^^ ^^ P^ A
» EPA
United States
Air Quality Criteria for
Ozone and Related
Photochemical Oxidants
(First External Review Draft)
Volume I of
-------
EPA/600/R-05/004aA
January 2005
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
Volume I
National Center for Environmental Assessment-RTF Office
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC
-------
DISCLAIMER
This document is an external review draft for review purposes only and does not
constitute U.S. Environmental Protection Agency policy. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
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 primarily on the scientific air quality criteria for
O3 and, to a lesser extent, on those for other photochemical oxidants such as hydrogen peroxide
and the peroxyacyl nitrates, as have subsequent revised versions of the ozone document.
I-ii
-------
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 then released in August 1986; and a supplement, Summary of
Selected 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 some of the newer scientific data that became available after completion of the
1986 criteria document. Such literature was assessed in the next periodic revision of the O3 air
quality criteria document, which was completed in 1996 and provided scientific bases supporting
the setting by EPA in 1997 of an 8-h O3 NAAQS that is currently in force together with the 1-h
O3 standard.
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 Ozone Air Quality Criteria Document (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, some other scientific data are also
presented and evaluated 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 assesses pertinent literature available
through 2004.
The present draft document (dated January 2005) is being released for public comment and
review by the Clean Air Scientific Advisory Committee (CASAC) to obtain comments on the
organization and structure of the document, the issues addressed, the approaches employed in
assessing and interpreting the newly available information on O3 exposures and effects, and the
key findings and conclusions arrived at as a consequence of this assessment. Public comments
and recommendations will be taken into account making any appropriate further revisions to this
document for incorporation into a Second External Review Draft. That draft will be released for
further public comment and CASAC review before last revisions are made in response and
incorporated into a final version to be completed by early 2006. Evaluations contained in the
present document will be drawn on to provide inputs to associated PM Staff Paper analyses
I-iii
-------
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. The document also evaluates ambient O3 environmental
effects on vegetation and ecosystems, man-made materials, and surface level solar UV radiation
flux and global climate change, 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 draft document.
I-iv
-------
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
(First External Review Draft)
VOLUME I
Executive Summary E-l
1. INTRODUCTION 1-1
2. PHYSICS AND CHEMISTRY OF OZONE IN THE ATMOSPHERE 2-1
CHAPTER 2 ANNEX (ATMOSPHERIC PHYSICS/CHEMISTRY) AX2-1
3. ENVIRONMENTAL CONCENTRATIONS, PATTERNS, AND
EXPOSURE ESTIMATES 3-1
CHAPTER 3 ANNEX (AIR QUALITY AND EXPOSURE) AX3-1
I-v
-------
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
(First External Review Draft)
(cont'd)
VOLUME II
4. DOSIMETRY, SPECIES HOMOLOGY, SENSITIVITY, AND
ANIMAL-TO-HUMAN EXTRAPOLATION 4-1
CHAPTER 4 ANNEX (DOSIMETRY) AX4-1
5. TOXICOLOGICAL EFFECTS OF OZONE AND RELATED
PHOTOCHEMICAL OXIDANTS IN LABORATORY ANIMALS
AND IN VITRO TEST SYSTEMS 5-1
CHAPTER 5 ANNEX (ANIMAL TOXICOLOGY) AX5-1
6. CONTROLLED HUMAN EXPOSURE STUDIES OF OZONE AND
RELATED PHOTOCHEMICAL OXIDANTS 6-1
CHAPTER 6 ANNEX (CONTROLLED HUMAN EXPOSURE) AX6-1
7. EPIDEMIOLOGICAL STUDIES OF HUMAN HEALTH EFFECTS
ASSOCIATED WITH AMBIENT OZONE EXPOSURE 7-1
CHAPTER 7 ANNEX (EPIDEMIOLOGY) AX7-1
8. INTEGRATIVE SYNTHESIS 8-1
I-vi
-------
Air Quality Criteria for Ozone and Related
Photochemical Oxidants
(First External Review Draft)
(cont'd)
VOLUME
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. EFFECT OF OZONE ON MAN-MADE MATERIALS 11-1
I-vii
-------
Table of Contents
Page
List of Tables I-xiii
List of Figures I-xvii
Authors, Contributors, and Reviewers I-xxviii
U.S. Environmental Protection Agency Project Team for Development of Air Quality
Criteria for Ozone and Related Photochemical Oxidants I-xxix
U.S. Environmental Protection Agency Science Advisory Board (SAB) Staff Office
Clean Air Scientific Advisory Committee (CASAC) Ozone Review Panel I-xxx
Abbreviations and Acronyms I-xxxi
Executive Summary E-l
1. INTRODUCTION 1-1
1.1 LEGAL AND HISTORICAL BACKGROUND 1-1
1.1.1 Legislative Requirements 1-1
1.1.2 Criteria and NAAQS Review Process 1-3
1.1.3 Regulatory Chronology 1-4
1.2 CURRENT OZONE CRITERIA AND NAAQS REVIEW 1-8
1.2.1 Key Milestones and Procedures for Document Preparation 1-8
1.3 ORGANIZATIONAL STRUCTURE OF THE DOCUMENT 1-11
1.3.1 General Document Format 1-11
1.3.2 Organization and Content of the Document 1-12
REFERENCES 1-14
2. PHYSICS AND CHEMISTRY OF OZONE IN THE ATMOSPHERE 2-1
2.1 INTRODUCTION 2-1
2.2 CHEMICAL PROCESSES INVOLVED IN OZONE FORMATION
AND DESTRUCTION 2-1
2.3 METEOROLOGICAL PROCESSES AFFECTING OZONE 2-6
2.4 RELATIONS OF OZONE TO ITS PRECURSORS 2-12
2.5 THE ROLE OF CHEMISTRY-TRANSPORT MODELS IN
UNDERSTANDING ATMOSPHERIC OZONE 2-17
2.6 TECHNIQUES FOR MEASURING OZONE AND ITS PRECURSORS 2-21
2.7 SUMMARY 2-22
REFERENCES 2-25
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
I-viii
-------
Table of Contents
(cont'd)
Pas
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
AX2.3 PHYSICAL PROCESSES INFLUENCING THE ABUNDANCE
OF OZONE AX2-58
AX2.3.1 Stratospheric-Tropospheric Ozone Exchange (STE) AX2-59
AX2.3.2 Deep Convection in the Troposphere AX2-68
AX2.3.2.1 Observations of the Effects of Convective
Transport AX2-70
AX2.3.2.2 Modeling the Effects of Convection AX2-73
AX2.3.3 Nocturnal Low-Level Jets AX2-76
AX2.3.4 Intercontinental Transport of Ozone and Other Pollutants AX2-80
AX2.3.4.1 The Atmosphere/Ocean Chemistry Experiment,
AEROCE AX2-80
AX2.3.4.2 The North Atlantic Regional Experiment,
NARE AX2-83
AX2.3.5 The Relation of Ozone to Solar Ultraviolet Radiation,
Aerosols, and Air Temperature AX2-86
AX2.3.5.1 Solar Ultraviolet Radiation and Ozone AX2-86
AX2.3.5.2 Impact of Aerosols on Radiation and Photolysis
Rates and Atmospheric Stability AX2-87
AX2.3.5.3 Temperature and Ozone AX2-88
AX2.4 THE RELATION OF OZONE TO ITS PRECURSORS AND
OTHER OXIDANTS AX2-92
AX2.4.1 Summary of Results for the Relations Among Ozone, its
Precursors and Other Oxidants from Recent Field Experiments .... AX2-95
AX2.4.1.1 Results from the Southern Oxidant Study
and Related Experiments AX2-95
AX2.4.1.2 Results from Studies on Biogenic and
Anthropogenic Hydrocarbons and Ozone
Production AX2-99
I-ix
-------
Table of Contents
(cont'd)
Page
AX2.4.1.3 Results of Studies on Ozone Production in
Mississippi and Alabama AX2-100
AX2.4.1.4 The Nocturnal Urban Plume Over Portland,
Oregon AX2-101
AX2.4.1.5 Effects of VOC's in Houston on Ozone
Production AX2-101
AX2.4.1.6 Chemical and Meteorological Influences on
the Phoenix Urban Ozone Plume AX2-101
AX2.4.1.7 Transport of Ozone and Precursors on the
Regional Scale AX2-102
AX2.4.1.8 Model Calculations and Aircraft Observations
of Ozone Over Philadelphia AX2-103
AX2.4.1.9 The Two-Reservoir System AX2-103
AX2.5 METHODS USED TO CALCULATE RELATIONS BETWEEN
OZONE AND ITS PRECURSORS AX2-105
AX2.5.1 Chemistry-Transport Models AX2-106
AX2.5.2 Emissions of Ozone Precursors AX2-120
AX2.5.3 Observationally Based Models AX2-126
AX2.5.4 Chemistry-Transport Model Evaluation AX2-128
AXA.5.4.1 Evaluation of Emissions Inventories AX2-140
AX2.5.4.2 Availability and Accuracy of Ambient
Measurements AX2-142
AX2.6 TECHNIQUES FOR MEASURING OZONE AND ITS PRECURSORS AX2-143
AX2.6.1 Sampling and Analysis of Ozone AX2-143
AX2.6.2 Sampling and Analysis of Nitrogen Oxides AX2-145
AX2.6.2.1 Calibration Standards AX2-146
AX2.6.2.2 Measurement of Nitric Oxide AX2-146
AX2.6.2.3 Measurements of Nitrogen Dioxide AX2-148
AX2.6.2.4 Monitoring for NO2 Compliance Versus
Monitoring for Ozone Formation AX2-149
AX2.6.3 Measurements of Nitric Acid Vapor, HNO3 AX2-149
AX2.6.4 Sampling and Analysis of Volatile Organic Compounds AX2-151
AX2.6.4.1 Polar Volatile Organic Compounds AX2-152
REFERENCES AX2-154
3. ENVIRONMENTAL CONCENTRATIONS PATTERNS, AND EXPOSURE ESTIMATES .3-1
3.1 INTRODUCTION 3-1
3.2 AMBIENT AIR QUALITY DATA FOR OZONE 3-5
3.3 DIURNAL VARIABILITY OF OZONE 3-28
3.4 SPATIAL VARIABILITY OF OZONE 3-31
3.5 TRENDS IN OZONE CONCENTRATIONS 3-33
3.6 RELATIONSHIPS BETWEEN OZONE AND OTHER SPECIES 3-36
3.7 POLICY RELEVANT BACKGROUND OZONE CONCENTRATIONS 3-43
3.8 INDOOR SOURCES AND EMISSIONS OF OZONE 3-51
I-x
-------
Table of Contents
(cont'd)
Page
3.9 HUMAN EXPOSURE TO OZONE 3-58
3.10 SUMMARY OF KEY POINTS 3-66
REFERENCES 3-69
AX3. ENVIRONMENTAL CONCENTRATIONS, PATTERNS, AND EXPOSURE
ESTIMATES AX3-1
AX3.1 INTRODUCTION AX3-1
AX3.2 SURFACE OZONE CONCENTRATIONS AX3-8
AX3.2.1 Urban Area Concentrations AX3-9
AX3.2.2 Nonurban Area Concentrations AX3-35
AX3.2.3 Ozone Concentrations Observed at Relatively Remote
Monitoring Sites AX3-39
AX3.3 DIURNAL PATTERNS IN OZONE CONCENTRATION AX3-68
AX3.3.1 Introduction AX3-68
AX3.3.2 Diurnal Patterns in Urban Areas AX3-71
AX3.3.3 Diurnal Patterns in Nonurban Areas AX3-77
AX3.4 SEASONAL PATTERNS IN OZONE CONCENTRATIONS AX3-80
AX3.4.1 Urban Area Seasonal Patterns AX3-80
AX3.4.2 Seasonal Patterns in Nonurban Areas AX3-83
AX3.5 SPATIAL VARIABILITY IN OZONE CONCENTRATIONS AX3-85
AX3.5.1 Spatial Variability of Ozone Concentrations in Urban Areas .. AX3-85
AX3.5.2 Urban-Nonurban Concentration Differences AX3-100
AX3.5.3 Ozone Concentrations at High Elevations AX3-100
AX3.6 TRENDS IN OZONE CONCENTRATIONS AX3-110
AX3.6.1 National Assessment of Trends AX3-110
AX3.6.2 Regional Trends AX3-110
AX3.6.3 Disproportionate Reductions in Hourly Average
Concentrations AX3-123
AX3.6.4 Trends in National Parks in the United States AX3-125
AX3.7 RELATIONS BETWEEN OZONE, OTHER OXIDANTS, AND
OXIDATION PRODUCTS AX3-130
AX3.8 RELATIONSHIP BETWEEN SURFACE OZONE AND OTHER
POLLUTANTS AX3-134
AX3.8.1 Introduction AX3-134
AX3.8.2 Co-Occurrence of Ozone with Nitrogen Oxides AX3-137
AX3.8.3 Co-Occurrence of Ozone with Sulfur Dioxide AX3-138
AX3.8.4 Co-Occurrence of Ozone and Daily PM25 AX3-141
AX3.8.5 Co-Occurrence of Ozone with Acid Precipitation AX3-141
AX3.8.6 Co-Occurrence of Ozone with Acid Cloudwater AX3-144
AX3.9 THE METHODOLOGY FOR DETERMINING POLICY RELEVANT
BACKGROUND OZONE CONCENTRATIONS AX3-144
AX3.9.1 Introduction AX3-144
AX3.9.2 CAPABILITY OF GLOBAL MODELS TO SIMULATE
TROPOSPHERIC OZONE AX3-160
I-xi
-------
Table of Contents
(cont'd)
Pas
AX3.9.3 Mean Background Concentrations: Spatial and Seasonal
Variation AX3-164
AX3.9.4 Frequency of High-Ozone Occurrences at Remote Sites AX3-166
AX3.10 INDOOR SOURCES AND CONCENTRATIONS AX3-174
AX3.10.1 Sources and Emissions of Indoor Ozone AX3-174
AX3.10.2 Factors Affecting Ozone Concentrations Indoors AX3-175
AX3.10.3 Ozone Concentrations in Microenvironments AX3-181
AX3.11 HUMAN EXPOSURE TO OZONE AND RELATED
PHOTOCHEMICAL OXIDANTS AX3-195
AX3.11.1 Introduction AX3-195
AX3.11.2 Summary of the Information Presented in the Exposure
Discussion in the 1996 Ozone Criteria Document AX3-196
AX3.11.3 Concepts of Human Exposure AX3-197
AX3.11.4 Quantification of Exposure AX3-197
AX3.11.5 Methods To Estimate Personal Exposure AX3-198
AX3.11.6 Exposures in Microenvironments AX3-201
AX3.11.7 Differences in Concentrations Among Microenvironments ... AX3-202
AX3.11.8 Activity Patterns/Time Spent in Microenvironments AX3-203
AX3.11.9 Personal Exposure Monitors and Measurements AX3-205
AX3.11.10 Trends in Concentrations Within Microenvironments AX3-208
AX3.11.11 Modeling Ozone Exposure AX3-208
AX3.11.11.1 Issues of Terminology AX3-208
AX3.11.11.2 A General Framework for Assessing
Exposure to Ozone AX3-210
AX3.11.11.3 Characterization of Ambient Concentrations
of Ozone AX3-213
AX3.11.11.4 Calculation of Microenvironmental
Concentrations AX3-214
AX3.11.12 Population Exposure Models: Considerations of Activity
Events and Population Demographics AX3-217
AX3.11.13 Recent Developments: Activity Events and
Inhalation Intake AX3-221
AX3.11.14 Characterization of Exposure AX3-223
AX3.11.14.1 Use of Ambient Ozone Concentrations AX3-223
AX3.11.14.2 Exposure Selection in Controlled
Exposure Studies AX3-225
AX3.11.14.3 Exposure to Related Photochemical Agents . . AX3-225
AX3.11.14.4 Exposure to Sensitive Populations AX3-226
REFERENCES AX3-230
I-xii
-------
List of Tables
Number Page
1-1 National Ambient Air Quality Standards (NAAQS) for Ozone 1-5
1-2 Key Milestones for Development of Revised Ozone Air Quality
Criteria Document 1-10
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
(adapted from Atkinson and Arey, 2003) AX2-25
AX2-3 Hydroxyl Rate Constants and Atmospheric Lifetimes of Mono- and
Di-cyclic Aromatic Hydrocarbons (adapted from Atkinson 2000) AX2-34
AX2-4 Chemistry-Transport Models (CTM) Contributing to the Oxcomp Evaluation
of Predicting Tropospheric O3 and OH (Prather and Ehhalt, 2001) AX2-119
AX2-5 Emissions of Nitrogen Oxides by Various Sources in the United States
in 1999 AX2-120
AX2-6 Emissions of Volatile Organic Compounds by Various Sources in the
United States in 1999 AX2-121
AX2-7 Emissions of Ammonia by Various Sources in the United States in 1999 AX2-122
AX2-8 Emissions of Carbon Monoxide by Various Sources in the United States
in 1999 AX2-123
3-1 Ozone Monitoring Seasons by State 3-4
3-2 The Second Highest Daily Maximum One-Hour Ozone Concentration (ppm)
by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan Statistical
Area (CMSA) for the Years 1999 to 2001 3-7
3-3 The Fourth Highest Daily Maximum Eight-Hour Ozone Concentration (ppm)
by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan Statistical
Area (CMSA) for the Years 1999 to 2001 3-15
3-4 Seasonal (April-October) Percentile Distribution of Hourly Ozone Concentrations,
Number of Hourly Mean Ozone Occurrences > 0.08 and > 0.10, SUM06, and
W126 Values for Selected Rural Ozone Monitoring Sites in 2001 3-22
I-xiii
-------
List of Tables
(cont'd)
Number Page
3-5 Seasonal (April - 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 > 75% 3-23
3-6 Summary Statistics for O3 (in ppm) Spatial Variability in Selected Urban Areas
in the United States 3-32
3-7 Trends at National Parks in the United States (1981 - 2001 or available
data period) 3-37
3-8 Previous Estimates of Background O3 in Surface Air Over the United States 3-47
3-9 Air Exchange Rates in Residences by Season and Region of the Country 3-53
3-10 Indoor/Outdoor Ozone Concentrations in Various Microenvironments 3-54
AX3-1 Ozone Monitoring Seasons by State AX3-7
AX3-2 Summary of Percentiles of Hourly Average Concentrations (ppm) for the
April to October Period AX3-10
AX3-3 The Second Highest Daily Maximum One-Hour Ozone Concentration (ppm)
by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan
Statistical Area (CMSA) for the Years 1999 to 2001 AX3-13
AX3-4 Percentiles of Hourly Average Concentrations (ppm) and Fourth Highest
Eight-Hour Daily Maximum Concentration for the April to October Period AX3-21
AX3-5 The Fourth Highest Daily Maximum Eight-Hour Ozone Concentration (ppm)
by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan
Statistical Area (CMSA) for the Years 1999 to 2001 AX3-23
AX3-6 Summary of Percentiles of Pooled Data Across Monitoring Sites for
1996 - 2000 AX3-33
AX3-7 Seasonal (April through October) Percentile Distribution of Hourly Ozone
Concentrations, Number of Hourly Mean Ozone Occurrences > 0.08 and
> 0.10, SUM06, and W126 Values for Selected Rural Ozone Monitoring
Sites in 2001 AX3-37
I-xiv
-------
List of Tables
(cont'd)
Number Page
AX3-8 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% AX3-51
AX3-9 The Top 10 Daily Maximum 8-h Average Concentrations (ppm) for Sites
Experiencing Low Maximum Hourly Average Concentrations with Data
Capture of > 75% AX3-56
AX3-10 Trends (Using Kendall's tau) for the 3Oth Percentile at National Park
Service Monitoring Sites AX3-63
AX3-11 Long-Term Trend of Background Ozone AX3-65
AX3-12 Summary Statistics for Ozone (in ppm) Spatial Variability in Selected U.S.
Urban Areas AX3-87
AX3-13 Description of Mountain Cloud Chemistry Program Sites AX3-102
AX3-14 Seasonal (April-October) Percentiles, SUM06, SUM08, and W126 Values for
the MCCP Sites AX3-103
AX3-15 Summary Statistics for 11 Integrated Forest Study Sites AX3-108
AX3-16 Trends at National Parks in the United States (1981 to 2001 or available
data period) AX3-127
AX3-17 Range of Annual (January-December) Hourly Ozone Concentrations (ppb)
at Background Sites Around the World (CMDL, 2004) AX3-150
AX3-18 Range of annual (January-December) Hourly Median and Maximum Ozone
Concentrations (ppb) at Background Stations in Protected Areas of the United
States (CASTNet, 2004) AX3-150
AX3-19 Range of annual (January-December) Hourly Median and Maximum Ozone
Concentrations (ppb) at Canadian Background Stations (CAPMoN, 2003) AX3-150
AX3-20 Number of Hours > 0.05 ppm for Selected Rural O3 Monitoring in the United
States by Month for the Period 1988 to 2001 AX3-152
AX3-21 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 AX3-154
I-xv
-------
List of Tables
(cont'd)
Number
AX3-22
AX3-23
AX3-24
AX3-25
AX3-26
AX3-27
AX3-28
AX3-29
AX3-30
AX3-31
AX3-32
Global Budgets of Tropospheric Ozone (Tg year"1) for the Present-day
Atmosphere
Description of Simulations Used for Source Attribution (Fiore et al, 2003 a) . . .
Number of Hours with Ozone Above 50 or 60 ppbv at U.S. CASTNet Sites
in 2001
Air Exchange Rates in Residences by Season and Region of the Country
Rate Constants (Ir1) for the Removal of Ozone by Surfaces in Different
Indoor Environments
Indoor/Outdoor Ozone Ratios
Indoor and Outdoor O3 Concentrations in Boston MA
Indoor and Outdoor O3 Concentrations in Hong Kong
Indoor and Outdoor Ozone Concentrations
Attributes of pNEM, SHEDS, APEX, and MENTOR/SHEDS Population
Exposure Models
Personal Exposure Concentrations
Page
AX3-159
. . AX3-163
. . AX3-168
. . AX3-178
. . AX3-180
. . AX3-182
AX3-187
AX3-189
. . AX3-192
. . AX3-219
. . AX3-227
I-xvi
-------
List of Figures
Number Page
2-1 Schematic overview of O3 photochemistry in the stratosphere and troposphere 2-4
2-2a Surface weather chart showing sea level (MSL) pressure (kPa), and surface fronts . . . 2-7
2-2b Vertical cross section along dashed line (a-a') from northwest to the southeast
(CYYC = Calgary, Alberta; LBF = North Platte, NB; LCH = Lake Charles, LA) 2-7
2-3 The diurnal evolution of the planetary boundary layer while high pressure
prevails over land 2-10
2-4 Locations of low level j et occurrences in decreasing order of prevalence
(most frequent, common, observed) 2-10
2-5 Conceptual two-reservoir model showing conditions in the PEL and in the lower
free troposphere during a multiday O3 episode 2-11
2-6 A scatter plot of daily 1 -h maximum O3 concentrations in New York, NY versus
daily maximum temperature 2-13
2-7 A scatter plot of daily 1-h maximum O3 concentrations in Phoenix, AZ versus
daily maximum temperature 2-13
2-8 Measured values of O3 and NOZ (NOy - NOX) during the afternoon at rural sites
in the eastern United States (grey circles) and in urban areas and urban plumes
associated with Nashville, TN (gray dashes); Paris, France (black diamonds);
and Los, Angeles CA (Xs) 2-16
2-9 Main components of a comprehensive atmospheric chemistry modeling
system, such as Models-3 2-18
AX2-1 Schematic overview of ozone 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 AX2-29
AX2-5 Products from the reaction of terpenes with ozone AX2-30
AX2-6 Initial steps in the photooxidation mechanism of toluene initiated by its
reactionwith OH radicals AX2-39
I-xvii
-------
List of Figures
(cont'd)
Number Page
AX2-7a Cross section through a tropopause folding event on March 13, 1978 at
0000 GMT AX2-61
AX2-7b Ozone mixing ratios pphm (parts per hundred million) corresponding to
Figure AX2-7A AX2-64
AX2-7c Condensation nuclei concentrations (particles cm"3) corresponding to
Figure AX2-7a AX2-65
AX2-8 Schematic diagram of a meteorological mechanism involved in high
concentrations of ozone found in spring in the lower troposphere off the
American east coast AX2-68
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 AX2-72
AX2-10 The diurnal evolution of the planetary boundary layer while high pressure
prevails over land AX2-77
AX2-11 Locations of low level j et occurrences in decreasing order of prevalence
(most frequent, common, observed) AX2-78
AX2-12 Schematic diagram showing the diurnal behavior of ozone and the
development of secondary ozone maxima resulting from downward
transport from the residual layer when a low level jet is present AX2-79
AX2-13 The nocturnal low level jet occupies a thin slice of the atmosphere near the
Earth's surface AX2-79
AX2-14 A scatter plot of daily maximum ozone concentration in (a) Atlanta, GA and
(b) New York NY, versus daily maximum temperature AX2-89
AX2-15 A scatter plot of daily maximum ozone concentration in (a) Detroit, MI and
(b) Phoenix, AZ versus daily maximum temperature AX2-90
AX2-16 A scatter plot of daily maximum ozone concentration versus daily maximum
temperature for four nonurban sites AX2-91
AX2-17 Measured values of O3 and NOZ (NOy - NOX) during the afternoon at rural sites
in the eastern United States (grey circles) and in urban areas and urban plumes
associated with Nashville, TN (gray dashes), Paris, FR (black diamonds) and
Los Angeles, CA (Xs) AX2-95
I-xviii
-------
List of Figures
(cont'd)
Number Page
AX2-18 Conceptual two-reservoir model showing conditions in the PEL and in the
lower free troposphere during a multi-day ozone episode AX2-104
AX2-19 Seasonal variability in ozone concentrations observed at a number of pressure
surfaces at six ozonesonde sites and the predictions of 13 global scale
chemistry-transport models AX2-117
AX2-20 Seasonal variability in ozone concentrations observed at a number of pressure
surfaces at six ozonesonde sites and the predictions of 13 global scale
chemistry-transport models AX2-118
AX2-21a,b Impact of model uncertainty on control strategy predictions for ozone for two
days (August 10[a] and 11 [b], 1992) in Atlanta, GA AX2-130
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 AX2-131
AX2-23a Time series for measured gas-phase species in comparison with results from
a photochemical model AX2-132
AX2-23b Time series for measured gas-phase species in comparison with results from
a photochemical model AX2-133
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 AX2-136
AX2-25a,b Evaluation of model versus measured O3 versus NOy for two model scenarios
for Atlanta AX2-137
AX2-26a,b Evaluation of model versus: (a) measured O3 versus NOZ and (b) O3 versus
the sum 2H2O2 + NOZ for Nashville, TN AX2-138
AX2-27 Time series of concentrations of RO2, HO2, and OH radicals, local ozone
photochemical production rate and concentrations of NOX from measurements
made during BERLIOZ AX2-139
3-1 Second highest daily maximum 1-h O3 concentrations 3-5
3-2 Fourth highest 8-h daily maximum O3 concentration 3-14
3-3a-h Seasonal variations in O3 concentrations as indicated by the 1-h maximum in
each month at selected sites in 2002 3-30
I-xix
-------
List of Figures
(cont'd)
Number Page
3-4 Trends for 1983 - 2002 period for the annual second maximum 1-h
average concentration 3-34
3-5 Trends for 1983 - 2002 period for the annual fourth maximum 8-h
average concentration 3-34
3-6 Conditional mean PM25 concentrations versus O3 concentrations
observed at Fort Meade, MD from July 1999 to July 2001 3-40
3-7 The co-occurrence pattern for O3 and nitrogen dioxide using 2001 data from
the AQS 3-42
3-8 The co-occurrence pattern for O3 and sulfur dioxide using 2001 data from AQS .... 3-42
3-9 The co-occurrence pattern for O3 and PM25 using 2001 data from AQS 3-43
3-1 Oa Monthly maximum hourly average O3 concentrations at Yellowstone National
Park, Wyoming, in 1998, 1999, 2000, and 2001 3-44
3-1 Ob Hourly average O3 concentrations at Yellowstone National Park, Wyoming for
the period January to December 2001 3-45
3-11 Estimates of background contribution to surface afternoon (13 to 17 LT) O3
concentrations in the United States as a function of local O3 concentration,
site altitude, and season 3-48
3-12 Time-series of hourly average ozone concentrations observed at five national
parks: Denali, AK; Voyageur, MN; Olympic, WA; Glacier, MT; and
Yellowstone, WY 3-50
3-13 Ozone exposure time profile 3-58
3-14 Conceptual overview of pNEM 3-63
AX3-1 Second highest daily maximum 1-h O3 concentrations in 2001 AX3-20
AX3-2 Fourth highest 8-h daily maximum O3 concentration in 2001 AX3-30
AX3-3 The relationship between the second highest 1-h average daily maximum
concentration and the fourth highest 8-h average daily maximum concentration
at urban sites for 2001 for the O, season AX3-31
I-xx
-------
List of Figures
(cont'd)
Number Page
AX3-4 The relationship between the second highest 1-h average daily maximum
concentration and the fourth highest 8-h average daily maximum concentration
at rural sites for 2001 for the O3 season AX3-31
AX3-5 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 AX3-40
AX3-6 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 AX3-41
AX3-7 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 AX3-42
AX3-8 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 AX3-43
AX3-9 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 AX3-44
AX3-10 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 AX3-45
AX3-11 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative W126 exposure index for 2001 for the eastern United States AX3-46
AX3-12 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative SUM06 exposure index for 2001 for the eastern United States AX3-46
AX3-13 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the eastern United States AX3-47
AX3-14 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative W126 exposure index for 2001 for the central United States AX3-47
AX3-15 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative SUM06 exposure index for 2001 for the central United States AX3-48
I-xxi
-------
List of Figures
(cont'd)
Number Page
AX3-16 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the central United States AX3-48
AX3-17 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative W126 exposure index for 2001 for the western United States AX3-49
AX3-18 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative SUM06 exposure index for 2001 for the western United States AX3-49
AX3-19 The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the western United States AX3-50
AX3-20 Three-year average of the fourth highest 8-h daily maximum concentration at
Olympic (WA), Glacier (MT), Yellowstone (WY), and Denali (AK) National
Parks for 1999 to 2001 AX3-60
AX3-21 Seasonal SUM06 and W126 exposure indices for the Ouachita National Forest
for the period of 1991 to 2001 AX3-61
AX3-22 The fourth highest 8-hour concentration at Yellowstone National Park AX3-63
AX3-23 Hourly average O3 concentrations at Yellowstone National Park, Wyoming
for the period of January to December 2001 AX3-64
AX3-24 The comparison of the seasonal (April-October) diurnal patterns for
urban-influenced (Jefferson County, KY) and a rural-influenced
(Oliver County, ND) monitoring sites using 2002 hourly data AX3-71
AX3-25a-d Diurnal behavior of O3 at rural sites in the United States in July AX3-72
AX3-26a,b Diurnal pattern of 1-h O3 concentration on August 3, 2004 at various sites
in Maryland AX3-73
AX3-27 Diurnal and 1-month composite diurnal variations in O3 concentrations,
Washington, DC July 1981 AX3-74
AX3-28 Diurnal and 1-month composite diurnal variations in O3 concentrations,
St. Louis County, MO September 1981 AX3-75
AX3-29 Diurnal and 1-month composite diurnal variations in O3 concentrations,
Alton, IL October 1981 (fourth quarter) AX3-75
AX3-30 Composite diurnal patterns of O3 concentrations by quarter, Alton, IL
October 1981 AX3-76
I-xxii
-------
List of Figures
(cont'd)
Number Page
AX3-31a-f Quarterly composite diurnal patterns of O3 concentrations at selected sites
representing potential for exposure of major crops, 1981 AX3-78
AX3-32 Composite diurnal O3 pattern at a rural National Crop Loss Assessment
Network site in Argonne, IL August 6 through September 30, 1980 AX3-79
AX3-33 Composite diurnal O3 pattern at selected national forest sites in the
United States using 2002 hourly average concentration data AX3-79
AX3-34a,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 AX3-81
AX3-35a-h Seasonal variations in O3 concentrations as indicated by the 1-h maximum in
each month at selected sites, 2002 AX3-82
AX3-36 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Charlotte, NC-Gastoma-Rock Hill, SC MSA AX3-88
AX3-37 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Baton Rouge, LA MSA AX3-89
AX3-38 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Detroit-Ann Arbor-Flint, MI CMS A AX3-90
AX3-39 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the St. Louis, MO-IL MSA AX3-91
AX3-40 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Phoenix-Mesa, AZ MSA AX3-93
AX3-41 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Fresno, CA MSA AX3-94
AX3-42 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Bakersfield, CA MSA AX3-95
AX3-43 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Los Angeles-Orange County, CA CMS A AX3-96
AX3-44 Locations of O3 sampling sites (a) by AQS ID# (b) and intersite correlation
statistics (c) for the Riverside-Orange County, CA CMS A AX3-98
I-xxiii
-------
List of Figures
(cont'd)
Number Page
AX3-45a-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 AX3-105
AX3-46a-e Integrated exposures for three non-Mountain Cloud Chemistry Program
Shenandoah National Park sites, 1983 to 1987 AX3-106
AX3-47 Trends for 1983 to 2002 period for the annual second maximum 1-h average
O3 concentration AX3-111
AX3-48 Trends for 1983 to 2002 period for the annual fourth maximum 8-h average
O3 concentration AX3-111
AX3-49 Trend in 1-h O3 level (based on annual second highest daily maximum
concentration) for 1983 to 2002 averaged across EPA regional office
boundaries AX3-112
AX3-50 Trend in 8-h O3 level (based on annual fourth highest maximum concentration)
for 1983 to 2002 averaged across EPA regional office boundaries AX3-112
AX3-51 Comparison of actual and meteorological adjusted 8-h O3 trends for the
period 1993 to 2003 AX3-114
AX3-52 Trend in the 4th highest daily maximum 1-h O3 concentration over 3 years for
the period of 1980 to 2001 for a monitoring site in Los Angeles, CA
(060371301) AX3-114
AX3-53 Trend in the 4th highest daily maximum 1-h O3 concentration over
3 years for the period of 1980 to 2001 for a monitoring site in Phoenix,
AZ (040133002) AX3-115
AX3-54 Trend in the 4th highest daily maximum 1-h O3 concentration over 3 years
for the period of 1980 to 2001 for a monitoring site in Denver, CO
(080590002) AX3-115
AX3-55 Trend in the 4th highest daily maximum 1-h O3 concentration over 3 years for
the period of 1980 to 2001 for a monitoring site in Chicago, IL (170317002) . . . AX3-116
AX3-56 Trend in the 4th highest daily maximum 1-h O3 concentration over 3 years for
the period of 1980 to 2001 for a monitoring site in Detroit, MI (260990009) . . . AX3-116
AX3-57 Trend in the 4th highest daily maximum 1-h O3 concentration over 3 years for
the period of 1980 to 2001 for a monitoring site in Houston, TX (482011037) . . AX3-117
I-xxiv
-------
List of Figures
(cont'd)
Number Page
AX3-58 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period of 1980 to 2001 for a monitoring site in Fairfield, CN
(090013007) AX3-117
AX3-59 Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Washington, DC (110010025) AX3-118
AX3-60 Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Los Angeles, CA (060371301) AX3-119
AX3-61 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period of 1980 to 2001 for a monitoring site in Phoenix, AZ
(040133002) AX3-119
AX3-62 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period of 1980 to 2001 for a monitoring site in Denver, CO
(080590002) AX3-120
AX3-63 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period in 1980 to 2001 for a monitoring site in Chicago, IL
(170317002) AX3-120
AX3-64 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period of 1980 to 2001 for a monitoring site in Detroit, MI
(260990009) AX3-121
AX3-65 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period of 1980 to 2001 for a monitoring site in Houston, TX
(482011037) AX3-121
AX3-66 Trend in the 4th highest daily maximum 8-h O3 concentration averaged over
3 years for the period of 1980 to 2001 for a monitoring site in Fairfield, CN
(090013007) AX3-122
AX3-67 Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Washington, DC (110010025) AX3-122
AX3-68 Changes in the distribution of hourly average O3 concentrations over time
for monitoring sites in Ventura County (CA), Cook County (IL), Bristol
County (MA), and Boone County (KY) AX3-123
I-xxv
-------
List of Figures
(cont'd)
Number Page
AX3-69a-d Measured O3 (ppbv) versus PAN (pptv) in Tennessee, including (a) aircraft
measurements, and (b, c, and d) suburban sites near Nashville AX3-132
AX3-70 Measured correlation between benzene and NOy at a measurement site in
Boulder, CO AX3-133
AX3-71 Conditional mean PM25 concentrations versus conditional mean O3
concentrations observed at Fort Meade, MD from July 1999 to July 2001 AX3-135
AX3-72 The co-occurrence pattern for O3 and NO2 AX3-138
AX3-73 The co-occurrence pattern for O3 and NO2 using 2001 data from the AQS AX3-139
AX3-74 The co-occurrence pattern for O3 and SO2 AX3-140
AX3-75 The co-occurrence pattern for O3 and SO2 using 2001 data from AQS AX3-141
AX3-76 The co-occurrence pattern for O3 and PM25 using 2001 data from AQS AX3-142
AX3-77a Monthly maximum hourly average O3 concentrations at Yellowstone
National Park, Wyoming in 1998, 1999, 2000, and 2001 AX3-146
AX3-77b Hourly average O3 concentrations at Yellowstone National Park, Wyoming
for the period January to December 2001 AX3-146
AX3-78 (a) Contour plot of CO mixing ratios (ppbv) observed in and near the
June 15, 1985, mesoscale convective complex in eastern Oklahoma AX3-148
AX3-80 Schematic diagram of a meteorological mechanism involved in high
concentrations of ozone found in spring in the lower troposphere off the
American East Coast AX3-158
AX3-81 Ozone vertical profile at Boulder, Colorado on May 6, 1999 at 1802 UTC
(1102LST) AX3-161
AX3-82 CASTNet stations in the continental United States for 2001 AX3-165
AX3-83 Monthly mean afternoon (1300 to 1700 hours LT) concentrations (ppbv)
in surface air averaged over the CASTNet stations (Figure AX3-82) in
each U.S. quadrant for March to October 2001 AX3- 1 66
I-xxvi
-------
List of Figures
(cont'd)
Number Page
AX3-84 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) AX3-170
AX3-85 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 AX3-170
AX3-86 Same as Figure AX3-85 but for Yellowstone National Park, Wyoming in
March to May 2001 AX3-172
AX3-87 Same as Figure AX3-86 but for March of 2001 at selected western (left
column) and southeastern (right column) sites AX3-173
AX3-88 Ozone decay processes versus time measured for several indoor rooms AX3-181
AX3-89 Air exchange rates and outdoor and indoor O3 concentrations during the
summer at telephone switching station in Burbank, CA AX3-191
AX3-90 Air exchange rates and outdoor and indoor O3 concentrations during the fall
at a telephone switching station in Burbank, CA AX3-192
AX3-91 Indoor and outdoor O3 concentration in moving cars AX3-194
AX3-92 Measured outdoor O3 concentrations (thin line) and modeled indoor
concentrations (bold line) AX3-196
AX3-93a Detailed diagram illustrating components of APEX exposure model AX3-212
AX3-93b Detailed diagram illustrating components of APEX exposure model AX3-213
I-xxvii
-------
Authors, Contributors, and Reviewers
To be inserted in Second External Review draft.
I-xxviii
-------
U.S. Environmental Protection Agency Project Team
for Development of Air Quality Criteria for Ozone
and Related Photochemical Oxidants
To be inserted in Second External Review draft.
I-xxix
-------
U.S. Environmental Protection Agency Science Advisory Board (SAB)
Staff Office Clean Air Scientific Advisory Committee (CASAC)
Ozone Review Panel
To be inserted in Second External Review draft.
I-xxx
-------
Abbreviations and Acronyms
AC
AER
AEROCE
AGL
AHC
AHCs
AirPEX
AirQUIS
AIRS
AP-CIMS
APEX
AQCD
AQS
air conditioning
air exchange rate
Atmospheric/Ocean Ocean Chemistry Experiment
above ground level
anthropogenic hydrocarbons
aromatic hydrocarbons
Air Pollution Exposure (model)
Air Quality Information System (model)
Aerometric Information Retrieval System
Atmospheric Pressure Chemical lonization Mass Spectrometer
Air Pollutants Exposure Model
Air Quality Criteria Document
Air Quality System
ATLAS
A/V
BC
BEIS
BHC
BME
C
CAA
CAAA
CADS
RAMS
CASAC
CASTNet
CBL
CEPEX
surface-to-volume ratio
black carbon
Biogenic Emission Inventory System
biogenic hydrocarbons
Bayesian Maxim Eutropy
carbon
Clean Air Act
Clean Air Act Amendments of 1990
Cincinnati Activity Diary Study
Regional Atmospheric Modeling System
Clean Air Scientific Advisory Committee
Clean Air Status and Trends Network
convective boundary layer
Central Equatorial Pacific Experiment
I-xxxi
-------
CFD
computational fluid dynamics
CG
cloud to ground
CHAD
Consolidated Human Activities Database
CH4
methane
CIMS
Chemical lonization Mass Spectroscopy
CL
chemiluminescence
CLM
chemiluminescence method
CMAQ
Community Model for Air Quality
CMBO
chloromethylbutenone
CMSA
consolidated metropolitan statistical area
CN
condensation nuclei
CO
carbon monoxide
CO9
carbon dioxide
COD
coefficient of divergence
CTM
Chemistry Transport Model
DA
dry airstream
DI
dry intrusion
DIAL
differential absorption lidar
DOAS
differential optical absorption spectroscopy/spectrometry
EDMAS
Exposure and Dose Modeling and Analysis System
EOF
empirical orthogonal function
EPA
U.S. Environmental Protection Agency
EPEM
Event Probability Exposure Model
EPRI
Electric Power Research Institute
ERAQS
Eastern Regional Air Quality Study
EVR
equivalent ventilation rate
FID
flame ionization detection
FR
Federal Register
FTIR
Fourier transform infrared absorption spectroscopy
I-xxxii
-------
GC
gas chromatography
GCE
Goddard Cumulus Ensemble
CG-FID
gas chromatography - flame ionization detection
GEOS-1 DAS
Goddard Earth Observing System Data Assimilation System
GEOS-CHEM
H+
hydrogen ion
HC
hydrocarbon
HCFC
hydrochlorofluorocarbon
H2CO, HCHO
formaldehyde
HNO2, HONO
nitrous acid
HNO,
nitric acid
HO
hydroxyl
HO9
hydroperoxyl; hydroperoxy
hydrogen peroxide
HPLC
high-performance liquid chromatography
sulfuric acid
IBM
individual-based model
1C
intracloud
ID
identification (number)
ICEM
Indoor Chemistry and Exposure Model
I/O
indoor/outdoor
IPCC
Intergovernmental Panel on Climate Change
IPMMI
International Photolysis Frequency Measurement and Modeling
Intercomparison
ISCCP
International Satellite Cloud Climatology Project
LFT
lower free troposphere
LIF
laser-induced fluorescence
LLJ
low level jet
LSI
local standard time
LT
local time
I-xxxiii
-------
LWC
liquid water content
MAQSIP
Multiscale Air Quality Simulation Platform
MBL
marine boundary layer
MBTH
3 -methyl-2-benzothiazolone hy drazone
MCCP
Mountain Cloud Chemistry Program
MCM
master chemical mechanism
MENTOR
Modeling Environment for Total Risk Studies
MENTOR-
OPERAS
Modeling Environment for Total Risk Studies — Ozone and Particles
Exposure Risk Analysis System
MET
metabolic equivalent
MIESR
matrix isolation on ESR spectroscopy
MM5
Mesoscale Model, version 5
MoOx
molybdenum oxides
MOZAIC
Measurement of Ozone and Water Vapor by Airbus In-Service Aircraft
MPAN
peroxymethacryloyl nitrate; peroxy-methacrylic nitric anhydride
MS
mass spectrometry
MSA
Metropolitan Statistical Area
MS/MS
tandem mass spectrometry
N100
number of hours at an ozone concentration of > 0.10 ppm
dinitrogen pentoxide
NA, N/A
not available
NAAQS
National Ambient Air Quality Standards
NADP
National Atmospheric Deposition Program
NAMS
National Air Monitoring Station
NAPAP
National Acid Precipitation Assessment Program
NAPBN
National Air Pollution Background Network
NARE
NCAR
North Atlantic Regional Experiment
NBS National Bureau of Standards; now National Institute of Standards and
Technology
National Center for Atmospheric Research
I-xxxiv
-------
NCEA-RTP
NCLAN
ND
NESCAUM
NDDN
NEM
NF
NH3
NHAPS
MST
NM
NMHC
NMOC
NMVOC
NO
NO2
N2O
NO3
NOX
N0y
NOZ
NOAA
NP
NPAN
NRC
NS
NTRMs
0('D)
National Center for Environmental Assessment Division in Research
Triangle Park, NC
National Crop Loss Assessment Network
not detectable; not detected
Northeast States for Coordinated Air Use Management
National Dry Deposition Network
National Ambient Air Quality Standards Exposure Model
national forest
ammonia
National Human Activity Pattern Survey
National Institute of Standards and Technology
national monument
nonmethane hydrocarbon
nonmethane organic compound
nonmethane volatile organic compound
nitric oxide
nitrogen dioxide
nitrous oxide
nitrate
nitrogen oxides
reactive nitrogen system components; sum of NOX and NOZ; odd nitrogen
species
difference between NOy and NOX
National Oceanic and Atmospheric Administration
national park
methacryloylperoxynitrate
National Research Council
nonsignificant
NIST Traceable Reference Materials
electronically excited oxygen atom
I-xxxv
-------
0(3P)
03
OAQPS
OEMs
Obs.
OH
OPE
ox
ground-state oxygen atom
ozone
Office of Air Quality Planning and Standards
observationally based methods
observations
hydroxyl; hydroxy
ozone production efficiency
odd oxygen species
P90 values for the 90th percentile
PAHs
PAMS
PAN
PANs
poly cyclic aromatic hydrocarbons
Photochemical Aerometric Monitoring System
peroxyacetyl nitrate; peroxyacetic nitric anhydride
peroxyacyl nitrates
PAR
p-ATP
PEL
PBM
PCA
PEM
pH
PM
PM25
pNEM
POC
PPN
PPP
PRB
PTR-MS
PV
/7-acetamidophenol
planetary boundary layer
population-based model
principal component analysis
personal exposure monitor
hydrogen ion concentration
particulate matter
fine particulate matter
Probabilistic National Ambient Air Quality Standard Exposure Model
particulate organic carbon
peroxypropionyl nitrate; peroxypropionic nitric anhydride
power plant plume
policy relevant background
proton-transfer-reaction mass spectroscopy
potential vorticity
I-xxxvi
-------
linear regression correlation coefficient; Pearson correlation coefficient
R2
multiple correlation coefficient
RACM
Regional Air Chemistry Mechanism
RADM
Regional Acid Deposition Model
RDBMS
Relational Database Management Systems
REHEX
Regional Human Exposure Model
RH
relative humidity
RMR
resting metabolic rate
RO9
organic peroxyl; organic peroxy
RRMS
Relatively Remote Monitoring Sites
SAB
Science Advisory Board
SAI
Systems Applications International
SAPRC
SHEDS
Statewide Air Pollution Research Center, University of California, Riverside
SAROAD Storage and Retrieval of Aerometric Data (U.S. Environmental Protection
Agency centralized database; superseded by Aerometric Information Retrieval
System [AIRS])
Simulation of Human Exposure and Dose System
SLAMS
State and Local Air Monitoring Station
SO9
sulfur dioxide
SO2
sulfate
SOA
secondary organic aerosol
SOS
Southern Oxidant Study
SRM
standard reference material
STE
stratospheric-tropospheric exchange
STEP
Stratospheric-Tropospheric-Exchange Project
STPD
standard temperature and pressure, dry
STRF
Spatio-Temporal Random Field
SUM06
seasonal sum of all hourly average concentrations > 0.06 ppm
SUM07
seasonal sum of all hourly average concentrations > 0.07 ppm
SUM08
seasonal sum of all hourly average concentrations > 0.08 ppm
I-xxxvii
-------
SURE
TAR
TOLAS
Tg
Sulfate Regional Experiment Program
Third Assessment Report
tunable-diode laser absorption spectroscopy
teragram
TOMS
TOPSE
TPLIF
TRIM
TTFMS
TVA
UAM
UTC
UV
UV-B
UV-DIAL
VOC
WFM
WMO/UNEP
W126
Tropospheric Ozone Production About the Spring Equinox
two-photon laser-induced fluorescence
Total Risk Integrated Methodology (model)
two-tone frequency-modulated spectroscopy
Tennessee Valley Authority
Urban Airshed Model
Coordinated Universal Time
ultraviolet
ultraviolet radiation of wavelengths 280 to 320 nm
Ultraviolet Differential Absorption Lidar
volatile organic compound
White Face Mountain
World Meteorological Organization/United Nations Environment Program
cumulative integrated exposure index with a sigmoidal weighting function
I-xxxviii
-------
i EXECUTIVE SUMMARY
2
3
4 To be prepared after CAS AC review of this First External Review Draft and then inserted
5 in Second External Review Draft to be released for further public comment and CASAC review.
January 2005 E-l DRAFT-DO NOT QUOTE OR CITE
-------
i 1. INTRODUCTION
2
3
4 This is an update revision of the document, "Air Quality Criteria for Ozone and Related
5 Photochemical Oxidants" published by the U.S. Environmental Protection Agency (EPA) in
6 1996 (U.S. Environmental Protection Agency, 1996). That 1996 Ozone Air Quality Criteria
7 Document (O3 AQCD) provided scientific bases for Congressionally-mandated periodic review
8 by the EPA of the National Ambient Air Quality Standards for Ozone (O3 NAAQS), which
9 culminated in promulgation of new O3 NAAQS by EPA in 1997.
10 The present document critically assesses the latest scientific information relative to
11 determining the health and welfare effects associated with the presence of various concentrations
12 of O3 and related oxidants in ambient air. It builds upon the previous 1996 EPA O3 AQCD,
13 by focusing on evaluation and integration of information relevant to O3 NAAQS criteria
14 development that has become available since that covered by the 1996 criteria review; and it will
15 provide scientific bases for the current periodic review of the O3 NAAQS.
16 This introductory chapter of the revised O3 AQCD presents: (a) background information
17 on legislative requirements, the criteria and NAAQS review process, and the history of O3
18 NAAQS reviews (including a chronology of changes in key elements of the O3 standards);
19 (b) an overview of the current O3 criteria review process and projected schedule (including
20 approaches and procedures used to prepare this document, as well as projected key milestones);
21 and (c) an orientation to the general organizational structure and content of the document.
22
23
24 1.1 LEGAL AND HISTORICAL BACKGROUND
25 1.1.1 Legislative Requirements
26 Two sections of the Clean Air Act (CAA) govern the establishment, review, and revision
27 of National Ambient Air Quality Standards (NAAQS). Section 108 (42 U.S.C. 7408) directs the
28 Administrator of the U.S. Environmental Protection Agency (EPA) to identify ambient air
29 pollutants that may be reasonably anticipated to endanger public health or welfare and to issue
30 air quality criteria for them. These air quality criteria are to reflect the latest scientific
January 2005 1-1 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
information useful in indicating the kind and extent of all identifiable effects on public health or
welfare that may be expected from the presence of a given pollutant in ambient air.
Section 109(a) of the CAA (42 U.S.C. 7409) directs the Administrator of EPA to propose
and promulgate primary and secondary NAAQS for pollutants identified under Section 108.
Section 109(b)(l) defines a primary standard as one that, in the judgment of the Administrator, is
requisite to protect the public health (see inset below) based on the criteria and allowing for an
adequate margin of safety. The secondary standard, as defined in Section 109(b)(2), must
specify a level of air quality that, in the judgment of the Administrator, is requisite to protect the
public welfare (see inset below) from any known or anticipated adverse effects associated with
the presence of the pollutant in ambient air, based on the criteria.
PUBLIC HEALTH EFFECTS
i D Effects on the health of the general population,
or identifiable groups within the population,
who are exposed to pollutants in ambient air
i D Effects on mortality
i D Effects on morbidity
i D Effects on other health conditions including
indicators of:
• pre-morbid processes,
• risk factors, and
• disease
PUBLIC WELFARE EFFECTS
i D Effects on personal comfort and well-being
i D Effects on economic values
i D Deterioration of property
i D Hazards to transportation
i D Effects on the environment, including:
animals
climate
crops
materials
soils
1 vegetation
1 visibility
1 water
1 weather
1 wildlife
Section 109(d) of the CAA (42 U.S.C. 7409) requires periodic review and, if appropriate,
revision of existing criteria and standards. If, in the Administrator's judgment, the Agency's
review and revision of criteria make appropriate the proposal of new or revised standards, such
standards are to be revised and promulgated in accordance with Section 109(b). Alternatively,
the Administrator may find that revision of the standards is inappropriate and conclude the
review by leaving the existing standards unchanged. Section 109(d)(2) of the 1977 CAA
Amendments also requires that an independent scientific review committee be established to
advise the EPA Administrator on NAAQS matters, including the scientific soundness of criteria
(scientific bases) supporting NAAQS decisions. This role is fulfilled by the Clean Air Scientific
Advisory Committee (CAS AC) of EPA's Science Advisory Board (SAB).
January 2005
1-2
DRAFT-DO NOT QUOTE OR CITE
-------
1 1.1.2 Criteria and NAAQS Review Process
2 Periodic reviews by EPA of criteria and NAAQS for a given criteria air pollutant progress
3 through a number of steps, beginning with preparation by EPA's National Center for
4 Environmental Assessment Division in Research Triangle Park, NC (NCEA-RTP) of an air
5 quality criteria document (AQCD). The AQCD provides a critical assessment of the latest
6 available scientific information upon which the NAAQS are to be based. Drawing upon the
7 AQCD, staff of EPA's Office of Air Quality Planning and Standards (OAQPS) prepare a Staff
8 Paper that evaluates policy implications of the key studies and scientific information contained
9 in the AQCD and presents the conclusions and recommendations of the staff for standard-setting
10 options for the EPA Administrator to consider. The Staff Paper is intended to help "bridge the
11 gap" between the scientific assessment contained in the AQCD and the judgments required of
12 the Administrator in determining whether it is appropriate to retain or to revise the NAAQS.
13 Iterative drafts of both the AQCD and the Staff Paper (as well as other analyses, such as
14 exposure and/or risk assessments, supporting the Staff Paper) are made available for public
15 comment and CAS AC review. The final versions of the AQCD and Staff Paper incorporate
16 changes made in response to CAS AC and public review. Based on the information in these
17 documents, the Administrator proposes decisions on whether to retain or revise the NAAQS,
18 taking into account public comments and CASAC advice and recommendations. The
19 Administrator's proposed decisions are published in the Federal Register, with a preamble that
20 presents the rationale for the decisions and solicits public comment. The Administrator
21 makes a final decision after considering comments received on the proposed decisions. The
22 Administrator's final decisions are promulgated in a Federal Register notice that addresses
23 significant comments received on the proposal.
24 NAAQS decisions involve consideration of the four basic elements of a standard:
25 indicator, averaging time, form, and level. The indicator defines the pollutant to be measured in
26 the ambient air for the purpose of determining compliance with the standard. The averaging
27 time defines the time period over which air quality measurements are to be obtained and
28 averaged, considering evidence of effects associated with various time periods of exposure.
29 The form of a standard defines the air quality statistic that is to be compared to the level of the
30 standard (i.e., an ambient concentration of the indicator pollutant) in determining whether an
31 area attains the standard. The form of the standard specifies the air quality measurements that
January 2005 1 -3 DRAFT-DO NOT QUOTE OR CITE
-------
1 are to be used for compliance purposes (e.g., the 98th percentile of an annual distribution of
2 daily concentrations; the annual arithmetic average), the monitors from which the measurements
3 are to be obtained (e.g., one or more population-oriented monitors in an area), and whether the
4 statistic is to be averaged across multiple years. These basic elements of a standard are the
5 primary focus of the staff conclusions and recommendations in the Staff Paper and in the
6 subsequent rulemaking, building upon the policy-relevant scientific information assessed in the
7 AQCD and on the policy analyses contained in the Staff Paper. These four elements taken
8 together determine the degree of public health and welfare protection afforded by the NAAQS.
9
10 1.1.3 Regulatory Chronology1
11 On April 30, 1971, the EPA promulgated primary and secondary NAAQS for
12 photochemical oxidants under Section 109 of the CAA (36 FR 8186). These were set at an
13 hourly average of 0.08 ppm total photochemical oxidants, not to be exceeded more than 1 h per
14 year. On April 20, 1977, the EPA announced (42 FR 20493) the first review and updating of the
15 1970 Air Quality Criteria Document for Photochemical Oxidants in accordance with Section
16 109(d) of the CAA. In preparing that AQCD, the EPA made two external review drafts of the
17 document available for public comment, and these drafts were peer reviewed by the
18 Subcommittee on Scientific Criteria for Photochemical Oxidants of EPA's Science Advisory
19 Board (SAB). A final revised AQCD for ozone (O3) and other photochemical oxidants was
20 published on June 22, 1978.
21 Based on the 1978 revised AQCD and taking into account the advice and recommendations
22 of the SAB Subcommittee and public comments, the EPA announced (44 FR 8202) a final
23 decision to revise the NAAQS for photochemical oxidants on February 8, 1979. That final
24 rulemaking revised the primary standard from 0.08 ppm to 0.12 ppm, set the secondary standard
25 to be the same as the primary standard, changed the chemical designation of the standards from
26 photochemical oxidants to O3, and revised the definition of the point at which the standard is
27 attained as indicated in Table 1-1.
28
'This following text is excerpted and adapted from the "Proposed Decision on the National Ambient
Air Quality Standards for Ozone," 57 FR 35542, 35542-35557 (August, 10, 1992) and the "National Ambient Air
Quality Standards for Ozone; Final Rule," 62 FR 38856, 83356-38896 (July 18, 1997).
January 2005 1 -4 DRAFT-DO NOT QUOTE OR CITE
-------
Table 1-1. National Ambient Air Quality Standards (NAAQS) for Ozone
Date of Promulgation Primary and Secondary NAAQS Averaging Time
February 8, 1979 0.12 ppnf (23 5 ug/m3) 1 hb
July 18, 1997 0.08 ppma (157 ug/m3) 8 hc
al ppm = 1962 ug/m3, 1 ug/m3 = 5.097 x 10'4 ppm @ 25 °C, 760 mm Hg.
bThe standard is attained when the expected number of days per calendar year with a maximum hourly average
concentration above 235 ug/m3 (0.12 ppm) is equal to or less than one.
"Based on the 3-year average of the annual fourth-highest daily maximum 8-h average concentration measured
at each monitor within an area.
Source: Federal Register (1979, 1997).
1 On March 17, 1982, in response to requirements of Section 109(d) of the CAA, the EPA
2 announced (47 FR 11561) that it planned to revise the existing 1978 AQCD for O3 and Other
3 Photochemical Oxidants, and on August 22, 1983, it announced (48 FR 38009) that review of the
4 primary and secondary NAAQS for O3 had been initiated. The EPA provided a number of
5 opportunities for expert review and public comment on revised chapters of the AQCD, including
6 two public peer-review workshops in December 1982 and November 1983. Comments made at
7 both workshops were considered by EPA in preparing the First External Review Draft that was
8 made available (49 FR 29845) on July 24, 1984, for public review.
9 On February 13, 1985 (50 FR 6049) and then on April 2, 1986 (51 FR 11339), the EPA
10 announced two public CAS AC meetings, which were held on March 4-6, 1985 and April 21-22,
11 1986, respectively. At these meetings, the CAS AC reviewed external review drafts of the
12 revised AQCD for O3 and Other Photochemical Oxidants. After these two reviews, the Chair
13 summarized CASAC's consensus view in an October 1986 letter to the EPA Administrator,
14 which stated that the document "represents a scientifically balanced and defensible summary of
15 the extensive scientific literature." Taking into account public and CAS AC comments on the
16 two external review drafts, revisions were made by EPA and the final document was released by
17 EPA in August 1986.
18 The first draft of the Staff Paper "Review of the National Ambient Air Quality Standards
19 for Ozone: Assessment of Scientific and Technical Information" drew upon key findings and
20 conclusions from the AQCD and was reviewed by CAS AC at an April 21-22, 1986 public
January 2005 1-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 meeting. At that meeting, the CASAC recommended that new information on prolonged O3
2 exposure effects be considered in a second draft of the Staff Paper. The CASAC reviewed the
3 resulting second draft and also heard a presentation of new and emerging information on the
4 health and welfare effects of O3, at a public review meeting held on December 14-15, 1987. The
5 CASAC concluded that sufficient new information existed to recommend incorporation of
6 relevant new data into a supplement to the 1986 AQCD (O3 Supplement) and in a third draft of
7 the Staff Paper.
8 A draft O3 Supplement, "Summary of Selected New Information on Effects of Ozone on
9 Health and Vegetation: Draft Supplement to Air Quality Criteria for Ozone and Other
10 Photochemical Oxidants," and the revised Staff Paper were made available to CASAC and to the
11 public in November 1988. The O3 Supplement assessed selected literature concerning exposure-
12 and concentration-response relationships observed for health effects in humans and experimental
13 animals and for vegetation effects that appeared in papers published or in-press from 1986
14 through early 1989. On December 14-15, 1988, CASAC held a public meeting to review these
15 documents. The CASAC sent the EPA Administrator a letter, dated May 1, 1989, which stated
16 that the draft O3 Supplement, the 1986 AQCD, and the draft Staff Paper "provide an adequate
17 scientific basis for the EPA to retain or revise the primary and secondary standards of ozone."
18 The CASAC concluded (a) that it would be some time before sufficient new information on the
19 health effects of multihour and chronic exposure to O3 would be published in scientific journals
20 to receive full peer review and, thus, be suitable for inclusion in a criteria document and (b) that
21 such information could be considered in the next review of the O3 NAAQS. A final version of
22 the O3 Supplement was published in 1992 (U.S. Environmental Protection Agency, 1992).
23 On October 22, 1991, the American Lung Association and other plaintiffs filed suit to
24 compel the Agency to complete the review of the criteria and standards for O3 in accordance
25 with the CAA. The U.S. District Court for the Eastern District of New York subsequently issued
26 an order requiring the EPA to announce its proposed decision on whether to revise the standards
27 for O3 by August 1, 1992, and to announce its final decision by March 1, 1993.
28 The proposed decision on O3, appearing in the Federal Register on August 10, 1992 (57 FR
29 35542), indicated that revision of the existing 1-h NAAQS was not appropriate at this time.
30 A public hearing on this decision took place in Washington, DC on September 1, 1992, and
31 public comments were received through October 9, 1992. The final decision not to revise the
January 2005 1 -6 DRAFT-DO NOT QUOTE OR CITE
-------
1 1-h NAAQS was published in the Federal Register on March 9, 1993 (58 FR 13008). However,
2 that decision did not take into consideration a number of more recent studies on the health
3 and welfare effects of O3 that had been published since the last of the literature assessed in the
4 O3 Supplement (i.e., studies available through 1985 and into early 1986).
5 The Agency initiated consideration of such studies as part of the next congress!onally-
6 mandated periodic review of criteria and NAAQS for Ozone. The new studies were assessed in
7 revised workshop draft O3 AQCD chapters that were peer reviewed in July and September 1993,
8 followed by public release of the First External Review Draft in February 1994 and CAS AC
9 review on July 20-21, 1994. Further drafts of the O3 AQCD, revised in response to public
10 comments and CASAC review, were reviewed by CAS AC on March 21-25, 1995, and at a final
11 CASAC review meeting on September 19-20, 1995. The scientific soundness of the revised O3
12 AQCD was recognized by CASAC in a November 28, 1995 letter to the EPA Administrator;
13 and the final AQCD for O3 was published in July 1996.
14 The first draft of the associated Staff Paper, "Review of the National Ambient Air Quality
15 Standards for Ozone: Assessment of Scientific and Technical Information," was also reviewed
16 by CASAC at the March 21-22, 1995 public meeting. CASAC also reviewed subsequent drafts
17 of the Staff Paper at public meetings on September 19-20, 1995 and March 21, 1996, with
18 completion of CASAC review of the primary and secondary standard portions of the draft Staff
19 Paper being communicated in letters to the EPA Administrator dated November 30, 1995 and
20 April 4, 1996, respectively. The final O3 Staff Paper was published in June 1996.
21 On December 13, 1996 EPA published its proposed decision to revise the O3 NAAQS
22 (61 FR 65716). EPA provided extensive opportunities for public comment on the proposed
23 decision, including several public hearings and two national satellite telecasts. EPA's final
24 decision to promulgate a new 8-h O3 NAAQS (see Table 1-1) was published on July 18, 1997
25 (62 FR 38856).
26 Following promulgation of the new standards, numerous petitions for review of the
27 standards were filed in the U.S. Court of Appeals for the District of Columbia Circuit (D.C.
28 Circuit)2. On May 14, 1999, the Court remanded the O3 NAAQS to EPA, finding that section
29 109 of the CAA, as interpreted by EPA, effected an unconstitutional delegation of legislative
:''American Trucking Associations v'. EPA, No. 97-1441
January 2005 1 -7 DRAFT-DO NOT QUOTE OR CITE
-------
1 authority3. In addition, the Court directed that, in responding to the remand, EPA should
2 consider the potential beneficial health effects of O3 pollution in shielding the public from the
3 effects of solar ultraviolet (UV) radiation. On January 27, 2000, EPA petitioned the U.S.
4 Supreme Court for certiorari on the constitutional issue (and two other issues), but did not
5 request review of the D.C. Circuit ruling regarding the potential beneficial health effects of O3.
6 On February 27, 2001 the U.S. Supreme Court unanimously reversed the judgment of the D.C.
7 Circuit on the constitutional issue, holding that section 109 of the CAA does not delegate
8 legislative power to the EPA in contravention of the Constitution, and remanded the case to the
9 D.C. Circuit to consider challenges to the O3 NAAQS that had not been addressed by that Court's
10 earlier decisions4. On March 26, 2002, the D.C. Circuit issued its final decision, finding the
11 1997 O3 NAAQS to be "neither arbitrary nor capricious," and denied the remaining petitions
12 for review5.
13 On November 14, 2001 EPA proposed to respond to the Court's remand to consider the
14 potential beneficial health effects of O3 pollution in shielding the public from the effects of solar
15 UV radiation by leaving the 1997 8-h NAAQS unchanged. Following a review of information in
16 the record and the substantive comments received on the proposed response, EPA issued a final
17 response to the remand, reaffirming the 8 h O3 NAAQS (68 FR 614, January 6, 2003).
18
19
20 1.2 CURRENT OZONE CRITERIA AND NAAQS REVIEW
21 1.2.1 Key Milestones and Procedures for Document Preparation
22 It is important to note at the outset that development of the present document has and will
23 continue to include substantial external expert review and opportunities for public input through
24 (a) public workshops involving the general aerosol scientific community, (b) iterative reviews of
25 successive drafts by CASAC, and (c) comments from the public on successive drafts. The
26 extensive external input received through such reviews will help to ensure that the review of the
3 American Trucking Associations v. EPA, 175 F.3d 1027 (D.C. Cir, 1999)
^Whitman v. American Trucking Associations, 531 U.S. 457 (2001)
^American Trucking Associations v. EPA, 283 F.3d 355, (D.C. Cir. 2002)
January 2005 1-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 O3 standards will be based on critical assessment in this document of the latest available
2 pertinent science.
3 The procedures for developing this revised O3 AQCD build on experience derived from the
4 other recent criteria document preparation efforts, with key milestones for development of this
5 O3 AQCD being listed in Table 1-2. Briefly, the respective responsibilities for production of this
6 O3 AQCD and key milestones are as follows. An NCEA-RTP Ozone Team is responsible for the
7 creation and implementation of a project plan for developing the O3 AQCD, taking into account
8 input from individuals in other EPA program and policy offices identified as part of the EPA
9 Ozone Work Group. The resulting plan, i.e., the Project Work Plan for Revised Air Criteria for
10 Ozone and Related Photochemical Oxidants (November 2002), was discussed with CAS AC in
11 January 2003. An ongoing literature search that was underway prior to initiation of work on this
12 document has continued throughout its preparation to identify pertinent O3 literature published
13 since early 1996. Under the processes established in Sections 108 and 109 of the CAA, the EPA
14 officially initiated the current criteria and NAAQS review by announcing the commencement of
15 the review in the Federal Register (65 FR 57810, September, 2000) with a call for information.
16 That Federal Register notice included (1) a request asking for recently available research
17 information on O3 that may not yet have been published and (2) a request for individuals with the
18 appropriate type and level of expertise to contribute to the writing of O3 AQCD materials to
19 identify themselves. The specific authors of chapters or sections of the proposed document
20 included both EPA and non-EPA scientific experts, who were selected on the basis of their
21 expertise on the subject areas and their familiarity with the relevant literature. The project team
22 defined critical issues and topics to be addressed by the authors and provided direction in order
23 to emphasize evaluation of those studies most clearly identified as important for standard setting.
24 As with other NAAQS reviews, critical assessment of relevant scientific information is
25 presented in this updated O3 AQCD. The main focus of this document is the evaluation and
26 interpretation of pertinent atmospheric science information, air quality data, human exposure
27 information, and health and welfare effects information newly published since that assessed in
28 the 1996 O3 AQCD. Draft versions of AQCD chapter materials were evaluated via expert peer-
29 consultation workshop discussions (see Table 1-2) that focused on the selection of pertinent
30 studies to be included in the chapters, the potential need for additional information to be added to
31 the chapters, and the quality of the characterization and interpretation of the literature. The
January 2005 1 -9 DRAFT-DO NOT QUOTE OR CITE
-------
Table 1-2. Key Milestones for Development of Revised Ozone Air Quality
Criteria Document"
Major Milestones
1. Literature Search
2. Federal Register Call for Information
3. Draft Project Plan Available for Public Comment
4. Revised Draft Project Plan Released for CASAC Review
5. CASAC Review of Draft Project Work Plan
6. Peer-Consultation Workshop on Draft Ecological Effects Materials
7. Peer-Consultation Workshops on Draft Atmospheric Science/Exposure
and Dosimetry/Health Chapters
8. First External Review Draft of O3 AQCD
9. Public Comment Period (90 days)
10. CASAC Public Review Meeting (First External Review Draft)
11. Second External Review Draft of O3 AQCD
12. Public Comment Period (90 days)
13. CASAC Public Review Meeting
14. Final O3 AQCD
Target Dates
Ongoing
September 2000
Dec 2001 -March 2002
December 2002
January 2003
April 2003
July 2004
January 2005
Feb - April 2005
April/May 2005
Aug/Sept 2005
Sept - Nov 2005
December 2005
February 2006
"Proposed schedule will be modified from time to time, as necessary, to reflect actual project requirements
and progress.
1 authors of the draft chapters then revised them on the basis of the workshop and/or other expert
2 review comments6. These and other integrative materials have been incorporated into this First
3 External Review Draft of the O3 AQCD (January 2005), which is being made available for
4 public comment and for CASAC review (see Table 1-2).
5 Following the upcoming April/May 2005 CASAC meeting, EPA plans to incorporate
6 revisions into the draft O3 AQCD in response to comments from CASAC and the public and to
7 make a Second External Review Draft available for further public comment and CASAC review
8 according to the schedule projected in Table 1-2. More specifically, that Second External
6It should be noted that materials contributed by non-EPA authors have, at times, been modified by EPA
Ozone Team staff in response to internal and/or external review comments, and that EPA is responsible for the
ultimate content of this O3 AQCD.
January 2005
1-10
DRAFT-DO NOT QUOTE OR CITE
-------
1 Review Draft is expected to be made available in August 2005 for public comment (90 days) and
2 to be reviewed by CAS AC at a public meeting in December 2005. The final O3 AQCD is to be
3 completed by February 28, 2006 and will be made publicly available electronically via an EPA
4 website and will then be subsequently printed. Its availability will be announced in the Federal
5 Register.
6 Once the CAS AC has reviewed the First External Review Draft of the revised O3 AQCD,
7 thus providing a preliminary basis for review of the existing standards, the EPA's Office of Air
8 Quality Planning and Standards (OAQPS) staff will prepare a draft O3 Staff Paper drawing upon
9 key information contained in the draft criteria document and presenting staff recommendations
10 on whether to retain or, if appropriate, revise the O3 NAAQS. After review of that draft Staff
11 Paper by the public and by CASAC, EPA will take public and CASAC comments into account in
12 producing a Second Draft Staff Paper. This Second Draft Staff Paper will be made available for
13 further public comment and CASAC review before EPA produces a final Staff Paper by
14 September 30, 2006.
15
16
17 1.3 ORGANIZATIONAL STRUCTURE OF THE DOCUMENT
18 1.3.1 General Document Format
19 The authors were provided with copies of the 1996 O3 AQCD (U.S. Environmental
20 Protection Agency, 1996) and were instructed to open each new section for the updated
21 document with concise summarization of key findings and conclusions from the previous 1996
22 O3 AQCD. After presentation of such background information, the remainder of each section
23 typically attempts to provide an updated discussion of newer literature and resulting key
24 conclusions. In some cases where no new information is available, the summary of key findings
25 and conclusions from the previous criteria document must suffice as the basis for current key
26 conclusions. Increased emphasis is placed in the main chapters of this revised O3 AQCD on
27 interpretative evaluation and integration of evidence pertaining to a given topic than has been
28 typical of previous EPA air quality criteria documents, with more detailed descriptions of
29 individual studies being provided in a series of accompanying annexes.
30 A list of references published since completion of the 1996 criteria document was made
31 available to the authors. The references were selected from information data base searches
January 2005 1-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 conducted by EPA. Additional references have been added to the list (e.g., missed or recently
2 published papers or "in press" publications) as work has proceeded in creating the draft
3 document materials. As an aid in selecting pertinent new literature, the authors were also
4 provided with a summary of issues that need to be addressed in the revised air quality criteria
5 document for O3. These issues were identified by authors and reviewers of the previous
6 documents and continue to be expanded, as appropriate, based on public discussions, workshops,
7 or other comments received by EPA.
8
9 1.3.2 Organization and Content of the Document
10 This revised AQCD for O3 and Related Photochemical Oxidants critically assesses
11 scientific information on the health and welfare effects associated with exposure to the
12 concentrations of these pollutants in ambient air. The document does not provide a detailed
13 literature review; but, rather, discusses cited references that reflect the current state of knowledge
14 on the most relevant issues pertinent to the NAAQS for O3. Although emphasis is placed on
15 discussion of health and welfare effects information, other scientific data are presented and
16 evaluated in order to provide a better understanding of the nature, sources, distribution,
17 measurement, and concentrations of O3 and related photochemical oxidants in ambient air,
18 as well as the measurement of population exposure to these pollutants.
