^^ ^^ P^ A
> EPA
United States
Air Quality Criteria for
Ozone and Related
Photochemical Oxidants
Volume II of
-------
EPA 600/R-05/004bF
February 2006
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
Volume II
National Center for Environmental Assessment-RTF Office
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
PREFACE
National Ambient Air Quality Standards (NAAQS) are promulgated by the United States
Environmental Protection Agency (EPA) to meet requirements set forth in Sections 108 and 109
of the U.S. Clean Air Act (CAA). Sections 108 and 109 require the EPA Administrator
(1) to list widespread air pollutants that reasonably may be expected to endanger public health
or welfare; (2) to issue air quality criteria for them that assess the latest available scientific
information on nature and effects of ambient exposure to them; (3) to set "primary" NAAQS to
protect human health with adequate margin of safety and to set "secondary" NAAQS to protect
against welfare effects (e.g., effects on vegetation, ecosystems, visibility, climate, manmade
materials, etc); and (5) to periodically review and revise, as appropriate, the criteria and NAAQS
for a given listed pollutant or class of pollutants.
In 1971, the U.S. Environmental Protection Agency (EPA) promulgated National Ambient
Air Quality Standards (NAAQS) to protect the public health and welfare from adverse effects of
photochemical oxidants. The EPA promulgates the NAAQS on the basis of scientific
information contained in air quality criteria issued under Section 108 of the Clean Air Act.
Following the review of criteria as contained in the EPA document, Air Quality Criteria for
Ozone and Other Photochemical Oxidants published in 1978, the chemical designation of the
standards was changed from photochemical oxidants to ozone (O3) in 1979 and a 1-hour O3
NAAQS was set. The 1978 document focused mainly on the air quality criteria for O3 and, to a
lesser extent, on those for other photochemical oxidants (e.g., hydrogen peroxide and the
peroxyacyl nitrates), as have subsequent revised versions of the document.
To meet Clean Air Act requirements noted above for periodic review of criteria and
NAAQS, the O3 criteria document, Air Quality Criteria for Ozone and Other Photochemical
Oxidants, was next revised and released in August 1986; and a supplement, Summary of Selected
Il-ii
-------
New Information on Effects of Ozone on Health and Vegetation, was issued in January 1992.
These documents were the basis for a March 1993 decision by EPA that revision of the existing
1-h NAAQS for O3 was not appropriate at that time. That decision, however, did not take into
account newer scientific data that had become available after completion of the 1986 criteria
document. Such literature was assessed in the next periodic revision of the O3 air quality criteria
document (O3 AQCD) which has completed in 1996 and provided scientific bases supporting the
setting by EPA in 1997 of the current 8-h O3 NAAQS.
The purpose of this revised air quality criteria document for O3 and related photochemical
oxidants is to critically evaluate and assess the latest scientific information published since that
assessed in the above 1996 O3 AQCD, with the main focus being on pertinent new information
useful in evaluating health and environmental effects data associated with ambient air O3
exposures. However, other scientific data are also discussed in order to provide a better
understanding of the nature, sources, distribution, measurement, and concentrations of O3 and
related photochemical oxidants and their precursors in the environment. The document mainly
assesses pertinent literature published through 2004, but also includes assessment of a few
additional important studies published or accepted for publication in 2005.
A First External Review Draft of this O3 AQCD (dated January 2005) was released for
public comment and was reviewed by the Clean Air Scientific Advisory Committee (CASAC)
in May, 2005 to obtain. Public comments and CASAC recommendations were then taken into
account in making revisions to the document for incorporation into a Second External Review
Draft (dated August, 2005), which underwent further public comment and CASAC review at a
December, 2005 public meeting. Public comments and CASAC advice derived from review of
that Second External Review Draft were considered in making revisions incorporated into this
final version of the document (dated February, 2006). Evaluations contained in the present
document will be drawn on to provide inputs to associated O3 Staff Paper analyses prepared by
EPA's Office of Air Quality Planning and Standards (OAQPS) to pose options for consideration
by the EPA Administrator with regard to proposal and, ultimately, promulgation of decisions on
potential retention or revision, as appropriate, of the current O3 NAAQS.
Preparation of this document was coordinated by staff of EPA's National Center for
Environmental Assessment in Research Triangle Park (NCEA-RTP). NCEA-RTP scientific
staff, together with experts from other EPA/ORD laboratories and academia, contributed to
-------
writing of document chapters. Earlier drafts of document materials were reviewed by non-EPA
experts in peer consultation workshops held by EPA. The document describes the nature,
sources, distribution, measurement, and concentrations of O3 in outdoor (ambient) and indoor
environments. It also evaluates the latest data on human exposures to ambient O3 and
consequent health effects in exposed human populations, to support decision making regarding
the primary, health-related O3 NAAQS. Lastly, the document also evaluates ambient O3
environmental effects on vegetation and ecosystems, surface level solar UV radiation flux and
global climate change, and man-made materials to support decision making on secondary
O3 NAAQS.
NCEA acknowledges the valuable contributions provided by authors, contributors, and
reviewers and the diligence of its staff and contractors in the preparation of this document.
Il-iv
-------
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
VOLUME I
Executive Summary E-l
1. INTRODUCTION 1-1
2. PHYSICS AND CHEMISTRY OF OZONE IN THE ATMOSPHERE 2-1
3. ENVIRONMENTAL CONCENTRATIONS, PATTERNS, AND
EXPOSURE ESTIMATES 3-1
4. DOSIMETRY, SPECIES HOMOLOGY, SENSITIVITY, AND
ANIMAL-TO-HUMAN EXTRAPOLATION 4-1
5. TOXICOLOGICAL EFFECTS OF OZONE AND RELATED
PHOTOCHEMICAL OXIDANTS IN LABORATORY ANIMALS
AND IN VITRO TEST SYSTEMS 5-1
6. CONTROLLED HUMAN EXPOSURE STUDIES OF OZONE AND
RELATED PHOTOCHEMICAL OXIDANTS 6-1
7. EPIDEMIOLOGICAL STUDIES OF HUMAN HEALTH EFFECTS
ASSOCIATED WITH AMBIENT OZONE EXPOSURE 7-1
8. INTEGRATIVE SYNTHESIS: EXPOSURE AND HEALTH EFFECTS 8-1
9. ENVIRONMENTAL EFFECTS: OZONE EFFECTS ON
VEGETATION AND ECOSYSTEMS 9-1
10. TROPOSPHERIC OZONE EFFECTS ON UV-B FLUX AND
CLIMATE CHANGE PROCESSES 10-1
11. OZONE EFFECTS ON MAN-MADE MATERIALS 11-1
II-v
-------
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
(cont'd)
VOLUME II
CHAPTER 2 ANNEX (ATMOSPHERIC PHYSICS/CHEMISTRY) AX2-1
CHAPTER 3 ANNEX (AIR QUALITY AND EXPOSURE) AX3-1
CHAPTER 4 ANNEX (DOSIMETRY) AX4-1
CHAPTER 5 ANNEX (ANIMAL TOXICOLOGY) AX5-1
CHAPTER 6 ANNEX (CONTROLLED HUMAN EXPOSURE) AX6-1
CHAPTER 7 ANNEX (EPIDEMIOLOGY) AX7-1
VOLUME III
CHAPTER 9 ANNEX (ENVIRONMENTAL EFFECTS) AX9-1
Il-vi
-------
age
Table of Contents
List of Tables II-xiv
List of Figures II-xix
Authors, Contributors, and Reviewers II-xxxi
U.S. Environmental Protection Agency Project Team for Development of
Air Quality Criteria for Ozone and Related Photochemical Oxidants II-xxxviii
U.S. Environmental Protection Agency Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
Ozone Review Panel II-xli
ABBREVIATIONS AND ACRONYMS II-xliv
AX2. PHYSICS AND CHEMISTRY OF OZONE IN THE ATMOSPHERE AX2-1
AX2.1 INTRODUCTION AX2-1
AX2.2 TROPOSPHERIC OZONE CHEMISTRY AX2-2
AX2.2.1 Atmospheric Structure AX2-2
AX2.2.2 Overview of Ozone Chemistry AX2-3
AX2.2.3 Initiation of the Oxidation of VOCs AX2-6
AX2.2.4 Chemistry of Nitrogen Oxides in the Troposphere .... AX2-11
AX2.2.5 The Methane Oxidation Cycle AX2-14
AX2.2.6 The Atmospheric Chemistry of Alkanes AX2-20
AX2.2.7 The Atmospheric Chemistry of Alkenes AX2-23
AX2.2.8 The Atmospheric Chemistry of Aromatic
Hydrocarbons AX2-31
AX2.2.8.1 Chemical Kinetics and Atmospheric
Lifetimes of Aromatic
Hydrocarbons AX2-32
AX2.2.8.2 Reaction Products and Mechanisms
of Aromatic Hydrocarbon Oxidation . . . AX2-36
AX2.2.8.3 The Formation of Secondary
Organic Aerosol as a Sink for
Ozone Precursors AX2-44
AX2.2.9 Importance of Oxygenated VOCs AX2-44
AX2.2.10 Influence of Multiphase Chemical Processes AX2-45
AX2.2.10.1 HOX and Aerosols AX2-47
AX2.2.10.2 NOX Chemistry AX2-50
AX2.2.10.3 Halogen Radical Chemistry AX2-52
AX2.2.10.4 Reactions on the Surfaces of Crustal
Particles AX2-56
AX2.2.10.5 Reactions on the Surfaces of
Aqueous H2SO4 Solutions AX2-57
AX2.2.10.6 Oxidant Formation in Particles AX2-58
AX2.2.10.7 Ozone Reaction Indoors AX2-60
-------
Table of Contents
(cont'd)
age
AX2.3 PHYSICAL PROCESSES INFLUENCING THE ABUNDANCE
OF OZONE AX2-60
AX2.3.1 Stratospheric-Tropospheric Ozone Exchange AX2-62
AX2.3.2 Deep Convection in the Troposphere AX2-71
AX2.3.2.1 Observations of the Effects of
Convective Transport AX2-73
AX2.3.2.2 Modeling the Effects of Convection ... AX2-76
AX2.3.3 Nocturnal Low-Level Jets AX2-79
AX2.3.4 Intercontinental Transport of Ozone and Other
Pollutants AX2-83
AX2.3.4.1 The Atmosphere/Ocean Chemistry
Experiment, AEROCE AX2-83
AX2.3.4.2 The North Atlantic Regional
Experiment, NARE AX2-86
AX2.3.5 Small-Scale Circulation Systems AX2-89
AX2.3.5.1 Land-Sea Breeze AX2-90
AX2.3.6 The Relation of Ozone to Solar Ultraviolet
Radiation, Aerosols, and Air Temperature AX2-90
AX2.3.6.1 Solar Ultraviolet Radiation
and Ozone AX2-90
AX2.3.6.2 Impact of Aerosols on Radiation
and Photolysis Rates and
Atmospheric Stability AX2-91
AX2.3.6.3 Temperature and Ozone AX2-92
AX2.4 THE RELATION OF OZONE TO ITS PRECURSORS AND
OTHER OXIDANTS AX2-95
AX2.4.1 Summary of Results for the Relations Among
Ozone, its Precursors and Other Oxidants from
Recent Field Experiments AX2-99
AX2.4.1.1 Results from the S outhern Oxidant
Study and Related Experiments AX2-99
AX2.4.1.2 Results from Studies on Biogenic and
Anthropogenic Hydrocarbons and
Ozone Production AX2-103
AX2.4.1.3 Results of Studies on Ozone
Production in Mississippi and
Alabama AX2-103
AX2.4.1.4 The Nocturnal Urban Plume over
Portland, Oregon AX2-104
AX2.4.1.5 Effects of VOCs in Houston on
Ozone Production AX2-104
-------
Table of Contents
(cont'd)
age
AX2.4.1.6 Chemical and Meteorological
Influences on the Phoenix Urban
Ozone Plume AX2-105
AX2.4.1.7 Transport of Ozone and Precursors
on the Regional Scale AX2-105
AX2.4.1.8 Model Calculations and Aircraft
Observations of Ozone over
Philadelphia AX2-106
AX2.4.1.9 The Two-Reservoir System AX2-107
AX2.5 METHODS USED TO CALCULATE RELATIONS BETWEEN
OZONE AND ITS PRECURSORS AX2-107
AX2.5.1 Chemistry-Transport Models AX2-109
AX2.5.2 Emissions of Ozone Precursors AX2-123
AX2.5.3 Observationally-Based Models AX2-130
AX2.5.4 Chemistry-Transport Model Evaluation AX2-131
AX2.5.4.1 Evaluation of Emissions
Inventories AX2-144
AX2.5.4.2 Availability and Accuracy of
Ambient Measurements AX2-146
AX2.6 TECHNIQUES FOR MEASURING OZONE AND ITS
PRECURSORS AX2-147
AX2.6.1 Sampling and Analysis of Ozone AX2-147
AX2.6.2 Sampling and Analysis of Nitrogen Oxides AX2-149
AX2.6.2.1 Calibration Standards AX2-150
AX2.6.2.2 Measurement of Nitric Oxide AX2-151
AX2.6.2.3 Measurements of Nitrogen Dioxide . . AX2-152
AX2.6.2.4 Monitoring forNO2 Compliance
Versus Monitoring for Ozone
Formation AX2-153
AX2.6.3 Measurements of Nitric Acid Vapor, HNO3 AX2-154
AX2.6.4 Sampling and Analysis of Volatile Organic
Compounds AX2-156
AX2.6.4.1 Polar Volatile Organic Compounds . . . AX2-156
REFERENCES AX2-159
AX3. ENVIRONMENTAL CONCENTRATIONS, PATTERNS, AND
EXPOSURE ESTIMATES AX3-1
AX3.1 INTRODUCTION AX3-1
AX3.2 SURFACE OZONE CONCENTRATIONS AX3-4
AX3.2.1 Nationwide Distribution of Metrics for Characterizing
Exposures of Vegetation to Ozone AX3-27
Il-ix
-------
Table of Contents
(cont'd)
Page
AX3.3 SPATIAL VARIABILITY IN OZONE CONCENTRATIONS AX3-30
AX3.3.1 Spatial Variability of Ozone Concentrations in
Urban Areas AX3-30
AX3.3.2 Small-scale Horizontal and Spatial Variability in
Ozone Concentrations AX3-56
AX3.3.3 Ozone Concentrations at High Elevations AX3-58
AX3.4 DIURNAL PATTERNS IN OZONE CONCENTRATION AX3-68
AX3.4.1 Introduction AX3-68
AX3.4.2 Diurnal Patterns in Urban Areas AX3-70
AX3.4.3 Diurnal Patterns in Nonurban Areas AX3-87
AX3.5 SEASONAL VARIATIONS IN OZONE CONCENTRATIONS .... AX3-95
AX3.5.1 Seasonal Variations in Urban Areas AX3-95
AX3.5.2 Seasonal Variations in Nonurban Areas AX3-98
AX3.6 TRENDS IN OZONE CONCENTRATIONS AX3-104
AX3.7 RELATIONS BETWEEN OZONE, OTHER OXIDANTS,
AND OXIDATION PRODUCTS AX3-114
AX3.8 RELATIONSHIP BETWEEN SURFACE OZONE AND
OTHER POLLUTANTS AX3-121
AX3.8.1 Introduction AX3-121
AX3.8.2 Co-Occurrence of Ozone with Nitrogen Oxides AX3-123
AX3.8.3 Co-Occurrence of Ozone with Sulfur Dioxide AX3-125
AX3.8.4 Co-Occurrence of Ozone and Daily PM25 AX3-127
AX3.8.5 Co-Occurrence of Ozone with Acid Precipitation .... AX3-128
AX3.8.6 Co-Occurrence of Ozone with Acid Cloudwater AX3-130
AX3.9 THE METHODOLOGY FOR DETERMINING POLICY
RELEVANT BACKGROUND OZONE CONCENTRATIONS . . . AX3-131
AX3.9.1 Introduction AX3-131
AX3.9.2 Capability of Global Models to Simulate
Tropospheric Ozone AX3-147
AX3.9.3 Mean Background Concentrations: Spatial and
Seasonal Variation AX3-150
AX3.9.4 Frequency of High-Ozone Occurrences at
Remote Sites AX3-153
AX3.10 OZONE EXPOSURE IN VARIOUS MICROENVIRONMENTS . . . AX3-161
AX3.10.1 Introduction AX3-161
AX3.10.2 Summary of the Information Presented in
the Exposure Discussion in the 1996 Ozone
Criteria Document AX3-161
AX3.10.3 Concepts of Human Exposure AX3-162
AX3.10.4 Quantification of Exposure AX3-162
II-x
-------
Table of Contents
(cont'd)
age
AX3.10.5 Methods to Estimate Personal Exposure AX3-163
AX3.10.5.1 Direct Measurement Method AX3-163
AX3.10.5.2 Indirect Measurement Method AX3-165
AX3.10.6 Ozone Exposure Models AX3-167
AX3.10.6.1 Population Exposure Models AX3-172
AX3.10.6.2 Ambient Concentrations Models AX3-180
AX3.10.6.3 Microenvironmental Concentration
Models AX3-182
AX3.10.7 Measured Exposures and Monitored
Concentrations AX3-186
AX3.10.7.1 Personal Exposure Measurements .... AX3-186
AX3.10.7.2 Monitored Ambient
Concentrations AX3-190
AX3.10.7.3 Ozone Concentrations in
Microenvironments AX3-191
AX3.10.8 Trends in Concentrations within
Microenvironments AX3-216
AX3.10.9 Characterization of Exposure AX3-217
AX3.10.9.1 Use of Ambient Ozone
Concentrations AX3-217
AX3.10.9.2 Exposure Selection in Controlled
Exposure Studies AX3-219
AX3.10.9.3 Exposure to Related Photochemical
Agents AX3-219
REFERENCES AX3-223
AX4. DOSIMETRY OF OZONE IN THE RESPIRATORY TRACT AX4-1
AX4.1 INTRODUCTION AX4-1
AX4.2 EXPERIMENTAL OZONE DOSIMETRY INVESTIGATIONS AX4-3
AX4.2.1 Bolus-Response Studies AX4-3
AX4.2.2 General Uptake Studies AX4-9
AX4.2.3 Dosimetry Modeling AX4-11
AX4.3 SPECIES HOMOLOGY, SENSITIVITY AND
ANIMAL-TO-HUMAN EXTRAPOLATION AX4-17
REFERENCES AX4-20
AX5. ANNEX TO CHAPTER 5 OF OZONE AQCD AX5-1
REFERENCES AX5-62
Il-xi
-------
Table of Contents
(cont'd)
age
AX6. CONTROLLED HUMAN EXPOSURE STUDIES OF OZONE AND
RELATED PHOTOCHEMICAL OXIDANTS AX6-1
AX6.1 INTRODUCTION AX6-1
AX6.2 PULMONARY FUNCTION EFFECTS OF OZONE EXPOSURE
IN HEALTHY SUBJECTS AX6-2
AX6.2.1 Introduction AX6-2
AX6.2.2 Acute Ozone Exposures for Up to 2 Hours AX6-3
AX6.2.3 Prolonged Ozone Exposures AX6-10
AX6.2.3.1 Effect of Exercise Ventilation Rate
on FEV] Response to 6.6 h Ozone
Exposure AX6-13
AX6.2.3.2 Exercise Ventilation Rate as a
Function of Body/Lung Size on FEV]
Response to 6.6 h Ozone Exposure .... AX6-14
AX6.2.3.3 Comparison of 6.6 h Ozone Exposure
Pulmonary Responses to Those
Observed in 2 h Intermittent Exercise
Ozone Exposures AX6-16
AX6.2.4 Triangular Ozone Exposures AX6-17
AX6.2.5 Mechanisms of Pulmonary Function Responses AX6-21
AX6.2.5.1 Pathophysiologic Mechanisms AX6-22
AX6.2.5.2 Mechanisms at a Cellular and
Molecular Level AX6-28
AX6.3 PULMONARY FUNCTION EFFECTS OF OZONE EXPOSURE
IN SUBJECTS WITH PREEXISTING DISEASE AX6-29
AX6.3.1 Subjects with Chronic Obstructive Pulmonary
Disease AX6-29
AX6.3.2 Subjects with Asthma AX6-34
AX6.3.3 Subjects with Allergic Rhinitis AX6-38
AX6.3.4 Subjects with Cardiovascular Disease AX6-40
AX6.4 INTERSUBJECT VARIABILITY AND REPRODUCIBILITY
OF RESPONSE AX6-41
AX6.5 INFLUENCE OF AGE, GENDER, ETHNIC, ENVIRONMENTAL
AND OTHER FACTORS AX6-45
AX6.5.1 Influence of Age AX6-45
AX6.5.2 Gender and Hormonal Influences AX6-51
AX6.5.3 Racial, Ethnic, and Socioeconomic Status Factors .... AX6-58
AX6.5.4 Influence of Physical Activity AX6-59
AX6.5.5 Environmental Factors AX6-60
AX6.5.6 Oxidant-Antioxidant Balance AX6-65
AX6.5.7 Genetic Factors AX6-68
-------
Table of Contents
(cont'd)
Page
AX6.6 REPEATED EXPOSURES TO OZONE AX6-70
AX6.7 EFFECTS ON EXERCISE PERFORMANCE AX6-81
AX6.7.1 Introduction AX6-81
AX6.7.2 Effect on Maximal Oxygen Uptake AX6-82
AX6.7.3 Effect on Endurance Exercise Performance AX6-84
AX6.8 EFFECTS ON AIRWAY RESPONSIVENESS AX6-86
AX6.9 EFFECTS ON INFLAMMATION AND HOST DEFENSE AX6-97
AX6.9.1 Introduction AX6-97
AX6.9.2 Inflammatory Responses in the Upper
Respiratory Tract AX6-113
AX6.9.3 Inflammatory Responses in the Lower
Respiratory Tract AX6-115
AX6.9.4 Adaptation of Inflammatory Responses AX6-120
AX6.9.5 Effect of Anti-Inflammatory and Other
Mitigating Agents AX6-122
AX6.9.6 Changes in Host Defense Capability Following
Ozone Exposure AX6-124
AX6.10 EXTRAPULMONARY EFFECTS OF OZONE AX6-126
AX6.11 OZONE MIXED WITH OTHER POLLUTANTS AX6-129
AX6.11.1 Ozone and Sulfur Oxides AX6-129
AX6.11.2 Ozone and Nitrogen-Containing Pollutants AX6-135
AX6.11.3 Ozone and Other Pollutant Mixtures Including
Particulate Matter AX6-137
AX6.12 CONTROLLED STUDIES OF AMBIENT AIR EXPOSURES .... AX6-138
AX6.12.1 Mobile Laboratory Studies AX6-138
AX6.12.2 Aircraft Cabin Studies AX6-139
REFERENCES AX6-142
AX7. EPIDEMIOLOGIC STUDIES OF HUMAN HEALTH EFFECTS
ASSOCIATED WITH AMBIENT OZONE EXPOSURE AX7-1
AX7.1 TABLES OF EPIDEMIOLOGIC STUDIES OF HUMAN
HEALTH EFFECTS ASSOCIATED WITH AMBIENT
OZONE EXPOSURE AX7-2
LIST OF STUDIES BY AUTHOR AX7-3
AX7.2 DESCRIPTION OF SUMMARY DENSITY CURVES AX7-121
AX7.3 ESTIMATING THE GOMPERTZ CONCENTRATION-
RESPONSE MODEL AX7-130
-------
List of Tables
Number Page
AX2-1 Comparison of the Atmospheric Lifetimes (t) of Low Molecular Weight
Hydrocarbons Due to Reaction with OH, NO3, Cl, Br, and O3 AX2-7
AX2-2 Calculated Atmospheric Lifetimes of Biogenic Volatile Organic
Compounds AX2-25
AX2-3 Hydroxyl Rate Constants and Atmospheric Lifetimes of Mono- and
Di-cyclic Aromatic Hydrocarbons AX2-34
AX2-4 Chemistry-Transport Models (CTM) Contributing to the Oxcomp
Evaluation of Predicting Tropospheric O3 and OH AX2-122
AX2-5 Emissions of Nitrogen Oxides by Various Sources in the United States
in 1999 AX2-124
AX2-6 Emissions of Volatile Organic Compounds by Various Sources in the
United States in 1999 AX2-125
AX2-7 Emissions of Ammonia by Various Sources in the United States
in 1999 AX2-126
AX2-8 Emissions of Carbon Monoxide by Various Sources in the United States
in 1999 AX2-127
AX3-1 Ozone Monitoring Seasons by State AX3-3
AX3-2 Summary of Percentiles of Pooled Data Across Monitoring Sites for
May to September 2000-2004 Concentrations are in ppb AX3-11
AX3-3 Seasonal (April to October) Percentile Distribution of Hourly Ozone
Concentrations (ppm), Number of Hourly Mean Ozone Occurrences
>0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and
W126 Values for Sites Experiencing Low Maximum Hourly Average
Concentrations with Data Capture of > 75% AX3-18
AX3-4 The Top 10 Daily Maximum 8-h Average Concentrations (ppm) for
Sites Experiencing Low Maximum Hourly Average Concentrations
with Data Capture of >75% AX3-23
AX3-5 Summary Statistics for Ozone (in ppm) Spatial Variability in Selected
U.S. Urban Areas AX3-42
-------
List of Tables
(cont'd)
Number Page
AX3-6 Description of Mountain Cloud Chemistry Program Sites AX3-60
AX3-7 Seasonal (April-October) Percentiles, SUM06, SUM08, and W126
Values for the MCCP Sites AX3-61
AX3-8 Summary Statistics for 11 Integrated Forest Study Sites AX3-66
AX3-9 Trends in Warm Season (May to September) Daily Maximum 8-h O3
Concentrations at National Parks in the United States (1990 to 2004) AX3-113
AX3-10 Range of Annual (January-December) Hourly Ozone Concentrations
(ppb) at Background Sites Around the World (CMDL, 2004) AX3-137
AX3-11 Range of annual (January-December) Hourly Median and Maximum
Ozone Concentrations (ppb) at Background Stations in Protected Areas
of the United States (CASTNet, 2004) AX3-137
AX3-12 Range of Annual (January-December) Hourly Median and Maximum
Ozone Concentrations (ppb) at Canadian Background Stations
(CAPMoN, 2003) AX3-138
AX3-13 Number of Hours > 0.05 ppm for Selected Rural O3 Monitoring in the
United States by Month for the Period 1988 to 2001 AX3-139
AX3-14 Number of Hours >0.06 ppm for Selected Rural O3 Monitoring Sites
in the United States by Month for the Period of 1988 to 2001 AX3-141
AX3-15 Global Budgets of Tropospheric Ozone (Tg year"1) for the Present-day
Atmosphere AX3-146
AX3-16 Description of Simulations Used for Source Attribution AX3-151
AX3-17 Number of Hours with Ozone Above 50 or 60 ppbv at U.S.
CASTNet Sites in 2001 AX3-155
AX3-18 Activity Pattern Studies Included in the Consolidated Human Activity
Database (CHAD) AX3-174
AX3-19 Personal and Population Exposure Models for Ozone AX3-179
AX3-20 Personal Exposure Measurements AX3-187
II-xv
-------
List of Tables
(cont'd)
Number Page
AX3-21 Indoor/Outdoor Ozone Ratios AX3-192
AX3-22 Indoor and Outdoor O3 Concentrations in Boston, MA AX3-197
AX3-23 Indoor and Outdoor O3 Concentrations in Hong Kong AX3-199
AX3-24 Indoor and Outdoor Ozone Concentrations AX3-202
AX3-25 Rate Constants (tf1) for the Removal of Ozone by Surfaces in
Different Indoor Environments AX3-211
AX4-1 New Experimental Human Studies on Ozone Dosimetry AX4-4
AX4-2 New Ozone Dosimetry Model Investigations AX4-13
AX5-1 Cellular Targets of Ozone Interaction AX5-2
AX5-2 Effects of Ozone on Lung Monooxygenases AX5-3
AX5-3 Antioxidants, Antioxidant Metabolism, and Mitochondrial Oxygen
Consumption AX5-4
AX5-4 Lipid Metabolism and Content of the Lung AX5-5
AX5-5 Effects of Ozone on Protein Synthesis AX5-7
AX5-6 Effects of Ozone on Differential Gene Expression AX5-7
AX5-7 Effects of Ozone on Lung Host Defenses AX5-8
AX5-8 Effects of Ozone on Lung Permeability and Inflammation AX5-17
AX5-9 Effects of Ozone on Lung Structure: Acute and Subchronic Exposures .... AX5-27
AX5-10 Effects of Ozone on Lung Structure: Subchronic and Chronic Exposures . . . AX5-31
AX5-11 Effects of Ozone on Pulmonary Function AX5-34
AX5-12 Effects of Ozone on Airway Responsiveness AX5-36
AX5-13 Effects of Ozone on Genotoxicity/Carcinogenicity AX5-43
-------
List of Tables
(cont'd)
Number Page
AX5-14 Systemic Effects of Ozone AX5-44
AX5-15 Interactions of Ozone With Nitrogen Dioxide AX5-51
AX5-16 Interactions of Ozone with Formaldehyde AX5-53
AX5-17 Interactions of Ozone with Tobacco Smoke AX5-54
AX5-18 Interactions Of Ozone With Particles AX5-55
AX5-19 Effects of Other Photochemical Oxidants AX5-61
AX6-1 Controlled Exposure of Healthy Humans to Ozone for 1 to 2 Hours
During Exercise AX6-5
AX6-2 Pulmonary Function Effects after Prolonged Exposures to Ozone AX6-11
AX6-3 Ozone Exposure in Subjects with Preexisting Disease AX6-30
AX6-4 Classification of Asthma Severity AX6-37
AX6-5 Age Differences in Pulmonary Function Responses to Ozone AX6-46
AX6-6 Gender and Hormonal Differences in Pulmonary Function Responses
to Ozone AX6-52
AX6-7 Influence of Ethnic, Environmental, and Other Factors AX6-62
AX6-8 Changes in Forced Expiratory Volume in One Second After Repeated
Daily Exposure to Ozone AX6-71
AX6-9 Pulmonary Function Effects with Repeated Exposures to Ozone AX6-73
AX6-10 Ozone Effects on Exercise Performance AX6-83
AX6-11 Airway Responsiveness Following Ozone Exposures AX6-88
AX6-12 Studies of Respiratory Tract Inflammatory Effects from Controlled
Human Exposure to Ozone AX6-99
AX6-13 Studies of Effects on Host Defense, on Drug Effects and Supportive
In Vitro Studies Relating to Controlled Human Exposure to Ozone AX6-106
-------
List of Tables
(cont'd)
Number Page
AX6-14 Ozone Mixed with Other Pollutants AX6-131
AX6-15 Acute Effects of Ozone in Ambient Air in Field Studies with a
Mobile Laboratory AX6-140
AX7-1 Effects of Acute O3 Exposure on Lung Function and Respiratory
Symptoms in Field Studies AX7-9
AX7-2 Effects of Acute O3 Exposure on Cardiovascular Outcomes in
Field Studies AX7-33
AX7-3 Effects of O3 on Daily Emergency Department Visits AX7-40
AX7-4 Effects of O3 on Daily Hospital Admissions AX7-52
AX7-5 Effects of Acute O3 Exposure on Mortality AX7-71
AX7-6 Effects of Chronic O3 Exposure on Respiratory Health AX7-102
AX7-7 Effects of Chronic O3 Exposure on Mortality and Incidence of Cancer .... AX7-118
AX7-8 Ozone-Associated Cardiovascular Mortality Risk Estimates per
Standardized Increment AX7-126
AX7-9 Estimated Parameters of the Gompertz Model AX7-131
-------
List of Figures
Number Page
AX2-1 Schematic overview of O3 photochemistry in the stratosphere
and troposphere AX2-4
AX2-2 General chemical mechanism for the oxidative degradation of VOCs AX2-21
AX2-3 Hydroxyl radical initiated oxidation of a) propane and b) propene AX2-24
AX2-4 Structures of a selected number of terpene and sesquiterpene compounds . . . AX2-27
AX2-5 Products from the reaction of terpenes with O3 AX2-30
AX2-6 Initial steps in the photooxidation mechanism of toluene initiated by
its reaction with OH radicals AX2-39
AX2-7a Cross section through a tropopause folding event on March 13, 1978
at 0000 GMT AX2-64
AX2-7b Ozone mixing ratios pphm (parts per hundred million) corresponding
to Figure AX2-7a AX2-67
AX2-7c Condensation nuclei concentrations (particles cm"3) corresponding to
Figure AX2-7a AX2-68
AX2-8 Schematic diagram of a meteorological mechanism involved in high
concentrations of O3 found in spring in the lower troposphere off the
American east coast AX2-71
AX2-9a,b (a) Contour plot of CO mixing ratios (ppbv) observed in and near the
June 15, 1985, mesoscale convective complex in eastern Oklahoma AX2-75
AX2-10 The diurnal evolution of the planetary boundary layer while high
pressure prevails over land AX2-80
AX2-11 Locations of low level jet occurrences in decreasing order of prevalence
(most frequent, common, observed) AX2-81
AX2-12 Schematic diagram showing the diurnal behavior of O3 and the
development of secondary O3 maxima resulting from downward
transport from the residual layer when a low-level jet is present AX2-82
AX2-13 The nocturnal low-level jet occupies a thin slice of the atmosphere
near the Earth's surface AX2-82
-------
List of Figures
(cont'd)
Number Page
AX2-14 A scatter plot of daily maximum 8-h O3 concentration versus daily
maximum temperature in the Baltimore, MD Air Quality Forecast Area . . . AX2-93
AX2-15 A scatter plot of daily maximum 1 -h average O3 concentration versus
daily maximum temperature in the Baltimore, MD Air Quality
Forecast Area AX2-93
AX2-16 A scatter plot of daily maximum 8-h average O3 concentrations versus
daily maximum temperature downwind of Phoenix, AZ AX2-94
AX2-17 Measured values of O3 and NOZ (NOy - NOX) during the afternoon at
rural sites in the eastern United States (gray circles) and in urban areas
and urban plumes associated with Nashville, TN (gray dashes), Paris, FR
(black diamonds) and Los Angeles, CA (X's) AX2-98
AX2-18 Conceptual two-reservoir model showing conditions in the PBL and
in the lower free troposphere during a multiday O3 episode AX2-108
AX2-19 Seasonal variability in O3 concentrations observed at a number of
pressure surfaces at six ozonesonde sites and the predictions of
13 global scale chemistry-transport models AX2-120
AX2-20 Seasonal variability in O3 concentrations observed at a number of
pressure surfaces at six ozonesonde sites and the predictions of
13 global scale chemistry-transport models AX2-121
AX2-21a,b Impact of model uncertainty on control strategy predictions for O3 for
two days (August 10[a] and 1 l[b], 1992) in Atlanta, GA AX2-134
AX2-22 Ozone isopleths (ppb) as a function of the average emission rate
for NOX and VOC (1012 molec. cm"2 s~!) in zero dimensional box
model calculations AX2-135
AX2-23a Time series for measured gas-phase species in comparison with results
from a photochemical model AX2-136
AX2-23b Time series for measured gas-phase species in comparison with results
from a photochemical model AX2-137
AX2-24 Correlations for O3 versus NOZ (NOy - NOX) in ppb from chemical
transport models for the northeast corridor, Lake Michigan, Nashville,
the San Joaquin Valley, and Los Angeles AX2-139
II-xx
-------
List of Figures
(cont'd)
Number Page
AX2-25a,b Evaluation of model versus measured O3 versus NOy for two model
scenarios for Atlanta AX2-140
AX2-26a,b Evaluation of model versus: (a) measured O3 versus NOZ and
(b) O3 versus the sum 2H2O2 + NOZ for Nashville, TN AX2-142
AX2-27 Time series of concentrations of RO2, HO2, and OH radicals, local
O3 photochemical production rate and concentrations of NOX from
measurements made during BERLIOZ AX2-143
AX3-1 Countywide mean daily maximum 8-h O3 concentrations, May to
September 2002 to 2004 AX3-6
AX3-2 Countywide 95th percentile value of daily maximum 8-h O3
concentrations, May to September 2002 to 2004 AX3-7
AX3-3 Locations of monitoring sites used for calculating countywide averages
across the United States AX3-8
AX3-4 Distribution of nationwide daily maximum 1-h average O3 concentrations
from May to September 2000 to 2004 AX3-9
AX3-5 Distribution of nationwide daily maximum 8-h average O3 concentrations
from May to September 2000 to 2004 AX3-9
AX3-6 Distribution of nationwide 24-h average O3 concentrations from
May to September 2000 to 2004 AX3-10
AX3-7 Box plots showing O3 averaged by month from 1993 to 2002 in the five
regions in the eastern United States derived by Lehman et al. (2004) AX3-12
AX3-8 Hourly average O3 concentrations observed at selected rural-agricultural
sites from April to October 2001 AX3-13
AX3-9 Hourly average O3 concentrations observed at selected rural-forest sites
from April to October 2001 AX3-14
AX3-10 Hourly average O3 concentrations observed at selected rural-commercial
or -residential sites from April to October 2001 AX3-15
AX3-1 la-d Daily 8-h maximum O3 concentrations observed at selected national
park sites AX3-16
-------
List of Figures
(cont'd)
Number Page
AX3-12 Seasonal SUM06 and W126 exposure indices for the Ouachita National
Forest for the period of 1991 to 2001 AX3-27
AX3-13 Six-month (April to September) 24-h cumulative W126 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the eastern United States AX3-29
AX3-14 Six-month (April to September) 24-h cumulative SUM06 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the eastern United States AX3-31
AX3-15 Six-month (April to September) 24-h cumulative W126 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the central United States AX3-32
AX3-16 Six-month (April to September) 24-h cumulative SUM06 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the central United States AX3-33
AX3-17 Six-month (April to September) 24-h cumulative W126 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the western United States AX3-34
AX3-18 Six-month (April to September) 24-h cumulative SUM06 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the western United States AX3-35
AX3-19 The 95% confidence interval for the 6-month (April to September)
24-h cumulative W126 exposure index for 2001 for the eastern
United States AX3-36
AX3-20 The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the eastern
United States AX3-36
AX3-21 The 95% confidence interval for the 6-month (April to September)
24-h cumulative N100 exposure index for 2001 for the eastern
United States AX3-37
AX3-22 The 95% confidence interval for the 6-month (April to September)
24-h cumulative W126 exposure index for 2001 for the central
United States AX3-37
-------
List of Figures
(cont'd)
Number Page
AX3-23 The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the central
United States AX3-38
AX3-24 The 95% confidence interval for the 6-month (April to September)
24-h cumulative N100 exposure index for 2001 for the central
United States AX3-38
AX3-25 The 95% confidence interval for the 6-month (April to September)
24-h cumulative W126 exposure index for 2001 for the western
United States AX3-39
AX3-26 The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the western
United States AX3-39
AX3-27 The 95% confidence interval for the 6-month (April to September)
24-h cumulative N100 exposure index for 2001 for the western
United States AX3-40
AX3-28 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Charlotte, NC-Gastonia-Rock Hill,
SC MSA AX3-44
AX3-29 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Baton Rouge, LA MSA AX3-45
AX3-30 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Detroit-Ann Arbor-Flint, MI CMSA AX3-46
AX3-31 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the St. Louis, MO-IL MSA AX3-47
AX3-32 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Phoenix-Mesa, AZ MSA AX3-49
AX3-33 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Fresno, CA MSA AX3-50
AX3-34 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Bakersfield, CA MSA AX3-51
-------
List of Figures
(cont'd)
Number Page
AX3-35 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Los Angeles-Orange County,
CA CMSA AX3-52
AX3-36 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Riverside-Orange County, CA CMSA AX3-54
AX3-37 Vertical profile of O3 obtained over low vegetation AX3-58
AX3-38 Vertical profile of O3 obtained in a spruce forest AX3-59
AX3-39a-d Seven- and 12-h seasonal means at (a) Whiteface Mountain and
(b) Shenandoah National Park for May to September 1987, and
integrated exposures at (c) Whiteface Mountain and (d) Shenandoah
National Park for May to September 1987 AX3-64
AX3-40a-e Cumulative exposures for three non-Mountain Cloud Chemistry
Program Shenandoah National Park sites, 1983 to 1987 AX3-65
AX3-41 Composite, nationwide diurnal variability in hourly averaged O3 in
urban areas AX3-70
AX3-42 Composite, nationwide diurnal variability in 8 hour average O3 in
urban areas AX3-71
AX3-43a-f Diurnal variability in hourly averaged O3 in selected urban areas AX3-72
AX3-43g-l Diurnal variability in hourly averaged O3 in selected urban areas AX3-73
AX3-44a-f Diurnal variability in 8 hour averaged O3 in selected urban areas AX3-74
AX3-44g-l Diurnal variability in 8 hour averaged O3 in selected urban areas AX3-75
AX3-45a-d Time of occurrence of daily maximum 1-h O3 concentration in four
cities, averaged from April to October, 2000 to 2004 AX3-77
AX3-46a-d Time of occurrence of daily maximum 8-h average O3 concentration in
four cities, averaged from April to October, 2000 to 2004 AX3-78
AX3-47a-d Diurnal variations in hourly averaged O3 on weekdays and weekends
in four cities AX3-80
II-xxiv
-------
List of Figures
(cont'd)
Number Page
AX3-47e-h Diurnal variations in hourly averaged O3 on weekdays and weekends
in four cities AX3-81
AX3-48a-d Diurnal variations in 8-h averaged O3 on weekdays and weekends in
four cities AX3-82
AX3-48e-h Diurnal variations in 8-h averaged O3 on weekdays and weekends in
four cities AX3-83
AX3-49a Diurnal variations in hourly averaged O3 at a site in downtown
Detroit, MI AX3-84
AX3-49b Diurnal variations in hourly averaged O3 at a site downwind of
downtown Detroit AX3-84
AX3-50a Diurnal variations in hourly averaged O3 at a site in downtown
St. Louis, MO AX3-85
AX3-50b Diurnal variations in hourly averaged O3 at a site downwind of
downtown St. Louis AX3-85
AX3-51a Diurnal variations in hourly averaged O3 at a site in
San Bernadino, CA AX3-86
AX3-51b Diurnal variations in hourly averaged O3 at a site in Riverside
County well downwind of sources AX3-87
AX3-52a Diurnal variations in 8-h average O3 at a site in downtown Detroit, MI .... AX3-88
AX3-52b Diurnal variations in 8-h average O3 at a site downwind of downtown
Detroit, MI AX3-88
AX3-53a Diurnal variations in 8-h average ozone at a site in downtown
St. Louis, MO AX3-89
AX3-53b Diurnal variations in 8-h average O3 at a site downwind of downtown
St. Louis, MO AX3-89
AX3-54a Diurnal variations in 8-h average O3 at a site in San Bernadino, CA AX3-90
AX3-54b Diurnal variations in 8-h average O3 at a site in Riverside County
well downwind of sources AX3-90
II-XXV
-------
List of Figures
(cont'd)
Number Page
AX3-55 Composite diurnal variability in hourly O3 concentrations observed
at CASTNET sites AX3-91
AX3-56 Composite diurnal variability in 8-h average O3 concentrations
observed at CASTNET sites AX3-91
AX3-57 The comparison of the seasonal diurnal patterns for urban-influenced
(Jefferson County, KY) and a rural-influenced (Oliver County, ND)
monitoring sites using 2002 hourly data for April-October AX3-93
AX3-58a-d Diurnal behavior of O3 at rural sites in the United States in July AX3-94
AX3-59 Composite diurnal O3 pattern at selected national forest sites in the
United States using 2002 hourly average concentration data AX3-95
AX3-60a,b Composite diurnal pattern at (a) White face Mountain, NY and
(b) the Mountain Cloud Chemistry Program Shenandoah National
Park site for May to September 1987 AX3-96
AX3-61a-h Seasonal variations in O3 concentrations as indicated by the 1-h
maximum in each month at selected sites, 2002 AX3-97
AX3-62a-f Diurnal variability in 1-h average O3 concentrations in EPA's 12 cities .... AX3-99
AX3-62g-l Diurnal variability in 1-h average O3 concentrations in EPA's 12 cities . . . AX3-100
AX3-63a-f Diurnal variability in 8-h average O3 concentrations in EPA's 12 cities . . . AX3-101
AX3-63g-l Diurnal variability in 8-h average O3 concentrations in EPA's 12 cities . . . AX3-102
AX3-64 Year-to-year variability in nationwide mean daily maximum 8-h O3
concentrations AX3-105
AX3-65 Year-to-year variability in nationwide 95th percentile value of the
daily maximum 8-h O3 concentrations AX3-106
AX3-66a-h Year-to-year variability in mean daily maximum 8-h O3 concentrations
at selected national park (NP), national wildlife refuge (NWR), and
national monument (NM) sites AX3-107
II-xxvi
-------
List of Figures
(cont'd)
Number Page
AX3-66i-p Year-to-year variability in mean daily maximum 8-h O3 concentrations
at selected national park (NP), national wildlife refuge (NWR), and
national monument (NM) sites AX3-108
AX3-66q-v Year-to-year variability in mean daily maximum 8-h O3 concentrations
at selected national park (NP), national wildlife refuge (NWR), and
national monument (NM) sites AX3-109
AX3-67a-h Year-to-year variability in 95th percentile of daily maximum 8-h O3
concentrations at selected national park (NP), national wildlife refuge
(NWR), and national monument (NM) sites AX3-110
AX3-67i-p Year-to-year variability in 95th percentile of daily maximum 8-h O3
concentrations at selected national park (NP), national wildlife refuge
(NWR), and national monument (NM) sites AX3-111
AX3-67q-v Year-to-year variability in 95th percentile of daily maximum 8-h O3
concentrations at selected national park (NP), national wildlife refuge
(NWR), and national monument (NM) sites AX3-112
AX3-68a-d Measured O3 (ppbv) versus PAN (pptv) in Tennessee, including
(a) aircraft measurements, and (b, c, and d) suburban sites near
Nashville AX3-117
AX3-69 Measured correlation between benzene and NOy at a measurement site
in Boulder, CO AX3-118
AX3-70 Binned mean PM2 5 concentrations versus binned mean O3
concentrations observed at Fort Meade, MD from July 1999 to
July 2001 AX3-121
AX3-71 The co-occurrence pattern for O3 and NO2 AX3-124
AX3-72 The co-occurrence pattern for O3 and NO2 using 2001 data from
the AQS AX3-125
AX3-73 The co-occurrence pattern for O3 and SO2 AX3-126
AX3-74 The co-occurrence pattern for O3 and SO2 using 2001 data from AQS .... AX3-127
AX3-75 The co-occurrence pattern for O3 and PM25 using 2001 data from AQS . . . AX3-128
II-xxvii
-------
List of Figures
(cont'd)
Number Page
AX3-76a Monthly maximum hourly average O3 concentrations at Yellowstone
National Park, Wyoming in 1998, 1999, 2000, and 2001 AX3-133
AX3-76b Hourly average O3 concentrations at Yellowstone National Park,
Wyoming for the period January to December 2001 AX3-133
AX3-77 (a) Contour plot of CO mixing ratios (ppbv) observed in and near the
June 15, 1985, mesoscale convective complex in eastern Oklahoma AX3-135
AX3-78 Maximum hourly average O3 concentrations at rural monitoring sites in
Canada and the United States in February from 1980 to 1998 AX3-143
AX3-79 Schematic diagram of a meteorological mechanism involved in high
concentrations of O3 found in spring in the lower troposphere off the
American East Coast AX3-145
AX3-80 Ozone vertical profile at Boulder, Colorado on May 6, 1999 at 1802
UTC (1102 LST) AX3-148
AX3-81 CASTNet stations in the continental United States for 2001 AX3-152
AX3-82 Monthly mean afternoon (1300 to 1700 hours LT) concentrations
(ppbv) in surface air averaged over the CASTNet stations
(Figure AX3-81) in each U.S. quadrant for March to October 2001 AX3-153
AX3-83 Probability distributions of daily mean afternoon (1300 to 1700 LT)
O3 concentrations in surface air for March through October 2001 at
U.S. CASTNet sites (Figure AX3-83): observations (thick solid line)
are compared with model results (thin solid line) AX3-157
AX3-84 Daily mean afternoon (13 to 17 LT) O3 concentrations in surface air
at Voyageurs National Park (NP), Minnesota in mid-May through
June of 2001 AX3-157
AX3-85 Same as Figure AX3-84 but for Yellowstone National Park, Wyoming
in March to May 2001 AX3-159
AX3-86 Same as Figure AX3-84 but for March of 2001 at selected western
(left column) and southeastern (right column) sites AX3-160
AX3-87a Detailed diagram illustrating components of an exposure model AX3-170
-------
List of Figures
(cont'd)
Number Page
AX3-87b Detailed diagram illustrating components of an exposure model AX3-171
AX3-88 Measured outdoor O3 concentrations (thin line) and modeled indoor
concentrations (bold line) AX3-185
AX3-89 Air exchange rates and outdoor and indoor O3 concentrations during
the summer at a telephone switching station in Burbank, CA AX3-200
AX3-90 Air exchange rates and outdoor and indoor O3 concentrations during
the fall at a telephone switching station in Burbank, CA AX3-201
AX3-91 Diurnal variation of indoor and outdoor O3 and PAN concentrations
measured in a private residence, Freising, Germany, August 11-12, 1995 . . AX3-203
AX3-92 Indoor and outdoor O3 concentration in moving cars AX3-205
AX3-93 Indoor/outdoor concentration ratios for PAN at 10 southern
California museums AX3-206
AX3-94 Ozone decay processes versus time measured for several indoor rooms . . . AX3-212
AX6-1 FEV] decrements as a function of O3 concentration following a 2 h
exposure with incremental exercise (15 min intervals) or rest AX6-8
AX6-2 Average FEV] decrements (±SE) for prolonged 6.6 h exposures to
0.12 ppm O3 as a function of exercise VE AX6-15
AX6-3 The forced expiratory volume in 1 s (FEV]) is shown in relation to
exposure duration (hours) under three exposure conditions AX6-18
AX6-4a,b Recovery of spirometric responses following a 2 h exposure to 0.4 ppm
O3 with IE AX6-23
AX6-5 Plot of the mean FEV] (% baseline) vs. time for ozone exposed cohorts . . . AX6-26
AX6-6 Frequency distributions of FEV] decrements following 6.6-h exposures
to O3 or filtered air AX6-42
AX6-7 Effect of O3 exposure (0.42 ppm for 1.5 h with IE) on FEV] as a function
of subject age AX6-50
II-xxix
-------
List of Figures
(cont'd)
Number Page
AX6-8 Regression curves were fitted to day-by-day postexposure FEV] values
obtained after repeated daily acute exposures to O3 for 2 to 3 h with
intermittent exercise at a VE of 24 to 43 L/min (adaptation studies) AX6-81
AX7-1 Density curves of the O3-associated excess risk of cardiovascular
mortality in the warm season per standardized increment AX7-127
AX7-2 Fitted derivative curve of the Gompertz model to the regression
coefficients (P = log relative risk per 1,000 ppb) for the association
between mortality and acute exposure to O3 plotted against community-
specific long-term average 24-h avg O3 concentrations using data from
the U.S. 95 communities study AX7-131
II-XXX
-------
Authors, Contributors, and Reviewers
CHAPTER 2 ANNEX (A TMOSPHERIC PHYSICS/CHEMISTR Y)
Principal Author
Dr. Joseph Pinto—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Russell Dickerson—Department of Atmospheric and Oceanic Sciences, University of
Maryland, College Park, MD
Contributing Authors
Dr. Brooke Hemming—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Daniel Jacob—Department of Earth and Planetary Sciences, Harvard University,
Cambridge, MA
Dr. William Keene—Department of Environmental Sciences, University of Virginia,
Charlottesville, VA
Dr. Tadeusz Kleindienst—National Exposure Research Laboratory, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Jennie Moody—Department of Environmental Sciences, University of Virginia,
Charlottesville, VA
Mr. Charles Piety—Department of Atmospheric and Oceanic Sciences, University of Maryland,
College Park, MD
Dr. Sandy Sillman—Department of Atmospheric, Oceanic, and Space Sciences, University of
Michigan, Ann Arbor, MI
Dr. Jeffrey Stehr—Department of Atmospheric and Oceanic Sciences, University of Maryland,
College Park, MD
Dr. Bret Taubman—Department of Atmospheric Sciences, Pennsylvania State University,
State College, PA
II-xxxi
-------
Authors, Contributors, and Reviewers
(cont'd)
Contributors and Reviewers
Dr. Christoph Bruhl—Max Planck Institute for Atmospheric Chemistry, Mainz, Germany
Dr. Mohammed Elshahawy—Department of Meteorology and Astronomy, Cairo University,
Giza, Egypt.
Dr. Arlene Fiore—National Oceanic and Atmospheric Administration/Geophysical Fluid Dynamics
Laboratory, Princeton, NJ
Mr. Chris Geron—National Risk Management Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC
Dr. David Golden—Department of Chemistry, Stanford University, Palo Alto, CA
Dr. John Merrill—Graduate School of Oceanography, University of Rhode Island, Kingston, RI
Dr. Sam Oltmans—National Oceanic and Atmospheric Administration/Climate Monitoring and
Diagnostic Laboratory, Boulder, CO
Dr. David Parrish—National Oceanic and Atmospheric Administration/Aeronomy Laboratory,
Boulder, CO
Dr. Perry Samson—Department of Atmospheric, Ocean, and Space Sciences, University of
Michigan, Ann Arbor, MI
Dr. Sandy Sillman—Department of Atmospheric, Ocean, and Space Sciences, University of
Michigan, Ann Arbor, MI
Dr. Melvin Shapiro—National Center for Atmospheric Research, Boulder, CO
CHAPTER 3 ANNEX (AIR QUALITY AND EXPOSURE)
Principal Authors
Ms. Beverly Comfort—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Joseph Pinto—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
II-xxxii
-------
Authors, Contributors, and Reviewers
(cont'd)
Principal Authors
(cont'd)
Dr. Arlene Fiore—National Oceanic and Atmospheric Administration/Geophysical Fluid Dynamics
Laboratory, Princeton, NJ
Dr. Daniel Jacob—Department of Earth and Planetary Sciences, Harvard University,
Cambridge, MA
Dr. Alan S. Lefohn—ASL &Associates, Helena, MT
Dr. Clifford Weisel—Environmental and Occupational Health Sciences Institute, Rutgers
University, New Brunswick, NJ
Contributing Authors
Dr. Jee Young Kim National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Dennis Kotchmar—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Timothy Lewis—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Mr. Thomas McCurdy—National Exposure Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC
Contributors and Reviewers
Dr. Christoph Bruehl—Max Planck Institute for Atmospheric Chemistry, Mainz, Germany
Dr. Russell Dickerson—Department of Atmospheric and Oceanic Sciences, University of
Maryland, College Park, MD
Dr. Judith Graham—American Chemistry Council, Washington, D.C.
Dr. Laszlo Horvath—Hungarian Meteorological Service, Budapest, Hungary
Dr. Ted Johnson—TRJ Associates, Durham, NC
-------
Authors, Contributors, and Reviewers
(cont'd)
Contributors and Reviewers
(cont'd)
Dr. John Merrill—Graduate School of Oceanography, University of Rhode Island, Kingston, RI
Dr. Jennie Moody—Department of Environmental Sciences, University of Virginia,
Charlottesville, VA
Dr. Sam Oltmans—National Oceanic and Atmospheric Administration/Climate Monitoring and
Diagnostic Laboratory, Boulder, CO
Dr. Michiel G.M. Roemer—TNO, The Netherlands
Dr. Sandy Sillman—Department of Atmospheric, Ocean, and Space Sciences, University of
Michigan, Ann Arbor, MI
Dr. Tamas Weidinger—Department of Meteorology, University of Budapest, Budapest, Hungary
CHAPTER 4 ANNEX (DOSIMETR Y)
Principal Authors
Dr. John Overton—U.S. Environmental Protection Agency, National Health and Environmental
Effects Research Laboratory-Research Triangle Park, NC 27711 (retired)
Dr. James S. Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Lori White—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Contributors and Reviewers
Dr. Gary Hatch—U.S. Environmental Protection Agency, National Health and Environmental
Effects Research Laboratory, NC
II-xxxiv
-------
Authors, Contributors, and Reviewers
(cont'd)
CHAPTER 5 ANNEX (ANIMAL TOXICOLOGY)
Principal Authors
Dr. Lori White—National Center for Environmental Assessment (B243-01), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711
Mr. James Raub—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711 (Retired)
Dr. Deepak Bhalla—Department of Occupational and Environmental Health Sciences,
Wayne State University, Detroit, MI
Dr. Carroll Cross—School of Medicine, University of California, Davis, CA
Dr. Mitch Cohen—NYU School of Medicine, New York University, New York, NY
Contributors and Reviewers
Dr. Steven Kleeberger—National Institute of Environmental Health Sciences, Research
Triangle Park, NC 27711
Dr. George Liekauf—Department of Environmental Health, University of Cincinnati,
Cincinnati, OH
Dr. David Basset—Department of Occupational and Environmental Health Sciences, Wayne State
University, Detroit, MI
Dr. E.M. Postlethwait—Department of Environmental Health Sciences, University of Texas
Medical Branch, Galveston, TX
Dr. Kent Pinkerton—Center for Health and the Environment, University of California, Davis, CA
Dr. Edward Schelegle—Department of Anatomy, Physiology, and Cell Biology, University of
California, Davis, CA
Dr. Judith Graham—American Chemical Council, Arlington, VA
Dr. Paul Reinhart—National Center for Environment (B243-03), U.S. Environmental Protection
Agency, Research Triangle Park, NC 27711
II-XXXV
-------
Authors, Contributors, and Reviewers
(cont'd)
CHAPTER 6 ANNEX (CONTROLLED HUMAN EXPOSURE)
Principal Authors
Dr. James S. Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Mr. James Raub—National Center for Environmental Assessment (B243-01), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711 (Retired)
Dr. William C. Adams—Human Performance Laboratory, University of California, Davis, CA
(retired)
Dr. Milan J. Hazucha—Center for Environmental Medicine, Asthma, and Lung Biology,
University of North Carolina, Chapel Hill, NC
Dr. E. William Spannhake—Department of Environmental Health Sciences, Johns Hopkins
University, Baltimore, MD
Contributors and Reviewers
Dr. Edward Avol—Department of Preventive Medicine, University of Southern California,
Los Angeles, CA
Dr. Jane Q. Koenig—Department of Environmental and Occupational Health, University of
Washington, Seattle, WA
Dr. Michael Madden—National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Chapel Hill, NC
Dr.William McDonnell—National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Chapel Hill, NC
II-xxxvi
-------
Authors, Contributors, and Reviewers
(cont'd)
CHAPTER 7 ANNEX (EPIDEMIOLOGY)
Principal Authors
Dr. Dennis Kotchmar—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Jee Young Kim—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. David Svendsgaard—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Kazuhiko Ito—New York University School of Medicine, Nelson Institute of
Environmental Medicine, Tuxedo, NY
Dr. Patrick Kinney—Columbia University, Mailman School of Public Health, New York, NY
Reviewers
Dr. Richard Burnett—Health Canada, Ottawa, Canada
Dr. Vic Hasselblad—Duke University, Durham, NC
Dr. Lucas Neas—National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Chapel Hill, NC
II-xxxvii
-------
U.S. Environmental Protection Agency Project Team
for Development of Air Quality Criteria for Ozone
and Related Photochemical Oxidants
Executive Direction
Dr. Lester D. Grant (Director)—National Center for Environmental Assessment-RTF Division,
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Scientific Staff
Dr. Lori White(Ozone Team Leader)—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Joseph Pinto—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Beverly Comfort—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Brooke Hemming—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. James S. Brown—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Dennis Kotchmar—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Jee Young Kim—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. David Svendsgaard—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Srikanth Nadadur—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. Timothy Lewis—National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Dr. William Hogsett—National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Corvallis, OR
II-xxxviii
-------
U.S. Environmental Protection Agency Project Team
for Development of Air Quality Criteria for Ozone
and Related Photochemical Oxidants
(cont'd)
Scientific Staff
(cont'd)
Dr. Christian Andersen—National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Corvallis, OR
Dr. Jay Garner—National Center for Environmental Assessment (B243-01), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711 (retired)
Mr. Bill Ewald—National Center for Environmental Assessment (B243-01), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711 (retired)
Mr. James Raub—National Center for Environmental Assessment (B243-01), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711 (retired)
Technical Support Staff
Ms. Nancy Broom—Information Technology Manager, National Center for Environmental
Assessment (B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Mr. Douglas B. Fennell—Technical Information Specialist, National Center for Environmental
Assessment (B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Emily R. Lee—Management Analyst, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Diane H. Ray—Program Specialist, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
Ms. Donna Wicker—Administrative Officer, National Center for Environmental Assessment
(B243-01), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711 (retired)
Mr. Richard Wilson—Clerk, National Center for Environmental Assessment (B243-01),
U.S. Environmental Protection Agency, Research Triangle Park, NC 27711
II-xxxix
-------
U.S. Environmental Protection Agency Project Team
for Development of Air Quality Criteria for Ozone
and Related Photochemical Oxidants
(cont'd)
Document Production Staff
Ms. Carolyn T. Perry—Manager, Computer Sciences Corporation, 2803 Slater Road, Suite 220,
Morrisville, NC 27560
Mr. John A. Bennett—Technical Information Specialist, Library Associates of Maryland,
11820 Parklawn Drive, Suite 400, Rockville, MD 20852
Mr. William Ellis—Records Management Technician, InfoPro, Inc., 8200 Greensboro Drive,
Suite 1450, McLean, VA 22102
Ms. Sandra L. Hughey—Technical Information Specialist, Library Associates of Maryland,
11820 Parklawn Drive, Suite 400, Rockville, MD 20852
Mr. Matthew Kirk—Graphic Artist, Computer Sciences Corporation, 2803 Slater Road, Suite 220,
Morrisville, NC 27560
Dr. Barbara Liljequist—Technical Editor, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560
Ms. Rosemary Procko—Senior Word Processor, TekSystems, 1201 Edwards Mill Road, Suite 201,
Raleigh, NC 27607
Ms. Faye Silliman—Publication/Graphics Specialist, InfoPro, Inc., 8200 Greensboro Drive, Suite
1450, McLean, VA 22102
Mr. Carlton Witherspoon—Graphic Artist, Computer Sciences Corporation, 2803 Slater Road,
Suite 220, Morrisville, NC 27560
II-xl
-------
U.S. Environmental Protection Agency Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
Ozone Review Panel
Chair
Dr. Rogene Henderson*, Scientist Emeritus, Lovelace Respiratory Research Institute, 2425
Ridgecrest Drive SE, Albuquerque, NM, 87108, Phone: 505-348-9464, Fax: 505-348-8541,
(rhenders@lrri.org) (FedEx: Dr. Rogene Henderson, Lovelace Respiratory Research Institute,
2425 Ridgecrest Drive SE, Albuquerque, NM, 87108, Phone: 505-348-9464)
Members
Dr. John Balmes, Professor, Department of Medicine, University of California San Francisco,
University of California - San Francisco, San Francisco, California, 94143, Phone: 415-206-8953,
Fax: 415-206-8949, (jbalmes@itsa.ucsf.edu)
Dr. Ellis Cowling*, University Distinguished Professor-at-Large, North Carolina State University,
Colleges of Natural Resources and Agriculture and Life Sciences, North Carolina State University,
1509 Varsity Drive, Raleigh, NC, 27695-7632, Phone: 919-515-7564 , Fax: 919-515-1700,
(ellis_cowling@ncsu.edu)
Dr. James D. Crapo*, Professor, Department of Medicine, National Jewish Medical and Research
Center. 1400 Jackson Street, Denver, CO, 80206, Phone: 303-398-1436, Fax: 303- 270-2243,
(crapoj @nj c. org)
Dr. William (Jim) Gauderman, Associate Professor, Preventive Medicine, University of
Southerm California, 1540 Alcazar #220, Los Angeles, CA, 91016, Phone: 323-442-1567,
Fax: 323-442-2349, (jimg@usc.edu)
Dr. Henry Gong, Professor of Medicine and Preventive Medicine, Medicine and Preventive
Medicine, Keck School of Medicine, University of Southern California, Environmental Health
Service, MSB 51, Rancho Los Amigos NRC, 7601 East Imperial Highway, Downey, CA, 90242,
Phone: 562-401-7561, Fax: 562-803-6883, (hgong@ladhs.org)
Dr. Paul J. Hanson, Senior Research and Development Scientist, Environmental Sciences Division,
Oak Ridge National Laboratory (ORNL), Bethel Valley Road, Building 1062, Oak Ridge, TN,
37831-6422, Phone: 865-574-5361, Fax: 865-576-9939, (hansonpz@comcast.net)
Dr. JackHarkema, Professor, Department of Pathobiology, College of Veterinary Medicine,
Michigan State University, 212 Food Safety & Toxicology Center, East Lansing, MI, 48824,
Phone: 517-353-8627, Fax: 517-353-9902, (harkemaj@msu.edu)
-------
U.S. Environmental Protection Agency Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
Ozone Review Panel
(cont'd)
Members
(cont'd)
Dr. Philip Hopke, Bayard D. Clarkson Distinguished Professor, Department of Chemical
Engineering, Clarkson University, Box 5708, Potsdam, NY, 13699-5708, Phone: 315-268-3861,
Fax: 315-268-4410, (hopkepk@clarkson.edu) (FedEx: 8 Clarkson Avenue, Potsdam, NY
136995708)
Dr. Michael T. Kleinman, Professor, Department of Community & Environmental Medicine, 100
FRF, University of California - Irvine, Irvine, CA, 92697-1825, Phone: 949-824-4765, Fax:
949-824-2070, (mtkleinm@uci.edu)
Dr. Allan Legge, President, Biosphere Solutions, 1601 11th Avenue NW, Calgary, Alberta,
CANADA, T2N 1H1, Phone: 403-282-4479, Fax: 403-282-4479, (allan.legge@shaw.ca)
Dr. Morton Lippmann, Professor, Nelson Institute of Environmental Medicine, New York
University School of Medicine, 57 Old Forge Road, Tuxedo, NY, 10987, Phone: 845-731-3558,
Fax: 845-351-5472, (lippmann@env.med.nyu.edu)
Dr. Frederick J. Miller*, Consultant, 911 Queensferry Road, Cary, NC, 27511, Phone:
919-467-3194, (fjmiller@nc.rr.com)
Dr. Maria Morandi, Assistant Professor of Environmental Science & Occupational Health,
Department of Environmental Sciences, School of Public Health, University of Texas - Houston
Health Science Center, 1200 Herman Pressler Street, Houston, TX, 77030, Phone: 713-500-9288,
Fax: 713-500-9249, (mmorandi@sph.uth.tmc.edu) (FedEx: 1200 Herman Pressler, Suite 624)
Dr. Charles Plopper, Professor, Department of Anatomy, Physiology and Cell Biology, School of
Veterinary Medicine, University of California - Davis, Davis, California, 95616, Phone:
530-752-7065, (cgplopper@ucdavis.edu)
Mr. Richard L. Poirot*, Environmental Analyst, Air Pollution Control Division, Department of
Environmental Conservation, Vermont Agency of Natural Resources, Bldg. 3 South, 103 South
Main Street, Waterbury, VT, 05671-0402, Phone: 802-241-3807, Fax: 802-241-2590,
(rich.poirot@state.vt.us)
Dr. Armistead (Ted) Russell, Georgia Power Distinguished Professor of Environmental
Engineering, Environmental Engineering Group, School of Civil and Environmental Engineering,
Georgia Institute of Technology, 311 Ferst Drive, Room 3310, Atlanta, GA, 30332-0512, Phone:
404-894-3079, Fax: 404-894-8266, (trussell@ce.gatech.edu)
-------
U.S. Environmental Protection Agency Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
Ozone Review Panel
(cont'd)
Members
(cont'd)
Dr. Elizabeth A. (Lianne) Sheppard, Research Associate Professor, Biostatistics and Environmental
& Occupational Health Sciences, Public Health and Community Medicine, University of
Washington, Box 357232, Seattle, WA, 98195-7232, Phone: 206-616-2722, Fax: 206 616-2724,
(sheppard@u. washington.edu)
Dr. Frank Speizer*, Edward Kass Professor of Medicine, Channing Laboratory, Harvard Medical
School, 181 Longwood Avenue, Boston, MA, 02115-5804, Phone: 617-525-2275, Fax:
617-525-2066, (frank.speizer@channing.harvard.edu)
Dr. James Ultman, Professor, Chemical Engineering, Bioengineering program, Pennsylvania State
University, 106 Fenske Lab, University Park, PA, 16802, Phone: 814-863-4802, Fax:
814-865-7846, (jsu@psu.edu)
Dr. Sverre Vedal, Professor of Medicine, Department of Environmental and Occupational Health
Sciences, School of Public Health and Community Medicine, University of Washington, 4225
Roosevelt Way NE, Suite 100, Seattle, WA, 98105-6099, Phone: 206-616-8285, Fax:
206-685-4696, (svedal@u.washington.edu)
Dr. James (Jim) Zidek, Professor, Statistics, Science, University of British Columbia, 6856
Agriculture Rd., Vancouver, BC, Canada, V6T 1Z2, Phone: 604-822-4302, Fax: 604-822-6960,
(jim@stat.ubc.ca)
Dr. Barbara Zielinska*, Research Professor , Division of Atmospheric Science, Desert Research
Institute, 2215 Raggio Parkway, Reno, NV, 89512-1095, Phone: 775-674-7066, Fax:
775-674-7008, (barbz@dri.edu)
Science Advisory Board Staff
Mr. Fred Butterfield, CASAC Designated Federal Officer, 1200 Pennsylvania Avenue, N.W.,
Washington, DC, 20460, Phone: 202-343-9994, Fax: 202-233-0643 (butterfield.fred@epa.gov)
(Physical/Courier/FedEx Address: Fred A. Butterfield, III, EPA Science Advisory Board Staff
Office (Mail Code 1400F), Woodies Building, 1025 F Street, N.W., Room 3604, Washington,
DC 20004, Telephone: 202-343-9994)
*Members of the statutory Clean Air Scientific Advisory Committee (CASAC) appointed by
the EPA Administrator
-------
ABBREVIATIONS AND ACRONYMS
AA
AA
al-AT
AC
ACh
ADSS
AED
AER
AEROCE
AF
AGL
AH2
AHCs
AHR
AirPEX
AirQUIS
AIRS
ALI
AM
ANOVA
AP
AP-CIMS
APEX
APHEA
AQCD
AQS
ASC
ASTM
ATLAS
arachidonic acid
ascorbic acid
al-antitrypsin
air conditioning
acetylcholine
aged and diluted sidestream cigarette smoke
aerodynamic diameter
air exchange rate
Atmospheric/Ocean Chemistry Experiment
adsorbed fraction
above ground level
ascorbic acid
aromatic hydrocarbons
airway hyperreactivity
Air Pollution Exposure (model)
Air Quality Information System (model)
Aerometric Information Retrieval System
air-liquid interface
alveolar macrophage
analysis of variance
alkaline phosphatase
Atmospheric Pressure Chemical lonization Mass Spectrometer
Air Pollutants Exposure (model)
Air Pollution on Health: European Approach (study)
Air Quality Criteria Document
Air Quality System
ascorbate
American Society for Testing and Materials
atmospheric model (by Kurucz)
-------
A/V
P
P2-AR
BAL
BALF
BALT
B[a]P
BC
BEIS
BERLIOZ
BF
BHC
BM
BME
BMZ
BP
bpm
BrdU
BS
BSA
BSA
TD
-'-'scatter
c
C3a
CAAA
CADS
CAPMoN
CAPs
CAR
CARB
surface-to-volume ratio
log relative risk
beta-2-adrenergic receptor
bronchoalveolar lavage
bronchoalveolar lavage fluid
bronchus-associated lymphoid tissue
benzo[a]pyrene
black carbon
Biogenic Emission Inventory System
Berlin Ozone Experiment
black female
biogenic hydrocarbons
black male
Bayesian Maxim Eutropy
basement membrane zone
blood pressure
breaths per minute
bromodeoxyuridine
black smoke
body surface area
bovine serum albumin
back scatter
concentration
complement protein fragment
Clean Air Act Amendments of 1990
Cincinnati Activity Diary Study
Canadian Air and Precipitation Monitoring Network
concentrated ambient particles
centriacinar region
California Air Resources Board
II-xlv
-------
CASTNet, CASTNET
CAT
CB
CBL
CBU
% c/c
CC16
CCh
CCSP
Cdyn
CE
CEPEX
CFD
CFR
CG
CH3
CH4
C2H4
C5H8
C6H6
CHAD
CH2=C(CH3)-CHO
CH3-CC13
C6H5CH3
CH3-CHO
CH3CH(ONO2)CHO
CHC13
CH3CN
CH3-CO
CH3-C(O)-CH=CH2
Clean Air Status and Trends Network
catalase
carbon black
convective boundary layer
cumulative breath units
percent carbon of total carbon
Clara cell secretory protein
carbachol
Clara cell secretory protein
dynamic lung compliance
continuous exercise
Central Equatorial Pacific Experiment
computational fluid dynamics
Code of Federal Regulations
cloud-to-ground (flash)
ammonia
methane
ethene
isoprene
benzene
Consolidated Human Activities Database
methacrolein
methyl chloroform
toluene
acetaldehyde
2-nitratopropanol
chloroform
acetonitrile
acetyl
methyl vinyl ketone
-------
CH3C(O)CH2ONO2
CH3-C(0)02,
CH3-C(O)OO
CHO
CH3O
CH3O2
CH3OH
CH3-O(O)CH3
CH3OOH
CI
CIMS
CINC
CIU
CL
CMAQ
CMBO
CMD
CMSA
CN
CO
CO2
COD
ConA
COPD
CSA
CTM
CV
CYP
cyt.
A
2-D
1 -nitratopropanone
acetyl peroxy, peroxyacetyl
Chinese hamster ovary (cells)
methoxy
methyl peroxy
methanol
acetone
methyl hydroperoxide
confidence interval
Chemical lonization Mass Spectroscopy
cytokine-induced neutrophil chemoattractant
cumulative inhalation units
chemiluminescence
Community Model for Air Quality
chl oromethy Ibutenone
count mean diameter
consolidated metropolitan statistical area
condensation nuclei
carbon monoxide
carbon dioxide
coefficient of divergence
concanavalin A
chronic obstructive pulmonary disease
consolidated statistical area
chemistry transport model
coefficient of variation
cytochrome P-450
cytochrome
delta; change in a variable
two-dimensional
-------
DA
DA
DD
DHBA
DI
DIAL
DNA
DOAS
DOC
DOE
DOPAC
DPPC
DR
DTPA
EEC
EC
ECG
EC-SOD
EDMAS
EE
eGPx
ELF
EM
ENA-78
eNOS
EOFs
EPA
EPEM
EPR
dopamine
dry airstream
doubling dose
2,3-dihydroxybenzoic acid
dry intrusion
differential absorption lidar (system)
deoxyribonucleic acid
differential optical absorption spectroscopy
dissolved organic carbon
Department of Energy
3,4-dihydroxyphenylacetic acid
dipalmitoylglycero-3-phosphocholine
disulfide reductase
diethylenetriaminepentaacetic acid
exhaled breath condensate (fluid)
elemental carbon
0.05% excess risk in mortality
el ectrocardi ographi c
extracellular superoxide dismutase
Exposure and Dose Modeling and Analysis System
energy expenditure
extracellular glutathione peroxidase
epithelial lining fluid
electron microscopy
epithelial cell-derived neutrophil-activating peptide 78
endothelial nitric oxide synthase
empirical orthogonal functions
U.S. Environmental Protection Agency
Event Probability Exposure Model
electron paramagnetic resonance (spectroscopy)
-------
EPRI Electric Power Research Institute
ER emergency room
ERAQS Eastern Regional Air Quality Study
ESR electron spin resonance (spectroscopy)
ET endotracheal
ETS environmental tobacco smoke
EVR equivalent ventilation rate
F female
f,/ fB frequency of breathing
F344 Fischer 344 (rat)
FA filtered air
FA fractional absorption; absorbed fraction
FAA Federal Aviation Administration
FEF forced expiratory flow
25.75, FEF250/0.75o/0 forced expiratory flow between 25 and 75% of vital capacity
forced expiratory flow after X% vital capacity (e.g., after 25, 50, or
75% vital capacity)
FEVj forced expiratory volume in 1 second
FGFR fibroblast growth factor receptor
FIVC forced inspiratory vital capacity
Fn fibronectin
FP fluticasone propionate
FRC functional residual capacity
FS field stimulation
FTIR Fourier Transform Infrared Spectroscopy
FVC forced vital capacity
GAM General Additive Model
GCE Goddard Cumulus Ensemble (model)
GC-FID gas chromatography-flame ionization detection
GCM general circulation model
GC/MS gas chromatography-mass spectrometry
FEF
-------
GDI
GEE
GEOS-CHEM
GEOS-1 DAS
GM-CSF
GMT
G6PD
GSH
GSHPx, GPx
GSSG
GSTM1
GSTMlnull
H+
3H
HCO
H2CO, HCHO
HDMA
HF
H202
H2SO4
HCs
HHP-C9
5-HIAA
HIST
HLA
HNE
HNO2
HNO3
HNO4
HO
glutathione-disulfidetranshydrogenase
Generalized Estimating Equations
three-dimensional model of atmospheric composition driven by
assimilated Goddard Earth Orbiting System observations
NASA Goddard Earth Orbiting System Data Assimilation System
granulocyte-macrophage colony stimulating factor
Greenwich mean time
glucose-6-phosphatedehydrogenase
glutathione; reduced glutathione
glutathione peroxidase
glutathione disulfide
glutathione ^-transferase |i-l (genotype)
glutathione S-transferase |i-l null (genotype)
hydrogen ion; symbol for acid
radiolabeled hydrogen
formyl
formaldehyde
house dust mite allergen
Howland Forest site
hydrogen peroxide
sulfuric acid
hydrocarbons
1 -hydroxy-1 -hydroperoxynonane
5-hydroxyindolacetic acid
histamine
human leukocyte antigen
4-hydroxynonenal
nitrous acid
nitric acid
pernitric acid
hydroxy
II-l
-------
HO2
H2O2
H3O+
HOCH2OOH
HONO
HOONO
HO2NO2
HOX
HPLC
HR
HRV
5-HT
hv
HVAC
IAS
IBM
1C
1C
ICAM
ICEM
ICS
ID#
IE
IFN
Ig
IL
IN
INF
inh
hydroperoxy
hydrogen peroxide
protonated water
hydroxymethylhydroperoxide
nitrous acid
pernitrous acid
peroxynitric acid
hydrogen oxides
high-pressure liquid chromatography
heart rate
maximum heart rate
heart rate variability
5 -hy droxytryptamine
solar ultraviolet proton
heating, ventilaltion, and air conditioning
interalveolar septum
individual-based model or modeling
inspiratory capacity
intracloud (flash)
intracellular adhesion molecule
Indoor Chemistry and Exposure Model
inhaled steroids
identification number
intermittent exercise
interferon
immunoglobulin (e.g., IgA, IgE, IgG, IgM)
interleukin (e.g., IL-1, IL-6, IL-8)
intranasal
interferon
inhalation
Il-li
-------
iNOS
I/O
ip
IPCC
IPMMI
IQR
ISCCP
IT
IU
iv
J(N02)
j(03)
Ka
KM
KO
KTB
LDH
LFHFR
LFT
LI
LIDAR
LIF
LIS
LLJ
LM
LOESS
LPS
LSI
LT
LT
inducible nitric oxide synthase
indoor-to-outdoor ratio
intraperitoneal
Intergovernmental Panel on Climate Change
International Photolysis Frequency Measurement and
Modeling Intercomparison
interquartile range
International Satellite Cloud Climatology Project
intratracheal
International Units
intravenous
photolysis rate coefficient for O3 to NO2
photolysis rate coefficient for O3 to O(1D)
intrinsic mass transfer coefficient/parameter
particle optical reflectance
knockout
tracheobronchial region overall mass transfer coefficient
lactate dehydrogenase
low frequency/high frequency (ratio)
lower free troposphere
labeling index
Light Detection And Ranging
laser-induced fluorescence
lateral intercellular space
low-level jet
light microscopy
locally estimated smoothing splines
lipopolysaccharide
local standard time
leukotriene (e.g., LTB4, LTC4, LTD4, LTE4)
local time
-------
LTA
LWC
M
M
MAQSIP
MBL
NBTH
MCCP
MCh
MCM
MCP
MEF50
MENTOR
MET
MHC
MI
MIESR
MIP
MLN
MM
MM5
MMAD
MMEF
MMMD
MoOx
MOZAIC
MPAN
mRNA
MS
lymphotoxin-alpha
liquid water content
air molecule
male
Multiscale Air Quality Simulation Platform
marine boundary layer
3-methyl-2-benzothiazolinone acetone azine
Mountain Cloud Chemistry Program
methacholine
master chemical mechanism
monocyte chemotactic protein
maximal expiratory flow at 50% of vital capacity
Modeling Environment for Total Risk Studies
metabolic equivalent of work
major histocompatibility
myocardial infarction
matrix isolation electron spin resonance (spectroscopy)
macrophage inflammatory protein
mediastinal lymph node
Mt. Mitchell site
National Center for Atmospheric Research/Penn State Mesoscale
Model
mass median aerodynamic diameter
maximal midexpiratory flow
mean maximum mixing height depth
molybdenum oxides
Measurement of Ozone and Water Vapor by Airbus
In-Service Aircraft
methacryloylperoxynitrate; peroxy-methacrylic nitric anhydride
messenger ribonucleic acid
mass spectrometry
-------
MS
MSA
MS/MS
MT, Mt
Mtn
MW
n,N
N100
NA
NA, N/A
NAAQS
NADH
NADP
NADPH
NADPH-CR
NAMS/SLAMS
NAPBN
NARE
NASA
NBS
NCICAS
NCLAN
ND
NEM
NESCAUM
NF
NF-KB
NH3
NH4+
Mt. Moosilauke site
metropolitan statistical area
tandem mass spectrometry
metallothionein
mountain
molecular weight
number
number of hours >0.10 ppm
noradrenaline
not available
National Ambient Air Quality Standards
reduced nicotinamide adenine dinucleotide
National Atmospheric Deposition Program
reduced nicotinamide adenine dinucleotide phosphate
reduced nicotinamide adenine dinucleotide phosphate-
cytochrome c reductase
National Ambient Monitoring Stations and State and Local Air
Monitoring Stations
National Air Pollution Background Network
North Atlantic Regional Experiment
National Aeronautics and Space Administration
National Bureau of Standards
National Cooperative Inner-City Asthma Study
National Crop Loss Assessment Network
not detectable
National Ambient Air Quality Standards Exposure Model
Northeast States for Coordinated Air Use Management
national forest
nuclear factor kappa B
ammonia
ammonium ion
Il-liv
-------
NHAPS
NHBE
(NH4)2S04
NIH
NIST
NK
NL
NLF
NM
NMHCs
NMMAPS
NMOCs
NMVOCs
NNK
nNOS
NO
NO2
NO3
NO3
N205
NOAA
NOAEL
NOS
NOX
NOy
NOZ
NP
NQO1
NQOlwt
NR
National Human Activity Pattern Survey
cultured human bronchial epithelial (cells)
ammonium sulfate
National Institutes of Health
National Institute of Standards and Technology
natural killer (cells)
nasal lavage
nasal lavage fluid
national monument
nonmethane hydrocarbons
National Morbidity, Mortality and Air Pollution Study
nonmethane organic compounds
nonmethane volatile organic compounds
4-(N-nitrosomethylamino)-1 -(3 -pyridyl)-1 -butanone
neuronal nitric oxide synthase
nitric oxide
nitrogen dioxide
nitrate (radical)
nitrate (ion)
dinitrogen pentoxide
National Oceanic and Atmospheric Administration
non-observable-adverse-effect level
nitric oxide synthase
nitrogen oxides
total reactive nitrogen; sum of NOX and NOZ; odd nitrogen species
nitrogen-containing species, the sum of the products of the
oxidation of NOX
national park
NAD(P)H-quinone oxidoreductase (genotype)
NAD(P)H-quinone oxidoreductase wild type (genotype)
not reported
II-lv
-------
NRC
NS
NS
NS
NSAID
NSBR
NTE
NTRMs
NWR
NZW
02
02
03
18Q
oc
OAQPS
OEMs
OCD)
OH
8-OHdG
OLS
0(3P)
OPE
OVA
Ox
P
"90
PAF
PAHs
PAMS, PAMs
National Research Council
national seashore
nonsignificant
nonsmoker
non-steroidal anti-inflammatory agent
nonspecific bronchial responsiveness
nasal trubinate epithelial (cells)
NIST Traceable Reference Materials
national wildlife refuge
New Zealand white (rabbit)
molecular oxygen
superoxide
ozone
radiolabeled oxygen atom
organic carbon
Office of Air Quality Planning and Standards
observationally based methods
electronically excited oxygen atom
hydroxy
8-hydroxy-2'-deoxyguanosine
ordinary least squares
ground-state oxygen atom
ozone production efficiency
ovalbumin
odd oxygen species; total oxidants
probability
values of the 90th percentile absolute difference in concentrations
platelet-activating factor
polycyclic aromatic hydrocarbons
Photochemical Aerometric Monitoring System
-------
PAN
Pa02
PAR
/7-ATP
PEL
PEL
PBM
PEN
PBPK
PC20
PC
50
PCA
PC-ALF
pCNEM
pCO2
PD
PD100SRaw
PE
PEF
PEF0.75
PEFR
PEG-CAT
PEG-SOD
PEM
PG
6PGD
PGHS-2
peroxyacetyl nitrate
partial pressure of arterial oxygen
proximal alveolar region
/>ara-acetamidophenol
peripheral blood lymphocytes
planetary boundary layer
population-based model or modeling
C-phenyl N-tert-butyl nitrone
physiologically based pharmacokinetic (approach)
provocative concentration that produces a 20% decrease in forced
expiratory volume in 1 second
provocative concentration that produces a 50% decrease in forced
expiratory volume in 1 second
principal component analysis
1 -palmitoyl-2-(9-oxonononoyl)-5«-glycero-3 -phosphocholine
Canadian version of National Ambient Air Quality Standards
Exposure Model
partial pressure of carbon dioxide
provovative dose that produces a 20% decrease in FEVj
provovative concentration that produces a 20% decrease in FEVj
provocative dose that produces a 100% increase in SRaw
postexposure
peak expiratory flow
peak expiratory flow in 0.75 second
peak expiratory flow rate
polyethylene glycol-catalase
polyethylene glycol-superoxide dismutase
personal exposure monitor
prostaglandin (e.g., PGD2, PGE, PGEj, PGE2 PGFla, PGF2(X)
6-phosphogluconate dehydrogenase
prostaglandin endoperoxide G/H synthase 2
-------
PHA
PIF
PK
PM
PM
2.5
PM
10
PM
10-2.5
PM15
PM-CAMx
PMNs
PMT
PND
pNEM
PNN,
50
POC
polyADPR
ppb
ppbv
pphm
ppm
ppmv
PPN
PPPs
pptv
PRB
phytohemagglutinin
peak inspiratory flow
pharmacokinetics
particulate matter
pressure at mouth at 0. 1 second of inspiration against a transiently
occluded mouthpiece (an index of inspiratory drive)
fine particulate matter (mass median aerodynamic diameter
<2.5
combination of coarse and fine particulate matter (mass median
aerodynamic diameter <10 jim)
coarse particulate matter (mass median aerodynamic diameter
between 10 and 2.5 jim)
combination of coarse and fine particulate matter (mass median
aerodynamic diameter <15 jim)
Particulate Matter Comprehensive Air Quality Model with
Extensions
polymorphonuclear leukocytes; neutrophils
photomultiplier tube
postnatal day
Probabilistic National Ambient Air Quality Standard Exposure
Model
proportion of adjacent N-N intervals differing by more than 50 ms
particulate organic carbon
poly(adenosinediphosphate-ribose)
parts per billion
parts per billion by volume
parts per hundred million
parts per million
parts per million by volume
peroxypropionyl nitrate
power plant plumes
parts per trillion by volume
policy relevant background
-------
PTR-MS
PU, PUL
PUFAs
PV
PVOCs
PWM
R
r
R2
9
r
RACM
RADM
rALP
RAMS
RANTES
Raw, Raw
RB
RBC
RDBMS
REHEX
RER
RH
RIOPA
RL
RMR
rMSSD
RO2
ROI
RONO2
ROONO2, RO2NO2
proton-transfer-reaction mass spectroscopy
pulmonary
polyunsaturated fatty acids
potential vorticity
photochemical volatile organic compounds
pokeweed mitogen
intraclass correlation coefficient
correlation coefficient
multiple correlation coefficient
correlation coefficient
Regional Air Chemistry Mechanism
Regional Acid Deposition Model
recombinant antileukoprotease
Regional Atmospheric Modeling System
regulated on activation, normal T cell-expressed and -secreted (cells)
airway resistance
respiratory bronchiole
red blood cell
Relational Database Management Systems
Regional Human Exposure Model
rough endoplasmic reticulum
relative humidity
Relationship of Indoor, Outdoor, and Personal Air (study)
total pulmonary resistance
resting metabolic rate
square root of the mean of the squared difference between adjacent
normal R-R intervals
organic peroxy
reactive oxygen intermediate/superoxide anion
organic nitrate
peroxy nitrate
Il-lix
-------
ROS
RR
RRMS
RT
RT
°g
S
SAC
SAI
sao2
SAPRC
SAROAD
SAWgrp
sc
sc
SCAQS
SD
SD
SDNN
SE
SEM
SES
SGaw
SH
SHEDS
sICAM
SO2
S042
reactive oxygen species
relative risk
relatively remote monitoring sites
respiratory tract
transepithelial resistance
sigma-g; geometric standard deviation
smoker
Staphylococcus aureus Cowan 1 strain
Systems Applications International
oxygen saturation of arterial blood
Statewide Air Pollution Research Center, University of
California, Riverside
Storage and Retrieval of Aerometric Data (U.S. Environmental
Protection Agency centralized database; superseded by Aerometric
Information Retrieval System [AIRS])
small airway function group
stratum corneum
subcutaneous
Southern California Air Quality Study
Sprague-Dawley
standard deviation
standard deviation of normal R-R intervals
standard error
standard error of the mean
socioeconomic status
specific airway conductance
Shenandoah National Park site
Simulation of Human Exposure and Dose System
soluble intracellular adhesion molecule
sulfur dioxide
sulfate
II-lx
-------
SOA
SOD
SOS
SOX
SP
SP
SRaw, sRaw
SRBC
SRM
STE
STEP
STPD
STRF
SUM06
SUM07
SUM08
SURE
T
T
t
TAR
TB
TB
TEA
TEARS
99mTc-DTPA
T
ico
TOLAS
secondary organic aerosol
superoxide dismutase
Southern Oxidant Study
sulfur oxides
substance P
surfactant protein (e.g., SP-A, SP-D)
specific airway resistance
sheep red blood cell
standard reference material
stratospheric-tropospheric exchange
Stratospheric-Tropospheric-Exchange Proj ect
standard temperature and pressure, dry
Spatio-Temporal Random Field
seasonal sum of all hourly average concentrations >0.06 ppm
seasonal sum of all hourly average concentrations >0.07 ppm
seasonal sum of all hourly average concentrations >0.08 ppm
Sulfate Regional Experiment Program
tau; atmospheric lifetime
time (duration of exposure)
t-test statistical value; t statistic
triiodothyronine
thyroxine
Third Assessment Report
terminal bronchiole
tracheobronchial (region)
thiobarbituric acid
thiobarbituric acid reactive substances
radiolabeled diethylenetriaminepentaacetic acid
core temperature
tunable-diode laser absorption spectroscopy
-------
TEM
Tg
TLC
TLR
TNF
TOMS
TOPSE
TRIM
TRIM.EXPO
TPLIF
TSH
TSP
TTFMS
TVA
TWA
TX
UA
UAM
ULLI
URT
UT
UTC
UV
UV-A
UV-B
UV-DIAL
VC
VCAM
expiratory time
transmission electron microscopy
teragram
transforming growth factor-beta 1
inspiratory time
total lung capacity
Toll-like receptor
tumor necrosis factor
Total Ozone Mapping Satellite; total ozone mapping spectrometer
Tropospheric Ozone Production About the Spring Equinox
Total Risk Integrated Methodology (model)
Total Risk Integrated Methodology Exposure Event (model)
two-photon laser-induced fluorescence
thyroid-stimulating hormone
total suspended particulate
two-tone frequency-modulated spectroscopy
Tennessee Valley Authority
time-weighted average
thromboxane (e.g., TXA2, TXB2)
uric acid
Urban Airshed Model
unit length labeling index
upper respiratory tract
Universal Time
Coordinated Universal Time
ultraviolet
ultraviolet radiation of wavelengths 320 to 400 nm
ultraviolet radiation of wavelengths 280 to 320 nm
Ultraviolet Differential Absorption Lidar
vital capacity
vascular cell adhesion molecule
-------
Emax
max25%
max50%
VD
VD
VE
V;
v,
v,
* max75%
VMD
V02
V02max
VOCs
VT
vra
Tmax
W126
WBGT
WCB
WF
WF, WFM
WM
WT
WT
anatomic dead space
deposition velocity
minute ventilation; expired volume per minute
maximum minute ventilation
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
oxygen consumption
maximal oxygen uptake (maximal aerobic capacity)
volatile organic compounds
tidal volume
tracheobronchial region volume
maximum tidal volume
cumulative integrated exposure index with a sigmoidal
weighting function
wet bulb globe temperature
warm conveyor belt
white female
White Face Mountain site
white male
White Top Mountain site
wild type
-------
ANNEX AX2. PHYSICS AND CHEMISTRY OF OZONE
IN THE ATMOSPHERE
AX2.1 INTRODUCTION
This annex (Annex AX2) provides detailed supporting information for Chapter 2 on the
physics and chemistry of ozone (O3) in the atmosphere. The organization of the material in this
annex follows that used in prior Air Quality Criteria Documents, i.e., material is presented in
sections and subsections. This annex provides material supporting Chapter 2 of the current draft
Air Quality Criteria Document for Ozone.
Section AX2.2 focuses on the chemistry of O3 formation. A very brief overview of
atmospheric structure is presented in Section AX2.2.1. An overview of O3 chemistry is given in
Section AX2.2.2. Information about reactive chemical species that initiate the oxidation of
VOCs is given in Section AX2.2.3. The chemistry of nitrogen oxides is then discussed briefly in
Section AX2.2.4. The oxidation of methane, the simplest hydrocarbon is outlined in
Section AX2.2.5.
The photochemical cycles leading to O3 production are best understood by considering the
oxidation of methane, structurally the simplest VOC. The CH4 oxidation cycle serves as a model
which can be viewed as representing the chemistry of the relatively clean or unpolluted
troposphere (although this is a simplification because vegetation releases large quantities of
complex VOCs, such as isoprene, into the atmosphere). Although the chemistry of the VOCs
emitted from anthropogenic and biogenic sources in polluted urban and rural areas is more
complex, a knowledge of the CH4 oxidation reactions aids in understanding the chemical
processes occurring in the polluted atmosphere because the underlying chemical principles are
the same. The oxidation of more complex hydrocarbons (alkanes, alkenes, and aromatic
compounds) is discussed in Sections AX2.2.6, AX2.2.7, and AX2.2.8, respectively. The
chemistry of oxygenated species is addressed in Section AX2.2.9. Greater emphasis is placed on
the oxidation of aromatic hydrocarbons in this section because of the large amount of new
information available since the last Air Quality Criteria for Ozone document (AQCD 96) was
published (U.S. Environmental Protection Agency, 1996) and because of their importance in O3
formation in polluted areas. Multiphase chemical processes influencing O3 are discussed in
AX2-1
-------
Section AX2.2.10. Meteorological processes that control the formation of O3 and other oxidants
and that govern their transport and dispersion, and the sensitivity of O3 to atmospheric
parameters are given in Section AX2.3. Greater emphasis is placed on those processes for which
a large amount of new information has become available since AQCD 96. The role of
stratospheric-tropospheric exchange in determining O3 in the troposphere is presented in Section
AX2.3.1. The importance of deep convection in redistributing O3 and its precursors and other
oxidants throughout the troposphere is given in Section AX2.3.2. The possible importance of
nocturnal low-level jets in transporting O3 and other pollutants is presented in Section AX2.3.3.
Information about the mechanisms responsible for the intercontinental transport of pollutants and
for the interactions between stratospheric-tropospheric exchange and convection is given in
Section AX2.3.4. Much of the material in this section is based on results of field programs
examining atmospheric chemistry over the North Atlantic ocean. The sensitivity of O3 to solar
ultraviolet radiation and temperature is given in Section AX2.3.5. The relations of O3 to its
precursors and to other oxidants based on field and modeling studies are discussed in Section
AX2.4. Methods used to calculate relations between O3 its precursors and other oxidants are
given in Section AX2.5. Chemistry-transport models are discussed in Section AX2.5.1.
Emissions of O3 precursors are presented in Section AX2.5.2. Issues related to the evaluation of
chemistry-transport models and emissions inventories are presented in Section AX2.5.3.
Measurement methods are summarized in Section AX2.6. Methods used to monitor ground-
level O3 are given in Section AX2.6.1, NO and NO2 in Section AX2.6.2, HNO3 in
Section AX2.6.3 and some important VOCs in Section AX2.6.4.
AX2.2 TROPOSPHERIC OZONE CHEMISTRY
AX2.2.1 Atmospheric Structure
The atmosphere can be divided into several distinct vertical layers, based primarily on the
major mechanism by which that portion of the atmosphere is heated or cooled. The lowest major
layer is the troposphere, which extends from the earth's surface to about 8 km above polar
regions and to about 16 km above tropical regions. The troposphere is heated by convective
transport from the surface, and by the absorption of infrared radiation emitted by the surface,
principally by water vapor and CO2. The planetary boundary layer (PEL) is the sublayer of the
AX2-2
-------
troposphere that mixes with surface air on time scales of a few hours or less. It typically extends
to 1-2 km altitude and is often capped by a temperature inversion. The sublayer of the
troposphere above the PEL is called the free troposphere. Ventilation of the PEL with free
tropospheric air takes place on a time scale of a week. Vertical mixing of the whole troposphere
takes place on a time scale of a month or two. The stratosphere extends from the tropopause, or
the top of the troposphere, to about 50 km in altitude. The upper stratosphere is heated by the
absorption of solar ultraviolet radiation by O3, while dissipation of wave energy transported
upwards from the troposphere is a primary heating mechanism in the lower stratosphere.
Heating of the stratosphere is balanced by radiative cooling due to infrared emissions to space
by CO2, H2O, and O3. As a result of heating of the upper stratosphere, temperatures increase
with height, inhibiting vertical mixing. A schematic overview of the major chemical cycles
involved in O3 formation and destruction in the stratosphere and troposphere is shown in Figure
AX2-1. The figure emphasizes gas phase processes, but the importance of multiphase processes
is becoming apparent. The sequences of reactions shown in the lower right quadrant of the
figure will be discussed in Section AX2.2. The reader is referred to any of the large number of
texts on atmospheric chemistry, such as Wayne (2000) or Seinfeld and Pandis (1998), for an
introduction to stratospheric photochemistry, including the impact of O3-destroying compounds.
AX2.2.2 Overview of Ozone Chemistry
Ozone is found not only in polluted urban atmospheres but throughout the troposphere,
including remote areas of the globe. Even without ground-level production, some O3 would be
found in the troposphere due to downward transport from the stratosphere. Tropospheric
photochemistry leading to the formation of O3 and other photochemical air pollutants is
complex, involving thousands of chemical reactions and thousands of stable and reactive
intermediate products. Other photochemical oxidants, such as peroxyacetyl nitrate (PAN), are
among the reactive products. Ozone can be photolyzed in the presence of water to form
hydroxyl radical (OH), which is responsible for the oxidation of NOX and SOX to form
nitric (HNO3) and sulfuric acid (H2SO4), respectively. Ozone participates directly in the
oxidation of unsaturated hydrocarbons, via the ozonolysis mechanism, yielding secondary
organic compounds that contribute to aerosol formation and mass, as well as formaldehyde
(H2CO) and other carbonyl compounds, such as aldehydes and ketones.
AX2-3
-------
Stratosphere
Non-Pola
Regions
Figure AX2-1. Schematic overview of O3 photochemistry in the stratosphere
and troposphere.
There is a rapid photochemical cycle in the troposphere that involves the photolysis of
nitrogen dioxide (NO2) by solar UV-A radiation to yield nitric oxide (NO) and a ground-state
oxygen atom, O(3P),
NO2 + hv -» NO + O(3P),
(AX2-1)
O(3P) then reacts with molecular oxygen to form O3: A molecule from the surrounding air
collides with the newly-formed O3 molecule, removing excess energy to allow it to stabilize.
AX2-4
-------
O(3P) + O2 + M -» O3 + M, where M = an air molecule. (AX2-2)
Reaction AX2-2 is the only significant reaction forming O3 in the troposphere.
NO and O3 react to reform NO2:
+ 02. (AX2-3)
This reaction is responsible for O3 decreases found near sources of NO (e.g., highways)
especially at night. The oxidation of reactive VOCs leads to the formation of reactive radical
species that allow the conversion of NO to NO2 without the participation of O3 (as in
reaction AX2-3).
HO-,% RO7-
NO —-—> NO7.
(AX2-4)
O3 can, therefore, accumulate as NO2 photolyzes as in reaction AX2-1 followed by reaction
AX2-2.
It is often convenient to speak about families of chemical species that are defined in terms
of members which interconvert rapidly among themselves on time scales that are shorter than
that for formation or destruction of the family as a whole. For example, an "odd oxygen" (Ox)
family can be defined as £ (O(3P) +O(1D)+ O3 + NO2) in much the same way as the NOX
(NO + NO2) family is defined. We can then see that production of Ox occurs by the schematic
reaction AX2-4, and that the sequence of reactions given by reactions AX2-1 through AX2-3
represents no net production of Ox. Definitions of species families and methods for constructing
families are discussed in Jacobson (1999) and references therein. Other families that include
nitrogen containing species, and will be referred to later in this chapter, are NOZ which is the
sum of the products of the oxidation of NOX = £ (HNO3 + PAN (CH3CHO-OO-NO2) + HNO4 +
other organic nitrates + particulate nitrate); and NOy, which is the sum of NOX and NOZ.
AX2-5
-------
AX2.2.3 Initiation of the Oxidation of VOCs
The key reactive species in the troposphere is the OH radical. OH radicals are
responsible for initiating the photochemical oxidation of CO and most anthropogenic and
biogenic VOCs, including those responsible for depleting stratospheric O3 (e.g., CH3Br,
hydroclorofluorocarbons), and those which contribute to the greenhouse effect (e.g., CH4).
Because of their role in removing so many potentially damaging species, OH radicals have
sometimes been referred to as the atmosphere's detergent. In the presence of NO, reactions of
OH with VOCs lead to the formation of O3. In addition to OH radicals, there are several other
atmospheric species such as NO3, Cl, and Br radicals and O3 that are capable of initiating VOC
oxidation. Rate coefficients and estimated atmospheric lifetimes (the e-folding time) for
reactions of a number of alkanes, alkenes and dienes involved in O3 formation with these
oxidants at concentrations characteristic of the relatively unpolluted planetary boundary layer are
given in Table AX2-1. As can be seen from Table AX2-1, there is a wide range of lifetimes
calculated for the different species. However, under certain conditions the relative importance of
these oxidants can change from those shown in the table. For hydrocarbons whose atmospheric
lifetime is much longer than a day, diurnally averaged concentrations of oxidant concentrations
can be used, but for those whose lifetime is much shorter than a day it is more appropriate to use
either daytime or nighttime averages depending on when the oxidant is at highest concentrations.
During these periods, these averages are of the order of twice the values used in Table AX2-1.
The main source of OH radicals is the photolysis of O3 by solar ultraviolet radiation at
wavelengths < 340 nm (solar radiation at wavelengths < 320 nm is also referred to as UV-B) to
generate electronically excited O(1D) atoms (Jet Propulsion Laboratory, 2003),
O3 + hv -> O2 + O(1D). (AX2-5)
The O(JD) atoms can either be deactivated to the ground state O(3P) atom by collisions with N2
and O2, or they react with water vapor to form two OH radicals:
H2O -» 2(-OH) (AX2-6)
AX2-6
-------
Table AX2-1. Comparison of the Atmospheric Lifetimes (u) of Low Molecular Weight Hydrocarbons Due to Reaction with
OH, NO3, Cl, Br, and O3
k, cm3 molecule"1 s"1
OH
Hydrocarbon
Alkanes
Ethane
Propane
2-Methylpropane
w-Butane
2-Methylbutane
x
to w-Pentane
2,2-Dimethylbutane
2,3 -Dimethylbutane
2-Methylpentane
3-Methylpentane
w-Hexane
2,2,4-Trimethylpentane
k x 1012
0.24
1.1
2.1
2.3
4
3.8
2.7
6.4
5.6
5.8
5.2
3.8
T
48 d
lid
5.6 d
5.2 d
2.9 d
3.0 d
4.3d
1.8d
2.1 d
2.0 d
2.2 d
3.0 d
NO3
k xlO12
<1.0xlO~5
0.00021
< 0.00007
0.000046
0.00016
0.000081
NA
0.00041
0.000017
0.00002
0.00011
0.000075
T
>13y
>0.60
>18y
2.8 y
0.79 y
1.6 y
NA
110 d
7.5 y
6.3 y
1.2 y
1.7 y
Cl
k x 1010
0.57
1.3
1.3
2.3
2
2.5
NA
2
2.5
2.5
3.1
2.3
T
6.7 mo
90 d
90 d
50 d
60 d
46 d
NA
60 d
47 d
46d
38 d
50 d
k x 1012
3.1 x 1Q-7
0
<1.0x
<1.0x
NA
NA
NA
0.0064
NA
NA
NA
0.0068
Br
T
i.o x io6y2
6.5 x 103y2
>3.2x 107y2
>3.2 x 107y2
NA
NA
NA
50 y
NA
NA
NA
47 y
03
k x 1018
<0.01
<0.01
<0.01
<0.01
NA
NA
NA
NA
NA
NA
NA
NA
T
>3.2y
>3.2y
>3.2y
>3.2y
NA
NA
NA
NA
NA
NA
NA
NA
-------
Table AX2-1 (cont'd). Comparison of the Atmospheric Lifetimes (u) of Low Molecular Weight Hydrocarbons Due to
Reaction with OH, NO3, Cl, Br and O3
k, cm3 molecule"1 s"1
OH
Hydrocarbon
Alkenes
Ethene
Propene
2-Methylpropene
1-Butene
/ra«s-2-Butene
^ c/s-2-Butene
to
oo 1,3 -Butadiene
Isoprene
2-Methyl-2-butene
1-Pentene
/ra«s-2-Pentene
c/s-2-Pentene
2,4,4-Trimethyl-l-pentene
k x 1012
8.5
26
51
31
64
56
67
100
87
31
67
65
65
T
33 h
llh
5.4 h
9.0 h
4.3 h
5.0 h
4.1 h
2.8 h
3.2 h
9.0 h
4.1 h
4.3 h
4.3 h
NO3
k x 1012
0
0.01
0.34
0.013
0.39
0.35
0.1
0.68
9.4
0.7
1.6
1.4
0.51
T
230 d
4.9 d
3.3h
3.6 d
2.8 h
3.2 h
llh
1.6 h
0.12 h
1.6 h
0.69 h
0.79 h
2.2 h
Cl
k x 1010
0.99
2.3
0.42
1.4
NA
NA
4.2
5.1
NA
NA
NA
NA
NA
T
3.8m
50 d
9.0m
65 d
NA
NA
28 d
23d
NA
NA
NA
NA
NA
Br
k x 1012
0.18
5.3
NA
3.4
0.23
6.3
57
74
19
NA
NA
NA
NA
T
1.8 y
22 d
NA
34 d
1.4 y
18 d
2.0 d
1.6 d
6.1 d
NA
NA
NA
NA
03
k x 1018
1.6
10
11
9.6
190
125
6.3
13
400
11
320
210
NA
T
7.2 d
1.2 d
Lid
1.2 d
1.5 h
2.3 h
1.8 d
21 h
0.69 h
Lid
0.86 h
1.3 h
NA
Notes: NA = Reaction rate coefficient not available. Rate coefficients were calculated at 298k and 1 atmosphere, y = year, d = day.
OH = 1 x lOVcm3; NO3 = 2.5 x 108/cm3; Cl = 1 x 103/cm3; Br = 1 x 105/cm3; O3 = 1 x 1012/cm3. Value for Br calculated based on equilibrium with BrO = 1 ppt.
1 Rate Coefficients were Obtained from the NIST Online Kinetics Database for Reactions of Alkanes and for all Cl and Br Reactions.
All Other Rate Coefficients were Obtained from the Evaluation
ofCalvertetal. (2000).
2 Lifetimes should be regarded as lower limits.
Sources: NIST online kinetics database (http://kinetics.nist.gov/index.php).
-------
The O(3P) atoms formed directly in the photolysis of O3 in the Huggins and Chappuis bands or
formed from deactivation of O(1D) atoms reform O3 through reaction AX2-2. Hydroxyl radicals
produced by reactions AX2-5 and AX2-6 can react further with species such as carbon monoxide
and with many hydrocarbons (for example, CH4) to produce HO2 radicals.
Measurements of OH radical concentrations in the troposphere (Poppe et al., 1995; Eisele
et al., 1997; Brune et al., 1999; Martinez et al., 2003; Ren et al., 2003) show that, as expected,
the OH radical concentrations are highly variable in space and time, with daytime maximum
concentrations of > 3 x io6 molecules /cm3 in urban areas. A global, mass-weighted mean
tropospheric OH radical concentration also can be derived from the estimated emissions and
measured atmospheric concentrations of methylchloroform (CH3CC13) and the rate constant for
the reaction of the OH radical with CH3CC13. Krol et al. (1998) derived a global average OH
concentration of 1.07 x IO6 molecules /cm3 for 1993 along with an upward trend of about
0.5%/yr between 1978 and 1993. Using an integrated data set of observed O3, H2O, NOy, CO,
VOCs, temperature and cloud optical depth, Spivakovsky et al. (2000) calculated a global annual
mean OH concentration of 1.16 x IO6 molecules cm"3, consistent to within 10% of the value
obtained by Krol et al. (1998).
HO2 radicals do not initiate the oxidation of hydrocarbons, but serve to recycle OH mainly
by way of reaction with NO, O3, and itself (the latter produces H2O2, which can photolyze to
yield OH). The HO2 radicals also react with organo-peroxy radicals produced during the
oxidation of VOCs to form organo-peroxides (cf, Section AX2.2.5, reaction AX2-20). Organo-
peroxides undergo wet or dry deposition (Wesely and Hicks, 2000) or degrade further by
photolysis and reaction with OH (Jet Propulsion Laboratory, 2003).
At night, NO3 assumes the role of primary oxidant (Wayne et al., 1991). Although it is
generally less reactive than OH, its high abundance in the polluted atmosphere compensates for
its lower reactivity. For several VOCs, however, including dimethylsulfide, isoprene, some
terpenes (a-pinene, limonene, linalool) and some phenolic compounds (phenol, o-cresol),
oxidation by NO3 at night is competitive with oxidation by OH during the day, making it an
important atmospheric removal mechanism for these compounds (Wayne et al., 1991) (see
Table AX2-1). The role of NO3 radicals in the chemistry of the remote marine boundary layer
has been examined recently by Allen et al., (2000) and in the polluted continental boundary layer
by Geyer and Platt (2002).
AX2-9
-------
Cl atoms, derived from products of multiphase processes can initiate the oxidation of most
of the same VOCs as OH radicals, however, the rate coefficients for the reactions of alkanes with
Cl atoms are usually much higher. Cl will also oxidize alkenes and aromatic compounds, but
with a significantly lower rate constant than for OH reactions. Following the initial reaction
with Cl, the degradation of the hydrocarbon proceeds as with OH and NO3, generating an
enhanced supply of odd hydrogen radicals leading to O3 production in the presence of
sufficient NOX. The corresponding reactions of Br with hydrocarbons proceed in a similar
manner, but with rate coefficients that can be substantially lower or higher.
Chlorine and bromine radicals will also react directly with O3 to form CIO and BrO
radicals, providing a sink for odd oxygen if they do not react with NO to form NO2 (e.g.,
Pszenny et al., 1993). As with other oxidants present in the atmosphere, Cl chemistry provides a
modest net sink for O3 when NOX is less than 20 pptv, and is a net source at higher NOX. Kasting
and Singh (1986) estimated that as much as 25% of the loss of nonmethane hydrocarbons in the
nonurban atmosphere can occur by reaction with Cl atoms, based on the production of Cl atoms
from gas phase photochemical reactions involving chlorine containing molecules (HC1, CH3C1,
CHC13, etc.). Elevated concentrations of atomic Cl and other halogen radicals can be found in
polluted coastal cities where precursors are emitted directly from industrial sources and/or are
produced via acid-catalyzed reactions involving sea-salt particles (Tanaka et al., 2000; Spicer
etal., 2001).
Substantial chlorine-VOC chemistry has been observed in the cities of Houston and
Beaumont/Port Arthur, Texas (Tanaka et al., 2000; Chang et al., 2002; Tanaka et al., 2003).
Industrial production activities in those areas frequently result in large releases of chlorine gas
(Tanaka et al., 2000). Chloromethylbutenone (CMBO), the product of the oxidation of isoprene
by atomic Cl and a unique marker for chlorine radical chemistry in the atmosphere (Nordmeyer
et al., 1997), has been found at significant mixing ratios (up to 145 pptv) in ambient Houston air
(Riemer and Apel, 2001). However, except for situations in which there are strong local sources
such as these, the evidence for the importance of Cl as an oxidizing agent is mixed. Parrish et al.
(1992, 1993) argued that ratios of selected hydrocarbons measured at Pt. Arena, CA were
consistent with loss by reaction with OH radicals and that any deviations could be attributed to
mixing processes. Finlayson-Pitts (1993), on the other hand had suggested that these deviations
could have been the result of Cl reactions. McKeen et al. (1996) suggested that hydrocarbon
AX2-10
-------
ratios measured downwind of anthropogenic source regions affecting the western Pacific Basin
are consistent with loss by reaction with OH radicals only. Rudolph et al. (1997), based on data
for several pairs of hydrocarbons collected during a cruise in the western Mediterranean Sea, the
eastern mid- and North Atlantic Ocean and the North Sea during April and May of 1991, also
found that ratios of hydrocarbons to each other are consistent with their loss given mainly by
reaction with OH radicals without substantial contributions from reactions with Cl. Their best
estimate, for their sampling conditions was a ratio of Cl to OH of about 1CT3, implying a
concentration of Cl of about 103/cm3 using the globally averaged OH concentration of
about lOVcm3 given above. In contrast Wingenter et al. (1996) and Singh et al. (1996a) inferred
significantly higher concentrations of atomic Cl (104 to 10s cm"3) based on relative concentration
changes in VOCs measured over the eastern North Atlantic and Pacific Oceans, respectively.
Similar approaches employed over the high-latitude southern ocean yielded lower estimates of
Cl concentrations (103 cm"3; Wingenter et al., 1999). Taken at face value, these observations
indicate substantial variability in Cl concentrations and uncertainty in "typical" values.
AX2.2.4 Chemistry of Nitrogen Oxides in the Troposphere
In the troposphere, NO, NO2, and O3 are interrelated by the following reactions:
+ U2 (AX2-3)
NO2 + hv -» NO + O(3P) (AX2-1)
0(3P) + 02 + M -» 03 + M (AX2-2)
The reaction of NO2 with O3 leads to the formation of the nitrate (NO3) radical,
N02 + 03 -» N03- + 02, (AX2-7)
which in the lower troposphere is nearly in equilibrium with dinitrogen pentoxide (N2O5):
NO3' + NO2 <-^-» N2 O5. (AX2-8)
AX2-11
-------
However, because the NO3 radical photolyzes rapidly (with a lifetime of «5 s for an overhead
sun [Atkinson et al., 1992]),
NO3' + hv -> NO + O2 (10%) (AX2-9a)
(90%) (AX2-9b)
its concentration remains low during daylight hours, but can increase after sunset to nighttime
concentrations of < 5 x 107 to 1 x 1010 molecules cm"3 (< 2 to 430 ppt) over continental areas
influenced by anthropogenic emissions of NOX (Atkinson et al., 1986). This leads to an increase
of N2O5 concentrations during the night by reaction (AX2-8).
The tropospheric chemical removal processes for NOX involve the reaction of NO2 with the
OH radical and the hydrolysis of N2O5 in aqueous aerosol solutions to produce HNO3.
•OH + N02 -^-^ HN03 (AX2-10)
N2O5 2 w > HNO3 (AX2-11)
The gas-phase reaction of the OH radical with NO2 initiates the major and ultimate removal
process for NOX in the troposphere. This reaction removes radicals (OH and NO2) and competes
with hydrocarbons for OH radicals in areas characterized by high NOX concentrations, such as
urban centers (see Section AX2.4). In addition to gas-phase nitric acid, Golden and Smith
(2000) have concluded that, pernitrous acid (HOONO) is also produced by the reaction of NO2
and OH radicals on the basis of theoretical studies. However, a recent assessment (Jet
Propulsion Laboratory, 2003) has concluded that this channel represents a minor yield
(approximately 15% at the surface). HOONO will thermally decompose or photolyze.
Gas-phase HNO3 formed from reaction AX2-10 undergoes wet and dry deposition to the surface
and uptake by ambient aerosol particles. The tropospheric lifetime of NOX due to reaction
AX2-10 ranges from a few hours to a few days. Geyer and Platt (2002) concluded that reaction
AX2-11 constituted about 10% of the removal of NOX at a site near Berlin, Germany during
AX2-12
-------
spring and summer. However, during winter the relative importance of reaction AX2-11 could
be much higher because of the much lower concentration of OH radicals and the enhanced
stability of N2O5 due to lower temperatures and intensity of sunlight. Note that reaction AX2-11
surely proceeds as a heterogeneous reaction.
OH radicals also can react with NO to produce nitrous acid (HNO2):
•OH + NO M >HN02. (AX2-12)
In the daytime, HNO2 is rapidly photolyzed back to the original reactants:
HNO2 + hv ->• 'OH + NO. (AX2-13)
At night, HNO2 can be formed by heterogeneous reactions of NO2 in aerosols or at the earth's
surface (Lammel and Cape, 1996; Jacob, 2000; Sakamaki et al., 1983; Pitts et al., 1984;
Svensson et al., 1987; Jenkin et al., 1988; Lammel and Perner, 1988; Notholt et al., 1992a,b).
This results in accumulation of HNO2 during nighttime. Modeling studies suggest that
photolysis of this HNO2 following sunrise, could provide an important early-morning source of
OH radicals to drive O3 formation (Harris et al., 1982).
Another important process controlling NOX concentrations is the formation of organic
nitrates. Oxidation of VOCs produces organic peroxy radicals (RO2), as discussed in the
hydrocarbon chemistry subsections to follow. Reaction of these RO2 radicals with NO and NO2
produces organic nitrates (RONO2) and peroxynitrates (RO2NO2):
RO2' + NO M > RONO2 (AX2-14)
RO2' + NO2 M > RO2NO2 (AX2-15)
Reaction (AX2-14) is a minor branch for the reaction of RO2 with NO (the major branch
produces RO and NO2, as discussed in the next section). The organic nitrate yield increases with
carbon number (Atkinson, 2000).
AX2-13
-------
The organic nitrates may react further, depending on the functionality of the R group, but
they will typically not return NOX and can therefore be viewed as a permanent sink for NOX.
This sink is usually small compared to HNO3 formation, but the formation of isoprene nitrates
may be a significant sink for NOX in the United States in summer (Liang et al., 1998).
The peroxynitrates produced by (AX2-15) are thermally unstable and most have very short
lifetimes (less than a few minutes) against thermal decomposition to the original reactants. They
are thus not effective sinks of NOX. Important exceptions are the peroxyacylnitrates (PANs)
arising from the peroxyacyl radicals RC(O)OO produced by oxidation and photolysis of
carbonyl compounds. PANs have lifetimes ranging from ~1 hour at room temperature to several
weeks at 250K. They can thus provide an effective sink of NOX at cold temperatures, but also a
reservoir allowing eventual release of NOX as air masses warm, in particular by subsidence. By
far the most important of these PANs compounds is peroxyacetylnitrate (PAN), with formula
CH3C(O)OONO2. PAN is a significant product in the oxidation of most VOCs. It is now well
established that PAN decomposition provides a major source of NOx in the remote troposphere
(Staudt et al., 2003). PAN decomposition in subsiding Asian air masses over the eastern Pacific
could make an important contribution to O3 enhancement in the U.S. from Asian pollution
(Hudman et al., 2004).
AX2.2.5 The Methane Oxidation Cycle
The photochemical cycles leading to O3 production are best understood by considering the
oxidation of methane, structurally the simplest VOC. The CH4 oxidation cycle serves as a model
which describes the chemistry of the relatively clean or unpolluted troposphere (although this is
a simplification because vegetation releases large quantities of complex VOCs into the
atmosphere). Although the chemistry of the VOCs emitted from anthropogenic and biogenic
sources in polluted urban and rural areas is more complex, a knowledge of the CH4 oxidation
reactions aids in understanding the chemical processes occurring in the polluted atmosphere
because the underlying chemical principles are the same.
Methane is emitted into the atmosphere as the result of anaerobic microbial activity in
wetlands, rice paddies, the guts of ruminants, landfills, and from mining and combustion of
fossil fuels (Intergovernmental Panel on Climate Change, 2001). The major tropospheric
removal process for CH4 is by reaction with the OH radical,
AX2-14
-------
•OH + CH4 -» H2O + CH3. (AX2-16)
In the troposphere, the methyl radical reacts solely with O2 to yield the methyl peroxy (CH3O2-)
radical (Atkinson et al., 1992):
CH3 + O2 M > CH3O2-. (AX2-17)
In the troposphere, the methyl peroxy radical can react with NO, NO2, HO2 radicals, and
other organic peroxy (RO^ radicals, with the reactions with NO and HO2 radicals being the most
important (see, for example, World Meteorological Organization, 1990). The reaction with NO
leads to the formation of the methoxy (CH36) radical,
CH3O2- + NO -» CH36 + NO2. (AX2-18)
The reaction with the HO2 radical leads to the formation of methyl hydroperoxide
(CH3OOH),
CH302- + H02- -» CH3OOH + O2, (AX2-19)
which can photolyze or react with the OH radical (Atkinson et al., 1992):
CH3OOH + hv -» CH36 + -OH. (AX2-20)
•OH + CH3OOH -> H2O + CH3O2- (AX2-21 a)
or
-^ H2O + CH2OOH fast decomposition
(AX2-21b)
CH2OOH + M -> H2CO + -OH
AX2-15
-------
Methyl hydroperoxide is much less soluble than hydrogen peroxide (H2O2), and so wet
deposition after incorporation into cloud droplets is much less important as a removal process
than it is for H2O2. CH3OOH can also be removed by dry deposition to the surface or transported
by convection to the upper troposphere. The lifetime of CH3OOH in the troposphere due to
photolysis and reaction with the OH radical is estimated to be «2 days. Methyl hydroperoxide is
then a temporary sink of radicals, with its wet or dry deposition representing a loss process for
tropospheric radicals.
The only important reaction for the methoxy radical (CH36) is
CH3 + O2 -» H2CO + HCy. (AX2-22)
HCy + NO -» -OH + NO2 (AX2-23)
The HO2 radicals produced in (AX2-22) can react with NO, O3, or other HO2 radicals according
to,
H02- + 03 -» -OH + 202 (AX2-24)
H02- + H02- -» H202 + 02. (AX2-25)
Formaldehyde (H2CO) produced in reaction AX2-22 can be photolyzed:
(55%) (AX2-26a)
(45%) (AX2-26b)
Formaldehyde also reacts with the OH radical,
•OH + H2CO -> H2O + HCO. (AX2-27)
AX2-16
-------
The H atom and HCO (formyl) radical produced in these reactions react solely with O2 to form
the HO2 radical:
•H + O2 + M -» HO2' + M (AX2-28)
HCO + O2 ->• HO2- + CO. (AX2-29)
The lifetimes of H2CO due to photolysis and reaction with OH radicals are «4 h and 1.5 days,
respectively, leading to an overall lifetime of slightly less than 4 hours for H2CO for overhead
sun conditions (Rogers, 1990).
The final step in the oxidation of CH4 involves the oxidation of CO by reaction with the
OH radical to form CO2:
CO + -OH -> C02 + -H (AX2-30)
HCO + 02 ->> H02' + CO. (AX2-29)
The lifetime of CO in the lower troposphere is «2 months at midlatitudes.
NO and HO2 radicals compete for reaction with CH3O2 and HO2 radicals, and the reaction
route depends on the rate constants for these two reactions and the tropospheric concentrations
of HO2 and NO. The rate constants for the reaction of the CH3O2- radicals with NO (reaction
AX2-18) and HO2 radicals (reaction AX2-19) are of comparable magnitude (e.g., Jet Propulsion
Laboratory, 2003). Based on expected HO2 radical concentrations in the troposphere, Logan
et al. (1981) calculated that the reaction of the CH3O2- radical with NO dominates for NO mixing
ratios of >30 ppt. For NO mixing ratios <30 ppt, the reaction of the CH3O2- radical with HO2
dominates. The overall effects of methane oxidation on O3 formation for the case when
NO >30 ppt can be written as:
AX2-17
-------
CH4 + -OH + 02 -> CH302- + H20 (AX2-16, AX2-17)
CH302- +NO -> CH30- +N02 (AX2-18)
CH3O + O2 -> H2CO + HO2- (AX2-22)
HO2« + NO -» -OH + NO2 (AX2-23)
2(NO2 + hv -» NO + O(3P)) (AX2-1)
2(O(3P) + O2 + M -» O3 + M) (AX2-2)
Net: CH4 + 4O2 + 2hv -> H2CO + 2O3 + H2O (AX2-31)
Further O3 formation occurs, based on the subsequent reactions of H2CO, e.g.,
H2CO + hv + 202 -» 2H02- + CO (AX2-26b; AX2-28; AX2-29)
2(H02- + NO -> -OH + N02) (AX2-23)
0(3P)) (AX2-1)
03+M) (AX2-2)
Net: H2CO + 4O2 + hv -> CO + 2O3 + 2-OH. (AX2-32)
Reactions in the above sequence lead to the production of two OH radicals which can further
react with atmospheric constituents (e.g., Crutzen, 1973). There is also a less important
pathway:
AX2-18
-------
hv->H2 + CO (AX2-26a)
CO + -OH -> C02 + -H (AX2.30)
•H + O2 + M -» HO2* + M (AX2-28)
H02- + NO-»OH + N02 (AX2-23)
NO2 + hv -> NO + O(3P) (AX2-1)
O(3P) + O2 + M -» O3 + M (AX2-2)
Net: H2CO + 2O2 + hv-»CO2 + O3+H2 (AX2-33)
These reaction sequences are important for tropospheric chemistry because formaldehyde is an
intermediate product of the oxidation of most VOCs. The reaction of O3 and HO2 radicals leads
to the net destruction of tropospheric O3:
H02- + 03 -> -OH + 202 (AX2-24)
Using the rate constants reported for reactions AX2-23 and AX2-24 (Atkinson et al., 1992) and
the background tropospheric O3 mixing ratios given above, the reaction of HO2 radicals with NO
dominates over reaction with O3 for NO mixing ratios >10 ppt. The rate constant for
reaction AX2-25 is such that an NO mixing ratio of this magnitude also means that the HO2
radical reaction with NO will be favored over the self-reaction of HO2 radicals.
Consequently, there are two regimes in the "relatively clean" troposphere, depending on
the local NO concentration: (1) a "very low-NOx" regime in which HO2 and CH3O2 radicals
combine (reaction AX2-19), and HO2 radicals undergo self-reaction (to form H2O2) and react
with O3 (reactions AX2-25 and AX2-24), leading to net destruction of O3 and inefficient OH
radical regeneration (see also Ehhalt et al., 1991; Ayers et al., 1992); and (2) a "low-NOx"
regime (by comparison with much higher NOX concentrations found in polluted areas) in
which HO2 and CH3O2 radicals react with NO to convert NO to NO2, regenerate the OH radical,
AX2-19
-------
and, through the photolysis of NO2, produce O3. In the "low NOX" regime there still may be
significant competition from peroxy-peroxy reactions, depending on the local NO concentration.
Nitric oxide mixing ratios are sufficiently low in the remote marine boundary layer
relatively unaffected by transport of NOX from polluted continental areas (< 15 ppt) that
oxidation of CH4 will lead to net destruction of O3, as discussed by Carroll et al. (1990) and
Ayers et al. (1992). In continental and marine areas affected by transport of NOX from
combustion sources, NO mixing ratios are high enough (of the order of-one to a few hundred
ppt) for the oxidation of CH4, nonmethane hydrocarbons (NMHCs) and CO to lead to net O3
formation (e.g., Carroll et al., 1990; Dickerson et al., 1995). Generally, NO mixing ratios
increase with altitude and can be of the order of fifty to a few hundred ppt in the upper
troposphere depending on location. The oxidation of peroxides, carbon monoxide and acetone
transported upward by convection, in the presence of this NO, can lead to local O3 formation
(e.g., Singh et al., 1995; McKeen et al., 1997; Wennberg et al., 1998; Bruhl et al., 2000).
AX2.2.6 The Atmospheric Chemistry of Alkanes
The same basic processes by which CH4 is oxidized occur in the oxidation of other, even
more reactive and more complex VOCs. As in the CH4 oxidation cycle, the conversion of NO
to NO2 during the oxidation of VOCs results in the production of O3 and the efficient
regeneration of the OH radical, which in turn can react with other VOCs (Figure AX2-2). The
chemistry of the major classes of VOCs important for O3 formation such as alkanes, alkenes
(including alkenes from biogenic sources), and aromatic hydrocarbons will be summarized in
turn.
Reaction with OH radicals represents the main loss process for alkanes and as also
mentioned earlier, reaction with nitrate and chlorine radicals are additional sinks for alkanes.
For alkanes having carbon-chain lengths of four or less, the chemistry is well understood and the
reaction rates are slow in comparison to alkenes and other VOCs of similar structure and
molecular weight. See Table AX2-1 for a comparison of reaction rate constants for several
small alkanes and their alkene and diene homologues. For alkanes larger than C5, the situation
is more complex because the products generated during the degradation of these compounds are
usually not well characterized. Branched alkanes have rates of reaction that are highly
AX2-20
-------
Anthropogenic
BiogenicVOC + OH
ROOM -*-
Carbonyl R°2
Alcohol
R-
|02
NO2
R02 9 * ROON02
RON02
Decomposition
02 ) Isomerization
Products
Figure AX2-2. General chemical mechanism for the oxidative degradation of VOCs.
Source: Atkinson (2000).
dependent on carbon backbone structure. Stable products of alkane photooxidation are known to
include carbonyl compounds, alkyl nitrates, and hydroxycarbonyls.
Alkyl nitrates form primarily as an alternate product of reaction AX2-34 (below). Several
modeling studies have predicted that large fractions of NOy exist as alkyl and hydroxy alkyl
nitrates (Calvert and Madronich, 1987; Atherton and Penner, 1988; Trainer et al., 1991).
In NOX- and VOC-rich urban atmospheres, 100 different alkyl and 74 different hydroxy alkyl
nitrate compounds have been predicted and identified (Calvert and Madronich, 1987; Schneider
and Ballschmiter, 1999; Schneider et al., 1998). Uncertainties in the atmospheric chemistry of
the alkanes include the branching ratio of reaction AX2-34, i.e., the extent to which alkyl nitrates
form versus RO and NO2. These uncertainties affect modeling predictions of NOX
concentrations, NO-to-NO2 conversion and O3 formation during photochemical degradation of
the VOCs. Discrepancies between observations and theory have been found in aircraft
measurements of NOy (Singh et al., 1996b). Recent field studies conducted by Day et al. (2003)
have shown that large fractions of organic nitrates, which may be associated with isoprene
AX2-21
-------
oxidation products, are present in urban and rural atmosphere that have not been previously
measured and considered in NOy calculations to date.
Alcohols and ethers in ambient air react almost exclusively with the OH radical, with the
reaction proceeding primarily via H-atom abstraction from the C-H bonds adjacent to the
oxygen-containing function group in these compounds (Atkinson and Arey, 2003).
The following list of general reactions, analogous to those described for methane,
summarizes the role of alkane oxidation in tropospheric O3 formation.
2 (AX2_34)
•R + O2 + M ->• RO2' + M (AX2-3 5)
RO2' + NO -> RO + NO2 (AX2-3 6)
HO2' + NO -> -OH + NO2 (AX2-23)
RO + O2 -> R'CHO + HO2- (AX2-3 7)
2(NO2 + hv -» NO + O(3p)) (AX2- 1 )
2(O(3p) + O2 + M -> O3 + M) (AX2-2)
Net: RH + 4O2 + 2hv -> R'CHO + 2O3 + H2O (AX2-3 8)
The oxidation of alkanes can also be initiated by other oxidizing agents such as NO3 and Cl
radicals. In this case, there is net production of an OH radical which can reinitiate the oxidation
sequence. The reaction of OH radicals with aldehydes forms acyl (R'CO) radicals, and acyl
peroxy radicals (R'C(O)O2) are formed by the addition of O2. As an example, the oxidation of
ethane (C2H5-H) yields acetaldehyde (CH3-CHO). Acetyl (CH3-CO) and acetylperoxy
(CH3-C(O)O2) radicals can then be formed. Acetylperoxy radicals can combine with NO2 to
form peroxyacetyl nitrate (PAN) via:
CH3C(O)O2' + NO2 + M o CH3C(O)O2NO2 + M (AX2-39)
AX2-22
-------
PAN can act as a temporary reservoir for NO2. Upon the decomposition of PAN, either locally
or elsewhere, NO2 is released to participate in the O3 formation process again. During the
oxidation of propane, the relatively long-lived intermediate acetone (CH3-C(O) CH3) is formed,
as shown in Figure AX2-3. The photolysis of acetone can be an important source of OH
radicals, especially in the upper troposphere (e.g., Singh et al., 1995). Examples of oxidation
mechanisms of more complex alkanes and other classes of hydrocarbons can be found in
comprehensive texts such as Seinfeld and Pandis (1998).
AX2.2.7 The Atmospheric Chemistry of Alkenes
As shown in Figure AX2-3, the presence of a double carbon-carbon bond, i.e., > C = C <,
in a VOC can greatly increase the range of potential reaction intermediates and products,
complicating the prediction of O3 production. The alkenes emitted from anthropogenic sources
are mainly ethene, propene, and the butenes, with lesser amounts of the > C5 alkenes. The major
biogenic alkenes emitted from vegetation are isoprene (2-methyl-1,3-butadiene) and C10H16
monoterpenes (Atkinson and Arey, 2003), and their tropospheric chemistry is currently the focus
of much attention (Zhang et al., 2002; Sauer et al., 1999; Geiger et al., 2003; Sprengnether et al.,
2002; Witter et al., 2002; Bonn and Moortgat, 2003; Berndt et al., 2003; Fick et al., 2003;
Kavouras et al., 1999; Atkinson and Arey, 2003).
Alkenes react in ambient air with OH and NO3 radicals and with O3. The mechanisms
involved in their oxidation have been discussed in detail by Calvert et al. (2000). All three
processes are important atmospheric transformation processes, and all proceed by initial addition
to the > C = C < bonds or, to a much lesser extent, by H atom extraction. Products of alkene
photooxidation include carbonyl compounds, hydroxy alkyl nitrates and nitratocarbonyls, and
decomposition products from the high energy biradicals formed in alkene-O3 reactions.
Table AX2-2 provides estimated atmospheric lifetimes for biogenic alkenes with respect to
oxidation by OH, NO3 and O3. The structures of most of the compounds given in Table AX2-2
are shown in Figure AX2-4.
Uncertainties in the atmospheric chemistry of the alkenes concern the products and
mechanisms of their reactions with O3, especially the yields of OH radicals, H2O2, and secondary
organic aerosol in both outdoor and indoor environments. However, many product analyses of
important biogenic and anthropogenic alkenes in recent years have aided in the narrowing of
AX2-23
-------
a. Propane
•OH +
HO
HO2« + CH2=O
CH3O« + NO2
b. Propene
H3C CH +HO2*
CH2=O + HO2*
Figure AX2-3. Hydroxyl radical initiated oxidation of a) propane and b) propene.
Source: Calvert et al. (2000).
AX2-24
-------
Table AX2-2. Calculated Atmospheric
Compounds (adapted from
Lifetimes of Biogenic Volatile Organic
Atkinson and Arey, 2003)a
Lifetime for Reaction with
Biogenic VOC
Isoprene
Monoterpenes
Camphene
2-Carene
3-Carene
Limonene
Myrcene
cis-/trans-Qcimene
a-Phellandrene
p-Phellandrene
a-Pinene
p-Pinene
Sabinene
a-Terpinene
y-Terpinene
Terpinolene
Sesquiterpenes
p-Caryophyllene
a-Cedrene
a-Copaene
a-Humulene
Longifolene
Oxygenates
Acetone6
Camphor
1,8-Cineole
c/s-3-Hexen-l-ol
c/s-3-Hexenyl acetate
Linalool
OHb
1.4 h
2.6 h
1.7 h
16h
49min
39 min
33 min
27 min
50 min
2.6 h
1.8 h
1.2 h
23 min
47 min
37 min
42 min
2.1 h
1.5 h
28 min
2.9 h
61 df
2.5 dh
l.Od1
1.3 hk
18 hk
52 mink
03C
0.92 d
13d
1.2 h
8.0 h
1.4 h
35 min
31 min
5.6 min
5.9 h
3.2 h
0.77 d
3.4 h
0.7 min
2.0 h
9.1 min
1 .4 min
9.8 h
1.8 h
1 .4 min
>23d
>3.2y8
> 165 dh
>77dJ
4.3 hk
5.1hk
39 min k
NO3d
1.6 h
1.7 h
4 min
7 min
5 min
6 min
3 min
0.9 min
8 min
11 min
27 min
7 min
0.5 min
2 min
0.7 min
3 min
8 min
4 min
2 min
1.6h
>8yf
>300dh
1.5y'
4.1 hk
4.Shk
6 mink
AX2-25
-------
Table AX2-2 (cont'd). Calculated Atmospheric Lifetimes of Biogenic Volatile Organic
Compounds (adapted from Atkinson and Arey, 2003)a
Lifetime for Reaction with
Biogenic VOC
Oxygenates (cont'd)
Methanol
2-Methyl-3 -buten-2-ol
6-Methyl-5 -hepten-2-one
OHb
12 df
2.4 h1
53 min°
CV
>3.2yg
1.2dm
0.7 h°
NO3d
2.0yf
7.7 dn
9min°
aRate coefficients from Calvert et al. (2000) unless noted otherwise.
b Assumed OH radical concentration: 1.0 x 106 molecule cm"3.
0 Assumed O3 concentration: 1 * 1012 molecule cm"3, 24-h average.
d Assumed NO3 radical concentration: 2.5 * 108 molecule cm"3, 12-h nighttime average.
e Photolysis will also occur with a calculated photolysis lifetime of -60 day for the lower troposphere, July, 40° N
(Meyrahnetal., 1986).
f Atkinson etal. (1999).
g Estimated.
hReisselletal. (2001).
1 Corchnoy and Atkinson (1990).
J Atkinson etal. (1990).
k Atkinson etal. (1995).
'Papagnietal. (2001).
mGrosjean and Grosjean (1994).
"Rudichetal. (1996).
"Smithetal. (1996).
these uncertainties. The reader is referred to extensive reviews by Calvert et al. (2000) and
Atkinson and Arey (2003) for detailed discussions of these products and mechanisms.
Oxidation by OH
As noted above, the OH radical reactions with the alkenes proceed mainly by OH radical
addition to the > C = C < bonds. As shown in Figure AX2-3, for example, the OH radical
reaction with propene leads to the formation of two OH-containing radicals. The subsequent
reactions of these radicals are similar to those of the alkyl radicals formed by H-atom abstraction
from the alkanes. Under high NO conditions, CH3CHCH2OH continues to react — producing
several smaller, "second generation," reactive VOCs.
AX2-26
-------
Camphene 2-Carene 3-Carene Limonene Myrcene
c;'s-Ocimene a-Phellandrene p-Phellandrene a-Pinene p-Pinene
Sabinene
a-Terpinene y-Terpinene Terpinolene
\
p-Caryophyllene
u-Cedrene
u-Copaene
a-Humulene
Longifolene
OH
Camphor
1,8-Cineole
Linalool
Figure AX2-4. Structures of a selected number of terpene and sesquiterpene compounds.
Source: Atkinson and Arey (2003).
AX2-27
-------
For the simple 1 ppb. The
values are much lower for lower NOX concentrations. The situation is much better for
methacrolein. Observed products can account for more than 90% of the reacted carbon.
The rates of formation of condensible, oxidation products of biogenic compounds that may
contribute to secondary organic aerosol formation is an important matter for the prediction of
ambient aerosol concentrations. Claeys et al. (2004a,b) found that 2-methyltetrols are formed
from the oxidation of isoprene in yields of about 0.2% on a molar basis, or 0.4% on a mass basis.
AX2-28
-------
These are semivolatile compounds that can condense on existing particles. On the other hand,
pinene oxidation leads to substantial organic aerosol formation.
Oxidation by Nitrate Radical
NO3 radical reacts with alkenes mainly by addition to the double bond to form a
b-nitrooxyalkyl radical (Atkinson 1991, 1994, 1997). The abstraction pathway may account for
up to 20% of the reaction. For propene, the initial reaction is followed by a series of reactions
NO3- + CH3CH=CH2 -» CH3CHCH2ONO2
(AX2-40)
-» CH3CH(ONO2)CH2
that (Atkinson, 1991) to lead to the formation of, among others, carbonyls and nitrato-carbonyls
including formaldehyde (HCHO), acetaldehyde (CH3CHO), 2-nitratopropanal
(CH3CH(ONO2)CHO), and 1-nitratopropanone (CH3C(O)CH2ONO2). By analogy to OH,
conjugated dienes like butadiene and isoprene will react with NO3 to form d-nitrooxyalkyl
radicals (Atkinson, 2000). If NO3 is available for reaction in the atmosphere, then NO
concentrations will be low, owing to the rapid reaction between NO3 and NO. Consequently,
nitrooxyalkyl peroxy radicals are expected to react primarily with NO2, yielding thermally
unstable peroxy nitrates, NO3, HO2, and organoperoxy radicals (Atkinson, 2000).
Several studies have undertaken the quantification of the products of NO3 initiated
degradation of several of the important biogenic alkenes in O3 and secondary organic aerosol
formation, including isoprene, a- and p-pinene, 3-carene, limonene, linalool, and 2-methyl-3-
buten-2-ol. See Figure AX2-4 for the chemical structures of these and other biogenic
compounds. The results of these studies have been tabulated by Atkinson and Arey (2003).
Oxidation by Ozone
Unlike other organic compounds in the atmosphere, alkenes react at significant rates
with O3. Ozone initiates the oxidation of alkenes by addition across carbon-carbon double
bonds, at rates that are competitive with reaction with OH (see Table AX2-1). The addition
AX2-29
-------
of O3 across the double bond yields an unstable ozonide, a 5-member ring including a single
carbon-carbon bond linked to the three oxygen atoms, each singly bound. The ozonide
rearranges spontaneously and then fragments to form an aldehyde or ketone, depending on the
original position of the double bond, and a high energy Criegee biradical. Collisional energy
transfer may stabilize the radical, preventing it from decomposing. Low pressure studies of the
decomposition of the Criegee biradical have shown high yields of the OH radical.
At atmospheric pressures, the rates of OH production have not been reliably established, due to
complications arising from subsequent reactions of the OH produced with the ozonide fragments
(Calvert et al., 2000).
The ozonolysis of larger biogenic alkenes yields high molecular weight oxidation products
with sufficiently low vapor pressures to allow condensation into the particle phase. Many
oxidation products of larger biogenic alkenes have been identified in ambient aerosol,
eliminating their further participation in O3 production. Figure AX2-5 shows the chemical
structures of the oxidation products of a-pinene and illustrates the complexity of the products.
Carbonyl containing compounds are especially prevalent. A summary of the results of product
yield studies for several biogenic alkenes can be found in Atkinson and Arey (2003).
CHO
CHO
CHO
O-OH
-CHO
CO2H
C02H
-CO2H
CO2H
CO9H
CO2H
Figure AX2-5. Products from the reaction of terpenes with O3.
Source: Atkinson and Arey (2003).
AX2-30
-------
NO2 also participates to a very small degree in the oxidation of alkenes by addition to
double bonds in a manner similar to O3. Rate constants for reactions of this type range
from 1CT18 to 1CT24 for dienes and monoalkenes (King et al., 2002). It should also be noted
that O3 reacts with terpenoid compounds released from household products such as air fresheners
and cleaning agents in indoor air to produce ultrafine particles (Wainman et al., 2000; Sarwar
et al., 2002)
AX2.2.8 The Atmospheric Chemistry of Aromatic Hydrocarbons
Aromatic hydrocarbons represent a major class of compounds found in gasoline and other
liquid fuels. Upon vaporization, most of these compounds react rapidly in the atmosphere
(Davis et al., 1975) and following a series of complex processes, involving molecular oxygen
and oxides of nitrogen, produce O3. Reaction with OH radicals serves as the major atmospheric
loss process of aromatic hydrocarbons. Atmospheric losses of alkyl aromatic compounds by O3
and nitrate radicals have been found to be minor processes for most monocyclic aromatic
hydrocarbons. (However, the reaction with of the nitrate radical with substituted
hydroxybenzenes, such as phenol or o-, m-, />-cresol, can be an important atmospheric loss
process for these compounds.) Much of the early work in this field focused on the temperature
dependence of the OH reactions (Perry et al., 1977; Tully et al., 1981) using absolute rate
techniques. Typically two temperature regions were observed for a large number of aromatic
compounds and the complex temperature profile suggested that two mechanisms were operative.
In the high temperature region, hydrogen (H)-atom abstraction from the aromatic ring
dominates, and in the temperature regime less than 320K, OH addition to the aromatic ring is the
dominant process. Thus, at normal temperatures and pressures in the lower troposphere, ring
addition is the most important reactive process followed by H-atom abstraction from any alkyl
substituents. The kinetics of monocyclic aromatic compounds are generally well understood and
there is generally broad consensus regarding the atmospheric lifetimes for these compounds. By
contrast, there is generally a wide range of experimental results from product studies of these
reactions. This leads to a major problem in model development due to a general lack of
understanding of the product identities and yields for even the simplest aromatic compounds,
which is due to the complex reaction paths following initial reaction with OH, primarily by the
addition pathway.
AX2-31
-------
Two comprehensive reviews, which provide a detailed understanding of the current state-
of-science of aromatic hydrocarbons have been written in the past five years. Atkinson (2000)
reviewed the atmospheric chemistry of volatile organic compounds, of which aromatic
hydrocarbons are included in one section of the review. More recently Calvert et al. (2002)
conducted a highly comprehensive examination of the reaction rates, chemical mechanisms,
aerosol formation, and contributions to O3 formation for monocyclic and polycyclic aromatic
hydrocarbons.
AX2.2.8.1 Chemical Kinetics and Atmospheric Lifetimes of Aromatic Hydrocarbons
Rate constants for the reaction of species in the atmosphere with aromatic hydrocarbons
vary widely depending on the number of aromatic rings and substituent groups. Reactions of O3
with aromatic hydrocarbons (AHCs) are generally slow except for monocyclic aromatic
hydrocarbons having unsaturated substituent groups. For example, indene and styrene have
atmospheric lifetimes of 3.3 h and 23 h with respect to reaction with O3, which are much longer
than that due to reactive loss with either OH or NO3. Thus, the atmospheric lifetimes and
reaction products of O3 and aromatic hydrocarbons will be ignored in this discussion. In
addition to chemical reaction, some organic compounds photolyze in the lower atmosphere.
Virtually all aromatic precursors are not subject to photolysis, although many of the ring
fragmentation products having multiple carbonyl groups can photolyze in the troposphere.
The reaction rates and atmospheric lifetimes of monocyclic aromatic compounds due to
reaction with OH radicals are generally dependent on the number and types of substituent groups
associated with the ring. These reaction rates have been found to be highly temperature and
pressure dependent. The temperature regimes are governed by the processes involved and show
a quite complex appearance. At room temperature (-300 K), both addition to the aromatic ring
and H-atom abstraction occur with the addition reaction being dominant. For the two smallest
monocyclic aromatic hydrocarbons, the initial addition adduct is not completely stabilized at
total pressures below 100 torr.
Numerous studies have been conducted to measure the OH + benzene rate constant over
a wide range of temperatures and pressures. An analysis of absolute rate data taken at
approximately 100 torr argon and not at the high pressure limit yielded a value of 1.2 x 10~12 cm3
AX2-32
-------
molecfJ s"1. Atkinson (1989) recommended a value of 1.4 x 10~12 cm3molecfJ s"1 at room
temperature and atmospheric pressure. This recommendation has been refined only slightly and
is reflected in the recent value recommended by Calvert et al. (2002) which is given as
1.39 x 10"12 cm3 molec'V1. This recommended value for the reaction of OH + benzene together
with values for other monocyclic aromatic hydrocarbons is given in Table AX2-3.
In general, it is observed that the OH rate constants with monocyclic alkyl aromatic
hydrocarbons are strongly influenced by the number of substituent groups found on the aromatic
ring. (That is, the identity of the alkyl substituent groups has little influence on the overall
reactions rate constant.) Single substituent single-ring aromatic compounds which include
toluene, ethyl benzene, n-propylbenzene, isopropylbenzene, and t-butylbenzene have average
OH reaction rate constants ranging from 4.5 to 7.0 x 10"12 cm3 molec"1 s"1 at room temperature
and atmospheric pressure. These rate constants lead to atmospheric lifetimes (see below) that
are still greater than 1 day. Rate constants for monocyclic aromatic compounds with greater
than 10 carbon atoms or more are generally not available.
The dominant monocyclic aromatic compounds with two substituents are TO-, o-, and
/>-xylene. Their recommended OH rate constants range from 1.4 to 2.4 x I0~n cm3 molec"1 s"1.
Similarly, the three isomers of ethyltoluene have recommended OH rate constants ranging from
1.2 tol.9 x I0~n cm3 molec"1 s"1. The only other two substituent single-ring aromatic compound
for which the OH rate constant has been measured is/>-cymene (para-isopropyltoluene), giving a
value of 1.5 x io~n cm3 molec"1 s"1.
OH rate constants for the C9 trimethyl substituted aromatic hydrocarbons (1,2,3-; 1,2,4-;
1,3,5-trimethylbenzene) are higher by a factor of approximately 2.6 over the di-substituted
compounds. Rate constants for the three isomers range from 3.3 to 5.7 x I0~n cm3 molec"1 s"1.
While concentrations for numerous other trisubstituted benzene compounds have been reported
(e.g., l,2-dimethyl-4-ethylbenzene), OH rate constants for trimethylbenzene isomers are the only
tri substituted aromatic compounds that have been reported.
Aromatic hydrocarbons having substituent groups with unsaturated carbon groups have
much higher OH rate constants than their saturated analogues. The smallest compound in this
group is the C8 AHC, styrene. This compound reacts rapidly with OH and has a recommended
rate constant of 5.8 x I0~n cm3 molec"1 s"1 (Calvert, 2002). Other methyl substituted styrene-
AX2-33
-------
Table AX2-3. Hydroxyl Rate Constants and Atmospheric Lifetimes of Mono- and
Di-cyclic Aromatic Hydrocarbons (adapted from Atkinson 2000)
Compound
Benzene
Toluene
Ethylbenzene
«-Propylbenzene
Isopropylbenzene
^-Butylbenzene
o-Xylene
m-Xylene
/•-Xylene
o-Ethyltoluene
m-Ethyltoluene
/•-Ethyltoluene
p-Cymene
1,2,3-Trimethylbenzene
1 ,2,4-Trimethylbenzene
1 , 3 ,5 -Trimethylbenzene
Indan
Styrene3
a-Methylstyrene
Napthalene
1 -Methylnapthalene
2-Methylnapthalene
OH Rate Constant
(xlO12)
1.4
5.6
7
5.8
6.3
4.5
14
23
14
12
19
12
14
33
33
57
19
58
51
23
53
52
TOH
(as indicated)
8.3d*
2.1 d
1.7 d
2.0 d
1.8 d
2.6 d
20 h
12 h
19 h
23 h
15 h
24 h
19 h
8.4 h
8.6 h
4.8 h
15 h
4.8 h
5.4 h
12 h
4.8 h
8.8 h
'Rate coefficients given as cmVmolec-sec.
2Lifetime for zero and single alkyl substituted aromatic based on OH concentration of 1 x 106 molec cm"3.
3 Lifetime for reaction of styrene with NO3 is estimated to be 44 min based on a nighttime NO3 concentration
of 2.5 x 108 molec cm"3 and a rate coefficient of 1.5 x lCT12cm3/molec-sec.
AX2-34
-------
type compounds (e.g., a-methylstyrene) have OH rate constants within a factor of two of that
with styrene. However, for unsaturated monocyclic aromatic hydrocarbons other processes
including atmospheric removal by NO3 radicals can also be important, particularly at night when
photolysis does not substantially reduce the NO3 radical concentration (see below).
Polycyclic aromatic hydrocarbons are found to a much lesser degree in the atmosphere
than are the monocyclic aromatic hydrocarbons. For example, measurements made in Boston
during 1995 (Fujita et al., 1997) showed that a single PAH (napthalene) was detected in the
ambient morning air at levels of approximately 1% (C/C) of the total monocyclic aromatic
hydrocarbons. 1-methyl and 2-methylnaphthalene have sufficient volatility to be present in the
gas phase. Other higher molecular weight PAHs (< 3 aromatic rings), if present, are expected to
exist in the gas phase at much lower concentrations than napthalene and are not considered here.
OH rate constants for napthalene and the two methyl substituted napthalene compounds have
been reviewed by Calvert et al. (2002). The values recommended (or listed) by Calvert et al.
(2002) are given in Table AX2-3. As seen in the monocyclic aromatic hydrocarbons, the
substitution of methyl groups on the aromatic ring increases the OH rate constant, in this case by
a factor of 2.3.
Some data is available for the reaction of OH with aromatic oxidation products. (In this
context, aromatic oxidation products refer to those products which retain the aromatic ring
structure.) These include the aromatic carbonyl compound, benzaldehyde, 2,4-; 2,5-; and
3,4-dimethyl-benzaldehyde, and t-cinnamaldehyde. Room temperature rate constants for these
compounds range from 1.3 x 10~n cm3 molec"1 s"1 (benzaldehyde) to 4.8 x 10~n cm3 molec"1 s"1
(t-cinnamaldehyde). While the yields for these compounds are typically between 2 to 6%, they
can contribute to the aromatic reactivity for aldehydes having high precursor concentration (e.g.,
toluene, 1,2,4-trimethylbenzene). OH also reacts rapidly with phenolic compounds. OH
reaction rates with phenols and o-, m-, and/>-cresol are typically rapid (2.7 to 6.8 x 10~n cm3
molec"1 s"1) at room temperature. Five dimethylphenols and two trimethylphenols have OH
reaction rates ranging between 6.6 x 10~n and 1.25 x 10"10 cm3 molec"1 s"1. Finally, unlike the
aromatic aldehydes and phenols, reaction rates for OH + nitrobenzene and OH + w-nitrotoluene
are much lower than the parent molecules, given their electron withdrawing behavior from the
aromatic ring. The room temperature rate constants are 1.4 x 10"13 and 1.2 x 10~12, respectively.
AX2-35
-------
The NO3 radical is also known to react with selected AHCs and aromatic photooxidation
products. Reaction can either occur by hydrogen atom abstraction or addition to the aromatic
ring. However, these reactions are typically slow for alkyl aromatic hydrocarbons and the
atmospheric removal due to this process is considered negligible. For AHCs having substituent
groups with double bonds (e.g., styrene, a-methylstyrene), the reaction is much more rapid, due
to the addition of NO3 to the double bond. For these compounds, NO3 rate constants are on the
order of 10"12 cm3 molec"1 s"1. This leads to atmospheric lifetimes on the order of about 1 h for
typical night time atmospheric NO3 levels of 2.5 x 108 molec cm"3 (Atkinson, 2000).
The most important reactions of NO3 with AHCs are those which involve phenol and
methyl, dimethyl, and trimethyl analogs. These reactions can be of importance due to the high
yields of phenol for the atmospheric benzene oxidation and o-, w-,/?-cresol from toluene
oxidation. The NO3 + phenol rate has been given as 3.8 x 10~12 cm3 molec"1 s"1. Similarly, the
cresol isomers each has an extremely rapid reaction rate with NO3 ranging from 1.1 to
1.4 x 1CT11 cm3 molec"1 s"1. As a result, these compounds, particularly the cresol isomers, can
show rapid nighttime losses due to reaction with NO3 with nighttime lifetimes on the order of a
few minutes. There is little data for the reaction of NO3 with dimethylphenols or
trimethylphenols which have been found as products of the reaction of OH + m-, />-xylene and
OH+ 1,2,4-; 1,3,5-trimethylbenzene.
AX2.2.8.2 Reaction Products and Mechanisms of Aromatic Hydrocarbon Oxidation
An understanding of the mechanism of the oxidation of AHCs is important 1 if O3 is to be
accurately predicted in urban atmospheres through modeling studies. As noted above, most
monocyclic aromatic hydrocarbons are removed from the atmosphere through reaction with OH.
Thus, product studies of the OH + AHC should provide the greatest information regarding the
AHC oxidation products. However, the effort to study these reactions has been intractable over
the past two decades due to a number of difficulties inherent in the OH-aromatic reaction
system. There are several reasons for the slow progress in understanding these mechanisms.
(1) Product yields for OH-aromatic systems are poorly understood; for the most studied system,
OH-toluene, approximately 50% of the reaction products have been identified under conditions
where NO2 reactions do not dominate the removal of the OH-aromatic adduct. (2) As noted, the
reaction mechanism can change as the ratio of NO2 to O2 changes in the system (Atkinson and
AX2-36
-------
Aschmann, 1994). Thus, reaction product distributions that may be measured in the laboratory
at high NO2 (or NOX) concentrations may not be applicable to atmospheric conditions. This also
limits the usefulness of models to predict O3 formation to the extent that secondary aromatic
reactions are not completely parameterized in the system. (3) Aromatic reactions produce highly
polar compounds for which there are few calibration standards available. In most cases,
surrogate compounds have to be used in GC/MS calibrations. Moreover, it is not at all clear
whether the present sampling techniques or analytical instruments are appropriate to measure the
highly polar products produced in these systems. (4) Finally for benzene and toluene in
particular, reaction rates of the products are substantially faster than that of the parent
compounds. Thus, it is difficult to measure yields accurately without substantial interferences
due to secondary reactions. Even given these difficulties, over the past decade a body of
knowledge has been developed whereby the initial steps in the OH-initiated photooxidation have
been established and a wide range of primary products from each of the major reaction systems
have been catalogued.
Benzene is one of the most important aromatic hydrocarbons released into the atmosphere
and is a recognized carcinogen. However, its reaction with OH is extremely slow and its
contribution to urban O3 formation is generally recognized to be negligible (Carter, 1994). As a
result, relatively few studies have been conducted on the OH reaction mechanism of benzene.
Major products of the oxidation of benzene have been found to be phenol and glyoxal (Berndt
et al., 1999; Tuazon et al., 1986).
Most of the product analysis and mechanistic work on alkyl aromatic compounds in the gas
phase has focused on examining OH reactions with toluene. The primary reaction of OH with
toluene follows either of two paths, the first being an abstraction reaction from the methyl group
and the second being addition to the ring. It has previously been found that H-atom abstraction
from the aromatic ring is of minor importance (Tully et al., 1981). A number of studies have
examined yields of the benzyl radical formed following OH abstraction from the methyl group.
This radical forms the benzyl peroxy radical, which reacts with nitric oxide (NO) leading to the
stable products benzaldehyde, with an average yield of 0.06, and benzyl nitrate, with an average
yield less than 0.01 (Calvert et al., 2002). Thus, the overall yield for the abstraction channel is
less than approximately 7%.
AX2-37
-------
It is now generally recognized that addition of OH to the aromatic ring is the major process
removing toluene from the atmosphere and appears to account for more than 90% of the reaction
yield for OH + toluene. The addition of OH to the ring leads to an intermediate OH-toluene
adduct that can be stabilized or can redissociate to the reactant compounds. For toluene, OH
addition can occur at any of the three possible positions on the ring (ortho, meta, or para) to form
the adduct. Addition of OH to the toluene has been shown to occur predominately at the ortho
position (yield of 0.81) with lesser amounts at the meta (0.05) and para (0.14) positions (Kenley
et al., 1981). The initial steps for both the abstraction and addition pathways in toluene have
been shown in Figure AX2-6; only the path to form the ortho-adduct is shown, viz. reaction (2).
The OH-toluene adduct formed is an energy-rich intermediate that must be stabilized by
third bodies in the system to undergo further reaction. Stabilization has been found to occur at
pressures above 100 Torr for most third bodies (Perry et al., 1977; Tully et al., 1981). Therefore,
at atmospheric pressure, the adduct will not substantially decompose back to its reactants as
indicated by reaction (-2). The stabilized adduct (I) is removed by one of three processes:
H-atom abstraction by O2 to give a cresol, as in reaction (5); an addition reaction with O2, as in
reaction (6); or reaction with NO2 to give w-nitrotoluene as in reaction (7).
The simplest fate for the adduct (I) is reaction with O2to form o-cresol. Data from a
number of studies (e.g., Kenley et al., 1981; Atkinson et al., 1980; Smith et al., 1998; Klotz
et al., 1998; summarized by Calvert et al., 2002) over a wide range of NO2 concentrations
(generally above 1 ppmv) show an average yield of approximately 0.15 for o-cresol. Most of the
measurements suggest the o-cresol yield is independent of total pressure, identity of the third
body, and NO2 concentration (Atkinson and Aschmann, 1994; Moschonas et al., 1999), but the
data tend to be scattered. This finding suggests that the addition of NO2 to the hydroxy
methylcyclo-hexadienyl radical does not contribute to the formation of phenolic-type
compounds. Fewer studies have been conducted for m and/?-cresol yields, but the results of two
studies indicate the yield is approximately 0.05 (Atkinson et al., 1980; Gery et al., 1985; Smith
et al., 1998). The data suggests good agreement between the relative yields of the cresols from
the product studies at atmospheric pressure and studies at reduced pressures. Thus, H-atom
abstraction from adducts formed at all positions appears to represent approximately 20% of the
total yield for toluene.
AX2-38
-------
CH3
+ -OH
Toluene
CH3
+ -OH
CH2OO'
+ NO
OH
02
+ O2
+ NO,
CH200'
+ H2O
CH3
OH
(I)
CHO
CH2ON02
CH3
OH
(1)
NO9 (3)
(4)
+ HOo (5)
(6)
H9O (7)
Figure AX2-6. Initial steps in the photooxidation mechanism of toluene initiated by its
reaction with OH radicals.
AX2-39
-------
The OH-toluene adduct also reacts with O2 to form a cyclohexadienyl peroxy radical (III),
shown as a product of reaction (6) after rearrangement. This radical can undergo a number of
possible processes. Most of these processes lead to ring fragmentation products, many of which
have been seen in several studies (Dumdei and O'Brien, 1984; Shepson et al., 1984).
Ring-fragmentation products are frequently characterized by multiple double bonds and/or
multiple functional groups. As such, these products are highly reactive and extremely difficult
to detect and quantify.
Klotz et al. (1997; 1998) have suggested that the intermediate could also follow through a
mechanism where toluene oxide/oxepin could be formed following the addition of O2 to the
OH-aromatic adduct. Recent experiments suggest that the formation of o-cresol through the
photolysis of toluene oxide/oxepin is only a minor contributor to the overall o-cresol that has
been measured (Klotz et al., 1998). This result contrasts to the high yield observed for the
formation of phenol from the photolysis of benzene oxide/oxepin (Klotz et al., 1997). Recently,
Berndt et al. (1999) used a flow tube to test the hypothetical formation of benzene oxide/oxepin
from the OH + benzene reaction at pressures below 100 torr. They saw very little evidence for
its formation.
A few studies have been conducted to identify fragmentation products using a variety of
instruments. Several approaches have been used that employ structural methods, particularly
mass spectrometry (MS), to identify individual products formed during the photooxidation.
In one approach (Dumdei and O'Brien, 1984), the walls of the reaction chamber were extracted
following an extended irradiation. In this study, the analysis was conducted by tandem mass
spectrometry (MS/MS), which allowed products to be separated without the use of a
chromatographic stationary phase. The investigators reported 27 photooxidation products from
toluene, with 15 reportedly from ring fragmentation processes. However, the study was purely
qualitative and product yields could not be obtained. No distinction could be made between
primary and secondary products from the reaction because extended irradiations and species in
various isotopic forms could not be differentiated. More refined approaches using atmospheric
pressure ionization-tandem mass spectrometry has been used to study toluene (Dumdei et al.,
1988) and m- and/?-xylene (Kwok et al., 1997) photooxidation.
In another study, Shepson et al. (1984) demonstrated that a number of these fragmentation
products could be analyzed by gas chromatography. Fragmentation products detected in two
AX2-40
-------
investigations (Dumdei and O'Brien, 1984; Shepson et al., 1984) included glyoxal, methyl
glyoxal, butenedial, 4-oxo-2-pentenal, hydroxybutenedial, l-pentene-3,4-dione, l-butene-3,4-
dione, and methyl vinyl ketone. Additional evidence (Shepson et al., 1984) for fragmentation
processes came from the detection of 2-methylfuran and furfural. These compounds, although
cyclic in structure, result from a bridged oxygen intermediate. Yields of the detected
fragmentation products were subsequently measured in a number of studies (e.g., Bandow et al.,
1985a,b; Tuazon et al., 1986; Smith et al., 1998), were typically under 15% on a reacted carbon
basis.
An additional possible pathway for reaction of the OH-toluene adduct is by reaction
withNO2 to give isomers of nitrotoluene. A yield of approximately 0.015 atNO2 concentrations
of about 1 ppmv has been measured (Atkinson et al., 1991). Although this yield itself is fairly
minor, the investigators reported a positive intercept in plotting the nitrotoluene concentration
against the NO2 concentration; however, the data were considerably scattered. The positive
intercept has been interpreted as suggesting that the OH-toluene adduct does not add O2. This
finding would require, therefore, another mechanism than that described above to be responsible
for the fragmentation products.
The results of this study can be compared to experiments which directly examined the OH
radical loss in reactions of OH with toluene and other aromatic compounds. Knipsel and co-
workers (Knispel et al., 1990) have found a double exponential decay for toluene loss in the
presence of added O2, a rapid decay reflective of the initial adduct formation and a slower decay
reflecting loss of the adduct by O2 or other scavengers. From the decay data in the presence
of O2, they determine a loss rate for the OH-toluene of 5.4 x 10"16 cm3 molec"1 s"1. Use of this
rate constant suggests that the loss rate of 2500 s"1 for the adduct in the presence of air at
atmospheric pressure. This loss rate compares to a loss due to NO2 (with a nominal atmospheric
concentration of 0.1 ppmv) of about 100 s"1. This finding suggests that removal of the OH-
toluene adduct by O2 is a far more important loss process than removal by NO2 under
atmospheric conditions which is in contrast other findings (Atkinson et al., 1991). This finding
was confirmed by the recent experiments from Moschonas et al. (1999).
Therefore, studies on the disposition of toluene following OH reaction can be summarized
as follows. It is generally accepted that H-atom abstraction from the methyl group by OH is a
relatively minor process accounting for a 6 to 7% yield in the OH reaction with toluene.
AX2-41
-------
Addition of OH to toluene to form an intermediate OH-toluene adduct is the predominant
process. At atmospheric pressure, ring-retaining products such as the cresol and nitrotoluenes
account for another 20% of the primary reaction products (Smith et al., 1999). The remaining
70 to 75% of the products are expected to be ring fragmentation products in the gas phase,
having an uncertain mechanism for formation. Many of these fragmentation products have been
detected, but appear to form at low yields, and relatively little quantitative information on their
formation yields exists. As noted earlier, some of these products contain multiple double bonds,
which are likely to be highly reactive with OH or photolyze which enhances the reactivity of
systems containing aromatics. Mechanisms that cannot adequately reflect the formation of
fragmentation products are likely to show depressed reactivity for the oxidation of toluene and
other aromatic compounds.
The number of studies of the multiple-substituted alkyl aromatics, such as the xylenes or
the trimethylbenzenes, is considerably smaller than for toluene. Kinetic studies have focused on
the OH rate constants for these compounds. For the xylenes, this rate constant is typically a
factor of 2 to 5 greater than that for OH + toluene. Thus, the OH reactivity of the fragmentation
products is similar to that of the parent compounds, potentially making the study of the primary
products of the xylenes less prone to uncertainties from secondary reactions of the primary
products than is the case for toluene.
Products from the OH reaction with the three xylenes have been studied most
comprehensively in a smog chamber using long-path FTIR (Bandow and Washida, 1985a) and
gas chromatography (Shepson et al., 1984; Atkinson and Aschmann, 1994; Smith et al., 1999).
Ring-fragmentation yields of 41, 55, and 36% were estimated for o-, m-, and/?-xylene,
respectively, based on the dicarbonyl compounds, glyoxal, methyl glyoxal, biacetyl, and 3-
hexene-2,5-dione detected during the photooxidation. These values could be lower limits, given
that Shepson et al. (1984) report additional fragmentation products from o-xylene, including
l-pentene-4,5-dione, butenedial, 4-oxo-pentenal, furan, and 2-methylfuran. In the earlier
studies, aromatic concentrations were in the range of 5 to 10 ppmv with NOX at 2 to 5 ppmv.
At atmospheric ratios of NO2 and O2, the observed yields could be different. Smith et al. (1999)
examined most of the ring retaining products in the OH + w-xylene and OH +/>-xylene systems.
In each case, tolualdehyde isomers, dimethylphenol isomers, and nitro xylene isomers specific
for each system were detected. The total ring retaining yield for OH + w-xylene was 16.3%; the
AX2-42
-------
yield for OH +/>-xylene it was 24.5%. A mass balance approach suggests that respective ring-
fragmentation yields of 84% and 76%, respectively. Kwok et al. (1997) also measured products
from the OH + m- and/?-xylene systems using atmospheric pressure ionization-tandem mass
spectrometry. Complementary ring-fragmentation products to glyoxal, methylglyoxal, and
biacetyl were detected from the parent ion peaks, although the technique did not permit the
determination of reaction yields.
Smith et al. (1999) also studied ring fragmentation products from the reaction of OH
with 1,2,4- and 1,3,5-trimethylbenzene. Ring-retaining products from the reaction with
1,2,4-trimethylbenzene gave three isomers each of dimethylbenzaldehyde and trimethylphenol
as expected by analogy with toluene. However, the ring-retaining products only accounted for
5.8% of the reacted carbon. Seven additional ring-fragmentation products were also detected
from the reaction, although the overall carbon yield was 47%. For 1,3,5-trimethylbenzene, its
reaction with OH leads to only two ring-retaining products, 3,5-dimethyl-benzaldehyde and
2,4,6-trimethylphenol, given its molecular symmetry. Only a single fragmentation product was
detected, methyl glyoxal, at a molar yield of 90%. The overall carbon yield in this case was
61%. The formation of relatively low yields of aromatic aldehydes and methylphenols suggests
that NOX removal by these compound in these reaction systems will be minimized (see below).
In recent years, computational chemistry studies have been applied to reaction dynamics of
the OH-aromatic reaction systems. Bartolotti and Edney (1995) used density functional-based
quantum mechanical calculations to help identify intermediates of the OH-toluene adduct. These
calculations were consistent with the main addition of OH to the ortho position of toluene
followed by addition of O2 to the meta position of the adduct. The reaction energies suggested
the formation of a carbonyl epoxide which was subsequently detected in aromatic oxidation
systems by Yu and Jeffries, (1997). Andino et al. (1996) conducted ab initio calculations using
density functional theory with semiempirical intermediate geometries to examine the energies of
aromatic intermediates and determine favored product pathways. The study was designed to
provide some insight into the fragmentation mechanism, although only a group additivity
approach to calculate AH^ was used to investigate favored reaction pathways. However, the
similarity in energies of the peroxy radicals formed from the O2 reaction with the OH-aromatic
adduct were very similar in magnitude making it difficult to differentiate among structures.
AX2-43
-------
A detailed analysis of toluene oxidation using smog chamber experiments and chemical
models (Wagner et al., 2003) shows that there are still large uncertainties in the effects of
toluene on O3 formation. A similar situation is likely to be found for other aromatic
hydrocarbons.
AX2.2.8.3 The Formation of Secondary Organic Aerosol as a Sink for Ozone Precursors
Aromatic hydrocarbons are known to generate secondary organic aerosol (SOA) following
their reaction with OH or other reactive oxidants. Secondary organic aerosol refers to the
formation of fine particulate matter either through nucleation processes or through condensation
onto existing particles. Over the last 12 years numerous experiments have been conducted in
environmental chambers to determine the yield of secondary organic aerosol as a function of the
reacted aromatic hydrocarbon. A review of the results of these studies can be found in the latest
Air Quality for Particulate Matter Document (U.S. Environmental Protection Agency, 2003).
The extent to which aromatic reaction products are removed from the gas phase and
become incorporated in the particle phase will influence the extent to which oxygenated organic
compounds will not be available for participation in the aromatic mechanisms that lead to O3
formation. However, this may be overstated to some degree for products of aromatic precursors.
First, at atmospheric loading levels of organic particulate matter, the SOA yields of the major
aromatic hydrocarbons are in the low percent range. Second, the aromatic products that are
likely to condense on particles are likely to be highly oxygenated and have OH reaction rates
that make them largely unreactive. Thus, while there may be some reduction of O3 formation, it
is not expected to be large.
AX2.2.9 Importance of Oxygenated VOCs
The role of oxygenated VOCs in driving O3 production has generated increasing interest
over the past decade. These VOCs include carbonyls, peroxides, alcohols, and organic acids.
They are produced in the atmosphere by oxidation of hydrocarbons, as discussed above, but are
also directly emitted to the atmosphere, in particular by vegetation (Guenther et al., 2000).
In rural and remote atmospheres, oxygenated VOCs often dominate over nonmethane
hydrocarbons in terms of total organic carbon mass and reactivity (Singh et al., 2004). The most
abundant by mass of these oxygenated VOCs is usually methanol, which is emitted by
AX2-44
-------
vegetation and is present in U.S. surface air at concentrations of typically 1-10 ppbv (Heikes
et al., 2002).
Most oxygenated VOCs react with OH to drive O3 production in a manner similar to the
hydrocarbon chemistry discussed in the previous sections. In addition, carbonyl compounds
(aldehydes and ketones) photolyze to produce peroxy radicals that can accelerate O3 production,
thus acting as a chemical amplifier (Jaegle et al., 2001). Photolysis of formaldehyde by
(AX2-26b) was discussed in section AX2.2.5. Also of particular importance is the photolysis of
acetone (Blitz et al., 2004):
20->
(CH3)2C(O) + hv - ?_» CH3C(O)O2- (AX2-41)
producing organic peroxy radicals that subsequently react with NO to produce O3. The
peroxyacetyl radical CH3C(O)OO can also react with NO2 to produce PAN, as discussed in
Section AX2.2.4. Photolysis of acetone are a minor but important source of HO2 radicals in the
upper troposphere (Arnold et al., 2004).
AX2.2.10 Influence of Multiphase Chemical Processes
In addition to reactions occurring in the gas phase, reactions occurring on the surfaces of or
within cloud droplets and airborne particles also occur. Their collective surface area is huge,
implying that collisions with gas phase species occur on very short time scales. The integrated
aerosol surface area ranges from 4.2 x 1CT7 cm2/cm3 for clean continental conditions to
1.1 x 1CT5 cm2/cm3 for urban average conditions (Whitby, 1978). There have been substantial
improvements in air quality especially in urban areas since the time these measurements were
made and so the U.S. urban values should be scaled downward by roughly a factor of two to
four. The resulting surface area is still substantial and the inferred collision time scale of a
gaseous molecule with a particle ranges from a few seconds or less to a few minutes. These
inferred time scales imply that heterogenous reactions will generally be much less important than
gas phase reactions for determining radical concentrations especially when reaction probabilities
much less then unity are considered. A large body of research has accumulated recently
AX2-45
-------
regarding chemical processes in cloud droplets, snow and ice crystals, wet (deliquesced)
inorganic particles, mineral dust, carbon chain agglomerates and organic carbon-coated particles.
Jacob's (2000) comprehensive review of the potential influences of clouds and aerosols on
tropospheric O3 cycling provides the starting point for this section. Updates to that review will
also be provided. Jacob's review evaluates the literature available through late 1999, discusses
major areas of uncertainty, recommends experiments to reduce uncertainties, and (based on then
current information) recommends specific multiphase pathways that should be considered in
models of O3 cycling. In regard to the latter, Jacob's recommendations should be viewed as
conservative. Specifically, only reasonably well constrained pathways supported by strong
observational evidence are recommended for inclusion in models. Several poorly resolved
and/or controversial pathways that may be significant in the ambient troposphere lack sufficient
constraints for reliable modeling. Some of these areas are discussed in more detail below.
It should be noted at the outset that many of the studies described in this section involve either
aerosols that are not found commonly throughout the United States (e.g., marine aerosol) or
correspond to unaged particles (e.g., soot, mineral dust). In many areas of the United States,
particles accrete a layer of hydrated H2SO4 which will affect the nature of the multiphase
processes occurring on particle surfaces.
Major conclusions from this review are summarized as follows (comments are given in
parentheses):
HOX Chemistry
(1) Catalytic O3 loss via reaction of O2 + O3(aq) in clouds appears to be inefficient.
(2) Aqueous-phase loss of HCHO in clouds appears to be negligible (see also Lelieveld and
Crutzen, 1990).
(3) Scavenging of HO2 by cloud droplets is significant and can be acceptably parameterized
with a reaction probability of YHo2 = 0.2, range 0.1 to 1, for HO2 -» 0.5 H2O2. However,
this approach may overestimate HO2 uptake because the influence of HO2(aq) on the
magnitude (and direction) of the flux is ignored.
(4) The uptake of alkyl peroxy radicals by aerosols is probably negligible.
(5) Hydrolysis of CH3C(O)OO in aqueous aerosols may be important at night in the
presence of high PAN and aerosol surface area; YcH3c(opo = 4 x io~3 is recommended.
AX2-46
-------
NOX Chemistry
(6) Hydrolysis of N2O5 to HNO3 in aqueous aerosols is important (Section AX2.2.4)
(and can be parameterized with JHN03 = 0.01 to 0.1 [Schutze and Herrmann, 2002;
Hallquist et al., 2003]).
(7) Although the mechanism is uncertain, heterogeneous conversion of NO2 to HONO
on aerosol surfaces should be considered with YNo2 = 10 4 (range 10~6 to 10~3) for
NO2 ->• 0.5 HONO + 0.5 HNO3. (This reaction also occurs on snow, Crawford et al.,
2001). Wet and dry deposition sinks for HONO should also be considered although
scavenging by aerosols appears to be negligible.
(8) There is no evidence for significant multiphase chemistry involving PAN.
(9) There is no evidence for significant conversion of HNO3 to NOX in aerosols.
Heterogeneous ozone loss
(10) There is no evidence for significant loss of O3 to aerosol surfaces (except during
dust storms observed in East Asia, e.g., Zhang and Carmichael, 1999).
Halogen radical chemistry
(11) There is little justification for considering BrOx and C1OX chemistry (except perhaps
in limited areas of the United States and nearby coastal areas).
Most of the above conclusions remain valid but, as detailed below, some should be
qualified based on recently published findings and on reevaluation of results form earlier
investigations.
AX2.2.10.1 HOX and Aerosols
Field measurements of HOX reviewed by Jacob (2000) correspond to regions with
relatively low aerosol concentrations (e.g., Mauna Loa [Cantrell et al., 1996]; rural Ontario
[Plummer et al., 1996]; and the upper troposphere [Jaegle et al., 1999]). In all cases, however,
significant uptake of HO2 or HO2 + RO2 radicals by aerosols was inferred based on imbalances
between measured concentrations of peroxy radicals and photochemical models of gas-phase
chemistry. Laboratory studies using artificial aerosols (both deliquesced and solid) confirm
uptake but the actual mechanism remains unclear. Several investigations report significant HOX
and H2O2 production in cloud water (e.g., Anastasio et al., 1994). However the potential
importance of this source is considered unlikely because measurements in continental air show
AX2-47
-------
no evidence of missing sources for HOX or H2O2. No investigations involving the potential
influences of marine aerosols as sources or sinks for HOX were considered in the above analysis.
Relative to conservative seawater tracers such as Mg2+ and Na+, organic C associated with
sea-salt aerosols is typically enriched by 2 to 3 orders of magnitude in both polluted (e.g.,
Hoffman and Duce, 1976, 1977; Turekian et al., 2003) and remote regions (Chesselet et al.,
1981). This organic C originates from three major sources: 1) organic surfactants concentrated
from bulk seawater on walls of subsurface bubbles (Tseng et al., 1992), 2) the surface microlayer
of the ocean (Gershey, 1983), and 3) condensation of organic gases (Pun et al., 2000).
Coagulation of chemically distinct aerosols (e.g., via cloud processing) may also contribute
under some conditions.
Resolving chemical processes involving particles in the marine boundary layer (MBL) is
constrained by the relative scarcity of measurements of particulate organic carbon (POC)
(Penner, 1995) and its molecular composition (Saxena et al., 1995). In MBL regions impacted
by direct continental outflow, POC may constitute more that half of the total dry aerosol mass
(Hegg et al., 1997). Carbon isotopic compositions in the polluted North Atlantic MBL indicate
that, on average, 35% to 40% of POC originates from primary (direct injection) and secondary
(condensation of gases) marine sources (Turekian et al., 2003).
The photolysis of dissolved organic compounds is a major source for OH, H2O2, and
C-centered radicals in both the surface ocean (e.g., Blough and Zepp, 1995; Blough, 1997;
Mopper and Kieber, 2000) and in marine aerosols (e.g., McDow et al., 1996). Relative to the
surface ocean, however, production rates in the aerosol are substantially greater per unit volume
because organic matter is highly enriched (Turekian et al., 2003) and aerosol pH is much lower
(Keene et al., 2002). Lower pHs increase rates of many reactions including acid-catalyzed
pathways such as the breakdown of the HOC1" radical (King et al., 1995), the formation of H2O2
from the photolysis of phenolic compounds (Anastasio et al., 1997), and the photolysis of organic
acids.
To provide a semi-quantitative context for the potential magnitude of this source, we
assume a midday OH production rate in surface seawater of 10~n M sec"1 (Zhou and Mopper,
1990) and a dissolved organic carbon enrichment of 2 to 3 orders of magnitude in sea-salt
aerosols. This yields an estimated OH production rate in fresh (alkaline) sea-salt aerosols of 10~9
to 10~8 M sec"1. As discussed above, rapid (seconds to minutes) acidification of the aerosol
AX2-48
-------
should substantially enhance these production rates. Consequently, the midday OH production
rates from marine-derived organic matter in acidified sea-salt aerosols may rival or perhaps
exceed midday OH scavenging rates from the gas phase (approximately 1CT7 M sec"1; [Chameides
and Stelson, 1992]). Scavenging is the only significant source for OH in acidified sea-salt
aerosols considered by many current models.
Limited experimental evidence indicates that these pathways are important sources of HOX
and ROX in marine air and possibly in coastal cities. For example, the absorption of solar energy
by organic species dissolved in cloud water (e.g., Faust et al., 1993; Anastasio et al., 1997) and in
deliquesced sea-salt aerosols (Anastasio et al., 1999) produces OH, HO2, and H2O2. In addition,
Fe(III) complexation by oxalate and similar ligands to metal such as iron can greatly enhance
radical production through ligand to metal charge transfer reactions (Faust, 1994; Hoigne et al.,
1994). Oxalate and other dicarboxylic anions are ubiquitous components of MBL aerosols in
both polluted (e.g., Turekian et al., 2003) and remote regions (Kawamura et al., 1996).
Substantial evidence exists for washout of peroxy radicals. Near solar noon, mixing ratios
of total HOX plus ROX radicals generally fall in the 50 ppt range, but during periods of rain these
values dropped to below the detection limit of 3 to 5 ppt (Andres Hernandez et al., 2001; Burkert
et al., 2001a,b, 2003). Such low concentrations cannot be explained by loss of actinic radiation,
because nighttime radical mixing ratios were higher.
Burkert et al. (2003) investigated the diurnal behavior of the trace gases and peroxy radicals
in the clean and polluted MBL by comparing observations to a time dependant, zero-dimensional
chemical model. They identified significant differences between the diurnal behavior of RO2*
derived from the model and that observed possibly attributable to multiphase chemistry. The
measured HCHO concentrations differed from the model results and were best explained by
reactions involving low levels of Cl.
Finally, photolytic NO3 reduction is important in the surface ocean (Zafiriou and True,
1979) and could contribute to OH production in sea-salt aerosols. Because of the
pH-dependence of HNO3 phase partitioning, most total nitrate (HNO3 + particulate NO3 ) in
marine air is associated with sea salt (e.g., Huebert et al., 1996; Erickson et al., 1999). At high
mM concentrations of NO3 in sea-salt aerosols under moderately polluted conditions (e.g.,
Keene et al., 2002) and with quantum yields for OH production of approximately 1% (Jankowski
et al., 2000), this pathway would be similar in magnitude to that associated with scavenging
AX2-49
-------
from the gas phase and with photolysis of dissolved organics. Experimental manipulations of
marine aerosols sampled under relatively clean conditions on the California coast confirms that
this pathway is a major source for OH in sea-salt solutions (Anastasio et al., 1999).
Although largely unexplored, the potential influences of these poorly characterized radical
sources on O3 cycling in marine air are probably significant. At minimum, the substantial
inferred concentrations of HO2 in aerosol solutions would diminish and perhaps reverse HO2
scavenging by marine aerosols and thereby increase O3 production relative to models based on
Jacob's (2000) recommended reaction probability.
AX2.2.10.2 NOX Chemistry
Jacob (2000) recommended as a best estimate, YN o = 0.1 for the reaction probability
of N2O5 on aqueous aerosol surfaces with conversion to HNO3. Recent laboratory studies on
sulfate and organic aerosols indicates that this reaction probability should be revised downward,
to a range 0.01-0.05 (Kane et al., 2001; Hallquist et al., 2003; Thornton et al., 2003). Tie et al.
(2003) found that a value of 0.04 in their global model gave the best simulation of observed NOX
concentrations over the Arctic in winter. A decrease in N2O5 slows down the removal of NOX
and thus increases O3 production. Based on the consistency between measurements of NOy
partitioning and gas-phase models, Jacob (2000) considers it unlikely that significant HNO3 is
recycled to NOX in the lower troposphere. However, only one of the reviewed studies (Schultz
et al., 2000) was conducted in the marine troposphere and none were conducted in the MBL.
An investigation over the equatorial Pacific reported discrepancies between observations and
theory (Singh et al., 1996b) that might be explained by HNO3 recycling. It is important to
recognize that both Schultz et al. (2000) and Singh et al. (1996b) involved aircraft sampling
which, in the MBL, significantly under represents sea-salt aerosols and thus most total NO3
(HNO3 + NO3 ) and large fractions of NOy in marine air (e.g., Huebert et al., 1996).
Consequently, some caution is warranted when interpreting constituent ratios and NOy budgets
based on such data.
Recent work in the Arctic has quantified significant photochemical recycling of NO3
to NOX and perturbations of OH chemistry in snow (Honrath et al., 2000; Dibb et al., 2002;
Domine and Shepson, 2002), which suggests the possibility of similar multiphase pathways
occurring in aerosols. As mentioned above, recent evidence also indicates that NO3 is
AX2-50
-------
photolytically reduced to NO2 (Zafariou and True, 1979) in acidic sea-salt solutions (Anastasio
et al., 1999). Further photolytic reduction of NO2" to NO (Zafariou and True, 1979) could
provide a possible mechanism for HNO3 recycling. Early experiments reported production
of NOX during the irradiation of artificial seawater concentrates containing NO3 (Petriconi and
Papee, 1972). Based on the above, we believe that HNO3 recycling in sea-salt aerosols is
potentially important and warrants further investigation. Other possible recycling pathways
involving highly acidic aerosol solutions and soot are reviewed by Jacob (2000).
Ammann et al. (1998) reported the efficient conversion of NO2 to HONO on fresh soot
particles in the presence of water. They suggest that interaction between NO2 and soot particles
may account for high mixing ratios of HONO observed in urban environments. Conversion
of NO2 to HONO and subsequent photolysis to NO + OH would constitute an NOx-catalyzed O3
sink involving snow. High concentrations of HONO can lead to the rapid growth in OH
concentrations shortly after sunrise, giving a "jump start" to photochemical smog formation.
Prolonged exposure to ambient oxidizing agents appears to deactivate this process. Broske et al.
(2003) studied the interaction of NO2 on secondary organic aerosols and concluded that the
uptake coefficients were too low for this reaction to be an important source of HONO in the
troposphere.
Choi and Leu (1998) evaluated the interactions of nitric acid on a model black carbon soot
(FW2), graphite, hexane and kerosene soot. They found that HNO3 decomposed to NO2
and H2O at higher nitric acid surface coverages, i.e., P(HNO3) > = 10~4 Torr. None of the soot
models used were reactive at low nitric acid coverages, at P(HNO3) = 5 x io~7 Torr or at lower
temperatures (220K). They conclude that it is unlikely that aircraft soot in the upper
troposphere/lower stratosphere reduces HNO3.
Heterogeneous production on soot at night is believed to be the mechanism by which
HONO accumulates to provide an early morning source of HOX in high NOX environments
(Harrison et al., 1996; Jacob, 2000). HONO has been frequently observed to accumulate to
levels of several ppb overnight, and has been attributed to soot chemistry (Harris et al., 1982;
Calvert et al., 1994; Jacob, 2000).
Longfellow et al. (1999) observed the formation of HONO when methane, propane, hexane
and kerosene soots were exposed to NO2. They estimate that this reaction may account for some
part of the unexplained high levels of HONO observed in urban areas. They comment that
AX2-51
-------
without details about the surface area, porosity and amount of soot available for this reaction,
reactive uptake values cannot reliably be estimated. They comment that soot and NO2 are
produced in close proximity during combustion, and that large quantities of HONO have been
observed in aircraft plumes.
Saathoff et al. (2001) studied the heterogeneous loss of NO2, HNO3, NO3/N2O5,
HO2/HO2NO2 on soot aerosol using a large aerosol chamber. Reaction periods of up to several
days were monitored and results used to fit a detailed model. They derived reaction probabilities
at 294 K and 50% RH for NO2, NO3, HO2 and HO2NO2 deposition to soot, HNO3 reduction
to NO2, and N2O5 hydrolysis. When these probabilities were included in photochemical box
model calculations of a 4-day smog event, the only noteworthy influence of soot was a 10%
reduction in the second day O3 maximum, for a soot loading of 20 jig m"3, i.e., a factor of 2 to
10 times observed black carbon loadings seen during extreme U.S. urban pollution events,
although such concentrations are observed routinely in the developing world.
Mufioz and Rossi (2002) conducted Knudsen cell studies of HNO3 uptake on black and
grey decane soot produced in lean and rich flames, respectively. They observed HONO as the
main species released following nitric acid uptake on grey soot, and NO and traces of NO2 from
black soot. They conclude that these reactions would only have relevance in special situations in
urban settings where soot and HNO3 are present in high concentrations simultaneously.
AX2.2.10.3 Halogen Radical Chemistry
Barrie et al. (1988) first suggested that halogen chemistry on snow surfaces in the Arctic
could lead to BrOx formation and subsequent O3 destruction. More recent work suggests that
halogen radical reactions may influence O3 chemistry in mid latitudes as well.
The weight of available evidence supports the hypothesis that halogen radical chemistry
significantly influences O3 cycling over much of the marine boundary layer at lower latitudes
and in at least some other regions of the troposphere. However, proposed chemical mechanisms
are associated with substantial uncertainties and, based on available information, it appears
unlikely that a simple parameterization (analogous to those recommended by Jacob (2000) for
other multiphase transformations) would adequately capture major features of the underlying
transformations.
AX2-52
-------
Most of the Cl and Br in the marine boundary layer are produced in association with
sea-salt aerosols by wind stress at the ocean surface (e.g., Gong et al., 1997). Fresh aerosols
rapidly dehydrate towards equilibrium with ambient water vapor and undergo other chemical
processes involving the scavenging of reactive gases, aqueous-phase transformations, and
volatilization of products. Many of these processes are strongly pH-dependent (Keene et al.,
1998). Throughout most of the marine boundary layer, sea-salt alkalinity is tritrated rapidly
(seconds to minutes) by ambient acids (Chameides and Stelson, 1992; Erickson et al., 1999) and,
under a given set of conditions, the pHs of the super-jam, sea-salt size fractions are buffered to
similar values via HC1 phase partitioning (Keene and Savoie, 1998; 1999; Keene et al., 2002).
Model calculations based on the autocatalytic halogen activation mechanism (Vogt et al.,
1996; Keene et al., 1998; Sander et al., 1999; von Glasow et al., 2002a,b; Pszenny et al., 2004;
Sander et al., 2003) predict that most particulate Br associated with acidified sea-salt aerosol
would react to form Br2 and BrCl, which subsequently volatilize and photolyze in sunlight to
produce atomic Br and Cl. Most Br atoms recycle in the gas phase via
•Br + O3 -> BrO + O2 (AX2-42)
BrO + HO2' ->> HOBr + O2 (AX2-43)
HOBr + hv -^ 'OH + 'Br (AX2-44)
and thereby catalytically destroy O3, analogous to Br cycling in the stratosphere (e.g.,
Mozurkewitch, 1995; Sander and Crutzen, 1996). Side reactions with HCHO and other
compounds produce HBr, which is either scavenged and recycled through the aerosol or lost to
the surface via wet and dry deposition (Dickerson et al., 1999).
Cl-radical chemistry influences O3 in two ways (e.g., Pszenny et al., 1993). Some atomic
Cl in marine air reacts directly with O3 initiating a catalytic sequence analogous to that of Br
(AX2-42 through AX2-44 above). However, most atomic Cl in the MBL reacts with
hydrocarbons (which, relative to the stratosphere, are present at high concentrations) via
hydrogen extraction to form HC1 vapor. The enhanced supply of odd hydrogen radicals from
AX2-53
-------
hydrocarbon oxidation leads to O3 production in the presence of sufficient NOX. Thus, Cl
chemistry represents a modest net sink for O3 when NOX is less than 20 pptv and a net source at
higher NOX. Although available evidence suggest that significant Cl-radical chemistry occurs in
clean marine air, its net influence on O3 appears to be small relative to that of Br and I.
In addition to Br and Cl, several lines of recent evidence suggests that an autocatalytic
cycle also sustains I-radical chemistry leading to significant net O3 destruction in marine air
(Vogt et al., 1996, 1999; von Glasow et al., 2002a). The cycle is initiated by photolysis of
organoiodine compounds emitted from the ocean surface to generate atomic I (Carpenter et al.,
1999). Iodine atoms react almost exclusively with O3 to form IO. Most IO photodissociates in
sunlight to generate I and atomic O, which rapidly recombines with O2 to form O3.
Consequently, this cycle has no net effect on O3 (Stutz et al., 1999). However, alternative
reaction pathways analogous to reactions AX2-42 through AX2-44 above lead to catalytic O3
destruction. Model calculations suggest that HOI recycles via acid-catalyzed aerosol scavenging
to form IC1 and IBr, which subsequently volatilize and photolyze to form halogen atoms. The
net effect of this multiphase pathway is to increase concentrations of volatile reactive I. The self
reaction of IO to form I and OIO may further enhance O3 destruction (Cox et al., 1999;
Ashworth et al., 2002). IO also reacts with NO2 to form INO3, which can be scavenged by
aqueous aerosols. This pathway has been suggested as a potentially important sink for NOX in
the remote MBL and would, thus, contribute indirectly to net O3 destruction (McFiggans et al.,
2000).
Various lines of observational evidence support aspects of the above scenarios. Most
measurements of particulate Br in marine air reveal large depletions relative to conservative sea-
salt tracers (e.g., Sander et al., 2003) and, because HBr is highly soluble in acidic solution, these
deficits cannot be explained by simple acid-displacement reactions (e.g., Ayers et al., 1999).
Observed Br depletions are generally consistent with predictions based on the halogen activation
mechanism. In contrast, available, albeit limited, data indicate that I is highly enriched in marine
aerosols relative to bulk seawater (e.g., Sturges and Barrie, 1988), which indicates active
multiphase iodine chemistry.
Direct measurements of BrO in marine air by differential optical absorption spectroscopy
(DOAS) reveal mixing ratios that are near or below analytical detection limits of about 1 to 3 ppt
(Honninger, 1999; Pszenny et al., 2003; Leser et al., 2003) but within the range of model
AX2-54
-------
predictions. Column-integrated DOAS observations from space reveal substantial mixing ratios
of tropospheric BrO (e.g., Wagner and Platt, 1998). Although the relative amounts in the MBL
cannot be resolved, these data strongly suggest active destruction of tropospheric O3 via the
reaction sequence of AX2-42 through AX2-44. Similarly, measurements of IO (McFiggans
et al., 2000) and OIO (Allan et al., 2001) indicate active O3 destruction by an analogous pathway
involving atomic I. In addition, anticorrelations on diurnal time scales between total volatile
inorganic Br and particulate Br and between volatile inorganic I and particulate I have been
reported (e.g., Rancher and Kritz, 1980; Pszenny et al., 2004). Although the lack of speciation
precludes unambiguous interpretation, these relationships are also consistent with predictions
based on the halogen activation mechanism.
Large diurnal variabilities in O3 measured over the remote subtropical Atlantic and Indian
Oceans (Dickerson et al., 1999; Burkert et al., 2003) and early morning depletions of O3
observed in the remote temperate MBL (Galbally et al., 2000) indicate that only about half of the
inferred O3 destruction in the MBL can be explained by conventional HOX/NOX chemistry.
Model calculations suggest that Br and I radical chemistry could account for a "missing" O3
sink of this magnitude (Dickerson et al., 1999; Stutz et al., 1999; McFiggans et al., 2000; von
Glasow et al., 2002b). In addition to the pathway for O3 destruction given by R AX2-39 to R
AX2-41, in areas with high concentrations of halogen radicals the following generic loss
pathways for O3 can occur in the Arctic at the onset of spring and also over salt flats near the
Dead Sea (Hebestreit et al., 1999) and the Great Salt Lake (Stutz et al., 2002) analogous to their
occurrence in the lower stratosphere (Yung et al., 1980).
X = Br,Cl,I) (AX2-45)
Y + 03 -» YO + 02 (Y = Br, Cl, I) (AX2-46)
2 (AX2-47)
Net: 203^302 (AX2-48)
Note that the self reaction of CIO radicals is likely to be negligible in the troposphere. There are
three major reaction pathways involved in reaction AX2-47. Short-lived radical species are
AX2-55
-------
produced. These radicals rapidly react to yield monoatomic halogen radicals. In contrast to the
situation in marine air, where DOAS measurements indicate BrO concentrations of 1 to 3 ppt,
Stutz et al. (2002) found peak BrO concentrations of about 6 ppt and peak CIO concentrations of
about 15 ppt. They also derived a correlation coefficient of -0.92 between BrO and O3 but much
smaller values of r between CIO and O3. Stutz et al. attributed the source of the reactive
halogens to concentrated high molality solutions or crystalline salt around salt lakes, conditions
that do not otherwise occur in more dilute or ocean salt water. They also suggest that halogens
may be released from saline soils. The inferred atmospheric concentrations of Cl are about
105/cm3, or about a factor of 100 higher than found in the marine boundary layer by Rudolph
et al. (1997) indicating that, under these conditions, the Cl initiated oxidation of hydrocarbons
could be substantial.
Most of the well-established multiphase reactions tend to reduce the rate of O3 formation in
the polluted troposphere. Direct reactions of O3 and atmospheric particles appears to be too slow
to reduce smog significantly. Removal of HO2 onto hydrated particles will decrease the
production of O3 by the reaction of HO2 with NO. The uptake of NO2 and HNO3 will also result
in the production of less O3. Conditions leading to high concentrations of Br, Cl, and I radicals
can lead to O3 loss. The oxidation of hydrocarbons (especially alkanes) by Cl radicals,
in contrast, may lead to the rapid formation of peroxy radicals and faster smog production in
coastal environments where conditions are favorable for the release of gaseous Cl from the
marine aerosol. There is still considerable uncertainty regarding the role of multiphase processes
in tropospheric photochemistry and so results should be viewed with caution and an appreciation
of their potential limitations.
AX2.2.10.4 Reactions on the Surfaces of Crustal Particles
Field studies have shown that O3 levels are reduced in plumes containing high particle
concentrations (e.g., DeReus et al.; 2000; Berkowitz et al., 2001; Gaffney et al., 2002).
Laboratory studies of the uptake of O3 on untreated mineral surfaces (Hanisch and Crowley,
2003; Michel et al., 2002, 2003) have shown that O3 is lost by reaction on these surfaces and this
loss is catalytic. Values of y of 1.2 ± 0.4 x io~4 were found for reactive uptake on a-A!2O3
and 5 ± 1 x icr5 for reactive uptake on SiO2 surfaces. Usher et al. (2003) found mixed behavior
for O3 uptake on coated surfaces with respect to untreated surfaces. They found that y drops
AX2-56
-------
from 1.2 ± 0.4 x io~4 to 3.4 ± 0.6 x io~5 when a-A!2O3 surfaces are coated with NO3 derived
from HNO3, whereas they found that y increases to 1.6 ± 0.2 x 10~4 after these surfaces have
been pretreated with SO2. Usher et al. also pretreated surfaces of SiO2 with either a C8-alkene or
a C8-alkane terminated organotrichlorosilane. They found that y increased to 7 ± 2 x 10~5 in the
case of treatment with the alkene, but that it decreased to 3 ± 1 x io~5 for treatment with the
alkane. Usher et al. (2003) suggested, on the basis of these results that mineral dust particles
coated with nitrates or alkanes will affect O3 less than dust particles that have accumulated
coatings of sulfite or alkenes. These studies indicate the importance of aging of airborne
particles on their ability to take up atmospheric gases. Reactions such as these may also be
responsible for O3 depletions observed in dust clouds transiting the Pacific Ocean.
Underwood et al. (2001) studied the uptake of NO2 and HNO3 on the surfaces of dry
mineral oxides (containing Al, Ca, Fe, Mg, Si and Ti) and naturally occurring mineral dust.
A wide range of values of y(NO2) were found, ranging from < 4 x 10"10 for SiO2 to 2 x 10~5 for
CaO, with most other values ~10~6. Values of y for Chinese loess and Saharan dust were also of
the order of 10~6. They found that as the reaction of NO2 proceeds on the surfaces that reduction
to NO occurs. They recommended a value of y for HNO3 of about 1 x 10~3. Not surprisingly,
the values of y increased from those given above if the surfaces were wetted. Underwood et al.
(2001) also suggested that the uptake of NO2 was likely to be only of marginal importance but
that uptake of HNO3 could be of significance for photochemical oxidant cycles.
Li et al. (2001) examined the uptake of acetaldehyde, acetone and propionaldehyde on the
same mineral oxide surfaces listed above. They found that these compounds weakly and
reversibly adsorb on SiO2 surfaces. However, on the other oxide surfaces, they irreversibly
adsorb and can form larger compounds. They found values of y ranging from 10~6 to 10~4.
These reactions may reduce O3 production efficiency in areas of high mineral dust concentration
such as the American Southwest or in eastern Asia as noted earlier.
AX2.2.10.5 Reactions on the Surfaces of Aqueous H2SO4 Solutions
The most recent evaluation of Photochemical and Chemical Data by the Jet Propulsion
Laboratory (Jet Propulsion Laboratory, 2003) includes recommendations for uptake coefficients
of various substances on a variety of surfaces including aqueous H2SO4 solutions. Although
much of the data evaluated have been obtained mainly for stratospheric applications, there are
AX2-57
-------
studies in which the range of environmental parameters is compatible with those found in the
troposphere. In particular, the uptake of N2O5 on the surface of aqueous H2SO4 solutions has
been examined over a wide range of values. Typical values of y are of the order of 0.1 (e.g., Jet
Propulsion Laboratory, 2003). Values of y for NO2 are much lower (5 x 10"7 to within a factor
of three) and thus the uptake of NO2 on the surface of aqueous H2SO4 solutions is unlikely to be
of importance for oxidant cycles. The available data indicate that uptake of OH and HO2
radicals could be significant under ambient conditions with values of y of the order of 0.1 or
higher for OH, and perhaps similar values for HO2.
AX2.2.10.6 Oxidant Formation in Particles
Water is a major component of submicron particles in the atmosphere. However,
photochemical reactions in particles have not been studied to the same extent as they have in
hydrometeors (e.g., Lelieveld and Crutzen, 1991). Friedlander and Yeh (1998) point out
that H2O2 and hydroxymethylhydroperoxide (HOCH2OOH) are especially likely to be found in
the aqueous component of atmospheric particles, based on observed gas-phase concentrations
and Henry's law solubility data; the concentrations in particles could be higher if the condensed
hydroperoxides form peroxyhydrate complexes (Wexler and Sarangapani, 1998). Laboratory
studies have found that UV irradiation of dissolved organic carbon (DOC) in collected
cloudwater samples is a source of free radicals to the aqueous phase (Faust et al., 1992, 1993)
but the mechanisms involved and the atmospheric fate of these radicals are unclear. Chemical
reactions involving dissolved transition metal ions could also provide significant sources of
radicals in particles (Jacob, 2000). However, only about 10 to 15% of the mass of organic
compounds in particles are quantified typically, but many of the compounds, in particular
aldehydes, could photolyze to produce free radicals. There are three basic mechanisms for the
formation of SOA (Pandis et al., 1992; Seinfeld and Pankow, 2003). These are (1) condensation
of oxidized end-products of photochemical reactions (e.g., ketones, aldehydes, organic acids, and
hydroperoxides), (2) adsorption of semivolatile organic compounds (e.g., polycyclic aromatic
hydrocarbons) onto existing organic particles, and (3) dissolution of water-soluble gases that can
then undergo subsequent reactions in particles (e.g., aldehydes). The first and third mechanisms
are expected to be of major importance during the summer when photochemistry is at its peak.
Information about the chemistry of formation of secondary organic aerosol (SOA) was reviewed
AX2-58
-------
in Section 3.3.1 and available information about the composition of organic compounds in
particles was summarized in Appendix 3C of the latest PM AQCD (U.S. Environmental
Protection Agency, 2004). It should be noted that reactive oxygen species (ROS) are also
formed during the production of SO A. In addition, polymerization of the products of the
reaction of a-pinene with O3 occurs (Tolocka et al., 2004).
Recent measurements of aerosol-phase ROS in Rubidoux, CA and New York City have
revealed relatively high concentrations, of the order of 5 to 6 x 10~7 in Rubidoux and 1 x 10~7 M
nT3 in New York City, expressed as equivalent H2O2 (Venkatachari et al., 2005a,b). The ROS
were found in particles of all sizes, with particularly high concentrations in the ultrafme range.
However, this finding could result from the condensation of vapors onto particles occurring
during adiabatic expansion in the nano stages of the sampler. A weak correlation was found
with O3, but large ROS concentrations were still found at night and in winter. The composition
and sources of the ROS are not clear. Millimolar concentrations of hydroperoxides, as estimated
by Friedlander and Yeh (1998), would contribute only 10~12 M nT3 based on a typical liquid
water volume fraction in air of 10~9. Formation of peroxyhydrates would lead to higher values
but would have to be very large to account for the ROS observations. Ozone and PAN are
orders of magnitude less water-soluble than the hydroperoxides (Jacob, 2000) and would not
contribute significantly to the ROS. Radical oxidants (e.g., OH or the superoxide ion O2 ) do not
seem to be present in sufficient abundance in the atmosphere to possibly account for the ROS
(Jacob, 2000). Low-volatility organic peroxides produced from the oxidation of large
substituted organic compounds could possibly make a major contribution. Formation of these
peroxides in the aerosol phase could be facilitated by photochemical reactions of dissolved
organic components (Anastasio et al., 1997) and by reactions of transition metals (Jacob, 2000).
Transition metals participate in the Haber-Weiss set of reactions, including Fenton's reaction,
generating free radicals from hydrogen peroxide even in the dark. ROS would also be formed in
the reaction of isoprene with O3 (Claeys et al., 2004b).
Reactions of ozone with monoterpenes have been shown to produce oxidants in the aerosol
phase. Docherty et al. (2005) found evidence for the substantial production of organic
hydroperoxides in SOA resulting from the reaction of monoterpenes with O3. Analysis of the
SOA formed in their environmental chamber indicated the SOA was mainly organic
hydroperoxides. In particular, they obtained yields of 47% and 85% of organic peroxides from
AX2-59
-------
the oxidation of a- and p-pinene. The hydroperoxides then react with aldehydes in particles to
form peroxyhemiacetals, which can either rearrange to form other compounds such as alcohols,
esters, and acids or revert back to the hydroperoxides. The aldehydes may be produced in
addition to the hydroperoxides during the oxidation of the monoterpenes. Monoterpenes also
react with OH radicals resulting, however, in the production of more lower molecular weight
products than in their reaction with O3. Bonn et al. (2004) estimated that hydroperoxides lead to
63% of global SOA formation from the oxidation of terpenes. The oxidation of anthropogenic
aromatic hydrocarbons by OH radicals may also produce organic hydroperoxides in SOA
(Johnson et al., 2004). Although the results of chamber and modeling studies indicate
substantial production of organic hydroperoxides, it should be noted that data for organic
hydroperoxides in ambient aerosol samples are sparse.
AX2.2.10.7 Ozone Reaction Indoors
Ozone chemical reactions indoors are analogous to those occurring in ambient air. Ozone
reacts with unsaturated VOCs in the indoor environment, primarily terpenes or terpene-related
compounds from cleaning products, air fresheners, and wood products. The reactions are
dependent on the O3 indoor concentration, the indoor temperature and, in most cases, the air
exchange rate/ventilation rate. Some of the reaction products may more negatively impact
human health and artifacts in the indoor environment than their precursors (Wolkoff et al., 1999;
Wilkins et al., 2001; Weschler et al., 1992; Weschler and Shields, 1997; Rohr et al., 2002;
N0jgaard et al., 2005). Primary reaction products are Criegee biradicals, nitrate radicals, and
peroxyacetyl radicals. Secondary reaction products are hydroxy, alkyl, alkylperoxy,
hydroperoxy, and alkoxy radicals. Reactions with alkenes can produce aldehydes, ketones, and
organic acids (Weschler and Shields, 2000; Weschler et al., 1992).
AX2.3 PHYSICAL PROCESSES INFLUENCING THE ABUNDANCE
OF OZONE
The abundance and distribution of O3 in the atmosphere is determined by complex
interactions between meteorology and chemistry. This section will address these interactions,
based mainly on the results of field observations. The importance of a number of transport
AX2-60
-------
mechanisms, whose understanding has undergone significant advances since the last AQCD
for O3, will be discussed in this section.
Major episodes of high O3 concentrations in the eastern United States and in Europe are
associated with slow moving, high pressure systems. High pressure systems during the warmer
seasons are associated with the sinking of air, resulting in warm, generally cloudless conditions,
with light winds. The sinking of air results in the development of stable conditions near the
surface which inhibit or reduce the vertical mixing of O3 precursors. The combination of
inhibited vertical mixing and light winds minimizes the dispersal of pollutants emitted in urban
areas, allowing their concentrations to build up. Photochemical activity involving these
precursors is enhanced because of higher temperatures and the availability of sunlight. In the
eastern United States, high O3 concentrations during a large scale episode can extend over a
hundred thousand square kilometers for several days. These conditions have been described in
greater detail in AQCD 96. The transport of pollutants downwind of major urban centers is
characterized by the development of urban plumes. However, the presence of mountain barriers
can limit mixing as in Los Angeles and Mexico City and will result in even longer periods and a
higher frequency of days with high O3 concentrations. Ozone concentrations in southern urban
areas, such as Houston, TX and Atlanta, GA tend to follow this pattern and they tend to decrease
with increasing wind speed. In northern cities, like Chicago, IL; New York, NY; and Boston,
MA the average O3 concentrations over the metropolitan areas increase with wind speed
indicating that transport of O3 and its precursors from upwind areas is important (Husar and
Renard, 1998; Schichtel and Husar, 2001).
Aircraft observations indicate that there can be substantial differences in mixing ratios of
key species between the surface and the atmosphere above (Fehsenfeld et al., 1996a; Berkowitz
and Shaw, 1997). Convective processes and small scale turbulence transport O3 and other
pollutants both upward and downward throughout the planetary boundary layer and the free
troposphere. Ozone and its precursors were found to be transported vertically by convection into
the upper part of the mixed layer on one day, then transported overnight as a layer of elevated
mixing ratios and then entrained into a growing convective boundary layer downwind and
brought back down to the surface. High concentrations of O3 showing large diurnal variations at
the surface in southern New England were associated with the presence of such layers
(Berkowitz et al., 1998). Because of wind shear, winds several hundred meters above the ground
AX2-61
-------
can bring pollutants from the west, even though surface winds are from the southwest during
periods of high O3 in the eastern United States (Blumenthal et al., 1997). Low level nocturnal
jets can also transport pollutants hundreds of kilometers. Turbulence associated with them can
bring these pollutants to the surface and in many locations result in secondary O3 maxima in the
early morning (Corsmeier et al., 1997). Based on analysis of the output of model studies
conducted by Kasibhatla and Chameides (2000), Hanna et al. (2001) concluded that O3 can be
transported over thousands of kilometers in the upper boundary layer of the eastern half of the
United States during specific O3 episodes.
Stratospheric-tropospheric exchange (STE) will be discussed in Section AX2.3.1. The
vertical redistribution of O3 and other pollutants by deep, or penetrating convection is discussed
in Section AX2.3.2. The potential importance of transport of O3 and precursors by low-level jets
is the topic of Section AX2.3.3. Issues related to the transport of O3 from North America are
presented in Section AX2.3.4. Relations of O3 to solar ultraviolet radiation and temperature will
then be discussed in Section AX2.3.5.
AX2.3.1 Stratospheric-Tropospheric Ozone Exchange (STE)
In the stratosphere, O3 formation is initiated by the photodissociation of molecular
oxygen (O2) by solar ultraviolet radiation at wavelengths less than 242 nm. Almost all of this
radiation is absorbed in the stratosphere (except for regions near the tropical tropopause),
preventing this mechanism from occurring in the troposphere. Some of the O3 in the
stratosphere is transported downward into the troposphere. The potential importance of this
source of tropospheric O3 has been recognized since the early work of Regener (1941), as cited
by Junge (1963). Stratospheric-tropospheric exchange (STE) of O3 and stratospheric
radionuclides produced by the nuclear weapons tests of the 1960s is at a maximum during late
winter and early spring (e.g., Ludwig et al., 1977). Since AQCD 96 on O3 substantial new
information from numerical models, field experiments and satellite-based observations has
become available. The following sections outline the basic atmospheric dynamics and
thermodynamics of stratosphere/troposphere exchange and review these new developments.
There are several important mechanisms for injecting stratospheric O3 into the troposphere,
they include tropopause folds (Reed, 1955; Danielsen, 1968), cutoff lows (Price and Vaughan,
1993), clear air turbulence, mesoscale convective complexes and thunderstorms, breaking
AX2-62
-------
gravity waves (Poulida et al., 1996; Langford and Reid, 1998; Stohl et al., 2003) and streamers.
Streamers are dry, stratospheric intrusions visible in satellite water vapor imagery that are
sheared into long filamentary structures that often roll into vortices and exhibit visible evidence
of the irreversible mixing of moist subtropical tropospheric and dry polar stratospheric air
(Appenzeller et al., 1996; Wimmers et al., 2003). They are often present at a scale that eludes
capture in large scale dynamical models of the atmosphere that cannot resolve features less than
1 degree (-100 km). Empirical evidence for stratospheric intrusions comes from observations of
indicators of stratospheric air in the troposphere. These indicators include high potential
vorticity, low water vapor mixing ratios, high potential temperature, enhancements in the ratio
of 7Be to 10Be in tropospheric aerosols, as well as enhancements in O3 mixing ratios and total
column amounts. These quantities can be observed with in situ aircraft and balloons, as well as
remotely sensed from aircraft and ground-based lidars and both geostationary and polar (low
earth orbiting) space platforms.
The exchange of O3 between the stratosphere and the troposphere in middle latitudes
occurs to a major extent by tropopause folding events (Reiter, 1963, 1975; Reiter and Mahlman,
1965; Danielsen, 1968, 1980; Danielsen and Mohnen, 1977). The term, tropopause folding is
used to describe a process in which the tropopause intrudes deeply into the troposphere along a
sloping frontal zone bringing air from the lower stratosphere with it. Tropopause folds occur
with the formation of upper level fronts associated with transverse circulations that develop
around the core of the polar jet stream. South of the jet stream core, the tropopause is higher
than to the north of it. The tropopause can be imagined as wrapping around the jet stream core
and folding beneath it and extending into the troposphere (cf, Figure AX2-7a). Although drawn
as a heavy solid line, the tropopause should not be imagined as a material surface, through which
there is no exchange. Significant intrusions of stratospheric air occur in "ribbons" -200 to 1000
km in length, 100 to 300 km wide and about 1 to 4 km thick (Hoskins, 1972; Wimmers et al.,
2003). These events occur throughout the year and their location follows the seasonal
displacement of the polar jet stream.
The seasonal cycle of O3 exchange from the stratosphere into the troposphere is not caused
by a peak in the seasonal cycle of upper tropospheric cyclone activity. Instead, it is related to the
large scale pattern of tracer transport in the stratosphere. During winter in the Northern
Hemisphere, there is a maximum in the poleward, downward transport of mass, which moves O3
AX2-63
-------
o>
(A
(A
0)
150
200
300
400
500
600
700
800
900
1000
330
320
360
350
340
330
13
12
11
10
9
8 §
7 £
O)
6 '55
I
5
4
3
2
1
121° VBG
119°
SAN 117°
UCC 115° ELY
113°
Figure AX2-7a. Cross section through a tropopause folding event on March 13,1978 at
0000 GMT. Potential temperatures (K) are represented by thin solid
lines. Wind speeds (m s'1) are given by thin dashed lines. The hatched
area near the center of the figure indicates the location of the jet stream
core. The tropopause defined by a potential vorticity of 100 x 10~7 K
mb"1 s"1 is shown as the heavy solid line. The two Sabreliner flight tracks
through this cross section are shown as a heavy solid line with filled
arrows and heavy dashed line with open arrows. Longitude is shown
along the x-axis . Upper air soundings were taken at Vandenberg AFB,
CA (VBG); San Diego, CA (SAN); Winnemucca, NV (UCC); and Ely, NV
(ELY).
Source: Adapted from Shapiro (1980).
from the tropical upper stratosphere to the lower stratosphere of the polar- and midlatitudes.
This global scale pattern is controlled by the upward propagation of large-scale and small-scale
waves generated in the troposphere. As the energy from these disturbances dissipates, it drives
this stratosphere circulation. As a result of this process, there is a springtime maximum in the
total column abundance of O3 over the poles. The concentrations of O3 (and other trace
AX2-64
-------
substances) build up in the lower stratosphere until their downward fluxes into the lower
stratosphere are matched by increased fluxes into the troposphere. Thus, there would be a
springtime maximum in the flux of O3 into the troposphere even if the flux of stratospheric air
through the tropopause by tropopause folding remained constant throughout the year (Holton
et al., 1995). Indeed, cyclonic activity in the upper troposphere is active throughout the entire
year in transporting air from the lower stratosphere into the troposphere (Mahlman, 1997).
Oltmans et al. (1996) and Moody et al. (1996) provide evidence that stratospheric intrusions
contribute to the O3 abundance in the upper troposphere over the North Atlantic even during
the summer.
There are a number of techniques that have been used to quantify the amount of O3 in the
free troposphere or even the amount of O3 reaching the surface that can be attributed to
downward transport from the stratosphere. Earlier work, cited in AQCD 96 relied mainly on the
use of 7Be as a tracer of stratospheric air. However, its use is ambiguous because it is also
formed in the upper troposphere. Complications also arise because its production rate is also
sensitive to solar activity (Lean, 2000). The ratio of 7Be to 10Be provides a much more sensitive
tracer of stratospheric air than the use of 7Be alone (Jordan et al., 2003). More recent work than
cited in AQCD 96 has focused on the use of potential vorticity (PV) as a tracer of stratospheric
air. Potential vorticity is a dynamical tracer used in meteorology. Generally, PV is calculated
from wind and temperature observations and represents the rotational tendency of a column of
air weighted by the static stability, which is just the distance between isentropic surfaces. This
quantity is a maximum in the lower stratosphere where static stability is great and along the jet
stream where wind shear imparts significant rotation to air parcels. As air moves from the
stratosphere to the troposphere, PV is conserved, and therefore it traces the motion of O3. The
static stability is lower in the troposphere, so to preserve PV, fluid rotation will increase. This is
why STE is associated with cyclogenesis, or the formation of storms along the polar jet stream.
Dynamical models clearly capture this correspondence between the location of storm tracks and
preferred regions for STE. However, because PV is destroyed at a faster rate with increasing
depth, it is not useful as a tracer of stratospheric air reaching the surface. Appenzeller et al.
(1996) found that maps of PV coupled with satellite images of humidity can provide indications
of the intrusion of stratospheric air into the troposphere, however, they had no measurements of
O3. Even if measurements of O3 were available, the extrapolation of any relations to other events
AX2-65
-------
would still be problematic as Olsen et al. (2002) have noted that there are seasonal and
geographic variations in the relation between O3 and PV. Recent flights of the NCAR C130
during the TOPSE campaign measured in situ O3, and curtains of O3 above and below the
aircraft observed with a lidar and clearly showed a correspondence between high PV and
stratospheric levels of O3 and satellite depictions of dry air indicating the presence of tropopause
folding (Wimmers and Moody, 2004a,b).
Detailed cross sections through a tropopause folding event showing atmospheric structure,
O3 mixing ratios and condensation nuclei (CN) counts are given in Figures AX2-7a, AX2-7b,
and AX2-7c (Shapiro, 1980). Flight tracks of an NCAR Sabreliner obtaining data through the
tropopause fold are also shown. The core of the jet stream is indicated by the hatched area near
the center of Figure AX2-7a. As can be seen from Figure AX2-7 a and b, there is a strong
relation between the folding of the tropopause, indicated by the heavy solid line and O3. CN
counts during the portions of the flights in the lower troposphere were tropospheric were
typically of the order of several * 10 3 cm"3 and 100 or less in the stratospheric portion.
However, it is clear that CN counts in the fold are much higher than in the stratosphere proper,
suggesting that there was active mixing between tropospheric and stratospheric air in the fold.
Likewise, it can also be seen from Figure AX2-7b that O3 is being mixed outside the fold into the
middle and upper troposphere. The two data sets shown in Figures AX2-7b and 7c indicate that
small scale turbulent processes were occurring to mediate this exchange and that the folds are
mixing regions whose chemical characteristics lie between those of the stratosphere and the
troposphere (Shapiro, 1980). Chemical interactions between stratospheric and tropospheric
constituents are also possible within tropopause folds. These considerations also imply that in
the absence of turbulent mixing, tropopause folding can be a reversible process.
Several recent papers have attempted to demonstrate that the atmosphere is a fluid
composed of relatively distinct airstreams with characteristic three-dimensional motions and
corresponding trace gas signatures. Based on aircraft observations, satellite imagery, and back
trajectories, it has been shown that dry airstreams, or dry intrusions (DA or DI) always advect
stratospheric O3 into the middle and upper troposphere (Cooper et al, 2001; Cooper et al.,
2002a), however the seasonal cycle of O3 in the lowermost stratosphere allows greater quantities
of O3 to enter the troposphere during spring (Cooper et al., 2002b). Other work has focused on
the signatures of PV to show specific instances of STE (Olsen and Stanford, 2001). This
AX2-66
-------
150
1000
121° VBG
119°
SAN 117°
UCC 115°
ELY
113°
Figure AX2-7b. Ozone mixing ratios pphm (parts per hundred million) corresponding
to Figure AX2-7a. The two Sabreliner flight tracks through this cross
section are shown as a heavy solid line with filled arrows and a heavy
dashed line with open arrows. Longitude is shown along the x-axis.
Upper air soundings were taken at Vandenberg AFB, CA (VBG);
San Diego, CA (SAN); Winnemucca, NV (UCC); and Ely, NV (ELY).
Source: Adapted from Shapiro (1980).
correlation between TOMS gradients and PV was also used to derive the annual mass flux of O3
from STE and generated an estimate somewhat higher (500 Tg/yr over the Northern
Hemisphere) than the estimates of most general circulation models. The IPCC has reported a
large range of model estimates of STE, expressed as the net global flux of O3 in Tg/yr, from a
low of 390 to a high of 1440 (reproduced as Table AX3-15). A few other estimates have been
made based on chemical observations in the lower stratosphere, or combined chemistry and
dynamics (450 Tg/yr Murphy and Fahey, 1994; 510 Tg/yr extratropics only, Gettleman et al.,
1997; and 500 Tg/yr midlatitude NH only (30 to 60N) (Olsen et al., 2002). These values
illustrate the large degree of uncertainty that remains in quantifying this important source of O3.
AX2-67
-------
150
121° VBG
119°
SAN 117°
UCC 115° ELY
113°
Figure AX2-7c. Condensation nuclei concentrations (particles cm"3) corresponding to
Figure AX2-7a. The two flight Sabreliner flight tracks through this cross
section are shown as a heavy solid line with filled arrows and a heavy
dashed line with open arrows. Longitude is shown along the x-axis.
Upper air soundings were taken at Vandenberg AFB, CA (VBG);
San Diego, CA (SAN); Winnemucca, NV (UCC); and Ely, NV (ELY).
Source: Adapted from Shapiro (1980).
Based on the concept of tracing airstream motion, a number of Lagrangian model studies
have resulted in climatologies that have addressed the spatial and temporal variability in
stratosphere to troposphere transport (Stohl, 2001; Wernli and Borqui, 2002; Seo and Bowman,
2002; James et al., 2003a,b; Sprenger and Wernli, 2003; Sprenger et al., 2003). Both Stohl
(2001) and Sprenger et al. (2003) produced one year climatologies of tropopause folds based on
a 1° by 1° gridded meteorological data set. They each found the probability of deep folds
(penetrating to the 800 hPa level) was a maximum during winter (December through February).
The highest frequency of folding extended from Labrador down the east coast of North America.
AX2-68
-------
However, these deep folds occurred less than 1% of the six hour intervals for which
meteorological data is assimilated for grid points in the United States. They observed a higher
frequency of more shallow folds (penetrating to the upper troposphere) and medium folds
(penetrating to levels between 500 and 600 hPa) of about 10% and 1 to 2% respectively. These
events occur preferentially across the subtropics and the southern United States. At higher
latitudes other mechanisms such as the erosion of cutoff lows and the breakup of stratospheric
streamers are likely to play an important role in STE. Stohl (2001) also described the region of
strong stirring in the upper extratropical troposphere related to the midlatitude storm tracks.
Stohl (2001) demonstrated that airstreams with strong vertical motion are all highly incoherent,
they stir their air parcels into a new environment, producing filamentary tracer structures and
paving the way for subsequent mixing. A 15-year climatology by Sprenger and Wernli (2003)
shows the consistent pattern of STE occurring over the primary storm tracks in the Pacific and
Atlantic along the Asian and North American coasts. This climatology, and the one of James
et al. (2003a,b) both found that recent stratospheric air associated with deep intrusions are
relatively infrequent occurrences in these models. Thus, stratospheric intrusions are most likely
to directly affect the middle and upper troposhere and not the planetary boundary layer.
However, this O3 can still exchange with the planetary boundary layer through convection as
described later in this subsection and in Section AX2.3.2, AX2.3.3 and AX2.3.4.
Interannual variations in STE are related to anomalies in large-scale circulation such as the
North Atlantic Oscillation which causes changes in storm track positions and intensities, and the
El Nino-Southern Oscillation, which results in anomalous strong convection over the eastern
Pacific (James et al., 2003a,b). It should also be remembered that the downward flux of O3 into
the troposphere is related to the depletion of O3 within the wintertime stratospheric polar
vortices. The magnitude of this depletion and the transport of O3 depleted air to midlatitudes in
the stratosphere (Mahlman et al., 1994; Hadjinicolaou and Pyle, 2004) shows significant
interannual variability which may also be reflected in the downward flux of O3 into the
troposphere. All of these studies, from the analysis of individual events to multiyear
climatologies are based on the consideration of the three-dimensional motion of discrete
airstreams in the atmosphere. However, there is a significant body of work that reports that
airstreams are not entirely independent of each other (Cooper et al., 2004a,b). Midlatitude
cyclones typically form in a sequential manner, some trailing in close proximity along a
AX2-69
-------
quasi-stationary frontal boundary, with each system influenced by remnants of other systems.
For example, a rising stream of air ahead of a cold front (also known as a warm conveyor belt or
WCB) on the back (western) side of a surface anticyclone may entrain air that has subsided
anticyclonically into the surface high pressure system from the upper troposphere and the lower
stratosphere (also known as a Dry Airstream or DA) that intruded into the mid-troposphere in a
cyclone that is further downstream. Convective mixing of the boundary layer in the WCB will
distribute this enhanced O3 throughout the lower troposphere and down to the surface (Davies
and Shuepbach, 1994; Cooper and Moody, 2000). The net effect is that the DA of one cyclone
may feed into the WCB of the system immediately upwind. Similarly, the lofting of warm moist
air in the WCB may inject surface emissions into the upper troposphere adjacent to the western
side of the subsiding Dry Airstream of the storm system immediately downwind, with
subsequent interleaving of these two airstreams (Prados et al., 1999; Parrish et al., 2000; Cooper
et al., 2004a,b) as illustrated schematically in Figure AX2-8. The ultimate mixing of these
airstreams, which inevitably occurs at a scale that is not resolved by current models confounds
our ability to attribute trace gases to their sources.
These studies suggest that both downward transport from the stratosphere and upward
transport from the atmospheric boundary layer act in concert with their relative roles determined
by the balance between the amount of O3 in the lower stratosphere and the availability of free
radicals to initiate the photochemical processes forming O3 in the boundary layer. Dickerson
et al. (1995) pointed out that springtime maxima in O3 observed in Bermuda correlate well with
maxima in carbon monoxide. Carbon monoxide, O3 and its photochemical precursors may have
been transported into the upper troposphere from the polluted continental boundary layer by
deep convection. The photochemical processes involve the buildup of precursors during the
winter at Northern mid- and high latitudes. Parrish et al. (1999) have noted that reactions
occurring during the colder months may tend to titrate O3. However, as NOX and its reservoirs
are transported sourthward they can initiate O3 formation through reactions described in
Section AX2.2 (see also Stroud et al., 2003).
AX2-70
-------
Altitude
Stratospheric Air
Stratospheric Air
Cold Front (B)
West
Subsidence x \
Polluted Air
-1500
Distance (km)
Bermuda
East
Figure AX2-8. Schematic diagram of a meteorological mechanism involved in high
concentrations of O3 found in spring in the lower troposphere off the
American east coast. Subsidence behind the first cold front meets
convection ahead of a second cold front such that polluted air and O3
from the upper troposphere/lower stratosphere are transported in close
proximity (or mixed) and advected over the north Atlantic Ocean. The
vertical scale is about 10 km; the horizontal scale about 1500 km.
(Note that not all cold fronts are associated with squall lines and that
mixing occurs even in their absence.)
Source: Prados (2000).
AX2.3.2 Deep Convection in the Troposphere
Much of the upward motion in the troposphere is driven by convergence in the boundary
layer and deep convection. Deep convection, as in developing thunderstorms can transport
pollutants rapidly to the middle and upper troposphere (Dickerson et al., 1987). The outflow
from these systems results in the formation of layers with distinctive chemical properties in the
middle troposphere. In addition, layers are formed as the result of stratospheric intrusions.
Layers ranging in thickness typically from 0.3 to about 2 km in the middle troposphere (mean
altitudes between 5 and 7 km) are ubiquitous and occupy up to 20% of the troposphere to 12 km
(Newell et al., 1999). The origin of these layers can be judged by analysis of their chemical
AX2-71
-------
composition (typically by comparing ratios of H2O, O3 and CO to each other) or dynamical
properties (such as potential vorticity). Thus, pollutants that have been transported into the
middle and upper troposphere at one location can then be transported back down into the
boundary layer somewhere else.
Crutzen and Gidel (1983), Gidel (1983), and Chatfield and Crutzen (1984) hypothesized
that convective clouds played an important role in rapid atmospheric vertical transport of trace
species and first tested simple parameterizations of convective transport in atmospheric chemical
models. At nearly the same time, evidence was shown of venting of the boundary layer by
shallow fair weather cumulus clouds (e.g., Greenhut et al., 1984; Greenhut, 1986). Field
experiments were conducted in 1985, which resulted in verification of the hypothesis that deep
convective clouds are instrumental in atmospheric transport of trace constituents (Dickerson
et al., 1987; Luke, 1997). Once pollutants are lofted to the middle and upper troposphere, they
typically have a much longer chemical lifetime and with the generally stronger winds at these
altitudes they can be transported large distances from their source regions. Photochemical
reactions occur during this long-range transport. Pickering et al. (1990) demonstrated that
venting of boundary layer pollutants by convective clouds (both shallow and deep) causes
enhanced O3 production in the free troposphere. Therefore, convection aids in the
transformation of local pollution into a contribution to global atmospheric pollution. Downdrafts
within thunderstorms tend to bring air with less pollution from the middle troposphere into the
boundary layer.
Field studies have established that downward transport of larger O3 and NOX mixing ratios
from the free troposphere to the boundary layer is an important process over the remote oceans
(e.g., Piotrowicz et al., 1991), as well as the upward transport of very low O3 mixing ratios from
the boundary layer to the upper troposphere (Kley et al., 1996). Global modeling by Lelieveld
and Crutzen (1994) suggests that the downward mixing of O3 into the boundary layer (where it is
destroyed) is the dominant global effect of deep convection. Some indications of downward
transport of O3 from higher altitudes (possibly from the stratosphere) in the anvils of
thunderstorms have been observed (Dickerson et al., 1987; Poulida et al., 1996; Suhre et al.,
1997). Ozone is most effective as a greenhouse gas in the vicinity of the tropopause. Therefore,
changes in the vertical profile of O3 in the upper troposphere caused by deep convection have
important radiative forcing implications for climate.
AX2-72
-------
Other effects of deep convection include perturbations to photolysis rates, which include
enhancement of these rates in the upper portion of the thunderstorm anvil. In addition,
thunderstorms are effective in the production of NO by lightning and in wet scavenging of
soluble species.
AX2.3.2.1 Observations of the Effects of Convective Transport
Some fraction of shallow fair weather cumulus clouds actively vent boundary layer
pollutants to the free troposphere (Stull, 1985). The first airborne observations of this
phenomenon were conducted by Greenhut et al. (1984) over a heavily urbanized area, measuring
the in-cloud flux of O3 in a relatively large cumulus cloud. An extension of this work was
reported by Greenhut (1986) in which data from over 100 aircraft penetrations of isolated
nonprecipitating cumulus clouds over rural and suburban areas were obtained. Ching and
Alkezweeny (1986) reported tracer (SF6) studies associated with nonprecipitating cumulus (fair
weather cumulus and cumulus congestus). Their experiments showed that the active cumulus
clouds transported mixed layer air upward into the overlying free troposphere and suggested that
active cumuli can also induce rapid downward transport from the free troposphere into the mixed
layer. A UV-DIAL (Ultraviolet Differential Absorption Lidar) provided space-height cross
sections of aerosols and O3 over North Carolina in a study of cumulus venting reported by Ching
et al. (1988). Data collected on evening flights showed regions of cloud debris containing
aerosol and O3 in the lower free troposphere in excess of background, suggesting that significant
vertical exchange had taken place during afternoon cumulus cloud activity. Efforts have also
been made to estimate the vertical transport by ensembles of nonprecipitating cumulus clouds in
regional chemical transport models (e.g., Vukovich and Ching, 1990).
The first unequivocal observations of deep convective transport of boundary layer
pollutants to the upper troposphere were documented by Dickerson et al. (1987).
Instrumentation aboard three research aircraft measured CO, O3, NO, NOX, NOy, and
hydrocarbons in the vicinity of an active mesoscale convective system near the
Oklahoma/Arkansas border during the 1985 PRE-STORM experiment. Anvil penetrations about
two hours after maturity found greatly enhanced mixing ratios of all of the aforementioned
species compared with outside of the cloud. Among the species measured, CO is the best tracer
of upward convective transport because it is produced primarily in the boundary layer and has an
AX2-73
-------
atmospheric lifetime much longer than the timescale of a thunderstorm. In the observed storm,
CO measurements exceeded 160 ppbv as high as 11 km, compared with -70 ppbv outside of the
cloud (Figure AX2-9a). Cleaner middle tropospheric air appears to have descended in
downdrafts forming a pool of lower mixing ratio CO beneath the cloud. Nonmethane
hydrocarbons (NMHC) with moderate lifetimes can also serve as tracers of convective transport
from the boundary layer. Ozone can also be an indicator of convective transport. In the polluted
troposphere large O3 values will indicate upward transport from the boundary layer, but in the
clean atmosphere such values are indicative of downward transport from the uppermost
troposphere or lowermost stratosphere. In this case measured O3 in the upper rear portion of the
anvil peaked at 98 ppbv, while boundary layer values were only -65 ppbv (Figure AX2-9b). It is
likely that some higher-O3 stratospheric air mixed into the anvil.
The large amount of vertical trace gas transport noted by Dickerson et al. (1987) cannot,
however, be extrapolated to all convective cells. Pickering et al. (1988) reported airborne
measurements of trace gases taken in the vicinity of a line of towering cumulus and
cumulonimbus clouds that also occurred during PRE-STORM. In this case trace gas mixing
ratios in the tops of these clouds were near ambient levels. Meteorological analyses showed that
these clouds were located above a cold front, which prevented entry of air from the boundary
layer directly below or near the clouds. Instead, the air entering these clouds likely originated in
the layer immediately above the boundary layer which was quite clean. Luke et al. (1992)
summarized the air chemistry data from all 18 flights during PRE-STORM by categorizing each
case according to synoptic flow patterns. Storms in the maritime tropical flow regime
transported large amounts of CO, O3, and NOy into the upper troposphere with the
midtroposphere remaining relatively clean. During frontal passages a combination of stratiform
and convective clouds mixed pollutants more uniformly into the middle and upper levels; high
mixing ratios of CO were found at all altitudes.
Prather and Jacob (1997) and Jaegle et al. (1997) noted that in addition to the primary
pollutants (e.g., NOX, CO, VOCs), precursors of HOX are also transported to the upper
troposphere by deep convection. Precursors of most importance are water vapor, formaldehyde,
hydrogen peroxide, methylhydroperoxide, and acetone. HOX is critical for oxidizing NO to NO2
in the O3 production process.
AX2-74
-------
a. Carbon Monoxide
Surface
98°Wf 97'
Oklahoma City
96° 95° 94°
Longitude
b. Ozone
Surface
97C
Oklahoma City
96° 95° 94°
Longitude
93°
92°
Figure AX2-9a,b. (a) Contour plot of CO mixing ratios (ppbv) observed in and near the
June 15,1985, mesoscale convective complex in eastern Oklahoma.
Heavy line shows the outline of the cumulonimbus cloud. Dark shading
indicates high CO and light shading indicates low CO. Dashed contour
lines are plotted according to climatology since no direct measurements
were made in that area, (b) Same as (a) but for O3 (ppbv).
Source: Dickerson et al. (1987).
AX2-75
-------
Over remote marine areas the effects of deep convection on trace gas distributions differ
from that over moderately polluted continental regions. Chemical measurements taken by the
NASA ER-2 aircraft during the Stratosphere-Troposphere Exchange Project (STEP) off the
northern coast of Australia show the influence of very deep convective events. Between
14.5 and 16.5 km on the February 2 to 3, 1987 flight, perturbations in the chemical profiles were
noted that included pronounced maxima in CO, water vapor, and CCN and minima of NOy,
and O3 (Pickering et al., 1993). Trajectory analysis showed that these air parcels likely were
transported from convective cells 800 to 900 km upstream. Very low boundary layer mixing
ratios of NOy and O3 in this remote region were apparently transported upward in the convection.
A similar result was noted in CEPEX (Central Equatorial Pacific Experiment; Kley et al., 1996)
where a series of ozonesonde ascents showed very low upper tropospheric O3 following deep
convection. It is likely that similar transport of low-O3 tropical marine boundary layer air to the
upper troposphere occurs in thunderstorms along the east coast of Florida. Convection over the
Pacific will likely transport halogens to the upper troposphere where they may aid in the
destruction of O3. This low-O3 convective outflow will likely descend in the subsidence region
of the eastern Pacific, leading to some of the cleanest air that arrives at the west coast of the
United States.
AX2.3.2.2 Modeling the Effects of Convection
The effects of deep convection may be simulated using cloud-resolving models, or in
regional or global models in which the convection is parameterized. The Goddard Cumulus
Ensemble (GCE) model (Tao and Simpson, 1993) has been used by Pickering et al. (1991,
1992a,b, 1993, 1996), Scala et al. (1990) and Stenchikov et al. (1996) in the analysis of
convective transport of trace gases. The cloud model is nonhydrostatic and contains detailed
representation of cloud microphysical processes. Two- and three-dimensional versions of the
model have been applied in transport analyses. The initial conditions for the model are usually
from a sounding of temperature, water vapor and winds representative of the region of storm
development. Model-generated wind fields can be used to perform air parcel trajectory analyses
and tracer advection calculations. Once transport calculations are performed for O3 precursors, a
1-D photochemical model was employed to estimate O3 production rates in the outflow air from
AX2-76
-------
the convection. These rates were then compared with those prior to convection to determine an
enhancement factor due to convection.
Such methods were used by Pickering et al. (1992b) to examine transport of urban plumes
by deep convection. Transport of the Oklahoma City plume by the 10-11 June 1985
PRE-STORM squall line was simulated with the 2-D GCE model. In this event forward
trajectories from the boundary layer at the leading edge of the storm showed that almost 75% of
the low level inflow was transported to altitudes exceeding 8 km. Over 35% of the air parcels
reached altitudes over 12 km. Tracer transport calculations were performed for CO, NOX, O3,
and hydrocarbons. The 3-D version of the GCE model has also been run for the 10-11 June
1985 PRE-STORM case. Free tropospheric O3 production enhancement of a factor of 2.5 for
Oklahoma rural air and -4 for the Oklahoma City case were calculated.
Stenchikov et al. (1996) used the 2-D GCE model to simulate the North Dakota storm
observed by Poulida et al. (1996). This storm showed the unusual feature of an anvil formed
well within the stratosphere. The increase of CO and water vapor above the altitude of the
preconvective tropopause was computed in the model. The total mass of CO across the model
domain above this level increased by almost a factor of two during the convective event. VOCs
injected into the lower stratosphere could enhance O3 production there. Downward transport of
O3 from the stratosphere was noted in the simulation in the rear anvil.
Regional estimates of deep convective transport have been made through use of a traveling
1-D model, regional transport models driven by parameterized convective mass fluxes from
mesoscale meteorological models, and a statistical-dynamical approach. Pickering et al. (1992c)
developed a technique which uses a combination of deep convective cloud cover statistics from
the International Satellite Cloud Climatology Project (ISCCP) and convective transport statistics
from GCE model simulations of prototype storms to estimate the amount of CO vented from the
planetary boundary layer (PEL) by deep convection. This statistical-dynamical approach was
used by Thompson et al. (1994) to estimate the convective transport component of the boundary
layer CO budget for the central United States (32.5°-50° N, 90°-105° W) for the month of June.
They found that the net upward deep convective flux (-18 x 105 kg-CO/month) and the shallow
convective flux (-16 x io5 kg-CO/month) to the free troposphere accounted for about 80% of the
loss of CO from the PEL. These losses roughly balanced horizontal transport of CO (-28 x 10s
kg-CO/month), the oxidation of hydrocarbons (-8 x 10s kg-CO/month) and anthropogenic and
AX2-77
-------
biogenic emissions (~8 + ~1 x 10s kg-CO/month) into the PEL in the central United States.
In this respect the central United States acts as a "chimney" for venting CO and other pollutants.
Regional chemical transport models have been used for applications such as simulations of
photochemical O3 production, acid deposition, and fine particulate matter. Walcek et al. (1990)
included a parameterization of cloud-scale aqueous chemistry, scavenging, and vertical mixing
in the chemistry model of Chang et al. (1987). The vertical distribution of cloud microphysical
properties and the amount of subcloud-layer air lifted to each cloud layer are determined using a
simple entrainment hypothesis (Walcek and Taylor, 1986). Vertically-integrated O3 formation
rates over the northeast U.S. were enhanced by -50% when the in-cloud vertical motions were
included in the model.
Wang et al. (1996) simulated the 10-11 June 1985 PRE-STORM squall line with the
NCAR/Penn State Mesoscale Model (MM5; Grell et al., 1994; Dudhia et al., 1993). Convection
was parameterized as a subgrid-scale process in MM5 using the Kain and Fritsch (1993) scheme.
Mass fluxes and detrainment profiles from the convective parameterization were used along with
the 3-D wind fields in CO tracer transport calculations for this convective event. The U.S.
Environmental Protection Agency has developed a Community Multiscale Air Quality (CMAQ)
modeling system that uses MM5 with the Kain-Fritsch convective scheme as the dynamical
driver (Ching et al., 1998).
Convective transport in global chemistry and transport models is treated as a subgrid-scale
process that is parameterized typically using cloud mass flux information from a general
circulation model or global data assimilation system. While GCMs can provide data only for a
"typical" year, data assimilation systems can provide "real" day-by-day meteorological
conditions, such that CTM output can be compared directly with observations of trace gases.
The NASA Goddard Earth Observing System Data Assimilation System (GEOS-1 DAS and
successor systems; Schubert et al., 1993; Bloom et al., 1996) provides archived global data sets
for the period 1980 to present, at 2° x 2.5° or better resolution with 20 layers or more in the
vertical. Convection is parameterized with the Relaxed Arakawa-Schubert scheme (Moorthi and
Suarez, 1992). Pickering et al. (1995) showed that the cloud mass fluxes from GEOS-1 DAS are
reasonable for the 10-11 June 1985 PRE-STORM squall line based on comparisons with the
GCE model (cloud-resolving model) simulations of the same storm. In addition, the GEOS-1
DAS cloud mass fluxes compared favorably with the regional estimates of convective transport
AX2-78
-------
for the central U.S. presented by Thompson et al. (1994). However, Allen et al. (1997) have
shown that the GEOS-1 DAS overestimates the amount and frequency of convection in the
tropics and underestimates the convective activity over midlatitude marine storm tracks.
AX2.3.3 Nocturnal Low-Level Jets
Nocturnal low-level jets (LLJ) are coincident with synoptic weather patterns involved with
high O3 episodes implying that they may play an important role in the formation of severe O3
events (Rao and Zurbenko, 1994). LLJ can transport pollutants hundreds of kilometers from
their sources. Figure AX2-10 shows the evolution of the planetary boundary layer (PEL) over
land during periods when high-pressure weather patterns prevail (Stull, 1988). During synoptic
weather patterns with stronger zonal flow, a schematic of the boundary layer could look quite
different with generally more uniform mixing present. As can be seen from Figure AX2-10, the
PEL can be divided into three sublayers: a turbulent mixed layer (typically present during
daylight hours), a less turbulent residual layer which occupies space that was formerly the mixed
layer, and a nocturnal, stable boundary layer that has periods of sporadic turbulence (Stull,
1988). The LLJ forms in the residual layer. It is important to note, that during the nighttime, the
PEL often comprises thin, stratified layers with different physical and chemical properties
(Stull, 1988).
At night, during calm conditions, the planetary boundary layer is stably stratified and as a
result verticle mixing is inhibited. On cloud-free evenings the LLJ begins to form shortly after
sunset. The wedge of cool air in the stable nocturnal boundary layer decouples the surface layer
from the residual layer and acts like a smooth surface allowing the air just above it (in the
residual layer) to flow rapidly past the inversion mostly unencumbered by surface friction (Stull,
1988). As the sun rises, its energy returns to heat the land and the lower atmosphere begins to
mix as the warm air rises. The jet diminishes as the nocturnal temperature inversion erodes and
surface friction slows wind speeds. If stable synoptic conditions persist, the same conditions the
next night could allow the low-level jet to reform with equal strength and similar consequences.
LLJ formation results in vertical wind shear that induces mixing between the otherwise stratified
layers.
AX2-79
-------
2-
o>
•53 1
0-
Free Atmosphere
Cloud Layer
Cloud Layer
Stable Nocturnal Boundary Layer
Mixed
Layer
Surface Layer
Afternoon
Sunset
Midnight
Sunrise
I
Noon
Figure AX2-10. The diurnal evolution of the planetary boundary layer while high
pressure prevails over land. Three major layers exist (not including the
surface layer): a turbulent mixed layer; a less turbulent residual layer
which contains air formerly in the mixed layer; and a nocturnal, stable
boundary layer which is characterized by periods of sporadic turbulence.
Source: Adapted from Stall (1988) Figures 1.7 and 12.1.
LLJs are often associated with mountain ranges. Mountains and pressure gradients on
either side of a developing LLJ help concentrate the flow of air into a corridor or horizontal
stream (Hobbs et al., 1996). Figure AX2-11 shows that LLJs commonly form east of the Rocky
Mountains and east of the Appalachian Mountains (Bonner, 1968). There may be other locations
in the U.S. where LLJs occur. The width of the jet can vary from location to location and from
one weather pattern to another, but is typically less than several hundred km not greater than
1000 km long. In extreme cases, winds in a LLJ can exceed 60 ms"1 but average speeds are
typically in the range of 10 to 20 ms"1.
Nocturnal low-level jets are not unique to the United States; they have been detected in
many other parts of the world (Corsmeier, 1997, Reitebuch, et al., 2000). Corsmeier et al.
(1997) observed secondary maxima in surface O3 at nighttime at a rural site in Germany,
supporting the notion that downward transport from the residual layer was occurring. The
secondary maxima were, on average, 10% of the next day's O3 maximum but at times could be
as much as 80% of the maximum (Corsmeier et al., 1997). The secondary O3 maxima were well
AX2-80
-------
Figure AX2-11. Locations of low-level jet occurrences in decreasing order of prevalence
(most frequent, common, observed). These locations are based on 2-years
radiosonde data obtained over limited areas. With better data coverage,
other low-level jets may well be observed elsewhere in the United States.
