFD:  Braun. S:  Fluckiger. W. (2006).  Effects of simultaneous ozone exposure and nitrogen loads on
        carbohydrate concentrations, biomass, growth, and nutrient concentrations of young beech trees
        (Fagus sylvatica). Environ Pollut 143:  341-354. http://dx.doi.Org/10.1016/i.envpol.2005.11.036.
Tian. H: Melillo. J: Lu. C:  Kicklighter. D: Liu. M: Ren. W: Xu. X: Chen. G: Zhang.  C: Pan. S: Liu. J: Running. S.
        (2011). China's terrestrial carbon balance: Contributions from multiple global change factors. Global
        Biogeochem Cycles 25: GB1007. http://dx.doi.org/10.1029/201OGB003838.
Tingev. DT: Hogsett. WE: Lee. EH: Herstrom. AA: Azevedo. SH. (1991). An evaluation of various alternative
        ambient ozone standards based on crop yield loss data.  In RL Berglund; DR Lawson; DJ McKee
        (Eds.), Tropospheric Ozone and the Environment (pp. 272-288). Los Angeles, CA: Air & Waste
        Management Association.
Tingev, DT: McVeetv, BD: Waschmann, R: Johnson, MG:  Phillips, PL: Rygiewicz, PT: Olszyk, DM. (1996). A
        versatile sun-lit controlled-environment facility for studying plant and soil processes. J Environ Qual 25:
        614-625.
Tingev, DT: Rodecap, KD: Lee,  EH: Hogsett, WE: Gregg,  JW. (2002). Pod development increases the ozone
        sensitivity of Phaseolus vulgaris. Water Air Soil Pollut 139: 325-341.
Tingev. DT: Hogsett. WE: Lee. EH: Laurence. JA.  (2004).  Stricter ozone ambient air quality standard has
        beneficial effect on ponderosa pine in California. J Environ Manage 34:  397-405.
Tingev. DT: Johnson. MG: Lee.  EH: Wise. C: Waschmann. R: Olszvk. DM: Watrud. LS: Donegan. KK. (2006).
        Effects of elevated CO2 and O3 on soil respiration under ponderosa pine. Soil Biol Biochem 38: 1764-
        1778.  http://dx.doi.0rg/10.1016/i.soilbio.2005.12.003.
Tissue. DT: Thomas. RB: Strain. BR. (1997). Atmospheric CO2 enrichment increases growth and
        photosynthesis of Pinus taeda: A4 year experiment in the field. Plant Cell Environ 20: 1123-1134.
        http://dx.doi.0rg/10.1046/i.1365-3040.1997.d01-140.x.
Tissue. DT: Griffin.  KL: Ball. T. (1999). Photosynthetic adjustment in field-grown  ponderosa pine trees after six
        years  of exposure to  elevated CO2. Tree Physiol 19: 221-228.
Draft - Do Not Cite or Quote                        9-177                                  September 2011

-------
Tioelker. MG: Volin. JC: Oleksvn. J: Reich. PB. (1995). Interaction of ozone pollution and light effects on
        photosynthesis in a forest canopy experiment. Plant Cell Environ 18: 895-905.
        http://dx.doi. orq/10.1111/1.1365-3040.1995.tb00598.x.
Tobiessen. P. (1982).  Dark opening of stomata in successional trees. Oecologia 52: 356-359.
        http://dx.doi. org/10.1007/BF00367959.
Toet. S: Ineson. P: Peacock. S: Ashmore. M. (2011). Elevated ozone reduces methane emissions from peatland
        mesocosms. Global Change Biol 17: 288-296. http://dx.doi.Org/10.1111/i. 1365-2486.2010.02267.x.
long, D: Mathur, R: Schere, K: Kang, D: Yu, S. (2007). The use of air quality forecasts to assess impacts of air
        pollution on crops:  Methodology and case study. Atmos Environ 41: 8772-8784.
        http://dx.doi.ora/10.1016/i.atmosenv.2007.07.060.
long, DQ: Mauzerall,  PL. (2008). Summertime state-level source-receptor relationships between nitrogen
        oxides emissions and surface ozone concentrations over the continental United States. Environ Sci
        Technol 42: 7976-7984. http://dx.doi.org/10.1021/es7027636.
Topa, MA: Vanderklein, DW: Corbin, A.  (2001). Effects of elevated ozone and low light on diurnal and seasonal
        carbon gain in sugar maple.  Plant Cell Environ 24: 663-677.
Torsethaugen. G: Pell. EJ: Assmann. SM. (1999). Ozone inhibits guard cell K+ channels implicated in stomatal
        opening. PNAS 96: 13577-13582.
Tosti. N: Pasgualini. S: Borgogni. A: Ederli. L: Falistocco. E: Crispi. S: Paolocci. F. (2006). Gene expression
        profiles of OS-treated Arabidopsis plants. Plant Cell Environ 29: 1686-1702.
        http://dx.doi.0rg/10.1111/i.1365-3040.2006.01542.x.
Turnipseed. AA: Burns. SP: Moore. DJP: Hu. J: Guenther. AB: Monson. RK. (2009). Controls over ozone
        deposition to a high elevation subalpine forest. Agr Forest Meteorol 149: 1447-1459.
        http://dx.doi.0rg/10.1016/i.agrformet.2009.04.001.
U.S. EPA. (U.S. Environmental Protection Agency). (1978a). Air quality criteria for ozone and other
        photochemical oxidants. (EPA/600/8-78/004). Washington,  DC.
U.S. EPA. (U.S. Environmental Protection Agency). (1984). Air quality criteria for ozone and other
        photochemical oxidants, Vol. 3. (EPA/600/8-84/020A). Research Triangle Park, NC.
        http://nepis.epa. gov/Exe/ZvPURL.cgi?Dockev=2000AVEV.txt.
U.S. EPA. (U.S. Environmental Protection Agency). (1986). Air quality criteria for ozone and other
        photochemical oxidants. (EPA-600/8-84-020aF - EPA-600/8-84-020eF). Research  Triangle Park,  NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1996b). Air quality criteria for ozone and related
        photochemical oxidants, Vol. II of III. (EPA/600/P-93/004BF). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1996c). Air quality criteria for ozone and related
        photochemical oxidants, Vol. Ill of III. (EPA/600/P-93/004cF). Research Triangle  Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (1996e).  Review of national ambient air quality standards
        for ozone: Assessment of scientific and technical information: OAQPS staff paper.  (EPA/452/R-96/007).
        Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2007b).  Review of the national ambient air quality
        standards for ozone: Policy assessment of scientific and technical  information: OAQPS staff paper.
        (EPA/452/R-07/003). Research Triangle Park,  NC.
Uddling. J: Teclaw. RM: Kubiske. ME: Pregitzer. KS: Ellsworth. PS. (2008).  Sap flux in pure aspen and mixed
        aspen-birch forests exposed to elevated concentrations of carbon dioxide and ozone. Tree Physiol 28:
        1231-1243. http://dx.doi.org/18519254.
Uddling. J: Teclaw. RM: Pregitzer. KS: Ellsworth. PS. (2009). Leaf and canopy conductance in aspen and
        aspen-birch forests under free-air enrichment of carbon dioxide and ozone. Tree Physiol 29: 1367-
        1380. http://dx.doi.org/10.1093/treephys/tppQ70.
Uddling. J: Hogg. AJ:  Teclaw. RM: Carroll.  MA: Ellsworth. PS. (2010). Stomatal uptake of O3 in aspen and
        aspen-birch forests under free-air CO2 and O3 enrichment.  Environ Pollut 158: 2023-2031.
        http://dx.doi.0rg/10.1016/i.envpol.2009.12.001.
Draft - Po Not Cite or Quote                        9-178                                  September 2011

-------
UNECE. (United Nations Economic Commission for Europe). (1988). ECE critical levels workshop; March; Bad
        Harzburg, Germany. In. Geneva, Switzerland.
UNEP. (United Nations Environment Programme). (2003). Millennium Ecosystem Assessment: Ecosystems and
        human well-being: A framework for assessment. Washington, DC: Island Press.
University of Illinois. (2010). SoyFACE, from http://soyface.illinois.edu/
Unsworth. MH: Heagle. AS: Heck. WW. (1984a). Gas exchange in open-top field chambers: I. Measurement
        and analysis of atmospheric resistances to gas exchange. Atmos Environ 18: 373-380.
        http://dx.doi.org/10.1016/0004-6981(84)90111-2.
Unsworth. MH: Heagle. AS: Heck. WW. (1984b). Gas exchange in open-top field chambers: II. Resistances to
        ozone uptake by soybeans. Atmos Environ 18: 381-385. http://dx.doi.orq/10.1016/0004-
        6981(84)90112-4.
USDA. (U.S. Department of Agriculture). (2011). Ozone biomonitoring program, from
        http://www.nrs.fs.fed.us/fia/topics/ozone/
Vagaggini, B; Bartoli, MLE; Cianchetti, S; Costa, F; Bacci, E; Dente,  FL; Pi Franco, A;  Malagrino, L; Paggiaro, P.
        (2010). Increase in markers of airway inflammation after ozone exposure can be observed also in
        stable treated asthmatics with minimal functional response to ozone. Respir Res 11:5.
        http://dx.doi.org/10.1186/1465-9921-11-5.
Vahisalu. T: Kollist. H: Wang. Y. -F: Nishimura. N: Chan. W. -Y: Valerio. G: Lamminmaki. A: Brosche.  M:  Moldau.
        H: Desikan. R: Schroeder. Jl: Kangasiarvi. J. (2008). SLAC1 is required for  plant guard cell S-type
        anion channel function in stomatal signalling. Nature 452: 487-491.
        http://dx.doi. orq/10.1038/nature06608.
Valkama. E:  Koricheva. J: Oksanen. E. (2007).  Effects of elevated O3, alone and in combination with elevated
        CO2, on tree leaf chemistry and insect herbivore performance: A meta-analysis. Global Change Biol
        13: 184-201. http://dx.doi.Org/10.1111/i.1365-2486.01284.x.
van Buuren.  ML: Guidi. L: Fornale. S:  Ghetti. F: Franceschetti.  M: Soldatini. GF:  Bagni. N. (2002). Ozone-
        response mechanisms in tobacco: Implications of polyamine metabolism. New Phytol 156: 389-398.
        http://dx.doi. orq/10.1046/i. 1469-8137.2002.00539.x.
Van Dingenen. R: Dentener. FJ: Raes. F: Krol. MC: Emberson. L: Cofala. J. (2009). The global impact of ozone
        on agricultural crop yields under current and future air quality legislation. Atmos Environ 43: 604-618.
        http://dx.doi. orq/10.1016/i.atmosenv.20Q8.10.033.
Vandermeiren. K: Black. C: Pleiiel. H: de Temmerman. L. (2005). Impact of rising tropospheric ozone on potato:
        Effects on photosynthesis, growth, productivity and yield quality. Plant Cell Environ 28: 982-996.
        http://dx.doi.0rg/10.1111/i.1365-3040.2005.01316.x.
Velikova. V: Pinelli. P: Pasgualini. S: Reale. L: Ferranti. F: Loreto. F.  (2005). Isoprene decreases the
        concentration of nitric oxide in leaves exposed to elevated ozone. New Phytol 166: 419-426.
        http://dx.doi.0rg/10.1111/i.1469-8137.2005.01409.x.
Vickers. CE: Possell. M: Coiocariu. Cl: Velikova. VB:  Laothawornkitkul. J: Ryan. A: Mullineaux. PM: Hewitt. CN.
        (2009). Isoprene synthesis protects transgenic tobacco plants from oxidative stress.  Plant Cell Environ
        32: 520-531. http://dx.doi.Org/10.1111/i.1365-3040.2009.01946.x.
Vigue. LM: Lindroth. RL. (2010). Effects of genotype, elevated CO2 and elevated O3 on aspen phytochemistry
        and aspen leaf beetle Chrysomela crotchi performance. Agr Forest Entomol 12: 267-276.
        http://dx.doi.0rg/10.1111/i.1461-9563.2010.00475.x.
Volk. M: Geissmann. M: Blatter. A: Contat. F: Fuhrer. J. (2003). Design and performance of a free-air exposure
        system to study long-term effects of ozone on grasslands. Atmos Environ 37:  1341-1350.
Volk. M: Bungener. P: Contat. F: Montani.  M: Fuhrer. J. (2006). Grassland yield declined by a quarter in 5 years
        of free-air ozone fumigation. Global Change Biol 12: 74-83.  http://dx.doi.Org/10.1111/i. 1365-
        2486.2005.01083.x.
Volk. M: Obrist.  D: Novak. K: Giger. R: Bassin. S:  Fuhrer. J. (2011). Subalpine grassland carbon dioxide fluxes
        indicate substantial carbon losses under increased nitrogen deposition,  but  not at elevated ozone
        concentration. Global Change Biol 17:  366-376. http://dx.doi.org/10.1111/j. 1365-2486.2010.02228.x.
Vollenweider. P: Woodcock. H: Keltv. MJ: Hofer. R. -M. (2003).  Reduction of stem growth and site dependency
        of leaf injury in Massachusetts black cherries exhibiting ozone symptoms. Environ Pollut 125: 467-480.
Draft - Do Not Cite or Quote                        9-179                                 September 2011

-------
Vollsnes. AV: Kruse. OMO: Eriksen. AB: Oxaal. U: Futsaether. CM. (2010). In vivo root growth dynamics of
        ozone exposed Trifolium subterraneum. Environ Exp Bot 69: 183-188.
        http://dx.doi.0rg/10.1016/i.envexpbot.2010.03.007.
Vuorinen. T: Nerg. A. -M: Holopainen. JK. (2004). Ozone exposure triggers the emission of herbivore-induced
        plant volatiles, but does not disturb tritrophic signalling. Environ Pollut 131: 305-311.
        http://dx.doi.0rg/10.1016/i.envpol.2004.02.027.
Wallin. G: Skarbv. L. (1992). The influence of ozone on the stomatal and non-stomatal limitation of
        photosynthesis in Norway spruce, Picea abies (L.) Karst, exposed to soil moisture deficit. Trees Struct
        Funct 6: 128-136.  http://dx.doi.org/10.1007/BF00202428.
Wang. L: He. X: Chen. W. (2009b). Effects of elevated ozone on photosynthetic CO2 exchange and chlorophyll
        a fluorescence in leaves of Quercus mongolica grown in urban area. Bull Environ Contam Toxicol 82:
        478-481. http://dx.doi.orq/10.1007/s00128-008-9606-3.
Wang. X: Mauzerall. PL. (2004). Characterizing distributions of surface ozone and its impact on grain
        production in China, Japan and South Korea: 1990 and 2020. Atmos  Environ 38: 4383-4402.
        http://dx.doi.0rg/10.1016/i.atmosenv.2004.03.067.
Wang. X: Zheng. Q: Feng. Z: Xie. J: Ouvang. Z: Manning. WJ. (2008). Comparison of a diurnal vs steady-state
        ozone exposure profile on growth and yield of oilseed rape (Brassica napus L.) in open-top chambers
        in the Yangtze Delta, China. Environ Pollut 156: 449-453.
        http://dx.doi.0rg/10.1016/i.envpol.2008.01.027.
Wang, X: Taub,  DR. (2010). Interactive effects of elevated carbon dioxide and environmental stresses on root
        mass fraction in plants: A meta-analytical synthesis using pairwise techniques. Oecologia 163: 1-11.
        http://dx.doi. orq/10.1007/s00442-010-1572-x.
Watanabe,  M: Yamaguchi,  M: Tabe, C:  Iwasaki, M: Yamashita, R: Funada, R: Fukami, M: Matsumura, H:  Kohno,
        Y:  Izuta. T. (2007). Influences of nitrogen load on the growth and photosynthetic responses of Quercus
        serrata seedlings to O3. Trees Struct Funct 21: 421-432. http://dx.doi.orq/10.1007/s00468-007-0134-2.
Weinstein, DA: Laurence, JA: Retzlaff, WA: Kern, JS: Lee, EH: Hogsett, WE: Weber, J. (2005). Predicting the
        effects  of tropospheric ozone on regional productivity of ponderosa pine and white fir. For Ecol Manage
        205: 73-89. http://dx.doi.orq/10.1016/i.foreco.20Q4.10.007.
Werner, H:  Fabian, P. (2002). Free-air fumigation of mature trees: A novel system for controlled ozone
        enrichment in grown-up beech and spruce canopies. Environ Sci Pollut Res Int 9: 117-121.
Weselv. ML: Hicks. BB.  (2000). A review of the current status of knowledge on dry deposition [Review]. Atmos
        Environ 34: 2261-2282. http://dx.doi.Org/10.1016/S1352-2310(99)00467-7.
Whitfield. CP: Davison. AW: Ashenden. TW. (1996). Interactive effects of ozone and soil volume on Plantago
        major. New Phytol 134: 287-294. http://dx.doi.Org/10.1111/i.1469-8137.1996.tb04633.x.
Whitfield, CP: Davison, AW: Ashenden, TW. (1997). Artificial selection and heritability of ozone resistance in two
        populations of Plantago major. New Phytol 137: 645-655.
Wieser. G: Havranek. WM. (1995). Environmental control of ozone uptake in Larix decidua Mill: A comparison
        between different altitudes. Tree Physiol 15: 253-258.
Wieser. G: Manning. WJ: Tausz. M: Bvtnerowicz. A. (2006). Evidence for potential impacts of ozone on Pinus
        cembra L. at mountain  sites in Europe: An overview. Environ Pollut 139: 53-58.
        http://dx.doi.0rg/10.1016/i.envpol.2005.04.037.
Wilkinson. S: Davies. WJ. (2010). Drought, ozone, ABA and ethylene:  New insights from cell to plant to
        community. Plant Cell Environ  33: 510-525.  http://dx.doi.Org/10.1111/i.1365-3040.2009.02052.x.
Will, RE: Ceulemans, R. (1997). Effects of elevated CO2 concentration on photosynthesis, respiration and
        carbohydrate status of coppice Populus hybrids. Physiol Plant 100: 933-939.
        http://dx.doi.0rg/10.1111/i.1399-3054.1997.tb00020.x.
Winner, WE: Lefohn, AS: Cotter, IS: Greitner, CS: Nellessen, J:  McEvov, LR, Jr: Olson, Rl_: Atkinson, CJ:
        Moore.  LD. (1989). Plant responses to elevational gradients of O3 exposures in Virginia. PNAS 86:
        8828-8832.
Wittig, VE: Ainsworth, EA: Long, SP (2007). To what extent do current and projected increases in surface ozone
        affect photosynthesis and stomatal conductance of trees? A meta-analytic review of the last 3 decades
        of experiments [Review]. Plant Cell Environ  30: 1150-1162. http://dx.doi.Org/10.1111/i. 1365-
        3040.2007.01717.x.
Draft - Do Not Cite or Quote                        9-180                                  September 2011

-------
Wittig. VE: Ainsworth. EA: Naidu. SL: Karnoskv. DF: Long. SP. (2009). Quantifying the impact of current and
        future tropospheric ozone on tree biomass, growth, physiology and biochemistry: A quantitative meta-
        analysis. Global Change Biol 15: 396-424. http://dx.doi.Org/10.1111/i.1365-2486.2008.01774.x.
Woo. SY: Hincklev. TM. (2005). The effects of ozone on growth and stomatal response in the F-2 generation of
        hybrid poplar (Populus trichocarpa x Populus deltoides). Biol Plantarum 49: 395-404.
        http://dx.doi. orq/10.1007/s10535-005-0014-9.
Wright. GA: Lutmerding. A: Dudareva. N: Smith. BH. (2005). Intensity and the ratios of compounds in the scent
        of snapdragon flowers affect scent discrimination by honeybees (Apis mellifera). J Comp Physiol A
        Neuroethol Sens Neural Behav Physiol 191: 105-114.
Yamaguchi. M: Watanabe. M: Iwasaki. M: Tabe. C: Matsumura. H: Kohno. Y: Izuta. T. (2007). Growth and
        photosynthetic responses of Fagus crenata seedlings to O3 under different nitrogen loads. Trees Struct
        Funct21: 707-718. http://dx.doi.ora/10.1007/s00468-007-0163-x.
Yan. K: Chen. W: He. XY: Zhang. GY: Xu. S: Wang. LL. (2010). Responses of photosynthesis, lipid peroxidation
        and antioxidant system in leaves of Quercus mongolica to elevated O3. Environ Exp Bot 69: 198-204.
        http://dx.doi.org/10.1016/i.envexpbot.2010.03.008.
Yoshida. S: Tamaoki. M: loki. M: Ogawa. D: Sato. Y: Aono. M:  Kubo. A: Sail. S:  Sail. H:  Satoh. S: Nakaiima. N.
        (2009). Ethylene and salicylic acid control glutathione biosynthesis  in ozone-exposed Arabidopsis
        thaliana. Physiol Plant 136: 284-298. http://dx.doi.Org/10.1111/i. 1399-3054.2009.01220.x.
Younglove. T: McCool. PM: Musselman. RC: Kahl. ME.  (1994). Growth-stage dependent crop yield response to
        ozone exposure. Environ Pollut 86: 287-295. http://dx.doi.org/10.1016/0269-7491(94)90169-4.
Yuan.  JS: Himanen. SJ: Holopainen. JK: Chen. F: Stewart. CN. Jr. (2009). Smelling global climate change:
        Mitigation of function for plant volatile organic compounds. Trends Ecol Evol 24: 323-331.
        http://dx.doi.0rg/10.1016/i.tree.2009.01.012.
Yun. S. -C: Laurence. JA. (1999). The response of sensitive and tolerant clones of Populus tremuloides to
        dynamic ozone exposure under controlled environmental conditions. New Phytol 143: 305-313.
Zak, PR: Holmes, WE: Pregitzer, KS. (2007). Atmospheric CO2 and O3 alter the flow of N15 in developing
        forest ecosystems. Ecology 88: 2630-2639.
Zhang. C: Tian.  HQ: Chappelka. AH: Ren. W: Chen. H: Pan. SF: Liu.  ML: Stvers. DM: Chen. GS: Wang. YH.
        (2007a). Impacts of climatic and atmospheric changes on carbon dynamics in the Great Smoky
        Mountains National  Park. Environ Pollut 149: 336-347. http://dx.doi.Org/10.1016/i.envpol.2007.05.028.
Zhang. J: Schaub. M: Ferdinand. JA: Skellv. JM:  Steiner. KG: Savage. JE. (201 Oa). Leaf age affects the
        responses of foliar injury and gas exchange to tropospheric ozone in Prunus serotina seedlings.
        Environ Pollut 158: 2627-2634. http://dx.doi.Org/10.1016/i.envpol.2010.05.003.
Zheng. F: Wang. X: Lu. F: Hou. P: Zhang.  W: Duan. X: Zhou. X: Ai. Y: Zheng. H: Ouvang. Z: Feng. Z. (2011).
        Effects of elevated ozone concentration  on  methane emission from a rice paddy in Yangtze River Delta,
        China. Global Change Biol 17: 898-910. http://dx.doi.Org/10.1111/i.1365-2486.2010.02258.x.
Draft - Do Not Cite or Quote                       9-181                                  September 2011

-------
      10   THE  ROLE  OF  TROPOSPHERIC  OZONE  IN
          CLIMATE  CHANGE  AND  UV-B  EFFECTS
          10.1   Introduction

 1                  Atmospheric O3 plays an important role in the Earth's energy budget by interacting with
 2                  incoming solar radiation and outgoing infrared radiation. Over mid-latitudes,
 3                  approximately 90% of the total atmospheric O3 column is located in the stratosphere (Kar
 4                  etal.. 2010; Crist etal.. 1994). Therefore, tropospheric O3 makes up a relatively small
 5                  portion (-10%) of the total column of O3 over mid-latitudes, but it does play an
 6                  important role in the overall radiation budget. The next section (Section 10.2) briefly
 7                  describes the physics of the earth's radiation budget, providing background material for
 8                  the subsequent two sections assessing how perturbations in tropospheric O3 might affect
 9                  (1) climate through its role as a greenhouse gas (Section 10.3), and (2) health, ecology
10                  and welfare through its role in shielding the  earth's surface from solar ultraviolet
11                  radiation (Section 10.4).
          10.2  Physics of the Earth's Radiation Budget

12                  Radiant energy from the sun enters the atmosphere in a range of wavelengths, but peaks
13                  strongly in the visible (400 nm up to 750 nm) part of the spectrum. Longer wavelength
14                  infrared (750 nm up to ~1 mm) and shorter wavelength ultraviolet (400 nm down to
15                  100 nm) radiation are also present in the solar electromagnetic spectrum. Since the
16                  energy possessed by a photon is inversely proportional to its wavelength, infrared (IR)
17                  radiation carries the least energy per photon, and ultraviolet (UV) radiation carries the
18                  most energy per photon. UV radiation is further subdivided into classes based on
19                  wavelength: UV-A refers to wavelengths from 400-315 nm; UV-B from 315-280 nm; and
20                  UV-C from 280-100 nm. By the same argument above describing the relationship
21                  between photon wavelength and energy, UV-A radiation is the least energetic and UV-C
22                  is the most energetic band in the UV spectrum.

23                  The wavelength of radiation also determines how the photons interact with the complex
24                  mixture of gases, clouds, and particles present in the atmosphere (see Figure 10-1). UV-A
25                  radiation can be scattered but is not absorbed to any meaningful degree by atmospheric
26                  gases including O3. UV-B radiation is absorbed and scattered in part within the
27                  atmosphere. UV-C is almost entirely blocked by the Earth's upper atmosphere, where it
      Draft - Do Not Cite or Quote                      10-1                                September 2011

-------
 1
 2
          participates in photoionization and photodissociation processes including absorption by
          stratospheric O3.
*
                   Backscattered
                     Radiation
                                       Incident Solar UV Radiation
                                          Stratospheric O3
       Source: 2006 O3 AQCD.

      Figure 10-1    Diagram of the factors that determine human exposure to
                      ultraviolet radiation.
 4
 5
 6
 1
 8
 9
10
11
12
13
           Since UV-A radiation is less energetic and does not interact with O3 in the troposphere or
           the stratosphere and UV-C radiation is almost entirely blocked by stratospheric O3, UV-
           B radiation is the most important band to consider in relation to tropospheric O3
           shielding. Furthermore, tropospheric O3 plays a "disproportionate" role in absorbing UV-
           B radiation compared with stratospheric O3 on a molecule per molecule basis (Balis et
           al.. 2002; Zerefos et al.. 2002; Crist etal. 1994; Bruhl and Crutzen. 1989). This effect
           results from the higher atmospheric pressure present in the troposphere, resulting in
           higher concentrations of gas molecules present that can absorb or scatter radiation. For
           this reason, the troposphere is referred to as a "multiple scattering" regime for UV
           absorption, compared to the "single scattering" regime in the stratosphere. Thus, careful
           quantification of atmospheric absorbers and scatterers, along with a well-resolved
      Draft - Do Not Cite or Quote
                                          10-2
September 2011

-------
 1                  description of the physics of these interactions, is necessary for predicting the impact of
 2                  tropospheric O3 on UV-B flux at the surface.

 3                  Solar flux at all wavelengths has a temporal dependence, while radiative scattering and
 4                  absorption have strong wavelength, path length, and gas/particle concentration
 5                  dependencies. These combine to create nonlinear effects on UV flux at the Earth's
 6                  surface. Chapter 10 of the 2006 O3 AQCDQJ.S. EPA. 2006b) describes in detail several
 7                  key factors that influence the spatiotemporal distribution of ground-level UV radiation
 8                  flux, including: (1) long-term solar activity including sunspot cycle; (2) solar rotation; (3)
 9                  the position of the Earth in its orbit around the sun; (4) atmospheric absorption and
10                  scattering of UV radiation by gas molecules and aerosol particles; (5) absorption and
11                  scattering by stratospheric and tropospheric clouds; and (6) surface albedo. The
12                  efficiencies of absorption and scattering are highly dependent on the concentration of the
13                  scattering medium, particle size (for aerosols and clouds), and the altitude at which these
14                  processes are occurring. These properties are sensitive to meteorology, which introduces
15                  additional elements of temporal dependency in ground-level UV radiation flux.

16                  About 30% of incoming solar radiation is directly reflected back to space, mainly by
17                  clouds or surfaces with high albedo (reflectivity), such as snow, ice, and desert sand.
18                  Radiation that does penetrate to the  Earth's surface and is absorbed can be re-emitted in
19                  the longwave (infrared) portion of the spectrum (750 nm up to ~1 mm); the rest goes into
20                  evaporating water or soil moisture or emerges as sensible heat. The troposphere is opaque
21                  to the outgoing longwave radiation. Polyatomic gases such as CO2,  CH4, and O3 absorb
22                  and re-emit the radiation upwelling  from the Earth's surface, reducing the efficiency with
23                  which that energy returns to space. In effect, these gases act as a blanket warming the
24                  Earth's surface. This phenomenon, known as the "Greenhouse Effect," was first
25                  quantified in the 19th century (Arrhenius.  1896). and gives rise to the term "greenhouse
26                  gas."
          10.3  Effects of Tropospheric Ozone on Climate
      Background

27                  As a result of its interaction with incoming solar radiation and outgoing longwave
28                  radiation, tropospheric O3 is a major greenhouse gas, and increases in its abundance may
29                  contribute to climate change (IPCC. 2007b). Models estimate that the global average
30                  concentration of O3 in the troposphere has doubled since the preindustrial era (Gauss et
31                  al.. 2006). while observations indicate that in some regions tropospheric O3 may have
      Draft - Do Not Cite or Quote                      10-3                                September 2011

-------
 1                  increased by factors as great as 4 or 5 (Marenco et al., 1994; Staehelin et al., 1994). These
 2                  increases are tied to the rise in emissions of O3 precursors from human activity, mainly
 3                  fossil fuel consumption and agricultural processes.

 4                  The impact on climate of the tropospheric O3 change since preindustrial times has been
 5                  estimated to be about 25-40% of anthropogenic CO2 impact and about 75% of
 6                  anthropogenic CH4 impact (IPCC. 2007b). ranking it third in importance of the
 7                  greenhouse gases. In the 21st century as the Earth's population continues to grow and
 8                  energy technology spreads to developing countries, a further rise in the global
 9                  concentration of tropospheric O3 is likely, with associated consequences for human
10                  health and ecosystems relating to climate change.

11                  To examine the science of a changing climate and to provide balanced and rigorous
12                  information to policy makers, the World Meteorological Organization (WMO) and the
13                  United Nations Environment Programme (UNEP) formed the Intergovernmental Panel on
14                  Climate Change (IPCC) in 1988. The IPCC supports the work of the Conference of
15                  Parties (COP) to the United Nations Framework Convention on Climate Change
16                  (UNFCCC). The IPCC periodically brings together climate scientists from member
17                  countries of WMO and the United Nations to review knowledge of the physical climate
18                  system, past and future climate change, and evidence of human-induced climate change.
19                  IPCC climate assessment reports are issued every five to seven years.

20                  This section draws in part on the fourth IPCC Assessment Report (AR4) (IPCC. 2007b).
21                  as well as other peer-reviewed published research. Section 10.3.1 reviews  evidence of
22                  climate change in the recent past and projections of future climate change. It also offers a
23                  brief comparison of tropospheric O3 relative to other greenhouse gases. Section 10.3.2
24                  describes factors that influence the magnitude of tropospheric O3 effects on climate.
25                  Section  10.3.3 considers the competing effects of O3 precursors on climate. Finally,
26                  Section  10.3.4 describes the effects of changing tropospheric O3 concentrations on
27                  present-day climate. Downstream effects resulting from climate change, such as
28                  ecosystem responses, are outside the scope of this assessment, which focuses on the
29                  direct effects of tropospheric O3 on climate.
            10.3.1  Climate Change Evidence and the Influence of Tropospheric Ozone
                    10.3.1.1   Climate Change in the Recent Past

30                  From the end of the Last Ice Age 12,000 years ago until the mid-1800s, observations
31                  from ice cores show that concentrations of the long-lived greenhouse gases CO2, CH4,
      Draft - Do Not Cite or Quote                      10-4                               September 2011

-------
 1                   and N2O have been relatively stable. Unlike these greenhouse gases, O3 is not preserved
 2                   in ice, and no record of it before the late 1800s exists. Models, however, suggest that it,
 3                   too, has remained relatively constant during this time period (Thompson et al., 1993;
 4                   Thompson.  1992). The stable mix of greenhouse gases in the atmosphere has kept the
 5                   global mean temperature of the Earth close to 15°C. Without the presence of greenhouse
 6                   gases in the atmosphere, the Earth's temperature would be about 30°C cooler, or -15°C.

 7                   Since the  start of the Industrial Revolution, human activity has led to significant increases
 8                   of greenhouse gases in the atmosphere, mainly through fossil fuel combustion. According
 9                   to the IPCC AR4 (IPCC. 2007b). there is now "very high confidence" that the net effect
10                   of anthropogenic greenhouse gas emissions since 1750 has led to warming, and it is "very
11                   likely" that human activity contributed to the 0.76°C rise in global mean temperature
12                   observed over the last century. The increase of tropospheric O3 may have contributed
13                   0.1-0.3°C warming to the global climate during this time period (Hansen et al.. 2005;
14                   Mickley et al., 2004). Global cooling due to anthropogenic aerosols (IPCC. 2007b) has
15                   likely masked the full warming effect of the anthropogenic greenhouse gases. Emissions
16                   of aerosols and their precursors in the United States and other developed countries are
17                   presently decreasing rapidly due to regulatory policies. The consequences of such
18                   decreases on regional climate could be large, as indicated by observations (e.g., Philipona
19                   et al.. 2009; Ruckstuhl et al.. 2008) and models (e.g.. Kloster et al.. 2009; Micklev et al..
20                   In Press).
                     10.3.1.2   Projections of Future Climate Change

21                   The IPCC AR4 projects a warming of ~0.2°C per decade for the remainder of the 21st
22                   century (IPCC. 2007b). Even at constant concentrations of greenhouse gases in the
23                   atmosphere, temperatures are expected to increase by about 0.1°C per decade, due to the
24                   slow response of oceans to the warming applied so far. It is likely that the Earth will
25                   experience longer and more frequent heat waves in the 21st century, together with more
26                   frequent droughts and/or heavy precipitation events in some regions, due to perturbations
27                   in the hydrological cycle that result from changing temperatures (IPCC. 2007b).  Sea
28                   levels could increase by 0.3-0.8 m by 2300 due to thermal expansion of the oceans.  The
29                   extent of Arctic sea ice is expected to decline, and contraction of the Greenland ice sheet
30                   could further contribute to the sea level rise (IPCC. 2007b).

31                   Projections of future climate change are all associated with some degree of uncertainty. A
32                   major uncertainty involves future trends in the anthropogenic emissions of greenhouse
3 3                   gases or their precursors. For the IPCC AR4 climate proj ections, a set of distinct
34                   "storylines" or emission pathways was developed (IPCC. 2000). Each storyline took into
      Draft - Do Not Cite or Quote                       10-5                                September 2011

-------
 1                   account factors such as population growth, mix of energy technologies, and the sharing of
 2                   technology between developed and developing nations, and each resulted in a different
 3                   scenario for anthropogenic emissions. When these trends in emissions are applied to
 4                   models, these scenarios yield a broad range of possible climate trajectories for the 21st
 5                   century.

 6                   A second factor bringing large uncertainty to model projections of future climate is the
 7                   representation of climate and, especially, climate feedbacks. A rise in surface
 8                   temperatures would perturb a suite of other processes in the earth-atmosphere-ocean
 9                   system, which may in turn either amplify the temperature increase (positive feedback) or
10                   diminish it (negative feedback). One important feedback involves the increase of water
11                   vapor content of the atmosphere that would accompany higher temperatures (Bony et al..
12                   2006). Water vapor is a potent greenhouse gas; accounting for the water vapor feedback
13                   may increase the climate sensitivity to a doubling of CO2 by nearly a factor of two (Held
14                   and Soden. 2000). The ice-albedo feedback is also strongly positive; a decline in snow
15                   cover and sea ice extent would diminish the Earth's albedo, allowing more solar energy
16                   to be deposited to the surface (Holland and Bitz. 2003; Rind etal.. 1995). A final
17                   example of a climate feedback involves the effects of changing cloud cover in a warming
18                   atmosphere. Models disagree on the magnitude and  even the sign of this feedback on
19                   surface temperatures (Soden and Held. 2006).
                     10.3.1.3   Metrics of Potential Climate Change

20                   Two different metrics are frequently used to estimate the potential climate impact of
21                   some perturbation such as a change in greenhouse gas concentration: (1) radiative
22                   forcing; and (2) global warming potential (GWP).
23                   Radiative forcing is a change in the radiative balance at a particular level of the
24                   atmosphere or at the surface when a perturbation is introduced in the earth-atmosphere-
25                   ocean system. In the global mean, radiative forcing of greenhouse gases at the tropopause
26                   (top of the troposphere) is roughly proportional to the surface temperature response
27                   (Hansen et al., 2005; NRC. 2005). It thus provides a useful metric for policymakers for
28                   assessing the response of the earth's surface temperature to a given change in the
29                   concentration of a greenhouse gas. Positive values of radiative forcing indicate warming
30                   in a test case relative to the control; negative values indicate  cooling. The units of
31                   radiative forcing are energy flux per area, or W/m2.

32                   Radiative forcing requires just a few model years to calculate, and it shows consistency
33                   from model to model. However, radiative forcing does not take into account the climate
34                   feedbacks  that could amplify or dampen the actual surface temperature response,

      Draft - Do Not Cite or Quote                       10-6                                 September 2011

-------
 1                  depending on region. Quantifying the change in surface temperature requires a climate
 2                  simulation in which all important feedbacks are accounted for. As these processes are not
 3                  well understood, the surface temperature response to a given radiative forcing is highly
 4                  uncertain and can vary greatly among models and even from region to region within the
 5                  same model.

 6                  GWP indicates the integrated radiative forcing over a specified period (usually 100 years)
 7                  from a unit mass pulse emission of a greenhouse gas or its precursor, and are reported as
 8                  the magnitude of this radiative forcing relative to that of CO2. GWP is most useful for
 9                  comparing the potential climate impacts of long-lived gases, such as N2O or CH4. Since
10                  tropospheric O3 has a lifetime on the order of weeks to months, GWP is not seen as a
11                  valuable metric for quantifying the importance  of O3 on  climate (Torster et al.. 2007).
                     10.3.1.4  Tropospheric Ozone as a Greenhouse Gas

12                   Tropospheric O3 differs in important ways from other greenhouse gases. It is not emitted
13                   directly, but is produced through photochemical oxidation of CO, CH4, and nonmethane
14                   volatile organic compounds (VOCs) in the presence of nitrogen oxide radicals (NOX =
15                   NO + NO2; see Section 3.2 for further details on the chemistry of O3 formation). It is also
16                   supplied by vertical transport from the stratosphere. The lifetime of O3 in the troposphere
17                   is typically a few weeks, resulting in an inhomogeneous distribution that varies
18                   seasonally; the distribution of the long-lived greenhouse gases like CO2 and CH4 are
19                   much more uniform. The longwave radiative forcing by O3 is mainly due to absorption in
20                   the 9.6 urn window, where absorption by water vapor is weak. It is therefore less
21                   sensitive to local humidity than the radiative forcing by CO2  or CH4, for which there is
22                   much more overlap with the water absorption bands (Lenoble. 1993). And unlike other
23                   major greenhouse gases, O3 absorbs in the shortwave as well as the longwave part of the
24                   spectrum.

25                   Figure 10-2 shows the main steps involved in the influence of tropospheric O3 on
26                   climate. Emissions of O3 precursors including CO, VOCs, CH4, and NOX lead to
27                   production of tropospheric O3. A change in the abundance of tropospheric O3 perturbs
28                   the radiative balance of the atmosphere,  an effect quantified by the radiative forcing
29                   metric. The earth-atmosphere-ocean system responds to the radiative forcing with a
30                   climate response, typically expressed as  a change in surface temperature. Finally, the
31                   climate response causes downstream climate-related health and ecosystem impacts, such
32                   as redistribution of diseases or ecosystem characteristics due to temperature changes.
33                   Feedbacks from both the climate response and downstream impacts can, in turn, affect
34                   the abundance of tropospheric O3 and O3 precursors through multiple feedback
      Draft - Do Not Cite or Quote                      10-7                                September 2011

-------
1
2
3
mechanisms. Direct feedbacks are discussed further in Section 10.3.3.4; the downstream
climate impacts and their feedbacks are extremely complex and outside the scope of this
assessment.
                                   Precursor Emissions of
                                   CO, VOCs,CH,,NON
                                         (Tg/y)
                                          I
                                     Troposphenc O,
                                       Abundance
                                         CIS)
                                          I
                                     Radiative Forcing
                                     Due to O, Change
                                         (WATT)  "
                                     Climate Response
                                          fQ
                                  I   Climate Impacts    ,
    Figure 10-2   Schematic illustrating the effects of tropospheric ozone on climate.
                  Figure includes the relationship between precursor emissions,
                  tropospheric ozone abundance, radiative forcing, climate response,
                  and climate impacts. Units shown are those typical for each
                  quantity illustrated. Feedbacks from both the climate response and
                  climate impacts can, in turn, affect the abundance of tropospheric
                  ozone and ozone precursors through multiple feedback
                  mechanisms. Climate impacts are deemphasized in the figure since
                  these downstream effects are extremely complex and  outside the
                  scope of this assessment.
4
5
6
7
The IPCC (2007b) reported a radiative forcing of 0.35 W/m2 for the change in

tropospheric O3 since the preindustrial era, ranking it third in importance after the

greenhouse gases CO2 (1.66 W/m2) and CH4 (0.48 W/m2). Figure 10-3 shows the global

average radiative forcing estimates and uncertainty ranges in 2005 for anthropogenic

CO2, CH4, O3 and other important agents and mechanisms. The error bars encompassing
    Draft - Do Not Cite or Quote
                           10-8
September 2011

-------
1
2
the tropospheric O3 radiative forcing estimate in the figure range from 0.25 to 0.65 W/m2,
making it relatively more uncertain than the long-lived greenhouse gases.
                                     RADIATIVE FORCING COMPONENTS
                 Rr Terns
                  Long-lived
              greenhouse gases ^
                      Ozone


              Stratospheric water
               vapour Irom CHa

                 Surface albedo


I                    Direct ettect


                   Cloud albedo
                       effect


                 Linear contrails
                Solar irradiance
                     Total not
                 anthropogenic
                                           RF values (W of*) Spatial scale LOSU
                                           1.66 [1.49 to 1.83]

                                           0.48 [0.43 to 0.53]
                                           0.16 [0.14 to 0.18]
                                           0.34

                                           -0.05 [-0.15 to 0.05]
                                           0.35 [0.25 to 0.65]

                                           0.07 [0.02 to 0.12]

                                            •O.2 [-0.4 to 0.0]
                                            0.1 [0.0 to 0.2]


                                            -0.5 [-0.9 to-0.1]


                                           41.7 [-1.8 to -0.3]


                                           0.01 [0.003 to 0.03]
                                                             0.12 [0.06 to 0.30]
                                            1.6 [0.6 to 2.4]
                                                                           Global
 Global


Continental
 to global


 Global


 Local to
continental

Continental
 to global

Continental
 to global


Continental
                                                                           Global
                                                                                  High
                                                                                  High
                                                                                  Met)
Low
Med
- Low
Med
-Low
Low
                         -2      -1012
                                Radiative Forcing (W rtv2)

      Source: Used with permission from Cambridge University Press, IPCC (2007b)

     Figure 10-3   Global average radiative forcing (RF) estimates and uncertainty
                    ranges in 2005 for anthropogenic 062, ChU, ozone and other
                    important agents and mechanisms. Figure shows the typical
                    geographical extent (spatial scale) of the radiative forcing and the
                    assessed level of scientific understanding (LOSU). The net
                    anthropogenic radiative forcing and its range are also shown.
                    These require summing asymmetric uncertainty estimates from the
                    component terms, and cannot be obtained by simple addition.
                    Additional radiative forcing factors not included here are
                    considered to have a very  low LOSU.
     Draft - Do Not Cite or Quote
                              10-9
        September 2011

-------
            10.3.2  Factors that Influence the Effect of Tropospheric Ozone on Climate

 1                   This section describes the main factors that influence the magnitude of the climate
 2                   response to changes in tropospheric O3. They include: (1) trends in the concentration of
 3                   tropospheric O3; (2) the effect of surface albedo on O3 radiative forcing; (3) the effect of
 4                   vertical distribution on O3 radiative forcing; (4) feedback factors that can alter the climate
 5                   response to O3 radiative forcing; and (5) the indirect effects of tropospheric O3 on the
 6                   carbon cycle. Trends in stratospheric O3 may also affect temperatures at the Earth's
 7                   surface, but aside from issues relating STE discussed in Chapter 3, Section 3.4.2,
 8                   stratospheric O3 assessment is beyond the scope of this document.
                     10.3.2.1  Trends in the Concentration of Tropospheric Ozone

 9                   To first order, the effect of tropospheric O3 on climate is proportional to the change in
10                   tropospheric O3 concentration. The earth's surface temperatures are most sensitive to O3
11                   perturbations in the mid to upper troposphere. This section therefore focuses mainly on
12                   observed O3 trends in the free troposphere or in regions far from O3 sources, where a
13                   change in O3 concentrations may indicate change throughout the troposphere.  Data from
14                   ozonesondes, mountaintops, and remote surface sites are discussed, as well as  satellite
15                   data.
                    Observed Trends in Ozone Since the Preindustrial Era

16                  Measurements of O3 at two European mountain sites dating from the late 1800s to early
17                  1900s show values at about 10 ppb, about one-fifth the values observed today at similar
18                  sites (Pavelin etal. 1999; Marenco et al., 1994). The accuracy of these early
19                  measurements is questionable however, in part because they exhibit O3 concentrations
20                  equivalent to or only a couple of parts per billion greater than those observed at nearby
21                  low-altitude sites during the same time period (Mickley et al.. 2001; Volz and Kiev.
22                  1988). A larger vertical gradient in tropospheric O3 would be expected because of its
23                  stratospheric source and its longer lifetime aloft. In another study, Staehelin et al. (1994)
24                  revisited observations made in the Swiss mountains during the 1950s and found a
25                  doubling in O3 concentrations from that era to 1989-1991.

26                  Routine observations of O3 in the troposphere began in the 1970s with the use of balloon-
27                  borne ozonesondes, but even  this record is sparse. Trends from ozonesondes have been
28                  highly variable and dependent on  region (Logan et al.. 1999). Over most sites in the U.S.,
29                  ozonesondes reveal little trend. Over Canada, observations show a decline in O3 between
30                  1980 and 1990, then a rebound in the following decade (Tarasick et al.. 2005).
      Draft - Do Not Cite or Quote                      10-10                               September 2011

-------
 1                   Ozonesondes over Europe give a mixed picture. European ozonesondes showed
 2                   significant increases in the 1970s and 1980s, with smaller increases or even declines
 3                   since then (Oltmans et al., 2006; Logan et al., 1999). Over Japan, O3 in the lower
 4                   troposphere increased about 0.2-0.4 ppb/y during the 1990s (Naja and Akimoto. 2004).

 5                   Ground-based measurements in remote regions provide a record of tropospheric O3
 6                   extending as far back as the late 1960s or, for ship measurements, the late 1970s. A long-
 7                   term record of O3 in the San Bernardino Mountains of California reveals that the number
 8                   of high O3 days (defined as days with daily maximum  O3 levels above 95 ppb) rose from
 9                   about 100 per summer in 1969 to over 160 in 1978 (Lee et al.. 2003a). Over the next 20
10                   years, the number of high O3 days dropped slowly, to well below 100 per summer by the
11                   end of the record in 1999. Springtime O3 observations  from several other mountain sites
12                   in the western U.S. show a positive trend of about of 0.5-0.7 ppb/y since the 1980s
13                   (Cooper et al.. 2010; Jaffe et al.. 2003). Ship-borne O3  measurements for the time period
14                   1977 to 2002 indicate increases of 0.1-0.7 ppb/y over the tropical and South Atlantic, but
15                   no significant change over the North Atlantic (Lelieveld et al.. 2004). The lack of trend
16                   for the North Atlantic would seem at odds with O3 observations at Mace Head on the
17                   west coast of Ireland, which show a significant positive trend of about 0.5 ppb/y from
18                   1987 to 2003 (Simmonds et al.. 2004). Over Japan, O3 at a remote mountain site has
19                   increased 1 ppb/y from 1998 to 2003 (Tanimoto. 2009). a rate more than double that
20                   recorded by ozonesondes in the lower troposphere over Japan during the 1990s (Najaand
21                   Akimoto. 2004). At Zugspitze,  a mountain site in Germany, O3 increased by 12% per
22                   decade during the 1970s and 1980s, consistent with European ozonesondes  (Oltmans et
23                   al.. 2006). Since then, O3 continues to increase at Zugspitze, but more slowly. What little
24                   data exist for the Southern Hemisphere point to significant increases in tropospheric O3
25                   in recent decades, as much as -15% at Cape Grim in the 1989-2004 time period (Oltmans
26                   et al..  2006).

27                   The satellite record is now approaching a length that can be useful for diagnosing trends
28                   in the total tropospheric O3 column (details on the use  of satellites to measure
29                   tropospheric O3 are covered in  Section 3.5.5.5). In contrast to the surface data from ships,
30                   tropospheric O3 columns from the Total Ozone Mapping Spectrometer (TOMS) show no
31                   trend over the tropical Atlantic  for the period 1980-1990 (Thompson and Hudson.  1999).
32                   Over the Pacific, a longer, 25 year record of TOMS data again reveals no trend over the
33                   tropics, but shows increases in tropospheric column O3 of about 2-3 Dobson Units (DU)1
34                   at midlatitudes in both hemispheres (Ziemke et al., 2005).
        1 The Dobson Unit is a typical unit of measure for the total O3 in a vertical column above the Earth's surface. One DU is equivalent
      to the amount of O3 that would exist in a 1 |jm (1CT5 m) thick layer of pure O3 at standard temperature (0°C) and pressure (1 atm),
      and corresponds to a column of O3 containing 2.69 x 1020 molecules/m2. Atypical value for the amount of ozone in a column of the
      Earth's atmosphere, although highly variable, is 300 DU and approximately 10% (30 DU) of that exists in the troposphere at mid
      latitudes.
      Draft - Do Not Cite or Quote                       10-11                                 September 2011

-------
 1                   Interpreting these recent trends in tropospheric O3 is challenging. The first difficulty is
 2                   reconciling apparently contradictory trends in the observations, e.g., over tropical oceans.
 3                   A second difficulty is that the O3 trends depend on several factors, not all of which can be
 4                   well characterized. These factors include (1) trends in emissions of O3 precursors, (2)
 5                   variation in the stratospheric source of O3, (3) changes in solar radiation resulting from
 6                   stratospheric O3 depletion, and (4) trends in tropospheric temperatures (Tusco and Logan.
 7                   2003). The trends in O3 in the San Bernardino Mountains reported by Lee et al. (2003a)
 8                   likely reflects regional increases in population and motor vehicles usage, and subsequent
 9                   implementation of more stringent motor vehicle emissions controls. More recent positive
10                   trends in the western U.S. and over Japan are consistent with the rapid increase in
11                   emissions of O3 precursors from mainland Asia and transport of pollution across the
12                   Pacific (Cooper et al.. 2010; Tanimoto. 2009). The satellite trends over the northern mid-
13                   latitudes are consistent with this picture as well (Ziemke et al.. 2005). Increases in
14                   tropospheric O3 in the Southern Hemisphere are also likely due to increased
15                   anthropogenic NOX emissions, especially from biomass burning (Fishman et al.. 1991).
16                   Recent declines in summertime O3 over Europe can be partly explained by decreases in
17                   O3 precursor emissions there (Jonson et al., 2005). while springtime increases at some
18                   European sites are likely linked to changes in stratospheric dynamics (Ordonez et al..
19                   2007). Over Canada, Fusco and Logan (2003) found that O3 depletion in the lowermost
20                   stratosphere may have reduced the stratospheric flux of O3 into the troposphere by  as
21                   much as 30% from the early 1970s to the mid 1990s, consistent with the trends in
22                   ozonesondes there.


                     Calculation of Ozone Trends for the Recent Past

23                   Attempts to simulate trends in tropospheric O3 allow us to test current knowledge of O3
24                   processes and to predict with greater confidence trends in future O3 concentrations.
25                   Time-dependent emission inventories of O3 precursors have also been developed (for
26                   1850-2000. Lamarque et al.. 2010; for 1890-1990. Van Aardenne et al..  2001). These
27                   inventories allow for the calculation of changing O3 concentration over  time.

28                   One recent multi-model study calculated an increase in the O3 concentration since
29                   preindustrial times of 8-14 DU, or about 30-70% (Gauss et al.. 2006). The large spread in
30                   modeled estimates reveals our limited knowledge of processes in the pristine  atmosphere.
31                   Models typically overestimate the late nineteenth and early twentieth century
32                   observations available in surface air and at mountain sites by 50-100% (Lamarque et al..
33                   2005: Shindell et al.. 2003: Micklev et al.. 2001: Kiehletal.. 1999). Reconciling the
34                   differences between models and measurements will require more accurate simulation of
35                   the natural sources of O3 (Micklev et al.. 2001) and/or implementation of novel sinks
      Draft - Do Not Cite or Quote                       10-12                                September 2011

-------
 1                   such as bromine radicals, which may reduce background O3 in the pristine atmosphere by
 2                   as much as 30% (Yang et al.. 2005c).

 3                   For the more recent past (since 1970), application of time-dependent emissions reveals an
 4                   equatorward shift in the distribution of tropospheric O3 in the Northern Hemisphere due
 5                   to the industrialization of societies at low-latitudes (Lamarque et al.. 2005; Berntsen et
 6                   al., 2000). By constraining a model with historical (1950s-2000) observations, Shindell et
 7                   al. (2002) calculated a large increase of 8.2 DU in tropospheric O3 over polluted
 8                   continental regions since 1950. Their result appears consistent with the large change in
 9                   tropospheric O3 since preindustrial times implied by the observations from the late 1800s
10                   (Pavelin et al.. 1999; Marenco et al.. 1994).
                     10.3.2.2  The Effect of Surface Albedo on Ozone Radiative Forcing

11                   The Earth's surface albedo plays a role in O3 radiative forcing. Through most of the
12                   troposphere, absorption of incoming shortwave solar radiation by O3 is small relative to
13                   its absorption of outgoing longwave terrestrial radiation. However, over surfaces
14                   characterized by high albedo (e.g., over snow, ice, or desert sand), incoming radiation is
15                   more likely to be reflected than over darker surfaces, and the probability that O3 will
16                   absorb shortwave solar radiation is therefore larger. In other words, energy that would
17                   otherwise return to space may instead be deposited in the atmosphere. Several studies
18                   have shown that transport of O3 to the Arctic from mid-latitudes leads to radiative forcing
19                   estimates greater than  1.0 W/m2 in the region, especially in summer (Shindell et al., 2006;
20                   Liao et al.. 2004b: Mickley et al.. 1999). Because the Arctic is especially sensitive to
21                   radiative forcing through the ice-albedo feedback, the large contribution in the shortwave
22                   solar spectrum to the total radiative forcing in the region may be important.
                     10.3.2.3  The Effect of Vertical Distribution on Ozone Radiative
                                Forcing

23                   In the absence of feedbacks, O3 increments near the tropopause produce the largest
24                   increases in surface temperature (Lacis et al.. 1990; Wang et al.. 1980). This is a result of
25                   the colder temperature of the tropopause relative to the rest of the troposphere and
26                   stratosphere. Since radiation emitted by the atmosphere is approximately proportional to
27                   the fourth power of its temperature2, the colder the added O3 is relative to the earth's
        2 As described by the Stefan-Boltzmann law, an ideal blackbody-which the atmosphere approximates-absorbs at all wavelengths
      and re-radiates proportional to the fourth power of its temperature.
      Draft - Do Not Cite or Quote                       10-13                                 September 2011

-------
 1                   surface, the weaker the radiation emitted and the greater the "trapping" of longwave
 2                   radiation in the troposphere.
                     10.3.2.4   Feedback Factors that Alter the Climate Response to
                                Changes in Ozone Radiative Forcing

 3                   Estimates of radiative forcing provide a first-order assessment of the effect of
 4                   tropospheric O3 on climate. In the atmosphere, climate feedbacks and transport of heat
 5                   alter the sensitivity of Earth's surface temperature to addition of tropospheric O3.
 6                   Assessment of the full climate response to increases in tropospheric O3 requires use of a
 7                   climate model to simulate these interactions.

 8                   Due to its short lifetime, O3 is heterogeneously distributed through the troposphere.
 9                   Sharp horizontal gradients exist in the radiative forcing of O3, with the greatest radiative
10                   forcing since preindustrial times occurring over the northern mid-latitudes (more on this
11                   in Section  10.3.4). If climate feedbacks are particularly powerful, they may obscure or
12                   even erase the correlation between regional radiative forcing and climate response
13                   (Harvey. 2004; Boer and Yu. 2003). For example, several model studies have reported
14                   that the horizontal pattern of surface temperature response from 2000-2100 trends in
15                   predicted short-lived species (including O3) closely matches the pattern from the trends
16                   in the  long-lived greenhouse gases over the same time period (Levy et al.,  2008; Shindell
17                   et al..  2008; Shindell et al.. 2007). This correspondence occurs even though the patterns
18                   of radiative forcing for the short-lived and long-lived species differ significantly. In a
19                   separate paper, Shindell (2007) found that Arctic temperatures are especially sensitive to
20                   the mid-latitude radiative forcing from tropospheric O3.

21                   Other studies have found that the signature of warming due to tropospheric O3 does show
22                   some consistency with the O3 radiative forcing. For example, Mickley et al. (2004)
23                   examined the change in O3 since preindustrial times and found greater warming in the
24                   Northern Hemisphere than in the Southern Hemisphere (+0.4°C versus +0.2°C), as well
25                   as higher surface temperatures downwind of Europe and Asia and over the North
26                   American interior in summer. For an array of short-lived species including O3, Shindell
27                   and Faluvegi (2009) found that radiative forcing applied over northern mid-latitudes yield
28                   more localized responses due to local cloud, water vapor, and albedo feedbacks than
29                   radiative forcing applied over the tropics.

30                   Climate feedbacks can also alter the sensitivity of surface temperature to the  vertical
31                   distribution of tropospheric O3.  The previous section (Section 10.3.2.3) described the
32                   greater impact of O3 added to the upper troposphere (near the tropopause) on radiative
33                   forcing, relative to additions in the mid- to lower troposphere. However, warming
      Draft - Do Not Cite or Quote                      10-14                                September 2011

-------
 1                  induced by increased O3 in the upper troposphere could stabilize the atmosphere to some
 2                  extent, limiting the transport of heat to the Earth's surface and mitigating the impact of
 3                  the added O3 on surface temperature (Joshi etal.. 2003; Christiansen. 1999). Hansen et
 4                  al. (1997) determined that allowing cloud feedbacks in a climate model meant that O3
 5                  enhancements in the mid-troposphere had the greatest effect on surface temperature.

 6                  Finally, climate feedbacks can amplify or diminish the climate response of one
 7                  greenhouse gas relative to another. For example, Mickley et al. (2004) found a greater
 8                  temperature response to CO2 radiative forcing than to an O3 radiative forcing of similar
 9                  global mean magnitude, due in part to the relatively weak ice-albedo feedback for O3.
10                  Since CO2 absorbs in the same bands as water vapor, CO2 radiative forcing saturates in
11                  the middle troposphere and is also shifted toward the drier poles. A poleward shift in
12                  radiative forcing amplifies the ice-albedo feedback in the case of CO2, and the greater
13                  mid-troposphere radiative forcing allows for greater surface temperature response,
14                  relative to that for O3.
                     10.3.2.5  Indirect Effects of Tropospheric Ozone on the Carbon Cycle

15                   A proposed indirect effect of tropospheric O3 on climate involves the carbon cycle. By
16                   directly damaging plant life in ways discussed in Chapter 9, increases in tropospheric O3
17                   may depress the land-carbon sink of CO2, leading to accumulation of CO2 in the
18                   atmosphere and ultimately warming of the Earth's surface. Sitch et al. (2007) calculated
19                   that this indirect warming effect of O3 on climate has about the same magnitude as the
20                   O3 direct effect. Their results suggest a doubled sensitivity of surface temperatures to O3
21                   radiative forcing, compared to current model estimates.
             10.3.3  Competing Effects of Ozone Precursors on Climate

22                   Changes in O3 precursors affect not just O3 concentrations, but also other species that
23                   have importance to the radiative balance of the earth's climate system. More specifically,
24                   O3 and its precursors exert a strong control on the oxidizing capacity of the troposphere
25                   (Derwent et al.. 2001). For example, an increase in CO or VOCs would lead to a decrease
26                   in hydroxyl  (OH) concentrations.  Since OH is a major sink of the greenhouse gas CH4, a
27                   decline in OH would lengthen the CH4 lifetime,  enhance the CH4 concentration, and
28                   amplify surface warming. A rise in NOX emissions, on the other hand, could lead to an
29                   increase in OH in certain locations, shortening the CH4 lifetime and leading to surface
30                   cooling (Fuglestvedt et al.. 1999). O3 can itself generate OH through (1) photolysis
      Draft - Do Not Cite or Quote                      10-15                                September 2011

-------
 1                   leading to excited oxygen atoms followed by reaction with water vapor and (2) reaction
 2                   withHO2.

 3                   Analyzing the net radiative forcing per unit emission for a suite of O3 precursors,
 4                   Shindell and Faluvegi (2009) calculated positive (+0.25 W/m2) radiative forcing from the
 5                   increase in anthropogenic emissions of CO and VOCs since preindustrial times, as well
 6                   as for CH4 (+1 W/m2). These species also contribute to warming via their eventual
 7                   contribution to CO2. In contrast, Shindell and Faluvegi (2009) found negative (-
 8                   0.29 W/m2) radiative forcing from anthropogenic emissions of NOX due mainly to the
 9                   link between NOX and CH4. These results are consistent with those of Forster et al.
10                   (2007) who reported a net warming of+0.27 W/m2 for combined anthropogenic CO and
11                   VOCs emissions and a net cooling of -0.21 W/m2 for anthropogenic NOX emissions.
12                   Other studies have found a near cancellation of the positive O3 radiative forcing and the
13                   negative CH4 radiative forcing that arise from an incremental increase in anthropogenic
14                   NOX emissions (Naiket al.. 2005; Fiore et al.. 2002; Fuglestvedt et al.. 1999).

15                   The net effect of aircraft NOX on climate is complex. While Isaksen et al. (2001) reported
16                   that the net radiative forcing effect of aircraft NO emissions is near zero, Wild et al.
17                   (2001) calculated a net warming due to increased O3 production efficiency in the upper
18                   troposphere. More recently, Stevenson et al. (2004) completed a detailed analysis of the
19                   OH budget in the years following a pulse of aircraft NOX emissions.  They calculated that
20                   while such a pulse leads initially to warming through O3 production over a few months,
21                   the long-term effect is cooling through the effects on CH4.  Both aircraft NOX and the O3
22                   it generates enhance OH concentrations, with the longer-lived O3 responsible for
23                   transferring the oxidizing effects of aircraft emissions away from flight corridors.

24                   Finally, OH production from O3 precursors can affect regional sulfate air quality and
25                   climate forcing by increasing gas-phase oxidation rates of SO2. Using the A1B scenario
26                   in the IPCC AR4, Unger et al. (2006) reported that at 2030, enhanced OH from the  A IB
27                   O3 precursors increased surface sulfate aerosol concentrations by up to 20% over India
28                   and China, relative to the present-day, with  a corresponding increase in radiative cooling
29                   over these regions. In this way, O3 precursors may  impose  an indirect cooling via sulfate
30                   (Unger. 2006).

31                   Taken together, these results point out the need for careful assessment of net radiative
32                   forcing involving multiple pollutants in developing climate change policy (Unger et al.,
3 3                   2008). Naik et al. (2005) calculated that a carefully combined reduction of CO, VOCs,
34                   and NOX emissions could lead to net cooling, especially over the tropics. Several studies
3 5                   point to CH4 as a particularly attractive target for emissions control since CH4 is  itself an
36                   important precursor of O3 (West et al.. 2007; Fiore et al.. 2002). Shindell et al. (2005)
37                   calculated that the emissions-based radiative forcing of anthropogenic CH4, which
      Draft - Do Not Cite or Quote                       10-16                                 September 2011

-------
 1                  includes both its own radiative forcing and that of CH4-generated O3, is 0.8-0.9 W/m2,
 2                  about double that of the CH4 abundance-based radiative forcing. Fiore et al. (2002) found
 3                  that reducing anthropogenic CH4 emissions by  50% would lead to a global negative (-
 4                  0.37 W/m2) radiative  forcing, mostly from CH4. In later research, Fiore et al. (2008)
 5                  reported that CH4 reductions would most strongly affect tropospheric O3 column
 6                  amounts in a zonal band centered around 30 N, a region of strong downwelling and NOX-
 7                  saturated conditions near the surface.
             10.3.4 Calculating Radiative Forcing and Climate Response to Past Trends
                    in Tropospheric Ozone

 8                  The magnitude of the radiative forcing from the change in tropospheric O3 since the
 9                  preindustrial era is uncertain. This uncertainty derives in part from the scarcity of early
10                  measurements and in part from our limited knowledge regarding processes in the natural
11                  atmosphere. As noted previously, the IPCC AR4 reports a radiative forcing of 0.35 W/m2
12                  from the change in tropospheric O3 since 1750 (Torster et al.. 2007). ranking it third in
13                  importance among greenhouse gases after CO2 and CH4. The O3 radiative forcing could,
14                  in fact, be as large as 0.7 W/m2, if reconstructions of preindustrial and mid-20th century
15                  O3 based on the measurement record are valid (Shindell and Faluvegi. 2002; Mickley et
16                  al.. 2001). In any event, Unger et al. (2010) showed that present-day O3 radiative forcing
17                  can be attributed to emissions from many economic sectors, including on-road vehicles,
18                  household biofuel, power generation, and biomass burning. As much as one-third of the
19                  radiative forcing from the 1890 to 1990 change in tropospheric O3 could be due to
20                  increased biomass burning (Ito etal. 2007a).

21                  These calculated radiative forcing estimates can be compared to those obtained from
22                  satellite data.  Using data from TOMS, Worden et al. (2008) estimated a reduction in
23                  clear-sky outgoing longwave radiation of 0.48 W/m2 by O3 in the upper troposphere over
24                  oceans in 2006. This radiative forcing includes contributions from both anthropogenic
25                  and natural O3. Assuming that the concentration of O3 has roughly doubled since
26                  preindustrial times (Gauss et al.. 2006). the total  O3 radiative forcing estimated with
27                  TOMS is consistent with that obtained from models estimating just the anthropogenic
28                  contribution.

29                  Calculation of the climate response to the O3 radiative forcing is challenging due to
30                  complexity of feedbacks,  as mentioned in Sections 10.3.1.2 and 10.3.2.4. In their
31                  modeling study, Mickley et al. (2004) reported a global mean increase of 0.28°C since
32                  preindustrial times, with values as large as 0.8°C in continental interiors. For the time
33                  period since 1870, Hansen et al. (2005) estimated a much smaller increase in global mean
      Draft - Do Not Cite or Quote                      10-17                                September 2011

-------
 1                   surface temperature (0.11°C), but they implemented 1880s anthropogenic emissions in
 2                   their base simulation and also took into account trends in both stratospheric and
 3                   tropospheric O3; the modeled decline of lower stratospheric O3, especially over polar
 4                   regions, cooled surface temperatures in this study, counteracting the warming effect of
 5                   increasing tropospheric O3.

 6                   Figure 10-4 shows the Hansen et al. (2005) results as reported in Shindell et al. (2006). In
 7                   that figure, summertime O3 has the largest radiative impact over the continental interiors
 8                   of the Northern Hemisphere. Shindell et al. (2006) estimated that the change in
 9                   tropospheric O3 over the 20th century could have contributed about 0.3°C to annual mean
10                   Arctic warming and as much as 0.4-0.5°C during winter and spring. Over eastern China,
11                   Chang et al. (2009) calculated a surface temperature increase of 0.4°C to the 1970-2000
12                   change in tropospheric O3. It is not clear, however, to what degree regional changes in
13                   O3 concentration influenced this response, as opposed to more global changes.
      Draft - Do Not Cite or Quote                      10-18                                September 2011

-------
         Annual surface air temperature
                                                 Annual radiative forcing
         .41
-1.1  -.9  -.7  -J5  -.3  -.1  .1   .3  .5  .7  .9  1.1   0    .1    .2    .3   .4   .5   .6   .7

 Summer (JJA) surface air temperature     .10    Winter (DJF) surface air temperature
                                                                                                  .11
       -1.1  -.9  -.7  -.5  -.3  -.1   .1    .3   .5   .7  .9  1.5  -1.1  -.9  -.7  %5  -.3   -.1   .1   .3   .5   .7  .9  1.4
       Source: Used with permission from American Geophysical Union (Shindell et al.. 2006)
       Figure includes the input radiative forcing (W/m2), as computed by the NASA GISS chemistry-climate model. Values are surface
     temperature trends for the annual average (top left), June-August (bottom left), and December-February (bottom right) and annual
     average tropopause instantaneous radiative forcing from 1880 to 1990 (top right). Temperature trends greater than about 0.1 °C are
     significant over the oceans, while values greater than 0.3°C are typically significant over land, except for northern middle and high
     latitudes during winter where values in excess of about 0.5°C are significant. Values in the top right corner give area-weighted global
     averages in the same units as the plots.


     Figure  10-4    Ensemble average 1900-2000 surface temperature trends (°C per
                      century) in response to  tropospheric ozone changes.
         10.4  UV-B Related Effects and Tropospheric Ozone
i

2
3
4

5
            10.4.1  Background
             UV radiation emitted from the Sun contains sufficient energy when it reaches the Earth to

             break (photolyze) chemical bonds in molecules, thereby leading to damaging effects on

             living organisms and materials. Atmospheric O3 plays a crucial role in reducing exposure

             to solar UV radiation at the Earth's surface. Stratospheric O3 is responsible for the

             majority of this shielding effect, as approximately 90% of total atmospheric O3 is located

             there over mid-latitudes (Kar et al., 2010; Crist et al.. 1994). Investigation of the
     Draft - Do Not Cite or Quote
                                            10-19
September 2011

-------
 1                   supplemental shielding of UV-B radiation provided by tropospheric O3 is necessary for
 2                   quantifying UV-B exposure and the incidence of related human health effects, ecosystem
 3                   effects, and materials damage. The role of tropospheric O3 in shielding of UV-B radiation
 4                   is discussed in this section.
             10.4.2  Human Exposure and Susceptibility to Ultraviolet Radiation

 5                   The factors that potentially influence UV radiation exposure were discussed in detail in
 6                   Chapter 10 of the 2006 O3 AQCD and are summarized here. These factors included
 7                   outdoor activity, occupation, age, gender, geography, and protective behavior. Outdoor
 8                   activity and occupation both influenced the amount of time people spend outdoors during
 9                   daylight hours, the predominant factor for exposure to solar UV radiation. Participation in
10                   outdoor sports (e.g., basketball, soccer, golf, swimming, cycling) significantly increased
11                   UV radiation exposure (Thieden et al.. 2004a: Thieden et al.. 2004b: Moehrle. 2001;
12                   Moehrle et al.. 2000). Occupations that substantially increased exposure to UV radiation
13                   included farming (Schenker et al.. 2002; Airey et al.. 1997). fishing (Rosenthal et al..
14                   1988). landscaping (Rosenthal et al.. 1988). construction (Gies and Wright. 2003).
15                   physical education (Vishvakarman et al.. 2001). mail delivery (Vishvakarman et al..
16                   2001). and various other occupations that require workers to spend the majority of their
17                   day outdoors during peak UV radiation hours.

18                   Age and gender were found to be factors that influence human exposure to UV radiation,
19                   particularly by influencing other factors  of exposure such as outdoor activity and risk
20                   behavior. Studies indicated that females  generally spent less time outdoors and,
21                   consequently, had lower UV radiation exposure compared to males (Godar etal.. 2001;
22                   Gies et al.. 1998; Shoveller et al.. 1998). The lowest exposure to UV radiation among
23                   Americans in the Godar et al. (2001) study was received in females during their child
24                   raising years (age 22-40 years); the highest exposure was observed in males  aged
25                   41-59 years. A similar Canadian survey found that younger adult males had the greatest
26                   exposures to UV radiation (Shoveller et  al..  1998).

27                   Geography influences the degree of solar UV flux to the surface, and hence exposure to
28                   UV radiation. In the U.S. study by Godar et al. (2001). northerners and southerners were
29                   found to spend an equal amount of time  outdoors; however, the higher solar flux at lower
30                   latitudes significantly increased the annual UV radiation dose for southerners. The annual
31                   UV radiation doses in southerners were 25 and 40% higher in females and males,
32                   respectively, compared to northerners. Other studies also have shown that altitude and
33                   latitude influence personal exposure to UV radiation (Rigel etal.. 1999; Kimlin et al..
34                   1998).
      Draft - Do Not Cite or Quote                      10-20                                September 2011

-------
 1                   Protective behaviors such as using sunscreen (Nole and Johnson. 2004). wearing
 2                   protective clothing (Rosenthal et al.. 1988). and spending time in shaded areas (TVIoise et
 3                   al.. 1999) were shown to reduce exposure to UV radiation. In one study, the use of
 4                   sunscreen was associated with extended intentional UV radiation exposure (Autier et al..
 5                   1999); however, a follow-up study indicated that sunscreen use increased duration of
 6                   exposures to doses of UV radiation that were below the threshold level for erythema
 7                   (Autier etal.. 2000).

 8                   Given these and other factors that potentially influence UV radiation exposure, the 2006
 9                   O3 AQCD listed the following subpopulations potentially at risk for higher exposures to
10                   UV radiation:

11                      •  Individuals who engage in high-risk behavior (e.g., sunbathing);
12                      •  Individuals who participate in outdoor sports and activities;
13                      •  Individuals who work outdoors with inadequate shade (e.g., farmers,
14                         construction workers, etc.); and
15                      •  Individuals living in geographic areas with higher solar flux including lower
16                         latitudes (e.g., Honolulu, HI) and higher altitudes (e.g., Denver, CO).

17                   The risks associated with all these factors are, of course, highly dependent on season and
18                   region (Sliney and Wengraitis. 2006).
             10.4.3  Human Health Effects due to UV-B Radiation

19                   Chapter 10 of the 2006 O3 AQCD covered in detail the human health effects associated
20                   with solar UV-B radiation exposure. These effects include erythema, skin cancer, ocular
21                   damage, and immune system suppression. These adverse effects, along with protective
22                   effects of UV radiation through increased production of vitamin D are summarized in this
23                   section. For additional details, the reader is referred to Chapter 10 of the 2006 O3 AQCD
24                   (U.S. EPA. 2006b) and references therein.

25                   The most conspicuous and well-recognized acute response to UV radiation is erythema,
26                   or the reddening of the skin. Erythema is likely caused by direct damage to DNA by UV
27                   radiation (Matsumura and Ananthaswamy. 2004). Many studies discussed in the 2006 O3
28                   AQCD found skin type to be a significant risk factor for erythema. Additional risk factors
29                   include atopic dermatitis (ten Berge et al.. 2009).

30                   Skin cancer is another prevalent health effect associated with UV radiation. Exposure to
31                   UV radiation is considered to be a major risk factor for all forms of skin cancer (Diepgen
32                   and Mahler. 2002; Gloster and Brodland. 1996). Ultraviolet radiation is especially
      Draft - Do Not Cite or Quote                      10-21                                September 2011

-------
 1                   effective in inducing genetic mutations and acts as both a tumor initiator and promoter.
 2                   Keratinocytes have evolved DNA repair mechanisms to correct the damage induced by
 3                   UV; however, mutations can occur, leading to skin cancers that are appearing with
 4                   increasing frequency (Hildesheim and Fornace. 2004). The relationship between skin
 5                   cancer and chronic exposure to UV radiation is further explored in Chapter 10 of the
 6                   2006 O3 AQCD (U.S. EPA. 2006b).

 7                   Ocular damage from UV radiation exposure includes effects on the cornea, lens, iris, and
 8                   associated epithelial and conjunctival tissues. The region of the eye affected by exposure
 9                   to UV radiation depends on the wavelength of the incident UV radiation. Depending on
10                   wavelength, common health effects associated with UV radiation include photokeratitis
11                   (snow blindness; short wavelengths) and cataracts (opacity of the lens; long
12                   wavelengths).

13                   Experimental studies have shown that exposure to UV radiation may suppress local and
14                   systemic immune responses to a variety of antigens (Clydesdale et al.. 2001; Garssen and
15                   Van Loveren. 2001; Selgrade et al., 1997). In rodent models, these effects have been
16                   shown to worsen the course and outcome of some infectious diseases and cancers
17                   (Granstein and  Matsui. 2004; Norval et al., 1999). Results from human clinical studies
18                   suggest that immune suppression induced by UV radiation may be a risk factor
19                   contributing to  skin cancer induction (Ullrich. 2005; Caforio et al.. 2000; Lindelof et al..
20                   2000). There is also evidence that UV radiation has indirect involvement in viral
21                   oncogenesis through the human papillomavirus (Pfister. 2003). dermatomyositis (Okada
22                   et al.. 2003). human immunodeficiency virus (Breuer-McHam  et al.. 2001) and other
23                   forms of immunosuppression (Selgrade et al.. 2001).

24                   A potential health benefit of increased UV-B exposure relates to the production of
25                   vitamin D in humans.  Most humans depend on sun exposure to satisfy their requirements
26                   for vitamin D (Holick. 2004). Vitamin D deficiency can cause metabolic bone disease
27                   among children and adults, and also may increase the risk of many common chronic
28                   diseases, including type I diabetes mellitus and rheumatoid arthritis (Holick. 2004).
29                   Substantial in vitro and toxicological evidence also support a role for vitamin D activity
30                   against the incidence or progression of various forms of cancer (Giovannucci. 2005; John
31                   etal.. 2005; Smedbv et al.. 2005; Grant and Garland. 2004; Hughes et al.. 2004;
32                   Freedman et al.. 2002: Grant. 2002a. b; John etal.. 1999: Studzinski and Moore. 1995:
33                   Lefkowitz and Garland. 1994: Hanchette and Schwartz. 1992: Garland et al..  1990:
34                   Gorham et al.. 1990). In some studies, UV-B related production of vitamin D had
3 5                   potential beneficial immunomodulatory effects on multiple sclerosis, insulin-dependent
36                   diabetes mellitus, and  rheumatoid arthritis (Ponsonby et al.. 2002: Cantorna. 2000). More
      Draft - Do Not Cite or Quote                      10-22                                September 2011

-------
 1                  details on UV-B protective studies are provided in Chapter 10 of the 2006 O3 AQCD
 2                  (U.S. EPA. 2006b).

 3                  In establishing guidelines on limits of exposure to UV radiation, the International
 4                  commission on Non-Ionizing Radiation Protection (ICNIRP) agreed that some low-level
 5                  exposure to UV radiation has health benefits (ICNIRP. 2004). However, the adverse
 6                  health effects of higher UV exposures necessitated the development of exposure limits
 7                  for UV radiation. The ICNIRP recognized the challenge in establishing exposure limits
 8                  that would achieve a realistic balance between beneficial and adverse health effects. As
 9                  concluded by ICNIRP (2004). "[t]he present understanding of injury mechanisms and
10                  long-term effects of exposure to [UV radiation] is incomplete, and awaits further
11                  research."
             10.4.4  Ecosystem and Materials Damage Effects Due to UV-B Radiation

12                   A 2009 progress report on the environmental effects of O3 depletion from the UNEP,
13                   Environmental Effects Assessment Panel (UNEP. 2009) lists many ecosystem and
14                   materials damage effects from UV-B radiation. An in-depth assessment of the global
15                   ecosystem and materials damage effects from UV-B radiation per se is out of the scope of
16                   this assessment. However, a brief summary of some mid-latitude effects is provided in
17                   this section to provide context for UV-B related issues pertaining to tropospheric O3. The
18                   reader is referred to the UNEP report (UNEP. 2009) and references therein for further
19                   details. All of these UV-B related ecosystem and materials effects can also be influenced
20                   by climate change through temperature and other meteorological alterations, making
21                   quantifiable predictions of UV-B effects difficult.

22                   Terrestrial ecosystem effects from increased UV-B radiation include reduced plant
23                   productivity and plant cover, changes in biodiversity, susceptibility to infection, and
24                   increases in natural UV protective responses. In general, however, these effects are small
25                   for moderate UV-B increases at mid-latitudes. A field study on wheat in southern Chile
26                   found no substantial changes in crop yield with moderate increases in UV-B radiation
27                   (Calderini et al.. 2008). Similarly, field studies on silver birch (Betulapendula) in
28                   Finland found no significant  effects in photosynthetic function with increases in UV-B
29                   radiation (Aphalo et al.. 2009). Subtle, but important, changes in habitat and biodiversity
30                   have also been linked to increases in UV-B radiation (Mazza et al.. 2010; Obaraet al..
31                   2008; Wahl. 2008). Some plants have natural coping mechanisms for dealing with
32                   changes in UV-B radiation (Favory et al.. 2009; Jenkins. 2009; Brown and Jenkins. 2008;
33                   loki et al.. 2008). but these defenses may have costs in terms of reduced growth (Snell et
34                   al.. 2009: Clarke and Robinson. 2008: Semerdiieva et al.. 2003: Phoenix et al.. 2000).
      Draft - Do Not Cite or Quote                      10-23                                September 2011

-------
 1                   Aquatic ecosystem effects from increased UV-B radiation include sensitivity in
 2                   growth, immune response, and behavioral patterns of aquatic organisms. One study
 3                   looking at coccolithophores, an abundant phytoplankton group, found a 25% reduction in
 4                   cellular growth with UV-B exposure (Gao et al.. 2009a). Exposure to relevant levels of
 5                   UV-B radiation has been shown to modify immune response, blood chemistry, and
 6                   behavior in certain species of fish (Markkula et al.. 2009; Holtby and Bothwell. 2008;
 7                   Jokinen et al.. 2008). Adverse effects on growth and development from UV-B radiation
 8                   have also been observed for amphibians, sea urchins, mollusks, corals, and zooplankton
 9                   (Garcia etal.. 2009; Romansic et al.. 2009; Croteau et al.. 2008a: Croteau et al.. 2008b:
10                   Marquis et al.. 2008; Marquis and Miaud. 2008; Oromi  et al.. 2008). Increases in the flux
11                   of UV-B radiation may also result in an increase in the catalysis of trace metals including
12                   mercury, particularly in clear oligotrophic lakes with low levels of dissolved organic
13                   carbon to stop the penetration of UV-B radiation (Schindler et al.. 1996). This could then
14                   alter the mobility of trace metals including the potential for increased mercury
15                   volatilization and transport within and among ecosystems.

16                   Biogeochemical cycles, particularly the carbon cycle, can also be influenced by
17                   increased UV-B radiation. A study on high latitude wetlands found UV-induced increases
18                   in CO2 uptake through soil respiration (Haapala et al.. 2009) while studies on arid
19                   terrestrial ecosystems found evidence for UV-induced release of CO2 through
20                   photodegradation of above-ground plant litter (Brandt et al.. 2009; Henry et al.. 2008;
21                   Caldwell et al.. 2007; Zepp et al., 2007). Changes in solar UV radiation may also have
22                   effects on carbon cycling and CO2 uptake in the oceans (Brewer and Peltzer. 2009;
23                   Meador et al.. 2009; Fritz et al.. 2008; Zepp et al.. 2008; Hader et al.. 2007) as well as
24                   release of dissolved organic matter from sediment and algae (Mayer et al.. 2009;
25                   Riggsbee  et al.. 2008). Additional studies showing effects on these and additional
26                   biogeochemical cycles including the water cycle and halocarbon cycle can be found in
27                   the UNEP report (UNEP. 2009) and references therein.

28                   Materials damage from increased UV-B radiation include UV-induced
29                   photodegradation of wood (Kataoka et al.. 2007) and plastics (Pickett et al.. 2008). These
30                   studies and others summarizing photo-resistant coatings and materials designed to reduce
31                   photodegradation of materials are summarized in the UNEP report (UNEP. 2009) and
32                   references therein.
      Draft - Do Not Cite or Quote                      10-24                               September 2011

-------
             10.4.5  UV-B Related Effects Associated with Changes in Tropospheric
                     Ozone Concentrations

 1                   There are multiple complexities in attempting to quantify the relationship between
 2                   changes in tropospheric O3 concentrations and UV radiation exposure. Quantifying the
 3                   relationship between UV radiation and health or welfare effects is complicated by the
 4                   uncertainties involved in the selection of an action spectrum and appropriate
 5                   characterization of dose (e.g., peak or cumulative levels of exposure, timing of exposures,
 6                   etc.) The lack of published studies that critically examine these issues together-that is the
 7                   incremental health or welfare effects attributable specifically to UV-B changes resulting
 8                   from reductions in tropospheric O3 concentrations—reflects the significant challenges in
 9                   this field.

10                   As reported in the 2006 O3 AQCD, one analysis by Lutter and Wolz (1997) attempted to
11                   estimate the effects of a nationwide 10 ppb reduction in seasonal average tropospheric O3
12                   on the incidence of nonmelanoma and melanoma skin cancers and cataracts in humans.
13                   Their estimate, however, depended upon several simplifying assumptions, ranging from
14                   an assumed generalized 10-ppb reduction in O3 column density, national annual average
15                   incidence rates for the two types of skin cancer, and simple, linear biological
16                   amplification factors. Specifically, the decrease of 10 ppbv in seasonally averaged O3
17                   concentrations is likely an overestimate since it doesn't account for the influence of
18                   background O3 coming from the global accumulation or generation of regional chemistry
19                   (Adamowicz et al.. 2004). Further, the methodologies used in this analysis have ignored
20                   area-specific factors that are important in estimating the extent to which small, variable
21                   changes in ground-level O3 mediate long-term exposures to UV-B radiation.

22                   A more recent study by Madronich et al. (2011) used CMAQ to estimate UV radiation
23                   response to changes in tropospheric O3 under different control scenarios projected out to
24                   2020. This study focused on southeastern U.S. and accounted for spatial and temporal
25                   variation in tropospheric O3 reductions, an important consideration since most controls
26                   are focused on reducing O3 in populated urban areas. The contrasting control strategies
27                   considered in this study included a historical scenario designed to meet an 84 ppb 8-h
28                   daily max standard and a reduced scenario designed to  bring areas predicted to exceed a
29                   similarly designed 70 ppb standard into attainment. A biologically effective irradiance
30                   was estimated by multiplying the modeled UV irradiance by a sensitivity function (action
31                   spectrum) for the induction of nonmelanoma skin cancer in mice corrected for human
32                   skin transmission, then integrating over UV wavelengths. The average relative change in
33                   skin cancer-weighted surface UV radiation between the two scenarios was about 0.11%
34                   over June, July and August. Weighting by population, this estimate increased to 0.19%.
3 5                   Madronich et al. (2011) report that their estimated UV  radiation increment is an order of
      Draft - Do Not Cite or Quote                      10-25                                September 2011

-------
 1                   magnitude less than that by Lutter and Wolz (1997) with the main reason for the
 2                   discrepancy coming from the unrealistic uniform 10 ppb reduction in O3 assumed in the
 3                   former study. Madronich et al. (2011) did not attempt to link their predicted increase in
 4                   UV radiation to a predicted increase in skin cancer incidence, however, due to several
 5                   remaining and substantial uncertainties.

 6                   A handful of additional studies have addressed the relationship between changes in
 7                   tropospheric pollutant concentrations and UV-B radiation exposure, providing some
 8                   additional insight. A study  by Palancar and Toselli (2002) looked at changes in measured
 9                   UV-B radiation in relation  to ground-level air pollutants during several air pollution
10                   episodes in Cordoba,  Argentina. They found that changes in aerosol concentrations
11                   explained the majority of UV-B radiation fluctuations, and that changes in tropospheric
12                   O3 and SO2 had little effect. Repapis et al. (1998) performed a similar study on UV-B
13                   exposures during high and  low air pollution days in Athens, Greece. They found cloud
14                   cover and aerosols to be the major factors in observed UV-B exposures reductions.
15                   Studies by Acosta and Evans (2000) in Mexico City and Koronakis et al. (2002) in
16                   Athens, Greece  both found significant reductions in surface-level UV exposures during
17                   pollution episodes. Both these studies include tropospheric O3 as a potential driver for the
18                   reductions,  but neither study was able to quantify the influence of individual atmospheric
19                   components involved in the observed attenuation in UV-B radiation.

20                   In the absence of reliable studies specifically addressing UV-B related health effects from
21                   a reduction in tropospheric O3, inferences were made in the 2006 O3 AQCD on the basis
22                   of studies focused on stratospheric O3 depletion. Studies included in that review
23                   examined the potential effect of stratospheric O3 depletion on the risk of erythema
24                   (Longstreth et al.. 1998). skin cancer (Urbach. 1997; Slaperetal.. 1996; De Gruiil  1995;
25                   Longstreth  et al.. 1995; Madronich and De Gruijl. 1993). nonmelanoma skin cancer
26                   (Slaperetal.. 1996; Longstreth et al.. 1995). and cataracts (Longstreth et al.. 1995). Note
27                   that several of the concerns expressed above in relation to the Lutter and Wolz (1997)
28                   analysis are relevant to these  analyses as well. Furthermore, these studies have a high
29                   degree of uncertainty due to inadequate information on the action spectrum and dose-
30                   response relationships. As a result, caution is advised when assessing and interpreting the
31                   quantitative results of health risks due to stratospheric O3 depletion in the context of
32                   tropospheric O3 shielding.

33                   Although the UV-B related health effects attributed to marginal reductions in
34                   tropospheric or ground-level  O3 that would result from reductions in O3 concentrations
3 5                   have not been directly assessed, they would be expected to be small given the above
36                   findings and the fact that tropospheric O3 makes up only -10% of the total atmospheric
37                   O3 column  at mid-latitudes (Kar et al.. 2010). Furthermore, O3 present in the planetary
      Draft - Do Not Cite or Quote                       10-26                                 September 2011

-------
 1                  boundary layer makes up only -10% of tropospheric O3 (Thompson et al., 2007) and the
 2                  NAAQS has only a fractional influence on those ground-level O3 concentrations. The net
 3                  result is a very small influence on total column O3 through attainment of the O3 standard.
 4                  In addition, the health benefits of UV-B in the production of vitamin D suggests that
 5                  increased risks of human disease due to a slight excess in UV-B radiation exposure may
 6                  be offset by the benefits of enhanced vitamin D production. However, as with other
 7                  impacts of UV-B on human health, this beneficial effect of UV-B has not been studied in
 8                  sufficient detail to allow for a credible health benefits assessment. Hence, the above
 9                  mentioned health and welfare effects associated with UV-B exposures resulting from
10                  changes in ground-level O3 concentrations would likely be small or nonexistent based on
11                  current information.

12                  More reasonable estimates of the human health impacts of enhanced UV-B penetration
13                  following reduced ground-level O3 concentrations require both (a) a solid understanding
14                  of the multiple factors that define the extent of human exposure to UV-B, and (b) well-
15                  defined and quantifiable links between human disease and UV-B exposure. Within the
16                  uncertain context of presently available information on UV-B surface fluxes, a risk
17                  assessment of UV-B-related health effects would need to factor in human habits (e.g.,
18                  daily activities, recreation, dress, and skin care) in order to adequately estimate UV-B
19                  exposure levels. Little is known about the impact of variability in these human factors on
20                  individual exposure to UV radiation. Furthermore, detailed information does not exist
21                  regarding the relevant type (e.g., peak or cumulative) and time period (e.g., childhood,
22                  lifetime,  or current) of exposure, wavelength dependency of biological responses, and
23                  inter-individual variability in UV resistance. In conclusion, the effect of changes in
24                  surface-level O3 concentrations on UV-induced health outcomes cannot yet be critically
25                  assessed within reasonable uncertainty. The reader is referred to the U.S. EPA 2002 Final
26                  Response to Court Remand (U.S. EPA. 2003) for detailed discussions of the data and
27                  scientific issues associated with the determination of public health benefits resulting from
28                  the attenuation of UV-B by surface-level O3.
          10.5  Summary
            10.5.1 Summary of the Effects of Tropospheric Ozone on Climate

29                  Tropospheric O3 is a major greenhouse gas, third in importance after CO2 and CH4.
30                  While the developed world has successfully reduced emissions of O3 precursors in recent
31                  decades, many developing countries have experienced large increases in precursor
32                  emissions and these trends are expected to continue, at least in the near term. Projections
      Draft - Do Not Cite or Quote                      10-27                               September 2011

-------
 1                   of radiative forcing due to changing O3 over the 21st century show wide variation, due in
 2                   large part to the uncertainty of future emissions of source gases. In the near-term (2000-
 3                   2030), projections of O3 radiative forcing range from near zero to +0.3 W/m2, depending
 4                   on the emissions scenario (Stevenson et al.. 2006). Reduction of tropospheric O3
 5                   concentrations could therefore provide an important means to slow climate change in
 6                   addition to the added benefit improving surface air quality.

 7                   It is clear that increases in tropospheric O3 lead to warming. However the precursors of
 8                   O3 also have competing effects on the greenhouse gas CH4, complicating emissions
 9                   reduction strategies. A decrease in CO or VOC emissions would enhance OH
10                   concentrations, shortening the lifetime of CH4, while a decrease in NOX emissions could
11                   depress OH concentrations in  certain regions and lengthen the CH4 lifetime. Recent
12                   research, however, has indicated that a carefully combined reduction of CO, VOCs,  and
13                   NOX emissions could lead to net cooling (Naik et al.. 2005). They calculate that such
14                   reductions  would have the greatest impact for developing countries in tropical regions.

15                   Abatement of CH4 emissions would likely provide the most straightforward means to
16                   address climate change since CH4 is itself an important precursor of background O3
17                   (West et al.. 2007; West et al.. 2006;  Fiore et al.. 2002). A reduction of CH4 emissions
18                   would also improve air quality on its own right. A set of global abatement measures
19                   identified by West and Fiore (2005) could reduce CH4 emissions by 10% at a cost
20                   savings, decrease background O3 by about 1 ppb in the Northern Hemisphere summer,
21                   and lead to a global net cooling of 0.12 W/m2. Unlike measures to reduce NOX, which
22                   would have immediate impacts on surface O3 but little net radiative forcing, the cooling
23                   effects of CH4 controls would be realized gradually, over -12 years. West et al. (2007)
24                   explored further the benefits of CH4 abatement, finding that a 20% reduction in global
25                   CH4 emissions would lead to significantly greater cooling per unit reduction in surface
26                   O3, compared to 20% reductions in VOCs or CO.

27                   Important uncertainties remain regarding the impact of tropospheric O3 on future climate
28                   change. To address these uncertainties, further research is needed to: (1) enhance our
29                   knowledge of the natural atmosphere; (2) interpret observed trends of O3 in the free
30                   troposphere and remote regions; (3) improve our understanding of the CH4 budget,
31                   especially emissions from wetlands and agricultural sources, (4) understand the
32                   relationship between regional  O3 radiative forcing and regional climate change; and (5)
33                   determine the optimal mix of emissions reductions that would act to limit future climate
34                   change.
      Draft - Do Not Cite or Quote                      10-28                                September 2011

-------
            10.5.2 Summary of UV-B Related Effects on Human Health, Ecosystems,
                    and Materials Relating to Changes in Tropospheric Ozone
                    Concentrations

 1                  UV radiation emitted from the Sun contains sufficient energy when it reaches the Earth to
 2                  break (photolyze) chemical bonds in molecules, thereby leading to damaging effects on
 3                  living organisms and materials. Atmospheric O3 plays a crucial role in reducing exposure
 4                  to solar UV radiation at the Earth's surface. Ozone in the stratosphere is responsible for
 5                  the majority of this shielding effect, as approximately 90% of total atmospheric O3 is
 6                  located there over mid-latitudes. Ozone in the troposphere provides supplemental
 7                  shielding of radiation in the wavelength band from 280-315 nm, referred to as UV-B
 8                  radiation. UV-B radiation has important effects on human health and ecosystems, and is
 9                  associated with materials damage.

10                  Adverse human health effects associated with solar UV-B radiation exposure include
11                  erythema, skin cancer, ocular damage, and immune system suppression. A potential
12                  human health benefit of increased UV-B exposure involves the UV-induced production
13                  of vitamin D which may help reduce the risk of metabolic bone disease, type I diabetes,
14                  mellitus, and rheumatoid arthritis, and may provide beneficial immunomodulatory effects
15                  on multiple sclerosis, insulin-dependent diabetes mellitus, and rheumatoid arthritis.

16                  Adverse ecosystem and materials damage effects associated with solar UV-B radiation
17                  exposure include terrestrial and aquatic ecosystem impacts, alteration of biogeochemical
18                  cycles, and degradation of man-made materials. Terrestrial ecosystem effects from
19                  increased UV-B radiation include reduced plant productivity and plant cover, changes in
20                  biodiversity, susceptibility to infection, and increases in natural UV protective responses.
21                  In general, however, these effects are small for moderate UV-B increases at mid-
22                  latitudes. Aquatic ecosystem effects from increased UV-B radiation include sensitivity in
23                  growth, immune response, and behavioral patterns of aquatic organisms and the potential
24                  for increased catalysis and mobility of trace metals. Biogeochemical cycles, particularly
25                  the carbon cycle, can also be influenced by increased UV-B radiation with effects ranging
26                  from UV-induced increases in CO2 uptake through soil respiration to UV-induced release
27                  of CO2 through photodegradation of above-ground plant litter. Changes in solar UV
28                  radiation may also have effects  on carbon cycling and CO2 uptake in the oceans as well
29                  as release of dissolved organic matter from sediment and algae. Finally, materials damage
30                  from increased UV-B radiation includes UV-induced photodegradation of wood and
31                  plastic.