19 The main focus of the scientific information discussed in the text comes from literature
20 published since completion of the 1996 O3 AQCD (U.S. Environmental Protection Agency,
21 1996). Emphasis is placed on studies conducted at or near O3 concentrations found in ambient
22 air. Other studies are included if they contain unique data, such as the documentation of a
23 previously unreported effect or of a mechanism for an observed effect; or if they were multiple-
24 concentration studies designed to provide exposure-response relationships. Generally, this is not
25 an issue for human clinical or epidemiology studies. However, for animal toxicology studies,
26 consideration is given mainly to those studies conducted at less than 1 ppm O3. Key information
27 from studies assessed in the previous O3 AQCD and whose data impacted the derivation of the
28 current NAAQS are briefly summarized in the text, along with specific citations to the previous
29 document. Prior studies are also discussed if they (1) are open to reinterpretation in light of
30 newer data, or (2) are potentially useful in deriving revised standards for O3. Generally, only
January 2005 1-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 information that has undergone scientific peer review and has been published (or accepted for
2 publication) through 2004 is included in the draft document.
3 The document will consist of three volumes. The first volume will include an Executive
4 Summary and Conclusions, as well as Chapters 1 through 3 and their accompanying annexes.
5 This introductory chapter (Chapter 1) presents background information on the purpose of the
6 document, legislative requirements, and the history of past O3 NAAQS regulatory actions, as
7 well as an overview of the organization and content of the document. Chapter 2 provides
8 information on the physics and chemistry of O3 and related photochemical oxidants in the
9 atmosphere. Chapter 3 covers tropospheric O3 environmental concentrations, patterns, and
10 exposure estimates.
11 Health information pertinent to derivation of the primary O3 NAAQS is then mainly
12 covered in the next several chapters (Chapters 4 through 8) contained in Volume II. Chapter 4
13 discusses O3 dosimetry aspects, and Chapters 5, 6, and 7 discuss animal toxicological studies,
14 human health effects from controlled-exposure studies, and epidemiologic studies of ambient air
15 exposure effects on human populations, respectively. Chapter 8 then provides an integrative and
16 interpretive evaluation of key information relevant to O3 exposure and health risks, of most
17 pertinence to the review of primary O3 NAAQS.
18 Volume III contains those chapters that assess welfare effects information pertinent to the
19 review of secondary O3 NAAQS. Chapter 9 deals with ecological and other environmental
20 effects of O3 and related photochemical oxidants. Chapter 10 assesses tropospheric O3
21 involvement in climate change processes, including determination of solar UV flux in Earth's
22 lower atmosphere. Lastly, Chapter 11 discusses O3 effects on man-made materials as a third
23 type of welfare effect of potential concern.
24
January 2005 1-13 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2
3 Federal Register. (1979) National primary and secondary ambient air quality standards: revisions to the National
4 Ambient Air Quality Standards for photochemical oxidants. F. R. (February 8) 44: 8202-8237.
5 Federal Register (1992) Proposed Decision on the National Ambient Air Quality Standards for Ozone, Preamble.
6 Federal Register 57: 35542-35557.
7 Federal Register (1997) National Ambient Air Quality Standards for Ozone; Final Rule. 40 CFR 50; Federal
8 Register 62: 38856-38896
9 U.S. Code (1999) Clean Air Act, title Ill-general, Section 302, definitions (h) effects on welfare. U.S.C. 42: § 7602.
10 U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
11 Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
12 Assessment Office; EPA report nos. EPA-600/8-84-020aF-eF. Available from NTIS, Springfield, VA;
13 PB87-142949.
14 U.S. Environmental Protection Agency. (1992) Summary of selected new information on effects of ozone on health
15 and vegetation: supplement to 1986 air quality criteria for ozone and other photochemical oxidants. Research
16 Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment
17 Office; EPA report no. EPA/600/8-88/105F. Available from NTIS, Springfield, VA; PB92-235670.
18 U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants.
19 Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF. 3v.
20 Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, and PB96-185608. Available online at:
21 www.epa.gov/ncea/ozone.htm.
22
23
January 2005 1-14 DRAFT-DO NOT QUOTE OR CITE
-------
i 2. PHYSICS AND CHEMISTRY OF OZONE
2 IN THE ATMOSPHERE
3
4 2.1 INTRODUCTION
5 Ozone (O3) and other oxidants, such as peroxacyl nitrates and hydrogen peroxide (H2O2)
6 form in polluted areas mainly by atmospheric reactions involving two classes of precursor
7 pollutants, volatile organic compounds (VOCs) and nitrogen oxides (NOX). Ozone is thus a
8 secondary pollutant. The formation of O3 other oxidants and oxidation products from these
9 precursors is a complex, nonlinear function of many factors: the intensity and spectral
10 distribution of sunlight; atmospheric mixing and processing on cloud and aerosol particles; the
11 concentrations of the precursors in ambient air; and the rates of chemical reactions of the
12 precursors. Information contained in this chapter and in greater detail in Annex AX2 describes
13 these processes, numerical models that incorporate these processes to calculate O3
14 concentrations, and techniques for measuring concentrations of ambient oxidants.
15 The atmosphere can be divided into several distinct vertical layers, based primarily on the
16 major mechanisms by which they are heated and cooled. The lowest major layer is the
17 troposphere, which extends from the earth's surface to about 8 km above polar regions and to
18 about 16 km above tropical regions. The planetary boundary layer (PEL) is the lower sub-layer
19 of the troposphere, extending from the surface to about 1 or 2 km and is most strongly affected
20 by surface conditions. The stratosphere extends from the tropopause, or the top of the
21 troposphere, to about 50 km in altitude (Annex AX2.2.1). The emphasis in this chapter is placed
22 on chemical and physical processes occurring in the troposphere, in particular in the PEL.
23
24
25 2.2 CHEMICAL PROCESSES INVOLVED IN OZONE FORMATION
26 AND DESTRUCTION
27 Ozone occurs not only in polluted urban atmospheres but also throughout the troposphere,
28 even in remote areas of the globe. The same basic processes, involving sunlight driven reactions
29 of NOX and VOCs contribute to O3 formation throughout the troposphere. These processes lead
30 also to the formation of other photochemical products, such as peroxyacetyl nitrate (PAN), nitric
January 2005 2-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 acid (HNO3), and sulfuric acid (H2SO4), and to other compounds, such as formaldehyde (HCHO)
2 and other carbonyl compounds, such as aldehydes and ketones.
3 The photochemical formation of O3 in the troposphere proceeds through the oxidation of
4 nitric oxide (NO) to nitrogen dioxide (NO2) by organic (RO2) or hydro-peroxy (HO2) radicals.
5 The photolysis of NO2 yields nitric oxide (NO) and a ground-state oxygen atom, O(3P), which
6 then reacts with molecular oxygen to form O3. The free radicals oxidizing NO to NO2 are
7 formed during the oxidation of VOCs (Annex AX2.2.2).
8 The term VOC refers to all carbon-containing gas-phase compounds in the atmosphere,
9 both biogenic and anthropogenic in origin, excluding carbon monoxide (CO) and carbon dioxide
10 (CO2). Classes of compounds important for the photochemical formation of O3 include alkanes,
11 alkenes, aromatic hydrocarbons, carbonyl compounds (e.g., aldehydes and ketones), alcohols,
12 organic peroxides, and halogenated organic compounds (e.g., alkyl halides). This array of
13 compounds encompasses a wide range of chemical properties and lifetimes: isoprene has an
14 atmospheric lifetime of approximately an hour, whereas methane has an atmospheric lifetime of
15 about a decade.
16 In urban areas, compounds representing all classes of VOCs are important for O3
17 formation. In nonurban vegetated areas, biogenic VOCs emitted from vegetation tend to be the
18 most important. In the remote troposphere, CH4 and CO are the main carbon-containing
19 precursors to O3 formation. CO also can play an important role in O3 formation in urban areas.
20 The oxidation of VOCs is initiated mainly by reaction with hydroxyl (OH) radicals. The primary
21 source of OH radicals in the atmosphere is the reaction of electronically excited O atoms, O(JD),
22 with water vapor. O(JD) is produced by the photolysis of O3 in the Hartley bands. In polluted
23 areas, the photolysis of aldehydes (e.g., HCHO), nitrous acid (HONO) and hydrogen peroxide
24 (H2O2) can also be significant sources of OH or HO2 radicals that can rapidly be converted to OH
25 (Eisele et al., 1997). Ozone can oxidize alkenes; and at night, when they are most abundant,
26 NO3 radicals also oxidize alkenes. In coastal environments and other selected environments,
27 atomic Cl and Br radicals can also initiate the oxidation of VOCs (Annex AX2.2.3).
28 There are a large number of oxidized nitrogen containing compounds in the atmosphere
29 including NO, NO2, NO3, HNO2, HNO3, N2O5, HNO4, PAN and its homologues, other organic
30 nitrates and particulate nitrate. Collectively these species are referred to as NOy. Oxidized
31 nitrogen compounds are emitted to the atmosphere mainly as NO which rapidly interconverts
January 2005 2-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 with NO2 and so NO and NO2 are often "lumped" together into their own group or family, or
2 NOX. NOX can be oxidized to reservoir and termination species (PAN and its homologues,
3 organic nitrates, HNO3, HNO4 and particulate nitrate). These reservoir and termination species
4 are referred to as NOZ. The major reactions involving inter-conversions of oxidized nitrogen
5 species are discussed in Annex AX2.2.4.
6 The photochemical cycles by which the oxidation of hydrocarbons leads to O3 production
7 are best understood by considering the oxidation of methane, structurally the simplest VOC.
8 The CH4 oxidation cycle serves as a model for the chemistry of the relatively clean or unpolluted
9 troposphere (although this is a simplification because vegetation releases large quantities of
10 complex VOCs, such as isoprene, into the atmosphere). In the polluted atmosphere, the
11 underlying chemical principles are the same, as discussed in Annex AX2.2.5. The conversion of
12 NO to NO2 occurring with the oxidation of VOCs is accompanied by the production of O3 and
13 the efficient regeneration of the OH radical, which in turn can react with other VOCs.
14 A schematic overview showing the major processes involved in O3 production and loss in the
15 troposphere and stratosphere is given in Figure 2-1.
16 The oxidation of alkanes and alkenes in the atmosphere has been treated in depth in CD96
17 and is updated in Annexes AX2.2.6 and AX2.2.7. In contrast to simple hydrocarbons containing
18 one or two carbon atoms, detailed kinetic information about the gas phase oxidation pathways of
19 many anthropogenic hydrocarbons (e.g., aromatic compounds, such as benzene and toluene),
20 biogenic hydrocarbons (e.g., isoprene, the monoterpenes), and their intermediate oxidation
21 products (e.g., epoxides, nitrates, and carbonyl compounds) is lacking. Reaction with OH
22 radicals represents the major loss process for alkanes. Reaction with chlorine atoms is an
23 additional sink for alkanes. Stable products of alkane photooxidation are known to include
24 carbonyl compounds, alkyl nitrates, and J-hydroxycarbonyls. Major uncertainties in the
25 atmospheric chemistry of the alkanes concern the chemistry of alkyl nitrate formation; these
26 uncertainties affect the amount of NO-to-NO2 conversion occurring and, hence, the amounts of
27 O3 formed during photochemical degradation of the alkanes.
28 The reaction of OH radicals with aldehydes produced during the oxidation of alkanes
29 forms acyl (R'CO) radicals, and acyl peroxy radicals (R'C(O)-O2) are formed by the further
30 addition of O2. As an example, the oxidation of ethane (C2H5-H) yields acetaldehyde
31 (CH3- CHO). The reaction of CH3- CHO with OH radicals yields acetyl radicals (CH3- CO).
January 2005 2-3 DRAFT-DO NOT QUOTE OR CITE
-------
Non-Polar
Mj^wm. Regions
Emissions Other Organic
ofNOxand Products
VOCs
Figure 2-1. Schematic overview of O3 photochemistry in the stratosphere
and troposphere.
1 The acetyl radicals will then participate with O2 in a termolecular recombination reaction to form
2 acetyl peroxy radicals, which can then react with NO to form CH3 + CO2 (or they can react with
3 NO2 to form PAN). PAN acts as a temporary reservoir for NO2. Upon the thermal
4 decomposition of PAN, either locally or elsewhere, NO2 is released to participate in the O3
5 formation process again.
6 Alkenes react in ambient air with OH, NO3, and Cl radicals and with O3. All of these
7 reactions are important atmospheric transformation processes, and all proceed by initial addition
January 2005
2-4
DRAFT-DO NOT QUOTE OR CITE
-------
1 to the >C=C< bonds. Products of alkene photooxidation include carbonyl compounds,
2 hydroxynitrates and nitratocarbonyls, and decomposition products from the energy-rich
3 biradicals formed in alkene-O3 reactions. Major uncertainties in the atmospheric chemistry of
4 the alkenes concern the products and mechanisms of their reactions with O3, especially the yields
5 of free radicals that participate in O3 formation. Examples of oxidation mechanisms of complex
6 alkanes and alkenes can be found in comprehensive texts such as Seinfeld and Pandis (1998).
7 The oxidation of aromatic hydrocarbons constitutes an important component of the
8 chemistry of O3 formation in urban atmospheres (Annex AX2.2.8). Virtually all of the important
9 aromatic hydrocarbon precursors emitted in urban atmospheres are lost through reaction with the
10 hydroxyl radical. Loss rates for these compounds vary from slow (i.e., benzene) to moderate
11 (e.g., toluene), to very rapid (e.g., xylene and trimethylbenzene isomers). These loss rates are
12 very well understood at room temperature and atmospheric pressure and numerous experiments
13 have been conducted that verify this. However, the mechanism for the reaction of OH with
14 aromatic hydrocarbons is poorly understood as evident from the poor mass balance of the
15 reaction products. The mechanism for the oxidation of toluene has been studied most thoroughly
16 and there is general agreement on the initial steps in the mechanism. However, at present there
17 is no promising approach for resolving the remaining issues concerning the later steps. The
18 oxidation of aromatic hydrocarbons also leads to particle formation which could remove gas-
19 phase constituents that participate in O3 formation. The chemistry of secondary organic aerosol
20 formation from gaseous precursors was summarized in the latest AQCD for particulate matter.
21 The reactions of oxygenated VOCs are also important components of O3 formation (Annex
22 AX2.2.9). They may be produced either by the oxidation of hydrocarbons or they may be
23 present in ambient air as the result of direct emissions. For example, motor vehicles and some
24 industrial processes emit formaldehyde and vegetation emits methanol.
25 As much as 30% of the carbon in hydrocarbons in many urban areas is in the form of
26 aromatic compounds. Yet, mass balance analyses performed on irradiated smog chamber
27 mixtures of aromatic hydrocarbons indicate that only about one-half of the carbon is in the form
28 of compounds that can be identified. The situation is not much better for some smaller
29 anthropogenic hydrocarbons. For example, only about 60% of the initial carbon can be
30 accounted for in the OH initiated oxidation of 1,3-butadiene. About two-thirds of the initial
31 carbon can be identified in product analyses of isoprene oxidation. Adequate analytical
January 2005 2-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 techniques needed to identify and quantify key intermediate species are not available for many
2 compounds. In addition, methods to synthesize many of the suspected intermediate compounds
3 are not available so that laboratory studies of their reaction kinetics cannot be performed.
4 Similar considerations apply to the oxidation of biogenic hydrocarbons such as isoprene.
5 In addition to reactions occurring in the gas phase, reactions occurring on the surfaces of or
6 within cloud droplets and airborne particles also occur. Their collective surface area is huge
7 implying that collisions with gas phase species occur on very short time scales. In addition to
8 hydrometeors (e.g., cloud and fog droplets and snow and ice crystals) there are also potential
9 reactions involving atmospheric particles of varying composition (e.g., wet [deliquesced]
10 inorganic particles, mineral dust, carbon chain agglomerates and organic carbon particles) to
11 consider. Most of the well-established multiphase reactions tend to reduce the rate of O3
12 formation in the polluted troposphere. Removal of HOX and NOX onto hydrated particles will
13 reduce the production of O3. The reactions of Br and Cl containing radicals deplete O3 in
14 selected environments such as the Arctic during spring, the tropical marine boundary layer and
15 inland salt lakes. Direct reactions of O3 and atmospheric particles appear to be too slow to
16 reduce O3 formation significantly at typical ambient PM levels. In addition, the oxidation of
17 hydrocarbons by Cl radicals could lead to the rapid formation of peroxy radicals and higher rates
18 of O3 production in selected coastal environments. It should be stressed that knowledge of
19 multiphase processes is still evolving and there are still many questions that remain to be
20 answered as outlined in Annex AX2.2.10.
21
22
23 2.3 METEOROLOGICAL PROCESSES AFFECTING OZONE
24 Since CD96, substantial new information about transport processes has become available
25 from numerical models, field experiments and satellite-based observations. Ozone is produced
26 naturally by photochemical reactions in the stratosphere as shown in Figure 2-1. Some of this O3
27 is transported downward into the troposphere throughout the year, with maximum contributions
28 during late winter and early spring mainly in a process known as tropopause folding. Figure
29 2-2a shows a synoptic situation associated with a tropopause folding event. A vertical cross
30 section taken through the atmosphere from a to a' is shown in Figure 2-2b. In this figure the
31 tropopause fold is shown folding downward above and slightly behind the surface cold front,
January 2005 2-6 DRAFT-DO NOT QUOTE OR CITE
-------
12UTC, 23Feb94
103.6kPa
Figure 2-2a. Surface weather chart showing sea level (MSL) pressure (kPa), and
surface fronts.
Source: Stall (2000).
U)
'53
a CYYC
LBF
LCH a'
Figure 2-2b. Vertical cross section along dashed line (a-a') from northwest to the
southeast (CYYC = Calgary, Alberta; LBF = North Platte, NB; LCH = Lake
Charles, LA). The approximate location of the jet stream core is indicated
by the hatched area. The position of the surface front is indicated by the
cold-frontal symbols and the frontal inversion top by the dashed line.
Note: This is 12 h later than the situations shown in Figure 2-2a.
Source: Adapted from Stull (2000).
January 2005
2-7
DRAFT-DO NOT QUOTE OR CITE
-------
1 bringing stratospheric air with it. Although the tropopause is drawn with a solid line, it should
2 not be taken to mean that it is a material surface, through which there is no exchange. Rather
3 these folds should be though of as regions in which mixing of tropospheric and stratospheric air
4 is occurring (Shapiro, 1980). This imported stratospheric air contributes to the natural
5 background of O3 in the troposphere especially in the free troposphere. It should be noted that
6 there is considerable uncertainty in the magnitude and distribution of this potentially important
7 source of tropospheric O3. Stratospheric intrusions that reach the surface are rare. Much more
8 common are intrusions which penetrate only to the middle and upper troposphere. However, O3
9 transported to the upper and middle troposphere can still affect surface concentrations through
10 various exchange mechanisms that mix air from the free troposphere with air in the planetary
11 boundary layer. Substantial photochemical production of O3 in the troposphere also begins in
12 late winter and early spring, therefore it cannot be assumed that O3 present at these times is only
13 stratospheric in origin. The basic atmospheric dynamics and thermodynamics of stratospheric-
14 tropospheric exchange are outlined in Annex AX2.3.1.
15 Our understanding of the meterological processes associated with summertime O3 episodes
16 remains basically the same as outlined in CD96. Major episodes of high O3 concentrations in the
17 eastern United States and in Europe are associated with slow moving, high pressure systems.
18 High pressure systems during the warmer seasons are associated with the sinking of air, resulting
19 in warm, generally cloudless skies, with light winds. The sinking of air results in the
20 development of stable conditions near the surface which inhibit or reduce the vertical mixing of
21 O3 precursors. The combination of inhibited vertical mixing and light winds minimizes the
22 dispersal of pollutants emitted in urban areas, allowing their concentrations to build up.
23 Photochemical activity involving these precursors is enhanced because of higher temperatures
24 and the availability of sunlight. In the eastern United States, high O3 concentrations during a
25 large scale episode can extend over hundreds of thousands square kilometers for several days.
26 These conditions have been described in greater detail in CD96. The transport of pollutants
27 downwind of major urban centers is characterized by the development of urban plumes.
28 However, the presence of mountain barriers limits mixing as in Los Angeles and Mexico City
29 and will result in even longer periods and a higher frequency of days with high O3
30 concentrations. Ozone concentrations in southern urban areas, such as Houston, TX and Atlanta,
31 GA tend to follow this pattern and they tend to decrease with increasing wind speed. In northern
January 2005 2-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 cities such as Chicago, IL; New York, NY; Boston, MA; and Portland, ME, the average O3
2 concentrations over the metropolitan areas increase with wind speed indicating that transport of
3 O3 and its precursors from upwind areas is important (Husar and Renard, 1998; Schichtel and
4 Husar, 2001).
5 Ozone and other secondary pollutants are determined by meteorological and chemical
6 processes extending typically over spatial scales of several hundred kilometers (e.g., Civerolo
7 et al., 2003; Rao et al., 2003). An analysis of the output of regional model studies conducted by
8 Kasibhatla and Chameides (2000) suggests that O3 can be transported over a few thousand
9 kilometers in the upper boundary layer of the eastern half of the United States during specific O3
10 episodes. Convection is capable of transporting O3 and its precursors vertically through the
11 troposphere as shown in Annex AX2.3.2. Nocturnal low level jets (LLJs) can also transport
12 pollutants hundreds of kilometers (Annex AX2.3.3). Schematic diagrams showing the
13 atmospheric conditions during the formation of low level jets and the regions in which they are
14 most prevalent are given in Figures 2-3 and 2-4. They have also been observed off the coast of
15 California. Turbulence associated with LLJs can bring these pollutants to the surface and result
16 in secondary O3 maxima in the early morning in many locations (Corsmeier et al., 1997).
17 Aircraft observations indicate that there can be substantial differences in mixing ratios of
18 key species between the surface and the atmosphere above (Fehsenfeld et al., 1996; Berkowitz
19 and Shaw, 1997). In particular, mixing ratios of O3 can be higher in the lower free troposphere
20 (aloft) than in the planetary boundary layer during multiday O3 episodes (Taubmann et al.,
21 2004). These conditions are illustrated schematically in Figure 2-5. Convective processes and
22 small scale turbulence transport O3 and other pollutants both upward and downward throughout
23 the planetary boundary layer and the free troposphere. Ozone and its precursors can be
24 transported vertically by convection into the upper part of the mixed layer on one day, then
25 transported overnight as a layer of elevated mixing ratios and then entrained into a growing
26 convective boundary layer downwind and brought back down to the surface. High
27 concentrations of O3 showing large diurnal variations at the surface in southern New England
28 were associated with the presence of such layers (Berkowitz et al., 1998). Zhang et al. (1997)
29 estimated that as much as 60% to 70% of O3 observed at the surface during one case study in
30 eastern Tennessee could have come from the upper boundary layer. Because of wind shear,
31 winds several hundred meters above the ground can bring pollutants from the west, even though
January 2005 2-9 DRAFT-DO NOT QUOTE OR CITE
-------
2-
•- 1-1
-------
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 2-5. 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 km. 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 relative humidity and the Angstrom exponent for aerosol scattering (a)
are higher. Larger, internally 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).
1 surface winds are from the southwest during periods of high O3 in the eastern United States
2 (Blumenthal et al., 1997). These considerations suggest that in many areas of the Unites States,
3 O3 formation involves processes occurring over hundreds if not thousands of square kilometers.
4 Although the vast majority of measurements are made near the Earth's surface, there is
5 substantial photochemistry and transport of O3 occurring above the boundary layer in the free
6 troposphere. In the free troposphere, pollutants are chemically more stable and can be
7 transported over much longer distances by westerly winds and O3 is produced more efficiently
January 2005
2-11
DRAFT-DO NOT QUOTE OR CITE
-------
1 than in the planetary boundary layer. Results from the Atmosphere/Ocean Ocean Chemistry
2 Experiment (AEROCE) indicated that springtime maxima in surface O3 over the western North
3 Atlantic Ocean result from tropopause folding in close proximity to convective clouds (Annex
4 AX2.3.4). The convection lifts O3 and its precursors to the free troposphere where they mix with
5 O3 from the stratosphere and the mixture is transported eastward. Results from the North
6 Atlantic Regional Experiment (Annex AX2.3.4) indicated that summertime air is transported
7 along the East Coast northeastward and upward ahead of cold fronts. New England and the
8 Maritime Provinces of Canada receive substantial amounts of O3 and other pollutants through
9 this mechanism. Pollutants transported in this way can then be entrained in stronger and more
10 stable westerly winds aloft and can travel long distances across the North Atlantic Ocean. The
11 pollutants can then be brought to the surface by subsidence in high pressure systems (typically
12 behind the cold front in advance of the one mentioned above). Thus, pollutants from North
13 America can be brought down either over the North Atlantic Ocean or in Europe. Pollutants can
14 be transported across the North Pacific Ocean from Asia to North America in a similar way.
15 Behind an advancing cold front, cold and dry stratospheric air is also being transported
16 downward and southward. Stratospheric constituents and tropospheric constituents can then mix
17 by small-scale turbulent exchange processes. The results of these studies suggest that the
18 mechanisms involved in the long-range transport of O3 and its precursors are closely tied to the
19 processes involved in stratospheric-tropospheric exchange.
20
21
22 2.4 RELATIONS OF OZONE TO ITS PRECURSORS
23 The local rate of O3 formation depends on atmospheric conditions such as the availability
24 of solar ultraviolet radiation capable of initiating photolysis reactions, air temperatures and the
25 concentrations of chemical precursors (Annex AX2.3.5). The dependence of daily 1-h
26 maximum surface O3 concentrations on daily maximum temperature is illustrated in Figure 2-6
27 for New York City. The notable trend is the apparent upper bound to O3 concentrations as a
28 function of temperature. However, this trend is absent in data from Phoenix, AZ (Figure 2-7).
29 At any given temperature, there is a wide range of O3 concentrations possible because other
30 factors (e.g., cloudiness, mixing height, wind speed) can also influence O3 production rates.
31 It should be noted, however, that the data shown in Figures 2-6 and 2-7 were obtained during the
January 2005 2-12 DRAFT-DO NOT QUOTE OR CITE
-------
240
-Q
a 180
O
_>,
Q
E
120-
.1 60
X
CO
1 1
1 1
1 1
New York, NY
L -L J
°°0
0
L
-------
1 1980s and different relations might be obtained with current data. Relations of O3 to precursor
2 variables are seen to be location specific and relations observed in one area can not be readily
3 extrapolated to another.
4 Rather than varying directly with emissions of its precursors, O3 changes in a nonlinear
5 fashion with the concentrations of its precursors (Annex AX2.4). At the low NOX concentrations
6 found in most environments, ranging from remote continental areas to rural and suburban areas
7 downwind of urban centers (low - NOX regime), the net production of O3 increases with
8 increasing NOX. At the high NOX concentrations found in downtown metropolitan areas,
9 especially near busy streets and roadways, and in power plant plumes there is net destruction
10 (titration) of O3 by reaction with NO (high - NOX regime). In between these two regimes there is
11 a transition stage in which O3 shows only a weak dependence on NOX concentrations. In the
12 high - NOX regime, NO2 scavenges OH radicals which would otherwise oxidize VOCs to
13 produce peroxy radicals, which in turn would oxidize NO to NO2. In this regime, O3 production
14 is limited by the availability of free radicals. The production of free radicals is in turn limited by
15 the availability of solar UV radiation capable of photolyzing O3 in the Hartley bands or
16 aldehydes; and/ or by the abundance of VOCs whose oxidation would produce more radicals
17 than they consume. In the low-NOx regime, the oxidation of VOCs generates (or at least does
18 not consume) free radicals, and O3 production varies directly with NOX. There are a number of
19 ways to refer to the chemistry in these two chemical regimes. Sometimes the terms VOC-
20 limited and NOx-limited are used to describe these two regimes. However, there are difficulties
21 with this usage because (1) VOC measurements are not as abundant as they are for nitrogen
22 oxides, (2) rate coefficients for reaction of individual VOCs with free radicals vary over an
23 extremely wide range, and (3) consideration is not given to CO nor to reactions that can produce
24 free radicals without invoking VOCs. The terms NOx-limited and NOx-saturated (e.g., Jaegle
25 et al., 2001) will be used wherever possible to more adequately describe these two regimes.
26 However, the terminology used in original articles will also be used here.
27 The chemistry of OH radicals, which are responsible for initiating the oxidation of
28 hydrocarbons, shows behavior similar to that for O3 with respect to NOX concentrations (Hameed
29 et al., 1979; Pinto et al., 1993; Poppe et al., 1993; Zimmerman and Poppe, 1993). These
30 considerations introduce a high degree of uncertainty into attempts to relate changes in O3
31 concentrations to emissions of precursors. There are no definitive rules governing the levels of
January 2005 2-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 NOX at which the transition from NOx-limited to NOx-saturated conditions occurs. The transition
2 between these two regimes is highly spatially and temporally dependent and depends also on the
3 nature and abundance of the hydrocarbons that are present.
4 Trainer et al. (1993) and Olszyna et al. (1994) have shown that O3 and NOy are highly
5 correlated in rural areas in the eastern United States. Trainer et al. (1993) also showed that O3
6 levels correlate even better with NOZ than with NOy, as may be expected because NOZ represents
7 the amount of NOX that has been oxidized, forming O3 in the process. NOZ is equal to the
8 difference between measured total reactive nitrogen, NOy, and NOX and represents the summed
9 products of the oxidation of NOX. NOZ is composed mainly of HNO3, PAN and other organic
10 nitrates, particulate nitrate, and HNO4.
11 Trainer et al. (1993) also suggested that the slope of the regression line between O3 and
12 NOZ can be used to estimate the rate of O3 production per NOX oxidized (also known as the O3
13 production efficiency, or OPE). Ryerson et al. (1998, 2001) used measured correlations between
14 O3 and NOZ to identify different rates of O3 production in plumes from large point sources.
15 A number of studies in the planetary boundary layer over the continental United States have
16 found that the OPE ranges typically from one to nearly ten. However, it may be higher in the
17 upper troposhere and in certain areas, such as the Houston-Galveston area. Observations
18 indicate that the OPE depends mainly on the abundance of NOX.
19 Various techniques have been proposed to use ambient NOX and VOC measurements to
20 derive information about the dependence of O3 production on their concentrations. For example,
21 it has been suggested that O3 formation in individual urban areas could be understood in terms of
22 measurements of ambient NOX and VOC concentrations during the early morning (e.g., National
23 Research Council, 1991). In this approach, the ratio of summed (unweighted) VOC to NOX is
24 used to determine whether conditions were NOx-limited or VOC limited. This procedure is
25 inadequate because it omits many factors that are important for O3 production such as the impact
26 of biogenic VOCs (which are typically not present in urban centers during early morning);
27 important differences in the ability of individual VOCs to generate free radicals (rather than just
28 total VOC) and other differences in O3 forming potential for individual VOCs (Carter et al.,
29 1995); and changes in the VOC to NOX ratio due to photochemical reactions and deposition as
30 air moves downwind from urban areas (Milford et al., 1994).
January 2005 2-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 Photochemical production of O3 generally occurs simultaneously with the production of
2 various other species such as nitric acid (HNO3), organic nitrates, and hydrogen peroxide. The
3 relative rate of production of O3 and other species varies depending on photochemical
4 conditions, and can be used to provide information about O3-precursor sensitivity. Sillman
5 (1995) and Sillman and He (2002) identified several secondary reaction products that show
6 different correlation patterns for NOx-limited and NOx-saturated conditions. The most important
7 correlations are for O3 versus NOy, O3 versus NOZ, O3 versus HNO3, and H2O2 versus HNO3.
8 The correlations between O3 and NOy, and O3 and NOZ are especially important because
9 measurements of NOy and NOX are more widely available than for VOCs. Measured O3 versus
10 NOZ (Figure 2-8) shows distinctly different patterns in different locations. In rural areas and
11 in urban areas such as Nashville, TN, O3is highly correlated with NOZ. By contrast, in
12 Los Angeles, CA, O3 is not as highly correlated with NOZ, and the rate of increase of O3 with
13 NOZ is lower and the O3 concentrations for a given NOZ value are generally lower. The different
14 O3 versus NOZ relations in Nashville, TN and Los Angeles, CA reflects the difference between
15 NOx-limited conditions in Nashville vs. an approach to NOX- saturated conditions in
16 Los Angeles.
NOZ (ppb)
Figure 2-8. Measured values of O3 and NOZ (NOy - NOX) during the afternoon at rural
sites in the eastern United States (grey circles) and in urban areas and urban
plumes associated with Nashville, TN (gray dashes); Paris, France (black
diamonds); and Los, Angeles CA (Xs).
Sources: Trainer etal. (1993), Sillman etal. (1997, 1998), Sillman and He (2002).
January 2005
2-16
DRAFT-DO NOT QUOTE OR CITE
-------
1 The difference between NOx-limited and NOx-saturated regimes is also reflected in
2 measurements of hydrogen peroxide (H2O2). Hydrogen peroxide production is highly sensitive
3 to the abundance of free radicals and is thus favored in the NOx-limited regime, typical of
4 summer conditions. Differences between these two regimes are also related to the preferential
5 formation of sulfate during summer and to the inhibition of sulfate and hydrogen peroxide
6 formation during winter (Stein and Lamb, 2003). Measurements in the rural eastern United
7 States (Jacob et al., 1995) Nashville, TN (Sillman et al., 1998), and Los Angeles, CA (Sakugawa
8 and Kaplan, 1989), show large differences in H2O2 concentrations between likely NOx-limited
9 and radical-limited locations.
10
11
12 2.5 THE ROLE OF CHEMISTRY-TRANSPORT MODELS IN
13 UNDERSTANDING ATMOSPHERIC OZONE
14 Chemistry-transport models (CTMs) are used to improve understanding of atmospheric
15 chemical processes and to develop control strategies (Annex AX2.5). The main components of a
16 CTM are summarized in Figure 2-9. Models such as the CMAQ (Community Model for Air
17 Quality) system incorporate the processes shown in Figure 2-9 as numerical algorithms to
18 predict time dependent concentration fields of a wide variety of gaseous and paniculate phase
19 pollutants. Also shown in Figure 2-9 is the meteorological model used to provide the inputs for
20 calculating the transport of species in the CTM. The meteorological models such as the MM5
21 model, which supplies these inputs to the CTMs mentioned above, also provide daily weather
22 forecasts. The domains of these models extend typically over areas of millions of square
23 kilometers. Because these models are computationally intensive, it is often impractical to run
24 them over larger domains without sacrificing some features. For these reasons, both the
25 meteorological model and the CTM rely on boundary conditions that allow processes occurring
26 outside the model domain to influence their predictions. The entire system of meteorological
27 model emissions processors and output processors shown in Figure 2-9 constitutes
28 the framework of EPA's Models-3.
29 Because of the large number of chemical species and reactions that are involved in
30 considering the oxidation of realistic mixtures of anthropogenic and biogenic hydrocarbons,
31 condensed mechanisms must be used in atmospheric models. These mechanisms are tested by
January 2005 2-17 DRAFT-DO NOT QUOTE OR CITE
-------
Initial/Boundary
Conditions and
Continuous Updates
of Met. Fields
from Observations
Meteorological
Model
Emissions
Model
Anthropogenic
(point, area sources)
Biogenic Emissions
X
' \
1
Aerosol
Chemistry
and
Microphysics
Chemistry Transport Model
Visualization of Output
Process Analyses
Figure 2-9. Main components of a comprehensive atmospheric chemistry modeling
system, such as Models-3.
1 comparison with smog chamber data. However, the existing chemical mechanisms neglect
2 many important processes such as the formation and subsequent reactions of long lived carbonyl
3 compounds, the incorporation of the most recent information about intermediate compounds, and
4 heterogeneous reactions involving cloud droplets and aerosol particles.
5 Emissions inventories are compiled for O3 precursors ( NOX, VOCs, and CO). Recent
6 estimates and more detailed discussions of the estimates are given in Annex AX2.5.2.
7 Anthropogenic NOX emissions are associated with combustion processes. Most emissions are in
8 the form of NO, which is formed at high combustion temperatures from atmospheric nitrogen
9 and oxygen and from fuel nitrogen. The two largest sources of NOX are electric power
10 generation plants and motor vehicles. Emissions of NOX therefore are highest in areas having a
January 2005
2-18
DRAFT-DO NOT QUOTE OR CITE
-------
1 high density of power plants and in urban regions having high traffic density. Natural NOX
2 sources include stratospheric intrusions, lightning, soils, and wildfires. Lightning, fertilized
3 soils, and wildfires are the major natural sources of NOX in the United States. Both nitrifying
4 and denitrifying organisms in the soil can produce NOX, mainly in the form of NO. Emission
5 rates depend mainly on fertilization levels and soil temperature and moisture. Spatial and
6 temporal variability in soil NOX emissions leads to considerable uncertainty in emissions
7 estimates. About 60% of lightning generated NOX occurs in the southern United States and
8 about 60% the total NOX emitted by soils occurs in the central corn belt of the United States.
9 The oxidation of NH3 emitted mainly by livestock and soils, leads to the formation of a small
10 amount of NO. Uncertainties in natural NOX inventories are much larger than for anthropogenic
11 NOX emissions.
12 Hundreds of VOCs, containing mainly two to about twelve carbon atoms, are emitted by
13 evaporation and combustion processes from a large number of anthropogenic sources. The two
14 largest source categories in the U.S. EPA's emissions inventories are industrial processes and
15 transportation. Emissions of VOCs from highway vehicles account for almost 75% of the
16 transportation-related emissions.
17 The accuracy of VOC emission estimates is difficult to determine, both for stationary and
18 mobile sources. Evaporative emissions, which depend on temperature and other environmental
19 factors, compound the difficulties of assigning accurate emission factors. In assigning VOC
20 emission estimates to the mobile source category, models are used that incorporate numerous
21 input parameters (e.g., type of fuel used, type of emission controls, age of vehicle), each of
22 which has some degree of uncertainty. Data for the ratio of CO to NOX and NMOC to NOX in
23 traffic tunnels (e.g., Pierson et al., 1990) indicated that emissions of NMHCs and CO from motor
24 vehicles have been underestimated by as much as a factor of two (based on the assumption that
25 emissions of NOX were reasonably well represented in the inventories). However, the results of
26 more recent studies have been mixed, with many studies showing agreement to within ±50%
27 (summarized in Air Quality Criteria for Carbon Monoxide [U.S. Environmental Protection
28 Agency, 2000]). Remote sensing data (Stedman et al., 1991) indicate that about 50% of NMHC
29 and CO emissions are produced by about 10% of the vehicles. These "super-emitters" are
30 typically poorly maintained vehicles. Vehicles of any age engaged in off-cycle operations (e.g.,
31 rapid accelerations) emit much more than if operated in normal driving modes.
January 2005 2-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 Vegetation emits significant quantities of VOCs such as terpenoid compounds (isoprene,
2 -2-methyl-3-buten-2-ol, monoterpenes), compounds in the hexanal family, alkenes, aldehydes,
3 organic acids, alcohols, ketones, and alkanes. The major species emitted by plants are isoprene
4 (35%), 19 other terpenoid compounds and 17 non-terpenoid compounds including oxygenated
5 compounds (40%) (Guenther et al., 2000). Coniferous forests represent the largest source on a
6 nationwide basis, because of their extensive land coverage. Most biogenic emissions occur
7 during the summer, because of their dependence on temperature and incident sunlight. Biogenic
8 emissions are also higher in southern states than in northern states for these reasons and because
9 of species variations. The uncertainty in natural emissions is about 50% for isoprene under
10 midday summer conditions and could be as much as a factor often higher for some compounds
11 (Guenther et al., 2000). Uncertainties in both biogenic and anthropogenic VOC emission
12 inventories prevent determination of the relative contributions of these two categories at least in
13 many urban areas. On the regional and global scales, emissions of VOCs from vegetation are
14 much larger than those from anthropogenic sources.
15 The performance of CTMs must be evaluated by comparison with field data as part of a
16 cycle of model improvements and subsequent evaluations. Discrepancies between model
17 predictions and observations can be used to point out gaps in current understanding of
18 atmospheric chemistry and to spur improvements in parameterizations of atmospheric chemical
19 and physical processes. Model evaluation does not merely involve a straightforward comparison
20 between model predictions and the concentration field of the pollutant of interest. Such
21 comparisons may not be meaningful because it is difficult to determine if agreement between
22 model predictions and observations truly represents an accurate treatment of physical and
23 chemical processes in the CTM or the effects of compensating errors in complex model routines.
24 Ideally, each of the model components (emissions inventories, chemical mechanism,
25 meteorological driver) should be evaluated individually, however this is rarely done in practice.
26 A comparison between free radical concentrations predicted by parameterized chemical
27 mechanisms and observations, suggests that radical concentrations were overestimated by
28 current chemical mechanisms for NOX concentrations < ~5 ppb for that set of environmental
29 conditions (Volz-Thomas et al., 2003).
30 In addition to comparisons between concentrations of calculated and measured species,
31 comparisons of correlations between measured primary VOCs and NOX and modeled VOCs and
January 2005 2-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 NOX are especially useful for evaluating results from chemistry-transport models. Likewise,
2 comparisons of correlations between measured species and modeled species can be used to
3 provide information about the chemical state of the atmosphere and to evaluate model
4 representations (including O3 production per NOX, O3-NOX-VOC sensitivity, and the general
5 accuracy of photochemical representations). A CTM that demonstrates the accuracy of both its
6 computed VOC and NOX in comparison with ambient measurements and the spatial and temporal
7 relations among the critical secondary species associated with O3 has a higher probability of
8 representing O3-precursor relations correctly than one that does not.
9
10
11 2.6 TECHNIQUES FOR MEASURING OZONE AND ITS PRECURSORS
12 Several techniques have been developed for sampling and measurement of O3 in the
13 ambient atmosphere at ground level. Although the chemiluminescence method (CLM) using
14 ethylene is designated as the Federal Reference Method for measuring O3, monitoring in the
15 NAMS/SLAMS networks is conducted mainly with UV absorption spectrometry using
16 commercial short path instruments. The primary reference standard instrument is a relatively
17 long-path UV absorption spectrometer maintained under carefully controlled conditions at NIST
18 (e.g., Fried and Hodgeson, 1982). Episodic measurements are made with a variety of other
19 techniques based on the principles of chemiluminescence, electrochemistry, differential optical
20 absorption spectroscopy (DOAS), and LIDAR.
21 In principle, each of these methods is subject to interference. Kleindienst et al. (1993)
22 found that water vapor could cause a positive interference in the CLM with an average positive
23 deviation of 3% ozone/% water vapor at 25 C. The UV absorption spectrometers are subject to
24 positive interference by atmospheric constituents, such as certain aromatic aldehydes that absorb
25 at the 253.7 nm Hg resonance line and are at least partially removed by the MnO2 scrubber.
26 Parrish and Fehsenfeld (2000) did not find any evidence for significant interference (> 1%) in
27 flights through the Nashville urban plume. They also did not find any significant interference in
28 extensive airborne and ground-based comparisons in the Houston, TX area. However, more
29 extensive data in urban areas are lacking.
30 By far, most measurements of NO are made using the CLM, based on its reaction with O3.
31 Commercial instruments for measuring NO and NO2 are constructed with an internal converter
January 2005 2-21 DRAFT-DO NOT QUOTE OR CITE
-------
1 for reducing NO2 to NO and then measuring NO by the CLM. In principle, this technique yields
2 a measurement of NOX with NO2 found by difference between NOX and NO. However, these
3 converters also reduce NOZ compounds thereby introducing a positive interference in the
4 measurement of NO2. Other methods for measuring NO2 are available, such as photolytic
5 reduction followed by CLM, laser-induced fluorescence and DOAS. However, they require
6 further development before they can be used for routine monitoring in the NAMS/SLAMS
7 networks. More detailed descriptions of the issues and techniques discussed above and
8 techniques for measuring HNO3 and VOCs can be found in Annex AX2.6.
9
10
11 2.7 SUMMARY
12 Ozone (O3) is formed by atmospheric reactions involving two classes of precursor
13 compounds, volatile organic compounds (VOCs) and nitrogen oxides (NOX). Ozone is thus a
14 secondary pollutant. Ozone is ubiquitous throughout the atmosphere, it is present even in remote
15 areas of the globe. The photochemical oxidation of almost all anthropogenic and biogenic VOCs
16 is initiated by reaction with hydroxyl (OH) radicals. At night, when they are most abundant,
17 NO3 radicals also oxidize alkenes. In coastal and other select environments, Cl and Br radicals
18 can also initiate the oxidation of VOCs.
19 In urban areas, basically all classes of VOCs (alkanes, alkenes, aromatic hydrocarbons,
20 carbonyl compounds, etc.) and CO are important for ozone formation. Although knowledge of
21 the oxidative mechanisms of VOCs has improved over the past several years, gaps in knowledge
22 involving key classes, such as aromatic hydrocarbons, still remain. For example, only about half
23 of the carbon initially present in aromatic hydrocarbons in smog chamber studies form
24 compounds that can be identified.
25 In addition to gas phase reactions, reactions also occur on the surfaces of, or within cloud
26 droplets and airborne particles. Most of the well-established multiphase reactions tend to reduce
27 the rate of O3 formation in polluted environments. Reactions of Cl and Br containing radicals
28 deplete O3 in selected environments such as the Arctic during spring, the tropical marine
29 boundary layer and inland salt lakes. Direct reactions of O3 and atmospheric particles appear to
30 be too slow to reduce O3 formation significantly at typical ambient PM levels.
January 2005 2-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 Our basic understanding of the meteorological processes associated with summertime
2 ozone episodes has not changed over the past several years. However, the realization that long-
3 range transport processes are important for determining ozone concentrations at the surface is
4 growing. In addition to synoptic scale flow fields, Nocturnal Low-Level Jets are capable of
5 transporting pollutants hundreds of km from their sources in either the upper boundary layer or
6 the lower free troposphere. Turbulence then brings ozone and other pollutants to the surface.
7 On larger scales, important progress has been made in identifying the mechanisms of
8 intercontinental transport of ozone and other pollutants.
9 Some ozone would be found near the earth's surface as the result of its downward transport
10 from the stratosphere, even in the absence of photochemical reactions in the troposphere.
11 Intrusions of stratospheric ozone that reach the surface are rare. Much more common are
12 intrusions that penetrate to the middle and upper troposphere. However, O3 transported to the
13 middle and upper troposphere can still affect surface concentrations through various mechanisms
14 that mix air between the planetary boundary layer and the free troposphere above.
15 The formation of ozone and associated compounds is a complex, nonlinear function of
16 many factors including the intensity and spectral distribution of sunlight; atmospheric mixing
17 and other atmospheric processes; and the concentrations of the precursors in ambient air. At the
18 low NOX concentrations found in most environments, ranging from remote continental areas to
19 rural and suburban areas downwind of urban centers, the net production of O3 increases with
20 increasing NOX. At the high concentrations found in downtown metropolitan areas, especially
21 near busy streets and highways, and in power plant plumes there is net destruction of O3 by
22 reaction with NO. In between these two regimes there is a transition stage, in which O3
23 production shows only a weak dependence on NOX concentrations. The efficiency of ozone
24 production per NOX oxidized is generally highest in areas where NOX concentrations are lowest
25 and decrease with increasing NOX concentration.
26 Chemistry transport models are used to improve understanding of atmospheric chemical
27 and physical processes as well as to develop air pollution control strategies. The performance of
28 these models must be evaluated by comparison with field data as part of a cycle of model
29 improvements and subsequent evaluations. Discrepancies between model predictions and
30 observations can be used to point out gaps in current understanding and thus to improve
31 parameterizations of atmospheric chemical and physical processes. Model evaluation does not
January 2005 2-23 DRAFT-DO NOT QUOTE OR CITE
-------
1 merely involve a straightforward comparison between model predictions and the concentration
2 fields of a pollutant of interest (e.g., O3). Such comparisons may not be meaningful because it is
3 difficult to determine if agreement between measurements and model predictions truly represents
4 an accurate treatment of physical and chemical processes in the model or the effects of
5 compensating errors in model routines.
6 The main methods in use for routine monitoring of ambient ozone are based on
7 chemiluminescence or UV absorption. Measurements at most ambient monitoring sites are
8 based on UV absorption. Both of these methods are subject to interference by other atmospheric
9 components. A few studies conducted in urban plumes have not found evidence for significant
10 positive interference in the UV absorption technique. However, more extensive data are not
11 available to completely resolve this issue in urban areas.
12
January 2005 2-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2
3 Berkowitz, C. M; Shaw, W. J. (1997) Airborne measurements of boundary layer chemistry during the Southern
4 Oxidant Study: a case study. J. Geophys. Res. [Atmos.] 102: 12,795-12,804.
5 Berkowitz, C. M.; Fast, J. D.; Sprinston, S. R.; Larsen, R. J.; Spicer, C. W.; Doskey, P. V.; Hubbe, J. M.;
6 Plastridge, R. (1998) Formation mechanisms and chemical characteristics of elevated photochemical layers
7 over the northeast United States. J. Geophys. Res. [Atmos.] 103: 10,631-10,647.
8 Blumenthal, D. L.; Lurmann, F. W.; Kumar, N.; Dye, T. S.; Ray, S. E.; Korc, M. E.; Londergan, R.; Moore, G.
9 (1997) Transport and mixing phenomena related to ozone exceedances in the northeast U.S. (analysis based
10 on NARSTO-northeast data). Available: http://capita.wustl.edu/otag/reports/otagrept/otagrept.html
11 (30 October 2003).
12 Bonner, W. D. (1968) Climatology of the low level jet. Mon. Weather Rev. 96: 833-850.
13 Carter, W. P. L. (1995) Computer modeling of environmental chamber studies of maximum incremental reactivities
14 of volatile organic compounds. Atmos. Environ. 29: 2513.
15 Civerolo, K. L.; Mao, H. T.; Rao, S. T. (2003) The airshed for ozone and fine paniculate pollution in the eastern
16 United States. Pure Appl. Geophys. 160: 81-105.
17 Corsmeier, U.; Kalthhoff, N.; Kolle, O.; Motzian, M.; Fiedler, F. (1997) Ozone concentration jump in the stable
18 nocturnal boundary layer during a LLJ-event. Atmos. Environ. 31: 1977-1989.
19 Eisele, F. L.; Mount, G. H.; Tanner, D.; Jefferson, A.; Shelter, R.; Harder, J. W.; Williams, E. J. (1997)
20 Understanding the production and interconversion of the hydroxyl radical during the tropospheric OH
21 photochemistry experiment. J. Geophys. Res. 102: 6457-6465.
22 Fehsenfeld, F. C.; Trainer, M.; Parrish, D. D.; Volz-Thomas, A.; Penkett, S. (1996) North Atlantic Regional
23 Experiment (NARE) 1993 summer intensive: foreword. J. Geophys. Res. [Atmos.] 101: 28,869-28,875.
24 Fried, A.; Hodgeson, J. (1982) Laser photoacoustic detection of nitrogen dioxide in the gas-phase titration of nitric
25 oxide with ozone. Anal. Chem. 54: 278-282.
26 Guenther, A.; Geron, C.; Pierce, T.; Lamb, B.; Harley, P.; Fall, R. (2000) Natural emissions of non-methane volatile
27 organic compounds, carbon monoxide, and oxides of nitrogen from North America. Atmos. Environ.
28 34:2205-2230.
29 Hameed, S.; Pinto, J. P.; Stewart, R. W. (1979) Sensitivity of the predicted CO-OH-CH4 perturbation to tropospheric
30 NO-j concentrations. J. Geophys. Res. C: Oceans Atmos. 84: 763-768.
31 Husar, R. B.; Renard, W. P. (1998) Ozone as a function of local wind speed and direction: Evidence of local and
32 regional transport. Presented at: 91st annual meeting and exhibition of the Air & Waste Management
33 Association; June; San Diego, CA. Pittsburgh, PA: Air & Waste Management Association; online paper no.
34 98-A922. Available: http://capita.wustl.edu/capita/CapitaReports/REPORTSl.HTML (13 November 2003).
35 Jacob, D. J.; Horowitz, L. W.; Munger, J. W.; Heikes, B. G.; Dickerson, R. R.; Artz, R. S.; Keene, W. C. (1995)
3 6 Seasonal transition from NOX- to hydrocarbon-limited conditions for ozone production over the eastern
37 United States in September. J. Geophys. Res. 100: 9315-9324.
3 8 Kasibhatla, P.; Chameides, W. L. (2000) Seasonal modeling of regional ozone pollution in the eastern United States.
39 Geophys. Res. Lett. 27: 1415-1418.
40 Kleindienst, T. E.; Hudgens, E. E.; Smith, D. F.; McElroy, F. F.; Bufalini, J. J. (1993) Comparison of
41 chemiluminescence and ultraviolet ozone monitor responses in the presence of humidity and photochemical
42 pollutants. Air Waste 43: 213-222.
43 Milford, J. B.; Gao, D.; Sillman, S.; Blossey, P.; Russell, A. G. (1994) Total reactive nitrogen (NOy) as an indicator
44 of the sensitivity of ozone to reductions in hydrocarbon and NOX emissions. J. Geophys. Res. 99: 3533-3542.
45 National Research Council. (1991) Rethinking the ozone problem in urban and regional air pollution. Washington,
46 DC: National Academy Press. Available: http://www.nap.edu/books/0309046319/html/ [26 March, 2004].
47 Olszyna, K. J.; Bailey, E. M.; Simonaitis, R.; Meagher, J. F. (1994) O3 and NOy relationships at a rural site.
48 J. Geophys. Res. [Atmos.] 99: 14,557-14,563.
49 Parrish, D. D.; Fehsenfeld, F. C. (2000) Methods for gas-phase measurements of ozone, ozone precursors and
50 aerosol precursors. Atmos. Environ. 34: 1921-1957.
51 Pierson, W. R.; Gertler, A. W.; Bradow, R. L. (1990) Comparison of the SCAQS tunnel study with historical data.
52 Presented at: 83rd annual meeting & exhibition of the Air and Waste Management Association; June;
53 Pittsburgh, PA. Pittsburgh, PA: Air and Waste Management Association; paper no. 90-175.3.
54 Pinto, J. P.; Bruhl, C.; Thompson, A. M. (1993) The current and future envirionmental role of atmospheric methane.
55 In: Khalil, M. A. K., ed. Atmospheric methane sources, sinks, and role in global change, p. 514-531.
56 (NATO ASI Series, v. 113).
January 2005 2-25 DRAFT-DO NOT QUOTE OR CITE
-------
1 Poppe, D.; Wallasch, M.; Zimmermann, J. (1993) The dependence of the concentration of OH on its precursors
2 under moderately polluted conditions: a model study. J. Atmos. Chem. 16: 61-78.
3 Rao, S. T.; Ku, J.-Y.; Herman, S.; Zhang, K.; Mao, H. (2003) Summertime characteristics of the atmospheric
4 boundary layer and relationships to ozone levels over the eastern United States. Pure Appl. Geophys.
5 160:21-55.
6 Ryerson, T. B.; Buhr, M. P.; Frost, G. J.; Goldan, P. D.; Holloway, J. S.; Htibler, G.; Jobson, B. T.; Kuster, W. C;
7 McKeen, S. A.; Parrish, D. D.; Roberts, J. M.; Sueper, D. T.; Trainer, M.; Williams, J.; Fehsenfeld, F. C.
8 (1998) Emissions lifetimes and ozone formation in power plant plumes. J. Geophys. Res. 103(D17):
9 22,569-22,583.
10 Ryerson, T. B.; Trainer, M.; Holloway, J. S.; Parrish, D. D.; Huey, L. G.; Sueper, D. T.; Frost, G. J.; Donnelly,
11 S. G.; Schauffler, S.; Atlas, E. L.; Kuster, W. C.; Goldan, P. D.; Htibler, G.; Meagher, J. F.; Fehsenfeld, F. C.
12 (2001) Observations of ozone formation in power plant plumes and implications for ozone control strategies.
13 Science (Washington, DC) 292: 719-723.
14 Sakugawa, H.; Kaplan, I. R. (1989) H2O2 and O3 in the atmosphere of Los Angeles and its vicinity: factors
15 controlling their formation and their role as oxidants of SO2. J. Geophys. Res. [Atmos.] 94: 12,957-12,973.
16 Schichtel, B. A.; Husar, R. B. (2001) Eastern North American transport climatology during high- and low-ozone
17 days. Atmos. Environ. 35: 1029-1038.
18 Seinfeld, J. H.; Pandis, S. N. (1998) Atmospheric chemistry and physics: from air pollution to climate change.
19 New York, NY: John Wiley & Sons, Inc.
20 Shapiro, M. A. (1980) Turbulent mixing within tropopause folds as a mechanism for the exchange of chemical
21 constituents between the stratosphere and troposphere. J. Atmos. Sci. 37: 994-1004.
22 Sillman, S. (1995) The use of NOy, H2O2 and HNO3 as indicators for ozone-NOx-hydrocarbon sensitivity in urban
23 locations. J. Geophys. Res. 100: 14,175-14,188.
24 Sillman, S.; He, D.-Y. (2002) Some theoretical results concerning O3-NOX-VOC chemistry and NOX-VOC
25 indicators. J. Geophys. Res. (Atmos.) 107: 10.1029/2001JDOO1123.
26 Sillman, S.; He, D.; Cardelino, C.; Imhoff, R. E. (1997) The use of photochemical indicators to evaluate
27 ozone-NOx-hydrocarbon sensitivity: case studies from Atlanta, New York, and Los Angeles. J. Air Waste
28 Manage. Assoc. 47: 1030-1040.
29 Sillman, S.; He, D.; Pippin, M. R.; Daum, P. H.; Imre, D. G.; Kleinman, L. I.; Lee, J. H.; Weinstein-Lloyd, J. (1998)
3 0 Model correlations for ozone, reactive nitrogen, and peroxides for Nashville in comparison with
31 measurements: implications for O3-NOx-hydrocarbon chemistry. J. Geophys. Res. [Atmos.]
32 103:22,629-22,644.
33 Stedman, D. H.; Bishop, G.; Peterson, J. E.; Guenther, P. L. (1991) On-road CO remote sensing in the Los Angeles
34 Basin: final report. Sacramento, CA: California Air Resources Board, ARB Contract No. A932-189.
3 5 Stein, A. F.; Lamb, D. (2003) Empirical evidence for the low- and high-NOx photochemical regimes of sulfate and
36 nitrate formation. Atmos. Environ. 37: 3615-3625.
37 Stull, R. B. (1999) An introduction to boundary layer meteorology. Dordrecht, The Netherlands: Kluwer Academic
38 Publishers; pp. 9-15, 500-505.
39 Stull, R. B. (2000) Meteorology for scientists and engineers: a technical companion book with Ahrens' Meteorology
40 Today. 2nd ed. Pacific Grove, CA: Brooks/Cole.
41 Taubman, B. F.; Marufu, L. T.; Piety, C. A.; Doddridge, B. G.; Stehr, J. W.; Dickerson, R. R. (2004) Airborne
42 characterization of the chemical, optical, and meteorological properties, and origins of a combined
43 ozone-haze episode over the eastern United States. J. Atmos. Sci. 61: 1781-1793.
44 Trainer, M.; Parrish, D. D.; Buhr, M. P.; Norton, R. B.; Fehsenfeld, F. C.; Anlauf, K. G.; Bottenheim, J. W.; Tang,
45 Y. Z.; Wiebe, H. A.; Roberts, J. M.; Tanner, R. L.; Newman, L.; Bowersox, V. C.; Meagher, J. F.; Olszyna,
46 K. J.; Rodgers, M. O.; Wang, T.; Berresheim, H.; Demerjian, K. L.; Roychowdhury, U. K. (1993) Correlation
47 of ozone with NOy in photochemically aged air. J. Geophys. Res. [Atmos.] 98: 2917-2925.
48 U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants.
49 Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF. 3v.
50 Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, and PB96-185608. Available:
51 http://cfpub2.epa.gov/ncea/.
52 U.S. Environmental Protection Agency. (2000) Air quality criteria for carbon monoxide. Research Triangle Park,
53 NC: National Center for Environmental Assessment; report no. EPA/600/P-99/00IF. Available:
54 http://www.epa.gov/ncea/pdfs/coaqcd.pdf (7 May 2003).
55 Volz-Thomas, A.; Geiss, H.; Hofzumahaus, A.; Becker, K.-H. (2003) Introduction to special section: photochemistry
56 experiment in BERLIOZ. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/JD002029.
January 2005 2-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 Zhang etal., 1997
2 Zimmermann, J.; Poppe, D. (1993) Nonlinear chemical couplings in the tropospheric NOX-HOX gas phase chemistry.
3 J. Atmos. Chem. 17: 141-155.
4
5
January 2005 2-27 DRAFT-DO NOT QUOTE OR CITE
-------
i ANNEX AX2. PHYSICS AND CHEMISTRY OF OZONE
2 IN THE ATMOSPHERE
3
4
5 AX2.1 INTRODUCTION
6 This annex (Annex AX2) provides detailed supporting information for Chapter 2 on the
7 physics and chemistry of ozone (O3) in the atmosphere. The organization of the material in this
8 annex follows that used in prior Air Quality Criteria Documents, i.e., material is presented in
9 sections and subsections. This annex provides material supporting Chapter 2 of the current draft
10 Air Quality Criteria Document for Ozone.
11 Section AX2.2 focuses on the chemistry of O3 formation. A very brief overview of
12 atmospheric structure is presented in Section AX2.2.1. An overview of O3 chemistry is given in
13 Section AX2.2.2. Information about reactive chemical species that initiate the oxidation of
14 VOCs is given in Section AX2.2.3. The chemistry of nitrogen oxides is then discussed briefly in
15 Section AX2.2.4. The oxidation of methane, the simplest hydrocarbon is outlined in
16 Section AX2.2.5.
17 The photochemical cycles leading to O3 production are best understood by considering the
18 oxidation of methane, structurally the simplest VOC. The CH4 oxidation cycle serves as a model
19 which can be viewed as representing the chemistry of the relatively clean or unpolluted
20 troposphere (although this is a simplification because vegetation releases large quantities of
21 complex VOCs, such as isoprene, into the atmosphere). Although the chemistry of the VOCs
22 emitted from anthropogenic and biogenic sources in polluted urban and rural areas is more
23 complex, a knowledge of the CH4 oxidation reactions aids in understanding the chemical
24 processes occurring in the polluted atmosphere because the underlying chemical principles are
25 the same. The oxidation of more complex hydrocarbons (alkanes, alkenes, and aromatic
26 compounds) is discussed in Sections AX2.2.6, AX2.2.7, and AX2.2.8, respectively. The
27 chemistry of oxygenated species is addressed in Section AX2.2.9. Greater emphasis is placed on
28 the oxidation of aromatic hydrocarbons in this section because of the large amount of new
29 information available since the last Air Quality Criteria for Ozone document (AQCD 96) was
30 published (U.S. Environmental Protection Agency, 1996) and because of their importance in O3
31 formation in polluted areas. Multiphase chemical processes influencing O3 are discussed in
January 2005 AX2-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 Section AX2.2.10. Meteorological processes that control the formation of O3 and other oxidants
2 and that govern their transport and dispersion, and the sensitivity of O3 to atmospheric
3 parameters are given in Section AX2.3. Greater emphasis is placed on those processes for which
4 a large amount of new information has become available since AQCD 96. The role of
5 stratospheric-tropospheric exchange in determining O3 in the troposphere is presented in Section
6 AX2.3.1. The importance of deep convection in redistributing O3 and its precursors and other
7 oxidants throughout the troposphere is given in Section AX2.3.2. The possible importance of
8 nocturnal low level jets in transporting O3 and other pollutants is presented in Section AX2.3.3.
9 Information about the mechanisms responsible for the intercontinental transport of pollutants and
10 for the interactions between stratospheric-tropospheric exchange and convection is given in
11 Section AX2.3.4. Much of the material in this section is based on results of field programs
12 examining atmospheric chemistry over the North Atlantic ocean. The sensitivity of O3 to solar
13 ultraviolet radiation and temperature is given in Section AX2.3.5. The relations of O3 to its
14 precursors and to other oxidants based on field and modeling studies are discussed in Section
15 AX2.4. Methods used to calculate relations between O3 its precursors and other oxidants are
16 given in Section AX2.5. Chemistry-transport models are discussed in Section AX2.5.1.
17 Emissions of O3 precursors are presented in Section AX2.5.2. Issues related to the evaluation of
18 chemistry-transport models and emissions inventories are presented in Section AX2.5.3.
19 Measurement methods are summarized in Section AX2.6. Methods used to monitor ground-
20 level O3 are given in Section AX2.6.1, NO and NO2 in Section AX2.6.2, HNO3 in Section
21 AX2.6.3 and some important VOCs in Section AX2.6.4.
22
23
24 AX2.2 TROPOSPHERIC OZONE CHEMISTRY
25 AX2.2.1 Atmospheric Structure
26 The atmosphere can be divided into several distinct vertical layers, based primarily on the
27 major mechanism by which that portion of the atmosphere is heated or cooled. The lowest major
28 layer is the troposphere, which extends from the earth's surface to about 8 km above polar
29 regions and to about 16 km above tropical regions. The troposphere is heated by convective
30 transport from the surface, and by the absorption of infrared radiation emitted by the surface,
31 principally by water vapor and CO2. The planetary boundary layer (PEL) is the sublayer of the
January 2005 AX2-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 troposphere that mixes with surface air on time scales of a few hours or less. It typically extends
2 to 1-2 km altitude and is often capped by a temperature inversion. The sublayer of the
3 troposphere above the PEL is called the free troposphere. Ventilation of the PEL with free
4 tropospheric air takes place on a time scale of a week. Vertical mixing of the whole troposphere
5 takes place on a time scale of a month or two. The stratosphere extends from the tropopause, or
6 the top of the troposphere, to about 50 km in altitude. The upper stratosphere is heated by the
7 absorption of solar ultraviolet radiation by O3, while dissipation of wave energy transported
8 upwards from the troposphere is a primary heating mechanism in the lower stratosphere.
9 Heating of the stratosphere is balanced by radiative cooling due to infrared emissions to space by
10 CO2, H2O, and O3. As a result of heating of the upper stratosphere, temperatures increase with
11 height, inhibiting vertical mixing. A schematic overview of the major chemical cycles involved
12 in O3 formation and destruction in the stratosphere and troposphere is shown in Figure AX2-1.
13 The figure emphasizes gas phase processes, but the importance of multiphase processes is
14 becoming apparent. The sequences of reactions shown in the lower right quadrant of the figure
15 will be discussed in Section AX2.2. The reader is referred to any of the large number of texts on
16 atmospheric chemistry, such as Wayne (2000) or Seinfeld and Pandis (1998), for an introduction
17 to stratospheric photochemistry, including the impact of O3-destroying compounds.
18
19 AX2.2.2 Overview of Ozone Chemistry
20 Ozone is found not only in polluted urban atmospheres but throughout the troposphere,
21 including remote areas of the globe. Even without ground-level production, some O3 would be
22 found in the troposphere due to downward transport from the stratosphere. Tropospheric
23 photochemistry leading to the formation of O3 and other photochemical air pollutants is
24 complex, involving thousands of chemical reactions and thousands of stable and reactive
25 intermediate products. Other photochemical oxidants, such as peroxyacetyl nitrate (PAN), are
26 among the reactive products. Ozone can be photolyzed in the presence of water to form
27 hydroxyl radical (OH), which is responsible for the oxidation of NOX and SOX to form nitric
28 (HNO3) and sulfuric acid (H2SO4), respectively. Ozone participates directly in the oxidation of
29 unsaturated hydrocarbons, via the ozonolysis mechanism, yielding secondary organic
30 compounds that contribute to aerosol formation and mass, as well as formaldehyde (H2CO) and
31 other carbonyl compounds, such as aldehydes and ketones.
January 2005 AX2-3 DRAFT-DO NOT QUOTE OR CITE
-------
Stratosphere
Non-Pola
Regions
Figure AX2-1. Schematic overview of ozone photochemistry in the stratosphere
and troposphere.
1
2
3
4
5
6
7
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.
January 2005
AX2-4
DRAFT-DO NOT QUOTE OR CITE
-------
O(3P) + O2 + M -» O3 + M, where M = an air molecule. (AX2-2)
1
2 Reaction AX2-2 is the only significant reaction forming O3 in the troposphere.
3
4 NO and O3 react to reform NO2:
5
6
+ 2. (AX2-3)
7
8
9 This reaction is responsible for O3 decreases found near sources of NO (e.g., highways)
10 especially at night. The oxidation of reactive VOCs leads to the formation of reactive radical
1 1 species that allow the conversion of NO to NO2 without the participation of O3 (as in
12 reaction AX2-3).
13
HO-,% RO2-
NO - = - > N02. (AX2.4)
14
15 O3 can, therefore, accumulate as NO2 photolyzes as in reaction AX2-1 followed by reaction
16 AX2-2.
17 It is often convenient to speak about families of chemical species, that are defined in terms
18 of members which interconvert rapidly among themselves on time scales that are shorter than
19 that for formation or destruction of the family as a whole. For example, an 'odd oxygen' (Ox)
20 family can be defined as £ (O(3P) +O(1D)+ O3 + NO2) in much the same way as the NOX
21 (NO + NO2) family is defined. We can then see that production of Ox occurs by the schematic
22 reaction AX2-4, and that the sequence of reactions given by reactions AX2-1 through AX2-3
23 represents no net production of Ox. Definitions of species families and methods for constructing
24 families are discussed in Jacobson (1999) and references therein. Other families that include
25 nitrogen containing species, and will be referred to later in this chapter, are NOZ which is the
26 sum of the products of the oxidation of NOX = £ (HNO3 + PAN (CH3CHO-OO-NO2) + HNO4 +
27 other organic nitrates + particulate nitrate); and NOy, which is the sum of NOX and NOZ.
January 2005 AX2-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX2.2.3 Initiation of the Oxidation of VOCs
2 The key reactive species in the troposphere is the OH radical. OH radicals are
3 responsible for initiating the photochemical oxidation of CO and most anthropogenic and
4 biogenic VOCs, including those responsible for depleting stratospheric O3 (e.g., CH3Br,
5 hydroclorofluorocarbons), and those which contribute to the greenhouse effect (e.g., CH4).
6 Because of their role in removing so many potentially damaging species, OH radicals have
7 sometimes been referred to as the atmosphere's detergent. In the presence of NO, reactions of
8 OH with VOCs lead to the formation of O3. In addition to OH radicals, there are several other
9 atmospheric species such as OH, NO3, Cl and Br radicals and O3 that are capable of initiating
10 VOC oxidation. Rate coefficients and estimated atmospheric lifetimes (the e-folding time) for
11 reactions of a number of alkanes, alkenes and dienes involved in O3 formation with these
12 oxidants at concentrations characteristic of the relatively unpolluted planetary boundary layer are
13 given in Table AX2-1. As can be seen from Table AX2-1, there is a wide range of lifetimes
14 calculated for the different species. However, under certain conditions the relative importance of
15 these oxidants can change from those shown in the table. For hydrocarbons whose atmospheric
16 lifetime is much longer than a day, diurnally averaged concentrations of oxidant concentrations
17 can be used, but for those whose lifetime is much shorter than a day it is more appropriate to use
18 either daytime or night-time averages depending on when the oxidant is at highest
19 concentrations. During these periods, these averages are of the order of twice the values used
20 here.
21 The main source of OH radicals is the photolysis of O3 by solar ultraviolet radiation at
22 wavelengths < 340 nm (solar radiation at wavelengths < 320 nm is also referred to as UV-B) to
23 generate electronically excited O(1D) atoms (Jet Propulsion Laboratory, 2003),
24
O3 + hv -> O2 + O(1D). (AX2-5)
25
26 The O(JD) atoms can either be deactivated to the ground state O(3P) atom by collisions with N2
27 and O2, or they react with water vapor to form two OH radicals:
28
29 O(1D) + H2O ->• 2(-OH) (AX2-6)
January 2005 AX2-6 DRAFT-DO NOT QUOTE OR CITE
-------
3
Table AX2-1. Comparison of the Atmospheric Lifetimes (u) of Low Molecular Weight Hydrocarbons Due to Reaction with
OH, NO3, Cl, Br and O3
<<
to
o
o
X
to
1
o
£j
H
6
o
0
H
O
o
0
o
H
W
k, cm3 molecule"1 s"1
OH
Hydrocarbon k x 1012
Alkanes
Ethane 0.24
Propane 1.1
2-Methylpropane 2.1
n -Butane 2.3
2-Methylbutane 4
w-Pentane 3.8
2,2-Dimethylbutane 2.7
2,3-Dimethylbutane 6.4
2-Methylpentane 5.6
3-Methylpentane 5.8
w-Hexane 5.2
2,2,4-Trimethylpentane 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
Br O3
k x 1012 T kx 1018
3.1 x 1(T7 l.OxlOV <0.01
0 6.5xl03y2 <0.01
<1.0x >3.2xl07y2 <0.01
<1.0x >3.2xl07y2 <0.01
NA NA NA
NA NA NA
NA NA NA
0.0064 50 y NA
NA NA NA
NA NA NA
NA NA NA
0.0068 47 y NA
T
>3.2y
>3.2y
>3.2y
>3.2y
NA
NA
NA
NA
NA
NA
NA
NA
-------
3
O
H
W
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
<<
to
o
o
X
to
1
oo
o
§
H
6
O
0
H
O
k, cm3 molecule"1 s"1
OH
Hydrocarbon
Alkenes
Ethene
Propene
2-Methylpropene
1-Butene
/ra«s-2-Butene
c/s-2-Butene
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).
-------
1 The O(3P) atoms formed directly in the photolysis of O3 in the Huggins and Chappuis bands or
2 formed from deactivation of O(1D) atoms reform O3 through reaction AX2-2. Hydroxyl radicals
3 produced by reactions AX2-5 and AX2-6 can react further with species such as carbon monoxide
4 and with many hydrocarbons (for example, CH4) to produce HO2 radicals.
5 Measurements of OH radical concentrations in the troposphere (Poppe et al., 1995; Eisele
6 et al., 1997; Brune et al., 1999; Martinez et al., 2003; Ren et al., 2003) show that, as expected,
7 the OH radical concentrations are highly variable in space and time, with daytime maximum
8 concentrations of > 3 x io6 molecules /cm3 in urban areas. A global, mass-weighted mean
9 tropospheric OH radical concentration also can be derived from the estimated emissions and
10 measured atmospheric concentrations of methyl chloroform (CH3CC13) and the rate constant for
11 the reaction of the OH radical with CH3CC13. Krol et al. (1998) derived a global average OH
12 concentration of 1.07 x IO6 molecules /cm3 for 1993 along with an upward trend of about
13 0.5%/yr between 1978 and 1993. Using an integrated data set of observed O3, H2O, NOy, CO,
14 VOCs, temperature and cloud optical depth, Spivakovsky et al. (2000) calculated a global annual
15 mean OH concentration of 1.16 x IO6 molecules cm"3, consistent to within less than 10% of the
16 value obtained by Krol et al. (1998).