Source: Bonner (1968).
correlated with an increase in wind speed and wind shear. The increased vertical shear over the
very thin layer results in mechanical mixing that leads a downward flux of O3 from the residual
to the near surface layer (see Low-level jets AX2-12 and AX2-13). Analysis of wind profiles
from aerological stations in northeastern Germany revealed the spatial extent of that particular
LLJ was up to 600 km in length and 200 km in width. The study concluded the importance of O3
transport by low-level jets was twofold: O3 and other pollutants could be transported hundreds
of kilometers at the jet core level during the night and then mixed to the ground far from their
source region. Salmond and McKendry (2002) also observed secondary O3 maxima (in the
Lower Fraser Valley, British Columbia) associated with low-level jets that occasionally
exceeded half the previous day's maximum O3 concentration. The largest increases in surface
O3 concentration occurred when boundary layer turbulence coincided with O3 levels greater than
80 parts per billion were observed aloft. In addition, the study suggests horizontal transport
efficiency during a low-level jet event could be as much as six times greater than transport with
light winds without an LLJ. Reitebuch et al. (2000) observed secondary O3 maxima associated
with low-level jet evolution in an urban area in Germany. The notion that O3 was transported
AX2-81
-------
o
"•5
(0
2 1.0-
Surface Ozone During Low Level Jet Periods
o
O
O
•o
N
I
o
0.5-
0.0-
Secondary
63 Maxima
o
o
CD
O
O
O
o
o
o
o
o
o
66
o
o
CM
o
o
o
o
o
o
CO
o
o
o
CD
o
Figure AX2-12.
Time of Day (Local)
Schematic diagram showing the diurnal behavior of O3 and the
development of secondary O3 maxima resulting from downward
transport from the residual layer when a low-level jet is present.
Source: Adapted from Reitbuch et al. (2000); Corsmeier et al. (1997); and Salmond and McKendry (2002).
Boundary Layer Winds During
a Nocturnal Low Level Jet
1.0-
f 0.5-
0-
M
£ O
m N
10 15 20
Wind Speed (ms-1) Wind Direction
Figure AX2-13. The nocturnal low-level jet occupies a thin slice of the atmosphere near
the Earth's surface. Abrupt changes in wind speed and wind direction
with height associated with the low-level jet create conditions favorable
for downward transport of air to the surface layer.
Source: Singh et al. (1997); Corsmeier et al. (1997).
AX2-82
-------
downward from the residual layer to the surface was supported by observed decreases in
concentrations of NO, NO2 and CO in the residual layer during secondary O3 maxima. Unlike
O3 in the residual layer, concentrations of NO, NO2, and CO should be lower than those found
nearer the surface (Reitebuch et al., 2000; Seinfeld and Pandis, 1998). As in other studies, wind
speed and directional shear were detected during these events. Calculations of the average wind
speed and duration of the LLJ suggested that pollutants were transported several hundred
kilometers. A study of the PEL and the vertical structure of O3 observed at a costal site in Nova
Scotia described how temperature and differences of surface roughness in a marine environment
can induce LLJ formation and pollution transport (Gong et al., 2000). In this case, rather strong
horizontal sea surface temperature gradients provided the necessary baroclinic forcing.
While the studies mentioned above have shed light on the possible role of the LLJ in the
transport of O3 and its precursors, quantitative statements about the significance of the LLJ in
affecting local and regional O3 budgets cannot yet be made. This inability reflects the lack of
available data for wind profiles in the planetary boundary layer in areas where LLJ are likely to
occur and because of the inadequacy of numerical models in simulating their occurrence.
AX2.3.4 Intercontinental Transport of Ozone and Other Pollutants
AX2.3.4.1 The Atmosphere/Ocean Chemistry Experiment, AEROCE
The AEROCE experiment, initiated in the early 1990s set out to examine systematically
the chemistry and meteorology leading to the trace gas and aerosol composition over the North
Atlantic Ocean. One particular focus area was to determine the relative contribution of
anthropogenic and natural processes to the O3 budget and oxidizing capacity of the troposphere
over the North Atlantic Ocean. Early results using isentropic back trajectories suggested that
periodic pulses of O3 mixing ratios up to 80 ppb were associated with large-scale subsidence
from the mid-troposphere, favoring a natural source (Oltmans and Levy, 1992). Moody et al.
(1995) extended this work with a five-year seasonal climatology and found the highest
concentrations of O3 were always associated with synoptic scale postfrontal subsidence off the
North American continent behind cold fronts, and this pattern was most pronounced in the
April-May time frame. These postfrontal air masses had uniformly low humidity and high
concentrations of 7Be, a cosmogenic tracer produced in the upper troposphere and lower
stratosphere. However, the pulsed occurrence of these postfrontal air masses also frequently
AX2-83
-------
delivered enhanced concentrations of species such as SO42", NO3 , 210Pb, etc. suggesting a
component originating in North America. In a subsequent analysis of data from one year (1992)
when CO observations were available, Dickerson et al. (1995) concluded that anthropogenic
sources made a significant contribution to surface O3, and using a simple mixing model they
determined that 57% of the air had a continental boundary layer origin.
Based on these observations of the synoptically modulated concentrations, AEROCE
conducted an aircraft and ozonesonde intensive in the spring of 1996. The intention was to
adopt a meteorologically informed sampling strategy to clearly distinguish the characteristics of
air masses ahead of and behind eastward progressing cold fronts. Sixteen research flights were
conducted with the University of Wyoming King Air research aircraft. The goal was to
differentiate the sources of enhanced O3 mixing ratios observed on Bermuda after the passage of
cold fronts, and to identify the major processes controlling the highly variable O3 mixing ratios
in the mid-to-upper troposphere over eastern North America and the North Atlantic Ocean
during April and May. In addition to aircraft flights, near-daily ozonesondes were launched in a
quasi-zonal transect from Purdue, Indiana, to Charlottesville, Virginia to Bermuda. An effort
was made to time the release of ozonesondes to cleanly differentiate pre- and postfrontal air
masses.
In several aircraft flights, the presence at altitude of distinct layers of air with elevated
concentrations of nonmethane hydrocarbons (NMHCs) attested to the dynamic vertical mixing
associated with springtime frontal activity. Layers of mid-tropospheric air of high O3 (140 ppb)
and low background NMHC mixing ratios (1.44 ppbv ethane, 0.034 ppbv propene, 0.247 ppbv
propane, and 0.034 ppbv isobutene, 0.041 ppbv n-butane, 0.063 ppbv benzene, 0.038 ppbv
toluene) were indicative of descending, stratospherically influenced air on a flight to the east of
Norfolk, VA on April 24 (alt 4600m). However layers of elevated NMHC concentrations
(1.88 ppbv ethane, 0.092 ppbv propene, 0.398 ppbv propane, 0.063 ppbv isobutene, 0.075 ppbv
n-butane, 0.106 ppbv benzene, 0.0102 ppbv toluene) occurred along with 60-70 ppbv of O3 on a
flight west of Bermuda April 28 (alt. 4100m), indicating air had been lofted from the continental
boundary layer. Meteorological evidence, supported by ozonesonde observations and earlier
King Air flights, indicated that stratosphere/troposphere exchange associated with an upstream
frontal system had injected and advected dry, O3-rich air into the mid-troposhere region over the
continent. This subsiding air mass provided deep layers of enhanced O3 in the offshore,
AX2-84
-------
postfrontal area. Convection from a developing (upwind) system lifted continental boundary
layer air into the proximity of the dry, subsiding air layer (Prados, et al., 1999). This resulted in
a mixture of high concentrations of anthropogenic pollutants along with naturally enhanced O3.
Ozone mixing ratios exceeded those attributable to boundary layer venting or in-transit
photochemical production. These meteorological processes led to pollution and stratospherically
enhanced O3 co-occurring in postfrontal air masses over the North Atlantic Ocean. A similar
event in February 1999 was observed by Parrish et al. (2000). It confirmed the occurrence of
thin layers of anthropogenic and stratospheric air that subsequently mix. These results, along
with recent modeling studies suggest that North American pollution clearly does contribute to
the periodic influx of less-than-pristine air observed in the marine boundary layer over Bermuda
(e.g., Li et al., 2002) and yet these incursions are not inconsistent with observing enhancements
in O3 due to stratospheric exchange.
The ozonesonde climatology of AEROCE clearly established that O3 mixing ratios were
always enhanced and increased with height in postfrontal air masses. Postfrontal O3 in the lower
troposphere over Bermuda originates in the postfrontal midtroposphere over the continent,
supporting the hypothesis that naturally occurring stratospheric O3 makes a contribution to air in
the marine boundary layer (Cooper et al., 1998). A schematic of the meteorological processes
responsible for the close proximity of natural and man-made O3 can be seen in Figure AX2-8
from Prados (2000). Cold fronts over North America tend to be linked in wave-like patterns
such that the subsidence behind one front may occur above with intrusions of convection ahead
of the next cold front. Pollutants, including VOC and NOx, precursors to O3, may be lofted into
the mid-to-upper troposphere where they have the potential to mix with layers of air descending
from O3-rich but relatively unpolluted upper troposphere and lower stratosphere. Through this
complex mechanism, both stratospheric and photochemically produced O3 may be transported to
the remote marine environment where they have large-scale impacts on the radiative and
chemical properties of the atmosphere. Recent three-dimensional modeling studies of air mass
motion over the Pacific provide further evidence that these complex mechanisms are indeed
active (Cooper et al., 2004b).
AX2-85
-------
AX2.3.4.2 The North Atlantic Regional Experiment, NARE
NARE was established by the International Global Atmospheric Chemistry Project to study
the chemical processes occurring in the marine troposphere of the North Atlantic, the marine
region expected to be the most impacted by industrial emissions from eastern North America and
western Europe. Surface measurements from several surface sites were initiated in 1991, with
major field intensives in summer 1993, spring 1996, early autumn 1997 and a few winter flights
in 1999. In the summer of 1993, airborne and ground-based measurements of O3 and O3
precursors were made in the North Atlantic region by an international team of scientists to
determine how the continents that rim the North Atlantic are affecting atmospheric composition
on a hemispheric scale (Fehsenfeld et al., 1996a,b). The focus of NARE was to investigate the
O3 budget of the North Atlantic region. Previous observations indicated that the O3 produced
from anthropogenic sources is greater than that reaching the lower troposphere from the
stratosphere and that O3 derived from anthropogenic pollution has a hemisphere wide effect at
northern mid latitudes. This study was performed to better quantify the contribution of
continental sources to the O3 levels over the North Atlantic.
Buhr et al. (1996) measured O3, CO, NO, and NOy as well as meteorological parameters
aboard the NCAR King Air in August 1993 during 16 flights over and near the Gulf of Maine.
They found that O3 produced from anthropogenic precursors was dominant throughout the
experimental region below 1500 m, in altitude.
The National Research Council of Canada Twin Otter aircraft was used to measure the O3
and related compounds in the summertime atmosphere over southern Nova Scotia (Kleinman
et al., 1996a,b). Forty-eight flights were performed, primarily over the surface sampling site in
Chebogue Point, Nova Scotia, or over the Atlantic Ocean. They found that a wide variety of air
masses with varying chemical content impact Nova Scotia. The effect depends on flow
conditions relative to the locations of upwind emission regions and the degree of photochemical
processing associated with transport times ranging from about 1 - 5 d. Moist continental
boundary layer air with high concentrations of O3 and other anthropogenic pollutants was
advected to Nova Scotia in relatively thin vertical layers, usually with a base altitude of several
hundred meters. Dry air masses with high concentrations of O3 often had mixed boundary layer
and upper atmosphere source regions. When a moist and dry air mass with the same
photochemical age and O3 concentration were compared, the dry air mass had lower
AX2-86
-------
concentrations of NOy and aerosol particles, which was interpreted as evidence for the selective
removal of soluble constituents during vertical lifting.
Due to strong, low-level temperature inversions over the North Atlantic, near surface air is
often unrepresentative of the eastward transport of the North American plume because of a
decoupling from the air transported aloft (Kleinman et al., 1996a; Daum et al., 1996; Angervine
et al., 1996). Pollution plumes were observed in distinct strata up to 1 km. Plume chemical
compositions were consistent with the occurrence of considerable photochemical processing
during transit from source regions over the eastern seaboard of the U.S. Ozone concentrations
reached 150 ppbv, NOX conversion to its oxidation products exceeded 85%, and high hydrogen
peroxide concentrations were observed (median 3.6 ppbv, maximum 11 ppbv). CO and O3
concentrations were well correlated (R2 = 0.64) with a slope (0.26) similar to previous
measurements in photochemically aged air (Parrish et al., 1998). Ozone depended nonlinearly
on the NOX oxidation product concentration, but there was a correlation (r2 = 0.73) found
between O3 and the concentration of radical sink species as represented by the quantity
((NOy-NOx) + 2H202)
Banic et al. (1996) determined that the average mass of O3 transported through an area
1 m in horizontal extent and 5 km in the vertical over the ocean near Nova Scotia to be 2.8 g s"1,
moving from west to east. Anthropogenic O3 accounted for half of the transport below 1 km,
35 to 50% from 1 to 3 km, 25 to 50% from 3 to 4 km, and only 10% from 4 to 5 km. Merrill and
Moody (1996) analyzed the meteorological conditions during the NARE intensive period
(August 1 to September 13, 1993). They determined the ideal meteorological scenario for
delivering pollution plumes from the U.S. East Coast urban areas over the Gulf of Maine to the
Maritime Provinces of Canada to be warm sector flow ahead of an advancing cold front. In the
winter phase of NARE, O3 and CO were measured from the NOAA WP-3D Orion aircraft from
St. John's, Newfoundland, Canada, and Keflavik, Iceland, from February 2 to 25, 1999 (Parrish
et al., 2000). In the lower troposphere over the western North Atlantic Ocean, the close
proximity of air masses with contrasting source signatures was remarkable. High levels of
anthropogenic pollution immediately adjacent to elevated O3 of stratospheric origin were
observed, similar to those reported by Prados et al. (1999). In air masses with differing amounts
of anthropogenic pollution, O3 was negatively correlated with CO, which indicates that
AX2-87
-------
emissions from surface anthropogenic sources had reduced O3, in this wintertime period, even in
air masses transported into the free troposphere.
The influence of the origin and evolution of airstreams on trace gas mixing ratios has been
studied in great detail for NARE aircraft data. The typical midlatitude cyclone is composed of
four major component airstreams, the warm conveyor belt, the cold conveyor belt, the post cold-
front airstream and the dry airstream (Cooper et al., 2001). The physical and chemical
processing of trace species was characterized for each airstream, and a conceptual model of a
midlatitude cyclones was developed (Cooper et al., 2002a). This showed how airstreams within
midlatitude cyclones drew and exported trace gases from the polluted continental boundary
layer, and the stratospherically enhanced mid-troposphere. Using back trajectories, airstream
composition was related to the origin and transport history of the associated air mass. The
lowest O3 values were associated with airstreams originating in Canada or the Atlantic Ocean
marine boundary layer; the highest O3 values were associated with airstreams of recent
stratospheric origin. The highest NOy values were seen in polluted outflow from New England
in the lower troposphere. A steep and positive O3/NOy slope was found for all airstreams in the
free troposphere regardless of air mass origin. Finally, the seasonal variation of photochemistry
and meteorology and their impact on trace gas mixing ratios in the conceptual cyclone model
was determined (Cooper et al., 2002b). Using a positive O3/CO slope as an indicator of
photochemical O3 production, O3 production during late summer-early autumn is associated with
the lower troposphere post-cold-front airstream and all levels of the WCB, especially the lower
troposphere. However, in the early spring, there is no significant photochemical O3 production
for airstreams at any level, and negative slopes in the dry air airstream indicate STE causes the
O3 increase in the mid- and upper troposphere.
Stohl et al. (2002) analyzed total odd nitrogen (NOy) and CO data taken during NARE in
spring 1996 and fall 1997. They studied the removal timescales of NOy originating from surface
emissions of NOX and what fraction reached the free troposphere. NOX limits O3 production in
the free troposphere and can be regenerated from NOy after the primary NOx has been
exhausted. It was determined that < 50% of the NOy observed above 3 km came from
anthropogenic surface emissions. The rest had to have been emitted in situ.
Several studies (e.g., Stohl and Trickl, 1999; Brunner et al., 1998; Schumann et al., 2000;
Stohl et al., 2003; Traub et al., 2003) have identified plumes that have originated in North
-------
America over Europe and over the eastern Mediterranean basin (e.g., Roelofs et al., 2003; Traub
et al., 2003). Modeling studies indicate that North American emissions contribute roughly 20%
to European CO levels and 2 to 4 ppb to surface O3, on average. Episodic events, such as forest
fires in North America have also been found to result in elevated CO and O3 levels and visible
haze layers in Europe (Volz-Thomas, et al., 2003). The O3 is either transported from North
America or formed during transport across the North Atlantic Ocean, perhaps as the result of
interactions between the photochemical degradation products of acetone with emissions of NOX
from aircraft (Bruhl et al., 2000; Arnold et al., 1997). In addition, North American and European
pollution is exported to the Arctic. Eckhardt et al. (2003) show that this transport is related to
the phase of the North Atlantic Oscillation which has a period of about 20 years.
AX2.3.5 Small-Scale Circulation Systems
Sub-synoptic scale circulation systems are especially important for determining pollution
levels. These systems include sea-land breezes, mountain-valley breezes and circulations driven
by the urban heat island. The circulations associated with these phenomena typically occur on
spatial scales of tens of kilometers.
AX2.3.5.1 Land-Sea Breeze
Of particular interest is the sea-land breeze, as many urban areas such as those in the
Northeast Corridor (Washington, Baltimore, Philadelphia, New York and Boston), Chicago,
Houston, and Los Angeles in which high O3 values are found (see Chapter 3) are located in or
near coastal zones.
During the day, heating of the land surface results in upward motion that is compensated
by air flowing in from the adjacent water body, i.e., the sea breeze. Winds gradually rotate with
height to produce a return flow aloft. This circulation generally reaches maximum strength in
late afternoon. Afterwards, the sense of the circulation is reversed and a land breeze develops,
which reaches maximum strength shortly after the land-sea temperature contrast is largest. This
also implies that winds are rotating at the surface as the sense of the circulation changes. The
circulation can interact with the larger synoptic scale flow pattern to either attenuate surface
winds or to increase them.
AX2-89
-------
Because of these effects, sea-land breezes can exert significant effects on concentrations of
pollutants emitted in coastal areas. If there is onshore flow (sea breeze) when there is opposing
large scale flow, a transition zone (sometimes called a sea breeze front) in which convergence of
the opposing flows forms. In this zone, horizontal winds are weak. Furthermore, air pollutants
are concentrated and are transported upwards. This situation was found during O3 episodes
occurring in Houston in August 2000, in which a "wall of pollution" formed (Banta et al., 2005).
If the sea breeze is dominant, it can transport pollutants well inland, even to central-eastern
Texas. During the land breeze phase, pollutants at the surface tend to be transported out over the
adjacent water body, resulting in dilution and dispersion of pollutants. Observational studies of
the effects of small scale circulations on O3 concentrations are lacking.
AX2.3.6 The Relation of Ozone to Solar Ultraviolet Radiation, Aerosols,
and Air Temperature
AX2.3.6.1 Solar Ultraviolet Radiation and Ozone
The effects of sunlight on photochemical oxidant formation, aside from the role of solar
radiation in meteorological processes, are related to its intensity and its spectral distribution.
Intensity varies diurnally, seasonally, and with latitude, but the effect of latitude is strong only in
the winter. Ultraviolet radiation from the sun plays a key role in initiating the photochemical
processes leading to O3 formation and affects individual photolytic reaction steps. However,
there is little empirical evidence in the literature, directly linking day-to-day variations in
observed UV radiation levels with variations in O3 levels.
In urban environments the rate of O3 formation is sensitive to the rate of photolysis of
several species including H2CO, H2O2, O3, and especially NO2. Monte Carlo calculations
suggest that model calculations of photochemical O3 production are most sensitive to uncertainty
in the photolysis rate coefficient for NO2 (Thompson and Stewart, 1991; Baumann et al., 2000).
The International Photolysis Frequency Measurement and Modeling Intercomparison (IPMMI)
hosted recently by NCAR in Boulder, CO brought together more than 40 investigators from
8 institutions from around the world (Bais et al., 2003; Cantrell et al., 2003 and Shelter et al.,
2003). They compared direct actinometric measurements, radiometric measurements, and
numerical models of photolysis rate coefficients, focusing on O3 to O(JD) and NO2, referred to as
j(03)andj(N02).
AX2-90
-------
The combination of direct measurements and comparisons to model calculations indicated
that for clear skies, zenith angles less than 70°, and low aerosol loadings, the absolute value of
the j(NO2) at the Earth's surface is known to better than 10% with 95% confidence. The results
suggest that the cross sections of Harder et al. (1997a) may yield more accurate values when
used in model calculations of j(NO2). Many numerical models agreed among themselves and
with direct measurements (actinometers) and semi-direct measurements (radiometers) when
using ATLAS extraterrestrial flux from Groebner and Kerr (2001). The results of IPMMI
indicate numerical models are capable of precise calculation of photolysis rates at the surface
and that uncertainties in calculated chemical fields arise primarily from uncertainties in the
variation of actinic flux with altitude in addition to the impact of clouds and aerosols on
radiation.
AX2.3.6.2 Impact of Aerosols on Radiation and Photolysis Rates and
Atmospheric Stability
Because aerosol particles influence the UV flux there is a physical link between particles
and gases that depends on the concentration, distribution, and refractive index of the particles.
Scattering of UV radiation by tropospheric aerosol particles can strongly impact photolysis rates
and thus photochemical O3 production or destruction. The effect shows high sensitivity to the
properties of the aerosol. Particles in the boundary layer can accelerate photochemistry if the
single scattering albedo is near unity, such as for sulfate and ammoniated sulfate aerosols, or
inhibit O3 production if the single scattering albedo is low, such as for mineral dust or soot
(Dickerson et al., 1997; Jacobson, 1998; Liao et al., 1999; Castro et al., 2001; Park et al., 2001).
Any aerosol layer in the free troposphere will reduce photolysis rates in the boundary layer.
The interaction of aerosols, photochemistry, and atmospheric thermodynamic processes
can impact radiative transport, cloud microphysics, and atmospheric stability with respect to
vertical mixing. Park et al. (2001) developed a single-column chemical transport model that
simulates vertical transport by convection, turbulent mixing, photochemistry, and interactive
calculations of radiative fluxes and photolysis rates. Results from simulations of an episode over
the eastern United States showed strong sensitivity to convective mixing and aerosol optical
depth. The aerosol optical properties observed during the episode produced a surface cooling of
up to 120 W/m2 and stabilized the atmosphere suppressing convection. This suggests two
AX2-91
-------
possible feedbacks mechanisms between aerosols and O3-reduced vertical mixing would tend to
increase the severity of O3 episodes, while reduced surface temperatures would decrease it.
AX2.3.6.3 Temperature and Ozone
An association between surface O3 concentrations and temperature has been demonstrated
from measurements in outdoor smog chambers and from measurements in ambient air.
Numerous ambient studies done over more than a decade have reported that successive
occurrences or episodes of high temperatures characterize high O3 years (Clark and Karl, 1982;
Kelly et al., 1986). The relation of daily maximum 8-h average O3 concentration to daily
maximum temperature from May to September 1994 to 2004 is illustrated in Figure AX2-14 for
the Baltimore Air Quality Forecast Area. The relation, based on daily maximum 1-h average O3
concentration is illustrated in Figure AX2-15. The relations are very similar in the two figures,
reflecting the high degree of correlation (r = 0.98) between the daily maximum 1-h and 8-h O3
concentrations. The relation of daily maximum 8-h average O3 to daily maximum temperature
from May to September 1994 to 2004 is illustrated in Figure AX2-15 for the three sites
downwind of Phoenix, AZ on high O3 days (cf, Figure AX3-32). As can be seen from a
comparison of Figures AX2-14 and AX2-16, O3 concentrations in the Phoenix area are not as
well correlated with daily maximum temperature (r = 0.14) as they are in the Baltimore Area
(r = 0.74). There appears to be an upper-bound on O3 concentrations that increases with
temperature. Likewise, Figure AX2-16 shows that a similar qualitative relationship exists
between O3 and temperature even at a number of nonurban locations.
The notable trend in these plots is the apparent upper-bound to O3 concentrations as a
function of temperature. It is clear that, at a given temperature, there is a wide range of possible
O3 concentrations because other factors (e.g., cloudiness, precipitation, wind speed) can reduce
O3 production rates. The upper edge of the curves may represent a practical upper bound on the
maximum O3 concentration achieved under the most favorable conditions. Relationships
between peak O3 and temperature also have been recorded by Wunderli and Gehrig (1991) for
three locations in Switzerland. At two sites near Zurich, peak O3 increased 3 to 5 ppb/°C for
diurnal average temperatures between 10 and 25 °C, and little change in peak O3 occurred for
temperatures below 10 °C. At the third site, a high-altitude location removed from
anthropogenic influence, a much smaller variation of O3 with temperature was observed.
AX2-92
-------
a.
a
ip
c
u
o
O
o
f
CO
3
E
B 40
S HI-
Q
„ °B °D
a D[
a n^fig[yi
D Hn n B|B|f
0 a n nSBaanD nHB'
DD|"| D| J|D
o°
U
Hn
|[
|p
nl
'
•o|
y
aH
I
II
D0a
E
DB
BD
Rg
1
IB
I
1 B
a riin-
O O
aooH|nn|pl
°° |n B°B 1°°°
illi^B1:
ga| a
I j|gB°a
lll|ol^S P
l^g °
pB|D
Q
gnnn
yn
u nn
10 15 20 25 30 35 40 45
Tmax at BWI °C
Figure AX2-14. A scatter plot of daily maximum 8-h O3 concentration versus daily
maximum temperature in the Baltimore, MD Air Quality Forecast Area.
Source: Piety (2005)
200
a.
a.
o
~
ro
•M
c
o>
o
c
o
o
a
Q
10
15 20 25
Tmax at BWI °C
30
35
40
Figure AX2-15. A scatter plot of daily maximum 1-h average O3 concentration versus
daily maximum temperature in the Baltimore, MD Air Quality Forecast
Area.
Source: Piety (2005)
AX2-93
-------
160'
£ 140'
a
a.
.S 120'
15
S 100'
u
C
o
o
60
S 20
10 15 20 25 30 35 40 45 50
Figure AX2-16. A scatter plot of daily maximum 8-h average O3 concentrations versus
daily maximum temperature downwind of Phoenix, AZ.
Source: Piety (2005)
Some possible explanations for the correlation of O3 with temperature include:
(1) Increased photolysis rates under meterological conditions associated with higher
temperatures;
(2) Increased H2O concentrations with higher temperatures as this will lead to greater OH
production via R(2-6);
(3) Enhanced thermal decomposition of PAN and similar compounds to release NOX at
higher temperatures;
(4) Increase of anthropogenic hydrocarbon (e.g., evaporative losses) emissions or NOX,
emissions with temperature or both;
(5) Increase of natural hydrocarbon emissions (e.g., isoprene) with temperature; and
(6) Relationships between high temperatures and stagnant circulation patterns.
(7) Advection of warm air enriched with O3.
AX2-94
-------
Cardelino and Chameides (1990) and Sillman and Samson (1995) both identified the
temperature-dependent thermal decomposition of PAN as the primary cause of the observed
O3-temperature relationship. When temperatures are low, PAN is relatively stable. Formation of
PAN represents a significant sink for NOX (in low NOX rural areas) and radicals (in high NOX
urban areas). This has the effect of slowing the rate of O3 production. Sillman and Samson
found that the impact of the PAN decomposition rate could explain roughly half of the observed
correlation between O3 and temperature. Jacob et al. (1993) found that warm events in summer
in the United States were likely to occur during stagnant meteorological conditions, and the
concurrence between warm temperatures and meteorological stagnation also explained roughly
half of the observed O3-temperature correlation. Other possible causes include higher solar
radiation during summer, the strong correlation between biogenic emission of isoprene and
temperature, and the somewhat weaker tendency for increased anthropogenic emissions
coinciding with warmer temperatures.
However, it should also be noted that a high correlation of O3 with temperature does not
necessarily imply a causal relation. Extreme episodes of high temperatures (a heat wave) are
often multiday events, high O3 episodes are also multiday events, concentrations build,
temperatures rise, but both are being influenced by larger-scale regional or synoptic
meteorological conditions. It also seems apparent, that while there is a trend for higher O3
associated with higher temperatures, there is also much greater variance in the range of O3
mixing ratios at higher temperatures.
AX2.4 THE RELATION OF OZONE TO ITS PRECURSORS AND
OTHER OXIDANTS
Ozone is unlike many other species whose rates of formation vary directly with the
emissions of their precursors. Ozone changes in a nonlinear fashion with the concentrations of
its precursors. At the low NOX concentrations found in most environments, ranging from remote
continental areas to rural and suburban areas downwind of urban centers the net production of O3
increases with increasing NOX. At the high NOX concentrations found in downtown metropolitan
areas, especially near busy streets and roadways, and in power plant plumes there is net
destruction (titration) of O3 by reaction with NO. In between these two regimes there is a
AX2-95
-------
transition stage in which O3 shows only a weak dependence on NOX concentrations. In the high
NOX regime, NO2 scavenges OH radicals which would otherwise oxidize VOCs to produce
peroxy radicals, which in turn would oxidize NO to NO2. In the low NOX regime, the oxidation
of VOCs generates, or at least does not consume, free radicals and O3 production varies directly
with NOX. Sometimes the terms VOC limited and NOX limited are used to describe these two
regimes. However, there are difficulties with this usage because (1) VOC measurements are not
as abundant as they are for nitrogen oxides, (2) rate coefficients for reaction of individual VOCs
with free radicals vary over an extremely wide range, and (3) consideration is not given to CO
nor to reactions that can produce free radicals without invoking VOCs. The terms NOx-limited
and NOx-saturated (e.g., Jaegle et al., 2001) will be used wherever possible to describe these two
regimes more adequately. However, the terminology used in original articles will also be kept.
The chemistry of OH radicals, which are responsible for initiating the oxidation of hydrocarbons,
shows behavior similar to that for O3 with respect to NOX concentrations (Hameed et al., 1979;
Pinto et al., 1993; Poppe et al., 1993; Zimmerman and Poppe, 1993). These considerations
introduce a high degree of uncertainty into attempts to relate changes in O3 concentrations to
emissions of precursors.
Various analytical techniques have been proposed that use ambient NOX and VOC
measurements to derive information about O3 production and O3-NOX-VOC sensitivity. It has
been suggested that O3 formation in individual urban areas could be understood in terms of
measurements of ambient NOX and VOC concentrations during the early morning (e.g., National
Research Council, 1991). In this approach, the ratio of summed (unweighted by chemical
reactivity) VOC to NOX is used to determine whether conditions were NOx-sensitive or VOC
sensitive. This procedure is inadequate because it omits many factors that are recognized as
important for O3 production: the impact of biogenic VOCs (which are not present in urban
centers during early morning); important individual differences in the ability of VOCs to
generate free radicals (rather than just total VOC) and other differences in O3 forming potential
for individual VOCs (Carter, 1995); the impact of multiday transport; and general changes in
photochemistry as air moves downwind from urban areas (Milford et al., 1994).
Jacob et al. (1995) used a combination of field measurements and a chemistry-transport
model (CTM) to show that the formation of O3 changed from NOx-limited to NOx-saturated as
the season changed from summer to fall at a monitoring site in Shenandoah National Park, VA.
AX2-96
-------
Photochemical production of O3 generally occurs simultaneously with the production of various
other species: nitric acid (HNO3), organic nitrates, and 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.
There are no hard and fast rules governing the levels of NOX at which the transition from
NOx-limited to NOx-saturated conditions occurs. The transition between these two regimes is
highly spatially and temporally dependent. Similar responses to NOX additions from commercial
aircraft have also been found for the upper troposphere (Bruhl et al., 2000). Bruhl et al. (2000)
found that the NOX levels for O3 production versus loss are highly sensitive to the radical sources
included in model calculations. They found that the inclusion of only CH4 and CO oxidation
leads to a decrease in net O3 production in the North Atlantic flight corridor due to NO emissions
from aircraft. However, the inclusion of acetone photolysis was found to shift the maximum in
O3 production to higher NOX mixing ratios, thereby reducing or eliminating areas in which there
is a decrease in O3 production rates due to aircraft emissions.
Trainer et al. (1993) suggested that the slope of the regression line between O3 and
summed NOX oxidation products (NOZ, equal to the difference between measured total reactive
nitrogen, NOy, and NOX) can be used to estimate the rate of O3 production per NOX (also known
as the O3 production efficiency, or OPE). Ryerson et al. (1998, 2001) used measured
correlations between O3 and NOZ to identify different rates of O3 production in plumes from
large point sources.
Sillman (1995) and Sillman and He (2002) identified several secondary reaction products
that show different correlation patterns for NOx-limited conditions 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 widely available. Measured O3 versus
NOZ (Figure AX2-17) shows distinctly different patterns in different locations. In rural areas and
in urban areas such as Nashville, TN, O3 shows a strong correlation with NOZ and a relatively
steep slope to the regression line. By contrast, in Los Angeles O3 also increases with NOZ, but
the rate of increase of O3 with NOZ is lower and the O3 concentrations for a given NOZ value are
generally lower.
AX2-97
-------
0
x
x
x
x
X
X
x—
X
X
X
X X
X
10
20
NOZ (ppb)
30
x
40
Figure AX2-17. Measured values of O3 and NOZ (NOy-NOx) during the afternoon at rural
sites in the eastern United States (gray circles) and in urban areas and
urban plumes associated with Nashville, TN (gray dashes), Paris, FR
(black diamonds) and Los Angeles, CA (X's).
Sources: Trainer etal. (1993), Sillmanetal. (1997, 1998, 2003).
The difference between NOx-limited and NOx-saturated regimes is also reflected in
measurements of hydrogen peroxide (H2O2). Hydrogen peroxide production is highly sensitive
to the abundance of free radicals and is thus favored in the NOx-limited regime, typical of
summer conditions. Differences between these two regimes are also related to the preferential
formation of sulfate during summer and to the inhibition of sulfate and hydrogen peroxide
during winter (Stein and Lamb, 2003). Measurements in the rural eastern United States (Jacob
et al., 1995) Nashville (Sillman et al., 1998), and Los Angeles (Sakugawa and Kaplan, 1989)
show large differences in H2O2 concentrations between likely NOx-limited and NOx-saturated
locations.
The discussion in Section AX2.4.1 centers mainly on the relations among O3, NOX and its
oxidation products, represented as NOZ (NOy-NOx) and VOCs derived from the results of field
studies. Most of these studies examined processes occurring in power plant and urban plumes.
AX2-98
-------
AX2.4.1 Summary of Results for the Relations Among Ozone, its Precursors
and Other Oxidants from Recent Field Experiments
AX2.4.1.1 Results from the Southern Oxidant Study and Related Experiments
The Southern Oxidant Study (SOS) was initiated to describe the sources, variation, and
distribution of O3 and its precursors in the southeastern United States during the summer season
(Hubler et al., 1998; Meagher et al., 1998; Goldan et al., 2000). Specific issues that were
addressed included: (1) the role of biogenic VOC and NOX emissions on local and regional O3
production, (2) the effect of urban-rural exchange/interchange on local and regional O3
production, (3) sub-grid-scale photochemical and meteorological processes, and (4) the
provision of a high-quality chemical and meteorological data set to test and improve observation
and emission-based air quality forecast models. Some of the more significant findings of the
1994 to 1995 studies include the following: (1) Ozone production in Nashville was found to be
close to the transition between NOx-limited and NOx-saturated regimes. (2) The number of
molecules of O3 produced per molecule of NOX oxidized in power plant plumes, or the O3
production efficiency (OPE) was found to be inversely proportional to the NOX emission rate,
with the plants having the highest NOX emissions exhibiting the lowest OPE. (3) During
stagnant conditions, winds at night dominated pollutant transport and represent the major
mechanism for advecting urban pollutants to rural areas—specific findings follow.
As part of SOS, the Tennessee Valley Authority's instrumented helicopter conducted
flights over Atlanta, Georgia to investigate the evolution of the urban O3 plume (Imhoff et al.,
1995). Ozone peak levels occurred at 20 - 40 km downwind of the city center. The OPE
obtained from five afternoon flights ranged between 4 and 10 molecules of O3 per molecule
ofNOx.
Berkowitz and Shaw (1997) measured O3 and its precursors at several altitudes over a
surface site near Nashville during SOS to determine the effects of turbulent mixing on
atmospheric chemistry. Early morning measurements of O3 aloft revealed values near 70 ppb,
while those measured at the surface were closer to 25 ppb. As the daytime mixed layer
deepened, surface O3 values steadily increased until they reached 70 ppb. The onset of
turbulence increased isoprene mixing ratios aloft by several orders of magnitude and affected the
slope of O3 as a function of NOy for each of the flight legs. Measurements from nonturbulent
flight legs yielded slopes that were considerably steeper than those from measurements made in
AX2-99
-------
turbulence. This study shows that the concentration of O3 precursors aloft is dependent on the
occurrence of turbulence, and turbulent mixing could explain the evolution of O3 concentrations
at the surface. In general, conclusions regarding pollutant concentrations must account for both
chemical and local dynamic processes.
Gillani et al. (1998) analyzed data from instrumented aircraft during SOS that flew through
the plumes of three large, tall-stack, base-load, Tennessee Valley Authority (TVA) coal-fired
power plants in northwestern Tennessee. They determined that plume chemical maturity and
peak O3 and NOZ production occurred within 30 to 40 km and 4 hours of summer daytime
convective boundary layer (CBL) transport time for a coal-fired power plant in the Nashville,
TN urban O3 nonattainment area (Gallatin). For a rural coal-fired power plant in an isoprene-
rich forested area about 100 km west of Nashville (Cumberland), plume chemical maturity and
peak O3 and NOZ production were realized within approximately 100 km and 6 hours of CBL
transport time. Their findings included approximately 3 molecules of O3 and more than
0.6 molecules of NOZ may be produced in large isolated rural power plant plumes (PPPs) per
molecule of NOX release; the corresponding peak yields of O3 and NOZ may be significantly
greater in urban PPPs. Both power plants can contribute as much as 50 ppb of excess O3 to the
Nashville area, raising the local levels to well above 100 ppb. Also using aircraft data collected
during SOS, Ryerson et al. (1998) concluded that the lower and upper limits to O3 production
efficiency in the Cumberland and Paradise PPPs (located in rural Tennessee) were 1 and
2 molecules of O3 produced per molecule of NOX emitted. The estimated lower and upper limits
to O3 production efficiency in the Johnsonville PPP (also located in rural Tennessee) were
higher, at 3 and 7.
The NOAA airborne O3 lidar provided detailed, three-dimensional lower tropospheric O3
distribution information during June and July 1995 in the Nashville area (Senff et al., 1998;
Alvarez et al., 1998). The size and shape of power plant plumes as well as their impacts on O3
concentration levels as the plume is advected downwind were studied. Specific examples
include: the July 7 Cumberland plume that was symmetrical and confined to the boundary layer,
and the July 19 Cumberland plume that was irregularly shaped with two cores, one above and
the other within the boundary layer. The disparate plume characteristics on these two days were
the result of distinctly different meteorological conditions. Ozone in the plume was destroyed at
a rate of 5 to 8 ppbv h'1 due to NOX titration close to the power plant, while farther downwind,
AX2-100
-------
O3 was produced at rates between 1.5 and 4 ppbv h"1. The lidar O3 measurements compared
reasonably well with in situ values, with the average magnitude of the offsets over all the flights
at 4.3 ppbv (7%).
The highest O3 concentrations observed during the 1995 SOS in middle Tennessee
occurred during a period of strong, synoptic-scale stagnation from July 11 through July 15. This
massive episode covered most of the eastern United States (e.g., Ryan et al., 1998). During this
time, the effects of vertical wind profiles on the buildup and transport of O3 were studied by
Banta et al. (1998) using an airborne differential absorption lidar (DIAL) system. Vertical cross
sections showed O3 concentrations exceeding 120 ppb extending to nearly 2 km above ground
level, but that O3 moved little horizontally. Instead, it formed a dome of pollution over or near
Nashville. Due to the stagnant daytime conditions (boundary layer winds ~1 to 3 m s"1),
nighttime transport of O3 became disproportionately important. At night, in the layer between
100 and 2000 m AGL (which had been occupied by the daytime mixed layer), the winds could
be accelerated to 5 to 10 m s"1 as a result of nocturnal decoupling from surface friction. Data
from surface and other aircraft measurements taken during this period suggest that the
background air and the edges of the urban plume were NOX sensitive and the core of the urban
plume was hydrocarbon sensitive (Valente et al., 1998). Also revealed was the fact that the
surface monitoring network failed to document the maximum surface O3 concentrations. Thus,
monitoring networks, especially in medium-sized urban areas under slow transport conditions,
may underestimate the magnitude and frequency of urban O3 concentrations greater than
120 ppb.
Nunnermacker et al. (1998) used both aircraft and surface data from SOS to perform a
detailed kinetic analysis of the chemical evolution of the Nashville urban plume. The analysis
revealed OH concentrations around 1.2 x 107 cm"3 that consumed 50% of the NOX within
approximately 2 hours, at an OPE of 2.5 to 4 molecules for each molecule of NOX.
Anthropogenic hydrocarbons provided approximately 44% of the fuel for O3 production by the
urban plume.
Surface and aircraft observations of O3 and O3 precursors were compared during SOS to
assess the degree to which midday surface measurements may be considered representative of
the larger planetary boundary layer (PEL) (Luke et al., 1998). Overall agreement between
surface and aircraft O3 measurements was excellent in the well-developed mixed layer
AX2-101
-------
(r2 = 0.96), especially in rural-regional background air and under stagnant conditions, where
surface concentrations change only slowly. Vertical variations in trace gas concentrations were
often minimal in the well-mixed PEL, and measurements at the surface always agreed well with
aircraft observations up to the level of measurements (460 m above ground level). Under
conditions of rapidly varying surface concentrations (e.g., during episodes of power plant plume
fumigation and early morning boundary layer development), agreement between surface and
aloft was dependent upon the spatial (aircraft) and temporal (ground) averaging intervals used in
the comparison. Under these conditions, surface sites were representative of the PEL only to
within a few kilometers horizontally.
On four days during SOS, air samples were taken in the plume of the Cumberland Power
Plant in central Tennessee using an instrumented helicopter to investigate the evolution of
photochemical smog (Luria et al., 1999, 2000). Twelve crosswind air-sampling traverses were
made between 35 and 116 km from this Power Plant on 16 July 1995. Winds, from the west-
northwest during the sampling period, directed the plume toward Nashville. Ten of the traverses
were performed upwind of Nashville, where the plume was isolated, and two were made
downwind of the city. The results indicated that even six hours after the plume left the stacks,
excess O3 production was limited to the edges of the plume. Excess O3 production within the
plume was found to vary from 20 ppb up to 55 ppb. It was determined that this variation
corresponded to differences in ambient isoprene levels. Excess O3 (up to 109 ppbv, 50 to
60 ppbv above background), was produced in the center of the plume when there was sufficient
mixing upwind of Nashville. The power plant plume apparently mixed with the urban plume
also, producing O3 up to 120 ppbv 15 to 25 km downwind of Nashville.
Nunnermacker et al. (2000) used data from the DOE G-l aircraft to characterize emissions
from a small power plant plume (Gallatin) and a large power plant plume (Paradise) in the
Nashville region. Observations made on July 3, 7, 15, 17, and 18, 1995, were compiled, and a
kinetic analysis of the chemical evolution of the power plant plumes was performed. OPEs were
found to be 3 in the Gallatin and 2 in the Paradise plumes. Lifetimes for NOX (2.8 and 4.2 hours)
and NOy (7.0 and 7.7 hours) were determined in the Gallatin and Paradise plumes, respectively.
These NOX and NOy lifetimes imply rapid loss of NOZ (assumed to be primarily HNO3), with a
lifetime determined to be 3.0 and 2.5 hours for the Gallatin and Paradise plumes, respectively.
AX2-102
-------
AX2.4.1.2 Results from Studies on Biogenic and Anthropogenic Hydrocarbons and
Ozone Production
Williams et al. (1997) made the first airborne measurements of peroxy-methacrylic nitric
anhydride (MPAN), which is formed from isoprene-NOx chemistry and is an indicator of recent
O3 production from isoprene and therefore biogenic hydrocarbons (BHC). They also measured
peroxyacetic nitric anhydride (PAN), peroxypropionic nitric anhydride (PPN), and O3 to estimate
the contributions of anthropogenic hydrocarbons (AHC) and BHC to regional tropospheric O3
production.
Airborne measurements of MPAN, PAN, PPN, and O3 were made during the 1994 and
1995 Nashville intensive studies of SOS to determine the fraction of O3 formed from
anthropogenic NOX and BHC (Roberts et al., 1998). It was found that PAN, a general product of
hydrocarbon-NOx photochemistry, could be well represented as a simple linear combination of
contributions from BHC and AHC as indicated by MPAN and PPN, respectively. The
PAN/MPAN ratios, characteristic of BHC-dominated chemistry, ranged from 6 to 10. The
PAN/PPN ratios, characteristic of AHC-dominated chemistry, ranged from 5.8 to 7.4. These
ratios were used to estimate the contributions of AHC and BHC to regional tropospheric O3
production. It was estimated that substantial O3 (50 to 60 ppbv) was produced from BHC when
high NOX from power plants was present in areas of high BHC emission.
AX2.4.1.3 Results of Studies on Ozone Production in Mississippi and Alabama
Aircraft flights made in June 1990 characterized the variability of O3 and reactive nitrogen
in the lower atmosphere over Mississippi and Alabama. The variety and proximity of sources
and the photochemical production and loss of O3 were found to be contributing factors (Ridley
et al., 1998). Urban, biomass burning, electrical power plant, and paper mill plumes were all
encountered during these flights. Urban plumes from Mobile, AL had OPEs as high as 6 to
7 ppbv O3 per ppbv of NOX. Emissions measured from biomass burning had lower efficiencies
of 2 to 4 ppbv O3 per ppbv of NOX, but the average rate of production of O3 was as high as
58 ppbv hr1 for one fire where the plume was prevented from vertical mixing. Near-source
paper mill and power plant plumes showed O3 titration, while far-field observations of power
plant plumes showed net O3 production. Early morning observations below a nocturnal
inversion provided evidence for the nighttime oxidation of NOX to reservoir species.
AX2-103
-------
Aircraft measurements of O3 and oxides of nitrogen were made downwind of Birmingham,
AL to estimate the OPE in the urban plume (Trainer et al., 1995). NOX emission rates were
estimated at 0.6 x 1025 molecules s"1 with an uncertainty of a factor of 2. During the
summertime it was determined that approximately seven O3 molecules could be formed for
every molecule of NOX emitted by the urban and proximately located power plant plumes. The
regional O3, the photochemical production of O3 during the oxidation of the urban emissions, and
wind speed and direction all combined to dictate the magnitude and location of the peak O3
concentrations observed in the vicinity of the Birmingham metropolitan area.
Aircraft observations of rural U.S. coal-fired power plant plumes in the middle Mississippi
and Tennessee Valleys were used to quantify the nonlinear dependence of tropospheric O3
formation on plume NOX concentration, determined by plant NOX emission rate and atmospheric
dispersion (Ryerson et al., 2001). The ambient availability of reactive VOCs, primarily biogenic
isoprene, was also found to affect O3 production rate and yield in these rural plumes. Plume O3
production rates and yields as a function of NOX and VOC concentrations differed by a factor of
2 or more. These large differences indicate that power plant NOX emission rates and geographic
locations play a large role in tropospheric O3 production.
AX2.4.1.4 The Nocturnal Urban Plume over Portland, Oregon
Aircraft observations of aerosol surface area, O3, NOy and moisture were made at night in
the Portland, Oregon urban plume (Berkowitz et al., 2001). Shortly after sunset, O3, relative
humidity, NOy and aerosol number density were all positively correlated. However, just before
dawn, O3 mixing ratios were highly anti-correlated with aerosol number density, NOy and
relative humidity. Back-trajectories showed that both samples came from a common source to
the northwest of Portland. The predawn parcels passed directly over Portland, while the other
parcels passed to the west of Portland. Several hypotheses were put forward to explain the loss
of O3 in the parcels that passed over Portland, including homogeneous gas-phase mechanisms
and a heterogeneous mechanism on the aerosol particle surface.
AX2.4.1.5 Effects of VOCs in Houston on Ozone Production
Aircraft Observations of O3 and O3 precursors over Houston, TX, Nashville, TN; New
York, NY; Phoenix, AZ, and Philadelphia, PA showed that despite similar NOX concentrations,
AX2-104
-------
high concentrations of VOCs in the lower atmosphere over Houston led to calculated O3
production rates that were 2 to 5 times higher than in the other 4 cities (Kleinman et al., 2002).
Concentrations of VOCs and O3 production rates are highest in the Ship Channel region of
Houston, where one of the largest petrochemical complexes in the world is located. As a result,
Houston lays claim to the highest recorded hourly average O3 concentrations in the United States
within the last 5 years (in excess of 250 ppb).
AX2.4.1.6 Chemical and Meteorological Influences on the Phoenix Urban Ozone Plume
The interaction of chemistry and meteorology for western cities can contrast sharply with
that of eastern cities. A 4-week field campaign in May and June of 1998 in the Phoenix area
comprised meteorological and chemical measurements (Fast et al., 2000). Data from models and
observations revealed that heating of the higher terrain north and east of Phoenix produced
regular, thermally driven circulations during the afternoon from the south and southwest through
most of the boundary layer, advecting the urban O3 plume to the northeast. Deep mixed layers
and moderate winds aloft ventilated the Phoenix area during the study period so that multiday
buildups of locally produced O3 did not appear to contribute significantly to O3 levels.
Sensitivity simulations estimated that 20% to 40% of the afternoon surface O3 mixing ratios
(corresponding to 15 to 35 ppb) was due to the entrainment of O3 reservoirs into the growing
convective boundary layer. The model results also indicated that O3 production in this arid
region is NOx-saturated, unlike most eastern U.S. sites.
AX2.4.1.7 Transport of Ozone and Precursors on the Regional Scale
Instrumented aircraft flights by the University of Maryland in a Cessna 172 and Sonoma
Technology, Inc. in a Piper Aztec measured the vertical profiles of trace gases and
meteorological parameters in Virginia, Maryland, and Pennsylvania on July 12-15, 1995 during
a severe O3 episode in the mid-Atlantic region (Ryan et al., 1998). Ozone measured upwind of
the urban centers reached 80 to 110 ppbv. Layers of high O3 aloft were associated with local
concentration maxima of SO2 and NOy, but not CO or NOX. This, together with a back trajectory
analysis, implicated coal-fired power plants in the industrialized Midwest as the source of the
photochemically aged air in the upwind boundary of the urban centers. When the PEL over the
Baltimore-Washington area deepened, the O3 and O3 precursors that had been transported from
AX2-105
-------
the west and northwest mixed with the local emissions and O3 in excess of 125 ppbv was
measured at the surface.
During the blackout of August 14, 2003 Marufu et al. (2004) measured profiles of O3, SO2
and CO over areas in western Pennsylvania, Maryland and Virginia. They found notable
decreases in O3, SO2, and NOX, over areas affected by the blackout but not over those that were
not affected. They also found that CO concentrations aloft were comparable over areas affected
and not affected by the blackout. They attributed the differences in concentrations between what
was observed and what was expected to the reduction in emissions from power plants mainly in
the Ohio Valley. They also reasoned that the CO concentrations were relatively unaffected
because they arise from traffic emissions, which may have been largely unaffected by the
blackout. However, the blackout also disrupted many industries, small scale emission sources,
and rail and air transportation.
The Department of Energy G-l aircraft flew in the New York City metropolitan area in
the summer of 1996 as part of the North American Research Strategy for Tropospheric
Ozone-Northeast effort to ascertain the causes leading to high O3 levels in the northeastern
United States (Kleinman et al., 2000). Ozone, O3 precursors, and other photochemically active
trace gases were measured upwind and downwind of New York City to characterize the O3
formation process and its dependence on NOX and VOCs. During two flights, the wind was
south southwesterly and O3 levels reached 110 ppb. On two other flights, the wind was from the
north-northwest and O3 levels were not as high. When the G-l observed O3 around 110 ppb, the
NOx/NOy ratio measured at the surface was between 0.20 and 0.30, indicating an aged plume.
AX2.4.1.8 Model Calculations and Aircraft Observations of Ozone Over Philadelphia
Regional-scale transport and local O3 production over Philadelphia was estimated using a
new meteorological-chemical model (Fast et al., 2002). Surface and airborne meteorological and
chemical measurements made during a 30-day period in July and August of 1999 as part of the
Northeast Oxidant and Particulate Study were used to evaluate the model performance. Both
research aircraft and ozonesondes, during the morning between 0900 and 1100 LST, measured
layers of O3 above the convective boundary layer. The model accounted for these layers through
upwind vertical mixing the previous day, subsequent horizontal transport aloft, and NO titration
of O3 within the stable boundary layer at night. Entrainment of the O3 aloft into the growing
AX2-106
-------
convective boundary contributed to surface O3 concentrations. During the study period, most of
the O3 appeared to result from local emissions in the vicinity of Philadelphia and the Chesapeake
Bay area, but during high O3 episodes, up to 30 to 40% of the O3 was due to regional transport
from upwind sources.
AX2.4.1.9 The Two-Reservoir System
Studies described above and over 500 aircraft flights over the mid-Atlantic region show
that a two-reservoir system illustrated schematically in Figure AX2-18 may represent both the
dynamics and photochemistry of severe, multiday haze and O3 episodes over the eastern United
States (Taubman et al., 2004, 2005). The first reservoir is the PEL, where most precursor
species are injected, and the second is the lower free troposphere (LFT), where photochemical
processes are accelerated and removal via deposition is rare. Bubbles of air lifted from urban
and industrial sources were rich in CO and SO2, but not O3, and contained greater numbers of
externally mixed primary sulfate and black carbon (BC) particles. Correlations among O3, air
parcel altitude, particle size, and relative humidity suggest that greater O3 concentrations and
relatively larger particles are produced in the LFT and mix back down into the PEL. Backward
trajectories indicated source regions in the Midwest and mid-Atlantic urban corridor, with
southerly transport up the urban corridor augmented by the Appalachian lee trough and nocturnal
low-level jet (LLJ). This concept of two-reservoirs may facilitate the numerical simulation of
multiday events in the eastern United States. A relatively small number of vertical layers will be
required if accurate representation of the sub-gridscale transport can be parameterized to
represent the actual turbulent exchange of air between the PEL and lower free troposphere.
AX2.5 METHODS USED TO CALCULATE RELATIONS BETWEEN
OZONE AND ITS PRECURSORS
Atmospheric chemistry and transport models are the major tools used to calculate the
relations between O3, its precursors, and other oxidation products. Other techniques, involving
statistical relations between O3 and other variables have also been used. Chemistry-transport
models (CTM) are driven by emissions inventories for O3 precursor compounds and by
meterological fields. Emissions of precursor compounds can be divided into anthropogenic and
AX2-107
-------
Two Reservoir Model
>
QJ
NO
SO2
O
CO
0
BC
0
0
> Actinic
Flux
a £1.9
.•' < Actinic
Flux
a a 1.9
0
VOCs Primary
Particles
Figure AX2-18. Conceptual two-reservoir model showing conditions in the PEL and
in the lower free troposphere during a multiday O3 episode. The dividing
line, the depth of the mixed layer, is about 1000 m. Emissions occur in
the PEL, where small, unmixed black carbon, sulfate, and crustal
particles in the PM2 5 size range are also shown. Ozone concentrations as
well as potential temperature (0) and actinic flux are lower in the PEL
than in the lower free troposphere, while RH is higher. Larger, mixed
sulfate and carbonaceous particles (still in the PM2 5 size range) and more
O3 exist in the lower free troposphere.
Source: Taubman et al. (2004, 2005).
natural source categories. Natural sources can be further divided into biotic (vegetation,
microbes, animals) and abiotic (biomass burning, lightning) categories. However, the distinction
between natural sources and anthropogenic sources is often difficult to make as human activities
affect directly, or indirectly, emissions from what would have been considered natural sources
during the preindustrial era. Emissions from plants and animals used in agriculture are usually
referred to as anthropogenic. Wildfire emissions may be considered natural, except that forest
management practices may have led to the buildup of fuels on the forest floor, thereby altering
the frequency and severity of forest fires. Needed meteorological quantities such as winds and
AX2-108
-------
temperatures are taken from operational analyses, reanalyses, or circulation models. In most
cases, these are off-line analyses, i.e., they are not modified by radiatively active species such as
O3 and particles generated by the model.
A brief overview of atmospheric chemistry-transport models is given in Section AX2.5.1.
A discussion of emissions inventories of precursors that are used by these models is given in
Section AX2.5.2. Uncertainties in emissions estimates have also been discussed in Air Quality
Criteria for Particulate Matter (U.S. Environmental Protection Agency, 2000). So-called
"observationally based models" which rely more heavily on observations of the concentrations
of important species are discussed in Section AX2.5.3. Chemistry-transport model evaluation
and an evaluation of the reliability of emissions inventories are presented in Section AX2.5.4.
AX2.5.1 Chemistry-Transport Models
Atmospheric chemistry-transport models (CTMs) are used to obtain better understanding
of the processes controlling the formation, transport, and destruction of O3 and other air
pollutants; to understand the relations between O3 concentrations and concentrations of its
precursors such as NOX and VOCs; and to understand relations among the concentration patterns
of O3 and other oxidants that may also exert health effects. Detailed examination of the
concentrations of short-lived species in a CTM can provide important insights into how O3 is
formed under certain conditions and can suggest likely avenues for data analysis and future
experiments and field campaigns. The dominant processes leading to the formation of O3 in a
particular time period, questions about whether NOX or VOCs were more important, the
influence of meteorology and of emissions from a particular geographic region, and the
transformation or formation of other pollutants could be examined using a CTM.
CTMs are also used for determining control strategies for O3 precursors. However, this
application has met with varying degrees of success because of the highly nonlinear relations
between O3 and emissions of its precursors. CTMs include mathematical descriptions of
atmospheric transport, emissions, the transfer of solar radiation through the atmosphere,
chemical reactions, and removal to the surface by turbulent motions and precipitation for
chemical species of interest. Increasingly, the trend is for these processes to be broken down and
handled by other models or sub-models, so a CTM will likely use emissions and meteorological
data from at least two other models.
AX2-109
-------
There are two major formulations of CTMs in current use. In the first approach,
grid-based, or Eulerian, air quality models, the region to be modeled (the modeling domain) is
subdivided into a three-dimensional array of grid cells. Spatial derivatives in the species
continuity equations are cast in finite-difference form over this grid, and a system of equations
for the concentrations of all the chemical species in the model are solved numerically at each
grid point. The modeling domain may be limited to a particular airshed or provide global
coverage and extend through several major atmospheric layers. Time dependent continuity
(mass conservation) equations are solved for each species including terms for transport, chemical
production and destruction, and emissions and deposition (if relevant), in each cell. Chemical
processes are simulated with ordinary differential equations, and transport processes are
simulated with partial differential equations. Because of a number of factors such as the
different time scales inherent in different processes, the coupled, nonlinear nature of the
chemical process terms, and computer storage limitations, all of the terms in the equations are
not solved simultaneously in three dimensions. Instead, a technique known as operator splitting,
in which terms involving individual processes are solved sequentially, is used. In the second
application of CTMs, trajectory or Lagrangian models, a large number of hypothetical air parcels
are specified as following wind trajectories. In these models, the original system of partial
differential equations is transformed into a system of ordinary differential equations.
A less common approach is to use a hybrid Lagrangian/Eulerian model, in which certain
aspects of atmospheric chemistry and transport are treated with a Lagrangian approach and
others are treaded in a Eulerian manner (e.g., Stein et al., 2000). Both modeling approaches have
their advantages and disadvantages. The Eulerian approach is more general in that it includes
processes that mix air parcels and allows integrations to be carried out for long periods during
which individual air parcels lose their identity. There are, however, techniques for including the
effects of mixing in Lagrangian models such as FLEXPART (e.g., Zanis et al., 2003), ATTILA
(Reithmeir and Sausen, 2002), and CLaMS (McKenna et al., 2002).
Major modeling efforts within the U.S. Environmental Protection Agency center on the
ModelsS/Community Modeling for Air Quality (CMAQ, Byun et al., 1998) and the Multi Scale
Air Quality Simulation Platform (MAQSIP, Odman and Ingram, 1996) whose formulations are
based on the regional acid deposition model (RADM, Chang et al., 1987). A number of other
modeling platforms using the Lagrangian and Eulerian frameworks have been reviewed in
AX2-110
-------
AQCD 96. CTMs currently in use are summarized in the review by Russell and Dennis (2000).
Evaluations of the performance of CMAQ are given in Arnold et al. (2003) and Fuentes and
Raftery (2005). The domains of MAQSIP and CMAQ are flexible and can extend from several
hundred km to the hemispherical scale. In addition, both of these classes of models allow the
resolution of the calculations over specified areas to vary. CMAQ and MAQSIP are both driven
by the MM5 mesoscale meteorological model (Seaman, 2000), though both may be driven by
other meteorological models (e.g., RAMS and Eta). Simulations of regional O3 episodes have
been performed with a horizontal resolution of 4 km. In principle, calculations over limited
domains can be accomplished to even finer scales. However, simulations at these higher
resolutions require better parameterizations of meteorological processes such as boundary layer
fluxes, deep convection and clouds (Seaman, 2000), and knowledge of emissions. Resolution at
finer scales will likely be necessary to resolve smaller-scale features such as the urban heat
island; sea, bay, and land breezes; and the nocturnal low-level jet.
Currently, the most common approach to setting up the horizontal domain is to nest a finer
grid within a larger domain of coarser resolution. However, a number of other strategies are
currently being developed, such as the stretched grid (e.g., Fox-Rabinowitz et al., 2002) and the
adaptive grid. In a stretched grid, the grid's resolution continuously varies throughout the
domain, thereby eliminating any potential problems with the sudden change from one resolution
to another at the boundary. One must be careful in using such a formulation, because certain
parameterizations that are valid on a relatively coarse grid scale (such as convection, for
example) are not valid or should not be present on finer scales. Adaptive grids are not set at the
start of the simulation, but instead adapt to the needs of the simulation as it evolves (e.g., Hansen
et al., 1994). They have the advantage that, if the algorithm is properly set up, the resolution is
always sufficient to resolve the process at hand. However, they can be very slow if the situation
to be modeled is complex. Additionally, if one uses adaptive grids for separate meteorological,
emissions, and photochemical models, there is no reason a priori why the resolution of each grid
should match; and the gains realized from increased resolution in one model will be wasted in
the transition to another model. The use of finer and finer horizontal resolution in the
photochemical model will necessitate finer-scale inventories of land use and better knowledge of
the exact paths of roads, locations of factories, and, in general, better methods for locating
AX2-111
-------
sources. The present practice of locating a source in the middle of a county or distributing its
emissions throughout a county if its location is unknown will likely not be adequate in the future.
The vertical resolution of these models continues to improve as more layers are added to
capture atmospheric processes and structures. This trend will likely continue because a model
with 25 vertical layers, for example, may have layers that are 500 m thick at the top of the
planetary boundary layer. Though the boundary layer height is generally determined through
other methods, the chemistry in the model is necessarily confined by such layering schemes.
Because the height of the boundary layer is of critical importance in simulations of air quality,
improved resolution of the boundary layer height would likely improve air quality simulations.
The difficulty of properly establishing the boundary layer height is most pronounced when
considering tropopause folding events, which are important in determining the chemistry of the
background atmosphere. In the vicinity of the tropopause, the vertical resolution of most any
large scale model is quite unlikely to be able to capture such a feature. Additionally, any current
model is likely to have trouble adequately resolving fine scale features such as the low-level jet.
Finally, models must be able to treat emissions, meteorology, and photochemistry differently in
different areas. Emissions models are likely to need better resolution near the surface and
possibly near any tall stacks. Photochemical models, on the other hand, may need better
resolution away from the surface and be more interested in resolving the planetary boundary
layer height, terrain differences, and other higher altitude features. Meteorological models share
some of the concerns of photochemical models, but are less likely to need sufficient resolution to
adequately treat a process such as dry deposition beneath a stable nocturnal boundary layer.
Whether the increased computational power necessary for such increases in resolution will be
ultimately justified by improved results in the meteorological and subsequent photochemical
simulations remains to be seen.
CTMs require time dependent, three-dimensional wind fields for the time period of
simulation. The winds may be either generated by a model using initial fields alone or four
dimensional data assimilation can be used to improve the model's meteorological fields (i.e.,
model equations can be updated periodically [or "nudged"] to bring results into agreement with
observations). Most modeling efforts have focused on simulations of several days duration (a
typical time scale for individual O3 episodes), but there have been several attempts at modeling
longer periods. For example, Kasibhatla and Chameides (2000) simulated a four month period
AX2-112
-------
from May to September of 1995 using MAQSIP. The current trend appears to be toward
simulating longer time periods. This will impose additional strains on computational resources,
as most photochemical modeling until recently has been performed with an eye toward
simulating only summertime episodes of peak O3. With the shift toward modeling an entire year
being driven by the desire to understand observations of periods of high wintertime PM (e.g.,
Blanchard et al., 2002), models will be further challenged to simulate air quality under
conditions for which they may not have been used previously.
Chemical kinetics mechanisms (a set of chemical reactions) representing the important
reactions that occur in the atmosphere are used in air quality models to estimate the net rate of
formation of each pollutant simulated as a function of time. Chemical mechanisms that
explicitly treat the chemical reactions of each individual reactive species are too lengthy and
demanding of computer resources to be incorporated into three-dimensional atmospheric models.
As an example, a master chemical mechanism includes approximately 10,500 reactions
involving 3603 chemical species (Derwent et al., 2001). Instead, "lumped" mechanisms, that
group compounds of similar chemistry together, are used. The chemical mechanisms used in
existing photochemical O3 models contain significant uncertainties that may limit the accuracy
of their predictions; the accuracy of each of these mechanisms is also limited by missing
chemistry. Because of different approaches to the lumping of organic compounds into surrogate
groups, chemical mechanisms, can produce somewhat different results under similar conditions.
The CB-IV chemical mechanism (Gery et al., 1989), the RADM II mechanism (Stockwell et al.,
1990), the SAPRC (e.g., Wang et al., 2000a,b; Carter, 1990) and the RACM mechanisms can be
used in CMAQ. Jimenez et al. (2003) provide brief descriptions of the features of the main
mechanisms in use and they compared concentrations of several key species predicted by seven
chemical mechanisms in a box model simulation over 24 h. The average deviation from the
average of all mechanism predictions for O3 and NO over the daylight period was less than 20%,
and 10% for NO2 for all mechanisms. However, much larger deviations were found for HNO3,
PAN, HO2, H2O2, C2H4 and C5H8 (isoprene). An analysis for OH radicals was not presented.
The large deviations shown for most species imply differences between the calculated lifetimes
of atmospheric species and the assignment of model simulations to either NOX limited or radical
limited regimes between mechanisms. Gross and Stockwell (2003) found small differences
between mechanisms for clean conditions with differences becoming more significant for
AX2-113
-------
polluted conditions, especially for NO2 and organic peroxy radicals. They caution modelers to
consider carefully the mechanisms they are using.
As CTMs incorporate more processes and knowledge of aerosol- and gas-phase chemistry
improves, a "one atmosphere" approach is evolving. For example, CMAQ and PM-CAMx now
incorporate some aerosol processes, and several attempts are currently underway to study
feedbacks of chemistry on atmospheric dynamics using meteorological models, usually MM5
(e.g., Grell et al., 2000; Liu et al., 2001a; Lu et al., 1997; Park et al., 2001). This coupling may
be necessary to accurately simulate cases such as the heavy aerosol loading found in forest fire
plumes (Lu et al., 1997; Park et al., 2001).
Spatial and temporal characterizations of anthropogenic and biogenic precursor emissions
must be specified as inputs to a CTM. Emissions inventories have been compiled on grids of
varying resolution for many hydrocarbons, aldehydes, ketones, CO, NH3, and NOX. Emissions
inventories for many species require the application of some algorithm for calculating the
dependence of emissions on physical variables such as temperature. For many species,
information concerning the temporal variability of emissions is lacking, so long term (e.g.,
annual or O3-season) averages are used in short term, episodic simulations. Annual emissions
estimates are often modified by the emissions model to produce emissions more characteristic of
the time of day and season. Significant errors in emissions can occur if an inappropriate time
dependence or a default profile is used. Additional complexity arises in model calculations
because different chemical mechanisms are based on different species, and inventories
constructed for use with another mechanism must be adjusted to reflect these differences. This
problem also complicates comparisons of the outputs of these models because one chemical
mechanism will necessarily produce species that are different from those in another and neither
output will necessarily agree with the measurements.
The effects of clouds on atmospheric chemistry are complex and introduce considerable
uncertainty into CTM calculations. Thunderstorm clouds are optically very thick and have
major effects on radiative fluxes and thus on photolysis rates. Madronich (1987) provided
modeling estimates of the effects of clouds of various optical depths on photolysis rates. In the
upper portion of a thunderstorm anvil, photolysis is likely to be enhanced (as much as a factor of
2 or more) due to multiple reflections off the ice crystals. In the lower portion of the cloud and
beneath the cloud, photolysis is substantially decreased. Thunderstorm updrafts, which contain
AX2-114
-------
copious amounts of water, are regions where efficient scavenging of soluble species occurs
(Balkanski et al., 1993). Direct field measurements of the amounts of specific trace gases
scavenged in observed storms are sparse. Pickering et al. (2001) used a combination of model
estimates of soluble species that did not include wet scavenging and observations of these
species from the upper tropospheric outflow region of a major line of convection observed near
Fiji. Over 90% of the nitric acid and hydrogen peroxide in the outflow air appeared to have been
removed by the storm. Walcek et al. (1990) included a parameterization of cloud-scale aqueous
chemistry, scavenging, and vertical mixing in the regional scale, chemistry-transport model of
Chang et al. (1987). The vertical distribution of cloud microphysical properties and the amount
of subcloud-layer air lifted to each cloud layer were determined using a simple entrainment
hypothesis (Walcek and Taylor, 1986). Vertically-integrated O3 formation rates over the
northeastern United States were enhanced by -50% when the in-cloud vertical motions were
included in the model.
In addition to wet deposition, dry deposition (the removal of chemical species from the
atmosphere by interaction with ground-level surfaces) is an important removal process for
pollutants on both urban and regional scales and must be included in CTMs. The general
approach used in most models is the three-resistance method, in which where dry deposition is
parameterized with a deposition velocity, which is represented as vd = (ra + rb + rj"1 where ra, rb,
and rc represent the resistance due to atmospheric turbulence, transport in the fluid sublayer very
near the elements of surface such as leaves or soil, and the resistance to uptake of the surface
itself. This approach works for a range of substances although it is inappropriate for species
with substantial emissions from the surface or for species whose deposition to the surface
depends on its concentration at the surface itself. The approach is also modified somewhat for
aerosols: the terms rb and rc are replaced with a surface deposition velocity to account for
gravitational settling. In their review, Wesley and Hicks (2000) point out several shortcomings
of current knowledge of dry deposition. Among those shortcomings are difficulties in
representing dry deposition over varying terrain where horizontal advection plays a significant
role in determining the magnitude of ra and difficulties in adequately determining a deposition
velocity for extremely stable conditions such as those occurring at night (e.g., Mahrt, 1998).
Under the best of conditions, when a model is exercised over a relatively small area where dry
deposition measurements have been made, models still commonly show uncertainties at least as
AX2-115
-------
large as ± 30% (e.g., Massman et al., 1994; Brook et al., 1996; Padro, 1996). Wesley and
Hicks (2000) state that an important result of these comparisons is that the current level of
sophistication of most dry deposition models is relatively low and relies heavily on empirical
data. Still larger uncertainties exist when the surface features are not well known or when the
surface comprises a patchwork of different surface types, as is common in the eastern United
States.
The initial conditions, i.e., the concentration fields of all species computed by a model, and
the boundary conditions, i.e., the concentrations of species along the horizontal and upper
boundaries of the model domain throughout the simulation must be specified at the beginning of
the simulation. It would be best to specify initial and boundary conditions according to
observations. However, data for vertical profiles of most species of interest are sparse.
Ozonesonde data have been used to specify O3 fields, but the initial and boundary values of
many other species are often set equal to zero because of a lack of observations. Further,
ozonesondes are thought to be subject to errors in measurement and differences arising from
improper corrections for pump efficiency and the solutions used (e.g., Hilsenrath et al., 1986;
Johnson et al., 2002). The results of model simulations over larger, preferably global, domains
can also be used. As may be expected, the influence of boundary conditions depends on the
lifetime of the species under consideration and the time scales for transport from the boundaries
to the interior of the model domain (Liu et al., 2001b).
Each of the model components described above has an associated uncertainty, and the
relative importance of these uncertainties varies with the modeling application. The largest
errors in photochemical modeling are still thought to arise from the meteorological and
emissions inputs to the model (Russell and Dennis, 2000). Within the model itself, horizontal
advection algorithms are still thought to be significant source of uncertainty (e.g., Chock and
Winkler, 1994) though more recently those errors are thought to have been reduced (e.g., Odman
et al., 1996). There are also indications that problems with mass conservation continue to be
present in photochemical and meteorological models (e.g., Odman and Russell, 1999); these can
result in significant simulation errors. Uncertainties in meteorological variables and emissions
can be large enough that they would lead one to make the wrong decision when considering
control strategies (e.g., Russell and Dennis, 2000; Sillman et al., 1995). The effects of errors in
initial conditions can be minimized by including several days "spin-up" time in a simulation to
AX2-116
-------
allow species to come to chemical equilibrium with each other before the simulation of the
period of interest begins.
While the effects of poorly specified boundary conditions propagate through the model's
domain, the effects of these errors remain undetermined. Many regional models specify constant
O3 profiles (e.g., 35 ppb) at their lateral and upper boundaries; ozonesonde data, however,
indicate that the mixing ratio of O3 increases vertically in the troposphere (to over 100 ppb at the
tropopause) and into the stratosphere (e.g., Newchurch et al., 2003). The practice of using
constant O3 profiles strongly reduces the potential effects of vertical mixing of O3 from above
the planetary boundary layer (via mechanisms outlined in Section AX2.3) on surface O3 levels.
The use of an O3 climatology (e.g., Fortuin and Kelder, 1998) might reduce the errors that would
otherwise be incurred. Because many meteorological processes occur on spatial scales which
are smaller than the grid spacing (either horizontally or vertically) and thus are not calculated
explicitly, parameterizations of these processes must be used and these introduce additional
uncertainty.
Uncertainty also arises in modeling the chemistry of O3 formation because it is highly
nonlinear with respect to NOX concentrations. Thus, the volume of the grid cell into which
emissions are injected is important because the nature of O3 chemistry (i.e., O3 production or
titration) depends in a complicated way on the concentrations of the precursors and the OH
radical. The use of ever-finer grid spacing allows regions of O3 titration to be more clearly
separated from regions of O3 production. The use of grid spacing fine enough to resolve the
chemistry in individual power-plant plumes is too demanding of computer resources for this to
be attempted in most simulations. Instead, parameterizations of the effects of subgrid scale
processes such as these must be developed; otherwise serious errors can result if emissions are
allowed to mix through an excessively large grid volume before the chemistry step in a model
calculation is performed. In light of the significant differences between atmospheric chemistry
taking place inside and outside of a power plant plume (e.g., Ryerson et al., 1998 and Sillman,
2000), inclusion of a separate, meteorological module for treating large, tight plumes is
necessary. Because the photochemistry of O3 and many other atmospheric species is nonlinear,
emissions correctly modeled in a tight plume may be incorrectly modeled in a more dilute
plume. Fortunately, it appears that the chemical mechanism used to follow a plume's
development need not be as detailed as that used to simulate the rest of the domain, as the
AX2-117
-------
inorganic reactions are the most important in the plume (e.g., Kumar and Russell, 1996). The
need to include explicitly plume-in-grid chemistry disappears if one uses the adaptive grid
approach mentioned previously, though such grids are more computationally intensive. The
differences in simulations are significant because they can lead to significant differences in the
calculated sensitivity of O3 to its precursors (e.g., Sillman et al., 1995).
Because the chemical production and loss terms in the continuity equations for individual
species are coupled, the chemical calculations must be performed iteratively until calculated
concentrations converge to within some preset criterion. The number of iterations and the
convergence criteria chosen also can introduce error.
The importance of global transport of O3 and its contribution to regional O3 levels in the
United States is slowly becoming apparent. There are presently on the order of 20
three-dimensional global models that have been developed by various groups to address
problems in tropospheric chemistry. These models resolve synoptic meteorology, O3-NOX-CO-
hydrocarbon photochemistry, wet and dry deposition, and parameterize sub-grid scale vertical
mixing such as convection. Global models have proven useful for testing and advancing
scientific understanding beyond what is possible with observations alone. For example, they can
calculate quantities of interest that we do not have the resources to measure directly, such as
export of pollution from one continent to the global atmosphere or the response of the
atmosphere to future perturbations to anthropogenic emissions.
The finest horizontal resolution at which global simulations are typically conducted is
-200 km2 although rapid advances in computing power continuously change what calculations
are feasible. The next generation of models will consist of simulations that link multiple
horizontal resolutions from the global to the local scale. Finer resolution will only improve
scientific understanding to the extent that the governing processes are more accurately described
at that scale. Consequently there is a critical need for observations at the appropriate scales to
evaluate the scientific understanding represented by the models.
Observations of specific chemical species have been useful for testing transport schemes.
Radon-222 simulations in sixteen global models have been evaluated with observations to show
that vertical mixing is captured to within the constraints offered by the mean observed
concentrations (Jacob et al., 1997). Tracers such as cosmogenic 7Be and terrigenic 210Pb have
been used to test and constrain model transport and wet deposition (e.g., Liu et al., 2001b).
AX2-118
-------
Other chemical species obtained from various platforms (surface measurements, aircraft,
satellites) are useful for evaluating the simulation of chemical and dynamical processing in
global models. For example, Emmons et al. (2000) compiled available measurements of
12 species relevant to O3 photochemistry from a number of aircraft campaigns in different
regions of the world and used this data composite to evaluate two global models. They
concluded that one model (MOZART) suffered from weak convection and an underestimate of
nitrogen oxide emissions from biomass burning, while another model (IMAGES) transported too
much O3 from the stratosphere to the troposphere (Emmons et al., 2000). The global coverage
available from satellite observations offers new information for testing models. Recent efforts
are using satellite observations to evaluate the emission inventories of O3 precursors that are
included in global models; such observations should help to constrain the highly uncertain
natural emissions of isoprene and nitrogen oxides (e.g., Palmer et al., 2003; Martin et al., 2003).
A comparison of numerous global chemistry-transport models developed by groups around
the world was included in Section 4.4 of the recent report of the Intergovernmental Panel on
Climate Change (Prather and Ehhalt, 2001). In that report, monthly mean O3 (O3) and carbon
monoxide (CO) simulated by the various models was evaluated with O3 observations from global
ozonesonde stations at 700, 500, and 300 hPa and with surface CO measurements from
17 selected NOAA/CMDL sites. The relevant figures (Figures AX2-4-10 and AX2-4-11) are
reproduced here (as Figures AX2-19 for O3 and AX2-20 for CO) along with the references in
their Table AX2-10 (as Table AX2-4). Overall, the models capture the general features of the O3
and CO seasonal cycles but meet with varying levels of success at matching the observed
concentrations and the amplitude of the observed seasonal cycle. For O3, the models show less
disagreement in the lower troposphere than in the upper troposphere, reflecting the difficulty of
representing the exchange between the stratosphere and troposphere and the loose constraints on
the net O3 flux that are provided by observations.
An evaluation of five global models with data from the Measurement of Ozone and Water
Vapor by Airbus In-Service Aircraft (MOZAIC) project over New York City and Miami
indicates that the models tend to underestimate the summer maximum in the middle and lower
troposphere over northern mid-latitude cities such as New York City and to underestimate the
variability over coastal cities such as Miami which are strongly influenced by both polluted
continental and clean marine air masses (Law et al., 2000). Local maxima and minima are
AX2-119
-------
J FMAMJ JASOND
J FMAMJ JASOND
J FMAMJ JASOND
J FMAM J JASOND
J FMAM J JASOND
J FMAM J JASOND
J FMAMJ JASOND
J FMAMJ JASOND
J FMAMJ JASOND
/^iqq
_¥— UIOPO
WOIQ
.... IAQR
„„_ U'MIV/ll
MO71
iyyin79
I ir^M
I IID
I IKMO
I II AO
^ I IPAM
........ ^y|p|p
Figure AX2-19. Seasonal variability in O3 concentrations observed at a number of
pressure surfaces at six ozonesonde sites and the predictions of 13 global
scale chemistry-transport models.
Source: IPCC Third Assessment Report (2001).
AX2-120
-------
J FMAM J J ASON D
J FMAM J J ASON D
J FMAM J J ASON D
J FMAM J J ASON D
J FMAM J J ASON D
J FMAM J J ASON D
observ.
GISS
HGEO
IASB
KNMI
MOZ1
MOZ2
ucn
UIO
UKMO
ULAQ
-*- UCAM
MPIC
J FMAM J J ASON D
Figure AX2-20. Seasonal variability in O3 concentrations observed at a number of
pressure surfaces at six ozonesonde sites and the predictions of 13 global
scale chemistry-transport models.
Source: IPCC Third Assessment Report (2001).
difficult to reproduce with global models because processes are averaged over an entire model
grid cell. Much of the spatial and temporal variability in surface O3 over the United States is
modulated by synoptic meteorology (e.g., Logan, 1989; Eder et al., 1993; Vukovich, 1995, 1997;
Cooper and Moody, 2000) which is resolved in the current generation of global models.
For example, an empirical orthogonal function analysis on observed and simulated fields over
AX2-121
-------
Table AX2-4. Chemistry-Transport Models (CTM) Contributing to the Oxcomp
Evaluation of Predicting Tropospheric O3 and OH (Prather and Ehhalt, 2001)
CTM
GISS
HGEO
HGIS
IASB
KNMI
MOZ1
MOZ2
MPIC
UCI
UIO
UIO2
UKMO
ULAQ
UCAM
Institute
GISS
Harvard U.
Harvard U.
lAS/Belg.
KNMI/Utrecht
NCAR/CNRS
NCAR
MPI/Chem
UC Irvine
U. Oslo
U. Oslo
UK Met Office
U. L. Aquila
U. Cambridge
Contributing Authors
Shindell/Grenfell
Bey/Jacob
Mickley/Jacob
Mulller
van Weele
Hauglustaine/Brasseur
Horowitz/Brasseur
Kuhlmann/Lawrence
Wild
Berntsen
Sundet
Stevenson
Pitari
Plantevin/Johnson
(TOMCAT)
References
Hansenetal. (1997)
Beyetal. (200 la)
Mickleyetal. (1999)
Miiller and Brasseur
(1995, 1999)
Jeukenetal. (1999),
Houweling et al. (2000)
Brasseur etal. (1998),
Hauglustaine et al. (1998)
Brasseur etal. (1998),
Hauglustaine et al.(1998)
Crutzen etal. (1999),
Lawrence et al. (1999)
Hannegan etal. (1998),
Wild and Prather (2000)
Berntsen and Isaksen (1997),
Fuglestvedt etal. (1999)
Sundet (1997)
Collins etal. (1997),
Johnson etal. (1999)
Pitari etal. (1997)
Law etal. (1998, 2000)
the eastern United States in summer has shown that a 2° x 2.5° horizontal resolution global
model (GEOS-CHEM) captures the synoptic-scale processes that control much of the observed
variability (Fiore et al., 2003). Further evaluation of the same model showed that it can also
capture many of the salient features of the observed distributions of O3 as well as its precursors
in surface air over the United States in summer, including formaldehyde concentrations and
AX2-122
-------
correlations between O3 and the oxidation products of nitrogen oxides (O3: NOy-NOx), all of
which indicate a reasonable photochemical simulation (Fiore et al., 2002).
A significant amount of progress in evaluating the performance of three-dimensional
global models with surface, aircraft, and satellite data has been made in recent years.
Disagreement among model simulations mainly stems from differences in the driving
meteorology and emissions. The largest discrepancies amongst models and between models and
observations occur in the upper troposphere and likely reflect uncertainties in exchange between
the stratosphere and troposphere and photochemical processes there; the models agree better
with observations closer to the surface. Synoptic-scale meteorology is resolved in these models,
enabling them to simulate much of the observed variability in pollutants in the lower
troposphere.
AX2.5.2 Emissions of Ozone Precursors
Estimated annual emissions of nitrogen oxides, VOCs, CO, andNH3 for 1999 (U.S.
Environmental Protection Agency, 2001) are shown in Tables AX2-5, AX2-6, AX2-7, and
AX2-8. Methods for estimating emissions of criteria pollutants, quality assurance procedures
and examples of emissions calculated by using data are given in U.S. Environmental Protection
Agency (1999).
Emissions of nitrogen oxides associated with combustion arise from contributions from
both fuel nitrogen and atmospheric nitrogen. Sawyer et al. (2000) have reviewed the factors
associated with NOX emissions by mobile sources. Estimates of NOX emissions from mobile
sources are generally regarded as fairly reliable although further work is needed to clarify this
point (Sawyer et al., 2000). Both nitrifying and denitrifying bacteria in the soil can produce
NOX, mainly in the form of NO. Emission rates depend mainly on fertilization levels and soil
temperature. About 60% of the total NOX emitted by soils occurs in the central corn belt of the
United States. The oxidation of NH3 emitted mainly by livestock and soils, leads to the
formation of NO. Estimates of emissions from natural sources are less certain than those from
anthropogenic sources.
Natural sources of oxides of nitrogen include lightning, oceans, and soil. Of these, as
reviewed in AQCD 96, only soil emissions appear to have the potential to impact surface O3 over
the U.S. On a global scale, the contribution of soil emissions to the oxidized nitrogen budget is
AX2-123
-------
Table AX2-5. Emissions of Nitrogen Oxides by Various Sources in the
United States in 1999
Source
Emissions1
(1012 g/y)
Notes
On-road vehicle exhaust 7.8
Non-road vehicle exhaust 5
Fossil fuel combustion 9.1
Industrial Processes 0.76
Biomass burning 0.35
Waste disposal 0.053
Natural sources2 3.1
Total 26
Gasoline (58%) and diesel (42%) vehicles.
Diesel (49%) and gasoline (3%) vehicles; railroads (22%);
marine vessels (18%); other sources (8%).
Electric utilities (57%); industry (31%); commercial,
institutional and residential combustion (12%).
Mineral products (43%); petrochemical products (17%);
chemical mfg. (16%); metal processing (11%); misc.
industries (12%).
Residential wood burning (11%);
open burning (8%); wildfires (81%).
Non-biomass incineration.
Lightning (50%); soils(50%).
'Emissions are expressed in terms of NO2.
Estimated on the basis of data given in Guenther et al. (2000).
Source: U.S. Environmental Protection Agency (2001).
on the order of 10% (van Aardenne et al., 2001; Finlayson-Pitts and Pitts, 2000; Seinfeld and
Pandis, 1998), but attempts to quantify emissions of NOX from fertilized fields show great
variability. Soil NO emissions can be estimated from the fraction of the applied fertilizer
nitrogen emitted as NOX, but the flux varies strongly with land use and temperature. The fraction
nitrogen. Estimated globally averaged fractional applied nitrogen loss as NO varies from 0.3%
(Skiba et al., 1997) to 2.5% (Yienger and Levy, 1995). Variability within biomes to which
fertilizer is applied, such as shortgrass versus tallgrass prairie, accounts for a factor of three in
uncertainty (Williams et al., 1992; Yienger and Levy, 1995; Davidson and Kingerlee, 1997).
The local contribution can be much greater than the global average, particularly in summer
especially where corn is grown extensively. Williams et al. (1992) estimated that contributions
from soils in Illinois contribute about 26% of the emissions from industrial and commercial
AX2-124
-------
Table AX2-6. Emissions of Volatile Organic Compounds by Various Sources in the
United States in 1999
Source
On-road vehicles
Non-road vehicles
Fossil fuel combustion
Chemical industrial
processes
Petroleum industrial
processes
Other industrial
processes
Emissions
(1012 g/y)
4.8
2.9
0.27
0.36
0.39
0.48
Notes
Exhaust and evaporative losses from gasoline (95%)
diesel (5%) vehicles.
Exhaust and evaporative losses from gasoline (80%)
diesel (12%) vehicles; aircraft and other sources (8%
and
and
).
Electrical utilities; industrial, commercial, institutional,
and residential sources.
Mfg. of organic chemicals, polymers and resins, and
products.
Oil and gas production (64%); refining (36%).
Metal processing (15%); wood processing (32%);
agricultural product processing (21%); misc.
processes (18%).
misc.
Solvent volatilization
4.4 Surface coatings (44%); other industrial uses (20%); non-
industrial uses (e.g., pesticide application, consumer
solvents) (36%).
Storage and transport
of volatile compounds
Biomass burning
Waste disposal
Biogenic sources1
Total
1.1
1.2
0.53
4.4
21
Evaporative losses from petroleum products and other
organic compounds.
Residential wood combustion (37%); open burning
(22%); agricultural burning (22%); wildfires (19%).
Residential burning (63%); waste water (23%); landfills
(6%); non-biomass incineration (8%).
Approximately 98% emitted by vegetation. (Isoprene
[35%], monoterpenes [25%], and all other reactive and
non-reactive compounds [40%]).
'Estimated on the basis of data given in Guenther et al. (2000).
Source: U.S. Environmental Protection Agency (2001).
AX2-125
-------
Table AX2-7. Emissions of Ammonia by Various Sources in the United States in 1999
Source
Emissions
(1012 g/y)
Notes
Exhaust from on-road
and non-road engines and
vehicles
Fossil fuel combustion
Industry
0.25 Exhaust from on-road (96%) and non-road (4%) vehicles.
0.044 Combustion by electric utilities, industry, commerce,
institutions, residences.
0.18 Chemical manufacturing (67%); petroleum refining (9%);
other industries (25%).
Agriculture
Waste disposal and
recycling
Natural sources
Total
3.9
0.08
0.032
4.5
Livestock (82%); fertilizer application (18%).
Wastewater treatment (99%).
Unmanaged soils;
wild animals.
Source: U.S. Environmental Protection Agency (2001).
processes in that State. In Iowa, Kansas, Minnesota, Nebraska, and South Dakota soil emissions
may dominate. Conversion of ammonium to nitrate (nitrification) in aerobic soils appears to be
the dominant pathway to NO. The mass and chemical form of nitrogen (reduced or oxidized)
applied to soils, the vegetative cover, temperature, soil moisture, and agricultural practices such
as tillage all influence the amount of fertilizer nitrogen released as NO.
As pointed out in the previous AQCD for O3, emissions of NO from soils peak in summer
when O3 formation is at a maximum. A recent NRC report outlined the role of agricultural in
emissions of air pollutants including NO and NH3 (NRC, 2002). That report recommends
immediate implementation of best management practices to control these emissions, and further
research to quantify the magnitude of emissions and the impact of agriculture on air quality.
Civerolo and Dickerson (1998) report that use of the no-till cultivation technique on a fertilized
cornfield in Maryland reduced NO emissions by a factor of seven.
Annual global production of NO by lightning is the most uncertain source of reactive
nitrogen. In the last decade literature values of the production rate range from 2 to 20 Tg-N per
year. However, the most likely range is from 3 to 8 Tg-N per year, because the majority of the
AX2-126
-------
Table AX2-8. Emissions of Carbon Monoxide by Various Sources in the
United States in 1999
Source
Emissions
(1012 g/y)
Notes
On-road vehicle exhaust
Non-road engines and
vehicle exhaust
Fossil fuel combustion
Industrial Processes
Biomass burning
50
25
3.7
16
Gasoline-fueled light-duty cars (54%) and trucks (32%),
heavy-duty trucks (9%); diesel vehicles (5%);
motorcycles (0.4%).
Gasoline-fueled (lawn and garden [44%], light
commercial [17%], recreational [14%], logging [4%],
industry and construction [6%, other [1%]); diesel-fueled
(5%); aircraft (4%); other (5%).
Electric utilities (22%); industry (58%); commercial,
institutional and residential combustion (20%).
Metal processing (45%); chemical mfg. (29%);
petrochemical production; (10%); mineral products (5%);
wood products (10%); misc. industries (1%).
Residential wood burning (21%); open burning (21%);
agricultural burning (41%); wildfires (17%).
Waste disposal
Other
Biogenic emissions1
Total
0.42
0.19
4.7+
702+
Non-biomass incineration.
Structural fires (45%); storage and transport (38%);
misc. sources (17%).
Primary emissions from vegetation and soils;
secondary formation (?).
'Estimated on the basis of data given in Guenther et al. (2000).
Source: U.S. Environmental Protection Agency (2001).
recent estimates fall in this range. The large uncertainty stems from several factors: (1) a large
range of NO production rates per flash (as much as two orders of magnitude); (2) the open
question of whether cloud-to-ground (CG) flashes and intracloud flashes (1C) produce
substantially different amounts of NO; (3) the global flash rate; and (4) the ratio of the number of
1C flashes to the number of CG flashes. Estimates of the amount of NO produced per flash have
been made based on theoretical considerations (e.g., Price et al., 1997), laboratory experiments
(e.g., Wang et al., 1998a); field experiments (e.g., Stith et al., 1999; Huntrieser et al., 2002), and
AX2-127
-------
through a combination of cloud-resolving model simulations, observed lightning flash rates, and
anvil measurements of NO (e.g., DeCaria et al., 2000). The latter method was also used by
Pickering et al. (1998), who showed that only -5% to 20% of the total NO production by
lightning in a given storms exists in the boundary layer at the end of a thunderstorm. Therefore,
the direct contribution to boundary layer O3 production by lightning NO is thought to be small.
However, lightning NO production can contribute substantially to O3 production in the middle
and upper troposphere. DeCaria et al. (2000) estimated that up to 7 ppbv of O3 were produced in
the upper troposphere in the first 24 hours following a Colorado thunderstorm due to the
injection of lightning NO. A major uncertainty in mesoscale and global chemical transport
models is the parameterization of lightning flash rates. Model variables such as cloud top height,
convective precipitation rate, and upward cloud mass flux have been used to estimate flash rates.
Allen and Pickering (2002) have evaluated these methods against observed flash rates from
satellite, and examined the effects on O3 production using each method.
Literally tens of thousands of organic compounds have been identified in plant tissues.
However, most of these compounds either have sufficiently low volatility or are constrained so
that they are not emitted in significant quantities. Less than 40 compounds have been identified
by Guenther et al. (2000) as being emitted in large enough quantities to affect atmospheric
composition. These compounds include terpenoid compounds (isoprene, 2-methyl-3-buten-2-ol,
monoterpenes), compounds in the hexanal family, alkenes, aldehydes, organic acids, alcohols,
ketones and alkanes. As can be seen from Table AX2-6, the major species emitted by plants are
isoprene (35%), 19 other terpenoid compounds (25%) and 17 non-terpenoid compounds (40%)
(Guenther et al., 2000). Of the latter, methanol contributes 12% of total emissions.
Because isoprene has been identified as the most abundant of biogenic VOCs (Guenther
et al., 1995, 2000; Geron et al., 1994), it has been the focus of air quality model analyses in
many published studies (Roselle, 1994; Sillman et al., 1995). The original Biogenic Emission
Inventory System (BEIS) of Pierce and Waidruff (1991) used a branch-level isoprene emission
factor of 14.7 jig (g-foliar dry mass)"1 h"1 for high isoprene emitting species (e.g., oaks, or North
American Quercus species). When considering self-shading of foliage within branch enclosures,
this is roughly equivalent to a leaf level emission rate of 20 to 30 |ig-C (g-foliar dry mass)"1 h"1
(Guenther at al, 1995). Geron et al (1994) reviewed studies between 1990 and 1994 and found
that a much higher leaf-level rate of 70 |ig-C (g-foliar dry mass)"1 h"1 + 50% was more realistic,
AX2-128
-------
and this rate was used in BEIS2 for high isoprene emitting tree species. BEIS3 (Guenther et al.,
2000) applied similar emission factors at tree species levels (Geron et al 2000a, 2001) and more
recent canopy environment models to estimate isoprene fluxes.
The results from several studies of isoprene emission measurements made at leaf, branch,
tree, forest stand, and landscape levels have been used to test the accuracy of BEIS2 and BEIS3.
These comparisons are documented in Geron et al. (1997) and Guenther et al. (2000). The
results of these studies support the higher emission factors used in BEIS2 and BEIS3. Typically,
leaf emission factors (normalized to standard conditions of PAR = 1000 jimol nT2 and leaf
temperature of 30 °C) measured at the top of tree canopies equal or exceed those used in
BEIS2/3, while those in more shaded portions of the canopy tend to be lower than those assumed
in the models, likely due to differences in developmental environments of leaves within the
canopy (Monson et al., 1994; Sharkey et al., 1996; Harley et al., 1996; Geron et al., 2000b).
Uncertainty in isoprene emissions due to variability in forest composition and leaf area remain in
BVOC emission models and inventories. Seasonality and moisture stress also impact isoprene
emission, but algorithms to simulate these effects are currently fairly crude (Guenther et al,
2000). The bulk of biogenic emissions occur during the summer, because of their dependence
on temperature and incident sunlight. Biogenic emissions are also higher in southern states than
in northern states for these reasons. The uncertainty associated with natural emissions ranges
from about 50% for isoprene under midday summer conditions to about a factor often for other
compounds (Guenther et al., 2000). In assessing the relative importance of these compounds, it
should be borne in mind that the oxidation of many of the classes of compounds result in the
formation of secondary organic aerosol and that many of the intermediate products may be
sufficiently long lived to affect O3 formation in areas far removed from where they were emitted.
The oxidation of isoprene can also contribute about 10% of the source of CO (U.S.
Environmental Protection Agency, 2000). Direct emissions of CO by vegetation is of much
smaller importance. Soil microbes both emit and take up atmospheric CO, however, soil
microbial activity appears to represent a net sink for CO.
Emissions from biomass burning depend strongly on the stage of combustion. Smoldering
combustion, especially involving forest ecosystems favors the production of CH4, NMHC and
CO at the expense of CO2, whereas active combustion produces more CO2 relative to the other
compounds mentioned above. Typical emissions ratios (defined as moles of compound per
AX2-129
-------
moles of emitted CO2 expressed as a percentage) range from 6 to 14% for CO, 0.6 to 1.6% for
CH4, and 0.3 to 1.1% for NMHCs (Andreae, 1991). Most NMHC emissions are due to
emissions of lighter compounds, containing 2 or 3 carbon atoms.
AX2.5.3 Observationally-Based Models
As an alternative to chemistry-transport models, observationally-based methods (OEMs),
which seek to infer O3-precursor relations by relying more heavily on ambient measurements,
can be used. Observationally-based methods are intuitively attractive because they provide an
estimate of the O3-precursor relationship based directly on observations of the precursors. These
methods rely on observations as much as possible to avoid many of the uncertainties associated
with chemistry/transport models (e.g., emission inventories and meteorological processes).
However, these methods have large uncertainties with regards to photochemistry. As originally
conceived, the observation-based approaches were intended to provide an alternative method for
evaluating critical issues associated with urban O3 formation. The proposed OEMs include
calculations driven by ambient measurements (Chameides et al., 1992; Cardelino et al., 1995)
and proposed "rules of thumb" that seek to show whether O3 is primarily sensitive to NOX or to
VOC concentrations (Sillman, 1995; Chang et al., 1997; Tonnesen and Dennis, 2000a,b;
Blanchard et al., 1999; Blanchard, 2000). These methods are controversial when used as
"stand-alone" rules, because significant uncertainties and possible errors have been identified for
all the methods (Chameides et al., 1988, Lu and Chang, 1998, Sillman and He, 2002; Blanchard
and Stoeckenius, 2001). Methods such as these are most promising for use in combination with
chemistry /transport models principally for evaluating the accuracy of model predictions.
Recent results (Tonnesen and Dennis, 2000a; Kleinman et al., 1997; 2000, 2001;
Kleinman, 2000) suggest that ambient VOC and NOX data can be used to identify the
instantaneous production rate for O3 and how the production rate varies with concentrations of
NOX and VOCs. The instantaneous production rate for O3 is only one of the factors that affect
the total O3 concentration, because O3 concentrations result from photochemistry and transport
over time periods ranging from several hours to several days in regional pollution events. Ozone
concentrations can be affected by distant emissions and by photochemical conditions at upwind
locations, rather than instantaneous production at the site. Despite this limitation, significant
information can be obtained by interpreting ambient NOX and VOC measurements. Kleinman
AX2-130
-------
et al. (1997, 2000, 2001) and Tonnesen and Dennis (2000a) both derived simple expressions that
relate the NOX-VOC sensitivity of instantaneous O3 production to ambient VOC and NOX. These
expressions usually involve summed VOC weighted by reactivity.
Cardelino et al. (1995, 2000) developed a method that seeks to identify O3-NOX-VOC
sensitivity based on ambient NOX and VOC data. Their method involves an area-wide sum of
instantaneous production rates over an ensemble of measurement sites, which serve to represent
the photochemical conditions associated with O3 production in metropolitan areas. Their
method, which relies on routine monitoring methods, is especially useful because it permits
evaluation for a full season rather than just for individual episodes.
AX2.5.4 Chemistry-Transport Model Evaluation
The comparison of model predictions with ambient measurements represents a critical task
for establishing the accuracy of photochemical models and evaluating their ability to serve as the
basis for making effective control strategy decisions. The evaluation of a model's performance,
or its adequacy to perform the tasks for which it was designed can only be conducted within the
context of measurement errors and artifacts. Not only are there analytical problems, but there
are also problems in assessing the representativeness of monitors at ground level for comparison
with model values which represent typically an average over the volume of a grid box.
Chemistry-transport models for O3 formation at the urban/regional scale have traditionally
been evaluated based on their ability to correctly simulate O3. A series of performance statistics
that measure the success of individual model simulations to represent the observed distribution
of ambient O3, as represented by a network of surface measurements were recommended in U.S.
Environmental Protection Agency (1991; see also Russell and Dennis, 2000). These statistics
consist of the following:
• Unpaired peak O3 within a metropolitan region (typically for a single day).
• Normalized bias equal to the summed difference between model and measured hourly
concentrations divided by the sum of measured hourly concentrations.
• Normalized gross error, equal to the summed unsigned (absolute value) difference
between model and measured hourly concentrations divided by the sum of measured
hourly concentrations.
AX2-131
-------
Unpaired peak prediction accuracy, Au;
^p(X,l)max t-o\X •>'• /max
AU= 77777: *100%, (AX2-49)
^*-O\ ' /max
Normalized bias, D;
,
Z . ' ' = 1, 24. (AX2-50)
Gross error, Ed (for hourly observed values of O3 >60 ppb)
/=1,24. (AX2-51)
The following performance criteria for regulatory models were recommended in U.S.
Environmental Protection Agency (1991): unpaired peak O3 to within ±15% or ±20%;
normalized bias within ±5% to ±15%; and normalized gross error less than 30% to 35%, but
only when O3 >60 ppb. This can lead to difficulties in evaluating model performance since
nighttime and diurnal cycles are ignored. A major problem with this method of model
evaluation is that it does not provide any information about the accuracy of O3-precursor
relations predicted by the model. The process of O3 formation is sufficiently complex that
models can predict O3 correctly without necessarily representing the O3 formation process
properly. If the O3 formation process is incorrect, then the modeled source-receptor relations
will also be incorrect.
Studies by Sillman et al. (1995, 2003), Reynolds et al. (1996) and Pierce et al. (1998) have
identified instances in which different model scenarios can be created with very different
O3-precursor sensitivity, but without significant differences in the predicted O3 fields.
Figures AX2-21a,b provide an example. Referring to the O3-NOX-VOC isopleth plot
(Figure AX2-22), it can be seen that similar O3 concentrations can be found for photochemical
conditions that have very different sensitivity to NOX and VOCs.
AX2-132
-------
Global-scale chemistry-transport models have generally been evaluated by comparison
with measurements for a wide array of species, rather than just for O3 (e.g., Wang et al., 1998b;
Emmons et al., 2000; Bey et al., 2001b; Hess, 2001; Fiore et al., 2002). These have included
evaluation of major primary species (NOX, CO, and selected VOCs) and an array of secondary
species (HNO3, PAN, H2O2) that are often formed concurrently with O3. Models for
urban/regional O3 have also been evaluated against a broader ensemble of measurements in a
few cases, often associated with measurement intensives (e.g., Jacobson et al., 1996, Lu et al.,
1997; Sillman et al., 1998). The results of a comparison between observed and computed
concentrations from Jacobson et al. (1996) for the Los Angeles Basin are shown in
Figures AX2-23a,b.
The highest concentrations of primary species usually occur in close proximity to emission
sources (typically in urban centers) and at times when dispersion rates are low. The diurnal
cycle includes high concentrations at night, with maxima during the morning rush hour, and low
concentrations during the afternoon (Figure AX2-23a). The afternoon minima are driven by the
much greater rate of vertical mixing at that time. Primary species also show a seasonal
maximum during winter, and are often high during fog episodes in winter when vertical mixing ,
is suppressed. By contrast, secondary species such as O3 are typically highest during the
afternoon (the time of greatest photochemical activity), on sunny days and during summer.
During these conditions concentrations of primary species may be relatively low. Strong
correlations between primary and secondary species are generally observed only in downwind
rural areas where all anthropogenic species are high simultaneously. The difference in the
diurnal cycles of primary species (CO, NOX and ethane)and secondary species (O3, PAN and
HCHO) is evident in Figure AX2-23b.
Models for urban/regional O3 have been evaluated less extensively than global-scale
models in part because the urban/regional context presents a number of difficult challenges.
Global-scale models typically represent continental-scale events and can be evaluated effectively
against a sparse network of measurements. By contrast, urban/regional models are critically
dependent on the accuracy of local emission inventories and event-specific meteorology, and
must be evaluated separately for each urban area that is represented.
AX2-133
-------
.Q
Q.
Q.
O
3
•o
o:
CO
O
25
20
~ 10
-5
-t
>*
;
*
£
xx>
100 120 140 160
Ozone (ppb)
180
200
35
-£ 30
.Q
Q.
^
C
CO
25
20
| ^
| 10
0
x x
X
XX
X
X
100 120 140 160
Ozone (ppb)
180
200
Figure AX2-21a,b.
Impact of model uncertainty on control strategy predictions for O3
for two days (August 10 [a] and ll[b], 1992) in Atlanta, GA. The
figures show the predicted reduction in peak O3 resulting from 35%
reductions in anthropogenic VOC emissions (crosses) and from 35%
reductions in NOX (solid circles) in a series of model scenarios with
varying base case emissions, wind fields, and mixed layer heights.
Source: Results are plotted from tabulated values published in Sillman et al. (1995, 1997).
AX2-134
-------
~ 10.00
E
o
3.16
o
es
O
C, 1.0
0)
•+•»
(0
a:
c
o
0.316
LU
x
O
0.1
a.
a.
CO
o
1.0 3.16 10.0 31.6 100.0
VOC Emission Rate (1012 molec. cm-2 s-1)
Figure AX2-22. Ozone isopleths (ppb) as a function of the average emission rate for
NOX and VOC (1012 molec. cm~2 s'1) in zero dimensional box model
calculations. The isopleths (solid lines) represent conditions during
the afternoon following 3-day calculations with a constant emission
rate, at the hour corresponding to maximum O3. The ridge line
(shown by solid circles) lies in the transition from NOx-saturated to
NOv-limited conditions.
The evaluation of urban/regional models is also limited by the availability of data.
Measured NOX and speciated VOC concentrations are widely available through the EPA PAMs
network, but questions have been raised about the accuracy of those measurements and the data
have not yet been analyzed thoroughly. Evaluation of urban/regional models versus
measurements has generally relied on results from a limited number of field studies in the United
States. Short term research-grade measurements for species relevant to O3 formation, including
VOCs, NOX, PAN, nitric acid (HNO3) and hydrogen peroxide (H2O2) are also widely available at
AX2-135
-------
Q.
a.
o
44
(0
O)
c
Q.
Q.
(Q
o:
D)
C
'x
Q.
Q.
(0
cr
_c
'R
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.30
0.25
0.20
0.15
0.10
0.05
0.00
6
5
4
3
2
1
0
Reseda
O3(g)
Predicted
Observed
8 16 24 32 40 48 56 64 72
Hour After First Midnight
Reseda
N0x(g)
Predicted
Observed
8 16 24 32 40 48 56 64 72
Hour After First Midnight
Riverside
C0(g)
Predicted
Observed
\ /
8 16 24 32 40 48 56 64 72
Hour After First Midnight
Figure AX2-23a. Time series for measured gas-phase species in comparison with
results from a photochemical model. The dashed lines represent
measurements, and solid lines represent model predictions (in parts per
million, ppmv) for August 26-28,1988 at sites in southern California.
The horizontal axis represents hours past midnight, August 25. Results
represent O3 and NOX at Reseda and CO at Riverside.
Source: Jacobson et al. (1996).
AX2-136
-------
0.060
| 0.050
Q.
3 0.040
o
'-§ 0.030
OL
0) 0.020
_c
~ 0.010
0.000
0.030
| 0.025
Q.
3 0.020
O
~ 0.015
a:
0) 0.010
~ 0.005
0.000
0.020
£ 0.016
Q.
a.
"g" 0.012
«rf
(0
o:
0.008
O)
_c
!B 0.004
§
0.000
Claremont
Ethane (g)
Predicted
O Observed
_L
16 24 32 40 48 56
Hour After First Midnight
64 72
Claremont
Formaldehyde (g)
Predicted
O Observed
I
16 24 32 40 48 56
Hour After First Midnight
64 72
Los Angeles
PAN (g)
Predicted
Observed
_L
16 24 32 40 48 56
Hour After First Midnight
64 72
Figure AX2-23b.
Time series for measured gas-phase species in comparison with results
from a photochemical model. The circles represent measurements, and
solid lines represent model predictions (in parts per million, ppmv) for
August 26 - 28,1988 at sites in southern California. The horizontal axis
represents hours past midnight, August 25. Results represent ethane
and formaldehyde at Claremont, and PAN at Los Angeles.
Source: Jacobsonetal. (1996).
AX2-137
-------
rural and remote sites (e.g., Daum et al., 1990, 1996; Martin et al., 1997; Young et al., 1997;
Thompson et al., 2000; Hoell et al., 1996, 1997; Fehsenfeld et al., 1996a; Emmons et al., 2000;
Hess, 2001; Carroll et al., 2001). The equivalent measurements are available for some polluted
rural sites in the eastern United States but only at a few urban locations (Meagher et al., 1998;
Hubler et al., 1998; Kleinman et al., 2000, 2001; Fast et al., 2002; new SCAQS-need reference).
Extensive measurements have also been made in Vancouver (Steyn et al., 1997) and in several
European cities (Staffelbach et al., 1997; Prevot et al., 1997, Dommen et al., 1999; Geyer et al.,
2001; Thielman et al., 2001; Martilli et al., 2002; Vautard et al., 2002).
The results of straightforward comparisons between observed and predicted concentrations
of O3 can be misleading because of compensating errors, although this possibility is diminished
when a number of species are compared. Ideally, each of the main modules of a chemistry-
transport model system (for example, the meteorological model and the chemistry and radiative
transfer routines) should be evaluated separately. However, this is rarely done in practice.
To better indicate how well physical and chemical processes are being represented in the model,
comparisons of relations between concentrations measured in the field and concentrations
predicted by the model can be made. These comparisons could involve ratios and correlations
between species. For example, correlation coefficients could be calculated between primary
species as a means of evaluating the accuracy of emission inventories; or between secondary
species as a means of evaluating the treatment of photochemistry in the model. In addition,
spatial relations involving individual species (correlations, gradients) can also be used as a
means of evaluating the accuracy of transport parameterizations. Sillman and He (2002)
examined differences in correlation patterns between O3 and NOZ in Los Angeles, CA, Nashville,
TN and various sites in the rural United States. Model calculations (Figure AX2-24) show
differences in correlation patterns associated with differences in the sensitivity of O3 to NOX
and VOCs. Primarily NOx-sensitive ( NOx-limited) areas in models show a strong correlation
between O3 and NOZ with a relatively steep slope, while primarily VOC-sensitive (NOX-
saturated) areas in models show lower O3 for a given NOZ and a lower O3-NOZ slope. They
found that differences found in measured data ensembles were matched by predictions from
chemical transport models. Measurements in rural areas in the eastern U.S. show differences in
the pattern of correlations for O3 versus NOZ between summer and autumn (Jacob et al., 1995;
AX2-138
-------
250
200
150
n
o
100
0
20
NOZ (ppb)
Figure AX2-24. Correlations for O3 versus NOZ (NOy-NOx) in ppb from chemical
transport models for the northeast corridor, Lake Michigan, Nashville,
the San Joaquin Valley, and Los Angeles. Each location is classified as
NOx-limited or NOx-sensitive (circles), NOx-saturated or VOC-sensitive
(crosses), mixed or with near-zero sensitivity (squares), and dominated by
NOX titration (asterisks) based on the model response to reduced NOX and
VOC.
Source: Sillman and He (2002).
Hirsch et al., 1996), corresponding to the transition from NOx-limited to NOx-saturated patterns,
a feature which is also matched by chemistry-transport models.
The difference in correlations between secondary species in NOx-limited to NOx-saturated
environments can also be used to evaluate the accuracy of model predictions in individual
applications. Figures AX2-25a and AX2-25b show results for two different model scenarios for
Atlanta. As shown in the figures, the first model scenario predicts an urban plume with high
NOy and O3 formation apparently suppressed by high NOy. Measurements show much lower
AX2-139
-------
10
20
NOy (ppb)
30
40
.Q
Q.
a.
CO
O
0
Figure AX2-25a,b. Evaluation of model versus measured O3 versus NOy for two
model scenarios for Atlanta. The model values are classified as NOX
limited (circles), NOx-saturated (crosses), or mixed or with low
sensitivity to NOX (squares). Diamonds represent aircraft
measurements.
Source: Sillmanetal. (1997).
AX2-140
-------
NOy in the Atlanta plume. This error was especially significant because the model locations
with high NOy were not sensitive to NOX, while locations with lower NOy were primarily
sensitive to NOX. The second model scenario (with primarily NOx-sensitive conditions) shows
much better agreement with measured values. Figure AX2-26a,b shows model-measurement
comparisons for secondary species in Nashville, showing better agreement with measured
conditions. Greater confidence in the predictions made by chemistry-transport models will be
gained by the application of techniques such as these on a more routine basis.
The ability of chemical mechanisms to calculate the concentrations of free radicals under
atmospheric conditions was tested in the Berlin Ozone Experiment, BERLIOZ (Volz-Thomas
et al., 2003) during July and early August at a site located about 50 km NW of Berlin. (This
location was chosen as O3 episodes in central Europe are often associated with SE winds.)
Concentrations of major compounds such as O3, hydrocarbons, etc., were fixed at observed
values. In this regard, the protocol used in this evaluation is an example of an observationally
based method. Figure AX2-27 compares the concentrations of RO2 (organic peroxy), HO2
(hydroperoxy) and OH (hydroxyl) radicals predicted by RACM (regional air chemistry
mechanism; Stockwell et al., 1997) and MCM (master chemical mechanism; Jenkin et al, 1997
with updates) with observations made by the laser induced fluorescence (LIF) technique and by
matrix isolation ESR spectroscopy (MIESR). Also shown are the production rates of O3
calculated using radical concentrations predicted by the mechanisms and those obtained by
measurements, and measurements of NOX concentrations. As can be seen, there is good
agreement between measurements of organic peroxy, hydroperoxy and hydroxyl radicals with
values predicted by both mechanisms at high concentrations of NOX (>10 ppb). However, at
lower NOX concentrations, both mechanisms substantially overestimate OH concentrations and
moderately overestimate HO2 concentrations. Agreement between models and measurements is
generally better for organic peroxy radicals, although the MCM appears to overestimate their
concentrations somewhat. In general, the mechanisms reproduced the HO2 to OH and RO2 to
OH ratios better than the individual measurements. The production of O3 was found to increase
linearly with NO (for NO <0.3 ppb) and to decrease with NO (for NO >0.5 ppb).
OH and HO2 concentrations measured during the PM2 5 Technology Assessment and
Characterization Study conducted at Queens College in New York City in the summer of 2001
were also compared with those predicted by RACM (Ren et al., 2003). The ratio of observed to
AX2-141
-------
160
140
0
160
140
10
20
NOZ (ppb)
30
40
0
10 20 30
2H2O2 + NOZ (ppb)
40
Figure AX2-26a,b.
Evaluation of model versus: (a) measured O3 versus NOZ and (b) O3
versus the sum 2H2O2 + NOZ for Nashville, TN. The model values are
classified as NOx-limited (gray circles), NOx-saturated (*s), mixed or
with near-zero sensitivity (squares), or dominated by NOX titration
(filled circles). Diamonds represent aircraft measurements.
Source: Sillmanetal. (1998).
AX2-142
-------
CO
I
o
CO 4
O
£. 2
CM
O
x 0
o
CD
O
10
^ 5
0
10
Q.
Q.
X 5
CO
o
Q.
a.
X
O
0
20
10
0
O LIF
• MIESR
J(d1D)*106(s-1)
8
10 12 14
UT 20.7.98
16
Figure AX2-27. Time series of concentrations of RO2, HO2, and OH radicals, local
O3 photochemical production rate and concentrations of NOX from
measurements made during BERLIOZ. Also shown are comparisons
with results of photochemical box model calculations using the RACM
and MCM chemical mechanisms.
Source: Volz-Thomas et al. (2003).
AX2-143
-------
predicted HO2 concentrations over a diurnal cycle was 1.24 and the ratio of observed to
predicted OH concentrations was about 1.10 during the day, but the mechanism significantly
underestimated OH concentrations during the night.
AX2.5.4.1 Evaluation of Emissions Inventories
Comparisons of emissions model predictions with observations have been performed in a
number of environments. A number of studies of ratios of concentrations of CO to NOX and
NMOC to NOX during the early 1990s in tunnels and ambient air (summarized in Air Quality
Criteria for Carbon Monoxide [U.S. Environmental Protection Agency, 2000]) indicated that
emissions of CO and NMOC were systematically underestimated in emissions inventories.
However, the results of more recent studies have been mixed in this regard, with many studies
showing agreement to within ± 50% (U.S. Environmental Protection Agency, 2000).
Improvements in many areas have resulted from the process of emissions model development,
evaluation, and further refinement. It should be remembered that the conclusions from these
reconciliation studies depend on the assumption that NOX emissions are predicted correctly by
emissions factor models. Road side remote sensing data indicate that over 50% of NMHC and
CO emissions are produced by less than about 10% of the vehicles (Stedman et al., 1991). These
"super-emitters" are typically poorly maintained vehicles. Vehicles of any age engaged in off-
cycle operations (e.g., rapid accelerations) emit much more than if operated in normal driving
modes. Bishop and Stedman (1996) found that the most important variables governing CO
emissions are fleet age and owner maintenance.
Emissions inventories for North America can be evaluated with comparisons to measured
long-term trends and or ratios of pollutants in ambient air. A decadal field study of ambient CO
at a rural cite in the Eastern U.S. (Hallock-Waters et al., 1999) indicates a downward trend
consistent with the downward trend in estimated emissions over the period 1988 to 1999 (U.S.
Environmental Protection Agency, 1997), even when a global downward trend is accounted for.
Measurements at two urban areas in the United States confirmed the decrease in CO emissions
(Parrish et al., 2002). That study also indicated that the ratio of CO to NOX emissions decreased
by almost a factor of three over 12 yr (such a downward trend was noted in AQCD 96).
Emissions estimates (U.S. Environmental Protection Agency, 1997) indicate a much smaller
decrease in this ratio, suggesting that NOX emissions from mobile sources may be
AX2-144
-------
underestimated and/or increasing. The authors conclude that O3 photochemistry in U.S. urban
areas may have become more NOx-limited over the past decade.
Pokharel et al. (2002) employed remotely-sensed emissions from on-road vehicles and fuel
use data to estimate emissions in Denver. Their calculations indicate a continual decrease in CO,
HC, and NO emissions from mobile sources over the 6 yr study period. Inventories based on the
ambient data were 30 to 70% lower for CO, 40% higher for HC, and 40 to 80% lower for NO
than those predicted by the recent MOBILE6 model.
Stehr et al. (2000) reported simultaneous measurements of CO, SO2 and NOy at an East
Coast site. By taking advantage of the nature of mobile sources (they emit NOX and CO but little
SO2) and power plants (they emitNOx and SO2 but little CO), the authors evaluated emissions
estimates for the eastern United States. Results indicated that coal combustion contributes 25 to
35% of the total NOX emissions in agreement with emissions inventories (U.S. Environmental
Protection Agency, 1997).
Parrish et al. (1998) and Parrish and Fehsenfeld (2000) proposed methods to derive
emission rates by examining measured ambient ratios among individual VOC, NOX and NOy.
There is typically a strong correlation among measured values for these species (e.g., Figure
AX2-14) because emission sources are geographically collocated, even when individual sources
are different. Correlations can be used to derive emissions ratios between species, including
adjustments for the impact of photochemical aging. Investigations of this type include
correlations between CO and NOy (e.g., Parrish et al., 1991), between individual VOC species
and NOy (Goldan et al., 1995, 1997, 2000) and between various individual VOC (Goldan et al.,
1995, 1997; McKeen and Liu, 1993; McKeen et al., 1996). Buhr et al. (1992) derived emission
estimates from principal component analysis (PCA) and other statistical methods. Many of these
studies are summarized in Trainer et al. (2000), Parrish et al. (1998), and Parrish and Fehsenfeld
(2000). Goldstein and Schade (2000) also used species correlations to identify the relative
impacts of anthropogenic and biogenic emissions. Chang et al. (1996, 1997) and
Mendoza-Dominguez and Russell (2000, 2001) used the more formal techniques of inverse
modeling to derive emission rates, in conjunction with results from chemistry-transport models.
Another concern regarding the use of emissions inventories is that emissions from all significant
sources have been included. This may not always be the case. As an example, hydrocarbon
AX2-145
-------
seeps from offshore oil fields may represent a significant source of reactive organic compounds
in near by coastal areas (Quigley et al., 1999).
AX2.5.4.2 Availability and Accuracy of Ambient Measurements
The use of methods such as observationally based methods or source apportionment
models, either as stand-alone methods or as a basis for evaluating chemistry /transport models,
is often limited by the availability and accuracy of measurements. Measured speciated VOC and
NOX are widely available in the United States through the PAMS network. However, challenges
have been raised about both the accuracy of the measurements and their applicability.
Parrish et al. (1998) and Parrish and Fehsenfeld (2000) developed a series of quality
assurance tests for speciated VOC measurements. Essentially these tests used ratios among
individual VOC with common emission sources to identify whether the variations in species
ratios were consistent with the relative photochemical lifetimes of individual species. These
tests were based on a number of assumptions: the ratio between ambient concentrations of
long-lived species should show relatively little variation among measurements affected by a
common emissions sources; and the ratio between ambient concentrations of long-lived and
short-lived species should vary in a way that reflects photochemical aging at sites more different
from source regions. Parrish et al. used these expectations to establish criteria for rejecting
apparent errors in measurements. They found that the ratios among alkenes at many PAMS sites
did not show variations that would be expected due to photochemical aging.
The PAMs network currently includes measured NO and NOX. However, Cardelino and
Chameides (2000) reported that measured NO during the afternoon was frequently at or below
the detection limit of the instruments (1 ppb), even in large metropolitan regions (Washington,
DC; Houston, TX; New York, NY). NOX measurements are made with commercial
chemilluminescent detectors with molybdenum converters. However these measurements
typically include some organic nitrates in addition to NOX, and cannot be interpreted as a "pure"
NOX measurement (see summary in Parrish and Fehsenfeld, 2000).
Total reactive nitrogen (NOy) is included in the PAMS network only at a few sites. The
possible expansion of PAMS to include more widespread NOy measurements has been suggested
(McClenny, 2000). A major issue concerning measured NOy is the possibility that HNO3,
a major component of NOy, is sometimes lost in inlet tubes and not measured (Luke et al., 1998;
AX2-146
-------
Parrish and Fehsenfeld, 2000). This problem is especially critical if measured NOy is used to
identify NOx-limited versus NOx-saturated conditions. The correlation between O3 and NOy
differs for NOx-limited versus NOx-saturated locations, but this difference is driven primarily by
differences in the ratio of O3 to HNO3. If HNO3 were omitted from the NOy measurements, than
the measurements would represent a biased estimate and their use would be problematic.
AX2.6 TECHNIQUES FOR MEASURING OZONE AND ITS
PRECURSORS
AX2.6.1 Sampling and Analysis of Ozone
Numerous techniques have been developed for sampling and measurement of O3 in the
ambient atmosphere at ground level. As noted above, sparse surface networks tend to
underestimate maximum O3 concentrations. Today, monitoring is conducted almost exclusively
with UV absorption spectrometry with commercial short path (0.15 to 0.3 m) instruments, a
method that has been thoroughly evaluated in clean air. The ultimate reference method is a
relatively long-path (3 m path length) UV absorption instrument maintained under carefully
controlled conditions atNIST (e.g., Fried and Hodgeson, 1982). Episodic measurements are
made with a variety of other techniques based on the principles of chemiluminescence,
electrochemistry, DO AS, and LIDAR. The rationale, history, and calibration of O3
measurements were summarized in AQCD 96, so this section will focus on the current state of
ambient O3 measurement, tests for artifacts, and on new developments.
Several reports in the reviewed scientific literature have investigated interferences in O3
detection via UV radiation absorption. Kleindienst et al. (1993) investigated the effects of water
vapor and VOCs on instruments based on both UV absorption and chemiluminescence. They
concluded that water vapor had no significant impact on UV absorption-based instruments, but
could cause a positive interference of up to 9% in chemiluminescence-based detectors at high
humidities (dew point of 24 C). In smog chamber studies, aromatic compounds and their
oxidation products were found to generate a positive but small interference in the UV absorption
instruments. Kleindienst et al. concluded that "when the results are scaled back to ambient
concentrations of toluene and NOX, the effect appears to be very minor (ca. 3 percent under the
study conditions)." Arshinov et al. (2002) reported a positive interference in UV absorption
AX2-147
-------
instruments from ambient particles at high concentrations, but this interference is eliminated by
use of particle filters less subject to absorption/adsorption artifacts such as used in the
NAMS/SLAMS network along with replacement of these filters to avoid artifacts associated
with particle loading on the filters. The possibility for substantive interferences in O3 detection
exists, but such interferences have not been observed even in urban plumes. Ryerson et al.
(1998) measured O3 with UV absorption and chemiluminescence instruments operated off a
common inlet on the NOAA WP-3 research aircraft. As reported by Parrish and Fehsenfeld
(2000) "Through five field missions over four years, excellent correlations were found between
the measurements of the two instruments, although the chemiluminescence instrument was
systematically low (5%) throughout some flights. The data sets include many passes through the
Nashville urban plume. There was never any indication (<1%) that the UV instrument measured
systematically higher in the urban plume." The same group tested the air of Houston, El Paso,
Nashville, Los Angeles, San Francisco, and the East Coast. They observed only one instance of
substantive positive interference defined as the UV absorption technique showing more than a
few ppb more than the CL. This occurred in Laporte, TX under heavily polluted conditions and
a low inversion, at night (Jobson et al., 2004).
Leston et al. (2005) reported positive and negative interferences in UV absorption
techniques for measuring O3 (relative to the CL technique) in Mexico City and in a smog
chamber study. They suggested that O3 measured in ambient air could be too high by 20 to
40 ppb under specific conditions due to positive interference by a number of organic compounds,
mainly those produced during the oxidation of aromatic hydrocarbons and some primary
compounds such as styrene and naphthalene. However, the concentrations of these compounds
were many times higher in both of these environments than are typically found at ambient air
monitoring sites in the United States. Although Hg is also potentially a strong interfering agent,
because the Hg resonance line is used in this technique, its concentration would also have to be
many times higher than is typically found in ambient air, e.g., as might be found in power plant
plumes. Thus, it seems unlikely that such interferences would amount to more than one or two
ppb (within the design specifications of the FEM), except under conditions conducive to
producing high concentrations of the substances they identified as causing interference. These
conditions might be found next to a plant producing styrene, for example. Leston et al. (2005)
AX2-148
-------
also noted that the use of alternative materials in the scrubber could alleviate many potential
problems under these conditions.
Ozone can also be detected by differential optical absorption spectroscopy (DOAS) at a
variety of wavelengths in the UV and visible parts of the spectrum. Prior comparisons of DOAS
results to those from a UV absorption instrument showed good agreement, on the order of 10%
(Stevens et al., 1993). Reisinger (2000) reported a positive interference due to an unidentified
absorber in the 279 to 289 nm spectral region used by many commercial short-path DOAS
systems for the measurement of O3. Results of that study suggest that effluent from wood
burning, used for domestic heating, may be responsible. Vandaele et al. (2002) reported good
agreement with other methods in the detection of O3 (and SO2) over the course of several years
in Brussels. While the DOAS method remains attractive due to its sensitivity and speed of
response further intercomparisons and interference tests are recommended.
Electrochemical methods are commonly employed where sensor weight is a problem, such
as in balloon borne sondes, and these techniques have been investigated for ambient monitoring.
Recent developments include changes in the electrodes and electrolyte solution (Knake and
Hauser, 2002) and selective removal of O3 for a chemical zero (Penrose et al., 1995).
Interferences from other oxidants such as NO2 and HONO remain potential problems and further
comparisons with UV absorption are necessary. Because of potential interferences from water
vapor (ASTM, 2003 a,b), all instruments should be either calibrated and zeroed with air humidity
near ambient or demonstrated to be insensitive to humidity.
Change in the vibration frequency of a piezoelectric quartz crystal has been investigated as
a means of detecting O3. Ozone reacts with polybutadiene coated onto the surface of a crystal,
and the resulting change in mass is detected as a frequency change (Black et al., 2000). While
this sensor has advantages of reduced cost power consumption and weight, it is lacks the lifetime
and absolute accuracy for ambient monitoring.
AX2.6.2 Sampling and Analysis of Nitrogen Oxides
Reactive nitrogen is generally released as NO but quickly converted in ambient air to NO2
and back again, thus these two species are often referred to together as NOX (NO + NO2). The
photochemical interconversion of NO and NO2 leads to O3 formation. Because NO2 is a health
hazard at sufficiently high concentrations, it is itself a criteria pollutant. In EPA documents,
AX2-149
-------
emissions of NOX are expressed in units of mass of NO2 per unit time, i.e., the total mass of NOX
that would be emitted if all the NO were converted to NO2. Ambient air monitors have been
required to demonstrate compliance with the standard for NO2 and thus have focused on
measuring this gas or determining an upper limit for its concentration.
NOX can be oxidized to species such as nitrous acid (HNO2), nitric acid (HNO3), aerosol
nitrate (NO3 ), and organo-nitrates such as alkyl nitrates (RONO2) and peroxy acetyl nitrate,
PAN, (CH3C(O)O2NO2). The sum of these species (explicitly excluding N2, N2O, and reduced N
such as NH3 and HCN) is called NOy. Nitrates play important roles in acid rain, and nutrient
cycling including over nitrification of surface ecosystems and in the formation of fine particulate
matter, but are generally inactive photochemically. Some studies refer specifically to the
oxidized or processed NOy species, NOy-NOx, as NOZ because this quantity is related to the
degree of photochemical aging in the atmosphere. Several NOZ species such as PAN and HONO
can be readily photolyzed or thermally dissociated to NO or NO2 and thus act as reservoirs for
NOX. This discussion focuses on current methods and on promising new technologies, but no
attempt is made here to cover the extensive development of these methods or of methods such as
wet chemical techniques, no longer in widespread use. More detailed discussions of the histories
of these methods may be found elsewhere (U.S. Environmental Protection Agency, 1993, 1996).
AX2.6.2.1 Calibration Standards
Calibration gas standards of NO, in nitrogen (certified at concentrations of approximately
5 to 40 ppm) are obtainable from the Standard Reference Material (SRM) Program of the
National Institute of Standards and Technology (NIST), formerly the National Bureau of
Standards (NBS), in Gaithersburg, MD. These SRMs are supplied as compressed gas mixtures
at about 135 bar (1900 psi) in high-pressure aluminum cylinders containing 800 L of gas at
standard temperature and pressure, dry (STPD; National Bureau of Standards, 1975; Guenther
et al., 1996). Each cylinder is supplied with a certificate stating concentration and uncertainty.
The concentrations are certified to be accurate to ±1 percent relative to the stated values.
Because of the resources required for their certification, SRMs are not intended for use as daily
working standards, but rather as primary standards against which transfer standards can be
calibrated.
AX2-150
-------
Transfer stand-alone calibration gas standards of NO in N2 (in the concentrations indicated
above) are obtainable from specialty gas companies. Information as to whether a company
supplies such mixtures is obtainable from the company, or from the SRM Program of NIST.
These NIST Traceable Reference Materials (NTRMs) are purchased directly from industry and
are supplied as compressed gas mixtures at approximately 135 bars (1,900 psi) in high-pressure
aluminum cylinders containing 4,000 L of gas at STPD. Each cylinder is supplied with a
certificate stating concentration and uncertainty. The concentrations are certified to be accurate
to within ±1 percent of the stated values (Guenther et al., 1996). Additional details can be found
in the previous AQCD for O3 (U.S. Environmental Protection Agency, 1996).
AX2.6.2.2 Measurement of Nitric Oxide
Gas-Phase Chemiluminescence (CL) Methods
Nitric oxide, NO, can be measured reliably using the principle of gas-phase
chemiluminescence induced by the reaction of NO with O3 at low pressure. Modern commercial
NOX analyzers have sufficient sensitivity and specificity for adequate measurement in urban and
many rural locations (U.S. Environmental Protection Agency, 1996). The physics of the method,
detection limits, interferences, and comparisons under field comparisons have been thoroughly
reviewed in the previous AQCD. Research grade CL instruments have been compared under
realistic field conditions to spectroscopic instruments, and the results indicate that both methods
are reliable (at concentrations relevant to smog studies) to better than 15 percent with 95 percent
confidence. Response times are on the order of 1 minute. For measurements meaningful for
understanding O3 formation, emissions modeling, and N deposition, special care must be taken
to frequently zero and calibrate the instrument. A chemical zero, by reacting the NO up stream
and out of view of the PMT, is preferred because it accounts for unsaturated hydrocarbon or
other interferences. Calibration should be performed with NIST-traceable reference material of
compressed NO in N2. Standard additions of NO at the inlet will account for NO loss or
conversion to NO2 in the lines. In summary CL methods, when operated in an appropriate
manner, can be suitable for measuring or monitoring NO (e.g., Crosley, 1996).
AX2-151
-------
Spectroscopic Methods for Nitric Oxide
Nitric oxide has also been successfully measured in ambient air with direct spectroscopic
methods; these include two-photon laser-induced fluorescence (TPLIF), tunable diode laser
absorption spectroscopy (TDLAS), and two-tone frequency-modulated spectroscopy (TTFMS).
These were reviewed thoroughly in the previous AQCD and will be only briefly summarized
here. The spectroscopic methods demonstrate excellent sensitivity and selectivity for NO with
detection limits on the order of 10 ppt for integration times of 1 min. Spectroscopic methods
compare well with the CL method for NO in controlled laboratory air, ambient air, and heavily
polluted air (e.g., Walega et al., 1984; Gregory et al., 1990; Kireev et al., 1999). These
spectroscopic methods remain in the research arena due to their complexity, size, and cost, but
are essential for demonstrating that CL methods are reliable for monitoring NO concentrations
involved in O3 formation—from 100's of ppb to around 20 ppt.
Atmospheric pressure laser ionization followed by mass spectroscopy has also been
reported for detection of NO and NO2. Garnica et al. (2000) describe a technique involving
selective excitation at one wavelength followed by ionization at a second wavelength. They
report good selectivity and detection limits well below 1 ppb. The practicality of the instrument
for ambient monitoring has yet to be demonstrated.
AX2.6.2.3 Measurements of Nitrogen Dioxide
Gas-Phase Chemiluminescence Methods
Since the previous AQCD, photolytic reduction followed by CL has been improved and the
method of laser-induced fluorescence has been developed. Ryerson et al. (2000) developed a
photolytic converter based on a Hg lamp with increased radiant intensity in the region of peak
NO2 photolysis (350 to 400 nm) and producing conversion efficiencies of 70% or more in less
than 1 s. Because the converter produces little radiation at wavelengths less than 350 nm,
interferences from HNO3 and PAN are minimal.
Alternative methods to photolytic reduction followed by CL are desirable to test the
reliability of this widely used technique. In any detector based on conversion to another species
interferences can be a problem. Several atmospheric species, PAN and HO2NO2 for example,
dissociate to NO2 at higher temperatures.
AX2-152
-------
Laser induced fluorescence for NO2 detection involves excitation of atmospheric NO2 with
laser light emitted at wavelengths too long to induce photolysis. The resulting excited molecules
relax in a photoemissive mode and the fluorescing photons are counted. Because collisions
would rapidly quench electronically excited NO2, the reactions are conducted at low pressure
(Cohen, 1999; Thornton et al., 2000; Day et al., 2002). For example Cleary et al. (2002)
describe field tests of a system that uses continuous, supersonic expansion followed by
excitation at 640 nm with a commercial cw external-cavity tunable diode laser. Sensitivity is
adequate for measurements in most continental environments (145 ppt in 1 min) and no
interferences have been identified.
Matsumi et al. (2001) describe a comparison of laser-induced fluorescence with a
photofragmentation chemiluminescence instrument. The laser-induced fluorescence system
involves excitation at 440 nm with a multiple laser system. They report sensitivity of 30 ppt in
10s and good agreement between the two methods under laboratory conditions at mixing ratios
up to 1.0 ppb. This high-sensitivity laser-induced fluorescence system has yet to undergo long-
term field tests.
NO2 can be detected by differential optical absorption spectroscopy (DOAS) in an open,
long-path system (Kim and Kim, 2001). Vandaele et al. (2002) reported that the DOAS
technique measured higher NO2 concentrations than were reported by other techniques in a
three-year study conducted in Brussels. Harder et al. (1997b) conducted an experiment in rural
Colorado involving simultaneous measurements of NO2 with DOAS and photolysis followed by
chemiluminescence. The found differences of as much as 110% in clean air from the west, but
for NO2 mixing ratios in excess of 300 ppt, the two methods agreed to better than 10%. Stutz
and Platt (1996) report less uncertainty.
AX2.6.2.4 Monitoring for NO2 Compliance Versus Monitoring for Ozone Formation
Observations of NO2 have been focused on demonstrating compliance with the NAAQS for
NO2. Today, few locations violate that standard, but NO2 and related NOy compounds remain
among the most important atmospheric trace gases to measure and understand. Commercial
instruments for NO/NOX detection are generally constructed with an internal converter for
reduction of NO2 to NO, and generate a signal referred to as NOX. These converters, generally
constructed of molybdenum oxides (MoOx), reduce not only NO2 but also most other NOy
AX2-153
-------
species (Fehsenfeld et al., 1987; Crosley, 1996; Nunnermacker et al., 1998). Thus the NOX
signal is more accurately referred to as NOy. Unfortunately with an internal converter, the
instruments may not give a faithful indication of NOy either — reactive species such as HNO3
will adhere to the walls of the inlet system. Most recently, commercial vendors such as Thermo
Environmental (Franklin, MA) have offered NO/NOy detectors with external Mo converters.
If such instruments are calibrated through the inlet with a reactive nitrogen species such as
propyl nitrate, they should give accurate measurements of total NOy, suitable for evaluation of
photochemical models. States should be encouraged to make these NOy measurements where
ever possible.
AX2.6.3 Measurements of Nitric Acid Vapor, HNO3
Accurate measurement of nitric acid vapor, HNO3, has presented a long-standing analytical
challenge to the atmospheric chemistry community. In this context, it is useful to consider the
major factors that control HNO3 partitioning between the gas and deliquesced-particulate phases
in ambient air. In equation form,
77 a
HNO3g <^ [HNO3aq] o [H+] + [NO3-] (AX2-52)
where KH is the Henry's Law constant in M atm-1 and Ka is the acid dissociation constant in M.
Thus, the primary controls on HNO3 phase partitioning are its thermodynamic properties
(KH, Ka, and associated temperature corrections), aerosol liquid water content (LWC), solution
pH, and kinetics. Aerosol LWC and pH are controlled by the relative mix of different acids and
bases in the system, hygroscopic properties of condensed compounds, and meteorological
conditions (RH, temperature, and pressure). It is evident from relationship XX that, in the
presence of chemically distinct aerosols of varying acidities (e.g., super-Jim predominantly sea
salt and sub-jam predominantly S aerosol), HNO3 will partition preferentially with the less-acidic
particles, which is consistent with observations (e.g., Huebert et al., 1996; Keene and Savoie,
1998; Keene et al., 2002). Kinetics are controlled by atmospheric concentrations of HNO3 vapor
and particulate NO3 and the size distribution and corresponding atmospheric lifetimes of
particles against deposition. Sub-jim-diameter aerosols typically equilibrate with the gas phase
AX2-154
-------
in seconds to minutes while super-um aerosols require hours to a day or more (e.g., Meng and
Seinfeld, 1996; Erickson et al., 1999). Consequently, smaller aerosol size fractions are typically
close to thermodynamic equilibrium with respect to HNO3 whereas larger size fractions (for
which atmospheric lifetimes against deposition range from hours to a few days) are often
undersaturated (e.g., Erickson et al., 1999; Keene and Savioe, 1998).
Many sampling techniques for HNO3 (e.g., standard filterpack and mist-chamber samplers)
employ upstream prefilters to remove particulate species from sample air. However, when
chemically distinct aerosols with different pHs (e.g., sea salt and S aerosols) mix together on a
bulk filter, the acidity of the bulk mixture will be greater than that of the less acidic aerosols with
which most NO3 is associated. This change in pH may cause the bulk mix to be supersaturated
with respect to HNO3 leading to volatilization and, thus, positive measurement bias in HNO3
sampled downstream. Alternatively, when undersaturated super-um size fractions (e.g., sea salt)
accumulate on a bulk filter and chemically interacts over time with HNO3 in the sample air
stream, scavenging may lead to negative bias in HNO3 sampled downsteam. Because the
magnitude of both effects will vary as functions of the overall composition and thermodynamic
state of the multiphase system, the combined influence can cause net positive or net negative
measurement bias in resulting data. Pressure drops across particle filters can also lead to artifact
volatilization and associated positive bias in HNO3 measured downstream.
Widely used methods for measuring HNO3 include standard filterpacks configured with
nylon or alkaline-impregnated filters (e.g., Goldan et al., 1983; Bardwell et al., 1990;
respectively) and standard mist chambers (Talbot et al., 1990). Samples are typically analyzed
by ion chromatography. Intercomparisons of these measurement techniques (e.g., Hering et al.,
1988; Tanner et al., 1989; Talbot et al., 1990) report differences of a factor of two or more.
More recently, sensitive HNO3 measurements based on the principle of Chemical
lonization Mass Spectroscopy (CIMS) have been reported (e.g., Huey et al., 1998; Mauldin
et al., 1998; Furutani and Akimoto, 2002; Neuman et al., 2002). CIMS relies on selective
formation of ions such as SiF5"-HNO3 or HSO4"-HNO3 followed by detection via mass
spectroscopy. Two CIMS techniques and a filter pack technique were intercompared in Boulder,
CO (Fehsenfeld et al., 1998). Results indicated excellent agreement (within 15%) between the
two CIMS instruments and between the CIMS and filterpack methods under relatively clean
conditions with HNO3 mixing ratios between 50 and 400 pptv. In more polluted air, the
AX2-155
-------
filterpack technique generally yielded higher values than the CIMS suggesting that interactions
between chemically distinct particles on bulk filters is a more important source of bias in
polluted continental air. Differences were also greater at lower temperature when particulate
NO3 corresponded to relatively greater fractions of total NO3 .
AX2.6.4 Sampling and Analysis of Volatile Organic Compounds
Hydrocarbons can be measured with gas chromatography followed by flame ionization
detection (GC-FID). Detection by mass spectroscopy is sometimes used to confirm species
identified by retention time (Westberg and Zimmerman, 1993; Dewulf and Van Langenhove,
1997). Preconcentration is typically required for less abundant species. Details are available in
AQCD 96.
Because of their variety, nonmethane hydrocarbons pose special analytical problems,
and several laboratory and field studies have recently addresses the uncertainty of VOC
measurements. An intercomparison conducted with 16 components among 28 laboratories,
showed agreement on the order of 10s of percents (Apel et al., 1994). In a more recent
intercomparison (Apel et al., 1999) 36 investigators from around the world were asked to
identify and quantify C2 to C10 hydrocarbons (HCs) in a mixture in synthetic air. Calibration was
based on gas standards of individual compounds, such as propane in air, and a 16-compound
mixture of C2 to C16 -alkanes, all prepared by NIST and certified to ± 3 percent. The
top-performing laboratories, including several in the United States, identified all the compounds
correctly, and obtained agreement of generally better than 20 percent for the 60 compounds.
Intercomparison of NMHCs in ambient air has only recently been reported by a European group
of 12 - 14 laboratories (Slemr et al., 2002). Some compounds gave several groups difficulties,
including isobutene, butadiene, methyl pentanes, and trimethyl benzenes. These
intercomparisons illustrated the need for reliable, multicomponent calibration standards.
AX2.6.4.1 Polar Volatile Organic Compounds
Many of the more reactive oxygen- and nitrogen-containing organic compounds play a role
in O3 formation and are included among the list of 189 hazardous air pollutants specified in the
1990 CAAA (U.S. Congress, 1990). These compounds are emitted directly from a variety of
sources including biogenic processes, biomass burning, industry, vehicles, and consumer
AX2-156
-------
products. Some can also be formed in the atmosphere by photochemical oxidation of
hydrocarbons. Although these compounds have been referred to collectively as PVOCs, their
reactivity and water solubility, rather than just polarity, make sampling and measurement
challenging. As indicated in the earlier AQCD, few ambient data exist for these species, but that
database has grown. The previous AQCD discusses two analytical methods for
PVOCs—cryogenic trapping techniques similar to those used for the nonpolar hydrocarbon
species, and adsorbent material for sample preconcentration. Here we discuss recently
developed methods.
Several techniques for sampling, preconcentrating and detecting oxygenated volatile
organic compounds were inter-compared during the 1995 Southern Oxidants Study Nashville
Intensive (Apel et al., 1998). Both chemical traps and derivatization followed by HPLC and
preconcentration and gas chromatography followed by mass spectrometric of flame ionization
were investigated. Both laboratory and field tests were conducted for formaldehyde,
acetaldehyde, acetone, and propanal. Substantial differences were observed indicating that
reliable sampling and measurement of PVOCs remains an analytical challenge and high research
priority.
Chemical ionization-mass spectroscopy, such as proton-transfer-reaction mass
spectroscopy (PTR-MS) can also be used for fast-response measurement of volatile organic
compounds including acetonitrile (CH3CN), methanol (CH3OH), acetone (CH3COCH3),
acetaldehyde (CH3CHO), benzene (C6H6) and toluene (C6H5CH3) (e.g., Hansel et al., 1995a,b;
Lindinger et al., 1998; Leibrock and Huey, 2000; Warneke et al., 2001). The method relies on
gas phase proton transfer reactions between H3O+ primary ions and volatile trace gases with a
proton affinity higher than that of water. Into a flow drift tube continuously flushed with
ambient air, H3O+ ions (from a hollow cathode ion source) are injected. On collisions between
H3O+ ions and organic molecules protons H+ are transferred thus charging the reagent. Both
primary and product ions are analyzed in a quadrupole mass spectrometer and detected by a
secondary electron multiplier/pulse counting system. The instrument has been successfully
employed in several field campaigns and compared to other techniques including gas
chromatography and Atmospheric Pressure Chemical Ionization Mass Spectrometer (AP-CIMS)
(Crutzen et al., 2000; Sprung et al., 2001). Sufficient sensitivity was observed for urban and
rural measurements; no interferences were discovered, although care must be exercised to avoid
AX2-157
-------
sampling losses. Commercial instruments are becoming available, but their price still precludes
widespread monitoring.
AX2-158
-------
REFERENCES
Allan, B. I; Plane, J. M. C.; McFiggans, G. (2001) Observations of OIO in the remote marine boundary layer.
Geophys. Res. Lett. 28: 1945-1948.
Allen, D. J.; Pickering, K. E. (2002) Evaluation of lightning flash rate parameterizations for use in a global chemical
transport model. J. Geophys. Res. [Atmos.] 107(D23): 10.1029/2002JD002066.
Allen, D. J.; Pickering, K. E.; Molod, A. (1997) An evaluation of deep convective mixing in the Goddard Chemical
Transport Model using ISCCP cloud parameters. J. Geophys. Res. [Atmos.] 102: 25,467-25,476.
Allen, D.; Pickering, K.; Stenchikov, G.; Thompson, A.; Kondo, Y. (2000) A three-dimensional total odd nitrogen
(NOy) simulation during SONEX using a stretched-grid chemical transport model. J. Geophys. Res. [Atmos.]
105: 3851-3876.
Alvarez, II; Senff, C. J.; Hardesty, R. M.; Parrish, D. D.; Luke, W. T.; Watson, T. B.; Daum, P. H.; Gillani, N.
(1998) Comparisons of airborne lidar measurements of ozone with airborne in situ measurements during the
1995 SouthernOxidants study. J. Geophys. Res. [Atmos.] 103: 31155-35171.
American Society for Testing and Materials (ASTM) International. (2003a) Standard test method for ozone in the
atmosphere: continuous measurement by ethylene chemiluminescence. In: Annual Book of ASTM Standards
2003, v. 11.03: D 5149-02. West Conshohocken, PA: American Society for Testing and Materials; pp.
443-447.
American Society for Testing and Materials (ASTM) International. (2003b) Standard test methods for continuous
measurement of ozone in ambient, workplace, and indoor atmospheres (ultraviolet absorption). In: Annual
Book of ASTM Standards 2003, v. 11.03: D 5156-02. West Conshohocken, PA: American Society for
Testing and Materials; pp. 448-454.
Ammann, M.; Kalberer, M.; Jost, D. T.; Tobler, L.; Rossler, E.; Piguet, D.; Gaggeler, H. W.; Baltensperger,
U. (1998) Heterogeneous production of nitrous acid on soot in polluted air masses. Nature (London)
395: 157-160.
Anastasio, C.; Faust, B. C.; Allen, J. M. (1994) Aqueous phase photochemical formation of hydrogen peroxide in
authentic cloud waters. J. Geophys. Res. [Atmos.] 99: 8231-8248.
Anastasio, C.; Faust, B.C.; Rao, C. J. (1997) Aromatic carbonyl compounds as aqueous-phase photochemical
sources of hydrogen peroxide in acidic sulfate aerosols, fogs, and clouds. 1. Non-phenolic
methoxybenzaldehydes and methoxyacetophenones with reductants (phenols). Environ. Sci. Technol.
31:218-232.
Anastasio, C.; Newberg, J. T.; Williams, D. K.; Chu, G. B.; Mathew, B. M. (1999) Photoformation of hydroxyl
radical in sea salt particles. EOS Trans. Am. Geophys. Union 80: F147.
Andino, J. M.; Smith, J. N.; Flagan, R. C.; Goddard, W. A., Ill; Seinfeld, J. H. (1996) Mechanism of the atmospheric
photooxidation of aromatics: a theoretical study. J. Phys. Chem. 100: 10,967-10,980.
Andreae, M. O. (1991) Biomass burning: its history, use, and distribution and its impact on environmental quality
and global climate. In: Levine, J. S., ed. Global biomass burning: atmospheric, climatic, and biospheric
implications. Cambridge, MA: MIT Press; pp. 1-21.
Andres Hernandez, M. D.; Burkert, J.; Reichert, L.; Stobener, D.; Meyer-Arnek, J.; Dickerson, R. R.; Doddridge,
B. G.; Burrows, J. P. (2001) Marine boundary layer peroxy radical chemistry during the AEROSOLS99
campaign: measurements and analysis. J. Geophys. Res. (Atmos.) 106(D18): 20,833-20,846.
Angevine, W. M.; Buhr, M. P.; Holloway, J. S.; Trainer, M.; Parrish, D. D.; MacPherson, J. I.; Kok, G. L.;
Schillawski, R. D.; Bowlby, D. H. (1996) Local meteorological features affecting chemical measurements at a
North Atlantic coastal site. J. Geophys. Res. (Atmos.) 101(D22): 28,935-28,946.
Apel, E. C.; Calvert, J. G.; Fehsenfeld, F. C. (1994) The nonmethane hydrocarbon intercomparison experiment
(NOMHICE): tasks 1 and 2. J. Geophys. Res. [Atmos.] 99(D8): 16,651-16,664.
Apel, E. C.; Calvert, J. G.; Riemer, D.; Pos, W.; Zika, R.; Kleindienst, T. E.; Lonneman, W. A.; Fung, K.; Fujita, E.;
Shepson, P. B.; Starn, T. K.; Roberts, P. T. (1998) Measurements comparison of oxygenated volatile organic
compounds at a rural site during the 1995 SOS Nashville Intensive. J. Geophys. Res. (Atmos.) 103(D17):
22,295-22,316.
Apel, E. C.; Calvert, J. G.; Gilpin, T. M.; Fehsenfeld, F. C.; Parrish, D. D.; Lonneman, W. A. (1999) The
Nonmethane Hydrocarbon Intercomparison Experiment (NOMHICE): task 3. J. Geophys. Res. [Atmos.]
104(021): 26,069-26,086.
Appenzeller, C.; Davies, H. C.; Norton, W. A. (1996) Fragmentation of stratospheric intrusions. J. Geophys. Res.
[Atmos.] 101: 1435-1456.
AX2-159
-------
Arnold, F.; Burger, V.; Droste-Fanke, B.; Grimm, F.; Krieger, A.; Schneider, I; Stilp, T. (1997) Acetone in the
upper troposphere and lower stratosphere: impact on trace gases and aerosols. Geophys. Res. Lett.
24: 3017-3020.
Arnold, J. R.; Dennis, R. L.; Tonnesen, G. S. (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.
Arnold, S. R.; Chipperfield, M. P.; Blitz, M. A.; Heard, D. E.; Pilling, M. J. (2004) Photodissociation of acetone:
atmospheric implications of temperature-dependent quantum yields. Geophys. Res. Lett. 31(L07110):
10.1029/2003GL019099.
Arshinov, M. Y.; Belan, B. D.; Krasnov, O. A.; Kovalevskii, V. K.; Pirogov, V. A.; Plotnikov, A. P.; Tolmachev,
G. N.; Fofonov, A. V. (2002) Comparison of ultraviolet and chemiluminescent ozonometers. Atmos. Oceanic
Opt. 15: 656-658.
Ashworth, S. H.; Allan, B. J.; Plane, J. M. C. (2002) High resolution spectroscopy of the OIO radical: implications
for the ozone-depleting potential of iodine. Geophys. Res. Lett. 29: 65-1 - 65-4.
Atherton, C. S.; Penner, J. E. (1988) The transformation of nitrogen oxides in the polluted troposphere. Tellus B
40: 380-392.
Atkinson, R. (1989) Kinetics and mechanisms of the gas-phase reactions of the hydroxyl radical with organic
compounds. Washington, DC: American Chemical Society. (J. Phys. Chem. Ref. Data Monograph no. 1).
Atkinson, R. (1990) Gas-phase tropospheric chemistry of organic compounds: a review. Atmos. Environ. Part A
24: 1-41.
Atkinson, R. (1991) Kinetics and mechanisms of the gas-phase reactions of the NO3 radical with organic
compounds. J. Phys. Chem. Ref. Data 20: 459-507.
Atkinson, R. (1994) Gas-phase tropospheric chemistry of organic compounds. Washington, DC: American Chemical
Society. (J. Phys. Chem. Ref. Data Monograph no. 2).
Atkinson, R. (1997) Atmospheric reactions of alkoxy and p-hydroxyalkoxy radicals. Int. J. Chem. Kinet. 29: 99-111.
Atkinson, R. (2000) Atmospheric chemistry of VOCs and NOX. Atmos. Environ. 34: 2063-2101.
Atkinson, R.; Arey, J. (2003) Gas-phase tropospheric chemistry of biogenic volatile organic compounds: a review.
Atmos. Environ.: 10.1016/81352-2310(03)00391-1.
Atkinson, R.; Aschmann, S. M. (1994) Products of the gas-phase reactions of aromatic hydrocarbons: effect of NO2
concentration. Int. J. Chem. Kinet. 26: 929-944.
Atkinson, R.; Carter, W. P. L.; Darnall, K. R.; Winer, A. M.; Pitts, J. N., Jr. (1980) A smog chamber and modeling
study of the gas phase NOx-air photooxidation of toluene and the cresols. Int. J. Chem. Kinet. 12: 779-836.
Atkinson, R.; Winer, A. M.; Pitts, J. N., Jr. (1986) Estimation of night-time N2O5 concentrations from ambient NO2
and NO3 radical concentrations and the role of N2O5 in night-time chemistry. Atmos. Environ. 20: 331-339.
Atkinson, R.; Arey, J.; Zielinska, B.; Aschmann, S. M. (1990) Kinetics and nitro-products of the gas-phase OH and
NO3 radical-initiated reactions of naphthalene-
-------
Bais, A. F.; Madronich, S.; Crawford, I; Hall, S. R.; Mayer, B.; Van Weele, M.; Lenoble, I; Calvert, J. G.; Cantrell,
C. A.; Shelter, R. E.; Hofzumahaus, A.; Koepke, P.; Monks, P. S.; Frost, G.; McKenzie, R.; Krotkov, N.;
Kylling, A.; Swartz, W. H.; Lloyd, S.; Pfister, G.; Martin, T. J.; Roeth, E.-P.; Griffioen, E.; Ruggaber, A.;
Krol, M.; Kraus, A.; Edwards, G. D.; Mueller, M.; Lefer, B. L. (2003) International Photolysis Frequency
Measurement and Model Intercomparison (IPMMI): Spectral actinic solar flux measurements and modeling.
J. Geophys. Res. [Atmos.] 108(D16): 10.1029/2002JD002891.
Balkanski, Y. J.; Jacob, D. J.; Gardener, G. M.; Graustein, W. C.; Turekian, K. K. (1993) Transport and residence
times of tropospheric aerosols inferred from a global three-dimensional simulation of 210Pb. J. Geophys. Res.
[Atmos.] 98: 20,573-20,586.
Bandow, H.; Washida, N. (1985a) Ring-cleavage reactions of aromatic hydrocarbons studied by FT-IR
spectroscopy. II photooxidation of o-, m-, and^-xylenes in the NOx-air system. Bull. Chem. Soc. Jpn.
58: 2541-2548.
Bandow, H.; Washida, N. (1985b) Ring-cleavage reactions of aromatic hydrocarbons studied by FT-IR
spectroscopy. Ill photooxidation of 1,2,3-, 1,2,4-, and 1,3,5-trimethylbenzene in the NOx-air system. Bull.
Chem. Soc. Jpn. 58: 2549-2555.
Banic, C. M.; Leaitch, W. R.; Isaac, G. A.; Couture, M. D.; Kleinman, L. L; Springston, S. R.; MacPherson, J. I.
(1996) Transport of ozone and sulfur to the North Atlantic atmosphere during the North Atlantic Regional
Experiment. J. Geophys. Res. [Atmos.] 101: 29,091-29,104.
Banta, R. M.; Senff, C. J.; White, A. B.; Trainer, M.; McNider, R. T.; Valente, R. J.; Mayor, S. D.; Alvarez, R. J.;
Hardesty, R. M.; Parrish, D.; Fehsenfeld, F. C. (1998) Daytime buildup and nighttime transport of urban
ozone in the boundary layer during a stagnation episode. J. Geophys. Res. [Atmos.] 103: 22,519-22,544.
Banta, R. M.; Senff, C. J.; Nielsen-Gammon, J.; Darby, L. S.; Ryerson, T. B.; Alvarez, R. J.; Sandberg, S. P.;
Williams, E. J.; Trainer, M. (2005) A bad air day in Houston. Bull. Am. Meteorol. Soc. 86: 657-669.
Bardwell, C. A.; Maben, J. R.; Hurt, J. A.; Keene, W. C.; Galloway, J. N.; Boatman, J. F.; Wellman, D. (1990) A
technique using high-flow, dichotomous filter packs for measuring major atmospheric chemical constituents.
Glob. Biogeochem. Cycles 4: 151-163.
Barrie, L. A.; Bottenheim, J. W.; Crutzen, P. J.; Rasmussen, R. A. (1988) Ozone destruction at polar sunrise in the
lower Arctic atmosphere. Nature 334: 138-141.
Bartolotti, L. J.; Edney, E. O. (1995) Density functional theory derived intermediates from the OH initiated
atmospheric oxidation of toluene. Chem. Phys. Lett. 245: 119-122.
Baumann, K.; Williams, E. J.; Angevine, W. M.; Roberts, J. M.; Norton, R. B.; Frost, G. J.; Fehsenfeld, F. C.;
Springston, S. R.; Bertman, S. B.; Hartsell, B. (2000) Ozone production and transport near Nashville,
Tennessee: results from the 1994 study at New Hendersonville. J. Geophys. Res. [Atmos.] 105(D7):
9137-9153.
Berkowitz, C. M.; Shaw, W. J. (1997) Airborne measurements of boundary layer chemistry during the Southern
Oxidant Study: a case study. J. Geophys. Res. [Atmos.] 102: 12,795-12,804.
Berkowitz, C. M.; Fast, J. D.; Sprinston, S. R.; Larsen, R. J.; Spicer, C. W.; Doskey, P. V.; Hubbe, J. M.; Plastridge,
R. (1998) Formation mechanisms and chemical characteristics of elevated photochemical layers over the
northeast United States. J. Geophys. Res. [Atmos.] 103: 10,631-10,647.
Berkowitz, C. M.; Zaveri, R. A.; Bian, X.; Zhong, S.; Disselkamp, R. S.; Laulainen, N. S.; Chapman, E. G. (2001)
Aircraft observations of aerosols, O3 and NO,, in a nighttime urban plume. Atmos. Environ. 35: 2395-2404.
Berndt, T.; Boge, O.; Herrmann, H. (1999) On the formation of benzene oxide/oxepin in the gas-phase reaction of
OH radicals with benzene. Chem. Phys. Lett. 314: 435-442.
Berndt, T.; Boge, O.; Stratmann, F. (2003) Gas-phase ozonolysis of a-pinene: gaseous products and particle
formation. Atmos. Environ. 37: 3933-3945.
Berntsen, T.; Isaksen, I. S. A. (1997) A global three-dimensional chemical transport model for the troposphere.
1. Model description and CO and ozone results. J. Geophys. Res. [Atmos.] 102: 21,239-21,280.
Bey, I.; Jacob, D. J.; Logan, J. A.; Yantosca, R. M. (2001a) Asian chemical outflow to the Pacific in spring: origins,
pathways, and budgets. J. Geophys. Res. [Atmos.] 106: 23,097-23,113.
Bey, I.; Jacob, D. J.; Yantosca, R. M.; Logan, J. A.; Field, B.; Fiore, A. M.; Li, Q.; Liu, H.; Mickley, L. J.; Schultz,
M. G. (200 Ib) Global modeling of tropospheric chemistry with assimilated meteorology: model description
and evaluation. J. Geophys. Res. [Atmos.] 106: 23,073-23,095.
Bishop, G. A.; Stedman, D. H. (1996) Measuring the emissions of passing cars. Ace. Chem. Res. 29: 489-495.
Black, D. R.; Harley, R. A.; Hering, S. V.; Stolzenburg, M. R. (2000) A new, portable, real-time monitor. Environ.
Sci. Technol. 34:3031-3040.
AX2-161
-------
Blanchard, C. L. (2000) Ozone process insights from field experiments. Part III: extent of reaction and ozone
formation. Atmos. Environ. 34: 2035-2043.
Blanchard, C. L.; Stoeckenius, T. (2001) Ozone response to precursor control: comparision of data analysis methods
with the predictions of photochemical air quality simulation models. Atmos. Environ. 35: 1203-1215.
Blanchard, C. L.; Lurmann, F. W.; Roth, P. M; Jeffries, H. E.; Korc, M. (1999) The use of ambient data to
corroborate analyses of ozone control strategies. Atmos. Environ. 33: 369-381.
Blanchard, P.; Brook, J. R.; Brazil, P. (2002) Chemical characterization of the organic fraction of atmospheric
aerosol at two sites in Ontario, Canada. J. Geophys. Res. [Atmos.] 107(D21): 10.1029/2001JD000627.
Blitz, M. A.; Heard, D. E.; Pilling, M. J.; Arnold, S. R.; Chipperfield, M. P. (2004) Pressure and
temperature-dependent quantum yields for the photodissociation of acetone between 279 and 327.5 nm.
Geophys. Res. Lett. 31: 10.1029/2003GL018793.
Bloom, S. C.; Takacs, L. L.; Da Silva, A. M.; Ledvina, D. (1996) Data assimilation using incremental analysis
updates. Mon. Weather Rev. 124: 1256-1271.
Blough, N. V. (1997) Photochemistry in the sea-surface microlayer. In: Liss, P. S.; Duce, R. A., eds. The sea surface
and global change. New York, NY: Cambridge University Press, pp. 383-424.
Blough, N. V.; Zepp. R. G. (1995) Reactive oxygen species in natural waters. In: Foote, C. S.; Valentine, A.;
Greenberg, A.; Liebman, J. F., eds. Active Oxygen in Chemistry. New York, NY: Chapman and Hall, Ch.8,
pp. 280-333.
Blumenthal, D. L.; Lurmann, F. W.; Kumar, N.; Dye, T. S.; Ray, S. E.; Korc, M. E.; Londergan, R.; Moore, G.
(1997) Transport and mixing phenomena related to ozone exceedances in the northeast U.S. (analysis based
onNARSTO-northeastdata). Available: http://capita.wustl.edu/otag/reports/otagrept/otagrept.html (30
October 2003).
Bonn, B.; Moortgat, G. K. (2003) Sesquiterpene ozonolysis: origin of atmospheric new particle formation from
biogenic hydrocarbons. Geophys. Res. Lett. 30: 10.1029/2003GL017000.
Bonn, B.; Von Kuhlmann, R.; Lawrence, M. G. (2004) High contribution of biogenic hydroperoxides to secondary
organic aerosol formation. Geophys. Res. Lett. 31: L10108, 10.1029/2003GL019172.
Bonner, W. D. (1968) Climatology of the low level jet. Mon. Weather Rev. 96: 833-850.
Brasseur, G. P.; Hauglustaine, D. A.; Walters, S.; Rasch, P. J.; Mtiller, J.-F.; Granier, C.; Tie, X. X. (1998)
MOZART, a global transport model for ozone and related chemical tracers: 1. model description. J. Geophys.
Res. [Atmos.] 103: 28,265-28,289.
Brook, J. R.; Sirois, A.; Clarke, J. F. (1996) Comparison of dry deposition velocities for SO2, HNO3, and SO42"
estimated with two inferential models. Water Air Soil Pollut. 87: 205-218.
Broske, R.; Kleffmann, J.; Wiesen, P. (2003) Heterogeneous conversion of NO2 on secondary organic aerosol
surfaces: a possible source of nitrous acid (HONO) in the atmosphere? Atmos. Chem. Phys. 3: 469-474.
Briihl, C.; Poschl, U.; Crutzen, P. J.; Steil, B. (2000) Acetone and PAN in the upper troposphere: impact on ozone
production from aircraft emissions. Atmos. Environ. 34: 3931-3938.
Brune, W. H.; Tan, D.; Faloona, I. F.; Jaegle, L.; Jacob, D. J.; Heikes, B. G.; Snow, J.; Kondo, Y.; Shelter, R.;
Sachse, G. W.; Anderson, B.; Gregory, G. L.; Vay, S.; Singh, H. B.; Davis, D. D.; Crawford, J. H.; Blake,
D. R. (1999) OH and HO2 chemistry in the North Atlantic free troposphere. Geophys. Res. Lett. 26:
3077-3080.
Brunner, D.; Staehelin, J.; Jeker, D. (1998) Large-scale nitrogen oxide plumes in the tropopause region and
implications for ozone. Science (Washington, DC) 282: 1305-1309.
Buhr, M. P.; Trainer, M.; Parrish, D. D.; Sievers, R. E.; Fehsenfeld, F. C. (1992) Assessment of pollutant emission
inventories by principal component analysis of ambient air measurements. Geophys. Res. Lett.
19: 1009-1012.
Buhr, M.; Sueper, D.; Trainer, M.; Goldan, P.; Kuster, B.; Fehsenfeld, F. (1996) Trace gas and aerosol
measurements using aircraft data from the North Atlantic Regional Experiment (NARE 1993). J. Geophys.
Res. [Atmos.] 101: 29,013-29,027.
Burkert, J.; Behmann, T.; Andres-Hernandez, M. D.; Stobener, D.; Weilknmeyer, M.; Perner, D.; Burrows, J. P.
(200 la) Measurements of peroxy radicals in a forested area of Portugal. Chemosphere Glob. Change Sci.
3: 327-338.
Burkert, J.; Andres-Hernandez, M. D.; Stobener, D.; Burrows, J. P.; Weilknmeyer, M.; Kraus, A. (200Ib) Peroxy
radical and related trace gas measurements in the boundary layer above the Atlantic Ocean. J. Geophys. Res.
(Atmos.) 106: 5457-5477.
AX2-162
-------
Burkert, I; Andres-Hernandez, M. D.; Reichert, L.; Meyer-Arnek, J.; Burrows, J. P.; Doddridge, B. G.; Dickerson,
R. R.; Mtihle, J.; Zhan, A.; Carsey, T. (2003) Trace gas and radical diurnal behavior in the marine boundary
layer during INDOEX 1999. J. Geophys. Res. (Atmos.) 108(D8): 10.1029/2002JD002790.
Byun, D.; Young, J.; Gipson, G.; Schere, K.; Godowitch, J. (1998) Urban air quality simulation with community
multi-scale air quality (CMAQ) modeling system. Research Triangle Park, NC: U.S. Environmental
Protection Agency, National Exposure Research Laboratory; report no., EPA--600/A-98/115. Available from:
NTIS, Springfield, VA; PB99-102246.
Calvert, J. G.; Madronich, S. (1987) Theoretical study of the initial products of the atmospheric oxidation of
hydrocarbons. J. Geophys. Res. [Atmos.] 92: 2211-2220.
Calvert, J. G.; Yarwood, G.; Dunker, A. M. (1994) An evaluation of the mechanism of nitrous acid formation in the
urban atmosphere. Res. Chem. Intermed. 20: 463-502.
Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich, S.; Moortgat, G. K.; Wallington, T. J.; Yarwood, G. (2000) The
mechanisms of atmospheric oxidation of the alkenes. New York, NY: Oxford University Press, Inc.
Calvert, J. G.; Atkinson, R.; Becker, K. H.; Kamens, R. M.; Seinfeld, J. H.; Wallington, T. J.; Yarwood, G. (2002)
The mechanisms of atmospheric oxidation of aromatic hydrocarbons. New York, NY: Oxford University
Press, Inc.
Cantrell, C. A.; Shelter, R. E.; Gilpin, T. M.; Calvert, J. G.; Eisele, F. L.; Tanner, D. J. (1996) Peroxy radical
concentrations measured and calculated from trace gas measurements in the Mauna Loa Observatory
Photochemistry Experiment 2. J. Geophys. Res. (Atmos.) 101: 14,653-14,664.
Cantrell, C. A.; Calvert, J. G.; Bais, A.; Shelter, R. E.; Lefer, B. L.; Edwards, G. D. (2003) Overview and
conclusions of the International Photolysis Frequency Measurement and Modeling Intercomparison (IPMMI)
study. J. Geophys. Res. (Atmos.) 108(D16): 10.1029/2002JD002962.
Cardelino, C. A.; Chameides, W. L. (1990) Natural hydrocarbons, urbanization, and urban ozone. J. Geophys. Res.
[Atmos.] 95: 13,971-13,979.
Cardelino, C. A.; Chameides, W. L. (1995) An observation-based model for analyzing ozone precursor relationships
in the urban atmosphere. J. Air Waste Manage. Assoc. 45: 161-180.
Cardelino, C. A.; Chameides, W. L. (2000) The application of data from photochemical assessment monitoring
stations to the observation-based model. Atmos. Environ. 34: 2325-2332.
Carpenter, L. J.; Stages, W. T. Liss, P. S.; Penkett, S. A.; Alicke, B.; Hebestreit, K.; Platt, U. (1999) Short-lived
alkyl iodides and bromides at Mace Head, Ireland: links to biogenic sources and halogen oxide production.
J. Geophys. Res. (Atmos.) 104: 1679-1689.
Carroll, M. A.; Hastie, D. R.; Ridley, B. A.; Rodgers, M. O.; Torres, A. L.; Davis, D. D.; Bradshaw, J. D.;
Sandholm, S. T.; Schiff, H. L; Karecki, D. R.; Harris, G. W.; Mackay, G. L; Gregory, G. L.; Condon, E. P.;
Trainer, M.; Hubler, G.; Montzka, D. D.; Madronich, S.; Albritton, D. L.; Singh, H. B.; Beck, S. M.;
Shipham, M. C.; Bachmeier, A. S. (1990) Aircraft measurement of NOX over the eastern Pacific and
continental United States and implications for ozone production. J. Geophys. Res. [Atmos.] 95:
10,205-10,233.
Carroll, M. A.; Bertman, S. B.; Shepson, P. B. (2001) Overview of the program for research on oxidants,
photochemistry, emissions and transport (PROPHET) summer 1998 measurements intensive. J. Geophys.
Res. [Atmos.] 106: 24,275-24,288.
Carter, W. P. L. (1990) A detailed mechanism for the gas-phase atmospheric reactions of organic compounds.
Atmos. Environ. Part A 24: 481-518.
Carter, W. P. L. (1994) Development of ozone reactivity scales for volatile organic compounds. J. Air Waste
Manage. Assoc. 44: 881-899.
Carter, W. P. L. (1995) Computer modeling of environmental chamber studies of maximum incremental reactivities
of volatile organic compounds. Atmos. Environ. 29: 2513.
Castro, T; Madronich, S; Rivale, S; Muhlia, A; Mar, B. (2001) The influence of aerosols on photochemical smog in
Mexico City. Atmos. 35: 1765-1772..
Chameides, W. L.; Stelson, A. W. (1992) Aqueous-phase chemical processes in deliquescent sea-salt aerosols: A
mechanism that couples the atmospheric cycles of S and sea salt. J. Geophys. Res. [Atmos.] 97(D18):
20,565-20,580.
Chameides, W. L.; Lindsay, R. W.; Richardson, J.; Kiang, C. S. (1988) The role of biogenic hydrocarbons in urban
photochemical smog: Atlanta as a case study. Science (Washington, DC) 241: 1473-1475.
Chameides, W. L.; Fehsenfeld, F.; Rodgers, M. O.; Cardelino, C.; Martinez, J.; Parrish, D.; Lonneman, W.; Lawson,
D. R.; Rasmussen, R. A.; Zimmerman, P.; Greenberg, J.; Middleton, P.; Wang, T. (1992) Ozone precursor
relationships in the ambient atmosphere. J. Geophys. Res. [Atmos.] 97: 6037-6055.
AX2-163
-------
Chang, J. S.; Brost, R. A.; Isaken, I. S. A.; Madronich, S.; Middleton, P.; Stockwell, W. R.; Walcek, C. J. (1987)
A three-dimensional Eulerian acid deposition model: physical concepts and formulation. J. Geophys. Res.
[Atmos.] 92: 14,681-14,700.
Chang, M. E.; Cardelino, C.; Hartley, D.; Chang, W. L. (1996) Inverse modeling of biogenic emissions. Geophys.
Res. Lett. 23: 3007ff.
Chang, M. E.; Hartley, D. E.; Cardelino, C.; Hass-Laursen, D.; Chang, W. L. (1997) On using inverse methods for
resolving emissions with large spatial inhomogeneities. J. Geophys. Res. (Atmos.) 102: 16,023-16,036.
Chang, S.; McDonand-Buller, E. C.; Kimura, Y.; Russell, M. M.; Tanaka, P. L.; Allen, D. T. (2002) Spatial and
temporal impacts of chlorine chemistry on ozone formation in southeastern Texas. Austin, TX: The Texas
Natural Resource Conservation Commission; Contract No. 9880077600-18. Available:
ftp://ftp.tnrcc.state.tx. us/pub/OEPAA/TAD/Modeling/HGAQSE/Contract_Reports/Others/Impacts_Chlorine_
on_Ozone_Formation.pdf (17 December 2003).
Chatfield, R. B.; Crutzen, P. J. (1984) Sulfur dioxide in remote oceanic air: cloud transport of reactive precursors.
J. Geophys. Res. 89: 7111-7132.
Chesselet, R.; Fontugne, M.; Buat-Menard, P.; Ezat, U.; Lambert, C. E. (1981) The origin of paniculate organic
carbon in the marine atmosphere as indicated by its stable carbon isotopic composition. Geophys. Res. Lett.
8: 345-348.
Ching, J. K. S.; Alkezweeny, A. J. (1986) Tracer study of vertical exchange by cumulus clouds. J. Clim. Appl.
Meteorol. 25: 1702-1711.
Ching, J. K. S.; Shipley, S. T.; Browell, E. V. (1988) Evidence for cloud venting of mixed layer ozone and aerosols.
Atmos. Environ. 22: 225-242.
Ching, J. K. S.; Byun, D.; Young, J.; Binkowski, F. S.; Pleim, J.; Roselle, S.; Godowitch, J.; Benjey, W.; Gipson, G.
(1998) Science features inModels-3 Community Multiscale Air Quality System. In: Preprints of the 10th
Joint AMS/AWMA Conference on Applications of Air Pollution Meteorology; Phoenix, AZ. Pittsburgh, PA:
Air & Waste Management Association.
Chock, D. P.; Winkler, S. L. (1994) A comparison of advection algorithms coupled with chemistry. Atmos. Environ.
28: 2659-2675.
Choi, W.; Leu, M.-T. (1998) Nitric acid uptake and decomposition on black carbon (soot) surfaces: its implications
for the upper troposphere and lower stratosphere. J. Phys. Chem. A 102: 7618-7630.
Civerolo, K.; Dickerson, R. R. (1998) Nitric oxide soil emissions from tilled and untilled cornfields. Agric. For.
Meteorol. 90:307-311.
Claeys, M.; Graham, B.; Vas, G.; Wang, W.; Vermeylen, R.; Pashynska, V.; Cafmeyer, J.; Guy on, P.; Andreae,
M. O.; Artaxo, P.; Maenhaut, W. (2004a) Formation of secondary organic aerosols through photooxidation of
isoprene. Science (Washington, DC) 303: 1173-1176.
Claeys, M.; Wang, W.; Ion, A. C.; Kourtchev, I.; Gelencser, A.; Maenhaut, W. (2004b) Formation of secondary
organic aerosols from isoprene and its gas-phase oxidation products through reaction with hydrogen peroxide.
Atmos. Environ. 38: 4093-4098.
Clark, T. L.; Karl, T. R. (1982) Application of prognostic meteorological variables to forecasts of daily maximum
one-hour ozone concentrations in the northeastern United States. J. Appl. Meteorol. 21: 1662-1671.
Cleary, P. A.; Wooldridge, P. J.; Cohen, R. C. (2002) Laser-induced fluorescence detection of atmospheric NO2 with
a commercial diode laser and a supersonic expansion. Appl. Optics 41: 6950-6956.
Cohen, R. C. (1999) Laser-induced fluorescence detection of atmospheric NO2 at parts per trillion mixing ratios:
implications for nitrogen oxide photochemistry in the stratosphere and troposphere. Abstr. Pap. Am. Chem.
Soc. (Phys.)218(Pt. 2): 262.
Collins, W. J.; Stevenson, D. S.; Johnson, C. E.; Derwent, R. G. (1997) Tropospheric ozone in a global-scale
three-dimensional Lagrangian model and its response to NO^ emission controls. J. Atmos. Chem. 26:
223-274.
Cooper, O. R.; Moody, J. L. (2000) Meteorological controls on ozone at an elevated eastern United States regional
background monitoring site. J. Geophys. Res. [Atmos.] 105: 6855-6869.
Cooper, O. R.; Moody, J. L.; Davenport, J. C.; Oltmans, S. J.; Johnson, B. J.; Chen, X.; Shepson, P. B.; Merrill, J. T.
(1998) Influence of springtime weather systems on vertical ozone distributions over three North American
sites. J. Geophys. Res. [Atmos.] 103: 22,001-22,013.
Cooper, O. R.; Moody, J. L.; Parrish, D. D.; Trainer, M.; Ryerson, T. B.; Holloway, J. S.; Htibler, G.; Fehsenfeld,
F. C.; Oltmans, S. J.; Evans, M. J. (2001) Trace gas signatures of the airstreams within North Atlantic
cyclones: case studies from the North Atlantic Regional Experiment (NARE '97) aircraft intensive.
J. Geophys. Res. [Atmos.] 106: 5437-5456.
AX2-164
-------
Cooper, O. R.; Moody, J. L.; Parrish, D. D.; Trainer, M; Ryerson, T. B.; Holloway, J. S.; Htibler, G.; Fehsenfeld,
F. C.; Evans, M. J. (2002a) Trace gas composition of mid-latitude cyclones over the western North Atlantic
Ocean: a conceptual model. J. Geophys. Res. [Atmos.] 107(D7): 10.1029/2000JD000901.
Cooper, O. R.; Moody, J. L.; Parrish, D. D.; Trainer, M.; Holloway, J. S.; Htibler, G.; Fehsenfeld, F. C.; Oltmans,
S. J.; Evans, M. J. (2002b) Trace gas composition of mid-latitude cyclones over the western North Atlantic
Ocean: a seasonal comparison of O3 and CO. J. Geophys. Res. [Atmos.] 107(D7): 10.1029/2001JD000902.
Cooper, O.; Forster, C.; Parrish, D.; Dunlea, E.; Htibler, G.; Fehsenfeld, F.; Holloway, J.; Oltmans, S.; Johnson, B.;
Wimmers, A.; Horowitz, L. (2004a) On the life cycle of a stratospheric intrusion and its dispersion into
polluted warm conveyor belts. J. Geophys. Res. (Atmos.) 109(D23S09): 10.1029/2003 JD004006.
Cooper, O. R.; Forster, C.; Parrish, D.; Trainer, M.; Dunlea, E.; Ryerson, T.; Htibler, G.; Fehsenfeld, F.; Nicks, D.;
Holloway, J.; De Gouw, J.; Warneke, C.; Roberts, J. M.; Flocke, F.; Moody, J. (2004b) A case study of
transpacific warm conveyor belt transport: the influence of merging airstreams on trace gas import to North
America. J. Geophys. Res. (Atmos.) 109(D23S08): 10.1029/2003JD003624.
Corchnoy, S. B.; Atkinson, R. (1990) Kinetics of the gas-phase reactions of OH and NO3 radicals with 2-carene,
l,8-cineole,/>-cymene, and terpinolene. Environ. Sci. Technol. 24: 1497-1502.
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.
Cox, R. A.; Bloss, W. J.; Jones, R. L.; Rowley, D. M. (1999) OIO and the atmospheric cycle of iodine. Geophys.
Res. Lett. 26: 1857-1860.
Crawford, J. H.; Davis, D. D.; Chen, G.; Buhr, M.; Oltmans, S.; Weller, R.; Mauldin, L.; Eisele, F.; Shelter, R.;
Lefer, B.; Arimoto, R.; Hogan, A. (2001) Evidence for photochemical production of ozone at the South Pole
surface. Geophys. Res. Lett. 28: 3641-3644.
Crosley, D. R. (1996) NOyblue ribbon panel. J. Geophys. Res. [Atmos.] 101(D1): 2049-2052.
Crutzen, P. (1973) A discussion of the chemistry of some minor constituents in the stratosphere and troposphere.
Pure Appl. Geophys. 106-108: 1385-1399.
Crutzen, P. J.; Gidel, L. T. (1983) A two-dimensional photochemical model of the atmosphere. 2. The tropospheric
budgets of the anthropogenic chlorocarbons, CO, CH4, CH3C1 and the effect of various NOX sources on
tropospheric ozone. J. Geophys. Res. [Atmos.] 88: 6641-6661.
Crutzen, P. J.; Lawrence, M. G.; Poschl, U. (1999) On the background photochemistry of tropospheric ozone. Tellus
51A-B: 123-146.
Crutzen, P. J.; Williams, J.; Poschl, U.; Hoor, P.; Fischer, H.; Warneke, C.; Holzinger, R.; Hansel, A.; Lindinger,
W.; Scheeren, B.; Lelieveld, J. (2000) High spatial and temporal resolution measurements of primary organics
and their oxidation products over the tropical forest of Surinam. Atmos. Environ. 34: 1161-1165.
Danielsen, E. F. (1968) Stratospheric-tropospheric exchange based on radioactivity, ozone and potential vorticity.
J. Atmos. Sci. 25:502-518.
Danielsen, E. F. (1980) Stratospheric source for unexpectedly large values of ozone measured over the Pacific Ocean
during Gametag, August 1977. J. Geophys. Res. C: Oceans Atmos. 85: 401-412.
Danielsen, E. F. (1993) In situ evidence of rapid, vertical, irreversible transport of lower tropospheric air into the
lower tropical stratosphere by convective cloud turrets and by larger-scale upwelling in tropical cyclones.
J. Geophys. Res. [Atmos.] 98(D5): 8665-8681.
Danielsen, E. F.; Mohnen, V. A. (1977) Project Dustorm report: ozone transport, in situ measurements, and
meteorological analyses of tropopause folding. J. Geophys. Res. [Atmos.] 82: 5867-5877.
Daum, P. H.; Kleinman, L. I.; Hills, A. J.; Lazrus, A. L.; Leslie, A. C. D.; Busness, K.; Boatman, J. (1990)
Measurement and interpretation of concentrations of H2O2 and related species in the upper midwest during
summer. J. Geophys. Res. [Atmos.] 95: 9857-9871.
Daum, P. H.; Kleinman, L. I.; Newman, L.; Luke, W. T.; Weinstein-Lloyd, J.; Berkowitz, C. M.; Busness, K. M.
(1996) Chemical and physical properties of plumes of anthropogenic pollutants transported over the North
Atlantic during the North Atlantic Regional Experiment. J. Geophys. Res. [Atmos.] 101: 29,029-29,042.
Davidson, E. A.; Kingerlee, W. (1997) A global inventory of nitric oxide emissions from soils. Nutr. Cycling
Agroecosyst. 48: 37-50.
Davies, T. D.; Schuepbach, E. (1994) Episodes of high ozone concentrations at the earth's surface resulting from
transport down from the upper troposphere/lower stratosphere: a review and case studies. Atmos. Environ.
28: 53-68.
Davis, D. D.; Bellinger, W.; Fischer, S. (1975) A kinetics study of the reaction of the OH free radical with aromatic
compounds I. Absolute rate constants for reaction with benzene and toluene at 300°K. J. Phys. Chem.
79: 293-294.
AX2-165
-------
Day, D. A.; Wooldridge, P. I; Dillon, M. B.; Thornton, J. A.; Cohen, R. C. (2002) A thermal dissociation
laser-induced fluorescence instrument for in situ detection of NO2, peroxy nitrates, alkyl nitrates, and HNO3.
J. Geophys. Res. (Atmos.) 107(D5-D6): 10.1029/2001JD000779.
Day, D. A.; Dillon, M. B.; Wooldridge, P. J.; Thornton, J. A.; Rosen, R. S.; Wood, E. C.; Cohen, R. C. (2003) On
alkyl nitrates, O3, and the "missing NOy". J. Geophys. Res. [Atmos.] 108(D16): 10.1029/2003JD003685.
DeCaria, A. J.; Pickering, K. E.; Stenchikov, G. L.; Scala, J. R.; Stith, J. L.; Dye, J. E.; Ridley, B. A.; Laroche, P.
(2000) A cloud-scale model study of lightning-generate NOX in an individual thunderstorm during
STERAO-A. J. Geophys. Res. [Atmos.] 105: 11601-11616.
De Reus, M.; Dentener, F.; Thomas, A.; Borrmann, S.; Strom, J.; Lelieveld, J. (2000) Airborne observations of dust
aerosol over the North Atlantic Ocean during ACE 2: indications for heterogeneous ozone destruction.
J. Geophys. Res. [Atmos.] 105: 15,263-15,275.
Derwent, R. G.; Collins, W. J.; Johnson, C. E.; Stevenson, D. S. (2001) Transient behaviour of tropospheric ozone
precursors in a global 3-D CTM and their indirect greenhouse effects. Climatic Change 49: 463-487.
Dewulf, J.; Van Langenhove, H. (1997) Analytical techniques for the determination and measurement data of 7
chlorinated Cl- and C2-hydrocarbons and 6 monocyclic aromatic hydrocarbons in remote air masses: an
overview. Atmos. Environ. 31: 3291-3307.
Dibb, J. E.; Arsenault, M.; Peterson, M. C. (2002) Fast nitrogen oxide photochemistry in Summit, Greenland snow.
Atmos. Environ. 36: 2501-2511.
Dickerson, R. R.; Huffman, G. J.; Luke, W. T.; Nunnermacker, L. J.; Pickering, K. E.; Leslie, A. C. D.; Lindsey,
C. G.; Slinn, W. G. N.; Kelly, T. J.; Damn, P. H.; Delany, A. C.; Greenberg, J. P.; Zimmerman, P. R.;
Boatman, J. F.; Ray, J. D.; Stedman, D. H. (1987) Thunderstorms: an important mechanism in the transport of
air pollutants. Science (Washington, DC) 235: 460-465.
Dickerson, R. R.; Doddridge, E.G.; Kelley, P.; Rhoads, K. P. (1995) Large-scale pollution of the atmosphere over
the remote Atlantic Ocean: evidence from Bermuda. J. Geophys. Res. [Atmos.] 100: 8945-8952.
Dickerson, R. R.; Kondragunta, S.; Stenchikov, G.; Civerolo, K. L.; Doddridge, B. G.; Holben, B. N. (1997) The
impact of aerosols on solar ultraviolet radiation and photochemical smog. Science (Washington, DC)
278: 827-830.
Dickerson, R. R.; Rhoads, K. P.; Carsey, T. P.; Oltmans, S. J.; Burrows, J. P.; Crutzen, P. J. (1999) Ozone in the
remote marine boundary layer: a possible role for halogens. J. Geophys. Res. [Atmos.] 104: 21,385-21,395.
Docherty, K. S.; Wu, W.; Lim, Y. B.; Ziemann, P. J. (2005) Contributions of organic peroxides to secondary aerosol
formed from reactions of monoterpenes with O3. Environ. Sci. Technol. 39: 4049-4059.
Domine, F.; Shepson, P. B. (2002) Air-snow interactions and atmospheric chemistry. Science (Washington, DC)
297: 1506-1510.
Dommen, J.; Prevot, A. S. H.; Hering, A. M.; Staffelbach, T.; Kok, G. L.; Schilawsky, R. D. (1999) Photochemical
production and aging of an urban air mass. J. Geophys. Res. (Atmos.) 104: 5493-5506.
Dudhia, J. (1993) A nonhydrostatic version of the Perm State-NCAR mesoscale model: validation tests and
simulation of an Atlantic cyclone and cold front. Mon. Weather Rev. 121: 1493-1513.
Dumdei, B. E.; O'Brien, R. J. (1984) Toluene's degradation products under simulated atmospheric conditions. Nature
(London) 311: 248-250.
Dumdei, B. E.; Kenny, D. V.; Shepson, P. B.; Kleindienst, T. E.; Nero, C. M.; Cupitt, L. T.; Claxton, L. D. (1988)
MS/MS analysis of the products of toluene photooxidation and measurement of their mutagenic activity.
Environ. Sci. Technol. 22: 1493-1498.
Eckhardt, S.; Stohl, A.; Beirle, S.; Spichtinger, N.; James, P.; Forster, C.; Junker, C.; Wagner, T.; Platt, U.; Jennings,
S. G. (2003) The North Atlantic oscillation controls air pollution transport to the Arctic. Atmos. Chem. Phys.
3: 1769-1778.
Eder, B. K.; Davis, J. M.; Bloomfield, P. (1993) A characterization of the spatiotemporal variability of non-urban
ozone concentrations over the Eastern United States. Atmos. Environ. 27A: 2645-2668.
Ehhalt, D. H.; Dorn, H.-P.; Poppe, D. (1991) The chemistry of the hydroxyl radical in the troposphere. In: Last,
F. T.; Watling, R., eds. Acidic deposition: its nature and impacts. Proceedings of the international
symposium; September 1990; Glasgow, United Kingdom. Proc. R. Soc. Edinburgh, Sect. B: Biol. Sci.
97: 17-34.
Eisele, F. L.; Mount, G. H.; Tanner, D.; Jefferson, A.; Shelter, R.; Harder, J. W.; Williams, E. J. (1997)
Understanding the production and interconversion of the hydroxyl radical during the tropospheric OH
photochemistry experiment. J. Geophys. Res. 102: 6457-6465.
AX2-166
-------
Emmons, L. K.; Hauglustaine, D. A.; Mtiller, J.-F.; Carroll, M. A.; Brasseur, G. P.; Brunner, D.; Staehelin, J.;
Thouret, V.; Marenco, A. (2000) Data composites of airborne observations of tropospheric ozone and its
precursors. J. Geophys. Res. [Atmos.] 105: 20,497-20,538.
Erickson, D. J.; Seuzaret, C.; Keene, W. C.; Gong, S.-L. (1999) A general circulation model based calculation of
HC1 and CINO2 production from sea-salt dechlorination: reactive chlorine emissions inventory. J. Geophys.
Res. [Atmos.] 104: 8347-8372.
Fast, J. D.; Doran, J.; Shaw, W. J.; Coulter, R. L.; Martin, T. J. (2000) The evolution of the boundary layer and its
effect on air chemistry in the Phoenix area. J. Geophys. Res. [Atmos.] 105: 22,833-22,848.
Fast, J. D.; Zaveri, R. A.; Bian, R. X.; Chapman, E. G.; Easter, R. C. (2002) Effect of regional-scale transport on
oxidants in the vicinity of Philadelphia during the 1999 NE-OPS Field Campaign. J. Geophys. Res. [Atmos.]
107(D16): 10.1029/2001JD000980.
Faust, B. C. (1994) A review of the photochemical redox reactions of iron(III) species in atmospheric, oceanic, and
surface waters: influences on geochemical cycles and oxidant formation. In: Helz, G. R.; Zepp, R. G.; Crosby,
D. G., eds. Aquatic and Surface Photochemistry. Boca Raton, FL: Lewis Publishers; pp. 3-37.
Faust, B. E.; Allen, J. M.; Anastasio, C. (1992) Aqueous-phase photochemical formation of peroxides peroxyl
radicals and singlet molecular oxygen in authentic cloud waters [abstract]. Presented at: 203rd national
meeting of the American Chemical Society; April; San Francisco, CA. Abstr. Pap. Am. Chem. Soc. 203:
ENVR94.
Faust, B. C.; Anastasio, C.; Allen, J. M.; Arakaki, T. (1993) Aqueous-phase photochemical formation of peroxides
in authentic cloud and fog waters. Science 260: 73-75.
Fehsenfeld, F. C.; Dickerson, R. R.; Htibler, G.; Luke, W. T.; Nunnermacker, L. J.; Williams, E. J.; Roberts, J. M.;
Calvert, J. G.; Curran, C. M.; Delany, A. C.; Eubank, C. S.; Fahey, D. W.; Fried, A.; Gandrud, B. W.;
Langford, A. O.; Murphy, P. C.; Norton, R. B.; Pickering, K. E.; Ridley, B. A. (1987) A ground-based
intercomparison of NO, NOX, and NOy measurement techniques. J. Geophys. Res. [Atmos.] 92:
14,710-14,722.
Fehsenfeld, F. C.; Trainer, M.; Parrish, D. D.; Volz-Thomas, A.; Penkett, S. (1996a) North Atlantic Regional
Experiment (NARE) 1993 summer intensive: foreword. J. Geophys. Res. [Atmos.] 101: 28,869-28,875.
Fehsenfeld, F. C.; Damn, P.; Leaitch, W. R.; Trainer, M.; Parrish, D. D.; Htibler, G. (1996b) Transport and
processing of O3 and O3 precursors over the North Atlantic: an overview of the 1993 North Atlantic Regional
Experiment (NARE) summer intensive. J. Geophys. Res. [Atmos.] 101(D22): 28877-28891.
Fehsenfeld, F. C.; Huey, L. G.; Sueper, D. T.; Norton, R. B.; Williams, E. J.; Eisele, F. L.; Mauldin, R. L., Ill;
Tanner, D. J. (1998) Ground-based intercomparison of nitric acid measurement techniques. J. Geophys. Res.
[Atmos.] 103(03): 3343-3353.
Pick, J.; Pommer, L.; Nilsson, C.; Andersson, B. (2003) Effect of OH radicals, relative humidity, and time on the
composition of the products formed in the ozonolysis of a-pinene. Atmos. Environ. 37: 4087-4096.
Finlayson-Pitts, B. J. (1993) Comment on "Indications of photochemical histories of Pacific air masses from
measurements of atmospheric trace species at Point Arena, California" by D. D. Parrish et al. J. Geophys.
Res. [Atmos.] 98(D8): 14,991-14,993.
Finlayson-Pitts, B. J.; Pitts, J. N., Jr. (2000) Chemistry of the upper and lower atmosphere theory, experiments and
applications. San Diego, CA: Academic Press.
Fiore, A. M.; Jacob, D. J.; Bey, L; Yantosca, R. M.; Field, B. D.; Fusco, A. C.; Wilkinson, J. G. (2002) Background
ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys.
Res. (Atmos.) 107(D15): 10.1029/2001JD000982.
Fiore, A. M.; Jacob, D. J.; Mathur, R.; Martin, R. V. (2003) Application of empirical orthogonal functions to
evaluate ozone simulations with regional and global models. J. Geophys. Res. (Atmos.) 108(D14):
10.1029/2002JD003151.
Fortuin, J. P. F.; Kelder, H. (1998) An ozone climatology based on ozonesonde and satellite measurements.
J. Geophys. Res. 103: 31,709-31,734.
Fox-Rabinovitz, M. S.; Takacs, L. L.; Govindaraju, R. C. (2002) A variable-resolution stretched-grid general
circulation model and data assimilation system with multiple areas of interest: studying the anomalous
regional climate events of 1998. J. Geophys. Res. [Atmos.] 107(D24): 10.1029/2002JD002177.
Fried, A.; Hodgeson, J. (1982) Laser photoacoustic detection of nitrogen dioxide in the gas-phase titration of nitric
oxide with ozone. Anal. Chem. 54: 278-282.
Friedlander, S. K.; Yeh, E. K. (1998) The submicron atmospheric aerosol as a carrier of reactive chemical species:
case of peroxides. Appl. Occup. Environ. Hyg. 13: 416-420.
AX2-167
-------
Fuentes, M; Raftery, A. E. (2005) Model evaluation and spatial interpolation by Bayesian combination of
observations with outputs from numerical models. Biometrics 61: 36-45.
Fuglestvedt, J. S.; Berntsen, T. K.; Isaksen, I. S. A.; 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.
Fujita, E. M.; Lu, Z.; Sheetz, L.; Harshfield, G.; Zielinska, B. (1997) Determination of mobile source emission
source fraction using ambient field measurements. Final report. Reno, NV: Desert Research Institute;
prepared for Coordinating Research Council, Inc. CRC project no. E5-1.
Furutani, H.; Akimoto, H. (2002) Development and characterization of a fast measurement system for gas-phase
nitric acid with a chemical ionization mass spectrometer in the marine boundary layer. J. Geophys. Res.
(Atmos.) 107(D1-D2): 10.1029/2000JD000269.
Gaffney, J. S.; Marley, N. A.; Drayton, P. J.; Doskey, P. V.; Kotamarthi, V. R.; Cunningham, M. M.; Baird, J. C.;
Dintaman, J.; Hart, H. L. (2002) Field observations of regional and urban impacts on NO2, ozone, UVB, and
nitrate radical production rates in the Phoenix air basin. Atmos. Environ. 36: 825-833.
Galbally, I. E.; Bentley, S. T.; Meyer, C. P. (2000) Mid-latitude marine boundary-layer ozone destruction at visible
sunrise observed at Cape Grim, Tasmania, 41°S. Geophys. Res. Lett. 27: 3841-3844.
Garnica, R. M.; Appel, M. F.; Eagan, L. (2000) A REMPI method for the ultrasensitive detection of NO and NO2
using atmospheric pressure laser ionization mass spectrometry. Anal. Chem. 72: 5639-5646.
Geiger, H.; Barnes, I.; Bejan, I.; Benter, T.; Spittler, M. (2003) The tropospheric degradation of isoprene: an updated
module for the regional atmospheric chemistry mechanism. Atmos. Environ. 37: 1503-1519.
Geron, C. D.; Guenther, A. B.; Pierce, T. E. (1994) An improved model for estimating emissions of volatile organic
compounds from forests in the eastern United States. J. Geophys. Res. [Atmos.] 99: 12,773-12,791.
Geron, C. D.; Nie, D.; Arnts, R. R.; Sharkey, T. D.; Singsaas, E. L.; Vanderveer, P. J.; Guenther, A.; Sickles, J. E.,
II; Kleindienst, T. E. (1997) Biogenic isoprene emission: model evaluation in a southeastern United States
bottomland deciduous forest. J. Geophys. Res. [Atmos.] 102(D15): 18,889-18,901.
Geron, C.; Guenther, A.; Sharkey, T.; Arnts, R.R. (2000a) Temporal variability in basal isoprene emission factor.
Tree Physiol. 20: 799-805.
Geron, C.; Rasmussen, R.; Arnts, R. R.; Guenther, A. (2000b) A review and synthesis of monoterpene speciation
from forests in the United States. Atmos. Environ. 34: 1761-1781.
Geron, C. D.; Harley, P.; Guenther, A. (2001) Isoprene emission capacity for US tree species. Atmos. Environ.
35: 3341-3352.
Gershey, R. M. (1983) Characterization of seawater organic matter carried by bubble-generated aerosols. Limnol.
Oceanogr. 28:309-319.
Gery, M. W.; Fox, D. L.; Jeffries, H. E.; Stockburger, L.; Weathers, W. S. (1985) A continuous stirred tank reactor
investigation of the gas-phase reaction of hydroxyl radicals and toluene. Int. J. Chem. Kinet. 17: 931-955.
Gery, M. W.; Whitten, G. Z.; Killus, J. P.; Dodge, M. C. (1989) A photochemical kinetics mechanism for urban and
regional scale computer modeling. J. Geophys. Res. [Atmos.] 94: 12,925-12,956.
Gettelman, A.; Holton, J. R.; Rosenlof, K. H. (1997) Mass fluxes of O3, CH4, N2O, and CF2C12 in the lower
stratosphere calculated from observational data. J. Geophys. Res. 102(D15): 19,149-19,159.
Geyer, A.; Alicke, B.; Konrad, S.; Schmitz, T.; Stutz, J.; Platt, U. (2001) Chemistry and oxidation capacity of the
nitrate radical in the continental boundary layer near Berlin. J. Geophys. Res. (Atmos.) 106: 8013-8025.
Geyer, A.; Platt, U. (2002) Temperature dependence of the NO3 loss frequency: A new indicator for the contribution
of NO3 to the oxidation of monoterpenes and NOX removal in the atmosphere. J. Geophys. Res. (Atmos.)
107(D20): 10.1029/2001JD001215.
Gidel, L. T. (1983) Cumulus cloud transport of transient tracers. J. Geophys. Res. [Atmos.] 88: 6587-6599.
Gillani, N. V.; Meagher, J. F.; Valente, R. J.; Imhoff, R. E.; Tanner, R. L.; Luria, M. (1998) Relative production of
ozone and nitrates in urban and rural power plant plumes. 1. Composite results based on data from 10 field
measurement days. J. Geophys. Res. [Atmos.] 103: 22,593-22,615.
Goldan, P. D.; Kuster, W. C.; Albritton, D. L.; Fehsenfeld, F. C.; Connell, P. S.; Norton, R. B.; Huebert, B. J. (1983)
Calibration and tests of the filter-collection method for measuring clean-air, ambient levels of nitric acid.
Atmos. Environ. 17: 1355-1364.
Goldan, P. D.; Trainer, M.; Kuster, W. C.; Parrish, D. D.; Carpenter, J.; Roberts, J. M.; Yee, J. E.; Fehsenfeld, F. C.
(1995) Measurements of hydrocarbons, oxygenated hydrocarbons, carbon monoxide, and nitrogen oxides in
an urban basin in Colorado: implications for emission inventories. J. Geophys. Res. [Atmos.] 100:
22,771-22,783.
AX2-168
-------
Goldan, P. D.; Kuster, W. C.; Fehsenfeld, F. C. (1997) Non-methane hydrocarbon measurements during the
tropospheric OH photochemistry experiment. J. Geophys. Res. (Atmos.) 102: 6315-6324.
Goldan, P. D.; Parrish, D. D.; Kuster, W. C.; Trainer, M; McKeen, S. A.; Holloway, J.; Jobson, B. T.; Sueper, D. T.;
Fehsenfeld, F. C. (2000) Airborne measurements of isoprene, CO, and anthropogenic hydrocarbons and their
implications. J. Geophys. Res. [Atmos.] 105: 9091-9105.
Golden, D. M.; Smith, G. P. (2000) Reaction of OH + NO2 + M: a new view. J. Phys. Chem. A 104: 3991-3997.
Goldstein, A. H.; Schade, G. W. (2000) Quantifying biogenic and anthropogenic contributions to acetone mixing
ratiosin a rural environment. Atmos. Environ. 34: 4997-5006.
Gong, S. L.; Barrie, L. A.; Blanchet, J.-P. (1997) Modeling sea-salt aerosols in the environment: 1. model
development. J. Geophys. Res. 102: 3805-3818.
Gong, W.; Mickle, R. E.; Bottenheim, F. F.; Beauchamp, S.; Waugh, D. (2000) Marine/coastal boundary layer and
vertical structure of ozone observed at a coastal site in Nova Scotia during the 1996 NARSTO-CE field
campaign. Atmos. Environ. 34: 4139-4154.
Greenhut, G. K. (1986) Transport of ozone between boundary layer and cloud layer by cumulus clouds. J. Geophys.
Res. [Atmos.] 91: 8613-8622.
Greenhut, G. K.; Ching, J. K. S.; Pearson, R., Jr.; Repoff, T. P. (1984) Transport of ozone by turbulence and clouds
in an urban boundary layer. J. Geophys. Res. [Atmos.] 89: 4757-4766.
Gregory, G. L.; Hoell, J. M., Jr.; Torres, A. L.; Carroll, M. A.; Ridley, B. A.; Rodgers, M. O.; Bradshaw, J.;
Sandholm, S.; Davis, D. D. (1990) An intercomparison of airborne nitric oxide measurements: a second
opportunity. J. Geophys. Res. [Atmos.] 95: 10,129-10,138.
Grell, G. A.; Dudhia, J.; Stauffer, D. R. (1994) Description of the fifth-generation Perm State/NCAR mesoscale
model (MM5). Boulder, CO: National Center for Atmospheric Research, Mesoscale and Microscale
Meteorology Division; report no. NCAR/TN-398+STR. Available from: NTIS, Springfield, VA;
PB95-206348.
Grell, G. A.; Emeis, S.; Stockwell, W. R.; 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.
Grobner, J.; Kerr, J. B. (2001) Ground-based determination of the spectral ultraviolet extraterrestrial solar irradiance:
providing a link between space-based and ground-based solar UV measurements. J. Geophys. Res. (Atmos.)
106(07): 7211-7217.
Grosjean, E.; Grosjean, D. (1994) Rate constants for the gas-phase reactions of ozone with unsaturated aliphatic
alcohols. Int. J. Chem. Kinetics 26: 1185-1191.
Gross, A.; Stockwell, W. R. (2003) Comparison of the EMEP, RADM2 and RACM Mechanisms. J. Atmos. Chem.
44: 151-170.
Guenther, A.; Hewitt, C. N.; Erickson, D.; Fall, R.; Geron, C.; Graedel, T.; Harley, P.; Klinger, L.; Lerdau, M.;
McKay, W. A.; Pierce, T.; Scholes, B.; Steinbrecher, R.; Tallamraju, R.; Taylor, J.; Zimmerman, P. (1995) A
global model of natural volatile organic compound emissions J. Geophys. Res. (Atmos.) 100(D5): 8873-8892.
Guenther, F. R.; Dorko, W. D.; Miller, W. R.; Rhoderick, G. C. (1996) The NIST traceable reference material
program for gas standards. Washington, DC: U.S. Department of Commerce, National Institute of Standards
and Technology; NIST special publication 260-126.
Guenther, A.; Geron, C.; Pierce, T.; Lamb, B.; Harley, 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.
Hadjinicolaou, P.; Pyle, J. A. (2004) The impact of Arctic ozone depletion on northern middle latitudes: interannual
variability and dynamical control. J. Atmos. Chem. 47: 25-43.
Hallock-Waters, K. A.; Doddridge, B. G.; Dickerson, R. R.; Spitzer, S.; Ray, J. D. (1999) Carbon monoxide in the
U.S. mid-Atlantic troposphere: evidence for a decreasing trend. Geophys. Res. Lett. 26: 2861-2864.
Hallquist, M.; Stewart, D. J.; Stephenson, S. K.; Cox, R. A. (2003) Hydrolysis of N2O5 on sub-micron sulfate
aerosols. Phys. Chem. Chem. Phys. 5: 3453-3463.
Hameed, S.; Pinto, J. P.; Stewart, R. W. (1979) Sensitivity of the predicted CO-OH-CH4 perturbation to tropospheric
NO-j concentrations. J. Geophys. Res. C: Oceans Atmos. 84: 763-768.
Hanisch, F.; Crowley, J. N. (2003) Ozone decomposition on Saharan dust: an experimental investigation. Atmos.
Chem. Phys. 3: 119-130.
AX2-169
-------
Hanna, A. F.; Mathur, R.; Pinto, I; Jang, C. (2001) Seasonal modeling of the export of pollutants from North
America using the Multi Scale Air Quality Simulation Pattern (MAQSIP). Presented at: Workshop on
photooxidants, particles, and haze across the Arctic and North Atlantic: observations and models; June; New
York, NY. Palisades, NY: Center for International Earth Science Information Network (CIESIN). Available:
http://www.ciesin.org/pph/uploads/H_A_Seasonal_Modeling_PDF.pdf (29 December 2003).
Hannegan, B.; Olsen, S.; Prather, M; Zhu, X.; Rind, D.; Lerner, J. (1998) The dry stratosphere: a limit on cometary
water influx. Geophys. Res. Lett. 25: 1649-1652.
Hansel, A.; Jordan, A.; Taucher, J. (1995a) Analysis of volatile organic compounds using
proton-transfer-reaction-mass spectrometry (PTR-MS) in the ppb range. In: Becker, K. H.; Carr, W. E.;
Kunhardt, E. E., eds. Proceedings of the XXII International conference on phenomena in ionized gases;
July-August; Hoboken, NJ. Hoboken, NJ: Stevens Institute of Technology; pp. 181-182.
Hansel, A.; Jordan, A.; Holzinger, R.; Prazeller, P.; Vogel, W.; Lindinger, W. (1995b) Proton transfer reaction mass
spectrometry: on-line trace gas analysis at the ppb level. Int. J. Mass Spectrom. lonProc. 149/150: 609-619.
Hansen, K.; Draaijers, G. P. J.; Ivens, W. P. M. F.; Gundersen, P.; vanLeeuwen, N. F. M. (1994) Concentration
variations in rain and canopy throughfall collected sequentially during individual rain events. Atmos. Environ.
28: 3195-3205.
Hansen, J.; Sato, M.; Ruedy, R.; Lacis, A.; Asamoah, K.; Beckford, K.; Borenstein, S.; Brown, E.; Cairns, B.;
Carlson, B.; Curran, B.; de Castro, S.; Druyan, L.; Etwarrow, P.; Ferede, T.; Fox, M.; Gaffen, D.; Glascoe, J.;
Gordon, H.; Hollandsworth, S.; Jiang, X.; Johnson, C.; Lawrence, N.; Lean, J.; Lerner, J.; Lo, K.; Logan, J.;
Luckett, A.; McCormick, M. P.; McPeters, R.; Miller, R.; Minnis, P.; Ramberran, L; Russell, G.; Russell, P.;
Stone, P.; Tegen, I.; Thomas, S.; Thomason, L.; Thompson, A.; et al. (1997) Forcings and chaos in
interannual to decadal climate change. J. Geophys. Res. [Atmos.] 102: 25,679-25,720.
Harder, J. W.; Brault, J. W.; Johnston, P. V.; Mount, G. H. (1997a) Temperature dependent NO2 cross sections at
high spectral resolution. J. Geophys. Res. [Atmos.] 102: 3861-3879.
Harder, J. W.; Williams, E. J.; Baumann, K.; Fehsenfeld, F. C. (1997b) Ground-based comparison of NO2, H2O, and
O3 measured by long-path and in situ techniques during the 1993 Tropospheric OH Photochemistry
Experiment. J. Geophys. Res. 102: 6227-6243.
Harley, P.; Guenther, A.; Zimmerman, P. (1996) Effects of light, temperature and canopy position on net
photosynthesis and isoprene emission from sweetgum (Liquidambar styraciflua) leaves. Tree Physiol.
16: 25-32.
Harris, G. W.; Carter, W. P. L.; Winer, A. M.; Pitts, J. N, Jr.; Platt, U.; Perner, D. (1982) Observations of nitrous
acid in the Los Angeles atmosphere and implications for predictions of ozone-precursor relationships.
Environ. Sci. Technol. 16: 414-419.
Harrison, R. M.; Peak, J. D.; Collins, G. M. (1996) Tropospheric cycle of nitrous acid. J. Geophys. Res. (Atmos.)
101: 14,429-14,439.
Hauglustaine, D. A.; Brasseur, G. P.; Walters, S.; Rasch, P. J.; Mtiller, J.-F.; Emmons, L. K.; Carroll, M. A. (1998)
MOZART, a global chemical transport model for ozone and related chemical tracers 2. model results and
evaluation. J. Geophys. Res. (Atmos.) 103: 28291-28335.
Hebestreit, K.; Stutz, J.; Rosen, D.; Matveiv, V.; Peleg, M.; Luria, M.; Platt, U. (1999) DOAS measurements of
tropospheric bromine oxide in mid-latitudes. Science 283: 55-57.
Hegg, D. A.; Livingston, J.; Hobbs, P. V.; Novakov, T.; Russell, P. (1997) Chemical apportionment of aerosol
column optical depth off the mid-Atlantic coast of the United States. J. Geophys. Res. (Atmos.) 102:
25,293-25,303.
Heikes, B. G.; Chang, W.; Pilson, M. E. Q.; Swift, E.; Singh, H. B.; Guenther, A.; Jacob, D. J.; Field, B. D.; Fall, R.;
Riemer, D.; Brand, L. (2002) Atmospheric methanol budget and ocean implication. Glob. Biogeochem.
Cycles 16: 10.1029/2002GB001895.
Hering, S. V.; Lawson, D. R.; Allegrini, I.; Febo, A.; Perrino, C.; Possanzini, M.; Sickles, J. E., II; Anlauf, K. G.;
Wiebe, A.; Appel, B. R.; John, W.; Ondo, J.; Wall, S.; Braman, R. S.; Sutton, R.; Cass, G. R.; Solomon,
P. A.; Eatough, D. J.; Eatough, N. L.; Ellis, E. C.; Grosjean, D.; Hicks, B. B.; Womack, J. D.; Horrocks, J.;
Knapp, K. T.; Ellestad, T. G.; Paur, R. J.; Mitchell, W. J.; Pleasant, M.; Peake, E.; MacLean, A.; Pierson,
W. R.; Brachaczek, W.; Schiff, H. I.; Mackay, G. I.; Spicer, C. W.; (1988) The nitric acid Shootout: field
comparison of measurement methods. Atmos. Environ. 22: 1519-1539.
Hess, P. G. (2001) Model and measurement analysis of springtime transport and chemistry of the Pacific Basin.
J. Geophys. Res. (Atmos.) 106: 12,689-12,717.
AX2-170
-------
Hilsenrath, E.; Attmannspacher, W.; Bass, A.; Evans, W.; Hagemeyer, R.; Barnes, R. A.; Komhyr, W.;
Mauersberger, K.; Mentall, J.; Proffitt, M; Robbins, D.; Taylor, S.; Torres, A.; Weinstock, E. (1986) Results
from the ballon ozonr intercomparison campaign (BOIC). J. Geophys. Res. 91: 13,137-13,152.
Hirsch, A. I.; Munger, J. W.; Jacob, D. J.; Horowitz, L. W.; Goldstein, A. H. (1996) Seasonal variation of the ozone
production efficiency per unit NOX at Harvard Forest, Massachusetts. J. Geophys. Res. [Atmos.]
101: 12,659-12,666.
Hobbs, P. V.; Locatelli, J. D.; Martin, J. E. (1996) A new conceptual model for cyclones generated in the lee of the
Rocky Mountains. Bull. Am. Meteorol. Soc. 77: 1169-1178.
Hoell, J. M.; Davis, D. D.; Jacob, D. J.; Rodgers, M. O.; Newell, R. E.; Fuelberg, H. E.; McNeal, R. J.; Raper, J. L.;
Bendura, R. J. (1996) Pacific exploratory mission in the tropical Pacific: PEM-Tropics A, August-September
1996. J. Geophys. Res. (Atmos.) 104: 5567-5583.
Hoell, J. M.; Davis, D. D.; Liu, S. C.; Newell, R. E.; Akimoto, H.; McNeal, R. J.; Bendura, R. J. (1997) The Pacific
exploratory mission A west phase B: February-March, 1994. J. Geophys. Res. [Atmos.] 102(D23):
28223-28239.
Hoffman, E. J.; Duce, R. A. (1976) Factors influencing the organic carbon content of marine aerosols: a laboratory
study. J. Geophys. Res. (Atmos.) 81: 3667-3670.
Hoffman, E. J.; Duce, R. A. (1977) Organic carbon in marine atmospheric paniculate matter: concentration and
particle size distribution. Geophys. Res. Lett. 4: 449-452.
Hoigne, J.; Zuo, Y.; Nowell, L. (1994) Photochemical reactions in atmospheric waters: role of dissolved iron
species. In: Helz, G. R.; Zepp, R. G.; Crosby, D. C. Aquatic and Surface Photochemistry. Boca Raton, FL:
Lewis Publishers.
Holton, J. R.; Haynes, P. H.; Mclntyre, M. E.; Douglass, A. R.; Rood, R. B.; Pfister, L. (1995)
Stratosphere-troposphere exchange. Rev. Geophys. 33: 403-439.
Honninger, G. (1999) Referenzspektren reaktiver halogenverbindungen fur DOAS-messungen. Heidelberg,
Germany: Ruprecht Karls University.
Honrath, R. E.; Guo, S.; Peterson, M. C. (2000) Photochemical production of gas phase NOX from ice crystal NO3".
J. Geophys. Res. (Atmos.) 105: 24,183-24,190.
Horowitz, L. W.; Liang, J.; Gardner, G. M.; Jacob, D. J . (1998) Export of reactive nitrogen from North America
during summertime: sensitivity to hydrocarbon chemistry. J. Geophys. Res. 103(D11): 13451-13476.
Hoskins, B. J. (1972) Non-boussinesq effects and further development in a model of upper tropospheric
frontogenesis. Q. J. Roy. Meteorol. Soc. 98: 532-541.
Houweling, S.; Dentener, F.; Lelieveld, J.; Walter, B.; Dlugokencky, E. (2000) The modeling of tropospheric
methane: how well can point measurements be reproduced by a global model? J. Geophys. Res. (Atmos.)
105: 8981-9002.
Htibler, G.; Alvarez, R., II; Daum, P.; Dennis, R.; Gillani, N.; Kleinman, L.; Luke, W.; Meagher, J.; Rider, D.;
Trainer, M.; Valente, R. (1998) An overview of the airborne activities during the Southern Oxidants Study
(SOS) 1995 Nashville/Middle Tennessee ozone study. J. Geophys. Res. [Atmos.] 103: 22,245-22,259.
Hudman, R. C.; Jacob, D. J.; Cooper, O. C.; Evans, M. J.; Heald, C. L.; Park, R. J.; Fehsenfeld, F.; Flocke, F.;
Holloway, J.; Htibler, G.; Kita, K.; Koike, M.; Kondo, Y.; Neuman, A.; Nowak, J.; Oltmans, S.; Parrish, D.;
Roberts, J. M.; Ryerson, T. (2004) Ozone production in transpacific Asian pollution plumes and implications
for ozone air quality in California. J. Geophys. Res. (Atmos.) 109(D23): 10.1029/2004JD004974.
Huebert, B. J.; Zhuang, L.; Howell, S.; Noone, K.; Noone, B. (1996) Sulfate, nitrate, methanesulfonate, chloride,
ammonium, and sodium measurements from ship, island, and aircraft during the Atlantic Stratocumulus
Transition Experiment/Marine Aerosol Gas Exchange. J. Geophys. Res. (Atmos.) 101: 4413-4423.
Huey, L. G.; Dunlea, E. J.; Lovejoy, E. R.; Hanson, D. R.; Norton, R. B.; Fehsenfeld, F. C.; Howard, C. J. (1998)
Fast time response measurements of HNO3 in air with a chemical ionization mass spectrometer. J. Geophys.
Res. 103(03): 3355-3360.
Huntrieser, H.; Feigl, C.; Schlager, H.; Schroder, F.; Gerbig, C.; Van Velthoven, P.; Flatey, F.; Thery, C.; Petzold,
A.; Holler, H.; Schumann, U. (2002) Airborne measurements of NOX, tracer species, and small particles
during the European Lightning Nitrogen Oxides Experiment. J. Geophys. Res. 107(D11):
10.1029/2000JD000209.
Husar, R. B.; Renard, W. P. (1998) Ozone as a function of local wind speed and direction: Evidence of local and
regional transport. Presented at: 91st annual meeting and exhibition of the Air & Waste Management
Association; June; San Diego, CA. Pittsburgh, PA: Air & Waste Management Association; online paper no.
98-A922. Available: http://capita.wustl.edu/capita/CapitaReports/REPORTSl.HTML (13 November 2003).
AX2-171
-------
Imhoff, R. E.; Valente, R.; Meagher, J. F.; Luria, M. (1995) The production of O3 in an urban plume - airborne
sampling of the Atlanta urban plume. Atmos. Environ. 29: 2349-2358.
Intergovernmental Panel on Climate Change (IPCC). (2001) Climate change 2001: the scientific basis. Contribution
of working group I to the third assessment report of the Intergovernmental Panel on Climate Change.
Cambridge, United Kingdom: Cambridge University Press.
Jacob, D. J. (2000) Heterogeneous chemistry and tropospheric ozone. Atmos. Environ. 34: 2131-2159.
Jacob, D. J.; Logan, J. A.; Gardner, G. M.; Yevich, R. M.; Spivakovsky, C. M.; Wofsy, S. C.; Sillman, S.; Prather,
M. J. (1993) Factors regulating ozone over the United States and its export to the global atmosphere.
J. Geophys. Res. [Atmos.] 98: 14,817-14,826.
Jacob, D. J.; Horowitz, L. W.; Munger, J. W.; Heikes, B. G.; Dickerson, R. R.; Artz, R. S.; Keene, W. C. (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, D. J.; Prather, M. J.; Rasch, P. J.; Shia, R.-L.; Balkanski, Y. J.; Beagley, S. R.; Bergman, D. J.; Blackshear,
W. T.; Brown, M.; Chiba, M.; Chipperfield, M.; De Grandpre, J.; Dignon, J. E.; Feichter, J.; Genthon, C.;
Grose, W.L.; Kasibhatla, P. S.; Kohler, L; Kritz, M. A.; Law, K.; Penner, J. E.; Ramonet, M.; Reeves, C. E.;
Rotman, D. A.; Stockwell, D. Z.; VanVelthoven, P. F. J.; Verver, G.; Wild, O.; Yang, H.; Zimmermann, P.
(1997) Evaluation and intercomparison of global atmospheric transport models using 222Rn and other
short-lived tracers. J. Geophys. Res. [Atmos.] 102: 5953-5970.
Jacobson, M. Z. (1998) Studying the effects of aerosols on vertical photolysis rate coefficient and temperature
profiles over an urban airshed. J. Geophys. Res. [Atmos.] 103: 10,593-10,604.
Jacobson, M. Z. (1999) Isolating nitrated and aromatic aerosols and nitrated aromatic gases as sources of ultraviolet
light absorption. J. Geophys. Res. (Atmos.) 104: 3527-3542.
Jacobson, M. Z.; Lu, R.; Turco, R. P.; Toon, O. P. (1996) Development and application of a new air pollution
modeling system-Part I: gas-phase simulations. Atmos. Environ. 30: 1939-1963.
Jaegle, L.; Jacob, D. J.; Wennberg, P. O.; Spivakovsky, C. M.; Hanisco, T. F.; Lanzendorf, E. J.; Hintsa, E. J.;
Fahey, D. W.; Keim, E. R.; Proffitt, M. H.; Atlas, E. L.; Flocke, F.; Schauffler, S.; McElroy, C. T.;
Midwinter, C.; Pfister, L.; Wilson, J. C. (1997) Observed OH and HO2 in the upper troposphere suggests a
major source from convective injection of peroxides. Geophys. Res. Lett. 24: 3181-3184.
Jaegle, L.; Jacob, D. J.; Brune, W. H.; Faloona, I. C.; Tan, D.; Kondo, Y.; Sachse, G. W.; Anderson, B.; Gregory, G.
L.; Vay, S.; Singh, H. B.; Blake, D. R.; Shelter, R. (1999) Ozone production in the upper troposphere and the
influence of aircraft during SONEX: approach of NOx-saturated conditions. Geophys. Res. Lett. 26:
3081-3084.
Jaegle, L.; Jacob, D. J.; Brune, W. H.; Wennberg, P. O. (2001) Chemistry of HOX radicals in the upper troposphere.
Atmos. Environ. 35: 469-489.
James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003a) A 15-year climatology of
stratosphere-troposphere exchange with a Lagrangian particle dispersion model: 1. Methodology and
validation. J. Geophys. Res. 108(D12): 10.1029/2002JD002637.
James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003b) A 15-year climatology of
stratosphere-troposphere exchange with a Lagrangian particle dispersion model. 2. Mean climate and seasonal
variability. J. Geophys. Res. (Atmos.) 108(D12): 10.1029/2002JD002639.
Jankowski, J. J.; Kieber, D. J.; Mopper, K.; Neale, P. J. (2000) Development and intercalibration of ultraviolet solar
actinometers. Photochem. Photobiol. 71: 431-440.
Jenkin, M. E.; Cox, R. A.; Williams, D. J. (1988) Laboratory studies of the kinetics of formation of nitrous acid from
the thermal reaction of nitrogen dioxide and water vapour. Atmos. Environ. 22: 487-498.
Jenkin, M. E.; Saunders, S. M.; Pilling, M. J. (1997) The tropospheric degradation of volatile organic compounds: a
protocol for mechanism development. Atmos. Environ. 31: 81-104.
Jet Propulsion Laboratory. (2003) Chemical kinetics and photochemical data for use in atmospheric studies.
Pasadena, CA: California Institute of Technology; JPL publication no. 02-25. Available:
http://jpldataeval.jpl.nasa.gov/pdf/JPL_02-25_l_intro_rev0.pdf (18 December 2003).
Jeuken, A. B. M.; Eskes, H. J.; Van Velthoven, P. F. J.; Kelder, H. M.; Holm, E. V. (1999) Assimilation of total
ozone satellite measurements in a three-dimensional tracer transport model. J. Geophys. Res. (Atmos.)
104: 5551-5563.
Jimenez, P.; Baldasano, J. M.; Dabdub, D. (2003) Comparison of photochemical mechanisms for air quality
modeling. Atmos. Environ. 37: 4179-4194.
AX2-172
-------
Jobson, B. T.; Berkowitz, C. M; Kuster, W. C; Goldan, P. D.; Williams, E. J.; Fehsenfeld, F. C.; Apel, E. C; Karl,
T.; Lonneman, W. A.; Riemer, D. (2004) Hydrocarbon source signatures in Houston, Texas: influence of the
petrochemical industry. J. Geophys. Res. 109: D24305: 10.1029/2004JD004887.
Johnson, C. E.; Collins, W. J.; Stevenson, D. S.; Derwent, R. G. (1999) The relative roles of climate and emissions
changes on future oxidant concentrations. J. Geophys. Res. (Atmos.) 104: 18,631-18,645.
Johnson, B. J.; Oltmans, S. J.; Vomel, H.; Smit, H. G. J.; Deshler, T.; Kroger, C. (2002) Electrochemical
concentration cell (ECC) ozonesonde pump efficiency measurements and tests on the sensitivity to ozone of
buffered and unbuffered ECC sensor cathode solutions. J. Geophys. Res. 107(D19): 10.1029/2001JD000557.
Johnson, D.; Jenkin, M. E.; Wirtz, K.; Martin-Riviejo, M. (2004) Simulating the formation of secondary organic
aerosol from the photooxidation of toluene. Environ. Chem. 1: 150-165.
Jordan, C. E.; Dibb, J. E.; Finkel, R. C. (2003) 10Be/7Be tracer of atmospheric transport and stratosphere-troposphere
exchange. J. Geophys. Res. 108(D8): 10.1029/2002JD002395.
Junge, C. E. (1963) Air chemistry and radioactivity. New York, NY: Academic Press. (Van Mieghem, J.; Hales,
A. L., eds. International geophysics series: v. 4).
Kain, J. S.; Fritsch, J. M. (1993) Convective parameterization in mesoscale models: the Kain-Fritsch scheme. In:
Emanuel, K. A.; Raymond, D. J., eds. The Representation of Cumulus Convection in Numerical Models.
Boston, MA: American Meteorological Society; pp. 165-170. (Meteorological Monographs, v. 24, no. 46).
Kane, S. M.; Caloz, F.; Leu, M.-T. (2001) Heterogeneous uptake of gaseous N2O5 by (NH4)2SO4 NH4HSO4, and
H2SO4 aerosols. J. Phys. Chem. A 105: 6465-6470.
Kasibhatla, P.; Chameides, W. L. (2000) Seasonal modeling of regional ozone pollution in the eastern United States.
Geophys. Res. Lett. 27: 1415-1418.
Kasting, J. F.; Singh, H. B. (1986) Nonmethane hydrocarbons in the troposphere: impact on the odd hydrogen and
odd nitrogen chemistry, J. Geophys. Res. 91: 13,239-13,256.
Kavouras, I. G.; Mihalopoulus, N.; Stephanou, E. G. (1999) Secondary organic aerosol formation vs primary organic
aerosol emission: in situ evidence for the chemical coupling between monoterpene acidic photooxidation
products and new particle formation over forests. Environ. Sci. Technol. 33: 1028-1037.
Kawamura, K.; Semere, R.; Imai, Y.; Fuji, Y.; Hayashi, M. (1996) Water soluble dicarboxylic acids and related
compounds in Antarctic aerosols. J. Geophys. Res. 101: 18,721-18,728.
Keene, W. C.; Savoie, D. L. (1998) The pH of deliquesced sea-salt aerosol in polluted marine air. Geophys. Res.
Lett. 25: 2181-2194.
Keene, W. C.; Savoie, D. L. (1999) Correction to "The pH of deliquesced sea-salt aerosol in polluted marine air."
Geophys. Res. Lett. 26: 1315-1316.
Keene, W. C.; Sander, R.; Pszenny, A. A. P.; Vogt, R.; Crutzen, P. J.; Galloway, J. N. (1998) Aerosol pH in the
marine boundary layer: a review and model evaluation. J. Aerosol Sci. 29: 339-356.
Keene, W. C.; Pszenny, A. A. P.; Maben, J. R.; Sander, R. (2002) Variation of marine aerosol acidity with particle
size. Geophys. Res. Lett. 29(7): 10.1029/2001GL013881.
Kelly, N. A.; Ferman, M. A.; Wolff, G. T. (1986) The chemical and meteorological conditions associated with high
and low ozone concentrations in southeastern Michigan and nearby areas of Ontario. J. Air Pollut. Control
Assoc. 36: 150-158.
Kenley, R. A.; Davenport, J. E.; Hendry, D. G. (1981) Gas-phase hydroxyl radical reactions. Products and pathways
for the reaction of OH with aromatic hydrocarbons. J. Phys. Chem. 85: 2740-2746.
Kim, K. -H.; Kim, M. -Y. (2001) Comparison of an open path differential optical absorption spectroscopy system and
a conventional in situ monitoring system on the basis of long-term measurements of SO2, NO2, and O3.
Atmos. Environ. 35: 4059-4072.
King, D. W.; Lounsbury, H. A.; Millero, F. J. (1995) Rates and mechanisms of Fe(II) oxidation at nonmolar total
iron concentrations. Environ. Sci. Technol. 29: 818-824.
King, M. D.; Canosa-Mas, C. E.; Wayne, R. P. (2002) A structure activity relationship (SAR) for predicting rate
constants for the reaction of nitrogen dioxide (NO2) with alkenes. Phys. Chem. Chem. Phys. 4: 295-303.
Kireev, S. V.; Shnyrev, S. L.; Zhiganov, A. A. (1999) A laser fluorimeter for NO and NO2 in atmosphere. Instrum.
Exp. Tech. 42: 701-703.
Kleindienst, T. E.; Hudgens, E. E.; Smith, D. F.; McElroy, F. F.; Bufalini, J. J. (1993) Comparison of
chemiluminescence and ultraviolet ozone monitor responses in the presence of humidity and photochemical
pollutants. Air Waste 43: 213-222.
Kleinman, L. I. (2000) Ozone process insights from field experiments - part II: observation-based analysis for ozone
production. Atmos. Environ. 34: 2023-2034.
AX2-173
-------
Kleinman, L. I.; Daum, P. H.; Lee, Y.-N.; Springston, S. R.; Newman, L.; Leaitch, W. R.; Banic, C. M.; Isaac,
G. A.; MacPherson, J. I. (1996a) Measurement of O3 and related compounds over southern Nova Scotia. 1.
Vertical distributions. J. Geophys. Res. [Atmos.] 101: 29,043-29,060.
Kleinman, L. L; Daum, P. H.; Springston, S. R.; Leaitch, W. R.; Banic, C. M.; Isaac, G. A.; Jobson, T. B.; Niki, H.
(1996b) Measurement of O3 and related compounds over southern Nova Scotia. 2. Photochemical age and
vertical transport. J. Geophys. Res. [Atmos.] 101: 29,061-29,074.
Kleinman, L. I.; Daum, P. H.; Lee, J. H.; Lee, Y.-N.; Nunnermacker, S. R.; Springston, L.; Newman, J.;
Weinstein-Lloyd, J.; Sillman, S. (1997) Dependence of ozone production on NO and hydrocarbons in the
troposphere. Geophys. Res. Lett. 24: 2299-2302.
Kleinman, L. I.; Daum, P. H.; Imre, D. G.; Lee, J. H.; Lee, Y.-N.; Nunnermacker, L. J.; Springston, S. R.;
Weinstein-Lloyd, J.; Newman, L. (2000) Ozone production in the New York City urban plume. J. Geophys.
Res. [Atmos.] 105: 14,495-14,511.
Kleinman, L. L; Daum, P. H.; Lee, Y.-N.; Nunnermacker, L. J.; Springston, S. R.; Weinstein-Lloyd, J.; Rudolph, J.
(2001) Sensitivity of ozone production rate to ozone precursors. Geophys. Res. Lett. 28: 2903-2906.
Kleinman, L. L; Daum, P. H.; Imre, D.; Lee, Y.-N.; Nunnermacker, L. J.; Springston, S. R.; Weinstein-Lloyd, J.;
Rudolph, J. (2002) Ozone production rate and hydrocarbon reactivity in 5 urban areas: a cause of high ozone
concentration in Houston. Geophys. Res. Lett. 29: 10.1029/2001GL014569.
Kley, D.; Crutzen, P. J.; Smit, H. G. J.; Vomel, H.; Oltmans, S. J.; Grassl, H.; Ramanathan, V. (1996) Observations
of near-zero ozone concentrations over the convective Pacific: effects on air chemistry. Science 274: 230-233.
Klotz, B.; Barnes, I.; Becker, K. H.; Golding, B. T. (1997) Atmospheric chemistry of benzene oxide/oxepin.
J. Chem. Soc. Faraday Trans. 93: 1507-1516.
Klotz, B; Barnes, L; Becker, K. H. (1998) New results on the atmospheric photooxidation of simple alkylbenzenes.
Chem. Phys. 231:289-301.
Knake, R.; Hauser, P. C. (2002) Sensitive electrochemical detection of ozone. Anal. Chim. Acta 459: 199-207.
Knispel, R.; Koch, R.; Siese, M.; Zetzsch, C. (1990) Adduct formation of OH radicals with benzene, toluene, and
phenol and consecutive reactions of the adducts with NO^ and O2. Ber. Bunsen-Ges. Phys. Chem. 94:
1375-1379.
Krol, M.; Van Leeuwen, P. J.; Lelieveld, J. (1998) Global OH trend inferred from methylchloroform measurements.
J. Geophys. Res. [Atmos.] 103: 10,697-10,711.
Kumar, N.; Russell, A. G. (1996) Development of a computationally efficient, reactive subgrid-scale plume model
and the impact in the northeastern United States using increasing levels of chemical detail. J. Geophys. Res.
[Atmos.] 101: 16,737-16,744.
Kwok, E. S. C.; Aschmann, S. M.; Atkinson, R.; Arey, J. (1997) Products of the gas-phase reactions of o-, m- and
p-xylens with the OH radical in the presence and absence of NOX. J. Chem. Soc. Faraday Trans.
93:2847-2854.
Lammel, G.; Cape, J. N. (1996) Nitrous acid and nitrite in the atmosphere. Chem. Soc. Rev. 25: 361-369.
Lammel, G.; Perner, D. (1988) The atmospheric aerosol as a source of nitrous acid in the polluted atmosphere.
J. Aerosol Sci. 19: 1199-1202.
Langford, A. O.; Reid, S. J. (1998) Dissipation and mixing of a small-scale stratospheric intrusion in the upper
troposphere. J. Geophys. Res. 103(D23): 31,265-31,276.
Law, K. S.; Plantevin, P. H.; Shallcross, D. E.; Rogers, H. L.; Pyle, J. A.; Grouhel, C.; Thouret, V; Marenco, A.
(1998) Evaluation of modeled O3 using measurement of ozone by airbus in-service aircraft (MOZAIC) data.
J. Geophys. Res. (Atmos.) 103: 25,721-25,737.
Law, K. S.; Plantevin, P.-H.; Thouret, V.; Marenco, A.; Asman, W. A. H.; Lawrence, M.; Crutzen, P. J.; Muller,
J. F.; Hauglustaine, D. A.; Kanakidou, M. (2000) Comparison between global chemistry transport model
results and measurement of ozone and water vapor by airbus in-service aircraft (MOZAIC) data. J. Geophys.
Res. (Atmos.) 105: 1503-1525.
Lawrence, M. G.; Crutzen, P. J.; Rasch, P. J.; Eaton, B. E.; Mahowald, N. M. (1999) A model for studies of
tropospheric photochemistry: description, global distributions and evaluation. J. Geophys. Res. (Atmos.)
104: 26,245-26,277.
Lean, J. L. (2000) Short term, direct indices of solar variability. Space Science Reviews 94: 39-51.
Leibrock, E.; Huey, L. G. (2000) Ion chemistry for the detection of isoprene and other volatile organic compounds in
ambient air. Geophys. Res. Lett. 27: 1719-1722.
Lelieveld, J.; Crutzen, P. J. (1990) Influences of cloud photochemical processes on tropospheric ozone. Nature
(London) 343: 227-233.
Lelieveld, J.; Crutzen, P. J. (1991) The role of clouds in tropospheric photochemistry. J. Atmos. Chem. 12: 229-267.
AX2-174
-------
Lelieveld, I; Crutzen, P. J. (1994) Role of deep cloud convection in the ozone budget of the troposphere. Science
264: 1759-1761.
Leser, H.; Honninger, G.; Platt, U. (2003) MAX-DOAS measurements of BrO and NO2 in the marine boundary
layer. Geophys. Res. Lett. 30: 10.1029/2002GL015811.
Leston, A. R.; Ollinson, W. M; Spicer, C. W.; Satola, J. (2005) Potential interference bias in ozone standard
compliance monitoring. J. Air Waste Manage. Assoc.: in press.
Li, Q.; Jacob, D. J.; Logan, J. A.; Bey, I.; Yantosca, R. M.; Liu, H.; Martin, R. V.; Fiore, A. M.; Field, B. D.;
Duncan, B. N. (2001) A tropospheric ozone maximum over the Middle East. Geophys. Res. Lett. 28:
3235-3238.
Li, Q.; Jacob, D. J.; Fairlie, T. D.; Liu, H.; Martin, R. V.; Yantosca, R. M. (2002) Stratospheric versus pollution
influences on ozone at Bermuda: reconciling past analyses. J. Geophys. Res. 107(D22):
10.1029/2002JD002138.
Liang, J.; Horowitz, L. W.; Jacob, D. J.; Wang, Y.; Fiore, A. M.; Logan, J. A.; Gardner, G. M.; Munger, J. W. (1998)
Seasonal variations of reactive nitrogen species and ozone over the United States, and export fluxes to the
global atmosphere. J. Geophys. Res. (Atmos.) 103: 13,435-13,450.
Liao, H.; Yung, Y. L.; Seinfeld, J. H. (1999) Effects of aerosols on tropospheric photolysis rates in clear and cloudy
atmospheres. J. Geophys. Res. (Atmos.) 104: 23,697-23,707.
Lindinger, W.; Hansel, A.; Jordan, A. (1998) Proton-transfer-reaction mass spectrometry (PTR-MS): on-line
monitoring of volatile organic compounds atpptv levels. Chem. Soc. Rev. 27: 347-354.
Liu, X. H.; Hegg, D. A.; Stoelinga, M. T. (2001a) Numerical simulation of new particle formation over the
northwest Atlantic using the MM5 mesoscale model coupled with sulfur chemistry. J. Geophys. Res.
[Atmos.] 106: 9697-9715.
Liu, H.; Jacob, D. J.; Bey, I.; Yantosca, R. M. (200Ib) Constraints from 210Pb and 7Be on wet deposition and
transport in a global three-dimensional chemical tracer model driven by assimilated meteorological fields.
J. Geophys. Res. [Atmos.] 106: 12,109-12,128.
Logan, J. A. (1989) Ozone in rural areas of the United States. J. Geophys. Res. [Atmos.] 94: 8511-8532.
Logan, J. A.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B. (1981) Tropospheric chemistry: a global perspective.
J. Geophys. Res. C: Oceans Atmos. 86: 7210-7254.
Longfellow, C. A.; Ravishankara, A. R.; Hanson, D. R. (1999) Reactive uptake on hydrocarbon soot: focus on NO2.
J. Geophys. Res. (Atmos.) 104: 13,833-13,840.
Lu, C.-H.; Chang, J. S. (1998) On the indicator-based approach to assess ozone sensitivities and emissions features.
J. Geophys. Res. (Atmos.) 103: 3453-3462.
Lu, R.; Turco, R. P.; Jacobson, M. Z. (1997) An integrated air pollution modeling system for urban and regional
scales: 1. Structure and performance. J. Geophys. Res. [Atmos.] 102: 6063-6079.
Ludwig, F. L.; Reiter, E.; Shelar, E.; Johnson, W. B. (1977) The relation of oxidant levels to precursor emissions and
meteorological features: v. 1, analysis and findings. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards; report no. EPA-450/3-77-022a. Available
from: NTIS, Springfield, VA; PB-275 001.
Luke, W. T. (1997) Evaluation of a commercial pulsed fluorescence detector for the measurement of low-level SO2
concentrations during the Gas-Phase Sulfur Intercomparison Experiment. J. Geophys. Res. (Atmos.)
102: 16,255-16,265.
Luke, W. T.; Dickerson, R. R.; Ryan, W. F.; Pickering, K. E.; Nunnermacker, L. J. (1992) Tropospheric chemistry
over the lower Great Plains of the United States. 2. Trace gas profiles and distributions. J. Geophys. Res.
97(D18): 20647-20670.
Luke, W. T.; Watson, T. B.; Olszyna, K. J.; Gunter, R. L.; McMillen, R. T.; Wellman, D. L.; Wilkison, S. W. (1998)
A comparison of airborne and surface trace gas measurements during the Southern Oxidants Study (SOS).
J. Geophys. Res. [Atmos.] 103: 22,317-22,337.
Luria, M.; Valente, R. J.; Tanner, R. L.; Gillani, N. V.; Imhoff, R. E.; Mueller, S. F.; Olszyna, K. J.; meagher, J. F.
(1999) The evolution of photochemical smog in a power plant plume. Atmos. Environ. 33: 3023-3036.
Luria, M.; Tanner, R. L.; Imhoff, R. E.; Valente, R. J.; Bailey, E. M.; Mueller, S. F. (2000) Influence of natural
hydrocarbons on ozone formation in an isolated power plant plume. J. Geophys. Res. [Atmos.] 105:
9177-9188.
Madronich, S. (1987) Photodissociation in the atmosphere. 1. Actinic flux and the effects of ground reflections and
clouds. J. Geophys. Res. [Atmos.] 92: 9740-9752.
Mahlman, J. D. (1997) Dynamics of transport processes in the upper troposphere. Science 276: 1079-1083.
AX2-175
-------
Mahlman, J. D.; Pinto, J. P.; Umscheid, L. J. (1994) Transport, radiative, and dynamical effects of the Antarctic
ozone hole: a GFDL "SKYHI" model experiment. J. Atmos. Sci. 51: 489-508.
Mahrt, L. (1998) Stratified atmospheric boundary layers and breakdown of models. Theor. Comput. Fluid Dyn.
11:263-279.
Martilli, A.; Neftel, A.; Favaro, G.; Kirchner, F.; Sillman, S.; Clappier, A. (2002) Simulation of the ozone formation
in the northern part of the Po Valley. J. Geophys. Res. (Atmos.) 107(D22): 10.1029/2001JD000534.
Martin, D.; Tsivou, M; Bonsang, B.; Abonnel, C.; Carsey, T.; Springer-Young, M.; Pszenny, A.; Suhre, K. (1997)
Hydrogen peroxide in the marine atmospheric boundary layer during the Atlantic Stratocumulus Transition
Experiment/Marine Aerosol and Gas Exchange experiment in the eastern subtropical North Atlantic.
J. Geophys. Res. (Atmos.) 102: 6003-6015.
Martin, R. V.; Jacob, D. J.; Chance, K. V.; Kurosu, T. P.; Palmer, P. L; Evans, M. J. (2003) Global inventory of
nitrogen oxide emissions constrained by space-based observations of NO2 columns. J. Geophys. Res.
108(D17): 10.1029/2003 JD003453.
Martinez, M.; Harder, H.; Kovacs, T. A.; Simpas, J. B.; Bassis, J.; Lesher, R.; Brune, W. H.; Frost, G. J.; Williams,
E. J.; Stroud, C. A.; Jobson, B. T.; Roberts, J. M.; Hall, S. R.; Shelter, R. E.; Wert, B.; Fried, A.; Alicke, B.;
Stutz, J.; Young, V. L.; White, A. B.; Zamora, R. J. (2003) OH and HO2 concentrations, sources, and loss
rates during the Southern Oxidants study in Nashville, Tennessee, summer 1999. J. Geophys. Res. (Atmos.)
108(D19): 10.1029/2003 JD003551.
Marufu, L. T.; Taubman, B. F.; Bloomer, B.; Piety, C. A.; Doddridge, B. G.; Stehr, J. W.; Dickerson, R. R. (2004)
The 2003 North American electrical blackout: an accidental experiment in atmospheric chemistry. Geophys.
Res. Lett. 31(L13106): 10.1029/2004GL019771.
Massman, W. J.; Pederson, J.; Delany, A.; Grantz, D.; Denhartog, G.; Neumann, H. H.; Oncley, S. P.; Pearson, R.;
Shaw, R. H. (1994) An evaluation of the regional acid deposition model surface module for ozone uptake at
3 sites in the San-Joaquin valley of California. J. Geophys. Res. [Atmos.] 99: 8281-8294.
Matsumi, Y.; Murakami, S.-L; Kono, M. (2001) High-sensitivity instrument for measuring atmospheric NO2. Anal.
Chem. 73:5485-5493.
Mauldin, R. L., Ill; Tanner, D. J.; Eisele, F. L. (1998) A new chemical ionization mass spectrometer technique for
the fast measurement of gas phase nitric acid in the atmosphere. J. Geophys. Res. 103(D3): 3361-3367.
McClenny, W. A., ed. (2000) Recommended methods for ambient air monitoring of NO, NO2, NOy, and individual
NOZ species. Research Triangle Park, NC: U.S. Environmental Protection Agency, National Exposure
Research Laboratory; report no. GPRA APM #442.
McDow, S. R.; Jang, M.; Hong, Y.; Kamens, R. M. (1996) An approach to studying the effect of organic
composition on atmospheric aerosol photochemistry. J. Geophys. Res. (Atmos.) 101(D14): 19,593-19,600.
McFiggans, G.; Allan, B.; Coe, H.; Plane, H. M. C.; Carpenter, L. J.; O'Dowd, C. (2000) A modeling study of iodine
chemistry in the marine boundary layer. J. Geophys. Res. (Atmos.) 105: 14,371-14,385.
McKeen, S. A.; Liu, S. C. (1993) Hydrocarbon ratios and photochemical history of air masses. Geophys. Res. Lett.
20: 2363-2366.
McKeen, S. A.; Liu, S. C. (1997) Hydrocarbon ratios and the photochemical history of air masses. Geophys. Res.
Lett. 20: 2363-2366.
McKeen, S. A.; Liu, S. C.; Hsie, E.-Y.; Lin, X.; Bradshaw, J. D.; Smyth, S.; Gregory, G. L.; Blake, D. R. (1996)
Hydrocarbon ratios during PEM-WEST A: a model perspective. J. Geophys. Res. (Atmos.) 101(D1):
2087-2109.
McKeen, S. A.; Mount, G.; Eisele, F.; Williams, E.; Harder, J.; Goldan, P.; Kuster, W.; Liu, S. C.; Baumann, K.;
Tanner, D.; Fried, A.; Sewell, S.; Cantrell, C.; Shelter, R. (1997) Photochemical modeling of hydroxyl and its
relationship to other species during the Tropospheric OH Photochemistry Experiment. J. Geophys. Res.
(Atmos.) 102: 6467-6493.
McKenna, D. S.; Konopka, P.; GrooB, J.-U.; Gtinther, G.; Mtiller, R.; Spang, R.; Offermann, D.; Orsolini, Y. (2002)
A new chemical Lagrangian model of the stratosphere (CLaMS) 1. formulation of advection and mixing. J.
Geophys. Res. 107(D16): 10.1029/2000JD000114.
Meagher, J. F.; Cowling, E. B.; Fehsenfeld, F. C.; Parkhurst, W. J. (1998) Ozone formation and transport in
southeastern United States: overview of the SOS Nashville/Middle Tennessee Ozone Study. J. Geophys. Res.
[Atmos.] 103:22,213-22,223.
Mendoza-Dominguez, A.; Russell, A. G. (2000) Iterative inverse modeling and direct sensitivity analysis of a
photochemical air quality model. Environ. Sci. Technol. 3: 4974-4981.
Mendoza-Dominguez, A.; Russell, A. G. (2001) Estimation of emission adjustments from the application of
four-dimensional data assimilation to photochemical air quality modeling. Atmos. Environ. 35: 2879-2894.