32                  There is a lack of published  studies that critically examine the incremental health or
33                  welfare effects (adverse or beneficial) attributable specifically to changes in UV-B
34                  exposure resulting from perturbations in tropospheric O3 concentrations. While the
      Draft - Do Not Cite or Quote                      10-29                               September 2011

-------
1                   effects are expected to be small, they cannot yet be critically assessed within reasonable
2                   uncertainty.
         10.6  References

     Acosta. LR: Evans. WFJ. (2000). Design of the Mexico City UV monitoring network: UV-B measurements at
            ground level in the urban environment. J Geophys Res 105: 5017-5026.
            http://dx.doi. orq/10.1029/1999JD900250.
     Adamowicz.  V: Dales. R: Hale. BA: Hrudev. SE: Krupnick. A: Lippman. M: McConnell. J: Renzi. P. (2004).
            Report of an expert panel to review the socio-economic models and related components supporting the
            development of Canada-Wide Standards (CWS) for particulate matter (PM) and  ozone to the Royal
            Society of Canada. J Toxicol Environ Health B Grit Rev 7: 147-266.
            http://dx.doi. org/10.1080/10937400490253238.
     Airev. DK: Wong. JCF: Fleming. RA: Meldrum. LR. (1997). An estimate of the UV-B exposure for outdoor
            workers during a south-east Queensland summer. Health Phys 72: 544-549.
     Aphalo, PJ: Vapaavuori, EM: de la Rosa, TM: Lehto, T. (2009). Does supplemental UV-B  radiation affect gas
            exchange and RuBisCO activity of Betula pendula Roth,  seedlings grown in forest soil under
            greenhouse conditions? Plant Ecol Divers 2: 37-43.  http://dx.doi.orq/10.1080/17550870902780299.
     Arrhenius, S. (1896). On the influence of carbonic acid in the air upon the temperature of the ground. Philos
            Mag 41: 237-276.
     Autier. P:  Pore. JF: Negrier. S: Lienard.  D: Panizzon. R: Leieune. FJ: Guggisbera. D: Eaaermont. AM. (1999).
            Sunscreen use and duration of sun exposure: A double-blind, randomized trial. J Natl Cancer Inst 91:
            1304-1309.
     Autier. P:  Pore. JF: Reis. AC:  Grivegnee. A: Ollivaud. L: Truchetet. F: Chamoun. E: Rotmensz. N: Severi. G:
            Cesarini, JP (2000). Sunscreen use and intentional  exposure to  ultraviolet A and B radiation: A double
            blind randomized trial using personal dosimeters. Br J Cancer 83: 1243-1248.
     Balis. PS: Zerefos. CS: Kourtidis. K: Bais. AF: Hofzumahaus. A: Kraus. A: Schmitt. R:  Blumthaler. M: Gobbi. GP
            (2002). Measurements and modeling of photolysis rates  during the photochemical activity and
            ultraviolet radiation (PAUR) II campaign. J Geophys Res 107: 8138.
            http://dx.doi. orq/10.1029/2000JD000136.
     Berntsen, TK: Myhre, G: Stordal, F: Isaksen, ISA. (2000). Time evolution  of tropospheric ozone and its radiative
            forcing. J  Geophys Res 105: 8915-8930. http://dx.doi.orq/10.1029/1999JD901139.
     Boer. GJ: Yu. B. (2003). Climate sensitivity and response. Clim Dynam 20: 415-429.
            http://dx.doi. org/10.1007/s00382-002-0283-3.
     Bony. S: Colman. R: Kattsov. VM: Allan. RP: Bretherton. CS: Dufresne. JL: Hall. A: Halleqatte. S:  Holland. MM:
            Ingram. W: Randall. DA: Soden. BJ: Tselioudis. G: Webb. MJ. (2006). How well do we understand and
            evaluate climate change feedback processes? J Clim 19: 3445-3482.
     Brandt. LA: Bohnet. C: King. JY. (2009). Photochemically induced carbon dioxide production as a mechanism
            for carbon loss from plant litter in arid ecosystems. J Geophys Res 114: G02004.
            http://dx.doi. org/10.1029/2008ig000772.
     Breuer-McHam. J: Simpson. E: Dougherty.  I: Bonkobara. M: Ariizumi. K: Lewis. DE: Dawson. DB: Duvic. M:
            Cruz. PP. Jr. (2001). Activation of HIV in human skin by ultraviolet B  radiation and its inhibition by
            NF"kappa"B blocking agents. Photochem Photobiol  74: 805-810.
     Brewer. PG:  Peltzer. ET. (2009). Limits to marine life. Science 324: 347-348.
            http://dx.doi.org/10.1126/science.1170756.
     Brown, BA: Jenkins, Gl. (2008). UV-B signaling pathways with different fluence-rate response profiles are
            distinguished in mature Arabidopsis leaf tissue by requirement for UVR8,  HY5, and HYH. Plant Physiol
            146: 576-588. http://dx.doi.org/10.1104/pp.107.108456.
     Bruhl, C:  Crutzen,  PJ. (1989). On the disproportionate role of tropospheric ozone as a filter against solar UV-B
            radiation.  Geophys Res Lett 16: 703-706. http://dx.doi.org/10.1029/GL016i007p00703.
     Draft - Do Not Cite or Quote                       10-30                                 September 2011

-------
Caforio. ALP: Fortina. AB: piaserico. S: Alaibac. M: Tona. F: Feltrin. G: Pompei. E: Testolin. L: Gambino. A:
       Volta, SD: Thiene, G: Casarotto, D: Peserico, A. (2000). Skin cancer in heart transplant recipients: risk
       factor analysis and relevance of immunosuppressive therapy. Circulation 102: III-222 - III-227.
Calderini. DF: Lizana. XC: Hess. S: Jobet. CR: Zuniga. JA. (2008). Grain yield and quality of wheat under
       increased ultraviolet radiation (UV-B) at later stages of the crop cycle. J Agr Sci 146: 57-64.
       http://dx.doi. orq/10.1017/S0021859607007447.
Caldwell.  MM: Bornman. JF: Ballare.  CL: Flint. SD: Kulandaivelu. G.  (2007). Terrestrial ecosystems, increased
       solar ultraviolet radiation, and interactions with bother climate change factors. Photochem Photobiol Sci
       6: 252-266. http://dx.doi.orq/10.1039/B700019q.
Cantorna.  MT. (2000). Vitamin D and  autoimmunity: Is vitamin D status an environmental factor affecting
       autoimmune disease prevalence? Exp Biol Med 223: 230-233.
Chang. W: Liao. H: Wang. H.  (2009).  Climate responses to direct radiative forcing of anthropogenic aerosols,
       tropospheric ozone, and long-lived greenhouse gases in eastern China over 1951-2000. Adv Atmos Sci
       26: 748-762.  http://dx.doi.org/10.1007/s00376-009-9032-4.
Christiansen. B. (1999). Radiative forcing and climate sensitivity: The ozone experience. Q J Roy Meteorol Soc
       125:  3011-3035. http://dx.doi.ora/10.1002/qi.49712556011.
Clarke, LJ: Robinson, SA. (2008). Cell wall-bound ultraviolet-screening compounds explain the high ultraviolet
       tolerance of the Antarctic moss,  Ceratodon purpureus. New Phytol 179: 776-783.
       http://dx.doi. orq/10.1111/i.1469-8137.2008.02499.x.
Clydesdale, GJ: Dandle,  GW: Muller,  HK. (2001). Ultraviolet light induced injury: Immunological and
       inflammatory effects. Immunol Cell Biol 79: 547-568.
Cooper. OR: Parrish. DP: Stohl. A: Trainer. M: Nedelec. P: Thouret. V: Gammas.  JP: Oltmans. SJ: Johnson. BJ:
       Tarasick, D: Leblanc, T:  McDermid, IS: Jaffe, D: Gao, R: Stith, J: Ryerson, T: Aikin,  K: Campos, T:
       Weinheimer. A: Averv. MA. (2010). Increasing springtime ozone mixing ratios in the free troposphere
       over western North America. Nature 463: 344-348. http://dx.doi.orq/10.1038/nature08708.
Crist, KC: Carmichael, GR: John, K. (1994). UV-B exposure and atmospheric ozone - Evaluation of radiative
       flux to changes in ambient ozone levels. J Hazard Mater 37: 527-538.  http://dx.doi.orq/10.1016/0304-
       3894(93)E0096-K.
Croteau, MC: Martyniuk,  CJ: Trudeau, Vl_: Lean, DRS. (2008a). Chronic exposure of rana pipiens tadpoles to
       uvb radiation and the estrogenic chemical 4-tert-octylphenol. J Toxicol Environ Health A 71: 134-144.
Croteau. MC: Davidson.  MA: Lean. DRS: Trudeau. VL. (2008b). Global increases in ultraviolet B radiation:
       Potential impacts on amphibian development and metamorphosis. Physiol Biochem Zool 81: 743-761.
       http://dx.doi. orq/10.1086/591949.
De Gruiil. FR. (1995). Action spectrum for photocarcinogenesis. Recent Results Cancer Res 139: 21-30.
Derwent, RG: Collins, WJ: Johnson, CE: Stevenson, PS. (2001). Transient behaviour of tropospheric ozone
       precursors in a global 3-D CTM and their indirect greenhouse effects. Clim Change 49: 463-487.
Diepqen. TL:  Mahler. V (2002). The epidemiology of skin cancer.  Br J Dermatol 146: 1-6.
Favorv, JJ: Stec, A: Gruber, H: Rizzini, L: Oravecz, A: Funk, M: Albert, A: Cloix, C: Jenkins, Gl: Oakelev, EJ:
       Seidlitz. HK: Naqy. F: Ulm. R. (2009). Interaction of COP1 and UVR8 regulates UV-B-induced
       photomorphogenesis and stress acclimation in Arabidopsis.  EMBO J 28: 591-601.
       http://dx.doi. orq/10.1038/emboi.2009.4.
Fiore. AM: Jacob. DJ: Field. BD:  Streets. DG: Fernandes. SD: Jang.  C. (2002). Linking ozone pollution and
       climate change:  The case for controlling methane. Geophys Res Lett 29: 1919.
       http://dx.doi. orq/10.1029/2002GL015601.
Fiore. AM: West. JJ: Horowitz. LW: Naik. V: Schwartzkopf. MD. (2008). Characterizing the tropospheric ozone
       response to methane emission controls and the benefits to climate and air quality. J Geophys Res 113:
       D08307. http://dx.doi.orq/10.1029/2007JD009162.
Fishman. J: Fakhruzzaman. K: Cros.  B: Nqanqa. D. (1991). Identification of widespread pollution in the
       Southern Hemisphere deduced from satellite analyses. Science 252: 1693-1696.
       http://dx.doi.orq/10.1126/science.252.5013.1693.
Draft - Do Not Cite or Quote                        10-31                                  September 2011

-------
Forster. P: Ramaswamv. V: Artaxo. P: Berntsen. T: Betts. R: Fahev. DW: Havwood. J: Lean. J: Lowe. DC:
        Myhre, G: Nganga, J: Prinn, R: Raga, G: Schultz, M: Van Dorland, R. (2007). Changes in atmospheric
        constituents and in radiative forcing.  In S Solomon;  D Qin; M Manning; Z Chen; M Marquis; KB Averyt;
        M Tignor; HL Miller (Eds.), Climate Change 2007: The physical science basis: Contribution of Working
        Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 129-
        234). Cambridge, U.K. and New York, NY: Cambridge University Press.
Freedman. DM: Dosemeci. M: McGlvnn. K. (2002). Sunlight and mortality from  breast, ovarian, colon, prostate,
        and non-melanoma skin cancer: A composite death certificate based case-control study. Occup Environ
        Med 59:  257-262.
Fritz. JJ: Neale. PJ: Davis. RF: Peloquin. JA.  (2008). Response of Antarctic phytoplankton to solar UVR
        exposure: Inhibition and recovery of photosynthesis in coastal and pelagic assemblages. Mar Ecol
        Prog Ser 365: 1-16. http://dx.doi.org/10.3354/Meps07610.
Fuglestvedt. JS: Berntsen. TK:  Isaksen. ISA:  Mao.  H: Liang. X. -Z: Wang. W.  -C. (1999). Climatic forcing of
        nitrogen  oxides through changes in tropospheric ozone and methane: Global 3D model studies. Atmos
        Environ 33: 961-978. http://dx.doi.orq/10.1016/S1352-2310(98)00217-9.
Fusco. AC:  Logan. JA. (2003). Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
        midlatitudes with the GEOS-CHEM model. J Geophys Res 108: 4449.
        http://dx.doi. orq/10.1029/2002JD002742.
Gao. KS: Ruan. ZX: Villafane. VE: Gattuso. JP: Helbling. EW. (2009a). Ocean acidification exacerbates the
        effect of UV radiation on the calcifying phytoplankter Emiliania huxleyi.  Limnol Oceanogr 54: 1855-
        1862.
Garcia. TS: Paoletti.  DJ: Blaustein. AR.  (2009). Correlated trait responses to multiple selection pressures in
        larval amphibians reveal conflict avoidance strategies. Freshw Biol 54:  1066-1077.
        http://dx.doi. org/10.1111/i.1365-2427.2008.02154.x.
Garland. FC: Garland. CF: Gorham. ED: Young. JF. (1990). Geographic variation in breast cancer mortality in
        the United States: A hypothesis involving exposure to solar radiation. Prev Med 19: 614-622.
Garssen. J: Van Loveren. H. (2001). Effects of ultraviolet exposure on the immune system. Crit Rev Immunol
        21: 359-397.
Gauss, M: Myhre, G; Isaksen, ISA: Grewe, V; Pitari, G: Wild, O; Collins, WJ:  Dentener, FJ; Ellingsen, K:  Gohar,
        LK: Hauglustaine. DA: lachetti. D: Lamargue. JF: Mancini. E: Micklev. LJ: Prather. MJ: Pvle. JA:
        Sanderson.  MG: Shine. KP: Stevenson. PS: Sudo. K: Szopa. S: Zeng.  G. (2006). Radiative forcing
        since preindustrial times due to ozone change in the troposphere and the lower stratosphere. Atmos
        Chem Phys 6: 575-599.
Gies. P: Roy. C: Toomev. S: MacLennan. R: Watson. M. (1998). Solar UVR exposures of primary school
        children  at three locations in Queensland.  Photochem Photobiol 68: 78-83.
Gies. P: Wright. J. (2003). Measured solar ultraviolet radiation exposures of outdoor workers in Queensland in
        the building and construction industry. Photochem Photobiol  78: 342-348.
Giovannucci, E. (2005). The epidemiology of vitamin D and cancer incidence and mortality: A review (United
        States) [Review]. Cancer Causes Control 16: 83-95.
Gloster.  HM. Jr: Brodland. DG.  (1996). The epidemiology of skin cancer.  Dermatol Surg 22: 217-226.
Godar, DE:  Wengraitis, SP; Shreffler, J: Sliney, PH. (2001). UV doses of Americans. Photochem Photobiol 73:
        621-629.
Gorham. ED: Garland. FC: Garland. CF. (1990). Sunlight and breast cancer incidence in the USSR. Int J
        Epidemiol 19: 820-824. http://dx.doi.Org/10.1093/iie/19.4.820.
Granstein. RD: Matsui. MS. (2004). UV radiation-induced immunosuppression and skin cancer. Cutis 5: 4-9.
Grant. WB.  (2002a). An ecologic study of dietary and solar ultraviolet-B links to  breast carcinoma mortality rates.
        Cancer 94: 272-281.
Grant. WB.  (2002b). An estimate of premature cancer mortality in the US due to inadequate doses of solar
        ultraviolet-B radiation. Cancer 94: 1867-1875.
Grant, WB:  Garland, CF. (2004). A critical  review of studies on vitamin D  in relation to colorectal cancer
        [Review]. Nutr Cancer 48: 115-123.
Haapala. JK: Morskv. SK: Saarnio. S: Rinnan. R: Suokanerva. H: Kvr. E: Latola. K: Martikanen. PJ: Holopainen.
        T: Silvola, J. (2009). Carbon dioxide balance of a fen ecosystem in northern  Finland under elevated
        UV-B radiation. Global Change Biol 15: 943-954. http://dx.doi.Org/10.1111/i.1365-2486.2008.01785.x.
Draft - Do Not Cite or Quote                       10-32                                 September 2011

-------
Hader. DP: Kumar. HP: Smith. RC: Worrest. RC. (2007). Effects of solar UV radiation on aquatic ecosystems
        and  interactions with climate change. Photochem Photobiol Sci 6: 267-285.
        http://dx.doi. orq/10.1039/B700020k.
Hanchette. CL: Schwartz. GG. (1992). Geographic patterns of prostate cancer mortality. Cancer 70: 2861-2869.
Hansen, J: Sato, M: Ruedy, R: Nazarenko, l_: Lacis, A: Schmidt, GA: Russell, G: Aleinov, I: Bauer, M: Bauer, S:
        Bell. N: Cairns. B: Canuto. V: Chandler. M: Cheng. Y: Del Genio. A: Faluvegi. G: Fleming.  E: Friend. A:
        Hall. T: Jackman. C: Kellev.  M: Kiang. N: Koch. D: Lean. J: Lerner. J: Lo.  K: Menon. S: Miller. R: Minnis.
        P: Novakov, T: Oinas, V: Perlwitz, J: Rind, D: Romanou, A: Shindell, D: Stone, P: Sun, S: Tausnev, N:
        Thresher.  D: Wielicki. B: Wong. T: Yao. M: Zhang. S. (2005). Efficacy of climate forcings. J Geophys
        Res 110: D18104. http://dx.doi.ora/10.1029/2005JD005776.
Hansen, JE:  Sato,  M: Ruedy, R. (1997). Radiative forcing and climate response. J Geophys Res 102: 6831-
        6864. http://dx.doi.orq/10.1029/96JD03436.
Harvey. LDP. (2004). Characterizing the annual-mean climatic effect of anthropogenic CO2 and aerosol
        emissions in eight coupled atmosphere-ocean GCMs. Clim Dynam 23: 569-599.
        http://dx.doi. orq/10.1007/s00382-004-0455-4.
Held. IM: Soden. BJ. (2000). Water vapor feedback and global warming. Annual Review of Energy  and the
        Environment 25:  441-475.
Henry. HAL: Brizgvs. K: Field. CB. (2008). Litter decomposition in a California annual grassland: Interactions
        between photodegradation and litter layer thickness. Ecosystems 11: 545-554.
        http://dx.doi.org/10.1007/s10021-008-9141-4.
Hildesheim. J: Fornace. AJ. Jr. (2004). The dark side of light: The damaging effects of UV rays and the
        protective efforts of  MAP kinase signaling in the epidermis. DMA Repair 3: 567-580.
Holick, MR (2004). Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers,
        and  cardiovascular disease. Am J Clin Nutr 80: 1678S-1688S.
Holland. MM: Bitz.  CM. (2003). Polar amplification of climate change in coupled models. Clim Dynam 21: 221-
        232. http://dx.doi.org/10.1007/s00382-003-0332-6.
Holtbv. LB: Bothwell. ML.  (2008). Effects of solar ultraviolet radiation on the behaviour of juvenile coho salmon
        (Oncorhynchus kisutch): Avoidance, feeding, and agonistic interactions. Can J Fish Aquat  Sci 65: 701-
        711. http://dx.doi.0rg/10.1139/F08-013.
Hughes. AM: Armstrong. BK: Vaidic.  CM: Turner. J: Grulich. AE:  Fritschi. L:  Milliken. S: Kaldor. J: Benke. G:
        Kricker. A. (2004). Sun exposure may protect against non-Hodgkin lymphoma: A case-control study. Int
        J Cancer 112: 865-871.
ICNIRP  (International Commission on Non-Ionizing Radiation Protection). (2004). Guidelines on limits of
        exposure to ultraviolet radiation of wavelengths between 180 nm and 400 nm (incoherent optical
        radiation). In ICNIRP Guidelines (Vol. 87, pp. 171-186). Oberschleissheim, Germany.
loki. M: Takahashi. S: Nakaiima. N: Fuiikura. K: Tamaoki. M:  Sail. H: Kubo. A: Aono. M: Kanna.  M: Ogawa. D:
        Fukazawa. J: Oda. Y: Yoshida. S: Watanabe. M: Hasezawa. S: Kondo. N. (2008). An unidentified
        ultraviolet-B-specific photoreceptor mediates transcriptional activation of the  cyclobutane pyrimidine
        dimer photolyase gene in plants. Planta 229: 25-36. http://dx.doi.org/10.1007/s00425-008-0803-4.
IPCC. (Intergovernmental Panel on Climate Change). (2000). Special report on emissions scenarios: A special
        report of Working Group III of the Intergovernmental  Panel on Climate Change. Cambridge, UK:
        Intergovernmental Panel on Climate Change; Cambridge University Press.
        http://www.grida.no/climate/ipcc/emission/.
IPCC. (Intergovernmental Panel on Climate Change). (2007b). Summary for policymakers. In: The Physical
        Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the
        Intergovernmental Panel on Climate Change.  In Climate Change 2007. Cambridge, United Kingdom
        and  New York, NY, USA: Cambridge University Press.
        http://www.ipcc.ch/publications  and data/ar4/wg2/en/spm.html.
Isaksen. ISA: Berntsen. TK: Wang. WC. (2001). NOx emissions from aircraft: Its impact on the global
        distribution of CH4 and O3 and on radiative forcing. Terr Atmos Ocean Sci 12: 63-78.
Ito. A: Sudo.  K: Akimoto. H: Sillman.  S: Penner. JE. (2007a). Global modeling analysis of tropospheric ozone
        and  its radiative forcing from biomass burning emissions in the twentieth century. J Geophys Res 112:
        D24307. http://dx.doi.0rg/10.1029/2007JD008745.
Draft - Do Not Cite or Quote                       10-33                                 September 2011

-------
Jaffe. D: Price. H: Parrish. D: Goldstein. A: Harris. J. (2003). Increasing background ozone during spring on the
        west coast of North America. Geophys Res Lett 30: 1613. http://dx.doi.org/10.1029/2003GL017024.
Jenkins. Gl. (2009). Signal transduction in responses to UV-B radiation. Annu Rev Plant Biol 60: 407-431.
        http://dx.doi.org/10.1146/annurev.arplant.59.032607.092953.
John, EM: Schwartz, GG:  Dreon, DM: Koo, J. (1999). Vitamin D and breast cancer risk: the NHANES I
        Epidemiologic Follow-up Study, 1971-1975 to 1992.  Cancer Epidemiol Biomarkers Prev 8: 399-406.
John. EM: Schwartz. GG:  Koo. J: Van Den Berg. D: Ingles. SA. (2005). Sun exposure, vitamin D receptor gene
        polymorphisms, and risk of advanced prostate cancer. Cancer Res 65: 5470-5479.
        http://dx.doi. orq/10.1158/0008-5472.CAN-04-3134.
Jokinen. IE: Markkula. ES: Salo. HM: Kuhn.  P: Nikoskelainen. S: Arts. MT: Browman. HI. (2008). Exposure to
        increased ambient ultraviolet B radiation has negative effects on growth, condition and immune
        function of juvenile Atlantic salmon (Salmo salar). Photochem Photobiol 84: 1265-1271.
        http://dx.doi.ora/10.1111/i.1751-1097.2008.00358.x.
Jonson, JE: Simpson, D: Fagerli, H: Solberg, S. (2005). Can we explain the trends in European  ozone levels?
        Atmos Chem Phys 6: 51-66. http://dx.doi.org/10.5194/acp-6-51-2006.
Joshi. M: Shine. KP: Ponater. M: Stuber. N: Sausen. R: Li. (2003). A comparison of climate response to different
        radiative forcings in three general circulation models: Towards an improved metric of climate change.
        Clim Dynam 20: 843-854. http://dx.doi.org/10.1007/s00382-003-0305-9.
Kar. J: Fishman. J: Creilson. JK: Richter. A: Ziemke. J: Chandra.  S. (2010). Are there urban signatures in the
        tropospheric ozone column products derived from satellite measurements? Atmos Chem Phys 10:
        5213-5222. http://dx.doi.org/10.5194/acp-10-5213-2010.
Kataoka. Y: Kiguchi. M: Williams. RS: Evans. PP. (2007). Violet light causes photodegradation of wood beyond
        the zone affected by ultraviolet radiation.  Holzforschung und Holzverwertung 61: 23-27.
        http://dx.doi.org/10.1515/HF.2007.005.
Kiehl. JT: Schneider. TL: Portmann.  RW: Solomon. S.  (1999). Climate forcing due to tropospheric and
        stratospheric ozone. J Geophys Res 104:  31239-31254.  http://dx.doi.org/10.1029/1999JD900991.
Kimlin. MG: Wong. JCF: Parisi. AV (1998). Simultaneous comparison  of the personal UV exposure of two
        human groups at different altitudes. Health Phys 74:  429-434.
Kloster, S: Dentener, F: Feichter, J: Raes, F: Lohmann, U: Roeckner,  E: Fischer-bruns,  I. (2009). A GCM study
        of future climate response to aerosol pollution reductions. Clim Dynam 34: 1177-1194.
        http://dx.doi. org/10.1007/s00382-009-0573-0.
Koronakis, PS: Sfantos, GK: Paliatsos, AG: Kaldellis, JK: Garofalakis,  JE: Koronaki, IP. (2002). Interrelations of
        UV-global/global/diffuse solar irradiance components and UV-global attenuation on air pollution episode
        days in Athens, Greece. Atmos Environ 36: 3173-3181.
Lacis, AA: Wuebbles, DJ:  Logan, JA. (1990). Radiative forcing of climate by changes in the vertical distribution
        of ozone. J Geophys Res 95: 9971-9981. http://dx.doi.org/10.1029/JD095iD07p09971.
Lamargue. JF:  Hess. P: Emmons.  L: Buia. L: Washington. W: Granier. C. (2005). Tropospheric ozone evolution
        between 1890 and 1990.  J Geophys Res 110: D08304. http://dx.doi.org/10.1029/2004JD005537.
Lamargue. JF:  Bond. TC:  Evring. V: Granier. C: Hell. A: Klimont. Z: Lee. D: Liousse. C:  Mieville.  A: Owen. B:
        Schultz. MG: Shindell. D:  Smith. SJ: Stehfest. E: Van Aardenne. J: Cooper. OR: Kainuma. M:
        Mahowald, N: McConnell, JR: Naik, V: Riahi, K: van  Vuuren, DP. (2010). Historical (1850-2000)
        gridded anthropogenic and biomass burning emissions of reactive gases and aerosols:  Methodology
        and application. Atmos Chem Phys Discuss 10:  4963-5019. http://dx.doi.org/10.5194/acpd-10-4963-
        2010.
Lee.  EH: Tingev. DT: Hogsett. WE: Laurence. JA. (2003a).  History of tropospheric ozone for the San Bernardino
        Mountains of southern California, 1963-1999. Atmos Environ 37: 2705-2717.
        http://dx.doi. org/10.1016/51352-2310(03)00203-6.
Lefkowitz. ES: Garland.  CF. (1994). Sunlight, vitamin D, and ovarian cancer mortality rates in US women. Int J
        Epidemiol 23: 1133-1136.
Lelieveld, J: van Aardenne, J: Fischer, H: de Reus, M: Williams, J: Winkler, P. (2004). Increasing ozone over the
        Atlantic Ocean. Science 304: 1483-1487. http://dx.doi.org/10.1126/science. 1096777.
Lenoble. J. (1993). Atmospheric radiative transfer. In. Hampton, VA: A. Deepak Publishing.
Draft - Do Not Cite or Quote                        10-34                                  September 2011

-------
Levy. H. II: Schwaarzkopf. MD: Horowitz. L: Ramaswamv. V: Findell. KL. (2008). Strong sensitivity of late 21st
        century climate to projected changes in short-lived air pollutants. J Geophys Res 113: D06102.
        http://dx.doi. orq/10.1029/2007JD009176.
Liao. H: Seinfeld. JH: Adams. PJ: Micklev. LJ. (2004b). Global radiative forcing of coupled tropospheric ozone
        and aerosols in a unified general circulation model. J Geophys Res 109: D16207.
        http://dx.doi. orq/10.1029/2003JD004456.
Lindelof. B: Sigurgeirsson. B: Gabel. H: Stern. RS. (2000). Incidence of skin cancer in 5356 patients following
        organ transplantation. Br J Dermatol 143: 513-519.
Logan. JA: Megretskaia.  IA: Miller. AJ: Tiao. GC: Choi. D: Zhang. L: Stolarski. RS: Labow. GJ: Hollandsworth.
        SM: Bodeker.  GE:  Claude. H: De Muer. D: Kerr. JB: Tarasick. DW: Oltmans.  SJ: Johnson. B: Schmidlin.
        F: Staehelin, J: Viatte, P: Uchino, O. (1999). Trends in the vertical distribution of ozone: A comparison
        of two analyses of ozonesonde data. J Geophys Res 104: 26373-26399.
        http://dx.doi. orq/10.1029/1999JD900300.
Longstreth, J: de Gruijl, FR: Kripke, ML: Abseck, S: Arnold,  F:  Slaper,  HI: Velders, G:  Takizawa, Y: van der
        Leun. JC. (1998). Health risks. J Photochem Photobiol B 46:  20-39.
Longstreth. JD: Gruiil. D: Kripke. ML: Takizawa. Y: van der Leun. JC. (1995). Effects of increased solar
        ultraviolet radiation on human health. Ambio 24: 153-165.
Lutter. R: Wolz. C. (1997). UV-B screening  by tropospheric ozone: Implications for the national ambient air
        quality standard. Environ Sci Technol 31: 142A-146A.
Madronich, S: De Gruiil, F. (1993). Skin cancer and  UV radiation [Letter/Response]. Nature 366: 23.
        http://dx.doi. orq/10.1038/366023aO.
Madronich. S: Wagner. M: Groth. P. (2011).  Influence of tropospheric ozone control on exposure to ultraviolet
        radiation at the surface.  Environ Sci Technol 45: 6919-6923. http://dx.doi.org/10.1021/es200701q.
Marenco. A:  Gouget. H: Nedelec. P: Pages. J. -P: Karcher.  F. (1994).  Evidence of a long-term increase in
        tropospheric ozone from Pic du Midi data series: Consequences: Positive radiative forcing. J Geophys
        Res 99:  16617-16632. http://dx.doi.org/10.1029/94JD00021.
Markkula. E: Salo. HM: Rikalainen. K: Jokinen.  IE. (2009). Long-term  UVB irradiation affects the immune
        functions of carp (Cyprinus carpio)  and rainbow trout (Oncorhynchus mykiss). Photochem Photobiol 85:
        347-352. http://dx.doi.0rg/10.1111/i.1751-1097.2008.00446.x.
Marguis. O: Miaud. C: Lena. JP (2008). Developmental responses to  UV-B radiation  in common frog Rana
        temporaria embryos from along an  altitudinal  gradient. Population  Ecology 50: 123-130.
        http://dx.doi.org/10.1007/s10144-007-0071-3.
Marguis. O: Miaud. C. (2008). Variation in UV sensitivity among common frog Rana temporaria populations
        along an altitudinal gradient. Zoology (Jena) 111: 309-317. http://dx.doi.Org/10.1016/i.zool.2007.09.003.
Matsumura, Y: Ananthaswamy, HN. (2004).  Toxic effects of ultraviolet radiation on the skin. Toxicol Appl
        Pharmacol 195:  298-308.
Mayer. LM: Schick. LL: Hardy. KR: Estapa. ML. (2009). Photodissolution and other photochemical changes
        upon irradiation of algal  detritus. Limnol Oceanogr 54: 1688-1698.
Mazza. CA: Izaguirre. MM:  Curiale. J: Ballare. CL. (2010). A look into the invisible: Ultraviolet-B sensitivity in an
        insect (Caliothrips  phaseoli) revealed through a behavioural action spectrum. Proc Biol Sci 277: 367-
        373.  http://dx.doi.0rg/10.1098/rspb.20Q9.1565.
Meador. JA:  Baldwin. AJ: Catala. P: Jeffrey. WH: Joux. F: Moss. JA: Pakulski. JD: Stevens. R: Mitchell. PL.
        (2009). Sunlight-induced DMA damage in marine micro-organisms collected  along a latitudinal gradient
        from 70 degrees N to 68 degrees S. Photochem Photobiol 85: 412-421.
        http://dx.doi.0rg/10.1111/i.1751-1097.2008.00462.x.
Micklev. LJ: Leibensperger. EM: Jacob. DJ:  Rind. D. (In Press) Regional warming from aerosol  removal over the
        United States: Results from a transient 2010-2050 climate simulation. Atmos Environ.
        http://dx.doi.0rg/10.1016/i.atmosenv.2011.07.030.
Micklev. LJ: Murti. PP: Jacob. DJ: Logan. JA: Koch. DM: Rind. D. (1999). Radiative forcing from tropospheric
        ozone calculated with a  unified chemistry-climate model. J Geophys Res 104: 30153-30172.
        http://dx.doi. orq/10.1029/1999JD900439.
Micklev. LJ: Jacob. DJ: Rind. D. (2001). Uncertainty in preindustrial abundance of tropospheric ozone:
        Implications for radiative forcing calculations.  J Geophys Res 106: 3389-3399.
        http://dx.doi. orq/10.1029/2000JD900594.
Draft - Do Not Cite or Quote                        10-35                                 September 2011

-------
Micklev. LJ: Jacob. DJ: Field. BD: Rind. D. (2004). Climate response to the increase in tropospheric ozone since
        preindustrial times: A comparison between ozone and equivalent CO2 forcings. J Geophys Res 109:
        D05106. http://dx.doi.orq/10.1029/2003JD003653.
Moehrle. M: Heinrich. L: Schmid. A: Garbe. C. (2000). Extreme UV exposure of professional cyclists.
        Dermatology 201: 44-45.
Moehrle. M. (2001). Ultraviolet  exposure in the Ironman triathlon. Med Sci Sports Exerc 33: 1385-1386.
Moise. AF: Buttner. PG: Harrison. SL. (1999). Sun exposure at school. Photochem Photobiol 70: 269-274.
Naik, V: Mauzerall, D: Horowitz, L: Schwarzkopf, MD: Ramaswamy, V: Oppenheimer, M. (2005). Net radiative
        forcing due to changes in regional emissions of tropospheric ozone precursors. J Geophys Res 110:
        D24306. http://dx.doi.orq/10.1029/2005JD005908.
Naja,  M: Akimoto, H. (2004). Contribution of regional pollution and long-range transport to the Asia-Pacific
        region: Analysis of long-term ozonesonde data over Japan. J Geophys Res 109: D21306.
        http://dx.doi. orq/10.1029/2004JD004687.
Nole,  G: Johnson, AW. (2004).  An analysis of cumulative lifetime solar ultraviolet radiation exposure and the
        benefits of daily sun protection. Dermatol Ther 17: 57-62.
Norval. M: Garssen. J: Van Loveren. H: EI-Ghorr. AA. (1999). UV-induced changes in the immune response to
        microbial infections in  human subjects and animal models. J Epidemiol 6: S84-S92.
NRC.  (National Research Council). (2005). Radiative forcing of climate change: Expanding the concept and
        addressing uncertainties. In. Washington, DC: The National Academies Press.
Obara, Y: Koshitaka, H: Arikawa, K.  (2008). Better mate in the shade: Enhancement of male mating behaviour
        in the cabbage butterfly, Pieris rapae crucivora, in a UV-rich environment. J Exp Biol 211: 3698-3702.
        http://dx.doi.org/10.1242/ieb.021980.
Okada, S: Weatherhead, E: Targoff,  IN: Wesley, R:  Miller, FW: Group, IMCS. (2003). Global surface ultraviolet
        radiation intensity may modulate the clinical and immunologic expression of autoimmune muscle
        disease. Arthritis Rheum 48: 2285-2293.
Oltmans, SJ: Lefohn, AS: Harris, JM: Galbally, I: Scheel, HE: Bodeker, G: Brunke, E: Claude,  H: Tarasick, D:
        Johnson. BJ: Simmonds. P: Shadwick.  D: Anlauf. K: Havden. K: Schmidlin. F: Fuiimoto. T: Akagi. K:
        Meyer. C: Nichol. S: Davies. J:  Redondas.  A: Cuevas. E. (2006). Long-term changes in tropospheric
        ozone. Atmos Environ 40: 3156-3173. http://dx.doi.Org/10.1016/i.atmosenv.2006.01.029.
Ordonez. C: Brunner. D: Staehelin. J: Hadiinicolaou. P:  Pvle. JA: Jonas.  M: Wernli. H: Prevot. ASH. (2007).
        Strong influence of lowermost stratospheric ozone on lower tropospheric  background  ozone changes
        over Europe. Geophys Res Lett 34: L07805. http://dx.doi.org/10.1029/2006GL029113.
Oromi. N: Marguis. O: Miaud. C: Sanuv. D. (2008).  Influence of ambient ultraviolet radiation on Bufo calamita
        egg development in a  semiarid zone (Catalonia, Spain). J Environ Biol 29: 135-137.
Palancar, GG: Toselli, BM. (2002). Erythemal ultraviolet irradiance in Cordoba, Argentina. Atmos Environ 36:
        287-292.
Pavelin. EG: Johnson. CE: Rughooputh. S: Toumi.  R. (1999). Evaluation of pre-industrial surface ozone
        measurements made using Schonbein's method. Atmos Environ 33: 919-929.
        http://dx.doi. org/10.1016/51352-2310(98)00257-X.
Pfister. H. (2003). Human papillomavirus and skin cancer. J Natl Cancer Inst Monographs No.  31: 52-56.
Philipona, R: Behrens, K: Ruckstuhl, C. (2009). How declining aerosols and rising greenhouse gases forced
        rapid warming in Europe since the 1980s. Geophys Res Lett 36: L02806.
        http://dx.doi. org/10.1029/2008GL036350.
Phoenix, GK: Gwynn-Jones, D: Lee, JA: Callaghan, TV. (2000). The impacts of UV-B radiation on the
        regeneration of a sub-arctic heath community. Plant Ecol 146: 67-75.
        http://dx.doi. org/10.1023/A: 1009839506658.
Pickett, JE: Gibson, DA: Gardner, MM. (2008). Effects of irradiation conditions on the weathering of engineering
        thermoplastics. Polym Degrad Stabil 93: 1597-1606.
        http://dx.doi.0rg/10.1016/i.polvmdegradstab.2008.02.009.
Ponsonby, AL: McMichael, A: Van der Mei, I. (2002). Ultraviolet radiation and autoimmune disease: Insights
        from epidemiological research. Toxicology  181/182: 71-78.
Repapis. CC: Mantis. HT:  Paliatsos. AG: Philandras. CM: Bais. AF: Meleti. C. (1998). Case study of UV-B
        modification during episodes of urban air pollution. Atmos Environ 38: 2203-2208.
Draft - Do Not Cite or Quote                       10-36                                 September 2011

-------
Riqel. PS: Riqel. EG: Riqel. AC. (1999). Effects of altitude and latitude on ambient UVB radiation. J Am Acad
        Dermatol40: 114-116.
Riggsbee. JA: Orr. CH: Leech. DM: Doyle. MW: Wetzel. RG. (2008). Suspended sediments in river ecosystems:
        Photochemical sources of dissolved organic carbon, dissolved organic nitrogen, and adsorptive
        removal of dissolved iron. J Geophys Res 113: G03019. http://dx.doi. org/10.1029/2007ig000654.
Rind. D: Healv. R:  Parkinson. C: Martinson. D. (1995). The role of sea ice in 2>2.0.CO:2.
Romansic. JM: Waggener. AA: Bancroft. BA: Blaustein. AR. (2009). Influence of ultraviolet-B radiation on
        growth, prevalence of deformities, and susceptibility to predation in Cascades frog (Rana cascadae)
        larvae. Hydrobiologia 624: 219-233. http://dx.doi.org/10.1007/s10750-009-9703-2.
Rosenthal. FS: Phoon. C: Bakalian. AE: Taylor.  HR. (1988). The ocular dose of ultraviolet radiation to outdoor
        workers. Invest Ophthalmol Vis Sci 29: 649-656.
Ruckstuhl, C:  Philipona, R: Behrens, K: Coen,  MC: Purr, B: Heimo, A: Matzler, C: Nyeki, S: Ohmura, A:
        Vuilleumier. L: Weller. M: Wehrli. C: Zelenka. A. (2008). Aerosol and cloud effects on solar brightening
        and recent rapid warming. Geophys Res Lett 35: L12708. http://dx.doi.orq/10.1029/2008GL034228.
Schenker, MB: Orenstein, MR: Samuels, SJ. (2002). Use of protective equipment among California farmers. Am
        J Ind  Med 42: 455-464.
Schindler. DW: Curtis. PJ: Parker. BR: Stainton. MP (1996). Consequences of climate warming and  lake
        acidification for UV-B penetration in North American boreal lakes. Nature 379: 705-708.
        http://dx.doi. orq/10.1038/379705aO.
Selgrade. M. -JK: Smith. MV: Oberhelman-Bragg. LJ: LeVee. GJ: Koren. HS: Cooper. KD. (2001). Dose
        response  for UV-induced  immune suppression in people of color: Differences based on erythemal
        reactivity rather than skin pigmentation. Photochem Photobiol 74: 88-95.
Selgrade. MK: Repacholi. MH: Koren. HS. (1997). Ultraviolet radiation-induced immune modulation:  Potential
        consequences for infectious, allergic, and autoimmune disease. Environ Health Perspect 105: 332-334.
Semerdiieva. SI: Phoenix. GK: Hares. D: Gwvnn-Jones. D: Callaghan. TV: Sheffield. E. (2003). Surface
        morphology, leaf and cuticle thickness  of four dwarf shrubs from a sub-Arctic heath following long-term
        exposure  to enhanced levels of UV-B. Physiol Plant 117: 289-294. http://dx.doi.org/10.1034/j. 1399-
        3054.2003.00006.x.
Shindell. D: Faluvegi. G: Lacis. A:  Hansen. J: Ruedv. R: Aguilar. E. (2006). Role of tropospheric ozone
        increases in 20th-century climate change.  J Geophys Res 111: D08302.
        http://dx.doi. orq/10.1029/2005JD006348.
Shindell. D: Faluvegi. G. (2009). Climate response to regional radiative forcing during the twentieth century. Nat
        Geosci 2:  294-300. http://dx.doi.Org/10.1038/ngeo473.
Shindell. DT: Faluvegi. G. (2002). An exploration of ozone changes and their radiative forcing prior to the
        chlorofluorocarbon era. Atmos Chem Phys Discuss 2: 363-374. http://dx.doi.org/10.5194/acp-2-363-
        2002.
Shindell. DT: Faluvegi. G: Bell. N.  (2003). Preindustrial-to-present-day radiative forcing by tropospheric ozone
        from improved simulations with the GISS chemistry-climate GCM. Atmos Chem Phys 3: 1675-1702.
        http://dx.doi.org/10.5194/acp-3-1675-2003.
Shindell. DT: Faluvegi. G: Bell. N:  Schmidt. GA. (2005). An emissions-based view of climate  forcing by methane
        and tropospheric ozone. Geophys Res Lett 32: L04803. http://dx.doi.org/10.1029/2004GL021900.
Shindell, DT: Faluvegi, G: Bauer, SE: Koch,  DM: Unger, N: Menon, S: Miller, RL: Schmidt, GA:  Streets, DG.
        (2007). Climate response to projected changes in  short-lived species under an A1B  scenario from
        2000-2050 in the GISS climate model.  J Geophys Res 112: D20103.
        http://dx.doi. org/10.1029/2007id008753.
Shindell. DT: Lew H. II: Schwarzkopf.  MD:  Horowitz. LW: Lamargue. JF:  Faluvegi. G. (2008). Multimodel
        projections of climate change from short-lived emissions due to human activities. J Geophys Res 113:
        D11109. http://dx.doi.org/10.1029/2007JD009152.
Shoveller. JA:  Lovato. CY: Peters. L: Rivers. JK. (1998). Canadian national survey on sun exposure  & protective
        behaviours: Adults at leisure. Cancer Prev Control 2: 111-116.
Draft - Do Not Cite or Quote                        10-37                                  September 2011

-------
Simmonds. PG: Derwent. RG: Manning. AL: Spain. G. (2004). Significant growth in surface ozone at Mace
        Head, Ireland, 1987-2003. Atmos Environ 38: 4769-4778.
        http://dx.doi.0rg/10.1016/i.atmosenv.2004.04.036.
Sitch. S: Cox. PM: Collins. WJ: Huntingford. C. (2007). Indirect radiative forcing of climate change through
        ozone effects on the land-carbon sink. Nature 448: 791-794. http://dx.doi.org/10.1038/nature06059.
Slaper. H: Velders. GJM: Daniel. JS: de Gruiil. FR: Van der Leun. JC. (1996). Estimates of ozone depletion and
        skin cancer incidence to examine the Vienna Convention achievements. Nature 384: 256-258.
Slinev, DH: Wengraitis, S. (2006). Is a differentiated advice by season and region necessary? Prog Biophys Mol
        Biol 92: 150-160. http://dx.doi.Org/10.1016/i.pbiomolbio.2006.02.007.
Smedbv. KE: Hialgrim. H: Melbve. M: Torrang. A: Rostgaard. K: Munksgaard. L: Adami. J:  Hansen. M: Porwit-
        MacDonald, A: Jensen, BA: Roos, G: pedersen, BB: Sundstrom, C: Glimelius, B: Adami, H, -O. (2005).
        Ultraviolet radiation exposure and risk of malignant lymphomas. J Natl Cancer Inst 97: 199-209.
Snell. KRS: Kokubun. T: Griffiths. H: Convey. P: Hodgson. DA: Newsham. KK.  (2009). Quantifying the metabolic
        cost to an Antarctic liverwort of responding to an abrupt increase in UVB radiation exposure. Global
        Change Biol 15: 2563-2573. http://dx.doi.Org/10.1111/i.1365-2486.2009.01929.x.
Soden. BJ:  Held. IM. (2006). An assessment of climate feedbacks in coupled ocean-atmosphere models. J Clim
        19: 3354-3360.
Staehelin. J: Thudium. J: Buehler. R: Volz-Thomas. A: Graber. W. (1994). Trends in surface ozone
        concentrations at Arosa (Switzerland). Atmos Environ 28: 75-87. http://dx.doi.org/10.1016/1352-
        2310(94)90024-8.
Stevenson. PS. (2004).  Radiative forcing from aircraft NO emissions: Mechanisms and seasonal dependence. J
        Geophys Res 109: D17307. http://dx.doi.org/10.1029/2004JD004759.
Stevenson, PS: Dentener, FJ: Schultz, MG: Ellingsen, K:  Van Noije, TPC: Wild, O: Zeng, G: Amann, M:
        Atherton. CS: Bell. N: Bergmann. DJ: Bey. I: Butler. T: Cofala. J: Collins. WJ: Derwent. RG: Dohertv.
        RM: Drevet. J: Eskes. HJ: Fiore. AM: Gauss. M: Hauglustaine. DA: Horowitz. LW: Isaksen. ISA: Krol.
        MC: Lamargue, JF: Lawrence, MG: Montanaro, V: Muller, JF: Pitari, G: Prather, MJ: Pyle, JA: Rast, S:
        Rodriguez. JM: Sanderson. MG: Savage. NH: Shindell. DT: Strahan. SE:  Sudo. K: Szopa. S. (2006).
        Multimodel ensemble simulations of present-day and near-future tropospheric ozone. J Geophys Res
        111: D08301. http://dx.doi.org/10.1029/2005JD006338.
Studzinski.  GP: Moore. DC. (1995). Sunlight—can it prevent as well  as cause cancer? Cancer Res 55: 4014-
        4022.
Tanimoto, H. (2009). Increase in springtime tropospheric ozone at a mountainous site in Japan for the period
        1998-2006. Atmos Environ 43: 1358-1363. http://dx.doi.Org/10.1016/i.atmosenv.2008.12.006.
Tarasick. DW: Fioletov. VE: Wardle. PI: Kerr. JB: Davies. J. (2005). Changes in the vertical distribution of ozone
        over Canada from ozonesondes: 1980-2001. J Geophys Res 110: D02304.
        http://dx.doi. org/10.1029/2004JD004643.
ten Berge. O: van Weelden. H: Bruiinzeel-Koomen. C: de Bruin-Weller. MS: Sigurdsson. V. (2009). Throwing a
        light on photosensitivity in atopic dermatitis: A retrospective study. Am  J Clin Dermatol 10: 119-123.
        http://dx.doi.org/10.2165/00128071-200910020-00004.
Thieden. E: Philipsen. PA: Sandbv-Moller. J:  Hevdenreich. J: Wulf. HC. (2004a). Proportion of lifetime UV dose
        received by children, teenagers and adults based on time-stamped personal dosimetry. J Invest
        Dermatol 123: 1147-1150.
Thieden. E: Philipsen. PA: Hevdenreich. J: Wulf. HC. (2004b). UV radiation exposure related to age, sex,
        occupation, and sun behavior based on time-stamped personal dosimeter readings. Arch Dermatol
        140: 197-203.
Thompson. AM. (1992). The oxidizing capacity of the Earth's atmosphere:  Probable past and future changes.
        Science 256: 1157-1165. http://dx.doi.org/10.1126/science.256.5060.1157.
Thompson. AM: Chappellaz. JA: Fung. IY: Kucsera. TL. (1993). The atmospheric CH4 increase since the last
        glacial maximum (2) Interactions with oxidants. Tellus B Chem Phys Meteorol 45: 242-257.
        http://dx.doi. org/10.1034/1.1600-0889.1993.t01 -2-00003.X.
Thompson. AM: Hudson. RD. (1999). Tropical tropospheric ozone (TTO) maps from  Nimbus 7 and Earth Probe
        TOMS by the modified-residual method: Evaluation with sondes, ENSO signals, and trends from
        Atlantic regional time series. J Geophys Res 104: 26961-26975.
        http://dx.doi. org/10.1029/1999JD900470.
Draft - Do Not Cite or Quote                       10-38                                 September 2011

-------
Thompson. AM: Stone. JB: Witte. JC: Miller. SK: Oltmans. SJ: Kucsera. TL: Ross. KL: Pickering. KE: Merrill. JT:
        Forbes, G: Tarasick, DW: Joseph, E:  Schmidlin,  FJ: McMillan, WW: Warner, J: Hintsa, EJ: Johnson, JE.
        (2007). Intercontinental Chemical Transport Experiment Ozonesonde Network study (IONS) 2004: 2
        Tropospheric ozone budgets and variability over northeastern North America. J Geophys Res 112:
        D12S13. http://dx.doi.org/10.1029/2006JD007670.
U.S. EPA. (U.S. Environmental Protection Agency). (2003). National ambient  air quality standards for ozone:
        Final response to remand. Fed Reg 68: 614-645.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
        photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
        Protection Agency, Office of Research and Development.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
Ullrich. SE. (2005).  Mechanisms underlying UV-induced immune suppression. Mutat Res 571: 185-205.
        http://dx.doi.ora/10.1016/i.mrfmmm.2004.06.059.
UNEP. (United Nations Environment Programme). (2009). Environmental effects of ozone depletion  and its
        interactions with climate change. Nairobi, Kenya.
        http://ozone.unep.org/Assessment Panels/EEAP/EEAP-Proqress-report-2009.pdf.
Unger, N. (2006). Cross influences of ozone and sulfate precursor emissions  changes on air quality  and climate.
        PNAS 103: 4377-4380. http://dx.doi.orq/10.1073/pnas.05Q8769103.
Unger. N: Shindell. DT: Koch. DM: Streets. DG. (2008). Air pollution radiative  forcing from specific emissions
        sectors at 2030. J Geophys Res 113: D02306. http://dx.doi.org/10.1029/2007JD008683.
Unger. N: Bond. TC: Wang. JS:  Koch. DM: Menon. S: Shindell. DT: Bauer. S.  (2010). Attribution of climate
        forcing to economic sectors.  PNAS 107:  3382-3387. http://dx.doi.orq/10.1073/pnas.0906548107.
Urbach, F (1997). Ultraviolet radiation and skin cancer of humans.  J Photochem Photobiol B 40: 3-7.
Van Aardenne. JA: Dentener. FJ: Olivier. JGJ: Klein Goldewiik. CGM: Lelieveld. J. (2001). A 1 x 1 resolution data
        set of historical anthropogenic trace gas emissions for the  period 1890-1990. Global Biogeochem
        Cycles 15: 909-928. http://dx.doi.org/10.1029/2000GB001265.
Vishvakarman. D: Wong. JCF: Boreham. BW. (2001). Annual occupational exposure to ultraviolet radiation in
        central Queensland. Health  Phys 81: 536-544.
Volz, A: Kiev, D. (1988). Evaluation of the Montsouris series  of ozone measurements made in the nineteenth
        century. Nature 332: 240-242. http://dx.doi.org/10.1038/332240aO.
Wahl. M. (2008). Ecological modulation of environmental stress: Interactions between ultraviolet radiation,
        epibiotic snail embryos, plants and herbivores. J Anim Ecol 77: 549-557.
        http://dx.doi.0rg/10.1111/i.1365-2656.2007.01352.x.
Wang. W. -C: Pinto. JP: Yung. YL. (1980). Climatic effects due to halogenated compounds in the earth's
        atmosphere. J Atmos Sci  37: 333-338. http://dx.doi.org/10.1175/1520-
        0469(1980)037<0333:CEPTHC>2.0.CO:2.
West. JJ: Fiore. AM. (2005). Management of tropospheric ozone by reducing  methane emissions. Environ Sci
        Technol39: 4685-4691.
West. JJ: Fiore. AM: Horowitz. LW: Mauzerall. PL. (2006). Global health benefits of mitigating ozone pollution
        with methane emission controls. PNAS 103: 3988-3993. http://dx.doi.org/10.1073/pnas.0600201103.
West, JJ: Fiore, AM: Naik, V: Horowitz, LW: Schwarzkopf, MD: Mauzerall, PL. (2007). Ozone air quality and
        radiative forcing consequences of changes in ozone precursor emissions. Geophys Res Lett 34:
        L06806. http://dx.doi.org/10.1029/2006GL029173.
Wild, O: Prather, MJ: Akimoto, H. (2001). Indirect long-term global radiative cooling from  NOX emissions.
        Geophys Res  Lett 28:  1719-1722. http://dx.doi.org/10.1029/2000GL012573.
Worden. HM: Bowman. KW: Worden. JR: Eldering. A: Beer.  R. (2008). Satellite measurements of the clear-sky
        greenhouse effect from tropospheric ozone. Nat Geosci 1:  305-308. http://dx.doi.org/10.1038/ngeo182.
Yang. X: Cox. RA: Warwick. NJ: Pvle. JA: Carver. GP:  O'connor. FM: Savage. NH. (2005c). Tropospheric
        bromine chemistry and its impacts on ozone: A model study. J Geophys  Res 110: P23311.
        http://dx.doi. org/10.1029/2005JP006244.
Zepp. RG: Erickson. PJ:  Paul. NP: Sulzberger. B. (2007). Interactive effects of solar UV radiation  and climate
        change on biogeochemical cycling. In The Environmental Effects of Ozone Pepletion and its
        Interactions with Climate Change: 2006 Assessment. Nairobi , Kenya: United Nations Environment
        Programme.
Draft - Po Not Cite or Quote                       10-39                                 September 2011

-------
Zepp. RG: Shank. GC: Stabenau. E: Patterson. KW: Cvterski. M: Fisher. W: Bartels. E: Anderson. SL. (2008).
        Spatial and temporal variability of solar ultraviolet exposure of coral assemblages in the Florida Keys:
        Importance of colored dissolved organic matter. Limnol Oceanogr 53: 1909-1922.
Zerefos. CS: Kourtidis. KA: Melas. D: Balis. D: Zanis. P: Katsaros. L: Mantis. HT: Repapis. C:  Isaksen. I:
        Sundet, J: Herman, J: Bhartia, PK: Calpini, B. (2002). Photochemical activity and solar ultraviolet
        radiation (PAUR) modulation factors: An overview of the project. J Geophys Res 107: 8134.
        http://dx.doi. orq/10.1029/2000JD000134.
Ziemke, JR: Chandra, S: Bhartia, PK. (2005). A 25-year data record of atmospheric ozone in the Pacific from
        Total Ozone Mapping Spectrometer (TOMS) cloud slicing: Implications for ozone trends in the
        stratosphere and troposphere. J Geophys Res 110: D15105. http://dx.doi.orq/10.1029/2004JD005687.
Draft - Do Not Cite or Quote                        10-40                                  September 2011

-------
11     REFERENCES
Abbey. DE: Nishino. N: McDonnell. WF: Burchette. RJ: Knutsen. SF: Beeson. WL: Yang. JX. (1999). Long-term
      inhalable particles and other air pollutants related to mortality in nonsmokers. Am J Respir Grit Care Med
      159: 373-382.
Abraham, WM: Delehunt, JC: Yerger, l_: Marchette, B: Oliver W, J, r. (1984). Changes in airway permeability and
      responsiveness after exposure to ozone. Environ Res 34: 110-119.
Acker, K: Febo, A: Trick, S: Perrino, C:  Bruno, P: Wiesen, P:  Moller; Wieprecht, W: Auel, R: Giusto, M: Geyer, A:
      Platt. U: Allegrini. I. (2006). Nitrous acid in the urban area of Rome. Atmos Environ 40: 3123-3133.
      http://dx.doi.0rg/10.1016/i.atmosenv.2006.01.028.
Acosta, LR: Evans, WFJ. (2000). Design of the Mexico City UV monitoring network: UV-B measurements at
      ground level in the urban environment. J Geophys Res 105: 5017-5026.
      http://dx.doi. org/10.1029/1999JD900250.
Adamkiewicz, G: Ebelt, S: Syring, M: Slater, J: Speizer, FE: Schwartz, J: Suh, H: Gold, DR. (2004). Association
      between air pollution exposure and exhaled nitric oxide  in an elderly population. Thorax 59: 204-209.
      http://dx.doi.org/10.1136/thorax.2003.006445.
Adamowicz, V: Dales, R: Hale, BA: Hrudey, SE: Krupnick, A: Lippman, M:  McConnell, J: Renzi, P. (2004).
      Report of an expert panel to review the socio-economic models and related components supporting the
      development of Canada-Wide Standards (CWS) for particulate matter (PM) and ozone to the Royal
      Society of Canada. J Toxicol Environ Health B Grit Rev 7: 147-266.
      http://dx.doi. org/10.1080/10937400490253238.
Adams, WC: Schelegle, ES.  (1983). Ozone and high ventilation effects on pulmonary function and endurance
      performance. J Appl Physiol 55: 805-812.
Adams, WC: Ollison, WM. (1997). Effects of prolonged simulated ambient ozone dosing patterns on human
      pulmonary function and symptomatology. In Presented at: 90th annual meeting of the Air & Waste
      Management Association; June; Toronto, Ontario, Canada Pittsburgh, PA: Air & Waste Management
      Association; paper no 97-MP902.
Adams, WC. (1998).  Dose-response effect of varied equivalent minute ventilation rates on pulmonary function
      responses during exposure to ozone. Washington, DC: American Petroleum Institute.
Adams. WC. (2002).  Comparison of chamber and face-mask 6.6-hour exposures to ozone on pulmonary
      function and symptoms responses. Inhal Toxicol 14: 745-764.
Adams, WC. (2003a). Comparison of chamber and face mask 6.6-hour exposure to 0.08 ppm ozone via square-
      wave and triangular profiles on pulmonary responses. Inhal Toxicol 15: 265-281.
Adams, WC. (2003b). Relation of pulmonary responses induced by 66-h exposures to 008 ppm ozone and 2-h
      exposures to 030 ppm ozone via chamber and face-mask inhalation. Inhal Toxicol 15: 745-759.
Adams. WC. (2006a). Comparison of chamber 6.6-h exposures to 0.04-0.08 PPM ozone via square-wave and
      triangular profiles on pulmonary responses. Inhal Toxicol 18: 127-136.
      http://dx.doi. org/10.1080/08958370500306107.
Adams. WC. (2006b). Human pulmonary responses with 30-minute time intervals of exercise and rest when
      exposed for 8 hours to 0.12 ppm ozone via square-wave and acute triangular profiles. Inhal Toxicol 18:
      413-422. http://dx.doi.org/10.1080/08958370600563599.
Agarwal, A; Saleh, RA; Bedaiwy, MA. (2003). Role of reactive oxygen species in the pathophysiology of human
      reproduction. Fertil Steril 79: 829-843.
Agrell, J; Kopper, BJ; McDonald, EP; Lindroth, RL (2005). CO2 and O3 effects on host plant preferences of the
      forest tent caterpillar (Malacosoma disstria). Global Change Biol 11: 588-599.
      http://dx.doi.0rg/10.1111/i.1365-2486.2005.00924.x.
Ahlfors, R; Brosche, M; Kollist,  H; Kangasjarvi, J. (2009).  Nitric oxide modulates ozone-induced cell death,
      hormone biosynthesis and gene expression in Arabidopsis thaliana. Plant J 58: 1-12.
      http://dx.doi.Org/10.1111/i.1365-313X.2008.03756.x.
Ahmad, S; Ahmad, A; McConville, G; Schneider, BK; Allen, CB; Manzer, R; Mason, RJ; White, CW. (2005). Lung
      epithelial cells release ATP during ozone exposure: Signaling for cell survival. Free Radic Biol Med  39:
      213-226. http://dx.doi.0rg/10.1016/i.freeradbiomed.2005.03.009.
Ahsan, N;  Nanjo, Y; Sawada, H; Kohno, Y; Komatsu, S. (2010). Ozone stress-induced proteomic changes in
      leaf total soluble and chloroplast proteins of soybean reveal that carbon allocation is involved in
      adaptation in the early developmental stage.  Proteomics 10: 2605-2619.
      http://dx.doi. org/10.1002/pmic.2Q1000180.
Aibo. PI: Birmingham. NP: Lewandowski. R:  Maddox. JF: Roth. RA: Ganev. PE: Wagner. JG: Harkema.  JR.
      (2010). Acute exposure to ozone exacerbates acetaminophen-induced liver injury in mice. Toxicol Sci
      115: 267-285. http://dx.doi.org/10.1093/toxsci/kfg034.
Ainsworth, EA; Long, SP (2005). What have we learned from 15 years of free-air CO2 enrichment (FACE)? A
      meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising
      CO2 [Review]. New Phytol 165: 351-371.
Draft - Do Not Cite or Quote                        11-1                                  September 2011

-------
Ainsworth, EA: Rogers, A. (2007). The response of photosynthesis and stomatal conductance to rising [CO2]:
      Mechanisms and environmental interactions [Review]. Plant Cell Environ 30: 258-270.
      http://dx.doi.0rg/10.1111/i.1365-3040.2007.01641.x.
Ainsworth, EA. (2008). Rice production in a changing climate: a meta-analysis of responses to elevated carbon
      dioxide and elevated ozone concentration. Global Change Biol 14: 1642-1650.
      http://dx.doi. org/10.1111/i.1365-2486.2008.01594.x.
Airev, DK: Wong, JCF: Fleming, RA: Meldrum, LR. (1997). An estimate of the UV-B exposure for outdoor
      workers during a south-east Queensland summer. Health Phys 72: 544-549.
Akinbami. LJ: Lynch. CD: Parker. JD: Woodruff. TJ. (2010). The association between childhood asthma
      prevalence and monitored air pollutants in metropolitan areas, United States, 2001-2004. Environ Res
      110: 294-301. http://dx.doi.Org/10.1016/i.envres.2010.01.001.
AI-Hegelan, M: Tighe, RM:  Castillo, C: Hollingsworth, JW. (2011). Ambient ozone and pulmonary innate
      immunity. Immunol Res 49: 173-191. http://dx.doi.Org/10.1007/s12026-010-8180-z.
Alexeeff.  SE: Litoniua. AA:  Suh. H: Sparrow. D: Vokonas. PS: Schwartz. J. (2007). Ozone exposure and lung
      function: Effect modified by obesity and airways hyperresponsiveness in the VA Normative Aging Study.
      Chest 132:  1890-1897.  http://dx.doi.org/10.1378/chest.07-1126.
Alexeeff.  SE: Litoniua. AA: Wright. RO: Baccarelli. A:  Suh. H: Sparrow. D: Vokonas. PS: Schwartz. J.  (2008).
      Ozone exposure, antioxidant genes, and lung  function in an elderly cohort: VA Normative Aging Study.
      Occup Environ Med  65: 736-742.  http://dx.doi.org/10.1136/oem.2007.035253.
Alexis, A: Garcia, A: Nvstrom, M: Rosenkranz, K.  (2001 a). The 2001 California almanac of emissions and air
      quality. Sacremento, CA: California Air Resources Board.
      http://www.arb.ca.gov/agd/almanac/almanac01/almanac01.htm.
Alexis. N: Urch. B: Tarlo. S: Corey. P: Pengellv. D: O'Bvrne. P: Silverman. F. (2000). Cyclooxygenase
      metabolites play a different role in ozone-induced pulmonary function decline in asthmatics compared to
      normals. Inhal  Toxicol 12:  1205-1224.
Alexis. N: Soukup. J:  Nierkens. S: Becker. S. (2001 b). Association between airway hyperreactivity and bronchial
      macrophage dysfunction in individuals with mild asthma. Am J Physiol  Lung Cell Mol Physiol 280: L369-
      L375.
Alexis, NE: Zhou, H: Lay, JC: Harris, B: Hernandez, ML: Lu, TS: Bromberg, PA:  Diaz-Sanchez,  D: Devlin,  RB:
      Kleeberger, SR: Peden, DB. (2009). The glutathione-S-transferase Mu 1  null genotype modulates ozone-
      induced airway inflammation in human subjects. J Allergy Clin Immunol 124: 1222-1228.
      http://dx.doi.0rg/10.1016/i.iaci.2009.07.036.
Alexis. NE: Lav. JC: Hazucha.  M: Harris. B: Hernandez.  ML: Bromberg. PA: Kehrl. H: Diaz-Sanchez.  D: Kim. C:
      Devlin, RB: Peden, DB. (2010). Low-level ozone exposure induces airways inflammation and modifies
      cell surface phenotypes in healthy humans. Inhal Toxicol 22: 593-600.
      http://dx.doi. orq/10.3109/08958371003596587.
Alfaro-Rodrfguez, A: Gonzalez-Pina, R. (2005). Ozone-induced paradoxical sleep decrease is related to
      diminished acetylcholine levels in  the medial preoptic area in rats. Chem  Biol Interact 151: 151-158.
      http://dx.doi.org/S0009-2797(04)00162-0 fpiil 10.1016/j.cbi.2004.10.001.
Alfaro, MF: Putney, L: Tarkington, BK:  Hatch, GE: Hyde, DM: Schelegle, ES. (2004). Effect of rapid shallow
      breathing on the distribution of 180-labeled ozone reaction product in the respiratory tract of the rat.
      Inhal Toxicol 16: 77-85.
Alfaro, MF: Walby, WF: Adams, WC: Schelegle, ES. (2007). Breath condensate  levels of 8-isoprostane and
      leukotriene B4 after  ozone inhalation are greater in sensitive versus nonsensitive subjects. Exp Lung Res
      33: 115-133. http://dx.doi.org/779284696 [pii]10.1080/01902140701364367.
Allen, EB: Temple, PJ: Bvtnerowicz, A: Arbaugh, MJ:  Sirulnik, AG: Rao, LE.  (2007).  Patterns of  understory
      diversity in mixed coniferous forests of southern California impacted by air pollution.
      ScientificWorldJournal 7: 247-263. http://dx.doi.org/10.1100/tsw.2007.72.
Alonso, R: Bermejo, V: Sanz, J: Vails,  B: Elvira, S: Gimeno, BS. (2007). Stomatal conductance  of semi-natural
      Mediterranean grasslands: Implications for the development of ozone critical levels. Environ Pollut 146:
      692-698. http://dx.doi.0rg/10.1016/i.envpol.2006.06.009.
Amthor, JS. (1988). Growth and maintenance respiration in leaves of bean (Phaseolus vulgaris L) exposed to
      ozone in open-top chambers in the field. New  Phytol 110: 319-325. http://dx.doi.orq/10.1111/1.1469-
      8137.1988.tb00268.x.
Andersen, CP: Wilson, R: Plocher, M:  Hogsett, WE. (1997). Carry-over effects of ozone on root growth and
      carbohydrate concentrations of ponderosa pine seedlings. Tree Physiol 17: 805-811.
Andersen, CP. (2003). Source-sink balance and carbon allocation below ground in plants exposed to ozone.
      New Phytol 157: 213-228.
Andersen. CP: Ritter.  W:  Gregg. J: Matvssek. R: Grams. TEE. (2010). Below-ground carbon allocation in
      mature beech and spruce trees following long-term, experimentally enhanced O3 exposure in  Southern
      Germany. Environ Pollut 158: 2604-2609. http://dx.doi.Org/10.1016/i.envpol.2010.05.008.
Anderson. GB: Bell. ML. (2010). Does one size fit all? The suitability of standard ozone exposure metric
      conversion  ratios and implications for epidemiology. J  Expo Sci Environ Epidemiol 20: 2-11.
      http://dx.doi. orq/10.1038/jes.2008.69.
Draft - Do Not Cite or Quote                        11-2                                  September 2011

-------
Anderson, HR: Armstrong, B: Hajat, S: Harrison, R: Monk, V: Poloniecki, J: Timmis, A: Wilkinson, P. (2010). Air
      pollution and activation of implantable cardioverter defibrillators in London. Epidemiology 21: 405-413.
      http://dx.doi.org/10.1097/EDE.Ob013e3181d61600.
Anderson, SE: Wells, JR: Fedorowicz, A: Butterworth, LF: Meade, BJ: Munson, AE. (2007). Evaluation of the
      contact and respiratory sensitization potential of volatile organic compounds generated by simulated
      indoor air chemistry. Toxicol Sci 97: 355-363. http://dx.doi.org/10.1093/toxsci/kfm043.
Andreae, MO. (1991). Biomass burning: its history, use, and distribution and its impact on environmental quality
      and global climate.  In JS Levine (Ed.), Global Biomass Burning: Atmospheric, Climatic, and Biospheric
      Implications (pp. 1-21). Cambridge, MA:  MIT Press.
Andrews, KM: Gibbons, JW: Jochimsen, DM. (2008). Ecological effects of roads on amphibians and reptiles: A
      literature review. In  JC Mitchell; REJ Brown; B Bartholomew (Eds.), Urban Herpetology (Vol. 3, pp. 121-
      143). Salt Lake City: Society for the Study of Amphibians and Reptiles.
Aneja, MK; Sharma,  S; Fleischmann,  F; Stich, S; Heller, W; Bahnweg, G;  Munch, JC; Schloter, M. (2007).
      Influence of ozone on litter quality and its subsequent effects on the initial structure of colonizing
      microbial communities. Microb Ecol 54: 151-160. http://dx.doi.org/10.1007/s00248-006-9183-0.
Angoa-Perez, M; Jiang, H; Rodrfguez, Al; Lemini, C; Levine, RA; Rivas-Arancibia, S. (2006). Estrogen
      counteracts ozone-induced oxidative stress and nigral neuronal death. Neuroreport 17: 629-633.
Anton. M: Lopez. M:  Vilaplana. JM: Kroon. M: McPeters. R: Banon. M: Serrano. A. (2009). Validation of OMI-
      TOMS and OMI-DOAS total ozone column using five Brewer spectroradiometers at the Iberian
      peninsula. J Geophys Res 114: D14307.  http://dx.doi.org/10.1029/2009JD012003.
Aoki, T; Tanabe, S. (2007). Generation of sub-micron particles and secondary pollutants from building materials
      by ozone reaction. Atmos Environ 41: 3139-3150. http://dx.doi.Org/10.1016/i.atmosenv.2006.07.053.
Aphalo. PJ: Vapaavuori. EM: de la Rosa. TM: Lehto. T. (2009). Does supplemental UV-B radiation affect gas
      exchange and RuBisCO activity of Betula pendula Roth, seedlings grown in forest soil under greenhouse
      conditions? Plant Ecol Divers 2: 37-43. http://dx.doi.org/10.1080/17550870902780299.
Appel. KW: Gilliland. A: Eder. B. (2005). An operational evaluation of the 2005 release of models-3 CMAQ
      version 45. Washington DC: National Oceanic and Atmospheric Administration-Air Resources.
Apte, MG; Buchanan, IS; Mendell, MJ. (2008). Outdoor ozone and building-related symptoms in the BASE
      study. Indoor Air 18: 156-170. http://dx.doi.Org/10.1111/i.1600-0668.2008.00521.x.
Araneda, S; Commin, L; Atlagich, M; Kitahama,  K; Parraguez, VH; Peguignot, JM; Dalmaz, Y. (2008). VEGF
      overexpression in the astroglial cells of rat brainstem following ozone exposure. Neurotoxicology 29:
      920-927. http://dx.doi.0rg/10.1016/i.neuro.2008.09.006.
Aranvi. C: Vana. SC: Thomas. PT: Bradof. JN: Renters.  JD: Graham. JA: Miller. FJ. (1983). Effects of
      subchronic exposure to a  mixture of O3,  SO2, and (NH4)2SO4 on  host defenses of mice. J Toxicol
      Environ Health 12: 55-71.
Arbaugh. M: Bvtnerowicz. A: Grulke. N: Fenn. M: Poth. M: Temple. P: Miller. P. (2003). Photochemical smog
      effects in mixed conifer forests along a natural gradient of ozone and nitrogen deposition in the San
      Bernardino Mountains. Environ Int29: 401-406. http://dx.doi.org/10.1016/50160-4120(02)00176-9.
Arbaugh, MJ; Miller,  PR; Carroll,  JJ; Takemoto,  BL; Proctor, T. (1998). Relationships of ozone exposure to pine
      injury in the Sierra Nevada and San  Bernardino Mountains of California,  USA. Environ Pollut 101: 291-
      301. http://dx.doi.org/10.1016/S0269-7491(98)00027-X.
Arbex. AM: de Souza Conceicao. GM: Perez Cendon. S: Arbex. FF: Lopes. AC: Providello Movses. E:  Santiago.
      SL; Saldiva, PHN; Pereira, LAA; Ferreira Braga, AL. (2009). Urban air pollution and COPD-related
      emergency room visits. J Epidemiol Community Health 966: 777-783.
      http://dx.doi.org/10.1136/iech.2008.078360.
Archibald, AT; Levine, JG;  Abraham, NL; Cooke, MC; Edwards, PM; Heard, DE; Jenkin,  ME; Karunaharan, A;
      Pike, RC; Monks, PS; Shallcross, DE; Telford, PJ; Whallev, LK; Pyle, JA. (2011). Impacts of HO x
      regeneration and recycling in the oxidation of isoprene: Consequences for the composition of past,
      present and future atmospheres. Geophys Res Lett 38: L05804.
      http://dx.doi. org/10.1029/201OGL046520.
Aris. RM: Christian. D: Hearne. PQ: Kerr. K: Finkbeiner. WE: Balmes. JR. (1993). Ozone-induced airway
      inflammation in human subjects as determined by airway lavage and biopsy. Am J Respir Grit Care Med
      148: 1363-1372.
Aris. RM: Tager. I: Christian. D: Kelly.  T: Balmes. JR. (1995). Methacholine responsiveness is not associated
      with OS-induced decreases in  FEV1. Chest 107:  621-628.
Arito, H; Uchiyama, I; Arakawa, H; Yokoyama, E. (1990). Ozone-induced bradycardia and arrhythmia and their
      relation to sleep-wakefulness in rats. Toxicol Lett 52: 169-178. http://dx.doi.org/10.1016/0378-
      4274(90)90151-6.
Arito. H: Uchivama. I: Yokovama. E. (1992). Acute effects of ozone on EEC activity, sleep-wakefulness and
      heart rate in rats. Ind Health 30: 23-34.
Arito, H; Takahashi, M; Iwasaki, T; Uchivama, I.  (1997). Age-related changes in ventilatory and heart rate
      responses to acute ozone exposure in the conscious rat. Ind Health 35:  78-86.
Draft - Do Not Cite or Quote                        11-3                                  September 2011

-------
Ariyaphanphitak, W: Chidthaisong, A: Sarobol, E: Bashkin, VN: Towprayoon, S. (2005). Effects of elevated
      ozone concentrations on Thai Jasmine rice cultivars (Oryza sativa L). Water Air Soil Pollut 167: 179-200.
      http://dx.doi.org/10.1007/s11270-005-8650-4.
Armstrong, BG. (2003). Fixed factors that modify the effects of time-varying factors: Applying the case-only
      approach. Epidemiology 14: 467-472.
Arnold. JR: Dennis. RL: Tonnesen. GS. (2003). Diagnostic evaluation of numerical air quality models with
      specialized ambient observations: testing the Community Multiscale Air Quality modeling system
      (CMAQ) at selected SOS 95 ground sites. Atmos Environ 37: 1185-1198.
Arrhenius. S. (1896). On the influence of carbonic acid in the air upon the temperature of the ground. Philos
      Mag 41: 237-276.
Arsalane, K: Gosset, P: Vanhee, D: Voisin,  C: Hamid, Q: Tonnel, A, -B: Wallaert,  B. (1995). Ozone stimulates
      synthesis of inflammatory cytokines  by alveolar macrophages in vitro. Am J Respir Cell Mol Biol 13: 60-
      68.
Arshinov. MY: Belan. BD: Krasnov. OA: Kovalevskii. VK: Pirogov. VA: Plotnikov. AP: Tolmachev. GN: Fofonov.
      AV. (2002).  Comparison of  ultraviolet and chemiluminescent ozonometers. Atmos Ocean 15: 656-658.
Ashmore, M: Emberson, L: Karlsson, PE: Pleijel, H. (2004a). Introduction for ozone deposition special issue.
      Atmos Environ 38: 2211-2212.
Ashmore. M: Emberson. L: Karlsson. PE: Pleiiel. H. (2004b). New directions: A new generation of ozone critical
      levels for the protection of vegetation in Europe (correspondence). Atmos Environ 38: 2213-2214.
Ashmore, MR:  Bell, JNB: Mimmack,  A. (1988). Crop growth along a gradient of ambient air pollution. Environ
      Pollut 53: 99-121. http://dx.doi.org/10.1016/0269-7491 (88)90028-0.
Ashmore, MR.  (2002).  Effects of oxidants at the whole plant and community level. In  JNB Bell;  M Treshow
      (Eds.), Air pollution and plant  life (pp. 89-118). London: Wiley.
ASHRAE. (American Society of Heating,  Refrigerating and Air-Conditioning Engineers Inc.). (2009). The 2009
      ASHRAE Handbook-Fundamentals. In. Atlanta, GA: American Society of Heating, Refrigerating and Air-
      Conditioning Engineers Inc.
Asplund. PT: Ben-Jebria. A:  Rigas. ML: Ultman. JS. (1996).  Longitudinal distribution of ozone absorption in the
      lung: Effect of continuous inhalation  exposure. Arch Environ Occup Health 51: 431-438.
Atkinson, RW:  Bremner, SA: Anderson, HR: Strachan,  DP: Bland, JM: Ponce de Leon, A. (1999).  Short-term
      associations between emergency  hospital admissions for respiratory and  cardiovascular disease and
      outdoor air pollution in London. Arch Environ Occup Health 54: 398-411.
ATMET (Atmospheric, Meteorological, and Environmental Technologies). (2011). Atmospheric, meteorological,
      and environmental technologies, from http://atmet.com/
ATS. (American Thoracic Society). (1991). Lung function testing: Selection of reference values and
      interpretative strategies. Am J Respir Grit Care Med 144: 1202-1218.
ATS. (American Thoracic Society). (2000a). Guidelines for methacholine and exercise challenge testing-1999.
      Am J Respir Grit Care Med 161: 309-329.
ATS. (American Thoracic Society). (2000b). What constitutes an adverse health effect of air pollution? This
      official statement of the American  Thoracic Society was adopted by the ATS Board of Directors, July
      1999. Am J Respir Grit Care Med  161: 665-673.
Auten. RL: Foster. WM. (In Press) Biochemical effects of ozone on asthma during postnatal development.
      Biochim Biophys Acta. http://dx.doi.Org/10.1016/i.bbaaen.2011.01.008.
Auten, RL: Potts, EN: Mason, SN: Fischer,  B: Huang, Y: Foster, WM. (2009). Maternal exposure to particulate
      matter increases postnatal  ozone-induced airway hyperreactivity in juvenile mice. Am J Respir Grit Care
      Med 180: 1218-1226. http://dx.doi.org/10.1164/rccm.200901-0116OC.
Autier, P: Pore, JF: Negrier, S: Lienard, D: Panizzon, R: Lejeune, FJ: Guggisberg, D: Eggermont, AM. (1999).
      Sunscreen use and duration of sun exposure: A double-blind, randomized trial. J Natl Cancer Inst 91:
      1304-1309.
Autier, P: Pore, JF: Reis, AC: Grivegnee, A: Ollivaud, L: Truchetet, F: Chamoun,  E: Rotmensz,  N: Severi, G:
      Cesarini. JP (2000). Sunscreen use and intentional exposure to ultraviolet A and B radiation: A double
      blind randomized trial using personal dosimeters. Br J Cancer 83: 1243-1248.
Avila-Costa, MR: Colin-Barengue, L: Fortoul, Tl: Machado-Salas, JP: Espinosa-Villanueva, J: Rugerio-Vargas,
      C: Rivas-Arancibia. S. (1999). Memory deterioration in an oxidative stress model and its correlation with
      cytological changes on rat hippocampus CA1. Neurosci Lett 270: 107-109.
Avissar, NE: Reed, CK: Cox, C: Frampton,  MW: Finkelstein, JN. (2000). Ozone, but not nitrogen dioxide,
      exposure decreases glutathione peroxidases in epithelial lining fluid of human lung. Am J Respir Grit
      Care Med 162: 1342-1347.
Avnerv, S: Mauzerall, PL: Liu, J: Horowitz, LW. (2011a). Global crop yield reductions  due to surface ozone
      exposure: 1. Year 2000 crop production losses and economic damage. Atmos Environ 45: 2284-2296.
      http://dx.doi. org/10.1016/i.atmosenv.2Q10.11.045.
Avnerv, S: Mauzerall, PL: Liu, J: Horowitz, LW. (2011b). Global crop yield reductions  due to surface ozone
      exposure: 2. Year 2030 potential crop production losses and economic damage under two scenarios of
      O3 pollution. Atmos Environ 45: 2297-2309. http://dx.doi.Org/10.1016/i.atmosenv.2011.01.002.
Draft - Do Not Cite or Quote                        11-4                                  September 2011

-------
Avol, EL: Linn, WS: Venet, TG: Shamoo, DA: Hackney, JD. (1984). Comparative respiratory effects of ozone
      and ambient oxidant pollution exposure during heavy exercise. J Air Waste Manag Assoc 34: 804-809.
Avol, EL: Trim, SC: Little, DE: Spier, CE: Smith, MM: Peng, RC: Linn, WS: Hackney, JD: Gross, KB: D'Arcv, JB:
      Gibbons, D: Higgins, ITT. (1990). Ozone exposure and lung function in children attending a southern
      California summer camp. In Proceedings of the 83rd A&WMA Annual Meeting (Vol. 8, pp. 90-150.153).
      Pittsburgh,  PA: Air & Waste Management Association.
Avol, EL: Trim, SC: Little, DE: Spier, CE: Smith, MM: Peng, RC: Linn, WS: Hackney, JD. (1991). Ozone
      exposure and lung function: A southern California summer camp study. In RL Berglund; DR Lawson; DJ
      McKee (Eds.), Tropospheric ozone and the environment: Papers from an international  conference; March
      1990; Los Angeles, CA(pp. 90-99). Los Angeles, CA: Air & Waste  Management Association.
Avol, EL; Navidi, WC; Rappaport,  EB; Peters, JM. (1998a). Acute effects  of ambient ozone on asthmatic,
      wheezy, and healthy children. (82). Topsfield, MA: Health Effects Institute; Flagship Press.
Avol, EL; Navidi, WC; Colome, SD. (1998b). Modeling ozone levels in and around southern California homes.
      Environ Sci Technol 32: 463-468.
Awasthi. YC: Yang. Y: Tiwari.  NK:  Patrick. B: Sharma. A: Li. J: Awasthi. S. (2004).  Regulation  of 4-
      hydroxynonenal-mediated signaling by gluathione S-transferases.  Free Radic Biol Med 37: 607-619.
      http://dx.doi.0rg/10.1016/i.freeradbiomed.2004.05.033.
Awmack. CS: Harrington. R: Lindroth. RL. (2004). Aphid individual performance may not predict population
      responses to elevated  CO2 or O3. Global Change Biol Biol.10: 1414-1423.
Awmack, CS; Mondor,  EB; Lindroth, RL.  (2007). Forest understory clover populations in enriched CO2 and O-3
      atmospheres: Interspecific, intraspecific, and indirect effects. Environ Exp Bot 59: 340-346.
      http://dx.doi.0rg/10.1016/i.envexpbot.2006.04.003.
Azevedo. JM: Goncalves. FL: de Fatima Andrade.  M. (2011). Long-range ozone transport and its impact on
      respiratory and cardiovascular health in the north of Portugal. Int J Biometeorol 55: 187-202.
      http://dx.doi. org/10.1007/s00484-010-0324-2.
Baccarelli. A: Zanobetti. A: Martinelli. I: Grillo. P: Hou. L: Lanzani.  G: Mannucci. PM: Bertazzi. PA: Schwartz. J.
      (2007). Air pollution, smoking, and plasma homocysteine. Environ  Health Perspect 115: 176-181.
Baccini, M; Biggeri, A; Accetta, G; Kosatskv, T;  Katsouyanni,  K; Analitis, A; Anderson,  HR; Bisanti, L; D'lppoliti,
      D; Danova, J; Forsberg, B; Medina,  S; Paldv, A;  Rabczenko, D; Schindler, C; Michelozzi, P. (2008). Heat
      effects on mortality in 15 European cities. Epidemiology 19: 711-719.
      http://dx.doi. org/10.1097/EDE.Ob013e318176bfcd.
Backus. GS: Howden. R: Fostel. J: Bauer. AK: Cho. HY: Marzec.  J: Peden. DB: Kleeberger. SR. (2010).
      Protective role of interleukin-10 in ozone-induced pulmonary inflammation.  Environ Health Perspect 118:
      1721-1727. http://dx.doi.org/10.1289/ehp. 1002182.
Bagard. M: Le Thiec. D: Delacote. E: Hasenfratz-Sauder. MP: Banvov. J:  Gerard. J: Dizengremel.  P: Jolivet. Y.
      (2008). Ozone-induced changes in photosynthesis and photorespiration of hybrid poplar in  relation to the
      developmental stage of the leaves. Physiol Plant 134:  559-574. http://dx.doi.Org/10.1111/i.1399-
      3054.2008.01160.x.
Baler, M; Kandlbinder, A; Golldack, D; Dietz, K. (2005).  Oxidative  stress and ozone: Perception; signalling and
      response. Plant Cell Environ 28: 1012-1020. http://dx.doi.Org/10.1111/i.1365-3040.2005.01326.x.
Baia. ES: Schwartz. JD: Wellenius. GA: Coull. BA:  Zanobetti. A: Vokonas. PS: Suh. HH. (2010). Traffic-related
      air pollution and QT interval: Modification by diabetes, obesity, and oxidative stress gene polymorphisms
      in the Normative Aging Study. Environ Health Perspect 118: 840-846.
      http://dx.doi.org/10.1289/ehp.0901396.
Balbi. B: Pignatti. P: Corradi. M: Baiardi.  P: Bianchi. L: Brunetti. G: Radaeli. A: Moscato. G: Mutti. A: Spanevello.
      A; Malerba, M. (2007). Bronchoalveolar lavage, sputum and exhaled clinically relevant inflammatory
      markers: Values in healthy  adults. Eur Respir J 30: 769-781.
      http://dx.doi.org/10.1183/09031936.00112306.
Baldantoni, D; Fagnano, M; Alfani, A. (2011). Tropospheric ozone effects  on chemical composition and
      decomposition rate of Quercus ilex L. leaves. Sci Total Environ 409: 979-984.
      http://dx.doi. org/10.1016/i.scitotenv.2Q10.11.022.
Balis, PS; Zerefos, CS: Kourtidis,  K; Bais, AF;  Hofzumahaus, A; Kraus, A; Schmitt, R; Blumthaler,  M;  Gobbi, GP
      (2002). Measurements and modeling of photolysis rates during the photochemical activity and ultraviolet
      radiation (PAUR) II campaign. J Geophys Res 107: 8138. http://dx.doi.org/10.1029/2000JD000136.
Ball, GR; Palmer-Brown, D; Fuhrer, J; Skarby,  L; Gimeno, BS; Mills, G. (2000). Identification of non-linear
      influences on the seasonal ozone dose-response of sensitive and  resistant clover clones using artificial
      neural networks. Ecol Modell 129: 153-168.
Ballester, F; Saez, M; Daponte, A; Ordonez, JM: Taracido, M;  Cambra, K; Arribas, F; Bellido, JB; Guillen, JJ;
      Aguinaga. I: Canada. A: Lopez. E: Iniguez. C: Rodriguez. P: Perez-Hovos.  S: Barcelo.  MA: Ocana. R:
      Aranguez. E. (2005). [The EMECAS Project: Spanish  multicentre study on  short-term health effects of air
      pollution]. Rev Esp Salud Publica 79: 229-242.
Draft - Do Not Cite or Quote                        11-5                                  September 2011

-------
Ballester, F: Rodriguez, P: Iniguez, C: Saez, M: Daponte, A: Galan, I: Taracido, M: Arribas, F: Bellido, J: Cirarda,
      FB: Canada, A: Guillen, JJ: Guillen-Grima, F: Lopez, E: Perez-Hoyos, S: Lertxundi, A: Toro, S. (2006). Air
      pollution and cardiovascular admisisons association in Spain: Results within the EMECAS project. J
      Epidemiol Community Health 60: 328-336.
Ballinger. CA: Cueto.  R: Squadrito. G: Coffin. JF: Velsor. LW: Prvor. WA: Postlethwait.  EM. (2005). Antioxidant-
      mediated augmentation of ozone-induced membrane oxidation. Free Radic Biol Med 38: 515-526.
      http://dx.doi. org/10.1016/i.freeradbiomed.20Q4.11.009.
Balls. GR: Palmer-Brown. D:  Sanders. GE. (1996). Investigating microclimatic influences on ozone injury in
      clover (Trifolium subterraneum) using artificial neural networks. New Phytol 132: 271-280.
      http://dx.doi.0rg/10.1111/i.1469-8137.1996.tb01846.x.
Balmes, JR: Chen, l_l_: Scannell, C: Tager, I: Christian, D: Hearne, PQ: Kelly, T: Aris, RM. (1996). Ozone-
      induced decrements in FEV1 and FVC do not correlate with measures of inflammation. Am J Respir Crit
      Care Med 153: 904-909.
Balmes. JR: Aris. RM: Chen.  LL: Scannell. C: Tager. IB:  Finkbeiner. W: Christian. D: Kelly. T: Hearne. PQ:
      Ferrando. R: Welch. B. (1997). Effects of ozone on normal and potentially sensitive human subjects.  Part
      I: Airway inflammation and responsiveness to ozone in normal and asthmatic subjects. Boston, MA:
      Health Effects  Institute.
Bandeff. JM: Pregitzer. KS: Lova. WM: Holmes. WE:  Zak.  DR. (2006). Overstory community composition and
      elevated atmospheric CO2 and O3 modify understory biomass production and nitrogen acquisition. Plant
      Soil 282: 251-259. http://dx.doi.org/10.1007/s11104-005-5930-0.
Bang, S: Kim, KY: Yoo, S: Kim, YG: Hwang, SW. (2007). Transient receptor potential A1 mediates
      acetaldehyde-evoked pain sensation. Eur J Neurosci 26: 2516-2523. http://dx.doi.org/10.1111/1.1460-
      9568.2007.05882.x.
Barnes. J: Zheng. Y: Lyons. T. (2002). Plant resistance to ozone: The role of ascorbate. In  K Omasa; H Saji; S
      Youssefian; N Kondo (Eds.), Air pollution and plant biotechnology - Prospects for phytomonitoring and
      phytoremediation (pp.  235-252). Tokyo: Springer-Verlag.
Barnes. PJ: Liew. FY.  (1995). Nitric oxide and asthmatic inflammation. Immunol Today 16: 128-130.
      http://dx.doi.org/10.1016/0167-5699(95)80128-6.
Barnett, AG: Williams, GM: Schwartz, J: Best, Tl_: Neller, AH: Petroeschevskv, Al_: Simpson, RW. (2006). The
      effects of air pollution on hospitalizations for cardiovascular disease in elderly people in Australian and
      New Zealand cities. Environ Health Perspect 114: 1018-1023.
Barraza-Villarreal. A: Sunver. J: Hernandez-Cadena.  L: Escamilla-Nunez. MC:  Sienra-Monge. JJ: Ramirez-
      Aguilar. M: Cortez-Lugo.  M: Holguin. F: Diaz-Sanchez. D: Olin. AC: Romieu. I. (2008). Air pollution,
      airway inflammation, and lung function in a cohort study of Mexico City schoolchildren. Environ Health
      Perspect 116: 832-838. http://dx.doi.org/10.1289/ehp. 10926.
Barrie. LA: Bottenheim. JW: Schnell. RC:  Crutzen. PJ: Rasmussen. RA. (1988). Ozone destruction and
      photochemical reactions at polar sunrise in the lower Arctic atmosphere. Nature 334: 138-141.
Barry, BE: Miller, FJ: Crapo, JD. (1983). Alveolar epithelial injury caused by inhalation of 025 ppm of ozone.  In
      74th Meeting of the Air Pollution Control Association (Vol. NJ). Pittsburgh, PA: Air Pollution Control
      Association.
Barry. BE: Miller. FJ: Crapo. JD. (1985). Effects of inhalation of 0.12 and 0.25 parts per million ozone on the
      proximal alveolar region of juvenile and adult rats. Lab Invest 53: 692-704.
Basha, MA: Gross, KB: Gwizdala, CJ: Haidar, AH: Popovich, J, Jr. (1994). Bronchoalveolar lavage neutrophilia
      in asthmatic and healthy volunteers after controlled exposure to ozone and filtered purified air. Chest
      106: 1757-1765.
Bassin, S: Volk, M: Fuhrer, J. (2007a). Factors affecting the ozone sensitivity of temperate  European
      grasslands: An overview. Environ Pollut 146: 678-691. http://dx.doi.Org/10.1016/i.envpol.2006.06.010.
Bassin, S: Volk, M: Suter,  M:  Buchmann, N: Fuhrer, J. (2007b). Nitrogen deposition but not ozone affects
      productivity and community composition of subalpine grassland after 3 yr of treatment. New Phytol 175:
      523-534. http://dx.doi.0rg/10.1111/i.1469-8137.2007.02140.x.
Bassin. S: Werner. RA: Sorgel. K: Volk.  M: Buchmann. N: Fuhrer. J. (2009). Effects of combined ozone and
      nitrogen deposition on the in situ properties of eleven key plant species  of a subalpine pasture.
      Oecologia 158: 747-756. http://dx.doi.org/10.1007/s00442-008-1191-v.
Bastackv. J: Lee. CY: Goerke. J: Koushafar. H: Yager. D: Kenaga. L: Speed. TP:  Chen. Y: Clements. JA.
      (1995). Alveolar lining  layer is thin and continuous: Low-temperature scanning electron  microscopy of rat
      lung. J Appl Physiol 79: 1615-1628.
Bates, ML: Brenza, TM: Ben-Jebria, A: Bascom, R: Ultman, JS. (2009). Longitudinal distribution of ozone
      absorption in the lung: Comparison of cigarette smokers and nonsmokers. Toxicol Appl Pharmacol 236:
      270-275.
Bauer. AK: Rondini. EA: Hummel. KA: Degraff. LM: Walker. C: Jedlicka. AE: Kleeberger. SR. (2011).
      Identification of candidate genes downstream  of TLR4 signaling after ozone exposure in mice: A role for
      heat shock protein 70. Environ Health Perspect 119: 1091-1097. http://dx.doi.org/10.1289/ehp. 1003326.
Draft - Do Not Cite or Quote                        11-6                                  September 2011

-------
Bauer, MR: Hultman, NE: Panek, JA: Goldstein, AH. (2000). Ozone deposition to a ponderosa pine plantation in
      the Sierra Nevada Mountains (CA): A comparison of two different climatic years. J Geophys Res 105:
      22,123-122,136. http://dx.doi.org/10.1029/2000JD900168.
Beckerman, B: Jerrett,  M: Brook, JR: Verma, DK: Arain, MA: Finkelstein, MM. (2008). Correlation of nitrogen
      dioxide with other traffic pollutants near a major expressway. Atmos Environ 42: 275-290.
Beckett. WS: McDonnell. WF: Horstman. DH: House. DE. (1985). Role of the parasympathetic nervous system
      in acute lung response to ozone. J Appl Physiol 59: 1879-1885.
Beelen.  R: Hoek. G: Pebesma. E: Vienneau. D: de Hoogh. K: Briggs. DJ. (2009). Mapping of background air
      pollution at a fine spatial scale across the European Union. Sci Total Environ 407: 1852-1867.
      http://dx.doi. org/10.1016/i.scitotenv.20Q8.11.048.
Beer, R. (2006). TES on the aura mission: Scientific objectives, measurements, and analysis overview.  IEEE
      Trans Geosci Remote Sens 44: 1102-1105. http://dx.doi.org/10.1109/TGRS.2005.863716.
Beeson, Wl_: Abbey, DE: Knutsen, SF. (1998).  Long-term concentrations of ambient air pollutants and incident
      lung cancer in California adults: Results from the AHSMOG study. Environ Health Perspect 106:  813-
      823.
Beko, G: Clausen, G: Weschler, CJ. (2007). Further studies of oxidation processes on filter surfaces: Evidence
      for oxidation products and the influence of time in service. Atmos Environ 41: 5202-5212.
      http://dx.doi.0rg/10.1016/i.atmosenv.2006.07.063.
Bell,  ML: McDermott, A: Zeger, Sl_: Samet, JM: Dominici, F (2004). Ozone and short-term mortality in 95 US
      urban communities,  1987-2000. JAM A 292: 2372-2378. http://dx.doi. org/10.1001/jama.292.19.2372.
Bell,  ML: Dominici, F: Samet, JM. (2005). A meta-analysis of time-series studies of ozone and mortality  with
      comparison to the national morbidity, mortality, and air pollution study. Epidemiology 16: 436-445.
Bell.  ML: Peng. RD: Dominici. F. (2006). The exposure-response curve for ozone and risk of mortality and the
      adequacy of current ozone regulations.  Environ Health Perspect 114: 532-536.
Bell,  ML. (2006). The use of ambient air quality modeling to estimate individual and population exposure for
      human health research: A case study of ozone in the Northern Georgia region of the United States.
      Environ Int 32: 586-593.
Bell,  ML: Kim, JY: Dominici, F. (2007). Potential confounding of particulate matter on the short-term association
      between ozone  and  mortality in multisite time-series studies. Environ Health Perspect 115:  1591-1595.
      http://dx.doi. org/10.1289/ehp. 10108.
Bell,  ML: Dominici, F. (2008). Effect modification by community characteristics on the short-term effects of
      ozone exposure and mortality in 98  US  communities. Am J Epidemiol 167: 986-997.
      http://dx.doi. org/10.1093/aie/kwm396.
Bell,  ML: Levy, JK: Lin, Z. (2008). The effect of sandstorms and air pollution on cause-specific hospital
      admissions in Taipei, Taiwan. Occup Environ Med 65: 104-111.
      http://dx.doi. org/10.1136/oem.2006.031500.
Belvisi, MG: Stretton, CD: Verleden, GM: Ledingham, SJ: Yacoub,  MH: Barnes, PJ. (1992). Inhibition of
      cholinergic neurotransmission in human airways by opioids. J Appl Physiol 72: 1096-1100.
Bender,  J: Muntifering,  RB: Lin,  JC: Weigel, HJ. (2006). Growth and nutritive quality of Poa pratensis as
      influenced by ozone and  competition. Environ Pollut 142: 109-115.
      http://dx.doi.0rg/10.1016/i.envpol.2005.09.012.
Bender.  J: Weigel. H. -J. (2011). Changes in atmospheric chemistry and crop health: A review [Review]. Agron
      Sustain Dev 31: 81-89. http://dx.doi.org/10.1051/agro/2010013.
Bennett. WD: Hazucha. MJ: Folinsbee. LJ: Bromberg. PA: Kissling. GE: London. SJ. (2007). Acute pulmonary
      function response to ozone in young adults as a function of body mass index. Inhal Toxicol  19: 1147-
      1154. http://dx.doi.org/10.1080/08958370701665475.
Benoit,  LF: Skelly, JM:  Moore, LD: Dochinger, LS. (1982). Radial growth reductions of Pinus strobus L
      correlated with foliar ozone sensitivity as an indicator of ozone-induced losses in eastern forests. Can J
      For Res 12: 673-678. http://dx.doi.org/10.1139/x82-101.
Bergamaschi. E: De Palma. G: Mozzoni. P: Vanni. S: Vettori. MV: Broeckaert. F: Bernard. A: Mutti. A. (2001).
      Polymorphism of quinone-metabolizing  enzymes and susceptibility to ozone-induced acute effects. Am J
      Respir Grit Care Med 163: 1426-1431.
Bergweiler. C: Manning. WJ: Chevone. Bl.  (2008). Seasonal and diurnal gas exchange differences in ozone-
      sensitive common milkweed (Asclepias syriaca L.) in relation to  ozone uptake. Environ Pollut 152: 403-
      415. http://dx.doi.0rg/10.1016/i.envpol.2007.06.019.
Bergweiler, CJ: Manning, WJ. (1999). Inhibition of flowering and reproductive success in spreading dogbane
      (Apocynum androsaemifolium) by exposure to ambient ozone. Environ Pollut 105: 333-339.
      http://dx.doi.org/10.1016/S0269-7491(99)00044-5.
Berhane. K: Gauderman. WJ: Stram. DO: Thomas. DC. (2004). Statistical  issues in studies of the long-term
      effects of air pollution: The Southern California Children's Health Study. Stat Sci 19: 414-449.
      http://dx.doi.org/10.1214/088342304000000413.
Berhane. K: Zhang. Y:  Linn. WS: Rappaport. EB: Bastain. TM: Salam. MT: Islam. T: Lurmann. F: Gilliland. FD.
      (2011). The effect of ambient air pollution on exhaled nitric oxide in the Children's Health Study. Eur
      Respir J 37: 1029-1036. http://dx.doi.org/10.1183/09031936.00081410.
Draft - Do Not Cite or Quote                        11-7                                  September 2011