17 HO2 radicals do not initiate the oxidation of hydrocarbons, but serve to recycle OH mainly
18 by way of reaction with NO, ozone, and itself (the latter produces H2O2, which can photolyze to
19 yield OH). The HO2 radicals also react with organo-peroxy radicals produced during the
20 oxidation of VOCs to form organo-peroxides (cf Section AX2.2.5, reaction AX2-20, e.g.).
21 Organo-peroxides may undergo wet or dry deposition (Wesely and Hicks, 2000), or degrade
22 further by photolysis and reaction with OH (Jet Propulsion Laboratory, 2003).
23 At night, NO3 assumes the role of primary oxidant (Wayne, 1991). Although it is generally
24 less reactive than OH, its high abundance in the polluted atmosphere compensates for its lower
25 reactivity. For several VOCs, however, including dimethylsulfide, isoprene, some terpenes
26 (a-pinene, limonene, linalool) and some phenolic compounds (phenol, o-cresol), oxidation by
27 NO3 at night is competitive with oxidation by OH during the day, making it an important
28 atmospheric removal mechanism for these compounds (Wayne, 1991) (see Table AX2-1). The
29 role of NO3 radicals in the chemistry of the remote marine boundary layer has been examined
30 recently by Allen et al., (2000) and in the polluted continental boundary layer by Geyer and
31 Platt(2002).
January 2005 AX2-9 DRAFT-DO NOT QUOTE OR CITE
-------
1 Cl atoms, derived from products of multiphase processes can initiate the oxidation of most
2 of the same VOCs as OH radicals, however, the rate coefficients for the reactions of alkanes with
3 Cl atoms are usually much higher. Cl will also oxidize alkenes and aromatic compounds, but
4 with a significantly lower rate constant than for OH reactions. Following the initial reaction
5 with Cl, the degradation of the hydrocarbon proceeds as with OH and NO3, generating an
6 enhanced supply of odd hydrogen radicals leading to O3 production in the presence of sufficient
7 NOX. The corresponding reactions of Br with hydrocarbons proceed in a similar manner, but
8 with rate coefficients that can be substantially lower or higher.
9 Chlorine and bromine radicals will also react directly with O3 to form CIO and BrO
10 radicals, providing a sink for odd oxygen if they do not react with NO to form NO2 (e.g.,
11 Pszenny et al., 1993). As with other oxidants present in the atmosphere, Cl chemistry provides a
12 modest net sink for O3 when NOX is less than 20 pptv, and is a net source at higher NOX. Kasting
13 and Singh (1986) estimated that as much as 25% of the loss of nonmethane hydrocarbons in the
14 nonurban atmosphere can occur by reaction with Cl atoms, based on the production of Cl atoms
15 from gas phase photochemical reactions involving chlorine containing molecules (HC1, CH3C1,
16 CHC13, etc.). Elevated concentrations of atomic Cl and other halogen radicals can be found in
17 polluted coastal cities where precursors are emitted directly from industrial sources and/or are
18 produced via acid-catalyzed reactions involving sea-salt particles (Tanaka et al., 2000; Spicer
19 etal., 2001).
20 Substantial chlorine-VOC chemistry has been observed in the cities of Houston and
21 Beaumont/Port Arthur, Texas (Tanaka et al., 2000; Chang et al., 2002; Tanaka et al., 2003a).
22 Industrial production activities in those areas frequently result in large releases of chlorine gas
23 (Tanaka et al., 2000). Chloromethylbutenone (CMBO), the product of the oxidation of isoprene
24 by atomic Cl and a unique marker for chlorine radical chemistry in the atmosphere (Nordmeyer
25 et al., 1997), has been found at significant mixing ratios (up to 145 pptv) in ambient Houston air
26 (Riemer and Apel, 2001). However, except for situations in which there are strong local sources
27 such as these, the evidence for the importance of Cl as an oxidizing agent is mixed. Parrish et al.
28 (1992, 1993) argued that ratios of selected hydrocarbons measured at Pt. Arena, CA were
29 consistent with loss by reaction with OH radicals and that any deviations could be attributed to
30 mixing processes. Finlayson-Pitts (1993), on the other hand had suggested that these deviations
31 could have been the result of Cl reactions. McKeen et al. (1996) suggested that hydrocarbon
January 2005 AX2-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 ratios measured downwind of anthropogenic source regions affecting the western Pacific Basin
2 are consistent with loss by reaction with OH radicals only. Rudolph et al. (1997), based on data
3 for several pairs of hydrocarbons collected during a cruise in the western Mediterranean Sea, the
4 eastern mid- and North Atlantic Ocean and the North Sea during April and May of 1991, also
5 found that ratios of hydrocarbons to each other are consistent with their loss given mainly by
6 reaction with OH radicals without substantial contributions from reactions with Cl. Their best
7 estimate, for their sampling conditions was a ratio of Cl to OH of about 1CT3, implying a
8 concentration of Cl of about 103/cm3 using the globally averaged OH concentration of about
9 lOVcm3 given above. In contrast Wingenter et al. (1996) and Singh et al. (1996a) inferred
10 significantly higher concentrations of atomic Cl (104 to 10s cm"3) based on relative concentration
11 changes in VOCs measured over the eastern North Atlantic and Pacific Oceans, respectively.
12 Similar approaches employed over the high-latitude southern ocean yielded lower estimates of
13 Cl concentrations (103 cm"3; Wingenter et al., 1999). Taken at face value, these observations
14 indicate substantial variability in Cl concentrations and uncertainty in "typical" values.
15
16 AX2.2.4 Chemistry of Nitrogen Oxides in the Troposphere
17 In the troposphere, NO, NO2, and O3 are interrelated by the following reactions:
18
26
+ U2 (AX2-3)
NO2 + hv -» NO + O(3P) (AX2-1)
0(3P) + 02 + M -» 03 + M (AX2-2)
19
20 The reaction of NO2 with O3 leads to the formation of the nitrate (NO3) radical,
21
22 N02 + 03 -»N03- + 02, (AX2-7)
23
24 which in the lower troposphere is nearly in equilibrium with dinitrogen pentoxide (N2O5):
25
NO3' + NO2 <-^-» N2 O5. (AX2-8)
January 2005 AX2-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 However, because the NO3 radical photolyzes rapidly (with a lifetime of «5 s for an overhead
2 sun [Atkinson et al., 1992a]),
3
NCy + hv -> NO + O2 (10%) (Ax2-9a)
-»NO2 + O(3P) (90%) (AX2-9b)
4
5 its concentration remains low during daylight hours, but can increase after sunset to nighttime
6 concentrations of < 5 x 107 to 1 x 1010 molecules cm"3 (< 2 to 430 ppt) over continental areas
7 influenced by anthropogenic emissions of NOX (Atkinson et al., 1986). This leads to an increase
8 of N2O5 concentrations during the night by reaction (AX2-8).
9 The tropospheric chemical removal processes for NOX involve the reaction of NO2 with the
10 OH radical and the hydrolysis of N2O5 in aqueous aerosol solutions to produce HNO3.
11
•OH + N02 -^-^ HN03 (AX2-10)
12
N2O5 H2°(1) > HNO3 (AX2-11)
13
14 The gas-phase reaction of the OH radical with NO2 initiates the major and ultimate removal
15 process for NOX in the troposphere. This reaction removes radicals (OH and NO2) and competes
16 with hydrocarbons for OH radicals in areas characterized by high NOX concentrations, such as
17 urban centers (see Section AX2.4). In addition to gas-phase nitric acid, Golden and Smith
18 (2000) have concluded that, pernitrous acid (HOONO) is also produced by the reaction of NO2
19 and OH radicals on the basis of theoretical studies. However, a recent assessment (Jet
20 Propulsion Laboratory, 2003) has concluded that this channel represents a minor yield
21 (approximately 15% at the surface). HOONO will thermally decompose or photolyze.
22 Gas-phase HNO3 formed from reaction AX2-10 undergoes wet and dry deposition to the surface
23 and uptake by ambient aerosol particles. The tropospheric lifetime of NOX due to reaction
24 AX2-10 ranges from a few hours to a few days. Geyer and Platt (2002) concluded that reaction
25 AX2-11 constituted about 10% of the removal of NOX at a site near Berlin, Germany during
January 2005 AX2-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 spring and summer. However, during winter the relative importance of reaction AX2-11 could
2 be much higher because of the much lower concentration of OH radicals and the enhanced
3 stability of N2O5 due to lower temperatures and intensity of sunlight. Note that reaction AX2-11
4 surely proceeds as a heterogeneous reaction.
5 OH radicals also can react with NO to produce nitrous acid (HNO2):
6
•OH + NO M >HN02. (AX2-12)
7
8 In the daytime, HNO2 is rapidly photolyzed back to the original reactants:
9
10 HNO2 + hv ->• 'OH + NO. (AX2-13)
11
12 At night, HNO2 can be formed by heterogeneous reactions of NO2 in aerosols or at the earth's
13 surface (Lammel and Cape, 1996; Jacob, 2000; Sakamaki et al., 1983; Pitts et al., 1984a;
14 Svensson et al., 1987; Jenkin et al., 1988; Lammel and Perner, 1988; Notholt et al., 1992a,b).
15 This results in accumulation of HNO2 during nighttime. Modeling studies suggest that
16 photolysis of this HNO2 following sunrise, could provide an important early-morning source of
17 OH radicals to drive ozone formation (Harris et al., 1982).
18 Another important process controlling NOX concentrations is the formation of organic
19 nitrates. Oxidation of VOCs produces organic peroxy radicals (RO2), as discussed in the
20 hydrocarbon chemistry subsections to follow. Reaction of these RO2 radicals with NO and NO2
21 produces organic nitrates (RONO2) and peroxynitrates (RO2NO2):
22
23 RO2' + NO M > RONO2 (AX2-14)
24
25 RO2' + NO2 M > RO2NO2 (AX2-15)
26
27 Reaction (AX2-14) is a minor branch for the reaction of RO2 with NO (the major branch
28 produces RO and NO2, as discussed in the next section). The organic nitrate yield increases with
29 carbon number (Atkinson, 2000).
January 2005 AX2-13 DRAFT-DO NOT QUOTE OR CITE
-------
1 The organic nitrates may react further, depending on the functionality of the R group, but
2 they will typically not return NOX and can therefore be viewed as a permanent sink for NOX.
3 This sink is usually small compared to HNO3 formation, but the formation of isoprene nitrates
4 may be a significant sink for NOX in the United States in summer (Liang et al., 1998).
5 The peroxynitrates produced by (AX2-15) are thermally unstable and most have very short
6 lifetimes (less than a few minutes) against thermal decomposition to the original reactants. They
7 are thus not effective sinks of NOX. Important exceptions are the peroxyacylnitrates (PANs)
8 arising from the peroxyacyl radicals RC(O)OO produced by oxidation and photolysis of
9 carbonyl compounds. PANs have lifetimes ranging from ~1 hour at room temperature to several
10 weeks at 250K. They can thus provide an effective sink of NOX at cold temperatures, but also a
11 reservoir allowing eventual release of NOX as air masses warm, in particular by subsidence. By
12 far the most important of these PANs compounds is peroxyacetylnitrate (PAN), with formula
13 CH3C(O)OONO2. PAN is a significant product in the oxidation of most VOCs. It is now well
14 established that PAN decomposition provides a major source of NOx in the remote troposphere
15 (Staudt et al., 2003). PAN decomposition in subsiding Asian air masses over the eastern Pacific
16 could make an important contribution to O3 enhancement in the U.S. from Asian pollution
17 (Hudman et al., 2004).
18
19 AX2.2.5 The Methane Oxidation Cycle
20 The photochemical cycles leading to O3 production are best understood by considering the
21 oxidation of methane, structurally the simplest VOC. The CH4 oxidation cycle serves as a model
22 which describes the chemistry of the relatively clean or unpolluted troposphere (although this is
23 a simplification because vegetation releases large quantities of complex VOCs into the
24 atmosphere). Although the chemistry of the VOCs emitted from anthropogenic and biogenic
25 sources in polluted urban and rural areas is more complex, a knowledge of the CH4 oxidation
26 reactions aids in understanding the chemical processes occurring in the polluted atmosphere
27 because the underlying chemical principles are the same.
28 Methane is emitted into the atmosphere as the result of anaerobic microbial activity in
29 wetlands, rice paddies, the guts of ruminants, landfills, and from mining and combustion of
30 fossil fuels (Intergovernmental Panel on Climate Change, 2001). The major tropospheric
31 removal process for CH4 is by reaction with the OH radical,
January 2005 AX2-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 -OH + CH4 -» H2O + CH3. (AX2-16)
2
3 In the troposphere, the methyl radical reacts solely with O2 to yield the methyl peroxy (CH3O2»
4 radical (Atkinson et al., 1992a):
CH3 + O2 M > CH3O2-. (AX2-17)
5
6 In the troposphere, the methyl peroxy radical can react with NO, NO2, HO2 radicals, and
7 other organic peroxy (RO^ radicals, with the reactions with NO and HO2 radicals being the most
8 important (see, for example, World Meteorological Organization, 1990). The reaction with NO
9 leads to the formation of the methoxy (CH36) radical,
10
CH3O2- + NO -» CH36 + NO2. (AX2-18)
11 The reaction with the HO2 radical leads to the formation of methyl hydroperoxide
12 (CH3OOH),
CH302- + H02- -» CH3OOH + O2, (AX2-19)
13 which can photolyze or react with the OH radical (Atkinson et al., 1992a):
CH3OOH + hv -» CH36 + -OH. (AX2-20)
•OH + CH3OOH -> H2O + CH3O2' (AX2-21 a)
14 or
15
-» H2O + CH2OOH fast decomposition
(AX2-21b)
CH2OOH + M -> H2CO + -OH
January 2005 AX2-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 Methyl hydroperoxide is much less soluble than hydrogen peroxide (H2O2) and so wet deposition
2 after incorporation into cloud droplets is much less important as a removal process than it is for
3 H2O2. CH3OOH can also be removed by dry deposition to the surface and it can also be
4 transported by conversion to the upper troposphere. The lifetime of CH3OOH in the troposphere
5 due to photolysis and reaction with the OH radical is estimated to be «2 days. Methyl
6 hydroperoxide is then a temporary sink of radicals, with its wet or dry deposition representing a
7 loss process for tropospheric radicals.
8 The only important reaction for the methoxy radical (CH36) is
9
CH3O + O2 -» H2CO + HCy. (AX2-22)
10
1 1 HCy + NO -> -OH + N02 (AX2-23)
12 The HO2 radicals produced in (AX2-22) can react with NO, O3, or other HO2 radicals according
13 to,
14
1 5 H02- + 03 -> -OH + 202 (AX2-24)
16
17 HO2- + HO2- -» H2O2 + O2. (AX2-25)
18 Formaldehyde (H2CO) produced in reaction AX2-22 can be photolyzed:
19
20 H2CO+hv^H2 + CO (55%) (AX2-26a)
21
(45%) (AX2-26b)
23 Formaldehyde also reacts with the OH radical,
24
25 -OH + H2CO -> H2O + HCO. (AX2-27)
January 2005 AX2-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 The H atom and HCO (formyl) radical produced in these reactions react solely with O2 to form
2 the HO2 radical:
3
•H + O2 + M -» HO2' + M (AX2-28)
HCO + O2 ->• HO2- + CO. (AX2-29)
4
5 The lifetimes of H2CO due to photolysis and reaction with OH radicals are «4 h and 1.5 days,
6 respectively, leading to an overall lifetime of slightly less than 4 hours for H2CO for overhead
7 sun conditions (Rogers, 1990).
8 The final step in the oxidation of CH4 involves the oxidation of CO by reaction with the
9 OH radical to form CO2:
10
11 CO +-OH-> C02 + «H (AX2-30)
12
1 HCO + 02 ->> H02' + CO. (AX2-29)
14
15 The lifetime of CO in the lower troposphere is «2 months at mid-latitudes.
16 NO and HO2 radicals compete for reaction with CH3O2 and HO2 radicals, and the reaction
17 route depends on the rate constants for these two reactions and the tropospheric concentrations
18 ofHO2andNO. The rate constants for the reaction of the CH3O2 radicals with NO (reaction
19 AX2-18) and HO2 radicals (reaction AX2-19) are of comparable magnitude (e.g., Jet Propulsion
20 Laboratory, 2003). Based on expected HO2 radical concentrations in the troposphere, Logan
21 et al. (1981) calculated that the reaction of the CH3O2 radical with NO dominates for NO mixing
22 ratios of > 30 ppt. For NO mixing ratios < 30 ppt, the reaction of the CH3O2 radical with HO2
23 dominates. The overall effects of methane oxidation on O3 formation for the case when
24 NO > 30 ppt can be written as:
25
January 2005 AX2-17 DRAFT-DO NOT QUOTE OR CITE
-------
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)
1
2 Further O3 formation occurs, based on the subsequent reactions of H2CO, e.g.,
3
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)
4
5 Reactions in the above sequence lead to the production of two OH radicals which can further
6 react with atmospheric constituents (e.g., Crutzen, 1973). There is also a less important
7 pathway:
January 2005 AX2-18 DRAFT-DO NOT QUOTE OR CITE
-------
2 H2CO + hv->H2 + CO (AX2.26a)
3 CO + -OH -> C02 + -H (AX2.30)
4
•H + O2 + M -» HO2* + M (AX2-28)
6 H02- + NO^-OH + N02 (AX2-23)
7 NO2 + hv -> NO + O(3P) (AX2-1)
8
O(3P) + O2 + M -» O3 + M (AX2-2)
10 Net: H2CO + 2O2 + hv-»CO2 + O3+H2 (AX2-33)
11
12
13 These reaction sequences are important for tropospheric chemistry because formaldehyde is an
14 intermediate product of the oxidation of most VOCs. The reaction of O3 and HO2 radicals leads
15 to the net destruction of tropospheric O3:
16
H02- + 03 -> -OH + 202 (AX2-24)
17
18 Using the rate constants reported for reactions AX2-23 and AX2-24 (Atkinson et al., 1992a) and
19 the background tropospheric O3 mixing ratios given above, the reaction of HO2 radicals with NO
20 dominates over reaction with O3 for NO mixing ratios > 10 ppt. The rate constant for
21 reaction AX2-25 is such that an NO mixing ratio of this magnitude also means that the HO2
22 radical reaction with NO will be favored over the self-reaction of HO2 radicals.
23 Consequently, there are two regimes in the "relatively clean" troposphere, depending on
24 the local NO concentration: (1) a "very low-NOx" regime in which HO2 and CH3O2 radicals
25 combine (reaction AX2-19), and HO2 radicals undergo self-reaction (to form H2O2) and react
26 with O3 (reactions AX2-25 and AX2-24), leading to net destruction of O3 and inefficient OH
27 radical regeneration (see also Ehhalt et al., 1991; Ayers et al., 1992); and (2) a "low-NOx"
28 regime (by comparison with much higher NOX concentrations found in polluted areas) in which
29 HO2 and CH3O2 radicals react with NO to convert NO to NO2, regenerate the OH radical, and,
January 2005 AX2-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 through the photolysis of NO2, produce O3. In the "low NOX" regime there still may be
2 significant competition from peroxy-peroxy reactions, depending on the local NO concentration.
3 Nitric oxide mixing ratios are sufficiently low in the remote marine boundary layer
4 relatively unaffected by transport of NOX from polluted continental areas (< 15 ppt) that
5 oxidation of CH4 will lead to net destruction of O3, as discussed by Carroll et al. (1990) and
6 Ayers et al. (1992). In continental and marine areas affected by transport of NOX from
7 combustion sources, NO mixing ratios are high enough (of the order of-one to a few hundred
8 ppt) for the oxidation of CH4, nonmethane hydrocarbons (NMHCs) and CO to lead to net O3
9 formation (e.g., Carroll et al., 1990; Dickerson et al., 1995). Generally, NO mixing ratios
10 increase with altitude and can be of the order of fifty to a few hundred ppt, in the upper
11 troposphere depending on location. The oxidation of peroxides, carbon monoxide and acetone
12 transported upward by convection, in the presence of this NO, can lead to local O3 formation
13 (e.g., Singh et al., 1995; McKeen et al., 1997; Wennberg et al., 1998; Bruhl et al., 2000).
14
15 AX2.2.6 The Atmospheric Chemistry of Alkanes
16 The same basic processes by which CH4 is oxidized occur in the oxidation of other, even
17 more reactive and more complex VOCs. As in the CH4 oxidation cycle, the conversion of NO to
18 NO2 during the oxidation of VOCs results in the production of O3 and the efficient regeneration
19 of the OH radical, which in turn can react with other VOCs (Figure AX2-2). The chemistry of
20 the major classes of VOCs important for O3 formation such as alkanes, alkenes (including
21 alkenes from biogenic sources), and aromatic hydrocarbons will be summarized in turn.
22 Reaction with OH radicals represents the main loss process for alkanes and as also
23 mentioned earlier, reaction with nitrate and chlorine radicals are additional sinks for alkanes.
24 For alkanes having carbon-chain lengths of four or less, the chemistry is well understood and the
25 reaction rates are slow in comparison to alkenes and other VOCs of similar structure and
26 molecular weight. See Table AX2-1 for a comparison of reaction rate constants for several
27 small alkanes and their alkene and diene homologues. For alkanes larger than C5, the situation
28 is more complex because the products generated during the degradation of these compounds are
29 usually not well characterized. Branched alkanes have rates of reaction that are highly
30 dependent on carbon backbone structure. Stable products of alkane photooxidation are known to
31 include carbonyl compounds, alkyl nitrates, and hydroxycarbonyls.
January 2005 AX2-20 DRAFT-DO NOT QUOTE OR CITE
-------
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).
1 Alkyl nitrates form primarily as an alternate product of reaction AX2-34 (below). Several
2 modeling studies have predicted that large fractions of NOy exist as alkyl and hydroxy alkyl
3 nitrates (Calvert and Madronich, 1987; Atherton and Penner, 1988; Trainer et al., 1991).
4 In NOX- and VOC-rich urban atmospheres, 100 different alkyl and 74 different hydroxy alkyl
5 nitrate compounds have been predicted and identified (Calvert and Madronich, 1987; Schneider
6 and Ballschmiter, 1999; Schneider et al., 1998). Uncertainties in the atmospheric chemistry of
7 the alkanes include the branching ratio of reaction AX2-34, i.e., the extent to which alkyl nitrates
8 form versus RO and NO2. These uncertainties affect modeling predictions of NOX
9 concentrations, NO-to-NO2 conversion and O3 formation during photochemical degradation of
10 the VOCs. Discrepancies between observations and theory have been found in aircraft
11 measurements of NOy (Singh et al., 1996b). Recent field studies conducted by Day et al. (2003)
12 have shown that large fractions of organic nitrates, which may be associated with isoprene
13 oxidation products, are present in urban and rural atmosphere that have not been previously
14 measured and considered in NOV calculations to date.
January 2005
AX2-21
DRAFT-DO NOT QUOTE OR CITE
-------
1 Alcohols and ethers in ambient air react almost exclusively with the OH radical, with the
2 reaction proceeding primarily via H-atom abstraction from the C - H bonds adjacent to the
3 oxygen-containing function group in these compounds (Atkinson and Arey, 2003).
4 The following list of general reactions, analogous to those described for methane,
5 summarizes the role of alkane oxidation in tropospheric O3 formation.
6
8 .OH + RH^H20 + .R (AX2.34)
10 .R + 02 + M^R02- + M (AX2-35)
12 R02' + NO -> RO + N02 (AX2-36)
13
14 H02-+NO^-OH + N02 (AX2-23)
15
16 RO + 02 -» R'CHO + H02- (AX2-37)
17
18 2(NO2 + hv -»NO + O) (AX2-1)
19
20 2(O + O2 + M -» O3 + M) (AX2-2)
21
22 Net: RH + 4O2 + 2hv-> R'CHO + 2O3 + H2O (AX2-38)
23
24
25 The oxidation of alkanes can also be initiated by other oxidizing agents such as NO3 and Cl
26 radicals. In this case, there is net production of an OH radical which can re-initiate the oxidation
27 sequence. The reaction of OH radicals with aldehydes forms acyl (R'CO) radicals, and acyl
28 peroxy radicals (R'C(O)O2) are formed by the addition of O2. As an example, the oxidation of
29 ethane (C2H5-H) yields acetaldehyde (CH3-CHO). Acetyl (CH3-CO) and acetylperoxy
30 (CH3-C(O)O2) radicals can then be formed. Acetylperoxy radicals can combine with NO2 to
31 form peroxyacetyl nitrate (PAN) via:
32
CH3C(O)O2- + NO2 + M c^ CH3C(O)O2NO2 + M (AX2-39)
33
34 PAN can act as a temporary reservoir for NO2. Upon the decomposition of PAN, either locally
35 or elsewhere, NO2 is released to participate in the O3 formation process again. During the
January 2005 AX2-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 oxidation of propane, the relatively long-lived intermediate acetone (CH3 - C(O) CH3) is formed,
2 as shown in Figure AX2-3. The photolysis of acetone can be an important source of OH
3 radicals, especially in the upper troposphere (e.g., Singh et al., 1995). Examples of oxidation
4 mechanisms of more complex alkanes and other classes of hydrocarbons can be found in
5 comprehensive texts such as Seinfeld and Pandis (1998).
6
7 AX2.2.7 The Atmospheric Chemistry of Alkenes
8 As shown in Figure AX2-3, the presence of a double carbon-carbon bond, i.e., > C = C <,
9 in a VOC can greatly increase the range of potential reaction intermediates and products,
10 complicating the prediction of O3 production. The alkenes emitted from anthropogenic sources
11 are mainly ethene, propene, and the butenes, with lesser amounts of the > C5 alkenes. The major
12 biogenic alkenes emitted from vegetation are isoprene (2-methyl-1,3-butadiene) and C10H16
13 monoterpenes (Atkinson and Arey, 2003), and their tropospheric chemistry is currently the focus
14 of much attention (Zhang et al., 2002; Sauer et al., 1999; Geiger et al., 2003; Sprengnether et al.,
15 2002; Witter et al., 2002; Bonn and Moortgat, 2003; Berndt et al., 2003; Fick et al., 2003;
16 Kavouras et al., 1999; Atkinson and Arey, 2003).
17 Alkenes react in ambient air with OH and NO3 radicals and with O3. The mechanisms
18 involved in their oxidation have been discussed in detail by Calvert et al. (2000). All three
19 processes are important atmospheric transformation processes, and all proceed by initial addition
20 to the > C = C < bonds or, to a much lesser extent, by H atom extraction. Products of alkene
21 photooxidation include carbonyl compounds, hydroxy alkyl nitrates and nitratocarbonyls, and
22 decomposition products from the high energy biradicals formed in alkene-O3 reactions.
23 Table AX2-2 provides estimated atmospheric lifetimes for biogenic alkenes with respect to
24 oxidation by OH, NO3 and O3. The structures of most of the compounds gives in Table AX2-2
25 are shown in Figure AX2-4.
26 Uncertainties in the atmospheric chemistry of the alkenes concern the products and
27 mechanisms of their reactions with O3, especially the yields of OH radicals, H2O2, and secondary
28 organic aerosol in both outdoor and indoor environments. However, many product analyses of
29 important biogenic and anthropogenic alkenes in recent years have aided in the narrowing of
30 these uncertainties. The reader is referred to extensive reviews by Calvert et al.(2000) and
31 Atkinson and Arey (2003) for detailed discussions of these products and mechanisms.
January 2005 AX2-23 DRAFT-DO NOT QUOTE OR CITE
-------
a. Propane
H2
•OH + H3C' \
H3C CH +CH3
HO2« + CH2=O
CH3O- + NO2
b. Propene
H3C—C
H3C CH +H02*
Figure AX2-3. Hydroxyl radical initiated oxidation of a) propane and b) propene.
Source: Calvert et al. (2000).
January 2005
AX2-24 DRAFT-DO NOT QUOTE OR CITE
-------
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
January 2005
AX2-25
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX-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°
03C
>3.2yg
1.2dm
0.7 h°
NO3d
2.0yf
7.7 dn
9min°
aRate coefficients rom 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).
1 Oxidation by OH
2 As noted above, the OH radical reactions with the alkenes proceed mainly by OH radical
3 addition to the > C = C < bonds. As shown in Figure AX2-3, for example, the OH radical
4 reaction with propene leads to the formation of two OH-containing radicals. The subsequent
5 reactions of these radicals are similar to those of the alkyl radicals formed by H-atom abstraction
6 from the alkanes. Under high NO conditions, CH3CHCH2OH continues to react — producing
7 several smaller, "second generation," reactive VOCs.
8 For the simple
-------
1 For the more complex alkenes of biogenic origin, multiple products may be possible from
2 the initial oxidation step. Each product will further react, following a distinct degradation
3 pathway. Formaldehyde (HCHO), methacrolein (CH2 = C(CH3)-CHO) and methyl vinyl ketone
4 (CH3-C(O)-CH=CH2 have been identified as the major products of the OH-isoprene reaction.
5 These products also react with OH radicals and undergo photolysis. Yields of these
6 products(and others) are sensitive to the concentration of NOX used in laboratory experiments.
7 For NOX -100 ppt, methacrolein, methyl vinyl ketone and formaldehyde are formed with yields
8 of roughly 20%, 16%, and 33% and yields of other carbonyl compounds are about 17%, based
9 on the results of Ruppert and Becker (2000) and references therein. Ruppert and Becker also
10 observed much lower yields of C5-unsaturated diols (2 to 5%), methanol and methyl
11 hydroperoxide indicating the presence of peroxy radical interactions. For NOX ~1 ppb, the yields
12 of methacrolein are similar to those for NOX -100 ppb, but the yields of methyl vinyl ketone
13 (-33%), formaldehyde (-60%) are much higher and the diols, methanol and methyl
14 hydroperoxide were not observed. Orlando et al. (1999) found that the major products of the
15 oxidation of methacrolein were CO, CO2, hydroxyacetone, formaldehyde and
16 methacryloylperoxynitrate (MPAN) in their experiment. Horowitz et al. (1998) suggested that
17 isoprene may be the principal precursor of PAN over the United States in summer. Hydroperoxy
18 and organic peroxy radicals formed during the oxidation of isoprene and its products can oxidize
19 NO to NO2, initiating photochemical O3 formation. It should be noted that only about two-thirds
20 of the carbon in isoprene can be accounted for on a carbon atom basis for NOX > 1 ppb. The
21 values are much lower for lower NOX concentrations. The situation is much better for
22 methacrolein. Observed products can account for more than 90% of the reacted carbon.
23 The rates of formation of condensible, oxidation products of biogenic compounds that may
24 contribute to secondary organic aerosol formation is an important matter for the prediction of
25 ambient aerosol concentrations. Isoprene photooxidation has been shown to make a very small
26 contribution to secondary organic aerosol formation under atmospheric conditions (Pandis et al.,
27 1991; Zhang et al., 1992). Claeys et al. (2004) found that 2-methyltetrols are formed from the
28 oxidation of isoprene in yields of about 0.2% on a molar basis, or 0.4% on a mass basis. These
29 are semi-volatile compounds that can condense on existing particles. On the other hand, pinene
30 oxidation leads to substantial organic aerosol formation.
31
January 2005 AX2-27 DRAFT-DO NOT QUOTE OR CITE
-------
1 Oxidation by Nitrate Radical
2 NO3 radical reacts with alkenes mainly by addition to the double bond to form a
3 b-nitrooxyalkyl radical.(Atkinson 1991, 1994, 1997). The abstraction pathway may account for
4 up to 20% of the reaction. For propene, the initial reaction is followed by a series of reactions
5
6
7 NO3- + CH3CH=CH2 -» CH3CHCH2ONO2
(AX2-40)
8 -» CH3CH(ONO2)CH2
9
10
11 that (Atkinson, 1991) to lead to the formation of, among others, carbonyls and nitrato-carbonyls
12 including formaldehyde (HCHO), acetaldehyde (CH3CHO), 2-nitratopropanal
13 (CH3CH(ONO2)CHO), and 1-nitratopropanone (CH3C(O)CH2ONO2). By analogy to OH,
14 conjugated dienes like butadiene and isoprene will react with NO3 to form d-nitrooxyalkyl
15 radicals. (Atkinson, 2000). If NO3 is available for reaction in the atmosphere, then NO
16 concentrations will be low, owing to the rapid reaction between NO3 and NO. Consequently,
17 nitrooxyalkyl peroxy radicals are expected to react primarily with NO2, yielding thermally
18 unstable peroxy nitrates, NO3, HO2, and organoperoxy radicals (Atkinson, 2000).
19 Several studies have undertaken the quantification of the products of NO3 initiated
20 degradation of several of the important biogenic alkenes in O3 and secondary organic aerosol
21 formation, including isoprene, a- and b-pinene, 3-carene, limonene, linalool, and 2-methyl-3-
22 buten-2-ol. See Figure AX2-4 for the chemical structures of these and other biogenic
23 compounds. The results of these studies have been tabulated by Atkinson and Arey (2003).
24
25 Oxidation by Ozone
26 Unlike other organic compounds in the atmosphere, alkenes react at significant rates with
27 O3. Ozone initiates the oxidation of alkenes by addition across carbon-carbon double bonds, at
28 rates that are competitive with reaction with OH (see Table AX2-1). The addition of O3 across
29 the double bond yields an unstable ozonide, a 5-member ring including a single carbon-carbon
30 bond linked to the three oxygen atoms, each singly bound. The ozonide rearranges
31 spontaneously and
January 2005 AX2-28 DRAFT-DO NOT QUOTE OR CITE
-------
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).
January 2005
AX2-29 DRAFT-DO NOT QUOTE OR CITE
-------
1 then fragments to form an aldehyde or ketone, depending on the original position of the double
2 bond, and a high energy Criegee biradical. Collisional energy transfer may stabilize the radical,
3 preventing it from decomposing. Low pressure studies of the decomposition of the Criegee
4 biradical have shown high yields of the OH radical. At atmospheric pressures, the rates of OH
5 production have not been reliably established, due to complications arising from subsequent
6 reactions of the OH produced with the ozonide fragments (Calvert et al., 2000).
7 The ozonolysis of larger biogenic alkenes yields high molecular weight oxidation products
8 with sufficiently low vapor pressures to allow condensation into the particle phase. Many
9 oxidation products of larger biogenic alkenes have been identified in ambient aerosol,
10 eliminating their further participation in O3 production. Figure AX2-5 shows the chemical
11 structures of the oxidation products of a-pinene and illustrates the complexity of the products.
12 Carbonyl containing compounds are especially prevalent. A summary of the results of product
13 yield studies for several biogenic alkenes can be found in Atkinson and Arey (2003).
CHO
CHO
CHO
O-OH
C02H
-CHO
-CO2H
CO2H
CO2H
CO9H
CO2H
Figure AX2-5. Products from the reaction of terpenes with ozone.
Source: Atkinson and Arey (2003).
January 2005
AX2-30 DRAFT-DO NOT QUOTE OR CITE
-------
1 NO2 also participates to a very small degree in the oxidation of alkenes by addition to
2 double bonds in a manner similar to O3. Rate constants for reactions of this type range from
3 10"18 to 10"24 for dienes and monoalkenes (King et al., 2002). It should also be noted that O3
4 reacts with terpenoid compounds released from household products such as air fresheners and
5 cleaning agents in indoor air to produce ultrafine particles (Wainman et al., 2000; Sarwar et al.,
6 2002).
7
8 AX2.2.8 The Atmospheric Chemistry of Aromatic Hydrocarbons
9 Aromatic hydrocarbons represent a major class of compounds found in gasoline and other
10 liquid fuels. Upon vaporization, most of these compounds react rapidly in the atmosphere
11 (Davis et al., 1975) and following a series of complex processes, involving molecular oxygen
12 and oxides of nitrogen, produce O3. The reaction of OH radicals with aromatic hydrocarbons
13 serves as their major atmospheric loss process. Atmospheric losses of alkyl aromatic
14 compounds by O3 and nitrate radicals have been found to be minor processes for most
15 monocyclic aromatic hydrocarbons. (However, the reaction with of the nitrate radical with
16 substituted hydroxybenzenes, such as phenol or o-,m-,p-cresol, can be an important atmospheric
17 loss process for these compounds.) Much of the early work in this field focused on the
18 temperature dependence of the OH reactions (Perry et al., 1977; Tully et al., 1981) using
19 absolute rate techniques. Typically two temperature regions were observed for a large number
20 of aromatic compounds and the complex temperature profile suggested that two mechanisms
21 were operative. In the high temperature region, hydrogen (H)-atom abstraction from the
22 aromatic ring dominates, and in the temperature regime less than 320K, OH addition to the
23 aromatic ring is the dominant process. Thus, at normal temperatures and pressures in the lower
24 troposphere, ring addition is the most important reactive process followed by H-atom abstraction
25 from any alkyl substituents. The kinetics of monocyclic aromatic compounds are generally well
26 understood and there is generally broad consensus regarding the atmospheric lifetimes for these
27 compounds. By contrast, there is generally a wide range of experimental results from product
28 studies of these reactions. This leads to a major problem in model development due to a general
29 lack of understanding of the product identities and yields for even the simplest aromatic
30 compounds, which is due to the complex reaction paths following initial reaction with OH,
31 primarily by the addition pathway.
January 2005 AX2-31 DRAFT-DO NOT QUOTE OR CITE
-------
1 In the past three years, two comprehensive reviews have been written which provide a
2 detailed understanding of the current state-of-science of aromatic hydrocarbons. Atkinson
3 (2000) reviewed the atmospheric chemistry of volatile organic compounds, of which aromatic
4 hydrocarbons are included in one section of the review. More recently Calvert et al. (2002)
5 conducted a highly comprehensive examination of the reaction rates, chemical mechanisms,
6 aerosol formation, and contributions to O3 formation for monocyclic and polycyclic aromatic
7 hydrocarbons.
8
9 AX2.2.8.1 Chemical Kinetics and Atmospheric Lifetimes of Aromatic Hydrocarbons
10 Rate constants for the reaction of species in the atmosphere with aromatic hydrocarbons
11 vary widely depending on the number of aromatic rings and substituent groups. Reactions of O3
12 with aromatic hydrocarbons (AHCs) are generally slow except for monocyclic aromatic
13 hydrocarbons having unsaturated substituent groups. For example, indene and styrene have
14 atmospheric lifetimes of 3.3 h and 23 h with respect to reaction with O3, which are much longer
15 than that due to reactive loss with either OH or NO3. Thus, the atmospheric lifetimes and
16 reaction products of O3 and aromatic hydrocarbons will be ignored in this discussion. In
17 addition to chemical reaction, some organic compounds photolyze in the lower atmosphere.
18 Virtually all aromatic precursors are not subject to photolysis, although many of the ring
19 fragmentation products having multiple carbonyl groups can photolyze in the troposphere.
20 The reaction rates and atmospheric lifetimes of monocyclic aromatic compounds due to
21 reaction with OH radicals are generally dependent on the number and types of substituent groups
22 associated with the ring. These reaction rates have been found to be highly temperature and
23 pressure dependent. The temperature regimes are governed by the processes involved and show
24 a quite complex appearance. At room temperature (-300 K), both addition to the aromatic ring
25 and H-atom abstraction occur with the addition reaction being dominant. For the two smallest
26 monocyclic aromatic hydrocarbons, the initial addition adduct is not completely stabilized at
27 total pressures below 100 torr.
28 Numerous studies have been conducted to measure the OH + benzene rate constant over a
29 wide range of temperatures and pressures. An analysis of absolute rate data taken at
30 approximately 100 torr argon and not at the high pressure limit yielded a value of 1.2 x 10~12 cm3
31 molec"1 s"1. Atkinson (1989) recommended a value of 1.4 x 10"12 cn^molec"1 s"1 at room
January 2005 AX2-32 DRAFT-DO NOT QUOTE OR CITE
-------
1 temperature and atmospheric pressure. This recommendation has been refined only slightly and
2 is reflected in the recent value recommended by Calvert et al. (2002) which is given as
3 1.39 x 10"12 cm3 molec'V1. This recommended value for the reaction of OH + benzene together
4 with values for other monocyclic aromatic hydrocarbons is given in Table AX2-3.
5 In general, it is observed that the OH rate constants with monocyclic alkyl aromatic
6 hydrocarbons are strongly influenced by the number of substituent groups found on the aromatic
7 ring. (That is, the identity of the alkyl substituent groups has little influence on the overall
8 reactions rate constant.) Single substituent single-ring aromatic compounds which include
9 toluene, ethyl benzene, n-propylbenzene, isopropylbenzene, and t-butylbenzene have average
10 OH reaction rate constants ranging from 4.5 to 7.0 x 10"12 cm3 molec"1 s"1 at room temperature
11 and atmospheric pressure. These rate constants lead to atmospheric lifetimes (see below) that
12 are still greater than 1 day. Rate constants for monocyclic aromatic compounds with greater
13 than 10 carbon atoms or more are generally not available.
14 The dominant monocyclic aromatic compounds with two substituents are m-,o-,p-xylene.
15 Their recommended OH rate constants range from 1.4 to 2.4 x I0~n cm3 molec"1 s"1. Similarly,
16 the three isomers of ethyltoluene have recommended OH rate constants ranging from
17 1.2 tol.9 x 10~n cm3 molec"1 s"1. The only other two substituent single-ring aromatic compound
18 for which the OH rate constant has been measured is p-cymene (para-isopropyltoluene), giving a
19 value of 1.5 x io~n cm3 molec"1 s"1.
20 OH rate constants for the C9 trimethyl substituted aromatic hydrocarbons (1,2,3-; 1,2,4-;
21 1,3,5-trimethylbenzene) are higher by a factor of approximately 2.6 over the di-substituted
22 compounds. Rate constants for the three isomers range from 3.3 to 5.7 x I0~n cm3 molec"1 s"1.
23 While concentrations for numerous other trisubstituted benzene compounds have been reported
24 (e.g., l,2-dimethyl-4-ethylbenzene), OH rate constants for trimethylbenzene isomers are the only
25 tri substituted aromatic compounds that have been reported.
26 Aromatic hydrocarbons having substituent groups with unsaturated carbon groups have
27 much higher OH rate constants than their saturated analogues. The smallest compound in this
28 group is the C8 AHC, styrene. This compound reacts rapidly with OH and has a recommended
29 rate constant of 5.8 x I0~n cm3 molec"1 s"1. (Calvert, 2002). Other methyl substituted styrene-
30 type compounds (e.g., a-m ethyl styrene) have OH rate constants within a factor of two of that
31 with styrene. However, for unsaturated monocyclic aromatic hydrocarbons other processes
January 2005 AX2-33 DRAFT-DO NOT QUOTE OR CITE
-------
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
1 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.
January 2005
AX2-34
DRAFT-DO NOT QUOTE OR CITE
-------
1 including atmospheric removal by NO3 radicals can also be an important process, particularly at
2 night when photolysis does not substantially reduce the NO3 radical concentration (see below).
3 Polycyclic aromatic hydrocarbons are found to a much lesser degree in the atmosphere
4 than are the monocyclic aromatic hydrocarbons. For example, measurements made in Boston
5 during 1995 (Fujita et al., 1995) showed that a single PAH (napthalene) was detected in the
6 ambient morning air at levels of approximately 1% (C/C) of the total monocyclic aromatic
7 hydrocarbons. 1-methyl and 2-methylnaphthalene have sufficient volatility to be present in the
8 gas phase. Other higher molecular weight PAHs (< 3 aromatic rings), if present, are expected to
9 exist in the gas phase at much lower concentrations than napthalene and are not considered here.
10 OH rate constants for napthalene and the two methyl substituted napthalene compounds the two
11 have been reviewed by Calvert et al. (2002). The values recommended (or listed) by Calvert
12 et al. (2002) are given in Table AX2-3. As seen in the monocyclic aromatic hydrocarbons, the
13 substitution of methyl groups on the aromatic ring increases the OH rate constant, in this case by
14 a factor of 2.3.
15 Some data is available for the reaction of OH with aromatic oxidation products. (In this
16 context, aromatic oxidation products refer to those products which retain the aromatic ring
17 structure.) These include the aromatic carbonyl compound, benzaldehyde, 2,4-; 2,5-; and
18 3,4-dimethyl-benzaldehyde, and t-cinnamaldehyde. Room temperature rate constants for these
19 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
20 (t-cinnamaldehyde). While the yields for these compounds are typically between 2 to 6%, they
21 can contribute to the aromatic reactivity for aldehydes having high precursor concentration (e.g.,
22 toluene, 1,2,4-trimethylbenzene). OH also reacts rapidly with phenolic compounds. OH
23 reaction rates with phenols and o-, m-, and p-cresol are typically rapid (2.7 to 6.8 x I0~n cm3
24 molec"1 s"1) at room temperature. Five dimethylphenols and two trimethylphenols have OH
25 reaction rates ranging between 6.6 x I0~n and 1.25 x 10"10 cm3 molec"1 s"1. Finally, unlike the
26 aromatic aldehydes and phenols, reaction rates for OH + nitrobenzene and OH + m-nitrotoluene
27 are much lower than the parent molecules, given their electron withdrawing behavior from the
28 aromatic ring. The room temperature rate constants are 1.4 x 10"13 and 1.2 x 10~12, respectively.
29 The NO3 radical is also known to react with selected AHCs and aromatic photooxidation
30 products. Reaction can either occur by hydrogen atom abstraction or addition to the aromatic
31 ring. However, these reactions are typically slow for alkyl aromatic hydrocarbons and the
January 2005 AX2-3 5 DRAFT-DO NOT QUOTE OR CITE
-------
1 atmospheric removal due to this process is considered negligible. For AHCs having substituent
2 groups with double bonds (e.g., styrene, a-methylstyrene), the reaction is much more rapid, due
3 to the addition of NO3 to the double bond. For these compounds, NO3 rate constants are on the
4 order of 10"12 cm3 molec"1 s"1. This leads to atmospheric lifetimes on the order of about 1 h for
5 typical night time atmospheric NO3 levels of 2.5 x 108 molec cm"3 (Atkinson, 2000).
6 The most important reactions of NO3 with AHCs are those which involve phenol and
7 methyl, dimethyl, and trimethyl analogs. These reactions can be of importance due to the high
8 yields of phenol for the atmospheric benzene oxidation and o-,m-,p-cresol from toluene
9 oxidation. The NO3 + phenol rate has been given as 3.8 x 10~12 cm3 molec"1 s"1. Similarly, the
10 cresol isomers each has an extremely rapid reaction rate with NO3 ranging from 1.1 to 1.4 x
11 10~n cm3 molec"1 s"1. As a result, these compounds, particularly the cresol isomers, can show
12 rapid nighttime losses due to reaction with NO3 with nighttime lifetimes on the order of a few
13 minutes. There is little data for the reaction of NO3 with dimethylphenols or trimethylphenols
14 which have been found as products of the reaction of OH + m-, p-xylene and OH + 1,2,4-; 1,3,5-
15 trimethylbenzene.
16
17 AX2.2.8.2 Reaction Products and Mechanisms of Aromatic Hydrocarbon Oxidation
18 An understanding of the mechanism of the oxidation of AHCs is important 1 if O3 is to be
19 accurately predicted in urban atmospheres through modeling studies. As noted above, most
20 monocyclic aromatic hydrocarbons are removed from the atmosphere through reaction with OH.
21 Thus, product studies of the OH + AHC should provide the greatest information regarding the
22 AHC oxidation products. However, the effort to study these reactions has been intractable over
23 the past two decades due to a number of difficulties inherent in the OH-aromatic reaction
24 system. There are several reasons for the slow progress in understanding these mechanisms.
25 (1) Product yields for OH-aromatic systems are poorly understood; for the most studied system,
26 OH-toluene, approximately 50% of the reaction products have been identified under conditions
27 where NO2 reactions do not dominate the removal of the OH-aromatic adduct. (2) As noted, the
28 reaction mechanism can change as the ratio of NO2 to O2 changes in the system (Atkinson and
29 Aschmann, 1994). Thus, reaction product distributions that may be measured in the laboratory
30 at high NO2 (or NOX) concentrations may not be applicable to atmospheric conditions. This also
31 limits the usefulness of models to predict O3 formation to the extent that secondary aromatic
January 2005 AX2-36 DRAFT-DO NOT QUOTE OR CITE
-------
1 reactions are not completely parameterized in the system. (3) Aromatic reactions produce highly
2 polar compounds for which there are few calibration standards available. In most cases,
3 surrogate compounds have to be used in GC/MS calibrations. Moreover, it is not at all clear
4 whether the present sampling techniques or analytical instruments are appropriate to measure the
5 highly polar products produced in these systems. (4) Finally for benzene and toluene in
6 particular, reaction rates of the products are substantially faster than that of the parent
7 compounds. Thus, it is difficult to measure yields accurately without substantial interferences
8 due to secondary reactions. Even given these difficulties, over the past decade a body of
9 knowledge has been developed whereby the initial steps in the OH-initiated photooxidation have
10 been established and a wide range of primary products from each of the major reaction systems
11 have been catalogued.
12 Benzene is one of the most important aromatic hydrocarbons released into the atmosphere
13 and is a recognized carcinogen. However, its reaction with OH is extremely slow and its
14 contribution to urban O3 formation is generally recognized to be negligible (Carter, 1994). As a
15 result, relatively few studies have been conducted on the OH reaction mechanism of benzene.
16 Major products of the oxidation of benzene have been found to be phenol and glyoxal (Berndt
17 et al., 1999; Tuazon et al., 1986).
18 Most of the product analysis and mechanistic work on alkyl aromatic compounds in the gas
19 phase has focused on examining OH reactions with toluene. The primary reaction of OH with
20 toluene follows either of two paths, the first being an abstraction reaction from the methyl group
21 and the second being addition to the ring. It has previously been found that H-atom abstraction
22 from the aromatic ring is of minor importance (Tully et al., 1981). A number of studies have
23 examined yields of the benzyl radical formed following OH abstraction from the methyl group.
24 This radical forms the benzyl peroxy radical, which reacts with nitric oxide (NO) leading to the
25 stable products benzaldehyde, with an average yield of 0.06, and benzyl nitrate, with an average
26 yield less than 0.01 (Calvert et al., 2002). Thus, the overall yield for the abstraction channel is
27 less than approximately 7%.
28 It is now generally recognized that addition of OH to the aromatic ring is the major process
29 removing toluene from the atmosphere and appears to account for more than 90% of the reaction
30 yield for OH + toluene. The addition of OH to the ring leads to an intermediate OH-toluene
31 adduct that can be stabilized or can redissociate to the reactant compounds. For toluene, OH
January 2005 AX2-37 DRAFT-DO NOT QUOTE OR CITE
-------
1 addition can occur at any of the three possible positions on the ring (ortho, meta, or para) to form
2 the adduct. Addition of OH to the toluene has been shown to occur predominately at the ortho
3 position (yield of 0.81) with lesser amounts at the meta (0.05) and para (0.14) positions (Kenley
4 et al., 1981). The initial steps for both the abstraction and addition pathways in toluene have
5 been shown in Figure AX2-6; only the path to form the ortho-adduct is shown, viz. reaction (2).
6 The OH-toluene adduct formed is an energy-rich intermediate that must be stabilized by
7 third bodies in the system to undergo further reaction. Stabilization has been found to occur at
8 pressures above 100 Torr for most third bodies (Perry et al., 1977; Tully et al., 1981). Therefore,
9 at atmospheric pressure, the adduct will not substantially decompose back to its reactants as
10 indicated by reaction (-2). The stabilized adduct (I) is removed by one of three processes:
11 H-atom abstraction by O2 to give a cresol, as in reaction (5); an addition reaction with O2, as in
12 reaction (6); or reaction with NO2 to give m-nitrotoluene as in reaction (7).
13 The simplest fate for the adduct (I) is reaction with O2to form o-cresol. Data from a
14 number of studies (e.g., Kenley et al., 1981; Atkinson et al., 1980; Smith et al., 1998; Klotz
15 et al., 1998; summarized by Calvert et al., 2002) over a wide range of NO2 concentrations
16 (generally above 1 ppmv) show an average yield of approximately 0.15 for o-cresol. Most of the
17 measurements suggest the o-cresol yield is independent of total pressure, identity of the third
18 body, and NO2 concentration (Atkinson and Aschmann, 1994; Moschonas et al., 1999), but the
19 data tend to be scattered. This finding suggests that the addition of NO2 to the hydroxy
20 methylcyclo-hexadienyl radical does not contribute to the formation of phenolic-type
21 compounds. Fewer studies have been conducted for m- and p-cresol yields, but the results of
22 two studies indicate the yield is approximately 0.05 (Atkinson et al., 1980; Gery et al., 1985;
23 Smith et al., 1998). The data suggests good agreement between the relative yields of the cresols
24 from the product studies at atmospheric pressure and studies at reduced pressures. Thus, H-atom
25 abstraction from adducts formed at all positions appears to represent approximately 20% of the
26 total yield for toluene.
27 The OH-toluene adduct also reacts with O2 to form a cyclohexadienyl peroxy radical (III),
28 shown as a product of reaction (6) after rearrangement. This radical can undergo a number of
29 possible processes. Most of these processes lead to ring fragmentation products, many of which
30 have been seen in several studies (Dumdei and O'Brien, 1984; Shepson et al., 1984). Ring-
31 fragmentation products are frequently characterized by multiple double bonds and/or multiple
January 2005 AX2-3 8 DRAFT-DO NOT QUOTE OR CITE
-------
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.
January 2005
AX2-39 DRAFT-DO NOT QUOTE OR CITE
-------
1 functional groups. As such, these products are highly reactive and extremely difficult to detect
2 and quantify.
3 Klotz et al. (1997; 1998) have suggested that the intermediate could also follow through a
4 mechanism where toluene oxide/oxepin could be formed following the addition of O2 to the
5 OH-aromatic adduct. Recent experiments suggest that the formation of o-cresol through the
6 photolysis of toluene oxide/oxepin is only a minor contributor to the overall o-cresol which has
7 been measure (Klotz et al., 1998). This result contrasts to the high yield observed for the
8 formation of phenol from the photolysis of benzene oxide/oxepin (Klotz et al., 1997). Recently,
9 Berndt et al. (1999) used a flow tube to test the hypothetical formation of benzene oxide/oxepin
10 from the OH + benzene reaction at pressures below 100 torr. They saw very little evidence for
11 its formation.
12 A few studies have been conducted to identify fragmentation products using a variety of
13 instruments. Several approaches have been used that employ structural methods, particularly
14 mass spectrometry (MS), to identify individual products formed during the photooxidation.
15 In one approach (Dumdei and O'Brien, 1984), the walls of the reaction chamber were extracted
16 following an extended irradiation. In this study, the analysis was conducted by tandem mass
17 spectrometry (MS/MS), which allowed products to be separated without the use of a
18 chromatographic stationary phase. The investigators reported 27 photooxidation products from
19 toluene, with 15 reportedly from ring fragmentation processes. However, the study was purely
20 qualitative and product yields could not be obtained. No distinction could be made between
21 primary and secondary products from the reaction because extended irradiations and species in
22 various isotopic forms could not be differentiated. More refined approaches using atmospheric
23 pressure ionization-tandem mass spectrometry has been used to study toluene (Dumdei et al.,
24 1988) and m- and p-xylene (Kwok et al., 1997) photooxidation.
25 In another study, Shepson et al. (1984) demonstrated that a number of these fragmentation
26 products could be analyzed by gas chromatography. Fragmentation products detected by the
27 both investigations (Dumdei and O'Brien, 1984; Shepson et al., 1984) included glyoxal, methyl
28 glyoxal, butenedial, 4-oxo-2-pentenal, hydroxybutenedial, l-pentene-3,4-dione, l-butene-3,4-
29 dione, and methyl vinyl ketone. Additional evidence (Shepson et al., 1984) for fragmentation
30 processes came from the detection of 2-methylfuran and furfural. These compounds, although
31 cyclic in structure, result from a bridged oxygen intermediate. Yields of the detected
January 2005 AX2-40 DRAFT-DO NOT QUOTE OR CITE
-------
1 fragmentation products were subsequently measured in a number of studies (e.g., Bandow et al.,
2 1985a,b; Tuazon et al., 1986; Smith et al., 1998), were typically under 15% on a reacted carbon
3 basis.
4 An additional possible pathway for reaction of the OH-toluene adduct is by reaction with
5 NO2 to give isomers of nitrotoluene. A yield of approximately 0.015 at NO2 concentrations of
6 about 1 ppmv has been measured (Atkinson et al., 1991). Although this yield itself is fairly
7 minor, the investigators reported a positive intercept in plotting the nitrotoluene concentration
8 against the NO2 concentration; however, the data were considerably scattered. The positive
9 intercept has been interpreted as suggesting that the OH-toluene adduct does not add O2. This
10 finding would require, therefore, another mechanism than that described above to be responsible
11 for the fragmentation products.
12 The results of this study can be compared to experiments which directly examined the OH
13 radical loss in reactions of OH with toluene and other aromatic compounds. Knipsel and co-
14 workers (Knispel et al., 1990) have found a double exponential decay for toluene loss in the
15 presence of added O2, a rapid decay reflective of the initial adduct formation and a slower decay
16 reflecting loss of the adduct by O2 or other scavengers. From the decay data in the presence of
17 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
18 constant suggests that the loss rate of 2500 s"1 for the adduct in the presence of air at
19 atmospheric pressure. This loss rate compares to a loss due to NO2 (with a nominal atmospheric
20 concentration of 0.1 ppmv) of about 100 s"1. This finding suggests that removal of the OH-
21 toluene adduct by O2 is a far more important loss process than removal by NO2 under
22 atmospheric conditions which is in contrast other findings (Atkinson et al., 1991). This finding
23 was confirmed by the recent experiments from Moschonas et al. (1999).
24 Therefore, studies on the disposition of toluene following OH reaction can be summarized
25 as follows. It is generally accepted that H-atom abstraction from the methyl group by OH is a
26 relatively minor process accounting for a 6 to 7% yield in the OH reaction with toluene.
27 Addition of OH to toluene to form an intermediate OH-toluene adduct is the predominant
28 process. At atmospheric pressure, ring-retaining products such as the cresol and nitrotoluenes
29 account for another 20% of the primary reaction products (Smith et al., 1999). The remaining
30 70 to 75% of the products are expected to be ring fragmentation products in the gas phase,
31 having an uncertain mechanism for formation. Many of these fragmentation products have been
January 2005 AX2-41 DRAFT-DO NOT QUOTE OR CITE
-------
1 detected, but appear to form at low yields, and relatively little quantitative information on their
2 formation yields exists. As noted earlier, some of these products contain multiple double bonds,
3 which are likely to be highly reactive with OH or photolyze which enhances the reactivity of
4 systems containing aromatics. Mechanisms that cannot adequately reflect the formation of
5 fragmentation products are likely to show depressed reactivity for the oxidation of toluene and
6 other aromatic compounds.
7 The number of studies of the multiple-substituted alkyl aromatics, such as the xylenes or
8 the trimethylbenzenes, is considerably smaller than for toluene. Kinetic studies have focused on
9 the OH rate constants for these compounds. For the xylenes, this rate constant is typically a
10 factor of 2 to 5 greater than that for OH + toluene. Thus, the OH reactivity of the fragmentation
11 products is similar to that of the parent compounds, potentially making the study of the primary
12 products of the xylenes less prone to uncertainties from secondary reactions of the primary
13 products than is the case for toluene.
14 Products from the OH reaction with the three xylenes have been studied most
15 comprehensively in a smog chamber using long-path FTIR (Bandow and Washida, 1985a) and
16 gas chromatography (Shepson et al., 1984; Atkinson and Aschmann, 1994; Smith et al., 1999).
17 Ring-fragmentation yields of 41, 55, and 36% were estimated for o-, m-, and p-xylene,
18 respectively, based on the dicarbonyl compounds, glyoxal, methyl glyoxal, biacetyl, and 3-
19 hexene-2,5-dione detected during the photooxidation. These values could be lower limits, given
20 that Shepson et al. (1984) report additional fragmentation products from o-xylene, including
21 l-pentene-4,5-dione, butenedial, 4-oxo-pentenal, furan, and 2-methylfuran. In the earlier
22 studies, aromatic concentrations were in the range of 5 to 10 ppmv with NOX at 2 to 5 ppmv.
23 At atmospheric ratios of NO2 and O2, the observed yields could be different. Smith et al. (1999)
24 examined most of the ring retaining products in the OH + m-xylene and OH + p-xylene systems.
25 In each case, tolualdehyde isomers, dimethylphenol isomers, and nitro xylene isomers specific
26 for each system were detected. The total ring retaining yield for OH + m-xylene was 16.3%; the
27 yield for OH + p-xylene it was 24.5%. A mass balance approach suggests that respective ring-
28 fragmentation yields of 84% and 76%, respectively. Kwok et al. (1997) also measured products
29 from the OH + m- and p-xylene systems using atmospheric pressure ionization-tandem mass
30 spectrometry. Complementary ring-fragmentation products to glyoxal, methylglyoxal, and
January 2005 AX2-42 DRAFT-DO NOT QUOTE OR CITE
-------
1 biacetyl were detected from the parent ion peaks, although the technique did not permit the
2 determination of reaction yields.
3 Smith et al. (1999) also studied ring fragmentation products from the reaction of OH
4 with 1,2,4- and 1,3,5-trimethylbenzene. Ring-retaining products from the reaction with
5 1,2,4-trimethylbenzene gave three isomers each of dimethylbenzaldehyde and trimethylphenol
6 as expected by analogy with toluene. However, the ring-retaining products only accounted for
7 5.8% of the reacted carbon. Seven additional ring-fragmentation products were also detected
8 from the reaction, although the overall carbon yield was 47%. For 1,3,5-trimethylbenzene, its
9 reaction with OH leads to only two ring-retaining products, 3,5-dimethyl-benzaldehyde and
10 2,4,6-trimethylphenol, given its molecular symmetry. Only a single fragmentation product was
11 detected, methyl glyoxal, at a molar yield of 90%. The overall carbon yield in this case was
12 61%. The formation of relatively low yields of aromatic aldehydes and methylphenols suggests
13 that NOX removal by these compound in these reaction systems will be minimized (see below).
14 In recent years, computational chemistry studies have been applied to reaction dynamics of
15 the OH-aromatic reaction systems. Bartolotti and Edney (1995) used density functional-based
16 quantum mechanical calculations to help identify intermediates of the OH-toluene adduct. These
17 calculations were consistent with the main addition of OH to the ortho position of toluene
18 followed by addition of O2 to the meta position of the adduct. The reaction energies suggested
19 the formation of a carbonyl epoxide which was subsequently detected in aromatic oxidation
20 systems by Yu and Jeffries, (1997). Andino et al. (1996) conducted ab initio calculations using
21 density functional theory with semiempirical intermediate geometries to examine the energies of
22 aromatic intermediates and determine favored product pathways. The study was designed to
23 provide some insight into the fragmentation mechanism, although only a group additivity
24 approach to calculate AH^ was used to investigate favored reaction pathways. However, the
25 similarity in energies of the peroxy radicals formed from the O2 reaction with the OH-aromatic
26 adduct were very similar in magnitude making it difficult to differentiate among structures.
27 A detailed analysis of toluene oxidation using smog chamber experiments and chemical
28 models (Wagner et al., 2003) shows that there are still large uncertainties in the effects of
29 toluene on O3 formation. A similar situation is likely to be found for other aromatic
30 hydrocarbons.
31
January 2005 AX2-43 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX2.2.8.3 The Formation of Secondary Organic Aerosol as a Sink for Ozone Precursors
2 Aromatic hydrocarbons are known to generate secondary organic aerosol (SOA) following
3 their reaction with OH or other reactive oxidants. Secondary organic aerosol refers to the
4 formation of fine particulate matter either through nucleation processes or through condensation
5 onto existing particles. Over the last 12 years numerous experiments have been conducted in
6 environmental chambers to determine the yield of secondary organic aerosol as a function of the
7 reacted aromatic hydrocarbon. A review of the results of these studies can be found in the latest
8 Air Quality for Parti culate Matter Document (U.S. Environmental Protection Agency, 2003).
9 The extent to which aromatic reaction products are removed from the gas phase and
10 become incorporated in the particle phase will influence the extent to which oxygenated organic
11 compounds will not be available for participation in the aromatic mechanisms that lead to O3
12 formation. However, this may be overstated to some degree for products of aromatic precursors.
13 First, at atmospheric loading levels of organic parti culate matter, the SOA yields of the major
14 aromatic hydrocarbons are in the low percent range. Second, the aromatic products that are
15 likely to condense on particles are likely to be highly oxygenated and have OH reaction rates
16 that make them largely unreactive. Thus, while there may be some reduction of O3 formation, it
17 is not expected to be large.
18
19 AX2.2.9 Importance of Oxygenated VOCs
20 The role of oxygenated VOCs in driving O3 production has generated increasing interest
21 over the past decade. These VOCs include carbonyls, peroxides, alcohols, and organic acids.
22 They are produced in the atmosphere by oxidation of hydrocarbons, as discussed above, but are
23 also directly emitted to the atmosphere, in particular by vegetation (Guenther et al., 2000).
24 In rural and remote atmospheres, oxygenated VOCs often dominate over nonmethane
25 hydrocarbons in terms of total organic carbon mass and reactivity (Singh et al., 2004). The most
26 abundant by mass of these oxygenated VOCs is usually methanol, which is emitted by
27 vegetation and is present in U.S. surface air at concentrations of typically 1-10 ppbv (Heikes
28 et al., 2002).
29 Most oxygenated VOCs react with OH to drive ozone production in a manner similar to the
30 hydrocarbon chemistry discussed in the previous sections. In addition, carbonyl compounds
31 (aldehydes and ketones) photolyze to produce peroxy radicals that can accelerate ozone
January 2005 AX2-44 DRAFT-DO NOT QUOTE OR CITE
-------
1 production, thus acting as a chemical amplifier (Jaegle et al., 2001). Photolysis of formaldehyde
2 by (A.26b) was discussed in section AX2.2.5. Also of particular importance is the photolysis of
3 acetone (Blitz et al., 2004):
4
?O
(CH3)2C(O) + hv ?—» CH3C(O)O2- (AX2-41)
5
6 producing organic peroxy radicals that subsequently react with NO to produce O3. The
7 peroxyacetyl radical CH3C(O)OO can also react with NO2 to produce PAN, as discussed in
8 Section AX2.2.4. Photolysis of acetone are a minor but important source of HO2 radicals in the
9 upper troposphere (Arnold et al., 2004).
10
11 AX2.2.10 Influence of Multiphase Chemical Processes
12 In addition to reactions occurring in the gas phase, reactions occurring on the surfaces of or
13 within cloud droplets and airborne particles also occur. Their collective surface area is huge,
14 implying that collisions with gas phase species occur on very short time scales. The integrated
15 aerosol surface area ranges from 4.2 x 10~7 cm2/cm3 for clean continental conditions to
16 1.1 x 10~5 cm2/cm3 for urban average conditions (Whitby, 1978). There have been substantial
17 improvements in air quality especially in urban areas since the time these measurements were
18 made and so the U.S. urban values should be scaled downward by roughly a factor of two to
19 four. The resulting surface area is still substantial and the inferred collision time scale of a
20 gaseous molecule with a particle ranges from a few seconds or less to a few minutes. These
21 inferred time scales imply that heterogenous reactions will generally be much less important than
22 gas phase reactions for determining radical concentrations especially when reaction probabilities
23 much less then unity are considered. A large body of research has accumulated recently
24 regarding chemical processes in cloud droplets, snow and ice crystals, wet (deliquesced)
25 inorganic particles, mineral dust, carbon chain agglomerates and organic carbon-coated particles.
26 Jacob's (2000) comprehensive review of the potential influences of clouds and aerosols on
27 tropospheric O3 cycling provides the starting point for this section. Updates to that review will
28 also be provided. Jacob's review evaluates the literature available through late 1999, discusses
29 major areas of uncertainty, recommends experiments to reduce uncertainties, and (based on then
January 2005 AX2-45 DRAFT-DO NOT QUOTE OR CITE
-------
1 current information) recommends specific multiphase pathways that should be considered in
2 models of O3 cycling. In regard to the latter, Jacob's recommendations should be viewed as
3 conservative. Specifically, only reasonably well-constrained pathways supported by strong
4 observational evidence are recommended for inclusion in models. Several poorly resolved
5 and/or controversial pathways that may be significant in the ambient troposphere lack sufficient
6 constraints for reliable modeling. Some of these areas are discussed in more detail below.
7 It should be noted at the outset that many of the studies described in this section involve either
8 aerosols that are not found commonly throughout the United States (e.g., marine aerosol) or
9 correspond to unaged particles (e.g., soot, mineral dust). In many areas of the United States,
10 particles accrete a layer of hydrated H2SO4 which will affect the nature of the multiphase
11 processes occurring on particle surfaces.
12 Major conclusions from this review are summarized as follows (comments are given in
13 parentheses):
14 HOX Chemistry
15 (1) Catalytic O3 loss via reaction of O2" + O3(aq) in clouds appears to be inefficient.
16 (2) Aqueous-phase loss of HCHO in clouds appears to be negligible (see also [Lelieveld
and Crutzen 1990]).
17 (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.
18 (4) The uptake of alkyl peroxy radicals by aerosols is probably negligible.
19 (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 icr3 is recommended.
20 NOX Chemistry
21 (6) Hydrolysis of N2O5 to HNO3 in aqueous aerosols is important (Section AX2.2.4)
(and can be parameterized with YnNo3 = 0-01 to 0.1, Schutze and Herrmann, 2002;
Hallquist et al., 2003).
22 (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.
January 2005 AX2-46 DRAFT-DO NOT QUOTE OR CITE
-------
1 (8) There is no evidence for significant multiphase chemistry involving PAN
2 (9) There is no evidence for significant conversion of HNO3 to NOX in aerosols.
3 Heterogeneous ozone loss
4 (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).
5 Halogen radical chemistry
6 (11) There is little justification for considering BrOx and C1OX chemistry (except perhaps
in limited areas of the United States and nearby coastal areas).
7 Most of the above conclusions remain valid but, as detailed below, some should be
8 qualified based on recently published findings and on reevaluation of results form earlier
9 investigations.
10
11 AX2.2.10.1 HOX and Aerosols
12 Field measurements of HOX reviewed by Jacob (2000) correspond to regions with
13 relatively low aerosol concentrations (e.g., Mauna Loa [Cantrell et al., 1996]; rural Ontario
14 [Plummer et al., 1996]; and the upper troposphere [Jaegle et al., 1999]). In all cases, however,
15 significant uptake of HO2 or HO2 + RO2 radicals by aerosols was inferred based on imbalances
16 between measured concentrations of peroxy radicals and photochemical models of gas-phase
17 chemistry. Laboratory studies using artificial aerosols (both deliquesced and solid) confirm
18 uptake but the actual mechanism remains unclear. Several investigations report significant HOX
19 and H2O2 production in cloud water (e.g., Anastasio et al., 1994). However the potential
20 importance of this source is considered unlikely because measurements in continental air show
21 no evidence of missing sources for HOX or H2O2. No investigations involving the potential
22 influences of marine aerosols as sources or sinks for HOX were considered in the above analysis.
23 Relative to conservative seawater tracers such as Mg2+ and Na+, organic C associated with
24 sea-salt aerosols is typically enriched by 2 to 3 orders of magnitude in both polluted (e.g.,
25 Hoffman and Duce, 1976, 1977; Turekian et al., 2003) and remote regions (Chesselet et al.,
26 1981). This organic C originates from three major sources: 1) organic surfactants concentrated
27 from bulk seawater on walls of subsurface bubbles (Tseng et al., 1992), 2) the surface microlayer
28 of the ocean (Gershey, 1983), and 3) condensation of organic gases (Pun et al., 2000).
January 2005 AX2-47 DRAFT-DO NOT QUOTE OR CITE
-------
1 Coagulation of chemically distinct aerosols (e.g., via cloud processing) may also contribute
2 under some conditions.
3 Resolving chemical processes involving particles in the marine boundary layer (MBL) is
4 constrained by the relative scarcity of measurements of particulate organic carbon (POC)
5 (Penner, 1995) and its molecular composition (Saxena et al., 1995). In MBL regions impacted
6 by direct continental outflow, POC may constitute more that half of the total dry aerosol mass
7 (Hegg et al., 1997). Carbon isotopic compositions in the polluted North Atlantic MBL indicate
8 that, on average, 35% to 40% of POC originates from primary (direct injection) and secondary
9 (condensation of gases) marine sources (Turekian et al., 2003).
10 The photolysis of dissolved organic compounds is a major source for OH, H2O2, and
11 C-centered radicals in both the surface ocean (e.g., Blough and Zepp, 1995; Blough, 1997;
12 Mopper and Kieber, 2000) and in marine aerosols (e.g., McDow et al., 1996). Relative to the
13 surface ocean, however, production rates in the aerosol are substantially greater per unit volume
14 because organic matter is highly enriched (Turekian et al., 2003) and aerosol pH is much lower
15 (Keene et al., 2002a). Lower pHs increase rates of many reactions including acid-catalyzed
16 pathways such as the breakdown of the HOC1" radical (King et al., 1995), the formation of H2O2
17 from the photolysis of phenolic compounds (Anastasio et al., 1997), and the photolysis of organic
18 acids.
19 To provide a semi-quantitative context for the potential magnitude of this source, we
20 assume a midday OH production rate in surface seawater of 10~n M sec"1 (Zhou and Mopper,
21 1990) and a dissolved organic carbon enrichment of 2 to 3 orders of magnitude in sea-salt
22 aerosols. This yields an estimated OH production rate in fresh (alkaline) sea-salt aerosols of 10~9
23 to 10~8 M sec"1. As discussed above, rapid (seconds to minutes) acidification of the aerosol
24 should substantially enhance these production rates. Consequently, the midday OH production
25 rates from marine-derived organic matter in acidified sea-salt aerosols may rival or perhaps
26 exceed midday OH scavenging rates from the gas phase (approximately 10~7 M sec"1; [Chameides
27 and Stelson, 1992]). Scavenging is the only significant source for OH in acidified sea-salt
28 aerosols considered by many current models.