AX2-176
-------
Meng, Z.; Seinfeld, J. H. (1996) Time scales to achieve atmospheric gas-aerosol equilibrium for volatile species.
Atmos. Environ. 30: 2889-2900.
Merrill, J. T.; Moody, J. L. (1996) Synoptic meteorology and transport during the North Atlantic Regional
Experiment (NARE) intensive: overview. J. Geophys. Res. [Atmos.] 101: 28,903-28,921.
Meyrahn, H.; Pauly, J.; Schneider, W.; Warneck, P. (1986) Quantum yields for photodissociation of acetone in air
and an estimate for the life time of acetone in the lower troposphere. J. Atmos. Chem. 4: 277-291.
Michel, A. E.; Usher, C. R.; Grassian, V. H. (2002) Heterogeneous and catalytic uptake of ozone on mineral oxides
and dusts: a Knudsen cell investigation. Geophys. Res. Lett. 29: DOI: 10.1029/2002GLO14896.
Michel, A. E.; Usher, C. R.; Grassian, V. H. (2003) Reactive uptake of ozone on mineral oxides and mineral dusts.
Atmos. Environ. 37: 3201-3211.
Mickley, L. J.; Murti, P. P.; Jacob, D. J.; Logan, J. A.; Koch, D. M.; Rind, D. (1999) Radiative forcing from
tropospheric ozone calculated with a unified chemistry-climate model. J. Geophys. Res. [Atmos.] 104:
30,153-30,172.
Milford, J. B.; Gao, D.; Sillman, S.; Blossey, P.; Russell, A. G. (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.
Monson, R. K.; Harley, P. C.; Litvak, M. E.; Wildermuth, M.; Guenther, A. B.; Zimmerman, P. R.; Fall, R. (1994)
Environmental and developmental controls over the seasonal pattern of isoprene emission from aspen leaves.
Oecologia 99: 260-270.
Moody, J. L.; Oltmans, S. J.; Levy, H., II; Merrill, J. T. (1995) Transport climatology of tropospheric ozone:
Bermuda, 1988-1991. J. Geophys. Res. [Atmos.] 100: 7179-7194.
Moody, J. L.; Davenport, J. C.; Merrill, J. T.; Oltmans, S. J.; Parrish, D. D.; Holloway, J. S.; Levy, H., II; Forbes,
G. L.; Trainer, M.; Buhr, M. (1996) Meteorological mechanisms for transporting O3 over the western North
Atlantic Ocean: a case study for August 24-29, 1993. J. Geophys. Res. 101(D22): 29,213-29,227.
Moorthi, S.; Suarez, M. J. (1992) Relaxed Arakawa-Schubert: a parameterization of moist convection for general
circulation models. Mon. Weather Rev. 120: 978-1002.
Mopper, K.; Kieber, D. J. (2000) Marine photochemistry and its impact on carbon cycling. In: Demers, S.; Vernet,
M., eds. Effects of UV Radiation on Marine Ecosystems. New York, NY: Cambridge University Press, Ch.4,
pp. 101-129.
Moschonas, N.; Danalatos, D.; Glavas, S. (1999) The effect of O2 and NO2 on the ring retaining products of the
reaction of toluene withhydroxyl radicals. Atmos. Environ. 33: 111-116.
Mozurkewich, M. (1995) Mechanisms for the release of halogens from sea-salt particles by free radical reactions.
J. Geophys. Res. (Atmos.) 100: 14,199-14,207.
Mtiller, J.-F.; Brasseur, G. (1995) IMAGES: a three-dimensional chemical transport model of the global troposphere.
J. Geophys. Res. (Atmos.) 100: 16,445-16,490.
Mtiller, J.-F.; Brasseur, G. (1999) Sources of upper tropospheric HOX: a three-dimensional study. J. Geophys. Res.
(Atmos.) 104: 1705-1715.
Munoz, M. S. S.; Rossi, M. J. (2002) Heterogeneous reactions of HNO3 with flame soot generated under different
combustion conditions. Reaction mechanism and kinetics. Phys. Chem. Chem. Phys. 4: 5110-5118.
National Bureau of Standards. (1975) Catalog of NBS standard reference materials, 1975-76 edition. Washington,
DC: U.S. Department of Commerce, National Bureau of Standards; NBS special publication no. 260.
National Institute of Standards and Technology (NIST). (2006) NIST chemical kinetics database on the web.
Available: http://kinetics.nist.gov/index.php [13 February, 2006].
National Research Council. (1991) Rethinking the ozone problem in urban and regional air pollution. Washington,
DC: National Academy Press.
National Research Council. (2002) Estimating the public health benefits of proposed air pollution regulations.
Washington, DC: National Academy of Sciences. Available at:
http://books.nap.edi^ooks/0309086094/html/index.html [7 November, 2002].
Neuman, J. A.; Huey, L. G.; Dissly, R. W.; Fehsenfeld, F. C.; Flocke, F.; Holecek, J. C.; Holloway, J. S.; Htibler, G.;
Jakoubek, R.; Nicks, D. K., Jr.; Parrish, D. D.; Ryerson, T. B.; Sueper, D. T.; Weinheimer, A. J. (2002)
Fast-response airborne in-situ measurements of HNO3 during the Texas Air Quality Study. J. Geophys. Res.
(Atmos.) 107: 10.1029/2001JD001437.
Newchurch, M. J.; Ayoub, M. A.; Oltmans, S.; Johnson, B.; Schmidlin, F. J. (2003) Vertical distribution of ozone at
four sites in the United States. J. Geophys. Res. 108(D1): 10.1029/2002JD002059.
Newell, R. E.; Thouret, V.; Cho, J. Y. N.; Stoller, P.; Marenco, A.; Smit, H. G. (1999) Ubiquity of quasi-horizontal
layers in the troposphere. Nature (London) 398: 316-319.
AX2-177
-------
Nejgaard, J. K.; Christensen, K. B.; Wolkoff, P. (2005) The effect on human eye blink frequency of exposure to
limonene oxidation products and methacrolein. Toxicol. Lett. 156: 241-251.
Nordmeyer, T.; Wang, W.; Ragains, M. L.; Finlayson-Pitts, B. J.; Spicer, C. W.; Plastridge, R. A. (1997) Unique
products of the reaction of isoprene with atomic chlorine: potential markers. Geophys. Res. Lett.
24: 1615-1618.
Notholt, J.; Hjorth, J.; Raes, F. (1992a) Formation of HNO2 on aerosol surfaces during foggy periods in the presence
of NO and NO2. Atmos. Environ. Part A 26: 211-217.
Notholt, J.; Hjorth, J.; Raes, F.; Schrems, O. (1992b) Simultaneous long path field measurements of HNO2, CH2O
and aerosol. Ber. Bunsen Ges. Phys. Chem. 96: 290-293.
Nunnermacker, L. J.; Imre, D.; Daum, P. H.; Kleinman, L.; Lee, Y.-N.; Lee, J. H.; Springston, S. R.; Newman, L.;
Weinstein-Lloyd, J.; Luke, W. T.; Banta, R.; Alvarez, R.; Senff, C.; Sillman, S.; Holdren, M.; Keigley, G. W.;
Zhou, X. (1998) Characterization of the Nashville urban plume on July 3 and July 18, 1995. J. Geophys. Res.
103(D21): 28,129-28,148.
Nunnermacker, L. J.; Kleinman, L. I.; Imre, D.; Daum, P. H.; Lee, Y.-N.; Lee, J. H.; Springston, S. R.; Newman, L.;
Gillani, N. (2000) NOy lifetimes and O3 production efficiencies in urban and power plant plumes: analysis of
field data. J. Geophys. Res. [Atmos.] 105: 9165-9176.
Odman, M. T.; Ingram, C. I. (1996) Multiscale air quality simulation platform (MAQSIP): source code
documentation and validation. Research Triangle Park, NC: MCNC; technical report ENV-96TR002-vl.O.
Available: http://www.ce.gatech.edu/~todman/maqsip.pdf (9 September 2003).
Odman, M. T.; Russell, A. G. (1999) Mass conservative coupling of non-hydrostatic meteorological models with air
quality models. In: Crying, S.-E.; Batchvarova, E., eds. Air Pollution Modeling and its Application XIII. New
York, NY: Plenum Press.
Olsen, M. A.; Stanford, J. L. (2001) Evidence of stratosphere-to-troposphere transport within a mesoscale model and
total ozone mapping spectrometer total ozone. J. Geophys. Res. [Atmos.] 106: 27,323-27,334.
Olsen, M. A.; Douglass, A. R.; Schoeberl, M. R. (2002) Estimating downward cross-tropopause flux using column
ozone and potential vorticity. J. Geophys. Res. 107(D22): 10.1029/2001JD002041.
Oltmans, S. J.; Levy, H., II. (1992) Seasonal cycle of surface ozone over the North Atlantic. Nature 358: 392-394.
Oltmans, S. J.; Levy, H., II; Harris, J. M.; Merrill, J. T.; Moody, J. L.; Lathrop, J. A.; Cuevas, E.; Trainer, M.;
O'Neill, M. S.; Prospero, J. M.; Vomel, H.; Johnson, B. J. (1996) Summer and spring ozone profiles over the
North Atlantic from ozonesonde observations. J. Geophys. Res. 101: 29,179-29,200.
Orlando, J. J.; Tyndall, G. S.; Paulson, S. E. (1999) Mechanism of the OH-initiated oxidation of methacrolein.
Geophys. Res. Lett. 26: 2191-2194.
Padro, J. (1996) Summary of ozone dry deposition velocity measurements and model estimates over vineyard,
cotton, grass and deciduous forest in summer. Atmos. Environ. 30: 2363-2369.
Palmer, P. I.; Jacob, D. J.; Fiore, A. M.; Martin, R. V.; Chance, K.; Kuruso, T. P. (2003) Mapping isoprene
emissions over North America using formaldehyde column observations from space. J. Geophys. Res.
108(D6): 10.1029/2002JD002153.
Pandis, S. N.; Harley, R. A.; Cass, G. R.; Seinfeld, J. H. (1992) Secondary organic aerosol formation and transport.
Atmos. Environ. Part A 26: 2269-2282.
Papagni, C.; Arey, J.; Atkinson, R. (2001) Rate constants for the gas-phase reactions of OH radicals with a series of
unsaturated alcohols. Int. J. Chem. Kinet. 33: 142-147.
Park, R. J.; Stenchikov, G. L.; Pickering; Dickerson, R. R.; Allen, D. J.; Kondragunta, S. (2001) Regional air
pollution and its radiative forcing: studies with a single column chemical and radiation transport model.
J. Geophys. Res. [Atmos.] 106: 28,751-28,770.
Parrish, D. D.; Fehsenfeld, F. C. (2000) Methods for gas-phase measurements of ozone, ozone precursors and
aerosol precursors. Atmos. Environ. 34: 1921-1957.
Parrish, D. D.; Trainer, M.; Buhr, M. P.; Watkins, B. A.; Fehsenfeld, F. C. (1991) Carbon monoxide concentrations
and their relatin to concentrations of total reactive oxided nitrogen at two rural U.S. sites. J. Geophys. Res.
(Atmos.) 96: 9309-9320.
Parrish, D. D.; Halm, C. J.; Williams, E. J.; Norton, R. B.; Fehsenfeld, F. C.; Singh, H. B.; Shelter, J. D.; Gandrud,
B. W.; Ridley, B. A. (1992) Indications of photochemical histories of Pacific air masses from measurements
of atmospheric trace species at Point Arena, California. J. Geophys. Res. [Atmos.] 97: 15,883-15,901.
Parrish, D. D.; Halm, C. J.; Williams, E. J.; Norton, R. B.; Fehsenfeld, F. C.; Singh, H. B.; Shelter, J. D.; Gandrud,
B. W.; Ridley, B. A. (1993) Reply [to Comment on "Indications of photochemical histories of Pacific air
masses from measurements of atmospheric trace species at Point Arena, California" by D. D. Parrish et al.].
J. Geophys. Res. (Atmos.) 98(D8): 14,995-14,997.
AX2-178
-------
Parrish, D. D.; Trainer, M.; Young, V.; Goldan, P. D.; Kuster, W. C.; Jobson, B. T.; Fehsenfeld, F. C.; Lonneman,
W. A.; Zika, R. D.; Farmer, C. T.; Riemer, D. D.; Rodgers, M. O. (1998) Internal consistency tests for
evaluation of measurements of anthropogenic hydrocarbons in the troposphere. J. Geophys. Res. (Atmos.)
103(D17): 22,339-22,359.
Parrish, D. D.; Ryerson, T. B.; Holloway, J. S.; Trainer, M.; Fehsenfeld, F. C. (1999) Does pollution increase or
decrease tropospheric ozone in winter-spring? Atmos. Environ. 33: 5147-5149.
Parrish, D. D.; Holloway, J. S.; Jakoubek, R.; Trainer, M.; Ryerson, T. B.; Htibler, G.; Fehsenfeld, F. C.; Moody,
J. L.; Cooper. O. R. (2000) Mixing of anthropogenic pollution with stratospheric ozone: a case study from the
North Atlantic wintertime troposphere. J. Geophys. Res. [Atmos.] 105: 24,363-24,374.
Parrish, D. D.; Trainer, M.; Hereid, D.; Williams, E. J.; Olszyna, K. J.; Harley, R. A.; Meagher, J. F.; Fehsenfeld,
F. C. (2002) Decadal change in carbon monoxide to nitrogen oxide ratio in U.S. vehicular emissions J.
Geophys. Res. (Atmos.) 107(D13): 10.1029/2001JD000720.
Penner, J. E. (1995) Carbonaceous aerosols influencing atmospheric radiation: black and organic carbon. In:
Charlson, R. J.; Heintzenberg, J., eds. Aerosol Forcing of Climate. New York, NY: John Wiley and Sons;
pp. 91-108.
Penrose, W. R..; Pan, L.; Stetter, J. R.; Ollison, W. M. (1995) Sensitive measurement of ozone using amperometric
gas sensors. Anal. Chim. Acta 313: 209-219.
Perry, R. A.; Atkinson, R.; Pitts, J. N., Jr. (1977) Kinetics and mechanism of the gas phase reaction of OH radicals
with aromatic hydrocarbons over the temperature range 296-473 K. J. Phys. Chem. 81: 296-304.
Petriconi, G. L.; Papee, H. M. (1972) On the photolytic separation of halogens from sea water concentrates. Water
Air Soil Pollut. 1: 117-131.
Pickering, K. E.; Dickerson, R. R.; Huffman, G. J.; Boatman, J. F.; Schanot, A. (1988) Trace gas transport in the
vicinity of frontal convective clouds. J. Geophys. Res. 93: 759-773.
Pickering, K. E.; Thompson, A. M.; Dickerson, R. R.; Luke, W. T.; McNamara, D. P.; Greenberg, J. P.;
Zimmerman, P. R. (1990) Model calculations of tropospheric ozone production potential following observed
convective events. J. Geophys. Res. [Atmos.] 95: 14,049-14,062.
Pickering, K. E.; Thompson, A. M.; Scala, J. R.; Tao, W.-K.; Simpson, J.; Garstang, M. (1991) Photochemical ozone
production in tropical squall line convection during NASA Global Tropospheric Experiment/Amazon
Boundary Layer Experiment 2A. J. Geophys. Res. 96: 3099-3114.
Pickering, K. E.; Thompson, A. M.; Scala, J. R.; Tao, W.-K.; Simpson, J. (1992a) Ozone production potential
following convective redistribution of biomass burning emissions. J. Atmos. Chem. 14: 297-313.
Pickering, K. E.; Thompson, A. M.; Scala, J. R.; Tao, W.-K.; Dickerson, R. R.; Simpson, J. (1992b) Free
tropospheric ozone production following entrainment of urban plumes into deep convection. J. Geophys. Res.
97(D16): 17985-18,000.
Pickering, K. E.; Scala, J. R.; Thompson, A. M.; Tao, W.-K.; Simpson, J. (1992c) A regional estimate of the
convective transport of CO from biomass burning. Geophys. Res. Lett. 19: 289-292.
Pickering, K. E.; Thompson, A. M.; Tao, W.-K.; Kucsera, T. L. (1993) Upper tropospheric ozone production
following mesoscale convection during STEP/EMEX. J. Geophys. Res. 98: 8737-8749.
Pickering, K. E.; Thompson, A. M.; Tao, W.-K.; Rood, R. B.; McNamara, D. P.; Molod, A. M. (1995) Vertical
transport by convective clouds: comparisons of three modeling approaches. Geophys. Res. Lett. 22:
1089-1092.
Pickering, K. E.; Thompson, A. M.; Wang, Y.; Tao, W.-K.; McNamara, D. P.; Kirchhoff, W. J. H.; Heikes, B. G.;
Sachse, G. W.; Bradshaw, J. D.; Gregory, G. L.; Blake, D. R. (1996) Convective transport of biomass burning
emissions over Brazil during TRACE A. J. Geophys. Res. [Atmos.] 101: 23,993-24,012.
Pickering, K. E.; Wang, Y.; Tao, W.-K.; Price, C.; Mtiller, J.-F. (1998) Vertical distributions of lightning NOX for
use in regional and global chemical transport models. J. Geophys. Res. [Atmos.] 103: 31,203-31,216.
Pickering, K. E.; Thompson, A. M.; Kim, H.; DeCaria, A. J.; Pfister, L.; Kucsera, T. L.; Witte, J. C.; Avery, M. A.;
Blake, D. R.; Crawford, J. H.; Heikers, B. G.; Sachse, G. W.; Sandholm, S. T.; Talbot, R. W. (2001) Trace
gas transport and scavenging ini PEM-Tropics B South Pacific Convergence Zone convection. J. Geophys.
Res. [Atmos.] 106: 32,591-32,602.
Pierce, T. E.; Waldruff, P. S. (1991) PC-BEIS: a personal computer version of the biogenic emissions inventory
system. J. Air Waste Manage. Assoc. 41: 937-941.
Pierce, T; Geron, C.; Bender, L.; Dennis, R.; Tonnesen, G.; Guenther, A. (1998) Influence of increased isoprene
emissions on regional ozone modeling. J. Geophys. Res. (Atmos.) 103: 25,611-25,630.
Piety, C. A. (2005) The relation between daily maximum ozone and daily maximum temperature [memorandum to
Dr. Joseph Pinto]. Research Triangle Park, NC: U.S. Environmental Protection Agency; July 18.
AX2-179
-------
Pinto, J. P.; Bruhl, C.; Thompson, A. M. (1993) The current and future envirionmental role of atmospheric methane.
In: Khalil, M. A. K., ed. Atmospheric methane sources, sinks, and role in global change, p. 514-531. (NATO
ASI Series, v. 113).
Piotrowicz, S. R.; Bezdek, H. F.; Harvey, G. R.; Springer-Young, M.; Hanson, K. J. (1991) On the ozone minimum
over the equatorial Pacific Ocean. J. Geophys. Res. 96: 18,679-18,687.
Pitari, G.; Grassi, B.; Visconti, G. (1997) Results of a chemical transport model with interactive aerosol
microphysics. In: Bojkov, R.; Visconti, G., eds. Proceedings of the XVIII Quadrennial Ozone Symposium;
September 1996; L'Aquila, Italy, pp. 759-762.
Pitts, J. N., Jr.; Sanhueza, E.; Atkinson, R.; Carter, W. P. L.; Winer, A. M.; Harris, G. W.; Plum, C. N. (1984) An
investigation of the dark formation of nitrous acid in environmental chambers. Int. J. Chem. Kinet. 16:
919-939.
Plummer, D. A.; McConnell, J. C.; Shepson, P. B.; Hastie, D. R.; Niki, H. (1996) Modeling of ozone formation at a
rural site in southern Ontario. Atmos. Environ. 30: 2195-2217.
Pokharel, S. S.; Bishop, G. A.; Stedman, D. H. (2002) An on-road motor vehicle emissions inventory for Denver: an
efficient alternative to modeling. Atmos. Environ. 36: 5177-5184.
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.
Poppe, D.; Zimmerman, J.; Dorn, H. P. (1995) Field data and model calculations for the hydroxyl radical. J. Atmos.
Sci. 52: 3402-3407.
Poulida, O.; Dickerson, R. R.; Heymsfield, A. (1996) Troposphere-stratosphere exchange in a midlatitude mesoscale
convective complex: 1. Observations. J. Geophys. Res. [Atmos.] 101: 6823-6836.
Prados, A. I. (2000) The export of ozone and other atmospheric trace species from North America: observations and
numerical simulations [dissertation]. College Park, MD: The University of Maryland.
Prados, A. L; Dickerson, R. R.; Doddridge, B. G.; mime, P. A.; Moody, J. L.; Merrill, J. T. (1999) Transport of
ozone and pollutants from North America to the North Atlantic Ocean during the 1996 AEROCE intensive
experiment. J. Geophys. Res. 104: 26,219-26,233.
Prather, M.; Ehhalt, D. (2001) Atmospheric chemistry and greenhouse gases. In: Houghton, J. T.; Ding, Y.; Griggs,
D. J.; Noguer, M.; van der Linden, P. J.; Dai, X.; Maskell, K.; Johnson, C. A., eds. Climate Change 2001:
The Scientific Basis; pp. 239-287. Cambridge, MA: Cambridge University Press. Available:
http://www.grida.no/crimate/ipcc_tar/wgl/ (20 January 2004).
Prather, M. J.; Jacob, D. J. (1997) A persistent imbalance in HOX and NOX photochemistry in the upper troposphere
driven by deep tropical convection. Geophys. Res. Lett. 24: 3189-3192.
Prevot, A. S. H.; Staehelin, J.; Kok, G. L.; Schillawski, R. D.; Neininger, B.; Staffelbach, T.; Neftel, A.; Wernli, H.;
Dommen, J. (1997) The Milan photooxidant plume. J. Geophys. Res. (Atmos.) 102: 23,375-23,388.
Price, J. D.; Vaughan, G. (1993) The potential for stratosphere-troposphere exchange in cut-off-low systems. Q. J. R.
Meteorol. Soc. 119: 343-365.
Price, C.; Penner, J.; Prather, M. (1997) NOX from lightning: 1. Global distribution based on lightning physics.
J. Geophys. Res. [Atmos.] 102: 5929-5941.
Pszenny, A. A. P.; Keene, W. C.; Jacob, D. J.; Fan, S.; Maben, J. R.; Zetwo, M. P.; Springer-Young, M.; Galloway,
J. N. (1993) Evidence of inorganic chlorine gases other than hydrogen chloride in marine surface air.
Geophys. Res. Lett. 20: 699-702.
Pszenny, A. A. P.; Moldanova, J.; Keene, W. C.; Sander, R.; Maben, J. R.; Martinez, M.; Crutzen, P. J.; Perner, D.;
Prinn, R. G. (2004) Halogen cycling and aerosol pH in the Hawaiian marine boundary layer. Atmos. Chem.
Phys. 4: 147-168.
Pun, B. K.; Seigneur, C.; Grosjean, D.; Saxena, P. (2000) Gas-phase formation of water-soluble organic compounds
in the atmosphere: a retrosynthetic analysis. J. Atmos. Chem. 35: 199-223.
Quigley, D. C.; Hornafius, J. S.; Luyendyk, B. P.; Francis, R. D.; Clark, J.; Washburn, L. (1999) Decrease in natural
marine hydrocarbon seepage near Coal Oil Point, California, associated with offshore oil production. Geol.
27: 1047-1050.
Rancher, J.; Kritz, M. A. (1980) Diurnal fluctuations of Br and I in the tropical marine atmosphere. J. Geophys. Res.
85: 5581-5587.
Rao, S. T.; Zurbenko, I. G. (1994) Detecting and tracking changes in ozone air quality. J. Air Waste Manage. Assoc.
44: 1089-1092.
Reed, R. J. (1955) A study of a characteristic type of upper-level frontogenesis. J. Meteorol. 12: 226-237.
Regener, E. (1941) Ozonschicht und atmospharische turbulenz. Ber. Dtsch. Wetterdienstes US-Zone No. 11: 45-57.
Reisinger, A. R. (2000) Unidentified interference inDOAS measurements of ozone. Appl. Spectrosc. 54: 72-79.
AX2-180
-------
Reissell, A.; Arey, I; Atkinson, R. (2001) Atmospheric chemistry of camphor. Int. J. Chem. Kinet. 33: 56-63.
Reitebuch, O.; Strassburger, A.; Emeis, S.; Kuttler, W. (2000) Nocturnal secondary ozone concentration maxima
analysed by sodar observation and surface measurements. Atmos. Environ. 34: 4315-4329.
Reiter, E. R. (1963) A case study of radioactive fallout. J. Appl. Meteorol. 2: 691-705.
Reiter, E. R. (1975) Stratospheric-tropospheric exchange processes. Rev. Geophys. Space Phys. 13: 459-474.
Reiter, E. R.; Mahlman, J. D. (1965) Heavy radioactive fallout over the southern United States, November 1962.
J. Geophys. Res. 70: 4501-4520.
Reithmeier, C.; Sausen, R. (2002) ATTILA: atmospheric tracer transport in a Lagrangian model. Tellus 54B:
278-299.
Ren, X. R.; Harder, H.; Martinez, M; Lesher, R. L.; Oliger, A.; Shirley, T.; Adams, J.; Simpas, J. B.; Brune, W. H.
(2003) HOx concentrations and OH reactivity observations in New York City during PMTACS-NY2001.
Atmos. Environ. 37: 3627-3637.
Reynolds, S.; Michaels, H.; Roth, P.; Tesche, T. W.; McNally, D.; Gardner, L.; Yarwood, G. (1996) Alternative base
cases in photochemical modeling: their construction, role, and value. Atmos. Environ. 30: 1977-1988.
Ridley, B. A.; Walega, J. G.; Lamarque, J.-F.; Grahek, F. E.; Trainer, M.; Htibler, G.; Lin, X.; Fehsenfeld, F. C.
(1998) Measurements of reactive nitrogen and ozone to 5-km altitude in June 1990 over the southeastern
United States. J. Geophys. Res. 103: 8369-8388.
Riemer, D. D.; Apel, E. C. (2001) Confirming the presence and extent of oxidation by Cl in the Houston, Texas
urban area using specific isoprene oxidation products as tracers. Miami, FL: University of Miami, Rosenstiel
School of Marine and Atmospheric Science. Available:
http://clearinghouse.ces.utexas.edu/TexAQS/DB_Files/ConfirmingPresenceandExtentOfOxidationByCI.pdf
(5 February 2004).
Roberts, J. M.; Williams, J.; Baumann, K.; Buhr, M. P.; Goldan, P. D.; Holloway, J.; Htibler, G.; Kuster, W. C.;
McKeen, S. A.; Ryerson, T. B.; Trainer, M.; Williams, E. J.; Fehsenfeld, F. C.; Bertman, S. B.; Nouaime, G.;
Seaver, C.; Grodzinsky, G.; Rodgers, M.; Young, V. L. (1998) Measurements of PAN, PPN, and MPAN
made during the 1994 and 1995 Nashville Intensives of the Southern Oxidant Study: implications for regional
ozone production from biogenic hydrocarbons. J. Geophys. Res. [Atmos.] 103: 22,473-22,490.
Roelofs, G. J.; Scheeren, H. A.; Heland, J.; Ziereis, H.; Lelieveld, J. (2003) A model study of ozone in the eastern
Mediterranean free troposphere during MINOS (August 2001). Atmos. Chem. Phys. 3: 1199-1210.
Rogers, J. D. (1990) Ultraviolet absorption cross sections and atmospheric photodissociation rate constants of
formaldehyde. J. Phys. Chem. 94: 4011-4015.
Rohr, A. C.; Wilkins, C. K.; Clausen, P. A.; Hammer, M.; Nielsen, G. D.; Wolkoff, P.; Spengler, J. D. (2002) Upper
airway and pulmonary effects of oxidation products of (+)-a-pinene, J-limonene, and isoprene in B ALB/c
mice. Inhalation Toxicol. 14: 663-684.
Roselle, S. J. (1994) Effects of biogenic emission uncertainties on regional photochemical modeling of control
strategies. Atmos. Environ. 28: 1757-1772.
Rudich, Y.; Talukdar, R. K.; Fox, R. W.; Ravishankara, A. R. (1996) Rate coefficients for reactions of NO3 with a
few olefms and oxygenated olefins. J. Phys. Chem. 100: 5374-5381.
Rudolph, J.; Ramacher, B.; Plass-Dtilmer, C.; Mtiller, K.-P.; Koppmann, R. (1997) The indirect determination of
chlorine atom concentration in the troposphere from changes in the patterns of non-methane hydrocarbons.
Tellus 649:592-601.
Ruppert, L.; Becker, K. H. (2000) A product study of the OH radical-initiated oxidation of isoprene: formation of
C5-unsaturated diols. Atmos. Environ. 34: 1529-1542.
Russell, A.; Dennis, R. (2000) NARSTO critical review of photochemical models and modeling. Atmos. Environ.
34: 2283-2324.
Ryan, W. F.; Doddridge, B. G.; Dickerson, R. R.; Morales, R. M.; Hallock, K. A.; Roberts, P. T.; Blumenthal, D. L.;
Anderson, J. A.; Civerolo, K. (1998) Pollutant transport during a regional O3 episode in the mid-Atlantic
states. J. Air Waste Manage. Assoc. 48: 786-797.
Ryerson, T. B.; Buhr, M. P.; Frost, G. J.; Goldan, P. D.; Holloway, J. S.; Htibler, G.; Jobson, B. T.; Kuster, W. C.;
McKeen, S. A.; Parrish, D. D.; Roberts, J. M.; Sueper, D. T.; Trainer, M.; Williams, J.; Fehsenfeld, F. C.
(1998) Emissions lifetimes and ozone formation in power plant plumes. J. Geophys. Res. 103(D17):
22,569-22,583.
Ryerson, T. B.; Williams, E J.; Fehsenfeld, F. C. (2000) An efficient photolysis system for fast-response NO2
measurements. J. Geophys. Res. (Atmos.) 105(D21): 26,447-26,461.
AX2-181
-------
Ryerson, T. B.; Trainer, M; Holloway, J. S.; Parrish, D. D.; Huey, L. G.; Sueper, D. T.; Frost, G. J.; Donnelly,
S. G.; Schauffler, S.; Atlas, E. L.; Kuster, W. C.; Goldan, P. D.; Htibler, G.; Meagher, J. F.; Fehsenfeld, F. C.
(2001) Observations of ozone formation in power plant plumes and implications for ozone control strategies.
Science (Washington, DC) 292: 719-723.
Saathoff, H.; Naumann, K. H.; Riemer, N.; Kamm, S.; Mohler, O.; Schurath, U.; Vogel, H.; Vogel, B. (2001) The
loss of NO2, HNO3,,NO3/N2O5 and HO2/HOONO2 on soot aerosol: a chamber and modeling study. Geophys.
Res. Lett. 28: 1957-1960.
Sakamaki, F.; Hatakeyama, S.; Akimoto, H. (1983) Formation of nitrous acid and nitric oxide in the heterogeneous
dark reaction of nitrogen dioxide and water vapor in a smog chamber. Int. J. Chem. Kinet. 15: 1013-1029.
Sakugawa, H.; Kaplan, I. R. (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. [Atmos.] 94: 12,957-12,973.
Salmond, J. A.; McKendry, I. G. (2002) Secondary ozone maxima in a very stable nocturnal boundary layer:
observations from the Lower Fraser Valley, BC. Atmos. Environ. 36: 5771-5782.
Sander, R.; Crutzen, P. J. (1996) Model study indicating halogen activation and ozone destruction in polluted air
masses transported to the sea. J. Geophys. Res. [Atmos.] 101: 9121-9138.
Sander, R.; Rudich, Y.; Von Glasow, R.; Crutzen, P. J. (1999) The role of BrNO3 in marine tropospheric chemistry:
a model study. Geophys. Res. Lett. 26: 2858-2860.
Sander, R.; Keene, W. C.; Pszenny, A. A. P.; Arimoto, R. Ayers, G. P.; Chainey, J. M.; Crutzen, P. J.; Duce, R. A.;
Huebert, B. J.; Maenhaut, W.; Turekian, V. C.; VanDingenan, R. (2003) Inorganic bromine in the marine
boundary layer: a critical review. Atmos. Chem. Phys. 3: 1301-1336.
Sarwar, G.; Corsi, R.; Kumura, Y.; Allen, D.; Weschler, C. J. (2002) Hydroxyl radicals in indoor environments.
Atmos. Environ. 36: 3973-3988.
Sauer, F.; Schaefer, C.; Neeb, P.; Horie, O.; Moortgat, G. (1999) Formation of hydrogen peroxide in the ozonolysis
of isoprene and simple alkenes. Atmos. Environ. 33: 229-241.
Sawyer, R. F.; Harley, R. A.; Cadle, S. H.; Norbeck, J. M.; Slott, R.; Bravo, H. A. (2000) Mobile sources critical
review: 1998 NARSTO assessment. Atmos. Environ. 34: 2161-2181.
Saxena, P.; Hildemann, L. M.; McMurry, P. H.; Seinfeld, J. H. (1995) Organics alter hygroscopic behavior of
atmospheric particles. J. Geophys. Res. [Atmos.] 100: 18,755-18,770.
Scala, J. M.; Garstang, M.; Tao, W.-K.; Pickering, K. E.; Thompson, A. M.; Simpson, J.; Kirchhoff, V W. J. H.;
Browell, E. V; Sachse, G. W.; Torres. A. L.; Gregory, G. L.; Rasmussen, R. W.; Khalil, M. A. K. (1990)
Cloud draft structure and trace gas transport. J. Geophys. Res. 95: 17,015-17,030.
Schichtel, B. A.; Husar, R. B. (2001) Eastern North American transport climatology during high- and low-ozone
days. Atmos. Environ. 35: 1029-1038.
Schneider, M.; Ballschmiter, K. (1999) C3-C14-alkyl nitrates in remote South Atlantic air. Chemosphere 38: 233-244.
Schneider, M.; Luxenhofer, O.; Deissler, A.; Ballschmiter, K. (1998) C{-C15 alkyl nitrates, benzyl nitrate, and
bifunctional nitrates: Measurements in California and South Atlantic air and global comparison using C2C14
and CHBr3 as marker molecules. Environ. Sci. Technol. 32: 3055-3062.
Schubert, S. D.; Rood, R. B.; Pfaendtner, J. (1993) An assimilated dataset for earth science applications. Bull. Am.
Meteorol. Soc. 74: 2331-2342.
Schultz, M. G.; Jacob, D. J.; Bradshaw, J. D.; Sandholm, S. T.; Dibb, J. E.; Talbot, R. W.; Singh, H. B. (2000)
Chemical NOX budget in the upper troposphere over the tropical South Pacific. J. Geophys. Res. 105:
6669-6679.
Schumann, U.; Schlager, H.; Arnold, F.; Ovarlez, J.; Kelder, H.; Hov, 0.; Hayman, G.; Isaksen, I. S. A.; Staehelin,
J.; Whitefield, P. D. (2000) Pollution from aircraft emissions in the North Atlantic flight corridor: overview
on the POLINAT projects. J. Geophys. Res. 105(D3): 10.1029/1999JD900941.
Schtitze, M.; Herrmann, H. (2002) Determination of phase transfer parameters for the uptake of HNO3, N2O5 and O3
on single aqueous drops. Phys. Chem. Chem. Phys. 4: 60-67.
Seaman, N. L. (2000) Meteorological modeling for air quality assessments. Atmos. Environ. 34: 2231-2259.
Seinfeld, J. H.; Pandis, S. N. (1998) Atmospheric chemistry and physics: from air pollution to climate change. New
York, NY: John Wiley & Sons, Inc.
Seinfeld, J. H.; Pankow, J. F. (2003) Organic atmospheric paniculate material. Ann. Rev. Phys. Chem. 54: 121-140.
Senff, C. J.; Hardesty, R. M.; Alvarez, R. J., II; Mayor, S. D. (1998) Airborne lidar characterization of power plant
plumes during the 1995 Southern Oxidants Study. J. Geophys. Res. [Atmos.] 103: 31,173-31,189.
Seo, K.-H.; Bowman, K. P. (2002) Lagrangian estimate of global stratosphere-troposphere mass exchange. J.
Geophys. Res. (Atmos.) 107(D21): 10.1029/2002JD002441.
AX2-182
-------
Shapiro, M. A. (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, T. D.; Singsaas, E. L.; Vanderveer, P. J.; Geron, C. D. (1996) Field measurements of isoprene emission
from trees in response to temperature and light. Tree Physiol. 16: 649-654.
Shepson, P. B.; Edney, E. O.; Corse, E. W. (1984) Ring fragmentation reactions on the photooxidations of toluene
and o-xylene. J. Phys. Chem. 88: 4122-4126.
Shelter, R. E.; Junkermann, W.; Swartz, W. H.; Frost, G. J.; Crawford, J. H.; Lefer, B. L.; Barrick, J. D.; Hall, S. R.;
Hofzumahaus, A.; Bais, A.; Calvert, J. G.; Cantrell, C. A.; Madronich, S.; Mtiller, M.; Kraus, A.; Monks,
P. S.; Edwards, G. D.; McKenzie, R.; Johnston, P.; Schmitt, R.; Griffioen, E.; Krol, M.; Kylling, A.;
Dickerson, R. R.; Lloyd, S. A.; Martin, T.; Gardiner, B.; Mayer, B.; Pfister, G.; Roeth, E. P.; Koepke, P.;
Ruggaber, A.; Schwander, H.; Van Weele, M. (2003) Photolysis frequency of NO2: Measurement and
modeling during the International Photolysis Frequency Measurement and Modeling intercomparison
(IPMMI). J. Geophys. Res. (Atmos.) 108(D16): 10.1029/2002JD002932.
Sillman, S. (1995) The use of NOy, H2O2 and HNO3 as indicators for O3-NOX-ROG sensitivity in urban locations.
J. Geophys. Res. 100: 14,175-14,188.
Sillman, S. (2000) Ozone production efficiency and loss of NOX in power plant plumes: photochemical model and
interpretation of measurements in Tennessee. J. Geophys. Res. [Atmos.] 105: 9189-9202.
Sillman, S.; He, D.-Y. (2002) Some theoretical results concerning O3-NOX-VOC chemistry and NOX-VOC
indicators. J. Geophys. Res. (Atmos.) 107: 10.1029/2001JDOO1123.
Sillman, S.; Samson, P. J. (1995) Impact of temperature on oxidant photochemistry in urban, polluted rural and
remote environments. J. Geophys. Res. [Atmos.] 100: 11,497-11,508.
Sillman, S.; Al-Wali, K. L; Marsik, F. J.; Nowacki, P.; Samson, P. J.; Rodgers, M. O.; Garland, L. J.; Martinez, J. E.;
Stoneking, C.; Imhoff, R.; Lee, J.-H.; Newman, L.; Weinstein-Lloyd, J.; Aneja, V. P. (1995) Photochemistry
of ozone formation in Atlanta, GA-models and measurements. Atmos. Environ. 29: 3055-3066.
Sillman, S.; He, D.; Cardelino, C.; Imhoff, R. E. (1997) The use of photochemical indicators to evaluate
ozone-NOx-hydrocarbon sensitivity: case studies from Atlanta, New York, and Los Angeles. J. Air Waste
Manage. Assoc. 47: 1030-1040.
Sillman, S.; He, D.; Pippin, M. R.; Daum, P. H.; Imre, D. G.; Kleinman, L. I.; Lee, J. H.; 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. [Atmos.] 103:
22,629-22,644.
Sillman, S.; Vautard, R.; Menut, L.; Kley, D. (2003) O3-NOX-VOC sensitivity and NOX-VOC indicators in Paris:
results from models and atmospheric pollution over the Paris area (ESQUIF) measurements. J. Geophys. Res.
(Atmos.) 108: 10.1029/2002JD001561.
Singh, H. B.; Kanakidou, M.; Crutzen, P. J.; Jacob, D. J. (1995) High concentrations and photochemical fate of
oxygenated hydrocarbons in the global troposphere. Nature (London) 378: 50-54.
Singh, H. B.; Gregory, G. L.; Anderson, B.; Browell, E.; Sachse, G. W.; Davis, D. D.; Crawford, J.; Bradshaw, J. D.;
Talbot, R.; Blake, D. R.; Thornton, D.; Newell, R.; Merrill, J. (1996a) Low ozone in the marine boundary
layer of the tropical Pacific Ocean: photochemical loss, chlorine atoms, and entrainment. J. Geophys. Res.
[Atmos.] 101: 1907-1917.
Singh, H. B.; Herlth, D.; Kolyer, R.; Salas, L.; Bradshaw, J. D.; Sandholm, S. T.; Davis, D. D.; Crawford, J.; Kondo,
Y.; Koike, M.; Talbot, R.; Gregory, G. L.; Sachse, G. W.; Browell, E.; Blake, D. R.; Rowland, F. S.; Newell,
R.; Merrill, J.; Heikes, B.; Liu, S. C.; Crutzen, P. J.; Kanakidou, M. (1996b) Reactive nitrogen and ozone over
the western Pacific: distribution, partitioning, and sources. J. Geophys. Res. 101: 1793-1808.
Singh, M. P.; McNider, R. T.; Meyers, R.; Shekhar, G. (1997) Nocturnal wind structure and dispersion of pollutants:
an analytical study. Atmos. Environ. 31: 105-115.
Singh, H. B.; Salas, L. J.; Chatfield, R. B.; Czech, E.; Fried, A.; Walega, J.; Evans, M. J.; Field, B. D.; Jacob, D. J.;
Blake, D.; Heikes, B.; Talbot, R.; Sachse, G.; Crawford, J. H.; Avery, M. A.; Sandholm, S.; Fuelberg, H.
(2004) Analysis of the atmospheric distribution, sources, and sinks of oxygenated volatile organic chemicals
based on measurements over the Pacific during TRACE-P. J. Geophys. Res. (Atmos.) 109(D15S07):
10.1029/2003 JD003883.
Skiba, U.; Fowler, D.; Smith, K.A. (1997) Nitric oxide emissions from agricultural soils in temperate and tropical
climates: sources, controls and mitigation options. Nutrient Cycling Agroecosystems. 48 (1-2): 139-153.
Slemr, J.; Slemr, F.; Partridge, R.; D'Souza, H.; Schmidbauer, N. (2002) Accurate Measuements of Hydrocarbons in
the Atmosphere (AMOHA): three European intercomparisons. J. Geophys. Res. [Atmos.]
107(019): 10.1020/2001JD001357.
AX2-183
-------
Smith, A. M; Rigler, E.; Kwok, E. S. C.; Atkinson, R. (1996) Kinetics and products of the gas-phase reactions of
6-methyl-5-hepten-2-one and /raws-cinnamaldehyde with OH and NO3 radicals and O3 at 29672K. Environ.
Sci. Technol. 30: 1781-1785.
Smith, D. F.; Mclver, C. D.; Kleindienst, T. E. (1998) Primary product distribution from the reaction of hydroxyl
radicals with toluene at ppb NOX mixing ratios. J. Atmos. Chem. 30: 209-228.
Smith, D. F.; Kleindienst, T. E.; Mclver, C. D. (1999) Primary product distributions from the reaction of OH with
m-, p-xylene, 1,2,4-and 1,3,5-trimethylbenzene. J. Atmos. Chem. 34: 339-364.
Spicer, C. W.; Billick, I. H.; Yanagisawa, Y. (2001) Nitrous acid interference with passive NO2 measurement
methods and the impact on indoor NO2 data. Indoor Air 11: 156-161.
Spivakovsky, C. M.; Logan, J. A.; Montzka, S. A.; Balkanski, Y. J.; Foreman-Fowler, M.; Jones, D. B. A.;
Horowitz, L. W.; Fusco, A. C.; Brenninkmeijer, C. A. M.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B.
(2000) Three-dimensional climatological distribution of tropospheric OH: update and evaluation. J. Geophys.
Res. (Atmos.) 105: 8931-8980.
Sprenger, M.; Wernli, H. (2003) A northern hemispheric climatology of cross-tropopause exchange for the ERA 15
time period (1979-1993). J. Geophys. Res. 108(D12): 10.1029/2002JD002636.
Sprenger, M.; Maspoli, M. C.; Wernli, H. (2003) Tropopause folds and cross-tropopause exchange: a global
investigation based upon ECMWF analyses for the time period March 2000 to February 2001. J. Geophys.
Res. (Atmos.) 108(D12): 10.1029/2002JD002587.
Sprengnether, M.; Demerjian, K. L.; Donahue, N. M.; Anderson, J. G. (2002) Product analysis of the OH oxidation
of isoprene and 1,3-butadiene in the presence of NO J. Geophys. Res. (Atmos.) 107(D15):
10.1029/2001JD000716.
Sprung, D.; Jost, C.; Reiner, T.; Hansel, A.; Wisthaler, A. (2001) Acetone and acetonitrile in the tropical Indian
ocean boundary layer and free troposphere: aircraft-based intercomparison of AP-CIMS and PTR-MS
measurements. J. Geophys. Res. 106: 28,511-28,528.
Staffelbach, T.; Neftel, A.; Blatter, A.; Gut, A.; Fahrni, M.; Stahelin, J.; Prevot, A.; Hering, A.; Lehning, M.;
Neininger, B.; Ba'umle, M.; Kok, G. L.; Dommen, J.; Hutterli, M.; Anklin, M. (1997) Photochemical oxidant
formation over southern Switzerland 1. results from summer 1994. J. Geophys. Res. (Atmos.) 102:
23,345-23,362.
Staudt, A. C.; Jacob, D. J.; Ravetta, F.; Logan, J. A.; Bachiochi, D.; Krishnamurti, T. N.; Sandholm, S. T.; Ridley, B.
A.; Singh, H. B.; Talbot, R. W. (2003) Sources and chemistry of nitrogen oxides over the tropical Pacific.
J. Geophys. Res. (Atmos.) 108(8239): 10.1029/2002JD002139.
Stedman, D. H.; Bishop, G.; Peterson, J. E.; Guenther, P. L. (1991) On-road CO remote sensing in the Los Angeles
Basin: final report. Sacramento, CA: California Air Resources Board, ARE Contract No. A932-189.
Stehr, J. W.; Dickerson, R. R.; Hallock-Waters, K. A.; Doddridge, B. G.; Kirk, D. (2000) Observations of NOy, CO,
and SO2 and the origin of reactive nitrogen J. Geophys. Res. 105(D3): 3553-3563.
Stein, A. F.; Lamb, D. (2003) Empirical evidence for the low- and high-NOx photochemical regimes of sulfate and
nitrate formation. Atmos. Environ. 37: 3615-3625.
Stein, A. F.; Lamb, D.; Draxler, R. R. (2000) Incorporation of detailed chemistry into a three-dimensional
Lagrangian-Eulerian hybrid model: application to regional tropospheric ozone. Atmos. Environ. 34:
4361-4372.
Stenchikov, G.; Dickerson, R.; Pickering, K.; Ellis, W. Jr.; Doddridge, B.; Kondragunta, S.; Poulida, O.; Scala, J.;
Tao, W.-K. (1996) Stratosphere-troposphere exchange in a midlatitude mesoscale convective complex. 2.
Numerical simulation. J. Geophys. Res. [Atmos.] 101: 6837-6851.
Stevens, R. K.; Drago, R. J.; Mamane, Y. (1993) A long path differential optical absorption spectrometer and
EPA-approved fixed-point methods intercomparison. Atmos. Environ. PartB 27: 231-236.
Steyn, D. J.; Bottenheim, J. W.; Thomson, R. B. (1997) Overview of tropospheric ozone in the Lower Fraser Valley,
and the Pacific '93 field study. Atmos. Environ. 31: 2025-2036.
Stith, J.; Dye, J.; Ridley, B.; Laroche, P.; Defer, E.; Baumann, K.; Htibler, G.; Zerr, R.; Venticinque, M. (1999) NO
signatures from lightning flashes. J. Geophys. Res. [Atmos.] 101: 6837-6851.
Stockwell, W. R.; Middleton, P.; Chang, J. S.; Tang, X. (1990) The second generation Regional Acid Deposition
Model chemical mechanism for regional air quality modeling. J. Geophys. Res. [Atmos.] 95: 16,343-16,367.
Stockwell, W. R.; Kirchner, F.; Kuhn, M.; Seefeld, S. (1997) A new mechanism for regional atmospheric chemistry
modeling. J. Geophys. Res. 102(D22): 25847-25879.
Stohl, A. (2001) A 1-yearLagrangian "climatology" of airstreams in the Northern Hemisphere troposphere and
lowermost stratosphere. J. Geophys. Res. (Atmos.) 106: 7263-7279.
AX2-184
-------
Stohl, A.; Trickl, T. (1999) A textbook example of long-range transport: simultaneous observation of ozone maxima
of stratospheric and North American origin in the free troposphere of Europe. J. Geophys. Res. 104(D23):
30,445-30,462.
Stohl, A.; Trainer, M; Ryerson, T. B.; Hollo way, J. S.; Parrish, D. D. (2002) Export of NOy from the North
American boundary layer during 1996 and 1997 North Atlantic Regional Experiments. J. Geophys. Res.
107(011): 10.1029/2001JD000519.
Stohl, A.; Bonasoni, P.; Cristofanelli, P.; Collins, W.; Feichter, J.; Frank, A.; Forster, C.; Gerasopoulos, E.;
Gaggeler, H.; James, P.; Kentarchos, T.; Kromp-Kolb, H.; Kriiger, B.; Land, C.; Meloen, J.; Papayannis, A.;
Priller, A.; Seibert, P.; Sprenger, M.; Roelofs, G. J.; Scheel, H. E.; Schnabel, C.; Siegmund, P.; Tobler, L.;
Trickl, T.; Wernli, H.; Wirth, V.; Zanis, P.; Zerefos, C. (2003) Stratosphere-troposphere exchange: a review
and what we have learned from STACCATO. J. Geophys. Res. 108(D12): 10.1029/2002JD002490.
Stroud, C.; Madronich, S.; Atlas, E.; Ridley, B.; Flocke, F.; Weinheimer, A.; Talbot, B.; Fried, A.; Wert, B.; Shelter,
R.; Lefer, B.; Coffey, M.; Heikes, B.; Blake, D. (2003) Photochemistry in the arctic free troposphere: NOX
budget and the role of odd nitrogen reservoir recycling. Atmos. Environ. 37: 3351-3364.
Stull, R. B. (1985) A fair-weather cumulus cloud classification scheme for mixed layer studies. J. Clim. Appl.
Meteorol. 24: 49-56.
Stull, R. B. (1988) An introduction to boundary layer meteorology. Dordrecht, The Netherlands: Kluwer Academic
Publishers, pp. 9-16; 499-506. (Atmospheric Sciences Library, v. 13).
Sturges, W. T.; Barrie, L. A. (1988) Chlorine Bromine and Iodine in Arctic aerosols. Atmos. Environ.
22: 1179-1194.
Stutz, J.; Platt, U. (1996) Numerical analysis and estimation of the statistical error of differential optical absorption
spectroscopy measurements with least-squares methods. Appl. Opt. 35: 6041-6053.
Stutz, J.; Hebestreit, K.; Alicke, B.; Platt, U. (1999) Chemistry of halogen ozides in the troposphere: comparison of
model calculations with recent field data. J. Atmos. Chem. 34: 65-85.
Stutz, J.; Ackermann, R.; Fast, J. D.; Barrie, L. (2002) Atmospheric reactive chlorine and bromine at the Great Salt
Lake, Utah. Geophys. Res. Lett. 29: 10.1029/2002GL014812.
Suhre, K.; Gammas, J.-P.; Nedelec, P.; Rosset, R.; Marenco, A.; Smit, H. G. J. (1997) Ozone-rich transients in the
upper equatorial Atlantic troposphere (letter). Nature 388: 661-663.
Sundet, J. K. (1997) Model studies with a 3-D global CTM using ECMWF data [thesis]. Oslo, Norway: University
of Oslo, Department of Geophysics.
Svensson, R.; Ljungstroem, E.; Lindqvist, O. (1987) Kinetics of the reaction between nitrogen dioxide and water
vapour. Atmos. Environ. 21: 1529-1539.
Talbot, R. W.; Vijgen, A. S.; Harris, R. C. (1990) Measuring tropospheric HNO3: problems and prospects for nylon
filter and mist chamber. J. Geophys. Res. [Atmos.] 95: 7553-7542.
Tanaka, P. L.; Oldfield, S.; Neece, J. D.; Mullins, C. B.; Allen, D. T. (2000) Anthropogenic sources of chlorine and
ozone formation in urban atmospheres. Environ. Sci. Technol. 34: 4470-4473.
Tanaka, P. L.; Riemer, D. D.; Chang, S.; Yarwood, G.; McDonald-Buller, E. C.; Apel, E. C.; Orlando, J. J.; Silva,
P. J.; Jimenez, J. L.; Canagaratna, M. R.; Neece, J. D.; Mullins, C. B.; Allen, D. T. (2003) Direct evidence for
chlorine-enhanced urban ozone formation in Houston, Texas. Atmos. Environ. 37: 1393-1400.
Tanner, R. L.; Kelly, T. J.; Dezaro, D. A.; Forrest, J. (1989) A comparison of filter, denuder, and real-time
chemiluminescence techniques for nitric acid determination in ambient air. Atmos. Environ. 23: 2213-2222.
Tao, W.-K.; Simpson, J. (1993) The Goddard Cumulus Ensemble Model. Part I: model description. Terr. Atmos.
Oceanic Sci. (TAO) 4: 35-71.
Taubman, B. F.; Marufu, L. T.; Piety, C. A.; Doddridge, E.G.; Stehr, J. W.; Dickerson, R. R. (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, B. F.; Hains, J. C.; Thompson, A. M.; Marufu, L. T.; Doddridge, B. G.; Stehr, J. W.; Piety, C. A.;
Dickerson, R. R. (2005) Aircraft vertical profiles of trace gas and aerosol pollution over the mid-Atlantic
U.S.: statistics and meteorological cluster analysis. J. Geophys. Res. (Atmos.): submitted.
Thielman, A.; Prevot, A. S. H.; Griiebler, F. C.; Staelhelin, J. (2001) Empirical ozone isopleths as a tool to identify
ozone production regimes. Geophys. Res. Lett. 28: 2369-2372.
Thompson, A. M.; Stewart, R. W. (1991) Effect of chemical kinetics uncertainties on calculated constituents in a
tropospheric photochemical model. J. Geophys. Res. 96(D7): 13,089-13,108.
Thompson, A. M.; Pickering, K. E.; Dickerson, R. R.; Ellis, W. G., Jr.; Jacob, D. J.; Scala, J. R.; Tao, W.-K.;
McNamara, D. P.; Simpson, J. (1994) Convective-transport over the central United States and its role in
regional CO and O3 budgets. J. Geophys. Res. [Atmos.] 99: 18,703-18,711.
AX2-185
-------
Thompson, A. M; Singh, H. B.; Schlager, H. (2000) Subsonic assessment ozone and nitrogen oxide experiment
(SONEX) and pollution from aircraft emissions in the North Atlantic Flight Corridor (POLINAT 2).
J. Geophys. Res. 105: 3595-3603.
Thornton, J. A.; Wooldridge, P. J.; Cohen, R. C. (2000) Atmospheric NO2: in situ laser-induced fluorescence
detection at parts per trillion mixing ratios. Anal. Chem. 72: 528-539.
Thornton, J. A.; Braban, C. F.; Abbatt, J. P. D. (2003) N2O5 hydrolysis on sub-micron organic aerosol: the effect of
relative humidity, particle phase and particle size. Phys. Chem. Chem. Phys. 5: 4593-4603.
Tie, X. X.; Emmons, L.; Horowitz, L.; Brasseur, G.; Ridley, ,B.; Atlas, E.; Stround, C.; Hess, P.; Klonecki, A.;
Madronich, S.; Talbot, R.; Dibb, J. (2003) Effect of sulfate aerosol on tropospheric NOX and ozone budgets:
model simulations and TOPSE evidence. J. Geophys. Res. (Atmos.) 108(D4): 10.1029/2001JD001508.
Tonnesen, G. S.; Dennis, R. L. (2000a) Analysis of radical propagation efficiency to assess ozone sensitivity to
hydrocarbons and NOX 1. Local indicators of odd oxygen production sensitivity. J. Geophys. Res. (Atmos.)
105: 9213-9225.
Tonnesen, G. S.; Dennis, R. L. (2000b) Analysis of radical propagation efficiency to assess ozone sensitivity to
hydrocarbons and NOX 2. Long-lived species as indicators of ozone concentration sensitivity. J. Geophys.
Res. 105: 9227-9241.
Trainer, M.; Buhr, M. P.; Curran, C. M.; Fehsenfeld, F. C.; Hsie, E. Y.; Liu, S. C.; Norton, R. B.; Parrish, D. D.;
Williams, E. J.; Gandrud, B. W.; Ridley, B. A.; Shelter, J. D.; Allwine, E. J.; Westberg, H. H. (1991)
Observations and modeling of the reactive nitrogen photochemistry at a rural site. J. Geophys. Res. [Atmos.]
96: 3045-3063.
Trainer, M.; Parrish, D. D.; Buhr, M. P.; Norton, R. B.; Fehsenfeld, F. C.; Anlauf, K. G.; Bottenheim, J. W.; Tang,
Y. Z.; Wiebe, H. A.; Roberts, J. M.; Tanner, R. L.; Newman, L.; Bowersox, V. C.; Meagher, J. F.; Olszyna,
K. J.; Rodgers, M. O.; Wang, T.; Berresheim, H.; Demerjian, K. L.; Roychowdhury, U. K. (1993) Correlation
of ozone with NOy in photochemically aged air. J. Geophys. Res. [Atmos.] 98: 2917-2925.
Trainer, M.; Ridley, B. A.; Buhr, M. P.; Kok, G.; Walega, J.; Htibler, G.; Parrish, D. D.; Fehsenfeld, F. C. (1995)
Regional ozone and urban plumes in the southeastern United States: Birmingham, a case study. J. Geophys.
Res. [Atmos.] 100: 18,823-18,834.
Trainer, M.; Parrish, D. D.; Golday, P. D.; Roberts, J.; Fehsenfeld, F. C. (2000) Review of observation-based
analysis of the regional factors influencing ozone concentrations. Atmos. Environ. 34: 2045-2061.
Traub, M.; Fischer, H.; De Reus, M.; Kormann, R.; Heland, J.; Ziereis, H.; Schlager, H.; Holzinger, R.; Williams, J.;
Warneke, C.; De Gouw, J.; Lelieveld, J. (2003) Chemical characteristics assigned to trajectory clusters during
the MINOS campaign. Atmos. Chem. Phys. 3: 459-468.
Tseng, R.-S.; Viechnicki, J. T.; Skop, R. A.; Brown, J. W. (1992) Sea-to-air transfer of surf ace-active organic
compounds by bursting bubbles. J. Geophys. Res. (Atmos.) 97: 5201-5206.
Tuazon, E. C.; MacLeod, H.; Atkinson, R.; Carter, W. P. L. (1986) a-Dicarbonyl yields from the NOx-air
photooxidations of a series of aromatic hydrocarbons in air. Environ. Sci. Technol. 20: 383-387.
Tully, F. P.; Ravishankara, A. R.; Thompson, R. L.; Nicovich, J. M.; Shah, R. C.; Kreutter, N. M.; Wine, P. H.
(1981) Kinetics of the reactions of hydroxyl radical with benzene and toluene. J. Phys. Chem. 85: 2262-2269.
Turekian, V. C.; Macko, S. A.; Keene, W. C. (2003) Concentrations, isotopic compositions, and sources of
size-resolved, paniculate organic carbon and oxalate in near-surface marine air at Bermuda during spring.
J. Geophys. Res. (Atmos.) 108(D5): 10.1029/2002JD002053.
U.S. Congress. (1990) Clean Air Act amendments of 1990, PL 101-549, November 15. Washington, DC: U.S.
Government Printing Office.
U.S. Environmental Protection Agency. (1991) Guideline for regulatory application of the urban airshed model.
Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA-450/4-91-013.
Available from: NTIS, Springfield, VA; PB92-108760/HSU.
U.S. Environmental Protection Agency. (1993) Air quality criteria for oxides of nitrogen. Research Triangle Park,
NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office; report
nos. EPA/600/8-9!/049aF-cF. 3v. Available from: NTIS, Springfield, VA; PB95-124533, PB95-124525, and
PB95-124517.
U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants, v.
I-III. Research Triangle Park, NC: Office of Research and Development; report no. EPA/600/P-93/004aF-cF.
Available from: NTIS, Springfield, VA; PB96-185582. PB96-185590, andPB96-185608. Available:
www.epa.gov/ncea/ozone.htm [25 February 2002].
AX2-186
-------
U.S. Environmental Protection Agency. (1997) National air pollutant emission trends 1990-1996. Research Triangle
Park, NC: Office of Air Quality Planning and Standards; report no. EPA/454/R-97/011. Available from:
NTIS, Springfield, VA.
U.S. Environmental Protection Agency. (1999) Control measure evaluations: the Control Measure Data Base for the
National Emissions Trends Inventory (ControlNET) [draft report]. Research Triangle Park, NC: Office of Air
Quality Planning and Standards; report no. EPA/452/D-01/001. Available from: NTIS, Springfield, VA;
PB2001-105812.
U.S. Environmental Protection Agency. (2000) Air quality criteria for carbon monoxide. Research Triangle Park,
NC: National Center for Environmental Assessment; report no. EPA/600/P-99/00 IF. Available:
http://www.epa.gov/ncea/pdfs/coaqcd.pdf (7 May 2003).
U.S. Environmental Protection Agency. (2001) National air quality and emissions trends report, 1999. Research
Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA 454/R-01-004. Available:
http://www.epa.gov/oar/aqtrnd99/ [7 May 2003].
U.S. Environmental Protection Agency. (2003) Air quality criteria for ozone and related photochemical oxidants.
Research Triangle Park, NC: Office of Research and Development: manuscript in preparation.
U.S. Environmental Protection Agency. (2004) Air quality criteria for paniculate matter. Research Triangle Park,
NC: National Center for Environmental Assessment; report no. EPA/600/P-99/002aF-bF. 2v. Available:
http://cfpub.epa.gov/ncea/ [9 November, 2004].
Underwood, G. M.; Song, C. H.; Phadnis, M.; Carmichael, G. R.; Grassian, V. H. (2001) Heterogeneous reactions of
NO2 and HNO3 on oxides and mineral dust: a combined laboratory and modeling study. J. Geophys. Res.
[Atmos.] 106: 18,055-18,066.
Usher, C. R.; Michel, A. E.; Stec, D.; Grassian, V. H. (2003) Laboratory studies of ozone uptake on processed
mineral dust. Atmos. Environ. 37: 5337-5347.
Valente, R. J.; Imhoff, R. E.; Tanner, R. L.; Meagher, J. F.; Damn, P. H.; Hardesty, R. M.; Banta, R. M.; Alvarez,
R. J.; McNider, R. T.; Gillani, N. V. (1998) Ozone production during an urban air stagnation episode over
Nashville, Tennessee. J. Geophys. Res. [Atmos.] 103: 22,555-22,568.
Van Aardenne, J. A.; Dentener, F. J.; Olivier, J. G. J.; Goldewijk, C. G. M. K.; Lelieveld, J. (2001) A 1 ° x 1°
resolution data set of historical anthropogenic trace gas emissions for the period 1980-1990. Glob.
Biogeochem. Cycles 15: 909-928.
Vandaele, A. C.; Tsouli, A.; Carleer, M.; Colin, R. (2002) UV Fourier transform measurements of tropospheric O3,
NO2, SO2, benzene, and toluene. Environ. Pollut. 116: 193-201.
Vautard, R.; Martin, D.; Beekman, M.; Drobinski, P.; Friedrich, R.; Jaubertie, A.; Kley, D.; Lattuati, M.; Moral, P.;
Neininger, B.; Theloke, J. (2002) Paris emission inventory diagnostics from the ESQUIF airborne
measurements and a chemistry transport model. J. Geophys. Res. (Atmos.) 108(D17):
10.1029/2002JD002797.
Venkatachari, P.; Hopke, P. K.; Grover, B. D.; Eatough, D. J. (2005a) Measurement of particle-bound reactive
oxygen species inRubidoux aerosols. J. Atmos. Chem. 50: 49-58.
Venkatachari, P.; Hopke, P .K.; Brune, W. H.; Ren, X.; Lesher, R.; Mao, J.; Mitchell, M. (2005b) Characterization
of reactive oxygen species trends in Flushing, New York. Atmos. Environ.: in press.
Vogt, R.; Crutzen, P. J.; Sander, R. (1996) A mechanism for halogen release from sea-salt aerosol in the remote
marine boundary layer. Nature 383: 327-330.
Vogt, R.; Sander, R.; Von Glasow, R.; Crutzen, P. J. (1999) Iodine chemistry and its role in halogen activation and
ozone loss in the marine boundary layer: a model study. J. Atmos. Chem. 32: 375-395.
Volz-Thomas, A.; Geiss, H.; Hofzumahaus, A.; Becker, K.-H. (2003) Introduction to special section: photochemistry
experiment in BERLIOZ. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/JD002029.
Von Glasow, R.; Sander, R.; Bott, A.; Crutzen, P. J. (2002a) Modeling halogen chemistry in the marine boundary
layer. 1. Cloud-free MBL. J. Geophys. Res. 107(D17): 10.1029/2001JD000942.
Von Glasow, Sander, R.; Bott, A.; Crutzen, P. J. (2002b) Modeling halogen chemistry in the marine boundary layer.
2. Interactions with sulfur and cloud-covered MBL. J. Geophys. Res. 107(D17): 10.1029/2001JD000943.
Vukovich, F. M. (1995) Regional-scale boundary layer ozone variations in the eastern United States and their
association with meteorological variations. Atmos. Environ. 29: 2259-2273.
Vukovich, F. M. (1997) Time scales of surface ozone variations in the regional, non-urban environment. Atmos.
Environ. 31:1513-1530.
Vukovich, R. M.; Ching, J. K. S. (1990) A semi-impirical approach to estimate vertical transport by nonprecipitating
convective clouds on a regional scale. Atmos. Environ. 24A: 2153-2168.
AX2-187
-------
Wagner, T.; Platt, U. (1998) Satellite mapping of enhanced BrO concentrations in the troposphere. Nature (London)
395: 486-490.
Wagner, V.; Jenkin, M. E.; Saunders, S. M; Stanton, K.; Wirtz, M.; Pilling, J. (2003) Modelling of the
photooxidation of toluene: conceptual ideas for validating detailed mechanisms. Atmos. Chem. Phys.
3: 89-106.
Wainman, T.; Zhang, J.; Weschler, C. J.; Lioy, P. J. (2000) Ozone and limonene in indoor air: a source of submicron
particle exposure. Environ. Health Perspect. 108: 1139-1145.
Walcek, C. J.; Taylor, G. R. (1986) A theoretical method for computing vertical distributions of acidity and sulfate
production within cumulus clouds. J. Atmos. Sci. 43: 339-355.
Walcek, C. J.; Stockwell, W. R.; Chang, J. S. (1990) Theoretical estimates of the dynamic, radiative and chemical
effects of clouds on tropospheric trace gases. Atmos. Res. 25: 53-69.
Walega, J. G.; Stedman, D. H.; Shelter, R. E.; Mackay, G. I.; Iguchi, T.; Schiff, H. I. (1984) Comparison of a
chemiluminescent and a tunable diode laser absorption technique for the measurement of nitrogen oxide,
nitrogen dioxide, and nitric acid. Environ. Sci. Technol. 18: 823-826.
Wang, Y.; Tao, W.-K.; Pickering, K. E.; Thompson, A. M.; Kain, J. S.; Adler, R. F.; Simpson, J.; Keehn, P. R.; Lai,
G. S. (1996) Mesoscale model simulations of TRACE A and preliminary regional experiment for storm-scale
operational and research meteorology convective systems and associated tracer transport. J. Geophys. Res.
[Atmos.] 101:24,013-24,027.
Wang, Y.; DeSilva, A. W.; Goldenbaum, G. C.; Dickerson, R. R. (1998a) Nitric oxide production by simulated
lightning: dependence on current, energy, and pressure. J. Geophys. Res. [Atmos.] 103: 19,149-19,159.
Wang, Y.; Logan, J. A.; Jacob, D. J. (1998b) Global simulation of tropospheric O3-NOx-hydrocarbon chemistry, 2.
Model evaluation and global ozone budget. J. Geophys. Res. (Atmos.) 103: 10,727-10,755.
Wang, L. H.; Milford, J. B.; Carter, W. P. L. (2000a) Reactivity estimates for aromatic compounds. Part I:
uncertainty in chamber-derived parameters. Atmos. Environ. 34: 4337-4348.
Wang, L. H.; Milford, J. B.; Carter, W. P. L. (2000b) Reactivity estimates for aromatic compounds. Part 2.
uncertainty in incremental reactivities. Atmos. Environ. 4349-4360.
Warneke, C.; Van Der Veen, C.; Luxembourg, S.; De Gouw, J. A.; Kok, A. (2001) Measurements of benzene and
toluene in ambient air using proton-transfer-reaction mass spectrometry: calibration, humidity dependence,
and field intercomparison. Int. J. Mass Spectrom. 207: 167-182.
Wayne, R. P. (2000) Chemistry of Atmospheres: an introduction to the chemistry of the atmospheres of Earth, the
planets, and their satellites. 3rd ed. New York, NY: Oxford University Press, Inc.
Wayne, R. P.; Barnes, I.; Biggs, P.; Burrows, J. P.; Canosa-Mas, C. E.; Hjorth, J.; Le Bras, G.; Moortgat, G. K.;
Perner, D.; Poulet, G.; Restelli, G.; Sidebottom, H. (1991) The nitrate radical: physics, chemistry, and the
atmosphere. Atmos. Environ. Part A 25: 1-203.
Wennberg, P. O.; Hanisco, T. F.; Jaegle, L.; Jacob, D. J.; Hintsa, E. J.; Lanzendorf, E. J.; Anderson, J. G.; Gao,
R.-S.; Keim, E. R.; Donnelly, S. G.; Del Negro, L. A.; Fahey, D. W.; McKeen, S. A.; Salawitch, R. J.;
Webster, C. R.; May, R. D.; Herman, R. L.; Proffitt, M. H.; Margitan, J. J.; Atlas, E. L.; Schauffler, S. M.;
Flocke, F.; McElroy, C. T.; Bui, T. P. (1998) Hydrogen radicals, nitrogen radicals, and the production of O3 in
the upper troposphere. Science (Washington, DC) 279: 49-53.
Wernli, H.; Bourqui, M. (2002) ALagrangian "1-year climatology" of (deep) cross-tropopause exchange in the
extratropical Northern Hemisphere. J. Geophys. Res. (Atmos.) 107(D1-D2): 10.1029/2001JD000812.
Weschler, C. J.; Shields, H. C. (1997) Potential reactions among indoor pollutants. Atmos. Environ. 31: 3487-3495.
Weschler, C. J.; Shields, H. C. (2000) The influence of ventilation on reactions among indoor pollutants: modeling
and experimental observations. Indoor Air. 10: 92-100.
Weschler, C. J.; Hodgson, A. T.; Wooley, J. D. (1992) Indoor chemistry: ozone, volatile organic compounds, and
carpets. Environ. Sci. Technol. 26: 2371-2377.
Wesely, M. L.; Hicks, B. B. (2000) A review of the current status of knowledge on dry deposition. Atmos. Environ.
34: 2261-2282.
Westberg, H.; Zimmerman, P. (1993) Analytical methods used to identify nonmethane organic compounds in
ambient atmospheres. In: Newman, L., ed. Measurement challenges in atmospheric chemistry. Washington,
DC: American Chemical Society, pp. 275-291.
Wexler, A. S.; Sarangapani, R. (1998) Particles do not increase vapor deposition in human airways. J. Aerosol Sci.
29: 197-204.
Whitby, K. T. (1978) The physical characteristics of sulfur aerosols. Atmos. Environ. 12: 135-159.
Wild, O.; Prather, M. J. (2000) Excitation of the primary tropospheric chemical mode in a global CTM. J. Geophys.
Res. (Atmos.) 105: 24,647-24,660.
AX2-188
-------
Wilkins, C. K.; Clausen, P. A.; Wolkoff, P.; Larsen, S. T.; Hammer, M.; Larsen, K.; Hansen, V.; Nielsen, G. D.
(2001) Formation of strong irritants in mixtures of isoprene/ozone and isoprene/ozone/nitrogen dioxide.
Environ. Health Perspect. 109: 937-941.
Williams, E. J.; Guenther, A.; Fehsenfeld, F. C. (1992) An inventory of nitric oxide emissions from soils in the
United States. J. Geophys. Res. [Atmos.] 97: 7511-7519.
Williams, J.; Roberts, J. M.; Fehsenfeld, F. C.; Bertman, S. B.; Buhr, M. P.; Goldan, P. D.; Hubler, G.; Kuster, W.
C.; Ryerson, T. B.; Trainer, M.; Young, V. (1997) Regional ozone frombiogenic hydrocarbons deduced from
airborne measurements of PAN, PPN, and MPAN. Geophys. Res. Lett. 24: 1099-1102.
Wimmers, A. J.; Moody, J. L. (2004a) Tropopause folding at satellite-observed spatial gradients: 1. Verification of
an empirical relationship. J. Geophys. Res. [Atmos.] 109(D19306): 10.1029/2003JD004145.
Wimmers, A. J.; Moody, J. L. (2004b) Tropopause folding at satellite-observed spatial gradients: 2. Development of
an empirical model. J. Geophys. Res. [Atmos.] 109(D19307): 10.1029/2003JD004146.
Wimmers, A. J.; Moody, J. L.; Browell, E. V.; Hair, J. W.; Grant, W. B.; Butler, C. F.; Fenn, M. A.; Schmidt, C. C.;
Li, J.; Ridley, B. A. (2003) Signatures of tropopause folding in satellite imagery. J. Geophys. Res. (Atmos.)
108(04): 10.1029/2001JD001358.
Wingenter, O. W.; Kubo, M. K.; Blake, N. J.; Smith, T. W., Jr.; Blake, D. R.; Rowland, F. S. (1996) Hydrocarbon
and halocarbon measurements as photochemical and dynamical indicators of atmospheric hydroxyl, atomic
chlorine, and vertical mixing obtained during Langrangian flights. J. Geophys. Res. (Atmos.) 101(D1):
4331-4340.
Wingenter, O. W.; Blake, D. R.; Blake, N. J.; Sive, B. C.; Rowland, F. S.; Atlas, E.; Flocke, F. (1999) Tropospheric
hydroxyl and atomic chlorine concentrations, and mixing timescales determined from hydrocarbon and
halocarbon measurements made over the Southern Ocean. J. Geophys. Res. (Atmos.) 104(D17):
21819-21828.
Witter, M.; Berndt, T.; Boge, O.; Stratmann, F.; Heintzenberg, J. (2002) Gas-phase ozonolysis: rate coefficients for a
series of terpenes and rate coefficients and OH yields for 2-methyl-2-butene and 2,3-dimethyl-2-butene. Int. J.
Chem. Kinet. 34: 394-403.
Wolkoff, P.; Clausen, P. A.; Wilkins, C. K.; Hougaard, K. S.; Nielsen, G. D. (1999) Formation of strong airway
irritants in a model mixture of (+)-a-pinene/ozone. Atmos. Environ. 33: 693-698.
World Meteorological Organization. (1990) Scientific assessment of stratospheric ozone: 1989. Volume II.
Appendix: AFEAS report. Geneva, Switzerland: Global Ozone Research Monitoring Project; report no. 20.
Wunderli, S.; Gehrig, R. (1991) Influence of temperature on formation and stability of surface PAN and ozone.
A two year field study in Switzerland. Atmos. Environ. Part A 25: 1599-1608.
Yienger, J. J.; Levy, H., II. (1995) Empirical model of global soil-biogenic NOX emissions. J. Geophys. Res.
100(D6): 11,447-11,464.
Young, V. L.; Kieser, B. N.; Chen, S. P.; Nii, H. (1997) Seasonal trends and local influences on nonmethane
hydrocarbon concentrations in the Canadian boreal forest. J. Geophys. Res. (Atmos.) 102: 5913-5918.
Yu, J.; Jeffries, H. E. (1997) Atmospheric photooxidation of alkylbenzenes II. evidence of formation of epoxide
intermediates. Atmos. Environ. 31: 2281-2287.
Yung, Y. L.; Pinto, J. P.; Watson, R. T.; Sander, S. P. (1980) Atmospheric bromine and ozone perturbations in the
lower stratosphere. J. Atmos. Sci. 37: 339-353.
Zafiriou, O. C.; True, M. B. (1979) Nitrate photolysis in seawaterby sunlight. Mar. Chem. 8: 33-42.
Zanis, P.; Trickl, T.; Stohl, A.; Wernli, H.; Cooper, O.; Zerefos, C.; Gaeggeler, H.; Schnabel, C.; Tobler, L.; Kubik,
P. W.; Priller, A.; Scheel, H. E.; Kanter, H. J.; Cristofanelli, P.; Forster, C.; James, P.; Gerasopoulos, E.;
Delcloo, A.; Papayannis, A.; Claude, H. (2003) Forecast, observation and modelling of a deep stratospheric
intrusion event over Europe. Atmos. Chem. Phys. 3: 763-777.
Zhang, Y.; Carmichael, G. R. (1999) The role of mineral aerosol in tropospheric chemistry in East Asia~a model
study. J. Appl. Meteorol. 38: 353-366.
Zhang, D.; Lei, W. F.; Zhang, R. Y. (2002) Mechanism of OH formation from ozonolysis of isoprene: kinetics and
product yields. Chem. Phys. Lett. 358: 171-179.
Zhou, X.; Mopper, K. (1990) Determination of photochemically produced hydroxyl radicals in seawater and
freshwater. Mar. Chem. 30: 71-88.
Zimmermann, J.; Poppe, D. (1993) Nonlinear chemical couplings in the tropospheric NOX-HOX gas phase chemistry.
J. Atmos. Chem. 17: 141-155.
AX2-189
-------
ANNEX AX3. ENVIRONMENTAL CONCENTRATIONS,
PATTERNS, AND EXPOSURE ESTIMATES
AX3.1 INTRODUCTION
Identification and Use of Existing Air Quality Data
Topics discussed in this annex include the characterization of ambient air quality data for
ozone (O3), the uses of these data in assessing the exposure of vegetation to O3, concentrations
of O3 in microenvironments, and a discussion of the currently available human exposure data and
exposure model development. The information contained in this chapter pertaining to ambient
concentrations is taken primarily from the U.S. Environmental Protection Agency (EPA) Air
Quality System (AQS; formerly the AIRS database). The AQS contains readily accessible
detailed, hourly data that has been subject to EPA quality control and assurance procedures.
Data available in AQS were collected from 1979 to 2001. As discussed in previous versions of
the O3 Air Quality Criteria Document or AQCD (U.S. Environmental Protection Agency, 1986,
1996a), the data available prior to 1979 may be unreliable due to calibration problems and
measurement uncertainties.
As indicated in the 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a), O3 is
the only photochemical oxidant other than nitrogen dioxide (NO2) that is routinely monitored
and for which a comprehensive database exists. Data for peroxyacetyl nitrate (PAN), hydrogen
peroxide (H2O2), and other oxidants either in the gas phase or particle phase typically have been
obtained only as part of special field studies. Consequently, no data on nationwide patterns of
occurrence are available for these non-O3 oxidants; nor are extensive data available on the
relationships of levels and patterns of these oxidants to those of O3. However, available data for
gas-phase and particle-phase oxidants will be discussed.
Characterizing Ambient Ozone Concentrations
The "concentration" of a specific air pollutant is typically defined as the amount (mass) of
that material per unit volume of air. However, most of the data presented in this annex are
expressed as "mixing ratios" in terms of a volume-to-volume ratio (parts per million [ppm] or
parts per billion [ppb]). Data expressed this way are often referred to as concentrations, both in
AX3-1
-------
the literature and in the text, following common usage. Human exposures are expressed in units
of mixing ratio times time.
Several different types of indicators are used for evaluating exposures of vegetation to O3.
The peak-weighted, cumulative exposure indicators used in this chapter for characterizing
vegetation exposures are SUM06 and SUM08 (the sums of all hourly average concentrations
>0.06 and 0.08 ppm, respectively) and W126 (the sum of the hourly average concentrations that
have been weighted according to a sigmoid function that is based on a hypothetical vegetation
response [see Lefohn and Runeckles, 1987]). Further discussion of these exposure indices is
presented in Chapter 9.
The U.S. Environmental Protection Agency (U.S. EPA) has established "ozone seasons"
for the required monitoring of ambient O3 concentrations for different locations within the
United States and U.S. territories (CFR, 2000). Table AX3-1 shows the O3 seasons during which
continuous, hourly averaged O3 concentrations must be monitored. Note that O3 monitoring is
optional outside of the "ozone season" and is monitored in many locations throughout the year.
In Section AX3.2, surface O3 concentrations are characterized and the difficulties of
characterizing background O3 concentrations for controlled exposure studies and for assessing
the health benefits associated with setting the NAAQS are discussed. In addition, hourly
averaged concentrations obtained by several monitoring networks have been characterized for
urban and rural areas. Spatial variations that occur within urban areas, between rural and urban
areas, as well as variations with elevation are discussed in Section AX3.3. The diurnal
variations for the various urban and rural locations are found in Section AX3.4, where urban and
rural patterns are described. In Section AX3.5 seasonal variations in 1-h and 8-h average
concentrations are discussed. Section AX3.6 of this annex summarizes the historical trends for
1980 to 2001 on a national scale and for selected cities. The most recent U.S. EPA trends results
are also presented. Section AX3.7 describes available information for the concentrations and
patterns of related photochemical oxidants. Section AX3.8 describes the co-occurrence patterns
of O3 with NO2, sulfur dioxide (SO2), and 24-h PM25. Indoor O3 concentrations, including
sources and factors affecting indoor O3 concentrations, are described in Section AX3.9.
Section AX3.10 describes human population exposure measurement methods, factors
influencing exposure, and exposure models.
AX3-2
-------
Table AX3-1. Ozone Monitoring Seasons by State
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Start Month — End
March -
April —
January
March -
January
March -
April —
April —
April —
March -
March -
January
April —
April —
April —
April —
April —
March -
January
April —
April —
April —
April —
April —
March -
April —
June —
April —
- October
October
— December
- November
— December
September
September
October
October
- October
- October
— December
October
October
September
October
October
- October
— December
September
October
September
September
October
- October
October
September
October
State
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico
Rhode Island
South Carolina
South Dakota
Tennessee
Texas1
Texas1
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
American Samoa
Guam
Virgin Islands
Start Month — End
January
April —
April —
January
April —
April —
May —
April —
— December
September
October
— December
October
October
September
October
March — November
May —
April —
January
April —
April —
June —
September
October
— December
September
October
September
March — October
January
— December
March — October
May —
April —
April —
May —
April —
April 15
April —
January
January
January
September
September
October
September
October
— October 15
October
— December
— December
— December
ozone season is defined differently in different sections of Texas.
Source: CFR(2000).
AX3-3
-------
AX3.2 SURFACE OZONE CONCENTRATIONS
Data for O3 concentrations in a number of different environments, ranging from urban to
remote, are summarized and characterized in this section. The main emphasis is placed on the
characterization of the variability of O3 concentrations in these different environments. Another
important issue relates to the determination of background concentrations. There are a number
of different uses of the term background depending on the context in which it is used. Various
definitions of background have been covered in the 1996 O3 AQCD (U.S. Environmental
Protection Agency, 1996a) and in Air Quality Criteria for Particulate Matter (PM AQCD; U.S.
Environmental Protection Agency, 1996b). This section deals with the characterization of
background O3 concentrations that are used for two main purposes: (1) performing experiments
relating the effects of exposure to O3 on humans, animals, and vegetation; and (2) assessing the
health benefits associated with setting different levels of the NAAQS for O3. Ozone background
concentrations used for NAAQS setting purposes are referred to as policy relevant background
(PRB) concentrations. PRB concentrations are defined by the U.S. EPA Office of Air Quality
Programs and Standards (OAQPS) as those concentrations that would be observed in the United
States if anthropogenic sources of O3 precursors were turned off in continental North America
(the United States, Canada and Mexico), i.e., the definition includes O3 formed from natural
sources everywhere in the world and from anthropogenic O3 precursors outside of North
America. The 1996 O3 AQCD considered two possible methods for quantifying background O3
concentrations for the two purposes mentioned above. The first method relied on mathematical
models and historical data. The second method used the distribution of hourly average O3
concentrations observed at clean, relatively remote monitoring sites (RRMS) in the United States
(i.e., those which experience low maximum hourly concentrations). At the time of the 1996 O3
AQCD, simulations of mathematical models were limited; therefore, the second method was
employed to quantify background O3 concentrations.
Sections AX3.2.1 and AX3.2.2 review data for O3 concentrations in urban and nonurban
(but influenced by urban emissions) environments. Section AX3.2.3 reviews the data from
relatively clean remote sites, addresses the issue of how to use these data to help set background
levels for controlled exposure studies, and presents evidence of trends in O3 concentrations at
these sites. The characterization of PRB O3 concentrations will be the subject of
Section AX3.2.4. Two alternative approaches for establishing PRB concentrations are
AX3-4
-------
presented: the first uses data from relatively clean, remote monitoring sites and the second uses
numerical models. The strengths and weaknesses of each approach are presented in the hopes of
stimulating discussion that will resolve issues related to the use of either of these alternative
methods.
Ozone Air Quality at Urban, Suburban, andNonurban Sites
Figure AX3-1 shows the mean daily maximum 8-h O3 concentrations and Figure AX3-2
shows the 95th percentile values of the daily maximum 8-h O3 concentrations, based on
countywide site-wise averages across the United States from May to September 2000 to 2004.
The locations of the monitoring sites, whose data were used in the calculation of background O3
concentrations, are shown in Figure AX3-3. The period from May to September was chosen
because, although O3 was monitored for different lengths of time across the country, all O3
monitors should be operational during these months. Data flagged because of quality control
issues were removed with concurrence of the local monitoring agency. Only days for which
there were 75% complete data (i.e., 18 of 24 hours) were kept, and a minimum of 115 of
153 days were required in each year. Cut points for the tertile distributions on each map were
chosen at the median and 95th percentile values. These cut points were chosen as they represent
standard metrics for characterizing important aspects of human exposure used by the EPA. Any
other percentiles or statistics that are believed to be helpful for characterizing human exposures
could also be used. Blank areas on the maps indicate no data coverage. It should be noted that
county areas can be much larger in the West than in the East, but monitors are not spread evenly
within a county. As a result, the assigned concentration range might not represent conditions
throughout a particular county and so large areas in western counties where there are not any
monitors were blanked out.
As shown in Figure AX3-1, the median of the countywide, mean daily maximum 8-h O3
concentration across the United States is 49 ppb, and 5% of these site-wise means exceeded
57 ppb. Although the median and 95th percentile value of the countywide means are fairly
close, these results cannot be taken to imply that mean O3 concentrations lie in a relatively
narrow range throughout the United States, because data coverage is not as complete in the West
as it is in the East. High mean daily maximum 8-h O3 concentrations are found in California and
states in the Southwest as well as in several counties in the East. As shown in Figure AX3-2, the
AX3-5
-------
Seasonal (May-September) Mean of Daily Maximum 8-Hour Values, 2002-2004
Concentration PPM
X < 0.049
0.049 £ X < 0.057
0.057 < X
Figure AX3-1. Countywide mean daily maximum 8-h O3 concentrations, May to
September 2002 to 2004.
Source: Fitz-Simons et al. (2005).
nationwide median of the county wide, 95th percentile value of the daily maximum 8-h O3
concentration is 73 ppb and 5% of these values are above 85 ppb. High values for the 95th
percentiles are found in California, Texas, and in the East, but not necessarily in the same
counties as shown for the mean daily maximum 8-h concentrations in Figure AX3-1.
Although mean O3 concentrations in Houston, TX were below the nationwide median, its
95th percentile value ranks in the highest 5% nationwide. Conversely, mean O3 concentrations
in southwestern states are among the highest in the United States, but peak values (i.e., 95th or
98th percentile values) in those counties are not among the highest peak values in the United
States. In other areas where the highest mean O3 concentrations occurred, such as California;
Dallas-Fort Worth, TX; and the Northeast Corridor, the highest peak values also were observed.
AX3-6
-------
Seasonal (May-September) 95th Percentile of Daily Maximum 8-Hour Values, 2002-2004
Concentration PPM
X < 0.073
0.073 < X < 0.085
0.085 < X
Figure AX3-2. Countywide 95th percentile value of daily maximum 8-h O3 concentrations,
May to September 2002 to 2004.
Source: Fitz-Simons et al. (2005).
Although countywide averages are shown, it should be noted that considerable spatial
variability can exist within a county, especially within urban areas as described in
Section AX3.3. In addition, there can also be differences in the diurnal profile of O3 among
monitors within counties.
Box plots showing the percentile distribution of nationwide O3 concentrations for
different averaging periods (1-h daily maximum, 8-h daily maximum and 24-h daily average) are
given in Figures AX3-4 to AX3-6 and the numerical values are given in Table AX3-2. The
differences between the 50th and 95th percentile values indicate the range of O3 levels between
"typical" O3 days and "high" O3 days. These differences are approximately 40, 30, and 25 ppb
AX3-7
-------
Figure AX3-3. Locations of monitoring sites used for calculating countywide averages
across the United States.
Source: Fitz-Simons et al. (2005).
for the daily 1-h, 8-h, maxima and daily averaged O3 concentrations. As might be expected, the
daily maximum 1-h and 8-h O3 concentrations are highly correlated.
Lehman et al. (2004) have shown that the eastern United States can be divided into five
regions, each of which exhibit spatial, relatively coherent spatial patterns of O3 at nonurban sites.
Only sites classified as being rural or suburban and with land usage of forest, agriculture, or
residential were included in the analyses. These criteria were chosen to avoid sites where O3 is
scavenged by NO that can be found in high concentrations near major sources, such as traffic in
urban cores. The five regions, shown in Figure AX3-7, are characterized by different patterns
of O3 properties such as temporal persistence and seasonal variability. Figure AX3-7 shows
nonurban, monthly average, daily maximum 8-h O3 concentrations in the five regions in the
eastern United States from April to October 1993 to 2002.
AX3-8
-------
1 Hour Daily Ozone Maximum Concentrations
0.24 r
0.22 -.
| 0.20 -
Q.
& 0.18-
0 0.16 -!
« 0.14 -
+•« :
g 0.10-
C 0.06 '-
O
N 0.04 -
0.02 \
0.00 '-
691012 367121 323891
X X
£ x_
_jj
p
t- -C
X
1- M
T t f
All Values
In a CSA
Not in a CSA
Figure AX3-4.
Distribution of nationwide daily maximum 1-h average O3 concentrations
from May to September 2000 to 2004. Medians, interquartile ranges,
minima and maxima and means (as dots) are shown. Numbers above the
boxes are the number of observations.
Source: Fitz-Simons et al. (2005).
8 Hour Daily Ozone Maximum Concentrations
0.16 :
•=• 0.14 :
Q. :
3 0.12 :
C ;
5 0.10 :
E ;
g 0-08 :
o
O 0 06
o
c 0.04 -
O :
O 0.02 -
0.00 -
6908
X
^
44 367
£
i_ -•
329 323
L. _jm
815
t T f
All Values
In a CSA
Not in a CSA
Figure AX3-5. Distribution of nationwide daily maximum 8-h average O3 concentrations
from May to September 2000 to 2004. Medians, interquartile ranges,
minima and maxima and means (as dots) are shown. Numbers above the
boxes are the number of observations.
Source: Fitz-Simons et al. (2005).
AX3-9
-------
24 Hour Average Ozone Concentrations
E
Q.
C
o
C
0)
o
C
o
o
0)
C
o
N
o
0.12 7
0,11 \
0.10 -
0.09 -
0.08 -!
0.07 -
0.06 -i
0.05 ^
0.04 ~
0.03 ^
0.02 -i
0.01 ^
0.00 '-
691
s
012 367
1
t
J
1
121 323
>
|
i_ j
891
<
I
l_
r f
All Values
In a CSA
Not in a CSA
Figure AX3-6. Distribution of nationwide 24-h average O3 concentrations from May to
September 2000 to 2004. Medians, interquartile ranges, minima and
maxima and means (as dots) are shown. Numbers above the boxes are the
number of observations.
Source: Fitz-Simons et al. (2005).
Regional differences are immediately apparent. Highest concentrations among all the
regions are generally found in the Mid-Atlantic region (mean of 52 ppb) with highest values
throughout the ozone concentration distribution except for the overall maximum. Lowest mean
concentrations (42 ppb) are found in Florida. In the northern regions (the Northeast, Great
Lakes) and the Mid-Atlantic region, highest median and peak concentrations are found in July,
whereas in the Southwest, highest median concentrations are found in August, with highest
peaks in June and September, i.e., outside the warmest summer months. In Florida, highest
monthly averaged median and peak concentrations are found during the spring. High O3
concentrations tend to be most persistent (3-4 days of persistence) in the southern regions, less
persistent in the Mid-Atlantic region (2-3 days) and least persistent in the northern regions (1 or
2 days). Such analyses could not be made for the western United States, in part because of the
presence of more extensive mountain ranges.
Box plots showing the distributions of hourly average O3 concentrations for different types
of rural sites for 2001 are given in Figures AX3-8 (rural-agricultural), AX3-9 (rural-forest),
AX3-10
-------
Table AX3-2. Summary of Percentiles of Data Pooled Across Monitoring Sites for May to September 2000-2004
Concentrations are in ppb.
Percentiles
Pooled Group/ Number
Avg. Time of Values Mean 1
10 25 30 50 70 75 90 95 99
Daily 1-h Maximum Concentrations
Monitors in 367,121 58 20
CSAs
Monitors not 323,891 55 20
in CSAs
29 34 44 46 56 66 70 84 94 116
28 33 43 45 54 64 67 79 87 104
X
OJ
8-h Daily Maximum Concentrations
Monitors in 367,029 50
CSAs
Monitors not 323,815 49
in CSAs
16 23 28 37
16 23
28
40
49 58 61 73 81 98
37 39 48 57 59 70 77 91
24-h Average Concentrations
Monitors in 367,121 33
CSAs
Monitors not 323,891 34
in CSAs
10 15 18 24 26 32 39 41 50 56 68
10 15 18 25 27 33 39 41 50 56 68
-------
13
£L
Q.
180-
160.
140-
120 •
100.
80.
60.
40-
20.
0
Great Lakes Region
JUN JUL
Month
a
Q.
Q.
180-
160.
140-
120.
100
80-
60.
40-
20-
0
Northeast Region
H
JUN JUL
Month
180-
160-
lid-
s' 120-
Q.
n
•s 100-
Southwest Region
o
N
O
80.
60-
40-
20.
0-
JUN JUL
Month
o
N
O
180-
160
140-
120-
100
80-
60-
40-
20-
0-
Mid-Atlantic Region
JUN JUL
Month
180-
160
140-
.Q 120
a.
Q.
— 100-
0)
N
O
60-
40.
20.
0-
Florida Region
JUN JUL
Month
AUG
SEP
OCT
Figure AX3-7. Box plots showing O3 averaged by month from 1993 to 2002 in the five
regions in the eastern United States derived by Lehman et al. (2004). The
boxes define the interquartile range and the whiskers, the extreme values.
Source: Lehman et al. (2004).
AX3-12
-------
Rural Agriculture
0.18-
0.17-
0.16-
—. 0.15-
| 0.14-
3 0.13-
c 0.12-
.2 0.11 -
2 0.10-
e 0.09 -
S 0.08-
§ 0.07 -
U 0.06 -
g 0.05 -
S 0.04-
O 0.03 -
0.02-
0.01 -
0.00-
No. of Hours -
Hours > 0.08 -
Hours a 0.10-
SUM06 (ppm-h) -
W126(ppm-h)-
1 1 1 n i
r~i i~i
JL
_ - _ U ~
I I T-*
y y y i x b
I I
H
3 i
5076 4367 5059 5116 5080 5055 4345
21 37 142 1 156 48 59
0 0 34 0 6
1
24.7 34.2 43.5 8.6 63.2 27.4 24.8
20.7 25.8 35.7 10.6 50.0 22.7 22.8
0°''
Figure AX3-8. Hourly average O3 concentrations observed at selected rural-agricultural
sites from April to October 2001. The whiskers represent minimum and
maximum concentrations. The boxes represent the 10th and 90th
percentile concentrations.
Source: Fitz-Simons et al. (2005).
and AX3-10 (rural-residential or commercial). Shown below the figures are the number of
observations and various metrics for characterizing vegetation exposures. Note that high O3
concentrations are found at sites that are classified as rural, as in Anne Arundel Co., MD;
Yosemite NP, CA; and Crestline, CA. Land use designations might not give an accurate picture
of exposure regimes in rural areas, because the land use characterization of "rural" does not
necessarily mean that a specific location is isolated from anthropogenic influences. Rather, the
characterization refers only to the current use of the land, not to the presence of sources.
Since O3 produced from emissions in urban areas is transported to more rural downwind
locations, elevated O3 concentrations can occur at considerable distances from urban centers.
In addition, major sources of O3 precursors such as power plants and highways are located in
nonurban areas and also produce O3 in these areas. Due to lower rates of chemical scavenging in
AX3-13
-------
Rural Forest
0.18-
0.17-
0.16-
—. 0.15-
| 0.14-
0. 0.13-
c 012-
~ 0.11-
2 o. ID-
'S 0.09-
0 0.08 -
0 0.07 -
O 0.06 -
g 0.05-
g 0.04 -
O 0.03 -
0.02-
0.01 -
0.00-
No. of Hours -
HoursaO.08-
Hours>0.10-
SUM06(ppm-h)-
W126 (ppm-h)-
T T T T
I I I
I I i | I
T P I V
— I—
4619 4536 5074 4813
106 77 267 157
6454
86.0 44.7 149.2 98.5
65.6 37.7 106.5 73.6
Figure AX3-9. Hourly average O3 concentrations observed at selected rural-forest sites
from April to October 2001. The whiskers represent minimum and
maximum concentrations. The boxes represent the 10th and 90th
percentile concentrations.
Source: Fitz-Simons et al. (2005).
nonurban areas, O3 tends to persist longer in nonurban than in urban areas, also tending to lead to
higher exposures in nonurban areas influenced by anthropogenic precursor emissions.
Ozone Air Quality Data at Relatively Remote Monitoring Sites (RRMS)
RRMS are sites that are located in the national parks that tend to be less affected by
obvious pollution sources than other sites. This does not mean that they are completely
unaffected by local pollution, because of the large number of visitors to these national parks.
It is important to characterize hourly average O3 concentrations at RRMS so that assessments of
the possible effects of O3 on human health and vegetation, use concentration ranges that mimic
the range found in the United States. Hourly average concentrations used as controls in studies
AX3-14
-------
Rural Other
8
o
0.18-
0.17-
0.16-
0.15-
0.13-
0.12-
0.11 -
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02-
0.01 -
0.00-
ours-
0.08-
0.10-
m-h) -
m-h) -
-,—
—
i D
4922 5088
933 9
369 0
174.8 21.4
144.9 20.3
-j—
-^-i
J
4879
111
10
45.4
334.4
Figure AX3-10. Hourly average O3 concentrations observed at selected rural-commercial
or -residential sites from April to October 2001. The whiskers represent
minimum and maximum concentrations. The boxes represent the 10th
and 90th percentile concentrations.
Source: Fitz-Simons et al. (2005).
of human health and vegetation exposures to O3 appear to be lower than those experienced at
RRMS in the United States or in other parts of the world (see Chapter 9). Typically, ambient air
is filtered to remove O3 before being admitted into the exposure chambers. As a result, O3
concentrations might only be a few ppb within these chambers.
Box plots in Figures AX3-1 la-d depicting the distributions of annual hourly average
measured O3 concentrations at four relatively remote monitoring sites (RRMS), show that annual
mean values of the daily 8-h maximum O3 concentration have not changed much over the past
10 years of available data. Mean values range typically from about 0.020 ppm to about
0.040 ppm. Concentrations only rarely exceed 0.080 ppm, in contrast to observations at other
"rural" sites shown in Figures AX3-8 to AX3-10.
AX3-15
-------
a. Theodore Roosevelt National Park
b. Yellowstone National Park
0.16- 3525 3162 3629 1808 3649 3651 3652 3672 3199 3667
0.14
0.12
0.06-
0.02-
0.00-
I I I I I I I I I I
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
0.16- 3276 3263 3408 3444 3385 3333 3478 3311 3343 3402
0.12-
0.04-
I I I I I I I I I I
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
c. Glacier National Park
d. Olympic National Park
0.16- 3365 3317 3169 3277 3232 3230 3320 3311 3157 3333
0.12
0.02-
0.00
I I I I I I I I I I
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
0.16- 3321 3438 3095 3416 3347 3351 3304 3361 3361 3276
0.10-
0.04-
0.02-
0.00-
I I I I I I I I I I
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Figure AX3-lla-d. Daily 8-h maximum O3 concentrations observed at selected national
park sites. The whiskers on the box plot represent the 10th and
90th percentile concentrations. The "X"s above and below the
whiskers are the values that fall below and above the 10th and 90th
percentile concentrations. The dots inside the box represent the
mean. The number of observations is shown above each box plot.
Source: Fitz-Simons et al. (2005).
AX3-16
-------
It is unlikely that distributions found at sites with low maximum hourly average
concentrations in the western United States represent those at sites in the eastern and midwestern
United States because of regional differences in sources of precursors and transport patterns.
Given the high density of sources in the eastern and midwestern United States, it is unclear
whether a site could be found in either of these regions that would not be influenced by the
transport of O3 from nearby urban areas. Thus, with the exception of the Voyageurs NP site in
Minnesota, observations at RRMS are limited to those obtained in the western United States.
However, not all national park sites in the West can be considered to be free of strong regional
pollution influences, e.g., Yosemite NP (CA), as can be seen from Figure AX3-9.
The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) concluded that the
annual average "background" concentration of O3 near sea level ranged from 0.020 to 0.035 ppm
and that, during the summer, the 1-h daily maximum ranged from 0.03 to 0.05 ppm. The
1996 O3 AQCD also included O3 hourly average concentrations measured at several clean,
RRMS mostly located in the western United States. Table AX3-3 provides a summary of the
characterization of the hourly average concentrations recorded from 1988 to 2001 at some of the
monitoring sites previously analyzed. The percentile distribution of the hourly average
concentrations (April to October), number of hourly average occurrences >0.08 and >0.10 ppm,
seasonal 7-h average concentrations, the SUM06, and W126 values were characterized for those
site years with a data capture of >75%. From 1988 to 2001, no hourly average concentrations
>0.08 ppm were observed at monitoring sites in Redwood NP (CA), Olympic NP (WA), Glacier
NP (MT), Denali NP (AK), Badlands (SD), and Custer NF (MT) during the months of April to
October. There were eight occurrences of hourly average O3 concentrations >0.08 ppm from
April to October of 1997 at the monitoring site in Theodore Roosevelt NP (ND). However, no
hourly average concentrations >0.08 ppm were observed from April to October in any other year
at this site. Except for 1988, the year in which there were major forest fires at Yellowstone NP
(WY), the monitoring site located there experienced no hourly average concentrations
>0.08 ppm. Logan (1989) noted that O3 hourly average concentrations rarely exceed 0.08 ppm
at remote monitoring sites in the western United States. In almost all cases for the above sites,
the maximum hourly average concentration was <0.075 ppm. The top 10 daily maximum 8-h
average concentrations for sites experiencing low maximum hourly average concentrations with
AX3-17
-------
Table AX3-3. Seasonal (April to October) Percentile Distribution of Hourly Ozone Concentrations (ppm), Number of
Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126 Values for Sites
Experiencing Low Maximum Hourly Average Concentrations with Data Capture of >75%
X
OJ
1
oo
Site Year
Redwood NP 1988
060150002
(California) 1989
235m
1990
1991
1992
1993
1994
Olympic NP 1989
(Washington)
530090012 1990
125m
1991
1993
1994
1995
1996
1997
1998
1999
2000
2001
Min.
0.002
0.000
0.000
0.001
0.000
0.000
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.002
10
0.011
0.010
0.011
0.012
0.010
0.010
0.011
0.003
0.005
0.006
0.004
0.006
0.006
0.006
0.005
0.008
0.006
0.006
0.009
30
0.018
0.017
0.018
0.019
0.017
0.017
0.018
0.010
0.012
0.014
0.010
0.013
0.014
0.013
0.010
0.014
0.014
0.013
0.017
50
0.023
0.022
0.023
0.025
0.021
0.022
0.024
0.015
0.018
0.019
0.016
0.019
0.020
0.019
0.015
0.019
0.019
0.019
0.023
Percentiles
70
0.029
0.027
0.028
0.031
0.026
0.027
0.028
0.022
0.023
0.024
0.021
0.025
0.027
0.025
0.022
0.025
0.026
0.025
0.028
90
0.038
0.034
0.035
0.038
0.035
0.035
0.035
0.030
0.030
0.033
0.029
0.033
0.037
0.034
0.035
0.033
0.036
0.035
0.036
95
0.041
0.038
0.038
0.041
0.039
0.038
0.038
0.035
0.034
0.036
0.034
0.038
0.040
0.038
0.040
0.037
0.039
0.039
0.041
99
0.046
0.042
0.043
0.045
0.045
0.042
0.043
0.046
0.043
0.044
0.041
0.043
0.048
0.043
0.046
0.044
0.044
0.045
0.046
No. of
Max. Obs.
0.06
0.047
0.053
0.054
0.055
0.054
0.050
0.065
0.064
0.056
0.064
0.062
0.077
0.058
0.057
0.063
0.050
0.061
0.055
4825
4624
4742
4666
4679
4666
4846
4220
4584
4677
4595
4044
4667
4811
4403
4792
4656
4676
4643
Hours
iO.08 iO.10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Seasonal
7-h
0.026
0.024
0.025
0.027
0.023
0.025
0.026
0.021
0.022
0.025
0.022
0.025
0.027
0.025
0.024
0.024
0.024
0.027
SUM06
(ppm-h)
1.8
1.0
1.2
1.7
1.1
1.1
0.0
0.7
0.8
0.9
0.7
0.2
0.8
0.0
0.3
0.0
0.1
0.0
W126
(ppm-h)
0.1
0.0
0.0
0.0
0.0
0.0
1.2
0.1
0.3
0.0
0.3
0.8
1.9
1.0
1.1
1.1
1.2
1.4
-------
>
X
Table AX3-3 (cont'd). Seasonal (April to October) Percentile Distribution of Hourly Ozone Concentrations (ppm), Number
of Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126 Values for
Sites Experiencing Low Maximum Hourly Average Concentrations with Data Capture of >75%
Site
Glacier NP
300298001
(Montana)
963m
Yellowstone NP
(Wyoming)
560391010
2484m
Year
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
1988
1989
1990
1991
1992
1993
1994
1995
Min.
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.002
0.002
0.000
0.004
0.001
0.000
0.003
0.004
10
0.003
0.003
0.001
0.001
0.000
0.001
0.000
0.002
0.000
0.003
0.002
0.001
0.000
0.020
0.018
0.015
0.020
0.018
0.018
0.022
0.022
30
0.015
0.014
0.014
0.013
0.010
0.014
0.010
0.013
0.008
0.013
0.015
0.011
0.013
0.029
0.027
0.023
0.030
0.029
0.028
0.033
0.033
50
0.026
0.026
0.027
0.025
0.020
0.026
0.022
0.025
0.017
0.025
0.026
0.023
0.025
0.037
0.036
0.029
0.037
0.036
0.036
0.040
0.040
Percentiles
70
0.036
0.035
0.036
0.033
0.029
0.036
0.031
0.035
0.026
0.035
0.035
0.033
0.033
0.044
0.044
0.036
0.042
0.042
0.042
0.046
0.045
90
0.046
0.044
0.046
0.043
0.040
0.046
0.041
0.046
0.041
0.047
0.046
0.044
0.042
0.054
0.052
0.043
0.048
0.051
0.047
0.053
0.052
95
0.050
0.047
0.049
0.048
0.044
0.050
0.045
0.051
0.045
0.051
0.051
0.048
0.044
0.058
0.057
0.046
0.051
0.056
0.050
0.056
0.055
99
0.058
0.052
0.056
0.055
0.050
0.056
0.051
0.058
0.053
0.058
0.058
0.055
0.049
0.070
0.063
0.053
0.057
0.064
0.054
0.062
0.059
No. of
Max. Obs.
0.067
0.066
0.062
0.077
0.058
0.061
0.066
0.065
0.058
0.064
0.068
0.062
0.057
0.098
0.071
0.061
0.064
0.075
0.060
0.072
0.065
4770
5092
5060
4909
5071
5072
4744
4666
4378
4649
4540
4551
4643
4257
4079
4663
4453
4384
4399
4825
4650
Hours
iO.08 iO.10
0
0
0
0
0
0
0
0
0
0
0
0
0
17
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Seasonal
7-h
0.036
0.036
0.036
0.033
0.029
0.036
0.023
0.035
0.027
0.036
0.035
0.033
0.033
0.043
0.042
0.034
0.042
0.042
0.041
0.046
0.045
SUM06
(ppm-h)
5.9
4.1
5.3
4.1
0.0
0.1
0.3
1.9
0.0
1.4
1.3
0.7
0.0
14.0
11.0
3.8
7.7
10.7
6.5
6.0
2.8
W126
(ppm-h)
1.8
1.3
0.7
1
2.3
5.4
2.3
5.4
2.3
5.6
5.4
3.8
2.7
8.9
6.7
0.5
1.2
6.3
0.2
15.2
12.5
-------
Table AX3-3 (cont'd). Seasonal (April to October) Percentile Distribution of Hourly Ozone Concentrations (ppm), Number
of Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126 Values for
Sites Experiencing Low Maximum Hourly Average Concentrations with Data Capture of >75%
Site
Yellowstone NP
(Wyoming)
560391011
2468m
Denali NP
(Alaska)
022900003
640m
X
OJ
1
to
o
Badlands NP
460711001
(South Dakota)
730m
Year
1997
1998
1999
2000
2001
1988
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
1989
1990
1991
Min.
0.005
0.004
0.012
0.009
0.012
0.003
0.003
0.005
0.003
0.002
0.003
0.001
0.002
0.001
0.004
0.002
0.003
0.002
0.006
0.006
0.005
10
0.026
0.029
0.033
0.031
0.034
0.018
0.017
0.018
0.016
0.017
0.017
0.013
0.015
0.015
0.018
0.016
0.014
0.016
0.020
0.019
0.020
30
0.035
0.038
0.040
0.039
0.041
0.024
0.024
0.024
0.023
0.023
0.022
0.019
0.022
0.023
0.023
0.024
0.019
0.023
0.027
0.027
0.028
50
0.040
0.043
0.046
0.045
0.046
0.028
0.029
0.028
0.028
0.028
0.027
0.025
0.028
0.030
0.030
0.029
0.025
0.029
0.034
0.032
0.033
Percentiles
70
0.045
0.048
0.051
0.050
0.050
0.033
0.034
0.034
0.034
0.033
0.033
0.032
0.035
0.038
0.036
0.036
0.029
0.036
0.041
0.037
0.040
90
0.051
0.055
0.059
0.057
0.057
0.044
0.040
0.041
0.044
0.041
0.042
0.042
0.044
0.045
0.048
0.045
0.034
0.048
0.049
0.044
0.047
95
0.054
0.058
0.062
0.060
0.060
0.050
0.043
0.043
0.047
0.043
0.045
0.044
0.047
0.048
0.050
0.048
0.036
0.051
0.053
0.048
0.050
99
0.060
0.064
0.069
0.065
0.065
0.053
0.048
0.047
0.050
0.048
0.049
0.052
0.052
0.051
0.055
0.054
0.038
0.055
0.060
0.054
0.056
No. of
Max. Obs.
0.068
0.073
0.079
0.074
0.078
0.056
0.050
0.057
0.054
0.055
0.053
0.059
0.063
0.084
0.058
0.058
0.049
0.068
0.071
0.063
0.066
4626
4827
4733
4678
4869
4726
3978
4809
4800
4773
4807
4825
4831
4053
4782
4868
4641
4868
4840
4783
4584
Hours
iO.08 iO.10
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Seasonal
7-h
0.043
0.046
0.049
0.047
0.048
0.031
0.030
0.030
0.031
0.030
0.030
0.028
0.031
0.032
0.032
0.032
0.025
0.032
0.040
0.037
0.038
SUM06
(ppm-h)
3.3
9.9
27.1
17.0
16.9
0.0
2.1
2.7
3.7
2.6
0.0
0.0
0.1
0.2
0.0
0.0
0.0
0.7
9.2
4.8
6.2
W126
(ppm-h)
12.4
20.0
29.8
23.4
25.6
4.0
0.0
0.0
0.0
0.0
2.9
3.0
4.1
4.0
6.0
4.7
1.0
11.1
3.1
0.8
0.7
-------
Table AX3-3 (cont'd). Seasonal (April to October) Percentile Distribution of Hourly Ozone Concentrations (ppm), Number
of Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126 Values for
Sites Experiencing Low Maximum Hourly Average Concentrations with Data Capture of >75%
Percentiles
Site
Theod. Roos.
NP
380530002
(North Dakota)
730m
>
X
OJ
1
hJ
Theod. Roos.
NP
380070002
(North Dakota)
808m
Custer NF, MT
300870101
(Montana)
1006m
Year
1984
1985
1986
1989
1992
1993
1994
1995
1996
1997
1999
2000
2001
1978
1979
1980
1983
Min.
0.000
0.000
0.004
0.004
0.005
0.004
0.000
0.000
0.003
0.000
0.007
0.002
0.002
0.000
0.010
0.010
0.010
10
0.017
0.019
0.017
0.023
0.019
0.018
0.018
0.018
0.022
0.016
0.024
0.021
0.023
0.010
0.025
0.025
0.025
30
0.025
0.026
0.027
0.032
0.027
0.025
0.028
0.028
0.031
0.029
0.031
0.031
0.031
0.020
0.035
0.035
0.035
50
0.032
0.032
0.033
0.039
0.033
0.031
0.035
0.035
0.037
0.037
0.037
0.036
0.036
0.035
0.040
0.040
0.040
70
0.039
0.038
0.039
0.045
0.039
0.037
0.041
0.041
0.043
0.044
0.042
0.043
0.042
0.040
0.045
0.050
0.045
90
0.047
0.046
0.047
0.054
0.047
0.045
0.049
0.050
0.051
0.053
0.049
0.050
0.049
0.050
0.050
0.055
0.05
95
0.050
0.049
0.050
0.058
0.050
0.048
0.052
0.053
0.054
0.058
0.052
0.053
0.052
0.055
0.055
0.060
0.055
99
0.059
0.054
0.056
0.065
0.056
0.055
0.058
0.058
0.059
0.069
0.058
0.058
0.058
0.060
0.060
0.065
0.060
No. of
Max. Obs.
0.068
0.061
0.062
0.073
0.063
0.064
0.079
0.064
0.064
0.082
0.070
0.066
0.064
0.070
0.075
0.070
0.065
4923
4211
4332
4206
4332
4281
4644
4242
3651
4344
5105
5105
5099
4759
5014
4574
4835
Hours
iO.08
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
iO.10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Seasonal
7-h
0.038
0.038
0.039
0.046
0.040
0.038
0.041
0.042
0.044
0.046
0.041
0.041
0.041
0.033
0.043
0.043
0.042
SUM06
(ppm-h)
7.0
5.0
5.5
14.2
6.1
4.6
1.1
1.2
1.8
11.8
1.6
2.3
1.9
3.0
7.3
22.4
4.2
W126
(ppm-h)
2.8
0.1
0.4
11.0
0.8
0.7
8.4
7.7
8.5
14.6
10
10.5
9.2
8.3
13.2
19.7
10.7
-------
a data capture of >75% are summarized in Table AX3-4. The highest 8-h daily maximum
concentrations do not necessarily all occur during the summer months. For example, at the
Yellowstone National Park site, the three highest 8-h daily maximum concentrations occurred in
April and May in 1998, and the fourth highest, 8-h daily maximum concentration did not occur
until July of that year. In 1999, the three highest, 8-h daily maximum concentrations were
observed in March and May, and the fourth highest value occurred in April. In 2000, the four
highest values occurred in May, June, July, and August.
The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) noted that the
7-month (April to October) average of the 7-h daily average concentrations (0900 to 1559 hours)
observed at the Theodore Roosevelt National Park monitoring site in North Dakota were 0.038,
0.039, and 0.039 ppm, respectively, for 1984, 1985, and 1986 and concluded that the range of
7-h seasonal averages for the Theodore Roosevelt National Park site was representative of the
range of maximum daily 8-h average O3 concentrations that may occur at other fairly clean sites
in the United States and other locations in the Northern Hemisphere. However, as shown in
Table AX3-4, the representative (as given by the fourth highest) daily maximum 8-h average O3
concentrations at fairly clean sites in the United States are higher than the 0.038 and 0.039 ppm
values cited in the 1996 O3 AQCD, and these updated values should be used to characterize O3 at
these sites.
As described in the 1996 O3 AQCD, the O3 monitoring site in the Ouachita National Forest,
AR experienced distributions of hourly average concentrations similar to some of the western
sites. However, since 1993, this site has seen significant shifts, both increases and decreases, in
hourly average concentrations. Figure AX3-12 shows the changes that have occurred from 1991
to 2001. The large changes in hourly average O3 concentrations observed at the Ouachita
National Forest may indicate that this rural site is influenced by the transport of pollution. Given
the high density of sources in the eastern and midwestern United States, it is unclear whether a
site could be found in either of these regions that would not be influenced by the transport of O3
from urban areas. Thus, with the exception of the Voyageurs National Park site, observations in
this section are limited to those obtained at relatively clean, remote sites in western North
America.
AX3-22
-------
Table AX3-4. The Top 10 Daily Maximum 8-h Average Concentrations (ppm) for Sites Experiencing Low Maximum
Hourly Average Concentrations with Data Capture of >75%
Site Year
Redwood NP 1988
060150002
(California) 1989
235 m 199Q
1991
1992
1993
1994
Olympic NP 1989
530090012
^ (Washington) 199°
S 125m 1991
to
w 1993
1994
1995
1996
1997
1998
1999
2000
2001
1
0.061
0.044
0.051
0.048
0.060
0.049
0.048
0.054
0.056
0.050
0.055
0.050
0.064
0.046
0.052
0.051
0.045
0.051
0.051
2
0.058
0.043
0.048
0.047
0.053
0.046
0.048
0.052
0.048
0.048
0.052
0.046
0.063
0.046
0.051
0.050
0.044
0.051
0.050
3
0.053
0.043
0.048
0.046
0.045
0.043
0.046
0.047
0.046
0.045
0.044
0.042
0.050
0.046
0.046
0.049
0.044
0.048
0.047
4
0.052
0.043
0.047
0.045
0.045
0.043
0.046
0.044
0.046
0.043
0.042
0.042
0.049
0.046
0.045
0.046
0.043
0.047
0.045
5
0.049
0.042
0.047
0.045
0.045
0.043
0.045
0.044
0.043
0.042
0.040
0.042
0.045
0.043
0.045
0.044
0.043
0.045
0.045
6
0.047
0.042
0.046
0.045
0.044
0.042
0.044
0.044
0.040
0.041
0.039
0.042
0.045
0.042
0.045
0.043
0.042
0.044
0.044
7
0.046
0.042
0.045
0.044
0.044
0.042
0.044
0.042
0.040
0.041
0.038
0.041
0.044
0.041
0.044
0.042
0.042
0.043
0.044
8
0.046
0.042
0.044
0.044
0.043
0.042
0.043
0.042
0.039
0.041
0.038
0.041
0.044
0.041
0.043
0.041
0.042
0.042
0.044
9
0.045
0.041
0.043
0.043
0.043
0.041
0.043
0.038
0.038
0.041
0.037
0.040
0.044
0.041
0.042
0.041
0.042
0.042
0.043
10
0.045
0.041
0.043
0.043
0.042
0.041
0.043
0.038
0.038
0.041
0.037
0.040
0.044
0.040
0.042
0.041
0.041
0.042
0.043
-------
Table AX3-4 (cont'd). The Top 10 Daily Maximum 8-h Average Concentrations (ppm) for Sites Experiencing Low
Maximum Hourly Average Concentrations with Data Capture of >75%
Site Year
Glacier NP 1989
300298001
(Montana) 199°
963m
1992
1993
1994
1995
1996
1997
^ 1998
i
to
-^ 1999
2000
2001
Yellowstone NP 1988
560391010
(Wyoming) 1989
2484 m
1991
1992
1993
1994
1995
1
0.062
0.058
0.060
0.062
0.055
0.057
0.061
0.059
0.056
0.060
0.065
0.059
0.054
0.068
0.067
0.057
0.059
0.066
0.057
0.067
0.064
2
0.061
0.057
0.057
0.056
0.052
0.057
0.055
0.059
0.054
0.059
0.065
0.058
0.052
0.068
0.065
0.056
0.058
0.064
0.054
0.063
0.062
3
0.060
0.055
0.057
0.055
0.051
0.056
0.053
0.058
0.052
0.058
0.060
0.058
0.049
0.067
0.064
0.054
0.058
0.064
0.054
0.063
0.061
4
0.059
0.054
0.057
0.054
0.051
0.056
0.052
0.058
0.052
0.058
0.058
0.056
0.049
0.066
0.063
0.054
0.057
0.063
0.054
0.061
0.060
5
0.058
0.053
0.056
0.054
0.050
0.055
0.052
0.057
0.052
0.056
0.056
0.054
0.049
0.066
0.063
0.053
0.056
0.063
0.053
0.061
0.059
6
0.057
0.053
0.055
0.054
0.050
0.055
0.052
0.057
0.051
0.056
0.055
0.052
0.048
0.066
0.061
0.052
0.056
0.061
0.053
0.061
0.059
7
0.056
0.052
0.055
0.053
0.049
0.055
0.051
0.055
0.050
0.055
0.055
0.051
0.047
0.064
0.061
0.050
0.056
0.061
0.053
0.061
0.059
8
0.056
0.052
0.054
0.053
0.049
0.055
0.051
0.056
0.050
0.055
0.055
0.050
0.047
0.064
0.061
0.050
0.055
0.059
0.052
0.061
0.059
9
0.056
0.052
0.054
0.053
0.049
0.054
0.051
0.055
0.050
0.055
0.055
0.050
0.047
0.063
0.061
0.049
0.055
0.059
0.052
0.059
0.059
10
0.056
0.052
0.053
0.053
0.048
0.053
0.050
0.055
0.050
0.054
0.054
0.050
0.047
0.061
0.060
0.048
0.055
0.058
0.052
0.059
0.058
-------
Table AX3-4 (cont'd). The Top 10 Daily Maximum 8-h Average Concentrations (ppm) for Sites Experiencing Low
Maximum Hourly Average Concentrations with Data Capture of >75%
X
OJ
I
to
Site
Yellowstone NP
560391011
(Wyoming)
2468m
Denali NP
022900003
(Alaska)
640m
Badlands NP
460711001
(South Dakota)
730m
Year
1997
1998
1999
2000
2001
1988
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
1989
1990
1991
1
0.065
0.069
0.078
0.070
0.068
0.055
0.049
0.054
0.053
0.053
0.053
0.058
0.058
0.054
0.057
0.056
0.046
0.061
0.069
0.061
0.058
2
0.065
0.068
0.074
0.069
0.068
0.054
0.048
0.054
0.052
0.053
0.051
0.056
0.053
0.053
0.056
0.056
0.046
0.058
0.066
0.059
0.058
3
0.062
0.066
0.073
0.067
0.066
0.054
0.048
0.050
0.052
0.051
0.049
0.056
0.053
0.052
0.056
0.054
0.044
0.057
0.064
0.055
0.056
4
0.061
0.066
0.071
0.065
0.066
0.053
0.048
0.050
0.051
0.048
0.049
0.054
0.053
0.051
0.055
0.054
0.044
0.055
0.063
0.055
0.056
5
0.061
0.063
0.070
0.065
0.065
0.053
0.048
0.047
0.050
0.048
0.049
0.051
0.052
0.051
0.054
0.054
0.044
0.055
0.060
0.054
0.056
6
0.060
0.063
0.070
0.065
0.064
0.053
0.047
0.046
0.050
0.047
0.048
0.050
0.052
0.050
0.054
0.053
0.043
0.055
0.058
0.052
0.055
7
0.057
0.061
0.070
0.064
0.064
0.052
0.047
0.046
0.049
0.047
0.048
0.049
0.052
0.050
0.054
0.053
0.043
0.053
0.057
0.052
0.055
8
0.056
0.061
0.069
0.064
0.064
0.052
0.046
0.046
0.049
0.046
0.048
0.046
0.052
0.049
0.054
0.053
0.042
0.053
0.057
0.051
0.054
9
0.056
0.061
0.068
0.063
0.064
0.052
0.046
0.045
0.049
0.046
0.048
0.046
0.051
0.049
0.053
0.052
0.042
0.053
0.057
0.051
0.054
10
0.056
0.060
0.067
0.063
0.063
0.052
0.046
0.044
0.049
0.046
0.048
0.046
0.051
0.049
0.053
0.051
0.042
0.053
0.057
0.050
0.053
-------
Table AX3-4 (cont'd). The Top 10 Daily Maximum 8-h Average Concentrations (ppm) for Sites Experiencing Low
Maximum Hourly Average Concentrations with Data Capture of >75%
X
OJ
1
to
ON
Site
Theod. Roos. NP
380530002
(North Dakota)
730m
Theod. Roos. NP
380070002
(North Dakota)
808m
Custer NF, MT
300870101
(Montana)
1006m
Year
1984
1985
1986
1989
1992
1993
1994
1995
1996
1997
1999
2000
2001
1978
1979
1980
1983
1
0.064
0.058
0.059
0.073
0.060
0.062
0.066
0.060
0.060
0.080
0.063
0.062
0.060
0.069
0.073
0.069
0.064
2
0.062
0.055
0.058
0.069
0.059
0.059
0.064
0.059
0.059
0.073
0.060
0.061
0.059
0.065
0.066
0.069
0.061
3
0.062
0.055
0.057
0.066
0.058
0.056
0.058
0.058
0.059
0.072
0.059
0.060
0.059
0.063
0.066
0.069
0.060
4
0.062
0.054
0.056
0.065
0.058
0.056
0.058
0.058
0.059
0.071
0.058
0.059
0.058
0.062
0.065
0.068
0.060
5
0.059
0.054
0.055
0.065
0.056
0.055
0.057
0.058
0.058
0.069
0.057
0.059
0.058
0.061
0.063
0.067
0.059
6
0.058
0.054
0.055
0.064
0.056
0.053
0.056
0.058
0.058
0.068
0.057
0.057
0.057
0.061
0.060
0.067
0.058
7
0.057
0.053
0.054
0.063
0.056
0.052
0.056
0.057
0.057
0.066
0.057
0.057
0.057
0.060
0.060
0.066
0.058
8
0.057
0.053
0.053
0.063
0.054
0.052
0.056
0.057
0.057
0.066
0.056
0.057
0.056
0.060
0.060
0.066
0.058
9
0.057
0.053
0.053
0.063
0.054
0.052
0.056
0.056
0.057
0.063
0.056
0.057
0.055
0.060
0.059
0.064
0.056
10
0.057
0.052
0.052
0.063
0.054
0.052
0.055
0.055
0.056
0.063
0.056
0.057
0.055
0.058
0.059
0.064
0.056
-------
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Year
Figure AX3-12. Seasonal SUM06 and W126 exposure indices for the Ouachita National
Forest for the period of 1991 to 2001.
AX3.2.1 Nationwide Distribution of Metrics for Characterizing Exposures
of Vegetation to Ozone
The previous O3 AQCD (U.S. Environmental Protection Agency, 1996a) concluded that
higher hourly average concentrations (>0.10 ppm) should be given greater weight than mid-level
(0.06 to 0.099 ppm) and lower hourly average concentrations in predicting injury and yield
reduction for agricultural crops and forests. The most recent findings concerning the importance
of the higher hourly average concentrations in comparison to the mid-level and lower values will
be discussed in Chapter 9. Because of a lack of air quality data collected at rural and remote
locations, interpolation techniques, such as kriging, have been applied to the estimation of O3
exposures across the United States (Reagan, 1984; Lefohn et al., 1987; Knudsen and Lefohn,
1988). "Kriging" (Matheron, 1963) has been used in the analyses of air quality data (Grivet,
1980; Faith and Sheshinski, 1979) and was used to provide estimates of seasonal O3 values for
the National Crop Loss Assessment Network (NCLAN) for 1978 through 1982 (May to
September of each year) (Reagan, 1984). These values, along with updated values, coupled with
exposure-response models, were used to predict agriculturally related economic benefits
anticipated by lower O3 levels in the United States (Adams et al., 1985, 1989). It should be
noted that although kriging has been used here to map unmeasured ozone concentrations, it has
shortcomings. In particular, the kriging variance underestimates the uncertainty in spatial
AX3-27
-------
prediction and relies on the assumption of spatial covariance isotropy. Newer methods
overcomes many of these shortcomings (e.g., Le et al., 1992, 1997, 2001). These methodologies
have been tested by Sun et al. (1998), who also show that the uncertainties in spatial fields can
be substantially underestimated using kriging.
Spatiotemporal distributions of O3 concentrations have alternatively been obtained using
methods of the "Spatio-Temporal Random Field" (STRF) theory (Christakos and Vyas,
1998a,b). The STRF approach interpolates monitoring data in both space and time
simultaneously. This method can analyze information on "temporal trends," which cannot be
incorporated directly in purely spatial interpolation methods such as standard kriging. Further,
the STRF method can optimize the use of data that are not uniformly sampled in either space or
time. The STRF theory was further extended in the Bayesian Maximum Entropy (BME)
framework and applied to O3 interpolation studies (Christakos and Hristopulos, 1998; Christakos
and Kolovos, 1999; Christakos, 2000). The BME framework can use prior information in the
form of "hard data" (measurements), probability law descriptors (type of distribution, mean and
variance), interval estimation (maximum and minimum values) and even constraint from
physical laws. According to these researchers, both STRF and BME were found to successfully
reproduce O3 fields when adequate monitor data are available.
For 2001, ordinary kriging was used to estimate the seasonal W126, SUM06, and number
of hours >0.10 ppm (N100), using hourly average concentrations accumulated over a 24-h
period. As discussed in Chapter 9, the correlation between the number of occurrences of hourly
average concentrations >0.10 ppm and the magnitude of the W126 and SUM06 values is not
strong. Because of this, the N100 was also estimated, along with the W126 and SUM06
exposure indices. For the period of April through September, the estimates of the seasonal
W126, SUM06, and N100 exposure index values were made for each 0.5° by 0.5° cell in the
continuous United States. The kriged values, the variance, and the 95% error bound for each
0.5° by 0.5° cell were estimated. Because of the concern for inner-city depletion caused by NOX
scavenging, data from specific monitoring stations located in large metropolitan areas were not
included in the analysis.
Figure AX3-13 shows the kriged values for the 24-h cumulative seasonal W126 exposure
index and the N100 index for 2001 for the eastern United States. Note that for some of the areas
with elevated W126 values (e.g., >35 ppm-h), the number of hourly average concentrations was
AX3-28
-------
N100 (hours)
* * "#
* * * \
+ 4 » •'•-,
'.if * * *"
«( . » »
0 125 250
N
500
750
1,000
^^ Miles
1:13,936,494
Figure AX3-13.
Six-month (April to September) 24-h cumulative W126 exposure index
with the number of hourly average concentrations >0.10 ppm (N100)
occurring during 2001 for the eastern United States.
AX3-29
-------
estimated to be <22. Figure AX3-14 illustrates the kriged values using the 24-h cumulative
seasonal SUM06 exposure index and the N100 index for 2001 for the eastern United States.
Figures AX3-15 and AX3-16 show the W126 and SUM06 values, respectively, with the N100
values for the central United States region. For 2001, the number of hourly average
concentrations >0.10 ppm was usually <22 for the 6-month period. Figures AX3-17 and
AX3-18 illustrate the W126 and SUM06 values, respectively, for the western United States
region. Note that in the Southern California and Central California areas, the number of hourly
average concentrations >0.10 ppm was in the range of 48 to 208 for the 6-month period. This is
considerably greater than the frequency of occurrences for the higher hourly average
concentrations observed in the eastern and central United States.
Due to the scarcity of monitoring sites across the United States, especially in the Rocky
Mountain region, the uncertainty in the estimates for the various exposure indices vary.
Figures AX3-19 through AX3-27 illustrate the 95% confidence intervals associated with the
indices by region.
AX3.3 SPATIAL VARIABILITY IN OZONE CONCENTRATIONS
The spatial variability of O3 concentrations in different environments in the United States
occurring across a variety of spatial scales is characterized in this section. This information will
be useful for understanding the influence of regional or altitudinal differences in O3 exposure on
vegetation and for establishing the spatial variations in O3 concentrations as they are used in
epidemiologic studies. Intracity variations in O3 concentrations are described in Section
AX3.3.1. Small scale horizontal and vertical variations in O3 concentrations are discussed in
Section AX3.3.2. Ozone concentrations at high elevations are characterized in Section AX3.3.3.
AX3.3.1 Spatial Variability of Ozone Concentrations in Urban Areas
Many processes contribute to spatial variability in O3 concentrations in urban areas. Ozone
formation occurs more or less continuously downwind of sources of precursors, producing a
gradient in O3 concentrations. Ozone "titration" by reaction with NO can deplete O3 levels near
NO sources such as highways and busy streets. Differences in surface characteristics affect the
AX3-30
-------
• • • • I . ,• • •
\mmmmmmm
t
* *
N100 (hours)
SUM06 (ppm-h)
0 125 250
1:13,934,062
1,000
Miles
Figure AX3-14. Six-month (April to September) 24-h cumulative SUM06 exposure index
with the number of hourly average concentrations ^0.10 ppm (N100)
occurring during 2001 for the eastern United States.
AX3-31
-------
N100 (hours)
• 1-2
» >2-7
• >7-21
• >21 - 48
A >48 - 208
W126 (ppm-h)
I i 1-19
j j >19-35
| I >35 - 68
>63- 146
v
i:
I • ••!!••***
i +...-• • • • • Vwiwi^ipui,
* * » • • » *••• * *
•»»•***»+•••
\ * * * * •
...#
JL
*
*
.
*^
• •
* • 1
» V
• * *
\ ' *
:^:
"*>
*
_*
*
»
M-y
-.-.—^
N
100 200
400
600
800
^3 Miles
1:11,998,827
Figure AX3-15. Six-month (April to September) 24-h cumulative W126 exposure index
with the number of hourly average concentrations ^0.10 ppm (N100)
occurring during 2001 for the central United States.
AX3-32
-------
"X
N1 00 (hours)
• 1-2
» >2-7
• >7-21
• >21 - 48
A >48 - 208
SUM06 (ppm-h)
[~] 1-18
>55-110
' • « » « • •f^A?= •»»»»/
f * * J| ^ • • t t *'XX
P
• • SJL-fc-4 1
t • • » »>-• » + ' • »f
*»»••»«•••* t ••• t » \» • /
%•_ y -x • • • » • »••••• %?^ *| A» ?'^'~--v
4^-4
N
0 100 200 400 600 800
1:11,998,827
Figure AX3-16. Six-month (April to September) 24-h cumulative SUM06 exposure index
with the number of hourly average concentrations ^0.10 ppm (N100)
occurring during 2001 for the central United States.
AX3-33
-------
800
Miles
1:13,052,120
Figure AX3-17. Six-month (April to September) 24-h cumulative W126 exposure index
with the number of hourly average concentrations ^0.10 ppm (N100)
occurring during 2001 for the western United States.
AX3-34
-------
>30 - 55
>55-110
800
Miles
1:13,052,120
Figure AX3-18. Six-month (April to September) 24-h cumulative SUM06 exposure index
with the number of hourly average concentrations ^0.10 ppm (N100)
occurring during 2001 for the western United States.
AX3-35
-------
W126 95% Conf. Interval
3-10 ppm-h
| >10-15 ppm-h
j >15-25 ppm-h
>25 - 90 ppm-h
Figure AX3-19. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative W126 exposure index for 2001 for the eastern United States.
SUM06 95% Conf. Interval
i 12-12 ppm-h
| i >12 -15 ppm-h
>15 -20 ppm-h
[ > 20 - 73 ppm-h
Figure AX3-20. The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the eastern
United States.
AX3-36
-------
N100 95% Conf. Interval
1 0-lOh
>10-15h
>15-20h
[ >20 - 95 h
i >95-12?h
Figure AX3-21. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the eastern United States.
W126 95% Conf. Interval
: 3- 10ppm-h
| >10 -15 ppm-h
i >15 - 25 ppm-h
>25 - 90 ppm-h
Figure AX3-22. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative W126 exposure index for 2001 for the central United States.
AX3-37
-------
SUM06 95% Conf. Interval
2- 12ppm-h
>12 -15 ppm-h
>15 - 20 ppm-h
> 20 - 73 ppm-h
Figure AX3-23. The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the central
United States.
N100 95% Conf. Interval
| 0- 10h
; >10-1Sh
>20 - 95 h
>95 -127 h
Figure AX3-24. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the central United States.
AX3-38
-------
W126 95% Conf. Interval
'y^vi 3- 10ppm-h
j >10 -15 ppm-h
>15 - 25 ppm-h
>25 - 90 ppm-h
Figure AX3-25. The 95% confidence interval for the 6-month (April to September)
24-h cumulative W126 exposure index for 2001 for the western
United States.
SUM06 95% Conf. Interval
I >12-15 ppm-ti
>15 - 20 ppm-h
> 20 - 73 ppm-h
Figure AX3-26. The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the western
United States.
AX3-39
-------
N100 95% Conf. Interval
0- 10 h
>15-20h
>20 - 95 h
>95-127h
Figure AX3-27. The 95% confidence interval for the 6-month (April to September)
24-h cumulative N100 exposure index for 2001 for the western
United States.
rate of deposition of O3. Mixing of O3 from aloft can also lead to local increases in O3
concentration.
The spatial variability in O3 concentrations in 24 MSAs across the United States is
characterized in this section. These areas were chosen to provide analyses to help guide risk
assessment, to provide a general overview of the spatial variability of O3 over different regions
of the country, and also to provide insight into the spatial distribution of O3 in cities where health
outcome studies have been conducted. Statistical analyses of the human health effects of
airborne pollutants based on aggregate population time-series data have often relied on ambient
concentrations of pollutants measured at one or more central sites in a given metropolitan area.
In the particular case of ground-level O3 pollution, central-site monitoring has been justified as a
regional measure of exposure partly on grounds that correlations between concentrations at
neighboring sites measured over time are usually high (U.S. Environmental Protection Agency,
1996a). In analyses where multiple monitoring sites provide ambient O3 concentrations, a
AX3-40
-------
summary measure such as an averaged concentration has often been regarded as adequately
characterizing the exposure distribution. Indeed, a number of studies have referred to
multiple-site averaging as the method for estimating O3 exposure (U.S. Environmental
Protection Agency, 1996a). It is hoped that the analyses presented here will shed some light on
the suitability of this practice. Earlier analyses were reported in the previous O3 AQCD (U.S.
Environmental Protection Agency, 1996a). The analyses presented there concluded that the
extent of spatial homogeneity varies with the MSA under study. In particular, cities with low
traffic densities that are located downwind of major sources of precursors are heavily influenced
by long range transport and tend to show smaller spatial variability (e.g., New Haven, CT) than
those source areas with high traffic densities located upwind (e.g., New York, NY).
Metrics for characterizing spatial variability include the use of Pearson correlation
coefficients (r), values of the 90th percentile (P90) of the absolute difference in concentrations,
and coefficients of divergence (COD)1. These methods of analysis follow those used for
characterizing PM25 and PM10_25 concentrations in Pinto et al. (2004) and in the latest edition of
the PM AQCD (U.S. Environmental Protection Agency, 2004a). Data were aggregated over
local O3 seasons, whose length varies from state to state. In several southwestern states, it lasts
all year long. However, it typically lasts for 6 months in other areas, such as New England, the
mid-Atlantic states, the Midwest, and the Northwest (see Table AX3-1).
Table AX3-5 shows the chosen urban areas, the range of 24-h average O3 concentrations
over the local O3 season from 1999 to 2001, the range of intersite correlation coefficients, the
range of P90 differences in O3 concentrations between site pairs, and the range in COD values. A
COD of zero implies that values in both data sets are identical, and a COD of one indicates that
the two data sets are uncorrelated with no identical values from either data set. In general,
statistics were calculated for partial MSAs. This was done so as to obtain reasonable lower
estimates of the spatial variability that is present, as opposed to examining the consolidated
MSAs. In Boston, MA and New York, NY, this could not be readily done, and so statistics were
The COD is defined as follows:
where xtj and xit represent the 24-h average PMj 5 concentration for day i at site/ and site k and/) is the number of observations.
AX3-41
-------
Table AX3-5. Summary Statistics for Ozone (in ppm) Spatial Variability in Selected U.S. Urban Areas
Urban Area
Boston, MA
New York, NY
Philadelphia, PA
Washington, DC
Charlotte, NC
Atlanta, GA
Tampa, FL
Detroit, MI
Chicago, IL
Milwaukee, WI
^ St. Louis, MO
OJ
_k_ Baton Rouge, LA
Dallas, TX
Houston, TX
Denver, CO
El Paso, TX
Salt Lake City, UT
Phoenix, AZ
Seattle, WA
Portland, OR
Fresno, CA
Bakersfield, CA
Los Angeles, CA
Riverside, CA
Number of
Sites
18
29
12
20
8
12
9
7
24
9
17
7
10
13
8
4
8
15
5
5
6
8
14
18
Minimum
Mean Cone.
0.021
0.015
0.020
0.022
0.031
0.023
0.024
0.022
0.015
0.027
0.022
0.018
0.028
0.016
0.022
0.022
0.029
0.021
0.015
0.015
0.030
0.028
0.010
0.018
Maximum
Mean Cone.
0.033
0.041
0.041
0.041
0.043
0.047
0.035
0.037
0.039
0.038
0.038
0.031
0.043
0.036
0.044
0.032
0.048
0.058
0.038
0.036
0.047
0.047
0.042
0.054
Minimum
Corr. Coeff.
0.46
0.45
0.79
0.72
0.48
0.63
0.74
0.74
0.38
0.73
0.78
0.81
0.67
0.73
0.60
0.81
0.52
0.29
0.63
0.73
0.90
0.23
0.42
0.38
Maximum
Corr. Coeff.
0.93
0.96
0.95
0.97
0.95
0.94
0.94
0.96
0.96
0.96
0.96
0.95
0.95
0.96
0.92
0.94
0.92
0.95
0.94
0.91
0.97
0.96
0.95
0.95
Minimum
P9oa
0.012
0.0080
0.011
0.010
0.012
0.013
0.011
0.0090
0.0080
0.0090
0.0090
0.0090
0.011
0.0090
0.013
0.012
0.012
0.011
0.0080
0.011
0.0090
0.013
0.010
0.013
Maximum
0.041
0.044
0.036
0.032
0.038
0.045
0.025
0.027
0.043
0.025
0.031
0.029
0.033
0.027
0.044
0.023
0.043
0.057
0.024
0.025
0.027
0.052
0.053
0.057
Minimum
COD"
0.17
0.17
0.23
0.17
0.17
0.24
0.20
0.19
0.16
0.18
0.15
0.23
0.16
0.20
0.16
0.24
0.13
0.15
0.16
0.20
0.17
0.20
0.22
0.15
Maximum
COD
0.45
0.55
0.46
0.45
0.32
0.55
0.35
0.36
0.50
0.33
0.41
0.41
0.36
0.38
0.46
0.31
0.51
0.61
0.46
0.50
0.40
0.58
0.59
0.64
T90 = 90th percentile absolute difference in concentrations.
bCOD = coefficient of divergence for different site pairs.
-------
calculated for the consolidated MSAs. More detailed calculations for a subset of nine MSAs are
given in Figures AX3-28 through AX3-36.
As can be seen, there are no clearly discernible regional trends in the ranges of parameters
shown. Additional urban areas would need to be examined to discern broadscale patterns. The
data indicate considerable variability in the concentration fields. Mean O3 concentrations vary
within individual urban areas from factors of 1.4 to 4.0.
The highest annual mean O3 concentration (0.058 ppm) is found in the Phoenix, AZ MSA
at a site which is located in the mountains well downwind of the main urban area. The lowest
annual mean O3 concentration (0.010 ppm) was found in Lynwood in the urban core of the
Los Angeles MSA. CO and NOX monitors at this site recorded the highest concentrations in
California, indicating that titration of O3 by NO freshly emitted from tail pipes of motor vehicles
is responsible for the low O3 values that are found. Ratios of highest to lowest mean O3
concentrations in these two MSAs are among the highest shown in Table AX3-5. Both of these
MSAs are characterized by sunny, warm climates; sources of precursors that are associated
with O3 titration to varying degrees in their urban centers; and with maximum O3 found well
downwind of the urban centers. Intersite correlation coefficients show mixed patterns, i.e.,
in some urban areas all pairs of sites are moderately to highly correlated, while other areas show
a very large range of values. As may be expected, those areas which show smaller ratios of
seasonal mean concentrations also exhibit a smaller range of intersite correlation coefficients.
Within the examined urban areas, P90 values were evenly distributed between all site pairs
considered. The CODs indicate variability among site pairs. However, there are a number of
cases where sites in an urban area may be moderately to highly correlated but showed substantial
differences in absolute concentrations. In many cases, values for P90 equaled or exceeded
seasonal mean O3 concentrations. This was reflected in both values for P90 and for the COD.
It is instructive to compare the metrics for spatial variability shown in Table AX3-5 to
those calculated for PM2 5 and PM10_2 5 in the PM AQCD (U.S. Environmental Protection
Agency, 2004a). The values for concentrations and concentration differences are unique to the
individual species, but the intersite correlation coefficients and the COD values can be directly
compared.
AX3-43
-------
Charlotte - Gastonia - Rock Hill, NC - SC MSA
a.
c.
40 20
0
40
Kilometers
b.
AIRS Site ID
Site A
Site B
SiteC
Site D
Site E
Site F
SiteG
SiteH
37-1
37-1
37-1
37-1
37-1
37-1
37-1
09-0004
19-0041
19-1005
19-1009
59-0021
59-0022
79-0003
45-091-0006
80
Site
mean
(ppm)
A
B
C
D
E
F
G
H
A B
0.034 0.034
1 0.87
0.020
0.28
4508
1
C
0.031
0.86
0.021
0.29
4824
0.93
0.015
0.20
4706
1
D
0.035
0.88
0.019
0.28
4834
0.95
0.012
0.17
4720
0.91
0.017
0.21
5045
1
E
0.039
0.88
0.021
0.28
4821
0.91
0.018
0.23
4503
0.85
0.023
0.28
4821
0.93
0.015
0.21
4832
1
F
0.040
0.86
0.023
0.31
4677
0.87
0.020
0.26
4366
0.82
0.025
0.30
4677
0.89
0.018
0.25
4688
0.89
0.016
0.22
4684
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
G
0.038
0.84
0.023
0.29
4664
0.90
0.017
0.24
4341
0.87
0.020
0.27
4657
0.87
0.019
0.25
4664
0.88
0.018
0.22
4651
0.81
0.022
0.26
4510
1
H
0.035
0.85
0.021
0.29
4749
0.82
0.021
0.27
4621
0.82
0.022
0.30
4955
0.81
0.022
0.27
4968
0.83
0.021
0.24
4744
0.81
0.021
0.23
4608
0.83
0.019
0.26
4582
1
Figure AX3-28.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Charlotte, NC-Gastonia-Rock Hill, SC
MSA. The mean hourly measured O3 concentration at each site is given
above its letter code. For each data pair, the Pearson correlation
coefficient, 90th percentile difference in absolute concentrations, the
coefficient of divergence, and number of observations are given.
AX3-44
-------
Baton Rouge, LA MSA
c.
25
0
Kilometers
b.
AIRS Site ID
Site A 22-005-0004
Site B 22-033-0003
Site C 22-033-0009
Site D 22-033-0013
SiteE 22-033-1001
Site F 22-063-0002
Site G 22-121-0001
50
i
Site
mean
(ppm)
A
B
C
D
E
F
G
A B
0.022 0.023
1 0.88
0.015
0.32
8378
1
C
0.023
0.86
0.016
0.33
8242
0.95
0.009
0.23
8159
1
D
0.025
0.90
0.015
0.32
8342
0.87
0.015
0.29
8266
0.87
0.015
0.30
8224
1
E
0.026
0.87
0.017
0.36
8381
0.92
0.012
0.27
8301
0.92
0.012
0.27
8171
0.93
0.011
0.24
8266
1
F
0.025
0.94
0.011
0.28
8382
0.86
0.016
0.32
8302
0.86
0.017
0.32
8260
0.93
0.012
0.24
8379
0.89
0.015
0.29
8306
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired
observations
G
0.023
0.86
0.016
0.33
8360
0.93
0.010
0.26
8281
0.94
0.010
0.23
8157
0.88
0.015
0.28
8251
0.94
0.011
0.25
8290
0.86
0.016
0.31
8292
1
Figure AX3-29.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Baton Rouge, LA MSA. The mean hourly
measured O3 concentration at each site is given above its letter code.
For each data pair, the Pearson correlation coefficient, 90th percentile
difference in absolute concentrations, the coefficient of divergence, and
number of observations are given.
AX3-45
-------
Detroit -Ann Arbor - Flint, Ml CMSA
b.
Kilometers
AIRS Site ID
Site A
SiteB
SiteC
SitcD
SiteE
SlteF
SiteG
SiteH
Site I
Site J
SiteK
26-049-
26-049-
26-091
26-099
26-099-
26-125-
26-147-
26-161-
26-163-
26-163-
26-163-
0021
2001
0007
0009
1003
0001
0005
0008
0001
0016
0019
50
100
C. Site
mean
(ppm)
0.033 0,035 0.037 0.032 0,031 0.028 0.031 0.032 0.026 0.028 0.028
1 0.93 0.84 0.84 0.86 0.86 0.81 0.86 0.82 0.84 0.85
0012 0019 0019 0019 0020 0020 0019 0.024 0.022 0.020
0.20 0.24 026 0.27 0.30 0.26 0.27 0.35 0.32 0.29
4333 4151 4253 4341 4103 4338 4163 4228 4324 4247
1
0.84 0.86 0.85 0.84 0.82 0.84 0.81 0.81 0.84
0.019 0.017 0.020 0.023 0.019 0.021 0.025 0.024 0.021
0.22 025 0.28 0.33 0.25 0.28 0.37 036 0.32
4152 4252 4341 4104 4339 4164 4230 4323 4248
1
085 0.86 0.87 0.78 0.92 0.84 086 0.86
0.020 0020 0.022 0.023 0015 0.025 0.023 0.021
0.27 0.28 0.32 0.27 0.25 0.37 0.35 0.32
4072 4159 3930 4156 4164 4049 4143 4068
1 0.91 0.88 0.86 0.86 0.84
0.014 0.019 0.017 0.019 0.022
0.26 0.31 0.25 0.27 0.34
4260 4022 4259 4086 4148
1 094 083 089 088
0.014 0.020 0.017 0.020
0.24 0.27 0.25 0.31
4112 4346 4172 4235
1
0.83 0.91 0.85
0.021 0.015 0.020
0.29 0.25 0.31
4109 3944 4000
1
0.79 0.76
0.022 0,024
0.28 0.36
4169 4233
1 0.88
0.019
0.34
4062
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
0.87 0.90
0.020 0.016
0.31 0.27
4245 4193
094 0.96
0.013 0.010
024 0.19
4330 4255
0.83 0.93
0.013 0.014
0.23 0.24
4093 4025
080 0.83
0.022 0.020
0.33 0.29
4329 4255
0.89 0.89
0.017 0.017
0.29 0.27
4153 4081
0.89 0.90
0.016 0.018
029 0.29
4217 4144
1 0.95
0.011
0.21
4242
1
Figure AX3-30.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Detroit-Ann Arbor-Flint, MI CMSA.
The mean hourly measured O3 concentration at each site is given above
its letter code. For each data pair, the Pearson correlation coefficient,
90th percentile difference in absolute concentrations, the coefficient of
divergence, and number of observations are given.
AX3-46
-------
St. Louis, MO - IL CMSA
a.
b.
60
i
30
0
i
60
i
120
i
Kilometers
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
SiteG
SiteH
Site I
Site J
SiteK
SiteL
SiteM
SiteN
SiteO
SiteP
SiteQ
17-083-
17-119-
17-119-
17-119-
17-119-
17-163-
29-099-
29-183-
29-183-
29-189-
29-189-
29-189-
29-189-
29-189-
29-510-
29-510-
29-510-
1001
0008
1009
2007
3007
0010
0012
1002
1004
0004
0006
3001
5001
7003
0007
0072
0086
Figure AX3-31.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the St. Louis, MO-IL MSA. The mean hourly
measured O3 concentration at each site is given above its letter code.
For each data pair, the Pearson correlation coefficient, 90th percentile
difference in absolute concentrations, the coefficient of divergence, and
number of observations are given.
AX3-47
-------
c.
St. Louis, MO - IL CMSA
Site
mean
(ppm)
A B
0.037 0.035
C
0.031
D
0.029
E
0.031
F G
0.028 0.035
H 1
0.034 0.035
J K
0.032 0.031
L
0.025
M N
0.030 0.030
0
0.028
P Q
0.024 0.030
M
O
Q
1 0.88 0.84 0.84 0.85 0.83 0.81 0.89 0.91 0.82 0.84 0.80 0.86 0.85 0.80 0.82 0.85
0.015 0.020 0.022 0.021 0.025 0.021 0.016 0.014 0.022 0.021 0.029 0.022 0.021 0.026 0.027 0.022
0.19 0.26 0.29 0.29 0.38 0.28 0.21 0.20 0.28 0.29 0.37 0.31 0.32 0.36 0.40 0.34
5055 4935 5032 5021 5045 5049 5052 4735 5077 5073 5075 5077 5071 5035 5034 5065
1
0.87 0.90 0.93 0.86 0.82 0.93 0.90 0.83 0.85 0.84 0.87 0.84 0.84 0.87 0.88
0.017 0.016 0.013 0.021 0.019 0.012 0.014 0.020 0.018 0.024 0.019 0.020 0.022 0.022 0.018
0.26 0.26 0.23 0.34 0.28 0.18 0.19 0.27 0.27 0.34 0.30 0.30 0.33 0.35 0.31
4963 5061 5048 5077 5078 5082 4766 5106 5102 5105 5106 5101 5064 5064 5094
1
0.92 0.88 0.92 0.87 0.90 0.88 0.88 0.87 0.85 0.90 0.88 0.86 0.88 0.91
0.012 0.015 0.014 0.018 0.015 0.016 0.016 0.016 0.020 0.015 0.016 0.018 0.017 0.014
0.24 0.26 0.29 0.26 0.23 0.24 0.25 0.26 0.32 0.26 0.27 0.30 0.32 0.27
4941 4933 4952 4958 4963 4680 4985 4981 4983 4985 4979 4943 4942 4973
1 0.93 0.90 0.85 0.91 0.88 0.86 0.84 0.84 0.87 0.85 0.86 0.88 0.90
0.012 0.014 0.019 0.016 0.017 0.018 0.018 0.020 0.016 0.017 0.017 0.016 0.015
0.22 0.29 0.26 0.23 0.25 0.27 0.27 0.30 0.27 0.28 0.28 0.29 0.27
5024 5050 5055 5058 4742 5083 5079 5081 5085 5077 5042 5040 5073
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
0.90 0.84 0.93 0.90 0.85 0.84 0.88 0.88 0.86 0.88 0.89 0.90
0.015 0.019 0.013 0.015 0.018 0.018 0.018 0.016 0.017 0.016 0.017 0.015
0.29 0.27 0.22 0.23 0.27 0.29 0.32 0.29 0.29 0.29 0.30 0.28
5039 5045 5046 4728 5069 5065 5067 5069 5065 5027 5027 5057
1 090 090 089 091 088 090 092 091 092 092 094
0.018 0.018 0.018 0.016 0.017 0.016 0.014 0.015 0.013 0.015 0.012
0.28 0.32 0.32 0.29 0.30 0.32 0.27 0.27 0.26 0.28 0.23
5067 5070 4754 5095 5091 5093 5095 5091 5053 5053 5083
1 0.87 0.86 0.95 0.91 0.88 0.88 0.88 0.89 0.86 0.91
0.017 0.017 0.012 0.015 0.022 0.018 0.017 0.020 0.023 0.016
0.23 0.24 0.22 0.25 0.33 0.27 0.26 0.29 0.35 0.26
5075 4756 5100 5096 5098 5100 5094 5058 5057 5088
1 0.94 0.87 0.88 0.87 0.92 0.90 0.86 0.87 0.91
0.011 0.017 0.016 0.022 0.015 0.015 0.020 0.022 0.015
0.15 0.24 0.24 0.32 0.26 0.27 0.31 0.35 0.29
4759 5103 5099 5101 5103 5097 5062 5060 5091
1 0.86 0.90 0.88 0.91 0.90 0.85 0.86 0.90
0.017 0.015 0.022 0.016 0.016 0.021 0.022 0.017
0.28 0.25 0.33 0.27 0.28 0.32 0.35 0.30
4784 4782 4782 4784 4778 4742 4743 4773
1 0.92 0.91 0.91 0.91 0.90 0.88 0.93
0.013 0.018 0.015 0.014 0.016 0.019 0.013
0.26 0.30 0.24 0.23 0.25 0.30 0.22
5124 5126 5128 5122 5086 5085 5116
1 0.87 0.90 0.91 0.86 0.87 0.90
0.019 0.015 0.014 0.018 0.020 0.015
0.29 0.23 0.23 0.28 0.30 0.25
5122 5124 5118 5082 5081 5112
1 0.88 0.90 0.90 0.88 0.92
0.018 0.017 0.015 0.016 0.016
0.28 0.28 0.27 0.27 0.26
5126 5120 5084 5083 5114
1 0.96 0.87 0.87 0.95
0.009 0.018 0.019 0.011
0.18 0.26 0.29 0.21
5122 5086 5085 5116
1 0.87 0.86 0.94
0.017 0.019 0.011
0.26 0.31 0.20
5080 5079 5110
1 0.92 0.92
0.014 0.014
0.26 0.25
5045 5074
1 0.92
0.016
0.28
5074
1
Figure AX3-31 (cont'd).
AX3-48
-------
Phoenix - Mesa, AZ MSA
50 0
2QO
Kilometers
c.
Site A B
H I
mean
(ppm)
0.021 0.025 0.028 0.046 0.021 0.024 0.024 0.028 0.038 0.041 0.024
1 0.87 0.91 0.59 0.93 0.89 0.91 0.90 0.78 078 0.95
0.017 0.018 0.046 0.012 0.016 0.014 0.017 0034 0.037 0.011
0.38 0.35 0.59 0.31 0.35 0.32 0.34 0.54 055 0.27
7580 8031 7991 7925 7593 7901 6651 8020 8081 7920
1 0.90 0.69 0.89 0.92 0.89 0.93 0.83 085 0.90
0.018 0.038 0.016 0.013 0.015 0.014 0027 0.029 0.016
0.25 0.46 0.41 0.33 0.30 0.22 0.38 041 0.32
7962 7912 7872 7502 7826 6768 7895 8011 7858
1 0.62 0.92 0.92 0.91 0.92 0.82 082 0.95
0.040 0.019 0.016 0.017 0.016 0027 0.030 0.013
0.47 0.40 0.34 0.28 0.24 0.39 041 0.28
8377 8326 7965 8273 7022 8367 8468 8319
1 058 063 058 061 067 079 058
0.047 0.043 0.042 0.043 0025 0.018 0.045
0.61 0.56 0.51 0.49 0.23 016 0.56
8286 7925 8231 6992 8332 8429 8281
1 0.92 0.94 0.93 0.82 080 0.95
0.014 0.012 0.015 0033 0.037 0.012
035 035 037 057 058 031
7894 8180 6938 8279 8369 8238
1 0.90 0.94 0.82 084 0.93
0.015 0.012 0030 0.033 0.013
0.35 0.30 0.50 052 0.31
7863 6599 7913 8009 7846
1 0.93 0.81 080 0.93
0.014 0029 0.033 0.013
0.25 0.44 046 0.29
6882 8225 8322 8166
1 0.82 083 0.93
0029 0032 0013
0.41 043 0.25
6963 7078 6912
1 086 0.81
0.016 0.031
017 0.50
8411 8265
1 0.80
0.035
0.51
8367
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
Figure AX3-32.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Phoenix-Mesa, AZ MSA. The mean
hourly measured O3 concentration at each site is given above its letter
code. For each data pair, the Pearson correlation coefficient, 90th
percentile difference in absolute concentrations, the coefficient of
divergence, and number of observations are given.
AX3-49
-------
Fresno, CA MSA
c.
Site
mean
(ppm)
A
B
A B
0.031 0.031
1 0.97
0.010
0.19
7804
1
C
0.047
0.94
0.027
0.34
8124
0.95
0.026
0.40
7777
D
0.040
0.93
0.022
0.25
7950
0.92
0.022
0.33
7624
E
0.036
0.95
0.016
0.19
8066
0.97
0.013
0.23
7726
F
0.030
0.92
0.015
0.25
8055
0.92
0.017
0.25
7710
1 0.92 0.93 0.94
0.019 0.022 0.027
0.20 0.29 0.40
7939 8055 8036
50 0 Kilometers 200 D
i i i i i i i i i i
b.
AIRS Site ID
Site A 06-019-0007
SiteB 06-019-0008
SiteC 06-019-0242
SiteD 06-019-4001
SiteE 06-019-5001
Site F 06-039-0004
E
F
1 0.92 0.9;
0.017 0.02
0.22 0.3^
7876 787
1 0.9C
0.02
0.2£
799
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
Figure AX3-33. Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Fresno, CA MSA. The mean hourly
measured O3 concentration at each site is given above its letter code. For
each data pair, the Pearson correlation coefficient, 90th percentile
difference in absolute concentrations, the coefficient of divergence, and
number of observations are given.
In general, the variability in O3 concentrations is larger than for PM2 5 concentrations and
comparable to that obtained for PM10_2 5. Intersite correlation coefficients in some areas (e.g.,
Philadelphia, PA; Atlanta, GA; Portland, OR) can be very similar for both PM2 5 and for O3.
However, there is much greater variability in the concentration fields of O3 as evidenced by the
much higher COD values. Indeed, COD values are higher for O3 than for PM25 in each of the
urban areas examined. In all of the urban areas examined for O3 some site pairs are always very
highly correlated with each other (i.e., r > 0.9) as seen for PM2 5. These sites also show less
AX3-50
-------
Bakersfield, CA MSA
a.
®
b.
Kilometers
AIRS Site ID
Site A 06-029-0007
Site B 06-029-0008
SiteC 06-029-0010
SiteD 06-029-0011
SiteE 06-029-0014
Site F 06-029-0232
SiteG 06-029-5001
Site H 06-029-6001
C.
200
Site
mean
(ppm)
A
B
A B
0.039 0.047
1 0.87
0.026
0.28
8299
1
C
0.029
0.82
0.030
0.40
7835
0.76
0.042
0.47
7779
D
0.047
0.62
0.038
0.35
8127
0.68
0.028
0.28
8079
E
0.030
0.80
0.032
0.45
8307
0.75
0.044
0.51
8251
F
0.038
0.88
0.019
0.24
7915
0.87
0.025
0.29
7880
G
0.043
0.87
0.022
0.24
8283
0.89
0.021
0.20
8244
H
0.030
0.80
0.031
0.40
8061
0.75
0.042
0.47
8006
D
0.62 0.95 0.88 0.84 0.94
0.047 0.014 0.026 0.032 0.015
0.48 0.30 0.37 0.43 0.29
7612 7791 7469 7772 7549
1
0.62 0.69 0.68 0.61
0.047 0.033 0.032 0.047
0.49 0.34 0.30 0.46
8096 7695 8066 7903
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
0.86 0.84 0.96
0.028 0.033 0.013
0.41 0.46 0.28
7862 8244 8024
1 0.89 0.85
0.021 0.028
0.25 0.37
7843 7626
1 0.84
0.033
0.42
8005
1
Figure AX3-34.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Bakersfield, CA MSA. The mean hourly
measured O3 concentration at each site is given above its letter code.
For each data pair, the Pearson correlation coefficient, 90th percentile
difference in absolute concentrations, the coefficient of divergence, and
number of observations are given.
variability in concentration and are probably influenced most strongly by regional production
mechanisms.
AX3-51
-------
a.
Los Angeles - Orange County, CA CMSA
b.
50
I
0
I
100
I
Kilometers
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
SiteG
SiteH
Site I
Sited
SiteK
SiteL
SiteM
SiteN
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
06-037-
0002
0016
•0030
0113
1002
1103
1201
1301
•1601
•1701
2005
4002
5001
9002
Figure AX3-35.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Los Angeles-Orange County, CA CMSA.
The mean hourly measured O3 concentration at each site is given above
its letter code. For each data pair, the Pearson correlation coefficient,
90th percentile difference in absolute concentrations, the coefficient of
divergence, and number of observations are given.
AX3-52
-------
Los Angeles - Orange County, CA CMSA
c.
Site
mean
(ppm)
A
P
A B
0.021 0.029
1 095
0017
032
8378
1
C
0.021
087
0020
033
6723
082
0029
036
6722
1
D
0.020
064
0028
041
8387
060
0034
042
8386
0.75
0.021
0.32
6731
E
0.018
088
0018
038
8202
086
0028
044
8201
0.88
0.018
0.35
6546
F
0.019
085
0020
036
8292
080
0029
045
8291
0.95
0.010
0.22
6636
G H
0.021 0.013
0 82 0 72
0021 0029
0 37 0 44
8364 8362
0 80 0 67
0027 0038
0 42 0 53
8363 8361
0.84 0.83
0.019 0.020
0.29 0.37
6708 6729
I
0.016
089
0019
0 41
8381
084
0030
052
8380
0.91
0.013
0.37
6725
J
0.013
091
0022
046
8386
088
0032
059
8385
0.85
0.020
0.49
6730
K
0.021
091
0015
033
8384
087
0024
039
8383
0.90
0.017
0.24
6728
L
0.022
061
0029
044
8379
057
0033
047
8378
0.72
0.022
0.31
6723
M N
0.027 0.042
0 55 0 58
0035 0048
0 45 0 47
8341 6321
0 53 0 56
0034 0042
0 46 0 42
8340 6321
0.68 0.61
0.030 0.048
0.37 0.45
6686 4998
H
K
M
N
0.76 0.82 0.73 0.77 0.77 0.69 0.73 0.75 0.81 0.62
0.023 0.018 0.023 0.023 0.022 0.027 0.024 0.021 0.025 0.046
0.31 0.31 0.31 0.36 0.35 0.43 0.34 0.31 0.30 0.40
8210 8300 8372 8370 8389 8394 8392 8387 8349 6321
1 0.90 0.86 0.79 0.90 0.87 0.91 0.67 0.66 0.59
0.014 0.019 0.023 0.015 0.019 0.014 0.027 0.034 0.050
0.29 0.31 0.35 0.33 0.38 0.28 0.37 0.37 0.45
8184 8187 8185 8204 8210 8207 8202 8164 6322
1
0.85 0.85 0.93 0.86 0.91 0.76 0.74 0.66
0.018 0.019 0.012 0.019 0.015 0.021 0.029 0.047
0.31 0.36 0.33 0.42 0.26 0.35 0.40 0.50
8277 8275 8294 8300 8297 8292 8254 6325
1 0.74 0.86 0.84 0.86 0.67 0.68 0.70
0.026 0.020 0.023 0.017 0.026 0.028 0.042
0.36 0.35 0.45 0.31 0.36 0.36 0.43
8347 8366 8371 8369 8364 8349 6326
1 0.86 0.74 0.80 0.84 0.73 0.61
0.017 0.018 0.025 0.021 0.033 0.053
0.26 0.37 0.38 0.35 0.43 0.51
8364 8369 8367 8362 8324 6299
1 0.89 0.93 0.77 0.71 0.66
0.016 0.015 0.022 0.032 0.048
0.31 0.30 0.34 0.39 0.48
8388 8386 8381 8343 6321
1 0.90 0.67 0.62 0.63
0.021 0.028 0.037 0.053
0.43 0.46 0.50 0.56
8391 8386 8348 6323
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
0.71 0.65 0.65
0.025 0.031 0.045
0.37 0.40 0.46
8384 8346 6321
1 0.74 0.62
0.025 0.043
0.33 0.42
8341 6321
1 0.63
0.038
0.36
6299
1
Figure AX3-35 (cont'd).
AX3-53
-------
Riverside - Orange County, CS CMSA
a.
b.
50
i i
0
I
Kilometers
200
i
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
SiteG
SiteH
Sitel
Site J
SiteK
SiteL
SiteM
SiteN
SiteO
SiteP
SiteQ
SiteR
06-065-0012
06-065-2002
06-065-5001
06-065-6001
06-065-8001
06-065-9001
06-071-0001
06-071-0005
06-071-0012
06-071-0017
06-071-0306
06-071-1004
06-071-1234
06-071-2002
06-071-4001
06-071-4003
06-071-9002
06-071-9004
Figure AX3-36.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Riverside-Orange County, CA CMSA.
The mean hourly measured O3 concentration at each site is given above
its letter code. For each data pair, the Pearson correlation coefficient,
90th percentile difference in absolute concentrations, the coefficient of
divergence, and number of observations are given.
AX3-54
-------
c.
Site
Riverside - Orange County, CS CMSA
ABODE FGH I JKLMNOPQR
mean
(ppm)
0.037 0.035 0.042 0.033 0.026 0.036 0.030 0.047 0.048 0.044 0.032 0.022 0.038 0.020 0.037 0.032 0.040 0.026
0.71 0.77 0.84 0.79 0.81 0.76 0.70 0.69 0.69 0.75 0.76 0.55 0.75 0.76 0.81 0.53 0.80
0.032 0.030 0.026 0.034 0.027 0.031 0.038 0.037 0.034 0.031 0.037 0.036 0.040 0.029 0.028 0.037 0.033
0.33 0.35 0.34 0.42 0.33 0.38 0.38 0.39 0.37 0.36 0.47 0.38 0.54 0.35 0.35 0.38 0.41
8397 8392 8394 8386 8035 7741 8393 8051 7844 7498 8385 7815 8394 7997 8391 7887 8364
1 0.85 0.69 0.64 0.66 0.75 0.62 0.64 0.81 0.72 0.60 0.65 0.56 0.71 0.61 0.52 0.61
0.023 0.035 0.039 0.036 0.031 0.041 0.038 0.029 0.031 0.043 0.031 0.046 0.031 0.038 0.035 0.040
0.25 0.34 0.42 0.33 0.33 0.33 0.34 0.30 0.34 0.46 0.31 0.51 0.31 0.37 0.33 0.42
8394 8395 8388 8040 7919 8393 8233 8017 7678 8389 7990 8395 8179 8392 8059 8369
1
M
O
0.73 0.66 0.69 0.74 0.67 0.72 0.81 0.73 0.61 0.67 0.59 0.75 0.65 0.65 0.65
0.037 0.044 0.036 0.035 0.036 0.030 0.025 0.034 0.049 0.030 0.052 0.030 0.039 0.031 0.045
0.38 0.47 0.35 0.39 0.32 0.32 0.28 0.38 0.52 0.30 0.56 0.32 0.40 0.33 0.46
8393 8385 8034 7752 8391 8059 7852 7506 8384 7823 8393 8005 8390 7896 8362
1 0.90 0.91 0.79 0.70 0.68 0.67 0.77 0.87 0.57 0.84 0.75 0.87 0.47 0.89
0.023 0.019 0.029 0.041 0.041 0.039 0.031 0.028 0.039 0.032 0.033 0.023 0.040 0.024
0.33 0.25 0.34 0.41 0.45 0.42 0.37 0.42 0.41 0.51 0.38 0.33 0.44 0.35
8387 8036 7742 8393 8049 7842 7496 8386 7813 8395 7995 8392 7885 8364
1 0.88 0.76 0.66 0.63 0.60 0.74 0.92 0.50 0.93 0.69 0.87 0.42 0.94
0.026 0.031 0.048 0.046 0.045 0.032 0.017 0.042 0.018 0.037 0.025 0.042 0.015
0.36 0.38 0.51 0.54 0.51 0.42 0.34 0.50 0.43 0.46 0.39 0.52 0.36
8028 7735 8385 8042 7835 7490 8378 7806 8387 7988 8384 7878 8356
1 0.76 0.64 0.62 0.63 0.74 0.83 0.54 0.81 0.70 0.82 0.44 0.85
0.031 0.042 0.039 0.037 0.031 0.032 0.037 0.034 0.032 0.026 0.039 0.028
0.33 0.37 0.39 0.36 0.37 0.43 0.36 0.53 0.35 0.31 0.38 0.35
7402 8034 7709 7487 7171 8027 7610 8036 7640 8033 7545 8010
1 0.68 0.68 0.75 0.85 0.72 0.68 0.68 0.78 0.72 0.46 0.73
0.042 0.041 0.035 0.022 0.035 0.033 0.038 0.029 0.032 0.038 0.033
0.40 0.43 0.40 0.33 0.41 0.39 0.49 0.36 0.37 0.41 0.40
7740 8067 7758 7684 7740 7669 7744 7861 7739 7592 7720
1 0.72 0.65 0.74 0.66 0.59 0.65 0.80 0.71 0.55 0.70
0.030 0.033 0.038 0.051 0.037 0.054 0.030 0.041 0.036 0.047
0.26 0.26 0.40 0.56 0.27 0.59 0.31 0.39 0.28 0.46
8047 7840 7494 8384 7811 8393 7993 8390 7883 8363
1 0.73 0.70 0.63 0.64 0.60 0.78 0.68 0.63 0.65
0.022 0.039 0.049 0.027 0.052 0.030 0.040 0.024 0.046
0.19 0.43 0.59 0.24 0.61 0.34 0.42 0.16 0.48
7983 7836 8047 7940 8051 8059 8046 7899 8027
1 0.72 0.57 0.73 0.54 0.74 0.60 0.57 0.58
0.035 0.048 0.024 0.051 0.028 0.041 0.025 0.046
0.41 0.56 0.21 0.59 0.32 0.39 0.21 0.45
7576 7840 7677 7844 7826 7839 7689 7820
1 0.70 0.64 0.67 0.84 0.70 0.49 0.72
0.036 0.032 0.039 0.024 0.033 0.037 0.034
0.44 0.40 0.51 0.34 0.41 0.42 0.42
7494 7430 7498 7626 7493 7352 7477
1 0.46 0.95 0.68 0.88 0.38 0.92
0.044 0.013 0.041 0.027 0.045 0.018
0.54 0.40 0.50 0.43 0.55 0.40
7814 8386 7993 8383 7885 8357
1 0.42 0.65 0.49 0.50 0.49
0.048 0.028 0.039 0.023 0.042
0.57 0.31 0.40 0.22 0.45
7815 7817 7811 7665 7796
1 0.64 0.87 0.38 0.93
0.0430.0290.0470.019
0.54 0.50 0.58 0.45
7996 8392 7887 8364
1 0.72 0.58 0.70
0.033 0.030 0.037
0.37 0.33 0.43
7992 7842 7973
1 0.46 0.92
0.039 0.020
0.41 0.33
7882 8361
. 1 0.45
0.041
0.46
7866
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
Figure AX3-36 (cont'd).
AX3-55
-------
AX3.3.2 Small-scale Horizontal and Spatial Variability in Ozone
Concentrations
Ozone concentrations near roadways
Apart from the larger scale variability in surface O3 concentrations, there is also significant
variability on the micro-scale (< a few hundred meters), especially near roadways and other
sources of emissions that react with O3. These sources are not confined to urban areas. Sources
of emissions that react with O3 such as highways and power plants are also found in rural areas.
Johnson (1995) described the results of studies examining O3 upwind and downwind of
roadways in Cincinnati, OH. In these studies, O3 upwind of the roadway was about 50 ppb and
values as high as this were not found again until distances of about 100 m downwind. The O3
profile varied inversely with that of NO, as might be expected. For peak NO concentrations of
30 ppb, the O3 mixing ratio was about 36 ppb, or about 70% of the upwind value. The
magnitude of the downwind depletion of O3 depends on the emissions of NO, the rate of mixing
of NO from the roadway and ambient temperatures. So depletions of O3 downwind of roadways
are expected, but with variable magnitude, although scavenging of O3 by NO near roadways was
more pronounced before the implementation of stringent NOx emissions controls.
Guidance for the placement of O3 monitors (U.S. Environmental Protection Agency, 1998)
states a separation distance that depends on traffic counts. For example, a minimum separation
distance of 100 m from a road with 70,000 vehicles per day is recommended for siting an O3
monitor to avoid interference that would mean a site is no longer representative of the
surrounding area. An average rate of about 3,000 vehicles per hour passing by a monitoring site
implies a road with rather heavy traffic. As noted earlier in Section AX3.3.1 for the Lakewood,
CA monitoring, O3 levels are lower at sites located near traffic than those located some distance
away and the scavenging of O3 by emissions of NO from roadways is a major source of spatial
variability in O3 concentrations.
Vertical Variations in Ozone Concentrations
In addition to horizontal variability in O3 concentrations, consideration must also be given
to variations in the vertical profile of O3 in the lowest layers of the atmosphere. The planetary
boundary layer consists of an outer and an inner portion. The inner part extends from the surface
to about one-tenth the height of the planetary boundary layer. Winds and transported pollutants,
AX3-56
-------
such as O3, are especially susceptible to interactions with obstacles, such as buildings and trees
in the inner boundary layer (atmospheric surface layer) (e.g., Garratt, 1992). Inlets to ambient
monitors (typically at heights of 3 to 5 meters) are located in, and human and vegetation
exposures occur in this part of the boundary layer.
Photochemical production and destruction of O3 occurs throughout the planetary boundary
layer. However, O3 is also destroyed on the surfaces of buildings, vegetation, etc. On most
surfaces, O3 is destroyed with every collision. In addition, O3 is scavenged by NO emitted by
motor vehicles and soils. These losses imply that the vertical gradient of O3 should always be
directed downward. The magnitude of the gradient is determined by the intensity of turbulent
mixing in the surface layer.
Most work characterizing the vertical profile of O3 near the surface has been performed in
nonurban areas with the aim of calculating fluxes of O3 and other pollutants through forest
canopies and to crops and short vegetation etc. Corresponding data are sparse for urban areas.
However, monitoring sites are often set up in open areas such as parks and playgrounds where
surface characteristics may resemble those in rural areas more than those in the surrounding
urban area. The vertical profile of O3 measured over low vegetation is shown in Figure AX3-37.
These measurements were obtained as part of a field campaign to measure the fluxes of several
gas and aerosol phase pollutants using the gradient-flux technique in a remote area in Hortobagy
National Park in Hungary during late spring of 1994 (Horvath et al., 1995). The labels stable
and unstable in the figure refer to atmospheric stability conditions and average represents the
overall average. Ozone concentrations were normalized relative to their values at a height of
4 m. As can be seen from the figure, there was a decrease of about 20% in going from a height
of 4 m down to 0.5 m above the surface during stable conditions, but O3 decreased by only about
7% during unstable conditions. The average decrease was about 10% for all measurements. As
might be expected, O3 concentrations at all heights were very highly correlated with one another.
Of course, these values represent averages and there is scatter about them, particularly under
strongly stable conditions. However, these conditions tend to occur mainly during night and the
stability regime during the day in urban areas tends more towards instability because of the
urban heat island effect. Figure AX3-38 shows the vertical profile of O3 obtained in a spruce
forest in northwestern Hungary in late summer 1991 (Horvath et al., 2003). The fall off in O3 for
AX3-57
-------
4-i
+•»
.c
O)
"i
I
1 -
0.5-
0
^ Average (n = 1797)
+ Stable (n= 937)
• Unstable (n = 860)
x
y 4
Y
0.5
0.6 0.7 0.8
Relative O3 Concentration
0.9
Figure AX3-37. Vertical profile of O3 obtained over low vegetation. Values shown are
relative to concentrations at 4 m above the surface. Ozone
concentrations for stable and unstable conditions were 41.3 and 24.1
ppb, and average O3 concentration weighted by stability class was 33.1
ppb at 4 m.
Source. Horvathetal. (1995).
this case is due to uptake by trees, reaction with ambient NO and with NO emitted by the soil in
the forest in addition to deposition on the surface.
AX3.3.3 Ozone Concentrations at High Elevations
The distributions of hourly average concentrations experienced at high-elevation cities are
similar to those experienced in low-elevation cities. For example, the distribution of hourly
average concentrations for several O3 sites located in Denver were similar to distributions
observed at many low-elevation sites elsewhere in the United States. However, the use of
absolute concentrations (e.g., in units of micrograms per cubic meter) in assessing the possible
impacts of O3 on vegetation at high-elevation sites instead of mixing ratios (e.g., parts per
million) may be an important consideration (see Chapter 9, for further considerations about
exposure and effective dose considerations for vegetation assessments).
AX3-58
-------
19-,
15-
+* 10-
O)
"i
I
2,5-
0.5-
^ Average (n = 1760)
+ Stable (n = 1272)
• Unstable (n = 488)
0.5
0.6 0.7 0.8 0.9
Relative O3 Concentration
Figure AX3-38. Vertical profile of O3 obtained in a spruce forest. Values shown are
relative to concentrations at 19 m above the surface. Mean tree height is
14.5 m. Ozone concentrations for stable and unstable conditions were
36.7 and 33.8 ppb, and the average O3 concentration weighted by stability
class was 34.6 ppb at 19 m.
Source: Horvath et al. (2003).
Concentrations of O3 vary with altitude and latitude. Although a number of reports contain
data on O3 concentrations at high altitudes (e.g., Coffey et al., 1977; Reiter, 1977; Singh et al.,
1977; Evans et al., 1985; Lefohn and Jones, 1986), fewer reports present data for different
elevations in the same locality. Monitoring data collected by the Mountain Cloud Chemistry
Project (MCCP) provide useful information about O3 exposure differences at different
elevations. When applying different exposure indices to the MCCP data, there appears to be no
consistency in the relationship between O3 exposure and elevation.
Lefohn et al. (1990a) summarized the characterization of gaseous exposures at rural sites in
1986 and 1987 at several MCCP high-elevation sites (see Table AX3-6). Aneja and Li (1992)
have summarized the O3 concentrations for 1986 to 1988 (see Table AX3-7). Table AX3-7
summarizes the concentrations and exposures that occurred at several of the sites for the period
1987 to 1988.
AX3-59
-------
Table AX3-6. Description of Mountain Cloud Chemistry Program Sites
Site Elevation (m)
X
OJ
ON
Rowland Forest (HF1), ME
Mt. Moosilauke (MSI), NH
Whiteface Mountain (WF1), NY
Shenandoah NP (SHI), VA
Shenandoah NP (SH2), VA
Shenandoah NP (SH3), VA
Whitetop Mountain (WT1), VA
Mt. Mitchell (MM 1),NC
Mt. Mitchell (MM2), NC
65
1000
1483
1015
716
524
1689
2006
1760
458°
438°
448°
38°
38°
38°
36°
35°
35°
Latitude
ir
59'
23'
37'
37'
37'
38'
44'
45'
18"
26"
12"
30"
45"
20"
15"
68°
71°
73°
78°
78°
78°
81°
82°
82°
Longitude
46'
48'
51'
20'
21'
21'
36'
17'
15'
28"
34"
48"
13"
28"
21"
15"
-------
Table AX3-7. Seasonal (April-October) Percentiles, SUM06, SUM08, and W126 Values for the MCCP Sites
X
Site
Howland Forest, ME
(HF1)
Mt. Moosilauke, NH
(MSI)
Whiteface Mountain,
NY (WF1)
(36-031-0002)
Whiteface Mountain,
NY (WF3)
Whiteface Mountain,
NY (WF4)
Mt. Mitchell, NC
(MM1)
Mt. Mitchell, NC
(MM2)
Shenandoah Park,
VA (SHI)
Shenandoah Park,
VA (SH2)
Year
1987
1988
1987
1988
1987
1988
1987
1987
1987
1988
1989
1992
1987
1988
1987
1988
1987*
1988
Min.
0.000
0.000
0.006
0.010
0.011
0.014
0.010
0.000
0.008
0.011
0.010
0.005
0.017
0.009
0.000
0.006
0.003
0.006
10
0.013
0.012
0.027
0.026
0.029
0.025
0.025
0.011
0.034
0.038
0.038
0.036
0.032
0.029
0.023
0.024
0.027
0.029
30
0.021
0.021
0.036
0.033
0.037
0.033
0.033
0.023
0.044
0.054
0.047
0.043
0.042
0.041
0.036
0.036
0.040
0.042
50
0.028
0.028
0.045
0.043
0.046
0.043
0.039
0.031
0.051
0.065
0.054
0.048
0.049
0.050
0.044
0.047
0.049
0.054
70
(ppm)
0.035
0.036
0.053
0.055
0.053
0.056
0.047
0.041
0.058
0.075
0.059
0.053
0.056
0.060
0.054
0.058
0.059
0.064
90
0.046
0.047
0.065
0.076
0.067
0.078
0.064
0.056
0.067
0.095
0.068
0.063
0.067
0.080
0.069
0.077
0.071
0.083
95
0.052
0.054
0.074
0.087
0.074
0.089
0.075
0.065
0.074
0.106
0.072
0.069
0.073
0.092
0.076
0.087
0.077
0.095
99
0.065
0.076
0.086
0.113
0.087
0.110
0.091
0.081
0.085
0.126
0.081
0.081
0.083
0.110
0.085
0.103
0.086
0.108
Max.
0.076
0.106
0.102
0.127
0.104
0.135
0.117
0.117
0.105
0.145
0.147
0.096
0.096
0.162
0.135
0.140
0.145
0.145
No.
Obs.
4766
4786
4077
2835
4703
4675
4755
4463
3539
2989
2788
3971
3118
2992
3636
3959
2908
4661
SUM06
5.9
10.9
45.0
51.9
63.5
94.4
45.4
23.8
59.4
145.1
54.8
37.8
47.0
68.7
54.2
80.9
55.7
133.8
SUM08
(ppm-h)
0.0
2.9
9.5
21.2
12.2
40.8
14.4
5.1
7.8
69.7
3.5
4.4
5.1
28.1
8.5
29.6
7.8
55.8
W126
7.7
11.6
40.1
43.4
50.5
78.3
40.3
21.3
46.5
116.6
40.7
36.7
37.4
57.7
42.0
67.2
41.8
109.4
-------
X
OJ
ON
to
Table AX3-7 (cont'd). Seasonal (April-October) Percentiles, SUM06, SUM08, and W126 Values for the MCCP Sites
Site
Shenandoah Park,
VA (SH3)
Whitetop Mountain,
VA (WT1)
Year
1987
1988
1987
1988
Min.
0.000
0.000
0.01
0.000
10
0.018
0.020
0.038
0.030
30
0.029
0.031
0.051
0.046
50
0.037
0.040
0.059
0.058
70
(ppm)
0.047
0.051
0.066
0.068
90
0.061
0.067
0.078
0.084
95
0.068
0.076
0.085
0.094
99
0.080
0.097
0.096
0.119
Max.
0.108
0.135
0.111
0.163
No.
Obs.
3030
4278
4326
3788
SUM06
23.1
52.3
147.7
133.8
SUM08
(ppm-h)
2.6
15.6
32.4
51.0
W126
19.2
44.2
105.7
102.8
*Calculations based on a May-September season.
-------
In 1987, the 7- and 12-h seasonal means were similar at the Whiteface Mountain WF1 and
WF3 sites (Figure AX3-39a). The 7-h mean values were 0.0449 and 0.0444 ppm, respectively,
and the 12-h mean values were 0.0454 and 0.0444 ppm, respectively. Note that, in some cases,
the 12-h mean was slightly higher than the 7-h mean value. In fact, at higher elevations, the 24-h
level was larger still (Table AX3-8). This resulted when the 7-h mean period (0900 to
1559 hours) did not capture the period of the day when the highest hourly mean O3
concentrations were measured. A similar observation was made, using the 1987 data, for the
MCCP Shenandoah National Park sites. The 7-h and 12-h seasonal means were similar for the
SHI and SH2 sites (Figure AX3-39b). Based on cumulative indices, the Whiteface Mountain
summit (1483-m) site (WF1) experienced a higher exposure than the WF3 (1026-m) site
(Figure AX3-39c). The site at the base of Whiteface Mountain (WF4) experienced the lowest
exposure of the three O3 sites. Among the MCCP Shenandoah National Park sites, marginally
higher O3 exposures were found at the SH2 site, based on the index that sums all of the hourly
average concentrations (referred to as "total dose" in Figure AX3-39c) and sigmoidal values,
than the SHI high-elevation site (Figure AX3-39d). The reverse was true for concentrations
>0.07 ppm (SUM07) and the number of hourly concentrations >0.07 ppm the sums of the
concentrations >0.07 ppm and the number of hourly concentrations >0.07 ppm. When the Big
Meadows, Dickey Ridge, Sawmill Run, and Shenandoah National Park data for 1983 to 1987
were compared, it again was found that the 7-h and 12-h seasonal means were insensitive to the
different O3 exposure patterns. Cumulative exposures at three sites in Shenandoah National Park
are shown in Figure AX3-40. There was no evidence that the highest elevation site, Big
Meadows, consistently had experienced higher O3 exposures than the other sites. In 2 of the
5 years, the Big Meadows site experienced lower exposures than the Dickey Ridge and Sawmill
Run sites, based on the sum of all concentration or sigmoidal indices. For 4 of the 5 years, the
SUM07 index yielded the same result.
Taylor et al. (1992) indicated that the forests they monitored experienced differences in O3
exposure. The principal spatial factors underlying this variation were elevation, proximity to
anthropogenic sources of oxidant precursors, regional-scale meteorological conditions, and
airshed dynamics between the lower free troposphere and the surface boundary layer.
Table AX3-8 summarizes the exposure values for the 10 EPRI Integrated Forest Study sites
located in North America.
AX3-63
-------
0.05,
0.00
0.05, b.
WF 1
WF3
WF4
0.00
SH 1
SH2
SH3
200 n C.
• Total Dose
• Sigmoidal
D Sum 20.07 ppm
200n
S 100
-------
a. May - September 1983
300 -i
Total Dose
Big Meadows Dickey Ridge Sawmill Run
c. June - September 1985
300 -i
D Sum >0.07 ppm £•
£ 200 -
Q.
Q.
0 -b
• Total Dose
GSigmoidal
DSum>0.07 ppm
Big Meadows Dickey Ridge Sawmill Run
b. May - September 1984
300 -i
Total Dose
DSum>0.07 ppm -r-
d. May - September 1986
300 -i
Total Dose
Big Meadows Dickey Ridge Sawmill Run
Big Meadows Dickey Ridge Sawmill Run
e. May - September 1987
300 -i
• Total Dose
| Sigmoidal
DSum>0.07 ppm
Big Meadows Dickey Ridge Sawmill Run
Figure AX3-40a-e. Cumulative exposures for three non-Mountain Cloud Chemistry
Program Shenandoah National Park sites, 1983 to 1987.
Source: Lefohnetal. (1990b).
AX3-65
-------
Table AX3-8. Summary Statistics for 11 Integrated Forest Study Sites3
Site Year
HIGH ELEVATION SITES
Whiteface Mtn, NY
Great Smoky Mtns NP
Coweeta Hydrologic Lab,
NC
LOW ELEVATION SITES
Huntington Forest, NY
Howland, ME
Oak Ridge, TN
Thompson Forest, WA
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
Quarter
2
3
2
3
2
o
5
2
o
3
2
3
2
3
2
3
2
o
5
2
o
3
2
3
2
3
2
o
3
2
o
5
2
3
24-h
42
45
49
44
54
53
71
59
50
47
61
57
36
24
40
37
34
26
36
24
42
29
40
32
36
30
32
32
12-h
43
44
50
43
52
51
70
57
48
44
59
54
42
32
46
46
39
32
41
30
53
44
57
47
43
36
39
39
7-h
42
43
49
43
49
49
68
55
47
42
59
51
42
33
46
48
39
31
41
30
50
41
58
51
41
34
37
36
1-h Max.
104
114
131
119
99
95
119
120
85
95
104
100
88
76
106
91
69
76
90
71
112
105
104
122
103
94
103
140
SUM06
103 ppb-h
13.2
30.1
33.5
22.6
57.1
34.3
126.3
74.7
32.4
24.1
81.6
63.6
9.8
5.4
19.2
18.6
1.9
3.8
8.1
1.7
39.5
24.3
26.4
19.7
10.7
10.3
8.1
13.5
SUM08
103 ppb-h
2.5
11.8
13.9
10.4
10.9
8.8
61.2
22.2
2.6
2.4
18.5
19.8
0.9
0.2
6.1
2.7
0.0
0.0
2.9
0.0
13.5
9.0
9.8
7.7
3.6
2.1
2.3
6.7
AX3-66
-------
Table AX3-8 (cont'd). Summary Statistics for 11 Integrated Forest Study Sites3
Site
LOW ELEVATION SITES
B.F. Grant Forest, GA
Gainesville, FL
Duke Forest, NC
Nordmoen, Norway
Year
(cont'd)
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
Quarter
2
3
2
3
2
o
3
2
o
3
2
3
2
3
2
o
3
2
o
3
24-h
32
33
47
32
42
29
35
20
38
52
54
38
32
14
22
11
12-h
46
52
63
47
53
44
48
29
48
59
69
51
40
18
28
15
7-h
48
54
64
48
50
41
51
30
52
50
75
54
41
20
29
16
1-h Max.
99
102
127
116
b
b
84
70
100
124
115
141
75
32
53
30
SUM06
103 ppb-h
26.1
31.3
53.1
24.1
b
b
23.4
1.9
29.2
b
b
52.9
2.4
0.0
0.0
0.0
SUM08
103 ppb-h
5.1
10.3
21.9
7.4
b
b
0.5
0.1
7.8
b
b
23.4
0.0
0.0
0.0
0.0
"Concentration in ppb.
bData were insufficient to calculate statistic.
Source: Taylor etal. (1992).
the temperature of a gas is held constant, the volume occupied by the gas varies inversely with
the pressure (i.e., as pressure decreases, volume increases). This pressure effect must be
considered when measuring absolute pollutant concentrations. At any given sampling location,
normal atmospheric pressure variations have very little effect on air pollutant measurements.
However, when mass/volume units of concentration are used and pollutant concentrations
measured at significantly different altitudes are compared, pressure (and, hence, volume)
adjustments are necessary. In practice, the summit site at Whiteface Mountain had a slightly
AX3-67
-------
higher O3 exposure than the two low-elevation sites (Lefohn et al., 1991). However, at
Shenandoah National Park sites, the higher elevation site experienced lower exposures than
lower elevation sites in some years.
These exposure considerations are relatively unimportant at low-elevation sites. However,
the differences between exposure-effects at high- and low-elevation sites, may be significant
(Lefohn et al., 1990b). In particular, assuming the sensitivity of the biological target to be
identical at low- and high-elevations, some adjustment will be necessary when attempting to link
experimental data obtained at low-elevation sites with air quality data monitored at the high-
elevation stations. This topic is further discussed in Annex AX9 when considering effective
dose considerations for predicting vegetation effects associated with O3.
AX3.4 DIURNAL PATTERNS IN OZONE CONCENTRATION
AX3.4.1 Introduction
Diurnal variations in O3 at a given location are controlled by a number of factors such as
the relative importance of transport versus local photochemical production and loss rates, the
timing for entrainment of air from the nocturnal residual boundary layer and the diurnal
variability in mixing layer height.
The form of an average diurnal pattern provides some information on sources, transport,
and chemical formation and destruction effects at various sites (Lefohn, 1992). Atmospheric
conditions leading to limited transportation from source regions will produce early afternoon
peaks. However, long-range transport processes will influence the actual timing of a peak, from
afternoon to evening or early morning hours. Ozone is rapidly depleted near the surface below
the nocturnal inversion layer (Berry, 1964). Mountainous sites, which are above the nocturnal
inversion layer, do not necessarily experience this depletion (Stasiuk and Coffey, 1974). Taylor
and Hanson (1992) reported similar findings, using data from the Integrated Forest Study. The
authors reported that intraday variability was most significant for the low-elevation sites due to
the pronounced daily amplitude in O3 concentration between the predawn minimum and
mid-afternoon-to-early evening maximum. The authors reported that interday variation was
more significant in the high-elevation sites. Ozone trapped below the inversion layer is depleted
by dry deposition and chemical reactions if other reactants are present in sufficient quantities
AX3-68
-------
(Kelly et al., 1984). Above the nocturnal inversion layer, dry deposition does not generally
occur, and the concentration of O3 scavengers is generally lower, so O3 concentrations remain
fairly constant (Wolff et al., 1987). A flat diurnal pattern is usually interpreted as indicating a
lack of efficient scavenging of O3 or a lack of photochemical precursors, whereas a strongly
varying diurnal pattern is taken to indicate the opposite.
An analysis that identified when the highest hourly average concentrations were observed
at rural agricultural and forested sites was described in 1996 O3 AQCD. A review of the hourly
average data collected at all rural agricultural and forested sites in Environmental Protection
Agency's AQS database for 1990 to 1992 was undertaken to evaluate the percentage of time
hourly average concentrations >0.1 ppm occurred during the period of 0900 to 1559 hours in
comparison with the 24-h period. It was found that 70% of the rural-agricultural and forested
sites used in the analysis experienced at least 50% of the occurrences >0.1 ppm during the period
of 0900 to 1559 hours when compared to the 24-h period. When O3 monitoring sites in
California were eliminated, approximately 73% of the remaining sites experienced at least 50%
of the occurrences >0.10 ppm during the daylight 7-h period when compared with the
24-h period.
Diurnal Variations in the Nationwide Data Set
Composite urban, diurnal variations in hourly averaged O3 for April through October 2000
to 2004 are shown in Figure AX3-41. As can be seen from Figure AX3-41, daily 1-h O3 maxima
tend to occur in mid-afternoon and daily 1-h O3 minima tend to occur during the early morning.
However, there is also considerable spread in these times. Therefore, some caution must be
exercised in extrapolating results from one city to another and when attempting to judge the time
of day when the daily 1-h maximum occurs.
Corresponding data for 8 hour average O3 data are shown in Figure AX3-42. As can be
seen from Figure AX3-42, daily maximum eight hour O3 concentrations tend to occur from
about 10 a.m. to about 6 p.m. As can be seen from Figure AX3-42, they can also occur at
slightly different times and the variation in the 8-h averages is smoother than for the 1-h
averages. The minima in the 8 h averages tend to occur starting at about midnight.
AX3-69
-------
Urban Sites
0.200 -i
-p 0.150
Q.
a.
o
c
to
o
o
O
0)
c
8
o
0.100 -
0.050 -
0.000 -I
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-41. Composite, nationwide diurnal variability in hourly averaged O3 in urban
areas. Values shown are averages from April to October 2000 to 2004.
Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
AX3.4.2 Diurnal Patterns in Urban Areas
Diurnal Variations in EPA's 12 Cities
The diurnal variability of hourly averaged O3 in the twelve urban areas considered for
inclusion in EPA's human health exposure assessment-risk assessment for the current review is
illustrated in Figure AX3-43a-l for April to October. Daily maximum 1-h concentrations tend to
occur in mid-afternoon. However, as can be seen from the figures, the diurnal patterns vary
from city to city, with high values (>0.100 ppm) occurring either late in the evening as in
Boston, past midnight as in Los Angeles and Sacramento, or mid-morning as in Houston.
Typically, high values such as these are found during the daylight hours in mid to late afternoon.
The reasons for the behavior of O3 during the night at the above mentioned locations are not
clear. Measurement issues may be involved or there may be physical causes such as transport
AX3-70
-------
Urban Sites
0.100 -
Q.
-------
a. Boston-Worcester-Manchester, MA-NH
b. New York-Newark-Bridgeport, NY-NJ-CT-PA
— 0.125
I
£ 0-075
o
O
E 0.125-
O
S3 0.075
£ 0.025
O
230001 02 03 0405 06 07 08 091011 12 13 14 15 15 17 18 19 20 21 22230001
hour
22230001 02 03 04 05 06 07 08 09 1011 12 13 14 15 16 17 18 19 20 21 22230001
hour
c. Philadelphia-Camden-Vineland, PA-NJ-DE-ME
d. Washington-Baltimore-Northern Virginia, DC-MD-VA-WV
— 0.125 •
~ 0.125 •
C 0.100 •
a
I
I °°7!H
s
o
0 0.050-1
i
O
N
0 0.025-1
22230001020304050607
08091011 121314 151617 1819202122230001
hour
22230001 02030405 06 07 08 09 10 11 12 13 14 15 16 17 18 1920 21 22230001
hour
e. Atlanta-Sandy Springs-Gainesville, GA-AL
— 0.125-
O
O
f. Cleveland-Akron-Elyria, OH
*~- 0.125 •
0 0,050-
I
22 23 00 01 02 03 04 05 C6 07 OS 09 10 11 12 13 14 15 16 17 18 '9 20 21 22 23 00 01
hour
22230001 0203 04 00 06 Of 08 09 1011 12 13 14 15 16 17 18 192021 22230001
hour
Figure AX3-43a-f. Diurnal variability in hourly averaged O3 in selected urban areas.
Values shown are averages from April to October 2000 to 2004.
Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
AX3-72
-------
g. Detroit-Warren-Flint, MI
h. Chicago-Naperville-Michigan City, IL-IN-WI
o
o
o
o
O 0,050
22 23 00 01 02 03 34 05 06 07 08 09 10 11 12 13 14 15 16 17 18 16 20 21 22 23 00 01
hour
22 23 00 01 C2 03040506070809 10 11 12 13 14 15 16 17 18192021 22 230301
hour
i. St. Louis-St. Charles-Farmington, MO-IL
s
S 0100
o
o
era
j. Houston-Baytown-Huntsville, TX
o
o
22 23 00 01 02 03 34050607 08 09 10 11 12 13 14 15 16 17 18 192021 22 23 00 01
hour
2223 0301 02 03 04 0506070809 10 11 12 13 14 15 16 1718 1920 21 222300 01
hour
k. Sacramento-Arden-Arcade-Truckee, CA-NV
o
1
o
o
I. Los Angeles-Long Beach-Riverside, CA
1 0203340506070809 1011 12 13 14 15 16 17 18 1S 20 21 22230001
hour
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-43g-l. Diurnal variability in hourly averaged O3 in selected urban areas.
Values shown are averages from April to October 2000 to 2004.
Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
AX3-73
-------
a. Boston-Worcester-Manchester, MA-NH
b. New York-Newark-Bridgeport, NY-NJ-CT-PA
22230001 02030405060708091011 12 13 14 15 16 1? 18 19 20 21 22230001
hour
22230001 02 03 04 05 08 07 OS 09 10 11 12131415161718192021 22 23 00 C1
hour
c. Philadelphia-Camden-Vineland, PA-NJ-DE-ME
d. Washington-Baltimore-Northern Virginia, DC-MD-VA-WV
22 23 00 01 02 03 3405 06 07 08 09 10 11 12 13 14 15 16 17 18 18 20 21 22 23 00 01
hour
E c.100 •
o
o
22230001 02030435060708091011 12 13 14 15 16 17 18 19 2C 2' 22230001
hour
e. Atlanta-Sandy Springs-Gainesville, GA-AL
f. Cleveland-Akron-Elyria, OH
E c.100 •
o
8 0.02H
22 23 CO 01 02030405060708091011 12'3 14 15 16 17 18 19 20 21 22 23 CO 01
hour
2223 0001 02 03 04 35080708 09 10 11 12 13 14 15 16 17 18 1920 2' 22 2300 C1
hour
Figure AX3-44a-f. Diurnal variability in 8 hour averaged O3 in selected urban areas.
Values shown are averages from April to October 2000 to 2004.
Boxes define the interquartile range and the whiskers, the minima
and maxima. The hour refers to the start of the 8-h averaging period.
Source: Fitz-Simons et al. (2005).
AX3-74
-------
g. Detroit-Warren-Flint, MI
h. Chicago-Naperville-Michigan City, IL-IN-WI
—• 0.125 •
O
o
•—. 0.125
O
o
22 23 00 01 02 03 34 05 06 07 08 09 10 11 12 13 14 15 16 17 18 16 20 21 22 23 00 01
hour
22 23 00 01 C2 03040506070809 10 11 12 13 14 15 16 17 18192021 22 230301
hour
i. St. Louis-St. Charles-Farmington, MO-IL
j. Houston-Baytown-Huntsville, TX
.-* 0 1?5
s
S 0.075
O
o
22 23 00 01 02 03 34050607 08 09 10 11 12 13 14 15 16 17 18 192021 22 23 00 01
hour
o
o
2223 0301 02 03 04 D5 06 0708 09 10 11 12 13 14 15 16 1718 1920 21 222300 01
hour
k. Sacramento-Arden-Arcade-Truckee, CA-NV
I. Los Angeles-Long Beach-Riverside, CA
.—. 0.125 •
O
1
O
o
22230001 0203340506070809 1011 12 13 14 15 16 17 18 1S 20 21 22230001
hour
— 0 125 -
B 0100
O
1
91011 12131415161718192021 22230001
hour
Figure AX3-44g-l. Diurnal variability in 8 hour averaged O3 in selected urban areas.
Values shown are averages from April to October 2000 to 2004.
Boxes define the interquartile range and the whiskers, the minima
and maxima. The hour refers to the start of the 8-h averaging period.
Source: Fitz-Simons et al. (2005).
AX3-75
-------
On days with high 1-h daily maximum concentrations (e.g., >0.12 ppm) the maxima tend
to occur in a smaller time window centered in the middle of the afternoon, compared to days in
which the maximum is lower. For example, on the high O3 days the 1-h maximum occurs from
about 11 a.m. to about 6 p.m. However, on days in which the 1-h daily maximum is
<0.080 ppm, the daily maximum can occur at any time during the day or night, with only a 50%
probability that it occurs between 1 and 3 p.m., in each of the 12 cities. Photochemical reactions
in combination with diurnal emissions patterns are expected to produce mid-afternoon peaks in
urban areas. These results suggest that transport from outside the urban airshed plays a major
role in determining the timing of the daily maxima for low peak O3 levels. This pattern is more
typical for the Los Angeles-Long Beach-Riverside, CA area even for high O3 days.
The same general timing patterns are found for 1-h daily maximum O3 concentrations as
for the daily maximum 8-h average O3 concentration. As mentioned above, the daily maximum
8-h O3 concentrations are generally found between the hours of 10 a.m and 6 p.m. However,
there are a significant number of days when this is not the case for high values, as in Houston,
TX and Los Angeles, CA, or in general for low values at any of the cities examined. Although
the 8-h average O3 concentration is highly correlated with the daily maximum 1-h average O3
concentration, there are situations where the daily maximum 8-h average O3 concentration may
be driven by very high values in the daily maximum 1-h average O3 concentration as illustrated
in Figure AX3-43J. In cases such as these, the predicted 8-h average might overestimate the
short-term O3 concentration later in the day.
As an aid to better understanding the nature of the diurnal patterns shown in the figures for
EPA's 12 cities, Figures AX3-45a-d show the hours in which the 1-h daily maximum O3
concentration occurs in four of the cities. As can be seen from Figures AX3-45a-c for the
Philadelphia, Atlanta, and Houston areas, the maximum tends to occur from about 2 p.m. to
4 p.m. about half of the time, and most values occur between about 12 p.m to 6 p.m at higher
values of the daily maximum 1-h O3 concentration. Although values at Houston can occur
earlier, these are most likely due to episodic releases from the petrochemical industries.
For small values of the 1-h daily maximum, most of the daily maxima still occur in the
afternoon, but maxima can also occur at any time of the day or night. In the Los Angeles area,
as shown in Figure AX3-45d high values of daily 1-h O3 maxima can occur at any time during
the day or night but with most values occurring during the afternoon.
AX3-76
-------
a. Philadelphia-Camden-Vineland, PA-NJ-DE-MD b. Atlanta-Sandy Springs-Gainesville, GA-AL
24 -i 24-1
18-
12 -
6-
O-l
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
c. Houston-Baytown-Huntsville, TX
24 •
18-
12 -
6-
0-
E
]E
3E
3E
3E
3E
3E
llii
]TTT
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
d. Los Angeles-Long Beach-Riverside, CA
24 -i
18 -
12 -
6 -
0 -I
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
Figure AX3-45a-d. Time of occurrence of daily maximum 1-h O3 concentration in four
cities, averaged from April to October, 2000 to 2004.
Source: Fitz-Simons et al. (2005).
Figures AX3-46a-d show the hours in which the 8-h daily maximum O3 concentration
begins. The mean time is about 10 a.m. at these four cities indicating that the 8-h daily
maximum tends to occur on average from about 10 a.m to 6 p.m. However, there can be
deviations from these times. The same general pattern in which the maxima tend to occur within
a narrower time frame at high values than at low values is found in the four cities shown.
AX3-77
-------
a. Philadelphia-Camden-Vineland, PA-NJ-DE-MD b. Atlanta-Sandy Springs-Gainesville, GA-AL
24 •
18-
12 -
6-
0-
E
]E
] E
3!
.
--
3!
.
3^
24 •
18 -
3 E^S e H
6 -
, . . ^- 0 -
C
H C
::
]C
4^Hfa
<0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14
c. Houston-Baytown-Huntsville, TX
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
d. Los Angeles-Long Beach-Riverside, CA
24 •
18-
-
12 -
-
6-
0-
L
i E
b c
] C
p c
DC
.
DC
DE
3E
24 •
18 -
3
212-
3 9~ ;
6 -
, 0 -
C
1 C
1 C
1 C
D C
D C
H C
_
i cp i±i L±I
^
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
< 0.07 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 > 0.14
Figure AX3-46a-d. Time of occurrence of daily maximum 8-h average O3 concentration
in four cities, averaged from April to October, 2000 to 2004. The
hour refers to the start of the 8-h averaging period.
Source: Fitz-Simons et al. (2005).
The patterns of diurnal variability for both 1-h and 8-h averages have remained quite stable
over the 15 year period from 1990 to 2004, with times of occurrence of the daily maxima
varying by no more than an hour from year to year in each of the 12 cities.
AX3-78
-------
Weekday/Weekend Differences
In addition to varying diurnally, O3 concentrations differ from weekdays to weekends.
Heuss et al. (2003) described the results of a nationwide analysis of weekday/weekend
differences that demonstrated significant variation in these differences across the United States.
Weekend 1-h or 8-h maximum O3 concentrations varied from 15% lower to 30% higher than
weekday levels across the U.S. The weekend O3 increases were primarily found in and around
large coastal cities in California and large cities in the Midwest and Northeast Corridor. Many
sites that experienced elevated weekday O3 concentrations also had higher O3 on weekends even
though the traffic and O3 precursor levels were substantially reduced on weekends. The authors
reported that detailed studies of this phenomenon indicated that the primary cause of the
higher O3 on weekends was the reduction in oxides of nitrogen emissions on weekends in a
volatile organic compound (VOC)-limited (NOx-saturated) chemical regime (cf, Chapter 2).
Heuss et al. (2003) hypothesized that the lower O3 on weekends in other locations may result
from NOX reductions in a NOx-limited regime (cf, Chapter 2).
Pun et al. (2003) described the day-of-week behavior for O3 in Chicago, Philadelphia, and
Atlanta. In Chicago and Philadelphia, maximum 1-h average O3 increases on weekends. In
Atlanta, O3 builds up from Mondays to Fridays and declines during the weekends. Fujita et al.
(2003) pointed out that since the mid-1970s, O3 levels in portions of California's South Coast
Air Basin on weekends have been as high as or higher than levels on weekdays, even though
emissions of O3 precursors are lower on weekends. Blanchard and Tanebaum (2003) noted that
despite significantly lower O3 precursor levels on weekends, 20 of 28 South Coast Air Basin
sites showed statistically significant higher mean O3 levels on Sundays than on weekdays.
Chinkin et al. (2003) noted that ambient O3 levels in California's South Coast Air Basin can be
as much as 55% higher on weekends than on weekdays under comparable meteorological
conditions.
Figures AX3-47a-h show the contrast in the patterns of hourly averaged O3 in the greater
Philadelphia, Atlanta, Houston and Los Angeles areas from weekdays to weekends from May to
September 2004. Daily maximum concentrations occur basically at the same time on either
weekdays or weekends. Mean O3 concentrations at midday are about the same on weekdays and
weekends in Atlanta, Philadelphia, and Houston, but are higher on weekends in the Los Angeles
area. Figures AX3-48a-h show the weekday/weekend differences for the 8-h averages. As can
AX3-79
-------
a. Philadelphia-Camden-Vineland, PA-NJ-DE-MD
(week day)
E
a.
a.
o
o
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
c. Atlanta-Sandy Springs-Gainesville, GA-AL
(week day)
B.
D.
O
O
b. Philadelphia-Camden-Vineland, PA-NJ-DE-MD
(week end)
~ 0,125
Q.
a
_o
S
o
O
o
8
22230001 02 03 04 05 06 07 OB 09 10 11 12131415161718192021 22230001
hour
d. Atlanta-Sandy Springs-Gainesville, GA-AL
(week end)
-~ 0.125 •
Q.
O.
O
I
S 0.075 •
-------
e. Houston-Baytown-Huntsville, TX
(week day)
f. Houston-Baytown-Huntsville, TX
(week end)
o.
Q.
O
O
Biaa
II
E
g 0,150
_o
I
£ 0,100 •
o
o
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
g. Los Angeles-Long Beach-Riverside, CA
(week day)
22230001 02 03 04 05 06 07 OB 09 10 11 12131415161718192021 22230001
hour
h. Los Angeles-Long Beach-Riverside, CA
(week end)
O
o
O 0.050 •
N
O
£
S 0.100
o
o
O 0.050
N
O
2223 00 01 02 0304 05 0607 0809 10 11 12 13 14 15 16 17 18 192021 22 2300 01
hour
2223 0001 02 03 0405 06 0708 09 10 11 12 13 14 15 16 17 18 1920 21 22 230001
hour
Figure AX3-47e-h. Diurnal variations in hourly averaged O3 on weekdays and weekends
in four cities. Values shown represent averages from
May to September of 2004. Boxes define the interquartile range and
the whiskers, the minima and maxima.
Source: Fitz-Simons et al. (2005).
week). Otherwise, the maximum O3 concentrations could be seen to occur during the week as
they do in Philadelphia and Atlanta, in contrast to Los Angeles.
Spatial Variability in Diurnal Patterns
Daily maxima in either the 1-h or 8-h averages do not necessarily occur at the same time of
day at each site in the 12 cities, and the diurnal pattern observed at individual sites can vary from
AX3-81
-------
a. Philadelphia-Camden-Vineland, PA-NJ-DE-MD
(week day)
E 0,100
a.
o
S5 0075
2
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
c. Atlanta-Sandy Springs-Gainesville, GA-AL
(week day)
£ 0.100
B.
D.
O
5 0.075
b. Philadelphia-Camden-Vineland, PA-NJ-DE-MD
(week end)
E 0.100
a.
22230001 02 03 04 05 06 07 OB 09 10 11 12131415161718192021 22230001
hour
d. Atlanta-Sandy Springs-Gainesville, GA-AL
(week end)
£ 0.100
Q.
O.
o
g 0.025
2223 00 01 02 0304 05 0607 0809 10 11 12 13 14 15 16 17 18 192021 22 2300 01
hour
2223 0001 02 03 0405 06 0708 09 10 11 12 13 14 15 16 17 18 1920 21 22 230001
hour
Figure AX3-48a-d. Diurnal variations in 8-h averaged O3 on weekdays and weekends in
four cities. Values shown represent averages from May to September
of 2004. The hour refers to the start of the 8-h averaging period.
Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
the composites shown in Figures AX3-41 and 42. Differences in diurnal patterns between sites
are related to differences in transport times from sources of precursors and chemical reactions, in
particular, titration of O3 by NO from local sources. Figure AX3-49a shows the diurnal pattern
of 1-h average O3 at a site in downtown Detroit, MI (cf, Site J in Figure AX3-30). This site is
affected by nearby traffic emissions. Figure AX3-49b shows the diurnal pattern at a site well
AX3-82
-------
e. Houston-Baytown-Huntsville, TX
(week day)
0,000
22 23 00 01 02
0.150 -i
— 0.125
O
o
g. Los Angeles-Long Beach-Riverside, CA
(week day)
f. Houston-Baytown-Huntsville, TX
(week end)
0.000 -I ,, 77 7 7 7777777777.7.
22230001 02030405060708091011 12 '3 14 15 16 1? 18 19 20 21 22230001
hour
h. Los Angeles-Long Beach-Riverside, CA
(week end)
— 0.125
O
O
T;
22230001 02 030405 03 07 08 09 10 11 12 13 14 15 1
hour
' 1S 192021 22 230001
2223 00 01 02 030405060708 09 10 11 12 13 14 15 16 17 18 19 20 21 22 2300 01
hour
Figure AX3-48e-h. Diurnal variations in 8-h averaged O3 on weekdays and weekends in
four cities. Values shown represent averages from May to September
of 2004. The hour refers to the start of the 8-h averaging period.
Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
downwind (cf., Site D in Figure AX3-30). The peak 1-h average O3 concentrations tend be
higher at the downwind site than at the site in the urban core. Figure AX3-50a shows the diurnal
patterns at a site in downtown St. Louis (cf., Site P in Figure AX3-31) and Figure AX3-50b
shows the diurnal pattern at a site downwind (Site A in Figure AX3-31). The same general
relations are observed at the two sites in the St. Louis urban area as in the Detroit urban area.
Figure AX3-51a shows the diurnal pattern observed at a site in San Bernadino, CA. This site is
AX3-83
-------
Q.
a.
c
.2 0.075 -
C
O 0.050
Detroit-Warren-Flint, Ml
Site: 261630016
T
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-49a. Diurnal variations in hourly averaged O3 at a site in downtown Detroit,
MI. Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
Detroit-Warren-Flint, Ml
Site: 260990009
C
o
o
o
I
I
±
IT
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-49b. Diurnal variations in hourly averaged O3 at a site downwind of
downtown Detroit. Boxes define the interquartile range and the
whiskers, the minima and maxima.
Source: Fitz-Simons et al. (2005).
AX3-84
-------
a.
.2 0.060 -
c
0)
u
c
O
O 0.040 -
C
o
8
St. Louis-St. Charles-Farmington, MO-IL
Site: 295100072
TTTUU
±
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-50a. Diurnal variations in hourly averaged O3 at a site in downtown
St. Louis, MO. Boxes define the interquartile range and the whiskers,
the minima and maxima.
Source: Fitz-Simons et al. (2005).
Q.
a.
c
.2 0.060
13
1)
u
c
o
O 0.040
01
c
o
8
St. Louis-St. Charles-Farmington, MO-IL
Site: 170831001
+
±
±
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-50b. Diurnal variations in hourly averaged O3 at a site downwind of
downtown St. Louis. Boxes define the interquartile range and the
whiskers, the minima and maxima.
Source: Fitz-Simons et al. (2005).
AX3-85
-------
•p 0.150-
Q.
Q.
C
o
5 0.100 H
o
C
o
O
®
c
O
0.050 -
Los Angeles-Long Beach-Riverside, CA
Site: 060719004
22 23 00 01 02 03 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-51a. Diurnal variations in hourly averaged O3 at a site in
San Bernadino, CA. Boxes define the interquartile range and the
whiskers, the minima and maxima.
Source: Fitz-Simons et al. (2005).
affected by transport of precursors from Los Angeles County, production from local precursors
and by nearby NO sources, driving O3 to very low levels at night as shown. A relatively high
peak 1-h O3 concentrations is reached at 2 to 3 p.m. Figure AX3-51b shows the diurnal pattern
of O3 at a site about 80 km to the east of the site mentioned above (cf, Site Q in Figure AX3-36)
in a relatively unpopulated area. The diurnal pattern of hourly averaged O3 is much flatter and
the 1-h peak concentration is reached about 5 or 6 p.m., on average. The cause of the rise in
concentrations at 2 a.m. is not clear.
The diurnal variation in the 8-h averages observed at the two contrasting sites in these three
areas are shown in Figures AX3-52a,b, 53a,b, and 54a,b. As might be expected the patterns are
somewhat flatter than for the 1-h averages. This implies that the difference in 8-h averages can
be substantial (i.e., over a factor of two) during early morning, afternoon, and evening.
The general pattern that emerges from the site to site variability within the urban areas
examined is that peaks in 1-h average concentrations are higher and tend to occur later at
AX3-86
-------
— 0.150-
Q.
a.
c
o
0.100-
o
c
o
u
ID
C
8
O
Los Angeles-Long Beach-Riverside, CA
Site: 060719002
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-51b. Diurnal variations in hourly averaged O3 at a site in Riverside County
well downwind of sources. Boxes define the interquartile range and the
whiskers, the minima and maxima.
Source: Fitz-Simons et al. (2005).
downwind sites than in the urban cores. To the extent that monitoring site are either near to or
remote from sources of precursors in urban/suburban areas, the behavior of O3 will follow these
basic patterns. Similar relations are found for the 8-h average O3 concentrations.
AX3.4.3 Diurnal Patterns in Nonurban Areas
Composite diurnal patterns of O3 are shown in Figure AX3-55 for hourly averaged O3 and
in Figure AX3-56 for 8 hour average O3 at rural (CASTNET) sites. As can be seen from a
comparison of Figures AX3-55 and AX3-56 with Figures AX3-42 and AX3-43, diurnal patterns
of O3 are smoother and shallower at the rural sites than at the urban sites. Maxima in hourly
averaged O3 concentrations also tend to occur in afternoon. However, highest concentrations
observed during any particular hour at night at the CASTNET sites (-0.130 ppm) are
substantially higher than observed in urban areas (<0.100 ppm) and daily 1-h maxima at
CASTNET sites have exceeded 0.150 ppm. The diurnal variations in 8-h average O3
AX3-87
-------
a
_a
c
.2 0.600
O 0.400
o
Detroit-Warren-Flint, Ml
Site: 261630016
0
iwy
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-52a. Diurnal variations in 8-h average O3 at a site in downtown Detroit, ML
Boxes define the interquartile range and the whiskers, the minima
and maxima.
Source: Fitz-Simons et al. (2005).
o
O
Detroit-Warren-Flint, Ml
Site: 260990009
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-52b. Diurnal variations in 8-h average O3 at a site downwind of downtown
Detroit, MI. Boxes define the interquartile range and the whiskers, the
minima and maxima.
Source: Fitz-Simons et al. (2005).
-------
u
c
o
O
0)
e
o
St. Louis-St. Charles-Farmington, MO-IL
Site: 295100072
T
T
T
±
T
±
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-53a. Diurnal variations in 8-h average ozone at a site in downtown
St. Louis, MO. Boxes define the interquartile range and the whiskers,
the minima and maxima.
Source: Fitz-Simons et al. (2005).
St. Louis-St. Charles-Farmington, MO-IL
Site: 170831001
— 0.600
Q.
O
o
CD
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-53b. Diurnal variations in 8-h average O3 at a site downwind of downtown
St. Louis, MO. Boxes define the interquartile range and the whiskers,
the minima and maxima.
Source: Fitz-Simons et al. (2005).
AX3-89
-------
0.
S 0.100-
c
O
o
0>
c
o
s
Los Angeles-Long Beach-Riverside, CA
Site: 060719004
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-54a. Diurnal variations in 8-h average O3 at a site in San Bernadino, CA.
Boxes define the interquartile range and the whiskers, the minima and
maxima.
Source: Fitz-Simons et al. (2005).
Los Angeles-Long Beach-Riverside, CA
Site: 060719002
Q.
S 0.100-
5 0.075
u
c
o
o
Q>
| 0.050
6
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-54b. Diurnal variations in 8-h average O3 at a site in Riverside County well
downwind of sources. Boxes define the interquartile range and the
whiskers, the minima and maxima.
Source: Fitz-Simons et al. (2005).
AX3-90
-------
Rural (CASTNET) Sites
0.200 n
E
g. 0.150
c
o
c
ffl
U
O
o
ffl
o
0.100 -
0,050 -
0.000 -
22 23 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
Figure AX3-55. Composite diurnal variability in hourly O3 concentrations observed at
CASTNET sites. Values shown are averages from April to October 2000
to 2004. Boxes define the interquartile range and the whiskers, the
minima and maxima.
Source: Fitz-Simons et al. (2005).
Rural (CASTNET) Sites
0.200 -\
£
g. 0.150
g 0.100 -
o
c
O
o
-------
concentrations are also much smaller at the CASTNET sites than at the urban sites. Note also
that the maxima in 8-h average O3 concentrations are higher at the CASTNET sites than at the
urban sites.
The diurnal variability of O3 in urban/suburban areas or in areas affected by local power
plants and highways is usually much greater than in other more isolated areas. The diurnal
variability of two sites that are characteristic of these two patterns is shown in Figure AX3-57.
The Jefferson County, KY site is characterized as suburban-residential in the AQS database and
is near Louisville, KY. High levels of O3 and NO can be found there. The Oliver County,
ND site is characterized as rural-agricultural in the AQS database. This site is fairly isolated
from combustion sources of precursors and is not near any large urban area. As can be seen
from Figure AX3-57, the diurnal variability of O3 is much smaller at the North Dakota site than
it is at the Kentucky site. The Kentucky site is influenced strongly by emissions of NO that
scavenge O3 during the night and by photochemical reactions that form O3 during the day. These
sources are lacking in the vicinity of the North Dakota site, and O3 observed there arrives mainly
from transport from distant source regions.
Logan (1989) described the diurnal variability of O3 at several rural locations, shown in
Figure AX3-58, and noted that on average, daily profiles show a broad maximum from about
noon to about 6 p.m. at all the eastern sites, except for the peak of Mt. Washington. Further
results that document the diurnal behavior of O3 in the United States during the past few decades
can be found in the previous AQCD for O3. Figure AX3-59 shows diurnal patterns for several
national forest sites in the EPA AQS database for 2002. Several of the sites analyzed exhibit
fairly flat average diurnal patterns. Such a pattern is based on average concentrations calculated
over an extended period and caution is urged in drawing conclusions concerning whether some
monitoring sites illustrated in the figure experience higher cumulative O3 exposures than other
sites. Variation in O3 concentration occurs from hour to hour on a daily basis, and, in some
cases, elevated hourly average concentrations are experienced either during daytime or nighttime
periods (Lefohn and Mohnen, 1986; Lefohn and Jones, 1986; Logan, 1989; Lefohn et al., 1990a;
Taylor et al., 1992). Because the diurnal patterns represent averaged concentrations calculated
over an extended period, the smoothing from the averaging tends to mask the elevated hourly
average concentrations.
AX3-92
-------
Jefferson Co. KY
Oliver Co. ND
1 3 5 7 9 11 13 15 17 19 21 23
Hour of Day
Figure AX3-57. The comparison of the seasonal diurnal patterns for urban-influenced
(Jefferson County, KY) and a rural-influenced (Oliver County, ND)
monitoring sites using 2002 hourly data for April-October.
Lefohn et al. (1990b) characterized O3 concentrations at high-elevation monitoring sites.
The authors reported that a fairly flat diurnal pattern for the Whiteface Mountain summit site
(WF1) was observed (Figure AX3-60a), with the maximum hourly average concentrations
occurring in the late evening or early morning hours. A similar pattern was observed for the
mid-elevation site at Whiteface Mountain (WF3). In contrast, the site at the base of Whiteface
Mountain (WF4) showed the typical diurnal pattern expected from sites that experience some
degree of O3 scavenging. More variation in the diurnal pattern for the highest Shenandoah
National Park sites occurred than for the higher elevation Whiteface Mountain sites, with the
typical variation for urban-influenced sites in the diurnal pattern at the lower elevation
Shenandoah National Park site (Figure AX3-60b). Aneja and Li (1992), in their analysis of the
five high-elevation Mountain Cloud Chemistry Program (MCCP) sites, noted the presence of the
flat diurnal pattern typical of high-elevation sites that has been described previously in the
literature. Aneja and Li (1992) noted that the peak of the diurnal patterns over the period of May
to October (1986 to 1988) occurred between 1800 and 2400 hours for the five sites, whereas
AX3-93
-------
.Q
Q.
a.
CO
o
50
40
30
20
10
60
50
40
30
60
50
40
30
20
10
0
80
70
60
50
40
30
20
10
0
_aj
— A7 _-^r
~--~^-~^'^'i^^^^ NOR
1 1 1 1
' MT
./'•*
— "••*-.,
__ ~~~--
1
WFM, NY
J;N(R) PA
I
I
I
8 12 16
Time of Day (h)
20
24
Figure AX3-58a-d. Diurnal behavior of O3 at rural sites in the United States in July.
Sites are identified by the state in which they are located.
(a) Western National Air Pollution Background Network sites
(NAPBN); (b) Whiteface Mountain (WFM) located at 1.5 km above
sea level; (c) eastern NAPBN sites; and (d) sites selected from the
Electric Power Research Institute's Sulfate Regional Air Quality
Study. IN(R) refers to Rockport.
Source: Logan (1989).
AX3-94
-------
Acadia NP
Great Smoky Mtn. NP
Glacier NP
Yellowstone NP
Grand Canyon NP
0
1 3 5 7 9 11 13 15 17 19 21 23
Time of Day
Figure AX3-59. Composite diurnal O3 pattern at selected national forest sites in the
United States using 2002 hourly average concentration data.
Source: U.S. Environmental Protection Agency (2003).
the minimum was observed between 0900 and 1200 hours. However, it is important to note that,
as indicated by Lefohn et al. (1990b), the flat diurnal pattern was not observed for all
high-elevation sites. As mentioned earlier, nonurban areas only marginally affected by nearby
sources usually have a flatter diurnal profile than sites located in urban areas. Nonurban O3
monitoring sites experience differing types of diurnal patterns, as shown in this section. The
difference in diurnal patterns may influence the potential for O3 exposures to affect vegetation.
AX3.5 SEASONAL VARIATIONS IN OZONE CONCENTRATIONS
AX3.5.1 Seasonal Variations in Urban Areas
Seasonal Variability
Figures AX3-61a-h show maximum 1-h O3 concentrations by month for selected urban
sites for 2002. As can be seen from the figure, maximum 1-h O3 concentrations tend to occur
mainly in July and August, but may also occur in other months. For example, they occurred in
AX3-95
-------
E
Q_
o>
c
o
N
O
0.06 -\
0.05
0.04 -
0.03
0.02 -
0.01 -
0.00
0.06
WF1 (1483 m)
WF3 (1026m)
WF4 (604 m)
8 10 12 14 16 18 20 22 24
Hour
SH1 (1015m)
SH3 (716 m)
SH3 (524 m)
0.00
8 10 12 14 16 18 20 22 24
Hour
Figure AX3-60a,b. Composite diurnal pattern at (a) Whiteface Mountain, NY and
(b) the Mountain Cloud Chemistry Program Shenandoah National
Park site for May to September 1987.
Source: Lefohnetal. (1990a).
AX3-96
-------
a. Phoenix, AZ
2002
0401330019
Ozone (ppm)
0.18
0.16-
0.14-
0.12-
0.10-
0.08
0.06
0.04-
0.02-
0.00-
•
ro
ZS
£Z
ro
— >
February
.c
o
ro
S
Q.
<
m m 1
III
III
III
III
>"• *
i i 3
||
|
|
1
August
September
October
November
December
b. Denver, CO
2002
080310014
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
I_
• _
II I
•
• ||
III
III
>^>.^= ^QJ >'"S
roro^Q-^^^^
zj zs ro<(^^~ en
-™ "§ <
_Q
E
OJ
U
0
tn
October
November
December
c. Atlanta, GA
2002
131210055
. Tampa, FL
2002
120571065
Q.
Q.
1"
=
S
O
0.20
0.18
0.16
0.14
0.12
e. Washington, D.C. 2002 110010043
0.08
0.06
0.04
0.02
0.00
0.20-
f. Boston, MA
2002
250250041
0.20
0.18
•P 0.16
§.0.14
a.0.12
"^0.10
c 0.08
g 0.06
O 0.04
0.02
0.00
g. Los Angeles, CA 2002 060376012
^ = >,
S q. ^
€
O
E E
0.20
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
h. Houston, TX
2002
482011034
Mill
E S
CD O
Q. O
E E
at o>
1 2
Figure AX3-61a-h.
Seasonal variations in O3 concentrations as indicated by the 1-h
maximum in each month at selected sites, 2002.
Source: U.S. Environmental Protection Agency (2003).
AX3-97
-------
June in Washington, DC and Denver, CO. The number of months for which data are shown
depends on local preference for the length of monitoring during the year. Due to a number of
factors, the absolute magnitude and the timing of the maximum hourly average concentrations
varies from year to year.
It should not be assumed that highest O3 levels are confined to the summer. Highest
average O3 concentrations generally occur at RRMS during the second quarter (i.e., during April
or May) versus the third quarter of the year as for urban sites or for nonurban sites heavily
affected by regional pollution sources.
The seasonal behavior of O3 varies across the 12 cities and high O3 values are also found at
some of the 12 cities outside of summer (e.g., Houston and Los Angeles). Figures AX3-62a-l
show the diurnal variability of hourly average O3 averaged over November through March for
EPA's 12 cities. Daily maxima tend to occur between about 1 and 2 p.m. standard time is used
across the U.S. accounting for the one hour shift from the warm season. As expected, maximum
values tend to be lower than during the warmer months. The diurnal patterns are not as clear as
in the warmer season as there is a greater tendency for highest values to occur throughout the
day and not only during early afternoon. In most northern cities, the extreme values of the daily
maximum 8-h average O3 concentration are a little more than half of those during the
warm season and the ratio of the medians are more similar as can be judged by comparison of
Figures AX3-41a-l with Figures AX3-62a-l. Differences are even smaller for the southern cities.
Indeed, some of the highest values are found in the Houston CSA outside of summer.
Figures AX3-63a-l show the diurnal variability of 8-h average O3 averaged over November
through March for EPA's 12 cities.
AX3.5.2 Seasonal Variations in Nonurban Areas
In assessing the effects of O3 on vegetation, it is important to characterize the seasons in
which the highest O3 concentrations would be expected to occur in nonurban areas. It should not
be assumed that highest O3 concentrations occur at all locations during the summer. For
example, places where highest average O3 concentrations are observed during the spring (i.e., the
months of April or May) versus the summer (Evans et al., 1983; Singh et al., 1978; Lefohn et al.,
2001) are found at many national parks in the West. Figure AX3-23 shows the hourly average
concentrations for Yellowstone National Park (WY) for the period of January to December 2001.
AX3-98
-------
a. Boston-Worcester-Manchester. MA-NH
b. New York-Newark-Bridgeport, NY-NJ-CT-PA
o
Q
mmY.
O
o
22 2300 01 02 03 3405 C6 07 OS 09 1011 12 13 14 15 16 17 18 '92021 2223 0001
hour
22 23 CO 01 02 0304050607 Oa 09 1011 12 '3 14 15 16 17 18 192021 2
hour
c. Philadelphia-Camden-Vineland. PA-NJ-DE-ME
d, Washington-Baltimore-Northern Virginia, DC-MD-VA-WV
1 02 03 Of 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01
hour
22230001 0203 04050607 0809 10 V 12 13 14 15 16 17 18 1920 2' 22230001
hour
e. Atlanta-Sandy Springs-Gainesville, GA-AL
f. Cleveland-Akron-Elyria, OH
IT
o
O
22 23 00 01 020304 05 06 07 OS 09 J0 11 12 13 14 15 16 17 18 19 2021 22 230001
hour
22230001 02 03 04 0506 D? OS 09 1011 12 '3 14 15 16 17 18 1920 2
hour
Figure AX3-62a-f. Diurnal variability in 1-h average O3 concentrations in EPA's 12 cities.
Values shown represent averages from November through March, 2000
to 2004. Boxes define the interquartile range and the whiskers, the
minima and maxima.
Source: Fitz-Simons et al., (2005)
AX3-99
-------
g. Detroit-Warren-Flint, MI
h. Chicago-Naperville-Michigan City, IL-IN-WI
0.060 •
— 0.050 •
E
a.
CL
C 0,040 •
g
1
g O.Q30 •
u
o
U 0.020 - fl
O
C
o
® 0010 •
0 000 •
nnnr^nr
T I T
T
JLJLJL
nn
^
nil ~
rj _
-Mm
i
Mri
-
TuUUUUUuL
22 23 CO 01 02 03 04 05 06 07 08 09 10 11 12 13 14151617-819202122230001
hour
222300 01 02030405 06 070809 10 11 12 13 14 15 16 17 1B 192021 2223 0001
hour
i. St. Louis-St. Charles-Farmington, MO-IL
j. Houston-Baytown-Huntsville, TX
0.
i
g- 0,150
2223 CO 01 02 03 040506 07 0809 10 11 12 13 14 15 16 17 '8 1920 21 22 230001
hour
22230001 02 030405 0607 CS 09 10 11 12 13 14 15 16 17 13 192021 222300 01
hour
k. Sacramento-Arden-Arcade-Truckee, CA-NV
I. Los Angeles-Long Beach-Riverside, CA
E c.ioo •
E 0,130 •
¥ t"T'Tl'i"T1LJU'r
ITYUU
TULJLJ
22 23 CO 01 02 03 04 05 Oe 07 06 09 10 11 12 13 14 15 16 17 -6 19 20 21 22 23 00 01
hour
22 23 00 31 02 03040506 07080910 11 12 13 14 15 16 17 18 1920 21 222300 01
hour
Figure AX3-62g-l. Diurnal variability in 1-h average O3 concentrations in EPA's 12 cities.
Values shown represent averages from November through March, 2000
to 2004. Boxes define the interquartile range and the whiskers, the
minima and maxima.
Source: Fitz-Simons et al., (2005)
AX3-100
-------
a. Boston-Worcester-Manchester, MA-NH
b. New York-Newark-Bridgeport, NY-NJ-CT-PA
0.060 •
1= 0.050 .
a
a
c
o
'•5 0.020 •
eO
C
01
§ 0 020 -
O
-------
g. Detroit-Warren-Flint, Ml
h. Chicago-Naperville-Michigan City, IL-IN-WI
*
E C 050
2223 DO 01 02 030405 36 07 OS 09 10 11 12 13 14 15 '6 17 18 '923 21 22230001
hour
22 23 0001 02 030405 060708 09 10 11 12 13 14 15 18 17 18 1920 21 22 23 0001
hour
i. St. Louis-St. Charles-Farmington, MO-IL
j. Houston-Baytown-Huntsville, TX
22 23 00 01 02 03 04 05 36 07 OB G
10 11 12 13 14 15 '6 17 18 '92321 22230001
hour
2223 0001 02 030405 060708 09 10 11 12 13 1
hour
15 16 17 18 19 20 21 22 23 00 01
k. Sacramento-Arden-Arcade-Truckee, CA-NV
I. Los Angeles-Long Beach-Riverside, CA
£ C.050
nn
w
£ 0 050 •
o
o
? 73 0001 0? 03 0405 06070809 1011 1? '3 14 15 16 1718 1920 ?1 ?? 73 00 01
hour
22 23 00 01 02 03 04 05 OS 07 08 09 10 11 121314 15 16 17 1S 19 20 21 22230001
hour
Figure AX3-63g-l. Diurnal variability in 8-h average O3 concentrations in EPA's 12 cities.
Values shown represent averages from November through March,
2000 to 2004. Boxes define the interquartile range and the whiskers,
the minima and maxima. The hour refers to the start of the 8-h
averaging period.
Source: Fitz-Simons et al. (2005).
AX3-102
-------
Note that, at the Yellowstone National Park site, the highest hourly average concentrations tend
to occur during April and May. Lefohn et al. (2001) and Monks (2000) noted that this was also
observed for other RRMS in North America and northern Europe.
During spring and summer, O3 concentrations in rural areas of the eastern United States are
affected differently by anthropogenic and natural emissions of NOX and hydrocarbons. These
differences can affect the timing of occurrence of local peak O3 concentrations (Logan, 1989).
O3 episodes occur when the weather is particularly conducive to photochemical formation of O3.
Taylor et al. (1992) reported that the temporal patterns of O3 during quarterly or annual periods
did not exhibit definitive patterns at 10 forest sites in North America. Based on the exposure
index selected, different patterns were reported. Meagher et al. (1987) reported that for
rural O3 sites in the southeastern United States, the daily maximum 1-h average concentration
was found to peak during the summer months. Taylor and Norby (1985) reported that at
Shenandoah National Park, both the highest frequency of episodes and the highest mean duration
of exposure events are found during the month of July.
Aneja and Li (1992) reported that the maximum monthly O3 levels at several rural sites
occurred in either the spring or the summer (May to August), and the minimum occurred in the
fall (September and October). The timing of the maximum monthly values differed across sites
and years. However, in 1988, an exceptionally high O3 concentration year occurred, and the
highest monthly average concentration occurred in June for almost all of the five sites
investigated. June 1988 was also the month in which the greatest number of O3 episodes
occurred in the eastern United States.
Lefohn et al. (1990a) characterized the O3 concentrations for several sites in the United
States exhibiting low maximum hourly average concentrations. Of the three western national
forest sites evaluated by Lefohn et al. (1990a), Apache National Forset (AZ), Custer National
Forest (MT), and Ochoco National Forest (OR), only at Apache National Forest (AZ) did
maximum monthly mean concentrations occur in the spring. The Apache National Forest site
was above mean nocturnal inversion height, and no decrease of concentrations occurred during
the evening hours. Highest hourly maximum concentration, as well as the highest
W126 O3 exposures were also found at this site. Most of the maximum monthly mean
concentrations occurred in the summer at the other two sites. Maximum monthly mean O3
AX3-103
-------
concentrations were found at the White River Oil Shale site in Colorado during the spring and
summer months.
The W126 sigmoidal weighting exposure index was also used to identify the month of
highest exposure of vegetation to O3. A somewhat more variable pattern was observed than
when the maximum monthly average concentration was used. In some cases, the highest W126
exposures occurred earlier in the year than was indicated by the maximum monthly
concentration. For example, in 1979, the Custer National Forest site experienced its highest
W126 exposure in April, although the maximum monthly mean occurred in August. In 1980, the
reverse occurred.
There was no consistent pattern for those sites located in the continental United States.
Maximum exposures to O3 during the spring and summer at the Theodore Roosevelt NP,
Ochoco, and Custer National Forest sites and the White River Oil Shale site. Likewise, the sites
at which highest exposures to O3 occurred during the period from fall to spring did not always
have the lowest exposures to O3.
AX3.6 TRENDS IN OZONE CONCENTRATIONS
Evidence for Trends in Ozone Concentrations at Rural Sites in the United States
Year-to-year variability in the nationwide May to September, mean daily maximum 8-h O3
concentrations are shown in Figure AX3-64. Data flagged because of quality control issues was
removed with concurrence of the local monitoring agency. Only days for which there was
75% data capture (i.e., 18 of 24 hours) were kept, and a minimum of 115 of 153 days (i.e.,
75% data capture) were required in each year. Data for missing years at individual sites were
filled in using simple linear interpolation, as done in EPA Trends reports. Year-to-year
variability in the corresponding 95th percentile values of the daily maximum 8-h O3
concentrations are shown in Figure AX3-65. Sites considered in this analysis are shown in the
map in Figure AX3-3. Mean O3 concentrations were slightly lower in 2003 and 2004 than in
earlier years, and as was shown in Figures AX3-1 and AX3-2, most sites are located in the East.
The summer of 2003 was slightly cooler than normal in the East (Levinson and Waple, 2004)
and the summer of 2004 was much cooler than normal in the East (Levinson, 2005) accounting
in part for the dip in O3 during these 2 years. Trends in compliance metrics such as the fourth
AX3-104
-------
Nationwide Trends, May to September
Mean of Daily Maximum 8-Hour Values, 1990 - 2004
0.12-
0.11 -
E
_£ 0.09 -_
O 0.08 -
£ 0.07 -
c
8 0.06 -
c
O 0.05 -.
0)
= 0.04 -
w
N
O 0.03 -.
0.02-
0.01 -
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Year
Figure AX3-64. Year-to-year variability in nationwide mean daily maximum 8-h O3
concentrations. The whiskers on the box plot represent the 10th and
90th percentile concentrations. The "X"s above and below the whiskers
are the values that fall below and above the 10th and 90th percentile
concentrations. The dots inside the box represent the mean, for the
statistic, at all sites.
Source: Fitz-Simons et al. (2005).
highest daily maximum 8-h average O3 concentration can be found in the EPA Trends reports
and so are not repeated here.
Figures AX3-66a-h show year-to-year variability in mean daily 8-h O3 concentrations
observed at selected national park sites across the United States. Figures AX3-67a-h show year-
to-year variability in the 95th percentile value of daily maximum 8-h O3 concentrations at the
same sites shown in Figures AX3-66a-h. The same criteria used for calculating values in
Figures AX3-64 and AX3-65 were used for calculating the May to September seasonal averages
for the national parks shown in Figures AX3-66a-h and 67a-h. Trends at these national parks are
shown in Table AX3-9. However, several monitoring sites were moved during the period from
1990 to 2004. Sites were moved at Acadia NP in 1996, Joshua Tree NP in 1993, Mammoth
AX3-105
-------
Nationwide Trends, May to September
95th Percentile of Daily Maximum 8-Hour Values, 1990 - 2004
E
Q.
Q.
C
o
C
0)
o
o
o
0)
o
N
o
0.18-
0.17-
0.16-
0.15-
0.14-
0.13-
QA2-
0.11 -
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02-
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Year
Figure AX3-65. Year-to-year variability in nationwide 95th percentile value of the daily
maximum 8-h O3 concentrations. The whiskers on the box plot represent
the 10th and 90th percentile values for the statistic. The "X"s above and
below the whiskers are the values that fall below and above the 10th and
90th percentile values. The dots inside the box represent the mean, for the
statistic, at all sites.
Source: Fitz-Simons et al. (2005).
Cave NP in 1996, Voyageurs NP in 1996, and Yellowstone NP in 1996 and offsets in O3
concentrations have resulted. As a result, trends are not shown for these sites.
As noted in The Ozone Report—Measuring Progress through 2003 (U.S. Environmental
Protection Agency, 2004b), O3 trends in national parks in the South and the East are similar to
nearby urban areas and reflect the regional nature of O3 pollution. For example, O3 trends in
Charleston, SC and Charlotte, NC track those in nearby Cowpens NP and Cape Romaine NP in
South Carolina; O3 in Knoxville and Nashville, TN tracks O3 in Great Smoky NP; O3 in
Philadelphia, PA and Baltimore, MD tracks Brigantine NP in New Jersey; and New York, NY
and Hartford, CT track O3 in Cape Cod NS. The situation is not as clear in the West, where
national parks are affected differently by pollution sources that are located at varying distances
away (e.g., Lassen Volcanic National Park and Yosemite National Park, CA). However, data
AX3-106
-------
May to September Mean of Daily Maximum 8-Hour Values, 1990 - 2004
0.09 i
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02-
a. AcadiaNP
0.01 V-
1990
1996 1998
Year
1996 1998
Year
d. Shenandoah NP
0.09 :
0.08
0.07 -
0.06 -
0.05 -
0.04 -
0.03 -
002 -
0.01 :
e. Great Smoky Mountains NP
0.09 :
0.08 -
0.07 -
0.06 -
0.05 -
0.04 -
0.03 -
0.02 -
0.01 -
g. Cape Roma in NWR
Year
1996 1998
Year
0.09
0.08 •
£ 0.07 :
Q.
^ 0.06 -
_o
£ 0.05 -
| 0.04 -
5 0.03 :
0.02 :
0.01 :
h. Mammoth Cave NP
1996 1998
Year
Figure AX3-66a-h. Year-to-year variability in mean daily maximum 8-h O3 concentrations
at selected national park (NP), national wildlife refuge (NWR), and
national monument (NM) sites.
Source: Fitz-Simons et al. (2005).
AX3-107
-------
May to September Mean of Daily Maximum 8-Hour Values, 1990 - 2004
0.08-
e 0.07 -
c
a.
°: 0.06 -
c
o
1 0.05 -
I 0.04 -
E
O
0 0.03
0.02-
0 01
i. Voyageurs NP
^__, . . __•—-»-"'*'•---•
. — S
0.08 •
c 0.07 •
c
a.
°: 006 •
c
o
1 0.05 •
§ 0.04 •
C
O
0 0.03
0.02 •
n i"H
j. Theodore Roosevelt NP
*— *—*~_^_-^-_^-*-~V
t^^_^^^.* *-"* ~~*~~* • tr^^ ^^
1980 1992 1994 1996 1998 2000 2002 2004 1990 1992 1994 1996 1998 2000 2002 20&
Year Year
0 00 n nn
0.08 -
E 0.07 •
a.
^ 0.06 -
_o
1 0.05 •
| 0.04 -
o
0 0.03 -
0.02 •
0 01 •
k. Rocky Mountain NP
. //~*\.
^ *~ ^s>Vv«xX^\ /
•— *— \y
•/
0.08 •
E 0.07 •
a.
^ 0.06 •
_o
'| 0.05 •
$ 0.04 •
D
0 0.03 •
0.02 -
n n-i
I. Yellowstone NP
* — • — *-^_--*^* •—+—* V^
. . — r^
IT
1990 1992 1994 1996 1998 2000 2002 2004 1990 1992 1994 1996 1998 2000 2002 200
Year Year
0 09 A nn
0.08
£ 0.07 -
Q.
Q.
£ 0.06 -
o
ys
2 0.05 -
E
| 0.04 -
o
o
0.03 -
0.02 -
0 01 -
m. Glacier NP
_
, , ^v J>L A~~«__ ^-«XX^\,
^s^/ 1" \ /
•V \J
0.08
£ 0.07 •
a.
a.
£ 0.06 •
o
| 0.05 •
c
| 0.04 •
o
o
0.03
0.02 •
n n-i
n. Grand Canyon NP
x*~ — *^^
1990 1992 1994 1996 1998 2000 2002 2004 1990 1992 1994 1996 1998 2000 2002 200
Year Year
0 00 A nn
0.08 •
c 0.07 -
c
a
^ 0.06 -
_o
g 0.05 -
1 0.04 -
c
o
0 0.03
0.02
n n-f
o. Sag u arc NM
^^^_^*—~m /*\
^~~+~~*^
0.08 •
- 0.07 •
c
a
^ 0.06 -
_o
jj 0.05 •
1 0.04 -
c
o
0 0.03 •
0.02 •
n m
p. Chirlcahua NM
• — • — •— -^ ~~-+~~^^+-~^^~i^
1992 1994 1996 1998 2000 2002 2004
Year
1990 1992 1994 1996 1998 2000 2002 2004
Year
Figure AX3-66i-p. Year-to-year variability in mean daily maximum 8-h O3 concentrations
at selected national park (NP), national wildlife refuge (NWR), and
national monument (NM) sites.
Source: Fitz-Simons et al. (2005).
AX3-108
-------
May to September Mean of Daily Maximum 8-Hour Values, 1990 - 2004
0.09
0.08 :
0.07
0.06:
0.05:
0.04:
0.03:
0.02 :
. Sequoia/Kings Canyon NP
0.01 V-
1990
1992
1994
1996 1998
Year
2000 2002
1996 1998
Year
0.09 :
0.08 -
0.07 :
0.06 :
0.05 :
0.04 :
0.03 :
0.02 :
0.01
u. Olympic NP
1990
1992
1994
1998
2000 2002
0.09
2004
2004
0.08 •
0.07 •
0.06 :
0.05 :
0.04 -
0.03 •
0.02 :
0.01
r. Pinnacles NM
1990
1992
1994
1996 1998
Year
2000 2002
2004
0.09
0.08 •
0.07 :
0.06 :
0.05 :
0.04 •
0.0:
0.02 :
0.01
t. Denali NP
1996 1998
Year
0.09
0.08 •
0.07 •
0.06 :
0.05
0.04 -
0.03 •
0.02 :
0.01
v. Lassen Volcanic NP
Year
1996 1998
Year
Figure AX3-66q-v. Year-to-year variability in mean daily maximum 8-h O3 concentrations
at selected national park (NP), national wildlife refuge (NWR), and
national monument (NM) sites.