-------
Berkey, CS: Hoaglin, DC: Antczak-Bouckoms, A: Mosteller, F: Colditz, GA. (1998). Meta-analysis of multiple
      outcomes by regression with random effects. Stat Med 17: 2537-2550.
      http://dx.doi.org/10.1002/(SICI)1097-0258(19981130)17:22<2537::AID-SIM953>3.0.CO:2-C.
Berkowitz, CM: Shaw, WJ. (1997). Airborne measurements of boundary layer chemistry during the Southern
      Oxidant Study: Acase study. J Geophys  Res 102: 12,795-712,804. http://dx.doi.orq/10.1029/97JD00417.
Berkowitz. CM: Fast. JD: Sprinston. SR: Larsen. RJ: Spicer. CW: Doskev. PV: Hubbe. JM: Plastridge. R. (1998).
      Formation mechanisms and chemical characteristics of elevated photochemical layers over the northeast
      United States. J Geophys Res 103: 10,631-610,647.
Bernacchi. CJ: Morgan. PB:  Ort.  PR:  Long. SR  (2005). The growth of soybean under free air CO2 enrichment
      (FACE) stimulates photosynthesis while decreasing in vivo Rubisco capacity. Planta 220: 434-446.
      http://dx.doi. org/10.1007/s00425-004-1320-8.
Bernacchi, CJ: Leaky, ADB:  Heady, LE: Morgan, PB: Dohleman, FG: McGrath, JM: Gillespie, KM: Wittig, VE:
      Rogers, A: Long, SP:  Ort,  DR.  (2006). Hourly and seasonal variation in photosynthesis and stomatal
      conductance of soybean grown at future  CO2 and ozone concentrations for 3 years under fully open-air
      field conditions. Plant Cell Environ 29: 2077-2090. http://dx.doi.Org/10.1111/i.1365-3040.2006.01581.x.
Berntsen, TK: Myhre,  G: Stordal, F: Isaksen, ISA. (2000). Time evolution of tropospheric ozone and its radiative
      forcing. J Geophys Res 105: 8915-8930. http://dx.doi.orq/10.1029/1999JD901139.
Berry. M: Liov. PJ: Gelperin.  K: Buckler. G:  Klotz. J. (1991). Accumulated exposure to ozone and measurement
      of health  effects in children and counselors at two summer camps. Environ Res 54: 135-150.
Betzelberger, AM: Gillespie,  KM:  McGrath, JM: Koester, RP: Nelson, Rl_: Ainsworth, EA. (2010). Effects of
      chronic elevated ozone concentration on antioxidant capacity, photosynthesis and seed yield of 10
      soybean cultivars. Plant Cell Environ 33:  1569-1581. http://dx.doi.Org/10.1111/i.1365-3040.2010.02165.x.
Bhalla. DK: Gupta. SK. (2000). Lung injury, inflammation, and inflammatory stimuli in rats exposed to ozone. J
      Toxicol Environ Health 59: 211-228.
Bidart-Bouzat, MG: Imeh-Nathaniel, A. (2008). Global change effects on plant chemical defenses against insect
      herbivores. J Integr Plant Biol 50: 1339-1354. http://dx.doi.Org/10.1111/i.1744-7909.2008.00751.x.
Biggeri. A: Baccini. M: Bellini. P: Terracini. B. (2005). Meta-analysis of the Italian studies of short-term effects of
      air pollution (MISA), 1990-1999. Int J Occup Environ Health 11: 107-122.
Bignami, G: Musi, B: Dell'Omo, G: Laviola,  G: Alleva, E. (1994). Limited effects of ozone exposure during
      pregnancy on physical and neurobehavioral development of CD-1 mice. Toxicol Appl Pharmacol 129:
      264-271.  http://dx.doi.org/10.1006/taap. 1994.1251.
Billings. WD. (1978). Plants and the ecosystem. In. Belmont, CA: Wadsworth Publishing Company, Inc.
Binkowski. F: Roselle. S. (2003). Models-3 Community Multiscale Air Quality(CMAQ) model aerosol component
      1. Model  description.  J Geophys Res 108: 4183.  http://dx.doi.org/10.1029/2001JD001409.
Binkowski. FS: Arunachalam. S: Adelman. Z: Pinto. JP (2007).  Examining photolysis rates with a prototype
      online photolysis module in CMAQ. J Appl Meteor Climatol 46: 1252-1256.
Bishop, GA: Stedman, PH. (2008). A decade of on-road emissions measurements. Environ Sci Technol 42:
      1651-1656. http://dx.doi.org/10.1021/es702413b.
Biswas, DK: Xu, H: Li, YG: Sun, JZ: Wang, XZ:  Han, XG: Jiang, GM. (2008). Genotypic differences in leaf
      biochemical, physiological and growth responses to ozone in 20 winter wheat cultivars released over the
      past 60 years. Global Change  Biol 14: 46-59. http://dx.doi.Org/10.1111/i.1365-2486.2007.01477.x.
Black. VJ:  Black. CR:  Roberts. JA: Stewart. CA. (2000). Impact of ozone on the reproductive development of
      plants. New Phytol 147: 421-447.
Black. VJ:  Stewart. CA: Roberts. JA: Black. CR. (2007). Ozone affects gas exchange, growth and reproductive
      development in Brassica campestris (Wisconsin Fast Plants). New Phytol 176:  150-163.
      http://dx.doi.0rg/10.1111/i.1469-8137.2007.02163.x.
Black, VJ:  Stewart, CA: Roberts, JA: Black, CR. (2010). Direct effects of ozone on reproductive development in
      Plantago  major L.  populations  differing in sensitivity. Environ Exp Bot 69: 121-128.
      http://dx.doi.0rg/10.1016/i.envexpbot.2010.04.006.
Blande. JD: Holopainen. JK: Li. T (2010). Air pollution impedes plant-to-plant communication by volatiles. Ecol
      Lett 13: 1172-1181. http://dx.doi.Org/10.1111/i.1461-0248.2010.01510.x.
Blanken, PD:  Dillon, J: Wismann, G. (2001). The impact of an air quality advisory program on voluntary mobile
      source air pollution reduction. Atmos Environ 35: 2417-2421.
Block. ML: Calderon-Garciduenas. L.  (2009). Air pollution: Mechanisms of neuroinflammation and CNS disease.
      Trends Neurosci 32: 506-516.  http://dx.doi.Org/10.1016/i.tins.2009.05.009.
Blomberg, A: Mudwav, IS: Nordenhall, C: Hedenstrom, H: Kelly, FJ:  Frew, AJ:  Holgate, ST: Sandstrom, T.
      (1999). Ozone-induced lung function decrements do not correlate with early airway inflammatory or
      antioxidant responses. Eur Respir J 13: 1418-1428.
Blondeau.  P: lordache. V: Poupard. O: Genin. D: Allard. F. (2005). Relationship between outdoor and indoor air
      quality in  eight French schools. Indoor Air 15: 2-12.
Bloom, B: Cohen, RA: Freeman,  G. (2008). Summary health statistics for U.S. children: National Health
      Interview Survey, 2008. Washington, DC: National Center for Health Statistics.
Draft - Do Not Cite or Quote                        11-8                                  September 2011

-------
Bloomer, BJ: Stehr, JW: Piety, CA: Salawitch, RJ: Dickerson, RR. (2009). Observed relationships of ozone air
      pollution with temperature and emissions. Geophys Res Lett 36: L09803.
      http://dx.doi. org/10.1029/2009GL037308.
Blumenthal, PL: Lurmann, FW: Kumar, N: Dye, TS: Ray, SE: Korc, ME: Londergan, R: Moore, G. (1997).
      Transport and mixing  phenomena related to ozone exceedances in the northeast US (analysis based on
      NARSTO-northeast data). Santa Rosa, CA: Sonoma Technology.
      http://capita.wustl.edu/otag/reports/otagrept/otagrept.html.
Bobak. M. (2000). Outdoor air pollution, low birth weight, and prematurity. Environ Health Perspect 108: 173-
      176.
Boer, GJ: Yu, B. (2003). Climate sensitivity and response. Clim Dynam 20: 415-429.
      http://dx.doi. org/10.1007/s00382-002-0283-3.
Bohler, S: Bagard, M: Oufir, M: Planchon, S: Hoffmann, l_: Jolivet, Y: Hausman, JF: Dizengremel, P: Renaut, J.
      (2007). A DICE analysis of developing poplar leaves subjected to ozone reveals major changes in carbon
      metabolism. Proteomics 7: 1584-1599. http://dx.doi.org/10.1002/pmic.200600822.
Bohler. S: Sergeant. K: Lefevre. I: Jolivet. Y: Hoffmann. L: Renaut. J: Dizengremel. P: Hausman. JF. (2010).
      Differential  impact of chronic ozone exposure on expanding and fully expanded poplar leaves. Tree
      Physiol 30: 1415-1432. http://dx.doi.org/10.1093/treephvs/tpg082.
Bonn. B: Von Kuhlmann. R: Lawrence. MG. (2004). High contribution of biogenic hydroperoxides to secondary
      organic aerosol formation. Geophys Res Lett 31: L10108. http://dx.doi.org/10.1029/2003GL019172.
Bony, S: Colman, R: Kattsov, VM: Allan, RP: Bretherton, CS: Dufresne, JL: Hall, A: Hallegatte, S: Holland, MM:
      Ingram, W: Randall, DA: Soden, BJ: Tselioudis, G: Webb, MJ. (2006). How well do we understand and
      evaluate climate change feedback processes? J Clim 19: 3445-3482.
Booker.  F: Muntifering. R: McGrath. M: Burkev.  K: Decoteau. D: Fiscus. E: Manning. W:  Krupa. S:  Chappelka.
      A: Grantz. D. (2009). The ozone component of global change: Potential effects on agricultural and
      horticultural plant yield, product quality and interactions with invasive species. J Integr Plant Biol 51:  337-
      351. http://dx.doi.0rg/10.1111/i.1744-7909.2008.00805.x.
Booker.  FL: Reid. CD: Brunschon-Harti. S: Fiscus. EL: Miller. JE. (1997). Photosynthesis and photorespiration in
      soybean [Glycine max (L) Merr] chronically exposed to elevated carbon dioxide and ozone. J Exp Bot 48:
      1843-1852.
Booker,  FL: Fiscus, EL: Miller, JE. (2004a). Combined effects of elevated atmospheric carbon dioxide and
      ozone on soybean whole-plant water use. Environ Manage 33: S355-S362.
      http://dx.doi.org/10.1007/s00267-003-9144-z.
Booker.  FL: Burkev. KO: Overmver. K: Jones. AM. (2004b). Differential responses of G-protein Arabidopsis
      thaliana mutants to ozone. New Phytol 162: 633-641.
Booker.  FL: Prior. SA: Torbert. HA: Fiscus. EL: Purslev. WA: Hu. S. (2005). Decomposition of soybean grown
      under elevated concentrations of CO2 and O3.  Global Change Biol 11: 685-698.
      http://dx.doi.0rg/10.1111/i.1365-2486.2005.00939.x.
Booker,  FL: Fiscus, EL. (2005). The role of ozone flux and antioxidants in the suppression of ozone injury by
      elevated CO2 in soybean. J Exp Bot 56: 2139-2151. http://dx.doi.org/10.1093/ixb/eri214.
Booker,  FL: Burkev, KO: Purslev, WA: Heagle, AS. (2007). Elevated carbon dioxide and ozone effects on
      peanut: I. Gas-exchange,  biomass, and leaf chemistry. Crop Sci 47: 1475-1487.
      http://dx.doi.org/10.2135/cropsci2006.08.0537.
Boorman, GA: Hailey, R:  Grumbein, S: Chou, BJ: Herbert, RA: Goehl, T: Mellick, PW: Roycroft, JH: Haseman,
      JK: Sills.  R. (1994). Toxicology and carcinogenesis studies of ozone and ozone 4-(N-
      nitrosomethylamino)-1-(3-pyridyl)-1-butanone in Fischer-344/N rats. Toxicol Pathol 22: 545-554.
Bornholdt, J: Dybdahl, M: Vogel,  U: Hansen, M: Loft, S: Wallin,  H. (2002). Inhalation of ozone induces DMA
      strand breaks and inflammation in mice.  DMA Repair 520: 63-71.
Borowiak, K: Rucinska-Sobkowiak, R: Rymer, K: Gwozdz, EA: Zbierska, J. (2009). Biochemical  markers of
      tropospheric ozone: Experimentation with test-plants.  Polish Journal  of Ecology 57: 3-14.
Bosson. J: Stenfors. N: Bucht. A: Helledav. R: Pourazar.  J: Holgate. ST: Kelly. FJ: Sandstrom. T: Wilson.  S:
      Frew. AJ: Blomberg. A. (2003). Ozone-induced  bronchial epithelial cytokine expression differs between
      healthy and asthmatic subjects. Clin Exp Allergy 33: 777-782.
Bosson. J: Blomberg. A: Pourazar. J:  Mudwav. IS: Frew. AJ: Kelly. FJ: Sandstrom. T. (2009).  Early suppression
      of NFkappaB and  IL-8 in bronchial epithelium after ozone exposure in healthy human  subjects. Inhal
      Toxicol 21:  913-919. http://dx.doi.Org/10.1080/08958370802657389.
Bou Jaoude, M: Katerji, N: Mastrorilli, M: Rana, G. (2008). Analysis of the ozone effect on soybean in the
      Mediterranean region II. The consequences on  growth, yield and water use efficiency. Eur J Agron 28:
      519-525. http://dx.doi.0rg/10.1016/i.eia.2007.09.001.
Bou Jaoude. M: Katerii. N: Mastrorilli. M: Rana. G. (2008). Analysis of the effect of ozone on  soybean in the
      Mediterranean region I: The consequences on crop-water status. Eur J Agron 28: 508-518.
      http://dx.doi.0rg/10.1016/i.eia.2007.09.002.
Boussouar. A: Araneda. S: Hamelin. C: Soulage. C: Kitahama. K: Dalmaz. Y. (2009).  Prenatal ozone exposure
      abolishes stress activation of Fos and tyrosine hydroxylase  in the nucleus tractus solitarius  of adult rat.
      Neurosci Lett 452: 75-78.
Draft - Do Not Cite or Quote                        11-9                                  September 2011

-------
Brandt, LA: Bohnet, C: King, JY. (2009). Photochemically induced carbon dioxide production as a mechanism
      for carbon loss from plant litter in arid ecosystems. J Geophys Res 114: G02004.
      http://dx.doi. org/10.1029/2008ig000772.
Brauer, M: Blair, J: Vedal, S. (1996). Effect of ambient ozone exposure on lung function in farm workers. Am J
      Respir Grit Care Med 154: 981-987.
Brauer. M: Brook. JR. (1997). Ozone personal exposures and health effects for selected groups residing in the
      Fraser Valley. Atmos Environ 31: 2113-2121.
Brauer. M: Brumm. J: Vedal. S: Petkau. AJ. (2002). Exposure misclassification and threshold concentrations in
      time series analyses of air pollution health effects. Risk Anal 22: 1183-1193.
Brauer, M: Lencar, C: Tamburic, l_: Koehoorn, M: Demers, P: Karr,  C. (2008). A cohort study of traffic-related air
      pollution impacts on birth outcomes. Environ Health Perspect 116: 680-686.
Braun-Fahrlander, C, h: Kunzli, N: Domenighetti, G: Carell, CF: Ackermann-Liebrich, U. (1994). Acute effects of
      ambient ozone on respiratory function of Swiss schoolchildren after a 10-minute heavy exercise. Pediatr
      Pulmonol  17:  169-177. http://dx.doi.orq/10.1002/ppul. 1950170306.
Bravo. MA: Bell.  ML. (2011). Spatial heterogeneity  of PM10 and O3 in Sao Paulo,  Brazil,  and implications for
      human health studies. J Air Waste  Manag Assoc61: 69-77.
Breen. MS: Breen. M: Williams. RW: Schultz. BD. (2010). Predicting residential air exchange rates from
      questionnaires and meteorology: Model evaluation in central North Carolina. Environ Sci Technol 44:
      9349-9356. http://dx.doi.org/10.1021/es101800k.
Bresnahan, BW:  Dickie, M: Gerking, S. (1997). Averting behavior and urban air pollution. Land Econ 73: 34-57.
Breton, CV: Salam, MT: Vora, H: Gauderman, WJ:  Gilliland, FD. (2011). Genetic variation in the glutathione
      synthesis  pathway, air pollution, and children's lung function growth. Am J Respir Grit Care Med 183:
      243-248. http://dx.doi.ora/10.1164/rccm.201006-0849OC.
Breuer-McHam. J: Simpson. E: Dougherty. I: Bonkobara. M: Ariizumi. K: Lewis. DE: Dawson. DB: Duvic. M:
      Cruz, PD, Jr. (2001). Activation of HIV in human  skin by ultraviolet B radiation and its inhibition by
      NF"kappa"B blocking agents. Photochem Photobiol 74: 805-810.
Brewer. PG: Peltzer. ET. (2009). Limits to  marine life. Science 324: 347-348.
      http://dx.doi.org/10.1126/science.1170756.
Briggs, DJ: Collins, S: Elliott, P: Fischer, P: Kingham, S: Lebret, E:  Pryl,  K: Van Reeuwijk, H: Smallbone,  K: Van
      Per Veen, A. (1997). Mapping urban air pollution using GIS: A regression-based approach. Int J Geogr
      Inform Sci 11: 699-718.
Broadmeadow. MSJ: Jackson. SB. (2000). Growth responses of Quercus petraea, Fraxinus excelsior and Pinus
      sylvestris to elevated carbon dioxide, ozone and water supply. New Phytol 146: 437-451.
      http://dx.doi. org/10.1046/1.1469-8137.2000.00665.x.
Brodin. M: Helmig. D: Oltmans. S. (2010). Seasonal ozone behavior along an elevation gradient in the Colorado
      Front Range Mountains. Atmos Environ 44:  5305-5315.
      http://dx.doi.0rg/10.1016/i.atmosenv.2010.06.033.
Broeckaert, F: Clippe, A: Wattiez, R: Falmagne, P:  Bernard, A. (2003). Lung hyperpermeability, Clara-cell
      secretory potein (CC16), and susceptibility to ozone of five inbred strains of mice.  Inhal Toxicol 15: 1209-
      1230.
Brook. JR: DiGiovanni. F: Cakmak. S: Meyers. TP  (1997). Estimation of dry deposition velocity using inferential
      models and site-specific meteorology—uncertainty due to siting of meteorological towers. Atmos Environ
      31:  3911-3919.
Brooks, EG. (2010).  Correspondence from Dr. Brooks - Re: Clarifications in 2008 J Occup Environ Med article
Brown. BA: Jenkins.  Gl. (2008). UV-B signaling pathways with different fluence-rate response profiles are
      distinguished  in mature Arabidopsis leaf tissue by requirement for UVR8, HY5, and HYH. Plant Physiol
      146: 576-588. http://dx.doi.org/10.1104/pp.107.108456.
Brown, JS:  Bateson, TF: McDonnell, WF.  (2008). Effects of exposure to  0.06 ppm  ozone on FEV1 in humans: A
      secondary analysis of existing data. Environ Health Perspect 116: 1023-1026.
      http://dx.doi.org/10.1289/ehp.11396.
Brown. K: Sarnat. J:  Suh. H: Coull. B: Koutrakis. P. (2009). Factors influencing relationships between personal
      and ambient concentrations of gaseous and particulate pollutants. Sci Total Environ  407: 3754-3765.
Bruhl. C: Crutzen. PJ. (1989). On the disproportionate role of tropospheric ozone as a filter against solar UV-B
      radiation. Geophys Res Lett 16: 703-706. http://dx.doi.org/10.1029/GL016i007p00703.
Brunekreef, B: Hoek, G: Breugelmans, O: Leentvaar, M. (1994). Respiratory effects of low-level photochemical
      air pollution in amateur cyclists. Am J Respir Grit Care Med  150: 962-966.
Buadong, D: Jinsart, W: Funatagawa, I: Karita, K: Yano, E. (2009). Association between PM10 and O3 levels
      and hospital visits for cardiovascular diseases in Bangkok, Thailand. J Epidemiol 19: 182-188.
      http://dx.doi.org/10.2188/iea.JE20080047.
Bulbovas.  P: de Souza. SR: de Moraes. RM: Luizao.  F: Artaxo. P. (2007). Soybean Tracaja' seedlings exposed
      to ozone under controlled conditions. Pesqui Agropecu Bras 42: 641-646.
      http://dx.doi. org/10.1590/S0100-204X2007000500005.
Burgard. DA: Bishop. GA: Stedman. DH: Gessner.  VH:  Daeschlein. C. (2006). Remote sensing of in-use heavy-
      duty diesel trucks. Environ Sci Technol 40: 6938-6942. http://dx.doi.org/10.1021/es060989a.
Draft - Do Not Cite or Quote                        11-10                                  September 2011

-------
Burke, JM: Zufall, MJ: Ozkaynak, H. (2001). A population exposure model for particulate matter: Case study
      results for PM2.5 in Philadelphia, PA. J Expo Sci Environ Epidemiol 11: 470-489.
Burkey, KO: Eason, G: Fiscus, EL.  (2003). Factors that affect leaf extracellular ascorbic acid content and redox
      status. Physiol Plant 117: 51-57. http://dx.doi.Org/10.1034/1.1399-3054.2003.1170106.x.
Burkev. KO: Neufeld. HS: Souza. L: Chappelka. AH: Davison. AW. (2006). Seasonal profiles of leaf ascorbic
      acid content and redox state in ozone-sensitive wildflowers. Environ Pollut 143: 427-434.
      http://dx.doi.0rg/10.1016/i.envpol.2005.12.009.
Burkev. KO: Booker. FL: Purslev. WA: Heagle. AS. (2007). Elevated carbon dioxide and ozone effects on
      peanut: II. Seed yield and quality. Crop Sci 47: 1488-1497.
      http://dx.doi.org/10.2135/cropsci2006.08.0538.
Burleson, GR: Keyes, l_l_: Stutzman, JD. (1989). Immunosuppression of pulmonary natural killer activity by
      exposure to ozone. Immunopharmacol Immunotoxicol 11: 715-735.
      http://dx.doi.org/10.3109/08923978909005397.
Burlev. JD: Ray. JD. (2007).  Surface ozone in Yosemite National Park. Atmos Environ 41: 6048-6062.
Burnett.  R: Raizenne. M: Krewski. D. (1990). Acute health effects of transported air pollution: A study of children
      attending a residential summer camp. Can J Stat 18: 367-373. http://dx.doi.org/10.2307/3315843.
Burra. TA: Moineddin. R: Agha. MM: Glazier. RH. (2009). Social disadvantage, air pollution, and asthma
      physician visits in Toronto, Canada. Environ Res 109: 567-574.
      http://dx.doi.0rg/10.1016/i.envres.2009.03.004.
Bush, ML: Asplund, PT: Miles, KA:  Ben-Jebria, A: Ultman, JS. (1996). Longitudinal distribution of O3 absorption
      in the lung: gender differences and intersubject variability. J Appl Physiol  81: 1651-1657.
Bush, ML: Zhang, W: Ben-Jebria, A: Ultman, JS. (2001). Longitudinal distribution of ozone and chlorine in the
      human respiratory tract: Simulation of nasal and oral breathing with the single-path diffusion model.
      Toxicol Appl  Pharmacol 173: 137-145. http://dx.doi.orq/10.1006/taap.20Q1.9182.
Buzica, D: Gerboles, M: Plaisance, H. (2008). The equivalence of diffusive samplers to reference methods for
      monitoring O3, benzene and NO2 in ambient air. J Environ Monit 10: 1052-1059.
Bvtnerowicz. A: Arbaugh. M: Schilling. S: Fraczek. W: Alexander. D. (2008). Ozone distribution and phytotoxic
      potential in mixed conifer forests of the San Bernardino Mountains, Southern California. Environ Pollut
      155: 398-408. http://dx.doi.Org/10.1016/i.envpol.2008.01.046.
Byun, D: Schere, KL. (2006). Review of the governing equations, computational algorithms, and other
      components of the models-3 community multiscale air quality (CMAQ) modeling system [Review]. Appl
      Mech Rev 59: 51-77.
Bvun. DW: Ching. JKS. (1999). Science algorithms of the EPA models-3 community multiscale air quality
      (CMAQ) modeling system. (EPA/600-R-99-030). Washington, DC: U.S. Environmental Protection
      Agency. http://www.epa.gov/asmdnerl/CMAQ/CMAQscienceDoc.html.
CAA. (Air quality criteria and control techniques, Section 108 of the Clean Air Act). 42. § 7408, (1990a).
CAA. (National primary and secondary ambient air quality standards,  Section 109 of the Clean Air Act). 42. §
      7409, (1990b).
CAA. (Definitions, Section 302 of the Clean Air Act). 42. § 7602, (2005).
Caceres, Al:  Brackmann, M: Ella, MD: Bessac, BF: del Camino, D: D'Amours, M: Witek, JS: Fanger, CM:
      Chong. JA: Havward. NJ:  Homer. RJ:  Cohn. L: Huang. X: Moran. MM:  Jordt. SE. (2009). A sensory
      neuronal ion channel  essential for airway inflammation and hyperreactivity in asthma. PNAS 106: 9099-
      9104. http://dx.doi.org/10.1073/pnas.0900591106.
Caforio. ALP:  Fortina. AB: piaserico. S: Alaibac. M: Tona. F: Feltrin. G: Pompei. E: Testolin. L: Gambino. A:
      Volta. SD: Thiene. G: Casarotto. D:  Peserico. A. (2000).  Skin cancer in heart transplant recipients:  risk
      factor analysis and relevance of immunosuppressive therapy. Circulation 102:  III-222- III-227.
Caird, MA: Richards, JH: Donovan, LA. (2007). Nighttime stomatal conductance and transpiration in C-3 and C-
      4 plants. Plant Physiol 143: 4-10. http://dx.doi.Org/10.1104/pp. 106.092940.
Cakmak, S: Dales, RE: Judek, S. (2006a).  Do gender, education, and income modify the effect of air pollution
      gases on cardiac disease? J Occup Environ Med 48: 89-94.
      http://dx.doi.org/10.1097/01.iom.0000184878.11956.4b.
Cakmak, S: Dales, RE: Judek, S. (2006b).  Respiratory health effects of air pollution gases: Modification by
      education and income. Arch Environ Occup Health 61: 5-10.
Cakmak. S: Dales. RE: Vidal. CB. (2007). Air pollution and mortality in Chile: Susceptibility among the elderly.
      Environ Health Perspect 115: 524-527.
Cakmak, S: Dales, RE: Angelica  Rubio, M: Blanco Vidal, C. (2011). The risk of dying on days of higher air
      pollution among the socially disadvantaged elderly. Environ Res 111: 388-393.
      http://dx.doi.0rg/10.1016/i.envres.2011.01.003.
Cal/EPA. (California Environmental Protection Agency). (2010). Air quality data branch main page, from
      http://www.arb.ca.gov/agd/agdpage.htm
Calatayud, A: Alvarado, JW:  Barreno, E. (2002). Similar effects of ozone on four cultivars of lettuce in open top
      chambers during winter. Photosynthetica 40: 195-200. http://dx.doi.org/10.1023/A: 1021333305592.
Draft - Do Not Cite or Quote                        11-11                                  September 2011

-------
Calatayud, A: Pomares, F: Barreno, E. (2006). Interactions between nitrogen fertilization and ozone in
      watermelon cultivar Reina de Corazones in open-top chambers. Effects on chlorophyll alpha
      fluorescence, lipid peroxidation, and yield. Photosynthetica 44:  93-101. http://dx.doi.org/10.1007/s11099-
      005-0163-2.
Calatavud. V: Cervero. J: Sanz. MJ. (2007a). Foliar, physiologial and growth responses of four maple species
      exposed to ozone. Water Air Soil Pollut 185: 239-254. http://dx.doi.org/10.1007/s11270-007-9446-5.
Calatavud, V: Sanz, MJ: Calvo, E: Cervero, J: Ansel, W: Klumpp, A. (2007b). Ozone biomonitoring with Bel-W3
      tobacco plants in the city of Valencia (Spain). Water Air Soil  Pollut 183: 283-291.
      http://dx.doi.ora/10.1007/s11270-007-9376-2.
Calderini, DF: Lizana, XC: Hess, S: Jobet, CR: Zuniga, JA. (2008). Grain yield and quality of wheat under
      increased  ultraviolet radiation (UV-B) at later stages of the crop cycle. J Agr Sci 146: 57-64.
      http://dx.doi. org/10.1017/S0021859607007447.
Calderon-Garciduenas, l_: Mora-Tiscareno, A: Fordham, LA: Chung, CJ: Valencia-Salazar, G: Flores-Gomez, S:
      Solt. AC: Gomez-del Campo. A: Jardon-Torres. R: Henriquez-Roldan. C: Hazucha. MJ: Reed. W. (2006).
      Lung radiology and pulmonary function of children chronically exposed to air pollution. Environ Health
      Perspect 114: 1432-1437.
Calderon Guzman. D: Barragan Meiia. G: Hernandez Garcia. E: Juarez Olguin. H. (2006). Effect of nutritional
      status and ozone exposure on some biomarkers of oxidative stress in rat brain regions. Nutr Cancer 55:
      195-200. http://dx.doi.org/10.1207/s15327914nc5502 11.
Calderon Guzman, D: Hernandez Islas, Jl_: Mejfa, GB: Santamarfa del Angel, D: Hernandez Garcfa,  E: Juarez
      Olgufn, H.  (2005). Effect of nutritional status and ozone exposure on  Na+/K+ ATPpase and lipid
      peroxidation in rat brain. Proc West Pharmacol Soc 48: 118-121.
Caldwell. MM: Bornman. JF: Ballare. CL: Flint. SD: Kulandaivelu. G. (2007). Terrestrial ecosystems, increased
      solar ultraviolet radiation, and interactions with bother climate change factors. Photochem  Photobiol Sci
      6: 252-266. http://dx.doi.org/10.1039/B700019g.
Campbell. SJ: Wanek. R: Coulston. JW. (2007). Ozone injury in west coast forests: 6 years of monitoring -
      Introduction. Portland, OR: U.S. Department of Agriculture.
Campos-Bedolla, P: Vargas, MH: Montano, LM. (2002). Effect of acute ozone exposure on pregnant  rat uterus
      contractile responses. Reprod Toxicol 16: 269-273.
Cannon, WN. (1990). Olfactory response of eastern spruce budworm larvae to red spruce needles exposed to
      acid rain and elevated levels of ozone. J Chem Ecol 16: 3255-3261.
      http://dx.doi. org/10.1007/BF00982096.
Cantorna.  MT. (2000). Vitamin  D and autoimmunity: Is vitamin D status an environmental factor affecting
      autoimmune disease prevalence? Exp Biol Med 223: 230-233.
Carbaial-Arrovo. L: Miranda-Soberanis. V: Medina-Ramon. M: Roias-Bracho. L: Tzintzun. G: Solfs-Gutierrez. P:
      Mendez-Ramfrez. I: Hurtado-Dfaz. M: Schwartz. J: Romieu. I. (2011). Effect of PM10 and O3 on infant
      mortality among residents in the Mexico City Metropolitan Area: A case-crossover analysis, 1997-2005. J
      Epidemiol  Community Health 65: 715-721. http://dx.doi.org/10.1136/iech.2009.101212.
Garde, RT: Haynes, KF. (2004). Stucture of the pheromone communication channel in moths. In Advances in
      insect chemical ecology (pp. 283-332). Cambridge: Cambridge  University Press.
Carey. SA: Minard. KR: Trease. LL: Wagner. JG: Garcia.  GJ: Ballinger. CA: Kimbell. JS: Plopper.  CG: Corlev.
      RA: Postlethwait. EM: Harkema. JR: Einstein. DR. (2007). Three-dimensional mapping of ozone-induced
      injury in the nasal airways of monkeys using magnetic resonance imaging and morphometric techniques.
      Toxicol Pathol 35: 27-40. http://dx.doi.org/10.1080/01926230601072343.
Carey. SA: Ballinger. CA: Plopper. CG: McDonald. RJ: Bartolucci. AA:  Postlethwait. EM:  Harkema. JR. (2011).
      Persistent rhinitis and epithelial remodeling induced by cyclic ozone exposure in the  nasal  airways of
      infant monkeys. Am J Physiol Lung Cell Mol Physiol 300:  L242-L254.
      http://dx.doi.org/10.1152/aiplung.00177.2010.
Carpagnano, GE: Foschino Barbaro, MP:  Cagnazzo, M: Pi Gioia, G: Giliberti, T: Pi Matteo,  C: Resta, O. (2005).
      Use of exhaled breath condensate in the study of airway inflammation after hypertonic saline solution
      challenge. Chest 128: 3159-3166. http://dx.doi.Org/10.1378/chest.128.5.3159.
Carter, WPL. (1995). Computer modeling of environmental chamber studies of maximum incremental
      reactivities of volatile organic compounds. Atmos Environ 29: 2513-2527.
Castagna, R; Davis, PA; Vasu, VT; Soucek, K; Cross, CE; Greci, L; Valacchi, G.  (2009). Nitroxide radical
      TEMPO reduces ozone-induced chemokine IL-8 production in lung epithelial cells. Toxicol  In Vitro 23:
      365-370. http://dx.doi.0rg/10.1016/i.tiv.2008.12.016.
Casteel, CL: O'Neill, BF: Zavala, JA: Bilgin, DP: Berenbaum, MR: DeLucia,  EH. (2008). Transcriptional profiling
      reveals elevated CO2 and elevated O-3 alter resistance of soybean (Glycine max) to Japanese beetles
      (Popillia japonica). Plant Cell Environ 31: 419-434. http://dx.doi.org/10.1111/i. 1365-3040.2008.01782.x.
Castilleios. M: Gold. PR: Damokosh. Al: Serrano. P: Allen.  G: McDonnell.  WF: Dockerv. D: Velasco. SR:
      Hernandez, M: Hayes, C. (1995). Acute effects of ozone on  the pulmonary function of exercising
      schoolchildren from Mexico City. Am J Respir Grit Care Med 152: 1501-1507.
Catalano. PJ: Rogus. J: Ryan. LM. (1995a). Consequences of prolonged inhalation of ozone on F344/N rats:
      collaborative studies Part X: robust composite scores based on median polish analysis.
Draft - Do Not Cite or Quote                        11-12                                  September 2011

-------
Catalano, PJ: L-YL, C: Harkema, JR: Kaden, DA: Last, JA: Mellick, PW: Parks, WC: Pinkerton, KE:
      Radhakrishnamurthy, B: Ryan, LM: Szarek, JL. (1995b). Consequences of prolonged inhalation of ozone
      on F344/N rats: Collaborative studies Part XI: Integrative summary. Cambridge, MA: Health Effects
      Institute.
CEMPD. (University of North Carolina at Chapel Hill, Center for Environmental Modeling for Policy
      Development). (2011). SMOKE (Version 2.7) [Computer Program]. Chapel Hill, NC. Retrieved from
      http://www.smoke-model.org/index.cfm
Chan. C. -C: Chuang. K. -J: Su. T. -C: Lin. L. -Y. (2005a). Association between nitrogen dioxide and heart rate
      variability in a susceptible population. Eur J Cardiovasc Prev Rehabil 12: 580-586.
Chan, C, -C: Wu, T, -H. (2005). Effects of ambient ozone exposure on mail carriers' peak expiratory flow rates.
      Environ Health Perspect 113: 735-738.
Chan, C, -C: Chuang, K, -J: Chien, L, -C: Chen, W, -J: Chang, W, -T. (2006). Urban air pollution and emergency
      admissions for cerebrovascular diseases in Taipei, Taiwan. Eur Heart J 27: 1238-1244.
Chan. E: Vet. RJ. (2010). Baseline levels and trends of ground level ozone in Canada and the United States.
      Atmos Chem Phys 10: 8629-8647. http://dx.doi.org/10.5194/acp-10-8629-2010.
Chan, WR:  Nazaroff, WW: Price, PN: Sohn, MD: Gadgil, AJ. (2005b). Analyzing a database of residential air
      leakage in the United States. Atmos Environ 39: 3445-3455.
Chang. C. -C: Tsai. S. -S: Ho. S. -C: Yang. C. -Y. (2005). Air pollution and hospital admissions for cardiovascular
      disease in Taipei, Taiwan. Environ Res 98: 114-119.
Chang, L: Miller, FJ: Ultman, J: Huang, Y: Stockstill,  Bl_: Grose, E: Graham, JA: Ospital, JJ: Crapo,  JD. (1991).
      Alveolar epithelial cell injuries by subchronic exposure to low concentrations of ozone correlate with
      cumulative exposure. Toxicol Appl Pharmacol 109: 219-234. http://dx.doi.org/10.1016/0041-
      008X(91)90170-J.
Chang. L. -T: Koutrakis. P: Catalano. PJ: Suh.  HH. (2000). Hourly personal exposures to fine particles and
      gaseous pollutants—Results from Baltimore, Maryland. J Air Waste Manag Assoc 50: 1223-1235.
Chang. L. -Y: Huang. Y: Stockstill. BL: Graham. JA: Grose. EC: Menache. MG: Miller. FJ: Costa.  PL: Crapo. JD.
      (1992). Epithelial injury and interstitial fibrosis in the proximal alveolar regions  of rats chronically exposed
      to a simulated pattern of urban ambient ozone. Toxicol Appl Pharmacol 115: 241-252.
      http://dx.doi.org/10.1016/0041-008X(92)90329-Q.
Chang, L, -Y: Stockstill, BL: Menache, MG: Mercer, RR: Crapo, JD. (1995). Consequences of prolonged
      inhalation of ozone on F344/N rats: Collaborative studies - Part VIII: morphometric analysis of structural
      alterations in alveolar regions.
Chang. MM. -J: Wu. R: Plopper. CG: Hyde. DM. (1998). IL-8 is one of the major chemokines produced by
      monkey airway epithelium after ozone-induced injury. Am J Physiol 275: L524-L532.
Chang. W: Liao. H: Wang. H. (2009).  Climate responses to direct radiative forcing of  anthropogenic aerosols,
      tropospheric ozone, and long-lived greenhouse gases in eastern China over 1951-2000. Adv Atmos Sci
      26: 748-762. http://dx.doi.org/10.1007/s00376-009-9032-4.
Chapman, JA: King, JS: Pregitzer, KS: Zak, DR. (2005). Effects of elevated concentrations of atmospheric CO2
      and tropospheric O-3 on decomposition of fine roots. Tree  Physiol 25:  1501-1510.
Chappelka,  A: Skelly, J: Somers, G: Renfro, J:  Hildebrand, E. (1999a). Mature black cherry used  as a
      bioindicator of ozone  injury. Water Air Soil Pollut 116: 261-266.
Chappelka.  A: Somers. G: Renfro. J. (1999b). Visible ozone injury on forest trees in Great Smoky Mountains
      National Park, USA. Water Air Soil Pollut 116: 255-260.
Chappelka.  AH: Samuelson. LJ. (1998). Ambient ozone effects on forest trees of the  eastern United States: A
      review [Review]. New Phytol 139: 91-108. http://dx.doi.Org/10.1046/i.1469-8137.1998.00166.x.
Chappelka,  AH. (2002). Reproductive development of blackberry  (Rubus cuneifolius) as influenced by ozone.
      New Phytol 155: 249-255. http://dx.doi.org/10.1046/j. 1469-8137.2002.00464.x.
Chappelka,  AH: Somers, Gl_: Renfro, JR. (2007). Temporal patterns of foliar ozone symptoms on tall milkweed
      (Asclepias exaltata L) in Great Smoky Mountains National  Park. Environ Pollut 149: 358-365.
      http://dx.doi.0rg/10.1016/i.envpol.2007.05.015.
Charpin. D:  Pascal. L: Birnbaum. J: Armengaud. A: Sambuc. R: Lanteaume. A: Vervloet. D. (1999).  Gaseous air
      pollution and atopy. Clin Exp Allergy 29: 1474-1480.
Chen. C: Ariomandi. M: Qin. H: Balmes. J: Tager. I: Holland. N. (2006a). Cytogenetic damage in buccal epithelia
      and peripheral lymphocytes of young healthy individuals exposed to ozone. Mutagenesis 21: 131-137.
      http://dx.doi. org/10.1093/mutage/gel007.
Chen, C: Ariomandi, M: Balmes, J: Tager, I: N,  H. (2007). Effects of chronic and acute ozone exposure on lipid
      peroxidation and antioxidant capacity in healthy young adults. Environ Health  Perspect 115:  1732-1737.
      http://dx.doi. org/10.1289/ehp. 10294.
Chen. CW:  Tsai. WT: Lucier. AA. (1998). A model of air-tree-soil system for ozone impact analysis. Ecol Modell
      111:  207-222.
Chen, J, -C: Schwartz, J. (2009). Neurobehavioral effects of ambient air pollution on cognitive performance in
      US adults. Neurotoxicology 30: 231-239. http://dx.doi.Org/10.1016/i.neuro.2008.12.011.
Chen. L: Jennison. BL: Yang. W: Omave. ST. (2000). Elementary school absenteeism and air pollution. Inhal
      Toxicol 12: 997-1016. http://dx.doi.org/10.1080/08958370050164626.
Draft - Do Not Cite or Quote                        11-13                                  September 2011

-------
Chen, L: Yang, W: Jennison, Bl_: Goodrich, A: Omaye, ST. (2002). Air pollution and birth weight in northern
      Nevada, 1991-1999. Inhal Toxicol 14: 141-157.
Chen, L: Bell, EM: Caton, AR: Druschel, CM: Lin, S. (201 Oa). Residential mobility during pregnancy and the
      potential for ambient air pollution exposure misclassification. Environ Res 110: 162-168.
      http://dx.doi.0rg/10.1016/i.envres.2009.11.001.
Chen. LH: Knutsen. SF: Shavlik. D: Beeson. WL: Petersen. F: Ghamsarv. M: Abbey. D. (2005). The association
      between fatal coronary heart disease and ambient particulate  air pollution: Are females at greater risk?
      Environ Health Perspect 113: 1723-1729.
Chen, P, -C;  Lai, Y, -M; Chan, C, -C;  Hwang, J, -S; Yang, C, -Y; Wang, J, -D.  (1999). Short-term effect of ozone
      on the pulmonary function of children in primary school. Environ Health Perspect 107: 921-925.
      http://dx.doi.org/10.1289/ehp.99107921.
Chen, X: Hopke, PK: Carter, WP (2011). Secondary organic aerosol  from ozonolysis of biogenic volatile organic
      compounds: Chamber studies of particle and reactive oxygen  species formation. Environ Sci Technol 45:
      276-282. http://dx.doi.org/10.1021/es102166c.
Chen. XQ: Yang. J: Hu. SP: Nie. HX:  Mao. GY:  Chen. HB. (2006b). Increased expression of CD86 and reduced
      production of IL-12 and  IL-10 by monocyte-derived dendritic cells from allergic asthmatics and their
      effects on Th1- and Th2-type cytokine balance. Respiration 73: 34-40.
      http://dx.doi.org/10.1159/000087457.
Chen, Z: Gallie, DR. (2005). Increasing tolerance to ozone by elevating foliar ascorbic acid confers greater
      protection against ozone than increasing avoidance. Plant Physiol 138: 1673-1689.
      http://dx.doi.org/10.1104/pp.105.062000.
Chen, Z: Wang, XK: Feng, ZZ: Xiao, Q: Duan, XN. (2009). Impact of  elevated O-3 on soil microbial community
      function under wheat crop. Water Air Soil Pollut 198: 189-198. http://dx.doi.org/10.1007/s11270-008-
      9838-1.
Chen, Z: Wang, XK: Yao, FF: Zheng,  FX: Feng, ZZ.  (201 Ob). Elevated ozone changed soil microbial community
      in a rice paddy. Soil Sci  Soc Am J 74: 829-837. http://dx.doi.org/10.2136/sssai2009.0258.
Cheng. FY: Burkev. KO: Robinson. JM: Booker. FL.  (2007). Leaf extracellular ascorbate in relation to O-3
      tolerance of two soybean cultivars. Environ Pollut 150: 355-362.
      http://dx.doi.0rg/10.1016/i.envpol.2007.01.022.
Chhabra, SK: Yasir, A:  Chaudhrv, K:  Shah,  B. (2010). Effect of ozone on response to ovalbumin & its modulation
      by vitamins C & E in sensitized guinea pigs. Indian J Med Res 132: 87-93.
Chimenti. L:  Morici. G:  Paterno. A: Bonanno. A:  Vultaggio. M: Bellia. V: Bonsignore. MR. (2009). Environmental
      conditions,  air pollutants, and airway cells in runners: A longitudinal field study. J Sports Sci 27: 925-935.
      http://dx.doi. org/10.1080/02640410902946493.
Ching. J: Herwehe. J: Swall. J. (2006). On joint  deterministic grid modeling and sub-grid variability conceptual
      framework for model evaluation. Atmos Environ 40: 4935-4945.
Chiu,  HF: Yang, CY (2009). Air pollution and emergency room visits for arrhythmias: Are there potentially
      sensitive groups? J Toxicol Environ Health A 72: 817-823.
      http://dx.doi. org/10.1080/15287390902800405.
Chiu,  HF: Cheng,  MH: Yang, CY. (2009). Air pollution and hospital admissions for pneumonia in a subtropical
      city: Taipei,  Taiwan. Inhal Toxicol 21: 32-37.
Cho. H. -Y: Zhang. L. -Y: Kleeberger. SR. (2001). Ozone-induced lung inflammation and hyperreactivity are
      mediated via tumor necrosis factor-"alpha" receptors. Am J Physiol 280:  L537-L546.
Cho. HY: Hotchkiss. JA: Harkema. JR. (1999). Inflammatory and epithelial responses during the development of
      ozone-induced mucous cell metaplasia in the nasal epithelium of rats. Toxicol Sci 51: 135-145.
Cho, HY: Kleeberger, SR. (2007). Genetic mechanisms of susceptibility to oxidative lung injury in mice. Free
      Radic Biol Med 42: 433-445. http://dx.doi.org/10.1016/i.freeradbiomed.20Q6.11.021.
Cho, HY: Morgan,  PL:  Bauer, AK: Kleeberger, SR. (2007). Signal transduction pathways of tumor necrosis
      factor—mediated lung injury induced by ozone in mice. Am J Respir Crit Care Med 175: 829-839.
      http://dx.doi.org/10.1164/rccm.200509-1527OC.
Cho. K: Shibato. J: Agrawal. GK: Jung. YH: Kubo. A: Jwa. NS: Tamogami. S: Satoh. K: Kikuchi. S: Higashi. T:
      Kimura, S: Sail,  H: Tanaka, Y: Iwahashi,  H: Masuo, Y: Rakwal, R. (2008). Integrated transcriptomics,
      proteomics, and metabolomics analyses to survey ozone responses in the leaves of rice seedling. J
      Proteome Res 7: 2980-2998.  http://dx.doi.org/10.1021/pr800128g.
Choi,  JH: Xu, QS:  Park, SY: Kim, JH: Hwang, SS: Lee, KH: Lee, HJ:  Hong, YC. (2007). Seasonal variation of
      effect of air pollution on  blood pressure. J Epidemiol Community Health 61: 314-318.
Christakos, G: Was, VM. (1998a). A composite space/time approach to studying ozone distribution over eastern
      United States. Atmos Environ 32: 2845-2857. http://dx.doi.org/10.1016/S1352-2310(98)00407-5.
Christakos. G: Vvas. VM. (1998b). A novel method for studying population health  impacts of spatiotemporal
      ozone distribution. Soc Sci Med 47: 1051-1066.
Christian, PL: Chen, LL: Scannell, CH: Ferrando, RE: Welch, BS: Balmes, JR. (1998). Ozone-induced
      inflammation is attenuated with multiday  exposure. Am J Respir Crit Care Med 158: 532-537.
Christiansen. B. (1999). Radiative forcing and climate sensitivity: The ozone experience. Q J Roy Meteorol Soc
      125: 3011-3035. http://dx.doi.org/10.1002/gi.49712556011.
Draft - Do Not Cite or Quote                        11-14                                  September 2011

-------
Christman, MA: Donovan, LA: Richards, JH. (2009). Magnitude of nighttime transpiration does not affect plant
      growth or nutrition in well-watered Arabidopsis. Physiol Plant 136: 264-273.
      http://dx.doi.0rg/10.1111/i.1399-3054.2009.01216.x.
Chuang, GC: Yang, Z: Westbrook, DG:  Pompilius, M: Ballinger, CA: White, RC: Krzywanski, DM: Postlethwait,
      EM: Ballinger. SW. (2009). Pulmonary ozone exposure induces vascular dysfunction, mitochondrial
      damage, and atherogenesis. Am J  Physiol Lung Cell Mol Physiol 297: L209-L216.
      http://dx.doi. org/10.1152/ajplung.00102.2009.
Chuang. K. -J: Chan.  C. -C: Su. T. -C: Lee. C. -T: Tang. C. -S. (2007a). The effect of urban air pollution on
      inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults. Am J Respir Grit
      Care Med 176: 370-376.
Chuang, K, -J: Yan, Y, -H: Cheng, T, -J.  (2010).  Effect of air pollution on blood pressure, blood lipids, and blood
      sugar: A population-based approach. J Occup Environ Med 52: 258-262.
      http://dx.doi. org/10.1097/JOM.Ob013e3181 ceff7a.
Chuang. KJ: Chan. CC: Su. TC:  Lin. LY: Lee. CT. (2007b). Associations between particulate sulfate and organic
      carbon exposures and heart rate variability in patients with or at risk for cardiovascular diseases. J
      Occup Environ Med 49: 610-617.
Chuang. KJ: Yan. YH: Chiu. SY:  Cheng. TJ. (2011). Long-term air pollution exposure and risk factors for
      cardiovascular diseases among the elderly in Taiwan. Occup Environ Med 68: 64-68.
      http://dx.doi.org/10.1136/oem.2009.052704.
Chung, HG: Zak, PR: Lilleskov, EA. (2006). Fungal community composition and metabolism under elevated
      CO2 and O-3. Oecologia  147: 143-154. http://dx.doi.org/10.1007/s00442-005-0249-3.
Civerolo, KL: Mao, HT: Rao, ST.  (2003). The airshed for ozone and  fine particulate pollution in the eastern
      United  States.  Pure Appl Geophys  160: 81-105.
Clark. NA: Demers. PA: Karr. CJ: Koehoorn. M: Lencar. C: Tamburic. L: Brauer. M. (2010). Effect of early life
      exposure to air pollution on development of childhood asthma.  Environ Health Perspect 118: 284-290.
      http://dx.doi.org/10.1289/ehp.0900916.
Clarke. LJ: Robinson. SA. (2008). Cell wall-bound ultraviolet-screening compounds explain the high ultraviolet
      tolerance of the Antarctic  moss, Ceratodon purpureus. New Phytol 179: 776-783.
      http://dx.doi.0rg/10.1111/i.1469-8137.2008.02499.x.
demons, GK: Garcia, JF. (1980). Changes in thyroid function after short-term ozone exposure in rats. J Environ
      Pathol Toxicol Oncol 4: 359-369.
Clydesdale. GJ: Dandle.  GW: Mullen HK.  (2001). Ultraviolet light induced injury: Immunological and
      inflammatory effects. Immunol Cell Biol 79: 547-568.
Cockcroft, DW:  Davis, BE: Todd, DC: Smycniuk, AJ. (2005). Methacholine challenge: Comparison of two
      methods. Chest 127: 839-844.
Coffin. PL: Blommer. EJ: Gardner. DE: Holzman. R. (1967). Effect of air pollution on alteration of susceptibility
      to pulmonary infection. Cincinnati, OH: U.S. Department of Health, Education, and Welfare.
Coffin, PL: Gardner, DE.  (1972).  Interaction of biological agents and chemical air pollutants. Ann  Occup Hyg 15:
      219-234.
Cohen-Hubal, EA: Kimbell, JS: Fedkiw,  PS. (1996). Incorporation of nasal-lining mass-transfer resistance into a
      CFD model for prediction  of ozone dosimetry in the upper respiratory tract. Inhal Toxicol 8: 831-857.
Cohen. MD: Sisco. M: Baker. K:  Li. Y: Lawrence. D: Van Loveren. H: Zelikoff. JT: Schlesinger. RB. (2002).
      Effects of inhaled ozone on pulmonary immune cells  critical to antibacterial responses in situ. Inhal
      Toxicol 14: 599-619.
Cole. MP: Freeman. BA.  (2009).  Promotion of cardiovascular disease by exposure to the air pollutant ozone.
      Am J Physiol Lung Cell Mol Physiol 297: L209-L216.
Coleridge, HM: Coleridge, JCG:  Ginzel, KH: Baker, DG: Banzett,  RB:  Morrison, MA. (1976). Stimulation of
      "irritant1 receptors and afferent C-fibers in the lungs by prostaglandins. Nature 264: 451-453.
Coleridge, JCG: Coleridge, HM: Schelegle, ES: Green, JF. (1993). Acute inhalation of ozone stimulates
      bronchial C-fibers and rapidly adapting receptors in dogs. J Appl Physiol 74:  2345-2352.
Colfn-Barengue. L: Dorado-Martinez. C: Rivas-Arancibia. S: Avila-Costa. MR: Fortoul. Tl. (2005). Morphological
      recovery of the granule cells from the olfactory bulb after the cessation of acute ozone exposure. Int J
      Neurosci 115: 411-421. http://dx.doi.org/10.1080/00207450590521028.
Colls. JJ: Unsworth. MH. (1992). Air pollution interactions with natural  stressors. In JR Barker; DT Tingey (Eds.),
      Air pollution effects on biodiversity. New York, NY: Van Nostrand Reinhold.
Conrad, R: Seller, W.  (1985). Influence of temperature, moisture, and  organic carbon on the flux of H2 and CO
      between soil and atmosphere: Field studies in subtropical regions. J Geophys Res 90: 5699-5709.
Conti, DV: Cortessis, V: Molitor, J: Thomas, DC. (2003). Bayesian modeling of complex metabolic pathways.
      Hum Hered 56: 83-93. http://dx.doi.org/10.1159/000073736.
Cooper. OR: Oltmans. SJ: Johnson. BJ: Brioude. J: Angevine. W: Trainer. M: Parrish. DP: Rverson. TR: Pollack.
      I: Cullis, PD: Ives, MA: Tarasick,  DW: Al-Saadi, J: Stajner, I. (In Press) Measurement of western U.S.
      baseline ozone from the surface  to the tropopause and assessment of downwind impact regions. J
      Geophys Res.
Draft - Do Not Cite or Quote                        11-15                                  September 2011

-------
Cooper, OR: Parrish, DP: Stohl, A: Trainer, M: Nedelec, P: Thouret, V: Gammas, JP: Oltmans, SJ: Johnson, BJ:
      Tarasick, D: Leblanc, T: McDermid, IS: Jaffe, D: Gao, R: Stith, J: Ryerson, T: Aikin, K: Campos, T:
      Weinheimer, A: Avery, MA. (2010). Increasing springtime ozone mixing ratios in the free troposphere over
      western North America. Nature 463: 344-348. http://dx.doi.Org/10.1038/nature08708.
Cornelissen. T. (2011). Climate change and its effects  on terrestrial insects and herbivory patterns. Neotrop
      Entomol 40:  155-163. http://dx.doi.orq/10.1590/S1519-566X2011000200001.
Corradi, M: Alinovi,  R: Goldoni, M: Vettori, M: Folesani, G:  Mozzoni, P: Cavazzini, S: Bergamaschi, E: Rossi,  L:
      Mutti. A. (2002). Biomarkers of oxidative stress after controlled human exposure to ozone. Toxicol Lett
      134: 219-225.
Corradi, M: Folesani, G: Andreoli, R: Manini,  P: Bodini, A: Piacentini, G: Carraro, S: Zanconato, S: Baraldi, E.
      (2003). Aldehydes and glutathione in exhaled breath condensate of children with asthma exacerbation.
      Am J  Respir Grit Care Med 167: 395-399. http://dx.doi.org/10.1164/rccm.200206-507OC.
Corsmeier, U: Kalthhoff, N: Kolle, O: Motzian, M: Fiedler, F. (1997). Ozone concentration jump in the stable
      nocturnal boundary layer during a LLJ-event. Atmos Environ 31: 1977-1989.
Costa. PL: Schafrank. SN: Wehner. RW: Jellett. E. (1985). Alveolar permeability to protein in rats differentially
      susceptible to ozone. J  Appl Toxicol 5: 182-186. http://dx.doi.org/10.1002/iat.25500503Q9.
Costa. PL: Folinsbee. LJ: Raub.  JA: Tilton. B: Tingev.  DT. (1992). Summary of selected new information on
      effects of ozone on health and vegetation: Supplement to 1986 air quality criteria for ozone and other
      photochemical oxidants. (EPA/600/8-88/105F).  Research Triangle Park, NC: U.S. Environmental
      Protection Agency.
Coulston, JW: Smith, GC: Smith, WD. (2003). Regional assessment of ozone sensitive tree species using
      bioindicator plants. Environ Monit Assess 83: 113-127.
Cremillieux. Y: Servais. S: Berthezene. Y: Dupuich. D: Boussouar. A: Stupar. V: Pequignot. JM. (2008). Effects
      of ozone exposure in rat lungs investigated with hyperpolarized 3He MRI. J Magn Reson Imaging 27:
      771-776. http://dx.doi.org/10.1002/imri.21216.
Crist.  KG: Carmichael. GR: John. K. (1994). UV-B exposure and atmospheric ozone - Evaluation of radiative
      flux to changes in  ambient ozone  levels. J Hazard Mater 37: 527-538. http://dx.doi.org/10.1016/0304-
      3894(93)E0096-K.
Cross, CE: Motchnik, PA: Bruener, BA: Jones, DA: Kaur, H: Ames, BN: Halliwell, B. (1992). Oxidative damage
      to plasma constituents by ozone.  FEBS Lett 298: 269-272.  http://dx.doi.org/10.1016/0014-
      5793(92)80074-Q.
Croteau. MC: Martvniuk.  CJ: Trudeau. VL: Lean. DRS. (2008a). Chronic exposure of rana pipiens tadpoles to
      uvb radiation and the estrogenic chemical 4-tert-octylphenol. J Toxicol Environ Health A 71: 134-144.
Croteau, MC: Davidson,  MA: Lean, DRS: Trudeau, VL. (2008b). Global increases in ultraviolet B radiation:
      Potential impacts on amphibian development and metamorphosis. Physiol Biochem Zool 81: 743-761.
      http://dx.doi. org/10.1086/591949.
Crous, KY: Vandermeiren, K: Ceulemans, R. (2006). Physiological responses to cumulative ozone uptake  in two
      white clover (Trifolium repens L. cv. Regal) clones with  different ozone sensitivity. Environ Exp Bot 58:
      169-179. http://dx.doi.0rg/10.1016/i.envexpbot.2005.07.007.
Curriero, FC: Heiner, KS: Samet, JM: Zeger,  SL: Strug, L: Patz, JA. (2002). Temperature and mortality in 11
      cities of the eastern United States. Am J Epidemiol  155: 80-87. http://dx.doi. org/10.1093/aie/155.1.80.
D'Anna. B: Jammoul. A: George. C: Stemmler. K: Fahrni. S: Ammann. M: Wisthaler. A. (2009). Light-induced
      ozone depletion by humic acid films and submicron aerosol particles. J Geophys Res 114: D12301.
      http://dx.doi.org/10.1029/2008JD011237.
D'Haese. D:  Vandermeiren. K:  Asard. H:  Horemans. N. (2005). Other factors than apoplastic ascorbate
      contribute to the differential ozone tolerance of  two clones of Trifolium repens L. Plant Cell Environ 28:
      623-632. http://dx.doi.0rg/10.1111/i.1365-3040.2005.01308.x.
Dadvand, P: Rankin, J: Rushton, S: Pless-Mulloli, T. (2011). Ambient air pollution and congenital heart disease:
      A register-based study.  Environ Res 111: 435-441. http://dx.doi.Org/10.1016/i.envres.2011.01.022.
Dahl. AR. (1990). Dose concepts for inhaled vapors and gases. Toxicol Appl Pharmacol 103: 185-197.
Dahl.  M: Bauer. AK: Arredouani.  M: Soininen. R: Tryggvason.  K: Kleeberger. SR: Kobzik. L. (2007). Protection
      against inhaled oxidants through scavenging of oxidized lipids by macrophage receptors MARCO and
      SR-AI/II. J Clin Invest 117: 757-764. http://dx.doi.org/10.1172/JCI29968.
Dales. R: Burnett. RT: Smith-Doiron.  M: Stieb. DM: Brook. JR. (2004). Air pollution and sudden infant death
      syndrome. Pediatrics 113: 628-631.
Dales, R: Chen, L: Frescura, AM: Liu, L:  Villeneuve, PJ. (2009). Acute effects of outdoor air pollution on forced
      expiratory volume in 1 s: A panel study of schoolchildren with asthma. Eur Respir J 34: 316-323.
      http://dx.doi. org/10.1183/09031936.00138908.
Dales. RE: Cakmak. S: Doiron. MS. (2006).  Gaseous  air pollutants and hospitalization for respiratory disease in
      the neonatal period. Environ Health Perspect 114: 1751-1754. http://dx.doi.org/10.1289/ehp.9044.
Dallmann, TR: Harlev, RA. (2010). Evaluation of mobile source emission trends in the United States. J Geophys
      Res 115: D14305. http://dx.doi.Org/10.1029/201OJD013862.
Dalstein. L: Vas. N.  (2005). Ozone concentrations and ozone-induced symptoms on coastal and alpine
      mediterranean pines in  southern France. Water Air Soil Pollut 160: 181-195.
Draft - Do Not Cite or Quote                        11-16                                 September 2011

-------
Damera, G; Zhao, H; Wang, M; Smith, M; Kirby, C; Jester, WF; Lawson, JA; Panettieri, RA, Jr. (2009). Ozone
      modulates IL-6 secretion in human airway epithelial and smooth muscle cells. Am J Physiol Lung Cell
      Mol Physiol 296: L674-L683. http://dx.doi.org/10.1152/aiplung.90585.2008.
Damera, G: Jester William, F: Jiang, M: Zhao, H: Fogle Homer, W: Mittelman, M: Haczku, A: Murphy, E: Parikh,
      I: Panettieri Reynold. A. (2010). Inhibition of myristoylated alanine-rich C kinase substrate (MARCKS)
      protein inhibits ozone-induced airway neutrophilia and inflammation. Exp Lung Res 36: 75-84.
      http://dx.doi.org/10.3109/01902140903131200.
Dammgen. U: Grunhage. L: Haenel. H. -D: Jager. H. -J. (1993). Climate and stress in ecotoxicology: A coherent
      system of definitions and terms. J Appl Bot Food Qual 67: 157-162.
Darbah, JNT: Kubiske, ME: Neilson, N; Oksanen, E: Vaapavuori, E: Karnosky, DF. (2007). Impacts of elevated
      atmospheric CO2 and O3 on paper birch (Betula papyrifera): Reproductive fitness.
      ScientificWorldJournal 7: 240-246. http://dx.doi.Org/10.110O/tsw.2007.42.
Darbah, JNT: Kubiske, ME: Nelson, N; Oksanen, E: Vapaavuori, E: Kamoskv, DF. (2008). Effects of decadal
      exposure to interacting elevated CO2 and/or O-3 on paper birch (Betula papyrifera) reproduction.
      Environ  Pollut 155: 446-452. http://dx.doi.Org/10.1016/i.envpol.2008.01.033.
Darrow, LA: Klein, M:  Flanders, WD: Waller, LA: Correa, A: Marcus, M:  Mulholland, JA: Russell, AG: Tolbert, PE.
      (2009). Ambient air pollution and preterm birth: A time-series analysis. Epidemiology 20: 689-698.
Darrow. LA: Klein. M:  Strickland.  MJ:  Mulholland. JA: Tolbert. PE. (2011a). Ambient air pollution and birth weight
      in full-term infants  in Atlanta, 1994-2004. Environ Health Perspect 119: 731-737.
      http://dx.doi.org/10.1289/ehp.1002785.
Darrow, LA: Klein, M:  Sarnat, JA: Mulholland, JA: Strickland,  MJ: Sarnat,  SE: Russell, AG: Tolbert, PE. (2011 b).
      The use of alternative pollutant metrics in time-series studies of ambient air pollution and respiratory
      emergency department visits. J Expo Sci Environ Epidemiol 21:  10-19.
      http://dx.doi.org/10.1038/ies.2009.49.
David, GL: Romieu, I: Sienra-Monge, JJ: Collins, WJ: Ramirez-Aguilar,  M: Del Rio-Navarro, BE: Reyes-Ruiz, Ml:
      Morris. RW: Marzec. JM: London. SJ.  (2003). Nicotinamide adenine dinucleotide (phosphate)
      reduced:quinone oxidoreductase and glutathione s-transferase ml polymorphism and childhood asthma.
      Am J Respir Grit Care Med 168: 1199-1204.
Davidson, A. (1993). Update of ozone trends in California's South Coast Air Basin. J Air Waste Manag Assoc 43:
      226-227.
Davis, DP: Orendovici, T. (2006). Incidence of ozone symptoms on vegetation within a National Wildlife Refuge
      in New Jersey, USA. Environ Pollut 143: 555-564. http://dx.doi.orq/10.1016/i.envpol.20Q5.10.051.
Davis. DP. (2007a). Ozone-induced symptoms on vegetation within the Moosehorn National Wildlife Refuge in
      Maine. Northeast Nat 14: 403-414. http://dx.doi.org/10.1656/1092-
      6194(2007)14r403:OSOVWT12.0.CO:2.
Davis. DP. (2007b). Ozone injury to plants within the Seney National Wildlife Refuge in northern Michigan.
      Northeast Nat 14: 415-424.
Davis, DP. (2009). Ozone-induced stipple on plants in the Cape Remain National Wildlife Refuge, South
      Carolina. Southeastern Naturalist 8: 471-478.
Pawson, TE: Burgess, SS: Tu, KP: Oliveira, RS: Santiago, LS: Fisher, JB: Simonin, KA: Ambrose, AR. (2007).
      Nighttime transpiration in woody plants from contrasting ecosystems. Tree Physiol 27: 561-575.
      http://dx.doi. orq/10.1093/treephvs/27.4.561.
de Gouw, JA: Brock, CA: Atlas, EL: Bates, TS: Fehsenfeld, FC: Goldan, PP: JS, H: Kuster, WC: Lerner, BM:
      Matthew. BM: Middlebrook. AM: Onasch. TB: Peltier. RE: Quinn. PK: Senff. CJ: Stohl. A: Sullivan. AP:
      Trainer.  M: Warneke. C: Weber. RJ: Williams. EJ. (2008). Sources  of particulate matter in the
      northeastern United States in summer: 1. Pirect emissions and secondary formation of organic matter in
      urban plumes.  J Geophys Res 113: P08301.
Pe Gruijl, FR. (1995). Action spectrum for photocarcinogenesis. Recent Results Cancer Res 139: 21-30.
Pe Temmerman, L: Legrand, G: Vandermeiren, K. (2007). Effects of ozone on sugar beet grown in open-top
      chambers. Eur J Agron 26: 1-9. http://dx.doi.Org/10.1016/i.eia.2006.08.001.
de Lourdes de  Bauer. M: Hernandez-Teieda.  T. (2007). A review of ozone-induced  effects on the forests of
      central Mexico [Review]. Environ Pollut 147: 446-453.  http://dx.doi.Org/10.1016/i.envpol.2006.12.020.
Pe Pablo. F: Lopez. A: Soriano. LR: Tomas. C: Piego.  L: Gonzalez. M:  Barrueco. M. (2006). Relationships of
      daily mortality and hospital admissions to air pollution in Castilla-Leon, Spain. Atmosfera 19: 23-39.
Pegl'lnnocenti, E: Guidi, L: Soldatini, GF (2007). Effects of elevated ozone on chlorophyll a fluorescence in
      symptomatic and asymptomatic leaves of two tomato genotypes. Biol Plantarum 51: 313-321.
      http://dx.doi. org/10.1007/s10535-007-0061 -5.
Pelaunois, A: Segura, P: Montano, LM: Vargas, MH: Ansay, M: Gustin,  P.  (1998). Comparison of ozone-induced
      effects on lung mechanics and hemodynamics in the rabbit. Toxicol Appl Pharmacol 150: 58-67.
Pelfino. RJ: Coate. BP: Zeiger. RS: Seltzer. JM: Street. PH: Koutrakis.  P.  (1996). Paily asthma severity in
      relation to personal ozone exposure and outdoor fungal spores.  Am J Respir Crit Care Med 154: 633-
      641.
Pelfino, RJ; Zeiger, RS; Seltzer, JM; Street, PH; Matteucci, RM; Anderson, PR; Koutrakis,  P. (1997). The effect
      of outdoor fungal spore concentrations on daily asthma severity. Environ Health Perspect 105: 622-635.
Draft - Po Not Cite or Quote                        11-17                                  September 2011

-------
Delfino, RJ; Zeiger, RS; Seltzer, JM; Street, DH; McLaren, CE. (2002). Association of asthma symptoms with
      peak particulate air pollution and effect modification by anti-inflammatory medication use. Environ Health
      Perspect 110: A607-A617.
Delfino, RJ: Gone, H: Linn, WS: Pellizzari, ED: Hu, Y. (2003). Asthma symptoms in Hispanic children and daily
      ambient exposures to toxic and criteria air pollutants.  Environ Health Perspect 111: 647-656.
      http://dx.doi.org/10.1289/ehp.5992.
Delfino, RJ: Quintana, PJE: Floro, J: Gastanaga, VM: Samimi, BS: Kleinman, MT: Liu, LJS: Bufalino, C: Wu,
      CF: McLaren. CE. (2004). Association of FEV1 in asthmatic children with personal and
      microenvironmental exposure to airborne particulate matter. Environ Health Perspect 112: 932-941.
Delfino, RJ: Staimer, N: Tjoa, T: Arhami, M: Polidori,  A: Gillen, PL: George, SC: Shafer, MM: Schauer, JJ:
      Sioutas, C. (201 Oa). Associations of primary and secondary organic aerosols with airway and aystemic
      inflammation in an elderly panel cohort. Epidemiology 21: 892-902.
      http://dx.doi. org/10.1097/EDE.Ob013e3181 f20e6c.
Delfino. RJ: Tioa. T: Gillen. PL: Staimer. N: Polidori. A: Arhami. M: Jamner. L: Sioutas. C: Longhurst. J. (201 Ob).
      Traffic-related air pollution and  blood pressure in elderly subjects with coronary artery disease.
      Epidemiology 21: 396-404. http://dx.doi.org/10.1097/EDE.Ob013e3181d5e19b.
Delfino. RJ: Gillen. PL: Tioa. T: Staimer. N: Polidori. A: Arhami. M: Sioutas. C: Longhurst. J. (2011).
      Electrocardiographic ST-segment depression  and exposure to traffic-related aerosols in elderly subjects
      with coronary artery disease. Environ Health Perspect 119: 196-202.
      http://dx.doi.org/10.1289/ehp.1002372.
Dell'Omo, G: Wolfer, D: Alleva, E: Lipp, H, -P.  (1995). Developmental exposure to  ozone induces subtle
      changes in swimming navigation of adult mice. Toxicol Lett 81: 91-99.
DeLucia. AJ: Adams. WC. (1977). Effects of O3 inhalation during exercise on pulmonary function and blood
      biochemistry. J Appl Physiol 43: 75-81.
Dempsey, J, . A.: Johnson, B, . P.: Saupe, K, . W. (1990). Adaptations and limitations in the pulmonary system
      during exercise. Chest 97: 81S-87S.
Dempsev. J. . A.: McKenzie.  D. .  C.: Haverkamp. H. . C.: Eldridge. M. . W. (2008).  Update in the understanding
      of respiratory limitations to exercise performance in fit, active  adults. Chest 134: 613-622.
      http://dx.doi. org/10.1378/chest. 07-2730.
Dennekamp, M: Akram, M: Abramson, MJ: Tonkin, A: Sim, MR: Fridman, M: Erbas, B. (2010). Outdoor air
      pollution as a trigger for out-of-hospital cardiac arrests. Epidemiology 21: 494-500.
      http://dx.doi. org/10.1097/EDE.Ob013e3181 e093db.
Depuvdt. P: Joos. GF:  Pauwels. RA. (1999). Ambient ozone concentrations induce airway hyperresponsiveness
      in some rat strains. Eur Respir  J 14: 125-131.
Dermodv. O: O'Neill. BF: Zangerl. AR: Berenbaum. MR: DeLucia. EH. (2008). Effects of elevated CO2 and O3
      on leaf damage and insect abundance  in a soybean agroecosystem. Arthropod-Plant Inte 2: 125-135.
Derwent, RG: Collins, WJ: Johnson, CE: Stevenson, PS. (2001). Transient behaviour of tropospheric ozone
      precursors in a global 3-D CTM and their indirect greenhouse effects. Clim  Change 49: 463-487.
Devlin, RB:  McDonnell, WF:  Mann, R: Becker, S: House,  DE: Schreinemachers, D: Koren, HS.  (1991).
      Exposure of humans to ambient levels  of ozone for 6.6 hours causes cellular and biochemical changes in
      the lung. Am J Respir Cell Mol  Biol  4: 72-81.
Devlin. RB:  McDonnell. WF:  Becker. S: Madden. MC: McGee. MP: Perez. R: Hatch.  G: House.  DE: Koren.  HS.
      (1996). Time-dependent changes of inflammatory mediators in the lungs of humans exposed to 0.4 ppm
      ozone for 2 hr: A comparison of mediators found in bronchoalveolar lavage fluid 1 and 18 hr after
      exposure. Toxicol Appl Pharmacol 138:  176-185.
Devlin, RB:  Folinsbee,  LJ: Biscardi, F: Hatch,  G:  Becker, S: Madden, MC: Robbins, M: Koren, HS. (1997).
      Inflammation and cell damage  induced by repeated exposure of humans to ozone. Inhal Toxicol 9: 211-
      235.
Pi Baccio, D: Castagna, A: Paoletti, E: Sebastiani, L: Ranieri, A. (2008). Could the differences in O3 sensitivity
      between two poplar clones be related to a difference in antioxidant defense and secondary metabolic
      response to O3 influx? Tree Physiol 28: 1761-1772.
Diaz, J: Linares, C: Garcia-Herrera, R: Lopez, C: Trigo, R. (2004). Impact of temperature and air pollution on the
      mortality of children in Madrid. J Occup Environ Med 46: 768-774.
Dickerson. RR: Rhoads. KP: Carsev. TP: Oltmans. SJ: Burrows. JP:  Crutzen. PJ.  (1999). Ozone in the remote
      marine boundary layer: A possible role  for halogens. J Geophys Res 104: 21,385-321,395.
Dickson, RE: Lewin, KF: Isebrands, JG: Coleman, MD: Heilman, WE: Riemenschneider, DE:  Sober, J: Host,
      GE: Zak, PR: Hendrey, GR:  Pregitzer, KS: Karnosky,  PR (2000). Forest Atmosphere Carbon Transfer
      and Storage (FACTS-II) the Aspen Free-Air CO2 and O3 Enrichment (FACE) project: An overview.
      (General Technical Report NC-214). St. Paul,  MN: U.S. Dept.  of Agriculture, Forest Service.
      http://nrs.fs.fed.us/pubs/278.
Diemer, T: Allen, JA: Hales, KH: Hales, DB. (2003). Reactive oxygen disrupts mitochondria in MA-10 tumor
      Leydig cells and inhibits steroidogenic acute regulatory (StAR) protein and  steroidogenesis.
      Endocrinology 144: 2882-2891. http://dx.doi.org/10.1210/en.2002-0090.
Diepgen, TL: Mahler, V. (2002). The epidemiology of skin cancer. Br  J Dermatol 146: 1-6.
Draft - Do Not Cite or Quote                       11-18                                 September 2011

-------
Dietert, RR: DeWitt, JC: Germolec, PR: Zelikoff, JT. (2010). Breaking patterns of environmentally influenced
      disease for health risk reduction: Immune perspectives. Environ Health Perspect 118:  1091-1099.
      http://dx.doi.org/10.1289/ehp.1001971.
Dimeo, MJ: Glenn, MG: Holtzman, MJ: Sheller, JR: Nadel, JA: Boushey, HA. (1981). Threshold concentration of
      ozone causing an increase in bronchial reactivity in humans and adaptation with repeated exposures. Am
      Rev Respir Dis 124: 245-248.
Dimitriadis, VK. (1992). Carbohydrate cytochemistry of bonnet monkey (Macaca radiata) nasal epithelium:
      Response to ambient levels of ozone. Histol Histopathol 7: 479-488.
Ditchkoff. SS: Lewis. JS: Lin. JC: Muntifering. RB: Chappelka. AH. (2009). Nutritive quality of highbush
      blackberry (Rubus argutus) exposed to tropospheric ozone. Rangeland Ecol Manag 62: 364-370.
Dizengremel, P: Sasek, T: Brown, K: Richardson,  C. (1994). Ozone-induced changes in primary carbon
      metabolism enzymes of loblolly pine needles. J Plant Physiol 144: 300-306.
Dizengremel, P: Le Thiec, D: Bagard, M: Jolivet, Y. (2008). Ozone risk assessment for plants: Central role of
      metabolism-dependent changes in reducing power. Environ Pollut 156: 11-15.
      http://dx.doi.0rg/10.1016/i.envpol.2007.12.024.
Dizengremel, P: Le Thiec, D: Hasenfratz-Sauder, MP: Vaultier, MM: Bagard, M: Jolivet, Y.  (2009). Metabolic-
      dependent changes in plant cell redox power after ozone exposure. Plant Biol (Stuttg) 11: 35-42.
      http://dx.doi.0rg/10.1111/i.1438-8677.2009.00261.x.
Dobson, HEM. (1994). Floral volatiles in insect biology. In EA Bernays (Ed.), Insect-plant interactions: Vol 5 (pp.
      47-82). Boca Raton, FL: CRC Press.
Docherty, KS: Wu, W: Lim, YB: Ziemann, PJ. (2005). Contributions of organic peroxides to secondary aerosol
      formed from reactions of monoterpenes with O3. Environ Sci Technol 39: 4049-4059.
      http://dx.doi. org/10.1021/es050228s.
Dockerv. DW: Pope.  CA. Ill:  Xu. X: Spengler. JD: Ware. JH: Fay. ME: Ferris. BG. Jr: Speizer. FE. (1993). An
      association between air pollution and mortality in six US cities. N Engl J  Med 329: 1753-1759.
Dockerv. DW: Luttmann-Gibson. H: Rich. DQ: Link. MS: Mittleman. MA: Gold.  PR: Koutrakis. P: Schwartz. JD:
      Verrier. RL. (2005). Association of air pollution with increased incidence of ventricular tachyarrhythmias
      recorded by implanted cardioverter defibrillators. Environ Health Perspect 113: 670-674.
Dohm, MR: Mautz, WJ:  Looby, PG: Gellert, KS: Andrade, JA. (2001). Effects of ozone on evaporative water loss
      and thermoregulatory behavior of marine toads (Bufo marinus). Environ  Res 86: 274-286.
Dohm, MR: Mautz, WJ: Andrade, JA: Gellert, KS: Salas-Ferguson, LJ: Nicolaisen,  N: Fujie, N. (2005). Effects of
      ozone exposure on nonspecific phagocytic capacity of pulmonary macrophages from an amphibian, Bufo
      marinus. Environ Toxicol Chem 24: 205-210.
Dohm, MR: Mautz, WJ:  Doratt, RE: Stevens, JR. (2008). Ozone exposure affects feeding  and locomotor
      behavior of adult Bufo marinus. Environ Toxicol Chem 27: 1209-1216. http://dx.doi.org/10.1897/07-388.1.
Dohrmann. AB: Tebbe. CC. (2005). Effect of elevated tropospheric ozone on the structure of bacterial
      communities inhabiting the rhizosphere of herbaceous plants native to Germany. Appl Environ Microbiol
      71:  7750-7758. http://dx.doi.org/10.1128/AEM.71.12.7750-7758.2005.
Dorado-Martinez, C:  Parades-Carbajal, C: Mascher, D: Borgonio-Perez, G: Rivas-Arancibia, S. (2001). Effects
      of different ozone doses on memory, motor activity and lipid peroxidation levels, in  rats. Int J Neurosci
      108: 149-161.
Dormans. JAM. A: Van Bree. L: Boere. AJF: Marra. M: Rombout. PJA. (1999).  Interspecies differences in time
      course of pulmonary toxicity following repeated exposure to ozone. Inhal Toxicol 11: 309-329.
Poster!  C: Petrilli. V: Van Bruggen. R: Steele. C: Mossman. BT: Tschopp. J. (2008). Innate immune activation
      through NalpS inflammasome sensing of asbestos and silica. Science 320: 674-677.
Doyle, M: Sexton, KG: Jeffries, H: Bridge, K: Jaspers, I. (2004). Effects of 1,3-butadiene, isoprene, and their
      photochemical degradation products on human lung cells. Environ Health Perspect 112: 1488-1495.
Doyle, M: Sexton, KG: Jeffries, H: Jaspers, I. (2007). Atmospheric photochemical transformations enhance 1,3-
      butadiene-induced inflammatory responses in human epithelial cells: The role of ozone and other
      photochemical degradation products. Chem Biol  Interact 166: 163-169.
      http://dx.doi.0rg/10.1016/i.cbi.2006.05.016.
Drechsler-Parks, DM. (1995). The dose-response relationship in older men exposed to ozone. Exp Gerontol 30:
      65-75.
Driscoll.  KE: Vollmuth. TA: Schlesinger. RB. (1987). Acute and subchronic ozone inhalation in the rabbit:
      Response of alveolar macrophages. J Toxicol Environ Health 21: 27-43.
      http://dx.doi. org/10.1080/15287398709531 OOP.
Drogoudi, PD: Ashmore,  MR. (2000). Does elevated ozone have differing effects in flowering and deblossomed
      strawberry? New Phytol 147: 561-569. http://dx.doi.org/10.1046/j. 1469-8137.2000.00718.x.
Drogoudi. PD: Ashmore.  M. (2001). 14C-allocation of flowering and deblossomed strawberry in response to
      elevated ozone. New Phytol 152: 455-461.  http://dx.doi.Org/10.1046/i.0028-646X.2001.00270.x.
Dryden,  DM: Spooner, CH: Stickland, MK: Vandermeer, B: Tjosvold, L: Bialv, L: Wong, K:  Rowe, BH.  (2010).
      Exercise-induced bronchoconstriction and asthma. In Evidence Report/Technology Assessment. (AHRQ
      Publication No. 10-E001).  Rockville, MD: Agency for Healthcare Research and Quality.
Draft - Do Not Cite or Quote                       11-19                                 September 2011

-------
Duan, X: Buckpitt, AR: Plopper, CG. (1993). Variation in antioxidant enzyme activities in anatomic
      subcompartments within rat and rhesus monkey lung. Toxicol Appl Pharmacol 123: 73-82.
Duan, X: Buckpitt, AR: Pinkerton, KE: Ji,  C: Plopper, CG. (1996). Ozone-induced alterations in glutathione in
      lung subcompartments of rats and monkeys. Am J Respir Cell Mol Biol 14: 70-75.
Dudareva. N: Negre. F: Nagegowda. DA: Orlova. I. (2006). Plant volatiles: Recent advances and future
      perspectives [Review]. Grit Rev Plant Sci 25: 417-440.
Dugandzic, R: Dodds, l_:  Stieb, D: Smith-Doiron, M. (2006). The association between low level exposures to
      ambient air pollution and term low birth weight: A retrospective cohort study. Environ Health 5: 3.
      http://dx.doi.org/10.1186/1476-069X-5-3.
Duncan, BN: Yoshida, Y:  Olson, JR: Sillman, S: Martin, RV: Lamsal, L: Hu, Y:  Pickering, KE: Retscher, C: Allen,
      DJ. (2010). Application of OMI observations to a space-based indicator of NOx and VOC  controls on
      surface ozone formation. Atmos Environ 44: 2213-2223.
      http://dx.doi.0rg/10.1016/i.atmosenv.2010.03.010.
Dunker. A: Yarwood. G: Ortmann. J: Wilson. G. (2002). The decoupled direct method for sensitivity analysis in a
      three-dimensional air quality model implementation, accuracy, and efficiency. Environ Sci Technol  36:
      2965-2976. http://dx.doi.org/10.1021/es0112691.
Dunker. AM. (1981). Efficient calculation  of sensitivity coefficients for complex atmospheric models. Atmos
      Environ  15: 1155-1161.  http://dx.doi.orq/10.1016/0004-6981 (81 )90305-X.
Dunlea, EJ: Herndon,  SC: Nelson, DP: Volkamer,  RM: Lamb, BK: Allwine, EJ: Grutter, M: Ramos Villegas, CR:
      Marguez, C:  Blanco, S:  Cardenas, B: Kolb, CE: Molina, LT: Molina, MJ. (2006). Technical note:
      Evaluation of standard ultraviolet absorption ozone monitors in a polluted urban  environment. Atmos
      Chem Phys Discuss 6: 2241-2279.
Duramad.  P: Tager. IB: Holland. NT. (2007). Cytokines and other immunological biomarkers in children's
      environmental health studies. Toxicol Lett 172: 48-59.
Ebeling, D: Patel, V: Findlav, M: Stetter, J. (2009). Electrochemical ozone sensor and instrument with
      characterization of the electrode and gas flow effects. Sens Actuators B 137: 129-133.
      http://dx.doi. orq/10.1016/i.snb.20Q8.10.038.
Eder, B: Yu, S.  (2005). A  performance evaluation of the 2004 release of Models-3 CMAQ. Atmos Environ  40:
      4811-4824. http://dx.doi.0rg/10.1016/i.atmosenv.2005.08.045.
Ederli, L: Morettini, R: Borgogni, A: Wasternack, C: Miersch, O: Reale, L:  Ferranti, F: Tosti, N:  Pasgualini, S.
      (2006). Interaction between nitric oxide and ethylene in the induction of alternative oxidase  in ozone-
      treated tobacco plants. Plant Physiol 142: 595-608. http://dx.doi.orq/10.1104/pp. 106.085472.
Edwards. IP: Zak. DR. (2011).  Fungal community composition and function after long-term exposure of northern
      forests to elevated atmospheric CO2 and tropospheric O3. Global  Change Biol 17: 2184-2195.
      http://dx.doi.0rg/10.1111/i.1365-2486.2010.02376.x.
Eisele. FL: Mount. GH: Tanner. D: Jefferson. A: Shetter. R: Harder. JW: Williams. EJ. (1997). Understanding the
      production and interconversion of the hydroxyl radical during the tropospheric OH photochemistry
      experiment. J Geophys Res 102: 6457-6465.  http://dx.doi.Org/10 1029/96JD02207.
Eiswerth, ME: Shaw, WD: Yen, ST. (2005).  Impacts of ozone on the activities of asthmatics: Revisiting the data.
      J Environ Manage 77: 56-63. http://dx.doi.Org/10.1016/i.ienvman.2005.02.010.
Ellenson. JL: Amundson.  RG. (1982). Delayed light imaging for the early detection of plant stress.  Science 215:
      1104-1106. http://dx.doi.orq/10.1126/science.215.4536.1104.
Ellsworth,  PS: Reich,  PB: Naumburg, ES: Koch, GW: Kubiske,  ME: Smith, SD. (2004).  Photosynthesis,
      carboxylation and  leaf nitrogen responses of 16 species to elevated pCO2 across four free-air CO2
      enrichment experiments in forest,  grassland and desert. Global Change Biol  10: 2121-2138.
      http://dx.doi.0rg/10.1111/i.1365-2486.2004.00867.x.
Eltayeb, AE: Kawano, N:  Badawi, GH: Kaminaka,  H: Sanekata, T: Morishima,  I: Shibahara, T:  Inanaga, S:
      Tanaka,  K. (2006). Enhanced tolerance to ozone and drought stresses in transgenic tobacco
      overexpressing dehydroascorbate reductase in cytosol. Physiol Plant 127: 57-65.
      http://dx.doi.0rg/10.1111/i.1399-3054.2005.00624.x.
Eltaveb. AE: Kawano. N:  Badawi. GH: Kaminaka.  H: Sanekata. T: Shibahara.  T: Inanaga. S: Tanaka. K. (2007).
      Overexpression of monodehydroascorbate reductase in transgenic tobacco confers enhanced tolerance
      to ozone, salt and  polyethylene glycol stresses. Planta 225: 1255-1264.
      http://dx.doi. orq/10.1007/s00425-006-0417-7.
Emberson, L: Ashmore, MR: Cambridge, HM: Simpson, D: Tuovinen, J, -P. (2000a). Modelling stomatal ozone
      flux across Europe. Environ Pollut 109: 403-413. http://dx.doi.org/10.1016/S0269-7491 (00)00043-9.
Emberson, LD: Wieser, G: Ashmore, MR. (2000b). Modelling of stomatal conductance and ozone flux of Norway
      spruce: Comparison with field data. Environ Pollut 109: 393-402. http://dx.doi.org/10.1016/S0269-
      7491(00)00042-7.
Emmerson. KM: Evans. MJ. (2009). Comparison of tropospheric gas-phase chemistry schemes  for use within
      global models.  Atmos Chem Phys 9: 1831-1845. http://dx.doi.org/10.5194/acpd-8-19957-2008.
Emmons. K: Foster. WM. (1991). Smoking cessation and acute airway response to ozone. Arch  Environ Occup
      Health 46: 288-295. http://dx.doi.org/10.1080/00039896.1991.9934389.
Draft - Do Not Cite or Quote                        11-20                                  September 2011

-------
Enami, S: Hoffmann, MR: Colussi, AJ. (2008a). Acidity enhances the formation of a persistent ozonide at
      aqueous ascorbate/ozone gas interfaces.  PNAS 105: 7365-7369.
      http://dx.doi.org/10.1073/pnas.0710791105.
Enami, S: Hoffmann, MR: Colussi, AJ. (2008b). Ozonolysis of uric acid at the air/water interface. J Phys Chem
      B 112: 4153-4156. http://dx.doi.org/10.1021/ip712010k.
Enami. S: Hoffmann. MR: Colussi. AJ. (2009a). How phenol and alpha-tocopherol react with ambient ozone at
      gas/liquid interfaces. J Phys Chem A 113:  7002-7010.  http://dx.doi.org/10.1021/ip901712k.
Enami. S: Hoffmann. MR: Colussi. AJ. (2009b). Ozone oxidizes glutathione to a sulfonic acid. Chem Res Toxicol
      22: 35-40. http://dx.doi.org/10.1021/tx800298i10.1021/tx800298i.
Enami, S: Hoffmann, MR: Colussi, AJ. (2009c). Simultaneous detection of cysteine sulfenate, sulfinate, and
      sulfonate during cysteine interfacial ozonolysis. J Phys Chem B 113: 9356-9358.
      http://dx.doi.org/10.1021/ip904316n.
Enders, G. (1992). Deposition of ozone to a mature spruce forest: Measurements and comparison to models.
      Environ  Pollut 75: 61-67. http://dx.doi.orq/10.1016/0269-7491 (92)90057-H.
Engel. LA. (1985). Intraregional gas mixing and distribution. In Gas  Mixing  and Distribution in the Lung (pp. 287-
      358). New York: Marcel Dekker.
ENVIRON. (ENVIRON Holdings Inc.). (2005). CAMx,  from http://www.camx.com/over/
Ercan. H: Birben. E: Dizdar. EA: Keskin. O: Karaaslan. C: Sover. OU: Put.  R: Sackesen. C:  Besler. T:  Kalavci.
      O.. (2006). Oxidative stress and genetic and epidemiologic determinants of oxidant injury in childhood
      asthma. J Allergy Clin Immunol 118: 1097-1104.  http://dx.doi.Org/10.1016/i.iaci.2006.08.012.
Escalante-Membrillo, C: Gonzalez-Maciel, A: Reynoso-Robles,  R: Gonzalez-Pina, R. (2005). Brain thiobarbituric
      acid-reactive substances in rats after short periods of ozone exposure. Environ Res 99:  68-71.
      http://dx.doi.0rg/10.1016/i.envres.2005.02.006.
Escamilla-Nunez. MC:  Barraza-Villarreal. A: Hernandez-Cadena. L:  Moreno-Macias. H: Ramirez-Aguilar. M:
      Sienra-Monge, JJ: Cortez-Lugo, M: Texcalac, Jl_: del Rio-Navarro, B: Romieu, I. (2008). Traffic-related air
      pollution and respiratory symptoms among asthmatic children, resident in Mexico City: The EVA cohort
      study. Respir Res 9:  74. http://dx.doi.org/10.1186/1465-9921-9-74.
Esperschutz, J: Pritsch, K: Gattinger, A: Welzl, G: Haesler, F: Buegger, F: Winkler, JB: Munch, JC: Schloter, M.
      (2009). Influence of chronic ozone stress on carbon translocation pattern into rhizosphere microbial
      communities of beech trees (Fagus sylvatica L.) during a growing season. Plant Soil  323: 85-95.
      http://dx.doi.org/10.1007/s11104-009-0090-2.
Esther. CR: Peden. DB: Alexis. NE: Hernandez. ML. (2011). Airway  purinergic responses in  healthy, atopic
      nonasthmatic, and atopic asthmatic subjects exposed  to ozone. Inhal Toxicol 23: 324-330.
      http://dx.doi.org/10.3109/08958378.2011.572096.
Eustis.  SL: Schwartz. LW: Kosch. PC: Dungworth. PL. (1981). Chronic bronchiolitis in nonhuman primates after
      prolonged ozone exposure. Am J Pathol 105: 121-137.
Evans,  M: Fiore, A: Jacob, DJ. (2003a). The GEOS-CHEM chemical mechanism: Version 5-07-8. Leeds, UK:
      University of Leeds.
Evans,  MJ: Fanucchi, MV: Baker, GL: Van Winkle, LS: Pantle, LM: Nishio,  SJ: Schelegle, ES: Gershwhin, LJ:
      Miller, LA:  Hyde, DM: Sannes,  PL: Plopper, CG.  (2003b). Atypical development of the tracheal basement
      membrane zone of infant rhesus monkeys exposed to ozone and allergen. Am J Physiol 285: L931-L939.
      http://dx.doi.org/10.1152/aiplung.00175.2003.
Evans,  MJ: Fanucchi, MV: Baker, GL: Van Winkle, LS: Pantle, LM: Nishio,  SJ: Schelegle, ES: Gershwin, LJ:
      Miller. LA:  Hyde. DM: Plopper.  CG. (2004). The remodelled tracheal basement membrane zone of infant
      rhesus monkeys after 6 months of recovery. Clin Exp Allergy  34: 1131-1136.
      http://dx.doi.0rg/10.1111/i.1365-2222.2004.02004.x CEA2004.
Fabbri, LM: Aizawa, H: O'Byrne, PM: Bethel, RA: Walters, EH: Holtzman, MJ: Nadel, JA. (1985). An anti-
      inflammatory drug (BW755C) inhibits airway hyperresponsiveness induced by ozone in dogs. J Allergy
      Clin Immunol 76: 162-166. http://dx.doi.org/10.1016/0091-6749(85)90695-5.
Fakhri.  AA: Ilic. LM: Wellenius. GA: Urch. B: Silverman. F: Gold. PR: Mittleman. MA. (2009). Autonomic effects
      of controlled fine particulate exposure in young healthy adults: Effect modification by  ozone. Environ
      Health Perspect 117: 1287-1292. http://dx.doi.org/10.1289/ehp.0900541.
Fakhrzadeh. L: Laskin. JD: Laskin. PL. (2002). Deficiency in inducible nitric oxide synthase  protects mice from
      ozone-induced lung inflammation and tissue injury. Am J Respir Cell Mol Biol 26: 413-419.
Fakhrzadeh, L: Laskin, JD: Laskin, PL. (2008). Regulation of caveolin-1 expression, nitric oxide production and
      tissue injury by tumor necrosis factor-alpha following ozone inhalation. Toxicol Appl Pharmacol  227: 380-
      389. http://dx.doi.0rg/10.1016/i.taap.2007.11.012.
Fang, Y: Fiore, AM: Horowitz, LW: Levy II, H: Hu, Y: Russell, AG. (2010). Sensitivity of the NOy budget over the
      United States to anthropogenic and lightning NOx in summer. J Geophys Res 115: D18312.
      http://dx.doi. orq/10.1029/201OJD014079.
Fanucchi, MV:  Wong, VJ: Hinds, D: Tarkington, BK: Van Winkle, LS: Evans, MJ: Plopper, CG.  (2000).  Repeated
      episodes of exposure to ozone alters postnatal development  of distal conducting airways in infant rhesus
      monkeys [Abstract]. Am J Respir Grit Care Med 161: A615.
Draft - Do Not Cite or Quote                        11-21                                  September 2011