29 Limited experimental evidence indicates that these pathways are important sources of HOX
30 and ROX in marine air and possibly in coastal cities. For example, the absorption of solar energy
31 by organic species dissolved in cloud water (e.g., Faust et al., 1993; Anastasio et al., 1997) and in
January 2005 AX2-48 DRAFT-DO NOT QUOTE OR CITE
-------
1 deliquesced sea-salt aerosols (Anastasio et al., 1999) produces OH, HO2, and H2O2. In addition,
2 Fe(III) complexation by oxalate and similar ligands to metal such as iron can greatly enhance
3 radical production through ligand to metal charge transfer reactions (Faust, 1994; Hoigne et al.,
4 1994). Oxalate and other dicarboxylic anions are ubiquitous components of MBL aerosols in
5 both polluted (e.g., Turekian et al., 2003) and remote regions (Kawamura et al., 1996).
6 Substantial evidence exists for washout of peroxy radicals. Near solar noon, mixing ratios
7 of total HOX plus ROX radicals generally fall in the 50 ppt range, but during periods of rain these
8 values dropped to below the detection limit of 3 to 5 ppt (Andres-Hernandez et al., 2001; Burkert
9 et al., 2001a; Burkert et al., 2001b; Burkert et al., 2003). Such low concentrations cannot be
10 explained by loss of actinic radiation, because nighttime radical mixing ratios were higher.
11 Burkert et al. (2003) investigated the diurnal behavior of the trace gases and peroxy radicals
12 in the clean and polluted MBL by comparing observations to a time dependant, zero-dimensional
13 chemical model. They identified significant differences between the diurnal behavior of RO2*
14 derived from the model and that observed possibly attributable to multiphase chemistry. The
15 measured HCHO concentrations differed from the model results and were best explained by
16 reactions involving low levels of Cl.
17 Finally, photolytic NO3 reduction is important in the surface ocean (Zafiriou and True,
18 1979) and could contribute to OH production in sea-salt aerosols. Because of the
19 pH-dependence of HNO3 phase partitioning, most total nitrate (HNO3 + particulate NO3 ) in
20 marine air is associated with sea salt (e.g., Huebert et al., 1996; Erickson et al., 1999). At high
21 mM concentrations of NO3 in sea-salt aerosols under moderately polluted conditions (e.g.,
22 Keene et al., 2002) and with quantum yields for OH production of approximately 1% (Jankowski
23 et al., 2000), this pathway would be similar in magnitude to that associated with scavenging
24 from the gas phase and with photolysis of dissolved organics. Experimental manipulations of
25 marine aerosols sampled under relatively clean conditions on the California coast confirms that
26 this pathway is a major source for OH in sea-salt solutions (Anastasio et al., 1999).
27 Although largely unexplored, the potential influences of these poorly characterized radical
28 sources on O3 cycling in marine air are probably significant. At minimum, the substantial
29 inferred concentrations of HO2 in aerosol solutions would diminish and perhaps reverse HO2
30 scavenging by marine aerosols and thereby increase O3 production relative to models based on
31 Jacob's (2000) recommended reaction probability.
January 2005 AX2-49 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX2.2.10.2 NOX Chemistry
2 Jacob (2000) recommended as a best estimate, YN o = 0.1 for the reaction probability of
3 N2O5 on aqueous aerosol surfaces with conversion to HNO3. Recent laboratory studies on
4 sulfate and organic aerosols indicates that this reaction probability should be revised downward,
5 to a range 0.01-0.05 (Kane et al., 2001; Hallquist et al., 2003; Thornton et al., 2003). Tie et al.
6 (2003) found that a value of 0.04 in their global model gave the best simulation of observed NOX
7 concentrations over the Arctic in winter. A decrease in N2O5 slows down the removal of NOX
8 and thus increases O3 production. Based on the consistency between measurements of NOy
9 partitioning and gas-phase models, Jacob (2000) considers it unlikely that significant HNO3 is
10 recycled to NOX in the lower troposphere. However, only one of the reviewed studies (Schultz
11 et al., 2000) was conducted in the marine troposphere and none were conducted in the MBL.
12 An investigation over the equatorial Pacific reported discrepancies between observations and
13 theory (Singh et al., 1996b) that might be explained by HNO3 recycling. It is important to
14 recognize that both Schultz et al. (2000) and Singh et al. (1996b) involved aircraft sampling
15 which, in the MBL, significantly under represents sea-salt aerosols and thus most total NO3
16 (HNO3 + NO3 ) and large fractions of NOy in marine air (e.g., Huebert et al., 1996).
17 Consequently, some caution is warranted when interpreting constituent ratios and NOy budgets
18 based on such data.
19 Recent work in the Arctic has quantified significant photochemical recycling of NO3 to
20 NOX and perturbations of OH chemistry in snow (Honrath et al., 2000; Dibb et al., 2002; Domine
21 and Shepson, 2002), which suggests the possibility of similar multiphase pathways occurring in
22 aerosols. As mentioned above, recent evidence also indicates that NO3 is photolytically reduced
23 to NO2" (Zafariou and True, 1979) in acidic sea-salt solutions (Anastasio et al., 1999). Further
24 photolytic reduction of NO2" to NO (Zafariou and True, 1979) could provide a possible
25 mechanism for HNO3 recycling. Early experiments reported production of NOX during the
26 irradiation of artificial seawater concentrates containing NO3 (Petriconi and Papee, 1972).
27 Based on the above, we believe that HNO3 recycling in sea-salt aerosols is potentially important
28 and warrants further investigation. Other possible recycling pathways involving highly acidic
29 aerosol solutions and soot are reviewed by Jacob (2000).
30 Ammann et al. (1998) reported the efficient conversion of NO2 to HONO on fresh soot
31 particles in the presence of water. They suggest that interaction between NO2 and soot particles
January 2005 AX2-50 DRAFT-DO NOT QUOTE OR CITE
-------
1 may account for high mixing ratios of HONO observed in urban environments. Conversion of
2 NO2 to HONO and subsequent photolysis to NO + OH would constitute an NOx-catalyzed O3
3 sink involving snow. High concentrations of HONO can lead to the rapid growth in OH
4 concentrations shortly after sunrise, giving a "jump start" to photochemical smog formation.
5 Prolonged exposure to ambient oxidizing agents appears to deactivate this process. Broske et al.
6 (2003) studied the interaction of NO2 on secondary organic aerosols and concluded that the
7 uptake coefficients were too low for this reaction to be an important source of HONO in the
8 troposphere.
9 Choi and Leu (1998) evaluated the interactions of nitric acid on a model black carbon soot
10 (FW2), graphite, hexane and kerosene soot. They found that HNO3 decomposed to NO2 and
11 H2O at higher nitric acid surface coverages, i.e., P(HNO3) > = 10~4 Torr. None of the soot
12 models used were reactive at low nitric acid coverages, at P(HNO3) = 5 x 10~7 Torr or at lower
13 temperatures (220K). They conclude that it is unlikely that aircraft soot in the upper
14 troposphere/lower stratosphere reduces HNO3.
15 Heterogeneous production on soot at night is believed to be the mechanism by which
16 HONO accumulates to provide an early morning source of HOX in high NOX environments
17 (Harrison et al., 1996; Jacob, 2000). HONO has been frequently observed to accumulate to
18 levels of several ppb over night, and has been attributed to soot chemistry (Harris et al., 1982;
19 Calvert et al., 1994; Jacob, 2000).
20 Longfellow et al. (1999) observed the formation of HONO when methane, propane, hexane
21 and kerosene soots were exposed to NO2. They estimate that this reaction may account for some
22 part of the unexplained high levels of HONO observed in urban areas. They comment that
23 without details about the surface area, porosity and amount of soot available for this reaction,
24 reactive uptake values cannot reliably be estimated. They comment that soot and NO2 are
25 produced in close proximity during combustion, and that large quantities of HONO have been
26 observed in aircraft plumes.
27 Saathoff et al. (2001) studied the heterogeneous loss of NO2, HNO3, NO3/N2O5,
28 HO2/HO2NO2 on soot aerosol using a large aerosol chamber. Reaction periods of up to several
29 days were monitored and results used to fit a detailed model. They derived reaction probabilities
30 at 294 K and 50% RH for NO2, NO3, HO2 and HO2NO2 deposition to soot, HNO3 reduction to
31 NO2, and N2O5 hydrolysis. When these probabilities were included in photochemical box model
January 2005 AX2-51 DRAFT-DO NOT QUOTE OR CITE
-------
1 calculations of a 4-day smog event, the only noteworthy influence of soot was a 10% reduction
2 in the 2nd day O3 maximum, for a soot loading of 20 jig m"3, i.e., a factor of 2 to 10 times
3 observed black carbon loadings seen during extreme U.S. urban pollution events, although such
4 concentrations are observed routinely in the developing world.
5 Mufioz and Rossi (2002) conducted Knudsen cell studies of HNO3 uptake on black and
6 grey decane soot produced in lean and rich flames, respectively. They observed HONO as the
7 main species released following nitric acid uptake on grey soot, and NO and traces of NO2 from
8 black soot. They conclude that these reactions would only have relevance in special situations in
9 urban settings where soot and HNO3 are present in high concentrations simultaneously.
10
11 AX2.2.10.3 Halogen Radical Chemistry
12 Barrie et al. (1988) first suggested that halogen chemistry on snow surfaces in the Arctic
13 could lead to BrOx formation and subsequent O3 destruction. More recent work suggests that
14 halogen radical reactions may influence O3 chemistry in mid latitudes as well.
15 The weight of available evidence supports the hypothesis that halogen radical chemistry
16 significantly influences O3 cycling over much of the marine boundary layer at lower latitudes
17 and in at least some other regions of the troposphere. However, proposed chemical mechanisms
18 are associated with substantial uncertainties and, based on available information, it appears
19 unlikely that a simple parameterization (analogous to those recommended by Jacob (2000) for
20 other multiphase transformations) would adequately capture major features of the underlying
21 transformations.
22 Most of the Cl and Br in the marine boundary layer are produced in association with
23 sea-salt aerosols by wind stress at the ocean surface (e.g., Gong et al., 1997). Fresh aerosols
24 rapidly dehydrate towards equilibrium with ambient water vapor and undergo other chemical
25 processes involving the scavenging of reactive gases, aqueous-phase transformations, and
26 volatilization of products. Many of these processes are strongly pH-dependent (Keene et al.,
27 1998). Throughout most of the marine boundary layer, sea-salt alkalinity is tritrated rapidly
28 (seconds to minutes) by ambient acids (Chameides and Stelson, 1992; Erickson et al., 1999) and,
29 under a given set of conditions, the pHs of the super-jam, sea-salt size fractions are buffered to
30 similar values via HC1 phase partitioning (Keene and Savoie, 1998; 1999; Keene et al., 2002).
January 2005 AX2-52 DRAFT-DO NOT QUOTE OR CITE
-------
1 Model calculations based on the autocatalytic halogen activation mechanism (Vogt et al.,
2 1996; Keene et al., 1998; Sander et al., 1999; von Glasow et al., 2002a,b; Pszenny et al., 2003;
3 Sander et al., 2003) predict that most particulate Br- associated with acidified sea-salt aerosol
4 would react to form Br2 and BrCl, which subsequently volatilize and photolyze in sunlight to
5 produce atomic Br and Cl. Most Br atoms recycle in the gas phase via
6
•Br + O3 -> BrO + O2 (AX2-42)
BrO- + HCy -» HOBr + O2 (AX2-43)
8
HOBr + hv -» 'OH + -Br (AX2-44)
9
10 and thereby catalytically destroy O3, analogous to Br cycling in the stratosphere (e.g.,
1 1 Mozurkewitch, 1995; Sander and Crutzen, 1996). Side reactions with HCHO and other
12 compounds produce HBr, which is either scavenged and recycled through the aerosol or lost to
13 the surface via wet and dry deposition (Dickerson et al., 1999).
14 Cl-radical chemistry influences O3 in two ways (e.g., Pszenny et al., 1993). Some atomic
15 Cl in marine air reacts directly with O3 initiating a catalytic sequence analogous to that of Br
16 (AX2-42 through AX2-44 above). However, most atomic Cl in the MBL reacts with
17 hydrocarbons (which, relative to the stratosphere, are present at high concentrations) via
18 hydrogen extraction to form HC1 vapor. The enhanced supply of odd hydrogen radicals from
19 hydrocarbon oxidation leads to O3 production in the presence of sufficient NOX. Thus, Cl
20 chemistry represents a modest net sink for O3 when NOX is less than 20 pptv and a net source at
21 higher NOX. Although available evidence suggest that significant Cl-radical chemistry occurs in
22 clean marine air, its net influence on O3 appears to be small relative to that of Br and I.
23 In addition to Br and Cl, several lines of recent evidence suggests that an autocatalytic
24 cycle also sustains I-radical chemistry leading to significant net O3 destruction in marine air
25 (Vogt et al., 1996, 1999; von Glasow et al., 2002a). The cycle is initiated by photolysis of
26 organoiodine compounds emitted from the ocean surface to generate atomic I (Carpenter et al.,
27 1999). Iodine atoms react almost exclusively with O3 to form IO. Most IO photodissociates in
January 2005 AX2-53 DRAFT-DO NOT QUOTE OR CITE
-------
1 sunlight to generate I and atomic O, which rapidly recombines with O2 to form O3.
2 Consequently, this cycle has no net effect on O3 (Stutz et al., 1999). However, alternative
3 reaction pathways analogous to reactions AX2-42 through AX2-44 above lead to catalytic O3
4 destruction. Model calculations suggest that HOI recycles via acid-catalyzed aerosol scavenging
5 to form IC1 and IBr, which subsequently volatilize and photolyze to form halogen atoms. The
6 net effect of this multiphase pathway is to increase concentrations of volatile reactive I. The self
7 reaction of IO to form I and OIO may further enhance O3 destruction (Cox et al., 1999;
8 Ashworth et al., 2002). IO also reacts with NO2 to form INO3, which can be scavenged by
9 aqueous aerosols. This pathway has been suggested as a potentially important sink for NOX in
10 the remote MBL and would, thus, contribute indirectly to net O3 destruction (McFiggans et al.,
11 2000).
12 Various lines of observational evidence support aspects of the above scenarios. Most
13 measurements of particulate Br in marine air reveal large depletions relative to conservative sea-
14 salt tracers (e.g., Sander et al., 2003) and, because HBr is highly soluble in acidic solution, these
15 deficits cannot be explained by simple acid-displacement reactions (e.g., Ayers et al., 1999).
16 Observed Br depletions are generally consistent with predictions based on the halogen activation
17 mechanism. In contrast, available, albeit limited, data indicate that I is highly enriched in marine
18 aerosols relative to bulk seawater (e.g., Sturges and Barrie, 1988), which indicates active
19 multiphase iodine chemistry.
20 Direct measurements of BrO in marine air by differential optical absorption spectroscopy
21 (DOAS) reveal mixing ratios that are near or below analytical detection limits of about 1 to 3 ppt
22 (Honninger, 1999; Pszenny et al., 2003; Leser et al., 2003) but within the range of model
23 predictions. Column-integrated DOAS observations from space reveal substantial mixing ratios
24 of tropospheric BrO (e.g., Wagner and Platt, 1998). Although the relative amounts in the MBL
25 cannot be resolved, these data strongly suggest active destruction of tropospheric O3 via the
26 reaction sequence of AX2-42 through AX2-44. Similarly, measurements of IO (McFiggans
27 et al., 2000) and OIO (Allan et al., 2001) indicate active O3 destruction by an analogous pathway
28 involving atomic I. In addition, anticorrelations on diurnal time scales between total volatile
29 inorganic Br and particulate Br and between volatile inorganic I and particulate I have been
30 reported (e.g., Rancher and Kritz, 1980; Pszenny et al., 2003). Although the lack of speciation
January 2005 AX2-54 DRAFT-DO NOT QUOTE OR CITE
-------
1 precludes unambiguous interpretation, these relationships are also consistent with predictions
2 based on the halogen activation mechanism.
3 Large diurnal variabilities in O3 measured over the remote subtropical Atlantic and Indian
4 Oceans (Dickerson et al., 1999; Burkert et al., 2003) and early morning depletions of O3
5 observed in the remote temperate MBL (Galbally et al., 2000) indicate that only about half of the
6 inferred O3 destruction in the MBL can be explained by conventional HOX/NOX chemistry.
7 Model calculations suggest that Br- and I-radical chemistry could account for a 'missing' O3 sink
8 of this magnitude (Dickerson et al., 1999; Stutz et al., 1999; McFiggans et al., 2000; von Glasow
9 et al., 2002b). In addition to the pathway for O3 destruction given by R AX2-39 to R AX2-41, in
10 areas with high concentrations of halogen radicals the following generic loss pathways for O3
1 1 can occur in the Arctic at the onset of spring and also over salt flats near the Dead Sea
12 (Hebestreit et al., 1999) and the Great Salt Lake (Stutz et al., 2002) analogous to their occurrence
13 in the lower stratosphere (Yung et al., 1980).
14
X = Br,Cl,I) (AX2-45)
• Y + 03 -» YO + 02 (Y = Br, Cl, I) (AX2-46)
2 (AX2-47)
Net: 203^302 (AX2-48)
15
16 Note that the self reaction of CIO radicals is likely to be negligible in the troposphere. There are
17 three major reaction pathways involved in reaction AX2-47. Short-lived radical species are
18 produced. These radicals rapidly react to yield monoatomic halogen radicals. In contrast to the
19 situation in marine air, where DO AS measurements indicate BrO concentrations of 1 to 3 ppt,
20 Stutz et al. (2002) found peak BrO concentrations of about 6 ppt and peak CIO concentrations of
21 about 15 ppt. They also derived a correlation coefficient of -0.92 between BrO and O3 but much
22 smaller values of r between CIO and O3. Stutz et al. attributed the source of the reactive
23 halogens to concentrated high molality solutions or crystalline salt around salt lakes, conditions
24 that do not otherwise occur in more dilute or ocean salt water. They also suggest that halogens
25 may be released from saline soils. The inferred atmospheric concentrations of Cl are about
January 2005 AX2-55 DRAFT-DO NOT QUOTE OR CITE
-------
1 10s / cm3, or about a factor of 100 higher than found in the marine boundary layer by Rudolph
2 et al. (1997) indicating that, under these conditions, the Cl initiated oxidation of hydrocarbons
3 could be substantial.
4 Most of the well-established multiphase reactions tend to reduce the rate of O3 formation in
5 the polluted troposphere. Direct reactions of O3 and atmospheric particles appears to be too slow
6 to reduce smog significantly. Removal of HO2 onto hydrated particles will decrease the
7 production of O3 by the reaction of HO2 with NO. The uptake of NO2 and HNO3 will also result
8 in the production of less O3. Conditions leading to high concentrations of Br, Cl, and I radicals
9 can lead to O3 loss. The oxidation of hydrocarbons (especially alkanes) by Cl radicals,
10 in contrast, may lead to the rapid formation of peroxy radicals and faster smog production in
11 coastal environments where conditions are favorable for the release of gaseous Cl from the
12 marine aerosol. There is still considerable uncertainty regarding the role of multiphase processes
13 in tropospheric photochemistry and so results should be viewed with caution and an appreciation
14 of their potential limitations.
15
16 AX2.2.10.4 Reactions on the Surfaces of Crustal Particles
17 Field studies have shown that O3 levels are reduced in plumes containing high particle
18 concentrations (e.g., DeReus et al.; 2000; Berkowitz et al., 2001; Gaffney et al., 2002).
19 Laboratory studies of the uptake of O3 on un-treated mineral surfaces (Hanisch and Crowley,
20 2002; Michel et al., 2002,2003) have shown that O3 is lost by reaction on these surfaces and this
21 loss is catalytic. Values of y of 1.2 ± 0.4 x 10~4 were found for reactive uptake on a-A12O3
22 and 5 ± 1 x icr5 for reactive uptake on SiO2 surfaces. Usher et al. (2003) found mixed behavior
23 for O3 uptake on coated surfaces with respect to untreated surfaces. They found that y drops
24 from 1.2 ± 0.4 x 1CT4 to 3.4 ± 0.6 x 1CT5 when a-A12O3 surfaces are coated with NO3 derived
25 from HNO3, whereas they found that y increases to 1.6 ± 0.2 x icr4 after these surfaces have
26 been pre-treated with SO2. Usher et al. also pre-treated surfaces of SiO2 with either a C8-alkene
27 or a C8-alkane terminated organotrichlorosilane. They found that y increased to 7 ± 2 x 1CT5 in
28 the case of treatment with the alkene, but that it decreased to 3 ± 1 x icr5 for treatment with the
29 alkane. Usher et al. (2003) suggested, on the basis of these results that mineral dust particles
30 coated with nitrates or alkanes will affect O3 less than dust particles that have accumulated
31 coatings of sulfite or alkenes. These studies indicate the importance of aging of airborne
January 2005 AX2-56 DRAFT-DO NOT QUOTE OR CITE
-------
1 particles on their ability to take up atmospheric gases. Reactions such as these may also be
2 responsible for O3 depletions observed in dust clouds transiting the Pacific Ocean.
3 Underwood et al. (2001) studied the uptake of NO2 and HNO3 on the surfaces of dry
4 mineral oxides (containing Al, Ca, Fe, Mg, Si and Ti) and naturally occurring mineral dust.
5 A wide range of values of y(NO2) were found, ranging from < 4 x 10"10 for SiO2 to 2 x 10~5 for
6 CaO, with most other values ~10~6. Values of y for Chinese loess and Saharan dust were also of
7 the order of 10~6. They found that as the reaction of NO2 proceeds on the surfaces that reduction
8 to NO occurs. They recommended a value of y for HNO3 of about 1 x 10~3. Not surprisingly,
9 the values of y increased from those given above if the surfaces were wetted. Underwood et al.
10 (2001) also suggested that the uptake of NO2 was likely to be only of marginal importance but
11 that uptake of HNO3 could be of significance for photochemical oxidant cycles.
12 Li et al. (2001) examined the uptake of acetaldehyde, acetone and propionaldehyde on the
13 same mineral oxide surfaces listed above. They found that these compounds weakly and
14 reversibly adsorb on SiO2 surfaces. However, on the other oxide surfaces, they irreversibly
15 adsorb and can form larger compounds. They found values of y ranging from 10~6 to 10~4.
16 These reactions may reduce O3 production efficiency in areas of high mineral dust concentration
17 such as the American Southwest or in eastern Asia as noted earlier.
18
19 AX2.2.10.5 Reactions on the Surfaces of Aqueous H2SO4 solutions
20 The most recent evaluation of Photochemical and Chemical Data by the Jet Propulsion
21 Laboratory (Jet Propulsion Laboratory, 2003) includes recommendations for uptake coefficients
22 of various substances on a variety of surfaces including aqueous H2SO4 solutions. Although
23 much of the data evaluated have been obtained mainly for stratospheric applications, there are
24 studies in which the range of environmental parameters is compatible with those found in the
25 troposphere. In particular, the uptake of N2O5 on the surface of aqueous H2SO4 solutions has
26 been examined over a wide range of values. Typical values of y are of the order of 0.1 (e.g., Jet
27 Propulsion Laboratory, 2003). Values of y for NO2 are much lower (5 x 10"7to within a factor of
28 three) and thus the uptake of NO2 on the surface of aqueous H2SO4 solutions is unlikely to be of
29 importance for oxidant cycles. The available data indicate that uptake of OH and HO2 radicals
30 could be significant under ambient conditions with values of y of the order of 0.1 or higher for
31 OH, and perhaps similar values for HO2.
January 2005 AX2-57 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX2.3 PHYSICAL PROCESSES INFLUENCING THE ABUNDANCE
2 OF OZONE
3 The abundance and distribution of O3 in the atmosphere is determined by complex
4 interactions between meteorology and chemistry. This section will address these interactions,
5 based mainly on the results of field observations. The importance of a number of transport
6 mechanisms, whose understanding has undergone significant advances since the last AQCD for
7 O3, will be discussed in this section.
8 Major episodes of high O3 concentrations in the eastern United States and in Europe are
9 associated with slow moving, high pressure systems. High pressure systems during the warmer
10 seasons are associated with the sinking of air, resulting in warm, generally cloudless conditions,
11 with light winds. The sinking of air results in the development of stable conditions near the
12 surface which inhibit or reduce the vertical mixing of O3 precursors. The combination of
13 inhibited vertical mixing and light winds minimizes the dispersal of pollutants emitted in urban
14 areas, allowing their concentrations to build up. Photochemical activity involving these
15 precursors is enhanced because of higher temperatures and the availability of sunlight. In the
16 eastern United States, high O3 concentrations during a large scale episode can extend over a
17 hundred thousand square kilometers for several days. These conditions have been described in
18 greater detail in AQCD 96. The transport of pollutants downwind of maj or urban centers is
19 characterized by the development of urban plumes. However, the presence of mountain barriers
20 can limit mixing as in Los Angeles and Mexico City and will result in even longer periods and a
21 higher frequency of days with high O3 concentrations. Ozone concentrations in southern urban
22 areas, such as Houston, TX and Atlanta, GA tend to follow this pattern and they tend to decrease
23 with increasing wind speed. In northern cities, like Chicago, IL; New York, NY; and Boston,
24 MA the average O3 concentrations over the metropolitan areas increase with wind speed
25 indicating that transport of O3 and its precursors from upwind areas is important (Husar and
26 Renard, 1998; Schichtel and Husar, 2001).
27 Aircraft observations indicate that there can be substantial differences in mixing ratios of
28 key species between the surface and the atmosphere above (Fehsenfeld et al., 1996a; Berkowitz
29 and Shaw, 1997). Convective processes and small scale turbulence transport O3 and other
30 pollutants both upward and downward throughout the planetary boundary layer and the free
31 troposphere. Ozone and its precursors were found to be transported vertically by convection into
January 2005 AX2-58 DRAFT-DO NOT QUOTE OR CITE
-------
1 the upper part of the mixed layer on one day, then transported overnight as a layer of elevated
2 mixing ratios and then entrained into a growing convective boundary layer downwind and
3 brought back down to the surface. High concentrations of O3 showing large diurnal variations at
4 the surface in southern New England were associated with the presence of such layers
5 (Berkowitz et al., 1998). Zhang et al. (1997) estimated that as much as 60% to 70% of O3
6 observed at the surface during one case study in eastern Tennessee could have come from the
7 upper boundary layer. Because of wind shear, winds several hundred meters above the ground
8 can bring pollutants from the west, even though surface winds are from the southwest during
9 periods of high O3 in the eastern United States (Blumenthal et al., 1997). Low level nocturnal
10 jets can also transport pollutants hundreds of kilometers. Turbulence associated with them can
11 bring these pollutants to the surface and in many locations result in secondary O3 maxima in the
12 early morning (Corsmeier et al., 1997). Based on analysis of the output of model studies
13 conducted by Kasibhatla and Chameides (2000), Hanna et al. (2001) concluded that O3 can be
14 transported over thousands of kilometers in the upper boundary layer of the eastern half of the
15 United States during specific O3 episodes.
16 Stratospheric-tropospheric exchange (STE) will be discussed in Section AX2.3.1. The
17 vertical redistribution of O3 and other pollutants by deep, or penetrating convection is discussed
18 in Section AX2.3.2. The potential importance of transport of O3 and precursors by low level jets
19 is the topic of Section AX2.3.3. Issues related to the transport of O3 from North America are
20 presented in Section AX2.3.4. Relations of O3 to solar ultraviolet radiation and temperature will
21 then be discussed in Section AX2.3.5.
22
23 AX2.3.1 Stratospheric-Tropospheric Ozone Exchange (STE)
24 In the stratosphere, O3 formation is initiated by the photodissociation of molecular oxygen
25 (O2) by solar ultraviolet radiation at wavelengths less than 242 nm. Almost all of this radiation
26 is absorbed in the stratosphere (except for regions near the tropical tropopause), preventing this
27 mechanism from occurring in the troposphere. Some of the O3 in the stratosphere is transported
28 downward into the troposphere. The potential importance of this source of tropospheric O3 has
29 been recognized since the early work of Regener (1941), as cited by Junge (1963).
30 Stratospheric-tropospheric exchange (STE) of O3 and stratospheric radionuclides produced by
31 the nuclear weapons tests of the 1960s is at a maximum during late winter and early spring (e.g.,
January 2005 AX2-59 DRAFT-DO NOT QUOTE OR CITE
-------
1 Ludwig et al., 1977 and references therein). Since AQCD 96 on O3 substantial new information
2 from numerical models, field experiments and satellite-based observations has become available.
3 The following sections outline the basic atmospheric dynamics and thermodynamics of
4 stratosphere/troposphere exchange and review these new developments.
5 There are several important mechanisms for injecting stratospheric O3 into the troposphere,
6 they include tropopause folds (Reed, 1955; Danielsen, 1968), cut-off lows (Price and Vaughan,
7 1993), clear air turbulence, mesoscale convective complexes and thunderstorms, breaking
8 gravity waves (Poulida et al., 1996; Langford and Reid, 1998; Stohl et al., 2003) and streamers.
9 Streamers are dry, stratospheric intrusions visible in satellite water vapor imagery that are
10 sheared into long filamentary structures that often roll into vortices and exhibit visible evidence
11 of the irreversible mixing of moist subtropical tropospheric and dry polar stratospheric air
12 (Appenzeller et al., 1996; Wimmers et al., 2003). They are often present at a scale that eludes
13 capture in large scale dynamical models of the atmosphere that cannot resolve features less than
14 1 degree (-100km). Empirical evidence for stratospheric intrusions comes from observations of
15 indicators of stratospheric air in the troposphere. These indicators include high potential
16 vorticity, low water vapor mixing ratios, high potential temperature, enhancements in the ratio of
17 7Be to 10Be in tropospheric aerosols, as well as enhancements in O3 mixing ratios and total
18 column amounts. These quantities can be observed with in situ aircraft and balloons, as well as
19 remotely sensed from aircraft and ground-based lidars and both geostationary and polar (low
20 earth orbiting) space platforms.
21 The exchange of O3 between the stratosphere and the troposphere in middle latitudes
22 occurs to a major extent by tropopause folding events (Reiter, 1963; Reiter and Mahlman, 1965;
23 Danielsen, 1968; Reiter, 1975; Danielsen andMohnen, 1977; Danielsen, 1980). The term,
24 tropopause folding is used to describe a process in which the tropopause intrudes deeply into the
25 troposphere along a sloping frontal zone bringing air from the lower stratosphere with it.
26 Tropopause folds occur with the formation of upper level fronts associated with transverse
27 circulations that develop around the core of the polar jet stream. South of the jet stream core, the
28 tropopause is higher than to the north of it. The tropopause can be imagined as wrapping around
29 the jet stream core and folding beneath it and extending into the troposphere (cf. Figure
30 AX2-7a). Although drawn as a heavy solid line, the tropopause should not be imagined as a
31 material surface, through which there is no exchange. Significant intrusions of stratospheric air
January 2005 AX2-60 DRAFT-DO NOT QUOTE OR CITE
-------
150
1000
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 Vandenburg AFB,
CA (VBG); San Diego, CA (SAN); Winnemucca, NV (UCC); and Ely, NV
(ELY).
Source: Adapted from Shapiro (1980).
1 occur in "ribbons" -200 to 1000 km in length, 100 to 300 km wide and about 1 to 4 km thick
2 (Hoskins, 1972; Wimmers et al., 2003). These events occur throughout the year and their
3 location follows the seasonal displacement of the polar jet stream.
4 The seasonal cycle of O3 exchange from the stratosphere into the troposphere is not caused
5 by a peak in the seasonal cycle of upper tropospheric cyclone activity. Instead, it is related to the
January 2005
AX2-61
DRAFT-DO NOT QUOTE OR CITE
-------
1 large scale pattern of tracer transport in the stratosphere. During winter in the Northern
2 Hemisphere, there is a maximum in the poleward, downward transport of mass, which moves O3
3 from the tropical upper stratosphere to the lower stratosphere of the polar and mid-latitudes This
4 global scale pattern is controlled by the upward propagation of large-scale and small-scale waves
5 generated in the troposphere. As the energy from these disturbances dissipates, it drives this
6 stratosphere circulation. As a result of this process, there is a springtime maximum in the total
7 column abundance of O3 over the poles. The concentrations of O3 (and other trace substances)
8 build up in the lower stratosphere until their downward fluxes into the lower stratosphere are
9 matched by increased fluxes into the troposphere. Thus, there would be a springtime maximum
10 in the flux of O3 into the troposphere even if the flux of stratospheric air through the tropopause
11 by tropopause folding remained constant throughout the year (Holton et al., 1995). Indeed,
12 cyclonic activity in the upper troposphere is active throughout the entire year in transporting air
13 from the lower stratosphere into the troposphere (Mahlman, 1997; and references therein).
14 Oltmans et al. (1996) and Moody et al. (1996) provide evidence that stratospheric intrusions
15 contribute to the O3 abundance in the upper troposphere over the North Atlantic even during the
16 summer.
17 There are a number of techniques that have been used to quantify the amount of O3 in the
18 free troposphere or even the amount of O3 reaching the surface that can be attributed to
19 downward transport from the stratosphere. Earlier work, cited in AQCD 96 relied mainly on the
20 use of 7Be as a tracer of stratospheric air. However, its use is ambiguous because it is also
21 formed in the upper troposphere. Complications also arise because its production rate is also
22 sensitive to solar activity (Lean, 2000). The ratio of 7Be to 10Be provides a much more sensitive
23 tracer of stratospheric air than the use of 7Be alone (Jordan et al., 2003 and references therein).
24 More recent work than cited in AQCD 96 has focused on the use of potential vorticity (PV) as a
25 tracer of stratospheric air. Potential vorticity is a dynamical tracer used in meteorology.
26 Generally, PV is calculated from wind and temperature observations and represents the
27 rotational tendency of a column of air weighted by the static stability, which is just the distance
28 between isentropic surfaces. This quantity is a maximum in the lower stratosphere where static
29 stability is great and along the jet stream where wind shear imparts significant rotation to air
30 parcels. As air moves from the stratosphere to the troposphere, PV is conserved, and therefore it
31 traces the motion of O3. The static stability is lower in the troposphere, so to preserve PV, fluid
January 2005 AX2-62 DRAFT-DO NOT QUOTE OR CITE
-------
1 rotation will increase. This is why STE is associated with cyclogenesis, or the formation of
2 storms along the polar jet stream. Dynamical models clearly capture this correspondence
3 between the location of storm tracks and preferred regions for STE. However, because PV is
4 destroyed at a faster rate with increasing depth, it is not useful as a tracer of stratospheric air
5 reaching the surface. Appenzeller et al. (1996) found that maps of PV coupled with satellite
6 images of humidity can provide indications of the intrusion of stratospheric air into the
7 troposphere, however, they had no measurements of O3. Even if measurements of O3 were
8 available, the extrapolation of any relations to other events would still be problematic as Olsen
9 et al. (2002) have noted that there are seasonal and geographic variations in the relation between
10 O3 and PV. Recent flights of the NCAR C130 during the TOPSE campaign measured in situ O3,
11 and curtains of O3 above and below the aircraft observed with a lidar and clearly showed a
12 correspondence between high PV and stratospheric levels of O3 and satellite depictions of dry air
13 indicating the presence of tropopause folding (Wimmers and Moody, 2004a,b).
14 Detailed cross sections through a tropopause folding event showing atmospheric structure,
15 O3 mixing ratios and condensation nuclei (CN) counts are given in Figures AX2-7a, AX2-7b,
16 and AX2-7c (Shapiro, 1980). Flight tracks of an NCAR Sabreliner obtaining data through the
17 tropopause fold are also shown. The core of the jet stream is indicated by the hatched area near
18 the center of Figure AX2-7a. As can be seen from Figure AX2-7 a and b, there is a strong
19 relation between the folding of the tropopause, indicated by the heavy solid line and O3. CN
20 counts during the portions of the flights in the lower troposphere were tropospheric were
21 typically of the order of several * 10 3 cm"3 and 100 or less in the stratospheric portion.
22 However, it is clear that CN counts in the fold are much higher than in the stratosphere proper,
23 suggesting that there was active mixing between tropospheric and stratospheric air in the fold.
24 Likewise, it can also be seen form Figure AX2-7b that O3 is being mixed outside the fold into the
25 middle and upper troposphere. The two data sets shown in Figures AX2-7b and 7c indicate that
26 small scale turbulent processes were occurring to mediate this exchange and that the folds are
27 mixing regions whose chemical characteristics lie between those of the stratosphere and the
28 troposphere (Shapiro, 1980). Chemical interactions between stratospheric and tropospheric
29 constituents are also possible within tropopause folds. These considerations also imply that in
30 the absence of turbulent mixing, tropopause folding can be a reversible process.
31
January 2005 AX2-63 DRAFT-DO NOT QUOTE OR CITE
-------
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 Vandenburg AFB, CA (VBG);
San Diego, CA (SAN); Winnemucca, NV (UCC); and Ely, NV (ELY).
Source: Adapted from Shapiro (1980).
1 Several recent papers have attempted to demonstrate that the atmosphere is a fluid
2 composed of relatively distinct airstreams with characteristic three-dimensional motions and
3 corresponding trace gas signatures. Based on aircraft observations, satellite imagery, and back
4 trajectories, it has been shown that dry airstreams, or dry intrusions (DA or DI) always advect
5 stratospheric O3 into the middle and upper troposphere (Cooper et al, 2001; Cooper et al.,
6 2002a), however the seasonal cycle of O3 in the lowermost stratosphere allows greater quantities
7 of O3 to enter the troposphere during spring (Cooper et al., 2002b). Other work has focused on
8 the signatures of PV to show specific instances of STE (Olsen and Stanford, 2001). This
9 correlation between TOMS gradients and PV was also used to derive the annual mass flux of O3
10 from STE and generated an estimate somewhat higher (500 Tg/yr over the Northern
January 2005
AX2-64
DRAFT-DO NOT QUOTE OR CITE
-------
150
121° VBG
119°
SAN 117°
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 Vandenburg AFB, CA (VBG);
San Diego, CA (SAN); Winnemucca, NV (UCC); and Ely, NV (ELY).
Source: Adapted from Shapiro (1980).
1 Hemisphere) than the estimates of most general circulation models. The IPCC has reported a
2 large range of model estimates of STE, expressed as the net global flux of O3 in Tg/yr, from a
3 low of 390 to a high of 1440 (reproduced as Table AX2-3C-1). A few other estimates have been
4 made based on chemical observations in the lower stratosphere, or combined chemistry and
5 dynamics (450 Tg/yr Murphy and Fahey, 1994; 510 Tg/yr extratropics only, Gettleman et al.,
6 1997; and 500 Tg/yr midlatitude NH only (30 to 60N) (Olsen et al., 2002). These values
7 illustrate the large degree of uncertainty that remains in quantifying this important source of O3.
January 2005
AX2-65
DRAFT-DO NOT QUOTE OR CITE
-------
1 Based on the concept of tracing airstream motion, a number of Lagrangian model studies
2 have resulted in climatologies that have addressed the spatial and temporal variability in
3 stratosphere to troposphere transport (Stohl, 2001; Wernli and Borqui, 2002; Seo and Bowman,
4 2002; James et al., 2003a,b; Sprenger and Wernli, 2003; Sprenger et al., 2003). Both Stohl
5 (2001) and Sprenger et al. (2003) produced one year climatologies of tropopause folds based on
6 a 1° by 1° gridded meteorological data set. They each found the probability of deep folds
7 (penetrating to the 800 hPa level) was a maximum during winter (December through February).
8 The highest frequency of folding extended from Labrador down the east coast of North America.
9 However, these deep folds occurred less than 1% of the six hour intervals for which
10 meteorological data is assimilated for grid points in the United States. They observed a higher
11 frequency of more shallow folds (penetrating to the upper troposphere) and medium folds
12 (penetrating to levels between 500 and 600 hPa) of about 10% and 1 to 2% respectively. These
13 events occur preferentially across the subtropics and the southern United States. At higher
14 latitudes other mechanisms such as the erosion of cut-off lows and the breakup of stratospheric
15 streamers are likely to play an important role in STE. Stohl (2001) also described the region of
16 strong stirring in the upper extratropical troposphere related to the mid-latitude storm tracks.
17 Stohl (2001) demonstrated that airstreams with strong vertical motion are all highly incoherent,
18 they stir their air parcels into a new environment, producing filamentary tracer structures and
19 paving the way for subsequent mixing. A 15-year climatology by Sprenger and Wernli (2003)
20 shows the consistent pattern of STE occurring over the primary storm tracks in the Pacific and
21 Atlantic along the Asian and North American coasts. This climatology, and the one of James
22 et al. (2003a,b) both found that recent stratospheric air associated with deep intrusions are
23 relatively infrequent occurrences in these models. Thus, stratospheric intrusions are most likely
24 to directly affect the middle and upper troposhere and not the planetary boundary layer.
25 However, this O3 can still exchange with the planetary boundary layer through convection as
26 described later in this sub-section and in Section AX2.3.2, AX2.3.3 and AX2.3.4.
27 Interannual variations in STE are related to anomalies in large-scale circulation such as the
28 North Atlantic Oscillation which causes changes in storm track positions and intensities, and the
29 El Nino-Southern Oscillation, which results in anomalous strong convection over the eastern
30 Pacific (James et al., 2003a,b). It should also be remembered that the downward flux of O3 into
31 the troposphere is related to the depletion of O3 within the wintertime stratospheric polar
January 2005 AX2-66 DRAFT-DO NOT QUOTE OR CITE
-------
1 vortices. The magnitude of this depletion and the transport of ozone depleted air to mid-latitudes
2 in the stratosphere (Mahlman et al., 1994; Hadjinicolaou and Pyle, 2004) shows significant
3 interannual variability which may also be reflected in the downward flux of O3 into the
4 troposphere. All of these studies, from the analysis of individual events to multiyear
5 climatologies are based on the consideration of the three-dimensional motion of discrete
6 airstreams in the atmosphere. However, there is a significant body of work that reports that
7 airstreams are not entirely independent of each other (Cooper et al., 2004a,b). Midlatitude
8 cyclones typically form in a sequential manner, some trailing in close proximity along a quasi-
9 stationary frontal boundary, with each system influenced by remnants of other systems. For
10 example, a rising stream of air ahead of a cold front (also known as a warm conveyor belt or
11 WCB) on the back (western) side of a surface anticyclone may entrain air that has subsided
12 anticyclonically into the surface high pressure system from the upper troposphere and the lower
13 stratosphere (also known as a Dry Airstream or DA) that intruded into the mid-troposphere in a
14 cyclone that is further downstream. Convective mixing of the boundary layer in the WCB will
15 distribute this enhanced O3 throughout the lower troposphere and down to the surface (Davies
16 and Shuepbach, 1994; Cooper and Moody, 2000). The net effect is that the DA of one cyclone
17 may feed into the WCB of the system immediately upwind. Similarly, the lofting of warm moist
18 air in the WCB may inject surface emissions into the upper troposphere adjacent to the western
19 side of the subsiding Dry Airstream of the storm system immediately downwind, with
20 subsequent interleaving of these two airstreams (Prados et al., 1999; Parrish et al., 2000; Cooper
21 et al., 2004a,b) as illustrated schematically in Figure AX2-8. The ultimate mixing of these
22 airstreams, which inevitably occurs at a scale that is not resolved by current models confounds
23 our ability to attribute trace gases to their sources.
24 These studies suggest that both downward transport from the stratosphere and upward
25 transport from the atmospheric boundary layer act in concert with their relative roles determined
26 by the balance between the amount of O3 in the lower stratosphere and the availability of free
27 radicals to initiate the photochemical processes forming O3 in the boundary layer. Dickerson
28 et al. (1995) pointed out that springtime maxima in O3 observed in Bermuda correlate well with
29 maxima in carbon monoxide. Carbon monoxide, O3 and its photochemical precursors may have
30 been transported into the upper troposphere from the polluted continental boundary layer by
31 deep convection. The photochemical processes involve the buildup of precursors during the
January 2005 AX2-67 DRAFT-DO NOT QUOTE OR CITE
-------
Altitude
Stratospheric Air
»--
Cold Front (A)
Stratospheric Air
Cold Front (B)
West
-1500
Distance (km)
East
Figure AX2-8. Schematic diagram of a meteorological mechanism involved in high
concentrations of ozone 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 ozone
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).
1
2
3
4
5
9
10
11
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).
AX2.3.2 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.
January 2005
AX2-68
DRAFT-DO NOT QUOTE OR CITE
-------
1 Layers ranging in thickness typically from 0.3 to about 2 km in the middle troposphere (mean
2 altitudes between 5 and 7 km) are ubiquitous and occupy up to 20% of the troposphere to 12 km
3 (Newell et al., 1999). The origin of these layers can be judged by analysis of their chemical
4 composition (typically by comparing ratios of H2O, O3 and CO to each other) or dynamical
5 properties (such as potential vorticity). Thus, pollutants that have been transported into the
6 middle and upper troposphere at one location can then be transported back down into the
7 boundary layer somewhere else.
8 Crutzen and Gidel (1983), Gidel (1983), and Chatfield and Crutzen (1984) hypothesized
9 that convective clouds played an important role in rapid atmospheric vertical transport of trace
10 species and first tested simple parameterizations of convective transport in atmospheric chemical
11 models. At nearly the same time, evidence was shown of venting of the boundary layer by
12 shallow fair weather cumulus clouds (e.g., Greenhut et al., 1984; Greenhut, 1986). Field
13 experiments were conducted in 1985, which resulted in verification of the hypothesis that deep
14 convective clouds are instrumental in atmospheric transport of trace constituents (Dickerson
15 et al., 1987; Luke et al., 1997). Once pollutants are lofted to the middle and upper troposphere,
16 they typically have a much longer chemical lifetime and with the generally stronger winds at
17 these altitudes they can be transported large distances from their source regions. Photochemical
18 reactions occur during this long-range transport. Pickering et al. (1990) demonstrated that
19 venting of boundary layer pollutants by convective clouds (both shallow and deep) causes
20 enhanced O3 production in the free troposphere. Therefore, convection aids in the
21 transformation of local pollution into a contribution to global atmospheric pollution. Downdrafts
22 within thunderstorms tend to bring air with less pollution from the middle troposphere into the
23 boundary layer.
24 Field studies have established that downward transport of larger O3 and NOX mixing ratios
25 from the free troposphere to the boundary layer is an important process over the remote oceans
26 (e.g., Piotrowicz et al., 1991), as well as the upward transport of very low O3 mixing ratios from
27 the boundary layer to the upper troposphere (Kley et al., 1996). Global modeling by Lelieveld
28 and Crutzen (1994) suggests that the downward mixing of O3 into the boundary layer (where it is
29 destroyed) is the dominant global effect of deep convection. Some indications of downward
30 transport of O3 from higher altitudes (possibly from the stratosphere) in the anvils of
31 thunderstorms have been observed (Dickerson et al., 1987; Poulida et al., 1996; Suhre et al.,
January 2005 AX2-69 DRAFT-DO NOT QUOTE OR CITE
-------
1 1997). Ozone is most effective as a greenhouse gas in the vicinity of the tropopause. Therefore,
2 changes in the vertical profile of O3 in the upper troposphere caused by deep convection have
3 important radiative forcing implications for climate.
4 Other effects of deep convection include perturbations to photolysis rates, which include
5 enhancement of these rates in the upper portion of the thunderstorm anvil. In addition,
6 thunderstorms are effective in the production of NO by lightning and in wet scavenging of
7 soluble species.
8
9 AX2.3.2.1 Observations of the Effects of Convective Transport
10 Some fraction of shallow fair weather cumulus clouds actively vent boundary layer
11 pollutants to the free troposphere (Stull, 1985). The first airborne observations of this
12 phenomenon were conducted by Greenhut et al. (1984) over a heavily urbanized area, measuring
13 the in-cloud flux of O3 in a relatively large cumulus cloud. An extension of this work was
14 reported by Greenhut (1986) in which data from over 100 aircraft penetrations of isolated
15 nonprecipitating cumulus clouds over rural and suburban areas were obtained. Ching and
16 Alkezweeny (1986) reported tracer (SF6) studies associated with nonprecipitating cumulus (fair
17 weather cumulus and cumulus congestus). Their experiments showed that the active cumulus
18 clouds transported mixed layer air upward into the overlying free troposphere and suggested that
19 active cumuli can also induce rapid downward transport from the free troposphere into the mixed
20 layer. A UV-DIAL (Ultraviolet Differential Absorption Lidar) provided space-height cross
21 sections of aerosols and O3 over North Carolina in a study of cumulus venting reported by Ching
22 et al. (1988). Data collected on evening flights showed regions of cloud debris containing
23 aerosol and O3 in the lower free troposphere in excess of background, suggesting that significant
24 vertical exchange had taken place during afternoon cumulus cloud activity. Efforts have also
25 been made to estimate the vertical transport by ensembles of nonprecipitating cumulus clouds in
26 regional chemical transport models (e.g., Vukovich and Ching, 1990).
27 The first unequivocal observations of deep convective transport of boundary layer
28 pollutants to the upper troposphere were documented by Dickerson et al. (1987).
29 Instrumentation aboard three research aircraft measured CO, O3, NO, NOX, NOy, and
30 hydrocarbons in the vicinity of an active mesoscale convective system near the
31 Oklahoma/Arkansas border during the 1985 PRE-STORM experiment. Anvil penetrations about
January 2005 AX2-70 DRAFT-DO NOT QUOTE OR CITE
-------
1 two hours after maturity found greatly enhanced mixing ratios of all of the aforementioned
2 species compared with outside of the cloud. Among the species measured, CO is the best tracer
3 of upward convective transport because it is produced primarily in the boundary layer and has an
4 atmospheric lifetime much longer than the timescale of a thunderstorm. In the observed storm,
5 CO measurements exceeded 160 ppbv as high as 11 km, compared with -70 ppbv outside of the
6 cloud (Figure AX2-9a). Cleaner middle tropospheric air appears to have descended in
7 downdrafts forming a pool of lower mixing ratio CO beneath the cloud. Nonmethane
8 hydrocarbons (NMHC) with moderate lifetimes can also serve as tracers of convective transport
9 from the boundary layer. Ozone can also be an indicator of convective transport. In the polluted
10 troposphere large O3 values will indicate upward transport from the boundary layer, but in the
11 clean atmosphere such values are indicative of downward transport from the uppermost
12 troposphere or lowermost stratosphere. In this case measured O3 in the upper rear portion of the
13 anvil peaked at 98 ppbv, while boundary layer values were only -65 ppbv (Figure AX2-9b). It is
14 likely that some higher-O3 stratospheric air mixed into the anvil.
15 The large amount of vertical trace gas transport noted by Dickerson et al. (1987) cannot,
16 however, be extrapolated to all convective cells. Pickering et al. (1988) reported airborne
17 measurements of trace gases taken in the vicinity of a line of towering cumulus and
18 cumulonimbus clouds that also occurred during PRE-STORM. In this case trace gas mixing
19 ratios in the tops of these clouds were near ambient levels. Meteorological analyses showed that
20 these clouds were located above a cold front, which prevented entry of air from the boundary
21 layer directly below or near the clouds. Instead, the air entering these clouds likely originated in
22 the layer immediately above the boundary layer which was quite clean. Luke et al. (1992)
23 summarized the air chemistry data from all 18 flights during PRE-STORM by categorizing each
24 case according to synoptic flow patterns. Storms in the maritime tropical flow regime
25 transported large amounts of CO, O3, and NOy into the upper troposphere with the
26 midtroposphere remaining relatively clean. During frontal passages a combination of stratiform
27 and convective clouds mixed pollutants more uniformly into the middle and upper levels; high
28 mixing ratios of CO were found at all altitudes.
29 Prather and Jacob (1997) and Jaegle et al. (1997) noted that in addition to the primary
30 pollutants (e.g., NOX, CO, VOCs), precursors of HOX are also transported to the upper
31 troposphere by deep convection. Precursors of most importance are water vapor, formaldehyde,
January 2005 AX2-71 DRAFT-DO NOT QUOTE OR CITE
-------
a. Carbon Monoxide
Surface
98°W| 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 ozone (ppbv).
Source: Dickerson et al. (1987).
January 2005
AX2-72
DRAFT-DO NOT QUOTE OR CITE
-------
1 hydrogen peroxide, methylhydroperoxide, and acetone. HOX is critical for oxidizing NO to NO2
2 in the O3 production process.
3 Over remote marine areas the effects of deep convection on trace gas distributions differ
4 from that over moderately polluted continental regions. Chemical measurements taken by the
5 NASA ER-2 aircraft during the Stratosphere-Troposphere Exchange Project (STEP) off the
6 northern coast of Australia show the influence of very deep convective events. Between
7 14.5 and 16.5 km on the February 2 to 3, 1987 flight, perturbations in the chemical profiles were
8 noted that included pronounced maxima in CO, water vapor, and CCN and minima of NOy, and
9 O3 (Pickering et al., 1993). Trajectory analysis showed that these air parcels likely were
10 transported from convective cells 800 to 900 km upstream. Very low boundary layer mixing
11 ratios of NOy and O3 in this remote region were apparently transported upward in the convection.
12 A similar result was noted in CEPEX (Central Equatorial Pacific Experiment; Kley et al., 1996)
13 where a series of ozonesonde ascents showed very low upper tropospheric O3 following deep
14 convection. It is likely that similar transport of low-O3 tropical marine boundary layer air to the
15 upper troposphere occurs in thunderstorms along the east coast of Florida. Convection over the
16 Pacific will likely transport halogens to the upper troposphere where they may aid in the
17 destruction of O3. This low-O3 convective outflow will likely descend in the subsidence region
18 of the eastern Pacific, leading to some of the cleanest air that arrives at the west coast of the
19 United States.
20
21 AX2.3.2.2 Modeling the Effects of Convection
22 The effects of deep convection may be simulated using cloud-resolving models, or in
23 regional or global models in which the convection is parameterized. The Goddard Cumulus
24 Ensemble (GCE) model (Tao and Simpson, 1993) has been used by Pickering et al. (1991;
25 1992a,b; 1993; 1996), Scala et al. (1990) and Stenchikov et al. (1996) in the analysis of
26 convective transport of trace gases. The cloud model is nonhydrostatic and contains detailed
27 representation of cloud microphysical processes. Two and three dimensional versions of the
28 model have been applied in transport analyses. The initial conditions for the model are usually
29 from a sounding of temperature, water vapor and winds representative of the region of storm
30 development. Model-generated wind fields can be used to perform air parcel trajectory analyses
31 and tracer advection calculations. Once transport calculations are performed for O3 precursors, a
January 2005 AX2-73 DRAFT-DO NOT QUOTE OR CITE
-------
1 1-D photochemical model was employed to estimate O3 production rates in the outflow air from
2 the convection. These rates were then compared with those prior to convection to determine an
3 enhancement factor due to convection.
4 Such methods were used by Pickering et al. (1992b) to examine transport of urban plumes
5 by deep convection. Transport of the Oklahoma City plume by the 10 - 11 June 1985
6 PRE-STORM squall line was simulated with the 2-D GCE model. In this event forward
7 trajectories from the boundary layer at the leading edge of the storm showed that almost 75% of
8 the low level inflow was transported to altitudes exceeding 8 km. Over 35% of the air parcels
9 reached altitudes over 12 km. Tracer transport calculations were performed for CO, NOX, O3,
10 and hydrocarbons. The 3-D version of the GCE model has also been run for the 10-11 June
11 1985 PRE-STORM case. Free tropospheric O3 production enhancement of a factor of 2.5 for
12 Oklahoma rural air and -4 for the Oklahoma City case were calculated.
13 Stenchikov et al. (1996) used the 2-D GCE model to simulate the North Dakota storm
14 observed by Poulida et al. (1996). This storm showed the unusual feature of an anvil formed
15 well within the stratosphere. The increase of CO and water vapor above the altitude of the
16 preconvective tropopause was computed in the model. The total mass of CO across the model
17 domain above this level increased by almost a factor of two during the convective event. VOCs
18 injected into the lower stratosphere could enhance O3 production there. Downward transport of
19 O3 from the stratosphere was noted in the simulation in the rear anvil.
20 Regional estimates of deep convective transport have been made through use of a traveling
21 1-D model, regional transport models driven by parameterized convective mass fluxes from
22 mesoscale meteorological models, and a statistical-dynamical approach. Pickering et al. (1992c)
23 developed a technique which uses a combination of deep convective cloud cover statistics from
24 the International Satellite Cloud Climatology Project (ISCCP) and convective transport statistics
25 from GCE model simulations of prototype storms to estimate the amount of CO vented from the
26 planetary boundary layer (PEL) by deep convection. This statistical-dynamical approach was
27 used by Thompson et al. (1994) to estimate the convective transport component of the boundary
28 layer CO budget for the central United States (32.5° - 50° N, 90° - 105° W) for the month of
29 June. They found that the net upward deep convective flux (-18 x 10s kg-CO/month) and the
30 shallow convective flux (-16 x io5 kg-CO/month) to the free troposphere accounted for about
31 80% of the loss of CO from the PEL. These losses roughly balanced horizontal transport of CO
January 2005 AX2-74 DRAFT-DO NOT QUOTE OR CITE
-------
1 (-28 x 105 kg-CO/month), the oxidation of hydrocarbons (~8 x 10s kg-CO/month) and
2 anthropogenic and biogenic emissions (~8 + ~1 x 10s kg-CO/month) into the PEL in the central
3 United States. In this respect the central United States acts as a "chimney" for venting CO and
4 other pollutants.
5 Regional chemical transport models have been used for applications such as simulations of
6 photochemical O3 production, acid deposition, and fine particulate matter. Walcek et al. (1990)
7 included a parameterization of cloud-scale aqueous chemistry, scavenging, and vertical mixing
8 in the chemistry model of Chang et al. (1987). The vertical distribution of cloud microphysical
9 properties and the amount of subcloud-layer air lifted to each cloud layer are determined using a
10 simple entrainment hypothesis (Walcek and Taylor, 1986). Vertically-integrated O3 formation
11 rates over the northeast U.S. were enhanced by -50% when the in-cloud vertical motions were
12 included in the model.
13 Wang et al. (1996) simulated the 10-11 June 1985 PRE-STORM squall line with the
14 NCAR/Penn State Mesoscale Model (MM5; Grell et al., 1994; Dudhia et al., 1993). Convection
15 was parameterized as a subgrid-scale process in MM5 using the Kain and Fritsch (1993) scheme.
16 Mass fluxes and detrainment profiles from the convective parameterization were used along with
17 the 3-D wind fields in CO tracer transport calculations for this convective event. The U.S.
18 Environmental Protection Agency has developed a Community Multiscale Air Quality (CMAQ)
19 modeling system that uses MM5 with the Kain-Fritsch convective scheme as the dynamical
20 driver (Ching et al., 1998).
21 Convective transport in global chemistry and transport models is treated as a subgrid-scale
22 process that is parameterized typically using cloud mass flux information from a general
23 circulation model or global data assimilation system. While GCMs can provide data only for a
24 "typical" year, data assimilation systems can provide "real" day-by-day meteorological
25 conditions, such that CTM output can be compared directly with observations of trace gases.
26 The NASA Goddard Earth Observing System Data Assimilation System (GEOS-1 DAS and
27 successor systems; Schubert et al., 1993; Bloom et al., 1996) provides archived global data sets
28 for the period 1980 to present, at 2° x 2.5° or better resolution with 20 layers or more in the
29 vertical. Convection is parameterized with the Relaxed Arakawa-Schubert scheme (Moorthi and
30 Suarez, 1992). Pickering et al. (1995) showed that the cloud mass fluxes from GEOS-1 DAS are
31 reasonable for the 10-11 June 1985 PRE-STORM squall line based on comparisons with the
January 2005 AX2-75 DRAFT-DO NOT QUOTE OR CITE
-------
1 GCE model (cloud-resolving model) simulations of the same storm. In addition, the GEOS-1
2 DAS cloud mass fluxes compared favorably with the regional estimates of convective transport
3 for the central U.S. presented by Thompson et al. (1994). However, Allen et al. (1997) have
4 shown that the GEOS-1 DAS overestimates the amount and frequency of convection in the
5 tropics and underestimates the convective activity over midlatitude marine storm tracks.
6
7 AX2.3.3 Nocturnal Low-Level Jets
8 Nocturnal low level jets (LLJ) are coincident with synoptic weather patterns involved with
9 high O3 episodes implying that they may play an important role in the formation of severe O3
10 events (Rao and Zurbenko, 1994). LLJ can transport pollutants hundreds of kilometers from
11 their sources. Figure AX2-10 shows the evolution of the planetary boundary layer (PEL) over
12 land during periods when high-pressure weather patterns prevail (Stull, 1999). During synoptic
13 weather patterns with stronger zonal flow, a schematic of the boundary layer could look quite
14 different with generally more uniform mixing present. As can be seen from Figure AX2-10, the
15 PEL can be divided into three sub-layers: a turbulent mixed layer (typically present during
16 daylight hours), a less turbulent residual layer which occupies space that was formerly the mixed
17 layer, and a nocturnal, stable boundary layer that has periods of sporadic turbulence (Stull,
18 1999). The LLJ forms in the residual layer. It is important to note, that during the nighttime, the
19 PEL often comprises thin, stratified layers with different physical and chemical properties (Stull,
20 1988).
21 At night, during calm conditions, the planetary boundary layer is stably stratified and as a
22 result verticle mixing is inhibited. On cloud-free evenings the LLJ begins to form shortly after
23 sunset. The wedge of cool air in the stable nocturnal boundary layer decouples the surface layer
24 from the residual layer and acts like a smooth surface allowing the air just above it (in the
25 residual layer) to flow rapidly past the inversion mostly unencumbered by surface friction (Stull,
26 1999). As the sun rises, its energy returns to heat the land and the lower atmosphere begins to
27 mix as the warm air rises. The jet diminishes as the nocturnal temperature inversion erodes and
28 surface friction slows winds speeds. If stable synoptic conditions persist, the same conditions
29 the next night could allow the low-level jet to reform with equal strength and similar
30 consequences. LLJ formation results in vertical wind shear that induces mixing between the
31 otherwise stratified layers.
January 2005 AX2-76 DRAFT-DO NOT QUOTE OR CITE
-------
2-
« H
0-
Free Atmosphere
Cloud Layer
Cloud Layer
Stable Nocturnal Boundary Layer
Surface Layer
\
Afternoon
I
Sunset
Midnight
r
Sunrise
Mixed
Layer
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 (1999) Figures 1.7 and 1.12.
1
2
3
4
5
6
7
8
9
10
1 1
12
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 AX2-1 1 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
January 2005
AX2-77
DRAFT-DO NOT QUOTE OR CITE
-------
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).
1 secondary maxima were, on average, 10% of the next day's O3 maximum but at times could be
2 as much as 80% of the maximum (Corsmeier et al., 1997). The secondary O3 maxima were well
3 correlated with an increase in wind speed and wind shear. The increased vertical shear over the
4 very thin layer results in mechanical mixing that leads a downward flux of O3 from the residual
5 to the near surface layer (see Figures AX2-12 and AX2-13). Analysis of wind profiles from
6 aerological stations in northeastern Germany revealed the spatial extent of that particular LLJ
7 was up to 600 km in length and 200 km in width. The study concluded the importance of O3
8 transport by low level jets was twofold: O3 and other pollutants could be transported hundreds
9 of kilometers at the j et core level during the night and then mixed to the ground far from their
10 source region. Salmond and McKendry (2002) also observed secondary O3 maxima (in the
11 Lower Fraser Valley, British Columbia) associated with low level jets that occasionally
January 2005
AX2-78
DRAFT-DO NOT QUOTE OR CITE
-------
c
o
Surface Ozone During Low Level Jet Periods
5 1.0-
0)
o
c
o
o
n
O
•a
o
o
q
(N
o
o
q
66
o
CM
o
q
o
o
o
q
CO
o
o
q
cb
o
Figure AX2-12.
Time of Day (Local)
Schematic diagram showing the diurnal behavior of ozone and the
development of secondary ozone 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- 5 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).
January 2005
AX2-79
DRAFT-DO NOT QUOTE OR CITE
-------
1 exceeded half the previous day's maximum O3 concentration. The largest increases in surface O3
2 concentration occurred when boundary layer turbulence coincided with O3 levels greater than
3 80 parts per billion were observed aloft. In addition, the study suggests horizontal transport
4 efficiency during a low level jet event could be as much as six times greater than transport with
5 light winds without an LLJ. Reitebuch et al. (2000) observed secondary O3 maxima associated
6 with low-level jet evolution in an urban area in Germany. The notion that O3 was transported
7 downward from the residual layer to the surface was supported by observed decreases in
8 concentrations of NO, NO2 and CO in the residual layer during secondary O3 maxima. Unlike
9 O3 in the residual layer, concentrations of NO, NO2, and CO should be lower than those found
10 nearer the surface (Reitebuch et al., 2000; Seinfeld and Pandis, 1998). As in other studies, wind
11 speed and directional shear were detected during these events. Calculations of the average wind
12 speed and duration of the LLJ suggested that pollutants were transported several hundred
13 kilometers. A study of the PEL and the vertical structure of O3 observed at a costal site in Nova
14 Scotia described how temperature and differences of surface roughness in a marine environment
15 can induce LLJ formation and pollution transport (Gong et al., 2000). In this case, rather strong
16 horizontal sea surface temperature gradients provided the necessary baroclinic forcing.
17 While the studies mentioned above have shed light on the possible role of the LLJ in the
18 transport of O3 and its precursors, quantitative statements about the significance of the LLJ in
19 affecting local and regional O3 budgets can not yet be made. This inability reflects the lack of
20 available data for wind profiles in the planetary boundary layer in areas where LLJ are likely to
21 occur and because of the inadequacy of numerical models in simulating their occurrence.
22
23 AX2.3.4 Intercontinental Transport of Ozone and Other Pollutants
24 AX2.3.4.1 The Atmosphere/Ocean Chemistry Experiment, AEROCE
25 The AEROCE experiment, initiated in the early 1990's set out to examine systematically
26 the chemistry and meteorology leading to the trace gas and aerosol composition over the North
27 Atlantic Ocean. One particular focus area was to determine the relative contribution of
28 anthropogenic and natural processes to the O3 budget and oxidizing capacity of the troposphere
29 over the North Atlantic Ocean. Early results using isentropic back trajectories suggested that
30 periodic pulses of O3 mixing ratios up to 80 ppb were associated with large-scale subsidence
31 from the mid-troposphere, favoring a natural source (Oltmans and Levy, 1992). Moody et al.
January 2005 AX2-80 DRAFT-DO NOT QUOTE OR CITE
-------
1 (1995) extended this work with a five-year seasonal climatology and found the highest
2 concentrations of O3 were always associated with synoptic scale post-frontal subsidence off the
3 North American continent behind cold fronts, and this pattern was most pronounced in the
4 April-May time frame. These post-frontal air masses had uniformly low humidity and high
5 concentrations of 7Be, a cosmogenic tracer produced in the upper troposphere and lower
6 stratosphere. However, the pulsed occurrence of these postfrontal air masses also frequently
7 delivered enhanced concentrations of species such as SO4=, NO3 , 210Pb, etc. suggesting a
8 component originating in North America. In a subsequent analysis of data from one year (1992)
9 when CO observations were available, Dickerson et al. (1995) concluded that anthropogenic
10 sources made a significant contribution to surface O3, and using a simple mixing model they
11 determined that 57% of the air had a continental boundary layer origin.
12 Based on these observations of the synoptically modulated concentrations, AEROCE
13 conducted an aircraft and ozonesonde intensive in the spring of 1996. The intention was to
14 adopt a meteorologically informed sampling strategy to clearly distinguish the characteristics of
15 air masses ahead of and behind eastward progressing cold fronts. Sixteen research flights were
16 conducted with the University of Wyoming King Air research aircraft. The goal was to
17 differentiate the sources of enhanced O3 mixing ratios observed on Bermuda after the passage of
18 cold fronts, and to identify the major processes controlling the highly variable O3 mixing ratios
19 in the mid-to-upper troposphere over eastern North America and the North Atlantic Ocean
20 during April and May. In addition to aircraft flights, near-daily ozonesondes were launched in a
21 quasi-zonal transect from Purdue, Indiana, to Charlottesville, Virginia to Bermuda. An effort
22 was made to time the release of ozonesondes to cleanly differentiate pre and post-frontal air
23 masses.
24 In several aircraft flights, the presence at altitude of distinct layers of air with elevated
25 concentrations of nonmethane hydrocarbons (NMHCs) attested to the dynamic vertical mixing
26 associated with springtime frontal activity. Layers of mid-tropospheric air of high O3 (140 ppb)
27 and low background NMHC mixing ratios (1.44 ppbv ethane, 0.034 ppbv propene, 0.247 ppbv
28 propane, and 0.034 ppbv isobutene, 0.041 ppbv n-butane, 0.063 ppbv benzene, 0.038 ppbv
29 toluene) were indicative of descending, stratospherically influenced air on a flight to the east of
30 Norfolk, VA on April 24 (alt 4600m). However layers of elevated NMHC concentrations
31 (1.88 ppbv ethane, 0.092 ppbv propene, 0.398 ppbv propane, 0.063 ppbv isobutene, 0.075 ppbv
January 2005 AX2-81 DRAFT-DO NOT QUOTE OR CITE
-------
1 n-butane, 0.106 ppbv benzene, 0.0102 ppbv toluene) occurred along with 60-70 ppbv of O3 on a
2 flight west of Bermuda April 28 (alt. 4100m), indicating air had been lofted from the continental
3 boundary layer. Meteorological evidence, supported by ozonesonde observations and earlier
4 King Air flights, indicated that stratosphere/troposphere exchange associated with an upstream
5 frontal system had injected and advected dry, O3-rich air into the mid-troposhere region over the
6 continent. This subsiding air mass provided deep layers of enhanced O3 in the offshore,
7 postfrontal area. Convection from a developing (upwind) system lifted continental boundary
8 layer air into the proximity of the dry, subsiding air layer (Prados, et al., 1999). This resulted in
9 a mixture of high concentrations of anthropogenic pollutants along with naturally enhanced O3.
10 Ozone mixing ratios exceeded those attributable to boundary layer venting or in-transit
11 photochemical production. These meteorological processes led to pollution and stratospherically
12 enhanced O3 co-occurring in post-frontal air masses over the North Atlantic Ocean. A similar
13 event in February 1999 was observed by Parrish et al. (2000). It confirmed the occurrence of
14 thin layers of anthropogenic and stratospheric air that subsequently mix. These results, along
15 with recent modeling studies suggest that North American pollution clearly does contribute to
16 the periodic influx of less-than-pristine air observed in the marine boundary layer over Bermuda
17 (e.g., Li et al., 2002) and yet these incursions are not inconsistent with observing enhancements
18 in O3 due to stratospheric exchange.
19 The ozonesonde climatology of AEROCE clearly established that O3 mixing ratios were
20 always enhanced and increased with height in post-frontal air masses. Postfrontal O3 in the
21 lower troposphere over Bermuda originates in the postfrontal midtroposphere over the continent,
22 supporting the hypothesis that naturally occurring stratospheric O3 makes a contribution to air in
23 the marine boundary layer (Cooper et al., 1998). A schematic of the meteorological processes
24 responsible for the close proximity of natural and man-made O3 can be seen in Figure AX2-8
25 from Prados (2000). Cold fronts over North America tend to be linked in wave-like patterns
26 such that the subsidence behind one front may occur above with intrusions of convection ahead
27 of the next cold front. Pollutants, including VOC and NOx, precursors to O3, may be lofted into
28 the mid-to-upper troposphere where they have the potential to mix with layers of air descending
29 from O3-rich but relatively unpolluted upper troposphere and lower stratosphere. Through this
30 complex mechanism, both stratospheric and photochemically produced O3 may be transported to
31 the remote marine environment where they have large-scale impacts on the radiative and
January 2005 AX2-82 DRAFT-DO NOT QUOTE OR CITE
-------
1 chemical properties of the atmosphere. Recent three-dimensional modeling studies of air mass
2 motion over the Pacific provide further evidence that these complex mechanisms are indeed
3 active (Cooper et al., 2004b).
4
5 AX2.3.4.2 The North Atlantic Regional Experiment, NARE
6 NARE was established by the International Global Atmospheric Chemistry Project to study
7 the chemical processes occurring in the marine troposphere of the North Atlantic, the marine
8 region expected to be the most impacted by industrial emissions from eastern North America and
9 western Europe. Surface measurements from several surface sites were initiated in 1991, with
10 major field intensives in summer 1993, spring 1996, early autumn 1997 and a few winter flights
11 in 1999. In the summer of 1993, airborne and ground-based measurements of O3 and O3
12 precursors were made in the North Atlantic region by an international team of scientists to
13 determine how the continents that rim the North Atlantic are affecting atmospheric composition
14 on a hemispheric scale (Fehsenfeld et al., 1996a; Fehsenfeld et al., 1996b). The focus of NARE
15 was to investigate the O3 budget of the North Atlantic region. Previous observations indicated
16 that the O3 produced from anthropogenic sources is greater than that reaching the lower
17 troposphere from the stratosphere and that O3 derived from anthropogenic pollution has a
18 hemisphere wide effect at northern mid latitudes. This study was performed to better quantify
19 the contribution of continental sources to the O3 levels over the North Atlantic.
20 Buhr et al. (1996) measured O3, CO, NO, and NOy as well as meteorological parameters
21 aboard the NCAR King Air in August 1993 during 16 flights over and near the Gulf of Maine.
22 They found that O3 produced from anthropogenic precursors was dominant throughout the
23 experimental region below 1500 m, in altitude.
24 The National Research Council of Canada Twin Otter aircraft was used to measure the O3
25 and related compounds in the summertime atmosphere over southern Nova Scotia (Kleinman
26 et al., 1996a; Kleinman et al., 1996b). Forty-eight flights were performed, primarily over the
27 surface sampling site in Chebogue Point, Nova Scotia, or over the Atlantic Ocean. They found
28 that a wide variety of air masses with varying chemical content impact Nova Scotia. The effect
29 depends on flow conditions relative to the locations of upwind emission regions and the degree
30 of photochemical processing associated with transport times ranging from about 1 - 5 d. Moist
31 continental boundary layer air with high concentrations of O3 and other anthropogenic pollutants
January 2005 AX2-83 DRAFT-DO NOT QUOTE OR CITE
-------
1 was advected to Nova Scotia in relatively thin vertical layers, usually with a base altitude of
2 several hundred meters. Dry air masses with high concentrations of O3 often had mixed
3 boundary layer and upper atmosphere source regions. When a moist and dry air mass with the
4 same photochemical age and O3 concentration were compared, the dry air mass had lower
5 concentrations of NOy and aerosol particles, which was interpreted as evidence for the selective
6 removal of soluble constituents during vertical lifting.