Source: Fitz-Simons et al. (2005).
obtained at these sites still provide valuable information about the variability in regional
background concentrations, especially since the West has not been broken down into regions as
has been done by Lehman et al. (2004) and shown in Figure AX3-7.
AX3-109
-------
May to September 95th Percentile of Daily Maximum 8-Hour Values, 1990 - 2004
0.11 -
0.10:
0.09:
0.08:
0.07:
0.06:
0.05:
0.04:
0.03-
0.02^
a. Acadia NP
2000 2002 2004
0.11
0.10
0.09:
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02
b. Cape Cod NS
1990 1992 1994
1996
2000 2002 2004
Year
0.11
2004
2004
1990 1992 1994 1996 1998
Year
2002 2004
Figure AX3-67a-h.
Year-to-year variability in 95th percentile of daily maximum 8-h O3
concentrations at selected national park (NP), national wildlife refuge
(NWR), and national monument (NM) sites.
Source: Fitz-Simons et al. (2005).
AX3-110
-------
May to September 95th Percentile of Daily Maximum 8-Hour Values, 1990 - 2004
0.11 -
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02 JT
i. Voyageurs NP
0.10 :
0.09:
0.08:
0.07:
0.06
0.05:
0.04:
0.03-
0.02
j. Theodore Roosevelt NP
1990 1992 18
1998 2000 2002 2004
1990 1992 1994
1996 1998
Year
2000 2002 2004
0.11
0.10;
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02
k. Rocky Mountain NP
0.11
0.10:
0.09:
0.08:
0.07:
0.06:
0.05
0.04 :
0.03:
0.02
I. Yellowstone NP
1990 1992 1994
1996 1998
Year
2000 2002 2004
1990 1992 1994
1996 1998
Year
2000 2002 2004
0.10-
0.09-
0.08-
0.07-
0.06-
0.05
0.04:
0.03-
0.02
m. Glacier NP
0.10:
0.09:
0.08-
0.07
0.06:
0.05-
0.04:
0.03-
0.02
n. Grand Canyon NP
1990 1992 1994
1996 1998
Year
2002 2004
1990 1992 1994
1996 1998
Year
2000 2002 2004
£
£ 0.08:
0.04-
0.03-
o. Saguaro NM
0.11
0.10:
0.09:
0.08:
0.07:
0.06:
0.05:
0.04:
0.03 :
0.02
p. Chiricahua NM
1990 1992 1994
1996 1998 2000
Year
2002 2004
1990 1992 1994
1996 1998
Year
2000 2002 2004
Figure AX3-67i-p. Year-to-year variability in 95th percentile of daily maximum 8-h O3
concentrations at selected national park (NP), national wildlife refuge
(NWR), and national monument (NM) sites.
Source: Fitz-Simons et al. (2005).
AX3-111
-------
May to September 95th Percent!le of Daily Maximum 8-Hour Values, 1990 - 2004
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
0.02
q. Sequoia/Kings Canyon NP
0.11
1996 1998
Year
1990
1992
1994
1996
2000 2002
0.11
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04-
0.03-
u. Olympic NP
0.02V—
1990
1996 1998
Year
0.10-
0.09-
0.08
0.07-
0.06-
0.05-
0.04-
0.03-
0.02
0.11
2004
r. Pinnacles NM
1996 1998
Year
0.10-
0.09-
0.08-
0.07-
0.06-
0.05-
0.04
0.03-
0.02
t. Denali NP
1990
1992
1994
1996
2000 2002
2004
0.11
0.10-
0.09-
0.08-
0.07
0.06-
0.05-
0.04-
0.03-
0.02
v. Lassen Volcanic NP
1996 1998
Year
Figure AX3-67q-v. Year-to-year variability in 95th percentile of daily maximum 8-h O3
concentrations at selected national park (NP), national wildlife refuge
(NWR), and national monument (NM) sites.
Source: Fitz-Simons et al. (2005).
Caution should be exercised in using trends calculated at national parks to infer
contributions from distant sources either inside or outside of North America, because of the
influence of local and regional pollution. For example, using a 15-year record of O3 from Lassen
AX3-112
-------
Table AX3-9. Trends in Warm Season (May to September) Daily Maximum 8-h O3 Concentrations at National Parks
in the United States (1990 to 2004). Trends are given as ppb yr'1.
X
Mean
Site
AcadiaNP(ME)
Brigantine NWR (NJ)
Cape Cod NS (MA)
Cape Remain NWR (SC)
ChiricahuaNM(AZ)
Cowpens NB (SC)
Denali NP (AK)
Glacier NP (MT)
Grand Cany on NP (AZ)
Great Smoky Mountains NP (NC-TN)
Joshua Tree NP (CA)3
Lassen Volcanic NP (CA)
Mammoth Cave NP (KY)3
Olympic NP (WA)
Pinnacles NM (CA)
Rocky Mountain NP (CO)
Saguaro NM (AZ)
Sequoia/Kings Cany on NP (CA)
ShenandoahNP(VA)
Theodore Roosevelt NP (ND)
Voyageurs NP (MN)3
Yellowstone NP (WY)3
trend
0.0401
-0.802
0
0.71
0.221
0
0.17
0
0.25
0.29
—
0.25
—
0.14
-0.1
0.911
-0.2
0.38
0
0.381
—
—
p-value
0.037
0.014
0.423
0.046
0.046
0.423
0.12
0.5
0.07
0.248
—
0.141
—
0.141
0.218
0.004
0.279
0.218
0.461
0.023
—
—
P95
trend
l.O1
-1.72
0
l.O1
0.2
0.1
0.61
0.1
0
0.9
—
0.2
—
0.31
-0.5
l.O1
-0.3
0
-0.2
0.2
—
—
p-value
0.037
0.004
0.349
0.01
0.084
0.349
0.01
0.349
0.5
0.19
—
0.19
—
0.037
0.07
0.014
0.19
0.461
0.385
0.19
—
—
P98
trend
-0.2
-1.92
-0.5
1.0
0.15
0.4
0.61
0.27
0.13
0.4
—
0
—
0.2
-0.56
0.88
-0.38
0
0.33
0.2
—
—
p-value
0.349
0.003
0.19
0.07
0.218
0.349
0.002
0.19
0.218
0.423
—
0.5
—
0.19
0.057
0.07
0.141
0.539
0.279
0.141
—
—
1 Upward trend, significant at p = 0.05 level.
2Downward trend, significant at p = 0.05 level.
3 Site moved. See text for details.
-------
Volcanic National Park, a rural elevated site in northern California; data from two aircraft
campaigns; and observations spanning 18 years from five U.S. West Coast, marine boundary
layer sites, Jaffe et al. (2003) reported that O3 in air arriving from the eastern Pacific in spring
has increased by approximately 10 ppb from the mid-1980s. They concluded that this positive
trend is due to increases of emissions of O3 precursors in Asia. They found positive trends in O3
in all seasons. They also noted that diurnal variations during summer were about 21 ppb, but
only about 6 ppb during spring. Although Lassen Volcanic National Park site is not close to any
major emission sources or urban centers, the site experiences maximum hourly average O3
concentrations above 0.080 ppm during April to May and above 0.100 ppm during the summer
(U.S. Environmental Protection Agency, 2003), suggesting local photochemical production, at
least during summer. However, local springtime photochemical production cannot be ruled out.
The authors suggested that the likely cause for the spring increases is transport from Asia,
because emissions of precursors have decreased in California over the monitoring period. The
springtime increases appears to be inconsistent with the summer increases, when there is
evidence for the occurrence of more localized photochemical activity. Although emissions of O3
precursors may have decreased in California as a whole over the monitoring period, there still
may be regional increases in areas that could affect air quality in Lassen.
AX3.7 RELATIONS BETWEEN OZONE, OTHER OXIDANTS, AND
OXIDATION PRODUCTS
Tables of measurements of PAN and peroxypropionyl nitrate (PPN, CH3CH2C(O)OONO2)
concentrations were given in the 1996 O3 AQCD (U.S. Environmental Protection Agency,
1996a). Measurements were summarized for rural and urban areas in the United States, Canada,
France, Greece, and Brazil. The use of measurements from aboard serve to illustrate or support
certain U.S. results as well as to demonstrate the widespread presence of PANs in the
atmosphere. Additional data for H2O2 were also presented in the 1996 O3 AQCD. Data for these
species are obtained as part of specialized field studies and not as part of routine monitoring
operations and thus are highly limited in their ranges of applicability. As a result, it is difficult
to relate the concentrations of O3, other oxidants, and oxidation products on the basis of rather
sparse data sets. This information is simply not available for a large number of environments.
AX3-114
-------
Instead, it might be more instructive to examine the relations between O3 and other products of
atmospheric reactions on the basis of current understanding of atmospheric photochemical
processes.
In order to understand co-occurrence between atmospheric species, an important
distinction must be made between primary (directly emitted) species and secondary
(photochemically produced) species. In general, it is more likely that primary species will be
more highly correlated with each other, and that secondary species will be more highly
correlated with each other than will species from mixed classes. By contrast, primary and
secondary species are less likely to be correlated with each other. Secondary reaction products
tend to correlate with each other, but there is considerable variation. Some species (e.g., O3 and
organic nitrates) are closely related photochemically and correlate with each other strongly.
Others (e.g., O3 and H2O2) show a more complex correlation pattern.
Although NO2 is produced mainly by the reaction of directly emitted NO with O3 with
some contributions from direct emissions, in practice, it behaves like a primary species. The
timescale for conversion of NO to NO2 is fast (5 min or less), so NO and NO2 ambient
concentrations rapidly approach values determined by the photochemical steady state. The sum
NO + NO2 (NOX) behaves like a typical primary species, while NO and NO2 reflect some
additional complexity based on photochemical interconversion. As a primary species, NO2
generally does not correlate with O3 in urban environments. In addition, chemical interactions
among O3, NO and NO2 have the effect of converting O3 to NO2 and vice versa, which can result
in a significant anti-correlation between O3 and NO2.
Organic nitrates consist of PAN, a number of higher-order species with photochemistry
similar to PAN (e.g., PPN), and species such as alkyl nitrates with somewhat different
photochemistry. These species are produced by a photochemical process very similar to that
of O3. Photochemical production is initiated by the reaction of primary and secondary VOCs
with OH radicals, the resulting organic radicals subsequently react with NO2 (producing PAN
and analogous species) or with NO (producing alkyl nitrates). The same sequence (with organic
radicals reacting with NO) leads to the formation of O3.
In addition, at warm temperatures, the concentration of PAN forms a photochemical steady
state with its radical precursors on a timescale of roughly 30 minutes. This steady state value
increases with the ambient concentration of O3 (Sillman et al., 1990). Ozone and PAN may
AX3-115
-------
show different seasonal cycles, because they are affected differently by temperature. Ambient
O3 increases with temperature, driven in part by the photochemistry of PAN (see description
above). By contrast, the photochemical lifetime of PAN decreases rapidly with increasing
temperature. The ratio, O3/PAN, should show seasonal changes, with highest ratios in summer,
although there is no evidence from measurements. Measured ambient concentrations
(Figures AX3-68a-d) show a strong nonlinear association between O3 and PAN, and between O3
and other organic nitrates (Pippin et al., 2001; Roberts et al., 1998). Moreover, the uncertainty
in the relationship between O3 and PAN grows as the level of PAN increases.
Individual primary VOCs are generally highly correlated with each other and with NOX
(Figure AX3-69). A summary of the results of a number of field studies of the concentrations of
precursors including NOX and nonmethane organic compounds (NMOCs) are summarized in the
1996O3AQCD.
Formation of H2O2 takes place by self-reaction of photochemically generated HO2 radicals,
so that there is large seasonal variation of H2O2 concentrations and values in excess of 1 ppb are
mainly limited to the summer months (Kleinman, 1991). Although H2O2 is produced from
photochemistry that is closely related to O3, it does not show a consistent pattern of correlation
with O3. Hydrogen peroxide is produced in abundance along with O3 only when O3 is produced
under NOx-limited conditions. When the photochemistry is NOx-saturated much less H2O2 is
produced. In addition, increasing NOX tends to slow the formation of H2O2 under NOx-limited
conditions.
Measurements of gas phase peroxides in the atmosphere were reviewed by Lee et al.
(2000). Ground level measurements of H2O2 taken during the 1970s indicated values of 180 ppb
in Riverside, CA and 10 to 20 ppb during smog episodes in Claremont and Riverside, with
values approaching 100 ppb in forest fire plumes. However, later surface measurements always
found much lower values. For example, in measurements made in Los Angeles and nearby areas
in the 1980s, peak values were always less than about 2 ppb and in a methods intercomparison
study in Research Triangle Park, NC in June 1986, concentrations were <2.5 ppbv. Higher
values ranging up to 5 ppb were found in a few other studies in Kinterbish, Alabama and
Meadview, Arizona. Several of these studies found strong diurnal variations (typically about a
factor three) with maximum values in the mid-afternoon and minimum values in the early
AX3-116
-------
140
0 1000 2000 3000 4000 0 1000 2000 3000 4000 5000
PAN (pptv)
Figure AX3-68a-d. Measured O3 (ppbv) versus PAN (pptv) in Tennessee, including
(a) aircraft measurements, and (b, c, and d) suburban sites near
Nashville.
Source: Roberts et al. (1998).
morning. Mean concentrations of organic hydroperoxides at the surface at Niwot Ridge, CO in
the summer of 1988 and State Park, GA during the summer of 1991 were all less than a few ppb.
Early aircraft measurements of H2O2 over the eastern United States were reported by
Heikes et al. (1987). More recent aircraft measurements of hydroperoxide (H2O2, CH3OOH and
HOCH2OOH) concentrations were made as part of the Southern Oxidants Study intensive
campaign in Nashville, TN in July 1995 (Weinstein-Lloyd et al., 1998). The median
AX3-117
-------
1.5-
• Sp2< 10 ppbvi
o"sb"2>"10 "ppbvT
0,0
20
40 60 80
NOv (ppbv)
100
120
Figure AX3-69. Relationship between benzene and NOy at a measurement site in Boulder,
CO. Instances with SO2 >10 ppb are identified separately (open circles),
because these may reflect different emission sources.
Source: Goldan et al. (1995).
concentration of total hydroperoxides in the boundary layer between 1100 and 1400 CDT was
about 5 ppbv, with more than 50% contribution from organic hydroperoxides. Median O3 was
about 70 ppbv at the same time. The concentrations of the hydroperoxides depended strongly on
wind direction. For example, values were about 40% lower when winds originated from the
N/NW as opposed to the S/SW.
Elevated O3 is generally accompanied by elevated HNO3, although the correlation is not as
strong as between O3 and organic nitrates. Ozone often correlates with HNO3, because they have
the same precursor (NOX). However, HNO3 can be produced in significant quantities in winter,
even when O3 is low. The ratio between O3 and HNO3 also shows great variation in air pollution
events, with NOx-saturated environments having much lower ratios of O3 to HNO3 (Ryerson
et al., 2001). Aerosol nitrate is formed primarily by the combination of nitrate (supplied
by HNO3) with ammonia, and may be limited by the availability of either nitrate or ammonia.
AX3-118
-------
Nitrate is expected to correlate loosely with O3 (see above), whereas ammonia is not expected to
correlate with O3.
In addition to nitrate, other oxidants are present in airborne cloud droplets, rain drops and
particulate matter. Measurements of hydroperoxides, summarized by Reeves and Penkett
(2003), are available mainly for hydrometeors, but are sparse for ambient particles.
Venkatachari et al. (2005a) sampled the concentrations of total reactive oxygen species (ROS) in
particles using a cascade impactor in Rubidoux, CA during July 2003. Although the species
constituting ROS were not identified, the results were reported in terms of equivalent H2O2
concentrations. Unlike O3 and gas phase H2O2 which show strong diurnal variability (i.e., about
a factor of three variation between afternoon maximum and early morning minimum), the
diurnal variation of particle phase ROS was found to be much weaker (i.e., less than about 20%)
at least for the time between 8 a.m. and midnight. Because the ROS were measured in the fine
aerosol size fraction, which has a lifetime with respect to deposition of much greater than a day,
little loss is expected but their concentrations might also be expected to increase because of
nighttime chemistry, perhaps involving NO3 radicals. The concentration of ROS, expressed as
equivalent H2O2 (5.2 to 6.1 x 10'7 M/m3, ranged from 20% to 100% that of O3 (diurnal average:
30%), with highest values at night. The ratio was likely higher at the early morning minimum
for O3. In a companion study conducted in Queens, NY during January and early February 2004,
Venkatachari et al. (2005b) found much lower concentrations of ROS of about 1 x 10"7 M/m3.
However, O3 levels were also substantially lower leading to ROS concentrations about 20%
those of O3. It is of interest to note that gas phase OH concentrations measured at the same time
ranged from about 7.5 x lOVcm3 to about 1.8 x 106/cm3, implying the presence of significant
photochemical activity even in New York City during winter.
Peroxyacetylnitrate (PAN) is produced during the photochemical oxidation of a wide range
of VOCs in the presence of NOX. It is removed by thermal decomposition and also by uptake to
vegetation (Sparks et al., 2003; Teklemariam and Sparks, 2004). PAN is the dominant member
of the broader family of peroxyacylnitrates (PANs), which includes as other significant
atmospheric components peroxypropionyl nitrate (PPN) of anthropogenic origin and
peroxymethacrylic nitrate (MPAN) produced from oxidation of isoprene. Measurements and
models show that PAN in the United States includes major contributions from both
anthropogenic and biogenic VOC precursors (Horowitz et al., 1998; Roberts et al., 1998).
AX3-119
-------
Measurements in Nashville during the 1999 summertime Southern Oxidants Study (SOS)
showed PPN and MPAN amounting to 14% and 25% of PANs, respectively (Roberts et al.,
2002). Measurements during the TexAQS 2000 study in Houston indicated PAN concentrations
of up to 6.5 ppbv (Roberts et al., 2003). PAN measurements in southern California during the
SCOS97-NARSTO study indicated peak concentrations of 5-10 ppbv, which can be contrasted to
values of 60-70 ppbv measured back in 1960 (Grosjean, 2003). Vertical profiles measured from
aircraft over the United States and off the Pacific coasts show PAN concentrations above the
boundary layer of only a few hundred pptv, although there are significant enhancements
associated with long-range transport of pollution plumes from Asia (Kotchenruther et al., 2001a;
Roberts et al., 2004). Decomposition of this anthropogenic PAN as it subsides over North
America can lead to significant O3 production, enhancing the O3 background (Kotchenruther
et al., 2001b; Hudman et al., 2004).
Relations between primary and secondary components discussed above are illustrated by
considering data for O3 and PM2 5. Ozone and PM2 5 concentrations observed at a monitoring site
in Fort Meade, MD are plotted as binned means in Figure AX3-70. These data were collected
between July 1999 and July 2001. As can be seen from the figure, PM25, regarded as a function
of O3, increases to the left of the inflection point (at about 30 ppbv O3) and also increases
with O3 to the right of the inflection point. Data to the left of the minimum in PM2 5 were
collected mainly during the cooler months of the year, while data to the right of the minimum
were collected during the warmer months. This situation arises because PM2 5 contains a large
secondary component during the summer and has a larger primary component during winter.
During the winter, O3 comes mainly from the free troposphere, above the planetary boundary
layer and, thus, may be considered a tracer for relatively clean air. Unfortunately, data for PM2 5
and O3 are collected concurrently at relatively few sites in the United States throughout an entire
year, so these results, while highly instructive are not readily extrapolated to areas where
appreciable photochemical activity occurs throughout the year. Ito et al. (2005) showed the
relation between PM10 and O3 on a seasonal basis in several urban areas (cf, Figure 7-24).
Although PM10 contains proportionately more primary material than does PM2 5, relations similar
to those shown in Figure AX3-70 are found, reflecting the dominant contribution from PM25
to PM10.
AX3-120
-------
40
35"
*E 30"
| 25-
i1
°- 20 f
c
(0
S 15 t
T3
0)
C
.E 10 f
m
I.Wit
10 20 30 40 50
Binned Mean 03 (ppbv)
60
70
Figure AX3-70. Binned mean PM2 5 concentrations versus binned mean O3 concentrations
observed at Fort Meade, MD from July 1999 to July 2001.
Source: Chen (2002).
AX3.8 RELATIONSHIP BETWEEN SURFACE OZONE AND
OTHER POLLUTANTS
AX3.8.1 Introduction
Several attempts have been made to characterize gaseous air pollutant mixtures (Lefohn
and Tingey, 1984; Lefohn et al., 1987). The characterization of co-occurrence patterns under
ambient conditions is important for relating human health and vegetation effects to controlled
chamber studies and to ambient conditions. Lefohn et al. (1987) discussed the various patterns
of pollutant exposures. Pollutant combinations can occur at or above a threshold concentration
either together or temporally separated from one another. Patterns that show air pollutant pairs
appearing at the same hour of the day at concentrations equal to or greater than a minimum
AX3-121
-------
hourly mean value were defined as simultaneous-only daily co-occurrences. When pollutant
pairs occurred at or above a minimum concentration during the 24-h period, without occurring
during the same hour, a "sequential-only" co-occurrence was defined. During a 24-h period, if
the pollutant pair occurred at or above the minimum level at the same hour of the day and at
different hours during the period, the co-occurrence pattern was defined as "complex-
sequential."
For characterizing the different types of co-occurrence patterns for O3/NO2, O3/SO2,
and NO2/SO2, Lefohn and Tingey (1984) used a 0.05 ppm threshold to identify the number of
hourly simultaneous-only co-occurrences for the period May through September at a large
number of air quality urban monitoring sites along with rural sites. The selection of a 0.05-ppm
threshold concentration was based on vegetation effects considerations. Data used in the
analysis included hourly averaged (1) Environmental Protection Agency Storage and Retrieval
of Aerometric Data (SAROAD; now AQS) data for 1981, (2) EPRI-SURE and Eastern Regional
Air Quality Study (ERAQS) data for 1978 and 1979, and (3) Tennessee Valley Authority (TVA)
data from 1979 to 1982. Lefohn and Tingey (1984) concluded, for the pollutant combinations,
that (1) the co-occurrence of two-pollutant mixtures lasted only a few hours per episode and (2)
the time interval between episodes was generally large (weeks, sometimes months).
Lefohn et al. (1987), using a 0.03-ppm threshold, grouped air quality data from rural and
RRMS (as characterized in the EPA database) within a 24-h period starting at 0000 hours and
ending at 2359 hours. Data were analyzed for the May to September period. Data used in the
analysis included hourly averaged (1) Environmental Protection Agency AQS (SAROAD) data
from 1978 to 1982, (2) EPRI-SURE and -ERAQS data for 1978 and 1979, and (3) TVA data
from 1979 to 1982. Patterns that showed air pollutant pairs appearing at the same hour of the
day at concentrations equal to or greater than a minimum hourly mean value were defined as
simultaneous-only daily co-occurrences. When pollutant pairs occurred at or above a minimum
concentration during the 24-h period, without occurring during the same hour, a "sequential-
only" co-occurrence was defined. During a 24-h period, if the pollutant pair occurred at or
above the minimum level at the same hour of the day and at different hours during the period,
the co-occurrence pattern was defined as "complex-sequential." A co-occurrence was not
indicated if one pollutant exceeded the minimum concentration just before midnight and the
other pollutant exceeded the minimum concentration just after midnight. As will be discussed
AX3-122
-------
below, studies of the joint occurrence of gaseous NO2/O3 and SO2/O3 reached two conclusions:
(1) hourly simultaneous and daily simultaneous-only co-occurrences are fairly rare and (2) when
co-occurrences are present, complex-sequential and sequential-only co-occurrence patterns
predominate. The authors reported that year-to-year variability was found to be insignificant;
most of the monitoring sites experienced co-occurrences of any type less than 12% of the
153 days.
Since 1999, monitoring stations across the United States have been routinely measuring the
24-h average concentrations of PM25. Because of the availability of the PM25 data, daily
co-occurrence of PM25 and O3 over a 24-h period was characterized. Because PM25 data are
mostly summarized as 24-h average concentrations in the AQS data base, a daily co-occurrence
of O3 and PM25 was subjectively defined as when an hourly average O3 concentration >0.05 ppm
and a PM2 5 24-h concentration >40 |ig/m3 occurred over the same 24-h period.
For exploring the co-occurrence of O3 and other pollutants (e.g., acid precipitation and
acidic cloudwater), limited data are available. In most cases, routine monitoring data are not
available from which to draw general conclusions. However, published results are reviewed and
summarized for the purpose of assessing an estimate of the possible importance of co-occurrence
patterns of exposure.
AX3.8.2 Co-Occurrence of Ozone with Nitrogen Oxides
Ozone occurs frequently at concentrations equal to or greater than 0.05 ppm at many rural
and remote monitoring sites in the United States (U.S. Environmental Protection Agency,
1996a). Therefore, for many rural locations in the United States, the co-occurrence patterns
observed by Lefohn and Tingey (1984) for O3 and NO2 were defined by the presence or absence
of NO2. Lefohn and Tingey (1984) reported that most of the sites analyzed experienced fewer
than 10 co-occurrences (when both pollutants were present at an hourly average concentration
>0.05 ppm). Figure AX3-71 summarizes the simultaneous co-occurrence patterns reported by
Lefohn and Tingey (1984). The authors noted that several urban monitoring sites in the South
Coast Air Basin experienced more than 450 co-occurrences. For more moderate areas of the
country, Lefohn et al. (1987) reported that even with a threshold of 0.03 ppm O3, the number of
co-occurrences with NO, was small.
AX3-123
-------
100
80
55 60
"S
o
n
40
20
ll
+++-
I
occoc:icoo
CM co -^- 10 0.05 ppm were
characterized. The data were not segregated by location settings categories (i.e., rural, suburban,
and urban and center city) or land use types (i.e., agricultural, commercial, desert, forest,
industrial, mobile, or residential). Data capture was not a consideration in the analysis. The data
were characterized over the EPA-defined O3 season (Table AX3-1). In 2001, there were
341 monitoring sites that co-monitored O3 and NO2. Because of possible missing hourly average
concentration data during periods when co-monitoring may have occurred, no attempt was made
to characterize the number of co-occurrences in the 0 category. Thus, co-occurrence patterns
were identified for those monitoring sites that experienced one or more co-occurrences.
Figure AX3-72 illustrates the results of the analysis. Similar to the analysis summarized
by Lefohn and Tingey (1984), most of the collocated monitoring sites analyzed, using the 2001
data, experienced fewer than 10 co-occurrences (when both pollutants were present at an hourly
average concentration >0.05 ppm).
AX3-124
-------
tfl
**
u—
O
1_
Qi
E
40-
35-
30-
25-
20-
15-
10-
5-
cn
CM
o
CM
CD
CO
o
oo
CD
u
cn
CD
o
CD
CO
CO
o
00
CD
CD
I
o
Cn
cn cn
O i-
cn
CM
cn
CO
cn
in
o
o
o
CM
O
CO
o
-=1-
o
m
Number of Co-Occurances (Hours)
Figure AX3-72. The co-occurrence pattern for O3 and NO2 using 2001 data from
the AQS.
AX3.8.3 Co-Occurrence of Ozone with Sulfur Dioxide
Because elevated SO2 concentrations are mostly associated with industrial activities (U.S.
Environmental Protection Agency, 1992), co-occurrence observations are usually associated
with monitors located near these types of sources. Lefohn and Tingey (1984) reported that,
for the rural and nonrural monitoring sites investigated, most sites experienced fewer than
10 co-occurrences of SO2 and O3. Lefohn et al. (1987) reported that even with a threshold of
0.03 ppm O3, the number of co-occurrences with SO2 was small. Figure AX3-73 illustrates the
simultaneous co-occurrence results reported by Lefohn and Tingey (1984).
Meagher et al. (1987) reported that several documented O3 episodes at specific rural
locations appeared to be associated with elevated SO2 levels. The investigators defined the
co-occurrence of O3 and SO2 to be when hourly mean concentrations were >0.10 and 0.01 ppm,
respectively.
The above discussion was based on the co-occurrence patterns associated with the presence
or absence of hourly average concentrations of pollutant pairs. Taylor et al. (1992) have
discussed the joint occurrence of O3, nitrogen, and sulfur in forested areas using cumulative
exposures of O3 with data on dry deposition of sulfur and nitrogen. The authors concluded in
AX3-125
-------
100
80
(A
W 60
"5
| 40
z
20
n
-
-
-
II..
oooooooo
Number of Co-Occurrences
Figure AX3-73. The co-occurrence pattern for O3 and SO2.
Source: Lefohn and Tingey (1984).
their study that the forest landscapes with the highest loadings of sulfur and nitrogen via dry
deposition tended to be the same forests with the highest average O3 concentrations and largest
cumulative exposure. Although the authors concluded that the joint occurrences of multiple
pollutants in forest landscapes were important, nothing was mentioned about the hourly
co-occurrences of O3 and SO2 or of O3 and NO2.
Using 2001 data from the EPA AQS database, patterns that showed air pollutant pairs
of O3/SO2 appearing at the same hour of the day at concentrations >0.05 ppm were characterized.
The data were not segregated by location settings categories (i.e., rural, suburban, and urban and
center city) or land use types (i.e., agricultural, commercial, desert, forest, industrial, mobile,
or residential). Data capture was not a consideration in the analysis. In 2001, there were
246 monitoring sites that co-monitored O3 and SO2. Because of possible missing hourly average
concentration data during periods when co-monitoring may have occurred, no attempt was made
to characterize the number of co-occurrences in the 0 category. Thus, co-occurrence patterns
AX3-126
-------
were identified for those monitoring sites that experienced one or more co-occurrences.
Figure AX3-74 shows the results from this analysis for the simultaneous co-occurrence of O3
and SO2. Similar to the analysis summarized by Lefohn and Tingey (1984), most of the
collocated monitoring sites analyzed, using the 2001 data, experienced fewer than 10 co-
occurrences (when both pollutants were present at an hourly average concentration >0.05 ppm).
60-
50-
40-
o
i- 30H
0)
E 20-
3
z
10-
CD
CM
O
CM
G>
CO
O
co
O)
"
o
in
o
o
(35
O) O)
O T-
Ol
CM
O)
co
en
if!
o
o
o
CM
O
CO
Number of Co-Occurances (Hours)
Figure AX3-74. The co-occurrence pattern for O3 and SO2 using 2001 data from AQS.
AX3.8.4 Co-Occurrence of Ozone and Daily PM2 5
Using 2001 data from the EPA AQS, the daily co-occurrence of PM2 5 and O3 over a 24-h
period was characterized. There were 362 sites where PM25 and O3 monitors were collocated.
As described in the introduction selection of this annex, a daily co-occurrence of O3 and PM25 is
subjectively defined as an hourly average O3 concentration >0.05 ppm and a PM2 5 24-h
concentration >40 |ig/m3 occurring over the same 24-h period. Figure AX3-75 illustrates the
daily co-occurrence patterns observed. Using 2001 data from the AQS, the daily co-occurrence
of PM25 and O3 was infrequent.
AX3-127
-------
0)
160--
140-
120-
100-
80-
40-
20-
0--
cn
CM
CD
CM
CD
co
O
CO
CD
co
o
m
CD
OO
O
OO
o
o
CD
CN
o
CN
CD
CO
O
OO
cn
in
o
IO
Number of Co-Occurances (Hours)
Figure AX3-75. The co-occurrence pattern for O3 and PM2 5 using 2001 data from AQS.
AX3.8.5 Co-Occurrence of Ozone with Acid Precipitation
Concern has been expressed about the possible effects on vegetation from co-occurring
exposures of O3 and acid precipitation (Prinz et al., 1985; National Acid Precipitation
Assessment Program, 1987; Prinz and Krause, 1989). Little information has been published
concerning the co-occurrence patterns associated with the joint distribution of O3 and acidic
deposition (i.e., H+). Lefohn and Benedict (1983) reviewed the EPA SAROAD monitoring data
for 1977 through 1980 and, using National Atmospheric Deposition Program (NADP) and EPRI
wet deposition data, evaluated the frequency distribution of pH events for 34 NADP and 8 EPRI
chemistry monitoring sites located across the United States. Unfortunately, there were few sites
where O3 and acidic deposition were co-monitored.
As a result, Lefohn and Benedict (1983) focused their attention on O3 and acidic deposition
monitoring sites that were closest to one another. In some cases, the sites were as far apart as
144 km. Using hourly O3 monitoring data and weekly and event acidic deposition data from the
NADP and EPRI databases, the authors identified specific locations where the hourly mean O3
concentrations were >0.1 ppm and 20% of the wetfall daily or weekly samples were below pH
4.0. Elevated levels of O3 were defined as hourly mean concentrations equal to or greater than
0.1 ppm. Although for many cases, experimental research results of acidic deposition on
AX3-128
-------
agricultural crops show few effects at pH levels >3.5 (National Acid Precipitation Assessment
Program, 1987), it was decided to use a pH threshold of 4.0 to take into consideration the
possibility of synergistic effects between O3 and acidic deposition.
Based on their analysis, Lefohn and Benedict (1983) identified five sites with the potential
for agricultural crops to experience additive, less than additive, or synergistic (i.e., greater than
additive) effects from elevated O3 and H+ concentrations. The authors stated that they believed,
based on the available data, the greatest potential for interaction between acid rain and O3
concentrations in the United States, with possible effects on crop yields, may be in the most
industrialized areas (e.g., Ohio and Pennsylvania). However, they cautioned that, because no
documented evidence existed to show that pollutant interaction had occurred under field growth
conditions and ambient exposures, their conclusions should only be used as a guide for further
research.
In their analysis, Lefohn and Benedict (1983) found no collocated sites. The authors
rationalized that data from non-co-monitoring sites (i.e., O3 and acidic deposition) could be used
because O3 exposures are regional in nature. However, work by Lefohn et al. (1988) has shown
that hourly mean O3 concentrations vary from location to location within a region, and that
cumulative indices, such as the percent of hourly mean concentrations >0.07 ppm, do not form a
uniform pattern over a region. Thus, extrapolating hourly mean O3 concentrations from known
locations to other areas within a region may provide only qualitative indications of actual O3
exposure patterns.
In the late 1970s and the 1980s, both the private sector and the government funded research
efforts to better characterize gaseous air pollutant concentrations and wet deposition. The event-
oriented wet deposition network, EPRI/Utility Acid Precipitation Study Program, and the weekly
oriented sampling network, NADP, provided information that can be compared with hourly
mean O3 concentrations collected at several co-monitored locations. No attempt was made to
include H+ cloud deposition information. In some cases, for mountaintop locations (e.g.,
Clingman's Peak, Shenandoah, Whiteface Mountain, and Whitetop Mountain), the H+ cloud
water deposition is greater than the H+ deposition in precipitation (Mohnen, 1989), and the
co-occurrence patterns associated with O3 and cloud deposition will be different from those
patterns associated with O3 and deposition by precipitation.
AX3-129
-------
Smith and Lefohn (1991) explored the relationship between O3 and H+ in precipitation,
using data from sites that monitored both O3 and wet deposition simultaneously and within
one-minute latitude and longitude of each other. The authors reported that individual sites
experienced years in which both H+ deposition and total O3 exposure were at least moderately
high (i.e., annual H+ deposition >0.5 kg ha"1 and an annual O3 cumulative, sigmoidally weighted
exposure (W126) value >50 ppm-h). With data compiled from all sites, it was found that
relatively acidic precipitation (pH <4.31 on a weekly basis or pH <4.23 on a daily basis)
occurred together with relatively high O3 levels (i.e., W126 values >0.66 ppm-h for the same
week or W126 values >0.18 ppm-h immediately before or after a rainfall event) approximately
20% of the time, and highly acidic precipitation (i.e, pH <4.10 on a weekly basis or pH <4.01 on
a daily basis) occurred together with a high O3 level (i.e., W126 values > 1.46 ppm-h for the
same week or W126 values >0.90 ppm-h immediately before or after a rainfall event)
approximately 6% of the time. Whether during the same week or before, during, or after a
precipitation event, correlations between O3 level and pH (or H+ deposition) were weak to
nonexistent. Sites most subject to relatively high levels of both H+ and O3 were located in the
eastern United States, often in mountainous areas.
AX3.8.6 Co-Occurrence of Ozone with Acid Cloudwater
In addition to the co-occurrence of O3 and acid precipitation, results have been reported on
the co-occurrence of O3 and acidic cloudwater in high-elevation forests. Vong and Guttorp
(1991) characterized the frequent O3-only and pH-only, single-pollutant episodes, as well as the
simultaneous and sequential co-occurrences of O3 and acidic cloudwater. The authors reported
that both simultaneous and sequential co-occurrences were observed a few times each month
above the cloud base. Episodes were classified by considering hourly O3 average concentrations
>0.07 ppm and cloudwater events with pH <3.2. The authors reported that simultaneous
occurrences of O3 and pH episodes occurred two to three times per month at two southern sites
(Mitchell, NC and Whitetop, VA) and the two northern sites (Whiteface Mountain, NY and
Moosilauke, NH) averaged one episode per month. No co-occurrences were observed at the
central Appalachian site (Shenandoah, VA), due to a much lower cloud frequency. Vong and
Guttorp (1991) reported that the simultaneous occurrences were usually of short duration
AX3-130
-------
(mean =1.5 h/episode) and were followed by an O3-only episode. As would be expected,
O3-only episodes were longer than co-occurrences and pH episodes, averaging an 8-h duration.
AX3.9 THE METHODOLOGY FOR DETERMINING POLICY
RELEVANT BACKGROUND OZONE CONCENTRATIONS
AX3.9.1 Introduction
Background O3 concentrations used for NAAQS-setting purposes are referred to as Policy
Relevant Background (PRB) O3 concentrations. Policy Relevant Background concentrations are
those concentrations that would result in the United States in the absence of anthropogenic
emissions in continental North America (the United Sates, Canada and Mexico). Policy
Relevant Background concentrations include contributions from natural sources everywhere in
the world and from anthropogenic sources outside these three countries. For the purposes of
informing decisions about O3 NAAQS, EPA assesses risks to human health and environmental
effects to O3 levels in excess of PRB concentrations. Issues concerning the methodology for
estimating PRB O3 concentrations are described in detail in Annex AX3, Section AX3.9.
Contributions to PRB O3 include: photochemical interactions involving natural emissions
of VOCs, NOX, and CO; the long-range transport of O3 and its precursors from outside North
America; and stratospheric-tropospheric exchange (STE). Processes involved in STE are
described in detail in Annex AX2.3. Natural sources of O3 precursors include biogenic
emissions, wildfires, and lightning. Biogenic emissions from agricultural activities are not
considered in the formation of PRB O3.
Most of the issues concerning the calculation of PRB O3 center on the origin of springtime
maxima in surface O3 concentrations observed at monitoring sites in relatively unpolluted areas
of the United States and on the capability of the current generation of global-scale, three-
dimensional chemistry transport models to correctly simulate their causes. These issues are
related to the causes of the occurrence of high O3 values, especially those averaged over 1-h to
8-h observed at O3 monitoring sites during late winter through spring (i.e., February to June).
The issues raised do not affect interpretations of the causes of summertime O3 episodes as
strongly. Summertime O3 episodes are mainly associated with slow-moving high-pressure
AX3-131
-------
systems characterized by limited mixing between the planetary boundary layer and the free
troposphere (Section AX2.3).
Springtime maxima are observed at national parks mainly in the western United States that
are relatively clean (Section AX3.2.2; Figures AX3-76a,b) and at a number of other relatively
unpolluted monitoring sites throughout the Northern Hemisphere. Spring maxima in
tropospheric O3 were originally attributed to transport from the stratosphere by Regener (1941)
as cited by Junge (1963). Junge (1963) also cited measurements of springtime maxima in O3
concentrations at Mauna Loa (elevation 3400 m) and at Arkosa, Germany (an alpine location,
elevation 1860 m). Measurements of radioactive debris transported downward from the
stratosphere as the result of nuclear testing during the 1960s also show springtime maxima
(Ludwig et al., 1977). However, more recent studies (Lelieveld and Dentener, 2000; Browell
et al., 2003) attribute the springtime maximum in tropospheric O3 concentrations to tropospheric
production rather than transport from the stratosphere. It should be noted here that O3 in the free
troposphere is subject to chemical loss on time scales much shorter than for decay of most
radio-isotopes produced by nuclear testing that were used as tracers of stratospheric air such
as 14C, 137Cs and 90Sr.
Springtime O3 maxima were observed in low-lying surface measurements during the late
19th century. However, these measurements are quantitatively highly uncertain, and extreme
caution should be exercised in their use. Concentrations of approximately 0.036 ppm for the
daytime average and of 0.030 ppm for the nighttime averages were reported for Zagreb, Croatia
using the Schonbein method during the 1890s (Lisac and Grubisic, 1991). Of the numerous
measurements of tropospheric O3 made in the 19th century, only the iodine catalyzed oxidation
of arsenite has been verified with modern laboratory methods. Kley et al. (1988) reconstructed
the apparatus used between 1876 and 1910 in Montsouris, outside Paris, and evaluated it for
accuracy and specificity. They concluded that O3 mixing ratios ranged from 5 to 16 ppb with
uncertainty of ±2 ppb. Interferences from SO2 were avoided as the Montsouris data were
selected to exclude air from Paris, the only source of high concentrations of SO2 at that time.
Uncertainties in the humidity correction to the Schonbein reading will lead to considerable
inaccuracies in the seasonal cycle established by this method (Pavelin et al., 1999). Because of
the uncertainties in the earlier methods, it is difficult to quantify the differences between
AX3-132
-------
Yellowstone National Park
Maximum Hourly Concentration
1998-2001
Month
I 1998
I 1999 D 2000 D 2001
Figure AX3-76a. Monthly maximum hourly average O3 concentrations at Yellowstone
National Park, Wyoming in 1998,1999, 2000, and 2001.
ooooooooooooooooooooooooooo
Time
Figure AX3-76b. Hourly average O3 concentrations at Yellowstone National Park,
Wyoming for the period January to December 2001.
Source: U.S. Environmental Protection Agency (2003).
AX3-133
-------
surface O3 concentrations measured in the last half of the 19th century at certain locations in
either Europe or North America with those currently monitored at remote locations in the world.
Observations of O3 profiles at a large number of sites indicate a positive gradient in O3
mixing ratios with increasing altitude in the troposphere and a springtime maximum in O3
concentrations in the upper troposphere (Logan, 1999). As discussed in Section AX2.3.1, STE
affects the middle and upper troposphere more than the lower troposphere. It is, therefore,
reasonable to suppose that the main cause of this positive gradient is STE. However, deep
convection transports pollutants upward and can result in an increase in the pollutant mixing
ratio with altitude downwind of surface source regions as shown in Figure AX3-77. This effect
can be seen in differences in ozonesonde profiles as one moves eastward across the United States
(Newchurch et al., 2003). In addition, O3 formed by lightning-generated NOX also contributes to
the vertical O3 gradient (Lelieveld and Dentener, 2000). This O3 could be either background or
not, depending on the sources of radical precursors. Another contributing factor is the increase
of O3 lifetime with altitude (Wang et al., 1998). Free-tropospheric O3 is not predominantly of
stratospheric origin, nor is it all natural; it is mostly controlled by production within the
troposphere and includes a major anthropogenic enhancement (e.g., Berntsen et al., 1997;
Roelofs et al., 1997; Wild and Akimoto, 2001).
Stohl (2001), Wernli and Borqui (2002), Seo and Bowman (2002), James et al. (2003a,b),
Sprenger and Wernli (2003), and Sprenger et al. (2003) addressed the spatial and temporal
variability in stratosphere to troposphere transport. Both Stohl (2001) and Sprenger et al. (2003)
produced 1-year climatologies of tropopause folds based on a 1° by 1° gridded meteorological
model data set. They each found that the probability of deep folds (penetrating to the 800 hPa
level) was maximum during winter (December through February) with the highest frequency of
folding extending from Labrador down the east coast of North America. However, these deep
folds occurred in <1% of the 6-h intervals for which meteorological data was assimilated for grid
points in the continental United States, with a higher frequency in Canada. They observed a
higher frequency of more shallow folds (penetrating to the upper troposphere) and medium folds
(penetrating to levels between 500 and 600 hPa) of about 10% and 1 to 2%, respectively. These
events occur preferentially across the subtropics and the southern United States. At higher
latitudes, other mechanisms such as the erosion of cut-off lows and the breakup of stratospheric
AX3-134
-------
a. Carbon Monoxide
Surface
98°Wf 97C
Oklahoma City
96° 95° 94°
Longitude
b. Ozone
g>
'o
Surface
98°W| 97C
Oklahoma City
96° 95° 94°
Longitude
Figure AX3-77. (a) Contour plot of CO mixing ratios (ppbv) observed in and near the
June 15,1985, mesoscale convective complex in eastern Oklahoma.
Heavy line shows the outline of the cumulonimbus cloud. Dark shading
indicates high CO and light shading indicates low CO. Dashed contour
lines are plotted according to climatology since no direct measurements
were made in that area, (b) Same as (a) but for O3 (ppbv).
Source: Dickerson et al. (1987).
AX3-135
-------
streamers are likely to play an important role in STE. A 15-year model climatology by Sprenger
and Wernli (2003) showed the consistent pattern of STE occurring over the primary storm tracks
along the Asian and North American coasts. This climatology, and the one of James et al.
(2003a,b) both found that recent stratospheric air associated with deep intrusions are relatively
infrequent occurrences in these models. Thus, stratospheric intrusions are most likely to directly
affect the middle and upper troposphere, not the planetary boundary layer. However, this O3 can
still exchange with the planetary boundary layer through convection or through large-scale
subsidence as described later in this subsection and in Sections AX2.3.2, AX2.3.3, and AX2.3.4.
These results are in accord with the observations of Ludwig et al. (1977) over the western United
States. It should also be remembered that stratospheric O3 injected into the upper troposphere is
subject to chemical destruction as it is transported downward toward the surface.
Ozone concentrations measured at RRMS in the Northern Hemisphere have been compiled
by Vingarzan (2004) and are reproduced here in Tables AX3-10, AX3-11, and AX3-12. Data
for annual mean/median concentrations show a broad range, as do annual maximum 1-h
concentrations. Generally, concentrations increase with elevation and the highest concentrations
are found during spring. The overall average of the annual median O3 concentrations at all sites
in the continental United States is about 30 ppb and excluding higher elevation sites it is about
24 ppb. Maximum concentrations may be related to stratospheric intrusions, wildfires, and
intercontinental or regional transport of pollution. However, it should be noted that all of these
sites are affected by anthropogenic emissions to some extent making an interpretation based on
these data alone problematic.
Daily 1-h maximum O3 concentrations exceeding 50 or 60 ppb are observed during late
winter and spring in southern Canada and at sites in national parks as shown in
Tables AX3-13, AX3-14, and Figure AX3-78. That these high values can occur during late
winter when there are low sun angles and cold temperatures may imply a negligible role for
photochemistry and a major role for stratospheric intrusions. However, active photochemistry
occurs even at high latitudes during late winter. Rapid O3 loss, apparently due to multiphase
chemistry involving bromine atoms (see Section 2.2.10) occurs in the Arctic marine boundary
layer. The Arctic throughout much of winter is characterized by low light levels, temperatures,
and precipitation, and can act as a reservoir for O3 precursors such as PAN and alkyl nitrates,
AX3-136
-------
Table AX3-10. Range of Annual (January-December) Hourly Ozone Concentrations
(ppb) at Background Sites Around the World (CMDL, 2004)
Location
Pt. Barrow, Alaska
Ny Alesund, Svalbard, Spitsbergen*
Mauna Loa, Hawaii0
Elevation (m)
11
475
3397
Period of Record
1992-2001
1989-1993
1992-2001
Range of Annual Means
23-29
28-33b
37-46d
"University of Stockholm Meteorological Institute.
bAnnual medians
10:00-18:00 UTC.
dHigh elevation site.
Source: Vingarzan (2004).
Table AX3-11. Range of Annual (January-December) Hourly Median and Maximum
Ozone Concentrations (ppb) at Background Stations in Protected Areas of the
United States (CASTNet, 2004)
Location
Denali NP, Alaska
Glacier NP, Montana
Voyageurs NP, Minnesota
Theodore Roosevelt NP,
North Dakota
Yellowstone NP, Wyoming
Rocky Mountain NP, Colorado
Olympic NP, Washington
North Cascades NP, Washington
Mount Rainier NP, Washington
Lassen NP, California
Virgin Islands NP,
U.S. Virgin Islands
Elevation
(m)
640
976
429
850
2469
2743
125
109
421
1756
80
Period of
Record
1998-2001
1989-2001
1997-2001
1983-2001
1996-2001
1994-2001
1998-2001
1996-2001
1995-2001
1995-2001
1998-2001
Range of Annual
Medians
29-34
19-27
28-35
29-43
37-45a
40-47a
19-22
14-18
38371
38-43a
19-24
Range of Annual
Maxima
49-68
57-77
74-83
61-82
68-79a
68-102a
50-63
48-69
54-98
81-109a
50-64
aHigh elevation site.
Source: Vingarzan (2004).
AX3-137
-------
Table AX3-12. Range of Annual (January-December) Hourly Median and Maximum
Ozone Concentrations (ppb) at Canadian Background Stations (CAPMoN% 2003)
Location
Kejimkujik, Nova Scotiab
Montmorency, Quebec
Algoma, Ontario3
Chalk River, Ontario
Egbert, Ontario3
E.L.A., Ontario
Bratt's Lake, Saskatchewan
Esther, Alberta
Saturna Island, British Columbia
Elevation
(m)
127
640
411
184
253
369
588
707
178
Period of
Record
1989-2001
1989-1996
1988-2001
1988-1996
1989-2001
1989-2001
1999-2001
1995-2001
1992-2001
Range of Annual
Medians
25-34
28-32
27-33
25-31
27-32
28-33
26-29
26-31
23-27
Range of Annual
Maxima
76-116
73-99
76-108
79-107
90-113
64-87
63-68
63-78
65-82
"Canadian Air and Precipitation Monitoring Network.
bStations affected by long-range transport of anthropogenic emissions.
Source: Vingarzan (2004).
which build up and can then photolyze when sun angles are high enough during late winter and
early spring. Long-range transport of total odd nitrogen species (NOy) (defined in AX2.2.2) and
VOCs to Arctic regions can occur from midlatitude-source regions. In addition, O3 can be
transported from tropical areas in the upper troposphere followed by its subsidence at mid and
high latitudes (Wang et al., 1998).
Penkett (1983), and later Penkett and Brice (1986), first observed a spring peak in PAN at
high northern latitudes and hypothesized that winter emissions transported into the Arctic would
be mixed throughout a large region of the free troposphere and transformed into O3 as solar
radiation returned to the Arctic in the spring. Subsequent observations (Dickerson, 1985)
confirmed the presence of strata of high concentrations of reactive nitrogen compounds at high
latitudes in early spring. Bottenheim et al. (1990, 1993) observed a positive correlation
between O3 and NO2 in the Arctic spring. Jaffe et al. (1991) found NOy concentrations
approaching 1 ppb in Barrow, Alaska, in the spring and attributed them to long-range transport.
AX3-138
-------
Table AX3-13. Number of Hours >0.05 ppm for Selected Rural O3 Monitoring in the United States
by Month for the Period 1988 to 2001
X
OJ
1
OJ
VO
Site Name
Denali National Park, Alaska
Denali National Park, Alaska
Denali National Park, Alaska
Denali National Park, Alaska
Denali National Park, Alaska
Yellowstone National Park,
Wyoming
Yellowstone National Park,
Wyoming
Yellowstone National Park,
Wyoming
Yellowstone National Park,
Wyoming
Yellowstone National Park,
Wyoming
Month
February
March
April
May
June
February
March
April
May
June
1988
0
0
217
26
0
0
194
228
225
58
1989 1990
0 0
0 0
0 2
1 0
0 0
0
2
17
2
67
1991
0
0
0
24
0
11
4
16
10
139
1992
0
0
64
10
0
3
95
217
196
33
1993
0
0
10
17
0
0
26
62
47
28
1994
0
0
31
1
0
21
285
311
180
116
1995
0
0
21
54
0
6
14
185
193
81
1996
52
12
97
27
0
7
65
212
94
1997
0
0
51
35
0
1
98
163
216
149
1998
0
122
236
79
0
5
150
385
289
78
1999
0
17
119
29
22
252
509
517
458
212
2000
0
0
0
0
0
23
286
242
240
181
2001
14
24
302
98
6
77
307
461
350
172
-------
Table AX3-13 (cont'd). Number of Hours >0.05 ppm for Selected Rural O3 Monitoring in the United States by Month
for the Period 1988 to 2001
X
OJ
1
o
Site Name
Glacier National Park,
Montana
Glacier National Park,
Montana
Glacier National Park,
Montana
Glacier National Park,
Montana
Glacier National Park,
Montana
Voyageurs National Park,
Minnesota
Voyageurs National Park,
Minnesota
Voyageurs National Park,
Minnesota
Voyageurs National Park,
Minnesota
Voyageurs National Park,
Minnesota
Month
February
March
April
May
June
February
March
April
May
June
1988 1989 1990
0
49
31
20
24
300
620
48 0 31
183 33 14
92 2
1991
0
9
64
81
37
0
0
22
10
0
1992
0
24
29
67
31
0
1
27
174
55
1993
0
10
5
41
5
43
94
56
78
50
1994
0
23
45
66
29
22
10
65
96
66
1995
8
40
16
51
13
0
39
30
107
190
1996
0
35
46
51
119
0
49
64
111
37
1997
0
9
52
4
0
23
220
128
146
221
1998
0
6
49
122
3
0
40
254
191
25
1999
0
17
128
103
0
6
215
221
247
23
2000
0
5
16
63
6
32
60
175
143
28
2001
0
4
0
23
0
0
0
0
62
95
-------
Table AX3-14. Number of Hours >0.06 ppm for Selected Rural O3 Monitoring Sites in the United States by Month
for the Period of 1988 to 2001
Site Name
Denali National Park,
Alaska
Denali National Park,
Alaska
Denali National Park,
Alaska
Denali National Park,
Alaska
Denali National Park,
Alaska
;> Yellowstone National
X Park, Wyoming
i
£ Yellowstone National
^ Park, Wyoming
Yellowstone National
Park, Wyoming
Yellowstone National
Park, Wyoming
Yellowstone National
Park, Wyoming
Glacier National Park,
Montana
Glacier National Park,
Month
February
March
April
May
June
February
March
April
May
June
February
March
1988 1989 1990
000
000
000
000
000
0 0
37 0
59 0
20 0
8 7
0
1
1991
0
0
0
0
0
0
0
0
0
18
0
0
1992
0
0
0
0
0
0
0
29
61
2
0
0
1993
0
0
0
0
0
0
0
0
3
1
0
0
1994
0
0
0
0
0
0
0
20
42
13
0
0
1995
0
0
0
0
0
0
0
4
24
0
0
0
1996
0
0
2
0
0
0
0
38
0
0
0
1997
0
0
0
0
0
0
1
0
26
22
0
0
1998
0
0
0
0
0
0
0
64
54
4
0
0
1999
0
0
0
0
0
6
120
158
169
27
0
0
2000
0
0
0
0
0
0
1
11
49
43
0
0
2001
0
0
0
9
2
0
4
77
139
18
0
0
Montana
-------
Table AX3-14 (cont'd). Number of Hours >0.06 ppm for Selected Rural O3 Monitoring Sites in the United States by Month
for the Period of 1988 of 2001
X
OJ
1
to
Site Name
Glacier National Park,
Montana
Glacier National Park,
Montana
Glacier National Park,
Montana
Voyageurs National
Park, Minnesota
Voyageurs National
Park, Minnesota
Voyageurs National
Park, Minnesota
Voyageurs National
Park, Minnesota
Voyageurs National
Park, Minnesota
Month 1988 1989
April
May
June
February 1 0
March 0 0
April 9 0
May 77 6
June 30
1990
0
2
0
0
0
1
0
0
1991
0
7
3
0
0
0
0
0
1992
1
13
1
0
0
0
40
28
1993
0
0
0
1
34
5
9
17
1994
0
0
1
8
0
8
40
5
1995
0
4
0
0
5
0
2
113
1996
0
5
16
0
2
17
27
12
1997
0
0
0
0
15
2
46
115
1998
2
16
0
0
0
57
53
0
1999
1
19
0
0
9
24
139
5
2000
0
8
0
0
4
41
43
0
2001
0
0
0
0
0
0
6
32
-------
54
Figure AX3-78. Maximum hourly average O3 concentrations at rural monitoring sites in
Canada and the United States in February from 1980 to 1998.
Source: Lefohnetal. (2001).
Beine et al. (1997) and Honrath et al. (1996) measured O3, PAN, and NOX in Alaska and
Svalbard, Norway and concluded that PAN decomposition can lead to photochemical O3
production. At Poker Flat, Alaska, O3 production was directly observable. Herring et al. (1997)
tracked springtime O3 maxima in Denali National Park, Alaska, an area one might presume to be
pristine. They measured NOX and hydrocarbons and concluded that, in the spring, O3 was
produced predominantly by photochemistry at a calculated rate of 1 to 4 ppb/day, implying that
the O3 observed could be produced on timescales ranging from about a week to a month.
Solberg et al. (1997) tracked the major components of NOy in remote Spitsbergen, Norway for
the first half of the year 1994. They observed high concentrations of PAN (800 ppt) peaking
simultaneously with O3 (45 to 50 ppb) and attributed this to the long-range transport of pollution
AX3-143
-------
and to photochemical smog chemistry. These investigators concluded, in general, that large
regions of the Arctic store high concentrations of O3 precursors in the winter and substantial
quantities of O3 are produced by photochemical reactions in the spring. Although reactions with
high-activation-energy barriers may be ineffective, reactions with low- or no activation-energy
barriers (such as radical-radical reactions) or negative temperature dependencies will still
proceed. Indeed, active photochemistry is observed in the coldest regions of the stratosphere and
mesosphere. While it is expected that photochemical production rates of O3 will increase with
decreasing solar zenith angle as one moves southward from the locations noted above, it should
not be assumed that photochemical production of O3 does not occur during late winter and spring
at mid- and high-latitudes.
Perhaps the most thorough set of studies investigating causes of springtime maxima in
surface O3 has been performed as part of the AEROCE and NARE studies (cf, Sections
AX2.3.4a,b) and TOPSE (Browell et al., 2003). These first two studies found that elevated or
surface O3 > 40 ppb at Bermuda, at least, arises from two distinct sources: the polluted North
American continent and the stratosphere. It was also found that these sources mix in the upper
troposphere before descending as shown in Figure AX3-79. (In general, air descending behind
cold fronts contains contributions from intercontinental transport and the stratosphere.) These
studies also concluded that it is impossible to determine sources of O3 without ancillary data that
could be used either as tracers of sources or to calculate photochemical production and loss rates.
In addition, subsiding back trajectories do not necessarily imply a free-tropospheric or
stratospheric origin for O3 observed at the surface, since the subsiding conditions are also
associated with strong inversions and clear skies that promote O3 production within the boundary
layer. Thus, it would be highly problematic to use observations alone as estimates of PRB O3
concentrations, especially for sites at or near sea level.
The IPCC Third Assessment Report (TAR) (2001) gave a large range of values for terms in
the tropospheric O3 budget. Estimates of O3 STE of O3 ranged over a factor of three from 391 to
1440 Tg/year in the twelve models included in the intercomparison; many of the models
included in that assessment overestimated O3 STE. However, the overestimates likely reflected
errors in assimilated winds in the upper troposphere (Douglass et al., 2003; Schoeberl et al.,
2003; Tan et al., 2004; van Noije et al., 2004). The budgets of tropospheric O3 calculated since
the IPCC TAR are shown in Table AX3-15. Simulation of stratospheric intrusions is
AX3-144
-------
Altitude
Stratospheric Air
,*-
Cold Front (A)
Stratospheric Air
Cold Front (B)
West
-1500
Distance (km)
East
Figure AX3-79. Schematic diagram of a meteorological mechanism involved in high
concentrations of O3 found in spring in the lower troposphere off
the American East Coast. Subsidence behind the first cold front meets
convection ahead of a second cold front such that polluted air and O3
from the upper troposphere/ lower stratosphere are transported in close
proximity (or mixed) and advected over the North Atlantic Ocean. The
vertical scale is about 10 km; the horizontal scale about 1500 km. (Note
that not all cold fronts are associated with squall lines and that mixing
occurs even in their absence.)
Source: Prados (2000).
notoriously difficult in global models, and O3 STE is generally parameterized in these models.
However, as can be seen from inspection of Table AX3-15, improvements in assimilation
techniques have improved and narrowed estimates of STE. A model intercomparison looking at
actual STE events found significant variations in model results that depended significantly on the
type and horizontal resolution of the model (Meloen et al., 2003; Cristofanelli et al., 2003).
In particular, it was found that the Lagrangian perspective (as opposed to the Eulerian
perspective used in most global scale CTMs) was necessary to characterize the depths and
residence times of individual events (Sprenger and Wernli, 2003; James et al., 2003a,b). A few
studies of the magnitude of the O3 STE have been made based on chemical observations in the
lower stratosphere or combined chemistry and dynamics (e.g., 450 Tg/year net global [Murphy
and Fahey, 1994]; 510 Tg/year net global extratropics only [Gettelman et al., 1997]; and
550 ± 140 Tg/year [Olsen et al., 2002]).
AX3-145
-------
Table AX3-15. Global Budgets of Tropospheric Ozone (Tg year *) for the Present-day Atmosphere
.
X
OJ
1
ON
Reference
TAR4
Lelieveld and
Dentener (2000)
Beyetal. (200 la)5
Horowitz et al.
(2003)
Von Kuhlmann
et al. (2003)
Shindell et al. (2003)
Park et al. (2004)
Rotman et al. (2004)
Wong et al. (2004)
Model
1 1 models
GEOS-CHEM
MOZART-2
MATCH-MPIC
GISS
UMD-CTM
IMPACT
SUNYA/UiO GCCM
Stratosphere-
Troposphere
Exchange
(STE)
770 ± 400
570
470
340
540
417
480
660
600
Chemical
Production 2
3420 ± 770
3310
4900
5260
4560
NR6
NR
NR
NR
Chemical
Loss2
3470 ± 520
3170
4300
4750
4290
NR
NR
NR
NR
Dry
Deposition
770 ± 180
710
1070
860
820
1470
1290
830
1100
Burden
(Tg)
300 ± 30
350
320
360
290
349
340
NR
376
Lifetime
(days)3
24 ±2
33
22
23
21
NR
NR
NR
NR
1 From global CTM simulations describing the atmosphere of the last decade of the 20th century.
2 Chemical production and loss rates are calculated for the odd oxygen family, usually defined as Ox = O3 + O + NO2 + 2NO3 + 3N2O5 + HNO4 +
peroxyacylnitrates (and sometimes HNO3), to avoid accounting for rapid cycling of O3 with short-lived species that have little implication for its budget.
Chemical production is mainly contributed by reactions of NO with peroxy radicals, while chemical loss is mainly contributed by the O('D) + H2O reaction
and by the reactions of O3 with HO2', -OH, and alkenes. Several models in this table do not report production and loss separately ("NR" entry in the table),
reporting instead net production. However, net production is not a useful quantity for budget purposes, because (1) it represents a small residual between
large production and loss, (2) it represents the balance between STE and dry deposition, both of which are usually parameterized as a flux boundary
condition.
3 Calculated as the ratio of the burden to the sum of chemical and deposition losses
4 Means and standard deviations from an ensemble of 11 CTM budgets reported in the IPCC TAR. The mean budget does not balance exactly because
only 9 CTMs reported chemical production and loss statistics.
5 The Martin et al. (2003) more recent version of GEOS-CHEM gives identical rates and burdens.
6 Not reported.
-------
Even if the magnitude of cross-tropopause O3 fluxes in global CTMs are calculated
correctly in an annual mean sense, it should be noted that stratospheric intrusions occur
episodically following the passage of cold fronts at midlatitudes. Of major concern is the ability
of global-scale CTMs to simulate individual intrusions and the effects on surface O3
concentrations that may result during these events. As noted in Section AX2.3.1, these
intrusions occur in "ribbons" ~ 200 to 1000 km long, 100 to 300 km wide, and 1 to 4 km thick.
An example of a stratospheric intrusion occurred in Boulder, CO (EPA AQS Site 080130011;
formally AIRS) on May 6, 1999 (Lefohn et al., 2001). At 1700 UTC (1000 hours LSI)
an hourly average concentration of 0.060 ppm was recorded and by 2100 UTC (1400 hours
LST), the maximum hourly average O3 concentration of 0.076 ppm was measured. At 0200
UTC on May 7, 1999 (1900 hours LST on May 6), the hourly average concentration declined to
0.059 ppm. Figure AX3-80 shows the O3 vertical profile that was recorded at Boulder, CO on
May 6, 1999, at 1802 UTC (1102 hours LST). The ragged vertical profile of O3 at > 4 km
reflects stratospheric air that has spiraled downward around an upper-level low and mixed with
tropospheric air along the way. Thus, stratospheric air which is normally extremely cold and dry
and rich in O3, loses its characteristics as it mixes downward. This process was described in
Section AX2.3.1 and illustrated in Figures AX2-7a, b, and c.
The dimensions given above imply that individual intrusions are not resolved properly in
the current generation of global-scale CTMs (Figure AX3-80). However, as noted in Section
AX2.3.1, penetration of stratospheric air directly to the planetary boundary layer rarely occurs in
the continental United States. Rather, intrusions are more likely to affect the middle and upper
troposphere, providing a reservoir for O3 that can exchange with the planetary boundary layer.
In this regard, it is important that CTMs be able to spatially and temporally resolve the exchange
between the planetary boundary layer and the lower free troposphere properly.
AX3.9.2 Capability of Global Models to Simulate Tropospheric Ozone
The current generation of global CTMs includes detailed representation of tropospheric
O3-NOX-VOC chemistry. Meteorological information is generally provided by global data
assimilation centers. The horizontal resolution is typically a few hundred km, the vertical
resolution is 0.1 to 1 km, and the effective temporal resolution is a few hours. These models can
simulate most of the observed variability in O3 and related species, although the coarse
AX3-147
-------
-100
Temperature and Frost Point (C)
-80 -60 -40 -20 0
20
40
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
Ozone (ppbv)
Figure AX3-80. Ozone vertical profile at Boulder, Colorado on May 6,1999 at 1802 UTC
(1102 LST).
Source: Lefohnetal. (2001).
resolution precludes simulation of fine-scale structures or localized extreme events. On the
synoptic scale, at least, all evidence indicates that global models are adequate tools to investigate
the factors controlling tropospheric O3. Stratosphere-troposphere exchange of O3 in global
models is generally parameterized. The parameterizations are typically constrained to match the
global mean O3 cross-tropopause flux, which is in turn constrained by a number of observational
proxies (550 ± 140 Tg O3 year1 [Olsen et al., 2002]). The model simulations are routinely
compared to ozonesonde observations in the middle and upper troposphere to test the simulation
of stratospheric influence on tropospheric O3 (Logan, 1999). Such evaluations show that the
parameterized cross-tropopause O3 flux in global models results in a good simulation of
tropospheric O3, at least in a mean sense; and that the current generation of models can
AX3-148
-------
reproduce the tropospheric ozonesonde climatology to within 5 to 10 ppbv, even at mid- and
high-northern latitudes in winter, with the correct seasonal cycle.
Fiore et al. (2003a) used the GEOS-CHEM global tropospheric chemistry model to
quantify PRB O3 concentrations across the United States. A net global O3 flux of 490 Tg O3
year1 from the stratosphere to the troposphere is imposed in the GEOS-CHEM model,
consistent with the range constrained by observations (Olsen et al., 2002). Previous applications
of the model have demonstrated that it simulates the tropospheric ozonesonde climatology
(Logan, 1999) generally to within 5 to 10 ppbv, including at mid- and high-latitudes (Bey et al.,
2001a) over Bermuda in spring (Li et al., 2002) and at sites along the Asian Pacific rim (Liu
et al., 2002). The phase of the seasonal cycle is reproduced to within 1 to 2 months (Bey et al.,
2001a; Li et al., 2002; Liu et al., 2002). An analysis of the 210Pb-7Be-O3 relationships observed
in three aircraft missions over the western Pacific indicates that the model does not
underestimate the stratospheric source of O3 (Liu et al., 2004). These studies and others (Li
et al., 2001; Bey et al., 2001b; Fusco and Logan, 2003) demonstrate that the model provides an
adequate simulation of O3 in the free troposphere at northern midlatitudes, including the mean
influence from the stratosphere. However, it cannot capture the structure and enhancements
associated with stratospheric intrusions, leading to mean O3 under-prediction in regions of
preferred stratospheric downwelling.
Fiore et al. (2002a, 2003b) presented a detailed evaluation of the model simulation for O3
and related species in surface air over the United States for the summer of 1995. They showed
that the model reproduces important features of observations including the high tail of O3
frequency distributions at sites in the eastern United States (although sub-grid-scale local peaks
are underestimated), the O3 to (NOy - NOX) relationships, and that the highest O3 values exhibit
the largest response to decreases in U.S. fossil fuel emissions from 1980 to 1995 (Lefohn et al.,
1998). Empirical orthogonal functions (EOFs) for the observed regional variability of O3 over
the eastern United States are also well reproduced, indicating that GEOS-CHEM captures the
synoptic-scale transport processes modulating surface O3 concentrations (Fiore et al., 2003b).
One model shortcoming relevant for the discussion below is that excessive convective mixing
over the Gulf of Mexico and the Caribbean leads to an overestimate of O3 concentrations in
southerly flow over the southeastern United States. Comparison of GEOS-CHEM with the
Multiscale Air Quality Simulation Platform (MAQSIP) regional air quality modeling system
AX3-149
-------
(Odman and Ingram, 1996) at 36 km2 horizontal resolution showed that the models exhibit
similar skill at capturing the observed variance in O3 concentrations with comparable model
biases (Fiore et al., 2003b).
Simulations to Quantify Background Ozone Over the United States
The sources contributing to the O3 background over the United States were quantified by
Fiore et al. (2003a) with three simulations summarized in Table AX3-16: (1) a standard
simulation, (2) a background simulation in which North American anthropogenic NOX,
NMVOC, and CO emissions are set to zero, and (3) a natural O3 simulation in which global
anthropogenic NOX, NMVOC and CO emissions are set to zero and the CH4 concentration is set
to its 700 ppbv pre-industrial value. Anthropogenic emissions of NOX, nonmethane volatile
organic compounds (NMVOCs), and CO include contributions from fuel use, industry, and
fertilizer application. The difference between the standard and background simulations
represents regional pollution, i.e., the O3 enhancement from North American anthropogenic
emissions. The difference between the background and natural simulations represents
hemispheric pollution, i.e., the O3 enhancement from anthropogenic emissions outside North
America. Methane and NOX contribute most to hemispheric pollution (Fiore et al., 2002b).
A tagged O3 tracer simulation (Fiore et al., 2002a) was used to isolate the stratospheric
contribution to the background and yielded results that were quantitatively consistent with those
from a simulation in which O3 transport from the stratosphere to the troposphere was suppressed
(Fusco and Logan, 2003). All simulations were initialized in June 2000; results are reported for
March through October 2001.
The standard and background simulations were conducted at 2° x 2.5° horizontal
resolution, but the natural simulation was conducted at 4° x 5° resolution to save on
computational time. There was no significant bias between 4° x 5° and 2° x 2.5° simulations
(Fiore et al., 2002a), particularly for a natural O3 simulation where surface concentrations were
controlled by large-scale processes.
AX3.9.3 Mean Background Concentrations: Spatial and Seasonal Variation
The analysis of Fiore et al. (2003a) focused on the 2001 observations from the Clean Air
Status and Trends Network (CASTNet) of rural and remote U.S. sites (Lavery et al., 2002)
AX3-150
-------
Table AX3-16. Description of Simulations Used for Source Attribution
(Fiore et al., 2003a)
Simulation Description Horizontal Resolution
Standard Present-day emissions as described in the text 2° x 2.5°
Background North American anthropogenic NOX, NMVOC, and CO emissions 2° x 2.5°
set to zero
Natural Global anthropogenic NOX, NMVOC, and CO emissions set to zero 4° x 5°
and CH4 concentration set to its 700 ppbv preindustrial value
Stratospheric Tagged O3 tracer originating from the stratosphere in standard 2° x 2.5°
simulation
(Figure AX3-81). Figure AX3-82 shows the mean seasonal cycle in afternoon (1300 to
1700 hours LT) O3 concentrations averaged over the CASTNet stations in each U.S. quadrant.
Measured O3 concentrations (asterisks) are highest in April to May, except in the Northeast
where they peak in June. Model results (triangles) are within 3 ppbv and 5 ppbv of the
observations for all months in the Northwest and Southwest, respectively. Model results for the
Northeast are too high by 5 to 8 ppbv when sampled at the CASTNet sites; the model is lower
when the ensemble of grid squares in the region are sampled (squares). The model is 8 to
12 ppbv too high over the Southeast in summer for reasons discussed in Section AX3.9.2.
Results from the background simulation (no anthropogenic emissions in North America; see
Table AX3-16) are shown as diamonds in Figure AX3-82. Mean afternoon background O3
ranges from 20 ppbv in the Northeast in summer to 35 ppbv in the Northwest in spring. It is
higher in the West than in the East because of higher elevation, deeper mixed layers, and
longer O3 lifetimes due to the arid climate (Fiore et al., 2002a). It is also higher in spring than in
summer, in part because of the seasonal maximum of stratospheric influence (Figure AX3-82)
and in part because of the longer lifetime of O3 (Wang et al., 1998).
Results from the natural O3 simulation (no anthropogenic emissions anywhere;
Table AX3-16) are shown as crosses in Figure AX3-82. Natural O3 concentrations are also
highest in the West and in spring when the influence of stratospheric O3 on the troposphere
peaks (e.g., Holton et al., 1995). Monthly mean natural O3 concentration ranges are 18 to 23,
18 to 27, 13 to 20, and 15 to 21 ppbv in the Northwest, Southwest, Northeast, and Southeast,
AX3-151
-------
X
-x' •
Y^k
X
b.Sx
.- x
„
voy-.;.-.-
.-•• .-•; •.cow v
CADcVL '.• '• -
' ^ A O * *
GAS '-'
'--.-*
Figure AX3-81. CASTNet stations in the continental United States for 2001. Sites
discussed in Section AX3.9.4 are labeled: VOY = Voyageurs NP, MI;
COW = Coweeta, NC; YEL = Yellowstone NP, WY; CAD = Caddo
Valley, AR; CVL = Coffeeville, MS; GAS = Georgia Station, GA;
GB = Great Basin, NV; GRC = Grand Canyon, AZ; CAN = Canyonlands,
UT; MEV = Mesa Verde, CO. Crosses denote sites > .5 km altitude.
Source: Fiore et al. (2003a).
respectively. The stratospheric contribution (X's) ranges from 7 ppbv in spring to 2 ppbv in
summer.
The difference between the background and natural simulations in Figure AX3-82
represents the increases in monthly mean hemispheric pollution. This increase ranges from 5 to
12 ppbv depending on region and season. It peaks in spring due to a longer O3 lifetime (Wang
et al., 1998) and to e efficient ventilation of pollution from the Asian continent (Liu et al., 2003).
In contrast to hemispheric pollution, the regional pollution influence (O3 produced from North
American anthropogenic emissions, shown as the difference between the squares and diamonds)
peaks in summer and is highest in the East. For the data in Figure AX3-82, it ranges from
8 ppbv in the northern quadrants in March to over 30 ppbv in the eastern quadrants in summer.
Monthly mean observed O3 concentrations are influenced by both regional and hemispheric
pollution in all U.S. regions from March through October.
AX3-152
-------
80
60
40
•£> 20
.0
Q.
Q. 0
C 80
O
N
° 60
40
20
NW
sw
J FMAMJ JASOND J FMAMJ JASOND
Month of 2001
Figure AX3-82.
Monthly mean afternoon (1300 to 1700 hours LT) concentrations (ppbv)
in surface air averaged over the CASTNet stations (Figure AX3-81) in
each U.S. quadrant for March to October 2001. Observations (asterisks)
are compared with model values from the standard simulation sampled
at the CASTNet sites (triangles) and sampled for the entire quadrant
(squares). The vertical lines show the standard deviation in the observed
and simulated values. Monthly mean model results for the background
(diamonds), natural (crosses), and stratospheric (X's) contributions
(Table AX3-16) to surface O3 are shown. The U.S. quadrants are
centered at 98.75° W and 37° N.
Source: Fiore et al. (2003a).
AX3.9.4 Frequency of High-Ozone Occurrences at Remote Sites
Lefohn et al. (2001) pointed out the frequent occurrence of high-O3 events (>50 and
60 ppbv) at remote northern U.S. sites in spring. Fiore et al. (2003a) replicated the analysis of
Lefohn et al. (2001) at the four CASTNet sites that they examined: Denali National Park
(Alaska), Voyageurs National Park (Minnesota), Glacier National Park (Montana), and
Yellowstone National Park (Wyoming). The number of times that the hourly O3 observations at
AX3-153
-------
the sites are >50 and 60 ppbv for each month from March to October 2001 were then calculated
(see results in Table AX3-17) and compared with the same statistics for March to June
1988 to 1998 from Lefohn et al. (2001), to place the 2001 statistics in the context of other years.
More incidences of O3 above both thresholds occur at Denali National Park and Yellowstone
National Park in 2001 than in nearly all of the years analyzed by Lefohn et al. (2001). The
statistics at Glacier National Park, Montana indicate that 2001 had fewer than average incidences
of high-O3 events. At Voyageurs National Park in Minnesota, March and April 2001 had
lower-than-average frequencies of high-O3 events, but May and June were more typical.
Overall, 2001 was considered to be a suitable year for analysis of high-O3 events. Ozone
concentrations >70 and 80 ppbv occurred most often in May through August in 2001 and were
found to be associated with regional pollution by Fiore et al. (2003a).
Fiore et al. (2003a) focused their analysis on mean O3 concentrations during the afternoon
hours (1300 to 1700 LT), as the comparison of model results with surface observations is most
appropriate in the afternoon when the observations are representative of a relatively deep mixed
layer (Fiore et al., 2002a). In addition, the GEOS-CFffiM model does not provide independent
information on an hour-to-hour basis, because it is driven by meteorological fields that are
updated every 6-h and then interpolated. Fiore et al. (2003a) tested whether an analysis
restricted to these mean 1300 to 1700 LT surface concentrations captures the same frequency
of O3 >50 and 60 ppbv that emerges from an analysis of the individual hourly concentrations
over 24 hours. Results are reproduced here in Table AX3-17, which shows that the percentage
of individual afternoon (1300 to 1700 LT) hours when O3 >50 and 60 ppbv at the CASTNet sites
is always greater than the percentage of all hourly occurrences above these thresholds, indicating
that elevated O3 concentrations preferentially occur in the afternoon. Furthermore,
Table AX3-17 shows that the frequency of observation of high-O3 events is not diminished when
4-h average (1300 to 1700 LT) concentrations are considered, reflecting persistence in the
duration of these events. Model frequencies of high-O3 events from 1300 to 1700 LT at the
CASTNet sites are similar to observations in spring, as shown in Table AX3-17, and about 10%
higher in the summer, largely because of the positive model bias in the Southeast discussed in
Section AX3.9.2.
AX3-154
-------
Table AX3-17. Number of Hours with Ozone Above 50 or 60 ppbv
at U.S. CASTNet Sites in 2001
Site
Mar
Apr
May
Observations ^
Denali NP, Alaska
(64° N, 149° W, 0.6 km)
Voyageurs NP, Minnesota
(48° N, 93° W, 0.4km)
Glacier NP, Montana
(49° N, 114° W, 1.0km)
Yellowstone NP, Wyoming
(45° N, 110° W, 2.5km)
All CASTNet sites (71)
All sites, 1300-1700 LT
only (hourly data)
All sites, 1300-1700 LT
mean (4-hour average)
All sites, model
1300-1700 LT mean
24
0
4
307
5468
(11%)
1817
(21%)
435
(20%)
254
(12%)
302
0
0
461
15814
(32%)
4684
(56%)
1153
(55%)
1249
(59%)
98
62
23
350
17704
(36%)
5174
(61%)
1295
(61%)
1527
(69%)
Observations ^
Denali NP
Voyageurs NP
Glacier NP
Yellowstone NP
All sites
All sites, 1300-1700 LT
only
All sites,
1300-1700 LT mean
All sites, model
1300-1700 LT mean
0
0
0
4
519
(1%)
235
(3%)
56
(3%)
13
(1%)
0
0
0
77
4729
(10%)
1798
(22%)
428
(20%)
377
(18%)
9
6
0
139
8181
(16%)
2808
(33%)
697
(33%)
834
(38%)
Jun
50 ppbv
6
95
0
172
16150
(33%)
4624
(56%)
1147
(55%)
1505
(71%)
60 ppbv
2
32
0
18
8199
(17%)
2721
(33%)
671
(32%)
964
(45%)
Jul
0
14
6
140
14489
(29%)
4613
(54%)
1161
(54%)
1475
(67%)
0
0
0
6
5705
(11%)
2235
(26%)
550
(26%)
910
(41%)
Aug
0
17
12
261
15989
(32%)
5075
(60%)
1283
(60%)
1500
(68%)
0
0
0
26
7407
(15%)
2758
(33%)
677
(32%)
834
(38%)
Sep
0
33
0
173
9874
(20%)
3343
(40%)
841
(40%)
1080
(51%)
0
15
0
1
3492
(7%)
1416
(17%)
358
(17%)
502
(24%)
Oct
0
0
0
77
5642
(11%)
1945
(23%)
478
(22%)
591
(27%)
0
0
0
2
2073
(4%)
878
(10%)
214
(10%)
204
(9%)
Data from 71 U.S. CASTNet sites are included in this analysis: those in Figure AX3-105 plus Denali NP.
Percentages of total occurrences are shown in parentheses.
NP = National Park; LT = Local Time.
Reproduced from Fiore et al. (2003a).
AX3-155
-------
NATURAL VERSUS ANTHROPOGENIC CONTRIBUTIONS TO
HIGH-OZONE OCCURRENCES
Figure AX3-83, reproduced from Fiore et al. (2003a), shows probability distributions of
daily mean afternoon (1300 to 1700 LT) O3 concentrations in surface air at the CASTNet sites
for March through October 2001. Model distributions for background, natural, and stratospheric
O3 (Table AX3-16) are also shown. The background (long-dashed line) ranges from 10 to
50 ppbv with most values in the 20 to 35 ppbv range. The full 10 to 50 ppbv range of
background predicted here encompasses the previous 25 to 45 ppbv estimates shown in
Table 3-8. However, background estimates from observations tend to be at the higher end of the
range (25 to 45 ppbv), while these results, as well as those from prior modeling studies
(Table 3-8) indicate that background O3 concentrations in surface air are usually below 40 ppbv.
The background O3 concentrations derived from observations may be overestimated if
observations at remote and rural sites contain some influence from regional pollution (as shown
below to occur in the model), or if the O3 versus NOy - NOX correlation is affected by different
relative removal rates of O3 and NOy (Trainer et al., 1993). Natural O3 concentrations
(short-dashed line) are generally in the 10 to 25 ppbv range and never exceed 40 ppbv. The
range of the hemispheric pollution enhancement (the difference between the background and
natural O3 concentrations) is typically 4 to 12 ppbv and only rarely exceeds 20 ppbv (< 1% total
incidences). The stratospheric contribution (dotted line) is always less than 20 ppbv and usually
below 10 ppbv. Time series for specific sites are presented below.
CASE STUDIES: INFLUENCE OF THE BACKGROUND ON ELEVATED OZONE
EVENTS IN SPRING
High-O3 events were previously attributed to natural processes by Lefohn et al. (2001) at:
Voyageurs National Park, Minnesota in June and Yellowstone National Park, Wyoming in
March through May. Fiore et al. (2003a) used observations from CASTNet stations in
conjunction with GEOS-CHEM model simulations to deconstruct the observed concentrations
into anthropogenic and natural contributions.
At Voyageurs National Park in 2001, O3 concentrations >60 ppbv occurred frequently in
June but rarely later in summer (Table AX3-17). A similar pattern was observed in 1995 and
1997 and was used to argue that photochemical activity was probably not responsible for these
events (Lefohn et al. 2001). Figure AX3-84 from Fiore et al. (2003a) shows that GEOS-CFffiM
AX3-156
-------
0.10n
Figure AX3-83.
.D
Q.
-F 0.05-
ro
.a
o
0 10 20 30 40 50 60 70 80 90
Ozone (ppbv)
Probability distributions of daily mean afternoon (1300 to 1700 LT) O3
concentrations in surface air for March through October 2001 at U.S.
CASTNet sites (Figure AX3-83): observations (thick solid line) are
compared with model results (thin solid line). Additional probability
distributions are shown for the simulated background (long-dashed line),
natural (short-dashed line), and stratospheric (dotted line) contributions
to surface O3 (Table AX3-23).
Source: Fiore et al. (2003a).
ja
OL
o
c
O
N
O
80
60
40
20
Voyageurs HP, Minnesota (93W, 48N)
10
20
30
May
19
June
29
Figure AX3-84.
Daily mean afternoon (13 to 17 LT) O3 concentrations in surface air at
Voyageurs National Park (NP), Minnesota in mid-May through June of
2001. Observations (asterisks) are compared with model values from the
standard simulation (triangles). The simulated contributions from
background (diamonds), natural (crosses), and stratospheric (X's) O3
are also shown.
Source: Fiore et al. (2003a).
AX3-157
-------
Figure AX3-84 shows that regional pollution drives most of the simulated day-to-day variability
and explains all events above 50 captures much of the day-to-day variability in observed
concentrations from mid-May through June, including the occurrence and magnitude of high-O3
events. The simulated background contribution (diamonds) ranges from 15 to 36 ppbv with a 25
ppbv mean. The natural O3 level (crosses) is 15 ppbv on average and varies from 9 to 23 ppbv.
The stratospheric contribution (X's) is always < 7 ppbv. The dominant contribution to the
high-O3 events on June 26 and 29 is from regional pollution (44 and 50 ppbv on June 26 and 29,
respectively, calculated as the difference between the triangles and diamonds in Figure AX3-84).
The background contribution (diamonds) is < 30 ppbv on both days, and is composed of a
20 ppbv natural contribution (which includes 2 ppbv of stratospheric origin) and a 5 ppbv
enhancement from hemispheric pollution (the difference between the diamonds and crosses).
Beyond these two high-O3 events, ppbv. In 2001, monthly mean observed and simulated O3
concentrations are lower in July (37 and 42 ppbv, respectively) and August (35 and 36 ppbv)
than in June (44 and 45 ppbv). Fiore et al. (2003a) hypothesized that the lower mean O3 and the
lack of O3 >60 ppbv in July and August reflects a stronger Bermuda high-pressure system
sweeping pollution from southern regions eastward before it could reach Voyageurs National
Park.
Frequently observed concentrations of O3 between 60 to 80 ppbv at Yellowstone NP in
spring (Figures AX3-76a,b) have been attributed by Lefohn et al. (2001) to natural sources,
because they occur before local park traffic starts and back-trajectories do not suggest influence
from long-range transport of anthropogenic sources. More hours with O3 >60 ppbv occur in
April and May of 2001 (Table AX3-17) than in the years analyzed by Lefohn et al. (2001). Fiore
et al. (2003a) used GEOS-CFffiM to interpret these events; results are shown in Figure AX3-85.
The mean background, natural, and stratospheric O3 contributions in March to May are higher
at Yellowstone (38, 22, and 8 ppbv, respectively) as compared to 27, 18, and 5 ppbv at the two
eastern sites previously discussed. The larger stratospheric contribution at Yellowstone reflects
the high elevation of the site (2.5 km). Fiore et al. (2003a) argued that the background at
Yellowstone National Park should be considered an upper limit for U.S. PJAB O3 concentrations,
because of its high elevation. While Yellowstone receives a higher background concentration
than the eastern sites, the model shows that regional pollution from North American
anthropogenic emissions (difference between the triangles and diamonds) contributes an
AX3-158
-------
80
"> 60
.Q
Q.
a
7 40
o
O 20
0
Yellowstone NP, Wyoming (11OW, 45N)
-i
20
March
29
April
19
May
Figure AX3-85. Same as Figure AX3-84 but for Yellowstone National Park, Wyoming in
March to May 2001. Observations (asterisks) are compared with model
values from the standard simulation (triangles). The simulated
contributions from background (diamonds), natural (crosses), and
stratospheric (X's) O3 are also shown.
Source: Fiore et al. (2003a).
additional 10 to 20 ppbv to the highest observed concentrations in April and May. One should
not assume that regional photochemistry is inactive in spring.
Higher-altitude western sites are more frequent recipients of subsidence events that
transport high concentrations of O3 from the free troposphere to the surface. Cooper and Moody
(2000) cautioned that observations from elevated sites are not generally representative of
lower-altitude sites. At Yellowstone, the background O3 rarely exceeds 40 ppbv, but it is even
lower in the East. This point is illustrated in Figure AX3-86, from Fiore et al. (2003a), with time
series at representative western and southeastern CASTNet sites for the month of March, when
the relative contribution of the background should be high. At the western sites, the background
is often near 40 ppbv but total surface O3 concentrations are rarely above 60 ppbv. While
variations in the background play a role in governing the observed total O3 variability at these
sites, regional pollution also contributes. Background concentrations are lower (often <30 ppbv)
in the southeastern states where regional photochemical production drives much of the observed
variability. Cooper and Moody (2000) have previously shown that the high O3 concentrations at
an elevated, regionally representative site in the eastern United States in spring coincide with
AX3-159
-------
&
Q.
O
80
60
40
20
0
I
80
60
40
20
80
60
40
20
0
I
80
60
40
20
0
Great Basin NP, Nevada (114W.39N)
0 10 20
Grand Canyon NP, Arizona (112W.36N)
10 20
Canyonlands NP, Utah (110W.38N)
0 10 20
Mesa Verde NP, Colorado (108W,37N)
10
20
80
60
40
20
0
I
80
60
40
20
0
I
80
60
40
20
0
I
80
60
40
20
0
Caddo Valley NP, Arkansas (93W,34N)
0 10 20
Coffeeville, Mississippi (90W,34N)
0 10 20
Georgia Station, Georgia (84W,33N)
0 10 20
Coweeta, North Carolina (83W,35N)
10
20
March 2001
Figure AX3-86. Same as Figure AX3-84 but for March of 2001 at selected western
(left column) and southeastern (right column) sites. Observations
(asterisks) are compared with model values from the standard simulation
(triangles). The simulated contributions from background (diamonds),
natural (crosses), and stratospheric (X's) O3 are also shown.
Source: Fiore et al. (2003a).
AX3-160
-------
high temperatures and anticyclonic circulation, conditions conducive to photochemical O3
production. Peak O3 concentrations in this region, mainly at lower elevations, are associated
with lower background concentrations because chemical and depositional loss during stagnant
meteorological conditions suppress mixing between the boundary layer and the free troposphere
(Fiore et al., 2002a). Surface O3 concentrations >80 ppbv could conceivably occur when
stratospheric intrusions reach the surface. However, based on information given in
Section AX2.3.2, these events are rare.
AX3.10 OZONE EXPOSURE IN VARIOUS MICROENVIRONMENTS
AX3.10.1 Introduction
There are many definitions of exposure. Human exposure to O3 and related photochemical
oxidants are based on the measured O3 concentrations in the individual's breathing zone as the
individual moves through time and space. Epidemiological studies generally use the ambient
concentrations as surrogates for exposure. Therefore, human exposure data and models provide
the best link between ambient concentrations (from measurements at monitoring sites or
estimated with atmospheric transport models), lung deposition and clearance, and estimates of
air concentration-exposure-dose relationships.
This section discusses the current information on the available human exposure data and
exposure model development. This includes information on (a) the relationships between O3
measured at ambient monitoring sites and personal exposures and (b) factors that affect these
relationships. The information presented in this section is intended to provide critical links
between ambient monitoring data and O3 dosimetry as well as between the toxicological and
epidemiologic studies presented in Annexes AX4, AX5, AX6, and AX7 of this document.
AX3.10.2 Summary of the Information Presented in the Exposure
Discussion in the 1996 Ozone Criteria Document
The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a), based on then
currently available information, indicated that less emphasis should be placed on O3
concentrations measured at ambient monitoring stations. Fixed monitoring stations are generally
used for monitoring associated with air quality standards and do not provide a realistic
AX3-161
-------
representation of individual exposures. Indoor/outdoor O3 ratios reported in the literature were
summarized for residences, hospitals, offices, art galleries, and museums. The differences in
residential I/O were found to be a function of ventilation conditions. The I/O ratios were less
than unity. In most cases, indoor and in-transit concentrations of O3 were significantly different
from ambient O3 concentrations. Ambient O3 varied from O3 concentrations measured at fixed-
site monitors. Very limited personal exposure measurements were available at the time the
1996 O3 AQCD was published, so estimates of O3 exposure or evaluated models were not
provided. The two available personal exposure studies indicated that only 40% of the variability
in personal exposures was explained by the exposure models using time-weighted indoor and
outdoor concentrations. The discussion addressing O3 exposure modeling primarily addressed
work reported by McCurdy (1994) on population-based models (PBMs). Literature published
since publication of the 1996 O3 AQCD has also focused on PBMs. A discussion of individual-
based models (IBMs) will be included in the description of exposure modeling in this document
to improve our mechanistic understanding of O3 source-to-exposure events and to evaluate their
usefulness in providing population-based estimates.
AX3.10.3 Concepts of Human Exposure
Human exposure to O3 and related photochemical oxidants occurs when individuals come
in contact with the pollutant through "(a) the visible exterior of the person (skin and openings
into the body such as mouth and nostrils) or (b) the so-called exchange boundaries where
absorption takes place (skin, mouth, nostrils, lung, gastrointestinal tract)" (Federal Register,
1986). Consequently, exposure to a chemical, in this case O3, is the contact of that chemical
with the exchange boundary (U.S. Environmental Protection Agency, 1992). Therefore,
inhalation exposure to O3 is based on measurements of the O3 concentration near the individual's
breathing zone that is not affected by exhaled air.
AX3.10.4 Quantification of Exposure
Quantification of inhalation exposure to any air pollutant starts with the concept of the
variation in the concentration of the air pollutant in the breathing zone, unperturbed by exhaled
breath, as measured by a personal exposure monitor as a person moves through time and space.
Since the concentrations of O3 and related photochemical oxidants vary with time and location
AX3-162
-------
and since people move among locations and activities, the exposure and dose received changes
during the day. Furthermore, the amount of pollutant delivered to the lung is dependent upon the
person's minute ventilation rate. Thus, the level of exertion is an important consideration in
determining the potential exposure and dose. Inhalation exposure has been defined as the
integral of the concentration as a function of time over the time period of interest for each
individual (Ott, 1982, 1985; Lioy, 1990):
t2
E =1 c(t}dt (AX3-2)
where E is inhalation exposure, eft) is the breathing zone concentration as a function of time and
tj and t2 the starting and ending time of the exposure, respectively.
AX3.10.5 Methods to Estimate Personal Exposure
There are two approaches for measuring personal exposure; direct and indirect methods
(Ott, 1982, 1985; Navidi et al., 1999). Direct approaches measure the contact of the person with
the chemical concentration in the exposure media over an identified period of time. For the
direct measurement method, a personal exposure monitor (PEM) is worn near the breathing zone
for a specified time to either continually collect for subsequent analysis or directly measure the
concentrations of the pollutant and the exposure levels. The indirect approach models
concentrations of a pollutant in specific microenvironments. Both methods are subject to
measurement error.
AX3.10.5.1 Direct Measurement Method
The passive monitors commonly used in the direct method provides integrated personal
exposure information. The monitor's sensitivity to wind velocity, badge placement, and
interference with other copollutants may result in measurement error.
Modified passive samplers have been developed for use in determining O3 exposure. The
difficulty in developing a passive O3 monitor is in identifying a chemical or trapping reagent that
can react with O3. Zhou and Smith (1997) evaluated the effectiveness of sodium nitrite,
3-methyl-2-benzothiazolinone acetone azine (MBTH), />-acetamidophenol (p-ATP), and indigo
AX3-163
-------
carmine as O3-trapping reagents. Only sodium nitrite and MBTH gave sensitive, linear
responses at environmentally relevant concentrations. However, MBTH overestimated the O3
concentrations significantly, suggesting an interference effect. Sodium nitrite was found to be a
valid reagent when an effective diffusion barrier was used. Scheeren and Adema (1996) used an
indigo carmine-coated glass-fiber filter to collect spectrophotometrically measured O3. The
detection limit was 23 ppb for a 1-h exposure, with no interfering oxidants identified. The
reagent was valid for a relative humidity range of 20 to 80%. The uptake rate was wind velocity
dependent. However, wind velocity dependencies was compensated for by using a small
battery-operated fan that continuously blew air across the face of the monitor at a speed of
1.3 m/s. The overall accuracy of the sampler, after correcting for samples collected under low-
wind conditions, was 11 ± 9% in comparison to a continuous UV-photometric monitor. Sample
stability was > 25 days in a freezer. Bernard et al. (1999) employed a passive sampler consisting
of a glass-fiber filter coated with a l,2-di(4-pyridyl)ethylene solution. The sample was analyzed
spectrophotometrically after color development by the addition of 3-methyl-2-benzothiazolinone
hydrazone hydrochloride. The sampler was used at 48 sites in Montpellier, France. The
correlation coefficient was r = 0.9, p < 0.0001. Detection limits were 17 ppb for 12-h and 8 ppb
for 24-h samples with an overall variation coefficient of 5% for field-tested paired samples. The
imprecision was estimated to be 1.0 ppb.
A series of studies have been conducted using a passive sampler developed by Koutrakis
et al. (1993) at the Harvard School of Public Health. The sampler used sodium nitrate as the
trapping reagent and included a small fan to assure sufficient movement of air across the face of
the badge when sampling was done indoors. The passive sampler has been evaluated against the
standard UV absorption technique used in studies in southern California (Avol et al., 1998a;
Geyh et al., 1999, 2000; Delfmo et al., 1996), Baltimore, MD (Sarnat et al., 2000), and Canada
(Brauer and Brook, 1997).
In a study discussed later in this chapter, Avol et al. (1998a) used nitrite-coated passive
samplers to measure O3 air concentrations indoors and outdoors of 126 homes between February
and December 1994 in the Los Angeles metropolitan area. The detection limit of the method
was near 5 ppb. The inconsistent sampler response due to changes in wind pattern and changes
in personal activity made the sampler unacceptable for widespread use.
AX3-164
-------
Geyh et al. (1997, 1999) compared passive and active personal O3 air samplers based on
nitrite-coated glass-fiber filters. The active sampler was more sensitive allowing for the
collection of short-term, 2.6-h samples. Comparison between the two samplers and UV
photometric O3 monitors demonstrated generally good agreement (bias for active personal
sampler of-6%). The personal sampler also had high precision (4% for duplicate analyses) and
good compliance when used by children attending summer day camp in Riverside, CA.
AX3.10.5.2 Indirect Measurement Method
The indirect method determines and measures the concentrations in all of the locations or
"microenvironments" a person encounters or determines the exposure levels through the use of
models or biomarkers. The concept of microenvironments is critical for understanding human
exposure and aids in the development of procedures for exposure modeling using data from
stationary monitors (indoor and outdoor). Microenvironments were initially defined as
individual or aggregate locations (and sometimes even as activities taking place within a
location) where a homogeneous concentration of the pollutant is encountered for a specified
period of time. Thus, a microenvironment has often been identified with an "ideal" (i.e.,
perfectly mixed) compartment of classical compartmental modeling. More recent and general
definitions view the microenvironment as a "control volume," indoors or outdoors, that can be
fully characterized by a set of either mechanistic or phenomenological governing equations,
when properly parameterized, given appropriate initial and boundary conditions. The boundary
conditions typically reflect interactions with the ambient air and with other microenvironments.
The parameterizations of the governing equations generally include the information on attributes
of sources and sinks within each microenvironment. This type of general definition allows for
the concentration within a microenvironment to be nonhomogeneous, provided its spatial profile
and mixing properties can be fully predicted or characterized. By adopting this definition, the
number of microenvironments used in a study is kept manageable, while existing variabilities in
concentrations are still taken into account. The "control volume" variation could result in a
series of microenvironments in the same location. If there are large spatial gradients within a
location for the same time period, the space should be divided into the number of
microenvironments needed to yield constant pollutant concentrations; the alternative offered by
the control volume approach is to provide concentration as a function of location within it,
AX3-165
-------
so that the appropriate value is selected for calculating exposure. Thus, exposure to a person in a
microenvironment is calculated using a formula analogous to equation AX3-3, but as the sum of
the discrete products of measured or modeled concentrations (specific to the receptor and/or
activity of concern) in each microenvironment by the time spent there. The equation is
expressed as:
(AX3-3)
/=!
where / specifies microenvironments from 1 to n, ci is the concentration in the rth
microenvironment, and A ti is the duration spent in the rth microenvironment. The total exposure
for any time interval for an individual is the sum of the exposures in all microenvironments
encountered within that time interval. The concentration and time component in this approach
can contribute to measurement error. However, this method should provide an accurate
determination of exposure provided that all microenvironments that contribute significantly to
the total exposures are included and the concentration assigned to the microenvironment is
appropriate for the time period spent in those environments. Results from the error analysis
models developed by Navidi et al. (1999) suggested that neither the microenvironmental nor
personal sampling approach would give reliable health effect estimates when measurement
errors were uncorrected. Using their error analysis models, nondifferential measurement error
biased the effect estimates toward zero under the model assumptions. When the measurement
error is correlated with the health response, a bias away from the null could result. These
researchers proposed the use of bidirectional crossover and multi-level analytic statistical
methods for estimating acute effects of environmental exposures.
Microenvironments typically used to determine O3 exposures include indoor residences,
other indoor locations, outdoors near roadways, other outdoor locations, and in-vehicles.
Outdoor locations near roadways are segregated from other outdoor locations because N2O
emissions from automobiles alter O3 and related photochemical oxidant concentrations compared
to concurrent typical ambient levels. Indoor residences are typically separated from other indoor
locations, because of the time spent there and potential differences between the residential
environment and the work/public environment. A special concern for O3 and related
AX3-166
-------
photochemical oxidants is their diurnal weekly (weekday-weekend) and seasonal variability.
Few indoor O3 sources exist, but include electronic equipment, O3 generators, and copying
machines. Some secondary reactions of O3 take place indoors that produce related
photochemical oxidants that could extend the exposures to those species above the estimates
obtained from O3 alone. (See discussion on O3 chemistry and indoor sources and concentrations
later in this Annex.)
AX3.10.6 Ozone Exposure Models
Measurement efforts to assess population exposures or exposures of large numbers of
individuals over long time periods is labor intensive and costly, so exposure modeling is often
done for large populations evaluated over time. Predicting (or reconstructing) human exposure
to O3 through mechanistic models is complicated by the fact that O3 (and associated
photochemical oxidants) is formed in the atmosphere through a series of chemical reactions that
are nonlinear and have a wide range of characteristic reaction timescales. Furthermore, these
reactions require the precursors VOCs and NOX that are emitted by a wide variety of both
anthropogenic and natural (biogenic) emission sources. This makes O3 a secondary pollutant
with complex nonlinear and multiscale dynamics in time and space. Concentration levels
experienced by individuals and populations exposed to O3 are therefore affected by (1) emission
levels and spatiotemporal patterns of the gaseous precursors: VOCs and NOX, that can be due to
sources as diverse as a power plant in a different state, automobiles on a highway five miles
away, and the gas stove in one's own kitchen; (2) ambient atmospheric as well as indoor
microenvironmental transport, removal and mixing processes (convective, advective, dispersive
and molecular/diffusional); and (3) chemical transformations that take place over a multitude of
spatial scales, ranging from regional/sub-continental (100 to 1000 km), to urban (10 to 100 km),
to local (1 to 10 km), to neighborhood (< 1 km), and to microenvironmental/personal. These
transformations depend on the presence of co-occurring pollutants in gas and aerosol phases,
both primary and secondary, and on the nature of surfaces interacting with the pollutants.
Further, the strong temporal variability of O3, both diurnal and seasonal, makes it critical
that definitions of integrated or time-averaged exposure employ appropriate averaging times in
order to produce scientifically defensible analyses for either causes of O3 production or health
effects that result from O3 exposure. An understanding of the effect of temporal profiles of
AX3-167
-------
concentrations and contacts with human receptors is essential. Short-term integrated metrics,
such as hourly averages, 8-h running averages, etc., are needed to understand the relationship
between O3 exposure and observed health and other effects.
Health effects associated with O3 have mostly been considered effects of acute exposures.
Peak O3 and related photochemical oxidants concentrations typically occur towards later in the
day during the summer months. Elevated concentrations can last for several hours.
Furthermore, O3 participates in multiphase (gas/aerosol) chemical reactions in various
microenvironments. Several recent studies show that O3 reacts indoors with VOCs and NOX in
an analogous fashion to that occurring in the ambient atmosphere (Lee and Hogsett, 1999;
Wainman et al., 2000; Weschler and Shields, 1997). These reactions produce secondary
oxidants and other air toxics that could play a significant role in cumulative human exposure and
health-related effects within the microenvironment.
Terminology
Models of human exposure to O3 can be characterized and differentiated based upon a
variety of attributes. For example, exposure models can be classified as (1) potential exposure
models, typically maximum outdoor concentration versus "actual" exposure, including locally
modified microenvironmental outdoor and indoor exposures; (2) population versus "specific
individual"-based exposure models; (3) deterministic versus probabilistic models; and
(4) observation versus mechanistic air quality model-driven estimates of spatial and temporal
correlations.
Some points should be made regarding terminology and the directions of exposure
modeling research (as related specifically to O3 exposure assessments) before proceeding to
discuss specific recent activities and developments. First, it must be understood that significant
variation exists in the definitions for much of the terminology used in the published literature.
The science of exposure modeling is an evolving field and the development of a "standard" and
commonly accepted terminology is a process in evolution. Second, very often procedures/efforts
listed in the scientific literature as "exposure models/exposure estimates," may in fact address
only a subset of the steps or components required for a complete exposure assessment. For
example, some efforts focus solely on refining the subregional or local spatiotemporal dynamics
of local O3 concentrations starting from "raw" data representing monitor observations or regional
AX3-168
-------
grid-based model estimates. It is recognized that ambient air concentrations are used as
surrogates for exposure in some epidemiological studies. Nevertheless, such efforts are included
in the discussion of the next subsection, as they can provide essential components of a complete
exposure assessment. On the other hand, formulations that are identified as exposure models,
but focus only on ambient air quality predictions, are not included in the discussion that follows,
as they do not provide true exposure estimates. These models are reviewed in an earlier section
of this annex. Third, O3-exposure modeling is very often identified explicitly with population-
based modeling, while models describing the specific mechanisms affecting the exposure of an
individual to O3, and possibly some of the co-occurring gas and/or aerosol phase pollutants, are
usually associated with studies focusing on indoor chemistry modeling. Finally, in recent years,
the focus of either individual- or population-based exposure modeling research has shifted from
O3 to other pollutants, mostly airborne toxics and particulate matter. However, many of the
modeling components that have been developed in these efforts are directly applicable to O3
exposure modeling and are, therefore, mentioned in the following discussion.
A General Framework for Assessing Exposure to Ozone
First, the individual and relevant activity locations for Individual Based Modeling, or the
population and associated spatial (geographical) domain for Population Based Modeling are
defined. The temporal framework of the analysis (period, resolution) and the comprehensive
modeling of individual/population exposure to O3 (and related pollutants) generally require
several steps (or components, as some of them do not have to be performed in sequence). The
steps represent a "composite" outline based on frameworks described in the literature over the
last 20 years (Ott, 1982, 1985; Lioy, 1990; Georgopoulos and Lioy, 1994; U.S. Environmental
Protection Agency, 1992, 1997) as well as on the structure of various existing inhalation
exposure models (McCurdy, 1994; Johnson et al., 1992; Nagda et al., 1987; U.S. Environmental
Protection Agency, 1996c; ICF Consulting, 2003; Burke et al., 2001; McCurdy et al., 2000;
Georgopoulos et al., 2002a,b; Freijer et al., 1998; Clench-Aas et al., 1999; Kunzli et al., 1997).
The conceptional frameworks of the models are similar. Figures AX3-87a,b provides a
conceptual overview of an exposure model. The steps involved in defining exposure models
include (1) estimation of the background or ambient levels of O3 through space-time analysis of
fixed monitor data, or emissions-based, photochemical, air quality modeling; (2) estimation of
AX3-169
-------
X
Sector/Location data I
(latitude/longitude) 1
^ \
Site-specific
air quality and
temperature data
i
r
Defined Study Area
(sectors within a city
radiusand with air and
temperature data)
Step 2
Generate N number
of simulated
individuals ,
' Population data /
, (age/gender/race) I
^ r
Population within
a study area
Age-specific physiological
distribution data (body
weight, height, etc)
Distribution functions for
profile variables
(e.g, probability of air
conditioner)
^^_
^^
[Com muting flow data/ \
(origin/destination
sectors) \J
^^-^Stoct
r
iastic^~~^^^
^^^^ profile generator^ —
Age group specific
employment
probabilities
Distribution functions
for daily varying profile
variables (e.g., window
status, car speed)
/A simulated individual with \
the following profile:
• Home sector
• Work sector
• Age
• Gender
• Face
• Employment status
• Gas stove
• Pilot gas light
• Air conditioner
• Height
• Weight
• Amount of hemoglobin
• Lung diffusivity
• Endogenous CO production rate
V • Plus 12 more variables J
f r \ - National
(^ ~^) -Simulation <^
step —
- Area-speciTic
J> - Data processor
| -intermediate step
^^^]J - Output data
Figure AX3-87a. Detailed diagram illustrating components of an exposure model.
Source: ICF Consulting and ManTech Environmental Technology, Inc. (2003)
-------
X
/ Diary evenfs/activlties /
and personal information
I (e.g., from CHAD) I
Activity diary pools by day
type/temperature category
The selected diary records
for each day in the
simulation period
Step 5
Determine exposure
for each simulated
individual
aleulate concentrator
in microenvironments
for each simulated
individual
Identify
concentrations in
microenviroments
visited
Microerwironiments denned by
grouping of CHAD location codes
Select calculation methods for each
microenvironment:
» Factors
* Mass balance
Hourly ambient air quality data for
the home and work sectors
Each day in the simulation
period, is assigned with
activity pool # based on
daily max/mean temperature
A sequence of events
(i.e., microenvironments
visited and minutes spent)
in the simulation period
Hourly concentrations in ail
microenvironments during
simulation period
Hourly concentrations and
minutes spent in each
microenviroment visited by
the simulated individual
Maximum/mean daily
temperature data
J Physiological parameters
from profile
Average exposure
concentrations
for simulated person;
* Hourly
* Daily
» Monthly
Annual
Average does for
simulated person;
* Hourly
* Daily
* Monthly
Annual
Figure AX3-87b. Detailed diagram illustrating components of an exposure model.
Source: ICF Consulting and ManTech Environmental Technology, Inc. (2003)
-------
levels and temporal profiles of O3 in various outdoor and indoor microenvironments such as
street canyons, residences, offices, restaurants, vehicles, etc. through linear regression of
available observational data sets, simple mass balance models, detailed (nonlinear) gas or
gas/aerosol chemistry models, or detailed combined chemistry and computational fluid dynamics
models; (3) characterization of relevant attributes of individuals or populations under study (age,
gender, weight, occupation, etc.); (4) development of activity event (or exposure event)
sequences for each member of the sample population or for each cohort for the exposure period;
(5) calculation of appropriate inhalation (in general intake) rates for the individuals of concern,
or the members of the sample population, reflecting/combining the physiological attributes of the
study subjects and the activities pursued during the individual exposure events; (6) combination
of intake rates and microenvironmental concentrations for each activity event to assess dose;
(7) calculation of event-specific exposure and intake dose distributions for selected time periods
(1-h and 8-h daily maximum, O3 season averages, etc.); and (8) use of PBM to extrapolate
population sample (or cohort) exposures and doses to the entire populations of interest. This
process should aim to quantify, to the maximum extent feasible, variability and uncertainty in
the various components, assessing their effects on the estimates of exposure.
Implementation of the above components of comprehensive exposure modeling has
benefitted significantly from recent advances and expanded availability of computational
technologies such as Relational Database Management Systems (RDBMS) and Geographic
Information Systems (Purushothaman and Georgopoulos, 1997, 1999a,b).
AX3.10.6.1 Population Exposure Models
Existing comprehensive inhalation exposure models treat human activity patterns as
sequences of exposure events in which each event is defined by a geographic location and
microenvironment. The U.S. EPA has supported the most comprehensive efforts in this area,
leading to the development of the National Ambient Air Quality Standard Exposure Model and
Probabilistic National Ambient Air Quality Standard Exposure Model (NEM and pNEM)
(Johnson, 2003) and the Modeling Environment for Total Risk Studies/Simulation of Human
Exposure and Dose System (MENTOR/SHEDS) (McCurdy et al., 2000). The Total Risk
Integrated Methodology Inhalation Exposure (TREVI.Expo) model, also referred to as the Air
Pollutants Exposure (APEX) model, was developed by the U.S. EPA as a tool for estimating
AX3-172
-------
human population exposure to criteria and air toxic pollutants. APEX serves as the human
inhalation exposure model within the Total Risk Integrated Methodology (TRIM) framework
(ICF Consulting and ManTech Environmental Technology, Inc. [2003]). APEX, a PC-based
model derived from the probabilistic NAAQS Exposure Model (pNEM), was used in the last O3
NAAQS review (Johnson et al., 1996a, 1996b). Over the past five years, APEX has undergone
several significant improvements in the science reflected in the model and in the databases input
to the model.
Recent European efforts have produced some formulations that have similar general
attributes as the above models but generally involve major simplifications in some of their
components. Examples of recent European models addressing O3 exposures include the AirPEx
(Air Pollution Exposure) model (Freijer et al., 1998), which basically replicates the pNEM
approach, and the AirQUIS (Air Quality Information System) model (Clench-Aas et al., 1999).
The NEM/pNEM, APEX, and MENTOR/SHEDS families of models provide exposure
estimates, defined by concentration and minute ventilation rate for each individual exposure
event, and provide distributions of exposure and O3 dose over any averaging period of concern
from 1 h to an entire O3 season. The above families of models also simulate certain aspects of
the variability and uncertainty in the principal factors affecting exposure. pNEM divides the
population of interest into representative cohorts based on the combinations of demographic
characteristics (age, gender, employment), home/work district, residential cooking fuel, and then
assigns activity diary records (Glen et al., 1997) to each cohort according to demographic
characteristic, season, day-type (weekday/weekend), and temperature. APEX and
MENTOR/SHEDS generates a population demographic file containing a user-defined number of
person-records for each census tract of the population based on proportions of characteristic
variables (age, gender, employment, housing) obtained for the population of interest, and then
assigns the matching activity information based on the characteristic variables. A discussion of
databases on time-activity data, and their influence on estimates of long-term ambient O3
exposure, can be found in Kunzli et al. (1997), McCurdy (2000), and McCurdy et al. (2000).
More recent exposure models are designed (or have been redesigned) to obtain such
information from CHAD (Consolidated Human Activities Database; www.epa.gov/chadnetl:
see Table AX3-18). There are now about 22,600 person-days of sequential daily activity pattern
data in CHAD. All ages of both genders are represented in CHAD. The data for each subject
AX3-173
-------
Table AX3-18. Activity Pattern Studies Included in the Consolidated Human Activity Database (CHAD)
X
OJ
Study Name
Baltimore
CARB: Adolescents and Adults
CARB: Children
Cincinnati (EPRI)
Denver (EPA)
Los Angeles: Elem. School
Children
Los Angeles: High School
Adoles.
National: NHAPS-A8
National: NHAPS-B8
University of Michigan:
Children
Valdez, AK
Washington, DC (EPA)
Calendar Time
Period
of the Study
Jan-Feb 1997
Jul-Aug 1998
Oct 1987-Sept 1988
Apr 1989-Feb 1990
Mar- Apr and Aug
1985
Nov 1982-Feb 1983
Oct 1989
Sept-Oct 1990
Sept 1992-Oct 1994
As above
Feb-Dec 1997
Nov 1990-Oct 1991
Nov 1982-Feb 1983
Diary
Age1
65+
12-94
0-11
0-86
18-70
10-12
13-17
0-93
0-93
0-13
11-71
18-98
Days2
391
1762
1200
2614
805
51
43
4723
4663
5616
401
699
Type3
Diary; 15-min
blocks
Retrospective
Retrospective
Diary
Diary
Diary
Diary
Retrospective
Retrospective
Retrospective
Retrospective
Diary
Time4
24-h Standard
24-h Standard
24-h Standard
24 h; nominal
7 pm-7 am
24 h; nominal
7 pm-7 am
24-h Standard
24-h Standard
24-h Standard
24-h Standard
24-h Standard
Varying 24-h
period
24 h; nominal
7 pm-7 am
Rate5
No
No
No
Yes
No
Yes
Yes
No9
No9
No
No
No
Documentation or
Reference
Williams et al. (2000a,b)
Robinson et al. (1991)
Wiley etal. (1991a)
Wiley etal. (1991b)
Johnson (1989)
Akland etal. (1985)
Johnson (1984)
Spier etal. (1992)
Spier etal. (1992)
Klepeisetal. (1995)
Tsang and Klepeis (1996)
As above
Institute for Social
Research (1997)
Goldstein et al. (1992)
Akland etal. (1985)
Hartwell et al. (1984)
Notes
Multiple days, varying from 5-15;
part of a PM25 PEM study
3 consecutive days; 186 P-D
removed7
Part of CO PEM6 study; 2 consec.
days; 55 P-D removed7
7 P-D removed7
23 P-D removed7
A national random-probability
survey
As above
2 days of data: one is a weekend
day
4 P-D removed7
Part of a CO PEM6 study; 6 P-D
removed7
1 All studies included both genders. The age range depicted is for the subjects actually included; in most cases, there was not an upper limit for the adult studies. Ages are inclusive. Age 0 =
babies < 1 year old.
2 The actual number of person-days of data in CHAD after the "flagging" and removal of questionable data. See the text for a discussion of these procedures.
3 Retrospective: a "what did you do yesterday" type of survey; also known as an ex post survey. Diary: a "real-time" paper diary that a subject carried as he or she went through the day.
4 Standard = midnight to midnight.
5 Was activity-specific breathing rate data collected?
6 PEM = a personal monitoring study. In addition to the diary, a subject carried a small CO or PM2 5 monitor throughout the sampling period.
7 P-D removed = The number of person-days of activity pattern data removed from consolidated CHAD because of missing activity and location information; completeness criteria are listed in
the text.
8 National Human Activity Pattern Study; A = the air version; B = the water version. The activity data obtained on the two versions are identical.
9 A question was asked regarding which activities (within each 6-h time block in the day) involved "heavy breathing," lifting heavy objects, and running hard.
Source: U.S. Environmental Protection Agency (2004a)
-------
consist of one or more days of sequential activities, in which each activity is defined by
start time, duration, activity type (140 categories), and microenvironment classification
(110 categories). Activities vary from 1 min to 1 h in duration. Activities longer than 1 h are
subdivided into clock-hour durations to facilitate exposure modeling. A distribution of values
for the ratio of oxygen uptake rate to body mass (referred to as metabolic equivalents or METs)
is provided for each activity type listed. The forms and parameters of these distributions were
determined through an extensive review of the exercise and nutrition literature. The primary
source of distributional data was Ainsworth et al. (1993), a compendium developed specifically
to "facilitate the coding of physical activities and to promote comparability across studies."
Other information on activity patterns has been reported by Klepeis et al. (1996, 2001); Avol
et al. (1998b); Adams (1993); Shamoo et al. (1994); Linn et al. (1996); Kunzli et al. (1997).
Use of the information in CHAD provides a rational way for incorporating realistic intakes into
exposure models by linking inhalation rates to activity information. As mentioned earlier, an
exposure event sequence derived from activity-diary data is assigned to each population unit
(cohort for pNEM- or REHEX-type models, or individual for APEX or MENTOR/SHEDS-type
models). Each exposure event is typically defined by a start and duration time, a geographic
location and microenvironment, and activity level. The most recent pNEM, APEX, and
MENTOR/SHEDS models have defined activity levels using the activity classification coding
scheme incorporated into CHAD. A probabilistic module within the APEX and
MENTOR/SHEDS-type models converts the activity classification code of each exposure event
to an energy expenditure rate, which in turn is converted into an estimate of oxygen uptake rate.
The oxygen uptake rate is then converted into an estimate of ventilation rate (VE), expressed in
L/min. Johnson (2001) reviewed the physiological principles incorporated into the algorithms
used in pNEM and APEX to convert each activity classification code to an oxygen uptake rate
and describes the additional steps required to convert oxygen uptake to VE.
McCurdy (1997a,b, 2000) recommended that ventilation rate be estimated as a function of
energy expenditure rate. The energy expended by an individual during a particular activity can
be expressed as:
EE = (MET)(RMR) (AX3 -4)
AX3-175
-------
where EE is the average energy expenditure rate (kcal/min) during the activity, MET (metabolic
equivalent of work) is a ratio specific to the activity and is dimensionless, and RMR is the
resting metabolic rate of the individual expressed in terms of number of energy units expended
per unit of time (kcal/min). If RMR is specified for an individual, then the above equation
requires only an activity-specific estimate of MET to produce an estimate of the energy
expenditure rate for a given activity. McCurdy et al. (2000) developed MET distributions for the
activity classifications appearing in the CHAD database.
An important source of uncertainty in existing exposure modeling involves the creation of
multiday, seasonal, or year long exposure activity sequences based on 1- to 3-day activity data
for any given individual from CHAD. Currently, appropriate longitudinal data are not available
and the existing models use various rules to derive longer-term activity sequences using 24-h
activity data from CHAD.
The pNEM family of models used by the EPA has evolved considerably since the
introduction of the first NEM model in the 1980s (Biller et al., 1981). The first such
implementations of pNEM/O3 in the 1980s used a reduced form of a mass balance equation to
estimate indoor O3 concentrations from outdoor concentrations. The second generation of
pNEM/O3 was developed in 1992 and used a simple mass balance model to estimate indoor O3
concentrations. Subsequent enhancements to pNEM/O3 and its input databases included
revisions to the methods used to estimate equivalent ventilation rates (ventilation rate divided by
body surface), to determine commuting patterns, and to adjust ambient O3 levels to simulate
attainment of proposed NAAQS. During the mid-1990s, the EPA applied updated versions of
pNEM/O3 to three different population groups in nine selected urban areas (Chicago, Denver,
Houston, Los Angeles, Miami, New York, Philadelphia, St. Louis, and Washington): (1) the
general population of urban residents, (2) outdoor workers, and (3) children who tended to spend
more time outdoors than the average child. Reports by Johnson et al. (1996a,b,c) describe these
versions of pNEM/O3 and summarize the results of the application of the model to the nine urban
areas. These versions of pNEM/O3 used a revised probabilistic mass balance model to determine
O3 concentrations over 1-h periods in indoor and in-vehicle microenvironments (Johnson, 2003).
The model assumed that there are no indoor sources of O3, that the outdoor O3 concentration and
AER during the clock hour is constant at a specified value, and that O3 decays at a rate
proportional to the outdoor O3 concentration and the indoor O3 concentration.
AX3-176
-------
The new pNEM-derived model, APEX, differs from earlier pNEM models in that the
probabilistic features of the model are incorporated into a Monte Carlo framework. Instead of
dividing the population of interest into a set of cohorts, APEX generates individuals as if they
were being randomly sampled from the population. APEX provides each generated individual
with a demographic profile that specifies values for all parameters required by the model. The
values are selected from distributions and databases that are specific to the age, gender, and other
specifications stated in the demographic profile. The APEX model has not been evaluated,
however, the pCNEM, a Canadian conceptional version of the NEM model, has been evaluated
for estimation of PM10. Since pCNEM is similar to the pNEM/NEM, the APEX model should
work as well as pCNEM. (See discussion on pCNEM later is this section.) The EPA plans to
develop future versions of APEX applicable to O3 and other criteria pollutants.
The latest version of APEX allows for finer geographical units such as census tracts and
automatically assigns population to the nearest monitor within a cutoff distance. Exposure
district-specific temperatures can be specified and the user can select the variables that affect
each parameter (e.g., the AER parameter in certain indoor microenvironments may depend on air
conditioning status or window position). The mass balance algorithms have been enhanced to
allow window position or vehicle speed to also be considered in determining AERs.
The APEX model simulates individual movement through time and space to provide an
estimate of exposure to a given pollutant in the indoor, outdoor, and in-vehicle
microenvironments. The model is highly versatile, allowing input data for specific applications.
APEX provides a good balance in terms of precision and resource expenditure compare with the
more narrowly focused site-specific model and the broadly applicable national screening-level
models.
A key strength of APEX is its ability to estimate hourly exposures and doses for all
simulated individuals in the sampled population. APEX is capable of estimating exposures of
workers in the geographic area where they work, in addition to the geographic area where they
live. APEX is able to represent much of the variability in the exposure estimates resulting from
the variability of the factors affecting human exposure by incorporating stochastic processes
representing the natural variability of personal profile characteristics, activity patterns, and
microenvironment parameters.
AX3-177
-------
A limitation of APEX is that uncertainty in the predicted distributions has not been
addressed. Certain aspects of the personal profiles are held constant (e.g., age) which could be
an issue for simulations with long timeframes. The combined data set for activity patterns
(CHAD) are from a number of different studies and may not constitute a representative sample.
However, the largest portion of CHAD (about 40 percent) is from a study of national scope and
research has shown that activity patterns are generally similar once you take into account age,
gender, day of week, and season/temperature. The commuting data addresses only home-to-
work travel and may not accurately reflect current commuting patterns. The population not
employed outside the home is assumed to always remain in the residential census tract.
Although several of the APEX microenvironments account for time spent in travel, the travel is
assumed to always occur in basically a composite of the home and work tract. Seasonal or year
long sequences for a simulated individual are created by sampling human activity data from
more than one subject, possibly causing an underestimation of the variability from person to
person and an overestimation of the day to day variability for any given individual. The model
does not capture certain correlations among human activities that can impact
microenvironmental concentrations (e.g., cigarette smoking leading to an individual opening a
window, which in turn affects the amount of outdoor air penetrating the residence).
MENTOR/SHEDS estimates the population distribution of pollutant exposure by randomly
sampling from various input distributions. MENTOR/SHEDS is capable of simulating
individuals exposures in eight microenvironments (outdoors, residence, office, school, store,
restaurant, bar, and vehicles) using spatial concentration data for each census tract for outdoor
pollutant concentrations. The indoor and in-vehicle pollutant concentrations are calculated using
specific equations for the microenvironment and ambient pollutant concentration relationship.
Model simulations use demographic data at the census tract level. Randomly selected
characteristics for a fixed number of individual are selected to match demographics within the
census tract for age, gender, employment status, and housing type. Smoking prevalence
statistics by gender and age is randomly selected for each individual in the simulation. Diaries
for activity patterns are matched for the simulated individual by demographic characteristics.
The essential attributes of some of the O3 exposure models and approaches are summarized in
Table AX3-19.
AX3-178
-------
Table AX3-19. Personal and Population Exposure Models for Ozone
Model Name
Model Type
Microenvironments or Predictors
Notes
Reference
pNEM
APEX
Mentor/SHEDS
X
VO
REHEX
Probabilistic General population, outdoor workers,
outdoor children
PC-based Outdoors, indoor residence,
in-vehicle
Probabilistic General population, outdoors, indoors,
in-vehicle
General population
Provides estimates of exposure within a defined
population for a specified period of time.
Uses activity records from CHAD and CADS
Simulates movement through time and space.
Estimates hourly exposures and doses.
Uncertainties in predicted distributions have not
been addressed.
Employees detailed person oriented exposure
approach that includes personal activity data,
physiology, and microenvironmental conditions.
Allows calculation of exposure and dose for
each activity event.
Provides estimates of exposure for 1-day to
3 yrs. Can represent variability in activities of
the population to capture extremes in exposure
distributions.
Johnson et al.
(1996a,b,c)
ICF Consulting and
ManTech
Environmental
Technology, Inc.
(2003)
Georgopoulos et al.
(2005)
Lurmann and Colome
(1991)
-------
Zidek et al. (2000, 2005) described a methodology for predicting human exposure to
environmental pollutants. The methodology builds on earlier models such as SHEDS and
pNEM/NEM and provides a WWW platform for developing a wide variety of models. pCNEM,
a platform model developed from this methodology, is a Canadian PC version of NEM. pCNEM
was used to estimate a conditional predictive exposure distribution for PM10 in London. An
important feature of pCNEM is its ability to estimate the effects of reductions in ambient levels
of pollutants.
Rifai et al. (2000) compared applications of an updated version of REHEX, REHEX-II.
The applications used NHAPS data for the southern states and the 48-state NHAPS or the
Houston-specific time-activity pattern data. The results indicated a sensitivity to the specificity
of the activity data: using Houston-specific data resulted in higher estimates of human exposure
in some of the scenarios. For example, using NHAPS data lead to an estimated 275 thousand-
exposure-hours between 120 to 130 ppb, while use of the Houston-specific activity data lead to
an estimated 297 thousand-exposure-hours between 120 and 130 ppb (8% higher). Using the
Houston-specific activity data in the model resulted in about 2,400 person-exposure-hours above
190 to 200 ppb O3 while no exposure above this threshold was estimated when the NHAPS
activity were used in the model.
Of the above families of models only NEM/pNEM implementations have been
extensively applied to O3 studies. However, it is anticipated that APEX will be useful as an
exposure modeling tool for assessing both criteria and hazardous air pollutants in the future. The
1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) focused on the pNEM/O3
family of models, referring to the review by McCurdy (1994) for the fundamental principles
underlying its formulation and listing, in addition to the "standard" version, three pNEM/O3-
derived models (the Systems Applications International NEM [SAI/NEM]; the Regional Human
Exposure Model [REHEX]; and the Event Probability Exposure Model [EPEM]).
AX3.10.6.2 Ambient Concentrations Models
As mentioned earlier, background and regional outdoor concentrations of pollutants over a
study domain may be calculated either through emissions-based mechanistic modeling or
through ambient-data-based modeling. Emissions-based models calculate the spatiotemporal
fields of the pollutant concentrations using precursor emissions and meteorological conditions as
AX3-180
-------
inputs. The ambient-data-based models typically calculate spatial or spatiotemporal
distributions of the pollutant through the use of interpolation schemes, based on either
deterministic or stochastic models for allocating monitor station observations to the nodes of a
virtual regular grid covering the region of interest. (See later discussion on population exposure
models). Kriging, a geostatistical technique, provides standard procedures for generating an
interpolated O3 spatial distribution for a given time period, using data from a set of observation
points. The kriging approach, with parameters calculated specifically for each hour of the period
of concern, was compared to the Urban Airshed Model (UAM-IV), a comprehensive
photochemical grid-based model for deriving concentration fields. The concentration fields
were then linked with corresponding population data to calculate potential outdoor population
exposure. Higher exposure estimates were obtained with the photochemical grid-based model
when O3 concentrations were <120 ppb, however, the situation was reversed when O3
concentrations exceeded 120 ppb. The authors concluded that kriging O3 values at the locations
studied can reconstruct aspects of population exposure distributions (Georgopoulos et al.,
1997a,b).
Carroll et al. (1997a,b) developed a spatial-temporal model, with a deterministic trend
component, to model hourly O3 levels with the capacity to predict O3 concentrations at any
location in Harris County, Texas during the time period between 1980 and 1993. A fast model-
fitting method was developed to handle the large amount of available data and the substantial
amount of missing data. Ozone concentration data used consisted of hourly measurements from
9 to 12 monitoring stations for the years 1980 to 1993. Using information from the census tract,
the authors estimated that exposure of young children to O3 declined by approximately 20% over
the analysis period. The authors also suggested that the O3 monitors are not sited in locations to
adequately measure population exposures. Several researchers have questioned the suitability of
the model for addressing spatial variations in O3 (Guttorp et al., 1997; Cressie, 1997; Stein and
Fang, 1997).
Spatiotemporal distributions of O3 concentrations have alternatively been obtained using
methods of the "Spatio-Temporal Random Field" (STRF) theory (Christakos and Vyas,
1998a,b). The STRF approach interpolates monitoring data in both space and time
simultaneously. This method can analyze information on "temporal trends," which cannot be
incorporated directly in purely spatial interpolation methods such as standard kriging. Further,
AX3-181
-------
the STRF method can optimize the use of data that are not uniformly sampled in either space or
time. The STRF theory was further extended in the Bayesian Maximum Entropy (BME)
framework and applied to O3 interpolation studies (Christakos and Hristopulos, 1998; Christakos
and Kolovos, 1999; Christakos, 2000). The BME framework can use prior information in the
form of "hard data" (measurements), probability law descriptors (type of distribution, mean and
variance), interval estimation (maximum and minimum values) and even constraint from
physical laws. According to these researchers, both STRF and BME were found to successfully
reproduce O3 fields when adequate monitor data are available.
Zidek et al. (1998, 2002), Sun et al. (1998, 2000), Le et al. (1997), and Li et al. (1999,
2001) used a hierarchical Bayesian approach to predict average hourly concentrations of ambient
O3 and PM10 when monitoring data was not available. A trend model and the Gaussian
stochastic residual was used at the primary level to model the logarithmic field. Space-time
modeling occurs at the second level of the hierarchical prior model so that uncertainty about the
model parameters is expressed at the first level. A special feature of this method is that it does
not require all sites to monitor the same set of pollutants. Also, the model can be updated as new
data become available.
AX3.10.6.3 Microenvironmental Concentration Models
Once specific ambient/local spatiotemporal O3 concentration patterns have been derived,
microenvironments that can represent either outdoor or indoor settings must be characterized.
This process can involve modeling of various local sources and sinks as well as
interrelationships between ambient/local and microenvironmental concentration levels. Three
approaches have been used in the past to model microenvironmental concentrations: empirical,
mass balance, and detailed computational fluid dynamics (CFD).
The empirical fitting approach has been used to summarize the findings of recent field
studies (Liu et al., 1995, 1997; Avol et al., 1998a). These empirical relationships could provide
the basis for future, "prognostic" population exposure models.
Mass balance modeling has ranged from very simple formulations, assuming ideal
(homogeneous) mixing and only linear physicochemical transformations with sources and sinks,
to models that take into account complex multiphase chemical and physical interactions and
nonidealities in mixing. Mass balance modeling is the most common approach used to model
AX3-182
-------
pollutant concentrations in enclosed microenvironments. As discussed earlier, the simplest
microenvironmental setting is a homogeneously mixed compartment in contact with possibly
both outdoor/local environments as well as with other microenvironments. The air quality of
this idealized microenvironment is affected primarily by transport processes (including
infiltration of outdoor air into indoor air compartments, advection between microenvironments,
and convective transport); sources and sinks (local outdoor emissions, indoor emissions, surface
deposition); and local outdoor and indoor gas and aerosol phase chemistry transformation
processes (such as the formation of secondary organic and inorganic aerosols).
Numerous indoor air quality modeling studies have been reported in the literature;
however, depending on the modeling scenario, only a limited number address physical and
chemical processes that affect O3 concentrations indoors (Nazaroff and Cass, 1986; Hayes, 1989,
1991). An example of a mass balance indoor air model for O3 and benzene can be found in the
work of Freijer and Bloemen (2000). They used outdoor O3 measurements to parameterize a
simplified linearized formulation of transport, transformation, and sources and sinks in the
indoor microenvironment.
The pNEM/O3 model includes a sophisticated mass balance model for enclosed (indoor and
vehicle) microenvironments. The general form of this mass balance model is a differential
equation that accounts for outdoor concentration, AER, penetration rate, decay rate, and indoor
sources. To address uncertainty, each of these parameters is represented by a probability
distribution or by a dynamic relationship to other parameters that may change according to time
of day, temperature, air conditioning status, window status, or other factors (Johnson, 2003).
The simplest form of the model is represented by the following differential equation for a
perfectly mixed microenvironment without an air cleaner:
= vCOUT + - vCjx (AX3-5)
where dCIN is the indoor pollutant concentration (mass/volume), t is time in hours, v is the air
exchange rate, COUT is the outdoor pollutant concentration (mass/volume), Vis the volume of the
microenvironment, and S is the indoor source emission rate.
AX3-183
-------
Nazaroff and Cass (1986) extended the mass balance model to include multiple
compartments and interactions between different compounds. The extended model takes into
account the effects of ventilation, filtration, heterogeneous removal, direct emission, and
photolytic and thermal and chemical reactions. A more in-depth discussion of the mass balance
model may be found in Shair and Heitner (1974) and in Nazaroff and Cass (1986).
Freijer and Bloemen (2000) used the one-compartment mass balance model to examine the
relationship between O3 I/O ratios as influenced by time patterns in outdoor concentrations,
ventilation rate, and indoor emissions. The microenvironment was 250 m3. Three different
ventilation patterns with the same long-term average AER (0.64 IT1) were used. A source
pattern (direct emissions) of zero was used, because O3 sources are not common. The time series
for outdoor O3 concentrations consisted of 100-day periods during the summer, with hourly
measured concentrations. The following assumptions were made: (1) O3 concentration in the air
that enters the microenvironment is equal to the concentration of the outside air minus the
fraction removed by filtration, (2) the O3 concentration that leaves the microenvironment equals
the O3 concentration in the microenvironment, (3) the decay processes in the microenvironment
are proportional to the mass of the pollutant, and (4) addition or removal of O3 in the
microenvironment also may occur through independent sources and sinks.
Figure AX3-88 represents the measured outdoor O3 concentrations and modeled indoor O3
concentrations. The indoor modeled O3 concentrations were found to equal approximately 33%
of the outdoor monitored concentrations.
Few indoor air models have considered detailed nonlinear chemistry, which, however, can
have a significant effect on the indoor air quality, especially in the presence of strong indoor
sources. Indeed, the need for more comprehensive models that can take into account the
complex, multiphase processes that affect indoor concentrations of interacting gas phase
pollutants and particulate matter has been recognized and a number of formulations have
appeared in recent years. For example, the Exposure and Dose Modeling and Analysis System
(EDMAS) (Georgopoulos et al., 1997a) included an indoor model with detailed gas-phase
atmospheric chemistry. This indoor model accounts for interactions of O3 with indoor sinks and
sources (surfaces, gas releases) and with entrained gas. The indoor model was dynamically
coupled with the outdoor photochemical air quality models, UAM-IV and UAM-V (Urban
Airshed Models), which provided the gas-phase composition of entrained air, and with a
AX3-184
-------
d>
8
o
63
51 -
37.6 -
12.5-
100
101
102 103
Time (day)
104
105
Figure AX3-88. Measured outdoor O3 concentrations (thin line) and modeled indoor
concentrations (bold line).
Source: Adapted from Freijer and Bloemen (2000).
physiologically based O3 uptake and dosimetry model. Subsequent work (Isukapalli et al., 1999)
expanded the framework and features of the EDMAS model to incorporate alternative
representations of gas-phase chemistry as well as multiphase O3 chemistry and gas/aerosol
interactions. The new model is a component of the integrated Modeling Environment for Total
Risk studies (MENTOR).
Sarwar et al. (2001, 2002) modeled estimates of indoor hydroxyl radical concentrations
using a new indoor air quality model, Indoor Chemistry and Exposure Model (ICEM). The
ICEM uses a modified SAPRC-99 atmospheric chemistry mechanism to simulate indoor
hydroxyl radical production and consumption from reactions of alkenes with O3. It also allows
for the simulation of transport processes between indoor and outdoor environments, indoor
emissions, chemical reactions, and deposition. Indoor hydroxyl radicals, produced from O3 that
penetrates indoors, can adversely impact indoor air quality through dark chemistry to produce
photochemical oxidants.
S0rensen and Weschler (2002) used CFD modeling to examine the production of a
hypothetical product from an O3-terpene reaction under two different ventilation scenarios. The
computational grid used in the model was nonuniform. There were significant variations in the
concentrations of reactants between locations in the room, resulting in varying reaction rates.
AX3-185
-------
Because the "age of the air" differed at different locations in the room, the time available for
reactions to occur also differed between locations.
Very few studies have focused on mechanistic modeling of outdoor microenvironments.
Fraigneau et al. (1995) developed a simple model to account for fastNO-O3 reaction/dispersion
in the vicinity of a motorway. Proyou et al. (1998) applied a simple three-layer photochemical
box model to an Athens street canyon. However, the adjustments of O3 levels for sources, sinks,
and mixing in outdoor microenvironments are done in a phenomenological manner in existing
exposure models, driven by limited available observations. Ongoing research is evaluating
approaches for quantifying local effects on outdoor O3 chemistry in specific settings.
Finally, one issue that should be mentioned is that of evaluating comprehensive prognostic
exposure modeling studies, for either individuals or populations, with field data. Attempts had
been made to evaluate pNEM/O3-type models using personal exposure measurements (Johnson
et al., 1990). Although databases that would be adequate for performing a comprehensive
evaluation are not expected to be available any time soon, a number of studies are building the
necessary information base, as discussed previously. Some of these studies report field
observations of personal, indoor, and outdoor O3 concentrations and describe simple
semiempirical personal exposure models that are parameterized using observational data and
regression techniques.
AX3.10.7 Measured Exposures and Monitored Concentrations
AX3.10.7.1 Personal Exposure Measurements
Passive O3 monitors have been used in several field studies to determine average daily O3
exposure as well as in scripted studies to evaluate O3 exposures over one to several-hour time
periods. Table AX3-20 lists the results of O3 exposure studies. Delfino et al. (1996) measured
12-h personal daytime O3 exposures in asthmatic patients in San Diego from September through
October 1993. They found that the mean personal exposures were 27% of the mean outdoor
concentrations. Individual exposure levels among the 12 asthmatic subjects aged 9 to 16 years
varied greatly. Mean personal O3 exposure levels were lower on Friday, Saturday, and Sunday
than on other days of the week (10 versus 13 ppb), while the ambient air concentrations were
higher Friday through Sunday. The authors suggested that the differences were due to higher
weekday NO emissions from local traffic which titrated the ambient O3 levels. The lower
AX3-186
-------
Table AX3-20. Personal Exposure Measurements
Location, Population, Sample Duration
Personal Exposure Mean3
(range) (ppb)
Reference
San Diego, CA, Asthmatics ages 9-18 years, 12
12 hour
Vancouver, Canada, Adult Workers, Daily 585
High indoor time
Moderate indoor time
Only outdoor
Southern California, Subjects 10-38 years 24
Spring
Fall
Montpellier, France, Adults, Hourly 16
Winter
Summer
Souther California, Children 6-12 years, 169
> 6 days
Upland - winter
- summer
Mountain - winter
- summer
Baltimore, MD, Technician, Hourlyb 1
Winter
Summer
Baltimore, MD, Adults 75 ± 7 years, Daily 20
Winter
Summer
12 ± 12 (0-84)
10 weekend
12 weekday
(ND-9)
(ND-12)
(2-44)
13.6 ±2.5 (-to 80)
10.5 ±2.5 (-to 50)
34.3 ± 17.6 (6.5-88)
15.4 ±7.7 (6.5-40)
44.1 ± 18.2(11-88)
6.2 ±4.7 (0.5-41)
19 ±18 (0.5-63)
5.7 ±4.2 (0.5-31)
25 ± 24 (0.5-72)
3.5 ± 7.5 (ND-49)
15 ± 18 (ND-76)
3.5±3.0(ND-9.9)
0. ± 1.8 (ND-2.8)
Delfinoetal. (1996)
Brauer and Brook
(1997)
Liu etal. (1997)
Bernard etal. (1999)
Geyh et al. (2000)
Chang et al. (2000)
Sarnat et al. (2000)
aND = not detected.
bMeasurements made following scripted activities for 15 days.
personal exposure levels on Friday, Saturday, and Sunday may have been an artifact of greater
noncompliance, with the badges remaining indoors and, therefore, being exposed to lower O3
concentrations. The overall correlation between the personal exposure concentrations between
any two individuals and with the outdoor stationary site was only moderate (r = 0.45; range:
0.36 to 0.69). The O3 concentrations at the stationary site exceeded the personal levels by an
average of 31 ppb. Avol et al. (1998b) observed a poor correlation between personal exposure
and fixed-site monitoring concentrations (r = 0.28, n = 1336 pairs) for a cohort of children
AX3-187
-------
(healthy, wheezy, and asthmatic). Personal exposure measurements were generally lower than
integrated hourly data. Sarnat et al. (2000) measured personal O3 exposures during a 12-day
longitudinal study of 20 older adults (>64 years) in Baltimore, MD. The subjects spent >94% of
the time indoors. Ten subjects participated in the summer and winter exposures and the
remaining 10 participated in either the summer or winter exposure. No statistically significant
overall correlations were identified between the personal and the ambient O3 concentrations
during either the winter or summer. Only a single individual (n = 14 summer and 13 winter) had
a significant correlation with outdoor concentrations. Geyh et al. (2000) measured indoor and
outdoor concentration and personal O3 exposures in 169 elementary school children from
116 homes during a year-long sampling protocol in 2 communities in southern California
(Upland and Mountain communities). Samples were collected for 1 week per month. Boys had
higher O3 exposure than girls, probably reflecting the greater amount of time boys spent
outdoors compared to girls (3.8 versus 3.2 h for the spring/summer and 2.9 versus 2.2 h for the
fall/winter). The average personal O3 exposures were lower than the levels measured at the
nearest monitor stations retrieved from the AQS. There was no significant difference in the O3
exposure for both groups during the non-O3 season (6.2 and 5.7 ppb for Upland and the
mountain communities, respectively), however, children in the mountainous region were
exposed to 35% more O3 than children in Upland during the O3 season (18.8 and 25.4 ppb for
Upland and the mountain communities) (two-tailed t-test,p < 0.01). During the O3 season,
differences were found in indoor concentrations and personal O3 exposures between the two
communities participating in the study based on ambient air concentrations and differences in air
exchange rates in the homes.
Brauer and Brook (1997) conducted personal exposure evaluations in three groups in Frazer
Valley, Vancouver, Canada. The groups were divided by the amount of time spent outdoors:
(1) the majority of the workday was spent indoors or commuting (25 medical clinic workers), (2)
an intermediate amount of time was spent outdoors (25 overnight camp staff members), and (3)
the entire exposure monitoring period was outdoors (15 adult farm workers). Time-activity data
were collected for the first two groups to assess the proportion of time spent outdoors.
For groups 1 and 2, the lowest quartile of participants based on the fraction of time spent
outdoors (0 to 25% and 7.5 to 45%, respectively) had significantly lower O3 exposure (mean
personal exposure to outdoor concentration ratio = 0.18 and 0.35, respectively) compared to
AX3-188
-------
those in the upper quartile (mean ratio = 0.51 and 0.58, respectively; p < 0.05; Bonferroni
multiple range test). The mean ratio was 0.96 with values ranging from near 0 to 2 for group 3,
the group that spent the entire exposure-monitoring period outdoors. The authors attributed the
extreme low ratios to random measurement error at low O3 air concentrations (estimated at
35%), local variability in O3 concentrations, and to differences between ground-level
concentrations (where the personal samples were collected) 3-m above ground level (where the
continuous monitors were located). The highest ratios may be due to either locale variability in
O3 concentrations or to an interference affecting the personal O3 samplers, particularly at the
lower concentration range, leading to a small positive error. Temporal plots of O3 for the mean
daily personal exposures and ambient concentrations showed the same general trend with
general agreement between the personal exposures and ambient air concentrations for group 3.
However, for groups 1 and 2, the day-to-day variability of the personal exposures and ambient
O3 concentrations had consistent patterns, suggesting that the ambient air was the primary source
for O3 exposure. The day-to-day variability in personal and continuous measurements was 0.60,
0.42, and 0.64 for groups 1 through 3, respectively. The actual O3 concentrations measured in
the personal air space were always considerably lower than the ambient concentrations. Bernard
et al. (1999) assessed O3 personal exposure and in the home and outdoor O concentration for up
to 110 subjects. Measurements were conducted over 5-day periods between June 1995 and
October 1996. As anticipated, O3 concentrations were higher during the warmer months. Mean
O3 concentrations for 70 subjects were 22, 35, 17.4, 40.5, and 18 ppb for personal, outdoor
home, indoor home, outdoor work, and indoor work, respectively for measurements made during
the warmer months.
In a study by Liard et al. (1999), 55 mild to moderate asthmatic adults (18 to 65 y old) and
39 children (7 to 15 y old) were monitored for O3 exposure. Subjects were monitored during
three periods, 4 days per monitoring period and asked to keep a diary of time spent outdoors and
in a car. Many of the study subjects O3 exposures were often below the level of detection for the
method used. Ozone exposure levels correlated with the hours spent outdoors. Ozone personal
exposure correlated poorly with the ambient monitoring measurements, however, the mean
values for all subjects correlated with those measurements from the ambient monitoring site
(r = 0.83, p < 0.05). Linn et al. (1996) estimated short-term O3 exposures in 269 children from
three communities in the Los Angeles Basin by monitoring air at head level in school class
AX3-189
-------
rooms, on the roof of one level school buildings, and in personal microenvironments of selected
individuals. Monitoring was carried out for six weeks in the fall, winter, and spring, two
successive weeks per season at each of three schools. Each subject was monitored for one week
in each season over a two year period. According to the authors there were meaningful
associations between personal exposures and central monitoring site O3 measurements (r = 0.61).
Based on information reported in the questionnaires, outdoor activity increased slightly in
communities/seasons with higher pollution.
Lee et al. (2004) found that personal O3 exposure was positively correlated with time spent
outdoors (r = 0.19, p < 0.01) and negatively correlated with time spent indoors (r = -0.17,
p <0.01) in elementary school children. Thirty-three elementary school children from two
Nashville, TN area school districts participated in a six week long O3 monitoring study during
the summer vacation. The study participants maintained a dairy of daily activities during the
study period. An additional 62 children from the same school completed a telephone interview
on time/activity at least eight times during the study period. Study participants wore a passive
sampler during their non-sleep time and the sampler was placed near their bed at night.
A passive monitor also was placed outside and inside of the home. Personal exposure correlated
with the amount of time spent outdoors. Exposures ranged from 0.0013 to 0.0064 ppm for
indoor concentrations compared to ambient concentrations of 0.011 to 0.042 ppm O3.
AX3.10.7.2 Monitored Ambient Concentrations
Ozone has been measured more extensively than the other photochemical oxidants.
Ambient monitors have been established in most areas of the country, with extensive monitoring
in regions that have been in noncompliance with the previous 1-h daily NAAQS. Monitoring
station-measured hourly concentrations also have been used as surrogates of exposure in
epidemiological studies and in evaluating exposure-related health effects. According to the
Guideline on Ozone Monitoring Site Selection (U.S. Environmental Protection Agency, 1998),
when designing an O3 monitoring network, consideration should be given to (1) proximity to
combustion emission sources, (2) distance from primary emission sources, and (3) the general
wind direction and speed to determine the primary transport pathways of O3 and its precursors.
Finally, the 1-h daily maximum and 8-h average O3 concentrations can have different spatial
patterns with elevated daily 8-h O3 concentrations typically being over a wider spatial area.
AX3-190
-------
Therefore, O3 monitoring networks should determine the highest concentrations expected to
occur in the area, representative concentrations for high population density areas, the impact of
sources or source categories on air pollution levels, and general background concentration levels
(U.S. Environmental Protection Agency, 1998).
The guideline also states that the monitor's O3 inlet probe should be placed at a height and
location that best approximates where people are usually found. However, complicating factors
(e.g., security considerations, availability) sometimes result in the probe placement being
elevated 3 to 15 m above ground level, a different location than the breathing zone (1 to 2 m) of
the populace. Although there are some commonalities in the considerations for the sampling
design for monitoring and for determining population exposures, differences also exist. These
differences between the location and height of the monitor compared to the locations and
breathing zone heights of people can result in different O3 concentrations between what is
measured at the monitor and exposure and, therefore, should be considered when using ambient
air monitoring data as a surrogate for exposure in epidemiological studies and risk assessments.
Further, since most people spend the majority of their time indoors, where O3 levels tend to be
much lower than outdoor ambient levels, the use of ambient monitoring data to determine
exposure generally overestimates true personal O3 exposure, resulting in exposure estimates
biased towards the null. Information on monitored ambient concentrations of O3 and other
photochemical oxidants appears earlier in this chapter.
AX3.10.7.3 Ozone Concentrations in Microenvironments
The 1996 O3 AQCD for Ozone (U.S. Environmental Protection Agency, 1996a) reported
O3 I/O ratios for a variety of indoor environments including homes, office/laboratories, a
hospital, museums, a school room, and automobiles and other vehicles. Ozone I/O ratios ranged
from 0.09 to 1.0 for residences, 0.19 to 0.8 for offices/laboratories, hospital and school rooms,
and 0.03 to 0.87 for museums and art galleries. The higher O3 ratios were generally noted in
indoor environments with high AERs or 100% outside air intake. Studies published since
completion of the 1996 O3 AQCD are discussed in this section. The findings of the more recent
studies on O3 I/O ratios are included in Table AX3-21.
Northeast States for Coordinated Air Use Management (NESCAUM, 2002) monitored
levels of O3 inside and outside of nine schools located in the New England states. The schools
AX3-191
-------
Table AX3-21. Indoor/Outdoor Ozone Ratios
Location and Ventilation
Conditions
Indoor/Outdoor Ratio
Mean (range)
Comments
Reference
X
VO
to
Toronto, Canada, Homes
Winter - weekly (68)
Summer - weekly (38)
Summer-12 h/day (128)
- 12 h/night (36)
Boston, MA, Homes (26)
Winter - continuously 24 h
Summer - continuously 24 h
Mexico City, School
Windows/Doors Open (27)
Windows/Doors Closed,
cleaner off (41)
Windows/Doors Closed,
cleaner on (47)
Los Angeles, CA
Homes (95)
Other locations (57)
0.07 ±0.10 (ND- 0.63)
0.40 ± 0.29 (ND-1.15)
0.30 ± 0.32 (ND - 1.42)
0.43 ± 0.54 (ND - 2.89)
0.30 ± 0.42 (ND-1.31)
0.22 ± 0.25 (ND - 0.88)
0.73 ± 0.04
0.17 ±0.02
0.15 ±0.02
0.28
0.18
Electrostatic air cleaners were present in about 50% of the homes. Liu et al.
Air conditioners were present in about 80% of the homes, most were (1995)
central units that used recycled air. Air conditioners used in only 13 of
the 40 homes on a daily basis. Measurements were made both inside
and outside of the homes for 5 consecutive 24-h periods. Homes with
electrostatic air cleaners had higher I/O ratios during the winter months.
The mean average weekly AER for all homes during the winter months
was 0.69 ± 0.88 h"1 with 50% of the homes with an AER of less than
0.41 h~'. For the summer months, the mean average AER was 1.04 ±
1.28 h"1 with 50% of the homes with an AER of less than 0.52 h.
Study examined the potential for O3 to react with VOCs to form acid Reiss et al.
aerosols. Carbonyls were formed. No clear trend of O3 with AERs. (1995)
The average AER was 0.9 h"1 during the winter and 2.6 h"1 during the
summer. Four residences in winter and nine in summer with 24 h
average concentrations. Air concentrations varied from 0-34.2 ppb
indoors and 4.4-40.5 ppb outdoors.
Study conducted over 4-day period during winter months. Two-min Gold et al.
averaged measurements were taken both inside and outside of the (1996)
school every 30 min from 10 a.m. to 4 p.m. Estimated air exchange rates
were 1.1, 2.1, and 2.5 h ' for low, medium, and high flow rates. Ozone
concentrations decreased with increasing relative humidity.
Study conducted in September. Monitored O3 concentrations consisted Johnson
of twenty-one 24-h periods beginning at 7 p.m. and ending at 7 p.m. on (1997)
the following day. Ozone concentrations were higher at the fixed
monitoring sites during the afternoon. The weather was sunny and the
temperature was high. I/O ratio was lower when windows were closed.
The effect of air conditioning on the I/O was varied.
-------
Table AX3-21 (cont'd). Indoor/Outdoor Ozone Ratios
Location and Ventilation
Conditions
Indoor/Outdoor Ratio
Mean (range)
Comments
Reference
X
I
^o
Mexico City
Homes (237)
Schools (59)
Los Angeles, Homes (239)
Summer
Winter
California
Homes
no AC, window opened (20)
AC, windows closed (3)
Munich Germany
Office
Gymnasium
Classroom
Residence
Bedroom
Livingroom
Hotel room
Car
La Rochelle, France
0.20 ±0.18 (0.04-0.99)
0.1-0.3
0.3-0.4
0.37 ±0.25 (0.06-1.5)
0.43 ±0.29
0.32 ±0.21
0.68 (n = 20)
0.09 (n= 3)
0.4-0.9
0.49-0.92
0.54-0.77
0.47 - 1.0
0.74 - 1.0
0.02
0.4-0.6
0-0.45
Ozone monitoring occurred between September and July. Study
included 3 schools and 145 homes. Most of the homes were large and
did not have air conditioning. Ninety-two percent of the homes had
carpeting, 13% used air filters, and 84% used humidifiers. Thirty-five
percent opened windows frequently, 43% sometimes, and 22% never
between 10 a.m. and 4 p.m. Ozone was monitored at the schools sites
from 8 a.m. to 1 p.m. daily for 14 consecutive days. Homes
were monitored for continuous 24-h periods for 14 consecutive days.
I/O based on 1-h average concentrations.
Four hundred and eighty-one samples collected inside and immediately
outside of home from February to December. Ratios based on 24-h
average O3 concentrations indoors and outdoors. Low outdoor
concentrations resulted in low indoor concentrations. However, high
outdoor concentrations resulted in a range of indoor concentrations
and ratios. I/O ratios were highest during the summer months.
I/O ratio was determined for 20 homes. Only 3 of the homes operated
the air conditioning. I/O ratios based on 24-h continuous ambient
concentrations and 0.5-1 h average indoor concentrations.
Indoor concentrations were dependent on the type of ventilation.
I/O ratio determined for 8 schools. Monitoring conducted for a 2-wk
period. Schools located in various areas with different ambient O3
concentrations, types of ventilation systems, and the presumed
air-tightness of building envelop. Building air-tightness and ambient O3
concentration influenced indoor O3 I/O.
Romieu
etal. (1998)
Avol et al.
(1998a)
Lee et al.
(1999)
Jakobi and
Fabian
(1997)
Blondeau
et al. (2005)
-------
Table AX3-21 (cont'd). Indoor/Outdoor Ozone Ratios
Location and Ventilation
Conditions
Indoor/Outdoor Ratio
Mean (range)
Comments
Reference
X
I
^o
Montpellier, France, Homes (110)
Southern CA, Homes
Upland Mountains
Krakow, Poland
Museums (5)
0.41
0.68 ±0.18
(windows open)
0.07-0.11
(AC used)
0.13 -0.42
Buildings, Greece
Thessaloniki
Athens
0.24 ±0.18 (0.01 to 0.75)
Ozone measurements were made over 5-day periods in and outside of 21 Bernard
homes during the summer and winter months. The winter I/O ratio was et al. (1999)
0.31 compared to 0.46 during the summer months.
Ozone measurements were taken at 119 homes (57 in Upland and 62 in Geyh et al.
towns located in the mountains) during April and May. I/O ratios were (2000);
based on average monthly outdoor concentrations and average weekly Lee et al.
indoor concentrations. I/O ratio was associated with the home location, (2002)
number of bedrooms, and the presence of an air conditioner. I/O ratios
based on subset of the homes.
Ozone continuously monitored at five museums and cultural centers. Salmon et al.
Monitoring conducted during the summer months for 21 -46 h or 28-3 3 (2000)
days at each of the sites. The I/O was found to be dependent on the
ventilation rate, i.e., when the ventilation rate was high, the I/O ratio
approached unity, while in rooms sequestered from the outdoor air or
where air was predominantly recycled through charcoal filters, the O3
levels indoors were greatly reduced resulting in low I/O ratios.
There was no heating/air conditioning system in the building at Drakou et al.
Thessaloniki. Windows were kept closed during the entire monitoring (1998)
period. Complete air exchange took place every 3 h. I/O ratio ranged
from 0.5-0.90, due to low deposition velocity on indoor surfaces. The air
conditioning system in continuous use at the Athens site recirculated
the air. Complete air exchange was estimated to be 1 h. Monitoring
lasted for 30 days at each site, but only the 7 most representative days
were used for calculation of the I/O ratio.
Southern California, Museum
0.19 ±0.05
Measurements made over a 2-wk period (24-h avg). Ratio for
concentrations at the buffer zone with the roll-up door closed.
Hisham and
Grosjean
(1991)
ND = not detectable.
-------
represented a variety of environmental conditions in terms of ambient O3 concentration, sources,
geographic location, population density, traffic patterns, and building types. Schools were
monitored during the summer months to establish baseline O3 concentrations and again during
the fall after classes started. A monitor was placed directly outside of the school entrance and
50 feet away from the entrance in the hall. Where available, monitors were placed at locations
identified as "problem" classrooms, classrooms with carpeting, or in rooms close to outdoor
sources of O3. As expected, outdoor concentrations of O3 were higher than those found indoors.
Averaged O3 concentrations were low during the early morning hours (7:30 a.m.) but peaked to
approximately 25 and 40 ppb around 1:30 p.m. indoors and outdoors, respectively.
Gold et al. (1996) compared indoor and outdoor O3 concentrations in classrooms in Mexico
City under three different ventilation conditions: windows/doors open, air cleaner off;
windows/doors closed, air cleaner off; and windows/doors closed, air cleaner on. Two-minute
averaged outdoor O3 levels varied between 64 and 361 ppb, while indoor O3 concentrations
ranged from 0 to 247 ppb. The highest indoor O3 concentrations were noted when the
windows/doors were open. The AERs were estimated to be 1.1, 2.1, and 2.5 IT1 for low,
medium, and high air flow rates, respectively. The authors indicated that the indoor levels, and
therefore O3 exposure to students in schools, can be reduced to < 80 ppb by closing windows and
doors even when ambient O3 levels reach 300 ppb.
In a second Mexico City study, Romieu et al. (1998) measured O3 concentrations inside
and outside of 145 homes and three schools. Measurements were made between November and
June. Most of the homes were large and did not have air conditioning. Ninety-five percent of
the homes had carpeting, 13% used air filters, and 84% used humidifiers. Thirty-five percent of
the homeowners reported that they opened windows frequently between the hours of 10 a.m. and
4 p.m., while 43% opened windows sometimes and 22% reported that they never opened
windows during that time period. Homes were monitored for continuous 24-h periods for
14 consecutive days. Schools were monitored from 8 a.m. to 1 p.m. or continuously for 24 h.
During the school monitoring periods the windows were frequently left open and the doors were
constantly being opened and closed. The mean indoor and outdoor O3 concentrations during 5-h
measurements at the schools were 22 ppb and 56 to 73 ppb, respectively. The mean indoor and
outdoor O3 concentrations for measurements made over a 7- and 14-day period at the test homes
were 5 and 27 ppb and 7 and 37 ppb, respectively. Ozone concentrations inside of homes were
AX3-195
-------
dependent on the presence of carpeting, the use of air filters, and whether the windows were
open or closed. Air exchange rates were not reported.
Reiss et al. (1995) compared indoor and outdoor O3 concentrations for residences in the
Boston, MA area. Four residences were monitored during the winter months and nine residences
during the summer months. Outside monitors were placed ~1 m away from the house.
Monitoring was conducted over a continuous 24-h period. There were no indoor sources of O3.
Indoor O3 concentrations were higher during the summer months, with concentrations reaching
34.2 ppb. Indoor O3 concentrations reached as high as 3.3 ppb during the winter monitoring
period. In one instance, O3 concentrations were higher indoors than outdoors. The authors
attributed that finding to analytical difficulties. Outdoor O3 concentrations were generally higher
during the summer monitoring period, with concentrations reaching 51.8 ppb. Indoor
concentrations were dependent on the outdoor O3 concentrations and AER. Indoor and
outdoor O3 concentrations, including the AERs at the times of the monitoring are included in
Table AX3-22.
Avol et al. (1998a) monitored 126 home in the Los Angeles metropolitan area during
February and December. Uniformly low ambient O3 concentrations were present during the
non-O3 seasons. Indoor O3 concentrations were always below outdoor O3 concentrations. The
mean indoor and outdoor O3 concentrations over the sampling period were 13 ± 12 ppb and 37 ±
19 ppb, respectively. There was a correlation between indoor O3 concentration and ambient
temperatures. The effect was magnified when the windows were open. When a central
refrigerant recirculating air conditioner was used, indoor O3 concentrations declined. The
authors were able to predict indoor O3 levels with nearly the same accuracy using measurements
made at regional stations coupled with window conditions as with measurements made directly
outside the homes coupled with window conditions, suggesting that monitoring station data may
be useful in helping to reduce errors associated with exposure misclassification. The authors
cautioned that varying results may occur at different locations with different housing stock or at
different times of the year.
Lee et al. (2002) reported indoor and outdoor O3 concentrations in 119 homes of school
children in two communities in southern California: Upland and San Bernadino county.
Measurements were made over one year. Outdoor and indoor O3 concentrations were based on
AX3-196
-------
Table AX3-22. Indoor and Outdoor O3 Concentrations in Boston, MA
Indoor Data
Residence
Indoor
Ozone
(Ppb)
Outdoor
Ozone
(Ppb)
AER
(IT1)
Relative
Humidity
(%)
Temp. (°F)
Outdoor Data
Relative
Humidity
(%)
Temp. (°F)
Winter
I
I
2
2
3
3
4
4
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
3.3
0
2.6
1.7
20.4
3.1
2.2
0.3
11.2
15
24.4
4.4
15.6
24.5
11.4
20.7
1
0.8
1
1
1
0.9
0.7
0.7
25-45
22-40
3-19
3-8
8-24
13-19
26-37
29-38
67-75
65-76
67-71
55-70
62-69
64-70
60-72
61-70
88
62
44
40
33
51
57
77
25
27
17
17
36
38
39
36
Summer
1
1
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
-Day 1
-Day 2
5.6
0.6
0.8
5
34.2
6.9
4.3
4.9
1.4
1.9
0.8
1.7
3.9
0
22.9
23.5
1.6
1.2
32.4
13.4
14.3
24.1
38.9
14
30
40.5
17.5
19
8.2
18.6
40.1
33.9
51.8
31.6
20.9
25
3
2.3
2.4
2.1
4.6
3.1
1.4
1.8
5.1
3.5
0.5
0.7
1.1
1.1
3.2
6.3
2.1
1.7
28-44
37-44
48-54
46-60
48-63
45-53
37-60
38-68
30-50
39-63
59-73
43-66
57-70
58-73
66-81
43-67
N/A
33-52
71-74
70-74
73-79
72-78
64-80
65-69
66-75
67-79
69-74
N/A
74-77
76-78
70-77
72-75
71-77
66-79
N/A
75-79
44
59
54
64
52
52
51
64
54
40
76
47
51
64
75
48
37
70
65
60
70
73
71
62
67
67
58
64
72
76
75
72
75
72
70
66
N/A = not available
Adapted from: Reiss et al. (1995)
AX3-197
-------
cautioned that varying results may occur at different locations with different housing stock or at
different times of the year.
Lee et al. (2002) reported indoor and outdoor O3 concentrations in 119 homes of school
children in two communities in southern California: Upland and San Bernadino county.
Measurements were made over one year. Outdoor and indoor O3 concentrations were based on
monthly and weekly averages, respectively. Housing characteristics were not found to affect
indoor O3 concentrations. Indoor O3 concentrations were significantly lower than outdoor O3
concentrations. Average indoor and outdoor O3 concentrations were 14.9 and 56.5 ppb. Homes
with air conditioning had lower O3 concentrations, suggesting decreased ventilation or greater
O3 removal.
Chao (2001) evaluated the relationship between indoor and outdoor levels of various air
pollutants, including O3, in 10 apartments in Hong Kong during May to June. Air monitoring
was conducted over a 48-h period. All participants had the habit of closing the windows during
the evenings and using the air conditioner during the sleep hours. Windows were partially open
during the morning. The air conditioners were off during the working hours. Indoor O3
concentrations were low in all of the monitored apartments, ranging from 0 to 4.9 ppb with an
average of 2.65 ppb. Outdoor O3 concentrations ranged from 1.96 to 15.68 ppb. Table AX3-23
provides information on the indoor and outdoor O3 concentrations and the apartment
characteristics.
Drakou et al. (1995) demonstrated the complexity of the indoor environment.
Measurements of several pollutants, including O3, were made inside and outside of two
nonresidential buildings in Thessaloniki and Athens, Greece. The building in Thessaloniki was a
58-m3 metal structure. The ceiling and walls were covered with colored corrugated plastic
sheeting, and the flooring was plastic tile. There was no heating/air conditioning system and the
building was closed during the monitoring period. The AER ranged from 0.3 to 0.35 h"1. The
building in Athens was a 51-m3 concrete structure. The air conditioning system (recirculated air)
worked continuously during the monitoring period. A window was opened slightly to
accommodate the monitors' sampling lines. The AER was approximately 1 h"1. Monitoring
lasted for 30 days at both locations, however, only data from the 7 most representative days were
reported. Indoor O3 concentrations closely followed the outdoor concentrations at the
Thessaloniki building. The averaged 7-day indoor and outdoor O3 concentrations were 9.39 and
AX3-198
-------
Table AX3-23. Indoor and Outdoor O3 Concentrations in Hong Kong
Apartment
No.
1
2
o
5
4
5
6
7
8
9
10
Floor
Area
40
47
140
67
86
43
47
30
26
20
Floor
No.
14
13
2
5
11
32
9
6
35
15
Window Opening
Frequency
Seldom
Sometimes
Sometimes
Seldom
Sometimes
Sometimes
Always
Seldom
Sometimes
Seldom
AER (IT1)
1.44
1.97
0.83
5.27
1.64
15.83
15.91
3.25
2.1
5.5
Indoor
O3 Cone.
0
4.96*
1.0*
4.96*
3.0*
4.01*
0
2.05*
4.9
4.01*
Outdoor
O3 Cone.
1.96
6.01*
6.96*
8.76*
7.80*
8.76*
3.0*
3.0*
15.68
4.96*
*Estimated Concentration.
Adapted from Chao (2001).
15.48 ppb, respectively. The indoor O3 concentrations at the Athens location were highly
variable compared to the outdoor concentration. The authors suggested that a high hydrocarbon
intrusion may be the reason for the variability in O3 concentrations noted at this site. The
averaged 7-day indoor and outdoor O3 concentrations were 8.14 and 21.66 ppb, respectively.
Weschler et al. (1994) reported that indoor O3 concentrations closely tracked outdoor O3
concentrations at a telephone switching station in Burbank, CA. The switching building was
flat-roofed, two-story (first floor and basement), uncarpeted, with unpainted brick walls. Each
floor was 930 m2 with a volume of 5095 m3. Indoor O3 concentrations were measured on the
first floor using a perfluorocarbon tracer or an UV photometric analyzer. The AER were
obtained by dividing the volumetric flow to the first floor by the volume of the first floor space
or by perfluorocarbon tracer techniques. The AER ranged from 1.0 to 1.9 h"1.
The major source of O3 at the switching station was transport from outdoors. Indoor O3
concentrations closely tracked outdoor concentrations, measuring from 30 to 70% of the outdoor
concentrations. Indoor O3 concentrations frequently exceed 50 ppb during the summer months,
AX3-199
-------
but seldom exceeded 25 ppb during the winter. During the early spring through late fall, indoor
O3 concentrations peaked during the early afternoon and approach zero after sunset. Ozone sinks
included a surface removal rate between 0.8 and 1.0 h'1 and reactions with NOX.
Figures AX3-89 and AX3-90 compare the outdoor and indoor O3 concentrations, including the
AER, for two 1-week periods during the study.
July, 1992
£150
0)
I 10°
O
Figure AX3-89. Air exchange rates and outdoor and indoor O3 concentrations during the
summer at a telephone switching station in Burbank, CA.
Source: Weschleret al. (1994).
The relationships between indoor and outdoor O3 concentrations in five museums and
cultural centers (Wawel Castle, Matejko Museum, National Museum, Collegium Maius, and
Cloth Hall) in Krakow, Poland were reported by Salmon et al. (2000). Measurements were
made for up to 46 h and up to 33 days. Air exchange measurements were only made for the
Matejko Museum and Wawel Castle. Both were naturally ventilated. However, the summertime
AER for the Matejko Museum was more than twice that of the Wawel Castle site, 1.26 to
1.44 h"1 compared to 0.56 to 0.66 h"1. The highest indoor O3 concentrations were noted at the
AX3-200
-------
September, 1992
Figure AX3-90. Air exchange rates and outdoor and indoor O3 concentrations during the
fall at a telephone switching station in Burbank, CA.
Source: Weschleretal. (1994).
Matejko Museum during the summer. The findings are included in Table AX3-24 for those
locations with reported AERs.
Figure AX3-91 shows O3 and PAN concentrations in a private residence in Germany.
All measurements were made in naturally ventilated rooms.
Johnson (1997) conducted a scripted study using four trained technicians to measure
hourly average O3 concentrations between 07:00 and 19:00 h in Los Angeles, CA during
September and October 1995. The ratio of the microenvironmental concentrations to the fixed
site monitor on days when the O3 levels >20 ppb were as follows: indoor residence, 0.28; other
locations indoors, 0.18; outdoor near roadways, 0.58; other locations outdoors, 0.59; and
in-vehicle, 0.21. The concentrations indoors and within vehicles varied depending on whether
the windows were opened (higher) or closed (lower) and the use of air conditioning. The lower
outdoor concentrations, particularly near roadways, probably reflect the reaction of O3 with NO
emitted by automobiles.
AX3-201
-------
Table AX3-24. Indoor and Outdoor Ozone Concentrations
Location
Matejko Museum
Outdoors (Town Hall Tower)
Indoors (third floor, west)
Wawel Castle
Outdoors (Loggia)
Indoors (Senator's Hall)
Wawel Castle, outdoors
Wawel Castle, Room 15
Wawel Castle, Senator's Hall
Matejko Museum, outdoors
Matejko Museum, Indoor Gallery
Duration
(hours) AER
26
1.26
43
0.63
31.8
31
31.8
26.9
26.9
Average O3 (ppb)
1988-2000
20
8.5
14.7
2.5
42a
8
7
21b
9
"On Loggia of Piano Nobile Level, high above the street.
bAt street level.
Adapted from Salmon et al. (2000).
A study of the effect of elevation on O3 concentrations found that concentrations increased
with increasing elevation. The ratio of O3 concentrations at street level (3 m) compared to the
rooftop (25 m) was between 0.12 and 0.16, though the actual concentrations were highly
correlated (r = 0.63) (Vakeva et al., 1999). Differential O3 exposures may, therefore, exist in
apartments that are on different floors. Differences in elevation between the monitoring sites in
Los Angeles and street level samples may have contributed to the lower levels measured by
Johnson (1997). Furthermore, since O3 monitors are frequently located on rooftops in urban
settings, the concentrations measured there may overestimate the exposure to individuals
outdoors in streets and parks, locations where people exercise and maximum O3 exposure is
likely to occur.
Chang et al. (2000) conducted a scripted exposure study in Baltimore during the summer of
1998 and winter of 1999, during which 1-h O3 samples were collected by a technician who also
changed his or her activity every hour. The activities chosen were selected to simulate older
AX3-202
-------
50-
0. 40
30 -I
10-
r 0.7
•0.5
1-0.3
•0.1
00:00
12:00
00^00
Time
12:00
Figure AX3-91. Diurnal variation of indoor and outdoor O3 and PAN concentrations
measured in a private residence, Freising, Germany, August 11-12,1995.
Source: Jaboki and Fabian (1997).
(> 65 years) adults, based on those activities reported in the National Human Activity Pattern
Survey (NHAPS). The scripted activities took place in five different microenvironments:
indoor residential (apartment in 24-h air conditioned high rise building), indoor other (restaurant,
post office, hospital, shopping mall, and bingo parlor), outdoor near roadway, outdoor away
from roadway, and inside motor vehicle. Personal O3 exposures were significantly lower in
indoor than the outdoor microenvironments, because more time was spent indoors. Mean
summer concentrations were 15.0 ± 18.3 ppb, with a maximum of 76.3 ppb. Significant
correlation was noted for the indoor nonresidential microenvironments and ambient O3 (r = 0.34
in summer, r = 0.46 in winter), however, the authors indicated that this finding was unclear due
to the low personal/ambient ratios. The personal O3 exposure levels within the outdoor and in-
vehicle microenvironments were significantly correlated with ambient concentration, although
the ratio of personal exposure to ambient levels was less than one, with only the top 5% of the
ratios exceeding one, indicating that the ambient measurements lead to the maximum
concentrations and exposures. The indoor concentrations did not correlate with outdoor
AX3-203
-------
measurements (r = 0.09 and r = 0.05 for summer and winter, respectively). The correlation for
outdoor near roadway and outdoor away from roadway was moderate to high (0.68
-------
11:30 11:45 12:00 12:15
Time
o
12:30 12:45
August 23, 1995
Figure AX3-92. Indoor and outdoor O3 concentration in moving cars.
Source: Jaboki and Fabian (1997).
at a copy rate of 8 pages/mm. There was no detectable change in O3 concentrations from the use
of the laser printer.
The U.S. Environmental Protection Agency (Steiber, 1995) measured O3 concentrations
from the use of three home/office O3 generators. The O3 generators were placed in a 27 m3room
with doors and windows closed and the heating, ventilating and air conditioning system off; the
AER was 0.3 h"1. The units were operated for 90 min. Ozone concentrations at the low output
setting ranged from nondetectable to 14 ppb (averaged output). At the high output setting,
averaged output O3 concentrations exceeded 200 ppb in several cases and had spike
concentrations as high as 480 ppb.
Figure AX3-93 includes PAN indoor/outdoor ratios for 10 museums. Four of the museums
were equipped with HVAC and chemical infiltration systems.
AX3.10.7.4 Factors Affecting Ozone Concentrations Indoors
In the absence of an indoor source, O3 concentrations in indoor environments will depend
on the outdoor concentration, the air exchange rate (AER) or outdoor infiltration, indoor
circulation rate, removal by indoor surfaces, reactions with other indoor pollutants, and
AX3-205
-------
1.0-
0.6-
0.4-
0.2-
0.0
3456
9 10 11
Figure AX3-93.
Indoor/outdoor concentration ratios for PAN at 10 southern California
museums. Code to locations: 1 El Pueblo Historical Park, Los Angeles;
2 Olivas Adobe House, Ventura; 3 Southwest Museum, Los Angeles;
4 Page Museum, Los Angeles; 5 Museum of Natural History,
Los Angeles; 6 Research Library, University of California, Los Angeles;
7 Scott Gallery, Huntington Museum, Pasadena; 8 The J. Paul Getty
Museum, Malibu; 9 Los Angeles County Museum of Art, Los Angeles;
10 Gene Autry Western Heritage Museum, Los Angeles (buffer zone);
11 Gene Autry Museum (Conservation Room).
Source: Hisham and Grosjean (1991).
temperature and humidity. Indoor concentrations generally closely track outdoor O3
concentrations. Limited information on PAN concentrations indoors also indicate that indoor
PAN concentrations track outdoor concentrations (Jakobi and Fabian, 1997; Hisham and
Grosjean, 1991). Since outdoor concentrations of photochemical oxidants are generally higher
during the warmer months, indoor concentrations will also be highest during that time period.
(See discussion on ambient concentrations of O3 earlier in this chapter.)
AX3-206
-------
Air Exchange Rates
Indoor O3 increases with increasing air exchange rate (Gold et al., 1996; Lee et al., 1999;
Jakobi and Fabian, 1997). The AER is the balance of the flow of air in and out of a
microenvironment. Infiltration through unintentional openings in the building envelope is the
dominant mechanism for residential air exchange. Duct systems account for 30% of the total
leakage area of a house. Natural ventilation, airflow through opened windows and doors, also
influences air exchange in residential buildings. Forced or mechanical ventilation is the
dominant mechanism for air exchange in nonresidential buildings.
Air exchange rates vary depending on the temperature differences, wind effects,
geographical region, type of heating/mechanical ventilation system, and building type (U.S.
Environmental Protection Agency, 1997; Weschler and Shields, 2000; Colome et al., 1994;
Johnson et al., 2004). Air exchange rates are generally higher during the summer and lower
during the winter months (Wilson et al., 1996; Murray and Burmaster, 1995; Colome et al.,
1994; Research Triangle Institute, 1990). The Gas Research Institute, Pacific Gas and Electric
Company, San Diego Gas and Electric Company, and Southern California Gas Company
measured the air exchange rates in a subset of 293 randomly selected homes in California as part
of an air pollution monitoring study. The average AER varied by type of heating system (wall
furnaces >forced-air > electric) and building type (multifamily units > single-family units)
(Billick et al., 1984, 1996; Colome et al., 1994).
Howard-Reed et al. (2002) determined that opening a window or exterior door causes the
greatest increase in AERs with differences between the indoor and outdoor temperature being
important when the windows were closed. Johnson and Long (2004) conducted a pilot study to
evaluate the frequency that windows were left open in a community. They found that a visual
2-h survey could be used to estimate the frequency that windows are left open. The occupancy,
season, housing density, absence of air conditioning, and wind speed were factors in whether the
windows were open.
Johnson et al. (2004) conducted a study using scripted ventilation conditions to identify
those factors that affected air exchange inside a test house in Columbus, OH. The test house was
a wood-framed, split-level structure with aluminum siding covering the wood outer walls. The
house had one exterior door located in the front and another at the rear of the house, single-pane
glazed windows, central gas heat, a window air-conditioning unit, and ceiling fans in three
AX3-207
-------
rooms. Eighteen scenarios with unique air flow conditions were evaluated to determine the
effect on the AER. The elements of the scenarios included: exteriors doors open/closed, interior
doors open/closed, heater on/off, air conditioner on/off, a ceiling fans on/off The lower level
was sealed off during the study. The various scenarios were evaluated during the winter season.
The average AER for all scenarios ranged from 0.36 to 15.8 h"1. When all windows and doors
were closed, the AER ranged from 0.36 to 2.29 h"1 (0.77 h"1 geometric mean). When at least one
exterior door or window was open the AER ranged from 0.5 to 15.8 h"1 (1.98 h"1 geometric
mean).
Williams et al. (2003a, 2003b) reported air exchange rates ranging from 0.001 h"1 to
4.87 h"1 (overall arithmetic mean of 0.72 h"1) in houses in the Research Triangle Park area in
North Carolina. Air exchange measurements were made in 37 homes as part of a year long study
PM panel study.
Air exchange rates for homes in Houston, TX, Los Angeles County, CA, and Elizabeth, NJ
were reported by Meng et al. (2005) as part of the Relationship of Indoor, Outdoor and Personal
Air (RIOPA) study. The RIOPA study was designed to determine indoor (residual), outdoor,
and personal exposure to several classes of pollutants. Approximately 100 homes from each of
the areas were sampled across all four seasons. The mean air exchange rate for the Los Angeles
County homes was 1.22 h"1, similar to the air exchange rate (1.51 h"1) homes in Los Angeles
previously reported by Wilson et al. (1996). The mean air exchange rate for Houston and
Elizabeth was 0.71 h"1 and 1.22 h"1, respectively. Air exchange rates for New Jersey were
higher than other reported values in the northeast region. The authors attributed these
differences to differences in the age of housing stock in the various areas.
Chan et al. (2005) compared air leakage measurements for more than 70,000 houses across
the United States, classified as low-income households, energy program houses, and convention
houses, to the building size, construction date and various construction characteristics, and
geographical location. The construction date and building size were the two most significant
predictors of leakage areas. Older and smaller houses had higher normalized leakage areas than
the newer and larger houses. Based on their evaluation of new and older residential
constructions, Sherman and Matson (1997) found that existing home stock (pre-1980) was quite
leaky with an AER of approximately 20.0 h"1. Newer constructions were considerably more
airtight.
AX3-208
-------
Murray and Burmaster (1995) conducted an analysis of data compiled by Brookhaven
National Laboratories on AERs for 2,844 residential structures in four climatic regions based on
heating degree days. (Region 1: IN, MN, MT, NH, NY1, VT, WI; Region 2: CO, CT, IL, NJ,
NY2, OH, PA, WA; Region 3: CAS, MD, OR, WA; Region 4: AZ, CA4, FL, TX). Data were
also separated by seasons (winter, spring, summer, and fall). The highest overall AERs occurred
during the spring and fall season. However, air exchange rates were variable within and between
seasons and between regions. Data from the warmest region during the summer months should
be reviewed with caution because many of the measurements were made in southern California
where windows were more likely to be open than in other areas of the country where air-
conditioning is used.
Air exchange rates for 49 nonresidential buildings (14 schools, 22 offices, and 13 retail
establishments) in California were reported by Lagus Applied Technology, Inc. (1995). Average
mean (median) AERs were 2.45 (2.24), 1.35 (1.09), and 2.22 (1.79) h'1 for schools, offices, and
retail establishments, respectively. Air infiltration rates for 40 of the 49 buildings were 0.32,
0.31, and 1.12 h'1 for schools, offices, and retail establishments, respectively. Air exchange
rates for 40 nonresidential buildings in Oregon and Washington (Turk et al., 1989) averaged
1.5 (1.3) h'1 (mean median). The geometric mean of the AERs for six garages was 1.6 h'1 (Marr
et al., 1998). Park et al. (1998) reported AERs for three stationary cars (cars varied by age)
under different ventilation conditions. Air exchange rates ranged from 1.0 to 3.0 h'1 for
windows closed and fan off, 13.3 to 23.5 h'1 for windows opened and fan off, 1.8 to 3.7 h'1 for
windows closed and fan on recirculation (two cars tested), and 36.2 to 47.5 h'1 for windows
closed and fan on fresh air (one car tested). An average AER of 13.1 h'1 was reported by Ott
et al. (1992) for a station wagon moving at 20 mph with the windows closed.
Ozone Removal Processes
The most important removal process for O3 in the indoor environment is deposition on and
reaction with indoor surfaces. The rate of deposition is material specific. The removal rate
indoors will depend on the indoor dimensions, surface coverings, and furnishings. Smaller
rooms generally have larger surface-to-volume ratios (A/V) and remove O3 faster. Fleecy
materials, such as carpets, have larger A/V ratios and remove O3 faster than smooth surfaces
(Weschler, 2000). O3 can react with carpet, decreasing O3 concentrations and increasing
AX3-209
-------
emissions of formaldehyde, acetaldehyde, and other C5-C10 aldehydes. Off-gassing of
4-phenylcyclohexene, 4-vinylcyclohexene, and styrene was reduced (Weschler et al., 1992). The
rate of O3 reaction with carpet diminishes with cumulative O3 exposure (Morrison and Nazaroff,
2000, 2002). Reiss et al. (1995) reported significant quantities of acetic acid and smaller
quantities of formic acid off-gassing from O3 reactions with latex paint. The emission rate also
was relative humidity-dependent, increasing with higher relative humidity. Klen0 et al. (2001)
evaluated O3 removal by several aged (1- to 120-month) but not used building materials (nylon
carpet, linoleum, painted gypsum board, hand polished stainless steel, oiled beech parquet,
melamine-coated particle board, and glass plate). Initially, O3 removal was high for all
specimens tested with the exception of the glass plate. Ozone removal for one carpet specimen
and the painted gypsum board remained high throughout the study. For the oiled beech parquet
and melamine-coated particle board, O3 removal leveled off to a moderate rate. Morrison et al.
(1998) reported that small amounts of O3 (-9%) are removed by lined ductwork of ventilation
systems. The removal efficiency decreases with continued exposure to O3. Unlined ductwork is
less efficient in removing O3. Ozone is scavenged by fiberglass insulation (Liu and Nazaroff,
2001). More O3 was scavenged (60 to 90%) by fiberglass that had not been previously exposed.
Table AX3-25 lists the removal rates for O3 in different indoor environments.
Jaboki and Fabian (1997) found the O3 decays exponentially. PAN decay is dependent on
room temperature and possibly the properties and structure of the materials it comes in contact
with. They examined O3 and PAN decay in several closed rooms and in a car after a period of
intensive ventilation. Figure AX3-94 plots the O3 and PAN decay rates in these environments.
Several studies have examined factors within homes that may scavenge O3 and lead to
decreased O3 concentrations (Lee et al., 1999; Wainman et al., 2000; Weschler and Shields,
1997). These reactions produce related oxidant species while reducing indoor O3 levels. Lee
et al. (1999) studied 43 homes in Upland, CA in the Los Angeles metropolitan region and
reported that O3 declined faster in homes with more bedrooms, greater amounts of carpeting, and
lower ceilings (all of which alter the A/V ratio) and with the use of air conditioning. Homes
with air conditioning had indoor/outdoor (I/O) ratios of 0.07, 0.09, and 0.11. Homes without air
conditioning had an I/O ratio of 0.68 ±0.18. Closed windows and doors combined with the use
of an air cleaner resulted in an I/O ratio of 0.15. The O3 I/O ratio was < 0.2 in homes with gas
stoves.
AX3-210
-------
Table AX3-25. Rate Constants (h l) for the Removal of Ozone by Surfaces in Different
Indoor Environments
Indoor environment
Aluminum Room ( 1 1 . 9 m3)
Stainless Steel Room (14.9m3)
Bedroom (40. 8m3)
Office (55.2m3)
Home (no forced air)
Home (forced air)
Department Store
Office (24. 1m3)
Office (20.7 m3)
Office/Lab
Office/Lab
Office/Lab
13 Buildings, 24 ventilation systems
Museum
Museum
Office/Lab
Office/Lab
Office
Lab
Cleanroom
Telephone Office
43 Homes
Surface Removal Rate,
k,, (A/V), h-1
3.2
1.4
7.2
4.0
2.9
5.4
4.3
4.0
4.3
4.3
3.2
3.6
3.6a
4.3
4.3
4
3.2
2.5
2.5
7.6
0.8- 1.0b
2.8±1.3
Reference
Mueller et at. (1973)
Sabersky etal. (1973)
Thompson et al. (1973)
Allen etal. (1978)
Shair and Heitner( 1974)
Shair(1981)
Nazaroff and Cass ( 1986)
Weschler etal. (1989)
Weschler et at. (1994)b
Lee etal. (1999)
a A/V = assumes surface area to volume ratio = 2.8 m '
b Large office, small A/V
Source: Weschler (2000).
Ozone Removal Through Chemical Processes
Ozone chemical reactions in the indoor environment are analogous to those reactions
occurring in the ambient air (See Annex AX2). Ozone reactions with unsaturated VOCs in the
indoor environment are dependent on the O3 indoor concentration, the indoor temperature and, in
most cases, the air exchange rate/ventilation rate and mixing factor. The air exchange rate
AX3-211
-------
PAN - residence, parquet
10
20
30 40 50 60 70 80
Elapsed Time (minutes)
90 100 110 120
Figure AX3-94. Ozone decay processes versus time measured for several indoor rooms.
Room temperature in all cases was -27 °C and 29 °C in the car.
Adapted from: Jaboki and Fabian (1997).
determines the amount of time available for chemical reactions to take place. At low air
exchange rate, the residence time for the pollutants is longer, the reaction time is greater, and the
concentration of the products produced by O3 chemistry is greater (Weschler and Shields, 2000,
2003). Since O3 is primarily an outdoor pollutant, the air exchange rate will influence the
amount of O3 occurring indoors.
Most unsaturated VOCs in the indoor environment are terpenes or terpene-related
compounds from cleaning products, air fresheners, and wood products. Some of the reaction
products may more negatively impact human health and artifacts in the indoor environment than
their precursors (Wolkoff et al., 1999; Wilkins et al., 2001; Weschler et al., 1992; Weschler and
Shields, 1997; Rohr et al., 2002; N0jgaard et al., 2005). The reactions products of O3 and
terpenes are Criegee biradicals, nitrate radicals, and peroxyacyl radicals. Secondary reaction of
AX3-212
-------
the pollutants may form hydroxy, alkyl, alkylperoxy, hydroperoxy, and alkoxy radicals. The
reaction of O3 with alkenes can produce aldehydes, ketones, and organic acids (Weschler and
Shields, 2000; Weschler et al., 1992).
The indoor chemistry of O3 is described by the following equations. The initial reaction
produces an ozonide which rapidly decomposes into one of the two possible combinations.
O3 + R1C(O)R2C = CR3R4 -» ozonide (AX3-6)
ozonide ^- R!C(O)R2 + [R3R4C»OO»]* (AX3-7)
or
ozonide -> [R^C • OO •]* + R3C(O)R4 (AX3-8)
The biradical (*) then rearranges or reacts to produce the highly reactive intermediate products
(hydroxy, hydroperoxy, and alkylperoxy radicals) and stable products (aldehydes, ketones and
organic acids).
Hydroxy radicals formed from the reaction of O3 with VOCs are lost to further reaction
with VOCs. For each molecule of O3 consumed, approximately one molecule of the hydroxy
radical is produced. The hydroxy radical also is formed through the reaction of nitric oxide and
hydroperoxy, and other intermediate products formed from the reaction of O3 with unsaturated
VOCs (Sarwar et al., 2002; Orzechowska and Paulson, 2002). The hydroxy radical can react
with various nitrogen compounds, sulfur dioxide, carbon monoxide and other compounds to
produce significantly more toxic compounds. In studies by Pick et al. (2003, 2004), the
formation of norpinonic and pinonic acid in a ventilation system injected with a-pinene in the
presence of O3 was reported to be almost exclusively the result of oxidation by hydroxy radicals.
Fan et al. (2003) attributed the formation of some secondary organic aerosols from the reaction
of O3 with 23 VOCs (as toluene) to reactions with hydroxy radicals. Van den Bergh et al. (2000)
found that formaldehyde, acetaldehyde, acetone, campholenealdehyde, and pinonaldehyde are
generated from the reaction of a-pinene with hydroxy radicals. A list of the VOCs occurring in
AX3-213
-------
the indoor environment known to react with O3 and OH radicals is found in Weschler (2000) and
Nazaroff and Weschler (2004).
The reaction between O3 and terpenes has been shown to increase the concentration of
indoor particles (Weschler and Shields, 1999, 2003; Weschler, 2004; Clausen et al., 2001; Fan
et al., 2003; Wainman et al., 2000). Sarwar et al. (2002) suggested that the hydroxy radical
reacts with terpenes to produce products with low vapor pressures that contribute to fine particle
growth. The acidity of particles was found to enhance the yield of secondary organic aerosols
when a-pinene ozonolysis was carried out in the presence of ammonium sulfate and sulfuric
acid. There was almost a 40% increase in organic carbon particles when a-pinene ozonolysis
occurred in the presence of sulfuric acid aerosols compared to ammonium sulfate aerosols
(linuma et al., 2004). Rohr et al. (2002, 2003) measured particle formation in a plexiglas
chamber as part of a mouse bioassay study. They found an increase in ultrafine particle numbers
as the result of O3 and a-pinene reactions. When O3 was introduced into the test chamber,
particle concentrations increased to >107 particles/cm3. Clausen found that the reaction of
limonene vapor with O3 increased the particle concentration to 3 x 10s from a background
concentration of <103 particles/cm3. Poupard et al. (2005) and Blondeau et al. (2005) found a
negative correlation between O3 and particle concentration in eight school buildings in France.
The researchers assumed that the increased particle concentration and decreased O3
concentration was likely the result of homogeneous processes involving O3. However, the
assumption could not be verified because the particles were not analyzed for chemical
composition.
Weschler et al. (1992) suggested that the reaction been O3 and NO2 in the indoor
environment may be a significant source of HNO3. When there are elevated concentrations of
both O3 and NO2 in the indoor environment, the following reaction sequence may occur:
O3 + NO2 - NO3 + O2
NO3 + NO2 ^ N2O5
(AX3-9)
N2O5 + Hap-* 2HNO3
NO3 + ORG - HNO3 + ORG
AX3-214
-------
PAN and PPN are thermally unstable and will decomposed in the indoor environment to
peroxyacetyl radicals and NO2 (See equation AX3-10). Decomposition and formation of PAN in
the indoor environment is influenced by NO2 and NO.
CH3C(O)OONO2 * CH3C(O)OO + NO2 (AX3-10)
When the concentration ratio of NO/NO2 is greater than 7, less than 10% of the
peroxyaceyl radicals will revert to PAN. Decomposition of PAN is expected to be a relatively
fast process when indoor O3 levels are low and when motor vehicle emissions are large or there
is an indoor source of NOX (Weshcler and Shields, 1997).
Sources and Emissions of Indoor Ozone
Ozone enters the indoor environment primarily through infiltration from outdoors through
cracks and crevices in the building envelope (unintentional and uncontrolled ventilation) and
through building components such as windows and doors and ventilation systems (natural and
controlled ventilation). Natural ventilation is driven by the natural forces of wind and
temperature. Possible indoor sources of O3 are office equipment (photocopiers, facsimile
machines, and laser printers) and air cleaners (electrostatic air filters and precipitators and O3
generators). Generally O3 emissions from photocopiers and laser printers are limited due to
installed filtering systems (Black and Worthan, 1999; Leovic et al., 1996; Aldrich et al., 1995).
However, emissions increase under improper maintenance conditions (Leovic et al., 1996).
Well-maintained photocopiers and laser printers usually emit low levels of O3 by catalytically
reducing the O3 to oxygen (Aldrich et al., 1995). Leovic et al. (1996, 1998) measured O3
emissions from four dry-process photocopiers. Ozone emissions ranged from 1300 to
7900 |ig/h.
Most electrostatic air filters and precipitators are designed to minimize the production
of O3. However, if excessive arcing occurs, the units can emit a significant amount of O3 into
the indoor environment (Weschler, 2000). Niu et al. (2001) measured O3 emissions from 27 air
cleaners that used ionization processes to remove particulates. The test were conducted in a
2 x 2 x 1.60 m3 stainless steel environmental chamber. The tests were terminated after 1.5 h
if no increase in O3 concentration was noted. If an increase in O3 was noted, the test was
AX3-215
-------
continued, in some cases, for up to 20 h. Most of the evaluated units emitted no O3 or only
small amounts. Five units were found to emit O3 ranging from 56 to 2757 |ig/h.
There are many brands and models of O3 generators commercially available. The amount
of O3 emitted by each unit depends on the size of the unit. Ozone emission rates have been
reported to range from tens to thousands of micrograms per hour (Weschler, 2000; Steiber,
1995). Ozone emissions supposedly can be regulated using the units control features. However,
available information suggests that O3 output may not be proportional to the control setting.
Some units are equipped with a sensor that automatically controls O3 output by turning the unit
on and off. The effectiveness of the sensor is unknown (U.S. Environmental Protection
Agency, 2002).
Peroxyacyl nitrates (PAN and PPN) have no known direct emission sources and are
formed in the atmosphere from the reaction of NO2 and hydrocarbons (Grosjean et al., 1996).
Peroxyacyl nitrates primarily occur in the indoor environment from infiltration through the
building envelop and through openings in the building envelopment. Peroxyacyl nitrates also
may be formed in the indoor environment through radical chemistry. PAN may be formed from
the reaction of the OH- or NO3 with acetaldehyde to form the acetyl radical, CH3CO. The acetyl
radical then reacts with oxygen to form and acetylperoxy radical which reacts with NO2 to form
PAN.
OH«(or NO3) + CH3CHO -> CH3CO (AX3-11)
CH3CO + O2 -» CH3C(O)OO (AX3-12)
CH3C(0)00- + N02 -> CH3C(0)00 NO2 (AX3-13)
PPN is formed from when the reaction of the OH- with propionaldehyde (Weschler and
Shields, 1997).
AX3.10.8 Trends in Concentrations within Microenvironments
There have not been sufficient numbers of measurements of personal exposures or indoor
O3 concentrations to directly document trends over time or location. However, since O3
AX3-216
-------
concentrations in all microenvironments are primarily derived from ambient sources, trends in
ambient air concentrations should reflect trends in personal exposure unless there are differences
in activity patterns over time or locations. Overall, a significant downward trend in ambient O3
concentrations has occurred from 1980 in most locations in the United States, although the trend
in the latter part of the 1990s suggests that continued improvements in air quality may have
leveled off. Greater declines in ambient O3 concentrations appear to have occurred in urban
centers than in rural regions. The decline in daily and weekly average O3 concentrations from
1989 to 1995 in rural regions was 5% and 7%, respectively (Wolff et al., 2001; Lin et al., 2001;
Holland et al., 1999). A detailed discussion of O3 trends appears earlier in this annex.
AX3.10.9 Characterization of Exposure
AX3.10.9.1 Use of Ambient Ozone Concentrations
The use of ambient air monitoring stations is still the most common surrogate for assigning
exposure in epidemiological studies. Since the primary source of O3 exposure is the ambient air,
monitoring concentration data would provide the exposure outdoors while exercising, a potential
important exposure to evaluate in epidemiological studies as well as a relative assignment of
exposure with time if the concentration were uniform across the region; the time-activity pattern
were the same across the population; and the housing characteristics, such as ventilation rates
and the O3 sinks contributing to its indoor decay rates, were constant for the study area. Since
these factors vary by population and location there will be errors in not only the magnitude of the
total exposure, but also in the relative total exposure assignment based solely on ambient
monitoring data. As discussed earlier in this section, spatial differences in O3 concentrations
within a city and between the height of the monitor and the breath zone (1 to 2 m) exist,
increasing uncertainties. The potential exists to obtain more complete exposure assignments for
both individuals and populations by modeling O3 exposure based on ambient air concentration to
account for spatial variations outdoors and for time spent indoors, provided housing
characteristics and activity patterns can be obtained. For cohort studies, measurement of
personal O3 exposures using passive monitors is also possible.
The potential for error in determining pollutant exposure was expressed by Krzyzanowski
(1997), who indicated that while the typical estimate of exposure in epidemiological studies is
"an average concentration of the pollutant calculated from the data routinely collected in the area
AX3-217
-------
of residence of the studied population. This method certainly lacks precision and, in most cases,
the analyses that use it will underestimate the effect of specific concentrations of a pollutant on
health." It is further stated that when estimating exposure for epidemiological studies it is
important to define: (1) representativeness of exposure or environmental data for the population
at risk, (2) appropriateness of the averaging time for the health outcome being examined, and
(3) the relationship between the exposure surrogate and the true exposure relative to the
exposure-response function used in the risk assessment. Zeger et al. (2000) suggested that the
largest biases will occur because of errors between ambient and average personal exposure
measures. Sheppard et al. (2005a) further indicated that for non-reactive pollutants with non-
ambient sources, exposure variability will introduce a large exposure error. However, for acute
effects, time series studies using ambient concentration measurements are adequate (Sheppard,
2005b).
Numerous air pollutants can have common ambient air sources resulting in strong
correlations among pollutant ambient air concentrations. As a result, some observed
associations between an air pollutant and health effects may be due to confounding by other air
pollutants. Sarnat et al. (2001) found that while ambient air concentrations of some air
pollutants were correlated, personal PM2 5 and several gaseous air pollutant (O3, SO2, NO2, CO,
and exhaust-related VOCs) exposures were not generally correlated. The findings were based on
the results of a multipollutant exposure study of 56 children and elderly adults in Baltimore, MD
conducted during both the summer and winter months. Ambient pollutant concentrations were
not associated with corresponding personal exposures, except for PM2 5. The gaseous pollutants
were found to be surrogates of PM25 and were generally not correlated. The authors concluded
that multipollutant models in epidemiologic studies of PM2 5 may not be suitable, and health
effects attributed to the gaseous pollutants may be the result of PM25 exposure. It should be
noted that the 95th percentile O3 concentrations in the study was lower than 60 ppb, an O3
concentration at which respiratory effects are noted. It would be important to examine whether
O3 is a surrogate for personal PM2 5 at high O3 levels when attributing adverse health effects to
O3 or PM2 5.
Kiinzli et al. (1996) assessed potential lifetime exposure to O3 based on the responses to a
standardized questionnaire completed by 175 college freshmen in California. Questions
addressed lifetime residential history, schools attended, general and outdoor activity patterns,
AX3-218
-------
driving habits and job history. The purpose was to determine what O3 monitoring data to use for
each time period of their lives, the potential correction factor for indoor levels and periods of
high activity to account for differential doses to the lung due to physical activity. The reliability
of the responses was checked by having each respondent complete the questionnaire twice, on
different days, and the results compared. A lifetime O3 exposure history was generated for each
participant and a sensitivity analysis performed to evaluate which uncertainties would cause the
greatest potential misclassification of exposure. Assigned lifetime O3 concentrations from the
nearest monitor yielded highly reliable cumulative values, although the reliability of residential
location decreased with increasing residential locations. Individuals involved in moderate and
heavy exercise could be reliably identified. Such an approach can be used to evaluate health
outcomes associated with chronic exposures to O3.
AX3.10.9.2 Exposure Selection in Controlled Exposure Studies
Ozone exposures in the environment are variable over time due to changes in the ambient
concentrations during the day as the photochemical reactions proceed and also because people
move between microenvironments that have different concentrations (Johnson, 1997).
Exposures are repeated on sequential days since weather conditions that produce O3 can move
slowly through or become stagnant within a region. For simplicity, most controlled-exposure
experiments are conducted at a single concentration for a fixed time period, with a limited
number of studies being repeated on a single individual. Few studies have been conducted using
multipollutants or photochemical agents other than O3, to better represent "real-world" exposures
with the exception of NO2. Studies by Hazucha et al. (1992), Adams (2003), and Adams and
Ollison (1997) examined the effect of varying O3 exposure concentrations on pulmonary
function. A description of, and findings in, the studies appears in Annex AX5 of this document.
AX3.10.9.3 Exposure to Related Photochemical Agents
Exposures to other related photochemical agents have not been measured using personal
samplers nor are these agents routinely measured at O3 monitoring stations. Photochemical
agents produced in the ambient air can penetrate indoors and react with other pollutants to
produce other potentially irritating compounds. Reiss et al. (1995) reported that organic acids,
aldehydes and ketones were produced indoors by reactions of O3 with VOCs. The produced
compounds included oxidants that can be respiratory irritants. The indoor concentrations were
AX3-219
-------
dependent upon the O3 concentrations indoors and the AER within the building. Weschler and
Shields (1997) summarized indoor air chemical reactions that depend directly or indirectly on
the presence of O3. They indicated that O3 concentrations are lower indoors than outdoors partly
because of gas-phase reactions that produce other oxidants in an analogous fashion to
photochemical smog in ambient air. The production of these species indoors is a function of the
indoor O3 concentration and the presence of the other necessary precursors, VOCs, and NO2,
along with an optimal AER. A variety of the photochemical oxidants related to O3 that are
produced outdoors, such as PAN and PPN, can penetrate indoors. These oxidants are thermally
unstable and can decompose indoors to peroxacetyl radicals and NO2 through thermal decay.
PAN removal increases with increasing temperature, and at a given temperature, with increasing
NO/NO2 concentration ratio (Grosjean et al., 2001). Other free radicals that can form indoors
include HO and HO2'. These free radicals can produce compounds that are known or suspected
to be irritating. Little is known about exposure to some of these agents, as not all have been
identified and collection and analytical methodologies have not be developed for their routine
determination. Lee et al. (1999) reported that homogeneous (gas phase) and heterogenous (gas
phase/solid surface) reactions occur between O3 and common indoor air pollutants such as NO
and VOCs to produce secondary products whose production rate depends on the AER and
surface area within the home. Wainman et al. (2000) found that O3 reacts indoors with
J-limonene, emitted from air fresheners, to form fine particles in the range of 0.1 to 0.2 |im and
0.2 to 0.3 |im. The indoor process also produces compounds that have been identified in the
ambient atmosphere. These species, plus others that may form indoors from other terpenes or
unsaturated compounds can present an additional exposure to oxidants, other than O3, at higher
concentrations than present in ambient air, even as the O3 concentration is being reduced
indoors.
The announcement of smog alerts or air quality indices may influence individuals to alter
behaviors (avoidance behavior). Neidell (2004), in his evaluation of the effect of pollution on
childhood asthma, examined the relationship between the issuance of smog alerts or air quality
indices for several counties in California and hospital admissions for asthma in children under
age 18 years (not including newborns). Smog alerts are issued in California on days when O3
concentrations exceed 200 ppb. There was a significant reduction in the number of asthma-
AX3-220
-------
related hospital admissions in children ages 1 to 12 years on smog alert days, indicating that
avoidance behavior might be present on days of high O3 concentrations.
Ozone exposure modeling has been conducted for the general population and sensitive
subgroups. The pNEM/O3 model takes into consideration the temporal and spatial distribution
of people and O3 throughout the area of consideration, variations in O3 concentrations within
microenvironments, and the effects of exertion/exercise (increased ventilation) on O3 uptake.
The pNEM/O3 model consists of two principal parts: the cohort exposure program and the
exposure extrapolation program. The methodology incorporated much of the general framework
described earlier in this section on assessing O3 exposure and consists of five steps: (1) define
the study area, population of interest, subdivisions of the study area, and exposure period;
(2) divide population of interest into a set of cohorts; (3) develop exposure event sequence for
each cohort for the exposure period; (4) estimate pollutant concentration and ventilation rate for
each exposure event; and (5) extrapolate cohort exposures to population of interest (U.S.
Environmental Protection Agency, 1996b).
There are three versions of the pNEM/O3 model: general population (Johnson et al.,
1996a), outdoor workers (Johnson et al., 1996b), and outdoor children (Johnson et al., 1996c,
1997). These three versions of the model have been applied to nine urban areas. The model also
has been applied to a single summer camp (Johnson et al., 1996c). The general population
version of the model uses activity data from the Cincinnati Activity Diary Study (CADS;
Johnson, 1989). Time-activity studies (Wiley et al., 1991a; Johnson, 1984; Linn et al., 1993;
Shamoo et al., 1991; Goldstein et al., 1992; Hartwell et al., 1984) were combined with the CADS
data for the outdoor worker version of the model. Additional time-activity data (Goldstein et al.,
1992; Hartwell et al., 1984; Wiley et al., 1991a,b; Linn et al., 1992; Spier et al., 1992) were also
added to CADS for the outdoor children of the model (U.S. Environmental Protection Agency,
1996b). Home-work commuting patterns are based on information gathered by the U.S. Census
Bureau. Ozone ambient air concentration data from monitoring stations were used to estimate
the outdoor exposure concentrations associated with each exposure event. Indoor O3 decay rate
is assumed to be proportional to the indoor O3 concentration. An algorithm assigns the
equivalent ventilation rate (EVR) associated with each exposure event. The outdoor children
model uses an EVR-generator module to generate an EVR value for each exposure event based
on data on heart rate by Spier et al. (1992) and Linn et al. (1992). The models produce exposure
AX3-221
-------
estimates for a range of O3 concentrations at specified exertion levels. The models were used to
estimate exposure for nine air quality scenarios (U.S. Environmental Protection Agency, 1996b).
Korc (1996) used the REHEX-II model, a general purpose air pollution exposure model
based on a microenvironmental approach modified to account for the influence of physical
activity along and spatial and the temporal variability of outdoor air pollution. Ozone exposure
was estimated by demographic groups across 126 geographic subregions for 1980 to 1982, and
for 142 geographic subregions for 1990 to 1992. Simulation results were determined for
population race, ethnicity, and per capita income and included indoor, in-transit, and outdoor
microenvironments. Exposure modeling was stratified by age because of differences in time-
activity patterns. Exposure distributions by regional activity pattern data were not considered,
rather it was assumed that all individuals within a county had the same exposure distribution by
race, ethnicity, and socioeconomic status. Model results for southern California indicated that
the segment of the population with the highest exposures were children 6 to 11 years old.
Individuals living in low income districts may have greater per capita hours of exposure to O3
above the NAAQS than those living in higher income districts. The author indicated that O3
exposure differences by race and ethnicity have declined over time. The noninclusion of details
on activity patterns for different populations in the model limit the extrapolations that can be
made from the model results.
Children appear to have higher exposures than adults and the elderly. Asthmatics appear to
ventilate more than healthy individuals, but tend to protect themselves by decreasing their
outdoor exercise (Linn et al., 1992). Additional data are still needed to identify and better define
exposures to potentially susceptible populations and improve exposure models for the general
population and subpopulation of concern.
AX3-222
-------
REFERENCES
Adams, W. C. (1993) Measurement of breathing rate and volume in routinely performed daily activities [final
report]. Sacramento, CA: California Environmental Protection Agency, Air Resources Board; contract no.
A033-205.
Adams, W. C. (2003) Comparison of chamber and face mask 6.6-hour exposure to 0.08 ppm ozone via square-wave
and triangular profiles on pulmonary responses. Inhalation Toxicol. 15: 265-281.
Adams, W. C.; Ollison, W. M. (1997) Effects of prolonged simulated ambient ozone dosing patterns on human
pulmonary function and symptomatology. Presented at: 90th annual meeting of the Air & Waste
Management Association; June; Toronto, Ontario, Canada. Pittsburgh, PA: Air & Waste Management
Association; paper no. 97-MP9.02.
Adams, R. M.; Hamilton, S. A.; McCarl, B. A. (1985) An assessment of the economic effects of ozone on U.S.
agriculture. J. AirPollut. Control Assoc. 35: 938-943.
Adams, R. M.; Glyer, J. D.; Johnson, S. L.; McCarl, B. A. (1989) A reassessment of the economic effects of ozone
on U.S. agriculture. JAPCA 39: 960-968.
Ainsworth; Haskell, W. L.; Leon, A. S.; Jacobs, D. R., Jr.; Montoye, H. J.; Sallis, J. F.; Paffenbarger, R. S., Jr.
(1993) Compendium of physical activities: classification of energy costs of human physical activities. Med.
Sci. Sports Exer. 25: 71-80.
Akland, G. G.; Hartwell, T. D.; Johnson, T. R.; Whitmore, R. W. (1985) Measuring human exposure to carbon
monoxide in Washington, D.C., and Denver, Colorado, during the winter of 1982-1983. Environ. Sci.
Technol. 19: 911-918.
Aldrich, F. D.; Brooks, B. O.; Davis, W. F.; DeBroy, J. A.; Li, S. T.; Schimke, R.; Utter, G. M. (1995) Health and
productivity impact of organic volatiles from electronic office equipment. Presented at: Pacific Rim
conference on occupational and environmental health; October; Sydney, Australia.
Allen, R. J.; Wadden, R. A.; Ross, E. D. (1978) Characterization of potential indoor sources of ozone. Am. Ind. Hyg.
Assoc. J. 39: 466-471.
Aneja, V. P.; Li, Z. (1992) Characterization of ozone at high elevation in the eastern United States: trends, seasonal
variations, and exposure. J. Geophys. Res. [Atmos.] 97: 9873-9888.
Avol, E. L.; Navidi, W. C.; Colome, S. D. (1998a) Modeling ozone levels in and around southern California homes.
Environ. Sci. Technol. 32: 463-468.
Avol, E. L.; Navidi, W. C.; Rappaport, E. B.; Peters, J. M. (1998b) Acute effects of ambient ozone on asthmatic,
wheezy, and healthy children. Cambridge, MA: Health Effects Institute; research report no. 82.
Beine, H. J.; Jaffe, D. A.; Herring, J. A.; Kelley, J. A.; Krognes, T.; Stordal, F. (1997) High-latitude springtime
photochemistry. Part I: NOX, PAN and ozone relationships. J. Atmos. Chem. 27: 127-153.
Bernard, N. L.; Gerber, M. J.; Astre, C. M.; Saintot, M. J. (1999) Ozone measurement with passive samplers:
validation and use for ozone pollution assessment in Montpellier, France. Environ. Sci. Technol.
33:217-222.
Berntsen, T. K.; Isaksen, I. S. A.; Myhre, G.; Fuglestvedt, J. S.; Stordal, F.; Larsen, T. A.; Freckleton, R. S.; Shine,
K. P. (1997) Effects of anthropogenic emissions on tropospheric ozone and its radiative forcing.
J. Geophys. Res. [Atmos.] 102: 28,101-28,126.
Berry, C. R. (1964) Differences in concentrations of surface oxidant between valley and mountaintop conditions in
the southern Appalachians. J. AirPollut. Control Assoc. 14: 238-239.
Bey, I.; Jacob, D. J.; Yantosca, R. M.; Logan, J. A.; Field, B.; Fiore, A. M.; Li, Q.; Liu, H.; Mickley, L. J.; Schultz,
M. G. (2001a) Global modeling of tropospheric chemistry with assimilated meteorology: model description
and evaluation. J. Geophys. Res. (Atmos.) 106: 23,073-23,095.
Bey, I.; Jacob, D. J.; Logan, J. A.; Yantosca, R. M. (200Ib) Asian chemical outflow to the Pacific in spring: origins,
pathways, and budgets. J. Geophys. Res. [Atmos.] 106: 23,097-23,113.
Biller, W. F.; Feagans, T. B.; Johnson, T. R.; Duggan, G. M.; Paul, R. A.; McCurdy, T.; Thomas, H. C. (1981)
A general model for estimating exposure associated with alternative NAAQS. Presented at: 74th annual
meeting of the Air Pollution Control Association; June; Philadelphia, PA. Pittsburgh, PA: Air Pollution
Control Association; paper no. 81-18.4.
Billick, I.; Johnson, D.; Moschandreas, D.; Relwani, S. (1984) An investigation of operational factors that influence
emission rates from gas appliances. In: Berglund, B.; Lindvall, T.; Sundell, J., eds. Indoor air. Volume 4:
chemical characterization and personal exposure. Proceedings of the 3rd international conference on indoor
air quality and climate; August; Stockholm, Sweden. Stockholm, Sweden: Swedish Council for Building
Research; pp. 181-187.
AX3-223
-------
Billick, I. H.; Behrens, D. W.; Becker, E. W.; Garrison, C. A.; Wilson, A. L.; Colome, S. D.; Tian, Y. (1996)
California residential indoor air quality study: an overview. In: Proceedings of the 1995 international gas
research conference; November 1995; Cannes, France. Rockville, MD: Government Institutes, Inc.;
pp. 1707-1716.
Black, M. S.; Worthan, A. W. (1999) Emissions from office equipment. In: Raw, G.; Aizlewood, C.; Warren, P.,
eds. Indoor Air '99, v. 2. London, England: Construction Research Communications Ltd.; pp. 454-459.
Black, D. R.; Harley, R. A.; Hering, S. V.; Stolzenburg, M. R. (2000) A new, portable, real-time monitor. Environ.
Sci. Technol. 34:3031-3040.
Blanchard, C. L.; Tanenbaum, S. J. (2003) Differences between weekday and weekend air pollutant levels in
southern California. J. Air Waste Mange. Assoc. 53: 816-828.
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.
Bohm, M.; McCune, B.; Vandetta, T. (1991) Diurnal curves of troposheric ozone in the western United States.
Atmos. Environ. Part A 25: 1577-1590.
Bottenheim, J. W.; Barrie, L. A.; Atlas, E.; Heidt, L. E.; Niki, H.; Rasmussen, R. A.; Shepson, P. B. (1990)
Depletion of lower tropospheric ozone during Arctic spring: The Polar Sunrise Experiment 1988.
J. Geophys. Res. (Atmos.) 95: 18,555-18,568.
Bottenheim, J. W.; Barrie, L. A.; Atlas, E. (1993) The partitioning of nitrogen oxides in the lower Arctic troposphere
during Spring 1988. J. Atmos. Chem. 17: 15-27.
Brauer, M.; Brook, J. R. (1997) Ozone personal exposures and health effects for selected groups residing in the
Fraser Valley. In: Steyn, D. G.; Bottenheim, J. W., eds. The Lower Fraser Valley Oxidants/Pacific '93 Field
Study. Atmos. Environ. 31: 2113-2121.
Browell, E. V.; Hair, J. W.; Butler, C. F.; Grant, W. B.; De Young, R. J.; Fenn, M. A.; Brackett, V. G.; Clayton,
M. B.; Brasseur, L. A.; Harper, D. B.; Ridley, B. A.; Klonecki, A. A.; Hess, P. G.; Emmons, L. K.; Tie, X.;
Atlas, E. L.; Cantrell, C. A.; Wimmers, A. J.; Blake, D. R.; Coffey, M. T.; Hannigan, J. W.; Dibb, J. E.;
Talbot, R. W.; Flocke, F.; Weinheimer, A. J.; Fried, A.; Wert, B.; Snow, J. A.; Lefer, B. L. (2003) Ozone,
aerosol, potential vorticity, and trace gas trends observed at high-latitudes over North America from
February to May 2000. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/2001JD001390.
Burke, J. M.; Zufall, M. J.; Ozkaynak, H. (2001) A population exposure model for paniculate matter: case study
results forPM25 in Philadelphia, PA. J. Exposure Anal. Environ. Epidemiol. 11: 470-489.
Carroll, R. J.; Chen, R.; George, E. L; Li, T. H.; Newton, H. J.; Schmiediche, H.; Wang, N. (1997a) Ozone exposure
and population density in Harris County, Texas. J. Am. Stat. Assoc. 92: 392-404.
Carroll, R. J.; Chen, R.; George, E. L; Li, T. H.; Newton, H. J.; Schmiediche, H.; Wang, N. (1997b) Ozone exposure
and population density in Harris County, Texas (rejoinder). J. Am. Stat. Assoc. 92: 414-415.
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, W. R.; Nazaroff, W. W.; Price, P. N.; Sohn, M. D.; Gadgil, A. J. (2005) Analyzing a database of residential
air leakage in the United States. Atmos. Environ. 39: 3445-3455.
Chang, L.-T.; Koutrakis, P.; Catalano, P. J.; Suh, H. H. (2000) Hourly personal exposures to fine particles and
gaseous pollutants—results from Baltimore, Maryland. J. Air Waste Manage. Assoc. 50: 1223-1235.
Chao, C. Y. (2001) Comparison between indoor and outdoor air contaminant levels in residential buildings from
passive sampler study. Build. Environ. 36: 999-1007.
Chen, L.-W. A. (2002) Urban fine paniculate matter: chemical composition and possible origins (dissertation).
College Park, MD: University of Maryland, Department of Chemical Physics. Available from: University
Microfilms, Ann Arbor, MI; AADAA-I3078297.
Chinkin, L. R.; Coe, D. L.; Funk, T. H.; Hafner, H. R.; Roberts, P. T.; Ryan, P. A. (2003) Weekday versus weekend
activity patterns for ozone precursor emissions in California's South Coast Air Basin. J. Air Waste Manage.
Assoc. 53:829-843.
Christakos, G. (2000) Modern spatiotemporal geostatistics. New York, NY: Oxford University Press. (International
Association for Mathematical Geology; studies in mathematical geology: 6).
Christakos, G.; Hristopulos, D. T. (1998) Spatiotemporal environmental health modelling. A tractatus stochasticus.
Boston, MA: Kluwer Academic Publishers.
Christakos, G.; Kolovos, A. (1999) A study of the spatiotemporal health impacts of ozone exposure. J. Exposure
Anal. Environ. Epidemiol. 9: 322-335.
Christakos, G.; Vyas, V. M. (1998a) A composite space/time approach to studying ozone distribution over eastern
United States. Atmos. Environ. 32: 2845-2857.
AX3-224
-------
Christakos, G.; Vyas, V. M. (1998b) A novel method for studying population health impacts of spatiotemporal
ozone distribution. Soc. Sci. Med. 47: 1051-1066.
Clausen, P. A.; Wilkins, C. K.; Wolkoff, P.; Nielsen, G. D. (2001) Chemical and biological evaluation of a reaction
mixture of R-(+)-limonene/ozone: formation of strong airway irritants. Environ. Int. 26: 511-522.
Clench-Aas, I; Bartonova, A.; Behler, T.; Grenskei, K. E.; Sivertsen, B.; Larssen, S. (1999) Air pollution exposure
monitoring and estimating. Part I. Integrated air quality monitoring system. J. Environ. Monit. 1: 313-319.
Code of Federal Regulations. (2000) Appendix D to part 58—Network design for state and local air monitoring
stations (SLAMS), national air monitoring stations (NAMS), and photochemical assessment monitoring
stations (PAMS). C. F. R. 40: pt. 58, app. D.
Coffey, P.; Stasiuk, W.; Mohnen, V. (1977) Ozone in rural and urban areas of New York State: part I. In:
Dimitriades, B., ed. International conference on photochemical oxidant pollution and its control -
proceedings: volume I; September 1976; Raleigh, NC. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Environmental Sciences Research Laboratory; pp. 89-96; report no.
EPA-600/3-77-001a. Available from: NTIS, Springfield, VA; PB-264232.
Colome, S. D.; Wilson, A. L.; Tian, Y. (1994) California residential indoor air quality study. Volume 2. Carbon
monoxide and air exchange rate: an univariate and multivariate analysis. Chicago, IL: Gas Research
Institute; report no. GRI-93/0224.3.
Cooper, O. R.; Moody, J. L. (2000) Meteorological controls on ozone at an elevated eastern United States regional
background monitoring site. J. Geophys. Res. [Atmos.] 105: 6855-6869.
Cressie, N. (1997) Ozone exposure and population density in Harris County, Texas [comment]. J. Am. Stat. Assoc.
92:411-413.
Cristofanelli, P.; Bonasoni, P.; Collins, W.; Feichter, J.; Forster, C.; James, P.; Kentarchos, A.; Kubik, P. W.; Land,
C.; Meloen, J.; Roelofs, G. J.; Siegmund, P.; Sprenger, M.; Schnabel, C.; Stohl, A.; Tobler, L.; Tositti, L.;
Trickl, T.; Zanis, P. (2003) Stratosphere-troposphere exchange: a model and method evaluation. J.
Geophys. Res. 108(D12): 10.1029/2002JD002600.
Delfino, R. J.; Coate, B. D.; Zeiger, R. S.; Seltzer, J. M.; Street, D. H.; 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.
Dickerson, R. R. (1985) Reactive nitrogen compounds in the Arctic. J. Geophys. Res. 90: 10,739-10,743.
Dickerson, R. R.; Huffman, G. J.; Luke, W. T.; Nunnermacker, L. J.; Pickering, K. E.; Leslie, A. C. D.; Lindsey,
C. G.; Slinn, W. G. N.; Kelly, T. J.; Damn, P. H.; Delany, A. C.; Greenberg, J. P.; Zimmerman, P. R.;
Boatman, J. F.; Ray, J. D.; Stedman, D. H. (1987) Thunderstorms: an important mechanism in the transport
of air pollutants. Science (Washington, DC) 235: 460-465.
Douglass, A. R.; Schoeberl, M. R.; Rood, R. B.; Pawson, S. (2003) Evaluation of transport in the lower tropical
stratosphere in a global chemistry and transport model. J. Geophys. Res. [Atmos.] 108(D9):
10.1029/2002JD002696.
Drakou, G.; Zerefos, C.; Ziomas, I. (1995) A preliminary study on the relationship between outdoor and indoor air
pollution levels. Fresenius' Environ. Bull. 4: 689-694.
Drakou, G.; Zerefos, C.; Ziomas, I.; Voyatzaki, M. (1998) Measurements and numerical simulations of indoor O3
and NO-j in two different cases. Atmos. Environ. 32: 595-610.
Evans, G.; Finkelstein, P.; Martin, B.; Possiel, N.; Graves, M. (1983) Ozone measurements from a network of
remote sites. J. AirPollut. Control Assoc. 33: 291-296.
Evans, E. G.; Rhodes, R. C.; Mitchell, W. J.; Puzak, J. C. (1985) Summary of precision and accuracy assessments
for the state and local air monitoring networks: 1982. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Environmental Monitoring Systems Laboratory; report no. EPA-600/4-85-031.
Available from: NTIS, Springfield, VA; PB85-208171.
Faith, R.; Sheshinski, R. (1979) Misspecification of trend in spatial random-function interpolation with application
to oxidant mapping. Philadelphia, PA: Siam Institute for Mathematics and Society; technical report no. 28.
Fan, Z.; Lioy, P.; Weschler, C.; Fiedler, N.; Kipen, H.; Zhang, J. (2003) Ozone-initiated reactions with mixtures of
volatile organic compounds under simulated indoor conditions. Environ. Sci. Technol. 37: 1811-1821.
Federal Register. (1986) Guidelines for estimating exposures. F. R. (September 24) 51: 34,042-34,054.
Pick, J.; Pommer, L.; Nilsson, C.; Andersson, B. (2003) Effect of OH radicals, relative humidity, and time on the
composition of the products formed in the ozonolysis of a-pinene. Atmos. Environ. 37: 4087-4096.
Pick, J.; Nilsson, C.; Andersson, B. (2004) Formation of oxidation products in a ventilation system. Atmos. Environ.
38: 5895-5899.
AX3-225
-------
Fiore, A. M; Jacob, D. I; Bey, I.; Yantosca, R. M; Field, B. D.; Fusco, A. C.; Wilkinson, J. G. (2002a) Background
ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys.
Res. (Atmos.) 107(D15): 10.1029/2001JD000982.
Fiore, A. M.; Jacob, D. J.; Field, B. D.; Streets, D. G.; Fernandes, S. D.; Jang, C. (2002b) Linking ozone pollution
and climate change: the case for controlling methane. Geophys. Res. Lett. 29(19): 10.1029/2002GL015601.
Fiore, A.; Jacob, D. J.; Liu, H.; Yantosca, R. M.; Fairlie, T. D.; Li, Q. (2003a) Variability in surface ozone
background over the United States: implications for air quality policy. J. Geophys. Res. (Atmos.) 108(D24):
10.1029/2003 JD003855.
Fiore, A. M.; Jacob, D. J.; Mathur, R.; Martin, R. V. (2003b) Application of empirical orthogonal functions to
evaluate ozone simulations with regional and global models. J. Geophys. Res. (Atmos.) 108(D14):
10.1029/2002JD003151.
Fitz-Simons, T.; McCluney, L.; Rizzo, M. (2005) Analysis of 2004 ozone data for the ozone NAAQS review
[memorandum to Dr. Joseph Pinto]. Research Triangle Park, NC: U.S. Environmental Protection Agency;
August 22.
Fraigneau, Y. C.; Gonzalez, M.; Coppalle, A. (1995) Dispersion and chemical reaction of a pollutant near a
motorway. Sci. Total Environ. 169: 83-91.
Freijer, J. I.; Bloemen, H. J. T. (2000) Modeling relationships between indoor and outdoor air quality. J. Air Waste
Manage. Assoc. 50: 292-300.
Freijer, J. I.; Bloemen, H. J. T.; de Loos, S.; Marra, M.; Rombout, P. J. A.; Steentjes, G. M.; Van Veen, M. P. (1998)
Modelling exposure of the Dutch population to air pollution. J. Haz. Mat. 61: 107-114.
Fujita, E. M.; Stockwell, W. R.; Campbell, D. E.; Keislar, R. E.; Lawson, D. R. (2003) Evolution of the magnitude
and spatial extent of the weekend ozone effect in California's South Coast Air Basin, 1981-2000. J. Air
Waste Manage. Assoc. 53: 802-815.
Fusco, A. C.; Logan, J. A. (2003) Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
midlatitudes with the GEOS-CHEM model. J. Geophys. Res. (Atmos.) 108: 10.1029/2002JD002742.
Garratt, J. R. (1992) The atmospheric boundary layer. Cambridge, United Kingdom: Cambridge University Press.
(Houghton, J. T.; Rycroft, M. J.; Dessler, A. J., eds. Cambridge atmospheric and space science series).
Georgopoulos, P. G.; Lioy, P. J. (1994) Conceptual and theoretical aspects of human exposure and dose assessment.
J. Exposure Anal. Environ. Epidemiol. 4: 253-285.
Georgopoulos, P. G.; Arunachalam, S.; Wang, S. (1997a) Alternative metrics for assessing the relative effectiveness
of NOX and VOC emission reductions in controlling ground-level ozone. J. Air Waste Manage. Assoc.
47: 838-850.
Georgopoulos, P. G.; Walia, A.; Roy, A.; Lioy, P. J. (1997b) Integrated exposure and dose modeling and analysis
system. 1. Formulation and testing of microenvironmental and pharmacokinetic components. Environ. Sci.
Technol. 31: 17-27.
Georgopoulos, P.; Yonone-Lioy, M. J.; Opeikum, R. E.; Lioy, P. (2002a) Environmental dynamics and human
exposure to copper. Volume I: environmental dynamics and human exposure issues. New York, NY:
International Copper Association.
Georgopoulos, P.; Wang, S. W.; Vyas, V. M.; Lioy, P. J.; Tan, H. C. (2002b) Environmental dynamics and human
exposure to copper. Volume 2: framework and data sources for assessing human exposure to copper in the
United States. New York, NY: International Copper Association.
Georgopoulos, P. G.; Wang, S.-W.; Vyas, V. M.; Sun, Q.; Burke, J.; Vedantham, R.; McCurdy, 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. Exposure Anal. Environ. Epidemiol. 15: 10.1038/sj.jea.7500422.
Gettelman, A.; Holton, J. R.; Rosenlof, K. H. (1997) Mass fluxes of O3, CH4, N2O, and CF2C12 in the lower
stratosphere calculated from observational data. J. Geophys. Res. [Atmos.] 102: 19,149-19,159.
Geyh, A. S.; Wolfson, J. M.; Koutrakis, P.; Mulik, J. D.; Avol, E. L. (1997) Development and evaluation of a small
active ozone sampler. Environ. Sci. Technol. 31: 2326-2330.
Geyh, A. S.; Roberts, P. T.; Lurmann, F. W.; Schoell, B. M.; Avol, E. L. (1999) Initial field evaluation of the
Harvard active ozone sampler for personal ozone monitoring. J. Exposure Anal. Environ. Epidemiol.
9: 143-149.
Geyh, A. S.; Xue, J.; Ozkaynak, H.; Spengler, J. D. (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.
Glen, G.; Lakkadi, Y.; Tippett, J. A.; del Valle-Torres, M. (1997) Development of NERL/CHAD: the National
Exposure Research Laboratory consolidated human activity database. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Research and Development; contract no. 68-D5-0049.
AX3-226
-------
Gold, D. R.; Allen, G.; Damokosh, A.; Serrano, P.; Hayes, C.; Castillejos, M. (1996) Comparison of outdoor and
classroom ozone exposures for school children in Mexico City. J. Air Waste Manage. Assoc. 46: 335-342.
Goldan, P. D.; Trainer, M.; Kuster, W. C.; Parrish, D. D.; Carpenter, J.; Roberts, J. M.; Yee, J. E.; Fehsenfeld, F. C.
(1995) Measurements of hydrocarbons, oxygenated hydrocarbons, carbon monoxide, and nitrogen oxides in
an urban basin in Colorado: implications for emission inventories. J. Geophys. Res. [Atmos.] 100:
22,771-22,783.
Goldstein, B. R.; Tardiff, G.; Hoffnagle, G.; Kester, R. (1992) Valdez health study: summary report. Anchorage,
AK: Alyeska Pipeline Service Company.
Grivet, C. D. (1980) Modeling and analysis of air quality data [dissertation]. Palo Alto, CA: Stanford University.
Available from: University Microfilms International; Ann Arbor, MI; publication no. 80-16,824.
Grosjean, D. (2003) Ambient PAN and PPN in southern California from 1960 to the SCOS97-NARSTO. Atmos.
Environ. 37(suppl. 2): S221-S238.
Grosjean, E.; Grosjean, D.; Fraser, M. P.; Cass, G. R. (1996) Air quality model evaluation data for organics: 3.
Peroxyacetyl nitrate and peroxypropionyl nitrate in Los Angeles air Environ. Sci. Technol. 30: 2704-2714.
Grosjean, E.; Grosjean, D.; Woodhouse, L. F. (2001) Peroxyacetyl nitrate and peroxypropionyl nitrate during SCOS
97-NARSTO. Environ. Sci. Technol. 35: 4007-4014.
Guttorp, P.; Meiring, W.; Sampson, P. D. (1997) Ozone exposure and population density in Harris County, Texas
[comment]. J. Am. Stat. Assoc. 92: 405-408.
Hartwell, T. D.; Clayton, C. A.; Ritchie, R. M.; Whitmore, R. W.; Zelon, H. S.; Jones, S. M.; Whitehurst, D. A.
(1984) Study of carbon monoxide exposure of residents of Washington, DC and Denver, Colorado.
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Research and Development,
Environmental Monitoring Systems Laboratory; report no. EPA-600/4-84-031.
Hayes, S. R. (1989) Estimating the effect of being indoors on total personal exposure to outdoor air pollution.
JAPCA39: 1453-1461.
Hayes, S. R. (1991) Use of an indoor air quality model (IAQM) to estimate indoor ozone levels. J. Air Waste
Manage. Assoc. 41: 161-170.
Hazucha, M. J.; Folinsbee, L. J.; Seal, E., Jr. (1992) Effects of steady-state and variable ozone concentration profiles
on pulmonary function. Am. Rev. Respir. Dis. 146: 1487-1493.
Heikes, B. G.; Kok, G. L.; Walega, J. G.; Lazrus, A. L. (1987) H2O2, O3 and SO2 measurements in the lower
troposphere over the eastern United States during fall. J. Geophys. Res. 92: 915-931.
Helaleh, M. I. H.; Ngudiwaluyo, S.; Korenaga, T.; Tanaka, K. (2002) Development of passive sampler technique for
ozone monitoring. Estimation of indoor and outdoor ozone concentration. Talanta 58: 649-659.
Herring, J. A.; Jaffe, D. A.; Beine, H. J.; Madronich, S.; Blake, D. R. (1997) High-latitude springtime
photochemistry. Part II: sensitivity studies of ozone production. J. Atmos. Chem. 27: 155-178.
Heuss, J. M.; Kahlbaum, D. F.; Wolff, G. T. (2003) Weekday/weekend ozone differences: what can we learn from
them? J. Air Waste Manage. Assoc. 53: 772-788.
Hisham, M. W. M.; Grosjean, D. (1991) Sulfur dioxide, hydrogen sulfide, total reduced sulfur, chlorinated
hydrocarbons and photochemical oxidants in southern California museums. Atmos. Environ. Part 25:
1497-1505.
Holland, D. M.; Principe, P. P.; Vorburger, L. (1999) Rural ozone: trends and exceedances at CASTNet sites.
Environ. Sci. Technol. 33: 43-48.
Holton, J. R.; Haynes, P. H.; Mclntyre, M. E.; Douglass, A. R.; Rood, R. B.; Pfister, L. (1995)
Stratosphere-troposphere exchange. Rev. Geophys. 33: 403-439.Honrath, R. E.; Hamlin, A. J.; Merrill, J. T.
(1996) Transport of ozone precursors from the Arctic troposphere to the North Atlantic region. J. Geophys.
Res. 101:29,335-29,351.
Honrath, R. E.; Hamlin, A. J.; Merrill, J. T. (1996) Transport of ozone precursors from the Arctic troposphere to the
North Atlantic region. J. Geophys. Res. 101: 29,335-29,351.
Horowitz, L. W.; Liang, J.; Gardner, G. M.; Jacob, D. J . (1998) Export of reactive nitrogen from North America
during summertime: sensitivity to hydrocarbon chemistry. J. Geophys. Res. [Atmos.] 103(D11):
13451-13476.
Horowitz, L. W.; Walters, S.; Mauserall, D. L.; Emmons, L. K.; Rasch, P. J.; Granier, C.; Tie, X.; Lamarque, J.-F.;
Schultz, M. G.; Tyndall, G. S.; Orlando, J. J.; Brasseur, G. P. (2003) A global simulation of tropospheric
ozone and related tracers: description and evaluation of MOZART, version 2. J. Geophys. Res. [Atmos.]
108(024): 10.1029/2002JD002853.
AX3-227
-------
Horvath, L.; Nagy, Z.; Weidinger, T.; Artz, R.; Luke, W. T.; Valigura, R.; Pinto, J. P.; Womack, J. (1995)
Measurement effluxes of trace gases (O3, NOX, SO2, CO2, HNO3), paniculate sulfate and nitrate, water
vapour over short vegetation by gradient and eddy correlation techniques in Hungary. EGS XX. General
Assembly; April; Hamburg, Germany. Ann. Geophys. 13(suppl. 2): C490.
Horvath, L.; Pinto, J.; Weidinger, T. (2003) Estimate of the dry deposition of atmospheric nitrogen and sulfur
species to spruce forest. Idojaras (Q. J. Hung. Meteorol. Serv.) 107: 249-255.
Howard-Reed, C.; Wallace, L. A.; Ott, W. R. (2002) The effect of opening windows on air change rates in two
homes. J. Air Waste Manage Assoc. 52: 147-159.
Hudman, R. C.; Jacob, D. J.; Cooper, O. R.; Evans, M. J.; Heald, C. L.; Park, R. J.; Fehsenfeld, F.; Flocke, F.;
Holloway, J.; Htibler, G.; Kita, K.; Koike, M.; Kondo, Y.; Neuman, A.; Nowak, J.; Oltmans, S.; Parrish, D.;
Roberts, J. M.; Ryerson, T. (2004) Ozone production in transpacific Asian pollution plumes and
implications for ozone air quality in California. J. Geophys. Res. (Atmos.) 109(D23):
10.1029/2004 JD004974.
ICF Consulting; ManTech Environmental Technology, Inc. (2003) Total Risk Integrated Methodology
TRIM.ExpoWialation user's document. Volume I: air pollutants exposure model (APEX, version 3) user's
guide. Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Available: http://www.epa.gov/ttn/fera/data/apex322/apexusersguidevoli4-24-03.pdf
[7 September, 2004].
linuma, Y.; Boge, O.; Gnauk, T.; Herrmann, H. (2004) Aerosol-chamber study of the a-pinene/O3 reaction:
influence of particle acidity on aerosol yields and products. Atmos. Environ. 38: 761-773.
Institute for Social Research. (1997) Child development supplement to the panel study of income dynamics. Ann
Arbor, MI: University of Michigan. Available at: www.isr.umich.edu/src/child-development/home.html.
Intergovernmental Panel on Climate Change (IPCC). (2001) Climate change 2001: the scientific basis. Contribution
of working group I to the third assessment report of the Intergovernmental Panel on Climate Change.
Cambridge, United Kingdom: Cambridge University Press.
Isukapalli, S. S.; Purushothaman, V.; Georgopoulos, P. G. (1999) Mechanistic modeling of the interrelationships
between indoor/outdoor air quality and human exposure in a GIS framework. Presented at: 92nd Air &
Waste Management Association Annual Meeting; June; St. Louis, MO. Pittsburgh, PA: Air & Waste
Management Association.
Ito, K.; De Leon, S. F.; Lippmann, M. (2005) Associations between ozone and daily mortality, analysis and
meta-analysis. Epidemiology 16: 446-457.
Jaffe, D. A.; Honrath, R. E.; Herring, J. A.; Li, S.-M.; Kahl, J. D. (1991) Measurements of nitrogen oxides at
Barrow, Alaska during spring: evidence for regional and northern hemispheric sources of pollution.
J. Geophys. Res. [Atmos.] 96: 7395-7405.
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: 10.1029/2003GL017024.
Jakobi, G.; Fabian, P. (1997) Indoor/outdoor concentrations of ozone and peroxyacetyl nitrate (PAN). Int. J.
Biometeorol. 40: 162-165.
James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003a) A 15-year climatology of
stratosphere-troposphere exchange with a Lagrangian particle dispersion model: 1. Methodology and
validation. J. Geophys. Res. 108(D12): 10.1029/2002JD002637.
James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003b) A 15-year climatology of
stratosphere-troposphere exchange with a Lagrangian particle dispersion model. 2. Mean climate and
seasonal variability. J. Geophys. Res. (Atmos.) 108(D12): 10.1029/2002JD002639.
Johnson, T. (1984) A study of personal exposure to carbon monoxide in Denver, Colorado. Research Triangle Park,
NC: U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory; report no.
EPA-600/4-84-014.
Johnson, T. A. (1989) Human activity patterns in Cincinnati, Ohio. Palo Alto, CA: Electric Power Research
Institute; report no. EPRIEN-6204.
Johnson, T. R. (1995) Recent advances in the estimation of population exposure to mobile source pollutants.
J. Exposure Anal. Environ. Epidemiol. 5: 551-571.
Johnson, T. (1997) A pilot study in Los Angeles to measure personal ozone exposures during scripted activities.
Washington, DC: American Petroleum Institute, Health and Environmental Sciences Department; API
publication no. DR 218.
Johnson, T. (2001) A guide to selected algorithms, distributions, and databases used in exposure models developed
by the Office of Air Quality Planning and Standards (DRAFT). Research Triangle Park, NC: U.S.
Environmental Protection Agency, CERM report TR01.
AX3-228
-------
Johnson, T. (2003) A guide to selected algorithms, distributions, and databases used in exposure models developed
by the Office of Air Quality Planning and Standards. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Research and Development; EPA grant no. CR827033. Available:
http://www.epa.gov/ttn/fera/data/human/report052202.pdf [9 April, 2004].
Johnson, T.; Long, T. (2004) Determining the frequency of open windows in residences: a pilot study in Durham,
North Carolina during varying temperature conditions. J. Exposure Anal. Environ. Epidemiol.:
10.1038/sj.jea.7500409.
Johnson, T. R.; Paul, R. A.; Capel, J. E.; McCurdy, T. (1990) Estimation of ozone exposure in Houston using a
probabilistic version of NEM. Presented at: 83rd annual meeting and exhibition of the Air & Waste
Management Association; June; Pittsburgh, PA. Pittsburgh, PA: Air & Waste Management Association;
paper no. 90-150.1.
Johnson, T. R.; Wijnberg, L.; Capel, J. E.; Vostal, J. J. (1992) The use of activity diary data to estimate the
probability of exposure to air pollution. In: Berglund, R. L., ed. Tropospheric ozone and the environment II:
effects, modeling and control [papers from an international specialty conference]; November; Atlanta, GA.
Pittsburgh, PA: Air & Waste Management Association; pp. 713-724. (A&WMA transactions series no. 20).
Johnson, T.; Capel, J.; McCoy, M. (1996a) Estimation of ozone exposures experienced by urban residents using a
probabilistic version of NEM and 1990 population data. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards; contract no. 68-DO-0062.
Johnson, T.; Capel, J.; McCoy, M.; Mozier, J. W. (1996b) Estimation of ozone exposures experienced by outdoor
workers in nine urban areas using a probabilistic version of NEM. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards; contract no. 63-D-30094,
work assignment nos. 0-1 and 1-4.
Johnson, T.; Capel, J.; Mozier, J.; McCoy, M. (1996c) Estimation of ozone exposures experienced by outdoor
children in nine urban areas using a probabilistic version of NEM. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards; contract no. 63-D-30094.
Johnson, T.; Mozier, J.; Capel, J. (1997) Supplement to "Estimation of ozone exposures experienced by outdoor
children in nine urban areas using a probabilistic version of NEM (April 1996)". Research Triangle Park,
NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
Johnson, T.; Myers, J.; Kelly, T.; Wisbith, A.; Ollison, W. (2004) A pilot study using scripted ventilation conditions
to identify key factors affecting indoor pollutant concentrations and air exchange rate in a residence. J.
Exposure Anal. Environ. Epidemiol. 14: 1-22.
Junge, C. E. (1963) Air chemistry and radioactivity. New York, NY: Academic Press. (Van Mieghem, J.; Hales,
A. L., eds. International geophysics series: v. 4).
Kelly, N. A.; Wolff, G. T.; Ferman, M. A. (1984) Sources and sinks of ozone in rural areas. Atmos. Environ.
18: 1251-1266.
Kleinman, L. I. (1991) Seasonal dependence of boundary-layer peroxide concentration - the low and high NOX
regimes. J. Geophys. Res. (Atmos.) 96(D11): 20721-20733.
Klene, J. G.; Clausen, P. A.; Weschler, C. J.; Wolkoff, P. (2001) Determination of ozone removal rates by selected
building products using the FLEC emission cell. Environ. Sci. Technol. 35: 2548-2553.
Klepeis, N. E.; Ott, W. R.; Switzer, P. (1995) Modeling the time series of respirable suspended particles and carbon
monoxide from multiple smokers: validation in two public smoking lounges. Presented at: 88th annual
meeting and exhibition of the Air & Waste Management Association; June; San Antonio, TX. Pittsburgh,
PA: Air & Waste Management Association; paper no. 95-MPH.03.
Klepeis, N. E.; Tsang, A. M.; Behar, J. V. (1996) Analysis of the national human activity pattern survey (NHAPS)
respondents from a standpoint of exposure assessment. Washington, DC: U.S. Environmental Protection
Agency, Office of Research and Development; report no. EPA/600/R-96/074.
Klepeis, N. E.; Nelson, W. C.; Ott. W. R.; Robinson, J. P. Tsang, A. M.; Switzer, P.; Behar, J. V.; Hern, S. C.;
Engelmann, W. H. (2001) The National Human Activity Pattern Survey (NHAPS): a resource for assessing
exposure to environmental pollutants. J. Exposure Anal. Environ. Epidemiol. 11: 231-252.
Kley, D.; Volz, A.; Mulheims, F. (1988) Ozone measurements in historic perspective. In: Isaksen, I. S. A., ed.
Tropospheric ozone - regional and global scale interactions: proceedings of the NATO advanced workshop
on regional and global ozone interaction and its environmental consequences; June 1987; Lillehammer,
Norway. Dordrecht, The Netherlands: D. Reidel Publishing Company; pp. 63-72. [NATO Advanced
Science Institutes (ASI) studies: series C, mathematical and physical sciences v. 227].
AX3-229
-------
Knudsen, H. P.; Lefohn, A. S. (1988) The use of geostatistics to characterize regional ozone exposures. In: Heck, W.
W.; Taylor, O. C.; Tingey, D. T., eds. Assessment of crop loss from air pollutants: proceedings of an
international conference; October 1987; Raleigh, NC. Essex, United Kingdom: Elsevier Science Publishers,
Ltd.; pp. 91-105.
Korc, M. E. (1996) A socioeconomic assessment of human exposure to ozone in the South Coast Air Basin of
California. J. Air Waste Manage. Assoc. 46: 547-557.
Kotchenruther, R. A.; Jaffe, D. A.; Beine, H. J.; Anderson, T. L.; Bottenheim, J. W.; Harris, J. M.; Blake, D. R.;
Schmitt, R. (2001a) Observations of ozone and related species in the northeast Pacific during the PHOBEA
campaigns: 2. Airborne observations. J. Geophys. Res. (Atmos.) 106: 7463-7483.
Kotchenruther, R. A.; Jaffe, D. A.; Jaegle, L. (200Ib) Ozone photochemistry and the role of peroxyacetyl nitrate in
the springtime northeastern Pacific troposphere: results from the Photochemical Ozone Budget of the
Eastern North Pacific Atmosphere (PHOBEA) campaign. J. Geophys. Res. (Atmos.) 106: 28,731-28,742.
Koutrakis, P.; Sioutas, C.; Ferguson, S. T.; Wolfson, J. M.; Mulik, J. D.; Burton, R. M. (1993) Development and
evaluation of a glass honeycomb denuder/filter pack system to collect atmospheric gases and particles.
Environ. Sci. Technol. 27: 2497-2501.
Krzyzanowski, M. (1997) Methods for assessing the extent of exposure and effects of air pollution. Occup. Environ.
Med. 54: 145-151.