-------
Fanucchi, MV: Plopper, CG: Evans, MJ: Hyde, DM: Van Winkle, LS: Gershwin, LJ: Schelegle, ES. (2006).
      Cyclic exposure to ozone alters distal airway development in infant rhesus monkeys. Am J Physiol Lung
      Cell Mol Physiol 291: L644-L650. http://dx.doi.org/10.1152/aiplung.00027.2006.
Fares, S: Barta, C: Brilli, F: Centritto, M: Ederli, L: Ferranti, F: Pasgualini, S: Reale, L: Tricoli, D: Loreto, F.
      (2006). Impact of high ozone on isoprene emission, photosynthesis and histology of developing Populus
      alba leaves directly or indirectly exposed to the pollutant. Physiol Plant 128: 456-465.
      http://dx.doi.0rg/10.1111/i.1399-3054.2006.00750.x.
Fares. S: Loreto. F: Kleist. E: Wildt. J. (2008). Stomatal uptake and stomatal deposition of ozone in isoprene
      and monoterpene emitting plants. Plant Biol (Stuttg) 10: 44-54. http://dx.doi.orq/10.1055/S-2007-965257.
Fares, S: McKay, M: Holzinger, R: Goldstein, AH. (201 Oa). Ozone fluxes in a Pinus ponderosa ecosystem are
      dominated by non-stomatal  processes:  Evidence from long-term continuous measurements. Agr Forest
      Meteorol 150: 420-431. http://dx.doi.Org/10.1016/i.agrformet.2010.01.007.
Fares, S: Oksanen, E: Lannenpaa, M: Julkunen-Tiitto, R: Loreto, F. (201 Ob). Volatile emissions and phenolic
      compound concentrations along a vertical profile of Populus nigra leaves exposed to realistic ozone
      concentrations. Photosynth  Res 104: 61-74. http://dx.doi.org/10.1007/s11120-010-9549-5.
Fares, S: Gentner, PR:  Park, JH: Ormeno, E: Karlik, J: Goldstein, AH. (2011). Biogenic emissions from Citrus
      species in California. Atmos Environ 45: 4557-4568. http://dx.doi.Org/10.1016/i.atmosenv.2011.05.066.
Farrai. AK: Bovkin. E: Ledbetter. A: Andrews. D: Gavett. SH.  (2010). Increased lung resistance after diesel
      particulate and ozone co-exposure not associated with enhanced lung inflammation in allergic mice.
      Inhal Toxicol 22:  33-41. http://dx.doi.org/10.3109/08958370902862434.
Favorv, JJ: Stec, A: Gruber,  H: Rizzini, L: Oravecz, A:  Funk, M: Albert, A: Cloix, C: Jenkins, Gl: Oakelev, EJ:
      Seidlitz, HK: Nagy, F: Ulm,  R. (2009). Interaction of COP1 and UVR8 regulates UV-B-induced
      photomorphogenesis and stress acclimation in Arabidopsis.  EMBO J 28: 591-601.
      http://dx.doi. orq/10.1038/emboi.2009.4.
Fehsenfeld, FC: Trainer, M:  Parrish, DP: Volz-Thomas, A: Penkett,  S. (1996). North Atlantic Regional
      Experiment (NARE) 1993 summer  intensive: Foreword. J Geophys Res 101: 28869-28875.
Fehsenfeld. FC: Ancellet. G: Bates. TS: Goldstein. AH: Hardestv. RM: Honrath. R: Law. KS: Lewis. AC: Leaitch.
      R: McKeen, S: Meagher, J: Parrish, DP: Pszenny, AAP: Russell, PB: Schlager, H: Seinfeld, J: Talbot, R:
      Zbinden, R. (2006). International consortium for atmospheric research on transport and transformation
      (ICARTT): North America to Europe: Overview of the 2004 summer field study. J Geophys Res 111:
      D23S01.21-D23S01.36. http://dx.doi.org/10.1029/2006JD007829.
Feichtinger. W: Papalambrou. K: Poehl. M: Krischker.  U: Neumann. K. (1997). Smoking and in vitro fertilization:
      A meta-analysis. J Assist Reprod Genet 14: 596-599.  http://dx.doi.org/10.1023/A: 1022584802711.
Felicity, H: Gina,  M: Laurence, J: Mike, A.  (2010). Does a simulated upland grassland community respond to
      increasing background, peak or accumulated exposure of ozone? Atmos Environ  44: 4155-4164.
      http://dx.doi.0rg/10.1016/i.atmosenv.2010.07.037.
Felzer, B: Kicklighter, D: Melillo, J: Wang,  C: Xhuang,  Q: Prinn, R.  (2004). Effects of ozone on net primary
      production and carbon sequestration in the conterminous United States using a biogeochemistry model.
      Tellus B Chem Phys Meteorol 56: 230-248. http://dx.doi.Org/10.1111/i.1600-0889.2004.00097.x.
Felzer, B: Reilly, J: Melillo, J: Kicklighter, D: Sarofim, M: Wang, C: Prinn, R: Zhuang, Q. (2005). Future effects of
      ozone on carbon sequestration and climate change policy using a global biogeochemical model. Clim
      Change 73: 345-373. http://dx.doi.org/10.1007/s10584-005-6776-4.
Felzer, BS: Cronin, TW: Melillo, JM: Kicklighter, DW: Schlosser, CA. (2009). Importance of carbon-nitrogen
      interactions and ozone on ecosystem hydrology during the 21st century. J Geophys Res 114: G01020.
      http://dx.doi. org/G0102010.1029/2008ig000826.
Feng, R: He, W: Ochi, H: Castranova, V. (2006). Ozone exposure impairs antigen-specific immunity but
      activates IL-7-induced proliferation of CD4-CD8-thymocytes in BALB/c mice. J Toxicol Environ Health A
      69: 1511-1526. http://dx.doi.org/10.1080/15287390500468696.
Feng, YW: Komatsu, S: Furukawa, T: Koshiba, T: Kohno, Y. (2008a). Proteome analysis of proteins responsive
      to ambient and elevated ozone in rice seedlings. Agric Ecosyst Environ 125: 255-265.
      http://dx.doi.0rg/10.1016/i.agee.2008.01.018.
Feng, Z: Pang, J: Nouchi, I:  Kobayashi, K: Yamakawa, T: Zhu, J. (2010). Apoplastic ascorbate contributes to the
      differential ozone sensitivity in two varieties of winter wheat under fully open-air field  conditions. Environ
      Pollut 158: 3539-3545. http://dx.doi.Org/10.1016/i.envpol.2010.08.019.
Feng, ZZ: Kobayashi, K: Ainsworth, EA. (2008b).  Impact of elevated ozone concentration on growth, physiology,
      and yield of wheat (Triticum aestivum L.): A meta-analysis. Global Change Biol 14: 2696-2708.
      http://dx.doi.0rg/10.1111/i.1365-2486.2008.01673.x.
Feng, ZZ: Kobayashi, K. (2009). Assessing the impacts of current and future concentrations of surface ozone on
      crop yield with meta-analysis. Atmos Environ 43: 1510-1519.
      http://dx.doi.0rg/10.1016/i.atmosenv.2008.11.033.
Fenn, ME: Poth,  MA: Johnson, DW. (1996). Evidence for nitrogen saturation in the San Bernardino Mountains in
      southern California. For Ecol Manage 82: 211-230. http://dx.doi.org/10.1016/0378-1127(95)03668-7.
Fenn. ME: de Bauer. LI: Hernandez-Teieda. T. (2002). Summary of air pollution impacts on forests in the Mexico
      City air basin. In  Urban air pollution and forests (pp. 337-355). New York, NY: Springer-Verlag.
Draft - Do Not Cite or Quote                        11-22                                  September 2011

-------
Feo Brito, F: Mur Gimeno, P: Martinez, C: Tobias, A: Suarez, l_: Guerra, F: Borja, JM: Alonso, AM. (2007). Air
      pollution and seasonal asthma during the pollen season: A cohort study in Puertollano and Ciudad Real
      (Spain). Allergy 62: 1152-1157.
Ferdinands, JM: Crawford, CA: Greenwald, R: Van Sickle, D: Hunter, E: league, WG. (2008). Breath
      acidification in adolescent runners exposed to atmospheric pollution: A prospective, repeated measures
      observational study. Environ Health Global Access Sci Source 7: 11. http://dx.doi.org/10.1186/1476-
      069X-7-10.
Ferng. S. -F: Castro. CE: Afifi. AA: Bermudez. E: Mustafa. MG. (1997). Ozone-induced DNAstrand breaks in
      guinea pig tracheobronchial epithelial cells. J Toxicol  Environ Health 51: 353-367.
Findlev,  DA:  Keever, GJ: Chappelka, AH: Eakes, DJ: Gillian, DJ. (1997). Differential responses of buddleia
      (Buddleia davidii  Franch) to ozone. Environ Pollut 98: 105-111.
Finkelstein, PL: Ellestad, TG: Clarke, JF: Meyers, TP: Schwede, DB: Hebert, EO: Neal, JA. (2000). Ozone and
      sulfur dioxide dry deposition to forests: Observations  and model evaluation. J Geophys Res 105: 15365-
      15377.
Finlavson-Pitts. BJ: Pitts. JN. Jr. (1986). Atmospheric chemistry: Fundamentals and experimental techniques.
      In. New York, NY: John Wiley & Sons.
Finnan. JM: Jones.  MB:  Burke. Jl. (1996). A time-concentration study on the effects of ozone on spring wheat
      (Triticum aestivum L): 2. A comparison of indices. Agric  Ecosyst Environ 57: 169-177.
      http://dx.doi. org/10.1016/0167-8809(95)01004-1.
Finnan, JM: Burke, JL: Jones, MB. (1997). An evaluation of indices that describe the impact of ozone on the
      yield of spring wheat (Triticum aestivum L). Atmos Environ 31: 2685-2693.
      http://dx.doi. org/10.1016/51352-2310(97)00105-2.
Fiore. A: Jacob. DJ:  Liu. H: Yantosca. RM: Fairlie. TD:  Li. Q. (2003). Variability in surface ozone background
      over the United States: Implications for air quality  policy. J Geophys Res 108: 4787.
      http://dx.doi. org/10.1029/2003JD003855.
Fiore. A: Dentener. F: Wild. O: Cuvelier. C: Schultz.  M: Hess. P: Textor. C: Schulz.  M: Dohertv. R: Horowitz. L:
      MacKenzie. I: Sanderson. M: Shindell. D: Stevenson. D: Szopa. S: Van Dingenen. R: Zeng. G: Atherton.
      C: Bergmann, D:  Bey, I: Carmichael, G: Collins, W: Duncan, B: Faluvegi, G: Folberth, G:  Gauss, M:
      Gong, S: Hauglustaine, D: Hollowav, T: Isaksen, I: Jacob, D: Jonson, J: Kaminski, J: Keating, T: Lupu, A:
      Marmer, E: Montanaro, V: Park, R: Pitari, G: Pringle,  K: Pyle, J: Schroeder,  S: Vivanco, M: Wind, P:
      Wojcik, G: Wu, S: Zuber, A. (2009). Multimodel estimates of intercontinental source-receptor
      relationships for ozone pollution. J Geophys Res 114: D04301. http://dx.doi.orq/10.1029/2008JD010816.
Fiore. AM: Jacob. DJ: Field. BD: Streets. DG: Fernandes. SD: Jang. C. (2002). Linking ozone pollution and
      climate change: The case for controlling methane. Geophys Res Lett 29: 1919.
      http://dx.doi. orq/10.1029/2002GL015601.
Fiore. AM: West. JJ: Horowitz. LW: Naik. V: Schwartzkopf. MD. (2008). Characterizing the tropospheric ozone
      response to methane emission controls and the benefits to climate and air quality. J Geophys Res 113:
      D08307. http://dx.doi.org/10.1029/2007JD009162.
Fischer,  EV. (2004).  Summertime ozone at Mount Washington: Meteorological controls at the highest peak in
      the northeast. J Geophys Res 109: D24303. http://dx.doi.org/10.1029/2004JD004841.
Fiscus. EL: Philbeck. R: Britt. AM: Booker. FL. (1999). Growth of Arabidopsis flavonoid mutants under solar
      radiation and  UV filters. Environ Exp Bot41: 231-245. http://dx.doi.org/10.1016/50098-8472(99)00011-8.
Fiscus, EL: Booker, FL:  Burkev,  KO. (2005). Crop responses to ozone: Uptake, modes of action, carbon
      assimilation and partitioning. Plant Cell Environ 28: 997-1011.
Fishman. J: Fakhruzzaman. K: Cros. B: Nganga. D. (1991).  Identification of widespread pollution in the
      Southern Hemisphere deduced from satellite analyses.  Science 252:  1693-1696.
      http://dx.doi.org/10.1126/science.252.5013.1693.
Fishman, J: Bowman, KW: Burrows, JP: Richter, A: Chance, KV: Edwards, DP: Martin, RV: Morris, GA: Pierce,
      RB: Ziemke, JR: Al-Saadi, JA: Creilson, JK:  Schaack, TK: Thompson,  AM. (2008).  Remote sensing of
      tropospheric pollution from space. Bull Am Meteorol Soc 89: 805-821.
      http://dx.doi. org/10.1175/2008BAMS2526.1.
Fishman, J: Creilson, JK: Parker, PA: Ainsworth, EA: Vining, GG: Szarka, J:  Booker, FL: Xu, XJ. (2010). An
      investigation of widespread ozone damage to the soybean crop in the upper Midwest determined from
      ground-based and satellite measurements. Atmos Environ 44: 2248-2256.
      http://dx.doi.0rg/10.1016/i.atmosenv.2010.01.015.
FLAG. (Federal Land Manager's Air Quality Related Values Workgroup). (2000).  Phase I report. Lakewood, CO:
      U.S. Forest Service.
Flagler, RB. (1998). Recognition of air pollution injury to vegetation: A pictorial atlas. In (2nd ed.). Pittsburgh, PA:
      Air & Waste Management Association.
Flowers. MD: Fiscus. EL: Burkev. KO: Booker. FL: Dubois. JJB. (2007). Photosynthesis, chlorophyll
      fluorescence, and yield of snap bean (Phaseolus vulgaris L.) genotypes differing in sensitivity to ozone.
      Environ Exp Bot 61: 190-198. http://dx.doi.Org/10.1016/i.envexpbot.2007.05.009.
Folinsbee. LJ: Silverman. F: Shephard. RJ. (1977). Decrease of maximum work performance following ozone
      exposure. J Appl  Physiol 42: 531-536.
Draft - Do Not Cite or Quote                        11-23                                  September 2011

-------
Folinsbee, LJ: Drinkwater, Bl_: Bedi, JF: Horvath, SM. (1978). The influence of exercise on the pulmonary
      function changes due to exposure to low concentrations of ozone. In LJ Folinsbee; JA Wagner; JF
      Borgia; BL Drinkwater; JAGIiner; JF Bedi (Eds.), Environmental stress: individual human adaptations
      (pp. 125-145). New York, NY: Academic Press.
Folinsbee. LJ: Bedi. JF:  Horvath. SM.  (1980). Respiratory responses in humans repeatedly exposed to low
      concentrations of ozone. Am Rev Respir Dis 121: 431-439.
Folinsbee, LJ; Bedi, JF;  Horvath, SM.  (1984). Pulmonary function changes after 1 h continuous heavy exercise
      in 0.21 ppm ozone. J Appl Physiol 57: 984-988.
Folinsbee. LJ: McDonnell. WF: Horstman.  PH.  (1988). Pulmonary function and symptom responses after 6.6-
      hour exposure to 0.12 ppm ozone with moderate exercise. J Air Waste Manag Assoc 38: 28-35.
Folinsbee, LJ; Hazucha, MJ. (1989). Persistence of ozone-induced changes in lung function and airway
      responsiveness. In Atmospheric ozone research and its policy implications (pp. 483-492). Amsterdam,
      The Netherlands: Elsevier.
Folinsbee. LJ: Devlin. RB: Abdul-Salaam. S: Koren. HS. (1993). Repeated severe ozone exposure causes
      depressed baseline spirometry [Abstract]. Am Rev Respir Dis 147: A638.
Folinsbee, LJ; Horstman, DH;  Kehrl, HR; Harder, S; Abdul-Salaam, S; Ives, PJ. (1994). Respiratory responses
      to repeated prolonged exposure to 0.12  ppm ozone. Am J Respir Crit Care Med 149: 98-105.
Folinsbee. LJ: Devlin. RB: Robbins. MK: Biscardi. FH: Abdul-Salaam. S: Koren. HS. (1998). Repeated exposure
      of humans to ozone: Pulmonary function and symptom responses. Research Triangle Park, NC: U.S.
      Environmental Protection Agency.
Folinsbee, LJ; Hazucha, MJ. (2000). Time  course of response to ozone exposure in healthy adult females. Inhal
      Toxicol 12: 151-167.
Fontan. JA: Minga. A: Lopez. A: Druilhet. A. (1992). Vertical ozone profiles in a pine forest. Atmos Environ 26:
      863-869. http://dx.doi.ora/10.1016/0960-1686(92)90245-G.
Forbes, LJ; Patel, MD; Rudnicka, AR;  Cook, DG; Bush, T; Stedman, JR; Whincup, PH; Strachan, DP; Anderson,
      RH. (2009a). Chronic exposure to outdoor air pollution and markers of systemic inflammation.
      Epidemiology 20: 245-253. http://dx.doi.org/10.1097/EDE.Ob013e318190ea3f.
Forbes, LJL; Kapetanakis, V; Rudnicka, AR; Cook, DG; Bush, T; Stedman, JR; Whincup, PH;  Strachan, DP;
      Anderson, HR. (2009b). Chronic exposure to outdoor air pollution and lung function in adults. Thorax 64:
      657-663.
Forster, P; Ramaswamy, V; Artaxo, P; Berntsen, T; Betts, R;  Fahey, DW; Haywood, J; Lean, J; Lowe, DC;
      Mvhre. G: Nganga. J: Prinn. R: Raga. G: Schultz. M: Van Dorland. R.  (2007). Changes in atmospheric
      constituents and in radiative forcing. In S Solomon; D Qin; M Manning; Z Chen; M Marquis; KB Averyt; M
      Tignor; HL Miller (Eds.), Climate Change 2007: The physical science basis: Contribution of Working
      Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 129-
      234). Cambridge, U.K. and New York, NY: Cambridge University Press.
Fortino, V; Maioli, E; Torricelli, C; Davis, P; Valacchi, G. (2007). Cutaneous MMPs are differently modulated by
      environmental stressors in old and young mice. Toxicol Lett 173: 73-79.
      http://dx.doi.0rg/10.1016/i.toxlet.2007.06.004.
Foster, WM; Silver, JA;  Groth, ML. (1993).  Exposure to ozone alters regional function and particle dosimetry in
      the human lung. J Appl Physiol 75:  1938-1945.
Foster. WM: Stetkiewicz. PT (1996). Regional clearance of solute from the respiratory epithelia: 18-20 h
      post exposure to ozone.  J Appl  Physiol 81: 1143-1149.
Foster. WM: Weinmann. GG: Menkes. E: Macri. K. (1997). Acute exposure of humans to ozone impairs small
      airway function. Ann Occup Hyg 1: 659-666.
Foster, WM; Freed, AN. (1999). Regional clearance of solute from peripheral airway epithelia: Recovery after
      sublobar exposure to ozone. J Appl Physiol 86: 641-646.
Foster, WM; Brown, RH; Macri, K; Mitchell, CS. (2000). Bronchial reactivity of healthy subjects: 18-20 h
      postexposure to ozone.  J Appl  Physiol 89: 1804-1810.
Fox. GA. (1991). Practical causal inference for  ecoepidemiologists. J Toxicol Environ Health A 33: 359-373.
      http://dx.doi. org/10.1080/15287399109531535.
Fox, SD; Adams, WC; Brookes, KA; Laslev, BL. (1993). Enhanced response to ozone exposure during the
      follicular phase of the menstrual cycle. Environ Health Perspect 101: 242-244.
Foxcroft. WJ: Adams. WC. (1986). Effects  of ozone exposure on four consecutive days on work performance
      and VO2max. J Appl Physiol 61: 960-966.
Foyer, CH; Noctor, G. (2005a). Oxidant and antioxidant signalling in plants: A re-evaluation  of the concept of
      oxidative stress in a physiological context. Plant Cell Environ 28: 1056-1071.
Foyer, CH; Noctor, G. (2005b). Redox homeostasis and antioxidant signaling: A metabolic interface between
      stress perception and physiological responses. Plant Cell 17: 1866-1875.
      http://dx.doi. org/10.1105/tpc. 105.033589.
Frampton, MW;  Morrow, PE; Torres, A; Voter, KZ; Whitin, JC; Cox, C; Speers, DM; Tsai, Y; Utell, MJ. (1997a).
      Effects of ozone on normal and potentially sensitive human subjects Part II: airway inflammation and
      responsiveness to ozone in nonsmokers and smokers. Boston, MA: Health Effects Institute.
Draft - Do Not Cite or Quote                       11-24                                 September 2011

-------
Frampton, MW:  Morrow, PE: Torres, A: Cox, C: Voter, KZ: Utell, MJ: Gibb, FR: Speers, DM. (1997b). Ozone
      responsiveness in smokers and nonsmokers. Am J Respir Grit Care Med 155: 116-121.
Frampton, MW:  Pryor, WA: Cueto, R: Cox, C: Morrow, PE: Utell, MJ. (1999). Ozone exposure increases
      aldehydes in epithelial lining fluid in human lung. Am J Respir Grit Care Med 159: 1134-1137.
Frank. R: Liu. MC: Spannhake. EW: Mlvnarek. S: Maori. K: Weinmann. GG. (2001). Repetitive ozone exposure
      of young  adults: Evidence of persistent small airway dysfunction. Am J Respir Crit Care Med 164:  1253-
      1260.
Franklin. M: Schwartz. J. (2008). The impact of secondary particles on the association between ambient ozone
      and mortality. Environ Health Perspect 116: 453-458. http://dx.doi.orq/10.1289/ehp. 10777.
Franze, T: Weller,  MG: Niessner, R:  Poschl,  U. (2005). Protein nitration by polluted air.  Environ Sci Technol 39:
      1673-1678.
Fredericksen, TS:  Joyce, BJ: Skellv, JM: Steiner, KG: Kolb, TE: Kouterick, KB: Savage, JE: Snvder, KR. (1995).
      Physiology, morphology, and  ozone uptake of leaves of black cherry seedlings, saplings, and canopy
      trees. Environ Pollut 89: 273-283. http://dx.doi.org/10.1016/0269-7491 (94)00077-Q.
Fredericksen. TS:  Kolb. TE: Skellv. JM: Steiner. KG: Joyce. BJ: Savage. JE. (1996). Light environment alters
      ozone uptake per net photosynthetic  rate in black cherry trees. Tree Physiol 16: 485-490.
Freed. AN: Chou. CL: Fuller. SD: Croxton. TL. (1996). Ozone-induced vagal reflex modulates airways reactivity
      in rabbits. Respir Physiol Neurobiol 105: 95-102.
Freed, AN: Cueto, R: Pryor, WA.  (1999). Antioxidant transport modulates peripheral airway reactivity and
      inflammation during ozone exposure. J Appl Physiol 87:  1595-1603.
Freedman, DM:  Dosemeci, M: McGlynn, K. (2002). Sunlight and mortality from  breast, ovarian, colon, prostate,
      and non-melanoma skin cancer: A composite death certificate based case-control study. Occup Environ
      Med 59: 257-262.
Freiwald. V: Haikio. E: Julkunen-Tiitto. R: Holopainen. JK: Oksanen. E. (2008).  Elevated ozone modifies the
      feeding behaviour of the common leaf weevil on hybrid aspen through shifts in developmental, chemical,
      and structural properties of leaves. Entomol  Exp Appl 128: 66-72. http://dx.doi.orq/10.1111/1.1570-
      7458.2008.00677.x.
Friedman, M: Gallo, JM: Nichols, HP: Bromberg, PA. (1983). Changes in inert gas rebreathing parameters after
      ozone exposure in dogs. Am  Rev Respir Dis 128: 851-856.
Frischer, T: Studnicka,  M:  Gartner, C: Tauber, E: Horak, F: Veiter, A: Spengler, J: Kuhr, J: Urbanek, R. (1999).
      Lung function growth and  ambient ozone: A three-year population study  in school children. Am J Respir
      Crit Care Med 160: 390-396.
Frischer. T: Studnicka.  M:  Halmerbauer. G:  Horak. F: Gartner. C: Tauber. E: Koller. DY. (2001). Ambient ozone
      exposure is associated with eosinophil activation in healthy children. Clin Exp Allergy 31: 1213-1219.
Fritz. JJ: Neale.  PJ: Davis. RF: Peloquin. JA. (2008). Response of Antarctic phytoplankton to solar UVR
      exposure: Inhibition and recovery of photosynthesis in coastal and pelagic assemblages. Mar Ecol Prog
      Ser 365:  1-16. http://dx.doi.org/10.3354/Meps07610.
Frush, S: Li, Z: Potts, EN:  Du, W: Eu, JP: Garantziotis, S: He, YW: Foster, WM: Hollingsworth, JW. (In Press)
      The role of the extracellular matrix protein mindin in airway response to environmental airways injury.
      Environ Health Perspect. http://dx.doi.org/10.1289/ehp. 1003339.
Fuentes. JD: Gillespie. TJ: den Hartog. G: Neumann. HH. (1992). Ozone deposition onto a deciduous forest
      during dry and wet conditions. Agr Forest Meteorol 62: 1-18. http://dx.doi.orq/10.1016/0168-
      1923(92)90002-L.
Fuentes. JD: Wang. D: Bowling. PR: Potosnak. M: Monson. RK: Goliff.  WS: Stockwell. WR. (2007). Biogenic
      hydrocarbon chemistry within and above a mixed deciduous forest. J Atmos Chem 56: 165-185.
      http://dx.doi. org/10.1007/s10874-006-9048-4.
Fuentes, M: Rafterv, AE. (2005).  Model evaluation and spatial interpolation by Bayesian combination of
      observations with outputs from numerical models. Biometrics 61: 36-45.
Fuglestvedt, JS: Berntsen, TK: Isaksen, ISA: Mao, H: Liang, X,  -Z: Wang, W, -C. (1999). Climatic forcing  of
      nitrogen oxides through changes in tropospheric ozone and methane: Global 3D model studies. Atmos
      Environ 33: 961-978. http://dx.doi.orq/10.1016/S1352-2310(98)00217-9.
Fuhrer, J. (1994). Effects of ozone on managed pasture: 1. Effects of open-top  chambers on microclimate,
      ozone flux,  and plant growth.  Environ Pollut 86: 297-305.
Fuhrer. J: Skarbv.  L: Ashmore. MR. (1997). Critical levels for ozone effects on vegetation in Europe. Environ
      Pollut 97: 91-106. http://dx.doi.org/10.1016/30269-7491(97)00067-5.
Fujinaka, LE: Hyde, DM: Plopper, CG: Tyler, WS: Dungworth, PL: Lollini, LO. (1985). Respiratory bronchiolitis
      following  long-term ozone exposure in bonnet monkeys: A morphometric study. Exp Lung Res 8: 167-
      190.
Fuiita. M: Sasavama. S: Ohno. A: Nakaiima. H: Asanoi. H. (1987). Importance of angina for development of
      collateral circulation. Heart 57: 139-143.
Funabashi, H: Shima, M: Kuwaki, T: Hiroshima, K: Kuriyama, T. (2004). Effects of repeated ozone exposure on
      pulmonary function and bronchial responsiveness in mice sensitized with ovalbumin. Toxicology 204: 75-
      83.  http://dx.doi.0rg/10.1016/i.tox.2004.06.047.
Draft - Do Not Cite or Quote                        11-25                                  September 2011

-------
Fung, KY: Luginaah, I: Gorey, KM: Webster, G. (2005). Air pollution and daily hospital admissions for
      cardiovascular diseases in Windsor, Ontario. Can J Public Health 96: 29-33.
Fusco, AC: Logan, JA. (2003). Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
      midlatitudes with the GEOS-CHEM model. J Geophys Res 108: 4449.
      http://dx.doi. orq/10.1029/2002JD002742.
Gackiere. F: Saliba. L: Baude. A: Bosler. O: Strube. C. (2011). Ozone inhalation activates stress-responsive
      regions of the CMS. J Neurochem 117: 961-972. http://dx.doi.org/10.1111/i1471-4159.2011.07267.x.
Gao. KS: Ruan. ZX: Villafane. VE: Gattuso. JP: Helbling. EW. (2009a). Ocean acidification exacerbates the
      effect of UV radiation on the calcifying phytoplankter Emiliania huxleyi. Limnol Oceanogr 54: 1855-1862.
Gao, X:  Raghavamenon, AC: D'Auvergne, O: Uppu, RM. (2009b). Cholesterol secoaldehyde induces apoptosis
      in J774 macrophages via mitochondrial pathway but not involving reactive oxygen species as mediators.
      Biochem Biophys Res Commun 389: 382-387. http://dx.doi.Org/10.1016/i.bbrc.2009.09.005.
Garantziotis, S:  Li, Z:  Potts, EN: Kimata, K: Zhuo, L: Morgan, PL: Savani, RC: Noble, PW: Foster, WM:
      Schwartz. DA:  Hollingsworth. JW. (2009). Hyaluronan mediates ozone-induced airway
      hyperresponsiveness in mice. J Biol Chem 284: 11309-11317. http://dx.doi.org/10.1074/ibe. M802400200.
Garantziotis, S:  Li, Z:  Potts, EN: Lindsev, JY: Stober, VP: Polosukhin, VV: Blackwell, TS: Schwartz, DA: Foster,
      WM: Hollingsworth. JW. (2010). TLR4 is necessary for hyaluronan-mediated airway
      hyperresponsiveness after ozone inhalation. Am J Respir Crit Care Med 181: 666-675.
      http://dx.doi.org/10.1164/rccm.200903-0381OC.
Garcia, TS: Paoletti, DJ: Blaustein, AR. (2009). Correlated trait responses to multiple selection pressures in
      larval amphibians reveal conflict avoidance strategies. Freshw Biol 54: 1066-1077.
      http://dx.doi.0rg/10.1111/i.1365-2427.2008.02154.x.
Garland. FC: Garland. CF: Gorham. ED: Young. JF. (1990). Geographic variation in breast cancer mortality in
      the United States: A hypothesis involving exposure to solar radiation. Prev  Med 19: 614-622.
Garssen, J: Van Loveren, H. (2001). Effects of ultraviolet exposure on the immune system. Crit Rev Immunol
      21: 359-397.
Gate. IM: McNeill.  S: Ashmore. MR. (1995). Effects of air pollution on the searching behaviour of an insect
      parasitoid. Water Air Soil Pollut 85:  1425-1430. http://dx.doi.org/10.1007/BF00477181.
Gauderman, WJ. (2001). Sample size reguirements for matched case-control studies of gene-environment
      interaction.  Stat Med 21: 35-50. http://dx.doi.org/10.1002/sim.973.
Gauderman, WJ. (2002). Sample size reguirements for association studies of gene-gene interaction. Am J
      Epidemiol 155: 478-484.
Gauderman. WJ: Avol. E: Gilliland. F:  Vora. H: Thomas. D: Berhane. K: McConnell. R: Kuenzli. N: Lurmann. F:
      Rappaport,  E: Margolis, H:  Bates, D: Peters, J. (2004). The effect of air pollution  on lung development
      from 10 to 18 years of age.  N Engl J Med 351:  1057-1067.
Gauss. M:  Mvhre.  G: Isaksen. ISA: Grewe. V: Pitari. G: Wild. O: Collins. WJ: Dentener.  FJ: Ellingsen. K: Gohar.
      LK:  Hauglustaine, DA: lachetti, D: Lamargue, JF: Mancini, E:  Mickley, LJ: Prather, MJ: Pyle, JA:
      Sanderson, MG: Shine, KP: Stevenson, PS: Sudo, K: Szopa, S: Zeng, G. (2006). Radiative forcing since
      preindustrial times due to ozone change in the troposphere and the lower stratosphere. Atmos Chem
      Phys 6: 575-599.
Gavdos. TM: Pinder. R: Koo. B:  Fahev. KM: Yarwood. G: Pandis. SN. (2007). Development and application of a
      three-dimensional aerosol chemical transport model, PMCAMx. Atmos Environ 41: 2594-2611.
Gee, GC: Payne-Sturges, DC. (2004). Environmental health disparities: A framework integrating psychosocial
      and environmental concepts. Environ Health Perspect 112: 1645-1653.
      http://dx.doi.org/10.1289/ehp.7074.
Geiser, LH: Neitlich, PN. (2007). Pollution  and climate gradients in western Oregon and Washington indicated
      by epiphytic macrolichens. Environ Pollut 145:  203-218. http://dx.doi.Org/10.1016/i.envpol.2006.03.024.
Generoso, S: Bey, I: Attie, J, -L:  Breon, F, -M. (2007). A satellite- and model-based assessment of the 2003
      Russian fires: Impact on the arctic region. J Geophys Res 112: 5302.
Gent. JF: Triche. EW: Holford. TR: Belanger. K: Bracken. MB: Beckett. WS: Leaderer. BP (2003). Association of
      low-level ozone and fine particles with respiratory symptoms in children with asthma. JAMA 290: 1859-
      1867. http://dx.doi.org/10.1001/iama.290.14.1859.
Georgopoulos.  PG: Purushothaman. V: Chiou. R. (1997). Comparative evaluation of methods for estimating
      potential  human exposure to ozone: Photochemical modeling and ambient  monitoring. J Expo Sci
      Environ Epidemiol 7: 191-215.
Georgopoulos,  PG: Wang, S, -W: Was, VM: Sun, Q: Burke, J: Vedantham, R: McCurdv, T: Ozkaynak, H.
      (2005). A source-to-dose assessment of population exposures to fine PM and ozone in Philadelphia, PA,
      during a summer 1999 episode. J Expo Sci Environ Epidemiol 15: 439-457.
Gerosa. G: Marzuoli. R: Rossini. M: Panigada. C: Meroni. M: Colombo. R: Faoro.  F: Iriti. M. (2009). A flux-based
      assessment of the effects of ozone on foliar injury, photosynthesis, and yield of bean (Phaseolus vulgaris
      L. cv. Borlotto Nano Lingua di Fuoco) in open-top chambers. Environ Pollut 157:  1727-1736.
      http://dx.doi.0rg/10.1016/i.envpol.2008.06.028.
Gerritv. TR: Weaver. RA: Berntsen. J: House.  DE:  O'Neil. JJ. (1988). Extrathoracic and  intrathoracic removal of
      O3 in tidal-breathing humans. J Appl Physiol 65: 393-400.
Draft - Do Not Cite or Quote                       11-26                                 September 2011

-------
Gerrity, TR: McDonnell, WF: House, DE. (1994). The relationship between delivered ozone dose and functional
      responses in humans. Toxicol Appl Pharmacol 124: 275-283.
Gerrity, TR: Biscardi, F: Strong, A: Garlington, AR: Brown, JS: Bromberg, PA. (1995). Bronchoscopic
      determination of ozone uptake  in humans. J Appl Physiol 79: 852-860.
Gershwin. LJ: Osebold. JW: Zee. YC.  (1981). Immunoglobulin E-containing cells in mouse lung following
      allergen inhalation and ozone exposure. International Arch Allergy Appl Immunol 65: 266-277.
Geyh, AS: Wolfson, JM: Koutrakis, P:  Mulik, JD: Avol, EL. (1997). Development and evaluation of a small active
      ozone sampler. Environ Sci Technol 31: 2326-2330.
Gevh. AS: Roberts. PT: Lurmann. FW: Schoell. BM: Avol. EL. (1999). Initial field evaluation of the Harvard
      active ozone sampler for personal ozone monitoring. J Expo Sci Environ Epidemiol 9: 143-149.
Gevh, AS: Xue, J: Ozkaynak, H: Spengler, JD. (2000). The Harvard Southern California chronic ozone exposure
      study: Assessing ozone exposure of grade-school-age children in two southern California communities.
      Environ Health Perspect 108: 265-270.
Giamalva. D: Church. DF: Prvor. WA.  (1985). A comparison of the rates of ozonation of biological antioxidants
      and oleate and linoleate esters. Biochem Biophys Res Commun 133: 773-779.
Gielen, B: Vandermeiren, K: Horemans, N: D'Haese, D:  Serneels, R: Valcke, R. (2006). Chlorophyll a
      fluorescence imaging of ozone-stressed Brassica napus L. plants differing in glucosinolate
      concentrations. Plant Biol (Stuttg) 8: 698-705. http://dx.doi.org/10.1055/S-2006-924150.
Gielen, MH: Van Per Zee, SC: Van Wijnen, JH: Van Steen, CJ: Brunekreef, B. (1997). Acute effects of summer
      air pollution on respiratory health of asthmatic children. Am J Respir Grit Care Med 155: 2105-2108.
Gies, P: Roy, C: Toomey, S: MacLennan, R: Watson, M. (1998). Solar UVR exposures  of primary school
      children at three locations  in Queensland. Photochem Photobiol 68: 78-83.
Gies. P: Wright.  J. (2003). Measured solar ultraviolet radiation exposures of outdoor workers in Queensland in
      the building and construction industry. Photochem Photobiol 78: 342-348.
Gilboa, SM: Mendola, P: Olshan,  AF:  Langlois, PH: Savitz, DA: Loomis, D: Herring, AH: Fixler, DE. (2005).
      Relation between ambient air quality and selected birth defects, seven county study, Texas, 1997-2000.
      Am J Epidemiol 162: 238-252.
Gilliland, AB: Hogrefe, C: Pinder,  RW:  Godowitch, JM: Foley, KL: Rao, ST. (2008). Dynamic evaluation of
      regional air quality models: Assessing changes in O3 stemming from changes in emissions and
      meteorology. Atmos  Environ 42: 5110-5123.
Gilliland, F: Avol, E: Kinnev, P: Jerrett, M: Dvonch, T: Lurmann, F: Buckley, T: Brevsse, P: Keeler, G: de Villiers,
      T: McConnell. R. (2005). Air pollution exposure assessment for epidemiologic studies of pregnant women
      and children: Lessons learned  from the Centers for Children's Environmental Health and Disease
      Prevention Research. Environ  Health Perspect 113: 1447-1454.
Gilliland. FD: McConnell. R: Peters. J: Gong Jr. H. (1999). Atheoretical basis for investigating ambient air
      pollution and children's respiratory health. Environ Health Perspect 107: 403-407.
Gilliland, FD: Berhane, K: Rappaport,  EB: Thomas, DC:  Avol,  E: Gauderman, WJ: London, SJ: Margolis, HG:
      McConnell, R: Islam, KT: Peters, JM. (2001). The effects of ambient air pollution on school absenteeism
      due to respiratory illnesses. Epidemiology 12: 43-54.
Gilliland, FD: Rappaport,  EB: Berhane, K: Islam, T: Dubeau, L: Gauderman, WJ: McConnell, R. (2002). Effects
      of glutathione S-Transferase P1, M1, and T1 on acute respiratory illness in school children. Am J Respir
      Grit Care Med 166: 346-351.
Gilmour, Ml: Jakab, GJ. (1991). Modulation of immune function in mice exposed to 08 ppm ozone. Inhal Toxicol
      3: 293-308.
Giovannelli. L: Pitozzi. V: Moretti. S: Boddi. V: Dolara. P. (2006). Seasonal variations of DMA damage in human
      lymphocytes: Correlation with different environmental variables. Mutat Res-Fundam  Mol Mech Mutagen
      593: 143-152. http://dx.doi.Org/10.1016/i.mrfmmm.2005.07.002.
Giovannucci, E.  (2005). The epidemiology of vitamin D and cancer incidence and  mortality: A review (United
      States) [Review]. Cancer Causes Control 16: 83-95.
Girardot. SP: Ryan. PB: Smith. SM: Davis. WT: Hamilton. CB: Obenour. RA: Renfro. JR:  Tromatore. KA: Reed.
      GD. (2006). Ozone and PM2.5 exposure and acute pulmonary health effects: A study of hikers in the
      Great Smoky Mountains National Park. Environ Health Perspect 113: 612-617.
      http://dx.doi.org/10.1289/ehp.8637.
Gitav. H: Brown. S: Easterling. W: Jallow. B. (2001). Ecosystems and their goods and services. In Climate
      change 2001: Impacts, adaptation and vulnerability: Contribution of Working Group II to the third
      assessment report of the Intergovernmental Panel on Climate Change (pp. 237-342). Cambridge, United
      Kingdom: Cambridge University Press.
Glen, G: Smith, L: Isaacs, K: Mccurdy, T: Langstaff, J. (2008). A new method of longitudinal diary assembly for
      human exposure modeling. J Expo Sci Environ Epidemiol 18: 299-311.
      http://dx.doi.org/10.1038/si.ies.7500595.
Gloster, HM, Jr:  Brodland, DG. (1996). The epidemiology of skin cancer. Dermatol Surg 22: 217-226.
Godar. DE: Wengraitis. SP:  Shreffler.  J: Slinev. PH. (2001). UV doses of Americans. Photochem  Photobiol 73:
      621-629.
Draft - Do Not Cite or Quote                       11-27                                 September 2011

-------
Godowitch, JM: Gilliland, AB: Draxler, RR: Rao, ST. (2008). Modeling assessment of point source NOx emission
      reductions on ozone air quality in the eastern United States. Atmos Environ 42: 87-100.
Gold, PR:  Damokosh, Al: III, PC: Dockery, DW: McDonnell, WF: Serrano, P: Retama, A: Castillejos, M. (1999).
      Particulate and ozone pollutant effects on the respiratory function of children  in southwest Mexico City.
      Epidemiology 10: 8-16.
Goldberg.  MS: Burnett. RT: Yale. JF: Valois. MF: Brook. JR. (2006). Associations between ambient air pollution
      and daily mortality among persons with diabetes and cardiovascular disease. Environ Res 100: 255-267.
Goldberg.  MS: Giannetti. N: Burnett. RT: Mayo. NE: Valois. MF: Brophv. JM. (2008). A panel study in congestive
      heart failure to estimate the short-term effects from personal factors and environmental conditions on
      oxygen saturation and pulse rate. Occup Environ Med 65: 659-666.
      http://dx.doi.org/10.1136/oem.2007.034934.
Goldman, GT: Mulholland, JA: Russell, AG: Strickland, MJ: Klein, M: Waller, LA: Tolbert, PE. (2011). Impact of
      exposure measurement error in air pollution epidemiology: Effect of error type in time-series studies.
      Environ Health Global Access Sci Source 10: 61. http://dx.doi.org/10.1186/1476-069X-10-61.
Goldstein.  A: Galballv. I. (2007).  Known and unexplored organic constituents in the earth's atmosphere. Environ
      Sci  Technol 41: 1514-1521.  http://dx.doi.org/10.1021/es072476p.
Goldstein.  AH: Millet. DB: McKay. M: Jaegle. L: Horowitz. L: Cooper. O: Hudman. R: Jacob.  DJ: Oltmans. S:
      Clarke. A. (2004). Impact of Asian emissions on observations at Trinidad Head, California, during ITCT
      2K2. J  Geophys Res 109: D23S17. http://dx.doi.org/10.1029/2003JD004406.
Gombert, S: Asta, J: Seaward, MRP. (2006). Lichens and tobacco plants as complementary biomonitors of air
      pollution in the Grenoble  area (Isere, southeast France). Ecol Indicat 6: 429-443.
      http://dx.doi.0rg/10.1016/i.ecolind.2005.06.001.
Gong. H. Jr: Bradley. PW: Simmons. MS: Tashkin. DP. (1986). Impaired exercise performance and pulmonary
      function in elite cyclists during low-level ozone exposure in a hot environment. Am J Respir Crit Care
      Med 134: 726-733.
Gong. H. Jr: McManus. MS: Linn. WS. (1997a). Attenuated response to repeated daily ozone exposures in
      asthmatic subjects. Arch  Environ Occup Health 52: 34-41.
Gong, H, Jr: Shamoo, DA: Anderson, KR: Linn, WS. (1997b). Responses of older men with  and without chronic
      obstructive pulmonary disease to prolonged ozone exposure. Arch Environ Occup Health 52: 18-25.
Gong, H, Jr: Wong, R: Sarma, RJ: Linn, WS: Sullivan, ED: Shamoo, DA: Anderson,  KR: Prasad, SB. (1998).
      Cardiovascular effects of ozone exposure in human volunteers. Am J Respir Crit Care Med 158: 538-
      546.
Gonzalez-Fernandez. I:  Kaminska.  A: Dodmani. M: Goumenaki. E:  Quarrie. S: Barnes.  JD. (2010). Establishing
      ozone flux-response relationships for winter wheat: Analysis of uncertainties based on data for UK and
      Polish genotypes. Atmos  Environ 44: 621-630. http://dx.doi.Org/10.1016/i.atmosenv.2009.11.021.
Gonzalez-Pina. R: Escalante-Membrillo. C: Alfaro-Rodriguez. A: Gonzalez-Maciel. A. (2008). Prenatal exposure
      to ozone disrupts cerebellar  monoamine contents in newborn rats. Neurochem Res 33: 912-918.
      http://dx.doi. org/10.1007/s11064-007-9534-3.
Gorham, ED:  Garland, FC: Garland, CF. (1990). Sunlight and breast cancer incidence in the USSR. Int J
      Epidemiol 19: 820-824. http://dx.doi.Org/10.1093/iie/19.4.820.
Gottardini. E:  Cristofori. A: Cristofolini. F: Ferretti. M. (2010). Variability of ozone concentration in a montane
      environment, northern Italy. Atmos Environ  44: 147-152.
      http://dx.doi. org/10.1016/i.atmosenv.20Q9.10.017.
Goumenaki. E: Tavbi. T: Borland. A: Barnes. J. (2010). Mechanisms underlying the impacts  of ozone on
      photosynthetic performance. Environ Exp Bot 69: 259-266.
      http://dx.doi.0rg/10.1016/i.envexpbot.2010.04.011.
Gouveia, N: Bremner, SA: Novaes,  HMD. (2004). Association between ambient air pollution and birth weight in
      Sao Paulo, Brazil. J Epidemiol Community  Health 58: 11-17.
Graham, DE:  Koren, HS. (1990). Biomarkers of inflammation in ozone-exposed humans: Comparison of the
      nasal and bronchoalveolar lavage. Am J Respir Crit Care Med 142: 152-156.
Graham. JA: Menzel. DB: Miller. FJ: Illing. JW: Gardner. DE. (1981). Influence of ozone on  pentobarbital-
      induced sleeping time in mice, rats, and hamsters. Toxicol Appl Pharmacol 61: 64-73.
      http://dx.doi. org/10.1016/0041-008X(81)90008-9.
Granstein. RD: Matsui. MS. (2004). UV radiation-induced immunosuppression and skin cancer. Cutis 5: 4-9.
Grant, WB. (2002a). An ecologic study of dietary and solar ultraviolet-B links to breast carcinoma mortality  rates.
      Cancer 94: 272-281.
Grant, WB. (2002b). An estimate of premature cancer mortality in the US due to inadequate doses of solar
      ultraviolet-B radiation. Cancer 94: 1867-1875.
Grant. WB: Garland. CF. (2004). A critical review of studies on vitamin D in relation to colorectal cancer
      [Review]. Nutr Cancer 48: 115-123.
Grantz, DA: Zhang, XJ: Massman, WJ: Den Hartog, G: Neumann, HH: Pederson, JR. (1995). Effects of
      stomatal conductance and surface wetness on ozone deposition in field-grown grape. Atmos Environ 29:
      3189-3198. http://dx.doi.org/10.1016/1352-2310(95)00129-M.
Draft - Do Not Cite or Quote                       11-28                                 September 2011

-------
Grantz, DA: Zhang, XJ:  Massman, W: Delany, A: Pederson, R. (1997). Ozone deposition to a cotton
      (Gossypium hirsutum L) field: Stomatal and surface wetness effects during the California Ozone
      Deposition experiment. Agr Forest Meteorol 85: 19-31. http://dx.doi.org/10.1016/30168-1923(96)02396-
      9.
Grantz. DA: Gunn. S: Vu. HB. (2006). O3 impacts on plant development: A meta-analysis of root/shoot
      allocation and growth. Plant Cell  Environ 29: 1193-1209. http://dx.doi.orq/10.1111/1.1365-
      3040.2006.01521.x.
Grantz. DA: Shrestha. A. (2006). Tropospheric ozone and interspecific competition between yellow nutsedge
      and pima cotton.  Crop Sci 46: 1879-1889. http://dx.doi.ora/10.2135/cropsci2005.06.0167.
Grantz, DA: Shrestha, A: Vu, HB. (2008). Early vigor and ozone response in horseweed (Conyza canadensis)
      biotypes differing in glyphosate resistance. Weed Sci 56: 224-230. http://dx.doi.org/10.1614/ws-07-130.1.
Grantz, DA: Vu, HB. (2009). O3 sensitivity in a potential C4 bioenergy crop: Sugarcane in California. Crop Sci
      49: 643-650.
Grantz. DA: Vu. HB: Aguilar. C: Rea. MA. (201 Oa). No interaction between methyl jasmonate and ozone in Pima
      cotton: Growth and allocation respond independently to both. Plant Cell Environ 33: 717-728.
      http://dx.doi.0rg/10.1111/i.1365-3040.2009.02096.x.
Grantz. DA: Shrestha. A: Vu. HB. (201 Ob). Ozone impacts on assimilation and allocation to  reproductive sinks in
      the vegetatively propagated C-4 weed, yellow nutsedge. Crop Sci 50: 246-252.
      http://dx.doi.org/10.2135/cropsci2009.03.0127.
Grebenc, T: Kraigher, H. (2007). Changes in the community of ectomycorrhizal fungi and increased fine root
      number under adult beech trees chronically fumigated with double ambient ozone concentration. Plant
      Biol (Stuttg) 9: 279-287. http://dx.doi.org/10.1055/s-2006-924489.
Greenberg. JP: Guenther. AB: Turnipseed. A. (2009). Tethered balloon-based soundings of ozone, aerosols,
      and solar radiation  near Mexico City during MIRAGE-MEX. Atmos Environ 43: 2672-2677.
      http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.019.
Gregg. JW: Jones. CG:  Dawson. TE. (2003). Urbanization effects on tree growth in the vicinity of New York City
      [Letter/Response]. Nature 424: 183-187. http://dx.doi.org/10.1038/nature01728.
Gregg, JW: Jones, CG:  Dawson, TE. (2006). Physiological and developmental effects of O3 on cottonwood
      growth in urban and rural sites. Ecol Appl 16:  2368-2381. http://dx.doi.org/10.1890/1051-
      0761(2006)016f2368:PADEOO12.0.CO:2.
Grell, GA: Emeis, S: Stockwell, WR: Schoenemeyer, T:  Forkel, R: Michalakes, J: Knoche, R: Seidl, W. (2000).
      Application of a multiscale, coupled MM5/chemistry model to the complex terrain of the VOTALP valley
      campaign. Atmos Environ 34: 1435-1453.
Grennfelt, P. (2004). New directions: Recent research findings may change ozone control policies. Atmos
      Environ 38: 2215-2216.
Groppa. MD: Benavides. MP (2008). Polyamines and abiotic stress: Recent advances. Amino Acids 34: 35-45.
      http://dx.doi.org/10.1007/s00726-007-0501-8.
Grosjean, D: Hisham, MWM. (1992). A passive sampler for atmospheric ozone. J Air Waste Manag Assoc 42:
      169-173.
Gross,  EA: Swenberg, JA: Fields, S: Popp, JA. (1982). Comparative morphometry of the nasal  cavity in rats and
      mice. J Anat 135: 83-88.
Gross.  EA: Starr. TB:  Randall. HW: Morgan. KT (1987). Morphometric analysis of the primate nasal cavity
      [Abstract]. Toxicologist 7: 193.
Grulke. N: Neufeld. H: Davison. A: Roberts. M: Chappelka. A. (2007a). Stomatal behavior of ozone-sensitive
      and -insensitive coneflowers (Rudbeckia laciniata van digitata) in Great Smoky Mountains National Park.
      New Phytol 173:  100-109. http://dx.doi.Org/10.1111/i.1469-8137.2006.01872.x.
Grulke, NE: Lee, EH.  (1997). Assessing visible ozone-induced injury in  ponderosa pine. Can J For Res 27:
      1658-1668.
Grulke, NE. (1999). Physiological responses of ponderosa pine to gradients of environmental stressors. In PR
      Miller; JR McBride (Eds.), Oxidant air pollution impacts in the montane forests of Southern California.
      New York, NY: Springer-Verlag.
Grulke, NE: Preisler, HK: Rose, C: Kirsch, J: Balduman, L. (2002).  O3 uptake and drought stress effects on
      carbon acquisition of ponderosa pine in natural stands. New Phytol 154: 621-631.
      http://dx.doi. org/10.1046/i. 1469-8137.2002.00403.x.
Grulke, NE: Johnson, R: Esperanza, A: Jones, D: Nguyen, T: Posch, S: Tausz, M. (2003a).  Canopy
      transpiration of Jeffrey pine in mesic and xeric microsites: O3 uptake and injury response. Trees Struct
      Funct 17: 292-298.
Grulke, NE: Johnson, R: Monschein, S:  Nikolova, P: Tausz, M. (2003b). Variation in morphological and
      biochemical O3 injury attributes of mature Jeffrey pine within canopies and between  microsites. Tree
      Physiol 23: 923-929.
Grulke, NE: Alonso, R: Nguyen,  T: Cascio, C: Dobrowolski, W. (2004). Stomata open at night in pole-sized and
      mature ponderosa pine: Implications for O3 exposure metrics. Tree Physiol 24: 1001-1010.
Draft - Do Not Cite or Quote                        11-29                                  September 2011

-------
Grulke, NE: Dobrowolski, W:  Mingus, P: Fenn,  ME. (2005). California black oak response to nitrogen
      amendment at a high O3, nitrogen-saturated site. Environ Pollut 137: 536-545.
      http://dx.doi.0rg/10.1016/i.envpol.2005.01.039.
Grulke, NE: Paoletti, E: Heath, RL. (2007b). Chronic vs. short-term acute O3 exposure effects on nocturnal
      transpiration in two Californian oaks. ScientificWorldJournal 7: 134-140.
      http://dx.doi.ora/10.1100/tsw.2007.33.
Grulke, NE: Paoletti, E: Heath, RL. (2007c). Comparison of calculated and measured foliar O3 flux in crop and
      forest species. Environ Pollut 146: 640-647. http://dx.doi.Org/10.1016/i.envpol.2006.04.014.
Grunewald. J: Eklund. A. (2007). Role of CD4+ T cells in sarcoidosis. Proc Am Thorac Soc 4: 461-464.
Grunhage, l_: Haenel, H, -D. (1997). PLATIN (PLant-ATmosphere-INteraction)  I: A model of plant-atmosphere
      interaction for estimating absorbed doses of gaseous air pollutants. Environ Pollut 98: 37-50.
      http://dx.doi.org/10.1016/S0269-7491(97)00114-0.
Grunhage, l_: Jager, H, -J. (2003). From critical levels to critical loads for ozone: a discussion of a new
      experimental and modelling approach for establishing flux-response relationships for agricultural crops
      and native plant species. Environ Pollut  125: 99-110.
Grunhage, l_: Krupa, SV: Legge, AH: Jager, H, -J. (2004). Ambient flux-based critical values of ozone for
      protecting vegetation:  Differing spatial scales and uncertainties in risk assessment. Atmos Environ 38:
      2433-2437. http://dx.doi.Org/10.1016/i.atmosenv.2003.12.039.
Gryparis, A: Forsberg, B: Katsouyanni,  K: Analitis, A: Touloumi, G: Schwartz, J: Samoli, E: Medina, S:  Anderson,
      HR: Niciu, EM: Wichmann, HE: Khz, B: Kosnik, M: Skorkovskv, J: Vonk, JM: Dortbudak, Z. (2004). Acute
      effects of ozone on mortality from the "Air pollution and health: A European approach" project. Am J
      Respir Grit Care Med 170:  1080-1087. http://dx.doi.org/10.1164/rccm.200403-333OC.
Guav. M: Stieb. DM: Smith-Doiron. M.  (2011). Assessment of long-term exposure to air pollution in a longitudinal
      national health survey. J Expo Sci Environ Epidemiol 21: 337-342. http://dx.doi.org/10.1038/ies.2010.37.
Guderian, R. (1985). Air pollution by photochemical oxidants: Formation, transport, control, and effects on
      plants. In. New York: Springer-Verlag.
Guenther. A: Geron. C: Pierce. T:  Lamb. B:  Harlev. P: Fall. R. (2000). Natural emissions of non-methane volatile
      organic compounds, carbon monoxide, and oxides of nitrogen from North America. Atmos Environ 34:
      2205-2230. http://dx.doi.org/10.1016/S1352-2310(99)00465-3.
Guenther, A: Karl, T: Harlev, P: Wiedinmyer, C:  Palmer, PI: Geron, C. (2006). Estimates of global terrestrial
      isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature). Atmos
      Chem Phys 6:  3181-3210.  http://dx.doi.org/10.5194/acp-6-3181-2006.
Guerrero. AL: Dorado-Martinez. C: Rodriguez. A: Pedroza-Rios. K: Borgonio-Perez. G: Rivas-Arancibia. S.
      (1999). Effects of vitamin E on ozone-induced memory deficits and lipid peroxidation in rats. Neuroreport
      10: 1689-1692.
Guevara-Guzman. R: Arriaga. V: Kendrick.  KM: Bernal. C: Vega. X: Mercado-Gomez. OF: Rivas-Arancibia. S.
      (2009). Estradiol prevents ozone-induced increases in brain lipid peroxidation and impaired social
      recognition memory in female rats. Neuroscience 159: 940-950.
      http://dx.doi.0rg/10.1016/i.neuroscience.2009.01.047.
Guidi, L:  Degl'lnnocenti, E. (2008). Ozone effects on high light-induced photoinhibition in Phaseolus vulgaris.
      Plant Sci 174: 590-596. http://dx.doi.Org/10.1016/i.plantsci.2008.03.003.
Guidi. L:  Degl'lnnocenti. E:  Martinelli. F: Piras. M. (2009). Ozone effects on carbon metabolism in sensitive and
      insensitive Phaseolus  cultivars. Environ  Exp Bot 66: 117-125.
      http://dx.doi.0rg/10.1016/i.envexpbot.2008.12.005.
GDI. H: Gaga. EO: Dogeroglu. T: Ozden. O: Avvaz. O: Ozel.  S: Gungor. G. (2011). Respiratory health  symptoms
      among students exposed to different levels of air pollution in a Turkish city. Int J Environ Res Public
      Health 8: 1110-1125. http://dx.doi.org/10.3390/iierph8041110.
Gumpertz, ML: Rawlings, JO. (1992). Nonlinear regression with variance components: Modeling effects of
      ozone on crop yield. Crop Sci 32: 219-224.
Gunderson. CA: Sholtis. JD: Wullschleger. SD:  Tissue.  DT: Hanson. PJ: Norbv. RJ. (2002). Environmental and
      stomatal control of photosynthetic enhancement in the canopy of a sweetgum (Liquidambar styraciflua L)
      plantation during 3 years of CO2 enrichment.  Plant Cell Environ 25: 379-393.
      http://dx.doi.0rg/10.1046/i.0016-8025.2001.00816.x.
Gunnison. AF: Hatch. GE: Crissman. K: Bowers. A. (1996). Comparative sensitivity of lactating and virgin
      female rats to ozone-induced pulmonary inflammation. Inhal Toxicol 8: 607-623.
Gunnison, AF: Hatch, GE. (1999). OS-induced inflammation in prepregnant,  pregnant, and lactating rats
      correlates with O3 dose estimated by 18O. Am J  Physiol 276: L332-L340.
Ha, E, -H: Hong, Y, -C: Lee, B, -E: Woo, B, -H: Schwartz, J: Christiani, DC. (2001). Is air pollution a risk factor
      for low birth weight in Seoul? Epidemiology 12: 643-648.
Ha. E. -H: Lee. J. -T: Kim. H:  Hong. Y.  -C: Lee.  (2003). Infant susceptibility of mortality to air pollution in Seoul,
      South Korea. Pediatrics 111: 284-290.
Haapala. JK: Morskv.  SK: Saarnio. S: Rinnan. R: Suokanerva. H: Kvr.  E: Latola.  K: Martikanen. PJ: Holopainen.
      T: Silvola. J. (2009). Carbon dioxide  balance of a fen ecosystem in northern Finland under elevated UV-B
      radiation. Global Change Biol 15: 943-954. http://dx.doi.Org/10.1111/i.1365-2486.2008.01785.x.
Draft - Do Not Cite or Quote                        11-30                                  September 2011

-------
Haberer, K: Herbinger, K: Alexou, M: Rennenberg, H: Tausz,  M. (2008). Effects of drought and canopy ozone
      exposure on antioxidants in fine roots of mature European beech (Fagus sylvatica). Tree Physiol 28:
      713-719. http://dx.doi.0rg/10.1093/treephys/28.5.713.
Hack, M: Fanaroff, AA. (1999). Outcomes of children of extremely low birth weight and gestational age in the
      1990s. Early Hum Dev 53: 193-218. http://dx.doi.org/10.1016/30378-3782(98)00052-8.
Hackney. JD: Linn. WS: Mohler. JG: Pedersen. EE: Breisacher. P: Russo. A. (1975). Experimental studies on
      human health effects of air pollutants: II. Four-hour exposure to ozone alone and in combination with
      other pollutant gases. Arch Environ Occup Health 30: 379-384.
Hackney. JD: Linn. WS: Mohler. JG: Collier. CR. (1977). Adaptation to short-term respiratory effects of ozone in
      men exposed repeatedly. J Appl Physiol 43: 82-85.
Hader, DP: Kumar, HP: Smith, RC: Worrest, RC. (2007). Effects of solar UV radiation on aquatic ecosystems
      and interactions with climate change. Photochem Photobiol Sci 6: 267-285.
      http://dx.doi. org/10.1039/B700020k.
Hains. JC: Taubman.  BF: Thompson. AM: Stehr. JW: Marufu. LT: Doddridge. BG: Dickerson. RR. (2008).
      Origins of chemical pollution derived from Mid-Atlantic aircraft profiles using a clustering technique.
      Atmos Environ 42: 1727-1741. http://dx.doi.org/10.1016/i.atmosenv.20Q7.11.052.
Haiat. S: Armstrong. B: Wilkinson. P: Busby. A: Polk. H. (2007). Outdoor air pollution and infant mortality:
      Analysis of daily time-series data in 10 English cities. J Epidemiol Community Health 61: 719-722.
      http://dx.doi.org/10.1136/iech.2006.053942.
Halonen, Jl: Lanki, T: Tiittanen, P: Niemi, JV: Loh,  M: J, P. (2009). Ozone and cause-specific cardiorespiratory
      morbidity and mortality.  J Epidemiol Community Health 64: 814-820.
      http://dx.doi. org/10.1136/jech.2009.087106.
Hamade. AK: Rabold. R: Tankerslev. CG. (2008). Adverse cardiovascular effects with acute particulate matter
      and ozone exposures: Interstrain variation in mice. Environ Health Perspect 116: 1033-1039.
      http://dx.doi.org/10.1289/ehp.10689.
Hamade. AK: Tankerslev. CG. (2009). Interstrain variation in cardiac and respiratory adaptation to repeated
      ozone and particulate matter exposures. Am J Physiol Regul Integr Comp Physiol 296:  R1202-R1215.
      http://dx.doi.org/10.1152/aipregu.90808.2008.
Hamade, AK: Misra, V: Rabold, R: Tankerslev, CG. (2010). Age-related changes in cardiac and respiratory
      adaptation to acute ozone and carbon black exposures: Interstrain variation in mice. Inhal Toxicol 22: 84-
      94.  http://dx.doi.org/10.3109/08958378.2010.503974.
Hameed. S:  Pinto. JP: Stewart. RW. (1979). Sensitivity of the predicted CO-OH-CH4 perturbation to
      tropospheric NOx concentrations. J Geophys Res 84:  763-768.
Hamel, LP: Miles, GP: Samuel, MA: Ellis, BE: Seguin, A: Beaudoin, N. (2005). Activation of stress-responsive
      mitogen-activated protein kinase pathways in hybrid poplar (Populus trichocarpa x Populus deltoides).
      Tree Physiol 25: 277-288. http://dx.doi.org/10.1093/treephvs/25.3.277.
Hamilton, JG: Dermody, O: Aldea, M: Zangerl, AR: Rogers, A: Berenbaum, MR:  DeLucia, EH. (2005).
      Anthropogenic changes in tropospheric composition increase susceptibility of soybean to insect
      herbivory. Environ Entomol 34: 479-485.
Hamilton, RF: Li, l_: Eschenbacher, Wl_: Szweda, l_: Holian, A. (1998). Potential  involvement of 4-
      hydroxynonenal in the response of human lung cells to ozone. Am J  Physiol 274: L8-L16.
Han. SG: Andrews. R: Gairola. CG: Bhalla. DK. (2008). Acute pulmonary effects of combined exposure to
      carbon nanotubes and ozone in mice. Inhal Toxicol 20: 391-398.
      http://dx.doi. org/10.1080/08958370801904014.
Han. SG: Bhoopalan. V: Akinbivi. T: Gairola. CG: Bhalla. DK. (2011). In utero tobacco smoke  exposure alters
      pulmonary responses of newborn rats to ozone. J Toxicol Environ Health A 74: 668-677.
      http://dx.doi. org/10.1080/15287394.2011.539133.
Hanchette, Cl_: Schwartz,  GG. (1992). Geographic patterns of prostate cancer mortality. Cancer 70: 2861-2869.
Handley, T: Grulke, NE. (2008). Interactive effects of O3 exposure on California black oak (Quercus kelloggii
      Newb.) seedlings with and without N amendment. Environ Pollut 156: 53-60.
      http://dx.doi.0rg/10.1016/i.envpol.2008.01.002.
Hanene, C: Jihene, l_: Jame, A: Kamel, H: Agnes, H. (2007). Association of GST genes polymorphisms with
      asthma in Tunisian children. Mediators Inflamm 19564: 6. http://dx.doi.org/10.1155/2007/19564.
Hansen. C: Neller. A: Williams. G: Simpson. R.  (2006). Maternal exposure to low levels of ambient air pollution
      and preterm birth in Brisbane, Australia.  BJOG 113:  935-941.
Hansen, C: Neller, A: Williams, G: Simpson, R.  (2007). Low levels of ambient air pollution during  pregnancy  and
      fetal growth among term neonates in Brisbane, Australia.  Environ Res 103: 383-389.
Hansen, C: Luben, TJ: Sacks,  JD: Olshan, A: Jeffav, S: Strader, l_: Perreault, SD. (2010). The effect of ambient
      air pollution on sperm quality. Environ Health Perspect 118: 203-209.
      http://dx.doi.org/10.1289/ehp.0901022.
Hansen, CA: Barnett, AG:  Pritchard, G. (2008). The effect of ambient air pollution during early  pregnancy on
      fetal ultrasonic measurements during mid-pregnancy. Environ Health Perspect 116: 362-369.
Hansen. CA: Barnett. AG:  Jalaludin. B: Morgan. G. (2009). Ambient air pollution and birth defects in Brisbane,
      Australia. PLoS ONE 4: e5408. http://dx.doi.org/10.1371/iournal.pone.0005408.
Draft - Do Not Cite or Quote                        11-31                                  September 2011

-------
Hansen, J: Sato, M: Ruedy, R: Nazarenko, L: Lacis, A: Schmidt, GA: Russell, G: Aleinov, I: Bauer, M: Bauer, S:
      Bell,  N: Cairns, B: Canuto, V: Chandler,  M: Cheng, Y: Del Genio, A: Faluvegi, G: Fleming, E: Friend, A:
      Hall,  T: Jackman, C: Kelley, M: Kiang, N: Koch, D: Lean,  J:  Lerner, J:  Lo, K: Menon, S: Miller, R: Minnis,
      P: Novakov, T: Oinas, V: Perlwitz, J: Rind, D: Romanou, A:  Shindell, D: Stone, P: Sun, S: Tausnev, N:
      Thresher. D: Wielicki. B: Wong. T: Yao. M: Zhang. S. (2005). Efficacy of climate forcings. J Geophys Res
      110:  D18104.  http://dx.doi.orq/10.1029/2005JD005776.
Hansen, JE: Sato, M: Ruedy, R. (1997). Radiative forcing and climate response. J Geophys Res 102: 6831-
      6864. http://dx.doi.orq/10.1029/96JD03436.
Hanson. PJ: Wullschleger. SD: Norbv. RJ:  Tschaplinski. TJ: Gunderson. CA. (2005). Importance of changing
      CO2, temperature, precipitation, and ozone on carbon and water cycles of an upland-oak forest:
      incorporating experimental results into model simulations. Global Change Biol 11: 1402-1423.
      http://dx.doi.0rg/10.1111/i.1365-2486.2005.00991.x.
Hague, R: Umstead,  TM: Ponnuru, P: Quo, X: Hawgood, S: Phelps, PS: Floros, J. (2007). Role of surfactant
      protein-A(SP-A) in lung injury in response to acute ozone exposure of SP-A deficient mice. Toxicol Appl
      Pharmacol 220: 72-82. http://dx.doi.org/10.1016/i.taap.2006.12.017.
Hague, R: Umstead,  TM: Freeman, WM: Floros, J: Phelps, PS.  (2009). The impact of surfactant protein-Aon
      ozone-induced changes in the mouse bronchoalveolar lavage proteome. Proteome Science 7: 12.
      http://dx.doi.org/10.1186/1477-5956-7-12.
Harkema, JR:  Plopper, CG: Hyde, DM: StGeorge, JA: Dungworth,  PL (1987a). Effects of an ambient level of
      ozone on primate nasal epithelial mucosubstances: quantitative histochemistry. Am J  Pathol 127: 90-96.
Harkema, JR:  Plopper, CG: Hyde, DM: StGeorge, JA: Wilson, DW: Dungworth, PL. (1987b). Response of the
      macaque nasal epithelium to ambient levels of ozone: A morphologic and morphometric study of the
      transitional and respiratory epithelium. Am J Pathol 128: 29-44.
Harkema. JR:  Plopper. CG: Hyde. DM: StGeorge. JA: Wilson. DW: Dungworth. PL. (1993). Response of
      macaque bronchiolar epithelium to ambient concentrations  of ozone. Am J Pathol 143: 857-866.
Harkema. JR:  Morgan. KT: Gross. EA: Catalano. PJ: Griffith. WC.  (1994).  Consequences of prolonged
      inhalation of ozone on F344/N rats:  Collaborative studies Part VII: Effects on the nasal mucociliary
      apparatus (Vol. 65). Cambridge, MA: Health Effects Institute.
Harkema, JR:  Catalano, PJ: Hotchkiss, JA. (1997a). Consequences of prolonged inhalation of ozone on F344/N
      rats:  Collaborative studies: Part XII Atrophy of bone in nasal turbinates. Cambridge, MA: Health Effects
      Institute.
Harkema. JR:  Hotchkiss. JA: Griffith. WC.  (1997b). Mucous cell  metaplasia in rat nasal epithelium after a 20-
      month exposure to ozone: A morphometric study of epithelial differentiation. Am J Respir Cell Mol Biol
      16: 521-530.
Harkema. JR:  Hotchkiss. JA: Barr. EB: Bennett. CB: Gallup. M: Lee. JK: Basbaum. C. (1999). Long-lasting
      effects of chronic ozone exposure on rat nasal  epithelium. Am J Respir Cell Mol Biol 20: 517-529.
Harkema, JR:  Wagner, JG. (2005). Epithelial and inflammatory responses in the airways of laboratory rats
      coexposed to  ozone and biogenic substances:  Enhancement of toxicant-induced airway injury. Exp
      Toxicol Pathol 57: 129-141. http://dx.doi.Org/10.1016/i.etp.2005.05.013.
Harley, RA:  Marr, LC: Lehner, JK: Giddings, SN. (2005). Changes  in motor vehicle emissions on diurnal to
      decadal time scales and effects on atmospheric composition. Environ Sci Technol 39: 5356-5362.
Haro.  R: Paz. C. (1993). Effects of ozone exposure during pregnancy on ontogeny of sleep in rats. Neurosci
      Lett 164: 67-70. http://dx.doi.Org/10.1016/0304-3940(93)90859-J.
Harvard University. (201 Oa). GEOS-Chem Model [Computer Program]. Cambridge,  MA. Retrieved from
      http://acmg.seas.harvard.edu/geos/
Harvard University. (201 Ob). GEOS-Chem Overview,  from
      http://acmg.seas.harvard.edu/geos/geos overview.html
Harvard University. (2011a). Anthropogenic emissions, from http://wiki.seas.harvard.edu/geos-
      chem/index.php/Anthropogenic  emissions
Harvard University. (2011b). Appendix 9: GEOS-Chem version history, from
      http://acmg.seas.harvard.edu/geos/doc/man/appendix 9.html
Harvard University. (2011c). Overview of GMAO met data products, from http://wiki.seas.harvard.edu/geos-
      chem/index.php/Overview of GMAO met data products
Harvard University. (2011d). Scale factors for anthropogenic emissions, from http://wiki.seas.harvard.edu/geos-
      chem/index.php/Scale factors for  anthropogenic emissions
Harvey, LPP. (2004). Characterizing the annual-mean climatic effect of anthropogenic CO2 and aerosol
      emissions in eight coupled atmosphere-ocean  GCMs. Clim  Pynam  23: 569-599.
      http://dx.doi. org/10.1007/s00382-004-0455-4.
Harward. M: Treshow. M. (1975). Impact of ozone on the growth and reproduction of understorey plants in the
      Aspen zone of western USA. Environ Conserv  2: 17-23. http://dx.doi.org/10.1017/S0376892900000564.
Hassan, R:  Scholes,  R: Ash, N. (2005). Ecosystems and human well-being: Current state and trends. In
      Millennium ecosystem assessment  series (Vol. 1). Washington, PC: Island Press.
Hassett. C:  Mustafa.  MG: Coulson. WF: Elashoff. RM. (1985). Murine lung carcinogenesis following exposure to
      ambient ozone concentrations. J Natl Cancer Inst 75: 771-777.
Draft - Po Not Cite or Quote                       11-32                                 September 2011

-------
Hatch, GE: Slade, R: Stead, AG: Graham, JA. (1986). Species comparison of acute inhalation toxicity of ozone
      and phosgene. J Toxicol Environ Health 19: 43-53. http://dx.doi.org/10.1080/15287398609530905.
Hatch, GE: Wiester, MJ: Overton, JH, Jr: Aissa, M. (1989). Respiratory tract dosimetry of [18]O-labeled ozone in
      rats:  Implications for a rat-human extrapolation of ozone dose. In Atmospheric ozone research and its
      policy implications (Vol. 35, pp. 553-560). Nijmegend, the  Netherlands: Elsevier Science Publishers B.V.
Hatch. GE. (1992). Comparative biochemistry of airway lining fluid. In RA Parent (Ed.), Comparative biology of
      the normal lung (Vol. 1, pp. 617-632). Boca Raton, FL: CRC Press, Inc.
Hatch. GE: Slade. R: Harris. LP:  McDonnell. WF: Devlin. RB: Koren. HS: Costa. PL: McKee. J. (1994). Ozone
      dose and effect in humans and rats: A comparison using oxygen-18 labeling and bronchoalveolar lavage.
      Am J Respir Grit Care Med 150: 676-683.
Hayes, F: Jones, MLM: Mills, G: Ashmore, M. (2007). Meta-analysis of the relative sensitivity of semi-natural
      vegetation species to ozone. Environ  Pollut 146: 754-762. http://dx.doi.Org/10.1016/i.envpol.2006.06.011.
Haves, F: Mills, G: Ashmore, M. (2009). Effects of ozone on inter- and intra-species competition and
      photosynthesis in mesocosms of Lolium perenne and Trifolium repens. Environ Pollut 157: 208-214.
      http://dx.doi.0rg/10.1016/i.envpol.2008.07.002.
Hazbun, ME: Hamilton, R: Holian, A: Eschenbacher, WL. (1993). Ozone-induced increases in substance P and
      8-epi-prostaglandin F2 alpha in the airways of human subjects. Am J Respir Cell Mol Biol 9: 568-572.
      http://dx.doi.Org/10.1165/aircmb/9.5.568.
Hazucha, MJ: Bates, DV: Bromberg, PA. (1989). Mechanism of action of ozone on the human lung. J Appl
      Physiol67: 1535-1541.
Hazucha, MJ: Folinsbee, LJ: Seal, E, Jr. (1992). Effects of steady-state and variable ozone concentration
      profiles on pulmonary function. Am J Respir Grit Care Med 146: 1487-1493.
Hazucha. MJ: Madden. M: Pape. G:  Becker. S: Devlin. R: Koren. HS: Kehrl. H: Bromberg.  PA. (1996). Effects of
      cyclo-oxygenase inhibition on ozone-induced respiratory inflammation and lung function changes. Eur J
      Appl Physiol 73: 17-27.
Hazucha. MJ: Folinsbee. LJ: Bromberg. PA.  (2003).  Distribution and reproducibility of spirometric response to
      ozone by gender and age. J Appl Physiol 95:  1917-1925.
He, XY: Ruan, YN: Chen, W: Lu, T. (2006). Responses of anti-oxidative system in leaves of Ginkgo biloba to
      elevated ozone concentration in urban area. Botanical Studies 47: 409-416.
He, XY: Fu, Sl_: Chen, W: Zhao, TH: Xu, S: Tuba, Z. (2007). Changes in effects of ozone exposure on growth,
      photosynthesis, and respiration of Ginkgo biloba in Shenyang urban area. Photosynthetica 45: 555-561.
      http://dx.doi. org/10.1007/s11099-007-0095-0.
Heagle. AS: Body. DE: Heck. WW. (1973). An open-top field chamber to assess the impact of air pollution on
      plants. J Environ Qual 2:  365-368.
Heagle. AS. (1979). Effects of growth media, fertiliser rate and hour and season  of exposure on sensitivity of
      four soybean cultivars to  ozone. Environ Pollut 18: 313-322.
Heagle, AS: Letchworth, MB: Mitchell, CA. (1983). Effects of growth medium  and fertilizer rate on the yield
      response of soybeans exposed to chronic doses of ozone. Phytopathology 73: 134-139.
      http://dx.doi. org/10.1094/Phyto-73-134.
Heagle, AS: Heck, WW: Lesser, VM: Rawlings, JO. (1987). Effects of daily ozone exposure duration and
      concentration fluctuation  on yield of tobacco.  Phytopathology 77: 856-862.
      http://dx.doi. org/10.1094/Phvto-77-856.
Heagle, AS: Kress, LW: Temple, PJ:  Kohut, RJ:  Miller, JE: Heggestad, HE. (1988). Factors influencing ozone
      dose-yield response relationships in open-top field chamber studies. In WW Heck; OC Taylor; DT Tingey
      (Eds.), Assessment of crop loss from air pollutants: Proceedings of an international  conference (pp. 141-
      179). Raleigh, NC: Elsevier Applied Science.
Heagle, AS. (1989). Ozone and  crop yield*. Annu Rev Phytopathol 27: 397-423.
      http://dx.doi.org/10.1146/annurev.py.27.090189.002145.
Heagle, AS; Miller, JE; Rawlings, JO; Vozzo, SF. (1991). Effect of growth stage on soybean response to chronic
      ozone exposure. J Environ Qual 20: 562-570.
      http://dx.doi.org/10.2134/ieg1991.00472425002000030010x.
Heagle, AS; Brandenburg, RL; Burns, JC; Miller, JE. (1994a). Ozone and carbon dioxide effects on spider mites
      in white clover and peanut. J Environ  Qual 23: 1168-1176.
      http://dx.doi.org/10.2134/ieg1994.00472425002300060006x.
Heagle, AS; Miller, JE; Sherrill, DE. (1994b). A white clover system to estimate effects of tropospheric ozone on
      plants. J Environ Qual 23: 613-621. http://dx.doi.org/10.2134/jeg1994.00472425002300030030x.
Heagle, AS; Reinert,  RA; Miller,  JE. (1996). Response of white clover to ozone in different  environments. J
      Environ Qual 25: 273-278. http://dx.doi.org/10.2134/jeg1996.0047242500250002001 Ox.
Heath. RL. (2008). Modification of the biochemical pathways of plants induced by ozone: What are the varied
      routes to change? Environ Pollut 155: 453-463. http://dx.doi.Org/10.1016/i.envpol.2008.03.010.
Heath, RL: Lefohn, AS; Musselman,  RC. (2009). Temporal processes that contribute to nonlinearity in
      vegetation responses to ozone exposure and dose. Atmos Environ 43: 2919-2928.
      http://dx.doi.0rg/10.1016/i.atmosenv.2009.03.011.
Draft - Do Not Cite or Quote                       11-33                                  September 2011

-------
Heck, WW: Philbeck, RB: Dunning, JA. (1978). A continuous stirred tank reactor (CSTR) system for exposing
      plants to gaseous air contaminants: Principles, specifications, construction, and operation. Washington,
      DC: U.S. Government Printing Office.
Heck, WW: Taylor, PC: Adams, R: Bingham, G: Miller, J: Preston, E: Weinstein, L. (1982). Assessment of crop
      loss from ozone.  J Air Pollut Control Assoc 32: 353-361.
Heck. WW: Cure. WW: Rawlings. JO: Zaragoza. LJ: Heagle. AS: Heggestad. HE: Kohut. RJ: Kress. LW:
      Temple, PJ. (1984). Assessing impacts of ozone on agricultural crops: II. Crop yield functions and
      alternative exposure statistics. J Air Pollut Control Assoc 34: 810-817.
Heck. WW: Heagle. AS: Miller. JE: Rawlings. JO. (1991). A national program (NCLAN) to assess the impact of
      ozone on agricultural resources. In RL Berglund; DR Lawson; DJ McKee (Eds.), Tropospheric ozone and
      the environment: papers from an international conference;  March 1990; Los Angeles, CA(pp. 225-254).
      Pittsburgh, PA: Air & Waste Management Association.
Heck, WW; Cowling, EB. (1997). The need for a long term cumulative secondary ozone standard - An
      ecological  perspective. EM  January: 23-33.
Heggestad. HE. (1991). Origin of Bel-W3, Bel-C and Bel-B tobacco varieties and their use as indicators of
      ozone. Environ Pollut 74: 263-291. http://dx.doi.org/10.1016/0269-7491 (91)90076-9.
HEI. (Health Effects Institute). (2003). Revised analyses of time-series studies of air pollution and health:
      Revised analyses of the National Morbidity, Mortality, and Air Pollution Study (NMMAPS), Part II.
      Cambridge, MA.  http://pubs.healtheffects.org/view.php?id=4.
Heidenfelder, BL; Reif, DM; Harkema, JR; Cohen Hubal, EA; Hudgens, EE; Bramble,  LA; Wagner, JG;
      Morishita, M; Keeler, GJ;  Edwards, SW; Gallagher, JE. (2009). Comparative microarray analysis and
      pulmonary changes in brown Norway rats exposed to ovalbumin and concentrated air particulates.
      ToxicolSci 108: 207-221.
Heidenreich. B: Haberer. G: Mayer. K: Sandermann. H:  Ernst. D. (2005). CDNA array-analysis of mercury- and
      ozone-induced genes in Arabidopsis thaliana. Acta Physiologiae Plantarum 27: 45-51.
      http://dx.doi. org/10.1007/s11738-005-0035-1.
Held. IM: Soden. BJ. (2000). Water vapor feedback and global warming. Annual  Review of Energy and the
      Environment 25:  441-475.
Hemmingsen, A; Fryer, AA; Hepple, M; Strange, RC; Spiteri, MA. (2001). Simultaneous identification of GSTP1
      Ne105->Val105 and Ala114->Val114 substitutions using an  amplification refractory mutation system
      polymerase chain reaction assay: Studies in patients with asthma. Respir Res 2: 255-260.
      http://dx.doi.org/10.1186/rr64.
Henderson. BH: Pinder. RW: Crooks. J: Cohen. RC: Hutzell. WT:  Sarwar. G: Goliff. WS: Stockwell. WR: Fahr. A:
      Mathur,  R; Carlton, AG; Vizuete, W. (2010). Evaluation of simulated photochemical partitioning of
      oxidized nitrogen in the upper troposphere. Atmos Chem Phys 10: 20125-20165.
      http://dx.doi. org/10.5194/acpd-10-20125-2010.
Henderson, FW; Dubovi, EJ; Harder, S; Seal, E, Jr; Graham, D. (1988). Experimental  rhinovirus infection  in
      human volunteers exposed to ozone. Am J Respir Grit Care Med 137: 1124-1128.
Hendrev, GR; Kimball, BA. (1994). The FACE program.  Agr Forest Meteorol 70: 3-14.
      http://dx.doi. org/10.1016/0168-1923(94)90044-2.
Hendrev. GR: Ellsworth. PS: Lewin. KF: Nagv. J. (1999). A free-air enrichment system for exposing tall  forest
      vegetation to elevated atmospheric CO2. Global  Change Biol 5:  293-309.
      http://dx.doi. org/10.1046/j. 1365-2486.1999.00228.x.
Henrotin. JB: Besancenot. JP: Beiot. Y: Giroud. M. (2007). Short-term effects of ozone air pollution on
      ischaemic stroke occurrence: A case-crossover analysis from a 10-year population-based study  in  Dijon,
      France.  Occup Environ Med 64: 439-445.
Henrotin, JB; Zeller, M; Lorgis, L; Cottin, Y; Giroud, M; Beiot, Y. (2010).  Evidence of the role of short-term
      exposure to ozone on ischaemic cerebral and cardiac events: The Dijon Vascular Project (DIVA). Heart
      96: 1990-1996. http://dx.doi.org/10.1136/hrt.2Q10.200337.
Henry. GT: Gordon. CS. (2003). Driving less for better air: Impacts of a  public information campaign. J  Policy
      Anal Manage 22: 45-63. http://dx.doi.org/10.1002/pam. 10095.
Henry, HAL; Brizgys, K;  Field, CB. (2008). Litter decomposition in a California annual grassland: Interactions
      between photodegradation  and litter layer thickness. Ecosystems 11: 545-554.
      http://dx.doi.org/10.1007/s10021-008-9141-4.
Herbert, RA; Hailey, JR; Grumbein, S; Chou, BJ; Sills, RC; Haseman, JK;  Goehl, T; Miller,  RA; Roycroft, JH;
      Boorman, GA. (1996). Two-year and lifetime toxicity and carcinogenicity studies of ozone in B6C3F1
      mice. Toxicol Pathol 24: 539-548.
Hernandez-Cadena, L; Holguin,  F; Barraza-Villarreal, A; Del Rfo-Navarro, BE; Sienra-Monge, JJ; Romieu, I.
      (2009). Increased levels of  outdoor air pollutants are associated  with reduced bronchodilation in children
      with asthma. Chest 136: 1529-1536. http://dx.doi.org/10.1378/chest.08-1463.
Hernandez, ML; Lay, JC; Harris, B; Esther, CR;  Brickev, WJ; Bromberg, PA; Diaz-Sanchez, D; Devlin, RB;
      Kleeberger. SR: Alexis. NE: Peden. DB. (2010). Atopic asthmatic subjects but not atopic subjects without
      asthma  have enhanced inflammatory response to ozone. J Allergy Clin Immunol 126: 537-544.
      http://dx.doi.0rg/10.1016/i.iaci.2010.06.043.
Draft - Do Not Cite or Quote                        11-34                                  September 2011

-------
Heroux, ME: Clark, N; Van Ryswyk, K: Mallick, R: Gilbert, ML: Harrison, I: Rispler, K: Wang, D:
      Anastassopoulos, A: Quay, M:  MacNeill, M: Wheeler, AJ. (2010). Predictors of indoor air concentrations
      in smoking and non-smoking residences. Int J Environ Res Public Health 7: 3080-3099.
      http://dx.doi.org/10.3390/iierph7083080.
HEW. (U.S. Department of Health, Education and Welfare). (1964). Smoking and  health: Report of the advisory
      committee to the surgeon general of the public health service. Washington, DC: U.S. Department of
      Health, Education, and Welfare.
HHS. (U.S. Department of Health and Human Services, Office of the Surgeon General). (2004). The health
      consequences of smoking: A report of the Surgeon General. Washington, DC: U.S. Department of Health
      and Human Services, http://www.surgeongeneral.gov/library/smokingconsequences/.
Hicks, A: Kourteva, G: Hilton, H: Li, H: Lin, T: Liao, W: Li, Y: Wei, X: March, T: Benson, J: Renzetti, L. (201 Oa).
      Cellular and molecular characterization of ozone-induced pulmonary  inflammation in the Cynomolgus
      monkey. Inflammation 33: 144-156. http://dx.doi.org/10.1007/s10753-009-9168-5.
Hicks. A: Goodnow. R. Jr: Cavallo. G: Tannu. SA: Ventre. JD: Lavelle. D: Lora. JM: Satiawatcharaphong. J:
      Brovarnev. M: Dabbagh. K: Tare. NS: Oh. H: Lamb. M: Sidduri. A: Dominigue.  R: Qiao. Q: Lou. JP:
      Gillespie, P: Fotouhi,  N; Kowalczyk, A: Kurylko, G: Hamid, R: Wright,  MB: Pamidimukkala, A: Egan, T:
      Gubler. U: Hoffman. AF: Wei. X: Li. YL: O'Neil. J: Marcano. R: Pozzani.  K:  Molinaro. T: Santiago. J:
      Singer. L: Hargaden.  M: Moore. D: Catala. AR: Chao. LC: Benson. J:  March. T: Venkat. R: Mancebo. H:
      Renzetti,  LM. (201 Ob). Effects of LTB4 receptor antagonism on pulmonary  inflammation  in rodents and
      non-human primates. Prostaglandins Other Lipid Mediat 92: 33-43.
      http://dx.doi. org/10.1016/j. prostaglandins. 2010.02.003.
Higgins, ITT: D'Arcy, JB: Gibbons, PI: Avol, EL: Gross, KB.  (1990). Effect of  exposures to ambient ozone on
      ventilatory lung function in children. Am J Respir Crit Care Med 141: 1136-1146.
Hildebrand. E: Skellv. JM: Fredericksen. TS. (1996). Foliar response of ozone-sensitive hardwood tree species
      from 1991 to 1993 in the Shenandoah National Park, Virginia. Can J For Res 26: 658-669.
Hildesheim. J: Fornace. AJ. Jr. (2004). The dark side of light: The damaging  effects of UV rays and the
      protective efforts of MAP kinase signaling in the epidermis. DMA Repair 3:  567-580.
Hill, AB. (1965). The environment and disease: Association or causation? Proc R  Soc Med  58: 295-300.
Hillstrom, M: Meehan, TD: Kelly, K: Lindroth, RL. (2010a). Soil carbon and nitrogen mineralization following
      deposition of insect frass and greenfall from forests under elevated CO2 and  O3. Plant Soil 336: 75-85.
      http://dx.doi.org/10.1007/s11104-010-0449-4.
Hillstrom. ML: Lindroth. RL. (2008). Elevated atmospheric carbon dioxide and ozone alter forest insect
      abundance and community composition. Insect Conservation and Diversity 1: 233-241.
      http://dx.doi.0rg/10.1111/i.1752-4598.2008.00031.x.
Hillstrom. ML: Vigue. LM: Covle. PR: Raffa. KF: Lindroth. RL. (201 Ob). Performance of the  invasive weevil
      Polydrusus sericeus is influenced by atmospheric CO2 and host species. Agr Forest Entomol 12:  285-
      292. http://dx.doi.0rg/10.1111/i.1461-9563.2010.00474.x.
Hiltermann, JTN: Lapperre, TS: Van Bree, L: Steerenberg, PA: Brahim, JJ: Sont, JK: Sterk, PJ:  Hiemstra, PS:
      Stolk, J. (1999).  Ozone-induced inflammation assessed in sputum and bronchial lavage  fluid from
      asthmatics: A new noninvasive tool in epidemiologic studies on air pollution and asthma. Free Radic Biol
      Med 27: 1448-1454.
Hiltermann. TJN: Stolk. J: Hiemstra. PS:  Fokkens. PHB: Rombout.  PJA: Sont. JK: Sterk. PJ: Diikman. JH.
      (1995). Effect of ozone exposure on maximal airway narrowing in non-asthmatic and asthmatic subjects.
      ClinSci(Lond)89:619-624.
Hiltermann. TJN: de Bruiine. CR:  Stolk. J: Zwinderman. AH: FThM. S: Roemer.  W: Steerenberg. PA: Fischer.
      PH: van Bree, L:  Hiemstra,  PS. (1997). Effects of photochemical air pollution and allergen exposure on
      upper respiratory tract inflammation in asthmatics. Am J  Respir Crit Care Med 156: 1765-1772.
Hiltermann, TJN; Peters, EA; Alberts,  B; Kwikkers, K; Borggreven, PA; Hiemstra, PS; Dijkman, JH;  van Bree,
      LA; Stolk, J. (1998). Ozone-induced airway hyperresponsiveness in patients with asthma: Role of
      neutrophil-derived serine proteinases. Free Radic Biol Med 24: 952-958.
Himanen. SJ: Nerg. AM: Holopainen. JK. (2009a). Degree of herbivore feeding  damage as an important
      contributor to multitrophic plant-parasitoid signaling under climate change.  4: 249-251.
Himanen. SJ: Nerg. AM: Nissinen. A: Pinto. DM: Stewart. CN: Poppy. GM: Holopainen. JK. (2009b). Effects of
      elevated carbon dioxide and ozone on volatile terpenoid emissions and multitrophic communication of
      transgenic insecticidal oilseed rape (Brassica napus). New Phytol 181: 174-186.
      http://dx.doi.0rg/10.1111/i.1469-8137.2008.02646.x.
Himanen, SJ; Nerg, AM; Nissinen, A; Stewart, CN; Poppy, GM;  Holopainen,  JK. (2009c). Elevated atmospheric
      ozone increases concentration of insecticidal Bacillus thuringiensis (Bt) Cry 1 Ac protein in Bt Brassica
      napus and reduces feeding of a Bt target herbivore on the non-transgenic parent. Environ Pollut 157:
      181-185.  http://dx.doi.0rg/10.1016/i.envpol.2008.07.006.
Draft - Do Not Cite or Quote                       11-35                                 September 2011

-------
Himes, BE: Hunninghake, GM: Baurley, JW: Rafaels, NM: Sleiman, P: Strachan, DP: Wilk, JB: Willis-Owen,
      SAG: Klanderman, B: Lasky-Su, J: Lazarus, R: Murphy, AJ: Soto-Quiros, ME: Avila, l_: Beaty, T: Mathias,
      RA: Ruczinski, I: Barnes, KG: Celedon, JC: Cookson, WOC: Gauderman, WJ: Gilliland, FD: Hakonarson,
      H: Lange, C: Moffatt, MF: O'Connor, GT: Raby, BA: Silverman, EK: Weiss, ST. (2009). Genome-wide
      Association Analysis Identifies PDE4D as an Asthma-Susceptibility Gene. Am J Hum Genet 84: 581-593.
      http://dx.doi.0rg/10.1016/i.aiha.2009.04.006.
Hinwood, Al_: De Klerk, N: Rodriguez, C: Jacoby, P:  Runnion, T: Rye, P: Landau,  l_: Murray, F: Feldwick, M:
      Spickett. J.  (2006). The relationship between  changes in daily air pollution  and hospitalizations in Perth,
      Australia 1992-1998: A case-crossover study. Int J Environ Health Res 16:  27-46.
      http://dx.doi.org/10.1080/09603120500397680.
Hocking, WK: Carey-Smith, T: Tarasick, DW: Argall,  PS: Strong, K:  Rochon, Y: Zawadzki, I: Taylor, PA. (2007).
      Detection of stratospheric ozone intrusions by windprofiler radars. Nature 450: 281-284.
      http://dx.doi.org/10.1038/nature06312.
Hoek. G: Brunekreef. B: Kosterink. P: Van den  Berg. R: Hofschreuder. P. (1993). Effect of ambient ozone on
      peak expiratory flow of exercising children in the Netherlands. Arch Environ Occup Health 48: 27-32.
      http://dx.doi. org/10.1080/00039896.1993.9938390.
Hoek. G: Brunekreef. B. (1995). Effect  of photochemical air pollution on acute respiratory symptoms in children.
      Am J Respir Grit Care Med 151: 27-32.
Hoek, G: Beelen, R:  de Hoogh, K: Vienneau, D: Gulliver, J: Fischer, P: Briggs,  D.  (2008). A review of land-use
      regression models to assess spatial variation of outdoor air pollution [Review]. Atmos Environ 42: 7561-
      7578.
Hofzumahaus, A: Rohrer, F: Keding, l_: Bohn, B: Brauers, T: Chih-Chung, C: Fuchs, H: Holland, F: Kita, K:
      Kondo. Y: Xin. L: Shengrong. L: Min. S:  Limin. Z: Wahner. A: Yuanhang. Z. (2009). Amplified trace gas
      removal in the troposphere. Science 324: 1702-1704. http://dx.doi.org/10.1126/science.1164566.
Hogg, A: Uddling, J:  Ellsworth, D: Carroll, MA: Presslev, S: Lamb, B: Vogel, C.  (2007). Stomatal and non-
      stomatal fluxes of  ozone to a northern mixed hardwood forest.  Tellus B Chem Phys Meteorol 59: 514-
      525. http://dx.doi.0rg/10.1111/i.1600-0889.2007.00269.x.
Hogsett, WE: Tingey, DT: Holman, SR. (1985). A programmable exposure control system for determination of
      the effects of pollutant exposure regimes on plant growth. Atmos Environ 19: 1135-1145.
      http://dx.doi.org/10.1016/0004-6981(85)90198-2.
Hogsett, WE: Olszyk, D: Ormrod, DP: Taylor, GE, Jr: Tingev, DT. (1987a). Air pollution exposure systems and
      experimental protocols: Volume 2: Description of facilities. (EPA/600/3-87/037b). Corvallis, OR: U.S.
      Environmental Protection Agency, Environmental Research Laboratory.
      http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=30000KQH.txt.
Hogsett. WE: Olszvk. D: Ormrod. DP: Taylor. GE. Jr: Tingev. DT. (1987b). Air pollution exposure systems and
      experimental protocols: Volume I: A review and evaluation of performance. (EPA/600/3-87/037a).
      Corvallis, OR: U.S. Environmental Protection  Agency.
Hogsett, WE: Tingev, DT: Lee, EH. (1988). Ozone exposure indices: Concepts for development and evaluation
      of their use. In Assessment of crop loss  from air pollutants: Proceedings of an  international conference
      (pp. 107-138). New York: Elsevier Applied Science.
Hogsett. WE: Weber. JE: Tingev. D: Herstrom. A: Lee. EH: Laurence.  JA. (1997).  Environmental auditing: An
      approach for characterizing tropospheric ozone risk to forests.  J Environ Manage 21: 105-120.
      http://dx.doi. org/10.1007/s002679900010.
Hogsett. WE: Tingev. DT: Lee. EH: Beedlow. PA: Andersen. CP (2008). An approach for evaluating the
      effectiveness  of various ozone Air Quality Standards for protecting trees. Environ Manage 41: 937-948.
      http://dx.doi. org/10.1007/s00267-007-9057-3.
Hoigne,  J: Bader, H.  (1983). Rate constants of reactions of ozone with organic and inorganic compounds in
      water- II: Dissociating organic compounds. Water Res 17: 185-194. http://dx.doi.org/10.1016/0043-
      1354(83)90099-4.
Holick. MF. (2004). Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers,
      and cardiovascular disease. Am J Clin Nutr 80: 1678S-1688S.
Holland, MM: Bitz, CM. (2003). Polar amplification of climate change in coupled models. Clim Dynam 21: 221-
      232. http://dx.doi.org/10.1007/s00382-003-0332-6.
Hollingsworth. JW: Maruoka. S:  Li. Z: Potts. EN: Brass.  DM: Garantziotis. S: Fong. A: Foster. WM: Schwartz.
      DA. (2007). Ambient ozone primes pulmonary innate immunity in mice. J Immunol 179: 4367-4375.
Hollingsworth, JW: Free,  ME:  Li, Z: Andrews, LN: Nakano, H: Cook, DN. (2010). Ozone activates pulmonary
      dendritic cells and promotes allergic sensitization through  a Toll-like receptor 4-dependent mechanism. J
      Allergy Clin Immunol 125: 1167-1170.
Hollingsworth. JW. II: Cook. DN: Brass. DM: Walker. JKL: Morgan. PL: Foster.  WM: Schwartz. DA. (2004). The
      role of Toll-like receptor 4 in environmental airway injury in mice. Am J Respir Crit Care Med 170: 126-
      132.
Holmes. WE: Zak. PR: Pregitzer. KS: King. JS. (2006). Elevated CO2 and O3 alter soil nitrogen transformations
      beneath trembling aspen, paper birch, and sugar maple. Ecosystems 9: 1354-1363.
      http://dx.doi.org/10.1007/s10021-006-0163-5.
Draft - Do Not Cite or Quote                       11-36                                 September 2011