7 Due to strong, low-level temperature inversions over the North Atlantic, near surface air is
8 often unrepresentative of the eastward transport of the North American plume because of a
9 decoupling from the air transported aloft (Kleinman et al., 1996a; Daum et al., 1996; Angervine
10 et al., 1996). Pollution plumes were observed in distinct strata up to 1 km. Plume chemical
11 compositions were consistent with the occurrence of considerable photochemical processing
12 during transit from source regions over the eastern seaboard of the U.S. Ozone concentrations
13 reached 150 ppbv, NOX conversion to its oxidation products exceeded 85%, and high hydrogen
14 peroxide concentrations were observed (median 3.6 ppbv, maximum 11 ppbv). CO and O3
15 concentrations were well correlated (R2 = 0.64) with a slope (0.26) similar to previous
16 measurements in photochemically aged air (Parrish et al., 1998). Ozone depended nonlinearly
17 on the NOX oxidation product concentration, but there was a correlation (r2 = 0.73) found
18 between O3 and the concentration of radical sink species as represented by the quantity
19 ((NOy-NOx) + 2H2O2)
20 Banic et al. (1996) determined that the average mass of O3 transported through an area
21 1 m in horizontal extent and 5 km in the vertical over the ocean near Nova Scotia to be 2.8 g s"1,
22 moving from west to east. Anthropogenic O3 accounted for half of the transport below 1 km,
23 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
24 Moody (1996) analyzed the meteorological conditions during the NARE intensive period
25 (August 1 to September 13, 1993). They determined the ideal meteorological scenario for
26 delivering pollution plumes from the U.S. East Coast urban areas over the Gulf of Maine to the
27 Maritime Provinces of Canada to be warm sector flow ahead of an advancing cold front. In the
28 winter phase of NARE, O3 and CO were measured from the NOAA WP-3D Orion aircraft from
29 St. John' s, Newfoundland, Canada, and Keflavik, Iceland, from February 2 to 25, 1999 (Parrish
30 et al., 2000). In the lower troposphere over the western North Atlantic Ocean, the close
31 proximity of air masses with contrasting source signatures was remarkable. High levels of
January 2005 AX2-84 DRAFT-DO NOT QUOTE OR CITE
-------
1 anthropogenic pollution immediately adjacent to elevated O3 of stratospheric origin were
2 observed, similar to those reported by Prados et al., (1999). In air masses with differing amounts
3 of anthropogenic pollution, O3 was negatively correlated with CO, which indicates that
4 emissions from surface anthropogenic sources had reduced O3, in this wintertime period, even in
5 air masses transported into the free troposphere.
6 The influence of the origin and evolution of airstreams on trace gas mixing ratios has been
7 studied in great detail for NARE aircraft data. The typical mid-latitude cyclone is composed of
8 four major component airstreams, the warm conveyor belt, the cold conveyor belt, the post cold-
9 front airstream and the dry airstream (Cooper et al., 2001). The physical and chemical
10 processing of trace species was characterized for each airstream, and a conceptual model of a
11 mid-latitude cyclones was developed (Cooper et al., 2002a). This showed how airstreams within
12 midlatitude cyclones drew and exported trace gases from the polluted continental boundary
13 layer, and the stratospherically enhanced mid-troposphere. Using back trajectories, airstream
14 composition was related to the origin and transport history of the associated air mass. The
15 lowest O3 values were associated with airstreams originating in Canada or the Atlantic Ocean
16 marine boundary layer; the highest O3 values were associated with airstreams of recent
17 stratospheric origin. The highest NOy values were seen in polluted outflow from New England
18 in the lower troposphere. A steep and positive O3/NOy slope was found for all airstreams in the
19 free troposphere regardless of air mass origin. Finally, the seasonal variation of photochemistry
20 and meteorology and their impact on trace gas mixing ratios in the conceptual cyclone model
21 was determined (Cooper et al., 2002b). Using a positive O3/CO slope as an indicator of
22 photochemical O3 production, O3 production during late summer-early autumn is associated with
23 the lower troposphere post-cold-front airstream and all levels of the WCB, especially the lower
24 troposphere. However, in the early spring, there is no significant photochemical O3 production
25 for airstreams at any level, and negative slopes in the dry air airstream indicate STE causes the
26 O3 increase in the mid- and upper troposphere.
27 Stohl et al. (2002) analyzed total odd nitrogen (NOy) and CO data taken during NARE in
28 spring 1996 and fall 1997. They studied the removal timescales of NOy originating from surface
29 emissions of NOX and what fraction reached the free troposphere. NOX limits O3 production in
30 the free troposphere and can be regenerated from NO after the primary NOx has been
January 2005 AX2-85 DRAFT-DO NOT QUOTE OR CITE
-------
1 exhausted. It was determined that < 50% of the NOy observed above 3 km came from
2 anthropogenic surface emissions. The rest had to have been emitted in situ.
3 Several studies (e.g., Stohl and Trickl, 1999; Brunner et al., 1998; Schumann et al., 2000;
4 Stohl et al., 2003; Traub et al., 2003) have identified plumes that have originated in North
5 America over Europe and over the eastern Mediterranean basin (e.g., Roelofs et al., 2003; Traub
6 et al., 2003). Modeling studies indicate that North American emissions contribute roughly 20%
7 to European CO levels and 2 to 4 ppb to surface O3, on average. Episodic events, such as forest
8 fires in North America have also been found to result in elevated CO and O3 levels and visible
9 haze layers in Europe (Volz-Thomas, et al., 2003). The O3 is either transported from North
10 America or formed during transport across the North Atlantic Ocean, perhaps as the result of
11 interactions between the photochemical degradation products of acetone with emissions of NOX
12 from aircraft (Bruhl et al., 2000; Arnold et al., 1997). In addition, North American and European
13 pollution is exported to the Arctic. Eckhardt et al. (2003) show that this transport is related to
14 the phase of the North Atlantic Oscillation which has a period of about 20 years.
15
16 AX2.3.5 The Relation of Ozone to Solar Ultraviolet Radiation, Aerosols,
17 and Air Temperature
18 AX2.3.5.1 Solar Ultraviolet Radiation and Ozone
19 The effects of sunlight on photochemical oxidant formation, aside from the role of solar
20 radiation in meteorological processes, are related to its intensity and its spectral distribution.
21 Intensity varies diurnally, seasonally, and with latitude, but the effect of latitude is strong only in
22 the winter. Ultraviolet radiation from the sun plays a key role in initiating the photochemical
23 processes leading to O3 formation and affects individual photolytic reaction steps. However,
24 there is little empirical evidence in the literature, directly linking day-to-day variations in
25 observed UV radiation levels with variations in O3 levels.
26 In urban environments the rate of O3 formation is sensitive to the rate of photolysis of
27 several species including H2CO, H2O2, O3, and especially NO2. Monte Carlo calculations
28 suggest that model calculations of photochemical O3 production are most sensitive to uncertainty
29 in the photolysis rate coefficient for NO2 (Thompson and Stewart, 1991; Baumann et al., 2000).
30 The International Photolysis Frequency Measurement and Modeling Intercomparison (IPMMI)
31 hosted recently by NCAR in Boulder, CO brought together more than 40 investigators from
January 2005 AX2-86 DRAFT-DO NOT QUOTE OR CITE
-------
1 8 institutions from around the world (Bais et al., 2003; Cantrell et al., 2003 and Shelter et al.,
2 2003). They compared direct actinometric measurements, radiometric measurements, and
3 numerical models of photolysis rate coefficients, focusing on O3 to O(1D) and NO2, referred to as
4 j(03)andj(N02).
5 The combination of direct measurements and comparisons to model calculations indicated
6 that for clear skies, zenith angles less than 70°, and low aerosol loadings, the absolute value of
7 the j(NO2) at the Earth's surface is known to better than 10% with 95% confidence. The results
8 suggest that the cross sections of Harder et al. (1997a) may yield more accurate values when
9 used in model calculations of j(NO2). Many numerical models agreed among themselves and
10 with direct measurements (actinometers) and semi-direct measurements (radiometers) when
11 using ATLAS extraterrestrial flux from Groebner and Kerr (2001). The results of IPMMI
12 indicate numerical models are capable of precise calculation of photolysis rates at the surface
13 and that uncertainties in calculated chemical fields arise primarily from uncertainties in the
14 variation of actinic flux with altitude in addition to the impact of clouds and aerosols on
15 radiation.
16
17 AX2.3.5.2 Impact of Aerosols on Radiation and Photolysis Rates and
18 Atmospheric Stability
19 Because aerosol particles influence the UV flux there is a physical link between particles
20 and gases that depends on the concentration, distribution, and refractive index of the particles.
21 Scattering of UV radiation by tropospheric aerosol particles can strongly impact photolysis rates
22 and thus photochemical O3 production or destruction. The effect shows high sensitivity to the
23 properties of the aerosol. Particles in the boundary layer can accelerate photochemistry if the
24 single scattering albedo is near unity, such as for sulfate and ammoniated sulfate aerosols, or
25 inhibit O3 production if the single scattering albedo is low, such as for mineral dust or soot
26 (Dickerson et al., 1997; Jacobson, 1998; Liao et al., 1999; Castro et al., 2001; Park et al., 2001).
27 Any aerosol layer in the free troposphere will reduce photolysis rates in the boundary layer.
28 The interaction of aerosols, photochemistry, and atmospheric thermodynamic processes
29 can impact radiative transport, cloud microphysics, and atmospheric stability with respect to
30 vertical mixing. Park et al. (2001) developed a single-column chemical transport model that
31 simulates vertical transport by convection, turbulent mixing, photochemistry, and interactive
32 calculations of radiative fluxes and photolysis rates. Results from simulations of an episode over
January 2005 AX2-87 DRAFT-DO NOT QUOTE OR CITE
-------
1 the eastern United States showed strong sensitivity to convective mixing and aerosol optical
2 depth. The aerosol optical properties observed during the episode produced a surface cooling of
3 up to 120 W/m2 and stabilized the atmosphere suppressing convection. This suggests two
4 possible feedbacks mechanisms between aerosols and O3-reduced vertical mixing would tend to
5 increase the severity of O3 episodes, while reduced surface temperatures would decrease it.
6
7 AX2.3.5.3 Temperature and Ozone
8 An association between surface O3 concentrations and temperature has been demonstrated
9 from measurements in outdoor smog chambers and from measurements in ambient air.
10 Numerous ambient studies done over more than a decade have reported that successive
11 occurrences or episodes of high temperatures characterize high O3 years (Clark and Karl, 1982;
12 Kelly et al., 1986). Figures AX2-14 and AX2-15 show the daily maximum O3 concentrations
13 versus maximum daily temperature for warm months (May to October), 1988 to 1990, for
14 Atlanta and New York City, NY, and for Detroit, MI, and Phoenix, AZ, respectively. There
15 appears to be an upper-bound on O3 concentrations that increases with temperature. Likewise,
16 Figure AX2-16 shows that a similar qualitative relationship exists between O3 and temperature
17 even at a number of nonurban locations.
18 The notable trend in these plots is the apparent upper-bound to O3 concentrations as a
19 function of temperature. It is clear that, at a given temperature, there is a wide range of possible
20 O3 concentrations because other factors (e.g., cloudiness, precipitation, wind speed) can reduce
21 O3 production rates. The upper edge of the curves may represent a practical upper bound on the
22 maximum O3 concentration achieved under the most favorable conditions. It can be seen that
23 this quantity does increase with temperature, at least for temperatures greater than about 20 °C
24 for all of the cities shown in Figures AX2-14 to AX2-16 except perhaps for Phoenix, AZ.
25 Relationships between peak O3 and temperature also have been recorded by Wunderli and
26 Gehrig (1991) for three locations in Switzerland. At two sites near Zurich, peak O3 increased
27 3 to 5 ppb/°C for diurnal average temperatures between 10 and 25 °C, and little change in peak
28 O3 occurred for temperatures below 10 °C. At the third site, a high-altitude location removed
29 from anthropogenic influence, a much smaller variation of O3 with temperature was observed.
January 2005 AX2-88 DRAFT-DO NOT QUOTE OR CITE
-------
240
U Atlanta, GA
o
E
x
re
're
O
180--
120--
60
S_.
o
I---Q--
10 15 20 25 30 35 40
240
bj New York, NY
£1
O.
O
E
x
re
re
Q
180
120--
60 ,
°°
0 0
_ js f
_8?8_8_°
o__
60
:--V%— -
0 0
-e-o-
-o
o
10 15 20 25 30 35
Daily Maximum Temperature (°C)
40
Figure AX2-14. A scatter plot of daily maximum ozone concentration in (a) Atlanta, GA
and (b) New York NY, versus daily maximum temperature.
1 Some possible explanations for the correlation of O3 with temperature include:
2 (1) Increased photolysis rates under meterological conditions associated with higher
temperatures;
3 (2) Increased H2O concentrations with higher temperatures as this will lead to greater OH
production via R(2-6);
4 (3) Enhanced thermal decomposition of PAN and similar compounds to release NOX at
higher temperatures;
January 2005
AX2-89 DRAFT-DO NOT QUOTE OR CITE
-------
.Q
Q.
to
O
X
(0
(Q
Q
240
180--
120--
£1
Q.
a.
CO
O
E
'x
S
_>,
Q
240
180 --
120--
10
15 20 25 30 35
Daily Maximum Temperature (°C)
40
Figure AX2-15. A scatter plot of daily maximum ozone concentration in (a) Detroit, MI
and (b) Phoenix, AZ versus daily maximum temperature.
1 (4) Increase of anthropogenic hydrocarbon (e.g., evaporative losses) emissions or NOX,
emissions with temperature or both;
2 (5) Increase of natural hydrocarbon emissions (e.g., isoprene) with temperature; and
3 (6) Relationships between high temperatures and stagnant circulation patterns.
4 (7) Advection of warm air enriched with O3.
January 2005
AX2-90
DRAFT-DO NOT QUOTE OR CITE
-------
210
-^ 180-
Q.
^
to
O
E
3
E
1C
re
Q
150 -
120 -
• Ann Arbor, Ml
n Williamsport, PA
^ Mammoth Cave, KY
O McKensie City, ND
Daily Maximum Temperature (°C)
Figure AX2-16. A scatter plot of daily maximum ozone concentration versus daily
maximum temperature for four nonurban sites. The relation with
temperature is still apparent, although the slope is reduced from that
of the urban areas.
1 Cardelino and Chameides (1990) and Sillman and Samson (1995) both identified the
2 temperature-dependent thermal decomposition of PAN as the primary cause of the observed
3 O3-temperature relationship. When temperatures are low PAN is relatively stable. Formation of
4 PAN represents a significant sink for NOX (in low NOX rural areas) and radicals (in high NOX
5 urban areas). This has the effect of slowing the rate of O3 production. Sillman and Samson
6 found that the impact of the PAN decomposition rate could explain roughly half of the observed
7 correlation between O3 and temperature. Jacob et al. (1993) found that warm events in summer
8 in the United States were likely to occur during stagnant meteorological conditions, and the
9 concurrence between warm temperatures and meteorological stagnation also explained roughly
10 half of the observed O3-temperature correlation. Other possible causes include higher solar
11 radiation during summer, the strong correlation between biogenic emission of isoprene and
12 temperature, and the somewhat weaker tendency for increased anthropogenic emissions
13 coinciding with warmer temperatures.
January 2005
AX2-91
DRAFT-DO NOT QUOTE OR CITE
-------
1 However, it should also be noted that a high correlation of O3 with temperature does not
2 necessarily imply a causal relation. Extreme episodes of high temperatures (a heat wave) are
3 often multi-day events- high O3 episodes are also multi-day events, concentrations build,
4 temperatures rise, but both are being influenced by larger-scale regional or synoptic
5 meteorological conditions. It also seems apparent, that while there is a trend for higher O3
6 associated with higher temperatures, there is also much greater variance in the range of O3
7 mixing ratios at higher temperatures.
10 AX2.4 THE RELATION OF OZONE TO ITS PRECURSORS AND
11 OTHER OXIDANTS
12 Ozone is unlike many other species whose rates of formation vary directly with the
13 emissions of their precursors. Ozone changes in a nonlinear fashion with the concentrations of
14 its precursors. At the low NOX concentrations found in most environments, ranging from remote
15 continental areas to rural and suburban areas downwind of urban centers the net production of O3
16 increases with increasing NOX. At the high NOX concentrations found in downtown metropolitan
17 areas, especially near busy streets and roadways, and in power plant plumes there is net
18 destruction (titration) of O3 by reaction with NO. In between these two regimes there is a
19 transition stage in which O3 shows only a weak dependence on NOX concentrations. In the high
20 NOX regime, NO2 scavenges OH radicals which would otherwise oxidize VOCs to produce
21 peroxy radicals, which in turn would oxidize NO to NO2. In the low NOX regime, the oxidation
22 of VOCs generates, or at least does not consume, free radicals and O3 production varies directly
23 with NOX. Sometimes the terms VOC limited and NOX limited are used to describe these two
24 regimes. However, there are difficulties with this usage because (1) VOC measurements are not
25 as abundant as they are for nitrogen oxides, (2) rate coefficients for reaction of individual VOCs
26 with free radicals vary over an extremely wide range, and (3) consideration is not given to CO
27 nor to reactions that can produce free radicals without invoking VOCs. The terms NOx-limited
28 and NOx-saturated (e.g., Jaegle et al., 2001) will be used wherever possible to describe these two
29 regimes more adequately. However, the terminology used in original articles will also be kept.
30 The chemistry of OH radicals, which are responsible for initiating the oxidation of hydrocarbons,
31 shows behavior similar to that for O3 with respect to NOX concentrations (Hameed et al., 1979;
January 2005 AX2-92 DRAFT-DO NOT QUOTE OR CITE
-------
1 Pinto et al., 1993; Poppe et al., 1993; Zimmerman and Poppe, 1993). These considerations
2 introduce a high degree of uncertainty into attempts to relate changes in O3 concentrations to
3 emissions of precursors.
4 Various analytical techniques have been proposed that use ambient NOX and VOC
5 measurements to derive information about O3 production and O3-NOX-VOC sensitivity. It has
6 been suggested that O3 formation in individual urban areas could be understood in terms of
7 measurements of ambient NOX and VOC concentrations during the early morning (e.g., National
8 Research Council, 1991). In this approach, the ratio of summed (unweighted by chemical
9 reactivity) VOC to NOX is used to determine whether conditions were NOx-sensitive or VOC
10 sensitive. This procedure is inadequate because it omits many factors that are recognized as
11 important for O3 production: the impact of biogenic VOCs (which are not present in urban
12 centers during early morning); important individual differences in the ability of VOCs to
13 generate free radicals (rather than just total VOC) and other differences in O3 forming potential
14 for individual VOCs (Carter, 1995); the impact of multiday transport; and general changes in
15 photochemistry as air moves downwind from urban areas (Milford et al., 1994).
16 Jacob et al. (1995) used a combination of field measurements and a chemistry-transport
17 model (CTM) to show that the formation of O3 changed from NOx-limited to NOx-saturated as
18 the season changed from summer to fall at a monitoring site in Shenandoah National Park, VA.
19 Photochemical production of O3 generally occurs simultaneously with the production of various
20 other species: nitric acid (HNO3), organic nitrates, and hydrogen peroxide. The relative rate of
21 production of O3 and other species varies depending on photochemical conditions, and can be
22 used to provide information about O3-precursor sensitivity.
23 There are no hard and fast rules governing the levels of NOX at which the transition from
24 NOx-limited to NOx-saturated conditions occurs. The transition between these two regimes is
25 highly spatially and temporally dependent. Similar responses to NOX additions from commercial
26 aircraft have also been found for the upper troposphere (Bruhl et al., 2000). Bruhl et al. (2000)
27 found that the NOX levels for O3 production versus loss are highly sensitive to the radical sources
28 included in model calculations. They found that the inclusion of only CH4 and CO oxidation
29 leads to a decrease in net O3 production in the North Atlantic flight corridor due to NO emissions
30 from aircraft. However, the inclusion of acetone photolysis was found to shift the maximum in
January 2005 AX2-93 DRAFT-DO NOT QUOTE OR CITE
-------
1 O3 production to higher NOX mixing ratios, thereby reducing or eliminating areas in which there
2 is a decrease in O3 production rates due to aircraft emissions.
3 Trainer et al. (1993) suggested that the slope of the regression line between O3 and
4 summed NOX oxidation products (NOZ, equal to the difference between measured total reactive
5 nitrogen, NOy, and NOX) can be used to estimate the rate of O3 production per NOX (also known
6 as the O3 production efficiency, or OPE). Ryerson et al. (1998, 2001) used measured
7 correlations between O3 and NOZ to identify different rates of O3 production in plumes from
8 large point sources.
9 Sillman (1995) and Sillman and He (2002) identified several secondary reaction products
10 that show different correlation patterns for NOx-limited conditions and NOx-saturated conditions.
11 The most important correlations are for O3 versus NOy, O3 versus NOZ, O3 versus HNO3, and
12 H2O2 versus HNO3. The correlations between O3 and NOy, and O3 and NOZ are especially
13 important because measurements of NOy and NOX are widely available. Measured O3 versus
14 NOZ (Figure AX2-17) shows distinctly different patterns in different locations. In rural areas and
15 in urban areas such as Nashville, TN, O3 shows a strong correlation with NOZ and a relatively
16 steep slope to the regression line. By contrast, in Los Angeles O3 also increases with NOZ, but
17 the rate of increase of O3 with NOZ is lower and the O3 concentrations for a given NOZ value are
18 generally lower.
19 The difference between NOx-limited and NOx-saturated regimes is also reflected in
20 measurements of hydrogen peroxide (H2O2). Hydrogen peroxide production is highly sensitive
21 to the abundance of free radicals and is thus favored in the NOx-limited regime, typical of
22 summer conditions. Differences between these two regimes are also related to the preferential
23 formation of sulfate during summer and to the inhibition of sulfate and hydrogen peroxide
24 during winter (Stein and Lamb, 2003). Measurements in the rural eastern United States (Jacob
25 et al., 1995) Nashville (Sillman et al., 1998), and Los Angeles (Sakugawa and Kaplan, 1989)
26 show large differences in H2O2 concentrations between likely NOx-limited and NOx-saturated
27 locations.
28 The discussion in Section AX2.4.1 centers mainly on the relations among O3, NOX and its
29 oxidation products, represented as NOZ (NOy - NOX) and VOCs derived from the results of field
30 studies. Most of these studies examined processes occurring in power plant and urban plumes.
31
January 2005 AX2-94 DRAFT-DO NOT QUOTE OR CITE
-------
x
X
x
X
X X
X
X
X
X
X
10
20
NOZ (ppb)
30
40
Figure AX2-17. Measured values of O3 and NOZ (NOy - NOX) during the afternoon at
rural sites in the eastern United States (grey 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 et al. (1993), Sillmanet al. (1997, 1998, 2003).
1 AX2.4.1 Summary of Results for the Relations Among Ozone, its Precursors
2 and Other Oxidants from Recent Field Experiments
3 AX2.4.1.1 Results from the Southern Oxidant Study and Related Experiments
4 The Southern Oxidant Study (SOS) was initiated to describe the sources, variation, and
5 distribution of O3 and its precursors in the southeastern United States during the summer season
6 (Hiibler et al., 1998; Meagher et al., 1998; Goldan et al., 2000). Specific issues that were
7 addressed included: (1) the role of biogenic VOC and NOX emissions on local and regional O3
8 production, (2) the effect of urban-rural exchange/interchange on local and regional O3
9 production, (3) sub-grid-scale photochemical and meteorological processes, and (4) the
10 provision of a high-quality chemical and meteorological data set to test and improve observation
11 and emission-based air quality forecast models. Some of the more significant findings of the
12 1994 to 1995 studies include the following: (1) Ozone production in Nashville was found to be
13 close to the transition between NOx-limited and NOx-saturated regimes. (2) The number of
14 molecules of O3 produced per molecule of NOX oxidized in power plant plumes, or the ozone
January 2005
AX2-95
DRAFT-DO NOT QUOTE OR CITE
-------
1 production efficiency (OPE) was found to be inversely proportional to the NOX emission rate,
2 with the plants having the highest NOX emissions exhibiting the lowest OPE. (3) During
3 stagnant conditions, winds at night dominated pollutant transport and represent the major
4 mechanism for advecting urban pollutants to rural areas — specific findings follow.
5 As part of SOS, the Tennessee Valley Authority's instrumented helicopter conducted
6 flights over Atlanta, Georgia to investigate the evolution of the urban O3 plume (Imhoff et al.,
7 1995). Ozone peak lev els occurred at 20 - 40 km downwind of the city center. The OPE
8 obtained from five afternoon flights ranged between 4 and 10 molecules of O3 per molecule
9 ofNOx.
10 Berkowitz and Shaw (1997) measured O3 and its precursors at several altitudes over a
11 surface site near Nashville during SOS to determine the effects of turbulent mixing on
12 atmospheric chemistry. Early morning measurements of O3 aloft revealed values near 70 ppb,
13 while those measured at the surface were closer to 25 ppb. As the daytime mixed layer
14 deepened, surface O3 values steadily increased until they reached 70 ppb. The onset of
15 turbulence increased isoprene mixing ratios aloft by several orders of magnitude and affected the
16 slope of O3 as a function of NOy for each of the flight legs. Measurements from nonturbulent
17 flight legs yielded slopes that were considerably steeper than those from measurements made in
18 turbulence. This study shows that the concentration of O3 precursors aloft is dependent on the
19 occurrence of turbulence, and turbulent mixing could explain the evolution of O3 concentrations
20 at the surface. In general, conclusions regarding pollutant concentrations must account for both
21 chemical and local dynamic processes.
22 Gillani et al. (1998) analyzed data from instrumented aircraft during SOS that flew through
23 the plumes of three large, tall-stack, base-load, Tennessee Valley Authority (TVA) coal-fired
24 power plants in northwestern Tennessee. They determined that plume chemical maturity and
25 peak O3 and NOZ production occurred within 30 to 40 km and 4 hours of summer daytime
26 convective boundary layer (CBL) transport time for a coal-fired power plant in the Nashville,
27 TN urban O3 nonattainment area (Gallatin). For a rural coal-fired power plant in an isoprene-
28 rich forested area about 100 km west of Nashville (Cumberland), plume chemical maturity and
29 peak O3 and NOZ production were realized within approximately 100 km and 6 hours of CBL
30 transport time. Their findings included approximately 3 molecules of O3 and more than
31 0.6 molecules of NOZ may be produced in large isolated rural power plant plumes (PPPs) per
January 2005 AX2-96 DRAFT-DO NOT QUOTE OR CITE
-------
1 molecule of NOX release; the corresponding peak yields of O3 and NOZ may be significantly
2 greater in urban PPPs. Both power plants can contribute as much as 50 ppb of excess O3 to the
3 Nashville area, raising the local levels to well above 100 ppb. Also using aircraft data collected
4 during SOS, Ryerson et al. (1998) concluded that the lower and upper limits to O3 production
5 efficiency in the Cumberland and Paradise PPPs (located in rural Tennessee) were 1 and
6 2 molecules of O3 produced per molecule of NOX emitted. The estimated lower and upper limits
7 to O3 production efficiency in the Johnsonville PPP (also located in rural Tennessee) were
8 higher, at 3 and 7.
9 The NOAA airborne O3 lidar provided detailed, three-dimensional lower tropospheric O3
10 distribution information during June and July 1995 in the Nashville area (Senff et al., 1998;
11 Alvarez et al., 1998). The size and shape of power plant plumes as well as their impacts on O3
12 concentration levels as the plume is advected downwind were studied. Specific examples
13 include: the July 7 Cumberland plume that was symmetrical and confined to the boundary layer,
14 and the July 19 Cumberland plume that was irregularly shaped with two cores, one above and
15 the other within the boundary layer. The disparate plume characteristics on these two days were
16 the result of distinctly different meteorological conditions. Ozone in the plume was destroyed at
17 a rate of 5 to 8 ppbv h'1 due to NOX titration close to the power plant, while farther downwind,
18 O3 was produced at rates between 1.5 and 4 ppbv h"1. The lidar O3 measurements compared
19 reasonably well with in situ values, with the average magnitude of the offsets over all the flights
20 at 4.3 ppbv (7%).
21 The highest O3 concentrations observed during the 1995 SOS in middle Tennessee
22 occurred during a period of strong, synoptic-scale stagnation from July 11 through July 15. This
23 massive episode covered most of the eastern United States (e.g., Ryan et al., 1998). During this
24 time, the effects of vertical wind profiles on the buildup and transport of O3 were studied by
25 Banta et al. (1998) using an airborne differential absorption lidar (DIAL) system. Vertical cross
26 sections showed O3 concentrations exceeding 120 ppb extending to nearly 2 km above ground
27 level, but that O3 moved little horizontally. Instead, it formed a dome of pollution over or near
28 Nashville. Due to the stagnant daytime conditions (boundary layer winds ~1 to 3 m s"1),
29 nighttime transport of O3 became disproportionately important. At night, in the layer between
30 100 and 2000 m AGL (which had been occupied by the daytime mixed layer), the winds could
31 be accelerated to 5 to 10 m s"1 as a result of nocturnal decoupling from surface friction. Data
January 2005 AX2-97 DRAFT-DO NOT QUOTE OR CITE
-------
1 from surface and other aircraft measurements taken during this period suggest that the
2 background air and the edges of the urban plume were NOX sensitive and the core of the urban
3 plume was hydrocarbon sensitive (Valente et al., 1998). Also revealed was the fact that the
4 surface monitoring network failed to document the maximum surface O3 concentrations. Thus,
5 monitoring networks, especially in medium-sized urban areas under slow transport conditions,
6 may underestimate the magnitude and frequency of urban O3 concentrations greater than
7 120 ppb.
8 Nunnermacker et al. (1998) used both aircraft and surface data from SOS to perform a
9 detailed kinetic analysis of the chemical evolution of the Nashville urban plume. The analysis
10 revealed OH concentrations around 1.2 x 107 cm"3 that consumed 50% of the NOX within
11 approximately 2 hours, at an OPE of 2.5 to 4 molecules for each molecule of NOX.
12 Anthropogenic hydrocarbons provided approximately 44% of the fuel for O3 production by the
13 urban plume.
14 Surface and aircraft observations of O3 and O3 precursors were compared during SOS to
15 assess the degree to which mid-day surface measurements may be considered representative of
16 the larger planetary boundary layer (PEL) (Luke et al., 1998). Overall agreement between
17 surface and aircraft O3 measurements was excellent in the well-developed mixed layer
18 (r2 = 0.96), especially in rural-regional background air and under stagnant conditions, where
19 surface concentrations change only slowly. Vertical variations in trace gas concentrations were
20 often minimal in the well-mixed PEL, and measurements at the surface always agreed well with
21 aircraft observations up to the level of measurements (460 m above ground level). Under
22 conditions of rapidly varying surface concentrations (e.g., during episodes of power plant plume
23 fumigation and early morning boundary layer development), agreement between surface and
24 aloft was dependent upon the spatial (aircraft) and temporal (ground) averaging intervals used in
25 the comparison. Under these conditions, surface sites were representative of the PEL only to
26 within a few kilometers horizontally.
27 On four days during SOS, air samples were taken in the plume of the Cumberland Power
28 Plant in central Tennessee using an instrumented helicopter to investigate the evolution of
29 photochemical smog (Luria et al., 1999,2000). Twelve crosswind air-sampling traverses were
30 made between 35 and 116 km from this Power Plant on 16 July 1995. Winds, from the west-
31 northwest during the sampling period, directed the plume toward Nashville. Ten of the traverses
January 2005 AX2-98 DRAFT-DO NOT QUOTE OR CITE
-------
1 were performed upwind of Nashville, where the plume was isolated, and two were made
2 downwind of the city. The results indicated that even six hours after the plume left the stacks,
3 excess O3 production was limited to the edges of the plume. Excess O3 production within the
4 plume was found to vary from 20 ppb up to 55 ppb. It was determined that this variation
5 corresponded to differences in ambient isoprene levels. Excess O3 (up to 109 ppbv, 50 to
6 60 ppbv above background), was produced in the center of the plume when there was sufficient
7 mixing upwind of Nashville. The power plant plume apparently mixed with the urban plume
8 also, producing O3 up to 120 ppbv 15 to 25 km downwind of Nashville.
9 Nunnermacker et al. (2000) used data from the DOE G-l aircraft to characterize emissions
10 from a small power plant plume (Gallatin) and a large power plant plume (Paradise) in the
11 Nashville region. Observations made on July 3, 7, 15, 17, and 18, 1995, were compiled, and a
12 kinetic analysis of the chemical evolution of the power plant plumes was performed. OPEs were
13 found to be 3 in the Gallatin and 2 in the Paradise plumes. Lifetimes for NOX (2.8 and 4.2 hours)
14 and NOy (7.0 and 7.7 hours) were determined in the Gallatin and Paradise plumes, respectively.
15 These NOX and NOy lifetimes imply rapid loss of NOZ (assumed to be primarily HNO3), with a
16 lifetime determined to be 3.0 and 2.5 hours for the Gallatin and Paradise plumes, respectively.
17
18 AX2.4.1.2 Results from Studies on Biogenic and Anthropogenic Hydrocarbons and
19 Ozone Production
20 Williams et al. (1997) made the first airborne measurements of peroxy-methacrylic nitric
21 anhydride (MPAN), which is formed from isoprene-NOx chemistry and is an indicator of recent
22 O3 production from isoprene and therefore biogenic hydrocarbons (BHC). They also measured
23 peroxyacetic nitric anhydride (PAN), peroxypropionic nitric anhydride (PPN), and O3 to estimate
24 the contributions of anthropogenic hydrocarbons (AHC) and BHC to regional tropospheric O3
25 production.
26 Airborne measurements of MPAN, PAN, PPN, and O3 were made during the 1994 and
27 1995 Nashville intensive studies of SOS to determine the fraction of O3 formed from
28 anthropogenic NOX and BHC (Roberts et al., 1998). It was found that PAN, a general product of
29 hydrocarbon-NOx photochemistry, could be well represented as a simple linear combination of
30 contributions from BHC and AHC as indicated by MPAN and PPN, respectively. The
31 PAN/MPAN ratios, characteristic of BHC-dominated chemistry, ranged from 6 to 10. The
January 2005 AX2-99 DRAFT-DO NOT QUOTE OR CITE
-------
1 PAN/PPN ratios, characteristic of AHC-dominated chemistry, ranged from 5.8 to 7.4. These
2 ratios were used to estimate the contributions of AHC and BHC to regional tropospheric O3
3 production. It was estimated that substantial O3 (50 to 60 ppbv) was produced from BHC when
4 high NOX from power plants was present in areas of high BHC emission.
5
6 AX2.4.1.3 Results of Studies on Ozone Production in Mississippi and Alabama
7 Aircraft flights made in June 1990 characterized the variability of O3 and reactive nitrogen
8 in the lower atmosphere over Mississippi and Alabama. The variety and proximity of sources
9 and the photochemical production and loss of O3 were found to be contributing factors (Ridley
10 et al., 1998). Urban, biomass burning, electrical power plant, and paper mill plumes were all
11 encountered during these flights. Urban plumes from Mobile, AL had OPEs as high as 6 to
12 7 ppbv O3 per ppbv of NOX. Emissions measured from biomass burning had lower efficiencies
13 of 2 to 4 ppbv O3 per ppbv of NOX, but the average rate of production of O3 was as high as
14 58 ppbv hr1 for one fire where the plume was prevented from vertical mixing. Near-source
15 paper mill and power plant plumes showed O3 titration, while far-field observations of power
16 plant plumes showed net O3 production. Early morning observations below a nocturnal
17 inversion provided evidence for the nighttime oxidation of NOX to reservoir species.
18 Aircraft measurements of O3 and oxides of nitrogen were made downwind of Birmingham,
19 AL to estimate the OPE in the urban plume (Trainer et al., 1995). NOX emission rates were
20 estimated at 0.6 x 1025 molecules s"1 with an uncertainty of a factor of 2. During the
21 summertime it was determined that approximately 7 O3 molecules could be formed for every
22 molecule of NOX emitted by the urban and proximately located power plant plumes. The
23 regional O3, the photochemical production of O3 during the oxidation of the urban emissions, and
24 wind speed and direction all combined to dictate the magnitude and location of the peak O3
25 concentrations observed in the vicinity of the Birmingham metropolitan area.
26 Aircraft observations of rural U.S. coal-fired power plant plumes in the middle Mississippi
27 and Tennessee Valleys were used to quantify the nonlinear dependence of tropospheric O3
28 formation on plume NOX concentration, determined by plant NOX emission rate and atmospheric
29 dispersion (Ryerson et al., 2001). The ambient availability of reactive VOC's, primarily
30 biogenic isoprene, was also found to affect O3 production rate and yield in these rural plumes.
31 Plume O3 production rates and yields as a function of NOX and VOC concentrations differed by a
January 2005 AX2-100 DRAFT-DO NOT QUOTE OR CITE
-------
1 factor of 2 or more. These large differences indicate that power plant NOX emission rates and
2 geographic locations play a large role in tropospheric O3 production.
3
4 AX2.4.1.4 The Nocturnal Urban Plume Over Portland, Oregon
5 Aircraft observations of aerosol surface area, O3, NOy and moisture were made at night in
6 the Portland, Oregon urban plume (Berkowitz et al., 2001). Shortly after sunset, O3, relative
7 humidity, NOy and aerosol number density were all positively correlated. However, just before
8 dawn, O3 mixing ratios were highly anti-correlated with aerosol number density, NOy and
9 relative humidity. Back-trajectories showed that both samples came from a common source to
10 the northwest of Portland. The pre-dawn parcels passed directly over Portland, while the other
11 parcels passed to the west of Portland. Several hypotheses were put forward to explain the loss
12 of O3 in the parcels that passed over Portland, including homogeneous gas-phase mechanisms
13 and a heterogeneous mechanism on the aerosol particle surface.
14
15 AX2.4.1.5 Effects of VOC'sin Houston on Ozone Production
16 Aircraft Observations of O3 and O3 precursors over Houston, TX, Nashville, TN; New
17 York, NY; Phoenix, AZ, and Philadelphia, PA showed that despite similar NOX concentrations,
18 high concentrations of VOC's in the lower atmosphere over Houston led to calculated O3
19 production rates that were 2 to 5 times higher than in the other 4 cities (Kleinman et al., 2002).
20 Concentrations of VOC's and O3 production rates are highest in the Ship Channel region of
21 Houston, where one of the largest petrochemical complexes in the world is located. As a result,
22 Houston lays claim to the highest recorded hourly average O3 concentrations in the United States
23 within the last 5 years (in excess of 250 ppb).
24
25 AX2.4.1.6 Chemical and Meteorological Influences on the Phoenix Urban Ozone Plume
26 The interaction of chemistry and meteorology for western cities can contrast sharply with
27 that of eastern cities. A 4-week field campaign in May and June of 1998 in the Phoenix area
28 comprised meteorological and chemical measurements (Fast et al., 2000). Data from models and
29 observations revealed that heating of the higher terrain north and east of Phoenix produced
30 regular, thermally driven circulations during the afternoon from the south and southwest through
31 most of the boundary layer, advecting the urban O3 plume to the northeast. Deep mixed layers
January 2005 AX2-101 DRAFT-DO NOT QUOTE OR CITE
-------
1 and moderate winds aloft ventilated the Phoenix area during the study period so that multi-day
2 buildups of locally produced O3 did not appear to contribute significantly to O3 levels.
3 Sensitivity simulations estimated that 20% to 40% of the afternoon surface O3 mixing ratios
4 (corresponding to 15 to 35 ppb) was due to the entrainment of O3 reservoirs into the growing
5 convective boundary layer. The model results also indicated that O3 production in this arid
6 region is NOx-saturated, unlike most eastern U.S. sites.
7
8 AX2.4.1.7 Transport of Ozone and Precursors on the Regional Scale
9 Instrumented aircraft flights by the University of Maryland in a Cessna 172 and Sonoma
10 Technology, Inc. in a Piper Aztec measured the vertical profiles of trace gases and
11 meteorological parameters in Virginia, Maryland, and Pennsylvania on July 12 - 15, 1995 during
12 a severe O3 episode in the mid-Atlantic region (Ryan et al., 1998). Ozone measured upwind of
13 the urban centers reached 80 to 110 ppbv. Layers of high O3 aloft were associated with local
14 concentration maxima of SO2 and NOy, but not CO or NOX. This, together with a back trajectory
15 analysis, implicated coal-fired power plants in the industrialized Midwest as the source of the
16 photochemically aged air in the upwind boundary of the urban centers. When the PEL over the
17 Baltimore-Washington area deepened, the O3 and O3 precursors that had been transported from
18 the west and northwest mixed with the local emissions and O3 in excess of 125 ppbv was
19 measured at the surface.
20 During the blackout of August 14, 2003 Marufu et al. (2004) measured profiles of ozone,
21 SO2 and CO over areas in western Pennsylvania, Maryland and Virginia. They found notable
22 decreases in O3, SO2, and NOX, over areas affected by the blackout but not over those that were
23 not affected. They also found that CO concentrations aloft were comparable over areas affected
24 and not affected by the blackout. They attributed the differences in concentrations between what
25 was observed and what was expected to the reduction in emissions from power plants mainly in
26 the Ohio Valley. They also reasoned that the CO concentrations were relatively unaffected
27 because they arise from traffic emissions, which may have been largely unaffected by the
28 blackout. However, the blackout also disrupted many industries, small scale emission sources,
29 and rail and air transportation.
30 The Department of Energy G-l aircraft flew in the New York City metropolitan area in
31 the summer of 1996 as part of the North American Research Strategy for Tropospheric
January 2005 AX2-102 DRAFT-DO NOT QUOTE OR CITE
-------
1 Ozone-Northeast effort to ascertain the causes leading to high O3 levels in the northeastern
2 United States (Kleinman et al., 2000). Ozone, O3 precursors, and other photochemically active
3 trace gases were measured upwind and downwind of New York City to characterize the O3
4 formation process and its dependence on NOX and VOCs. During two flights, the wind was
5 south southwesterly and O3 levels reached 110 ppb. On two other flights, the wind was from the
6 north-northwest and O3 levels were not as high. When the G-l observed O3 around 110 ppb, the
7 NOx/NOy ratio measured at the surface was between 0.20 and 0.30, indicating an aged plume.
8
9 AX2.4.1.8 Model Calculations and Aircraft Observations of Ozone Over Philadelphia
10 Regional-scale transport and local O3 production over Philadelphia was estimated using a
11 new meteorological-chemical model (Fast et al., 2002). Surface and airborne meteorological and
12 chemical measurements made during a 30-day period in July and August of 1999 as part of the
13 Northeast Oxidant and Particulate Study were used to evaluate the model performance. Both
14 research aircraft and ozonesondes, during the morning between 0900 and 1100 LST, measured
15 layers of O3 above the convective boundary layer. The model accounted for these layers through
16 upwind vertical mixing the previous day, subsequent horizontal transport aloft, and NO titration
17 of O3 within the stable boundary layer at night. Entrainment of the O3 aloft into the growing
18 convective boundary contributed to surface O3 concentrations. During the study period, most of
19 the O3 appeared to result from local emissions in the vicinity of Philadelphia and the Chesapeake
20 Bay area, but during high O3 episodes, up to 30 to 40% of the O3 was due to regional transport
21 from upwind sources.
22
23 AX2.4.1.9 The Two-Reservoir System
24 Studies described above and aircraft observations made in August 2002 over the mid-
25 Atlantic region show that a two-reservoir system illustrated schematically in Figure AX2-18 may
26 realistically represent both the dynamics and photochemistry of severe, multi-day haze and O3
27 episodes over the eastern United States (Taubman et al., 2004). The first reservoir is the PEL,
28 where most precursor species are injected, and the second is the lower free troposphere (LFT),
29 where photochemical processes are accelerated and removal via deposition is rare. Bubbles of
30 air lifted from urban and industrial sources were rich in CO and SO2, but not O3, and contained
31 greater numbers of externally mixed primary sulfate and black carbon (BC) particles.
January 2005 AX2-103 DRAFT-DO NOT QUOTE OR CITE
-------
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 multi-day ozone 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 ozone exist in the lower free troposphere.
Source: Taubman et al. (2004).
1 Correlations among O3, air parcel altitude, particle size, and relative humidity suggest that
2 greater O3 concentrations and relatively larger particles are produced in the LFT and mix back
3 down into the PEL. Backward trajectories indicated source regions in the Midwest and mid-
4 Atlantic urban corridor, with southerly transport up the urban corridor augmented by the
5 Appalachian lee trough and nocturnal low-level jet (LLJ). This concept of two-reservoirs may
6 facilitate the numerical simulation of multiday events in the eastern United States. A relatively
7 small number of vertical layers will be required if accurate representation of the sub-gridscale
January 2005
AX2-104
DRAFT-DO NOT QUOTE OR CITE
-------
1 transport can be parameterized to represent the actual turbulent exchange of air between the PEL
2 and lower free troposphere.
3
4
5 AX2.5 METHODS USED TO CALCULATE RELATIONS BETWEEN
6 OZONE AND ITS PRECURSORS
7 Atmospheric chemistry and transport models are the major tools used to calculate the
8 relations between O3, its precursors, and other oxidation products. Other techniques, involving
9 statistical relations between O3 and other variables have also been used. Chemistry-transport
10 models (CTM) are driven by emissions inventories for O3 precursor compounds and by
11 meterological fields. Emissions of precursor compounds can be divided into anthropogenic and
12 natural source categories. Natural sources can be further divided into biotic (vegetation,
13 microbes, animals) and abiotic (biomass burning, lightning) categories. However, the distinction
14 between natural sources and anthropogenic sources is often difficult to make as human activities
15 affect directly, or indirectly, emissions from what would have been considered natural sources
16 during the pre-industrial era. Emissions from plants and animals used in agriculture are usually
17 referred to as anthropogenic. Wildfire emissions may be considered natural, except that forest
18 management practices may have led to the buildup of fuels on the forest floor, thereby altering
19 the frequency and severity of forest fires. Needed meteorological quantities such as winds and
20 temperatures are taken from operational analyses, reanalyses, or circulation models. In most
21 cases, these are off-line analyses, i.e., they are not modified by radiatively active species such as
22 O3 and particles generated by the model.
23 A brief overview of atmospheric chemistry-transport models is given in Section AX2.5.1.
24 A discussion of emissions inventories of precursors that are used by these models is given in
25 Section AX2.5.2. Uncertainties in emissions estimates have also been discussed in Air Quality
26 Criteria for Particulate Matter (U.S. Environmental Protection Agency, 2000). So-called
27 'observationally based models' which rely more heavily on observations of the concentrations of
28 important species are discussed in Section AX2.5.3. Chemistry-transport model evaluation and
29 an evaluation of the reliability of emissions inventories are presented in Section AX2.5.4.
30
January 2005 AX2-105 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX2.5.1 Chemistry-Transport Models
2 Atmospheric chemistry-transport models (CTMs) are used to obtain better understanding
3 of the processes controlling the formation, transport, and destruction of O3 and other air
4 pollutants; to understand the relations between O3 concentrations and concentrations of its
5 precursors such as NOX and VOCs; and to understand relations among the concentration patterns
6 of O3 and other oxidants that may also exert health effects. Detailed examination of the
7 concentrations of short-lived species in a CTM can provide important insights into how O3 is
8 formed under certain conditions and can suggest likely avenues for data analysis and future
9 experiments and field campaigns. The dominant processes leading to the formation of O3 in a
10 particular time period (e.g., whether NOX or VOCs were more important, the influence of
11 meteorology, and the influence of emissions from a particular geographic region) and the
12 transformation or formation of other pollutants that would not be possible in an experiment or
13 even the most intensive field campaign could be examined using a CTM. As such, CTMs could
14 play a major role in the interpretation of experimental results by allowing selected aspects of
15 atmospheric chemistry to be examined on a scale not feasible in laboratory or in field
16 experiments.
17 CTMs are also used for determining control strategies for O3 precursors. However, this
18 application has met with varying degrees of success because of the highly nonlinear relations
19 between O3 and emissions of its precursors. CTMs include mathematical descriptions of
20 atmospheric transport, emissions, the transfer of solar radiation through the atmosphere,
21 chemical reactions, and removal to the surface by turbulent motions and precipitation for
22 chemical species of interest. Increasingly, the trend is for these processes to be broken down and
23 handled by other models or sub-models, so a CTM will likely use emissions and meteorological
24 data from at least two other models.
25 There are two major formulations of CTMs in current use. In the first approach,
26 grid-based, or Eulerian, air quality models, the region to be modeled (the modeling domain) is
27 subdivided into a three-dimensional array of grid cells. Spatial derivatives in the species
28 continuity equations are cast in finite-difference form over this grid, and a system of equations
29 for the concentrations of all the chemical species in the model are solved numerically at each
30 grid point. The modeling domain may be limited to a particular airshed or provide global
31 coverage and extend through several major atmospheric layers. Time dependent continuity
January 2005 AX2-106 DRAFT-DO NOT QUOTE OR CITE
-------
1 (mass conservation) equations are solved for each species including terms for transport, chemical
2 production and destruction, and emissions and deposition (if relevant), in each cell. Chemical
3 processes are simulated with ordinary differential equations, and transport processes are
4 simulated with partial differential equations. Because of a number of factors such as the
5 different time scales inherent in different processes, the coupled, nonlinear nature of the
6 chemical process terms, and computer storage limitations all of the terms in the equations cannot
7 be solved simultaneously in three dimensions. Instead, a technique known as operator splitting,
8 in which terms involving individual processes are solved sequentially, is used. In the second
9 application of CTMs, trajectory or Lagrangian models, a large number of hypothetical air parcels
10 are specified as following wind trajectories. In these models, the original system of partial
11 differential equations is transformed into a system of ordinary differential equations.
12 A less common approach is to use a hybrid Lagrangian/Eulerian model, in which certain
13 aspects of atmospheric chemistry and transport are treated with a Lagrangian approach and
14 others are treaded in a Eulerian manner (e.g., Stein et al., 2000). Both modeling approaches have
15 their advantages and disadvantages. The Eulerian approach is more general in that it includes
16 processes that mix air parcels and allows integrations to be carried out for long periods during
17 which individual air parcels lose their identity. There are, however, techniques for including the
18 effects of mixing in Lagrangian models such as FLEXPART (e.g., Zanis et al., 2003), ATTILA
19 (Reithmeir and Sausen, 2002), and CLaMS (McKenna et al., 2002).
20 Major modeling efforts within the U.S. Environmental Protection Agency center on the
21 ModelsS/Community Modeling for Air Quality (CMAQ, Byun et al., 1998) and the Multi Scale
22 Air Quality Simulation Platform (MAQSIP, Odman and Ingram, 1996) whose formulations are
23 based on the regional acid deposition model (RADM, Chang et al., 1987). A number of other
24 modeling platforms using the Lagrangian and Eulerian frameworks have been reviewed in
25 AQCD 96. CTMs currently in use are summarized in the review by Russell and Dennis (2000).
26 The domains of MAQSIP and CMAQ are flexible and can extend from several hundred km to
27 the hemispherical scale. In addition, both of these classes of models allow the resolution of the
28 calculations over specified areas to vary. CMAQ and MAQSIP are both driven by the MM5
29 mesoscale meteorological model (Seaman, 2000 and references therein), though both may be
30 driven by other meteorological models (e.g., RAMS and Eta). Simulations of regional O3
31 episodes have been performed with a horizontal resolution of 4 km. In principle, calculations
January 2005 AX2-107 DRAFT-DO NOT QUOTE OR CITE
-------
1 over limited domains can be accomplished to even finer scales. However, simulations at these
2 higher resolutions require better parameterizations of meteorological processes such as boundary
3 layer fluxes, deep convection and clouds (Seaman, 2000), and knowledge of emissions.
4 Resolution at finer scales will likely be necessary to resolve smaller-scale features such as the
5 urban heat island; sea, bay, and land breezes; and the nocturnal low-level jet.
6 Currently, the most common approach to setting up the horizontal domain is to nest a finer
7 grid within a larger domain of coarser resolution. However, a number of other strategies are
8 currently being developed, such as the stretched grid (e.g., Fox-Rabinowitz et al., 2002) and the
9 adaptive grid. In a stretched grid, the grid's resolution continuously varies throughout the
10 domain, thereby eliminating any potential problems with the sudden change from one resolution
11 to another at the boundary. One must be careful in using such a formulation, because certain
12 parameterizations that are valid on a relatively coarse grid scale (such as convection, for
13 example) are not valid or should not be present on finer scales. Adaptive grids are not set at the
14 start of the simulation, but instead adapt to the needs of the simulation as it evolves (e.g., Hansen
15 et al., 1994). They have the advantage that, if the algorithm is properly set up, the resolution is
16 always sufficient to resolve the process at hand. However, they can be very slow if the situation
17 to be modeled is complex. Additionally, if one uses adaptive grids for separate meteorological,
18 emissions, and photochemical models, there is no reason a priori why the resolution of each grid
19 should match; and the gains realized from increased resolution in one model will be wasted in
20 the transition to another model. The use of finer and finer horizontal resolution in the
21 photochemical model will necessitate finer-scale inventories of land use and better knowledge of
22 the exact paths of roads, locations of factories, and, in general, better methods for locating
23 sources. The present practice of locating a source in the middle of a county or distributing its
24 emissions throughout a county if its location is unknown will likely not be adequate in the future.
25 The vertical resolution of these models continues to improve as more layers are added to
26 capture atmospheric processes and structures. This trend will likely continue because a model
27 with 25 vertical layers, for example, may have layers that are 500 m thick at the top of the
28 planetary boundary layer. Though the boundary layer height is generally determined through
29 other methods, the chemistry in the model is necessarily confined by such layering schemes.
30 Because the height of the boundary layer is of critical importance in simulations of air quality,
31 improved resolution of the boundary layer height would likely improve air quality simulations.
January 2005 AX2-108 DRAFT-DO NOT QUOTE OR CITE
-------
1 The difficulty of properly establishing the boundary layer height is most pronounced when
2 considering tropopause folding events, which are important in determining the chemistry of the
3 background atmosphere. In the vicinity of the tropopause, the vertical resolution of most any
4 large scale model is quite unlikely to be able to capture such a feature. Additionally, any current
5 model is likely to have trouble adequately resolving fine scale features such as the low level jet.
6 Finally, models must be able to treat emissions, meteorology, and photochemistry differently in
7 different areas. Emissions models are likely to need better resolution near the surface and
8 possibly near any tall stacks. Photochemical models, on the other hand, may need better
9 resolution away from the surface and be more interested in resolving the planetary boundary
10 layer height, terrain differences, and other higher altitude features. Meteorological models share
11 some of the concerns of photochemical models, but are less likely to need sufficient resolution to
12 adequately treat a process such as dry deposition beneath a stable nocturnal boundary layer.
13 Whether the increased computational power necessary for such increases in resolution will be
14 ultimately justified by improved results in the meteorological and subsequent photochemical
15 simulations remains to be seen.
16 CTMs require time dependent, three-dimensional wind fields for the time period of
17 simulation. The winds may be either generated by a model using initial fields alone or four
18 dimensional data assimilation can be used to improve the model's meteorological fields (i.e.,
19 model equations can be updated periodically [or "nudged'] to bring results into agreement with
20 observations). Most modeling efforts have focused on simulations of several days duration (a
21 typical time scale for individual O3 episodes), but there have been several attempts at modeling
22 longer periods. For example, Kasibhatla and Chameides (2000) simulated a four month period
23 from May to September of 1995 using MAQSIP. The current trend appears to be toward
24 simulating longer time periods. This will impose additional strains on computational resources,
25 as most photochemical modeling until recently has been performed with an eye toward
26 simulating only summertime episodes of peak O3. With the shift toward modeling an entire year
27 being driven by the desire to understand observations of periods of high wintertime PM (e.g.,
28 Blanchard et al., 2002), models will be further challenged to simulate air quality under
29 conditions for which they may not have been used previously.
30 Chemical kinetics mechanisms (a set of chemical reactions) representing the important
31 reactions that occur in the atmosphere are used in air quality models to estimate the net rate of
January 2005 AX2-109 DRAFT-DO NOT QUOTE OR CITE
-------
1 formation of each pollutant simulated as a function of time. Chemical mechanisms that
2 explicitly treat the chemical reactions of each individual reactive species are too lengthy and
3 demanding of computer resources to be incorporated into three-dimensional atmospheric models.
4 As an example, a master chemical mechanism includes approximately 10,500 reactions
5 involving 3603 chemical species (Derwent et al., 2001 and references therein). Instead,
6 "lumped" mechanisms, that group compounds of similar chemistry together, are used. The
7 chemical mechanisms used in existing photochemical O3 models contain significant uncertainties
8 that may limit the accuracy of their predictions; the accuracy of each of these mechanisms is also
9 limited by missing chemistry. Because of different approaches to the lumping of organic
10 compounds into surrogate groups, chemical mechanisms, can produce somewhat different results
11 under similar conditions. The CB-IV chemical mechanism (Gery et al., 1989), the RADM II
12 mechanism (Stockwell et al., 1990), the SAPRC (e.g., Wang et al., 2000a; Wang et al., 2000b;
13 Carter, 1990) and the RACM mechanisms can be used in CMAQ. Jimenez et al. (2003)
14 provide brief descriptions of the features of the main mechanisms in use and they compared
15 concentrations of several key species predicted by seven chemical mechanisms in a box model
16 simulation over 24 h. The average deviation from the average of all mechanism predictions for
17 O3 and NO over the daylight period was less than 20%, and 10% for NO2 for all mechanisms.
18 However, much larger deviations were found for HNO3, PAN, HO2, H2O2, C2H4 and C5H8
19 (isoprene). An analysis for OH radicals was not presented. The large deviations shown for most
20 species imply differences between the calculated lifetimes of atmospheric species and the
21 assignment of model simulations to either NOX limited or radical limited regimes between
22 mechanisms. Gross and Stockwell (2003) found small differences between mechanisms for
23 clean conditions with differences becoming more significant for polluted conditions, especially
24 for NO2 and organic peroxy radicals. They caution modelers to consider carefully the
25 mechanisms they are using.
26 As CTMs incorporate more processes and knowledge of aerosol- and gas-phase chemistry
27 improves, a "one atmosphere" approach is evolving. For example, CMAQ and PM-CAMx now
28 incorporate some aerosol processes in its code PM-CAMx, and several attempts are currently
29 underway to incorporate chemistry into meteorological models, usually MM5 (e.g., Grell et al.,
30 2000; Liu et al., 2001; Lu et al., 1997 and Park et al., 2001) to examine feedbacks on
January 2005 AX2-110 DRAFT-DO NOT QUOTE OR CITE
-------
1 meteorological processes. This coupling is necessary for cases such as the heavy aerosol loading
2 found in forest fire plumes (Lu et al., 1997 and Park et al., 2001).
3 Spatial and temporal characterizations of anthropogenic and biogenic precursor emissions
4 must be specified as inputs to a CTM. Emissions inventories have been compiled on grids of
5 varying resolution for many hydrocarbons, aldehydes, ketones, CO, NH3, and NOX. Emissions
6 inventories for many species require the application of some algorithm for calculating the
7 dependence of emissions on physical variables such as temperature. For many species,
8 information concerning the temporal variability of emissions is lacking, so long term (e.g.,
9 annual or O3-season) averages are used in short term, episodic simulations. Annual emissions
10 estimates are often modified by the emissions model to produce emissions more characteristic of
11 the time of day and season. Significant errors in emissions can occur if an inappropriate time
12 dependence or a default profile is used. Additional complexity arises in model calculations
13 because different chemical mechanisms are based on different species, and inventories
14 constructed for use with another mechanism must be adjusted to reflect these differences. This
15 problem also complicates comparisons of the outputs of these models because one chemical
16 mechanism will necessarily produce species that are different from those in another and neither
17 output will necessarily agree with the measurements.
18 The effects of clouds on atmospheric chemistry are complex and introduce considerable
19 uncertainty into CTM calculations. Thunderstorm clouds are optically very thick and have
20 major effects on radiative fluxes and thus on photolysis rates. Madronich (1987) provided
21 modeling estimates of the effects of clouds of various optical depths on photolysis rates. In the
22 upper portion of a thunderstorm anvil, photolysis is likely to be enhanced (as much as a factor of
23 2 or more) due to multiple reflections off the ice crystals. In the lower portion of the cloud and
24 beneath the cloud, photolysis is substantially decreased. Thunderstorm updrafts, which contain
25 copious amounts of water, are regions where efficient scavenging of soluble species occurs
26 (Balkanski et al., 1993). Direct field measurements of the amounts of specific trace gases
27 scavenged in observed storms are sparse. Pickering et al. (2001) used a combination of model
28 estimates of soluble species that did not include wet scavenging and observations of these
29 species from the upper tropospheric outflow region of a major line of convection observed near
30 Fiji. Over 90% of the nitric acid and hydrogen peroxide in the outflow air appeared to have been
31 removed by the storm. Walcek et al. (1990) included a parameterization of cloud-scale aqueous
January 2005 AX2-111 DRAFT-DO NOT QUOTE OR CITE
-------
1 chemistry, scavenging, and vertical mixing in the regional scale, chemistry-transport model of
2 Chang et al. (1987). The vertical distribution of cloud microphysical properties and the amount
3 of subcloud-layer air lifted to each cloud layer were determined using a simple entrainment
4 hypothesis (Walcek and Taylor, 1986). Vertically-integrated O3 formation rates over the
5 northeastern United States were enhanced by -50% when the in-cloud vertical motions were
6 included in the model.
7 In addition to wet deposition, dry deposition (the removal of chemical species from the
8 atmosphere by interaction with ground-level surfaces) is an important removal process for
9 pollutants on both urban and regional scales and must be included in CTMs. The general
10 approach used in most models is the three-resistance method, in which where dry deposition is
11 parameterized with a deposition velocity, which is represented as vd = (ra + rb + rj"1 where ra, rb,
12 and rc represent the resistance due to atmospheric turbulence, transport in the fluid sublayer very
13 near the elements of surface such as leaves or soil, and the resistance to uptake of the surface
14 itself. This approach works for a range of substances although it is inappropriate for species
15 with substantial emissions from the surface or for species whose deposition to the surface
16 depends on its concentration at the surface itself. The approach is also modified somewhat for
17 aerosols: the terms rb and rc are replaced with a surface deposition velocity to account for
18 gravitational settling. In their review, Wesley and Hicks (2000) point out several shortcomings
19 of current knowledge of dry deposition. Among those shortcomings are difficulties in
20 representing dry deposition over varying terrain where horizontal advection plays a significant
21 role in determining the magnitude of ra and difficulties in adequately determining a deposition
22 velocity for extremely stable conditions such as those occurring at night (e.g., Mahrt, 1998).
23 Under the best of conditions, when a model is exercised over a relatively small area where dry
24 deposition measurements have been made, models still commonly show uncertainties at least as
25 large as ± 30% (e.g., Massman et al., 1994; Brook et al., 1996; Padro, 1996). Wesley and
26 Hicks (2000) state that an important result of these comparisons is that the current level of
27 sophistication of most dry deposition models is relatively low and relies heavily on empirical
28 data. Still larger uncertainties exist when the surface features are not well known or when the
29 surface comprises a patchwork of different surface types, as is common in the eastern United
30 States.
January 2005 AX2-112 DRAFT-DO NOT QUOTE OR CITE
-------
1 The initial conditions, i.e., the concentration fields of all species computed by a model, and
2 the boundary conditions, i.e., the concentrations of species along the horizontal and upper
3 boundaries of the model domain throughout the simulation must be specified at the beginning of
4 the simulation. It would be best to specify initial and boundary conditions according to
5 observations. However, data for vertical profiles of most species of interest are sparse.
6 Ozonesonde data have been used to specify O3 fields, but the initial and boundary values of
7 many other species are often set equal to zero because of a lack of observations. Further,
8 ozonesondes are thought to be subject to errors in measurement and differences arising from
9 improper corrections for pump efficiency and the solutions used (e.g., Hilsenrath et al., 1986;
10 Johnson et al., 2002). The results of model simulations over larger, preferably global, domains
11 can also be used. As may be expected, the influence of boundary conditions depends on the
12 lifetime of the species under consideration and the time scales for transport from the boundaries
13 to the interior of the model domain (Liu et al., 2001).
14 Each of the model components described above has an associated uncertainty, and the
15 relative importance of these uncertainties varies with the modeling application. The largest
16 errors in photochemical modeling are still thought to arise from the meteorological and
17 emissions inputs to the model (Russell and Dennis, 2000). Within the model itself, horizontal
18 advection algorithms are still thought to be significant source of uncertainty (e.g., Chock and
19 Winkler, 1994) though more recently those errors are thought to have been reduced (e.g., Odman
20 et al., 1996). There are also indications that problems with mass conservation continue to be
21 present in photochemical and meteorological models (e.g., Odman and Russell, 1999); these can
22 result in significant simulation errors. Uncertainties in meteorological variables and emissions
23 can be large enough that they would lead one to make the wrong decision when considering
24 control strategies (e.g., Russell and Dennis, 2000; Sillman et al., 1995). The effects of errors in
25 initial conditions can be minimized by including several days "spin-up" time in a simulation to
26 allow species to come to chemical equilibrium with each other before the simulation of the
27 period of interest begins.
28 While the effects of poorly specified boundary conditions propagate through the model's
29 domain, the effects of these errors remain undetermined. Many regional models specify constant
30 O3 profiles (e.g., 35 ppb) at their lateral and upper boundaries; ozonesonde data, however,
31 indicate that the mixing ratio of O3 increases vertically in the troposphere (to over 100 ppb at the
January 2005 AX2-113 DRAFT-DO NOT QUOTE OR CITE
-------
1 tropopause) and into the stratosphere (e.g., Newchurch et al., 2003). The practice of using
2 constant O3 profiles strongly reduces the potential effects of vertical mixing of O3 from above
3 the planetary boundary layer (via mechanisms outlined in Section AX2.3) on surface O3 levels.
4 The use of an O3 climatology (e.g., Fortuin and Kelder, 1998) might reduce the errors that would
5 otherwise be incurred. Because many meteorological processes occur on spatial scales which
6 are smaller than the grid spacing (either horizontally or vertically) and thus are not calculated
7 explicitly, parameterizations of these processes must be used and these introduce additional
8 uncertainty.
9 Uncertainty also arises in modeling the chemistry of O3 formation because it is highly
10 nonlinear with respect to NOX concentrations. Thus, the volume of the grid cell into which
11 emissions are injected is important because the nature of O3 chemistry (i.e., O3 production or
12 titration) depends in a complicated way on the concentrations of the precursors and the OH
13 radical. The use of ever-finer grid spacing allows regions of O3 titration to be more clearly
14 separated from regions of O3 production. The use of grid spacing fine enough to resolve the
15 chemistry in individual power-plant plumes is too demanding of computer resources for this to
16 be attempted in most simulations. Instead, parameterizations of the effects of subgrid scale
17 processes such as these must be developed; otherwise serious errors can result if emissions are
18 allowed to mix through an excessively large grid volume before the chemistry step in a model
19 calculation is performed. In light of the significant differences between atmospheric chemistry
20 taking place inside and outside of a power plant plume (e.g., Ryerson et al., 1998 and Sillman,
21 2000), inclusion of a separate, meteorological module for treating large, tight plumes is
22 necessary. Because the photochemistry of O3 and many other atmospheric species is nonlinear,
23 emissions correctly modeled in a tight plume may be incorrectly modeled in a more dilute
24 plume. Fortunately, it appears that the chemical mechanism used to follow a plume's
25 development need not be as detailed as that used to simulate the rest of the domain, as the
26 inorganic reactions are the most important in the plume (e.g., Kumar and Russell, 1996). The
27 need to include explicitly plume-in-grid chemistry disappears if one uses the adaptive grid
28 approach mentioned previously, though such grids are more computationally intensive. The
29 differences in simulations are significant because they can lead to significant differences in the
30 calculated sensitivity of O3 to its precursors (e.g., Sillman et al., 1995).
January 2005 AX2-114 DRAFT-DO NOT QUOTE OR CITE
-------
1 Because the chemical production and loss terms in the continuity equations for individual
2 species are coupled, the chemical calculations must be performed iteratively until calculated
3 concentrations converge to within some preset criterion. The number of iterations and the
4 convergence criteria chosen also can introduce error.
5 The importance of global transport of O3 and its contribution to regional O3 levels in the
6 United States is slowly becoming apparent. There are presently on the order of 20
7 three-dimensional global models that have been developed by various groups to address
8 problems in tropospheric chemistry. These models resolve synoptic meteorology, O3-NOX-CO-
9 hydrocarbon photochemistry, wet and dry deposition, and parameterize sub-grid scale vertical
10 mixing such as convection. Global models have proven useful for testing and advancing
11 scientific understanding beyond what is possible with observations alone. For example, they can
12 calculate quantities of interest that we do not have the resources to measure directly, such as
13 export of pollution from one continent to the global atmosphere or the response of the
14 atmosphere to future perturbations to anthropogenic emissions.
15 The finest horizontal resolution at which global simulations are typically conducted is
16 -200 km2 although rapid advances in computing power continuously change what calculations
17 are feasible. The next generation of models will consist of simulations that link multiple
18 horizontal resolutions from the global to the local scale. Finer resolution will only improve
19 scientific understanding to the extent that the governing processes are more accurately described
20 at that scale. Consequently there is a critical need for observations at the appropriate scales to
21 evaluate the scientific understanding represented by the models.
22 Observations of specific chemical species have been useful for testing transport schemes.
23 Radon-222 simulations in sixteen global models have been evaluated with observations to show
24 that vertical mixing is captured to within the constraints offered by the mean observed
25 concentrations (Jacob et al., 1997). Tracers such as cosmogenic 7Be and terrigenic 210Pb have
26 been used to test and constrain model transport and wet deposition (e.g., Liu et al., 2001).
27 Other chemical species obtained from various platforms (surface measurements, aircraft,
28 satellites) are useful for evaluating the simulation of chemical and dynamical processing in
29 global models. For example, Emmons et al. (2000) compiled available measurements of
30 12 species relevant to O3 photochemistry from a number of aircraft campaigns in different
31 regions of the world and used this data composite to evaluate two global models. They
January 2005 AX2-115 DRAFT-DO NOT QUOTE OR CITE
-------
1 concluded that one model (MOZART) suffered from weak convection and an underestimate of
2 nitrogen oxide emissions from biomass burning, while another model (IMAGES) transported too
3 much O3 from the stratosphere to the troposphere (Emmons et al., 2000). The global coverage
4 available from satellite observations offers new information for testing models. Recent efforts
5 are using satellite observations to evaluate the emission inventories of O3 precursors that are
6 included in global models; such observations should help to constrain the highly uncertain
7 natural emissions of isoprene and nitrogen oxides (e.g., Palmer et al., 2003; Martin et al., 2003).
8 A comparison of numerous global chemistry-transport models developed by groups around
9 the world was included in Section 4.4 of the recent report of the Intergovernmental Panel on
10 Climate Change (Prather and Ehhalt, 2001). In that report, monthly mean O3 (O3) and carbon
11 monoxide (CO) simulated by the various models was evaluated with O3 observations from global
12 ozonesonde stations at 700, 500, and 300 hPa and with surface CO measurements from
13 17 selected NOAA/CMDL sites. The relevant figures (Figures AX2-4-10 and AX2-4-11) are
14 reproduced here (as Figures AX2-19 for O3 and AX2-20 for CO) along with the references in
15 their Table AX2-10 (as Table AX2-4). Overall, the models capture the general features of the O3
16 and CO seasonal cycles but meet with varying levels of success at matching the observed
17 concentrations and the amplitude of the observed seasonal cycle. For O3, the models show less
18 disagreement in the lower troposphere than in the upper troposphere, reflecting the difficulty of
19 representing the exchange between the stratosphere and troposphere and the loose constraints on
20 the net O3 flux that are provided by observations.
21 An evaluation of five global models with data from the Measurement of Ozone and Water
22 Vapor by Airbus In-Service Aircraft (MOZAIC) project over New York City and Miami
23 indicates that the models tend to underestimate the summer maximum in the middle and lower
24 troposphere over northern mid-latitude cities such as New York City and to underestimate the
25 variability over coastal cities such as Miami which are strongly influenced by both polluted
26 continental and clean marine air masses (Law et al., 2000). Local maxima and minima are
27 difficult to reproduce with global models because processes are averaged over an entire model
28 grid cell. Much of the spatial and temporal variability in surface O3 over the United States is
29 modulated by synoptic meteorology (e.g., Logan, 1989; Eder et al., 1993; Vukovich, 1995, 1997;
30 Cooper and Moody, 2000) which is resolved in the current generation of global models.
31 For example, an empirical orthogonal function analysis on observed and simulated fields over
January 2005 AX2-116 DRAFT-DO NOT QUOTE OR CITE
-------
J FMAM J J A SON D
J FMAM J J A SON D
J FMAMJ JASOND
J FMAM J JASOND
J FMAM J JASOND
J FMAMJ JASOND
J FMAM J JASOND
J FMAM J JASOND
J FMAMJ JASOND
observ. HGIS MOZ1 UIO -*- UCAM
GISS IASB * MOZ2 UKMO MPIC
-*- HGEO —-- KNMI UCI1 ULAQ
Figure AX2-19. Seasonal variability in ozone 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).
1 the eastern United States in summer has shown that a 2° x 2.5° horizontal resolution global
2 model (GEOS-CHEM) captures the synoptic-scale processes that control much of the observed
3 variability (Fiore et al., 2003). Further evaluation of the same model showed that it can also
January 2005
AX2-117
DRAFT-DO NOT QUOTE OR CITE
-------
.a
a.
a.
CO
o
.a
a.
a.
100
80
60
4
20
0
100
80
60
4
20
J FMAMJ JA SOND JFMAMJJASOND JFMAMJJASOND
1001 200
J FMAMJ JA SOND JFMAMJJASOND JFMAMJJASOND
O
observ.
GISS
HGEO
IASB
KNMI
MOZ1
ucn
UIO
UKMO
- - ULAQ
-*- UCAM
MPIC
J FMAMJ JASOND
Figure AX2-20. Seasonal variability in ozone 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).
1
2
3
4
5
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
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).
January 2005
AX2-118
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX2-4. Chemistry-Transport Models (CTM) Contributing to the Oxcomp
Evaluation of Predicting Tropospheric O3 and OH (Prather and Ehhalt, 2001)
1
2
3
4
5
6
7
CTM
GISS
HGEO
HGIS
IASB
KNMI
MOZ1
MOZ2
MPIC
UCI
UIO
UI02
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)
A significant amount of progress in evaluating the performance
global models
Disagreement
with surface, aircraft, and
among model simulations
meteorology and emissions. The largest
satellite data has been made
References
Hansen et al. (1997)
Beyetal. (200 la)
Mickley et al. (1999)
Miiller and Brasseur
(1995, 1999)
Jeuken et al. (1999),
Houweling et al. (2000)
Brasseur et al. (1998),
Hauglustaine et al. (1998)
Brasseur et al. (1998),
Hauglustaine et al.(1998)
Crutzen et al. (1999),
Lawrence et al. (1999)
Hannegan et al. (1998),
Wild and Prather (2000)
Berntsen and Isaksen (1997),
Fuglesrvedt et al. (1999)
Sundet (1997)
Collins et al. (1997),
Johnson et al. (1999)
Pitari et al. (1997)
Law et al. (1998, 2000)
of three-dimensional
in recent years.
mainly stems from differences in the driving
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;
with observations closer to the surface.
the models agree better
Synoptic-scale meteorology is resolved in these models,
January 2005
AX2-119
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
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).