Ktinzli, N.; Lurmann, F.; Segal, M.; Ngo, L.; Balmes, J.; Tager, I. B. (1996) Reliability of lifetime residential history
and activity measures as elements of cumulative ambient ozone exposure assessment. J. Exposure Anal.
Environ. Epidemiol. 6: 289-310.
Ktinzli, N.; Kelly, T.; Balmes, J.; Tager, I. B. (1997) Reproducibility of retrospective assessment of outdoor
time-activity patterns as an individual determinant of long-term ambient ozone exposure. Int. J. Epidemiol.
26: 1258-1271.
Lagus Applied Technology, Inc. (1995) Air change rates in non-residential buildings in California. Sacramento, CA:
California Energy Commission; contract no. 400-91-034; July.
Lavery, T. F.; Rogers, C. M.; Howell, H. K.; Burnett, M. C.; Wanta, C. A. (2002) Clean air status trends network
(CASTNet) 2000 annual report. Available: http://www.epa.gov/castnet/library/annualOO.html (9 September
2003).
Le, N. D.; Zidek, J. V. (1992) Interpolation with uncertain spatial covariances: a Bayesian alternative to Kriging.
J. Multivariate Anal. 43: 351-374.
Le, N. D.; Sun, W.; Zidek, J. V. (1997) Bayesian multivariate spatial interpolation with data missing by design. J. R.
Stat. Soc. 659:501-510.
Le, N. D.; Sun, L.; Zidek, J. V. (2001) Spatial prediction and temporal backcasting for environmental fields having
monotone data patterns. Can. J. Stat. 29: 529-554.
Lee, E. H.; Hogsett, W. E. (1999) Role of concentrations and time of day in developing ozone exposure indices for a
secondary standard. J. Air Waste Manage. Assoc. 49: 669-681.
Lee, K.; Vallarino, J.; Dumyahn, T. Ozkaynak, H.; Spengler, J. (1999) Ozone decay rates in residences. J. Air Waste
Manage. Assoc. 49: 1238-1244.
Lee, M.; Heikes, B. G.; O'Sullivan, D. W. (2000) Hydrogen peroxide and organic hydroperoxide in the troposphere.
Atmos. Environ. 34: 3475-3494.
Lee, K.; Xue, J.; Geyh, A. S.; Ozkaynak, H.; Leaderer, B. P.; Weschler, C. J.; Spengler, J. D. (2002) Nitrous acid,
nitrogen dioxide, and ozone concentrations in residential environments. Environ. Health Perspect.
110: 145-150.
Lee, K.; Parkhurst, W. J.; Xue, J.; Ozkaynak, H.; Neuberg, D.; Spengler, J. D. (2004) Outdoor/indoor/personal ozone
exposures of children in Nashville, Tennessee. J. Air Waste Manage. Assoc. 54: 352-359.
Lefohn, A. S. (1992) The characterization of ambient ozone exposures. In: Lefohn, A. S., ed. Surface level ozone
exposures and their effects on vegetation. Chelsea, MI: Lewis Publishers, Inc.; pp. 31-92.
Lefohn, A. S.; Benedict, H. M. (1983) The potential for the interaction of acidic precipitation and ozone pollutant
doses affecting agricultural crops. Presented at: 76th annual meeting of the Air Pollution Control
Association; June; Atlanta, GA. Pittsburgh, PA: Air Pollution Control Association; paper no. 83-2.2.
Lefohn, A. S.; Jones, C. K. (1986) The characterization of ozone and sulfur dioxide air quality data for assessing
possible vegetation effects. J. AirPollut. Control Assoc. 36: 1123-1129.
Lefohn, A. S.; Mohnen, V. A. (1986) The characterization of ozone, sulfur dioxide, and nitrogen dioxide for selected
monitoring sites in the Federal Republic of Germany. J. Air Pollut. Control Assoc. 36: 1329-1337.
Lefohn, A. S.; Runeckles, V. C. (1987) Establishing standards to protect vegetation - ozone exposure/dose
considerations. Atmos. Environ. 21: 561-568.
AX3-230
-------
Lefohn, A. S.; Tingey, D. T. (1984) The co-occurrence of potentially phytotoxic concentrations of various gaseous
air pollutants. Atmos. Environ. 18: 2521-2526.
Lefohn, A. S.; Davis, C. E.; Jones, C. K.; Tingey, D. T.; Hogsett, W. E. (1987) Co-occurrence patterns of gaseous air
pollutant pairs at different minimum concentrations in the United States. Atmos. Environ. 21: 2435-2444.
Lefohn, A. S.; Knudsen, H. P.; McEvoy, L. R., Jr. (1988) The use of kriging to estimate monthly ozone exposure
parameters for the southeastern United States. Environ. Pollut. 53: 27-42.
Lefohn, A. S.; Krupa, S. V.; Winstanley, D. (1990a) Surface ozone exposures measured at clean locations around the
world. Environ. Pollut. 63: 189-224.
Lefohn, A. S.; Shadwick, D. S.; Mohnen, V. A. (1990b) The characterization of ozone concentrations at a select set
of high-elevation sites in the eastern United States. Environ. Pollut. 67: 147-178.
Lefohn, A. S.; Benkovitz, C. M; Tanner, R. L.; Smith, L. A.; Shadwick, D. S. (1991) Air quality measurements and
characterizations for terrestrial effects research. In: Irving, P. M., ed. Acidic deposition: state of science and
technology, volume I, emissions, atmospheric processes, and deposition. Washington, DC: The U.S.
National Acid Precipitation Assessment Program. (State of science and technology report no. 7).
Lefohn, A. S.; Shadwick, D. S.; Ziman, S. D. (1998) The difficult challenge of attaining EPA's new ozone standard.
Environ. Sci. Technol. 32: 276A-282A.
Lefohn, A. S.; Oltmans, S. J.; Dann, T.; Singh, H. B. (2001) Present-day variability of background ozone in the
lower troposphere. J. Geophys. Res. [Atmos.] 106: 9945-9958.
Lehman, J.; Swinton, K.; Bortnick, S.; Hamilton, C.; Baldridge, E.; Ender, B.; Cox, B. (2004) Spatio-temporal
characterization of tropospheric ozone across the eastern United States. Atmos. Environ. 38: 4357-4369.
Lelieveld, J.; Dentener, F. J. (2000) What controls tropospheric ozone? J. Geophys. Res. [Atmos.] 105: 3531-3551.
Leovic, K. W.; Sheldon, L. S.; Whitaker, D. A.; Hetes, R. G.; Calcagni, J. A.; Baskir, J. N. (1996) Measurement of
indoor air emissions from dry-process photocopy machines. J. Air Waste Manage. Assoc. 46: 821-829.
Leovic, K. W.; Whitaker, D.; Brockmann, C.; Bayer, C. W.; Sojka, P. E. (1998) Evaluation of low-emitting products
for use in the indoor environment. Presented at: 91st annual meeting of the Air & Waste Management
Association; June; San Diego, CA. Pittsburgh, PA: Air & Waste Management Association; paper no.
98-TP48.02 (A292).
Levinson, D. H., ed. (2005) State of the climate in 2004. Bull. Am. Meteorol. Soc. 86: S1-S86.
Levinson, D. H.; Waple, A. M., eds. (2004) State of the climate in 2003. Bull. Am. Meteorol. Soc. 85: S1-S72.
Li, K. H.; Le, N. D.; Sun, L.; Zidek, J. V. (1999) Spatial-temporal models for ambient hourly PM10 in Vancouver.
Environmetrics 10: 321-338.
Li, Q.; Jacob, D. J.; Logan, J. A.; Bey, I.; Yantosca, R. M.; Liu, H.; Martin, R. V.; Fiore, A. M.; Field, B. D.;
Duncan, B. N. (2001) A tropospheric ozone maximum over the Middle East. Geophys. Res. Lett.
28: 3235-3238.
Li, Q.; Jacob, D. J.; Fairlie, T. D.; Liu, H.; Martin, R. V.; Yantosca, R. M. (2002) Stratospheric versus pollution
influences on ozone at Bermuda: reconciling past analyses. J. Geophys. Res. 107(D22):
10.1029/2002JD002138.
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.
Lin, C.-Y. C.; Jacob, D. J.; Fiore, A. M. (2001) Trends in exceedances of the ozone air quality standard in the
continental United States, 1980-1998. Atmos. Environ. 35: 3217-3228.
Linn, W. S.; Shamoo, D. A.; Hackney, J. D. (1992) Documentation of activity patterns in 'high-risk' groups exposed
to ozone in the Los Angeles area. In: Tropospheric ozone and the environment II: effects, modeling and
control: papers from an Air & Waste Management Association international specialty conference;
November, 1991; Atlanta, GA. Pittsburgh, PA: Air & Waste Management Association; pp. 701-712.
(A&WMA publication TR-20).
Linn, W. S.; Spier, C. E.; Hackney, J. D. (1993) Activity patterns in ozone-exposed construction workers. J. Occup.
Med. Toxicol. 2: 1-14.
Linn, W. S.; Shamoo, D. A.; Anderson, K. R.; Peng, R.-C.; Avol, E. L.; Hackney, J. D.; Gong, H., Jr. (1996)
Short-term air pollution exposures and responses in Los Angeles area schoolchildren. J. Exposure Anal.
Environ. Epidemiol. 6: 449-472.
Lioy, P. J. (1990) Assessing total human exposure to contaminants: a multidisciplinary approach. Environ. Sci.
Technol. 24: 938-945.
Lisac, I.; Grubisic, V. (1991) An analysis of surface ozone data measured at the end of the 19th century in Zagreb,
Yugoslavia. Atmos. Environ. Part A 25: 481-486.
Liu, D.-L.; Nazaroff, W. W. (2001) Modeling pollutant penetration across building envelopes. Atmos. Environ.
35: 4451-4462.
AX3-231
-------
Liu, L.-J. S.; Koutrakis, P.; Leech, I; Broder, I. (1995) Assessment of ozone exposures in the greater metropolitan
Toronto area. J. Air Waste Manage. Assoc. 45: 223-234.
Liu, L.-J. S.; Delfino, R.; Koutrakis, P. (1997) Ozone exposure assessment in a southern California community.
Environ. HealthPerspect. 105: 58-65.
Liu, H.; Jacob, D. J.; Chan, L. Y.; Oltmans, S. J.; Bey, I.; Yantosca, R. M; Harris, J. M; Duncan, B. N.; Martin, R.
V. (2002) Sources of tropospheric ozone along the Asian Pacific Rim: an analysis of ozonesonde
observations. J. Geophys. Res. [Atmos.] 107(D21): 10.1029/2001JD002005.
Liu, H.; Jacob, D. J.; Bey, I.; Yantosca, R. M.; Duncan, B. N.; Sachse, G. W. (2003) Transport pathways for Asian
pollution outflow over the Pacific: interannual and seasonal variations. J. Geophys. Res. (Atmos.)
108(D20): 10.1029/2002JD003102.
Liu, H.; Jacob, D. J.; Dibb, J. E.; Fiore, A. M.; Yantosca, R. M. (2004) Constraints on the sources of tropo spheric
ozone from 210Pb-7Be-O3 correlations. J. Geophys. Res. (Atmos.) 109(D7): 10.1029/2003JD003988.
Logan, J. A. (1989) Ozone in rural areas of the United States. J. Geophys. Res. [Atmos.] 94: 8511-8532.
Logan, J. A. (1999) An analysis of ozonesonde data for the troposphere: recommendations for testing 3-D models
and development of a gridded climatology for tropospheric ozone. J. Geophys. Res. 104: 16,115-16,149.
Ludwig, F. L.; Reiter, E.; Shelar, E.; Johnson, W. B. (1977) The relation of oxidant levels to precursor emissions and
meteorological features: v. I, analysis and findings. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards; report no. EPA-450/3-77-022a. Available
from: NTIS, Springfield, VA; PB-275 001.
Lurmann, F.; Colome, S. (1991) An assessment of current and future human exposure to ozone in the South Coast
Air Basin. Santa Rosa, CA: Sonoma Technology, Inc.; report no. STI-90070-1022-FR; 1991 AQMP
revision, technical report V-H.
Marr, L. C.; Morrison, G. C.; Nazaroff, W. W.; Harley, R. A. (1998) Reducing the risk of accidental death due to
vehicle-related carbon monoxide poisoning. J. Air Waste Manage. Assoc. 48: 899-906.
Martin, R. V.; Jacob, D. J.; Yantosca, R. M.; Chin, M.; Ginoux, P. (2003) Global and regional decreases in
tropospheric oxidants from photochemical effects of aerosols. J. Geophys. Res. [Atmos.] 108(D3):
10.1029/2002JD002622.
Matheron, G. (1963) Principles of geostatistics. Econ. Geol. 58: 1246-1266.
McCurdy, T. R. (1994) Human exposure to ambient ozone. In: McKee, D. J., ed. Tropospheric ozone: human health
and agricultural impacts. Boca Raton, FL: Lewis Publishers; pp. 85-127.
McCurdy, T. (1997a) Human activities that may lead to high inhaled intake doses in children aged 6-13. Environ.
Toxicol. Pharmacol. 4: 251-260.
McCurdy, T. (1997b) Modeling the dose profile in human exposure assessments: ozone as an example. Rev.
Toxicol. 1: 3-23.
McCurdy, T. (2000) Conceptual basis for multi-route intake dose modeling using an energy expenditure approach.
J. Exposure Anal. Environ. Epidemiol. 10: 86-97.
McCurdy, T.; Glen, G.; Smith, L.; Lakkadi, Y. (2000) The National Exposure Research Laboratory's Consolidated
Human Activity Database. J. Exposure Anal. Environ. Epidemiol. 10: 566-578.
Meagher, J. F.; Lee, N. T.; Valente, R. J.; Parkhurst, W. J. (1987) Rural ozone in the southeastern United States.
Atmos. Environ. 21: 605-615.
Meloen, J.; Siegmund, P.; Van Velthoven, P.; Kelder, H.; Sprenger, M.; Wernli, H.; Kentarchos, A.; Roelofs, G.;
Feichter, J.; Land, C.; Forster, C.; James, P.; Stohl, A.; Collins, W.; Cristofanelli, P. (2003)
Stratosphere-troposphere exchange: a model and method intercomparison. J. Geophys. Res. 108(D12):
10.1029/2002JD002274.
Meng, Q. Y.; Turpin, B. J.; Korn, L.; Weisel, C. P.; Morandi, M.; Colome, S.; Zhang, J.; Stock, T.; Spektor, D.;
Winer, A.; Zhang, L.; Lee, J. H.; Giovanetti, R.; Cui, W.; Kwon, J.; Alimokhtari, S.; Shendell, D.; Jones, J.;
Farrar, C.; Maberti, S. (2005) Influence of ambient (outdoor) sources on residential indoor and personal
PM25 concentrations: analyses of RIOPA data. J. Exposure Anal. Environ. Epidemiol. 15: 17-28.
Mohnen, V. A. (1989) Mountain cloud chemistry project: wet, dry and cloud water deposition. Research Triangle
Park, NC: U.S. Environmental Protection Agency, Atmospheric Research and Exposure Assessment
Laboratory; report no. EPA/600/3-89/009. Available from: NTIS, Springfield, VA; PB89-148597.
Monks, P. S. (2000) A review of the observations and origins of the spring ozone maximum. Atmos. Environ.
34: 3545-3561.
Morrison, G. C.; Nazaroff, W. W. (2000) The rate of ozone uptake on carpets: experimental studies. Environ. Sci.
Technol. 34: 4963-4968.
Morrison, G. C.; Nazaroff, W. W. (2002) The rate of ozone uptake on carpet: mathematical modeling. Atmos.
Environ. 36: 1749-1756.
AX3-232
-------
Morrison, G. C.; Nazaroff, W. W.; Cano-Ruiz, J. A.; Hodgson, A. T.; Modera, M. P. (1998) Indoor air quality
impacts of ventilation ducts: ozone removal and emissions of volatile organic compounds. J. Air Waste
Manage. Assoc. 48: 941-952.
Mueller, F. X.; Loeb, L.; Mapes, W. H. (1973) Decomposition rates of ozone in living areas. Environ. Sci. Technol.
7: 342-346.
Murphy, D. M.; Fahey, D. W. (1994) An estimate of the flux of stratospheric reactive nitrogen and ozone into the
troposphere. J. Geophys. Res. (Atmos.) 99: 5325-5332.
Murray, D. M.; Burmaster, D. E. (1995) Residential air exchange rates in the United States: empirical and estimated
parametric distributions by season and climatic region. Risk Anal. 15: 459-465.
Nagda, N. L.; Rector, H. E.; Koontz, M. D. (1987) Guidelines for monitoring indoor air quality. Washington, DC:
Hemisphere Publishing Corporation.
National Acid Precipitation Assessment Program. (1987) Interim assessment: the causes and effects of acidic
deposition, volume I, executive summary. Washington, DC: Office of the Director of Research.
Navidi, W.; Thomas, D.; Langholz, B.; Stram, D. (1999) Statistical methods for epidemiologic studies of the health
effects of air pollution. Cambridge, MA: Health Effects Institute; research report no. 86.
Nazaroff, W. W.; Cass, G. R. (1986) Mathematical modeling of chemically reactive pollutants in indoor air.
Environ. Sci. Technol. 20: 924-934.
Nazaroff, W. W.; Weschler, C. J. (2004) Cleaning products and air fresheners: exposure to primary and secondary
air pollutants. Atmos. Environ. 38: 2841-2865.
Neidell, M. J. (2004) Air pollution, health, and socio-economic status: the effect of outdoor air quality on childhood
asthma. J. Health Econ. 23: 1209-1236.
Newchurch, M. J.; Ayoub, M. A.; Oltmans, S.; Johnson, B.; Schmidlin, F. J. (2003) Vertical distribution of ozone at
four sites in the United States. J. Geophys. Res. 108(D1): 10.1029/2002JD002059.
Niu, J.; Tung, T. C. W.; Burnett, J. (2001) Ozone emission rate testing and ranking method using environmental
chamber. Atmos. Environ. 35: 2143-2151.
Nejgaard, J. K.; Christensen, K. B.; Wolkoff, P. (2005) The effect on human eye blink frequency of exposure to
limonene oxidation products and methacrolein. Toxicol. Lett. 156: 241-251.
Northeast States for Coordinated Air Use Management (NESCAUM). (2002) Indoor/outdoor school air monitoring
project. Boston, MA. Available: http://www.nescaum.org/pdf/schoolmonitoring.pdf [29 October, 2003].
Odman, M. T.; Ingram, C. I. (1996) Multiscale air quality simulation platform (MAQSIP): source code
documentation and validation. Research Triangle Park, NC: MCNC; technical report ENV-96TR002-vl.O.
Available: http://www.ce.gatech.edu/~todman/maqsip.pdf (9 September 2003).
Olsen, M. A.; Douglass, A. R.; Schoeberl, M. R. (2002) Estimating downward cross-tropopause flux using column
ozone and potential vorticity. J. Geophys. Res. 107(D22): 10.1029/2001JD002041.
Orzechowska, G. E.; Paulson, S. E. (2002) Production of OH radicals from the reactions of C4-C6 internal alkenes
and styrenes with ozone in the gas phase. Atmos. Environ. 36: 571-581.
Ott, W. R. (1982) Concepts of human exposure to air pollution. Environ. Int. 7: 179-196.
Ott, W. R. (1985) Total human exposure: an emerging science focuses on humans as receptors of environmental
pollution. Environ. Sci. Technol. 19: 880-886.
Ott, W.; Langan, L.; Switzer, P. (1992) A time series model for cigarette smoking activity patterns: model validation
for carbon monoxide and respirable particles in a chamber and an automobile. In: Measuring,
understanding, and predicting exposures in the 21st century: proceedings of the conference; November
1991; Atlanta, GA. J. Exposure Anal. Environ. Epidemiol. 2(suppl. 2): 175-200.
Park, J.-H.; Spengler, J. D.; Yoon, D.-W.; Dumyahn, T.; Lee, K.; Ozkaynak, H. (1998) Measurement of air
exchange rate of stationary vehicles and estimation of in-vehicle exposure. J. Exposure Anal. Environ.
Epidemiol. 8: 65-78.
Park, R. J.; Pickering, K. E.; Allen, D. J.; Stenchikov, G. L.; Fox-Rabinovitz, M. S. (2004) Global simulation of
tropospheric ozone using the University of Maryland Chemical Transport Model (UMD-CTM): 1. Model
description and evaluation. J. Geophys. Res. [Atmos.] 109(D9): 10.1029/2003 JD004266.
Pavelin, E. G.; Johnson, C. E.; Rughooputh, S.; Toumi, R. (1999) Evaluation of preindustrial surface ozone
measurements made using Schonbein's method. Atmos. Environ. 33: 919-929.
Penkett, S. A. (1983) PAN in the natural atmosphere. Nature (London) 302: 293-294.
Penkett, S. A.; Brice, K. A. (1986) The spring maximum in photo-oxidants in the Northern Hemisphere troposphere.
Nature (London) 319: 655-657.
Pinto, J. P.; Lefohn, A. S.; Shadwick, D. S. (2004) Spatial variability of PM2 s in urban areas in the United States.
J. Air Waste Manage. Assoc. 54: 440-449.
AX3-233
-------
Pippin, M. R.; Bertman, S.; Thornberry, T.; Town, M.; Carroll, M. A.; Sillman, S. (2001) Seasonal variation of
PAN, PPN, and O3 at the upper Midwest PROPHET site. J. Geophys. Res. (Atmos.) 106: 24,451-24,463.
Poupard, O.; Blondeau, P.; lordache, V.; Allard, F. (2005) Statistical analysis of parameters influencing the
relationship between outdoor and indoor air quality in schools. Atmos. Environ. 39: 2071-2080.
Prados, A. I. (2000) The export of ozone and other atmospheric trace species from North America: observations and
numerical simulations [dissertation]. College Park, MD: The University of Maryland.
Prinz, B.; Krause, G. H. M. (1989) State of scientific discussion about the causes of the novel forest decline in the
Federal Republic of Germany and surrounding countries. In: Bucher, J. B.; Bucher-Wallin, I., eds. Air
pollution and forest decline, v. I: Proceedings of the 14th international meeting for specialists in air
pollution effects on forest ecosystems; October 1988; Interlaken, Switzerland. Birmensdorf, Switzerland:
Eidgenossische Anstalt fur das forstliche Versuchswesen (EAFV); pp. 27-34.
Prinz, B.; Krause, G. H. M.; Jung, K.-D. (1985) Untersuchungen der LIS Essen zur Problematik der Waldschaden
[Investigations of the LIS Essen into the problem of forest damage]. In: Kortzfleisch, G., ed. Waldschaden:
Theorie und Praxis auf der Suche nach Antworten [Forest damage: theory and practice in the search for
answers]. Mtinchen, Germany: R. Oldenbourg Verlag; pp. 143-194.
Proyou, A. G.; Ziomas, I. C.; Stathopoulos, A. (1998) Application of a three-layer photochemical box model in an
Athens street canyon. J. Air Waste Manage. Assoc. 48: 427-433.
Pun, B. K.; Seigneur, C.; White, W. (2003) Day-of-week behavior of atmospheric ozone in three U.S. cities. J. Air
Waste Manage. Assoc. 53: 789-801.
Purushothaman, V.; Georgopoulos, P. G. (1997) Computational tools to aid the estimation and visualization of
potential human exposure to ozone. In: Proceedings of the Air & Waste Management Association Specialty
Conference on Computing in Environmental Resource Management; December, 1996; Research Triangle
Park, NC. Pittsburgh, PA: A&WMA; VIP-68.
Purushothaman, V.; Georgopoulos, P. G. (1999a) Evaluation of regional emissions control strategies for ozone
utilizing population exposure metrics: a new GIS-based approach applied to the July 1995 episode over the
OTAG domain. Piscataway, NJ: Environmental and Occupational Health Sciences Institute, Ozone
Research Center; ORC technical report, ORC-TR99-02.
Purushothaman, V.; Georgopoulos, P. G. (1999b) Integrating photochemical modeling, geostatistical techniques and
geographical information systems for ozone exposure assessment. Piscataway, NJ: Environmental and
Occupational Health Sciences Institute, Ozone Research Center; ORC technical report, ORC-TR99-01.
Reagan, J. (1984) Air quality data interpretation. In: Heck, W. W.; Taylor, O. C.; Adams, R. M.; Bingham, G. E.;
Miller, J. E.; Preston, E. M.; Weinstein, L. H., eds. National Crop Loss Assessment Network (NCLAN)
1982 annual report. Corvallis, OR: U.S. Environmental Protection Agency, Corvallis Environmental
Research Laboratory; pp. 198-219; report no. EPA-600/4-84-049. Available from: NTIS, Springfield, VA;
PB84-169358.
Reeves, C. E.; Penkett, S. A. (2003) Measurements of peroxides and what they tell us. Chem. Rev. 103: 5199-5218.
Regener, E. (1941) Ozonschicht und atmospharische turbulenz. Ber. Dtsch. Wetterdienstes US-Zone No. 11: 45-57.
Reiss, R.; Ryan, P. B.; Koutrakis, P.; Tibbetts, S. J. (1995) Ozone reactive chemistry on interior latex paint. Environ.
Sci. Technol. 29: 1906-1912.
Reiter, E. R. (1977) The role of stratospheric import on tropospheric ozone concentrations. In: Dimitriades, B., ed.
International conference on photochemical oxidant pollution and its control - proceedings: volume I;
September 1976; Raleigh, NC. Research Triangle Park, NC: U.S. Environmental Protection Agency,
Environmental Sciences Research Laboratory; pp. 393-410; report no. EPA-600/3-77-001a. Available from:
NTIS, Springfield, VA; PB-264232.
Research Triangle Institute. (1990) An investigation of infiltration and indoor air quality. Albany, NY: New York
State Energy Research and Development Authority; report no. NYERDA 90-11.
Riediker, M.; Williams, R.; Devlin, R.; Griggs, T.; Bromberg, P. (2003) Exposure to paniculate matter, volatile
organic compounds, and other air pollutants inside patrol cars. Environ. Sci. Technol. 37: 2084-2093.
Rifai, H. S.; Hopkins, L.; Mehdizadeh, F. (2000) Developing human exposure estimates for indoor/outdoor air
pollution in Houston. In: Environmental Institute of Houston -1999 Annual Report. Houston, TX:
Environmental Institute of Houston.
Roberts, J. M.; Williams, J.; Baumann, K.; Buhr, M. P.; Goldan, P. D.; Holloway, J.; Htibler, G.; Kuster, W. C.;
McKeen, S. A.; Ryerson, T. B.; Trainer, M.; Williams, E. J.; Fehsenfeld, F. C.; Bertman, S. B.; Nouaime,
G.; Seaver, C.; Grodzinsky, G.; Rodgers, M.; Young, V. L. (1998) Measurements of PAN, PPN, and
MPAN made during the 1994 and 1995 Nashville Intensives of the Southern Oxidant Study: implications
for regional ozone production from biogenic hydrocarbons. J. Geophys. Res. [Atmos.] 103: 22,473-22,490.
AX3-234
-------
Roberts, J. M; Flocke, F.; Stroud, C. A.; Hereid, D.; Williams, E.; Fehsenfeld, F.; Brane, W.; Martinez, M.; Harder,
H. (2002) Ground-based measurements of peroxycarboxylic nitric anhydrides (PANs) during the 1999
Southern Oxidants Study Nashville Intensive. J. Geophys. Res. [Atmos.] 107(D21):
10.1029/2001JD000947.
Roberts, J. M.; Jobson, B. T.; Kuster, W.; Goldan, P.; Murphy, P.; Williams, E.; Frost, G.; Riemer, D.; Apel, E.;
Stroud, C.; Wiedinmyer, C.; Fehsenfeld, F. (2003) An examination of the chemistry of peroxycarboxylic
nitric anhydrides and related volatile organic compounds during Texas Air Quality Study 2000 using
ground-based measurements. J. Geophys. Res. [Atmos.] 108(D16): 10.1029/2003JD003383.
Roberts, J. M.; Flocke, F.; Chen, G.; De Gouw, J.; Holloway, J. S.; Htibler, G.; Neuman, J. A.; Nicks, D. K., Jr.;
Nowak, J. B.; Parrish, D. D.; Ryerson, T. B.; Sueper, D. T.; Warneke, C.; Fehsenfeld, F. C. (2004)
Measurement of peroxycarboxylic nitric anhydrides (PANs) during the ITCT 2K2 aircraft intensive
experiment. J. Geophys. Res. [Atmos.] 109(D23S21): 10.1029/2004JD004960.
Robinson, J. P.; Wiley, J. A.; Piazza, T.; Garrett, K.; Cirksena, K. (1991) Activity patterns of California residents
and their implications for potential exposure to pollution (final report). Sacramento, CA: California Air
Resources Board; report no. CARB-A6-177-33.
Roelofs, G. J.; Lelieveld, J.; Van Borland, R. (1997) A three-dimensional chemistry/general circulation model
simulation of anthropogenically derived ozone in the troposphere and its radiative climate forcing.
J. Geophys. Res. [Atmos.] 102: 23,389-23,401.
Rohr, A. C.; Wilkins, C. K.; Clausen, P. A.; Hammer, M.; Nielsen, G. D.; Wolkoff, P.; Spengler, J. D. (2002) Upper
airway and pulmonary effects of oxidation products of (+)-a-pinene, J-limonene, and isoprene in B ALB/c
mice. Inhalation Toxicol. 14: 663-684.
Rohr, A. C.; Shore, S. A.; Spengler, J. D. (2003) Repeated exposure to isoprene oxidation products causes enhanced
respiratory tract effects in multiple murine strains. Inhalation Toxicol. 15: 1191-1207.
Romieu, I.; Lugo, M. C.; Colome, S.; Garcia A. M.; Avila, M. H.; Geyh, A.; Velasco, S. R.; Rendon, E. P. (1998)
Evaluation of indoor ozone concentration and predictors of indoor-outdoor ratio in Mexico City. J. Air
Waste Manage. Assoc. 48: 327-335.
Rotman, D. A.; Atherton, C. S.; Bergmann, D. J.; Cameron-Smith, P. J.; Chuang, C. C.; Connell, P. S.; Dignon,
J. E.; Franz, A.; Grant, K. E.; Kinnison, D. E.; Molenkamp, C. R.; Proctor, D. D.; Tannahill, J. R. (2004)
IMPACT, the LLNL 3-D global atmospheric chemical transport model for the combined troposphere and
stratosphere: model description and analysis of ozone and other trace gases. J. Geophys. Res. [Atmos.]
109(04): 10.1029/2002JD003155.
Ryerson, T. B.; Trainer, M.; Holloway, J. S.; Parrish, D. D.; Huey, L. G.; Sueper, D. T.; Frost, G. J.; Donnelly,
S. G.; Schauffler, S.; Atlas, E. L.; Kuster, W. C.; Goldan, P. D.; Htibler, G.; Meagher, J. F.; Fehsenfeld,
F. C. (2001) Observations of ozone formation in power plant plumes and implications for ozone control
strategies. Science (Washington, DC) 292: 719-723.
Sabersky, R. H.; Sinema, D. A.; Shair, F. H. (1973) Concentrations, decay rates, and removal of ozone and their
relation to establishing clean indoor air. Environ. Sci. Technol. 7: 347-353.
Salmon, L. G.; Cass, G. R.; Bruckman, K.; Haber, J. (2000) Ozone exposure inside museums in the historic central
district of Krakow, Poland. Atmos. Environ. 34: 3823-3832.
Sarnat, J. A.; Koutrakis, P.; Suh, H. H. (2000) Assessing the relationship between personal paniculate and gaseous
exposures of senior citizens living in Baltimore, MD. J. Air Waste Manage. Assoc. 50: 1184-1198.
Sarnat, J. A.; Schwartz, J.; Catalano, P. J.; Suh, H. H. (2001) Gaseous pollutants in paniculate matter epidemiology:
confounders or surrogates? Environ. Health Perspect. 109: 1053-1061.
Sarnat, J. A.; Brown, K. W.; Schwartz, J.; Coull, B. A.; Koutrakis, P. (2005) Ambient gas concentrations and
personal paniculate matter exposures: implications for studying the health effects of particles.
Epidemiology 16: 385-395.
Sarwar, M.; Corsi, R.; Kimura, Y.; Allen, D.; Weschler, C. (2001) Hydroxyl radicals in indoor environments. In:
Proceedings of the Air & Waste Management Association's 94th Annual Conference & Exhibition; June;
Orlando, FL. Pittsburgh, PA: Air & Waste Management Association.
Sarwar, G.; Corsi, R.; Kumura, Y.; Allen, D.; Weschler, C. J. (2002) Hydroxyl radicals in indoor environments.
Atmos. Environ. 36: 3973-3988.
Scheeren, B. A.; Adema, E. H. (1996) Monitoring ambient ozone with a passive measurement technique method,
field results and strategy. Water Air Soil Pollut. 91: 335-350.
Schoeberl, M. R.; Douglass, A. R.; Zhu, Z.; Pawson, S. (2003) A comparison of the lower stratospheric age spectra
derived from a general circulation model and two data assimilation systems. J. Geophys. Res. (Atmos.)
108(03): 10.1029/2002JD002652.
AX3-235
-------
Seo, K.-H.; Bowman, K. P. (2002) Lagrangian estimate of global stratosphere-troposphere mass exchange.
J. Geophys. Res. (Atmos.) 107(D21): 10.1029/2002JD002441.
Shair, F. H. (1981) Relating indoor pollutant concentrations of ozone and sulfur dioxide to those outside: economic
reduction of indoor ozone through selective filtration of the make-up air. ASHRAE Trans.
87(pt. 1): 116-139.
Shair, F. H.; Heitner, K. L. (1974) Theoretical model for relating indoor pollutant concentrations to those outside.
Environ. Sci. Technol. 8: 444-451.
Shamoo, D. A.; Johnson, T. R.; Trim, S. C.; Little, D. E.; Linn, W. S.; Hackney, J. D. (1991) Activity patterns in a
panel of outdoor workers exposed to oxidant pollution. J. Exposure Anal. Environ. Epidemiol. 1: 423-438.
Shamoo, D. A.; Linn, W. S.; Peng, R.-C.; Solomon, J. C.; Webb, T. L.; Hackney, J. D.; Gong, H., Jr. (1994)
Time-activity patterns and diurnal variation of respiratory status in a panel of asthmatics: implications for
short-term air pollution effects. J. Exposure Anal. Environ. Epidemiol. 4: 133-148.
Sheppard, L.; Slaughter, J. C.; Schildcrout, J.; Liu, L.-J. S.; Lumley, T. (2005a) Exposure and measurement
contributions to estimates of acute air pollution effects. J. Exposure Anal. Environ. Epidemiol. 15: 366-376.
Sheppard, L. (2005b) Acute air pollution effects: consequences of exposure distribution and measurements.
J. Toxicol. Environ. Health Part A 68: 1127-1135.
Sherman, M. H.; Matson, N. E. (1997) Residential ventilation and energy characteristics. Presented at: 1997
American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) winter meeting;
February; Philadelphia, PA. Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning
Engineers, Inc. Report no. CONF-9702141. ASHRAE Trans. 103: 717-730.
Shindell, D. T.; 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.
Sillman, S.; Logan, J. A.; Wofsy, S. C. (1990) The sensitivity of ozone to nitrogen oxides and hydrocarbons in
regional ozone episodes. J. Geophys. Res. [Atmos.] 95: 1837-1851.
Singh, H. B.; Ludwig, F. L.; Johnson, W. B. (1977) Ozone in clean remote atmospheres: concentrations and
variabilities. New York, NY: Coordinating Research Council, Inc.; report no. CRC-APRAC CAPA-15-76.
Available from: NTIS, Springfield, VA; PB-272290.
Singh, H. B.; Ludwig, F. L.; Johnson, W. B. (1978) Tropospheric ozone: concentrations and variabilities in clean
remote atmospheres. Atmos. Environ. 12: 2185-2196.
Smith, L. A.; Lefohn, A. S. (1991) Co-occurence of ozone and wet deposited hydrogen ion in the United States.
Atmos. Environ. Part A 25: 2707-2716.
Solberg, S.; Krognes, T.; Stordal, F.; Hov, 0.; Beine, H. J.; Jaffe, D. A.; Clemitshaw, K. C.; Penkett, S. A. (1997)
Reactive nitrogen compounds at Spitsbergen in the Norwegian Arctic. J. Atmos. Chem. 28: 209-225.
Serensen, D. N.; Weschler, C. J. (2002) Modeling-gas phase reactions in indoor environments using computational
fluid dynamics. Atmos. Environ. 36: 9-18.
Sparks, J. P.; Roberts, J. M.; Monson, R. K. (2003) The uptake of gaseous organic nitrogen by leaves: a significant
global nitrogen transfer process. Geophys. Res. Lett. 30(23): 10.1029/2003GL018578.
Spier, C. E.; Little, D. E.; Trim, S. C.; Johnson, T. R.; Linn, W. S.; Hackney, J. D. (1992) Activity patterns in
elementary and high school students exposed to oxidant pollution. J. Exposure Anal. Environ. Epidemiol.
2: 277-293.
Sprenger, M.; Wernli, H. (2003) A northern hemispheric climatology of cross-tropopause exchange for the ERA 15
time period (1979-1993). J. Geophys. Res. 108(D12): 10.1029/2002JD002636.
Sprenger, M.; Maspoli, M. C.; Wernli, H. (2003) Tropopause folds and cross-tropopause exchange: a global
investigation based upon ECMWF analyses for the time period March 2000 to February 2001. J. Geophys.
Res. (Atmos.) 108(D12): 10.1029/2002JD002587.
Stasiuk, W. N., Jr.; Coffey, P. E. (1974) Rural and urban ozone relationships in New York State. J. Air Pollut.
Control Assoc. 24: 564-568.
Steiber, R. S. (1995) Ozone generators in indoor air settings. Research Triangle Park, NC: U.S. Environmental
Protection Agency, National Risk Management Research Laboratory; report no. EPA-600/R-95-154.
Available from: NTIS, Springfield, VA; PB96-100201.
Stein, M. L.; Fang, D. (1997) Ozone exposure and population density in Harris County, Texas [comment]. J. Am.
Stat. Assoc. 92:408-411.
Stohl, A. (2001) A 1-year Lagrangian "climatology" of airstreams in the Northern Hemisphere troposphere and
lowermost stratosphere. J. Geophys. Res. (Atmos.) 106: 7263-7279.
Sun, W.; Le, N. D.; Zidek, J. V.; Burnett, R. (1998) Assessment of a Bayesian multivariate interpolation approach
for health impact studies. Environmetrics 9: 565-586.
AX3-236
-------
Sun, L.; Zidek, J. V.; Le, N. D.; Ozkaynak, H. (2000) Interpolating Vancouver's daily ambient PM10 field.
Environmetrics 11: 651-663.
Tan, W. W.; Geller, M. A.; Pawson, S.; Da Silva, A. (2004) A case study of excessive subtropical transport in the
stratosphere of a data assimilation system. J. Geophys. Res. (Atmos.) 109(D11102):
10.1029/2003 JD004057.
Taylor, G. E., Jr.; Hanson, P. J. (1992) Forest trees and tropospheric ozone: role of canopy deposition and leaf
uptake in developing exposure-response relationships. Agric. Ecosyst. Environ. 42: 255-273.
Taylor, G. E., Jr.; Norby, R. J. (1985) The significance of elevated levels of ozone on natural ecosystems of North
America. In: Lee, S. D., ed. Evaluation of the scientific basis for ozone/oxidants standards: proceedings of
an APCA international specialty conference; November 1984; Houston, TX. Pittsburgh, PA: Air Pollution
Control Association; pp. 152-175. (APCA transactions: TR-4).
Taylor, G. E., Jr.; Ross-Todd, B. M.; Allen, E.; Conklin, P.; Edmonds, R.; Joranger, E.; Miller, E.; Ragsdale, L.;
Shepard, J.; Silsbee, D.; Swank, W. (1992) Patterns of tropospheric ozone in forested landscapes of the
Integrated Forest Study. In: Johnson, D. W.; Lindberg, S. E., eds. Atmospheric deposition and forest
nutrient cycling: a synthesis of the integrated forest study. New York, NY: Springer-Verlag; pp. 50-71.
(Billings, W. D.; Golley, F.; Lange, O. L.; Olson, J. S.; Remmert, H., eds. Ecological studies: analysis and
synthesis: v. 91).
Teklemariam, T. A.; Sparks, J. P. (2004) Gaseous fluxes of peroxyacetyl nitrate (PAN) into plant leaves. Plant Cell
Environ. 27: 1149-1158.
Thompson, C. R.; Hensel, E. G.; Kats, G. (1973) Outdoor-indoor levels of six air pollutants. J. AirPollut. Control
Assoc. 23: 881-886.
Trainer, M.; Parrish, D. D.; Buhr, M. P.; Norton, R. B.; Fehsenfeld, F. C.; Anlauf, K. G.; Bottenheim, J. W.; Tang,
Y. Z.; Wiebe, H. A.; Roberts, J. M.; Tanner, R. L.; Newman, L.; Bowersox, V. C.; Meagher, J. F.; Olszyna,
K. J.; Rodgers, M. O.; Wang, T.; Berresheim, H.; Demerjian, K. L.; Roychowdhury, U. K. (1993)
Correlation of ozone with NOy in photochemically aged air. J. Geophys. Res. [Atmos.] 98: 2917-2925.
Tsang, A. M.; Klepeis, N. E. (1996) Descriptive statistics tables from a detailed analysis of the national human
activity pattern survey (NHAPS) data. Las Vegas, NV: U.S. Environmental Protection Agency, Office of
Research and Development; report no. EPA/600/R-96/148.
Turk, B. H.; Grimsrud, D. T.; Brown, J. T.; Geisling-Sobotka, K. L.; Harrison, J.; Prill, R. J. (1989) Commercial
building ventilation rates and particle concentrations. ASHRAE Trans. 95(part 1): 422-433.
U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office; report nos. EPA-600/8-84-020aF-eF. 5v. Available from: NTIS, Springfield, VA;
PB87-142949.
U.S. Environmental Protection Agency. (1992) National air quality and emissions trends report, 1991. Research
Triangle Park, NC: U.S. Environmental Protection Agency,Office of Air Quality Planning and Standards;
report no. EPA/450-R-92-001. Available from: NTIS, Springfield, VA; PB93-143998.
U.S. Environmental Protection Agency. (1996a) Air quality criteria for ozone and related photochemical oxidants.
Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF.
3v. Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, andPB96-185608. Available:
http://cfpub2.epa.gov/ncea/.
U.S. Environmental Protection Agency. (1996b) Air quality criteria for paniculate matter. Research Triangle Park,
NC: National Center for Environmental Assessment-RTF Office; report nos. EPA/600/P-95/001aF-cF. 3v.
U.S. Environmental Protection Agency. (1996c) Review of national ambient air quality standards for ozone:
assessment of scientific and technical information. OAQPS staff paper. Research Triangle Park, NC: Office
of Air Quality Planning and Standards; report no. EPA/452/R-96/007. Available from: NTIS, Springfield,
VA; PB96-203435. Available: http://www.epa.gov/ttn/naaqs/standards/ozone/s_o3_pr_sp.html
(9 September 2003).
U.S. Environmental Protection Agency. (1997) Responses to significant comments on the 1996 proposed rule on the
national ambient air quality standards for paniculate matter (December 13, 1996; 61 FR 65638). Research
Triangle Park, NC: National Center for Environmental Assessment; docket number A-95-54; July.
ECAO-CD-92-0671; STD-081.
U.S. Environmental Protection Agency. (1998) Guideline on ozone monitoring site selection. Research Triangle
Park, NC: U.S. Environmental Protection Agency, National Center for Environmental Assessment;
EPA-454/R-98-002.
AX3-237
-------
U.S. Environmental Protection Agency. (2002) Latest findings on national air quality: 2001 status and trends.
Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA-454/K-02-001.
Available: http://www.epa.gov/air/aqtrnd01/ (22 July 2003).
U.S. Environmental Protection Agency. (2003) Technology Transfer Network: Air Quality System (AQS).
Washington, DC: Office of Air and Radiation. Available: http://www.epa.gov/ttn/airs/airsaqs/ [24 August,
2005].
U.S. Environmental Protection Agency. (2004a) Air quality criteria for paniculate matter. Research Triangle Park,
NC: National Center for Environmental Assessment; report no. EPA/600/P-99/002aF-bF. 2v. Available:
http://cfpub.epa.gov/ncea/ [9 November, 2004].
U.S. Environmental Protection Agency. (2004b) The ozone report: measuring progress through 2003. Cincinnati,
OH: U.S. Environmental Protection Agency; report no. EPA454/K-04-001.
Vakeva, M; Ha'meri, K.; Kulmala, M; Lahdes, R.; Ruuskanen, I; Laitinen, T. (1999) Street level versus rooftop
concentrations of submicron aerosol particles and gaseous pollutants in an urban street canyon. Atmos.
Environ. 33: 1385-1397.
Van den Bergh, V.; Vanhees, I.; De Boer, R.; Compernolle, F.; Vinckier, C. (2000) Identification of the oxidation
products of the reaction between a-pinene and hydroxyl radicals by gas and high-performance liquid
chromatography with mass spectrometric detection. J. Chromatogr. A 896: 135-148.
Van Noije, T. P. C.; Eskes, H. J.; Van Weele, M.; Van Velthoven, P. F. J. (2004) Implications of enhanced
Brewer-Dobson circulation in European Centre for medium-range weather forecasts reanalysis ERA-40 for
the stratosphere-troposphere exchange of ozone in global chemistry transport models. J. Geophys. Res.
(Atmos.) 109(D19308): 10.1029/2004 JD004586.
Venkatachari, P.; Hopke, P. K.; Grover, B. D.; Eatough, D. J. (2005a) Measurement of particle-bound reactive
oxygen species inRubidoux aerosols. J. Atmos. Chem. 50: 49-58.
Venkatachari, P.; Hopke, P .K.; Brune, W. H.; Ren, X.; Lesher, R.; Mao, J.; Mitchell, M. (2005b) Characterization
of reactive oxygen species trends in Flushing, New York. Atmos. Environ.: in press.
Vingarzan, R. (2004) A review of surface ozone background levels and trends. Atmos. Environ. 38: 3431-3442.
VonKuhlmann, R.; Lawrence, M. G.; Crutzen, P. J.; Rasch, P. J. (2003) A model for studies of tropospheric ozone
and nonmethane hydrocarbons: model description and ozone results. J. Geophys. Res. [Atmos.] 108(D9):
10.1029/2002JD002893.
Vong, R. J.; Guttorp, P. (1991) Co-occurence of ozone and acidic cloudwater in high-elevation forests. Environ. Sci.
Technol. 25: 1325-1329.
Wainman, T.; Zhang, J.; Weschler, C. J.; Lioy, P. J. (2000) Ozone and limonene in indoor air: a source of submicron
particle exposure. Environ. Health Perspect. 108: 1139-1145.
Wang, Y.; Logan, J. A.; Jacob, D. J. (1998) Global simulation of tropospheric O3-NOx-hydrocarbon chemistry 2.
Model evaluation and global ozone budget. J. Geophys. Res. (Atmos.) 103: 10,727-10,755.
Weinstein-Lloyd, J. B.; Lee, J. H.; Damn, P. H.; Nunnermacker, L. J.; Springston, S. R.; Newman, L. (1998)
Measurements of peroxides and related species during the 1995 summer intensive of the Southern Oxidants
Study in Nashville, Tennessee. J. Geophys. Res. (Atmos.) 103: 22361-22373.
Wernli, H.; Bourqui, M. (2002) ALagrangian "1-year climatology" of (deep) cross-tropopause exchange in the
extratropical Northern Hemisphere. J. Geophys. Res. (Atmos.) 107(D2): 10.1029/2001JD000812.
Weschler, C. J. (2000) Ozone in indoor environments: concentration and chemistry. Indoor Air 10: 269-288.
Weschler, C. J. (2004) Chemical reactions among indoor pollutants: what we've learned in the new millennium.
Indoor Air 14(suppl. 7): 184-194.
Weschler, C. J.; Shields, H. C. (1997) Potential reactions among indoor pollutants. Atmos. Environ. 31: 3487-3495.
Weschler, C. J.; Shields, H. C. (1999) Indoor ozone/terpene reactions as a source of indoor particles. Atmos.
Environ. 33:2301-2312.
Weschler, C. J.; Shields, H. C. (2000) The influence of ventilation on reactions among indoor pollutants: modeling
and experimental observations. Indoor Air. 10: 92-100.
Weschler, C. J.; Shields, H. C. (2003) Experiments probing the influence of air exchange rates on secondary organic
aerosols derived from indoor chemistry. Atmos. Environ. 37: 5621-5631.
Weschler, C. J.; Shields, H. C.; Naik, D. V (1989) Indoor ozone exposures. JAPCA 39: 1562-1568.
Weschler, C. J.; Hodgson, A. T.; Wooley, J. D. (1992) Indoor chemistry: ozone, volatile organic compounds, and
carpets. Environ. Sci. Technol. 26: 2371-2377.
Weschler, C. J.; Shields, H. C.; Naik, D. V. (1994) Indoor chemistry involving O3, NO, and NO2 as evidenced by
14 months of measurements at a site in southern California. Environ. Sci. Technol. 28: 2120-2132.
Wild, O.; Akimoto, H. (2001) Intercontinental transport of ozone and its precursors in a three-dimensional global
CTM. J. Geophys. Res. [Atmos.] 106: 27,729-27,744.
AX3-238
-------
Wiley, J. A.; Robinson, J. P.; Piazza, T.; Garrett, K.; Cirksena, K.; Cheng, Y.-T.; Martin, G. (199la) Activity
patterns of California residents. Final report. Sacramento, CA: California Air Resources Board; report no.
ARB/R93/487. Available from: NTIS, Springfield, VA.; PB94-108719.
Wiley, J. A.; Robinson, J. P.; Cheng, Y.-T.; Piazza, T.; Stork, L.; Pladsen, K. (1991b) Study of children's activity
patterns: final report. Sacramento, CA: California Air Resources Board; report no. ARB-R-93/489.
Wilkins, C. K.; Clausen, P. A.; Wolkoff, P.; Larsen, S. T.; Hammer, M; Larsen, K.; Hansen, V.; Nielsen, G. D.
(2001) Formation of strong irritants in mixtures of isoprene/ozone and isoprene/ozone/nitrogen dioxide.
Environ. Health Perspect. 109: 937-941.
Williams, R.; Creason, J.; Zweidinger, R.; Watts, R.; Sheldon, L.; Shy, C. (2000a) Indoor, outdoor, and personal
exposure monitoring of paniculate air pollution: the Baltimore elderly epidemiology-exposure pilot study.
Atmos. Environ. 34: 4193-4204.
Williams, R.; Suggs, J.; Creason, J.; Rodes, C.; Lawless, P.; Kwok, R.; Zweidinger, R.; Sheldon, L. (2000b) The
1998 Baltimore Paniculate Matter Epidemiology-Exposure Study: part 2. Personal exposure assessment
associated with an elderly study population. J. Exposure Anal. Environ. Epidemiol. 10: 533-543.
Williams, R.; Suggs, J.; Rea, A.; Leovic, K.; Vette, A.; Croghan, C.; Sheldon, L.; Rodes, C.; Thornburg,, J.; Ejire,
A.; Herbst, M.; Sanders, W., Jr. (2003a) The Research Triangle Park paniculate matter panel study: PM
mass concentration relationships Atmos. Environ. 37: 5349-5363.
Williams, R.; Suggs, J.; Rea, A.; Sheldon, L.; Rodes, C.; Thornburg, J. (2003b) The Research Triangle Park
paniculate matter panel study: modeling ambient source contribution to personal and residential PM mass
concentrations. Atmos. Environ. 37: 5365-5378.
Wilson, A. L.; Colome, S. D.; Tian, Y.; Becker, E. W.; Baker, P. E.; Behrens, D. W.; Billick, I. H.; Garrison, C. A.
(1996) California residential air exchange rates and residence volumes. J. Exposure Anal. Environ.
Epidemiol. 6:311-326.
Wolff, G. T.; Lioy, P. J.; Taylor, R. S. (1987) The diurnal variations of ozone at different altitudes on a rural
mountain in the eastern United States. JAPCA 37: 45-48.
Wolff, G. T.; Dunker, A. M.; Rao, S. T.; Porter, P. S.; Zurbenko, I. G. (2001) Ozone air quality over North America:
part I—a review of reported trends. J. Air Waste Manage. Assoc. 51: 273-282.
Wolkoff, P.; Clausen, P. A.; Wilkins, C. K.; Hougaard, K. S.; Nielsen, G. D. (1999) Formation of strong airway
irritants in a model mixture of (+)-a-pinene/ozone. Atmos. Environ. 33: 693-698.
Wong, S.; Wang, W.-C.; Isaksen, I. S. A.; Berntsen, T. K.; Sundet, J. K. (2004) A global climate-chemistry model
study of present-day tropospheric chemistry and radiative forcing from changes in tropospheric O3 since the
preindustrial period. J. Geophys. Res. [Atmos.] 109(D11): 10.1029/2003JD003998.
Zeger, S. L.; Thomas, D.; Dominici, F.; Samet, J. M.; Schwartz, J.; Dockery, D.; Cohen, A. (2000) Exposure
measurement error in time-series studies of air pollution: concepts and consequences. Environ. Health
Perspect. 108: 419-426.
Zhou, J.; Smith, S. (1997) Measurement of ozone concentrations in ambient air using a badge-type passive monitor.
J. Air Waste Manage. Assoc. 47: 697-703.
Zidek, J. V.; White, R.; Le, N. D.; Sun, W.; Burnett, R. T. (1998) Imputing unmeasured explanatory variables in
environmental epidemiology with application to health impact analysis of air pollution. Environ. Ecol. Stat.
5:99-115.
Zidek, J. V.; Meloche, J.; Le, N. D.; Sun, L. (2000) Combining statistical and computer models for health risk
assessment (exposure analysis). In: Nunez-Anton, V.; Feffeirra, E., eds. Proceedings of New Trends in
Statistical Modelling, the 15th international workshop on statistical modelling; Bilbao, Spain. Stat. Model.:
pp. 95-106.
Zidek, J.; Sun, L.; Le, N.; Ozkaynak, H. (2002) Contending with space-time interaction in the spatial prediction of
pollution: Vancouver's hourly ambient PM10 field. Environmetrics 13: 595-613.
Zidek, J. V.; Shaddick, G.; White, R.; Meloche, J.; Chatfield, C. (2005) Using a probabilistic model (pCNEM) to
estimate personal exposure air pollution. Environmetrics 16: 481-493.
AX3-239
-------
ANNEX AX4. DOSIMETRY OF OZONE IN
THE RESPIRATORY TRACT
AX4.1 INTRODUCTION
This annex serves to provide supporting material for Chapter 4—Dosimetry, Species,
Homology, Sensitivity, and Animal-to-Human Extrapolation. It includes tables that summarize
new literature published since the 1996 Ozone Air Quality Criteria Document or O3 AQCD (U.S.
Environmental Protection Agency, 1996). In addition, it provides descriptions of the new
findings, in many cases, with more detail than is provided in the chapter.
Dosimetry refers to the measurement or estimation of the quantity of or rate at which a
chemical and/or it reaction products are absorbed and retained at target sites within the
respiratory tract (RT). The compound most directly responsible for toxic effects may be the
inhaled gas, ozone (O3), or one of its chemical reaction products (e.g., aldehydes or peroxides).
Complete identification of the actual toxic agents and their integration into dosimetry is a
complex issue that has not been fully resolved. Thus, most dosimetry investigations are
concerned with the dose of the primary inhaled chemical. In this context, a further confounding
aspect can be the units of dose (e.g., mass retained per breath, mass retained per breath per body
weight, mass retained per breath per RT surface area). That is, when comparing dose between
species, what is the relevant measure of dose? This question has not been answered; units are
often dictated by the type of experiment or by a choice made by the investigators. There is also
some lack of agreement as to what constitutes "dose." Dahl's seminal paper (1990) classified O3
as a reactive gas and discussed the characterization of dose measurement by parameters,
including: (1) inhaled O3 concentration; (2) amount of O3 inhaled as determined by minute
volume, vapor concentration, and exposure duration; (3) uptake or the amount of O3 retained
(i.e., not exhaled); (4) O3 or its active metabolites delivered to target cells or tissues; (5) O3 or its
reactive metabolites delivered to target biomolecules or organelles; and (6) O3 or its metabolites
participating in the ultimate toxic reactions-the effective dose. This characterization goes from
AX4-1
-------
least complex to greatest, culminating in measurement of the fraction of the inhaled O3 that
participates in the effects of cellular perturbation and/or injury.
Understanding dosimetry as it relates to O3-induced injury is complex due to the fact
that O3 interacts primarily with the epithelial lining fluid (ELF), which contains surfactant and
antioxidants. Ozone reacts with molecules in the ELF to create reactive metabolites, which can
then diffuse within the lung or be transported out of the lung to generate systemic effects.
Antioxidant defenses can vary between species, region of the RT, and animal age. Section 5.3.1
contains further information on the cellular targets of O3 interactions and antioxidants.
Experimental dosimetry studies in laboratory animals and humans, as well as theoretical
(dosimetry modeling) studies, have been used to obtain information on O3 dose. Since the
1996 O3 AQCD (U.S. Environmental Protection Agency, 1996), most new experiments have
been carried out in humans to obtain direct measurements of absorbed O3 in the RT, the upper
RT (URT) region proximal to the tracheal entrance, and in the tracheobronchial (TB) region.
Experimentally obtaining dosimetry data is extremely difficult in smaller RT regions or
locations, such as specific airways or the centriacinar region (CAR; junction of conducting
airways and gas exchange region), where lesions caused by O3 occur. Nevertheless,
experimentation is important for determining dose, making dose comparisons between
subpopulations and between different species, assessing hypotheses and concepts, and validating
mathematical models that can be used to predict dose at specific RT sites and under more
general conditions.
Theoretical studies are based on the use of mathematical models developed for the
purposes of simulating the uptake and distribution of absorbed gases in the tissues and fluids of
the RT. Because the factors affecting the transport and absorption of gases are applicable to all
mammals, a model that uses appropriate species or disease-specific anatomical and ventilatory
parameters can be used to describe absorption in the species and in different-sized, aged, or
diseased members of the same species. More importantly, models also may be used to make
interspecies and intraspecies dose comparisons, to compare and reconcile data from different
experiments, to predict dose in conditions not possible or feasible experimentally, and to better
understand the processes involved in toxicity. However, virtually all of these models have
assumed that the reaction rate of O3 in the liquid lining layer and in tissues is quasi first-order
AX4-2
-------
with respect to O3 concentration. There is considerable discrepancy between rate constants used
in models. Both the uptake of O3 at the gas-liquid interface and the fraction of O3 (or its reaction
products) that reach epithelial cells are sensitive to the value of these reaction rate constants.
A large disparity in rate constants between studies illustrates the limitations in our current state
of knowledge and can potential lead to erroneous interpretation of O3 model predictions.
A review (Miller, 1995) of the factors influencing RT uptake of O3 stated that structure of
the RT region, ventilation, and gas transport mechanisms were important. Additionally, local
dose is the critical link between exposure and response. A more detailed discussion of
experimental and theoretical dosimetry studies is available in the 1996 O3 AQCD (Volume III,
Chapter 8, U.S. Environmental Protection Agency, 1996).
AX4.2 EXPERIMENTAL OZONE DOSIMETRY INVESTIGATIONS
There have been some advances in understanding human O3 dosimetry that better enable
quantitative extrapolation from laboratory animal data. The next two sections review the
available new experimental studies on O3 dosimetry, which involve only human subjects and are
all from the same laboratory. Of the studies considered in the following discussion, five
involved the use of the bolus response method as a probe to obtain information about the
mechanism of O3 uptake in the URT and TB regions. Of the remaining two investigations, one
focused on total uptake by the RT and the other on uptake by the nasal cavities. Table AX4-1
provides a summary of the newer human studies.
AX4.2.1 Bolus-Response Studies
The bolus-response method has been used as a probe to explore the effects of physiological
and anatomical differences or changes on the uptake of O3 by human beings.
Asplund et al. (1996) studied the effects of continuous O3 inhalation on O3 bolus uptake,
and Rigas et al. (1997) investigated the potential effects of continuous coexposure to O3,
nitrogen dioxide (NO2), or sulfur dioxide (SO2) on O3 absorption. In both of these studies,
AX4-3
-------
Table AX4-1. New Experimental Human Studies on Ozone Dosimetrya
X
-k
Purpose/Objective
Determine the effect
of continuous O3
inhalation on O3
uptake
Subject
Characteristics
8 male,
3 female,
22-3 1 years old,
166-186 cm,
64-93 kg
Region of
Interest
Central
conducting
airways
(70-120 mL
from lips)
Breathing
Patterns/Exposure
2 h of continuous exposure at
rest: 0.0, 0.12, and 0.36 ppm
O3. Spontaneous breathing.
Bolus test breaths every
30 min using 250 mL/sec
constant flow rate.
Results Reference
Averaged over all subjects and the four Asplund et al.
measurement intervals, the absorbed fraction (1996)
(AF) changed +0.045, -0.005, and -0.03 for
the 0, 0.12, and 0.36 ppm continuous
exposures, respectively. These changes are
approximately +6, - 1, and -4 % based on an
average AF value of 0.7 in the range of 70 to
120 mL.b Both nonzero exposures were
significantly different than the air exposure.
Evaluate the
influence of VD on
intersubject variation
of O3 dose.
Investigate the
effect of continuous
exposure to O3,
nitrogen dioxide and
sulfur dioxide on O3
absorption.
10 male, Conducting
22-30 years old, airways
163-186 cm,
64-92 kg;
10 female,
22-35 years old,
149-177 cm,
48-81 kg
6 male, 21-29 years Lower
Bolus-response test (VT = 500
mL at 250 mL/sec constant
flow rate). Fowler single-
breath N2 washout method to
determine VD.
old, 165-185 cm,
60-92 kg;
6 female,
19-33 years old,
152-173 cm,
48-61 kg
conducting
airways
(70-120 mL
from lips)
2 h of continuous exposure
at rest: O3 (0, 0.36 ppm),
SO2 (0, 0.36 ppm), or NO2
(0, 0.36, 0.75 ppm).
5-min Bolus test every
30 min: VT = 500 mL;
250 mL/sec constant
flow rate.
On average, for the same VP, women had a Bush et al.
larger AF than men; women had a smaller VD (1996a)
than men. However, for the same value of
VP/VD, AF for men and women were
indistinguishable. Further analysis indicated
"that previously reported gender differences
may be due to a failure in properly accounting
for tissue surface within the conducting
airways."
Averaging over all subjects or by gender, all Rigas et al.
exposures except O3 resulted in an increase of (1997)
AF. Based on an AF reference valueb, the
change in AF ranged from - 3 to +7 %.
Only the O3 and the NO2 (0.36 ppm)
exposures were significantly different from
the air exposures.
-------
Table AX4-1 (cont'd). New Experimental Human Studies on Ozone Dosimetry!
Purpose/Objective
Compare the
absorption of
chlorine and O3.
Determine how the
physical-chemical
properties of these
compounds affect
their uptake
distribution in
the RT.
Subject
Characteristics
5 male, 2 1-26 years
old, 168-198 cm,
64-95 kg; 5 female,
18-28 years old,
162-178 cm,
55-68 kgc
Region of
Interest
Conducting
airways.
Nasal and
oral routes
Breathing
Patterns/Exposure
Bolus-response technique;
VT = 500 mL; 3 constant flow
rates: 150, 250, and
1000 mL/sec.
Results
Ozone dose to the URT was sensitive to the
mode of breathing and to the respiratory rate.
With increased airflow rate, O3 retained by the
upper airways decreased from 95 to 50%.
TB region dose ranged from 0 to 35%.
Mass transfer theory indicated that the
diffusion resistance of the tissue phase is
important for O3. The gas phase resistances
were the same for O3 and C12.
Reference
Nodelman
and Ultman
(1999a)c
X
To determine O3
uptake relative to
inhaled O3 dose.
5 male,
5 female,
18-35 years old,
175 ± 13 (SD) cm,
72 ± 13 (SD) kg
Respiratory Breath-by-breath calculation
tract; oral of O3 retention based on data
breathing from fast response analyzers
for O3 and airflow rates.
Oral breathing: 0.2 or
0.4 ppm O3 at VE of
approximately 20 L/min
for 60 min or 40 L/min for
30 min.
The FA for all breaths was 0.86.
Concentration, minute volume, and time have
small but statistically significant effects on AF
when compared to intersubject variability.
The investigators concluded: for a given
subject, constant O3 exposure, a given exercise
level, and time <2 h, inhaled dose is a
reasonable surrogate for the actual uptake
ofO3. However, the actual doses may vary
considerably among individuals who are
exposed to similar inhaled doses.
Rigas et al.
(2000)
-------
Table AX4-1 (cont'd). New Experimental Human Studies on Ozone Dosimetry!
Purpose/Objective
Subject
Characteristics
Region of
Interest
Breathing
Patterns/Exposure
Results
Reference
X
Study the effect of
gas flow rate and O3
concentration on O3
uptake in the nose.
Evaluate intersubject
variability in O3
uptake; correlate
differences in
breathing pattern and
lung anatomy with O3
uptake
7 male, 3 female,
26 ± years,
170 ± 11 (SD)cm,
75 ± 20 (SD) kg
nonsmokers,
32 male, 22.9 ± 0.8
years old,178±l cm,
80.6 ± 2.5kg;
28 female,
22.4 ± 0.9 years old,
166 ± 1 cm,
62.1 ±2.2 kg
Nasal For a given flow rate and
cavities exposure concentration, the
subjects inhaled through one
nostril and exhale through the
other. For two 1-h sessions,
a series of 9-12 measurements
of AF were carried out for
10 s each: (1) O3 exposure
concentration = 0.4 ppm; flow
rates = 3, 5, 8, and 15 L/min.
(2) O3 exposure = 0.1, 0.2,
and 0.4 ppm; flow rate =
15 L/min. (3) O3 exposure
= 0.4 ppm, flow rate
= 15 L/min; AF was measured
every 5 min for 1 h.
Respiratory Continuous: 1 h exposure to
tract; oral 0.25 ppm, exercising at
breathing 30L/min.
Bolus: breath-by-breath
calculation of O3 retention.
Timing of bolus varied to
create penetration volumes of
10to250mL. Peak inhaled
bolus of ~1 ppm.
(1) With the exposure concentration at
0.4 ppm O3, AF decreased from 0.80 to
0.33 when the flow rate was increased from
3 to 15 L/min. (2) At a flow rate of 15 L/min,
the AF changed from 0.36 to 0.32 when the
exposure concentration increased from 0.1 to
0.4 ppm O3. (3) Statistical analysis indicated
that the AF was not associated with the time at
which the measurement was taken.
Continuous: Fractional O3 uptake efficiency
ranged from 0.70 to 0.98 (mean 0.89 ± 0.06).
Inverse correlation between uptake and
breathing frequency. Direct correlation
between uptake and tidal volume. Intersubject
differences in forced respiratory responses not
due to differences in O3 uptake.
Bolus: The penetration volume at which 50%
of the bolus was taken up was 90.4 mL in
females and 107 mL in males. Distribution of
O3 shifts distally as the size of the airway
increases.
Santiago
etal. (2001)
Ultman et al.
(2004)
a See Appendix A for abbreviations and acronyms.
b Fig. 4, Hu et al. (1994), for the 250 mL/s curve and penetration volume range of 70 to 120 mL; the average AF is approximately 0.7.
0 Subject characteristics are from Nodelman and Ultman (1999b).
-------
subjects were exposed "continuously" to a gas for 2 h. Every 30 min, breathing at 250 mL/s,
a series of bolus test breaths was performed targeted at the lower conducting airways.
Differences in bolus-response absorbed fraction from an established baseline indicated the
degree to which the "continuous" gas exposure affected the absorption of O3. Depending on the
gas and concentration, changes in absorbed fraction ranged from -3 to +7 % (see Table AX4-1).
Continuous O3 exposure lowered the uptake of O3, whereas NO2 and SO2 increased the uptake
of O3. The investigators concluded that in the tested airways, NO2 and SO2 increased the
capacity to absorb O3 because more of the compounds oxidized by O3 were made available.
On the other hand, they conjectured that continuous O3 exposure depleted these compounds,
thereby reducing O3 uptake.
Bush et al. (1996a) investigated the effect of lung anatomy and gender on O3 absorption in
the conducting airways during oral breathing using the bolus-response technique. Absorption
was measured using this technique applied to 10 men and 10 women. Anatomy was defined in
terms of forced vital capacity (FVC), total lung capacity (TLC), and dead space (VD). The
absorbed fraction data were analyzed in terms of a function of penetration volume, airflow rate,
and an "intrinsic mass transfer parameter" called the overall mass transfer coefficient (Ka),
which was determined for each subject and found to be highly correlated with VD, but not with
height, weight, age, gender, FVC, or TLC. That is, in all subjects, whether men or women,
dosimetry differences could be explained by differences in VD. Combining their data with those
of Hu et al. (1994), where absorbed fraction was determined for several flow rates, Bush et al.
(1996a) inferred that Ka was proportional to flow rate/VD. The investigators point out that the
applicability of their results may be limited because of their assumptions that Ka was
independent of location in the RT and that there was no mucous resistance. They also suggested
that the dependence of Ka on flow rate and VD be restricted to flow rates < 1000 mL/s until
studies at higher rates have been performed.
With flow rates of 150, 250, and 1000 mL/s, Nodelman and Ultman (1999b) used the
bolus-response technique (a) to compare the uptake distributions of O3 and chlorine gas (C12) and
(b) to investigate how their uptakes were affected by their physical and chemical properties.
Ozone dose to the URT was found to be sensitive to the mode of breathing and to the airflow
rate. With increased rate, O3 retained by the upper airways decreased from 95 to 50% and TB
AX4-7
-------
region dose increased from 0 to 35%. At the highest flow rate only 10% of the O3 reached the
pulmonary region. Data were analyzed by expressing the overall diffusion resistance between
the respired gas and the epithelial cells as the sum of individual gas-phase and liquid-phase
resistances. The liquid-phase resistance made an important contribution to the overall resistance,
implying that changes in the reaction rate of O3 in ELF would cause concomitant changes in
the O3 uptake. The gas-phase resistance was inversely related to the volumes of the oral and
nasal cavities during oral and nasal breathing, respectively.
Ultman et al. (2004) used separate bolus and continuous exposure sessions to test the
hypotheses that differences in O3 uptake in lungs are responsible for variation in O3-induced
changes in lung function parameters and that differences in O3 uptake are due to variations in
breathing patterns and lung anatomy. Thirty-two males and 28 female nonsmokers were
exposed to bolus penetration volumes ranging from 10 to 250 mL, which was determined by the
timing of the bolus injection. The subjects controlled their breathing to generate a target respired
flow of 1000 mL/sec. At this high minute ventilation, there was very little uptake in the upper
airway and most of the O3 reached areas where gas exchange takes place. To quantify
intersubject differences in O3 bolus uptake, they measured the penetration volume at which 50%
of the O3 was taken up. Values for penetration volume ranged from 69 to 134 mL and were
directly correlated with the subjects' values for anatomic dead space volume. A better
correlation was seen when the volume of the upper airways was subtracted. The penetration
volume at which 50% of the bolus was taken up was 90.4 mL in females and 107 mL in males.
This significant difference in uptake suggests to the authors that, in females, the smaller airways
and associated larger surface-to-volume ratio enhance local O3 uptake and cause reduced
penetration of O3 into the distal lung. Data from the continuous exposure sessions indicated that
overall O3 uptake was related to breathing pattern but was not correlated with VD. Thus, these
findings indicate that overall O3 uptake is not related to airway size, but that the distribution
of O3 shifts distally as the size of the airway in increased.
General comment on estimating mass transfer coefficients. Bush et al. (1996b) and Nodelman
and Ultman (1999a) used a simple model to analyze their bolus-response data. This model,
presented by Hu et al. (1992, 1994), assumed steady-state mass transfer by convection (but no
AX4-8
-------
dispersion) and the mass transfer of O3 to the walls of a tube of uniform cross-sectional area.
These assumptions led to an analytical solution (for the absorbed fraction) which was a function
of an "overall mass transfer coefficient," penetration volume, and airflow rate. As the
investigators have shown, the model is very useful for statistical analysis and hypothesis testing.
Given the absorbed fraction data, the model overall mass transfer coefficients were estimated for
each flow rate. In those bolus-response studies that used this method to analyze data, there was
no discussion of the models' "accuracy" in representing mass transfer in the human RT with
respect to omitting dispersion. In addition, the formulation of the gas phase mass transfer
coefficient does not take into account that it has a theoretical lower limit greater than zero as the
airflow rate goes to zero (Miller et al., 1985; Bush et al., 2001). As a consequence, there is no
way to judge the usefulness of the values of the estimated mass transfer coefficients for
dosimetry simulations that are based on convection-dispersion equations, or whether or not the
simple model's mass transfer coefficients, as well as other parameters derived using these
coefficients, are the same as actual physiological parameters.
AX4.2.2 General Uptake Studies
Rigas et al. (2000) performed an experiment to determine the ratio of O3 uptake to the
quantity of O3 inhaled (fractional absorption, FA). Five men and five women were exposed
orally to 0.2 or 0.4 ppm O3 while exercising at a minute volume of approximately 20 L/min for
60 min or 40 L/min for 30 min. Ozone retention was calculated from breath-by-breath data
taken from fast response analyzers of O3 and airflow rates. The FA was statistically analyzed in
terms of subject, exposure concentration, minute volume, and exposure time.
Fractional absorption ranged from 0.56 to 0.98, with a mean ± SD of 0.85 ± 0.06 for all
2000 recorded breaths. Intersubject differences had the largest influence on FA, resulting in a
variation of approximately 10%. Statistical analysis indicated that concentration, minute
volume, and exposure time had statistically significant effects on FA. However, relatively large
changes in these variables were estimated to result in relatively small changes in FA. Note: the
quantity of O3 retained by the RT is equal to FA times the quantity of O3 inhaled; thus, relatively
large changes in concentration, minute volume, or exposure time may result in relatively large
AX4-9
-------
changes in the amount of O3 retained by the RT or absorbed locally. Also, according to Overton
et al. (1996), differences in PAR dose due to anatomical variability may be considerably larger
than corresponding small changes in FA would indicate.
Santiago et al. (2001) studied the effects of airflow rate and O3 concentration on O3 uptake
in the nasal cavities of three women and seven men. Air was supplied at a constant flow rate to
one nostril and exited from the other nostril while the subject kept the velopharyngeal aperture
closed by raising the soft palate. Thus, a constant unidirectional flow of air plus O3 was
restricted to the nasal cavities. The fraction of O3 absorbed was calculated using the inlet and
outlet concentrations. Inlet concentration and airflow rate were varied in order to determine their
effect on O3 uptake.
The mean FA decreased from 0.80 to 0.33, with an increase in flow rate from 3 to
15 L/min. The effect of both flow rate and subject on FA was statistically significant. Further
analysis indicated that the overall mass transfer coefficient was highly correlated with the flow
rate and that the gas phase resistance contributed from 6.3% (15 L/min) to 23% (3 L/min) of the
total resistance to O3 transfer to the nasal cavity surface. Concentration had a small, but
statistically significant effect on FA, when the inlet concentration was increased from 0.1 to
0.4 ppm O3, FA decreased from 0.36 to 0.32. The investigators observed that differences in FA
among subjects were important; generally, subject variability accounted for approximately half
of the total variation in FA.
As mentioned above, Ultman et al. (2004) tested hypotheses that differences in O3 uptake
in lungs are responsible for variation in O3-induced changes in lung function parameters and that
differences in O3 uptake are due to variations in breathing patterns and lung anatomy. Thirty-
two males and 28 female nonsmokers were exposed continuously for 1 h to either clean air or
0.25 ppm O3 while exercising at a target minute ventilation of 30 L/min. Ultman et al. first
determined the forced expiratory response to clean air, then evaluated O3 uptake measuring dead
space volume, cross-sectional area of peripheral lung (Ap) for CO2 diffusion, FEVj, FVC,
and FEF250/0.75o/0. The fractional O3 uptake efficiency ranged from 0.70 to 0.98, with a mean of
0.89 ± 0.06. They found an inverse correlation between uptake and breathing frequency and a
direct correlation between uptake and tidal volume. There was a small, but statistically
significant decrease in uptake efficiency during the four sequential 15 min intervals of the 1 h
AX4-10
-------
exposure (0.906 ± 0.058 vs. 0.873 ± 0.088, first and last interval, respectively), in part due to the
increased breathing frequency and decreasing tidal volume occurring over the same period.
Ozone uptake rate correlated with individual %Ap, but did not correlate with individual "/oFEVj.
Neither of these parameters correlated with the penetration volume determined in the bolus
studies mentioned above. The authors concluded that the intersubject differences in forced
respiratory responses were not due to differences in O3 uptake. However, these data did partially
support the second hypothesis, i.e., that the differences in cross-sectional area available for gas
diffusion induce differences in O3 uptake.
Plopper et al. (1998) examined the relationship between O3 dose and epithelial injury in
Rhesus monkeys. Using 18O content in lung tissues, the respiratory bronchioles were confirmed
as the site receiving the greatest O3 dose (mass 18O per dry lung weight). Furthermore, the
greatest cellular injury occurred in the vicinity of the respiratory bronchioles and was dependent
on the delivered O3 dose to these tissues. After a 2 h exposure, the antioxidant glutathione
(GSH) was increased in the proximal intrapulmonary bronchus after 0.4 ppm O3 and decreased
in the respiratory bronchiole after 1.0 ppm O3. Perhaps an adaptive response, chronic O3
exposure leads to increased GSH levels in distal bronchioles of both rats and monkeys relative to
GSH levels in FA-exposed animals (Duan et al., 1996). Differences in the levels of antioxidants
between species and regions of the lung do not appear to be the primary factor determining
susceptibility to O3-induced tissue injury (Duan et al ., 1993, 1996). Plopper et al
. (1998) concluded that, in monkeys, there was a close association between site-specific O3 dose,
the degree of epithelial injury, and reduced-glutathione depletion. Within a species, antioxidant
defenses against O3 can also vary with animal age (Servias et al ., 2005) and exposure history
(Duanetal., 1996).
AX4.2.3 Dosimetry Modeling
When all of the animal and human in vivo O3 uptake efficiency data are compared, there is
a good degree of consistency across data sets (U.S. Environmental Protection Agency, 1996).
This agreement raises the level of confidence with which these data sets can be used to support
dosimetric model formulations.
AX4-11
-------
Schelegle et al. (2001) reported that the terminal bronchioles supplied by short and long
paths had similar epithelial injury in rats exposed to 1 ppm O3 for 8 h. Interestingly, in rats with
a C-fiber conduction block to prevent O3-induced rapid shallow breathing, it was the long path
terminal bronchioles that received the greatest epithelial injury. Overall, O3-induced rapid
shallow breathing appears to protect the large conducting airways while producing a more even
distribution of injury to the terminal bronchioles (load et al ., 2000; Schelegle et al ., 2001).
Postlethwait et al . (2000) have also identified the conducting airways as a primary site of
acute O3-induced cell injury. Such data must be considered when developing models that
attempt to predict site-specific locations of O3-induced injury. The early models computed
relationships between delivered regional dose and response with the assumption that O3 was the
active agent responsible for injury. It is now known that reactive intermediates, e.g., aldehydes
and hydrohydroxyperoxides are important agents mediating the response to O3 (further discussed
in Section 5.3.1). Thus, models must consider O3 reaction/diffusion in the ELF and ELF-derived
reactions products.
Table AX4-2 presents a summary of new theoretical studies on the uptake of O3 by the RTs
(or regions) of humans and laboratory animals that have become available since the 1996 O3
AQCD review. They are discussed below.
Overton and Graham (1995) described the development and simulation results of a
dosimetry model that was applied to a TB region anatomical model that had asymmetrically
branching airways, but which had identical single-path pulmonary units distal to each terminal
bronchiole. The anatomical model of the TB region was based on Raabe et al. (1976), which
reported lung cast data for the TB region of a 330 g rat. In general, the modeled O3 dose to lung
tissues varied among anatomically equivalent ventilatory units as a function of path length from
the trachea, with shorter paths showing greater dose. This conflicts with Schelegle et al . (2001),
who exposed rats to 1 ppm O3 for 8 h and found that the terminal bronchioles supplied by
short and long paths had similar epithelial injury. In rats with a C-fiber conduction block to
prevent O3-induced rapid shallow breathing, it was the long path terminal bronchioles that
received the greatest epithelial injury.
AX4-12
-------
Table AX4-2. New Ozone Dosimetry Model Investigations"
Purpose/Objective
Type of mass transport
model/Anatomical model b
Species/RT region of
interest/Regional
anatomical models
Ventilation and
Exposure
Results
Reference
To describe an RT
dosimetry model that
uses a branching TB
region anatomical
model and to
illustrate the results
of its application to a
rat exposed to O3.
To incorporate into
the CFD model of
Kimbelletal. (1993)
resistance to mass
transfer in the nasal
lining and to
investigate the
effects of this lining
on O3 uptake.
To determine if the
single-path model is
able to simulate
bolus inhalation data
recorded during oral
breathing at quiet
respiratory flow.
Rat/RT/URT:
Patraetal. (1987).
TB: multiple path
model of Raabe et al.
(1976).
PUL: Mercer etal.
(1991).
One-dimensional (along axis
of airflow), time-dependent,
convection-dispersion
equation of mass transport
applied to each airway or
model segment. URT: single
path; TB: asymmetric
branching airways. PUL:
single path anatomical model
distal to each terminal
bronchiole.
Three dimensional steady-
state Navier-Stokes equations
for solving air velocity flow
field. Three dimensional
steady-state convection-
diffusion equation for O3
transport.
Three-dimensional CFD
model of the nasal passages
of a rat.
Single-path, one-dimensional Human/RT/URT
Rat/nasal passages
Nasal passages:
Kimbelletal. (1993).
(along axis of airflow),
time-dependent, convection-
dispersion equation of mass
transport. Single-path
anatomical model.
(oral): Olson et al.
(1973).
LRT: Weibel(1963).
f= ISObpm;
VT=1.5,2.0,
2.5 mL.
One constant
concentration.
Steady-state
unidirectional
("inhalation")
flow rate
= 576mLl/min.
One constant
concentration.
VT = 500 mL,
f= 15bpm,
constant flow
rate = 250 mL/s.
Bolus-response
simulations.
(protocol used is
described by Hu
etal., 1992).
(1) For VT = 2.0 mL, f = 150 bpm: The general shape
of the dose versus generation plot along any path from
the trachea to a sac is independent of path: generally
the tissue dose decreases with increasing generation
index. In the TB region, the coefficient of variation for
dose ranges from 0 to 34 %, depending on generation.
The maximum ratio of the largest to smallest dose in
the same generation is 7; the average ratio being 3.
In the first PUL region model segment, the coefficient
of variation for the dose is 29 %. (2) The average dose
to the first PUL region model segment increases with
increasing VT.
Predictions of overall uptake were within the range of
measured uptake. Results suggest that mucus resistance
is important for describing O3 dosimetry and this
thickness may play a role for determining patterns
of O3-induced lesions in the rat nasal passage.
Simulations are sensitive to conducting airway volume
but are relative insensitive to characteristics of the
respiratory airspace. Although the gas-phase resistance
to lateral diffusion limits O3 absorption during quiet
breathing, diffusion through mucus may become
important at the large respiratory flows that are
normally associated with exercise. The single-path
convection-diffusion model was a reasonable approach
/to simulate the bolus-response data.
Overton and
Graham
(1995)
Cohen-
Hubal et al.
(1996)
Bush et al.
(1996b)
-------
Table AX4-2 (cont'd). New Ozone Dosimetry Model Investigations a
>
Purpose/Objective
To assess age- and
gender-specific
differences in
regional and
systemic uptake.
To examine the
impact on predictions
due to the value used
for the TB region
volume at FRC and
due to TB region
volume change
during respiration.
To make parameter
modifications so that
a single-path model
would simulate AF
from bolus-response
experiments
involving O3
(and C12).
Type of mass transport
model/Anatomical model b
PBPK, at ages 1, 3, 6, months
and 1,5, 10, 15,25, 50, and
75 years.
Single-path, one-dimensional
(along axis of airflow),
time-dependent, convection-
dispersion equation of mass
transport. Single-path
anatomical model.
Single-path, one-dimensional
(along axis of airflow), time-
dependent, convection-
dispersion equation of mass
transport. Single-path
anatomical model.
Species/RT region of
interest/Regional
anatomical models
Human/ET/TB
Human/RT/
URT: Nunnetal.
(1959)
LRT: Weibel(1963)
Rat/RT/URT:
Patraetal. (1987)
TB: Yehetal. (1979)
PUL: Mercer etal.
(1991)
Human/RT/URT
(oral): Olson et al.
(1973).
URT (nasal):
Olson etal. (1973)
and Guilmette et al.
(1989)
LRT: Weibel(1963).
Ventilation and
Exposure
Pulmonary
ventilation ranged
from 34 mL/s
(in 1 -month-old)
to 190 mL/s
(in 15-year-old).
VT varied with age.
Human: VT = 500,
2250mL;f=15,
30 bpm.
Rat: VT=1.4,
2.4mL;f=96,
157 bpm.
One constant
concentration.
Oral and nasal
breathing.
Flow rates =150,
250, 1000 mL/s,
VT = 500 mL.
Bolus-response
simulations.
Results
Regional extraction is insensitive to age. Extraction per
unit surface area is 2- to 8-fold higher in infants
compared to adults. PU and ET regions have a large
increase in unit extraction with increasing age. Early
postnatal period is time of largest differences in PK,
due to immaturity of metabolic enzymes.
(1) A better understanding and characterization of the
role of TB region expansion (mainly the rat) and
volume is important for an improved understanding of
respiratory-tract dosimetry modeling of reactive gases.
(2) Extrapolations based on dose in the PAR can differ
significantly from those based on exposure
concentration or total uptake.
(3) Human subjects who appear similar outwardly may
have very different PAR doses and potentially different
responses to the same exposure.(Uptake by the URT
was not considered.)
(Simulation results for O3only) (1) Using parameter
values from the literature and assuming that absorption
was gas-phase controlled, the simulations of O3 data
were realistic at flow rate = 250 mL/s, but not realistic
at 1000 mL/s. (2) Accurate simulations at 250 mL/s
required modification of mass transfer coefficients
reported in the literature for the conducting airways.
(3) It was necessary to include a diffusion resistance for
the ELF based on an assumed O3 reaction rate constant
that was much greater than in in vitro estimates.
(4) Partial validation of the final parameters
(determined at 250 mL/s) was obtained by simulations
of bolus-response data at flow rates of 150 and 1000
mL/s. Validation was obtained also by simulating
internal measurements of O3 in subjects exposed during
quiet breathing.
Reference
Sarangapani
et al. (2003)
Overton
etal. (1996)
Bush et al.
(2001)
aSee Appendix A for abbreviations and acronyms.
bThe anatomical models used in an investigation generally differ from those described in the references, e.g., dimensions are often scaled to dimensions appropriate to the
dosimetry investigation; or the original structure may be simplified, keeping or scaling the original dimensions.
-------
Rat effects data (from the PAR) are available that are identified with the lobe and the
generation in the lobe from which tissue samples were obtained (Pinkerton et al., 1995, 1998).
Models, like Overton et al. (1996), can be helpful in understanding the distribution of the
magnitude of such effects as well as suggesting sampling sites for future experiments.
Using computational fluid dynamics (CFD), Cohen-Hubal et al. (1996) explored the effect
of the mucus layer thickness in the nasal passage of a rat. The nasal lining was composed of
mucus and tissue layers in which mass transfer was by molecular diffusion with first order
chemical reaction. Physicochemical parameters for O3 were obtained from the literature. Three
scenarios were considered: 10 jim thick mucus layer, no mucus layer, and two nasal passage
regions each with a different mucus layer thickness. Predictions of overall uptake were within
the range of measured uptake. Predicted regional O3 flux was correlated with measured cell
proliferation for the CFD simulation that incorporated two regions, each with a different mucus
thickness. The reaction rate constant used by Cohen-Hubal and co-workers may be too low.
Using bolus-response data, Hu et al. (1994) and Bush et al. (2001) estimated a reaction rate
constant that is more than a 1000 times as large as that used by Cohen-Hubal et al. (1996).
A rate constant this large could result in conclusions different than those based on the smaller
constant.
With an RT dosimetry model, Overton et al. (1996) investigated the sensitivity of
absorbed fraction (AF), proximal alveolar region (PAR) dose, and PAR dose ratio to TB region
volume (V-YB) and TB region expansion in human beings and rats. The PAR was defined as the
first generation distal to terminal bronchioles and the PAR dose ratio was defined as the ratio of
a rat's predicted PAR dose to a human's predicted PAR dose. This ratio relates human and rat
exposure concentrations so that both species receive the same PAR dose. In rats the PAR is a
region of major damage from O3. For each species, three literature values of Vra were used:
a mean value and the mean ± twice the SD. The following predictions were obtained:
(1) The sensitivity of AF and PAR dose to V^ depends on species, ventilation, TB region
overall mass transfer coefficient (k^), and expansion. For k^ = 0.26 cm/s and quiet breathing,
AF was predicted to vary by less than 3% for the ±2 SD range of VTB. In contrast, the PAR
dose predicted for the smallest VTB is five times larger than the PAR dose predicted with the
AX4-15
-------
largest VTB. The effect of V^ is much less during heavy exercise: the ratio of maximum to
minimum PAR dose was approximately 1.5. In any case, the simulations predicted that
fractional changes in AF due to different V^ are not, in general, a good predictor of the
fractional changes in PAR doses.
(2) Relative to no expansion in the TB region, expansion decreases both AF and PAR
dose. The largest effect of including expansion in the human simulations was to decrease the AF
by «8%; in rats, the maximum decrease was -45%. The PAR doses decreased relatively more,
25 and 65% in human beings and rat, respectively.
(3) The authors attempted to obtain an understanding as to uncertainty or variability in
estimates of exposure concentrations (that give the same PAR dose in both species) if the
literature mean value of VTB was used. For various values off, VT, 1%,, and expansion, the PAR
dose ratios at upper and lower values of VTB deviated in absolute values from the PAR dose ratio
calculated at the mean values of VTB by as little as 10% to as large as 310%. The smallest
deviation occurred at the largest VT and smallest kra for both species; whereas, the largest
deviation occurred at the smallest VT and largest kra for both species.
Bush et al. (2001) modified the single-path model of Bush et al. (1996b) in order to be able
to simulate absorbed fraction data for O3 (and C12, which is not considered) for three airflow
rates and for oral and nasal breathing. By adjusting several parameters, reasonable agreement
between predicted and experimental values was obtained. On the other hand, the O3 plots of the
experimental and predicted values of absorbed fraction versus penetration volume (e.g.,
Figures 4 and 5 of Bush et al., 2001) show sequential groups composed of only positive or only
negative residuals, indicating a lack of fit. Possibly adjusting other parameters would eliminate
this. To obtain an independent validation of the model, Bush et al. (2001) simulated
measurements of O3 concentrations made by Gerrity et al. (1995) during both inhalation and
exhalation at four locations between the mouth and the bronchus intermedius of human subjects.
Simulated and experimental values obtained are in close agreement. Note, however, that Bush
et al. (2001) made no quantitative assessment of how well their simulations agreed with the
experimental data; assessments were made on the basis of visual inspection of experimental and
simulated values plotted on the same figure. Thus, that evaluation of the model was subjective.
AX4-16
-------
Recently, Sarangapani et al. (2003) used physiologically based pharmacokinetic (PBPK)
modeling to characterize age- and gender-specific differences in both regional and systemic
uptake of O3 in humans. This model indicated that regional extraction of O3 is relatively
insensitive to age, but extraction per unit surface area is 2- to 8-fold higher in infants compared
to adults, due to the region-specific mass transfer coefficient not varying with age. The PU and
ET regions have a large increase in unit extraction with increasing age because both regions
increase in surface area. Males and females in this model have similar trends in regional
extraction and regional unit extraction. In early childhood, dose metrics were as much as
12 times higher than adult levels, but these differences leveled out with age, such that inhalation
exposures varied little after age 5. These data suggest that the early postnatal period is the time
of the largest difference in pharmacokinetics observed, and this difference is primarily due to the
immaturity of the metabolic enzymes used to clear O3 from the RT.
Mudway and Kelly (2004) attempted to model O3 dose-inflammatory response using a
meta-analysis of 23 exposures in published human chamber studies. The O3 concentrations
ranged from 0.08 to 0.6 ppm and the exposure durations ranged from 60 to 396 min. The
analysis showed linear relationships between O3 dose and neutrophilia in bronchoalveolar lavage
fluid (BALF). Linear relationships were also observed between O3 dose and protein leakage
intoBALF.
AX4.3 SPECIES HOMOLOGY, SENSITIVITY AND ANIMAL-TO-
HUMAN EXTRAPOLATION
Biochemical differences among species are becoming increasingly apparent, and these
differences may factor into a species' susceptibility to O3 exposure effects. Lee et al. (1998)
compared SD rats and rhesus monkeys to ascertain species differences in the various isoforms of
CYP moonoxygenases in response to O3 exposure (discussed in more detail in Section 5.3.1.2).
Differences in activities between rat and monkey were 2- to 10-fold, depending on the isoform
and the specific lung region assayed. This study supports the view that differential expression
of CYPs is a key factor in determining the toxicity of O3. As further characterization of
species- and region-specific CYP enzymes occurs, a greater understanding of the differences
AX4-17
-------
in response may allow more accurate extrapolation from animal exposures to human exposures
and toxic effects.
Arsalane et al. (1995) compared guinea pig and human AM recovered in BALF and
subsequently exposed in vitro to 0.1 to 1 ppm for 60 min. Measurement of inflammatory
cytokines showed a peak at 0.4 ppm in both species. Guinea pig AM had an increase in IL-6
and TNF-a while human AM had increases in TNF-a, IL-lb, IL-6 and IL-8. This exposure also
caused an increase in mRNA expression for TNF-a, IL-lb, IL-6 and IL-8 in human cells.
At 0.1 ppm exposures, only TNF-a secretion was increased. These data suggest similar cytokine
responses in guinea pigs and humans, both qualitatively and quantitatively.
Dormans et al. (1999) continuously exposed rats, mice, and male guinea pigs to filtered air,
0.2, or 0.4 ppm O3 for 3 to 56 days or for 28 days to follow recovery at 3, 7, and 28 days PE.
Depending on the endpoint studied, the species varied in sensitivity. Greater sensitivity was
shown in the mouse as determined by biochemical endpoints, persistence of bronchiolar
epithelial hypertrophy, and recovery time. Guinea pigs were more sensitive in terms of the
inflammatory response, though all three species had increases in the inflammatory response after
three days that did not decrease with exposure. In all species the longest exposure to the highest
dose caused increased collagen in ductal septa and large lamellar bodies in Type II cells, but that
response also occurred in rats and guinea pigs at 0.2 ppm. No fibrosis was seen at the shorter
exposure times and the authors question whether fibrosis occurs in healthy humans after
continuous exposure. The authors do not rule out the possibility that some of these differences
may be attributable to differences in total inhaled dose or dose actually reaching a target site.
Overall, the authors rated mice as most susceptible, followed by guinea pigs and rats.
Comparisons of airway effects in rats, monkeys and ferrets resulting from exposures of
1.0 ppm O3 for 8 h (Sterner-Kock et al. 2000) demonstrated that monkeys and ferrets had a
similar inflammatory responses and epithelial necrosis. The response of these two species was
more severe than that seen in rats. These data suggest that ferrets are a good animal model
for O3-induced airway effects due to the similarities in pulmonary structure between primates
and ferrets.
AX4-18
-------
The rat is a key species used in O3 toxicological studies, but Watkinson and Gordon,
(1993) suggest that, because the rat has both behavioral and physiological mechanisms that can
lower core temperature in response to acute exposures, extrapolation of these exposure data to
humans may be limited. Another laboratory (Iwasaki et al., 1998) has demonstrated both
cardiovascular and thermoregulatory responses to O3 at exposure to 0.1, 0.3, and 0.5 ppm O3
8 h/day for 4 consecutive days. A dose-dependent disruption of HR and Tco were seen on the
first and second days of exposure, which then recovered to control values. Watkinson et al.
(2003) exposed rats to 0.5 ppm O3 and observed this hypothermic response which included
lowered HR, lowered Tco, and increased inflammatory components in BALF. The authors
suggest that the response is an inherent reflexive pattern that can possibly attenuate O3 toxicity in
rodents. They discuss the cascade of effects created by decreases in Tco, which include:
(1) lowered metabolic rate, (2) altered enzyme kinetics, (3) altered membrane function,
(4) decreased oxygen consumption and demand, (5) reductions in minute ventilation, which
would act to limit the dose of O3 delivered to the lungs. These effects are concurrent with
changes in HR which lead to: (1) decreased CO, (2) lowered BP, (3) decreased tissue perfusion,
all of which may lead to functional deficits. The hypothermic response has not been observed in
humans, except at very high O3 exposures.
AX4-19
-------
REFERENCES
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, P. T.; Ben-Jebria, A.; Rigas, M. L.; Ultman, J. S. (1996) Longitudinal distribution of ozone absorption in
the lung: effect of continuous inhalation exposure. Arch. Environ. Health 51: 431-438.
Bush, M. L.; Asplund, P. T.; Miles, K. A.; Ben-Jebria, A.; Ultman, J. S. (1996a) Longitudinal distribution of O3
absorption in the lung: gender differences and intersubject variability. J. Appl. Physiol. 81: 1651-1657.
Bush, M. L.; Raybold, T.; Abeles, S.; Hu, S.-C.; Ben-Jebria, A.; Ultman, J. S. (1996b) Longitudinal distribution of
ozone absorption in the lung: simulation with a single-path model. Toxicol. Appl. Pharmacol. 140: 219-226.
Bush, M. L.; Zhang, W.; Ben-Jebria, A.; Ultman, J. S. (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.
Cohen-Hubal, E. A.; Kimbell, J. S.; Fedkiw, P. S. (1996) Incorporation of nasal-lining mass-transfer resistance into a
CFD model for prediction of ozone dosimetry in the upper respiratory tract. Inhalation Toxicol. 8: 831-857.
Dahl, A. R. (1990) Dose concepts for inhaled vapors and gases. Toxicol. Appl. Pharmacol. 103: 185-197.
Dormans, J. A. M. A.; VanBree, L.; Boere, A. J. F.; Marra, M.; Rombout, P. J. A. (1999) Interspecies differences in
time course of pulmonary toxicity following repeated exposure to ozone. Inhalation Toxicol. 11: 309-329.
Duan, X.; Buckpitt, A. R.; Plopper, C. G. (1993) Variation in antioxidant enzyme activities in anatomic
subcompartments within rat and rhesus monkey lung. Toxicol. Appl. Pharmacol. 123: 73-82.
Duan, X.; Buckpitt, A. R.; Pinkerton, K. E.; Ji, C.; Plopper, C. G. (1996) Ozone-induced alterations in glutathione in
lung subcompartments of rats and monkeys. Am. J. Respir. Cell Mol. Biol. 14: 70-75.
Gerrity, T. R.; Biscardi, F.; Strong, A.; Garlington, A. R.; Brown, J. S.; Bromberg, P. A. (1995) Bronchoscopic
determination of ozone uptake in humans. J. Appl. Physiol. 79: 852-860.
Guilmette, R. A.; Wicks, J. D.; Wolff, R. K. (1989) Morphometry of human nasal airways in vivo using magnetic
resonance imaging. J. Aerosol. Med. 2: 365-377.
Hu, S. C.; Ben-Jebria, A.; Ultman, J. S. (1992) Longitudinal distribution of ozone absorption in the lung: quiet
respiration in healthy subjects. J. Appl. Physiol. 73: 1655-1667.
Hu, S.-C.; Ben-Jebria, A.; Ultman, J. S. (1994) Longitudinal distribution of ozone absorption in the lung: effects of
respiratory flow. J. Appl. Physiol. 77: 574-583.
Iwasaki, T.; Takahashi, M.; Saito, H.; Arito, H. (1998) Adaptation of extrapulmonary responses to ozone exposure in
conscious rats. Ind. Health 36: 57-60.
Joad, J. P.; Brie, J. M.; Weir, A. J.; Putney, L.; Hyde, D. M.; Postlewait, E. M.; Plopper, C. G. (2000) Effect of
respiratory pattern on ozone injury to the airways of isolated rat lungs. Toxicol. Appl. Pharmacol. 169: 26-32.
Kimbell, J. S.; Gross, E. A.; Joyner, D. R.; Godo, M. N.; Morgan, K. T. (1993) Application of computational fluid
dynamics to regional dosimetry of inhaled chemicals in the upper respiratory tract of the rat. Toxicol. Appl.
Pharmacol. 121: 253-263.
Lee, C.; Watt, K. C.; Chang, A. M.; Plopper, C. G.; Buckpitt, A. R.; Pinkerton, K. E. (1998) Site-selective
differences in cytochrome P450 isoform activities: comparison of expression in rat and rhesus monkey lung
and induction in rats. Drug Metab. Dispos. 26: 396-400.
Mercer, R. R.; Anjilvel, S.; Miller, F. J.; Crapo, J. D. (1991) Inhomogeneity of ventilatory unit volume and its effects
on reactive gas uptake. J. Appl. Physiol. 70: 2193-2205.
Miller, F. J. (1995) Uptake and fate of ozone in the respiratory tract. Toxicol. Lett. 82/83: 277-285.
Miller, F. J.; Overton, J. H., Jr.; Jaskot, R. H.; Menzel, D. B. (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.
Mudway, I. S.; Kelly, F. J. (2004) An investigation of inhaled ozone dose and the magnitude of airway inflammation
in healthy adults. Am. J. Respir. Crit. Care Med. 169: 1089-1095.
Nodelman, V.; Ultman, J. S. (1999a) Longitudinal distribution of chlorine absorption in human airways: comparison
of nasal and oral quiet breathing. J. Appl. Physiol. 86: 1984-1993.
Nodelman, V.; Ultman, J. S. (1999b) Longitudinal distribution of chlorine absorption in human airways: a
comparison to ozone absorption. J. Appl. Physiol. 87: 2073-2080.
Nunn, J. F.; Campbell, E. J. M.; Peckett, B. W. (1959) Anatomical subdivisions of the volume of respiratory dead
space and the effect of position of the jaw. J. Appl. Physiol. 14: 174-176.
AX4-20
-------
Olson, D. E.; Sudlow, M. F.; Horsfield, K.; Filley, G. F. (1973) Convective patterns of flow during inspiration. Arch.
Intern. Med. 131:51-57.
Overton, J. H.; Graham, R. C. (1995) Simulation of the uptake of a reactive gas in a rat respiratory tract model with
an asymmetric tracheobronchial region patterned on complete conducting airway cast data. Comput. Biomed.
Res. 28: 171-190.
Overton, J. H.; Graham, R. C.; Menache, M. G.; Mercer, R. R.; Miller, F. J. (1996) Influence of tracheobronchial
region expansion and volume on reactive gas uptake and interspecies dose extrapolations. Inhalation Toxicol.
8: 723-745.
Patra, A. L.; Menache, M. G.; Shaka, N. B.; Gooya, A. (1987) A morphometric study of nasal-pharyngeal growthfor
particle deposition in the rat. Am. Ind. Hyg. Assoc. J. 48: 556-562.
Pinkerton, K. E.; Menache, M. G.; Plopper, C. G. (1995) Consequences of prolonged inhalation of ozone onF344/N
rats: collaborative studies. Part IX. Changes in the tracheobronchial epithelium, pulmonary acinus, and lung
antioxidant enzyme activity. Cambridge, MA: Health Effects Institute; pp. 41-98; research report no. 65.
Available from: NTIS, Springfield, VA; PB95-261996.
Pinkerton, K. E.; Weller, B. L.; Menache, M. G.; Plopper, C. G. (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. Cambridge, MA: Health Effects Institute;
research report no. 65.
Postlethwait, E. M.; Joad, J. P.; Hyde, D. M.; Schelegle, E. S.; Brie, J. M.; Weir, A. J.; Putney, L. F.; Wong, V. J.;
Velsor, L. W.; Plopper, C. G. (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.
Raabe, O. G.; Yeh, H. C.; Schum, G. M.; Phalen, R. F. (1976) Tracheobronchial geometry: human, dog, rat,
hamster. Albuquerque, NM: Lovelace Foundation; report no. LF-53.
Rigas, M. L.; Ben-Jebria, A.; Ultman, J. S. (1997) Longitudinal distribution of ozone absorption in the lung: effects
of nitrogen dioxide, sulfur dioxide, and ozone exposures. Arch. Environ. Health 52: 173-178.
Rigas, M. L.; Catlin, S. N.; Ben-Jebria, A.; Ultman, J. S. (2000) Ozone uptake in the intact human respiratory tract:
relationship between inhaled dose and actual dose. J. Appl. Physiol. 88: 2015-2022.
Santiago, L. Y.; Harm, M. C.; Ben-Jebria, A.; Ultman, J. S. (2001) Ozone adsorption in the human nose during
unidirectional airflow. J. Appl. Physiol. 91: 725-732.
Sarangapani, R.; Gentry, P. R.; Covington, T. R.; Teeguarden, J. G.; Clewell, H. J., III. (2003) Evaluation of the
potential impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of vapors.
Inhalation Toxicol. 15: 987-1016.
Schelegle, E. S.; Alfaro, M. F.; Putney, L.; Stovall, M.; Tyler, N.; Hyde, D. M. (2001) Effect of C-fiber-mediated,
ozone-induced rapid shallow breathing on airway epithelial injury in rats. J. Appl. Physiol. 91: 1611-1618.
Sterner-Kock, A.; Kock, M.; Braun, R.; Hyde, D. M. (2000) Ozone-induced epithelial injury in the ferret is similar
to nonhuman primates. Am. J. Respir. Crit. Care Med. 162: 1152-1156.
Ultman, J. S.; Ben-Jebria, A.; Arnold, S. F. (2004) Uptake distribution of ozone in human lungs: intersubject
variability in physiologic response. Boston, MA: Health Effects Institute; research report no. 125. Available:
http://www.healtheffects.org/Pubs/Ultman.pdf [29 July, 2005].
U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants.
Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF. 3v.
Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, andPB96-185608. Available online at:
www.epa.gov/ncea/ozone.htm.
Watkinson, W. P.; Gordon, C. J. (1993) Caveats regarding the use of the laboratory rat as a model for acute
lexicological studies: modulation of the toxic response via physiological and behavioral mechanisms.
Toxicology 81: 15-31.
Watkinson, W. P.; Campen, M. J.; Wichers, L. B.; Nolan, J. P.; Costa, D. L. (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, E. R. (1963) Morphometry of the human lung. New York, NY: Academic Press Inc.
Yeh, H. C.; Schum, G. M.; Duggan, M. T. (1979) Anatomic models of the tracheobronchial and pulmonary regions
of the rat. Anat. Rec. 195: 483-492.
AX4-21
-------
AX5. ANNEX TO CHAPTER 5 OF OZONE AQCD
AX5-1
-------
Table AX5-1. Cellular Targets of Ozone Interaction
Concentration
ppm Duration
0.22 4 h with
exercise
0.5-10 both 1 h
with and
without 5% CO2
0.25-1.0 30-60 min
Species
Rat, male, SD, 90 days
old, 300-330 g,
n = 6/group
Rat, male, SD, 90 days
old, 300-330 g,
n = 6-14/group
Rat, male, SD,
250-275 g,
Effects
Demonstrated the ozonation of PUFA to form nonanal and hexanal in rat B ALF.
Increases in nonanal not accompanied by significant changes in lung function,
epithelial permeability, or airway inflammation. Hexanal levels did not increase
significantly. Levels of both aldehydes returned to baseline by 18 h PE.
O3 plus CO2 increased the VT and the yield of aldehydes with a maximal
aldehyde yield at 2.5 ppm for 1 h. 0.5 ppm O3 with 5% CO2, levels of hexanal
and nonanal increased at 30 min, decreased slightly from that level at 60 min,
was maximal at 90 min and then dropped to 60 min levels at 120 min. Levels of
heptanal did not change appreciably during this time course. Suggests that
levels of these aldehydes were dependent on a dynamic relationship between
their production and the disappearance from the ELF.
PUFAs directly react with O3. The amount of bioactive lipids produced is
inversely related to AA availability.
Reference
Frampton
et al. (1999)
Pryor et al.
(1996)
Postlethwait
et al. (1998)
X
-------
Table AX5-2. Effects of Ozone on Lung Monooxygenases
X
Concentration
ppm Duration Species
NA
Effects
Reference
1.0
1.0
0.8
1
8h
90 days
8 h/day for
90 days
2h
Rat, male, SD,
350-600 g,
n= 3-6/group
Rat, male, SD,
275-300 g
Mice, male, 2-3 months
old, Clara Cell Secretory
Protein deficient,
WT strain 129
n = 3/group
Increases CYP2E1 activity in lobar-bronchi and major daughter airway
with 8 h exposure. Decreased CYP2E1 activities in both major and minor
daughter airways with 90-day exposure. O3 does not result in consistent
dramatic alterations in CYP2E1 activities.
CYP2B activity increased. Linked to Clara cells in distal lung only — not in
trachea or proximal airway.
CCSP^mice had increases in IL-6 and MT mRNA that preceded decreases
in Clara cell CYP2F2 mRNA. WT mice had levels change, but to a lesser
degree.
Watt et al.
(1998)
Paige et al.
(2000a)
Mango et al.
(1998)
Rat, male, SD, adult
monkey, Rhesus, 0.75 to
9.7 years old
Microdissection for regiospecific and species-specific differences in
isoforms of CYPs. Rat parenchyma: both CYP1A1 and CYP2B
were highest. Rat airways: CYP2E1 was highest. Rat airways and
parenchyma: P450 reductase activities were high, and conversely, low in
trachea. Monkeys: did not exhibit site-selective differences in CYP2B1,
CYP1A1, or P450 reductase; however, they had high CYP2E1 activity
in parenchyma and distal bronchioles.
Lee et al.
(1998)
CYP = Cytochrome P450
WT = wild-type
MT = metallothionein
CCSP = Clara Cell Secretory Protein
-------
X
-k
Table AX5-3. Antioxidants, Antioxidant Metabolism, and Mitochondrial Oxygen Consumption
Concentration
ppm
1.0
Duration Species
6 h/day, Rat, male Fischer F344,
5 days/week 30-32 days old, n = 4
for 2 or
3 months
Effects
Immunohistochemistry and immunogold labeling studies. In epithelial
cells in airways and parenchyma: reduced Cu-Zn SOD labeling with
O3 -exposure. In the CAR regions, in both AMs and Type II epithelial
cells: significantly increased levels of Mn SOD. Mn SOD levels were not
increased in Type I epithelial cells, fibroblasts, or Clara cells. Suggests
that the increased levels of Mn SOD in Type II cells in the proximal
alveolar duct confer tolerance and protection from further O3-induced
injury.
Reference
Weller et al.
(1997)
0.2
6h
Dog, male, mongrel,
~/9 kg, n = 6/group
0,0.1,0.25,
0.5, 1.0 or 1.5
30-720 min
Model of continually
mixed, interfacial exposure
Blocking antioxidant transport with probenecid caused heterogeneously Freed et al.
distributed increases in peripheral airway resistance and reactivity. (1999)
Probenecid inhibited O3-induced neutrophilic inflammation, which is
evidence of a dissociation between airway function and inflammation.
Probenecid caused a 50-60% decrease in plasma urate, a decrease in
ascorbate, and a decrease in B ALF protein. Suggests probenecid has
either a direct or indirect effect on either cytokine or leukotriene transport.
Modeled the interactions of O3 with three ELF antioxidants, AA, UA and Mudway and
GSH. Ranking of reactivity with O3 was UA>AA>GSH. Antioxidants Kelly (1998)
caused no changes in sample pH and protein carbonyl formations.
Consumption of the antioxidants occurred in a linear fashion with time
and a positive relationship to O3 concentration. Suggests that GSH is
not an important substrate for O3; UA appeared to be the most important
substrate.
Cu = Copper
Zn = Zinc
Mn = Manganese
SOD = Superoxide Dismutase
AM = Alveolar Macrophage
BALF = Bronchoalveolar Lavage Fluid
AA = Ascorbic Acid
UA = Uric Acid
GSH = Reduced Glutathione
-------
Table AX5-4. Lipid Metabolism and Content of the Lung
Concentration
ppm
Duration
Species/Cell Line
Effects
Reference
NA
NA
60 min
Cultured human epithelial
cells (BEAS-2B)
Cultured human bronchial
epithelial cells (NHBE)
and BEAS-2B cells
>
X
0.06, 0.125, and
0.25 ppm
2 to 48 h
Lung (calf surfactant)
30 min
Human red blood cell
(RBC) model
Incubation with 10 uM lipid oxidation products caused significant Pryor et al.
release of arachidonic acid. Suggests that lipid oxidation products cause (1995)
activation of specific lipases, which then trigger the activation of second
messenger pathways (e.g., phospholipase A2 or phospholipase C).
Incubation with lipid ozonation product Kafoury et al.
1 -palmitoyl-2-(9-oxononanoyl)-sn-glycero-3 -phosphocholine (PC-ALF) (1999)
and 1-hydroxy-l-hydroperoxynonane (HHP-C9). PC-ALF elicited
release of (PAF) and prostaglandin E2, but not IL-6. HHP-C9 caused
release of PAF and IL-6 in these cells, but not prostaglandin E2.
Suggests that O3-induced production of lipid ozonation products
causes release of proinflarnmatory mediators that then generate an
early inflammatory response.
Dose- and time-dependent increase in the formation of Uhlson et al.
1 -palmitoyl-2-(9'-oxo-nonanoyl)-glycerophosphocholine (2002)
(16:Oa/9-al-GPCho), an oxidized phospholipid, which possessed
biological activity in three assays. The 16:Oa/9-al-GPCho:
(1) decreased macrophage viability by necrosis at 6 uM, (2) induced
apoptosis in pulmonary epithelial-like A549 cells at 100-200 uM,
and (3) elicited release of IL-8 from A549 cells at 50-100 uM.
Human RBCs intermittently covered by an aqueous film consisting of Ballinger et al.
rat BALF or BALF plus added reagents that included AA, UA, GSH, (2005)
Trolox (a vitamin E analog), SOD, catalase, desferrioxamine,
deithylenetriaminepentaacectic acid, mannitol or BSA. Oxidation of
lipids in BALF or on membranes was assayed by measuring TEARS
accumulation and loss of acetylcholinesterase activity. AA and GSH
induced dose-dependent oxidative damage to the cell membrane proteins
and lipids via secondary oxidant formation. Conclusion: early in O3
exposure, ELF antioxidants are high enough to drive reactive absorption
of O3 into the ELF and to concurrently quench secondary reaction
products, thus limiting cell injury. With continued exposure, O3 flux
would decrease, cellular injury would increase due to depleted
antioxidants levels. They hypothesize that especially in areas where
O3 deposition is high, unreacted O3 and cytotoxic products can diffuse
to the cell membranes, causing injury.
-------
Table AX5-4 (cont'd). Lipid Metabolism and Content of the Lung
Concentration
ppm Duration
Species/Cell Line
Effects
Reference
X
2.0
4h
0.2, 0.5, or 1.0 1 h
Rat BALF exposed to O3
in vitro.
Cultured human bronchial
epithelial cells (16-HBE)
0.5, 1.0,2.0,
or 3.0
Mice, C57BL/6J,
8-12 weeks old female.
n=4
The major reaction products identified were Pulfer and
5-hydroperoxy-B-homo-6-oxa-cholestan-3,7a-diol, Murphy
5P,6p-epoxycholesterol, and 3p-hydroxy-5oxo-5,6-seco-cholestan-6-al. (2004)
Extracted lipid reaction products either immediately or 24 h PE.
Higher levels of 5p,6p-epoxycholesterol were recovered in the extract
immediately PE. Both 5 P,6 P-epoxy cholesterol and its most abundant
metabolite, cholestan-6-oxo-3p,5oc-diol, were cytotoxic to 16-HBE cells
at physiologically relevant concentrations. Both reaction products were
also shown to be potent inhibitors of cholesterol synthesis.
In BALF, lavaged cells, and whole lung homogenate there was a Pulfer et al.
dose-dependent increase in formation cholestan-6-oxo-3 P,5oc-diol at (2005)
exposure levels of 0.5, 1.0 and 2.0 ppm O3, respectively. In BALF
and lavaged cells O3 induced increases in 5p,6p-epoxycholesterol at
exposure levels of 0.5 and 1.0, respectively.
PAF = Platelet-Activating Factor
RBC= Red Blood Cell
AA = Ascorbic Acid
GSH = Reduced Glutathione
IL-6 = Interleukin-6
IL-8 = Interleukin-8
ELF = Epithelial Lining Fluid
BALF = Bronchoalveolar Lavage Fluid
BSA = Bovine Serum Albumin
TEARS = thiobarbituric acid
UA = Uric Acid
-------
Table AX5-5. Effects of Ozone on Protein Synthesis
Concentration
ppm
Duration
Species
Effects
Reference
0.4
1,3,7,28, or
56 days
Rat, male, Wistar,
200 g,
n = 5/group
Centriacinar thickening of septa after 7 days of exposure, which
progressed at 28 and 56 days of exposure. After 28 days of O3, the
increase in collagen content in ductular septa was apparent and it
increased progressively until the 56-day time point. Collagen content
decreased with PE recovery, the structural fibrotic changes in ductular
septa did not return to control levels. Respiratory bronchioles were
present at an increasing degree, which persisted after an 80-d recovery
period. Suggests that subchronic O3 exposures in rats creates a
progression of structural lung injury that can evolve into a more chronic
form, which included fibrosis.
van Bree et al.
(2001)
X
Table AX5-6. Effects of Ozone on Differential Gene Expression
Concentration
ppm
Duration
Species
Effects
Reference
1
8 h for three
consecutive
nights
Mice, C57BL/6,
20-25 g,
n = 5-6/group
Ozone exposure induced changes in expression of 260 genes ( 80%
repressed and 20% induced). Cell cycle genes upregulated: S-adenosyl
methionine decarboxylase 3, ribonucleotide reductase, and clusterin.
NF-KB-induced genes upregulated: serum amyloid protein, topoisomerase
IIcc, monocyte chemoattractant protein, platelet-derived growth factor,
and inhibitor of apoptosis. Downregulation of transcripts for isoforms
of myosins and actins. CYP family genes downregulated: 2a4, and2el,
and 2f2, as were aryl-hydrocarbon receptor and several glutathione
transferases. Metallothionein 1 and 2 and lactotransferrin upregulated.
Major histocompatibility complex genes and lymphocyte specific
proteins downregulated.
Gohil et al.
(2003)
-------
Table AX5-7. Effects of Ozone on Lung Host Defenses
Concentration
ppm
Duration
Species
Effects
Reference
Microbiologic Endpoints
0.1,0.3
4 h/day,
5 days/week,
1 or 3 weeks
Rat, male, Fischer F344,
200-250 g
3h
Rat, male, Fischer F344,
8 weeks old,
n = 4-10/group
No effect on cumulative mortality from subsequent lung infection Cohen et al.
with 4-8 x 106 Listeria monocytogenes, but concentration-related (2001, 2002)
effects on morbidity onset and persistence. One-week exposed rats:
listeric burdens trended higher than in controls; 0.3 ppm rats displayed
continual burden increases and no onset of resolution; in situ IL-loc,
TNF-oc, and IFN-y levels 48 and 96 h post-infection (4 x 106 level)
higher than controls. Three-week exposed rats: no O3-related change
in bacterial clearance; IL-loc, TNF-oc, and IFN-y levels higher than
control only at 48 h post-infection (4 x 106) and only with 0.3 ppm rats.
Single exposure to S. zooepidemicus led to differential clearance Dong et al.
patterns in exposed rats maintained on ad libitum or O3-mitigating (1998)
calorie-restricted diets.
jx^ Clearance Endpoints (Non-Microbial)
oo 0.01-1.0 lOmin Rat, SD
0.4
6h
Dog, male, mixed breed,
1-3 years old,
n = 7/group
Single 10 min exposure of trachea! explants, followed by 1 h incubation Churg et al.
with particles, led to dose-related increases in uptake of amosite (1996)
asbestos and titanium dioxide particles. Effect inhibited by added
catalase or desferoxamine, but not by superoxide dismutase.
Increased trachea! permeability to 99mTc-DTPA after direct sublobar Foster and
exposure to O3. Clearance halftimes remained significantly lower for Freed (1999)
1-7 days PE, but recovered by 14 days PE.
Alveolar Macrophage Endpoints (General)
0.8
0.8
1.0
3h
3h
4h
Rat, male, SD,
250-275 g,
n = 5-6/group
Rat, male, SD,
250-300 g,
n = 5/group
Mice cell line
(WEH1-3)
Increased ex vivo AM adherence to epithelial cultures mitigated by
cell pretreatment with anti-CD 1 Ib or anti-ICAM-1 antibodies.
Increased ex vivo AM adherence to epithelial cell cultures mitigated
by cell pretreatment with anti-TNF-oc/IL-la antibodies.
Increased intracellular calcium resting levels in WEHI-3 cells.
Decreased rates of calcium influx due to digitonin.
Bhalla (1996)
Pearson and
Bhalla (1997)
Cohen et al.
(1996)
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
Concentration
ppm
Duration
Species
Effects
Reference
Alveolar Macrophage Endpoints (Functional)
0.1,0.3
0.1,0.3
0.3
X
4 h/day,
5 days/week,
1 or 3 weeks
4 h/day,
5 days/week,
1 or 3 weeks
5 h/day,
5 days/week,
4 weeks
3h
Rat, male, Fischer F344,
200-250 g
Rat, male, Fischer F344,
200-250 g
Rat, male, Fischer F344,
200-250 g,
n = 10/group
Rat, male, SD, 250-
275 g, n = 5-6/group
Superoxide anion: increased AM production (1 week; 0.1,0.3 ppm); Cohen et al.
no intergroup differences noted after IFN-y stimulation. (2001)
H2O2: reduced production (1 week; 0.1, 0.3 ppm); further reduced
production after treatment with IFN-y (0.1, 0.3 ppm, 1 and 3 weeks).
Increased AM superoxide anion production (1 week; 0.1, 0.3 ppm), Cohen et al.
Lower H2O2 production (1 week; 0.1,0.3 ppm). Reduced production (2002)
after treatment with IFN-y - superoxide (0.3 ppm, 1 week) and H2O2
(0.1 ppm, 1 week) - relative to cells without IFN-y treatment.
No effects from 3-week exposures.
No effect on AM endotoxin-stimulated IL-lcc, IL-6, or TNF-cc Cohen et al.
production. Decrease in stimulated, but not spontaneous, superoxide (1998)
formation; variable effects on H2O2 formation. No effect on AM
spontaneous, endotoxin-, or IFN-y-stimulated, NO formation.
Increased AM motility in response to chemotaxin; effect mitigated Bhalla (1996)
by cell pretreatment with anti-CD 1 Ib or anti-ICAM-1 antibodies.
1.0
3h
24 h/day,
3 days
Rat, male, Fischer F344,
8 weeks old,
n = 4-10/group
Mice, female,
(B6J129SV)
(C57BL/6X 129 NOS"'"),
8-16 weeks old,
n= 3-12/group
Rat, male, Wistar,
8-12 weeks,
Decrease in AM phagocytic activity.
Increased AM spontaneous and IFN-y + LPS-induced NOS expression
and NO production and PGE2 release. Initial decrease in ROI
production, with eventual rebound. Knockout (NOS"'") mice AM
incapable of similar response to O3 - no inducible NO or PGE2
above control levels and consistent decreased ROI production.
BALF from exposed rats subsequently inhibited IFN-y-induced
AM NO production.
Dong et al.
(1998)
Fakhrzadeh
et al. (2002)
Koike et al.
(1998,1999)
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
X
Concentration
ppm
Duration
Cytokines, Chemokines: Production,
0.1,0.3
0.1,0.3
0.3
0.3
1.0
2.5
4 h/day,
5 days/week,
1 or 3 weeks
4 h/day,
5 days/week,
1 or 3 weeks
5 h/day,
5 days/week,
4 weeks
24 or 96 h
1, 2, or4h,
or24h
Species
Effects
Reference
Binding, and Inducible Endpoints
Rat, male, Fischer F344,
200-250 g,
Rat, male, Fischer F344,
200-250 g,
n= 3-5/group
Rat, male, Fischer F344,
200-250 g,
n = 10/group
Mice, male, C57BL/6J,
adult, n = 3/group
Superoxide anion: no intergroup differences noted after IFN-y
stimulation. H2O2: reduced production after treatment with IFN-y.
Decreased expression of CDS among lung lymphocytes (0. 1 ppm only;
3 weeks); effect exacerbated by stimulation with IFN-cc (but not with
IL-lcc). Decreased expression of CD25 (IL-2R) on CD3+ lymphocytes
(0.3 ppm only; 3 weeks); effect worsened by treatment with IL-lcc
(0.1, 0.3 ppm; 3 weeks). No effects on IL-2-inducible lympho-
proliferation. Reduced AM production of ROIs after treatment with
IFN-y; superoxide anion (0.3 ppm, 1 week) and H2O2 (0. 1 ppm, 1 week)
- relative to untreated cells.
No effect on AM endotoxin-stimulated IL-lcc, IL-6, or TNF-cc
production.
0.3 ppm: Increased lung: MIP-2, MCP-1, and eotaxin mRNA
expression.
1 f\ rvr\nT A -ftf^r A V\ in/^rv^ooi^/l Inner* A^TT5_0 A^f^T5_1 i^/~\tovin on/1
Cohen et al.
(2001)
Cohen et al.
(2002)
Cohen et al.
(1998)
Johnston et al.
(1999a)
IL-6 mRNA expression.
2.5 ppm: After 2 h, increased lung: MIP-2, MCP-1, eotaxin, and
IL-6 mRNA expression.
No exposure-related increases in lung IL-lcc, IL-lp, IL-lRcc, IL-10,
IL-12, or IFN-y mRNA expression.
1.0 4h Mice cell line
(WEH1-3)
1.0 6h Rat, male, SD,
200-250 g,
n= 3-6/group
Decreased binding of IFN-y by WEHI-3 cells. Decreased superoxide
anion production by IFN-y-treated cells; no similar effect on H2O2
production. Decreased IFN-y -stimulated phagocytic activity. No effect
on IFN-y-inducible la (MHC Class II) antigen expression.
Increased AM MIP-lcc, CINC, TNF-cc, and IL-lp mRNA expression.
Induced increase in MIP-lcc and CINC mRNA temporally inhibited
by cell treatment with anti-TNF-cc/IL-lp antibodies.
Cohen et al.
(1996)
Ishii et al.
(1997)
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
Concentration
ppm Duration Species
Effects
Reference
Cytokines, Chemokines: Production, Binding, and Inducible Endpoints (cont'd)
>
X
(J\
1
^
1.0 24h/day, Rat, male, Wistar,
3 days 8-12 weeks old
1.0 24 h Mice, male, C57BL/6J,
8 weeks old, n = 3/group
1.0 24 h Mice, male, C57BL/6J,
8 weeks old, n = 3/group
1.0 8 h/day, Mice (C57BL/6)
3 days (C57BL/6Ai" NOS"'")
n= 3-10/group
1.0, 2.5 4 or 24 h Mice, male, C57BL/6J,
adult, n = 3/group
0.6, 2.0 3 h Mice, C57BL/6, Rat,
Wistar, 14-16 weeks
old, n = 4-6/group
0.8 3 h Rat, male, SD,
250-300 g, n = 5/group
0.8 3 h Mice, female,
B6J129SV,
C57BL/6X 129 NOS"',
8-16 weeks,
n= 3-12/group
BALF from exposed rats subsequently inhibited: ConA-stimulated
lymphocyte IFN-y production, but had no effect on IL-2 production;
IL-2 -induced lymphoproliferation; and, IFN-y-induced AM NO
production.
Increased lung: MIP-2 (4 h PE) and MCP-1 (4 and 24 h PE) mRNA
expression.
Increased lung MIP-2 and MCP-1 mRNA expression (4 and 24 h PE);
no effects on mRNA levels of IL-lcc, IL-lp, IL-lRcc, IL-6, MIF,
MTP-la, MIP-lp, eotaxin, or RANTES at either time point in recovery
period. Enhanced expressions of some cytokines/chemokines were
maintained longer than normal by coexposure to endotoxin.
Knockout (NOS"'") mice have more lavageable MIP-2 after exposure
than wild-type; both greater than control.
Dose-related increases in cytokine/chemokine induction. Increased
lung MIP-lcc, MIP-2, eotaxin (4 and 24 h), IL-6 (4 h only), and iNOS
mRNA expression.
Increased lung MIP-2 (4 h PE) and MCP-1 mRNA expression
(24 h PE); PMN and monocyte increased accumulation in lungs
consistent with sequential expression of the chemokines.
NF-kB activation also increased 20-24 h PE.
Increased ex vivo AM adherence to epithelial cells mitigated by cell
treatment with anti-TNF-a or IL-loc antibodies.
Increased AM IFN-y + LPS-induced NOS expression and NO
production, as well as induced PGE2 release. Knockout (NOS"'")
mice AM incapable of similar response to O3 - no inducible NO or
PGE2 above control levels.
Koike et al.
(1998, 1999)
Johnston et al.
(2001)
Johnston et al.
(2002)
Kenyon et al.
(2002)
Johnston et al.
(2000a)
Zhao et al.
(1998)
Pearson and
Bhalla (1997)
Fakhrzadeh
et al. (2002)
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
X
to
Concentration
ppm
Duration
Cytokines, Chemokines: Production,
0.8
2.0
0.08-0.25
OVA
Alveolar
0.3
0.8
0.8
3h
3h
and 1% 4 h,
3 times/week,
4 weeks
Macrophage/Lung NO- and
5 h/day,
5days/week,
4 weeks
3h
3h
Species
Effects
Reference
Binding, and Inducible Endpoints (cont'd)
Mice, female,
B6J129SV,
C57BL/6X 129 NOS"'",
8-16 weeks,
n= 3-4/group
Rat, female, SD,
200-225 g,
n = 4-6/group
Mice, female, BALB/c,
C57BL/6,
6-8 weeks old,
n = 4-12/group
iNOS-Related Endpoints
Rat, male, Fischer F344,
200-250 g,
n = 10/group
Mice, female,
B6J129SV,
C57BL/6X 129 NOS"'",
8-16 weeks,
n= 3-12/group
Mice, female,
B6J129SV,
Increased AM IFN-y + LPS-induced NOS expression and NO
production.
Increased AM spontaneous and IFN-y + LPS-induced NOS expression
and NO production. AM from exposed rats showed rapid
onset/prolonged activation of NF-KB.
O3 - dose-dependent increases in IgE, IL-4, IL-5; recruitment of
eosinophils and lymphocytes in BALB/c; O3 + OVA - increased IgG,
antibody liters, leukotrienes, airway responsiveness, immediate
cutaneous hypersensitivity reactions in BALB/c. In C57BL/6 only
O3 + OVA caused cutaneous hypersensitivity and altered IgG responses.
No effect on AM spontaneous, endotoxin-, or IFN-y -stimulated,
NO formation.
Increased AM IFN-y + LPS-induced NOS expression and NO
production and PGE2 release. Knockout (NOS"'") mice AM incapable of
similar response to O3 - no inducible NO or PGE2 above control levels.
Increased AM spontaneous and IFN-y + LPS-induced NOS expression
and NO production. AM from exposed mice showed rapid and
Laskin et al.
(2002)
Laskin et al.
(1998a)
Neuhaus-
Steinmetz
et al. (2000)
Cohen et al.
(1998)
Fakhrzadeh
et al. (2002)
Laskin et al.
(2002)
C57BL/6X 129 NOS"'",
8-16 weeks,
n= 3-4/group
prolonged activation of NF-KB, STAT-1 (expression, activity),
phosphoinositide 3-kinase, and protein kinase B.
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
X
OJ
Concentration
ppm
Alveolar
1.0
1.0
1.0,2.5
2.0
2.0
3.0
0.12,0.5,
Duration
Macrophage/Lung NO- and
8 h/day,
3 days
24 h/day,
3 days
4 or 24 h
3h
3h
6h
or2 3h
Species
iNOS-Related Endpoints
Mice, C57BL/6,
C57BL/6Ai' NOS ' ,
n= 3-10/group
Rat, male, Wistar,
8-12 weeks, n = 2/group
Mice, male, C57BL/6J,
adult, n = 3/group
Rat, female, SD,
200-250 g,
n= 1-3 /group
Rat, female, SD,
200-225 g,
n= 3-6/group
Rat, female, Brown
Norway, 250-300 g,
n = 4-8/group
Mice, female, BALB/c,
5-6 weeks
Effects
(cont'd)
Knockout (NOS"'-) mice have more lavageable PMN, MTP-2,
and protein in lungs after exposure than wild-type.
BALF from exposed rats subsequently inhibited IFN-y-induced
AM NO production.
Dose-related increase in lung iNOS mRNA expression.
Increased AM spontaneous, IFN-y, and LPS-induced NO production,
as well as spontaneous and LPS-induced NOS expression. Effect
somewhat ameliorated by pretreatment with bacterial endotoxin.
Increased AM spontaneous and IFN-y + LPS-induced NOS
expression and NO production. AM from exposed rats showed
rapid onset/prolonged activation of NF-icB.
Increased lung iNOS mRNA expression. Effect blocked by
pretreatment with dexamethasone.
Dose-dependent increases in nitrate and Penh; increases in nNOS, but
not iNOS or eNOS.
Reference
Kenyon et al.
(2002)
Koike et al.
(1998, 1999)
Johnston et al.
(2000a)
Pendino et al.
(1996)
Laskin et al.
(1998b)
Haddad et al.
(1995)
Jang et al.
(2002)
Surface Marker-Related Endpoints
0.8
1.0
1.0
3h
4h
2h
Rat, male, SD,
250-275 g,
n = 5-6/group
Mice cell line
(WEH1-3)
Rat, female, SD,
170-210 g,
n = 8-12/group
Increased expression of AM CD 1 Ib, but no effect on ICAM- 1 .
No effect on IFN-y-inducible la (MHC Class II) antigen expression
on WEHI-3 cells.
Decreased expression of integrins CD 1 8 on AM and CD 1 Ib on PMN.
No effect on PMN CD62L selection.
Bhalla (1996)
Cohen et al.
(1996)
Hoffer et al.
(1999)
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
Concentration
ppm
Duration
Species
Effects
Reference
Surface Marker-Related Endpoints (cont'd)
1.0
0.1,0.3
3 days
4 h/day,
5 days/week,
1 or 3 weeks
Rat, male, Wistar and
Fischer F344,
8-10 weeks, n = 3/group
Rat, male, Fischer F344,
200-250 g
Increased expression of surface markers associated with antigen Koike et al.
presentation: la (MHC Class II) antigen, B7.1, B7.2, and CDllb/c (2001)
on BAL cells. Effect attributed to influx of monocytes.
Decreased expression of CDS among lung lymphocytes (0.1 ppm only; Cohen et al.
3 weeks); effect exacerbated by stimulation of cells with IFN-cc (but not (2002)
with IL-lcc). Decreased expression of CD25 (IL-2R) on CD3+
lymphocytes (0.3 ppm only; 3 weeks); effect worsened by treatment
of cells with IL-lcc (0.1 and 0.3 ppm; 3 weeks).
NK- and Lymphocyte-Related Endpoints
0.1,0.3
X
4 h/day,
5 days/week,
1 or 3 weeks
Rat, male, Fischer F344,
200-250 g
0.4,0.8, 1.6
1.0
12 h
24 h/day,
3 days
Mice, male, BALB/c,
6-8 weeks old,
n = 5-8/group
Rat, male, Wistar,
8-12 weeks old
Decreased expression of CD3 among lung lymphocytes (0.1 ppm only; Cohen et al.
3 weeks); effect exacerbated by stimulation of cells with IFN-cc (2002)
(but not with IL-lcc). Decreased expression of CD25 (IL-2R) on
CD3+ lymphocytes (0.3 ppm only; 3 weeks); effect worsened by
treatment of cells with IL-lcc (0.1 and 0.3 ppm; 3 weeks).
Lymphoproliferation: no effect on spontaneous or IL-2-inducible
forms; 0.1 ppm increased response to ConA mitogen (1 week only);
0.3 ppm - decreased response to ConA (1 week only).
Decreased pulmonary delayed-type hypersensitivity reactions to low Garssen et al.
MW agents, likely via activation of TH2-dependent pathways. (1997)
Lavage fluid from exposed rats subsequently inhibited ConA-stimulated Koike et al.
lymphocyte IFN-y production, but had no effect on IL-2 production; (1999)
material also inhibited IL-2-induced lymphoproliferation.
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
Concentration
ppm
Duration
Species
Effects
Reference
Susceptibility Factors
0.3 24 to 72 h
4h
24-72 h
X
o.i
2h
Mice
C57BL/6J
C3H/HeJ
C3H/HeOuJ, 6-8 weeks
old, n = 4-8/group
CHO-K1 cell line SP-A
Mice, male, C57BL/6J,
C3H/HeJ, C3H/HeOuJ,
6-8 weeks old,
n = 5-16/group
Mice, male, C57BL/6,
6-8 weeks old
Lavageable protein concentration lowered by inhibition of iNOS and by Kleeberger
targeted disruption o£Nos2; reduced Nos2 and Tlr4 mRNA levels in the et al. (200 Ib)
O3-resistant C3H/HeJ mice.
Differences exist biochemically and functionally in SP-A variants. Wang et al.
O3 exposure affects the ability of variants to stimulate TNF-oc and IL-8. (2002)
Identified a candidate gene on chromosome 4, Toll-like receptor 4 Kleeberger
(Tlr4), a gene implicated in endotoxin susceptibility and innate et al. (2000)
immunity. O3-resistant strain C3H/HeJ and C3H/HeOuJ (differing from
the O3-resistant strain by a polymorphism in the coding region of Tlr4)
were exposed, greater protein concentrations were demonstrated in the
OuJ strain. Differential expression of Tlr4 mRNA with O3 exposure.
Suggests quantitative trait locus on chromosome 4 is responsible for a
significant portion of the genetic variance in O3 -induced lung
hyperpermeability; potential interaction between the innate and acquired
immune system.
Sensitized the mice to OVA by intratracheal instillation of OVA-pulsed Depuydt et al.
dendritic cells (the principal antigen-presenting cells in airways). (2002)
Created Th2 lymphocyte-dependent eosinophilic airway inflammation.
Groups of mice exposed to O3 during sensitization by OVA-pulsed
dendritic cells showed no modification of the allergic sensitization
process, whereas, previously sensitized mice exposed to O3,
demonstrated increases in allergen-induced airway inflammation.
Suggests that dendritic cells are an important component of O3-induced
eosinophilic airway inflammation.
-------
Table AX5-7 (cont'd). Effects of Ozone on Lung Host Defenses
Concentration
ppm
Duration
Species
Effects
Reference
X
Susceptibility Factors (cont'd)
0.1,0.5, and 1.0 2h
Human lymphocytes
NA
RatBALF;
Murine macrophage
cell line (RAW 264.7)
Subsequent to O3 exposure, when lymphocytes were stimulated Becker et al.
with pokeweed mitogen (PWM, a T-cell-dependent stimulus) (1991)
or Staphylococcus aureus Cowan 1 strain (SAC, a T-cell-independent
stimulus), both B and T cells were found to be affected by O3
preexposure. T cells also demonstrated an increase in IL-6 and
a decrease in IL-2, suggesting that O3 may have direct effects on
IgG-producing cells and concurrently an effect that is mediated
by altered production of T cell immunoregulatory molecules.
Both SP-A and SP-D directly protected surfactant phospholipids Bridges et al.
and macrophages from oxidative damage. Both proteins blocked (2000)
accumulation of TEARS and conjugated dienes generated during
oxidation of surfactant lipids or low density lipoprotein particles by
a mechanism that does not involve metal chelation or oxidative
modification of the proteins.
AM = Alveolar macrophage; PE = Postexposure (i.e., time after O3 exposure ceased); MIP = macrophage inflammatory protein;
PMN = Polymorphonuclear leukocyte; MLN = Mediastinal lymph node; CINC = cytokine-induce neutrophil chemoattractant;
BAL = Bronchoalveolar lavage; DTPA = diethylenetriaminepentaacetic acid; ROI = reactive oxygen intermediate/superoxide anion;
IFN = Interferon; BALT = Bronchus-associated lymphoid tissue; MCP = monocyte chemoattractant protein; CON A = Concanavalin A;
OVA = Ovalbumin; SP-A = Surfactant Protein A; SP-D = Surfactant Protein D
-------
Table AX5-8. Effects of Ozone on Lung Permeability and Inflammation
X
Concentration
ppni Duration
0.1 0. 5 h, in vitro
0.2
0.5
0.1 1 h, in vitro
0.2
0.4
1.0
0.2 23 h/day for 1 week
0.4
0.8
Species
Rat, SD, primary alveolar
Type II cells
Guinea pig, male and
female, Hartley, and human
alveolar macrophages
Guinea pigs, female
(Hartley), 260-330 g
n = 4-10/group
Effects
Decreased transepithelial resistance (R^) after 0.5 ppm from 2 to 24 h
PE and at 48 h in monolayers subjected to PMNs. Significantly lower
Rt after PMN treatment at 0.2 and 0.5 ppm.
Exposure of guinea pig alveolar macrophages to 0.4 ppm for 60 min
produced a significant increase in IL-6 and TNF-a, and an exposure of
human alveolar macrophages to identical O3 concentration increased
TNF-a, IL-lp, IL-6 and IL-8 protein and mRNA expression.
Increase in BALF protein and albumin immediately after
0.8 ppm exposure, with no effect of ascorbate deficiency in diet.
O3 -induced increase in BALF PMN number was only slightly
augmented by ascorbate deficiency.
Reference
Cheek et al.
(1995)
Arsalane et
(1995)
al.
Kodavanti et al.
(1995)
0.26
0.3
0.1
0.3
1.0
0.3
2.0
0.3
8 h/day, 5 days/week
for 1-90 days
48 h and 72 h.
Exposures repeated
after 14 days
60 min
72 h
3h
48 h
Mice, male (mast
cell-deficient and
-sufficient), 6-8 weeks old
n = 4-8/group
Mice, male, C57BL/6J and
C3H/HeJ, 6-8 weeks old
Rat basophilic leukemia cell
line (RBL-2H3)
Mice, male and female,
C57BL/6J and C3H/HeJ
Mice, C57BL/6J and
C3H/HeJ, 6-8 weeks old
Greater increases in lavageable macrophages, epithelial cells and Kleeberger
PMNs in mast cell -sufficient and mast cell-deficient mice repleted of et al. (2001b)
mast cells than in mast cell-deficient mice. O3-induced permeability
increase was not different in genotypic groups.
Greater BALF protein, inflammatory cell and LDH response in Paquette et al.
C57BL/6J than in C3H/HeJ after initial exposure. Repeated exposure (1994)
caused a smaller increase in BALF protein and number
of macrophages, lymphocytes and epithelial cells in both strains,
but PMN number was greater in both strains of mice compared to
initial exposure.
O3 inhibited IgE- and A23187 - indued degranulation. Spontaneous Peden and
release of serotonin and modest generation of PGD2 occurred only Dailey (1995)
under conditions that caused cytotoxicity.
Greater PMN response in C57BL/6J than in C3H/HeJ after acute and Tankersley and
subacute exposures. Responses of recombinant mice were discordant Kleeberger,
and suggested two distinct genes controlling acute and subacute (1994)
responses. Genes termed Inf-1 andInf-2.
Susceptibility to O3 is linked to a quantitative trait locus, and TNF-a Kleeberger
is identified as a candidate gene. et al. (1997)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppm Duration
Species
Effects
Reference
X
oo
0.3
1.0
2.5
24 or 48 h
I,2or4h
2, 4 or 24 h
Mice, male, C57BL/6J, 0.3 ppm for 24 h caused increase in mRNA for eotaxin, MlP-la and
8 weeks old MIP-2.
n = 3/group
1 ppm for 4 h caused increase in mRNA for eotaxin, MlP-la, MIP-2,
and IL-6.
2.5 ppm for 2 and 4 h caused increase in mRNA for MlP-la, MIP-2
and IL-6 and metallothionein. Greater increases and lethality after 24
h.
Johnston et al.
(1999a)
0.3
0.4
0.15,0.3,
or 0.5
0.5
0.5
1.0
2.0
72 h
5 weeks
3h
4h, 12-4PMfor
daytime and 7-11
PM for nighttime
exposures. Exposures
repeated 16 h later.
Mice, male {HeJ, OuJ,
Nos2 (+/+) [C57BL/6J-
Nos2 (+/+)], and Nos2
(-/-) [C57BL/6J-Nos2
(-/-)]}, 6-8 weeks old
Guinea pigs, male,
(Hartley), 5 weeks old (350-
450 g)
n = 7-8/group
Rat, male, SD
6-8 weeks old,
n = 2-6/group
Rat, male, Wistar,
60-90 days old
n = 5-15/group
Rat, male, Fisher,
90 days old
n = 6-12/group
O3 induced permeability was decreased by pretreatment with a nitric
oxide synthase inhibitor and in animals with iNOS gene knocked out.
Ovalbumin instillation in the nose caused an increase in O3 -induced
infiltration of eosinophils in nasal epithelium.
Time-related increase in permeability and inflammation, with a peak
at 8 h PE, after 0.5 ppm. No change following exposure to 0. 15 or 3
ppm.
Significantly greater increase in IL-6, but not inflammation, following
a nighttime exposure compared to daytime exposure. An initial
nighttime exposure resulted in lesser inflammation following a
subsequent exposure. Pretreatment with IL-6 receptor antibody
abolished cellular adaptive response without affecting inflammatory
response induced by initial nighttime exposure.
Increase in BALF protein and albumin occurred immediately
after 2 ppm exposure, and at 18 h after 1 ppm. No increase after
0.5 ppm. The movement of water and protein into airspace were not
coupled.
Kleeberger
etal. (200 la)
lijima et al.
(2001)
Bhalla and
Hoffman
(1997)
McKinney et al.
(1998)
Cheng et al.
(1995)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppm
Duration
Species
Effects
Reference
1.0-2.0
3h
0.5
0.5
X
(Si
I
VO
24 h folio wing a 3-d (6
h/day) exposure to
cigarette smoke
8 h during nighttime
2h and 6 h
Mice, C57BL/6, 6-8 weeks
old and rats, Wistar,
14-16 weeks old
n = 4-6/group
Mice, male, B6C3F1,
25 ± 2 g, 10 weeks old,
n = 6/group
Rat, male, Wistar, SD and
Fischer F344, 90 days old
n = 3-8/group
Rats, male, Fisher F344,
Juvenile (2 months;
180-250 g),
Adult (9 months;
370-420 g),
Old (18 months;
375-425 g), Senescent
(24 months; 400-450 g)
n = 2/group
Steady state MCP-1 mRNA increase after 0.6 ppm, with maximal Zhao et al.
increase after 2 ppm. After 2 ppm, MTP-2 mRNA peaked at 4 h PE (1998)
and MCP-1 mRNA peaked at 24 h PE. HALF neutrophils and
monocytes peaked at 24 and 72 h PE, respectively. BALF MCP-1
activity induced by CX was inhibited by an anti-MCP-1 antibody.
J J J J J
BALF protein, neutrophils and lymphocytes were increased in animals Yu et al. (2002)
exposed to smoke and then to O3. Macrophages from this group also
responded with greater release of TNF-a upon LPS stimulation.
Exposure resulted in a significantly greater injury, inflammation and Dye et al.
BALF levels of IL-6 in Wistar than in SD or F344 rats. (1999)
Comparable effect on the leakage of alveolar protein in rats Vincent et al.
of different age groups, but a greater increase occurred in interleukin-6 (1996).
and N-acetyl-beta-D-glucosaminidase in senescent animals than in
juvenile and adult rats.
0.8 3h
0.8 3h
0.8 3h
Rat, male, SD,
6-8 weeks old,
n = 5/group
Rat, male, SD
6-8 weeks old,
n = 5/group
Rat, male, SD,
200-225 g,
n = 2-9/group
Increased adhesion of macrophages from exposed animals to
rat alveolar type II epithelial cells in culture. Treatment with
anti-TNF-a + anti-IL-la antibody decreased adhesion in vitro, but not
permeability in vivo.
Increase in fibronectin protein in BALF and lung tissue, and
fibronectin mRNA in lung tissue. The increase produced by O3 was
amplified in animals pre-treated intra-tracheally with rabbit serum to
induce inflammation.
Treatment of animals with IL-10 prior to O3 exposure caused a
reduction in O3 induced BALF protein, albumin and fibronectin and
tissue fibronectin mRNA.
Pearson and
Bhalla(1997)
Gupta et al.
(1998)
Reinhart et al.
(1999)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
X
to
o
Concentration
ppm
0.8
0.8
1.0-2.0
0.8
0.8
1.0
1.0
0.2
0.5
1.0
1.0
Duration
8h
48 h
3h
8h
3h
5 min exposure of
airway segments
following
bronchoscopy
8 h, assayed 1 and
2hPE
In vitro at liquid/air
interface
3h
Species
Monkey, male, Rhesus,
3 years 8 months-3 years
10 months old
n = 2-6/group
Rat, male, SD,
6-8 weeks old,
200-225 g,
n = 3-8/group
Monkeys, male, Rhesus,
3 years 8 months-3 years
10 months old (5. 1-7. 6 kg)
n = 2-6/group
Mice, female,
C57BL/6X 129 NOS''-
knockout and wild-type
B6J129SVF2,
8-16 weeks old,
n = 3-12/group
Dogs, male, Mongrel,
Adult,
n = 1 -4/group
Monkeys (Rhesus)
Primary TBE, BEAS-2b S
andHBEl
Rat, male, SD,
6-8 weeks old,
n = 4-5/group
Effects
Pretreatment of monkeys with a monoclonal anti-CD18 antibody
resulted in a significant inhibition of O3-induced neutrophil emigration
and accumulation of necrotic airway epithelial cells.
Cyclophosphamide treatment ameliorated O3 -induced BALF
neutrophils and albumin after short term and 1-day exposure. Anti-
neutrophil serum reduced lavageable neutrophils but did not affect
permeability.
Tracheal epithelium of exposed animals expressed b6 integrin. The
integrin expression was reduced or undetectable in animals treated
with CD- 18 antibody.
Alveolar macrophages from O3 exposed wild-type mice produced
increased amounts of NO, per oxy nitrite, superoxide anion, and PGE2.
Nitrogen intermediates were not produced and PGE2 was at control
level in exposed NOSH knockout mice. These mice were also
protected from O3 -induced inflammation and injury.
Mast cells from O3 -exposed airways of ascaris sensitive dogs released
significantly less histamine and PGD2 following in vitro challenge
with ascaris antigen or calcium ionophore.
Increase in steady state IL-8 mRNA in airway epithelium. Increase in
IL-8 protein staining declined at 24 h after exposure.
Dose related increase in IL-8 release in the conditioned media. Ozone
produced greater toxicity in cell lines than in primary cultures.
Time-related increase in BALF protein, fibronectin (Fn), and alkaline
phosphatase (AP) activity. Fn mRNA detected in macrophages, and
AP in Type II cells and in BALF PMNs from exposed animals only.
Reference
Hyde et al.
(1999)
Bassett et al.
(2001)
Miller et al.
(2001)
Fakhrzadeh
et al. (2002)
Spannhake
(1996)
Chang et al.
(1998)
Bhalla et al.
(1999)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppni Duration Species
1 2 h Rats, female, SD,
170-210 g
n = 8-12/group
1 3 h Rat, male, SD,
6-8 weeks old,
n = 5/group
1 3 h Rat, male, SD,
250-275 g,
n = 6/group
Effects
The expression of GDI 8 on alveolar macrophages and GDI Ib on
blood PMNs was lowered by exposure, but CD62L expression on
blood PMNs was not affected.
Time-related increase in BALF albumin, PMNs, MTP-2 and ICAM-1,
and increase in MTP-2 mRNA only at early time point in BALF
macrophages. MIP-2 mRNA not detected in lung tissue.
Ozone induced increase in BALF albumin, fibronectin and PMN
number was associated with an increase in expression of TNF-a, IL-
la, IL-6 and IL-10 mRNA. Pretreatment with anti-TNF-a antibody
caused downregulation of gene expression and reduction of BALF
albumin and PMN number, but not fibronectin.
Reference
Hoffer et al.
(1999)
Bhalla and
Gupta (2000)
Bhalla et al.
(2002)
X
(Si
to
0.5
1.0
2.5
1.0
1.0
6h
4h
8 h/night for three
nights
4h
Rat, male, SD,
200-250 g,
n = 3-6/group
Mice, male
(129 wild-type or Clara Cell
Secretory Protein"'"),
2-5 months old,
n = 3/group
Mice, (C57BL/6 wild-type
and iNOS knockout)
n = 3-10/group
Mice, male (129 strain,
wild-type and Clara Cell
Secretory Protein-deficient),
2-3 months old,
n = 3/group
Increase in number of macrophages with mRNA transcripts and Ishii et al.
immunocytochemical staining of IL-1, TNF-a, MIP-2 and cytokine- (1997)
induced neutrophil chemoattractact (CINC). Chemokine activities
were reduced by treatment of macrophages with anti-IL-1P and anti-
TNF-a antibodies.
Increases in IL-6 and metallothionein mRNA by 2 h after exposure to Mango et al.
1 ppm. mRNA increases were further enhanced in CCSP"'" mice. (1998)
O3 exposure produced greater injury, as determined by measurement Kenyon et al.
of MIP-2, matrix metalloproteinases, total protein, cell content and (2002)
tyrosine nitration of whole lung protein, in iNOS knockout mice than
in wild-type mice.
Increases in abundance of mRNAs encoding eotaxin, MlP-la Johnston et al.
and MIP-2 in CCSP"'", but no change in wild-type mice. (1999b)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
X
to
to
Concentration
ppni Duration
1.2 6h
2.0 3h
2.0 3h
1.1 8h
0.32 48 h
(subacute)
3h
(acute)
0.3 24 to 72 h
2.0 3h
Species
Rat, male, Brown Norway,
200-250 g,
n = 4/group
Mice, male, C57BL/6J,
6-8 weeks old,
n = 5-10/group
Rat, female, SD,
6-8 weeks old
Rat, Wistar, -depleted of
neutrophils, 43 days old,
n = 9-10/group
Mice, male, C57BL/6J,
WT
TNRF1KO
TNRF2KO
6-8 weeks old,
n = 3-12/group
Mice
C57BL/6J
C3H/HeJ
C3H/HeOuJ
6-8 weeks old,
n = 5-16/group
Mice C3H/HeJ, A/J,
C57BL/6J, 129/SvIm,
CAST/Ei, BTBR, DBA/2J,
FVB/NJ, BALB/cJ,
n = 6-24/group
Effects
Eotaxin mRNA expression in the lungs increased 1 .6-fold immediately
after and 4-fold at 20 h. Number of lavageable eosinophils increased
3- and 15-fold respectively at these time points. Alveolar
macrophages and bronchial epithelial cells stained positively for
eotaxin.
O3-induced increase in protein and PMNs in BALF, and pulmonary
epithelial cell proliferation were significantly reduced in animals pre-
treated with UK-74505, a platelet activating factor-receptor antagonist.
BALF cells from exposed animals released 2 to 3 times greater IL-1
and TNF-a, and greater fibronectin. Immunocytochemistry showed
greater staining of these mediators in lung tissue from exposed rats.
Epithelial necrosis in the nasal cavity, bronchi, and distal airways.
Proliferation of terminal bronchiolar epithelial cells also decreased
by O3 exposure, suggesting a role for neutrophils in the repair process.
TNFR1 and TNFR2 KOs less sensitive to subacute O3 exposure than
WT. With acute exposures, airway hyperreactivity was diminished in
KO mice compared to WT mice, but lung inflammation and
permeability were increased.
Differential expression of Tlr4 mRNA.
Two strains consistently O3 -resistant: C3H/HeJ and A/J. Two strains
consistently O3-vulnerable: C57BL/6Jand 129/SvIm. Five strains
with inconsistent phenotypes with intermediate responses: CAST/Ei,
BTBR, DBA/2J, FVB/NJ, and BALB/cJ.
Reference
Ishii et al.
(1998)
Longphre et al.
(1999)
Pendino et al.
(1994)
Vesely et al.
(1999)
Cho et al.
(2001)
Kleeberger
et al. (2000)
Savov et al.
(2004)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppm
Duration
Species
Effects
Reference
1
0.4
X
(Si
to
1.0 or 2.5
3 h- examined at 3 h PE Guinea pigs, male,
Dunkin-Hartley,
OVA-sensitized
1 h examined 24 h PE
1 or 5 days/ 12 h/day,
recovery period in fresh
air of 5, 10, 15, or 20
days after the 5-d
preexposure
2 h, examined 2, 12,
and 48 h PE
4, 20, or 24 h,
examined immediately
PE
lOmin
3 days, continuous
Rat, Wistar, male,
7 weeks old,
n = 5/group
Rat, female, SD,
200-250 g
n = 4-7/group
Mice, C57BL/6J, 36 h and !
weeks old
Endotoxin (10 ng)
n = 3/group
Rat, male, SD,
30 days old,
n = 6/group
PMN levels significantly increased, without any change in BAL Sun et al.
protein levels, suggesting a lack of correlation between the two (1997)
endpoints. Increased AHR.
Increase in PMN, no increase in BAL protein levels. No increased
AHR, suggesting a dissociation between PMN levels and AHR.
Exposure for 5 days caused lower BALF proteins, fibronectin, IL-6, Van Bree
and inflammatory cells than animals exposed for 1 day. Postexposure et al. (2002)
challenge with single O3 exposures at different time points showed
that a recovery of susceptibility to O3 (as measured by BALF levels of
albumin, IL-6, and the number of macrophages and neutrophils)
occurred at -15-20 days, but total protein and fibronectin levels
remained attenuated even at 20 days post-5-day exposure. The
recovery with regards to BrdU labeling occurred in 5-10 days after the
5-day exposure.
Adherence of neutrophils to pulmonary vascular endothelium was Lavnikova et al.
maximal within 2 h after exposure and returned to control levels by 12 (1998)
hPE.
RPA for IL-12, IL-10, IL-la, IL-lp, IL-IRa, MIF, IFN-y, MlP-la, Johnston et al.
MIP-2JL-6, and Mt. Newborn mice: increased Mt mRNA only. (2000b)
8-week-old mice: increased MlP-la, MIP-2, IL-6, and Mt mRNA.
Both age groups had similar cytokine/chemokine profiles with
endotoxin exposure, suggesting that the responses to endotoxin, which
does not cause epithelial injury, and the responses to O3, which does,
demonstrate that differences in inflammatory control between
newborn and adult mice is secondary to epithelial injury.
General dietary restriction to 20% of the freely-fed diet for Elsayed (2001)
60 days caused an extreme reduction in body weight and higher
survivability. Levels of antioxidants and detoxifying enzymes
increased less than in freely fed animals.
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppm
Duration
Species
Effects
Reference
1.8
0.11
X
(Si
to
lor 3
lor 3
0.1-2
3h
24 h/day for up to
3 days, assays
immediately or at 16 h
PE
2-6 h, assayed at 2,
8 and 24 h PE
3h
3 h, assayed 6 h PE
(exposure-response)
3 h, assayed 0, 2, 6, or
24 h PE (time-course)
Mice, female, C57BL/6J
CBAC3H/HeJAKR/J
SJL/J, 6-8 weeks old,
n = 4-7/group
Rat, female, Brown
Norway, 250-300 g,
n = 4-8/group
Rat, male, Brown Norway,
200-250 g,
n = 3/group
Mice,
C57BL/6,
n = 2-4/group
Both exposure levels caused a transient increase in CC16 in Broeckaert
serum that correlated with BALF changes in protein, LDH, and et al. (2003)
inflammatory cells. Inverse relationship between preexposure levels
of CC16 in BALF and epithelial damage based on serum CC16 levels
and BALF markers of inflammation. Inverse relationship between
preexposure levels of albumin in BALF and lung epithelium damage.
Suggests that a major determinant of susceptibility to O3 is basal lung
epithelial permeability. C57BL/6J mice had lower levels of CC16a
(the more acidic form) than C3H/HeJ. Both the strains had similar
levels of CC16b. Suggests that basal lung epithelial permeability is a
major determinant of susceptibility; greater epithelial permeability
observed in C57BL/6J may be due to difference in the expression of
CC16a and possibly other antioxidant/inflammatory proteins.
MIP-2 peaked at 2 h and rapidly declined. PMNs in BALF increased Haddad et al.
at 2, 8 and 24 h. No significant increase in AMs, eosinophils, (1995)
lymphocytes or epithelial cells. MIP expression preceded increase in
PMN. Both responses suppressed by dexamethasone, which suggests
a mechanism of glucocorticoid regulation of inflammation.
Increase in lung CINC mRNA within 2 h PE exposure. Koto et al.
Significant increase in PMNs in BALF 24 h PE. Anti-CINC antibody (1997)
(1 mg, i.v.) suppressed neutrophilia but not the increase in AHR
to acetylcholine. Anti-CINC antibody inhibited BALF neutrophilia
induced at 3 ppm AHR. Results suggest that CINC causes O3-induced
neutrophil chemoattraction, but is not involved in the induction of
ozone-induced AHR.
O3 > 1 ppm increased MIP-2 mRNA and recruitment of neutrophils. Driscoll et al.
MIP-2 increase was immediate and decreased to control by 24 h PE. (1993)
-------
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppm
Duration
Species
Effects
Reference
0.12,
0.24,
0.5
3 h, NF-KB assayed 0 h
PE, TNF-a assayed 0,
1,2,4, 16, 20 or 22 h
PE
Cultures of human nasal
epithelial (HNE) cells
Electron spin resonance signal suggested free radical production.
Small dose-response activation of NF-KB coincided with O3-induced
free radical production. TNF-a increased with exposures to 0.24 and
0.5 ppm at 16 h PE. Results suggest that the human airway epithelium
plays a role in directing the inflammatory response to inhaled O3 via
free radical-mediated NF-KB.
Nichols et al.
(2001)
3h
X
(Si
to
Rats, SD, 200-250 g,
n = 5-6/group
AMs from O3-exposed rats exhibited greater motility and greater
adhesion in cultures of epithelial cells (ARL-14). O3-induced motility
and adhesion were attenuated with AMs incubated in the presence of
Mabs to leukocyte adhesion molecules, GDI Ib, or epithelial cell
adhesion molecules, ICAM-1. Increased surface expression of GDI Ib
but no change in ICAM-1 expression in AMs from O3-exposed rats.
Demonstration of alteration of AM functions following O3 exposure.
Possible dependence of these functions on the biologic characteristics,
rather than the absolute expression, of cell adhesion molecules.
Bhalla(1996)
3h
Mice, male and female,
C57BL/6J, assayed
immediately or 3, 6, 9, or 21
hPE
Increase in tissue expression of ICAM-1 3 -9 h PE, remaining until 21
h PE. Bronchioles and terminal bronchiole/alveolar duct regions:
Enhanced 1C AM-MR 0 - 3 h PE, returning to baseline by 21 and 9 h,
respectively. Lung parenchyma: maximal ICAM-1 expression and
PMN influx concurrent 3 h, followed by transepithelial migration of
PMNs to the airway lumen. Suggests regional variations in airway
inflammatory activity; upregulation of ICAM-1 may play a role in
local regulation of PMN influx to the airways after acute exposure.
Takahashi et al.
(1995a)
4h
Rat, SD, male, 225-250 g,
treated with 10 mg/kg
ebselen every 12 h from 1 h
before O3 exposure,
n = 4/group
Ebselen significantly decreased pulmonary inflammation (albumin and
PMN in BALF) 18 h PE without altering AM expression of iNOS.
Ebselen inhibited the nitration reaction of tyrosine residues and
enhanced expression of Cu-, Zn-,and Mn SOD. Suggests that ebselen
scavenges peroxynitrite during O3-induced inflammation and may
protect against acute lung injuries by modulating the oxidant-related
inflammatory process.
Ishii et al.
(2000a)
-------
X
(Si
to
Table AX5-8 (cont'd). Effects of Ozone on Lung Permeability and Inflammation
Concentration
ppm
3
Duration
2h
Species
Human transformed
Effects
NO donors increased IL-8 production dose-dep(
Reference
sndently. TNF-a plus Inoue et al.
0.12,
0.5,
l,or
2
3h
bronchial epithelial cells
(16-HBE)
Guinea pigs, Hartley, male,
450-550 g,
n = 3-5/group
Mice, female, BALB/c, 5-6
weeks old
IL-1 P plus INFy increased IL-8 in culture supernatant of epithelial (2000)
cells. NOS inhibitors (aminoguanidine plus NG-nitro-L-arginine
methyl ester) attenuated the cytokine-induced IL-8 production.
O3 induced AHR to acetylcholine and increased PMN in BALF.
Persisting for 5 h. Pretreatment with NOS inhibitors did not affect
AHR or PMN accumulation 0 h PE, but, inhibited at 5 h PE.
Suggests that endogenous NO, through upregulation of IL-8,
modulates O3-induced airway inflammation and AHR.
O3 exposure caused dose -dependent increases in nitrate (indicative of
in vivo NO generation). Increases in enhanced pause (P^ were also
dose-dependent. Increases in NOS-1, but not in NOS-3 or iNOS
isoforms. Suggest that NOS-1 may induce airway responsiveness by
neutrophilic airway inflammation.
Jang et al.
(2002)
PMN = Polymorphonuclear leukocyte
PE = Postexposure (time after O3 exposure ceased)
BAL = Bronchoalveolar lavage
BALF = Bronchoalveolar lavage fluid
-------
Table AX5-9. Effects of Ozone on Lung Structure: Acute and Subchronic Exposures
X
to
Concentration
ppm
0.1
0.5
1.0
0.2
0.4
0.2
0.4
0.8
0.4
Duration
8 h/day * 1 day
8 h/day x 1 day
8 h/day x 1, 10,
75, and 90 days
3, 7, 28, and
56 days; 3-, 7-,
and 28-day
recovery from
28 days of
exposure
23 h/day for
7 days
12 h/day;
1- or 7-day
exposure
Species
Rat; male, SD,
350-600 g,
n = 3-6/group
Mice, male, NIH,
Rat, male, Wistar
RIV:Tox
Guinea pig,
male, Hartley
Crl:(HA)BR,
7 weeks old,
n= 3-9/group
Guinea pig, female,
Hartley; ±AH2 diet
Rat, Wistar
RiV:TOX,
male and female,
1, 3, 9, and 18 months
of age,
n = 5-6/group
Effects
No dose-related response on CYP2E1, one of six P450 enzymes identified
in respiratory tissue. CYP2E1 activity was elevated (250% and 280%)
in the lobar bronchi/major daughters airways immediately after 1.0 ppm O3
exposure for 1 day and 10 days, respectively, but not in the trachea or distal
bronchioles; CYPE1 activity was unchanged and decreased after 1.0 ppm
O3 exposure for 75 and 90 days, respectively.
Concentration-related centriacinar inflammation, with a maximum after
3 days of exposure; number of alveolar macrophages and pulmonary cell
density increased progressively until 56 days of exposure, with the guinea
pig the most sensitive species. Concentration and exposure-time dependent
hypertrophy of bronchiolar epithelium in mouse only. Exposure to 0.2 ppm
for 3 and 7 days caused significant histological and morphometric changes
in all 3 species; exposure for 56 days caused alveolar duct fibrosis in rat
and guinea pigs. Total recovery in rats after 28-d exposure, but not in
guinea pigs or mice.
Treatment-related lesions were observed after exposure to 0.4 and 0.8 ppm
O3; lesions were primarily seen in the terminal bronchioles and consisted
of mononuclear cell and neutrophilic infiltrate and thickening of the
peribronchiolar interstitium. Effects were only marginally exacerbated
by the AH2 (ascorbic acid) deficient diet and lesions were resolved after
1 week in FA.
Centriacinar inflammation (increased alveolar macrophages and PMNs;
increased proximal and ductular septal density) was greatest in young rats
(1 month and 3 months for 1-and 7-day exposures, respectively) and
decreased with age. No major gender differences were noted.
Reference
Watt et al.
(1998)
Dormans
et al. (1999)
Kodavanti
et al. (1995)
Dormans
et al. (1996)
-------
Table AX5-9 (cont'd). Effects of Ozone on Lung Structure: Acute and Subchronic Exposures
Concentration
ppm Duration
Species
Effects
Reference
0.4
1.0
X
to
oo
2h
Monkey;
adult male Rhesus
Reduced glutathione (GSH) increased in the proximal intrapulmonary
bronchus after 0.4 ppm O3 and decreased in the respiratory bronchiole after
1.0 ppm O3. Local O3 dose (measured as excess 18O) varied by as much as a
factor of three in different airways of monkeys exposed to 1.0 ppm, with
respiratory bronchioles having the highest concentration and the parenchyma
the lowest concentration. After exposure to 0.4 ppm, the O3 dose was 60%
to 70% less and epithelial injury was minimal, except in the respiratory
bronchiole, where cell loss and necrosis occurred, but was 50% less than
found at 1.0 ppm.
Plopper et al.
(1998)
0.5 8h
+ BrdU to label
epithelial cells
0.5 8 h/day,
3 or 5 days;
+ fluticasone
propionate (FP)
intranasally
0.5 8 h/day,
3 days
+ endotoxin
(100 ug/mL)
intranasally
Rat, male, Fischer
F344, 13 weeks old,
n = 6/group
Rat, male, Fischer
F344, 203-232 g,
n = 6/group
Rat, male, Fischer
F344/N Hsd, 12
weeks old,
n = 6/group
O3 exposure induced a transient influx of neutrophils and a significant (17%)
loss of NTE cells 2-4 h after exposure. Increased epithelial DNA synthesis
was first detected 12 h PE. LI and ULLI indices of epithelial cell DNA
synthesis were greatest 20-24 h and still elevated 36 h PE; numeric density
of NTE cells returned to control levels 20-24 h PE.
No significant difference of fluticasone propionate on morphometry of the
maxilloturbinates; O3 exposure caused neutrophilic rhinitis with 3.3- and
1.6-fold more intraepithelial neutrophils (3-day and 5-day exposure,
respectively) and marked mucous cell metaplasia (5-day exposure only)
with numerous mucous cells and approximately 60 times more
intraepithelial mucosubstances in the nasal transitional epithelium;
FP-treated rats exposed to O3 had minimal nasal inflammation and mucous
cell metaplasia.
Endotoxin-induced neutrophilia in nasal mucosa with NTE; mucous cell
metaplasia was not detected in air/endotoxin-exposed rats, was observed
in O3/saline-exposed rats, and was most severe in O3/endotoxin-exposed
rats.
Hotchkiss
et al. (1997)
Hotchkiss
et al. (1998)
Fanucchi
et al. (1998)
-------
Table AX5-9 (cont'd). Effects of Ozone on Lung Structure: Acute and Subchronic Exposures
Concentration
ppm
0.5
Duration
8h/day,
1, 2, or 3 days
+ BrdU to label
epithelial cells
+ antirat
neutrophil
antiserum
Species
Rat, male, Fischer
F344/N, 10-12 weeks
old, n = 6-8/group
Effects
Acute O3 exposure induced a rapid increase in rMuc-5AC mRNA levels
prior to the onset of mucous cell metaplasia; neutrophilic inflammation
coincided with epithelial DNA synthesis and upregulation, but was
resolved when mucous cell metaplasia first appeared in the NTE.
Maxilloturbinates lined with NTE determined the epithelial labeling
index, numeric densities of neutrophils, total epithelial and mucous secretory
Reference
Cho et al.
(1999a, 2000)
0.5
X
to
VO
8h/day,
3 days
+ endotoxin
Rat, male, Fischer
F-344, 10-12 weeks
old, n = 6/group
cells, amount of stored intraepithelial mucosubstances, and steady-state
rMUC-5AC mRNA levels. Four days after a 3-d exposure,
antiserum-treated, O3-exposed rats had 66% less stored intraepithelial
mucosubstances and 58% fewer mucous cells in their NTE than did controls.
Antiserum treatment had no effects on O3-induced epithelial cell
proliferation or mucin mRNA upregulation.
Enhanced epithelial lesions in the NTE and respiratory epithelium of the
nose and conducting airways by endotoxin and O3 exposures, respectively;
synergistic effects of coexposure mediated by neutrophils. Endotoxin
increased rMuc-5 AC mRNA levels in the NTE of O3-exposed rats;
neutrophil depletion, however, had no effect on endotoxin-induced
upregulation of mucin gene mRNA levels. Endotoxin enhanced the
O3-induced increase in stored mucosubstances (4-fold increase),
but only in neutrophil-sufficient rats.
Wagner et al.
(2001a,b)
0.5 8 h/day,
1 and 3 days
+ OVA
(1%, 50 uL/nasal
passage)
1 8h
Rat, male, Brown
Norway, 10-12 weeks
old, n = 6/group
Rat, male, SD,
10 weeks old,
Ferret, male, 18
months old,
Monkey, male,
Rhesus, 4 years old,
n = 4-8/group
O3 enhanced the appearance of eosinophils in the maxilloturbinates of
OVA-challenged rats but did not increase inflammation in other nasal
tissues; O3/OVA coexposures for 3 days increased the number of epithelial
cells as well as the appearance of mucus-containing cells in the NTE lining
the maxilloturbinates.
Severe, acute infiltration of neutrophils along with necrotic bronchiolar
epithelium in all lung regions, especially in the centriacinar region; necrosis
and inflammation was more severe in ferrets and monkeys than in rats.
Wagner et al.
(2002)
Sterner-Kock
et al. (2000)
-------
Table AX5-9 (cont'd). Effects of Ozone on Lung Structure: Acute and Subchronic Exposures
Concentration
ppm Duration
Species
Effects
Reference
0.5 8 h/day for 3 days,
assayed 2 h or
4 days PE
Rat, male, Fischer
F344/N, with or
without prior
exposure to
100 ug/day
endotoxin,
10-12 weeks old,
n = 8/group
2 h PE: Endotoxin/O3 rats had 48 and 3 times more PMNs in the NTE than
did saline/air- and saline/O3-exposed rats, respectively at 2 h PE. O3-only
rats had 35% more NTE cells and 2-fold more mucin mRNA than did
saline/air-exposed rats.
4 days PE: Endotoxin/O3 rats had 5 and 2 times more IM and mucous cells,
respectively, than did saline/air- and saline/O3- rats.
Cho et al.
(1999b)
8 h/day for
90 days
>
X
(Si
Rats, SD, male,
275-300 g, treated
i.p. with
1 -nitronaphthalene
(0, 50, or 100 mg/kg)
No mucous cell metaplasia was present in those rats killed at 4 days
postexposure. Suggests that pre-existing rhinitis augments O3-induced
mucous cell metaplasia.
1-nitronaphthalene (a pulmonary toxicant requiring metabolic activation)-
treated rats exposed to O3 showed greater histopathologic and morphometric
effects in the centriacinar region of the lung. Caused denudation of the
basement membrane and necrosis of remaining epithelial cells. Increased
severity of ciliated cell toxicity in O3-exposed rats. No differences in the
intrapulmonary airways or trachea in sensitivity to 1-nitronaphthalene,
suggesting a site-selective synergy between O3 and 1-nitronaphthalene.
Paige et al.
(2000b)
AM = Alveolar macrophage
PE = Postexposure (i.e., time after O3 exposure ceased)
LM = Light microscopy
EM = Electron microscopy
RB = Respiratory bronchiole
TB = Terminal bronchiole
IAS = Interalveolar septum
PMN = Polymorphonuclear leukocyte
-------
Table AX5-10. Effects of Ozone on Lung Structure: Subchronic and Chronic Exposures
>
X
Concentration
ppm
Mexico City
Ambient:
0.018
(>0.12for
18 1-h
intervals)
0.12
0.5
1.0
0.12
0.50
1.0
0.12
1.0
Duration
23 h/day for
7 weeks
6 h/day,
5 days/week
for 20 months
6 h/day,
5 days/week
for 24 and
30 months
6 h/day,
5 days/week,
for 2 or
3 months
Species
Rat, male and
female, Fischer
F344,
8 weeks old
Rat, male,
Fischer F344,
6-8 weeks old
Mice, male and
female, B6C3F1,
6-7 weeks old,
n = 50/group
Rat, male,
Fischer F344/N,
4-5 weeks old,
n = 4/group
Effects
No inflammatory or epithelial lesions in nasal airways or respiratory tract.
LM morphometry of CAR remodeling. Thickened tips of alveolar septa
lining ADs (alveolar entrance rings) 0.2 mm from TB in rats exposed to 0. 12 ppm
and to 0.6 mm in rats exposed to 1.0 ppm. At 0.5 and 1.0 ppm, atrophy of nasal
turbinates, mucous cell metaplasia in NTE, increased volume of interstitium and
epithelium along ADs due to epithelial metaplasia, and bronchiolar epithelial
hyperplasia. At 1.0 ppm, increased AMs and mild fibrotic response (increase in
interstitial matrix and cellular interstitium; the latter due to increase in volume
in interstitial fibroblasts). More effects in PAR than in terminal bronchioles.
Effects not influenced by gender or by aging. Effects similar to, or model of,
early fibrotic human disease (e.g., idiopathic pulmonary fibrosis).
Effects in the nose and centriacinar region of the lung at 0.5 and 1.0 ppm.
Nasal lesions were mild: hyaline degeneration, hyperplasia, squamous
metaplasia, fibrosis, suppurative inflammation of transitional and respiratory
epithelium; and atrophy of olfactory epithelium. Lung lesions:
alveolar/bronchiolar epithelial metaplasia and histiocytosis in terminal
bronchioles, alveolar ducts, and proximal alveoli. Severity was greatest in
mice exposed to 1.0 ppm O3, but there was minimal interstitial fibrosis.
Morphometric changes (epithelial thickening, bronchiolarization) occurred after
2 or 3 months exposure to 1.0 ppm O3; effects were similar to those found with
20 months exposure (see Pinkerton et al., 1995)
Reference
Moss et al.
(2001)
Catalano et al.
(1995a,b);
Chang et al.
(1995);
Harkema et al.
(1994, 1997a,b)
Pinkerton et al.
(1995);
Plopper et al.
(1994);
Stockstill et al.
(1995)
Herbert et al.
(1996)
Pinkerton et al.
(1998)
-------
Table AX5-10 (cont'd). Effects of Ozone on Lung Structure: Subchronic and Chronic Exposures
Concentration
ppm
0.011
0.25
0.5
0.4
Duration
6 months
8h/day,
7 days/week
for 13 weeks
23.5 h/day
for 1, 3, 7, 28,
or 56 days
Species
Rat, male,
Wistar, 2 months
Rat, male,
Fischer F344/N
HSD,
10-14 weeks old
n = 5-6/group
Rat, Wistar,
7 weeks old,
n = 5/group
Effects
CO = 1.25 ppm; PM = 35.18 ug/m3; SO2 = 29.05 ug/m3. Induction of secretory
hypertrophy, acidic mucous secretion, and ciliary damage.
Mucous cell hyperplasia in nasal epithelium after exposure to 0.25 and 0.5 ppm
O3; still evident after 13 weeks recovery from 0.5 ppm O3 exposure. Mucous cell
metaplasia found only after 0.5 ppm O3, but still detectable 13 weeks PE.
Acute inflammatory response (increased PMNs and plasma protein in B ALF)
reached a maximum at day 1 and resolved within 6 days during exposure;
AMs in B ALF increased progressively up to day 56, and slowly returned to near
Reference
Lemos et al.
(1994)
Harkema et al.
(1999)
Van Bree et al.
(2002)
>
X
(Si
I
to
control levels with PE recovery. Histological examination and morphometry
of the lungs revealed CAR inflammatory responses throughout O3 exposure;
thickening of septa was observed at day 7. Ductular septa thickened progressively
at day 7, 28, and 56 of exposure; showed increased collagen at day 28, which was
further enhanced at day 56. Increased RBs with continuous exposure. Collagen
and bronchiolization remained present after a recovery period.
0.5 8 h/day for
1, 3, and
6 months
0.5 8 h/day for
5 days,
every 5 days
for a total of
1 1 episodes
Rat, male,
Fischer F344/N
Monkey;
Rhesus,
30-day-olds,
n = 6/group
Increased Bcl-2, a regulator of apoptosis, after 1 month, decreasing somewhat
thereafter, returning to baseline by 13 weeks PE; increased number of metaplastic
mucous cells in NTE after 3 and 6 months.
Increased density and distribution of goblet cells in RB whole mounts stained
with AB/PAS; extensive remodeling of distal airway with O3 and O3 + HDMA
challenge; increased airways resistance and reactivity, and respiratory motor
adaptation also occurred. Authors conclude that periodic cycles of acute injury
and repair associated with the episodic nature of environmental patterns of O3
exposure alters postnatal morphogenesis and epithelial differentiation in the
distal lung of infant primates.
Tesfaigzi et al.
(1998)
Schelegle et al.
(2003);
Chen et al.
(2003);
Plopper and
Fanucchi (2000)
-------
Table AX5-10 (cont'd). Effects of Ozone on Lung Structure: Subchronic and Chronic Exposures
>
X
Concentration
ppm Duration Species
Effects
Reference
0.8
0.5
0.5
8 h/day for
90 days
+ 1-NN
(100 mg/kg)
1 1 episodes of
5 days each,
8 h/day
followed by
9 days of
recovery
1 1 episodes of
5 days each,
8 h/day
followed by
9 days of
recovery
Rat, male, SD,
275-301 g
Monkey,
Macaca mulatta,
30 days old
Monkey,
Rhesus,
30 days old,
n = 6/group
Increased O3 -induced centriacinar toxicity (histopathology, TEM, morphometry)
of 1-Nitronaphthalene (1-NN), a pulmonary cytotoxicant requiring metabolic
activation, especially to ciliated cells.
In small conducting airways O3 caused decrements in density of airway epithelial
nerves. Reduction greater with HDMA + O3. O3 or HDMA + O3 caused increase
in number of PGP 9.5 (pan-neuronal marker) in airway. CGRP-IR nerves were
in close contact with the PGP9.5 positive cells. Appearance of clusters of
PGP9.5+/CGRP cells. Suggests episodic O3 alters developmental pattern of
neural innervation of epithelial compartment.
Abnormalities in the BMZ included: (1) irregular and thin collagen throughout
the BMZ; (2) perclecan depeleted or severely reduced; (3) FGFR-1
immunoreactivity was reduced; (4) FGF-2 immunoreactivity was absent in
perlecan-deficient BMZ, but was present in the lateral intercelluar space (LIS),
in basal cells, and in attenuated fibroblasts; (5) syndecan-4 immunoreactivity
was increased in basal cells.
Paige et al.
(2000b)
Larson et al.
(2004)
Evans et al.
(2003)
TB = Terminal bronchiole
PE = Postexposure (i.e., time after O3 exposure ceased)
AM = Alveolar macrophage
LM = Light microscopy
TEM = Transmission Electron Microscopy
BMZ = Basement Membrane Zone
EM = Electron microscopy
RB = Respiratory bronchiole
IAS = Interalveolar septum
C x T = Product of concentration and time
-------
Table AX5-11. Effects of Ozone on Pulmonary Function
Concentration
ppm
Duration
Species
Effects
Reference
0.5
>
X
0.3
6 or 23 h/day
over 5 days
3 h, assayed 1
and24hPE
Rats, male,
Fischer 344,
90 days old,
n = 28-36/group,
ambient temperature
10, 22, or 34 °C
Mice, male,
C57BL/6J C3H/HeJ,
53 days old,
n = 5-7/group
challenged by CO2
(5 or 8%)
48 and 72 h, with Mice,
re-exposure after C57BL/6J,
14 days of C3H/HeJ,
recovery 6-8 weeks old
2-3 h, assayed
OhPE
Mice, male,
C57BL/6J,
C3H/HeJ,
4 or 11-12 weeks old,
n = 4-6/group
Toxicity increased with decreases in temperature. At 10 °C: decreased body
weight, total lung capacity, BALF protein, alkaline phosphatase activity,
% PMN, and lysozyme. Ozone-induced changes in lung volume were
attenuated during the 5 exposure days and returned to control levels after
7 days recovery. The responses to repeated O3 exposure in rats were
exacerbated by reduced ambient temperature, presumably as a result of
increased metabolic activity.
C57BL/6J mice: CO2-induced changes in VE were attenuated 1 h after O3
exposure; VT was reduced 1 h after O3 exposure; the diminished VT 1 h
after O3 was coincident with reduced/ mean inspiratory flow, and slope of
VE-to-%CO2 relationship compared with FA.; VE partially reversed 24 h after
O3 relative to FA.
C3H/HeJ: VT was reduced 1 h after O3 exposure; increased/to sustain the
hypercapnic VE response similar to air exposure.
Suggests that control of ventilation during response to CO2 is governed,
in part, by genetic factors in these two strains of mice, implying differential
O3 susceptibility.
VE and/were measured before and immediately after exposure. Normocapnic
VE was greater following subacute O3 exposure in C57BL/6J mice than in
C3H/HeJ mice, due to increased/and reduced VT, respectively. Ventilatory
responses to both normocapnia and hypercapnia were similar after O3
reexposure in both strains. Suggests that: increased VT in C57BL/6J mice
may contribute to the increased susceptibility to lung injury due to a greater
dose of O3 reaching the lower lung; mechanistic separation of airway
inflammation and ventilation.
Using 18O-labeled O3. C3H/HeJ mice had 46% less 18O in lungs and 61% less
in trachea, than C57BL/6J. C3H/HeJ mice had a greater body temperature
decrease following O3 exposure than C57BL/6J mice. Suggests that the
differences in susceptibility to O3 are due to differences the ability to decrease
body temperature and, consequently decrease the dose of O3 to the lung.
Wiester et al.
(1996)
Tankersley
etal. (1993)
Paquette
etal. (1994)
Slade et al.
(1997)
-------
Table AX5-11 (cont'd). Effects of Ozone on Pulmonary Function
Concentration
ppm Duration
Species
Effects
Reference
>
X
3 h, assayed 6 h
after exposure
Mice, male and
female, AKR/J,
C3H/HeJ, CBA/J,
129/J, NJ, C57BL/6J,
C3HeB/FeJ, SJL/J
Measured trachea! transepithelial potential in the six strains and in progeny of Takahashi
B6 and C3 strain mice. Fl mice and second generation backcrosses with the et al. (1995b)
resistant parent were O3 resistant. Ratios of 1:1 (resistant: susceptible) were
obtained with second generation backcrosses with the susceptible parent,
suggesting simple autosomal recessive inheritance of susceptibility.
Susceptible phenotype: 129/J, A/J, B6, C3HeB/FeJ, and SJL/J.
Resistant phenotype: AICR/J, C3, and CBA/J.
Different pattern of susceptibility than with inflammation, suggesting that
the responses are controlled by disparate genetic factors.
Mice, C3H/HeJ, A/J, Used whole body plethysmography and enhanced pause index (Penh) Savov et al.
C57BL/6J, 129/SvIm, evaluations. (2004)
CAST/Ei, BTBR,
DBA/2J, FVB/NJ, C57BL/6J, BALB/cJ, 129/Svlm, BTBR: were highly sensitive to O3;
B ALB/cJ, exhibited significant increases in Penh to MCh at 6 and 24 h after exposure
n = 6-24/group to O3.
DBA/2J, A/J, FVB/NJ, CAST/Ei, C3H/HeJ: increases in sensitivity to MCh
at 6 h after exposure, return to near baseline by 24 h after exposure to O3.
VE = Minute ventilation
VT = Tidal volume
/= Frequency of breathing
FA = Filtered air
MCh = Methacholine
-------
Table AX5-12. Effects of Ozone on Airway Responsiveness
>
X
Concentration
ppm
0.1
0.3
0.15
0.30
0.60
1.2
0.3
0.5
1
1
2
Exposure
Duration
4h/day,
4 days/week
for 24 weeks
4h
4 h/day for 1,
3, 6, 12, 24,
or 3 8 days
8 h/day for 5 days,
repeated every
14 days for
6 months
Ih
Ih
2h
Species, Sex, Strain,
and Age
Guinea pig, male and
female, Hartley,
200-250 g,
n = 10-20/group
Guinea pig, male
Hartley,
500-600 g,
n = 5-8/group
Guinea pig, male
Hartley,
500-600 g,
n = 6-7/group
Rhesus monkey,
male, 30 days old,
n = 6/group
Guinea pig, male,
Dunkin-Hartley ,
250-300 g,
n = 6-7/group
Mice, male,
C57BL/6,
6 weeks old,
n= 10-3 I/group
Rat, male, Fischer
F344,
14 months old,
n = 6/group
Observed Effect(s)
O3 exposure did not produce airway hyperresponsiveness to ACh in
nonsensitized animals; in OVA-sensitized animals, there was increased
responsiveness to both nonspecific (ACh) and specific (OVA) airway
challenge that persisted for 4 weeks after exposure 0. 1 and 0.3 ppm O3.
Effects were not gender specific and were not associated with BALF
inflammatory indicators, but were associated with antigen-specific
antibodies in blood.
Increased airway responsiveness to Hist, but not ACh, 16-18 h after
1.2 ppm O3 exposure only. Increased responsiveness to SP occurred after
exposure to > 0.3 ppm O3.
Increased airway responsiveness to SP occurred 16-18 h after exposure
to 0.3 ppm O3 for 1, 3, 6, 12, and 24 days; but not after 48 days. Highly
significant correlation between airway responsiveness and BALF total
cells, AMs, neutrophils, and eosinophils, suggesting that airway
inflammation is involved.
Increased airway responsiveness to Hist after 10 episodes of exposure to
O3 + HDMA in sensitized infant monkeys.
Increased bronchial responsiveness at 3 h, but not 24 h after O3; OVA had
no effect on baseline, but enhanced airway responsiveness 24 h after O3.
Ozone caused increased Cdyn and VE, and decreased PaO2 in OVA-
sensitized mice.
Increased airway responsiveness to MCh 2 h PE.
Reference
Schlesinger
etal.
(2002a,b)
Segura et al.
(1997)
Vargas et al.
(1998)
Schelegle
et al. (2003)
Sun et al.
(1997)
Yamauchi
et al. (2002)
Dye et al.
(1999)
-------
Table AX5-12 (cont'd). Effects of Ozone on Airway Responsiveness
Concentration Exposure
ppm Duration
Species, Sex, Strain,
and Age
Observed Effect(s)
Reference
0.3-3.0
3h
0.3
5h
3h
X
2h
Mice, AJ, male and
female,
aged 2, 4, 8, or
12 weeks,
n = 42-50/group
Mice, BALB/c,
3 weeks old,
n = 5/group
Mice, C57BL/6J, and
ob/ob,
8-12 weeks old,
n = 6-7/group
Rats, male and
female, SD,
2, 4, 6, 8, or 12
weeks,
n= 4-19/group
Guinea pigs, male,
Hartley, 400-600 g,
n = 6/group
Nose-only exposure plethysmographs. VE decreased with increasing age. Shore et al.
O3 caused concentration-related decrease in VE at all ages, but with less (2002)
response in the 2-weeks old. Younger mice with less decrease in
O3-induced VE demonstrated 3- to 4-times greater inhaled dose when
normalized for body weight. The 2- to 4-weeks mice showed no AHR at
any dose, while the 8- and 12-week old mice demonstrated dose-related
increases in AHR. Older mice demonstrated increased levels of IL-6 and
MIP-2. Suggests that young mice are less sensitive to O3 for the endpoints
of two cytokine and AHR.
Mice OVA-sensitized day 7-14; exposed day 21-23, assayed day 24 or
25 Decrease in Penh in rats exposed to O3, as a function of MCh
concentration.
Compared C57BL/6J and ob/ob mice ( strain obese due to defect in gene
coding for satiety hormone leptin). Intravenous MCh challenge induced
AHR and inflammation in both groups, but was greater in obese mice.
Dose per gram of lung tissue was greater in obese mice. Suggests obese
mice get greater dose of O3.
Nose-only-exposure plethysmographs exposure.
8,12 week rats: O3-induced 40-50% decreases in VE (primarily due
to decrease in VT).
6 week rats: Cyinduced changes in VE were significantly less.
2, 4 week rats: no O3-induced changes in VE.
BALF protein and PGE2 were greater than in older rats. Suggests higher
delivered dose to younger rats, decreased ventilatory response, and
greater lung injury.
AHR to MCh peaked 2 h PE; PMN in BALF increased until 6 h PE. Igarashi et al.
Tazanolast (a mast cell stabilizing drug, doses 30, 100, or 300 mg/kg) (1998)
administered before O3 exposure inhibited O3-induced AHR dose-
dependently. Suggests that mast cells may play in role in the
development of AHR.
Goldsmith
et al. (2002)
Shore et al.
(2003)
Shore et al.
(2000)
-------
Table AX5-12 (cont'd). Effects of Ozone on Airway Responsiveness
Concentration Exposure
ppm Duration
1 or 3 4 h, assayed 4 to
72 PE
2 2h
Species, Sex, Strain,
and Age
Mice,
normal WBB6F1
(1/1) and mast cell-
deficient WBB6F1-
kit W /kit W-v
(kit W /kit W-v)
8-12 weeks old,
n= 3-1 I/group
Cat, 2-3 kg,
n = 5/group
Observed Effect(s)
Demonstrated O3-induced cutaneous, as well as bronchial, mast cell
degranulation. PMN influx observed at 1 ppm only in normal mice.
AHR in response to MCh with 1 ppm observed in both. Suggests that
mast cells are involved in O3 -induced PMN influx but not AHR.
Cats anesthetized and mechanically ventilated, challenged with ACh.
Pretreated with polyethylene glycol-superoxide dismutase (PEG-SOD)
orPEG-catalase (PEG-CAT) 5 min before O3 exposure. PEG-SOD
partially prevented O3-induced AHR, PEG-CAT did not. Suggests
superoxide involvement in O3-induced AHR.
Reference
Noviski et al.
(1999)
Takahashi
etal. (1993)
0.75
X
oo
4 h, MCh challenge
6hPE
2h, assayed 2 hPE
Mice
FVB/N, and
FVB/N with
P2-AR transgene,
10-14 weeks old,
n = 10/group
Rat, male, SD,
2.5-3.5 months old,
n = 5/group
Targeted expression (using CCSP promoter) of p2-adrenergic receptors
(P2-AR) to airway epithelium to mimic agonist activation. Heterozygous
mice from generations 2 to 4 used. MCh challenge dose needed to increase
Penh was greater in CCSP-P2-AR mice. CCSP-P2-AR mice less responsive
to O3. Suggests that p2-ARs regulate airway responsiveness and that
P-agonists induce bronchodilation through activation of receptors on
smooth muscle cells and epithelial cells.
Neonatal rats treated with capsaicin. Challenged with MCh following O3.
Capsaicin-treated rats: O3 had no effect on pulmonary conductance;
decreased dynamic compliance; increase in AHR.
During O3 exposure: 50% decrease in HR and 2.5 °C decrease in core
temperature in both controls and capsaicin rats.
Suggests that C-fibers inhibit O3-induced AHR but do not modulate HR
or core temperature.
McGraw et al.
(2000)
Jimba et al.
(1995)
-------
Table AX5-12 (cont'd). Effects of Ozone on Airway Responsiveness
Concentration
ppm
1
Exposure
Duration
3 h, assayed 4 h PE
Species, Sex, Strain,
and Age
Rat, SD, treated
with capsaicin
or tachykinin
antagonists,
n = 6/group
Observed Effect(s)
Rats treated with CP-99994 (neurokinin-1 receptor antagonist) and
SR-48968 (neurokinin-2-receptor antagonist). O3 induced greater
numbers of PMN in B ALF of treated rats. The antagonists has no
effects on pulmonary mechanics or airway responsiveness.
Reference
Takebayashi
et al. (1998)
>
X
(Si
I
VO
3 h, ACh challenge
Oand24hPE
Mice, 8 weeks old,
15-25 g.
DBA/2J, AKR/J,
A/J, C3H/HeJ,
C57BL/6J, SJL/H,
129/J,
n = 6-8/group
Suggests that tachykinins are involved in the protective effects of C-fibers
against O3-induced inflammation.
Capsaicin treatment induced increased PMN in BALF. O3 exposure
reduced VE in both vehicle and capsaicin-treated rats, but the capsaicin
treatment caused a greater, more immediate reduction.
Suggests that the increase in BALF PMN is not due to a greater inhaled
dose of O3 reaching the lung.
Differing susceptibility to O3: Hyperreactive-DBA/2J, AKR/J, A/J;
Hyporeactive-C3H/HeJ, C57BL/6J, SJL/H; Intermediate 129/J.
ACh challenge 25 or 50 ug/kg.
ACh 24 h PE: Airway responses increased in A/J strain at 25 ug/kg and
in C57BL/6J and SJL/J strains at 50 ug/kg.
Ozone did not alter reactivity in other strains. MCh or carbachol challenge
not affected by O3, suggesting that cholinesterase function is affected by
O3. C57BL/6J and A/J mice treated with cyclophosphamide
(an immunosuppressant) or anti-PMN caused decreased Cyinduced
PMN levels but did not alter AHR, suggesting that O3-induced ACh
hyperreactivity correlates with susceptibility and that PMN influx
is independent of AHR.
Zhang et al.
(1995)
-------
Table AX5-12 (cont'd). Effects of Ozone on Airway Responsiveness
Concentration
ppm
Exposure
Duration
Species, Sex, Strain,
and Age
Observed Effect(s)
Reference
0.05
4 h, challenged with Rats, male, Long-
iv 5-HT
0.2
7 h, assayed 3 h PE
X
0.4
4 h, assayed by
ACh, SP, or
histamine challenge
0 or 48 h PE
Evans, SD, Fisher
344, Brown-Norway,
BDII, BDE, DA,
Lewis and Wistar,
6-8 weeks old,
n = 10/group
Rabbits, New
Zealand white, 5 kg,
n = 5-7/group
Rabbits, New
Zealand white, male
and female, 2.5-3 kg,
n = 4-6/group
AHR: developed in Lewis, BDII and Long-Evans rats 90 min after O3. Depuydt et al.
Baseline AHR differed among strains; did not correlate with O3-induced (1999)
AHR.
PMN influx: did not occur in any strain.
LE rats: AHR lasted > 12 h PE with no change in BALF PMN, LDH,
alkaline phosphatase, or protein.
Suggests that O3.induced AHR occurs without airway inflammation and
that genetic factors may alter the sensitivity to O3.
O3-induced decrease in tracheal transepithelial potential difference, but no Freed et al.
change in lung resistance. (1996)
ACh challenge: no change in compartmentalized lung resistance;
140% increase in O3-induced lung resistance.
Bilateral vagotomy: no change in compartmentalized lung resistance;
enhancement of O3-induced peripheral lung reactivity.
Suggests that O3 exposure may affect tracheal epithelial function and
increase central airway reactivity, possibly through vagally-mediated
mechanisms.
Used isolated perfused lung model allowing partitioning of the total Delaunois
pressure gradient into arterial, pre- and postcapillary, and venous et al. (1998)
components. O3-induced inhibition of pulmonary mechanical reactivity to
ACh, SP, and histamine. No change in baseline pulmonary resistance or
dynamic compliance. At 48 h PE O3 altered vasoreactivity of the vascular
bed.. Ozone-induced modification of the vasoreactivity of the vascular bed
at 48 h PE and elevation of arterial segmental pressure. Suggests that O3
can directly induce vascular constriction both immediately and 2 days
following exposure and can inhibit ACh-, SP-, and histamine-induced
changes in lung mechanics.
-------
Table AX5-12 (cont'd). Effects of Ozone on Airway Responsiveness
Concentration
ppm
0,0.12, 0.5, or
1.0
Exposure
Duration
6 h/day,
5 days/week for
20 months
Species, Sex, Strain,
and Age
Rat, male and
female,
Fischer 344,
6-7 weeks old
Observed Effect(s)
Isolated eighth generation airways following O3 exposure. Circumferential
tension development was measured in response to bethanechol,
acetylcholine, and electrical field stimulation and normalized to smooth
muscle area. 0.5 ppm caused an increase in smooth muscle area.
Maximum responses of the small bronchi of male rats were significantly
reduced after exposure to 0. 12 and 0.5 ppm O3. Suggests that O3 -induced
increases in airway responsiveness do not persist with near-lifetime
exposure and that chronic exposure alters smooth muscle cell function.
Reference
Szarek et al.
(1995)
0.5
8 h/day for 7 days
Guinea pig, male,
Hartley, 5 weeks old
X
-k
0.5
8 h/day for 5 days
followed by 9 days
of FA; for
11 episodes
Monkey, Rhesus,
30-d old,
n = 6/group
Repeated exposure increased rapidly adapting receptor activity to
substance P, methacholine, and hyperinflation; no significant effects on
baseline or substance P- and methacholine-induced changes in lung
compliance and resistance.
Suggest that because agonist-induced changes in receptor activity precede
lung function changes, the responsiveness of rapidly adapting receptors
was enhanced.
Half of the monkeys were sensitized to house dust mite allergen (HDMA)
at 14 and 28 days of age before exposure. HDMA and histamine aerosol
challenges administered until Raw doubled. Baseline Raw elevated after
10 exposure episodes in the HDMA + O3 group compared to the FA,
HDMA, and O3 exposure groups. Aerosol challenge with HDMA at the
end of the 10th episode did not significantly affect Raw, VT,/ or SaO2.
AHR appeared to develop following episode 6. Aerosol challenge with
HDMA at the end of the 10th episode did not significantly affect R^,, VT,/
or SaO2. HDMA + O3 group: had increases in serum IgG, histamine and
eosinophils; greater alteration in airway structure and content suggesting
that O3 can enhance the structural remodeling and allergic effects of
HDMA sensitization exposure groups.
Joad et al.
(1998)
Schelegle
et al. (2003)
-------
Table AX5-12 (cont'd). Effects of Ozone on Airway Responsiveness
Concentration
ppm
1
Exposure
Duration
8h
Species, Sex, Strain,
and Age
Rat, Wistar,
some treated
with capsaicin,
n = 6-8/group
Observed Effect(s)
Vehicle-treated: O3-induced rapid shallow breathing pattern; BrdU label
started at the bifurcation of the main stem bronchi and increased distally.
Capsaicin treated: no O3-induced changes in respiratory frequency;
Reference
Schelegle
etal. (2001)
2h
X
to
Rat, male, Wistar,
100-120 days old,
n = 7-10/group
reduced BrdU labeling density in the terminal bronchioles supplied by
short airway paths. Suggests that O3-induced rapid shallow breathing
is protective of conducting airways and allows distribution of injury to
more distal regions.
Examined the site-specific deposition of 18O at breathing frequencies of 80,
120, 16, or 200 bpm at a VT to produce a constant minute ventilation of
72.8 ml/mm/100 gbody weight.
All frequencies: parenchyma! areas had a lower content of 18O than trachea
and bronchi; right caudal parenchyma! levels did not change.
80 to 160 bpm: deposition reduced in midlevel trachea and increased in
both mainstream bronchi; increased deposition in parenchyma supplied by
short (cranial) airway paths.
200 bpm: increased deposition in trachea increased; increased deposition
in right cranial and caudal bronchi regions: decreased content in right
cranial parenchyma!.
Suggests that the effect of rapid, shallow breathing is to create a more
evenly distributed injury pattern with less deposition of O3) in the trachea
and a small effect on deposition in the parenchyma.
Alfaro et al.
(2004)
MCh = Methylcholine, ACh = Acetylcholine, Hist = Histamine, 5-HT = 5-Hydroxytryptamine, SP = Substance P, FS = Field Stimulation, CCh = Carbachol,
TX = Thromboxane, KC1 = Potassium Chloride, Pt = Platinum; Route: iv = intravenous, inh = inhalation., sc = subcutaneous, ip = intraperitoneal,
OVA = Ovalbumin, BALF = bronchoalveolar lavage fluid, HDMA = House Dust Mite Allergen, Cdyn = Dynamic Lung Compliance, VE = Minute Ventilation,
PaO2 = partial pressure of arterial oxygen, AHR = airway hyperreactivity, VT = Tidal Volume, Raw = airway resistance,/= frequency of breathing,
SaO2 = oxygen saturation of arterial blood.
-------
Table AX5-13. Effects of Ozone on Genotoxicity/Carcinogenicity
X
Concentration
ppm
1
Ior2
2
0.12,0.50,
and 1.0
0.5
0,0.12,0.5,or
1.0
Exposure
Duration
0, 12, 24, 48,
72, or 96 h
90min
90 min/day
for 5 days
6 h/day,
5 days/week for
up to 9 months
6 h/day,
5 days/week for
12 weeks
6 hr/day,
5 days/week,
to ppm 2-yr
and lifetime
Species,
Sex, Strain,
and Age
Guinea pigs,
Dunkin-Hartley ,
male, 2 mo-old,
n = 4/group
Mice, female,
BALB/c
(20.6 g) or
Muta™ (26.0 g)
n= 11 -21 /group
Mice, female,
A/I
n = 29-35/group
Mice, male and
female, B6C3Fl5
5-6 weeks old,
n = 20 M and 20
F
Rats, male and
female, Fischer
F-344/N,
n = 5M+5F
(acute)
n = 50m + 50
F/group
Observed Effect(s)
Following O3 exposure, two main bronchi were removed and tracheobronchial cells were isolated
and assayed for DNA strand breaks using fluormetric analysis. Ozone induced an increase in
BALF protein and in DNA strand breaks, but did not change cell yield or viability. The amount
of DNA in alkali ly sates was decreased at 72 h, which suggests an increase in strand breaks at
that time point.
O3-induced increase in strand breaks up to 200 min following exposure. No effects after 200 min.
O3 did not affect the level of oxidized amino acids in lung or the level of 8-oxo-deoxyguanosine
in nuclear DNA. O3-induced induction of IL-6 mRNA following DNA strand breaks, which does
not support inflammation causing DNA damage.
Mutagenic mice had no O3-induced mutations in ell transgene.
At 5 months, no difference in lung tumor multiplicity or incidence. At 9 months, no differences
in lung tumor multiplicity between control mice and mice exposed to any concentration of O3.
The highest, and only statistically significant lung tumor incidence, was found in the mice exposed
to 0.5 ppm O3. In the O3-exposed mice allowed to recover in filtered air, only the mice exposed to
0.12 ppm O3 had increases in lung tumor incidence and multiplicity. Authors consider the results
to be spurious and of no significance for data interpretation.
Sporadic differences in mean body weight between O3-exposed mice and air-exposed controls,
as well as significant differences in the mean absolute and relative weights of liver, spleen,
kidney, testes, and ovary. No O3-related increased incidence of neoplasms in lung tissue.
Oviductal carcinomas observed upon histopathologic examination in 30% of O3-exposed
female mice.
Cocarcinogenicity study with subcutaneous administration of 0, 0.1, or 1.0 mg/kg body weight
of 4-(N-nitrosomethylamino)-l-(3-pyridyl)-l-butanone (NNK) and inhalation of 0 or 0.5 ppm
O3 to male rats.
O3 caused dose-related increase in inflammation in CAR; increased fibrosis; extension of the
bronchiolar epithelium to the proximal alveoli; but no increase in neoplasms.
NNK (1.0 mg/kg) alone causes increased bronchiolar/alveolar neoplasms.
Reference
Femg et al.
(1997)
Bornholdt
et al. (2002)
Witschi et al.
(1999)
Kim et al.
(2001)
Boorman
etal. (1994)
NNK + O3: no enhancement of neoplasms.
Suggests that O3 is not carcinogenic; does not enhance neoplasms growth; and creates mild
site-specific lesions which persist with continued exposure.
-------
Table AX5-14. Systemic Effects of Ozone
X
Concentration
ppm Duration
NEUROBEHAVIORAL EFFECTS
0.1 4h
0.2
0.5
1.0
0.1 4h
0.4
0.7
1.1
1.5
0.3 30 days
0.6
Species
Rat, male, Wistar,
47-50 days old,
n = 25/group
Rat, male, Wistar,
300-350 g,
n = 10/group
Mice, CD-I
M, F, 28-33 g,
n = 6-7/group
Effects
Rats exposed for 4 h to 0.2, 0.5, and 1 ppm O3 showed long-term memory
deterioration and decreased motor activity, which was reversed 24 h later.
Brain and pulmonary Cu/Zn SOD levels were increased in animals exposed
to 0.1, 0.2, and 0.5 ppm O3, but decreased in animals exposed to 1 ppm O3.
O3 caused memory impairment at >0.7 ppm (one trial passive avoidance test),
decreased motor activity at > 1. 1 ppm, and increased lipid peroxidation at
>0.4 ppm. Lipid perioxidation levels from the frontal cortex, hippocampus,
striatum and cerebellum increased with increasing O3 concentration.
O3 exposure slightly but selectively affected neurobehavioral performance in
male mice assessed with a 5-min open-field test on exposure day 4 and 19 and
on day 3 after the end of the exposure. O3 exposure, however, did not grossly
Reference
Rivas-Arancibia
et al. (1998)
Dorado-Martinez
etal. (2001)
Sorace et al.
(2001)
affect neurobehavioral development. Reversal learning in the Morris water
maze test was consistently impaired in both prenatally and adult exposed
mice. In addition, longer latency to step-through in the first trial of the
passive avoidance test and a decrease in wall rearing in the hot-plate test were
recorded in O3 prenatally exposed mice. Except for the first open-field test,
altered responses were observed only in animals exposed to 0.3 ppm O3.
0.35 12 h
0.75
1.5
0.7 4h
0.7 4h
Rat, male, Wistar,
270 g,
n = 10/group
Rat, male, Wistar,
47-50 days old,
n = 6-10/group
Rat, male, Wistar,
27 months old,
n = 3-4/group
O3 exposure decreased paradoxical sleep after 2 h of exposure, and increased
slow wave sleep after 12 h of exposure at all O3 concentrations; 5-HT
concentrations in the pons increased with increasing O3 concentration.
Vitamin E administered before or after O3 exposure blocked memory
deterioration (passive avoidance)and increases in lipid peroxidation levels
in the striatum, hippocampus and frontal cortex that were associated with
oxidative stress.
O3 exposure increased ultrastructural alterations in the hippocampus and
prefrontal cortex in aged rats compared with controls. These areas are related
to learning and memory functions, which are the first degenerative aging
changes observed.
Paz and Huitron-
Resendiz (1996)
Guerrero et al.
(1999)
Nino-Cabrera
et al. (2002)
-------
Table AX5-14 (cont'd). Systemic Effects of Ozone
Concentration
ppm Duration Species
Effects
Reference
NEUROBEHAVIORAL EFFECTS (cont'd)
0.7
0.8
X
1.5
0.5
4h
12 h/day
during dark
period
4h
24 h
20 h/day
for 5 days
Rat, male, Wistar,
47, 540, or 900
days old,
n = 10-30/group
Rat, female, adult,
n = 6/group
Rat, male, Wistar,
n = 24/group
Rat, male, Wistar,
275 g,
n = 10/group
Rat, male, Wistar,
n = 1 I/group
Rat, SD,
220-240 g,
n = 10-20/group
Taurine (43 mg/kg) given before or after O3 exposure improved memory
deterioration in an age-specific manner. Old rats showed peroxidation in all
control groups and an improvement in memory with taurine. When taurine
was applied before O3, peroxidation levels were high in the frontal cortex
of old rats and the hippocampus of young rats; in the striatum, peroxidation
caused by O3 was blocked when taurine was applied either before or after
exposure.
O3 exposure during pregnancy affects the neural regulation of paradoxical
sleep and circadian rhythm of rat pups 30, 60, and 90 days after birth.
O3 caused alterations in long-term memory and a significant reduction of
dendritic spines. Results provide evidence that deterioration in memory is
probably due to the reduction in spine density in the pyramidal neurons of the
hippocampus.
O3 or its reaction products affect the metabolism of major neurotransmitter
systems as rapidly as after 1 h of exposure. There were significant
increases in dopamine (DA), and its metabolites noradrenaline (NA)
and 3,4 dihydroxyphenylacetic acid (DOPAC), and 5-hydroxyindolacetic
acid (5-HIAA) in the midbrain and the striatum.
Adult rats exposed to O3 spend decreased time in wakefulness and
paradoxical sleep and a significant increase in time in slow-wave sleep.
Neurochemical changes include increased metabolism of serotonin in
the medulla oblongata, pons, and midbrain.
O3 produced marked neural disturbances in structures involved in the
integration of chemosensory inputs, arousal, and motor control.
O3 inhibited tyrosine hydroxylase activity in noradrenergic brainstem cell
groups, including the locus ceruleus (-62%) and the caudal A2 subset (-57%).
Catecholamine turnover was decreased by O3 in the cortex (- 49%) and
striatum (-18%) but not in the hypothalamus.
Rivas-Arancibia
et al. (2000)
Haro and Paz
(1993)
Avila-Costa et al.
(1999)
Gonzalez-Pina
and Paz (1997)
Huitron-Resendiz
et al. (1994)
Cottet-Emard
et al. (1997)
-------
Table AX5-14 (cont'd). Systemic Effects of Ozone
Concentration
ppm Duration Species
Effects
Reference
NEUROBEHAVIORAL
0.4, 0.8, or 1.2 24 h
0.75, 1.5 and 3.0 4h
1-1.5
4h
X
Cat, male, adult,
n = 5/group
Rat, male, Wistar,
250-300 g,
n= 15/group
Rat, male, Wistar,
-250 g
n = 5/group
Evaluated EEG of sleep-wake organization in cats. 0.4 O3 did not change the Paz and Bazan-
amount of sleep parameters, did decrease paradoxical sleep during first 8 h Perkins (1992)
of exposure. At 1.2 ppm paradoxical sleep was reduced during O3 exposure,
followed by a dose-related increase of slow-wave sleep. Suggests O3-induced
changes in sleep patterns.
Recorded evoked potential in visual cortex and lateral geniculate nucleus. Custodio-
Pl, Nl and P2 components delayed in the visual cortex and lateral geniculate Ramierez and
nucleus at 3.0 ppm O3. Nl component in the visual cortex affected at 1.5 ppm Paz (1997)
O3. Suggest O3-induced alterations in synaptic excitability and conduction
mechanisms in the visual pathway.
O3-induced loss of dendritic spines on primary and secondary dendrites of Colin-Barenque
granule cells; swelling of Golgi apparatus and mitochondrion; dilation cisterns et al.
of the rough endoplasmic reticulum; vacuolation of neuronal cytoplasm. (1999)
Suggests O3-induced oxidative stress creates alterations in the granule layer
of the olfactory bulb and possible modifications of function.
NEUROENDOCRINE EFFECTS
0.5 to 3.0
1.0
3h
24 h
Rat, male, SD, Hyperthyroid, T4-treated rats (0.1-1.0 mg/kg/day for 7 days) had increased
44-47 days old, pulmonary injury (BALF LDH, albumin, PMNs) at 18 h PE compared to
n = 4-6/group control rats.
Rat, male, SD, Hyperthyroid, T3-treated rats had increased metabolic activity and O3-induced
3-4 months old pulmonary injury, but lipid peroxidation, as assessed by alkane generation,
was not affected.
Huffman et al.
(2001)
Sen etal. (1993)
-------
Table AX5-14 (cont'd). Systemic Effects of Ozone
X
Concentration
ppm Duration
CARDIOVASCULAR EFFECTS
0.1 5h
0.3
0.5
0.1 8h/day
0.3 for 4 days
0.5
0.25 to 2.0 2 h to 5 days
Species
Rat, Wistar
young (4-6
months) and old
(22-24 months)
n = 9-14/group
Rat, male, Wistar,
10 weeks old,
n = 9/group
Rat, Fischer F344,
Mice, C57BL/6J,
Effects
Transient rapid shallow breathing with slightly increased HR appeared
1-2 min after the start of O3 exposure, possibly due to olfactory sensation;
persistent rapid shallow breathing with a progressive decrease in HR occurred
with a latent period of 1-2 h. The last 90-min averaged values for relative
minute ventilation tended to decrease with the increase in the level of
exposure to O3 and these values for young rats were significantly lower than
those for old rats. An exposure of young rats to 0. 1 ppm O3 for shorter than
5 h significantly decreased the tidal volume and HR and increased breathing
frequency, but no significant changes were observed in old rats. There were
no differences between young and old rats in non-observable-adverse-effect-
levels (NOAELs) for the O3 -induced persistent ventilatory and HR responses,
when the NOAELs were determined by exposure to 0.3 and 0.5 ppm O3.
Circadian rhythms of HR and core body temperature were significantly
decreased on the first and second O3 exposure days in a concentration
dependent manner, and returned to control levels on the third and fourth days.
Robust and consistent decreases in HR and core body temperature; smaller
decreases in metabolism, minute ventilation, blood pressure, and cardiac
Reference
Aritoetal. (1997)
Iwasaki et al.
(1998)
Watkinson et al.
(2001)
0.5
6h/day
23 h/day
for 5 days
C3H/HeJ, Guinea
pig, Hartley,
n = 4-10/group
Rat, male, Fischer
F-344,
100-120 days old,
n = 4-6/group
output that vary inversely with ambient temperature and body mass.
Minimal extrapulmonary effects were observed at a core body temperature of
34 °C; O3 exposures at 22 and 10 °C produced significant decreases in heart
rate (160 and 210 beats/min, respectively), core body temperature (2.0 and
3.5 °C, respectively), and body weight (15 and 40 g, respectively).
Decreases in these functional parameters reached their maxima over the first
2 exposure days and returned to control levels after the 3rd day of exposure.
Watkinson et al.
(1995);
Highfill and
Watkinson (1996)
-------
Table AX5-14 (cont'd). Systemic Effects of Ozone
X
oo
Concentration
ppm
Duration
CARDIOVASCULAR EFFECTS
0.5
0
0.5
8h
24 h/day
6h/day
Species
(cont'd)
Rat, male, Fischer
F-344,
270-330 g,
n = 6/group
Rat, male, Fischer
F-344 kept at one
of three
Effects
O3 exposure increased atrial natriuretic peptides in the heart, lung, and
circulation, suggesting they mediate the decreased BP and pulmonary
edema observed with similar O3 exposures.
0.5 ppm O3 for both 6 h/day and 23 h/day caused decreases in heart rate and
core temperature (termed hypothermic response) and increases in B ALF
inflammatory markers. Exercise in 0 ppm O3 caused increases in heart rate
Reference
Vesely et al.
(1994a,b,c)
Watkinson et al.
(2003)
temperatures:
10 °C, 22 °C,
0.5 23 h/day 34 °C at rest,
moderate or heavy
0.5 CO2-stimulated
ventilation,
100-120 days old,
n = 4-8/group
REPRODUCTIVE AND DEVELOPMENTAL EFFECTS
and core temperature, 0.5 ppm O3 decreases. CO2 and O3 induced the
greatest deficits. Dose, animal mass, and environmental stress are suggested
to modify the hypothermic response.
0.2
0.4
0.6
0.3
0.6
0.9
Continuous up
to days 17 of
pregnancy
Continuous up
to postnatal day
26
Mice, male and
female, CD-I,
25-27 g
Mice, male and
female, CD-I,
27-30 g,
n= 11-15/group
No significant effects on either reproductive performance, postnatal somatic
and neurobehavioral development (as assessed by a Fox test battery) or adult
motor activity (including within-session habituation); some subtle or
borderline behavioral deficits were noted, however.
O3 caused subtle CNS effects but did not affect the animals' capability to
learn a reflexive response (limb withdrawal); females exposed to 0.6 ppm
O3 showed a reduced preference for the right paw than both their same-sex
controls and 0.6 ppm males. The effect was more robust in the case of an
organized avoidance response (wall-rearing).
Petruzzi et al.
(1995)
Petruzzi et al.
(1999)
-------
Table AX5-14 (cont'd). Systemic Effects of Ozone
X
Concentration
ppm
Duration
REPRODUCTIVE AND DEVEL
0.3
0.6
0.4
0.8
1.2
0.6
Continuous
until gestational
day 17
Continuous
during gestation
day 7-17
Continuous
from birth to
weaning
Species
Effects
Reference
OPMENTAL EFFECTS (cont'd)
Mice, male and
female, CD-I,
28-30 g,
n = 6-9/group
Mice, CD-I
Mice, male and
female, CD-I,
25-27 g,
n= 13-16/group
Exposure to O3 did not grossly affect neurobehavioral development,
as assessed by somatic and sensorimotor development (postnatal day
(PND) 2-20), homing performance (PND 12), motor activity (PND 21),
passive avoidance (PND 22-23), water maze performances (PND 70-74),
and response to a nociceptive stimulus (PND 100).
No effect of O3 on reproductive performance; no significant somatic
developmental effects in O3-exposed pups except for a delay in eye
opening that was not concentration dependent.
Exposure to O3 did not produce any significant impairment of the acquisition
phase during swimming navigation, a sensitive indicator for hippocampal
damage; however, O3 slightly increased the swimming paths during the last
day of the reversal phase. Mice exposed to O3 showed a slightly but
significantly higher swimming speed during all the days, which was unrelated
to differences in body weight and to navigational performances. Moreover,
mice exposed to O3 (with the exception of one animal) had a strong tendency
to make turns to the left while the controls, independent of sex, preferred
clockwise turns.
Sorace et al.
(2001)
Bignami et al.
(1994)
DeH'Onio et al.
(1995a,b)
12 h/day for
entire gestation,
assayed at
postnatal day 0,
12, and 60
Rat, male and
female, Wistar,
n = 9-10/group
EFFECTS ON LIVER, SPLEEN, THYMUS
3h
1.0
2.0
Rat, female, SD,
200-225 g
Histological and planimetric analysis using sagittal sections of the anterior
cerebellar lobe.
PND 0: O3-induced cerebellar necrosis.
PND 12: diminished molecular layer; pale nucleoli and perinucleolar
bodies in Purkinjie cells. PND 60: Purkinjie cells with clumps of chromatin
around periphery. Suggests that gestational exposure to O3 induces permanent
cerebellar damage.
High O3 exposure stimulates hepatocytes to produce increased amounts
of nitric oxide as well as protein, possibly mediated by cytokines such
as TNF-oc produced by alveolar macrophages. When macrophage function
is blocked, hepatic injury induced by O3 is prevented.
Rivas-Manzano
and Paz( 1999)
Laskin et al.
(1994, 1996,
1998b);
Laskin and
Laskin (2001)
-------
Table AX5-14 (cont'd). Systemic Effects of Ozone
Concentration
ppm
Duration
Species
Effects
Reference
>
X
(Si
I
o
EFFECTS ON LIVER, SPLEEN, THYMUS (cont'd)
2.0
2h
Rat, male, Fischer
F-344,
2, 9, or 24 months
old,
n = 2/group
EFFECTS ON CUTANEOUS TISSUE
0.5
0.8
1.0
10.0
1.0
5.0
10.0
2h
6h
2h
2h
0, 0.8, 1, and 10 2 h
Mice, hairless
female
Mice, SKH-1
hairless
Mice, SKH-1
hairless
Mice
Mice, SKH-1
hairless
Utilizing electron paramagnetic resonance (EPR) spectroscopy of chloroform Vincent et al.
extracts of liver homogenates, a significant flux of hydrogen peroxide (1996)
produced from the reaction of O3 with lipids of the extracellular lining could
be a source of biologically relevant amounts of hydroxyl radical. EPR signals
for carbon-centered alkoxyl and alkyl adducts were detected with C-phenyl
N-fer/-butyl nitrone (PEN) in the liver of animals exposed to O3.
a tocopherol levels in the stratum corneum (SC) were not affected by O3 Valacchi et al.
exposure (0.5 ppm) alone, but were significantly depleted by combined (2000)
exposure to UV and O3.
Increased lipid peroxidation in the skin epidermis and dermis activated stress Valacchi et al.
proteins HSP27 and HO-1, and activated a proteolytic enzyme system (2003)
(MMP-9) related to matrix injury and repair processes.
High O3 depletes hydrophilic antioxidants in the SC: vit. C decreased to 80%, Weber et al.
GSH decreased to 41%, and uric acid decreased to 44% of control levels after (2000)
exposure to > 1.0 ppm O3.
High O3 exerts an oxidizing effect on the outermost layer of the skin (SC); Weber et al.
depletes low-molecular-weight antioxidants (a tocopherol, vit. C, glutathione, (2001)
uric acid) in a concentration dependent manner; increases malondialdehyde
levels associated with lipid peroxidation.
1 ppm O3 depleted SC levels of vitamin C ( 80%), GSH (41%), and UA Weber et al.
(44%). Suggests that hydrophilic antioxidants in the SC modulate the effects (1999)
of O3-induced oxidative stress.
RER = Rough endoplasmic reticulum
PE = Postexposure (i.e., time after O3 exposure ceased)
TSH = Thyroid stimulating hormone
T3 = Triiodothyronine
T4 = Thyroxine
cyt. = Cytochrome
NADPH = Reduced nicotinamide adenine dinucleotide phosphate
NADH = Reduced nicotinamide adenine dinucleotide
B[a]P = Benzo[a]pyrene
NK = Natural killer
PHA = Phytohemagglutin
ConA = Concanavalin A
LPS = Lipopolysaccharide
SRBC = Sheep red blood cell
TEA = Thiobarbituric acid
IgE = Immunoglobulin E.
PHA = Phytohemagglutin
-------
Table AX5-15. Interactions of Ozone With Nitrogen Dioxide
Concentration
ppm
O3 NO2
MORPHOLOGY
0.8 14.4
0.3 1.2
Duration
6 h/day,
7 days/week
for 90 days
Continuous
for 3 days
Species
Rat, male, SD,
10-12 weeks
old,
n = 4/group
Rat, male, SD,
3 months old,
n = 4/group
Endpoints Interaction
Morphometry of lung Syngeristic; more peripheral centriacinar lesion, but same
parenchyma; DNA probes after 7, 78, and 90 days of exposure.
for procollagen; in situ
mRNA hybridization.
DNA single strand breaks; None; effect due to O3.
polyADPR synthetase of
AMs; total cells, protein,
andLDHinBALF.
Reference
Farman et al.
(1999)
Bermudez
etal. (1999);
Bermudez
(2001)
>
X
BIOCHEMISTRY
0.8 14.4 6 h/day,
7 days/week
for 9 weeks
0.4
7 90 days
Rat, male, SD,
10-12 weeks
old
n = 4/group
Rat, male, SD,
200-225 g
Lung hydroxyproline,
hydrooxypyridinium,
DNA, and protein content
of whole lung; morphology
and labeling index.
PMN, pulmonary edema,
fibrosis, MIP-2.
Synergistic; fibrosis after 7-8 weeks of exposure;
50% mortality at -10 weeks.
1-3 days: enhanced MIP-2, IL-lp, TNF-oc, thioredoxin,
and IL-6 expression; pulmonary edema and PMN influx,
which reversed by day 8; activation of NF-icB.
15-45 days: no tissue responses observed, suggesting
adaptation.
60 and 90 days: increased lung collagen; increased
expression of transforming growth factor-13 and TNF-oc,
activation of NF-icB.
Farman et al.
(1997)
Ishii et al.
(2000b)
MnSOD and GPX not altered during exposure.
Suggests that cytokines play a role in the early responses
to combined O3 and NO2 and that pulmonary fibrosis is
more dependent on concentration than on cumulative dose.
-------
Table AX5-15 (cont'd). Interactions of Ozone With Nitrogen Dioxide
Concentration
ppm
NO2 Duration Species
Endpoints
Interaction
Reference
>
X
(Si
I
to
BIOCHEMISTRY (cont'd)
0.8 14.4 1,5, and
8 weeks
Rat, male, SD,
10-12 weeks
old,
n = 4/group
Immunohistochemisty and
morphometric analysis of
TNF-cc and MnSOD levels
in alveolar ducts.
Triphasic response. 1-3 weeks: initial inflammation;
TNF-cc increased in proximal area.
4-5 weeks: partial resolution.
6-8 weeks: rapidly progressive fibrosis; elevated MnSOD;
TNF-cc increased in proximal area.
TNF-cc increased in interstitial cells at all time points.
Suggests O3-induced increases in MnSOD in areas of
injury and in more protected areas; O3-induced increases
in TNF-cc correlate spatiotemporally with injury.
Weller et al.
(2000)
BAL = Bronchoalveolar lavage.
PG = Prostaglandin.
G-6-PD = Glucose-6-phosphate dehydrogenase.
GOT = GSH-disulfide transhydrogenase.
GSHPX = GSH peroxidase.
SOD = Superoxide dismutase.
DR = Disulfide reductase.
NADPH-CR = Reduced nicotinamide adenine dinucleotide phosphate-cytochrome c reductase.
GSH = Glutathione.
6-PG-D = 6-phosphogluconate dehydrogenase.
-------
Table AX5-16. Interactions of Ozone with Formaldehyde
03
0.4
HCHO
3.6
Duration
8 h/day for
3 days
Species
Rat, male,
Wistar,
n = 20/group
Endpoints
O3 or HCHO: no changes in formaldehyde dehydrogenase,
glutathione S-transferase, glutathione reductase, and glucose-6-
phosphate dehydrogenase activities.
Interaction
Proliferative
response:
No interactive
effects
Reference
Cassee and
Feron(1994)
>
X
0.6 10 3 h, with
exercise at
2* resting
ventilation
0.2-10 0.4-4 30 min
O3 + HCHO: slightly decreased enzyme activities
HCHO alone: rhinitis, necrosis, degeneration, hyperplasia and
squamous metaplasia of nasal respiratory epithelium.
O3: PMN influx; disarrangement, flattening and slight basal cell
hyperplasia of the nonciliated cuboidal epithelium. Proliferating
cell nuclear antigen levels elevated.
Rat, male, HCHO did not alter O3-induced changes in breathing pattern.
SD, 7 weeks Parenchyma! injury attributed to O3 alone.
Mice, Continuously measured/ VT, expiratory flow, T;, Te, and
BALB/c, respiratory patterns during acute exposures. HCHO: appeared to
n = 4 be a pure sensory irritant at lower concentrations. O3: induced
non-dose-dependent transient increase in rapid, shallow breathing.
No effect level of HCHO was 0.3 ppm, for O3 1.0 ppm. Suggests
that O3 and HCHO have the same respiratory effects in BALB/c
mice and humans, with similar sensitivities.
Additive effects
in transitional
epithelium
and trachea
Not additive
Mautz (2003)
Nielson et al.
(1999)
-------
Table AX5-17. Interactions of Ozone with Tobacco Smoke
>
X
03
1.5
HCHO
7 ml breaths at
33% challenge
Duration
Ih
Species
Guinea pig,
male, Hartley,
320-460 g,
n = 5 to 9
Endpoints
Pulmonary resistance and dynamic lung compliance were compared
pre- and post-O3 exposure.
Pre-O3: cigarette smoke induced bronchoconstriction after 1 min.
Interaction
Synergistic
Reference
Wu et al.
(1997)
0.5
ADSS
30 mg/m3
6 h/day
for 3 days
Mice, male,
B6C3F1,
10 weeks old,
n=ll
24 h
Post-O3: cigarette smoke induced bronchoconstriction more quickly
and for a longer period.
Selective antagonism of neurokinin 1 and 2 receptors blocked then
enhanced O3 -induced bronchoconstriction, suggesting that endogenous
tachykinins modulate O3 -induced bronchoconstriction.
Aged and diluted side stream cigarette smoke (ADSS) exposure
followed by O3.
ADSS + O3: greater increase in BALF cells, %PMN, and proteins
compared to alone; 402% increase in proliferating cells in CAR;
LPS-stimulated release of IL-6 decreased; LPS-stimulated release
of TNF-ct increased.
Synergism
Yu et al.
(2002)
ADSS alone: no change in number of proliferating cells in CAR;
LPS-stimulated release of TNF-cc increased.
O3 alone: 280% increase in proliferating cells in CAR;
LPS-stimulated release of IL-6 decreased.
Suggests that O3-induced lung injury is enhanced by prior ADSS
exposure.
-------
Table AX5-18. Interactions Of Ozone With Particles
>
X
Concentration
03
(ppm)
PM
(mg/m3) Duration
Species
Endpoints
Interaction
Reference
SULFURIC ACID
0.1
0.2
0.1
0.3
0.6
0.1
0.3
0.6
0.6
PARTICI
0.1
0.16
0.30
0.59
0.02-0.15 23. 5 h/day CM-
CO. 4-0.8 urn) intermittent 12 h/day
for up to 90 days
0.50 (0.3 urn) 3 h
0.125(0.3 urn)
0.50 (0.3 urn) 3 h
0.125(0.3 urn)
0.5 (0.06 and 4 h/day for 2 days
0.3 umMMD)
,E MIXTURES
Diesel PM (MIST 24 h (IT)
#2975) reacted with
O3 for 48 h
0.05-0.22 mg/m3 4 h/day, 3 days/week
ammonium bisulfate for 4 weeks
0.03-0. 10 mg/m3 C
0.1 1-0.39 pmNO2
0.02-0. 11 mg/m3 HNO3.
(O.SumMMAD)
Rat, male, SD,
250-275 g,
n = 6/group
Rabbit NZW
male, 3.5-4.5 kg,
n = 5/group
Rabbit NZW
male, 3.5-4.5 kg,
n = 5/group
Rat, male, SD,
250-300 g,
n = 10/group
Rat, male, SD,
250-300 g,
n = 4-13
Rat, male,
Fischer F344N,
1 1 weeks old,
n = 8-30
Morphology
Biochemistry
AM intracellular
pH homeostasis and
H+ extrusion
Airway responsiveness
(in vitro bronchial
rings + ACh)
Morphology: volume
percentage of total
parenchyma containing injured
alveolar septae;
bromodeoxyuridine cell
labeling index in the periacinar
region
Inflammation
breathing pattern, morphology,
lavagable protein, and
clearance
No interaction
Antagonism
Antagonism
Synergism: ultrafine + O3, but
not fine
Synergism: fine + O3
Synergism
Complex interactions, but
possible loss of typical
attenuation seen with O3 only
exposure, reflecting
persistence of inflammation.
Last and
Pinkerton
(1997)
Chen et al.
(1995)
El-Fawal
etal. (1995)
Kimmel et al.
(1997)
Madden et al.
(2000)
Mautz et al.
(2001)
-------
Table AX5-18 (cont'd). Interactions Of Ozone With Particles
>
X
Concentration
O3 PM
(ppm) (mg/m3)
PARTICLE MIXTURES (cont'd)
0.2 0.07 and 0.14 mg/m3
ammonium bisulfate
(0.45 urn MMMD);
0.05 and 0.10 mg/m3
carbon
0.2 0.50 mg/m3 ammonium
bisulfate (0.45 um
MMMD) and elemental
carbon)
0.3 0.063 to 1.57 mg/m3
CAPs (Boston) + ip
OVA sensitization
0.4 0.20 and 0.50 mg/m3
fine, H2O2-coated
carbon (0.26 um
MMMD)
0.5 Endotoxin (IN)
100 ug 24 hand
48 h after the 3rd
O3 exposure
0.5 OVA (IN)
50 ul (1%)
Duration
4 h/day, 3 days/week
for 4 weeks
4 h/day, 3 days/week
for 4 weeks; nose only
Sensitized at day 7
and 14. Challenged at
day 2 1,22, and 23.
5 h O3 exposure
4 h/day for 1 or 5 days
8 h/day for 3 days
8 h/day for 1 day or
3 consecutive days
Species
Rat, male,
Fischer F344,
22-24 months
old,
n = 5-10
Rat, Fischer
F344N-NIA,
22-24 months old
Mice, BALB/c,
n = 5 -6/group
Rat, SD,
300 g,
n=10
Rat, Fischer
F344,
10-12 weeks old,
n = 6/group
Rat, Brown,
Norway,
10-12 weeks old,
n = 6/group
Endpoints
BAL protein and albumin;
plasma hydroxylase and
fibronectin
DNA labeling of dividing lung
epithelial and interstitial cells
by 5-bromo-2-deoxyuridine
Airway function
Inflammation
Nasal morphology
Nasal morphology
Interaction
Questionable interaction.
No changes in BALF protein
or prolyl 4-hydroxylase in
blood. Small decease in
plasma fibronectin with
combined exposure.
Synergism. Increased AM
phagocytosis and respiratory
burst. Decreased lung
collagen.
Interaction: increased RL
and airway responsiveness in
normal and OVA-sensitized
mice.
Synergism for effect on day 5.
Greater response at high dose,
contrasts with O3 along where
inflammation was greatest at
0.4 ppm on day 1.
Synergism: increased
intraepithelial mucosubstances
and mucous cell metaplasia
Synergism: increased
intraepithelial mucosubstances
and mucous all metaplasia.
Reference
Bolarin et al.
(1997)
Kleinman
et al. (2000)
Kobzik et al.
(2001)
Kleinman
etal. (1999)
Fanucchi
etal. (1998)
Wagner et al.
(2001a,b)
Wagner et al.
(2002)
-------
Table AX5-18 (cont'd). Interactions Of Ozone With Particles
Concentration
O3 PM
(ppm) (mg/m3)\
Duration
Species
Endpoints
>
X
Interaction
Reference
PARTICLE MIXTURES (cont'd)
0.8 0.5 mg, 1.5 mg, or 2, 4, and 7 days after
5 mg of PM from IT instillation
Ottawa Canada
(EHC-93)
1 0.11 mg/m3 ultra fine 6 h
carbon (25 nm CMD)
+ endotoxin (IH)
1 Endotoxin (37.5 EU) 4, 20, or 24 h
for 10 min
1 Endotoxin (IN) 0, 2, 8 h, repeated after
or 20 ug in 120 uL 24 h
Rat, male, Inflammation
Wistar,
200-250 g,
n = 5/group
Rat, male, Inflammation
Fischer F3 44,
10 weeks and
22 months old,
n = 3/group
Mice, male,
TSK,
14-17 months
old,
n = 6-7/group
Mice, Inflammation
C57BL/6J, 36 h
old, 8 weeks old,
n = 12/group
Rat, male, Lung morphometric analysis
Fischer F344, and inflammation
10-12 weeks old,
n = 6/group
Interaction: increased TNF-cc
inBALF.
Interaction: increased
PMNs and ROS release from
BALF cells for old rats and
mice primed with endotoxin;
depressed in young rats.
Synergism: increased BALF
protein and PMNs.
Synergism: increased
BALF PMNs and mucin
glycoprotein; increased
intraepithelial
mucosubstances and
mucingene mRNA.
Ulrich et al.
(2002)
Elder et al.
(2000a,b)
Johnston
et al. (2000b,
2002)
Wagner et al.
(2003)
-------
Table AX5-18 (cont'd). Interactions Of Ozone With Particles
Concentration
03
(ppm)
PM
(mg/m3)
Duration Species
Endpoints
Interaction Reference
PARTICLE MIXTURES (cont'd)
0.8 EHC-93,5 mg/m3 or
50 mg/m3
4 h exposure, clean air
for 32 h, 3H injection
followed by assay at
90min
>
X
(Si
I
oo
1.8 (EHC93) at 50 mg/m3
4 h, exposure, clean
air for 32 h, 3H
injection followed by
assay at 90 min
Rat, male, Control rats: 3H labeling
Fischer F344, (indicative of proliferation)
200-250 g, low in bronchioles and
n = 6/group parenchyma.
ECH-93 alone: no induction
of proliferation.
O3 alone: increased cell
labeling in bronchioles and
parenchyma, suggestive of
reparative cell proliferation.
ECH-93 + O3: both doses
of EHC-93 potentiated
proliferation, especially
in epithelia of terminal
bronchioles and alveolar
ducts, but not distal
parenchyma. Suggests that
ambient PM can enhance
O3-induced proliferations
and exacerbate injury.
Rat, male, O3 +EHC-93: epithelial cell
Fischer F344, injury and proliferation
200-250 g, (3H-labeling) higher than
n = 4/group single exposures; higher in
periductal region than in
whole lung counts; greater
numbers of AMs and PMNs in
lung tissue compartment than
in single exposures. Suggests
that exposures to urban PM
have few effects alone, but can
potentiate O3-induced injury.
Synergism of proliferation
Vincent et al.
(1997)
Synergism
Adamson
etal. (1999)
-------
Table AX5-18 (cont'd). Interactions Of Ozone With Particles
Concentration
03
(ppm)
PM
(mg/m3)
Duration Species
Endpoints
Interaction Reference
PARTICLE MIXTURES (cont'd)
0.15
HNO350ug/m3
>
X
0.5
Carbon Black (CB),
0.5 or 1.5 mg/rat
4 h/day, 3 days/week
for 40 weeks,
nose-only exposure
Intratracheal CB
followed by 7 days
or 2 months of O3
Rat, male,
Fischer F344/N, 8
weeks old,
n = 4-5/group
Rat, male, Wistar,
7 weeks old,
n = 7-8/group
O3 alone: 28% increase in Synergism
lung putrescine.
HNO3 alone: 21% decrease in
lung putrescine
O3andO3+HNO3:
56% increase in lung
putrescine
Pulmonary spermidine and
spermine did not change with
any exposure.
Suggests role of putrescine in
regulation of inflammation.
Phagocytotic capacity: Synergism
decreased in CB-exposed
group, unchanged in O3 group.
Formation of superoxide anion
radicals and numbers of
ingested particles increased at
2 months in O3 group.
Chemotactic migration:
decreased in CB-treated group.
Suggests that CB impairs
phagocytosis and chemotactic
migration in AMs, whereas O3
stimulates these functions.
Sindhu et al.
(1998)
Creutzenberg
etal. (1995)
-------
Table AX5-18 (cont'd). Interactions Of Ozone With Particles
Concentration
O,
PM
(ppm) (mg/m3)
Duration
Species
Endpoints
Interaction
Reference
PARTICLE MIXTURES (cont'd)
0.018
CH2O
TSP
PM10
PM,,
3.3 ppb
0.068 mg/m3
0.032 mg/m3
0.016 mg/m3
23 h/day for 7 weeks
Rat, male and
female, Fischer
F344,
9 weeks old,
n = 5/group
Exposure to filtered and
unfiltered Mexico City air.
Histopathology revealed no
nasal lesions in exposed or
control rats; tracheal and lung
tissue from both groups
showed similar levels of
minor abnormalities.
Moss et al.
(2001)
X
(Si
I
ON
O
VMD = Volume median diameter
og = Geometric standard deviation
MMAD = Mass median aerodynamic diameter
CMD = Count median diameter
PM = Particulate Matter
OVA= Ovalbumin
RL = Total Pulmonary Resistance
AED = Aerodynamic diameter
BAL = Bronchoalveolar lavage
AM = Alveolar macrophage
IN = Intranasal
IT = Intratracheal
Unless indicated otherwise, whole-body exposures used
-------
Table AX5-19. Effects of Other Photochemical Oxidants
X
Concentration
ppm Duration
H2O2(10, 20, or 2h,
100 ppb) and/or assayed 0
(NH4)2S04 or24hPE
(429or215Ag/m3;
0.3-0.4 gmmass
median diameter)
PAN
0.13,0.66,1.31 3h
0.077,0.192,0.384 Ih
0, 15, 39, or 78 ppm Ih
3 ppm PAN 1 month
-300 ppb 4, 7, 10, or
12h
78 ppm 24 h
-80 ppm 3H-PAN 1 h
Species
Rat,
female, SD,
n = 4- 1 8/group
PEL from
mice, male,
CD- 1 , 5 weeks
old, n = 25
Mice, male,
B6C3F1,
6 weeks old,
n = 4
Hamster,
Chinese,
n = 4
Salmonella
TA100
Mice, Big
Blue®, male,
4 weeks old,
n=15
Effects
Exposures alone or in combination.
H2O2 + (NH4)2S04: no effect on BALF cell number, viability, protein or LDH at either
time point; increased PMNs adhered to vascular epithelium; increased TNF-a by AMs at
both time points; at 0 h PE, transient increase in superoxide anion by AMs; decreased NO
production; nitrotyrosine detected; heme oxygenase-1 expression upregulated in AMs;
H2O2 only: decreased NO production which persisted for 24 h.
Suggests that PM-induced effects are augmented by H2O2 and that PM-induced tissue
injury may be modulated by cytotoxic mediators produced by AMs.
In vitro: Exposed peripheral blood lympocytes (PEL) to PAN; assayed for chromosome
aberrations, sister chromatid exchanges and DNA damage. DNA damage was observed
at cytotoxic concentrations of PAN. No effects at noncytotoxic exposures.
In vivo: Nose-only exposures. No dose-related effects at any exposure level in any
assay.
Suggests that PAN, both in vitro and in vivo, is not a DNA damaging agent or clastogen.
Measured frequency of thioguanine-resistant lung fibroblasts and frequency of
micronuclei in either the bone marrow or the lungs.
Mutation frequencies not altered from control levels. No chromosome breakage was
observed in bone marrow or lung. Suggests no mutagenicity in vivo.
Gas phase exposure PAN-mduced mutants: 59% GC -> TA; 29% GC -> AT;
2% GC -> CG; 10% multiple mutations — primarily GG -> TT tandem-base
substitutions.
Nose-only exposure. Mutagenic at the lacl gene in the lung; no tandem-base mutations.
Nasal tissue: 3.9% of the radiolabel.
Lung: 0.3 % of radiolabel.
Suggests that PAN is weakly mutagenic in Salmonella (with a signature GG -> TT
transversion) and in mouse lung.
Reference
Morio et al.
(2001)
Kligerman
etal. (1995)
Heddle et al.
(1993)
DeMarini
et al. (2000)
-------
REFERENCES
Adamson, I. Y. R.; Vincent, R.; Bjarnason, S. G. (1999) Cell injury and interstitial inflammation in rat lung after
inhalation of ozone and urban particulates. Am. J. Respir. Cell Mol. Biol. 20: 1067-1072.
Alfaro, M. F.; Putney, L.; Tarkington, B. K.; Hatch, G. E.; Hyde, D. M; Schelegle, E. S. (2004) Effect of rapid
shallow breathing on the distribution of 18O-labeled ozone reaction product in the respiratory tract of the rat.
Inhalation Toxicol. 16: 77-85.
Arito, H.; Takahashi, M.; Iwasaki, T.; Uchiyama, I. (1997) Age-related changes in ventilatory and heart rate
responses to acute ozone exposure in the conscious rat. Ind. Health 35: 78-86.
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.
Avila-Costa, M. R.; Colin-Barenque, L.; Fortoul, T. L; Machado-Salas, J. P.; 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.
Ballinger, C. A.; Cueto, R.; Squadrito, G.; Coffin, J. F.; Velsor, L. W.; Pryor, W. A.; Postlethwait, E. M. (2005)
Antioxidant-mediated augmentation of ozone-induced membrane oxidation. Free Radical Biol. Med.
38: 515-526.
Bassett, D.; Elbon-Copp, C.; Otterbein, S.; Barraclough-Mitchell, H.; DeLorme, M.; Yang, H. (2001) Inflammatory
cell availability affects ozone-induced lung damage. J. Toxicol. Environ. Health A 64: 547-565.
Becker, S.; Quay, J.; Koren, H. S. (1991) Effect of ozone on immunoglobulin production by humanB cells in vitro.
J. Toxicol. Environ. Health 34: 353-366.
Bermudez, E. (2001) Detection of poly(ADP-ribose) synthetase activity in alveolar macrophages of rats exposed to
nitrogen dioxide and ozone. Inhalation Toxicol. 13: 69-84.
Bermudez, E.; Ferng, S. F.; Castro, C. E.; Mustafa, M. G. (1999) DNA strand breaks caused by exposure to ozone
and nitrogen dioxide. Environ. Res. 81: 72-80.
Bhalla, D. K. (1996) Alteration of alveolar macrophage chemotaxis, cell adhesion, and cell adhesion molecules
following ozone exposure of rats. J. Cell. Physiol. 169: 429-438.
Bhalla, D. K.; Gupta, S. K. (2000) Lung injury, inflammation, and inflammatory stimuli in rats exposed to ozone.
J. Toxicol. Environ. Health 59: 211-228.
Bhalla, D. K.; Hoffman, L. (1997) Time course of airway epithelial and inflammatory changes in rats exposed to
moderate levels of ozone. Inhalation Toxicol. 9: 829-842.
Bhalla, D. K.; Gupta, S. K.; Reinhart, P. G. (1999) Alteration of epithelial integrity, alkaline phosphatase activity,
and fibronectin expression in lungs of rats exposed to ozone. J. Toxicol. Environ. Health A 56: 329-343.
Bhalla, D. K.; Reinhart, P. G.; Bai, C.; Gupta, S. K. (2002) Amelioration of ozone-induced lung injury by anti-tumor
necrosis factor-a. Toxicol. Sci. 69: 400-408.
Bignami, G.; Musi, B.; DeH'Omo, G.; Laviola, G.; Alleva, E. (1994) Limited effects of ozone exposure during
pregnancy on physical and neurobehavioral development of CD-I mice. Toxicol. Appl. Pharmacol.
129: 264-271.
Bolarin, D. M.; Bhalla, D. K.; Kleinman, M. T. (1997) Effects of repeated exposures of geriatric rats to ozone and
particle-containing atmospheres: an analysis of bronchoalveolar lavage and plasma proteins. Inhalation
Toxicol. 9: 423-434.
Boorman, G. A.; Hailey, R.; Grumbein, S.; Chou, B. J.; Herbert, R. A.; Goehl, T.; Mellick, P. W.; Roycroft, J. H.;
Haseman, J. K.; Sills, R. (1994) Toxicology and carcinogenesis studies of ozone and ozone
4-(A^-nitrosomethylamino)-l-(3-pyridyl)-l-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 DNA
strand breaks and inflammation in mice. Mutat. Res. 520: 63-71.
Bridges, J. P.; Davis, H. W.; Demodarasamy, M.; Kuroki, Y.; Howies, G.; Hui, D. Y.; McCormack, F. X. (2000)
Pulmonary surfactant proteins A and D are potent endogenous inhibitors of lipid peroxidation and oxidative
cellular injury. J. Biol. Chem. 275: 38848-38855.
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. Inhalation Toxicol.
15: 1209-1230.
Cassee, F. R.; Feron, V. J. (1994) Biochemical and histopathological changes in nasal epithelium of rats after 3-day
intermittent exposure to formaldehyde and ozone alone or in combination. Toxicol. Lett. 72: 257-268.
AX5-62
-------
Catalano, P. I; Rogus, J.; Ryan, L. M. (1995a) Consequences of prolonged inhalation of ozone onF344/Nrats:
collaborative studies. Part X: robust composite scores based on median polish analysis. Cambridge, MA:
Health Effects Institute; research report no. 65.
Catalano, P. J.; Chang, L.-Y. L.; Harkema, J. R.; Kaden, D. A.; Last, J. A.; Mellick, P. W.; Parks, W. C.; Pinkerton,
K. E.; Radhakrishnamurthy, B.; Ryan, L. M.; Szarek, J. L. (1995b) Consequences of prolonged inhalation of
ozone on F344/N rats: collaborative studies. Part XI: Integrative summary. Cambridge, MA: Health Effects
Institute; research report no. 65.
Chang, L.-Y.; Stockstill, B. L.; Menache, M. G.; Mercer, R. R.; Crapo, J. D. (1995) Consequences of prolonged
inhalation of ozone onF344/N rats: collaborative studies. Part VIII: morphometric analysis of structural
alterations in alveolar regions. Cambridge, MA: Health Effects Institute; pp. 3-39; research report no. 65.
Chang, M. M.-J.; Wu, R.; Plopper, C. G.; Hyde, D. M. (1998) IL-8 is one of the major chemokines produced by
monkey airway epithelium after ozone-induced injury. Am. J. Physiol. 275: L524-L532.
Cheek, J. M.; McDonald, R. J.; Rapalyea, L.; Tarkington, B. K.; Hyde, D. M. (1995) Neutrophils enhance removal
of ozone-injured alveolar epithelial cells in vitro. Am. J. Physiol. 269: L527-L535.