-------
Holtby, LB: Bothwell, ML. (2008). Effects of solar ultraviolet radiation on the behaviour of juvenile coho salmon
      (Oncorhynchus kisutch): Avoidance, feeding, and agonistic interactions. Can J Fish Aquat Sci 65: 701-
      711. http://dx.doi.0rg/10.1139/F08-013.
Holtzman, MJ: Cunningham, JH: Sheller, JR: Irsigler, GB: Nadel, JA: Boushev, HA. (1979). Effect of ozone on
      bronchial reactivity in atopic and nonatopic subjects. Am Rev Respir Dis 120: 1059-1067.
Holtzman. MJ: Fabbri. LM: O'Bvrne. PM: Gold. BD: Aizawa. H: Walters. EH: Alpert. SE: Nadel. JA. (1983).
      Importance of airway inflammation for hyperresponsiveness induced by ozone. Am Rev Respir Dis 127:
      686-690.
Holz. O: Jorres.  RA: Timm.  P: Mucke. M: Richter. K: Koschvk. S: Magnussen. H. (1999). Ozone-induced airway
      inflammatory changes differ between individuals and are reproducible. Am J Respir Grit Care Med 159:
      776-784.
Holz, O: Mucke, M: Paasch, K: Bohme, S: Timm, P: Richter, K: Magnussen, H: Jorres, RA. (2002). Repeated
      ozone exposures enhance bronchial allergen responses in subjects with rhinitis or  asthma. Clin Exp
      Allergy 32: 681-689.
Holz. O: Tal-Singer. R: Kanniess. F: Simpson. KJ: Gibson. A: Vessev. RSJ: Janicki. S: Magnussen. H: Jorres.
      RA: Richter, K. (2005). Validation of the human ozone challenge model as a tool for assessing anti-
      inflammatory drugs in early development. J Clin Pharmacol 45: 498-503.
Hong. B: Weinstein. D: Swanev. D. (2006). Assessment of ozone effects on nitrate export from Hubbard Brook
      Watershed 6. Environ Pollut 141: 8-21. http://dx.doi.Org/10.1016/i.envpol.2005.08.030.
Hoppe, P: Praml, G: Rabe,  G: Lindner, J: Fruhmann, G: Kessel, R. (1995). Environmental ozone field study on
      pulmonary and subjective responses of assumed risk groups. Environ  Res 71: 109-121.
Hoppe, P: Peters, A:  Rabe,  G: Praml, G: Lindner, J: Jakobi, G: Fruhmann, G: Nowak, D. (2003). Environmental
      ozone effects in different  population subgroups. Int J Hyg Environ Health 206:  505-516.
      http://dx.doi.orq/10.1078/1438-4639-00250.
Horak, F, Jr: Studnicka, M: Gartner, C: Spengler, JD: Tauber, E: Urbanek,  R: Veiter, A: Frischer,  T. (2002).
      Particulate matter and lung function growth in children: a 3-yr follow-up study in Austrian schoolchildren.
      Eur Respir J 19: 838-845.
Horstman, DH: Folinsbee, LJ: Ives, PJ: Abdul-Salaam, S: McDonnell, WF. (1990). Ozone concentration and
      pulmonary response relationships for 6.6-hour exposures with five  hours of moderate exercise to 0.08,
      0.10, and 0.12 ppm.  Am J Respir Grit Care Med 142: 1158-1163.
Horstman, DH: Ball, BA: Brown, J: Gerritv, T: Folinsbee, LJ. (1995). Comparison of pulmonary responses of
      asthmatic and nonasthmatic subjects performing light exercise while exposed to a low level of ozone.
      Toxicol Ind Health 11: 369-385.
Horvath, L: Nagy, Z: Weidinger, T: Ariz, R: Luke, WT: Valigura, R: Pinto, JP: Womack, J. (1995). Measurement
      effluxes of trace gases (O3, NOX, SO2, CO2, HNO3), particulate sulfate  and nitrate, water vapour over
      short vegetation by gradient and eddy correlation techniques in Hungary. Ann Geophys 13: C490.
Horvath, SM: Gliner, JA: Matsen-Twisdale, JA. (1979).  Pulmonary function and maximum exercise responses
      following  acute ozone exposure. Aviat Space Environ Med 50: 901-905.
Horvath, SM: Gliner, JA: Folinsbee, LJ. (1981). Adaptation to ozone: Duration of  effect. Am Rev Respir Dis 123:
      496-499.
Hosseinpoor. AR: Forouzanfar. MH: Yunesian.  M: Asghari. F: Naieni. KH: Farhood. D. (2005). Air pollution and
      hospitalization due to angina pectoris in Tehran, Iran: A time-series study.  Environ Res 99: 126-131.
Hotchkiss, JA: Harkema, JR: Henderson, RF. (1991). Effect of cumulative  ozone exposure on ozone-induced
      nasal epithelial hyperplasia and secretory metaplasia in rats. Exp Lung Res 15: 589-600.
Houslev. DG: Eccles. R: Richards. RJ. (1996).  Gender difference in the concentration of the antioxidant uric
      acid in human nasal  lavage. Acta Otolaryngol 116: 751-754.
Howard, AR: van lersel, MW: Richards, JH: Donovan, LA. (2009). Night-time transpiration can decrease
      hydraulic redistribution. Plant Cell Environ 32: 1060-1070. http://dx.doi.Org/10.1111/i.1365-
      3040.2009.01988.x.
Howarth. PH: Persson. CG: Meltzer. EO: Jacobson. MR: Durham. SR: Silkoff. PE. (2005). Objective monitoring
      of nasal airway inflammation in rhinitis. J Allergy Clin Immunol 115:  S414-S441.
Hsu, J: Prather,  MJ. (2009). Stratospheric variability and tropospheric ozone.  J Geophys Res 114: D06102.
      http://dx.doi. org/10.1029/2008JD010942.
Hu. PC: Miller. FJ: Daniels.  MJ: Hatch. G. (1982). Protein accumulation in lung lavage fluid following ozone
      exposure. Environ Res 29: 377-388. http://dx.doi.org/10.10116/0013-9351 (82)90039-1.
Hu, S, -C: Ben-Jebria, A: Ultman, JS.  (1994). Longitudinal distribution of ozone absorption in the lung: Effects of
      respiratory flow. J Appl Physiol 77: 574-583.
Hu, SC: Ben-Jebria, A: Ultman, JS. (1992). Longitudinal distribution of ozone absorption in the lung: Quiet
      respiration in healthy subjects. J Appl Physiol 73: 1655-1667.
Hudman. RC:  Murray. LT: Jacob. DJ:  Millet. DB: Turquetv. S: Wu.  S: Blake. PR: Goldstein. AH:  Hollowav. J:
      Sachse, GW. (2008). Biogenic versus anthropogenic sources of CO in the United States. Geophys Res
      Lett 35: L04801. http://dx.doi.org/10.1029/2007gl032393.
Draft - Do Not Cite or Quote                       11-37                                 September 2011

-------
Huen, K: Gunn, L: Duramad, P: Jeng, M: Scalf, R: Holland, N. (2006). Application of a geographic information
      system to explore associations between air pollution and micronucleus frequencies in African American
      children and adults. Environ Mol  Mutagen 47: 236-246. http://dx.doi.Org/10.1002/em.2Q193.
Huffman, LJ: Beighley, CM: Frazer, DG: McKinney, WG: Porter, DW. (2006). Increased susceptibility of the
      lungs of hyperthyroid rats to oxidant injury: Specificity of effects. Toxicology 225: 119-127.
      http://dx.doi.0rg/10.1016/i.tox.2006.05.008.
Hughes, AM: Armstrong, BK: Vajdic, CM: Turner, J: Grulich, AE: Fritschi, L: Milliken, S:  Kaldor, J: Benke, G:
      Kricker. A. (2004). Sun exposure may protect against non-Hodgkin lymphoma: A case-control study. Int J
      Cancer 112: 865-871.
Hui, D: Sims, DA: Johnson, DW: Chang, W: Luo, Y. (2002). Effects of gradual  versus step increases in carbon
      dioxide on Plantago photosynthesis and growth in a microcosm study. Environ Exp Bot 47: 51-66.
Hunt, J. (2002). Exhaled breath condensate: An evolving tool for noninvasive evaluation of lung disease. J
      Allergy Clin Immunol 110: 28-34. http://dx.doi.org/10.1067/mai.2002.124966.
Hunt. JF: Fang. K: Malik. R: Snvder. A: Malhotra. N: Platts-Mills. TAE: Gaston. B.  (2000). Endogenous airway
      acidification: Implications for asthma pathophysiology. Am J Respir Grit Care Med 161: 694-699.
Hurst, DJ: Gardner, DE; Coffin, PL. (1970). Effect of ozone on acid hydrolases of the pulmonary alveolar
      macrophage. J Reticuloendothel Soc 8: 288-300.
Husar.  RB: Renard. WP. (1998) Ozone as a function of local wind speed and direction:  Evidence of local and
      regional transport 91st annual meeting and exhibition of the Air & Waste Management Association.  San
      Diego, CA.
Hutcheon, JA: Platt, RW. (2008). The missing data problem in birth weight percentiles and thresholds for "small-
      for-gestational-age". Am J Epidemiol 167: 786-792. http://dx.doi.org/10.1093/aje/kwm327.
Hwang. B. -F: Lee. Y. -L: Lin. Y. -C: Jaakkola. JJK: Guo. YL. (2005). Traffic related air pollution as a determinant
      of asthma among Taiwanese school children. Thorax 60: 467-473.
Hwang, B, -F: Jaakkola, JJK:  Lee, Y, -L: Lin, Y, -C: Y-LL, G. (2006). Relation between air pollution and allergic
      rhinitis in Taiwanese schoolchildren. Respir Res 7: 23.
Hwang. BF: Jaakkola. JJ. (2008). Ozone and other air pollutants and the risk of oral clefts. Environ Health
      Perspect 116: 1411-1415.
Hyde, DM: Plopper, CG: Harkema, JR: St George, JA: Tyler, WS: Dungworth,  PL. (1989). Ozone-induced
      structural changes in monkey respiratory system. In T Schneider; SD Lee;  GJR Wolters; LD Grant (Eds.),
      Atmospheric ozone research and its policy implications: Proceedings of the 3rd US-Dutch International
      Symposium, Nijmegen, the Netherlands May 9-13, 1988 (pp. 523-532). Nijmegen, the Netherlands:
      Elsevier Science Publishers.
Hyde, DM; Hubbard, WC; Wong, V; Wu, R; Pinkerton, K; Plopper, CG. (1992). Ozone-induced acute
      tracheobronchial epithelial injury: relationship to granulocyte emigration in  the lung. Am J Respir Cell Mol
      Biol 6: 481-497.
Hyttinen,  M; Pasanen,  P; Kalliokoski, P. (2006). Removal of ozone on clean, dusty and sooty supply air filters.
      Atmos Environ 40: 315-325.
IARC. (International Agency for Research on Cancer). (2006). Preamble to the IARC monographs. Lyon,
      France.  http://monographs.iarc.fr/ENG/Preamble/.
Ichinose.  T: Arakawa. K: Shimoio.  N: Sagai. M. (1988). Biochemical effects of combined gases of nitrogen
      dioxide and ozone: II Species differences in lipid peroxides and antioxidative protective enzymes in the
      lungs. Toxicol Lett 42: 167-176.
ICNIRP (International Commission on Non-Ionizing Radiation Protection).  (2004). Guidelines on limits of
      exposure to ultraviolet  radiation of wavelengths between 180 nm and 400 nm (incoherent optical
      radiation). In ICNIRP Guidelines  (Vol. 87, pp. 171-186). Oberschleissheim, Germany.
ICRP (International Commission on Radiological Protection). (1994). Human respiratory tract model for
      radiological protection: A report of a task group of the International Commission on Radiological
      Protection. Ann  ICRP 24: 1-482.
Ignatenko. AV:  Cherenkevich. SN. (1985). Reactivity of amino-acids and proteins  in reactions with ozone.  Kinet
      Catal26: 1145-1148.
Ihorst, G; Frischer, T; Horak, F; Schumacher, M; Kopp, M; Forster, J; Mattes, J; Kuehr, J. (2004). Long- and
      medium-term ozone effects on lung growth including a broad spectrum of exposure. Eur Respir J 23:
      292-299.
lijima, MK; Kobayashi,  T. (2004). Nasal allergy-like symptoms aggravated by ozone exposure in a
      concentration-dependent manner in guinea pigs. Toxicology 199: 73-83.
      http://dx.doi.0rg/10.1016/i.tox.2004.01.008.
Inman, RE;  Ingersoll, RB; Levy, EA. (1971). Soil: a natural sink for carbon monoxide. Science 172: 1229-1231.
      http://dx.doi.org/10.1126/science.172.3989.1229.
Innes. JL: Skellv. JM: Schaub. M. (2001). Ozone and broadleaved species: A guide to the identification of
      ozone-induced foliar injury.  In. Bern, Switzerland: Paul Haupt Publishers.
Inoue. K:  Takano. H: Kaewamatawong. T: Shimada. A: Suzuki. J: Yanagisawa. R: Tasaka. S: Ishizaka. A: Satoh.
      JVl (2008). Role of metallothionein in lung inflammation induced by ozone exposure in mice. Free Radic
      Biol Med 45: 1714-1722. http://dx.doi.Org/10.1016/i.freeradbiomed.2008.09.008.
Draft - Do Not Cite or Quote                        11-38                                  September 2011

-------
loannidis, JPA. (2008). Why most discovered true associations are inflated. Epidemiology 19: 640-648.
      http://dx.doi. org/10.1097/EDE.Ob013e31818131 e7.
loki, M: Takahashi, S: Nakajima, N: Fujikura, K: Tamaoki, M: Sail, H:  Kubo, A: Aono, M: Kanna, M: Ogawa, D:
      Fukazawa, J:  Oda, Y: Yoshida, S: Watanabe, M: Hasezawa, S: Kondo, N. (2008). An unidentified
      ultraviolet-B-specific photoreceptor mediates transcriptional activation of the cyclobutane pyrimidine
      dimer photolyase gene in plants. Planta 229: 25-36. http://dx.doi.ora/10.1007/s00425-008-0803-4.
IPCC. (Intergovernmental Panel on Climate Change). (2000). Special report on emissions scenarios: A special
      report of Working Group III of the Intergovernmental Panel on Climate Change. Cambridge, UK:
      Intergovernmental Panel on Climate Change; Cambridge University Press.
      http://www.grida.no/climate/ipcc/emission/.
IPCC. (Intergovernmental Panel on Climate Change). (2007a). Summary for policymakers. In: Impacts,
      Adaptation and Vulnerability. Contribution  of Working Group II to the Fourth Assessment Report of the
      Intergovernmental Panel on Climate Change. In Climate Change 2007. Cambridge, UK: Cambridge
      University Press.
IPCC. (Intergovernmental Panel on Climate Change). (2007b). Summary for policymakers. In: The Physical
      Science Basis. Contribution of Working Group  I to the Fourth Assessment Report of the
      Intergovernmental Panel on Climate Change. In Climate Change 2007. Cambridge, United Kingdom and
      New York, NY, USA:  Cambridge University Press.
      http://www.ipcc.ch/publications  and data/ar4/wg2/en/spm.html.
Iriti, M: Faoro, F. (2009). Chemical diversity and defence  metabolism: How plants cope with pathogens and
      ozone pollution. International Journal  of Molecular Sciences 10: 3371-3399.
      http://dx.doi.org/10.3390/iims10083371.
Iriti. M: Pi Maro. A: Bernasconi. S:  Burlini. N: Simonetti. P: Picchi. V:  Panigada. C: Gerosa. G: Parente. A:
      Faoro. F. (2009). Nutritional traits of bean  (Phaseolus vulgaris) seeds from plants chronically exposed to
      ozone pollution. J Agric Food Chem 57:  201-208. http://dx.doi.org/10.1021/jf802819m.
IRRI. (International Rice Research Institute). (2002). Annual Report.  Los Banos, Laguna in the Philippines.
      http://irri.org/.
Isaksen, ISA: Berntsen, TK: Wang, WC. (2001). NOx  emissions from aircraft: Its impact on the global
      distribution of CH4 and O3 and on radiative forcing. Terr Atmos Ocean Sci 12: 63-78.
Isebrands, JG:  Dickson, RE: Rebbeck, J: Karnoskv, DF. (2000). Interacting effects of multiple stresses on
      growth and  physiological processes in northern forest trees. In RA Mickler; RABirsdey; J Horn (Eds.),
      Responses of northern US forests to environmental change (pp. 149-180). New York, NY: Springer-
      Verlag.
Isebrands, JG:  McDonald, EP: Kruger,  E: Hendrev,  G: Percy, K: Pregitzer, K: Sober, J: Karnoskv, DF. (2001).
      Growth responses of Populus tremuloides clones to interacting carbon dioxide and tropospheric ozone.
      Environ  Pollut 115: 359-371.
Islam, T: Gauderman, WJ: Berhane, K: McConnell,  R: Avol, E:  Peters, JM: Gilliland, FD.  (2007). The
      relationship between air pollution, lung function and asthma in adolescents. Thorax 62: 957-963.
Islam, T: McConnell, R: Gauderman, WJ: Avol, E: Peters, JM: Gilliland, FD. (2008). Ozone, oxidant defense
      genes and risk of asthma during adolescence.  Am J Respir Grit Care Med 177: 388-395.
      http://dx.doi.org/10.1164/rccm.200706-863OC.
Islam. T: Berhane. K: McConnell. R: Gauderman. WJ: Avol. E:  Peters. JM: Gilliland. FD.  (2009). Glutathione-S-
      transferase (GST) P1, GSTM1, exercise, ozone and asthma incidence in school children. Thorax 64:
      197-202. http://dx.doi.org/10.1136/thx.2008.099366.
Ito. A: Sudo. K: Akimoto. H: Sillman. S: Penner. JE. (2007a). Global modeling analysis of tropospheric ozone
      and its radiative forcing from biomass burning emissions in the twentieth century. J Geophys Res 112:
      D24307. http://dx.doi.org/10.1029/2007JD008745.
Ito, K: De Leon, SF:  Lippmann, M. (2005). Associations between ozone and daily mortality, analysis and meta-
      analysis. Epidemiology 16: 446-457.
Ito. K: Thurston. GD: Silverman. RA. (2007b). Characterization of PM2.5, gaseous pollutants, and
      meteorological interactions in the context of time-series health effects models. J Expo Sci Environ
      Epidemiol 17: S45-S60.
Iwasaki. T: Takahashi. M: Saito. H: Arito. H. (1998). Adaptation of extrapulmonary responses to ozone exposure
      in conscious rats. Ind Health 36: 57-60.
Jackson, DM: Heagle, AS: Eckel, RVW. (1999). Ovipositional response of tobacco hornworm moths
      (Lepidoptera:  Sphyngidae) to tobacco plants grown under elevated levels of ozone.  Environ Entomol  28:
      566-571.
Jacob, DJ: Horowitz, LW: Munger,  JW: Heikes, BG: Dickerson, RR: Artz, RS: Keene, WC. (1995). Seasonal
      transition from NOx- to hydrocarbon-limited conditions for ozone production over the eastern United
      States in September. J Geophys Res  100: 9315-9324.
Jacob, DJ. (1999). Introduction to atmospheric chemistry.  In. New Jersey:  Princeton University Press.
Jacobson. MZ.  (2002). Atmospheric pollution: history,  science,  and regulation.  In. New York: Cambridge
      University Press.
Draft - Do Not Cite or Quote                       11-39                                 September 2011

-------
Jacobson, MZ. (2005). Fundamentals of atmospheric modeling.  ln(2ed.). New York: Cambridge University
      Press.
Jacquemin, B: Kauffmann, F: Pin, I: Le Moual, N: Bousquet, J: Gormand, F: Just, J: Nadif, R: Pison, C: Vervloet,
      D: Kunzli, N: Siroux, V. (In Press) Air pollution and asthma control in the epidemiological study on the
      genetics and environment of asthma. J Epidemiol Community Health.
      http://dx.doi. org/10.1136/iech.2Q10.130229.
Jaegle, l_: Jacob, DJ:  Brune, WH: Wennberg, PO. (2001). Chemistry of HOx radicals in the upper troposphere.
      Atmos Environ 35: 469-489. http://dx.doi.org/10.1016/S1352-2310(00)00376-9.
Jaffe. D: Price. H: Parrish. D: Goldstein. A:  Harris. J. (2003). Increasing background ozone during spring on the
      west coast of North America. Geophys Res Lett 30:  1613. http://dx.doi.org/10.1029/2003GL017024.
Jaffe, D: Chand, D: Hafner, W: Westerling,  A: Spracklen, D. (2008). Influence of fires on O-3 concentrations in
      the western US. Environ Sci Technol 42: 5885-5891. http://dx.doi.org/10.1021/es800084k.
Jaffe, D. (2011). Relationship between surface and free tropospheric ozone in the Western U.S. Environ Sci
      Technol 45: 432-438. http://dx.doi.org/10.1021/es1028102.
Jakab. GJ: Hmieleski. RR. (1988). Reduction of influenza virus pathogenesis by exposure to 0.5 ppm ozone.  J
      Toxicol Environ Health 23: 455-472. http://dx.doi.org/10.1080/15287398809531128.
Jakab. GJ: Bassett. DJP (1990). Influenza virus infection, ozone exposure, and fibrogenesis. Am J Respir Crit
      Care Med 141: 1307-1315.
Jalaludin, B: Mannes, T: Morgan, G: Lincoln,  D: Sheppeard, V: Corbett, S. (2007). Impact of ambient air
      pollution on gestational age is modified by season in Sydney, Australia. Environ Health 6: 16.
Jalaludin, BB: Chev, T: O'Toole, Bl: Smith, WT: Capon, AG: Leeder, SR. (2000). Acute effects of low levels of
      ambient ozone on peak expiratory flow rate in a cohort of Australian children. Int J Epidemiol 29: 549-
      557. http://dx.doi.Org/10.1093/iie/29.3.549.
Jalaludin. BB: O'Toole. Bl: Leeder. SR. (2004). Acute effects of urban ambient air pollution on respiratory
      symptoms, asthma medication use,  and doctor visits for asthma in a cohort of Australian  children.
      Environ Res 95: 32-42. http://dx.doi.org/10.1016/S0013-9351 (03)00038-0.
James. P: Stohl. A: Forster. C: Eckhardt. S: Seibert. P: Frank. A. (2003). A 15-year climatology of stratosphere-
      troposphere exchange with a Lagrangian particle dispersion model: 2. Mean climate and seasonal
      variability. J Geophys Res 108: D12. http://dx.doi.org/10.1029/2002JD002639.
Janero, DR. (1990). Malondialdehyde and thiobarbituric acid-reactivity as diagnostic indices of lipid peroxidation
      and peroxidative tissue injury. Free Radic Biol Med 9: 515-540. http://dx.doi.org/10.1016/0891-
      5849(90)90131-2.
Jang. AS: Choi. IS: Yang. SY: Kim. YG: Lee. JH: Park. SW: Park. CS. (2005). Antioxidant responsiveness in
      BALB/c mice exposed to ozone.  Respiration 72: 79-84. http://dx.doi.org/10.1159/000083405.
Janic, B; Umstead, TM; Phelps, DS; Floras, J. (2005).  Modulatory effects of ozone on THP-1 cells in response
      to SP-A stimulation. Am J Physiol Lung Cell Mol Physiol 288: L317-L325.
      http://dx.doi.org/10.1152/aiplung.00125.2004.
Jansson, M:  Bergstrom, A, -K: Drakare, S: Blomgvist, P. (2001).  Nutrient limitation of bacterioplankton and
      phytoplankton in humic lakes in northern Sweden. Freshw Biol 14: 76-85.
Janzik, I: Preiskowski, S: Kneifel, H. (2005). Ozone has dramatic effects on the regulation of the prechorismate
      pathway in tobacco (Nicotiana tabacum L. cv. Bel W3). Planta 223: 20-27.
      http://dx.doi. org/10.1007/s00425-005-0060-8.
Jedlinska-Krakowska, M: Bomba, G: Jakubowski,  K: Rotkiewicz, T: Jana, B: Penkowskii, A. (2006). Impact of
      oxidative stress and  supplementation with vitamins E and C on testes morphology in rats. J  Reprod Dev
      52: 203-209.
Jenkins, Gl. (2009). Signal transduction in responses to UV-B radiation. Annu Rev Plant Biol 60: 407-431.
      http://dx.doi.org/10.1146/annurev.arplant.59.032607.092953.
Jerrett, M: Burnett, RT: Pope, CA, III:  Ito, K: Thurston,  G: Krewski, D: Shi, Y: Calle, E: Thun, M. (2009). Long-
      term ozone exposure and mortality.  N  Engl J Med 360: 1085-1095.
      http://dx.doi.org/10.1056/NEJMoa0803894.
Jiang. D: Liang. J: Fan. J: Yu. S: Chen.  S: Luo. Y:  Prestwich.  GD: Mascarenhas. MM: Garg. HG: Quinn. DA:
      Homer, RJ: Goldstein, PR: Bucala, R:  Lee, PJ:  Medzhitov, R: Noble, PW. (2005). Regulation of lung
      injury and repair by Toll-like receptors and hyaluronan. Nat Med  11: 1173-1179.
      http://dx.doi. org/10.1038/nm1315.
Jiang, LL: Zhang,  YH: Song, GX: Chen, GH: Chen, BH: Zhao, NQ: Kan, HP. (2007). A time series analysis of
      outdoor air pollution  and  preterm birth  in Shanghai, China. Biomed  Environ Sci 20:  426-431.
Jimenez, JL: Jayne, JT: Shi, Q:  Kolb, CE: Worsnop, PR: Yourshaw, I: Seinfeld, JH: Flagan, RC: Zhang, X:
      Smith, KA.  (2003). Ambient aerosol sampling using the Aerodyne Aerosol Mass Spectrometer. J
      Geophys Res 108: 8425.
Jo. WK:  Park. JH. (2005). Characteristics of roadside air pollution in Korean metropolitan city (Daegu) over last
      5 to 6 years: temporal variations, standard exceedances,  and dependence on meteorological conditions.
      Chemosphere 59: 1557-1573. http://dx.doi.Org/10.1016/i.chemosphere.2004.12.021.
Joad. JP: Kott. KS: Brie. JM. (1996). The local C-fiber  contribution to ozone-induced effects on the  isolated
      guinea pig  lung. Toxicol Appl Pharmacol 141: 561-567.
Draft - Do Not Cite or Quote                        11-40                                  September 2011

-------
Joad, JP: Kott, KS: Brie, JM: Peake, JL: Plopper, CG: Schelegle, ES: Gershwin, LJ: Pinkerton, KE. (2006).
      Structural and functional localization of airway effects from episodic exposure of infant monkeys to
      allergen and/or ozone. Toxicol Appl Pharmacol 214: 237-243.
      http://dx.doi.0rg/10.1016/i.taap.2005.12.012.
Joad. JP: Kott. KS: Brie. JM: Schelegle. ES: Gershwin. LJ: Plopper. CG: Peake. JL: Pinkerton. KE. (2008). The
      effects of inhaled corticosteroids on intrinsic responsiveness and histology of airways from infant
      monkeys exposed to house dust mite allergen and ozone. Toxicol Appl Pharmacol 226: 153-160.
      http://dx.doi. org/10.1016/i.taap.2007.09.005.
Johansson. E: Wesselkamper. SC:  Shertzer. HG: Leikauf. GD: Dalton. TP: Chen. Y. (2010). Glutathione
      deficient C57BL/6J mice are not sensitized to ozone-induced lung injury. Biochem Biophys Res Commun
      396: 407-412. http://dx.doi.Org/10.1016/i.bbrc.2010.04.105.
John, EM: Schwartz, GG:  Dreon, DM: Koo, J. (1999). Vitamin D and breast cancer risk: the NHANES I
      Epidemiologic Follow-up Study, 1971-1975 to 1992. Cancer Epidemiol Biomarkers Prev 8: 399-406.
John. EM: Schwartz. GG:  Koo. J: Van Den Berg. D: Ingles. SA. (2005). Sun exposure, vitamin D receptor gene
      polymorphisms, and risk of advanced prostate cancer. Cancer Res 65: 5470-5479.
      http://dx.doi. org/10.1158/0008-5472.CAN-04-3134.
Johnson. D: Jenkin. ME: Wirtz. K: Martin-Rivieio. M. (2004). Simulating the formation of secondary organic
      aerosol from the photooxidation of toluene. Environ Chem 1: 150-165.
Johnson, RM: Pregitzer, KS. (2007). Concentration of sugars, phenolic acids, and amino acids in forest soils
      exposed to elevated atmospheric CO2 and O3. Soil Biol Biochem 39:  3159-3166.
      http://dx.doi.0rg/10.1016/i.soilbio.2007.07.010.
Johnson, TR. (1995). Recent advances in the estimation of population exposure to mobile source pollutants. J
      Expo Sci Environ Epidemiol  5: 551-571.
Johnston, C; Holm, B; Gelein, R; Finkelstein, J. (2006). Postnatal lung development: Immediate-early gene
      responses post  ozone and LPS exposure. Inhal Toxicol 18: 875-883.
      http://dx.doi. org/10.1080/08958370600822466.
Johnston. RA: Mizgerd. JP: Shore.  SA. (2005a). CXCR2 is essential for maximal neutrophil recruitment and
      methacholine responsiveness after ozone exposure. Am J Physiol Lung  Cell Mol Physiol 288: L61-L67.
      http://dx.doi.org/10.1152/ajplung.OQ101.2004 00101.2004.
Johnston, RA: Schwartzman, IN: Flynt, l_: Shore, SA. (2005b). Role of interleukin-6 in murine airway responses
      to ozone. Am J Physiol Lung Cell Mol Physiol 288: L390-L397.
      http://dx.doi.org/10.1152/aiplung.00007.2004.
Johnston. RA: Mizgerd. JP: Flvnt. L: Quinton. LJ: Williams. ES: Shore. SA. (2007). Type I interleukin-1  receptor
      is required for pulmonary responses to subacute ozone exposure in mice. Am J Respir Cell Mol  Biol 37:
      477-484. http://dx.doi.org/10.1165/rcmb.2006-0315OC.
Johnston. RA: Theman. TA: Lu. FL: Terry. RD: Williams. ES: Shore. SA. (2008).  Diet-induced obesity causes
      innate  airway hyperresponsiveness to methacholine and enhances  ozone-induced pulmonary
      inflammation. J Appl Psychol 104: 1727-1735.  http://dx.doi.org/10.1152/iapplphvsiol.00075.2008.
Jokinen, IE: Markkula,  ES: Salo,  HM: Kuhn,  P: Nikoskelainen, S: Arts, MT: Browman, HI. (2008). Exposure to
      increased ambient  ultraviolet B radiation has negative effects on growth, condition and immune function
      of juvenile Atlantic salmon (Salmo salar). Photochem Photobiol 84:  1265-1271.
      http://dx.doi.0rg/10.1111/i.1751-1097.2008.00358.x.
Jones,  ME: Paine, TD.  (2006). Detecting changes in insect herbivore communities along a pollution gradient.
      Environ Pollut 143: 377-387. http://dx.doi.Org/10.1016/i.envpol.2005.12.013.
Jones.  MLM:  Hodges. G: Mills. G. (2010). Nitrogen mediates above-ground effects of ozone but not below-
      ground effects in a  rhizomatous sedge. Environ Pollut 158: 559-565.
      http://dx.doi.0rg/10.1016/i.envpol.2009.08.002.
Jones,  SL: Kittelson, J: Cowan, JO: Flannerv, EM: Hancox, RJ: McLachlan, CR: Taylor, DR. (2001). The
      predictive value of exhaled nitric oxide measurements in assessing  changes in asthma control. Am J
      Respir Grit Care Med 164: 738-743.
Jones.  TG:  Freeman. C: Lloyd. A: Mills. G. (2009). Impacts of elevated atmospheric ozone on peatland below-
      ground doc characteristics. Ecol Eng 35: 971-977.  http://dx.doi.Org/10.1016/i.ecoleng.2008.08.009.
Jonson. JE: Simpson. D: Fagerli. H: Solberg. S. (2005). Can we explain the trends in European ozone levels?
      Atmos Chem Phys 6: 51-66. http://dx.doi.org/10.5194/acp-6-51-2006.
Joo,  JH: Wang, SY: Chen, JG: Jones, AM: Fedoroff, NV.  (2005). Different signaling and cell death roles of
      heterotrimeric G protein alpha and beta subunits in the arabidopsis  oxidative stress response to ozone.
      Plant Cell 17: 957-970. http://dx.doi.org/10.1105/tpc.104.029603.
Jorres, R: Nowak, D: Magnussen, H: Speckin, P: Koschyk, S. (1996). The  effect of ozone exposure on  allergen
      responsiveness in subjects with asthma or rhinitis. Am J Respir Crit Care Med 153: 56-64.
Jorres. RA: Holz. O: Zachgo.  W: Timm. P: Koschvk. S: Mullen  B: Grimminger. F: Seeger. W: Kelly. FJ: Punster.
      C: Frischer, T: Lubec, G: Waschewski, M: Niendorf, A: Magnussen,  H. (2000). The effect of repeated
      ozone  exposures on inflammatory markers in bronchoalveolar lavage  fluid and mucosal biopsies. Am J
      Respir Crit Care Med 161: 1855-1861.
Draft - Do Not Cite or Quote                        11-41                                  September 2011

-------
Joshi, M: Shine, KP: Ponater, M: Stuber, N: Sausen, R: Li. (2003). A comparison of climate response to different
      radiative forcings in three general circulation models: Towards an improved metric of climate change.
      Clim Dynam 20: 843-854. http://dx.doi.org/10.1007/s00382-003-0305-9.
Joss, U: Graber, WK. (1996). Profiles and simulated exchange of H2O, O3, NO2 between the atmosphere and
      the HartX Scots pine  plantation. Theor Appl Climatol 53:  157-172.
Jovan. S: McCune. B. (2006). Using epiphytic macrolichen communities for biomonitoring ammonia in forests of
      the greater Sierra Nevada, California. Water Air Soil Pollut 170: 69-93.
Just. J: Segala. C: Sahraoui. F: Priol. G: Grimfeld. A: Neukirch.  F. (2002). Short-term health effects of particulate
      and photochemical air pollution in asthmatic children. Eur Respir J 20: 899-906.
      http://dx.doi. org/10.1183/09031936.02.00236902.
Kabel, JR: Ben-Jebria, A: Ultman, JS. (1994). Longitudinal distribution of ozone absorption in the lung:
      Comparison of nasal  and oral quiet breathing. J Appl Physiol 77: 2584-2592.
Kafoury, RM: Pryor, WA: Squadrito, GL: Salgo, MG: Zou, X: Friedman,  M. (1998). Lipid ozonation products
      activate phospholipases A2, C, and D. Toxicol Appl Pharmacol 150: 338-349.
Kaiekar. R: Pieczarka. EM: Smiley-Jewell. SM: Schelegle. ES: Fanucchi. MV:  Plopper. CG.  (2007). Early
      postnatal exposure to allergen and ozone leads to hyperinnervation of the pulmonary epithelium. Respir
      Physiol  Neurobiol 155: 55-63. http://dx.doi.Org/10.1016/i.resp.2006.03.002.
Kan. H: London. SJ: Chen. G: Zhang. Y: Song. G: Zhao. N: Jiang. L: Chen. B. (2008). Season, sex, age, and
      education as modifiers of the effects of outdoor air pollution on daily mortality in Shanghai, China: The
      Public Health and Air  Pollution in Asia (PAPA) Study. Environ Health Perspect 116: 1183-1188.
Kanerva, T: Palojarvi, A: Ramo, K: Ojanpera,  K: Esala, M: Manninen, S. (2006). A 3-year exposure to CO2 and
      O3 induced minor changes in soil N cycling in a meadow ecosystem. Plant Soil 286:  61-73.
      http://dx.doi.ora/10.1007/s11104-006-9026-2.
Kanerva. T: Regina. K: Ramo. K: Oianpera. K: Manninen. S. (2007). Fluxes of N2O, CH4 and CO2 in a meadow
      ecosystem  exposed to elevated ozone and carbon dioxide for three years. Environ Pollut 145: 818-828.
      http://dx.doi.0rg/10.1016/i.envpol.2006.03.055.
Kanerva. T: Paloiarvi. A: Ramo. K: Manninen. S.  (2008). Changes in soil microbial community structure under
      elevated tropospheric O3 and CO2. Soil Biol Biochem 40: 2502-2510.
      http://dx.doi.0rg/10.1016/i.soilbio.2008.06.007.
Kangasjarvi, J: Jaspers, P: Kollist, H. (2005).  Signalling and cell death in ozone-exposed plants. Plant Cell
      Environ 28: 1021-1036.
Kannan. S: Misra.  DP: Dvonch. T: Krishnakumar. A. (2006). Exposures to airborne particulate matter and
      adverse perinatal outcomes: A biologically plausible mechanistic framework for exploring potential effect
      modification by nutrition. Environ Health Perspect 114: 1636-1642.
Kanno. S: Yanagisawa. Y.  (1992). Passive ozone/oxidant sampler with coulometric determination using
      iodine/nylon-6 charge-transfer complex. Environ Sci Technol 26: 744-749.
      http://dx.doi.org/10.1021/es00028a012.
Kanofskv, JR: Sima, PP. (1995). Reactive absorption of ozone by aqueous biomolecule solutions: Implications
      for the role  of sulfhydryl compounds as targets for ozone. Arch Biochem Biophys 316: 52-62.
Kar, J: Fishman, J: Creilson, JK: Richter, A: Ziemke, J: Chandra, S. (2010). Are there urban  signatures  in the
      tropospheric ozone column products derived from satellite measurements? Atmos Chem Phys 10: 5213-
      5222. http://dx.doi.org/10.5194/acp-10-5213-2010.
Karl, F: Hatch,  G: Slade, R: Crissman, K: Simeonova, PP: Luster, M. (1997). Dietary  restriction mitigates ozone-
      induced lung inflammation in rats: A role for endogenous antioxidants. Am J Respir Cell Mol Biol 17:  740-
      747.
Karlsson, PE: Sellden, G: Plaijel, H. (2003). Establishing ozone critical levels II UNECE workshop report.
      Gothenburg, Sweden: IVL Swedish Environmental Institute.
Karlsson, PE: Uddling, J: Braun, S: Broadmeadow, M: Elvira, S: Gimeno, BS:  Le Thiec, D: Okansen, E:
      Vandermeiren, K: Wilkinson, M: Emberson, L. (2004). New critical levels for ozone effects on young trees
      based on AOT40 and simulated cumulative leaf uptake of ozone. Atmos Environ 38: 2283-2294.
Karner. AA: Eisinger. PS: Niemeier. DA. (2010). Near-roadway air quality: Synthesizing the findings from real-
      world data.  Environ Sci Technol 44: 5334-5344. http://dx.doi.org/10.1021/es1 OOOOSx.
Karnoskv. DF:  Gagnon. ZE:  Dickson. RE: Coleman. MD: Lee. EH: Isebrands. JG. (1996). Changes in growth,
      leaf abscission, biomass associated with seasonal tropospheric ozone  exposures of  Populus tremuloides
      clones and  seedlings. Can J For Res 26: 23-37.
Karnoskv, DF:  Mankovska, B: Percy, K:  Dickson, RE: Podila, GK: Sober, J: Noormets, A: Hendrev, G: Coleman,
      MD: Kubiske, M: Pregitzer, KS: Isebrands, JG. (1999). Effects of tropospheric ozone on trembling aspen
      and interaction with CO2: Results from an OS-gradient and a FACE experiment. Water Air Soil Pollut 116:
      311-322.
Draft - Do Not Cite or Quote                        11-42                                  September 2011

-------
Karnosky, DF: Zak, PR: Pregitzer, KS: Awmack, CS: Bockheim, JG: Dickson, RE: Hendrey, GR: Host, GE: King,
      JS: Kopper, BJ: Kruger, EL: Kubiske,  ME: Lindroth, Rl_: Mattson, WJ: McDonald, EP: Noormets, A:
      Oksanen, E: Parsons, WFJ: Percy, KE: Podila, GK: Riemenschneider, DE: Sharma, P: Thakur, R: Sober,
      A: Sober, J: Jones, WS: Anttonen, S:  Vapaavuori, E: Mankovska, B: Heilman, W: Isebrands, JG. (2003).
      Tropospheric O3 moderates responses of temperate hardwood forests to elevated CO2: A synthesis of
      molecular to ecosystem results from the Aspen FACE project. Funct Ecol 17: 289-304.
Karnosky, DF: Pregitzer, KS: Zak, PR: Kubiske, ME: Hendrey, GR: Weinstein, D: Nosal, M: Percy, KE. (2005).
      Scaling ozone responses of forest trees to the ecosystem level  in a changing climate. Plant Cell Environ
      28: 965-981. http://dx.doi.Org/10.1111/i.1365-3040.2005.01362.x.
Karr, C:  Lumlev, T: Schreuder, A: Davis, R: Larson, T: Ritz, B: Kaufman, J. (2007). Effects of subchronic and
      chronic exposure to ambient air pollutants on infant bronchiolitis. Am J Epidemiol 165: 553-560.
Kasibhatla,  P: Chameides, WL. (2000). Seasonal modeling of regional ozone pollution in the eastern United
      States. Geophys Res Lett 27: 1415-1418. http://dx.doi.org/10.1029/1999GL011147.
Kasurinen. A: Keinanen. MM: Kaipainen. S: Nilsson. LO: Vapaavuori. E: Kontro. MH: Holopainen. T. (2005).
      Below-ground responses of silver birch trees exposed to elevated CO2 and O3 levels during three
      growing seasons. Global Change Biol 11: 1167-1179. http://dx.doi.Org/10.1111/i.1365-2486.2005.00970.x.
Kasurinen. A: Riikonen. J: Oksanen. E: Vapaavuori. E: Holopainen. T. (2006). Chemical composition  and
      decomposition of silver birch leaf litter produced under elevated CO2 and O3.  Plant Soil 282: 261-280.
      http://dx.doi.org/10.1007/s11104-005-6026-6.
Kasurinen, A: Peltonen, PA: Julkunen-Tiitto,  R: Vapaavuori, E: Nuutinen, V: Holopainen, T: Holopainen, JK.
      (2007). Effects of elevated CO2 and O3 on leaf litter phenolics and subsequent performance of litter-
      feeding soil macrofauna. Plant Soil 292: 25-43. http://dx.doi.Org/10.1007/s11104-007-9199-3.
Kataoka. Y: Kiguchi. M: Williams. RS: Evans. PP. (2007). Violet light causes photodegradation of wood beyond
      the zone affected by ultraviolet radiation. Holzforschung und Holzverwertung 61: 23-27.
      http://dx.doi.org/10.1515/HF.2007.005.
Katre. A: Ballinger. C: Akhter.  H: Fanucchi. M: Kim. DK:  Postlethwait. E: Liu. RM. (2011). Increased transforming
      growth factor beta 1 expression mediates ozone-induced airway fibrosis  in mice. Inhal Toxicol  23: 486-
      494.  http://dx.doi.org/10.3109/08958378.2011.584919.
Kats, G: Thompson, CR: Kuby, WC. (1976).  Improved ventilation of open top greenhouses. J Air Pollut Control
      Assoc26: 1089-1090.
Kats, G: Olszyk, DM: Thompson, CR. (1985). Open top experimental chambers for trees. J Air Waste Manag
      Assoc35: 1298-1301.
Katsouvanni. K: Touloumi. G:  Samoli. E: Gryparis. A: Le Tertre. A: Monopolis. Y: Rossi. G: Zmirou. D: Ballester.
      F: Boumghar, A: Anderson, HR: Woityniak, B: Paldy, A: Braunstein, R: Pekkanen, J: Schindler, C:
      Schwartz. J. (2001). Confounding and effect modification in the short-term effects of ambient particles on
      total  mortality: Results from 29 European cities within the APHEA2 project. Epidemiology 12: 521-531.
Katsouvanni, K: Samet, JM: Anderson, HR: Atkinson, R:  Le Tertre, A: Medina, S: Samoli, E: Touloumi, G:
      Burnett, RT: Krewski, D: Ramsay, T: Dominici, F: Peng, RD: Schwartz, J: Zanobetti, A. (2009). Air
      pollution and health: A European and  North American approach (APHENA). (Research Report 142).
      Boston,  MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=327.
Kavlock. R:  Daston. G: Grabowski. CT. (1979). Studies on the developmental toxicity of ozone. I. Prenatal
      effects. Toxicol Appl Pharmacol 48: 19-28. http://dx.doi.org/10.1016/S0041-008X(79)80004-6.
Kavlock, RJ: Meyer, E: Grabowski, CT. (1980). Studies on the developmental toxicity of ozone: Postnatal
      effects. Toxicol Lett 5: 3-9. http://dx.doi.orq/10.1016/0378-4274(80)90141-1.
Kavnak. B:  Hu. Y:  Martin. RV: Russell. AG: Choi. Y: Wang. Y. (2008). The effect of lightning NOx production on
      surface ozone in the continental United States. Atmos Chem Phys  8:  5151-5159.
Kehrl, HR: Hazucha, MJ: Solic, JJ: Bromberg, PA. (1985). Responses  of subjects with chronic obstructive
      pulmonary disease after exposures to 0.3 ppm ozone. Am Rev Respir Dis 131: 719-724.
Kehrl, HR: Vincent, LM: Kowalskv, RJ: Horstman, DH: O'Neil, JJ: McCartney, WH: Bromberg, PA. (1987).
      Ozone exposure increases respiratory epithelial permeability in humans. Am Rev Respir Dis 135: 1124-
      1128.
Kehrl, HR: Peden, DB: Ball, BA: Folinsbee, LJ: Horstman, DH. (1999). Increased specific airway reactivity  of
      persons with mild allergic asthma after 7.6 hours of exposure to 0.16 ppm ozone. J Allergy Clin Immunol
      104:  1198-1204.
Keller, T: Hasler, R. (1984).  The influence of  a fall fumigation with ozone on the  stomatal behavior of spruce and
      fir. Oecologia 64: 284-286.  http://dx.doi.org/10.1007/BF00376884.
Kellomaki, S: Wang, K, -Y. (1997). Effects of elevated O3 and CO2 concentrations on photosynthesis and
      stomatal conductance in Scots pine. Plant Cell Environ 20: 995-1006.  http://dx.doi.Org/10.1111/i.1365-
      3040.1997.tb00676.x.
Kenvon. NJ: Van Per Vliet. A: Schock. BC: Okamoto. T: McGrew. GM:  Last.  JA. (2002). Susceptibility to ozone-
      induced acute lung injury in iNOS-deficient mice. Am J Physiol 282: L540-L545.
Kenvon. NJ: Last.  MS: Eiserich. JP: Morrissev. BM: Temple.  LM:  Last.  JA. (2006). Differentiation of the roles of
      NO from airway epithelium and inflammatory cells in ozone-induced lung inflammation. Toxicol Appl
      Pharmacol 215: 250-259. http://dx.doi.Org/10.1016/i.taap.2006.03.005.
Draft - Do Not Cite or Quote                       11-43                                 September 2011

-------
Kermani, S: Ben-Jebria, A: Ultman, JS. (2006). Kinetics of ozone reaction with uric acid, ascorbic acid, and
      glutathione at physiologically relevant conditions. Arch Biochem Biophys 451: 8-16.
      http://dx.doi.0rg/10.1016/i.abb.2006.04.015.
Kerner, R: Winkler, J: Dupuy, J: Jurgensen, M: Lindermayr, C: Ernst, D:  Muller-starck, G. (2011). Changes in the
      proteome of juvenile European beech following three years exposure to free-air elevated ozone. iForest
      4: 69-76. http://dx.doi.ora/10.3832/ifor0570-004.
Keutgen, AJ: Noga, G:  Pawelzik,  E. (2005). Cultivar-specific impairment of strawberry growth, photosynthesis,
      carbohydrate and nitrogen accumulation by ozone. Environ Exp Bot 53: 271-280.
      http://dx.doi.ora/10.1016/i.envexpbot.2004.04.003.
Keutgen, N: Keutgen, AJ: Janssens, MJJ. (2008).  Sweet potato [Ipomoea batatas (L.) Lam.] cultivated as tuber
      or leafy vegetable supplier as affected by elevated tropospheric ozone. J Agric Food Chem 56: 6686-
      6690. http://dx.doi.org/10.1021/if8006272.
Kharitonov,  SA: Barnes, PJ. (2000). Clinical aspects of exhaled nitric oxide. Eur Respir J 16: 781-792.
Khatri. SB: Holguin. FC: Ryan. PB: Mannino.  D: Erzurum. SC: league. WG. (2009). Association of ambient
      ozone exposure with airway inflammation and allergy in adults with asthma. J Asthma 46: 777-785.
      http://dx.doi. org/10.1080/02770900902779284.
Kiehl. JT: Schneider. TL: Portmann. RW: Solomon. S. (1999). Climate forcing due to tropospheric and
      stratospheric ozone. J Geophys Res 104: 31239-31254.  http://dx.doi.org/10.1029/1999JD900991.
Kim, CS: Alexis, NE: Rappold, AG: Kehrl, H: Hazucha,  MJ: Lay, JC: Schmitt, MT: Case, M: Devlin, RB: Peden,
      DB: Diaz-Sanchez,  D. (2011). Lung function and inflammatory responses in healthy young adults
      exposed to 0.06 ppm ozone for 6.6 hours. Am J Respir Grit Care Med  183: 1215-1221.
      http://dx.doi.org/10.1164/rccm.201011-1813OC.
Kim. MY: Cho.  MY. (2009a). Toxicity and carcinogenicity of ozone in combination with 4-(N-methyl-N-
      nitrosamino)-1-(3-pyridyl)-1-butanone and dibutyl phthalate in B6C3F1 mice for 16 and 32 weeks.
      Biomed  Environ Sci 22: 216-222.
Kim. MY: Cho.  MY. (2009b). Tumorigenesis in B6C3F1 mice exposed to ozone in combination with 4-(N-methyl-
      N-nitrosamino)-1-(3-pyridyl)-1-butanone and dietary dibutyl phthalate. Toxicol Ind Health 25: 189-195.
      http://dx.doi. org/10.1177/0748233709106185.
Kimlin, MG: Wong, JCF: Parisi, AV. (1998). Simultaneous comparison of the personal UV exposure of two
      human groups at different altitudes. Health  Phys 74: 429-434.
King, GM. (1999). Characteristics  and significance of atmospheric carbon monoxide consumption by soils.
      Chemosphere 1: 53-63.
King. JS: Pregitzer. KS: Zak. PR: Sober. J: Isebrands. JG: Dickson. RE: Hendrev. GR: Karnoskv. DF. (2001).
      Fine-root biomass and fluxes of soil carbon in young stands of paper birch and trembling aspen as
      affected by elevated atmospheric CO2 and  tropospheric O3. Oecologia 128: 237-250.
King. JS: Kubiske. ME: Pregitzer. KS: Hendrev. GR: McDonald. EP: Giardina. CP: Quinn. VS: Karnoskv. DF.
      (2005). Tropospheric O3 compromises net  primary production in  young stands of trembling aspen, paper
      birch and sugar maple in response to elevated atmospheric CO2. New Phytol 168: 623-635.
      http://dx.doi.0rg/10.1111/i.1469-8137.2005.01557.x.
Kinney, PL:  Thurston, GD: Raizenne, M. (1996). The effects of ambient  ozone on lung function in children: A
      reanalysis of six summer camp studies. Environ Health Perspect 104:  170-174.
Kinnev. PL:  Lippmann.  M. (2000).  Respiratory effects of seasonal exposures to ozone and particles. Arch
      Environ  Occup Health 55: 210-216.
Kitao. M: Low. M: Heerdt. C: Grams. TEE:  Haberle. KH: Matvssek. R. (2009). Effects of chronic elevated ozone
      exposure on gas exchange responses of adult beech trees (Fagus sylvatica) as related to the within-
      canopy light gradient. Environ Pollut 157: 537-544. http://dx.doi.Org/10.1016/i.envpol.2008.09.016.
Kleeberger,  SR: Seiden, JE: Levitt, RC: Zhang, L,  -Y (1993). Mast cells modulate acute ozone-induced
      inflammation of the murine  lung. Am J  Respir Grit Care Med 148: 1284-1291.
Kleeberger,  SR: Levitt,  RC: Zhang, L, -Y: Longphre, M: Harkema, J: Jedlicka, A: Eleff, SM: DiSilvestre, D:
      Holrovd. KJ. (1997). Linkage analysis of susceptibility to ozone-induced lung inflammation in  inbred mice.
      Nat Genet 17: 475-478.
Kleeberger,  SR: Reddy, S: Zhang, L, -Y: Jedlicka,  AE. (2000). Genetic susceptibility to ozone-induced lung
      hyperpermeability: Role of toll-like receptor 4. Am J Respir Cell Mol Biol 22: 620-627.
Kleeberger.  SR: Reddv. SP: Zhang. L. -Y: Cho. H.  -Y: Jedlicka. AE. (2001). Toll-like receptor 4 mediates ozone-
      induced murine  lung hyperpermeability via  inducible nitric oxide synthase. Am J Physiol 280:  L326-L333.
Kleffmann, J: Lorzer, JC: Wiesen,  P: Kern, C: Trick, S: Volkamer, R: Rodenas, M: Wirtz, K. (2006).
      Intercomparison of the DOAS and LOPAP techniques for the detection of nitrous acid (HONO). Atmos
      Environ  40: 3640-3652.
Kleffmann. J: Wiesen. P. (2008). Technical note: Quantification of interferences of wet chemical  HONO LOPAP
      measurements under simulated polar conditions. Atmos Chem Phys 8: 6813-6822.
Kleindienst, TE: Hudgens,  EE: Smith, DF:  McElroy, FF:  Bufalini, JJ. (1993). Comparison of chemiluminescence
      and ultraviolet ozone monitor responses in  the presence of humidity and photochemical pollutants. Air
      Waste 43:  213-222.
Draft - Do Not Cite or Quote                       11-44                                 September 2011

-------
Klepeis, NE. (1999). An introduction to the indirect exposure assessment approach: Modeling human exposure
      using microenvironmental measurements and the recent National Human Activity Pattern Survey.
      Environ Health Perspect 107: 365-374.
Klepeis, NE: Nelson, WC: Ott, WR: Robinson, JP: Tsang, AM: Switzer, P: Behar, JV: Hern, SC:  Engelmann,
      WH. (2001). The National Human Activity Pattern Survey (NHAPS): A resource for assessing exposure to
      environmental pollutants. J Expo Sci Environ Epidemiol 11: 231-252.
Klestadt, D: Laval-Gillv, P: Foucaud, l_: Falla, J.  (2005). Influences of ozone  exposure  upon macrophage
      responsivity to N-formyl-methionyl-leucyl-phenylalanine: Mobility and  metabolic changes. Toxicol In Vitro
      19:  199-206. http://dx.doi.orq/10.1016/i.tiv.2004.08.004.
Kline, LJ: Davis, DP: Skellv, JM: Savage, JE: Ferdinand, J. (2008). Ozone sensitivity of 28 plant selections
      exposed to ozone under controlled conditions. Northeast Nat 15: 57-66. http://dx.doi.org/10.1656/1092-
      6194(2008)15f57:OSOPSE12.0.CO:2.
Kline, LJ: Davis, DP: Skellv, JM: Decoteau, DR. (2009). Variation in ozone sensitivity within Indian hemp and
      common milkweed selections from the Midwest. Northeast Nat 16: 307-313.
      http://dx.doi.orq/10.1656/045.016.0210.
Kloster, S: Dentener, F: Feichter, J: Raes, F: Lohmann, U: Roeckner, E: Fischer-bruns, I. (2009). A GCM study
      of future climate response to aerosol pollution reductions. Clim Dynam 34: 1177-1194.
      http://dx.doi. org/10.1007/s00382-009-0573-0.
Ko, FWS: Tarn, W: Wong, TW:  Lai, CKW. (2007). Effects of air pollution on asthma hospitalization rates in
      different age groups in Hong Kong. Clin Exp Allergy 37: 1312-1319.
Kodavanti, UP: Costa,  PL: Dreher, Kl_: Crissman, K: Hatch, GE. (1995). Ozone-induced tissue injury and
      changes in antioxidant homeostasis in normal and ascorbate-deficient guinea pigs.  Biochem Pharmacol
      50:  243-251. http://dx.doi.org/10.1016/0006-2952(95)00122-G.
Kodavanti. UP: Thomas.  R: Ledbetter. AD: Schladweiler. MC: Shannahan. JH: Wallenborn. JG:  Lund. AK:
      Campen, MJ: Butler, EO: Gottipolu, RR: Nvska, A: Richards, JE: Andrews, D: Jaskot, RH: McKee, J:
      Kotha. SR: Patel. RB: Parianandi. NL. (2011). Vascular and cardiac impairments in rats Inhaling ozone
      and diesel exhaust particles. Environ Health Perspect 119: 312-318.
      http://dx.doi. org/10.1289/ehp. 1002386.
Koenig, JQ: Covert, PS: Marshall, SG: Van Belle, G: Pierson, WE. (1987). The effects of ozone and nitrogen
      dioxide on pulmonary function in healthy and in asthmatic adolescents. Am J Respir Crit Care Med 136:
      1152-1157.
Kohut. R. (2007). Assessing the risk of foliar injury from ozone on vegetation in parks in the US National Park
      Service's Vital Signs Network. Environ Pollut 149: 348-357.
      http://dx.doi.0rg/10.1016/i.envpol.2007.04.022.
Kolb. TE: Fredericksen. TS: Steiner. KG: Skellv. JM. (1997).  Issues in scaling tree size and age responses to
      ozone: A review [Review]. Environ Pollut 98: 195-208. http://dx.doi.org/10.1016/S0269-7491 (97)00132-2.
Kollist, T: Moldau, H: Rasulov, B: Oja, V: Ramma, H: Huve, K: Jaspers, P: Kangasjarvi, J: Kollist, H. (2007). A
      novel device detects a rapid ozone-induced transient stomatal closure in intact Arabidopsis and its
      absence in abi2 mutant. Physiol Plant 129: 796-803. http://dx.doi.Org/10.1111/i.1399-3054.2006.00851.x.
Kooter, IM: Pennings, Jl_: Fokkens, PH: Leseman, PL: Boere, AJ: Gerlofs-Nijland, ME: Cassee, FR: Schalk, JA:
      Orzechowski. TJ: Schaap. MM: Breit. TM: Dormans. JA: van Oostrom. CT: de Vries. A: van Steeg. H.
      (2007). Ozone induces clear cellular and molecular responses in the  mouse lung independently of the
      transcription-coupled repair status. JAppI Physiol 102: 1185-1192.
      http://dx.doi.org/10.1152/iapplphvsiol.00796.2006.
Koren. HS: Devlin. RB: Graham. DE: Mann. R: McGee. MP: Horstman. DH:  Kozumbo. WJ: Becker. S: House.
      DE: McDonnell,  WF: Bromberg, PA. (1989). Ozone-induced inflammation in the lower airways of human
      subjects. Am J Respir Crit Care Med 139: 407-415.
Koronakis, PS: Sfantos, GK: Paliatsos, AG: Kaldellis, JK: Garofalakis, JE: Koronaki, IP. (2002).  Interrelations of
      UV-global/global/diffuse solar irradiance components and UV-global attenuation on air pollution episode
      days in Athens,  Greece. Atmos Environ 36: 3173-3181.
Korrick. SA: Neas. LM: Dockerv. DW: Gold. PR: Allen.  GA: Hill. LB: Kimball.  KD: Rosner. BA: Speizer. FE.
      (1998). Effects of ozone and other pollutants  on the pulmonary function of adult hikers. Environ Health
      Perspect 106: 93-99. http://dx.doi.org/10.1289/ehp.9810693.
Kostikas. K: Papatheodorou. G: Ganas. K: Psathakis. K: Panagou. P: Loukides. S. (2002). pH in expired breath
      condensate of patients with inflammatory airway diseases. Am J Respir Crit Care Med 165: 1364-1370.
Kostka-Rick, R: Hahn,  HU. (2005). Biomonitoring using tobacco Bel W3 provides supplemental information  for
      risk assessment of vegetation injury due to ozone. Gefahrstoffe Reinhaltung Der Luft 65: 485-491.
Koutrakis,  P: Wolfson,  JM: Bunyaviroch, A: Froehlich, SE: Hirano, K: Mulik, JD. (1993). Measurement of
      ambient ozone using a nitrite-coated filter. Anal  Chem 65: 209-214.
Kozovits. AR: Matvssek. R: Blaschke. H: Gottlein. A: Grams. TEE. (2005). Competition increasingly dominates
      the responsiveness of juvenile beech and spruce to elevated CO2 and/or O3 concentrations throughout
      two subsequent growing seasons. Global Change Biol 11: 1387-1401. http://dx.doi.Org/10.1111/i. 1365-
      2486.2005.00993.x.
Draft - Do Not Cite or Quote                       11-45                                 September 2011

-------
Kreit, JW: Gross, KB: Moore, TB: Lorenzen, TJ: D'Arcy, J: Eschenbacher, WL. (1989). Ozone-induced changes
      in pulmonary function and bronchial responsiveness in asthmatics. J Appl Physiol 66: 217-222.
Krewski, D: Jerrett, M: Burnett, RT: Ma, R: Hughes, E: Shi, Y: Turner, MC: 3rd, PA: Thurston, G: Calle, EE:
      Thun, MJ. (2009). Extended follow-up and spatial analysis of the American Cancer Society study linking
      particulate air pollution and mortality. (Report Nr. 140). Cambridge, MA: Health Effects Institute.
Krishna. MT: Sprinqall. D: Meng. Q. -H: Withers. N: Macleod. D: Biscione. G: Frew. A: Polak. J: Holgate. S.
      (1997). Effects of ozone on epithelium and sensory nerves in the bronchial mucosa of healthy  humans.
      Am J Respir Grit  Care Med 156: 943-950.
Krupa. SV:  Grunhage. L: Jager. H. -J: Nosal. M: Manning. WJ: Legge. AH: Hanewald. K. (1995). Ambient ozone
      (O3) and adverse crop response: A unified view of cause and effect.  Environ Pollut 87: 119-126.
      http://dx.doi.org/10.1016/80269-7491(99)80014-1.
Krupa, SV:  Nosal, M:  Peterson, PL. (2001). Use of passive ozone O3 samplers in vegetation effects
      assessment. Environ Pollut 112: 303-309.
Kubiske. ME: Quinn. VS: Heilman. WE: McDonald. EP: Marquardt. PE: Teclaw. RM: Friend. AL: Karnoskev. PR
      (2006). Interannual climatic variation mediates elevated CO2 and O3 effects on forest growth.  Global
      Change Biol 12: 1054-1068. http://dx.doi.Org/10.1111/i.1365-2486.2006.01152.x.
Kubiske. ME: Quinn. VS: Marguardt. PE: Karnoskv. PR (2007). Effects of elevated atmospheric CO2  and/or O3
      on intra- and interspecific competitive ability of aspen. Plant Biol (Stuttg) 9: 342-355.
      http://dx.doi. org/10.1055/S-2006-924760.
Kulle, TJ: Sauder, LR: Kerr, HP: Rarrell, BP: Bermel, MS: Smith, PM. (1982). Puration of pulmonary function
      adaptation to ozone in humans. Am Ind Hyg Assoc J 43: 832-837.
Kulle, TJ: Sauder, LR: Hebel, JR: Chatham, MP. (1985). Ozone response relationships in healthy nonsmokers.
      Am Rev Respir Pis 132: 36-41.
Kumarathasan. P: Blais.  E: Goegan. P: Yagminas.  A: Guenette. J: Adamson. IY: Crapo.  JP:  Mason. RJ:
      Vincent, R. (2005). 90-day repeated inhalation exposure of surfactant Protein-C/tumor necrosis factor-
      alpha, (SP-C/TNR-alpha) transgenic mice to air pollutants. Int J Toxicol 24: 59-67.
Kuo. HW: Lai. JS: Lee. MC: Tai. RC. (2002). Respiratory effects of air pollutants among  asthmatics in central
      Taiwan. Arch Environ Occup Health 57: 194-200.
Lacis, AA: Wuebbles, PJ: Logan, JA. (1990). Radiative forcing of climate by changes in the vertical distribution
      of ozone. J Geophys Res 95: 9971-9981. http://dx.doi.org/10.1029/JP095iP07p09971.
Laffray, X:  Rose, C: Garrec, JP. (2007). Estimation  of ozone concentration in a valley of the alps mountains
      based on bel-w3  tobacco leaf injury. Water Air Soil Pollut 186: 29-42. http://dx.doi.org/10.1007/s11270-
      007-9460-7.
Lagorio, S:  Rorastiere, R: Pistelli, R: lavarone,  I: Michelozzi, P: Rano, V: Marconi, A: Ziemacki, G: Ostro, BP.
      (2006). Air pollution and lung function among susceptible adult subjects: A panel study. Environ Health 5:
      11. http://dx.doi.org/10.1186/1476-069X-5-11.
Lam, Y: Ru, J. (2010). Corrigendum to "A novel downscaling technigue for the linkage of global and regional air
      guality modeling" published in Atmos. Chem. Phys., 9, 9169-9185, 2009. Atmos Chem Phys 10: 4013-
      4031. http://dx.doi.org/10.5194/acp-10-4013-2010.
Lamargue, JR: Hess, P:  Emmons, L: Buja, L: Washington, W: Granier, C.  (2005). Tropospheric ozone evolution
      between 1890 and 1990. J Geophys Res 110: P08304. http://dx.doi.org/10.1029/2004JP005537.
Lamargue. JR: Bond. TC: Evring. V: Granier. C: Hell. A: Klimont. Z: Lee. P: Liousse. C: Mieville. A: Owen.  B:
      Schultz, MG: Shindell, P: Smith, SJ: Stehfest, E: Van Aardenne, J: Cooper,  OR: Kainuma, M: Mahowald,
      N: McConnell. JR: Naik. V: Riahi. K: van Vuuren. PP. (2010). Historical (1850-2000) gridded
      anthropogenic and biomass burning emissions of reactive gases and aerosols: Methodology and
      application. Atmos Chem Phys Piscuss 10: 4963-5019. http://dx.doi.org/10.5194/acpd-10-4963-2010.
Lampl, M: Jeanty,  P.  (2003). Timing is everything: A reconsideration of fetal growth velocity patterns identifies
      the  importance of individual and sex differences. Am J Hum Biol 15: 667-680.
      http://dx.doi.org/10.1002/aihb.10204.
Langebartels. C: Kerner. K: Leonard!. S: Schraudner. M: Trost. M: Heller. W: Sandermann. H. Jr.  (1991).
      Biochemical plant responses to ozone:  I. Pifferential induction of polyamine and ethylene biosynthesis in
      tobacco. J Plant Physiol 95:  882-889. http://dx.doi.Org/10.1104/pp.95.3.882.
Langford. AO: Aikin. KC: Eubank. CS: Williams. EJ. (2009). Stratospheric contribution to high surface ozone in
      Colorado during springtime.  Geophys Res Lett 36: L12801. http://dx.doi.org/10.1029/2009gl038367.
Langstaff, JE. (2007). Analysis of uncertainty in ozone population exposure modeling [technical memorandum].
      Research Triangle Park,  NC: U.S. Environmental  Protection Agency, Office  of Air Quality Planning and
      Standards.
Lanki, T: Pekkanen, J: Aalto, P: Elosua, R: Berglind, N: P'lppoliti, P: Kulmala, M: Nyberg, R:  Peters, A: Picciotto,
      S: Salomaa. V: Sunver.  J: Tiittanen. P: Von Klot. S: Rorastiere. R. (2006). Associations of traffic-related air
      pollutants with hospitalisation for first acute myocardial infarction: The HEAPSS study. Occup Environ
      Med 63: 844-851.
Draft - Po Not Cite or Quote                        11-46                                  September 2011

-------
Larrieu, S: Jusot, JF: Blanchard, M: Prouvost, H: Declercq, C: Fabre, P: Pascal, L: Le Tertre, A: Wagner, V:
      Riviere, S: Chardon, B: Borelli, D: Cassadou, S: Eilstein, D: Lefranc, A. (2007). Short term effects of air
      pollution on hospitalizations for cardiovascular diseases in eight French cities: The PSAS program. Sci
      Total Environ 387: 105-112.
Larsen. ST: Matsubara. S: McConville. G: Poulsen. SS: Gelfand. EW. (2010). Ozone increases airway
      hyperreactivity and mucus hyperproduction in mice previously exposed to allergen. J Toxicol Environ
      Health A 73: 738-747. http://dx.doi.Org/10.1080/15287391003614034.
Larson. JL: Zak. PR: Sinsabaugh. RL. (2002).  Extracellular enzyme activity  beneath temperate trees growing
      under elevated carbon dioxide and ozone. Soil Sci Soc Am J 66: 1848-1856.
Larson, SD: Schelegle, ES: Walbv, WF: Gershwin, LJ: Fanuccihi,  MV: Evans, MJ: Joad, JP: Tarkington, BK:
      Hyde, DM: Plopper, CG. (2004). Postnatal remodeling of the neural components of the epithelial-
      mesenchymal trophic unit in the proximal  airways of infant rhesus monkeys exposed to ozone and
      allergen. Toxicol Appl Pharmacol 194: 211-220.
Laskin. PL: Pending. KJ: Punjabi.  CJ: del Valle. MR: Laskin. JD. (1994). Pulmonary and hepatic effects of
      inhaled ozone in rats. Environ Health Perspect 10: 61-64.
Laskin, PL: Heck, DE: Laskin, JD. (1998). Role of inflammatory cytokines and nitric oxide in hepatic and
      pulmonary toxicity. Toxicol Lett 102-103: 289-293.
Laskin. PL: Laskin. JD. (2001). Role of macrophages and inflammatory mediators in chemically induced toxicity.
      Toxicology 160: 111-118.
Laskin, JD: Heck, DE: Laskin, PL. (1996). Nitric oxide production in the lung and liver following inhalation of the
      pulmonary irritant ozone. Adv Exp Med  Biol 387:  141-146.
Last, JA: Reiser, KM: Tyler, WS: Rucker, RB. (1984). Long-term consequences of exposure to ozone. I. Lung
      collagen content. Toxicol Appl Pharmacol  72: 111-118.
Last. JA: Warren. PL: Pecguet-Goad. E: Witschi. H. (1987).  Modification by  ozone of lung tumor development in
      mice. J Natl Cancer Inst 78: 149-154.
Last. JA: Gelzleichter. TR: Harkema. J: Hawk.  S. (1994). Consequences of prolonged inhalation of ozone on
      Fischer-344/N rats: Collaborative studies. Part I: Content and cross-linking of lung collagen.
Last, JA: Gohil, K: Mathrani, VC: Kenyon, NJ. (2005). Systemic responses to inhaled ozone in mice: cachexia
      and down-regulation of liver xenobiotic metabolizing genes. Toxicol Appl Pharmacol 208: 117-126.
      http://dx.doi.0rg/10.1016/i.taap.2005.02.001.
Latzin, P: Roosli, M: Huss, A: Kuehni, CE: Frev, U. (2009). Air pollution during pregnancy and lung function in
      newborns: A birth cohort study. Eur Respir J 33: 594-603.
Lawlor. DW. (1998). Plant responses to global  change: Temperature and drought stress. In LJ De Kok; I Stulen
      (Eds.), Responses of plant  metabolism to air pollution and global change. Leiden, The Netherlands:
      Backhuys Publishers.
Lawrence. SO: Simpson-Haidaris. PJ. (2004).  Regulated de novo biosynthesis of fibrinogen in extrahepatic
      epithelial cells in response to inflammation. Thromb Haemostasis 92: 234-243.
      http://dx.doi.org/10.1160/TH04-01-0024.
Lav, JC: Alexis, NE: Kleeberger, SR: Roubey, RA: Harris, BD: Bromberg, PA: Hazucha, MJ: Devlin, RB: Peden,
      DB. (2007). Ozone enhances markers of innate immunity and antigen presentation on airway monocytes
      in healthy individuals. J Allergy Clin Immunol 120: 719-722. http://dx.doi.Org/10.1016/i.iaci.2007.05.005.
Leakey. ADB: Bernacchi. CJ: Ort. PR: Long. SP  (2006). Long-term growth of soybean at elevated CO2 does
      not cause acclimation of stomatal conductance under fully open-air conditions. Plant Cell Environ 29:
      1794-1800. http://dx.doi.0rg/10.1111/i.1365-3040.2006.01556.x.
Lee. EH: Tingev. DT: Hogsett. WE. (1987). Selection of the best exposure-response model using various 7-hour
      ozone  exposure statistics. Research Triangle Park, NC: U.S. Environmental Protection Agency.
Lee, EH: Tingev, DT: Hogsett, WE. (1988a). Evaluation of ozone-exposure indices for relating exposure to plant
      production and for estimating agricultural  losses. (EPA/600/3-88/039). Washington, DC:  U.S.
      Environmental Protection Agency.
Lee. EH: Tingev. DT: Hogsett. WE. (1988b). Evaluation of ozone exposure indices in exposure-response
      modeling. Environ Pollut 53: 43-62. http://dx.doi.org/10.1016/0269-7491 (88)90024-3.
Lee, EH: Tingev, DT: Hogsett, WE. (1989). Interrelation of experimental exposure and ambient air quality data
      for comparison of ozone exposure indices and estimating agricultural losses. (EPA/600/3-89/047).
      Corvallis, OR: U.S. Environmental Protection Agency.
Lee, EH: Hogsett, WE: Tingev,  DT. (1994). Attainment and effects issues regarding alternative secondary ozone
      air quality standards. J Environ Qual 23: 1129-1140.
      http://dx.doi.org/10.2134/ieg1994.00472425002300060002x.
Lee, EH: Hogsett, WE. (1996).  Methodology for calculating inputs for ozone secondary standard benefits
      anaylsis:  Part II. Research Triangle Park,  NC: U.S. Environmental Protection Agency.
Lee. EH: Hogsett. WE. (1999).  Role of concentrations and time of day in developing ozone exposure indices for
      a secondary standard. J Air Waste Manag Assoc 49: 669-681.
Lee. EH: Tingev. DT: Hogsett. WE: Laurence. JA. (2003a). History of tropospheric ozone for the San Bernardino
      Mountains of southern California,  1963-1999. Atmos Environ 37: 2705-2717.
      http://dx.doi. org/10.1016/51352-2310(03)00203-6.
Draft - Do Not Cite or Quote                        11-47                                  September 2011

-------
Lee, EH: Tingey, DT: Waschmann, RS: Phillips, PL: Olszyk, DM: Johnson, MG: Hogsett, WE. (2009a).
      Seasonal and long-term effects of CO2 and O-3 on water loss in ponderosa pine and their interaction
      with climate and soil moisture. Tree Physiol 29: 1381-1393. http://dx.doi.org/10.1093/treephys/tpp071.
Lee, IM: Tsai, SS: Ho, CK: Chiu, HF: Yang, CY. (2007). Air pollution and hospital admissions for congestive
      heart failure in a tropical city:  Kaohsiung, Taiwan. Inhal Toxicol 19: 899-904. http://dx.doi.org/781182105
      [pii]10.1080/08958370701479406.
Lee, IM: Tsai, SS: Ho, CK: Chiu, HF: Wu, TN: Yang, CY. (2008a). Air pollution and hospital admissions for
      congestive heart failure: Are there potentially sensitive groups? Environ Res 108: 348-353.
      http://dx.doi. org/10.1016/i.envres.2008.07.024.
Lee, J: Kim, KH: Kim, YJ. (2008b). Application of a long-path differential optical absorption spectrometer (LP-
      DOAS) on the measurements of NO(2), SO(2), O(3), and HNO(2) in Gwangju, Korea. J  Environ Manage
      86: 750-759.
Lee, JT: Kim, H: Cho, YS: Hong, YC: Ha, EH: Park, H. (2003b). Air pollution and hospital admissions for
      ischemic heart diseases among individuals 64+ years of age residing in Seoul, Korea. Arch Environ
      Health 58: 617-623.
Lee, JT: Son, JY: Kim,  H: Kim, SY (2006). Effect of air pollution on asthma-related hospital admissions for
      children by socioeconomic status associated with area of residence. Arch Environ Occup Health 61: 123-
      130.
Lee, K: Parkhurst, WJ:  Xue, J: Ozkaynak, H: Neuberg, D: Spengler, JD. (2004a). Outdoor/indoor/personal
      ozone exposures of children in Nashville, Tennessee. J Air Waste Manag Assoc 54: 352-359.
Lee, S: Yun,  SC. (2006). The  ozone  stress transcriptome of pepper (Capsicum annuum L). Molecules and Cells
      21: 197-205.
Lee. SJ: Haiat.  S: Steer. PJ: Filippi. V. (2008c). A time-series analysis of any short-term effects  of meteorological
      and air pollution factors on preterm births in London, UK.  Environ Res 106: 185-194.
Lee, WS:  Chevone, Bl: Seller, JR. (1990). Growth and gas exchange of loblolly pine seedlings  as influenced by
      drought  and air pollutants. Water Air Soil Pollut 51: 105-116. http://dx.doi.ora/10.1007/BF00211508.
Lee. Y -L: Lin. Y -C: Lee. Y -C: Wang. J. -Y: Hsiue. T. -R: Quo. YL. (2004b). Glutathione S-transferase P1 gene
      polymorphism and air pollution as interactive risk factors for childhood asthma. Clin Exp Allergy 34: 1707-
      1713.
Lee, YL: McConnell, R: Berhane, K:  Gilliland, FD. (2009b). Ambient ozone modifies the effect of tumor necrosis
      factor G-308A on bronchitic symptoms among children with asthma. Allergy 64: 1342-1348.
      http://dx.doi. orq/10.1111/i.1398-9995.2009.02014.x.
Leech. JA: Nelson. WC: Burnett. RT: Aaron. S: Raizenne. ME. (2002). It's about time: A comparison of
      Canadian and American time-activity patterns. J Expo Anal Environ Epidemiol 12: 427-432.
      http://dx.doi.org/10.1038/si.iea.7500244.
Lefkowitz. ES: Garland. CF. (1994). Sunlight, vitamin D, and ovarian cancer mortality rates in US women. Int J
      Epidemiol 23: 1133-1136.
Lefohn, AS: Benedict, HM.  (1982). Development of a mathematical index that describes ozone concentration,
      frequency and duration. Atmos Environ 16: 2529-2532. http://dx.doi.org/10.1016/0004-6981 (82)90145-7.
Lefohn, AS: Laurence, JA: Kohut, RJ. (1988). A comparison of indices that describe the relationship between
      exposure to ozone and reduction in the yield of agricultural crops. Atmos Environ 22: 1229-1240.
      http://dx.doi. org/10.1016/0004-6981(88)90353-8.
Lefohn, AS: Jackson, W:  Shadwick,  PS: Knudsen, HP. (1997). Effect of surface ozone exposures on vegetation
      grown in the southern Appalachian Mountains: Identification of possible areas of concern. Atmos Environ
      31: 1695-1708. http://dx.doi.org/10.1016/51352-2310(96)00258-0.
Lefohn, AS: Shadwick,  PS. (2000). Differences in trending estimates in the United States using several ozone
      metrics.  In Proceedings of the 93rd Air & Waste Management Association Annual Conference and
      Exhibition (pp. AS 1d-645).  Pittsburgh, PA: Air & Waste Management Association.
Lefohn, AS: Wernli,  H: Shadwick, D: Limbach,  S: Oltmans, SJ: Shapiro, M. (2011). The importance of
      stratospheric-tropospheric transport in affecting surface ozone concentrations in the western and
      northern tier of the United States. Atmos Environ 45: 4845-4857.
      http://dx.doi.0rg/10.1016/i.atmosenv.2011.06.014.
Legge. AH: Grunhage.  L: Nosal. M: Jager. H. -J: Krupa. SV. (1995). Ambient ozone and adverse crop  response:
      An evaluation of North American and European data as they relate to exposure indices and critical levels.
      J Appl Bot Food Qual 69: 192-205.
Legro, RS: Sauer, MV:  Mottla, Gl_: Richter, KS: Li, X: Dodson, WC: Liao, D. (2010). Effect of air quality on
      assisted human reproduction. Hum Reprod 25: 1317-1324. http://dx.doi.org/10.1093/humrep/deg021.
Leitao, L:  Delacote, E: Dizengremel, P: Le Thiec, D: Biolley, JP (2007a). Assessment of the impact of increasing
      concentrations of ozone on photosynthetic components of maize (Zea mays L.), a C-4 plant. Environ
      Pollut 146: 5-8. http://dx.doi.Org/10.1016/i.envpol.2006.05.019.
Leitao, L:  Maoret, JJ: Biollev,  JP. (2007b). Changes in PEP carboxylase, rubisco and rubisco activase mRNA
      levels from maize (Zea mays) exposed to a chronic ozone stress. Biol Res 40: 137-153.
      http://dx.doi.org/10.4067/S0716-97602007000200005.
Draft - Do Not Cite or Quote                        11-48                                  September 2011

-------
Leitao, L: Bethenod, O: Biolley, JP. (2007c). The impact of ozone on juvenile maize (Zea mays L.) plant
      photosynthesis: Effects on vegetative biomass, pigmentation, and carboxylases (PEPc and Rubisco).
      Plant Biol (Stuttg) 9: 478-488. http://dx.doi.org/10.1055/s-2007-964942.
Lelieveld, J: van Aardenne, J: Fischer, H: de Reus, M: Williams, J: Winkler, P. (2004). Increasing ozone over the
      Atlantic Ocean. Science 304: 1483-1487. http://dx.doi.orq/10.1126/science. 1096777.
Lenoble. J. (1993). Atmospheric radiative transfer. In. Hampton, VA: A. Deepak Publishing.
Leonard, RJ: Charpied, Gl_: Faddis, B. (1991). Effects of ambient inhaled ozone on vocal fold mucosa in Bonnet
      monkeys. J Voice 5: 304-309. http://dx.doi.org/10.1016/30892-1997(05)80060-8.
Lesser. VM: Rawlings. JO: Spruill. SE: Somerville. MC. (1990). Ozone effects on agricultural crops: Statistical
      methodologies and estimated dose-response relationships. Crop Sci 30: 148-155.
Leston, AR: Ollinson, WM: Spicer, CW: Satola, J. (2005). Potential interference bias in ozone standard
      compliance monitoring. J Air Waste Manag Assoc 55: 1464-1472.
Leuning, R: Unsworth, MH: Neumann, HN: King, KM. (1979). Ozone fluxes to tobacco and soil under field
      conditions. Atmos Environ 13: 1155-1163. http://dx.doi.org/10.1016/0004-6981 (79)90039-8.
Levine. JS: Pinto. JP.  (1998). The production of CO by biomass burning. In MAK Khalil; JP Pinto; MJ Shearer
      (Eds.),  Atmospheric carbon monoxide and its environmental effects: Proceedings of the international
      conference; December 1997; Portland, Oregon (pp. 251-256). Portland,  OR: U.S. Environmental
      Protection Agency, Office of Research and  Development.
Levy, H, II: Schwaarzkopf, MD; Horowitz, L; Ramaswamv, V; Findell, KL. (2008). Strong sensitivity of late 21st
      century climate to projected changes in short-lived air pollutants. J Geophys Res  113: D06102.
      http://dx.doi. org/10.1029/2007JD009176.
Levy, Jl; Chemerynski, SM; Sarnat, JA. (2005). Ozone exposure and mortality, an empiric Bayes
      metaregression analysis. Epidemiology 16: 458-468.
Lewis. JS: Ditchkoff. SS: Lin. JC: Muntifering. RB: Chappelka. AH. (2006). Nutritive quality of big bluestem
      (Andropogon gerardii) and eastern gamagrass (Tripsacum dactyloides) exposed to tropospheric ozone.
      Rangeland Ecol Manag 59: 267-274.
Lewis. TC: Robins. TG: Dvonch. JT: Keeler. GJ: Yip. FY: Mentz. GB: Lin. X: Parker. EA:  Israel. BA: Gonzalez. L:
      Hill, Y. (2005). Air pollution-associated changes in lung function among asthmatic children in Detroit.
      Environ Health Perspect 113: 1068-1075.
Li, H; Romieu, I; Sienra-Monge,  JJ; Ramirez-Aguilar,  M; Estela del Rio-Navarro, B; Kistner, EO;  Gjessing, HK;
      Irma del Carmen, LS;  Chiu, GY; London, SJ. (2006a). Genetic polymorphisms in arginase I and II and
      childhood asthma and atopy. J Allergy Clin  Immunol 117: 119-126.
Li. PH: Mane.  SP: Sioson. AA: Robinet. CV: Heath. LS: Bohnert. HJ: Grene. R. (2006b).  Effects  of chronic
      ozone exposure on gene expression in Arabidopsis thaliana ecotypes and in Thellungielia halophila.
      Plant Cell Environ 29: 854-868. http://dx.doi.Org/10.1111/i.1365-3040.2005.01465.x.
Li. Y: Lee. SR: Wu. CY. (2006c). UV-absorption-based measurements of ozone and mercury: An investigation
      on their mutual interferences. Aerosol Air Qual Res 6: 418-429.
Li, YF; Gauderman, WJ; Avol, E; Dubeau,  L; Gilliland, FD. (2006d). Associations of tumor necrosis factor G-
      308Awith childhood asthma and wheezing. Am J Respir Grit  Care Med 173: 970-976.
      http://dx.doi.org/10.1164/rccm.200508-1256OC.
Li. YF: Gauderman. WJ: Conti. DV: Lin. PC: Avol.  E: Gilliland. FD. (2008). Glutathione S-Transferase P1,
      Maternal Smoking,  and Asthma in Children: A Haplotype-Based Analysis. Environ Health  Perspect 116:
      409-415.
Li. Z: Potts. EN:  Piantadosi. CA: Foster. WM: Hollingsworth. JW. (2010). Hyaluronan fragments contribute to the
      ozone-primed immune response to lipopolysaccharide. J Immunol 185: 6891-6898.
      http://dx.doi. org/10.4049/iimmunol. 1000283.
Liao, D; Duan, Y; Whitsel,  EA; Zheng, Z, -J; Heiss, G; Chinchilli, VM; Lin, H, -M. (2004a). Association of higher
      levels of ambient criteria pollutants with impaired cardiac autonomic control: a population-based study.
      Am J Epidemiol 159: 768-777.
Liao. D: Heiss. G: Chinchilli. VM: Duan. Y: Folsom. AR: Lin. HM: Salomaa.  V. (2005). Association of criteria
      pollutants with  plasma hemostatic/inflammatory markers: A population-based study. J Expo Sci Environ
      Epidemiol 15: 319-328.
Liao. H: Seinfeld. JH: Adams. PJ: Micklev. LJ. (2004b). Global radiative forcing  of coupled tropospheric ozone
      and aerosols in a unified general circulation model. J Geophys Res 109: D16207.
      http://dx.doi. org/10.1029/2003JD004456.
Liard,  R; Zureik, M; Le Moullec, Y; Soussan, D; Glorian, M; Grimfeld, A; Neukirch, F. (1999). Use of personal
      passive samplers for measurement of NO2, NO, and O3 levels in panel studies. Environ Res 81: 339-
      348.
Lim. Y: Phung. AD:  Corbacho. AM: Aung. HH: Maioli.  E: Reznick. AZ: Cross. CE: Davis.  PA: Valacchi.  G. (2006).
      Modulation of cutaneous wound healing by ozone: Differences between  young and aged  mice.  Toxicol
      Lett 160: 127-134. http://dx.doi.Org/10.1016/i.toxlet.2005.06.013.
Lin. CA: Pereira. LAA: Nishioka. DC: Conceicao. QMS: Graga. ALF: Saldiva. PHN. (2004a). Air pollution and
      neonatal deaths in Sao Paulo, Brazil.  Braz J Med  Biol Res 37: 765-770.
Draft - Do Not Cite or Quote                       11-49                                 September 2011

-------
Lin, CM: Li, C, -Y: Yang, G, -Y: Mao, IF. (2004b). Association between maternal exposure to elevated ambient
      sulfur dioxide during pregnancy and term low birth weight. Environ Res 96: 41-50.
Lin, JC: Nosal, M: Muntifering, RB: Krupa, SV. (2007). Alfalfa nutritive quality for ruminant livestock as
      influenced by ambient air quality in west-central Alberta. Environ Pollut 149: 99-103.
      http://dx.doi.0rg/10.1016/i.envpol.2006.12.009.
Lin. M: Stieb. DM: Chen. Y. (2005). Coarse particulate matter and hospitalization for respiratory infections in
      children younger than  15 years in Toronto: A case-crossover analysis. Pediatrics 116: 235-240.
Lin. S: Fitzgerald. E: Hwang.  SA: Munsie. JP: Stark. A. (1999). Asthma hospitalization rates and socioeconomic
      status in New York State (1987-1993). J Asthma 36: 239-251.
Lin, S: Bell, EM: Liu, W: Walker, RJ: Kim, NK: Hwang, SA. (2008a). Ambient ozone concentration and hospital
      admissions due to childhood respiratory diseases in New York State, 1991-2001. Environ Res 108: 42-
      47.  http://dx.doi.0rg/10.1016/i.envres.2008.06.007.
Lin, S: Liu, X: Le, LH: Hwang, SA. (2008b). Chronic exposure to ambient ozone and asthma hospital
      admissions among children. Environ Health Perspect 116: 1725-1730.
      http://dx.doi.org/10.1289/ehp.11184.
Lin, S. (2010). E-mail correspondence from Shao Lin to Dennis Kotchmar dated  December 20, 2010
Linares. C: Diaz.  J.  (2010).  Short-term effect of concentrations of fine particulate matter on hospital admissions
      due to cardiovascular and respiratory causes among the over-75  age group in Madrid,  Spain. Public
      Health 124: 28-36. http://dx.doi.Org/10.1016/i.puhe.2009.11.007.
Lindelof, B: Sigurgeirsson, B: Gabel, H: Stern, RS. (2000).  Incidence of skin cancer in 5356 patients following
      organ transplantation.  Br J Dermatol 143: 513-519.
Lindroth, RL. (2010). Impacts of elevated atmospheric CO2 and O3 on forests: Phytochemistry, trophic
      interactions,  and ecosystem dynamics. J Chem Ecol 36: 21-Feb.  http://dx.doi.org/10.1007/s10886-009-
      9731-4.
Linn, WS:  Buckley,  RD: Spier, CE: Blessey, RL: Jones, MP: Fischer, DA: Hackney, JD. (1978). Health effects of
      ozone exposure in asthmatics. Am Rev Respir Dis 117: 835-843.
Linn. WS:  Medwav.  DA: Anzar. UT: Valencia.  LM: Spier. CE: FS-D. T: Fischer. DA: Hackney. JD. (1982a).
      Persistence of adaptation  to ozone in volunteers exposed repeatedly for six weeks. Am Rev Respir Dis
      125: 491-495.
Linn, WS:  Fischer, DA: Medwav,  DA: Anzar, UT:  Spier, CE: Valencia,  LM: Venet, TG: Hackney, JD. (1982b).
      Short-term respiratory effects of 0.12 ppm ozone exposure in volunteers with chronic obstructive
      pulmonary disease. Am Rev Respir Dis 125: 658-663.
Linn. WS: Shamoo. DA: Venet. TG: Spier. CE: Valencia.  LM: Anzar. UT:  Hackney. JD. (1983). Response to
      ozone in volunteers with chronic obstructive pulmonary disease. Arch Environ Occup Health 38: 278-283.
Linn. WS: Avol. EL:  Shamoo. DA: Spier. CE: Valencia. LM:  Venet. TG: Fischer. DA: Hackney. JD. (1986). A
      dose-response study of healthy, heavily exercising men exposed  to ozone at concentrations near the
      ambient air quality standard. Toxicol Ind Health 2: 99-112.
Linn, WS: Shamoo, DA: Anderson, KR: Peng, R, -C: Avol, EL: Hackney,  JD: Gong, H, Jr. (1996). Short-term air
      pollution exposures and responses in Los Angeles area schoolchildren. J Expo Sci Environ Epidemiol 6:
      449-472.
Linn. WS:  Rappaport. EB: Berhane. KT: Bastain. TM: Avol. EL: Gilliland. FD. (2009).  Exhaled  nitric oxide in a
      population-based study of southern California  schoolchildren.  Respir Res 10: 28.
Lipfert, FW: Perry, HM, Jr: Miller, JP: Baty, JD: Wyzga, RE: Carmody, SE. (2000). The Washington University-
      EPRI veterans' cohort mortality study:  Preliminary results. Inhal Toxicol 4: 41-73.
Lipfert. FW: Perry. HM. Jr: Miller. JP: Batv. JD: Wvzga. RE: Carmodv. SE. (2003). Air pollution, blood pressure,
      and their long-term associations with mortality. Inhal Toxicol 15: 493-512.
Lipfert, FW: Baty, JD: Miller, JP: Wyzga,  RE. (2006a). PM2.5 constituents and related air quality variables as
      predictors of survival in a cohort of U.S. military veterans. Inhal Toxicol 18: 645-657.
Lipfert, FW: Wvzga, RE: Baty, JD: Miller, JP. (2006b). Traffic density as a surrogate measure of environmental
      exposures in studies of air pollution health effects: Long-term mortality in a cohort of US veterans. Atmos
      Environ 40: 154-169.
Lisabeth, LD: Escobar, JD:  Dvonch, JT: Sanchez, BN: Majersik, JJ: Brown, PL: Smith, MA:  Morgenstern, LB.
      (2008). Ambient air pollution and risk for ischemic stroke and transient ischemic attack. Ann Neurol 64:
      53-59. http://dx.doi.org/10.1002/ana.21403.
Liu, L: Leech, JA: Urch, RB: Silverman, FS. (1997). In vivo  salicylate  hyroxylation: A potential biomarker for
      assessing acute ozone exposure and effects in humans. Am J Respir Grit Care Med 156: 1405-1412.
Liu, L: Leech, JA: Urch, RB: Poon, R: Zimmerman, B: Kubay, JM: Silverman, FS. (1999). A comparison of
      biomarkers of ozone exposure in human plasma,  nasal lavage, and sputum. Inhal Toxicol 11: 657-674.
Liu. L: King. J: Giardina. C.  (2005). Effects of elevated concentrations of atmospheric CO2 and tropospheric O3
      on leaf litter production and chemistry  in trembling aspen and paper birch communities. Tree Physiol 25:
      1511-1522.
Liu. L: King. JS: Giardina. CP (2007a). Effects of elevated atmospheric CO2 and tropospheric O3 on nutrient
      dynamics: Decomposition of leaf litter in termbling aspen and paper birch communities.  Plant  Soil 299:
      65-82.
Draft - Do Not Cite or Quote                        11-50                                  September 2011