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
Non-road vehicle exhaust
Fossil fuel combustion
Industrial Processes
Biomass burning
Waste disposal
Natural sources2
Total
9.1
0.76
0.35
0.053
3.1
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).
January 2005
AX2-120
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX2-6. Emissions of Volatile Organic Compounds by Various Sources in the
United States in 1999
Source
Emissions (1012 g/y) Notes
On-road vehicles
Non-road vehicles
Fossil fuel combustion
Chemical industrial
processes
Petroleum industrial
processes
Other industrial
processes
Solvent volatilization
Storage and transport
of volatile
compounds
Biomass burning
Waste disposal
Biogenic sources1
Total
4.8
2.9
0.27
0.36
0.39
0.48
4.4
1.1
1.2
0.53
4.4
21
Exhaust and evaporative losses from gasoline (95%)
and diesel (5%) vehicles.
Exhaust and evaporative losses from gasoline (80%)
and diesel (12%) vehicles; aircraft and other sources
Electrical utilities; industrial, commercial,
institutional, and residential sources.
Mfg. of organic chemicals, polymers and resins, and
misc. products.
Oil and gas production (64%); refining (36%).
Metal processing (15%); wood processing (32%);
agricultural product processing (21%); misc.
processes (18%).
Surface coatings (44%); other industrial uses (20%);
non-industrial uses (e.g., pesticide application,
consumer solvents) (36%).
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).
January 2005
AX2-121
DRAFT-DO NOT QUOTE OR CITE
-------
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
Agriculture
Waste disposal and recycling
Natural sources
Total
0.25
0.044
0.18
3.9
0.08
0.032
4.5
Exhaust from on-road (96%) and non-road
(4%) vehicles.
Combustion by electric utilities, industry,
commerce, institutions, residences.
Chemical manufacturing (67%); petroleum
refining (9%); other industries (25%).
Livestock (82%); fertilizer application
(18%).
Wastewater treatment (99%).
Unmanaged soils; wild animals.
Source: U.S. Environmental Protection Agency (2001).
1 Emissions of nitrogen oxides associated with combustion arise from contributions from
2 both fuel nitrogen and atmospheric nitrogen. Sawyer et al. (2000) have reviewed the factors
3 associated with NOX emissions by mobile sources. Estimates of NOX emissions from mobile
4 sources are generally regarded as fairly reliable although further work is needed to clarify this
5 point (Sawyer et al., 2000). Both nitrifying and denitrifying bacteria in the soil can produce
6 NOX, mainly in the form of NO. Emission rates depend mainly on fertilization levels and soil
7 temperature. About 60% of the total NOX emitted by soils occurs in the central corn belt of the
8 United States. The oxidation of NH3 emitted mainly by livestock and soils, leads to the
9 formation of NO. Estimates of emissions from natural sources are less certain than those from
10 anthropogenic sources.
11 Natural sources of oxides of nitrogen include lightning, oceans, and soil. Of these, as
12 reviewed in AQCD 96, only soil emissions appear to have the potential to impact surface O3 over
13 the U.S. On a global scale, the contribution of soil emissions to the oxidized nitrogen budget is
14 on the order of 10% (van Aardenne et al., 2001; Finlayson-Pitts and Pitts, 2000; Seinfeld and
15 Pandis, 1998), but attempts to quantify emissions of NOX from fertilized fields show great
16 variability. Soil NO emissions can be estimated from the fraction of the applied fertilizer
January 2005
AX2-122
DRAFT-DO NOT QUOTE OR CITE
-------
Source
Table AX2-8. Emissions of Carbon Monoxide by Various Sources in the
United States in 1999
Emissions (1012 g/y) Notes
On-road vehicle exhaust
Non-road engines and
vehicle exhaust
Fossil fuel combustion
Industrial Processes
Biomass burning
Waste disposal
Other
Biogenic emssions1
Total
50
25
3.7
16
0.42
0.19
4.7+
102+
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%).
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).
1 nitrogen emitted as NOX, but the flux varies strongly with land use and temperature. The fraction
2 nitrogen. Estimated globally averaged fractional applied nitrogen loss as NO varies from 0.3%
3 (Skiba et al., 1997) to 2.5% (Yienger and Levy, 1995). Variability within biomes to which
4 fertilizer is applied, such as shortgrass versus tallgrass prairie, accounts for a factor of
5 three in uncertainty (Williams et al., 1992; Yienger and Levy, 1995; Davidson and Kingerlee,
6 1997).
7 The local contribution can be much greater than the global average, particularly in summer
8 especially where corn is grown extensively. Williams et al. (1992) estimated that contributions
9 from soils in Illinois contribute about 26% of the emissions from industrial and commercial
January 2005
AX2-123
DRAFT-DO NOT QUOTE OR CITE
-------
1 processes in that State. In Iowa, Kansas, Minnesota, Nebraska, and South Dakota soil emissions
2 may dominate. Conversion of ammonium to nitrate (nitrification) in aerobic soils appears to be
3 the dominant pathway to NO. The mass and chemical form of nitrogen (reduced or oxidized)
4 applied to soils, the vegetative cover, temperature, soil moisture, and agricultural practices such
5 as tillage all influence the amount of fertilizer nitrogen released as NO.
6 As pointed out in the previous AQCD for O3, emissions of NO from soils peak in summer
7 when O3 formation is at a maximum. A recent NRC report outlined the role of agricultural in
8 emissions of air pollutants including NO and NH3 (NRC, 2002). That report recommends
9 immediate implementation of best management practices to control these emissions, and further
10 research to quantify the magnitude of emissions and the impact of agriculture on air quality.
11 Civerolo and Dickerson (1998) report that use of the no-till cultivation technique on a fertilized
12 cornfield in Maryland reduced NO emissions by a factor of seven.
13 Annual global production of NO by lightning is the most uncertain source of reactive
14 nitrogen. In the last decade literature values of the production rate range from 2 to 20 Tg-N per
15 year. However, the most likely range is from 3 to 8 Tg-N per year, because the majority of the
16 recent estimates fall in this range. The large uncertainty stems from several factors: (1) a large
17 range of NO production rates per flash (as much as two orders of magnitude); (2) the open
18 question of whether cloud-to-ground (CG) flashes and intracloud flashes (1C) produce
19 substantially different amounts of NO; (3) the global flash rate; and (4) the ratio of the number of
20 1C flashes to the number of CG flashes. Estimates of the amount of NO produced per flash have
21 been made based on theoretical considerations (e.g., Price et al., 1997), laboratory experiments
22 (e.g., Wang et al., 1998); field experiments (e.g., Stith et al., 1999; Huntrieser et al., 2002), and
23 through a combination of cloud-resolving model simulations, observed lightning flash rates, and
24 anvil measurements of NO (e.g., DeCaria et al., 2000). The latter method was also used by
25 Pickering et al. (1998), who showed that only -5% to 20% of the total NO production by
26 lightning in a given storms exists in the boundary layer at the end of a thunderstorm. Therefore,
27 the direct contribution to boundary layer O3 production by lightning NO is thought to be small.
28 However, lightning NO production can contribute substantially to O3 production in the middle
29 and upper troposphere. DeCaria et al. (2000) estimated that up to 7 ppbv of O3 were produced in
30 the upper troposphere in the first 24 hours following a Colorado thunderstorm due to the
31 injection of lightning NO. A major uncertainty in mesoscale and global chemical transport
January 2005 AX2-124 DRAFT-DO NOT QUOTE OR CITE
-------
1 models is the parameterization of lightning flash rates. Model variables such as cloud top height,
2 convective precipitation rate, and upward cloud mass flux have been used to estimate flash rates.
3 Allen and Pickering (2002) have evaluated these methods against observed flash rates from
4 satellite, and examined the effects on O3 production using each method.
5 Literally tens of thousands of organic compounds have been identified in plant tissues.
6 However, most of these compounds either have sufficiently low volatility or are constrained so
7 that they are not emitted in significant quantities. Less than 40 compounds have been identified
8 by Guenther et al. (2000) as being emitted in large enough quantities to affect atmospheric
9 composition. These compounds include terpenoid compounds (isoprene, 2-methyl-3-buten-2-ol,
10 monoterpenes), compounds in the hexanal family, alkenes, aldehydes, organic acids, alcohols,
11 ketones and alkanes. As can be seen from Table AX2-6, the major species emitted by plants are
12 isoprene (35%), 19 other terpenoid compounds (25%) and 17 non-terpenoid compounds (40%)
13 (Guenther et al., 2000). Of the latter, methanol contributes 12% of total emissions.
14 Because isoprene has been identified as the most abundant of biogenic VOCs (Guenther
15 et al., 1995, 2000; Geron et al., 1994), it has been the focus of air quality model analyses in
16 many published studies (Roselle, 1994; Sillman et al., 1995). The original Biogenic Emission
17 Inventory System (BEIS) of Pierce et al. (1991) used a branch-level isoprene emission factor of
18 14.7 jig (g-foliar dry mass)"1 h"1 for high isoprene emitting species (e.g., oaks, or North
19 American Quercus species). When considering self-shading of foliage within branch enclosures,
20 this is roughly equivalent to a leaf level emission rate of 20 to 30 |ig-C (g-foliar dry mass)"1 h"1
21 (Guenther at al, 1995). Geron et al (1994) reviewed studies between 1990 and 1994 and found
22 that a much higher leaf-level rate of 70 |ig-C (g-foliar dry mass)"1 h"1 + 50% was more realistic,
23 and this rate was used in BEIS2 for high isoprene emitting tree species. BEIS3 (Guenther et al.,
24 2000) applied similar emission factors at tree species levels (Geron et al 2000a, 2001) and more
25 recent canopy environment models to estimate isoprene fluxes.
26 The results from several studies of isoprene emission measurements made at leaf, branch,
27 tree, forest stand, and landscape levels have been used to test the accuracy of BEIS2 and BEIS3.
28 These comparisons are documented in Geron et al. (1997) and Guenther et al. (2000). The
29 results of these studies support the higher emission factors used in BEIS2 and BEIS3. Typically,
30 leaf emission factors (normalized to standard conditions of PAR = 1000 jimol nT2 and leaf
31 temperature of 30°C) measured at the top of tree canopies equal or exceed those used in BEIS2/3
January 2005 AX2-125 DRAFT-DO NOT QUOTE OR CITE
-------
1 while those in more shaded portions of the canopy tend to be lower than those assumed in the
2 models, likely due to differences in developmental environments of leaves within the canopy
3 (Monson et al., 1994; Sharkey et al., 1996; Harley et al., 1996; Geron et al., 2000b). Uncertainty
4 in isoprene emissions due to variability in forest composition and leaf area remain in B VOC
5 emission models and inventories. Seasonality and moisture stress also impact isoprene emission,
6 but algorithms to simulate these effects are currently fairly crude (Guenther et al, 2000). The
7 bulk of biogenic emissions occur during the summer, because of their dependence on
8 temperature and incident sunlight. Biogenic emissions are also higher in southern states than in
9 northern states for these reasons. The uncertainty associated with natural emissions ranges from
10 about 50% for isoprene under midday summer conditions to about a factor often for other
11 compounds (Guenther et al., 2000). In assessing the relative importance of these compounds, it
12 should be borne in mind that the oxidation of many of the classes of compounds result in the
13 formation of secondary organic aerosol and that many of the intermediate products may be
14 sufficiently long lived to affect O3 formation in areas far removed from where they were emitted.
15 The oxidation of isoprene can also contribute about 10% of the source of CO (U.S.
16 Environmental Protection Agency, 2000). Direct emissions of CO by vegetation is of much
17 smaller importance. Soil microbes both emit and take up atmospheric CO, however, soil
18 microbial activity appears to represent a net sink for CO.
19 Emissions from biomass burning depend strongly on the stage of combustion. Smoldering
20 combustion, especially involving forest ecosystems favors the production of CH4, NMHC and
21 CO at the expense of CO2, whereas active combustion produces more CO2 relative to the other
22 compounds mentioned above. Typical emissions ratios (defined as moles of compound per
23 moles of emitted CO2 expressed as a percentage) range from 6 to 14% for CO, 0.6 to 1.6% for
24 CH4, and 0.3 to 1.1% for NMHCs (Andreae, 1991). Most NMHC emissions are due to
25 emissions of lighter compounds, containing 2 or 3 carbon atoms.
26
27 AX2.5.3 Observationally Based Models
28 As an alternative to chemistry-transport models, observationally-based methods (OEMs),
29 which seek to infer O3-precursor relations by relying more heavily on ambient measurements,
30 can be used. Observationally-based methods are intuitively attractive because they provide an
31 estimate of the O3-precursor relationship based directly on observations of the precursors. These
January 2005 AX2-126 DRAFT-DO NOT QUOTE OR CITE
-------
1 methods rely on observations as much as possible to avoid many of the uncertainties associated
2 with chemistry/transport models (e.g., emission inventories and meteorological processes).
3 However, these methods have large uncertainties with regards to photochemistry. As originally
4 conceived, the observation-based approaches were intended to provide an alternative method for
5 evaluating critical issues associated with urban O3 formation. The proposed OEMs include
6 calculations driven by ambient measurements (Chameides et al., 1992; Cardelino et al., 1995)
7 and proposed "rules of thumb" that seek to show whether O3 is primarily sensitive to NOX or to
8 VOC concentrations (Sillman, 1995; Chang et al., 1997; Tonnesen and Dennis, 2000a,b;
9 Blanchard et al., 1999; Blanchard, 2000). These methods are controversial when used as
10 "stand-alone" rules, because significant uncertainties and possible errors have been identified for
11 all the methods (Chameides et al., 1988, Lu and Chang, 1998, Sillman and He, 2002; Blanchard
12 and Stockenius, 2001). Methods such as these are most promising for use in combination with
13 chemistry /transport models principally for evaluating the accuracy of model predictions.
14 Recent results (Tonnesen and Dennis, 2000a; Kleinman et al., 1997; 2000, 2001;
15 Kleinman, 2000) suggest that ambient VOC and NOX data can be used to identify the
16 instantaneous production rate for O3 and how the production rate varies with concentrations of
17 NOX and VOCs. The instantaneous production rate for O3 is only one of the factors that affect
18 the total O3 concentration, because O3 concentrations result from photochemistry and transport
19 over time periods ranging from several hours to several days in regional pollution events. Ozone
20 concentrations can be affected by distant emissions and by photochemical conditions at upwind
21 locations, rather than instantaneous production at the site. Despite this limitation, significant
22 information can be obtained by interpreting ambient NOX and VOC measurements. Kleinman
23 et al. (1997, 2000, 2001) and Tonnesen and Dennis (2000a) both derived simple expressions that
24 relate the NOX-VOC sensitivity of instantaneous O3 production to ambient VOC and NOX. These
25 expressions usually involve summed VOC weighted by reactivity.
26 Cardelino et al. (1995, 2000) developed a method that seeks to identify O3-NOX-VOC
27 sensitivity based on ambient NOX and VOC data. Their method involves an area-wide sum of
28 instantaneous production rates over an ensemble of measurement sites, which serve to represent
29 the photochemical conditions associated with O3 production in metropolitan areas. Their
30 method, which relies on routine monitoring methods, is especially useful because it permits
31 evaluation for a full season rather than just for individual episodes.
January 2005 AX2-127 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX2.5.4 Chemistry-Transport Model Evaluation
2 The comparison of model predictions with ambient measurements represents a critical task
3 for establishing the accuracy of photochemical models and evaluating their ability to serve as the
4 basis for making effective control strategy decisions. The evaluation of a model's performance,
5 or its adequacy to perform the tasks for which it was designed can only be conducted within the
6 context of measurement errors and artifacts. Not only are there analytical problems, but there
7 are also problems in assessing the representativeness of monitors at ground level for comparison
8 with model values which represent typically an average over the volume of a grid box.
9 Chemistry /transport models for O3 formation at the urban/regional scale have traditionally
10 been evaluated based on their ability to correctly simulate O3. A series of performance statistics
1 1 that measure the success of individual model simulations to represent the observed distribution
12 of ambient O3, as represented by a network of surface measurements were recommended in U.S.
13 Environmental Protection Agency (1991; see also Russell and Dennis, 2000). These statistics
14 consist of the following:
15 • Unpaired peak O3 within a metropolitan region (typically for a single day)
16 • Normalized bias equal to the summed difference between model and measured hourly
concentrations divided by the sum of measured hourly concentrations.
17 • 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.
1 8 Normalized bias, D;
19
20 C^Omax-Q^VOmax
21 A"= - 77777: - *100%, (AX2-49)
i^ov* >' Jrnax
22
23 Gross error, Ed (for hourly observed values of O3 > 60 ppb)
24
25 ! * {C(Xi,f) - C0(Xi,t)}
C =1,24. (AX2-50)
27
January 2005 AX2-128 DRAFT-DO NOT QUOTE OR CITE
-------
1 Unpaired peak prediction accuracy, Au
2
Cp(xi,f)-C0(xi,f)
S———. ' '=1,24. (AX2-51)
3
4 The following performance criteria for regulatory models were recommended in U.S.
5 Environmental Protection Agency (1991): unpaired peak O3 to within ± 15% or ± 20%;
6 normalized bias within ± 5% to ± 15%; and normalized gross error less than 30% to 35%, but
7 only when O3 > 60 ppb. This can lead to difficulties in evaluating model performance since
8 nighttime and diurnal cycles are ignored. A major problem with this method of model
9 evaluation is that it does not provide any information about the accuracy of O3-precursor
10 relations predicted by the model. The process of O3 formation is sufficiently complex that
11 models can predict O3 correctly without necessarily representing the O3 formation process
12 properly. If the O3 formation process is incorrect, then the modeled source-receptor relations
13 will also be incorrect.
14 Studies by Sillman et al. (1995, 2003), Reynolds et al. (1996) and Pierce et al. (1998) have
15 identified instances in which different model scenarios can be created with very different
16 O3-precursor sensitivity, but without significant differences in the predicted O3 fields.
17 Figures AX2-21a,b provide an example. Referring to the O3-NOX-VOC isopleth plot
18 (Figure AX2-22), it can be seen that similar O3 concentrations can be found for photochemical
19 conditions that have very different sensitivity to NOX and VOCs.
20 Global-scale chemistry-transport models have generally been evaluated by comparison
21 with measurements for a wide array of species, rather than just for O3 (e.g., Wang et al., 1998;
22 Emmons et al., 2000; Bey et al., 2001b; Hess, 2001; Fiore et al., 2002). These have included
23 evaluation of major primary species (NOX, CO, and selected VOCs) and an array of secondary
24 species (FDSTO3, PAN, H2O2) that are often formed concurrently with O3. Models for
25 urban/regional O3 have also been evaluated against a broader ensemble of measurements in a
26 few cases, often associated with measurement intensives (e.g., Jacobson et al., 1996, Lu et al.,
27 1997; Sillman et al., 1998). The results of a comparison between observed and computed
28 concentrations from Jacobson et al. (1996) for the Los Angeles Basin are shown in
29 Figures AX2-23a,b.
January 2005 AX2-129 DRAFT-DO NOT QUOTE OR CITE
-------
^.o
*>" 9n
£ 2D
Q.
tt 1S
C
.2 10
o
3
01
co n
0 °
c.
±J
^€
*
X
* •
• X
•
XXX*
**9\
1
•
, x •,
V
1
•
, .
<
100 120 140 160 180 200
Ozone (ppb)
otr
.Q
Q. OR _
3
C on
.2
o m
3 15
O -in
O 5 "
n
u
4
X X
.
' *•
**
X
XX
X
X
1
\ ,
. "
•.
X* '
X
X
X
<
X
100 120 140 160 180 200
Ozone (ppb)
Figure AX2-21a,b. Impact of model uncertainty on control strategy predictions for ozone
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).
1 The highest concentrations of primary species usually occur in close proximity to emission
2 sources (typically in urban centers) and at times when dispersion rates are low. The diurnal
3 cycle includes high concentrations at night, with maxima during the morning rush hour, and low
4 concentrations during the afternoon (Figure AX2-23a). The afternoon minima are driven by the
January 2005
AX2-130
DRAFT-DO NOT QUOTE OR CITE
-------
£- 10.00
v>
Cjl
E
u
3.16
_
o
E
c. 1.0
3
re
01
c
•jj; 0.316
(A
E
LU
x
g 0.1
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
NO-limited conditions.
1 much greater rate of vertical mixing at that time. Primary species also show a seasonal
2 maximum during winter, and are often high during fog episodes in winter when vertical mixing ,
3 is suppressed. By contrast, secondary species such as O3 are typically highest during the
4 afternoon (the time of greatest photochemical activity), on sunny days and during summer.
5 During these conditions concentrations of primary species may be relatively low. Strong
6 correlations between primary and secondary species are generally observed only in downwind
7 rural areas where all anthropogenic species are high simultaneously. The difference in the
8 diurnal cycles of primary species (CO, NOX and ethane)and secondary species (O3, PAN and
9 HCHO) is evident in Figure AX2-23b.
January 2005
AX2-131
DRAFT-DO NOT QUOTE OR CITE
-------
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).
January 2005
AX2-132
DRAFT-DO NOT QUOTE OR CITE
-------
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).
January 2005
AX2-133
DRAFT-DO NOT QUOTE OR CITE
-------
1 Models for urban/regional O3 have been evaluated less extensively than global-scale
2 models in part because the urban/regional context presents a number of difficult challenges.
3 Global-scale models typically represent continental-scale events and can be evaluated effectively
4 against a sparse network of measurements. By contrast, urban/regional models are critically
5 dependent on the accuracy of local emission inventories and event-specific meteorology, and
6 must be evaluated separately for each urban area that is represented.
7 The evaluation of urban/regional models is also limited by the availability of data.
8 Measured NOX and speciated VOC concentrations are widely available through the EPA PAMs
9 network, but questions have been raised about the accuracy of those measurements and the data
10 have not yet been analyzed thoroughly. Evaluation of urban/regional models versus
11 measurements has generally relied on results from a limited number of field studies in the United
12 States. Short term research-grade measurements for species relevant to O3 formation, including
13 VOCs, NOX, PAN, nitric acid (HNO3) and hydrogen peroxide (H2O2) are also widely available at
14 rural and remote sites (e.g., Daum et al., 1990, 1996; Martin et al., 1997; Young et al., 1997;
15 Thompson et al., 2000; Hoell et al., 1996, 1997; Fehsenfeld et al., 1996a; Emmons et al., 2000;
16 Hess, 2001; Carroll et al., 2001). The equivalent measurements are available for some polluted
17 rural sites in the eastern United States (e.g.) but only at a few urban locations (Meagher et al.,
18 1998; Hubler et al., 1998; Kleinman et al., 2000, 2001; Fast et al., 2002; new SCAQS-need
19 reference). Extensive measurements have also been made in Vancouver (Steyn et al., 1997) and
20 in several European cities (Staffelbach et al., 1997; Prevot et al., 1997, Dommen et al., 1999;
21 Geyer et al., 2001; Thielman et al., 2001; Martilli et al., 2002; Vautard et al., 2002).
22 The results of straightforward comparisons between observed and predicted concentrations
23 of O3 can be misleading because of compensating errors, although this possibility is diminished
24 when a number of species are compared. Ideally, each of the main modules of a chemistry -
25 transport model system (for example, the meteorological model and the chemistry and radiative
26 transfer routines) should be evaluated separately. However, this is rarely done in practice.
27 To better indicate how well physical and chemical processes are being represented in the model,
28 comparisons of relations between concentrations measured in the field and concentrations
29 predicted by the model can be made. These comparisons could involve ratios and correlations
30 between species. For example, correlation coefficients could be calculated between primary
31 species as a means of evaluating the accuracy of emission inventories; or between secondary
January 2005 AX2-134 DRAFT-DO NOT QUOTE OR CITE
-------
1 species as a means of evaluating the treatment of photochemistry in the model. In addition,
2 spatial relations involving individual species (correlations, gradients) can also be used as a
3 means of evaluating the accuracy of transport parameterizations. Sillman and He (2002)
4 examined differences in correlation patterns between O3 and NOZ in Los Angeles, CA, Nashville,
5 TN and various sites in the rural United States. Model calculations (Figure AX2-24) show
6 differences in correlation patterns associated with differences in the sensitivity of O3 to NOX
7 and VOCs. Primarily NOx-sensitive ( NOx-limited) areas in models show a strong correlation
8 between O3 and NOZ with a relatively steep slope, while primarily VOC-sensitive (NOX-
9 saturated) areas in models show lower O3 for a given NOZ and a lower O3-NOZ slope. They
10 found that differences found in measured data ensembles were matched by predictions from
11 chemical transport models. Measurements in rural areas in the eastern U.S. show differences in
12 the pattern of correlations for O3 versus NOZ between summer and autumn (Jacob et al., 1995;
13 Hirsch et al., 1996), corresponding to the transition from NOx-limited to NOx-saturated patterns,
14 a feature which is also matched by chemistry-transport models.
15 The difference in correlations between secondary species in NOx-limited to NOx-saturated
16 environments can also be used to evaluate the accuracy of model predictions in individual
17 applications. Figures AX2-25a and AX2-25b show results for two different model scenarios for
18 Atlanta. As shown in the figures, the first model scenario predicts an urban plume with high
19 NOy and O3 formation apparently suppressed by high NOy. Measurements show much lower
20 NOy in the Atlanta plume. This error was especially significant because the model locations
21 with high NOy were not sensitive to NOX, while locations with lower NOy were primarily
22 sensitive to NOX. The second model scenario (with primarily NOx-sensitive conditions) shows
23 much better agreement with measured values. Figure AX2-26a,b shows model-measurement
24 comparisons for secondary species in Nashville, showing better agreement with measured
25 conditions. Greater confidence in the predictions made by chemistry-transport models will be
26 gained by the application of techniques such as these on a more routine basis.
27 The ability of chemical mechanisms to calculate the concentrations of free radicals under
28 atmospheric conditions was tested in the Berlin Ozone Experiment, BERLIOZ (Volz-Thomas
29 et al., 2003) during July and early August at a site located about 50 km NW of Berlin. (This
30 location was chosen as O3 episodes in central Europe are often associated with SE winds.)
31 Concentrations of major compounds such as O3, hydrocarbons, etc., were fixed at observed
January 2005 AX2-13 5 DRAFT-DO NOT QUOTE OR CITE
-------
250
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).
1 values. In this regard, the protocol used in this evaluation is an example of an observationally
2 based method. Figure AX2-27 compares the concentrations of RO2 (organic peroxy), HO2
3 (hydroperoxy) and OH (hydroxyl) radicals predicted by RACM (regional air chemistry
4 mechanism; Stockwell et al., 1997) and MCM (master chemical mechanism; Jenkin et al, 1997
5 with updates) with observations made by the laser induced fluorescence (LIF) technique and by
6 matrix isolation ESR spectroscopy (MIESR). Also shown are the production rates of O3
7 calculated using radical concentrations predicted by the mechanisms and those obtained by
8 measurements, and measurements of NOX concentrations. As can be seen, there is good
9 agreement between measurements of organic peroxy, hydroperoxy and hydroxyl radicals with
10 values predicted by both mechanisms at high concentrations of NOX (> 10 ppb). However, at
January 2005
AX2-136
DRAFT-DO NOT QUOTE OR CITE
-------
.0
CL
Q.
10 20 30
NOy (Ppb)
40
10 20 30
NOy (Ppb)
40
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).
1 lower NOX concentrations, both mechanisms substantially overestimate OH concentrations and
2 moderately overestimate HO2 concentrations. Agreement between models and measurements is
3 generally better for organic peroxy radicals, although the MCM appears to overestimate their
January 2005
AX2-137
DRAFT-DO NOT QUOTE OR CITE
-------
160
140
160
10 20 30
NOZ (ppb)
40
10 20 30
2H202 + 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).
1 concentrations somewhat. In general, the mechanisms reproduced the HO2 to OH and RO2 to
2 OH ratios better than the individual measurements. The production of O3 was found to increase
3 linearly with NO (for NO < 0.3 ppb) and to decrease with NO (for NO > 0.5 ppb).
January 2005
AX2-138
DRAFT-DO NOT QUOTE OR CITE
-------
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
ozone 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).
January 2005
AX2-139
DRAFT-DO NOT QUOTE OR CITE
-------
1 OH and HO2 concentrations measured during the PM2 5 Technology Assessment and
2 Characterization Study conducted at Queens College in New York City in the summer of 2001
3 were also compared with those predicted by RACM (Ren et al., 2003). The ratio of observed to
4 predicted HO2 concentrations over a diurnal cycle was 1.24 and the ratio of observed to
5 predicted OH concentrations was about 1.10 during the day, but the mechanism significantly
6 underestimated OH concentrations during the night.
7
8 AXA.5.4.1 Evaluation of Emissions Inventories
9 Comparisons of emissions model predictions with observations have been performed in a
10 number of environments. A number of studies of ratios of concentrations of CO to NOX and
11 NMOC to NOX during the early 1990s in tunnels and ambient air (summarized in Air Quality
12 Criteria for Carbon Monoxide [U.S. Environmental Protection Agency, 2000]) indicated that
13 emissions of CO and NMOC were systematically underestimated in emissions inventories.
14 However, the results of more recent studies have been mixed in this regard, with many studies
15 showing agreement to within ± 50% (U.S. Environmental Protection Agency, 2000).
16 Improvements in many areas have resulted from the process of emissions model development,
17 evaluation, and further refinement. It should be remembered that the conclusions from these
18 reconciliation studies depend on the assumption that NOX emissions are predicted correctly by
19 emissions factor models. Road side remote sensing data indicate that over 50% of NMHC and
20 CO emissions are produced by less than about 10% of the vehicles (Stedman et al., 1991). These
21 "super-emitters" are typically poorly maintained vehicles. Vehicles of any age engaged in off-
22 cycle operations (e.g., rapid accelerations) emit much more than if operated in normal driving
23 modes. Bishop and Stedman (1996) found that the most important variables governing CO
24 emissions are fleet age and owner maintenance.
25 Emissions inventories for North America can be evaluated with comparisons to measured
26 long-term trends and or ratios of pollutants in ambient air. A decadal field study of ambient CO
27 at a rural cite in the Eastern U.S. (Hallock-Waters et al., 1999) indicates a downward trend
28 consistent with the downward trend in estimated emissions over the period 1988 to 1999 (U.S.
29 Environmental Protection Agency, 1997), even when a global downward trend is accounted for.
30 Measurements at two urban areas in the United States confirmed the decrease in CO emissions
31 (Parrish et al., 2002). That study also indicated that the ratio of CO to NOX emissions decreased
January 2005 AX2-140 DRAFT-DO NOT QUOTE OR CITE
-------
1 by almost a factor of three over 12 yr (such a downward trend was noted in AQCD 96).
2 Emissions estimates (U.S. Environmental Protection Agency, 1997) indicate a much smaller
3 decrease in this ratio, suggesting that NOX emissions from mobile sources may be
4 underestimated and/or increasing. The authors conclude that O3 photochemistry in U.S. urban
5 areas may have become more NOx-limited over the past decade.
6 Pokharel et al. (2002) employed remotely-sensed emissions from on-road vehicles and fuel
7 use data to estimate emissions in Denver. Their calculations indicate a continual decrease in CO,
8 HC, and NO emissions from mobile sources over the 6 yr study period. Inventories based on the
9 ambient data were 30 to 70% lower for CO, 40% higher for HC, and 40 to 80% lower for NO
10 than those predicted by the recent MOBILE6 model.
11 Stehr et al. (2000) reported simultaneous measurements of CO, SO2 and NOy at an East
12 Coast site. By taking advantage of the nature of mobile sources (they emit NOX and CO but little
13 SO2) and power plants (they emit NOX and SO2 but little CO), the authors evaluated emissions
14 estimates for the eastern United States. Results indicated that coal combustion contributes 25 to
15 35% of the total NOX emissions in agreement with emissions inventories (U.S. Environmental
16 Protection Agency, 1997).
17 Parrish et al. (1998) and Parrish and Fehsenfeld (2000) proposed methods to derive
18 emission rates by examining measured ambient ratios among individual VOC, NOX and NOy.
19 There is typically a strong correlation among measured values for these species (e.g., Figure
20 AX2-14) because emission sources are geographically collocated, even when individual sources
21 are different. Correlations can be used to derive emissions ratios between species, including
22 adjustments for the impact of photochemical aging. Investigations of this type include
23 correlations between CO and NOy (e.g., Parrish et al., 1991), between individual VOC species
24 and NOy (Goldan et al., 1995,1997, 2001; Harley et al., 1997) and between various individual
25 VOC (Goldan et al., 1995, 1997; McKeen and Liu, 1993; McKeen et al., 1996). Buhr et al.
26 (1992) derived emission estimates from principal component analysis (PCA) and other statistical
27 methods. Many of these studies are summarized in Trainer et al. (2000), Parrish et al. (1998),
28 and Parrish and Fehsenfeld (2000). Goldstein and Schade (2000) also used species correlations
29 to identify the relative impacts of anthropogenic and biogenic emissions. Chang et al. (1996,
30 1997) and Mendoza-Dominguez and Russell (2000, 2001) used the more formal techniques of
31 inverse modeling to derive emission rates, in conjunction with results from chemistry-transport
January 2005 AX2-141 DRAFT-DO NOT QUOTE OR CITE
-------
1 models. Another concern regarding the use of emissions inventories is that emissions from all
2 significant sources have been included. This may not always be the case. As an example,
3 hydrocarbon seeps from off-shore oil fields may represent a significant source of reactive
4 organic compounds in near by coastal areas (Quigley et al., 1999).
5
6 AX2.5.4.2 Availability and Accuracy of Ambient Measurements
7 The use of methods such as observationally based methods or source apportionment
8 models, either as stand-alone methods or as a basis for evaluating chemistry /transport models,
9 is often limited by the availability and accuracy of measurements. Measured speciated VOC and
10 NOX are widely available in the United States through the PAMS network. However, challenges
11 have been raised about both the accuracy of the measurements and their applicability.
12 Parrish et al. (1998) and Parrish and Fehsenfeld (2000) developed a series of quality
13 assurance tests for speciated VOC measurements. Essentially these tests used ratios among
14 individual VOC with common emission sources to identify whether the variations in species
15 ratios were consistent with the relative photochemical lifetimes of individual species. These
16 tests were based on a number of assumptions: the ratio between ambient concentrations of
17 long-lived species should show relatively little variation among measurements affected by a
18 common emissions sources; and the ratio between ambient concentrations of long-lived and
19 short-lived species should vary in a way that reflects photochemical aging at sites more different
20 from source regions. Parrish et al. used these expectations to establish criteria for rejecting
21 apparent errors in measurements. They found that the ratios among alkenes at many PAMS sites
22 did not show variations that would be expected due to photochemical aging.
23 The PAMs network currently includes measured NO and NOX. However, Cardelino and
24 Chameides (2000) reported that measured NO during the afternoon was frequently at or below
25 the detection limit of the instruments (1 ppb), even in large metropolitan regions (Washington,
26 DC; Houston, TX; New York, NY). NOX measurements are made with commercial
27 chemilluminescent detectors with molybdenum converters. However these measurements
28 typically include some organic nitrates in addition to NOX, and cannot be interpreted as a "pure"
29 NOX measurement (see summary in Parrish and Fehsenfeld, 2000).
30 Total reactive nitrogen (NOy) is included in the PAMS network only at a few sites. The
31 possible expansion of PAMS to include more widespread NOy measurements has been suggested
January 2005 AX2-142 DRAFT-DO NOT QUOTE OR CITE
-------
1 (McClenny, 2000). A major issue concerning measured NOy is the possibility that HNO3,
2 a major component of NOy, is sometimes lost in inlet tubes and not measured (Luke et al., 1998;
3 Parrish and Fehsenfeld, 2000). This problem is especially critical if measured NOy is used to
4 identify NOx-limited versus NOx-saturated conditions. The correlation between O3 and NOy
5 differs for NOx-limited versus NOx-saturated locations, but this difference is driven primarily by
6 differences in the ratio of O3 to HNO3. If HNO3 were omitted from the NOy measurements, than
7 the measurements would represent a biased estimate and their use would be problematic.
10 AX2.6 TECHNIQUES FOR MEASURING OZONE AND ITS
11 PRECURSORS
12 AX2.6.1 Sampling and Analysis of Ozone
13 Numerous techniques have been developed for sampling and measurement of O3 in the
14 ambient atmosphere at ground level. As noted above, sparse surface networks tend to
15 underestimate maximum O3 concentrations. Today, monitoring is conducted almost exclusively
16 with UV absorption spectrometry with commercial short path instruments, a method that has
17 been thoroughly evaluated in clean air. The ultimate reference method is a relatively long-path
18 UV absorption instrument maintained under carefully controlled conditions at NIST (e.g., Fried
19 and Hodgeson, 1982). Episodic measurements are made with a variety of other techniques based
20 on the principles of chemiluminescence, electrochemistry, DOAS, and LIDAR. The rationale,
21 history, and calibration of O3 measurements were summarized in AQCD 96, so this section will
22 focus on the current state of ambient O3 measurement, tests for artifacts, and on new
23 developments.
24 Several reports in the reviewed scientific literature have investigated interferences in O3
25 detection via UV radiation absorption. Kleindienst et al. (1993) investigated the effects of water
26 vapor and VOC's on instruments based on both UV absorption and chemiluminescence. They
27 concluded that water vapor had no significant impact on UV absorption-based instruments, but
28 could cause a positive interference of up to 9% in chemiluminescence-based detectors at high
29 humidities (dew point of 24 C). In smog chamber studies, aromatic compounds and their
30 oxidation products were found to generate a positive but small interference in the UV absorption
31 instruments. Kleindienst et al. concluded that "when the results are scaled back to ambient
January 2005 AX2-143 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations of toluene and NOX, the effect appears to be very minor (ca. 3 percent under the
2 study conditions)." More recently Narita et al. (1998) tested organic and inorganic compounds
3 and found response to several, but not at levels likely to interfere with accurate determination of
4 O3 in an urban environment. The possibility for substantive interferences in O3 detection exists,
5 but such interferences have not been observed even in urban plumes. Ryerson et al. (1998)
6 measured O3 with UV absorption and chemiluminescence instruments operated off a common
7 inlet on the NOAA WP-3 research aircraft. As reported by Parrish and Fehsenfeld (2000)
8 "Through five field missions over four years, excellent correlations were found between the
9 measurements of the two instruments, although the chemiluminescence instrument was
10 systematically low (5%) throughout some flights. The data sets include many passes through the
11 Nashville urban plume. There was never any indication (< 1%) that the UV instrument
12 measured systematically higher in the urban plume."
13 Ozone can also be detected by differential optical absorption spectroscopy (DOAS) at a
14 variety of wavelengths in the UV and visible parts of the spectrum. Prior comparisons of DOAS
15 results to those from a UV absorption instrument showed good agreement, on the order of 10%
16 (Stevens et al., 1993). Reisinger (2002) reported a positive interference due to an unidentified
17 absorber in the 279 to 289 nm spectral region used by many commercial short-path DOAS
18 systems for the measurement of O3. Results of that study suggest that effluent from wood
19 burning, used for domestic heating, may be responsible. Vandaele et al. (2002) reported good
20 agreement with other methods in the detection of O3 (and SO2) over the course of several years
21 in Brussels. While the DOAS method remains attractive due to its sensitivity and speed of
22 response further intercomparisons and interference tests are recommended.
23 Electrochemical methods are commonly employed where sensor weight is a problem, such
24 as in balloon borne sondes, and these techniques have been investigated for ambient monitoring.
25 Recent developments include changes in the electrodes and electrolyte solution (Knake and
26 Hauser, 2002) and selective removal of O3 for a chemical zero (Penrose et al., 1995).
27 Interferences from other oxidants such as NO2 and HONO remain potential problems and further
28 comparisons with UV absorption are necessary. Because of potential interferences from water
29 vapor in some instruments, it is recommended (ASTM, 2003a,b) that calibrations and scrubber
30 tests be conducted with air humidified to near ambient levels, rather than with dry compressed
31 gas.
January 2005 AX2-144 DRAFT-DO NOT QUOTE OR CITE
-------
1 Change in the vibration frequency of a piezoelectric quartz crystal has been investigated as
2 a means of detecting O3. Ozone reacts with polybutadiene coated onto the surface of a crystal,
3 and the resulting change in mass is detected as a frequency change (Black et al., 2000). While
4 this sensor has advantages of reduced cost power consumption and weight, it is lacks the lifetime
5 and absolute accuracy for ambient monitoring.
6 In summary, new techniques are being developed, but UV absorption remains the method
7 of choice for ambient O3 monitoring near the Earth's surface. These commercial UV absorption
8 detectors are available at a moderate price. They show good absolute accuracy with only minor
9 cross sensitivity in clean to moderately polluted environments; they are stable, reliable, and
10 sensitive.
11
12 AX2.6.2 Sampling and Analysis of Nitrogen Oxides
13 The role of nitrogen oxides in tropospheric O3 formation was reviewed thoroughly in the
14 previous AQCD and will be only briefly summarized here. Reactive nitrogen is generally
15 released as NO but quickly converted in ambient air to NO2 and back again, thus these two
16 species are often referred to together as NOX (NO + NO2). The photochemical interconversion of
17 NO and NO2 leads to O3 formation. Because NO2 is a health hazard at sufficiently high
18 concentrations, it is itself a criteria pollutant. In EPA documents, emissions of NOX are
19 expressed in units of mass of NO2 per unit time, i.e., the total mass of NOX that would be emitted
20 if all the NO were converted to NO2. Ambient air monitors have been required to demonstrate
21 compliance with the standard for NO2 and thus have focused on measuring this gas or
22 determining an upper limit for its concentration.
23 NOX can be further oxidized to species including nitrous acid (HNO2), nitric acid (HNO3),
24 aerosol nitrate (NO3 ), and organo-nitrates such as alkyl nitrates (RONO2) and peroxy acetyl
25 nitrate, PAN, (CH3C(O)O2NO2). The sum of these species (explicitly excluding N2, N2O, and
26 reduced N such as NH3 and HCN) is called NOy. Nitrates play important roles in acid rain, and
27 nutrient cycling including over nitrification of surface ecosystems and in the formation of fine
28 particulate matter, but are generally inactive photochemically. Some studies refer specifically to
29 the oxidized or processed NOy species, NOy-NOx, as NOZ because this quantity is related to the
30 degree of photochemical aging in the atmosphere. Several NOZ species such as PAN and HONO
31 can be readily photolyzed or thermally dissociated to NO or NO2 and thus act as reservoirs for
January 2005 AX2-145 DRAFT-DO NOT QUOTE OR CITE
-------
1 NOX. This discussion focuses on current methods and on promising new technologies, but no
2 attempt is made here to cover the extensive development of these methods or of methods such as
3 wet chemical techniques, no longer in widespread use. More detailed discussions of the histories
4 of these methods may be found elsewhere (U.S. Environmental Protection Agency, 1993, 1996).
5
6 AX2.6.2.1 Calibration Standards
7 Calibration gas standards of NO, in nitrogen (certified at concentrations of approximately
8 5 to 40 ppm) are obtainable from the Standard Reference Material (SRM) Program of the
9 National Institute of Standards and Technology (NIST), formerly the National Bureau of
10 Standards (NBS), in Gaithersburg, MD. These SRMs are supplied as compressed gas mixtures
11 at about 135 bar (1900 psi) in high-pressure aluminum cylinders containing 800 L of gas at
12 standard temperature and pressure, dry (STPD; National Bureau of Standards, 1975; Guenther
13 et al., 1996). Each cylinder is supplied with a certificate stating concentration and uncertainty.
14 The concentrations are certified to be accurate to ±1 percent relative to the stated values.
15 Because of the resources required for their certification, SRMs are not intended for use as daily
16 working standards, but rather as primary standards against which transfer standards can be
17 calibrated.
18 Transfer stand-alone calibration gas standards of NO in N2 (in the concentrations indicated
19 above) are obtainable from specialty gas companies. Information as to whether a company
20 supplies such mixtures is obtainable from the company, or from the SRM Program of NIST.
21 These NIST Traceable Reference Materials (NTRMs) are purchased directly from industry and
22 are supplied as compressed gas mixtures at approximately 135 bars (1,900 psi) in high-pressure
23 aluminum cylinders containing 4,000 L of gas at STPD. Each cylinder is supplied with a
24 certificate stating concentration and uncertainty. The concentrations are certified to be accurate
25 to within ±1 percent of the stated values (Guenther et al., 1996). Additional details can be found
26 in the previous AQCD for O3 (U.S. Environmental Protection Agency, 1996).
27
28 AX2.6.2.2 Measurement of Nitric Oxide
29 Gas-Phase Chemiluminescence (CL) Methods
30 Nitric oxide, NO, can be measured reliably using the principle of gas-phase
31 chemiluminescence induced by the reaction of NO with O3 at low pressure. Modern commercial
January 2005 AX2-146 DRAFT-DO NOT QUOTE OR CITE
-------
1 NOX analyzers have sufficient sensitivity and specificity for adequate measurement in urban and
2 many rural locations (U.S. Environmental Protection Agency, 1996). The physics of the method,
3 detection limits, interferences, and comparisons under field comparisons have been thoroughly
4 reviewed in the previous AQCD. Research grade CL instruments have been compared under
5 realistic field conditions to spectroscopic instruments, and the results indicate that both methods
6 are reliable (at concentrations relevant to smog studies) to better than 15 percent with 95 percent
7 confidence. Response times are on the order of 1 minute. For measurements meaningful for
8 understanding O3 formation, emissions modeling, and N deposition, special care must be taken
9 to frequently zero and calibrate the instrument. A chemical zero, by reacting the NO up stream
10 and out of view of the PMT, is preferred because it accounts for unsaturated hydrocarbon or
11 other interferences. Calibration should be performed with NIST-traceable reference material of
12 compressed NO in N2. Standard additions of NO at the inlet will account for NO loss or
13 conversion to NO2 in the lines. In summary CL methods, when operated in an appropriate
14 manner, can be suitable for measuring or monitoring NO (e.g., Crosley, 1996).
15
16 Spectroscopic Methods for Nitric Oxide
17 Nitric oxide has also been successfully measured in ambient air with direct spectroscopic
18 methods; these include two-photon laser-induced fluorescence (TPLIF), tunable diode laser
19 absorption spectroscopy (TDLAS), and two-tone frequency-modulated spectroscopy (TTFMS).
20 These were reviewed thoroughly in the previous AQCD and will be only briefly summarized
21 here. The spectroscopic methods demonstrate excellent sensitivity and selectivity for NO with
22 detection limits on the order of 10 ppt for integration times of 1 min. Spectroscopic methods
23 compare well with the CL method for NO in controlled laboratory air, ambient air, and heavily
24 polluted air (e.g., Walega et al., 1984; Gregory et al., 1990; Kireev et al., 1999). These
25 spectroscopic methods remain in the research arena due to their complexity, size, and cost, but
26 are essential for demonstrating that CL methods are reliable for monitoring NO concentrations
27 involved in O3 formation — from 100s of ppb to around 20 ppt.
28 Atmospheric pressure laser ionization followed by mass spectroscopy has also been
29 reported for detection of NO and NO2. Garnica et al. (2000) describe a technique involving
30 selective excitation at one wavelength followed by ionization at a second wavelength. They
January 2005 AX2-147 DRAFT-DO NOT QUOTE OR CITE
-------
1 report good selectivity and detection limits well below 1 ppb. The practicality of the instrument
2 for ambient monitoring has yet to be demonstrated.
3
4 AX2.6.2.3 Measurements of Nitrogen Dioxide
5 Gas-Phase Chemiluminescence Methods
6 Since the previous AQCD, photolytic reduction followed by CL has been improved and the
7 method of laser-induced fluorescence has been developed. Ryerson et al. (2000) developed a
8 photolytic converter based on a Hg lamp with increased radiant intensity in the region of peak
9 NO2 photolysis (350 to 400 nm) and producing conversion efficiencies of 70% or more in less
10 than 1 s. Because the converter produces little radiation at wavelengths less than 350 nm,
11 interferences from HNO3 and PAN are minimal.
12 Alternative methods to photolytic reduction followed by CL are desirable to test the
13 reliability of this widely used technique. In any detector based on conversion to another species
14 interferences can be a problem. Several atmospheric species, PAN and HO2NO2 for example,
15 dissociate to NO2 at higher temperatures.
16 Laser induced fluorescence for NO2 detection involves excitation of atmospheric NO2 with
17 laser light emitted at wavelengths too long to induce photolysis. The resulting excited molecules
18 relax in a photoemissive mode and the fluorescing photons are counted. Because collisions
19 would rapidly quench electronically excited NO2, the reactions are conducted at low pressure
20 (Cohen, 1999; Thornton et al., 2000; Day et al., 2002). For example Cleary et al. (2002)
21 describe field tests of a system that uses continuous, supersonic expansion followed by
22 excitation at 640 nm with a commercial cw external-cavity tunable diode laser. Sensitivity is
23 adequate for measurements in most continental environments (145 ppt in 1 min) and no
24 interferences have been identified.
25 Matsumi et al. (2001) describe a comparison of laser-induced fluorescence with a
26 photofragmentation chemiluminescence instrument. The laser-induced fluorescence system
27 involves excitation at 440 nm with a multiple laser system. They report sensitivity of 30 ppt in
28 10s and good agreement between the two methods under laboratory conditions at mixing ratios
29 up to 1.0 ppb. This high-sensitivity laser-induced fluorescence system has yet to undergo long-
30 term field tests.
January 2005 AX2-148 DRAFT-DO NOT QUOTE OR CITE
-------
1 NO2 can be detected by differential optical absorption spectroscopy (DOAS) in an open,
2 long-path system (Kim and Kim, 2001). Vandaele et al. (2002) reported that the DOAS
3 technique measured higher NO2 concentrations than were reported by other techniques in a
4 three-year study conducted in Brussels. Harder et al. (1997b) conducted an experiment in rural
5 Colorado involving simultaneous measurements of NO2 with DOAS and photolysis followed by
6 chemiluminescence. The found differences of as much as 110% in clean air from the west, but
7 for NO2 mixing ratios in excess of 300 ppt, the two methods agreed to better than 10%. Stutz
8 and Platt (1996) report less uncertainty.
9
10 AX2.6.2.4 Monitoring for NO2 Compliance Versus Monitoring for Ozone Formation
11 Observations of NO2 have been focused on demonstrating compliance with the NAAQS for
12 NO2. Today, few locations violate that standard, but NO2 and related NOy compounds remain
13 among the most important atmospheric trace gases to measure and understand. Commercial
14 instruments for NO/NOX detection are generally constructed with an internal converter for
15 reduction of NO2 to NO, and generate a signal referred to as NOX. These converters, generally
16 constructed of molybdenum oxides (MoOx), reduce not only NO2 but also most other NOy
17 species (Fehsenfeld et al., 1987; Crosley, 1996; Nunnermacker et al., 1998). Thus the NOX
18 signal is more accurately referred to as NOy. Unfortunately with an internal converter, the
19 instruments may not give a faithful indication of NOy either — reactive species such as HNO3
20 will adhere to the walls of the inlet system. Most recently, commercial vendors such as Thermo
21 Environmental (Franklin, MA) have offered NO/NOy detectors with external Mo converters.
22 If such instruments are calibrated through the inlet with a reactive nitrogen species such as
23 propyl nitrate, they should give accurate measurements of total NOy, suitable for evaluation of
24 photochemical models. States should be encouraged to make these NOy measurements where
25 ever possible.
26
27 AX2.6.3 Measurements of Nitric Acid Vapor, HNO3
28 Accurate measurement of nitric acid vapor, HNO3, has presented a long-standing analytical
29 challenge to the atmospheric chemistry community. In this context, it is useful to consider the
30 major factors that control HNO3 partitioning between the gas and deliquesced-particulate phases
31 in ambient air. In equation form,
January 2005 AX2-149 DRAFT-DO NOT QUOTE OR CITE
-------
K^ Ka
2 HN03g O [HN03aq] <=> [H+] + [NO3-] (AX2-52)
3
4 where KH is the Henry's Law constant in M atm-1 and Ka is the acid dissociation constant in M.
5 Thus, the primary controls on HNO3 phase partitioning are its thermodynamic properties
6 (KH, Ka, and associated temperature corrections), aerosol liquid water content (LWC), solution
7 pH, and kinetics. Aerosol LWC and pH are controlled by the relative mix of different acids and
8 bases in the system, hygroscopic properties of condensed compounds, and meteorological
9 conditions (RH, temperature, and pressure). It is evident from relationship XX that, in the
10 presence of chemically distinct aerosols of varying acidities (e.g., super-Jim predominantly sea
11 salt and sub-jam predominantly S aerosol), HNO3 will partition preferentially with the less-acidic
12 particles, which is consistent with observations (e.g., Huebert et al., 1996; Keene and Savoie,
13 1998; Keene et al., 2002). Kinetics are controlled by atmospheric concentrations of HNO3 vapor
14 and particulate NO3 and the size distribution and corresponding atmospheric lifetimes of
15 particles against deposition. Sub-jim-diameter aerosols typically equilibrate with the gas phase
16 in seconds to minutes while super-um aerosols require hours to a day or more (e.g., Meng and
17 Seinfeld, 1996; Erickson et al., 1999. Consequently, smaller aerosol size fractions are typically
18 close to thermodynamic equilibrium with respect to HNO3 whereas larger size fractions (for
19 which atmospheric lifetimes against deposition range from hours to a few days) are often
20 undersaturated (e.g., Erickson et al., 1999; Keene and Savioe, 1998).
21 Many sampling techniques for HNO3 (e.g., standard filterpack and mist-chamber samplers)
22 employ upstream prefilters to remove particulate species from sample air. However, when
23 chemically distinct aerosols with different pHs (e.g., sea salt and S aerosols) mix together on a
24 bulk filter, the acidity of the bulk mixture will be greater than that of the less acidic aerosols with
25 which most NO3 is associated. This change in pH may cause the bulk mix to be supersaturated
26 with respect to HNO3 leading to volatilization and, thus, positive measurement bias in HNO3
27 sampled downstream. Alternatively, when undersaturated super-um size fractions (e.g., sea salt)
28 accumulate on a bulk filter and chemically interacts over time with HNO3 in the sample air
29 stream, scavenging may lead to negative bias in HNO3 sampled downsteam. Because the
30 magnitude of both effects will vary as functions of the overall composition and thermodynamic
31 state of the multiphase system, the combined influence can cause net positive or net negative
January 2005 AX2-150 DRAFT-DO NOT QUOTE OR CITE
-------
1 measurement bias in resulting data. Pressure drops across particle filters can also lead to artifact
2 volatilization and associated positive bias in HNO3 measured downstream.
3 Widely used methods for measuring HNO3 include standard filterpacks configured with
4 nylon or alkaline-impregnated filters (e.g., Goldan et al., 1983; Bardwell et al., 1990;
5 respectively) and standard mist chambers (Talbot et al., 1990). Samples are typically analyzed
6 by ion chromatography. Intercomparisons of these measurement techniques (e.g., Hering et al.,
7 1988; Tanner et al., 1989; Talbot et al., 1990) report differences of a factor of two or more.
8 More recently, sensitive HNO3 measurements based on the principle of Chemical
9 lonization Mass Spectroscopy (CIMS) have been reported (e.g., Huey et al., 1998; Mauldin
10 et al., 1998; Furutani and Akimoto, 2002; Neuman et al., 2002). CIMS relies on selective
11 formation of ions such as SiF5"-HNO3 or HSO4"-HNO3 followed by detection via mass
12 spectroscopy. Two CIMS techniques and a filter pack technique were intercompared in Boulder,
13 CO (Fehsenfeld et al., 1998). Results indicated excellent agreement (within 15%) between the
14 two CIMS instruments and between the CIMS and filterpack methods under relatively clean
15 conditions with HNO3 mixing ratios between 50 and 400 pptv. In more polluted air, the
16 filterpack technique generally yielded higher values than the CIMS suggesting that interactions
17 between chemically distinct particles on bulk filters is a more important source of bias in
18 polluted continental air. Differences were also greater at lower temperature when particulate
19 NO3 corresponded to relatively greater fractions of total NO3 .
20
21 AX2.6.4 Sampling and Analysis of Volatile Organic Compounds
22 Hydrocarbons can be measured with gas chromatography followed by flame ionization
23 detection (GC-FID). Detection by mass spectroscopy is sometimes used to confirm species
24 identified by retention time (Westberg and Zimmerman, 1993; Dewulf and Van Langenhove,
25 1997). Preconcentration is typically required for less abundant species. Details are available in
26 AQCD 96.
27 Because of their variety, nonmethane hydrocarbons pose special analytical problems,
28 and several laboratory and field studies have recently addresses the uncertainty of VOC
29 measurements. An intercomparison conducted with 16 components among 28 laboratories,
30 showed agreement on the order of 10s of percents (Apel et al., 1994). In a more recent
31 intercomparison (Apel et al., 1999) 36 investigators from around the world were asked to
January 2005 AX2-151 DRAFT-DO NOT QUOTE OR CITE
-------
1 identify and quantify C2 to C10 hydrocarbons (HCs) in a mixture in synthetic air. Calibration was
2 based on gas standards of individual compounds, such as propane in air, and a 16-compound
3 mixture of C2 to C16 n-alkanes, all prepared by NIST and certified to ± 3 percent. The
4 top-performing laboratories, including several in the United States, identified all the compounds
5 correctly, and obtained agreement of generally better than 20 percent for the 60 compounds.
6 Intercomparison of NMHCs in ambient air has only recently been reported by a European group
7 of 12 - 14 laboratories (Slemr et al., 2002). Some compounds gave several groups difficulties,
8 including isobutene, butadiene, methyl pentanes, and trimethyl benzenes. These
9 intercomparisons illustrated the need for reliable, multicomponent calibration standards.
10
11 AX2.6.4.1 Polar Volatile Organic Compounds
12 Many of the more reactive oxygen- and nitrogen-containing organic compounds play a role
13 in O3 formation and are included among list of 189 hazardous air pollutants specified in the 1990
14 CAAA (U.S. Congress, 1990). These compounds are emitted directly from a variety of sources
15 including biogenic processes, biomass burning, industry, vehicles, and consumer products.
16 Some can also be formed in the atmosphere by photochemical oxidation of hydrocarbons.
17 Although these compounds have been referred to collectively as PVOCs, their reactivity and
18 water solubility, rather than just polarity, make sampling and measurement challenging. As
19 indicated in the earlier AQCD, few ambient data exist for these species, but that database has
20 grown. The previous AQCD discusses two analytical methods for PVOCs — cryogenic trapping
21 techniques similar to those used for the nonpolar hydrocarbon species, and adsorbent material
22 for sample preconcentration. Here we discuss recently developed methods.
23 Several techniques for sampling, preconcentrating and detecting oxygenated volatile
24 organic compounds were inter-compared during the 1995 Southern Oxidants Study Nashville
25 Intensive (Apel et al., 1998). Both chemical traps and derivatization followed by HPLC and
26 pre-concentration and gas chromatography followed by mass spectrometric of flame ionization
27 were investigated. Both laboratory and field tests were conducted for formaldehyde,
28 acetaldehyde, acetone, and propanal. Substantial differences were observed indicating that
29 reliable sampling and measurement of PVOCs remains an analytical challenge and high research
30 priority.
January 2005 AX2-152 DRAFT-DO NOT QUOTE OR CITE
-------
1 Chemical ionization-mass spectroscopy, such as proton-transfer-reaction mass
2 spectroscopy (PTR-MS) can also be used for fast-response measurement of volatile organic
3 compounds including acetonitrile (CH3CN), methanol (CH3OH), acetone (CH3COCH3),
4 acetaldehyde (CH3CHO), benzene (C6H6) and toluene (C6H5CH3) (e.g., Hansel et al., 1995a,b;
5 Lindinger et al., 1998; Leibrock and Huey, 2000; Warneke et al., 2001). The method relies on
6 gas phase proton transfer reactions between H3O+ primary ions and volatile trace gases with a
7 proton affinity higher than that of water. Into a flow drift tube continuously flushed with
8 ambient air, H3O+ ions (from a hollow cathode ion source) are injected. On collisions between
9 H3O+ ions and organic molecules protons H+ are transferred thus charging the reagent. Both
10 primary and product ions are analyzed in a quadrupole mass spectrometer and detected by a
11 secondary electron multiplier/pulse counting system. The instrument has been successfully
12 employed in several field campaigns and compared to other techniques including gas
13 chromatography and Atmospheric Pressure Chemical lonization Mass Spectrometer (AP-CEVIS)
14 (Crutzen et al., 2000; Sprung et al., 2001). Sufficient sensitivity was observed for urban and
15 rural measurements; no interferences were discovered, although care must be exercised to avoid
16 sampling losses. Commercial instruments are becoming available, but their price still precludes
17 wide spread monitoring.
18
19
January 2005 AX2-153 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2 Allan, B. I; McFiggans, G.; Plane, J. M. C. (2000) The nitrate radical in the remote marine boundary layer.
3 J. Geophys. Res. [Atmos.] 105(D19): 24,191-24,204.
4 Allan, B. J.; Plane, J. M. C.; McFiggans, G. (2001) Observations of OIO in the remote marine boundary layer.
5 Geophys. Res. Lett. 28: 1945-1948.
6 Allen, D. J.; Pickering, K. E. (2002) Evaluation of lightning flash rate parameterizations for use in a global chemical
7 transport model. J. Geophys. Res. [Atmos.] 107(D23): 10.1029/2002JD002066.
8 Allen, D. J.; Pickering, K. E.; Molod, A. (1997) An evaluation of deep convective mixing in the Goddard Chemical
9 Transport Model using ISCCP cloud parameters. J. Geophys. Res. [Atmos.] 102: 25,467-25,476.
10 Allen, D.; Pickering, K.; Stenchikov, G.; Thompson, A.; Kondo, Y. (2000) A three-dimensional total odd nitrogen
11 (NOy) simulation during SONEX using a stretched-grid chemical transport model. J. Geophys. Res. [Atmos.]
12 105: 3851-3876.
13 Altshuller, A. P. (1986) The role of nitrogen oxides in nonurban ozone formation in the planetary boundary layer
14 over N America, W Europe and adjacent areas of ocean. Atmos. Environ. 20: 245-268.
15 Altshuller, A. P.; Lefohn, A. S. (1996) Background ozone in the planetary boundary layer over the United States.
16 J. Air Waste Manage. Assoc. 46: 134-141.
17 Alvarez, II; Senff, C. J.; Hardesty, R. M.; Parrish, D. D.; Luke, W. T.; Watson, T. B.; Daum, P. H.; Gillani, N.
18 (1998) Comparisons of airborne lidar measurements of ozone with airborne in situ measurements during the
19 1995 SouthernOxidants study. J. Geophys. Res. [Atmos.] 103: 31155-35171.
20 American Society for Testing and Materials (ASTM) International. (2003a) Standard test method for ozone in the
21 atmosphere: continuous measurement by ethylene chemiluminescence. In: Annual Book of ASTM Standards
22 2003, v. 11.03: D 5149-02. West Conshohocken, PA: American Society for Testing and Materials;
23 pp. 443-447.
24 American Society for Testing and Materials (ASTM) International. (2003b) Standard test methods for continuous
25 measurement of ozone in ambient, workplace, and indoor atmospheres (ultraviolet absorption). In: Annual
26 Book of ASTM Standards 2003, v. 11.03: D 5156-02. West Conshohocken, PA: American Society for
27 Testing and Materials; pp. 448-454.
28 Ammann, M.; Kalberer, M.; Jost, D. T.; Tobler, L.; Rossler, E.; Piguet, D.; Gaggeler, H. W.; Baltensperger, U.
29 (1998) Heterogeneous production of nitrous acid on soot in polluted air masses. Nature (London)
30 395: 157-160.
31 Anastasio, C.; Faust, B. C.; Allen, J. M. (1994) Aqueous phase photochemical formation of hydrogen peroxide in
32 authentic cloud waters. J. Geophys. Res. [Atmos.] 99: 8231-8248.
33 Anastasio, C.; Faust, B.C.; Rao, C. J. (1997) Aromatic carbonyl compounds as aqueous-phase photochemical
34 sources of hydrogen peroxide in acidic sulfate aerosols, fogs, and clouds. 1. Non-phenolic
35 methoxybenzaldehydes and methoxyacetophenones with reductants (phenols). Environ. Sci. Technol.
36 31:218-232.
37 Anastasio, C.; Newberg, J. T.; Williams, D. K.; Chu, G. B.; Mathew, B. M. (1999) Photoformation of hydroxyl
38 radical in sea salt particles. EOS Trans. Am. Geophys. Union 80: F147.
39 Andino, J. M.; Smith, J. N.; Flagan, R. C.; Goddard, W. A., Ill; Seinfeld, J. H. (1996) Mechanism of the atmospheric
40 photooxidation of aromatics: a theoretical study. J. Phys. Chem. 100: 10,967-10,980.
41 Andreae, M. O. (1991) Biomass burning: its history, use, and distribution and its impact on environmental quality
42 and global climate. In: Levine, J. S., ed. Global biomass burning: atmospheric, climatic, and biospheric
43 implications. Cambridge, MA: MIT Press; pp. 1-21.
44 Andres Hernandez, M. D.; Burkert, J.; Reichert, L.; Stobener, D.; Meyer-Arnek, J.; Dickerson, R. R.; Doddridge,
45 E.G.; Burrows, J. P. (2001) Marine boundary layer peroxy radical chemistry during the AEROSOLS99
46 campaign: measurements and analysis. J. Geophys. Res. (Atmos.) 106(D18): 20,833-20,846.
47 Andronache, C.; Chameides, W. L.; Rodgers, M. O.; Martinez, J. E.; Zimmerman, P.; Greenberg, J. (1994) Vertical
48 distribution of isoprene in the lower boundary layer of the rural and urban southern United States. J. Geophys.
49 Res. [Atmos.]) 99: 16,989-17,000.
50 Angevine, W. M.; Buhr, M. P.; Holloway, J. S.; Trainer, M.; Parrish, D. D.; MacPherson, J. I.; Kok, G. L.;
51 Schillawski, R. D.; Bowlby, D. H. (1996) Local meteorological features affecting chemical measurements at a
52 North Atlantic coastal site. J. Geophys. Res. (Atmos.) 101(D22): 28,935-28,946.
53 Apel, E. C.; Calvert, J. G.; Fehsenfeld, F. C. (1994) The nonmethane hydrocarbon intercomparison experiment
54 (NOMHICE): tasks 1 and 2. J. Geophys. Res. [Atmos.] 99(D8): 16,651-16,664.
January 2005 AX2-154 DRAFT-DO NOT QUOTE OR CITE
-------
1 Apel, E. C.; Calvert, J. G.; Riemer, D.; Pos, W.; Zika, R.; Kleindienst, T. E.; Lonneman, W. A.; Fung, K.; Fujita, E.;
2 Shepson, P. B.; Starn, T. K.; Roberts, P. T. (1998) Measurements comparison of oxygenated volatile organic
3 compounds at a rural site during the 1995 SOS Nashville Intensive. J. Geophys. Res. (Atmos.)
4 103(D17): 22,295-22,316.
5 Apel, E. C.; Calvert, J. G.; Gilpin, T. M; Fehsenfeld, F. C.; Parrish, D. D.; Lonneman, W. A. (1999) The
6 Nonmethane Hydrocarbon Intercomparison Experiment (NOMHICE): task 3. J. Geophys. Res. [Atmos.]
7 104(021): 26,069-26,086.
8 Appenzeller, C.; Davies, H. C. (1992) Structure of stratospheric intrusions into the troposphere. Nature
9 358: 570-572.
10 Appenzeller, C.; Davies, H. C.; Norton, W. A. (1996) Fragmentation of stratospheric intrusions. J. Geophys. Res.
11 [Atmos.] 101: 1435-1456.
12 Arnold, F.; Burger, V.; Droste-Fanke, B.; Grimm, F.; Krieger, A.; Schneider, J.; Stilp, T. (1997) Acetone in the
13 upper troposphere and lower stratosphere: impact on trace gases and aerosols. Geophys. Res. Lett.
14 24:3017-3020.
15 Arnold, S. R.; Chipperfield, M. P.; Blitz, M. A.; Heard, D. E.; Pilling, M. J. (2004) Photodissociation of acetone:
16 atmospheric implications of temperature-dependent quantum yields. Geophys. Res. Lett.
17 31(L07110): 10.1029/2003GL019099.
18 Ashworth, S. H.; Allan, B. J.; Plane, J. M. C. (2002) High resolution spectroscopy of the OIO radical: implications
19 for the ozone-depleting potential of iodine. Geophys. Res. Lett. 29: 65-1 - 65-4.
20 Atherton, C. S.; Penner, J. E. (1988) The transformation of nitrogen oxides in the polluted troposphere. Tellus B
21 40:380-392.
22 Atkinson, R. (1989) Kinetics and mechanisms of the gas-phase reactions of the hydroxyl radical with organic
23 compounds. Washington, DC: American Chemical Society. (J. Phys. Chem. Ref. Data Monograph no. 1).
24 Atkinson, R. (1990) Gas-phase tropospheric chemistry of organic compounds: a review. Atmos. Environ. Part A
25 24: 1-41.
26 Atkinson, R. (1991) Kinetics and mechanisms of the gas-phase reactions of the NOX radical with organic
27 compounds. J. Phys. Chem. Ref. Data 20: 459-507.
28 Atkinson, R. (1994) Gas-phase tropospheric chemistry of organic compounds. Washington, DC: American Chemical
29 Society. (J. Phys. Chem. Ref. Data Monograph no. 2).
30 Atkinson, R. (1997) Atmospheric reactions of alkoxy and p-hydroxyalkoxy radicals. Int. J. Chem. Kinet. 29: 99-111.
31 Atkinson, R. (2000) Atmospheric chemistry of VOCs and NOX. Atmos. Environ. 34: 2063-2101.
32 Atkinson, R.; Arey, J. (2003) Gas-phase tropospheric chemistry of biogenic volatile organic compounds: a review.
33 Atmos. Environ.: 10.1016/81352-2310(03)00391-1.
34 Atkinson, R.; Aschmann, S. M. (1994) Products of the gas-phase reactions of aromatic hydrocarbons: effect of NO2
35 concentration. Int. J. Chem. Kinet. 26: 929-944.
36 Atkinson, R.; Carter, W. P. L.; Darnall, K. R.; Winer, A. M.; Pitts, J. N., Jr. (1980) A smog chamber and modeling
37 study of the gas phase NOx-air photooxidation of toluene and the cresols. Int. J. Chem. Kinet. 12: 779-836.
38 Atkinson, R.; Winer, A. M.; Pitts, J. N., Jr. (1986) Estimation of night-time N2O5 concentrations from ambient NO2
39 and NO3 radical concentrations and the role of N2O5 in night-time chemistry. Atmos. Environ. 20: 331-339.
40 Atkinson, R.; Aschmann, S. M.; Arey, J.; Carter, W. P. L. (1989) Formation of ring-retaining products from the OH
41 radical-initiated reactions of benzene and toluene. Int. J. Chem. Kinet. 21: 801-827.
42 Atkinson, R.; Arey, J.; Zielinska, B.; Aschmann, S. M. (1990) Kinetics and nitro-products of the gas-phase OH and
43 NO3 radical-initiated reactions of naphthalene-d8, fluoranthene-dlO and pyrene. Int. J. Chem. Kinet.
44 22:999-1014.
45 Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr, J. A.; Troe, J. (1992a) Evaluated kinetic and
46 photochemical data for atmospheric chemistry: supplement IV, IUPAC Subcommittee on Gas Kinetic Data
47 Evaluation for Atmospheric Chemistry. J. Phys. Chem. Ref. Data 21: 1125-1568.
48 Atkinson, R.; Aschmann, S. M.; Arey, J.; Shorees, B. (1992b) Formation of OH radicals in the gas phase reactions
49 of O3 with a series of terpenes. J. Geophys. Res. [Atmos.] 97: 6065-6073.
50 Atkinson, R.; Kwok, E. S. C.; Arey, J.; Aschmann, S. M. (1995) Faraday Discuss. 100: 23.
51 Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr, J. A.; Rossi, M. J.; Troe, J. (1999) Evaluated
52 kinetic and photochemical data for atmospheric chemistry, organic species: supplement VII. J. Phys. Chem.
53 Ref. Data 28: 191-393.
54 Ayers, G. P.; Penkett, S. A.; Gillett, R. W.; Bandy, B.; Galbally, I. E.; Meyer, C. P.; Elsworth, C. M.; Bentley, S. T.;
55 Forgan, B. W. (1992) Evidence for photochemical control of ozone concentrations in unpolluted marine air.
56 Nature (London) 360: 446-449.
January 2005 AX2-155 DRAFT-DO NOT QUOTE OR CITE
-------
1 Ayers, G. P.; Gillett, R. W.; McAiney, J. M.; Dick, A. L. (1999) Chloride and bromide loss from sea-salt particles in
2 the southern ocean air. J. Atmos. Chem. 33: 299-319.
3 Bais, A. F.; Madronich, S.; Crawford, J.; Hall, S. R.; Mayer, B.; Van Weele, M.; Lenoble, J.; Calvert, J. G.; Cantrell,
4 C. A.; Shelter, R. E.; Hofzumahaus, A.; Koepke, P.; Monks, P. S.; Frost, G.; McKenzie, R.; Krotkov, N.;
5 Kylling, A.; Swartz, W. H.; Lloyd, S.; Pfister, G.; Martin, T. J.; Roeth, E.-P.; Griffioen, E.; Ruggaber, A.;
6 Krol, M.; Kraus, A.; Edwards, G. D.; Mueller, M.; Lefer, B. L. (2003) International Photolysis Frequency
7 Measurement and Model Intercomparison (IPMMI): Spectral actinic solar flux measurements and modeling.
8 J. Geophys. Res. [Atmos.] 108(D16): 10.1029/2002JD002891.
9 Balkanski, Y. J.; Jacob, D. J.; Gardener, G. M.; Graustein, W. C.; Turekian, K. K. (1993) Transport and residence
10 times of tropospheric aerosols inferred from a global three-dimensional simulation of 210Pb. J. Geophys. Res.
11 [Atmos.] 98: 20,573-20,586.