Chen, L. C.; Qu, Q.; Amdur, M. O.; Schlesinger, R. B. (1995) Alteration of pulmonary macrophage intracellular pH
following inhalation exposure to sulfuric acid/ozone mixtures. Exp. Lung Res. 21: 113-128.
Chen, C.-Y.; Bonham, A. C.; Plopper, C. G.; Joad, J. P. (2003) Plasticity in respiratory motor control: selected
contribution: neuroplasticity in nucleus tractus solitarius neurons following episodic ozone exposure in infant
primates. J. Appl. Physiol. 94: 819-827.
Cheng, P.-W.; Boat, T. F.; Shaikh, S.; Wang, O.-L.; Hu, P.-C.; Costa, D. L. (1995) Differential effects of ozone on
lung epithelial lining fluid volume and protein content. Exp. Lung Res. 21: 351-365.
Cho, H. Y.; Hotchkiss, J. A.; Harkema, J. R. (1999a) 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, H. Y.; Hotchkiss, J. A.; Bennett, C. B.; Harkema, J. R. (1999b) Effects of pre-existing rhinitis on
ozone-induced mucous cell metaplasia in rat nasal epithelium. Toxicol. Appl. Pharmacol. 158: 92-102.
Cho, H. Y.; Hotchkiss, J. A.; Bennett, C. B.; Harkema, J. R. (2000) Neutrophil-dependent and
neutrophil-independent alterations in the nasal epithelium of ozone-exposed rats. Am. J. Respir. Crit. Care
Med. 162: 629-636.
Cho, H.-Y.; Zhang, L.-Y.; Kleeberger, S. R. (2001) Ozone-induced lung inflammation and hyperreactivity are
mediated via tumor necrosis factor-a receptors. Am. J. Physiol. 280: L537-L546.
Churg, A.; Brauer, M.; Keeling, B. (1996) Ozone enhances the uptake of mineral particles by tracheobronchial
epithelial cells in organ culture. Am. J. Respir. Crit. Care Med. 153: 1230-1233.
Cohen, M. D.; Zelikoff, J. T.; Qu, Q.; Schlesinger, R. B. (1996) Effects of ozone upon macrophage-interferon
interactions. Toxicology 114: 243-252.
Cohen, M. D.; Zelikoff, J. T.; Chen, L.-C.; Schlesinger, R. B. (1998) Immunotoxicologic effects of inhaled
chromium: role of particle solubility and co-exposure to ozone. Toxicol. Appl. Pharmacol. 152: 30-40.
Cohen, M. D.; Sisco, M.; Li, Y.; Zelikoff, J. T.; Schlesinger, R. B. (2001) Ozone-induced modulation of
cell-mediated immune responses in the lungs. Toxicol. Appl. Pharmacol. 171: 71-84.
Cohen, M. D.; Sisco, M.; Baker, K.; Li, Y.; Lawrence, D.; VanLoveren, H.; Zelikoff, J. T.; Schlesinger, R. B.
(2002) Effects of inhaled ozone on pulmonary immune cells critical to antibacterial responses in situ.
Inhalation Toxicol. 14: 599-619.
Colin-Barenque, L.; Avila-Costa, M. R.; Fortoul, T.; Rugerio-Vargas, C.; Machado-Salas, J. P.;
Espinosa-Villanueva, J.; Rivas-Arancibia, S. (1999) Morphologic alteration of the olfactory bulb after acute
ozone exposure in rats. Neurosci. Lett. 274: 1-4.
Connor, L. M.; Ballinger, C. A.; Albrecht, T. B.; Postlethwait, E. M. (2004) Interfacial phospholipids inhibit ozone
reactive absorption-mediated cytotoxicity in vitro. Am. J. Physiol.: 10.1152/ajplung.00397.2003.
Cottet-Emard, J.-M.; Dalmaz, Y.; Pequignot, J.; Peyrin, L.; Pequignot, J.-M. (1997) Long-term exposure to ozone
alters peripheral and central catecholamine activity in rats. Pfluegers Arch. 433: 744-749.
Creutzenberg, O.; Bellmann, B.; Klingebiel, R.; Heinrich, U.; Muhle, H. (1995) Phagocytosis and chemotaxis of rat
alveolar macrophages after a combined or separate exposure to ozone and carbon black. Exp. Toxicol. Pathol.
47: 202-206.
Custodio-Ramirez, V.; Paz, C. (1997) Ozone produces functional deficits in the rat visual pathway.
Electroencephalogr. Clin. Neurophysiol. 104: 269-273.
AX5-63
-------
DeMarini, D. M; Shelton, M. L.; Kohan, M. J.; Hudgens, E. E.; Kleindienst, T. E.; Ball, L. M; Walsh, D.; de Boer,
J. G.; Lewis-Bevan, L.; Rabinowitz, J. R.; Claxton, L. D.; Lewtas, J. (2000) Mutagenicity in lung of Big
Blue(R) mice and induction of tandem-base substitutions in Salmonella by the air pollutant peroxyacetyl
nitrate (PAN): predicted formation of intrastrand cross-links. Mutat. Res. 457: 41-55.
Delaunois, A.; Segura, P.; Montano, L. M.; Vargas, M. H.; Ansay, M.; Gustin, P. (1998) Comparison of
ozone-induced effects on lung mechanics and hemodynamics in the rabbit. Toxicol. Appl. Pharmacol.
150: 58-67.
Dell'Omo, G.; Fiore, M.; Petruzzi, S.; Alleva, E.; Bignami, G. (1995a) Neurobehavioral development of CD-I mice
after combined gestational and postnatal exposure to ozone. Arch. Toxicol. 69: 608-616.
DeH'Omo, G.; Wolfer, D.; Alleva, E.; Lipp, H.-P. (1995b) Developmental exposure to ozone induces subtle changes
in swimming navigation of adult mice. Toxicol. Lett. 81: 91-99.
Depuydt, P.; Joos, G. F.; Pauwels, R. A. (1999) Ambient ozone concentrations induce airway hyperresponsiveness
in some rat strains. Eur. Respir. J. 14: 125-131.
Depuydt, P. O.; Lambrecht, B. N.; Joos, G. F.; Pauwels, R. A. (2002) Effect of ozone exposure on allergic
sensitization and airway inflammation induced by dendritic cells. Clin. Exp. Allergy 32: 391-396.
Dong, W.; Selgrade, M. K.; Gilmour,M. L; Lange, R. W.; Park, P.; Luster, M. I.; Kari, F. W. (1998) Altered alveolar
macrophage function in calorie-restricted rats. Am. J. Respir. Cell Mol. Biol. 19: 462-469
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, J. A. M. A.; Boere, A. J. F.; vanLoveren, H.; Rombout, P. J. A.; Marra, M.; vanBree, L. (1996)
Age-related toxicity in rat lungs following acute and repeated ozone exposure. Inhalation Toxicol. 8: 903-925.
Dormans, J. A. M. A.; VanBree, L.; Boere, A. J. F.; Marra, M.; Rombout, P. J. A. (1999) Interspecies differences in
time course of pulmonary toxicity following repeated exposure to ozone. Inhalation Toxicol. 11: 309-329.
Driscoll, K. E.; Simpson, L.; Carter, J.; Hassenbein, D.; Leikauf, G. D. (1993) Ozone inhalation stimulates
expression of a neutrophil chemotactic protein, macrophage inflammatory protein 2. Toxicol. Appl.
Pharmacol. 119:306-309.
Dye, J. A.; Madden, M. C.; Richards, J. H.; Lehmann, J. R.; Devlin, R. B.; Costa, D. L. (1999) Ozone effects on
airway responsiveness, lung injury, and inflammation. Comparative rat strain and in vivo/in vitro
investigations. Inhalation Toxicol. 11: 1015-1040.
El-Fawal, H. A. N.; McGovern, T.; Schlesinger, R. B. (1995) Nonspecific bronchial responsiveness assessed in vitro
following acute inhalation exposure to ozone and ozone/sulfuric acid mixtures. Exp. Lung Res. 21: 129-139.
Elder, A. C. P.; Gelein, R.; Finkelstein J. N.; Cox, C.; Oberdorster, G. (2000a) Endotoxin priming affects the lung
response to ultrafine particles and ozone in young and old rats. In: Phalen, R. F., ed. Inhalation toxicology:
proceedings of the third colloquium on paniculate air pollution and human health (first special issue);
June, 1999; Durham, NC. Inhalation Toxicol. 12(suppl. 1): 85-98.
Elder, A. C. P.; Gelein, R.; Finkelstein, J. N.; Cox, C.; Oberdorster, G. (2000b) Pulmonary inflammatory response to
inhaled ultrafine particles is modified by age, ozone exposure, and bacterial toxin. In: Grant, L. D., ed.
PM2000: paniculate matter and health. Inhalation Toxicol. 12(suppl. 4): 227-246.
Elsayed, N. M. (2001) Diet restriction modulates lung response and survivability of rats exposed to ozone.
Toxicology 159: 171-182.
Evans, M. J.; Fanucchi, M. V; Baker, G. L.; Van Winkle, L. S.; Pantle, L. M.; Nishio, S. J.; Schelegle, E. S.;
Gershwhin, L. J.; Miller, L. A.; Hyde, D. M.; Sannes, P. L.; Plopper, C. G. (2003) Atypical development of
the trachea! basement membrane zone of infant rhesus monkeys exposed to ozone and allergen. Am. J.
Physiol. 285:L931-L939.
Fakhrzadeh, L.; Laskin, J. D.; Laskin, D. L. (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, M. V.; Hotchkiss, J. A.; Harkema, J. R. (1998) Endotoxin potentiates ozone-induced mucous cell
metaplasia in rat nasal epithelium. Toxicol. Appl. Pharmacol. 152: 1-9.
Farman, C. A.; Pinkerton, K. E.; Rajini, P.; Witschi, H.; Last, J. A. (1997) Evolution of lung lesions in rats exposed
to mixtures of ozone and nitrogen dioxide. Inhalation Toxicol. 9: 647-677'.
Farman, C. A.; Watkins, K.; Van Hoozen, B.; Last, J. A.; Witschi, H.; Pinkerton, K. E. (1999) Centriacinar
remodeling and sustained procollagen gene expression after exposure to ozone and nitrogen dioxide. Am. J.
Respir. Cell Mol. Biol. 20: 303-311.
Ferng, S.-F.; Castro, C. E.; Afifi, A. A.; Bermudez, E.; Mustafa, M. G. (1997) Ozone-induced DNA strand breaks in
guinea pig tracheobronchial epithelial cells. J. Toxicol. Environ. Health 51: 353-367.
AX5-64
-------
Foster, W. M; Freed, A. N. (1999) Regional clearance of solute from peripheral airway epithelia: recovery after
sublobar exposure to ozone. J. Appl. Physiol. 86: 641-646.
Frampton, M. W.; Pryor, W. A.; Cueto, R.; Cox, C.; Morrow, P. E.; Utell, M. J. (1999) Aldehydes (nonanal and
hexanal) in rat and human bronchoalveolar lavage fluid after ozone exposure. Cambridge, MA: Health Effects
Institute; research report no. 90. Available: www.healtheffects.org/Pubs/Frampton-C.pdf [2000, February 9].
Freed, A. N.; Chou, C. L.; Fuller, S. D.; Croxton, T. L. (1996) Ozone-induced vagal reflex modulates airways
reactivity in rabbits. Respir. Physiol. 105: 95-102.
Freed, A. N.; Cueto, R.; Pryor, W. A. (1999) Antioxidant transport modulates peripheral airway reactivity and
inflammation during ozone exposure. J. Appl. Physiol. 87: 1595-1603.
Garssen, J.; Van Bree, L.; Van Der Vliet, H.; Van Loveren, H. (1997) Ozone-induced impairment of pulmonary type
IV hypersensitivity and airway hyperresponsiveness in mice. Inhalation Toxicol. 9: 581-599.
Gohil, K.; Cross, C. E.; Last, J. A. (2003) Ozone-induced disruptions of lung transcriptomes. Biochem. Biophys.
Res. Commun. 305: 719-728.
Goldsmith, C.-A. W.; Ning, Y.-Y.; Qin, G.; Imrich, A.; Lawrence, J.; Murthy, G. G., K.; Catalano, P. J.; Kobzik, L.
(2002) Combined air pollution particle and ozone exposure increases airway responsiveness in mice.
Inhalation Toxicol. 14: 325-347.
Gonzalez-Pina, R.; Paz, C. (1997) Brain monoamine changes in rats after short periods of ozone exposure.
Neurochem. Res. 22: 63-66.
Guerrero, A. L.; 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.
Gupta, S. K.; Reinhart, P. G.; Bhalla, D. K. (1998) Enhancement of fibronectin expression in rat lung by ozone and
an inflammatory stimulus. Am. J. Physiol. 275: L330-L335.
Haddad, E.-B.; Liu, S. F.; Salmon, M.; Robichaud, A.; Barnes, P. J.; Chung, K. F. (1995) Expression of inducible
nitric oxide synthase mRNA in Brown Norway rats exposed to ozone: effect of dexamethasone. Eur. J.
Pharmacol. Environ. Toxicol. Pharmacol. Sect. 293: 287-290.
Hamilton, R. F.; Li, L.; Eschenbacher, W. L.; 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.
Harkema, J. R.; Morgan, K. T.; Gross, E. A.; Catalano, P. J.; Griffith, W. C. (1994) Consequences of prolonged
inhalation of ozone on F344/N rats: collaborative studies. Part VII: effects on the nasal mucociliary apparatus.
Cambridge, MA: Health Effects Institute; research report no. 65.
Harkema, J. R.; Catalano, P. J.; Hotchkiss, J. A. (1997a) Consequences of prolonged inhalation of ozone onF344/N
rats: collaborative studies. Part XII. Atrophy of bone in nasal turbinates. Cambridge, MA: Health Effects
Institute; research report no. 65.
Harkema, J. R.; Hotchkiss, J. A.; Griffith, W. C. (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, J. R.; Hotchkiss, J. A.; Barr, E. B.; Bennett, C. B.; Gallup, M.; Lee, J. K.; 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.
Heddle, J. A.; Shepson, P. B.; Gingerich, J. D.; So, K. W. (1993) Mutagenicity of peroxyacetyl nitrate (PAN)
in vivo: tests for somatic mutations and chromosomal aberrations. Environ. Mol. Mutagen. 21: 58-66.
Herbert, R. A.; Hailey, J. R.; Grumbein, S.; Chou, B. J.; Sills, R. C.; Haseman, J. K.; Goehl, T.; Miller, R. A.;
Roycroft, J. H.; Boorman, G. A. (1996) Two-year and lifetime toxicity and carcinogenicity studies of ozone in
B6C3F1 mice. Toxicol. Pathol. 24: 539-548.
Highfill, J. W.; Watkinson, W. P. (1996) Ozone toxicity in the rat. II. Modeling changes due to ambient temperatures
and duration. J. Appl. Physiol. 80: 1811-1818.
Hoffer, E.; Baum, Y.; Tabak, A.; Frevert, C. (1999) Adhesion molecules of blood polymorphonuclear leukocytes
and alveolar macrophages in rats: modulation by exposure to ozone. Hum. Exp. Toxicol. 18: 547-551.
Hotchkiss, J. A.; Harkema, J. R.; Johnson, N. F. (1997) Kinetics of nasal epithelial cell loss and proliferation in F344
rats following a single exposure to 0.5 ppm ozone. Toxicol. Appl. Pharmacol. 143: 75-82.
Hotchkiss, J. A.; Hilaski, R.; Cho, H.; Regan, K.; Spencer, P.; Slack, K.; Harkema, J. R. (1998) Fluticasone
propionate attenuates ozone-induced rhinitis and mucous cell metaplasia in rat nasal airway epithelium.
Am. J. Respir. Cell Mol. Biol. 18: 91-99.
AX5-65
-------
Huffman, L. I; Judy, D. I; Brumbaugh, K.; Frazer, D. G.; Reynolds, J. S.; McKinney, W. G.; Goldsmith, W. T.
(2001) Hyperthyroidism increases the risk of ozone-induced lung toxicity in rats. Toxicol. Appl. Pharmacol.
173: 18-26.
Huitron-Resendiz, S.; Custodio-Ramirez, V.; Escalante-Membrillo, C.; Gonzalez-Pina, R.; Paz, C. (1994) Sleep
alterations and brain regional changes of serotonin and its metabolite in rats exposed to ozone. Neurosci. Lett.
177: 119-122.
Hyde, D. M; Miller, L. A.; McDonald, R. J.; Stovall, M. Y.; Wong, V.; Pinkerton, K. E.; Wegner, C. D.; Rothlein,
R.; Plopper, C. G. (1999) Neutrophils enhance clearance of necrotic epithelial cells in ozone-induced lung
injury in rhesus monkeys. Am. J. Physiol. 277: L1190-L1198.
Igarashi, A.; lijima, H.; Tamura, G.; Shirato, K. (1998) Tazanolast inhibits ozone-induced airway
hyperresponsiveness in guinea pigs. Am. J. Respir. Crit. Care Med. 157: 1531-1535.
lijima, M. K.; Kobayashi, T.; Kamada, H.; Shimojo, N. (2001) Exposure to ozone aggravates nasal allergy-like
symptoms in guinea pigs. Toxicol. Lett. 123: 77-85.
Inoue, H.; Aizawa, H.; Nakano, H.; Matsumoto, K.; Kuwano, K.; Nadel, J. A.; Kara, N. (2000) Nitric oxide synthase
inhibitors attenuate ozone-induced airway inflammation in guinea pigs: possible role of interleukin-8. Am. J.
Respir. Crit. Care Med. 161: 249-256.
Ishii, Y.; Yang, H.; Sakamoto, T.; Nomura, A.; Hasegawa, S.; Hirata, F.; Bassett, D. J. P. (1997) Rat alveolar
macrophage cytokine production and regulation of neutrophil recruitment following acute ozone exposure.
Toxicol. Appl. Pharmacol. 147: 214-223.
Ishii, Y.; Shirato, M.; Nomura, A.; Sakamoto, T.; Uchida, Y.; Ohtsuka, M.; Sagai, M.; Hasegawa, S. (1998) Cloning
of rat eotaxin: ozone inhalation increases mRNA and protein expression in lungs of brown Norway rats. Am.
J. Physiol. 274: L171-L176.
Ishii, Y.; Hashimoto, K.; Hirano, K.; Morishima, Y.; Mochizuki, M.; Masuyama, K.; Nomura, A.; Sakamoto, T.;
Uchida, Y.; Sagai, M.; Sekizawa, K. (2000a) Ebselen decreases ozone-induced pulmonary inflammation in
rats. Lung 178: 225-234.
Ishii, Y.; Hirano, K.; Morishima, Y.; Masuyama, K.; Goto, Y.; Nomura, A.; Sakamoto, T.; Uchida, Y.; Sagai, M.;
Sekizawa, K. (2000b) Early molecular and cellular events of oxidant-induced pulmonary fibrosis in rats.
Toxicol. Appl. Pharmacol. 167: 173-181.
Iwasaki, T.; Takahashi, M.; Saito, H.; Arito, H. (1998) Adaptation of extrapulmonary responses to ozone exposure in
conscious rats. Ind. Health 36: 57-60.
Jang, A.-S.; Choi, I.-S.; Koh, Y.-L; Park, C.-S.; Lee, J.-S. (2002) The relationship between alveolar epithelial
proliferation and airway obstruction after ozone exposure. Allergy 57: 737-740.
Jimba, M.; Skornik, W. A.; Killingsworth, C. R.; Long, N. C.; Brain, J. D.; Shore, S. A. (1995) Role of C fibers in
physiological responses to ozone in rats. J. Appl. Physiol. 78: 1757-1763.
Joad, J. P.; Kott, K. S.; Bonham, A. C. (1998) Exposing guinea pigs to ozone for 1 wk enhances responsiveness of
rapidly adapting receptors. J. Appl. Physiol. 84: 1190-1197.
Johnston, C. J.; Stripp, B. R.; Reynolds, S. D.; Avissar, N. E.; Reed, C. K.; Finkelstein, J. N. (1999a) Inflammatory
and antioxidant gene expression in C57BL/6J mice after lethal and sublethal ozone exposures. Exp. Lung
Res. 25: 81-97.
Johnston, C. J.; Finkelstein, J. N.; Oberdorster, G.; Reynolds, S. D.; Stripp, B. R. (1999b) Clara cell secretory
protein-deficient mice differ from wild-type mice in inflammatory chemokine expression to oxygen and
ozone, but not to endotoxin. Exp. Lung Res. 25: 7-21.
Johnston, C. J.; Reed, C. K.; Avissar, N. E.; Gelein, R.; Finkelstein, J. N. (2000a) Antioxidant and inflammatory
response after acute nitrogen dioxide and ozone exposures in C57B1/6 mice. Inhalation Toxicol. 12: 187-203.
Johnston, C. J.; Oberdorster, G.; Gelein, R.; Finkelstein, J. N. (2000b) Newborn mice differ from adult mice in
chemokine and cytokine expression to ozone, but not to endotoxin. Inhalation Toxicol. 12: 205-224.
Johnston, C. J.; Oberdorster, G.; Finkelstein, J. N. (2001) Recovery from oxidant-mediated lung injury: response of
metallothionein, MIP-2, and MCP-1 to nitrogen dioxide, oxygen, and ozone exposures. Inhalation Toxicol.
13:689-702.
Johnston, C. J.; Oberdorster, G.; Gelein, R.; Finkelstein, J. N. (2002) Endotoxin potentiates ozone-induced
pulmonary chemokine and inflammatory responses. Exp. Lung Res. 28: 419-433.
Kafoury, R. M.; Pryor, W. A.; Squadrito, G. L.; Salgo, M. G.; Zou, X.; Friedman, M. (1999) Induction of
inflammatory mediators in human airway epithelial cells by lipid ozonation products. Am. J. Respir. Crit.
Care Med. 160: 1934-1942.
Kenyon, N. J.; VanDer Vliet, A.; Schock, B. C.; Okamoto, T.; McGrew, G. M.; Last, J. A. (2002) Susceptibility to
ozone-induced acute lung injury in iNOS-deficient mice. Am. J. Physiol. 282: L540-L545.
AX5-66
-------
Kim, M. Y.; Son, J. W.; Cho, M. H.; Choi, C. S.; Chae, C. H.; Lee, M. H. (2001) Oviductural carcinoma inB6C3Fl
female mice exposed to 0.5 ppm ozone. Vet. Hum. Toxicol. 43: 370-372.
Kimmel, T. A.; Chen, L. C.; Bosland, M. C.; Nadziejko, C. (1997) Influence of acid aerosol droplet size on
structural changes in the rat lung caused by acute exposure to sulfuric acid and ozone. Toxicol. Appl.
Pharmacol. 144: 348-355.
Kleeberger, S. R.; Levitt, R. C.; Zhang, L.-Y.; Longphre, M.; Harkema, J.; Jedlicka, A.; Eleff, S. M.; DiSilvestre, D.;
Holroyd, K. J. (1997) Linkage analysis of susceptibility to ozone-induced lung inflammation in inbred mice.
Nat. Genet. 17: 475-478.
Kleeberger, S. R.; Reddy, S.; Zhang, L.-Y.; Jedlicka, A. E. (2000) Genetic susceptibility to ozone-induced lung
hyperpermeability: role of toll-like receptor 4. Am. J. Respir. Cell Mol. Biol. 22: 620-627.
Kleeberger, S. R.; Ohtsuka, Y.; Ahang, L.-Y.; Longphre, M. (200la) Airway responses to chronic ozone exposure
are partially mediated through mast cells. J. Appl. Physiol. 90: 713-723.
Kleeberger, S. R.; Reddy, S. P.; Zhang, L.-Y.; Cho, H.-Y.; Jedlicka, A. E. (200Ib) Toll-like receptor 4 mediates
ozone-induced murine lung hyperpermeability via inducible nitric oxide synthase. Am. J. Physiol.
280: L326-L333.
Kleinman, M. T.; Mautz, W. J.; Bjarnason, S. (1999) Adaptive and non-adaptive responses in rats exposed to ozone,
alone and in mixtures, with acidic aerosols. Inhalation Toxicol. 11: 249-264.
Kleinman, M. T.; Bufalino, C.; Rasmussen, R.; Hyde, D.; Bhalla, D. K.; Mautz, W. J. (2000) Toxicity of chemical
components of ambient fine paniculate matter (PM25) inhaled by aged rats. J. Appl. Toxicol. 20: 357-364.
Kligerman, A. D.; Mottus, K.; Erexson, G. L. (1995) Cytogenetic analyses of the in vitro and in vivo responses of
murine cells to peroxyacetyl nitrate (PAN). Mutat. Res. 341: 199-206.
Kobzik, L.; Goldsmith, C.-A. W.; Ning, Y. Y.; Qin, G.; Morgan, B.; Imrich, A.; Lawrence J.; Murthy, G. G. K.;
Catalano, P. J. (2001) Effects of combined ozone and air pollution particle exposure in mice. Boston, MA:
Health Effects Institute; research report no. 106. Available: http://www.healtheffects.org/Pubs/Kobzik.pdf
[27 January 2003].
Kodavanti, U. P.; Hatch, G. E.; Starcher, B.; Giri, S. N.; Winsett, D.; Costa, D. L. (1995) Ozone-induced pulmonary
functional, pathological, and biochemical changes in normal and vitamin C-deficient guinea pigs. Fundam.
Appl. Toxicol. 24: 154-164.
Koike, E.; Kobayashi, T.; Nelson, D. J.; Me William, A. S.; Holt, P. G. (1998) Effect of ozone exposure on alveolar
macrophage-mediated immunosuppressive activity in rats. Toxicol. Sci. 41: 217-223.
Koike, E.; Kobayashi, T.; Murakami, M.; McWilliam, A. S.; Holt, P. G. (1999) Mechanisms of ozone-induced
inhibitory effect of bronchoalveolar lavage fluid on alveolar macrophage-mediated immunosuppressive
activity in rats. J. Leukoc. Biol. 66: 75-82.
Koike, E.; Kobayashi, T.; Shimojo, N. (2001) Ozone exposure enhances expression of cell-surface molecules
associated with antigen-presenting activity on bronchoalveolar lavage cells in rats. Toxicol. Sci. 63: 115-124.
Koto, H.; Salmon, M.; Haddad el-B.; Huang, T.-J.; Zagorski, J.; Chung, K. F. (1997) Role of cytokine-induced
neutrophil chemoattractant (CINC) in ozone-induced airway inflammation and hyperresponsiveness. Am. J.
Respir. Crit. Care Med. 156: 234-239.
Larson, S. D.; Schelegle, E. S.; Walby, W. F.; Gershwin, L. J.; Fanuccihi, M. V.; Evans, M. J.; Joad, J. P.;
Tarkington, B. K.; Hyde, D. M.; Plopper, C. G. (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, D. L.; Laskin, J. D. (2001) Role of macrophages and inflammatory mediators in chemically induced toxicity.
Toxicology (Ireland) 160: 111-118.
Laskin, D. L.; Pendino, K. J.; Punjabi, C. J.; del Valle, M. R.; Laskin, J. D. (1994) Pulmonary and hepatic effects of
inhaled ozone in rats. Environ. Health Perspect. 102(suppl. 10): 61-64.
Laskin, J. D.; Heck, D. E.; Laskin, D. L. (1996) Nitric oxide production in the lung and liver following inhalation of
the pulmonary irritant ozone. In: Snyder, R.; Kocsis, J. J.; Sipes, I. G.; Kalf, G. F.; Jollow, D. J.; Greim, H.;
Monks, T. J.; Witmer, C. M., eds. Biological Reactive Intermediates V: Basic Mechanistic Research in
Toxicology and Human Risk Assessment: proceedings of the Fifth International Symposium; January 1995;
Munich, Germany. Adv. Exp. Med. Biol. 387: 141-146.
Laskin, D. L.; Heck, D. E.; Laskin, J. D. (1998a) Role of inflammatory cytokines and nitric oxide in hepatic and
pulmonary toxicity. Toxicol. Lett. 102-103: 289-293.
Laskin, D. L.; Sunil, V.; Guo, Y.; Heck, D. E.; Laskin, J. D. (1998b) Increased nitric oxide synthase in the lung after
ozone inhalation is associated with activation of NF-KB. Environ. Health Perspect. 106(suppl. 5): 1175-1178.
AX5-67
-------
Laskin, D. L.; Fakhrzadeh, L.; Heck, D. E.; Gerecke, D.; Laskin, J. D. (2002) Upregulation of phosphoinositide
3-kinase and protein kinase B in alveolar macrophages following ozone inhalation. Role of NF-KB and
STAT-1 in ozone-induced nitric oxide production and toxicity. Mol. Cell. Biochem. 234-235: 91-98.
Last, J. A.; Pinkerton, K. E. (1997) Chronic exposure of rats to ozone and sulfuric acid aerosol: biochemical and
structural responses. Toxicology 116: 133-146.
Lavnikova, N.; Prokhorova, S.; Lakhotia, A. V.; Gordon, R.; Laskin, D. L. (1998) Distinct inflammatory responses
of adherent vascular lung neutrophils to pulmonary irritants. J. Inflammation 48: 56-66
Lee, C.; Watt, K. C.; Chang, A. M; Plopper, C. G.; Buckpitt, A. R.; Pinkerton, K. E. (1998) Site-selective
differences in cytochrome P450 isoform activities: comparison of expression in rat and rhesus monkey lung
and induction in rats. Drug Metab. Dispos. 26: 396-400.
Lemos, M.; Lichtenfels, A. J. F. C.; Amaro, E., Jr.; Macchione, M.; Martins, M. A.; King, M.; Bohm, G. M.;
Saldiva, P. H. N. (1994) Quantitative pathology of nasal passages in rats exposed to urban levels of air
pollution. Environ. Res. 66: 87-95.
Longphre, M.; Zhang, L.-Y.; Harkema, J. R.; Kleeberger, S. R. (1999) Ozone-induced pulmonary inflammation and
epithelial proliferation are partially mediated by PAF. J. Appl. Physiol. 86: 341-349.
Madden, M. C.; Richards, J. H.; Dailey, L. A.; Hatch, G. E.; Ohio, A. J. (2000) Effect of ozone on diesel exhaust
particle toxicity in rat lung. Toxicol. Appl. Pharmacol. 168: 140-148.
Mango, G. W.; Johnston, C. J.; Reynolds, S. D.; Finkelstein, J. N.; Plopper, C. G.; Stripp, B. R. (1998) Clara cell
secretory protein deficiency increases oxidant stress response in conducting airways. Am. J. Physiol.
275: L348-L356.
Mautz, W. J. (2003) Exercising animal models in inhalation toxicology: interactions with ozone and formaldehyde.
Environ. Res. 92: 14-26.
Mautz, W. J.; Kleinman, M. T.; Bhalla, D. K.; Phalen, R. F. (2001) Respiratory tract responses to repeated inhalation
of an oxidant and acid gas-particle air pollutant mixture. Toxicol. Sci. 61: 331-341.
McGraw, D. W.; Forbes, S. L.; Mak, J. C. W.; Witte, D. P.; Carrigan, P. E.; Leikauf, G. D.; Liggett, S. B. (2000)
Transgenic overexpression of p2-adrenergic receptors in airway epithelial cells decreases bronchoconstriction.
Am. J. Physiol. 279: L379-L389.
McKinney, W. J.; Jaskot, R. H.; Richards, J. H.; Costa, D. L.; Dreher, K. L. (1998) Cytokine mediation of
ozone-induced pulmonary adaptation. Am. J. Respir. Cell. Mol. Biol. 18: 696-705.
Miller, L. A.; Barnett, N. L.; Sheppard, D.; Hyde, D. M. (2001) Expression of the P6 integrin subunit is associated
with sites of neutrophil influx in lung epithelium. J. Histochem. Cytochem. 49: 41-48.
Morio, L. A.; Hooper, K. A.; Brittingham, J.; Li, T.-H.; Gordon, R. E.; Turpin, B. J.; Laskin, D. L. (2001) Tissue
injury following inhalation of fine paniculate matter and hydrogen perioxide is associated with altered
production of inflammatory mediators and antioxidants by alveolar macrophages. Toxicol. Appl. Pharmacol.
177: 188-199.
Moss, O. R.; Gross, E. A.; James, R. A.; Janszen, D. B.; Ross, P. W.; Roberts, K. C.; Howard, A. M.; Harkema,
J. R.; Calderon-Garciduenas, L.; Morgan, K. T. (2001) Respiratory tract toxicity in rats exposed to Mexico
City air. Cambridge, MA: Health Effects Institute; research report no. 100. Available:
http://www.healtheffects.org/pubs-research.htm [15 May, 2003].
Mudway, I. S.; Kelly, F. J. (1998) Modeling the interactions of ozone with pulmonary epithelial lining fluid
antioxidants. Toxicol. Appl. Pharmacol. 148: 91-100.
Neuhaus-Steinmetz, U.; Uffhausen, F.; Herz, U.; Renz, H. (2000) Priming of allergic immune responses by repeated
ozone exposure in mice. Am. J. Respir. Cell Mol. Biol. 23: 228-233.
Nichols, B. G.; Woods, J. S.; Luchtel, D. L.; Corral, J.; Koenig, J. Q. (2001) Effects of ozone exposure on nuclear
factor-KB activation and tumor necrosis factor-a expression in human nasal epithelial cells. Toxicol. Sci. 60:
356-362.
Nielsen, G. D.; Hougaard, K. S.; Larsen, S. T.; Hammer, M.; Wolkoff, P.; Clausen, P. A.; Wilkins, C. K.; Alarie, Y.
(1999) Acute airway effects of formaldehyde and ozone inBALB/c mice. Hum. Exp. Toxicol. 18: 400-409.
Nino-Cabrera, H. G.; Colin-Barenque, L.; Avila-Costa, M. R.; Espinosa-Villanueva, J.; Fortoul, T. I.;
Rivas-Arancibia, S. (2002) Differences between hippocampus and cerebral cortex in aged rats in an oxidative
stress model. Int. J. Neurosci. 112: 373-381.
Noviski, N.; Brewer, J. P.; Skornik, W. A.; Galli, S. J.; Drazen, J. M.; Martin, T. R. (1999) Mast cell activation is
not required for induction of airway hyperresponsiveness by ozone in mice. J. Appl. Physiol. 86: 202-210.
Paige, R. C.; Royce, F. H.; Plopper, C. G.; Buckpitt, A. R. (2000a) Long-term exposure to ozone increases acute
pulmonary centriacinar injury by 1-nitronaphthalene: I. Region-specific enzyme activity. J. Pharmacol. Exp.
Ther. 295: 934-941.
AX5-68
-------
Paige, R. C.; Wong, V.; Plopper, C. G. (2000b) Long-term exposure to ozone increases acute pulmonary centriacinar
injury by 1-nitronaphtnalene: II. Quantitative histopathology. J. Pharmacol. Exp. Ther. 295: 942-950.
Paquette, N. C.; Tankersley, C. G.; Zhang, L.-Y.; Kleeberger, S. R. (1994) Repeated subacute ozone exposure of
inbred mice: airway inflammation and ventilation. Exp. Lung Res. 20: 579-594.
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.
Pearson, A. C.; Bhalla, D. K. (1997) Effects of ozone onmacrophage adhesion in vitro and epithelial and
inflammatory responses in vivo: the role of cytokines. J. Toxicol. Environ. Health 50: 143-157.
Peden, D. B.; Dailey, L. (1995) Modulation of mast cell functions by in vitro ozone exposure. Am. J. Physiol.
268: L902-L910.
Pendino, K. J.; Shuler, R. L.; Laskin, J. D.; Laskin, D. L. (1994) Enhanced production of interleukin-1, tumor
necrosis factor-a, and fibronectin by rat lung phagocytes following inhalation of a pulmonary irritant. Am. J.
Respir. CellMol. Biol. 11: 279-286.
Pendino, K. J.; Gardner, C. R.; Shuler, R. L.; Laskin, J. D.; Durham, S. K.; Barton, D. S.; Ohnishi, S. T.; Ohnishi,
T.; Laskin, D. L. (1996) Inhibition of ozone-induced nitric oxide synthase expression in the lung by
endotoxin. Am. J. Respir. CellMol. Biol. 14: 516-525.
Petruzzi, S.; Fiore, M.; DeH'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-I mice after pre- and postnatal ozone exposure. Acta Neurobiol. Exp. 59: 115-122.
Pinkerton, K. E.; Menache, M. G.; Plopper, C. G. (1995) Consequences of prolonged inhalation of ozone onF344/N
rats: collaborative studies. Part IX. Changes in the tracheobronchial epithelium, pulmonary acinus, and lung
antioxidant enzyme activity. Cambridge, MA: Health Effects Institute; pp. 41-98; research report no. 65.
Available from: NTIS, Springfield, VA; PB95-261996.
Pinkerton, K. E.; Weller, B. L.; Menache, M. G.; Plopper, C. G. (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. Cambridge, MA: Health Effects Institute;
research report no. 65.
Plopper, C. G.; Fanucchi, M. V. (2000) Do urban environmental pollutants exacerbate childhood lung diseases?
Environ. Health Perspect. 108: A252-A253.
Plopper, C. G.; Chu, F.-P.; Haselton, C. J.; Peake, J.; Wu, J.; Pinkerton, K. E. (1994) Dose-dependent tolerance to
ozone: I. tracheobronchial epithelial reorganization in rats after 20 months' exposure. Am. J. Pathol.
144: 404-420.
Plopper, C. G.; Hatch, G. E.; Wong, V.; Duan, X.; Weir, A. J.; Tarkington, B. K.; Devlin, R. B.; Becker, S.;
Buckpitt, A. R. (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.
Postlethwait, E. M.; Cueto, R.; Velsor, L. W.; Pryor, W. A. (1998) O3-induced formation of bioactive lipids:
estimated surface concentrations and lining layer effects. Am. J. Physiol. 274: L1006-L1016.
Pryor, W. A.; Squadrito, G. L.; Friedman, M. (1995) A new mechanism for the toxicity of ozone. Toxicol. Lett.
82/83:287-293.
Pryor, W. A.; Bermudez, E.; Cueto, R.; Squadrito, G. L. (1996) Detection of aldehydes in bronchoalveolar lavage of
rats exposed to ozone. Fundam. Appl. Toxicol. 34: 148-156.
Pulfer, M. K.; Murphy, R. C. (2004) Formation of biologically active oxysterols during ozonolysis of cholesterol
present in lung surfactant. J. Biol. Chem. 279: 26331-26338.
Pulfer, M. K.; Taube, C.; Gelfand, E.; Murphy, R. C. (2005) Ozone exposure in vivo and formation of biologically
active oxysterols in the lung. J. Pharmacol. Exp. Ther. 312: 256-264.
Reinhart, P. G.; Gupta, S. K.; Bhalla, D. K. (1999) Attenuation of ozone-induced lung injury by interleukin-10.
Toxicol. Lett. 110:35-42.
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.
AX5-69
-------
Rivas-Arancibia, S.; Dorado-Martinez, C.; Borgonio-Perez, G.; Hiriart-Urdanivia, M.; Verdugo-Diaz, L.;
Duran-Vazquez, A.; Colin-Baranque, L.; Avila-Costa, M. R. (2000) Effects of taurine on ozone-induced
memory deficits and lipid peroxidation levels in brains of young, mature, and old rats. Environ. Res. 82: 7-17.
Rivas-Manzano, P.; Paz, C. (1999) Cerebellar morphological alterations in rats induced by prenatal ozone exposure.
Neurosci. Lett. 276: 37-40.
Savov, J. D.; Whitehead, G. S.; Wang, I; Liao, G.; Usuka, I; Peltz, G.; Foster, W. M.; Schwartz, D. A. (2004)
Ozone-induced acute pulmonary injury in inbred mouse strains. Am. J. Respir. Cell Mol. Biol. 31: 69-77.
Schelegle, E. S.; Alfaro, M. F.; Putney, L.; Stovall, M.; Tyler, N.; Hyde, D. M. (2001) Effect of C-fiber-mediated,
ozone-induced rapid shallow breathing on airway epithelial injury in rats. J. Appl. Physiol. 91: 1611-1618.
Schelegle, E. S.; Miller, L. A.; Gershwin, L. J.; Fanucchi, M. V.; Van Winkle, L. S.; Gerriets, J. E.; Walby, W. F.;
Mitchell, V; Tarkington, B. K.; Wong, V. J.; Baker, G. L.; Pantle, L. M.; Joad, J. P.; Pinkerton, K. E.;
Wu, R.; Evans, M. J.; Hyde, D. M.; Plopper, C. G. (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.
Schlesinger, R. B.; Cohen, M. D.; Gordon, T.; Nadziejko, C.; Zelikoff, J. T.; Sisco, M.; Regal, J. F.; Menache, M. G.
(2002a) Ozone differentially modulates airway responsiveness in atopic versus nonatopic guinea pigs.
Inhalation Toxicol. 14: 431-457.
Schlesinger, R. B.; Cohen, M.; Gordon, T.; Nadziejko, C.; Zelikoff, J. T.; Sisco, M.; Regal, J. F.; Menache, M. G.
(2002b) Ozone-induced modulation of airway hyperresponsiveness in guinea pigs. Boston, MA: Health
Effects Institute; research report no. 109.
Segura, P.; Montano, L. M.; Bazan-Perkins, B.; Gustin, P.; Vargas, M. H. (1997) Ozone at high-pollution urban
levels causes airway hyperresponsiveness to substance P but not to other agonists. Environ. Toxicol.
Pharmacol. 3: 91-95.
Sen, S.; Dulchavsky, S. A.; Dutta, S. (1993) Effects of triiodothyronine (T3) supplementation upon ozone-induced
lung injury. Free Radic. Res. Commun. 18: 299-308.
Shore, S. A.; Abraham, J. H.; Schwartzman, I. N.; Murthy, G. G.; Laporte, J. D. (2000) Ventilatory responses to
ozone are reduced in immature rats. J. Appl. Physiol. 88: 2023-2030.
Shore, S. A.; Johnston, R. A.; Schwartzman, I. N.; Chism, D.; Krishna Murthy, G. G. (2002) Ozone-induced airway
hyperresponsiveness is reduced in immature mice. J. Appl. Physiol. 92: 1019-1028.
Shore, S. A.; Rivera-Sanchez, Y. M.; Schwartzman, I. N.; Johnston, R. A. (2003) Responses to ozone are increased
in obese mice. J. Appl. Physiol. 95: 938-945.
Sindhu, R. K.; Mautz, W. J.; Kikkawa, Y. (1998) Chronic exposure to ozone and nitric acid vapor results in
increased levels of rat pulmonary putrescine. Arch. Toxicol. 72: 445-449.
Slade, R.; Watkinson, W. P.; Hatch, G. E. (1997) Mouse strain differences in ozone dosimetry and body temperature
changes. Am. J. Physiol. 272: L73-L77.
Sorace, A.; De Acetis, L.; Alleva, E.; Santucci, D. (2001) Prolonged exposure to low doses of ozone: short- and
long-term changes to behavioral performance in mice. Environ. Res. 85: 122-134.
Spannhake, E. W. (1996) Down-regulation of canine airway mast cell function following exposure to ozone in vivo.
Exp. Lung Res. 22: 163-178.
Sterner-Kock, A.; Kock, M.; Braun, R.; Hyde, D. M. (2000) Ozone-induced epithelial injury in the ferret is similar
to nonhuman primates. Am. J. Respir. Crit. Care Med. 162: 1152-1156.
Stockstill, B. L.; Chang, L.-Y.; Menache, M. G.; Mellick, P. W.; Mercer, R. R.; Crapo, J. D. (1995) Bronchiolarized
metaplasia and interstitial fibrosis in rat lungs chronically exposed to high ambient levels of ozone. Toxicol.
Appl. Pharmacol. 134: 251-263.
Sun, J.; Koto, H.; Chung, K. F. (1997) Interaction of ozone and allergen challenges on bronchial responsiveness and
inflammation in sensitised guinea pigs. Int. Arch. Allergy Immunol. 112: 191-195.
Szarek, J. L.; Stewart, N. L.; Zhang, J. Z.; Webb, J. A.; Valentovic, M. A.; Catalano, P. (1995) Contractile responses
and structure of small bronchi isolated from rats after 20 months' exposure to ozone. Fundam. Appl. Toxicol.
28: 199-208.
Takahashi, T.; Miura, M.; Katsumata, U.; Ichinose, M.; Kimura, K.; Inoue, H.; Takishima, T.; Shirato, K. (1993)
Involvement of superoxide in ozone-induced airway hyperresponsiveness in anesthetized cats. Am. Rev.
Respir. Dis. 148: 103-106.
Takahashi, N.; Yu, X.-Y.; Schofield, B. H.; Kleeberger, S. R.; Scott, A. L.; Hasegawa, S.; Spannhake, E. W. (1995a)
Expression of ICAM-1 in airway epithelium after acute ozone exposure in the mouse. J. Appl. Physiol.
79: 1753-1761.
AX5-70
-------
Takahashi, M; Kleeberger, S. R.; Croxton, T. L. (1995b) Genetic control of susceptibility to ozone-induced changes
in mouse tracheal electrophysiology. Am. J. Physiol. 269: L6-L10.
Takebayashi, T.; Abraham, J.; Murthy, G. G. K.; Lilly, C.; Rodger, L; Shore, S. A. (1998) Role of tachykinins in
airway responses to ozone in rats. J. Appl. Physiol. 85: 442-450.
Tankersley, C. G.; Kleeberger, S. R. (1994) Ozone-induced inflammation and altered ventilation in genetically
susceptible mice: a comparison of acute and subacute exposures. Toxicol Lett. 72: 279-289.
Tankersley, C. G.; Fitzgerald, R. S.; Mitzner, W. A.; Kleeberger, S. R. (1993) Hypercapnic ventilatory responses in
mice differentially susceptible to acute ozone exposure. J. Appl. Physiol. 75: 2613-2619.
Tesfaigzi, J.; Hotchkiss, J. A.; Harkema, J. R. (1998) Expression of the Bcl-2 protein in nasal epithelia of F344/N
rats during mucous cell metaplasia and remodeling. Am. J. Respir. Cell Mol. Biol. 18: 794-799.
Uhlson, C.; Harrison, K.; Allen, C. B.; Ahmad, S.; White, C. W.; Murphy, R. C. (2002) Oxidized phospholipids
derived from ozone-treated lung surfactant extract reduce macrophage and epithelial cell viability. Chem.
Res. Toxicol. 15: 896-906.
Ulrich, M. M. W.; Alink, G. M.; Kumarathasan, P.; Vincent, R.; Boere, A. J.; Cassee, F. R. (2002) Health effects
and time course of paniculate matter on the cardiopulmonary system in rats with lung inflammation.
J. Toxicol. Environ. Health Part A 65: 1571-1595.
Valacchi, G.; Weber, S. U.; Luu, C.; Cross, C. E.; Packer, L. (2000) Ozone potentiates vitamin E depletion by
ultraviolet radiation in the murine stratum corneum. FEES Lett. 466: 165-168.
Valacchi, G.; Pagnin, E.; Okamoto, T.; Corbacho, A. M.; Olano, E.; Davis, P. A.; Van der Vliet, A.; Packer, L.;
Cross, C. E. (2003) Induction of stress proteins and MMP-9 by 0.8 ppm of ozone in murine skin. Biochem.
Biophys. Res. Commun. 305: 741-746.
Van Bree, L.; Dormans, J. A. M. A.; Boere, A. J. F.; Rombout, P. J. A. (2001) Time study on development and
repair of lung injury following ozone exposure in rats. Inhalation Toxicol. 13: 703-717.
Van Bree, L.; Dormans, J. A. M. A.; Koren, H. S.; Devlin, R. B.; Rombout, P. J. A. (2002) Attenuation and recovery
of pulmonary injury in rats following short-term, repeated daily exposure to ozone. Inhalation Toxicol.
14: 883-900.
Vargas, M. H.; Romero, L.; Sommer, B.; Zamudio, P.; Gustin, P.; Montano, L. M. (1998) Chronic exposure to ozone
causes tolerance to airway hyperresponsiveness in guiea pigs: lack of SOD role. J. Appl. Physiol.
84: 1749-1755.
Vesely, D. L.; Giordano, A. T.; Raska-Emery, P.; Montgomery, M. R. (1994a) Ozone increases amino- and
carboxy-terminal atrial natriuretic factor prohormone peptides in lung, heart, and circulation. J. Biochem.
Toxicol. 9: 107-112.
Vesely, D. L.; Giordano, A. T.; Raska-Emery, P.; Montgomery, M. R. (1994b) Increase in atrial natriuretic factor in
the lungs, heart, and circulatory system owing to ozone. Chest 105: 1551-1554.
Vesely, D. L.; Giordano, A. T.; Raska-Emery, P.; Montgomery, M. R. (1994c) Ozone increases atrial natriuretic
peptides in heart, lung and circulation of aged vs. adult animals. Gerontology (Basel) 40: 227-236.
Vesely, K. R.; Schelegle, E. S.; Stovall, M. Y.; Harkema, J. R.; Green, J. F.; Hyde, D. M. (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.
Vincent, R.; Janzen, E. G.; Chen, G.; Kumarathasan, P.; Haire, D. L.; Guenette, J.; Chen, J. Z.; Bray, T. M. (1996)
Spin trapping study in the lungs and liver of F344 rats after exposure to ozone. Free Radical Res. 25: 475-488.
Vincent, R.; Bjarnason, S. G.; Adamson, I. Y. R.; Hedgecock, C.; Kumarathasan, P.; Guenette, J.; Potvin, M.;
Goegan, P.; Bouthillier, L. (1997) Acute pulmonary toxicity of urban paniculate matter and ozone. Am. J.
Pathol. 151: 1563-1570.
Wagner, J. G.; Hotchkiss, J. A.; Harkema, J. R. (200la) Effects of ozone and endotoxin coexposure on rat airway
epithelium: potentiation of toxicant-induced alterations. Environ. Health Perspect. 109(suppl. 4): 591-598.
Wagner, J. G.; VanDyken, S. J.; Hotchkiss, J. A.; Harkema, J. R. (200Ib) Endotoxin enhancement of ozone-induced
mucous cell metaplasia is neutrophil-dependent in rat nasal epithelium. Toxicol. Sci. 60: 338-347.
Wagner, J. G.; Hotchkiss, J. A.; Harkema, J. R. (2002) Enhancement of nasal inflammatory and epithelial responses
after ozone and allergen coexposure in brown Norway rats. Toxicol. Sci. 67: 284-294.
Wagner, J. G.; VanDyken, S. J.; Wierenga, J. R.; Hotchkiss, J. A.; Harkema, J. R. (2003) Ozone exposure enhances
endotoxin-induced mucous cell metaplasia in rat pulmonary airways. Toxicol. Sci. 74: 437-446.
Wang, G.; Umstead, T. M.; Phelps, D. S.; Al-Mondhiry, H.; Floras, J. (2002) The effect of ozone exposure on the
ability of human surfactant protein A variants to stimulate cytokine production. Environ. Health Perspect.
110:79-84.
AX5-71
-------
Watkinson, W. P.; Wiester, M. I; Highfill, J. W. (1995) Ozone toxicity in the rat. I. Effect of changes in ambient
temperature on extrapulmonary physiological parameters. J. Appl. Physiol. 78: 1108-1120.
Watkinson, W. P.; Campen, M. J.; Nolan, J. P.; Costa, D. L. (2001) Cardiovascular and systemic responses to
inhaled pollutants in rodents: effects of ozone and paniculate matter. Environ. Health Perspect.
109(suppl. 4): 539-546.
Watkinson, W. P.; Campen, M. J.; Wichers, L. B.; Nolan, J. P.; Costa, D. L. (2003) Cardiac and thermoregulatory
responses to inhaled pollutants in healthy and compromised rodents: modulation via interaction with
environmental factors. Environ. Res. 92: 35-47.
Watt, K. C.; Plopper, C. G.; Weir, A. J.; Tarkington, B.; Buckpitt, A. R. (1998) Cytochrome P450 2E1 in rat
tracheobronchial airways: response to ozone exposure. Toxicol. Appl. Pharmacol. 149: 195-202.
Weber, S. U.; Thiele, J. J.; Cross, C. E.; Packer, L. (1999) Vitamin C, uric acid, and glutathione gradients in murine
stratum corneum and their susceptibility to ozone exposure. J. Invest. Dermatol. 113:1128-1132.
Weber, S. U.; Jothi, S.; Thiele, J. J. (2000) High-pressure liquid chromatography analysis of ozone-induced
depletion of hydrophilic and lipophilic antioxidants in murine skin. Methods Enzymol. 319: 536-546.
Weber, S. U.; Han, N.; Packer, L. (2001) Ozone: an emerging oxidative stressorto skin. Curr. Probl. in Dermatol.
29: 52-61.
Weller, B. L.; Crapo, J. D.; Slot, J.; Posthuma, G.; Plopper, C. G.; Pinkerton, K. E. (1997) Site- and cell-specific
alteration of lung copper/zinc and manganese superoxide dismutases by chronic ozone exposure. Am. J.
Respir. Cell Mol. Biol. 17: 552-560.
Weller, B. L.; Witschi, H.; Pinkerton, K. E. (2000) Quantitation and localization of pulmonary manganese
superoxide dismutase and tumor necrosis factor a following exposure to ozone and nitrogen dioxide. Toxicol.
Sci. 54: 452-461.
Wiester, M. J.; Watkinson, W. P.; Costa, D. L.; Crissman, K. M.; Richards, J. H.; Winsett, D. W.; Highfill, J. W.
(1996) Ozone toxicity in the rat. III. Effect of changes in ambient temperature on pulmonary parameters.
J. Appl. Physiol. 81: 1691-1700.
Witschi, H.; Espiritu, L; Pinkerton, K. E.; Murphy, K.; Maronpot, R. R. (1999) Ozone carcinogenesis revisited.
Toxicol. Sci. 52: 162-167.
Wu, Z.-X.; Morton, R. F.; Lee, L.-Y. (1997) Role of tachykinins in ozone-induced airway hyperresponsiveness to
cigarette smoke in guinea pigs. J. Appl. Physiol. 83: 958-965.
Yamauchi, T.; Shima, M.; Kuwaki, T.; Ando, M.; Ohmichi, M.; Fukuda, Y.; Adachi, M. (2002) Acute effects of
ozone exposure on lung function in mice sensitized to ovalbumin. Toxicology (Ireland) 172: 69-78.
Yu, M.; Pinkerton, K. E.; Witschi, H. (2002) Short-term exposure to aged and diluted sidestream cigarette smoke
enhances ozone-induced lung injury inB6C3Fl mice. Toxicol. Sci. 65: 99-106.
Zhang, L.-Y.; Levitt, R. C.; Kleeberger, S. R. (1995) Differential susceptibility to ozone-induced airways
hyperreactivity in inbred strains of mice. Exp. Lung Res. 21: 503-518.
Zhao, Q.; Simpson, L. G.; Driscoll, K. E.; Leikauf, G. D. (1998) Chemokine regulation of ozone-induced neutrophil
and monocyte inflammation. Am. J. Physiol. 274: L39-L46.
AX5-72
-------
ANNEX AX6. CONTROLLED HUMAN EXPOSURE
STUDIES OF OZONE AND RELATED
PHOTOCHEMICAL OXIDANTS
AX6.1 INTRODUCTION
Results of ozone (O3) studies in laboratory animals and in vitro test systems were presented
in Chapter 5 and Annex 5. The extrapolation of results from animal studies is one mechanism
by which information on potential adverse human health effects from exposure to O3 is obtained.
More direct evidence of human health effects due to O3 exposure can be obtained through
controlled human exposure studies of volunteer subjects or through field and epidemiologic
studies of populations exposed to ambient O3. Controlled human exposure studies, discussed in
this chapter, typically use fixed concentrations of O3 under carefully regulated environmental
conditions and subject activity levels.
Most of the scientific information selected for review and evaluation in this chapter comes
from the literature published since 1996 which, in addition to further study of physiological
pulmonary responses and respiratory symptoms, has focused on mechanisms of inflammation
and cellular responses to injury induced by O3 inhalation. Older studies are discussed where
only limited new data are available and where new and old data are conflicting. The reader is
referred to both the 1986 and 1996 Air Quality Criteria documents (U.S. Environmental
Protection Agency, 1986, 1996) for a more extensive discussion of older studies. Summary
tables of the relevant O3 literature are included for each of the major subsections.
In summarizing the human health effects literature, changes from control are described if
statistically significant at a probability (p) value less than 0.05, otherwise trends are noted
as such.
AX6-1
-------
AX6.2 PULMONARY FUNCTION EFFECTS OF OZONE EXPOSURE IN
HEALTHY SUBJECTS
AX6.2.1 Introduction
The responses observed in young healthy nonsmoking human adults exposed to ambient O3
concentrations include decreased inspiratory capacity; mild bronchoconstriction; rapid, shallow
breathing pattern during exercise; and symptoms of cough and pain on deep inspiration.
In addition, O3 has been shown to result in airway hyperresponsiveness as demonstrated by an
increased physiological response to a nonspecific bronchoconstrictor, as well as airway injury
and inflammation assessed via bronchoalveolar lavage and biopsy. Reflex inhibition of
inspiration and consequent decrease in inspiratory capacity results in a decrease in forced vital
capacity (FVC) and total lung capacity (TLC) and, in combination with mild
bronchoconstriction, contributes to a decrease in the forced expiratory volume in 1 s (FEVj).
Given that both FEVj and FVC are subject to decrease with O3 exposures, changes in the
ratio (FEVj/FVC) become difficult to interpret and so are not discussed.
The majority of controlled human studies have investigated the effects of exposure to
variable O3 concentrations in healthy subjects performing continuous exercise (CE) or
intermittent exercise (IE) for variable periods of time. These studies have several important
limitations: (1) the ability to study only short-term, acute effects; (2) the inability to link short-
term effects with long-term consequences; (3) the use of a small number of volunteers that may
not be representative of the general population; and (4) the statistical limitations associated with
the small sample size. Nonetheless, studies reviewed in the 1996 EPA criteria document
(U.S. Environmental Protection Agency, 1996) provided a large body of data describing the
effects and dose-response characteristics of O3 as function of O3 concentration (C), minute
ventilation (VE), and duration or time (T) of exposure. In most of these studies, subjects were
exposed to O3 and to filtered air (FA [reported as 0 ppm O3]) as a control. The most salient
observations from these studies were: (1) healthy subjects exposed to O3 concentrations
>0.08 ppm develop significant reversible, transient decrements in pulmonary function if VE or
T are increased sufficiently, (2) there is a large degree of intersubject variability in physiologic
and symptomatic responses to O3 and these responses tend to be reproducible within a given
individual over a several months period, and (3) subjects exposed repeatedly to O3 over several
AX6-2
-------
days develop a tolerance to successive exposures, as demonstrated by an attenuation of
responses, which is lost after about a week without exposure.
In this section, the effects of single O3 exposures of 1- to 8-h in duration on pulmonary
function in healthy nonsmoking subjects are examined by reviewing studies that investigate:
(1) the O3 exposure-response relationship; (2) intersubject variability, individual sensitivity, and
the association between responses; and (3) mechanisms of pulmonary function responses and the
relationship between tissue-level events and functional responses. Discussion will largely be
limited to studies published subsequent to the 1996 EPA criteria document (U.S. Environmental
Protection Agency, 1996)
AX6.2.2 Acute Ozone Exposures for Up to 2 Hours
At-Rest Exposures. Exposure studies investigating the effects of O3 exposures on sedentary
subjects were discussed in the 1986 EPA criteria document (U.S. Environmental Protection
Agency, 1986). The lowest O3 concentration at which significant reductions in FVC and FEVj
were reported was 0.5 ppm (Folinsbee et al., 1978; Horvath et al., 1979). Based on the
average O3 responses in these two studies (corrected for FA responses), resting young adults
(n = 23, age = 22) exposed to 0.5 ppm O3 have a -4% reduction in FVC and a -7% reduction
FEVj. At lower O3 concentrations of 0.25 to 0.3 ppm, resting exposures did not significantly
affect lung function.
Exposures with Exercise. Collectively, the studies reviewed in the 1996 EPA criteria
document (U.S. Environmental Protection Agency, 1996) demonstrated that healthy young
adults performing moderate to heavy IE or CE of 1 to 2.5 h duration, exposed to 0.12 to
0.18 ppm O3 experienced statistically significant decrements in pulmonary function and
respiratory symptoms. As an example, 2 hr exposures to 0.12 and 0.18 ppm O3 during heavy IE
(exercise VE = 65 L/min) have resulted in FEVj decrements of 2.0 ± 0.8% (mean ± SE; n = 40)
and 9.5 ± 1.1% (n = 89), respectively (McDonnell and Smith, 1994). Significant decrements in
pulmonary function have been reported in heavily exercising healthy adults exposed for 1 h with
CE at O3 concentrations of 0.12 ppm (Gong et al., 1986), 0.16 ppm (Avol et al., 1984), and
0.2 ppm (Adams and Schelegle, 1983; Folinsbee et al., 1984).
In an attempt to describe O3 dose-response characteristics, many investigators modeled
acute responses as a function of total inhaled O3 dose (C * T x VE), which was found to be a
AX6-3
-------
better predictor of response than O3 concentration, VE, or T of exposure, alone. In an analysis of
6 studies with 1 to 2 h exposures to between 0.12 and 0.18 ppm O3 with exercise, Folinsbee et al.
(1988) reported a good correlation (r = 0.81) between total inhaled O3 dose and FEVj
decrements. For a given exposure duration, total inhaled O3 dose can be increased by increases
in C and/or VE . In exposures of fixed duration, results of several studies suggested that O3
concentration was a more important predictor of response or explained more of the variability in
response than VE (Adams et al., 1981; Folinsbee et al, 1978; Hazucha, 1987). Based on a review
of previously published studies, Hazucha (1987) noted that relative to the FEVj decrement
occurring at a given C and VE, doubling C (e.g., from 0.1 to 0.2 ppm) would increase the FEVj
decrement by 400%, whereas doubling the VE (e.g., from an exercise VE of 20 to 40 L/min)
which would only increase the FEVj decrement by 190%. Thus, C appears to have a greater
affect than VE on FEVj responses even when total inhaled O3 doses are equivalent.
New studies (i.e., not reviewed in the 1996 EPA criteria document) that provide
spirometric responses for up to 2 h exposures are summarized in Table AX6-1. Most of these
newer studies have investigated mechanisms affecting responses, inflammation, and/or effects in
diseased groups versus healthy adults, accordingly their findings may be summarized differently
in several sections of this chapter. Rather than being interested in responses due to O3 versus FA
exposures, many of the newer studies have tested the effects of a placebo versus treatment in
modulating responses to O3 exposure. Studies appearing in Table 1, but not discussed in this
section, are discussed in other sections of this chapter as indicated within the table.
McDonnell et al. (1997) pooled the results of eight studies entailing 485 healthy male
subjects exposed for 2 h on one occasion to one of six O3 concentrations (0.0, 0.12, 0.18, 0.24,
0.30, or 0.40 ppm) at rest or one of two levels of IE (VE of 25 and 35 L/min/m2 BSA). FEVj
was measured preexposure, after 1 h of exposure, and immediately postexposure. Decrements
in FEVj were modeled by sigmoid-shaped curve as a function of subject age, O3 concentration,
VE, and T. The modeled decrements reach a plateau with increasing T and dose rate (C * VE).
That is, for a given O3 concentration, exercise VE level, and after a certain length of exposure,
the FEVj response tends not to increase further with increasing duration of exposure. The
modeled FEVj responses increased with C * VE and T, decreased with subject age, but were only
AX6-4
-------
Table AX6-1. Controlled Exposure of Healthy Humans to Ozone for 1 to 2 Hours During Exercisea
X
Ozone
Concentration"
ppm Hg/m3
0.0 0
0.4 784
0.0 0
0.2 392
0.0 0
0.2 392
0.0 0
0.33 647
0.0 0
0.35 690
Exposure
Duration and
Activity
2hIE
4x15 min
on bicycle,
VE = 30 L/min
2hIE
4x15 min
at VE = 20
L/min/m2 BSA
2hIE
4x15 min
at VE = 20
L/min/m2 BSA
2hIE
4x15 min
on bicycle
ergometer
(600 kpm/min)
2.2 h IE
2 x 30 min
on treadmill
(VE <= 50 L/min)
Final 10 min rest
Exposure Number and
Conditions Gender of Subject
c Subjects Characteristics
NA 5 M, 4 F Healthy adults
25 ± 2 years old
6 M, 7 F Mild atopic
asthmatics
22 ± 0.7 years old
20 °C 8 M, 5 F Healthy NS
50% RH median age 23 years
20 °C 10 M, 12 F Healthy NS
50% RH mean age 24 years
NA 9 M Healthy NS
26.7 ± 7 years old
19-23°C 15 M Healthy NS
48-55% 25.4 ±2 years old
RH
Observed Effect(s)
Cyinduced reductions in FVC (12%, 10%) and FEV;
(13%, 11%) for asthmatic and healthy subjects.
Significant reductions in mid-flows in both asthmatics and
healthy subjects. Indomethacin pretreatment significantly
decreased FVC and FEV[ responses to O3 in healthy but not
asthmatic subjects. See Section AX6. 3. 2 and Tables AX6-3
andAX6-13.
Median O3-induced decrements of 70 mL, 190 mL, and
400 mL/s in FVC, FEVj, and FEF25.75, respectively.
Spirometric responses not predicted of inflammatory
responses. See Sections AX6. 2. 5. 2, AX6.5.6, andAX6.9.3
andTableAX6-12.
Significant O3-induced decrement in FEV[ immediately
postexposure but not significantly different from baseline
2 h later. No correlation between Clara cell protein (CC16)
and FEV; decrement. CC16 levels, elevated by O3 exposure,
remained high at 6 h postexposure, but returned to baseline
by 18 h postexposure. See Table AX6-13.
Cyinduced reductions in FVC (7%). FRC not altered by O3
exposure. Post FA, normal gradient in ventilation which
increased from apex to the base of the lung. Post O3,
ventilation shifted away from the lower-lung into middle
and upper-lung regions. The post O3 increase in ventilation
to mid-lung region was correlated with decrease in
midmaximal expiratory flow (r = 0.76, p < 0.05).
Pre- to post-O3, mean FVC and FEV; decreased by 12 and
14%, respectively. Following O3 exposure, there was a
pronounced slow phase evident in multibreath nitrogen
washouts which, on average, represented a 24% decrease
in the washout rate relative to pre-O3.
Reference
Alexis et al.
(2000)
Blomberg
etal. (1999)
Blomberg
et al. (2003)
Foster et al.
(1993)
Foster et al.
(1997)
AX6-5
-------
Table AX6-1 (cont'd). Controlled Exposure of Healthy Humans to Ozone for 1 to 2 Hours during Exercise"
Ozone
Concentration'
ppm
0.0
0.12
0.18
0.24
0.30
0.40
ug/m3
0
235
353
471
589
784
Exposure
Duration and
Activity
2 h rest or IE
(4 x 15 min
at VE = 25 or 35
L/min/m2 BSA)
Exposure Number and
Conditions Gender of
c Subjects
22 °C 485 M (each
40% RH subject
exposed at one
activity level
to one O3
concentration)
Subject
Characteristics
Healthy NS
18 to 36 years old
mean age 24 years
Observed Effect(s)
Statistical analysis of 8 experimental chamber studies
conducted between 1980 and 1993 by the U.S. EPA in
Chapel Hill, NC. Decrement in FEV; described by sigmoid-
shaped curve as a function of subject age, O3 concentration,
VE, and time. Response decreased with age, was minimally
affected by body size corrections, and was not more
Reference
McDonnell
etal. (1997)
sensitive to O3 concentration than VE. Also see Section
AX6.5.
0.4
X
Oi
784
2hIE
20 min mild-mod.
exercise,
10 min rest
NA
4 M, 5 F Healthy NS Subjects previously in Nightingale et al. (2000) study.
30 ± 3 years old Placebo-control: Immediately postexposure decrements
in FVC (9%) and FEV1 (14%) relative to pre-exposure
values. FEV; decrement only 9% at 1 hr postexposure.
By 3 h postexposure, recovery in FVC to 97% and FEVj
to 98% of preexposure values. Significant increases in
8-isoprostane at 4 h postexposure. Budesonide for
2 wk prior to exposure did not affect responses.
Montuschi
et al. (2002)
0.0 392 2hIE 20 °C
0.2 4x15 min 50% RH
at VE = 20
L/min/m2 BSA
0.4 784 2hIE NA
20 min mild-mod.
exercise,
10 min rest
6 M, 9 F Healthy adults
24 years old
9 M, 6 F Mild asthmatics
29 years old
6 M, 9 F Healthy NS
mean age ~3 1 years
O3-induced FEVj decrement (8%, healthy adults;
3% asthmatics) and PMN increase (20.6%, healthy adults;
15.2% asthmatics). Primary goal was to investigate
relationship between antioxidant defenses and O3
responses in asthmatics and healthy adults.
See Tables AX6-3 andAX6-13.
Placebo-control: O3 caused significant decrements in FEVj
(13.5%) and FVC (10%) immediately following exposure,
a small increase in MCh-reactivity, and increased PMNs and
myeloperoxidase in induced sputum at 4 h postexposure.
FEV; at 96% and FVC at 97% preexposure values at 3 h
postexposure. Budesonide for 2 wk prior to exposure did
not affect spirometric responses. See Section AX6.2.5 and
Table AX6-13.
Mudway
etal. (2001)
Stenfors
et al. (2002)
Nightingale
et al. (2000)
-------
Table AX6-1 (cont'd). Controlled Exposure of Healthy Humans to Ozone for 1 to 2 Hours during Exercise"
X
Ozone
Concentration" Exposure
Duration and
ppm Hg/m3 Activity
0.0 784 2hIE
0.4 4 x ISminat
VE=18L/min/m2
BSA
2 exposures:
25% subjects
exposed to air-air,
75% to O3-O3
0.0 784 2hIE
0.4 4 x 15min
at VE = 20
L/min/m2 BSA
0.0 490 1 h CE
0.25 VE = 30 L/min
Exposure Number and
Conditions Gender of Subject
c Subjects Charactenstics
21 °C Weak
40% RH responders Healthy NS
7M, 13F 20 to 59 years old
Strong
responders
21M,21F
20 °C Placebo group
40% RH 1 5 M, 1 F Healthy NS
mean age 27 years
Antioxidant
group
13M,2F
NA 32 M, 28 F Healthy NS
Face mask 22.6 ± 0.6 years old
exposure
Observed Effect(s)
Significant O3-induced decrements in spirometric lung
function. Young adults (<35 years) were significantly more
responsive than older individuals (>35 years). Sufentanil,
a narcotic analgesic, largely abolished symptom responses
and improved FEV[ in strong responders. Naloxone, an
opioid antagonist, did not affect O3 effects in weak
responders. See Section AX6. 2. 5.1.
Placebo and antioxidant groups had O3-induced decrements
in FEVj (20 and 14%) and FVC (13 and 10%), respectively.
Percent neutrophils and IL-6 levels in BAL fluid obtained
1 h postexposure were not different in the two treatment
groups. See Table AX6-13.
Mean O3-induced FEV; decrements of 15.9% in males and
9.4% in females (gender differences not significant). FEVj
decrements ranged from -4 to 56%; decrements >15% in
20 subjects and >40% in 4 subjects. Uptake of O3 greater in
males than females, but uptake not correlated with
spirometric responses.
Reference
Passannante
etal. (1998)
Samet et al.
(2001)
Steck-Scott
et al. (2004)
Ultman et al.
(2004)
aSee Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
'Studies conducted in exposure chamber unless otherwise indicated.
-------
minimally affected by body size corrections to VE. Fitted and experimental FEVj decrements
following a 2 h exposure at three nominal levels of VE are illustrated in Figure AX6-1 as a
function of O3 concentration. Their analysis indicated that C was marginally, but not
significantly, more important than VE in predicting FEVj response. Additionally, the McDonnell
et al. (1997) analysis revealed that some prior analyses of IE protocols may have over estimated
the relative importance of C over VE in predicting FEVj responses by considering only the VE
during exercise and ignoring the VE during periods of rest.
24-
20-
•£ 16-
E
« 12-|
o
I .H
HI
4-
0-
-4-
0.0
T = 2h
0.1
0.2 0.3
O3 (ppm)
Ve = 40 L/min (n)
31 L/min (o)
Ve = 10 L/min (A)
0.4
0.5
Figure AX6-1. FEVt decrements as a function of O3 concentration following a 2 h
exposure with incremental exercise (15 min intervals) or rest. Points are
experimental data (mean ± SE) and lines are model predictions for each
activity level. Minute ventilation (VE) represent average across intervals
of rest and exercise.
Source: McDonnell et al. (1997).
AX6-8
-------
Ultman et al. (2004) measured O3 uptake and pulmonary responses in 60 young heathy
nonsmoking adults (32 M, 28 F). A bolus technique was used to quantify the uptake of O3 as a
function of the volume into the lung which the bolus penetrated. From these measurements, the
volumetric depth at which 50% uptake occurred was calculated. This volumetric lung depth was
correlated with conducting airways volume, i.e., a greater fraction of O3 penetrated to deeper into
the lungs of individuals have larger conducting airways volumes. Two weeks after the bolus
measurements, subjects were exposed via a face mask to FA and subsequently two weeks later to
0.25 ppm O3 for 1 h with CE at a target VE of 30 L/min. The breath-by-breath uptake of O3 was
measured. There was a small but significant reduction in the breath-by-breath uptake of O3 from
90.6% on average for the first 15 minutes to 87.3% on average for the last 15 minutes of
exposure. The uptake fraction was significantly greater in males (91.4%) than females (87.1%),
which is consistent with the larger fB and smaller VT of the females than males. Uptake was
not correlated with spirometric responses. However, there was tendency for males to have
greater O3-induced FEVj decrements than females, 15.9% versus 9.4%, respectively. There was
considerable intersubject variability in FEVj decrements which ranged from -4 to 56% with
20 subjects having decrements of >15% and 4 subjects with >40% decrements (see Section
AX6.4for additional discussion regarding intersubject variability).
Few studies have measured the effect of ozone on ventilation distribution within the lung.
Foster et al. (1993) measured the effect of ozone on the vertical distribution of inspired air in the
lung using planar gamma scintigraphy. Nine healthy nonsmoking males (26.7 ± 7 years old)
were randomly exposed to FA or 0.33 ppm O3 for 2 h with IE. After each exposure session,
subjects inhaled a 2- to 4-ml bolus of xenon-133 while seated in from of a gamma camera.
Images were acquired at the end of the first inspiration and 5-6 breaths later after the xenon had
equilibrated between lung regions. Using these images, the distribution of ventilation and
volume between upper-, middle-, and lower-lung regions was quantified. Post-O3 relative to
post-FA, there were significant reductions in FVC (FA, 5.23 ± 0.5; O3, 4.88 ± 0.5 liters) and
midmaximal expiratory flow (FA, 3.82 ± 0.8; O3, 3.14 ± 0.9 liter/sec). Neither FRC nor the
distribution of volume (upper, 26.5%; middle, 42.5%; lower, 31%) between lung compartments
were affected by O3 exposure. After the FA exposure, the distribution of ventilation per unit
volume increased with progression from the apex to the base of the lung, i.e., the lower lung
regions received the greatest ventilation. Following O3 exposure, there was a significant
AX6-9
-------
reduction in the ventilation to the lower-lung and significant increases in ventilation to the
upper- and middle-lung regions relative to the FA values in 7 of the 9 subjects. The post-O3
increase in middle-lung ventilation was correlated with the decrease in midmaximal expiratory
flow (r = 0.76, p < 0.05).
Foster et al. (1997) measured the effect of ozone on ventilation distribution using a
multiple breath nitrogen washout. Fifteen healthy nonsmoking males (25.4 ± 2 years old) were
randomly exposed to FA or 0.35 ppm O3 for 2.2 h with IE. Subjects alternated between 30 min
periods of rest and treadmill exercise (VE « IQxFVC ~ 50 L/min). The final exercise period
was followed by 10 min rest period. Multiple breath nitrogen washout and spirometry were
measured pre- and immediately postexposure. At 24-h post-O3 exposure, 12 of 15 subjects
returned and completed an addition multibreath nitrogen washout maneuver. Pre- to post-O3
exposure, the mean FVC and FEVj were significantly decreased by 12 and 14%, respectively.
Exposure to FA did not appreciably affect spirometry or the multibreath nitrogen washout.
Following O3 exposure, the washout of nitrogen was delayed and resembled a two-compartment
washout, whereas pre-O3 exposure the log-linear clearance of nitrogen as a function of expired
volume resembled a single-compartment washout. The clearance rate of the slow compartment
was approximated as the slope (Ln[N2] per expired volume) of the nitrogen washout between
20% and 9% nitrogen. Post-O3, there was a pronounced slow phase evident in nitrogen washout
which, on average, represented a 24% decrease in the washout rate relative to pre-O3. Data for a
single subject (see Figure 6-1) allowed for the size of the slow compartment to be determined.
For this subject, the slow compartment represented 23% of the lung. This is fairly consistent
with Foster et al. (1993) where ventilation to the lower-lung (31% of volume) was reduced
post-O3. At 24-h post-O3, 6 of the 12 subjects who completed an additional nitrogen washout
maneuver had a delayed washout relative to the pre-O3 maneuver. This suggests a prolonged O3
effect on the small airways and ventilation distribution in some individuals.
AX6.2.3 Prolonged Ozone Exposures
Between 1988 and 1994, a number studies were completed that described the responses of
subjects exposed to relatively low (0.08 to 0.16 ppm) O3 concentrations for exposure durations
of 4 to 8 h. These studies were discussed in the 1996 criteria document (U.S. Environmental
Protection Agency, 1996) and only a select few are briefly discussed here. Table AX6-2 details
AX6-10
-------
Table AX6-2. Pulmonary Function Effects after Prolonged Exposures to Ozone"
X
Ozone Concentration11
ppm
ug/m3
Exposure
Duration
and Activity
Number and
Exposure Gender of Subject
Conditions Subjects Characteristics
Observed Effect(s)
Reference
Studies with 4 hr Exposures
0.18
0.0
0.20
0.2
0.0
0.24
353
0
392
392
0
470
4hIE
(4 x 50 min)
VE = 35L/min
4hIE
(4 x 50 min cycle
ergometry or
treadmill running
[VE = 40L/min])
4hIE
(4 x 50 min)
VE = 25 L/min/m2
BSA
4hIE
(4x15 min)
VE = 20 L/min
23 °C 2 M, 2 F Adults NS, 21
50% RH to 33 years old
20 °C FA: 1 1 M, 3 Adult NS, 19 to
50% RH F 41 years old
03: 9M, 3F
20 °C 42 M, 24 F Adults NS,
50% RH 18 to 50 years
old
24 °C 10 M Healthy NS,
40% RH 60 to 69 years
9 M COPD
59 to 71 years
FVC decreased 19% and FEVj decreased 29% in these four
pre-screened sensitive subjects.
Decrease in FVC, FEVj, VT, and SRaw and increase in fB with
O3 exposure compared with FA; total cell count and LDH
increased in isolated left main bronchus lavage and inflammatory
cell influx occurred with O3 exposure compared to FA exposure.
FEVj decreased by 18.6%; Pre-exposure methacholine
responsiveness was weakly correlated with the functional response
to O3 exposure. Symptoms were also weakly correlated with the
FEVj response (r = -0.31 to -0.37)
Healthy: small, 3.3%, decline in FEVj (p = 0.03 [not reported in
paper], paired-t on O3 versus FA pre-post FEVj). COPD: 8%
decline in FEVj (p = ns, O3 versus FA). Adjusted for exercise,
ozone effects did not differ significantly between COPD patients
and healthy subjects.
See Section AX6. 5.1.
Adams
(2000a)
Aris et al.
(1993)
Aris et al.
(1995)
Gong et al.
(1997a)
Studies with >6 hr Exposures
0.0
0.06
0.08
0.04 (mean,
peak of 0.05)
0.06 (mean,
peak of 0.09)
0.08 (mean,
peak of 0.15)
0.0
0.04
0.08
0.12
0
118
157
78
118
157
0
78
157
235
6.6 h
IE (6 x 50min)
VE = 20 L/min/m2
BSA
6.6 h
IE (6 x 50min)
VE = 20 L/min/m2
BSA
25 °C 15M, 15 F Healthy NS
40-60% RH Males
23.5 ± 3.0 yrs
Females
22.8 ± 1.2 yrs
23 °C 1 5 M, 1 5 F Healthy NS,
50% RH 22.4 ± 2.4 yrs
old
FEVj and symptom responses after 6.6 h exposure to 0.04 and
0.06 ppm not significantly different from FA. Following exposure
to 0.08 ppm, O3-induced FEVj (-6.1%, square-wave; -7.0%,
triangular) and symptom responses significantly greater than after
0.04 and 0.06 ppm exposures. Triangular exposure to 0.08 ppm
caused peak decrement in FEVj at 5.6 h of exposure, whereas peak
for square-wave exposure occurred at 6.6 h.
FEVj and total symptoms after 6.6 h exposure to 0.04 ppm not
significantly different from FA. FEVj (-6.4%) and total symptoms
significant at 6.6 h exposure to 0.08 ppm. FEVj (-15.4%) at 6.6 h
not significantly different between chamber and face mask
exposure to 0.12 ppm.
Adams (2006)
Adams (2002)
-------
Table AX6-2 (cont'd). Pulmonary Function Effects after Prolonged Exposures to Ozonea
X
ON
to
Ozone Concentration11
ppm
0.12
(a) 0.08
(b) 0.08
(mean) varied
from 0.03 to
0.15
(a) 0.08
(b) 0.30
(a) 0.12
(b)0.12
(mean) varied
from 0.07 to
0.16
(c)0.12
(mean) varied
from 0.11 to
0.13
(d)0.12
ug/m3
235
235
235
(mean)
157
588
235
235
(mean)
235
(mean)
235
Exposure Number and
Duration Exposure Gender of
and Activity Conditions Subjects
3 day-6.6h/day 23 °C 15M, 15 F
IE (6 x 50 min) 50% RH
VE= 17 L/min/m2,
20 L/min/m2
BSA, and 23
L/min/m2 BSA
6.6 h 23 °C 15 M
IE (6x50 min) 50% RH 15 F
VE = 20 L/min/m2
BSA
6.6 h 23 °C 15 M
IE (6x50 min) 50% RH 15 F
VE = 20 L/min/m2
BSA
2h
IE (4 x 15 min)
VE = 35 L/min/m2
BSA
6.6 h IE 23 °C 6 M, 6 F
(6 x 50 min) 50% RH
(a,b,c) VE = 20
L/min/m2 BSA
(d)VE=12
L/min/m2 BSA
Subject
Characteristics
Healthy NS, 18
to 3 1 years old
Healthy NS,
18 to 25 years
old
Healthy NS,
1 8 to 25 years
old
Healthy NS,
19 to 25 years
old
Observed Effect(s)
FEVj at 6.6 h decreased significantly by 9.3%, 11.7%, and 13.9%,
respectively at three different exercise VE rates, but were not
significantly different from each other. Total symptoms at the
highest VE protocol were significantly greater than for the lowest
VE protocol beginning at 4.6 h. Largest subjects (2.2 m2 BSA)
had significantly greater average FEVj decrement for the three
protocols, 18.5% compared to the smallest subjects (1.4 m2
BSA), 6.5%.
(a) FEVj decreased 6.2% after 6.6 h in square-wave exposures.
Total symptoms significantly increased at 5.6 and 6.6 h.
(b) FEVj decreased 5.6 to 6.2% after 4.6 to 6.6 h, respectively,
in varied exposure; total symptoms significantly increased also
after 4.6 to 6.6 h. No significant difference between face mask
and chamber exposures.
Significantly greater FEVj decrement (12.4%) for 2-h, 0.30 ppm
exposure than for 6.6-h, 0.08 ppm exposure (3.6%).
(a) FEVj decreased 11% at 6.6 h in square-wave exposure.
Total symptoms significant from 4.6 to 6.6 h.
(b) FEVj decreased 13% at 6.6 h; not significantly different from
square-wave exposure. Total symptoms significant from 4.6 to
6.6 h.
(c) FEVj decreased 10.3% at 6.6 h; not significantly different from
square-wave exposure. Total symptoms significant from 4.6 to
6.6 h.
(d) FEVj decreased 3.6% at 6.6 h; significantly less than for
20 L/min/m2 BSA protocols.
Reference
Adams
(2000b)
Adams
(2003a)
Adams
(2003b)
Adams and
Ollison(1997)
"See Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
-------
newer studies of healthy subjects undergoing prolonged exposures at O3 concentrations ranging
from 0.06 to 0.20 ppm. In most of these studies, statistically significant changes in pulmonary
function, symptoms, and airway responsiveness have been observed during and after exposures
to O3 concentrations of 0.08 ppm and higher. As with studies conducted at higher O3
concentrations for shorter periods of time, there is considerable intersubject variability in
response (see Section AX6.4).
Folinsbee et al. (1988) first reported the effects of a 6.6 h exposure to 0.12 ppm O3 in ten
young healthy adults (25 ± 4 yr) with quasi continuous exercise that was intended to simulate a
full workday of heavy physical labor. Except for a 35-min lunch break after 3 h, the subjects
exercised at a moderate level ( VE « 40 L/min) for 50 min of each hour. Ignoring the lunch
break during which lung function did not change appreciably, approximately linear decreases
were observed in FVC, FEVl3 and FEV25_75 with duration of O3 exposure. Correcting for FA
responses, decrements of 8.2, 14.9, and 26.8% in FVC, FEVl3 and FEV25.75 occurred as a result
of the O3 exposure. Using the same 6.6 h protocol, but a lower O3 concentration of 0.08 ppm,
Horstman et al. (1990) and McDonnell et al. (1991) observed decrements corrected for FA (and
averaged across studies) of 5, 8, and 11% in FVC, FEVl3 and FEV25_75, respectively, in 60 young
adults (25 ± 5 years old). Horvath et al. (1991) observed a 4% (p = 0.03)1 decrement in FEVj
using the forementioned protocol (i.e., 6.6 h and 0.08 ppm O3) in 11 healthy adults (37 ± 4 yr).
The smaller decrement observed by Horvath et al. (1991) versus Horstman et al. (1990) and
McDonnell et al. (1991) is consistent with response decreasing as subject age increases (see
Section AX6.5.1).
AX6.2.3.1 Effect of Exercise Ventilation Rate on FEVt Response to 6.6 h Ozone Exposure
It is well known that response to O3 exposure is a function of VE in studies of 2 h or less in
duration (See Section AX6.2.2). It is reasonable to expect that response to a prolonged 6.6-h O3
exposure is also function of VE, although quantitative analyses are lacking.
In an attempt to quantify this effect, Adams and Ollison (1997) exposed 12 young adults
to an average O3 concentration of 0.12 ppm for 6.6 h at varied exercise VE . They observed a
mean FEVj decrements of 10 to 11% in two protocols having a mean exercise VE of 33 L/min
'Based on two-tailed paired t-test of data in Table III of Horvath et al. (1991).
AX6-13
-------
and a 14% decrement in a protocol with a mean exercise VE of 36 L/min. These FEVj
decrements were significantly greater than the average decrement of 3.6% (not significantly
different from FA response) observed at an exercise VE of only 20 L/min. In a subsequent study
of 30 healthy adults (Adams, 2000b), the effect of smaller exercise VE differences on pulmonary
function and symptoms responses to 6.6 h exposure to 0.12 ppm O3 was examined. FEVj
decrements of 9.3, 11.7, and 13.9% were observed for the exercise VE of 30.2, 35.5, and
40.8 L/min, respectively. Along with the tendency for FEVj responses to increase with VE, total
symptoms severity was found to be significantly greater at the end of the highest VE protocol
relative to the lowest VE protocol. Although the FEVj responses were not significantly different
from each other, the power of the study to detect differences between the three VE levels was not
reported and no analysis was performed using all of the data (e.g., a mixed effects model). Data
from the Adams and Ollison (1997) and Adams (2000b) studies are illustrated in Figure AX6-2
with data from three older studies. There is a paucity of data below an exercise VE of 30 L/min.
Existing data for exposure to 0.12 ppm O3 suggest that FEVj responses increase with increasing
exercise VE until at least 35 L/min.
AX6.2.3.2 Exercise Ventilation Rate as a Function of Body/Lung Size on FEVt Response
to 6.6 h Ozone Exposure
Typically, with the assumption that the total inhaled O3 dose should be proportional to the
lung size of each individual, exercise VE in 6.6 h exposures has been set as a multiple of body
surface area (BSA) (McDonnell et al., 1991) or as a product of eight times FVC (Folinsbee et al.,
1988; Frank et al., 2001; Horstman et al., 1990). Utilizing previously published data, McDonnell
et al. (1997) developed a statistical model analyzing the effects of O3 concentration, VE, duration
of exposure, age, and body and lung size on FEVj response. They concluded that any effect of
BSA, height, or baseline FVC on percent decrement in FEVj in this population of 485 young
adults was small if it exists at all. This is consistent with Messineo and Adams (1990), who
examined pulmonary function responses in young adult women having small (n = 14) or large
(n= 14) lung sizes (mean FVC of 3.74 and 5.11 L, respectively). Subjects were exposed to
0.30 ppm O3 for 1 h with CE (VE = 47 L/min). There was no significant difference between the
AX6-14
-------
>
LU
25
~ 20-
I 1=
10H
5-
" Adams (2000b)
D Adams and Ollison (1997)
A Folinsbeeetal. (1988)
A Folinsbeeetal. (1994)
• Horstmanetal. (1990)
15 20 25 30 35
Minute Ventilation (L/min)
40
45
Figure AX6-2. Average FEVt decrements (±SE) for prolonged 6.6 h exposures to 0.12 ppm
•
O3 as a function of exercise VE. Since age affects response to O3 exposure,
selected studies had subjects with mean ages between 22 and 25 years.
FEVt decrements were calculated as mean O3 responses minus mean air
responses. The SE bars illustrate variability in FEVt responses (pre minus
post) on the O3 exposure day in all cases except for Folinsbee et al. (1994),
where post O3 exposure variability is illustrated. In one case, the SE for VE
of 33 L/Min (10.3% decrement) was taken as the SE of data from protocol
with VE of 33 L/min (11% decrement). All studies used a constant
0.12 ppm O3 exposure except two (*) which used 0.115 ppm O3 for hours
1-2 and 5-6 and 0.13 ppm O3 for hours 3-4 of exposure.
group FEVj decrements (22.1 and 25.6% for small and large lung, respectively). In addition,
Messineo and Adams (1990) also did a retrospective analysis of 36 young adult males who each
had completed similar 1 h exposures to 0.30 ppm O3 with CE (VE « 70 L/min) and found lung
size was not related to FEVj response.
Adams (2000b) studied a group of 30 young adult men and women exposed to
0.12 ppm O3 for 6.6 h on three occasions while exercising 50 min of each hour at one of three
different VE levels (viz., 17, 20, and 23 l/min/m2BSA). Their postexposure FEVj responses
were regressed as a function of BSA (which was directly related to the absolute amount of VE
AX6-15
-------
during exercise and, thus, primarily responsible for individual differences in total inhaled O3
dose). The slope was significantly different from zero (p = 0.01), meaning that the smallest
subjects, who had the lowest exercise VE (« 26 L/min), had a lower FEVj decrement (-5%)
than the largest subjects (-17%), whose exercise VE was -44 L/min. This relationship was not
a gender-based difference, as the mean female's FEVj decrement was -11.2%, which was not
significantly different from the male's -12.2% mean value. Similarly, when total symptoms
severity response was regressed against BSA, the slope was significantly different than zero
(p = 0.0001), with lower values for smaller subjects than for larger subjects. Results of this
study suggest that for the O3 concentration and exposure duration used, responses are more
closely related to VE than VE normalized to BSA. Further, this observation is in agreement with
McDonnell et al. (1997), who observed no evidence that measurements of lung or body size
were significantly related to FEVj response in 2 h IE exposures. These authors state that the
absence of an observed relationship between FEVj response and BSA, height, or FVC may be
due to the poor correlation between these variables and airway caliber (Collins et al., 1986;
Martin et al., 1987). Also, the O3 dosimetry study of Bush et al. (1996) indicated that
normalization of the O3 dose would be more appropriately applied as a function of anatomic
dead space.
AX6.2.3.3 Comparison of 6.6 h Ozone Exposure Pulmonary Responses to Those Observed
in 2 h Intermittent Exercise Ozone Exposures
It has been shown that greater O3 concentration (Horstman et al., 1990) and higher VE
(Adams, 2000b) each elicit greater FEVj response in prolonged, 6.6-h exposures, but data on the
relative effect of O3 concentration, VE, and T in prolonged exposures are very limited and have
not been systematically compared to data from shorter (<2-h) exposures. In a recent study
(Adams, 2003b), the group mean FEVj response for a 2-h IE exposure to 0.30 ppm O3 was
-12.4%, while that for a 6.6-h exposure to 0.08 ppm O3 was -3.5%. The total inhaled O3 dose
(as the simple product of C x T x VE ) was 1358 ppm-L for the 2-h exposure and 946 ppm-L for
the 6.6-h exposure. Thus, the FEVj decrement was 3.5 times greater and the total inhaled O3
dose was 1.44 times greater for the 2-h exposure compared to the 6.6-h exposure. This
difference illustrates the limitations of utilizing the concept of total O3 dose for comparisons
between studies of vastly different exposure durations.
AX6-16
-------
Adams (2003b) also examined whether prolonged 6.6 h exposure to a relatively low O3
concentration (0.08 ppm) and the 2-h IE exposure at a relatively high O3 concentration (0.30
ppm) elicited consistent individual subject effects, i.e, were those most or least affected in one
exposure also similarly affected in the other? Individual subject O3 exposure reproducibility was
first examined via a regression plot of the postexposure FEVj response to the 6.6-h chamber
exposure as a function of postexposure FEVj response to the 2-h chamber exposure. The R2 of
0.40, although statistically significant, was substantially less than that observed in a comparison
of individual FEVj response to two 2-h IE exposures by chamber and face mask, respectively
(R2 = 0.83). The Spearman rank order correlation for the chamber 6.6-h and chamber 2-h
exposure comparison was also substantially less (0.49) than that obtained for the two 2-h
exposures (0.85). The primary reason for the greater variability in the chamber 6.6-h exposure
FEVj response as a function of that observed for the two 2-h IE exposures is very likely related
to the increased variability in response upon repeated exposure to O3 concentrations lower than
0.18 ppm (R = 0.57, compared to a mean R of 0.82 at higher concentrations) reported by
McDonnell et al. (1985a). This rationale is supported by the lower R (0.60) observed by
Adams (2003b) for the FEVj responses found in 6.6 h chamber and face mask exposures to
0.08 ppm O3, compared to an R of 0.91 observed for responses found for the 2 h chamber and
face mask exposures to 0.30 ppm O3.
AX6.2.4 Triangular Ozone Exposures
To further explore the factors that determine responsiveness to O3, Hazucha et al. (1992)
designed a protocol to examine the effect of varying, rather than constant, O3 concentrations.
In this study, subjects were exposed to a constant level of 0.12 ppm O3 for 8 h and to an O3 level
that increased linearly from 0 to 0.24 ppm for the first 4 h and then decreased linearly from
0.24 to 0 over the second 4 h of the 8 h exposure (triangular concentration profile). Subjects
performed moderate exercise (VE -40 L/min) during the first 30 minutes of each hour. The total
inhaled O3 dose (i.e., C x T x VE) for the constant versus the triangular concentration profile was
almost identical. FEVj responses are illustrated in Figure AX6-3. With exposure to the constant
0.12 ppm O3, FEVj declined approximately 5% by the fifth hour of exposure and then remained
at that level. This observation clearly indicates a response plateau as suggested in other
AX6-17
-------
4.5
4.4
4.3
HI
U-
c
(0
4.2
4.1
4.0
3.9
0 ppm
0.12 ppm
Constant
Variable
246
Exposure Duration (h)
Figure AX6-3. The forced expiratory volume in 1 s (FEVt) is shown in relation to exposure
duration (hours) under three exposure conditions. Subjects exercised
(minute ventilation « 40 L/min) for 30 min during each hour; FEVt was
measured at the end of the intervening rest period. Standard error of the
mean for these FEVt averages (not shown) ranged from 120 to 150 mL.
Source: Hazuchaetal. (1992).
prolonged exposure studies (Horstman et al., 1990; McDonnell et al., 1991). However, with the
triangular O3 concentration profile after a minimal initial response over the first 3 h, Hazucha
et al. (1992) observed a substantial change in FEVj corresponding to the higher average O3
concentration that reached a maximum decrement of 10.3% at 6 h. Despite 2 h of continued
exposure to a lower O3 concentration (0.12 to 0.00 ppm, mean = 0.06 ppm), FEVj improved and
was only reduced by 6.3% (relative to the preexposure FEVj) at the end of the 8-h exposure.
The authors concluded that total inhaled O3 dose (C * VE * T) was not a sufficient index of O3
exposure and that, as observed by others (Adams et al., 1981; Folinsbee et al., 1978; Hazucha,
1987; Larsen et al., 1991), O3 concentration appears to be more important in determining
exposure effects than is either duration or the volume of air breathed during the exposure.
AX6-18
-------
However, it should be noted that the mean O3 concentration for Hazucha et al.'s triangular
exposure profile was 0.12 ppm at 4 h, 0.138 ppm at 5 h, 0.14 ppm at 6 h, and 0.133 ppm at 7 h,
before falling to 0.12 ppm at 8 h. The FEVj responses of the last 4 hours (Figure AX6-3) follow
a pattern similar to the total mean O3 concentration over the same time period.
It has become apparent that laboratory simulations of air-pollution risk-assessment need to
employ O3 concentration profiles that more accurately mimic those encountered during summer
daylight ambient air pollution episodes (Adams and Ollison, 1997; Lefohn and Foley, 1993;
Rombout et al., 1986). Neither square-wave O3 exposures or the one 8-h study by Hazucha et al.
(1992) that utilized a triangular shaped varied O3 exposure described above closely resembles
the variable diurnal daylight O3 concentration pattern observed in many urban areas experiencing
air-pollution episodes (Lefohn and Foley, 1993). Recently, 6.6 h less abrupt triangular O3
exposure profiles at lower concentrations more typical of outdoor ambient conditions have been
examined (Adams, 2003a, 2006; Adams and Ollison, 1997).
Using a face-mask inhalation system, Adams and Ollison (1997) observed no significant
differences in postexposure pulmonary function responses or symptoms between the 6.6-h,
0.12 ppm O3 square-wave exposure and those observed for a triangular O3 profile in which
concentration was increased steadily from 0.068 ppm to 0.159 ppm at 3.5 h and then decreased
steadily to 0.097 ppm at end exposure. Further, no attenuation in FEVj response during the last
2 h was observed in either the 6.6 h square-wave or the triangular exposures.
In a subsequent study (Adams, 2003a), no significant difference was observed in
pulmonary function responses or symptoms between face-mask and chamber exposure systems
either for a 6.6-h, 0.08 ppm O3 square-wave profile or for the triangular O3 exposure beginning
at 0.03 ppm, increasing steadily to 0.15 ppm in the fourth hour, and decreasing steadily to
0.05 ppm at 6.6 h (mean = 0.08 ppm). For the chamber-exposure comparison, postexposure
values for FEVj and symptoms were not significantly different from the responses for the
square-wave 0.08 ppm O3 exposure. However, analysis showed that FEVj response for the
square-wave protocol did not become statistically significant until the 6.6-h postexposure value,
while that for the triangular exposure protocol was significant at 4.6 h (when O3 concentration
was 0.15 ppm). Earlier significant FEVj responses for the triangular protocol were accompanied
by significant increases in symptoms at 4.6 h, which continued on through the fifth and sixth
AX6-19
-------
hours when the mean O3 concentration was 0.065 ppm. Symptoms for the square-wave 0.08
ppm exposure did not become statistically significant until 5.6 h.
The FEVj responses during the last two hours of the triangular exposure by Adams (2003 a)
did not decrease as dramatically as in the Hazucha et al. (1992) study (Figure AX6-3). The most
probable reason for differences in the triangular O3 profile observations of Hazucha et al. (1992)
and those of Adams (2003a) is that the increase and decrease in Hazucha et al.'s study (i.e., 0 to
0.24 ppm and back to 0) encompassed a much greater range of O3 concentrations than those used
by Adams (2003a), viz., 0.03 ppm to 0.15 ppm from 0 to 3.6 h, then decreasing to 0.05 ppm for
the final hour of exposure. Nonetheless, the greatest FEVj decrement was observed at 6 h of
Hazucha et al.'s 8 h triangular exposure (Figure AX6-3) corresponding to the time when total
mean O3 concentration was highest (0.14 ppm), with a very similar response at 7 h when total
mean O3 concentration was 0.138 ppm.
In a recent study, Adams (2006) employed three square-wave and three triangular chamber
exposure profiles. Ozone concentrations for the 6.6-h exposures were 0.00, 0.06, and 0.08 ppm
for square-wave profile and averaged 0.04, 0.06, and 0.08 ppm for the triangular profile. The
pattern and magnitude of hourly changes in FEVj and symptoms during filtered air (0.0 ppm)
and 0.04 ppm O3 exposure were essentially the same. Similarly, there was no significant
difference in FEVj and symptomatic response between the square-wave and triangular 0.06 ppm
exposures. While the 6.6-h postexposure responses to the triangular exposure (mean O3
concentration of 0.08 ppm) did not differ significantly from those observed in the square-wave
exposure, FEVj and symptoms were significantly different from preexposure at 4.6 h (when the
O3 concentration was 0.15 ppm) in the triangular exposure, but not until 6.6 h in the square-wave
exposure. These observations have confirmed and expanded findings of an earlier study
(Adams, 2003a) showing a clear divergence in the peak and hourly responses in FEVj between
the square-wave and triangular concentration profiles at the 0.08 ppm O3 level. At the lower
0.06 ppm O3 concentration, however, no temporal pattern differences in FEVj responses could
be discerned between the square-wave and triangular exposure profiles. The author concluded
that the results support previous evidence that O3 concentration has a greater singular effect in
the total inhaled O3 dose than do VE and exposure duration.
Adams (2006) observed that at O3 concentrations of 0.06 ppm and below differences in
prolonged exposure profiles do not induce statistically different hourly responses.
AX6-20
-------
Decrements in FEVj during square-wave O3 exposures between 0.08 to 0.12 ppm tend to
increase with time of exposure (i.e., with steadily increasing total inhaled dose), whereas FEVj
decrements during triangular exposures (Hazucha et al., 1992; Adams, 2003a) occurred 1 to 2 h
after the peak O3 concentration and 1 h to 2 h before the maximal total O3 inhaled dose occurred
at the end of exposure. This difference, especially because O3 concentration profiles during
summer daylight air-pollution episodes rarely mimic a square-wave, implies that triangular O3
exposure profiles most frequently observed during summer daylight hours merit further
investigation. These three studies suggest that depending upon the profile of the exposure,
the triangular exposure can potentially lead to higher FEVj responses than the square wave
exposures at the overall equivalent ozone dose.
AX6.2.5 Mechanisms of Pulmonary Function Responses
Inhalation of O3 for several hours while physically active elicits both subjective respiratory
tract symptoms and acute pathophysiologic changes. The typical symptomatic response
consistently reported in studies is that of tracheobronchial airway irritation. This is accompanied
by decrements in lung capacities and volumes, bronchoconstriction, airway hyperresponsiveness,
airway inflammation, immune system activation, and epithelial injury. The severity of
symptoms and the magnitude of response depend on inhaled dose, O3 sensitivity of an individual
and the extent of tolerance resulting from previous exposures. The development of effects is
time dependent during both exposure and recovery periods with considerable overlap of evolving
and receding effects.
Exposure to O3 initiates reflex responses manifested as a decline in spirometric lung
function parameters (1FVC, IFEVl3 1FEF25.75), bronchoconstriction (TSRaw) and altered
breathing pattern (1VT, t fB), which becomes more pronounced as exposure progresses and
symptoms of throat irritation, cough, substernal soreness and pain on deep inspiration develop.
The spirometric lung function decline and the severity of symptoms during a variable (ramp)
exposure profile seem to peak a short time (about 1 to 2 h) following the highest concentration
of O3 (Hazucha et al., 1992; Adams, 2003a). Exposure to a uniform O3 concentration profile
elicits the maximum spirometric response at the end of exposure (Hazucha et al., 1992; Adams,
2003a). Regardless of exposure concentration profile, as the exposure to O3 progresses, airway
inflammation begins to develop and the immune response at both cellular and subcellular level is
AX6-21
-------
activated. Airway hyperreactivity develops slower than pulmonary function effects, while
neutrophilic inflammation of the airways develops even more slowly and reaches the maximum
3 to 6 h postexposure. The cellular responses (e.g., release of immunoregulatory cytokines)
appear to still be active as late as 20 h postexposure (Torres et al., 2000). Following cessation of
exposure, the recovery in terms of breathing pattern, pulmonary function and airway
hyperreactivity progresses rapidly and is almost complete within 4 to 6 hours in moderately
responsive individuals. Persisting small residual lung function effects are almost completely
resolved within 24 hours. Following a 2 h exposure to 0.4 ppm O3 with IE, Nightingale et al.
(2000) observed a 13.5% decrement in FEVj. By 3 h postexposure, however, only a 2.7% FEVj
decrement persisted. As illustrated in Figure AX6-4, a similar postexposure recovery in FVC
was observed. In hyperresponsive individuals, the recovery takes longer and as much as
48 hours to return to baseline values. More slowly developing inflammatory and cellular
changes persist for up to 48 hours. The time sequence, magnitude and the type of responses of
this complex series of events, both in terms of development and recovery, indicate that several
mechanisms, activated at different times of exposure, must contribute to the overall lung
function response (U.S. Environmental Protection Agency, 1996).
AX6.2.5.1 Pathophysiologic Mechanisms
Breathing pattern changes
Human studies consistently report that inhalation of O3 alters the breathing pattern without
significantly affecting minute ventilation. A progressive decrease in tidal volume and a
"compensatory" increase in frequency of breathing to maintain steady minute ventilation during
exposure suggests a direct modulation of ventilatory control. These changes parallel a response
of many animal species exposed to O3 and other lower airway irritants (Tepper et al., 1990).
Although alteration of a breathing pattern could be to some degree voluntary, the presence of the
response in animals and the absence of perception of the pattern change by subjects, even before
appearance of the first subjective symptoms of irritation, suggests an involuntary reflex
mechanism.
Direct recording from single afferent vagal fibers in animals convincingly demonstrated
that bronchial C-fibers and rapidly adapting receptors are the primary vagal afferents responsible
for O3-induced changes in ventilatory rate and depth (Coleridge et al., 1993; Hazucha and
AX6-22
-------
a.
4.0-,
3.5-
LJJ
b.
o
2.5-
2.0-
Pre 01234
Time from Exposure (hours)
4.5-|
4.0-
3.5-
3.0-
24
Pre 0 1 2 3 4 24
Time from Exposure (hours)
Figure AX6-4a,b. Recovery of spirometric responses following a 2 h exposure to 0.4 ppm
O3 with IE. Immediately postexposure, there were significant
decrements (**p < 0.001, ***p < 0.0005) in FVC (10%) and FEVt
(13.5%) compared to preexposure values. At 3 h postexposure, FVC
and FEVj were at 96 and 97% of preexposure values, respectively.
Adapted from Nightingale et al. (2000).
Sant'Ambrogio, 1993). In spontaneously breathing dogs, an increase in VT/T; (T; decreased more
than VT) was attributed to an increased inspiratory drive due to stimulation of rapidly adapting
receptors and bronchial C-fibers by O3 (Schelegle et al., 1993). Folinsbee and Hazucha (2000)
also observed similar changes in VT/T; and other breath-timing parameters in humans exposed
to O3 implying activation of the same mechanisms. They also reported that Pm0A (pressure at
AX6-23
-------
mouth at 0.1 sec of inspiration against a transiently occluded mouthpiece which is considered an
index of inspiratory drive) increased during controlled hypercapnia without a change in the slope
of Pm0 j versus pCO2 relation suggesting that the primary mechanism is an increased inspiratory
drive. Since no significant within-individual differences in ventilatory response to CO2 between
air exposure and O3 exposure were found, the CO2 chemoreceptors did not modulate the
response. Therefore, the principal peripheral mechanism modulating changes in breathing
pattern appears to be direct and indirect stimulation of lung receptors and bronchial C-fibers
by O3 and/or its oxidative products. The activity of these afferents, centrally integrated with
input from other sensory pathways, drives the ventilatory controller, which determines the depth
and the frequency of breathing.
The potential modulation of breathing pattern by activation of sensory afferents located in
extrathoracic airways by O3 has not yet been studied in humans. Laboratory animal studies have
shown that the larynx, pharynx, and nasal mucosa are densely populated by free-ending,
unmyelinated sensory afferents resembling nociceptive C-fibers (Spit et al., 1993; Sekizawa and
Tsubone, 1994). They are almost certainly stimulated by O3 and likely contribute to overall
ventilatory and symptomatic responses. Nasal only exposure of rats produced O3-induced
changes in breathing pattern that are similar to changes found in humans (Kleinman et al., 1999).
Symptoms and lung function changes
As already discussed, in addition to changes in ventilatory control, O3 inhalation by
humans will also induce a variety of symptoms, reduce vital capacity (VC) and related functional
measures, and increase airway resistance. Hazucha et al. (1989) postulated that a reduction of
VC by O3 is due to a reflex inhibition of inspiration and not due to a voluntary reduction of
inspiratory effort. Recently, Schelegle et al. (2001) convincingly demonstrated that a reduction
of VC due to O3 is indeed reflex in origin and not a result of subjective discomfort and
consequent premature voluntary termination of inspiration. They reported that inhalation of an
aerosolized topical anesthetic tetracaine substantially reduced if not abolished O3-induced
symptoms that are known to be mediated in part by bronchial C-fibers. Yet, such local
anesthesia of the upper airway mucosa had a minor and irregular effect on pulmonary function
decrements and tachypnea, strongly supporting neural mediation, i.e., stimulation of both
AX6-24
-------
bronchial and pulmonary C-fibers, and not voluntary inhibition of inspiration (due to pain) as the
key mechanism.
The involvement of nociceptive bronchial C-fibers modulated by opioid receptors in
limiting maximal inspiration and eliciting subjective symptoms in humans was studied by
Passannante et al. (1998). The authors hypothesized that highly variable responses among
individuals might reflect the individual's inability or unwillingness to take a full inspiration.
Moreover, development of symptoms of pain on deep inspiration, cough and substernal soreness
suggested that nociceptive mechanism(s) might be involved in O3-induced inhibition of maximal
inspiration. If this were so, pain suppression or inhibition by opioid receptor agonists should
partially or fully reverse symptoms and lung functional impairment. Subjects for this study were
pre-screened with exposure to 0.42 ppm O3 and classified either as "weak" (FEVj >95% of
preexposure value), "strong" (FEVj < 85% of preexposure value), or "moderate" responders.
Sixty two (28 M, 34 F) healthy volunteers (18 to 59 yrs old), known from the previous screening
to be "weak" (n = 20) or "strong" (n = 42) O3-responders, participated in this double-blind
crossover study. Subjects underwent either two 2 h exposures to air, or two 2 h exposures to
0.42 ppm O3, with 15 min IE at 17.5 1/min/m2 BSA. Immediately following postexposure
spirometry the "weak" responders were given (in random order) either the potent opioid receptor
antagonist naloxone (0.15 mg/kg) or saline, while "strong" responders received (in random
order) either the potent, rapid-acting opioid agonist and analgesic sufentanil (0.2 jig/kg), or
physiologic saline administered through an indwelling catheter. Administration of saline or
naloxone had no significant effect on the relatively small decrements in FEVj observed in
"weak" responders. However, as hypothesized, sufentanil rapidly reversed both the O3-induced
symptomatic effects and spirometric decrements (FEV^ p < 0.0001) in "strong" responders
(Figure AX6-5). All the same, the reversal was not complete and the average post-sufentanil
FEVj remained significantly below (-7.3%) the preexposure value suggesting involvement of
non-opioid receptor modulated mechanisms as well. Uneven suppression of symptoms has
implied involvement of both A-6 and bronchial C-fibers. The plasma p-endorphin (a potent
pain suppressor) levels, though substantially elevated immediately postexposure and post-drug
administration, were not related to individuals' O3 responsiveness. These observations have
demonstrated that nociceptive mechanisms play a key role in modulating O3-induced inhibition
of inspiration. Moreover, these findings are consistent with and further support the concept that
AX6-25
-------
LJJ
0.5
1 1.5
Time (hours)
2.5
Figure AX6-5. Plot of the mean FEVt (% baseline) vs. time for ozone exposed cohorts.
Solid lines represent data for "strong" males (n = 14; solid squares) and
females (n = 15; solid circles) that received sufentanil and dotted lines
represent data for the same cohorts after receiving saline. Dashed lines
represent data for "weak" males (n = 5; open squares) and females (n = 10;
open circles) that received naloxone and dot-dash lines represent data for
the same cohorts after receiving saline. The arrow denotes the time of drug
administration (-2.1 hrs). Vertical bars associated with the symbols are
one-sided SEM.
Source: Adapted from Passannante et al. (1998).
the primary mechanism of O3-induced reduction in inspiratory lung function, is an inhibition of
inspiration elicited by stimulation of the C-fibers. The absence of effect of naloxone in "weak"
responders shows that the weak response is not due to excessive endorphin production in those
individuals. However, other neurogenic mechanisms not modulated by opioid receptors may
have some though limited role in inspiratory inhibition.
AX6-26
-------
Airway hyperreactivity
In addition to limitation of maximal inspiration and its effects on other spirometric
endpoints, activation of airway sensory afferents also plays a role in receptor-mediated
bronchoconstriction and an increase in airway resistance. Despite this common mechanism,
post-O3 pulmonary function changes and either early or late airway hyperresponsiveness (AHR)
to inhaled aerosolized methacholine or histamine are poorly correlated either in time or
magnitude. Fentanyl and indomethacin, the drugs that have been shown to attenuate O3-induced
lung function decrements in humans, did not prevent induction of AHR when administered to
guinea pigs prior to O3 exposure (Yeadon et al., 1992). Neither does post-O3 AHR seem to be
related to airway baseline reactivity. These findings imply that the mechanisms are either not
related or are activated independently in time. Animal studies (with limited support from human
studies) have suggested that an early post-O3 AHR is, at least in part, vagally mediated (Freed,
1996) and that stimulation of C-fibers can lead to increased responsiveness of bronchial smooth
muscle independently of systemic and inflammatory changes which may be even absent (load
et al., 1996). In vitro study of isolated human bronchi have reported that O3-induced airway
sensitization involves changes in smooth muscle excitation-contraction coupling (Marthan,
1996). Characteristic O3-induced inflammatory airway neutrophilia which at one time was
considered a leading AHR mechanism, has been found in a murine model to be only
coincidentally associated with AHR, i.e., there was no cause and effect relationship (Zhang et al.,
1995). However, this observation does not rule out involvement of other cells such as
eosinophils or T-helper cells in AHR modulation. There is some evidence that release of
inflammatory mediators by these cells can sustain AHR and bronchoconstriction. In vitro and
animal studies have also suggested that airway neutral endopeptidase activity can be a strong
modulator of AHR (Marthan et al., 1996; Yeadon et al., 1992). Late AHR observed in some
studies is plausibly due to a sustained damage of airway epithelium and continual release of
inflammatory mediators (Foster et al., 2000). Thus, O3-induced AHR appears to be a product of
many mechanisms acting at different time periods and levels of the bronchial smooth muscle
signaling pathways. [The effects ofO3 on AHR are described in Section AX6.8. ]
AX6-27
-------
AX6.2.5.2 Mechanisms at a Cellular and Molecular Level
Stimulation of vagal afferents by O3 and reactive products, the primary mechanism of lung
function impairment is enhanced and sustained by what can be considered in this context to be
secondary mechanisms activated at a cellular and molecular level. The complexity of these
mechanisms is beyond the scope of this section and the reader is directed to Section AX6.9 of
this chapter for greater details. A comprehensive review by Mudway and Kelly (2000) discusses
the cellular and molecular mechanisms of O3-induced pulmonary response in great detail.
Neurogenic airway inflammation
Stimulation of bronchial C-fibers by O3 not only inhibits maximal inspiration but, through
local axon reflexes, induces neurogenic inflammation. This pathophysiologic process is
characterized by release of tachykinins and other proinflammatory neuropeptides. Ozone
exposure has been shown to elevate C-fiber-associated tachykinin substance P in human
bronchial lavage fluid (Hazbun et al. 1993) and to deplete neuropeptides synthesized and
released from C-fibers in human airway epithelium rich in substance P-immunoreactive axons.
Substance P and other transmitters are known to induce granulocyte adhesion and subsequent
transposition into the airways, increase vascular permeability and plasma protein extravasation,
cause bronchoconstriction, and promote mucus secretion (Solway and Left, 1991). Although the
initial pathways of neurogenic, antigen-induced, and generally immune-mediated inflammation
are not the same, they eventually converge leading to further amplification of airway
inflammatory processes by subsequent release of cytokines, eicosanoids, and other mediators.
Significantly negative correlations between O3-induced leukotriene (LTC4/D4/E4) production and
spirometric decrements (Hazucha et al., 1996), and an increased level of postexposure PGE2, a
mediator known to stimulate bronchial C-fibers, show that these mediators play an important
role in attenuation of lung function due to O3 exposure (Mohammed et al., 1993; Hazucha et al.,
1996). Moreover, because the density of bronchial C-fibers is much lower in the small than
large airways, the reported post-O3 dysfunction of small airways assessed by decrement
in FEF25.75 (Weinman et al., 1995; Frank et al., 2001) may be due in part to inflammation.
Also, because of the relative slowness of inflammatory responses as compared to reflex
effects, O3-triggered inflammatory mechanisms are unlikely to initially contribute to progressive
lung function reduction. It is plausible, however, that when fully activated, they sustain and
AX6-28
-------
possibly further aggravate already impaired lung function. Indeed, a prolonged recovery of
residual spirometric decrements following the initial rapid improvement after exposure
termination could be due to slowly resolving airway inflammation. Bronchial biopsies
performed 6 h postexposure have shown that O3 caused a significant decrease in
immunoreactivity to substance P in the submucosa (Krishna et al., 1997a). A strong negative
correlation with FEVj also suggests that the release of substance P may be a contributing
mechanism to persistent post-O3 bronchoconstriction (Krishna et al., 1997a). Persistent
spirometry changes observed for up to 48 h postexposure could plausibly be sustained by
the inflammatory mediators, many of which have bronchoconstrictive properties (Blomberg
etal., 1999).
AX6.3 PULMONARY FUNCTION EFFECTS OF OZONE EXPOSURE IN
SUBJECTS WITH PREEXISTING DISEASE
This section examines the effects of O3 exposure on pulmonary function in subjects with
preexisting disease by reviewing O3 exposure studies that utilized subjects with (1) chronic
obstructive pulmonary disease (COPD), (2) asthma, (3) allergic rhinitis, and (4) ischemic heart
disease. Studies of subjects with preexisting disease exposed to O3, published subsequent to or
not included in the 1996 Air Quality Criteria Document (U.S. Environmental Protection Agency,
1996), are summarized in Table AX6-3. Studies examining increased airway responsiveness
after O3 exposure are discussed in Section AX6.8.
AX6.3.1 Subjects with Chronic Obstructive Pulmonary Disease
Five studies of O3-induced responses in COPD patients were available for inclusion in the
1996 criteria document (U.S. Environmental Protection Agency, 1996). The COPD patients in
these studies were exposed during light IE (4 studies) or at rest (1 study) for 1 to 2 hours to O3
concentrations between 0.1 and 0.3 ppm. None of theses studies found significant O3-induced
changes in pulmonary function. Of the four studies examining arterial oxygen saturation, two
reported small but statistically significant O3-induced decreases in the COPD patients. These
limited data suggest COPD patients experience minimal O3-induced effects for 0.3 ppm O3
AX6-29
-------
Table AX6-3. Ozone Exposure in Subjects with Preexisting Disease"
X
ON
OJ
O
Ozone
Concentration'
ppm
Hg/ni3
Exposure Duration
and Activity
Exposure Number and
Condition Gender of Subjects
Subject Characteristics
Observed Effect(s)
Reference
Subjects with Chronic Obstructive Pulmonary or Heart Disease
0.0
0.24
0.3
0
472
589
4hIE
15 min exercise
15 min rest
VE ~ 20 L/min
3hIE
VE ~ 30 L/min
24 °C 9 M
40% RH 10 M
22 °C 10 M
50% RH
6M
COPD patients
Age-matched healthy NS
All subjects 59-71 years old
Hypertension
42-61 years old
Healthy
41-49 years old
No significant changes in FEVj, FVC, or SRaw due to
ozone in COPD patients. Equivocal SaO2 decrement
during 2nd and 3rd hours of ozone exposure in COPD
patients. Adjusted for exercise, ozone effects did not
differ significantly between COPD patients and healthy
subjects.
O3-induced FEV; decrements of 6.7 and 7.6% in healthy
and hypertensive subjects, respectively. Significant
O3-induced reductions in alveolar-arterial oxygen tension
in both groups. No significant changes in cardiac
enzymes or ECG telemetry.
Gong et al.
(1997a)
Gong and
Tierney (1995)
Gong et al.
(1998)
Subjects with Allergic Rhinitis
0.0
0.2
0
392
IhCE
at VE = 25 L/min/m2
BSA
20 °C 13 M, 1 F
50% RH
Dust mite sensitized asthmatics
mean age 29 ± 5 years
FEVj decrement following O3 of 10% not significantly
different from the 4% decrement following FA. Subjects
received dust mite antigen challenge at 0.5 h FA and O,
Chen et al.
(2004)
0.0 0 1 h CE 20 °C 13 M, 1 F
0.2 392 at VE = 25 L/min/m2 50% RH
BSA
0.125 245 3hIE 27 °C 5 F, 6 M
0.250 490 (10 min rest, 15 min 50 % RH
exercise on bicycle) 6 F, 16 M
VE = 30 L/min
0.125 245 3hIEx4days
Dust mite sensitized asthmatics
mean age 29 ± 5 years
Mild bronchial asthma
20-53 years old
Allergic rhinitis
19-48 years old
FEVj decrement following O3 of 10% not significantly
different from the 4% decrement following FA. Subjects
received dust mite antigen challenge at 0.5 h FA and O3
postexposures and were lavaged 6 h post-challenge.
Amount of allergen producing 15% FEVj decrement was
decreased by O3 compared to FA in 9 of 14 subjects.
PMN in proximal airway lavage tended to be greater after
O3 than FA (p = 0.06).
Mean early-phase FEVj response and number of >20%
reductions in FEVj were significantly greater after
0.25ppmO3or4 x 0.125 ppm O3. Most of the >15%
late-phase FEVj responses occurred after 4 days of
exposure to 0.125 ppm O3, as well as significant
inflammatory effects, as indicated by increased sputum
eosinophils (asthma and allergic rhinitis) and increased
sputum lymphocytes, mast cell tryptase, histamine, and
LDH (asthma only).
Chen et al.
(2004)
Holz et al.
(2002)
0.0
0.25
0
490
3hIE,
VE = 30 L/min
15 min ex/10 min
rest/5 min no O3;
every 30 min.
27 °C
54% RH
mouthpiece
exposure
13 M, 11 F Atopic mild asthma
6 M, 6 F Positive allergen and IgE tests
5 M, 5 F Healthy NS
O3-induced FEVj decrements of 12.5, 14.1, and 10.2%
in asthmatics, allergic rhinitics and healthy subjects,
respectively (group differences not significant).
Methacholine responsiveness increased in asthmatics.
Allergen responsiveness: increased significantly after O3
exposure in asthmatics (~ 2 dose shift) and a smaller shift
is rhinitics. No change in healthy. Neither allergen or
methacholine response correlated with lung function and
were not correlated with each.
Jorres et al.
(1996)
-------
Table AX6-3 (cont'd). Ozone Exposure in Subjects with Preexisting Disease"
X
Ozone
Concentration'
ppni ug/ni3
Exposure Duration Exposure
and Activity Conditions
Number and
Gender of Subjects Subject Characteristics
Observed Effect(s)
Reference
Subjects with Asthma
0.4 784
0.0 0
0.4 784
0.12 236
0.0 0
0.2 392
0.4 784
0.0 0
0.12 235
0.4 784
2h IE NA
(15 min rest, 15 min exercise
on bicycle)
VE = 30 L/min
2h IE NA
4x15 min on bicycle,
VE = 40 L/min
Rest 22 °C
40% RH
6 h 22 °C
30 min rest/30 min exercise 50% RH
VE ~ 25 L/min
3h 6x15 min cycle ergometer 31 °C
VE » 32L/min 35% RH
5 consecutive days
1 h rest NA
air-antigen
O3-antigen
2 h IE Head mask
15 min exercise exposure
15 min rest »18°C
VE » 20L/min 60% RH
4 F, 5 M Healthy
(25 ± 2 years old)
7 F, 6 M Mild atopic asthma;
beta agonists only
(22 ± 0.7 years old)
15 Healthy adults
18-40 years old
9 Mild atopic asthmatics
18-40 years old
10 M, 5 F atopic asthma
5 M Healthy NS
5 M Asthmatics, physician
diagnosed,
All 18-45 years
8 M , 2 F Asthmatic NS adults
beta-agonist use only
19-48 years old
ATS criteria for asthma
9 M, 6 F Mild allergic asthma; 18 to
49 years of age.
5 M, 1 F Healthy adults
6 M Atopic asthmatics
Significant reductions in FVC (12%, 10%) and
FEV[ (13%, 1 1%) for asthmatic and healthy
subjects, respectively; attenuated by indomethacin
in healthy subjects only. Significant reductions in
mid-flows which tended to be greater in asthmatics
than healthy subjects. Indomethacin treatment
attenuated mid-flow-reductions somewhat more
in asthmatics than healthy subjects.
Sputum collected 24 h before and 4-6 h post-O3
exposure. Baseline GDI Ib expression positively
correlated with O3-induced PMN. Increased
expression of mCD14 on macrophages following
O3 compared to FA. Asthmatic PMN response
similar to healthy subjects (also see Table AX6-3).
No spirometric data available.
No effect of O3 on airway response to grass
allergen.
Similar spirometric responses in asthmatic and
healthy. However, preexposure FEV1 and FVC
were both —0.4 L lower on O3-day than FA day.
More PMN's in asthmatics. IL-8 and IL-6 higher
in asthmatics exposed to O3. No relationship of
spirometry and symptoms to inflammation.
FEVj decreased 35% on first exposure day.
Methacholine reactivity increased about ten-fold.
Also see Table AX6- 7 for repeated exposure results.
No effect of O3 on airway response to grass or
ragweed allergen.
FEVj responses of healthy and asthmatic similar
(~ 15% decrease). Maximal FEVj response to
methacholine increased similarly in both groups
(12 h postexposure). Larger increase in PC20 in
Alexis et al.
(2000)
Alexis et al.
(2004)
Ball et al.
(1996)
Basha et al.
(1994)
Gong et al.
(1997b)
Hanania
etal. (1998)
Hiltermann
etal. (1995)
healthy subjects. Both groups had increased
PMN's in sputum no correlation of PMN's
and lung function.
-------
Table AX6-3 (cont'd). Ozone Exposure in Subjects with Preexisting Disease"
X
ON
OJ
to
Ozone
Concentration'
ppm
Hg/ni3
Number and
Exposure Duration Exposure Gender of
and Activity Conditions Subjects
Subject
Characteristics
Observed Effect(s)
Reference
Adult Subjects with Asthma (cont'd)
0.0
0.15
0.25
0.25
0.0
0.16
0.0
0.25
0.16
0.25
0.40
0.0
0.2
0
294
490
490
0
314
0
490
314
490
784
0
392
3hIE 27±1°C 10M, 11F
15 min rest 56 ± 7 % RH
(VE = 7 L/min)
15 min on bicycle 5M, 10F
(VE = 26 L/min)
7.6 h 18 °C 13 M
25 min treadmill, 40% RH
25 min cycle/10 min rest 7 M, 10 F
per hour.
VE = 27-32 L/min
3hIE, 27 °C 13 M, 11 F
VE = 30 L/min 54% RH
15 min ex/ 10 min rest/5 mouthpiece 6 M, 6 F
min no O3; every 30 min exposure
5M, 5F
7.6 h 22 °C 4 M, 5 F
25 min treadmill, 25 min 40 % RH
cycle/10 min rest per
hour.
VE = 25 L/min
VE = 25-45 L/min NA 8 M, 4 F
8 M, 10 F
22 M, 16 F
2 h IE 20 °C 6 M, 9 F
4x15 min 50% RH
at VE = 20
L/min/m2 BSA 9 M, 6 F
Healthy NS
28 ± 5 years old
Mild Asthmatic
30 ± 8 years old
Healthy NS,
age 19-32 years.
Moderate Asthmatics,
physician diagnosed,
beta agonist users,
age 19-32 years.
Atopic mild asthma
Positive allergen and
IgE tests
Healthy NS
Mild atopic asthma;
no meds 12 h
pre-exposure
20-35 years old
Asthmatics
Allergic rhinitics
Healthy adults
All <26 years old
Healthy adults
24 years old
Mild asthmatics
29 years old
No significant O3-induced group differences in symptoms or
spirometry. After 0.25 ppm O3, there were significant
decrements in FEV; and FVC that tended to be greater in the
asthmatics than controls. Small but significant neutrophil
increase in asthmatics following 0.15 ppm O3. Significant
neutrophil increases following 0.25 ppm O3 that did not differ
between groups.
FEVj decreased 19% in asthmatics and only 10% in
nonasthmatics. High responders had worse baseline airway
status. More wheeze in asthmatics after O3.
O3-induced FEVj decrements of 12.5, 14.1, and 10.2% in
asthmatics, allergic rhinitics and healthy subjects, respectively
(group differences not significant). Methacholine
responsiveness increased in asthmatics. Allergen responsiveness
increased after O3 exposure in asthmatics (=2 dose shift),
a smaller shift occurred in rhinitics, no change occurred in
healthy subjects. Neither allergen nor methacholine responses
were correlated with each other or with lung function.
Significant FEVj decrease of 9.1 % following O3 exposure;
marked individual variability with responses ranging from 2
% to 26 %.
Healthy 12.2% decrease in FEV1; Rhinitics 10.1%,
asthmatics 12.4%
O3-induced FEVj decrement (8%, healthy adults; 3% asthmatics)
and PMN increase (20.6%, healthy adults; 15.2% asthmatics).
Primary goal was to investigate relationship between antioxidant
defenses and O3 responses in asthmatics and healthy adults
(see Tables AX6-3 andAX6 -13).
Holz et al.
(1999)
Horstman
etal. (1995)
Jorres et al.
(1996)
Kehrl et al.
(1999)
Magnussen
etal. (1994)
Mudway
etal. (2001)
Stenfors
et al. (2002)
-------
Table AX6-3 (cont'd). Ozone Exposure in Subjects with Preexisting Disease"
X
Ozone
Concentration'
Exposure Duration Exposure
ppm Hg/ni3 and Activity Conditions
Number and Subject
Gender of Subjects Characteristics
Observed Effect(s)
Reference
Adult Subjects with Asthma (cont'd)
0.2 396 2hIE 22 °C
( 1 5 min rest, 1 5 min 40 % RH
exercise on bicycle)
VE = 20 L/min/m2 BSA
0.4 784 2 h rest 21 °C
40% RH
0.16 314 7.6 h 18 °C
25 min treadmill, 40% RH
25 min cycle/ every
hour.
0.0 0 4h 21 °C
0.2 392 50 min exercise, 50% RH
10 min rest each hour.
VE » 45-50 L/min
5 F, 4 M Mild atopic asthma; no
meds 8 h pre-exposure
2 1-42 years old
1 1 M , 1 1 F Asthmatics sensitive
to D Farinae,
physician diagnosed,
18 to 35 years
8 M Mild asthmatics,
physician diagnosed,
reactive to dust mite
D. Farinae.
12 M, 6F 18 adult mild
asthmatics mostly
beta agonist users.
Significant decrease in FEVj and a trend toward decreases
in mean inspiratory flow, FEF25, and FEF75 after O3
exposure. No significant differences in FEF50, FVC, TLC,
Raw, or sRaw. No correlation between sputum neutrophils
at 6 h postexposure and FEVj immediately after exposure.
Ozone resulted in nasal inflammation (increased PMN's)
and caused augmented response to nasal allergen challenge.
Increased eosinophils and PMN's after O3 exposure more
in initial (bronchial) fraction. No correlation of eosinophils
and PMN's, FEVj & FVC decreased 14% and 9%
respectively.
FVC, FEVj decreased 17.6% and 25% respectively.
Trend for larger increase in SRaw in asthmatics. Larger
increase in PMN's and protein in asthmatics indicating
more inflammation. No increase in eosinophils.
Spirometry changes in asthmatics similar to healthy
subjects (Aris et al., 1995; Balmes et al., 1997).
Newson et al.
(2000)
Peden et al.
(1995)
Peden et al.
(1997)
Scannell et al.
(1996)
"See Appendix A for abbreviations and acronyms.
'Grouped by rest and exercise; within groups listed from lowest to highest O3 concentration.
-------
exposures less than 2 hours in duration. These findings are also consistent decreasing O3 effects
with increasing age (see Section AX6.5.7).
More recently, Gong et al. (1997a) exposed 9 COPD patients (age range, 59 to 71 years;
mean age 66 ± 4 years) and 10 healthy NS (age range, 60 to 69 years; mean age 65 ± 3 years)
to 0.24 ppm for 4 h with interment light exercise («20 L/min). COPD patients had decreases
in FEVj following both clean air (-11%, p = 0.06) and O3 (-19%, p < 0.01) exposures.
These FEVj decrements, presumably due to exercise, were primarily attributable to four of
the patients who lost greater than 14% of their FEVj following both the air and O3 exposures.
Relative to clean air, O3 caused a statistically insignificant FEVj decrement of -8% in COPD
patients which was not statistically different from the decrement of -3% in healthy subjects.
Ozone-induced symptoms, sRaw, SaO2, and postexposure bronchial activity also exhibited little
or no difference between the COPD patients and the healthy subjects.
AX6.3.2 Subjects with Asthma
Based on studies reviewed in the 1996 criteria document (U.S. Environmental Protection
Agency, 1996) asthmatics appear to be at least as sensitive to acute effects of O3 as healthy
nonasthmatic subjects. At rest, neither adolescent asthmatics nor healthy controls had significant
responses as a result of an hour exposure to 0.12 ppm O3. Exposure of adult asthmatics to
0.25 ppm O3 for 2 h at rest also caused no significant responses. Preexposure to between 0.10
and 0.25 ppm O3 for 1 hr with light IE does not appear to exacerbate exercise-induced asthma
(Fernandes et al., 1994; Weymer et al., 1994). At higher exposures (0.4 ppm O3 with heavy
IE for 2 h), Kreit et al. (1989) and Eschenbacher et al. (1989) demonstrated significantly
greater FEVj and FEF25.75 decrements in asthmatics than in healthy controls. With longer
duration exposures to lower O3 levels (0.12 ppm with moderate IE for 6.5 h), asthmatics have
also shown a tendency for greater FEVj decrements than healthy nonasthmatics (Linn et al.,
1994). Newer clinical studies (see Table AX6-3) continue to suggest that asthmatics are at least
as sensitive as healthy controls to O3-induced responses.
Studies of less than 3 h duration have reported similar or tendencies for increased
O3-induced spirometric responses up to O3 concentrations of 0.4 ppm. Similar group decrements
in FEVj and FVC were reported by Hiltermann et al. (1995), who exposed 6 asthmatics and
6 healthy subjects to 0.4 ppm O3 for 2 h with light IE. Alexis et al. (2000) exposed 13 mild
AX6-34
-------
atopic asthmatics and 9 healthy subjects for 2 h to 0.4 ppm O3 with IE ( VE = 30 L/min). Similar
O3-induced group decrements in FEVj and FVC were also reported by these investigators.
A tendency, however, for an increased O3-induced reduction in mid-flows (viz., FEF25, FEF50,
FEF60p, FEF75) was reported for the asthmatics relative to the healthy subjects. In a larger study,
Torres et al. (1996) exposed 24 asthmatics, 12 allergic rhinitis, and 10 healthy subjects to
0.25 ppm O3 for 3 h with IE. Statistically significant O3-induced decreases in FEVj occurred in
all groups, but tended to be lower in healthy controls (allergic rhinitis, -14.1%; asthmatics,
-12.5%; healthy controls, -10.2%). Holz et al. (1999) exposed 15 asthmatics and 21 healthy
controls to 0.15 and 0.25 ppm O3 for 3-h with light IE. After the 0.25 ppm O3 exposure, there
were significant decrements in FEVj and FVC that tended to be slightly greater in the asthmatics
than controls. One study reported that asthmatics tended to have less of an FEVj response to O3
than healthy controls (Mudway et al., 2001). In that study, however, the asthmatics also tended
to be older than the healthy subjects which could partially explain their lesser response.
Studies between 4 and 8 h duration, with O3 concentrations of 0.2 ppm or less, also suggest
a tendency for increased O3-induced pulmonary function responses in asthmatics relative to
healthy subjects. Scannell et al. (1996) exposed 18 asthmatics to 0.2 ppm O3 for 4 h with
IE (VE -25 L/min/m2 BSA). Baseline and hourly pulmonary function measurements of FEVb
FVC, and sRaw were obtained. Asthmatic responses were compared to 81 healthy subjects who
underwent similar experimental protocols (Aris et al., 1995; Balmes et al., 1996). Asthmatic
subjects experienced a significant O3-induced increase in sRaw, FEVj and FVC. The O3-induced
increase in sRaw tended to be greater in asthmatics than the healthy subjects, whereas similar
group decrements in FEVj and FVC were observed. Basha et al. (1994) also reported similar
spirometric responses between 5 asthmatic and 5 healthy subjects exposed to 0.2 ppm O3 for 6 h
with IE. However, the mean preexposure FEVj in the asthmatics was about 430 mL less (i.e.,
-12% decreased) on the O3-day relative to the air-day. In a longer exposure duration (7.6 h)
study, Horstman et al. (1995) exposed 17 asthmatics and 13 healthy controls to 0.16 ppm O3 or
FA with alternating periods of exercise (50 min, VE «30 L/min) and rest (10 min). Both groups
had significant O3-induced decrements in FEVl5 FVC, and FEV25.75. The asthmatic and healthy
subjects had similar O3-induced reductions in FVC. The FEVj decrement experienced by the
asthmatics was significantly greater in the healthy controls (19% versus 10%, respectively).
AX6-35
-------
There was also tendency for a greater O3-induced decrease in FEF25_75 in asthmatics relative to
the healthy subjects (24% versus 15%, respectively).
With repeated O3 exposures asthmatics, like healthy subjects (see Section AX6.6)
develop tolerance. Gong et al. (1997b) exposed 10 asthmatics to 0.4 ppm O3, 3 h per day with
IE (VE -32 L/min), for 5 consecutive days. Symptom and spirometric responses were greatest
on the first (-35 % FEVj) and second (-34 % FEVj) exposure days, and progressively
diminished toward baseline levels (-6 % FEVj) by the fifth exposure day. Similar to healthy
subjects, asthmatics lose their tolerance to O3 after 4 to 7 days without O3 exposure.
Other published studies with similar results (e.g., McBride et al., 1994; Basha et al., 1994;
Peden et al., 1995, 1997; Peden, 2001a; Scannell et al., 1996; Hiltermann et al., 1997, 1999;
Michelson et al., 1999; Vagaggini et al., 1999; Newson et al., 2000; Holz et al., 2002) also
reported that asthmatics have a reproducible and somewhat exaggerated inflammatory response
to acute O3 exposure (see Section AX6.9). For instance, Scannell et al. (1996) performed lavages
at 18 h post-O3 exposure to assess inflammatory responses in asthmatics. Asthmatic responses
were compared to healthy subjects who underwent a similar experimental protocol (Balmes
et al., 1996). Ozone-induced increases in BAL neutrophils and total protein were significantly
greater in asthmatics than healthy subjects. There was also a trend for an ozone related increased
IL-8 in the asthmatics relative to healthy subjects. Inflammatory responses do not appear to be
correlated with lung function responses in either asthmatic or healthy subjects (Balmes et al.,
1996, 1997; Holz et al., 1999). This lack of correlations between inflammatory and spirometric
responses may be due to differences in the time kinetics of these responses (Stenfors et al.,
2002). In addition, airway responsiveness to inhaled allergens is increased by O3 exposure in
subjects with allergic asthma for up to 24 h (see Section AX6.8).
One of the difficulties in comparing O3-induced spirometric responses of healthy subjects
versus asthmatics is the variability in responsiveness of asthmatics. Most of the asthma studies
were conducted on subjects with a clinical history of mild disease. However, classification of
asthma severity is not only based on functional assessment (e.g., percent predicted FEVj), but
also on clinical symptoms, signs, and medication use (Table AX6-4). Although "mild atopic
asthmatics" are frequently targeted as an experimental group, the criteria for classification has
varied considerably within and across the available published studies. Although the magnitude
of group mean changes in spirometry may not be significantly different between healthy and
AX6-36
-------
X
Table AX6-4. Classification of Asthma Severity1
Classification
Severe
persistent
Moderate
persistent
Days with
Step symptoms
4 Continual
3 Daily
Nights with
symptoms
Frequent
>l/week
Lung Function2
FEVlorPEF PEF
% predicted variability
oral (%)
<60 >30
between >30
60 and 80
Medications3
Daily
High-dose inhaled steroids (ICS)
and long-acting inhaled p2-agonist
If needed, add oral steroids
Low-to-medium-dose ICS and
long-acting p2-agonist (preferred)
Quick relief
Short-acting inhaled
p2-agonist, as needed; oral
steroids may be required
Short-acting inhaled
p2-agonist, as needed;
Mild
persistent
>2/week,
but2/week
>80
Mild
intermittent
<2/week
<2/month
^80
Or
Medium-dose ICS (another preferred
option for children ages <5 years)
Or
Low-to-medium-dose ICS and either
leukotriene modifier or theophylline
20-30 Low-dose inhaled steroids (preferred)
Or
Cromolyn leukotriene modifier,
or (except for children aged <5 years)
nedocromil or sustained release
theophylline to serum concentration
of 5-15 ug/mL
<20 No daily medicine needed
oral steroids may be
required
Short-acting inhaled
p2-agonist, as needed;
oral steroids may be
required
Short-acting inhaled
p2-agonist, as needed;
oral steroids may be
required
1 Sources: Centers for Disease Control (2003); National Institutes of Health (1997, 2003).
2For adults and children aged >5 years who can use a spirometer or peak flow meter.
3The medications listed here are appropriate for treating asthma at different levels of severity. The preferred treatments, dosage, and type of medication
recommended vary for adults and children and are detailed in the EPR-Update 2002 stepwise approach to therapy. The stepwise approach emphasizes that
therapy should be stepped up as necessary and stepped down when possible to identify the least amount of medication required to achieve goals of therapy.
The stepwise approach to care is intended to assist, not replace, the clinical decision-making required to meet individual patient needs.
-------
asthmatic subjects, many of the studies have reported clinically significant changes in some
individuals.
Alexis et al. (2000) explored the possibility that the mechanisms of O3-induced spirometric
responses may differ between asthmatics and healthy subjects. Physician-diagnosed mild
atopic asthmatics and healthy subjects were pretreated with 75 mg/day of indomethacin
(a COX inhibitor) or placebo and then exposed for 2 h to 0.4 ppm O3 or to FA during mild
IE ( VE =30 L/m). The number and severity of O3-induced symptoms were significantly
increased in both asthmatics and healthy subjects. These symptom responses were similar
between the subject groups and unaffected by indomethacin pretreatment. Asthmatics and
healthy subjects also had similar O3-induced reductions in FVC and FEVj. These restrictive-
type responses, occurring due to the combined effects of bronchoconstriction and reflex
inhibition of inspiration (see Section AX6.2.1\ were attenuated by indomethacin in the healthy
subjects but not the asthmatics. Thus, in healthy subjects but not asthmatics, COX metabolites
may contribute to O3-induced reductions in FVC and FEVj. As assessed by the magnitude of
reductions in mid-flows (viz. FEF25, FEF50, FEF60p, FEF75), the small airways of the asthmatics
tended to be more affected than the healthy subjects. This suggests asthmatics may be more
sensitive to small airway effects of O3, which is consistent with the observed increases in
inflammation and airway responsiveness. Indomethacin pretreatment attenuated some of
these O3-induced small airways effects (FEF50 in healthy subjects, FEF60p in asthmatics).
AX6.3.3 Subjects with Allergic Rhinitis
Most O3 exposure studies in humans with existing respiratory disease have focused on lung
diseases like COPD and asthma. However, chronic inflammatory disorders of the nasal airway,
especially allergic rhinitis, are very common in the population. People with allergic rhinitis have
genetic risk factors for the development of atopy that predispose them to increased upper airway
responsiveness to specific allergens as well as nonspecific air pollutants like O3. Studies
demonstrating the interaction between air pollutants and allergic processes in the human nasal
airways and rhinoconjunctival tissue have been reviewed by Peden (2001b) and Riediker et al.
(2001), respectively. Ozone exposure of subjects with allergic rhinitis has been shown to induce
nasal inflammation and increase airway responsiveness to nonspecific bronchoconstrictors,
although to a lesser degree than experienced by asthmatics.
AX6-38
-------
McDonnell et al. (1987) exposed nonasthmatic adults with allergic rhinitis to 0.18 ppm O3.
The allergic rhinitics were no more responsive to O3 than healthy controls, based on symptoms,
spirometry, or airway reactivity to histamine although they had a small but significantly greater
increase in SRaw. The data on subjects with allergic rhinitis and asthmatic subjects suggest that
both of these groups have a greater rise in Raw to O3 with a relative order of airway
responsiveness to O3 being normal < allergic < asthmatic.
Bascom et al. (1990) studied the upper respiratory response to acute O3 inhalation, nasal
challenge with antigen, and the combination of O3 plus antigen in subjects with allergic rhinitis.
Exposure to O3 caused significant increases in upper and lower airway symptoms, a mixed
inflammatory cell influx with a seven-fold increase in nasal lavage PMNs, a 20-fold increase in
eosinophils, and a 10-fold increase in mononuclear cells, as well as an apparent sloughing of
epithelial cells. McBride et al. (1994) also observed increased nasal PMN's after O3 exposure in
atopic asthmatics. Peden et al. (1995), who studied allergic asthmatics exposed to O3 found
that O3 causes an increased response to nasal allergen challenge in addition to nasal
inflammatory responses. Their data suggested that allergic subjects have an increased immediate
response to allergen after O3 exposure. In a follow-up study, Michelson et al. (1999) reported
that 0.4 ppm O3 did not promote early-phase-response mediator release or enhance the response
to allergen challenge in the nasal airways of mild, asymptomatic dust mite-sensitive asthmatic
subjects. Ozone did, however, promote an inflammatory cell influx, which helps induce a more
significant late-phase response in this population.
Torres et al. (1996) found that O3 causes an increased response to bronchial allergen
challenge in subjects with allergic rhinitis. This study also compared responses in subjects
with mild allergic asthma (see Sections AX6.3.2 andAX6.8). The subjects were exposed to
0.25 ppm O3 for 3 h with IE. Airway responsiveness to methacholine was determined 1 h before
and after exposure; responsiveness to allergen was determined 3 h after exposure. Statistically
significant decreases in FEVj occurred in subjects with allergic rhinitis (13.8%) and allergic
asthma (10.6%), and in healthy controls (7.3%). Methacholine responsiveness was statistically
increased in asthmatics, but not in subjects with allergic rhinitis. Airway responsiveness to an
individual's historical allergen (either grass and birch pollen, house dust mite, or animal dander)
was significantly increased after O3 exposure when compared to FA exposure. In subjects with
asthma and allergic rhinitis, a maximum percent fall in FEVj of 27.9 % and 7.8%, respectively,
AX6-39
-------
occurred 3 days after O3 exposure when they were challenged with of the highest common dose
of allergen. The authors concluded that subjects with allergic rhinitis, but without asthma, could
be at risk if a high O3 exposure is followed by a high dose of allergen.
Holz et al. (2002) extended the results of Torres et al. (1996) by demonstrating that
repeated daily exposure to lower concentrations of O3 (0.125 ppm for 4 days) causes an
increased response to bronchial allergen challenge in subjects with preexisting allergic airway
disease, with or without asthma. There was no major difference in the pattern of bronchial
allergen response between subjects with asthma or rhinitis, except for a 10-fold increase in the
dose of allergen required to elicit a similar response (>20% decrease in FEVj) in the asthmatic
subjects. Early phase responses were more consistent in subjects with rhinitis and late-phase
responses were more pronounced in subjects with asthma. There also was a tendency towards a
greater effect of O3 in subjects with greater baseline response to specific allergens chosen on the
basis of skin prick test and history (viz., grass, rye, birch, or alder pollen, house dust mite, or
animal dander). These data suggest that the presence of allergic bronchial sensitization, but not a
history of asthma, is a key determinant of increased airway allergen responsiveness with O3.
[A more complete discussion of airway responsiveness is found in Section AX6.8]
AX6.3.4 Subjects with Cardiovascular Disease
Superko et al. (1984) exposed six middle-aged males with angina-symptom-limited
exercise tolerance for 40 min to FA and to 0.2 and 0.3 ppm O3 while they were exercising
continuously according to a protocol simulating their angina-symptom-limited exercise training
prescription (meanVE= 35 L/min). No significant pulmonary function impairment or evidence
of cardiovascular strain induced by O3 inhalation was observed. Gong et al. (1998) exposed
hypertensive (n = 10) and healthy (n = 6) adult males, 41 to 78 years of age, to FA and on the
subsequent day to 0.3 ppm O3 for 3 h with IE at 30 L/min. The ECG was monitored by
telemetry, blood pressure by cuff measurement, and a venous catheter was inserted for
measurement of routing blood chemistries and cardiac enzymes. Pulmonary artery and radial
artery catheters were placed percutaneously for additional blood sampling and for measurement
of hemodynamic pressures, cardiac output, and SaO2. Other hemodynamic variables were
calculated, including cardiac index, stroke volume, pulmonary and systemic vascular resistance,
left and right ventricular stroke-work indices, and rate-pressure product. Spirometric volumes
AX6-40
-------
(FVC, FEVj) and symptoms of breathing discomfort were measured before and after the
O3 exposures. There were significant O3-induced FEVj decrements in both subject groups that
did not defer between groups (hypertensive, 7.6%; healthy, 6.7%). The overall results did not
indicate any major acute cardiovascular effects of O3 in either the hypertensive or normal
subjects. However, statistically significant O3 effects for both groups combined were increases
in HR, rate-pressure product, and the alveolar-to-arterial PO2 gradient, suggesting that impaired
gas exchange was being compensated for by increased myocardial work. These effects might be
more important in some patients with severe cardiovascular disease. [See Section AX6.10 for
discussion of extrapulmonary effects ofO3 exposure.]
AX6.4 INTERSUBJECT VARIABILITY AND REPRODUCIBILITY
OF RESPONSE
Analysis of the factors that contribute to intersubject variability is important for the
understanding of individual responses, mechanisms of response, and health risks associated with
acute O3 exposures. Bates et al. (1972) noted that variation between individuals in sensitivity
and response was evident in respiratory symptoms and pulmonary function following O3
exposure. A large degree of intersubject variability in response to O3 has been consistently
reported in the literature (Adams et al., 1981; Aris et al., 1995; Folinsbee et al., 1978; Kulle
et al., 1985; McDonnell et al., 1983). Kulle et al. (1985) noted that the magnitude of variability
between individuals in FEVj responses increases with O3 concentration. Similarly, McDonnell
et al. (1983) observed FEVj decrements ranging from 3 to 48% (mean 18%) in 29 young adult
males exposed to 0.40 ppm O3 for 2 h during heavy IE. At a lower O3 concentration of
0.18 ppm, 20 similarly exposed subjects had FEVj decrements ranging from 0 o 23%
(mean = 6%), while those exposed to FA (n = 20) had decrements ranging from -2% to 6%
(mean = 1%) (McDonnell et al., 1983). All of the subjects in these studies were young adult
males. (Intersubject variability related to age and gender is discussed in Sections AX6.5.1 and
AX6.5.2, respectively.)
More recently, McDonnell (1996) examined the FEVj response data from three 6.6 h
exposure studies of young adult males conducted at the EPA Health Effects Research Laboratory
in Chapel Hill, NC (Folinsbee et al., 1988; Horstman et al., 1990; McDonnell et al., 1991).
AX6-41
-------
The response distributions for subjects at each of four O3 concentrations (0.0, 0.08, 0.10, and
0.12 ppm) are illustrated in Figure AX6-6. It is apparent that the FEVj responses in FA are
small with most tightly grouped around zero. With increasing O3 concentration, the mean
response increases as does the variability about the mean. At higher O3 concentrations, the
distribution of response becomes asymmetric with a few individuals experiencing large FEVj
decrements. The response distribution in Figure AX6-6 allows estimates of the number or
percentage of subjects responding in excess of a certain level. With FA exposure, none of
87 subjects had a FEVj decrement in excess of 10%; however, 26%, 31%, and 46% exceeded a
10% decrement at 0.08, 0.10, and 0.12 ppm, respectively. FEVj decrements as large as 30 to
50% were even observed in some individuals. In 6.6-h face mask exposures of young adults
(half women) to 0.08 ppm O3, Adams (2002) found that 6 of 30 subjects (20%) had >10%
decrements in FEVj. The response distributions in Figure AX6-6 underlines the wide range of
response to O3 under prolonged exposure conditions and reinforces the observations by others
consequent to 2 h IE exposures at higher O3 concentrations (Horvath et al.,1981; McDonnell
etal., 1983).
o
-40-
(/) 30-
.2 20-
| 10-
0>
n n .
nn
-
0 ppm
n = 87
0%
n
-
-
.
-
0.08 ppm
n = 60
26%
n
I
[h™
-
-
.
-
0.10 ppm
n = 32
31%
-
n
n n
•
•
.
•
0.12 ppm
n = 49
46%
In
[tin n n
-10 0 10 20 30 40
-10 0 10 20 30 40
-10 0 10 20 30 40
-10 0 10 20 30 40
FEV, ("/(.Decrement)
Figure AX6-6. Frequency distributions of FEVt decrements following 6.6-h exposures
to O3 or filtered air. During each hour of the exposures, subjects were
engaged in moderate exercise for 50 minutes. With increasing O3
concentration, the distribution of responses becomes asymmetric, with a
few individuals exhibiting large FEVt decrements. The percentage in each
panel indicates the portion of subjects having a FEVt decrement in excess
of 10%.
Source: McDonnell (1996).
AX6-42
-------
Some of the intersubject variability in response to O3 inhalation may be due to intrasubject
variability, i.e., how reproducible the measured responses are in an individual between several
O3 exposures. The more reproducible the subject's response, the more precisely it indicates
his/her intrinsic responsiveness. McDonnell et al. (1985a) examined the reproducibility of
individual responses to O3 in healthy human subjects (n = 32) who underwent repeated
exposures within a period of 21 to 385 days (mean = 88 days; no median reported) at one of
five O3 concentrations ranging from 0.12 to 0.40 ppm. Reproducibility was assessed using the
intraclass correlation coefficient (R). The most reproducible responses studied were FVC
(R = 0.92) and FEVj (R = 0.91). However, at the lowest concentration, 0.12 ppm, relatively
poor FEVj reproducibility was observed (R = 0.58) due, in part, to a lack of specific O3 response
or a uniformly small response in the majority of subjects. McDonnell et al. (1985a) concluded
that for 2 h IE O3 exposures equal to or greater than 0.18 ppm, the intersubject differences in
magnitude of change in FVC and FEVj are quite reproducible over time and likely due to
differences in intrinsic responsiveness of individual subjects. Hazucha et al. (2003) exposed
47 subjects on three occasions for 1.5 h, with moderate intensity IE, to 0.40 to 0.42 ppm O3.
Reproducibility of FEVj responses was related to the length of time between re-exposures,
with a Spearman correlation R of 0.54 obtained between responses for exposures 1 and
2 (median = 105 days), and an R of 0.85 between responses for exposures 2 and 3
(median = 7 days).
Identification of mechanisms of response and health risks associated with acute O3
exposures are complicated by a poor association between various O3-induced responses.
For example, McDonnell et al. (1983) observed a very low correlation between changes in sRaw
and FVC (r = -0.16) for 135 subjects exposed to O3 concentrations ranging from 0.12 to
0.40 ppm for 2.5 h with IE. In a retrospective study of 485 male subjects (ages 18 to 36 yrs)
exposed for 2 h to one of six O3 concentrations at one of three activity levels, McDonnell et al.
(1999) observed significant, but low, Spearman rank order correlations between FEVj response
and symptoms of cough (R = 0.39), shortness of breath (R = 0.41), and pain on deep inspiration
(R = 0.30). The authors concluded from their data that the O3-induced responses are related
mechanistically to some degree, but that there is not a single factor which is responsible for the
observed individual differences in O3 responsiveness across the spectrum of symptom and lung
function responses. This conclusion is supported by differences in reproducibility observed by
AX6-43
-------
McDonnell et al., (1985a). Compared to the intraclass correlation coefficient for FEVj
(R = 0.91), relatively low but statistically significant R values for symptoms ranged from 0.37 to
0.77, with that for sRaw being 0.54. The reproducibility correlations for fB (R = -0.20) and VT
(R = -0.03) were not statistically significant.
The effect of this large intersubject variability on the ability to predict individual
responsiveness to O3 was demonstrated by McDonnell et al. (1993). These investigators
analyzed the data of 290 male subjects (18 to 32 years of age) who underwent repeat 2 h IE
exposures to one or more O3 concentrations ranging from 0.12 to 0.40 ppm in order to identify
personal characteristics (i.e., age, height, baseline pulmonary functions, presence of allergies,
and past smoking history) that might predict individual differences in FEVj response. Only age
contributed significantly to intersubject responsiveness (younger subjects were more
responsive), accounting for just 4% of the observed variance. Interestingly, O3 concentration
accounted for only 31% of the variance, strongly suggesting the importance of as yet undefined
individual characteristics that determine FEVj responsiveness to O3. A more general form of
this model was developed to investigate the O3 exposure FEVj response relationship (McDonnell
et al., 1997). These authors used data from 485 male subjects (age = 18 to 36 years) exposed
once for 2 h to one of six O3 concentrations (ranging from 0.0 to 0.40 ppm) at one of 3 activity
levels (rest, n = 78; moderate IE, n = 92; or heavy IE, n = 314). In addition to investigating the
influence of subject's age, the model focused on determining whether FEVj response was more
sensitive to changes in C than to changes in VE, and whether the magnitude of responses is
independent of differences in lung size. It was found that the unweighted proportion of the
variability in individual responses explained by C, VE, T, and age was 41%, with no evidence
that the sensitivity of FEVj response to VE was different than changes in C, and no evidence that
magnitude of response was related to measures of body or lung size. The authors concluded that
much inter-individual variability in FEVj response to O3 remains unexplained.
AX6-44
-------
AX6.5 INFLUENCE OF AGE, GENDER, ETHNIC, ENVIRONMENTAL
AND OTHER FACTORS
AX6.5.1 Influence of Age
On the basis of results reported from epidemiologic studies, children and adolescents are
considered to be at increased risk, but not necessarily more responsive, to ambient oxidants than
adults. However, findings of controlled laboratory studies that have examined the acute effects
of O3 on children and adolescents do not completely support this assertion (Table AX6-5).
Children experience about the same decrements in spirometric endpoints as young adults
exposed to comparable O3 doses (McDonnell et al., 1985b; Avol et al., 1987). In contrast to
young adults, however, they had no symptomatic response, which may put them at an increased
risk for continued exposure. Similarly, young adults (Linn et al., 1986; Avol et al., 1984) have
shown comparable spirometric function response when exposed to low O3 dose under similar
conditions. Among adults, however, it has been repeatedly demonstrated that older individuals
respond to O3 inhalation with less intense lung function changes than younger adults. Thus,
children, adolescents, and young adults appear to be about equally responsive to O3, but more
responsive than middle-aged and older adults when exposed to a comparable dose of O3 (U.S.
Environmental Protection Agency, 1996).
Gong et al. (1997a) studied ten healthy men (60 to 69 years old) and nine COPD patients
(59 to 71 years old) from the Los Angeles area who were exposed to 0.24 ppm O3 while
intermittently exercising every 15 min at a light load (-20 L/min) for 4 h. Healthy subjects
showed a small but significant O3-induced FEVj decrement of 3.3% (p = 0.03 [not reported in
paper] paired-t on O3 versus FA pre-post FEVj)2. Small but statistically nonsignificant changes
were also observed for respiratory symptoms, airway resistance and arterial O2 saturation. In the
COPD patients, there was an 8% FEVj decrement due to O3 exposure which was not
significantly different from the response in the healthy subjects. The authors have concluded
that typical ambient concentrations of O3 are unlikely to induce "a clinically significant acute
lung dysfunction" in exposed older men. However, they also acknowledged that the "worst
case" scenario of O3 exposure used in their study causes acute spirometric responses.
2Personal communication from authors, correction to Table 2 in Gong et al. (1997a), the %FEVl change at
the end of the ozone exposure for subject ID 2195 should read 4.9 and not the published value of -4.3, the mean and
standard deviation reported in the table are correct.
AX6-45
-------
Table AX6-5. Age Differences in Pulmonary Function Responses to Ozonea
X
Oi
-k
Oi
Ozone
Concentration1"
ppm
0.40
0.42
0.0
0.40
0.0
0.24
0.0
0.12
0.18
0.24
0.30
0.40
0.0
0.12
0.18
0.24
0.30
0.40
ug/m3
784
823
0
784
0
470
0
235
353
471
589
784
0
235
353
471
589
784
Exposure Duration Exposure
and Activity Conditions
2 h IE (15' ex/1 5' rest) -22 °C
VE * 33-45 L/min 40% RH
(47 subjects only) treadmill
1. 5 h IE (20' ex/10' rest)
VE * 33-45 L/min
(All subjects)
2 h, IE (15' ex/1 5' rest) 21 °C
VE « 1 8 L/min/m2 BSA 40% RH
2 exposures: 25%ofsubj. treadmill
exposed to air-air,
75% exposed to O3-O3
4 h, IE (15' ex/1 5' rest) 24 °C
VE = 20 L/min 40% RH
2 h rest or IE 22 °C
(4xl5min 40% RH
at VE = 25 or 35
L/min/m2 BSA)
2. 33 h IE 22 °C
(4xl5min 40% RH
atVE = 25
L/min/m2 BSA)
Number and
Gender of
Subjects
146 M
94 F
28 M
34 F
10 M
9M
485 WM (each
subject
exposed at one
activity level
to one O3
concentration)
371 (WM,
BM, WF, BF;
-25% per
group) each
subject
exposed to one
03
concentration
Subject
Characteristics
Healthy NS
18 to 60 years old
Healthy NS
18 to 57 years old
Healthy NS
18 to 59 years old
Healthy NS
60 to 69 years old
COPD
59 to 71 years old
Healthy NS
18 to 36 years old
mean age 24 years
Healthy NS
18 to 35 years old
mean age 24 years
Observed Effect(s)
Young individuals of both gender (<35 years)
significantly more responsive than older
subjects. Strong responses are less common
over the age of 35 years, especially in women.
The variability of an individual's responsiveness
to repeated exposures to O3 decreases with age.
Significant decrements in spirometric lung
function in all groups. Young males and females
(<35 years) were significantly more responsive
than older individuals (>35 years).
Healthy: small, 3.3%, decline in FEVj (p = 0.03 [not
reported in paper], paired-t on O3 versus FA pre-post
FEVj). COPD: 8% decline in FEVj (p = ns, O3 versus
FA). Adjusted for exercise, ozone effects did not differ
significantly between COPD patients and healthy
subjects.
Statistical analysis of 8 experimental chamber
studies conducted between 1980 and 1993 by the
U.S. EPA in Chapel Hill, NC. O3-induced
decrement in FEV[ predicted to decrease with
age. FEV; response of a 30 year old predicted to
be 50% the response of a 20 year old. Also see
Table 6-1
Statistical analysis of experimental data
collected between 1983 and 1990 by the U.S.
EPA in Chapel Hill, NC. Cyinduced decrement
in FEV; predicted to decrease with age. FEV;
response of a 30 year old predicted to be 65%
the response of a 20 year old. No effect of
menstrual cycle phase on FEV; response.
Inconsistent effect of social economic status on
FEV[ response.
Reference
Hazucha et al.
(2003)
Passannante
etal. (1998)
Gong et al.
(1997a)
McDonnell
etal. (1997)
Seal et al.
(1996)
-------
Table AX6-5 (cont'd). Age Differences in Pulmonary Function Responses to Ozone"
X
Ozone
Concentration1"
ppm
0.18
0.24
0.30
0.40
0.45
0.45
0.45
0.12
0.20
0.30
0.113C +
other
ambient
pollutants
ug/m3
353
470
588
784
882
882
882
235
392
588
221
Exposure Duration
and Activity
2.33h
IE
VE = 20L/min/m2BSA
lh,CE
VE « 26 L/min
2h,IE
VE « 26 L/min
2 h, IE (20' ex/20' rest)
Male: VE = 28.5L/min
Female: VE = 26.1 L/min
2 h, IE (20' ex/20' rest)
VE * 26 L/min
1 h IE (mouthpiece)
VE = 4 to 5 x resting
1 h (mouthpiece)
50' rest/10' ex for first 7
males, 20' rest/10' ex for
remaining subjects
Male: VE « 29 L/min
Female: VE « 23 L/min
1 h CE (bicycle)
VE * 22 L/min
Number and
Exposure Gender of
Conditions Subjects
NA 48 WF, 55 BF
«23°C 7M
58% RH 5 F
cycle/treadmill
23 °C 10 M,
46% RH 6 F
cycle/treadmill
-24 °C 8 M
63% RH
cycle
8F
22 °C 5 M, 7 F
75% RH
treadmill
-22 °C 9M, 10 F
>75% RH
treadmill
32.7 °C 33 M, 33 F
-43% RH
cycle
Subject
Characteristics
Healthy NS,
18 to 35 years old,
black and white
Healthy NS,
60 to 79 years old
(all in 60s except
one 79 years old)
Healthy NS,
60 to 89 years old
Healthy NS,
51 to 69 years old
Healthy NS,
56 to 76 years old
Healthy NS,
12 to 17 years old
Healthy NS,
55 to 74 years old
NS for both groups,
mean age =
9.4 years old
Observed Effect(s)
Older women had smaller changes in FEV[ than
younger women. No age- related differences in
SRaw or cough score.
Comparison of 1-h CE protocol and 2-h IE
protocol indicated no difference between the
changes in pulmonary function following the
two protocols.
Mean decrement in FEV; = 5.7%; eight subjects
had a 5% or greater difference between their
response to O3 and FA, and the other eight had
less than a 5% difference between their
responses to FA and 0.45 ppm O3.
13 subjects had decrements in FEV[ on three
separate exposures to 0.45 ppm within 5% of
their mean response to the three exposures. The
other three subjects were not reproducible.
Symptom reports did not correlate well with
pulmonary function changes.
No significant changes in any pulmonary
function in healthy subjects.
No spirometic changes for either group.
Females had 13% increase in RT at 3 and
22 min after 0.30-ppm exposure.
No differences in responses of boys and girls.
Similar decrements (<5% on average)
following both purified air and ambient air (O3
at 0. 1 1 ppm) exposures.
Reference
Seal et al.
(1993)
Drechsler-
Parks et al.
(1990)
Bedi et al.
(1989)
Bedi et al.
(1988)
Koenig et al.
(1988)
Reisenauer
etal. (1988)
Avol et al.
(1987)
-------
Table AX6-5 (cont'd). Age Differences in Pulmonary Function Responses to Ozone"
Ozone
Concentration1"
Exposure Duration
ppm Mg/m3 and Activity
0.45 882 2 h, IE (20' ex/20' rest)
VE * 26 L/min
0.12 235 40 min (mouthpiece)
IE, 10 min exercise at
VE = 32.6 L/min
0.18 353 40 min (mouthpiece)
IE, 10 min exercise at
VE = 41.3L/min
Number and
Exposure Gender of Subject
Conditions Subjects Characteristics
-23 °C 8 M, Healthy NS,
53% RH 8 F 51 to 76 years old
cycle
NA 3 M, 7 F Healthy NS,
treadmill 14 to 19 years old
4M, 6F
Observed Effect(s)
Mean decrement in FEV; = 5.6 ± 13%; range of
decrements = 0 to 12%.
No significant change in FEV^ increased RT
with exposure to 0. 1 8 ppm O3. Some subjects
responded to 5 to 10 mg/mL methacholine after
0.18-ppm O3 exposure, whereas none responded
to 25 mg/mL methacholine at baseline
bronchochallenge.
Reference
Drechsler-
Parks et al.
(1987a,b)
Koenig et al.
(1987)
aSee Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
j> GOzone concentration is the mean of a range of ambient concentrations.
X
Oi
oo
-------
Although Gong et al. (1997'a) and others (see Table AX6-5) have examined responses to O3
exposure in subjects of various ages, the exposure conditions differ between most studies
so that age effects remain uncertain. Three recent studies, which analyzed large data sets
(>240 subjects) of similarly exposed subjects, show clearly discernable changes in FEVj
responses to O3 as a function of age.
Seal et al. (1996) analyzed O3-induced spirometric responses in 371 young nonsmokers
(18 to 35 years of age). The subject population was approximately 25% white males, 25% white
females, 25% black males, and 25% black females. Each subject was exposed once to 0.0, 0.12,
0.18, 0.24, 0.30, or 0.40 ppm ozone for 2.3 h during IE at a VE of 25 L/min/m2 BSA. A logistic
function was used to model and test the significance of age, socioeconomic status (SES), and
menstrual cycle phase as predictors of FEVj response to O3 exposure. Menstrual cycle phase
was not a significant. SES was inconsistent with the greatest response observed in the medium
SES and the lowest response in high SES. FEVj responses decreased with subject age. On
average, regardless of the O3 concentration, the response of 25, 30, and 35 year old individuals
are predicted to be 83, 65, and 48% (respectively) of the response in 20 year olds. For example,
in 20 year old exposed to 0.12 ppm ozone (2.3 h IE, VE = 25 L/min/m2 BSA) a 5.4% decrement
in FEVj is predicted, whereas, a similarly exposed 35 yr old is only predicted to have a 2.6%
decrement. The Seal et al. (1996) model is limited to predicting FEVj responses immediately
postexposure in individuals exposed for 2.3 h during IE at a VE of 25 L/min/m2 BSA.
McDonnell et al. (1997) examined FEVj responses in 485 healthy white males (18 to
36 years of age) exposed once for 2 h to an O3 concentration of 0.0, 0.12, 0.18, 0.24, 0.30, or
0.40 ppm at rest or one of two levels of IE (VE of 25 and 35 L/min/m2 BSA). FEVj was
measured preexposure, after 1 h of exposure, and immediately postexposure. Decrements
in FEVj were modeled by sigmoid-shaped curve as a function of subject age, O3 concentration,
VE, and duration of exposure. Regardless of the O3 concentration or duration of exposure, the
average responses of 25, 30, and 35 year old individuals are predicted to be 69, 48, and 33%
(respectively) of the response in 20 year olds. The McDonnell et al. (1997) model is best suited
to predicting FEVj responses in while males exposed to O3 for 2 h or less under IE conditions.
Hazucha et al. (2003) analyzed the distribution of O3 responsiveness in subjects (146 M,
94 F) between 18 and 60 years of age. Subjects were exposed to 0.42 ppm O3 for 1.5 h with IE
AX6-49
-------
at VE = 20 L/min/m2 BSA. Figure AX6-7 illustrates FEVj responses to O3 exposure as a
function of subject age. Consistent with the discussion in Section 6.4, a large degree of
intersubject variability is evident in Figure AX6-7. Across all ages, 18% of subjects were weak
responders (<5% FEVj decrement), 39% were moderate responders, and 43% were strong
responders (> 15% FEVj decrement). Younger subjects (<35 years of age) were predominately
strong responders, whereas, older subjects (>35 years of age) were mainly weak responders.
In males, the FEVj responses of 25, 35, and 50 year olds are predicted to be 94, 83, and 50%
(respectively) of the average response in 20 year olds. In females, the FEVj responses of 25, 35,
and 50 year olds are predicted to be 82, 46, and 18% (respectively) of the average response in
20 year olds. The Hazucha et al. (1996) model is limited to predicting FEVj responses
immediately postexposure in individuals exposed to 0.42 ppm O3 for 1.5 h during IE at a VE of
20 L/min/m2 BSA.
110-
100-
90-
80-
70-
60-
50-
40-
30H
15 20 25 30 35 40 45 50 55 60
Age (years)
"53
in
03
J2
LU
110-
100-
90-
80-
70-
60-
50-
40-
30-1
.
J>'a o
o ° o
15 20 25 30 35 40 45 50 55 60
Age (years)
Figure AX6-7. Effect of O3 exposure (0.42 ppm for 1.5 h with IE) on FEVt as a function
of subject age. Left panel data for males (n = 146; 19 to 60 yrs old),
right panel data for females (n = 94; 18 to 59 yrs old). In a 45 yr old
male and 52 yr old female, FEVt decreased to 37 and 72% of baseline,
respectively. These two outliers, not illustrated above, demonstrate that
some individuals may still experience substantial responses despite
increasing age.
Source: Adapted from Hazucha et al. (2003).
AX6-50
-------
The pathophysiologic mechanisms behind the pronounced age-dependent, gender-
differential rate of loss of O3 responsiveness are unclear. Passannante et al. (1998) have
previously demonstrated that O3-induced spirometric decrements (FEVj) in healthy young and
middle-aged adults are principally neural in origin, involving opioid-modulated sensory
bronchial C-fibers. (The methodological details of this study are presented in Section AX6.2.3 of
this chapter} The peripheral afferents are most likely the primary site of action, which would be
compatible with a reflex action as well as a cortical mechanism. The pattern of progressive
decline, as well as the subsequent rate of recovery of spirometric lung function, suggest
involvement of both direct and indirect (possibly by PGE2a) stimulation and/or sensitization of
vagal sensory fibers. (For details, see Section AX6.2.3.1 of this chapter }
The additional pulmonary function data published since the release of last O3 criteria
document (U.S. Environmental Protection Agency, 1996) and reviewed in this section reinforce
the conclusions reached in that document. Children and adolescents are not more responsive
to O3 than young adults when exposed under controlled laboratory conditions. However, they
are more responsive than middle-aged and older individuals. Young individuals between the age
of 18 and 25 years appear to be the most sensitive to O3. With progressing age, the sensitivity
to O3 declines and at an older age (>60 yrs) appears to be minimal except for some very
responsive individuals. Endpoints other than FEVj may show a different age-related pattern
of responsiveness.
AX6.5.2 Gender and Hormonal Influences
The few late 1970 and early 1980 studies specifically designed to determine symptomatic
and lung function responses of females to O3 were inconsistent. Some studies have concluded
that females might be more sensitive to O3 than males, while others found no gender differences
(U.S. Environmental Protection Agency, 1996). During the subsequent decade, seven studies
designed to systematically explore gender-based differences in lung function following O3
exposure were completed (Table AX6-6). Protocols included mouthpiece and chamber
exposures, young and old individuals, normalization of ventilation to BSA or FVC, continuous
and intermittent exercise, control for menstrual cycle phase, and the use of equivalent effective
dose of O3 during exposures. These studies have generally reported no statistically significant
differences in pulmonary function between males and females (Adams et al., 1987; Drechsler-
AX6-51
-------
Table AX6-6. Gender and Hormonal Differences in Pulmonary Function Responses to Ozone"
Ozone
Concentration11
ppm ug/m3 and Activity Conditions'
0.0 490 1 h CE NA
0.25 VE = 30 L/min Face mask
exposure
Number and Gender Subject
of Subjects Characteristics
32 M, 28 F Healthy NS
22.6 ±0.6 years old
Observed Effect(s)
Mean O3-induced FEVj decrements of 15.9%
in males and 9.4% in females (gender
differences not significant). FEV; decrements
ranged from -4 to 56%; decrements >15% in
20 subjects and >40% in 4 subjects. Uptake
of O3 greater in males than females, but
uptake not correlated with spirometric
responses.
Reference
Ultman et al.
(2004)
X
ON
t^ft
to
0.40
0.42
0.0
0.35
0.0
0.4
0
0.12
0.24
0.30
0.40
784
823
0
686
0
784
0
235
470
588
784
2 h, IE (15' ex/1 5' rest)
VE = 33-45 L/min
1. 5 h IE (20' ex/10' rest)
VE = 33-45 L/min
1.25h, IE
(30' ex/1 5' rest/30' ex)
VE = 40 L/min
2 h, IE (15' ex/1 5' rest)
VE» 18L/min/m2BSA
2 exposures: 25% of
subj. exposed to air-air,
75% exposed to O3-O3
2.33 h IE (15' ex/15' rest)
VE = 20 L/min/m2 BSA
one exposure per subject
22 °C
40% RH
treadmill
22 °C
40% RH
treadmill
21 °C
40% RH
treadmill
22 °C
40% RH
treadmill
146 M Healthy NS, No significant gender differences in FEVj
94 F 1 8 to 60 years old among young (<3 5 years) and older
individuals. Strong responses are less
common over the age of 35 years, especially
in women.
19 F O3 responders FVC and FIVC changes about the same,
22.1 ± 2.7 years old - 13%, FEV; -20%. Increased airway
responsiveness to methacholine. Persistence
of small effects on both inspired and expired
spirometry past 18 h. Chemoreceptors not
activated but ventilatory drive was
accelerated.
28 M Healthy NS, 20-59 years old Significant decrements in spirometric lung
34 F function. No significant differences in FEV;
between young females and males and older
females and males either in responders or
nonresponders subgroups.
48WF, 55BF Healthy NS, Significant menstrual cycle phase * race
18 to 35 years old interaction for FEVj. No significant
menstrual cycle phase effect when blacks and
whites were analyzed separately.
No significant menstrual phase effects for
SRaw or cough score.
Hazucha
et al. (2003)
Folinsbee
and Hazucha
(2000)
Passannante
etal. (1998)
Seal et al.
(1996)
-------
Table AX6-6 (cont'd). Gender and Hormonal Differences in Pulmonary Function Responses to Ozone"
Ozone
Concentration11
ppm ug/m3 and Activity
0.0 0 2.15 h, IE
0.35 686 (30' ex/30' rest)
Exposure Number and Gender Subject
Conditions' of Subjects Characteristics
19-24 °C 12 M
48-55% RH 12 F
treadmill
Healthy NS,
5 F follicular and 7 luteal
phase exposure,
regular menstrual cycles, 18
to 35 years old
Observed Effect(s)
Changes in FVC, FEV1; FEF25.75, Vm!K50%, and
Vrmx25% were similar during both the follicular
and luteal phases. No significant difference
between males and females.
Reference
Weinmann
etal. (1995)
X
0.3
0
0.12
0.18
0.24
0.30
0.40
0
0.18
0.30
0.0
0.45
0
0.20
0.30
588
0
235
353
470
588
784
0
353
588
0
882
0
392
588
IhCE
VE = 50 L/min
2.33 h (15' ex/1 5' rest)
VE = 25 L/min/m2 BSA
(one exposure/subject)
1 h (mouthpiece), CE
VE ~ 47 L/min
exposures >4 days apart
2 h, IE (20' ex/20' rest)
VE = 28.5 L/min for M
VE = 26.1 L/min for F
repeated O3 exposures
1 h (mouthpiece) IE
(50' rest/10' ex first 7 M)
(20' rest/ 10' ex all others)
VE ~ 28 L/min for M
VE ~ 23 L/min for F
NA
22 °C
40% RH
treadmill
21 to 25 °C
45 to 60% RH
cycle
23.1 °C
46.1%RH
cycle/treadmill
=22 °C
>75% RH
treadmill
9F
30 to 33 F and 30 to
33 M in each
concentration group;
total of
372 individuals
participated
14 F
14 F
10M
6F
9M, 10 F
Healthy NS, regular
menstrual cycles,
20 to 34 years old
Healthy NS, 18 to 35 years
old, blacks and whites
FVC = 5. 11 ±0.53L,
NS, 20 to 24 years old
FVC = 3.74 ± 0.30 L,
NS, 19 to 23 years old
Healthy NS,
60 to 89 years old
Healthy NS,
64 to 71 years old
Healthy NS,
55 to 74 years old
FEV; decreased 13.1% during the mid-luteal
phase and 18.1% during the follicular phase.
Decrement in FEF25.75 was significantly larger
during the follicular phase than the mid-luteal
phase. Changes in FVC were similar in
both phases.
Decrements in FEV1; increases in SRaw and
cough, correlated with O3 concentration.
There were no significant differences between
the responses of males and females.
Small lung group, FVC = 3.74 ± 0.30 L.
Large lung group, FVC = 5.1 1 ± 0.53 L.
Significant concentration-response effect on
FVC and FEVjj lung size had no effect on
percentage decrements in FVC or FEVj.
Mean decrement in FEVj = 5.7%.
Decrements in FVC and FEVj were the only
pulmonary functions significantly altered by
O3 exposure. No significant differences
between responses of men and women.
No change in any spirometic measure for
either group. Females had 13% increase in RT
after 0.30-ppm exposure. Gender differences
not evaluated.
Fox etal.
(1993)
Seal et al.
(1993)
Messineo
and Adams
(1990)
Bedi et al.
(1989)
Reisenauer
etal. (1988)
-------
Table AX6-6 (cont'd). Gender and Hormonal Differences in Pulmonary Function Responses to Ozone"
Ozone
Concentration11
ppm ug/m3 and Activity
0.3 588 1 h (mouthpiece), CE
VE ~ 70 L/min for men
VE ~ 50 L/min for
women
0.0 0 2 h, IE (20' rest/ 20' ex)
0.45 882 VE = 27.9 L/min for M
VE = 25 .4 L/min for F
repeated O3 exposures
0.48 941 2h, IE
VE = 25 L/min
Oi
Exposure
Conditions'
21 to 25 °C
45 to 60% RH
cycle
24 °C
58% RH
cycle
21 °C
(WBGT)
cycle
Number and Gender Subject
of Subjects Characteristics
20 M NS, 18 to 30 years old
20 F NS, 19 to 25 years old
8M Healthy NS,
5 1 to 69 years old
8F Healthy NS,
56 to 76 years old
10 F Healthy NS,
19 to 36 years old
Observed Effect(s)
Significant decrements in FVC, FEVj, and
FEF25.75 following O3 exposure.
No significant differences between men and
women for spirometry or SRaw.
Range of responses in FEVj:
0 to - 12% (mean = -5.6%).
No significant difference in responses of men
and women.
Tendency for women to have greater effects.
Mean decrement in FEV; = 22.4%.
Significant decrements in all spirometric
measurements. Results not significantly
different from a similar study on males
(Drechsler-Parks et al., 1984).
Reference
Adams et al.
(1987)
Drechsler-
Parks et al.
(1987a,b)
Horvath
etal. (1986)
* See Appendix A for abbreviations and acronyms.
b Listed from lowest to highest O3 concentration.
CWBGT = 0.7 Twetbulb + 0.3 T^,^,^.
-------
Parks et al., 1987a; Messineo and Adams, 1990; Seal et al., 1993; Weinmann et al., 1995)
although in some studies females appeared to experience a slightly greater decline then males
(Drechsler-Parks et al., 1987a; Messineo and Adams, 1990). The comparative evaluations were
based on responses that included spirometry, airway resistance, nonspecific bronchial
responsiveness (NSBR) determinations, and changes in frequency and severity of respiratory
symptoms. However, depending on how the O3 dose was calculated and normalized, the
findings of at least three studies may be interpreted as showing that females are more sensitive
to O3 than males. The findings of the seven studies are presented in detail in Section 7.2.1.3 of
the previous O3 criteria document (U.S. Environmental Protection Agency, 1996).
Some support for a possible increased sensitivity of females to O3 comes from a study of
uric acid concentration in nasal lavage fluid (NLF). Housley et al. (1996) found that the NLF of
females contains smaller amounts of uric acid than the NLF of males. The primary source of
uric acid is plasma; therefore, lower nasal concentrations would reflect lower plasma
concentrations of this antioxidant. The authors have speculated that in females, both lower
plasma and NLF levels (of uric acid) can plausibly make them more susceptible to oxidant
injury, since local antioxidant protection may not be as effective as with higher levels of uric
acid, and consequently more free O3 can penetrate deeper into the lung.
Several studies also have suggested that anatomical differences in the lung size and the
airways between males and females, and subsequent differences in O3 distribution and
absorption, may influence O3 sensitivity and potentially differential O3 response. The study of
Messinio and Adams (1990) have, however, convincingly demonstrated that the effective dose to
the lung, and not the lung size, determines the magnitude of (FEVj) response. Furthermore,
the O3 dosimetry experiments of Bush et al. (1996) have shown that despite gender differences in
longitudinal distribution of O3, the absorption distribution in conducting airways was the same
for both sexes when expressed as a ratio of penetration to anatomic dead space volume. This
implies that gender differences, if any, are not due to differences in (normal) lung anatomy.
The data also have shown that routine adjustment of O3 dose for body size and gender
differences would be more important if normalized to anatomic dead space rather than the usual
FVCorBSA.
One of the secondary objectives of a study designed to examine the role of neural
mechanisms involved in limiting maximal inspiration following O3 exposure has been to
AX6-55
-------
determine if gender differences occur. A group of healthy males (n = 28) and females (n = 34)
were exposed to 0.42 ppm O3 for 2 h with IE. The methodological details of the study are
presented in Section AX6.2.5.1 of this document. As Figure AX6-4 shows, the differences
between males and females were, at any condition, measurement point, and O3 sensitivity status
only minimal and not significant (Passannante et al., 1998).
In another investigation, Folinsbee and Hazucha (2000) exposed a group of
19 O3-responsive young females (average age of 22 years, prescreened for O3 responsiveness by
earlier exposure) to air and 0.35 ppm O3. The randomized 75-min exposures included two
30-min exercise periods at a VE of 40 L/min. In addition to standard pulmonary function tests,
they employed several techniques used for the first time in human air pollution studies
assessment of O3 effects. The average lung function decline from a pre-exposure value was 13%
for FVC, 19.9 % for FEVj, and 30% for FEF25.75. The infrequently measured forced inspiratory
vital capacity (FIVC) was the same as FVC suggesting that the lung volume limiting
mechanisms are the same. The reduction in peak inspiratory flow (PIF) most likely reflects an
overall reduction in inspiratory effort associated with neurally mediated inhibition of inspiration.
Persistence of small inspiratory and expiratory spirometric effects, airway resistance, and airway
responsiveness to methacholine for up to 18 h postexposure suggests that recovery of pulmonary
function after O3 exposure involves more than the simple removal of an irritant. Incomplete
repair of damaged epithelium and still unresolved airway inflammation are the likely causes of
the residual effects that in some individuals persisted beyond 24 h postexposure. However, by
42 hours no residual effects were detected. No significant changes were found in ventilatory
response to CO2 between air and O3 exposures, suggesting that chemoreceptors were not affected
by O3. However, O3 inhalation did result in accelerated timing of breathing and a modest
increase in inspiratory drive. These observations are consistent with, and further supportive of,
the primary mechanisms of O3-induced reduction in inspiratory lung function, namely an
inhibition of inspiration elicited by stimulation of the C-fibers and other pulmonary receptors.
Because the measures of inspiratory and chemical drive to assess O3 effects were not reported in
any previous human study, no comparisons are possible. Because no male subjects were
recruited for the study, it is not possible to compare gender effects. Despite being O3-responsive,
however, the average post-O3 decline in expiratory lung function from preexposure (13% for
FVC; 19.9% for FEV^ 30% for FEF25.75) was similar to that seen in female cohorts studied by
AX6-56
-------
other investigators under similar conditions of exposure. These were the same studies that found
no gender differences in O3 sensitivity (Adams et al., 1987; Messineo and Adams, 1990).
The study by Hazucha et al. (2003), discussed in the previous section, has in addition to
aging also examined gender differences in O3 responsiveness. The male (n = 146) and female
(n = 94) cohorts were classified into young (19 to 35 year-old) and middle-aged (35 to 60 year-
old) groups. This classification was selected in order to facilitate comparison with data reported
previously by other laboratories. Using a linear regression spline model (with a break point at
35 years), the authors reported that the rate of loss of sensitivity is about three times as high in
young females as in young males (p < 0.003). In young females, the average estimated decline
in FEVj response is 0.71% per year, while in young males it is 0.19% per year. Middle-aged
groups of both genders show about the same rate of decline (0.36 to 0.39%, respectively).
At 60 years of age, the model estimates about a 5% post-O3 exposure decline in FEVj for males,
but only a 1.3% decline for females. These observations suggest that young females lose O3
sensitivity faster than young males, but by middle age, the rate is about the same for both
genders. Descriptive statistics show that there were practically no differences in the mean value,
standard error of the mean, and coefficient of variation for % FEVj decrement between the group
of young males (n = 125; 83.7 ± 1.1%; CV = 13.5%) and young females (n = 73; 83.4 ± 1.25%;
CV = 12.8%). A straight linear regression model of these data was illustrated in Figure AX6-7.
The slopes, significant in both males (r = 0.242; p = 0.003) and females (r = 0.488; p = 0.001),
represent the decline in responsiveness of 0.29% and 0.55% per year respectively, as assessed
by FEVi.
Two earlier studies of the effects of the menstrual cycle phase on O3 responsiveness have
reported conflicting results (U.S. Environmental Protection Agency, 1996). Weinmann et al.
(1995) found no significant lung function effects related to menstrual cycle, although during the
luteal phase the effects were slightly more pronounced than during the follicular phase; while
Fox et al., (1993) reported that follicular phase enhanced O3 responsiveness. In a more recent
investigation of possible modulatory effects of hormonal changes during menstrual cycle on O3
response, young women (n = 150) 18 to 35 years old were exposed once to one of multiple O3
concentrations (0.0, 0.12, 0.18, 0.24, 0.30, 0.40 ppm) for 140 min with IE at 35 L/min/m2 BSA.
The women's menstrual cycle phase was determined immediately prior to O3 exposure. Post-O3,
no significant differences in % predicted FEVj changes that could be related to the menstrual
AX6-57
-------
cycle phase were found. Admittedly, a less precise method of determining menstrual cycle
phase used in this study could have weakened the statistical power. Unfortunately, the direction
and magnitude of O3 response as related to the menstrual cycle phases were not reported (Seal
et al., 1996). Considering the inconclusiveness of findings of this study and the inconsistency of
results between the two earlier studies, it is not possible to make any firm conclusions about the
influence of the menstrual cycle on responses to O3 exposure.
Additional studies presented in this section clarify an open-ended conclusion reached in the
previous O3 criteria document (U.S. Environmental Protection Agency, 1996) regarding the
influence of age on O3 responsiveness. Healthy young males and females are about equally
responsive to O3, although the rate of loss of sensitivity is higher in females than in males.
Middle-aged men and women are generally much less responsive to O3 than younger individuals.
Within this range, males appear to be slightly more responsive than females, but the rate of age-
related loss in FEVj is about the same. The O3 sensitivity may vary during the menstrual cycle;
however, this variability appears to be minimal.
AX6.5.3 Racial, Ethnic, and Socioeconomic Status Factors
In the only laboratory study designed to compare spirometric responses of whites and
blacks exposed to a range of O3 concentrations (0 to 0.4 ppm), Seal et al. (1993) reported
inconsistent and statistically insignificant FEVj differences between white and black males and
females within various exposure levels. Perhaps, with larger cohorts the tendency for greater
responses of black than white males may become significant. Thus, based on this study it is still
unclear if race is a modifier of O3 sensitivity, although the findings of epidemiologic studies
reported in the previous criteria document "can be considered suggestive of an ethnic difference"
(U.S. Environmental Protection Agency, 1996). However, as Gwynn and Thurston (2001)
pointed out, it appears that it is more the socioeconomic status (SES) and overall quality of
healthcare that drives PM10- and O3-related hospital admissions than an innate or acquired
sensitivity to pollutants.
This assertion is somewhat supported by the study of Seal et al. (1996) who employed a
family history questionnaire to examine the influence of SES on the O3 responsiveness of
352 healthy, 18- to 35-year-old black and white subjects. Each subject was exposed once under
controlled laboratory conditions to either air or 0.12, 0.18, 0.24, 0.30, 0.40 ppm O3 for 140 min
AX6-58
-------
with 15 min IE at 35 L/min/m2 BSA. An answer to the "Education of the father" question was
selected as a surrogate variable for SES status. No other qualifying indices of SES were used or
potential bias examined. Of the three SES categories, individuals in the middle SES category
showed greater concentration-dependent decline in % predicted FEVj (4-5% @ 0.4 ppm O3) than
low and high SES groups. The authors did not have an "immediately clear" explanation for this
finding. The SES to %predicted FEVj relationship by gender-race group was apparently
examined as well; however, these results were not presented. Perhaps a more comprehensive
and quantitative evaluation of SES status would have identified the key factors and clarified the
interpretation of these findings. With such a paucity of data it is not possible to discern the
influence of racial or other related factors on O3 sensitivity.
AX6.5.4 Influence of Physical Activity
Apart from the importance of increased minute ventilation on the inhaled dose of O3 during
increased physical activity, including work, recreational exercise, and more structured exercise
like sports, no systematic effort has been made to study other potential physical factors that may
modulate O3 response. The typical physiologic response of the body to exercise is to increase
both the rate and depth of breathing, as well as increase other responses such as heart rate, blood
pressure, oxygen uptake, and lung diffusion capacity.
Physical activity increases minute ventilation in proportion to work load. At rest, and
during light exercise, the dominant route of breathing is through the nose. The nose not only
humidifies air, among other physiologic functions, but also absorbs O3 thus decreasing the
overall dose. As the intensity of exercise increases, the minute ventilation increases and the
breathing switches from nasal to oronasal mode. There is considerable individual variation in
the onset of oronasal breathing, which ranges from 24 to 46 L/min (Niinimaa et al., 1980).
During heavy exercise, ventilation is dominated by oral breathing. Consequently, the residence
time of inhaled air in the nose and the airways is shorter, reducing the uptake of O3 (Kabel et al.,
1994). Moreover, increasing inspiratory flow and tidal volume shifts the longitudinal
distribution of O3 to the peripheral airways, which are more sensitive to injury than the larger,
proximal airways. Ozone uptake studies of human lung showed that at simulated quiet
breathing, 50% of O3 was absorbed in the upper airways, 50% in the conducting airways, and
none reached the small airways (Hu et al. 1994). With ventilation simulating heavy exercise
AX6-59
-------
(60 L/min), the respective O3 uptakes were 10% (upper airways), 65% (conducting airways), and
25% (small airways). These observations imply that equal O3 dose (C x T x VE) will have a
greater effect on pulmonary function and inflammatory responses when inhaled during heavy
physical activity than when inhaled during lighter activity. Although, Ultman et al. (2004)
recently reported that spirometric response are not correlated with O3 uptake. (See Chapter 4 of
this document for more information on the dosimetry ofO3.)
Other physiologic factors activated in response to physical activity are unlikely to have as
much impact on O3 responsiveness as does minute ventilation; however, their potential influence
has not been investigated.
AX6.5.5 Environmental Factors
Since the 1996 O3 criteria document, not a single human laboratory study has examined the
potential influence of environmental factors such as rural versus urban environment, passive
cigarette smoke exposure, and bioactive admixtures such as endotoxin on healthy individual's
pulmonary function changes due to O3 (U.S. Environmental Protection Agency, 1996).
Some of the unresolved issues, e.g., health effects of ETS and O3 interaction, which need to
be examined in human studies were explored very recently in laboratory animal studies (see
Chapter 5 for more details). In one study on mice, preexposure of animals to sidestream
cigarette smoke (ETS surrogate), which elicited no immediate effects, resulted in a potentiation
of subsequent O3-induced inflammatory response. This finding suggests that typical adverse
effects of ETS do not necessarily have to elicit an immediate response to ETS, but may in fact
potentiate the effects of a subsequent exposure to another pollutant like O3 (Yu et al., 2002). The
key mechanism by which smoke inhalation may potentiate subsequent oxidant injury appears to
be damage to cell membranes and the resulting increase in epithelial permeability. Disruption of
this protective layer may facilitate as well as accelerate injury to subepithelial structures when
subsequently exposed to other pollutants (Bhalla, 2002). Although this may be a plausible
mechanism in nonsmokers and acute smokers exposed to ETS and other pollutants, studies
involving chronic smokers who most likely already have chronic airway inflammation do not
seems to show exaggerated response with exposure to O3.
More than 25 years ago, Hazucha et al. (1973) reported that the spirometric lung function
of smokers declined significantly less than that of nonsmokers when exposed to 0.37 ppm O3.
AX6-60
-------
The findings of this study have been confirmed and expanded (Table AX6-7). Frampton et al.
(1997a) found that exposure of current smokers (n = 34) and never smokers (n = 56) to
0.22 ppm O3 for 4 h with IE for 20 min of each 30 min period at 40 to 46 L/min, induced a
substantially smaller decline in FVC, FEVj and SGaw of smokers than never smokers. Smokers
also demonstrated a much narrower distribution of spirometric endpoints than never smokers.
Similarly, nonspecific airway responsiveness to methacholine was decreased in smokers.
However, both groups showed the consistency of response from exposure to exposure. It should
be noted that despite seemingly lesser response, the smokers were more symptomatic post air
exposure than never smokers, but the opposite was true for O3 exposure. This would suggest
that underlying chronic airway inflammation present in smokers has blunted stimulation of
bronchial C-fibers and other pulmonary receptors, the receptors substantially responsible for
post-O3 lung function decrements. In addition to desensitization, the other "protective"
mechanisms active in smokers may be an increase in the mucus layer conferring not only a
mechanical protection, but also acting as an O3 scavenger. Another plausible explanation of a
diminished responsiveness of smokers may be related to elevated levels of reduced glutathione
(GSH), an antioxidant, found in epithelial lining fluid of chronic but not acute smokers (MacNee
etal., 1996).
Despite some differences in a release of proinflammatory cytokines and subsequent
recruitment of inflammatory cells, both smokers and nonsmokers developed airway
inflammation following O3 exposure. This was demonstrated by the Torres et al. (1997) study
that involved exposures of about equal size cohorts of otherwise healthy young smokers,
nonsmoker O3 nonresponders (<5% FEVj post-O3 decrement) and nonsmoker O3 responders
(>15% FEVj post-O3 decrement) to air and two 0.22 ppm O3 atmospheres for 4 hours, alternating
20 min of moderate exercise (25 L/min/m2 BSA) with 10 min of rest. Both O3 exposures were
followed by nasal lavage (NL) and bronchoalveolar lavage (BAL) performed immediately post
one of exposures and 18 hr later following the other exposure. Neither O3 responsiveness nor
smoking alters the magnitude or the time course of O3-induced airway inflammation. The
overall cell recovery was lower immediately postexposure but higher, particularly in
nonsmokers, 18 h post-O3 exposure when compared to control (air) in all groups. Recovery of
lymphocytes, PMNs and AMs in both alveolar and bronchial lavage fluid showed the largest
increase in response to O3 in all groups, with nonsmokers showing greater relative increases than
AX6-61
-------
Table AX6-7. Influence of Ethnic, Environmental, and Other Factors
X
Oi
ON
to
Ozone
Concentration
ppm Hg/m3
0.0 0
0.4 780
0.0 0
0.12 235
0.0 0
0.22 431
0.22 431
0.0 0
0.12- 235-470a
0.24a
Exposure
Duration
and Activity
2h
IE, 15' ex/15' rest
VE = 20 L/min/m2
BSA
0.75 h
IE, 15' ex/15' rest
VE = 40-46 L/min
4h
IE, 20' ex/10' rest
VE = 40-46 L/min
2.17 h
IE, 10' ex/ 10' rest
VE = 36-39 L/min
Number
Exposure and Gender Subject
Conditions of Subjects Characteristics
20 °C 15 M, 1 F Placebo group:
40% RH healthy NS
avg age 27 yrs.
13 M, 2 F Antiox. Suppl.
Gr: Healthy NS
avg age 27 yrs.
60% RH 5 M, 12 F Asthmatics
sensitive to SO2
19 to 38 yrs old
21°C 25(M/F) Healthy NS
37% RH O3 responders and
nonresponders
18 to 40 yrs old
22 °C or 30 °C 5 M, 4 F Healthy NS
45-55% RH 24 to 32 yrs old
Observed Effect(s)
PF decrements in the supplementation
group were signif. smaller for FVC
(p < 0.046) and near significant for FEV{
(p < 0.055). The inflammatory response
(BAL) showed no significant differences
between the two groups either in the
recovery of cellular components or the
concentrations and types of inflammatory
cytokines.
No significant differences due to O3
between placebo and antioxidant
supplement cohort in either
spirometric responses or bronchial
hyperresponsiveness to 0.1 ppm SO2.
Glutathione peroxidase (GPx) activity and
eGPx protein level were significantly
(p = 0.0001) depleted in ELF for at least
18 h postexp. In BAL both endpoints
were elevated (ns). No association
between cell injury, PF, or GPx activity.
FEVj decreased (p < 0.5) by -8% at
22 °C and -6.5% at 30 °C. 19 h postexp
decline of 2.3% still signif. (p < 0.05).
SGaw signif. (p < 0.05) declined at 30 °C
but not at 22 °C. The AHR assessed 19 h
postexp. as PC50 sGaw methacholine
signif. (p < 0.05) higher at both
temperatures.
Reference
Samet et al.
(2001)
Trenga et al.
(2001)
Avissar et al.
(2000)b
Foster et al.
(2000)
-------
Table AX6-7 (cont'd). Influence of Ethnic, Environmental, and Other Factors
Ozone
Concentration
ppm Hg/ni3
0.0 0
0.40 780
Exposure Number
Duration Exposure and Gender Subject
and Activity Conditions of Subjects Characteristics
2 h 6 M, 9 F Healthy NS
IE, 20' ex/ 10' rest avg age 3 1 yrs.
Observed Effect(s)
Corticosteroid pretreatment had no effects
on post-O3 decline in PF, PMN response,
Reference
Nightingale
et al. (2000)
VE = mild to mod.
and sputum cell count under both the
placebo and treatment conditions.
Methacholine PC20 FEV{ was equally
decreased in both cond. 4 h after
exposure. No changes in exhaled NO
and CO.
X
Oi
0.0 0
0.22 431
0.0 0
0.22 431
0.22 431
4h 21 °C
IE 37% RH
20' ex/10' rest
VE = 40-46 L/min
4h 21 °C
IE, 20' ex/10' rest 37% RH
VE = 25 L/min/m2
BSA
90 M 56 never smokers
34 current
smokers
18 to 40 yrs. old
10 M, 2 F NS, O3 nonresp.,
avg age 25 yrs.;
10 M, 3 F NS, O3 resp.,
avg age 25 yrs;
1 1 M, 2 F smokers avg
age 28 yrs
Smokers are less responsive to O3 as
assessed by spirometric and
plethysmographic variables. Neither age,
gender, nor methacholine responsiveness
were predictive of O3 response.
Neither O3 responsiveness nor smoking
has altered the magnitude and the time
course of O3-induced airway
inflammation. Inflammation involved
all types of cells accessible by B AL.
The recovery profile of these cells over
time was very similar for all groups
showing highest values 18 h
postexposure.
Frampton et al.
(1997a,c)b
Torres et al.
(1997)b
Frampton et al.
(1997a,c)b
aRamp exposure from 0.12 ppm to 0.24 ppm and back to 0.12 ppm at the end of exposure.
bRelated studies, sharing of some subjects .
-------
smokers. Of the two cytokines, IL-6 and IL-8, IL-6 was substantially and significantly
(p < 0.0002) elevated immediately postexposure but returned back to control 18 h later in all
groups; but only nonsmokers' effects were significantly higher (p < 0.024). IL-8 showed a
similar pattern of response but the increase in all groups, though still significant (p < 0.0001),
was not as high as for IL-6. Between group differences were not significant. This inflammatory
response involved all types of cells present in BAL fluid and the recovery profile of these cells
over time was very similar for all groups. In contrast to BAL, NL did not prove to be a reliable
marker of airway inflammation. The lack of association between lung function changes
(spirometry) and airway inflammation for all three groups confirms similar observations
reported from other laboratories. This divergence of mechanisms is further enhanced by an
observation that a substantially different spirometric response between O3 responders and
nonresponders, the airway inflammatory response of the two groups was very similar, both in
terms of magnitude and pattern (Torres et al., 1997).
The influence of ambient temperature on pulmonary effects induced by O3 exposure in
humans has been studied infrequently under controlled laboratory conditions. Several
experimental human studies published more than 20 years ago reported additive effects of heat
and O3 exposure (see U.S. Environmental Protection Agency, 1986, 1996). In the study of
Foster et al. (2000) 9 young (mean age 27 years) healthy subjects (4F/5M) were exposed for
130 min (IE 10 min @ 36 to 39 1/min) to filtered air and to ramp profile O3 at 22° and 30 °C,
45-55% RH. The order of exposures was randomized. The O3 exposure started at 0.12 ppm,
reached the peak of 0.24 ppm mid-way through and subsequently declined to 0.12 ppm at the
end of exposure. Ozone inhalation decreased VT and increased fB as compared to baseline at
both temperatures. At the end of exposure FEVj decreased significantly (p < 0.5) by -8% at
22 °C and -6.5% at 30 °C. One day (19 h) later, the decline of 2.3% from baseline was still
significant (p < 0.05). FVC decrements were smaller and significant only at 22 °C immediately
postexposure. SGaw significantly (p < 0.05) declined at 30 °C but not at 22 °C. A day later,
sGaw was elevated above the baseline for all conditions. The nonspecific bronchial
responsiveness (NSBR) to methacloline assessed as PC50 sGaw was significantly (p < 0.05)
higher one day following O3 exposure at both temperatures but more so at 30 °C. Thus, these
findings indicate that elevated temperature has partially attenuated spirometric response but
enhanced airway reactivity. Numerous studies have reported an increase in NSBR immediately
AX6-64
-------
after exposure to O3. Whether the late NSBR reported in this study is a persistent residual effect
of an earlier increase in airway responsiveness, or is a true one day lag effect cannot be
determined from this study. Whatever the origin, however, a delayed increase in airway
responsiveness raises a question of potentially increased susceptibility of an individual to
respiratory impairment, particularly if the suggested mechanism of disrupted epithelial
membrane holds true.
AX6.5.6 Oxidant-Antioxidant Balance
Oxidant-antioxidant balance has been considered as one of the determinants of O3
responsiveness. Amateur cyclists who took antioxidant supplements (vitamins C, E, and
p-carotene) for three months showed no decrements in spirometric lung function when cycling
on days with high O3 levels. In contrast, matched control group of cyclists not pretreated with
vitamin supplements experienced an almost 2% decline in FVC and FEVj and >5% reduction in
PEF during the same activity period. Adjustment of data for confounders such as PM10 and NO2
did not change the findings. Apparently, substantially elevated levels of plasma antioxidants
may afford some protection against lung function impairment (Grievink et al.,1998, 1999).
Both laboratory animal and human studies have repeatedly demonstrated that antioxidant
compounds present the first line of defense against the oxidative stress. Thus, upregulation of
both enzymatic and nonenzymatic antioxidant systems is critical to airway epithelial protection
from exposure to oxidants such as O3 and NO2 (see Table AX6-7). As an extension of an earlier
study focused on pulmonary function changes (Frampton et al., 1997a), Avissar et al. (2000)
hypothesized that concentration of glutathione peroxidase (GPx), one of the antioxidants in
epithelial lining fluid (ELF), is related to O3 and NO2 responsiveness. They exposed healthy
young nonsmokers (n = 25), O3-responders, and nonresponders to filtered air and twice to
0.22 ppm O3 for 4 h (IE, 20' ex 710' rest, @ VE 40 to 46 L/min). In the NO2 part of the study,
subjects were exposed to air and twice to NO2 (0.6 and 1.5 ppm) for 3 h, with IE of 10 min of
each 30 min @ VE of 40 L/min. Ozone exposure elicited a typical pulmonary function response
with neutrophilic airway inflammation in both responders and nonresponders. The GPx activity
was significantly reduced (p = 0.0001) and eGPx protein significantly depleted (p = 0.0001) in
epithelial lining fluid (ELF) for at least 18 h postexposure. In contrast, both GPx and eGPx were
slightly elevated in bronchoalveolar lavage fluid (BALF). However, neither of the two NO2
AX6-65
-------
exposures had a significant effect on pulmonary function, airway neutrophilia, epithelial
permeability, GPx activity, or eGPx protein level in either ELF or BALF. The lack of a
significant response to NO2 has been attributed to the weak oxidative properties of this gas.
No association has been observed between cell injury, assessed by ELF albumin, or pulmonary
function and GPx activity for O3 exposure. Thus, it is unclear what role antioxidants may have
in modulation of O3-induced lung function and inflammatory responses. The authors found a
negative association between lower baseline eGPx protein concentration in ELF and post-O3
neutrophilia to be an important predictor of O3-induced inflammation; however, the causal
relationship has not been established.
The effects of dietary antioxidant supplementation on O3-induced pulmonary and
inflammatory response of young healthy individuals has been investigated by Samet et al.
(2001). Under controlled conditions, subjects received ascorbate restricted diet for three weeks.
After the first week of prescribed diet, subjects were randomly assigned into two groups, and
exposed to air (2 h, IE every 15 min at 20 L/min/m2 BSA). Thereafter, one group received daily
placebo pills and the other a daily supplement of ascorbate, a-tocopherol and a vegetable juice
for the next two weeks. At the end of a two week period subjects were exposed to 0.4 ppm O3
under otherwise similar conditions as in sham exposures. Serum concentration of antioxidants
determined prior to O3 exposure showed that subjects receiving supplements had substantially
higher concentrations of ascorbate, tocopherol and carotenoid in blood than the control group.
Plasma levels of glutathione and uric acid (cellular antioxidants) remained essentially the same.
Ozone exposure reduced spirometric lung function in both groups; however, the average
decrements in the supplementation group were smaller for FVC (p = 0.046) and FEVj
(p = 0.055) when compared to the placebo group. There was no significant correlation between
individual lung function changes and respective plasma levels of antioxidants. Individuals in
both groups experienced typical post-O3 subjective symptoms of equal severity. Similarly, the
inflammatory response as assessed by BALF showed no significant differences between the two
groups either in the recovery of cellular components or the types and concentrations of
inflammatory cytokines. Because of the complexity of protocol, the study was not designed as a
cross-over type. However, it is unlikely that the fixed air-O3 sequence of exposures influenced
the findings in any substantial way. Although the study did not elucidate the protective
mechanisms, it has demonstrated the value of dietary antioxidants in attenuating lung function
AX6-66
-------
effects of O3. This observation may appear to contradict the findings of Avissar's and colleagues
study (2000) discussed above; however, neither study found association between lung function
changes and glutathione levels. The lack of such association suggests that activation of
antioxidant protective mechanisms is seemingly independent of mechanisms eliciting lung-
function changes and that dietary antioxidants afford protection via a different pathway than
tissue-dependent antioxidant enzymes. Moreover, the findings of this study have provided
additional evidence that symptomatic, functional, inflammatory, and antioxidant responses are
operating through substantially independent mechanisms.
Further evidence that the levels and activity of antioxidant enzymes in ELF may not be
predictive or indicative of O3-induced lung function or inflammatory effects has been provided
by a study of Blomberg et al. (1999). No association was found between the respiratory tract
lining fluid redox potential level, an indicator of antioxidants balance, and either spirometric or
inflammatory changes induced by a moderate exposure of young individuals to O3 (0.2 ppm/2 h,
intermittent exercise at 20 L/min/m2 BSA). However, O3 exposure caused a partial depletion of
antioxidants (uric acid, GSH, EC-SOD) in nasal ELF and a compensatory increase in plasma uric
acid, affording at least some local protection (Mudway et al. 1999). More recently, Mudway
et al. (2001) investigated the effect of baseline antioxidant levels on response to a 2-h exposure
to 0.2 ppm O3 in 15 asthmatic and 15 healthy subjects. In the BALF of 15 healthy subjects,
significant O3-induced reductions in ascorbate and increases in glutathione disulphide and
EC-SOD were observed, whereas, levels were unaffected by O3 exposure in the asthmatics.
In both groups, BALF levels of uric acid and a-Tocopherol were unaffected by O3.
Trenga et al. (2001) studied the potential protective effects of dietary antioxidants (500 mg
vitamin C and 400 IU of vitamin E) on bronchial responsiveness of young to middle-aged
asthmatics. Recruited subjects were prescreened by exposure to 0.5 ppm SO2 for 10 min while
exercising on a treadmill and selected for study participation if they experienced a >8% decease
in FEVj. Prior to the 1st exposure, subjects took either two supplements or two placebo pills at
breakfast time for 4 weeks. They continued taking respective pills for another week when the
2nd exposure took place. The 45-min exposures to air and 0.12 ppm O3 (15 min IE, VE « 3x
resting rate) via mouthpiece were randomized. Each exposure was followed by two 10-min
challenges to 0.10 and 0.25 ppm SO2 with exercise to determine bronchial hyperresponsiveness.
Due to potential variability of baseline lung function between days, and the way the data have
AX6-67
-------
been presented, it is difficult to interpret the results. All spirometric measures (FEVb
FVC, FEF25_75, and PEF) were significantly decreased from baseline at subsequent time points
following both the FA and O3 exposures. Exposure to O3 caused significant decrements in FEVj
and PEF. The post FA and O3 exposure decrements in lung function were not affected by the
treatment regimen (placebo versus vitamin). Bronchial hyperresponsiveness to 0.1 ppm SO2 was
also unaffected by treatment regimen. Based on the prescreening SO2 challenge, subjects were
ranked by their bronchial responsiveness to SO2 as "less-severe" (8 to 16% FEVj decrements)
and "more-severe" (27 to 44% FEVj decrements). The authors concluded O3 exposure increases
bronchial responsiveness to SO2 in asthmatics and that antioxidant supplementation has a
protective effect against this responsiveness, especially in the "more-severe" responders.
AX6.5.7 Genetic Factors
It has been repeatedly postulated that genetic factors may play an important role in
individual responsiveness to ozone. Recent studies (Bergamaschi et al., 2001; Corradi et al,
2002; Romieu et al, 2004) have indeed found that genetic polymorphisms of antioxidant
enzymes, namely NAD(P)H:quinone oxidoreductase (NQO1) and glutathione-S-transferase Ml
(GSTM1), may play an important role in attenuating oxidative stress of airway epithelium.
Bergamaschi and colleagues (2001) studied young nonsmokers (15 F, 9 M; mean age 28.5 years)
who cycled for two hours on a cycling circuit in a city park on days with the average ozone
concentration ranging from 32 to 103 ppb. There was no control study group nor the intensity of
bicycling has been reported . Since spirometry was done within 30 min post-ride, it is difficult
to gage how much of the statistically significant (p = 0.026) mean decrement of 160 ml in FEVj
of 8/24 individuals with NQO1 wild type (NQOlwt) and GSTMlnull (GSTMlnull) genotypes
was due to ozone. Individuals with other genotype combinations including GSTMlnull had a
mean post-ride decrement of FEVj of only 40 mL. The post-ride serum level of Clara cell
protein (CC16), a biomarker of airway permeability, has been elevated in both subgroups. Only
a "susceptible" subgroup carrying NQOlwt in combination with GSTMlnull genotype, serum
concentration of CC16 showed positive correlation with ambient concentration of ozone and
negative correlation with FEVj changes. Despite some interesting observations, the study results
should be interpreted cautiously.
AX6-68
-------
A subsequent study from the same laboratory was conducted in a more controlled
environment (Corradi et al., 2002). Healthy young (mean 30.1 yrs) individuals (12 M, 10 F)
underwent a single exposure to 0.1 ppm O3 for 2 h while intermittently exercising at a moderate
load on a bicycle ergometer. The study design did not incorporate sham exposure, though the
authors have stated that in a separate experiment the effects of exercise on markers of
inflammation in blood and EEC were negligible. The eight subjects with NQOlwt and
GSTMlnull genotype, the "susceptible" group, indeed showed an increase in markers of
inflammation (IL-6, IL-8, TEARS, LTB4) and oxidative stress (8-isoprostane, H2O2)
immediately post and 18 hrs postexposure. The fourteen subjects with other combination of
genotypes showed small and inconsistent response in EEC and blood biomarkers, though PMN
activity in both groups was significantly increased by exposure. The DNA adduct 8-hydroxy-2'-
deoxyguanosine (8-OHdG), a marker of oxidative DNA damage, was elevated immediately
postexposure in both groups but only in the "susceptible" group the increase became significant.
The spirometric endpoints (not reported) were not affected by the exposure at any time point,
which contrasts the previous study. The incomplete study design calls for a careful
interpretation of the findings.
It is of interest to note, that human nasal mucosa biopsies of GSTM1 deficient subjects
showed higher antioxidant enzymes activity than biopsies of GSTM1 positive individuals when
incubated for 24 h in 120 ppb O3 environment (Otto-Knapp et al., 2003).
The influence of functional polymorphism of inflammatory and other genes on O3
susceptibility was studied by Yang et al. (2005). In this study 54 nonsmoking subjects
(11 healthy subjects, 15 mild asthmatics, 25 with rhinitis) were exposed to 250 ppb O3 for 3 h
(44 subjects), 200 ppb for 4 h (4 subjects), and 400 ppb for 2 h (3 subjects). During these
exposures subjects intermittently exercised (-14 L/min/m2 BSA). The pooled data of the tumor
necrosis factor a (TNF-a), lymphotoxin-a (LTA), Toll-like receptor 4 (TLR4 ), superoxide
dismutase (SOD2) and glutathione peroxidase (GPX1) genes appear to show only TNF-a as a
promising genetic factors of susceptibility. However, as the authors stated "the functional
significance of individual TNF-a polymorphisms remains controversial" (Yang et al., 2005).
More specific genotyping has shown that O3 responsiveness and asthma risk may be related
to the presence of variant Ser allele for NQO1. In a field study of susceptibility to ambient O3 in
Mexico City, 4-17 yrs old asthmatic children (n = 218) were genotyped, including variant alleles
AX6-69
-------
(David et al., 2003). The risk of asthma was related to the 1-h daily max ambient O3 which
ranged from 12 to 309 ppb. Relative to Pro/Pro genotype the presence of at least one NQO1
Ser allele variant lowered the risk of asthma in these children (RR = 0.8). In children with
GSTMlnull genotype combined with at least one NQO1 Ser allele variant the decreased risk of
asthma became statistically significant (RR = 0.4). The presence of Ser allele which renders
NQO1 less active, thus affecting the conjugation of quinones and formation of ROS
subsequently reducing the oxidative stress, may plausibly explain the protective effect of this
genotypic combination.
Another field study of asthmatic children (n = 158) exposed to ambient O3 (12-309 ppb
1-h max during the 12 week study period) has found that in children with genetic deficiency of
GSTM1 the decrements in FEF25.75 were related to the previous day 1-h daily max O3. The
association was more pronounced in moderate to severe asthmatics. Children with GSTMlpos
variant showed no significant decrement in lung function. Randomly administered antioxidant
supplementation (vit. C 250 mg/day and vit. E 50 mg/day) attenuated post-ozone lung function
response in GSTMlnull children (Romieu et al., 2004).
These recent studies have shown that individual's innate susceptibility to ozone may be
linked to genetic background of an individual. Although a number of potential ozone
susceptibility genes have been identified, additional better designed and controlled studies are
needed to ascertain the link between susceptibility and polymorphism.
AX6.6 REPEATED EXPOSURES TO OZONE
Repeated daily exposure to O3 in the laboratory for 4 or 5 days leads to attenuated changes
in pulmonary function responses and symptoms (Hackney et al., 1977a; U.S. Environmental
Protection Agency, 1986, 1996). A summary of studies investigating FEVj responses to
repeated daily exposure for up to 5 days is given in Table AX6-8. The FEVj responses to
repeated O3 exposure typically have shown an increased response on the second exposure day
(Day 2) compared to the initial (Day 1) exposure response. This is readily apparent in repeated
exposures to a range of concentrations from 0.4 to 0.5 ppm O3 accompanied by moderate
exercise (Folinsbee et al., 1980; Horvath et al., 1981; Linn et al., 1982), and at lower
concentrations, 0.20 to 0.35 ppm, when accompanied by heavy exercise (Brookes et al., 1989;
AX6-70
-------
Table AX6-8. Changes in Forced Expiratory Volume in One Second After Repeated Daily Exposure to Ozone"
X
Ozone
Concentration1"
ppm
0.12
0.20
0.20
0.20
0.20
0.25
0.35
0.35
0.35
0.35
0.40
0.40
0.4
0.4
0.42
0.45
0.45
0.47
0.5
0.5
ug/m3
235
392
392
392
392
490
686
686
686
686
784
784
784
784
823
882
882
921
980
980
Exposure Duration
and Activity0
6.6 h, IE (40)
2 h, IE (30)
2 h, IE (18 and 30)
2 h, IE (18 and 30)
1 h, CE (60)
1 h, CE (63)
2 h, IE (30)
1 h, CE (60)
1 h, CE (60)
1 h, CE (60)
3 h, IE (4-5 x resting)
3 h, IE (4-5 x resting)
2 h, IE (65)
3 h, IE (32)
2 h, IE (30)
2 h, IE (27)
2 h, IE (27)
2 h, IE (3 x resting)
2 h, IE (30)
2. 5 h, IE (2 x resting)
Number and Gender Percent Change in FEVj on Consecutive
of Subjects Exposure Days References'1
17 M
10M
8M, 13 F
9
15M
4M,2F
5M,2F
10M
8M
10M
10M
15M
13M5
HFf
8M
8 M, 2 Fh
24 M
1M, 5F
10M,6F
8 M, 2 Fg
8M
6
First
-12.79
+1.4
-3.0
-8.7
-5.02
-20.2
-18.8
-5.3
-31.0
-16.1
-14.4
-15.9
-9.2
-8.8
-18.0
-34.7
-21.1
-13.3
-5.8
-11.4
-8.7
-2.7
Second
-8.73
+2.7
-4.5
-10.1
-7.8
-34.8
—
-5.0
-41.0
-30.4
—
-24.6
-10.8
-12.9
-29.9
-31.1
-26.4
—
-5.6
-22.9
-16.5
-4.9
Third
-2.54
-1.6
-1.1
-3.2
—
—
-22.3
-2.2
-33.0
—
-20.6
—
-5.3
-4.1
-21.1
-18.5
-18.0
-22.8
-1.9
-11.9
-3.5
-2.4
Fourth Fifth
-0.6 0.2 Folinsbeeetal. (1994)
— — Folinsbeeetal. (1980)
— — Glineretal. (1983)
— — Glineretal. (1983)
— — Brookes etal. (1989)
— — Folinsbee and Horvath( 1986)
— —
— — Folinsbeeetal. (1980)
-25.0 — Foxcroft and Adams (1986)
— — Schonfeld etal. (1989)
— —
— — Brookes etal. (1989)
-0.7 -1.0 Kulle etal. (1982)
-3.0 -1.6 Kulle etal. (1982)
-7.0 -4.4 Folinsbeeetal. (1998)
-12.0 -6.2 Gong etal. (1997b)
-6.3 -2.3 Horvath etal. (1981)
— — Bedi etal. (1985)
— — Bedi etal. (1989)
-4.3 — Linn etal. (1982)
— — Folinsbeeetal. (1980)
-0.7 — Hackney etal. (1977a)
aSee Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
'Exposure duration and intensity of IE or CE were variable; VE (number in parentheses) given in liters per minute or as a multiple of resting ventilation.
*For a more complete discussion of these studies, see Table AX6-9 and U.S. Environmental Protection Agency (1986).
'Subjects were especially sensitive on prior exposure to 0.42 ppm O3 as evidenced by a decrease in FEV; of more than 20%. These nine subjects are a subset of the total group of
21 individuals used in this study.
'Bronchial reactivity to a methacholine challenge also was studied.
gSeven subjects completed entire experiment.
Subjects had mild asthma.
-------
Folinsbee and Horvath, 1986; Foxcroft and Adams, 1986; Schonfeld et al., 1989). Mechanisms
for enhanced pulmonary function responses on Day 2 have not been established, although
persistence of acute O3-induced damage for greater than 24 h may be important (Folinsbee et al.,
1993). An enhanced Day 2 FEVj response was less obvious or absent in exposures at lower
concentrations or those that caused relatively small group mean O3-induced decrements.
For example, Bedi et al. (1988) found no enhancement of the relatively small pulmonary
function responses in older subjects (median age, 65 years) exposed repeatedly to O3. Three
reports (Bedi et al., 1985; Folinsbee and Horvath, 1986; Schonfeld et al., 1989) demonstrated
that enhanced pulmonary function responsiveness was present within 12 h, lasted for at least
24 h and possibly 48 h, but was absent after 72 h.
After 3 to 5 days of consecutive daily exposures to O3, FEVj responses are markedly
diminished or absent. One study (Horvath et al., 1981) suggested that the rapidity of this decline
in FEVj response was related to the magnitude of the subjects' initial responses to O3 or their
"sensitivity." A summary of studies examining the effects of repeated exposures to O3 on FEVj
and other pulmonary function, symptoms, and airway inflammation is given in Table AX6-9.
Studies examining persistence of the attenuation of pulmonary function responses following
4 days of repeated exposure (Horvath et al., 1981; Kulle et al., 1982; Linn et al., 1982) indicate
that attenuation is relatively short-lived, being partially reversed within 3 to 7 days and typically
abolished within 1 to 2 weeks. Repeated exposures separated by 1 week (for up to 6 weeks)
apparently do not induce attenuation of the pulmonary function response (Linn et al., 1982).
Gong et al. (1997b) studied the effects of repeated exposure to 0.4 ppm O3 in a group of mild
asthmatics and observed a similar pattern of responses as those seen previously in healthy
subjects. The attenuation of pulmonary responses reached after 5 days of consecutive O3
exposure was partially lost at 4 and 7 days postexposure.
In addition to the significant attenuation or absence of pulmonary function responses after
several consecutive daily O3 exposures, symptoms of cough and chest discomfort usually
associated with O3 exposure generally are substantially reduced or absent (Folinsbee et al., 1980,
1994; Foxcroft and Adams, 1986; Linn et al., 1982). Airway responsiveness to methacholine is
increased with an initial O3 exposure (Holtzman et al., 1979; Folinsbee et al., 1988), may be
further increased with subsequent exposures (Folinsbee et al.,1994), and shows a tendency for
the increased response to diminish with repeated exposure (Dimeo et al., 1981; Kulle et al.,
AX6-72
-------
Table AX6-9. Pulmonary Function Effects with Repeated Exposures to Ozone"
X
Ozone
Concentration" Exposure
ppm Mg/ni3 Activity
0.25 490 2 h IE, (30 min
rest, 30 min
exercise),
VE = 39 L/min
0.2 392 4 h IE
(4 x 30 min
exercise), VE =
14.8 L/min/m2
BSA
0.2 392 4 h IE
(4 x 30 min
exercise), VE =
25 L/min/m2
BSA
0.4 784 3h/day for 5
days
IE (15 min rest,
15 min
exercise)
VE = 32 L/min
0.12 235 6.6 h
50 min
Number and
Exposure Gender of
Conditions Subjects
21.4 °C 5M, 3F
43.9% RH
4 days consecutive
FA exposure; 4 days
consecutive O3
exposure
1 day FA, 1 day, O3; 15 M, 8F
4 days consecutive
exposure to O3
20 °C 50% RH 9 M, 6 F
(1 day, O3; 4 days
consecutive exposure
toO3
31 °C 8M,2F
35% RH
5 consecutive days
plus follow up @ 4 or
7 days
18 °C 17 M
40% RH
Subject
Characteristics Observed Effect(s)
Healthy, NS FVC and FEVj decrements were significantly
attenuated on Day 4 of O3 exposure compared to day
1 of O3 exposure. Significant small airway function
depression accompanied by significant neutrophilia
in BALF one day following the end of O3 exposure.
Healthy, NS FEVj decrement and symptoms significantly
21 to 35 years old reduced on Day 4 of O3 exposure compared to Day
1 of O3 exposure. Airway inflammation of mucosa
persisted on Day 4 although some inflammatory
markers in BALF attenuated significantly.
Healthy, NS Significant decrease in FVC, FEVj, SRaw, and
23 to 37 years old symptoms on Day 4 of O3 exposure compared to a
single day of O3 exposure. Number of PMNs,
fibronectin, and IL6 in BALF were significantly
decreased on Day 4 compared to a single day of O3
exposure.
Mild asthma FEV; decreased 35% on day 1 and only 6% on day
adult 5. Bronchial reactivity increased after day 1 and
remained elevated. Adaptation of asthmatics is
similar to healthy subjects but may be slower and
less complete.
Healthy NS FEV; responses were maximal on first day of
exposure (-13%), less on second day (-9%), absent
Reference
Frank et al.
(2001)
Jorres et al.
(2000)
Christian et al.
(1998)
Gong et al.
(1997b)
Folinsbee et al.
(1994)
exercise/10 min five consecutive
rest, 30 min daily exposures
lunch
VE=38.8
L/min
thereafter. Symptoms only the first 2 days.
Methacholine airway responsiveness was at least
doubled on all exposure days, but was highest on the
second day of O3. Airway responsiveness was still
higher than air control after 5 days of O3 exposure.
Trend to lessened response, but it was not achieved
after 5 days.
-------
Table AX6-9 (cont'd). Pulmonary Function Effects with Repeated Exposures to Ozonea
X
Ozone
Concentrationb
ppm ug/m3 and Activity
0.4 784 2 h IE (15 min rest,
1 5 min exercise
VE « 60 L/min
0.45 882 2 h
IE
(3 x 20 min
exercise)
VE = 27 L/min
0.20/0.20 392/392 1 h
0.35/0.20 686/392 CE at 60 L/min
0.35/0.35 686/686
0.35 686 60minCE
VE = 60 L/min
Number and
Gender of
Exposure Conditions Subjects
5 days consecutive O3 16 M
exposure
23.3 °C 10M,6F
63% RH
Exposed for
3 consecutive days,
not exposed for 2 days,
then exposed to 0.45 ppm
again for 1 day
21 to 25 °C 15 M
40 to 60% RH
(three 2-day sets
of exposures)
21 to 25 °C 40 M
40 to 60% RH (4 groups
Subject
Characteristics
Healthy NS
Healthy NS
60 to 89 years old
median 65 years
old; mean FVC =
3.99 L; mean FEV,
= 3.01L;
FEVj/FVC range =
61 to 85%
Healthy aerobically
trained NS,
FVC = 4.24 to
6.98L
NS; nonallergic,
non-Los Angeles
Observed Effect(s)
O3-exposure FEV[ decrement was greater on
day 2, 29.9%, than day 1, 18.0%, then
decreased on day 3, 21.1%, day 4, 7% and
day 5, 4.4%
Overall increase in symptoms, but no single
symptom increased significantly. FVC
decreased 111 mL and 104 mL on Days 1
and 2, respectively. FEV[ fell by 171 and
164 mL, and FEV3 fell by 185 and 172 mL.
No significant changes on Days 3 and 4 or
with FA. FEV[ changes were -5.8, -5.6,
-1.9, and - 1 .7% on the four O3 days.
Consecutive days of exposure to 0.20 ppm
produced similar FEV; responses on each day
(-5.02, -7.80); 0.35/0.20 ppm pair caused
increased response to 0.20 ppm on second
day (-8.74); 0.35/0.35 ppm caused much
increased response on Day 2 (- 15.9, -24.6).
Symptoms were worse on the second
exposure to 0.35 ppm, but not with second
exposure to 0.20 ppm.
No differences between responses to
exposures separated by 72 or 120 h.
Reference
Folinsbee et al.
(1998)
Devlin et al.
(1997)
Bedi et al.
(1989)
Brookes et al.
(1989)
Schonfeld et al.
(1989)
(two exposures for each
subject separated by 24,
48, 72, or 120 h)
of 10) residents for >6 mo;
=25 years old
Enhanced FEV[ response at 24 h (-16.1% vs.
- 30.4%). Possible enhanced response at 48 h
(-14.4% vs. -20.6%). Similar trends
observed for breathing pattern and SRaw.
0.45 882 2 h IE
(3 x 20 min
exercise)
VE = 26 L/min
23.3 °C 8M, 8F
62.5% RH
(three exposures with
a minimum 1-week
interval)
Healthy NS, 61
years old for M and
65 years old for F
(FVC = 4.97 L for
MandS.llLforF)
Spirometric changes were not reproducible
from time to time after O3 exposure
R < 0.50). Repeat exposures to air yielded
consistent responses.
Bedi et al.
(1988)
-------
Table AX6-9 (cont'd). Pulmonary Function Effects with Repeated Exposures to Ozonea
X
Ozone
Concentrationb
ppm MS/m3 and Activity
0.18 353 2h
IE (heavy)
VE « 60 to 70 L/min
(35 L/min/m2 BSA)
0.45 882 2 h
(+0.30 IE (20 min rest,
PAN) 20 min exercise)
VE = 27 L/min
0.35 686 «lhCE
(see paper
Exposure Conditions
31 °C
35% RH
(screen exposures
in spring 1986;
second exposures in
summer/fall 1986 and
winter 1 987 and spring
1987 for responders and
nonresponders only)
22 °C
60% RH
5 days consecutive
exposure to PAN + O3
22 to 25 °C
35 to 50% RH
Number and
Gender of Subject
Subjects Characteristics
59 adult Responders:
Los Angeles 1 9 to 40 years old
residents 6 atopic,
12 responsive 2 asthmatic,
13 4 normal
nonresponsive
Nonresponders:
18 to 39 years old,
1 3 normal
3 M, 5 F Healthy NS,
Mean age =
24 years
8 M Aerobically trained
healthy NS (some
Observed Effect(s)
Responders had AFEVj = - 12.4% after initial
screening; nonresponders had no change.
Responders had nonsignificant response in
late summer or early winter, but were
responsive again in early spring (spring 1986,
-385 mL; Autumn 1986, - 17 mL; winter
1987, +16 mL; spring 1987, -347 mL).
Nonresponders did not change with season.
Suggests that responders responses may vary
with ambient exposure, but nonresponders
generally remain nonresponsive.
FEV; decreased « 1 9% with O3 alone, « 1 5%
on Day 1 of O3 + PAN, «5% on Day 5 of O3
+ PAN, -7% 3 days after 5 days of O3 +
PAN, * 1 5% after 5 days of O3 + PAN.
Similar to other repeated O3 exposure studies,
O3 responses peaked after 2 days, were
depressed 3 days later, and responses
returned 7 days later. PAN probably had no
effect on repeated to O3 exposure responses.
Largest FEV[ decrease on second of 4 days
O3 exposure (-40% mean decrease). Trend
Reference
Linn et al.
(1988)
(also see
Hackney et al.,
1989)
Drechsler-
Parks et al.
(1987b)
(also see
Table AX6- 14)
Foxcroft and
Adams (1986)
for details)
(1 day FA; 1 day O3;
4 days consecutive
exposure to O3)
were known O3
sensitive),
22.4 ± 2.2 years old
for attenuation of pulmonary function
response not complete in 4 days. VO2ma>[
decreased with single acute O3 exposure
(-6%) but was not significant after 4 days of
O3 exposure (-4%). Performance time was
less after acute O3 (211 s) exposure than after
FA (253 s).
aSee Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
-------
1982). The initially enhanced and then lessened response may be related to changes that occur
during the repair of pulmonary epithelia damaged as a consequence of O3 exposure.
Inflammatory responses (Koren et al., 1989a), epithelial damage, and changes in permeability
(Kehrl et al., 1987) might explain a portion of these responses. By blocking pulmonary function
responses and symptoms with indomethacin pretreatment, Schonfeld et al. (1989) demonstrated
that in the absence of an initial response, pulmonary function and symptoms effects were not
enhanced on Day 2 by repeated exposure to 0.35 ppm O3. These results suggest that airway
inflammation and the release of cyclooxygenase products of arachidonic acid play a role in the
enhanced pulmonary function responses and symptoms observed upon reexposure to O3 within
48 h.
Response to laboratory O3 exposure as a function of the season of the year in the South
Coast Air Basin of Los Angeles, CA, has been examined in several studies (Avol et al., 1988;
Hackney et al., 1989; Linn et al., 1988). Their primary purpose was to determine whether O3
responsive subjects would remain responsive after regular ambient exposure during the "smog
season". The subjects were exposed to 0.18 ppm O3 for 2 h with heavy IE on four occasions,
spring, fall, winter, and the following spring. The marked difference in FEVj response between
responsive and nonresponsive subjects seen initially (-12.4% versus +1%) no longer was present
after the summer smog season (fall test) or 3 to 5 months later (winter test). However, when the
subjects were exposed to O3 during the following spring, the responsive subjects again had
significantly larger changes in FEVl3 suggesting a seasonal variation in response.
Brookes et al. (1989) and Gliner et al. (1983) tested whether initial exposure to one O3
concentration could alter response to subsequent exposure to a different O3 concentration.
Gliner et al. (1983) showed that FEVj response to 0.40 ppm O3 was not influenced by previously
being exposed to 0.20 ppm O3 for 2 h on 3 consecutive days. Brookes et al. (1989) found
enhanced FEVj and symptoms upon exposure to 0.20 ppm after previous exposure to
0.35 ppm O3. These observations suggest that, although preexposure to low concentrations of O3
may not influence responses to higher concentrations, preexposure to a high concentration of O3
can significantly increase responses to a lower concentration on the following day.
Foxcroft and Adams (1986) demonstrated that decrements in exercise performance seen
after 1 h of exposure to 0.35 ppm O3 with heavy CE were significantly less after 4 consecutive
days exposure than they were after a single acute exposure. Further, exercise performance,
AX6-76
-------
VO2max, VEmax and HR^ were not significantly different after 4 days of O3 exposure compared to
those observed in a FA exposure. Despite the change in exercise performance, Foxcroft and
Adams (1986) did not observe a significant attenuation of FEVj response, although symptoms
were significantly reduced. However, these investigators selected known O3-sensitive subjects
whose FEVj decrements exceeded 30% on the first 3 days of exposure. The large magnitude of
these responses, the trend for the responses to decrease on the third and fourth day, the decreased
symptoms, and the observations by Horvath et al. (1981) that O3-sensitive subjects adapt slowly,
suggest that attenuation of response would have occurred if the exposure series had been
continued for another 1 or 2 days. These observations support the contention advanced by
Horvath et al. (1981) that the progression of attenuation of response is a function of initial "O3
sensitivity."
Drechsler-Parks et al. (1987b) examined the response to repeated exposures to 0.45 ppm O3
plus 0.30 ppm peroxyacetyl nitrate (PAN) in 8 healthy subjects and found similar FEVj
responses to exposures to O3 (-19%) and to O3 plus PAN (-15%). Thus, PAN did not increase
responses to O3. Further, repeated exposure to the PAN plus O3 mixture resulted in similar
changes to those seen with repeated O3 exposure alone. The FEVj responses fell to less than
-5% after the fifth day, with the attenuation of response persisting 3 days after the repeated
exposures, but being absent after 7 days. These observations suggest that PAN does not
influence the attenuation of response to repeated O3 exposure. If the PAN responses are
considered negligible, this study confirms the observation that the attenuation of O3 responses
with chamber exposures lasts no longer than 1 week. [More discussion on the interaction ofO3
with other pollutants can be found in Section AX6.11.]
Folinsbee et al. (1993) exposed a group of 16 healthy males to 0.4 ppm O3 for 2 h/day on
5 consecutive days. Subjects performed heavy IE (VE = 60 to 70 L/min). Decrements in FEVj
averaged 18.0, 29.9, 21.1, 7.0, and 4.4% on the 5 exposure days. However, baseline preexposure
FEVj decreased from the first day's preexposure measurement and was depressed by an average
of about 5% by the third day. This study illustrates that, with high-concentration and heavy-
exercise exposures, pulmonary function responses may not be completely recovered within 24 h.
During this study, BALF also was obtained immediately after the Day 5 exposure, with results
reported by Devlin et al. (1997). These authors found that some inflammation and cellular
responses associated with acute O3 exposure were also attenuated after 5 consecutive days of O3
AX6-77
-------
exposure (compared to historical data for responses after a single-day exposure), although
indicators of epithelial cell damage—not seen immediately after acute exposure—were present
in BALF after the fifth day of exposure. When reexposed again 2 weeks later, changes in BALF
indicated that epithelial cells appeared fully repaired (Devlin et al., 1997).
Frank et al. (2001) exposed 8 healthy young adults to 0.25 ppm O3 for 2 h with moderate
IE (exercise VE = 40 L/min) on 4 consecutive days. In addition to standard pulmonary function
measures, isovolumetric FEF25.75, Vmax50 and Vmax75 were grouped into a single value representing
small airway function (SAWgrp). Exercise ventilatory pattern was also monitored each day,
while peripheral airway resistance was measured by bronchoscopy followed by lavage on Day 5.
The authors observed two patterns of functional response in their subjects— attenuation and
persistent. Values of FVC and FEVj showed significant attenuation by Day 4 compared to Day
1 values. However, SAWgrp and rapid shallow breathing during exercise persisted on Day 4
compared to Day 1, and were accompanied by significant neutrophilia in BALF 1 day following
the end of O3 exposure. Frank et al. (2001) suggested that both types of functional response (i.e.,
attenuation and persistence) are linked causally to inflammation. They contend that the
attenuation component is attributable at least in part to a reduction in local tissue dose during
repetitive exposure that is likely to result from the biochemical, mechanical, and morphological
changes set in motion by inflammation. They speculated that the persistent component
represents the inefficiencies incurred through inflammation. Whether the persistent small airway
dysfunction is a forerunner of more permanent change in the event that oxidant stress is extended
over lengthy periods of time is unknown.
Early repeated multihour (6 to 8 h) exposures focused on exposures to low concentrations
of O3 between 0.08 and 0.12 ppm (Folinsbee et al., 1994; Horvath et al., 1991; Linn et al., 1994).
Horvath et al. (1991) exposed subjects for 2 consecutive days to 0.08 ppm using the 6.6-h
prolonged exposure protocol (see Table AX6-2). They observed small pre- to postexposure
changes in FEVj (-2.5%) on the first day, but no change on the second day. Linn et al. (1994)
observed a 1.7% decrease in FEVj in healthy subjects after 6.6 h exposure to 0.12 ppm O3.
A second consecutive day exposure to O3 yielded even smaller (<1%) responses. In a group of
asthmatics exposed under similar conditions (Linn et al., 1994), the FEVj response on the first
day was -8.6% which was reduced to -6.7% on day 2, both significantly greater than those
observed for the nonasthmatics group. The observations of Horvath et al. (1991) and Linn et al.
AX6-78
-------
(1994) elicited a somewhat different pattern of response (no enhancement of response after the
first exposure) than that seen at higher concentrations in 2 h exposures with heavy exercise
(Tables AX6-8 and AX6-9). However, the subjects studied by Horvath et al. (1991) were
exposed only to 0.08 ppm O3 and were somewhat older (30 to 43 yrs) than the subjects studied
by Folinsbee et al. (1994), mean age of 25 yrs, while the nonasthmatic subjects studied by Linn
et al. (1994) were also older (mean = 32 yrs), had lower exercise VE (-20%) and were residents
of Los Angeles who often encountered ambient levels of O3 at or above 0.12 ppm.
Folinsbee et al. (1994) exposed 17 subjects to 0.12 ppm O3 for 6.6 h, with 50 min of
moderately heavy exercise ( VE = 39 L/min) each hour, on 5 consecutive days. Compared with
FA, the percentage changes in FEVj over the five days were -12.8%, -8.7%, -2.5%, -0.06%,
and +0.18%. A parallel attenuation of symptoms was observed, but the effect of O3 in enhancing
airway responsiveness (measured by increase in SRaw upon methacholine challenge) over
5 days was not attenuated (3.67, 4.55, 3.99, 3.24, and 3.74, compared to 2.22 in FA control).
Nasal lavage revealed no increases in neutrophils except on the first O3 exposure day.
Christian et al. (1998) exposed 15 adults (6 females and 9 males; mean age = 29.1 yrs) to
4 consecutive days at 0.20 ppm O3 for 4 h, with 30 mm of IE (exercise VE = 25 L/min/m2) each
hour. Measures of FEVl3 FVC, and symptoms were all significantly reduced on Day 1, further
decreased on Day 2, and then attenuated to near FA control values on Day 4. The pattern of
SRaw response was similar, being greatest on Day 2 and no different from FA control on Day 4.
BAL was done on Day 5 and showed that neutrophil recruitment to the respiratory tract was
attenuated with repeated short-term exposures, compared to Day 1 control O3 exposure, while
airway epithelial injury appeared to continue as reflected by no attenuation of IL-6, IL-8, total
protein, and LDH. The authors concluded that such injury might lead to airway remodeling,
which has been observed in several animal studies (Brummer et al., 1977; Schwartz et al., 1976;
Tepper et al., 1989; Van Bree et al., 1989). In a similar study to that of Christian et al. (1998),
Torres et al. (2000) exposed 23 adults (8 females and 15 males; mean age = 27.9 yrs) on
4 consecutive days to 0.20 ppm O3 for 4 h, with 30 min of IE (exercise VE = 26 L/min) each
hour. The authors observed that FEVj was significantly reduced and symptoms were
significantly increased on Day 1. On Day 2, FEVj was further decreased, while symptoms
remained unchanged. By Day 4, both FEVj and symptoms were attenuated to near FA, control
values. Twenty hours after the Day 4 exposure, BAL and bronchial mucosal biopsies were
AX6-79
-------
performed. These authors found via bronchial mucosal biopsies that inflammation of the
bronchial mucosa persisted after repeated O3 exposure, despite attenuation of some inflammatory
markers in BALF and attenuation of lung function responses and symptoms. Further, Torres
et al. (2000) observed persistent although small decrease in baseline FEVj measured before
exposure, thereby suggesting that there are different time scales of the functional responses
to O3, which may reflect different mechanisms. The levels of protein remaining elevated after
repeated exposures confirms the findings of others (Christian et al., 1998; Devlin et al., 1997),
and suggests that there is ongoing cellular damage irrespective of the attenuation of cellular
inflammatory responses with the airways. [Further discussion on the inflammatory responses to
O3 can be found in Section AX6.9.]
Based on studies cited here and in the previous O3 criteria documents (U.S. Environmental
Protection Agency, 1986, 1996), several conclusions can be drawn about repeated 1- to 2-h O3
exposures. Repeated exposures to O3 can cause an enhanced (i.e., greater) lung function
response on the second day of exposure. This enhancement appears to be dependent on the
interval between the exposures (24 h is associated with the greatest increase) and is absent with
intervals >3 days. As shown in Figure AX6-8, an enhanced response also appears to depend
on O3 concentration and to some extent on the magnitude of the initial response. Small
responses to the first O3 exposure are less likely to result in an enhanced response on the second
day of O3 exposure. Repeated daily exposure also results in attenuation of pulmonary function
responses, typically after 3 to 5 days of exposure. This attenuated response persists for less than
1 week or as long as 2 weeks. In temporal conjunction with the pulmonary function changes,
symptoms induced by O3, such as cough and chest discomfort, also are attenuated with repeated
exposure. Ozone-induced changes in airway responsiveness attenuate more slowly than
pulmonary function responses and symptoms. Attenuation of the changes in airway
responsiveness appear to persist longer than changes in pulmonary function, although this has
been studied only on a limited basis. In longer-duration (6.6 h), lower-concentration studies that
do not cause an enhanced second-day response, the attenuation of response to O3 appears to
proceed more rapidly. Inflammatory markers from BALF on the day following both 2 h (Devlin
et al., 1997) and 4 h (Christian et al., 1998; Torres et al., 2000) repeated O3 exposure for 4 days
indicate that there is ongoing cellular damage irrespective of the attenuation of some cellular
inflammatory responses of the airways, lung function responses and symptoms.
AX6-80
-------
c
o
o
•s
HI
100-
95-
90-
85-
80-
75-
70-
65-
T I I I I I
0.0 0.5 1.0 1.5 2.0 2.5
Time (days)
i
3.0
I
3.5
4.0
Figure AX6-8. Regression curves were fitted to day-by-day postexposure FEVt values
obtained after repeated daily acute exposures to O3 for 2 to 3 h with
intermittent exercise at a VE of 24 to 43 L/min (adaptation studies).
Symbols represent the results from individual studies conducted at 0.2 ppm
for 2 h (+), 0.35 ppm for 2 h (•), 0.4 ppm for 2 h (+), 0.5 ppm for 2 h (#),
and 0.54 ppm for 3 h (A). Also shown for comparison are the FEVt values
obtained after exposure to 0.12 ppm O3 for 10 h (•).
Source: Modified from Hazucha (1993).
AX6.7 EFFECTS ON EXERCISE PERFORMANCE
AX6.7.1 Introduction
In an early epidemiologic study examining race performances in Los Angeles area high
school cross-country runners, Wayne et al. (1967) observed that endurance exercise performance
was depressed by inhalation of ambient oxidant air pollutants. The authors concluded that the
detrimental effects of oxidant air pollutants on race performance might have been related to the
associated discomfort in breathing, thus limiting the runners' motivation to perform at high
levels, although physiologic effects limiting O2 availability could not be ruled out.
AX6-81
-------
Subsequently, the effects of acute O3 inhalation on endurance exercise performance have been
examined in numerous controlled laboratory studies. These studies were discussed in the
previous O3 criteria document (U.S. Environmental Protection Agency, 1996) in two categories:
(1) those that examined the effects of acute O3 inhalation on maximal oxygen uptake (VO2max)
and (2) those that examined the effects of acute O3 inhalation on the ability to complete
strenuous continuous exercise protocols of up to 1 h in duration. In this section, major
observations in these studies are briefly reviewed with emphasis on reexamining the primary
mechanisms causing decrements in VO2max and endurance exercise performance consequent
to O3 inhalation. A summary of major studies of O3 inhalation effects on endurance exercise
performance, together with observed pulmonary function and symptoms of breathing discomfort
responses, is given in Table AX6-10.
AX6.7.2 Effect on Maximal Oxygen Uptake
Three early studies (Folinsbee et al., 1977; Horvath et al., 1979; Savin and Adams, 1979)
examining the effects of acute O3 exposures on VO2max were reviewed in an earlier O3 criteria
document (U.S. Environmental Protection Agency, 1986). Briefly, Folinsbee et al. (1977)
observed that VO2max was significantly decreased (10.5%) following a 2-h exposure to
0.75 ppm O3 with light (50 Watts) IE. Reduction in VO2max was accompanied by a decrease in
maximal ventilation, maximal heart rate, and a large decrease in maximal tidal volume.
In addition, the 2-h IE O3 exposure resulted in a 22.3% decrease in FEVj and significant
symptoms of cough and chest discomfort. In contrast, Horvath et al. (1979) did not observe a
change in VO2max or other maximal cardiopulmonary endpoints in subjects exposed for 2 h at rest
to either 0.50 or 0.75 ppm, although FVC was significantly decreased 10% following the latter
exposure. Without preliminary exposure to O3, Savin and Adams (1979) examined the effects of
a 30-min exposure to 0.15 and 0.30 ppm O3 while performing a progressively incremented
exercise test to volitional fatigue (mean = 31.5 min in FA). No significant effect on maximal
work time or VO2max was observed compared to that observed upon FA exposure. Further, no
significant effect on pulmonary function, maximal heart rate, and maximal tidal volume was
observed, although maximal VE was significantly reduced 7% in the 0.30 ppm O3 exposure.