-------
Liu, L: Poon, R: Chen, l_: Frescura, AM: Montuschi, P: Ciabattoni, G: Wheeler, A: Dales, R. (2009a). Acute
      effects of air pollution on pulmonary function, airway inflammation, and oxidative stress in asthmatic
      children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehp11813.
Liu, L, -JS: Koutrakis, P: Leech, J: Broder,  I. (1995). Assessment of ozone exposures in the greater metropolitan
      Toronto area. J Air Waste Manag Assoc 45: 223-234.
Liu. LL: King. JS: Giardina. CP: Booker. FL. (2009b). The influence of chemistry, production and community
      composition on leaf litter decomposition under elevated atmospheric CO2 and tropospheric O-3 in a
      northern hardwood ecosystem. Ecosystems 12: 401-416. http://dx.doi.ora/10.1007/s10021-009-9231-v.
Liu. S: Krewski. D: Shi. Y: Chen. Y: Burnett. R. (2007b). Association between maternal exposure to ambient air
      pollutants during pregnancy and fetal growth restriction. J Expo Sci Environ Epidemiol 17: 426-432.
Liu, X: Chance, K: Sioris, CE: Kurosu, TP:  Spurr,  RJD: Martin, RV: Fu, T, -M: Logan, JA: Jacob, DJ: Palmer, PI:
      Newchurch, MJ: Megretskaia,  IA: Chatfield, RB. (2006). First directly retrieved global distribution of
      tropospheric column ozone from GOME: Comparison with the GEOS-CHEM model. J  Geophys Res 111:
      D02308. http://dx.doi.org/10.1029/2005JD006564.
Liu. XH: Head. DA: Stoelinaa.  MT (2001).  Numerical simulation of new particle formation over the northwest
      Atlantic using the MM5  mesoscale model coupled with sulfur chemistry. J Geophys Res 106: 9697-9715.
Loats. KV: Rebbeck. J. (1999). Interactive effects of ozone and elevated carbon dioxide on the growth and
      physiology of black cherry, green ash, and yellow poplar seedlings. Environ Pollut 106: 237-248.
      http://dx.doi.org/10.1016/S0269-7491(99)00069-X.
Logan, JA: Megretskaia, IA: Miller, AJ: Tiao, GC: Choi, D: Zhang, L: Stolarski, RS: Labow, GJ: Hollandsworth,
      SM: Bodeker, GE: Claude, H: De Muer, D:  Kerr, JB: Tarasick, DW: Oltmans, SJ: Johnson, B: Schmidlin,
      F: Staehelin, J: Viatte, P: Uchino, O. (1999). Trends in the vertical distribution of ozone: A comparison of
      two analyses of ozonesonde data. J Geophys Res 104: 26373-26399.
      http://dx.doi. org/10.1029/1999JD900300.
London, SJ. (2007). Gene-air pollution interactions in asthma. Proc Am Thorac Soc 4: 217-220.
      http://dx.doi.orq/10.1513/pats.200701-031AW.
Long. NC: Suh. J:  Morrow. JD: Schiestl. RH: Krishna Murthv. GG: Brain. JD: Frei. B. (2001). Ozone causes lipid
      peroxidation but little antioxidant depletion in exercising and nonexercising hamsters. J Appl Physiol 91:
      1694-1700.
Long, SP (1991). Modification  of the response of photosynthetic productivity to rising temperature by
      atmospheric CO2 concentrations: Has its importance  been underestimated? Plant Cell Environ 14: 729-
      739. http://dx.doi.org/10.1111/i. 1365-3040.1991.tb01439.x.
Longphre. M: Zhang. L. -Y: Harkema. JR: Kleeberger. SR. (1999).  Ozone-induced pulmonary inflammation and
      epithelial proliferation are partially mediated by PAR J Appl Physiol 86: 341-349.
Longstreth. J: de Gruiil. FR: Kripke. ML: Abseck. S: Arnold. F: Slaper.  HI: Velders. G:  Takizawa. Y: van der
      Leun. JC. (1998). Health risks. J Photochem Photobiol B 46: 20-39.
Longstreth, JD: Gruiil, D: Kripke, ML:  Takizawa, Y: van der Leun, JC. (1995). Effects of increased  solar
      ultraviolet radiation on human health. Ambio 24: 153-165.
Loomis, D: Castillejos, M: Gold, PR: McDonnell, W: Borja-Aburto, VH. (1999). Air pollution and infant mortality in
      Mexico City. Epidemiology 10: 118-123.
Lopez-Aparicio. S: Smolfk. J: Maskova. L: Souckova. M: Gr0ntoft. T: Ondrackova. L:  Stankiewicz. J. (2011).
      Relationship of indoor and outdoor air pollutants in a naturally ventilated historical building  envelope.
      Build Environ 46: 1460-1468. http://dx.doi.Org/10.1016/i.buildenv.2011.01.013.
Lopez. I: Sanchez.  I: Bizarro. P: Acevedo. S: Ustarroz. M: Fortoul. T. (2008). Ultrastructural alterations during
      embryonic rats' lung development caused by ozone. J Electron Microsc (Tokyo) 57:  19-23.
      http://dx.doi. org/10.1093/imicro/dfm033.
Loranger, Gl: Pregitzer, KS: King, JS. (2004). Elevated CO2  and O3t concentrations differentially  affect selected
      groups of the fauna in temperate forest soils. Soil Biol Biochem 36: 1521-1524.
Lorenzini, G: Mali,  C. (1995). Analysis of vertical ozone and nitrogen oxides profiles in a Prunus cerasifera
      canopy. Int J Biometeorol 39: 1-4. http://dx.doi.org/10.1007/BF01320885.
Loreto. F: Velikova. V. (2001).  Isoprene produced by leaves protects the photosynthetic apparatus against
      ozone damage, quenches ozone products, and reduces lipid peroxidation of cellular membranes.  Plant
      Physiol  127: 1781-1787. http://dx.doi.org/10.1104/pp.010497.
Loreto. F: Fares. S. (2007). Is ozone flux inside leaves only a damage indicator? Clues from volatile isoprenoid
      studies. Plant Physiol 143: 1096-1100. http://dx.doi.org/10.1104/pp.106.091892.
Low, M: Herbinger, K: Nunn, AJ: Haberle, KH: Leuchner, M: Heerdt, C: Werner, H: Wipfler, P:  Pretzsch, H:
      Tausz,  M: Matyssek, R.  (2006). Extraordinary drought of 2003 overrules ozone impact on adult beech
      trees (Fagus sylvatica).  Trees Struct Funct 20: 539-548. http://dx.doi.org/10.1007/s00468-006-0069-z.
Lova. WM: Pregitzer. KS: Karberg.  NJ: King. JS: Giardina. CP. (2003). Reduction of soil carbon formation by
      tropospheric ozone under elevated carbon dioxide. Nature 425: 705-707.
Lu,  FL: Johnston, RA: Flynt, L: Theman, TA: Terry, RD: Schwartzman, IN:  Lee, A: Shore, SA.  (2006).  Increased
      pulmonary responses to acute ozone exposure in obese db/db mice. Am J Physiol Lung Cell Mol  Physiol
      290: L856-L865. http://dx.doi.org/10.1152/aiplung.00386.2005.
Draft - Do Not Cite or Quote                        11-51                                  September 2011

-------
Lu, R: Turco, RP: Jacobson, MZ. (1997). An integrated air pollution modeling system for urban and regional
      scales: 1 Structure and performance. J Geophys Res 102: 6063-6079.
Ludwikow, A: Gallois, P: Sadowski, J. (2004). Ozone-induced oxidative stress response in Arabidopsis:
      Transcription profiling by microarray approach.  Cell Mol Biol Lett 9: 829-842.
Ludwikow. A: Sadowski. J. (2008). Gene networks in plant ozone stress response and tolerance. J Integr Plant
      Biol 50:  1256-1267. http://dx.doi.Org/10.1111/i.1744-7909.2008.00738.x.
Ludwikow, A: Kierzek, D: Gallois, P: Zeef, L: Sadowski, J. (2009). Gene expression profiling of ozone-treated
      Arabidopsis abiltd insertional mutant: Protein phosphatase 2C ABU  modulates biosynthesis ratio of ABA
      and ethylene. Planta 230: 1003-1017. http://dx.doi.org/10.1007/s00425-009-1001-8.
Luecken, DJ: Phillips, S: Sarwar, G: Jang, C. (2008). Effects of using the  CB05 vs. SAPRC99 vs. CB4 chemical
      mechanism on model predictions: Ozone and gas-phase  photochemical precursor concentrations. Atmos
      Environ  42: 5805-5820.
Luo, Y: Reynolds, JF. (1999). Validity of extrapolating field CO2 experiments to predict carbon sequestration in
      natural ecosystems. Ecology 80: 1568-1583.
Luo. Y. (2001).  Transient ecosystem response to free-air CO2 enrichment (FACE): Experimental evidence and
      methods of analysis. New Phytol 152: 3-8.
Lutter. R: Wolz. C. (1997). UV-B screening by tropospheric ozone: Implications for the national ambient air
      quality standard. Environ Sci Technol 31: 142A-146A.
Lyons, TM: Barnes, JD. (1998).  Influence of plant age  on  ozone  resistance in Plantago major. New Phytol 138:
      83-89. http://dx.doi.0rg/10.1046/i.1469-8137.1998.00879.x.
M, I: M. (International Cooperative Programme on Modelling and Mapping). (2004). Mapping critical levels for
      vegetation.  In Manual on methodologies and criteria for modelling  and mapping critical loads and levels,
      and air pollution effects, risks and trends.
Madronich. S: De Gruiil. F (1993). Skin cancer and  UV radiation [Letter/Response]. Nature 366: 23.
      http://dx.doi. org/10.1038/366023aO.
Madronich. S: Wagner. M: Groth. P. (2011). Influence of tropospheric ozone control  on exposure to ultraviolet
      radiation at the surface. Environ Sci Technol 45: 6919-6923. http://dx.doi.org/10.1021/es200701g.
Maggio, A: Chiaranda, FQ: Cefariello, R: Fagnano, M.  (2009). Responses to ozone pollution of alfalfa exposed
      to increasing salinity levels. Environ Pollut  157: 1445-1452.
      http://dx.doi.0rg/10.1016/i.envpol.2008.09.013.
Mahajan, AS: Shaw, M: Oetjen,  H: Hornsbv, KE: Carpenter,  LJ: Kaleschke, L: Tian-Kunze, X: Lee, JD:  Moller,
      SJ: Edwards. P. (2010). Evidence of reactive iodine chemistry in the Arctic boundary layer. J Geophys
      Res 115: D20303. http://dx.doi.Org/10.1029/2009JD013665.
Mahalingam, R: Shah, N: Scrymgeour, A: Fedoroff, N.  (2005). Temporal evolution of the Arabidopsis oxidative
      stress response. Plant Mol Biol  57: 709-730.  http://dx.doi.org/10.1007/s11103-005-2860-4.
Mahalingam. R: Jambunathan. N: Gunian. SK:  Faustin. E: Weng. H: Avoubi. P. (2006). Analysis of oxidative
      signalling induced by ozone in Arabidopsis thaliana. Plant Cell Environ 29: 1357-1371.
      http://dx.doi.0rg/10.1111/i.1365-3040.2006.01516.x.
Maier-Maercker, U. (1998). Predisposition of trees to drought stress by ozone. Tree Physiol 19: 71-78.
Mandl, RH: Weinstein, LH: McCune, DC: Kevenv, M. (1973). A cylindrical, open-top chamber for the exposure of
      plants to air pollutants in the field. J Environ Qual 2: 371-376.
Mandl. RH: Laurence. JA: Kohut. RJ. (1989). Development and testing of open-top  chambers for exposing
      large,  perennial plants to air pollutants. J Environ Qual 18: 534-540.
      http://dx.doi. org/10.2134/ieg1989.00472425001800040026x.
Maniar-Hew. K: Postlethwait. EM: Fanucchi. MV:  Ballinger. CA: Evans. MJ:  Harkema. JR:  Carey. SA:
      McDonald, RJ: Bartolucci, AA: Miller, LA. (2011). Postnatal episodic ozone results in persistent
      attenuation of pulmonary and peripheral blood responses to LPS challenge. Am J Physiol Lung Cell Mol
      Physiol 300: L462-L471.  http://dx.doi.org/10.1152/aiplung.00254.2010.
Mafikovska, B: Percy, KE: Karnoskv, DF. (2005). Impacts  of greenhouse gases on epicuticular waxes of
      Populus tremuloides Michx.: Results from an open-air exposure and a natural O3 gradient. Environ Pollut
      137: 580-586. http://dx.doi.Org/10.1016/i.envpol.2005.01.043.
Mann, JK: Balmes, JR: Bruckner, TA: Mortimer, KM: Margolis, HG: Pratt,  B: Hammond, SK: Lurmann, F: Tager,
      IB. (2010). Short-term effects of air pollution on wheeze in asthmatic children in Fresno, California.
      Environ  Health Perspect 118: 1497-1502. http://dx.doi.org/10.1289/ehp.0901292.
Mannes,  T: Jalaludin, B: Morgan, G: Lincoln, D: Sheppeard, V: Corbett, S. (2005). Impact  of ambient air
      pollution on birth weight in Sydney, Australia. Occup Environ Med 62: 524-530.
Manning, WJ: Krupa, SV. (1992). Experimental methodology for studying the effects of ozone on crops and
      trees.  In AS Lefohn (Ed.), Surface level ozone exposures and their effects on vegetation (pp. 93-156).
      Chelsea, Ml: Lewis Publishers.
Manning. WJ. (2003). Detecting plant effects is necessary to give biological significance to ambient ozone
      monitoring data and predictive ozone standards. Environ  Pollut 126:  375-379.
Mansfield. CA:  Johnson. FR:  Van Houtven. GL. (2006). The missing piece: Valuing averting behavior for
      children's ozone exposures. Resource Energ Econ 28: 215-228.
      http://dx.doi.0rg/10.1016/i.reseneeco.2006.02.002.
Draft - Do Not Cite or Quote                        11-52                                  September 2011

-------
Manzer, R; Wang, J; Nishina, K; McConville, G; Mason, RJ. (2006). Alveolar epithelial cells secrete chemokines
      in response to IL-1beta and lipopolysaccharide but not to ozone. Am J Respir Cell Mol Biol 34: 158-166.
      http://dx.doi.org/10.1165/rcmb.2005-0205OC.
Mapp, CE: Fryer, AA: De Marzo, N: Pozzato, V: Padoan, M: Boschetto, P: Strange,  RC:  Hemmingsen, A:
      Spiteri. MA. (2002). Glutathione S-transferase GSTP1  is a susceptibility gene for  occupational asthma
      induced by isocyanates. J Allergy Clin Immunol 109: 867-872.
      http://dx.doi. org/10.1067/mai.20Q2.123234.
Mar. TF: Koenig. JQ. (2009). Relationship between visits to emergency departments for  asthma and  ozone
      exposure in greater Seattle, Washington. Ann Allergy Asthma Immunol  103: 474-479.
Marenco, A: Gouget, H: Nedelec, P:  Pages, J,  -P: Karcher, F. (1994). Evidence of a long-term increase in
      tropospheric ozone from Pic du Midi data series: Consequences: Positive radiative forcing. J Geophys
      Res 99: 16617-16632. http://dx.doi.org/10.1029/94JD00021.
Mariassy, AT:  Sielczak, MW: McCray, MM: Abraham, WM: Wanner, A. (1989). Effects of ozone on lamb tracheal
      mucosa: Quantitative glycoconjugate histochemistry. Am J Pathol 135: 871-879.
Mariassv. AT:  Abraham. WM: Phipps. RJ: Sielczak. MW: Wanner. A. (1990). Effect of ozone on the postnatal
      development of lamb  mucociliary apparatus. J Appl Physiol 68: 2504-2510.
Markkula. E: Salo. HM: Rikalainen. K: Jokinen. IE. (2009). Long-term UVB irradiation affects the immune
      functions of carp (Cyprinus carpio) and  rainbow trout (Oncorhynchus mykiss). Photochem Photobiol 85:
      347-352. http://dx.doi.0rg/10.1111/i.1751-1097.2008.00446.x.
Marguis, O: Miaud,  C: Lena, JP (2008). Developmental responses to UV-B radiation in common frog Rana
      temporaria embryos from along an altitudinal gradient. Population Ecology 50: 123-130.
      http://dx.doi.org/10.1007/s10144-007-0071-3.
Marguis. O: Miaud.  C. (2008). Variation in UV sensitivity among common frog Rana temporaria populations
      along an altitudinal gradient. Zoology (Jena) 111: 309-317. http://dx.doi.Org/10.1016/i.zool.2007.09.003.
Marshall, E: Harris,  G: Wartenberg, D. (2010).  Oral cleft defects and maternal  exposure to ambient air pollutants
      in New Jersey. Birth Defects Res A Clin  Mol Teratol 88: 205-215. http://dx.doi. org/10.1002/bdra.20650.
Marshall. JD:  Netherv. E: Brauer. M.  (2008). Within-urban variability  in ambient air pollution: Comparison of
      estimation methods. Atmos Environ 42:  1359-1369. http://dx.doi.Org/10.1016/i.atmosenv.2007.08.012.
Martin, MJ: Host, GE: Lenz,  KE: Isebrands, JG. (2001). Simulating the growth  response  of aspen to elevated
      ozone:  A mechanistic approach to scaling a leaf-level model of ozone effects on photosynthesis to a
      complex canopy architecture. Environ Pollut 115: 425-436.
Martfnez-Canabal. A: Angora-Perez. M. (2008). Effect of growth hormone on cyclooxygenase-2 expression in
      the hippocampus of rats chronically exposed to ozone. Int J Neurosci 118: 455-469.
      http://dx.doi. org/10.1080/00207450701593160.
Martinez. FD:  Wright. AL: Taussig. LM: Holberg. CJ: Halonen. M: Morgan. WJ: Associates. GHM. (1995).
      Asthma and  wheezing in the first six years of life. N Engl J Med 332: 133-138.
Martrette, JM: Thornton, SN: Trabalon, M. (2011). Prolonged ozone exposure effects behaviour, hormones and
      respiratory muscles in young female rats.  Physiol Behav 103: 302-307.
      http://dx.doi.0rg/10.1016/i.physbeh.2011.02.024.
Maruo, YY. (2007).  Measurement of ambient ozone using newly developed porous glass sensor. Sens Actuators
      B 126:  485-491. http://dx.doi.Org/10.1016/i.snb.2007.03.041.
Maruo. YY: Akaoka. K: Nakamura. J. (2010). Development and performance evaluation of ozone detection
      paper using azo dye orange I: Effect of  pH. Sens Actuators B 143: 487-493.
      http://dx.doi.0rg/10.1016/i.snb.2009.09.042.
Massman. WJ: Grantz. DA. (1995). Estimating canopy conductance to ozone uptake from observations of
      evapotranspiration at the canopy scale  and at the leaf scale. Global Change  Biol  1: 183-198.
      http://dx.doi.0rg/10.1111/i.1365-2486.1995.tb00020.x.
Massman, WJ: Musselman,  RC: Lefohn, AS. (2000). A conceptual ozone dose-response model to develop a
      standard to protect vegetation. Atmos Environ  34: 745-759. http://dx.doi.org/10.1016/S1352-
      2310(99)00395-7.
Massman. WJ. (2004). Toward an ozone standard to  protect vegetation based on effective dose: A review of
      deposition resistances and a possible metric [Review]. Atmos Environ 38: 2323-2337.
Mathur.  R. (2008). Estimating the impact of the 2004 Alaskan forest fires on episodic particulate matter pollution
      over the eastern United States through  assimilation of satellite-derived aerosol optical depths  in a
      regional air quality model. J Geophys Res 113: D17302.  http://dx.doi.org/10.1029/2007JD009767.
Matsumura, Y (1970). The effects of ozone, nitrogen  dioxide, and sulfur dioxide on the experimentally induced
      allergic respiratory disorder in guinea pigs: II. The effects of ozone on the absorption and the retention  of
      antigen in the lung. Am Rev Respir Dis  102: 438-443.
Matsumura. Y: Ananthaswamv. HN. (2004). Toxic effects of ultraviolet radiation on the skin. Toxicol Appl
      Pharmacol 195: 298-308.
Matvssek, R:  Gunthardt-Goerg, MS:  Maurer, S: Keller, T (1995). Nighttime exposure to ozone reduces whole-
      plant production in Betula pendula. Tree Physiol 15: 159-165.
Draft - Do Not Cite or Quote                       11-53                                  September 2011

-------
Matyssek, R: Le Thiec, D: Low, M: Dizengremel, P: Nunn, AJ: Haberle, KH. (2006). Interactions between
      drought and O3 stress in forest trees. Plant Biol (Stuttg) 8: 11-17. http://dx.doi.org/10.1055/s-2005-
      873025.
Matyssek, R: Sandermann,  H: Wieser, G: Booker,  F: Cieslik, S: Musselman, R: Ernst, D. (2008). The challenge
      of making ozone risk assessment for forest trees more mechanistic.  Environ Pollut 156: 567-582.
      http://dx.doi.0rg/10.1016/i.envpol.2008.04.017.
Matvssek, R: Wieser, G: Ceulemans, R: Rennenberg, H: Pretzsch, H: Haberer, K: Low, M: Nunn, AJ: Werner, H:
      Wipfler. P: Obwald. W: Nikolova. P: Hanke. DE: Kraigher. H: Tausz. M: Bahnweg. G: Kitao. M: Dieler. J:
      Sandermann. H: Herbinger. K: Grebenc. T:  Blumenrother. M: Deckmvn. G: Grams. TEE: Heerdt. C:
      Leuchner, M: Fabian,  P: Haberle, KH. (2010). Enhanced ozone strongly reduces carbon sink strength of
      adult beech (Fagus sylvatica): Resume from the free-air fumigation study at Kranzberg Forest. Environ
      Pollut 158: 2527-2532. http://dx.doi.Org/10.1016/i.envpol.2010.05.009.
Mautz, WJ: Dohm, MR. (2004). Respiratory and behavioral effects of ozone on a lizard and a frog. Comp
      Biochem  Physiol A Mol Integr Physiol 139: 371-377.
Mayer. LM: Schick. LL: Hardy. KR: Estapa. ML. (2009). Photodissolution and other photochemical changes
      upon irradiation of algal detritus. Limnol Oceanogr 54: 1688-1698.
Mazza. CA: Izaguirre.  MM: Curiale. J: Ballare. CL. (2010). A look into the invisible: Ultraviolet-B sensitivity in an
      insect (Caliothrips phaseoli) revealed through a behavioural action spectrum. Proc Biol Sci 277: 367-373.
      http://dx.doi. org/10.1098/rspb.20Q9.1565.
McAinsh, MR: Evans,  NH: Montgomery, LT: North, KA. (2002). Calcium signalling in stomatal responses to
      pollutants. New Phytol 153: 441-447.
McBride, DE: Koenig,  JQ: Luchtel, PL: Williams, PV: Henderson, WR, Jr. (1994). Inflammatory effects of ozone
      in the upper airways of subjects with asthma. Am J Respir Crit Care  Med 149: 1192-1197.
McBride. JR: Laven. RD. (1999).  Impact of oxidant air pollutants on forest succession in the mixed conifer
      forests of the San Bernardino Mountains. In PR Miller; JR McBride (Eds.), Oxidant air pollution impacts in
      the montane forests of southern California:  A case study of the San Bernardino Mountains (pp. 338-352).
      New York, NY: Springer-Verlag.
McBride, JT. (1992). Architecture  of the tracheobronchial tree. In RA Parent (Ed.), Comparative biology of the
      normal lung (pp. 49-61). Boca Raton,  FL: CRC Press.
McCarthy, HR: Oren, R: Johnsen, KH: Gallet-Budynek, A: Pritchard, SG: Cook, CW: LaDeau, Sl_: Jackson,  RB:
      Finzi, AC. (2009). Re-assessment of plant carbon dynamics at the Duke free-air CO2 enrichment site:
      Interactions of atmospheric [CO2] with nitrogen and water availability over stand development. New
      Phytol 185: 514-528. http://dx.doi.Org/10.1111/i.1469-8137.2009.03078.x.
McConnell, R: Berhane, K: Gilliland, F: London, SJ: Islam, T: Gauderman, WJ: Avol, E: Margolis, HG: Peters,
      JM. (2002). Asthma in exercising children exposed to ozone: A cohort study. Lancet 359: 386-391.
McConnell. R: Berhane. K: Yao. L: Lurmann. FW: Avol. E: Peters. JM. (2006). Predicting residential ozone
      deficits from nearby traffic. Sci Total Environ 363: 166-174.
McConnell, R: Islam, T: Shankardass, K: Jerrett, M: Lurmann, F: Gilliland, F: Gauderman, J: Avol, E: Kuenzli, N:
      Yao, L: Peters,  J: Berhane, K. (2010). Childhood incident asthma and traffic-related air pollution at home
      and school. Environ Health Perspect 118: 1021-1026. http://dx.doi.org/10.1289/ehp.0901232.
McCool. PM: Musselman. RC: Younglove. T: Teso. RR. (1988). Response of kidney bean to sequential ozone
      exposures. Environ Exp Bot 28: 307-313.
McCurdv, T: Glen, G: Smith,  L:  Lakkadi, Y (2000). The National Exposure Research Laboratory's consolidated
      human activity database. J Expo Sci Environ Epidemiol 10:  566-578.
McDermott. M: Srivastava. R: Croskell. S. (2006). Awareness of and compliance with air pollution advisories: A
      comparison of parents of asthmatics with other parents. J Asthma 43: 235-239.
      http://dx.doi.org/10.1080/02770900600567114.
McDonnell, WF:  Horstman, DH: Hazucha, MJ: Seal, E, Jr:  Haak, ED: Salaam, SA: House, DE. (1983).
      Pulmonary effects of ozone exposure during exercise: Dose-response characteristics. J Appl Physiol 54:
      1345-1352.
McDonnell. WF:  Kehrl. HR: Abdul-Salaam. S: Ives. PJ: Folinsbee. LJ:  Devlin. RB: O'Neil. JJ: Horstman. DH.
      (1991). Respiratory response of humans exposed to low levels of ozone for 66 hours. Arch Environ
      Occup Health 46: 145-150.
McDonnell. WF.  (1996). Individual variability in human lung  function responses to ozone exposure. Environ
      Toxicol Pharmacol 2: 171-175.
McDonnell, WF:  Stewart, PW: Andreoni, S: Seal, E, Jr: Kehrl,  HR: Horstman, DH: Folinsbee, LJ: Smith, MV.
      (1997). Prediction of ozone-induced FEV1 changes:  Effects of concentration, duration, and ventilation.
      Am J Respir Crit Care Med 156: 715-722.
McDonnell. WF:  Stewart. PW: Smith. MV: Pan. WK: Pan. J. (1999). Ozone-induced respiratory symptoms:
      Exposure-response models and association with lung function. Eur Respir J 14: 845-853.
McDonnell, WF:  Stewart, PW: Smith, MV. (2007). The temporal dynamics of ozone-induced FEV1 changes in
      humans: An exposure-response model. Inhal Toxicol 19: 483-494.
McDonnell. WF:  Stewart. PW: Smith. MV. (2010). Prediction of ozone-induced lung function responses in
      humans.  Inhal Toxicol 22: 160-168. http://dx.doi.org/10.3109/08958370903089557.
Draft - Do Not Cite or Quote                       11-54                                 September 2011

-------
McDonnell, WF, III: Horstman, DH: Abdul-Salaam, S: House, DE. (1985a). Reproducibility of individual
      responses to ozone exposure. Am Rev Respir Dis 131:  36-40.
McDonnell, WF, III: Chapman, RS: Leigh, MW: Strope, Gl_: Collier, AM. (1985b). Respiratory responses of
      vigorously exercising children to 012 ppm ozone exposure. Am Rev Respir Dis 132: 875-879.
McElrov. MB: Salawitch. RJ: Wofsv. SC: Logan. JA.  (1986). Reductions of Antarctic ozone due to synergistic
      interactions of chlorine and bromine. Nature 321: 759-762.
McFrederick, QS: Kathilankal, JC: Fuentes, JD. (2008). Air pollution modifies floral scent trails. Atmos Environ
      42:  2336-2348. http://dx.doi.org/10.1016/i.atmosenv.2007.12.033.
McFrederick. QS: Fuentes. JD: Roulston. T:  Kathilankal. JC: Lerdau. M. (2009). Effects of air pollution on
      biogenic volatiles and ecological interactions. Oecologia 160: 411-420. http://dx.doi.org/10.1007/s00442-
      009-1318-9.
McKinnev, WJ: Jaskot, RH: Richards, JH: Costa, PL: Dreher, KL. (1998). Cytokine mediation of ozone-induced
      pulmonary adaptation. Am J Respir Cell Mol Biol 18: 696-705.
Mclaughlin. SB: Nosal. M: Wullschleger. SD: Sun. G. (2007a). Interactive effects of ozone and climate on tree
      growth and water use in a southern Appalachian forest in the USA. New Phytol 174: 109-124.
      http://dx.doi.0rg/10.1111/i.1469-8137.2007.02018.x.
McLaughlin. SB: Wullschleger. SD: Sun. G: Nosal. M. (2007b). Interactive effects of ozone and climate on water
      use, soil moisture content and streamflow in a southern Appalachian forest in the USA. New Phytol 174:
      125-136. http://dx.doi.0rg/10.1111/i.1469-8137.2007.01970.x.
McLeod, AR: Long, SP. (1999). Free-air carbon dioxide enrichment (FACE) in global change research: A review.
      Adv Ecol Res 28: 1-56. http://dx.doi.org/10.1016/50065-2504(08)60028-8.
Meador, JA: Baldwin, AJ: Catala, P: Jeffrey, WH: Joux, F:  Moss, JA: Pakulski, JD: Stevens,  R: Mitchell, PL.
      (2009). Sunlight-induced DNA damage in  marine micro-organisms collected along a latitudinal gradient
      from 70 degrees N to 68 degrees S. Photochem Photobiol 85: 412-421.  http://dx.doi.orq/10.1111/1.1751-
      1097.2008.00462.x.
Medina-Ramon. M: Zanobetti. A: Schwartz. J. (2006). The effect of ozone and PM10 on hospital admissions for
      pneumonia and chronic obstructive pulmonary disease: A national multicity study. Am J Epidemiol  163:
      579-588. http://dx.doi.org/10.1093/aie/kwjQ78.
Medina-Ramon, M: Schwartz, J. (2008). Who is more vulnerable to die from ozone air pollution? Epidemiology
      19:  672-679.
Medlyn, BE: Barton, CVM: Broadmeadow, MSJ: Ceulemans, R: DeAngelis, P:  Forstreuter, M: Freeman, M:
      Jackson. SB: Kellomaki. S: Laitat. E: Rev. A:  Roberntz.  P: Sigurdsson. BD: Strassemever.  J: Wang. K:
      Curtis.  PS: Jarvis. PG. (2001). Stomatal conductance of forest species after long-term exposure to
      elevated CO2 concentration: A synthesis.  New Phytol 149: 247-264. http://dx.doi.Org/10.1046/j. 1469-
      8137.2001.00028.x.
Meehan. TD: Crosslev. MS: Lindroth. RL. (2010). Impacts of elevated CO2 and O3 on aspen leaf litter chemistry
      and earthworm and springtail productivity. Soil Biol Biochem 42: 1132-1137.
      http://dx.doi.0rg/10.1016/i.soilbio.2010.03.019.
Menendez, Al:  Romero, AM: Folcia, AM: Martinez-Ghersa, MA. (2009). Getting the interactions right: Will higher
      O3  levels interfere with induced defenses to aphid feeding?  Basic Appl Ecol 10: 255-264.
      http://dx.doi.0rg/10.1016/i.baae.2008.03.010.
Menendez. Al:  Romero. AM: Folcia. AM: Martinez-Ghersa. MA. (2010). Aphid and episodic O3 injury in arugula
      plants (Eruca sativa Mill) grown in open-top field chambers. Agric Ecosyst Environ  135: 10-14.
      http://dx.doi.0rg/10.1016/i.agee.2009.08.005.
Meng. YY: Wilhelm. M: Rull. RP: English. P:  Ritz. B. (2007). Traffic and outdoor air pollution levels near
      residences and poorly controlled asthma in adults. Ann Allergy Asthma Immunol 98: 455-463.
Meng, YY: Rull, RP: Wilhelm, M: Lombardi, C: Balmes, J:  Ritz, B. (2010). Outdoor air pollution and  uncontrolled
      asthma in the San Joaquin Valley, California.  J Epidemiol Community  Health 64: 142-147.
      http://dx.doi.org/10.1136/iech.2008.083576.
Mercer. RR: Aniilvel. S: Miller. FJ: Crapo. JD. (1991). Inhomogeneity of ventilatory unit volume and its effects on
      reactive gas uptake.  J Appl Physiol 70: 2193-2205.
Mercer, RR: Russell, ML: Crapo, JD. (1992). Mucous lining layers in human and rat airways [Abstract]. Am Rev
      Respir Dis 145: A355.
Mereu. S: Gerosa. G:  Finco. A: Fusaro. L:  Muvs. B:  Manes. F. (2009).  Improved sapflow methodology reveals
      considerable night-time ozone uptake by Mediterranean species. Biogeosciences 6: 3151-3162.
Messineo, TD: Adams, WC. (1990). Ozone inhalation effects in females varying widely in lung size: Comparison
      with males. J Appl Physiol 69:  96-103.
Metzger, KB: Tolbert, PE: Klein, M: Peel, JL: Flanders,  WD: Todd, KH: Mulholland, JA: Ryan, PB:  Frumkin, H.
      (2004). Ambient air pollution and cardiovascular emergency department visits. Epidemiology 15: 46-56.
Metzger. KB: Klein. M: Flanders. WD: Peel. JL: Mulholland. JA: Langberg. JJ: Tolbert.  PE. (2007). Ambient air
      pollution and cardiac arrhythmias in patients with implantable defibrillators. Epidemiology 18: 585-592.
      http://dx.doi. org/10.1097/EDE.Ob013e318124ffOe.
Michelson. PH: Dailev. L: Devlin. RB: Peden. DB. (1999). Ozone effects on the immediate-phase  response to
      allergen in the nasal airways of allergic asthmatic subjects. Otolaryngol Head Neck Surg 120: 225-232.
Draft - Do Not Cite or Quote                        11-55                                  September 2011

-------
Mickley, LJ: Leibensperger, EM: Jacob, DJ: Rind, D. (In Press) Regional warming from aerosol removal over the
      United States: Results from a transient 2010-2050 climate simulation. Atmos Environ.
      http://dx.doi.0rg/10.1016/i.atmosenv.2011.07.030.
Micklev, LJ: Murti, PP: Jacob, DJ: Logan, JA: Koch, DM: Rind,  D. (1999). Radiative forcing from tropospheric
      ozone calculated with a unified chemistry-climate model. J Geophys Res 104:  30153-30172.
      http://dx.doi. orq/10.1029/1999JD900439.
Micklev, LJ: Jacob, DJ: Rind,  D. (2001). Uncertainty in preindustrial abundance of tropospheric ozone:
      Implications for radiative forcing calculations. J Geophys Res 106: 3389-3399.
      http://dx.doi. orq/10.1029/2000JD900594.
Micklev, LJ: Jacob, DJ: Field,  BD: Rind, D. (2004). Climate response to the increase in tropospheric ozone since
      preindustrial times: A comparison between ozone and equivalent CO2 forcings. J Geophys Res 109:
      D05106. http://dx.doi.0rg/10.1029/2003JD003653.
Middleton, N; Yiallouros,  P: Kleanthous, S: Kolokotroni, O: Schwartz, J: Dockery, DW: Demokritou, P: Koutrakis,
      P. (2008). A 10-year time-series analysis of respiratory and cardiovascular morbidity in Nicosia, Cyprus:
      The effect of short-term changes in air pollution and dust storms. Environ Health 7: 39.
Mikerov, AN: Hague, R: Gan,  X: Quo, X: Phelps, PS: Floros, J. (2008a). Ablation of SP-A has a negative impact
      on the susceptibility of mice to Klebsiella pneumoniae infection after ozone exposure: Sex differences.
      Respir Res 9: 77.  http://dx.doi.org/10.1186/1465-9921-9-77.
Mikerov, AN: Umstead, TM: Gan, X: Huang, W: Quo, X: Wang, G: Phelps, PS: Floros, J. (2008b). Impact of
      ozone exposure on the phagocytic activity of human surfactant protein A (SP-A) and SP-A variants. Am J
      Physiol Lung Cell  Mol Physiol 294: L121-L130. http://dx.doi.org/10.1152/aiplung.00288.2007.
Mikerov, AN: Gan, X:  Umstead, TM: Miller, l_: Chinchilli, VM: Phelps, PS:  Floros, J. (2008c). Sex differences in
      the impact of ozone on survival and alveolar macrophage function of mice after Klebsiella pneumoniae
      infection. Respir Res 9: 24. http://dx.doi.org/10.1186/1465-9921-9-24.
Miles, GP: Samuel, MA: Zhang, Yl_: Ellis, BE. (2005). RNA interference-based (RNAi) suppression of AtMPK6,
      an Arabidopsis mitogen-activated protein kinase, results in hypersensitivity to ozone and misregulation of
      AtMPKS. Environ Pollut 138: 230-237. http://dx.doi.Org/10.1016/i.envpol.2005.04.017.
Milford, JB:  Gao, D: Sillman, S: Blossev, P: Russell, AG. (1994). Total reactive nitrogen (NOy) as an indicator of
      the sensitivity of ozone to reductions in hydrocarbon and NOx emissions. J Geophys Res 99: 3533-3542.
Miller, FJ: Illing, JW:  Gardner, DE. (1978). Effect of urban ozone levels on laboratory-induced respiratory
      infections. Toxicol Lett  2: 163-169.
Miller. FJ: McNeal. CA: Kirtz. JM: Gardner. DE: Coffin. PL: Menzel.  DB. (1979). Nasopharyngeal removal of
      ozone in rabbits and guinea pigs. Toxicology 14: 273-281.  http://dx.doi.org/10.1016/0300-
      483X(79)90009-X.
Miller. FJ: Overton. JH. Jr: Jaskot. RH: Menzel. DB. (1985). A model of the regional uptake of gaseous
      pollutants in the lung: I. The sensitivity of the uptake of ozone in the human lung to lower respiratory tract
      secretions and exercise. Toxicol Appl Pharmacol 79: 11-27. http://dx.doi.org/10.1016/0041-
      008X(85)90364-3.
Miller, FJ: Overton, JH: Gerritv, TR: Graham, RC. (1988). Interspecies dosimetry of reactive gases. In U Mohr; D
      Dungworth; R  McClellan; G Kimmerle; W Stober; J Lewkowski (Eds.), Inhalation toxicology: The design
      and interpretation  of inhalation studies and their use in risk assessment (pp. 139-155). New York, NY:
      Springer-Verlag.
Miller, FJ: Kimbell, JS. (1995). Regional dosimetry of inhaled reactive gases. In RO McClellan; RF  Henderson
      (Eds.), Concepts in inhalation toxicology (2nd ed., pp. 257-287). Washington, DC: Taylor & Francis.
Miller. FJ. (1995).  Uptake and fate of ozone in the respiratory tract. Toxicol Lett 82-83: 277-285.
Miller, LA: Gerriets, JE: Tyler,  NK: Abel, K: Schelegle, ES: Plopper, CG: Hyde,  DM. (2009). Ozone and allergen
      exposure during postnatal development alters the frequency and airway distribution of CD25+ cells in
      infant rhesus monkeys. Toxicol Appl Pharmacol 236: 39-48. http://dx.doi.Org/10.1016/i.taap.2008.12.031.
Miller, PL. (1973).  Oxidant-induced community change in a  mixed conifer forest. In JA Naegele (Ed.), Air
      pollution damage to vegetation (pp. 101-117). Washington, DC: American Chemical Society.
Miller. PR: Parmeter. JR. Jr: Taylor. OC: Cardiff. EA. (1963). Ozone injury to the foliage of Pinus ponderosa.
      Phytopathology 53: 1072-1076.
Miller. PR: McCutchan. MH: Ryan. BC. (1972). Influence of climate  and topography on oxidant air pollution
      concentrations that damage conifer forests in southern California.  Mitt Forstl Bundesversuchsanst Wien
      97: 585-607.
Miller, PR: Elderman, MJ. (1977). Photochemical oxidant air pollutant effects on a mixed conifer forest
      ecosystem: A progress report,  1976. Corvallis, Oregon:  U.S.  Environmental Protection Agency.
Miller, PR: Rechel, J.  (1999). Temporal changes in crown condition  indices, needle litterfall, and collateral
      needle injuries of Ponderosa and Jeffrey pines. In  PR Miller;  JR McBride (Eds.), Oxidant air pollution
      impacts in the  Montane forests of southern California: A case study of the San Bernardino Mountains (pp.
      164-178). New York, NY: Springer.
Mills. G: Ball. G: Haves. F: Fuhrer. J:  Skarbv. L: Gimeno.  B: De Temmerman. L. (2000). Development of a multi-
      factor model for predicting the  effects of ambient ozone  on the biomass of white clover. Environ Pollut
      109:  533-542.  http://dx.doi.org/10.1016/S0269-7491 (00)00057-9.
Draft - Do Not Cite or Quote                        11-56                                 September 2011

-------
Mills, G. (2002). Modification of plant response by environmental conditions. In JNB Bell; M Treshow (Eds.), Air
      pollution and plant life (2nd ed., pp. 343-358). Chichester, United Kingdom: John Wiley & Sons.
Mills, G: Haves, F: Jones,  MLM: Cinderby, S. (2007a). Identifying ozone-sensitive communities of (semi-)natural
      vegetation suitable  for mapping exceedance of critical levels. Environ Pollut 146: 736-743.
      http://dx.doi.0rg/10.1016/i.envpol.2006.04.005.
Mills. G: Buse. A: Gimeno. B: Bermeio. V: Holland. M:  Emberson. L: Pleiiel. H. (2007b). A synthesis of AOT40-
      based response functions and critical levels of ozone for agricultural and horticultural crops. Atmos
      Environ 41: 2630-2643.  http://dx.doi.Org/10.1016/i.atmosenv.2006.11.016.
Mills. G: Haves. F: Wilkinson. S: Davies. WJ. (2009). Chronic exposure to increasing background ozone impairs
      stomatal functioning in grassland species. Global Change Biol 15: 1522-1533.
      http://dx.doi.0rg/10.1111/i.1365-2486.2008.01798.x.
Miwa, T: Maruo, YY: Akaoka, K: Kunioka, T: Nakamura, J. (2009). Development of colorimetric ozone detection
      papers with high ultraviolet resistance using ultraviolet absorbers. J Air Waste Manag Assoc 59: 801-808.
      http://dx.doi.0rg/10.3155/1047-3289.59.7.801.
Moehrle. M: Heinrich. L: Schmid. A: Garbe. C. (2000).  Extreme UV exposure of professional cyclists.
      Dermatology 201: 44-45.
Moehrle. M. (2001). Ultraviolet exposure in the Ironman triathlon.  Med Sci Sports Exerc 33: 1385-1386.
Moffatt. RK: Hyde. DM: Plopper. CG: Tyler. WS: Putney.  LF. (1987). Ozone-induced adaptive and reactive
      cellular changes in  respiratory bronchioles of Bonnet monkeys. Exp Lung Res 12: 57-74.
Moise, AF: Buttner, PG:  Harrison,  SL. (1999). Sun exposure at school. Photochem Photobiol 70: 269-274.
Mokoena, ML: Harvey, BH: Oliver, DW: Brink, CB. (2010). Ozone modulates the effects of imipramine on
      immobility in the forced swim test, and nonspecific parameters of hippocampal oxidative stress in the rat.
      Metab Brain Dis 25: 125-133. http://dx.doi.org/10.1007/s11011-010-9189-7.
Molfino. NA: Wright. SC: Katz.  I: Tarlo. S: Silverman. F: McClean. PA: Szalai. JP:  Raizenne. M: Slutskv. AS:
      Zamel, N. (1991). Effect of low concentrations of ozone on inhaled allergen responses in asthmatic
      subjects. Lancet 338: 199-203.
Mollner. AK: Valluvadasan. S: Feng. L: Sprague. MK: Okumura. M: Milligan. DB: Bloss. WJ: Sander. SP:
      Martien, PT: Harlev, RA. (2010). Rate of gas phase association of hydroxyl radical and nitrogen dioxide.
      Science 330: 646-649. http://dx.doi.org/10.1126/science.1193030.
Monchaux, G: Morlier, JP:  Morin, M: Rochefort, P: Maximilien, R: Tredaniel, J. (1996). Co-carcinogenic effects
      in rats of combined exposure to radon and ozone. Environ Int221: S909-S915.
Mondor. EB: Tremblav. MN: Awmack. CS: Lindroth. RL. (2004). Divergent pheromone-mediated insect
      behaviour under global atmospheric change. Global Change Biol 10: 1820-1824.
Mondor, EB: Tremblav, MN: Awmack, CS: Lindroth, RL. (2005). Altered genotypic and phenotypic frequencies of
      aphid populations under enriched CO2 and O3  atmospheres. Global Change Biol 11: 1990-1996.
      http://dx.doi.0rg/10.1111/i.1365-2486.2005.01054.x.
Mondor, EB: Awmack, CS: Lindroth, RL. (2010). Individual growth rates do not predict aphid population
      densities under altered atmospheric conditions. Agr Forest Entomol 12: 293-299.
      http://dx.doi.0rg/10.1111/i.1461-9563.2010.00478.x.
Moon, JS: Kim, YS: Kim, JH: Son, BS: Kim, PS: Yang, W. (2009). Respiratory health effects among
      schoolchildren and  their relationship to air pollutants in  Korea. Int J Environ Health Res 19: 31-48.
      http://dx.doi.org/10.1080/09603120802272201.
Moore, K: Neugebauer, R: Lurmann, F: Hall, J: Brajer,  V: Alcorn, S: Tager, I. (2008). Ambient ozone
      concentrations cause increased hospitalizations for asthma in children: An 18-year study in Southern
      California. Environ Health Perspect 116: 1063-1070. http://dx.doi.org/10.1289/ehp.10497.
Morello-Frosch, R: Jesdale, BM: Sadd, JL: Pastor,  M. (2010). Ambient air pollution exposure and full-term birth
      weight in California. Environ Health 9: 44. http://dx.doi.org/10.1186/1476-069X-9-44.
Morgan, PB: Ainsworth,  EA: Long, SP. (2003). How does elevated ozone impact soybean? A meta-analysis of
      photosynthesis, growth and yield. Plant Cell Environ 26: 1317-1328.
Morgan. PB:  Bernacchi. CJ: Ort. PR: Long. SP. (2004). An in vivo analysis of the  effect of season-long open-air
      elevation of ozone to anticipated 2050 levels on photosynthesis in soybean. J Plant Physiol 135: 2348-
      2357.
Morgan. PB:  Mies. TA: Bollero. GA: Nelson. RL: Long. SP. (2006). Season-long elevation of ozone
      concentration to projected 2050 levels under fully open-air conditions substantially decreases the growth
      and production of soybean. New Phytol 170: 333-343. http://dx.doi.org/10.11111\. 1469-
      8137.2006.01679.x.
Morison, JIL: Lawlor, DW.  (1999).  Interactions between increasing CO2 concentration and temperature on  plant
      growth. Plant Cell Environ 22: 659-682. http://dx.doi.org/10.1046/j. 1365-3040.1999.00443.x.
Morris. CR: Poliakovic. M:  Lavrisha. L: Machado. L: Kuypers. FA: Morris. SM. Jr.  (2004). Decreased arginine
      bioavailability and increased serum arginase activity in asthma. Am J Respir Crit Care Med 170: 148-
      153. http://dx.doi.org/10.1164/rccm.200309-1304OC.
Morrow. JD: Hill. KE: Burk. RF: Nammour. TM: Badr. KF: Roberts. LJ. 2nd. (1990). A series of prostaglandin F2-
      like compounds are produced in vivo in humans by a non-cyclooxygenase, free radical-catalyzed
      mechanism. PNAS  87: 9383-9387.
Draft - Do Not Cite or Quote                        11-57                                  September 2011

-------
Morsky, SK: Haapala, JK: Rinnan, R: Tiiva, P: Saarnio, S: Silvola, J: Holopainen, T: Martikainen, PJ. (2008).
      Long-term ozone effects on vegetation, microbial community and methane dynamics of boreal peatland
      microcosms in open-field conditions. Global Change Biol 14: 1891-1903. http://dx.doi.org/10.1111/1.1365-
      2486.2008.01615.x.
Mortimer. K: Neugebauer. R: Lurmann. F: Alcorn. S: Balmes. J: Tager. I. (2008a). Air pollution  and pulmonary
      function in asthmatic children: Effects of prenatal and lifetime exposures. Epidemiology 19: 550-557.
      http://dx.doi. org/10.1097/EDE.Ob013e31816a9dcb.
Mortimer. K: Neugebauer. R: Lurmann. F: Alcorn. S: Balmes. J: Tager. I. (2008b). Early-lifetime exposure to air
      pollution and allergic sensitization in children with asthma. J Asthma 45: 874-881.
      http://dx.doi. org/10.1080/02770900802195722.
Mortimer, KM: Tager, IB: Dockery, DW: Neas, LM: Redline, S. (2000). The effect of ozone on inner-city children
      with asthma: Identification of susceptible subgroups. Am J Respir Crit Care  Med 162: 1838-1845.
Mortimer, KM: Neas, LM: Dockery, DW: Redline, S: Tager, IB. (2002). The effect of air pollution on inner-city
      children with asthma.  Eur  Respir J 19: 699-705. http://dx.doi.org/10.1183/09031936.02.00247102.
Mudd. JB. (1996). Biochemical basis for the toxicity of ozone. In M Yunus; M Igbal  (Eds.), Plant response to air
      pollution (pp. 267-283).  New York, NY: John Wiley & Sons.
Mudwav.  IS: Houslev. D: Eccles.  R: Richards. RJ: Datta. AK: Tetlev. TD: Kelly.  FJ. (1996). Differential depletion
      of  human respiratory tract antioxidants in response to ozone challenge. Free Radic Res 25: 499-513.
Mudwav,  IS: Kelly, FJ. (1998). Modeling the interactions of ozone with pulmonary epithelial lining fluid
      antioxidants. Toxicol Appl Pharmacol 148: 91-100.
Mudwav,  IS: Blomberg, A: Frew, AJ:  Holgate,  ST: Sandstrom, T: Kelly, FJ. (1999a). Antioxidant consumption
      and repletion kinetics in nasal lavage fluid following exposure of healthy human volunteers to ozone. Eur
      Respir J 13: 1429-1438.
Mudwav.  IS: Krishna.  MT: Frew. AJ:  MacLeod. D: Sandstrom. T: Holgate. ST: Kelly. FJ. (1999b). Compromised
      concentrations of ascorbate in fluid lining the respiratory tract in human subjects after exposure to ozone.
      Occup Environ Med 56: 473-481.
Mudwav.  IS: Kelly. FJ. (2000). Ozone and the lung: A sensitive issue. Mol Aspects Med 21: 1-48.
Mudwav,  IS: Stenfors, N: Blomberg,  A: Helledav, R: Punster, C: Marklund, Sl_: Frew, AJ: Sandstrom, T: Kelly,
      FJ. (2001). Differences in  basal airway antioxidant concentrations are not predictive of  individual
      responsiveness to ozone:  A comparison of healthy and  mild asthmatic subjects. Free Radic Biol Med 31:
      962-974.
Mudwav.  IS: Kelly. FJ. (2004a). An investigation of inhaled ozone dose and the magnitude of airway
      inflammation in healthy adults. Am J Respir Crit Care Med 169: 1089-1095.
Mudwav,  IS: Kelly, FJ. (2004b). An investigation of inhaled ozone dose and the magnitude of airway
      inflammation in healthy adults: Online data supplement. Am J Respir Crit Care  Med 169: 1089-1095.
      http://dx.doi.org/10.1164/rccm.200309-1325PP.
Mudwav,  IS: Behndig, AF: Helledav,  R: Pourazar, J: Frew, AJ: Kelly, FJ: Blomberg,  A. (2006). Vitamin
      supplementation does not protect against symptoms in  ozone-responsive subjects. Free Radic Biol Med
      40: 1702-1712.
Mueller, SF: Mallard, JW. (2011a). Contributions of natural emissions to ozone and PM 2.5 as  simulated by the
      Community Multiscale Air  Quality (CMAQ) model. Environ Sci Technol 45: 4817-4823.
      http://dx.doi. org/10.1021/es103645m.
Mueller, SF: Mallard, JW. (2011 b). Errata in 'Contributions of natural emissions to ozone and PM 2.5 as
      simulated  by the Community Multiscale Air Quality (CMAQ)  model' [Erratum]. Environ Sci Technol 45:
      7950.  http://dx.doi.org/10.1021/es2027086.
Muntifering, RB: Chappelka, AH:  Lin, JC: Karnoskv, DF: Somers, GL. (2006). Chemical composition and
      digestibility of Trifolium exposed to elevated ozone and  carbon  dioxide in a free-air (FACE) fumigation
      system.  Funct Ecol 20: 269-275. http://dx.doi.Org/10.1111/i.1365-2435.2006.01093.x.
Murphy, RC: Johnson, KM. (2008). Cholesterol, reactive oxygen species, and the formation of biologically  active
      mediators. J Biol  Chem 283: 15521-15525. http://dx.doi.org/10.1074/ibc.R700049200.
Murphy. SD: Ulrich. CE:  Frankowitz.  SH: Xintaras. C. (1964). Altered function in animals inhaling low
      concentrations of ozone and nitrogen dioxide. Am Ind Hyg Assoc J 25: 246-253.
Murugan. A: Prvs-Picard. C:  Calhoun. WJ. (2009). Biomarkers in asthma. Curr Opin Pulm Med 15: 12-18.
      http://dx.doi. org/10.1097/MCPOb013e32831 de235.
Musselman, RC:  McCool, PM:  Younglove, T. (1988). Selecting ozone exposure statistics for determining crop
      yield loss from air pollutants. Environ Pollut 53: 63-78. http://dx.doi.org/10.1016/0269-7491 (88)90025-5.
Musselman, RC:  Massman, WJ.  (1999). Ozone flux to vegetation and its relationship to plant response and
      ambient air guality standards. Atmos Environ 33: 65-73.
Musselman. RC:  Minnick. TJ. (2000). Nocturnal stomatal conductance and ambient air guality standards for
      ozone. Atmos Environ 34:  719-733. http://dx.doi.org/10.1016/S1352-2310(99)00355-6.
Musselman, RC:  Lefohn, AS: Massman, WJ: Heath, RL. (2006). A critical review and analysis  of the use of
      exposure- and flux-based  ozone indices for predicting vegetation effects [Review]. Atmos Environ 40:
      1869-1888. http://dx.doi.org/10.1016/i.atmosenv.20Q5.10.064.
Draft - Do Not Cite or Quote                        11-58                                  September 2011

-------
Myatt, L: Kossenjans, W: Sahay, R: Els, A: Brockman, D. (2000). Oxidative stress causes vascular dysfunction
      in the placenta. J Matern Fetal Med 9: 79-82. http://dx.doi.org/10.1002/(SICI)1520-
      6661(200001/02)9:1<79::AID-MFM16>3.0.CO:2-O.
Naeher, LP; Holford, TR; Beckett, WS; Belanger, K; Triche, EW; Bracken, MB; Leaderer, BP. (1999). Healthy
      women's PEF variations with ambient summer concentrations of PM10, PM2.5, SO42-, H+, and O3. Am
      J Respir Grit Care Med 160: 117-125.
Naik, V: Mauzerall,  D: Horowitz, L: Schwarzkopf, MD: Ramaswamy, V: Oppenheimer, M. (2005). Net radiative
      forcing due to changes in regional emissions of tropospheric ozone precursors. J Geophys Res 110:
      D24306. http://dx.doi.org/10.1029/2005JD005908.
Naja, M: Akimoto, H. (2004). Contribution of regional pollution and long-range transport to the Asia-Pacific
      region: Analysis of long-term ozonesonde data over Japan. J Geophys Res 109: D21306.
      http://dx.doi. org/10.1029/2004JD004687.
Nakamura, K: Matsunaga, K. (1998). Susceptibility of natural killer (NK) cells to reactive oxygen species (ROS)
      and their restoration by the mimics of superoxide dismutase (SOD). Cancer Biother Radiopharm 13: 275-
      290.
Mali, C: Balducci, E: Frati, L: Paoli, L: Loppi, S: Lorenzini, G. (2007). Integrated biomonitoring of air quality with
      plants and lichens: A case study on ambient ozone from central Italy. Chemosphere 67: 2169-2176.
      http://dx.doi. org/10.1016/i.chemosphere.20Q6.12.036.
NAPAP (National Acid Precipitation Assessment Program). (1987). Diagnosing injury to Eastern forest trees: A
      manual for identifying damage caused by air pollution, pathogens, insects, and abiotic stresses. In.
      University Park, PA: Pennsylvania State University.
Nash, TH, III.  (2008). Lichen sensitivity to air pollution. In Lichen Biology (pp. 299-314). Cambridge, United
      Kingdom: Cambridge University Press.
Nassar. R: Logan. JA: Worden. HM:  Megretskaia. IA: Bowman. KW: Osterman. GB:  Thompson. AM: Tarasick.
      DW: Austin, S: Claude, H: Dubey, MK: Hocking, WK: Johnson, BJ:  Joseph, E: Merrill, J: Morris, GA:
      Newchurch.  M: Oltmans. SJ: Posnv. F: Schmidlin. FJ: Vomel. H: Whiteman. DN: Witte. JC. (2008).
      Validation of Tropospheric Emission Spectrometer (TES) nadir ozone profiles using ozonesonde
      measurements. D15S17 (13 pp.). http://dx.doi.org/10.1029/2007id008819.
Neas, LM: Dockery, DW: Koutrakis, P: Tollerud,  DJ: Speizer, FE. (1995). The association of ambient air pollution
      with twice daily peak expiratory flow rate  measurements in children. Am J Epidemiol 141: 111-122.
Neas, LM: Dockery, DW: Koutrakis, P: Speizer, FE. (1999). Fine particles  and peak flow in children: Acidity
      versus mass. Epidemiology 10: 550-553.
Neidell. M. (2009). Information, avoidance behavior, and health: The effect of ozone  on asthma hospitalizations.
      Journal of Human Resources 44: 450-478.
Neidell. M. (2010). Air quality warnings and outdoor activities: Evidence from Southern California using a
      regression discontinuity design. J Epidemiol Community Health 64: 921-926.
      http://dx.doi.org/10.1136/iech.2008.081489.
Neidell, M: Kinney, PL. (2010). Estimates of the  association between ozone and asthma hospitalizations that
      account for behavioral responses to air quality information. Environ Sci Pol 13: 97-103.
      http://dx.doi.0rg/10.1016/i.envsci.2009.12.006.
Neuberger, M; Schimek, MG; Horak, F, Jr; Moshammer, H; Kundi, M; Frischer, T; Gomiscek, B; Puxbaum, H;
      Hauck, H; AUPHEP-Team. (2004). Acute effects of particulate matter on respiratory diseases, symptoms
      and functions: Epidemiological results of the Austrian Projects on Health Effects  of Particulate Matter
      (AUPHEP). Atmos Environ 38: 3971-3981.
Neufeld. HS: Renfro. JR: Hacker. WD: Silsbee. D. (1992). Ozone in Great Smoky Mountains National Park:
      Dynamics and effects on plants. In RL Berglund (Ed.), Tropospheric ozone and the environment II:
      Effects, modeling and control  (pp. 594-617). Atlanta, GA: Air & Waste Management Association.
Newell, RE: Thouret, V: Cho, JYN: Stoller, P: Marenco, A: Smit, HG. (1999). Ubiquity of quasi-horizontal layers
      in the troposphere. Nature 398: 316-319. http://dx.doi.org/10.1038/18642.
Newson. EJ: Krishna. MT: Lau. LCK: Howarth. PH: Holgate. ST: Frew. AJ. (2000). Effects of short-term
      exposure to 0.2 ppm ozone on biomarkers of inflammation in sputum, exhaled nitric oxide,  and lung
      function in subjects with mild atopic asthma. J Occup Environ Med 42: 270-277.
Nickmilder. M: De Burbure. C: Svlviane. C: Xavier. D: Alfred. B: Alain. D. (2007). Increase of exhaled nitric oxide
      in children exposed to low levels of ambient ozone. J Toxicol Environ Health A 70: 270-274.
Nightingale, JA: Rogers, DF: Chung, KF: Barnes, PJ. (2000). No effect of inhaled budesonide on the response
      to inhaled ozone in normal subjects. Am J Respir Grit Care Med 161: 479-486.
Nikolova, PS: Andersen, CP: Blaschke,  H: Matvssek, R: Haberle, KH. (2010). Belowground effects of enhanced
      tropospheric ozone and drought in a beech/spruce forest (Fagus sylvatica L./Picea abies [L.] Karst).
      Environ Pollut 158: 1071-1078. http://dx.doi.Org/10.1016/i.envpol.2009.07.036.
Nishivama. H: Ikeda. H: Kaneko. T: Fu.  L: Kudo. M: Ito. T: Okubo. T. (1998). Neuropeptides mediate the ozone-
      induced increase in the permeability of the tracheal mucosa in guinea pigs. Am J Physiol 275: L231-
      L238.
NOAA. (National Oceanic and Atmospheric Administration). (2010). The Rapid Update Cycle (RUC), from
      http://ruc.noaa.gov/
Draft - Do Not Cite or Quote                       11-59                                 September 2011

-------
Noctor, G: Foyer, CH. (1998). Ascorbate and glutathione: Keeping active oxygen under control. 49: 249-279.
Nodelman, V: Ultman, JS. (1999). Longitudinal distribution of chlorine absorption in human airways: A
      comparison to ozone absorption. J Appl Physiol 87: 2073-2080.
Nole, G: Johnson, AW. (2004). An analysis of cumulative lifetime solar ultraviolet radiation exposure and the
      benefits of daily sun protection. Dermatol Ther 17: 57-62.
Nolte. CG: Gilliland. AM: Hogrefe. C: Micklev. LJ. (2008). Linking global to regional models to assess future
      climate impacts on surface ozone levels in the United States. J Geophys Res 113: D14307.
      http://dx.doi.ora/10.1029/2007JD008497.
Norbv. RJ: DeLucia.  EH: Gielen. B:  Calfapietra. C: Giardina. CP: King. JS: Ledford. J: McCarthy. HR: Moore.
      DJP: Ceulemans, R: De Angelis, P: Finzi, AC: Karnosky, DF: Kubiske, ME: Lukac, M:  Pregitzer,  KS:
      Scarascia-Mugnozza, GE: Schlesinger, WH: Oren, R. (2005). Forest response to elevated CO2 is
      conserved across a broad range of productivity. PNAS 102: 18052-18056.
      http://dx.doi. org/10.1073/pnas.05Q9478102.
Norval. M: Garssen.  J: Van Loveren. H: EI-Ghorr. AA. (1999). UV-induced changes in the immune response to
      microbial infections in human subjects and  animal models. J Epidemiol 6: S84-S92.
Novak, K: Cherubini, P: Saurer, M: Fuhrer, J: Skellv, JM:  Krauchi, N: Schaub, M. (2007). Ozone air pollution
      effects on tree-ring growth, delta(13)C, visible foliar injury and leaf gas exchange in three ozone-sensitive
      woody plant species. Tree Physiol 27: 941-949.
Noviski, N: Brewer, JP: Skornik, WA: Galli, SJ: Drazen, JM: Martin, TR. (1999). Mast cell activation is not
      required for induction  of airway hyperresponsiveness by ozone in mice. J Appl Physiol 86: 202-210.
Noziere, B: Gonzalez, NJD: Borg-karlson, A, -K: Pel, Y: Redeby, JP: Krejci, R: Dommen, J: Prevot, ASH:
      Anthonsen, T. (2011).  Atmospheric chemistry in stereo: A new look at secondary organic aerosols from
      isoprene. Geophys Res  Lett  38: L11807. http://dx.doi.org/10.1029/2011GL047323.
NPS. (U.S. National  Park Service). (2006). Ozone bioindicators. Washington, DC.
      http://www.nature.nps.gov/air/Pubs/bioindicators/index.cfm.
NPS. (U.S. National  Park Service). (2007). Ozone effects studies.  Washington, DC: U.S. Department of the
      Interior, National Park Service. http://www.nature.nps.aov/air/studies/ecoOzone.cfm.
NPS. (U.S. National  Park Service). (2011). Portable Ozone Monitoring Systems (POMS). Washington,  DC.
      http://www.nature.nps.gov/air/studies/porto3.cfm.
NRC. (National Research Council).  (1991). Rethinking the ozone problem in urban and regional air pollution. In.
      Washington,  DC: The National Academies Press.
NRC. (National Research Council).  (2005). Radiative forcing of climate change: Expanding the concept and
      addressing uncertainties. In.  Washington, DC:  The National Academies Press.
NRC. (National Research Council).  (2007). Models in environmental regulatory decision making. In.
      Washington,  DC: National Academies Press.
NRC. (National Research Council).  (2009). Global sources of local pollution: An assessment of long-range
      transport of key air pollutants to and from the United States. Washington, DC: The National Academies
      Press, http://www.nap.edu/catalog.php7record id=12743.
NTP (National Toxicology Program). (1994).  Toxicology and carcinogenesis: Studies of ozone (CAS No 10028-
      15-6)  and ozone/NNK (CAS  No 10028-15-6/64091-91-4) in F344/N rats and B6C3F1  mice. (Technical
      Report No. 440). Research Triangle Park, NC. http://ntp.niehs.nih.gov/index.cfm?obiectid=070AOEBD-
      081E-B501-E38F640803C3542C.
Nussbaum, S: Geissmann, M: Fuhrer, J. (1995). Ozone exposure-response relationships for mixtures of
      perennial ryegrass and white clover depend on ozone exposure patterns. Atmos Environ 29: 989-995.
      http://dx.doi. org/10.1016/1352-2310(94)00368-U.
O'Byrne, P: Walters, E: Gold, B: Aizawa, H: Fabbri, L: Alpert, S:  Nadel, J: Holtzman,  M. (1983). Neutrophil
      depletion inhibits airway hyperresponsiveness induced by ozone exposure. Am Rev Respir Dis 130: 214-
      219.
O'Byrne, PM: Walters, EH: Aizawa,  H: Fabbri, LM: Holtzman, MJ: Nadel, JA. (1984). Indomethacin inhibits the
      airway hyperresponsiveness but not the neutrophil influx induced by ozone in dogs. Am Rev Respir Dis
      130: 220-224.
O'Connor, GT: Neas, L: Vaughn,  B:  Kattan, M: Mitchell, H: Grain, EF: III, ER: Gruchalla,  R: Morgan, W: Stout, J:
      Adams. GK: Lippmann. M. (2008). Acute respiratory health effects of air pollution on children with asthma
      in US  inner cities. J Allergy Clin Immunol 121: 1133-1139. http://dx.doi.Org/10.1016/i.iaci.2008.02.020.
O'Gara, PJ. (1922). Sulfur dioxide and fume  problems and their solution. J Ind Eng Chem 14: 744-745.
O'Neill, BF: Zangerl, AR: Delucia, EH: Berenbaum, MR. (2008).  Longevity and fecundity of Japanese beetle
      (Popillia japonica) on foliage  grown under elevated carbon dioxide. Environ Entomol 37: 601-607.
O'Neill, BF: Zangerl, AR: Dermodv, O: Bilgin, DP: Casteel, CL: Zavala, JA: DeLucia, EH: Berenbaum, MR.
      (2010). Impact of elevated levels of atmospheric CO2 and herbivory on flavonoids of soybean (Glycine
      max Linnaeus). J Chem Ecol 36: 35-45. http://dx.doi.org/10.1007/s10886-009-9727-0.
O'Neill, MS: Ramirez-Aguilar, M:  Meneses-Gonzalez, F: Hernandez-Avila, M: Geyh, AS: Sienra-Monge, JJ:
      Romieu. I. (2003).  Ozone exposure among Mexico City outdoor workers. J Air Waste  Manag Assoc 53:
      339-346.
Draft - Do Not Cite or Quote                       11-60                                 September 2011

-------
O-Keeffe, S: Fitzpatrick, C: Lewis, E. (2007). An optical fibre based ultra violet and visible absorption
      spectroscopy system for ozone concentration monitoring. Sens Actuators B 125: 372-378.
      http://dx.doi.0rg/10.1016/i.snb.2007.02.023.
Obara, Y: Koshitaka, H: Arikawa, K. (2008). Better mate in the shade: Enhancement of male mating behaviour
      in the cabbage butterfly, Pieris rapae crucivora, in a UV-rich environment. J Exp Biol 211: 3698-3702.
      http://dx.doi.org/10.1242/ieb.021980.
Ogawa:  Co. (Ogawa & Company).  (2007). Ambient air passive sampler for NO-NO2, NOx, SO2, O3, NH3.
      Pompano Beach, FL: Ogawa & Company USA, Inc. http://www.ogawausa.com/passive.html.
Ogawa.  D: Nakaiima. N: Sano. T: Tamaoki. M: Aono. M: Kubo. A: Kanna. M: loki. M: Kamada. H: Sail.  H.
      (2005). Salicylic acid accumulation under O-3 exposure is regulated by ethylene in tobacco plants. Plant
      Cell Physiol 46: 1062-1072.  http://dx.doi.org/10.1093/pcp/pci118.
Ohira, SI: Dasgupta, PK: Schug, KA. (2009). Fiber optic sensor for simultaneous determination of atmospheric
      nitrogen dioxide, ozone, and relative humidity. Anal Chem 81: 4183-4191.
      http://dx.doi.org/10.1021/ac801756z.
Okada. S: Weatherhead. E: Targoff. IN: Wesley. R: Miller. FW: Group. IMCS. (2003). Global surface ultraviolet
      radiation intensity may modulate the clinical and immunologic expression of autoimmune muscle
      disease. Arthritis Rheum 48: 2285-2293.
Oksanen. E: Holopainen. T. (2001). Responses of two birch (Betula pendula Roth) clones to different ozone
      profiles with similar AOT40 exposure. Atmos Environ 35: 5245-5254. http://dx.doi.org/10.1016/S1352-
      2310(01)00346-6.
Olaguer, EP: Rappengluck, B: Lefer, B: Stutz, J: Dibb, J: Griffin,  R: Brune, WH: Shauck, M: Buhr, M: Jeffries, H:
      Vizuete, W: Pinto, JP (2009). Deciphering  the role of radical precursors during the Second Texas Air
      Quality Study. J Air Waste Manag Assoc 59: 1258-1277. http://dx.doi.org/10.3155/1047-3289.59.11.1258.
Olbrich.  M: Betz. G: Gerstner. E: Langebartels.  C: Sandermann. H: Ernst. D. (2005). Transcriptome  analysis of
      ozone-responsive genes in leaves of European beech (Fagus sylvatica L). Plant Biol (Stuttg) 7: 670-676.
      http://dx.doi. org/10.1055/S-2005-873001.
Olbrich.  M: Gerstner. E: Welzl. G: Winkler. JB: Ernst. D. (2009). Transcript responses in leaves of ozone-treated
      beech saplings seasons at an outdoor free air model fumigation site over two growing seasons. Plant
      Soil 323: 61-74. http://dx.doi.Org/10.1007/s11104-009-0129-4.
Olbrich,  M: Gerstner, E: Bahnweg,  G: Haberle,  KH: Matvssek, R: Welzl, G: Heller, W: Ernst,  D. (2010).
      Transcriptional signatures in leaves  of adult European beech trees (Fagus sylvatica L.) in an
      experimentally enhanced free air ozone setting. Environ Pollut 158: 977-982.
      http://dx.doi.0rg/10.1016/i.envpol.2009.08.001.
Ollinger, SV: Aber, JD: Reich, PB. (1997a). Simulating ozone effects on forest productivity: Interactions among
      leaf- and stand-level processes. Ecol Appl  123: 351-358.
Ollinger. SV: Aber. JD: Reich. PB. (1997b). Simulating ozone effects on forest productivity: Interactions among
      leaf-, canopy-, and  stand-level processes.  Ecol Appl 7:  1237-1251.
Ollinger, SV: Aber, JD: Reich, PB: Freuder, RJ.  (2002).  Interactive effects of nitrogen deposition, tropospheric
      ozone, elevated CO2 and land use history  on the carbon  dynamics of northern hardwood forests. Global
      Change Biol 8: 545-562.  http://dx.doi.org/10.1046/j. 1365-2486.2002.00482.x.
Olszvk. DM: Kats. G: Dawson. PJ:  Bvtnerowicz. A: Wolf. J: Thompson. CR. (1986). Characteristics of air
      exclusion systems vs chambers for field  air pollution studies. J Environ Qual 15: 326-334.
Olszyna, KJ: Bailey, EM: Simonaitis, R: Meagher,  JF. (1994). O3 and NOy relationships at a rural site.  J
      Geophys Res 99: 14557-14563.
Oltmans. SJ: Lefohn. AS: Harris. JM: Galballv. I: Scheel. HE: Bodeker.  G: Brunke.  E: Claude. H: Tarasick. D:
      Johnson, BJ: Simmonds, P:  Shadwick, D: Anlauf, K:  Hayden, K: Schmidlin, F: Fujimoto, T: Akagi, K:
      Meyer, C: Nichol, S: Davies, J: Redondas,  A: Cuevas, E. (2006). Long-term changes in tropospheric
      ozone. Atmos Environ 40: 3156-3173. http://dx.doi.Org/10.1016/i.atmosenv.2006.01.029.
Oltmans, SJ: Lefohn, AS: Harris, JM: Shadwick, PS. (2008). Background ozone levels of air entering the west
      coast of the US and assessment of longer-term changes. Atmos Environ 42: 6020-6038.
      http://dx.doi.0rg/10.1016/i.atmosenv.2008.03.034.
Omasa,  K: Shimazaki, Kl: Aiga, I: Larcher, W: Onoe, M. (1987). Image  analysis of chlorophyll fluorescence
      transients for diagnosing the photosynthetic system of attached  leaves. Plant Physiol 84: 748-752.
      http://dx.doi.0rg/10.1104/pp.84.3.748.
Orazzo,  F: Nespoli, L: Ito,  K: Tassinari, D: Giardina, D: Funis, M: Cecchi, A: Trapani, C: Forgeschi, G: Vignini, M:
      Nosetti,  L: Pigna, S: Zanobetti, A. (2009). Air pollution, aeroallergens, and emergency room visits for
      acute respiratory diseases and gastroenteric disorders among young children in six Italian cities. Environ
      Health Perspect 117: 1780-1785. http://dx.doi.org/10.1289/ehp.0900599.
Ordonez. C: Brunner. D: Staehelin. J: Hadiinicolaou. P: Pvle. JA: Jonas. M: Wernli. H: Prevot. ASH.  (2007).
      Strong influence of lowermost stratospheric ozone on lower tropospheric background ozone changes
      over Europe. Geophys Res  Lett 34:  L07805. http://dx.doi.org/10.1029/2006GL029113.
Orendovici-Best. T: Skellv.  JM: Davis. DP:  Ferdinand. JA: Savage. JE:  Stevenson. RE. (2008). Ozone uptake
      (flux) as it relates to ozone-induced foliar symptoms of Prunus serotina and Populus maximowizii x
      trichocarpa. Environ Pollut 151: 79-92. http://dx.doi.Org/10.1016/i.envpol.2007.03.003.
Draft - Do Not Cite or Quote                        11-61                                  September 2011

-------
Orendovici, T: Skelly, JM: Ferdinand, JA: Savage, JE: Sanz, M, -J: Smith, GC. (2003). Response of native
      plants of northeastern United States and southern Spain to ozone exposures; determining
      exposure/response relationships. Environ Pollut 125: 31-40.
Oromi, N: Marquis, O: Miaud, C: Sanuv, D. (2008). Influence of ambient ultraviolet radiation on Bufo calamita
      egg development in a semiarid zone (Catalonia, Spain). J Environ Biol 29: 135-137.
Orvszczvn. MP: Bouzigon. E: Maccario. J: Siroux. V: Nadif. R: Wright. A: Kauffmann.  F. (2007).
      Interrelationships of quantitative  asthma-related phenotypes in the epidemiological study on the genetics
      and environment of asthma, bronchial hyperresponsiveness, and atopy. J Allergy Clin Immunol 119: 57-
      63.
Oshima, RJ: Poe, MP: Braegelmann, PK: Baldwin, DW: Van Way, V. (1976). Ozone dosage-crop loss function
      for alfalfa: A standardized method for assessing crop losses from air pollutants. J Air Pollut Control Assoc
      26: 861-865.
Oshima, RJ: Braegelmann, PK: Baldwin, DW: Van Way, V: Taylor, OC.  (1977). Reduction of tomato fruit size
      and yield by ozone. J Am Soc Hortic Sci 102: 289-293.
Oslund. KL: Hyde. DM:  Putney. LF: Alfaro. MF: Walbv. WF: Tyler. NK: Schelegle. ES.  (2008). Activation of
      neurokinin-1 receptors during ozone inhalation contributes to epithelial injury and repair. Am J Respir Cell
      Mol Biol 39: 279-288. http://dx.doi.ora/10.1165/rcmb.2008-0009OC.
Oslund. KL: Hyde. DM:  Putney. LF: Alfaro. MF: Walbv. WF: Tyler. NK: Schelegle. ES.  (2009). Activation of
      calcitonin gene-related peptide receptor during ozone inhalation contributes to airway epithelial injury and
      repair. Toxicol Pathol  37: 805-813. http://dx.doi.org/10.1177/0192623309345691.
Ostro, B: Lipsett, M: Mann, J: Braxton-Owens, H: White, M. (2001). Air pollution and exacerbation of asthma in
      African-American children in Los Angeles. Epidemiology 12:  200-208.
Ostro. B: Broadwin. R: Green.  S: Feng.  WY: Lipsett. M. (2006). Fine particulate air pollution and mortality in
      nine California counties: Results from CALFINE. Environ Health Perspect 114: 29-33.
Oudin, A: Stromberg, U: Jakobsson, K:  Stroh, E: Bjork, J. (2010). Estimation of short-term effects of air pollution
      on stroke hospital admissions in  southern Sweden. Neuroepidemiology 34: 131-142.
      http://dx.doi.ora/10.1159/000274807.
Overmyer, K: Tuominen, H: Kettunen, R: Betz, C: Langebartels, C: Sandermann, H, Jr: Kangasjarvi, J. (2000).
      Ozone-sensitive Arabidopsis rcdl mutant reveals opposite roles for ethylene and jasmonate signaling
      pathways  in regulating superoxide-dependent cell  death. Plant Cell 12:  1849-1862.
      http://dx.doi.org/10.1105/tpc.12.10.1849.
Overmver. K: Brosche. M: Pellinen. R: Kuittinen. T: Tuominen. H: Ahlfors. R: Keinanen. M: Saarma. M: Scheel.
      D: Kangasiarvi. J. (2005). Ozone-induced programmed cell death in the Arabidopsis radical-induced cell
      deathl mutant. Plant  Physiol  137: 1092-1104. http://dx.doi.org/10.1104/pp.104.055681.
Overmver. K: Kollist. H:  Tuominen. H: Betz. C: Langebartels. C: Wingsle. G:  Kangasiarvi. S: Brader. G:
      Mullineaux. P: Kangasiarvi. J.  (2008). Complex phenotypic profiles leading to ozone sensitivity in
      Arabidopsis thaliana mutants.  Plant Cell Environ 31: 1237-1249.  http://dx.doi.Org/10.1111/i.1365-
      3040.2008.01837.x.
Overton, JH: Graham, RC: Miller, FJ. (1987). A model of the regional uptake of gaseous pollutants in the lung: II.
      The sensitivity of ozone uptake in laboratory animal lungs to anatomical and ventilatory parameters.
      Toxicol Appl Pharmacol 88: 418-432. http://dx.doi.org/10.1016/0041-008X(87)90216-X.
Overton. JH: Graham. RC. (1989). Predictions of ozone absorption  in human lungs from newborn to adult.
      Health Physl: 29-36.
Overton. JH: Graham. RC: Menache. MG: Mercer. RR: Miller. FJ. (1996). Influence of tracheobronchial region
      expansion and volume on reactive gas uptake and interspecies dose  extrapolations. Inhal Toxicol 8: 723-
      745.
Oyarzun, M: Dussaubat, N: Gonzalez, S. (2005). Effect of 0.25 ppm ozone exposure on pulmonary damage
      induced by bleomycin. Biol Res 38: 353-358.
Palancar, GG: Toselli, BM. (2002). Erythemal ultraviolet irradiance in Cordoba, Argentina. Atmos Environ 36:
      287-292.
Palli. D: Sera. F:  Giovannelli. L: Masala. G: Grechi. D: Bendinelli. B: Caini. S: Dolara.  P: Saieva. C. (2009).
      Environmental ozone exposure and oxidative DMA damage in adult residents of Florence, Italy. Environ
      Pollut 157: 1521-1525. http://dx.doi.org/S0269-7491 (08)00472-7 rpii110.1016/i.envpol.2008.09.011.
Pan. YD: Birdsev. R: Horn. J: McCullough. K. (2009). Separating effects of changes in atmospheric composition,
      climate and land-use  on carbon sequestration of US Mid-Atlantic temperate forests. For Ecol Manage
      259: 151-164. http://dx.doi.Org/10.1016/i.foreco.2009.09.049.
Panek, J: Kurpius, MR:  Goldstein, AH. (2002). An evaluation of ozone exposure metrics for a seasonally
      drought-stressed ponderosa pine ecosystem. Environ Pollut 117: 93-100.
      http://dx.doi. org/10.1016/S0269-7491 (01 )00155-5.
Panek. JA: Goldstein. AH. (2001). Responses of stomatal conductance to drought in ponderosa pine:
      Implications for carbon and ozone uptake. Tree Physiol 21: 337-344.
Panek. JA. (2004). Ozone uptake, water loss and carbon exchange dynamics in annually drought-stressed
      Pinus ponderosa forests: Measured trends and parameters for uptake modeling. Tree Physiol 24: 277-
      290.
Draft - Do Not Cite or Quote                        11-62                                 September 2011

-------
Paolacci, AR: Miraldi, C: Tanzarella, OA: Badiani, M: Porceddu, E: Nail, C: Lorenzini, G: Ciaffi, M. (2007). Gene
      expression induced by chronic ozone in the Mediterranean shrub Phillyrea latifolia: Analysis by cDNA-
      AFLP. Tree Physiol 27: 1541-1550. http://dx.doi.Org/10.1093/treephys/27.11.1541.
Paoletti, E: Grulke, NE. (2005). Does living in elevated CO2 ameliorate tree response to ozone? A review on
      stomatal responses [Review]. Environ Pollut 137: 483-493.
      http://dx.doi.0rg/10.1016/i.envpol.2005.01.035.
Paoletti, E: Seufert, G: Delia Rocca, G: Thomsen, H. (2007).  Photosynthetic responses to elevated CO2 and O-
      3 in Quercus ilex leaves at a natural CO2 spring. Environ Pollut 147: 516-524.
      http://dx.doi.0rg/10.1016/i.envpol.2006.08.039.
Paoletti, E: Manning, WJ. (2007). Toward a biologically significant and usable standard for ozone that will also
      protect plants.  Environ Pollut 150: 85-95.  http://dx.doi.Org/10.1016/i.envpol.2007.06.037.
Paoletti, E: Grulke, NE. (2010). Ozone exposure and stomatal sluggishness in different plant physiognomic
      classes. Environ Pollut 158: 2664-2671. http://dx.doi.Org/10.1016/i.envpol.2010.04.024.
Paguette. NC: Zhang. L. -Y:  Ellis. WA: Scott. AL: Kleeberger. SR. (1996). Vitamin A deficiency enhances ozone-
      induced lung injury. Am J Physiol 270: L475-L482.
Park, JW: Taube, C: Swasey, C: Kodama, T: Joetham, A: Balhorn, A: Takeda, K: Miyahara, N: Allen, CB:
      Dakhama. A: Kim. SH: Dinarello. CA: Gelfand. EW. (2004). lnterleukin-1 receptor antagonist attenuates
      airway hyperresponsiveness following exposure to ozone. Am J Respir Cell Mol Biol 30: 830-836.
      http://dx.doi.org/10.1165/rcmb.2003-0373OC.
Park, JW: Lim, YH: Kyung, SY: An, CH: Lee, SP: Jeong, SH: Ju, S, -Y. (2005a). Effects of ambient particulate
      matter on peak expiratory flow rates and respiratory symptoms of asthmatics during Asian dust periods in
      Korea. Respirology 10: 470-476. http://dx.doi.Org/10.1111/i.1440-1843.2005.00728.x.
Park. RJ: Stenchikov. GL: Pickering: Dickerson.  RR: Allen. DJ: Kondragunta. S. (2001). Regional air pollution
      and its radiative forcing: Studies with a single column chemical and radiation transport model. J Geophys
      Res 106: 28,751-728,770.
Park. SK: O'Neill. MS: Vokonas. PS: Sparrow. D: Schwartz. J. (2005b). Effects of air pollution on heart rate
      variability: The VA Normative Aging Study. Environ Health Perspect 113: 304-309.
Park, SK: O'Neill, MS: Stunder, BJB: Vokonas, PS: Sparrow,  D: Koutrakis, P: Schwartz, J. (2007). Source
      location of air pollution and cardiac autonomic function: Trajectory cluster analysis for exposure
      assessment. J Expo Sci Environ Epidemiol 17: 488-497.
Park, SK: O'Neill, MS: Vokonas, PS: Sparrow, D: Wright, RO: Coull, B: Nie, H: Hu, H:  Schwartz, J. (2008). Air
      pollution and heart rate variability: Effect modification by chronic lead exposure. Epidemiology 19: 111-
      120. http://dx.doi.orq/10.1097/EDE.Ob013e31815c408a.
Parker, JD: Akinbami, LJ: Woodruff, TJ. (2009). Air pollution and childhood respiratory allergies in the United
      States. Environ Health Perspect 117: 140-147. http://dx.doi.org/10.1289/ehp.11497.
Parker. JD: Rich. DQ: Glinianaia. SV: Leem. JH: Wartenberg. D: Bell.  ML: Bonzini. M: Brauer. M: Darrow. L:
      Gehring, U: Gouveia, N: Grillo, P: Ha, E: van den Hooven, EH: Jalaludin, B: Jesdale, BM: Lepeule,  J:
      Morello-Frosch, R: Morgan, GG:  Slama, R: Pierik, FH: Pesatori, AC: Sathyanarayana, S: Seo, J:
      Strickland, M: Tamburic, L: Woodruff, TJ. (2011). The international collaboration on air pollution and
      pregnancy outcomes: Initial results. Environ Health Perspect 119:  1023-1028.
      http://dx.doi.org/10.1289/ehp.1002725.
Parrish.  DP.  (2006). Critical evaluation of US on-road vehicle emission inventories.  Atmos Environ 40: 2288-
      2300.
Parrish.  DP:  Millet. DB: Goldstein. AH. (2009). Increasing ozone in marine boundary layer inflow at the west
      coasts of North America and Europe. Atmos Chem Phys 9: 1303-1323. http://dx.doi.org/10.5194/acpd-8-
      13847-2008.
Parsons, WFJ: Bockheim, JG: Lindroth, RL. (2008). Independent, interactive, and species-specific responses of
      leaf litter decomposition to elevated CO2  and O3  in a northern  hardwood forest. Ecosystems 11: 505-
      519.
Passannante. AN: Hazucha. MJ: Bromberg. PA:  Seal. E: Folinsbee. L: Koch. G. (1998). Nociceptive
      mechanisms modulate ozone-induced human lung function decrements. J Appl Physiol 85: 1863-1870.
Paulu, C: Smith, AE. (2008). Tracking associations between ambient ozone and asthma-related emergency
      department visits using case-crossover analysis. J Public Health Manag Pract 14: 581-591.
Pavelin. EG:  Johnson. CE: Rughooputh. S: Toumi. R. (1999). Evaluation of pre-industrial surface ozone
      measurements made using Schonbein's method. Atmos Environ 33: 919-929.
      http://dx.doi. org/10.1016/51352-2310(98)00257-X.
Paz, C: Bazan-Perkins, B. (1992). Sleep-wake disorganization in cats exposed to ozone. Neurosci Lett 140:
      270-272.
Paz. C: Huitron-Resendiz. S. (1996). The effects of ozone exposure on the sleep-wake cycle and serotonin
      contents in the pons of the rat. Neurosci Lett 204: 49-52.
Peacock, JL: Anderson,  HR: Bremner, SA: Marston, L: Seemungal, TA: Strachan, DP: Wedzicha, JA. (2011).
      Outdoor air pollution and respiratory health in patients with COPD. Thorax 66: 591-596.
      http://dx.doi.org/10.1136/thx.2010.155358.
Draft - Do Not Cite or Quote                        11-63                                  September 2011

-------
Pearson, AC: Bhalla, DK. (1997). Effects of ozone on macrophage adhesion in vitro and epithelial and
      inflammatory responses in vivo: The role of cytokines. J Toxicol Environ Health 50: 143-157.
Pearson, M: Mansfield, TA. (1993). Interacting effects of ozone and water stress on the stomatal resistance of
      beech (Fagus sylvatica L). New Phytol 123: 351-358. http://dx.doi.org/10.1111/i. 1469-
      8137.1993.tb03745.x.
Pearson. S: Davison. AW: Reiling. K: Ashenden. T: Ollerenshaw. JH. (1996). The effects of different ozone
      exposures on three contrasting populations of Plantago major. New Phytol 132: 493-502.
      http://dx.doi.0rg/10.1111/i.1469-8137.1996.tb01869.x.
Peden. DB: Setzer. RW. Jr: Devlin. RB.  (1995). Ozone exposure has both a priming effect on allergen-induced
      responses and an intrinsic inflammatory action in the nasal airways of perennially allergic asthmatics. Am
      J Respir Grit Care Med 151: 1336-1345.
Peden, DB: Boehlecke, B: Horstman, D: Devlin, R. (1997). Prolonged acute exposure to 0.16 ppm ozone
      induces eosinophilic airway inflammation in asthmatic subjects with allergies. J Allergy Clin Immunol 100:
      802-808.
Peden. DB. (2001). Air pollution in asthma: Effect of pollutants on airway inflammation. Ann Allergy Asthma
      Immunol 3: 12-17.
Peden. DB. (2011). The role of oxidative stress and innate immunity in O(3) and endotoxin-induced human
      allergic airway disease. Immunol Rev 242: 91-105. http://dx.doi.Org/10.1111/i.1600-065X.2011.01035.x.
Peel, Jl_: Metzger, KB: Klein,  M: Flanders, WD: Mulholland, JA: Tolbert, PE. (2007). Ambient air pollution and
      cardiovascular emergency department visits in potentially sensitive groups. Am  J Epidemiol 165: 625-
      633.
Peeters, J: Nguyen, Tl_: Vereecken, L. (2009). HOx radical regeneration in the oxidation of isoprene. Phys
      Chem Chem Phys 11:  5935. http://dx.doi.ora/10.1039/B908511D.
Peeters. J: Mullen J. -F. (2010). HOx radical regeneration in isoprene oxidation via peroxy radical
      isomerisations. II: Experimental evidence and global impact. Phys Chem Chem Phys 12:  14227-14235.
      http://dx.doi.ora/10.1039/COCP00811G.
Pel. Z. -M: Murata. Y: Benning. G: Thomine. S: Klusener. B: Allen. GJ: Grill. E: Schroeder. Jl. (2000). Calcium
      channels activated by  hydrogen peroxide mediate abscisic acid signalling in guard cells. Nature 406:
      731-734.  http://dx.doi.org/10.1038/35021067.
Pell, EJ: Sinn, JP: Brendley, BW: Samuelson, L: Vinten-Johansen, C: Tien, M: Skillman, J. (1999). Differential
      response of four tree species to ozone-induced acceleration of foliar senescence. Plant Cell Environ 22:
      779-790.  http://dx.doi.org/10.1046/i. 1365-3040.1999.00449.x.
Pellegrino. R: Viegi. G: Brusasco. V: Crapo. RO: Burgos. F: Casaburi. R: Coates. A: van der Grinten. CP:
      Gustafsson, P: Hankinson, J: Jensen, R: Johnson, DC: Maclntyre, N:  McKay, R: Miller, MR: Navajas, D:
      Pedersen. OF: Wanger. J. (2005). Interpretative strategies for lung function tests. Eur Respir J 26: 948-
      968. http://dx.doi.org/10.1183/09031936.05.00035205.
Peltonen, PA: Julkunen-Tiitto, R: Vapaavuori, E: Holopainen, JK. (2006).  Effects of elevated carbon dioxide and
      ozone on aphid oviposition preference and birch bud exudate phenolics. Global Change Biol 12: 1670-
      1679. http://dx.doi.0rg/10.1111/i.1365-2486.2006.01226.x.
Peltonen, PA: Vapaavuori, E:  Heinonen, J: Julkunen-Tiitto, R:  Holopainen, JK. (2010). Do elevated atmospheric
      CO2 and O-3 affect food quality and performance of folivorous insects on silver birch? Global Change
      Biol 16: 918-935. http://dx.doi.Org/10.1111/i.1365-2486.2009.02073.x.
Peluso M Hainaut, P: Airoldi,  L: Autrup,  H: Dunning, A: Garte,  S: Gormallv, E:  Malaveille, C: Matullo, G:
      Munniaa. A: Riboli. E: investigators. VPE. (2005). Methodology of laboratory measurements in
      prospective studies on gene-environment interactions:  The experience of GenAir. DNA Repair 574: 92-
      104.
Penard-Morand, C: Charpin, D: Raherison, C:  Kopferschmitt, C: Caillaud, D:  Lavaud, F: Annesi-Maesano, I.
      (2005). Long-term exposure to background air pollution related to respiratory and allergic health in
      schoolchildren. Clin Exp Allergy 35: 1279-1287.
Peng. RD: Dominici. F: Pastor-Barriuso.  R: Zeger. SL: Samet. JM.  (2005). Seasonal analyses of air pollution
      and mortality in 100 US cities. Am J Epidemiol 161: 585-594.
Percy, KE: Nosal,  M: Heilman, W: Dann, T: Sober, J: Legge, AH: Karnosky, PR (2007).  New exposure-based
      metric approach for evaluating O3 risk to North American aspen forests. Environ Pollut 147: 554-566.
      http://dx.doi. org/10.1016/i.envpol.20Q6.10.009.
Pereira, FAC: De Assuncao, JV: Saldiva, PHN: Pereira,  LAA:  Mirra, AP:  Braga, ALF. (2005). Influence  of air
      pollution on the incidence of respiratory tract neoplasm. J Air Waste Manag Assoc 55:  83-87.
Pereira, LAA: Loomis, D: Conceicao, QMS: Braga, ALF: Areas, RM: Kishi, HS: Singer,  JM: Bohm, GM: Saldiva,
      PHN. (1998). Association between air pollution and intrauterine mortality in Sao Paulo, Brazil. Environ
      Health Perspect 106: 325-329.
Perepu. RS: Garcia. C: Postal. D: Sethi. R. (2010). Enhanced death signaling in ozone-exposed ischemic-
      reperfused hearts. Mol Cell Biochem 336: 55-64.  http://dx.doi.org/10.1007/s11010-009-0265-4.
Perevra-Munoz. N: Rugerio-Vargas. C: Angoa-Perez. M: Borgonio-Perez. G: Rivas-Arancibia. S. (2006).
      Oxidative damage in substantia nigra and striatum of rats chronically exposed to ozone. J Chem
      Neuroanat31: 114-123. http://dx.doi.Org/10.1016/i.ichemneu.2005.09.006.
Draft - Do Not Cite or Quote                        11-64                                 September 2011

-------
Perez-Gil, J. (2008). Structure of pulmonary surfactant membranes and films: The role of proteins and lipid-
      protein interactions. Biochim Biophys Acta 1778: 1676-1695.
      http://dx.doi.0rg/10.1016/i.bbamem.2008.05.003.
Perring, AE:  Bertram, TH: Wooldridge, PJ: Fried, A: Heikes, BG: Dibb, J: Crounse, JD: Wennberg,  PO: Blake,
      NJ: Blake. PR: Brune.  WH: Singh. HB: Cohen. RC. (2009). Airborne observations of total RONO2: New
      constraints on the yield and lifetime of isoprene nitrates. Atmos Chem Phys 9: 1451-1463.
Peters, A: Dockery,  DW: Muller, JE: Mittleman, MA. (2001). Increased particulate air pollution and the triggering
      of myocardial infarction. Circulation 103: 2810-2815.
Peters. JM: Avol. E: Navidi. W: London. SJ: Gauderman. WJ: Lurmann. F: Linn. WS: Margolis. H: Rappaport. E:
      Gong, H, Jr:  Thomas, DC. (1999a). A study of twelve southern California communities with differing
      levels and types of air  pollution I  Prevalence of respiratory morbidity. Am J Respir Crit Care  Med 159:
      760-767.
Peters, JM: Avol, E: Gauderman, WJ: Linn, WS: Navidi, W: London, SJ: Margolis, H: Rappaport, E: Vora, H:
      Gong. H. Jr:  Thomas. DC. (1999b). A study of twelve southern California communities with differing
      levels and types of air  pollution II Effects on pulmonary function. Am J Respir Crit Care Med 159: 768-
      775.
Peterson. DC: Andrews. HL. (1963). The role of ozone in radiation avoidance in the mouse. Radiat Res 19: 331-
      336.
Peterson, PL: Arbaugh,  MJ: Wakefield, VA: Miller, PR. (1987). Evidence of growth reduction in ozone-injured
      Jeffrey pine (Pinus Jeffrey! Grev and Balf) in Sequoia and Kings Canyon National Parks. J Air Waste
      Manag Assoc 37: 906-912.
Petroeschevskv, A:  Simpson,  RW: Thalib, l_: Rutherford, S. (2001). Associations between outdoor air pollution
      and hospital  admissions in Brisbane, Australia. Arch Environ Occup Health 56: 37-52.
Petruzzi.  S: Fiore. M: Dell'Omo. G: Bignami. G: Alleva. E. (1995). Medium and long-term behavioral effects in
      mice of extended gestational exposure to ozone. Neurotoxicol Teratol 17: 463-470.
Petruzzi.  S: De Acetis. L: Chiarotti. F: Sorace. A: Alleva. E.  (1999). Limited changes in handedness and
      morphine reactivity in CD-1 mice after pre- and postnatal  ozone exposure. Acta Neurobiol Exp (Wars) 59:
      115-122.
Pfister, G: Hess, PG: Emmons, LK: Lamarque, JF: Wiedinmyer, C: Edwards, DP: Petron,  G: Gille, JC: Sachese,
      GW. (2005).  Quantifying CO emissions  from the 2004 Alaskan wildfires using  MOPITT CO data.
      Geophys Res Lett 32:  L11809.
Pfister. H. (2003). Human papillomavirus and skin cancer. J Natl Cancer Inst Monographs No. 31: 52-56.
Pfleeger.  TG: Plocher. M: Bichel. P. (2010). Response of pioneer plant communities to elevated ozone
      exposure. Agric Ecosyst Environ 138: 116-126.  http://dx.doi.Org/10.1016/i.agee.2010.04.009.
Philipona. R: Behrens. K: Ruckstuhl. C. (2009). How declining aerosols and rising greenhouse gases forced
      rapid warming in  Europe since the 1980s. Geophys  Res Lett 36: L02806.
      http://dx.doi. org/10.1029/2008GL036350.
Phillips, PL: Johnson, MG: Tingey, DT: Storm,  MJ. (2009).  Elevated CO2 and O3 effects on fine-root
      survivorship  in ponderosa pine mesocosms. Oecologia 160: 827-837. http://dx.doi.org/10.1007/s00442-
      009-1339-4.
Phillips. RL: Zak. PR: Holmes. WE: White.  DC. (2002). Microbial community composition and function beneath
      temperate trees exposed to elevated atmospheric carbon dioxide and ozone. Oecologia 131: 236-244.
Phoenix,  GK: Gwynn-Jones, D:  Lee, JA: Callaghan, TV. (2000). The impacts of UV-B radiation on the
      regeneration of a sub-arctic heath community. Plant Ecol  146: 67-75.
      http://dx.doi. org/10.1023/A: 1009839506658.
Pichavant, M: Goya, S: Meyer, EH: Johnston, RA: Kim, HY: Matangkasombut, P: Zhu, M: Iwakura, Y: Savage,
      PB: DeKruyff, RH: Shore, SA:  Umetsu,  DT. (2008). Ozone exposure in a mouse model induces airway
      hyperreactivity that requires the presence of natural killer T cells and  IL-17. J Exp Med 205: 385-393.
      http://dx.doi. org/10.1084/jem.20071507.
Picher. M: Burch. LH: Boucher. RC. (2004). Metabolism of  P2 receptor agonists in human airways: Implications
      for mucociliary clearance and cystic fibrosis. J Biol Chem 279: 20234-20241.
      http://dx.doi. org/10.1074/jbc. M400305200.
Pickett. JE: Gibson. DA: Gardner. MM. (2008). Effects of irradiation conditions on the weathering of engineering
      thermoplastics. Polym  Degrad Stabil 93: 1597-1606.
      http://dx.doi.0rg/10.1016/i.polymdegradstab.2008.02.009.
Piikki, K:  Vorne, V: Ojanpera,  K: Pleijel,  H. (2007). Impact of elevated O-3 and CO2 exposure on potato
      (Solanum tuberosum L. cv. Bintje) tuber macronutrients (N, P,  K, Mg,  Ca). Agric Ecosyst Environ 118:  55-
      64. http://dx.doi.0rg/10.1016/i.agee.2006.04.012.
Piikki. K:  De Temmerman. L: Oianpera.  K: Danielsson. H: Pleiiel. H. (2008a). The grain quality of spring wheat
      (Triticum aestivum L.) in relation to elevated ozone uptake and carbon dioxide exposure. Eur J Agron  28:
      245-254. http://dx.doi.0rg/10.1016/i.eia.2007.07.004.
Piikki. K:  De Temmerman. L: Hogy. P: Pleiiel. H. (2008b). The open-top chamber impact on vapour pressure
      deficit and its consequences for stomatal ozone uptake. Atmos Environ 42: 6513-6522.
      http://dx.doi.0rg/10.1016/i.atmosenv.2008.04.014.
Draft - Do Not Cite or Quote                        11-65                                  September 2011