12 Bandow, H.; Washida, N. (1985a) Ring-cleavage reactions of aromatic hydrocarbons studied by FT-IR
13 spectroscopy. II photooxidation of o-, m-, and^-xylenes in the NOx-air system. Bull. Chem. Soc. Jpn.
14 58:2541-2548.
15 Bandow, H.; Washida, N. (1985b) Ring-cleavage reactions of aromatic hydrocarbons studied by FT-IR
16 spectroscopy. Ill photooxidation of 1,2,3-, 1,2,4-, and 1,3,5-trimethylbenzene in the NOx-air system. Bull.
17 Chem. Soc. Jpn. 58: 2549-2555.
18 Banic, C. M.; Leaitch, W. R.; Isaac, G. A.; Couture, M. D.; Kleinman, L. L; Springston, S. R.; MacPherson, J. I.
19 (1996) Transport of ozone and sulfur to the North Atlantic atmosphere during the North Atlantic Regional
20 Experiment. J. Geophys. Res. [Atmos.] 101: 29,091-29,104.
21 Banta, R. M.; Senff, C. J.; White, A. B.; Trainer, M.; McNider, R. T.; Valente, R. J.; Mayor, S. D.; Alvarez, R. J.;
22 Hardesty, R. M.; Parrish, D.; Fehsenfeld, F. C. (1998) Daytime buildup and nighttime transport of urban
23 ozone in the boundary layer during a stagnation episode. J. Geophys. Res. [Atmos.] 103: 22,519-22,544.
24 Bardwell, C. A.; Maben, J. R.; Hurt, J. A.; Keene, W. C.; Galloway, J. N.; Boatman, J. F.; Wellman, D. (1990)
25 A technique using high-flow, dichotomous filter packs for measuring major atmospheric chemical
26 constituents. Glob. Biogeochem. Cycles 4: 151-163.
27 Barrie, L. A.; Bottenheim, J. W.; Crutzen, P. J.; Rasmussen, R. A. (1988) Ozone destruction at polar sunrise in the
28 lower Arctic atmosphere. Nature 334: 138-141.
29 Bartolotti, L. J.; Edney, E. O. (1995) Density functional theory derived intermediates from the OH initiated
30 atmospheric oxidation of toluene. Chem. Phys. Lett. 245: 119-122.
31 Baugues, K. (1986) A review of NMOC, NOX and NMOC/NOX ratios measured in 1984 and 1985. Research
32 Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards;
33 report no. EPA-450/4-86-015. Available from: NTIS, Springfield, VA; PB87-166963/HSU.
34 Baumann, K.; Williams, E. J.; Angevine, W. M.; Roberts, J. M.; Norton, R. B.; Frost, G. J.; Fehsenfeld, F. C.;
35 Springston, S. R.; Bertman, S. B.; Hartsell, B. (2000) Ozone production and transport near Nashville,
36 Tennessee: results from the 1994 study at New Hendersonville. J. Geophys. Res. [Atmos.]
37 105(D7): 9137-9153.
3 8 Behnke, W.; George, C.; Sheer, V.; Zetzsch, C. (1997) Production and decay of C1NO2 from the reaction of gaseous
39 N2O5 with NaCl solution: bulk and aerosol experiments. J. Geophys. Res. 102: 3795-3804.
40 Bell, G. B. (1960) Meteorological conditions during oxidant episodes in coastal San Diego County in October and
41 November, 1959. Sacramento, CA: State of California, Department of Public Health.
42 Benjamin, S. G.; Carlson, T. N. (1986) Some effects of surface heating and topography on the region severe storm
43 environment. Part I: three dimensional simulations. Mon. Weather Rev. 114: 307-329.
44 Bergin, M. H.; Pandis, S. N.; Davidson, C. I.; Jaffrezo, J.-L.; Dibb, J. E.; Russell, A. G.; Kuhns, H. D. (1996)
45 Modeling of the processing and removal of trace gas and aerosol species by Arctic radiation fogs and
46 comparison with measurements. J. Geophys. Res. [Atmos.] 101: 14,465-14,478.
47 Berkowitz, C. M.; Shaw, W. J. (1997) Airborne measurements of boundary layer chemistry during the Southern
48 Oxidant Study: a case study. J. Geophys. Res. [Atmos.] 102: 12,795-12,804.
49 Berkowitz, C. M.; Fast, J. D.; Sprinston, S. R.; Larsen, R. J.; Spicer, C. W.; Doskey, P. V.; Hubbe, J. M.; Plastridge,
50 R. (1998) Formation mechanisms and chemical characteristics of elevated photochemical layers over the
51 northeast United States. J. Geophys. Res. [Atmos.] 103: 10,631-10,647.
52 Berkowitz, C. M.; Fast, J. D.; Easter, R. C. (2000) Boundary layer vertical exchange processes and the mass budget
53 of ozone: observations and model results. J. Geophys. Res. [Atmos.] 105: 14,789-14,805.
54 Berkowitz, C. M.; Zaveri, R. A.; Bian, X.; Zhong, S.; Disselkamp, R. S.; Laulainen, N. S.; Chapman, E. G. (2001)
55 Aircraft observations of aerosols, O3 and NO,, in a nighttime urban plume. Atmos. Environ. 35: 2395-2404.
56 Berndt, T.; Boge, O.; Herrmann, H. (1999) On the formation of benzene oxide/oxepin in the gas-phase reaction of
57 OH radicals with benzene. Chem. Phys. Lett. 314: 435-442.
January 2005 AX2-156 DRAFT-DO NOT QUOTE OR CITE
-------
1 Berndt, T.; Boge, O.; Stratmann, F. (2003) Gas-phase ozonolysis of a-pinene: gaseous products and particle
2 formation. Atmos. Environ. 37: 3933-3945.
3 Berntsen, T.; Isaksen, I. S. A. (1997) A global three-dimensional chemical transport model for the troposphere. 1.
4 Model description and CO and ozone results. J. Geophys. Res. [Atmos.] 102: 21,239-21,280.
5 Bey, I.; Jacob, D. J.; Logan, J. A.; Yantosca, R. M. (2001a) Asian chemical outflow to the Pacific in spring: origins,
6 pathways, and budgets. J. Geophys. Res. [Atmos.] 106: 23,097-23,113.
7 Bey, I.; Jacob, D. J.; Yantosca, R. M.; Logan, J. A.; Field, B.; Fiore, A. M.; Li, Q.; Liu, H.; Mickley, L. J.; Schultz,
8 M. G. (200Ib) Global modeling of tropospheric chemistry with assimilated meteorology: model description
9 and evaluation. J. Geophys. Res. [Atmos.] 106: 23,073-23,095.
10 Bierbach, A.; Barnes, I.; Becker, K. H.; Wiesen, E. (1994) Atmospheric chemistry of unsaturated carbonyls:
11 butenedial, 4-Oxo-2-Pentenal, 3-Hexene-2,5-Dione, Maleic Anhydride, 3H-Furan-2-one, and
12 5-Methyl-3H-Furan-2-one. Environ. Sci. Technol. 28: 715-729.
13 Bishop, G. A.; Stedman, D. H. (1990) On-road carbon monoxide emission measurement comparisons for the
14 1988-1989 Colorado oxy-fuels program. Environ. Sci. Technol. 24: 843-847.
15 Bishop, G. A.; Stedman, D. H. (1996) Measuring the emissions of passing cars. Ace. Chem. Res. 29: 489-495.
16 Black, D. R.; Harley, R. A.; Hering, S. V.; Stolzenburg, M. R. (2000) A new, portable, real-time monitor. Environ.
17 Sci. Technol. 34:3031-3040.
18 Blanchard, C. L. (2000) Ozone process insights from field experiments. Part III: extent of reaction and ozone
19 formation. Atmos. Environ. 34: 2035-2043.
20 Blanchard, C. L.; Lurmann, F. W.; Roth, P. M.; Jeffries, H. E.; Korc, M. (1999) The use of ambient data to
21 corroborate analyses of ozone control strategies. Atmos. Environ. 33: 369-381.
22 Blanchard, C. L.; Stoeckenius, T. (2001) Ozone response to precursor control: comparision of data analysis methods
23 with the predictions of photochemical air quality simulation models. Atmos. Environ. 35: 1203-1215.
24 Blanchard, P.; Brook, J. R.; Brazil, P. (2002) Chemical characterization of the organic fraction of atmospheric
25 aerosol at two sites in Ontario, Canada. J. Geophys. Res. [Atmos.] 107(D21): 10.1029/2001JD000627.
26 Bloom, S. C.; Takacs, L. L.; Da Silva, A. M.; Ledvina, D. (1996) Data assimilation using incremental analysis
27 updates. Mon. Weather Rev. 124: 1256-1271.
28 Blough, N. V. (1997) Photochemistry in the sea-surface microlayer. In: Liss, P. S.; Duce, R. A., eds. The sea surface
29 and global change. New York, NY: Cambridge University Press, pp. 383-424.
30 Blough, N. V.; Zepp. R. G. (1995) Reactive oxygen species in natural waters. In: Foote, C. S.; Valentine, A.;
31 Greenberg, A.; Liebman, J. F., eds. Active Oxygen in Chemistry. New York, NY: Chapman and Hall, Ch.8,
32 pp. 280-333.
33 Blumenthal, D. L.; Lurmann, F. W.; Kumar, N.; Dye, T. S.; Ray, S. E.; Korc, M. E.; Londergan, R.; Moore, G.
34 (1997) Transport and mixing phenomena related to ozone exceedances in the northeast U.S. (analysis based
35 on NARSTO-northeast data). Available: http://capita.wustl.edu/otag/reports/otagrept/otagrept.html
36 (30 October 2003).
37 Bonn, B.; Moortgat, G. K. (2003) Sesquiterpene ozonolysis: origin of atmospheric new particle formation from
38 biogenic hydrocarbons. Geophys. Res. Lett. 30: 10.1029/2003GL017000.
39 Bonner, W. D. (1968) Climatology of the low level jet. Mon. Weather Rev. 96: 833-850.
40 Bottenheim, J. W.; Barrie, L. A.; Atlas, E.; Heidt, L. E.; Niki, H.; Rasmussen, R. A.; Shepson, P. B. (1990)
41 Depletion of lower tropospheric ozone during Arctic spring: The Polar Sunrise Experiment 1988. J. Geophys.
42 Res. (Atmos.) 95: 18,555-18,568.
43 Brasseur, G. P.; Hauglustaine, D. A.; Walters, S.; Rasch, P. J.; Mtiller, J.-F.; Granier, C.; Tie, X. X. (1998)
44 MOZART, a global transport model for ozone and related chemical tracers: 1. model description. J. Geophys.
45 Res. [Atmos.] 103: 28,265-28,289.
46 Brook, J. R.; Sirois, A.; Clarke, J. F. (1996) Comparison of dry deposition velocities for SO2, HNO3, and SO42'
47 estimated with two inferential models. Water Air Soil Pollut. 87: 205-218.
48 Broske, R.; Kleffmann, J.; Wiesen, P. (2003) Heterogeneous conversion of NO2 on secondary organic aerosol
49 surfaces: a possible source of nitrous acid (HONO) in the atmosphere? Atmos. Chem. Phys. 3: 469-474.
50 Browell, E. V.; Fenn, M. A.; Butler, C. F.; Grant, W. B.; Ismail, S.; Ferrare, R. A.; Kooi, S. A.; Brackett, V. G.;
51 Clayton, M. B.; Avery, M. A.; Barrick, J. D. W.; Fuelberg, H. E.; Maloney, J. C.; Newell, R. E.; Zhu, Y.;
52 Mahoney, M. J.; Anderson, B. E.; Blake, D. R.; Brune, W. H.; Heikes, B. G.; Sachse, G. W.; Singh, H. B.;
53 Talbot, R. W. (2001) Large-scale air mass characteristics observed over the remote tropical Pacific Ocean
54 during March-April 1999: results from PEM-Tropics B field experiment. J. Geophys. Res. [Atmos.]
55 106: 32,481-32,501.
January 2005 AX2-157 DRAFT-DO NOT QUOTE OR CITE
-------
1 Brueggemann, N.; Schnitzler, J.-P. (2002) Comparison of isoprene emission, intercellular isoprene concentration and
2 photo synthetic performance with water-limited oak (Quercus pubescens Willd. and Quercus robur L.)
3 saplings. Plant Biol. 4: 456-463.
4 Briihl, C.; Poschl, U.; Crutzen, P. J.; Steil, B. (2000) Acetone and PAN in the upper troposphere: impact on ozone
5 production from aircraft emissions. Atmos. Environ. 34: 3931-3938.
6 Brune, W. H. et al. (1999) OH and HO2 chemistry in the North Atlantic free troposphere. Geophys. Res. Lett.
7 26: 3077-3080.
8 Brunner, D.; Staehelin, J.; Jeker, D. (1998) Large-scale nitrogen oxide plumes in the tropopause region and
9 implications for ozone. Science (Washington, DC) 282: 1305-1309.
10 Buhr, M. P.; Trainer, M.; Parrish, D. D.; Sievers, R. E.; Fehsenfeld, F. C. (1992) Assessment of pollutant emission
11 inventories by principal component analysis of ambient air measurements. Geophys. Res. Lett. 19:
12 1009-1012.
13 Buhr, M.; Parrish, D.; Elliot, J.; Holloway, J.; Carpenter, J.; Goldan, P.; Kuster, W.; Trainer, M.; Montzka, S.;
14 McKeen, S.; Fehsenfeld, F. C. (1995) Evaluation of ozone precursor source types using principal component
15 analysis of ambient air measurements in rural Alabama. J. Geophys. Res. [Atmos.] 100: 22,853-22,860.
16 Buhr, M.; Sueper, D.; Trainer, M.; Goldan, P.; Kuster, B.; Fehsenfeld, F. (1996) Trace gas and aerosol
17 measurements using aircraft data from the North Atlantic Regional Experiment (NARE 1993). J. Geophys.
18 Res. [Atmos.] 101: 29,013-29,027.
19 Burkert, J.; Behmann, T.; Andres-Hernandez, M. D.; Stobener, D.; WeiBenmeyer, M.; Perner, D.; Burrows, J. P.
20 (200 la) Measurements of peroxy radicals in a forested area of Portugal. Chemosphere Glob. Change Sci.
21 3:327-338.
22 Burkert, J.; Andres-Hernandez, M. D.; Stobener, D.; Burrows, J. P.; Weilknmeyer, M.; Kraus, A. (200Ib) Peroxy
23 radical and related trace gas measurements in the boundary layer above the Atlantic Ocean. J. Geophys. Res.
24 (Atmos.) 106: 5457-5477.
25 Burkert, J.; Andres-Hernandez, M. D.; Reichert, L.; Meyer-Arnek, J.; Burrows, J. P.; Doddridge, B. G.; Dickerson,
26 R. R.; Muehle, J.; Zhan, A.; Carsey, T. (2003) Trace gas and radical diurnal behavior in the marine boundary
27 layer during INDOEX 1999. J. Geophys. Res. (Atmos.) 108(D8): 10.1029/2002JD002790.
28 Byun, D.; Young, J.; Gipson, G.; Schere, K.; Godowitch, J. (1998) Urban air quality simulation with community
29 multi-scale air quality (CMAQ) modeling system. Research Triangle Park, NC: U.S. Environmental
30 Protection Agency, National Exposure Research Laboratory; report no., EPA--600/A-98/115. Available from:
31 NTIS, Springfield, VA; PB99-102246.
32 Calvert, J. G.; Madronich, S. (1987) Theoretical study of the initial products of the atmospheric oxidation of
33 hydrocarbons. J. Geophys. Res. [Atmos.] 92: 2211-2220.
34 Calvert, J. G.; Heywood, J. B.; Sawyer, R. F.; Seinfeld, J. H. (1993) Achieving acceptable air quality: some
35 reflections on controlling vehicle emissions. Science (Washington, DC) 261: 37-45.
36 Calvert, J. G.; Yarwood, G.; Dunker, A. M. (1994) An evaluation of the mechanism of nitrous acid formation in the
37 urban atmosphere. Res. Chem. Intermed. 20: 463-502.
38 Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich, S.; Moortgat, G. K.; Wallington, T. J.; Yarwood, G. (2000)
3 9 The mechanisms of atmospheric oxidation of the alkenes. New York, NY: Oxford University Press, Inc.
40 Calvert, J. G.; Atkinson, R.; Becker, K. H.; Kamens, R. M.; Seinfeld, J. H.; Wallington, T. J.; Yarwood, G. (2002)
41 The mechanisms of atmospheric oxidation of aromatic hydrocarbons. New York, NY: Oxford University
42 Press, Inc.
43 Cantrell, C. A.; Shelter, R. E.; Gilpin, T. M.; Calvert, J. G.; Eisele, F. L.; Tanner, D. J. (1996) Peroxy radical
44 concentrations measured and calculated from trace gas measurements in the Mauna Loa Observatory
45 Photochemistry Experiment 2. J. Geophys. Res. (Atmos.) 101: 14,653-14,664.
46 Cantrell, C. A.; Calvert, J. G.; Bais, A.; Shelter, R. E.; Lefer, B. L.; Edwards, G. D. (2003) Overview and
47 conclusions of the International Photolysis Frequency Measurement and Modeling Intercomparison (IPMMI)
48 study. J. Geophys. Res. (Atmos.) 108(D16): 10.1029/2002JD002962.
49 Cardelino, C. A.; Chameides, W. L. (1990) Natural hydrocarbons, urbanization, and urban ozone. J. Geophys. Res.
50 [Atmos.] 95: 13,971-13,979.
51 Cardelino, C. A.; Chameides, W. L. (1995) An observation-based model for analyzing ozone precursor relationships
52 in the urban atmosphere. J. Air Waste Manage. Assoc. 45: 161-180.
53 Cardelino, C. A.; Chameides, W. L. (2000) The application of data from photochemical assessment monitoring
54 stations to the observation-based model. Atmos. Environ. 34: 2325-2332.
55 Carpenter, L. J.; Stages, W. T. Liss, P. S.; Penkett, S. A.; Alicke, B.; Hebestreit, K.; Platt, U. (1999) Short-lived
56 alkyl iodides and bromides at Mace Head, Ireland: links to biogenic sources and halogen oxide production.
57 J. Geophys. Res. (Atmos.) 104: 1679-1689.
January 2005 AX2-158 DRAFT-DO NOT QUOTE OR CITE
-------
1 Carroll, M. A.; Hastie, D. R.; Ridley, B. A.; Rodgers, M. O.; Torres, A. L.; Davis, D. D.; Bradshaw, J. D.;
2 Sandholm, S. T.; Schiff, H. L; Karecki, D. R.; Harris, G. W.; Mackay, G. L; Gregory, G. L.; Condon, E. P.;
3 Trainer, M.; Hubler, G.; Montzka, D. D.; Madronich, S.; Albritton, D. L.; Singh, H. B.; Beck, S. M.;
4 Shipham, M. C.; Bachmeier, A. S. (1990) Aircraft measurement of NOX over the eastern Pacific and
5 continental United States and implications for ozone production. J. Geophys. Res. [Atmos.] 95:
6 10,205-10,233.
7 Carroll, M. A.; Bertman, S. B.; Shepson, P. B. (2001) Overview of the program for research on oxidants,
8 photochemistry, emissions and transport (PROPHET) summer 1998 measurements intensive. J. Geophys.
9 Res. [Atmos.] 106: 24,275-24,288.
10 Carter, W. P. L. (1990) A detailed mechanism for the gas-phase atmospheric reactions of organic compounds.
11 Atmos. Environ. Part A 24: 481-518.
12 Carter, W. P. L. (1994) Development of ozone reactivity scales for volatile organic compounds. J. Air Waste
13 Manage. Assoc. 44: 881-899.
14 Carter, W. P. L. (1995) Computer modeling of environmental chamber studies of maximum incremental reactivities
15 of volatile organic compounds. Atmos. Environ. 29: 2513.
16 Castro, T; Madronich, S; Rivale, S; Muhlia, A; Mar, B. (2001) The influence of aerosols on photochemical smog in
17 Mexico City. Atmos. 35: 1765-1772..
18 Chameides, W. L.; Stelson, A. W. (1992) Aqueous-phase chemical processes in deliquescent sea-salt aerosols:
19 A mechanism that couples the atmospheric cycles of S and sea salt. J. Geophys. Res. [Atmos.]
20 97(D18): 20,565-20,580.
21 Chameides, W. L.; Lindsay, R. W.; Richardson, J.; Kiang, C. S. (1988) The role of biogenic hydrocarbons in urban
22 photochemical smog: Atlanta as a case study. Science (Washington, DC) 241: 1473-1475.
23 Chameides, W. L.; Fehsenfeld, F.; Rodgers, M. O.; Cardelino, C.; Martinez, J.; Parrish, D.; Lonneman, W.; Lawson,
24 D. R.; Rasmussen, R. A.; Zimmerman, P.; Greenberg, J.; Middleton, P.; Wang, T. (1992) Ozone precursor
25 relationships in the ambient atmosphere. J. Geophys. Res. [Atmos.] 97: 6037-6055.
26 Chang, J. S.; Brost, R. A.; Isaken, I. S. A.; Madronich, S.; Middleton, P.; Stockwell, W. R.; Walcek, C. J. (1987)
27 A three-dimensional Eulerian acid deposition model: physical concepts and formulation. J. Geophys. Res.
28 [Atmos.] 92: 14,681-14,700.
29 Chang, M. E.; Cardelino, C.; Hartley, D.; Chang, W. L. (1996) Inverse modeling of biogenic emissions. Geophys.
30 Res. Lett. 23: 3007ff.
31 Chang, M. E.; Hartley, D. E.; Cardelino, C.; Hass-Laursen, D.; Chang, W. L. (1997) On using inverse methods for
32 resolving emissions with large spatial inhomogeneities. J. Geophys. Res. (Atmos.) 102: 16,023-16,036.
33 Chang, T. Y.; Chock, D. P.; Nance, B. L; Winkler, S. L. (1997) A photochemical extent parameter to aid ozone air
34 quality management. Atmos. Environ. 31: 2787-2794.
35 Chang, S.; McDonand-Buller, E. C.; Kimura, Y.; Russell, M. M.; Tanaka, P. L.; Allen, D. T. (2002) Spatial and
3 6 temporal impacts of chlorine chemistry on ozone formation in southeastern Texas. Austin, TX: The Texas
37 Natural Resource Conservation Commission; Contract No. 9880077600-18. Available:
38 ftp://ftp.tnrcc.state.tx.us/pub/OEP AA/TAD/Modeling/HGAQSE/Contract_Reports/Others/Impacts_Chlorine_
39 on_Ozone_Formation.pdf (17 December 2003).
40 Chatfield, R. B.; Crutzen, P. J. (1984) Sulfur dioxide in remote oceanic air: cloud transport of reactive precursors.
41 J. Geophys. Res. 89:7111-7132.
42 Chen, L.-W. A.; Chow, J. C.; Doddridge, B. G.; Dickerson, R. R.; Ryan, W. F.; Mueller, P. K. (2003) Analysis of
43 summertime PM25 and haze episode in the mid-Atlantic region. J. Air Waste Manage. Assoc. 53: 946-956.
44 Cheng, M.-D.; Hopke, P. K.; Zeng, Y. (1993) A receptor-oriented methodology for determining source regions of
45 paniculate sulfate observed at Dorset, Ontario. J. Geophys. Res. [Atmos.] 98: 16,839-16,849.
46 Chesselet, R.; Fontugne, M.; Buat-Menard, P.; Ezat, U.; Lambert, C. E. (1981) The origin of paniculate organic
47 carbon in the marine atmosphere as indicated by its stable carbon isotopic composition. Geophys. Res. Lett.
48 8:345-348.
49 Ching, J. K. S.; Alkezweeny, A. J. (1986) Tracer study of vertical exchange by cumulus clouds. J. Clim. Appl.
50 Meteorol. 25: 1702-1711.
51 Ching, J. K. S.; Shipley, S. T.; Browell, E. V. (1988) Evidence for cloud venting of mixed layer ozone and aerosols.
52 Atmos. Environ. 22: 225-242.
53 Ching, J. K. S.; Byun, D.; Young, J.; Binkowski, F. S.; Pleim, J.; Roselle, S.; Godowitch, J.; Benjey, W.; Gipson, G.
54 (1998) Science features inModels-3 Community Multiscale Air Quality System. In: Preprints of the 10th
55 Joint AMS/AWMA Conference on Applications of Air Pollution Meteorology; Phoenix, AZ. Pittsburgh, PA:
56 Air & Waste Management Association.
January 2005 AX2-159 DRAFT-DO NOT QUOTE OR CITE
-------
1 Chock, D. P.; Winkler, S. L. (1994) A comparison of advection algorithms coupled with chemistry. Atmos. Environ.
2 28:2659-2675.
3 Choi, W.; Leu, M.-T. (1998) Nitric acid uptake and decomposition on black carbon (soot) surfaces: its implications
4 for the upper troposphere and lower stratosphere. J. Phys. Chem. A 102: 7618-7630.
5 Choi, Y. J.; Ehrman, S. H. (2004) Investigation of sources of volatile organic carbon in the Baltimore area using
6 highly time-resolved measurements. Atmos. Environ. 38: 775-791.
7 Civerolo, K.; Dickerson, R. R. (1998) Nitric oxide soil emissions from tilled and untilled cornfields. Agric. For.
8 Meteorol. 90:307-311.
9 Civerolo, K. L.; Mao, H. T.; Rao, S. T. (2003) The airshed for ozone and fine paniculate pollution in the eastern
10 United States. Pure Appl. Geophys. 160: 81-105.
11 Claeys, M; Wang, W.; Ion, A. C.; Kourtchev, L; Gelencser, A.; Maenhaut, W. (2004) Formation of secondary
12 organic aerosols from isoprene and its gas-phase oxidation products through reaction with hydrogen peroxide.
13 Atmos. Environ. 38: 4093-4098.
14 Clark, T. L.; Karl, T. R. (1982) Application of prognostic meteorological variables to forecasts of daily maximum
15 one-hour ozone concentrations in the northeastern United States. J. Appl. Meteorol. 21: 1662-1671.
16 Cleary, P. A.; Wooldridge, P. J.; Cohen, R. C. (2002) Laser-induced fluorescence detection of atmospheric NO2 with
17 a commercial diode laser and a supersonic expansion. Appl. Optics 41: 6950-6956.
18 Cleveland, W. S.; Guarino, R.; Kleiner, B.; McRae, J. E.; Warner, J. L. (1976) The analysis of the ozone problem in
19 the northeast United States. In: Specialty conference on: ozone/oxidants-interactions with the total
20 environment; March; Dallas, TX. Pittsburgh, PA: Air Pollution Control Association; pp. 109-119.
21 Cleveland, W. S.; Kleiner, B.; McRae, J. E.; Warner, J. L. (1976) Photochemical air pollution: transport from
22 New York City area into Connecticut and Massachusetts. Science (Washington, DC) 191: 179-181.
23 Cohen, R. C. (1999) Laser-induced fluorescence detection of atmospheric NO2 at parts per trillion mixing ratios:
24 implications for nitrogen oxide photochemistry in the stratosphere and troposphere. Abstr. Pap. Am. Chem.
25 Soc. (Phys.) 218(Pt. 2): 262.
26 Collins, W. J.; Stevenson, D. S.; Johnson, C. E.; Derwent, R. G. (1997) Tropospheric ozone in a global-scale
27 three-dimensional Lagrangian model and its response to NO^ emission controls. J. Atmos. Chem.
28 26:223-274.
29 Comes, F. J.; Armerding, W.; Grigonis, R.; Herbert, A.; Spiekermann, M.; Walter, J. (1992) Tropospheric OH: local
30 measurements and their interpretations. Ber. Bunsen-Ges. Phys. Chem. 96: 284-286.
31 Cooper, O. R.; Moody, J. L. (2000) Meteorological controls on ozone at an elevated eastern United States regional
32 background monitoring site. J. Geophys. Res. [Atmos.] 105: 6855-6869.
33 Cooper, O. R.; Moody, J. L.; Davenport, J. C.; Oltmans, S. J.; Johnson, B. J.; Chen, X.; Shepson, P. B.; Merrill, J. T.
34 (1998) Influence of springtime weather systems on vertical ozone distributions over three North American
35 sites. J. Geophys. Res. [Atmos.] 103: 22,001-22,013.
36 Cooper, O. R.; Moody, J. L.; Parrish, D. D.; Trainer, M.; Ryerson, T. B.; Holloway, J. S.; Htibler, G.; Fehsenfeld,
37 F. C.; Oltmans, S. J.; Evans, M. J. (2001) Trace gas signatures of the airstreams within North Atlantic
3 8 cyclones: case studies from the North Atlantic Regional Experiment (NARE '97) aircraft intensive. J.
39 Geophys. Res. [Atmos.] 106: 5437-5456.
40 Cooper, O. R.; Moody, J. L.; Parrish, D. D.; Trainer, M.; Ryerson, T. B.; Holloway, J. S.; Htibler, G.; Fehsenfeld,
41 F. C.; Evans, M. J. (2002a) Trace gas composition of mid-latitude cyclones over the western North Atlantic
42 Ocean: a conceptual model. J. Geophys. Res. [Atmos.] 107(D7): 10.1029/2000JD000901.
43 Cooper, O. R.; Moody, J. L.; Parrish, D. D.; Trainer, M.; Holloway, J. S.; Htibler, G.; Fehsenfeld, F. C.; Oltmans,
44 S. J.; Evans, M. J. (2002b) Trace gas composition of mid-latitude cyclones over the western North Atlantic
45 Ocean: a seasonal comparison of O3 and CO. J. Geophys. Res. [Atmos.] 107(D7): 10.1029/2001JD000902.
46 Cooper, O.; Forster, C.; Parrish, D.; Dunlea, E.; Htibler, G.; Fehsenfeld, F.; Holloway, J.; Oltmans, S.; Johnson, B.;
47 Wimmers, A.; Horowitz, L. (2004a) On the life cycle of a stratospheric intrusion and its dispersion into
48 polluted warm conveyor belts. J. Geophys. Res. (Atmos.) 109(D23S09): 10.1029/2003 JD004006.
49 Cooper, O. R.; Forster, C.; Parrish, D.; Trainer, M.; Dunlea, E.; Ryerson, T.; Htibler, G.; Fehsenfeld, F.; Nicks, D.;
50 Holloway, J.; De Gouw, J.; Warneke, C.; Roberts, J. M.; Flocke, F.; Moody, J. (2004b) A case study of
51 transpacific warm conveyor belt transport: the influence of merging airstreams on trace gas import to North
52 America. J. Geophys. Res. (Atmos.) 109(D23S08): 10.1029/2003JD003624.
53 Corchnoy, S. B.; Atkinson, R. (1990) Kinetics of the gas-phase reactions of OH and NO3 radicals with 2-Carene,
54 l,8-cineole,/>-cymene, and terpinolene. Environ. Sci. Technol. 24: 1497-1502.
55 Cornell, S.; Rendell, A.; Jickells, T. (1995) Atmospheric inputs of dissolved organic nitrogen to the oceans. Nature
56 376:243-246.
January 2005 AX2-160 DRAFT-DO NOT QUOTE OR CITE
-------
1 Cornell, S. E.; Jickells, T. D.; Thornton, C. A. (1998) Urea in rainwater and atmospheric aerosols. Atmos. Environ.
2 32: 1903-1910.
3 Cornell, S.; Mace, K.; Coeppicus, S.; Duce, R.; Huebert, B.; Jickells, T.; Zhuang, L. Z. (2001) Organic nitrogen in
4 Hawaiian rain and aerosols. J. Geophys. Res. (Atmos.) 106: 7973-7983.
5 Corsmeier, U.; Kalthhoff, N.; Kolle, O.; Motzian, M.; Fiedler, F. (1997) Ozone concentration jump in the stable
6 nocturnal boundary layer during a LLJ-event. Atmos. Environ. 31: 1977-1989.
7 Cowling, E. B.; Chameides, W. L.; Kiang, C. S.; Fehsenfeld, F. C.; Meagher, J. F. (2000) Introduction to special
8 section: Southern Oxidant Study Nashville/Middle Tennessee Ozone Study, part 2. J. Geophys. Res. (Atmos.)
9 105: 9075-9077.
10 Cox, R. A.; Bloss, W. J.; Jones, R. L.; Rowley, D. M. (1999) OIO and the atmospheric cycle of iodine. Geophys.
11 Res. Lett. 26: 1857-1860.
12 Crawford, J. H.; Davis, D. D.; Chen, G.; Buhr, M.; Oltmans, S.; Weller, R.; Mauldin, L.; Eisele, F.; Shelter, R.;
13 Lefer, B.; Arimoto, R.; Hogan, A. (2001) Evidence for photochemical production of ozone at the South Pole
14 surface. Geophys. Res. Lett. 28: 3641-3644.
15 Crosley, D. R. (1996) NOyblue ribbon panel. J. Geophys. Res. [Atmos.] 101(D1): 2049-2052.
16 Crutzen, P. (1973) A discussion of the chemistry of some minor constituents in the stratosphere and troposphere.
17 Pure Appl. Geophys. 106-108: 1385-1399.
18 Crutzen, P. J.; Gidel, L. T. (1983) A two-dimensional photochemical model of the atmosphere. 2. The tropospheric
19 budgets of the anthropogenic chlorocarbons, CO, CH4, CH3C1 and the effect of various NOX sources on
20 tropospheric ozone. J. Geophys. Res. [Atmos.] 88: 6641-6661.
21 Crutzen, P. J.; Lawrence, M. G.; Poschl, U. (1999) On the background photochemistry of tropospheric ozone. Tellus
22 51A-B: 123-146.
23 Crutzen, P. J.; Williams, J.; Poschl, U.; Hoor, P.; Fischer, H.; Warneke, C.; Holzinger, R.; Hansel, A.; Lindinger,
24 W.; Scheeren, B.; Lelieveld, J. (2000) High spatial and temporal resolution measurements of primary organics
25 and their oxidation products over the tropical forest of Surinam. Atmos. Environ. 34: 1161-1165.
26 Cziczo, D. J.; Abbatt, J. P. D. (1999) Deliquescence, efflorescence, and supercooling of ammonium sulfate aerosols
27 at low temperature: implications for cirrus cloud formation and aerosol phase in the atmosphere. J. Geophys.
28 Res. [Atmos.] 104: 13,781-13,790.
29 Czoschke, N. M.; Jang, M.; Kamens, R. M. (2003) Effect of acidic seed onbiogenic secondary organic aerosol
30 growth. Atmos. Environ. 37: 4287-4299.
31 Danielsen, E. F. (1968) Stratospheric-tropospheric exchange based on radioactivity, ozone and potential vorticity.
32 J. Atmos. Sci. 25:502-518.
3 3 Danielsen, E. F. (1980) Stratospheric source for unexpectedly large values of ozone measured over the Pacific Ocean
34 during Gametag, August 1977. J. Geophys. Res. C: Oceans Atmos. 85: 401-412.
3 5 Danielsen, E. F. (1993) In situ evidence of rapid, vertical, irreversible transport of lower tropospheric air into the
3 6 lower tropical stratosphere by convective cloud turrets and by larger-scale upwelling in tropical cyclones.
37 J. Geophys. Res. [Atmos.] 98(D5): 8665-8681.
38 Danielsen, E. F.; Mohnen, V. A. (1977) Project Dustorm report: ozone transport, in situ measurements, and
39 meteorological analyses of tropopause folding. J. Geophys. Res. [Atmos.] 82: 5867-5877.
40 Danielsen, E. F., Hipskind, R. S.; Gaines, S. E.; Sachse, G. W.; Gregory, G. L.; Hill, G. F. (1987) Three-dimensional
41 analysis of potential vorticity associated with tropopause folds and observed variations of ozone and carbon
42 monoxide. J. Geophys. Res. [Atmos.] 92: 2103-2111.
43 Daum, P. H.; Kleinman, L. I.; Hills, A. J.; Lazrus, A. L.; Leslie, A. C. D.; Busness, K.; Boatman, J. (1990)
44 Measurement and interpretation of concentrations of H2O2 and related species in the upper midwest during
45 summer. J. Geophys. Res. [Atmos.] 95: 9857-9871.
46 Daum, P. H.; Kleinman, L. I.; Newman, L.; Luke, W. T.; Weinstein-Lloyd, J.; Berkowitz, C. M.; Busness, K. M.
47 (1996) Chemical and physical properties of plumes of anthropogenic pollutants transported over the North
48 Atlantic during the North Atlantic Regional Experiment. J. Geophys. Res. [Atmos.] 101: 29,029-29,042.
49 Davidson, E. A.; Kingerlee, W. (1997) A global inventory of nitric oxide emissions from soils. Nutr. Cycling
50 Agroecosyst. 48: 37-50.
51 Davies, T. D.; Schuepbach, E. (1994) Episodes of high ozone concentrations at the earth's surface resulting from
52 transport down from the upper troposphere/lower stratosphere: a review and case studies. Atmos. Environ.
53 28:53-68.
54 Davis, D. D.; Bellinger, W.; Fischer, S. (1975) A kinetics study of the reaction of the OH free radical with aromatic
55 compounds I. Absolute rate constants for reaction with benzene and toluene at 300°K. J. Phys. Chem.
56 79:293-294.
January 2005 AX2-161 DRAFT-DO NOT QUOTE OR CITE
-------
1 Day, D. A.; Wooldridge, P. I; Dillon, M. B.; Thornton, J. A.; Cohen, R. C. (2002) A thermal dissociation
2 laser-induced fluorescence instrument for in situ detection of NO2, peroxy nitrates, alkyl nitrates, and HNO3.
3 J. Geophys. Res. (Atmos.) 107(D5-D6): 10.1029/2001JD000779.
4 DeCaria, A. J.; Pickering, K. E.; Stenchikov, G. L.; Scala, J. R.; Stith, J. L.; Dye, J. E.; Ridley, B. A.; Laroche, P.
5 (2000) A cloud-scale model study of lightning-generate NOX in an individual thunderstorm during
6 STERAO-A. J. Geophys. Res. [Atmos.] 105: 11601-11616.
7 DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.; Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.;
8 Kolb, C. E.; Molina, M. J. (1992) Chemical kinetics and photochemical data for use in stratospheric
9 modeling: evaluation number 10. Pasadena, CA: National Aeronautics and Space Administration, Jet
10 Propulsion Laboratory; JPL publication no. 92-20.
11 Dennis, R. L.; Novak, J. H. (1992) EPA's third generation modeling system (Models-3): an overview. In: Berglund,
12 R. L., ed. Tropospheric ozone and the environment II: effects, modeling and control [papers from an
13 international specialty conference]; November; Atlanta, GA. Pittsburgh, PA: Air & Waste Management
14 Association; pp. 137-147. (A&WMA transactions series no. 20).
15 De Reus, M.; Dentener, F.; Thomas, A.; Borrmann, S.; Strom, J.; Lelieveld, J. (2000) Airborne observations of dust
16 aerosol over the North Atlantic Ocean during ACE 2: indications for heterogeneous ozone destruction.
17 J. Geophys. Res. [Atmos.] 105: 15,263-15,275.
18 Derwent, R. G.; Jenkin, M. E. (1991) Hydrocarbons and the long-range transport of ozone and PAN across Europe.
19 Atmos. Environ. Part A 25: 1661-1678.
20 Derwent, R. G.; Jenkin, M. E.; Saunders, S. M. (1996) Photochemical ozone creation potentials for a large number
21 of reactive hydrocarbons under European conditions. Atmos. Environ. 30: 181-199.
22 Derwent, R. G.; Jenkin, M. E.; Saunders, S. M.; Pilling, M. J. (1998) Photochemical ozone creation potentials for
23 organic compounds in northwest Europe calculated with a master chemical mechanism. Atmos. Environ.
24 32:2429-2441.
25 Derwent, R. G.; Collins, W. J.; Johnson, C. E.; Stevenson, D. S. (2001) Transient behaviour of tropospheric ozone
26 precursors in a global 3-D CTM and their indirect greenhouse effects. Climatic Change 49: 463-487.
27 Dewulf, J.; Van Langenhove, H. (1997) Analytical techniques for the determination and measurement data of
28 7 chlorinated Cl- and C2-hydrocarbons and 6 monocyclic aromatic hydrocarbons in remote air masses:
29 an overview. Atmos. Environ. 31: 3291-3307.
30 Dibb, J. E.; Arsenault, M.; Peterson, M. C. (2002) Fast nitrogen oxide photochemistry in Summit, Greenland snow.
31 Atmos. Environ. 36: 2501-2511.
32 Dickerson, R. R.; Huffman, G. J.; Luke, W. T.; Nunnermacker, L. J.; Pickering, K. E.; Leslie, A. C. D.; Lindsey,
33 C. G.; Slinn, W. G. N.; Kelly, T. J.; Damn, P. H.; Delany, A. C.; Greenberg, J. P.; Zimmerman, P. R.;
34 Boatman, J. F.; Ray, J. D.; Stedman, D. H. (1987) Thunderstorms: an important mechanism in the transport of
35 air pollutants. Science (Washington, DC) 235: 460-465.
36 Dickerson, R. R.; Doddridge, B. G.; Kelley, P.; Rhoads, K. P. (1995) Large-scale pollution of the atmosphere over
37 the remote Atlantic Ocean: evidence fromBermuda. J. Geophys. Res. [Atmos.]100: 8945-8952.
38 Dickerson, R. R.; Kondragunta, S.; Stenchikov, G.; Civerolo, K. L.; Doddridge, B. G.; Holben, B. N. (1997) The
3 9 impact of aerosols on solar ultraviolet radiation and photochemical smog. Science (Washington, DC)
40 278: 827-830.
41 Dickerson, R. R.; Rhoads, K. P.; Carsey, T. P.; Oltmans, S. J.; Burrows, J. P.; Crutzen, P. J. (1999) Ozone in the
42 remote marine boundary layer: a possible role for halogens. J. Geophys. Res. [Atmos.] 104: 21,385-21,395.
43 Djuric, D.; Damiani, M. S., Jr. (1980) On the formation of the low-level jet over Texas. Mon. Weather Rev. Vol.
44 108: 1854-1865.
45 Dodge, M. C. (1977) Combined use of modeling techniques and smog chamber data to derive ozone-precursor
46 relationships. In: Dimitriades, B., ed. International conference on photochemical oxidant pollution and its
47 control - proceedings: volume II; September 1976; Raleigh, NC. Research Triangle Park, NC: U.S.
48 Environmental Protection Agency, Environmental Sciences Research Laboratory; pp. 881-889; report no.
49 EPA-600/3-77-001b. Available from: NTIS, Springfield, VA; PB-264233.
50 Dodge, M. C. (1977) Effect of selected parameters on predictions of a photochemical model. Research Triangle
51 Park, NC: U.S. Environmental Protection Agency, Environmental Sciences Research Laboratory; EPA report
52 no. EPA-600/3-77-048. Available from: NTIS, Springfield, VA; PB-269858.
53 Domine, F.; Shepson, P. B. (2002) Air-snow interactions and atmospheric chemistry. Science (Washington, DC)
54 297: 1506-1510.
55 Dommen, J.; Prevot, A. S. H.; Hering, A. M.; Staffelbach, T.; Kok, G. L.; Schilawsky, R. D. (1999) Photochemical
56 production and aging of an urban air mass. J. Geophys. Res. (Atmos.) 104: 5493-5506.
January 2005 AX2-162 DRAFT-DO NOT QUOTE OR CITE
-------
1 Dudhia, J. (1993) A nonhydrostatic version of the Perm State-NCAR mesoscale model: validation tests and
2 simulation of an Atlantic cyclone and cold front. Mon. Weather Rev. 121: 1493-1513.
3 Dumdei, B. E.; O'Brien, R. J. (1984) Toluene's degradation products under simulated atmospheric conditions. Nature
4 (London) 311: 248-250.
5 Dumdei, B. E.; Kenny, D. V.; Shepson, P. B.; Kleindienst, T. E.; Nero, C. M; Cupitt, L. T.; Claxton, L. D. (1988)
6 MS/MS analysis of the products of toluene photooxidation and measurement of their mutagenic activity.
7 Environ. Sci. Technol. 22: 1493-1498.
8 Eckhardt, S.; Stohl, A.; Beirle, S.; Spichtinger, N.; James, P.; Forster, C.; Junker, C.; Wagner, T.; Platt, U.; Jennings,
9 S. G. (2003) The North Atlantic oscillation controls air pollution transport to the Arctic. Atmos. Chem. Phys.
10 3: 1769-1778.
11 Eder, B. K.; Davis, J. M.; Bloomfield, P. (1993) A characterization of the spatiotemporal variability of non-urban
12 ozone concentrations over the Eastern United States. Atmos. Environ. 27A: 2645-2668.
13 Eder, B. K.; Davis, J. M.; Bloomfield, P. (1994) An automated classification scheme designed to better elucidate the
14 dependence of ozone on meteorology. J. Appl. Meteorol. 33: 1182-1199.
15 Edney, E. O.; Driscoll, D. J.; Weathers, W. S.; Kleindienst, T. E.; Conver, T. S.; Mclver, C. D.; Li, W. (2001)
16 Formation of polyketones in irradiated toluene/propylene/NOx/air mixtures. Aerosol Sci. Technol.
17 35:998-1008.
18 Ehhalt, D. H.; Dorn, H.-P.; Poppe, D. (1991) The chemistry of the hydroxyl radical in the troposphere. In: Last,
19 F. T.; Watling, R., eds. Acidic deposition: its nature and impacts. Proceedings of the international
20 symposium; September 1990; Glasgow, United Kingdom. Proc. R. Soc. Edinburgh, Sect. B: Biol. Sci.
21 97: 17-34.
22 Eisele, F. L.; Mount, G. H.; Tanner, D.; Jefferson, A.; Shelter, R.; Harder, J. W.; Williams, E. J. (1997)
23 Understanding the production and interconversion of the hydroxyl radical during the tropospheric OH
24 photochemistry experiment. J. Geophys. Res. 102: 6457-6465.
25 Emmons, L. K.; Carroll, M. A.; Hauglustaine, D. A.; Brasseur, G. P.; Atherton, C.; Penner, J.; Sillman, S.; Levy, H.,
26 II; Rohrer, F.; Wauben, W. M. F.; van Velthoven, P. F. J.; Wang, Y.; Jacob, D.; Bakwin, P.; Dickerson, R.;
27 Doddridge, B.; Gerbig, C.; Honrath, R.; Htibler, G.; Jaffe, D.; Kondo, Y.; Munger, J. W.; Torres, A.;
28 Volz-Thomas, A. (1997) Climatologies of NOX and NOy: a comparison of data and models. Atmos. Environ.
29 31: 1837-1850.
30 Emmons, L. K.; Hauglustaine, D. A.; Mtiller, J.-F.; Carroll, M. A.; Brasseur, G. P.; Brunner, D.; Staehelin, J.;
31 Thouret, V.; Marenco, A. (2000) Data composites of airborne observations of tropospheric ozone and its
32 precursors. J. Geophys. Res. [Atmos.] 105: 20,497-20,538.
33 Erickson, D. J.; Seuzaret, C.; Keene, W. C.; Gong, S.-L. (1999) A general circulation model based calculation of
34 HC1 and CINO2 production from sea-salt dechlorination: reactive chlorine emissions inventory. J. Geophys.
35 Res. [Atmos.] 104: 8347-8372.
36 Fast, J. D.; Zhong, S. Y. (1998) Meteorological factors associated with inhomogeneous ozone concentrations within
37 the Mexico City basin. J. Geophys. Res. (Atmos.) 103: 18,927-18,946.
38 Fast, J. D.; Doran, J.; Shaw, W. J.; Coulter, R. L.; Martin, T. J. (2000) The evolution of the boundary layer and its
39 effect on air chemistry in the Phoenix area. J. Geophys. Res. [Atmos.] 105: 22,833-22,848.
40 Fast, J. D.; Zaveri, R. A.; Bian, R. X.; Chapman, E. G.; Easter, R. C. (2002) Effect of regional-scale transport on
41 oxidants in the vicinity of Philadelphia during the 1999NE-OPS Field Campaign. J. Geophys. Res. [Atmos.]
42 107(D16): 10.1029/2001JD000980.
43 Faust, B. C.; Anastasio, C.; Allen, J. M.; Arakaki, T. (1993) Aqueous-phase photochemical formation of peroxides
44 in authentic cloud and fog waters. Science 260: 73-75.
45 Faust, B. C. (1994) A review of the photochemical redox reactions of iron(III) species in atmospheric, oceanic, and
46 surface waters: influences on geochemical cycles and oxidant formation. In: Helz, G. R.; Zepp, R. G.; Crosby,
47 D. G., eds. Aquatic and Surface Photochemistry. Boca Raton, FL: Lewis Publishers; pp. 3-37.
48 Faust, B. C. (1994) Photochemistry of clouds, fogs, and aerosols. Environ. Sci. Technol. 28: 217A-222A.
49 Fehsenfeld, F. C.; Dickerson, R. R.; Htibler, G.; Luke, W. T.; Nunnermacker, L. J.; Williams, E. J.; Roberts, J. M.;
50 Calvert, J. G.; Curran, C. M.; Delany, A. C.; Eubank, C. S.; Fahey, D. W.; Fried, A.; Gandrud, B. W.;
51 Langford, A. O.; Murphy, P. C.; Norton, R. B.; Pickering, K. E.; Ridley, B. A. (1987) A ground-based
52 intercomparison of NO, NOX, and NOy measurement techniques. J. Geophys. Res. [Atmos.]
53 92: 14,710-14,722.
54 Fehsenfeld, F.; Calvert, J.; Fall, R.; Goldan, P.; Guenther, A. B.; Hewitt, C. N.; Lamb, B.; Liu, S.; Trainer, M.;
55 Westberg, H.; Zimmerman, P. (1992a) Emissions of volatile organic compounds from vegetation and the
56 implications for atmospheric chemistry. Global Biogeochem. Cycles 6: 389-430.
January 2005 AX2-163 DRAFT-DO NOT QUOTE OR CITE
-------
1 Fehsenfeld, F. C.; Trainer, M; Parrish, D. D.; Volz-Thomas, A.; Penkett, S. (1996a) North Atlantic Regional
2 Experiment (NARE) 1993 summer intensive: foreword. J. Geophys. Res. [Atmos.] 101: 28,869-28,875.
3 Fehsenfeld, F. C.; Damn, P.; Leaitch, W. R.; Trainer, M.; Parrish, D. D.; Htibler, G. (1996b) Transport and
4 processing of O3 and O3 precursors over the North Atlantic: an overview of the 1993 North Atlantic Regional
5 Experiment (NARE) summer intensive. J. Geophys. Res. [Atmos.] 101(D22): 28877-28891.
6 Fehsenfeld, F. C.; Huey, L. G.; Sueper, D. T.; Norton, R. B.; Williams, E. J.; Eisele, F. L.; Mauldin, R. L., Ill;
7 Tanner, D. J. (1998) Ground-based intercomparison of nitric acid measurement techniques. J. Geophys. Res.
8 [Atmos.] 103(03): 3343-3353.
9 Ferman, M. A.; Monson, P. R. (1978) Comparison of rural and urban air quality. Presented at: 71st annual meeting
10 of the Air Pollution Control Association; June; Houston, TX. Pittsburgh, PA: Air Pollution Control
11 Association; paper no. 78-10.4.
12 Pick, J.; Pommer, L.; Nilsson, C.; Andersson, B. (2003) Effect of OH radicals, relative humidity, and time on the
13 composition of the products formed in the ozonolysis of a-pinene. Atmos. Environ. 37: 4087-4096.
14 Finlayson-Pitts, B. J. (1993) Comment on "Indications of photochemical histories of Pacific air masses from
15 measurements of atmospheric trace species at Point Arena, California" by D. D. Parrish et al. J. Geophys.
16 Res. [Atmos.] 98(D8): 14,991-14,993.
17 Finlayson-Pitts, B. J.; Pitts, J. N., Jr. (1993) Atmospheric chemistry of tropospheric ozone formation: scientific and
18 regulatory implications. Air Waste 43: 1091-1100.
19 Finlayson-Pitts, B. J.; Pitts, J. N., Jr. (2000) Chemistry of the upper and lower atmosphere theory, experiments and
20 applications. San Diego, CA: Academic Press.
21 Fiore, A. M.; Jacob, D. J.; Bey, L; Yantosca, R. M.; Field, B. D.; Fusco, A. C.; Wilkinson, J. G. (2002) Background
22 ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys.
23 Res. (Atmos.) 107(D15): 10.1029/2001JD000982.
24 Fiore, A. M.; Jacob, D. J.; Mathur, R.; Martin, R. V. (2003) Application of empirical orthogonal functions to
25 evaluate ozone simulations with regional and global models. J. Geophys. Res. [Atmos.]: in press.
26 Fiore, A.; Jacob, D. J.; Liu, H.; Yantosca, R. M.; Fairlie, T. D.; Li, Q. (2003) Variability in surface ozone
27 background over the United States: implications for air quality policy. J. Geophys. Res. (Atmos.): submitted.
28 Fishman, J.; Wozniak, A. E.; Creilson, J. K. (2003) Global distribution of tropospheric ozone from satellite
29 measurements using the empirically corrected tropospheric ozone residual technique: identification of the
30 regional aspects of air pollution. Atmos. Chem. Phys. 3: 893-907.
31 Forstner, H. J. L.; Flagan, R. C.; Seinfeld, J. H. (1997) Secondary organic aerosol from the photooxidation of
32 aromatic hydrocarbons: molecular composition. Environ. Sci. Technol. 31: 1345-1358.
33 Fortuin, J. P. F.; Kelder, H. (1998) An ozone climatology based on ozonesonde and satellite measurements.
34 J. Geophys. Res. 103: 31,709-31,734.
35 Fox-Rabinovitz, M. S.; Takacs, L. L.; Govindaraju, R. C. (2002) A variable-resolution stretched-grid general
3 6 circulation model and data assimilation system with multiple areas of interest: studying the anomalous
37 regional climate events of 1998. J. Geophys. Res. [Atmos.] 107(D24): 10.1029/2002JD002177.
3 8 Fried, A.; Hodgeson, J. (1982) Laser photoacoustic detection of nitrogen dioxide in the gas-phase titration of nitric
39 oxide with ozone. Anal. Chem. 54: 278-282.
40 Fuglestvedt, J. S.; Berntsen, T. K.; Isaksen, I. S. A.; Mao, H.; Liang, X.-Z.; Wang, W.-C. (1999) Climatic forcing of
41 nitrogen oxides through changes in tropospheric ozone and methane; global 3D model studies. Atmos.
42 Environ. 33: 961-978.
43 Fujita, E. M.; Croes, B. E.; Bennett, C. L.; Lawson, D. R.; Lurmann, F. W.; Main, H. H. (1992) Comparison of
44 emission inventory and ambient concentration ratios of CO, NMOG, and NOX in California's south coast air
45 basin. J. Air Waste Manage. Assoc. 42: 264-276.
46 Fujita etal., 1995.
47 Furutani, H.; Akimoto, H. (2002) Development and characterization of a fast measurement system for gas-phase
48 nitric acid with a chemical ionization mass spectrometer in the marine boundary layer. J. Geophys. Res.
49 (Atmos.) 107(D1-D2): 10.1029/2000JD000269.
50 Fusco, A. C.; Logan, J. A. (2003) Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
51 midlatitudes with the GEOS-CHEM model. J. Geophys. Res. (Atmos.) 108: 10.1029/2002JD002742.
52 Gaffney, J. S.; Marley, N. A.; Drayton, P. J.; Doskey, P. V.; Kotamarthi, V. R.; Cunningham, M. M.; Baird, J. C.;
53 Dintaman, J.; Hart, H. L. (2002) Field observations of regional and urban impacts on NO2, ozone, UVB, and
54 nitrate radical production rates in the Phoenix air basin. Atmos. Environ. 36: 825-833.
55 Galbally, I. E.; Bentley, S. T.; Meyer, C. P. (2000) Mid-latitude marine boundary-layer ozone destruction at visible
56 sunrise observed at Cape Grim, Tasmania, 41°S. Geophys. Res. Lett. 27: 3841-3844.
January 2005 AX2-164 DRAFT-DO NOT QUOTE OR CITE
-------
1 Galloway, J. N.; Whelpdale, D. M. (1980) An atmospheric sulfur budget for eastern North America. Atmos.
2 Environ. 14: 409-417.
3 Galmarini, S.; Duynkerke, P. G.; de Arellano, J. V. G. (1997) Evolution of nitrogen oxide chemistry in the nocturnal
4 boundary layer. J. Appl. Meteorol. 36: 943-957.
5 Garnica, R. M.; Appel, M. F.; Eagan, L. (2000) A REMPI method for the ultrasensitive detection of NO and NO2
6 using atmospheric pressure laser ionization mass spectrometry. Anal. Chem. 72: 5639-5646.
7 Gauss, M.; Myhre, G.; Pitari, G.; Prather, M. J.; Isaksen, I. S. A.; Berntsen, T. K.; Brasseur, G. P.; Dentener, F. J.;
8 Derwent, R. G.; Hauglustaine, D. A.; Horowitz, L. W.; Jacob, D. J.; Johnson, M.; Law, K. S.; Mickley, L. J.;
9 Mtiller, J.-F.; Plantevin, P. H.; Pyle, J. A.; Rogers, H. L.; Stevenson, D. S.; Sundet, J. K.; Van Weele, M.;
10 Wild, O. (2003) Radiative forcing in the 21st century due to ozone changes in the troposphere and the lower
11 stratosphere. J. Geophys. Res. (Atmos.) 108(D9): 10.1029/2002JD002624.
12 Geiger, H.; Barnes, L; Bejan, L; Benter, T.; Spittler, M. (2003) The tropospheric degradation of isoprene: an updated
13 module for the regional atmospheric chemistry mechanism. Atmos. Environ. 37: 1503-1519.
14 Geron, C. D.; Guenther, A. B.; Pierce, T. E. (1994) An improved model for estimating emissions of volatile organic
15 compounds from forests in the eastern United States. J. Geophys. Res. [Atmos.] 99: 12,773-12,791.
16 Geron, C. D.; Nie, D.; Arnts, R. R.; Sharkey, T. D.; Singsaas, E. L.; Vanderveer, P. J.; Guenther, A.; Sickles, J. E.,
17 II; Kleindienst, T. E. (1997) Biogenic isoprene emission: model evaluation in a southeastern United States
18 bottomland deciduous forest. J. Geophys. Res. [Atmos.] 102(D15): 18,889-18,901.
19 Geron, C.; Guenther, A.; Sharkey, T.; Arnts, R.R. (2000a) Temporal variability in basal isoprene emission factor.
20 Tree Physiol. 20: 799-805.
21 Geron, C.; Rasmussen, R.; Arnts, R. R.; Guenther, A. (2000b) A review and synthesis of monoterpene speciation
22 from forests in the United States. Atmos. Environ. 34: 1761-1781.
23 Geron, C. D.; Harley, P.; Guenther, A. (2001) Isoprene emission capacity for US tree species. Atmos. Environ.
24 35: 3341-3352.
25 Gershenzon, Y. M.; Grigorieva, V. M.; Ivanov, A. V.; Remorov, R. G. (1995) O3 and OH sensitivity to
26 heterogeneous sinks of HOX and CH3O2 on aerosol particles. Faraday Discuss. 100: 83-100.
27 Gershenzon, M. Y.; Il'in, S.; Fedotov, N. G.; Gershenzon, Y. M.; Aparina, E. V.; Zelenov, V. V. (1999) The
28 mechanism of reactive NO3 uptake on dry NaX (X=C1, Br). J. Atmos. Chem. 34: 119-135.
29 Gershey, R. M. (1983) Characterization of seawater organic matter carried by bubble-generated aerosols. Limnol.
30 Oceanogr. 28:309-319.
31 Gertler, A. W.; Sagebiel, J. C.; Dippel, W. A.; O'Connor, C. M. (1999) The impact of California phase 2
32 reformulated gasoline on real-world vehicle emissions. J. Air Waste Manage. Assoc. 49: 1339-1346.
33 Gery, M. W.; Fox, D. L.; Jeffries, H. E.; Stockburger, L.; Weathers, W. S. (1985) A continuous stirred tank reactor
34 investigation of the gas-phase reaction of hydroxyl radicals and toluene. Int. J. Chem. Kinet. 17: 931-955.
35 Gery, M. W.; Whitten, G. Z.; Killus, J. P. (1988) Development and testing of the CMB-IV for urban and regional
36 modeling. Research Triangle Park, NC: U.S. Environmental Protection Agency, Atmospheric Sciences
37 Research Laboratory; report no. EPA-600/3-88-012. Available from: NTIS, Springfield, VA; PB88-180039.
38 Gery, M. W.; Whitten, G. Z.; Killus, J. P.; Dodge, M. C. (1989) A photochemical kinetics mechanism for urban and
39 regional scale computer modeling. J. Geophys. Res. [Atmos.] 94: 12,925-12,956.
40 Gettelman, A.; Holton, J. R.; Rosenlof, K. H. (1997) Mass fluxes of O3, CH4, N2O, and CF2C12 in the lower
41 stratosphere calculated from observational data. J. Geophys. Res. 102(D15): 19,149-19,159.
42 Geyer, A.; Alicke, B.; Konrad, S.; Schmitz, T.; Stutz, J.; Platt, U. (2001) Chemistry and oxidation capacity of the
43 nitrate radical in the continental boundary layer near Berlin. J. Geophys. Res. (Atmos.) 106: 8013-8025.
44 Geyer, A.; Platt, U. (2002) Temperature dependence of the NO3 loss frequency: A new indicator for the contribution
45 of NO3 to the oxidation of monoterpenes and NOX removal in the atmosphere. J. Geophys. Res. (Atmos.)
46 107(020): 10.1029/2001JD001215.
47 Gidel, L. T. (1983) Cumulus cloud transport of transient tracers. J. Geophys. Res. [Atmos.] 88: 6587-6599.
48 Gillani, N. V.; Pleim, J. E. (1996) Sub-grid-scale features of anthropogenic emissions of NOX and VOC in the
49 context of regional Eulerian models. Atmos. Environ. 30: 2043-2059.
50 Gillani, N. V.; Meagher, J. F.; Valente, R. J.; Imhoff, R. E.; Tanner, R. L.; Luria, M. (1998) Relative production of
51 ozone and nitrates in urban and rural power plant plumes. 1. Composite results based on data from 10 field
52 measurement days. J. Geophys. Res. [Atmos.] 103: 22,593-22,615.
53 Gilpin, T.; Apel, E.; Fried, A.; et al. (1997) Intercomparison of six ambient (CH2O) measurement techniques.
54 J. Geophys. Res. (Atmos.) 102(D17): 21,161-21,188.
55 Goldan, P. D.; Kuster, W. C.; Albritton, D. L.; Fehsenfeld, F. C.; Connell, P. S.; Norton, R. B.; Huebert, B. J. (1983)
5 6 Calibration and tests of the filter-collection method for measuring clean-air, ambient levels of nitric acid.
57 Atmos. Environ. 17: 1355-1364.
January 2005 AX2-165 DRAFT-DO NOT QUOTE OR CITE
-------
1 Goldan, P. D.; Kuster, W. C.; Fehsenfeld, F. C. (1997) Non-methane hydrocarbon measurements during the
2 tropospheric OH photochemistry experiment. J. Geophys. Res. (Atmos.) 102: 6315-6324.
3 Goldan, P. D.; Trainer, M; Kuster, W. C.; Parrish, D. D.; Carpenter, J.; Roberts, J. M; Yee, J. E.; Fehsenfeld, F. C.
4 (1995) Measurements of hydrocarbons, oxygenated hydrocarbons, carbon monoxide, and nitrogen oxides in
5 an urban basin in Colorado: implications for emission inventories. J. Geophys. Res. [Atmos.] 100:
6 22,771-22,783.
7 Goldan, P. D.; Parrish, D. D.; Kuster, W. C.; Trainer, M.; McKeen, S. A.; Holloway, J.; Jobson, B. T.; Sueper, D. T.;
8 Fehsenfeld, F. C. (2000) Airborne measurements of isoprene, CO, and anthropogenic hydrocarbons and their
9 implications. J. Geophys. Res. [Atmos.] 105: 9091-9105.
10 Golden, D. M.; Smith, G. P. (2000) Reaction of OH + NO2 + M: a new view. J. Phys. Chem. A 104: 3991-3997.
11 Goldstein, A. H.; Schade, G. W. (2000) Quantifying biogenic and anthropogenic contributions to acetone mixing
12 ratiosin a rural environment. Atmos. Environ. 34: 4997-5006.
13 Gong, S. L.; Barrie, L. A.; Blanchet, J.-P. (1997) Modeling sea-salt aerosols in the environment: 1. model
14 development. J. Geophys. Res. 102: 3805-3818.
15 Gong, W.; Mickle, R. E.; Bottenheim, F. F.; Beauchamp, S.; Waugh, D. (2000) Marine/coastal boundary layer and
16 vertical structure of ozone observed at a coastal site in Nova Scotia during the 1996 NARSTO-CE field
17 campaign. Atmos. Environ. 34: 4139-4154.
18 Gorzelska, K.; Galloway, J. N.; Watterson, K.; Keene, W. C. (1992) Water-soluble primary amine compounds in
19 rural continental precipitation. Atmos. Environ. 26: 1005-1018.
20 Goult, J. (1938) Vents en alititude a Fort Lamy (Tach). Ann. Phys. (Paris, France) 5: 373-379.
21 Grassi, B.; Pitari, G.; Visconti, G. (1997) Results of a chemical-transport model with interactive aerosol
22 microphysics. In: Bojkov, R.; Visconti, G. Proceedings of the XVIII Quadrennial Ozone Symposium;
23 September, 1996; L'Aquila, Italy.
24 Greenhut, G. K. (1986) Transport of ozone between boundary layer and cloud layer by cumulus clouds. J. Geophys.
25 Res. [Atmos.] 91: 8613-8622.
26 Greenhut, G. K.; Ching, J. K. S.; Pearson, R., Jr.; Repoff, T. P. (1984) Transport of ozone by turbulence and clouds
27 in an urban boundary layer. J. Geophys. Res. [Atmos.] 89: 4757-4766.
28 Gregg, J. W.; Jones, C. G.; Dawson, T. E. (2003) Urbanization effects on tree growth in the vicinity of New York
29 City [letter]. Nature 424: 183-187.
30 Gregory, G. L.; Hoell, J. M., Jr.; Torres, A. L.; Carroll, M. A.; Ridley, B. A.; Rodgers, M. O.; Bradshaw, J.;
31 Sandholm, S.; Davis, D. D. (1990) An intercomparison of airborne nitric oxide measurements: a second
32 opportunity. J. Geophys. Res. [Atmos.] 95: 10,129-10,138.
33 Grell, G. A.; Dudhia, J.; Stauffer, D. R. (1994) Description of the fifth-generation Perm State/NCAR mesoscale
34 model (MM5). Boulder, CO: National Center for Atmospheric Research, Mesoscale and Microscale
3 5 Meteorology Division; report no. NCAR/TN-398+STR. Available from: NTIS, Springfield, VA;
36 PB95-206348.
37 Grell, G. A.; Emeis, S.; Stockwell, W. R.; Schoenemeyer, T.; Forkel, R.; Michalakes, J.; Knoche, R.; Seidl, W.
3 8 (2000) Application of a multiscale, coupled MM5/chemistry model to the complex terrain of the VOTALP
39 valley campaign. Atmos. Environ. 34: 1435-1453.
40 Grobner, J.; Kerr, J. B. (2001) Ground-based determination of the spectral ultraviolet extraterrestrial solar irradiance:
41 providing a link between space-based and ground-based solar UV measurements. J. Geophys. Res. (Atmos.)
42 106(D7): 7211-7217.
43 Grosjean, E.; Grosjean, D. (1994) Rate constants for the gas-phase reactions of ozone with unsaturated aliphatic
44 alcohols. Int. J. Chem. Kinetics 26: 1185-1191.
45 Grosjean, D.; Williams, E. L., II; Grosjean, E.; Novakov, T. (1994) Evolved gas analysis of secondary organic
46 aerosols. Aerosol Sci. Technol. 21: 306-324.
47 Gross, A.; Stockwell, W. R. (2003) Comparison of the EMEP, RADM2 and RACM Mechanisms. J. Atmos. Chem.
48 44: 151-170.
49 Guenther, A.; Hewitt, C. N.; Erickson, D.; Fall, R.; Geron, C.; Graedel, T.; Harley, P.; Klinger, L.; Lerdau, M.;
50 McKay, W. A.; Pierce, T.; Scholes, B.; Steinbrecher, R.; Tallamraju, R.; Taylor, J.; Zimmerman, P. (1995) A
51 global model of natural volatile organic compound emissions J. Geophys. Res. (Atmos.) 100(D5): 8873-8892.
52 Guenther, A.; Baugh, W.; Davis, K.; Hampton, G.; Harley, P.; Klinger, L.; Vierling, L.; Zimmerman, P.; Allwine,
53 E.; Dilts, S.; Lamb, B.; Westberg, H.; Baldocchi, D.; Geron, C.; Pierce, T. (1996) Isoprene fluxes measured
54 by enclosure, relaxed eddy accumulation, surface layer gradient, mixed layer gradient, and mixed layer mass
55 balance techniques. J. Geophys. Res. (Atmos.) 101: 18,555-18,567.
January 2005 AX2-166 DRAFT-DO NOT QUOTE OR CITE
-------
1 Guenther, F. R.; Dorko, W. D.; Miller, W. R.; Rhoderick, G. C. (1996) The NIST traceable reference material
2 program for gas standards. Washington, DC: U.S. Department of Commerce, National Institute of Standards
3 and Technology; NIST special publication 260-126.
4 Guenther, A.; Geron, C.; Pierce, T.; Lamb, B.; Harley, P.; Fall, R. (2000) Natural emissions of non-methane volatile
5 organic compounds, carbon monoxide, and oxides of nitrogen from North America. Atmos. Environ.
6 34:2205-2230.
7 Hadjinicolaou, P.; Pyle, J. A. (2004) J. Atmos. Chem. 47: 25-43.
8 Hallock-Waters, K. A.; Doddridge, B. G.; Dickerson, R. R.; Spitzer, S.; Ray, J. D. (1999) Carbon monoxide in the
9 U.S. mid-Atlantic troposphere: evidence for a decreasing trend. Geophys. Res. Lett. 26: 2861-2864.
10 Hallquist, M; Stewart, D. J.; Stephenson, S. K.; Cox, R. A. (2003) Hydrolysis of N2O5 on sub-micron sulfate
11 aerosols. Phys. Chem. Chem. Phys. 5: 3453-3463.
12 Hameed, S.; Pinto, J. P.; Stewart, R. W. (1979) Sensitivity of the predicted CO-OH-CH4 perturbation to tropospheric
13 NO-j concentrations. J. Geophys. Res. C: Oceans Atmos. 84: 763-768.
14 Hanisch, F.; Crowley, J. N. (2002) Ozone decomposition on Saharan dust: an experimental investigation. Atmos.
15 Chem. Phys. Disc. 2: 1809-1845.
16 Hanna, A. F.; Mathur, R.; Pinto, J.; Jang, C. (2001) Seasonal modeling of the export of pollutants from North
17 America using the Multi Scale Air Quality Simulation Pattern (MAQSIP). Presented at: Workshop on
18 photooxidants, particles, and haze across the Arctic and North Atlantic: observations and models; June;
19 New York, NY. Palisades, NY: Center for International Earth Science Information Network (CIESIN).
20 Available: http://www.ciesin.org/pph/uploads/H_A_Seasonal_Modeling_PDF.pdf (29 December 2003).
21 Hannegan, B.; Olsen, S.; Prather, M.; Zhu, X.; Rind, D.; Lerner, J. (1998) The dry stratosphere: a limit on cometary
22 water influx. Geophys. Res. Lett. 25: 1649-1652.
23 Hansel, A.; Jordan, A.; Taucher, J. (1995) Analysis of volatile organic compounds using
24 proton-transfer-reaction-mass spectrometry (PTR-MS) in the ppb range. In: Becker, K. H.; Carr, W. E.;
25 Kunhardt, E. E., eds. Proceedings of the XXII International conference on phenomena in ionized gases;
26 July-August; Hoboken, NJ. Hoboken, NJ: Stevens Institute of Technology; pp. 181-182.
27 Hansel, A.; Jordan, A.; Holzinger, R.; Prazeller, P.; Vogel, W.; Lindinger, W. (1995) Proton transfer reaction mass
28 spectrometry: on-line trace gas analysis at the ppb level. Int. J. Mass Spectrom. Ion Proc. 149/150: 609-619.
29 Hansen, K.; Draaijers, G. P. J.; Ivens, W. P. M. F.; Gundersen, P.; vanLeeuwen, N. F. M. (1994) Concentration
30 variations in rain and canopy throughfall collected sequentially during individual rain events. Atmos. Environ.
31 28:3195-3205.
32 Hansen, J.; Sato, M.; Ruedy, R.; Lacis, A.; Asamoah, K.; Beckford, K.; Borenstein, S.; Brown, E.; Cairns, B.;
33 Carlson, B.; Curran, B.; de Castro, S.; Druyan, L.; Etwarrow, P.; Ferede, T.; Fox, M.; Gaffen, D.; Glascoe, J.;
34 Gordon, H.; Hollandsworth, S.; Jiang, X.; Johnson, C.; Lawrence, N.; Lean, J.; Lerner, J.; Lo, K.; Logan, J.;
35 Luckett, A.; McCormick, M. P.; McPeters, R.; Miller, R.; Minnis, P.; Ramberran, L; Russell, G.; Russell, P.;
36 Stone, P.; Tegen, I.; Thomas, S.; Thomason, L.; Thompson, A.; et al. (1997) Forcings and chaos in
37 interannual to decadal climate change. J. Geophys. Res. [Atmos.] 102: 25,679-25,720.
38 Kara, K.; Osada, K.; Hayashi, M.; Matsunaga, K.; Shibata, T.; Iwasaka, Y.; Furuya, K. (1999) Fractionationof
39 inorganic nitrates in winter Arctic troposphere: coarse aerosol particles containing inorganic nitrates.
40 J. Geophys. Res. (Atmos.) 104: 23,671-23,679.
41 Hard, T. M.; Mehrabzadeh, A. A.; Chan, C. Y.; O'Brien, R. J. (1992) PAGE measurements of tropospheric HO with
42 measurements and model of interferences. J. Geophys. Res. [Atmos.] 97: 9795-9817.
43 Harder, J. W.; Brault, J. W.; Johnston, P. V.; Mount, G. H. (1997a) Temperature dependent NO2 cross sections at
44 high spectral resolution. J. Geophys. Res. [Atmos.] 102: 3861-3879.