Results of these early studies suggest that VO2max is reduced if the incremented maximal exercise
AX6-82
-------
Table AX6-10. Ozone Effects on Exercise Performance"
Ozone
Concentration b
ppm
0.06-0.07
0.12-0.13
0.18
0.35
0.12
0.20
0.12
> 0.18
X 0.24
ON
i
oo
W 0.21
0.20
0.35
0.25
0.50
0.75
0.15
0.30
0.75
Hg/m3
120-140
245-260
353
686
235
392
235
353
470
412
392
686
490
980
1,470
294588
1,470
Exposure Duration Exposure
and Activity Conditions
CE 23 to 24.5 °C
(VE = 30 to 120 L/min) 50 to 53% RH
16 to 28 min progressive
maximum exercise
protocol
1 h CE or competitive NA
simulation at mean
VE = 94 L/min
50 min CE 22 to 25 °C
VE = 60 L/min 35 to 50% RH
IhCE 31 °C
VE = 89 L/min
1 h competitive 23 to 26 °C
simulation exposures at 45 to 60% RH
mean
VE = 87 L/min
1 h CE at 75% V02max 19 to 21 °C
60 to 70% RH
1 h CE or competitive 23 to 26 °C
simulation at mean 45 to 60% RH
VE = 77.5 L/min
2 h rest NA
—30 min, progressively 23 °C
incremented exercise to 50% RH
voluntary exhaustion
2 h IE NA
(4x 15 min light [50 W]
bicycle ergometry)
Number and
Gender of Subject
Subjects Characteristics
12 M, 12 F Athletic
# not given; Well-trained
all males distance runners
8 M Trained
nonathletes
15M, 2F Highly trained
competitive
cyclists
10 M Highly trained
competitive
cyclists
6 M, 1 F Well-trained
cyclists
10 M Well-trained
distance runners
8M, 5F
9 M Healthy, NS
21 to 44 years old
13 M 4 light S,
9NS
Observed Effect(s)
Reduced maximum performance time and increased symptoms
of breathing discomfort during O3 exposure.
Maximal treadmill runtime reduced from 71. 7 min in FA to
66.2 min during O3 exposure with no decease in arterial O2
saturation.
VT decreased, fB increased with 50-min O3 exposures; decrease
in FVC, FEVj, FEF25.75, performance time, VO2max, V^^, and
HR,,,,,, from FA to 0.35-ppm O3 exposure.
Decrease in VEmax, VO2max, VTmax, workload, ride time, FVC,
and FEVj with 0.20 ppm O3 exposure, but not significant with
0. 12-ppm O3 exposure, as compared to FA exposure.
Decrease in exercise time of 7.7 min and 10. 1 min for subjects
unable to complete the competitive simulation at 0.18 and 0.24
ppm O3, respectively; decrease in FVC and FEVj for 0.18-and
0.24-ppm O3 exposure compared with FA exposure.
Decrease in FVC, FEVj, FEF25.75, and MVV with 0.21 ppm O3
compared with FA exposure.
VT decreased and fB increased with continuous 50-min O3
exposures; decrease in FVC, FEVj, and FEF25.75 from FA to
0.20 ppm and FA to 0.35-ppm O3 exposure in all conditions;
three subjects unable to complete continuous and competitive
protocols at 0.35 ppm O3.
FVC decreased with 0.50- and 0.75-ppm O3 exposure
compared with FA; 4% nonsignificant decrease in mean
VO2max following 0.75 ppm O3 compared with FA exposure.
Exposure to 0. 15 and 0.30 ppm O3 did not decrease maximal
exercise performance or VO2max compared to FA.
No significant pulmonary function or symptom responses were
observed, although a trend (P < .10) was evident.
Decrease in FVC, FEVj, ERV, 1C, and FEF50% after 1-h
0.75-ppm O3 exposure; decrease in VO2max, VTmax, VEmax,
maximal workload, and HRmax following 0.75-ppm O3
exposure compared with FA.
Reference
Linder et al.
(1988)
Folinsbee et al.
(1986)
Foxcroft and
Adams (1986)
Gong et al.
(1986)
Schelegle and
Adams (1986)
Folinsbee et al.
(1984)
Adams and
Schelegle (1983)
Horvath et al.
(1979)
Savin and
Adams (1979)
Folinsbee et al.
(1977)
"See Appendix A for abbreviations and acronyms.
""Listed from lowest to highest O3 concentration.
-------
test is preceded by an O3 exposure of sufficient total inhaled dose of O3 to result in significant
pulmonary function decrements and symptoms of breathing discomfort.
Using trained nonathletes, Foxcroft and Adams (1986) observed significant (p < 0.05)
reductions in rapidly incremented VO2max exercise performance time (-16.7%), VO2max (-6.0%),
maximal VE (-15.0%), and maximal heart rate (-5.6%) immediately following an initial 50-min
exposure to 0.35 ppm O3 during heavy CE ( VE = 60 L/min). These decrements were
accompanied by a significant reduction in FEVj (-23%) and the occurrence of marked
symptoms of breathing discomfort. Similarly, Gong et al. (1986) found significant reductions in
rapidly incremented VO2max exercise performance time (-29.7%), VO2max (-16.4%), maximal VE
(-18.5%), and maximal workload (-7.8%) in endurance cyclists immediately following a 1-h
exposure to 0.20 ppm O3 with very heavy exercise ( VE 89 L/min), but not following exposure to
0.12 ppm. Gong et al. (1986) observed only a 5.6% FEVj decrement and mild symptoms
following exposure to 0.12 ppm, but a large decrement in FEVj (-21.6%) and substantial
symptoms of breathing discomfort following the 0.20 ppm exposure, which the authors
contended probably limited maximal performance and VO2max.
AX6.7.3 Effect on Endurance Exercise Performance
A number of studies of well trained endurance athletes exposed to O3 have consistently
observed an impairment of 1-h continuous heavy exercise performance of some individuals
(Adams and Schelegle, 1983; Avol et al., 1984; Folinsbee et al., 1984; Gong et al., 1986). The
performance impairment is indicated by an inability to complete the prescribed O3 exposures
(even at concentrations as low as 0.16 ppm) that subjects were able to complete in FA (Avol
et al., 1984). Other indications of impaired endurance exercise performance upon exposure to O3
include a -7.7% reduced endurance treadmill running time when exposed to 0.18 ppm O3
(Folinsbee et al., 1986), which was accompanied by significantly decreased FEVj and
significantly elevated symptoms of breathing discomfort. Another study (Schelegle and Adams,
1986) observed the failure of some trained endurance athletes to complete a 1-h competitive
simulation protocol upon exposure to O3 (30 min warm-up, followed immediately by 30 min at
the maximal workload that each subject could just maintain in FA; mean VE = 120 L/min).
In this study, all subjects (n = 10) completed the FA exposure, whereas one, five, and seven
AX6-84
-------
subjects could not complete the 0.12, 0.18, and 0.24 ppm O3 exposures, respectively. Following
the 0.18 ppm and 0.24 ppm O3 exposures, but not the 0.12 ppm exposure, FEVj was reduced
significantly and symptoms were significantly increased. Linder et al. (1988) also observed
small decrements in performance time (1 to 2 min) during a progressive maximal exercise test
(mean = 21.8 min) at O3 concentrations of 0.065 and 0.125 ppm. These small effects were
accompanied by a significant increase in subjective perception of overall effort at 0.125 ppm, but
with no significant reduction in FEVj at either O3 concentration. Collectively, reduced
endurance exercise performance and associated pulmonary responses are clearly related to the
total inhaled dose of O3 (Adams and Schelegle, 1983; Avol et al., 1984; Schelegle and Adams,
1986).
Mechanisms limiting VO2max and maximal exercise performance upon O3 exposure have
not been precisely identified. Schelegle and Adams (1986) observed no significant effect of O3
on cardiorespiratory responses, and there was no indirect indication that arterial O2 saturation
was affected. The latter is consistent with the observation that measured arterial O2 saturation at
the end of a maximal endurance treadmill run was not affected by O3 (Folinsbee et al., 1986).
In studies in which O3 inhalation resulted in a significant decrease in VO2max, and/or maximal
exercise performance, significantly decreased FEVj and marked symptoms of breathing
discomfort were observed (Adams and Schelegle, 1983; Avol et al., 1984; Folinsbee et al., 1977,
1984, 1986; Foxcroft and Adams, 1986; Gong et al., 1986; Schelegle and Adams, 1986).
However, Gong et al. (1986) observed rather weak correlations between FEVj impairment and
physiological variable responses during maximal exercise (R = 0.26 to 0.44). Rather, these
authors concluded that substantial symptoms of breathing discomfort consequent to 1 h of
very heavy exercise while exposed to 0.20 ppm O3, probably limited maximal performance
and VO2max either voluntarily or involuntarily (Gong et al., 1986). Strong support for this
contention is provided by the observation of significant increases in VO2max (4.7%) and maximal
performance time (8.8%) following four consecutive days of 1 h exposure to 0.35 ppm O3 with
heavy exercise ( VE =60 L/min) compared to initial O3 exposure (Foxcroft and Adams, 1986).
These improvements, which were not significantly different from those for FA, were
accompanied by a significant reduction in symptoms of breathing discomfort with no significant
attenuation of FEVj and other pulmonary function responses. In this regard, Schelegle et al.
(1987) observed a disparate effect of indomethacin pretreatment (an inhibitor of the cyclo-
AX6-85
-------
oxygenation of arachidonic acid to prostaglandins associated with inflammatory responses)
on O3-induced pulmonary function response (significant reduction) and an overall rating of
perceived exertion and symptoms of pain on deep inspiration and shortness of breath (no
significant effect).
AX6.8 EFFECTS ON AIRWAY RESPONSIVENESS
Increased airway responsiveness, also called airway hyperresponsiveness or bronchial
hyperreactivity, indicates that the airways are more reactive to bronchoconstriction induced by a
variety of stimuli (e.g., specific allergens, exercise, SO2, cold air) than they would be when
normoreactive. In order to determine the level of airway responsiveness, airway function
(usually assessed by spirometry or plethysmography) is measured after the inhalation of small
amounts of an aerosolized specific (e.g., antigen, allergen) or nonspecific (e.g., methacholine,
histamine) bronchoconstrictor agent or measured stimulus (e.g., exercise, cold air). The dose or
concentration of the agent or stimulus is increased from a control, baseline level (placebo) until a
predetermined degree of airway response, such as a 20% drop in FEVj or a 100% increase
in Raw, has occurred (Cropp et al., 1980; Sterk et al., 1993). The dose or concentration of the
bronchoconstrictor agent that produced the increased responsiveness often is referred to as the
"PD^FEVj" or "PC2QFEVJ" (i.e., the provocative dose or concentration that produced a 20%
drop in FEVj) or the "PD100SRaw" (i.e., the provocative dose that produced a 100% increase in
SRaw). A high level of bronchial responsiveness is a hallmark of asthma. The range of
nonspecific bronchial responsiveness, as expressed by the PD20 for example, is at least 1,000-
fold from the most sensitive asthmatics to the least sensitive healthy subjects. Unfortunately, it
is difficult to compare the PD2QFEVJ or PD100SRaw across studies because of the many different
ways of presenting dose response to bronchoconstrictor drugs, for example, by mg/mL,
units/mL, and molar solution; or by cumulative dose (CIU or CBU) and doubling dose (DD).
Further confounding comparisons by affecting the site of drug delivery, dose, and ultimately
bronchial responses, the size of aerosolized agents used in challenges can vary between
nebulizers and as a function of supply air pressure in otherwise identical systems. Other typical
bronchial challenge tests with nonspecific bronchoconstrictor stimuli are based on exercise
intensity or temperature of inhaled cold air.
AX6-86
-------
Increases in nonspecific airway responsiveness were previously reported as an important
consequence of exposure to O3 (e.g., Golden et al., 1978; Table AX6-11). Konig et al. (1980)
and Holtzman et al. (1979) found the increased airway responsiveness after O3 exposure in
healthy subjects appeared to be resolved after 24 h. Because atopic subjects had similar
increases in responsiveness to histamine after O3 exposure as nonatopic subjects, Holtzman et al.
(1979) concluded that the increased nonspecific bronchial responsiveness after O3 exposure was
not related to atopy. Folinsbee and Hazucha (1989) showed increased airway responsiveness in
18 female subjects 1 and 18 h after exposure to 0.35 ppm O3. Taken together, these studies
suggest that O3-induced increases in airway responsiveness usually resolve 18 to 24 h after
exposure, but may persist in some individuals for longer periods.
Gong et al. (1986) found increased nonspecific airway responsiveness in elite cyclists
exercising at competitive levels with O3 concentrations as low as 0.12 ppm. Folinsbee et al.
(1988) found an approximate doubling of the mean methacholine responsiveness in a group of
healthy volunteers exposed for 6.6 h to 0.12 ppm O3. Horstman et al. (1990) demonstrated
significant decreases in the PD100SRaw in 22 healthy subjects immediately after a 6.6-h exposure
to concentrations of O3 as low as 0.08 ppm. No relationship was found between O3-induced
changes in airway responsiveness and changes in FVC or FEVj (Folinsbee et al., 1988; Aris
et al., 1995), suggesting that changes in airway responsiveness and spirometric volumes occur by
different mechanisms.
Dimeo et al. (1981) were the first to investigate attenuation of the O3-induced increases in
nonspecific airway responsiveness after repeated O3 exposure. Over 3 days of a 2 h/day
exposure to 0.40 ppm O3, they found progressive attenuation of the increases in airway
responsiveness such that, after the third day of O3 exposure, histamine airway responsiveness
was no longer different from the sham exposure levels. Kulle et al. (1982) found that there was
a significantly enhanced response to methacholine after the first 3 days of exposure, but this
response slowly normalized by the end of the fifth day. Folinsbee et al. (1994) found a more
persistent effect of O3 on airway responsiveness which was only partially attenuated after
5 consecutive days of O3 exposure.
The occurrence and duration of increased nonspecific airway responsiveness following O3
exposure could have important clinical implications for asthmatics. Kreit et al. (1989)
investigated changes in airway responsiveness to methacholine that occur after O3 exposure in
AX6-87
-------
TABLE AX6-11. Airway Responsiveness Following Ozone Exposures"
Ozone
Concentrationb
ppm
ug/m3
Exposure Duration Exposure Number and Subject
and Activity Conditions Gender of Subjects Characteristics
Observed Effect(s)
Reference
0.125
0.250
0.125
245 3hIE
490 (10 min rest, 15 min
exercise on bicycle)
VE = 30 L/min
245 3hIEx4days
27 °C 5 F, 6 M Mild bronchial
50 % RH 20-53 years old asthma
6F, 16 M
1 9-48 years old Allergic rhinitis
Mean early-phase FEV; response and number
of >20% reductions in FEV! were
significantly greater after 0.25 ppm O3 or
4 x 0.125 ppm O3. Most of the >15% late-
phase FEV; responses occurred after
exposure to 4 x 0.125 ppm O3, as well as
significant inflammatory effects, as indicated
by increased sputum eosinophils (asthma and
allergic rhinitis) and increased sputum
lymphocytes, mast cell tryptase, histamine,
and LDH (asthma only).
Holz et al.
(2002)
X
ON
oo
oo
0.4
0.12
0.2
0.4
0.16
784 2 h IE NA
VE = 20 L/min/m2
BSA
235 45 min IE 60% RH
exercise, rest, exercise
VE = 3 x resting
392 4 h IE 20 °C
VE = 25 L/min/m2 62% RH
BSA
784 2 h IE NA
40 min/h @ 50 W
314 7.6 h IE 22°C
50 min/h 40% RH
VE s25 1/min
6F
1M
19-26 years old
12 F
5M
19-3 8 years old
4F
8M
23-47 years old
15 healthy subjects ;
9F,6M;31.1±2.1
years old
5F
4M
Stable mild asthma;
no meds 8 h
preexposure
Physician diagnosed
asthma; SO2-induced
airway
hyperreactivity
Healthy nonsmokers
Healthy; nonatopic
Mild atopic
asthma, HMD
sensitive, 20-35 years
old
Increased bronchial responsiveness to
methacholine 16 h after exposure; inhaled
apocynin treatment significantly reduced O3-
induced airway responsiveness.
The authors concluded O3 exposure increases
bronchial responsiveness to SO2 in asthmatics
and that antioxidant supplementation has a
protective effects against this responsiveness,
especially in the "more-severe" responders.
Increased sputum total cells, % neurtophils,
IL-6, and IL-8 at 18 h after exposure;
increased airway responsiveness to
methacholine 2 h after postexposure FEV1
returned to 5% of base-line; no anti-
inflammatory effect of azithromycin.
Decreased FEV; and FVC; increased
bronchial reactivity to methacholine
4 h postexposure; no protection from
inhaled corticosteroid, budesonide.
Mean 9.1% FEV; decrease 18 h after O3
exposure; provocative dose of dust mite
allergen decreased from 10.3 to 9.7 dose
units.
Peters et al.
(2001)
Trenga et al.
(2001)
Criqui et al.
(2000)
Nightingale
et al. (2000)
Kehrl et al.
(1999)
-------
TABLE AX6-11 (cont'd). Airway Responsiveness Following Ozone Exposures3
Ozone
Concentration1"
ppm
0.2
ug/m3
392
and Activity
4hIE
40 min/h @ 50 W
Exposure Number and
Conditions Gender of Subj ects
NA 10 asthmatic
(6 F, 4 M),
Subject
Characteristics
Mild atopic asthma;
nonatopic healthy
Observed Effect(s)
Decreased FEV; in asthmatic (9.3%)
and healthy (6.7%) subjects; increased
Reference
Nightingale
etal. (1999)
0.12 235
Air-antigen
0.4
0.2
784
392
X
oo
VO
0.4
784
0.12
0.25
236
490
1 h rest
2hIE
VE= 20 L/min/m2
BSA
4hIE
50 min/h
VE = 25 L/min/m2
BSA
3 h/d for 5 days;
alternating 15 min of rest
and exercise at
VE = 32 L/min
Rest
3hIE
VE = 30 L/min
15 min ex/
10 min rest/
5 min no O3; every
30 min.
NA
NA
20 °C
50% RH
31 °C
35% RH
22 °C
40% RH
27 °C
54% RH
mouthpiece
exposure
26.6 ± 2.3 years old;
10 healthy
(4 F, 6 M),
27.3 ±1.4 years old
6F
9M
5F
1M
18-27 years old
6F
12 M
18-36 years old
2F
8M
19-48 years old
5F
10M
24 mild asthmatics
11F/13M
12 allergic rhinitics
6M/6F
10 healthy 5 M/5F
subjects; no meds
8 weeks pre-exposure
Mild allergic asthma;
18 to 49 years if age
Stable mild asthma;
no meds 8 h
preexposure
Phy sician-diagno sed
mild asthma; no meds
prior to exposure
Mild asthma
requiring only
occasional
bronchodilator
therapy
atopic
asthma
atopic mild asthmatic
NS
sputum neutrophils in both groups (NS);
no change in methacholine airway reactivity
24 h postexposure.
No effect of O3 on airway response to grass Hanania
or ragweed allergen. et al. (1998)
Increased airway responsiveness to Hiltermann
methacholine 16 h postexposure; et al. (1998)
no effect of proteinase inhibitor, rALP.
Decreased FEV; and FVC, increased SRaw; Balmes et al.
lower respiratory Sx; increased (1997);
% neutrophils, total protein, LDH, Scannell
fibronectin, IL-8, GM-CSF, and MPO et al. (1996)
in BAL. Correlation between pre-exposure
methacholine challenge and O3-induced
SRaw increase.
Significant FE V[ and Sx response on 1 st and Gong et al.
2nd O3 exposure days, then diminishing with (1997b)
continued exposure; tolerance partially lost
4 and 7 days postexposure; bronchial
reactivity to methacholine peaked on 1 st O3
exposure day, but remained elevated with
continued exposure.
No effect of O3 on airway response to grass Ball et al.
allergen. (1996)
Increased allergen responsiveness afer O3 Jorres et al.
exposure in subjects with allergic airway (1996)
disease, with or without asthma.
-------
TABLE AX6-11 (cont'd). Airway Responsiveness Following Ozone Exposures3
X
Ozone
Concentration'
ppm
0.2
0.12
0.12
0.10
0.25
0.40
Air-antigen
0.12 ppm
O3-antigen
0.08
0.10
0.12
ug/m3
392
235
235
196
490
785
157
196
235
Exposure Duration
and Activity
4hIE
50min/10min
exercise/rest each hour
IhR
6.6 h, IE x 5 days
50 min exercise/10 min
rest, 30 min lunch
VE=38.8L/min
1 h light IE
2x15 min on treadmill
VE = 27 L/min
1 h at rest
6.6 h
IE at « 39 L/min
Exposure
Conditions
22 °C
50% RH
Ambient T
&RHfor
exposure;
23°C&
50% RH for
exercise
challenge
18 °C
40% RH
21 °C
40% RH
NA
18 °C
40% RH
Number and Subject
Gender of Subj ects Characteristics
42M/24F 18-50 years
NS healthy
8 F Mild stable asthma
7M
19-45 years old
17 M Healthy nonsmokers
25 ± 4 years old
9 F Stable mild
12 M asthmatics with FEV;
1 9-40 years old >70% and
methacholine
responsiveness
4 M, 3 F Asthmatic,
21 to 64 years old
22 M Healthy NS,
18 to 32 years old
Observed Effect(s)
FEV^- 18.6%), FVC (- 14.6%), decreased
after O3. Baseline PC100 for methacholine
was not related to changes in FVC, FEVl5
a weak association was seen for PC100 and
increased SRaw.
No significant difference in % fall FEV[ or
V40p; no increase in bronchial responsiveness
to exercise challenge.
FEV[ responses were maximal on 1st day of
exposure (-13%), less on second day (-9%),
absent thereafter. Sx responses only the first
2 days. Methacholine airway responsiveness
was at least doubled on all exposure days,
but was highest on the second day of O3.
No significant differences in FEV[ or FVC
were observed for 0.10 and 0.25 ppm O3-FA
exposures or postexposure exercise challenge;
12 subjects exposed to 0.40 ppm O3 showed
significant reduction in FEV^
Increased bronchoconstrictor response to
inhaled ragweed or grass after O3 exposure
compared to air.
33, 47, and 55% decreases in cumulative dose
of methacholine required to produce a 100%
increase in SRaw after exposure to O3 at 0.08,
0.10, and 0.12 ppm, respectively.
Reference
Aris et al.
(1995)
Fernandes
etal. (1994)
Folinsbee
etal. (1994)
Weymer
etal. (1994)
Molfino
etal. (1991)
Horstman
etal. (1990)
-------
TABLE AX6-11 (cont'd). Airway Responsiveness Following Ozone Exposures'1
X
Ozone
Concentration'
ppm Mg/m3
0.12 ppm
CylOOppb
S02
0.12 ppm
O3-0.12ppm
03
Air-100ppbSO2
0.35 686
0.40 784
0.12 235
0.12 235
0.20 392
0.40 784
0.20 392
0.40 784
0.40 784
— Exposure Duration Exposure
and Activity Conditions
45 min in first atmosphere 22 °C
and 1 5 min in second IE 75% RH
70 min with IE at NA
40 L/min
2 h with IE at 22 °C
VE = 53 to 55 L/min 50% RH
6. 6 h with IE at NA
«25 L/min/m2 BSA
lhat VE= 89 L/min 31 °C
followed by 3 to 4 min 35% RH
at -150 L/min
3 h/day for 5 days in a
row
2 h with IE at 2 x resting 22 °C
2 h with IE at 2 x resting 55% RH
2 h/day for 3 days
Number and Subject
Gender of Subj ects Characteristics
8 M, 5 F Asthmatic,
12 to 18 years old
18 F Healthy NS,
19 to 28 years old
8M, 10 F 9 asthmatics (5 F,
4M),
9 healthy (5 F, 4 M),
18 to 34 years old
10 M Healthy NS,
18 to 33 years old
15M,2F Elite cyclists,
19 to 30 years old
13M, 11 F Healthy NS,
1 9 to 46 years old
12 M, 7 F Healthy NS,
21 to 32 years old
Observed Effect(s)
Greater declines in FEV[ and Vmax5m(> and
greater increase in respiratory resistance after
O3-SO2 than after O3-O3 or air-SO2.
PD100 decreased from 59 CIU after air
exposure to 41 CIU and 45 CIU, 1 and
18 h after O3 exposure, respectively.
Decreased PC100SRaw from 33 mg/mL to
8.5 mg/mL in healthy subjects after O3.
PC100SRaw fell from 0.52 mg/mL to
0.19 mg/mL in asthmatic subjects after
exposure to O3 and from 0.48 mg/mL
to 0.27 mg/mL after exposure to air.
Approximate doubling of mean
methacholine responsiveness after exposure.
On an individual basis, no relationship
between O3-induced changes in airway
responsiveness and FEV; or FVC.
Greater than 20% increase in histamine
responsiveness in one subject at 0.12 ppm
O3 and in nine subjects at 0.20 ppm O3.
Enhanced response to methacholine after
first 3 days, but this response normalized by
Day5.
110% increase in ASRaw to a 10-breath
histamine (1.6%) aerosol challenge after
exposure to O3 at 0.40 ppm, but no change
Reference
Koenig et al.
(1990)
Folinsbee
and Hazucha
(1989)
Kreit et al.
(1989)
Folinsbee
etal. (1988)
Gong et al.
(1986)
Kulle et al.
(1982)
Dimeo et al.
(1981)
at 0.20 ppm. Progressive adaptation of this
effect over 3-day exposure.
-------
TABLE AX6-11 (cont'd). Airway Responsiveness Following Ozone Exposures3
>
ON
i
VO
to
Ozone
Concentration1"
ppm Mg/m3 and Activity Conditions Gender of Subjects Characteristics
0.10 196 2h NA 14 HealthNS,
0.32 627 24 ± 2 years old
1.00 1,960
0.60 1,176 2 h with IE at 2 x resting 22 °C 1 1 M, 5 F 9 atopic,
5 5 % RH 7 nonatopic ,
NS, 21 to 35 years
old
0.6 1,176 2 hat rest NA 5 M, 3 F Healthy NS,
22 to 30 years old
Observed Effect(s)
Increased airway responsiveness
to methacholine immediately after exposure
at the two highest concentrations of O3.
Ten-breath methacholine or histamine
challenge increased SRaw >150% in
16 nonasthmatics after O3. On average, the
atopic subjects had greater responses than
the nonatopic subjects. The increased
responsiveness resolved after 24 h. Atropine
premedication blocked the O3-induced
increase in airway responsiveness.
300% increase in histamine-induced ARaw
5 min after O3 exposure; 84 and 50%
increases 24 h and 1 week after exposure
(p > 0.05), respectively. Two subjects had
an increased response to histamine 1 week
after exposure.
Reference
Konig et al.
(1980)
Holtzman
etal. (1979)
Golden et al.
(1978)
aSee Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
-------
mild asthmatics. They found that the baseline PC100SRaw declined from 0.52 to 0.19 mg/mL
after a 2-h exposure to 0.40 ppm O3 as compared to a decline from 0.48 to 0.27 mg/mL after air
exposure; however, because of the large variability in responses of the asthmatics, the percent
decrease from baseline in mean PC100SRaw was not statistically different between healthy and
asthmatic subjects (74.2 and 63.5%, respectively).
Two studies examined the effects of preexposure to O3 on exacerbation of exercise-induced
bronchoconstriction (Fernandes et al., 1994; Weymer et al., 1994). Fernandes et al. (1994)
preexposed subjects with stable mild asthma and a history of >15% decline in FEVj after
exercise to 0.12 ppm O3 for 1 h at rest followed by a 6-min exercise challenge test and found no
significant effect on either the magnitude or time course of exercise-induced
bronchoconstriction. Similarly, Weymer et al. (1994) observed that preexposure to either 0.10 or
0.25 ppm O3 for 60 min while performing light IE did not enhance or produce exercise-induced
bronchoconstriction in otherwise healthy adult subjects with stable mild asthma. Although the
results suggested that preexposure to O3 neither enhances nor produces exercise-induced asthma
in asthmatic subjects, the relatively low total inhaled doses of O3 used in these studies limit the
ability to draw any definitive conclusions.
Gong et al. (1997b) found that subjects with asthma developed tolerance to repeated
O3 exposures in a manner similar to normal subjects; however, there were more persistent effects
of O3 on airway responsiveness, which only partially attenuated when compared to filtered air
controls. Volunteer subjects with mild asthma requiring no more than bronchodilator therapy
were exposed to filtered air or 0.4 ppm O3, 3 h/d for 5 consecutive days, and follow-up
exposures 4 and 7 days later. Symptom and FEVj responses were large on the 1st and 2nd
exposure days, and diminished progressively toward filtered air responses by the 5th exposure
day. A methacholine challenge was performed when postexposure FEVj returned to within 10%
of preexposure baseline levels. The first O3 exposure significantly decreased PD2QFEVJ by an
order of magnitude and subsequent exposures resulted in smaller decreases, but they were still
significantly different from air control levels. Thus, the effects of consecutive O3 exposures on
bronchial reactivity differ somewhat from the effects on lung function. The same conclusion
was drawn by Folinsbee et al. (1994) after consecutive 5-day O3 exposures in healthy subjects,
despite a much lower bronchial reactivity both before and after O3 exposure.
AX6-93
-------
A larger number of studies examined the effects of O3 on exacerbation of antigen-induced
asthma. Molfino et al. (1991) were the first to report the effects of a 1-h resting exposure to
0.12 ppm O3 on the response of subjects with mild, stable atopic asthma to a ragweed or grass
allergen inhalation challenge. Allergen challenges were performed 24 h after air and
O3 exposure. Their findings suggested that allergen-specific airway responsiveness of mild
asthmatics is increased after O3 exposure. However, Ball et al. (1996) and Hanania et al. (1998)
were unable to confirm the findings of Molfino et al. (1991) in a group of grass-sensitive mild
allergic asthmatics exposed to 0.12 ppm O3 for 1 h. The differences between Hanania et al.
(1998) and Molfino et al. (1991), both conducted in the same laboratory, were due to better, less
variable control of the 1 h 0.12 ppm O3 exposure and better study design by Hanania and
colleagues. In the original, Molfino et al. (1991) study, the control (air) and experimental (O3)
exposures were not randomized after the second subject because of long-lasting (3 months),
O3-induced potentiation of airway reactivity in that subject. For safety reasons, therefore, the air
exposures were performed prior to the O3 exposures for the remaining 5 of 7 subjects being
evaluated. It is possible that the first antigen challenge caused the significant increase in the
second (post-O3) antigen challenge.
Torres et al. (1996) later confirmed that higher O3 concentrations cause increased airway
reactivity to specific antigens in subjects with mild allergic asthma, and to a lesser extent in
subjects with allergic rhinitis, after exposure to 0.25 ppm O3 for 3 h. The same laboratory
repeated this study in separate groups of subjects with asthma and rhinitis and found similar
enhancement of allergen responsiveness after O3 exposure (Holz et al., 2002); however, the
effects of a 3-h exposure to 0.25 ppm O3 were more variable, most likely due to performing the
allergen challenges 20 h after exposure, rather than the 3 h used in the first study.
The timing of allergen challenges in O3-exposed subjects with allergic asthma is important.
Bronchial provocation with allergen, and subsequent binding with IgE antibodies on mast cells
in the lungs, triggers the release of histamine and leukotrienes and a prompt early-phase
contraction of the smooth muscle cells of the bronchi, causing a narrowing of the lumen of the
bronchi and a decrease in bronchial airflow (i.e., decreased FEVj). In many asthma patients,
however, the release of histamine and leukotrienes from the mast cells also attracts an
accumulation of inflammatory cells, especially eosinophils, followed by the production of mucus
and a late-phase decrease in bronchial airflow for 4 to 8 h.
AX6-94
-------
A significant finding from the study by Holz et al. (2002) was that clinically relevant
decreases in FEVj (>20%) occurred during the early-phase allergen response in subjects with
rhinitis after a consecutive 4-day exposure to 0.125 ppm O3. Kehrl et al. (1999) previously
found an increased reactivity to house dust mite antigen in asthmatics 16 to 18 h after exposure
to 0.16 ppm O3 for 7.6 hours. These important observations indicate that O3 not only causes
immediate increases in airway-antigen reactivity, but that this effect may persist for at least 18 to
20 h. Ozone exposure, therefore, may be a clinically important co-factor in the response to
airborne bronchoconstrictor substances in individuals with pre-existing allergic asthma. It is
plausible that this phenomenon could contribute to increased symptom exacerbations and, even,
consequent increased physician or ER visits, and possible hospital admissions (see Chapter 7).
A number of human studies, especially more recent ones, have been undertaken to
determine various aspects of O3-induced increases in nonspecific airway responsiveness, but
most studies have been conducted in laboratory animals (See the toxicology chapter, Section
5.3.4.4.). In humans, increased airway permeability (Kehrl et al., 1987; Molfmo et al., 1992)
could play a role in increased airway responsiveness. Inflammatory cells and mediators also
could affect changes in airway responsiveness. The results of a multiphase study (Scannell
et al., 1996; Balmes et al., 1997) showed a correlation between preexposure methacholine
responsiveness in healthy subjects and increased SRaw caused by a 4 h exposure to 0.2 ppm O3,
but not with O3-induced decreases in FEVj and FVC. The O3-induced increase in SRaw, in turn,
was correlated with O3-induced increases in neutrophils and total protein concentration in BAL
fluid. Subjects with asthma had a significantly greater inflammatory response to the same O3
exposures, but it was not correlated with increased SRaw, and nonspecific airway provocation
was not measured. Therefore, it is difficult to determine from this series of studies if underlying
airway inflammation plays a role in increased airway responsiveness to nonspecific
bronchoconstrictors. The study, however, confirmed an earlier observation (e.g., Balmes et al.,
1996) that O3-induced changes in airway inflammation and lung volume measurements are not
correlated.
Hiltermann et al. (1998) reported that neutrophil-derived serine proteinases associated
with O3-induced inflammation are not important mediators for O3-induced nonspecific airway
hyperresponsiveness. Subjects with mild asthma, prescreened for O3-induced airway
responsiveness to methacholine, were administered an aerosol of recombinant antileukoprotease
AX6-95
-------
(rALP) or placebo at hourly intervals two times before and six times after exposure to filtered air
or 0.4 ppm O3 for 2 h. Methacholine challenges were performed 16 h after exposure. Treatment
with rALP had no effect on the O3-induced decrease in FEVj or PC2QFEVJ in response to
methacholine challenge. The authors speculated that proteinase-mediated tissue injury caused
by O3 may not be important in the development of airway hyperresponsiveness of asthmatics
to O3. In a subsequent study using a similar protocol (Peters et al., 2001), subjects with mild
asthma were administered an aerosol of apocynin, an inhibitor of NADPH oxidase present in
inflammatory cells such as eosinophils and neutrophils, or a placebo. In this study, methacholine
challenge performed 16 h after O3 exposure showed treatment-related effects on PC2QFEVJ,
without an effect on FEVj. The authors concluded that apocynin could prevent O3-induced
bronchial hyperresponsiveness in subjects with asthma, possibly by preventing superoxide
formation by eosinophils and neutrophils in the larger airways.
Nightingale et al. (1999) reported that exposures of healthy subjects and subjects with mild
atopic asthma to a lower O3 concentration (0.2 ppm) for 4 h caused a similar neutrophilic lung
inflammation in both groups but no changes in airway responsiveness to methacholine measured
24 h after O3 exposure in either group. There were, however, significant decreases in FEVj of
6.7 and 9.3% immediately after O3 exposure in both healthy and asthmatic subjects, respectively.
In a subsequent study, a significant increase in bronchoresponsiveness to methacholine was
reported 4 h after healthy subjects were exposed to 0.4 ppm O3 for 2 h (Nightingale et al., 2000).
In the latter study, preexposure treatment with inhaled budesonide (a corticosteroid) did not
protect against O3-induced effects on spirometry, methacholine challenge, or sputum neutrophils.
These studies also confirm the earlier reported findings that O3-induced increases in airway
responsiveness usually resolve by 24 h after exposure.
Ozone-induced airway inflammation and hyperresponsiveness were used by Criqui et al.
(2000) to evaluate anti-inflammatory properties of the macrolide antibiotic, azithromycin. In a
double-blind, cross-over study, healthy volunteers were exposed to 0.2 ppm O3 for 4 h after
pretreatment with azithromycin or a placebo. Sputum induction 18 h postexposure resulted in
significantly increased total cells, percent neutrophils, IL-6, and IL-8 in both azithromycin- and
placebo-treated subjects. Significant pre- to postexposure decreases in FEVj and FVC also were
found in both subject groups. Airway responsiveness to methacholine was not significantly
different between azithromycin-treated and placebo-treated subjects when they were challenged
AX6-96
-------
2 h after postexposure FEVj decrements returned to within 5 % of baseline. Thus, azithromycin
did not have anti-inflammatory effects in this study.
The effects of dietary antioxidants on O3-induced bronchial responsiveness to SO2
provocation were evaluated in adult asthmatic subjects by Trenga et al. (2001). This study and
potential interpretative problems are discussed in detail in Section AX6.5.6. Briefly, 17 adult
asthmatic subjects sensitive to SO2 provocation took vitamin supplements (400 IU vitamin E and
500 mg vitamin C) or placebo once a day for 5 weeks. After the fourth and fifth weeks of
vitamin or placebo, subjects were randomly exposed to FA and 0.12 ppm O3 for 45 min during
IE (VE « 3x resting rate) followed by two sequential 10 min exposures to 0.1 and 0.25 ppm SO2.
Vitamin treatment was not associated with decreased bronchial responsiveness following the
0.1 ppm SO2 challenge. However, the change in spirometric responses (FEVj, FVC, FEF25.75,
and PEF) between the 0.1 and 0.25 ppm SO2 challenges were more severe for the placebo than
the vitamin treatment regimen (p = 0.009). The authors concluded O3 exposure increases
bronchial responsiveness to SO2 in asthmatics and that antioxidant supplementation has a
protective effect against this responsiveness.
AX6.9 EFFECTS ON INFLAMMATION AND HOST DEFENSE
AX6.9.1 Introduction
In general, inflammation can be considered as the host response to injury, and the
induction of inflammation can be accepted as evidence that injury has occurred. Several
outcomes are possible: (1) inflammation can resolve entirely; (2) continued acute inflammation
can evolve into a chronic inflammatory state; (3) continued inflammation can alter the structure
or function of other pulmonary tissue, leading to diseases such as fibrosis or emphysema;
(4) inflammation can alter the body's host defense response to inhaled microorganisms; and
(5) inflammation can alter the lung's response to other agents such as allergens or toxins.
At present, it is known that short-term exposure of humans to O3 can cause acute inflammation
and that long-term exposure of laboratory animals results in a chronic inflammatory state (see
Chapter 5). However, the relationship between repetitive bouts of acute inflammation in humans
caused by O3 and the development of chronic respiratory disease is unknown.
AX6-97
-------
Bronchoalveolar lavage (BAL) using fiberoptic bronchoscopy has been utilized to sample
cells and fluids lining the respiratory tract primarily from the alveolar region, although the use of
small volume lavages or balloon catheters permits sampling of the airways. Cells and fluid can
be retrieved from the nasal passages using nasal lavage (NL) and brush or scrape biopsy.
Several studies have analyzed BAL and NL fluid and cells from O3-exposed humans for
markers of inflammation and lung damage (see Tables AX6-12 and AX6-13). The presence of
neutrophils (PMNs) in the lung has long been accepted as a hallmark of inflammation and is an
important indicator that O3 causes inflammation in the lungs. It is apparent, however, that
inflammation within airway tissues may persist beyond the point that inflammatory cells are
found in BAL fluid. Soluble mediators of inflammation such as the cytokines (IL-6, IL-8) and
arachidonic acid metabolites (e.g., PGE2, PGF2a, thromboxane, and leukotrienes [LTs] such as
LTB4) have been measured in the BAL fluid of humans exposed to O3. In addition to their role
in inflammation, many of these compounds have bronchoconstrictive properties and may be
involved in increased airway responsiveness following O3 exposure.
Some recent evidence suggests that changes in small airways function may provide a
sensitive indicator of O3 exposure and effect (see Section AX6.2.5), despite the fact that inherent
variability in their measurement by standard spirometric approaches make their assessment
difficult. Observations of increased functional responsiveness of these areas relative to the more
central airways, and of persistent effects following repeated exposure, may indicate that further
investigation of inflammatory processes in these regions is warranted.
Under normal circumstances, the epithelia lining the large and small airways develop tight
junctions and restrict the penetration of exogenous particles and macromolecules from the
airway lumen into the interstitium and blood, as well as restrict the flow of plasma components
into the airway lumen. O3 disrupts the integrity of the epithelial cell barrier in human airways, as
measured by markers of plasma influx such as albumin, immunoglobulin, and other proteins into
the airways. Markers of epithelial cell damage such as lactate dehydrogenase (LDH) also have
been measured in the BAL fluid of humans exposed to O3. Other soluble factors that have been
studied include those involved with fibrin deposition and degradation (Tissue Factor, Factor VII,
and plasminogen activator), potential markers of fibrogenesis (fibronectin, platelet derived
growth factor), and components of the complement cascade (C3a).
AX6-98
-------
Table AX6-12. Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozone"
X
Ozone Concentration*
ppm |ig/m3 Duration
Activity Level
(VE)
Number and Gender
of Subjects
Observed Effect(s)
Reference
Upper Airway Studies
0.4 784 2 h
0.2 392 2h
0.4 980 2h
0.12 235 1.5 h
0.24 470
0.5 980 4h
0.4 784 2 h
0.5 980 4hon
2 consecutive days
At rest
IE
(15 min/30 min);
(VE) « 20
L/min/m2 BSA
At rest
IE
(20 L/min) at
15 -min intervals
Resting
IE
(70 L/min) at
15 -min intervals
Resting
12 mild,
asymptomatic dust
mite-sensitive
asthmatics;
18-35 years of age
8 M, 5 F healthy NS
20-3 1 years of age
10 mild NS
asthmatics
18-3 5 years old
5 M, 5 F, asthmatic;
4 M, 4 F,
nonasthmatic;
18 to 41 years old
6 M, 6 F,
allergic rhinitics,
31.4 ± 2.0 (SD)
years old
11 M,
18 to 35 years old
41 M
(21 O3-exposed,
20 air-exposed),
18 to 35 years old
Release of early-onset mast cell-derived mediators into NL in
response to allergen not enhanced following O3 exposure.
Neutrophil and eosinophil inflammatory mediators were not
increased after O3 exposure or enhanced after allergen
challenge. O3 increased eosinophil influx following allergen
exposure.
No neutrophilia in NL samples by 1 .5 h postexposure.
Depletion of uric acid in NL fluid by 30% during h 2 of
exposure with increase in plasma uric acid levels.
No depletion of ascorbic acid, reduced glutathione,
extracellular superoxide dismutase.
Response to allergen increased (NS). PMN and
eosinophils increased after O3 plus allergen challenge.
Ozone alone increased inflammation in the nose.
NL done immediately and 24 h after exposure.
Increased number of PMNs at both times in asthmatic subjects
exposed to 0.24 ppm O3; no change in nonasthmatic subjects.
No change in lung or nasal function.
NL done immediately after exposure. Increased upper
and lower respiratory symptoms and increased levels of PMNs,
eosinophils, and albumin in NL fluid.
NL done immediately before, immediately after, and 22 h after
exposure. Increased numbers of PMNs at both times after
exposure; increased levels of tryptase, a marker of mast cell
degranulation, immediately after exposure; increased levels
of albumin 22 h after exposure.
NL done immediately before and after each exposure
and 22 h after the second exposure. Increased levels of PMNs
at all times after the first exposure, with peak values occurring
immediately prior to the second exposure.
Michelson
etal. (1999)
Mudway et al.
(1999)
Peden et al.
(1995)
McBride et al.
(1994)
Bascom et al.
(1990)
Graham and
Koren(1990)
Koren et al.
(1990)
Graham et al.
(1988)
-------
Table AX6-12 (cont'd). Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozonea
X
O
O
Ozone Concentration11
ppm ug/m3 Duration (VE)
Number and Gender
of Subjects
Observed Effect(s)
Reference
Lower Airway Studies
0.2 392 2h IE
(15 min/30 min);
(VE) - 20
L/min/m2 BSA
0.1 196 2h mild IE
0.2 392 2h IE
(15 min/30 min);
(VE) - 20
L/min/m2 body
surface area
0.27 529 2h IE
(20 min/ 60 min);
(VE) - 25
L/min/m2 BSA
0.22 431 4 h IE
(15 min/30 min);
(VE) = 25
L/min/m2 BSA
6M, 6F healthy,
nonatopic and
9 M, 6F mild
asthmatic subjects,
19-48 years of age
12 M, 10 F
healthy subjects
mean age -30
years
6M, 9F healthy
subjects and
9 M, 6F mild
asthmatics
12 subjects with
intermittent-mild
asthma exhibiting a
dual response; 18-37
years of age
12 nonsmoker,
nonresponders;
1 3 nonsmoker,
responders;
13 smokers;
1 8-40 years of age
Significantly higher baseline expression of IL-4 and IL-5 in
bronchial mucosal biopsies from asthmatic vs. healthy subjects
6 h postexposure. Following O3 exposure, epithelial
expression of IL-5, GM-CSF, ENA-78, and IL-8 increased
significantly in asthmatics, as compared to healthy subjects.
Markers of exposure in exhaled breath condensate, including
increased 8-isoprostane, TEARS and LTB-4, and a marker of
ROS-DNA interaction in peripheral blood leukocytes
(8-OHdG), were increased in a sub-set of subjects bearing the
wild genotype for NAD(P)H:quinone oxidoreductase and the
null genotype for glutathione-S-transferase Ml .
No evidence seen for increased responsiveness to the
inflammatory effects of O3 in mild asthmatics versus healthy
subjects at 6 h following exposure. Used neutrophil
recruitment and exacerbation of pre-existing inflammation.
Exposure to O3 24 h following allergen challenge resulted in a
significant decrease in FEV1, FVC and VC and increase in
symptom scores compared to air exposure. The percentage of
eosinophils, but not neutrophils, in induced sputum was higher
6 h after O3 than after air.
Recovery of AM was approximately 3-fold higher in BAL
from smokers versus nonsmokers. Unstimulated AM from
smokers released ~2-fold greater amounts of superoxide anion
than from nonsmokers at 30 min and 18 h postexposure, but
release was not further enhanced by stimulation of the cells.
ROS generation by AM from nonsmokers decreased following
exposure at 18 h; markers of epithelial permeability increased.
No relationship was found between measures of ROS
production and lung function responsiveness to O3.
Bosson et al.
(2003)
Corradi et al.
(2002)
Stenfors et al.
(2002)
Vagaggini
et al. (2002)
Voter et al.
(2001)
-------
Table AX6-12 (cont'd). Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozonea
X
Ozone Concentration11
ppm ug/m3 Duration
Activity Level
(VE)
Number and Gender
of Subjects
Observed Effect(s)
Reference
Lower Airway Studies (cont'd)
0.2 392 2h
0.4 784 2 h
0.4 784 2 h
0.4 784 1 h
0.2 392 2h
0.4 784 2 h/day for 5 days,
2 h either 10 or
20 days later
IE
(15 min/30 min);
(VE) - 20
L/min/m2 BSA
IE
(15 min/30 min);
(VE) - 20
L/min/m2 BSA
At rest
Continuous
exercise;
(VE) - 30
L/min/m2 BSA
IE
(40 L/min) at
15 -min intervals
8M, 5F healthy
nonsmokers;
20-3 1 years of age
10M, 6F subjects
with intermittent
asthma;
19-35 years of age
12 mild,
asymptomatic dust
mite-sensitive
asthmatics;
18-35 years of age
4 healthy subjects
15 healthy
nonsmokers
16 M; 18 to 35 years
of age
Early (1 .5 h postexposure) increase in adhesion molecule
expression, submucosal mast cell numbers and alterations in
lining fluid redox status. No clear relationship between early
markers of response and lung function deficits. 2.5-fold
increase in % human leukocyte antigen (HLA)-DR+ alveolar
macrophages in BAL.
In a cross-over study, levels of eosinophil cationic protein, IL-8
and percentage eosinophils were found to be highly correlated
in induced sputum and BAL 16 h following O3 exposure.
Release of early-onset mast cell-derived mediators into NL in
response to allergen not enhanced following O3 exposure.
Neutrophil and eosinophil inflammatory mediators were not
increased after O3 exposure or enhanced after allergen
challenge. O3 increased eosinophil influx following allergen
exposure.
Apoptosis was observed in cells obtained by airway lavage
6 h following exposure. AM obtained by BAL showed the
presence of a 4-hydroxynonenal (HNE) protein adduct and the
stress proteins, 72-kD heat shock protein and ferritin. These
effects were replicated by in vitro exposure of AM to HNE.
Increased numbers of CD3+, CD4+, and CD8+ T lymphocyte
subsets, in addition to neutrophils, in BAL 6 h postexposure.
BAL done immediately after fifth day of exposure and again
after exposure 10 or 20 days later. Most markers of
inflammation (PMNs, IL-6, PGE2, fibronectin) showed partial
to complete attenuation; markers of damage (LDH, IL-8,
protein, al-antitrypsin, elastase,) did not. Reversal of
attenuation was not complete for some markers, even after
20 days.
Blomberg
etal. (1999)
Hiltermann
etal. (1999)
Michelson
etal. (1999)
Hamilton et al.
(1998)
Blomberg
etal. (1997)
Devlin et al.
(1997)
-------
Table AX6-12 (cont'd). Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozone"
X
O
to
Ozone Concentration11
ppm
Hg/m3
Exposure
Duration
Activity Level
(VE)
Number and Gender
of Subjects
Observed Effect(s)
Reference
Lower Airway Studies (cont'd)
0.22
0.12
0.16
0.2
0.4
0.4
0.2
431
235
314
392
784
784
392
4h
2h
7.6 h
4h
T = 20 °C
RH = 50%
2h
T = 22 °C
RH = 50%
2h
1 5 min,
ex/ 15 min, rest
4h
T = 20 °C
RH = 50%
IE
20 min ex/1 9 min
rest
(VE) * 39-45 1/min
IE
(15 min/30 min);
(VE) - 20
L/min/m2 BSA
IE
50 min/hr
(VE) = 25L/min
IE
(50 min/60 min);
(VE)«44L/min
1 5 min rest
1 5 min exercise
cycle ergometer
(VE) « 55 1/min
(VE) = 66 1/min
IE
(50 min/60 min);
(VE)«44L/min
31M,7F
smokers and
nonsmokers
9M, 3F healthy
nonsmokers; mean
age -28 years
8 asthmatics sensitive
to dust mites
14 M, 6 F
healthy NS
1 1 healthy
nonsmokers;
18-3 5 years
8M
healthy nonsmokers
17M,6F
mild asthmatics
Post-O3 exposure FEV; in 3 groups: Smokers (-13.9%);
nonresponders (- 1 .4%) and responders (-28.5%). PMN's
increased immediately and at 18 h in all groups. Eosinophils
and lymphocytes increased after O3. IL-6 increased more in
nonsmokers. No relationship of symptoms with inflammation,
lung function changes not related to inflammation. Nasal
lavage indicators did not predict bronchial or alveolar
inflammation.
Increase in the percentage of vessels expressing P-selectin in
bronchial biopsies at 1 .5 h postexposure. No changes in FEVj,
FVC, inflammatory cells or markers in BAL, or vessels
expressing VCAM-1, E-selectin or ICAM-1 in biopsies.
Increased numbers of eosinophils in BAL after O3 exposure.
Ozone increased PMN, protein, IL-8, for all subjects.
No relationship of inflammation with spirometric responses.
Mean FEVl5 change = - 10%. BAL occurred at 0, 2, or 4 h
postexposure. Small n limits statistical inference. Trend for
PMN's to be highest at 4 h. LTC4 increased at all time points.
No change in PGE2 or thromboxane.
Comparison of BAL at 1 h postexposure vs. 18 h postexposure.
At 1 h, PMN's, total protein, LDH, al-antitrypsin, fibronectin,
PGE2, thromboxane B2, C3a, tissue factor, and clotting factor
VII were increased. IL-6 and PGE2 were higher after 1 h than
18 h. Fibronectin and tissue plasminogen activator higher after
18 h. No time differences for PMN and protein.
Increased PMN, protein, IL-8, LDH, in BAL. Inflammatory
responses were greater than a group of nonasthmatics
(Balmes et al., 1996)
Frampton
etal. (1997a)
Torres et al.
(1997)
Krishna et al.
(1997b)
Peden et al.
(1997)
Balmes et al.
(1996)
Coffey et al.
(1996)
Devlin et al.
(1996)
(compare with
Koren et al.
(1989a)
Scannell et al.
(1996)
-------
Table AX6-12 (cont'd). Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozone"
X
Ozone Concentration11
ppm ug/m3
Exposure
Duration
Activity Level
(VE)
Number and Gender
of Subjects
Observed Effect(s)
Reference
Lower Airway Studies (cont'd)
0.4 784
0.2 392
0.08 157
0.10 196
0.4 784
0.3 588
0.40 784
0.40 784
0.4 784
2 h mouthpiece
exposure
20 °C
42% RH
4h
6.6 h
2h
Ih
(mouth-piece)
2h
2h
2h
1 5 min exercise
1 5 min rest
(VE) « 40 1/min
IE
(50 min/60 min);
(VE) = 40 L/min
IE
(50 min/60 min) +
35 min lunch;
(VE) = 40 L/min
IE
(15 min/30min);
(VE) = 70 L/min
CE (60 L/min)
IE
(15 min/30 min);
(VE) = 70 L/min
IE
(15 min/30 min);
(VE) = 70 L/min
IE
(15 min/30 min);
(VE) = 70 L/min
5M, 5F
healthy; age ~ 30
15 M, 13 F,
21 to 39 years old
18 M,
18 to 35 years of age
10 M,
18 to 35 years old
5M
11M, 18 to
35 years old
11 M,
18 to 35 years old
11 M,
18 to 35 years old
Sputum induction 4 h after O3 exposure 3-fold increase in
neutrophils and a decrease in macrophages after O3 exposure.
IL-6, IL-8, and myeleperoxidase increased after O3. Possible
relationship of IL-8 and PMN levels.
Bronchial lavage, bronchial biopsies, and BAL done 1 8 h
after exposure. BAL shows changes similar to other studies.
Airway lavage shows increased cells, LDH, IL-8. Biopsies
show increased number of PMNs.
BAL fluid 18 h after exposure to 0.1 ppm O3 had significant
increases in PMNs, protein, PGE2, fibronectin, IL-6, lactate
dehydrogenase, and a-1 antitrypsin compared with the same
subjects exposed to FA. Similar but smaller increases in all
mediators after exposure to 0.08 ppm O3 except for protein and
fibronectin. Decreased phagocytosis of yeast by alveolar
macrophages was noted at both concentrations.
BAL fluid 1 h after exposure to 0.4 ppm O3 had significant
increases in PMNs, protein, PGE2, TXB2, IL-6, LDH, a-1
antitrypsin, and tissue factor compared with the same subjects
exposed to FA. Decreased phagocytosis of yeast by alveolar
macrophages.
Significantly elevated PMNs in the BAL fluid 1, 6, and 24 h
after exposure, with peak increases at 6 h.
Macrophages removed 18 h after exposure had changes in the
rate of synthesis of 123 different proteins as assayed by
computerized densitometry of two-dimensional gel protein
profiles.
BAL fluid 18 h after exposure contained increased levels of the
coagulation factors, tissue factor, and factor VII. Macrophages
in the BAL fluid had elevated tissue factor mRNA.
BAL fluid 18 h after exposure had significant increases in
PMNs, protein, albumin, IgG, PGE2, plasminogen activator,
elastase, complement C3a, and fibronectin.
Fahy et al.
(1995)
Aris et al.
(1993)
Devlin et al.
(1990,1991)
Koren et al.
(1991)
Koren et al.
(1991)
Schelegle
etal. (1991)
Devlin and
Koren (1990)
McGee et al.
(1990)
Koren et al.
(1989a,b)
-------
Table AX6-12 (cont'd). Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozonea
X
Ozone Concentration11
ppm ug/m3 Duration
Activity Level
(VE)
Number and Gender
of Subjects
Observed Effect(s)
Reference
Repeated Exposure Studies
0.125 245 3 h exposures to
0.25 490 both O3 cones, and
to FA;
3 h on four
consecutive days to
0.125; study arms
separated by
>4 wks
0.25 490 2 h on four
consecutive days;
O3 and FA exposure
study arms
separated by
>3 wks
0.2 392 single,
4 h exposures to O3
and to FA;
4 h on four
consecutive days to
O3; study arms
separated by
>4 wks
0.2 392 single, 4 h
exposure;
4 h exposures on
four consecutive
days; study arms
separated by
>4 wks
IE
(15 min/30 min)
IE
(30 min/60 min);
(VE) ~ 8 times the
FVC/min
IE
(15 min/30 min);
(MeanVE) =
14.8 L/min/m2
BSA
IE
(30 min/60 min);
(MeanVE) = 25
L/min/m2 BSA
5M, 6F allergic
asthmatic and
16M, 6F allergic
rhinitic subjects;
19-53 years of age
5M, 3F healthy
subjects;
25-31 years of age
15M, 8F healthy
subjects;
21-35 years of age
9M, 6F healthy NS
23-37 years of age
All subjects underwent 4 exposure arms and were challenged
with allergen 20 h following the last exposure in each. Sputum
was induced 6-7 h later. In rhinitics, but not asthmatics, the
incidence and magnitude of early phase FEV; decrements to
Ag were greater after 0.25 and 4x 0. 125 ppm O3. Repeated
exposure caused increases in neutrophil and eosinophil
numbers in both subject groups, as well as increased
percentage and number of lymphocytes in the asthmatics.
Maximal mean reductions in FEV; and FVC were observed on
day 2, and became negligible by day 4. FEF25.75, VmaxSO, and
Vmax75 were combined into a single value representing small
airway function (SAWgrp). This variable was the only one to
show persistent depression of the 24 h postexposure baseline
from day 2 to day 5 measurements. Numbers of PMNs in BAL
fluid on day 5 were significantly higher in subjects following
O3, compared to air, exposures.
All subjects underwent 3 exposure arms with BAL and
bronchial mucosal biopsies performed 20 h following the last
exposure in each. After repeated exposure, functional and
BAL cellular responses were not different from those after FA,
whereas total protein, IL-6, IL-8, reduced glutathione and
ortho-tyrosine remained elevated. Also at this time,
macroscopic scores of inflammation and tissue neutrophils
were increased in mucosal biopsies. IL-10 was detected only
in BAL fluid following repeated O3 exposure.
Subjects were randomly assigned to each of the exposure
regimens in a crossover design. Compared to single exposure,
repeated exposure resulted in an initial progression followed by
an attenuation of decrements in FEVj, FVC and specific
airways resistance by day 4. Bronchial and BAL washings
showed decreases in the numbers of PMNs and fibronectin
levels and IL-6 was decreased in BAL fluid on day 4.
Holz et al.
(2002)
Frank et al.
(2001)
Jorres et al.
(2000)
Christian et al.
(1998)
-------
Table AX6-12 (cont'd). Studies of Respiratory Tract Inflammatory Effects from Controlled Human Exposure to Ozonea
Ozone Concentration11
ppm |ig/m3 Duration
Repeated Exposure Studies (cont'd)
0.4 784 2 h/day for 5 days,
2 h either 10 or
20 days later
0.40 784 2 h
0.60
Activity Level
(VE)
IE
(60 L/min) at
1 5 -min intervals
IE
(83 W for women,
1 00 W for men)
at 1 5-min intervals
Number and Gender
of Subj ects Observed Effect(s)
16 M; BAL done immediately after fifth day of exposure and again
18 to 35 years of age after exposure 10 or 20 days later. Most markers of
inflammation (PMNs, IL-6, PGE2, fibronectin) showed
complete attenuation; markers of damage (LDH, IL-8, protein,
al-antitrypsin, elastase) did not. Reversal of attenuation was
not complete for some markers, even after 20 days.
7M, 3F BAL fluid 3 h after exposure had significant increases
23 to 4 1 years of age in PMNs, PGE2, TXB2, and PGF2ct at both O3 concentrations.
Reference
Devlin et al.
(1997)
Seltzer et al.
(1986)
a See Appendix A for abbreviations and acronyms.
j> b Listed from lowest to highest O3 concentration.
X
Oi
-------
Table AX6-13. Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to Controlled
Human Exposure to Ozone"
X
Oi
O
Oi
Ozone
Concentration*
ppm ng/m3 Duration
Activity Level Number and
(VE) Gender of Subjects
Observed Effect(s)
Reference
Host Defense
0.2 392 2 h
0.3 588 6 h/day for
5 consecutive
days
0.2 382 2h
IE(15min/30 4M, 5F
min); mild atopic
(VE) ~ 20 asthmatics;
L/min/m2 BSA 2 1 -42 years of age
IE (light treadmill) 24 M
(12 O3, 12 air)
IE 10M,2F
(15 min/30 min); healthy NS
( VE) -30 L/min mean -28 years
of age
A significant decline in FEV; and VC immediately following exposure.
A 2-fold increase in percent PMNs, with no changes in other
biomarkers, was observed at 6 h postexposure. By 24 h postexposure,
PMNs had decreased, but albumin, total protein, myeloperoxidase and
eosinophil cationic protein had increased.
Subjects inoculated with type 39 rhinovirus prior to exposure. NL was
performed on the morning of Days 1 to 5, 8, 15, and 30. No difference
in virus titers in NL fluid of air and O3-exposed subjects at any time
tested. No difference in PMNs or interferon gamma in NL fluid, or in
blood lymphocyte proliferative response to viral antigen.
Subjects were exposed to O3 and FA in a cross-over design and
underwent BAL 6 h postexposure. O3 exposure induced a 3-fold
increase in % PMNs and epithelial cells, and increased IL-8, Gro-a,
and total protein in BAL fluid. % PMNs correlated positively with
chemokine levels. Exposure also resulted in a significant decrease in
the CD4+/CD8+ ratio and the % of activated CD4+ and CD8+ T cells
in BAL fluid.
Newson et al.
(2000)
Henderson
etal. (1988)
Krishna et al.
(1998)
Host Defense - Mucous Clearance
0.4 784 1 h
0.20 392 2 h
0.40 784
CE (40 L/min) 1 5 healthy NS
18 to 35 years old
IE (light treadmill) 7 M,
27.2 ± 6.0 (SD)
years old
Subjects inhaled radiolabeled iron oxide particles 2 h after exposure.
No significant O3-induced effect on clearance of particles during the
next 3 h or the following morning.
Subjects inhaled radiolabeled iron oxide particles immediately before
exposure. Concentration-dependent increase in rate of particle
clearance 2 h after exposure, although clearance was confined
primarily to the peripheral airways at the lower O3 concentration.
Gerrity et al.
(1993)
Foster et al.
(1987)
-------
Table AX6-13 (cont'd). Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to
Controlled Human Exposure to Ozone"
X
I
o
Ozone
Concentration*
ppm
77-,^*,^^-,-.-.^^
ug/m3 Duration
Activity Level
(VE)
Number and
Gender of Subjects
Observed Effects)
Reference
Host Defense - Epithelial Permeability
0.0
0.2
0.15
0.35
0.5
0.4
0 2h
392
294 130min
686
784 2.25 h
784 2h
IE
4x15 min
at VE = 20
L/min/m2 BSA
IE
10 exercise/
10 rest
(VE) « SxFVC
IE
(70 L/min) at
1 5-min intervals
IE
(70 L/min) at
1 5-min intervals
10 M, 12 F
Healthy NS;
mean 24 years of
age
8M,1F
NS
16 M,
20 to 30 years old
8M,
20 to 30 years old
Clara cell protein (CC16) levels, a biomarker of epithelial
permeability, were elevated by O3 exposure and remained high at 6 h
postexposure. By 18 h postexposure, CC16 levels had returned to
baseline. No correlation between CC16 and FEV; decrements.
Subjects inhaled 99mTc-DTPA 19 h after exposure to O3. Clearance
was increased in the lung periphery. Clearance was not related to
spirometry.
Similar design and results as earlier study (Kehrl et al., 1987).
For the combined studies the average rate of clearance was 60% faster
in O3-exposed subjects.
Subjects inhaled 99mTc-DTPA 75 min after exposure. Significantly
increased clearance of 99mTc-DTPA from the lung in O3-exposed
subjects. Subjects had expected changes in FVC and SRaw.
Blomberg et al.
(2003)
Foster and
Stetkiewicz
(1996)
Kehrl et al.
(1989)
Kehrl et al.
(1987)
Drug Effects on Inflammation
0.25
490 3 h IE
1 5-min
intervals
403
exposures:
screening,
placebo, and
two treatments
27 °C
56%RH
(values from Holz
etal. 1999)
14M, 4F
Healthy NS
ozone responders
3 1.4 ±8.4 years old
On average, the screening and placebo O3 exposures caused greater
than a 9-fold increase in sputum neutrophils relative to baseline levels.
Relative to placebo, the inhaled or oral corticosteroids significantly
reduced neutrophil levels by 62 and 64%, respectively. Post-O3,
spirometry not significantly different from baseline.
Holz et al.
(2005)
0.4
784
2h
23 healthy adults Subjects were exposed to O3 following random selection for a 2 wk
daily regimen of antioxidants, including vegetable juice high in the
carotenoid, lycopene, or placebo. Concentrations of lycopene in the
lungs of supplemented subjects increased by 12% following treatment.
Supplemented subjects showed a 20% decrease in epithelial cell DNA
damage as assessed by the Comet Assay. Effects attributable to
lycopene could not be separated from those of other antioxidants.
Arab et al.
(2002)
-------
Table AX6-13 (cont'd). Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to
Controlled Human Exposure to Ozone"
X
O
OO
Oz one
Concentration*
ppm ug/m3 Duration
Activity Level
(VE)
Number and
Gender of Subjects
Observed Effect(s)
Reference
Drug Effects on Inflammation (cont'd)
0.0 0 2hIE
0.4 784 20 min
mild-mod.
exercise,
10 min rest
0.2 392 2 h
All exposures
separated by at
least 2 wks
(mean ~ 30d)
0.4 784 2 h
0.27 529 2 h
All exposures
separated by at
least 1 wk
(mean ~ 14 d)
0.4 784 2 h
4M, 5F
IE
(15 min/30min);
(VE) - 20
L/min/m2 BSA
IE
1 5-min intervals;
VE ~ 20 L/min/m2
BSA
Continuous
exercise;
(VE) - 25
L/min/m2 BSA
IE
1 5-min intervals
VEmin
~ 30 L/min
Healthy NS
30 ± 3 years old
Healthy (6 M, 9 F)
and mild asthmatic
(9 M, 6 F) subjects
Placebo group
15 M, IF
Antioxidant group
13M,2F
Mean age 27 years
7 M, F subjects
with mild asthma;
20-50 years of age
5 M, 4 F healthy
6 M, 7 F asthmatics
Subjects previously in Nightingale et al. (2000) study. Placebo-
control: Immediately postexposure decrements in FVC (9%) and
FEV; (14%) relative to pre-exposure values. FEV; decrement only
9% at 1 hr postexposure. By 3 h postexposure, recovery in FVC to
97% and FEV[ to 98% of preexposure values. Significant increases in
8-isoprostane at 4 h postexposure. Budesonide for 2 wk prior to
exposure did not affect responses.
Comparison was made of responses in healthy subjects, who had
higher basal ascorbate (ASC) levels and lower glutathione disulfide
(GSSG) levels than those of asthmatics. 6 h after exposure, ASC levels
were decreased and GSSG levels were increased in BAL fluid of
normals, but not asthmatics. Despite these differences in basal
antioxidant levels and response to O3, decrements in FEV; and
neutrophil influx did not differ in the two subject groups.
All subjects were exposed to FA and then entered a 2 wk regimen
of placebo or 250 mg Vit C, 50IU a-tocopherol, and 12 oz veg.
cocktail/day prior to O3 exposure. O3-induced decrements in FEV[ and
FVC were 30% and 24% less, respectively, in supplemented subjects.
Percent neutrophils and IL-6 levels in BAL fluid obtained 1 h
postexposure were not different in the two treatment groups.
Subjects were randomly exposed to FA and to O3 before and after
4 wks of treatment with 400 ug budesonide, b.i.d. Budesonide did not
inhibit the decrement in FEV[ or increase in symptom scores, but
significantly reduced the increase in % neutrophils and IL-8 in sputum
induced 6 h postexposure.
Subjects were pretreated for 3 days prior to exposure with
indomethacin (75 mg/day) or placebo. Similar reductions in FEV; and
FVC were seen in both groups following placebo, whereas mid-flows
showed greater decline in asthmatics than normals. Indomethacin
attenuated decrements in FEV[ and FVC in normals, but not
asthmatics. Attenuation of decrements was seen for FEF60p in
asthmatics and for FEF50 in normals.
Montuschi et al.
(2002)
Mudway et al.
(2001)
Samet et al.
(2001)
Stech-Scott
et al. (2004)
Vagaggini et al.
(2001)
Alexis et al.
(2000)
-------
Table AX6-13 (cont'd). Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to
Controlled Human Exposure to Ozonea
X
O
VO
Ozone
Concentration*
ppm
T-> \ , • • , ^ 1
ug/m3 Duration ( VE)
Number and
Gender of Subjects
Observed Effect(s)
Reference
Drug Effects on Inflammation (cont'd)
0.4
0.0
0.4
784 2 h IE
(20 min/30 min);
workload @
50 watts
784 2hIE 21 °C
4x15 min at 40% RH
VE = 18
L/min/m2BSA
2 exposures:
25% subjects
exposed to air-
air, 75% to
6 M, 9 F healthy
NS
mean —31 years
of age
Weak responders
7M, 13F
Strong responders
21M,21F
Healthy NS
20 to 59 years old
Subjects were randomly exposed to FA and to O3 before and after
2 wks of treatment with 800 ug budesonide, b.i.d. O3 caused
significant decrements in FEV[ and FVC immediately following
exposure, and a small increase in MCh-reactivity and increases in
neutrophils and myeloperoxidase in sputum induced at 4 h
postexposure. No differences were detected between responses in the
two treatment groups.
Significant O3-induced decrements in spirometric lung function.
Young adults (<35 years) were significantly more responsive than
older individuals (>35 years). Sufentanil, a narcotic analgesic, largely
abolished symptom responses and improved FEV; in strong
responders. Naloxone, an opioid antagonist, did not affect O3 effects in
weak responders. See Section AX6. 2. 5. 1
Nightingale
et al. (2000)
Passannante
etal. (1998)
0.4 784 2h IE
(60 L/min) at
1 5-min intervals
0.4 784 2h IE
(15 min/ 30 min);
(VE)=30
L/min/m2 BSA
0.35 686 1 h Continuous
exercise;
(VE)
-------
Table AX6-13 (cont'd). Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to
Controlled Human Exposure to Ozonea
X
Ozone
Concentration*
ppm
ug/m3 Duration
Activity Level
(VE)
Number and
Gender of Subjects
Observed Effect(s)
Reference
Supportive In Vitro Studies
0.01 to
0.10
0.1
0.2
1
0.01 to
0.10
0.12
0.24
0.50
19.6 to 6h
196
196 24 h
392 3h
1,690 4 h
19.6 to 6h
196
235 3h
470
980
bronchial
epithelial cells
Nasal mucosa
Nasal epithelial
cells and airway
epithelial cell line
Macrophage-like
THP-1 cells
bronchial
epithelial cells
Nasal epithelial
cells
Nonatopic,
nonasthmatic and
atopic, mild
asthmatic bronchial
biopsy samples
Allergic and
nonallergic patients
Nonatopic,
nonasthmatic and
atopic, mild
asthmatic bronchial
biopsy samples
Exposure to 0.01-0.10 ppm O3 significantly decreased the electrical
resistance of cells from asthmatic sources, compared to nonasthmatic
sources. This range of O3 concentrations also increased the movement
of 14C-BSA across the confluent cultures of "asthmatic" cells to an
extent that was greater than that in "nonasthmatic" cells.
Increased concentrations of neurokinin A and substance P in medium
following O3 exposure. Levels of release of both neuropeptides were
higher from tissues derived from allergic compared to nonallergic
patients.
Synergistic effect of O3 exposure on rhino virus- induced release of IL-8
at 24 h through mechanisms abrogated by antioxidant pretreatment.
Additive enhancement of ICAM-1 expression.
THP-1 cells were treated with samples of human surfactant protein A
(SP-A) genetic variants (SP-A1 and SP-A2) that had been previously
exposed to O3. O3-exposed variants differed in their ability to stimulate
the production of TNF-a and IL-8 by these cells.
No difference in constitutive release of IL-8, GM-CSF, si CAM- 1 and
RANTES from cells from nonasthmatic and asthmatic sources, except
for detection of RANTES in latter cells only. Increased release of all
mediators 24 h after 0.05 to 0.10 ppm O3 in "asthmatic" cells, but only
IL-8 and si CAM- 1 in "nonasthmatic" cells.
Small dose-response activation of NF-KB coinciding with O3-induced
production of free radicals assessed by electron spin resonance.
Increased TNF-a at two higher concentrations of O3 at 16 h
postexposure.
Bayram et al.
(2002)
Schierhorn
et al. (2002)
Spannhake
et al. (2002)
Wang et al.
(2002)
Bayram et al.
(2001)
Nichols et al.
(2001)
-------
Table AX6-13 (cont'd). Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to
Controlled Human Exposure to Ozone"
X
Ozone
Concentration*
ppm
ug/m3 Duration
Activity Level
(VE)
Number and
Gender of Subjects Observed Effect(s)
Reference
Supportive In Vitro Studies (cont'd)
0.06 to 118 to 24 h
0.20
0.5
0.5
0.4
0.25
0.50
0.25
0.50
1.00
392
980 Ih
980 Ih
784 Ih
490 6h
980
490 Ih
980
1,960
Nasal mucosa
Lung fibroblast
cell line with an
airway epithelial
cell line
tracheal epithelial
cells
Lung fibroblasts;
airway epithelial
cell line
Human nasal
epithelial cells
Airway epithelial
cell line and
alveolar
macrophages
105 surgical Increased histamine release correlated with mast cell degranulation.
samples from Increased release of IL-1, IL-6, IL-8 and TNF-a following O3 exposure
atopic and at 0. 10 ppm. Release of IL-4, IL-6, IL-8 and TNF-a at
nonatopic patients this concentration was significantly greater from tissues from atopic
versus nonatopic patients.
BEAS-2B cells in the presence or absence of HFL-1 cells were
exposed and incubated for 11 or 23 h. Steady-state mRNA levels
of alpha 1 procollagens type I and II, as well as TGFJ31, were increased
in O3-exposed co-cultured fibroblasts compared to air controls. Data
support interactions between the cell types in the presence and the
absence of O3-exposure.
O3 exposure caused an increase in ROS formation and a decline in
PGE2 production. No differences in mRNA and protein levels of
prostaglandin endoperoxide G/H synthase 2 (PGHS-2) or the rate of its
synthesis were detected, suggesting a direct effect of O3-generated
oxidants on PGHS-2 activity.
Cells incubated with O3-exposed arachidonic acid (AA) were found to
contain DNA single strand breaks. Pretreatment of the exposed AA
solution with catalase eliminated the effect on DNA, indicating its
dependence on H2O2 production. The effect was potentiated by the
non-carbonyl component of ozonized AA.
Increased in ICAM-1, IL-6, IL-1, and TNF expression at 0.5 ppm.
No increase in IL-8 expression. No increases at 0.25 ppm.
Increased secretion of IL-6, IL-8, and fibronectin by epithelial cells,
even at lowest O3 concentration. No O3-induced secretion of these
compounds by macrophages.
Schierhorn
etal. (1999)
Lang et al.
(1998)
Alpert et al.
(1997)
Kozumbo et al.
(1996)
Beck et al.
(1994)
Devlin et al.
(1994)
-------
Table AX6-13 (cont'd). Studies of Effects on Host Defense, on Drug Effects and Supportive In Vitro Studies Relating to
Controlled Human Exposure to Ozone"
X
Oi
Ozone
Concentration*
ppm
Hg/m3 Duration
Activity Level
(VE)
Number and
Gender of Subjects Observed Effect(s)
Reference
Supportive In Vitro Studies (cont'd)
0.20 to
1.0
0.25
0.50
1.00
0.30
1.00
392 to 2 h or 4 h
1960
490 Ih
980
1,960
588 Ih
1,960
Airway epithelial
cell line
Airway epithelial
cell line
Alveolar
macrophages
O3 caused a dose-related loss in cellular replicative activity at exposure
levels that caused minimal cytotoxicity. DNAsingle strand breaks
were not detected. These effects were different from those of H2O2
and, thus, not likely related to production of this oxidant within the
cells.
Concentration-dependent increased secretion of PGE2, TXB2,
PGF2ct, LTB4, and LTD4. More secretion basolaterally than apically.
Concentration-dependent increases in PGE2 production, and decreases
in phagocytosis of sheep erythrocytes. No O3-induced secretion of IL-
l,TNF,orIL-6.
Gabrielson
etal. (1994).
McKinnon
etal. (1993)
Becker et al.
(1991)
aSee Appendix A for abbreviations and acronyms.
bListed from lowest to highest O3 concentration.
-------
Inflammatory cells of the lung such as alveolar macrophages (AMs), monocytes, and
PMNs also constitute an important component of the pulmonary host defense system. Upon
activation, they are capable of generating free radicals and enzymes with microbicidal
capabilities, but they also have the potential to damage nearby cells. More recently published
studies since the last literature review (U.S. Environmental Protection Agency, 1996) observed
changes in T lymphocyte subsets in the airways following exposure to O3 that suggest
components of the immune host defense also may be affected.
AX6.9.2 Inflammatory Responses in the Upper Respiratory Tract
The nasal passages constitute the primary portal for inspired air at rest and, therefore, the
first region of the respiratory tract to come in contact with airborne pollutants. Nikasinovic et al.
(2003) recently reviewed the literature of laboratory-based nasal inflammatory studies published
since 1985. Nasal lavage (NL) has provided a useful tool for assessing O3-induced inflammation
in the nasopharynx. Nasal lavage is simple and rapid to perform, is noninvasive, and allows
collection of multiple sequential samples. Graham et al. (1988) reported increased levels of
PMNs in the NL fluid of humans exposed to 0.5 ppm O3 at rest for 4 h on 2 consecutive days,
with NL performed immediately before and after each exposure, as well as 22 h after the second
exposure. Nasal lavage fluid contained elevated numbers of PMNs at all postexposure times
tested, with peak values occurring immediately prior to the second day of exposure. Bascom
et al. (1990) exposed subjects with allergic rhinitis to 0.5 ppm O3 at rest for 4 h, and found
increases in PMNs, eosinophils, and mononuclear cells following O3 exposure. Graham and
Koren (1990) compared inflammatory mediators present in both the NL and BAL fluids of
humans exposed to 0.4 ppm O3 for 2 h. Increases in NL and BAL PMNs were similar (6.6- and
eightfold, respectively), suggesting a qualitative correlation between inflammatory changes in
the lower airways (BAL) and the upper respiratory tract (NL), although the PMN increase in NL
could not quantitatively predict the PMN increase in BAL. Torres et al. (1997) compared NL
and BAL in smokers and nonsmokers exposed to 0.22 ppm O3 for 4 h. In contrast to Graham
and Koren (1990), they did not find a relationship between numbers or percentages of
inflammatory cells (PMNs) in the nose and the lung, perhaps in part due to the variability
observed in their NL recoveries. Albumin, a marker of epithelial cell permeability, was
increased 18 h later, but not immediately after exposure, as seen by Bascom et al. (1990).
AX6-113
-------
Tryptase, a constituent of mast cells, was also elevated after O3 exposure at 0.4 ppm for 2 h
(Koren et al., 1990). McBride et al. (1994) reported that asthmatic subjects were more sensitive
than nonasthmatics to upper airway inflammation at an O3 concentration (0.24 ppm for 1.5 h
with light IE) that did not affect lung or nasal function or biochemical mediators. A significant
increase in the number of PMNs in NL fluid was detected in the asthmatic subjects both
immediately and 24 h after exposure. Peden et al. (1995) also found that O3 at a concentration of
0.4 ppm had a direct nasal inflammatory effect, and reported a priming effect on the response to
nasal allergen challenge, as well. A subsequent study in dust mite-sensitive asthmatic subjects
indicated that O3 at this concentration enhanced eosinophil influx in response to allergen, but did
not promote early mediator release or enhance the nasal response to allergen (Michelson et al.,
1999). Similar to observations made in the lower airways, the presence of O3 molecular
"targets" in nasal lining fluid is likely to provide some level of local protection against exposure.
In a study of healthy subjects exposed to 0.2 ppm O3 for 2 h, Mudway and colleagues (1999)
observed a significant depletion of uric acid in NL fluid at 1.5 h following exposure.
An increasing number of studies have taken advantage of advances in cell and tissue
culture techniques to examine the role of upper and lower airway epithelial cells and mucosa in
transducing the effects of O3 exposure. Many of these studies have provided important insight
into the basis of observations made in vivo. One of the methods used enables the cells or tissue
samples to be cultured at the air-liquid interface (ALI), allowing cells to establish apical and
basal polarity, and both cells and tissue samples to undergo exposure to O3 at the apical surfaces
as would occur in vivo. Nichols and colleagues (2001) examined human nasal epithelial cells
grown at the ALI for changes in free radical production, based on electron spin resonance, and
activation of the NF-KB transcription factor following exposure to O3 at 0.12 to 0.5 ppm for 3 h.
They found a dose-related activation of NF-KB within the cells that coincided with O3-induced
free radical production and increased release of TNF-a at levels above 0.24 ppm. These data
confirm the importance of this oxidant stress-associated pathway in transducing the O3 signal
within nasal epithelial cells and suggest its role in directing the inflammatory response. In a
study of nasal mucosal biopsy plugs, Schierhorn et al. (1999) found that tissues exposed to O3 at
a concentration of 0.1 ppm induced release of IL-4, IL-6, IL-8, and TNF-a that was significantly
greater from tissues from atopic patients compared to nonatopic controls. In a subsequent study,
this same exposure regimen caused the release of significantly greater amounts of the
AX6-114
-------
neuropeptides, neurokinin A and substance P, from allergic patients, compared to nonallergic
controls, suggesting increased activation of sensory nerves by O3 in the allergic tissues
(Schierhorn et al., 2002).
AX6.9.3 Inflammatory Responses in the Lower Respiratory Tract
Seltzer et al. (1986) were the first to demonstrate that exposure of humans to O3 resulted in
inflammation in the lung. Bronchoalveolar lavage fluid (3 h postexposure) from subjects
exposed to O3 contained increased PMNs as well as increased levels of PGE2, PGF2a, and TXB2
compared to fluid from air-exposed subjects. Koren et al. (1989a,b) described inflammatory
changes 18 h after O3 exposure. In addition to an eightfold increase in PMNs, Koren et al.
reported a two-fold increase in BAL fluid protein, albumin, and immunoglobulin G (IgG) levels,
suggestive of increased epithelial cell permeability. There was a 12-fold increase in IL-6 levels,
a two-fold increase in PGE2, and a two-fold increase in the complement component, C3a.
Evidence for stimulation of fibrogenic processes in the lung was shown by significant increases
in coagulation components, Tissue Factor and Factor VII (McGee et al., 1990), urokinase
plasminogen activator and fibronectin (Koren et al., 1989a). Subsequent studies by Lang et al.
(1998), using co-cultures of cells of the BEAS-2B bronchial epithelial line and of the HFL-1
lung fibroblast line, provided additional information about O3-induced fibrogenic processes.
They demonstrated that steady-state mRNA levels of both alpha 1 and procollagens type I and
III in the fibroblasts were increased following O3 exposure and that this effect was mediated by
the O3-exposed epithelial cells. This group of studies demonstrated that exposure to O3 results in
an inflammatory reaction in the lung, as evidenced by increases in PMNs and proinflammatory
compounds. Furthermore, they demonstrated that cells and mediators capable of damaging
pulmonary tissue are increased after O3 exposure and provided early suggestion of the potential
importance of the epithelial cell-myofibroblast "axis" in modulating fibrotic and fibrinolytic
processes in the airways.
Isolated lavage of the mainstream bronchus using balloon catheters or BAL using small
volumes of saline have been used to assess O3-induced changes in the large airways. Studies
collecting lavage fluid from isolated airway segments after O3 exposure indicate increased
neutrophils in the airways (Aris et al., 1993; Balmes et al., 1996; Scannell et al., 1996). Other
evidence of airway neutrophil increase comes from studies in which the initial lavage fraction
AX6-115
-------
("bronchial fraction") showed increased levels of neutrophils (Schelegle et al., 1991; Peden
et al., 1997; Balmes et al., 1996; Torres et al., 1997). Bronchial biopsies show increased PMNs
in airway tissue (Aris et al.,1993) and, in sputum collected after O3 exposure, neutrophil numbers
are elevated (Fahy et al., 1995).
Increased BAL protein, suggesting O3-induced changes in epithelial permeability (Koren
et al., 1989a, 1991 and Devlin et al., 1991) supports earlier work in which increased epithelial
permeability, as measured by increased clearance of radiolabled diethylene triamine pentaacetic
acid (99mTc-DTPA) from the lungs of humans exposed to O3, was demonstrated (Kehrl et al.,
1987). In addition, Foster and Stetkiewicz (1996) have shown that increased permeability
persists for at least 18-20 h and the effect is greater at the lung apices than at the base. In a study
of mild atopic asthmatics exposed to 0.2 ppm O3 for 2 h, Newson et al. (2000) observed a 2-fold
increase in the percentage of PMNs present at 6 hours postexposure, with no change in markers
of increased permeability as assessed by sputum induction. By 24 h, the neutrophilia was seen
to subside while levels of albumin, total protein, myeloperoxidase, and eosinophil cationic
protein increased significantly. It was concluded that the transient PMN influx induced by acute
exposure of these asthmatic subjects was followed by plasma extravasation and the activation of
both PMNs and eosinophils within the airway tissues. Such changes in permeability associated
with acute inflammation may provide better access of inhaled antigens, particulates, and other
substances to the submucosal region.
Devlin et al. (1991) reported an inflammatory response in subjects exposed to 0.08
and 0.10 ppm O3 for 6.6 h. Increased numbers of PMNs and levels of IL-6 were found at
both O3 concentrations, suggesting that lung inflammation from O3 can occur as a consequence
of prolonged exposure to ambient levels while exercising. Interestingly, those individuals who
had the largest increases in inflammatory mediators in this study did not necessarily have the
largest decrements in pulmonary function, suggesting that separate mechanisms underlie these
two responses. The absence of a relationship between spirometric responses and inflammatory
cells and markers has been reported in several studies, including Balmes et al., 1996; Schelegle
et al., 1991; Torres et al., 1997; Hazucha et al., 1996; Blomberg et al., 1999. These observations
relate largely to disparities in the times of onset and duration following single exposures. The
relationship between inflammatory and residual functional responses following repeated or
chronic exposures may represent a somewhat different case (see Section AX6.9.4).
AX6-116
-------
As indicated above, a variety of potent proinflammatory mediators have been reported to
be released into the airway lumen following O3 exposure. Studies of human alveolar
macrophages (AM) and airway epithelial cells exposed to O3 in vitro suggest that most mediators
found in the BAL fluid of O3-exposed humans are produced by epithelial cells. Macrophages
exposed to O3 in vitro showed only small increases in PGE2 (Becker et al., 1991). In contrast,
airway epithelial cells exposed in vitro to O3 showed large concentration-dependent increases
in PGE2, TXB2, LTB4, LTC4, and LTD4 (McKinnon et al., 1993) and increases in IL-6, IL-8, and
fibronectin at O3 concentrations as low as 0.1 ppm (Devlin et al., 1994). Macrophages lavaged
from subjects exposed to 0.4 ppm (Koren et al., 1989a) showed changes in the rate of synthesis
of 123 different proteins, whereas AMs exposed to O3 in vitro showed changes in only six
proteins, suggesting that macrophage function was altered by mediators released from other
cells. Furthermore, recent evidence suggests that the release of mediators from AMs may be
modulated by the products of O3-induced oxidation of airway lining fluid components, such as
human surfactant protein A (Wang et al., 2002).
Although the release of mediators has been demonstrated to occur at exposure
concentrations and times that are minimally cytotoxic to airway cells, potentially detrimental
latent effects have been demonstrated in the absence of cytotoxicity. These include the
generation of DNA single strand breaks (Kozumbo et al., 1996) and the loss of cellular
replicative activity (Gabrielson et al., 1994) in bronchial epithelial cells exposed in vitro, and the
formation of protein and DNA adducts. A highly toxic aldehyde formed during O3-induced lipid
peroxidation is 4-hydroxynonenal (HNE). Healthy human subjects exposed to 0.4 ppm O3 for
1 h underwent BAL 6 h later. Analysis of lavaged alveolar macrophages by Western blot
indicated increased levels of a 32-kDa HNE-protein adduct, as well as 72-kDa heat shock protein
and ferritin, in O3- versus air-exposed subjects (Hamilton et al., 1998). In a recent study of
healthy subjects exposed to 0.1 ppm O3 for 2 h (Corradi et al., 2002), formation of 8-hydroxy-2'-
deoxyguanosine (8-OHdG), a biomarker of reactive oxidant species (ROS)-DNA interaction,
was measured in peripheral blood lymphocytes. At 18 h postexposure, 8-OHdG was
significantly increased in cells compared to pre-exposure levels, presumably linked to concurrent
increases in chemical markers of ROS. Of interest, the increase in 8-OHdG was only significant
in a subgroup of subjects with the wild genotype for NAD(P)H:quinone oxidoreductase and the
null genotype for glutathione-S-transferase Ml, suggesting that polymorphisms in redox
AX6-117
-------
enzymes may confer "susceptibility' to O3 in some individuals. The generation of ROS
following exposure to O3 has been shown to be associated with a wide range of responses. In a
recent study, ROS production by alveolar macrophages lavaged from subjects exposed to
0.22 ppm for 4 h was assessed by flow cytometry (Voter et al., 2001). Levels were found to be
significantly elevated 18 h postexposure and associated with several markers of increased
permeability. An in vitro study of human tracheal epithelial cells exposed to O3 indicated that
generation of ROS resulted in decrease in synthesis of the bronchodilatory prostaglandin, PGE2,
as a result of inactivation of prostaglandin endoperoxide G/H synthase 2 (Alpert et al., 1997).
These and similar studies indicate that the responses to products of O3 exposure in the airways
encompass a broad range of both stimulatory and inhibitory activities, many of which may be
modulated by susceptibility factors upstream in the exposure process, at the level of
compensating for the imposed oxidant stress.
The inflammatory responses to O3 exposure also have been studied in asthmatic subjects
(Basha et al., 1994; Scannell et al., 1996; Peden et al., 1997). In these studies, asthmatics
showed significantly more neutrophils in the BAL (18 h postexposure) than similarly exposed
healthy individuals. In one of these studies (Peden et al., 1997), which included only allergic
asthmatics who tested positive for Dematophagoides farinae antigen, there was an eosinophilic
inflammation (2-fold increase), as well as neutrophilic inflammation (3-fold increase). In a
study of subjects with intermittent asthma that utilized a 2-fold higher concentration of O3
(0.4 ppm) for 2 h, increases in eosinophil cationic protein, neutrophil elastase and IL-8 were
found to be significantly increased 16 h postexposure and comparable in induced sputum and
BAL fluid (Hiltermann et al, 1999). Scannell et al. (1996) also reported that IL-8 tended to be
higher in post-O3 exposure BAL in asthmatics compared to nonasthmatics (36 vs. 12 pg/mL,
respectively) suggesting a possible mediator for the significantly increased neutrophilic
inflammation in asthmatics relative to healthy subjects (12 vs. 4.5%, respectively). In a recent
study comparing the neutrophil response to O3 at a concentration and exposure time similar to
those of the latter three studies, Stenfors and colleagues (2002) were unable to detect a
difference in the increased neutrophil numbers between 15 mild asthmatic and 15 healthy
subjects by bronchial wash at the 6 h postexposure time point. These results suggest that, at least
with regard to neutrophil influx, differences between healthy and asthmatic individuals develop
gradually following exposure and may not become evident until later in the process.
AX6-118
-------
In another study, mild asthmatics who exhibited a late phase underwent allergen challenge
24 hrs before a 2 h exposure to 0.27 ppm O3 or filtered air in a cross-over design (Vagaggini
et al., 2002). At 6 h postexposure, eosinophil numbers in induced sputum were found to be
significantly greater after O3 than after air. Studies such as these suggest that the time course of
eosinophil and neutrophil influx following O3 exposure can occur to levels detectable within the
airway lumen by as early as 6 h. They also suggest that the previous or concurrent activation of
proinflammatory pathways within the airway epithelium may enhance the inflammatory effects
of O3. For example, in an in vitro study of epithelial cells from the upper and lower respiratory
tract, cytokine production induced by rhinovirus infection was enhanced synergistically by
concurrent exposure to O3 at 0.2 ppm for 3 h (Spannhake et al, 2002).
The use of bronchial mucosal biopsies has also provided important insight into the
modulation by O3 of existing inflammatory processes within asthmatics. In a study of healthy
and allergic asthmatic subjects exposed to 0.2 ppm O3 or filtered air for 2 h, biopsies were
performed 6 hr following exposure (Bosson et al., 2003). Monoclonal antibodies were used to
assess epithelial expression of a variety of cytokines and chemokines. At baseline (air
exposure), asthmatic subjects showed significantly higher expression of interleukins (IL)-4
and -5. Following O3 exposure, the epithelial expression of IL-5, IL-8, granulocyte-macrophage
colony-stimulating factor (GM-CSF) and epithelial cell-derived neutrophil-activating peptide
78 (ENA-78) was significantly greater in asthmatic subjects, as compared to healthy subjects.
In vitro studies of bronchial epithelial cells derived by biopsy from nonatopic, nonasthmatic
subjects and asthmatic subjects also demonstrated the preferential release of GM-CSF and also
of regulated on activation, normal T cell-expressed and -secreted (RANTES) from asthmatic
cells following O3 exposure.
The time course of the inflammatory response to O3 in humans has not been explored fully.
Nevertheless, studies in which BAL was performed 1-3 h (Devlin et al., 1990; Koren et al.,
1991; Seltzer et al., 1986) after exposure to 0.4 ppm O3 demonstrated that the inflammatory
response is quickly initiated, and other studies (Koren et al., 1989a,b; Torres et al., 1997;
Scannell et al., 1996; Balmes et al., 1996) indicated that, even 18 h after exposure, inflammatory
mediators such as IL-6 and PMNs were still elevated. However, different markers show peak
responses at different times. Ozone-induced increases in IL-8, IL-6, and PGE2 are greater
immediately after O3 exposure, whereas BAL levels of fibronectin and plasminogen activator are
AX6-119
-------
greater after 18 h. PMNs and some products (protein, Tissue Factor) are similarly elevated both
1 and 18 h after O3 exposure (Devlin et al., 1996; Torres et al., 1997). Schelegle et al. (1991)
found increased PMNs in the "proximal airway" lavage at 1, 6, and 24 h after O3 exposure, with
a peak response at 6 h. In a typical BAL sample, PMNs were elevated only at the later time
points. This is consistent with the greater increase 18 h after exposure seen by Torres et al.
(1997). In addition to the influx of PMNs and (in allergic asthmatics) eosinophils, lymphocyte
numbers in BAL were also seen to be elevated significantly at 6 h following exposure of healthy
subjects to 0.2 ppm O3 for 2 h (Blomberg et al., 1997). Analysis of these cells by flow cytometry
indicated the increased presence of CD3+, CD4+ and CD8+ T cell subsets. This same laboratory
later demonstrated that within 1.5 h following exposure of healthy subjects to the same O3
regimen, expression of human leukocyte antigen (HLA)-DR on lavaged macrophages underwent
a significant, 2.5-fold increase (Blomberg et al., 1999). The significance of these alterations in
immune system components and those in IL-4 and IL-5 expression described above in the
studies of Bosson et al. (2003) has not been fully explored and may suggest a role for O3 in the
modulation of immune inflammatory processes.
AX6.9.4 Adaptation of Inflammatory Responses
Residents of areas with high oxidant concentrations tend to have somewhat blunted
pulmonary function responses and symptoms to O3 exposure (Hackney et al., 1976, 1977b, 1989;
Avol et al., 1988; Linn et al., 1988). Animal studies suggest that while inflammation may be
diminished with repeated exposure, underlying damage to lung epithelial cells continues (Tepper
et al., 1989). Devlin et al. (1997) examined the inflammatory responces of humans repeatedly
exposed to 0.4 ppm O3 for 5 consecutive days. Several indicators of inflammation (e.g., PMN
influx, IL-6, PGE2, fibronectin, macrophage phagocytosis) were attenuated after 5 days of
exposure (i.e., values were not different from FA). Several markers (LDH, IL-8, total protein,
epithelial cells) did not show attenuation, indicating that tissue damage probably continues to
occur during repeated exposure. The recovery of the inflammatory response occurred for some
markers after 10 days, but some responses were not normalized even after 20 days. The
continued presence of markers of cellular injury indicates a persistent but not necessarily
perceived response to O3.
AX6-120
-------
Christian et al. (1998) randomly subjected heathy subjects to a single exposure and to
4 consecutive days of exposure to 0.2 ppm O3 for 4 h. As reported by others, they found an
attenuation of FEVl3 FVC and specific airway resistance when comparing the single exposure
with day 4 of the multiday exposure regimen. Similarly, both "bronchial" and "alveolar"
fractions of the BAL showed decreased numbers of PMNs and fibronectin concentration at day
4 versus the single exposure, and a decrease in IL-6 levels in the alveolar fraction. Following a
similar study design and exposure parameters, but with single day filtered air controls, Torres
et al. (2000) found a decrease in FEVj and increases in the percentages of neutrophils and
lymphocytes, in concentrations of total protein, IL-6, IL-8, reduced glutathione, ortho-tyrosine
and urate in BAL fluid, but no changes in bronchial biopsy histology following the single
exposure. Twenty hours after the day 4 exposure, both functional and BAL cellular responses
to O3 were abolished. However, levels of total protein, IL-6, IL-8, reduced glutathione and
ortho-tyrosine were still increased significantly. In addition, following the day 4 exposure,
visual scores for bronchitis, erythema and the numbers of neutrophils in the mucosal biopsies
were increased. Their results indicate that, despite reduction of some markers of inflammation
in BAL and measures of large airway function, inflammation within the airways persists
following repeated exposure to O3.
In another study, Frank and colleagues (2001) exposed healthy subjects to filtered air and
to O3 (0.25 ppm, 2 h) on 4 consecutive days each, with pulmonary function measurements being
made prior to and following each exposure. BAL was performed on day 5, 24 h following the
last exposure. On day 5, PMN numbers remained significantly higher in the O3 arm compared to
air control. Of particular note in this study was the observation that small airway function,
assessed by grouping values for isovolumetric FEF25.75, Vmax50, and Vmax75 into a single value,
showed persistent reduction from day 2 through day 5. These data suggest that methods to more
effectively monitor function in the most peripheral airway regions, which are known to be the
primary sites of O3 deposition in the lung, may provide important information regarding the
cumulative effects of O3 exposure. Holz et al. (2002) made a comparison of early and late
responses to allergen challenge following O3 in subjects with allergic rhinitis or allergic asthma.
With some variation, both early and late FEVj and cellular responses in the two subject groups
were significantly enhanced by 4 consecutive days of exposure to 0.125 ppm O3 for 3 h.
AX6-121
-------
AX6.9.5 Effect of Anti-Inflammatory and Other Mitigating Agents
Studies have shown that indomethacin, a non-steroidal anti-inflammatory agent (NSAID)
that inhibits the production of cyclooxygenase products of arachidonic acid metabolism, is
capable of blunting the well-documented decrements in pulmonary function observed in
humans exposed to O3 (Schelegle et al., 1987; Ying et al., 1990). Indomethacin did not alter
the O3-induced increase in bronchial responsiveness to methacholine (Ying et al., 1990).
Pretreatment of healthy subjects and asthmatics with indomethacin prior to exposure to 0.4 ppm
for 2 h significantly attenuated decreases in FVC and FEVj in normals, but not asthmatics
(Alexis et al., 2000). Subjects have also been given ibuprofen, another NSAID agent that blocks
cyclooxygenase metabolism, prior to O3 exposure. Ibuprofen blunted decrements in lung
function following O3 exposure (Hazucha et al., 1996). Subjects given ibuprofen also had
reduced BAL levels of the cyclooxygenase product PGE2 and thromboxane B2, as well as IL-6,
but no decreases were observed in PMNs, fibronectin, permeability, LDH activity, or
macrophage phagocytic function. These studies suggest that NSAIDs can blunt O3-induced
decrements in FEVj with selective (perhaps drug-specific) affects on mediator release and other
markers of inflammation.
At least two studies have looked at the effects of the inhaled corticosteroid, budesonide, on
the effects of O3, with differing outcome perhaps associated with the presence of preexistent
disease. Nightingale and colleagues (2000) exposed healthy nonsmokers to 0.4 ppm O3 for 2 h
following 2 wk of treatment with budesonide (800 micrograms, twice daily) or placebo in a
blinded, randomized cross-over study. This relatively high O3 exposure resulted in significant
decreases in spirometric measures and increases in methacholine reactivity and neutrophils and
myeloperoxidase in induced sputum. No significant differences were observed in any of these
endpoints following budesonide treatment versus placebo. In contrast, Vagaggini et al. (2001)
compared the effects of treatment with budesonide (400 micrograms, twice daily) for 4 wk on
the responses of mild asthmatic subjects to exposure to 0.27 ppm O3for 2 h. Prior to exposure,
at the midpoint and end of exposure, and at 6 h postexposure, FEVj was measured and a
symptom questionnaire was administered; at 6 h postexposure, sputum was induced.
Budesonide treatment did not inhibit the decrement in FEVj or alter symptom score, but
significantly blunted the increase in percent PMNs and concentration of IL-8 in the sputum. The
difference in subject health status between the two studies (healthy versus mild asthmatic) may
AX6-122
-------
suggest a basis for the differing outcomes; however, because of differences in the corticosteroid
dosage and O3 exposure levels, that basis remains unclear.
Holz et al. (2005) investigated the mitigation of O3-induced inflammatory responses in
subjects pretreated with single doses of inhaled fluticasone and oral prednisolone. Eighteen
healthy ozone-responders (>10% increase in sputum neutrophils from O3 exposure) received
corticosteroid treatment or placebo 1-h before being exposed for 3-h with IE (15 min periods
rest/exercise) to 0.25 ppm O3. Sputum was collected 3-h post-O3 exposure. The 18 ozone-
responders were selected from 35 screened subjects. Twelve subjects were disqualified from the
study (6 produced insufficient sputum and 6 had inadequate neutrophil responses to O3), the
remaining 5 subjects were [presumably] qualified but did not participate. The O3 exposure
caused small changes in FEVj (-3.6% ±6.8%) that were not significantly different from baseline
or between treatment groups (i.e., prescreening, placebo, fluticasone, and prednisolone).
On average, the prescreening and placebo O3 exposures caused greater than a 9-fold increase in
sputum neutrophils relative to baseline levels. Relative to placebo, the inhaled or oral
corticosteroids significantly reduced neutrophil levels by 62 and 64%, respectively. Total
protein levels were not altered by O3 or corticosteroid treatment. Authors concluded that the
pronounced anti-inflammatory effect of steroids in their study was due to administering the
highest single doses shown to be safe and well tolerated. Furthermore, steroids were
administered so that maximal plasma level would be reached at approximately the beginning of
the O3 exposure.
Because the O3 exerts its actions in the respiratory tract by virtue of its strong oxidant
activity, it is reasonable to assume that molecules that can act as surrogate targets in the airways,
as constituents of either extracellular fluids or the intracellular milieus, could abrogate the effects
of O3. Some studies have examined the ability of dietary "antioxidant" supplements to reduce
the risk of exposure of the lung to oxidant exposure. In a study of healthy, nonsmoking adults,
Samet and colleagues (2001) restricted dietary ascorbate and randomly treated subjects for
2 weeks with a mixture of vitamin C, a-tocopherol and vegetable cocktail high in carrot and
tomato juices or placebo. Responses to 0.4 ppm O3 for 2 h were assessed in both groups at the
end of treatment. O3 -induced decrements in FEVj and FVC were significantly reduced in the
supplemented group, whereas the inflammatory response, as assessed by percentage neutrophils
and levels of IL-6 in BAL fluid, were unaffected by antioxidant supplementation. In a study that
AX6-123
-------
focused on supplementation with a commercial vegetable cocktail high in the carotenoid,
lycopene, healthy subjects were exposed for 2 h to 0.4 ppm O3 after 2 wk of antioxidant
supplementation or placebo (Arab et al., 2002). These investigators observed that lung epithelial
cell DNA damage, as measured by the Comet Assay, decreased by 20% in supplemented
subjects. However, the relationships between the types and levels of antioxidants in airway
lining fluid and responsiveness to O3 exposure is likely to be complex. In another study where
differences in ascorbate and glutathione concentrations between healthy and mild asthmatic
subjects were exploited, no relationship between antioxidant levels and spirometric or cellular
responses could be detected (Mudway et al., 2001).
AX6.9.6 Changes in Host Defense Capability Following Ozone Exposure
Concern about the effect of O3 on human host defense capability derives from numerous
animal studies demonstrating that acute exposure to as little as 0.08 ppm O3 causes decrements
in antibacterial host defenses (see Chapter 5). A study of experimental rhinovirus infection in
susceptible human volunteers failed to show any effect of 5 consecutive days of O3 exposure on
the clinical evolution of, or host response to, a viral challenge (Henderson et al., 1988). Healthy
men were nasally inoculated with type 39 rhinovirus (103 TCID50). There was no difference
between the O3-exposed and control groups in rhinovirus liters in nasal secretions, in levels of
interferon gamma or PMNs in NL fluid, or in blood lymphocyte proliferative response to
rhinovirus antigen. However, subsequent findings that rhinovirus can attach to the intracellular
adhesion molecule (ICAM)-l receptor on respiratory tract epithelial cells (Greve et al., 1989)
and that O3 can up-regulate the ICAM-1 receptor on nasal epithelial cells (Beck et al., 1994)
suggest that more studies are needed to explore the possibility that prior O3 exposure can
enhance rhinovirus binding to, and infection of, the nasal epithelium.
In a single study, human AM host defense capacity was measured in vitro in AMs removed
from subjects exposed to 0.08 and 0.10 ppm O3 for 6.6 h while undergoing moderate exercise.
Alveolar macrophages from O3-exposed subjects had significant decrements in complement-
receptor-mediated phagocytosis of Candida albicans (Devlin et al., 1991). The impairment of
AM host defense capability could potentially result in decreased ability to phagocytose and kill
inhaled microorganisms in vivo. A concentration-dependent decrease in phagocytosis of AMs
exposed to 0.1 to 1.0 ppm O3 in vitro has also been shown Becker et al. (1991). Although the
AX6-124
-------
evidence is inconclusive at present, there is a concern that O3 may render humans and animals
more susceptible to a subsequent bacterial challenge.
Only two studies (Foster et al., 1987; Gerrity et al., 1993) have investigated the effect
of O3 exposure on mucociliary particle clearance in humans. Foster et al. (1987) had seven
healthy subjects inhale radiolabeled particles (5 jim MMAD) and then exposed these subjects to
FA or O3 (0.2 and 0.4 ppm) during light IE for 2 h. Gerrity et al. (1993) exposed 15 healthy
subjects to FA or 0.4 ppm O3 during CE (40 L/min) for 1 h; at 2 h post-O3 exposure, subjects
then inhaled radiolabeled particles (5 jim MMAD). Subjects in both studies had similar
pulmonary function responses (average FVC decrease of 11 to 12%) immediately postexposure
to 0.4 ppm O3. The Foster et al. (1987) study suggested there is a stimulatory affect of O3
on mucociliary clearance; whereas, Gerrity et al. (1993) found that in the recovery period
following O3 exposure, mucus clearance is similar to control, i.e., following a FA exposure.
The clearance findings in these studies are complementary not conflicting. Investigators in both
studies suggested that O3-induced increases in mucociliary clearance could be mediated by
cholinergic receptors. Gerrity et al. (1993) further suggested that transient clearance increases
might be coincident to pulmonary function responses; this supposition based on the return of
sRaw to baseline and the recovery of FVC to within 5% of baseline (versus an 11% decrement
immediately postexposure) prior to clearance measurements.
Insofar as the airway epithelial surface provides a barrier to entry of biological, chemical
and particulate contaminants into the submucosal region, the maintenance of barrier integrity
represents a component of host defense. Many of the studies of upper and lower respiratory
responses to O3 exposure previously cited above have reported increases in markers of airway
permeability after both acute exposures and repeated exposures. These findings suggest that O3
may increase access of airborne agents. In a study of bronchial epithelial cells obtained from
nonatopic and mild atopic asthmatic subjects (Bayram et al., 2002), cells were grown to
confluence and transferred to porous membranes. When the cultures again reached confluence,
they were exposed to 0.01-0.1 ppm O3 or air and their permeability was assessed by measuring
the paracellular flux of 14C-BSA. The increase in permeability 24 h following O3 exposure was
observed to be significantly greater in cultures of cells derived from asthmatics, compared to
healthy subjects. Thus, the late increase in airway permeability following exposure of asthmatic
AX6-125
-------
subjects to O3, of the sort described by Newson et al. (2000), may be related to an inherent
susceptibility of 'asthmatic' cells to the barrier-reducing effects of O3.
As referenced in Section 6.9.3, the O3-induced increase in the numbers of CD8+ T
lymphocytes in the airways of healthy subjects reported by Blomberg et al. (1997) poses several
interesting questions regarding possible alterations in immune surveillance processes following
exposure. In a subsequent study from the same group, Krishna et al. (1998) exposed healthy
subjects to 0.2 ppm O3 or filtered air for 2 h followed by BAL at 6 h. In addition to increased
PMNs and other typical markers of inflammation, they found a significant decrease in the
CD4+/CD8+ T lymphocyte ratio and in the proportion of activated CD4+ and CD8+ cells.
Studies relating to the effects of low-level O3 exposure on the influx and activity of
immunocompetent cells in the upper and lower respiratory tracts may shed additional light on
modulation of this important area of host defense.
AX6.10 EXTRAPULMONARY EFFECTS OF OZONE
Ozone reacts rapidly on contact with respiratory system tissue and is not absorbed or
transported to extrapulmonary sites to any significant degree as such. Laboratory animal studies
suggest that reaction products formed by the interaction of O3 with respiratory system fluids or
tissues may produce effects measured outside the respiratory tract—either in the blood, as
changes in circulating blood lymphocytes, erythrocytes, and serum, or as changes in the structure
or function of other organs, such as the parathyroid gland, the heart, the liver, and the central
nervous system. Very little is known, however, about the mechanisms by which O3 could cause
these extrapulmonary effects. (See Section 5.4 for a discussion of the systemic effects
ofO3 observed in laboratory animals.)
The results from human exposure studies discussed in the previous criteria documents
(U.S. Environmental Protection Agency, 1986, 1996) failed to demonstrate any consistent
extrapulmonary effects. Early studies on peripheral blood lymphocytes collected from human
volunteers did not find any significant genotoxic or functional changes at O3 exposures of 0.4 to
0.6 ppm for up to 4 h/day. Limited data on human subjects indicated that 0.5 ppm O3 exposure
for over 2 h caused transient changes in blood erythrocytes and sera (e.g., erythrocyte fragility
and enzyme activities), but the physiological significance of these studies remains questionable.
AX6-126
-------
The conclusions drawn from these early studies raise doubt that cellular damage or altered
function is occurring to circulating cells at O3 exposures under 0.5 ppm.
Other human exposure studies have attempted to identify specific markers of exposure
to O3 in blood. For example, Schelegle et al. (1989) showed that PGF2a was elevated after O3
exposure (0.35 ppm); however, no increase in a-1 protease inhibitor was observed by Johnson
et al. (1986). Foster et al. (1996) found a reduction in the serum levels of the free radical
scavenger a-tocopherol after O3 exposure. Vender et al. (1994) failed to find any changes in
indices of red blood cell antioxidant capacity (GSH, CAT) in healthy male subjects exposed to
0.16 ppm O3 for 7.5 h while intermittently exercising. Liu et al. (1997, 1999) used a salicylate
metabolite, 2,3-dihydroxybenzoic acid (DFffiA), to indicate increased levels of hydroxyl radical
which hydroxylates salicylate to DFffiA. Increased DFffiA levels after exposure to 0.12 and
0.40 ppm suggest that O3 increases production of hydroxyl radical. The levels of DFffiA were
correlated with changes in spirometry.
Only a few experimental human studies have examined O3 effects in other nonpulmonary
organ systems besides blood. Early studies on the central nervous system (Gliner et al., 1979,
1980) were not able to find significant effects on motor activity or behavior (vigilance and
psychomotor performance) from O3 exposures at rest up to 0.75 ppm (U.S. Environmental
Protection Agency, 1986). Drechsler-Parks et al. (1995) monitored ECG, FIR, cardiac output,
stroke volume, and systolic time intervals in healthy, older subjects (56 to 85 years of age)
exposed to 0.45 ppm O3 using a noninvasive impedance cardiographic method. No changes
were found at this high O3 concentration after 2 h of exposure while the subjects exercised
intermittently at 25 L/min.
Gong et al. (1998) monitored ECG, HR, cardiac output, blood pressure, oxygen saturation,
and chemistries, as well as calculating other hemodynamic variables (e.g., stroke volume,
vascular resistance, rate-pressure products) in both healthy (n = 6) and hypertensive (n = 10)
adult males (41-78 years old). Subjects were exposed for 3 h with IE (VE « 30 L/min) to FA
and on the subsequent day to 0.3 ppm O3. See Section AX6.3 for more details about this study.
Statistically significant O3 effects for both groups combined were increases in HR, rate-pressure
product, and the alveolar-to-arterial PO2 gradient. Gong et al. (1998) suggested that by
impairing alveolar-arterial oxygen transfer, the O3 exposure could potentially lead to adverse
cardiac events by decreasing oxygen supply to the myocardium. The subjects in the Gong et al.
AX6-127
-------
(1998) study had sufficient functional reserve so as to not experience significant ECG changes or
myocardial ischemia and/or injury. However, Gong et al. (1998) concluded that O3 exposure
could pose a cardiopulmonary risk to persons with preexisting cardiovascular disease, with or
without concomitant respiratory disease.
The mechanism for the decrease in arterial oxygen tension in the Gong et al. (1998) study
could be due to an O3 induced ventilation-perfusion mismatch. It is well recognized and
accepted that ventilation and perfusion per unit lung volume increase with progression from the
apex to the base of the lung in normal upright healthy humans (Inkley and Maclntyre, 1973;
Kaneko et al., 1966). But, Foster et al. (1993) demonstrated that even in relatively young
healthy adults (26.7 ± 7 years old), O3 exposure can cause ventilation to shift away from the well
perfused basal lung (see Section AX6.2.3 for more details). This effect of O3 on ventilation
distribution [and by association the small airways] may persist beyond 24-h postexposure (Foster
et al., 1997). Hypoxic pulmonary artery vasoconstriction acts to shift perfusion away from areas
of low ventilation and moderate ventilation-perfusion mismatches (Santak et al., 1998). This
arterial vasoconstriction is thought to be mediated by protein kinase C, (Barman, 2001; Tsai
et al., 2004). A more generalized (i.e., not localized to poor ventilated areas) increase in
pulmonary vascular resistance in response to O3 exposure would presumably act against the
ability of the hypoxic vasoconstriction in mediating ventilation-perfusion mismatches. Acute
arterial vasoconstriction has been observed clinically in humans (15 M, 10 F; 34.9 ± 10 years
old) exposed for 2-h to O3 (0.12 ppm) in tandem with fine particulate («150 |ig/m3) (Brook et al.,
2002). Delaunois et al. (1998) also found that O3 exposure increases total (arterial, capillary, and
venous segments) pulmonary vascular resistance in rabbits. Hence, vasoconstriction could
potentially be induced by mechanisms other than regional hypoxia during O3 exposure. This
notion is consistent with the O3-induced reduction in alveolar-arterial oxygen transfer observed
by Gongetal. (1998).
Effects of O3 exposure on alveolar-arterial oxygen gradients may be more pronounced in
patients with preexisting obstructive lung diseases. Relative to healthy elderly subjects, COPD
patients have increased heterogeneity in both regional ventilation and perfusion (Kronenberg
et al., 1973). King and Briscoe (1968) examined the distribution of ventilation and perfusion in
a group of eight patients with severe COPD (mean FEVj/FVC = 36%). In these patients, 68% of
the lung by volume received 45% of the cardiac output, but only 10% of the total alveolar
AX6-128
-------
ventilation. This distribution of ventilation and perfusion in the patients contributed to their low
mean SaO2 of only 82% (inspired oxygen, 20.93%). Thus, even prior to O3 exposure, COPD
patients may have reduced gas exchange and low SaO2. Based on model predictions, increasing
tidal volume increases the O3 dose to the proximal alveolar region (Overton et al., 1996).
Similarly, with 90% of the alveolar ventilation supplied to only 32% of lung's volume, the well
ventilated regions of the COPD lung will be subjected to increased peripheral O3 doses. Any
inflammatory or edematous responses due to O3 delivered to the well ventilated regions of the
COPD lung will likely further inhibit gas exchange and reduce oxygen saturation.
In addition to reducing alveolar-arterial oxygen transfer, O3 induced vasoconstriction could
also acutely induce pulmonary hypertension. Individuals with COPD and coexisting pulmonary
hypertension might be subpopulations sensitive to cardiac effects as a consequence of O3
exposure. Acute pulmonary hypertension could potentially affect cardiac function by increasing
right ventricular workloads. Oral or inhaled vasodilators are used in patients to reduce
pulmonary artery pressure to improve right ventricular function (Santak et al., 1998).
Consequently, inducing pulmonary vasoconstriction in these patients would perhaps worsen their
condition, especially if their right ventricular function was already compromised. There are
reduced spirometric and symptom responses to O3 exposure with age (see Section AX6.5.7). It is
conceivable, therefore, that COPD patients and elderly individuals (due to their decreased
symptomatic responses to ambient O3) might further increase their risk of adverse
cardiopulmonary responses by continuing their exposures beyond the point where young healthy
adults might experience discomfort and cease exposure.
AX6.11 OZONE MIXED WITH OTHER POLLUTANTS
Controlled laboratory studies simulating conditions of ambient exposures have failed for
the most part to demonstrate significant adverse effects either in healthy subjects, atopic
individuals, or in young and middle-aged asthmatics.
AX6.11.1 Ozone and Sulfur Oxides
The difference in solubilities and other chemical properties of O3 and SOX seems to limit
chemical interaction and formation of related species in the mixture of these pollutants either in
AX6-129
-------
liquid or gaseous phase. Laboratory studies reviewed in the previous O3 criteria document
(Table AX6-14) reported, except for one study (Linn et al., 1994), no significant effects on
healthy individuals exposed to mixtures of O3 and SO2 or H2SO4 aerosol. In the study of Linn
et al. (1994), which was a repeated 6.5 h exposure protocol, O3 alone and O3 + H2SO4 induced
significant spirometric decrements in healthy adults and asthmatics, but the magnitude of effects
between exposure atmospheres was not significant. Asthmatic and atopic subjects showed
somewhat enhanced or potentiated response to mixtures or sequential exposure, respectively;
however, the observed effects were almost entirely attributable to O3 (U.S. Environmental
Protection Agency, 1996). Thus, in both healthy and asthmatic subjects, the interactive effects
of O3 and other pollutants were marginal and the response was dominated by O3.
Since 1994, the only laboratory study that examined the health effects of a mixture of O3
and sulfur oxides (SO2 and H2SO4) has been that of Linn et al. (1997). In this study, the
investigators closely simulated ambient summer haze air pollution conditions in Uniontown, PA
as well as controlled the selection of study subjects with the objective to corroborate earlier
reported findings of an epidemiologic study of Neas et al. (1995). The subjects were 41 children
(22F/19M) 9 to 12 yrs old. Of these, 26 children had history of asthma or allergy. During a
14-day study period, children were exposed on the 4th and 11th day for 4 hrs (IE, 15 min @ avg.
VE 22 L/min) in random order to air and a mixture of 0.10 ppm O3, 0.10 ppm SO2 and 42 to
198 mg/m3 H2SO4 (mean cone. 101 mg/m3, 0.6 mm MMAD). The effects of controlled
exposures were assessed by spirometry. Except for exposure days, children used diaries to
record activity, respiratory symptoms, location, and PEFR. Thus, every exposure day was
bracketed by 3 days of monitoring. Spirometry, PEFR, and respiratory symptoms score showed
no meaningful changes between any condition for a total study population. The symptoms score
reported by a subset of asthmatic/allergic subjects was positively associated with the inhaled
concentration of H2SO4 (p = 0.01). However, the reported symptoms were different from the
ones reported in the Uniontown study (Neas et al., 1995). Although retrospective statistical
power calculations using these study observations for the symptoms score, PEFR, and
spirometric endpoints were sufficient to detect with >80% probability the same magnitude of
changes as observed in Uniontown, the effects were minimal and not significant. The divergent
observations of the two studies have been explained by the presence of an unidentified
environmental factor in Uniontown, differences in physico-chemical properties of acid,
AX6-130
-------
Table AX6-14. Ozone Mixed with Other Pollutants3
X
H
b
o
O
H
O
o
H
W
O
O
HH
H
W
Concentration11 Number and
Exposure Duration Exposure Gender of Subject
ppm Hg/in3 Pollutant and Activity Conditions' Subjects Characteristics
Observed Effect(s)
Reference
Sulfur-Containing Pollutants
0.0
0.1 +
0.1 +
0.2
0.3
0.12
0.30
0.05
0.12
0 Air 4h 25 °C
196"+ 03+ IE 15' ex/ 15' rest 50% RH
262" + S02 + VE = 22 L/min
101" H2S04
392 03 90 min. 21 °C
564 NO2 VE ~ 32 L/min 50% RH
H2SO4 IE 3x15 min
235 03 1. 5 h with IE for 22 °C
564 NO2 2 consecutive days; 65% RH
70 H2SO4 VE = 23.2L/min
HNO3
235 O3 6.5 h 21 °C
100 H2SO4 2 consecutive days 50% RH
50 min exercise/h
VE = 29 L/min
8 M, 7 F Healthy
1 M, 4 F Asthmatic
10M, 11 F Allergic
All NS,
9 to 12 yrs. old
24 Asthmatic NS,
(17 M, 7F) 11 to 18 years old
22 completed Asthmatic NS,
study; adolescents; NS, 12 to
15M, 7F 19 years old
8 M, 7 F Nonasthmatic NS,
22 to 41 years old
13M, 17 F Asthmatic NS,
18 to 50 years old
Spirometry, PEER and subjective
symptoms score showed no meaningful
changes between any condition for a
total study population. The symptoms
score reported by a subset of
asthmatics/allergies were positively
associated with inhaled concentration
ofH2SO4(p = 0.01).
H2SO4/O3/NO2, O3/NO2 and clean air
produced similar responses
No significant pulmonary function
changes following any exposure
compared to response to clean air.
Six additional subjects started the
study, but dropped out due to
uncomfortable symptoms.
Exposure to O3 or O3 + H2SO4 induced
significant decrements in
forced expiratory function. Differences
between O3 and O3 + H2SO4 were, at
best, marginally significant. O3 is the
more important pollutant for inducing
respiratory effects. A few asthmatic
and nonasthmatic subjects were more
responsive to O3 + H2SO4 than to
O, alone.
Linn et al.
(1997)
Linnetal.
(1995)
Koenig et al.
(1994)
Linn et al.
(1994)
0.08 157 O3 3-h exposure to aerosol,
0.12 235 O3 followed 24 h later by a
0.18 353 O3 3-h exposure to O3. IE
100 NaCl (10 min per half hour)
100 H2SO4 VE = 4 times resting
(30 to 364 min)
21 °C 16 M, 14 F Nonasthmatic NS, No significant changes in symptoms or
-40% RH 18 to 45 years old lung function with any aerosol/O3
10M, 20 F Asthmatic NS, combination in the healthy group.
21 to 42 years old In asthmatics, H2SO4 preexposure
enhanced the small decrements in FVC
that occurred following exposure to
0.18ppmO3. Asthmatics had no
significant changes on FEV; with any
O3 exposures, but symptoms were
greater.
Utell et al.
(1994)
Frampton
etal. (1995)
-------
Table AX6-14 (cont'd). Ozone Mixed with Other Pollutants3
X
to
H
6
o
o
H
O
o
H
W
O
O
HH
H
W
Concentration11
ppm
Hg/m3
Pollutant
Number and
Exposure Duration Exposure Gender of
and Activity Conditions' Subjects
Subject
Characteristics
Observed Effect(s)
Reference
Sulfur-Containing Pollutants (cont'd)
0.12
0.10
0.25
235
262
490
1,200
to
1,600
03
SO2
03
H2SO4
1 h (mouthpiece) 22 °C 8 M, 5 F
IE VE ~ 30 L/min 75% RH
45-min exposure to air or
O3, followed by 1 5-min
exposure to O3 or SO2
2h 35 °C 9M
IE 83% RH
VE = 30 to 32 L/min
Allergic asthmatics,
12 to 18 years old,
medications withheld
for at least 4 h before
exposures
Healthy NS,
19 to 29 years old
Prior exposure to O3 potentiated
pulmonary function responses to SO2;
decrements in FEVj were -3, -2, and
-8% for the air/O3, O3/O3, and O3/SO2
exposures, respectively.
No significant effects of exposure to O3
alone or combined with H2SO4 aerosol.
Koenig et al.
(1990)
Horvath
etal. (1987)
Nitrogen-Containing Pollutants
0.0
0.2
0.4
0.2+0.4
0.1
0.2
0.1+0.2
0.0
0.36
0.36
0.75
0.36
0.45
0.60
0
392
752
196
376
0
706
677
1,411
943
883
1,129
Air
03
NO2
O3+NO2
03
N02
03+N02
Air
03
NO2
NO2
SO2
03
NO2
3 h 9 M, 2 F
IE 10' ex/ 20' rest
VE =32 L/min
6h
IE 10' ex/ 20' rest
VE =32 L/min
T = 25°C
RH = 50%
2 h Head only 6 M, 6 F
rest exposure
2-h random exposures to 23.6 °C 6 M
FA, O3, NO2, and O3 + 62% RH 2 F
NO2; IE; VE = 26-29
L/min
Atopic asthmatics
22 to 41 yrs. old
Healthy NS
19 to 33 yrs. old
Healthy, NS, 56 to
85 years old
Exposure to NO2 alone had minimal
effects on FEVj. However, O3 alone or
in combination elicited significantly
greater decline in FEVj in a short (3 h)
exposure (higher concentrations) than a
long (6 h) exposure where the effects
were nonsignificant. Allergen
challenge inhalation significantly
reduced PD20 FEV; in all short but not
the long exposures. No additive or
potentiating effects have been
observed.
For NO2 and SO2 the absorbed fraction
of O3 increased relative (to baseline)
whereas after O3 exposure it decreased.
The differences explained by an
increased production of O3-reactive
substrate in ELF due to inflammation.
Exercise-induced cardiac output was
smaller with O3 + NO2 exposure
compared to FA or O3 alone.
Jenkins et al.
(1999)
Rigas et al.
(1997)
Drechsler-
Parks et al.
(1995)
-------
Table AX6-14 (cont'd). Ozone Mixed with Other Pollutants3
X
H
6
o
O
H
O
o
H
W
O
O
HH
H
W
Concentration11
ppm
(ig/m3 Pollutant
Exposure Duration
and Activity
Exposure
Conditions'
Number and
Gender of
Subjects
Subject
Characteristics
Observed Effect(s)
Reference
Nitrogen-Containing Pollutants (cont'd)
0.30
0.60
0.2
0.0
0.12
0.30
0.12+0.30
589 O3
1,129 N02
392 O3
500 HNO3
H20
0 Air
235 03
564 N02
03 + N02
2-h exposure to NO2 or 21 °C
FA, followed 3 h later by 40% RH
2-h exposure to O3, IE
VE = 20 L/min/m2 BSA
5 h 20 °C
IE (50 min/h exercise) 5% RH
VE ~ 40 L/min
2 h HNO3 or H2O fog or
air, followed by 1-h break,
followed by 3 h O3
1 h (mouthpiece) 22 °C
IE 75% RH
VE = 33 L/min
VE = 35 L/min
21 F Healthy NS,
18 to 34 years old
6 M, 4 F Healthy NS,
minimum of 1 0%
decrement in FEV;
after 3 h exposure to
0.20 ppm O3 with
50 min exercise/h
5 M, 7 F Healthy NS,
12 to 17 years old
9 M, 3 F Asthmatic
13 to 18 years old
No significant effect of NO2 exposures
on any measured parameter. Sequential
exposure of NO2 followed by O3
induced small but significantly larger
decrements in FEVj and FEF25.75 than
FA/O3 sequence. Subjects had
increased airway responsiveness to
methacholine after both exposures, with
significantly greater responsiveness
after the NO2/O3 sequences than after
the FA/O3 sequence.
Exposure to HNO3 or H2O fog followed
by O3 induced smaller pulmonary
function decrements than air followed
by03.
Findings inconsistent across cohorts
and atmospheres. No significant
differences in FEV; and RT between
asthmatics and healthy, or between
atmospheres and cohorts.
Hazucha
etal. (1994)
Aris etal.
(1991)
Koenig et al.
(1988)
0.30 589 O3 1 h (mouthpiece)
0.60 1,129 N02 CE
VE ~ 70 L/min for men
VB ~ 50 L/min for women
20 M, 20 F
Healthy NS,
21.4±1.5(SD)
years old for F,
22.7 ± 3.3 (SD) years
old for M
No differences between responses to
O3 and NO2 + O3 for spirometric
parameters. Increase in SRaw with
NO2 + O3 was significantly less than for
O, alone.
Adams etal.
(1987)
0.30
0.30
0.15
0.15
0.15
0.15
0.15
589
564
200
294
284
200
294
282
393
200
03
N02
H2S04
03
NO2
H2SO4
03
N02
S02
H2S04
2 hCE for 20 min 28 to 29 °C 6M
V = 25 L/min 50 to 60% RH
2 h, 60 min 6 M
total exercise
V ~ 25 L/min
2 h, 60 min 3 M
total exercise
V = 25 L/min
Healthy subjects, Possible small decrease in SGaw
some smokers
Possible small decrease in SGaw
Possible small decrease in FEV;
Kagawa
(1986)
-------
Table AX6-14 (cont'd). Ozone Mixed With Other Pollutants3
X
Concentration11
ppm
Hg/m3
Pollutant
Exposure Duration
and Activity
Number and
Exposure Gender of Subject
Conditions' Subjects Characteristics
Observed Effect(s)
Reference
Peroxyacetyl Nitrate
0.45
0.60
0.13
0.45
0.30
0.485
0.27
883
1,129
644
883
1,485
952
1,337
03
N02
PAN
03
PAN
03
PAN
2h
IE
VE = 25 L/min
2h
IE
VE = 27 L/min
2h
IE
VE = 25 L/min
24 °C 8M, 8F Healthy NS;
55 to 58% RH 8 M, 8 F 19 to 26 years old;
51 to 76 years old
22 °C 3 M, 5 F Healthy NS,
60% RH mean age = 24 years
21 °C 10 F Healthy NS,
WBGT 19 to 36 years old
No differences between
responses to O3 alone, O3 + NO2, O3 +
PAN, or 03 + N02 + PAN.
No differences between
responses to exposure to
O3 alone and O3 + PAN.
Exposure to the mixture of PAN + O3
induced decrements in FVC and FEVj
averaging 10% greater than observed
following exposure to O3 alone.
Drechsler-
Parks et al.
(1989)
Drechsler-
Parks et al.
(1987b)
Horvath
etal. (1986)
Particle-Containing Pollutants
0.0
0.12
0
235b +
153"
Air
03 +
PM25
2-2.5 h
rest
22 °C 15M, 10 F Healthy NS
30% RH 18to50yrs. old
Neither systolic nor diastolic pressure
has been affected by pollutants
exposure despite a significant brachial
artery constriction and a reduction in
arterial diameter when compared to
filtered air (p = 0.03). Absence of
flow- and nitroglycerin-mediated
brachial artery dilatation.
Brook et al.
(2002)
H
6
o
o
H
O
o
H
W
O
"See Appendix A for abbreviations and acronyms.
'Grouped by pollutant mixture.
O
HH
H
W
-------
differences in time course of exposure and history of previous exposure of children to pollutants,
psychological and physiological factors related to chamber exposures, and by other conjectures.
AX6.11.2 Ozone and Nitrogen-Containing Pollutants
Nitrogen dioxide is a key component of the photooxidation cycle and formation of O3.
Both gases are almost invariably present in ambient atmosphere. Compared to O3, NOX species
have limited solubility and moderate oxidizing capability. Both O3 and NO2 are irritants and
tissue oxidants and exert their toxic actions through many common mechanisms. The regional
dosimetry and the primary sites of action of O3 and NO2 overlap but are not the same. Since
these gases are relatively insoluble in water, they will likely penetrate into the peripheral airways
that are more sensitive to damage than better protected conducting airways. The controlled
studies reviewed in the previous O3 criteria document (Table AX6-14) generally reported only
small pulmonary function changes after combined exposures of NO2 or nitric acid (HNO3) with
O3, regardless if the interactive effects were potentiating or additive. In two of these studies, the
effects reached statistical significance, but they were not coherent. Preexposure with NO2
potentiated both spirometric and nonspecific airway reactivity response following subsequent O3
exposure (Hazucha et al., 1994); however, exposure to NO2 + O3 mixture blunted SRaw increase
as compared to O3 alone (Adams et al.,1987). As with O3 and SOX mixtures, the effects have
been dominated by O3 (U.S. Environmental Protection Agency, 1996).
Combined exposure to O3 and NO2 also blunted the exercise-induced increase in cardiac
output found with FA and O3 exposures alone (Drechsler-Parks, 1995). Eight healthy older
subjects (56 to 85 years of age) were exposed for 2 h to FA, 0.60 ppm NO2, 0.45 ppm O3, and to
0.60 ppm NO2 + 0.45 ppm O3 while alternating 20-min periods of rest and exercise. Cardiac
output, FIR, stroke volume, and systolic time intervals were measured by noninvasive impedance
cardiography at the beginning of each exposure, while the subjects were at rest, and again during
the last 5 min of exercise. Metabolic exercise data (VE, VO2, fB) also were measured. There
were no statistically significant differences between exposures for HR, VE, VO2, fB, stroke
volume, or systolic time intervals. Exercise increased cardiac output after all exposures;
however, the incremental increase over rest was significantly smaller for the combined O3
and NO2 exposures. The authors speculated that nitrate and nitrite reaction products from the
interaction of O3 and NO2 cross the air/blood interface in the lungs, causing peripheral
AX6-135
-------
vasodilation and a subsequent drop in cardiac output. No major cardiovascular effects of O3
only exposures have been reported in human subjects (see Section AX6.10).
Despite suggested potentiation of O3 response by NO2 in healthy subjects, it is unclear
what response, and at what dose, either sequential or combined gas exposures will induce
in asthmatics. Jenkins et al. (1999) exposed 11 atopic asthmatics in random order to air,
0.1 ppm O3, 0.2 ppm NO2, and 0.1 ppm O3 + 0.2 ppm NO2 for 6 h (IE for 10 min @ 32 L/min
every 40 min). Two weeks later, 10 of these subjects were exposed for 3 h to doubled
concentrations of these gases (i.e., 0.2 ppm O3, 0.4 ppm NO2, and 0.2 ppm O3 + 0.4 ppm NO2)
employing the same exercise regimen. Immediately following each exposure, subjects were
challenged with allergen (D. pteronyssinus) and PD20 FEVj was determined. Exposure to NO2
alone had minimal effects on FEVj or airway responsiveness. However, O3 alone or in
combination with NO2 elicited a significantly (p < 0.05) greater decline in FEVj in a short (3 h)
exposure (higher concentrations) than the long (6 h) exposure, where the effects were not
significant. Allergen challenge inhalation significantly (p = 0.018 to 0.002) reduced PD20 FEVj
in all short, but not the long, exposures. No associations were observed between pollutant
concentrations and physiologic endpoints. The statistical analyses of these data suggest that the
combined effect (O3 + NO2) on lung function (FVC, FEVj) was not significantly greater than the
effect of individual gases for 6-h exposures, thus no additive or potentiating effects have been
observed. Shorter 3-h exposures using twice as high NO2 concentrations, however, showed
significant FEVj decrements following exposures to atmospheres containing O3. The analysis
also suggests that it is the inhaled concentration, rather than total dose, that determines lung
airway responsiveness to allergen.
The potential for interaction between O3 and other gas mixtures was studied by Rigas et al.
(1997). They used an O3 bolus absorption technique to determine how exposures to O3, NO2,
and SO2 will affect distribution of O3 adsorption by airway mucosa. The selected O3 bolus
volume was set to reach lower conducting airways. Healthy young nonsmokers (6F/6M) were
exposed on separate days at rest in a head dome to 0.36 ppm O3, 0.36 ppm NO2, 0.75 ppm NO2
and 0.75 ppm SO2 for 2 h. The rationale for the selection of these gases was their differential
absorption. Because O3 and NO2 are much less soluble in liquid (i.e., ELF) than SO2, they are
expected to penetrate deeper into the lung than SO2 which is absorbed more quickly in the
epithelial lining fluid of the upper airways. The actual experimental measurements have shown
AX6-136
-------
that during continuous NO2 and SO2 exposure the absorbed fraction of an O3 bolus in lower
conducting airways increased relative to baseline, whereas during continuous O3 exposure the O3
bolus fraction in lower conducting airways decreased. The authors attempted to explain the
differences by suggesting that there may be increased production of an O3-reactive substrate
in epithelial lining fluid due to airway inflammation. As interpreted by the investigators,
during NO2 and SO2 exposures the substrate was not depleted by these gases and so could react
with the O3 bolus, whereas during O3 exposure the substrate was depleted, causing the fractional
absorption of the O3 bolus to decrease. Greater absorption in males than females for all gases
was attributed to anatomical differences in the bronchial tree.
AX6.11.3 Ozone and Other Pollutant Mixtures Including Particulate Matter
Almost all of the studies published over the last twenty years investigating the health
effects of mixtures of O3 with other air pollutants involved peroxyacetyl nitrate (PAN). These
studies on healthy individuals exposed under laboratory conditions came from the Horvath
laboratory at UC Santa Barbara (Table AX6-13). In the last of this series of studies, Drechsler-
Parks and colleagues (1989) found the same equivocal interaction of O3 and PAN as in previous
studies, which is attributable to O3 exposure alone (U.S. Environmental Protection Agency,
1996). Subsequently, only a couple of studies have investigated the effects of more complex air
pollutant mixtures on human pathophysiology under controlled conditions.
It is not only the interaction between air pollutants in ambient air; but, as Rigas et al.
(1997) has found, an uneven distribution of O3, SO2, and NO2 absorption in the lower conducting
airways of young healthy subjects may modulate pathophysiologic response as well. Exposure
to SO2 and NO2 increased, while exposure to O3 decreased, the absorbing capacity of the airways
for O3. The authors have suggested that SO2 or NO2-inflamed airways release additional
substrates into the epithelial lining fluid that react with O3, thus progressively removing O3 from
the airway lumen. This mechanism may explain findings of antagonistic response (e.g., Adams
et al., 1987; Dreschler-Parks, 1995) when the two gases are combined in an exposure
atmosphere.
The mechanisms by which inhalation exposure to other complex ambient atmospheres
containing particulate matter (PM) and O3 induce cardiac events frequently reported in
epidemiologic studies are rarely studied in human subjects under laboratory conditions.
AX6-137
-------
Recently, Brook et al. (2002) have reported changes in brachial artery tone and reactivity in
healthy nonsmokers following 2-h exposures to a mixture of 0.12 ppm O3 and 153 |ig/m3 of
concentrated ambient PM25, and a control atmosphere of filtered air with a trace of O3
administered in random order. Neither systolic nor diastolic pressure was affected by pollutant
exposure despite a significant brachial artery constriction and a reduction in arterial diameter
when compared to filtered air (p = 0.03). The authors postulate that changes in arterial tone may
be a plausible mechanism of air pollution-induced cardiac events. However, the observations of
no changes in blood pressure, and an absence of flow- and nitroglycerin- mediated brachial
artery dilatation, cast some doubt on the plausibility of this mechanism. A number of other
proposed mechanisms advanced to establish a link between cardiac events due to pollution and
changes in vasomotor tone based on the findings of this study are purely speculative.
AX6.12 CONTROLLED STUDIES OF AMBIENT AIR EXPOSURES
A large amount of informative O3 exposure-effects data has been obtained in controlled
laboratory exposure studies under a variety of different experimental conditions. However,
laboratory simulation of the variable pollutant mixtures present in ambient air is not practical.
Thus, the exposure effects of one or several artificially generated pollutants (i.e., a simple
mixture) on pulmonary function and symptoms may not explain responses to ambient air where
complex pollutant mixtures exist. Epidemiologic studies, which do investigate ambient air
exposures, do not typically provide the level of control and monitoring necessary to adequately
characterize short term responses. Thus, controlled exposures to ambient air using limited
numbers of volunteers have been used to try and bridge the gap between laboratory and
community exposures.
AX6.12.1 Mobile Laboratory Studies
As presented in previous criteria documents (U.S. Environmental Protection Agency, 1986;
1996), quantitatively useful information on the effects of acute exposure to photochemical
oxidants on pulmonary function and symptoms responses originated from field studies using a
mobile laboratory. These field studies involved subjects exposed to ambient air, FA without
pollutants, or FA containing artificially generated concentrations of O3 that are comparable to
AX6-138
-------
those measured in the ambient environment. As a result, measured pulmonary responses in
ambient air can be directly compared to those found in more artificial or controlled conditions.
However, the mobile laboratory shares some of the same limitations of stationary exposure
laboratories (e.g., limited number of both subjects and artificially generated pollutants for
testing). Further, mobile laboratory ambient air studies are dependent on ambient outdoor
conditions which can be unpredictable, uncontrollable, and not completely characterizable.
As summarized in Table AX6-15, investigators in California used a mobile laboratory and
demonstrated that pulmonary effects of ambient air in Los Angeles residents are related to O3
concentration and level of exercise (Avol et al., 1983, 1984, 1985a,b,c, 1987; Linn et al., 1980,
1983). Avol et al. (1987) observed no significant pulmonary function or symptoms responses in
children (8 to 11 years) engaged in moderate continuous exercise for 1 h while breathing
ambient air with an O3 concentration of 0.113 ppm. However, significant pulmonary function
decrements and increased symptoms of breathing discomfort were observed in healthy
exercising (1 h continuous) adolescents (Avol et al., 1985a,b), athletes, (Avol et al., 1984, 1985c)
and lightly exercising asthmatic subjects (Linn et al., 1980, 1983) at O3 concentrations averaging
from 0.144 to 0.174 ppm. Many of the healthy subjects with a history of allergy appeared to be
more responsive to O3 than "nonallergic" subjects (Linn et al., 1980, 1983), although a
standardized evaluation of atopic status was not performed. Comparative studies of exercising
athletes (Avol et al., 1984, 1985c) with chamber exposures to oxidant-polluted ambient air
(mean O3 concentration of 0.153 ppm) and purified air containing a controlled concentration of
generated O3 at 0.16 ppm showed similar pulmonary function responses and symptoms,
suggesting that acute exposures to coexisting ambient pollutants had minimal contribution to
these responses under the typical summer ambient conditions in Southern California. This
contention is similar to, but extends, the laboratory finding of no significant difference in
pulmonary function effects between O3 and O3 plus PAN exposures (Drechsler-Parks, 1987b).
Additional supporting evidence is provided in Section AX6.11.
AX6.12.2 Aircraft Cabin Studies
Respiratory symptoms and pulmonary function effects resulting from exposure to O3 in
commercial aircraft flying at high altitudes, and in altitude-simulation studies, have been
reviewed elsewhere (U.S. Environmental Protection Agency, 1986, 1996). Flight attendants,
AX6-139
-------
Table AX6-15. Acute Effects of Ozone in Ambient Air in Field Studies with a Mobile Laboratory"
Mean Ozone
Concentration" Ambient
Temperature0 Exposure Activity Level
ppm MS/m3 (°C) Duration (VE) Number of Subjects
Observed Effect(s)
Reference
X
0.113 ±.033 221 ±65 33 ± 1 Ih CE (22 L/min) 66 healthy children,
8 to 11 years old
0.144 ±.043 282 ±84 32 ± 1
0.153 ±.025 300 ±49
0.156 ±.055 306 ±107
0.165 ±.059 323 ±115
0.174 ±.068 341 ±133
32 ±2
33 ±4
33 ±3
33 ±2
1 h CE (32 L/min)
Ih
Ih
Ih
2h
CE (53 L/min)
CE (38 L/min)
CE (42 L/min)
59 healthy
adolescents,
12 to 15 years old
50 healthy adults
(competitive
bicyclists)
48 healthy adults,
50 asthmatic adults
60 "healthy" adults
(7 were asthmatic)
IE (2 times resting) 34 "healthy" adults,
at 15-min intervals 30 asthmatic adults
No significant changes in forced expiratory Avol et al.
function and symptoms of breathing (1987)
discomfort after exposure to 0.113 ppm O3 in
ambient air.
Small significant decreases in FVC (-2.1%), Avol et al.
FEV075 (-4.0%), FEV; (-4.2%), and PEFR (1985a,b)
(-4.4%) relative to control with no recovery
during a 1-h postexposure rest; no significant
increases in symptoms.
Mild increases in symptoms scores and Avol et al.
significant decreases in FEV; (-5.3%) and (1984, 1985c)
FVC; mean changes in ambient air were not
statistically different from those in purified air
containing 0.16 ppm O3.
No significant changes for total symptom Linn et al.
score or forced expiratory performance in (1983)
normals or asthmatics; however, FEV; Avol et al.
remained low or decreased further (-3%) 3 h (1983)
after ambient air exposure in asthmatics.
Small significant decreases in FEV[ (-3.3%) Linn et al.
and FVC with no recovery during a 1-h (1983)
postexposure rest; TLC decreased and AN2 Avol et al.
increased slightly. (1983)
Increased symptom scores and small Linn et al.
significant decreases in FEV; (-2.4%), FVC, (1980, 1983)
PEFR, and TLC in both asthmatic and healthy
subjects; however, 25/34 healthy subjects were
allergic and "atypically" reactive to polluted
ambient air.
aSee Appendix A for abbreviations and acronyms.
bRanked by lowest level of O3 in ambient air, presented as the mean ± SD.
cMean±SD.
-------
because of their physical activities at altitude, tend to receive higher exposures. In a series of
hypobaric chamber studies of nonsmoking subjects exposed to 1,829 m (6,000 ft) and O3 at
concentrations of 0.2 and 0.3 ppm for 3 or 4 h (Lategola et al., 1980a,b), increased symptoms
and pulmonary function decrements occurred at 0.3 ppm but not at 0.2 ppm.
Commercial aircraft cabin O3 levels were reported to be very low (average concentration
0.01 to 0.02 ppm) during 92 randomly selected smoking and nonsmoking flights in 1989 (Nagda
et al., 1989). None of these flights recorded O3 concentrations exceeding the 3-h time-weighted
average (TWA) standard of 0.10 ppm promulgated by the Federal Aviation Administration
(FAA, 1980), probably due to the use of O3-scrubbing catalytic filters (Melton, 1990). However,
in-flight O3 exposure can still occur because catalytic filters are not necessarily in continuous use
during flight. Other factors to consider in aircraft cabins, however, are erratic temperature
changes, lower barometric pressure and oxygen pressure, and lower humidity, often reaching
levels between 4 and 17% (Rayman, 2002).
Ozone contamination aboard high-altitude aircraft also has been an interest to the U.S. Air
Force because of complaints by crew members of frequent symptoms of dryness and irritation of
the eyes, nose, and throat and an occasional cough (Hetrick et al., 2000). Despite the lack of
ventilation system modifications as used in commercial aircraft, the O3 concentrations never
exceeded the FAA ceiling limit of 0.25 ppm and exceeded the 3-h TWA of 0.10 ppm only 7% of
the total monitored flight time (43 h). The authors concluded that extremely low average
relative humidity (12%) during flight operations was most likely responsible for the reported
symptoms.
AX6-141
-------
REFERENCES
Adams, W. C. (2000a) Feasibility study of prolonged ozone inhalation exposure via face mask. Inhalation Toxicol.
12:299-313.
Adams, W. C. (2000b) Ozone dose-response effects of varied equivalent minute ventilation rates. J. Exposure Anal.
Environ. Epidemiol. 10: 217-226.
Adams, W. C. (2002) Comparison of chamber and face-mask 6.6-hour exposures to ozone on pulmonary function
and symptoms responses. Inhalation Toxicol. 14: 745-764.
Adams, W. C. (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. Inhalation Toxicol. 15: 265-281.
Adams, W. C. (2003b) Relation of pulmonary responses induced by 6.6-h exposures to 0.08 ppm ozone and 2-h
exposures to 0.30 ppm ozone via chamber and face-mask inhalation. Inhalation Toxicol. 15: 745-759.
Adams, W. C. (2006) Comparison of chamber 6.6-h exposures to 0.04-0.08 PPM ozone via square-wave and
triangular profiles on pulmonary responses. Inhalation Toxicol. 18: 127-136.
Adams, W. C.; Ollison, W. M. (1997) Effects of prolonged simulated ambient ozone dosing patterns on human
pulmonary function and symptomatology. Presented at: 90th annual meeting of the Air & Waste Management
Association; June; Toronto, Ontario, Canada. Pittsburgh, PA: Air & Waste Management Association; paper
no. 97-MP9.02.
Adams, W. C.; Schelegle, E. S. (1983) Ozone and high ventilation effects on pulmonary function and endurance
performance. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 55: 805-812.
Adams, W. C.; Savin, W. M.; Christo, A. E. (1981) Detection of ozone toxicity during continuous exercise via the
effective dose concept. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 51: 415-422.
Adams, W. C.; Brookes, K. A.; Schelegle, E. S. (1987) Effects of NO2 alone and in combination with O3 on young
men and women. J. Appl. Physiol. 62: 1698-1704.
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. Inhalation Toxicol. 12: 1205-1224.
Alexis, N. E.; Becker, S.; Bromberg, P. A.; Devlin, R.; Peden, D. B. (2004) Circulating CDllb expression correlates
with the neutrophil response and airway mCD14 expression is enhanced following ozone exposure in
humans. Clin. Immunol. Ill: 126-131.
Alpert, S. E.; Walenga, R. W.; Jaspers, I.; Qu, Q.; Chen, L. C. (1997) Ozone inactivates Cyclooxygenase in human
tracheal epithelial cells without altering PGHS-2 mRNA or protein. Am. J. Physiol. 272: L879-L887.
Arab, L.; Steck-Scott, S.; Fleishauer, A. T. (2002) Lycopene and the lung. Exp. Biol. Med. 227: 894-899.
Aris, R.; Christian, D.; Sheppard, D.; Balmes, J. R. (1991) The effects of sequential exposure to acidic fog and
ozone on pulmonary function in exercising subjects. Am. Rev. Respir. Dis. 143: 85-91.
Aris, R. M.; Christian, D.; Hearne, P. Q.; Kerr, K.; Finkbeiner, W. E.; Balmes, J. R. (1993) Ozone-induced airway
inflammation in human subjects as determined by airway lavage and biopsy. Am. Rev. Respir. Dis.
148: 1363-1372.
Aris, R. M.; Tager, L; Christian, D.; Kelly, T.; Balmes, J. R. (1995) Methacholine responsiveness is not associated
with O3-induced decreases inFEVj. Chest 107: 621-628.
Avissar, N. E.; Reed, C. K.; Cox, C.; Frampton, M. W.; Finkelstein, J. N. (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, E. L.; Linn, W. S.; Shamoo, D. A.; Venet, T. G.; Hackney, J. D. (1983) Acute respiratory effects of
Los Angeles smog in continuously exercising adults. J. Air Pollut. Control Assoc. 33: 1055-1060.
Avol, E. L.; Linn, W. S.; Venet, T. G.; Shamoo, D. A.; Hackney, J. D. (1984) Comparative respiratory effects of
ozone and ambient oxidant pollution exposure during heavy exercise. J. Air Pollut. Control Assoc.
34: 804-809.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Valencia, L. M.; Anzar, U. T.; Venet, T. G.; Hackney, J. D. (1985a)
Respiratory effects of photochemical oxidant air pollution in exercising adolescents. Am. Rev. Respir. Dis.
132: 619-622.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Valencia, L. M.; Anzar, U. T.; Hackney, J. D. (1985b) Short-term health
effects of ambient air pollution in adolescents. In: Lee, S. D., ed. Evaluation of the scientific basis for
ozone/oxidants standards: transactions of an APCA international specialty conference; November 1984;
Houston, TX. Pittsburgh, PA: Air Pollution Control Association; pp. 329-336. (APCA international specialty
conference transactions: TR-4).
AX6-142
-------
Avol, E. L.; Linn, W. S.; Venet, T. G.; Shamoo, D. A.; Spier, C. E.; Hackney, J. D. (1985c) Comparative effects of
laboratory generated ozone and ambient oxidant exposure in continuously exercising subjects. In: Lee, S. D.,
ed. Evaluation of the scientific basis for ozone/oxidants standards: proceedings of an APCA international
specialty conference; November 1984; Houston, TX. Pittsburgh, PA: Air Pollution Control Association;
pp. 216-225. (APCA international specialty conference transactions: TR-4).
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Spier, C. E.; Valencia, L. M; Venet, T. G.; Trim, S. C.; Hackney, J. D.
(1987) Short-term respiratory effects of photochemical oxidant exposure in exercising children. JAPCA
37: 158-162.
Avol, E. L.; Linn, W. S.; Shamoo, D. A.; Hackney, J. D. (1988) Seasonal ozone reactivity in Los Angeles residents.
Presented at: 81st annual meeting of the Air Pollution Control Association; June; Dallas, TX. Pittsburgh, PA:
Air Pollution Control Association; paper no. 88-122.6.
Ball, B. A.; Folinsbee, L. J.; Peden, D. B.; Kehrl, H. R. (1996) Allergen bronchoprovocation of patients with mild
allergic asthma after ozone exposure. J. Allergy Clin. Immunol. 98: 563-572.
Balmes, J. R.; Chen, L. L.; Scannell, C.; Tager, I.; Christian, D.; Hearne, P. Q.; Kelly, T.; Aris, R. M. (1996)
Ozone-induced decrements in FEVj and FVC do not correlate with measures of inflammation. Am. J. Respir.
Crit. CareMed. 153: 904-909.
Balmes, J. R.; Aris, R. M.; Chen, L. L.; Scannell, C.; Tager, I. B.; Finkbeiner, W.; Christian, D.; Kelly, T.; Hearne,
P. Q.; 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. Cambridge, MA:
Health Effects Institute. Research report no. 78; pp 1-37, 81-99.
Barman, S. A. (2001) Effect of protein kinase C inhibition on hypoxic pulmonary vasoconstriction. Am. J. Physiol.
Lung Cell Mol. Physiol. 280: L888-L895.
Bascom, R.; Naclerio, R. M.; Fitzgerald, T. K.; Kagey-Sobotka, A.; Proud, D. (1990) Effect of ozone inhalation on
the response to nasal challenge with antigen of allergic subjects. Am. Rev. Respir. Dis. 142: 594-601.
Basha, M. A.; Gross, K. B.; Gwizdala, C. J.; Haidar, A. H.; 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.
Bates, D. V; Bell, G. M.; Burnham, C. D.; Hazucha, M.; Mantha, J.; Pengelly, L. D.; Silverman, F. (1972)
Short-term effects of ozone on the lung. J. Appl. Physiol. 32: 176-181.
Bayram, H.; Sapsford, R. J.; Abdelaziz, M. M.; Khair, O. A. (2001) Effect of ozone and nitrogen dioxide on the
release of proinflammatory mediators from bronchial epithelial cells of nonatopic nonasthmatic subjects and
atopic asthmatic patients in vitro. J. Allergy Clin. Immunol. 107: 287-294.
Bayram, H.; Rusznak, C.; Khair, O. A.; Sapsford, R. J.; Abdelaziz, M. M. (2002) Effect of ozone and nitrogen
dioxide on the permeability of bronchial epithelial cell cultures of non-asthmatic and asthmatic subjects.
Clin. Exp. Allergy 32: 1285-1292.
Beck, N.; Koenig, J. Q.; Luchtel, D. L.; Altaian, L. C.; Orsborn, M. T.; Kenney, J. S. (1994) Ozone can increase the
expression of intercellular adhesion molecule-1 and the synthesis of cytokines by human nasal epithelial cells.
Inhalation Toxicol. 6: 345-357.
Becker, S.; Madden, M. C.; Newman, S. L.; Devlin, R. B.; Koren, H. S. (1991) Modulation of human alveolar
macrophage properties by ozone exposure in vitro. Toxicol. Appl. Pharmacol. 110: 403-415.
Bedi, J. F.; Drechsler-Parks, D. M.; Horvath, S. M. (1985) Duration of increased pulmonary function sensitivity to
an initial ozone exposure. Am. Ind. Hyg. Assoc. J. 46: 731-734.
Bedi, J. F.; Horvath, S. M.; Drechsler-Parks, D. M. (1988) Reproducibility of the pulmonary function response of
older men and women to a 2-hour ozone exposure. JAPCA 38: 1016-1019.
Bedi, J. F.; Horvath, S. M.; Drechsler-Parks, D. M. (1989) Adaptation by older individuals repeatedly exposed to
0.45 parts per million ozone for two hours. JAPCA 39: 194-199.
Bergamaschi, E.; De Palma, G.; Mozzoni, P.; Vanni, S.; Vettori, M. V.; 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.
Bhalla, D. K. (2002) Interactive effects of cigarette smoke and ozone in the induction of lung injury. Toxicol. Sci.
65: 1-3.
Blomberg, A.; Helleday, R.; Pourazar, J.; Stenfors, N.; Kelly, F. J.; Frew, A. J.; Holgate, S. T.; Sandstrom, T. (1997)
Early airway and peripheral blood cell responses to 0.20 ppm ozone in healthy human subjects. Eur. Respir. J.
10(suppl 25): 274S.
AX6-143
-------
Blomberg, A.; Mudway, I. S.; Nordenhall, C.; Hedenstrom, H.; Kelly, F. I; Frew, A. I; Holgate, S. T.;
Sandstrom, T. (1999) Ozone-induced lung function decrements do not correlate with early airway
inflammatory or antioxidant responses. Eur. Respir. J. 13: 1418-1428.
Blomberg, A.; Mudway, I.; Svensson, M; Hagenbjork-Gustafsson, A.; Thomasson, L.; Helleday, R.; Dumont, X.;
Forsberg, B.; Nordberg, G.; Bernard, A. (2003) Clara cell protein as a biomarker for ozone-induced lung
injury in humans. Eur. Respir. J. 22: 883-888.
Bosson, J.; Stenfors, N.; Bucht, A.; Helleday, R.; Pourazar, J.; Holgate, S. T.; Kelly, f. J.; Sandstrom, T.; Wilson, S.;
Frew, A. J.; Blomberg, A. (2003) Ozone-induced bronchial epithelial cytokine expression differs between
healthy and asthmatic subjects. Clin. Exp. Allergy 33: 777-782.
Brook, R. D.; Brook, J. R.; Urch, B.; Vincent, R.; Rajagopalan, S.; Silverman, F. (2002) Inhalation of fine paniculate
air pollution and ozone causes acute arterial vasoconstriction in healthy adults. Circulation 105: 1534-1536.
Brookes, K. A.; Adams, W. C.; Schelegle, E. S. (1989) 0.35 ppm O3 exposure induces hyperresponsiveness on 24-h
reexposure to 0.20 ppmO3. J. Appl. Physiol. 66: 2756-2762.
Brummer, M. E. G.; Schwartz, L. W.; McQuillen, N. K. (1977) A quantitative study of lung damage by scanning
electron microscopy. Inflammatory cell response to high-ambient levels of ozone. In: Johari, O.; Becker,
R. P., eds. Scanning electron microscopy/1977/11 - biological applications of the SEM: proceedings of the
workshops on advances inbiomedical applications of the SEM & STEM; March-April; Chicago, IL. Chicago,
IL: IIT Research Institute; pp. 513-518.
Bush, M. L.; Asplund, P. T.; Miles, K. A.; Ben-Jebria, A.; Ultman, J. S. (1996) Longitudinal distribution of O3
absorption in the lung: gender differences and intersubject variability. J. Appl. Physiol. 81: 1651-1657.
Centers for Disease Control and Prevention. (2003) Key clinical activities for quality asthma care: recommendations
of the National Asthma Education and Prevention Program. Morb. Mortal. Wkly. Rep. 52(RR-6): 1-8.
Available: http://www.cdc.gov/mmwr/PDF/rr/rr5206.pdf (11 April 2003).
Chen, L. L.; Tager, I. B.; Peden, D. B.; Christian, D. L.; Ferrando, R. E.; Welch, B. S.; Balmes, J. R. (2004) Effect
of ozone exposure on airway responses to inhaled allergen in asthmatic subjects. Chest 125: 2328-2335.
Christian, D. L.; Chen, L. L.; Scannell, C. H.; Ferrando, R. E.; Welch, B. S.; Balmes, J. R. (1998) Ozone-induced
inflammation is attenuated with multiday exposure. Am. J. Respir. Crit. Care Med. 158: 532-537.
Coffey, M. J.; Wheeler, C. S.; Gross, K. B.; Eschenbacher, W. L.; Sporn, P. H. S.; Peters-Golden, M. (1996)
Increased 5-lipoxygenase metabolism in the lungs of human subjects exposed to ozone. Toxicology
114: 187-197.
Coleridge, J. C. G.; Coleridge, H. M.; Schelegle, E. S.; Green, J. F. (1993) Acute inhalation of ozone stimulates
bronchial C-fibers and rapidly adapting receptors in dogs. J. Appl. Physiol. 74: 2345-2352.
Collins, D. V.; Cutillo, A. G.; Armstrong, J. D.; Crapo, R. O.; Kanner, R. E.; Tocino, L; Renzetti, A. D., Jr. (1986)
Large airway size, lung size, and maximal expiratory flow in healthy nonsmokers. Am. Res. Respir. Dis.
134: 951-955.
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.
Criqui, G. I.; Solomon, C.; Welch, B. S.; Ferrando, R. E.; Boushey, H. A.; Balmes, J. R. (2000) Effects of
azithromycin on ozone-induced airway neutrophilia and cytokine release. Eur. Respir. J. 15: 856-862.
Cropp, G. J. A.; Bernstein, I. L.; Boushey, H. A., Jr.; Hyde, R. W.; Rosenthal, R. R.; Specter, S. L.; Townley, R. G.
(1980) Guidelines for bronchial inhalation challenges with pharmacologic and antigenic agents. ATS News
(Spring): 11-19.
David, G. L.; Romieu, I.; Sienra-Monge, J. J.; Collins, W. J.; Ramirez-Aguilar, M.; Del Rio-Navarro, B. E.;
Reyes-Ruiz, N. I.; Morris, R. W.; Marzec, J. M.; London, S. J. (2003) Nicotinamide adenine dinucleotide
(Phosphate) reduced:quinone oxidoreductase and glutathione s-transferase ml polymorphism and childhood
asthma. Am. J. Respir. Crit. Care Med. 168: 1199-1204.
Delaunois, A.; Segura, P.; Montano, L. M.; Vargas, M. H.; Ansay, M.; Gustin, P. (1998) Comparison of
ozone-induced effects on lung mechanics and hemodynamics in the rabbit. Toxicol. Appl. Pharmacol.
150: 58-67.
Devlin, R. B.; Koren, H. S. (1990) The use of quantitative two-dimensional gel electrophoresis to analyze changes in
alveolar macrophage proteins in humans exposed to ozone. Am. J. Respir. Cell Mol. Biol. 2: 281-288.
Devlin, R. B.; McDonnell, W. F.; Koren, H. S.; Becker, S. (1990) Prolonged exposure of humans to 0.10 and
0.08 ppm ozone results in inflammation in the lung. Presented at: 83rd annual meeting of the Air & Waste
Management Association; June; Pittsburgh, PA. Pittsburgh, PA: Air & Waste Management Association;
paper no. 90-150.2.
AX6-144
-------
Devlin, R. B.; McDonnell, W. F.; Mann, R.; Becker, S.; House, D. E.; Schreinemachers, D.; Koren, H. S. (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, R. B.; McKinnon, K. P.; Noah, T.; Becker, S.; Koren, H. S. (1994) Ozone-induced release of cytokines and
fibronectinby alveolar macrophages and airway epithelial cells. Am. J. Physiol. 266: L612-L619.
Devlin, R. B.; McDonnell, W. F.; Becker, S.; Madden, M. C.; McGee, M. P.; Perez, R.; Hatch, G.; House, D. E.;
Koren, H. S. (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, R. B.; Folinsbee, L. J.; Biscardi, F.; Hatch, G.; Becker, S.; Madden, M. C.; Robbins, M.; Koren, H. S. (1997)
Inflammation and cell damage induced by repeated exposure of humans to ozone. Inhalation Toxicol.
9:211-235.
Dimeo, M. J.; Glenn, M. G.; Holtzman, M. J.; Sheller, J. R.; Nadel, J. A.; Boushey, H. A. (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.
Drechsler-Parks, D. M. (1995) Cardiac output effects of O3 and NO2 exposure in healthy older adults. Toxicol. Ind.
Health 11: 99-109.
Drechsler-Parks, D. M.; Bedi, J. F.; Horvath, S. M. (1984) Interaction of peroxyacetyl nitrate and ozone on
pulmonary functions. Am. Rev. Respir. Dis. 130: 1033-1037.
Drechsler-Parks, D. M.; Bedi, J. F.; Horvath, S. M. (1987a) Pulmonary function responses of older men and women
to ozone exposure. Exp. Gerontol. 22: 91-101.
Drechsler-Parks, D. M.; Bedi, J. F.; Horvath, S. M. (1987b) Pulmonary function desensitization on repeated
exposures to the combination of peroxyacetyl nitrate and ozone. JAPCA37: 1199-1201.
Drechsler-Parks, D. M.; Bedi, J. F.; Horvath, S. M. (1989) Pulmonary function responses of young and older adults
to mixtures of O3, NO2 and PAN. Toxicol. Ind. Health 5: 505-517.
Drechsler-Parks, D. M.; Horvath, S. M.; Bedi, J. F. (1990) The "effective dose" concept in older adults exposed to
ozone. Exp. Gerontol. 25: 107-115.
Eschenbacher, W. L.; Ying, R. L.; Kreit, J. W.; Gross, K. B. (1989) Ozone-induced lung function changes in normal
and asthmatic subjects and the effect of indomethacin. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant,
L. D., eds. Atmospheric ozone research and its policy implications: proceedings of the 3rd US-Dutch
international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier
Science Publishers; pp. 493-499. (Studies in environmental science 35).
Fahy, J. V.; Wong, H. H.; Liu, J. T.; Boushey, H. A. (1995) Analysis of induced sputum after air and ozone
exposures in healthy subjects. Environ. Res. 70: 77-83.
Federal Aviation Administration. (1980) Airplane cabin ozone contamination. F. R. (January 21) 45: 3880-3883.
Fernandes, A. L. G.; Molfino, N. A.; McClean, P. A.; Silverman, F.; Tarlo, S.; Raizenne, M.; Slutsky, A. S.;
Zamel, N. (1994) The effect of pre-exposure to 0.12 ppm of ozone on exercise-induced asthma. Chest
106: 1077-1082.
Folinsbee, L. J.; Hazucha, M. J. (1989) Persistence of ozone-induced changes in lung function and airway
responsiveness. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research
and its policy implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen,
The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 483-492. (Studies in
environmental science 35).
Folinsbee, L. J.; Hazucha, M. J. (2000) Time course of response to ozone exposure in healthy adult females.
Inhalation Toxicol. 12: 151-167.
Folinsbee, L. J.; Horvath, S. M. (1986) Persistence of the acute effects of ozone exposure. Aviat. Space Environ.
Med. 57: 1136-1143.
Folinsbee, L. J.; Silverman, F.; Shephard, R. J. (1975) Exercise responses following ozone exposure. J. Appl.
Physiol. 38: 996-1001.
Folinsbee, L. J.; Silverman, F.; Shephard, R. J. (1977) Decrease of maximum work performance following ozone
exposure. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 42: 531-536.
Folinsbee, L. J.; Drinkwater, B. L.; Bedi, J. F.; Horvath, S. M. (1978) The influence of exercise on the pulmonary
function changes due to exposure to low concentrations of ozone. In: Folinsbee, L. J.; Wagner, J. A.; Borgia,
J. F.; Drinkwater, B. L.; Gliner, J. A.; Bedi, J. F., eds. Environmental stress: individual human adaptations.
New York, NY: Academic Press; pp. 125-145.
AX6-145
-------
Folinsbee, L. I; Bedi, J. F.; Horvath, S. M. (1980) Respiratory responses in humans repeatedly exposed to low
concentrations of ozone. Am. Rev. Respir. Dis. 121: 431-439.
Folinsbee, L. J.; Bedi, J. F.; Horvath, S. M. (1984) Pulmonary function changes after 1 h continuous heavy exercise
in0.21 ppmozone. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 57: 984-988.
Folinsbee, L. J.; Horstman, D. H.; Vorona, R. D.; Prince, J. M.; Berry, M. (1986) Determinants of endurance
performance during ozone inhalation. Presented at: 30th International Union of Physiological Sciences
congress; July; Vancouver, BC, Canada. Proc. Int. Union Physiol. Sci. 16: 176.
Folinsbee, L. J.; McDonnell, W. F.; Horstman, D. H. (1988) Pulmonary function and symptom responses after
6.6-hour exposure to 0.12 ppm ozone with moderate exercise. JAPCA 38: 28-35.
Folinsbee, L. J.; Devlin, R. B.; Abdul-Salaam, S.; Koren, H. S. (1993) Repeated severe ozone exposure causes
depressed baseline spirometry. Am. Rev. Respir. Dis. 147: A638.
Folinsbee, L. J.; Horstman, D. H.; Kehrl, H. R.; Harder, S.; Abdul-Salaam, S.; Ives, P. J. (1994) Respiratory
responses to repeated prolonged exposure to 0.12 ppm ozone. Am. J. Respir. Crit. Care Med. 149: 98-105.
Folinsbee, L. J.; Devlin, R. B.; Robbins, M. K.; Biscardi, F. H.; Abdul-Salaam, S.; Koren, H. S. (1998) Repeated
exposure of humans to ozone: pulmonary function and symptom responses. Research Triangle Park, NC:
U.S. Environmental Protection Agency; National Center for Environmental Assessment; unpublished data.
Foster, W. M.; Stetkiewicz, P. T. (1996) Regional clearance of solute from the respiratory epithelia: 18-20 h
postexposure to ozone. J. Appl. Physiol. 81: 1143-1149.
Foster, W. M.; Costa, D. L.; Langenback, E. G. (1987) Ozone exposure alters tracheobronchial mucociliary function
in humans. J. Appl. Physiol. 63: 996-1002.
Foster, W. M.; Silver, J. A.; Groth, M. L. (1993) Exposure to ozone alters regional function and particle dosimetry in
the human lung. J. Appl. Physiol. 75: 1938-1945.
Foster, W. M.; Wills-Karp, M.; Tankersley, C. G.; Chen, X.; Paquette, N. C. (1996) Bloodborne markers in humans
during multiday exposure to ozone. J. Appl. Physiol. 81: 794-800.
Foster, W. M.; Weinmann, G. G.; Menkes, E.; Macri, K. (1997) Acute exposure of humans to ozone impairs small
airway function. Ann. Occup. Hyg. 41(suppl. 1): 659-666.
Foster, W. M.; Brown, R. H.; Macri, K.; Mitchell, C. S. (2000) Bronchial reactivity of healthy subjects: 18-20 h
postexposure to ozone. J. Appl. Physiol. 89: 1804-1810.
Fox, S. D.; Adams, W. C.; Brookes, K. A.; Lasley, B. L. (1993) Enhanced response to ozone exposure during the
follicular phase of the menstrual cycle. Environ. Health Perspect. 101: 242-244.
Foxcroft, W. J.; Adams, W. C. (1986) Effects of ozone exposure on four consecutive days on work performance and
VO2max. J. Appl. Physiol. 61: 960-966.
Frampton, M. W.; Morrow, P. E.; Cox, C.; Levy, P. C.; Condemi, J. J.; Speers, D.; Gibb, F. R.; Utell, M. J. (1995)
Sulfuric acid aerosol followed by ozone exposure in healthy and asthmatic subjects. Environ. Res. 69: 1-14.
Frampton, M. W.; Morrow, P. E.; Torres, A.; Cox, C.; Voter, K. Z.; Utell, M. J.; Gibb, F. R.; Speers, D. M. (1997a)
Ozone responsiveness in smokers and nonsmokers. Am. J. Respir. Crit. Care Med. 155: 116-121.
Frampton, M. W.; Balmes, J. R.; Cox, C.; Krein, P. M.; Speers, D. M.; Tsai, Y.; Utell, M. J. (1997b) Effects of
ozone on normal and potentially sensitive human subjects. Part III: mediators of inflammation in
bronchoalveolar lavage fluid from nonsmokers, smokers, and asthmatic subjects exposed to ozone: a
collaborative study. Cambridge, MA: Health Effects Institute. Research report 78; pp. 73-99, 81-99.
Frank, R.; Liu, M. C.; Spannhake, E. W.; Mlynarek, S.; Macri, K.; Weinmann, G. G. (2001) Repetitive ozone
exposure of young adults: evidence of persistent small airway dysfunction. Am. J. Respir. Crit. Care Med.
164: 1253-1260.
Freed, A. N.; Chou, C. L.; Fuller, S. D.; Croxton, T. L. (1996) Ozone-induced vagal reflex modulates airways
reactivity in rabbits. Respir. Physiol. 105: 95-102.
Gabrielson, E. W.; Yu, X.-Y.; Spannhake, E. W. (1994) Comparison of the toxic effects of hydrogen peroxide and
ozone on cultured human bronchial epithelial cells. Environ. Health Perspect. 102: 972-974.
Gerrity, T. R.; Bennett, W. D.; Kehrl, H.; DeWitt, P. J. (1993) Mucociliary clearance of inhaled particles measured
at 2 h after ozone exposure in humans. J. Appl. Physiol. 74: 2984-2989.
Gliner, J. A.; Matsen-Twisdale, J. A.; Horvath, S. M. (1979) Auditory and visual sustained attention during ozone
exposure. Aviat. Space Environ. Med. 50: 906-910.
Gliner, J. A.; Horvath, S. M.; Sorich, R. A.; Hanley, J. (1980) Psychomotor performance during ozone exposure:
spectral and discriminant function analysis of EEG. Aviat. Space Environ. Med. 51: 344-351.
Gliner, J. A.; Horvath, S. M.; Folinsbee, L. J. (1983) Preexposure to low ozone concentrations does not diminish the
pulmonary function response on exposure to higher ozone concentrations. Am. Rev. Respir. Dis. 127: 51-55.
AX6-146
-------
Golden, J. A.; Nadel, J. A.; Boushey, H. A. (1978) Bronchial hyperirritability in healthy subjects after exposure to
ozone. Am. Rev. Respir. Dis. 118: 287-294.
Gomez, F. P.; Rodriguez-Roisin, R. (2002) Global initiative for chronic obstructive lung disease (GOLD) guidelines
for chronic obstructive pulmonary disease. Curr. Opin. Pulm. Med. 8: 81-86.
Gong, H., Jr.; Tierney, D. F. (1995) Evaluation of chronic obstructive pulmonary disease patients for ozone
sensitivity: validation of health advisories. Sacramento, CA.: California State Air Resources Board, Research
Division; CARB-A133-123. Available from: NTIS, Springfield, VA; PB98-153620.
Gong, H., Jr.; Bradley, P. W.; Simmons, M. S.; Tashkin, D. P. (1986) Impaired exercise performance and pulmonary
function in elite cyclists during low-level ozone exposure in a hot environment. Am. Rev. Respir. Dis.
134: 726-733.
Gong, H., Jr.; Shamoo, D. A.; Anderson, K. R.; Linn, W. S. (1997a) Responses of older men with and without
chronic obstructive pulmonary disease to prolonged ozone exposure. Arch. Environ. Health 52: 18-25.
Gong, H., Jr.; McManus, M. S.; Linn, W. S. (1997b) Attenuated response to repeated daily ozone exposures in
asthmatic subjects. Arch. Environ. Health 52: 34-41.
Gong, H., Jr.; Wong, R.; Sarma, R. J.; Linn, W. S.; Sullivan, E. D.; Shamoo, D. A.; Anderson, K. R.; Prasad, S. B.
(1998) Cardiovascular effects of ozone exposure in human volunteers. Am. J. Respir. Crit. Care Med.
158: 538-546.
Graham, D. E.; Koren, H. S. (1990) Biomarkers of inflammation in ozone-exposed humans: comparison of the nasal
and bronchoalveolar lavage. Am. Rev. Respir. Dis. 142: 152-156.
Graham, D.; Henderson, F.; House, D. (1988) Neutrophil influx measured in nasal lavages of humans exposed to
ozone. Arch. Environ. Health 43: 228-233.
Greve, J. M.; Davis, G.; Meyer, A. M.; Forte, C. P.; Yost, S. C; Marlor, C. W.; Kamarck, M. E.; McClelland, A.
(1989) The major human rhinovirus receptor is ICAM-1. Cell 56: 839-847.
Grievink, L.; Jansen, S. M. A.; Van't Veer, P.; Brunekreef, B. (1998) Acute effects of ozone on pulmonary function
of cyclists receiving antioxidant supplements. Occup. Environ. Med. 55: 13-17.
Grievink, L.; Zijlstra, A. G.; Ke, X.; Brunekreef, B. (1999) Double-blind intervention trial on modulation of ozone
effects on pulmonary function by antioxidant supplements. Am. J. Epidemiol. 149: 306-314.
Gwynn, R. C.; Thurston, G. D. (2001) The burden of air pollution: impacts among racial minorities. Environ. Health
Perspect. 109(suppl. 4): 501-506.
Hackney, J. D.; Linn, W. S.; Buckley, R. D.; Hislop, H. J. (1976) Studies in adaptation to ambient oxidant air
pollution: effects of ozone exposure in Los Angeles residents vs. new arrivals. Environ. Health Perspect.
18: 141-146.
Hackney, J. D.; Linn, W. S.; Mohler, J. G.; Collier, C. R. (1977a) Adaptation to short-term respiratory effects of
ozone in men exposed repeatedly. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 43: 82-85.
Hackney, J. D.; Linn, W. S.; Karuza, S. K.; Buckley, R. D.; Law, D. C.; Bates, D. V.; Hazucha, M.; Pengelly, L. D.;
Silverman, F. (1977b) Effects of ozone exposure in Canadians and southern Californians: evidence for
adaptation? Arch. Environ. Health 32: 110-116.
Hackney, J. D.; Linn, W. S.; Shamoo, D. A.; Avol, E. L. (1989) Responses of selected reactive and nonreactive
volunteers to ozone exposure in high and low pollution seasons. In: Schneider, T.; Lee, S. D.; Welters, G. J.
R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications: proceedings of the 3rd
US-Dutch international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands:
Elsevier Science Publishers; pp. 311-318. (Studies in environmental science 35).
Hamilton, R. F.; Li, L.; Eschenbacher, W. L.; 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.
Hanania, N. A.; Tarlo, S. M.; Silverman, F.; Urch, B.; Senathirajah, N.; Zamel, N.; Corey, P. (1998) Effect of
exposure to low levels of ozone on the response to inhaled allergen in allergic asthmatic patients. Chest
114:752-756.
Hazbun, M. E.; Hamilton, R.; Holian, A.; Eschenbacher, W. L. (1993) Ozone-induced increases in substance P and
8-epi-prostaglandin F2a in the airways of human subjects. Am. J. Respir. CellMol. Biol. 9: 568-572.
Hazucha, M. J. (1987) Relationship between ozone exposure and pulmonary function changes. J. Appl. Physiol.
62: 1671-1680.
Hazucha, M. J. (1993) Meta-analysis and "effective dose" revisited. In: Mohr, U., ed. Proceedings of the 3rd
International Inhalation Symposium on Advances in Controlled Clinical Inhalation Studies; October, 1991.
Berlin, Germany: Springer-Verlag; pp. 247-256.
Hazucha, M. J.; Sant'Ambrogio, G. (1993) Effects of ozone on the activity of slowly (SAR) and rapidly adapting
(RAR) receptors in cats. FASEB J. 7: 407A.
AX6-147
-------
Hazucha, M; Silverman, F.; Parent, C.; Field, S.; Bates, D. V. (1973) Pulmonary function in man after short-term
exposure to ozone. Arch. Environ. Health 27: 183-188.
Hazucha, M. J.; Bates, D. V.; Bromberg, P. A. (1989) Mechanism of action of ozone on the human lung. J. Appl.
Physiol. 67: 1535-1541.
Hazucha, M. J.; Folinsbee, L. J.; Seal, E., Jr. (1992) Effects of steady-state and variable ozone concentration profiles
on pulmonary function. Am. Rev. Respir. Dis. 146: 1487-1493.
Hazucha, M. J.; Folinsbee, L. J.; Seal, E.; Bromberg, P. A. (1994) Lung function response of healthy women after
sequential exposures to NO2 and O3. Am. J. Respir. Crit. Care Med. 150: 642-647.
Hazucha, M. J.; Madden, M.; Pape, G.; Becker, S.; Devlin, R.; Koren, H. S.; Kehrl, H.; Bromberg, P. A. (1996)
Effects of cyclo-oxygenase inhibition on ozone-induced respiratory inflammation and lung function changes.
Eur. J. Appl. Physiol. Occup. Med. 73: 17-27.
Hazucha, M. J.; Folinsbee, L. J.; Bromberg, P. A. (2003) Distribution and reproducibility of spirometric response to
ozone by gender and age. J. Appl. Physiol. 95: 1917-1925.
Henderson, F. W.; Dubovi, E. J.; Harder, S.; Seal, E., Jr.; Graham, D. (1988) Experimental rhinovirus infection in
human volunteers exposed to ozone. Am. Rev. Respir. Dis. 137: 1124-1128.
Hetrick, S. M.; Gould, W. D.; Christensen, D. E. (2000) Inflight cabin ozone aboard long duration C-5 airlift
missions: a historical issue revisited. Aviat. Space Environ. Med. 71: 408-414.
Hiltermann, T. J. N.; Stolk, J.; Hiemstra, P. S.; Fokkens, P. H. B.; Rombout, P. J. A.; Sont, J. K.; Sterk, P. J.;
Dijkman, J. H. (1995) Effect of ozone exposure on maximal airway narrowing in non-asthmatic and asthmatic
subjects. Clin. Sci. 89: 619-624.
Hiltermann, T. J. N.; de Bruijne, C. R.; Stolk, J.; Zwinderman, A. H.; Spieksma, F. Th. M.; Roemer, W.;
Steerenberg, P. A.; Fischer, P. H.; vanBree, L.; Hiemstra, P. S. (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, T. J. N.; Peters, E. A.; Alberts, B.; Kwikkers, K.; Borggreven, P. A.; Hiemstra, P. S.; Dijkman, J. H.;
van Bree, L. A.; Stolk, J. (1998) Ozone-induced airway hyperresponsiveness in patients with asthma: role of
neutrophil-derived serine proteinases. Free Radical Biol. Med. 24: 952-958.
Hiltermann, J. T. N.; Lapperre, T. S.; VanBree, L.; Steerenberg, P. A.; Brahim, J. J.; Sont, J. K.; Sterk, P. J.;
Hiemstra, P. S.; 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 Radical
Biol. Med. 27: 1448-1454.
Holtzman, M. J.; Cunningham, J. H.; Sheller, J. R.; Irsigler, G. B.; Nadel, J. A.; Boushey, H. A. (1979) Effect of
ozone on bronchial reactivity in atopic and nonatopic subjects. Am. Rev. Respir. Dis. 120: 1059-1067.
Holz, O.; Jorres, R. A.; Timm, P.; Mticke, M.; Richter, K.; Koschyk, 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.; Mticke, M.; Paasch, K.; Bohme, S.; Timm, P.; Richter, K.; Magnussen, H.; Jorres, R. A. (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, K. J.; Gibson, A.; Vessey, R. S. J.; Janicki, S.; Magnussen, H.;
Jorres, R. A.; 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, D. H.; Folinsbee, L. J.; Ives, P. J.; Abdul-Salaam, S.; McDonnell, W. F. (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. Rev. Respir. Dis. 142: 1158-1163.
Horstman, D. H.; Ball, B. A.; Brown, J.; Gerrity, T.; Folinsbee, L. J. (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, S. M.; Gliner, J. A.; Matsen-Twisdale, J. A. (1979) Pulmonary function and maximum exercise responses
following acute ozone exposure. Aviat. Space Environ. Med. 50: 901-905.
Horvath, S. M.; Gliner, J. A.; Folinsbee, L. J. (1981) Adaptation to ozone: duration of effect. Am. Rev. Respir. Dis.
123:496-499.
Horvath, S. M.; Bedi, J. F.; Drechsler-Parks, D. M. (1986) Effects of peroxyacetyl nitrate alone and in combination
with ozone in healthy young women. J. Air Pollut. Control Assoc. 36: 265-270.
Horvath, S. M.; Folinsbee, L. J.; Bedi, J. F. (1987) Combined effect of ozone and sulfuric acid on pulmonary
function in man. Am. Ind. Hyg. Assoc. J. 48: 94-98.
AX6-148
-------
Horvath, S. M; Bedi, J. F.; Drechsler-Parks, D. M.; Williams, R. E. (1991) Alterations in pulmonary function
parameters during exposure to 80 ppb ozone for 6.6 hours in healthy middle aged individuals. In: Berglund,
R. L.; Lawson, D. R.; McKee, D. J., eds. Tropospheric ozone and the environment: papers from an
international conference; March 1990; Los Angeles, CA. Pittsburgh, PA: Air & Waste Management
Association; pp. 59-70. (A&WMA transactions series: no. TR-19).
Housley, D. G.; Eccles, R.; Richards, R. J. (1996) Gender difference in the concentration of the antioxidant uric acid
in human nasal lavage. Acta Oto-Laryngol. 116: 751-754.
Hu, S.-C.; Ben-Jebria, A.; Ultman, J. S. (1994) Longitudinal distribution of ozone absorption in the lung: effects of
respiratory flow. J. Appl. Physiol. 77: 574-583.
Inkley, S. R.; Maclntyre, W. J. (1973) Dynamic measurement of ventilation-perfusion with xenon-133 at resting
lung volumes. Am. Rev. Respir. Dis. 107: 429-441.
Jenkins, H. S.; Devalia, J. L.; Mister, R. L.; Bevan, A. M.; Rusznak, C.; Davies, R. J. (1999) The effect of exposure
to ozone and nitrogen dioxide on the airway response of atopic asthmatics to inhaled allergen: dose- and
time-dependent effects. Am. J. Respir. Crit. Care Med. 160: 33-39.
Joad, J. P.; Kott, K. S.; Brie, J. M. (1996) The local C-fiber contribution to ozone-induced effects on the isolated
guinea pig lung. Toxicol. Appl. Pharmacol. 141: 561-567.
Johnson, D. A.; Winters, R. S.; Woolley, T.; Graham, D.; Henderson, F. W. (1986) Ozone effects on
alpha-1-proteinase inhibitor in vivo: blood plasma inhibitory activity is unchanged. Exp. Lung Res.
11:95-103.
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, R. A.; Holz, O.; Zachgo, W.; Timm, P.; Koschyk, S.; Mtiller, B.; Grimminger, F.; Seeger, W.; Kelly, F. J.;
Dunster, C.; Frischer, T.; Lubec, G.; Waschewski, M.; Niendorf, A.; Magnussen, H. (2000) The effect of
repeated ozone exposures on inflammatory markers inbronchoalveolar lavage fluid and mucosal biopsies.
Am. J. Respir. Crit. Care Med. 161: 1855-1861.
Kabel, J. R.; Ben-Jebria, A.; Ultman, J. S. (1994) Longitudinal distribution of ozone absorption in the lung:
comparison of nasal and oral quiet breathing. J. Appl. Physiol. 77: 2584-2592.
Kagawa, J. (1986) Experimental studies on human health effects of aerosol and gaseous pollutants. In: Lee, S. D.;
Schneider, T.; Grant, L. D.; Verkerk, P. J., eds. Aerosols: research, risk assessment and control strategies -
proceedings of the second U.S.-Dutch international symposium; May 1985; Williamsburg, VA. Chelsea, MI:
Lewis Publishers, Inc.; pp. 683-697.
Kaneko, K.; Milic-Emili, J.; Dolovich, M. B.; Dawson, A.; Bates, D. V. (1966) Regional distribution of ventilation
and perfusion as a function of body position. J. Appl. Physiol. 21: 767-777'.
Kehrl, H. R.; Vincent, L. M.; Kowalsky, R. J.; Horstman, D. H.; O'Neil, J. J.; McCartney, W. H.; Bromberg, P. A.
(1987) Ozone exposure increases respiratory epithelial permeability in humans. Am. Rev. Respir. Dis.
135: 1124-1128.
Kehrl, H. R.; Vincent, L. M.; Kowalsky, R. J.; Horstman, D. H.; O'Neil, J. J.; McCartney, W. H.; Bromberg, P. A.
(1989) Ozone-induced changes in the pulmonary clearance of 99mTc-DTPA in man. In: Schneider, T.; Lee, S.
D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research and its policy implications: proceedings
of the 3rd US-Dutch international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The
Netherlands: Elsevier Science Publishers; pp. 343-351. (Studies in environmental science 35).
Kehrl, H. R.; Peden, D. B.; Ball, B. A.; Folinsbee, L. J.; Horstman, D. H. (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.
King, T. K. C.; Briscoe, W. A. (1968) The distribution of ventilation, perfusion, lung volume and transfer factor
(diffusing capacity) in patients with obstructive lung disease. Clin. Sci. 35: 153-170.
Kleinman, M. T.; Mautz, W. J.; Bjarnason, S. (1999) Adaptive and non-adaptive responses in rats exposed to ozone,
alone and in mixtures, with acidic aerosols. Inhalation Toxicol. 11: 249-264.
Koenig, J. Q.; Covert, D. S.; Marshall, S. G.; Van Belle, G.; Pierson, W. E. (1987) The effects of ozone and nitrogen
dioxide on pulmonary function in healthy and in asthmatic adolescents. Am. Rev. Respir. Dis. 136:
1152-1157.
Koenig, J. Q.; Covert, D. S.; Smith, M. S.; Van Belle, G.; Pierson, W. E. (1988) The pulmonary effects of ozone and
nitrogen dioxide alone and combined in healthy and asthmatic adolescent subjects. Toxicol. Ind. Health
4: 521-532.
Koenig, J. Q.; Covert, D. S.; Hanley, Q. S.; Van Belle, G.; Pierson, W. E. (1990) Prior exposure to ozone potentiates
subsequent response to sulfur dioxide in adolescent asthmatic subjects. Am. Rev. Respir. Dis. 141: 377-380.
AX6-149
-------
Koenig, J. Q.; Covert, D. S.; Pierson, W. E.; Hanley, Q. S.; Rebolledo, V.; Dumler, K.; McKinney, S. E. (1994)
Oxidant and acid aerosol exposure in healthy subjects and subjects with asthma. Part I: effects of oxidants,
combined with sulfuric or nitric acid, on the pulmonary function of adolescents with asthma. Cambridge, MA:
Health Effects Institute; pp. 1-36; research report no. 70.
Konig, G.; Rommelt, H.; Kienle, H.; Dirnagl, K.; Polke, H.; Fruhmann, G. (1980) Anderung der
bronchomotorischen Reagibilitat des Menschen durch Einwirkung von Ozon [Changes in the bronchial
reactivity of humans caused by the influence of ozone]. Arbeitsmed. Sozialmed. Praeventivmed.
151:261-263.
Koren, H. S.; Devlin, R. B.; Graham, D. E.; Mann, R.; McGee, M. P.; Horstman, D. H.; Kozumbo, W. J.; Becker, S.;
House, D. E.; McDonnell, W. F.; Bromberg, P. A. (1989a) Ozone-induced inflammation in the lower airways
of human subjects. Am. Rev. Respir. Dis. 139: 407-415.
Koren, H. S.; Devlin, R. B.; Graham, D. E.; Mann, R.; McDonnell, W. F. (1989b) The inflammatory response in
human lung exposed to ambient levels of ozone. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D.,
eds. Atmospheric ozone research and its policy implications: proceedings of the 3rd US-Dutch international
symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier Science
Publishers; pp. 745-753. (Studies in environmental science 35).
Koren, H. S.; Hatch, G. E.; Graham, D. E. (1990) Nasal lavage as a tool in assessing acute inflammation in response
to inhaled pollutants. Toxicology 60: 15-25.
Koren, H. S.; Devlin, R. B.; Becker, S.; Perez, R.; McDonnell, W. F. (1991) Time-dependent changes of markers
associated with inflammation in the lungs of humans exposed to ambient levels of ozone. Toxicol. Pathol.
19:406-411.
Kozumbo, W. J.; Hanley, N. M.; Agarwas, S.; Thomas, M. J.; Madden, M. C. (1996) Products of ozonized
arachidonic acid potentiate the formation of DNA single strand breaks in cultured human lung cells. Environ.
Mol. Mutagen. 27: 185-195.
Kreit, J. W.; Gross, K. B.; Moore, T. B.; Lorenzen, T. J.; D'Arcy, J.; Eschenbacher, W. L. (1989) Ozone-induced
changes in pulmonary function and bronchial responsiveness in asthmatics. J. Appl. Physiol. 66: 217-222.
Krishna, M. T.; Springall, D.; Meng, Q.-H.; Withers, N.; Macleod, D.; Biscione, G.; Frew, A.; Polak, J.; Holgate, S.
(1997a) Effects of ozone on epithelium and sensory nerves in the bronchial mucosa of healthy humans. Am. J.
Respir. Crit. Care Med. 156: 943-950.
Krishna, M. T.; Blomberg, A.; Biscione, G. L.; Kelly, F.; Sandstrom, T.; Frew, A.; Holgate, S. (1997b) Short-term
ozone exposure upregulates P-selectin in normal human airways. Am. J. Respir. Crit. Care Med. 155:
1798-1803.
Krishna, M. T.; Madden, J.; Teran, L. M.; Biscione, G. L.; Lau, L. C. K.; Withers, N. J.; Sandstrom, T.; Mudway, L;
Kelly, F. J.; Walls, A.; Frew, A. J.; Holgate, S. T. (1998) Effects of 0.2 ppm ozone onbiomarkers of
inflammation in bronchoalveolar lavage fluid and bronchial mucosa of healthy subjects. Eur. Respir. J.
11: 1294-1300.
Kronenberg, R. S.; Drage, C. W.; Ponto, R. A.; Williams, L. E. (1973) The effect of age on the distribution of
ventilation and perfusion in the lung. Am. Rev. Respir. Dis. 108: 576-586.
Kulle, T. J.; Sauder, L. R.; Kerr, H. D.; Farrell, B. P.; Bermel, M. S.; Smith, D. M. (1982) Duration of pulmonary
function adaptation to ozone in humans. Am. Ind. Hyg. Assoc. J. 43: 832-837.
Kulle, T. J.; Sauder, L. R.; Hebel, J. R.; Chatham, M. D. (1985) Ozone response relationships in healthy
nonsmokers. Am. Rev. Respir. Dis. 132: 36-41.
Lang, D. S.; Torres, R. A.; Mticke, M.; Siegfried, W.; Magnussen, H. (1998) Interactions between human
bronchoepithelial cells and lung fibroblasts after ozone exposure in vitro. Toxicol. Lett. 96-97: 13-24.
Larsen, R. I.; McDonnell, W. F.; Horstman, D. H.; Folinsbee, L. J. (1991) An air quality data analysis system for
interrelating effects, standards, and needed source reductions: part 11. a lognormal model relating human lung
function decrease to O3 exposure. J. Air Waste Manage. Assoc. 41: 455-459.
Lategola, M. T.; Melton, C. E.; Higgins, E. A. (1980) Effects of ozone on symptoms and cardiopulmonary function
in a flight attendant surrogate population. Aviat. Space Environ. Med. 51: 237-246.
Lategola, M. T.; Melton, C. E.; Higgins, E. A. (1980) Pulmonary and symptom threshold effects of ozone in airline
passenger and cockpit crew surrogates. Aviat. Space Environ. Med. 51: 878-884.
Lefohn, A. S.; Foley, J. K. (1993) Establishing relevant ozone standards to protect vegetation and human health:
exposure/dose-response considerations. Air Waste 43: 106-112.
Linder, J.; Herren, D.; Monn, C.; Wanner, H.-U. (1988) Die Wirkung von Ozon auf die korperliche
Leistungsfahigkeit [The effect of ozone on physical activity]. Schweiz Z. Sportmed. 36: 5-10.
AX6-150
-------
Linn, W. S.; Buckley, R. D.; Spier, C. E.; Blessey, R. L.; Jones, M. P.; Fischer, D. A.; Hackney, J. D. (1978) Health
effects of ozone exposure in asthmatics. Am. Rev. Respir. Dis. 117: 835-843.
Linn, W. S.; Jones, M. P.; Bachmayer, E. A.; Spier, C. E.; Mazur, S. F.; Avol, E. L.; Hackney, J. D. (1980)
Short-term respiratory effects of polluted ambient air: a laboratory study of volunteers in a high-oxidant
community. Am. Rev. Respir. Dis. 121: 243-252.
Linn, W. S.; Fischer, D. A.; Medway, D. A.; Anzar, U. T.; Spier, C. E.; Valencia, L. M.; Venet, T. G.; Hackney,
J. D. (1982) 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, W. S.; Avol, E. L.; Hackney, J. D. (1983) Effects of ambient oxidant pollutants on humans: a movable
environmental chamber study. In: Lee, S. D.; Mustafa, M. G.; Mehlman, M. A., eds. International symposium
on the biomedical effects of ozone and related photochemical oxidants; March 1982; Pinehurst, NC.
Princeton, NJ: Princeton Scientific Publishers, Inc.; pp. 125-137. (Advances in modern environmental
toxicology: v. 5).
Linn, W. S.; Avol, E. L.; Shamoo, D. A.; Spier, C. E.; Valencia, L. M.; Venet, T. G.; Fischer, D. A.; Hackney, J. D.
(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, W. S.; Avol, E. L.; Shamoo, D. A.; Peng, R.-C.; Valencia, L. M.; Little, D. E.; Hackney, J. D. (1988) Repeated
laboratory ozone exposures of volunteer Los Angeles residents: an apparent seasonal variation in response.
Toxicol. Ind. Health 4: 505-520.
Linn, W. S.; Shamoo, D. A.; Anderson, K. R.; Peng, R.-C.; Avol, E. L.; Hackney, J. D. (1994) Effects of prolonged,
repeated exposure to ozone, sulfuric acid, and their combination in healthy and asthmatic volunteers. Am. J.
Respir. Crit. Care Med. 150: 431-440.
Linn, W. S.; Anderson, K. R.; Shamoo, D. A.; Edwards, S. A.; Webb, T. L.; Hackney, J. D.; Gong, H., Jr. (1995)
Controlled exposures of young asthmatics to mixed oxidant gases and acid aerosol. Am. J. Respir. Crit. Care
Med. 152: 885-891.
Linn, W. S.; Gong, H., Jr.; Shamoo, D. A.; Anderson, K. R.; Avol, E. L. (1997) Chamber exposures of children to
mixed ozone, sulfur dioxide, and sulfuric acid. Arch. Environ. Health 52: 179-187.
Liu, L.; Leech, J. A.; Urch, R. B.; Silverman, F. S. (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, J. A.; Urch, R. B.; Poon, R.; Zimmerman, B.; Kubay, J. M.; Silverman, F. S. (1999) A comparison of
biomarkers of ozone exposure in human plasma, nasal lavage, and sputum. Inhalation Toxicol. 11: 657-674.
MacNee, W.; Morrison, D.; Rahman, I.; Li, X. Y.; Donaldson, K. (1996) Cigarette smoke and ozone-induced
epithelial perturbation in vivo and in vitro: the role of glutathione. Chest 109(suppl.): 39S.
Magnussen, H.; Nowak, D.; Jorres, R. (1994) Experimentelle Schadstoffeffekte bei Gesundenund Asthmatikern
[Experimental effects of air pollutants in healthy and asthmatic subjects]. Pneumologie 48: 85-88.
Marthan, R.; Roux, E.; Savineau, J.-P. (1996) Human bronchial smooth muscle responsiveness after in vitro
exposure to oxidizing pollutants. CellBiol. Toxicol. 12: 245-249.
Martin, T. R.; Castile, R. G.; Fredberg, J. J.; Wohl, M. E. B.; Mead, J. (1987) Airway size is related to sex but not
lung size in normal adults. J. Appl. Physiol. 63: 2042-2047.
McBride, D. E.; Koenig, J. Q.; Luchtel, D. L.; Williams, P. V; Henderson, W. R., Jr. (1994) Inflammatory effects of
ozone in the upper airways of subjects with asthma. Am. J. Respir. Crit. Care Med. 149: 1192-1197.
McDonnell, W. F. (1996) Individual variability in human lung function responses to ozone exposure. Environ.
Toxicol. Pharmacol. 2: 171-175.
McDonnell, W. F.; Smith, M. V. (1994) Description of acute ozone response as a function of exposure rate and total
inhaled dose. J. Appl. Physiol. 76: 2776-2784.
McDonnell, W. F.; Horstman, D. H.; Hazucha, M. J.; Seal, E., Jr.; Haak, E. D.; Salaam, S. A.; House, D. E. (1983)
Pulmonary effects of ozone exposure during exercise: dose-response characteristics. J. Appl. Physiol.: Respir.
Environ. Exercise Physiol. 54: 1345-1352.
McDonnell, W. F., Ill; Horstman, D. H.; Abdul-Salaam, S.; House, D. E. (1985a) Reproducibility of individual
responses to ozone exposure. Am. Rev. Respir. Dis. 131: 36-40.
McDonnell, W. F., Ill; Chapman, R. S.; Leigh, M. W.; Strope, G. L.; Collier, A. M. (1985b) Respiratory responses
of vigorously exercising children to 0.12 ppm ozone exposure. Am. Rev. Respir. Dis. 132: 875-879.
McDonnell, W. F.; Horstman, D. H.; Abdul-Salaam, S.; Raggio, L. J.; Green, J. A. (1987) The respiratory responses
of subjects with allergic rhinitis to ozone exposure and their relationship to nonspecific airway reactivity.
Toxicol. Ind. Health 3: 507-517.
AX6-151
-------
McDonnell, W. F.; Kehrl, H. R.; Abdul-Salaam, S.; Ives, P. I; Folinsbee, L. I; Devlin, R. B.; O'Neil, J. I;
Horstman, D. H. (1991) Respiratory response of humans exposed to low levels of ozone for 6.6 hours. Arch.
Environ. Health 46: 145-150.
McDonnell, W. F.; Muller, K. E.; Bromberg, P. A.; Shy, C. M. (1993) Predictors of individual differences in acute
response to ozone exposure. Am. Rev. Respir. Dis. 147: 818-825.
McDonnell, W. F.; Stewart, P. W.; Andreoni, S.; Seal, E., Jr.; Kehrl, H. R.; Horstman, D. H.; Folinsbee, L. J.; Smith,
M. V. (1997) Prediction of ozone-induced FEVj changes: effects of concentration, duration, and ventilation.
Am. J. Respir. Crit. Care Med. 156: 715-722.
McDonnell, W. F.; Stewart, P. W.; Smith, M. V.; Pan, W. K.; Pan, J. (1999) Ozone-induced respiratory symptoms:
exposure-response models and association with lung function. Eur. Respir. J. 14: 845-853.
McGee, M. P.; Devlin, R.; Saluta, G.; Koren, H. (1990) Tissue factor and factor VII messenger RNAs in human
alveolar macrophages: effects of breathing ozone. Blood 75: 122-127.
McKinnon, K. P.; Madden, M. C.; Noah, T. L.; Devlin, R. B. (1993) In vitro ozone exposure increases release of
arachidonic acid products from a human bronchial epithelial cell line. Toxicol. Appl. Pharmacol.
118:215-223.
Melton, C. E. (1990) Airliner cabin ozone: an updated review. Washington, DC: Federal Aviation Administration,
Office of Aviation Medicine; report no. DOT/FAA/AM-89/13. Available from: NTIS, Springfield, VA;
AD-A219 264.
Messineo, T. D.; Adams, W. C. (1990) Ozone inhalation effects in females varying widely in lung size: comparison
with males. J. Appl. Physiol. 69: 96-103.
Michelson, P. H.; Dailey, L.; Devlin, R. B.; Peden, D. B. (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.
Mohammed, S. P.; Higenbottam, T. W.; Adcock, J. J. (1993) Effects of aerosol-applied capsaicin, histamine and
prostaglandin-E2 on airway sensory receptors of anaesthetized cats. J. Physiol. Lond. 46: 951-966.
Molfino, N. A.; Wright, S. C.; Katz, L; Tarlo, S.; Silverman, F.; McClean, P. A.; Szalai, J. P.; Raizenne, M.; Slutsky,
A. S.; Zamel, N. (1991) Effect of low concentrations of ozone on inhaled allergen responses in asthmatic
subjects. Lancet 338(8761): 199-203.
Molfino, N. A.; Slutsky, A. S.; Zamel, N. (1992) The effects of air pollution on allergic bronchial responsiveness.
Clin. Exp. Allergy 22: 667-672.
Montuschi, P.; Nightingale, J. A.; Kharitonov, S. A.; Barnes, P. J. (2002) Ozone-induced increase in exhaled
8-isoprostane in healthy subjects is resistant to inhaled budesonide. Free Radical Biol. Med. 33: 1403-1408.
Mudway, I. S.; Kelly, F. J. (2000) Ozone and the lung: a sensitive issue. Mol. Aspects. Med. 21: 1-48.
Mudway, I. S.; Blomberg, A.; Frew, A. J.; Holgate, S. T.; Sandstrom, T.; Kelly, F. J. (1999) Antioxidant
consumption and repletion kinetics in nasal lavage fluid following exposure of healthy human volunteers to
ozone. Eur. Respir. J. 13: 1429-1438.
Mudway, I. S.; Stenfors, N.; Blomberg, A.; Helleday, R.; Dunster, C.; Marklund, S. L.; Frew, A. J.; Sandstrom, T.;
Kelly, F. J. (2001) Differences in basal airway antioxidant concentrations are not predictive of individual
responsiveness to ozone: a comparison of healthy and mild asthmatic subjects. Free Radical Biol. Med.
31:962-974.
Nagda, N. L.; Fortmann, R. C.; Koontz, M. D.; Baker, S. R.; Ginevan, M. E. (1989) Airliner cabin environment:
contaminant measurements, health risks, and mitigation options. Washington, DC: U.S. Department of
Transportation, Office of the Secretary. Available from: NTIS, Springfield, VA; PB91-159384.
National Institutes of Health. (1997) Guidelines for the diagnosis and management of asthma: expert panel report 2.
Bethesda, MD: U.S. Department of Health and Human Services, National Heart, Lung, and Blood Institute;
publication no. 97-4051. Available: http://www.nhlbi.nih.gov/guidelines/asthma/asthgdln.pdf (11 April 2003).
National Institutes of Health. (2003) Expert panel report: guidelines for the diagnosis and management of asthma.
Update on selected topics 2002. Bethesda, MD: U.S. Department of Health and Human Services, National
Heart, Lung, and Blood Institute; NIH publication no. 02-5074. Available:
http://www.nhlbi.nih.gov/guidelines/asthma/asthmafullrpt.pdf [26 August, 2004].
Neas, L. M.; Dockery, D. W.; Koutrakis, P.; Tollerud, D. J.; Speizer, F. E. (1995) The association of ambient air
pollution with twice daily peak expiratory flow rate measurements in children. Am. J. Epidemiol.
141: 111-122.
Newson, E. J.; Krishna, M. T.; Lau, L. C. K.; Howarth, P. H.; Holgate, S. T.; Frew, A. J. (2000) Effects of
short-term exposure to 0.2 ppm ozone onbiomarkers of inflammation in sputum, exhaled nitric oxide, and
lung function in subjects with mild atopic asthma. J. Occup. Environ. Med. 42: 270-277'.
AX6-152
-------
Nichols, B. G.; Woods, J. S.; Luchtel, D. L.; Corral, I; Koenig, J. Q. (2001) Effects of ozone exposure on nuclear
factor-KB activation and tumor necrosis factor-a expression in human nasal epithelial cells. Toxicol. Sci.
60: 356-362.
Nightingale, J. A.; Rogers, D. F.; Barnes, P. J. (1999) Effect of inhaled ozone on exhaled nitric oxide, pulmonary
function, and induced sputum in normal and asthmatic subjects. Thorax 54: 1061-1069.
Nightingale, J. A.; Rogers, D. F.; Chung, K. F.; Barnes, P. J. (2000) No effect of inhaled budesonide on the response
to inhaled ozone in normal subjects. Am. J. Respir. Crit. Care Med. 161: 479-486.
Niinimaa, V.; Cole, P.; Mintz, S.; Shephard, R. J. (1980) The switching point from nasal to oronasal breathing.
Respir. Physiol. 42: 61-71.
Nikasinovic, L.; Momas, L; Seta, N. (2003) Nasal epithelial and inflammatory response to ozone exposure: a review
of laboratory-based studies published since 1985. J. Toxicol. Environ. Health B 6: 521-568.
Otto-Knapp, R.; Jurgovsky, K.; Schierhorn, K.; Kunkel, G. (2003) Antioxidative enzymes in human nasal mucosa
after exposure to ozone. Possible role of GSTM1 deficiency. Inflamm. Res. 52: 51-55.
Passannante, A. N.; Hazucha, M. J.; Bromberg, P. A.; Seal, E.; Folinsbee, L.; Koch, G. (1998) Nociceptive
mechanisms modulate ozone-induced human lung function decrements. J. Appl. Physiol. 85: 1863-1870.
Peden, D. B. (2001a) Air pollution in asthma: effect of pollutants on airway inflammation. Ann. Allergy Asthma
Immunol. 87(suppl. 3): 12-17.
Peden, D. B. (200 Ib) Effect of pollutants in rhinitis. Curr. Allergy Asthma Rep. 1: 242-246.
Peden, D. B.; Setzer, R. W., Jr.; Devlin, R. B. (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, D. B.; 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.
Peters, E. A.; Hiltermann, J. T.; Stolk, J. (2001) Effect of apocynin on ozone-induced airway hyperresponsiveness to
methacholine in asthmatics. Free Radical Biol. Med. 31: 1442-1447.
Rayman, R. B. (2002) Cabin air quality: an overview. Aviat. Space Environ. Med. 73: 211-215.
Reisenauer, C. S.; Koenig, J. Q.; McManus, M. S.; Smith, M. S.; Kusic, G.; Pierson, W. E. (1988) Pulmonary
response to ozone exposures in healthy individuals aged 55 years or greater. JAPCA 38: 51-55.
Riediker, M.; Monn, C.; Koller, T.; Stahel, W. A.; Wuthrich, B. (2001) Air pollutants enhance rhinoconjunctivitis
symptoms in pollen-allergic individuals. Ann. Allergy Asthma Immunol. 87: 311-318.
Rigas, M. L.; Ben-Jebria, A.; Ultman, J. S. (1997) Longitudinal distribution of ozone absorption in the lung: effects
of nitrogen dioxide, sulfur dioxide, and ozone exposures. Arch. Environ. Health 52: 173-178.
Rombout, P. J. A.; Dormans, J. A. M. A.; Marra, M.; VanEsch, G. J. (1986) Influence of exposure regimen on
nitrogen dioxide-induced morphological changes in the rat lung. Environ. Res. 41: 466-480.
Romieu, L; Sienra-Monge, J. J.; Ramirez-Aguilar, M.; Moreno-Macias, H.; Reyes-Ruiz, N. L; Estela del
Rio-Navarro, B.; Hernandez-Avila, M.; London, S. J. (2004) Genetic polymorphism ofGSTMl and
antioxidant supplementation influence lung function in relation to ozone exposure in asthmatic children in
Mexico City. Thorax 59: 8-10.
Samet, J. M.; Hatch, G. E.; Horstman, D.; Steck-Scott, S.; Arab, L.; Bromberg, P. A.; Levine, M.; McDonnell,
W. F.; Devlin, R. B. (2001) Effect of antioxidant supplementation on ozone-induced lung injury in human
subjects. Am. J. Respir. Crit. Care Med. 164: 819-825.
Santak, B.; Calzia, E.; Radermacher, P. (1998) Ventilation-perfusion relationships: clinical implications. Acta
Anaesthesiol. Scand. 42[suppl. 112]: 18-20.
Savin, W. M.; Adams, W. C. (1979) Effects of ozone inhalation on work performance and VO2max. J. Appl. Physiol.:
Respir. Environ. Exercise Physiol. 46: 309-314.
Scannell, C.; Chen, L.; Aris, R. M.; Tager, I.; Christian, D.; Ferrando, R.; Welch, B.; Kelly, T.; Balmes, J. R. (1996)
Greater ozone-induced inflammatory responses in subjects with asthma. Am. J. Respir. Crit. Care Med.
154: 24-29.
Schelegle, E. S.; Adams, W. C. (1986) Reduced exercise time in competitive simulations consequent to low level
ozone exposure. Med. Sci. Sports Exercise 18: 408-414.
Schelegle, E. S.; Adams, W. C.; Siefkin, A. D. (1987) Indomethacin pretreatment reduces ozone-induced pulmonary
function decrements in human subjects. Am. Rev. Respir. Dis. 136: 1350-1354.
Schelegle, E. S.; Adams, W. C.; Giri, S. N.; Siefkin, A. D. (1989) Acute ozone exposure increases plasma
prostaglandin F2a in ozone-sensitive human subjects. Am. Rev. Respir. Dis. 140: 211-216.
AX6-153
-------
Schelegle, E. S.; Siefkin, A. D.; McDonald, R. J. (1991) Time course of ozone-induced neutrophilia in normal
humans. Am. Rev. Respir. Dis. 143: 1353-1358.
Schelegle, E. S.; Carl, M. L.; Coleridge, H. M; Coleridge, J. C. G.; Green, J. F. (1993) Contribution of vagal
afferents to respiratory reflexes evoked by acute inhalation of ozone in dogs. J. Appl. Physiol. 74: 2338-2344.
Schelegle, E. S.; Eldridge, M. W.; Cross, C. E.; Walby, W. F.; Adams, W. C. (2001) Differential effects of airway
anesthesia on ozone-induced pulmonary responses in human subjects. Am. J. Respir. Crit. Care Med.
163: 1121-1127.
Schierhorn, K.; Zhang, M.; Matthias, C.; Kunkel, G. (1999) Influence of ozone and nitrogen dioxide on histamine
and interleukin formation in a human nasal mucosa culture system. Am. J. Respir. Cell Mol. Biol.
20: 1013-1019.
Schierhorn, K.; Hanf, G.; Fischer, A.; Umland, B.; Olze, H.; Kunkel, G. (2002) Ozone-induced release of
neuropeptides from human nasal mucosa cells. Int. Arch. Allergy Immunol. 129: 145-151.
Schonfeld, B. R.; Adams, W. C.; Schelegle, E. S. (1989) Duration of enhanced responsiveness upon re-exposure to
ozone. Arch. Environ. Health 44: 229-236.
Schwartz, L. W.; Dungworth, D. L.; Mustafa, M. G.; Tarkington, B. K.; Tyler, W. S. (1976) Pulmonary responses of
rats to ambient levels of ozone: effects of 7-day intermittent or continuous exposure. Lab. Invest. 34: 565-578.
Seal, E., Jr.; McDonnell, W. F.; House, D. E.; Salaam, S. A.; Dewitt, P. J.; Butler, S. O.; Green, J.; Raggio, L.
(1993) The pulmonary response of white and black adults to six concentrations of ozone. Am. Rev. Respir.
Dis. 147: 804-810.
Seal, E., Jr.; McDonnell, W. F.; House, D. E. (1996) Effects of age, socioeconomic status, and menstrual cycle on
pulmonary response to ozone. Arch. Environ. Health 51: 132-137.
Sekizawa, S. L; Tsubone, H. (1994) Nasal receptors responding to noxious chemical irritants. Respir. Physiol.
96: 37-48.
Seltzer, J.; Bigby, B. G.; Stulbarg, M.; Holtzman, M. J.; Nadel, J. A.; Ueki, I. F.; Leikauf, G. D.; Goetzl, E. J.;
Boushey, H. A. (1986) O3-induced change in bronchial reactivity to methacholine and airway inflammation in
humans. J. Appl. Physiol. 60: 1321-1326.
Solway, J.; Leff, A. R. (1991) Sensory neuropeptides and airway function. J. Appl. Physiol. 71: 2077-2087.
Spannhake, E. W.; Reddy, S. P. M.; Jacoby, D. B.; Yu, X.-Y.; Saatian, B.; Tian, J. (2002) Synergismbetween
rhinovirus infection and oxidant pollutant exposure enhances airway epithelial cell cytokine production.
Environ. Health Perspect. 110: 665-670.
Spit, B. J.; Bretschneider, F.; Hendriksen, E. G. J.; Kuper, C. F. (1993) Ultrastructure of free nerve endings in
respiratory and squamous epithelium on the rat nasal septum. Cell Tissue Res. 274: 329-335.
Steck-Scott, S.; Arab, L.; Craft, N. E.; Samet, J. M. (2004) Plasma and lung macrophage responsiveness to
carotenoid supplementation and ozone exposure in humans. Eur. J. Clin. Nutr. 1-9.
Stenfors, N.; Pourazar, J.; Blomberg, A.; Krishna, M. T.; Mudway, L; Helleday, R.; Kelly, F. J.; Frew, A. J.;
Sandstrom, T. (2002) Effect of ozone on bronchial mucosal inflammation in asthmatic and healthy subjects.
Respir. Med. 96: 352-358.
Sterk, P. J.; Fabbri, L. M.; Quanjer, P. H.; Cockcroft, D. W.; O'Byrne, P. M.; Anderson, S. D.; Juniper, E. F.; Malo,
J.-L. (1993) Airway responsiveness. Standardized challenge testing with pharmacological, physical and
sensitizing stimuli in adults. Report Working Party Standardization of Lung Function Tests, European
Community for Steel and Coal. Eur. Respir. J. 6(suppl. 16): 53-83.
Stevens, W. H. M.; Adelroth, E.; Wattie, J.; Woolley, M. J.; Ellis, R.; Dahlback, M.; O'Byrne, P. M. (1994) Effect of
inhaled budesonide on ozone-induced airway hyperresponsiveness and bronchoalveolar lavage cells in dogs.
J. Appl. Physiol. 77: 2578-2583.
Superko, H. R.; Adams, W. C.; Daly, P. W. (1984) Effects of ozone inhalation during exercise in selected patients
with heart disease. Am. J. Med. 77: 463-470.
Tepper, J. S.; Costa, D. L.; Lehmann, J. R.; Weber, M. F.; Hatch, G. E. (1989) Unattenuated structural and
biochemical alterations in the rat lung during functional adaptation to ozone. Am. Rev. Respir. Dis.
140: 493-501.
Tepper, J. S.; Wiester, M. J.; Weber, M. F.; Menache, M. G. (1990) Measurements of cardiopulmonary response in
awake rats during acute exposure to near-ambient concentrations of ozone. Fundam. Appl. Toxicol. 10: 7-15.
Torres, A.; Utell, M. J.; Morow, P. E.; Voter, K. Z.; Whitin, J. C.; Cox, C.; Looney, R. J.; Speers, D. M.; Tsai, Y.;
Frampton, M. W. (1997) Airway inflammation in smokers and nonsmokers with varying responsiveness to
ozone. Am. J. Respir. Crit. Care Med. 156: 728-736.
Trenga, C. A.; Koenig, J. Q.; Williams, P. V. (2001) Dietary antioxidants and ozone-induced bronchial
hyperresponsiveness in adults with asthma. Arch. Environ. Health 56: 242-249.
AX6-154
-------
Tsai, B. M; Wang, M.; Pitcher, J. M; Meldrum, K. K.; Meldram, D. R. (2004) Hypoxic pulmonary vasoconstriction
and pulmonary artery tissue cytokine expression are mediated by protein kinase C. Am. J. Physiol. Lung Cell
Mol. Physiol. 287: L1215-L1219.
U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office; report nos. EPA-600/8-84-020aF-eF. 5v. Available from: NTIS, Springfield, VA;
PB87-142949.
U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants.
Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF. 3v.
Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, andPB96-185608. Available:
www.epa.gov/ncea/ozone.htm.
Ultman, J. S.; Ben-Jebria, A.; Arnold, S. F. (2004) Uptake distribution of ozone in human lungs: intersubject
variability in physiologic response. Boston, MA: Health Effects Institute.
Utell, M. J.; Frampton, M. W.; Morrow, P. E.; Cox, C.; Levy, P. C.; Speers, D. M.; Gibb, F. R. (1994) Oxidant and
acid aerosol exposure in healthy subjects and subjects with asthma. Part II: effects of sequential sulfuric acid
and ozone exposures on the pulmonary function of healthy subjects and subjects with asthma. Cambridge,
MA: Health Effects Institute; pp. 37-93; research report no. 70.
Vagaggini, B.; Carnevali, S.; Macchioni, P.; Taccola, M.; Fornai, E.; Bacci, E.; Bartoli, M. L.; Cianchetti, S.; Dente,
F. L.; Di Franco, A.; Giannini, D.; Paggiaro. P. L. (1999) Airway inflammatory response to ozone in subjects
with different asthma severity. Eur. Respir. J. 13: 274-280.
Vagaggini, B.; Taccola, M.; Conti, L; Carnevali, S.; Cianchetti, S.; Bartoli, M. L.; Bacci, E.; Dente, F. L.; Di Franco,
A.; Giannini, D.; Paggiaro, P. L. (2001) Budesonide reduces neutrophilic but not functional airway response
to ozone in mild asthmatics. Am. J. Respir. Crit. Care Med. 164: 2172-2176.
Vagaggini, B.; Taccola, M.; Cianchetti, S.; Carnevali, S.; Bartoli, M. L.; Bacci, E.; Dente, F. L.; Di Franco, A.;
Giannini, D.; Paggiaro, P. L. (2002) Ozone exposure increases eosinophilic airway response induced by
previous allergen challenge. Am. J. Respir. Crit. Care Med. 166: 1073-1077.
Van Bree, L.; Rombout, P. J. A.; Rietjens, I. M. C. M.; Dormans, J. A. M. A.; Marra, M. (1989) Pathobiochemical
effects in rat lung related to episodic ozone exposure. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant,
L. D., eds. Atmospheric ozone research and its policy implications: proceedings of the 3rd US-Dutch
international symposium; May 1988; Nijmegen, The Netherlands. Amsterdam, The Netherlands: Elsevier
Science Publishers; pp. 723-732. (Studies in environmental science 35).
Vender, R. L.; Horstman, D. H.; Mangione, S. (1994) Red blood cell antioxidants in human volunteers exposed to
ozone. Toxicol. Ind. Health 10: 53-58.
Voter, K. Z.; Whitin, J. C.; Torres, A.; Morrow, P. E.; Cox, C.; Tsai, Y.; Utell, M. J.; Frampton, M. W. (2001)
Ozone exposure and the production of reactive oxygen species by bronchoalveolar cells in humans. Inhalation
Toxicol. 13:465-483.
Wang, G.; Umstead, T. M.; Phelps, D. S.; Al-Mondhiry, H.; Floras, J. (2002) The effect of ozone exposure on the
ability of human surfactant protein A variants to stimulate cytokine production. Environ. Health Perspect.
110:79-84.
Wayne, W. S.; Wehrle, P. F.; Carroll, R. E. (1967) Oxidant air pollution and athletic performance. JAMA J. Am.
Med. Assoc. 199: 151-154.
Weinmann, G. G.; Weidenbach-Gerbase, M.; Foster, W. M.; Zacur, H.; Frank, R. (1995) Evidence for
ozone-induced small-airway dysfunction: lack of menstrual-cycle and gender effects. Am. J. Respir. Crit.
Care Med. 152: 988-996.
Weymer, A. R.; Gong, H., Jr.; Lyness, A.; Linn, W. S. (1994) Pre-exposure to ozone does not enhance or produce
exercise-induced asthma. Am. J. Respir. Crit. Care Med. 149: 1413-1419.
Yang, I. A.; Holz, O.; Jorres, R. A.; Magnussen, H.; Barton, S. J.; Rodriguez, S.; Cakebread, J. A.; Holloway, J. W.;
Holgate, S. T. (2005) Association of tumor necrosis factor-a polymorphisms and ozone-induced change in
lung function. Am. J. Respir. Crit. Care Med. 171: 171-176.
Yeadon, M.; Wilkinson, D.; Darley-Usmar, V.; O'Leary, V. J.; Payne, A. N. (1992) Mechanisms contributing to
ozone-induced bronchial hyperreactivity in guinea-pigs. Pulm. Pharmacol. 5: 39-50.
Ying, R. L.; Gross, K. B.; Terzo, T. S.; Eschenbacher, W. L. (1990) Indomethacin does not inhibit the
ozone-induced increase in bronchial responsiveness in human subjects. Am. Rev. Respir. Dis. 142: 817-821.
Yu, M.; Pinkerton, K. E.; Witschi, H. (2002) Short-term exposure to aged and diluted sidestream cigarette smoke
enhances ozone-induced lung injury inB6C3Fl mice. Toxicol. Sci. 65: 99-106.
Zhang, L.-Y.; Levitt, R. C.; Kleeberger, S. R. (1995) Differential susceptibility to ozone-induced airways
hyperreactivity in inbred strains of mice. Exp. Lung Res. 21: 503-518.
AX6-155
-------
ANNEX AX7. EPIDEMIOLOGIC STUDIES OF HUMAN
HEALTH EFFECTS ASSOCIATED WITH AMBIENT
OZONE EXPOSURE
AX7-1
-------
AX7.1 TABLES OF EPIDEMIOLOGIC STUDIES OF HUMAN HEALTH
EFFECTS ASSOCIATED WITH AMBIENT OZONE EXPOSURE
AX7-2
-------
LIST OF STUDIES BY AUTHOR
Page
Abbey et al. (1999) AX7-119
Anderson et al. (1996) AX7-87
Anderson et al. (1997) AX7-61
Anderson et al. (1998) AX7-63
Anderson et al. (2001) AX7-88
Atkinson et al. (1999a) AX7-47
Atkinson et al. (1999b) AX7-63
Atkinson et al. (2001) AX7-61
Avol et al. (1998) AX7-10
Avol et al. (2001) AX7-107
Ballester et al. (2001) AX7-66
Beeson et al. (1998) AX7-119
Bell et al. (2004) AX7-74
Bell et al. (2005) AX7-71
Borja-Aburto et al. (1997) AX7-95
Borja-Aburto et al. (1998) AX7-95
Bourcier et al. (2003) AX7-48
Brauer and Brook (1997) AX7-18
Brauer et al. (1996) AX7-18
Bremner et al. (1999) AX7-87
Burnett et al. (1995) AX7-57
Burnett et al. (1997a) AX7-57
Burnett et al. (1997b) AX7-58
Burnett et al. (1999) AX7-58
Burnett et al. (2001) AX7-58
Calderon-Garciduefias et al. (1995) AX7-115
Cassino et al. (1999) AX7-41
Castellsague et al. (1995) AX7-48
Castillejos et al. (1995) AX7-27
Chan and Wu (2005) AX7-32
Chang et al. (2005) AX7-68
Charpin et al. (1999) AX7-111
Chen et al. (1998) AX7-31
Chen et al. (1999) AX7-32
Chen et al. (2000) AX7-14
Chen et al. (2002) AX7-110
Chew et al. (1999) AX7-51
Chock et al. (2000) AX7-82
Cifuentes et al. (2000) AX7-97
Cuijpers et al. (1994) AX7-25
Dab et al. (1996) AX7-89
De Leon et al. (2003) AX7-82
Delfmo et al. (1996) AX7-13
AX7-3
-------
LIST OF STUDIES BY AUTHOR (cont'd)
Page
Delfmo et al. (1997a) AX7-12
Delfmo et al. (1997b) AX7-45
Delfmo et al. (1998a) AX7-13
Delfmo et al. (1998b) AX7-46
Delfmo et al. (2003) AX7-12
Delfmo et al. (2004) AX7-13
Desqueyroux et al. (2002a) AX7-20
Desqueyroux et al. (2002b) AX7-20
Diaz et al. (1999) AX7-94
Dockery et al. (1992) AX7-78
Dockery et al. (2005) AX7-36
Dominici et al. (2003) AX7-75
Fairley (1999) AX7-77
Fairley (2003) AX7-77
Friedman et al. (2001) AX7-42
Frischer et al. (1993) AX7-22
Frischer et al. (1997) AX7-23
Frischer et al. (1999) AX7-113
Frischer et al. (2001) AX7-113
Fung et al. (2005) AX7-59
Galizia and Kinney (1999) AX7-102
Gamble (1998) AX7-78
Garcia-Aymerich et al. (2000) AX7-92
Gauderman et al. (2000; 2004a,b) AX7-105
Gauderman et al. (2002) AX7-105
Gent et al. (2003) AX7-16
Gielen et al. (1997) AX7-25
Gilliland et al. (2001) AX7-11
Gold et al. (1999) AX7-27
Gold et al. (2000) AX7-35
Gold et al. (2003) AX7-35
Goldberg et al. (2001) AX7-84
Goldberg et al. (2003) AX7-84
Gong et al. (1998b) AX7-109
Goss et al. (2004) AX7-102
Gouveia and Fletcher (2000a) AX7-66
Gouveia and Fletcher (2000b) AX7-96
Gouveia et al. (2004) AX7-115
Greer et al. (1993) AX7-103
Gryparis et al. (2004) AX7-85
Gwynn and Thurston (2001) AX7-56
Gwynn et al. (2000) AX7-56
Ha et al. (2001) AX7-116
AX7-4
-------
LIST OF STUDIES BY AUTHOR (cont'd)
Page
Ha et al. (2003) AX7-100
Hagen et al. (2000) AX7-65
Hajat et al. (1999; 2002) AX7-47
Hedley et al. (2002) AX7-101
Hernandez-Gardufio et al. (1997) AX7-49
Hiltermann et al. (1998) AX7-26
Hoek (2003) AX7-90
Hoek and Brunekreef (1995) AX7-26
Hoek et al. (2000) AX7-90
Hoek et al. (2001) AX7-90
Holguin et al. (2003) AX7-39
Hong et al. (2002) AX7-100
Hoppe et al. (1995a,b) AX7-23
Hoppe et al. (2003) AX7-24
Horak et al. (2002a,b) AX7-114
Huang et al. (2005) AX7-75
Hwang and Chan (2002) AX7-50
Ihorst et al. (2004) AX7-112
Ilabaca et al. (1999) AX7-50
Ito (2003; 2004) AX7-55
Ito and Thurston (1996) AX7-79
Ito et al. (2005) AX7-72
Jaffe et al. (2003) AX7-40
Jalaludin et al. (2000) AX7-30
Jalaludin et al. (2004) AX7-31
Jones et al. (1995) AX7-40
Just et al. (2002) AX7-21
Kim et al. (2004) AX7-98
Kinney and Lippmann (2000) AX7-103
Kinney and Ozkaynak (1991) AX7-76
Kinney et al. (1995) AX7-76
Kinney et al. (1996b) AX7-110
Klemm and Mason (2000) AX7-83
Koken et al. (2003) AX7-54
Kopp et al. (1999) AX7-24
Kopp et al. (2000) AX7-112
Korrick et al. (1998) AX7-17
Kunzli et al. (1997) AX7-108
Kuo et al. (2002) AX7-117
Kwon et al. (2001) AX7-99
Lagerkvist et al. (2004) AX7-21
Le Tertre et al. (2002a) AX7-88
Le Tertre et al. (2002b) AX7-62
AX7-5
-------
LIST OF STUDIES BY AUTHOR (cont'd)
Page
Lee and Schwartz (1999) AX7-99
Lee et al. (1999) AX7-99
Lee et al. (2002) AX7-67
Levy et al. (2005) AX7-73
Liao et al. (2004) AX7-33
Lin et al. (1999) AX7-50
Lin et al. (2003) AX7-59
Lin et al. (2004) AX7-60
Linn et al. (1996) AX7-11
Linn et al. (2000) AX7-53
Lipfert et al. (2000a) AX7-81
Lipfert et al. (2000b; 2003) AX7-118
Lippmann et al. (2000) AX7-55. AX7-80
Loomis et al. (1999) AX7-96
Luginaah et al. (2005) AX7-59
Mann et al. (2002) AX7-52
Martins et al. (2002) AX7-50
McConnell et al. (1999) AX7-105
McConnell et al. (2002) AX7-106
McConnell et al. (2003) AX7-106
McDonnell et al. (1999) AX7-104
Metzger et al. (2004) AX7-43
Moolgavkar (2000) AX7-80
Moolgavkar (2003) AX7-79
Moolgavkar et al. (1995) AX7-81
Moolgavkar et al. (1997) AX7-55
Morgan et al. (1998a) AX7-67
Morgan et al. (1998b) AX7-98
Mortimer et al. (2000) AX7-10
Mortimer et al. (2002) AX7-9
Naeher et al. (1999) AX7-17
Nauenberg and Basu (1999) AX7-53
Neas et al. (1995) AX7-15
Neas et al. (1999) AX7-16
Neidell (2004) AX7-52
Newhouse et al. (2004) AX7-14
Oftedal et al. (2003) AX7-65
Ostro (1995) AX7-77
Ostro (2000) AX7-77
Ostro et al. (1996) AX7-97
Ostro et al. (2001) AX7-12
O'Neill et al. (2004) AX7-96
Palli et al. (2004) AX7-114
AX7-6
-------
LIST OF STUDIES BY AUTHOR (cont'd)
Page
Park et al. (2005) AX7-34
Peel et al. (2005) AX7-43
Pereira et al. (1998) AX7-97
Pereira et al. (2005) AX7-120
Peters et al. (1999a,b) AX7-104
Peters et al. (2000a) AX7-33
Peters et al. (2000b) AX7-91
Peters et al. (2001) AX7-34
Petroeschevsky et al. (2001) AX7-67
Poloniecki et al. (1997) AX7-64
Ponce de Leon et al. (1996) AX7-63
Ponka and Virtanen (1996) AX7-66
Ponka et al. (1998) AX7-91
Pope et al. (2002) AX7-118
Prescott et al. (1998) AX7-64. AX7-88
Ramadour et al. (2000) AX7-111
Rich et al. (2004) AX7-37
Rich et al. (2005) AX7-37
Ritz and Yu (1999) AX7-107
Ritz et al. (2000) AX7-107
Ritz et al. (2002) AX7-108
Roemer and van Wijinen (2001) AX7-90
Romieu et al. (1996) AX7-28
Romieu et al. (1997) AX7-28
Romieu et al. (1998) AX7-29
Romieu et al. (2002) AX7-29
Romieu et al. (2004) AX7-30
Ross et al. (2002) AX7-15
Ruidavets et al. (2005) AX7-38
Saez et al. (1999) AX7-92
Saez et al. (2002) AX7-92
Saldiva et al. (1994) AX7-97
Saldiva et al. (1995) AX7-97
Samet et al. (2000) AX7-75
Sartor et al. (1995) AX7-89
Scarlett et al. (1996) AX7-19
Schindler et al. (2001) AX7-22
Schouten et al. (1996) AX7-65
Schwartz (1996) AX7-54
Schwartz (2005) AX7-76
Schwartz et al. (1996) AX7-55
Schwartz et al. (2005) AX7-35
Sheppard (2003) AX7-53
AX7-7
-------
LIST OF STUDIES BY AUTHOR (cont'd)
Page
Sheppard et al. (1999) AX7-53
Simpson et al. (1997) AX7-98
Stieb et al. (1996) AX7-46
Sunyer and Basagana (2001) AX7-93
Sunyer et al. (1996) AX7-93
Sunyer et al. (1997) AX7-47
Sunyer et al. (2002) AX7-94
Tager et al. (2005) AX7-109
Taggart et al. (1996) AX7-20
Tellez-Rojo et al. (2000) AX7-96
Tenias et al. (1998; 2002) AX7-49
Thompson et al. (2001) AX7-48
Thurston et al. (1997) AX7-17
Tobias et al. (1999) AX7-49
Tolbert et al. (2000) AX7-44
Touloumi et al. (1997) AX7-86
Tsai et al. (2003a) AX7-69
Tsai et al. (2003b) AX7-100
Ulmer et al. (1997) AX7-25
Vedal et al. (2003) AX7-83
Vedal et al. (2004) AX7-38
Verhoeff et al. (1996) AX7-91
Villeneuve et al. (2003) AX7-83
Ward et al. (2002) AX7-19
Weisel et al. (2002) AX7-42. AX7-56
Wilson et al. (2005) AX7-41
Wong et al. (1999a) AX7-69
Wong et al. (1999b) AX7-70
Wong et al. (2001) AX7-101
Wong et al. (2002) AX7-62
Yang et al. (2003) AX7-60
Yang et al. (2004a) AX7-68
Yang et al. (2004b) AX7-100
Zhu et al. (2003) AX7-44
Zmirou et al. (1996) AX7-89
Zmirou et al. (1998) AX7-86
AX7-8
-------
Table AX7-1. Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
X
United States
Mortimer et al. (2002)
Eight urban areas in
the U.S.:
St. Louis, MO;
Chicago, IL;
Detroit, MI;
Cleveland, OH;
Washington, DC;
Baltimore, MD;
East Harlem, NY;
Bronx, NY
Jun-Aug 1993
National Cooperative Inner-
City Asthma Study
(NCICAS) cohort. This panel
study examined 846
asthmatic children aged
4-9 years for O3 exposure
effects on PEF and morning
symptoms using linear mixed
effects models and GEE.
8-h avg O3
(10 a.m.-6p.m.):
48ppb
SD not provided.
Range of medians
across cities
shown in a figure:
Approximately
34 to 58 ppb.
<5%ofdays
exceeded 80 ppb.
PM10, NO2, No associations were seen between
SO2 single or multiday O3 measures and any
evening outcome measure. The effects
of O3 on morning outcomes increased
over several days with the strongest
associations seen for multiday lags.
Joint modeling of O3 with NO2 or SO2
resulted in slightly reduced estimates for
each pollutant. Closed cohort.
Approximately 60% of NCICAS cohort
returned diary, characterizations similar
to entire cohort.
8-h avg O3 (per 15 ppb):
% change in morning PEF:
Lag 1-5:
All areas: -0.59% (-1.05,-0.13)
St. Louis: -0.86% (-2.10, 0.38)
Chicago: -0.62% (-2.41, 1.16)
Detroit: -0.75% (-2.36, 0.86)
Cleveland: -0.62% (-2.23, 0.99)
Washington: -0.54% (-2.02, 0.93)
Baltimore: 0.24% (-0.95,1.43)
EastHarlem: -0.73% (-1.63, 0.17)
Bronx: -0.69% (-1.54, 0.15)
Odds ratios:
Morning symptoms:
Lag 1-4:
All areas: 1.16(1.02,1.30)
St. Louis: 0.82(0.59,1.14)
Chicago: 1.09(0.69,1.72)
Detroit: 1.72(1.12,2.64)
Cleveland: 1.20(0.81,1.79)
Washington: 1.11(0.72,1.72)
Baltimore: 1.19(0.89,1.60)
EastHarlem: 1.22 (0.97, 1.53)
Bronx: 1.23(0.98,1.54)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
X
United States (cont'd)
Mortimer et al. (2000)
Eight urban areas in
the U.S.:
St. Louis, MO;
Chicago, IL;
Detroit, MI;
Cleveland, OH;
Washington, DC;
Baltimore, MD;
East Harlem, NY;
Bronx, NY
Jun-Aug 1993
Avoletal. (1998)
Southern California
communities
Spring-summer 1994
A cohort of 846 asthmatic
children aged 4-9 years
examined for effects of
summer O3 exposure on PEF
and morning symptoms. Two
subgroups were compared:
(1) low birth weight or
premature and (2) normal
birth weight or full-term.
Analysis using GEE and
linear mixed effects models.
Panel study.
Three panels of children (age
10-12 years): (1) asthmatic
(n = 53); (2) wheezy (n = 54);
and (3) healthy (n = 103).
Examined for symptoms,
medication use, outdoor time,
physical activity, and
pulmonary function measures
in relation to O3 exposure, via
logistic regression and GLM.
8-h avg O3
(10 a.m.-6 p.m.):
48ppb
SD not provided.
See Mortimer
et al. (2002).
Stratified analysis
of low and high
03:
Fixed site 1-h
max O3:
Low: < 100 ppb
High: >100ppb
Personal 24-h
avg O3:
Low: < 15.6 ppb
High: >32.4ppb
None Low birth weight and premature
asthmatic children had greater declines
in PEF and higher incidence of morning
symptoms than normal birth weight and
full-term asthmatic children.
None The three groups responded similarly.
Few pulmonary function or symptom
associations. Asthmatic children had
the most trouble breathing, the most
wheezing, and the most inhaler use on
high O3 days in the spring. Ozone levels
were considered too low during the
period of the study. Noncompliance
by subjects may have been a problem.
Other analysis methods may have been
more appropriate.
8-h avg O3 (per 15 ppb):
% change in morning PEF:
Low birth weight:
Lag 1-5: -1.83% (-2.65,-1.01)
Normal birth weight:
Lag 1-5: -0.30% (-0.79, 0.19)
Interaction term for birth weight,
p<0.05
Odds ratios:
Morning symptoms:
Low birth weight:
Lag 1-4: 1.42(1.10,1.82)
Normal birth weight:
Lag 1-4: 1.09(0.95,1.24)
Interaction term for birth weight,
p<0.05
Multiple endpoints analyzed. Few
consistent or statistically significant
responses to O3 exposure reported.
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
X
United States (cont'd)
Gillilandetal. (2001)
12 Southern
California
communities
Jan-Jun 1996
Linnetal. (1996)
Three towns in
California: Rubidoux,
Upland, Torrance
Fall-spring 1992-1993
and 1993-1994
Time-series study of 1,933
4th grade children (age
9-10 years) followed for
school absences. Each
absence classified as illness-
related or not. Among
former, classified into
respiratory or gastrointestinal.
Respiratory absences further
classified into upper or lower.
Pollution measured in central
site in each town. Analysis of
distributed lag effects
controlling for time, day of
week, and temperature in a
Poisson model.
Panel study of 269 school
children, each followed for
morning/afternoon lung
function and symptoms for
1 week in fall, winter, and
spring during their 4th and
5th grade school years.
Personal exposure monitoring
in a subset. Analyzed
afternoon symptoms vs. same
day pollution and morning
symptoms vs. 1-day
lag pollution.
8-h avg O3
(10 a.m.-6 p.m.):
Levels not
reported.
Figure depicts
mean range of
approximately
35 to 55 ppb
across the
12 communities.
PM10,N02
24-h avg O3:
Personal:
5 ppb
SD3
Range 1-16
Central site:
23 ppb
SD12
Range 3-53
PM2 5, NO2
Ozone strongly associated with illness-
related and respiratory absences. PM10
only associated with upper respiratory
absences. Long distributed lag effects
for O3 raise questions about adequacy
of control for seasonal changes.
Central site O3 correlated with personal
exposures, r = 0.61. Ozone effects
observed on lung function but only
significant for FEV[ in one analysis.
No effects on symptoms. Ozone effects
were not robust to NO2 or PM2 5. Power
may have been limited by short followup
within seasons (limiting both person-
days and variability in exposures).
8-h avg O3 (per 20 ppb):
% change in school absences:
All illness:
62.9% (18.4, 124.1)
Nonrespiratory illnesses:
37.3% (5.7, 78.3)
Respiratory illnesses:
82.9% (3.9, 222.0)
Upper respiratory:
45.1% (21.3, 73.7)
Lower respiratory with wet cough:
173.9% (91.3, 292.3)
Change in lung function (per ppb):
FEV; next morning:
-0.26 mL (SE 0.25), p = 0.30
FEV[ afternoon:
-0.18 mL(SE 0.26), p = 0.49
FEV; crossday difference:
-0.58 mL(SE 0.23), p = 0.01
FVC next morning:
-0.21 mL(SE 0.22), p = 0.34
FVC afternoon:
-0.20 mL(SE 0.29), p = 0.48
FVC crossday difference:
-0.25 mL (SE 0.25), p = 0.32
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
United States (cont'd)
Ostroetal. (2001)
Central Los Angeles
and Pasadena, CA
Aug-Octl993
Panel study of 138 African- 1 -h max O3:
American children aged
8-13 years with doctor Los Angeles:
diagno sed asthma requiring 59.5 ppb
medication in past year SD 31.4
followed for daily respiratory
symptoms and medication Pasadena:
use. Lags of 0 to 3 days 95.8 ppb
examined. SD 49.0
PM10, NO2, Correlation between PM10 and O3 was
pollen, mold r = 0.35. Significant O3 effect seen for
extra medication use (above normal use).
No O3 effect on symptoms in expected
direction observed. Inverse association
seen for cough. PM10 effects seen at a
lag of 3 days. Time factors not explicitly
controlled in analysis; may have led to
confounding of O3 effects.
X
1-h max O3 (per 40 ppb):
Odds ratios:
Extra medication use:
Lagl: 1.15(1.12,1.19)
Respiratory symptoms:
Shortness of breath:
Lag 3: 1.01 (0.92,1.10)
Wheeze:
Lag 3: 0.94(0.88,1.00)
Cough:
Lag 3: 0.93(0.87,0.99)
to Delfino et al. (2003)
Los Angeles, CA
Nov 1999-Jan 2000
A panel study of 22 Hispanic
children with asthma aged
10-16 years. Filled out
symptom diaries in relation to
pollutant levels. Analysis
using GEE model.
l-hmaxO3:
25.4 ppb
SD9.6
N02, S02, CO,
volatile
organic
compounds,
PM,n
Support the view that air toxics in
the pollutant mix from traffic may
have adverse effects on asthma in
children.
1-h max O3 (per 14.0 ppb):
Odds ratio:
Symptoms interfering with
daily activities:
LagO: 1.99(1.06,3.72)
Delfino etal. (1997a)
Alpine, CA
May-Augl994
Panel study of 22 asthmatics
aged 9-46 years followed for
respiratory symptoms,
morning-afternoon PEF, and
(32 agonist inhaler use.
Personal O3 measured for
12 h/day using passive
monitors. GLM mixed
model.
Ambient:
12-h avg O3
(8 a.m.-8 p.m.):
64 ppb
SD17
Personal:
12-h avg O3:
(8 a.m.-8p.m.)
18 ppb
SD14
PM10, pollen,
fungi
No O3 effects observed.
No quantitative results for O3.
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
United States (cont'd)
Delfinoetal. (1998a)
Alpine, CA
Aug-Octl995
X Delfino et al. (2004)
7"1 Alpine, CA
u> Aug-Oct 1999, Apr-
Jun 2000
Delfinoetal. (1996)
San Diego, CA
Sep-0ctl993
A panel of 24 asthmatics aged
9-17 years followed for daily
symptoms. Analysis using
GEE model.
Panel study of 19 asthmatic
children (age 9-17 years)
followed daily for 2 weeks to
determine relationship
between air pollutants,
namely PM, and FEVp
Linear mixed model used for
analysis.
Panel study of 12 well-
characterized moderate
asthmatics aged 9-16 years
(7 males, 5 females) followed
over 6 weeks for medication
use and respiratory
symptoms. Allergy measured
at baseline with skin prick
tests. Personal O3
measured with passive badge.
Analysis with GLM mixed
model.
1-hmax O3:
90ppb
SD18
8-h max O3:
62.8 ppb
SD15.1
IQR22.0
PM,,
PM25,PM10,
NO,
Asthma symptoms were significantly
associated with both ambient O3 and
PM10 in single-pollutant models. Ozone
effects generally robust to PM10. Current
day O3 effects strongest in asthmatics not
on anti-inflammatory medication.
Effects of O3 and PM10 were largely
independent. The largest effects for
PM10 were seen for a 5-day distributed
lag. For O3 effects, there were no lag
day effects; current day results showed
the greatest effect.
Significant declines in FEV; associated
with various PM indices (personal,
indoor home, etc.), but not ambient O3
levels.
Ambient: PM2 5, SO42 , No effect of ambient O3 on symptom
1-h max O3: H+, HNO3, score. Personal O3 significant for
68 ppb pollen, fungal symptoms, but effect disappeared when
SD 30 spores confounding day-of-week effect was
controlled with weekend dummy
Ambient: variable. (32 inhaler used among
12-h avg O3: 7 subjects was significantly related to
43 ppb personal O3. Results of this small study
SD 17 suggest the value of personal exposure
data in providing more accurate
Personal: estimates of exposures. However, nearly
12-h avg O3: 50% of personal O3 measurements were
11.6 ppb below limits of detection, diminishing
SD 11.2 value of these data. Pollen and fine
particulate (low levels) were not
associated with any of the outcomes.
1-h max O3 (per 58 ppb):
Odds ratios:
O3 only model:
LagO: 1.54(1.02,2.33)
O3 with PM10 model:
LagO: 1.46(0.93,2.29)
No quantitative results for O3.
Change in (32-agonist inhaler use
(per ppb personal O3):
0.0152 puffs/day (SE 0.0075),
p = 0.04
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
United States (cont'd)
Chen et al. (2000)
Washoe County, NV
1996-1998
X
£
Newhouse et al.
(2004)
Tulsa, OK
Sep-Oct2000
Time-series study of school 1-h max O3: PM10, CO
absenteeism examined among 37.45 ppb
27,793 students (kindergarten SD 13.37
to 6th grade) from 57
elementary schools.
First-order autoregression
models used to assess
relationship between O3 and
school absenteeism after
adjusting for weather, day of
week, month, holidays, and
time trends. Ozone levels
from the current day, and
cumulative lags of 1-14 days,
1-21 days, and 1-28 days
examined.
Panel study of 24 subjects 24-h avg O3: PM25, CO,
aged 9-64 years with 30 ppb SO2, pollen,
physician diagnosis of Range 10-70 fungal spores
asthma. Performed PEF
twice daily (morning and
afternoon), and reported daily
respiratory symptoms and
medication use. Forward
stepwise multiple regression
models and Pearson
correlation analyses.
Multipollutant models were examined.
Ozone concentrations in the preceding
14 days were significantly associated
with school absenteeism for students
in grades 1 through 6, but not those in
kindergarten. Both PM10 and CO
concentrations on the concurrent day
were associated with school absenteeism,
but the estimate for PM10 was a negative
value.
Among ambient air pollutants, O3
seemed to be most significant factor.
Morning PEF values significantly
associated with average and maximum
O3 levels on the previous day. Individual
symptoms, including wheezing,
headache, and fatigue, also significantly
related to average and maximum daily
O3. Multiple regression analyses
produced complex models with different
predictor variables for each symptom.