-------
Pinkerton, KE: Brody, AR: Miller,  FJ: Crapo, JD. (1989). Exposure to low levels of ozone results in enhanced
      pulmonary retention of inhaled asbestos fibers. Am J Respir Grit Care Med 140: 1075-1081.
Pinkerton, KE: Menache, MG: Plopper, CG. (1995). Consequences of prolonged inhalation of ozone on F344/N
      rats: Collaborative studies Part IX Changes in the tracheobronchial epithelium, pulmonary acinus, and
      lung antioxidant enzyme activity.
Pinkerton. KE: Weller. BL: Menache. MG: Plopper. CG. (1998). Consequences of prolonged inhalation of ozone
      on F344/N rats: Collaborative studies Part XIII A comparison of changes in the tracheobronchial
      epithelium and pulmonary acinus in male rats at 3 and 20 months.
Pinto. DM: Blande. JD: Nvkanen. R: Dong. WX: Nerg. AM: Holopainen. JK. (2007a). Ozone degrades common
      herbivore-induced plant volatiles: Does this affect herbivore prey location by predators and parasitoids? J
      Chem Ecol 33: 683-694. http://dx.doi.Org/10.1007/s10886-007-9255-8.
Pinto, DM: Nerg, AM: Holopainen, JK. (2007b). The role of ozone-reactive compounds, terpenes, and green leaf
      volatiles (GLVs), in the orientation of Cotesia plutellae. J Chem Ecol 33: 2218-2228.
      http://dx.doi. orq/10.1007/s10886-007-9376-0.
Pinto. DM: Himanen. SJ: Nissinen. A: Nerg. AM: Holopainen. JK. (2008).  Host location behavior of Cotesia
      plutellae Kurdjumov (Hymenoptera: Braconidae) in ambient and moderately  elevated ozone in field
      conditions. Environ Pollut  156: 227-231. http://dx.doi.Org/10.1016/i.envpol.2007.12.009.
Pinto. DM: Blande. JD: Souza. SR: Nerg. AM: Holopainen. JK. (2010). Plant volatile organic compounds (VOCs)
      in ozone (O-3) polluted atmospheres: The ecological effects. J Chem Ecol 36: 22-34.
      http://dx.doi. org/10.1007/s10886-009-9732-3.
Pinto, J. (2009). Wyoming winter smog. Nat Geosci 2: 88-90. http://dx.doi.org/10.1038/ngeo430.
Pinto, JP: Lefohn, AS: Shadwick, PS. (2004). Spatial variability of PM2.5  in urban areas in the United States. J
      Air Waste Manag Assoc 54: 440-449.
Pleiiel. H: Oianpera.  K: Mortensen. L. (1997). Effects of tropospheric ozone on the yield and grain protein
      content of spring wheat (Triticum aestivum L) in the nordic countries. Acta Agric Scand B Soil Plant Sci
      47: 20-25. http://dx.doi.orq/10.1080/09064719709362434.
Pleiiel. H: Danielsson. H: Gelang. J: Slid. E: Sellden. G. (1998). Growth stage dependence of the grain yield
      response to ozone in spring wheat (Triticum aestivum L). Agric Ecosyst Environ 70: 61-68.
      http://dx.doi.org/10.1016/S0167-8809(97)00167-9.
Pleiiel, H: Danielsson, H: Oianpera, K: De Temmerman, L: Hogv, P: Badiani, M:  Karlsson,  PE.  (2004a).
      Relationships between ozone exposure and yield loss in European wheat and potato—a comparison of
      concentration- and flux-based exposure indices. Atmos Environ 38: 2259-2269.
Pleiiel. H: h. D: Oianpera. K: De Temmerman. L: Hogy. P. (2004b).  Relationships between ozone exposure and
      yield loss in wheat and potato - Suggestions of critical levels for ozone effects on crops. Atmos Environ
      38: 2259-2269. http://dx.doi.orq/10.1016/i.atmosenv.2003.09.076.
Pleis. JR: Lucas. JW: Ward. BW. (2009). Summary health statistics for U.S. adults:  National Health Interview
      Survey, 2008. In Vital and Health Statistics, 10 (Vol. 242). (DHHS 2010-1570). Hyattsville, MD: National
      Center for Health Statistics.
Plessl, M: Elstner, EF: Rennenberg, H: Habermeyer, J: Heiser,  I. (2007). Influence of elevated  CO2 and ozone
      concentrations on late blight resistance and growth of potato plants. Environ Exp Bot 60: 447-457.
      http://dx.doi.0rg/10.1016/i.envexpbot.2007.01.003.
Plochl. M: Lyons. T: Ollerenshaw. J: Barnes. J. (2000). Simulating ozone  detoxification in the leaf apoplast
      through the direct reaction with ascorbate. Planta 210: 454-467. http://dx.doi.org/10.1007/PL00008153.
Plopper. CG: Chu. F. -P: Haselton. CJ: Peake. J: Wu. J: Pinkerton. KE. (1994). Dose-dependent tolerance to
      ozone: I tracheobronchial  epithelial reorganization in rats after 20 months' exposure. Am J Pathol 144:
      404-420.
Plopper, CG: Hatch,  GE: Wong, V: Duan, X: Weir, AJ: Tarkington, BK: Devlin, RB: Becker, S: Buckpitt, AR.
      (1998). Relationship of inhaled ozone concentration to acute tracheobronchial epithelial injury, site-
      specific ozone dose and glutathione depletion in rhesus monkeys. Am J Respir Cell Mol Biol 19:  387-
      399.
Plopper. CG: Mango. GW: Hatch. GE: Wong. VJ: Toskala. E: Reynolds. SD: Tarkington. BK: Stripp. BR. (2006).
      Elevation of susceptibility to ozone-induced acute tracheobronchial injury in transgenic mice deficient in
      Clara cell secretory protein. Toxicol Appl Pharmacol 213: 74-85.
      http://dx.doi.0rg/10.1016/i.taap.2005.09.003.
Plopper, CG: Smiley-Jewell, SM: Miller, LA: Fanucchi, MV: Evans,  MJ: Buckpitt, AR: Avdalovic, M: Gershwin,
      LJ: Joad, JP:  Kajekar, R: Larson, S: Pinkerton, KE: Van Winkle, LS: Schelegle, ES: Pieczarka, EM: Wu,
      R: Hyde,  DM. (2007). Asthma/allergic airways disease: Does postnatal exposure to environmental
      toxicants promote airway pathobiology? Toxicol Pathol 35: 97-110.
      http://dx.doi. org/10.1080/01926230601132030.
Pokharel. SS: Bishop. GA: Stedman. PH. (2002). An on-road motor vehicle emissions inventory for Denver: An
      efficient alternative to modeling. Atmos Environ 36: 5177-5184.
Pokharel. SS: Bishop. GA: Stedman. PH. (2003). Emissions reductions as a result of automobile improvement.
      Environ Sci Technol 37: 5097-5101.
Draft - Do Not Cite or Quote                        11-66                                  September 2011

-------
Pollack, AK: Lindhjem, C: Stoeckenius, TE: Iran, C: Mansell, G: Jimenez, M: Wilson, G: Coulter-Burke, S.
      (2004). Final Report: Evaluation of the US EPA MOBILE6 highway vehicle emission factor model. (CRC
      Project E-64). Novato, CA: ENVIRON International Corporation.
Pollastrini, M: Desotgiu, R: Cascio, C:  Bussotti, F: Cherubini, P: Saurer, M: Gerosa, G: Marzuoli, R. (2010).
      Growth and physiological responses to ozone and mild drought stress of tree species with different
      ecological requirements. Trees  Struct Funct 24: 695-704. http://dx.doi.orq/10.1007/s00468-010-0439-4.
Polle, A: Pell, EJ. (1999). Role of carbon dioxide in modifying the plant response to ozone. In Y Luo; HA Mooney
      (Eds.), Carbon dioxide and environmental stress (pp. 193-213). San Diego, CA: Academic Press.
Poloniecki. JD: Atkinson. RW: Ponce de  Leon. A: Anderson. HR. (1997). Daily time series for cardiovascular
      hospital admissions and previous day's air pollution in London, UK. Occup Environ Med 54: 535-540.
Ponsonbv, Al_:  McMichael, A: Van der  Mei, I. (2002). Ultraviolet radiation and autoimmune disease: Insights
      from epidemiological research.  Toxicology 181/182: 71-78.
Pope CA, III: Burnett, RT: Thun, MJ: Calle, EE: Krewski, D: Ito, K: Thurston, GD. (2002). Lung cancer,
      cardiopulmonary mortality, and  long-term exposure to fine particulate air pollution. JAMA 287: 1132-1141.
Poppe. D: Wallasch. M: Zimmermann.  J. (1993). The dependence of the concentration of OH on its precursors
      under moderately polluted conditions: A model study. J Atmos Chem 16: 61-78.
Postlethwait. EM: Langford. SD: Bidani. A. (1994).  Determinants of inhaled ozone absorption in isolated rat
      lungs. Toxicol Appl Pharmacol 125: 77-89.
Postlethwait, EM: Cueto, R: Velsor, LW: Pryor, WA. (1998). OS-induced formation of bioactive lipids: Estimated
      surface concentrations and lining  layer effects. Am J Physiol 274: L1006-L1016.
Postlethwait, EM: Joad, JP: Hyde, DM: Schelegle,  ES: Brie, JM: Weir, AJ: Putney, LF: Wong, VJ: Velsor, LW:
      Plopper, CG. (2000). Three-dimensional mapping of ozone-induced acute cytotoxicity in tracheobronchial
      airways of isolated perfused rat lung. Am J Respir Cell  Mol Biol 22: 191-199.
Postlethwait. EM: Ultman. JS. (2001). Airspace surface chemistry mediates OS-induced lung injury. Hum Ecol
      Risk Assess 7: 1145-1159. http://dx.doi. org/10.1080/20018091094907.
Pregitzer. K: Lova. W: Kubiske. M: Zak. D. (2006).  Soil respiration in northern forests exposed to elevated
      atmospheric carbon dioxide and ozone. Oecologia 148: 503-516. http://dx.doi.org/10.1007/s00442-006-
      0381-8.
Pregitzer, KS: Burton, AJ: King, JS: Zak,  DR. (2008). Soil respiration,  root biomass, and root turnover following
      long-term exposure of northern  forests to elevated atmospheric Co-2 and tropospheric O-3. New Phytol
      180: 153-161. http://dx.doi.Org/10.1111/i.1469-8137.2008.02564.x.
Prescott. GJ: Cohen. GR: Elton. RA: Fowkes. FGR: Agius. RM. (1998). Urban air pollution and cardiopulmonary
      ill health: A 145 year time series study. Occup Environ Med 55: 697-704.
Pretzsch, H: Dieler, J: Matyssek, R: Wipfler,  P. (2010). Tree and stand growth of mature Norway spruce and
      European  beech under long-term  ozone fumigation. Environ Pollut 158:  1061-1070.
      http://dx.doi.0rg/10.1016/i.envpol.2009.07.035.
Pritsch,  K: Esperschuetz, J: Haesler, F: Raidl,  S: Winkler, B: Schloter, M.  (2009). Structure and activities of
      ectomycorrhizal and microbial communities  in the rhizosphere  of Fagus sylvatica under ozone and
      pathogen stress in a lysimeter study.  Plant Soil 323: 97-109. http://dx.doi.org/10.1007/s11104-009-9972-
      6.
Prvor. WA. (1976). Free radical reactions in biology: Initiation of lipid autoxidation by ozone and nitrogen
      dioxide. Environ Health Perspect 16:  180-181.
Prvor, WA: Giamalva, DH: Church, DF. (1984). Kinetics of ozonation. 2. Amino acids and model compounds in
      water and  comparisons to rates in nonpolar  solvents. J Am Chem Soc 106: 7094-7100.
Prvor. WA: Das. B: Church. DF. (1991). The ozonation of unsaturated fatty acids: Aldehydes and hydrogen
      peroxide as products and possible mediators of ozone toxicity.  Chem Res Toxicol 4: 341-348.
Pryor, WA. (1992). How far does ozone penetrate into the pulmonary  air/tissue boundary before it reacts? Free
      Radic Biol Med 12: 83-88. http://dx.doi.org/10.1016/0891-5849(92)90060-T.
Prvor, WA. (1994). Mechanisms of radical formation from reactions of ozone with target molecules in the lung.
      Free Radic Biol Med 17: 451-465.
Prvor. WA: Bermudez. E: Cueto. R: Sguadrito. GL.  (1996). Detection of aldehydes in bronchoalveolar lavage of
      rats exposed to ozone. Toxicol Sci 34: 148-156.
Puckette. MC: Tang. YH: Mahalingam.  R. (2008). Transcriptomic changes induced by acute ozone in resistant
      and sensitive  Medicago truncatula accessions. BMC Plant Biol 8: 46. http://dx.doi.org/10.1186/1471-
      2229-8-46.
Pujol Pereira, El:  Chung, H: Scow, K: Sadowskv, MJ: van Kessel, C: Six, J. (2011).  Soil nitrogen transformations
      under elevated atmospheric CO2 and O3 during the soybean growing season. Environ Pollut 159: 401-
      407. http://dx.doi.0rg/10.1016/i.envpol.2010.10.033.
Pulfer. MK: Murphy. RC. (2004). Formation of biologically active oxysterols during ozonolysis of cholesterol
      present in  lung surfactant. J Biol Chem  279: 26331-26338.
Pulfer, MK: Taube, C: Gelfand, E: Murphy, RC. (2005). Ozone exposure in vivo and formation of biologically
      active oxysterols in the lung. J Pharmacol Exp Ther 312: 256-264.
Qian. Z: Liao. D: Lin. H. -M: Whitsel. EA: Rose. KM: Duan. Y. (2005). Lung function and long-term exposure to
      air pollutants in middle-aged American adults. Arch Environ Occup Health 60: 156-163.
Draft - Do Not Cite or Quote                        11-67                                  September 2011

-------
Qian, Z: Lin, H, -M: Chinchilli, VM: Lehman, EB: Duan, Y: Craig, TJ: Wilson, WE: Liao, D: Lazarus, SC:
      Bascom, R. (2009). Interaction of ambient air pollution with asthma medication on exhaled nitric oxide
      among asthmatics. Arch Environ Occup Health 64: 168-176.
      http://dx.doi. org/10.1080/19338240903240616.
Que. LG: Stiles. JV: Sundv. JS: Foster. WM. (2011). Pulmonary function, bronchial reactivity, and epithelial
      permeability are response phenotypes to ozone and develop differentially in healthy humans. J Appl
      Physiol 111: 679-687. http://dx.doi.org/10.1152/iapplphysiol.00337.2011.
Rabinovitch. N: Zhang. LN: Murphy. JR: Vedal. S: Dutton. SJ: Gelfand. EW. (2004). Effects of wintertime
      ambient air pollutants on asthma exacerbations in urban minority children with moderate to severe
      disease. J Allergy Clin Immunol 114: 1131-1137.  http://dx.doi.Org/10.1016/i.iaci.2004.08.026.
Rage, E: Siroux, V: Kunzli, N: Pin, I: Kauffmann,  F. (2009a). Air pollution and asthma severity in adults. Occup
      Environ Med 66:  182-188. http://dx.doi.org/10.1136/oem.2007.038349.
Rage, E: Jacguemin, B: Nadif, R: Orvszczyn, MP: Siroux, V: Aguilera, I: Kauffmann, F: Kunzli, N. (2009b). Total
      serum IgE levels are associated with ambient ozone concentration in asthmatic adults. Allergy 64: 40-46.
Raizenne. M: Stern. B: Burnett. R: Spengler. J. (1987). Acute respiratory function and transported air pollutants:
      Observational studies (paper no. 87-32.6). In Proceedings of the 80th Annual Meeting of the Air Pollution
      Control Association (pp. 18). New York, NY: Air Pollution Control Association.
Raizenne. ME: Burnett.  RT: Stern. B: Franklin. CA: Spengler. JD. (1989). Acute lung function responses to
      ambient acid aerosol exposures in children. Environ Health Perspect 79:  179-185.
Ramadour, M: Burel, C: Lanteaume, A: Vervloet,  D: Charpin, D: Brisse, F: Dutau, H. (2000).  Prevalence of
      asthma and rhinitis in relation to long-term exposure to gaseous air pollutants. Allergy 55: 1163-1169.
Ramfrez-Aguilar, M: Barraza-Villarreal, A: Moreno-Macfas, H: Winer, AM:  Cicero-Fernandez, P: Velez-Marguez,
      MG: Cortez-Lugo. M:  Sienra-Monge. JJ: Romieu. I. (2008). Assessment of personal exposure to ozone in
      asthmatic children residing in Mexico City. Salud Publica Mex 50: 67-75.
Ramo,  K: Kanerva, T: Ojanpera, K: Manninen, S. (2007). Growth onset, senescence,  and  reproductive
      development of meadow species in mesocosms exposed to elevated O3 and CO2. Environ Pollut 145:
      850-860. http://dx.doi.0rg/10.1016/i.envpol.2006.03.054.
Rao, MV: Hale, BA: Ormrod, DP.  (1995). Amelioration of ozone-induced oxidative damage in wheat plants
      grown under high carbon dioxide: Role of antioxidant enzymes. J Plant Physiol 109: 421-432.
Rao, ST: Ku, J, -Y: Berman,  S: Zhang, K: Mao, H. (2003). Summertime characteristics of the atmospheric
      boundary layer and relationships to ozone levels over the eastern United States. Pure Appl Geophys
      160: 21-55.
Rappengluck. B: Dasgupta. PK: Leuchner. M: Li. Q: Luke. W. (2009). Formaldehyde and its relation to CO,
      PAN, and SO2 in the Houston-Galveston airshed. Atmos Chem Phys Discuss 9: 24193-24223.
      http://dx.doi. org/10.5194/acp-10-2413-2010.
Rapport. DJ: Whitford. WG. (1999). How ecosystems respond to stress: Common properties of arid and aquatic
      systems. Bioscience 49: 193-203.
Rashba-Step, J: Tatoyan, A:  Duncan, R: Ann, D:  Pushpa-Rehka, TR: Sevanian, A. (1997).  Phospholipid
      peroxidation induces cytosolic phospholipase A2 activity: Membrane effects versus enzyme
      phosphorylation. Arch Biochem Biophys 343: 44-54. http://dx.doi.org/10.1006/abbi. 1997.0134.
Rastigeiev. Y: Park. R: Brenner. MP: Jacob. DJ. (2010). Resolving intercontinental pollution plumes in global
      models of atmospheric transport. J Geophys Res 115: D02302. http://dx.doi.org/10.1029/2009JD012568.
Rawlings, JO: Cure, WW. (1985). The Weibull function as a dose-response model to describe ozone effects on
      crop yields. Crop Sci 25: 807-814.
Reeser. WH: Lee. GM: Taylor. A: Wang. L: Arnold. SF: Ultman. JS: Ben-Jebria. A. (2005). Uptake of ozone in
      human lungs and  its relationship to local physiological response. Inhal Toxicol 17: 699-707.
      http://dx.doi. org/10.1080/08958370500224433.
Reich, PB: Lassoie, JP (1984). Effects of low level O3 exposure on  leaf diffusive conductance and water-use
      efficiency in hybrid poplar. Plant Cell Environ 7: 661-668. http://dx.doi.org/10.1111/1365-
      304Q.ep11571645.
Reich. PB. (1987). Quantifying plant response to ozone: A unifying theory. Tree Physiol 3:  63-91.
      http://dx.doi. org/10.1093/treephys/3.1.63.
Reid. CD: Fiscus. EL. (2008). Ozone and density affect the response of biomass and seed yield to elevated
      CO2 in rice. Global Change Biol 14: 60-76. http://dx.doi.Org/10.1111/i.1365-2486.2007.01472.x.
Reid, N: Yap, D: Bloxam, R. (2008). The potential role of background ozone on current and emerging air issues:
      An overview. Air Qual Atmos Health 1: 19-29. http://dx.doi.org/10.1007/s11869-008-0005-z.
Reidmiller, PR: Fiore, AM: Jaffe, DA: Bergmann,  D: Cuvelier, C: Dentener, FJ: Duncan: Bryan, N: Folberth, G:
      Gauss, M: Gong, S: Hess, P: Jonson, JE:  Keating, T: Lupu, A: Marmer, E: Park, R:  Schultz, MG:
      Shindell. DT: Szopa. S: Vivanco. MG: Wild.  O: Zuber. A.  (2009). The influence of foreign vs. North
      American emissions on surface ozone in the US. Atmos Chem Phys 9: 5027-5042.
Reiling,  K: Davison, AW.  (1992). Effects of a short ozone exposure given at different stages in the development
      of Plantago major L. New Phytol 121: 643-647. http://dx.doi.Org/10.1111/i.1469-8137.1992.tb01135.x.
Draft - Do Not Cite or Quote                        11-68                                  September 2011

-------
Reiling, K: Davison, AW. (1994). Effects of exposure to ozone at different stages in the development of Plantago
      major L on chlorophyll fluorescence and gas exchange. New Phytol 128: 509-514.
      http://dx.doi.0rg/10.1111/i.1469-8137.1994.tb02998.x.
Reinert, RA: Ho, MC. (1995). Vegetative growth of soybean as affected by elevated carbon dioxide and ozone.
      Environ Pollut 89: 89-96. http://dx.doi.orq/10.1016/0269-7491 (94)00039-G.
Reinert. RA: Eason.  G: Barton. J. (1997). Growth and  fruiting of tomato as influenced by elevated carbon
      dioxide and ozone.  New Phytol 137: 411-420. http://dx.doi.Org/10.1046/1.1469-8137.1997.00846.x.
Reiser. KM: Tyler. WS: Hennessv. SM: Dominguez. JJ: Last. JA. (1987). Long-term consequences of exposure
      to ozone: II. Structural alterations in lung collagen of monkeys. Toxicol Appl Pharmacol 89: 314-322.
      http://dx.doi. org/10.1016/0041 -008X(87)90151-7.
Reisinger, AR. (2000). Unidentified interference in DOAS measurements  of ozone. Appl Spectres Rev 54: 72-
      79.
Reiss, R: Ryan, PB:  Tibbetts, SJ: Koutrakis, P. (1995a). Measurement of  organic acids, aldehydes, and ketones
      in residential environments and their relation to ozone.  J Air Waste Manag Assoc 45: 811-822.
Reiss. R: Ryan. PB:  Koutrakis. P: Tibbetts. SJ. (1995b). Ozone reactive chemistry on interior latex paint.
      Environ Sci Technol 29: 1906-1912.
Ren. C: Williams. GM: Mengersen. K: Morawska. L: long. S. (2008). Does temperature modify short-term
      effects of ozone on total mortality in 60 large eastern US communities? An assessment using the
      NMMAPS data. Environ Int 34: 451-458.
Ren, W: Tian, HQ: Liu, ML: Zhang, C: Chen, GS: Pan, SF: Felzer, B: Xu,  XF. (2007a). Effects of tropospheric
      ozone pollution on net primary productivity and carbon  storage in terrestrial ecosystems of China. J
      Geophys Res 112: D22S09. http://dx.doi.org/10.1029/2007id008521.
Ren. W: Tian. H: Chen. G: Liu. M: Zhang. C: Chappelka. AH: Pan. S. (2007b). Influence of ozone pollution and
      climate variability on net primary productivity and carbon storage in China's grassland ecosystems from
      1961 to 2000. Environ Pollut 149: 327-335.  http://dx.doi.Org/10.1016/i.envpol.2007.05.029.
Ren. W: Tian. H: Tao. B: Chappelka. A: Sun. G: Lu. C: Liu. M:  Chen. G: Xu. X. (2011). Impacts of tropospheric
      ozone and climate change on net primary productivity and net carbon exchange of China's forest
      ecosystems. Global Ecology and Biogeography 20: 391-406. http://dx.doi.org/10.1111/1.1466-
      8238.2010.00606.x.
Renzetti, G: Silvestre, G: D'Amario, C: Bottini, E: Gloria-Bottini, F: Bottini, N: Auais, A: Perez, MK: Piedimonte,
      G.. (2009). Less air  pollution leads to rapid reduction of airway inflammation and improved airway function
      in asthmatic children. Pediatrics 123:  1051-1058. http://dx.doi.org/10.1542/peds.2008-1153.
Repapis. CC: Mantis. HT:  Paliatsos. AG: Philandras. CM: Bais. AF:  Meleti. C. (1998). Case study of UV-B
      modification during  episodes of urban air pollution. Atmos  Environ  38: 2203-2208.
Revis. NW: Major. T: Dalbev. WE. (1981). Cardiovascular effects  of ozone and cadmium inhalation in the rat. In
      Proceedings of the  research planning workshop on  health  effects of oxidants. (EPA-600/9-81-001).
      Raleigh, NC: U.S. Environmental Protection Agency.
Rhea, L: King, J: Kubiske,  M: Saliendra,  N: Teclaw, R. (2010). Effects of elevated atmospheric CO2 and
      tropospheric O3 on tree branch growth and implications for hydrologic budgeting. Environ Pollut 158:
      1079-1087. http://dx.doi.0rg/10.1016/i.envpol.2009.08.038.
Rich. DQ: Schwartz. J: Mittleman. MA: Link. M: Luttmann-Gibson. H: Catalano.  PJ: Speizer. FE:  Dockerv. DW.
      (2005). Association of short-term ambient air pollution concentrations and ventricular arrhythmias. Am J
      Epidemiol  161: 1123-1132.
Rich. DQ: Kim. MH: Turner. JR: Mittleman. MA: Schwartz. J: Catalano. PJ: Dockerv. DW. (2006a). Association
      of ventricular arrhythmias detected by implantable cardioverter defibrillator and ambient air pollutants in
      the St Louis, Missouri metropolitan area. Occup Environ Med 63: 591-596.
Rich, DQ: Mittleman, MA: Link, MS: Schwartz, J: Luttmann-Gibson, H: Catalano, PJ: Speizer, FE: Gold,  PR:
      Dockerv, DW. (2006b). Increased risk of paroxysmal atrial  fibrillation episodes associated  with acute
      increases in ambient air pollution. Environ Health Perspect 114: 120-123.
Rich. DQ: Kipen.  HM: Zhang. J: Kamat. L: Wilson. AC: Kostis. JB. (2010). Triggering oftransmural infarctions,
      but not nontransmural infarctions, by  ambient fine particles. Environ Health Perspect 118:  1229-1234.
      http://dx.doi.org/10.1289/ehp.0901624.
Richards. NAD: Osterman. GB: Browell.  EV: Hair. JW: Averv. M: Qinbin. L. (2008). Validation of tropospheric
      emission spectrometer ozone profiles with aircraft observations during the intercontinental chemical
      transport experiment-B. J Geophys Res 113: D16S29. http://dx.doi.org/10.1029/2007id008815.
Riddell, J: Padgett, PE: Nash, TH,  III. (2010). Responses of the lichen Ramalina menziesii Tayl. to ozone
      fumigations. In TH Nash, III  (Ed.), Biology of lichens: Symbiosis, ecology, environmental monitoring,
      systematics, cyber applications (Vol. 105, pp. 113-123). Stuttgart: J. Cramer in der Gebruder Borntraeger
      Verlagsbuchhandlung.
Riediker. M: Monn. C: Koller. T: Stahel. WA:  Wuthrich. B. (2001).  Air pollutants  enhance rhinoconjunctivitis
      symptoms in pollen-allergic  individuals. Ann Allergy Asthma Immunol 87: 311-318.
Riediker. M: Williams. R: Devlin. R: Griggs. T: Bromberg. P. (2003).  Exposure to particulate matter, volatile
      organic compounds, and other air pollutants inside patrol cars. Environ Sci  Technol 37: 2084-2093.
      http://dx.doi. org/10.1021/esQ26264y.
Draft - Do Not Cite or Quote                        11-69                                 September 2011

-------
Rigas, ML: Ben-Jebria, A: Ultman, JS. (1997). Longitudinal distribution of ozone absorption in the lung: Effects
      of nitrogen dioxide, sulfur dioxide, and ozone exposures. Arch Environ Occup Health 52: 173-178.
Rigas, ML: Catlin, SN: Ben-Jebria, A: Ultman, JS. (2000). Ozone uptake in the intact human respiratory tract:
      Relationship between inhaled dose and actual dose. J Appl Physiol 88: 2015-2022.
Riqel. PS:  Riqel. EG: Riqel. AC. (1999). Effects of altitude and latitude on ambient UVB radiation. J Am Acad
      Dermatol40: 114-116.
Riggsbee,  JA: Orr, CH: Leech, DM: Doyle, MW: Wetzel, RG. (2008). Suspended sediments in river ecosystems:
      Photochemical sources of dissolved organic carbon, dissolved organic nitrogen, and adsorptive removal
      of dissolved iron. J Geophys Res 113: G03019. http://dx.doi.orq/10.1029/2007ia000654.
Riikonen, J: Kets, K: Darbah, J: Oksanen, E: Sober, A: Vapaavuori, E: Kubiske, ME: Nelson, N: Karnoskv, PR
      (2008). Carbon gain  and bud physiology in Populus tremuloides and Betula papyrifera grown under long-
      term exposure to elevated concentrations of CO2 and O3. Tree Physiol 28: 243-254.
      http://dx.doi. org/10.1093/treephys/28.2.243.
Riikonen. J: Maenpaa. M: Alavillamo. M: Silfver. T: Oksanen.  E. (2009). Interactive effect of elevated
      temperature and O3 on antioxidant capacity and gas exchange in Betula pendula saplings. Planta 230:
      419-427. http://dx.doi.org/10.1007/s00425-009-0957-8.
Rind. D: Healv. R: Parkinson.  C: Martinson.  D. (1995). The role of sea ice in 2>2.0.CO:2.
Ritz, B: Yu, F. (1999). The effect of ambient carbon monoxide on low birth weight among children born in
      southern California between 1989 and 1993.  Environ Health Perspect 107: 17-25.
Ritz, B: Yu, F: Chapa, G: Fruin, S. (2000). Effect of air pollution on preterm birth among children born in
      Southern California between 1989 and 1993. Epidemiology 11: 502-511.
Ritz. B: Yu. F: Fruin. S: Chapa. G: Shaw. GM: Harris. JA. (2002). Ambient air pollution and risk of birth defects  in
      Southern California.  Am J Epidemiol 155: 17-25.
Ritz. B: Wilhelm. M: Zhao. Y. (2006). Air pollution and infant death  in southern California,  1989-2000. Pediatrics
      118: 493-502.
Ritz, B: Wilhelm, M: Hoggatt, KJ: Ghosh, JK. (2007). Ambient air pollution and preterm birth in the environment
      and pregnancy outcomes study at the University of California, Los Angeles. Am J Epidemiol 166: 1045-
      1052.
Ritz, B: Wilhelm, M. (2008). Ambient air pollution and adverse birth outcomes: Methodologic issues in an
      emerging field. Basic Appl Ecol 102: 182-190.
Rivas-Arancibia. S: Vazquez-Sandoval. R: Gonzalez-Kladiano. D: Schneider-Rivas. S: Lechuga-Guerrero. A.
      (1998). Effects of ozone exposure in rats on memory and levels of brain and pulmonary superoxide
      dismutase. Environ Res 76: 33-39.
Rivas-Arancibia. S: Dorado-Martinez. C: Borgonio-Perez. G: Hiriart-Urdanivia. M: Verdugo-Diaz. L: Duran-
      Vazquez, A: Colin-Baranque, L: Avila-Costa, MR. (2000). Effects oftaurine on ozone-induced memory
      deficits and lipid peroxidation levels in brains of young,  mature, and old rats. Environ Res 82: 7-17.
      http://dx.doi. org/10.1006/enrs. 1999.3996.
Rivas-Arancibia, S: Guevara-Guzman, R: Lopez-Vidal, Y: Rodrfguez-Martfnez,  E: Gomes, MZ: Angoa-Perez, M:
      Raisman-Vozari.  R. (2010). Oxidative stress caused by ozone exposure induces loss of brain repair in
      the  hippocampus of adult rats. Toxicol Sci 113: 187-197. http://dx.doi.org/10.1093/toxsci/kfp252.
Rivas-Manzano, P: Paz, C.  (1999). Cerebellar morphological alterations in rats induced by  prenatal ozone
      exposure. Neurosci Lett 276: 37-40.
Rizzo. M: Bernard!. R: Salvini.  M: Mali. C: Lorenzini. G: Durante. M. (2007). Identification of differentially
      expressed genes induced by ozone stress in sensitive and tolerant poplar hybrids. J Plant Physiol 164:
      945-949. http://dx.doi.0rg/10.1016/i.iplph.2006.07.012.
Rodenkirchen,  H: Gottlein, A:  Kozovits,  AR:  Matvssek, R:  Grams,  TEE. (2009).  Nutrient contents and
      efficiencies of beech and spruce saplings as influenced by competition and O3/CO2 regime. European
      Journal of Forest Research 128:  117-128. http://dx.doi.org/10.1007/s10342-008-0221-v.
Rodes. CE: Holland. DM. (1981). Variations of NO, NO2 and O3 concentrations downwind  of a Los Angeles
      freeway. Atmos Environ 15: 243-250.
Rodriguez. C: Tonkin. R: Hevworth. J: Kusel. M: De  Klerk. N:  Sly.  PD: Franklin.  P: Runnion. T: Blocklev. A:
      Landau. L: Hinwood. AL. (2007). The relationship between outdoor air quality and respiratory symptoms
      in young children. Int J Environ Health Res 17: 351-360. http://dx.doi.org/10.1080/09603120701628669.
Rogers, A: Allen, DJ: Davev, PA: Morgan, PB: Ainsworth, EA: Bernacchi, CJ:  Cornic, G: Dermody,  OC:
      Dohleman, FG: Heaton, EA: Mahoney, J: Zhu, X, -G: Delucia, EH: Ort, PR: Long, SP (2004). Leaf
      photosynthesis and carbohydrate dynamics of soybean grown throughout their life-cycle under free-air
      carbon dioxide enrichment. Plant Cell Environ 27: 449-458.
Roias-Martinez. R: Perez-Padilla. R: Olaiz-Fernandez. G: Mendoza-Alvarado. L: Moreno-Macias. H: Fortoul. T:
      McDonnell, W: Loomis, D: Romieu, I. (2007). Lung function growth in children with long-term exposure to
      air pollutants in Mexico  City. Am J Respir Grit Care Med 176: 377-384.
Draft - Do Not Cite or Quote                        11-70                                  September 2011

-------
Romansic, JM: Waggener, AA: Bancroft, BA: Blaustein, AR. (2009). Influence of ultraviolet-B radiation on
      growth, prevalence of deformities, and susceptibility to predation in Cascades frog (Rana cascadae)
      larvae. Hydrobiologia 624: 219-233. http://dx.doi.org/10.1007/s10750-009-9703-2.
Romero-Velazquez, RM: Alfaro-Rodriguez, A: Gonzalez-Pina, R: Gonzalez-Maciel, A. (2002). Effect of ozone
      prenatal exposure on postnatal development of cerebellum. Proc West Pharmacol Soc45: 65-67.
Romero. R: Espinoza. J: Kusanovic. JP: Gotsch. F: Hassan. S: Erez.  O: Chaiworapongsa. T: Mazor. M. (2006).
      The preterm parturition syndrome. BJOG 113: 17-42. http://dx.doi.Org/10.1111/i.1471-0528.2006.01120.x.
Romieu. I: Meneses. F: Ruiz. S: Sienra. JJ: Huerta. J: White. MC: Etzel. RA. (1996). Effects of air pollution on
      the respiratory health of asthmatic children living in Mexico City. Am J Respir  Grit Care Med 154: 300-
      307.
Romieu, I: Meneses, F: Ruiz, S: Huerta, J: Sienra, JJ: White, M: Etzel, R: Hernandez, M. (1997). Effects of
      intermittent ozone exposure on peak expiratory flow and respiratory symptoms among asthmatic children
      in Mexico City. Arch Environ Occup Health 52: 368-376.
Romieu. I: Meneses. F: Ramirez. M: Ruiz. S: Padilla. RP: Sienra. JJ:  Gerber. M: Grievink. L: Dekker. R: Walda.
      I: Brunekreef.  B. (1998a). Antioxidant supplementation and respiratory functions among workers exposed
      to high levels of ozone. Am J Respir Grit Care Med 158: 226-232.
Romieu. I: Lugo. MC: Colome.  S: Garcia. AM: Avila. MH:  Gevh. A: Velasco. SR: Rendon. EP. (1998b).
      Evaluation of indoor ozone concentration and predictors of indoor-outdoor ratio in Mexico City. J Air
      Waste Manag Assoc 48: 327-335.
Romieu, I: Sienra-Monge, JJ: Ramirez-Aguilar, M: Tellez-Rojo, MM: Moreno-Macias, H:  Reyes-Ruiz, Ml: Del
      Rio-Navarro, BE: Ruiz-Navarro, MX: Hatch, G: Slade, R: Hernandez-Avila, M. (2002). Antioxidant
      supplementation and lung functions among children with asthma exposed to high levels of air pollutants.
      Am J  Respir Grit Care Med 166: 703-709.
Romieu. I: Sienra-Monge. JJ: Ramirez-Aguilar. M: Moreno-Macias. H: Reyes-Ruiz. Nl: Estela del  Rio-Navarro.
      B:  Hernandez-Avila, M: London, SJ. (2004a). Genetic polymorphism of GSTM1 and antioxidant
      supplementation influence lung function in relation to ozone exposure in asthmatic children in Mexico
      City. Thorax 59: 8-10.
Romieu, I: Ramirez-Aguilar, M: Moreno-Macias, H:  Barraza-Villarreal, A: Miller, P:  Hernandez-Cadena, L:
      Carbajal-Arroyo, LA: Hernandez-Avila, M. (2004b). Infant mortality and air pollution:  Modifying effect by
      social class. J Occup Environ Hyg 46: 1210-1216.
Romieu, I: Ramirez-Aguilar, M: Sienra-Monge, JJ: Moreno-Macias, H: Del Rio-Navarro,  BE: David, G: Marzec,
      J: Hernandez-Avila. M: London. S. (2006). GSTM1  and GSTP1 and respiratory health in asthmatic
      children exposed to ozone. Eur Respir J 28:  953-959. http://dx.doi.org/10.1183/09031936.06.00114905.
Romieu, I: Barraza-Villarreal, A: Escamilla-Nunez, C: Almstrand, AC:  Diaz-Sanchez,  D: Sly, PD: Olin, AC.
      (2008). Exhaled breath malondialdehyde as  a marker of effect of exposure to air pollution in children with
      asthma. J Allergy Clin Immunol 121: 903-909. http://dx.doi.Org/10.1016/i.iaci.2007.12.004.
Romieu, I: Barraza-Villarreal, A: Escamilla-Nunez, C: Texcalac-Sangrador, Jl_: Hernandez-Cadena, L: Dfaz-
      Sanchez, D: De Batlle, J: Del Rio-Navarro, BE. (2009). Dietary intake, lung function  and airway
      inflammation in Mexico City school children exposed to air pollutants. Respir Res 10: 122.
Rosenfeld, MA: Leonova, VB: Konstantinova, ML: Razumovskii, SD. (2009). Self-assembly of fibrin  monomers
      and fibrinogen aggregation during ozone oxidation. Biochemistry (Mosc) 74: 41-46.
      http://dx.doi. org/10.1134/S0006297909010064.
Rosenthal, FS: Phoon,  C: Bakalian, AE: Taylor, HR. (1988). The ocular dose of ultraviolet radiation to outdoor
      workers. Invest Ophthalmol Vis Sci 29: 649-656.
Ross. MA: Perskv. VW: Scheff. PA: Chung. J: Curtis. L: Ramakrishnan. V: Wadden.  RA:  Hrvhorczuk. DO.
      (2002). Effect of ozone and aeroallergens on the respiratory health of asthmatics. Arch Environ Occup
      Health 57: 568-578.
Rothman, KJ: Greenland, S. (1998). Modern epidemiology. In (2nd ed.). Philadelphia, PA: Lippincott, Williams,
      SWilkins.
Roux. E: Hvvelin. J. -M: Savineau. J. -P: Marthan. R. (1999). Human  isolated airway contraction: Interaction
      between air pollutants and passive sensitization. Am J Respir Grit Care Med 160: 439-445.
Rowland-Bamford, AJ. (2000).  Plant responses to changing carbon dioxide and temperature. In SN  Singh (Ed.),
      Trace gas emissions and plants (pp. 63-74). Dordecht, The Netherlands: Kluwer Academic Publishers.
Rozenfeld. MA: Leonova. VB: Konstantinova. ML: Razumovskii. SD: Makarov. VA: Nevedrova. OE:
      Belozerskaja,  GG. (2008). Disturbance of functional properties of fibrinogen under ozone oxidation. Dokl
      Biochem Biophys 422: 315-318. http://dx.doi.org/10.1134/S1607672908050165.
Rubes, J: Selevan, SG: Evenson,  DP: Zudova, D: Vozdova, M: Zudova, Z: Robbins,  WA: Perreault,  SD. (2005).
      Episodic air pollution is associated with increased  DNA fragmentation in human sperm without other
      changes in  semen quality. Hum Reprod 20: 2776-2783.
Rubio. C: Paz. C. (2003). Indomethacin reverts sleep disorders produced by ozone exposure in rats. Toxicology
      191: 89-96. http://dx.doi.org/10.1016/S0300-483X(03)00245-2.
Ruckstuhl. C: Philipona. R: Behrens. K:  Coen. MC:  Purr. B: Heimo. A: Matzler. C: Nveki. S: Ohmura. A:
      Vuilleumier. L: Weller. M: Wehrli.  C: Zelenka. A. (2008). Aerosol and cloud effects on solar brightening
      and recent rapid warming. Geophys Res Lett 35: L12708. http://dx.doi.org/10.1029/2008GL034228.
Draft - Do Not Cite or Quote                       11-71                                 September 2011

-------
Rudez, G: Janssen, NA: Kilinc, E: Leebeek, FW: Gerlofs-Nijland, ME: Spronk, HM: ten Gate, H: Cassee, FR: de
      Maat, MR (2009). Effects of ambient air pollution on hemostasis and inflammation. Environ Health
      Perspect 117: 995-1001.
Ruidavets, J, -B: Cassadou, S: Cournot, M: Bataille, V: Meybeck, M: Ferrieres, J. (2005a). Increased resting
      heart rate with pollutants in a  population based study. J Epidemiol Community Health 59: 685-693.
Ruidavets. J. -B: Cournot. M: Cassadou. S: Giroux. M: Mevbeck. M: Ferrieres. J. (2005b). Ozone air pollution is
      associated with acute myocardial infarction. Circulation 111: 563-569.
Russell. A: Dennis. R. (2000). NARSTO critical review of photochemical models  and modeling [Review]. Atmos
      Environ 34: 2283-2324. http://dx.doi.orq/10.1016/S1352-2310(99)00468-9.
Rutkowski, JM: Santiag, LY: Ben-Jebria, A: Ultman, JS. (2011). Comparison of ozone-specific (OZAC) and
      oxygen radical (ORAC) antioxidant capacity assays for use with  nasal lavage fluid. Toxicol In Vitro 25:
      1406-1413. http://dx.doi.0rg/10.1016/i.tiv.2011.04.008.
Ryan, A: Cojocariu, C: Possell, M: Davies, WJ: Hewitt, CN. (2009). Defining hybrid poplar (Populus deltoides x
      Populus trichocarpa) tolerance to ozone: Identifying key parameters. Plant Cell Environ 32: 31-45.
      http://dx.doi.ora/10.1111/i.1365-3040.2008.01897.x.
Ryan, PH: LeMasters, GK. (2007). A review of land-use regression models for characterizing intraurban air
      pollution exposure [Review]. Inhal Toxicol 19: 127.
Rvang. SZ: Woo. SY: Kwon. SY:  Kim. SH: Lee.  SH:  Kim. KN: Lee. DK.  (2009). Changes of net photosynthesis,
      antioxidant enzyme activities, and antioxidant contents of Liriodendron tulipifera under elevated ozone.
      Photosynthetica 47: 19-25. http://dx.doi.org/10.1007/s11099-009-0005-8.
Ryerson, TB: Buhr, MP: Frost, GJ: Goldan, PD:  Holloway, JS: Hubler, G: Jobson, BT: Kuster, WC: McKeen, SA:
      Parrish, DP: Roberts, JM:  Sueper, DT: Trainer, M: Williams, J: Fehsenfeld, FC. (1998). Emissions
      lifetimes and ozone formation in power plant plumes. J Geophys Res 103: 22569-22583.
      http://dx.doi.org/10.1029/98JD01620.
Ryerson, TB: Trainer, M: Holloway, JS: Parrish, DP:  Huey, LG: Sueper,  DT:  Frost, GJ: Donnelly, SG:  Schauffler,
      S: Atlas. EL: Kuster. WC: Goldan. PD: Hubler. G: Meagher. JF: Fehsenfeld. FC. (2001). Observations of
      ozone formation in  power plant plumes and implications for ozone control strategies. Science  292: 719-
      723. http://dx.doi.org/10.1126/science.1058113.
Sacks, JD: Stanek, LW: Luben, TJ: Johns, DO: Buckley, BJ: Brown, JS: Ross, M. (2011). Particulate  matter
      induced health effects: Who's susceptible? Environ Health Perspect 119:  446-454.
      http://dx.doi.org/10.1289/ehp.1002255.
Sagai. M: Arakawa. K: Ichinose. T: Shimoio. N. (1987). Biochemical effects on combined gases of nitrogen
      dioxide and ozone: I. Species differences of lipid peroxides and phospholipids in lungs. Toxicology 46:
      251-265. http://dx.doi.0rg/10.1016/0300-483X(87)90207-1.
Sakugawa. H: Kaplan. IR. (1989). H2O2 and O3 in the atmosphere of Los Angeles and its vicinity: Factors
      controlling their formation and their role as oxidants of SO2. J Geophys Res 94: 12957-12973.
Salam, MT: Millstein, J: Li, Y, -F: Lurmann, FW:  Margolis, HG: Gilliland, FD. (2005). Birth outcomes and prenatal
      exposure to ozone, carbon monoxide, and particulate matter: Results from the Children's Health Study.
      Environ Health Perspect 113:  1638-1644.
Salam, MT: Islam, T: Gauderman, WJ: Gilliland, FD. (2009). Roles of arginase variants, atopy, and ozone in
      childhood asthma. J Allergy Clin Immunol 123: 596-602. http://dx.doi.Org/10.1016/i.iaci.2008.12.020.
Samet. JM: Zeger. SL:  Dominici.  F: Curriero. F: Coursac. I: Dockerv. DW: Schwartz. J: Zanobetti. A. (2000). The
      national morbidity, mortality, and air pollution study. Part II: Morbidity, mortality, and air pollution in the
      United States. Cambridge, MA: Health Effects Institute.
Samet. JM: Hatch. GE: Horstman. D: Steck-Scott. S: Arab. L: Bromberg. PA: Levine. M: McDonnell. WF: Devlin.
      RB. (2001). Effect of antioxidant supplementation on ozone-induced lung  injury in human subjects. Am J
      Respir Grit Care Med 164: 819-825.
Samet, JM: Bodurow, CC. (2008). Improving the presumptive disability decision-making process for veterans. In
      JM Samet; CC Bodurow (Eds.). Washington,  DC: National Academies Press.
Samoli. E: Zanobetti. A: Schwartz. J: Atkinson. R: Le Tertre. A: Schindler. C: Perez. L:  Cadum. E:  Pekkanen. J:
      Paldv. A: Touloumi. G: Katsouvanni. K. (2009). The temporal pattern of mortality responses to ambient
      ozone in the APHEA project. J Epidemiol Community Health 63:  960-966.
      http://dx.doi.org/10.1136/iech.2008.084012.
Samuel. MA: Miles. GP: Ellis. BE. (2000). Ozone treatment rapidly activates MAP kinase signalling in plants.
      Plant J 22: 367-376. http://dx.doi.Org/10.1046/j. 1365-313x.2000.00741 .x.
Samuel, MA: Ellis, BE.  (2002). Double jeopardy:  Both overexpression and suppression of a redox-activated
      plant mitogen-activated protein kinase render tobacco plants ozone sensitive. Plant Cell 14: 2059-2069.
      http://dx.doi.org/10.1105/tpc.002337.
Samuel. MA: Walia. A: Mansfield. SD: Ellis. BE. (2005).  Overexpression of SIPK in tobacco enhances ozone-
      induced ethylene formation and blocks ozone-induced SA accumulation. J Exp Bot 56: 2195-2201.
      http://dx.doi. org/10.1093/jxb/eri219.
Samuelson. LJ: Kelly. JM. (1997). Ozone uptake in Prunus serotina, Acer rubrum and  Quercus rubra forest
      trees of different sizes. New Phytol  136: 255-264. http://dx.doi.Org/10.1046/i.1469-8137.1997.00734.x.
Draft - Do Not Cite or Quote                        11-72                                  September 2011

-------
Sanchez, MJS: Pena, GS: Lorente, VC: Gallego, TM: Albert, JC. (2001). La contaminacion atmosferica en los
      bosques: Gufa para la identificacfon de danos visibles causados por Ozono. In (Vol. 6). Madrid, Spain:
      Ministerio de Medio Ambiente.
Sandermann, H. (2008). Ecotoxicology of ozone: Bioactivation of extracellular ascorbate. Biochem Biophys Res
      Commun 366: 271-274. http://dx.doi.orq/10.1016/i.bbrc.2007.12.018.
Santiago-Lopez. D: Bautista-Martfnez. JA: Reyes-Hernandez. Cl: Aguilar-Martfnez. M:  Rivas-Arancibia. S.
      (2010). Oxidative stress, progressive damage in the substantia nigra and plasma dopamine oxidation, in
      rats chronically exposed to ozone. Toxicol Lett 197: 193-200.
      http://dx.doi. orq/10.1016/i.toxlet.2Q10.05.020.
Santiago, LY: Hann, MC: Ben-Jebria, A: Ultman, JS. (2001). Ozone absorption in the human nose during
      unidirectional airflow. J Appl Physiol 91: 725-732.
Santucci, D: Sorace, A: Francia, N: Aloe, L: Alleva, E. (2006). Prolonged prenatal exposure to low-level ozone
      affects aggressive behaviour as well as NGF and BDNF levels in the central nervous system of CD-1
      mice. Behav Brain Res 166: 124-130. http://dx.doi.Org/10.1016/i.bbr.2005.07.032.
Sanz. J: Muntifering. RB: Bermeio. V: Gimeno. BS: Elvira. S. (2005). Ozone and increased nitrogen supply
      effects on the yield and nutritive quality of Trifolium subterraneum. Atmos Environ 39: 5899-5907.
      http://dx.doi.orq/10.1016/i.atmosenv.2005.06.022.
Sanz. J: Bermeio.  V: Gimeno.  BS: Elvira. S: Alonso. R.  (2007). Ozone sensitivity of the Mediterranean terophyte
      trifolium striatum is modulated by soil nitrogen content. Atmos Environ 41:  8952-8962.
      http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.016.
Sarangapani, R: Gentry, PR: Covington, TR: Teeguarden, JG: Clewell HJ, III. (2003). Evaluation of the potential
      impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of vapors.
      Inhal Toxicol 15: 987-1016.
Sarkar. A: Rakwal. R: Agrawal. SB: Shibato. J: Ogawa.  Y: Yoshida. Y: Agrawal. GK: Agrawal. M. (2010).
      Investigating the impact of elevated levels of ozone on tropical wheat using integrated phenotypical,
      physiological, biochemical, and proteomics approaches. J Proteome Res 9: 4565-4584.
      http://dx.doi. orq/10.1021/PM 002824.
Sarnat, JA:  Koutrakis, P:  Suh, HH. (2000). Assessing the relationship between personal particulate and
      gaseous exposures of senior citizens living in Baltimore, MD. J Air Waste Manag Assoc 50: 1184-1198.
Sarnat, JA:  Schwartz, J: Catalano, PJ: Suh, HH. (2001). Gaseous pollutants in particulate matter epidemiology:
      Confounders or surrogates? Environ  Health Perspect 109: 1053-1061.
Sarnat. JA:  Brown. KW: Schwartz. J: Coull. BA: Koutrakis. P. (2005). Ambient gas concentrations and personal
      particulate matter exposures: implications for studying the health effects of particles. Epidemiology 16:
      385-395.
Sarnat. SE: Suh. HH: Coull. BA: Schwartz. J: Stone. PH: Gold. DR. (2006a). Ambient particulate air pollution
      and cardiac arrhythmia in a panel of older adults in Steubenville, Ohio. Occup Environ Med 63:  700-706.
Sarnat, SE: Coull,  BA: Schwartz, J: Gold, PR: Suh, HH. (2006b). Factors affecting the association between
      ambient concentrations and personal exposures to particles and gases. Environ Health  Perspect 114:
      649-654.
Sarnat, SE: Klein,  M: Sarnat,  JA: Flanders, WD: Waller, LA:  Mulholland, JA: Russell, AG: Tolbert, PE. (2010). An
      examination of exposure measurement error from air pollutant spatial variability  in time-series studies. J
      Expo Sci Environ Epidemiol 20: 135-146. http://dx.doi.orq/10.1038/ies.20Q9.10.
Sarwar, G: Roselle, SJ: Mathur, R: Appel, W: Dennis, RL: Vogel, B.  (2008). A comparison of CMAQ HONO
      predictions  with observations from the Northeast Oxidant and Particle Study. Atmos Environ 42: 5760-
      5770.
Sathishkumar, K: Hague, M:  Perumal, TE: Francis, J: Uppu, RM. (2005). A major ozonation product of
      cholesterol, 3beta-hydroxy-5-oxo-5,6-secocholestan-6-al, induces apoptosis in H9c2 cardiomyoblasts.
      FEBS Lett 579: 6444-6450.
Sathishkumar, K: XI, X: Martin, R: Uppu, RM. (2007a). Cholesterol secoaldehyde, an ozonation product of
      cholesterol, induces amyloid aggregation and apoptosis in murine GT1-7 hypothalamic neurons. J
      Alzheimers Dis 11: 261-274.
Sathishkumar, K: Murthy, SN: Uppu, RM. (2007b). Cytotoxic effects of oxysterols produced during ozonolysis of
      cholesterol  in murine GT1-7 hypothalamic neurons. Free Radic Res 41: 82-88.
      http://dx.doi. orq/10.1080/10715760600950566.
Sathishkumar, K: Gao, X:  Raghavamenon, AC:  Parinandi, N: Pryor, WA: Uppu, RM. (2009). Cholesterol
      secoaldehyde induces apoptosis in H9c2 cardiomyoblasts through reactive oxygen species involving
      mitochondrial and death receptor pathways. Free Radic Biol Med 47: 548-558.
      http://dx.doi.0rg/10.1016/i.freeradbiomed.2009.05.020.
Sato. S: Shimura.  S: Hirose.  T: Maeda. S: Kawakami. M: Takishima. T: Kimura. S. (1980). Effects  of long-term
      ozone exposure and dietary vitamin E in rats. Tohoku J Exp Med 130: 117-128.
Saviranta, NMM: Julkunen-Tiitto, R: Oksanen,  E: Karjalainen, RO. (2010). Leaf phenolic compounds in red
      clover (Trifolium pratense L.) induced by exposure to moderately elevated ozone. Environ Pollut 158:
      440-446. http://dx.doi.0rg/10.1016/i.envpol.2009.08.029.
Draft - Do Not Cite or Quote                        11-73                                  September 2011

-------
Sawada, H: Kohno, Y. (2009). Differential ozone sensitivity of rice cultivars as indicated by visible injury and
      grain yield. Plant Biol (Stuttg) 11: 70-75. http://dx.doi.Org/10.1111/i.1438-8677.2009.00233.x.
Sawyer, K: Brown, J: HazuchaM: Bennett, WD. (2007). The effect of exercise on nasal uptake of ozone in
      healthy human adults. J Appl Physiol 102: 1380-1386. http://dx.doi.org/10.1152/iapplphysiol.00269.2006.
Scannell. C: Chen. L: Aris. RM: Tager. I: Christian. D: Ferrando. R: Welch. B: Kelly. T: Balmes. JR. (1996).
      Greater ozone-induced inflammatory responses in subjects with asthma. Am J Respir Crit Care Med 154:
      24-29.
Scarlett, JF; Abbott, KJ; Peacock, JL; Strachan, DP; Anderson, HR. (1996). Acute effects of summer air
      pollution on respiratory function  in primary school children in southern England. Thorax 51: 1109-1114.
Scebba, F: Giuntini, D: Castagna, A: Soldatini, G: Ranieri, A. (2006). Analysing the impact of ozone on
      biochemical and physiological variables in plant species belonging to natural ecosystems. Environ Exp
      Bot 57: 89-97. http://dx.doi.Org/10.1016/i.envexpbot.2005.04.005.
Schaub, M: Skellv, JM: Zhang, JW: Ferdinand, JA: Savage, JE: Stevenson, RE: Davis,  DP:  Steiner, KG. (2005).
      Physiological and foliar symptom response in the crowns of Prunus serotina, Fraxinus americana and
      Acer rubrum canopy trees to ambient ozone under forest conditions. Environ Pollut 133: 553-567.
      http://dx.doi.0rg/10.1016/i.envpol.2004.06.012.
Schelegle. ES: Adams. WC. (1986). Reduced exercise time in competitive simulations consequent to low level
      ozone exposure. Med Sci  Sports Exerc 18: 408-414.
Schelegle, ES: Adams, WC: Siefkin, AD. (1987). Indomethacin pretreatment reduces ozone-induced pulmonary
      function decrements in human subjects. Am Rev Respir Dis 136: 1350-1354.
Schelegle, ES: Siefkin, AD: McDonald, RJ. (1991). Time course of ozone-induced neutrophilia in normal
      humans. Am J Respir Crit Care Med 143: 1353-1358.
Schelegle. ES: Carl. ML: Coleridge. HM: Coleridge. JCG: Green. JF. (1993). Contribution of vagal afferents to
      respiratory reflexes evoked by acute inhalation of ozone in dogs. J Appl Physiol 74: 2338-2344.
Schelegle, ES: Miller, LA: Gershwin, LJ: Fanucchi, MV: Van Winkle, LS: Gerriets, JE: Walby, WF: Mitchell,  V:
      Tarkington. BK: Wong. VJ: Baker. GL: Pantle. LM: Joad. JP: Pinkerton. KE: Wu.  R: Evans. MJ: Hyde.
      DM: Plopper. CG. (2003).  Repeated episodes of ozone inhalation amplifies the effects of allergen
      sensitization and inhalation on airway immune and structural development in Rhesus monkeys. Toxicol
      Appl Pharmacol 191: 74-85.
Schelegle, ES: Walby, WF: Adams, WC. (2007). Time course of ozone-induced changes in breathing pattern in
      healthy exercising humans. J Appl Physiol 102: 688-697.
      http://dx.doi.org/10.1152/iapplphvsiol.00141.2006.
Schelegle. ES: Morales. CA: Walbv. WF: Marion. S: Allen. RP (2009).  6.6-hour inhalation of ozone
      concentrations from 60 to  87 parts per billion in healthy humans. Am J Respir Crit Care Med 180: 265-
      272. http://dx.doi.org/10.1164/rccm.200809-1484OC.
Schenker.  MB: Orenstein.  MR: Samuels. SJ. (2002). Use of protective equipment among California farmers. Am
      J Ind Med 42: 455-464.
Schichtel, BA: Husar, RB. (2001). Eastern North American transport climatology during  high- and low-ozone
      days. Atmos Environ 35: 1029-1038.  http://dx.doi.org/10.1016/S1352-2310(00)00370-8.
Schildcrout, JS:  Sheppard,  L: Lumlev, T: Slaughter, JC: Koenig, JQ: Shapiro, GG. (2006). Ambient air pollution
      and asthma exacerbations in children: An eight-city analysis. Am J Epidemiol 164: 505-517.
      http://dx.doi. org/10.1093/aie/kwi225.
Schindler, DW: Curtis, PJ: Parker, BR: Stainton, MP (1996). Consequences of climate warming and lake
      acidification for UV-B penetration in North American boreal lakes. Nature 379: 705-708.
      http://dx.doi. org/10.1038/379705aO.
Schmekel, B: Ahlner, J: Malmstrom, M: Venge, P. (2001). Eosinophil cationic protein (ECP) in saliva: A new
      marker of disease activity  in bronchial asthma. Respir  Med 98: 670-675.
      http://dx.doi. org/10.1053/rmed.20Q1.1123.
Schmelzer, KR: Wheelock,  AM: Dettmer, K:  Morin, D: Hammock, BD. (2006). The role of inflammatory
      mediators in the synergistic toxicity of ozone and 1-nitronaphthalene in rat airways. Environ Health
      Perspect  114: 1354-1360.
Schnell, RC:  Oltmans, SJ: Neelv, RR: Endres, MS: Molenar, JV: White, AB. (2009). Rapid photochemical
      production of ozone at high concentrations in a rural site during winter. Nat Geosci 2: 120-122.
      http://dx.doi. org/10.1038/NGEO415.
Schopke, R: Wolfer, DP: Lipp, HP: Leisinger-Trigona, MC. (1991). Swimming navigation and structural
      variations of the infrapyramidal mossy fibers in the hippocampus of the mouse. Hippocampus 1: 315-
      328. http://dx.doi.org/10.1002/hipo.450Q10322.
Schraudner, M: Moeder, W: Wiese, C: Van Camp, W: Inze, D: Langebartels, C: Sandermann, H, Jr. (1998).
      Ozone-induced oxidative burst in the ozone biomonitor plant, tobacco Bel W3. Plant J 16: 235-245.
      http://dx.doi. org/10.1046/i. 1365-313x. 1998.00294.x.
Schroter, RC: Sudlow, MF (1969). Flow patterns in models of the human bronchial airways. Respir Physiol
      Neurobiol 7:  341-355.
Schwartz. J. (2005a). How sensitive is the association between ozone and daily deaths to control for
      temperature? Am J Respir Crit Care Med 171: 627-631.
Draft - Do Not Cite or Quote                       11-74                                 September 2011

-------
Schwartz, J: Litonjua, A: Suh, H: Verrier, M: Zanobetti, A: Syring, M: Nearing, B: Verrier, R: Stone, P:
      MacCallum, G: Speizer, FE: Gold, DR. (2005). Traffic related pollution and heart rate variability in a panel
      of elderly subjects. Thorax 60: 455-461.
Schwartz, J. (2005b). Who is sensitive to extremes of temperature? A case-only analysis. Epidemiology 16: 67-
      72. http://dx.doi.org/10.1097/01.ede.0000147114.25957.71.
Seal. E.  Jr:  McDonnell. WF:  House. DE: Salaam. SA: Dewitt. PJ: Butler. SO: Green. J: Raggio. L. (1993). The
      pulmonary response of white and black adults to six concentrations of ozone. Am J Respir Crit Care Med
      147:  804-810.
Seal. E.  Jr:  McDonnell. WF:  House. DE. (1996). Effects of age, socioeconomic status, and menstrual cycle on
      pulmonary response to ozone. Arch Environ Occup Health 51: 132-137.
Seaman, NL. (2000).  Meteorological modeling for air quality assessments. Atmos Environ 34: 2231-2259.
Seinfeld, JH: Pandis,  SN. (1998). Atmospheric chemistry and physics: From air pollution to climate change. In.
      New York: John Wiley & Sons.
Selevan. SG: Borkovec. L: Slott. VL: Zudova. Z: Rubes. J: Evenson. DP: Perreault. SD. (2000). Semen quality
      and reproductive health of young Czech men exposed to seasonal air pollution. Environ Health Perspect
      108:  887-894.
Belgrade. M. -JK: Smith. MV: Oberhelman-Bragg. LJ: LeVee. GJ: Koren. HS: Cooper. KD. (2001). Dose
      response for UV-induced immune suppression in people of color: Differences based on erythemal
      reactivity rather than skin pigmentation. Photochem Photobiol 74: 88-95.
Selgrade, MK: Daniels, MJ: Grose, EC. (1990). Acute, subchronic, and chronic exposure to a simulated urban
      profile of ozone: Effects on  extrapulmonary natural killer cell activity and lymphocyte mitogenic
      responses. Inhal Toxicol 2:  375-389.
Belgrade. MK: Repacholi. MH: Koren. HS. (1997). Ultraviolet radiation-induced immune modulation: Potential
      consequences for infectious, allergic, and autoimmune disease. Environ Health Perspect 105: 332-334.
Seltzer, J: Bigby, BG: Stulbarg, M: Holtzman, MJ: Nadel, JA: Ueki, IF: Leikauf, GD: Goetzl, EJ: Boushey, HA.
      (1986). OS-induced change in bronchial reactivity to methacholine and airway inflammation in humans. J
      Appl Physiol 60: 1321-1326.
Selwyn,  BJ: Stock, TH: Hardy, RJ: Chan, FA: Jenkins, DE: Kotchmar, DJ:  Chapman, RS.  (1985). Health effects
      of ambient ozone exposure in vigorously exercising adults. In Evaluation of the scientific basis for
      ozone/oxidants standards: Proceedings of an apca international specialty conference (pp. 281-296).
      Houston, TX: Air Pollution Control Association.
Semenza. JC: Wilson. DJ: Parra. J: Bontempo. BD: Hart. M: Sailor. DJ: George. LA. (2008). Public  perception
      and behavior change in relationship to hot weather and air pollution. Environ Res 107:  401-411.
      http://dx.doi.0rg/10.1016/i.envres.2008.03.005.
Semerdiieva. SI: Phoenix. GK: Hares. D: Gwvnn-Jones. D: Callaghan. TV: Sheffield.  E. (2003). Surface
      morphology,  leaf and  cuticle thickness of four dwarf shrubs from a sub-Arctic heath following long-term
      exposure to enhanced levels of UV-B. Physiol Plant  117: 289-294.  http://dx.doi.Org/10.1034/j. 1399-
      3054.2003.00006.x.
Servais, S: Boussouar, A: Molnar, A: Douki, T: Peguignot, JM: Favier, R. (2005). Age-related sensitivity to lung
      oxidative stress during ozone exposure. Free Radic  Res 39: 305-316.
      http://dx.doi. orq/10.1080/10715760400011098.
Severino. JF: Stich. K: Soia. G. (2007). Ozone stress and antioxidant substances in Trifolium repens and
      Centaureajacea leaves. Environ Pollut 146: 707-714.  http://dx.doi.Org/10.1016/i.envpol.2006.04.006.
Sexton.  KG: Jeffries. HE: Jang. M:  Kamens. RM: Doyle. M: Voicu. I: Jaspers. I. (2004). Photochemical products
      in urban mixtures enhance  inflammatory responses in lung cells. Inhal Toxicol  1: 107-114.
Shapiro, MA. (1980). Turbulent mixing within tropopause folds as a mechanism for the exchange of chemical
      constituents between the stratosphere and troposphere. J Atmos Sci 37: 994-1004.
Sharkey, TD: Wiberley, AE:  Donohue, AR. (2008). Isoprene emission from  plants: Why and how. Ann Bot 101: 5-
      18. http://dx.doi.0rg/10.1093/aob/mcm240.
Sharkhuu. T: Doerfler. PL: Copeland. C: Luebke. RW: Gilmour. Ml. (2011). Effect of maternal exposure to ozone
      on reproductive outcome and immune, inflammatory, and allergic responses in the offspring.  J
      Immunotoxicol 8: 183-194.  http://dx.doi.org/10.3109/1547691X.2011.568978.
Sheppard. L: Slaughter. JC:  Schildcrout. J: L-JS. L: Lumlev. T. (2005). Exposure and measurement
      contributions to estimates of acute air pollution effects. J Expo Sci Environ Epidemiol 15: 366-376.
Sherman, M: McWilliams, J.  (2007). Air leakage of U.S.  homes: Model prediction. (LBNL-62078).  Berkeley, CA:
      Lawrence Berkeley National Laboratory, http://epb.lbl.gov/publications/lbnl-62078.pdf.
Sherman, MH: Grimsrud, DT (1980). Infiltration-pressurization correlation: Simplified physical modeling. In
      ASHRAE Transactions (pp. 778-807). Denver, CO: Lawrence Berkeley Laboratory.
Sherry. B: Blanck. HM: Galuska. DA: Pan. L: Dietz. WH: Balluz. L. (2010). Vital signs: State-specific obesity
      prevalence among adults -  United States, 2009. MMWR Recomm Rep 59: 951-955.
Shindell, D: Faluvegi, G: Lacis, A:  Hansen, J: Ruedv,  R: Aguilar, E. (2006). Role of tropospheric ozone
      increases in 20th-century climate change. J Geophys Res 111: D08302.
      http://dx.doi. orq/10.1029/2005JD006348.
Draft - Do Not Cite or Quote                        11-75                                  September 2011

-------
Shindell, D: Faluvegi, G. (2009). Climate response to regional radiative forcing during the twentieth century. Nat
      Geosci 2: 294-300. http://dx.doi.Org/10.1038/ngeo473.
Shindell, DT: Faluvegi, G. (2002). An exploration of ozone changes and their radiative forcing prior to the
      chlorofluorocarbon era. Atmos Chem Phys Discuss 2: 363-374. http://dx.doi.org/10.5194/acp-2-363-
      2002.
Shindell. DT: Faluvegi. G: Bell. N. (2003). Preindustrial-to-present-day radiative forcing by tropospheric ozone
      from improved simulations with the GISS chemistry-climate GCM. Atmos Chem Phys 3: 1675-1702.
      http://dx.doi.org/10.5194/acp-3-1675-2003.
Shindell. DT: Faluvegi. G: Bell. N: Schmidt. GA. (2005). An emissions-based view of climate forcing by methane
      and tropospheric ozone. Geophys Res Lett 32: L04803. http://dx.doi.Org/10.1029/2004GL021900.
Shindell, DT: Faluvegi, G: Bauer, SE: Koch, DM: Unger, N: Menon, S: Miller, Rl_: Schmidt, GA: Streets, DG.
      (2007). Climate response to projected  changes in short-lived species under an A1B scenario from 2000-
      2050 in the GISS climate model. J Geophys Res 112: D20103. http://dx.doi.org/10.1029/2007id008753.
Shindell. DT: Lew H. II: Schwarzkopf.  MD: Horowitz. LW: Lamargue. JF: Faluvegi. G. (2008). Multimodel
      projections of climate change from short-lived emissions due to human activities. J Geophys Res 113:
      D11109. http://dx.doi.org/10.1029/2007JD009152.
Shore. SA: Schwartzman. IN: Le Blanc. B: Krishna Murthv. GG: Doerschuk. CM. (2001). Tumor necrosis factor
      receptor 2 contributes to ozone-induced airway hyperresponsiveness in mice. Am J Respir Crit Care Med
      164: 602-607.
Shore, SA: Rivera-Sanchez, YM: Schwartzman, IN:  Johnston, RA. (2003).  Responses to  ozone are increased in
      obese mice. J Appl Physiol 95: 938-945.
Shore, SA. (2007). Obesity and asthma: lessons from animal models. J Appl Physiol 102: 516-528.
Shore. SA: Lang. JE:  Kasahara. PI: Lu. FL: Verbout. NG: Si. H: Williams. ES: Terry. RD: Lee. A: Johnston. RA.
      (2009). Pulmonary responses to subacute ozone exposure in obese vs. lean mice. J Appl  Physiol 107:
      1445-1452. http://dx.doi.org/10.1152/iapplphysiol.00456.2009.
Shoveller. JA: Lovato. CY: Peters. L: Rivers. JK. (1998). Canadian national survey on sun exposure & protective
      behaviours: Adults at leisure. Cancer Prev Control 2: 111-116.
Shu, S: Morrison, GC. (2011). Surface  reaction rate and probability of ozone and alpha-terpineol  on glass,
      polyvinyl chloride, and latex  paint surfaces. Environ Sci Technol 45:  4285-4292.
      http://dx.doi.org/10.1021/es200194e.
Sienra-Monge, JJ: Ramirez-Aguilar, M:  Moreno-Macias, H: Reyes-Ruiz, Ml: Del Rio-Navarro, BE: Ruiz-Navarro,
      MX: Hatch. G:  Crissman. K:  Slade. R:  Devlin. RB: Romieu. I. (2004). Antioxidant supplementation and
      nasal inflammatory responses among young  asthmatics exposed to high levels of  ozone. Clin Exp
      Immunol 138: 317-322. http://dx.doi.Org/10.1111/i.1365-2249.2004.02606.x.
Sillman. S. (1995). The use of NOy, H2O2 and HNO3 as indicators for ozone-NOx-hydrocarbon sensitivity in
      urban locations. J Geophys  Res 100: 14175-14188.
Sillman, S: He, D: Pippin,  MR: Daum, PH: Imre, DG: Kleinman, LI: Lee, JH: Weinstein-Lloyd, J. (1998). Model
      correlations for ozone, reactive nitrogen, and peroxides for Nashville in comparison with measurements:
      implications for O3-NOx-hydrocarbon chemistry. J Geophys Res 103: 22629-22644.
Sillman, S: He, D, -Y.  (2002). Some theoretical results concerning O3-NOx-VOC chemistry and NOx-VOC
      indicators. J Geophys Res 107:  4659.  http://dx.doi.org/10.1029/2001JD001123.
Silverman. RA: Ito. K. (2010). Age-related association of fine particles and  ozone with severe acute asthma in
      New York City. J Allergy Clin Immunol  125: 367-373.e365. http://dx.doi.Org/10.1016/i.iaci.2009.10.061.
Silverman. RA: Ito. K: Freese. J: Kaufman. BJ: De Claro. D: Braun. J: Prezant. DJ. (2010). Association of
      ambient fine particles with out-of-hospital cardiac arrests in  New York City. Am J Epidemiol 172: 917-923.
      http://dx.doi.org/10.1093/aie/kwg217.
Simmonds, PG: Derwent, RG: Manning, Al_: Spain, G. (2004). Significant growth in surface  ozone at Mace
      Head, Ireland,  1987-2003. Atmos Environ 38: 4769-4778.
      http://dx.doi.0rg/10.1016/i.atmosenv.2004.04.036.
Simonian.  NA: Covle. JT (1996). Oxidative stress in neurodegenerative diseases. Annu Rev Pharmacol Toxicol
      36: 83-106. http://dx.doi.org/10.1146/annurev.pa.36.040196.000503.
Simpson, R: Williams, G: Petroeschevskv, A:  Best, T: Morgan, G:  Denison,  L: Hinwood, A: Neville, G. (2005).
      The short-term effects of air  pollution on hospital admissions in four Australian cities. Aust  N Z J Public
      Health 29: 213-221.
Sinclair, AH: Tolsma, D. (2004). Associations and lags between air pollution and acute respiratory visits in an
      ambulatory care setting: 25-month results from the aerosol research and inhalation epidemiological
      study. J Air Waste Manag Assoc 54: 1212-1218.
Sinclair, AH: Edgerton, ES: Wyzga, R: Tolsma, D. (2010). A two-time-period comparison of the effects of
      ambient air pollution on outpatient visits for acute respiratory illnesses. J Air Waste Manag Assoc 60:
      163-175. http://dx.doi.0rg/10.3155/1047-3289.60.2.163.
Singh, E: Tiwari, S: Agrawal, M. (2009). Effects of elevated ozone on  photosynthesis and  stomatal conductance
      of two soybean varieties: A case study to assess impacts of one component of predicted global climate
      change. Plant Biol (Stuttg) 11: 101-108. http://dx.doi.org/10.1111/1.1438-8677.2009.00263.X.
Draft - Do Not Cite or Quote                        11-76                                  September 2011

-------
Singh, E: Tiwari, S: Agrawal, M. (201 Oa). Variability in antioxidant and metabolite levels, growth and yield of two
      soybean varieties: An assessment of anticipated yield losses under projected elevation  of ozone. Agric
      Ecosyst Environ 135: 168-177. http://dx.doi.Org/10.1016/i.agee.2009.09.004.
Singh, HB: Anderson, BE: Brune, WH: Cai, C: Cohen,  RC: Crawford, JH: Cubison, MJ: Czech, EP: Emmons, l_:
      Fuelberg. HE. (201 Ob). Pollution influences on atmospheric composition and chemistry  at high northern
      latitudes: Boreal and California forest fire emissions. Atmos Environ 44: 4553-4564.
      http://dx.doi.0rg/10.1016/i.atmosenv.2010.08.026.
Sitch. S: Cox. PM: Collins. WJ: Huntingford. C.  (2007). Indirect radiative forcing  of climate change through
      ozone effects on the land-carbon sink. Nature 448: 791-794. http://dx.doi.org/10.1038/nature06059.
Skarbv, l_: Troeng, E: Bostrom, C, -A. (1987). Ozone uptake and effects on transpiration, net photosynthesis,
      and dark respiration in Scots pine. Forest Sci 33: 801-808.
Skarbv, l_: Ottosson, S: Karlsson, PE: Wallina, G:  Sellden,  G: Medina, EL: Pleijel, H. (2004). Growth of Norway
      spruce (Picea abies) in relation to different ozone exposure indices: a synthesis. Atmos  Environ 38:
      2225-2236.
Slade. R: Highfill. JW: Hatch. GE. (1989). Effects  of depletion of ascorbic acid or nonprotein sulfhydryls on the
      acute inhalation toxicity of nitrogen dioxide, ozone, and phosgene. Inhal Toxicol 1: 261-271.
Slade. R: Crissman. K: Norwood. J: Hatch. G. (1993). Comparison of antioxidant substances in bronchoalveolar
      lavage cells and fluid from humans, guinea pigs, and rats. Exp Lung Res  19: 469-484.
Slade, R: Watkinson, WP: Hatch, GE. (1997). Mouse strain differences in ozone dosimetry  and body
      temperature changes. Am J Physiol 272: L73-L77.
Slama, R: Darrow, l_: Parker, J: Woodruff, TJ: Strickland, M: Nieuwenhuijsen, M: Glinianaia, S:  Hoggatt, KJ:
      Kannan, S: Hurley, F: Kalinka, J: Sram, R:  Brauer, M: Wilhelm, M: Heinrich, J: Ritz, B. (2008). Meeting
      report: Atmospheric pollution and  human reproduction. Environ Health Perspect 116: 791-798.
Slaper. H: Velders. GJM: Daniel. JS: de Gruiil. FR: Van der Leun. JC. (1996). Estimates of ozone depletion and
      skin cancer incidence to examine the Vienna Convention achievements. Nature 384: 256-258.
Slinev. DH: Wengraitis. S. (2006). Is a differentiated advice by season and region necessary? Prog Biophys Mol
      Biol 92: 150-160. http://dx.doi.Org/10.1016/i.pbiomolbio.2006.02.007.
Smedby,  KE: Hjalgrim, H: Melbye, M: Torrang, A:  Rostgaard, K: Munksgaard, l_:  Adami, J: Hansen, M: Porwit-
      MacDonald, A: Jensen, BA:  Roos, G: pedersen, BB: Sundstrom,  C: Glimelius, B: Adami, H, -O. (2005).
      Ultraviolet radiation exposure and risk of malignant  lymphomas. J Natl Cancer Inst 97: 199-209.
Smith, G: Coulston, J: Jepsen, E: Prichard, T. (2003). A national ozone biomonitoring program: Results from
      field surveys of ozone sensitive plants in northeastern forests (1994-2000).  Environ Monit Assess 87:
      271-291.
Smith, KR: Jerrett, M: Anderson, HR: Burnett, RT: Stone, V: Derwent, R: Atkinson, RW:  Cohen, A: Shonkoff, SB:
      Krewski. D: Pope. CA. Ill: Thun. MJ: Thurston. G. (2009a). Public health benefits of strategies to reduce
      greenhouse-gas emissions: Health implications of short-lived greenhouse pollutants. Lancet 374: 2091-
      2103. http://dx.doi.org/10.1016/s0140-6736(09)61716-5.
Smith, RL: Xu, B: Switzer, P. (2009b). Reassessing the relationship between ozone and short-term mortality in
      U.S.  urban communities. Inhal Toxicol 21: 37-61. http://dx.doi.org/10.1080/08958370903161612.
Snell, KRS: Kokubun, T:  Griffiths, H: Convey, P: Hodgson,  DA:  Newsham, KK. (2009). Quantifying the metabolic
      cost to an Antarctic liverwort of responding to an abrupt increase in UVB radiation exposure. Global
      Change Biol 15: 2563-2573. http://dx.doi.Org/10.1111/i.1365-2486.2009.01929.x.
Soden, BJ: Held, IM. (2006). An assessment of climate feedbacks in coupled ocean-atmosphere models. J Clim
      19: 3354-3360.
Soia. G: Barnes. JD: Posch. M: Vandermeiren. K: Pleiiel. H: Mills. G. (2000). Phenological weighting of ozone
      exposures in the calculation of critical levels for wheat, bean and plantain. Environ Pollut 109: 517-524.
      http://dx.doi.org/10.1016/S0269-7491(00)00055-5.
Soia, G: Reichenauer, TG: Eid, M: Soia, A, -M: Schaber, R: Gangl, H. (2004). Long-term ozone exposure and
      ozone uptake of grapevines in open-top chambers. Atmos Environ 38: 2313-2321.
Sokol. RZ: Kraft. P: Fowler. IM: Mamet. R: Kim.  E: Berhane. KT. (2006). Exposure to environmental ozone alters
      semen quality. Environ Health Perspect 114:  360-365.
Solic, JJ: Hazucha, MJ: Bromberg, PA. (1982). The acute effects of 0.2 ppm ozone in patients  with chronic
      obstructive pulmonary disease. Am Rev Respir Dis  125: 664-669.
Somers. GL: Chappelka. AH: Rosseau. P: Renfro. JR. (1998). Empirical evidence  of growth decline related to
      visible ozone injury. For Ecol Manage 104: 129-137. http://dx.doi.org/10.1016/50378-1127(97)00252-1.
Son, JY: Cho, YS: Lee, JT (2008). Effects of air pollution on postneonatal infant mortality among firstborn
      infants in Seoul, Korea: Case-crossover and time-series analyses. Arch Environ Occup  Health 63: 10S-
      US.
Son, JY; Bell, ML; Lee, JT. (2010).  Individual exposure to air pollution and lung function  in Korea: Spatial
      analysis using multiple exposure approaches. Environ Res 110: 739-749.
      http://dx.doi.0rg/10.1016/i.envres.2010.08.003.
Soulage.  C: Perrin. D: Cottet-Emard. J. -M: Peguignot. J: Dalmaz. Y: Peguignot. J. -M. (2004).  Central and
      peripheral changes in catecholamine biosynthesis and turnover in rats after a short period of ozone
      exposure. Neurochem Int45: 979-986.
Draft - Do Not Cite or Quote                        11-77                                  September 2011

-------
Sousa, SI: Ferraz, C: Alvim-Ferraz, MC: Martins, FG: Vaz, LG: Pereira, MC. (2011). Spirometric tests to assess
      the prevalence of childhood asthma at Portuguese rural areas: Influence of exposure to high ozone
      levels. Environ Int 37: 474-478. http://dx.doi.Org/10.1016/i.envint.2010.11.014.
Sousa, SIV: Pereira, MMC: Martins, FG: Alvim-Ferraz,  CM. (2008). Identification of regions with high  ozone
      concentrations aiming the impact assessment on childhood asthma. Hum Ecol Risk Assess 14: 610-622.
      http://dx.doi. orq/10.1080/10807030802074147.
Sousa, SIV: Alvim-Ferraz,  MCM: Martins, FG: Pereira,  MC. (2009). Ozone exposure and its influence on the
      worsening of childhood asthma. Allergy 64: 1046-1055. http://dx.doi.org/10.1111/1.1398-
      9995.2009.01946.x.
Souza, L: Neufeld, HS: Chappelka, AH: Burkev, KO: Davison, AW. (2006). Seasonal development of  ozone-
      induced foliar injury on tall milkweed (Asclepias  exaltata) in Great Smoky Mountains National Park.
      Environ  Pollut 141:  175-183. http://dx.doi.Org/10.1016/i.envpol.2005.07.022.
Spannhake, EW: Reddv, SPM: Jacoby, DB: Yu,  X, -Y: Saatian, B: Tian, J. (2002). Synergism between rhinovirus
      infection and oxidant pollutant exposure enhances airway epithelial cell cytokine production. Environ
      Health Perspect 110: 665-670.
Spektor, DM: Lippmann, M: Liov, PJ: Thurston, GD: Citak, K: James, DJ: Bock, N; Speizer, FE: Hayes, C.
      (1988a). Effects of ambient ozone on respiratory function in active, normal children. Am Rev Respir Dis
      137: 313-320.
Spektor, DM: Lippmann, M: Thurston, GD: Liov, PJ: Stecko, J: O'Connor, G: Garshick, E: Speizer, FE: Hayes,
      C.. (1988b). Effects  of ambient ozone  on respiratory function in healthy adults exercising outdoors. Am
      Rev Respir Dis 138: 821-828.
Spektor, DM: Lippmann, M. (1991). Health effects of ambient ozone on healthy children at a summer camp. In
      RL Berglund; DR Lawson; DJ McKee (Eds.), Tropospheric Ozone and the Environment: Papers from an
      International Conference; March 1990; Los Angeles, CA(pp. 83-89). Pittsburgh, PA: Air & Waste
      Management Association.
Spicer. CW: Joseph. DW: Ollison. WM. (2010). A re-examination of ambient air ozone monitor interferences. J
      Air Waste Manag Assoc 60: 1353-1364. http://dx.doi.org/10.3155/1047-3289.60.11.1353.
Sram, RJ; Binkova, B; Rossner, P; Rubes, J; Topinka, J;  Dejmek, J. (1999). Adverse reproductive outcomes
      from exposure to environmental mutagens. Mutat  Res-Fundam Mol Mech Mutagen 428: 203-215.
      http://dx.doi. org/10.1016/51383-5742(99)00048-4.
SSDAN CensusScope. (Social Science Data Analysis Network, CensusScope). (201 Oa). United States: Age
      distribution [Data Set]. Ann Arbor, Michigan: Social Science Data Analysis Network. Retrieved  from
      http://www.censusscope.org/us/chart  age.html
SSDAN CensusScope. (Social Science Data Analysis Network, CensusScope). (201 Ob). United States:
      Population by race  [Data Set]. Ann Arbor, Michigan.  Retrieved from
      http://www.censusscope.org/us/chart  race.html
SSDAN CensusScope. (Social Science Data Analysis Network, CensusScope). (2010c). United States: Poverty
      by age [Data Set]. Ann Arbor, Michigan. Retrieved from
      http://www.censusscope.org/us/chart  povertv.html
Staehelin, J; Thudium, J; Buehler, R; Volz-Thomas, A; Graber, W. (1994). Trends in surface ozone
      concentrations at Arosa (Switzerland). Atmos Environ 28: 75-87.  http://dx.doi.org/10.1016/1352-
      2310(94)90024-8.
Stafoggia, M; Forastiere, F; Faustini, A; Biggeri, A; Bisanti,  L;  Cadum, E; Cernigliaro, A; Mallone, S; Pandolfi,  P;
      Serinelli. M: Tessari. R: Vigotti. MA: Perucci. CA. (2010). Susceptibility factors to ozone-related mortality:
      A population-based case-crossover analysis. Am J Respir Grit Care Med 182: 376-384.
      http://dx.doi.org/10.1164/rccm.200908-1269OC.
Stampfli, A; Fuhrer, J. (2010). Spatial heterogeneity confounded ozone-exposure experiment in semi-natural
      grassland. Oecologia 162: 515-522. http://dx.doi.org/10.1007/s00442-009-1462-2.
Stedman, DH;  Daby, EE; Stuhl,  F; Niki, H. (1972). Analysis of ozone and nitric oxide by a chemiluminescent
      method  in laboratory and atmospheric studies of photochemical smog. J Air Waste Manag Assoc 22:
      260-263.
Stedman, JR; Kent, AJ. (2008). An analysis of the spatial patterns of human health related surface ozone
      metrics  across the UK in 1995, 2003 and 2005. Atmos Environ 42: 1702-1716.
Steinvil. A: Kordova-Biezuner. L: Shapira. I: Berliner. S: Rogowski. O. (2008). Short-term exposure to air
      pollution and inflammation-sensitive biomarkers. Environ Res 106: 51-61.
Steinvil, A; Fireman, E; Kordova-Biezuner, L; Cohen, M; Shapira, I; Berliner, S; Rogowski, O. (2009).
      Environmental air pollution has decremental effects on pulmonary function test parameters up to one
      week after exposure. Am J Med Sci 338:  273-279. http://dx.doi.org/10.1097/MAJ.ObOl 3e3181 adb3ed.
Stenfors.  N: Pourazar. J: Blomberg. A: Krishna.  MT:  Mudwav. I: Helledav. R: Kelly. FJ:  Frew. AJ: Sandstrom. T.
      (2002). Effect of ozone on bronchial mucosal inflammation in asthmatic and healthy subjects. Respir Med
      96: 352-358.
Draft - Do Not Cite or Quote                       11-78                                 September 2011

-------
Stenfors, N: Bosson, J: Helleday, R: Behndig, AF: Pourazar, J: Tornqvist, H: Kelly, FJ: Frew, AJ: Sandstrom, T:
      Mudway, IS: Blomber, A. (2010). Ozone exposure enhances mast-cell inflammation in asthmatic airways
      despite inhaled corticosteroid therapy. Inhal Toxicol 22: 133-139.
      http://dx.doi.org/10.3109/08958370903005736.
Sterner-Kock. A: Kock. M: Braun. R: Hyde. DM. (2000). Ozone-induced epithelial injury in the ferret is similar to
      nonhuman primates. Am J Respir Crit Care Med 162: 1152-1156.
Stevens, R: Pinto, J: Mamane, Y: Ondov, J: Abdulraheem, M: AI-Majed, N: Sadek,  M: Gofer, W: Ellenson, W:
      Kellogg. R. (1993). Chemical and physical properties of emissions from Kuwaiti oil fires. Water Sci
      Technol 27: 223-233.
Stevenson, PS. (2004). Radiative forcing from aircraft NO emissions: Mechanisms and seasonal dependence. J
      Geophys Res 109: D17307.  http://dx.doi.org/10.1029/2004JD004759.
Stevenson, PS: Dentener, FJ: Schultz, MG: Ellingsen, K: Van Noije, TPC: Wild, O: Zeng, G: Amann, M:
      Atherton, CS: Bell, N: Bergmann, DJ: Bey, I: Butler, T: Cofala, J: Collins, WJ: Derwent, RG: Doherty, RM:
      Drevet. J: Eskes. HJ: Fiore. AM: Gauss. M: Hauglustaine. DA: Horowitz. LW: Isaksen. ISA: Krol. MC:
      Lamarque. JF: Lawrence. MG: Montanaro. V: Mullen JF: Pitari. G:  Prather. MJ: Pvle. JA: Rast. S:
      Rodriguez, JM: Sanderson, MG: Savage, NH: Shindell, DT: Strahan, SE: Sudo, K: Szopa, S. (2006).
      Multimodel ensemble simulations of present-day and near-future tropospheric ozone.  J Geophys Res
      111: D08301. http://dx.doi.orq/10.1029/2005JD006338.
Stewart, CA: Black, VJ: Black,  CR:  Roberts, JA. (1996). Direct effects of ozone on the reproductive
      development of Brassica species. J Plant Physiol 148: 172-178.
Stewart, CA. (1998) Impact of ozone on the reproductive biology of Brassica campestris L and Plantago major
      L Loughborough University of Technology, England. Retrieved from
      http://ethos.bl.uk/OrderDetails.do?did=1&uin=uk.bl. ethos. 299673
Stieb. DM: Szvszkowicz.  M: Rowe.  BH: Leech. JA. (2009). Air pollution and emergency department visits for
      cardiac and respiratory conditions: A multi-city time-series analysis. Environ Health Global Access Sci
      Source 8: 25. http://dx.doi.org/10.1186/1476-069X-8-25.
Stockstill. BL: Chang. L. -Y: Menache. MG: Mellick. PW: Mercer. RR: Crapo. JD. (1995). Bronchiolarized
      metaplasia and interstitial fibrosis in rat lungs chronically exposed to high ambient levels of ozone.
      Toxicol Appl Pharmacol 134: 251-263.
Stoelken, G: Pritsch, K: Simon, J: Mueller, CW: Grams, TEE: Esperschuetz, J: Gayler, S: Buegger, F:
      Brueggemann, N:  Meier, R: Zeller, B: Winkler, JB: Rennenberg, H. (2010). Enhanced ozone exposure of
      European beech (Fagus sylvatica) stimulates nitrogen mobilization from leaf litter and nitrogen
      accumulation in the soil. Plant Biosystems 144: 537-546. http://dx.doi.org/10.1080/11263500903429346.
Stokinger, HE.  (1962). Effects of air pollution in animals. In AC Stern (Ed.), Air pollution (Vol.  1, pp. 282-334).
      New York, NY: Academic Press.
Street. NR: James. TM: James. T: Mikael. B: Jaakko. K: Mark. B: Taylor. G. (2011). The physiological,
      transcriptional and genetic responses of an ozone-sensitive and an ozone tolerant poplar and selected
      extremes of their F2 progeny. Environ Pollut 159: 45-54. http://dx.doi.Org/10.1016/i.envpol.2010.09.027.
Strickland, MJ: Klein, M: Correa, A: Reller,  MD: Mahle, WT: Riehle-Colarusso, TJ: Botto, LD: Flanders, WD:
      Mulholland, JA: Siffel, C: Marcus, M: Tolbert,  PE. (2009). Ambient air pollution and cardiovascular
      malformations in Atlanta, Georgia, 1986-2003. Am J Epidemiol 169: 1004-1014.
Strickland. MJ: Darrow. LA: Klein. M: Flanders. WD:  Sarnat. JA: Waller. LA: Sarnat. SE:  Mulholland. JA: Tolbert.
      PE. (2010). Short-term associations between ambient air pollutants and pediatric asthma emergency
      department visits.  Am J Respir Crit Care Med 182: 307-316. http://dx.doi.org/10.1164/rccm.200908-
      1201OC.
Strickland, MJ: Darrow, LA: Mulholland, JA: Klein, M: Flanders, WD: Winguist, A: Tolbert, PE. (2011).
      Implications of different approaches for characterizing ambient air pollutant concentrations within the
      urban airshed for time-series studies and health benefits analyses. Environ Health Global Access Sci
      Source 10: 36. http://dx.doi.org/10.1186/1476-069X-10-36.
Studzinski. GP: Moore. DC. (1995). Sunlight-can it prevent as well as cause cancer? Cancer Res 55: 4014-
      4022.
Stutz, J: Ackermann, R: Fast, JD: Barrie, L. (2002). Atmospheric reactive chlorine and bromine at the Great Salt
      Lake, Utah. Geophys Res Lett 29: 1380. http://dx.doi.org/10.1029/2002GL014812.
Stutz. J: Oh. HJ: Whitlow. SI: Anderson. C: Dibb. JE: Flvnn. JH: Rappengluck. B: Lefe. B. (2009). Simultaneous
      DOAS and mist-chamber 1C measurements of HONO in Houston, TX. Atmos Environ TBD: TBD.
      http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.003.
Stylianou, M: Nicolich, MJ. (2009). Cumulative effects and threshold levels in air pollution mortality:  Data
      analysis of nine large US cities using the NMMAPS dataset. Environ Pollut 157: 2216-2223.
      http://dx.doi.0rg/10.1016/i.envpol.2009.04.011.
Suh. HH: Zanobetti. A. (2010). Exposure error masks the relationship between traffic-related air pollution and
      heart rate variability [Erratum]. J Occup Environ Med 52: 1138.
      http://dx.doi. org/10.1097/JOM.Ob013e3181 fd2632.
Sun. J: Koto. H: Chung. KF. (1997).  Interaction of ozone and allergen challenges on bronchial responsiveness
      and inflammation in sensitised guinea pigs. Int Arch Allergy Immunol 112: 191-195.
Draft - Do Not Cite or Quote                        11-79                                  September 2011