1-h max O3 (per 50 ppb):
Total absence rate:
O3 with PM10 and CO model:
Lag 1-14: 3.79% (1.04, 6.55)
Pearson correlation coefficient:
Morning PEF:
Mean O3 levels:
Lagl: -0.274, p< 0.05
Maximum O3 levels:
Lagl: -0.289, p< 0.05
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Copollutants
Considered
Findings, Interpretation
Effects
United States (cont'd)
Ross et al. (2002)
East Moline, IL and
nearby communities
May-Octl994
X
Neasetal. (1995)
Uniontown, PA
Summer 1990
Panel study of 59 asthmatics
aged 5-49 years recruited.
19 lost to follow-up, yielding
study population of 40.
Assessment of PEF
and respiratory symptoms.
Analytical methods unclear in
terms of control for
time factors.
Panel study of 83 4th and 5th
grade children reported twice
daily PEF and the presence of
cold, cough, or wheeze.
Relationship to pollutants was
analyzed by an autoregressive
linear regression model/GEE.
The number of hours each
child spent outdoors during
the preceding 12-h period was
evaluated.
41.5ppb
SD 14.2
IQR20
PM10, S02,
NO2, pollen,
fungi
12-h avg O3:
Daytime
(8 a.m.-8p.m.):
50.0 ppb
Overnight
(8p.m.-8 a.m.):
24.5 ppb
SO2, PM10, H+
Saw significant associations between
O3 and both PEF declines and
symptom increases. Most but not all
effects remained after controlling for
temperature, pollen and fungi. Ozone
effect on morning PEF disappeared after
adjusting for temperature. No PM10
effects observed.
Evening cough was associated with O3
levels weighted by hours spent outdoors
during the prior 12 h. A decrease in PEF
was associated with O3 levels weighted
by hours spent outdoors. When particle-
strong acidity was added to the model,
the decrement was decreased and no
longer significant.
8-h max O3 (per 20 ppb):
Change in PEF (L/min):
Morning:
Lag 0-1: -2.29 (-4.26,-0.33)
Afternoon:
LagO: -2.58 (-4.26,-0.89)
Symptom score (on scale of 0-3):
Morning:
Lag 1-3: 0.08(0.03,0.13)
Afternoon:
Lag 1-3: 0.08(0.04,0.12)
12-h avg O3 (per 30 ppb):
Evening PEF:
-2.24 L/min (-4.43,-0.05)
Odds ratio:
Evening cough:
1.36(0.86,2.13)
Concentration weighted by
proportion of time spent outdoors
during prior 12 h:
Evening PEF:
-2.79 L/min(-6.69,-1.12)
Odds ratio:
Evening cough:
2.20(1.02,4.75)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
X
Reference, Study
Location and Period
United States (cont'd)
Neasetal. (1999)
Philadelphia, PA
Jul-Sep 1993
Outcomes and Methods
Panel study of 156 children
aged 6-11 years at two
summer camps followed for
twice-daily PEF. Analysis
using mixed effects models
adjusting for autocorrelated
errors.
Mean O3 Levels
Daytime
12-h avg O3
(9 a.m-9p.m.):
SW camp:
57.5 ppb
IQR19.8
NE camp:
55. 9 ppb
IQR21.9
Copollutants
Considered Findings, Interpretation
H+, SO42~, Some O3 effects detected as well as PM
PM25, PM10, effects. Similar O3-related decrements
PM10_2 5 observed in both morning and afternoon
PEF. Ozone effects not robust to SO42~
in two-pollutant models, whereas SO42~
effects relatively robust to O3.
Effects
12-h avg O3 (per 20 ppb):
Morning and evening PEF:
O3 only models:
LagO:
-1.38 L/min(-2.81, 0.04)
Lag 1-5:
-2. 58 L/min (-4. 81, -0.35)
O3 with SO42~ model:
Gent et al. (2003)
Southern New
England
Apr-Sep2001
Panel study of 271 children
(age <12 years) with active,
doctor-diagnosed asthma
followed over 183 days for
respiratory symptoms. For
analysis, cohort split into two
groups: 130 who used
maintenance medication
during follow-up and
141 who did not, on
assumption that medication
users had more severe
asthma. Logistic regression
analyses performed.
1-hmax O3:
58.6 ppb
SD 19.0
51.3 ppb
SD15.5
Correlation between 1-h max O3 and
daily PM25 was 0.77 during this warm-
season study. Large numbers of
statistical tests performed. Significant
associations between symptoms and O3
seen only in medication users, a
subgroup considered to be more
sensitive. PM25 significant for some
symptoms, but not in two-pollutant
models. Ozone effects generally robust
to PM25. Study limitations include
limited control for meteorological factors
and the post-hoc nature of the population
stratification by medication use.
Lag not specified:
-0.04 L/min
1-h max O3 (per 50 ppb):
Odds ratios:
Regular medication users (n = 130):
Chest tightness:
O3 only model:
Lagl: 1.26(1.00,1.48)
O3 with PM2 5 model:
Lagl: 1.42(1.14,1.78)
Shortness of breath:
O3 only model:
Lagl: 1.22(1.02,1.45)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
United States (cont'd)
Korricketal. (1998)
Mount Washington,
NH
Summers 1991, 1992
Cross-sectional study
evaluating the acute effects of
ambient O3 on pulmonary
function of exercising adults.
530 hikers (age 15-64 years)
monitored before and after
their hike. Analysis using a
general linear regression
model.
Avg of hourly O3 PM25, smoke, With prolonged outdoor exercise low-
during each hike: acidity level exposures to O3 were associated
40 ppb with significant effects on pulmonary
SD 12 function. Hikers with asthma had a
Range 21 -74 4-fold greater responsiveness to exposure
toO3.
% change in lung function
(per 50 ppb O3):
-2.6% (-4.7, -0.4)
FVC: -2.2% (-3. 5, -0.8)
X
Thurston et al. (1997)
Connecticut River
Valley, CT
June 1991, 1992,1993
Panel study of children (age
7-13 years) with moderate-to-
severe asthma followed for
medication use, lung
function, and medical
symptoms at a summer
asthma camp for 1 week in
1991 (n= 52), 1992 (n= 58),
and 1993 (n = 56). Analysis
was conducted using both
Poisson modeling and GLM.
1-hmax O3:
1991: 114.0 ppb
1992: 52.2 ppb
1993: 84.6 ppb
1991-1993:
83.6 ppb
H+, SO42 Ozone was most consistently associated
with acute asthma exacerbation, chest
symptoms, and lung function
decrements. Pollen was poorly
associated with any adverse effect.
Consistent results were obtained between
the aggregate and individual analyses.
1-h max O3 (per 83.6 ppb):
Relative risks:
P2-agonist use: 1.46 (t = 3.57)
Chest symptoms: 1.50 (t = 4.77)
Change in PEF (per ppb):
-0.096 L/min(t= 1.92)
Naeheretal. (1999)
Vinton, VA
Summers 1995, 1996
Panel study evaluated the
relationship between O3 and
daily change in PEF studied
in a sample of 473
nonsmoking women aged
19-43 years who recently
delivered babies. PEF
performed twice daily for a
2-week period. Mixed linear
random coefficient model.
8-h max O3:
53.69 ppb
Range
17.00-87.63
24-h avg O3:
34.87 ppb
Range 8.74-56.63
PM25, PM10, Ozone was the only exposure related to
SO42~, H+ evening PEF with 5-day cumulative lag
exposure showing the greatest effect.
24-h avg O3 (per 30 ppb):
Morning PEF:
Lagl:
-2.13L/min(-4.72,0.46)
Lag 1-3:
-3.60 L/min (-7.66, 0.46)
Evening PEF:
LagO:
-2.49 L/min (-5.02, 0.04)
Lag 1-5:
-7.65 L/min (-13.0,-2.25)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Canada
Braueretal. (1996)
Fraser Valley, British
Columbia, Canada
Jun-Aug 1993
Panel study of 58 berry
pickers aged 10-69 years had
lung function measured
before and after a series
of outdoor work shifts
(average duration = 11 h)
over 59 days. Analysis using
pooled regression with
subject-specific intercepts,
with and without temperature
control.
1-hmax O3:
40.3 ppb
SD 15.2
Work shift O3:
26.0 ppb
SD11.8
PM2 5, SO42~, End shift FEV; and FVC significantly
NO3~, NH4+, diminished in relation to O3 levels.
H+ PM2 5 also related to lung function
declines, but O3 remained significant in
two-pollutant models. Next morning
lung function remained diminished
following high O3 days. Ozone effects
still evident at or below 40 ppb. There
was an overall decline of lung function
of roughly 10% over course of study,
suggesting subchronic effect. Levels of
other pollutants low during study.
Change in lung function (per ppb
l-hmaxO3):
Endshift lung function:
FEVp -3.8mL(SE0.4)
FVC: -5.4mL(SE0.6)
Next morning function:
FEVp -4.5mL(SE0.6)
FVC: -5.2mL(SE0.7)
X
oo
Brauer and Brook
(1997)
Fraser Valley, British
Columbia, Canada
Jun-Aug 1993
Additional analysis of Brauer
et al. (1996) with personal
exposure presented for three
groups, stratified by time
spent outdoors.
Group 1: 25 individuals who
spent most of the day indoors.
Group 2: 25 individuals who
spent much of the day
indoors, but still spent several
daylight hours outdoors.
Group 3: 15 individuals who
spent the entire work day
outdoors.
1-hmax O3:
Ambient:
40 ppb
SD15
Range 13-84
PM2 5, SO42~, Group 1 : 9.0% sampling time (24-h)
NO3~,NH4+, outdoors. Personal to ambient O3 ratio
H+ was 0.28.
Group 2: 25.8% sampling time (24-h)
outdoors. Personal to ambient O3 ratio
was 0.48.
Group 3: 100% sampling time
( 1 1 -h work shift) outdoors. Personal
to ambient O3 ratio was 0.96.
One of the first direct demonstrations
that magnitude of personal exposure to
O3 is related to amount of time spent
outdoors. Further showed that, on
average, outdoor fixed O3 monitors were
representative of day-to-day changes in
O3 exposure experienced by the study
population.
Same outcomes as reported in
Braueretal. (1996).
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Mean O, Levels
Copollutants
Considered
Findings, Interpretation
Effects
Europe
Scarlett et al. (1996)
Surrey, England
Jun-Jul 1994
X
VO
Ward et al. (2002)
Birmingham and
Sandwell, England
Jan-Mar 1997
May-Jul 1997
Panel study examined
154 children aged 7 years in a
primary school next to a
major motorway for O3
exposure effects on PEF0 75,
FVC, and FEV; using
autoregression for %
change in function.
A panel study of 162 children
(age 9 years at time of
enrollment in Sept 1996). 39
of 162 children (24%)
reported wheezing in the past
12 months. Examined
association of ambient acid
species with PEF and
symptoms. Single-day lags of
0 to 3 days and a 7-day
cumulative lag were
investigated. Linear
regression used for PEF and
logistic regression used for
symptoms.
50.7 ppb
SD 24.48
PM10, NO2,
pollen
24 h-avg O3:
Winter:
Median 13.0 ppb
Range 2-33
Summer:
Median 22.0 ppb
Range 10-41
PM10,PM25,
SO2, H+, CL,
HC1, HN03,
NH3, NH4+,
NO3- SO42~
No significant association was seen
between pulmonary function measures
and O3 levels. No pollen effects.
Pollutants levels were generally low,
even in the summer. Significant
associations were noted between
respiratory health outcomes and air
pollutants, but no consistent patterns
were identified. The association between
O3 and PEF was generally negative in the
summer and positive in the winter. More
associations between O3 and symptoms
were observed in the winter. Ozone was
associated with a significant increase in
cough, shortness of breath, and wheeze
during the winter. Results did not
indicate that children with atopy or a
history of recent wheezing were more
susceptible to short-term effects of air
pollutants.
Change in lung function (per ppb O3
weighted by inverse of variance):
Lagl: 0.01 mL (-0.12, 0.13)
FVC:
Lagl: 0.07 mL (-0.09, 0.23)
FEV075/FVC:
Lagl: -0.1% (-5.1,4.8)
24-h avg O3 (per 21.5 ppb for
winter; per 10.2 ppb for summer):
Change in PEF (L/min):
Morning (lag 0-6):
Winter: 17.53(6.56,28.52)
Summer: -5.66 (-11.21,-0.09)
Afternoon (lag 0-6):
Winter: 0.28 (-9.03, 9.79)
Summer: -0.14 (-5.34, 5.04)
Odds ratios:
Symptoms:
Cough (lag 0-6):
Winter: 0.88(0.42,1.81)
Summer: 0.95(0.76,1.19)
Shortness of breath (lag 0-6):
Winter: 2.79(1.56,4.95)
Summer: 1.35(0.95,1.94)
Wheeze (lag 0-6):
Winter: 1.59(0.77,3.31)
Summer: 0.88(1.38,0.57)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
X
to
o
Europe (cont'd)
Taggartetal. (1996)
Runcom and Widnes
in NW England
M-Sep 1993
Desqueyroux et al.
(2002a)
Paris, France
Nov 1995-Nov 1996
Panel study investigated the
relationship of asthmatic
bronchial
hyperresponsiveness and
pulmonary function with
summertime ambient air
pollution among 38 adult
nonsmoking asthmatics (age
18-70 years) using log-linear
models. Analysis limited
to investigation of within
subject variance of the
dependent variables.
Panel study of 60 severe
asthmatics (mean age 55
years) were monitored by
their physicians for asthma
attacks. Asthma attacks were
based on medical data
collected by a pulmonary
physician at time of clinical
examination. Analysis using
GEE.
1-h avg O3:
Maximum
61 |ig/m3
24-h avg O3:
Maximum
24.5 /m3
SO2, NO2,
smoke
8-h avg O3
(10 a.m.-6p.m.):
Summer:
41 ng/m3
SD18
Winter:
11 ng/m3
SD10
PMU
No association found for O3. Changes 24-havg O3 (per 10 |ig/m3):
in bronchial hyperresponsiveness were
found to correlate significantly with % change in bronchial
change in the levels of 24-h mean SO2, hyperresponsiveness:
Lagl: 0.3% (-16.6, 20.6)
Lag 2: 2.6% (-22.1, 34.9)
Significant associations between PM10,
O3, and incident asthma attacks were
found. Low O3 levels raise plausibility
8-havgO3(perlO|ig/m3):
Odds ratio:
Lag 2: 1.20(1.03,1.41)
Desqueyroux et al.
(2002b)
Paris, France
Oct 1995-Nov 1996
Panel study of 39 adult
patients with severe COPD
(mean age 67 years) followed
over 14 months by physicians
for exacerbations.
Logistic regression with GEE,
examining exposure lags
of 0 to 5 days.
8-h avg O3
(10 a.m.-6 p.m.):
Summer:
41 |ig/m3
SD18
Winter:
11 ng/m
SD10
PM10, SO2, 50 COPD exacerbations observed over
NO2 follow-up period. 1-, 2-, and 3-day lag
O3 significantly related to exacerbations.
No other pollutants significant. Low O3
levels raise plausibility and confounding
concerns.
8-h avg O3 (per 10 |ig/m3):
Odds ratio:
Lagl: 1.56(1.05,2.32)
Effects appeared larger among
smokers and those with worse gas
exchange lung function.
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
Europe (cont'd)
Just et al. (2002)
Paris, France
Apr- Jim 1996
X
to
Lagerkvist et al.
(2004)
Brussels, Belgium
May 2002
Panel study of 82 medically
diagnosed asthmatic children
(mean age 10.9 years)
followed for O3 exposure and
PEF, asthmatic attacks,
cough, supplementary use of
(32-agonists, and symptoms of
airway irritation. Analysis by
GEE.
Panel study of 57 children
(mean age 10.8 years)
stratified by swimming pool
attendance. Pulmonary
function test performed and
Clara cell protein levels
measured in blood before and
after light exercise outdoors
for 2 h. Analysis using
Student's t test and Pearson
correlation test. For dose
calculations, O3 levels indoors
assumed to be 50% of the
mean outdoor O3
concentration.
24-h avg O3:
58.9 ug/m3
SD24.5
Range 10.0-121.0
Daytime outdoor
03:
Range 77-116
ug/m3
Exposure dose:
Range 352-914
ug/m3-h
PM10, NO2 In asthmatic children, O3 exposure was
related to the occurrence of asthma
attacks and additional bronchodilator
use. O3 was the only pollutant associated
with changes in lung function, as shown
by an increase in PEF variability and
decrease in PEF.
None Ozone levels did not have any adverse
effect on FEV[ after 2 h of outdoor
exercise. In addition, no significant
differences were observed between Clara
cell protein levels before and after
exercise. A marginally significant
positive correlation between ambient O3
dose and Clara cell protein levels
observed among the nonswimmers,
indicating increased antioxidant activity
following O3 exposure in this group.
The lack of a clear relationship between
Clara cell protein levels and O3 dose may
be attributable to the short period of time
between measurements and diurnal
variability of the protein levels.
24-h avg O3 (per 10 ug/m3):
% change in daily PEF variability:
Lag 0-2: 2.6%, p = 0.05
Odds ratio:
Supplementary use of p2-agonist
on days on which no steroids
were used:
LagO: 1.41 (1.05,1.89)
Pearson correlation:
O3 exposure dose and Clara cell
protein levels in serum:
All subjects (n = 54):
r = 0.17, p = 0.21
Nonswimmers (n = 33):
r=0.34,p = 0.06
Swimmers (n = 21):
r =-0.08, p = 0.74
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study Copollutants
Location and Period Outcomes and Methods Mean O, Levels Considered
Findings, Interpretation
Effects
Europe (cont'd)
Schindleretal. (2001)
Eight communities
of Switzerland
May-Sep 1991
X
to
to
Frischer et al. (1993)
Umkirch, Germany
May-Octl991
A random sample of 3,912
adult never-smokers, aged
18-60 years, examined for
short-term O3-related changes
in lung function. Natural
logarithms of FVC, FEVl5
and FEF25.75 were regressed
against the individual
predictor variables and O3.
Spline functions were used to
control potential trends.
Sensitivity analyses for grass
and pollens, and NO2 and
TSP involved linear
time-trend variables.
Panel study of nasal lavage
repeatedly performed on
44 school children (age 9-11
years) according to protocol
published by Koren et al.
(1990). Samples collected
morning after "low" and
"high" O3 days. Nasal lavage
samples analyzed for
polymorphonuclear leukocyte
counts, albumin, tryptase,
eosinophil cationic protein,
and myeloperoxidase.
Analysis using individual
regression methods.
8 h-avg O3
(10 a.m.-6 p.m.):
90.3 ug/m3
Range 2.9-247.1
Stratified analysis
of '/2-h avg O3 at
3 p.m.:
Low:
<140 ug/m3
High:
>180 ug/m3
NO2, TSP Daily average concentrations of O3 were
associated with daily sample means of
FEV[ and FEF25.75 in this random adult
cross-sectional sample. The associations
between daily O3 levels and daily means
of lung function were smaller in
magnitude than the association between
annual O3 levels in the previous analyses
(Ackermann and Liebrich et al., 1997).
This analytic approach was designed to
filter out long-term components.
Sensitive analyses indicated that major
confounding by uncontrolled effects of
pollen, NO2, and TSP was unlikely.
None Significant higher polymorphonuclear
leukocyte counts after high O3 days.
In children without symptoms of rhinitis,
significantly elevated myeloperoxydase
and eosinophil cationic protein
concentrations detected. Results suggest
that ambient O3 produces an
inflammatory response in the upper
airways of healthy children.
8-havgO3(perlOug/m3):
% change in lung function:
FEVI: -0.51% (-0.88,-0.13)
FVC: -0.24% (-0.59, 0.11)
FEF25.75: -1.04% (-1.85,-0.22)
Children without symptoms of
rhinitis (n = 30):
My eloperoxy dase:
LowO3: median 77.39 ug/L
High O3: median 138.60 ug/L
p < 0.05; Wilcoxon sign rank test
Eosinophilic cationic protein:
LowO3: median 3.49 ug/L
High O3: median 5.39 ug/L
p < 0.05; Wilcoxon sign rank test
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Copollutants
Mean O3 Levels Considered
Findings, Interpretation
Effects
Europe (cont'd)
Frischeretal. (1997)
Umkirch, Germany
May-Octl991
X
to
Hoppeetal. (1995a,b)
Munich, Germany
Apr-Sep 1992-1994
Panel study examined 44
school children aged 9-11
years for ratio of ortho-
tyrosine to /wra-tyrosine in
nasal lavage as a marker of
hydroxyl radical attack.
Nasal lavage performed
according to protocol
published by Koren et al.
(1990). Concomitant lung
function tests performed.
Analysis using individual
regression methods.
Panel study of five study
groups (age 12-95 years):
(1) senior citizens (n = 41);
(2) juvenile asthmatics
(n = 43); (3) forestry workers
(n = 41); (4) athletes (n = 43);
and (5) clerks (n = 40) as a
control group. Examined for
lung function (FVC, FEVj,
PEF) and questions on
irritated airways. Each
subject tested 8 days, 4 days
with elevated or high O3 and
4 days with low O3. Analysis
using Wilcoxon matched
pairs signed rank test and
linear regression.
Stratified analysis None
of '/2-h avg O3 at
3 p.m.:
Low:
<140 ug/m3
High:
>180 ug/m3
!/2-h max O3
(1 p.m.-4 p.m.):
Seniors:
High: 70ppb
Low: 31 ppb
Asthmatics:
High: 74 ppb
Low: 34 ppb
Forestry workers:
High: 64 ppb
Low: 32 ppb
Athletes:
High: 71 ppb
Low: 28 ppb
Clerks:
High: 68 ppb
Low: 15 ppb
None
Ambient O3 was associated with the
generation of hydroxyl radicals in the
upper airways of healthy children and
significant lung function decrements.
However, the ortho/para ratio was not
related to polymorphonuclear leukocyte
counts. Passive smoking was not related
to outcomes.
No indication that senior citizens
represent a risk group in this study.
Senior citizens had the lowest ventilation
rate (mean 10 L/min). Athletes and
clerks experienced significant
decrements in lung function parameters.
Well-medicated juvenile asthmatics have
a trend toward large pulmonary
decrements. Forestry workers were
exposed to motor tool exhaust, which
might be a potential promoting factor.
FEV; (% predicted):
Low: 105.4 (SD 15.6)
High: 103.9 (SD 15.0)
A: 1.5, p = 0.031
Ortho/para ratio:
Low: 0.02 (SD 0.07)
High: 0.18(SD0.16)
A: 0.17, p = 0.0001
!/2-h max O3 (per 100 ppb):
Change in lung function:
Seniors:
FEVj: 0.034 L(SD 0.101)
PEF: 0.006 L/s (SD 0.578)
Asthmatics:
FEVj: -0.210 L (SD 0.281)
PEF: -0.712 L/s (SD 0.134)*
Forestry workers:
FEVI: -0.140 L(SD 0.156)
PEF: -1.154 L/s (SD 0.885)*
Athletes:
FEVI: -0.152 L(SD 0.136)*
PEF: -0.622 L/s (SD 0.589)*
Clerks:
FEVj: -0.158 L (SD 0.114)*
PEF: -0.520 L/s (SD 0.486)*
*p<0.05
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
Europe (cont'd)
Hoppe et al. (2003)
Munich, Germany
Apr-Sep 1992-1995
X
to
Koppetal. (1999)
Two towns in Black
Forest, Germany
Mar-Oct 1994
Three of the same study
groups examined in Hoppe
etal. (1995a,b): asthmatics
(n = 43), athletes (n = 43),
and elderly = 41). One
additional risk group, children
(n = 44), was examined.
Over 80% of the elderly and
asthmatic groups took
medications on a daily basis.
Eye and airway symptoms
were assessed, as were
pulmonary function test.
GLM analyses was
conducted.
Panel study of 170 school
children (median age
9.1 years) followed over
11 time points with nasal
lavage sampling. Subjects
were not sensitive to inhaled
allergens. Nasal lavage
samples analyzed for
eosinophil cationic protein,
albumen, and leukocytes.
Analysis using GEE.
!/2-h max O3
(1 p.m.-4p.m.):
Asthmatics:
High: 66.9ppb
Low: 32.5 ppb
Athletes:
High: 65.9 ppb
Low: 27.2 ppb
Children:
High: 70.4 ppb
Low: 29.8 ppb
Elderly:
High: 66.1 ppb
Low: 30.6 ppb
'/2-h max O3:
Villingen:
64 ug/m3
5th %-95th %
1-140
Freudenstadt:
105 ug/m3
5th %-95th %
45-179
NO,
PM10, N02,
SO,, TSP
Stratified analyses by group did not show
consistent O3 effects at the concentration
level of this study; lack of power may
have made it difficult to detect small O3
effects. Analyses on an individual basis
show clearly different patterns of O3
sensitivity. Ozone responders are
defined as individuals with relevant lung
function changes of at least 10% for
FEVj, FVC, and PEF, and 20% for
sRaw. Most of the responders were
found in the asthmatic and children
groups. In these responders, a significant
O3 concentration-response relationship
was observed in individual regressions
(results not presented). The sample size
may limit quantitative extrapolation to
larger populations, but may allow
cautious first estimates.
Eosinophil cationic protein and
leukocyte levels peaked soon after first
major O3 episode of summer, but did not
show response to later, even higher,
O3 episodes. These observations are
consistent with an adaptive response
in terms of nasal inflammation.
!/2-h max O3 (per 50 ppb):
% change in lung function, lag 0:
Asthmatics:
FEVI: 4.26% (-3.13, 11.66)
PEF: 6.67% (-1.55, 14.89)
Athletes:
FEVj: 0.01% (-0.13, 0.11)
PEF: -0.13% (-0.29, 0.03)
Children:
FEVj: -1.81% (-5.34, 1.73)
PEF: -11.88%(-18.98,-4.78)
Elderly:
FEVI: 2.10% (-4.65, 8.84)
PEF: 7.29% (-2.84, 17.43)
Ozone responders:
Asthmatics: 21%
Athletes: 5%
Children: 18%
Elderly: 5%
Change in log eosinophil cationic
protein concentration (per ug/m3
03):
Early summer:
0.97(0.03,1.92)
Late summer:
-0.43 (-1.34, 0.47)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Copollutants
Outcomes and Methods
Findings, Interpretation
Effects
X
to
Europe (cont'd)
Ulmeretal. (1997)
Freudenstadt and
Villingen, Germany
Mar-Oct 1994
Panel study of 135 children
aged 8-11 years in two towns
were evaluated. Pulmonary
function was associated with
the highest O3 concentration
in the previous 24 h.
An initial cross-sectional
analysis was followed by
a longitudinal analysis
using GEE with the data
at four time periods
(Apr, Jun, Aug, Sep).
Maastricht, the (age unspecified) randomly
Netherlands chosen from 535 reexamined
Nov-Dec 1990 for lung function and
(baseline), symptoms. Corrected
Aug 8-16, 1991 (smog baseline lung function
episode) compared by paired t test.
Difference in prevalence of
respiratory symptoms
examined.
!/2-h max O3:
Freudenstadt:
Median 50.6 ppb
10th%-90th%
22.5-89.7
Villingen:
Median 32.1 ppb
10th%-90th%
0.5-70.1
None
Cuijpers et al. (1994) During episode, 212 children Baseline:
8-h avg O3:
Range 2-56 ug/m3
Smog episode:
l-hmaxO3:
Exceeded
160 ug/m3 on
11 days
PM10, S02,
NO,
In the cross-sectional analysis,
a significant negative association
between O3 exposure and FVC was only
shown at the June testing. ForFEV1;
no significant associations were detected.
In contrast, the longitudinal analysis
obtained a statistically significant
negative correlation between O3
exposure, and FVC and FEV[ for the
subpopulation living in the town with
higher O3 levels, Freudenstadt. The
associations were more pronounced
in males than females.
Small decrements in FEV[ and FEF25.75
were found in the 212 children.
However, significant decreases in
resistance parameters also were noted.
Each day a different group of 30 children
were measured. The results of the lung
function are contradictory in that
spirometry suggest airflow obstruction
while impedance measurement suggest
otherwise. Respiratory symptoms
impacted by low response rate of 122 of
212 children due to summer holidays.
No increase was observed.
Change in lung function (per ug/m3
!/2-h max O3):
Freudenstadt:
-1. 13 mL,p = 0.002
Villingen:
-0.19mL,p = 0.62
FVC:
Freudenstadt:
- 1. 23 mL,p = 0.002
Villingen:
0.02 mL,p = 0.96
Change in lung function and
impedance between baseline and
smog episode:
-0.032 L(SD 0.226), p< 0.05
FEF25.75:
-0.086 L/s (SD 0.415), p < 0.01
Resistence at 8 Hz:
-0.47 cmH20/(L/s) (SD 1.17),
p < 0.05
Gielenetal. (1997)
Amsterdam, the
Netherlands
Apr-Jul 1995
Panel study of 61 children
aged 7-13 years from two
special schools for
chronically ill children,
followed for twice-daily PEF,
symptoms, and medication
usage. 77% of cohort had
doctor-diagnosed asthma.
l-hmaxO3:
77.3 ug/m3
SD15.7
8-h max O3:
67.0 ug/m3
SD 14.9
PM10, BS,
pollen
Morning PEF significantly associated
with 8-h max O3 at a lag of 2 days.
BS also associated with PEF. Among
14 symptom models tested, only one
yielded a significant O3 finding (for
upper respiratory symptoms). PM10
and BS, but not O3, were related to
(32-agonist inhaler use.
8-h max O3 (per 83.2 ug/m3):
% change in PEF:
Morning:
Lag 2: -1.86% (-3.58,-0.14)
Afternoon:
Lag 2: -1.88% (-3.94, 0.18)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Location and Period
Outcomes and Methods
Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
Europe (cont'd)
Hiltermann et al.
(1998)
Bilthoven, the
Netherlands
M-Octl995
X
to
Hoek and Brunekreef
(1995)
Deume and
Enkhuizen, the
Netherlands
Mar-M 1989
Panel study of 60 adult
nonsmoking intermittent to
severe asthmatics (age
18-55 years) followed over
96 days. Measured morning
and afternoon PEF,
respiratory symptoms, and
medication use. Analysis
controlled for time trends,
aeroallergens, environmental
tobacco smoke exposures,
day of week, temperature.
Lags of 0 to 2 days examined.
The occurrence of acute
respiratory symptoms
investigated in children aged
7-11 years (Deurne n = 241;
Enkhuizen n = 59).
Symptoms included cough,
shortness of breath, upper and
lower respiratory symptoms,
throat and eye irritation,
headache and nausea. Ozone-
related symptom prevalence
and incidence were examined.
Lags of 0 and 1 day, and
mean O3 concentration from
previous week were
investigated. Analyses using
1 st-order autoregressive
models and logistic regression
models.
80.1 ug/m3
Range 6-94
PM10, NO2,
SO,, BS
1-hmax O3:
Deume:
57ppb
SD20
Range 22-107
Enkhuizen:
59ppb
SD14
Range 14-114
PM10, NO2,
SO,
Ozone had strongest association with
symptoms of any pollutant analyzed.
PEF lower with O3 but not statistically
significant. No effect on medication use.
No effect modification by steroid use or
hyperresponsiveness.
No consistent association between
ambient O3 concentrations and the
prevalence or incidence of symptoms in
either city. The one significant positive
coefficient in Enkhuizen for prevalence
of upper respiratory symptoms was not
confirmed by the Deurne results.
No associations of daily symptom
prevalence or incidence found with any
of the other copollutants examined.
8-h max O3 (per 100 ug/m3):
Odds ratios:
Respiratory symptoms:
Shortness of breath:
LagO: 1.18(1.02,1.36)
Sleep disturbed by breathing:
LagO: 1.14(0.90,1.45)
Pain on deep inspiration:
LagO: 1.44(1.10,1.88)
Cough of phlegm:
LagO: 0.94(0.83,1.07)
Bronchodilator use:
LagO: 1.05(0.94,1.19)
1-h max O3 (per 50 ppb):
Prevalence of symptoms:
Deume:
Any respiratory symptom:
LagO: -0.06 (SE 0.04)
Cough:
LagO: -0.07 (SE 0.07)
Upper respiratory symptoms:
LagO: -0.06 (SE 0.05)
Enkhuizen:
Any respiratory symptom:
LagO: 0.12 (SE 0.07)
Cough:
LagO: -0.07 (SE 0.18)
Upper respiratory symptoms:
LagO: 0.18 (SE 0.09)*
*p<0.05
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Latin America
Castillejos et al.
(1995)
SW Mexico City
Augl990-Octl991
Gold etal. (1999)
SW Mexico City
1991
X
to
Panel study of children aged
7/2-11 years (22 males, 18
females) tested up to 8 times
for FEV; and FVC, before
and after exercise. Target
minute ventilation was
35 L/min/m2. Analysis using
GEE models.
Panel study of 40 school
children aged 8-11 years in
polluted community followed
for twice-daily PEF and
respiratory symptoms. PEF
deviations in
morning/afternoon from
child-specific means analyzed
in relation to pollution,
temperature, season, and time
trend. Morning symptoms
analyzed by Poission
regression.
1-hmax O3:
112.3ppb
Range 0-365
5th quintile mean
229.1ppb
24-h avg O3:
52.0 ppb
IQR 25
PM10
PM2 5, PM1
The mean % decrements in lung function
were significantly greater than zero only
in the fifth quintile of O3 exposure
(1 83-365 ppb).
Reported significant declines in PEF
in relation to 24-h avg O3 levels.
Effects did not vary by baseline
symptom history. Lags chosen to
maximize effects and varied by outcome.
Ozone generally robust to PM25.
Morning phlegm significantly related to
24-h avg O3 at a 1-day lag.
% change with exercise in
5th quintile of O3 exposure
(183-365 ppb):
j: -2.85% (-4.40, -1.31)
FVC: -1.43% (-2.81, -0.06)
24-h avg O3 (per 25 ppb):
% change in PEF:
Morning:
Lag 1-10: -3.8% (-5.8,-1.8)
Afternoon:
Lag 0-9: -4.6% (-7.0,-2.1)
% change in phlegm:
Morning:
Lagl: 1.1% (1.0, 1.3)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Latin America (cont'd)
Romieuetal. (1996)
N Mexico City
Apr-Jul 1991,
Nov 1991-Feb 1992
X
to
oo
Panel study of 71 mildly
asthmatic children aged
5-13 years followed for PEF
and respiratory symptoms.
Lower respiratory symptoms
included cough, phlegm,
wheeze and/or difficulty
breathing.
1 -h max O3: PM2 5, PM10, Ozone effects observed on both PEF and
190 ppb NO2, SO2 symptoms. Symptom, but not PEF,
SD 80 effects robust to PM10 in two-pollutant
models. Symptoms related to O3
included cough and difficulty breathing.
1-h max O3 (per 50 ppb):
Change in PEF (L/min):
Morning:
Lag 0: -2.44 (-4.40, -0.49)
Lagl: -0.23 (-0.41, 1.62)
Lag 2: -1.49 (-3.80, 0.80)
Afternoon:
LagO: -0.56 (-2.70, 1.60)
Lagl: -1.27 (-3.20, 0.62)
Lag 2: -1.92 (-4.50, 0.66)
Odds ratios:
Lower respiratory symptoms:
LagO: 1.09(1.03,1.15)
Lagl: 1.10(1.04,1.17)
Lag 2: 1.04(0.97,1.12)
Romieuetal. (1997)
SW Mexico City
Apr-Jul 1991,
Nov 1991-Feb 1992
Same period as Romieu et al.,
1996, but in different section
of city. 65 mildly asthmatic
children aged 5-13 years
followed for twice-daily PEF,
and respiratory symptoms.
Up to 2 months follow-up per
child. Analysis included
temperature and looked at
0- to 2-day lags. No time
controls. Lower respiratory
symptoms included cough,
phlegm, wheeze and/or
difficulty breathing. Panel
study.
1-h max O3:
196 ppb
SD78
PM10 Ozone had significant effects on PEF and 1-h max O3 (per 50 ppb):
symptoms, with largest effects at lags 0
and 1 day. Symptoms related to O3 Change in PEF (L/min):
included cough and phlegm. Ozone Morning:
effects stronger than those for PM10. LagO: -1.32 (-3.21, 0.57)
Lagl: -0.39 (-2.24, 1.47)
Lag 2: -0.97 (-2.94, 0.99)
Afternoon:
LagO: -1.81 (-3.60,-0.01)
Lagl: -2.32 (-4.17,-0.47)
Lag 2: -0.21 (-2.44, 2.02)
Odds ratios:
Lower respiratory symptoms:
LagO: 1.11 (1.05,1.19)
Lagl: 1.08(1.01,1.15)
Lag 2: 1.07(1.02,1.13)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
X
to
VO
Latin America (cont'd)
Romieuetal. (1998)
Mexico City
Mar-May 1996
(1st phase)
Jun-Aug 1996
(2nd phase)
Romieu et al. (2002)
Mexico City
Octl998-Apr2000
Panel study of 47 street
workers aged 18-58 years
randomly selected to take a
daily supplement (vitamin C,
vitamin E, and beta carotene)
or placebo during 1 st phase of
study. Following washout
period, the use of
supplements and placebos
was reversed during 2nd
phase. Pulmonary function
test performed twice a week
at end of workday. Plasma
concentrations of beta
carotene and a-tocopherol
measured. Analysis using
GEE models.
Panel study of 158 asthmatic
children aged 6-16 years
randomly given a vitamin (C
and E) supplement or placebo
followed for 12 weeks. Peak
flow was measured twice a
day and spirometry was
performed twice per week in
the morning. Double blind
study. Plasma concentration
of vitamin E levels measured.
Analysis using GEE models.
l-hmaxO3:
123 ppb
SD40
55% of days >110
ppb.
Workday hourly
average during
workday prior to
pulmonary
function test:
67.3 ppb
SD24
PM10, NO2 During the 1 st phase, O3 levels were
significantly associated with declines in
lung function parameters. No
associations were observed in the daily
supplement group. A significant
supplement effect was observed. Ozone-
related decrements also were observed
during the 2nd phase, however the
associations were not significant.
Supplementation with antioxidants
during the 1 st phase may have had a
residual protective effect on the lung.
l-hmaxO3: PM10,NO2 Ozone levels were significantly
102 ppb correlated with decrements in FEF25.75
SD 47 in the placebo group, but not in the
supplement group. When analysis was
restricted to children with moderate-to-
severe asthma, amplitudes of decrements
were larger and significant for FEV^
FEF25.75, and PEF in the placebo group.
Supplementation with antioxidants may
modulate the impact of O3 exposure on
the small airways of children with
moderate to severe asthma.
1-h max O3 (per 10 ppb):
Placebo group:
1st phase:
LagO: - 17.9 mL (SE 5.4)*
FVC:
LagO: - 14.8 mL(SE 7.1)*
2nd phase:
FEVI:
LagO: -3.3mL(SE6.5)
FVC:
LagO: -0.27 mL (SE 7.8)
No significant associations with O3
observed when taking supplements.
1-h max O3 (per 10 ppb):
Children with moderate to severe
asthma:
Placebo group:
O3 with PM10 and NO2 models:
FEVI:
Lagl: -4.59mL,p = 0.04
FEF2.5.75:
Lagl: -13.32mL/s,p<0.01
PEF:
Lagl: - 15.01 mL/s, p = 0.04
No association observed in the
vitamin supplement group.
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Latin America (cont'd)
Romieu et al. (2004)
Mexico City
Octl998-Apr2000
X
Additional analysis of
Romieu et al., 2002 with data
on glutathion S-transferase
Ml polymorphism (GSTM1
null genotype) in
158 asthmatic children.
Analysis performed using
GEE models, stratified by
GSTM1 genotype (null vs.
positive) within the two
treatment groups (placebo
and antioxidant
supplemented). Panel study.
l-hmaxO3:
102ppb
SD47
None In the placebo group, O3 exposure was
significantly and inversely associated
with FEF2 5.75 in children who had the
GSTM1 null genotype, with larger
effects seen in children with moderate-
to-severe asthma. No significant
decrements were seen in the GSTM1
positive children. These results provide
preliminary evidence that asthmatic
children who may be genetically
impaired to handle oxidative stress are
most susceptible to the effect of O3 on
small airways function.
1-h max O3 (per 50 ppb):
FEF25.75 in children with moderate to
severe asthma:
Placebo group:
GSTM1 null:
Lagl: -80.8mL/s,p = 0.002
GSTM1 positive:
Lagl: -34.4 mL/s,p> 0.10
Supplement group:
GSTM1 null:
Lag 1: -7.3 mL/s, p> 0.10
GSTM1 positive:
Lagl: 2.0 mL/s, p> 0.10
Australia
Jalaludin et al. (2000)
Sydney, Australia
Feb-Dec 1994
Panel study of three groups of
children (mean age 9.6
years): (1) n = 45 with
history of wheeze 12 months,
positive histamine challenge,
and doctor-diagnosed asthma;
(2) n = 60 with history of
wheeze and doctor-diagnosed
asthma; (3) n = 20 with only
history of wheeze. Examined
for evening PEF and daily O3
using GEE model and
population regression models.
Mean daytime O3 PM10, NO2 A significant negative association was
(6 a.m.-9 p.m.): found between daily mean deviation in
12 ppb PEF and same-day mean daytime O3
SD 6.8 concentration after adjusting for
copollutants, time trend, meteorological
Maximum variables, pollen count, and Alternaria
daytime O3 count. The association was stronger in
(6 a.m.-9 p.m.): a subgroup of children with bronchial
26 ppb hyper-reactivity and doctor-diagnosed
SD 14.4 asthma. In contrast, the same-day
maximum O3 concentration was not
statistically associated.
Change in PEF (per 10 ppb mean
daytime O3):
All children (n = 125):
O3 only model:
-0.9178 (SE 0.4192), p = 0.03
O3 with PM10 model:
-0.9195 (SE 0.4199), p = 0.03
O3 with PM10 and NO2 model:
-0.8823 (SE 0.4225), p = 0.04
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Australia (cont'd)
Jalaludin et al. (2004)
Sydney, Australia
Feb-Dec 1994
X
Chen etal. (1998)
Six communities in
Taiwan
1994-1995
Same three groups of children
as studied in Jalaludin et al.,
2000. Examined relationship
between O3 and evening
respiratory symptoms
(wheeze, dry cough, and wet
cough), evening asthma
medication use (inhaled
(32-agonist and inhaled
corticosteroids), and doctor
visits for asthma. Analysis
using GEE logistic regression
models. Panel study.
4,697 school children (age
unspecified) from a rural area
(Taihsi), urban areas
(Keelung and Sanchung), and
petrochemical industrial areas
(Jenwu, Linyuan, and Toufen)
cross-sectionally examined
for respiratory symptoms and
diseases using parent-
completed questionnaires.
Multiple logistic regression
models used to compare odds
of symptoms and diseases in
urban or petrochemical areas
to the rural area after
controlling for potential
confounding factors.
Cross-sectional panel study.
Mean daytime O3 PM10, NO2 No significant O3 effects observed on
(6 a.m.-9 p.m.): evening symptoms, evening asthma
12 ppb medication use, and doctors visits.
SD 6.8 Also, no differences in the response of
children in the three groups. A potential
Maximum limitation is that the use of evening
daytime O3 outcome measures rather than morning
(6 a.m.-9 p.m.): values may have obscured the effect of
26 ppb ambient air pollutants. Only consistent
SD 14.4 relationship was found between mean
daytime PM10 concentrations and doctor
visits for asthma.
24-h avg O3:
Rural area:
52.56 ppb
Urban area:
Mean range
38.34-41.90 ppb
Petrochemical
industrial area:
Mean range
52.12-60.64 ppb
SO2, CO,
PM10,N02
School children in urban communities,
but not in petrochemical industrial areas,
had significantly more respiratory
symptoms and diseases vs. those living
in the rural community. However, mean
O3 levels in the urban communities were
lower than that of the rural community.
No causal relationship could be derived
between O3 and respiratory symptoms
and diseases in this cross-sectional study.
Mean daytime O3 (per 8.3 ppb):
Odds ratios:
All children (n = 125):
Wheeze:
Lagl: 1.00(0.93,1.08)
Dry cough:
Lagl: 1.03(0.96,1.11)
Wet cough:
Lagl: 0.97(0.92,1.03)
Inhaled p2-agonist use:
Lagl: 1.02(0.97,1.07)
Inhaled corticosteroid use:
Lagl: 1.02(0.99,1.04)
Doctor visit for asthma:
Lagl: 1.05(0.77,1.43)
Urban areas vs. rural area:
Odds ratios:
Respiratory symptoms:
Morning cough:
1.33(0.98,1.80)
Day or night cough:
1.67(1.21,2.29)
Shortness of breath:
1.40(1.04,1.91)
Wheezing or asthma:
1.68(1.11,2.54)
-------
Table AX7-1 (cont'd). Effects of Acute O3 Exposure on Lung Function and Respiratory Symptoms in Field Studies
Reference, Study
Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Asia
X
to
Chen etal. (1999)
Three towns in
Taiwan: Sanchun,
Taihsi, Linyuan
May 1995-Jan 1996
Chan and Wu (2005)
Taichung City,
Taiwan
Sep2001
(Questionnaire
survey)
Nov-Dec 2001
(Field study)
Valid lung function data
obtained once on each of
895 children (age 8-13 years)
in three towns. Examined
relation between lung
function and pollution
concentrations on same day
and over previous 1, 2, and
7 days. Multipollutant
models examined.
Cross-sectional panel study.
A cohort of mail carriers
(mean age 39 years)
examined for effects of O3
exposure on PEF. Their
exposure periods were
between 9 a.m. and 5 p.m.
every working day. PEF
measurements taken twice
daily. Single-day lags from
0 to 3 days examined. A two-
step statistical model was
used, a multiple linear
regression without air
pollutants followed by a
linear mixed effects model to
estimate pollution effects.
l-hmaxO3:
Range 19.7-110.3
ppb
SD not provided.
S02, CO,
PM10,N02
FEV; and FVC significantly associated Change in lung function:
with day 1-day lag O3. FVC also related
8-h avg O3
(9 a.m.-5 p.m.):
35.6 ppb
SD12.1
Range 7.6-65.1
PM10,N02
to NO2, SO2, and CO. No PM10 effects
observed. Only O3 remained significant
in multipollutant models. No PM10
effects. A significant O3 effect was not
evident at O3 levels below 60 ppb.
Significant associations observed
between evening PEF and O3
concentrations at lags of 0, 1 and 2 days.
Largest effect observed at a day 1-day
lag. Similar O3 effects on morning PEF
also noted (data not presented). Neither
PM10 nor NO2 showed consistent
associations with PEF. Ozone results
were robust to adjustment for PM10 and
O3 only models:
Lagl:
FEVI: -0.64 mL/ppb(SE 0.34)*
FVC: -0.79 mL/ppb (SE 0.32)*
O3 with NO2 models:
Lagl:
FEVp -0.85 mL/ppb (SE 0.34)*
FVC: -0.91 mL/ppb (SE 0.37)*
*p<0.05
8-h avg O3 (per 10 ppb):
% change in evening PEF:
LagO: 0.54%,p<0.05
Lagl: 0.69%, p< 0.05
Lag 2: 0.52%,p<0.05
-------
Table AX7-2. Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study
Location and Period
Outcomes and Methods Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
X
United States
Liao et al. (2004)
Three locations in U.S.:
Minneapolis, MN;
Jackson, MS; Forsyth
County, NC
1996-1998
Cross-sectional study of
5,431 cohort members of
the Atherosclerosis Risk in
Communities study, men
and women aged 45-64
years at entry in 1987.
Association between O3 and
cardiac autonomic control
assessed using 5-min heart
rate variability indices
collected over a 4-h period.
Analysis using
multivariable linear
regression models, adjusting
for individual
cardiovascular disease risk
factors and meteorological
factors.
8-h avg O3 PM10, CO, Significant interaction between O3 and
(10 a.m.-6 p.m.): SO2, NO2 ethnicity in relation to high-frequency
41 ppb power (p < 0.05). Ambient O3
SD 16 significantly associated with high-
frequency power among whites, but not
blacks. No significant O3 effect on other
heart rate variability indices, including
low-frequency power and SD of normal
R-R intervals. More consistent
relationships observed between PM10
and heart rate variability indices.
8-h avg O3 (per 16 ppb):
Log-transformed high-frequency
power:
White race:
Lagl: -0.069 (SE 0.019)*
Black race:
Lagl: 0.047 (SE 0.034)
Log-transformed high-frequency
power:
Lagl: -0.010 (SE 0.016)
SD of normal R-R intervals:
Lagl: -0.336 (SE 0.290)
*p<0.05
Peters et al. (2000a)
Eastern Massachusetts
1995-1997
Records of detected
arrhythmias and therapeutic
interventions were
downloaded from
defibrillators implanted in
cardiac clinic patients aged
22-85 years (n = 100).
Analysis was restricted to
defibrillator discharges
precipitated by ventricular
tachycardias or fibrillation.
Data were analyzed by
logistic regression models
using fixed effect models
with individual intercepts.
24-h avg O3:
18.6 ppb
IQR 14.0
PM25,PM10,
BC, CO, N02,
SO,
No significant O3 effects observed for
defibrillator discharge interventions.
For patients with 10 or more
interventions, increased arrhythmias
were associated significantly with
PM2 5, CO, and NO2 at various lag
periods, but not O3.
24-h avg O3 (per 32 ppb):
Odds ratios:
Defibrillator discharges:
Patients with at least one event:
LagO: 0.96(0.47,1.98)
Patients with at least 10 events:
LagO: 1.23(0.53,2.87)
-------
Table AX7-2 (cont'd). Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study
Location and Period
Outcomes and Methods Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
United States (cont'd)
Peters etal. (2001)
Greater Boston area,
MA
Janl995-May 1996
Case-crossover study l-hmaxO3:
design used to investigate 19.8 ppb
association between air SD 148
pollution and triggering of 5th%-95th%
myocardial infarction in 1-46
772 patients (mean age
61.6 years). For each 24-h avg O3:
subject, one case period 19.9 ppb
was matched to three SD 10.0
control periods 24 h apart. 5th%-95th%
Conditional logistic 6-36
regression used for analysis.
PM25, PM10, None of the gaseous pollutants, including
PM10_25, BC, O3, were significantly associated with
CO, NO2, SO2 the triggering of myocardial infarctions.
Significant associations reported for
PM25andPM10. Note that the use of
unidirectional referent sampling may lead
to time trend bias and overlap bias (i.e.,
biased conditional logistic regression
estimating equations), which can result in
overestimated effects of exposure.
Odds ratios:
Myocardial infarctions:
2-h avg O3 (per 45 ppb):
Lag 1 h:
1.31 (0.85,2.03)
24-h avg O3 (per 30 ppb):
Lag 24 h:
0.94(0.60,1.49)
X Park et al. (2005)
7~" Greater Boston area,
£ MA
Nov 2000-Oct 2003
Cross-sectional study
examining the effect of O3
on heart rate variability in
497 adult males (mean age
72.7 years). Subjects were
monitored during a 4-min
rest period between 8 a.m.
and 1 p.m. Ozone levels
measured at central site 1
km from study site.
Exposure averaging times
of 4, 24, and 48 h
investigated. Modifying
effects of hypertension,
ischemic heart disease,
diabetes, and use of cardiac/
antihypertensive
medications also examined.
Linear regression analyses.
24-h avg O3: PM2 5, particle Of the pollutants examined, only PM25
23.0 ppb number and O3 showed significant associations
SD 13.0 concentration, with heart rate variability outcomes.
BC, NO2, SO2, The 4-h averaging period was most
CO strongly associated with heart rate
variability indices. The O3 effect was
robust in models including PM2 5. The
associations between O3 and heart rate
variability indices were stronger in
subjects with hypertension (n = 335) and
ischemic heart disease (n = 142).
In addition, calcium-channel blockers
significantly influenced the effect of O3
on low frequency power. Major
limitations of this study are the use of a
short 4-min period to monitor heart rate
variability and the lack of repeated
measurements for each subject.
4-h avg O3 (per 13 ppb):
Change in low frequency power:
All subjects:
-11.5% (-21.3, -0.4)
Subjects with hypertension:
-12.6% (-25.0, 1.9)
Subjects without hypertension:
-5.4% (-21.6, 14.1)
Subjects with ischemic heart
disease:
-25.8% (-41.9,-5.3)
Subjects without ischemic heart
disease:
-4.8% (-16.7, 8.8)
-------
Table AX7-2 (cont'd). Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study
Location and Period
Outcomes and Methods Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
United States (cont'd)
Gold et al. (2000;
reanalysis Gold et al.,
2003)
Boston, MA
Jun-Sep 1997
X
Panel study of repeated
measurements of heart rate
variability in subjects aged
53-87 years (n = 2 1,1 63
observations). Twenty-five
minute protocol included
5 min each of rest, standing,
exercise outdoors, recovery,
and 20 cycles of slow
breathing. Ozone levels
measured at central site
4.8 miles from study site.
Analyses using random
effects models and GAM
with stringent convergence
criteria.
1 -h avg O3 PM2 5, NO2, Increased levels of O3 were associated
during HRV SO2 with reduced r-MSSD (square root of the
measurement: mean of the squared differences between
25.7 ppb adjacent normal RR intervals) during the
IQR 23.0 slow breathing period after exercise
outdoors. The estimated O3 effects were
similar to those of PM25. Results suggest
that O3 exposure may decrease vagal
tone, leading to reduced heart rate
variability.
1-h avg O3 (per 23.0 ppb):
Change in r-MSSD:
During first rest period:
O3 only model:
-3.0 ms(SE 1.9), p = 0.12
During slow breathing period:
O3 only model:
-5.8ms(SE2.4),p = 0.02
O3 with PM2 5 model:
-5.4 ms(SE 2.5), p = 0.03
Schwartz et al. (2005)
Boston, MA
12 weeks during the
summer of 1999
A panel study of 28 elderly
subjects (age 61-89 years).
Various HRV parameters
were measured for 30 min
once a week. Analysis
using linear mixed models
with log-transformed HRV
measurements. To examine
heterogeneity of effects,
hierarchical model was
used.
1-h avg O3 BC, PM25, HRV parameters examined included:
during HRV CO, SO2, NO2 standard deviation of normal RR intervals
measurement: (SDNN), root mean squared differences
Median 34 ppb between adjacent R-R intervals
IQR 26 (r-MSSD), proportion of adjacent NN
intervals differing by more than 50 ms
(PNN50), and low frequency/high
frequency ratio (LFHFR). Ozone was
weakly associated with HRV parameters.
Strongest association seen for BC, an
indicator of traffic particles. The random
effects model indicated that the negative
effect of BC on HRV was not restricted
to a few subjects. Subjects with MI
experienced greater BC-related
decrements in HRV parameters. Authors
noted that in this study ambient O3 might
represent a secondary particle effect and
not a true O3 effect, and suggested that
personal exposure measurements might
be necessary to assess the effect of O3 on
cardiovascular outcomes.
1-h avg O3 (per 26 ppb):
Change in HRV parameters:
SDNN:
-3.1ms (-7.0, 0.9)
r-MSSD:
-8.5ms (-16.6, 0.3)
PNN50:
-6.5% (-18.9, 7.8)
-------
Table AX7-2 (cont'd). Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study
Location and Period
Outcomes and Methods Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
United States (cont'd)
Dockery et al. (2005)
Boston, MA
Ml 995-Jul 2002
X
Effect of air pollution on
incidence of ventricular
arrhythmias was examined
in 203 patients with
implantable cardioverter
defibrillators using time-
series methods. Mean
follow-up period was
3.1 year/subject. All
subjects located <40 km of
air pollution monitoring
site. Two-day mean air
pollution level used in
analysis. Results analyzed
by logistic regression using
GEE with random effects.
Modifying effects of
previous arrhythmia within
3 days also examined.
48-h avg O3:
Median 22.9 ppb
IQR 15.4
PM25, BC,
SO42~, particle
number, CO,
NO2, SO2
No associations were observed between
air pollutants and ventricular arrhythmias
when all events were considered.
When only examining ventricular
arrhythmias within 3 days of a prior
event, positive associations were found
for most pollutants except for O3.
Suggestive evidence of a concentration-
response relationship between ventricular
arrhythmias and increasing quintiles
of03.
48-h avg O3 (per 15.4 ppb):
Odds ratios:
All events:
1.09(0.93,1.29)
Prior arrhythmia event <3 days:
1.01 (0.76,1.35)
Prior arrhythmia event >3 days:
1.14(0.92,1.40)
-------
Table AX7-2 (cont'd). Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study
Location and Period
Outcomes and Methods Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
United States (cont'd)
Rich et al. (2005)
Boston, MA
Ml 995-Jul 2002
X
Same study population as
Dockery et al. (2005).
Case-crossover study design
used to examine association
between air pollution and
ventricular arrhythmias.
For each case period,
3-4 control periods were
selected. Various moving
average concentrations of
exposure considered - lags
0-2, 0-6, 0-23, and 0-47 h.
Analysis using conditional
logistic regression models.
1 -h avg O3: PM2 5, BC, Associations observed for PM2 5 and O3
Median 22.2 ppb CO, NO2, SO2 with a 24-h moving average, and for NO2
IQR21.7 and SO2 with a 48-h moving average. In
two-pollutant analyses, only PM25 and O3
24-h avg O3: appeared to act independently. In
Median 22.6 ppb contrast to results from other pollutants,
IQR 15.7 stratified analyses indicated that O3 was
associated with increased risk among
subjects without a recent event, but not
those with recent events. The odds ratio
for the 24-h moving average
concentration was larger than that for the
same-calendar day concentration. This
suggested that using calendar day
concentrations might result in greater
exposure misclassification which could
lead to underestimation of risk. Findings
of an association with 24-h moving
average concentrations but not with
shorter time periods could imply a
cumulative effect across the previous
24 h.
Odds ratios:
24-h moving average O3 (per
15.9 ppb):
O3 only model:
All events:
1.21 (1.00,1.45)
Prior arrhythmia event <3 days:
1.04(0.78,1.37)
Prior arrhythmia event >3 days:
1.28(1.05,1.58)
O3 with PM2 5 model:
All events:
1.18(0.94,1.49)
24-h calendar day O3
(per 15.7 ppb):
O3 only model:
All events:
0.96(0.80,1.15)
Canada
Rich et al. (2004)
Vancouver, British
Columbia, Canada
Feb-Dec 2000
Case-crossover study
design used to investigate
association between air
pollution and cardiac
arrhythmia in patients aged
15-85 years (n = 34) with
implantable cardioverter
defribillators. Controls
periods were selected 7 days
before and after each case
day. Analysis using
conditional logistic
regression.
l-hmaxO3
27.5 ppb
SD9.7
IQR 13.4
PM25,PM10,
EC, OC,
so/-, co,
NO2, SO2
No consistent association between any
of the air pollutants, including O3, and
implantable cardioverter defribillators
discharges. No significant association
observed in all-year data, however,
significant relationship found in winter
months at a 3-day lag. Overall, little
evidence that air pollutants affect risk
of cardiac arrhythmias, however, power
was limited to study subtle effects.
No quantitative results for O3.
-------
Table AX7-2 (cont'd). Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study
Location and Period
Outcomes and Methods Mean O3 Levels
Copollutants
Considered
Findings, Interpretation
Effects
Canada (cont'd)
Vedal et al. (2004)
Vancouver, British
Columbia, Canada
1997-2000
Retrospective, longitudinal
panel study of 50 patients
(age 12-77 years) with
implantable cardioverter
defribillators. Occurrence
of cardiac arrhythmia was
associated with air
pollutants over 4-year
period. GEE used for
analysis.
28.2 ppb
SD 10.2
IQR 13.8
PM10, CO,
N02, S02
No consistent association between any
of the air pollutants and % change in
arrhythmia. Among patients with at least
2 arrhythmia event-days per year, a
significant negative relationship between
O3 and arrhythmias was observed at lag
3-day during the summer, but no
associations were found during the
winter. These results do not provide
evidence for an O3 effect on cardiac
arrhythmias in susceptible patients.
No quantitative results for O3.
X
oo
Europe
Ruidavets et al. (2005)
Toulouse, France
Jan 1997 - Jun 1999
MONICA Project.
Examined short-term effects
of pollution on acute MI
using case-crossover study
design. The study
population included
395,744 inhabitants aged
35 to 64 years. Acute MI
was examined using
clinical, electrocardiograms,
and enzymatic data. Four
case definitions were used;
there were a total of 635
cases for the most inclusive
definition. Deaths were
validated. Age, gender,
history of ischemic heart
disease, and survival status
evaluated. Analyses using
fixed-effects method with
conditional logistic
regression.
8-h max O3: NO2, SO2 After adjustment for temperature, relative
74.8 ug/m3 humidity, and influenza epidemics, an
SD 28.1 association between O3 and acute MI was
Range 3.8-160.2 found for 0- and 1-day lags, but not for
longer lags. Older age was an important
risk factor. Subjects with no personal
history of ischemic heart disease yielded
a stronger association. Moderate levels
of NO2 and SO2 were observed. NO2 and
SO2 were not associated with acute MI.
No PM data was reported.
8-h max O3 (per 5 ug/m3):
Relative risk:
All cases (n = 635):
LagO: 1.05(1.01,1.08)
Lagl: 1.05(1.01,1.09)
Age group:
Age 35-54 years (n = 281):
LagO: 1.04(0.99,1.09)
Age 55-64 years (n = 283):
LagO: 1.06(1.01,1.12)
History of ischemic heart disease:
Yes (n = 127):
LagO: 1.03(0.96,1.12)
No (n = 437):
LagO: 1.05(1.01,1.09)
Age 55-64 years with no history
of ischemic heart disease (n =
225):
LagO: 1.07(1.01,1.13)
Lagl: 1.11 (1.04,1.19)
-------
Table AX7-2 (cont'd). Effects of Acute O3 Exposure on Cardiovascular Outcomes in Field Studies
Reference, Study Copollutants
Location and Period Outcomes and Methods Mean O3 Levels Considered
Findings, Interpretation
Effects
Latin America
Holguin et al. (2003)
Mexico City
Feb-Apr 2000
X
VO
Panel study of the
association between O3
and heart rate variability
examined in 34 elderly
subjects (mean age
79 years) in a nursing home.
Subjects were monitored
during a 5-min rest period
between 8 a.m. and 1 p.m.
every other day for a 3-mo
period. A total of 595
observations were collected.
Ambient O3 levels obtained
from central site 3 km
upwind from study site.
Analysis performed using
GEE models adjusting for
potential confounding
factors including age and
average heart rate.
l-hmaxO3
149 ppb
SD40
PM2 5 (indoor,
outdoor,
total), NO2,
SO2, CO
Only PM2 5 and O3 were significantly
associated with heart rate variability
outcomes. A significant effect of O3 on
heart rate variability was limited to
subjects with hypertension (n = 21).
In two-pollutant models, the magnitude
of the PM25 effect decreased slightly but
remained significant, whereas O3 was
no longer associated with heart rate
variability indices.
1-h max O3 (per 10 ppb):
Log10 high frequency
power/100,000 ms2:
All subjects:
-0.010 (-0.022, 0.001)
Subjects with hypertension:
-0.031 (-0.051,-0.012)
Subjects without hypertension:
0.002 (-0.012, 0.016)
Log10 low frequency
power/100,000 ms2:
All subjects:
-0.010 (-0.021, 0.001)
Subjects with hypertension:
-0.029 (-0.046,-0.011)
Subjects without hypertension:
0.001 (-0.012, 0.015)
AX7-39
-------
Table AX7-3. Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and
Design
Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
United States
Jaffe et al. (2003)
Cincinnati, Cleveland,
and Columbus, OH
Jun-Aug 1991-1996
X
Jones etal. (1995)
Baton Rouge, LA
Jun-Aug 1990
Daily time-series
study of emergency
department visits for
asthma among
Medicaid recipients
aged 5-34 years.
Daily emergency
department visits for
respiratory complaints
over a 3-month period
in pediatric
(age 0-17 years), adult
(age 18-60 years),
and geriatric
(age >60 years)
subgroups.
Time-series study.
8-h max O3:
Cincinnati:
60ppb
SD20
Cleveland:
SOppb
SD17
Columbus:
57ppb
SD16
PM10,N02,
SO,
1,2,3
l-hmaxO3:
69.1 ppb
SD28.7
24-h avg O3:
28.2 ppb
SD11.7
Mold, pollen Not specified.
Poisson regression with control
for city, day of week, week,
year, minimum temperature,
overall trend, and a dispersion
parameter. No specific effort
to control cycles, but
regression residuals were
uncorrelated, presumably due
to seasonal restriction. Results
shown for individual cities and
overall. PM10 available only
every 6th day. Positive
relationships between
emergency department visits
for asthma and 8-h max O3
levels lagged 2 to 3 days.
Results of borderline statistical
significance. Other pollutants
also related to asthma
emergency department visits
in single-pollutant models.
Relatively simple statistical
approach using ordinary least
squares regression to model
effects of O3 by itself and of O3
along with pollen counts, mold
counts, temperature, and
relative humidity. No direct
control of cycles but authors
reported nonsignificant
autocorrelations among model
residuals. Data restriction to
3-month period may have
reduced any cyclic behavior.
Significant associations
between respiratory emergency
department visits and O3
observed for adult age group
only in multiple regression
models.
8-h max O3 (per 30 ppb):
Cincinnati:
Lag 2: 1.16(1.00,1.37)
Cleveland:
Lag 2: 1.03(0.92,1.16)
Columbus:
Lag 3: 1.16(0.98,1.37)
Three cities: 1.09(1.00,
1.19)
24-h avg O3 (per 20 ppb):
Pediatric: 0.87(0.65,1.09)
Adult: 1.20(1.01,1.39)
Geriatric: 1.27(0.93,1.61)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and
Design
Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
X
United States (cont'd)
Wilson et al. (2005)
Portland, ME
1998-2000
Manchester, NH
1996-2000
Daily emergency
room visits for total
respiratory and asthma
examined for all ages
and age groups 0-14,
15-64, and 65+ years.
Time-series study.
8-h max O3:
Portland:
Spring:
43.7ppb
SD 10.2
Summer:
46.1 ppb
SD 15.4
Manchester:
Spring:
43.4 ppb
SD9.7
Summer:
42.8 ppb
SD 14.6
SO,
Poisson GAM with stringent
convergence criteria. Positive
associations for asthma in the
larger city, Portland. Authors
expressed the view that larger
cities might be needed to
conduct such studies.
8-h max O3 (per 30 ppb):
Warm season (Apr-Sep):
All ages:
Portland:
Total respiratory:
0.970(0.915,1.029)
Asthma:
1.094(1.032,1.160)
Manchester:
Total respiratory:
0.970(0.915,1.029)
Asthma:
0.970(0.863,1.092)
Cassinoetal. (1999)
New York City
Jul 1992-Dec 1995
Daily time-series
study of emergency
department visits in a
cohort of 1,115 adult
asthmatics aged
18-84. Stratified into
552 never-smokers,
278 light smokers, and
285 heavy smokers.
l-hmaxO3:
37.2 ppb
IQR28
24-h avg O3:
17.5 ppb
IQR14
CO, NO2,
SO,
0,1,2,3 Used Poisson GAM with
default convergence criteria.
No warm-season results
presented. Significant O3
effects seen only at lag 2
among heavy-smokers.
Copollutants did not have
effects. Short-term cycles and
episodic variations in asthma
may not have been controlled
adequately with 3-month
period LOESS. Multiple tests
performed, and inconsistent
results across smoking strata
and lags raise possibility of
chance findings. No PM
results included.
1-h max O3 (per 100 ppb):
Never smokers:
Lag 2: 0.80(0.58,1.14)
Light smokers:
Lag 2: 0.89(0.62,1.16)
Heavy smokers:
Lag 2: 1.59(1.08,2.36)
24-h avg O3 (per 50 ppb):
All subjects:
Lag 2: 1.22(0.98,1.53)
Never smokers:
Lag 2: 0.88(0.57,1.12)
Light smokers:
Lag 2: 0.88(0.60,1.28)
Heavy smokers:
Lag 2: 1.72(1.13,2.62)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
X
to
Reference, Study
Location and Period
United States (cont'd)
Weisel et al. (2002)
New Jersey
May-Aug 1995
Friedman etal. (2001)
Atlanta, GA
Jun-Sep 1996
Outcomes and
Design
Daily asthma
emergency department
visits for all ages.
Time-series study.
Emergency
department visits and
hospital admissions
for asthma in children
aged 1-16 years.
Outcomes measures
during 1996 Summer
Olympics were
compared to a
baseline period of
4 weeks before and
after the Olympic
Games. Time-series
study.
Mean O3 Levels
l-hmaxO3;
5-h avg O3
(10 a.m.-3 p.m.);
and 8-h avg O3
(2 p.m.-lO p.m.)
analyzed.
Levels not
reported.
l-hmaxO3:
Baseline:
81.3ppb
SD not provided.
Intervention
period:
58.6 ppb
SD not provided.
Copollutants Lag Structure Method, Findings,
Considered Examined Interpretation
Pollen, spores 0, 1, 2, 3 No control for time, but
authors report no
autocorrelation, which
alleviates concerns about lack
of control. Significant O3
effects reported, even after
adjusting for potential
confounding by pollen. All
three O3 indices gave
essentially same results.
NO2, SO2, 0,0-1,0-2 Analysis using Poisson GEE
CO, PM10, models addressing serial
mold autocorrelation. An overall
decrease was observed when
comparing the number of visits
and hospitalizations during the
Olympic Games to the baseline
period. However, significant
associations between O3 and
asthma events were found
during the Olympic Games.
Effects
(Relative Risk and 95% CI)
Slope estimate
(visits/day/ppb):
Excluding data from May
when pollen levels are high:
O3 only model:
LagO: 0.039, p = 0.049
O3 with pollen model:
LagO: 0.040, p = 0.014
1-h max O3 (per 50 ppb):
Pediatric emergency
departments:
LagO: 1.2(0.99,1.56)
Lag 0-1: 1.4(1.04,1.79)
Lag 0-2: 1.4(1.03,1.86)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
X
Reference, Study
Location and Period
United States (cont'd)
Metzger et al. (2004)
Atlanta, GA
1993-2000
Peel et al. (2005)
Atlanta, GA
1993-2000
Outcomes and
Design Mean O3 Levels
Emergency 8-h max O3:
department visits for
total and cause- Mar-Nov:
specific Median 53.9 ppb
cardiovascular 1 Oth %-90th %
diseases by age groups 13.2-44.7
19+ years and 65+
years. Time-series
study.
Emergency 8-h max O3:
department visits for
total and cause- Mar-Nov:
specific respiratory 55. 6 ppb
diseases by age groups SD 23.8
0-1,2-18, 19+, and
65+ years. Time-
series study.
Copollutants
Considered
NO2, SO2,
CO,PM25,
PM10,
PM10.2.5,
ultrafine PM
count, SO42~,
H+, EC, OC,
metals,
oxygenated
hydrocarbons
N02, S02,
CO,PM25,
PM10,PM10.
25, ultrafine
PM count,
SO42~, H+,
EC, OC,
metals,
oxygenated
hydrocarbons
Lag Structure Method, Findings,
Examined Interpretation
0-2 Poisson GLM regression used
for analysis. A priori models
specified a lag of 0 to 2 days.
Secondary analyses performed
to assess alternative pollutant
lag structures, seasonal
influences, and age effects.
Cardiovascular visits were
significantly associated with
several pollutants, including
N02, CO, and PM2 5, but
not O3.
0-2 Poisson GEE and GLM
regression used for analysis.
A priori models specified a lag
of 0 to 2 days. Also performed
secondary analyses estimating
the overall effect of pollution
over the previous 2 weeks.
Seasonal analyses indicated
stronger associations with
asthma in the warm months.
Quantitative results not
presented for multipollutant,
age-specific, and seasonal
analyses.
Effects
(Relative Risk and 95% CI)
8-h max O3 (per 25 ppb):
All ages:
Total cardiovascular:
1.008(0.987,1.030)
Dysrhythmia:
1.008(0.967,1.051)
Congestive heart failure:
0.965(0.918,1.014)
Ischemic heart disease:
1.019(0.981,1.059)
Peripheral vascular and
cerebro vascular disease:
1.028(0.985,1.073)
8-h max O3 (per 25 ppb):
All ages:
All available data:
Total respiratory:
1.024(1.008,1.039)
Upper respiratory infections:
1.027(1.009,1.045)
Asthma:
1.022(0.996,1.049)
Pneumonia:
1.015(0.981,1.050)
COPD:
1.029(0.977,1.084)
Warm season (Apr-Oct):
Asthma:
1.026(1.002,1.051)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
United States (cont'd)
Tolbert et al. (2000)
Atlanta, GA
Outcomes and
Design
Pediatric (aged
0-16 years) asthma
Mean O3 Levels
l-hmaxO3:
68.8 ppb
Copollutants
Considered
PM10, NO2,
mold, pollen
Lag Structure
Examined
1
Method, Findings,
Interpretation
A priori specification of
model, including a lag of 1 day
Effects
(Relative Risk and 95% CI)
8-h max O3 (per 20 ppb):
Jun-Aug 1993-1995
X
Zhu et al. (2003)
Atlanta, GA
Jun-Aug 1993-1995
emergency department SD 21.1
visits over three
summers in Atlanta. 8-h max O3:
A unique feature of 59.3 ppb
the study was SD 19.1
assignment of O3
exposures to zip code
centroids based on
spatial interpolation
from nine O3
monitoring stations.
Time-series study.
Asthma emergency
department visits in
children (age 0-16
years) over three
summers in Atlanta.
Time-series study.
8-h max O3:
Levels not
reported.
None
for all pollutants and
meteorological variables.
Secondary analysis using
logistic regression of the
probability of daily asthma
visits, referenced to total visits
(asthma and nonasthma).
Significant association with
O3 and PM10 in 1-, but not in
2-pollutant models (correlation
between O3 and PM10:
r = 0.75). Secondary analysis
indicated nonlinearity, with O3
effects clearly significant only
for days >100 ppb vs. days
<50 ppb.
Used Bayseian hierarchical
modeling to address model
variability and spatial
associations. Data were
analyzed at the zip code level
to account for spatially
misaligned longitudinal data.
A positive, but nonsignificant
relationship between O3 and
emergency room visits for
asthma.
Poisson regression:
O3 only model:
1.040(1.008,1.074)
Logistic regression:
O3 only model:
1.042(1.017,1.068)
O3 with PM10 model:
1.024(0.982,1.069)
8-h max O3 (per 20 ppb):
Posterior median:
1.016(0.984,1.049)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and
Design
Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Canada
Delfinoetal. (1997b)
Montreal, Quebec,
Canada
Jun-Sep 1992-1993
X
Daily time-series
ecologic study of
emergency department
visits for respiratory
complaints within five
age strata (<2, 2-18,
19-34,35-64,
>64 years).
1-h max O3:
1992:
33.2 ppb
SD 12.6
1993:
36.2 ppb
SD 13.8
8-h max O3:
1992:
28.8 ppb
SDH. 3
1993:
30.7 ppb
SD11.5
PM10,PM25
SO/-, H+
0, 1, 2 Used ordinary least squares,
with 19-day weighted moving
average pre-filter to control
cycles; weather also controlled.
Significant effects reported for
1-day lag O3 in 1993 only for
age >64 years. This O3 effect
reported to be robust in two-
pollutant models. LowO3
levels raise plausibility
concerns. Short data series,
multiple tests performed, and
inconsistent results across
years and age groups raise
possibility of chance findings.
1993 (age >64 years):
1-h max O3 (per 36.2 ppb):
Lag 1: 1.214(1.084,1.343)
8-h max O3 (per 30.7 ppb):
Lagl: 1.222(1.091,1.354)
No significant O3 effects in
1992 or in other age groups at
the p < 0.02 level (chosen to
take into account multiple
testing bias).
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and
Design
Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Canada (cont'd)
Delfinoetal. (1998b)
Montreal, Quebec,
Canada
Jun-Aug 1989-1990
X
Daily time-series
ecologic study of
emergency department
visits for respiratory
complaints across all
ages and within four
age strata (<2, 2-34,
35-64, >64 years).
l-hmaxO3:
1989:
44.1 ppb
SD 18.3
1990:
3 5. 4 ppb
SD 12.9
8-h max O3:
1989:
37.5 ppb
SD 15.5
Estimated
0, 1, 2 Same analytical approach used
in Delfino et al, 1997. Results
presented only for 1989.
Significant effects reported for
1-day lag O3 in 1989 only for
age >64 years. This O3 effect
reported to be robust in
2-pollutant models. Ozone
concentrations were
significantly higher in 1989
than 1990 (p< 0.001).
1989 (age >64 years):
1-h max O3 (per 44.1 ppb):
Lag 1: 1.187(0.969,1.281)
8-h max O3 (per 37.5 ppb):
Lagl: 1.218(1.097,1.338)
No significant O3 effects in
1990 or in other age groups at
the p < 0.05 level.
1990:
29.9 ppb
SD11.2
Stiebetal. (1996)
Saint John, New
Brunswick, Canada
May-Sep 1984-1992
Daily emergency
department visits for
asthma in all ages,
age<15 years and
15+years. Time-
series study.
1-h max O3: SO2, NO2, 0,1,2,3 Poisson analysis with control
41.6 ppb SO42~, TSP of time based on 19-day
Range 0-160 moving average filter. Also
95th % 75 controlled day of week and
weather variables. Ozone was
only pollutant consistently
associated with emergency
department visits for asthma,
but effect appeared nonlinear,
with health impacts evident
only above 75 ppb O3.
All ages:
1-h max O3 (per 40 ppb):
Lag 2: 1.09(1.00,1.19)
1-h max O3 >75 ppb vs. <75
ppb:
Lag 2: 1.33(1.10,1.56)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
X
Reference, Study
Location and Period
Europe
Sunyeretal. (1997)
Four European cities:
Barcelona, Helsinki,
London, and Paris
1986-1992
Atkinson etal. (1999a)
London, England
1992-1994
Hajatetal. (1999;
2002)
London, England
1992-1994
Outcomes and
Design
Emergency
admissions for asthma
in children (< 15
years) and adults (15-
64 years). Time-
series study.
Emergency
department visits
for respiratory
complaints, asthma for
all ages and age 0-14,
15-64, and 65+ years.
Time-series study.
Daily doctor consults
for asthma, lower
respiratory diseases,
and upper respiratory
diseases for age 0-14,
15-64, and >65 years.
Time-series study.
Mean O3 Levels
l-hmaxO3:
Barcelona:
Median 72 ug/m3
Range 7-283
Helsinki:
Median 27 ug/m3
Range 1-78
London:
Median 40 ug/m3
Range 1-188
Paris:
Median 36 ug/m3
Range 1-230
8-h max O3:
17.5 ppb
SD11.5
8-h max O3:
All year:
17.5 ppb
SD11.5
Warm season:
22.7 ppb
SD 12.2
Cold season:
12.1 ppb
SD7.6
Copollutants Lag Structure Method, Findings,
Considered Examined Interpretation
BS, SO2, NO2 0, 1, 2, 3 Poisson analysis using APHEA
methodology. Significant O3
effects on emergency
admissions for asthma
observed among 15-64 year
olds in Barcelona and London.
Across all cities, there was no
strong evidence for
associations involving O3.
NO2, SO2, 0, 1 , 2, 3 Poisson GLM regression used
CO, PM10 0-1,0-2,0-3 for analysis. No warm-season
analysis attempted. PM10
positively associated.
BS, SO2, 0-3 Used Poisson GAM with
NO2, CO, default convergence criteria.
PM10, pollen Conducted all-year and
seasonal analyses. Single- and
two-pollutant models analyzed.
Significant negative effects for
O3. This may reflect residual
confounding by seasonal
factors or highly negative
correlation with other
pollutants.
Effects
(Relative Risk and 95% CI)
1-h max O3 (per 50 ug/m3):
Weighted mean effect (best
lag selected for each city):
Age<15 years:
3-city pooled estimate
(Barcelona data not
available):
1.006(0.976,1.037)
Age 15-64 years:
4-city pooled estimate:
1.015(0.955,1.078)
8-h max O3 (per 25.7 ppb):
All ages:
Total respiratory:
Lagl: 1.017(0.991,1.043)
Asthma:
Lag 1: 1.027(0.983,1.072)
Upper respiratory diseases,
age >65 years:
All year:
8-h max O3 (per 25.7 ppb)
Lag 2: -8.3% (-13.3, -3.0)
Warm season:
8-h max O3 (per 28.5 ppb)
Lag 2: -0.6% (-6.1, 5.1)
Cool season:
8-h max O3 (per 19.8 ppb)
Lag 2: -7.9% (-12.9, 2.7)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and Copollutants Lag Structure
Design Mean O3 Levels Considered Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Europe (cont'd)
Thompson etal. (2001)
Belfast, N Ireland
1993-1995
Asthma emergency
department
admissions in children
(age unspecified).
Time-series study.
24-h avg O3:
Warm season:
18.7 ppb
IQR9
Cold season:
17.1 ppb
IQR12
PM10, S02:
NO2, CO,
benzene
0,0-1,0-2,0-3
GLM with sinusoids.
Pre-adjustment. Very low O3
levels in both seasons. No O3
effect in warm season.
Significant inverse O3
associations in full-year and
cold-season models. After
adjusting for benzene in model
O3 was no longer negatively
associated with asthma visits.
24-h avg O3 (per 10 ppb):
All year:
O3 only model:
Lag 0-1: 0.93(0.87,1.00)
O3 with benzene model:
Lag 0-1: 1.08(0.97,1.21)
Warm season:
O3 only model:
Lag 0-1: 0.99(0.89,1.10)
X
Cold season:
O3 only model:
Lag 0-1: 0.89(0.82,0.97)
oo
Bourcier et al. (2003)
Paris, France
Jan 1999-Dec 1999
Ophthalmological
emergency
examination;
conjunctivitis and
related ocular surface
problems. Time-
series study.
24-h avg O3:
35.7 ug/m3
Range 1-97
PM10, S02,
NO,
0,1,2,3
Logistic Regression
Results indicate a strong
relation to NO2 and NO.
24-h avg O3 (per 69 ug/m3):
Conjunctivitis:
LagO: 1.13(0.90,1.42)
Castellsague et al.
(1995)
Barcelona, Spain
1985-1989
Daily emergency
department visits for
asthma in persons
aged > 14 years.
Time-series study.
l-hmaxO3:
Summer:
43 ppb
IQR22
Winter:
29 ppb
IQR16
BS, SO2, NO2 Not specified. Poisson regression with year
and month dummy variables
and extensive control for
weather factors (minimum,
maximum, mean temperature,
relative humidity, dewpoint
temperature; continuous and
categorical parameterizations)
1-h max O3 (per 12.7 ppb):
Summer:
0.991 (0.
Winter:
1.055(0.
939,1.045)
998,1.116)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and
Design
Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Europe (cont'd)
Tobias etal. (1999)
Barcelona, Spain
1986-1989
Daily asthma
emergency department
visits. Investigated
sensitivity of results
to four alternative
methods for
controlling asthma
epidemics. Time-
series study.
Levels not
reported.
BS, NO2, SO2 Not specified.
Poisson analysis using
APHEA methodology.
Asthma epidemics either
not controlled, or controlled
with one, six, or individual
epidemic dummy variables.
O3 results were sensitive to
method used to control
asthma epidemics, with
regression coefficients
ranging over 5-fold
depending on the model.
Only 1 of 8 models reported
had a significant O3 effect.
X
Tenias et al.
(1998; 2002)
Valencia, Spain
1994-1995
Latin America
Hemandez-Garduno
etal. (1997)
Mexico City
May 1992-Jan 1993
Daily emergency
department visits for
asthma and COPD
in persons aged
>14 years. Time-
series study.
Visits to clinics for
respiratory diseases in
persons aged 1 month
to 92 years. Time-
series study.
l-hmaxO3: BS,NO2,
S02, CO
All year:
62.8 ug/m3
Range 13.3-157.3
Warm season:
74.0 ug/m3
Cool season:
5 1.4 ug/m3
Percent time SO2, NO2,
exceeding 1-h CO
max O3 of 120
ppb:
6.1-13.2% by
location
0, 1,2, 3, 4, 5 Poisson analysis using APHEA
methodology. Compared
warm and cold season effects.
GAM explored in sensitivity
analysis. For asthma, both O3
and NO2 significant in single-
and two-pollutant models, and
O3 effect larger in warm
season. For COPD, both O3
and CO significant in both
single- and two-pollutant
models and no difference in O3
effects by season.
0,1,2,3,4,5 GLM with pre-adjustment.
Ozone at lags 0 and 5 days
significantly associated with
daily visits for all ages,
age <14 years, and 15+ years.
Neither O3 nor NO2 significant
in two-pollutant model.
1-h max O3 (per 10 ug/m3):
Asthma:
All year:
Lag 1: 1.06(1.01,1.11)
Warm season:
Lag 1: 1.08(1.02,1.05)
Cold season:
Lagl: 1.04(0.97,1.11)
COPD:
All year:
Lag 5: 1.06(1.02,1.10)
1-h max O3 (per maximum
less average, value not
given):
LagO: 1.19(SE0.08),
p<0.05
Lag 5: 1.19 (SE 0.08),
p<0.05
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
X
Reference, Study
Location and Period
Latin America (cont'd)
Lin etal. (1999)
Sao Paulo, Brazil
May 1991 -Apr 1993
Martins et al. (2002)
Sao Paulo, Brazil
May 1996-Sep 1998
Ilabaca etal. (1999)
Santiago, Chile
Feb 1995-Aug 1996
Asia
Hwang and Chan
(2002)
50 communities
in Taiwan
1998
Outcomes and
Design
Daily pediatric (age
unspecified)
respiratory emergency
department visits.
Time-series study.
Daily emergency
department visits
for chronic lower
respiratory diseases
in persons
aged >64 years.
Time-series study.
The association
between pollutant
levels and emergency
visits for pneumonia
and other respiratory
illnesses among
children. Time-series
study.
Daily clinic visits for
lower respiratory
illnesses for all ages.
Time-series study.
Mean O3 Levels
l-hmaxO3:
34ppb
SD22
l-hmaxO3:
34ppb
SD21
IQR21
l-hmaxO3:
Warm season:
66.6 ug/m3
SD25.2
Cold season:
27.6 ug/m3
SD 20.2
l-hmaxO3:
Means for
50 communities:
54.2 ppb
SD 10.2
Range 38.9-78.3
Copollutants Lag Structure Method, Findings,
Considered Examined Interpretation
SO2, CO, 0,1,0-1,0-2, Seasonal control using month
PM10,NO2 0-3,0-4,0-5 dummy variables. Also
controlled day of week,
temperature. Both O3 and
PM10 associated with outcome
alone and together.
CO,NO2, 0-1,0-2,0-3, Analyzed using Poisson GAM
SO2, PM10 0-4, 0-5, 0-6. with default convergence
criteria. Only O3 and SO2
significant in single-pollutant
models. Ozone effect
remained significant when SO2
included in two-pollutant
model.
PM10, PM25, 1, 2, 3, 1-7 Poisson regression analysis.
SO2,NO2
NO2, SO2, 0, 1 Analysis using general linear
PM10, CO regressions with
moving-average residual
processes and Bayesian
hierarchical modeling. All
pollutants except O3 were
associated with daily clinic
visits.
Effects
(Relative Risk and 95% CI)
1-h max O3 (per 5 ppb):
O3 only model:
Lag 0-4: 1.022(1.016,1.028)
O3 with PM10 model:
Lag 0-4: 1.015(1.009,1.021)
1-h max O3 (per 18.26 ppb):
Lag 0-3: 1.14(1.04,1.23)
Warm season:
1-h max O3 (per 30 ug/m3):
Lag 2: 1.019(1.003,1.035)
Cold season:
1-h max O3 (per 24 ug/m3):
Lag 2: 0.995(0.978,1.011)
1-h max O3 (per 40 ppb):
Bayesian hierarchical model:
1.003(0.983,1.023)
-------
Table AX7-3 (cont'd). Effects of O3 on Daily Emergency Department Visits
Reference, Study
Location and Period
Outcomes and
Design
Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Asia (cont'd)
Chew etal. (1999)
Singapore
Jan 1990-Dec 1994
Emergency
department visits for
asthma in persons
aged 3-21 years.
Time-series study.
1 -h max O3: TSP, PM10, 0,1,2 Simplistic but probably
23 ppb SO2, NO2 adequate control for time by
SD 15 including 1-day lagged
outcome as covariate.
In adjusted models that
included covariates, O3
had no significant effect.
No quantitative results
presented for O3.
X
-------
Table AX7-4. Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design
Copollutants Lag Structure
Mean O3 Levels Considered Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
United States
Neidell (2004)
California
1992-1998
X
to
Mann et al. (2002)
South Coast air basin,
CA
1988-1995
Asthma hospital
admissions within five
age strata (0-1, 1-3,
3-6, 6-12, and
12-18 years).
Time-series study.
O3 (index not
specified):
38.9 ppb
SD 17.8
Low SES:
40.1 ppb
High SES:
38.3 ppb
Ischemic heart disease 8-h max O3:
admissions for age 50.3 ppb
40+years. Time-series SD30.1
study. IQR 39.6
CO, NO2, PM10;
multipollutant
models
Not specified.
PM10, CO, NO2
0,1,2,3,4,5,
0-1,0-2,0-3,
0-4
Statistical analysis using
naturally occurring seasonal
variations in pollutant
concentrations within zip
codes. Linear regression
analyses. Consistent
significant positive effects
only observed for CO.
Negative O3 effect observed
in all age groups. Number of
smog alerts was negatively
associated with asthma
hospitalizations, indicating
avoidance behavior on high
O3 days. Interaction term
with indicator variable for
low SES was positive in all
age groups and statistically
significant in age 3-6 years
and 12-18 years, after
adjusting for number of smog
alerts.
Poisson GAM with cubic
B-splines; co-adjustment.
No significant O3 effects
observed overall or in warm
season. CO and NO2
significant in full-year
analyses.
Slope estimate (adjusting for
number of smog alerts):
O3 with CO, NO2, and PM10
models:
Age 3-6 years:
-0.038 (SE 0.014)
Age 6-12 years:
-0.044 (SE 0.013)
Age 12-18 years:
-0.022 (SE 0.011)
O3 x low SES interaction
term:
Age 3-6 years:
0.092 (SE 0.026)
Age 6-12 years:
0.024 (SE 0.024)
Age 12-18 years:
0.042 (SE 0.019)
O3 coefficients all negative,
but no consistent, significant
effect.
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
Reference, Study
Location and Period
United States (cont'd)
Linn et al. (2000)
Los Angeles, CA
1992-1995
Nauenberg and Basu
(1999)
Los Angeles, CA
1991-1994
Sheppard et al.
(1999; reanalysis
Sheppard, 2003)
Seattle, WA
1987-1994
Outcomes and Design
Total respiratory and
total cardiovascular
admissions for age
30+ years. Time-series
study.
Unscheduled asthma
admissions for all ages.
Time-series study.
Asthma admissions for
age <65 years.
Time-series study.
Copollutants
Mean O3 Levels Considered
24-h avg O3: PM10, CO, NO2
Winter:
14 ppb
SD7
Spring:
32 ppb
SD10
Summer:
33 ppb
SD8
Fall:
15 ppb
SD9
24-h avg O3: PM10
All year:
19. 88 ppb
SD11.13
Nov-Mar:
7.52 ppb
SD4.34
8-h max O3: PM25, PM10,
30.4 ppb PM10.25, SO2,
IQR 20 CO
Lag Structure Method, Findings,
Examined Interpretation
0 Poisson GLM;
co-adjustment. Only
significant O3 effects
observed were inverse
associations with total
cardiac admission in full-year
and winter season, suggesting
residual confounding.
No significant effects of O3
on respiratory admissions.
0, 0-7 Poisson GLM with pre-
adjustment. No significant
effects of O3. No
warm-season results
presented.
1,2,3 Poisson GAM, reanalyzed
with stringent convergence
criteria; Poisson GLM.
Results stratified by season.
Ozone significant predictor
of outcome. No two-
pollutant model results
reported for O3.
Effects
(Relative Risk and 95% CI)
24-h avg O3 (per 10 ppb):
All year:
Respiratory:
1.008(1.000,1.016)
Cardiovascular:
0.993(0.987,0.999)
Summer:
Respiratory:
1.006(0.992,1.020)
Cardiovascular:
1.001(0.991,1.011)
24-h avg O3 (per 20 ppb):
Nov-Mar:
LagO: 1.01(0.93,1.08)
8-h max O3 (per 20 ppb):
GLM with natural splines:
Lag 2: 1.07(1.01,1.13)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
United States (cont'd)
Schwartz (1996)
Spokane, WA
Apr-Oct 1988-1990
X
Koken et al. (2003)
Denver, CO
M-Aug 1993-1997
Total respiratory
admissions in persons
aged 65+ years.
Cause-specific
cardiovascular
admissions for age
>65 years. Cause
categories include
acute MI, coronary
atherosclerosis,
pulmonary heart
disease, cardiac
dysrhythmia, and
congestive heart
failure. Time-series
study.
l-hmaxO3:
79 ug/m3
IQR23
24-h avg O3
56 ug/m3
IQR17
24-h avg O3:
25.0ppb
SD6.61
Range 5.4-40.2
PM,f
PM10,N02,
S02, CO
2 Poisson GAM with default
convergence criteria. Results
available only for warm
season. Ozone and PM10
both significant predictors of
outcome. No two-pollutant
models reported. Ozone
effects robust to more
extensive temperature
specification.
0,1, 2, 3,4 Analysis using Poisson
GLM. Results suggest that
O3 increases the risk of
hospitalization for coronary
atherosclerosis and
pulmonary heart disease.
No association was found for
PM10. Strong O3 effects
observed in this seasonal
study compared to other
studies examining year-round
data. Male gender and
higher temperatures were
found to be important risk
factors for cardiovascular
disease. No multipollutant
analyses were reported.
1-h max O3 (per 50 ug/m3):
Lag 2: 1.244(1.002,1.544)
24-h avg O3 (per 50 ug/m3):
Lag 2: 1.284(0.926,1.778)
24-h avg O3 (per 9.7 ppb):
Acute MI:
0.824 (0.733, 0.925)
Coronary atherosclerosis:
1.123(1.040,1.214)
Pulmonary heart disease:
1.214(1.040,1.418)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
X
United States (cont'd)
Moolgavkar et al.
(1997)
Minneapolis/St. Paul,
MN and Birmingham,
AL
1986-1991
Lippmann et al. (2000;
reanalysis Ito, 2003,
2004)
Detroit, MI
1992-1994
Pneumonia and COPD
admissions for age
65+ years. Time-series
study.
Pneumonia, COPD,
ischemic heart disease,
dysrhythmias, heart
failure, and stroke
admissions for age
65+ years. Time-series
study.
24-h avg O3:
Minneapolis/
St. Paul:
26.2 ppb
IQR15.3
Birmingham:
25.1 ppb
IQR 12.7
24-h avg O3:
25 ppb
Maximum 55
PM10, SO2, NO2
0,1,2,3
PM10.25,S042-,
H+, NO2, SO2,
CO
0,1,2,3,0-1,
0-2, 0-3
Poisson GLM with co-
adjustment. Both O3 and
PM10 significant in MN; not
in AL. Ozone, but not PM10,
effects were robust to NO2
and SO,.
Poisson GAM, reanalyzed
with stringent convergence
criteria; Poisson GLM. At
various lags, no consistent O3
effect observed for any
cause-specific hospital
admission. No two-pollutant
models reported.
24-h avg O3 (per 1 5 ppb):
Minneapolis/St. Paul:
Pneumonia and COPD:
Lag 1: 1.060(1.033,1.087)
Pneumonia:
Lagl: 1.066(1.034,1.098)
COPD:
LagO: 1.045(0.995,1.067)
Birmingham:
Pneumonia and COPD:
Lag 2: 1.006(0.971,1.042)
24-h avg O3 (per 28 ppb):
Poisson GLM:
Pneumonia:
LagO: 0.93(0.83,1.05)
COPD:
LagO: 0.94(0.83,1.08)
Ischemic heart disease:
LagO: 1.01(0.94,1.09)
Dysrhythmias:
Lag 0:0. 98 (0.86, 1.12)
Heart failure:
LagO: 1.00(0.91,1.09)
Stroke:
LagO: 0.99(0.90,1.09)
Schwartz et al. (1996)
Cleveland, OH
Apr-Oct 1988-1990
Total respiratory
admissions for age
65+ years. Time-series
study.
l-hmaxO3:
56 ppb
IQR 28
PM10, SO2
1-2 Poisson GLM with sinusoids;
co-adjustment. Results
available only for warm
season. Ozone and PM10
both significant predictors of
outcome. No two-pollutant
models reported.
1-h max O3 (per 100
1.09(1.02,1.16)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
X
United States (cont'd)
Gwynn and Thurston
(2001)
New York City
1988-1990
Gwynn et al. (2000)
Buffalo, NY
May 1988-Octl990
Respiratory admissions
for all ages, stratified
by race and insurance
status. Time-series
study.
24-h avg O3:
22.1 ppb
IQR14.1
Maximum 80.7
42~, PM1(
Total respiratory 24-h avg O3
admissions for all ages. 26.2 ppb
Time-series study. IQR 14.8
PM10, S042-,
H+, COH, CO,
N02, S02
0,1,2,3
GLM with high-pass filter.
Ozone associated with
respiratory admissions;
effects larger for nonwhites
and for those uninsured or on
medicaid.
Used Poisson with GAM
default convergence criteria
for control of temperature;
moving average control for
time. Ozone significant
predictor of outcome. No
two-pollutant models
reported.
24-h avg O3 (per 58.6 ppb):
White:
1.032(0.987,1.079)
Non-white:
1.122(1.074,1.172)
Uninsured:
1.138(1.084,1.194)
24-h avg O3 (per 14.8 ppb):
Lagl: 1.029(1.013,1.045)
Weisel et al. (2002)
New Jersey
May-Aug 1995
Asthma admissions for l-hmaxO3; Pollen, spores
all ages. Time-series 5-h avg O3
study. (10 a.m.-3 p.m.);
and 8-h avg O3
(2p.m.-10p.m.)
analyzed.
Levels not
reported.
0,1,2,3 No control for time, but
authors report no
autocorrelation, which
alleviates concerns about
lack of control. Significant
O3 effects reported after
adjusting for potential
confounding by pollen.
Slope estimate
(admissions/day /ppb):
5-h avg O3 and 8-h avg O3:
O3 only model:
Lag 2: 0.099, p = 0.057
All three O3 indices:
O3 with pollen model:
Lag 2: 0.1 l,p = 0.033
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Canada
Burnett etal. (1997a)
16 Canadian cities
1981-1991
X
Burnett etal. (1995)
168 Hospitals in
Ontario, Canada
1983-1988
Total respiratory
admissions for all ages,
age <65 years and
65+ years. Time-series
study.
Respiratory and
cardiovascular
admissions for all
ages and within age
strata. Study focused
mainly on testing for
sulfate effects. Time-
series study.
l-hmaxO3:
All year:
31 ppb
95th % 60
Mean range
across cities:
26-38 ppb
95th % 45-84
Jan-Mar:
26 ppb
Apr-Jun:
40 ppb
Jul-Sep:
38 ppb
Oct-Dec:
21 ppb
9.6% of O3 data
missing.
l-hmaxO3:
36.3 ppb
SO2, NO2,
CO, coefficient
of haze
0,1,2,0-1,0-2,
1-2
so/-
Poisson GLM with co-
adjustment. Results stratified
by season. Significant O3
effect observed in warm
season only. No O3 effects
on control outcomes. Results
consistent across cities.
GLM with pre-adjustment of
outcome variables. Results
stratified by season. Authors
report that O3 associated with
respiratory admission in
warm season only.
1-h max O3 (per 30 ppb):
All ages:
Jan-Mar:
Lag 1: 0.994(0.964,1.025)
Apr-Jun:
Lagl: 1.042(1.012,1.073)
Jul-Sep:
Lag 1: 1.050(1.026,1.074)
Oct-Dec:
Lagl: 1.028(0.998,1.059)
No quantitative results
presented for O3.
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
oo
Reference, Study
Location and Period
Canada (cont'd)
Burnett etal. (1997b)
Toronto, Ontario,
Canada
Summers 1992-1994
Burnett etal. (1999)
Toronto, Ontario,
Canada
1980-1994
Outcomes and Design Mean O3 Levels
Unscheduled l-hmaxO3:
respiratory and 41.2 ppb
cardiovascular IQR 22
admissions for all ages.
Time-series study.
Cause-specific 24-h avg O3:
respiratory and 19. 5 ppb
cardiovascular IQR 1 9
admissions for all ages.
Cause categories
included asthma,
COPD, respiratory
infections, heart failure,
ischemic heart disease,
and cerebrovascular
disease. Time-series
study.
Copollutants
Considered
PM25,PM10,
H+, SO/-, S02,
N02, CO,
coefficient
of haze
Estimated
PM2.5,PM10,
PM10.25,CO,
N02, S02
Lag Structure
Examined
0,1,2,3,4,
2 to 5 multiday
periods lagged
1 to 4 days
0,1,2,0-1,0-2,
1-2,1-3,2-3,
2-4
Method, Findings,
Interpretation
Poisson GLM with co-
adjustment. Results stratified
by season. Ozone and
coefficient of haze strongest
predictors of outcomes.
Ozone effects on both
outcomes were robust to PM.
PM effects were not robust to
03.
Poisson GAM with LOESS
pre-filter applied to pollution
and hospitalization data.
Ozone effects seen for
respiratory outcomes only.
Ozone effect robust to PM;
not vice versa. No seasonal
stratification.
Effects
(Relative Risk and 95% CI)
12-h avg O3 (8 a.m.-8 p.m.)
(per 11.5 ppb):
Models adjusted for
temperature and dewpoint:
Respiratory :
Lag 1-3: 1.064(1.039,1.090)
Cardiovascular:
Lag 2-4: 1.074(1.035,1.115)
24-h avg O3 (per 19.5 ppb):
Asthma:
Lag 1-3: 1.063(1.036,1.091)
COPD:
Lag 2-4: 1.073(1.038,1.107)
Respiratory infection:
Lag 1-2: 1.044(1.024,1.065)
Burnett etal. (2001)
Toronto, Ontario,
Canada
1980-1994
Acute respiratory
disease admissions for
age <2 years. Time-
series study.
1-h max O3: Estimated 0, 1, 2, 3, 4, Poisson GAM with LOESS
PM25, PM10, 5, 0-4 pre-filter applied to pollution
Summer: PM10_2 5, CO, and hospitalization data.
45.2 ppb NO2, SO2 Sensitivity analyses using co-
IQR 25 adjustment. Results stratified
by season. Ozone effects
significant only in summer.
Ozone effect robust to PM;
not vice versa.
1-h max O3 (per 45.2 ppb):
Summer:
O3 only model:
Lag 0-4: 1.348(1.193,1.524)
O3 with PM2 5 model:
Lag 0-4: 1.330(1.131,1.565)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
Reference, Study
Location and Period
Canada (cont'd)
Lin et al. (2003)
Toronto, Ontario,
Canada
1981-1993
Outcomes and Design
Asthma admission for
age 6-12 years.
Case-crossover design.
Mean O3 Levels
l-hmaxO3:
30ppb
IQR20
Copollutants
Considered
CO, SO2, NO2
Lag Structure
Examined
0,0-1,0-2,0-3,
0-4, 0-5, 0-6
Method, Findings,
Interpretation
Conditional logistic
regression model analysis.
No O3 effects observed.
Positive relations to CO, SO2
Effects
(Relative Risk and 95% CI)
1-h max O3 (per 20 ppb):
Odds ratios:
Males:
Fung et al. (2005)
Windsor, Ontario,
Canada
Apr 1995-Dec 2000
Luginaah et al. (2005)
Windsor, Ontario,
Canada
Apr 1995-Dec 2000
Cardiovascular hospital 1 -h max O3:
admissions for age 39.3 ppb
<65 and >65 years. SD 21.4
Time-series study. Range 1-129
Respiratory hospital
admissions by gender
for all ages and age
0-14, 15-64, and
65+ years. Time-series
study.
39.3 ppb
SD21.4
Range 1-129
NO2, SO2, CO,
PM10,
coefficient
of haze, total
reduced sulfur
compounds
NO2, SO2, CO,
PM10,
coefficient
of haze, total
reduced sulfur
compounds
and NO, observed.
0,0-1, 0-2 Conducted both time-series
analysis using Poisson GLM
with natural splines.
Strongest effect observed for
SO2 in individuals aged
>65 years. No associations
were found any other
pollutant, including O3.
0,0-1, 0-2 Conducted both time-series
analysis using Poisson GLM
with natural splines and
bidirectional case-crossover
analysis using conditional
logistic regression models.
For case-crossover analysis,
control periods selected 2
weeks before and after the
case period. Results were
consistent for the time-series
and case-crossover analyses.
Significant associations were
found for all pollutants
except O3 and total reduced
sulfur compounds.
LagO: 0.96(0.88,1.04)
Females:
LagO: 0.86(0.78,1.04)
1-h max O3 (per 29 ppb):
Age <65 years:
LagO: 0.999(0.913,1.093)
Lag 0-2: 1.042(0.923,1.177)
Age >65 years:
LagO: 0.974(0.924,1.027)
Lag 0-2: 1.014(0.941,1.092)
1-h max O3 (per 29 ppb):
All ages:
Time-series analysis:
Males:
LagO: 1.04(0.92,1.17)
Females:
LagO: 0.95(0.82,1.10)
Case-crossover analysis:
Males:
LagO: 1.06(0.93,1.22)
Females:
LagO: 1.01(0.77,1.34)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
Reference, Study
Location and Period
Canada (cont'd)
Lin et al. (2004)
Vancouver, British
Columbia, Canada
1987-1998
Outcomes and Design
Asthma admissions for
age 6-12 years. Time-
series study.
Mean O3 Levels
l-hmaxO3:
28.02 ppb
SD11.54
IQR 14.81
Copollutants
Considered
CO, SO2, NO2
Lag Structure
Examined
0,0-1,0-2,0-3,
0-4, 0-5, 0-6
Method, Findings,
Interpretation
Poisson GAM with LOESS
(using default convergence
criteria). Repeated analysis
with natural cubic splines
Effects
(Relative Risk and 95% CI)
1-h max O3 (per 14.8 ppb):
Males:
Low SES:
Yang et al. (2003)
Vancouver, British
Columbia, Canada
Jan 1986-Dec 1998
Daily respiratory
admissions in children
aged <3 years and
adults aged 65+ years.
Bidirectional
case-crossover.
Conditional logistic
regression.
24-h avg O3:
13.41 ppb
SD6.61
IQR 9.74
CO, N02, S02:
coefficient
of haze
1,2,3,4,5
using 1,000 iterations with
convergence criteria 10"15.
Results were similar for both
analyses. NO2 exposure
associated for males in low
SES but not high. No
association for CO and O3 in
either SES group.
Used bidirectional case-
crossover analysis,
comparing air pollution on
day of admission to levels
1 week before and after. SES
evaluated. Ozone was
positively associated with
respiratory hospital
admissions among young
children and the elderly.
Lag 1: 0.85(0.76,0.94)
High SES:
Lagl: 0.93(0.83,1.04)
Females:
Low SES:
Lagl: 1.11(0.97,1.28)
High SES:
Lag 1: 0.91 (0.78, 1.05)
24-h avg O3 (per 9.74 ppb):
Odds ratios:
Age <3 years:
Lag 4: 1.22(1.15,1.30)
Age 65+ years:
Lag 4: 1.13(1.09,1.18)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Europe
Anderson et al. (1997)
Five European cities:
London, Paris,
Amsterdam,
Rotterdam, Barcelona
Study periods vary by
city, ranging from
1977-1992
> Atkinson etal. (2001)
X Eight European cities:
i Barcelona,
^ Birmingham, London
Milan, Netherlands,
Paris, Rome, and
Stockholm
Study periods vary by
city, ranging from
early to middle 1990s
Emergency COPD
admissions for all ages.
Each city analyzed
previously by
individual teams.
Results combined here
viameta-analysis.
Time-series study.
Total respiratory,
asthma, and COPD
admissions for all ages
and age 0-14,
15-64 and 65+years.
Time-series study.
l-hmaxO3:
Median range
across five cities:
All year:
36-77 ug/m3
Warm season:
48-91 ug/m3
Cool season:
20-64 ug/m3
8-h max O3:
Mean range:
26.0 ug/m3
(Rome)
to 66.6 ug/m3
(Stockholm)
TSP, SO2, NO2,
BS
0,1,2,3,4,5
PM10,N02f
SO2, CO
N/A
Poisson GLM using APHEA
methodology. Results
stratified by season.
Ozone most consistent and
significant predictor of
admissions. Warm-season
effect larger.
Study focused on PM10
effects. Copollutants
included only as effect
modifiers. No direct O3
results shown. Ozone
appeared to modify the PM10
effect on respiratory
admissions for persons over
age 64 years of age.
1-h max O3 (per 50 ug/m3):
Weighted mean effect across
five cities (best lag selected
for each city):
All year:
1.03(1.01,1.05)
Warm season:
1.03(1.01,1.05)
Cool season:
1.01(0.98,1.05)
No results presented for O3.
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Europe (cont'd)
Le Tertre et al.
(2002b)
Eight European cities:
Barcelona,
Birmingham, London
Milan, Netherlands,
Paris, Rome, and
Stockholm
Study periods vary by
city, ranging from
early to middle 1990s
Wong et al. (2002)
r*j London, England
Cj 1992-1994
0-, Hong Kong
^ 1995-1997
Outcomes and Design
Total cardiovascular,
ischemic heart disease,
and stroke admissions
for all ages and age
0-64 and 65+ years.
Time-series study.
Total respiratory
(>64 years), asthma
(15-64 years), total
cardiovascular (all
ages), and ischemic
heart disease (all ages)
admissions. Time-
series study.
Mean O3 Levels
8-h max O3:
Mean range:
26.0 ug/m3
(Rome) to 66.6
ug/m3
(Stockholm)
8-h max O3:
London:
All year:
34.9 ug/m3
SD23.1
Warm season:
45.3 ug/m3
Cool season:
24.0 ug/m3
Hong Kong:
33.5 ug/m3
SD23.0
Warm season:
32.0 ug/m3
Cool season:
35.1 ug/m3
Copollutants Lag Structure Method, Findings,
Considered Examined Interpretation
PM10,BS,NO2, N/A Main focus on PM10 and BS.
SO2, CO Gaseous copollutants
evaluated as effect modifiers.
Greater PM10 effects seen in
cities with lower annual O3
levels. No risk estimates
presented for O3.
PM10, NO2, SO2 0, 1 , 2, 3 Poisson GAM with default
0-1 convergence criteria. Most
consistent associations found
with total respiratory
admissions. Ozone
associated with total
respiratory admissions in
all-year and warm-season
analyses in both cities.
Associations in cool-season
analyses observed only in
Hong Kong. All-year O3
effects robust to copollutants.
Effects
(Relative Risk and 95% CI)
No results presented for O3.
8-h max O3 (per 10 ug/m3):
Total respiratory:
London:
All year:
Lag 0-1: 1.008(1.
Warm season:
Lag 0-1: 1.010(1.
Cool season:
Lag 0-1: 1.002(0.
Hong Kong:
All year:
Lag 0-1: 1.008(1.
Warm season:
Lag 0-1: 1.008(1.
Cool season:
Lag 0-1: 1.010(1.
002,1.014)
003,1.017)
993,1.012)
003,1.013)
002,1.014)
002,1.017)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Outcomes and Design Mean O3 Levels
Copollutants
Considered
Lag Structure
Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Europe (cont'd)
Anderson et al. (1998)
London, England
1987-1992
X Atkinson etal. (1999b)
i London, England
So 1992-1994
Admissions for asthma 8-h max O,:
in all ages and
age 0-14, 15-64,
and 65+ years.
Time-series study.
Total and cause-
specific respiratory and
cardiovascular
admissions in all ages
and in all ages and
age 0-14, 15-64, and
65+ years. Time-series
study.
15.5 ppb
IQR13
l-hmaxO3:
20.6 ppb
IQR16
8-h max O3:
17.5 ppb
SD11.5
SO2, NO2, 0,1,2,0-1, Poisson GLM using APHEA
BS, pollens 0-2 method; co-adjustment.
Ozone significantly
associated with asthma
admissions in the warm
season for all ages and for
age 15-64 years. Warm-
season O3 effect robust in
two-pollutant models.
Inverse associations observed
in the cool season for some
age groups.
NO2, SO2, CO, 0, 1, 2, 3, 0-1, Poisson GLM using APHEA
PM10, BS 0-2,0-3 methodology. No significant
associations seen between O3
and respiratory admissions.
Ozone was positively
associated with total
cardiovascular admissions in
age 65+ years. Seasonal
analyses were not conducted.
8-h max O3 (per 10 ppb):
All ages:
Warm season:
Lag 1: 1.022(1.006,1.038)
Cool season:
Lagl: 0.968(0.946,0.992)
8-h max O3 (per 25.7 ppb):
All ages:
Total respiratory:
Lagl: 1.012(0.990,1.035)
Total cardiovascular:
Lag 2: 1.023(1.002,1.046)
Ponce de Leon et al.
(1996)
London, England
Apr 1987-Feb 1992
Total respiratory
admissions in several
age strata: all ages,
0-14, 15-64, 65+years.
Time-series study.
8-h avg O3
(9 a.m.-5 p.m.):
15.6 ppb
SD12
IQR14
BS, SO2, NO2
0,1,2,0-1,0-2,
0-3,
Poisson GLM using APHEA
co-adjustment methodology.
Ozone significant predictor
overall. Effect larger and
more significant in warm
season. Effect robust to
copollutants. Effects varied
by age.
All ages:
All year:
8-h avg O3 (per 26 ppb):
Lag 1: 1.029(1.011,1.048)
Warm season:
8-h avg O3 (per 29 ppb):
Lag 1: 1.048(1.025,1.073)
Cool season:
8-h avg O3 (per 20 ppb):
Lag 1: 0.996(0.972,1.021)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
Reference, Study
Location and Period
Europe (cont'd)
Poloniecki et al.
(1997)
London, England
Apr 1987-Mar 1994
Outcomes and Design
Cause-specific and total
circulatory admissions
for all ages. Time-
series study.
Mean O3 Levels
8-h avg O3
(9 a.m. -5 p.m.):
Median 1 3 ppb
Range 0-94
Copollutants
Considered
BS, NO2,
S02, CO
Lag Structure
Examined
0
Method, Findings,
Interpretation
Poisson regression using
APHEA methodology. No
association was found
between O3 and circulatory
Effects
(Relative Risk and 95% CI)
8-h avg O3 (per 25 ppb):
Total circulatory:
All year:
Prescott et al. (1998)
Edinburgh, Scotland
1992-1995
Total respiratory and
cardiovascular
admissions for age
<65 years and
65+ years. Time-series
study.
24-h avg O3:
14.5 ppb
Range 1-37
BS, PM10, N02
S02, CO
0,1,1-3
diseases in all-year analyses.
Results from acute MI
suggest potential seasonal
effect.
Poisson GLM, month dummy
variables; co-adjustment. No
O3 or other pollution effects
on respiratory admissions.
Significant inverse
association of O3 with
cardiac admissions in older
age group. Very low O3
concentrations.
0.9726(0.9436,1.0046)
Acute MI:
All year:
0.9825(0.9534,1.0142)
Warm season:
1.0126(0.9560,1.0228)
Cool season:
0.9680(0.9208,1.0202)
24-h avg O3 (per 10 ppb):
Respiratory:
Age <65 years:
Lag 1-3: 0.971(0.885,1.068)
Age 65+ years:
Lag 1-3: 1.009(0.916,1.111)
Cardiovascular:
Age <65 years:
Lag 1-3: 1.041(0.946,1.144)
Age 65+ years:
Lag 1-3: 0.941(0.886,0.999)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
a\
Reference, Study
Location and Period
Europe (cont'd)
Schouten et al. (1996)
Amsterdam and
Rotterdam,
the Netherlands
1977-1989
Hagen et al. (2000)
Drammen, Norway
Nov 1994-Dec 1997
Oftedal et al. (2003)
Drammen, Norway
1995-2000
Outcomes and Design
Unscheduled total
respiratory, asthma, and
COPD admissions in
all ages. Time-series
study.
Total respiratory
admissions for all ages.
Time-series study.
Admissions for
respiratory disease.
Time-series study.
Copollutants
Mean O3 Levels Considered
8-h max O3: SO2, NO2, BS
Amsterdam:
Summer:
86 ug/m3
5th % to 95th %
28-152
Winter:
53 ug/m3
5th % to 95th %
3-104
Rotterdam:
Summer:
81 ug/m3
5th % to 95th %
25-199
Winter:
45 ug/m3
5th % to 95th %
3-96
24-h avg O3: PM10,NO2,
44.48 ug/m3 SO2, benzene,
SD 18.40 toluene,
IQR 26.29 formaldehyde
24-h avg O3: Benzene,
44.6 ug/m3 formaldehyde,
SD19.2 toluene, PM10,
IQR 26. 9 NO2, SO2
Lag Structure Method, Findings,
Examined Interpretation
0,1,2,0-1, 0-2, Poisson GLM using APHEA
0-3, 0-4, 0-5 methodology; co-adjustment.
No consistent O3 effects.
Concern regarding multiple
comparisons.
0 Poisson GAM with partial
splines; co-adjustment.
Single and multipollutant
models evaluated. No O3
effects. Ozone levels low
and cycles may not have
been adequately controlled.
0 Benzene had the strongest
association.
Effects
(Relative Risk and 95% CI)
1-h max O3 (per 100 ug/m3):
Amsterdam and Rotterdam:
Total respiratory, all ages:
Summer:
Lag 2: 1.051(1.029,1.073)
Winter:
Lag 2: 0.976(0.951,1.002)
24-h avg O3
(per 26.29 ug/m3):
LagO: 0.964(0.899,1.033)
24-h avg O3 (per 26.9 ug/m3):
0.996(0.942,1.053)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
Oi
Reference, Study
Location and Period
Europe (cont'd)
Ponka and Virtanen
(1996)
Helsinki, Finland
1987-1989
Ballester et al. (2001)
Valencia, Spain
1994-1996
Latin America
Gouveia and Fletcher
(2000a)
Sao Paulo, Brazil
Nov 1992-Sep 1994
Outcomes and Design
Asthma admissions for
age 0-14 years and
15-64 years. Time-
series study.
Emergency total
cardiovascular
admissions for all ages.
Time-series study.
Total respiratory,
pneumonia, and asthma
admissions for age
<5 years. Time-series
study.
Copollutants
Mean O3 Levels Considered
8-h max O3: TSP, SO2, NO2
All year:
22 ug/m3
Winter:
20 ug/m3
Spring:
30 ug/m3
Summer:
23 ug/m3
Fall:
15 ug/m3
8-h max O3: SO2, NO2,
23 ppb CO, BS
Range 5-64
l-hmaxO3: PM10,NO2,
63.4 ug/m3 S02, CO
SD38.1
IQR 50.3
Lag Structure Method, Findings,
Examined Interpretation
0, 1, 2, 3, 4, 5 Poisson GLM using APHEA
methodology. Reported
significant O3 effect for age
0-14 years, but also for
control (digestive disease)
conditions. Ozone levels
very low.
0, 1, 2, 3, 4, 5 Poisson GLM using APHEA
methodology. Results
stratified by season. No O3
effects.
0,1,2 Poisson GLM with co-
adjustment using sine/cosine
waves. Significant O3 effects
on total respiratory and
pneumonia admissions.
Ozone effects fairly robust
to NO2 and PM10.
Effects
(Relative Risk and 95% CI)
Not quantitatively useful.
8-h max O3 (per 5 ppb):
Lag 2: 0.99(0.97,1.01)
1-h max O3 (per 1 19.6 ug/m3):
Total respiratory:
LagO: 1.054(1.003,1.107)
Pneumonia:
LagO: 1.076(1.014,1.142)
Asthma:
Lag 2: 1.011(0.899,1.136)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
X
Reference, Study
Location and Period
Australia
Morgan etal. (1998a)
Sydney, Australia
1990-1994
Petroeschevsky et al.
(2001)
Brisbane, Australia
1987-1994
Asia
Lee et al. (2002)
Seoul, Korea
Dec 1997-Dec 1999
Outcomes and Design
Admissions for asthma
(age 1-14 years,
1 5-64 years), COPD
(age 65+ years), and
heart disease (all ages,
0-64 years, 65+ years).
Time-series study.
Unscheduled asthma,
total respiratory and
total cardiovascular
admissions in several
age strata: all ages,
0-4,5-14,15-64,
65+ years. Time-
series study.
Asthma admissions
for age <1 5 years.
Time-series study.
Mean O3 Levels
l-hmaxO3:
25ppb
SD13
IQR11
8-h avg O3
(10 a.m.-6p.m.):
All year:
19.0ppb
Range 1.7-64.7
Winter:
16.1 ppb
Spring:
23.3 ppb
Summer:
19. 9 ppb
Fall:
16.7 ppb
l-hmaxO3:
36.0 ppb
SD 18.6
IQR21.7
Copollutants Lag Structure Method, Findings,
Considered Examined Interpretation
Bscatter, NO2 0, 1 , 2, 0-1 , Poisson with GEE.
0-2 No significant effects of O3
in single or multipollutant
models.
Bscalter, SO2, 0, 1 , 2, 3, 0-2, Poisson GLM using APHEA
NO2 0-4 co-adjustment methodology.
Seasonal variations in O3
effects also investigated.
Only a few significant
interactions observed between
O3 and season. Strong O3
effects observed in all-year
models. Ozone significantly
related to asthma and total
respiratory admissions, not
for cardiac admissions.
Effects varied by age group.
Ozone effects robust to
copollutants.
SO2, NO2, CO, 0, 1 , 2, 3, 4, Poisson GAM using default
PM10 0-1,1-2,2-3, convergence criteria. Ozone
3-4 associated with asthma
admissions in single- and
two-pollutant models.
Effects
(Relative Risk and 95% CI)
l-hmaxO3(per28ppb):
Asthma, age 1-14 years:
Lagl: 0.975(0.932,1.019)
Asthma, age 15-64 years:
LagO: 1.025(0.975,1.078)
COPD, age 65+ years:
LagO: 1.010(0.960,1.062)
Heart disease, all ages:
LagO: 1.012(0.990,1.035)
8-h avg O3 (per 10 ppb):
All year, all ages:
Total respiratory:
Lag 2: 1.023(1.003,1.043)
Asthma:
Lag 0-4: 1.090(1.042,1.141)
Total cardiovascular:
Lag 3: 0.987(0.971,1.002)
1-h max O3 (per 21 .7 ppb):
O3 only model:
Lag 1: 1.12(1.07,1.16)
O3 with PM10 model:
Lag 1: 1.10(1.05,1.15)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Copollutants Lag Structure
Outcomes and Design Mean O3 Levels Considered Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Asia (cont'd)
Chang et al. (2005)
Taipei, Taiwan
1997-2001
X
OO
Yang et al. (2004a)
Kaohsiung, Taiwan
1997-2000
Total cardiovascular
hospital admissions for
all ages. Cool days
(<20 °C) and warm
days (>20 °C) were
evaluated. Case-
crossover approach.
Total cardiovascular
hospital admissions for
all ages. Cool days
(<25 °C) and warm
days (>25 °C) were
evaluated. Case-
crossover approach.
24-h avg O3:
19.74ppb
IQR 10.87
Range 2.30-53.93
PM10, S02
NO2, CO
24-h avg O3:
25.02ppb
IQR 21.20
Range 1.25-83.00
PM10, S02,
NO2, CO
0-2 Conditional logistic
regression. All
cardiovascular admissions
chosen because similar risks
have been observed for major
subcategories and combining
counts provide greater power.
Subtropical climate in Taipei.
In the analysis of warm days
only, all pollutants except
SO2 were associated with
cardiovascular admissions.
Ozone effect slightly
diminished in two-pollutant
model adjusting for PM10.
0-2 Conditional logistic
regression. All pollutants
except SO2 associated with
cardiovascular admissions on
warm days. On cool days, O3
effect was diminished.
Ozone effect was robust to
adjustment to copollutants.
Results from tropical city
may restrict generalization to
other locations.
24-h avg O3 (per 10.87 ppb):
Odds ratios:
O3 only models:
Warm: 1.189(1.154,1.225)
Cool: 1.073(1.022,1.127)
O3 with PM10 models:
Warm: 1.066(1.038,1.094)
Cool: 0.980(0.924,1.039)
24-h avg O3 (per 21.20 ppb):
Odds ratios:
O3 only models:
Warm: 1.351(1.279,1.427)
Cool: 1.057(0.962,1.162)
O3 with PM10 models:
Warm: 1.308(1.219,1.404)
Cool: 0.820 (0.732, 0.912)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Asia (cont'd)
Tsai et al. (2003a)
Kaohsiung, Taiwan
1997-2000
Outcomes and Design
Stroke admissions
(subarachnoid
hemorrhagic stroke,
primary intracerebral
Mean O3 Levels
24-h avg O3:
25.02ppb
IQR21.20
Range 1.25-83. 00
Copollutants
Considered
PM10, S02,
NO2, CO
Lag Structure
Examined
0-2
Method, Findings,
Interpretation
Conditional logistic
regression. Warm day
associations were positive
while cool days were
Effects
(Relative Risk and 95% CI)
24-h avg O3 (per 21 .20 ppb):
Odds ratios:
X
VO
Wongetal. (1999a)
Hong Kong
1994-1995
hemorrhage, ischemic
stroke, and others) for
all ages. Cool days
(<20 °C) and warm
days (>20 °C) were
evaluated. Case-
crossover approach.
Total and cause-
specific respiratory and
cardiovascular
admissions in several
age strata: all ages,
0-4, 5-64, 65+ years.
Time-series study.
8-h max O3:
20.2 ug/m3
Median 24.15
IQR31.63
N02, S02.
PM,n
0,1,2,3,4,5,
0-1,0-2,0-3,
0-4, 0-5
generally negative with some
positive associations. Ozone
effect was robust to
adjustment for SO2 and CO,
butnotPM,n.
Poisson GLM using APHEA
methodology. Ozone
significantly associated with
total and cause specific
respiratory and cardiac
outcomes. Ozone results
robust to adjustment for high
PM10, but not high NO2.
Effects of O3 persisted in
cold season.
O3 only models:
Primary intracerebral
hemorrhage:
Warm: 1.20(1.06,1.35)
Cool: 0.57(0.24,1.34)
Ischemic stroke:
Warm: 1.15(1.07,1.23)
Cool: 0.88(0.50,1.53)
O3 with PM10 models:
Primary intracerebral
hemorrhage:
Warm: 0.98(0.85,1.14)
Ischemic stroke:
Warm: 0.96(0.88,1.05)
8-h max O3 (per 10 ug/m3):
All ages:
Total respiratory:
Lag 0-3: 1.022(1.015,1.029)
Total cardiovascular:
Lag 0-5: 1.013(1.005,1.021)
-------
Table AX7-4 (cont'd). Effects of O3 on Daily Hospital Admissions
Reference, Study
Location and Period
Copollutants Lag Structure
Outcomes and Design Mean O3 Levels Considered Examined
Method, Findings,
Interpretation
Effects
(Relative Risk and 95% CI)
Asia (cont'd)
Wongetal. (1999b)
Hong Kong
Jan 1995-Jun 1997
Cause-specific
cardiovascular
admissions for age
>65 years. Time-series
study.
8-h avg O3:
Warm season:
31.2|ig/m3
Cool season:
34.8 ng/m3
N02, S02,
respirable PM
0,1,2,3,4,5,
0-1,0-2,0-3,
0-4, 0-5
GLM with sinusoids;
co-adjustment. Ozone
significantly associated with
total and cause-specific
cardiovascular admissions in
cool season only, when O3
levels are higher in Hong
Kong. Details missing in
brief report.
03 (per 50 |ig/m3):
O3 with NO2 models:
Total cardiovascular:
All year:
Lag 0-1: 1.03(1.00,1.07)
Warm season:
Lag 0-1: 1.01 (0.95,1.07)
Cool season:
Lag 0-1: 1.08(1.02,1.14)
X
-------
Table AX7-5. Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Meta-analysis
Bell et al. (2005)
Various U.S. and
non-U.S. cities
Varying study periods
Outcome Measure Mean O3 Levels
All cause; Not applicable.
cardiovascular;
respiratory; all
ages; age 64+ or
65+ years.
Copollutants
Considered
Various PM
indices
Lag Structure
Reported
0,1,2, or 0-1
Method/Design
Meta-analysis.
Bayesian hierarchical
model; included up to
144 estimates from
39 studies.
Effect Estimates
24-h avg O3 (per
Posterior means:
All cause:
lOppb):
X
Risk estimates obtained
for yearly vs.
warm-season data;
cause-specific; PM
adjustment; U.S. vs.
non-U.S.;various lags;
and GAM vs. non-
GAM.
Comparisons with
NMMAPS estimates
(Bell etal., 2004).
All year: 0.87% (0.55,1.18)
Warm: 1.50% (0.72, 2.29)
Cardiovascular:
All year: 1.11% (0.68,1.53)
Warm: 2.45% (0.88, 4.10)
Respiratory:
All year: 0.47% (-0.51,
1.47)
O3 with PM model:
All cause:
All year: 0.97% (-0.03,
1.98)
Meta-analysis results were
consistently larger than those
from NMMAPS. In addition,
heterogeneity of city-specific
estimates
in the meta-analysis were
larger compared to
NMMAPS. These findings
indicate possible publication
bias.
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
to
Reference, Study
Location and Period
Meta-analysis (cont'd)
Ito et al. (2005)
Various U.S. and
Outcome Measure Mean O3 Levels
All cause Not applicable.
Copollutants
Considered
Various PM
indices
Lag Structure
Reported
Reported lags
up to 3-day lags
Method/Design
Meta-analysis.
DerSimonian-Laird
Effect Estimates
24-h avg O3 (per 20 ppb):
non-U.S. cities
Varying study periods
approach; included up
to 43 estimates from
38 studies.
Risk estimates obtained
for yearly vs.
warm-season data; PM
adjustment; correction
for asymmetry in
funnel plot; and GAM
vs. non-GAM.
Seven U.S. cities time-
series study with
various sensitivity
analyses.
O3 only model:
All year: 1.6% (1.1, 2.0)
Warm: 3.5% (2.1, 4.9)
O3 with PM model:
All year: 1.5% (0.8, 2.2)
Non-GAM-affected:
All year: 1.4% (0.8, 2.0)
GAM-affected:
All year: 1.9% (1.0, 2.8)
Correction for funnel plot
asymmetry:
All year: 1.4% (0.9, 1.9)
Seven U.S. cities analysis
found that including PM
in the model did not
substantially reduce the O3
risk estimates. However,
differences in the weather
adjustment model could
result in a 2-fold difference in
risk estimates.
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Meta-analysis (cont'd)
Levy et al. (2005)
Various U.S., Canadian
and European cities
Varying study periods
Copollutants
Outcome Measure Mean O3 Levels Considered
All cause Not applicable. PM10, PM2 5, SO2,
N02,CO
Lag Structure
Reported Method/Design
0, 1-2 Meta-analysis.
Empirical Bayes
metaregression;
included up to
48 estimates from
28 studies.
Effect Estimates
1-h max O3 (per 10 ug/m3):
O3 only model:
All year: 0.21% (0.16, 0.26)
Warm: 0.43% (0.29, 0.56)
Cool: -0.02 (-0.17, 0.14)
X
Risk estimates obtained
by season; copollutant
adjustment; North
America vs. Europe;
various lags;
temperature
adjustment; GAM vs.
non-GAM; and annual
mean O3.
Examined relationship
between O3
personal exposure and
ambient concentrations
using cooling
degree days and
residential central air-
conditioning
prevalence.
Non-GAM-affected:
All year: 0.23% (0.15, 0.31)
GAM-affected:
All year: 0.20% (0.13, 0.27)
Nonlinear specification of
temperature:
All year: 0.20% (0.13, 0.27)
Linear specification of
temperature:
All year: 0.06 (-0.03, 0.16)
In the metaregression,
air-conditioning prevalence
and lag time were the
strongest predictors
of between-study variability.
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
United States
Bell et al. (2004)
95 U.S. communities
1987-2000
Copollutants Lag Structure
Outcome Measure Mean O3 Levels Considered Reported Method/Design
All cause; 24-havgO3: PM10, PM25; two- 0,1,2,3,0-6 PoissonGLM;
cardiopulmonary; 26 ppb across the pollutant models Bayesian hierarchical
all ages; age 95 communities. model. Time-series
<65 years; study.
age 65-74 years; 55 communities
age >75 years with all-year data:
Median range:
14.38 ppb
(Newark, NJ)
to 37.30 ppb
(Bakersfield, CA)
40 communities
with warm-season
only data:
Median range:
20.41 ppb
(Portland, OR)
to 36. 15 ppb
(Memphis, TN)
Effect Estimates
24-h avg O3 (per 10 ppb):
Posterior means:
All cause, all ages:
All-available data:
LagO: 0.25% (0.12, 0.39)
Lag 0-6: 0.52% (0.27, 0.77)
Warm season:
LagO: 0.22% (0.08, 0.38)
Lag 0-6: 0.39% (0.1 3, 0.65)
All cause:
All-available data:
Age <65 years:
Lag 0-6: 0.50% (0.10, 0.92)
Age 65-74 years:
Lag 0-6: 0.70% (0.28, 1.12)
Age >75 years:
Lag 0-6: 0.52% (0.18, 0.87)
Cardiopulmonary, all ages:
All-available data:
Lag 0-6: 0.64% (0.31, 0.98)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
United States (cont'd)
Samet et al. (2000;
reanalysis Dominici et al.,
2003)
90 U.S. cities (80 U.S.
cities with O3 data)
1987-1994
Huang et al. (2005)
19 large U.S. cities
Jun-Sept 1987-1 994
Outcome Measure Mean O3 Levels
All cause; 24-h avg O3:
cardiopulmonary
Mean range:
Approximately
12 ppb
(Des Moines, IA)
to 36 ppb
(San Bernardino,
CA)
Cardiopulmonary 24-h avg O3:
Mean range:
Approximately
18 ppb (Oakland,
CA) to 56 ppb
(San Bernardino,
CA)
Daily range across
19 U.S. cities:
0-1 00 ppb
Copollutants Lag Structure
Considered Reported Method/Design Effect Estimates
PM10, NO2, SO2, 0,1,2 Poisson GAM, 24-h avg O3 (per 1 0 ppb):
CO; two-pollutant reanalyzed with
models stringent convergence Posterior means:
criteria; Poisson GLM.
Time-series study. All cause:
All available data:
Lag 0: 0.42% (0.22, 0.62)
Lagl: 0.19% (0.03, 0.35)
Summer:
Lagl: 0.51% (0.23, 0.78)
Winter:
Lagl: -0.53% (-1.10, 0.05)
PM10, PM2 5; two- 0, 1 , 2, 0-6 Poisson GLM; 24-h avg O3 (per 1 0 ppb):
pollutant models Bayesian hierarchical
model. Time-series Posterior means:
study.
Single-day lag models:
O3 only model:
LagO: 0.73% (0.27, 1.19)
O3 with PM10 model:
LagO: 0.74% (-0.33, 1.72)
Cumulative lag models:
O3 only model:
Lag 0-6: 1.25% (0.47, 2.03)
Model adjusted for heat
waves:
Lag 0-6: 1.11%(0.38, 1.86)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
United States (cont'd)
Schwartz (2005)
14 U.S. cities
1986-1993
Copollutants
Outcome Measure Mean O3 Levels Considered
All cause l-hmaxO3: PM10; two-
pollutant models
Median range:
35.1 ppb (Chicago,
IL) to 60.0 ppb
(Provo, UT)
Lag Structure
Reported Method/Design
0 Case-crossover
analysis; conditional
logical regression
controlled for
temperature using
nonlinear regression
splines and matching
Effect Estimates
1-h max O3 (per 10 ppb):
Analysis with temperature
regression splines:
All year:
0.19% (0.03, 0.35)
Warm season:
0.26% (0.07, 0.44)
Cold season:
0% (-0.27, 0.27)
X
Analysis with temperature
matched controls:
All year:
0.23% (0.01, 0.44)
Warm season:
0.37% (0.11,0.62)
Cold season:
-0.13% (-0.53, 0.28)
Kinney and Ozkaynak
(1991)
Los Angeles County, CA
1970-1979
All cause;
respiratory;
circulatory
Kinney etal. (1995)
Los Angeles County, CA
1985-1990
All cause
1-h max total
oxidants (Ox):
75 ppb
SD45
1-h max O3:
70 ppb
SD41
KM (particle
optical
reflectance), NO2,
SO2, CO;
multipollutant
models
PM10, NO2, SO2,
CO; two-pollutant
models
OLS (ordinary least
squares) on high-pass
filtered variables.
Time-series study.
Linear, log-linear, and
Poisson. Time-series
study.
All cause:
Multipollutant model:
Slope estimate:
0.030 deaths/ppb (SE 0.009),
p = 0.0005
Cardiovascular:
O3 only model:
Slope estimate:
0.023 deaths/ppb (SE 0.006),
p< 0.0001
1-h max O3 (per 143 ppb):
O3 only model:
2% (0, 5)
O3 with PM10 model:
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period Outcome Measure
United States (cont'd)
Ostro(1995) All cause
San Bernardino County and
Riverside County, CA
1980-1986
Ostro et al. (2000) All cause;
Coachella Valley, CA respiratory;
1989-1998 cardiovascular
Fairley (1999; reanalysis All cause;
Fairley, 2003) respiratory;
Santa Clara County, CA cardiovascular
1989-1996
Copollutants
Mean O3 Levels Considered
l-hmaxO3: PM25
140 ppb
Range 20-370
l-hmaxO3: PM10,PM25,
PM10.25, NO2, CO
Palm Springs:
67 ppb
Range 0-1 90
Indio:
62 ppb
Range 0-1 80
8-h max O3: PM10,PM25,
29 ppb PM10.2.5, SO/-,
SD 1 5 coefficient of
haze, NO3', NO2,
24-h avg O3: SO2; two-pollutant
16 ppb models
SD9
O3 ppb-h >60 ppb:
Levels not reported.
Lag Structure
Reported Method/Design
0 Autoregressive linear;
Poisson. Time-series
study.
0 Poisson GAM with
default convergence
criteria. Time-series
study.
0 Poisson GAM,
reanalyzed with
stringent convergence
criteria; Poisson GLM.
Time-series study.
Effect Estimates
1 -h max O3 (per 100 ppb):
Warm season:
2.0% (0.0, 5.0)
1-h max O3 (per 40 ppb):
All cause:
-l%(-4, 3)
Respiratory:
3% (-9, 16)
Cardiovascular:
-4% (-9,1)
GAM with stringent
convergence criteria:
All cause:
8-h max O3 (per 33 ppb):
3.1% (-0.3, 6. 6)
O3 ppb-h >60 ppb
(increment not reported):
3.8% (1.4, 6.3)
Cardiovascular:
8-h max O3 (per 33 ppb):
2.6% (-2.3, 7.8)
O3 ppb-h >60 ppb
(increment not reported):
4. 3% (0.4, 8.3)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study Copollutants Lag Structure
Location and Period Outcome Measure Mean O3 Levels Considered Reported Method/Design
United States (cont'd)
Gamble (1998) All cause; 24-h avg O3: PM10, NO2, SO2, 1-2 Poisson GLM. Time-
Dallas, TX respiratory; CO; two-pollutant series study.
1990-1994 cardiovascular; All year: models
cancer; other 22 ppb
Range 0-1 60
Summer:
30 ppb
Range 0-1 60
Winter:
12 ppb
^ Range 0-75
X
oo
Dockeryetal. (1992) All cause 24-h avg O3: PM10,PM25, 1 Poisson with GEE.
St. Louis, MO and Eastern SO42~, H+, NO2, Time-series study.
Tennessee St. Louis, MO: SO2
1985-1986 22.5 ppb
SD18.5
Eastern Tennessee:
23.0 ppb
SD11.4
Effect Estimates
All year:
24-h avg O3 (per 14.
All cause:
2.7% (0.6, 4.8)
Cardiovascular:
2.4% (-1.1, 6.0)
Summer:
24-h avg O3 (per 13.
All cause:
3. 5% (p< 0.05)
Cardiovascular:
3. 3% (p> 0.05)
Winter:
24-h avg O3 (per 7.7
All cause:
2.4% (p > 0.05)
Cardiovascular:
1.5%(p>0.05)
24-h avg O3 (per 20
St. Louis, MO:
0.6% (t = 0.38)
Eastern Tennessee:
-1.3%(t=-0.37)
7 ppb):
1 ppb):
ppb):
ug/m3):
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
United States (cont'd)
Ito and Thurston (1996)
Cook County, IL
1985-1990
Moolgavkar (2003)
Cook County, IL and Los
Angeles County, CA
1987-1995
Outcome Measure
All cause;
respiratory;
circulatory; cancer;
race/gender
subcategories
All cause;
cardiovascular
Copollutants Lag Structure
Mean O3 Levels Considered Reported Method/Design
l-hmaxO3: PM10, NO2, SO2, 0-1 Poisson GLM. Time-
38.1 ppb CO; two-pollutant series study.
SD 19.9 models
24-h avg O3: PM2 5, PM10, NO2, 0,1,2,3,4,5 Poisson GAM with
SO2, CO; two- default convergence
Cook County: pollutant models criteria. Time-series
Median 18 ppb study.
Range 0.2-67
Los Angeles
County:
Median 24 ppb
Range 0.6-77
Effect Estimates
1-h max O3 (per 100 ppb):
All cause:
O3 only model:
10% (6, 15)
O3 with PM10 model:
7% (1,12)
Circulatory (results given in
graphic format):
O3 only model:
12% (6, 20)
24-h avg O3 (per 100 ppb):
All cause, all year:
Cook County:
LagO: 1.4% (t= 6.3)
Los Angeles County
LagO: 0.4% (t = 2.3)
All cause, summer:
Cook County:
LagO: 2.9%(t= 7.2)
Los Angeles County
LagO: 1.0% (t = 2.8)
Cardiovascular, all year:
Cook County:
LagO: 1.8%(t=5.5)
Los Angeles County
LagO: 0.2%(t= 0.9)
Cardiovascular, summer:
Cook County:
LagO: 3.3%(t= 5.6)
Los Angeles County
LagO: 0.8%(t= 1.7)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
oo
o
Reference, Study
Location and Period
United States (cont'd)
Moolgavkar (2000)
Cook County, IL; Los
Angeles County, CA; and
Maricopa County, AZ
1987-1995
Lippmann et al. (2000;
reanalysis Ito, 2003, 2004)
Detroit, MI
1985-1990
1992-1994
Outcome Measure Mean O3 Levels
Cardiovascular; 24-h avg O3:
cerebro vascular;
COPD Cook County:
Median 1 8 ppb
Range 0.2-67
Los Angeles
County:
Median 24 ppb
Range 0.6-77
Maricopa County:
Median 25 ppb
Range 1-50
All cause; 24-h avg O3:
respiratory;
circulatory; 1985-1990:
cause-specific 20.9 ppb
IQR 12.0-27.5
1992-1994:
25 ppb
IQR 18-30
Copollutants
Considered
PM25, PM10, NO2,
SO2, CO; two- and
three-pollutant
models
PM10,PM25,
PM10.25,S042-,H+,
N02, S02, CO;
two-pollutant
models
Lag Structure
Reported Method/Design
0, 1, 2, 3, 4, 5 Poisson GAM with
default convergence
criteria. Time-series
study.
0, 1, 2, 3, 0-1, Poisson GAM,
0-2, 0-3 reanalyzed with
stringent convergence
criteria; Poisson GLM.
Time-series study.
Effect Estimates
24-h avg O3 (per 100 ppb):
Cook County:
Cardiovascular, all year:
LagO: 1.51% (0.78, 2.24)
COPD, all year:
LagO: 1.53% (-0.49, 3.55)
Los Angeles County and
Maricopa County:
O3 results not presented.
Noted as negative or small
and nonsignificant in all-year
and warm-season analyses.
Poisson GLM:
All lags and outcomes during
both study periods (n = 140):
24-h avg O3 (per 5th to
95th % increment):
Median 1.6%
Range -1.8-2. 6
1985-1990:
24-h avg O3 (per 36 ppb):
All cause:
LagO: 1.08% (-1.08, 3.30)
Circulatory:
LagO: 1.84% (-1.26, 5.04)
1992-1994
24-h avg O3 (per 28 ppb):
All cause:
LagO: 2.58% (-2.41, 7.82)
Circulatory:
LagO: 2.13% (-5.04, 9.85)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
oo
Reference, Study
Location and Period
United States (cont'd)
Lipfert et al. (2000a)
Seven counties in
Philadelphia, PA area
1991-1995
Outcome Measure
All cause;
respiratory;
cardiovascular;
all ages; age
65+ years; age
<65 years; various
subregional
boundaries
Mean O3 Levels
1-hmax O3:
44.76 ppb
SD 25.68
24-havg O3:
23. 44 ppb
SD 13.86
Copollutants
Considered
PM10,PM25,
PM10.25,S042-,
NO3-, other PM
indices, NO2, SO2,
CO; two-pollutant
models
Lag Structure
Reported Method/Design
0-1 Linear with 1 9-day
weighted average
Shumway filters.
Time-series study.
Effect Estimates
1-h max O3 (per 45 ppb less
background, not reported):
All cause, all ages:
O3 only model:
3.19%, p< 0.055
O3 with PM2 5 model:
3. 34%, p< 0.055
Cardiovascular, all ages:
O3 only model:
3.98%, p< 0.055
O3 with PM2 5 model:
5.35%, p< 0.055
Moolgavkar et al. (1995) All cause
Philadelphia, PA
1973-1988
24-h avg O3: TSP, SO2;
multipollutant
Spring: models
25. 9 ppb
Range 2.9-74.0
Summer:
35.5 ppb
Range 1.3-159.0
Fall:
16. 2 ppb
Range 0.2-63.8
Winter:
11. 9 ppb
Range 0.0-32.9
1 Poisson; GEE and 24-h avg O3 (per 100 ppb):
nonparametric
bootstrap methods. O3 with TSP and SO2 models:
Time-series study. Spring:
2.0% (-6.7, 11.5)
Summer:
14.9% (6.8, 23.6)
Fall:
-4. 5% (-13. 9, 5.9)
Winter:
0.4% (-15. 6, 19.4)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Outcome Measure Mean O3 Levels
Copollutants
Considered
Lag Structure
Reported
Method/Design
Effect Estimates
X
oo
to
United States (cont'd)
Chock et al. (2000)
Pittsburgh, PA
1989-1991
All cause; age
<74 years;
age 75+ years
De Leon et al. (2003)
New York City
1985-1994
Circulatory and
cancer with and
without
contributing
respiratory causes
l-hmaxO3:
Levels not reported.
PM10, NO2, SO2,
CO; two-, five-,
and six-pollutant
models
PoissonGLM. Time-
series study.
24-h avg O3:
21.59 ppb
5th %-95th %
7.00-44.97
PM10, N02, S02,
CO; two-pollutant
models
0 or 1 Poisson GAM with
stringent convergence
criteria; Poisson GLM.
Time-series study.
1-h max O3 (per 40 ppb):
Age <74 years:
O3 only model:
-1.5% (t=-0.68)
O3 with PM10 model:
-2.0% (t=-0.93)
Age 75+ years:
O3 only model:
-1.8% (t=-0.82)
O3 with PM10 model:
-2.2% (t=-0.98)
Quantitative results not given.
Circulatory deaths:
Larger O3 effect estimates
with contributing respiratory
causes than without
(RR nonsignificant).
Cancer deaths:
Smaller O3 effect estimates
with contributing respiratory
causes than without
(RR nonsignificant).
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
United States (cont'd)
Klemm and Mason (2000);
Klemm et al. (2004)
Atlanta, GA
Aug 1998- July 2000
Canada
^ Vedal et al. (2003)
oo Vancouver, British
00 Columbia, Canada
1994-1996
Villeneuve et al. (2003)
Vancouver, British
Columbia, Canada
1986-1999
Outcome Measure Mean O3 Levels
All cause; 8-h max O3:
respiratory; 47.03 ppb
cardiovascular; SD 24.71
cancer; other; age
<65 years; age
65+ years
All cause; 1 -h max O3:
respiratory; 27.3 ppb
cardiovascular SD 10.2
All cause; 24-h avg O3:
respiratory; 13. 4 ppb
cardiovascular; Range 0.6-38.6
cancer;
socioeconomic
status
Copollutants
Considered
PM2,,PM10.2,,
EC, OC, NO2,
so42-,
N03-, S02, CO
PM10, N02, S02,
CO
PM25,PM10,
PM2 5.10, TSP,
coefficient of
haze, SO42% SO2,
N02, CO
Lag Structure
Reported Method/Design
0- 1 Poisson GLM using
quarterly, monthly, or
biweekly knots for
temporal smoothing.
Time-series study.
0,1,2 Poisson GAM with
stringent convergence
criteria. Time-series
study.
0,1,0-2 Poisson GLM with
natural splines. Time-
series study.
Effect Estimates
All cause, age 65+ years:
Quarterly knots:
Slope estimate:
0.00079 (SE 0.00099),
t=0.80
Monthly knots:
Slope estimate:
0.00136 (SE 0.001 11),
t=1.22
1-h max O3 (per 10.2 ppb):
Summer:
All cause:
LagO: 4.0% (1.4, 6.7)
Respiratory:
LagO: 1.5% (-6.6, 9.6)
Cardiovascular:
LagO: 3.9% (-0.3, 8.0)
24-h avg O3 (per 13. 7 ppb):
All year:
All cause:
LagO: 1.4% (-0.9, 3.6)
Respiratory:
LagO: 1.6%(-4.5, 8.1)
Cardiovascular:
LagO: 0.7% (-2.7, 4.3)
Cancer:
LagO: 2.6% (-1.2, 6.5)
No effect modification of O3-
mortality effects by
socioeconomic status.
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
65 years
Mean O3 Levels
24-h avg O3:
29.0 ng/m3
SD17.1
24-h avg O3:
29.0 ng/m3
SD17.1
Copollutants
Considered
PM25, coefficient
of haze, SO42~,
S02, N02, CO
PM25, coefficient
of haze, SO2,NO2,
NO, CO
Lag Structure
Reported Method/Design
0,1,0-2 Poisson GLM with
natural splines. Time-
series study.
0,1,0-2 Poisson GAM with
default convergence
criteria. Time-series
study.
Effect Estimates
24-h avg O3 (per 21 .3 |ig/m3):
Congestive heart failure as
underlying cause of death:
Lag 0-2: 4. 54% (-5.64,
15.81)
Having congestive heart
failure 1 year prior to death:
Lag 0-2: 2.34% (-1.78, 6.63)
24-h avg O3 (per 21 .3 |ig/m3):
All cause, all year:
All ages:
Lag 0-2: 2.26% (1.23, 3.29)
Age <65 years:
Lag 0-2: 0.18% (-1.79,2.20)
Age >65 years:
Lag 0-2: 2.84% (1.66, 4.04)
Cardiovascular, all year:
All ages:
Lag 0-2: 3.00% (1.44, 4.59)
Age <65 years:
Lag 0-2: 1.33% (-2.30, 5.09)
Age >65 years:
Lag 0-2: 3.33% (1.62, 5.08)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
oo
Reference, Study
Location and Period
Europe
Gryparis et al. (2004)
23 European cities
Study periods vary by city,
ranging from 1990-1997
Copollutants Lag Structure
Outcome Measure Mean O3 Levels Considered Reported Method/Design
All cause; l-hmaxO3: PM10, NO2, SO2, 0-1 Poisson GAM with
respiratory; Median range: CO; two-pollutant stringent convergence
cardiovascular models criteria; Bayesian
Warm season: hierarchical model.
44 ppb (Tel Aviv, Time-series study.
Israel) to 117 ppb
(Torino, Italy)
Cold season:
1 1 ppb (Milan,
Italy) to 57 ppb
(Athens, Greece)
8-h max O3:
Median range:
Warm season:
30 ppb (Rome,
Italy) to 99 ppb
(Torino, Italy)
Effect Estimates
8-h max O3 (per 10 ug/m3):
Weighted mean effect across
21 cities with 8-h max O3
concentrations:
Random effects model:
All cause:
All year:
0.03% (-0.18, 0.21)
Warm season:
O3 only model:
0.31% (0.17, 0.52)
O3 with PM10 model:
0.27% (0.08, 0.49)
Cold season:
O3 only model:
0.12% (-0.12, 0.37)
O3 with PM10 model:
Cold season:
8 ppb (Milan, Italy)
to 49 ppb (Budapest,
Hungary)
0.22% (-0.08, 0.51)
Respiratory:
Warm season:
1.13% (0.74, 0.51)
Cold season:
0.26% (-0.50, 0.84)
Cardiovascular:
Warm season:
0.46% (0.22, 0.73)
Cold season:
0.07% (-0.28, 0.41)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Europe (cont'd)
Touloumietal. (1997)
Four European cities:
London, Paris
Barcelona, Athens
Study periods vary by city,
ranging from 1986- 1992
^
X
oo
ON
Zmirouetal. (1998)
Four European cities:
London, Paris, Lyon,
Barcelona
Study periods vary by city,
ranging from 1985-1992
Copollutants
Outcome Measure Mean O3 Levels Considered
All cause 1 -h max O3: BS, NO2; two-
pollutant models
London:
41.2 ug/m3
SD26.0
Paris:
46.1 ug/m3
SD 32.9
Barcelona:
72.4 ug/m3
SD 34.9
Athens:
93.8 ug/m3
SD42.8
Respiratory; 8-h avg O3 BS, TSP, SO2,
cardiovascular (9 a.m.-5 p.m.): NO2
London:
Cold: 2 1.0 ug/m3
Warm: 40.8 ug/m3
Paris:
Cold: 11.5 ug/m3
Warm: 42.7 ug/m3
Lyon:
Cold: 2 1.0 ug/m3
Warm: 40.8 ug/m3
Barcelona:
Cold: 51.5 ug/m3
Warm: 89.7 ug/m3
Lag Structure
Reported Method/Design Effect Estimates
0, 1,2, 3, 0-1, Poisson autoregressive. 1-h max O3 (per 50 ug/m3):
0-2, 0-3 Time-series study.
Weighted mean effect across
four cities (best lag selected
for each city):
Single-day lag, random
effects models:
O3 only model:
2. 9% (1.0, 4.9)
O3 with BS model:
2. 8% (0.5, 5.0)
Cumulative lag, fixed effects
model:
O3 only model:
2.4% (1.2, 3.7)
0,1,2,3,0-1, Poisson GLM. Time- 8-h avg O3 (per 50 ug/m3):
0-2, 0-3 series study.
Weighted mean effect across
four cities (best lag selected
for each city):
Random effects models:
Respiratory:
5% (2, 8)
Cardiovascular:
2% (0, 3)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Outcome Measure Mean O3 Levels
Copollutants
Considered
Lag Structure
Reported
Method/Design
Effect Estimates
Europe (cont'd)
Anderson et al. (1996)
London, England
1987-1992
All cause;
respiratory;
cardiovascular
X
oo
1-hmax O3:
20.6 ppb
SD 13.2
8-h avg O3
(9 a.m.-5 p.m.):
15.5 ppb
SD 10.9
BS, NO2, SO2;
two-pollutant
models
Poisson GLM. Time-
series study.
All year:
8-h avg O3 (per 26 ppb):
All cause:
2.43% (1.11, 3.76)
Respiratory:
6.03% (2.22, 9.99)
Cardiovascular:
1.44o/0 (-0.45, 3.36)
Warm season:
8-h avg O3 (per 29 ppb):
All cause:
3.48% (1.73, 5.26)
Respiratory:
5.41% (0.35, 10.73)
Cardiovascular:
3.55% (1.04, 6.13)
Cool season:
8-h avg O3 (per 20 ppb):
All cause:
0.77% (-0.88, 2.44)
Respiratory:
6.20(1.67,10.94)
Cardiovascular:
-1.69% (-3.99, 0.68)
Bremneretal. (1999)
London, England
1992-1994
All cause;
respiratory;
cardiovascular; all
cancer; all others;
all ages; age
specific (0-64, 65+,
65-74, 75+ years)
22.6 ppb
SD 13.4
17.5 ppb
SD11.5
BS, PM10, NO2,
SO2, CO; two-
pollutant models
Selected best
fromO, 1,2, 3,
(all cause);
0,1,2,3,0-1,
0-2, 0-3
(respiratory,
cardiovascular)
Poisson GLM. Time-
series study.
8-h max O3 (per 26 ppb):
All ages:
All cause:
Lag 2: -0.7% (-2.3, 0.9)
Respiratory:
Lag 2: -3.6% (-7.7, 0.8)
Cardiovascular:
Lag 2: 3.5% (0.5, 6.7)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
oo
oo
Reference, Study
Location and Period
Europe (cont'd)
Anderson et al. (2001)
West Midlands region,
England
1994-1996
Prescott et al. (1998)
Edinburgh, Scotland
1992-1995
Le Tertre et al. (2002a)
Le Havre, Lyon, Paris,
Rouen, Strasbourg, and
Toulouse, France
Study periods vary by city,
ranging from 1990-1995
Outcome Measure
All cause;
respiratory;
cardiovascular; all
ages; age 0-14
years; age 15-64
years; age 65+
years
All cause;
respiratory;
cardiovascular; all
ages; age <65
years; age >65
years
All cause;
respiratory;
cardiovascular
Copollutants Lag Structure
Mean O3 Levels Considered Reported Method/Design
8-h max O,,: PM10, PM25, 0-1 Poisson GAM with
24.0 ppb PM2 5.10, BS, SO/', default convergence
SD13.8 NO2, SO2,CO criteria. Time-series
study.
24-havgO3: BS,PM10,NO2, 0 Poisson GLM. Time-
14.5 ppb SO2, CO; two- series study.
SD 2 . 3 pollutant models
8-hmaxO3: BS,NO2, SO2 0-1 Poisson GAM with
default convergence
Le Havre: criteria. Time-series
43.4 ug/m3 study.
Lyon:
52.0 ug/m3
Paris:
26.0 ug/m3
Rouen:
57.9 ug/m3
Strasbourg:
37.0 ug/m3
Toulouse:
68.0 ug/m3
Effect Estimates
8-h max O3 (per 30.8 ppb):
All ages:
All cause:
2. 9% (-0.1, 6.0)
Respiratory:
2.2% (-5.4, 10.4)
Cardiovascular:
0.9% (-3.4, 5.4)
24-havgO3(perlOppb):
All cause, all ages:
-4.2% (-8.1, -0.1)
Cardiovascular, age >65
years:
2.2% (-5. 1,10. 3)
8-h max O3 (per 50 ug/m3):
Six-city pooled estimates:
All cause:
2.7% (1.3, 4.1)
Respiratory:
0.8% (-4.8, 6.2)
Cardiovascular:
2.4% (-0.3, 5.1)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
oo
VO
Reference, Study
Location and Period
Europe (cont'd)
Dab etal. (1996)
Paris, France
1987-1992
Zmirou etal. (1996)
Lyon, France
1985-1990
Sartor etal. (1995)
Belgium
Summer 1994
Copollutants
Outcome Measure Mean O3 Levels Considered
Respiratory 1 -h max O3: BS, PM13, NO2,
43.9 ug/m3 S02, CO
5th %-99th %
6.0-147.0
8-h max O3:
27.7 ug/m3
5th %-99th %
3-110
All cause; 1 -h max O3: PM13, SO2, NO2
respiratory; 15.23 ug/m3
cardiovascular; Range 0-1 52
digestive
8-h avg O3
(9 a.m.-5 p.m.):
9.94 ug/m3
Range 0-78. 92
All cause; age 24-h avg O3: TSP, NO, NO2,
<65 years; Geometric mean: SO2
age 65+ years
During heat wave
(42-day period):
72.4 ug/m3
Range 34. 5-1 11. 5
Before heat wave
(43-day period):
52.4 ug/m3
Range 30.7-92.0
After heat wave
(39-day period):
38.6 ug/m3
Range 18.8-64.9
Lag Structure
Reported Method/Design Effect Estimates
0 Poisson autoregressive. 1-h max O3 (per 100 ug/m3):
Time-series study. 1 .074 (0.934, 1 .235)
8-h max O3 (per 100 ug/m3):
1.040(0.934,1.157)
Selected best Poisson GLM. Time- 8-h avg O3 (per 50 ug/m3):
from 0, 1, 2, 3 series study.
All cause:
LagO: 3% (-5, 12)
Respiratory:
Lagl: l%(-8, 10)
Cardiovascular:
Lagl: 0%(-11,12)
0,1,2 Log-linear regression. No individual regression
Time-series study. coefficient for O3 alone;
interaction with temperature
suggested.
24-h avg O3 (from 18.8 to
111.5 ug/m3) and temperature
(from 10.0 to 27.5°C):
Age <65 years:
Lag 1 : 16% increase in
mortality (5.3% expected)
Age 65+ years:
Lagl: 36.5% increase in
mortality (4% expected)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Outcome Measure Mean O3 Levels
Copollutants
Considered
Lag Structure
Reported
Method/Design
Effect Estimates
Europe (cont'd)
Hoek et al. (2000;
reanalysis Hoek, 2003)
The Netherlands: entire
country, four urban areas
1986-1994
All cause; COPD;
pneumonia;
cardiovascular
8-h avg O3
(12 p.m.-8 p.m.):
Median: 47 ug/m3
Range 1-226
PM10, BS, SO/-,
NO/, NO2, SO2,
CO; two-pollutant
models
1,0-6 PoissonGAM,
reanalyzed with
stringent convergence
criteria; Poisson GLM.
Time-series study.
Poisson GLM:
All cause:
8-h avg O3 (per 150 ug/m3):
Lagl: 4.3% (2.4, 6.2)
8-h avg O3 (per 120 ug/m3):
Lag 0-6: 5.9% (3.1, 8.7)
X
Hoek etal. (2001;
reanalysis Hoek, 2003)
The Netherlands
1986-1994
Total
cardiovascular;
myocardial
infarction;
arrhythmia; heart
failure;
cerebro vascular;
thrombosis-related
8-h avg O3
(12p.m.-8p.m.):
Median: 47 ug/m3
Range 1-226
PM10, N02, S02.
CO
Poisson GAM,
reanalyzed with
stringent convergence
criteria; Poisson GLM.
Time-series study.
8-h avg O3 (per 150 ug/m3):
Poisson GLM:
Total cardiovascular:
6.2% (3.3, 9.2)
Myocardial infarction:
4.3% (0.1, 8.6)
Arrhythmia:
11.4% (-1.2, 25.5)
Heart failure:
10.2% (1.2, 19.9)
Cerebrovascular:
9.1% (2.9, 15.7)
Thrombo sis-related:
16.6% (2.8, 32.2)
Roemer and van Wijinen
(2001)
Amsterdam, the
Netherlands
1987-1998
All cause
8-h max O3:
Background sites:
43 ug/m3
Maximum 221
Traffic sites:
36 ug/m3
Maximum 213
BS,PM10,N02.
SO,, CO
1, 2, 0-6 Poisson GAM with
default convergence
criteria (only one
smoother). Time-series
study.
8-h max O3 (per 100 ug/m3):
Total population using
background sites:
Lagl: -0.3%(-4.1,3.7)
Total population using traffic
sites:
Lagl: 0.2% (-3.6, 4.2)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
Europe (cont'd)
Verhoeffetal. (1996)
Amsterdam, the
Netherlands
1986-1992
Peters et al. (2000b)
NE Bavaria, Germany
1982-1994
Coal basin in Czech
Republic
1993-1994
Ponkaetal. (1998)
Helsinki, Finland
1987-1993
Outcome Measure
All cause; all ages;
age 65+ years
All cause;
respiratory;
cardiovascular;
cancer
All cause;
cardiovascular; age
<65 years, age 65+
Mean O3 Levels
1-hmax O3:
43 ug/m3
Maximum 301
24-h avg O3:
Czech Republic:
40.3 ug/m3
SD25.0
Bavaria, Germany:
38.2 ug/m3
SD21.9
24-h avg O3:
Median 18 ug/m3
5th %-95th% 3-51
Copollutants Lag Structure
Considered Reported Method/Design
PM10, NO2, SO2, 0,1,2 Poisson. Time-series
CO; multipollutant study.
models
TSP, PM10, NO2, 0,1,2,3 Poisson GLM.
SO2, CO Time-series study.
TSP, PM10, NO2, 0, 1, 2, 3, 4, 5, Poisson GLM. Time-
SO2 6, 7 series study.
Effect Estimates
1-h max O3 (per 100 ug/m3)
All ages:
LagO: 1.8% (-3. 8, 7.8)
Lagl: 0.1% (-4.7, 5.1)
Lag 2: 4.9% (0.1, 10.0)
24-h avg O3 (per 100 ug/m3):
Czech Republic:
All cause:
Lag 2: 7.8% (-1.8, 18.4)
Bavaria, Germany:
All cause:
LagO: 8.2% (0.4, 16.7)
Cardiovascular:
LagO: 6.1% (-3.7, 17.0)
24-h avg O3 (per 20 ug/m3):
All cause:
years
Age <65 years:
Not significant, values not
reported.
Age 65+ years:
Not significant, values not
reported.
Cardiovascular:
Age <65 years:
LagO: -2.0% (-9.5, 6.1)
Lagl: 6.2%(-2.2, 15.5)
Age 65+ years:
Not significant, values not
reported.
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Europe (cont'd)
Saez et al. (2002)
Barcelona, Spain
1991-1995
Madrid, Spain
1992-1995
Valenica, Spain
1994-1996
j^j Garcia-Aymerich et al.
^ (2000)
\Q Barcelona, Spain
t° 1985-1989
Outcome Measure
All cause;
respiratory;
cardiovascular
All cause;
respiratory;
cardiovascular;
general population;
Mean O3 Levels
8-h max O3:
Barcelona:
67.5 ug/m3
SD 32.2
Madrid:
42.1 ug/m3
SD27.8
Valencia:
45.5 ug/m3
SD 19.7
l-hmaxO3:
Levels not reported.
Copollutants Lag Structure
Considered Reported Method/Design
NO2, PM, SO2, 0-5 Poisson GAM with
CO; multipollutant default convergence
models criteria. Time-series
study.
BS, NO2, SO2, Selected best Poisson GLM.
single-day lag Time-series study.
Effect Estimates
8-h max O3 (per 10 ug/m3):
Three-city pooled estimates:
All cause:
0.23% (-0.15, 0.61)
Respiratory:
0.29% (-0.05, 0.63)
Cardiovascular:
0.60% (0.08, 1.13)
1-h max O3 (per 50 ug/m3):
All cause:
General population:
patients with COPD
Saezetal. (1999)
Barcelona, Spain
1986-1989
Asthma mortality;
age 2-45 years
BS, NO2, SO2,
Levels not reported.
0-2 Poisson with GEE.
Time-series study.
Lag 5: 2.4% (0.6, 4.2)
COPD patients:
Lag 3: 4.0% (-4.7, 13.4)
Respiratory:
General population:
Lag 5: 3.5% (-1.9, 9.2)
COPD patients:
Lag 3: 5.7% (-7.9, 21.4)
Cardiovascular:
General population:
Lagl: 2.9% (0.4, 5.4)
COPD patients:
Lag 3: !.!%(-14.2,19.2)
Slope estimate:
0.021 (SE 0.01 l),p = 0.054
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Outcome Measure Mean O3 Levels
Copollutants
Considered
Lag Structure
Reported
Method/Design
Effect Estimates
X
Europe (cont'd)
Sunyeretal. (1996)
Barcelona, Spain
1985-1991
Sunyer and Basagana
(2001)
Barcelona, Spain
1990-1995
All cause;
respiratory;
cardiovascular;
all ages;
age 70+ years
Mortality in a
cohort of patients
with COPD
Summer:
86.5 ng/m3
Range 9.5-283.5
Winter:
55.2 ng/m3
Range 7-189.2
1 -h max O3:
Mean not reported
IQR 2 1 |ig/m3
BS, SO2,NO2
PM10, NO2, CO
Selected best
single-day lag
0-2
Autoregressive
Poisson. Time-series
stucjy
Conditional logistic
(case-crossover)
1-h max O3 (per 100 |ig/m3):
All cause, all ages:
All year:
LagO: 4.8%(1.2, 8.6)
Summer:
LagO: 5.8%(1.7, 10.1)
Winter:
LagO: 2.6%(-3.5, 9.1)
Respiratory, all ages:
All year:
Lag 5: 7.1% (-3.8, 19.2)
Summer:
Lag 5: 5.0% (-7.3, 18.8)
Winter:
Lag 5: 14.0% (-7.6,40.6)
Cardiovascular, all ages:
All year:
Lagl: 5.8%(0.9, 11.1)
Summer:
Lagl: 8.8%(2.8, 15.2)
Winter:
Lagl: -0.8% (-8.9, 7.9)
1-h max O3(per21 |ig/m3):
Odds ratio:
0.979(0.919,1.065)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Outcome Measure Mean O3 Levels
Copollutants
Considered
Lag Structure
Reported
Method/Design
Effect Estimates
X
Europe (cont'd)
Sunyer et al. (2002)
Barcelona, Spain
1986-1995
Diazetal. (1999)
Madrid, Spain
1990-1992
All cause,
respiratory, and
cardiovascular
mortality in a
cohort of patients
with severe asthma
All cause;
respiratory;
cardiovascular
Median 69.3 ug/m3
Range 6.6-283.0
Median 54.4 ug/m3
Range 3.9-244.5
24-h avg O3:
Levels not reported.
PM10, BS, SO2,
N02, CO, pollen
0-2
Conditional logistic
(case-crossover)
TSP, NO2, SO2.
CO
1,4,10
Autoregressive linear.
Time-series study.
1-hmax O3 (per 48 ug/m3):
Odds ratios:
Patients with only one
admission:
All cause:
1.096(0.820,1.466)
Cardiovascular:
1.397(0.854,2.285)
Patients with more than
one admission:
All cause:
1.688(0.978,2.643)
Cardiovascular:
1.331(0.529,3.344)
Patients admitted for both
asthma and COPD:
All cause:
0.946(0.674,1.326)
Cardiovascular:
0.985(0.521,1.861)
24-h avg O3 (per 25 ug/m3):
For O3 levels higher than 35
ug/m3:
All cause:
Lag 4: 12% (p < 0.01)
U-shaped (quadratic) O3-
mortality relationship with a
minimum of 35 ug/m3.
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Outcome Measure Mean O3 Levels
Copollutants
Considered
Lag Structure
Reported
Method/Design
Effect Estimates
Latin America
Borja-Aburtoetal. (1997)
Mexico City
1990-1992
X
All cause;
respiratory;
cardiovascular; all
ages; age <5 years;
age >65 years
l-hmaxO3:
Median 155 ppb
8-h max O3:
Median 94 ppb
10-havgO3
(8 a.m.-6 p.m.):
Median 87 ppb
24-h avg O3:
Median 54 ppb
TSP, S02, CO;
two-pollutant
models
0,1, 2 Poisson iteratively
weighted and filtered
least-squares method.
Time-series study.
1-h max O3 (per 100 ppb):
All ages:
O3 only model:
All cause:
LagO: 2.4%(1.1, 3.9)
Respiratory:
LagO: 2.3% (-1.9, 6.7)
Cardiovascular:
LagO: 3.6% (0.6, 6.6)
O3 with TSP model:
All cause:
LagO: -1.8% (-10.0, 6.4)
Respiratory:
LagO: -1.9%(-11.0,8.2)
Cardiovascular:
LagO: 2.4 (-4.4, 9.6)
Borja-Aburto et al. (1998)
SW Mexico City
1993-1995
All cause;
respiratory;
cardiovascular;
other; all ages;
age >65 years
1-hmax O3:
163 ppb
SD57
24-h avg O3:
44.0 ppb
SD15.7
PM2 5, NO2, SO2;
two-pollutant
models
0,1,2,3,4,5,
1-2
Poisson GAM with
default convergence
criteria (only one
smoother). Time series
study.
24-h avg O3 (per 10 ppb):
All cause, all ages:
Lag 1-2: 0.6% (-0.3, 1.5)
All cause, age > 65 years:
Lag 1-2: 0.8% (-0.4, 2.0)
Respiratory, all ages:
Lag 1-2: -0.7% (-3.6, 2.1)
Cardiovascular, all ages:
Lag 1-2: 1.8% (0.1, 3.5)
Other noninjury, all ages:
Lag 1-2: 0.3% (-0.9, 1.4)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
Latin America (cont'd)
O'Neill et al. (2004)
Mexico City
1996-1998
Tellez-Rojo et al. (2000)
Mexico City
1994
Loomisetal. (1999)
Mexico City
1993-1995
Gouveia and Fletcher
(2000b)
Sao Paulo, Brazil
1991-1993
Copollutants
Outcome Measure Mean O3 Levels Considered
All cause; all ages; 24-h avg O3: PM10
age 65+ years; 35.3ppb
SES gradient SD11.0
Respiratory; l-hmaxO3: PM10,NO2, SO2
COPD mortality; 1 34. 5 ppb
age 65+ years; SD33.4
within medical
unit; outside of
medical unit
Infant mortality 24-h avg O3: PM2 5, NO2
44.1 ppb
SD15.7
All ages (all cause); 1 -h max O3: PM10, NO2, SO2,
age <5 years 67.9 ug/m3 CO
(all cause, SD42.1
respiratory,
pneumonia);
age 65+ years
(all cause,
respiratory,
cardiovascular)
Lag Structure
Reported Method/Design
0- 1 Poisson GAM with
stringent convergence
criteria. Time-series
study.
1, 2, 3, 4, 5, 1-3, Poisson, iteratively
1-5,1-7 weighted and filtered
least-squares method.
Time-series study.
0,1,2,3,4,5, Poisson GAM with
2-3 default convergence
criteria. Time-series
study.
0,1,2 Poisson GLM. Time-
series study.
Effect Estimates
24-h avg O3 (per 10 ppb):
All ages:
0.65% (0.02, 1.28)
Age 65+ years:
1.39% (0.51, 2.28)
SES gradient did not show
any consistent pattern.
1-h max O3 (per 40 ppb):
Outside medical unit:
Respiratory:
Lag 1-5: 14.0% (4.1, 24.9)
COPD mortality:
Lag 1-5: 15. 6% (4.0, 28.4)
24-h avg O3 (per 10 ppb):
O3 only model:
2.45% (-0.54, 5.43)
O3 with PM2 5 model:
1.40% (-1.92, 4.72)
1-h max O3 (per 106 ug/m3):
All ages:
All cause:
LagO: 0.8%(-1. 1,2.7)
Age 65+ years:
All cause:
Lag 2: 2.3% (0,4.6)
Respiratory:
Lag 2: 5.1% (-0.6, 11.1)
Cardiovascular:
LagO: 3.1% (-0.4, 6.7)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
Latin America (cont'd)
Pereiraetal. (1998)
Sao Paulo, Brazil
1991-1992
Saldivaetal. (1994)
Sao Paulo, Brazil
1990-1991
Saldivaetal. (1995)
Sao Paulo, Brazil
1990-1991
Cifuentes et al. (2000)
Santiago, Chile
1988-1966
Ostroetal. (1996)
Santiago, Chile
1989-1991
Outcome Measure Mean O3 Levels
Intrauterine 1 -h max O3 :
mortality 67.5 ug/m3
SD45.0
Respiratory; age 24-h avg O3:
<5 years 12.14ppb
SD 9.94
All cause; 1-h max O3:
age 65+ years 38.3 ppb
SD29.7
24-h avg O3:
12.5 ppb
SD11.5
All cause 1 -h max O3:
Summer:
108.2 ppb
IQR48.0
All cause 1 -h max O3:
52.8 ppb
Range 11 -264
Copollutants
Considered
PM10, N02, S02,
CO
PM10, N02, S02,
CO; multipollutant
models
PM10, NO2, SO2,
CO; two-pollutant
models
PM25,PM10.25,
CO,SO2,NO2
PM10, NO2, SO2;
two-pollutant
models
Lag Structure
Reported Method/Design
0 Poisson, linear with
M-estimation. Time-
series study.
0-2 OLS of transformed
data. Time-series
study.
0-1 OLS; Poisson with
GEE. Time-series
study.
0,1,2,3,4,5, Poisson GAM with
1-2,1-3,1-4, default convergence
1-5 criteria; Poisson GLM.
Time-series study.
1 OLS, several other
methods. Time-series
study.
Effect Estimates
Slope estimate:
0.0000 (SE 0.0004)
Slope estimate:
0.01048 deaths/day/ppb
(SE 0.02481 ),p = 0.673
Slope estimate:
l-hmaxO3:
0.0280 deaths/day/ppb
(SE 0.0213), p> 0.05
24-h avg O3:
0.0093 deaths/day/ppb
(SE 0.0813), p> 0.05
1-h max O3 per (108.2 ppb):
Poisson GLM:
Summer:
O3 only model:
Lag 1-2: 0.3%(t=0.3)
Multipollutant model:
Lag 1-2: -0.1% (t= -0.1)
All year:
1-h max O3 (per 52.8 ppb):
-3% (-4, -2)
Summer:
1 -h max O3 (per 100 ppb):
4% (0, 9)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
oo
Reference, Study
Location and Period
Australia
Morgan etal. (1998b)
Sydney, Australia
1989-1993
Simpson etal. (1997)
Brisbane, Australia
1987-1993
Asia
Kim et al. (2004)
Seoul, Korea
1995-1999
Outcome Measure Mean O3 Levels
All cause; 1 -h max O3:
respiratory; 24 ppb
cardiovascular SD 13
All cause; 8-h avg O3
respiratory; (10 a.m.-6 p.m.):
cardiovascular;
all ages; age All year:
<65 years; 18.1 ppb
age 65+ years Range 1.7-63.4
Summer:
20.2 ppb
Range 2.7-63.4
Winter:
16.1 ppb
Range 1.7-56.9
All cause 1 -h max O3:
All year:
35. 16 ppb
SD 18.31
Summer:
46.87 ppb
SD 22.46
Winter:
2 1.26 ppb
SD6.92
Copollutants
Considered
PMby
nephelometer,
NO2;
multipollutant
models
PM10,PMby
nephelometer,
NO2, SO2, CO
PM10, N02, S02,
CO; two-pollutant
models
Lag Structure
Reported Method/Design
0 Poisson with GEE.
Time-series study.
0 Autoregressive Poisson
with GEE. Time-series
study.
1 Poisson GAM with
stringent convergence
criteria (linear model);
GLM with cubic
natural spline; GLM
with B-mode spline
(threshold model).
Time-series study.
Effect Estimates
l-hmaxO3(per28ppb):
All cause:
2.04% (0.37, 3.73)
Respiratory:
-0.84% (-7.16, 5.91)
Cardiovascular:
2.52% (-0.25, 5.38)
8-h avg O3 (per 10 ppb):
All cause, all ages:
All year:
2.4% (0.8, 4.0)
Summer:
3.0% (1.0, 5.0)
Winter:
1.3% (-1.4, 4.1)
1-h max O3 (per 21 .5 ppb):
All year:
Linear model:
2.6% (1.7, 3.5)
Threshold model:
3.4% (2.3, 4.4)
Summer:
Linear model:
1.9% (0.5, 3.3)
Threshold model:
3.8% (2.0, 5.7)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
Reference, Study
Location and Period
Asia (cont'd)
Leeetal. (1999)
Seoul and Ulsan, Korea
1991-1995
Lee and Schwartz (1999)
"^ Seoul, Korea
^ 1991-1995
VO
Kwonetal. (2001)
Seoul, Korea
1994-1998
Copollutants
Outcome Measure Mean O3 Levels Considered
All cause 1-h max O3: TSP, SO2
Seoul:
32.4 ppb
10th%-90th%
14-55
Ulsan:
26.0 ppb
10th%-90th%
16-39
All cause 1-h max O3: TSP, SO2
Seoul:
32.4 ppb
10th%-90th%
14-55
Mortality in a 1-h max O3: PM10, NO2, SO2,
cohort of patients 31.8 ppb CO
with congestive IQR 20.5
heart failure Range 4.3-102.8
Lag Structure
Reported Method/Design
0 Poisson with GEE.
Time-series study.
0 Conditional logistic
regression. Case-
crossover with
bidirectional control
sampling.
0 Time-series analysis
using Poisson GAM
with default
convergence criteria;
case-crossover analysis
using conditional
logistic regression.
Effect Estimates
1-h max O3 (per 50 ppb):
Seoul:
1.5% (0.5, 2.5)
Ulsan:
2.0% (-11. 1,17.0)
1-h max O3 (per 50 ppb):
Two controls, plus and minus
1 week:
1.5% (-1.2, 4.2)
Four controls, plus and minus
2 weeks:
2. 3% (-0.1, 4. 8)
1-h max O3 (per 20.5 ppb):
Odds ratios from case-
crossover study design:
General population:
1.9% (1.0, 2.9)
Congestive heart failure
cohort:
5.1% (-3. 6, 14.7)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
o
o
Reference, Study
Location and Period
Asia (cont'd)
Ha et al. (2003)
Seoul, Korea
1995-1999
Hong et al. (2002)
Seoul, Korea
1995-1998
Tsai et al. (2003b)
Kaohsiung, Taiwan
1994-2000
Yang et al. (2004b)
Taipei, Taiwan
1994-1998
Outcome Measure
All cause;
respiratory;
postneonatal
(1 month to 1 year);
age 2-64 years; age
>65 years
Acute stroke
mortality
All cause;
respiratory;
cardiovascular;
tropical area
All cause;
respiratory;
cardiovascular;
subtropical area
Copollutants
Mean O3 Levels Considered
8-h avg O3: PM10, NO2, SO2,
2 1.2 ppb CO
SD11.6
Range 2.9-69.1
8-h avg O3: PM10, NO2, SO2,
22.6 ppb CO
SD 12.4
IQR 9.3
24-h avg 03: PM10, SO2, NO2,
23. 6 ppb CO
Range 1.2-83.0
24-h avg O3: PM10, SO2, NO2,
17.18 ppb CO
Range 2. 3-43.47
Lag Structure
Reported Method/Design
0 Poisson GAM with
default convergence
criteria. Time-series
study.
0 Poisson GAM with
default convergence
criteria. Time-series
study.
0-2 Conditional logistic
regression. Case-
crossover analysis.
0-2 Conditional logistic
regression. Case-
crossover analysis.
Effect Estimates
8-h avg O3 (per 16.1 ppb):
Respiratory:
Postneonatal:
22.6% (-41.2, 155.8)
Age 2-64 years:
9.8% (6.8, 13.0)
Age >65 years:
3.7% (2.6, 4.8)
8-h avg O3 (per 9.3 ppb):
O3 only model:
2. 9% (0.3, 5.5)
PM10< median: 5.5%
PM10> median: -2.5%
24-h avg O3 (per 19. 2 ppb):
Odds ratios:
All cause:
0.994(0.995,1.035)
Respiratory:
0.996(0.848,1.169)
Cardiovascular:
1.005(0.919,1.098)
24-h avg O3 (per 9.34 ppb):
Odds ratios:
All cause:
0.999(0.972,1.026)
Respiratory:
0.991(0.897,1.094)
Cardiovascular:
1.004(0.952,1.058)
-------
Table AX7-5 (cont'd). Effects of Acute O3 Exposure on Mortality
X
Reference, Study
Location and Period
Asia (cont'd)
Wong etal. (2001)
Hong Kong
1995-1997
Copollutants
Outcome Measure Mean O3 Levels Considered
All cause; 8-h avg O3 PM10, NO2, SO2;
respiratory; (9 a.m.-5 p.m.): two-pollutant
cardiovascular models
Warm:
32.0 ppb
SD24.5
Range 0-1 68. 9
Cool:
35.1 ppb
SD21.3
Range 0-101. 6
Lag Structure
Reported Method/Design Effect Estimates
3, 4, 5 Poisson GAM with 8-h avg O3 (per 16.1 ppb):
default convergence
criteria. Time-series All cause:
study. Warm:
Lag 5: -l%(-3,2)
Cool:
Lag 5: 4% (1,6)
Respiratory:
Warm:
Lag 4: -l%(-6, 5)
Cool:
Lag 4: 8% (2, 15)
Cardiovascular:
Warm:
Lag 3: -2% (-6, 3)
Cool:
Lag 3: 5% (0,11)
Hedley et al. (2002)
Hong Kong
1985-1995
Intervention Jul 1990
(switch to low sulfur-
content fuel)
All cause;
cardiovascular;
respiratory;
neoplasms and
other causes;
all ages;
age 15-64 years;
age 65+ years
Average monthly
03:
Baseline:
18.5 ng/m3
SD7.5
1 year after
intervention:
21.3|ig/m3
SD9.1
2-5 years after
intervention:
22.1 ng/m3
SD 10.2
SO2 (main
pollutant of
interest, 45%
reduction observed
5 years after
intervention),
PM10, SO/', NO2
Monthly Poisson regression of
averages monthly averages to
considered estimate changes in the
without lags increase in deaths from
warm to cool season.
Annual proportional
change in death rate
before and after the
intervention was also
examined.
Declines observed in all
cause (2. l%,p = 0.001),
respiratory (3.9%, p = 0.001),
and cardiovascular (2.0%, p =
0.020) mortality after the
intervention.
Analysis not specific for O3
effects. As O3 levels did not
change before and after the
intervention, O3 likely did not
play a role in the decline in
observed mortality.
-------
Table AX7-6. Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
United States
Galizia and Kinney
(1999; expo sure data
Kinney et al., 1998)
U.S. nationwide
1995
l-hmaxO3:
10-year mean Jun-Aug:
61.2 ppb
SD15.5
Range 13-185
X
o
to
Goss et al. (2004)
U.S. nationwide
1999-2000
l-hmaxO3
51.0 ppb
SD7.3
Cross-sectional study of a nationwide sample of
520 young adults. Subjects were nonsmokers, aged
17-21 years, 50% males. Each subject provided one
spirometric lung function measurement in the spring
of their 1st year at Yale College in New Haven, CT,
and completed a questionnaire addressing
residential history, respiratory diseases, and activity
patterns. Long-term O3 exposure was treated as a
high/low dichotomous variable, with subjects
assigned to the high O3 category if they lived for
4+ years in counties with 10-year summer mean O3
levels greater than 80 ppb. Four lung function
variables (FVC, FEVl5 FEF25.75, FEF75) were
regressed on O3 exposure, controlling for
age, height, height squared, sex, race, parental
education, and maternal smoking history.
Respiratory symptom histories (cough, phlegm,
wheeze apart from colds, and composite index for
any of the three symptoms) were logistically
regressed on O3 exposure, controlling for sex, race,
parental education, and maternal smoking.
Cohort study of 11,484 cystic fibrosis patients over
the age of 6 years. Exposure to O3, PM25, PM10,
NO2, SO2, and CO assessed by linking Aerometric
Information Retrieval System with patients' home
zip code. Studied exacerbation and lung function.
Mortality was also of interest, but study was
underpowered to examine this outcome. Logistic
regression models were used to analyze the
exacerbations and multiple linear regression was
used to study lung function. O3 monitoring season
and regional effects also were examined.
Significant decrements in FEV[ and FEF25.75 in relation to O3
exposure were observed for all subjects and for males alone, but
not for females alone. Similar patterns observed for FVC and
FEF75, but not with statistical significance.
Percent difference in lung function for high vs. low O3 exposure
groups:
All subjects: -3.07% (-0.22, -5.92)
Females: -0.26% (3.79,-4.31)
Males: -4.71% (-0.66,-8.76)
FEF25.75:
All subjects: -8.11% (-2.32, -13.90)
Females: -1.96% (6.39,-10.30)
Males: -13.02% (-4.87,-21.17)
Wheeze and respiratory symptom index were significantly elevated
for high O3 exposure group.
Odds ratios for symptoms:
Wheeze: 1.97(1.06,3.66)
Respiratory symptom index: 2.00 (1.15, 3.46)
Ozone may increase the risk for pulmonary exacerbations in cystic
fibrosis patients.
Odds ratios for two or more exacerbations (per 10 ppb increase in
1-hmax O3):
O3 only model: 1.10(1.03,1.17)
O3 with PM2 5 model: 1.08(1.01,1.15)
PM25, but not O3, was significantly associated with declines in
lung function in these patients.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
X
o
oo
United States (cont'd)
Kinney and Lippmann
(2000)
Fort Sill, OK;
Fort Leonard Wood,
MO; Fort Dix, NJ;
Fort Benning, GA;
West Point, NY
Apr-Sep 1990
l-hmaxO3:
Mean during 5-week
summer training
period:
Fort Benning, GA:
55.6 ppb
(OhO3>100ppb)
Fort Dix, NJ:
71.3 ppb
(23hO3>100ppb)
Fort Leonard Wood,
MO:
55.4 ppb
(lhO3>100ppb)
Fort Sill, OK:
61.7 ppb
(lhO3>100ppb)
Prospective cohort study of 72 nonsmoking students
(mean age 20.25 years) at the U.S. Military
Academy at West Point, NY were measured for
lung function and respiratory symptoms before
(Apr) and after (Aug-Sep) taking part in an
intensive, largely outdoor, summer training over 5
weeks (Jul 11-Aug 15) at four U.S. military bases.
Ozone levels in the Fort Dix, NJ area were
consistently higher than at the three remaining three
locations. Analysis assessed change in lung
function and respiratory symptoms measured before
and soon after the summer training, and examined
whether adverse trends would be more pronounced
in students exposed to higher O3 levels during
summer training.
Mean FEV! declined significantly over the two measurement
points for all subjects combined, which may reflect combined
effects of O3 with exposures to dust, vehicle exhaust, and
environmental tobacco smoke as reported by subjects from all four
locations in the post-summer questionnaire. However, a larger
mean decline was seen at the higher O3 site, Fort Dix, than at the
remaining three sites, suggesting an influence of O3 exposures.
A larger decline was observed in subjects with post-summer
measurements in the 1 st and 2nd weeks after returning from
training vs. those measured in the 3rd and 4th weeks, which is
consistent with the lung function effects being somewhat transient.
Change in lung function over the summer:
All locations: -44 mL (SE21), p = 0.035
Fort Dix: -78 mL (SE41), p = 0.07
Forts Sill, Leonard Wood, and Benning combined:
-31mL(SE24),p = 0.21
Greeretal. (1993)
California
1973-1987
Annual mean O3:
Levels not reported.
Prospective cohort study of 3,914 nonsmoking
adults aged 25+ years at enrollment in 1977
completed questionnaires at two time points, 1977
and 1987. To be eligible, subjects had to have lived
10 or more years within 5 miles of current
residence. Residential histories used to interpolate
air pollution levels to zip centroids over a 20-year
period (1966-1987). New asthma cases defined as
answering yes to doctor diagnosed asthma at 1987
followup among those answering no at enrollment
in 1977. Multiple logistic regression used to test for
associations between new-onset asthma and long-
term exposures to air pollution, controlling for age,
education, pneumonia or bronchitis before age 16
years, and years worked with a smoker through
1987. All models stratified by gender.
There were 27 incident cases of asthma among 1,305 males and
51 incident cases among 2,272 females. In logistic regression
analyses, long-term O3 exposures were associated with increased
risk of incident asthma among males but not females.
Relative risks for incident cases of asthma (per 10 ppb increase
in annual mean O3):
Males: 3.12(1.61,5.85)
Females: 0.94(0.65,1.34)
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
United States (cont'd)
McDonnell et al.
(1999)
California
1973-1992
X
Peters etal. (1999a,b)
12 Southern
California
communities
1993-1994
8-h avg O3
(9 a.m.-5 p.m):
20-year mean:
46.5 ppb
SD15.3
l-hmaxO3:
Mean range:
1986-1990:
30.2 ppb (Santa Maria)
to 109.2 ppb
(San Dimas)
1994:
35.5 ppb (Santa Maria)
to 97.5 ppb
(Lake Gregory)
This prospective cohort study continued the work of
Greer et al. (1993). 3,091 nonsmoking adults
completed questionnaires at one additional time
point, 1992. Residential histories used to interpolate
air pollution levels to zip centroids over the period
1973-1992, yielding annual mean exposure
estimates for O3, PM10, SO2, and NO2. New asthma
cases defined as answering yes to doctor diagnosed
asthma at either 1987 or 1992. Multiple logistic
regression used to test for associations between
new-onset asthma and long-term exposures to air
pollution, controlling for age, education, pneumonia
or bronchitis before age 16, and ever smoking.
All models run separately for males and females.
3,676 children aged 9-16 years enrolled into
the 1st cohort of the Children's Health Study in
1993. Subjects provided questionnaire data on
respiratory disease histories and symptoms.
3,293 subjects also underwent pulmonary function
testing, of which 2,781 were used in air pollution
regressions. Air pollution data for O3, PM10, PM2 5,
NO2, and inorganic acid vapors analyzed from
1986-1990 and 1994. For cross-sectional analysis
of respiratory diseases, individual pollutants were
tested for associations with ever asthma, current
asthma, bronchitis, cough, and wheeze after
controlling for covariates. For analysis of lung
function, individual pollutants and pairs of
pollutants were regressed with FVC, FEV1; FEF25.75,
and PEF, controlling for usual demographic and
anthropometric covariates.
There were 32 incident cases of asthma among 972 males and
79 incident cases among 1,786 females. In logistic regression
analyses, long-term O3 exposures were associated with increased
risk of incident asthma among males but not females. Other
pollutants were neither associated with asthma incidence nor
were confounders of the O3 association in males.
Relative risks for incident cases of asthma (per 27 ppb increase
in annual mean 8-h avg O3):
Males: 2.09(1.03,4.16)
Females: 0.86(0.58,1.26)
Acids and NO2, but not O3, were associated significantly with
prevalence of wheeze. No associations of O3 with any of the
respiratory diseases or symptoms.
Decreased lung function was associated with multiple pollutants
among females but not males. For O3 exposure in females, all four
lung function variables declined with increasing exposure.
Associations were stronger for current (1994) exposure compared
to previous (1986-1990) exposure. In males who spent more time
outdoors, FVC and FEV[ declined significantly with higher current
exposure to O3.
Change in lung function (per 40 ppb 1-h max O3 from 1986-1990):
Females:
PEF: -187.2mL/s(SE50.1),p<0.005
FEF25.75: -102.2 mL/s (SE 28.8), p < 0.01
Males:
PEF: 31.1 mL/s (SE 48.8), p> 0.05
FEF25.75: 11.7 mL/s (SE 26.7), p > 0.05
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
X
United States (cont'd)
Gauderman et al.
(2000; 2004a,b)
12 Southern
California
communities
1993-2001
Gauderman et al.
(2002)
12 Southern
California
communities
1996-1999
McConnell et al.
(1999)
12 Southern
California
communities
1993
8-h avg O3
(10 a.m.-6 p.m.):
Mean range:
Approximately 28 ppb
(Long Beach) to 65 ppb
(Lake Arrowhead)
8-h avg O3
(10 a.m.-6 p.m.):
Mean range:
Approximately 27 ppb
(Long Beach) to 67 ppb
(Lake Gregory)
l-hmaxO3:
Estimated annual mean:
65.6 ppb
Range 35.5-97.5
Analysis of longitudinal lung function change
in relation to long-term air pollution levels in
the Children's Health Study cohort. Children from
4th (n = 1,498), 7th (n = 802), and 10th (n = 735)
grade enrolled in 1993. Children enrolled in 7th and
10th grade were followed until 1997; 4th graders
were followed until 2001. Baseline questionnaires
completed and annual pulmonary function tests
(FVC, FEVj, FEF25.75, FEF75) performed.
Air pollution monitoring stations established in the
12 study communities beginning in 1994 to measure
O3, NO2, PM10, PM25, and inorganic acid.
Analysis using adjusted linear regression models.
Second cohort of the longitudinal cohort Children's
Health Study. 2,081 4th graders (mean age 9.9
years) enrolled in 1996. Baseline questionnaires
were completed and annual pulmonary function
tests (FVC, FEVj, FEF25.75, FEF25.75/FVC, PEF)
were performed. 1,672 children had at least two
pulmonary function test data. Air pollutants
examined include O3, NO2, PM10, PM2 5, inorganic
acid, elemental carbon, and organic carbon.
Adjusted linear regression model was used.
First cohort of the Children's Health Study.
Association of O3 with prevalence of chronic lower
respiratory tract symptoms among children with a
history of asthma was examined in a cross-sectional
study in 12 communities. Questionnaires were
completed by parents of 3,676 4th, 7th, and 10th
graders, of which 493 had asthma. Exposure data
(O3 NO2, PM10, PM25, and inorganic acid vapors)
collected in 1994 used to estimate exposure.
Analysis using logistic regression method.
In the 7th and 10th grade cohorts, difference in lung function
growth from the least to the most polluted community was not
associated with any of the air pollutants, including O3. Among the
4th graders, decreased lung growth was associated with exposures
to PM25, NO2, acid vapor, and elemental carbon, but not with O3.
It should be noted, however, that there was only about a 2- to
2.5-fold difference in O3 concentrations from the least to most
polluted communities, compared to the other pollutants which had
a 4- to 8-fold difference in concentrations. In addition, these
results should be interpreted with caution given the potential for
substantial misclassification of O3 exposure.
In this cohort, a significant association between O3 and PEF and
FVC was noted in children spending more time outdoors.
Percent difference in annual increases in lung function from least
to most polluted community (per 36.6 ppb increase in annual mean
8-h avg 03):
PEF:
All children: -1.21% (-2.06, -0.36)
Children more outdoors: -1.62% (-2.93, -0.29)
Children less outdoors: -0.87% (-2.09, 0.37)
Children with asthma were much more likely to have bronchitis or
related symptoms than children without such history. Among the
asthmatic children, significant relationship were observed between
phlegm and all pollutants studied, with the exception of O3.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
X
United States (cont'd)
McConnell et al.
(2003)
12 Southern
California
communities
1996-1999
8-h avg O3
(10 a.m. -6 p.m.):
4-year average across
12 communities:
47.2 ppb
SD11.3
Range 28. 3-65. 8
Range of yearly
variability within the
12 communities:
5.3 ppb
SD3.2
Range 1.7-13.2
A total of 475 children with asthma from the 1st and
2nd cohorts of the Children's Health Study were
recruited to examine the relationship between
bronchitic symptoms and air pollutants. Analysis
involved three stages using logistic mixed effects
models. Within-community variability in air
pollution was assessed in the first stage; individual
level time-independent confounders were assessed
in the second stage; and the effects of 4-year
average air pollutants were examined in the third
stage. Other copollutants examined include NO2,
PM10, PM25, PM10_25, inorganic acid, organic acid,
EC, and OC.
Symptoms were generally associated with the various air
pollutants. Within-community effects were greater than the
between-community effects. Authors note that if the larger within-
community effect estimates are correct, then other cross-sectional
(between-community) studies may have underestimated the true
effect of air pollution on respiratory symptoms in children. These
differences may be attributable to confounding by poorly measured
or unmeasured risk factors that vary between communities. Ozone
effect estimates were markedly reduced in two-pollutant models
(odds ratios not provided).
Odds ratios for bronchitic symptoms (per ppb O3):
Within-community effects: 1 .06 (1 .00, 1 . 12)
Between-community effects: 0.99 (0.98, 1.01)
McConnell et al.
(2002)
12 Southern
California
communities
1993-1998
l-hmaxO3:
4-year average
(1994-1997):
Low-pollution
communities (n = 6):
50.1 ppb
Range 37.7-67.9
High-pollution
communities (n = 6):
75.4 ppb
Range 69.3-87.2
Prospective cohort study of 3,535 children (age
9-16 years) without a history of asthma recruited in
1993 and 1996, and followed with annual surveys
through 1998 to determine incidence of new onset
asthma. Participation in sports assessed at baseline.
Copollutants included PM10, PM25, NO2, and
inorganic acid vapors. Asthma incidence was
examined as a function of number of sports played
in high- and low-pollution communities, controlling
for age, sex, and ethnic origin.
Asthma incidence was not higher in the high-pollution
communities as compared with the low-pollution communities,
regardless of the pollutant used to define high/low. In fact,
the high-O3 communities had generally lower asthma incidence.
However, in high-O3 communities, there was an increased risk of
asthma in children playing three or more sports vs. those playing
no sports; no such increase was observed in the low-O3
communities. No other pollutant showed this association. These
results suggest that high levels of physical activity is associated
with risk of new asthma development for children living
in communities with high O3 levels.
Relative risk of developing asthma in children playing three or
more sports vs. those playing no sports:
Low-pollution communities: 0.8(0.4, 1.6)
High-pollution communities: 3.3 (1.9, 5.9)
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
X
United States (cont'd)
Avoletal. (2001)
12 Southern
California
communities and six
western states
Baseline 1994
Follow-up 1998
Ritz and Yu( 1999)
Southern California
1989-1993
Ritz et al. (2000)
Southern California
1989-1993
8-h avg O3
(10 a.m.-6 p.m.):
Mean range of baseline
levels:
30.4 ppb for Santa
Maria to 70.8 ppb for
Lake Gregory
Mean range of changes:
11.7 ppb increase to
27.0 ppb decrease in
O3 levels
Last trimester average:
20.9 ppb
Range 3.0-49.4
8-h avg O3
(9 a.m.-5 p.m.):
1st month of
pregnancy:
36.9 ppb
SD20.7
Range 2.6-124
6 weeks before birth:
36.9 ppb
SD 19.4
Range 3.3-117
110 children enrolled in Children's Health Study in
1993 and 1994 followed after moving to different
western states. Age 10-11 years at time of
enrollment). Follow-up pulmonary function testing
carried out in 1998. Change in lung function over
time tested in relation to change in exposures to
PM10, N02, and O3.
125,573 births within 2 miles of an air monitoring
station were examined to determine associations
between CO and low birth weight. Copollutants
included only as potential confounders.
Data on 97,158 singleton births over period
1989-1993 linked to air monitoring data during
different periods of pregnancy to determine
associations between pollution exposures and
preterm birth. Besides O3, pollutants of interest
included PM10, NO2, and CO. Multiple regression
analysis used, controlling for maternal age, race,
education, parity, and other factors.
Negative but nonsignificant associations were found between lung
function parameters and changes in O3. The relationship was
strongest with PM10. Subjects who moved to areas of lower PM10
showed greater increases in FEV1 compared subjects who moved
to areas with higher PM10.
Change in lung function (per 10 ppb increase in changes in annual
mean 8-h avg O3):
FEVI: 0.1 mL (-8.7, 8.9)
FVC: -1.4 mL (-10.8, 8.0)
MMEF (maximal midexpiratory flow): -3.4 mL/s (-23.6, 16.8)
PEF: -8.9 mL/s (-41.6, 23.8)
Exposure to higher levels of ambient CO during the last trimester
was associated with a significantly increased risk for low birth
weight. Effects of CO appeared more pronounced after adjustment
for concurrent exposures to NO2, PM10, and O3. Ozone effect
estimates were not reported.
Both PM10 and CO during early or late pregnancy were associated
with increased risk for preterm birth. No associations observed
with O,.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
United States (cont'd)
Rite et al. (2002)
Southern California
1987-1993
X
o
oo
Kilnzlietal. (1997);
Tageretal. (1998)
Los Angeles and San
Francisco, CA;
Berkeley, CA
1995
24-h avg O3:
Quartile means:
1 st month of gestation:
6.4,15.2,23.9,
34.2 ppb
2nd month:
6.4,15.6,24.2,
34.9 ppb
3rd month:
6.8,16.6,24.9,
34.9 ppb
8-h avg O3
(10 a.m.-6 p.m.):
Lifetime monthly
Los Angeles:
Median 51.5 ppb
Range 25-74
San Francisco:
Median 22.5 ppb
Range 16-33
The effect of air pollution on the occurrence of birth
defects was examined using a case-control study
design. Analyses focused on CO and O3. Six types
of cardiac birth defects were investigated - aortic
defects, defects of the atrium and atrium septum,
endocardial and mitral valve defects, pulmonary
artery and valve defects, conotruncal defects, and
ventricular septal defects. A 1:10 case-control ratio
was achieved for defect-specific analyses. Analyses
were conducted using polytomous logistic
regression. A two-stage hierarchical regression
model was used to adjust for multiple comparisons.
A pilot cohort study of 130 freshman students
(age 17-21 years) at the University of California at
Berkeley measured for lung function and histories
of residential locations and indoor/outdoor activity
patterns and levels. By design, students had
previously resided in one of two metropolitan areas
that differed greatly in O3 concentrations,
San Francisco or Los Angeles. A key goal was to
test whether measures of small airways function
(e.g., nitrogen washout, FEF25.75, FEF75) were
sensitive measures of long-term O3 impacts.
Lifetime exposures to O3, PM10 and NO2 assigned
by interpolation to sequence of residence locations
from available monitoring stations. Multiple
exposure measures were derived with varying
degrees of incorporation of time-activity
information, going from ecological concentration
to individual time-activity weighted exposure.
Performed linear regression analysis of lung
function on O3 exposures, controlling for height,
ethnicity, gender, and region.
Concentration-response patterns were observed for O3
concentrations during the 2nd month of gestation on aortic artery
and valve defects, pulmonary artery and valve anomalies, and
conotruncal defects. CO during the 2nd gestational month was
found to be associated with ventricular septal defects. The results
were inconclusive for NO2 and PM10. Findings from this study
suggest that there may be a vulnerable window of development to
human malformations.
Odds ratios for birth defects (per 10 ppb O3 during 2nd month of
gestation):
Aortic defects: 1.56(1.16,2.09)
Pulmonary valve defects: 1.34 (0.96, 1.87)
Conotruncal defects: 1.36 (0.91, 2.03)
Decreased FEF25.75 and FEF75 were associated with long-term O3
exposures. Results were similar whether O3 exposure was purely
ecologic or incorporated time-activity information. FVC, FEV^
and nitrogen washout were generally not associated with O3 levels.
No evidence for PM10 or NO2 main effects or confounding of O3.
Similar patterns results using O3 hours >60 ppb as exposure metric
instead of daily 8-h avg O3. Surprisingly, region of residence was
a major negative confounder as lung function was lower on
average among students from the low O3 city, San Francisco, than
among those who had lived in Los Angeles. Ozone exposures
were significant predictors only after controlling the regional
effect.
Change in lung function (per 20 ppb increase in lifetime mean
8-h avg O3):
FEF25.75: -420 mL/s (-886, 46); 7.2% of population mean
FEF75: - 334 mL/s (-657, -11); 14% decline of population mean
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
United States (cont'd)
Tager et al. (2005)
Los Angeles and
San Francisco, CA;
Berkeley, CA
2000-2002
X
Gongetal. (1998b)
Glendora, CA
1977-1987
8-h avg O3
(10 a.m.-6 p.m.):
Lifetime monthly
average:
Men:
36 ppb
Range 19-64
Women:
33 ppb
Range 18-65
Difference in median
exposure between
Los Angeles and
San Francisco:
17 ppb
l-hmaxO3:
Annual means range
(1983-1987):
109-134 ppb
Similar study design to that of Kilnzli et al. (1997).
A cohort study of 255 freshman students (age
16-19 years) at the University of California at
Berkeley who previously resided in San Francisco
or Los Angeles measured for lung function between
Feb and May. Subjects were lifetime never-smokers
without history of chronic respiratory diseases.
Histories of residential locations, and
indoor/outdoor activity patterns and levels also
evaluated. Lifetime exposures to O3, PM10, and NO2
assigned by interpolation to sequence of residence
locations from available monitoring stations.
Multiple exposure measures were derived with
varying degrees of incorporation of time-activity
information, going from ecological concentration
to individual time-activity weighted exposure. The
base lung function model consisted of gender-
specific linear regressions of each lung function
measure on age, height, and weight. Modification
of the air pollution effect on lung function by the
FEF25.75/F VC ratio (a measure of intrinsic airway
size) was also examined.
Longitudinal cohort study of 164 adults (mean age
45 years; 34% males) from a high O3 community
underwent lung function testing in 1986-1987 (T3).
Subjects were recruited from a cohort of 208
nonsmoking adults who had been tested on two
previous occasions: 1977-1978 (Tl) and 1982-1983
(T2). Analyzed changes in lung function at three
time points. Subjects were also asked to undergo
controlled exposures to 0.40 ppm O3 over 2 h with
intermittent exercise. 45 subjects agreed to
participate. Investigators hypothesized that acutely
responsive subjects would show more rapid declines
in function over the study period.
In the models examining only the main effects of O3, no
associations were observed between lifetime exposure to O3 and
lung function parameters in either men or women. However, after
including an interaction term for FEF25.75/FVC ratio, consistent
inverse associations between increasing lifetime exposure to O3
and FEF25.75 and FEF75 were observed for both men and women.
The adverse impact of increased lifetime exposure to O3 was found
to decrease with increasing FEF25.75/FVC ratio. Comparable
regressions with PM10 and NO2 showed similar results. When
PM10 and NO2 were added to the O3 model for FEF75, there were no
meaningful changes in the effect estimates for lifetime exposure to
O3 or the interaction term while the effect parameters for PM10 and
NO2 were reduced substantially. The potential effect of
measurement error from the use of questionnaire responses to
estimate exposure to O3 was assessed. The measurement error
correction did not have a meaningful effect on the magnitude of
the association with O3.
Change in lung function (per 17 ppb increase in lifetime mean 8-h
avg O3), after including interaction term for FEF25.75/FVC ratio:
-19% (-39, 1)
FEF75: -37% (-41,-32)
FEF25.75: -29% (-33,24)
Mean FVC and FEV[ showed nonsignificant increase from T2 to
T3, whereas an earlier analysis of the Tl to T2 change had found a
significant decline in function (Betels et al., 1987). There was
evidence for "regression to the mean," in that subject with larger
declines in function from Tl to T2 tended to have larger increases
in function from T2 to T3. A consistent decline in FEVj/FVC ratio
was observed at all three time points (p < 0.0001 by ANOVA).
Acute changes in lung function, determined using controlled O3
exposures, were not associated with chronic lung function changes.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
United States (cont'd)
Chen et al. (2002)
Washoe County, NV
1991-1999
Kinneyetal. (1996b)
New York City
1992-1993
X
8-h max O3: Birth weight for 36,305 single births analyzed
27.23 ppb in relation to mean PM10, O3, and CO levels in 1 st,
SD 10.62 2nd, and 3rd trimesters.
Range 2.76-62.44
1-h max O3: Study of 19 healthy adult joggers (age 23-38 years;
18 males) from the Governors Island U.S. Coast
Summer Guard facility in New York harbor underwent a
(Jul-Sep 1992): series of two bronchoalveolar lavages, first in the
58 ppb summer of 1992 and then again in the winter of
Maximum 100 1992. Because the summer of 1992 had lower than
average O3 levels, six subjects underwent a third
Winter bronchoalveolar lavage in the summer of 1993.
(Jan-Mar 1992): Study tested whether inflammatory markers in
32 ppb bronchoalveolar lavage fluid were elevated during
Maximum 64 the summer O3 season among adults who regularly
exercised outdoors. Outcomes included cell
Summer differentials, release of IL-8 and TNF-a in
(Jul-Sep 1993): bronchoalveolar lavage cells supernatants, release
69 ppb of reactive oxygen species by macrophages, and
Maximum 142 concentrations of protein, lactate dehydrogenase,
IL-8, fibronectin, al-antitrypsin (al-AT),
complement fragments (C3a), and PGE2 in
bronchoalveolar lavage fluids.
PM10 was the only air pollutant associated with decreased birth
weights. Ozone levels quite low throughout study.
There was no evidence of acute inflammation in the summer of
1992 compared to the winter; i.e., neutrophil differentials, IL-8 and
TNF-a showed no significant differences. However, a measure of
cell damage, lactate dehydrogenase, was elevated in the summer,
suggesting possible O3-mediated damage to the lung epithelium
with repeated exposures to O3 while exercising. O3 levels during
the summer of 1992 were atypically low for New York City.
Among six subjects who agreed to undergo a third bronchoalveolar
lavage test in the summer of 1993, lactate dehydrogenase was
again elevated compared to winter. In addition, IL-8 was elevated
in the summer of 1993, suggesting acute inflammation.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Mean O3 Levels
Study Description
Results and Comments
Europe
Charpinetal. (1999)
Seven towns in
SE France
Jan-Feb 1993
X
Ramadour et al.
(2000)
Seven towns in
SE France
Jan-Feb 1993
8-h max O3: Cross-sectional cohort study of 2,073 children (age
Range of means: 10-11 years) from seven towns tested for atopy
30.2-52.1 ug/m3 based on skin prick testing (house dust mite, cat
dander, grass pollen, cypress pollen, and
24-h avg O3: Alternaria). Towns represented a range of ambient
Range of means: O3 and other pollutant (NO2 and SO2) levels. Tested
20.1-42.1 ug/m3 hypothesis that atopy is greater in towns with higher
photochemical pollution levels. To be eligible,
subjects must have resided in current town for at
least 3 years. Authors stated that Jan to Feb
pollution levels correlated with levels observed
throughout the year, though no data was given to
support this.
8-h max O3: Cross-sectional cohort study of 2,445 children
Range of means: (age 13-14 years) who had lived at their current
30.2-52.1 ug/m3 residence for at least 3 years were recruited from
schools in seven towns in SE France. This region
has highest O3 levels in France. Subjects completed
ISAAC survey of asthma and respiratory symptoms.
In addition to O3 also collected data on SO2 and
NO2. Analyzed relationships between asthma and
other respiratory conditions with mean air pollution
levels across the seven towns using logistic
regression, controlling for family history of asthma,
personal history of early-life respiratory diseases,
and SES. Also did simple univariate linear
regressions.
In this cross-sectional analysis, no differences in atopy levels were
seen across the seven towns. Authors concluded that long-term
exposures to oxidant pollution do not favor increased allergy to
common allergens. The very low winter O3 levels monitored and
lack of long-term exposure data make it impossible to reach this
conclusion in a definitive manner.
In logistic regressions, no significant associations seen between O3
and 12-mo history of wheezing, history of asthma attack, exercise
induced asthma and/or dry cough in last 12 months.
In simple bivariate scatterplots of respiratory outcomes vs. mean
O3 levels in the seven towns, there appeared to be strong positive
relationships (r = 0.71 for wheezing in last 12 months and r = 0.96
for asthma attacks). No data on slope estimates given. Concerns
about potential confounding across towns limits the interpretation
of this study.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Study Description
Results and Comments
X
Europe (cont'd)
Ihorst et al. (2004)
Nine communities in
Lower Austria
Apr 1994-Oct 1997
Six communities in
Germany
Febl996-0ctl999
Kopp et al. (2000)
Ten communities in
Austria and SW
Germany
Mar 1994-Nov 1995
!/2-h avg O3:
Semi-annual mean:
Quartile ranges:
Summer:
Istquartile: 22-30 ppb
2ndquartile: 30-38
ppb
3rd quartile: 38-46 ppb
4th quartile: 46-54 ppb
Winter:
Istquartile: 4-12 ppb
2nd quartile: 12-20
ppb
3rd quartile: 20-28 ppb
4th quartile: 28-36 ppb
!/2-h avg O3:
Mean from Apr-Oct
1994:
Stratified by low,
medium, high
exposure:
Low: 24-33 ppb
Medium: 35-38 ppb
High: 44-52 ppb
Longitudinal cohort study of 2,153 children (median
age 7.6 years) were studied for the effects of semi-
annual and 3'/2-year mean O3 concentrations on
FVC and FEV^ As a measure of lung growth,
the difference between two consecutive values for
each child was divided by the number of days
between tests. The effect of O3 exposure on lung
growth was analyzed by linear regression models,
after adjusting for sex, age, height at start of the
time period, and passive smoking exposure.
Longitudinal cohort study of 797 children with a
mean age of 8.2 years. Four pulmonary function
tests (FVC, FEVj) performed on each child over two
summers. Examined association between average
daily lung function growth and exposure to O3,
PM10, NO2, and SO2. Analysis using linear
regression models.
Higher semi-annual mean O3 levels were associated with
diminished lung function growth during the summer, but
increased lung function growth in the winter.
Change in lung function (4th quartile vs. 1 st quartile semi-annual
O3 mean):
Summer:
FEVj (mL/100 days): -18.5 (-27.1,-9.8)
FVC (mL/100 days): -19.2 (-27.8, -10.6)
Winter:
FEV; (mL/100 days): 10.9(2.1,19.7)
FVC (mL/100 days): 16.4(8.3,24.6)
No associations between longer term O3 exposure (mean summer
O3 over a 3!/2-year period) and lung function growth was found.
Lower FVC and FEV[ increases were observed in children exposed
to high ambient O3 levels vs. those exposed to lower O3 levels
during the summer. During the winter, children in higher O3 areas
showed a slightly greater increase in FVC and FEV[ than those in
the low O3 areas, which might reflect that children make up for
lung function deficits during the winter season.
Change in lung function for high vs. low O3 exposure groups (per
ppb 03):
FEVI:
Summer of 1994: -0.303 mL/day, p = 0.007
Winter of 1994/1995: 0.158 mL/day, p = 0.006
Summer of 1995: -0.322 mL/day, p = 0.001
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Study Description
Results and Comments
X
Europe (cont'd)
Frischeretal. (1999)
Nine communities in
Austria
1994-1996
Frischeretal. (2001)
Nine communities in
Austria
Sep-Octl997
24-h avg O3:
Summer:
34.8ppb
SD8.7
Winter:
23.1ppb
SD7.7
!/2-h avg O3:
30-day mean:
31.57ppb
IQR 20.61
Longitudinal cohort study of communities from two
counties chosen to represent a broad range of O3
concentrations; a 2-fold range in mean levels was
observed. 1,150 children (mean age 7.8 years;
52% males) from grades 1 and 2 performed
spirometry in spring and fall over 3 years (total of
six measurements per child) to determine if seasonal
exposure to O3 would be associated with diminished
lung function growth, especially over the summer
seasons. Ozone levels were low during lung
function testing periods. Participation rates were
high. At baseline, respiratory histories were
collected and subjects were tested for allergy by
skin prick. Examined association between O3 levels
and change in lung function (FVC, FEVj, and
MEF50 [maximal expiratory flow at 50% of vital
capacity]) over each season, controlling for baseline
function, atopy, gender, site, environmental tobacco
smoke exposure, season, and change in height.
Other pollutants studied included PM10, SO2, and
NO,.
A cross-sectional cohort study of 877 school
children (mean age 11.2 years). Analyzed
for urinary eosinophil protein as a marker
of eosinophil activation determined from a single
spot urine sample using linear regression models.
Seasonal mean O3 exposures were associated with reductions in
growth in all three lung function measures. Inconsistent results
seen for other pollutants. Summer season lung function growth
decrements per unit O3 were larger when data restricted to children
who spent whole summer in their community. No evidence for
nonlinear O3 effect. No confounding of O3 effect by temperature,
ETS, or acute respiratory illnesses.
Change in lung function (per ppb O3):
FEV; (mL/day):
All subjects:
Summer: -0.029 (SE 0.005), p< 0.001
Winter: -0.024 (SE 0.006), p< 0.001
Restricted to subjects who stayed in community:
Summer: -0.034 (SE 0.009), p< 0.001
FVC (mL/day):
All subjects:
Summer: -0.018 (SE 0.005), p< 0.001
Winter: -0.010 (SE 0.006), p = 0.08
Restricted to subjects who stayed in community:
Summer: -0.033 (SE 0.007), p< 0.001
Log-transformed urinary eosinophil protein-X concentrations were
found to be significantly associated with O3 levels, after adjusting
for gender, site, and atopy.
Change in log urinary eosinophil protein-X (per ppb O3):
0.007 ug/mmol creatinine (SE 0.02), p < 0.001
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Study Description
Results and Comments
Europe (cont'd)
Horak et al. (2002a,b)
Eight communities in
Austria
1994-1997
X
Palli et al. (2004)
Florence, Italy
1993-1998
Summer:
31.8ppb
Range 18.7-49.3
Winter:
19.8ppb
Range 12.7-35.9
24-h avg O3:
Range of monthly
means from 1993-1998:
Approximately
25-125 ppb
This longitudinal cohort study continued the work
of Frischer et al., 1999 by including one additional
year of data, 1997. The major hypothesis
considered PM10. For this study, 80.6% of
the 975 children (mean age 8.11 years) performed
all six lung function tests. A total of 860 children
were included in the GEE analysis. Multipollutant
analysis for PM10, SO2, and NO2.
Cohort study of 320 residents (age 35-64 years) in
the metropolitan area of Florence enrolled in a study
investigating the correlation between levels of DNA
bulky adducts and cumulative O3 exposure.
One blood sample was collected for each subject.
Various time windows of exposure were examined,
ranging from 0-15 days to 0-90 days prior to the
blood draw. Simple Spearman correlations between
DNA adduct levels and different O3 exposure time
windows were calculated after stratifying by
smoking history, area of residence, and population
type (random sample or exposed workers).
Seasonal mean O3 showed a negative effect on lung function
growth, confirming the previous shorter study. Ozone effects were
robust to inclusion of PM10 into the model. However, for FEV; in
winter, the O3 effect slightly diminished after including PM10.
Taking into account only children who stayed at home the whole
summer period did not affect the results.
Change in lung function (per ppb O3):
FEV; (mL/day):
O3 only models:
Summer: -0.021, p< 0.001
Winter: -0.020, p < 0.001
O3 with PM10 models:
Summer: -0.020, p< 0.001
Winter: -0.012, p = 0.04
Consistent relationships between O3 exposure and DNA adduct
levels were observed only among never smokers. Correlations
were highest among never smokers who resided in the urban area
and were not occupationally exposed to vehicle traffic pollution.
Associations were significant up to a time window of 0-60 days
prior to the blood draw in the subgroup of never smokers, with
strongest relationships observed between 30-45 days prior.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period
Study Description
Results and Comments
X
Latin America
Calderon-Garciduenas
etal. (1995)
SW Mexico City
Nov 1993
Manzanillo, Mexico
Jan 1994
l-hmaxO3:
SW Mexico City
(urban):
4.4 h/day >120 ppb
Maximum 307 ppb
Manzanillo, Pacific
port (control):
No detectable air
pollutants.
A cohort study in which nasal lavage samples
collected from 38 urban (mean age 12.2 years) and
28 control (mean age 11.7 years) children. Urban
children monitored four times in November;
controls monitored twice in January. Samples
analyzed for polymorphonuclear leukocyte counts,
expression of human complement receptor type
3 (GDI Ib) on nasal polymorphonuclear leukocytes,
and nasal cytologies.
Nasal cytologies revealed that children from Mexico City had
abnormal nasal mucosae, including mucosal atrophy, marked
decreases in the numbers of ciliated-type cells and goblet cells,
and squamous metaplasia.
Exposed children had significantly higher nasal
polymorphonuclear leukocyte counts (p < 0.001) and nasal GDI Ib
expression (p < 0.001) vs. controls. However, the inflammatory
response did not seem to correlate with the previous day's O3
exposure, suggesting that there might be a competing
inflammatory mechanism at the bronchoalveolar level. Overall,
these results suggest that ambient O3 produces an inflammatory
response in chronically exposed children. However, it should be
noted that urban children were exposed to a complex pollution
mix, making it difficult to attribute effects to O3 specifically.
Gouveia et al. (2004)
Sao Paulo, Brazil
1997
l-hmaxO3
63.0 ppb
SD33.5
Birth weight for 179,460 single births analyzed
in relation to PM10, SO2, CO, NO2, and O3 levels in
1 st, 2nd, and 3rd trimesters. GAM and logistic
regression models used for analysis.
Exposures to PM10 and CO during 1 st trimester were found to
have significant negative associations with birth weight.
No associations observed for the other air pollutants, including O3.
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period Mean O3 Levels
Study Description
Results and Comments
Asia
Haetal. (2001)
Seoul, Korea
1996-1997
8-h avg O3:
1 st trimester:
Median 22.4 ppb
IQR 13.6
3rd trimester:
Median 23.3 ppb
IQR 16.1
X
Examined association between air pollution
exposure during pregnancy and low birth weight
among all full-term births for a 2-year period.
These associations were evaluated after adjusting
for gestational age, maternal age, parental
educational level, parity, and infant sex. Analysis
using GAM with default convergence criteria.
Exposures during the 1st and 3rd trimesters were initially
examined separately. For the 1st trimester exposure estimates,
positive associations with risk of low birth weight were observed
for CO, NO2, SO2, and TSP, but not O3. For the 3rd trimester
exposure estimates, a positive association was observed for O3, but
not the other pollutants. When exposures from both trimesters
were examined simultaneously, the risk of low birth weight
remained positive for CO, NO2, SO2, and TSP during the 1st
trimester. However for O3, the positive association with 3rd
trimester exposure was diminished. These results suggest that
exposures to CO, NO2, SO2, and TSP in the 1st trimester may be
risk factors for low birth weight.
Relative risk of low birth weight (per 13.6 ppb 8-h avg O3 for 1 st
trimester; per 16.1 ppb 8-h avg O3 for 3rd trimester):
Stratified analyses by trimester:
Isttrimester: 0.92(0.88,0.96)
3rd trimester: 1.09(1.04,1.14)
Combined analyses of both trimesters:
Isttrimester: 0.96 (0.87, 1.07)
3rd trimester: 1.06(0.94,1.18)
-------
Table AX7-6 (cont'd). Effects of Chronic O3 Exposure on Respiratory Health
Reference, Study
Location, and Period Mean O3 Levels
Study Description
Results and Comments
Asia (cont'd)
Kuo et al. (2002)
Central Taiwan
1996
l-hmaxO3:
Annual mean range
across 7 of 8 schools:
18.6-27.3 ppb
X
Cross-sectional study. Respiratory questionnaire
administered to 12,926 children aged 13-16 years at
eight junior high schools in central Taiwan, to
determine asthma prevalence. The association
between asthma prevalence and air pollution
exposure analyzed by simple Pearson correlations of
prevalence with annual mean air pollution levels
(O3, SO2, PM10, and NO2), and by multiple logistic
regression. The 775 asthmatics who were identified
then provided follow-up data on symptoms and
exacerbations over a 1-year period. Simple Pearson
correlations were computed between monthly
hospital admissions and air pollution levels, not
controlling for covariates such as season or weather.
Asthma prevalence ranged from 5.5% to 14.5% across the 8
schools. Based on simple Pearson correlations, mean O3 (r = 0.51)
and NO2 (r = 0.63) levels were correlated with variations in asthma
prevalence. However, only NO2 remained significant in multiple
logistic regression analyses after adjusting for various potential
confounding factors.
Longitudinal hospital admissions results are inconclusive due
to analytical limitations. Monthly correlations of hospital
admissions for asthmatics yielded variable results, all of which
would be confounded by temporal factors.
-------
Table AX7-7. Effects of Chronic O3 Exposure on Mortality and Incidence of Cancer
Reference, Location,
Study Period
Mean O3 Levels
Study Description
Results and Comments
United States
Pope et al. (2002)
U.S. nationwide
1982-1998
l-hmaxO3:
Mean from Jul-Sep:
59.7 ppb
SD 12.8
Prospective cohort study of approximately 500,000
members of American Cancer Society cohort enrolled
in 1982 and followed through 1998 for all cause,
cardiopulmonary, lung cancer, and all other cause
mortality. Age at enrollment was 30+ years. Air
pollution concentrations in urban area of residence at
time of enrollment assessed from 1982 through 1998.
Other pollutants considered include TSP, PM15, PM10,
PM25, PM15.25, SO/', S02, N02, and CO.
No significant effect of O3 on mortality risk, though the
association of Jul-Sep O3 concentrations with all cause
and cardiopulmonary mortality were positive and nearly
significant.
Residential location was known only at enrollment to
study in 1982. Thus, exposure misclassification is
likely to be high.
X
oo
Lipfert et al. (2000b;
2003)
32 Veterans
Administration hospitals
nationwide in the U.S.
1976-1996
95th % O3:
1960-1974: 132 ppb
1975-1981: 140 ppb
1982-1988: 94 ppb
1989-1996: 85 ppb
Cohort study of approximately 50,000 U.S. veterans
(all males) diagnosed with hypertension. Mean age at
recruitment was 51 years. Exposure to O3 during four
periods (1960-1974, 1975-1981, 1982-1988, 1989-
1996) associated with mortality over three periods
(1976-1981, 1982-1988, 1989-1996). Long-term
exposures to TSP, PM15, PM10, PM25, PM15.25, SO42~,
NO2, and CO also analyzed. Used Cox proportional
hazards regression, adjusting for race, smoking, age,
systolic and diastolic blood pressure, body mass index,
and socioeconomic factors.
Positive average concurrent responses for TSP, SO42~,
NO2, O3 in individual period analyses, but only O3 was
significant for overall. Two-pollutants analyses indicate
that responses to peak O3 are robust.
Relative risks (per mean 95th % O3 less estimated
background level, value not reported):
Averaged over all four periods:
Exposure concurrent with mortality:
O3 only model:
9.4% (SE 4.6), p < 0.05
O3 with NO2 model:
12.2%, p < 0.05
Exposure before mortality:
O3 only model:
-0.2% (SE 6.3) p> 0.05
Analyses were robust to the deletion of diastolic blood
pressure in the models, indicating that the association
between mortality and O3 was not mediated through
blood pressure.
-------
Table AX7-7 (cont'd). Effects of Chronic O3 Exposure on Mortality and Incidence of Cancer
Reference, Location,
Study Period
Mean O3 Levels
Study Description
Results and Comments
X
United States (cont'd)
Abbey etal. (1999)
Three California air
basins: San Francisco,
South Coast (Los Angeles
and eastward), San Diego
1977-1992
Beesonetal. (1998)
Three California air
basins: San Francisco,
South Coast (Los Angeles
and eastward), San Diego
1977-1992
24-h avg O3:
Monthly average from
1973 to censor date:
26.11ppb
SD 7.65
IQR 12.0
O3h/year>100ppb:
330 h/year
SD295
IQR 551
24-h avg O3:
Monthly average from
1973 to 1992:
26.2 ppb
SD7.7
O3h/year>100ppb:
333 h/year
SD297
IQR 556
Prospective cohort study of 6,338 nonsmoking non-
Hispanic white adult members of the Adventist Health
Study followed for all cause, cardiopulmonary,
nonmalignant respiratory, and lung cancer mortality.
Participants were aged 27-95 years at enrollment in
1977. 1,628 (989 females, 639 males) mortality events
followed through 1992. All results were stratified by
gender. Used Cox proportional hazards analysis,
adjusting for age at enrollment, past smoking,
environmental tobacco smoke exposure, alcohol use,
education, occupation, and body mass index. Analyzed
mortality from all natural causes, cardiopulmonary,
nonmalignant respiratory, and lung cancer. Ozone
results were presented for both metrics.
Prospective cohort study of 6,338 nonsmoking non-
Hispanic white adult members of the Adventist Health
Study aged 27-95 years at time of enrollment.
36 (20 females, 16 males) histologically confirmed lung
cancers were diagnosed through 1992. Extensive
exposure assessment, with assignment of individual
long-term exposures to O3, PM10, SO42", and SO2, was a
unique strength of this study. All results were stratified
by gender. Used Cox proportional hazards analysis,
adjusting for age at enrollment, past smoking,
education, and alcohol use.
Of 16 regressions involving O3 exposures (two genders;
four mortality causes; two O3 metrics), 11 were positive
and one was statistically significant, for lung cancer in
males for O3 h/year >100 ppb.
Relative risks for lung cancer mortality in males:
24-h avg O3 (per 12.0 ppb):
2.10(0.99,4.44)
O3 h/year >100 ppb (per 551 h/year):
4.19(1.81,9.69)
Inconsistency across outcomes and genders raises
possibility of spurious finding. The lack of
cardiopulmonary effects raises plausibility concerns.
Males, but not females, showed moderate association
for O3 and incident lung cancer risk.
Relative risks for lung cancer incident in males:
O3 h/year >100 ppb (per 556 h/year):
All males: 3.56(1.35,9.42)
Never smokers: 4.48(1.25,16.04)
Past smokers: 2.15 (0.42, 10.89)
-------
Table AX7-7 (cont'd). Effects of Chronic O3 Exposure on Mortality and Incidence of Cancer
Reference, Location,
Study Period
Mean O3 Levels
Study Description
Results and Comments
Latin America
Pereira et al. (2005)
Sao Paulo, Brazil
Exposure period
1981-1990
Case period
1997
Mean avg of days/year
when O3 levels exceeded
air quality standards (units
not provided):
Lapa: 40.2
Moema: 19.6
Mooca: 67.1
Se: 28.2
Annual records on larynx and lung cancer diseases
obtained from the Sao Paulo Cancer Registry. The
correlation between average air pollution data from
1981-1990 and cases of larynx and lung cancer from
1997 were assessed using Pearson correlation
coefficients. Other pollutants examined included PM10,
N02, NOX5 S02, and CO.
Results from this ecologic study provide limited
evaluation of the relationship between air pollution and
cancer. There was a significant difference in the
incidence of larynx and lung cancer among the Sao
Paulo city districts.
Of all the pollutants examined O3 was best correlated
with cases of larynx and lung cancer.
Pearson correlation coefficient:
X
Larynx cancer: 0.9929 (p = 0.007)
Lung cancer: 0.7234 (p = 0.277)
to
o
-------
AX7.2 DESCRIPTION OF SUMMARY DENSITY CURVES
Introduction
The summary density was used in various figures in Chapter 7. It is important that the
reader not confuse a summary density with an error density, which was also used in Chapter 7
figures. This section will explain the relationship between these two densities, previous use of
similar densities, the theory behind the summary density, and how to interpret the graph of the
summary density.
In explaining the equation for the summary density and the interpretation of it, the
discussion will explain the need for the summary density and its construction. First, the
preliminaries will be discussed. The statistically experienced reader can skip this portion except
to read the definition of the error density.
The summary density has been used before (Flachaire and Nunez, 2002). There, summary
density was weighted by the population size and was used in an economic context to estimate
income distributions. A meta-analytic method for use in the presence of non-normal
distributions of effects with varying precision has been developed (Burr and Doss, 2005). This
was used in an analysis of effects from multiple studies concerning the association of the platelet
P1A polymorphism of glycoprotein Ilia and risk of coronary heart disease. A related density
estimate, the kernel density estimate, has also been used in a publication (Kochi et al., 2005)
referenced in a White Paper (Dockins et al., 2004) presentation to the U.S. EPA Science
Advisory Board-Environmental Economics Advisory Board.
Preliminaries
The error density is the curve describing the distribution of uncertainty about the mean (or
posterior mean) or the slope estimate. For a normal density, this is sometimes referred to as the
"bell-shaped curve." With a two-sided test, the area under the curve and to the left of the
no-effect value for a positive (or right of the value for negative) estimate is less than 2.5% when
the effect estimate is statistically significant.
The log odds and the log relative risk estimates are usually considered to have a normal
distribution when the estimate is based on a large number of observations. Many times when the
health effect is a continuous variable, the estimate (or a transformation of it) is assumed to have
a normal distribution.
AX7-121
-------
The two-sided confidence intervals for a health effect estimate are based on the error
density of the estimate. The confidence intervals include 95% of the area under the error
density, with equal portions of area beyond the intervals.
In displaying confidence limits for effect estimates obtained under different conditions
(relative risk for different cities or studies), the stick diagram is often used (for an example, see
Figure 7-1 in Section 7.2.3.2). The length of the line (stick) represents the confidence limit, with
the estimated mean at the midpoint of this line. With confidence limits for relative risks or odds
ratios, the mean estimates may not be at the midpoint, because their distributions are skewed.
What is Summary Density?
The summary density is the average of the error densities at each possible value of the
estimate. Since each estimate is different, the summary density usually will not appear as a
normal density. The summary density may have many modes (bumps) or appear skewed.
The summary density is used to portray the distribution of the heterogeneous effects while
accounting for the differing error densities. Other graphics can also be used for the same
purpose. The stick diagram (also called Forest Plots) also portrays the heterogeneity, but it is not
easy to interpret. The stick diagram gives a distorted view, because the effects with the poorest
estimates, and consequently the least informative, have the longest confidence intervals; thus,
they are more eye catching than the most informative effects, which have shorter confidence
intervals. A histogram could be constructed from these estimates, but it weights each estimate
equally although some are more precise than others. Summary density curves can be viewed as
smoothed histograms. However, unlike a histogram, summary density curves account for the
varying standard errors of the individual mean effect estimates.
Many readers interpret stick diagrams by noting the fraction of significant effects. This
method has limitations, as there may be an overall significant effect detectable by meta-analysis
despite the many nonsignificant effects due to low power in the individual studies. The
summary density of these effects may show a mode at a value different from zero. This suggests
that the nonsignificant effects tend to cluster around a value other than zero. The summary
density can also show two or more modes, indicating disagreement between estimates or the
presence of a factor (or multiple factors) that varies between estimates.
AX7-122
-------
Any summary density that does not appear to look similar to a normal density may be
reflecting a distribution that is not normal. This is important because most meta-analyses
assume that the heterogeneity distribution is normal. However, it should be noted that there may
be too few estimates or too much uncertainty to determine the normality of the heterogeneity
distribution based on the shape of the summary density.
Another method for portraying a distribution of effects that have different precision is the
radial plot developed by Galbraith (1994). This is also a good method to summarize various
effect estimates, but it may require more statistical experience to understand.
Theory of Summary Density
Statistical science has studied Kernel density estimates (Silverman, 1986), and the
summary density belongs to the class called the Gaussian Variable Kernel Density. Gaussian is
another name for the normal distribution. The summary density has a variable kernel: variances
of each kernel differ because of the unequal lengths of the confidence intervals of the estimates.
The kernel is the shape of the distribution used for each estimate. For example, other than the
normal density, a triangular density could be used. The histogram is also a kernel density
estimate. The kernel in this case is rectangular (a uniform density). The rectangular shape is not
considered a good kernel, because as the number of observations increase, the histogram
converges more slowly to the true density compared to other kernels.
The summary density is presented as a graphical description of the heterogeneity. It does
not converge to the true distribution of the heterogeneity. To do so in an optimal way, the
variance of the error density would have to decrease with the number of effects being studied
and the standard deviation among them.
Rather than the summary density being only a descriptive tool, it can be used for inference.
Research has yielded a formula for fixing the variance of the Gaussian kernel so that there is less
than a 5% chance of erroneously concluding that a sample from a normal density is multimodal
(has more than one bump) (Jones, 1983). As the kernel of the summary density does not use this
formula, the significance of a multimodal summary density is unknown. Also because of the
varying kernel of the summary density, the formula does not apply and likely calculates too
small a value. However, a simulation could be carried out to compute the p value for each
application of the summary density to data.
AX7-123
-------
Extending the Summary Density to an Optimal Kernel Density Estimate
The Gaussian kernel density estimate is an average of normal density estimates and is
given by the equation
(AX7-1)
where x is the value (in this case a possible value of the effect) at which the density is to be
evaluated, ftt is the effect estimate, n is the number of samples (number of effects) and h is the
standard deviation of each Gaussian density. The value of h is a constant for each density
estimate but varies depending on n. As n increases, smaller values of h are used so that the
kernel density estimate converges to the true density.
The definition of the summary density is the same as the Gaussian kernel density except
hf = 0,- (AX7-2)
where <^ is the estimate of the standard error of the rth sample (effect estimate). In this case, the
summary density is called a variable kernel density estimate, as ht varies depending on the
standard error of each estimate.
To extend the summary density to an optimal density consider ht of the form
kA ,
A _^CT. (AX7-3)
l
where & is a constant to be determined, og is the geometric mean of the estimated standard
errors and
A = nw^cTiJQR, 1 1.34) (AX7-4)
where <^ is the standard deviation and IQRt is the interquartile range of the samples (effect
estimates). Both ^ and IQRtl 1. 34 are estimates of the true standard deviation when the
AX7-124
-------
distribution of the effects is normal. The extension of the summary density to an optimal kernel
density will be referred to as the extension.
Next, the choice of the ht for the extension will be explained. Jones (1983) has shown for
the (constant) kernel that
kA
(AX7-5)
where k = 1.25. A normally distributed sample will have a multimodal kernel estimate 5% of the
time with this choice ofk. However, for the extension, a larger k is required to achieve the same
critical value (5%) as Jones has achieved. This is due to the kernel being variable and prone to
having more than one mode. Also, A is used rather than the more common estimate of <7, since A
is used with the optimal kernel.
Another simpler, but somewhat unsatisfactory approach, is to use Jones' result directly on
the effect size (the ratio of the effect estimate to its standard error). The variability of this ratio
is approximately constant for effects estimated from long sampling periods.
Probably the best and yet somewhat subjective method is to adopt aspects of Silverman's
method (Silverman, 1986). A multiple of the standard deviation is considered. This multiple is
either increased or decreased until the density becomes visually multimodal. The effects are
simulated using A as defined by Equation AX7-4 for the standard error and the mean of the
effects for the error densities.
An Example of Calculating the Summary Density
Table AX7-8 shows the O3-associated excess risk estimates for cardiovascular mortality in
the warm season from select studies (see Figure 7-25 in Section 7.4.7 for the stick diagram of
these estimates).
The estimated log relative risk is considered normally distributed, so the percent change is
converted to log relative risk per standard unit (RR) by Equation AX7-6:
log(RR) = log[(% change/ 100) +1] (AX7-6)
AX7-125
-------
Table AX7-8. Ozone-Associated Cardiovascular Mortality Risk Estimates
per Standardized Increment"
% Excess Risk in Mortality
Reference Study Location (95% CI)
Moolgavkar (2003)" Los Angeles, CA 1.61 (-0.24,3.50)
Moolgavkar (2003)" Cook County, IL 6.82 (4.38,9.32)
Lippmann et al. (2000)c Detroit Area, MI 1.52 (-3.64,6.95)
Vedal et al. (2003) Vancouver, British Columbia, Canada 16.37 (-1.14, 36.98)
Anderson et al. (1996) London, England 4.46 (1.30,7.72)
Sunyer et al. (1996) Barcelona, Spain 6.70 (2.15, 11.50)
Simpson etal. (1997) Brisbane, Australia 7.37 (-3.41, 19.36)
a The standardized increment is 40 ppb for 1-h max O3; 30 ppb for 8-h max O3; and 20 ppb for 24 h avg O3.
b Indicates use of Poisson GAM with default convergence criteria.
c Reanalysis by Ito (2004) using Poisson GLM.
Then the above equation is applied to the confidence limits of the percent excess risks to
get confidence limits for log(RR). The difference of these limits divided by 3.92 is used as the
standard error of the log(RR).
The equation of the normal density is
(AX7-7)
For each of the seven effect estimates, the log(RR) is substituted for jut and its standard error is
substituted for at. These densities are calculated over a range of x. For Figure AX7-1, x has
been calculated every 0.0004 units of log(RR) from -0.2 to 0.2. Then the densities for each
value of x are averaged, resulting in the summary density.
Figure AX7-1 shows each of the seven error densities and the summary density curve.
The summary density is multiplied by the number of estimates (n = 7) to better illustrate the
shape of the curve. In the figure, the summary density appears bimodal, but there are too few
effects to confirm this statistically.
AX7-126
-------
Q
0 10 20
% Change in Cardiovascular Mortality
Figure AX7-1. Density curves of the O3-associated excess risk of cardiovascular mortality
in the warm season per standardized increment. The thicker curve is the
summary density curve of the seven effect estimates.
In Chapter 7, the summary density curves in Figures 7-29 and 7-30 were calculated using
equation AX7-3 with A equal to the estimated standard deviation of the effects. Simulation
indicated that the all-year curve in Figure 7-29 was not significantly multimodal.
Significance of a Summary Density
When there is no significant difference between effect estimates and the pooled analysis
finds the overall effect to be significantly positive, the error density of the overall effect would
have less than 2.5% of its area under the curve beyond the value of no effect. However, the
summary density of the individual error densities will generally show more than 2.5% beyond
the no-effect value. Results from Mortimer et al. (2002) reported in Chapter 7 found no
significant effects of O3 on PEF and incidence of symptoms in the individual cities, but indicated
a significant overall effect when the data from all cities were combined (see Figures 7-4 and 7-7
in Sections 7.2.3.2 and 7.2.4, respectively). The summary density of the error densities from the
eight cities had about 20 to 25% of the area beyond zero. Simulation under these conditions
verified that these larger percentages of the area beyond the value of no effect are common. One
of the reasons for this result is that the summary density does not treat each estimate as derived
from random estimates of the same single true estimate.
AX7-127
-------
Rather than treating the area beyond the value of no effect in a summary density as an
indication of significance, it should be treated as an indication of disagreement. For example,
there might be a mode at zero and a mode at a positive value. This could be due to some cities
being always below the threshold and other cities being above the threshold. In this case, there
could be appreciable area below zero while some of the cities indicate a positive effect. Even if
this simple case were the true state, the summary density might appear normal due to wide error
densities.
"Apples and Oranges" Issue
The argument that a summary analysis is mixing apples and oranges is based on a variety
of reasons. One reason is the uncertainty arising when the analysis includes effects that vary
with factors other than the factor of interest. Another reason is the combining of effects that lack
commonality. That is, the effects may vary with a number of factors, but an average effect is
estimated through some process. The summary density does not average effects; instead it
considers common location through a clustering of effects. This allows a weaker assumption
concerning the commonality of the effects.
Consider a summary density based on estimates of high precision that cluster together and
one estimate of low precision that is significantly different from the rest (the confidence intervals
do not overlap) and based on an older measurement method. The summary density could
average out the outlying value while forming a high mode based on the other effects. Such a
graph would lead one to ignore the results of the old measurement method. Disregarding
previous results when more precise measurements become available is often practical under such
circumstances.
Masking of Heterogeneity
When the variances are very large, the kernel density appears to be very similar to a normal
density. Thus, large error densities can mask the true pattern of heterogeneity. There may be
good reasons to suppose that the heterogeneity is other than normal, and the failure of the
summary density to show this pattern is due to wide error densities. When such masking occurs,
the summary density cannot reject the assumption of normality of heterogeneous effects in a
meta-analysis.
AX7-128
-------
Another reason that masking may occur with the summary density is the use of the
standard error of the effect for the standard deviation of the effects error density. Kernel density
theory permits decreasing the standard deviation when more effects are available. Narrower
error densities should clarify the heterogeneity distribution. However, the ideal reduction in
variance due to increasing observation size is not large for the numbers of effects usually
considered. For example, ideally h decreases as n115 increases, which is rather slowly.
Conclusion
The summary density is not new. As it stands, it is a kernel density estimate without a
fixed value of h. Others have fixed h either using graphics (Kochi et al., 2005) or ad hoc values
(Burr and Doss, 2005). Flachaire and Nunez (2002) used a weighted average of error densities,
with the weights based on the population size. This and other types of weighting should be
considered in the future. Also, efforts should be made to unmask the heterogeneity distribution
in a statistically justified manner.
The summary density is a simple graphical method for portraying the distribution of
heterogeneous effects in the presence of effects estimated with different precision. It has
graphical advantages over both the stick diagram and the histogram. The summary density can
be put on a more firm statistical footing. Inference concerning the presence of modes could be
made reliably if p values were generated from simulations. The summary density is a graphical-
diagnostic tool for the normality assumption in meta-analyses. A meta-analysis method has
been developed for use when the distribution of effects is not normal and the precision varies
(Burr and Doss, 2005). There is a need to develop other improvements to unmask the
heterogeneity distribution in a statistically justified manner.
The summary density overcomes some issues with reliance on statistical tests. If effects
were nonsignificant, one would expect them to cluster on either side of the no-effect value. If
the summary density indicates a mode at a positive effect value, a tentative conclusion is that
there is a positive nonrandom effect. Confirmation that the mode is statistically significant may
need further study, mindful that effects can include spurious components.
AX7-129
-------
AX7.3 ESTIMATING THE GOMPERTZ CONCENTRATION-
RESPONSE MODEL
Estimating the Gompertz Concentration-Response Model for Mortality
Risk and Long-Term O3 Concentrations
A three-parameter Gompertz concentration-response model was fit to examine the
association between mortality risk from acute O3 exposure and long-term O3 concentrations
using data from the U.S. 95 communities study (Bell et al., 2004). Most single-city studies fit a
linear relationship, not a Gompertz model, to mortality times-series data. This may be due to the
limited range of the O3 concentrations within a city, so that all the features (i.e., linear portion
and two flat regions) of the Gompertz model cannot be estimated. The mortality risk estimates
from the 95 communities are also linear effects, and they are assumed to be tangential to the
Gompertz curve for the long-term average 24-h avg O3. To estimate the Gompertz model
parameters, the effect estimates were fit to a derivative of the Gompertz model. Figure AX7-2
shows the fitted derivative curve using only communities with warm-season data (n = 40). The
derivative model does not have the flexibility to estimate negative P (log relative risk per
1,000 ppb) estimates and can only estimate them with a value near zero. The area under the
derivative curve is the Gompertz curve, which is shown in Figure 7-33 in Chapter 7.
The equation of the Gompertz model is as follows:
% excess risk= 100 x (eP - 1)
where (3 = bi x exp[-exp(-fo (O3 - ECo.os% - fog(fog(6/ /0,0005))))]
The estimated parameters of the Gompertz model are listed in Table AX7-9.
The result is a model with essentially a threshold at low concentrations and a flat curve at
higher concentrations. The curve has shallower slopes above the median effective concentration
(inflection point) than below that concentration point. The horizontal line is the 95% confidence
interval on the long-term average 24-h avg O3 associated with a 0.05% excess risk in mortality
(EC005o/0). The EC005o/0 is a risk that is believed to be low enough to be in the noise band. The
upper 95% confidence limit on the EC0050/0 was 27.6 ppb, which may be considered the lowest
concentration for which the data support a no-threshold association. The Gompertz model
presents a threshold of effects at low concentrations and a flat curve at higher concentrations.
AX7-130
-------
-5 -
10 20 30
Long-Term Average 24-h avg O3 (ppb)
40
Figure AX7-2. Fitted derivative curve of the Gompertz model to the regression
coefficients (P = log relative risk per 1,000 ppb) for the association
between mortality and acute exposure to O3 plotted against community-
specific long-term average 24-h avg O3 concentrations using data from the
U.S. 95 communities study (Bell et al., 2004). Only the 40 communities
with warm-season data are included.
Table AX7-9. Estimated Parameters of the Gompertz Model
Parameter
100 x b,
•k^'O.05%
ft,
Estimate
0.63
25.8
0.79
SE
0.21
0.9
0.36
(95% CI)
(0.20, 1.06)
(24.0, 27.6)
(0.06, 1.53)
The actual curve would not be flat indefinitely as the concentration increases, but there is no data
to model what is probably a very complex portion of curve. To the extent that the estimated
concentration associated with a 0.05% excess risk is realistic, the fit of the model supports the
observed relationship between greater mortality risk at lower long-term O3 concentrations.
AX7-131
-------
The best method of estimating the ECao5% would be to estimate the Gompertz model
directly from the pooled data from all the communities. Some of the disadvantages of estimating
the EC005o/0 by using the community-specific regression coefficients are (1) other possibly
important factors have not been used to adjust the estimate; (2) regression coefficients may not
be tangents to the curve at the long-term average O3 level, but rather have some other
relationship; and (3) the data points represent a sampling of communities rather than individuals.
To compare O3 levels from a 24-h avg to a 1-h max, one could multiply the x-axis by a
factor of two as is done to obtain comparable incremental changes between the metrics (a 20 ppb
increase in 24-h avg O3 corresponds to a 40 ppb increase in 1-h max O3). Many researchers
believe that short spike exposures are more toxic than longer averages. However, the 24-h avg
O3 may be a more stable exposure estimate and still correlate with the shorter metrics of
exposure.
AX7-132
-------
wEPA
Please make all necessary changes in the below label, PRESORTED STANDARD
detach copy or copy, and return to the address in the upper POSTAGE & FEES PAID
left-hand corner. EPA
United States •-, PERMIT No. G-35
Environmental Protection 'f y°uhdo not wish to ^ve these reports CHECK HERE |J:
detach copy or copy, and return to the address in the
Agency upper left-hand corner.
National Center for
Environmental Assessment
Research Triangle Park, NC 27711
Official Business
Penalty for Private Use
$300
EPA 600/R-05/004bF
February 2006
------- |