-------
Symons, JM: Wang, L: Guallar, E: Howell, E: Dominici, F: Schwab, M: Ange, BA: Samet, J: Ondov, J: Harrison,
      D: Geyh, A. (2006). A case-crossover study of fine particulate matter air pollution and onset of congestive
      heart failure symptom exacerbation leading to hospitalization. Am J Epidemiol 164:  421-433.
Szyszkowicz, M. (2008). Ambient air pollution and daily emergency department visits for ischemic stroke in
      Edmonton,  Canada. Int J Occup Med Environ Health 21: 295-300. http://dx.doi.org/10.2478/v10001-008-
      0029-5.
Tager, IB: Balmes, J: Lurmann, F: Ngo, L: Alcorn, S: Kunzli, N. (2005). Chronic exposure to ambient ozone and
      lung function in young adults. Epidemiology 16: 751-759.
      http://dx.doi.ora/10.1097/01.ede.0000183166.68809.bO.
Takeuchi, C: Galve, R: Nieva, J: Witter, DP: Wentworth, AD: Troseth, RP: Lerner, RA: Wentworth P, J, r. (2006).
      Proatherogenic effects of the cholesterol ozonolysis products, atheronal-Aand atheronal-B. Biochemistry
      45:  7162-7170. http://dx.doi.org/10.1021/bi0604330.
Talhelm, AF: Pregitzer, KS: Zak, DR. (2009). Species-specific responses to atmospheric carbon dioxide and
      tropospheric ozone mediate changes in soil carbon. Ecol Lett 12: 1219-1228.
      http://dx.doi.ora/10.1111/i.1461-0248.2009.01380.x.
Tamaoki, M: Nakajima, N: Kubo, A: Aono, M:  Matsuyama, T: Sail, H. (2003). Transcriptome analysis of OS-
      exposed Arabidopsis reveals that multiple signal pathways act mutually antagonistically to induce gene
      expression. Plant Mol Biol 53: 443-456. http://dx.doi.Org/10.1023/B:PLAN.0000019064.55734.52.
Tamer, L: Calikoglu, M: Ates, NA: Yildirim, H:  Ercan, B: Saritas, E: Unlu, A: Atik, U.  (2004).  Glutathione-S-
      transferase gene polymorphisms (GSTT1, GSTM1, GSTP1) as increased risk factors for asthma.
      Respirology 9: 493-498.
Tang, Q: Prather, MJ: Hsu, J. (2011). Stratosphere-troposphere exchange ozone flux related to deep
      convection. Geophys Res Lett 38: L03806. http://dx.doi.org/10.1029/201OGL046039.
Tanimoto.  H: Mukai. H: Hashimoto. S:  Morris. JE. (2006).  Intercomparison of ultraviolet photometry and gas-
      phase titration techniques for ozone reference standards at ambient levels.  J Geophys Res 111:  D16313.
      http://dx.doi. org/10.1029/2005JD006983.
Tanimoto.  H. (2009). Increase in springtime tropospheric ozone at a mountainous site in Japan for the period
      1998-2006. Atmos Environ 43: 1358-1363. http://dx.doi.Org/10.1016/i.atmosenv.2008.12.006.
Tankerslev, CG: Kleeberger, SR.  (1994). Ozone-induced inflammation and altered  ventilation in  genetically
      susceptible mice: A comparison of acute and subacute exposures. Toxicol Lett 72: 279-289.
Tankerslev, CG: Peng, RD: Bedga, D:  Gabrielson, K: Champion, HC. (2010). Variation  in echocardiographic and
      cardiac hemodynamic effects of PM and ozone inhalation exposure in strains related to Nppa and Npr1
      gene knock-out mice. Inhal Toxicol 22: 695-707. http://dx.doi.org/10.3109/08958378.2010.487549.
Tarasick, DW: Fioletov, VE: Wardle, PI: Kerr, JB: Davies, J. (2005). Changes in the vertical distribution of ozone
      over Canada from ozonesondes: 1980-2001. J Geophys Res 110: D02304.
      http://dx.doi. org/10.1029/2004JD004643.
Tarasick, DW: Slater, R. (2008). Ozone in the troposphere: Measurements,  climatology, budget, and trends.
      Atmos Ocean 46: 93-115.  http://dx.doi.org/10.3137/ao.460105.
Taubman,  BF:  Marufu,  LT: Piety, CA: Doddridge, BG: Stehr, JW:  Dickerson, RR. (2004). Airborne
      characterization of the chemical, optical, and meteorological properties, and origins of a combined
      ozone-haze episode over the eastern United States. J Atmos Sci 61: 1781-1793.
Taubman.  BF:  Hains. JC: Thompson. AM: Marufu. LT: Doddridge. BG: Stehr. JW:  Piety. CA: Dickerson. RR.
      (2006). Aircraft vertical profiles of trace gas and aerosol pollution over the mid-Atlantic United States:
      Statistics and meteorological cluster analysis. J Geophys Res 111: D10S07.
      http://dx.doi. org/10.1029/2005JD006196.
Taylor-Clark, TE: McAlexander, MA: Nassenstein, C: Sheardown, SA: Wilson, S: Thornton, J:  Carr, MJ:  Undem,
      BJ. (2008).  Relative contributions of TRPA1 and TRPV1 channels in the activation of vagal
      bronchopulmonary C-fibres by the endogenous autacoid 4-oxononenal. J Physiol 586:  3447-3459.
      http://dx.doi.org/10.1113/iphysiol.2008.153585.
Taylor-Clark. TE: Undem. BJ. (2010).  Ozone activates airway nerves via the selective stimulation of TRPA1 ion
      channels. J Physiol 588: 423-433. http://dx.doi.org/10.1113/iphvsiol.2009.183301.
Taylor, AB: Lee, GM: Nellore, K: Ben-Jebria, A: Ultman, JS. (2006). Changes in the carbon dioxide expirogram
      in response to ozone exposure. Toxicol Appl Pharmacol 213: 1-9.
      http://dx.doi.0rg/10.1016/i.taap.2005.09.009.
Taylor, AB: Borhan, A: Ultman, JS. (2007). Three-dimensional simulations of reactive gas uptake in single
      airway bifurcations. Ann Biomed Eng 35: 235-249. http://dx.doi.org/10.1007/s10439-006-9195-4.
Temple, PJ: Kupper, RS: Lennox, RW: Rohr, K. (1988). Injury and yield responses of differentially irrigated
      cotton to ozone. Agron J 80: 751-755.  http://dx.doi.org/10.2134/agroni1988.00021962008000050011x.
Temple. PJ: Riechers. GH: Miller. PR. (1992). Foliar injury responses of ponderosa pine seedlings to ozone, wet
      and dry acidic deposition,  and drought. Environ Exp Bot 32: 101-113. http://dx.doi.org/10.1016/0098-
      8472(92)90035-Z.
ten Berge. O: van  Weelden. H: Bruiinzeel-Koomen. C: de Bruin-Weller. MS: Sigurdsson. V. (2009). Throwing a
      light on photosensitivity in atopic dermatitis: A retrospective study. Am J Clin Dermatol 10: 119-123.
      http://dx.doi.org/10.2165/00128071-200910020-00004.
Draft - Do Not Cite or Quote                        11-80                                  September 2011

-------
Tepper, Jl_: Weiss, B: Cox, C. (1982). Microanalysis of ozone depression of motor activity. Toxicol Appl
      Pharmacol 64: 317-326.
Tepper, Jl_: Weiss, B: Wood, RW. (1983). Behavioral indices of ozone exposure. In.
Tepper, JS: Weiss, B: Wood, RW. (1985). Alterations in behavior produced by inhaled ozone or ammonia.
      Toxicol Sci 5: 1110-1118.
Tepper. JS: Costa. PL:  Lehmann. JR: Weber. MF: Hatch. GE. (1989). Unattenuated structural and biochemical
      alterations in the rat lung during functional adaptation to ozone. Am J Respir Crit Care Med 140: 493-
      501.
Tepper. JS: Costa. PL:  Fitzgerald. S: Doerfler. PL: Bromberg. PA. (1993). Role of tachykinins in ozone-induced
      acute lung injury in guinea pigs. J Appl Physiol 75: 1404-1411.
Thaller, El: Petronella,  SA: Hochman, D: Howard, S: Chhikara, RS: Brooks, EG. (2008). Moderate increases in
      ambient PM2.5 and ozone are associated with lung function decreases in beach lifeguards. J Occup
      Environ Med 50: 202-211. http://dx.doi.org/10.1097/JOM.Ob013e31816386b4.
Theis. N: Raguso.  RA.  (2005). The effect of pollination on floral fragrance in thistles. J Chem Ecol 31: 2581-
      2600.
Thieden,  E: Philipsen, PA: Sandbv-Moller, J: Heydenreich, J: Wulf,  HC. (2004a). Proportion of lifetime UV dose
      received by children, teenagers and adults based on time-stamped personal dosimetry.  J  Invest
      Dermatol 123: 1147-1150.
Thieden,  E: Philipsen, PA: Hevdenreich, J: Wulf,  HC. (2004b). UV radiation exposure related to age, sex,
      occupation, and sun behavior based on time-stamped personal dosimeter readings. Arch  Dermatol 140:
      197-203.
Thomas,  VFD: Braun, S:  Fluckiger, W. (2005). Effects of simultaneous ozone exposure and nitrogen loads  on
      carbohydrate concentrations,  biomass, and growth of young spruce trees (Picea abies). Environ Pollut
      137: 507-516. http://dx.doi.Org/10.1016/i.envpol.2005.02.002.
Thomas,  VFD: Braun, S:  Fluckiger, W. (2006). Effects of simultaneous ozone exposure and nitrogen loads  on
      carbohydrate concentrations,  biomass, growth, and nutrient  concentrations of young beech trees (Fagus
      sylvatica). Environ Pollut 143: 341-354. http://dx.doi.Org/10.1016/i.envpol.2005.11.036.
Thompson, AM. (1992). The oxidizing capacity of the Earth's atmosphere: Probable past and future changes.
      Science 256: 1157-1165. http://dx.doi.org/10.1126/science.256.5060.1157.
Thompson, AM: Chappellaz, JA: Fung, IY: Kucsera, TL. (1993). The atmospheric CH4 increase since the last
      glacial maximum (2) Interactions with oxidants. Tellus B Chem Phys Meteorol 45: 242-257.
      http://dx.doi. org/10.1034/1.1600-0889.1993.t01 -2-00003.X.
Thompson. AM: Hudson.  RD. (1999). Tropical tropospheric ozone (TTO) maps from Nimbus 7  and Earth Probe
      TOMS by the modified-residual method: Evaluation with sondes,  ENSO signals, and trends from Atlantic
      regional time series. J Geophys Res 104:  26961-26975. http://dx.doi.org/10.1029/1999JD900470.
Thompson. AM: Stone. JB: Witte. JC: Miller.  SK:  Oltmans. SJ: Kucsera.  TL: Ross. KL: Pickering. KE: Merrill. JT:
      Forbes, G: Tarasick, DW: Joseph, E: Schmidlin, FJ: McMillan, WW: Warner, J: Hintsa, EJ: Johnson,  JE.
      (2007). Intercontinental Chemical Transport Experiment Ozonesonde Network study (IONS) 2004: 2
      Tropospheric ozone budgets and variability over northeastern North America. J Geophys Res 112:
      D12S13. http://dx.doi.org/10.1029/2006JD007670.
Thompson. AM: Zanobetti. A: Silverman. F: Schwartz. J: Coull. B: Urch.  B: Speck. M: Brook. JR: Manno. M:
      Gold.  DR. (2010).  Baseline Repeated Measures from Controlled  Human Exposure Studies: Associations
      between Ambient Air Pollution Exposure and the Systemic Inflammatory Biomarkers IL-6 and Fibrinogen.
      Environ Health Perspect 118:  120-124. http://dx.doi.org/10.1289/ehp.0900550.
Thomson. E: Kumarathasan. P:  Goegan. P: Aubin. RA: Vincent. R.  (2005). Differential  regulation of the lung
      endothelin system by urban particulate matter and ozone. Toxicol Sci 88: 103-113.
Thomson, E: Kumarathasan, P:  Vincent, R. (2006). Pulmonary expression of preproET-1 and preproET-3
      mRNAs is altered  reciprocally in rats after inhalation of air pollutants. Exp Biol Med 231: 979-984.
Thomson, EM: Kumarathasan, P: Calderon-Garciduenas, l_: Vincent, R. (2007). Air pollution alters brain and
      pituitary endothelin-1 and inducible nitric oxide synthase gene expression. Environ Res  105: 224-233.
      http://dx.doi.0rg/10.1016/i.envres.2007.06.005.
Thornton, JA: Kercher,  JP: Riedel, TP: Wagner, Nl_: Cozic, J: Hollowav,  JS: Dube, WP: Wolfe, GM: Quinn, PK:
      Middlebrook. AM: Alexander. B: Brown. SS. (2010). A large atomic chlorine source  inferred from mid-
      continental reactive nitrogen chemistry. Nature 464: 271-274. http://dx.doi.org/10.1038/nature08905.
Thurston, GD: Lippmann, M: Scott, MB: Fine, JM. (1997). Summertime haze air pollution and children with
      asthma. Am J Respir Crit Care Med 155: 654-660.
Tian, H:  Melillo, J: Lu, C:  Kicklighter,  D: Liu, M: Ren, W: Xu, X: Chen, G: Zhang, C: Pan, S: Liu, J: Running, S.
      (2011). China's terrestrial carbon balance: Contributions from multiple global change factors. Global
      Biogeochem Cycles 25: GB1007. http://dx.doi.org/10.1029/201OGB003838.
Tingev. DT:  Hogsett. WE: Lee. EH: Herstrom. AA: Azevedo. SH. (1991). An evaluation  of various alternative
      ambient ozone standards based on crop yield loss data. In RL Berglund; DR Lawson; DJ McKee (Eds.),
      Tropospheric Ozone and the Environment (pp. 272-288). Los Angeles, CA: Air & Waste Management
      Association.
Draft - Do Not Cite or Quote                       11-81                                 September 2011

-------
Tingey, DT: McVeety, BD: Waschmann, R: Johnson, MG: Phillips, PL: Rygiewicz, PT: Olszyk, DM. (1996). A
      versatile sun-lit controlled-environment facility for studying plant and soil processes. J Environ Qual 25:
      614-625.
Tingey, DT: Rodecap, KD: Lee, EH: Hogsett, WE: Gregg, JW. (2002). Pod development increases the ozone
      sensitivity of Phaseolus vulgaris. Water Air Soil Pollut 139: 325-341.
Tingev. DT: Hogsett. WE: Lee. EH: Laurence. JA. (2004). Stricter ozone ambient air quality standard has
      beneficial effect on ponderosa pine in California. J Environ Manage 34: 397-405.
Tingey. DT: Johnson. MG: Lee. EH: Wise. C: Waschmann. R: Olszvk. DM: Watrud. LS: Donegan. KK. (2006).
      Effects of elevated CO2 and O3 on soil respiration under ponderosa pine. Soil Biol Biochem 38: 1764-
      1778. http://dx.doi.0rg/10.1016/i.soilbio.2005.12.003.
Tissue, DT: Thomas, RB: Strain,  BR.  (1997). Atmospheric CO2 enrichment increases growth and
      photosynthesis of Pinus taeda: A4 year experiment in the field. Plant Cell Environ 20: 1123-1134.
      http://dx.doi.0rg/10.1046/i.1365-3040.1997.d01-140.x.
Tissue. DT: Griffin. KL: Ball. T. (1999). Photosynthetic adjustment in field-grown ponderosa pine trees after six
      years of exposure to elevated CO2. Tree Physiol 19: 221-228.
Tjoelker, MG: Volin, JC: Oleksyn, J: Reich, PB. (1995). Interaction of ozone pollution and light effects on
      photosynthesis in a forest canopy experiment. Plant Cell Environ 18: 895-905.
      http://dx.doi.0rg/10.1111/i.1365-3040.1995.tb00598.x.
Tobiessen, P. (1982). Dark opening of stomata in successional trees. Oecologia 52: 356-359.
      http://dx.doi. org/10.1007/BF00367959.
Toet, S: Ineson,  P: Peacock, S: Ashmore, M. (2011). Elevated ozone reduces methane emissions from peatland
      mesocosms. Global Change Biol 17: 288-296. http://dx.doi.Org/10.1111/i.1365-2486.2010.02267.x.
Tolbert. PE: Klein.  M: Peel. JL: Sarnat. SE: Sarnat. JA. (2007). Multipollutant modeling  issues in a study of
      ambient air quality and emergency department visits in Atlanta. J Expo Sci Environ Epidemiol 17: S29-
      S35. http://dx.doi.org/10.1038/si.ies.7500625.
long. D: Mathur. R: Schere. K: Kang. D: Yu. S. (2007). The use of air quality forecasts to assess impacts of air
      pollution on crops: Methodology and case study. Atmos Environ 41: 8772-8784.
      http://dx.doi.0rg/10.1016/i.atmosenv.2007.07.060.
Tong, DQ: Mauzerall, PL. (2008). Summertime state-level source-receptor relationships between nitrogen
      oxides emissions and surface ozone concentrations over the continental United  States. Environ Sci
      Technol 42: 7976-7984. http://dx.doi.org/10.1021/es7027636.
Topa. MA: Vanderklein.  DW: Corbin. A. (2001). Effects of elevated ozone and low light on diurnal and seasonal
      carbon gain in sugar maple. Plant Cell Environ 24: 663-677.
Torres, A: Utell, MJ: Morow, PE: Voter, KZ: Whitin, JC:  Cox, C: Loonev, RJ: Speers, DM: Tsai, Y: Frampton,  MW.
      (1997). Airway inflammation in smokers and nonsmokers with varying responsiveness to ozone. Am J
      RespirCrit Care  Med 156: 728-736.
Torsethaugen, G: Pell, EJ: Assmann,  SM. (1999). Ozone inhibits guard cell K+ channels implicated in stomatal
      opening. PNAS 96: 13577-13582.
Tosti, N: Pasgualini, S: Borgogni, A: Ederli, l_: Falistocco, E: Crispi, S: Paolocci, F. (2006). Gene expression
      profiles of OS-treated Arabidopsis plants. Plant Cell  Environ 29: 1686-1702.
      http://dx.doi.0rg/10.1111/i.1365-3040.2006.01542.x.
Tovalin. H: Valverde. M: Morandi. MT: Blanco. S: Whitehead. L: Roias. E.  (2006). DMA  damage in outdoor
      workers occupationally exposed to environmental air pollutants. Occup Environ  Med 63: 230-236.
Trainer. M: Parrish. DP: Buhr. MP: Norton. RB: Fehsenfeld. FC: Anlauf. KG:  Bottenheim. JW: Tang. YZ: Wiebe.
      HA: Roberts. JM: Tanner. RL: Newman. L: Bowersox. VC: Meagher. JF: Olszvna. KJ: Rodgers. MO:
      Wang, T:  Berresheim, H:  Demerjian, KL: Roychowdhury, UK. (1993).  Correlation of ozone with NOy in
      photochemically  aged air. J Geophys Res 98: 2917-2925.
Tran, MU: Weir, AJ: Fanucchi, MV: Rodriguez, AE: Pantle,  LM: Smiley-Jewell, SM: Van Winkle, LS: Evans, MJ:
      Miller, LA: Schelegle, ES: Gershwin, LJ: Hyde, DM:  Plopper, CG. (2004). Smooth muscle hypertrophy in
      distal airways of sensitized infant rhesus monkeys exposed to house dust mite allergen. Clin Exp Allergy
      34: 1627-1633. http://dx.doi.Org/10.1111/i.1365-2222.2004.02057.x.
Trenga, CA:  Koenig, JQ: Williams, PV. (2001).  Dietary  antioxidants and ozone-induced bronchial
      hyperresponsiveness in adults with asthma. Arch Environ Occup Health 56: 242-249.
Trevisani. M: Siemens. J: Materazzi. S: Bautista. DM: Nassini. R: Campi.  B:  Imamachi.  N: Andre. E: Patacchini.
      R: Cottrell, GS: Gatti, R: Basbaum, Al: Bunnett,  NW: Julius,  D: Geppetti, P. (2007). 4-Hydroxynonenal, an
      endogenous aldehyde, causes pain and neurogenic inflammation through activation of the irritant
      receptor TRPA1. PNAS 104: 13519-13524. http://dx.doi.org/10.1073/pnas.07Q5923104.
Triche, EW: Gent, JF: Holford, TR: Belanger, K: Bracken, MB: Beckett, WS: Naeher, L:  McSharrv, J, -E:
      Leaderer. BP (2006). Low-level ozone exposure and respiratory symptoms in infants. Environ Health
      Perspect 114: 911-916. http://dx.doi.org/10.1289/ehp.8559.
Tsai, S, -S: Chen, C, -C: Hsieh, H, -J: Chang, C, -C: Yang, C, -Y.  (2006). Air pollution and postneonatal mortality
      in a tropical city:  Kaohsiung, Taiwan. Inhal Toxicol 18: 185-189.
Tsuiino. I: Kawakami. Y: Kaneko. A. (2005). Comparative simulation of gas transport in  airway models of rat,
      dog, and  human. Inhal Toxicol  17: 475-485. http://dx.doi.org/10.1080/08958370590964476.
Draft - Do Not Cite or Quote                        11-82                                  September 2011

-------
Turner, RM: Muscatello, DJ: Zheng, W: Willmore, A: Arendts, G. (2007). An outbreak of cardiovascular
      syndromes requiring urgent medical treatment and its association with environmental factors: an
      ecological study. Environ Health 6: 37. http://dx.doi.org/10.1186/1476-069X-6-37.
Turnipseed, AA: Burns, SP: Moore, DJP: Hu, J: Guenther, AB: Monson, RK. (2009). Controls over ozone
      deposition to a high elevation subalpine forest. Agr Forest Meteorol 149: 1447-1459.
      http://dx.doi.org/10.1016/Laqrformet.2009.04.001.
Tyler, WS: Tyler, NK: Last, JA: Gillespie, MJ: Barstow, TJ. (1988). Comparison of daily and seasonal exposures
      of young monkeys to ozone. Toxicology 50: 131-144.
U.S. Census Bureau.  (2010). U.S. population projections [Data Set]. Retrieved from
      http://www.census.gov/population/www/proiections/proiectionsagesex.html
U.S. Census Bureau.  (2011). U.S. Census Bureau, from http://www.census.gov/
U.S. EPA.  (U.S. Environmental Protection Agency). (1971). National primary and secondary ambient air quality
      standards. Fed Reg 36: 8186-8201.
U.S. EPA.  (U.S. Environmental Protection Agency). (1978a). Air quality criteria for ozone and other
      photochemical oxidants. (EPA/600/8-78/004). Washington, DC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1978b). Photochemical oxidants: Proposed revisions to the
      national ambient air quality standards.  Fed Reg 43: 26962-26971.
U.S. EPA.  (U.S. Environmental Protection Agency). (1979a). National primary and secondary ambient air quality
      standards: Revisions to the national ambient air quality standards for photochemical oxidants.  Fed  Reg
      44:  8202-8237.
U.S. EPA.  (U.S. Environmental Protection Agency). (1979b). Transfer standards for the calibration of ambient air
      monitoring analyzers for ozone: Technical  assistance document.  (EPA-600/4-79-056). Research Triangle
      Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1982). Air quality criteria document for ozone and other
      photochemical oxidants. Fed Reg 47: 11561.
U.S. EPA.  (U.S. Environmental Protection Agency). (1983). Review of the national ambient air quality standards
      for ozone. Fed Reg 48: 38009.
U.S. EPA.  (U.S. Environmental Protection Agency). (1984). Air quality criteria for ozone and other
      photochemical oxidants, Vol. 3. (EPA/600/8-84/020A). Research  Triangle Park,  NC.
      http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=2000AVEV.txt.
U.S. EPA.  (U.S. Environmental Protection Agency). (1986). Air quality criteria for ozone and other
      photochemical oxidants. (EPA-600/8-84-020aF - EPA-600/8-84-020eF). Research Triangle Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1989). Review of the national ambient air quality standards
      for ozone: Assessment of scientific and technical information: OAQPS staff report. (EPA/450/2-92-001).
      Research Triangle Park, NC. http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=2000LOW6.txt.
U.S. EPA.  (U.S. Environmental Protection Agency). (1992). National ambient air quality standards for ozone;
      Proposed decision. Fed Reg 57: 35542-35557.
U.S. EPA.  (U.S. Environmental Protection Agency). (1993). National ambient air quality standards for ozone -
      Final decision.  Fed Reg 58: 13008-13019.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996a). Air quality criteria for ozone and related
      photochemical oxidants. (EPA/600/P-93/004AF). Research Triangle  Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996b). Air quality criteria for ozone and related
      photochemical oxidants, Vol. II of III. (EPA/600/P-93/004BF). Research Triangle Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996c). Air quality criteria for ozone and related
      photochemical oxidants, Vol. Ill of III. (EPA/600/P-93/004cF). Research Triangle Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996d). National ambient air quality standards for ozone:
      Proposed decision. Fed Reg 61: 65716-65750.
U.S. EPA.  (U.S. Environmental Protection Agency). (1996e). Review of  national ambient air quality standards
      for ozone: Assessment of scientific and technical information: OAQPS staff paper. (EPA/452/R-96/007).
      Research Triangle Park, NC.
U.S. EPA.  (U.S. Environmental Protection Agency). (1997). National ambient air quality standards for ozone;
      final rule. Fed Reg 62: 38856-38896.
U.S. EPA.  (U.S. Environmental Protection Agency). (1998). Guidelines for ecological risk assessment.
      (EPA/630/R-95/002F). Washington, DC. http://www.epa.gov/raf/publications/guidelines-ecological-risk-
      assessment.htm.
U.S. EPA.  (U.S. Environmental Protection Agency). (2000). Air quality criteria for ozone and related
      photochemical oxidants. Fed Reg 65: 57810.
U.S. EPA.  (U.S. Environmental Protection Agency). (2001). National ambient air quality standards for ozone:
      Proposed response to remand. Fed Reg 66: 57268-57292.
U.S. EPA.  (U.S. Environmental Protection Agency). (2002). A framework for assessing and reporting on
      ecological condition: An SAB report. Washington, DC.
U.S. EPA.  (U.S. Environmental Protection Agency). (2003). National ambient air quality standards for ozone:
      Final response to remand. Fed Reg 68: 614-645.
Draft - Do Not Cite or Quote                       11-83                                 September 2011

-------
U.S. EPA. (U.S. Environmental Protection Agency). (2004). Final rule to implement the 8-hour ozone national
      ambient air quality standard-phase 1. Fed Reg 69: 23951-24000.
U.S. EPA. (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment.
      (EPA/630/P-03/001F). Washington, DC. http://www.epa.gov/cancerguidelines/.
U.S. EPA. (U.S. Environmental Protection Agency). (2006a). Aging and toxic response:  Issues relevant to risk
      assessment. (EPA/600/P-03/004A). Washington, DC.
      http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=156648.
U.S. EPA. (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
      photochemical oxidants. (EPA/600/R-05/004AF). Research Triangle Park, NC: U.S. Environmental
      Protection Agency, Office of Research and Development.
      http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923.
U.S. EPA. (U.S. Environmental Protection Agency). (2007a). National ambient air quality standards for ozone.
      Fed Reg 72:  37818-37919.
U.S. EPA. (U.S. Environmental Protection Agency). (2007b). Review of the national ambient air quality
      standards for ozone: Policy assessment of scientific and technical information: OAQPS staff paper.
      (EPA/452/R-07/003). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2008a). 2005 National Emissions Inventory data and
      documentation, from http://www.epa.gov/ttn/chief/net/2005inventorv.html
U.S. EPA. (U.S. Environmental Protection Agency). (2008b). Integrated science assessment for oxides of
      nitrogen: Health criteria. (EPA/600/R-08/071). Research Triangle Park, NC.
      http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645.
U.S. EPA. (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for sulfur oxides:
      Health criteria. (EPA/600/R-08/047F). Research Triangle Park, NC.
      http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=198843.
U.S. EPA. (U.S. Environmental Protection Agency). (2008d). National air quality: Status and trends through
      2007. (EPA/454/R-08/006). Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2008e). National ambient air quality standards for ozone.
      Fed Reg 73:  16436-16514.
U.S. EPA. (U.S. Environmental Protection Agency). (2008f). Notice of workshop and call for information on
      integrated science assessment for ozone. Fed Reg 73: 56581-56583.
U.S. EPA. (U.S. Environmental Protection Agency). (2009a). Consolidated Human Activity Database, from
      http://www.epa.gov/chadnet1/
U.S. EPA. (U.S. Environmental Protection Agency). (2009b). Human exposure modeling: Air pollutants exposure
      model (APEX/TRIM.Expo  Inhalation), from http://www.epa.gov/ttn/fera/human apex.html
U.S. EPA. (U.S. Environmental Protection Agency). (2009c). Integrated review plan for the ozone National
      Ambient Air Quality Standards review (external review draft). (EPA 452/D-09-001). Washington, DC.
      http://www.epa.gov/ttnnaags/standards/ozone/data/externalreviewdraftO3IRP093009.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2009d). Integrated science assessment for particulate
      matter. (EPA/600/R-08/139F). Research Triangle Park, NC.
      http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546.
U.S. EPA. (U.S. Environmental Protection Agency). (2009e). Risk and exposure assessment for review of the
      secondary National Ambient Air Quality Standards for oxides of nitrogen and oxides of sulfur.
      (EPA/452/R-09/008A). Research Triangle Park,  NC.
      http://www.epa.gov/ttnnaags/standards/no2so2sec/cr rea.html.
U.S. EPA. (U.S. Environmental Protection Agency). (2009f). The  U.S. Environmental Protection Agency's
      strategic plan for evaluating the toxicity of chemicals. (EPA/1 OO/K-09/001). Washington, DC.
      http://www.epa.gov/osa/spc/toxicitytesting/docs/toxtest strategy 032309.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (201 Oa). Air trends: Design values, from
      http://epa.gov/airtrends/values.html
U.S. EPA. (U.S. Environmental Protection Agency). (2010b). Biogenic Emissions Inventory System (BEIS)
      modeling, from http://www.epa.gov/AMD/biogen.html
U.S. EPA. (U.S. Environmental Protection Agency). (2010c). Integrated science assessment for carbon
      monoxide. (EPA/600/R-09/019F).  Research Triangle Park, NC.
      http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=218686.
U.S. EPA. (U.S. Environmental Protection Agency). (2010d). MOBILE6 vehicle emission modeling software,
      from http://www.epa.gov/otag/m6.htm
U.S. EPA. (U.S. Environmental Protection Agency). (2010e). Our nation's air:  Status and trends through 2008.
      (EPA-454/R-09-002). Research Triangle Park, NC.
      http://www.epa.gov/airtrends/2010/report/fullreport.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (201 Of). Transfer standards for calibration of air  monitoring
      analyzers for ozone. (EPA-454/B-10-001). Research Triangle Park, NC.
      http://www.epa.gov/ttn/amtic/files/ambient/gagc/OzoneTransferStandardGuidance.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2011a). AirNow, from http://www.airnow.gov/
Draft - Do Not Cite or Quote                       11-84                                 September 2011

-------
U.S. EPA. (U.S. Environmental Protection Agency). (2011 b). Exposure model for individuals, from
      http://www.epa.gov/heasd/products/emi/emi.html
U.S. EPA. (U.S. Environmental Protection Agency). (2011c). Integrated science assessment for ozone:
      Modeling for policy relevant background concentrations. Research Triangle Park, NC.
U.S. EPA. (U.S. Environmental Protection Agency). (2011d). Map monitoring sites, from
      http://www.epa.qov/airexplorer/monitor kmi.htm
U.S. EPA. (U.S. Environmental Protection Agency). (2011e). MOVES (Motor Vehicle Emission Simulator), from
      http://www.epa.aov/otaq/models/moves/index.htm
Uchivama. I: Simomura. Y: Yokovama. E. (1986). Effects of acute exposure to ozone on heart rate and blood
      pressure of the conscious rat. Environ Res 41: 529-537.
Uchivama, I: Yokovama, E. (1989). Effects of short- and long-term exposure to ozone on heart rate and blood
      pressure of emphysematous rats. Environ Res 48: 76-86.
Uddling, J: Teclaw, RM: Kubiske, ME: Pregitzer, KS: Ellsworth, PS. (2008). Sap flux in pure aspen and mixed
      aspen-birch forests exposed to elevated concentrations of carbon dioxide and ozone. Tree Physiol 28:
      1231-1243. http://dx.doi.orq/18519254.
Uddling, J: Teclaw, RM: Pregitzer, KS: Ellsworth,  PS. (2009). Leaf and canopy conductance in aspen and
      aspen-birch forests under free-air enrichment of carbon dioxide and ozone. Tree Physiol 29: 1367-1380.
      http://dx.doi. orq/10.1093/treephvs/tpp070.
Uddling, J: Hogg, AJ: Teclaw, RM: Carroll, MA: Ellsworth, PS. (2010). Stomatal uptake of O3 in aspen and
      aspen-birch forests under free-air CO2 and O3 enrichment. Environ Pollut 158: 2023-2031.
      http://dx.doi.0rg/10.1016/i.envpol.2009.12.001.
Ullrich, SE.  (2005).  Mechanisms underlying UV-induced immune suppression. Mutat Res 571: 185-205.
      http://dx.doi.0rg/10.1016/i.mrfmmm.2004.06.059.
Ulmer, C; Kopp, M; Ihorst, G; Frischer, T; Forster, J;  Kuehr, J. (1997). Effects of ambient ozone exposures
      during the spring and summer of 1994 on pulmonary function of schoolchildren. Pediatr Pulmonol 23:
      344-353. http://dx.doi.org/10.1002/(SICm099-0496(199705)23:5<344::AIP-PPUL6>3.0.CO:2-K.
Ultman. JS. (1985). Gas transport in  the conducting airways. In LA Engel; M Paiva (Eds.), Gas mixing and
      distribution in the lung (pp. 63-136). New York, NY: Marcel Pekker.
Ultman, JS: Anjilvel, S. (1990).  Monte Carlo simulation of ozone uptake in an asymmetric lung model. In PJ
      Schneck; CL Lucas (Eds.), Biofluid mechanics 3: Proceedings of the third Mid-Atlantic Conference on
      Biofluid Mechanics; October; Blacksburg, VA(pp. 45-52). New York, NY:  New York University Press.
Ultman. JS: Ben-Jebria. A: Hu.  S. -C. (1994). Noninvasive determination of respiratory ozone absorption: The
      bolus-response method. (HEI  Research Report 69). Cambridge,  MA:  Health Effects Institute.
Ultman, JS: Ben-Jebria, A; Arnold,  SF. (2004). Uptake distribution of ozone in human lungs: Intersubject
      variability in physiologic  response. (HEI Research Report 125). Boston, MA: Health Effects Institute.
      http://pubs.healtheffects.org/view.php?id=70.
UNECE. (United Nations Economic Commission for Europe). (1988). ECE critical levels workshop; March; Bad
      Harzburg, Germany. In. Geneva, Switzerland.
UNEP. (United Nations Environment  Programme). (2003). Millennium Ecosystem Assessment: Ecosystems and
      human well-being: A framework for assessment. Washington, PC: Island Press.
UNEP. (United Nations Environment  Programme). (2009). Environmental effects of ozone depletion and its
      interactions with climate change. Nairobi, Kenya.
      http://ozone.unep.org/Assessment  Panels/EEAP/EEAP-Progress-report-2009.pdf.
Unger. N. (2006). Cross influences of ozone and sulfate precursor emissions changes on air quality and climate.
      PNAS 103: 4377-4380. http://dx.doi.orq/10.1073/pnas.05Q8769103.
Unger, N; Shindell,  PT; Koch, PM; Streets, PG. (2008). Air pollution radiative forcing from specific emissions
      sectors at 2030. J Geophys Res 113: P02306. http://dx.doi.org/10.1029/2007JP008683.
Unger, N; Bond, TC; Wang, JS:  Koch, PM; Menon, S; Shindell, PT; Bauer, S. (2010). Attribution of climate
      forcing to economic sectors. PNAS 107: 3382-3387. http://dx.doi.org/10.1073/pnas.0906548107.
Univ of Leeds. NCAS.. (University of Leeds, National Centre for Atmospheric Science). (2010). The master
      chemical mechanism, from http://mcm.leeds.ac.uk/MCM/home.htt
University of Illinois. (2010). SoyFACE, from http://soyface.illinois.edu/
Unsworth. MH:  Heagle. AS: Heck. WW. (1984a). Gas exchange in open-top  field chambers: I.  Measurement
      and analysis of atmospheric resistances to gas exchange. Atmos Environ 18: 373-380.
      http://dx.doi.org/10.1016/0004-6981(84)90111-2.
Unsworth, MH;  Heagle, AS; Heck, WW. (1984b). Gas exchange in open-top  field chambers: II. Resistances to
      ozone uptake by soybeans. Atmos Environ 18: 381-385. http://dx.doi.org/10.1016/0004-6981 (84)90112-
      4.
Uppu. RM: Cueto. R: Sguadrito. GL:  Prvor. WA. (1995). What does ozone react with at the air/lung interface?
      Model studies using human red blood cell membranes. Arch Biochem Biophys 319: 257-266.
Urbach, F (1997). Ultraviolet radiation and skin cancer of humans. J Photochem Photobiol B 40: 3-7.
Urch.  B: Silverman. F: Corey. P: Brook. JR: Lukic. KZ: Raiagopalan. S: Brook. RP. (2005). Acute blood pressure
      responses in healthy adults during controlled  air pollution exposures.  Environ Health Perspect 113: 1052-
      1055.
Praft - Po Not Cite or Quote                       11-85                                 September 2011

-------
Urch, B: Speck, M: Corey, P: Wasserstein, D: Manno, M: Lukic, KZ: Brook, JR: Liu, L: Coull, B: Schwartz, J:
      Gold, PR: Silverman, F. (2010). Concentrated ambient fine particles and not ozone induce a systemic
      interleukin-6 response in humans.  Inhal Toxicol 22: 210-218.
      http://dx.doi. org/10.3109/08958370903173666.
USDA. (U.S. Department of Agriculture). (2011). Ozone biomonitoring program, from
      http://www.nrs.fs.fed.us/fia/topics/ozone/
Utembe, SR: Hansford, GM: Sanderson, MG: Freshwater, RA: Pratt, KFE: Williams, DE: Cox, RA: Jones, RL.
      (2006). An ozone monitoring instrument based on the tungsten trioxide (WO3) semiconductor. Sens
      Actuators B 114: 507-512. http://dx.doi.Org/10.1016/i.snb.2005.04.049.
Vagaggini, B: Carnevali, S: Macchioni, P:  Taccola, M: Fornai, E: Bacci, E:  Bartoli, ML: Cianchetti,  S: Dente, Fl_:
      Pi Franco, A: Giannini, D: PaggiaroPL. (1999). Airway inflammatory response to ozone in subjects with
      different asthma severity. Eur Respir J 13: 274-280.
Vagaggini, B: Taccola, M: Conti, I: Carnevali, S: Cianchetti, S: Bartoli, ML: Bacci, E:  Dente, Fl_: Pi Franco, A:
      Giannini. D: Paggiaro. PL. (2001).  Budesonide reduces neutrophilic but not functional airway response to
      ozone in mild asthmatics. Am J Respir Grit Care Med  164: 2172-2176.
Vagaggini, B: Taccola, M: Cianchetti,  S: Carnevali, S: Bartoli, ML: Bacci, E: Dente, Fl_: Pi Franco, A: Giannini, D:
      Paggiaro. PL. (2002). Ozone exposure increases eosinophilic airway response induced by previous
      allergen challenge. Am J Respir Grit Care Med 166: 1073-1077.
Vagaggini, B: Cianchetti, S: Bartoli, M: Ricci, M: Bacci, E: Dente,  FL: Pi Franco, A: Paggiaro, P. (2007).
      Prednisone blunts airway neutrophilic inflammatory response due to ozone exposure in asthmatic
      subjects. Respiration 74: 61-58. http://dx.doi.org/10.1159/000096078.
Vagaggini, B: Bartoli, MLE: Cianchetti, S:  Costa, F: Bacci, E: Dente, FL: Pi Franco, A: Malagrino,  L: Paggiaro, P.
      (2010). Increase in markers of airway inflammation after ozone exposure can be observed also in stable
      treated asthmatics with minimal functional response to ozone. Respir Res 11:5.
      http://dx.doi.org/10.1186/1465-9921-11-5.
Vahisalu. T: Kollist. H: Wang. Y. -F: Nishimura. N: Chan. W. -Y: Valerio. G: Lamminmaki. A: Brosche.  M:  Moldau.
      H: Desikan. R: Schroeder. Jl: Kangasiarvi. J. (2008). SLAC1 is required for plant guard cell S-type anion
      channel function in stomatal signalling.  Nature 452: 487-491. http://dx.doi.org/10.1038/nature06608.
Valacchi, G: Vasu, VT: Yokohama, W: Corbacho, AM: Phung, A: Lim, Y: Aung, HH:  Cross, CE: Davis, PA.
      (2007). Lung vitamin E transport processes are affected by both age and environmental oxidants in mice.
      Toxicol Appl Pharmacol 222: 227-234. http://dx.doi.org/10.1016/i.taap.2007.04.010.
Valacchi. G: Pecorelli. A: Mencarelli. M: Maioli. E: Davis.  PA. (2009). Beta-carotene prevents ozone-induced
      proinflammatory markers in  murine skin. Toxicol Ind Health 25: 241-247.
      http://dx.doi. org/10.1177/0748233709103030.
Valkama. E: Koricheva. J: Oksanen. E. (2007). Effects of elevated O3, alone and in combination with elevated
      CO2, on tree leaf chemistry and insect herbivore performance: A meta-analysis. Global Change Biol 13:
      184-201. http://dx.doi.0rg/10.1111/i.1365-2486.01284.x.
Van Bree, L: Dormans, JAM, A: Koren, HS: Devlin, RB: Rombout, PJA. (2002). Attenuation and recovery of
      pulmonary injury in rats following short-term, repeated daily exposure to ozone. Inhal Toxicol 14: 883-
      900.
Van Loveren. H: Krainc. El: Rombout. PJ: Blommaert. FA: Vos. JG. (1990). Effects of ozone,
      hexachlorobenzene, and bis(tri-n-butyltin)oxide on natural killer activity in the rat lung. Toxicol Appl
      Pharmacol 102: 21-33.
Van Aardenne. JA: Dentener. FJ: Olivier. JGJ: Klein Goldewiik. CGM:  Lelieveld. J. (2001). A 1 * 1 resolution data
      set of historical anthropogenic trace gas emissions for the period 1890-1990. Global Biogeochem Cycles
      15: 909-928. http://dx.doi.org/10.1029/2000GB001265.
van Buuren, ML: Guidi,  L: Fornale, S: Ghetti, F: Franceschetti, M: Soldatini, GF: Bagni, N. (2002). Ozone-
      response mechanisms in tobacco:  Implications of polyamine metabolism. New Phytol 156: 389-398.
      http://dx.doi. org/10.1046/1.1469-8137.2002.00539.x.
van der Werf. GR: Randerson. JT: Giglio. L: Collate.  GJ: Kasibhatla. PS: Arellano. AF. Jr. (2006). Interannual
      variability in global biomass burning emissions from 1997 to 2004. Atmos Chem Phys 6: 3423-3441.
Van Dingenen, R:  Dentener, FJ: Raes, F:  Krol, MC: Emberson, L: Cofala, J. (2009).  The global impact of ozone
      on agricultural crop yields under current and future air quality legislation. Atmos Environ 43: 604-618.
      http://dx.doi. org/10.1016/i.atmosenv.20Q8.10.033.
Van Loveren, H: Rombout, PJA: Wagenaar, SS: Walvoort, HC: Vos, JG. (1988). Effects of ozone on the  defense
      to a respiratory Listeria monocytogenes infection in the rat: Suppression of macrophage function  and
      cellular immunity and aggravation of histopathology in lung and liver during infection. Toxicol Appl
      Pharmacol 94: 374-393.
Van Winkle. LS: Baker.  GL: Chan. JK: Schelegle. ES: Plopper. CG. (2010). Airway mast cells in a rhesus model
      of childhood allergic airways disease. Toxicol Sci 116: 313-322.  http://dx.doi.org/10.1093/toxsci/kfg119.
Vancza, EM: Galdanes, K: Gunnison, A: Hatch, G: Gordon, T. (2009). Age, strain, and gender as factors for
      increased sensitivity of the mouse  lung to inhaled ozone. Toxicol Sci 107: 535-543.
      http://dx.doi. org/10.1093/toxsci/kfn253.
Draft - Do Not Cite or Quote                        11-86                                  September 2011

-------
Vandermeiren, K: Black, C: Pleijel, H: deTemmerman, L. (2005). Impact of rising tropospheric ozone on potato:
      Effects on photosynthesis, growth, productivity and yield quality. Plant Cell Environ 28: 982-996.
      http://dx.doi.0rg/10.1111/i.1365-3040.2005.01316.x.
Vanguilder, HP: Freeman, WM. (2011). The hippocampal neuroproteome with aging and cognitive decline: Past
      progress and future directions. Front Aging Neurosci 3: 8.  http://dx.doi.ora/10.3389/fnaai.2011.00008.
Vardoulakis. S: Lumbreras. J: Solazzo. E. (2009). Comparative evaluation of nitrogen oxides and ozone passive
      diffusion tubes for exposure studies. Atmos Environ 43:  2509-2517.
      http://dx.doi.ora/10.1016/i.atmosenv.2009.02.048.
Vasu. VT:  Oommen. S: Lim. Y: Valacchi. G: Hobson. B:  Eiserich. JP: Leonard. SW: Traber. MG: Cross. CE:
      Gohil, K. (2010). Modulation of ozone-sensitive genes in alpha-tocopherol transfer protein null mice.
      Inhal Toxicol 22: 1-16. http://dx.doi.orq/10.3109/08958370902838145.
Velikova, V: Pinelli, P: Pasqualini, S: Reale, L: Ferranti,  F: Loreto, F. (2005). Isoprene decreases the
      concentration of nitric oxide in leaves exposed to elevated ozone. New Phytol 166: 419-426.
      http://dx.doi. orq/10.1111/i.1469-8137.2005.01409.x.
Veninga. TS. (1967). Toxicity of ozone in comparison with ionizing radiation. Strahlentherapie 134: 469-477.
Verhein, KG: Hazari, MS: Moulton, BC: Jacoby, IW:  Jacoby, DB: Fryer, AD. (2011). Three days after a single
      exposure to ozone the mechanism of airway hyperreactivity is dependent upon substance P and nerve
      growth factor. Am J Physiol Lung Cell Mol Physiol 300: L176-L184.
      http://dx.doi.org/10.1152/aiplung.00060.2010.
Vesely, PL: Giordano, AT: Raska-Emery, P:  Montgomery, MR. (1994a). Increase in atrial natriuretic factor in the
      lungs, heart, and circulatory system owing to ozone. Chest 105: 1551-1554.
Vesely, PL: Giordano, AT: Raska-Emery, P:  Montgomery, MR. (1994b). Ozone increases amino- and carboxy-
      terminal atrial natriuretic factor prohormone peptides in lung, heart, and circulation. J Biochem Mol
      Toxicol 9: 107-112.
Vesely, PL: Giordano, AT: Raska-Emery, P:  Montgomery, MR. (1994c). Ozone increases atrial natriuretic
      peptides in heart, lung and circulation of aged vs adult animals. Gerontology 40:  227-236.
      http://dx.doi.org/10.1159/000213590.
Vesely, KR: Schelegle, ES: Stovall, MY: Harkema, JR: Green, JF: Hyde, DM. (1999). Breathing pattern
      response and epithelial labeling in ozone-induced airway injury in neutrophil-depleted rats. Am J Respir
      Cell Mol Biol 20: 699-709.
Viallon, J:  Moussay,  P: Morris, JE: Guenther, FR: Wielgosz, Rl. (2006). A study of systematic biases and
      measurement uncertainties in ozone mole fraction measurements with the NIST  Standard Reference
      Photometer. Metrologia 43:  441-450. http://dx.doi.Org/10.1088/0026-1394/43/5/016.
Vickers, CE: Possell, M: Cojocariu, Cl: Velikova, VB: Laothawornkitkul, J: Ryan, A: Mullineaux,  PM: Hewitt, CN.
      (2009). Isoprene synthesis protects transgenic tobacco  plants from oxidative stress. Plant Cell Environ
      32: 520-531. http://dx.doi.Org/10.1111/i.1365-3040.2009.01946.x.
Vigue, LM: Lindroth, RL. (2010). Effects of genotype, elevated  CO2 and elevated  O3 on aspen phytochemistry
      and aspen leaf beetle Chrysomela crotchi performance. Agr Forest Entomol 12: 267-276.
      http://dx.doi.0rg/10.1111/i.1461-9563.2010.00475.x.
Villeneuve, PJ: Chen, L: Stieb, D: Rowe, BH. (2006a). Associations between outdoor air pollution and
      emergency department visits for stroke in Edmonton, Canada.  Eur J Epidemiol21: 689-700.
Villeneuve. PJ: Doiron. M. -S: Stieb. D: Dales. R: Burnett. RT: Dugandzic. R. (2006b). Is outdoor air pollution
      associated with physician visits for allergic rhinitis among the elderly in Toronto, Canada? Allergy 61:
      750-758. http://dx.doi.0rg/10.1111/i.1398-9995.2006.01070.x.
Villeneuve. PJ: Chen. L: Rowe. BH: Coates. F. (2007). Outdoor air pollution and emergency department visits
      for  asthma among children and adults: A case-crossover study in northern Alberta, Canada. Environ
      Health Global Access Sci Source 6: 40. http://dx.doi.org/10.1186/1476-069X-6-40.
Vincent, R: Vu, D: Hatch, G: Poon, R: Dreher, K: Guenette, J: Bjarnason, S: Potvin, M: Norwood, J: McMullen,
      E_. (1996a). Sensitivity of lungs of aging Fischer 344 rats to ozone: Assessment by bronchoalveolar
      lavage. Am J  Physiol 271: L555-L565.
Vincent. R: Janzen. EG: Chen. G: Kumarathasan. P: Haire. PL: Guenette. J: Chen. JZ:  Bray. TM. (1996b). Spin
      trapping study in the lungs and liver of F344 rats  after exposure to ozone. Free Radic Res 25: 475-488.
Vishvakarman. D: Wong. JCF: Boreham. BW. (2001). Annual occupational exposure to  ultraviolet radiation in
      central Queensland. Health Phys 81: 536-544.
Vivier, E: Raulet, DH: Moretta, A: Caligiuri, MA: Zitvogel, L: Lanier, LL: Yokoyama, WM:  Ugolini, S. (2011).
      Innate or adaptive immunity? The example of natural killer cells. Science 331: 44-49.
      http://dx.doi.org/10.1126/science.1198687.
Volk, M: Geissmann, M: Blatter, A: Contat, F: Fuhrer, J. (2003). Design and performance of a free-air exposure
      system to study long-term effects of ozone on grasslands. Atmos Environ 37: 1341-1350.
Volk. M: Bungener. P: Contat. F: Montani. M: Fuhrer. J. (2006). Grassland yield declined by a quarter in 5 years
      of free-air ozone fumigation. Global Change Biol  12: 74-83. http://dx.doi.org/10.1111/1.1365-
      2486.2005.01083.x.
Draft - Do Not Cite or Quote                        11-87                                  September 2011

-------
Volk, M: Obrist, D: Novak, K: Giger, R: Bassin, S: Fuhrer, J. (2011). Subalpine grassland carbon dioxide fluxes
      indicate substantial carbon losses under increased nitrogen deposition, but not at elevated ozone
      concentration. Global Change Biol 17: 366-376.  http://dx.doi.Org/10.1111/i.1365-2486.2010.02228.x.
Vollenweider, P: Woodcock, H: Kelty, MJ: Hofer,  R, -M.  (2003). Reduction of stem growth and site dependency
      of leaf injury in Massachusetts black cherries exhibiting ozone symptoms. Environ Pollut 125: 467-480.
Vollsnes. AV: Kruse. OMO: Eriksen. AB: Oxaal. U: Futsaether. CM. (2010). In vivo root growth dynamics of
      ozone exposed Trifolium subterraneum. Environ Exp Bot 69: 183-188.
      http://dx.doi.0rg/10.1016/i.envexpbot.2010.03.007.
Volz. A:  Kiev. D. (1988). Evaluation of the Montsouris series of ozone measurements made in the nineteenth
      century. Nature 332: 240-242. http://dx.doi.org/10.1038/332240aO.
Von Klot, S:  Peters, A: Aalto, P: Bellander, T: Berglind, N: D'lppoliti, D: Elosua, R: Hormann, A: Kulmala, M:
      Lanki, T: Lowel, H: Pekkanen, J: Picciotto, S: Sunyer, J: Forastiere, F: Group, HEoPoSSS. (2005).
      Ambient air pollution is associated with increased risk of hospital cardiac readmissions of myocardial
      infarction survivors in five European cities. Circulation 112: 3073-3079.
      http://dx.doi.org/10.1161/CIRCULATIONAHA.105.548743.
Voynow, JA: Fischer, BM: Zheng, S: Potts, EN: Grover,  AR: Jaiswal, AK: Ghio, AJ: Foster, WM. (2009).
      NAD(P)H quinone oxidoreductase 1 is essential  for  ozone-induced oxidative stress in mice and humans.
      Am J Respir Cell Mol Biol 41: 107-113. http://dx.doi.org/10.1165/rcmb.2008-0381OC.
Vrijheid, M: Martinez, D: Manzanares, S: Dadvand, P: Schembari, A: Rankin, J:  Nieuwenhuijsen, M. (2011).
      Ambient air pollution and risk of congenital anomalies: A systematic review and  meta-analysis. Environ
      Health Perspect 119: 598-606. http://dx.doi.org/10.1289/ehp.1002946.
Vuorinen, T: Nerg, A, -M: Holopainen, JK. (2004). Ozone exposure triggers the emission of herbivore-induced
      plant volatiles, but does not disturb tritrophic signalling. Environ Pollut 131: 305-311.
      http://dx.doi.0rg/10.1016/i.envpol.2004.02.027.
Wagner, JG: Hotchkiss, JA: Harkema, JR. (2002). Enhancement of nasal inflammatory and epithelial responses
      after ozone and allergen coexposure in brown Norway rats. Toxicol Sci 67: 284-294.
Wagner. JG: Jiang. Q: Harkema. JR:  Illek. B: Patel. DP: Ames. BN: Peden. DB.  (2007). Ozone enhancement of
      lower airway allergic inflammation is prevented by gamma-tocopherol. Free Radic Biol Med 43:  11 ye-
      ll 88. http://dx.doi.0rg/10.1016/i.freeradbiomed.2007.07.013.
Wagner, JG: Harkema, JR: Jiang, Q:  Illek, B: Ames, BN: Peden,  DB. (2009). Gamma-tocopherol attenuates
      ozone-induced exacerbation of allergic rhinosinusitis in rats. Toxicol Pathol 37: 481-491.
      http://dx.doi. org/10.1177/0192623309335630.
Wahl. M. (2008). Ecological modulation of environmental stress: Interactions between ultraviolet radiation,
      epibiotic snail embryos, plants and herbivores. J Anim Ecol 77: 549-557.  http://dx.doi.org/10.1111 /j. 1365-
      2656.2007.01352.x.
Wallace. L: Williams. R: Suggs. J: Jones. P. (2006). Estimating contributions of outdoor fine particles to indoor
      concentrations and personal exposures: Effects  of household characteristics and personal activities.
      (EPA/600/R-06/023). Research Triangle Park, NC: U.S. Environmental Protection Agency.
Wallin, G: Skarbv,  L. (1992). The influence of ozone on  the stomatal and non-stomatal  limitation of
      photosynthesis in Norway spruce, Picea abies (L.) Karst, exposed to soil moisture deficit. Trees Struct
      Funct 6: 128-136. http://dx.doi.org/10.1007/BF00202428.
Wang. DJ: Zhou. WD: Dai. XJ: Yan. Y. (2007). Study on effect and mechanism of sodium ferulate in preventing
      and treating ozone induced lung injury in mice. Chin J Integr Med 13: 211-214.
      http://dx.doi.org/10.1007/s11655-007-0211-9.
Wang. HQ: Jacob. DJ: Le Sager. P: Streets. DG: Park.  RJ: Gilliland. AB: van Donkelaar. A.  (2009a). Surface
      ozone background in the United States: Canadian and Mexican pollution influences. Atmos Environ 43:
      1310-1319. http://dx.doi.0rg/10.1016/i.atmosenv.2008.11.036.
Wang, J: Christopher, SA: Nair, US:  Reid, JS: Prins, EM: Szykman, J: Hand, JL. (2006). Mesoscale modeling of
      Central American smoke transport to the United  States: 1. "Top-down" assessment of emission strength
      and diurnal variation impacts. J Geophys  Res 111: D05S17. http://dx.doi.org/10.1029/2005JD006416.
Wang. L: He. X: Chen. W. (2009b). Effects of elevated ozone on photosynthetic CO2 exchange and chlorophyll
      a fluorescence in leaves of Quercus mongolica grown in urban area. Bull Environ Contam Toxicol 82:
      478-481. http://dx.doi.org/10.1007/s00128-008-9606-3.
Wang. T. -N: Ko. Y. -C: Chao. Y. -Y: Huang. C. -C: Lin. R. -S. (1999). Association between indoor and outdoor
      air pollution and adolescent asthma from  1995 to 1996 in Taiwan. Environ Res 81: 239-247.
Wang, W, -C: Pinto,  JP: Yung, YL. (1980). Climatic effects due to halogenated compounds  in the earth's
      atmosphere. J Atmos Sci 37:  333-338. http://dx.doi.org/10.1175/1520-
      0469(1980)037<0333:CEDTHC>2.0.CO:2.
Wang. X: Mauzerall. PL.  (2004). Characterizing distributions of surface ozone and its impact on grain
      production in  China, Japan and South Korea: 1990 and 2020. Atmos Environ 38: 4383-4402.
      http://dx.doi.0rg/10.1016/i.atmosenv.2004.03.067.
Wang. X: Zheng. Q:  Feng. Z: Xie. J: Ouvang. Z:  Manning. WJ. (2008). Comparison of a diurnal vs steady-state
      ozone exposure profile on growth and yield of oilseed rape (Brassica napus L.) in open-top chambers  in
      the Yangtze Delta, China. Environ Pollut 156: 449-453. http://dx.doi.Org/10.1016/i.envpol.2008.01.027.
Draft - Do Not Cite or Quote                        11-88                                  September 2011

-------
Wang, X: Taub, DR. (2010). Interactive effects of elevated carbon dioxide and environmental stresses on root
      mass fraction in plants: A meta-analytical synthesis using pairwise techniques. Oecologia 163: 1-11.
      http://dx.doi. org/10.1007/s00442-010-1572-x.
Wang, XY: Hu, W: long, S. (2009c). Long-term exposure to gaseous air pollutants and cardio-respiratory
      mortality in Brisbane, Australia. Geospat Health 3: 257-263.
Ward. DJ: Roberts. KT: Jones. N:  Harrison. RM: Avres. JG:  Hussain. S: Walters. S. (2002). Effects of daily
      variation in outdoor particulates and ambient acid species in normal and asthmatic children. Thorax 57:
      489-502. http://dx.doi.0rg/10.1136/thorax.57.6.489.
Watanabe. M: Yamaguchi. M: Tabe. C: Iwasaki. M: Yamashita. R: Funada. R: Fukami. M: Matsumura. H: Kohno.
      Y: Izuta, T. (2007). Influences of nitrogen load on the growth and photosynthetic responses of Quercus
      serrata seedlings to O3. Trees Struct Funct 21:  421-432. http://dx.doi.org/10.1007/s00468-007-0134-2.
Watkinson, WP: Aileru, AA: Dowd, SM: Doerfler, PL: Tepper, JS: Costa, PL. (1993). Acute effects of ozone on
      heart  rate and body temperature in the unanesthetized, unrestrained rat maintained at different ambient
      temperatures. Inhal Toxicol 5: 129-147.
Watkinson. WP: Campen. MJ: Nolan. JP:  Costa.  PL. (2001). Cardiovascular and systemic responses to inhaled
      pollutants in rodents: Effects of ozone and  particulate matter. Environ Health Perspect 109: 539-546.
Watkinson. WP: Campen. MJ: Wichers. LB: Nolan. JP:  Costa. PL. (2003). Cardiac and thermoregulatory
      responses to inhaled pollutants in healthy and compromised rodents: Modulation via interaction with
      environmental factors. Environ Res 92: 35-47.
Wattiez, R: Noel-Georis, I: Cruvt, C: Broeckaert, F: Bernard, A: Falmagne, P. (2003). Susceptibility to oxidative
      stress: proteomic analysis of bronchoalveolar lavage from ozone-sensitive and ozone-resistant strains of
      mice.  Proteomics 3: 658-665. http://dx.doi.org/10.1002/pmic.20030Q417.
Webster. M:  Nam. J: Kimura. Y: Jeffries. H: Vizuete. W: Allen. PT (2007). The effect of variability in industrial
      emissions on ozone formation in Houston,  Texas. Atmos Environ 41: 9580-9593.
      http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.052.
Weibel.  ER.  (1980). Pesign and structure  of the human lung. In AP Fishman (Ed.), Assessment of pulmonary
      function. New York, NY: McGraw-Hill.
Weinmann, GG: Weidenbach-Gerbase, M: Foster, WM: Zacur, H: Frank, R. (1995a). Evidence for ozone-
      induced small-airway dysfunction: Lack of menstrual-cycle and gender effects. Am J Respir Crit Care
      Med 152: 988-996.
Weinmann, GG: Liu, MC: Proud, P: Weidenbach-Gerbase, M: Hubbard,  W: Frank, R. (1995b). Ozone exposure
      in humans: Inflammatory, small and peripheral airway responses. Am J Respir Crit Care Med  152: 1175-
      1182.
Weinmann, GG: Bowes, SM: Gerbase, MW: Kimball, AW: Frank, R.  (1995c). Response to acute ozone
      exposure in healthy men. Results of a screening procedure. Am J Respir Crit Care Med 151: 33-40.
Weinstein. PA: Laurence. JA: Retzlaff. WA: Kern.  JS: Lee. EH: Hogsett.  WE: Weber. J. (2005). Predicting the
      effects of tropospheric ozone on regional productivity of ponderosa pine and white fir.  For Ecol Manage
      205: 73-89. http://dx.doi.org/10.1016/i.foreco.20Q4.10.007.
Welch, RW:  Wang, Y: Grossman, A, Jr: Park, JB:  Kirk, KL: Levine, M. (1995). Accumulation of vitamin C
      (ascorbate) and its oxidized metabolite dehydroascorbic acid occurs by separate mechanisms. J  Biol
      Chem 270: 12584-12592. http://dx.doi.org/10.1074/ibc.270.21.12584.
Wellenius. GA: Bateson. TF: Mittleman. MA: Schwartz. J. (2005). Particulate air pollution and the rate of
      hospitalization for congestive heart failure among medicare beneficiaries in Pittsburgh, Pennsylvania. Am
      J Epidemiol 161: 1030-1036.
Wellenius. GA: Yeh. GY: Coull. BA: Suh. HH: Phillips. RS: Mittleman. MA. (2007). Effects of ambient air
      pollution on functional status in patients with chronic congestive heart failure: A repeated-measures
      study. Environ Health 6: 1-7.
Wen, XJ: Balluz, L: Mokdad, A. (2009). Association between media alerts of air quality index and change of
      outdoor activity among adult asthma in six  states, BRFSS, 2005. J Community Health 34: 40-46.
      http://dx.doi. org/10.1007/s10900-008-9126-4.
Wenten. M: Gauderman. WJ:  Berhane. K: Lin. PC: Peters. J: Gilliland. FP. (2009). Functional variants in the
      catalase and myeloperoxidase genes, ambient air pollution, and respiratory-related school absences: An
      example of epistasis in gene-environment interactions. Am J Epidemiol 170: 1494-1501.
      http://dx.doi.org/10.1093/aie/kwp310.
Wentworth, P, Jr: Nieva, J: Takeuchi, C: Galve, R: Wentworth, AP: Pillev, RB: PeLaria, GA: Saven, A: Babior,
      BM: Janda, KP: Eschenmoser, A: Lerner, RA. (2003). Evidence for ozone formation in human
      atherosclerotic arteries. Science 302: 1053-1056.
Werner, H: Fabian, P. (2002). Free-air fumigation of mature trees: A novel system for controlled ozone
      enrichment in grown-up beech and spruce  canopies.  Environ Sci  Pollut Res Int 9: 117-121.
Wernli. H: Bourgui. M. (2002). A Lagrangian "1-year climatology" of (deep) cross-tropopause exchange in the
      extratropical Northern Hemisphere. J Geophys Res 107: 4021. http://dx.doi.org/10.1029/2001JP000812.
Weschler. CJ: Shields. HC. (1997). Potential reactions among indoor pollutants. Atmos Environ 31: 3487-3495.
Weselv. ML:  Hicks. BB. (2000). A review of the current status of knowledge  on dry deposition [Review]. Atmos
      Environ 34: 2261-2282. http://dx.doi.org/10.1016/S1352-2310(99)00467-7.
Praft - Po Not Cite or Quote                        11-89                                  September 2011

-------
West, JJ: Fiore, AM. (2005). Management of tropospheric ozone by reducing methane emissions. Environ Sci
      Technol39: 4685-4691.
West, JJ: Fiore, AM: Horowitz, LW: Mauzerall, PL. (2006). Global health benefits of mitigating ozone pollution
      with methane emission controls. PNAS 103: 3988-3993. http://dx.doi.org/10.1073/pnas.0600201103.
West. JJ: Fiore. AM: Naik. V:  Horowitz. LW: Schwarzkopf. MD: Mauzerall. PL. (2007). Ozone air quality and
      radiative forcing consequences of changes in ozone precursor emissions. Geophys Res Lett 34: L06806.
      http://dx.doi. org/10.1029/2006GL029173.
Wheeler. A: Zanobetti. A: Gold. PR: Schwartz. J: Stone. P: Suh.  HH. (2006). The relationship between ambient
      air pollution and heart  rate variability differs for individuals with heart and pulmonary disease. Environ
      Health Perspect 114: 560-566.
White, IR:  Martin, P: Munoz, MP: Petersson, FK: Henshaw, SJ: Nickless, G: Lloyd-Jones, GC: Clemitshaw,  KG:
      Shallcross, PE. (2010). Use of reactive tracers to determine ambient OH radical concentrations:
      Application within the indoor environment. Environ  Sci Technol 44: 6269-6274.
      http://dx.doi.org/10.1021/es901699a.
Whitfield. CP: Pavison. AW: Ashenden. TW. (1996). Interactive effects of ozone and soil volume on Plantago
      major. New Phytol 134: 287-294. http://dx.doi.Org/10.1111/i.1469-8137.1996.tb04633.x.
Whitfield. CP: Pavison. AW: Ashenden. TW. (1997). Artificial selection and heritability of ozone resistance in two
      populations of Plantago major. New Phytol 137: 645-655.
Wieser,  G: Havranek, WM. (1995). Environmental control  of ozone uptake in Larix decidua Mill: A comparison
      between different altitudes. Tree Physiol  15: 253-258.
Wieser,  G: Manning, WJ: Tausz, M: Bytnerowicz, A. (2006). Evidence for potential impacts of ozone on Pinus
      cembra L. at mountain sites in Europe: An overview. Environ Pollut 139: 53-58.
      http://dx.doi.0rg/10.1016/i.envpol.2005.04.037.
Wiester. MJ: Williams. TB: King. ME: Menache.  MG: Miller. FJ. (1987). Ozone uptake in awake Sprague-Pawley
      rats. Toxicol Appl Pharmacol 89: 429-437. http://dx.doi.org/10.1016/0041-008X(87)90162-1.
Wiester. MJ: Tepper. JS: King. ME: Menache. MG:  Costa. PL. (1988). Comparative study of ozone (O3)  uptake
      in three strains of rats and in the guinea pig. Toxicol Appl Pharmacol 96: 140-146.
Wiester, MJ: Tepper, JS: Winsett, PW: Crissman, KM: Richards,  JH: Costa, PL. (1996a). Adaptation to ozone in
      rats and its association with ascorbic acid in the lung. Toxicol Sci 31: 56-64.
Wiester, MJ: Watkinson, WP: Costa, PL:  Crissman, KM: Richards,  JH: Winsett, PW:  Highfill, JW. (1996b).
      Ozone toxicity in the rat. III. Effect of changes in ambient temperature on pulmonary parameters.  J Appl
      Physiol 81: 1691-1700.
Wiester. MJ: Stevens. MA:  Menache. MG: McKee.  JL. Jr:  Gerritv. TR. (1996c). Ozone uptake in healthy  adult
      males during quiet breathing. Toxicol Sci 29: 102-109.
Wild. O: Prather.  MJ: Akimoto. H. (2001). Indirect long-term global radiative cooling from NOX emissions.
      Geophys Res Lett 28:  1719-1722. http://dx.doi.org/10.1029/2000GL012573.
Wilhelm, M: Ritz, B. (2005). Local variations in CO  and particulate air pollution and adverse birth outcomes in
      Los Angeles County, California, USA. Environ Health Perspect 113: 1212-1221.
Wilkinson, S: Pavies, WJ. (2010). Prought, ozone,  ABA and ethylene: New insights from cell to plant to
      community. Plant Cell  Environ 33: 510-525.  http://dx.doi.Org/10.1111/i.1365-3040.2009.02052.x.
Will. RE: Ceulemans. R. (1997). Effects of elevated CO2 concentration on photosynthesis, respiration and
      carbohydrate status of coppice Populus hybrids.  Physiol Plant 100: 933-939.
      http://dx.doi.0rg/10.1111/i.1399-3054.1997.tb00020.x.
Williams. AS: Issa. R: Leung. SY:  Puneeta. N: Gregory.  P: Ferguson. P: Brvdon. L: Bennett. I: Adcock. M:
      Chung. KF. (2007a). Attenuation of ozone-induced  airway inflammation and  hyper-responsiveness by c-
      Jun NH2 terminal kinase inhibitor SP600125. J Pharmacol Exp Ther 322: 351-359.
      http://dx.doi.org/10.1124/ipet.107.121624.
Williams, AS: Leung, SY: Nath, P: Khorasani, NM:  Bhavsar, P: Issa, R:  Mitchell, JA: Adcock, IM: Chung,  KF.
      (2007b). Role of TLR2, TLR4, and MyP88 in murine ozone-induced airway hyperresponsiveness  and
      neutrophilia. J Appl Physiol 103: 1189-1195. http://dx.doi.org/10.1152/iapplphvsiol.00172.2007.
Williams. AS: Nath. P: Leung. SY: Khorasani. N: McKenzie. ANJ: Adcock. IM: Chung. KF. (2008a). Modulation
      of ozone-induced airway hyperresponsiveness and inflammation by interleukin-13. Eur Respir J 32: 571-
      578. http://dx.doi.org/10.1183/09031936.00121607.
Williams. AS: Issa. R: Purham. A:  Leung. SY: Kapoun. A:  Medicherla. S: Higgins. LS: Adcock. IM: Chung. KF.
      (2008b). Role of p38 mitogen-activated protein kinase in ozone-induced airway hyperresponsiveness
      and inflammation. Eur J Pharmacol 600:  117-122. http://dx.doi.Org/10.1016/i.eiphar.2008.09.031.
Williams, AS: Eynott, PR: Leung, SY: Nath, P: Jupp, R:  Pe Sanctis, GT: Resnick, R: Adcock, IM: Chung, KF.
      (2009a). Role of cathepsin S in ozone-induced airway  hyperresponsiveness and inflammation. Pulm
      Pharmacol Ther 22: 27-32. http://dx.doi.Org/10.1016/i.pupt.2008.11.002.
Williams. EJ: Fehsenfeld. FC: Jobson. BT: Kuster. WC:  Goldan.  PP: Stutz. J: McClennv. WA. (2006).
      Comparison of ultraviolet absorbance, chemiluminescence,  and POAS instruments for ambient ozone
      monitoring. Environ Sci Technol 40: 5755-5762.  http://dx.doi.org/10.1021/es0523542.
Praft - Po Not Cite or Quote                        11-90                                  September 2011

-------
Williams, R: Rea, A: Vette, A: Croghan, C: Whitaker, D: Stevens, C: McDow, S: Fortmann, R: Sheldon, L:
      Wilson, H: Thornburg, J: Phillips, M: Lawless, P: Rodes, C: Daughtrey,  H. (2009b). The design and field
      implementation of the Detroit exposure and aerosol research study. J Expo Sci Environ Epidemiol 19:
      643-659. http://dx.doi.0rg/10.1038/jes.2008.61.
Wilson. KL: Birks. JW. (2006). Mechanism and elimination of a water vapor interference in the measurement of
      ozone by UV absorbance. Environ Sci Technol 40: 6361-6367. http://dx.doi.orq/10.1021/es052590c.
Wilson, WE: Suh, HH. (1997). Fine particles and coarse  particles: Concentration relationships relevant to
      epidemiologic studies. J Air Waste Manag Assoc 47: 1238-1249.
Wilson. WE: Mage. DT:  Grant. LD.  (2000). Estimating separately personal exposure to ambient and nonambient
      particulate matter for epidemiology and risk assessment: Why and how. J Air Waste Manag Assoc 50:
      1167-1183.
Winner, WE: Lefohn, AS: Cotter, IS: Greitner, CS: Nellessen, J: McEvov, LR, Jr: Olson, Rl_: Atkinson, CJ:
      Moore, LD. (1989). Plant responses to elevational gradients of O3 exposures in Virginia. PNAS 86: 8828-
      8832.
Wise. EK: Comrie. AC. (2005). Meteorologically adjusted urban air quality trends in the Southwestern United
      States. Atmos  Environ 39: 2969-2980. http://dx.doi.Org/10.1016/i.atmosenv.2005.01.024.
Witschi. H. (1991). Effects of oxygen and ozone on mouse lung tumorigenesis. Exp Lung Res 17: 473-483.
Witschi. H: Wilson.  DW: Plopper. CG. (1993). Modulation of N-nitrosodiethylamine-induced hamster lung tumors
      by ozone. Toxicology 77: 193-202.
Witschi, H: Espiritu, I: Pinkerton, KE: Murphy, K:  Maronpot, RR. (1999). Ozone carcinogenesis revisited. Toxicol
      Sci 52: 162-167.
Wittig, VE: Ainsworth, EA: Long, SP. (2007). To what extent do current and projected increases in surface ozone
      affect photosynthesis and stomatal conductance of trees? A meta-analytic review of the last 3 decades of
      experiments [Review]. Plant Cell Environ 30: 1150-1162. http://dx.doi.orq/10.1111/1.1365-
      3040.2007.01717.x.
Wittig. VE: Ainsworth. EA: Naidu. SL: Karnoskv. DF: Long. SP. (2009). Quantifying the impact of current and
      future tropospheric ozone on tree biomass, growth, physiology and biochemistry: A quantitative meta-
      analysis. Global Change Biol 15: 396-424. http://dx.doi.Org/10.1111/i.1365-2486.2008.01774.x.
Wiwatanadate, P; Trakultivakorn, M. (2010). Air pollution-related peak expiratory flow rates among asthmatic
      children in Chiang Mai, Thailand. Inhal Toxicol 22: 301-308.
      http://dx.doi.org/10.3109/08958370903300327.
Wiwatanadate. P: Liwsrisakun. C. (2011). Acute effects of air pollution on peak expiratory flow rates and
      symptoms among asthmatic patients in Chiang  Mai, Thailand. Int J Hyg Environ Health 214: 251-257.
      http://dx.doi.0rg/10.1016/i.iiheh.2011.03.003.
Wollmann. HA. (1998). Intrauterine growth restriction: Definition and etiology. Horm  Res 49: 1-6.
Wong. C. -M: Ma. S: AJ. H: Lam. T -H. (1999a). Does  ozone have any effect on daily hospital admissions for
      circulatory diseases? J Epidemiol Community Health 53: 580-581.
Wong, C, -M: Ou, C, -Q: Thach, T, -Q: Chau, Y, -K: Chan, K, -P: Ho, S, -Y: Chung, RY:  Lam, T, -H: Hedley, AJ.
      (2007). Does regular exercise protect against air pollution-associated mortality? Prev Med 44: 386-392.
Wong, CM: Ou,  CQ: Chan, KP: Chau, YK: Thach, TQ:  Yang, L: Chung,  RY: Thomas, GN: Peiris, JS: Wong, TW:
      Hedlev. AJ: Lam. TH. (2008). The effects of air pollution on mortality in socially deprived urban  areas in
      Hong Kong,  China. Environ Health Perspect 116:  1189-1194.
Wong, CM: Yang, L: Thach, TQ: Chau, PY: Chan, KP:  Thomas, GN: Lam, TH: Wong, TW: Hedlev, AJ: Peiris,
      JS. (2009). Modification by influenza  on health effects of air pollution in Hong Kong. Environ Health
      Perspect 117: 248-253. http://dx.doi.org/10.1289/ehp.11605.
Wong, CM: Vichit-Vadakan, N: Vajanapoom, N: Ostro,  B: Thach, TQ: Chau, PY: Chan,  EK: Chung, RY: Ou, CQ:
      Yang, L: Peiris, JS: Thomas, GN: Lam, TH: Wong, TW:  Hedlev, AJ:  Kan, H: Chen, B: Zhao, N: London,
      SJ: Song, G: Chen,  G: Zhang, Y: Jiang, L: Qian, Z: He,  Q: Lin, HM: Kong, L:  Zhou, D: Liang, S: Zhu, Z:
      Liao, D: Liu,  W: Bentley, CM: Dan, J:  Wang, B: Yang, N: Xu, S: Gong, J: Wei, H: Sun, H: Qin, Z. (2010).
      Part 5.  Public health and air pollution in Asia (PAPA): A  combined analysis of four studies of air pollution
      and mortality. In Public Health and Air Pollution  in  Asia (PAPA): Coordinated Studies of Short-Term
      Exposure to Air Pollution and Daily Mortality in Four Cities (Vol. 154). Boston, MA: Health Effects
      Institute.  http://pubs.healtheffects.org/view.php?id=348.
Wong. TW: Lau. TS: Yu. TS: Neller. A: Wong. SL: Tarn. W: Pang. SW. (1999b). Air pollution and hospital
      admissions for respiratory and cardiovascular diseases in Hong Kong. Occup Environ Med 56: 679-683.
Woo, SY: Hincklev, TM.  (2005). The effects of ozone on growth and stomatal response in the F-2 generation of
      hybrid poplar (Populus trichocarpa x Populus deltoides). Biol Plantarum 49: 395-404.
      http://dx.doi. orq/10.1007/s10535-005-0014-9.
Wood. AM: Harrison.  RM: Semple.  S: Avres. JG: Stocklev. RA. (2009). Outdoor air pollution is associated with
      disease severity in a1-antitrypsin deficiency. Eur Respir J 34: 346-353.
Woodruff, TJ: Darrow, LA: Parker, JD. (2008). Air pollution and postneonatal infant mortality in the United
      States, 1999-2002. Environ Health Perspect 116:  110-115.
Draft - Do Not Cite or Quote                        11-91                                  September 2011

-------
Woodruff, TJ: Parker, JD: Darrow, LA: Slama, R: Bell, ML: Choi, H: Glinianaia, S: Hoggatt, KJ: Karr, CJ: Lobdell,
      DT: Wilhelm,  M. (2009). Methodological issues in studies of air pollution and reproductive health. Environ
      Res 109: 311-320.
Woodruff, TJ: Parker, JD: Adams, K: Bell, ML: Gehring,  U: Glinianaia, S: Ha, EH: Jalaludin, B: Slama, R. (2010).
      International Collaboration on Air Pollution and Pregnancy Outcomes (ICAPPO). Int J  Environ Res Public
      Health 7: 2638-2652. http://dx.doi.org/10.3390/iierph7062638.
Worden, HM: Logan, JA: Worden, JR: Beer, R: Bowman, K: Clough, SA: Eldering, A: Fisher,  BM: Gunson, MR:
      Herman. RL:  Kulawik. SS: Lampel. MC: Luo. M:  Megretskaia. IA: Osterman. GB: Shephard. MW.
      (2007a). Comparisons of Tropospheric Emission Spectrometer (TES) ozone profiles to ozonesondes:
      Methods and  initial results. J Geophys Res 112:  D03309. http://dx.doi.Org/10.1029/2006JD007258.
Worden, HM: Bowman,  KW: Worden, JR: Eldering, A: Beer, R. (2008). Satellite measurements of the clear-sky
      greenhouse effect from tropospheric ozone. Nat  Geosci 1: 305-308. http://dx.doi.org/10.1038/ngeo182.
Worden, J: Liu, X: Bowman, K: Chance, K:  Beer, R: Eldering, A: Gunson, M: Worden, H. (2007b). Improved
      tropospheric ozone profile retrievals using OMI and TES radiances.  Geophys Res Lett 34:  L01809.
      http://dx.doi. orq/10.1029/2006GL027806.
Wright, GA: Lutmerding, A:  Dudareva, N: Smith, BH. (2005). Intensity and the  ratios of compounds in the scent
      of snapdragon flowers affect scent discrimination by honeybees  (Apis mellifera). J Comp Physiol A
      Neuroethol Sens Neural Behav Physiol 191: 105-114.
Wu, CF: Kuo, 1C: Su, TC: Li, YR: Lin, LY: Chan, CC: Hsu, SC. (2010). Effects of personal exposure to
      particulate matter and ozone on arterial stiffness and heart rate variability in healthy adults. Am J
      Epidemiol 171: 1299-1309. http://dx.doi.org/10.1093/aie/kwq060.
Wu, S: Mickley, LJ: Leibensperger, EM: Jacob, DJ: Rind, D: Streets, DG. (2008a). Effects of 2000-2050 global
      change on ozone air quality in the United States. J Geophys Res 113: D06302.
      http://dx.doi. orq/10.1029/2007JD008917.
Wu, W: Doreswamy, V:  Diaz-Sanchez, D: Samet, JM: Kesic, M: Dailey,  L: Zhang, W: Jaspers, I: Peden, DB.
      (2011). GSTM1 modulation of IL-8 expression in  human bronchial epithelial  cells exposed to ozone. Free
      Radic Biol Med 51: 522-529. http://dx.doi.Org/10.1016/i.freeradbiomed.2011.05.006.
Wu, ZX: Satterfield, BE: Dey,  RD. (2003). Substance P  released from intrinsic  airway neurons contributes to
      ozone-enhanced airway hyperresponsiveness in ferret trachea. J Appl Physiol 95: 742-750.
Wu, ZX: Barker, JS:  Batchelor, TP: Dey, RD. (2008b). Interleukin (IL)-1  regulates ozone-enhanced tracheal
      smooth muscle responsiveness by increasing substance P (SP)  production in intrinsic airway  neurons of
      ferret. Respir  Physiol Neurobiol 164: 300-311. http://dx.doi.Org/10.1016/i.resp.2008.07.019.
Xia. Y: long. H. (2006).  Cumulative effects of air pollution on public health. Stat Med 25: 3548-3559.
      http://dx.doi. org/10.1002/sim.2446.
Xue. J: McCurdv. T:  Spengler. J:  Ozkavnak. H. (2004). Understanding variability in time spent in selected
      locations for 7-12-year old children. J Expo Anal  Environ Epidemiol 14:  222-233.
      http://dx.doi. org/10.1038/sj.iea.750Q319.
Xue, J: Liu, SV: Ozkavnak,  H: Spengler, JD. (2005). Parameter evaluation and model validation of ozone
      exposure assessment using Harvard Southern California Chronic Ozone Exposure Study data. J Air
      Waste Manag Assoc 55: 1508-1515.
Yallop. D: Duncan. ER:  Norris. E: Fuller. GW: Thomas. N: Walters. J: Dick.  MC: Height. SE: Thein. SL: Rees.
      DC. (2007). The associations between air quality and the number of hospital admissions for acute pain
      and sickle-cell disease in an urban environment.  Br J Haematol 136: 844-848.
      http://dx.doi.0rg/10.1111/i.1365-2141.2007.06493.x.
Yamaguchi. M: Watanabe. M: Iwasaki. M: Tabe. C: Matsumura. H:  Kohno. Y: Izuta. T. (2007). Growth and
      photosynthetic responses of Fagus crenata seedlings to O3 under different nitrogen loads. Trees Struct
      Funct 21: 707-718. http://dx.doi.org/10.1007/s00468-007-0163-x.
Yan, K: Chen, W: He, XY: Zhang, GY: Xu, S: Wang, LL. (2010). Responses of photosynthesis, lipid peroxidation
      and antioxidant system in  leaves of Quercus mongolica to elevated O3. Environ Exp Bot 69: 198-204.
      http://dx.doi.0rg/10.1016/i.envexpbot.2010.03.008.
Yang. C. -Y:  Chen. Y. -S: Yang. C. -H: Ho. S. -C. (2004). Relationship between ambient air pollution and hospital
      admissions for cardiovascular diseases in Kaohsiung, Taiwan. J  Toxicol Environ Health A 67: 483-493.
Yang. C. -Y:  Hsieh. H. -J: Tsai. S. -S: Wu. T. -N: Chiu. H. -F. (2006). Correlation between air pollution and
      postneonatal  mortality in a subtropical city: Taipei, Taiwan. J Toxicol  Environ Health A 69: 2033-2040.
      http://dx.doi. org/10.1080/15287390600746181.
Yang, CY. (2008). Air pollution and hospital admissions  for congestive heart failure in a subtropical city: Taipei,
      Taiwan. J Toxicol Environ Health A 71: 1085-1090.
Yang, IA: Holz, O: Jorres, RA: Magnussen, H: Barton, SJ: Rodriguez, S: Cakebread, JA: Hollowav, JW: Holgate,
      ST. (2005a). Association of tumor necrosis factor alpha polymorphisms  and  ozone-induced change in
      lung function. Am J Respir Grit Care Med 171: 171-176.
Yang, Q: Chen, Y: Krewski, D: Burnett, RT: Shi, Y:  McGrail, KM. (2005b). Effect of short-term exposure to low
      levels of gaseous pollutants on chronic obstructive pulmonary disease hospitalizations. Environ Res 99:
      99-105. http://dx.doi.0rg/10.1016/i.envres.2004.09.014.
Draft - Do Not Cite or Quote                        11-92                                  September 2011

-------
Yang, Q: Cunnold, DM: Choi, Y: Wang, Y: Nam, J: Wang, HJ: Froidevaux, L: Thompson, AM: Bhartia, PK.
      (2010). A study of tropospheric ozone column enhancements over North America using satellite data and
      a global chemical transport model. J Geophys Res 115: D08302.
      http://dx.doi.org/10.1029/2009JD012616.
Yang. X: Cox. RA: Warwick. NJ: Pvle. JA: Carver. GD: O'connor. FM: Savage. NH. (2005c). Tropospheric
      bromine chemistry and its impacts on ozone: A model study. J Geophys Res 110: D23311.
      http://dx.doi. org/10.1029/2005JD006244.
Yokovama. E: Frank. R. (1972). Respiratory uptake of ozone in dogs. Arch Environ Occup Health 25: 132-138.
Yokovama. E: Uchivama. I: Arito. H. (1989). Extrapulmonary effects of low level ozone exposure. In T
      Schneider; SD Lee; GJR Wolters; LD Grant  (Eds.), Atmospheric ozone research and its policy
      implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The
      Netherlands (Vol. 35, pp. 301-309). Nijmegen, The Netherlands: Elsevier.
Yoon, HK;  Cho, HY; Kleeberger, SR. (2007). Protective role of matrix metalloproteinase-9 in ozone-induced
      airway inflammation. Environ Health Perspect 115: 1557-1563. http://dx.doi.org/10.1289/ehp.10289.
Yoshida. S: Tamaoki. M: loki. M: Ogawa. D: Sato. Y: Aono. M:  Kubo. A: Sail. S: Sail. H: Satoh. S: Nakaiima. N.
      (2009). Ethylene and salicylic acid control glutathione biosynthesis in ozone-exposed Arabidopsis
      thaliana. Physiol Plant 136: 284-298. http://dx.doi.Org/10.1111/i.1399-3054.2009.01220.x.
Yost. BL: Gleich.  GJ: Jacobv. DB: Fryer. AD. (2005). The changing  role of eosinophils in long-term
      hyperreactivity following a single ozone exposure. Am J Physiol Lung Cell Mol Physiol 289: L627-L635.
      http://dx.doi.org/10.1152/aiplung.00377.2004.
Younglove, T; McCool, PM; Musselman,  RC; Kahl,  ME. (1994). Growth-stage  dependent crop yield response to
      ozone exposure. Environ Pollut 86: 287-295. http://dx.doi.org/10.1016/0269-7491(94)90169-4.
Yu. M: Zheng. X: Witschi. H: Pinkerton. KE. (2002). The role of interleukin-6 in pulmonary inflammation and
      injury induced by exposure to environmental air pollutants. Toxicol Sci 68: 488-497.
Yuan, JS;  Himanen, SJ; Holopainen, JK; Chen, F; Stewart, CN, Jr.  (2009). Smelling global climate change:
      Mitigation  of function for plant volatile organic compounds. Trends Ecol Evol 24: 323-331.
      http://dx.doi.0rg/10.1016/i.tree.2009.01.012.
Yun, S, -C; Laurence, JA. (1999). The response of sensitive and tolerant clones of Populus tremuloides to
      dynamic ozone exposure under controlled environmental  conditions. New Phytol 143: 305-313.
Yung, YL;  Pinto, JP; Watson, RT; Sander, SP (1980). Atmospheric bromine and ozone perturbations in the
      lower stratosphere. J Atmos Sci 37: 339-353.
Zak. PR: Holmes. WE: Pregitzer. KS. (2007). Atmospheric CO2 and O3 alter the flow of N15 in developing
      forest ecosystems. Ecology 88:  2630-2639.
Zanobetti,  A; Schwartz, J. (In Press) Ozone and survival in four cohorts with potentially predisposing diseases.
      Am J Respir Grit Care Med.  http://dx.doi.org/10.1164/rccm.201102-0227OC.
Zanobetti.  A: Canner.  MJ: Stone. PH: Schwartz. J: Sher. D: Eagan-Bengston.  E: Gates. KA: Hartley. LH: Suh. H:
      Gold, DR. (2004). Ambient pollution and blood pressure in cardiac rehabilitation patients. Circulation 110:
      2184-2189. http://dx.doi.org/10.1161/01.cir.0000143831.33243.d8.
Zanobetti,  A; Schwartz, J. (2006). Air pollution and emergency admissions in Boston, MA. J Epidemiol
      Community Health 60: 890-895.
Zanobetti.  A: Schwartz. J. (2007). Particulate air pollution, progression, and survival after myocardial infarction.
      Environ Health Perspect 115: 769-775.
Zanobetti,  A; Schwartz, J. (2008a).  Is there adaptation in the ozone mortality relationship: A multi-city case-
      crossover analysis. Environ Health 7: 22. http://dx.doi.org/10.1186/1476-069X-7-22.
Zanobetti.  A: Schwartz. J. (2008b).  Mortality displacement in the association of ozone with mortality: An analysis
      of 48 cities in the United States. Am J Respir Grit Care Med  177: 184-189.
      http://dx.doi.org/10.1164/rccm.200706-823OC.
Zanobetti,  A; Bind, MAC; Schwartz, J. (2008). Particulate air pollution and survival in a COPD cohort. Environ
      Health Perspect 7: 48.
Zanobetti.  A: Gold.  PR: Stone.  PH:  Suh.  HH: Schwartz. J: Coull.  BA: Speizer.  FE. (2010). Reduction in heart
      rate variability with traffic and air pollution  in patients with  coronary artery disease. Environ Health
      Perspect 118: 324-330.
Zartarian.  VG: Schultz. BD. (2010). The EPA's human exposure  research program for assessing cumulative risk
      in communities. J Expo Sci Environ Epidemiol 20: 351-358.  http://dx.doi.org/10.1038/ies.2009.20.
Zauli Sajani, S; Hanninen, O; Marches!, S; Lauriola, P. (2011).  Comparison of  different exposure settings in a
      case-crossover study on air  pollution and daily mortality: Counterintuitive results. J Expo Sci Environ
      Epidemiol 21: 385-394. http://dx.doi.org/10.1038/jes.2Q10.27.
Zeger, SL; Thomas, D; Dominici, F; Samet, JM; Schwartz, J; Dockerv, D; Cohen, A. (2000). Exposure
      measurement error in time-series studies of air pollution: Concepts and consequences. Environ Health
      Perspect 108: 419-426.
Zelac, RE; Cromrov, HL; Bolch, WE, Jr; Dunavant,  BG;  Bevis,  HA.  (1971a). Inhaled ozone as a mutagen: I
      chromosome aberrations induced in Chinese hamster lymphocytes. Environ Res 4: 262-282.
Draft - Do Not Cite or Quote                        11-93                                  September 2011

-------
Zelac, RE: Cromroy, Hl_: Bolch, WE, Jr: Dunavant, BG: Bevis, HA. (1971b). Inhaled ozone as a mutagen: II
      effect on the frequency of chromosome aberrations observed in irradiated Chinese hamsters. Environ
      Res 4: 325-342.
Zepp, RG: Erickson, DJ: Paul, ND: Sulzberger, B. (2007). Interactive effects of solar UV radiation and climate
      change on biogeochemical cycling. In The Environmental Effects of Ozone Depletion and its Interactions
      with Climate Change: 2006 Assessment. Nairobi, Kenya: United Nations Environment Programme.
Zepp, RG: Shank, GC: Stabenau, E: Patterson, KW: Cvterski, M: Fisher, W: Bartels, E: Anderson, SL. (2008).
      Spatial and temporal variability of solar ultraviolet exposure of coral assemblages in the Florida Keys:
      Importance of colored dissolved organic matter. Limnol Oceanogr 53: 1909-1922.
Zerefos, CS: Kourtidis, KA:  Melas, D: Balis,  D: Zanis,  P: Katsaros, l_: Mantis, HT: Repapis, C: Isaksen, I:
      Sundet, J: Herman, J: Bhartia, PK: Calpini, B. (2002). Photochemical activity and solar ultraviolet
      radiation (PAUR) modulation factors: An overview of the project. J Geophys Res 107: 8134.
      http://dx.doi. org/10.1029/2000JD000134.
Zhang. C: Tian. HQ: Chappelka. AH: Ren. W: Chen. H: Pan. SF: Liu. ML: Stvers. DM:  Chen. GS: Wang. YH.
      (2007a). Impacts of climatic and atmospheric changes on carbon dynamics in the Great Smoky
      Mountains National Park. Environ Pollut 149: 336-347. http://dx.doi.Org/10.1016/i.envpol.2007.05.028.
Zhang. J: Schaub. M:  Ferdinand. JA: Skellv. JM: Steiner. KG: Savage. JE.  (201 Oa). Leaf age affects the
      responses of foliar injury and gas exchange to tropospheric ozone in Prunus serotina seedlings. Environ
      Pollut 158: 2627-2634. http://dx.doi.Org/10.1016/i.envpol.2010.05.003.
Zhang, K: Wexler, A. (2008). Modeling urban and regional aerosols: Development of the UCD Aerosol Module
      and implementation in CMAQ model. Atmos Environ  42: 3166-3178.
Zhang, l_: Jacob, DJ: Downey, NV: Wood, DA: Blewitt, D: Carouge, CC: Van donkelaar, A: Jones, DBA: Murray,
      LT: Wang. Y (In Press) Improved estimate of the policy-relevant background ozone in the United States
      using the GEOS-Chem global model with 1/2° * 2/3° horizontal resolution over North America. Atmos
      Environ. http://dx.doi.Org/10.1016/i.atmosenv.2011.07.054.
Zhang. L: Jacob. DJ: Boersma. KF: Jaffe. DA: Olson. JR: Bowman. KW:  Worden. JR: Thompson. AM: Averv.
      MA: Cohen. RC: Dibb. JE:  Flock. FM: Fuelberg. HE:  Huev. LG:  McMillan. WW:  Singh. HB: Weinheimer.
      AJ. (2008). Transpacific transport of ozone pollution and the effect of recent Asian emission increases on
      air quality in North America: An integrated analysis using satellite, aircraft, ozonesonde, and surface
      observations. Atmos  Chem Phys 8: 6117-6136.
Zhang, L: Jacob, DJ: Logan, JA: Chance, K:  Eldering,  A: Bojkov, BR. (201 Ob). Intercomparison methods for
      satellite measurements of atmospheric composition: Application to tropospheric ozone from TES and
      OMI. Atmos Chem Phys 10: 4725-4739. http://dx.doi.org/10.5194/acpd-10-1417-2010.
Zhang, Q: Jimenez, Jl_: Canagaratna, MR: Jayne, JT:  Worsnop, DR. (2005). Time- and size-resolved chemical
      composition of  submicron particles in  Pittsburgh: Implications for aerosol sources and processes.  J
      Geophys Res 110: 1-19.
Zhang, X: Zhuang, G:  Quo,  J: Yin, K: Zhang, P. (2007b). Characterization of aerosol over the Northern South
      China Sea during two cruises in 2003. Atmos Environ 41: 7821-7836.
Zheng, F: Wang, X: Lu, F: Hou, P: Zhang, W: Duan, X: Zhou, X: Ai, Y:  Zheng,  H: Ouvang, Z: Feng, Z. (2011).
      Effects of elevated ozone concentration on methane  emission from a rice paddy in Yangtze River Delta,
      China. Global Change Biol 17: 898-910. http://dx.doi.Org/10.1111/i.1365-2486.2010.02258.x.
Zidek. JV: Shaddick. G: Meloche. J: Chatfield. C: White. R. (2007). A framework for predicting personal
      exposures to environmental hazards.  Environ Ecol Stat 14: 411-431.
Ziemke. JR: Chandra. S:  Bhartia.  PK. (2005). A 25-year data record of atmospheric ozone in the Pacific from
      Total Ozone Mapping Spectrometer (TOMS) cloud slicing: Implications  for ozone trends in the
      stratosphere and troposphere. J Geophys Res  110: D15105. http://dx.doi.org/10.1029/2004JD005687.
Ziemke, JR: Chandra, S:  Duncan, BN: Froidevaux, L:  Bhartia, PK: Levelt, PF:  Waters, JW. (2006). Tropospheric
      ozone determined from Aura OMI and MLS: Evaluation of measurements and comparison with the
      Global Modeling Initiative's Chemical Transport Model. J Geophys Res 111: D19303.
      http://dx.doi. org/10.1029/2006JD007089.
Zimmermann. J: Poppe. D. (1993). Nonlinear chemical couplings in the tropospheric NOx-HOx gas phase
      chemistry. J Atmos Chem 17: 141-155.
Draft - Do Not Cite or Quote                       11-94                                 September 2011

-------