45 Harder, J. W.; Williams, E. J.; Baumann, K.; Fehsenfeld, F. C. (1997b) Ground-based comparison of NO2, H2O, and
46 O3 measured by long-path and in situ techniques during the 1993 Tropospheric OH Photochemistry
47 Experiment. J. Geophys. Res. 102: 6227-6243.
48 Harley, P.; Guenther, A.; Zimmerman, P. (1996) Effects of light, temperature and canopy position on net
49 photosynthesis and isoprene emission from sweetgum (Liquidambar styraciflua) leaves. Tree Physiol.
50 16:25-32.
51 Harris, G. W.; Carter, W. P. L.; Winer, A. M.; Pitts, J. N, Jr.; Platt, U.; Perner, D. (1982) Observations of nitrous
52 acid in the Los Angeles atmosphere and implications for predictions of ozone-precursor relationships.
53 Environ. Sci. Technol. 16: 414-419.
54 Harris, J. M.; Oltmans, S. J.; Bodeker, G. E.; Stolarski, R.; Evans, R. D.; Quincy, D. M. (2003) Long-term variations
55 in total ozone derived from Dobson and satellite data. Atmos. Environ. 37: 3167-3175.
56 Harrison, R. M.; Peak, J. D.; Collins, G. M. (1996) Tropospheric cycle of nitrous acid. J. Geophys. Res. (Atmos.)
57 101: 14,429-14,439.
January 2005 AX2-167 DRAFT-DO NOT QUOTE OR CITE
-------
1 Hauglustaine, D. A.; Brasseur, G. P.; Walters, S.; Rasch, P. I; Mtiller, J.-F.; Emmons, L. K.; Carroll, M. A. (1998)
2 MOZART, a global chemical transport model for ozone and related chemical tracers 2. model results and
3 evaluation. J. Geophys. Res. (Atmos.) 103: 28291-28335.
4 Hebestreit, K.; Stutz, J.; Rosen, D.; Matveiv, V.; Peleg, M.; Luria, M.; Platt, U. (1999) DOAS measurements of
5 tropospheric bromine oxide in mid-latitudes. Science 283: 55-57.
6 Hegg, D. A.; Livingston, J.; Hobbs, P. V.; Novakov, T.; Russell, P. (1997) Chemical apportionment of aerosol
7 column optical depth off the mid-Atlantic coast of the United States. J. Geophys. Res. (Atmos.) 102:
8 25,293-25,303.
9 Heikes, B. G.; Chang, W.; Pilson, M. E. Q.; Swift, E.; Singh, H. B.; Guenther, A.; Jacob, D. J.; Field, B. D.; Fall, R.;
10 Riemer, D.; Brand, L. (2002) Atmospheric methanol budget and ocean implication. Glob. Biogeochem.
11 Cycles 16: 10.1029/2002GB001895.
12 Henry, R. C.; Lewis, C. W.; Hopke, P. K.; Williamson, H. J. (1984) Review of receptor model fundamentals. Atmos.
13 Environ. 18: 1507-1515.
14 Hering, S. V.; Lawson, D. R.; Allegrini, I.; Febo, A.; Perrino, C.; Possanzini, M.; Sickles, J. E., II; Anlauf, K. G.;
15 Wiebe, A.; Appel, B. R.; John, W.; Ondo, J.; Wall, S.; Braman, R. S.; Sutton, R.; Cass, G. R.; Solomon, P.
16 A.; Eatough, D. J.; Eatough, N. L.; Ellis, E. C.; Grosjean, D.; Hicks, B. B.; Womack, J. D.; Horrocks, J.;
17 Knapp, K. T.; Ellestad, T. G.; Paur, R. J.; Mitchell, W. J.; Pleasant, M.; Peake, E.; MacLean, A.; Pierson,
18 W. R.; Brachaczek, W.; Schiff, H. I.; Mackay, G. I.; Spicer, C. W.; (1988) The nitric acid Shootout: field
19 comparison of measurement methods. Atmos. Environ. 22: 1519-1539.
20 Herlihy, L. J.; Galloway, J. N.; Mills, A. L. (1987) Bacterial utilization of formic and acetic acid in rainwater.
21 Atmos. Environ. 21: 2397-2402.
22 Hess, P. G. (2001) Model and measurement analysis of springtime transport and chemistry of the Pacific Basin.
23 J. Geophys. Res. (Atmos.) 106: 12,689-12,717.
24 Hilsenrath, E.; Attmannspacher, W.; Bass, A.; Evans, W.; Hagemeyer, R.; Barnes, R. A.; Komhyr, W.;
25 Mauersberger, K.; Mentall, J.; Proffitt, M.; Robbins, D.; Taylor, S.; Torres, A.; Weinstock, E. (1986) Results
26 from the ballon ozonr intercomparison campaign (BOIC). J. Geophys. Res. 91: 13,137-13,152.
27 Hirsch, A. I.; Munger, J. W.; Jacob, D. J.; Horowitz, L. W.; Goldstein, A. H. (1996) Seasonal variation of the ozone
28 production efficiency per unit NOX at Harvard Forest, Massachusetts. J. Geophys. Res. [Atmos.]
29 101: 12,659-12,666.
30 Hobbs, P. V.; Locatelli, J. D.; Martin, J. E. (1996) A new conceptual model for cyclones generated in the lee of the
31 Rocky Mountains. Bull. Am. Meteorol. Soc. 77: 1169-1178.
32 Hoell, J. M.; Davis, D. D.; Jacob, D. J.; Rodgers, M. O.; Newell, R. E.; Fuelberg, H. E.; McNeal, R. J.; Raper, J. L.;
33 Bendura, R. J. (1996) Pacific exploratory mission in the tropical Pacific: PEM-Tropics A, August-September
34 1996. J. Geophys. Res. (Atmos.) 104: 5567-5583.
35 Hoell, J. M.; Davis, D. D.; Liu, S. C.; Newell, R. E.; Akimoto, H.; McNeal, R. J.; Bendura, R. J. (1997) The Pacific
36 exploratory mission A west phase B: February-March, 1994. J. Geophys. Res. [Atmos.] 102(D23):
37 28223-28239.
3 8 Hoffman, E. J.; Duce, R. A. (1976) Factors influencing the organic carbon content of marine aerosols: a laboratory
39 study. J. Geophys. Res. (Atmos.) 81: 3667-3670.
40 Hoffman, E. J.; Duce, R. A. (1977) Organic carbon in marine atmospheric paniculate matter: concentration and
41 particle size distribution. Geophys. Res. Lett. 4: 449-452.
42 Hoigne, J.; Zuo, Y.; Nowell, L. (1994) Photochemical reactions in atmospheric waters: role of dissolved iron
43 species. In: Helz, G. R.; Zepp, R. G.; Crosby, D. C. Aquatic and Surface Photochemistry. Boca Raton, FL:
44 Lewis Publishers.
45 Holton, J. R. (1967) The diurnal boundary layer wind oscillation above sloping terrain. Tellus 19: 199-205.
46 Holton, J. R.; Haynes, P. H.; Mclntyre, M. E.; Douglass, A. R.; Rood, R. B.; Pfister, L. (1995)
47 Stratosphere-troposphere exchange. Rev. Geophys. 33: 403-439.
48 Honninger, G. (1999) Referenzspektren reaktiver halogenverbindungen fur DOAS-messungen. Heidelberg,
49 Germany: Ruprecht Karls University.
50 Honrath, R. E.; Guo, S.; Peterson, M. C. (2000) Photochemical production of gas phase NOX from ice crystal NO3".
51 J. Geophys. Res. (Atmos.) 105: 24,183-24,190.
52 Hopke, P. K. (1985) Receptor modeling in environmental chemistry. New York, NY: John Wiley & Sons.
53 Horowitz, L. W.; Liang, J.; Gardner, G. M.; Jacob, D. J . (1998) Export of reactive nitrogen from North America
54 during summertime: sensitivity to hydrocarbon chemistry. J. Geophys. Res. 103(D11): 13451-13476.
55 Hoskins, B. J. (1972) Non-boussinesq effects and further development in a model of upper tropospheric
56 frontogenesis. Q. J. Roy. Meteorol. Soc. 98: 532-541.
January 2005 AX2-168 DRAFT-DO NOT QUOTE OR CITE
-------
1 Houweling, S.; Dentener, F.; Lelieveld, J.; Walter, B.; Dlugokencky, E. (2000) The modeling of tropospheric
2 methane: how well can point measurements be reproduced by a global model? J. Geophys. Res. (Atmos.)
3 105:8981-9002.
4 Htibler, G.; Alvarez, R., II; Daum, P.; Dennis, R.; Gillani, N.; Kleinman, L.; Luke, W.; Meagher, J.; Rider, D.;
5 Trainer, M.; Valente, R. (1998) An overview of the airborne activities during the Southern Oxidants Study
6 (SOS) 1995 Nashville/Middle Tennessee ozone study. J. Geophys. Res. [Atmos.] 103: 22,245-22,259.
7 Hudman, R. C; Jacob, D. J.; Cooper, O. C; Evans, M. J.; Heald, C. L.; Park, R. J.; Fehsenfeld, F.; Flocke, F.;
8 Holloway, J.; Htibler, G.; Kita, K.; Koike, M.; Kondo, Y.; Neuman, A.; Nowak, J.; Oltmans, S.; Parrish, D.;
9 Roberts, J. M.; Ryerson, T. (2004) Ozone production in transpacific Asian pollution plumes and implications
10 for ozone air quality in California. J. Geophys. Res. (Atmos.) 109(D23): 10.1029/2004JD004974.
11 Huebert, B. J.; Zhuang, L.; Howell, S.; Noone, K.; Noone, B. (1996) Sulfate, nitrate, methanesulfonate, chloride,
12 ammonium, and sodium measurements from ship, island, and aircraft during the Atlantic Stratocumulus
13 Transition Experiment/Marine Aerosol Gas Exchange. J. Geophys. Res. (Atmos.) 101: 4413-4423.
14 Huey, L. G.; Dunlea, E. J.; Lovejoy, E. R.; Hanson, D. R.; Norton, R. B.; Fehsenfeld, F. C.; Howard, C. J. (1998)
15 Fast time response measurements of HNO3 in air with a chemical ionization mass spectrometer. J. Geophys.
16 Res. 103(03): 3355-3360.
17 Hunsaker, C.; Graham, R.; Turner, R. S.; Ringold, P. L.; Holdren, G. R., Jr.; Strickland, T. C. (1993) A national
18 critical loads framework for atmospheric deposition effects assessment: II. Defining assessment end points,
19 indicators, and functional subregions. Environ. Manage. 17: 335-341.
20 Huntrieser, H.; Feigl, C.; Schlager, H.; Schroder, F.; Gerbig, C.; Van Velthoven, P.; Flatey, F.; Thery, C.;
21 Petzold, A.; Holler, H.; Schumann, U. (2002) Airborne measurements of NOX, tracer species, and small
22 particles during the European Lightning Nitrogen Oxides Experiment. J. Geophys. Res.
23 107(011): 10.1029/2000JD000209.
24 Husain, L.; Coffey, P. E.; Meyers, R. E.; Cederwall, R. T. (1977) Ozone transport from stratosphere to troposphere.
25 Geophys. Res. Lett. 4: 363-365.
26 Husar, R. B.; Renard, W. P. (1998) Ozone as a function of local wind speed and direction: Evidence of local and
27 regional transport. Presented at: 91st annual meeting and exhibition of the Air & Waste Management
28 Association; June; San Diego, CA. Pittsburgh, PA: Air & Waste Management Association; online paper no.
29 98-A922. Available: http://capita.wustl.edu/capita/CapitaReports/REPORTSl.HTML (13 November 2003).
30 Imhoff, R. E.; Valente, R.; Meagher, J. F.; Luria, M. (1995) The production of O3 in an urban plume - airborne
31 sampling of the Atlanta urban plume. Atmos. Environ. 29: 2349-2358.
32 Intergovernmental Panel on Climate Change (IPCC). (1995) Climate change 1994: radiative forcing of climate
33 change and an evaluation of the IPCC IS92 emission scenarios. Cambridge, United Kingdom: University
34 Press.
35 Intergovernmental Panel on Climate Change (IPCC). (1996) Climate change 1995: the science of climate change:
3 6 contribution of working group I to the second assessment of the Intergovernmental Panel on Climate Change.
37 Cambridge, United Kingdom: Cambridge University Press.
38 Intergovernmental Panel on Climate Change (IPCC). (2001) Climate change 2001: the scientific basis. Contribution
39 of working group I to the third assessment report of the Intergovernmental Panel on Climate Change.
40 Cambridge, United Kingdom: Cambridge University Press.
41 Jacob, D. J. (2000) Heterogeneous chemistry and tropospheric ozone. Atmos. Environ. 34: 2131-2159.
42 Jacob, D. J.; Logan, J. A.; Gardner, G. M.; Yevich, R. M.; Spivakovsky, C. M.; Wofsy, S. C.; Sillman, S.; Prather,
43 M. J. (1993) Factors regulating ozone over the United States and its export to the global atmosphere.
44 J. Geophys. Res. [Atmos.] 98: 14,817-14,826.
45 Jacob, D. J.; Horowitz, L. W.; Munger, J. W.; Heikes, B. G.; Dickerson, R. R.; Artz, R. S.; Keene, W. C. (1995)
46 Seasonal transition from NOX- to hydrocarbon-limited conditions for ozone production over the eastern
47 United States in September. J. Geophys. Res. 100: 9315-9324.
48 Jacob, D. J.; Prather, M. J.; Rasch, P. J.; Shia, R.-L.; Balkanski, Y. J.; Beagley, S. R.; Bergman, D. J.; Blackshear,
49 W. T.; Brown, M.; Chiba, M.; Chipperfield, M.; De Grandpre, J.; Dignon, J. E.; Feichter, J.; Genthon, C.;
50 Grose, W.L.; Kasibhatla, P. S.; Kohler, I.; Kritz, M. A.; Law, K.; Penner, J. E.; Ramonet, M.; Reeves, C. E.;
51 Rotman, D. A.; Stockwell, D. Z.; VanVelthoven, P. F. J.; Verver, G.; Wild, O.; Yang, H.; Zimmermann, P.
52 (1997) Evaluation and intercomparison of global atmospheric transport models using 222Rn and other
53 short-lived tracers. J. Geophys. Res. [Atmos.] 102: 5953-5970.
54 Jacobi, C.; Becker, C.; Kratzsch, T.; Siemer, A.; Roth, R. (1994) Experimental investigations of nocturnal low-level
55 jets with some remarks on flight safety. Contrib. Atmos. Phys. 67: 71-82.
56 Jacobson, M. Z. (1997) Development and application of a new air pollution modeling system - Part III.
57 aerosol-phase simulations. Atmos. Environ. 31: 587-608.
January 2005 AX2-169 DRAFT-DO NOT QUOTE OR CITE
-------
1 Jacobson, M. Z. (1998) Studying the effects of aerosols on vertical photolysis rate coefficient and temperature
2 profiles over an urban airshed. J. Geophys. Res. [Atmos.] 103: 10,593-10,604.
3 Jacobson, M. Z. (1999) Isolating nitrated and aromatic aerosols and nitrated aromatic gases as sources of ultraviolet
4 light absorption. J. Geophys. Res. (Atmos.) 104: 3527-3542.
5 Jacobson, M. Z.; Lu, R.; Turco, R. P.; Toon, O. P. (1996) Development and application of a new air pollution
6 modeling system-Part I: gas-phase simulations. Atmos. Environ. 30: 1939-1963.
7 Jacobson, M. C.; Hansson, H.-C.; Noone, K. J.; Charlson, R. J. (2000) Organic atmospheric aerosols: review and
8 state of the science. Rev. Geophys. 38: 267-294.
9 Jaegle, L.; Jacob, D. J.; Wennberg, P. O.; Spivakovsky, C. M.; Hanisco, T. F.; Lanzendorf, E. J.; Hintsa, E. J.;
10 Fahey, D. W.; Keim, E. R.; Proffitt, M. H.; Atlas, E. L.; Flocke, F.; Schauffler, S.; McElroy, C. T.;
11 Midwinter, C.; Pfister, L.; Wilson, J. C. (1997) Observed OH and HO2 in the upper troposphere suggests a
12 major source from convective injection of peroxides. Geophys. Res. Lett. 24: 3181-3184.
13 Jaegle, L.; Jacob, D. J.; Brune, W. H.; Faloona, I. C.; Tan, D.; Kondo, Y.; Sachse, G. W.; Anderson, B.; Gregory,
14 G. L.; Vay, S.; Singh, H. B.; Blake, D. R.; Shelter, R. (1999) Ozone production in the upper troposphere and
15 the influence of aircraft during SONEX: approach of NOx-saturated conditions. Geophys. Res. Lett.
16 26: 3081-3084.
17 Jaegle, L.; Jacob, D. J.; Brune, W. H.; Faloona, I.; Tan, D.; Heikes, B. G.; Kondo, Y.; Sachse, G. W.; Anderson, B.;
18 Gregory, G. L.; Singh, H. B.; Pueschel, R.; Ferry, G.; Blake, D. R.; Shelter, R. E. (2000) Photochemistry of
19 HOX in the upper troposphere at northern midlatitudes. J. Geophys. Res. (Atmos.) 105(D3): 3877-3892.
20 Jaegle, L.; Jacob, D. J.; Brune, W. H.; Wennberg, P. O. (2001) Chemistry of HOX radicals in the upper troposphere.
21 Atmos. Environ. 35: 469-489.
22 Jaffe, D.; Price, H.; Parrish, D.; Goldstein, A.; Harris, J. (2003) Increasing background ozone during spring on the
23 west coast of North America. Geophys. Res. Lett. 30: 10,1029/2003GL017024.
24 Jaffe, D.; McKendry, I.; Anderson, T.; Price, H. (2003) Six 'new'episodes of trans-Pacific transport of air pollutants.
25 Atmos. Environ. 37: 391-404.
26 James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003a) A 15-year climatology of
27 stratosphere-troposphere exchange with a Lagrangian particle dispersion model: 1. Methodology and
28 validation. J. Geophys. Res. 108(D12): 10.1029/2002JD002637.
29 James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003b) A 15-year climatology of
30 stratosphere-troposphere exchange with a Lagrangian particle dispersion model. 2. Mean climate and seasonal
31 variability. J. Geophys. Res. (Atmos.) 108(D12): 10.1029/2002JD002639.
32 Jang, M.; Kamens, R. M. (2001) Characterization of secondary aerosol from the photooxidation of toluene in the
33 presence of NOX and 1-propene. Environ. Sci. Technol. 35: 3626-3639.
34 Jankowski, J. J.; Kieber, D. J.; Mopper, K.; Neale, P. J. (2000) Development and intercalibration of ultraviolet solar
35 actinometers. Photochem. Photobiol. 71: 431-440.
36 Jenkin, M. E.; Cox, R. A.; Williams, D. J. (1988) Laboratory studies of the kinetics of formation of nitrous acid from
37 the thermal reaction of nitrogen dioxide and water vapour. Atmos. Environ. 22: 487-498.
38 Jenkin, M. E.; Saunders, S. M.; Pilling, M. J. (1997) The tropospheric degradation of volatile organic compounds: a
3 9 protocol for mechanism development. Atmos. Environ. 31:81-104.
40 Jet Propulsion Laboratory. (2003) Chemical kinetics and photochemical data for use in atmospheric studies.
41 Pasadena, CA: California Institute of Technology; JPL publication no. 02-25. Available:
42 http://jpldataeval.jpl.nasa.gov/pdf/JPL_02-25_l_intro_rev0.pdf (18 December 2003).
43 Jeuken, A. B. M.; Eskes, H. J.; Van Velthoven, P. F. J.; Kelder, H. M.; Holm, E. V. (1999) Assimilation of total
44 ozone satellite measurements in a three-dimensional tracer transport model. J. Geophys. Res. (Atmos.)
45 104: 5551-5563.
46 Jimenez, P.; Baldasano, J. M.; Dabdub, D. (2003) Comparison of photochemical mechanisms for air quality
47 modeling. Atmos. Environ. 37: 4179-4194.
48 Johnson, W. B.; Viezee, W. (1981) Stratospheric ozone in the lower troposphere -1. presentation and interpretation
49 of aircraft measurements. Atmos. Environ. 15: 1309-1323.
50 Johnson, C. E.; Collins, W. J.; Stevenson, D. S.; Derwent, R. G. (1999) The relative roles of climate and emissions
51 changes on future oxidant concentrations. J. Geophys. Res. (Atmos.) 104: 18,631-18,645.
52 Johnson, B. J.; Oltmans, S. J.; Vomel, H.; Smit, H. G. J.; Deshler, T.; Kroger, C. (2002) Electrochemical
53 concentration cell (ECC) ozonesonde pump efficiency measurements and tests on the sensitivity to ozone of
54 buffered and unbuffered ECC sensor cathode solutions. J. Geophys. Res. 107(D19): 10.1029/2001JD000557.
55 Jordan, C. E.; Dibb, J. E.; Finkel, R. C. (2003) 10Be/7Be tracer of atmospheric transport and stratosphere-troposphere
56 exchange. J. Geophys. Res. 108(D8): 10.1029/2002JD002395.
57 Junge, C. E. (1962) Global ozone budget and exchange between stratosphere and troposphere. Tellus 14: 363-377.
January 2005 AX2-170 DRAFT-DO NOT QUOTE OR CITE
-------
1 Junge, C. E. (1963) Air chemistry and radioactivity. New York, NY: Academic Press. (Van Mieghem, I; Hales,
2 A. L., eds. International geophysics series: v. 4).
3 Kain, J. S.; Fritsch, J. M. (1993) Convective parameterization in mesoscale models: the Kain-Fritsch scheme. In:
4 Emanuel, K. A.; Raymond, D. J., eds. The Representation of Cumulus Convection in Numerical Models.
5 Boston, MA: American Meteorological Society; pp. 165-170. (Meteorological Monographs, v. 24, no. 46).
6 Kane, M. M.; Rendell, A. R.; Jickells, T. D. (1994) Atmospheric scavenging processes over the North Sea. Atmos.
7 Environ. 28: 2523-2530.
8 Kane, S. M.; Caloz, F.; Leu, M.-T. (2001) Heterogeneous uptake of gaseous N2O5 by (NH4)2SO4 NH4HSO4, and
9 H2SO4 aerosols. J. Phys. Chem. A 105: 6465-6470.
10 Karamchandani, P.; Santos, L.; Sykes, L; Zhang, Y.; Tonne, C.; Seigneur, C. (2000) Development and evaluation of
11 a state-of-the-science reactive plume model. Environ. Sci. Technol. 34: 870-880.
12 Karamchandani, P.; Seigneur, C.; Vijayaraghavan, K.; Wu, S.-Y. (2002) Development and application of a
13 state-of-the-science plume-in-grid model. J. Geophys. Res. (Atmos.) 107: 10.1019/2002D002123.
14 Kasibhatla, P.; Chameides, W. L. (2000) Seasonal modeling of regional ozone pollution in the eastern United States.
15 Geophys. Res. Lett. 27: 1415-1418.
16 Kasting, J. F.; Singh, H. B. (1986) Nonmethane hydrocarbons in the troposphere: impact on the odd hydrogen and
17 odd nitrogen chemistry, J. Geophys. Res. 91: 13,239-13,256.
18 Kavouras, I. G.; Mihalopoulus, N.; Stephanou, E. G. (1999) Secondary organic aerosol formation vs primary organic
19 aerosol emission: in situ evidence for the chemical coupling between monoterpene acidic photooxidation
20 products and new particle formation over forests. Environ. Sci. Technol. 33: 1028-1037.
21 Kawamura, K.; Semere, R.; Imai, Y.; Fuji, Y.; Hayashi, M. (1996) Water soluble dicarboxylic acids and related
22 compounds in Antarctic aerosols. J. Geophys. Res. 101: 18,721-18,728.
23 Keene, W. C.; Savoie, D. L. (1998) The pH of deliquesced sea-salt aerosol in polluted marine air. Geophys. Res.
24 Lett. 25: 2181-2194.
25 Keene, W. C.; Savoie, D. L. (1999) Correction to "The pH of deliquesced sea-salt aerosol in polluted marine air."
26 Geophys. Res. Lett. 26: 1315-1316.
27 Keene, W. C.; Sander, R.; Pszenny, A. A. P.; Vogt, R.; Crutzen, P. J.; Galloway, J. N. (1998) Aerosol pH in the
28 marine boundary layer: a review and model evaluation. J. Aerosol Sci. 29: 339-356.
29 Keene, W. C.; Pszenny, A. A. P.; Maben, J. R.; Sander, R. (2002) Variation of marine aerosol acidity with particle
30 size. Geophys. Res. Lett. 29(7): 10.1029/2001GL013881.
31 Kelly, N. A.; Wolff, G. T.; Ferman, M. A. (1982) Background pollutant measurements in air masses affecting the
32 eastern half of the United States -1. air masses arriving from the northwest. Atmos. Environ. 16: 1077-1088.
33 Kelly, N. A.; Ferman, M. A.; Wolff, G. T. (1986) The chemical and meteorological conditions associated with high
34 and low ozone concentrations in southeastern Michigan and nearby areas of Ontario. J. Air Pollut. Control
35 Assoc. 36: 150-158.
36 Kenley, R. A.; Davenport, J. E.; Hendry, D. G. (1981) Gas-phase hydroxyl radical reactions. Products and pathways
3 7 for the reaction of OH with aromatic hydrocarbons. J. Phys. Chem. 85: 2740-2746.
3 8 Kim, B. M.; Henry, R. C. (2000) Application of SAFER model to the Los Angeles PM10 data. Atmos. Environ.
39 34: 1747-1759.
40 Kim, K.-H.; Kim, M.-Y. (2001) Comparison of an open path differential optical absorption spectroscopy system and
41 a conventional in situ monitoring system on the basis of long-term measurements of SO2, NO2, and O3.
42 Atmos. Environ. 35: 4059-4072.
43 King, D. W.; Lounsbury, H. A.; Millero, F. J. (1995) Rates and mechanisms of Fe(II) oxidation at nonmolar total
44 iron concentrations. Environ. Sci. Technol. 29: 818-824.
45 King, M. D.; Canosa-Mas, C. E.; Wayne, R. P. (2002) A structure activity relationship (SAR) for predicting rate
46 constants for the reaction of nitrogen dioxide (NO2) with alkenes. Phys. Chem. Chem. Phys. 4: 295-303.
47 Kireev, S. V.; Shnyrev, S. L.; Zhiganov, A. A. (1999) A laser fluorimeter for NO and NO2 in atmosphere. Instrum.
48 Exp. Tech. 42: 701-703.
49 Kleindienst, T. E.; Hudgens, E. E.; Smith, D. F.; McElroy, F. F.; Bufalini, J. J. (1993) Comparison of
50 chemiluminescence and ultraviolet ozone monitor responses in the presence of humidity and photochemical
51 pollutants. Air Waste 43: 213-222.
52 Kleindienst, T. E.; Smith, D. F.; Li, W.; Edney, E. O.; Driscoll, D. J.; Speer, R. E.; Weathers, W. S. (1999)
53 Secondary organic aerosol formation from the oxidation of aromatic hydrocarbons in the presence of dry
54 submicronammonium sulfate aerosol. Atmos. Environ. 33: 3669-3681.
55 Kleinman, L. I. (2000) Ozone process insights from field experiments - part II: observation-based analysis for ozone
56 production. Atmos. Environ. 34: 2023-2034.
January 2005 AX2-171 DRAFT-DO NOT QUOTE OR CITE
-------
1 Kleinman, L. I.; Daum, P. H.; Lee, Y.-N.; Springston, S. R.; Newman, L.; Leaitch, W. R.; Banic, C. M.; Isaac,
2 G. A.; MacPherson, J. I. (1996a) Measurement of O3 and related compounds over southern Nova Scotia. 1.
3 Vertical distributions. J. Geophys. Res. [Atmos.] 101: 29,043-29,060.
4 Kleinman, L. L; Daum, P. H.; Springston, S. R.; Leaitch, W. R.; Banic, C. M.; Isaac, G. A.; Jobson, T. B.; Niki, H.
5 (1996b) Measurement of O3 and related compounds over southern Nova Scotia. 2. Photochemical age and
6 vertical transport. J. Geophys. Res. [Atmos.] 101: 29,061-29,074.
7 Kleinman, L. I.; Daum, P. H.; Lee, J. H.; Lee, Y.-N.; Nunnermacker, S. R.; Springston, L.; Newman, J.;
8 Weinstein-Lloyd, J.; Sillman, S. (1997) Dependence of ozone production on NOand hydrocarbons in the
9 troposphere. Geophys. Res. Lett. 24: 2299-2302.
10 Kleinman, L. I.; Daum, P. H.; Imre, D. G.; Lee, J. H.; Lee, Y.-N.; Nunnermacker, L. J.; Springston, S. R.;
11 Weinstein-Lloyd, J.; Newman, L. (2000) Ozone production in the New York City urban plume. J. Geophys.
12 Res. [Atmos.] 105: 14,495-14,511.
13 Kleinman, L. L; Daum, P. H.; Lee, Y.-N.; Nunnermacker, L. J.; Springston, S. R.; Weinstein-Lloyd, J.; Rudolph, J.
14 (2001) Sensitivity of ozone production rate to ozone precursors. Geophys. Res. Lett. 28: 2903-2906.
15 Kleinman, L. L; Daum, P. H.; Imre, D.; Lee, Y.-N.; Nunnermacker, L. J.; Springston, S. R.; Weinstein-Lloyd, J.;
16 Rudolph, J. (2002) Ozone production rate and hydrocarbon reactivity in 5 urban areas: a cause of high ozone
17 concentration in Houston. Geophys. Res. Lett. 29: 10.1029/2001GL014569.
18 Kley, D.; Crutzen, P. J.; Smit, H. G. J.; Vomel, H.; Oltmans, S. J.; Grassl, H.; Ramanathan, V. (1996) Observations
19 of near-zero ozone concentrations over the convective Pacific: effects on air chemistry. Science 274: 230-233.
20 Klotz, B.; Barnes, I.; Becker, K. H.; Golding, B. T. (1997) Atmospheric chemistry of benzene oxide/oxepin.
21 J. Chem. Soc. Faraday Trans. 93: 1507-1516.
22 Klotz, B; Barnes, L; Becker, K. H. (1998) New results on the atmospheric photooxidation of simple alkylbenzenes.
23 Chem. Phys. 231:289-301.
24 Knake, R.; Hauser, P. C. (2002) Sensitive electrochemical detection of ozone. Anal. Chim. Acta 459: 199-207.
25 Knispel, R.; Koch, R.; Siese, M.; Zetzsch, C. (1990) Adduct formation of OH radicals with benzene, toluene, and
26 phenol and consecutive reactions of the adducts with NO^ and O2. Ber. Bunsen-Ges. Phys. Chem.
27 94: 1375-1379.
28 Koch, T. G.; Banham, S. F.; Sodeau, J. R.; Horn, A. B.; McCoustra, M. R. S.; Chesters, M. A. (1997) Mechanisms
29 for the heterogeneous hydrolysis of hydrogen chloride, chlorine nitrate and dinitrogen pentoxide on water-rich
30 atmospheric particle surfaces. J. Geophys Res. (Atmos.) 102: 1513-1522.
31 Konrad, S.; Schmitz, T.; Buers, H.-J.; Houben, N.; Mannschreck, K.; Mihelcic, D.; Mtisgen, P.; Patz, H.-W.;
32 Holland, F.; Hofzumahaus, A.; Schafer, H.-J.; Schroder, S.; Volz-Thomas, A.; Bachmann, K.; Schlomski, S.;
33 Moortgat, G.; GroBmann, D. (2003) Hydrocarbon measurements at Pabstthum during the BERLIOZ
34 campaign and modeling of free radicals. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/2001JD000866.
35 Kotamarthi, V. R.; Gaffney, J. S.; Marley, N. A.; Doskey, P. V. (2001) Heterogeneous NOX chemistry in the polluted
36 PEL. Atmos. Environ. 35: 4489-4498.
37 Krol, M.; Van Leeuwen, P. J.; Lelieveld, J. (1998) Global OH trend inferred from methylchloroform measurements.
38 J. Geophys. Res. [Atmos.] 103: 10,697-10,711.
39 Kumar, N.; Russell, A. G. (1996) Development of a computationally efficient, reactive subgrid-scale plume model
40 and the impact in the northeastern United States using increasing levels of chemical detail. J. Geophys. Res.
41 [Atmos.] 101: 16,737-16,744.
42 Kwok, E. S. C.; Aschmann, S. M.; Atkinson, R.; Arey, J. (1997) Products of the gas-phase reactions of o-, m- and
43 p-xylens with the OH radical in the presence and absence of NOX. J. Chem. Soc. Faraday Trans.
44 93:2847-2854.
45 Lamb, B.; Gay, D.; Westberg, H.; Pierce, T. (1993) A biogenic hydrocarbon emission inventory for the U.S.A. using
46 a simple forest canopy model. Atmos. Environ. Part A 27: 1673-1690.
47 Lammel, G.; Cape, J. N. (1996) Nitrous acid and nitrite in the atmosphere. Chem. Soc. Rev. 25: 361-369.
48 Lammel, G.; Perner, D. (1988) The atmospheric aerosol as a source of nitrous acid in the polluted atmosphere.
49 J. Aerosol Sci. 19: 1199-1202.
50 Langford, A. O.; Reid, S. J. (1998) Dissipation and mixing of a small-scale stratospheric intrusion in the upper
51 troposphere. J. Geophys. Res. 103(D23): 31,265-31,276.
52 Lantz, K. O.; Shelter, R. E.; Cantrell, C. A.; Flocke, S. J.; Calvert, J. G.; Madronich, S. (1996) Theoretical,
53 actinometric, and radiometric determinations of the photolysis rate coefficient of NO2 during the Mauna Loa
54 Observatory Photochemistry Experiment 2. J. Geophys. Res. (Atmos.) 101(D9): 14,613-14629.
55 Law, K. S.; Plantevin, P. H.; Shallcross, D. E.; Rogers, H. L.; Pyle, J. A.; Grouhel, C.; Thouret, V; Marenco, A.
56 (1998) Evaluation of modeled O3 using measurement of ozone by airbus in-service aircraft (MOZAIC) data.
57 J. Geophys. Res. (Atmos.) 103: 25,721-25,737.
January 2005 AX2-172 DRAFT-DO NOT QUOTE OR CITE
-------
1 Law, K. S.; Plantevin, P.-H.; Thouret, V.; Marenco, A.; Asman, W. A. H.; Lawrence, M; Crutzen, P. J.; Muller,
2 J. F.; Hauglustaine, D. A.; Kanakidou, M. (2000) Comparison between global chemistry transport model
3 results and measurement of ozone and water vapor by airbus in-service aircraft (MOZAIC) data. J. Geophys.
4 Res. (Atmos.) 105: 1503-1525.
5 Lawrence, M. G.; Crutzen, P. J.; Rasch, P. J.; Eaton, B. E.; Mahowald, N. M. (1999) A model for studies of
6 tropospheric photochemistry: description, global distributions and evaluation. J. Geophys. Res. (Atmos.)
7 104:26,245-26,277.
8 Lean, J. L. (2000) Short term, direct indices of solar variability. Space Science Reviews 94: 39-51.
9 Lefohn, A. S.; Oltmans, S. J.; Dann, T.; Singh, H. B. (2001) Present-day variability of background ozone in the
10 lower troposphere. J. Geophys. Res. [Atmos.] 106: 9945-9958.
11 Leibrock, E.; Huey, L. G. (2000) Ion chemistry for the detection of isoprene and other volatile organic compounds in
12 ambient air. Geophys. Res. Lett. 27: 1719-1722.
13 Lelieveld, J.; Crutzen, P. J. (1990) Influences of cloud photochemical processes on tropospheric ozone. Nature
14 (London) 343: 227-233.
15 Lelieveld, J.; Crutzen, P. J. (1994) Role of deep cloud convection in the ozone budget of the troposphere. Science
16 264: 1759-1761.
17 Lelieveld, J.; Dentener, F. J. (2000) What controls tropospheric ozone? J. Geophys. Res. [Atmos.] 105: 3531-3551.
18 Leser, H.; Honninger, G.; Platt, U. (2003) MAX-DOAS measurements of BrO and NO2 in the marine boundary
19 layer. Geophys. Res. Lett. 30: 10.1029/2002GL015811.
20 Leu, M.-T. (2003) Laboratory studies of interaction between trace gases and sulphuric acid or sulphate aerosols
21 using flow-tube reactors. Int. Rev. Phys. Chem. 22: 341-376.
22 Li, Q.; Jacob, D. J.; Logan, J. A.; Bey, I.; Yantosca, R. M.; Liu, H.; Martin, R. V.; Fiore, A. M.; Field, B. D.;
23 Duncan, B. N. (2001) A tropospheric ozone maximum over the Middle East. Geophys. Res. Lett.
24 28: 3235-3238.
25 Li, Q.; Jacob, D. J.; Fairlie, T. D.; Liu, H.; Martin, R. V.; Yantosca, R. M. (2002) Stratospheric versus pollution
26 influences on ozone at Bermuda: reconciling past analyses. J. Geophys. Res. 107(D22):
27 10.1029/2002JD002138.
28 Liang, J.; Horowitz, L. W.; Jacob, D. J.; Wang, Y.; Fiore, A. M.; Logan, J. A.; Gardner, G. M.; Munger, J. W. (1998)
29 Seasonal variations of reactive nitrogen species and ozone over the United States, and export fluxes to the
30 global atmosphere. J. Geophys. Res. (Atmos.) 103: 13,435-13,450.
31 Liao, H.; Yung, Y. L.; Seinfeld, J. H. (1999) Effects of aerosols on tropospheric photolysis rates in clear and cloudy
32 atmospheres. J. Geophys. Res. (Atmos.) 104: 23,697-23,707.
33 Lillis, D.; Cruz, C. N.; Collett, J.; Richards, L. W.; Pandis, S. N. (1999) Production and removal of aerosol in a
34 polluted fog layer: model evaluation and fog effect on PM. Atmos. Environ. 33: 4797-4816.
35 Lin, C.; Milford, J. B. (1994) Decay-adjusted chemical mass balance receptor modeling for volatile organic
36 compounds. Atmos. Environ. 28: 3261-3276.
37 Lin, X.; Trainer, M.; Liu, S. C. (1988) On the nonlinearity of the tropospheric ozone production. J. Geophys. Res.
38 [Atmos.] 93: 15,879-15,888.
39 Lindinger, W.; Hansel, A.; Jordan, A. (1998) Proton-transfer-reaction mass spectrometry (PTR-MS): on-line
40 monitoring of volatile organic compounds at pptv levels. Chem. Soc. Rev. 27: 347-354.
41 Liu, M.-K.; Seinfeld, J. H. (1975) On the validity of grid and trajectory models of urban air pollution. Atmos.
42 Environ. 9: 555-574.
43 Liu, S. C.; Trainer, M.; Fehsenfeld, F. C.; Parrish, D. D.; Williams, E. J.; Fahey, D. W.; Htibler, G.; Murphy, P. C.
44 (1987) Ozone production in the rural troposphere and the implications for regional and global ozone
45 distributions. J. Geophys. Res. [Atmos.] 92: 4191-4207.
46 Liu, X. H.; Hegg, D. A.; Stoelinga, M. T. (2001) Numerical simulation of new particle formation over the northwest
47 Atlantic using the MM5 mesoscale model coupled with sulfur chemistry. J. Geophys. Res. [Atmos.]
48 106: 9697-9715.
49 Liu, H.; Jacob, D. J.; Bey, I.; Yantosca, R. M. (2001) Constraints from 210Pb and 7Be on wet deposition and transport
50 in a global three-dimensional chemical tracer model driven by assimilated meteorological fields. J. Geophys.
51 Res. [Atmos.] 106: 12,109-12,128.
52 Logan, J. A. (1989) Ozone in rural areas of the United States. J. Geophys. Res. [Atmos.] 94: 8511-8532.
53 Logan, J. A.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B. (1981) Tropospheric chemistry: a global perspective.
54 J. Geophys. Res. C: Oceans Atmos. 86: 7210-7254.
55 Logan, J. A. (1994) Trends in the vertical distribution of ozone: an analysis of ozonesonde data. J. Geophys. Res.
56 (Atmos.) 99: 25,553-25,585.
January 2005 AX2-173 DRAFT-DO NOT QUOTE OR CITE
-------
1 Longfellow, C. A.; Ravishankara, A. R.; Hanson, D. R. (1999) Reactive uptake on hydrocarbon soot: focus on NO2.
2 J. Geophys. Res. (Atmos.) 104: 13,833-13,840.
3 Lonneman, W. A.; Seila, R. L. (1993) Hydrocarbon compositions in Los Angeles and New York 20 years later.
4 Presented at: International symposium on measurement of toxic and related air pollutants. Pittsburgh, PA: Air
5 & Waste Management Association.
6 Lu, C.-H.; Chang, J. S. (1998) On the indicator-based approach to assess ozone sensitivities and emissions features.
7 J. Geophys. Res. (Atmos.) 103: 3453-3462.
8 Lu, R.; Turco, R. P.; Jacobson, M. Z. (1997) An integrated air pollution modeling system for urban and regional
9 scales: 1. Structure and performance. J. Geophys. Res. [Atmos.] 102: 6063-6079.
10 Ludwig, F. L.; Reiter, E.; Shelar, E.; Johnson, W. B. (1977) The relation of oxidant levels to precursor emissions and
11 meteorological features: v. 1, analysis and findings. Research Triangle Park, NC: U.S. Environmental
12 Protection Agency, Office of Air Quality Planning and Standards; report no. EPA-450/3-77-022a. Available
13 from: NTIS, Springfield, VA; PB-275 001.
14 Luke, W. T. (1997) Evaluation of a commercial pulsed fluorescence detector for the measurement of low-level SO2
15 concentrations during the Gas-Phase Sulfur Intercomparison Experiment. J. Geophys. Res. (Atmos.)
16 102: 16,255-16,265.
17 Luke, W. T.; Dickerson, R. R.; Ryan, W. F.; Pickering, K. E.; Nunnermacker, L. J. (1992) Tropospheric chemistry
18 over the lower Great Plains of the United States. 2. Trace gas profiles and distributions. J. Geophys. Res.
19 97(D18): 20647-20670.
20 Luke, W. T.; Watson, T. B.; Olszyna, K. J.; Gunter, R. L.; McMillen, R. T.; Wellman, D. L.; Wilkison, S. W. (1998)
21 A comparison of airborne and surface trace gas measurements during the Southern Oxidants Study (SOS).
22 J. Geophys. Res. [Atmos.] 103: 22,317-22,337.
23 Luria, M.; Valente, R. J.; Tanner, R. L.; Gillani, N. V.; Imhoff, R. E.; Mueller, S. F.; Olszyna, K. J.; meagher, J. F.
24 (1999) The evolution of photochemical smog in a power plant plume. Atmos. Environ. 33: 3023-3036.
25 Luria, M.; Tanner, R. L.; Imhoff, R. E.; Valente, R. J.; Bailey, E. M.; Mueller, S. F. (2000) Influence of natural
26 hydrocarbons on ozone formation in an isolated power plant plume. J. Geophys. Res. [Atmos.]
27 105:9177-9188.
28 Madronich, S. (1987) Photodissociation in the atmosphere. 1. Actinic flux and the effects of ground reflections and
29 clouds. J. Geophys. Res. [Atmos.] 92: 9740-9752.
30 Madronich, S.; McKenzie, R. L.; Caldwell, M. M.; Bjorn, L. O. (1995) Changes in ultraviolet radiation reaching the
31 Earth's surface. Ambio 24: 143-152.
32 Mahlman, J. D. (1997) Dynamics of transport processes in the upper troposphere. Science 276: 1079-1083.
33 Mahlman, J. D.; Pinto, J. P.; Umscheid, L. J. (1994) Transport, radiative, and dynamical effects of the Antarctic
34 ozone hole: a GFDL "SKYHI" model experiment. J. Atmos. Sci. 51: 489-508.
3 5 Mahrt, L. (1998) Stratified atmospheric boundary layers and breakdown of models. Theor. Comput. Fluid Dyn.
36 11:263-279.
37 Martilli, A.; Neftel, A.; Favaro, G.; Kirchner, F.; Sillman, S.; Clappier, A. (2002) Simulation of the ozone formation
38 in the northern part of the Po Valley. J. Geophys. Res. (Atmos.) 107(D22): 10.1029/2001JD000534.
39 Martin, D.; Tsivou, M.; Bonsang, B.; Abonnel, C.; Carsey, T.; Springer-Young, M.; Pszenny, A.; Suhre, K. (1997)
40 Hydrogen peroxide in the marine atmospheric boundary layer during the Atlantic Stratocumulus Transition
41 Experiment/Marine Aerosol and Gas Exchange experiment in the eastern subtropical North Atlantic.
42 J. Geophys. Res. (Atmos.) 102: 6003-6015.
43 Martin, R. V.; Jacob, D. J.; Chance, K. V.; Kurosu, T. P.; Palmer, P. I.; Evans, M. J. (2003) Global inventory of
44 nitrogen oxide emissions constrained by space-based observations of NO2 columns. J. Geophys. Res.
45 108(D17): 10.1029/2003 JD003453.
46 Martinez, M.; Harder, H.; Kovacs, T. A.; Simpas, J. B.; Bassis, J.; Lesher, R.; Brune, W. H.; Frost, G. J.; Williams,
47 E. J.; Stroud, C. A.; Jobson, B. T.; Roberts, J. M.; Hall, S. R.; Shelter, R. E.; Wert, B.; Fried, A.; Alicke, B.;
48 Stutz, J.; Young, V. L.; White, A. B.; Zamora, R. J. (2003) OH and HO2 concentrations, sources, and loss
49 rates during the Southern Oxidants study in Nashville, Tennessee, summer 1999. J. Geophys. Res. (Atmos.)
50 108(D19): 10.1029/2003 JD003551.
51 Marufu, L. T.; Taubman, B. F.; Bloomer, B.; Piety, C. A.; Doddridge, E.G.; Stehr, J. W.; Dickerson, R. R. (2004)
52 The 2003 North American electrical blackout: an accidental experiment in atmospheric chemistry. Geophys.
53 Res. Lett. 31(L13106): 10.1029/2004GL019771.
54 Massman, W. J.; Pederson, J.; Delany, A.; Grantz, D.; Denhartog, G.; Neumann, H. H.; Oncley, S. P.; Pearson, R.;
55 Shaw, R. H. (1994) An evaluation of the regional acid deposition model surface module for ozone uptake at
56 3 sites in the San-Joaquin valley of California. J. Geophys. Res. [Atmos.] 99: 8281-8294.
January 2005 AX2-174 DRAFT-DO NOT QUOTE OR CITE
-------
1 Matsumi, Y.; Murakami, S.-L; Kono, M. (2001) High-sensitivity instalment for measuring atmospheric NO2. Anal.
2 Chem. 73:5485-5493.
3 Mauldin, R. L., Ill; Tanner, D. J.; Eisele, F. L. (1998) A new chemical ionization mass spectrometer technique for
4 the fast measurement of gas phase nitric acid in the atmosphere. J. Geophys. Res. 103(D3): 3361-3367.
5 McClenny, W. A., ed. (2000) Recommended methods for ambient air monitoring of NO, NO2, NOy, and individual
6 NOZ species. Research Triangle Park, NC: U.S. Environmental Protection Agency, National Exposure
7 Research Laboratory; report no. GPRA APM #442.
8 McDow, S. R.; Jang, M.; Hong, Y.; Kamens, R. M. (1996) An approach to studying the effect of organic
9 composition on atmospheric aerosol photochemistry. J. Geophys. Res. (Atmos.) 101(D14): 19,593-19,600.
10 McFiggans, G.; Allan, B.; Coe, H.; Plane, H. M. C.; Carpenter, L. J.; O'Dowd, C. (2000) A modeling study of iodine
11 chemistry in the marine boundary layer. J. Geophys. Res. (Atmos.) 105: 14,371-14,385.
12 McFiggans, G.; Cox, R. A.; Mossinger, J. C.; Allan, B. J.; Plane, J. M. C. (2002) Active chlorine release from
13 marine aerosols: roles for reactive iodine and nitrogen species. J. Geophys. Res. 17(D15):
14 10.1029/2001JD000383.
15 McKeen, S. A.; Liu, S. C. (1993) Hydrocarbon ratios and photochemical history of air masses. Geophys. Res. Lett.
16 20:2363-2366.
17 McKeen, S. A.; Liu, S. C. (1997) Hydrocarbon ratios and the photochemical history of air masses. Geophys. Res.
18 Lett. 20: 2363-2366.
19 McKeen, S. A.; Liu, S. C.; Hsie, E.-Y.; Lin, X.; Bradshaw, J. D.; Smyth, S.; Gregory, G. L.; Blake, D. R. (1996)
20 Hydrocarbon ratios during PEM-WEST A: a model perspective. J. Geophys. Res. (Atmos.)
21 101(D1): 2087-2109.
22 McKeen, S. A.; Mount, G.; Eisele, F.; Williams, E.; Harder, J.; Goldan, P.; Kuster, W.; Liu, S. C.; Baumann, K.;
23 Tanner, D.; Fried, A.; Sewell, S.; Cantrell, C.; Shelter, R. (1997) Photochemical modeling of hydroxyl and its
24 relationship to other species during the Tropospheric OH Photochemistry Experiment. J. Geophys. Res.
25 (Atmos.) 102: 6467-6493.
26 McKenna, D. S.; Konopka, P.; GrooB, J.-U.; Gtinther, G.; Mtiller, R.; Spang, R.; Offermann, D.; Orsolini, Y. (2002)
27 A new chemical Lagrangian model of the stratosphere (CLaMS) 1. formulation of advection and mixing.
28 J. Geophys. Res. 107(D16): 10.1029/2000JD000114.
29 Meagher, J. F.; Cowling, E. B.; Fehsenfeld, F. C.; Parkhurst, W. J. (1998) Ozone formation and transport in
30 southeastern United States: overview of the SOS Nashville/Middle Tennessee Ozone Study. J. Geophys. Res.
31 [Atmos.] 103:22,213-22,223.
32 Mendoza-Dominguez, A.; Russell, A. G. (2000) Iterative inverse modeling and direct sensitivity analysis of a
33 photochemical air quality model. Environ. Sci. Technol. 3: 4974-4981.
34 Mendoza-Dominguez, A.; Russell, A. G. (2001) Estimation of emission adjustments from the application of
35 four-dimensional data assimilation to photochemical air quality modeling. Atmos. Environ. 35: 2879-2894.
36 Meng, Z.; Seinfeld, J. H. (1996) Time scales to achieve atmospheric gas-aerosol equilibrium for volatile species.
37 Atmos. Environ. 30: 2889-2900.
38 Meng, Z.; Dabdub, D.; Seinfeld, J. H. (1997) Chemical coupling between atmospheric ozone and paniculate matter.
39 Science (Washington, DC) 277: 116-119.
40 Mentel, T. F.; Sohn, M.; Wahner, A. (1999) Nitrate effect in the heterogeneous hydrolysis of dinitrogen pentoxide
41 on aqueous aerosols. Phys. Chem. Chem. Phys. 1: 5451-5457.
42 Merrill, J. T.; Moody, J. L. (1996) Synoptic meteorology and transport during the North Atlantic Regional
43 Experiment (NARE) intensive: overview. J. Geophys. Res. [Atmos.] 101: 28,903-28,921.
44 Meyers, T.; Sickles, J.; Dennis, R.; Rusell, K.; Galloway, J.; Church, T. (2001) Atmospheric nitrogen deposition to
45 coastal estuaries and their watersheds. In: Valigura, R. A.; Alexander, R. B.; Castro, M. S.; Meyers, T. P.;
46 Paerl, H. W.; Stacey, P. E.; Turner, R. E., eds. Nitrogen Loading in Coastal Water Bodies: an Atmospheric
47 Perspective. Washington, DC: AGU Press.
48 Meyrahn, H.; Pauly, I; Schneider, W.; Warneck, P. (1986) Quantum yields for photodissociation of acetone in air
49 and an estimate for the life time of acetone in the lower troposphere. J. Atmos. Chem. 4: 277-291.
50 Michel, A. E.; Usher, C. R.; Grassian, V. H. (2002) Heterogeneous and catalytic uptake of ozone on mineral oxides
51 and dusts: a Knudsen cell investigation. Geophys. Res. Lett. 29: DOI: 10.1029/2002GLO14896.
52 Michel, A. E.; Usher, C. R.; Grassian, V. H. (2003) Reactive uptake of ozone on mineral oxides and mineral dusts.
53 Atmos. Environ. 37: 3201-3211.
54 Mickley, L. J.; Murti, P. P.; Jacob, D. J.; Logan, J. A.; Koch, D. M.; Rind, D. (1999) Radiative forcing from
55 tropospheric ozone calculated with a unified chemistry-climate model. J. Geophys. Res. [Atmos.]
56 104: 30,153-30,172.
January 2005 AX2-175 DRAFT-DO NOT QUOTE OR CITE
-------
1 Mihelcic, D.; Holland, F.; Hofzumahaus, A.; Hoppe, L.; Konrad, S.; Mtisgen, P.; Patz, H.-W.; Schafer, H.-J.;
2 Schmitz, T.; Volz-Thomas, A.; Bachmann, K.; Schlomski, S.; Platt, U.; Geyer, A.; Alicke, B.; Moortgat,
3 G. K. (2003) Peroxy radicals during BERLIOZ and Pabstthum: measurements, radical budgets and ozone
4 production. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/2001JD001014.
5 Milanchus, M. L.; Rao, S. T.; Zurbenko. I. G. (1998) Evaluating the effectiveness of ozone management efforts in
6 the presence of meteorological variability. J. Air Waste Manage. Assoc. 48: 201-215.
7 Milford, J. B.; Russell, A. G.; McRae, G. J. (1989) A new approach to photochemical pollution control: implications
8 of spatial patterns in pollutant responses to reductions in nitrogen oxides and reactive organic gas emissions.
9 Environ. Sci. Technol. 23: 1290-1301.
10 Milford, J. B.; Gao, D.; Sillman, S.; Blossey, P.; Russell, A. G. (1994) Total reactive nitrogen (NOy) as an indicator
11 of the sensitivity of ozone to reductions in hydrocarbon and NOX emissions. J. Geophys. Res. 99: 3533-3542.
12 Milne, P. J.; Zika, R. G. (1993) Amino acid nitrogen in atmospheric aerosols: occurrence, sources, and
13 photochemical modification. J. Atmos. Chem. 16: 361-398.
14 Milne, P. J.; Prados, A. L; Dickerson, R. R.; Doddrige, B. G.; Riemer, D. D.; Zika, R. G.; Merrill, J. T.; Moody, J. L.
15 (2000) Nonmethane hydrocarbon mixing ratios in continental outflow air from Eastern North America: export
16 of ozone precursors to Bermuda. J. Geophys. Res. [Atmos.] 105: 9981-9990.
17 Mochida, M.; Hirokawa, J.; Kajii, Y.; Akimoto, H. (1998) Heterogeneous reactions of C12 with sea salts at ambient
18 temperature: implications for halogen exchange in the atmosphere. Geophys. Res. Lett. 25: 3927-3930.
19 Mohnen, V. A.; Reiter, E. R. (1977) International conference on oxidants, 1976 - analysis of evidence and
20 viewpoints: part III. the issue of stratospheric ozone intrusion. Research Triangle Park, NC: U.S.
21 Environmental Protection Agency, Environmental Sciences Research Laboratory; EPA report no.
22 EPA-600/3-77-115. Available from: NTIS, Springfield, VA; PB-279 010.
23 Monks, P. S. (2000) A review of the observations and origins of the spring ozone maximum. Atmos. Environ.
24 34: 3545-3561.
25 Monson, R. K.; Harley, P. C.; Litvak, M. E.; Wildermuth, M.; Guenther, A. B.; Zimmerman, P. R.; Fall, R. (1994)
26 Environmental and developmental controls over the seasonal pattern of isoprene emission from aspen leaves.
27 Oecologia 99: 260-270.
28 Moody, J. L.; Oltmans, S. J.; Levy, H., II; Merrill, J. T. (1995) Transport climatology of tropospheric ozone:
29 Bermuda, 1988-1991. J. Geophys. Res. [Atmos.] 100: 7179-7194.
30 Moody, J. L.; Davenport, J. C.; Merrill, J. T.; Oltmans, S. J.; Parrish, D. D.; Holloway, J. S.; Levy, H., II; Forbes,
31 G. L.; Trainer, M.; Buhr, M. (1996) Meteorological mechanisms for transporting O3 over the western North
32 Atlantic Ocean: a case study for August 24-29, 1993. J. Geophys. Res. 101(D22): 29,213-29,227.
33 Moody, J. L.; Wimmers, A. J.; Davenport, J. C. (1999) Remotely sensed specific humidity: development of a
34 derived product from the GOES Imager Channel 3. Geophys. Res. Lett. 26: 59-62.
35 Moorthi, S.; Suarez, M. J. (1992) Relaxed Arakawa-Schubert: a parameterization of moist convection for general
36 circulation models. Mon. Weather Rev. 120: 978-1002.
37 Mopper, K.; Kieber, D. J. (2000) Marine photochemistry and its impact on carbon cycling. In: Demers, S.; Vernet,
38 M., eds. Effects of UV Radiation on Marine Ecosystems. New York, NY: Cambridge University Press, Ch.4,
39 pp. 101-129.
40 Morales-Morales, R. (1998) Carbon monoxide, ozone, and hydrocarbons in the Baltimore metropolitan area
41 [dissertation]. College Park, MD: University of Maryland, Department of Meteorology. Available from:
42 University Microfilms International, Ann Arbor, MI; publication no. 9908993.
43 Moschonas, N.; Danalatos, D.; Glavas, S. (1999) The effect of O2 and NO2 on the ring retaining products of the
44 reaction of toluene withhydroxyl radicals. Atmos. Environ. 33: 111-116.
45 Mount, G. H.; Eisele, F. L. (1992) An intercomparison of tropospheric OH measurements at Fritz Peak Observatory,
46 Colorado. Science (Washington, DC) 256: 1187-1190.
47 Mount, G. H.; Williams, E. J. (1997) An overview of the tropospheric OH photochemistry experiment, Fritz
48 Peak/Idaho Hill, Colorado, fall 1993. J. Geophys. Res. [Atmos.] 102: 6171-6186.
49 Mozurkewich, M. (1995) Mechanisms for the release of halogens from sea-salt particles by free radical reactions.
50 J. Geophys. Res. (Atmos.) 100: 14,199-14,207.
51 Mtiller, J.-F.; Brasseur, G. (1995) IMAGES: a three-dimensional chemical transport model of the global troposphere.
52 J. Geophys. Res. (Atmos.) 100: 16,445-16,490.
53 Mtiller, J.-F.; Brasseur, G. (1999) Sources of upper tropospheric HOX: a three-dimensional study. J. Geophys. Res.
54 (Atmos.) 104: 1705-1715.
55 Munger, J. W.; Fan, S.-M.; Bakwin, P. S.; Goulden, M. L.; Goldstein, A. H.; Colman, A. S.; Wofsy, S. C. (1998)
56 Regional budgets of nitrogen oxides from continental sources: variations of rates for oxidation and deposition
57 with season and distance from source regions. J. Geophys. Res. (Atmos.) 103: 8355-8368.
January 2005 AX2-176 DRAFT-DO NOT QUOTE OR CITE
-------
1 Munoz, M. S. S.; Rossi, M. J. (2002) Heterogeneous reactions of HNO3 with flame soot generated under different
2 combustion conditions. Reaction mechanism and kinetics. Phys. Chem. Chem. Phys. 4:5110-5118.
3 Murphy, D.; Fahey, D. (1994) An estimate of the flux of stratospheric reactive nitrogen and ozone into the
4 troposphere. J. Geophys. Res. (Atmos.) 99: 5325-5332.
5 Narita S.; Uchiyama S.; Takemi K. (1998) The interference and cross sensitivity for an UV ozone monitor. Ozone
6 Sci. Eng. 20: 495-497.
7 National Bureau of Standards. (1975) Catalog of NBS standard reference materials, 1975-76 edition. Washington,
8 DC: U.S. Department of Commerce, National Bureau of Standards; NBS special publication no. 260.
9 National Research Council. (1991) Rethinking the ozone problem in urban and regional air pollution. Washington,
10 DC: National Academy Press.
11 National Research Council. (1999) Ozone-forming potential of reformulated gasoline. Washington, DC: National
12 Academy of Sciences.
13 National Research Council. (2002) Estimating the public health benefits of proposed air pollution regulations.
14 Washington, DC: National Academy of Sciences. Available at:
15 http://books.nap.edi^ooks/0309086094/html/index.html [7 November, 2002].
16 National Research Council. (2003) Air emissions from animal feeding operations: current knowledge and future
17 needs. Available: http://www.nap.edu/books/0309087058/html/ (5 February 2004).
18 Neff, J. C.; Holland, E. A.; Dentener, F. J.; McDowell, W. H.; Russell, K. M. (2002) Atmospheric organic nitrogen:
19 implications for the global N cycle. Biogeochemistry 57/58: 99-136.
20 Neuman, J. A.; Huey, L. G.; Dissly, R. W.; Fehsenfeld, F. C.; Flocke, F.; Holecek, J. C.; Holloway, J. S.; Htibler, G.;
21 Jakoubek, R.; Nicks, D. K., Jr.; Parrish, D. D.; Ryerson, T. B.; Sueper, D. T.; Weinheimer, A. J. (2002)
22 Fast-response airborne in-situ measurements of HNO3 during the Texas Air Quality Study. J. Geophys. Res.
23 (Atmos.) 107: 10.1029/2001JD001437.
24 Newberg, J.; Williams, D.; Chu, G.; Matthew, B.; Anastasio, C. (2000) Photoformation of hydroxyl radical in
25 sea-salt particles. In: Preprints of Extended Abstracts presented at the 219th American Chemical Society
26 National Meeting; March; San Francisco, CA. Columbus, OH: American Chemical Society; ENVR 176.
27 Newchurch, M. J.; Ayoub, M. A.; Oltmans, S.; Johnson, B.; Schmidlin, F. J. (2003) Vertical distribution of ozone at
28 four sites in the United States. J. Geophys. Res. 108(D1): 10.1029/2002JD002059.
29 Newell, R. E.; Thouret, V.; Cho, J. Y. N.; Stoller, P.; Marenco, A.; Smit, H. G. (1999) Ubiquity of quasi-horizontal
30 layers in the troposphere. Nature (London) 398: 316-319.
31 Nordmeyer, T.; Wang, W.; Ragains, M. L.; Finlayson-Pitts, B. J.; Spicer, C. W.; Plastridge, R. A. (1997) Unique
32 products of the reaction of isoprene with atomic chlorine: potential markers. Geophys. Res. Lett.
33 24: 1615-1618.
34 North American Research Strategy for Tropospheric Ozone (NARSTO) Synthesis Team. (2000) An assessment of
3 5 tropospheric ozone pollution: a North American perspective. Palo Alto, CA: Electric Power Research
36 Institute.
37 Notholt, J.; Hjorth, J.; Raes, F. (1992a) Formation of HNO2 on aerosol surfaces during foggy periods in the presence
38 of NO and NO2. Atmos. Environ. Part A 26: 211-217.
39 Notholt, J.; Hjorth, J.; Raes, F.; Schrems, O. (1992b) Simultaneous long path field measurements of HNO2, CH2O
40 and aerosol. Ber. Bunsen Ges. Phys. Chem. 96: 290-293.
41 Nunnermacker, L. J.; Imre, D.; Daum, P. H.; Kleinman, L.; Lee, Y.-N.; Lee, J. H.; Springston, S. R.; Newman, L.;
42 Weinstein-Lloyd, J.; Luke, W. T.; Banta, R.; Alvarez, R.; Senff, C.; Sillman, S.; Holdren, M.; Keigley, G. W.;
43 Zhou, X. (1998) Characterization of the Nashville urban plume on July 3 and July 18, 1995. J. Geophys. Res.
44 103(D21): 28,129-28,148.
45 Nunnermacker, L. J.; Kleinman, L. I.; Imre, D.; Daum, P. H.; Lee, Y.-N.; Lee, J. H.; Springston, S. R.; Newman, L.;
46 Gillani, N. (2000) NOy lifetimes and O3 production efficiencies in urban and power plant plumes: analysis of
47 field data. J. Geophys. Res. [Atmos.] 105: 9165-9176.
48 O'Brien, R. J.; Green, P. J.; Nguyen, N.-L.; Doty, R. A.; Dumdel, B. E. (1983) Carbon balances in simulated
49 atmospheric reactions: aromatic hydrocarbons. Environ. Sci. Technol. 17: 183-186.
50 Odman, M. T.; Ingram, C. I. (1996) Multiscale air quality simulation platform (MAQSIP): source code
51 documentation and validation. Research Triangle Park, NC: MCNC; technical report ENV-96TR002-vl.O.
52 Available: http://www.ce.gatech.edu/~todman/maqsip.pdf (9 September 2003).
53 Odman, M. T.; Russell, A. G. (1999) Mass conservative coupling of non-hydrostatic meteorological models with air
54 quality models. In: Crying, S.-E.; Batchvarova, E., eds. Air Pollution Modeling and its Application XIII.
5 5 New York, NY: Plenum Press.
56 Odum, J. R.; Hoffmann, T.; Bowman, F.; Collins, D.; Flagan, R. C.; Seinfeld, J. H. (1996) Gas/particle partitioning
57 and secondary organic aerosol yields. Environ. Sci. Technol. 30: 2580-2585.
January 2005 AX2-177 DRAFT-DO NOT QUOTE OR CITE
-------
1 Olsen, M. A.; Stanford, J. L. (2001) Evidence of stratosphere-to-troposphere transport within a mesoscale model and
2 total ozone mapping spectrometer total ozone. J. Geophys. Res. [Atmos.] 106: 27,323-27,334.
3 Olsen, M. A.; Douglass, A. R.; Schoeberl, M. R. (2002) Estimating downward cross-tropopause flux using column
4 ozone and potential vorticity. J. Geophys. Res. 107(D22): 10.1029/2001JD002041.
5 Olszyna, K. J.; Bailey, E. M.; Simonaitis, R.; Meagher, J. F. (1994) O3 and NOy relationships at a rural site.
6 J. Geophys. Res. [Atmos.] 99: 14,557-14,563.
7 Oltmans, S. J.; Levy, H., II. (1992) Seasonal cycle of surface ozone over the North Atlantic. Nature 358: 392-394.
8 Oltmans, S. J.; Levy, H., II; Harris, J. M.; Merrill, J. T.; Moody, J. L.; Lathrop, J. A.; Cuevas, E.; Trainer, M.;
9 O'Neill, M. S.; Prospero, J. M.; Vomel, H.; Johnson, B. J. (1996) Summer and spring ozone profiles over the
10 North Atlantic from ozonesonde observations. J. Geophys. Res. 101: 29,179-29,200.
11 Oltmans, S. J.; Lefohn, A. S.; Scheel, H. E.; Harris, J. M.; Levy, H., II; Galbally, I. E.; Brunke, E.-G.; Meyer, C. P.;
12 Lathrop, J. A.; Johnson, B. J.; Shadwick, D. S.; Cuevas, E.; Schmidlin, F. J.; Tarasick, D. W.; Claude, H.;
13 Kerr, J. B.; Uchino, O.; Mohnen, V. (1998) Trends of ozone in the troposphere. Geophys. Res. Lett.
14 25: 139-142.
15 Orlando, J. J.; Tyndall, G. S.; Paulson, S. E. (1999) Mechanism of the OH-initiated oxidation of methacrolein.
16 Geophys. Res. Lett. 26: 2191-2194.
17 Padro, J. (1996) Summary of ozone dry deposition velocity measurements and model estimates over vineyard,
18 cotton, grass and deciduous forest in summer. Atmos. Environ. 30: 2363-2369.
19 Palmer, P. I.; Jacob, D. J.; Fiore, A. M.; Martin, R. V.; Chance, K.; Kuruso, T. P. (2003) Mapping isoprene
20 emissions over North America using formaldehyde column observations from space. J. Geophys. Res.
21 108(D6): 10.1029/2002JD002153.
22 Pandis, S. N.; Paulson, S. E.; Seinfeld, J. H.; Flagan, R. C. (1991) Aerosol formation in the photooxidation of
23 isoprene and p-pinene. Atmos. Environ. Part A 25: 997-1008.
24 Pandis, S. N.; Harley, R. A.; Cass, G. R.; Seinfeld, J. H. (1992) Secondary organic aerosol formation and transport.
25 Atmos. Environ. Part A 26: 2269-2282.
26 Papagni, C.; Arey, J.; Atkinson, R. (2001) Rate constants for the gas-phase reactions of OH radicals with a series of
27 unsaturated alcohols. Int. J. Chem. Kinet. 33: 142-147.
28 Park, R. J.; Stenchikov, G. L.; Pickering; Dickerson, R. R.; Allen, D. J.; Kondragunta, S. (2001) Regional air
29 pollution and its radiative forcing: studies with a single column chemical and radiation transport model.
30 J. Geophys. Res. [Atmos.] 106: 28,751-28,770.
31 Parrish, D. D.; Fehsenfeld, F. C. (2000) Methods for gas-phase measurements of ozone, ozone precursors and
32 aerosol precursors. Atmos. Environ. 34: 1921-1957.
33 Parrish, D. D.; Trainer, M.; Buhr, M. P.; Watkins, B. A.; Fehsenfeld, F. C. (1991) Carbon monoxide concentrations
34 and their relatin to concentrations of total reactive oxided nitrogen at two rural U.S. sites. J. Geophys. Res.
35 (Atmos.) 96: 9309-9320.
36 Parrish, D. D.; Hahn, C. J.; Williams, E. J.; Norton, R. B.; Fehsenfeld, F. C.; Singh, H. B.; Shelter, J. D.; Gandrud,
37 B. W.; Ridley, B. A. (1992) Indications of photochemical histories of Pacific air masses from measurements
38 of atmospheric trace species at Point Arena, California. J. Geophys. Res. [Atmos.] 97: 15,883-15,901.
39 Parrish, D. D.; Hahn, C. J.; Williams, E. J.; Norton, R. B.; Fehsenfeld, F. C.; Singh, H. B.; Shelter, J. D.; Gandrud,
40 B. W.; Ridley, B. A. (1993) Reply [to Comment on "Indications of photochemical histories of Pacific air
41 masses from measurements of atmospheric trace species at Point Arena, California" by D. D. Parrish et al.].
42 J. Geophys. Res. (Atmos.) 98(D8): 14,995-14,997.
43 Parrish, D. D.; Trainer, M.; Young, V.; Goldan, P. D.; Kuster, W. C.; Jobson, B. T.; Fehsenfeld, F. C.; Lonneman,
44 W. A.; Zika, R. D.; Farmer, C. T.; Riemer, D. D.; Rodgers, M. O. (1998) Internal consistency tests for
45 evaluation of measurements of anthropogenic hydrocarbons in the troposphere. J. Geophys. Res. (Atmos.)
46 103(017): 22,339-22,359.
47 Parrish, D. D.; Ryerson, T. B.; Holloway, J. S.; Trainer, M.; Fehsenfeld, F. C. (1999) Does pollution increase or
48 decrease tropospheric ozone in winter-spring? Atmos. Environ. 33: 5147-5149.
49 Parrish, D. D.; Holloway, J. S.; Jakoubek, R.; Trainer, M.; Ryerson, T. B.; Htibler, G.; Fehsenfeld, F. C.; Moody,
50 J. L.; Cooper. O. R. (2000) Mixing of anthropogenic pollution with stratospheric ozone: a case study from the
51 North Atlantic wintertime troposphere. J. Geophys. Res. [Atmos.] 105: 24,363-24,374.
52 Parrish, D. D.; Trainer, M.; Hereid, D.; Williams, E. J.; Olszyna, K. J.; Harley, R. A.; Meagher, J. F.; Fehsenfeld,
53 F. C. (2002) Decadal change in carbon monoxide to nitrogen oxide ratio in U.S. vehicular emissions J.
54 Geophys. Res. (Atmos.) 107(D13): 10.1029/2001JD000720.
55 Paul, J. W.; Beauchamp, E. G. (1993) Nitrogen availability for corn in soils amended with urea, cattle slurry, and
56 solid and composted manures. Can. J. Soil Sci. 73: 253-266.
January 2005 AX2-178 DRAFT-DO NOT QUOTE OR CITE
-------
1 Paul, J. W.; Beauchamp, E. G.; Zhang, X. (1993) Nitrous and nitric oxide emissions during nitrification and
2 denitrification from manure-amended soil in the laboratory. Can. J. Soil Sci. 73: 539-553.
3 Penner, J. E. (1995) Carbonaceous aerosols influencing atmospheric radiation: black and organic carbon. In:
4 Charlson, R. J.; Heintzenberg, J., eds. Aerosol Forcing of Climate. New York, NY: John Wiley and Sons;
5 pp. 91-108.
6 Penrose, W. R..; Pan, L.; Stetter, J. R.; Ollison, W. M. (1995) Sensitive measurement of ozone using amperometric
7 gas sensors. Anal. Chim. Acta 313: 209-219.
8 Perry, R. A.; Atkinson, R.; Pitts, J. N., Jr. (1977) Kinetics and mechanism of the gas phase reaction of OH radicals
9 with aromatic hydrocarbons over the temperature range 296-473 K. J. Phys. Chem. 81: 296-304.
10 Petriconi, G. L.; Papee, H. M. (1972) On the photolytic separation of halogens from sea water concentrates. Water
11 Air Soil Pollut. 1: 117-131.
12 Pickering, K. E.; Dickerson, R. R.; Huffman, G. J.; Boatman, J. F.; Schanot, A. (1988) Trace gas transport in the
13 vicinity of frontal convective clouds. J. Geophys. Res. 93: 759-773.
14 Pickering, K. E.; Thompson, A. M.; Dickerson, R. R.; Luke, W. T.; McNamara, D. P.; Greenberg, J. P.;
15 Zimmerman, P. R. (1990) Model calculations of tropospheric ozone production potential following observed
16 convective events. J. Geophys. Res. [Atmos.] 95: 14,049-14,062.
17 Pickering, K. E.; Thompson, A. M.; Scala, J. R.; Tao, W.-K.; Simpson, J.; Garstang, M. (1991) Photochemical ozone
18 production in tropical squall line convection during NASA Global Tropospheric Experiment/Amazon
19 Boundary Layer Experiment 2A. J. Geophys. Res. 96: 3099-3114.
20 Pickering, K. E.; Thompson, A. M.; Scala, J. R.; Tao, W.-K.; Simpson, J. (1992a) Ozone production potential
21 following convective redistribution of biomass burning emissions. J. Atmos. Chem. 14: 297-313.
22 Pickering, K. E.; Thompson, A. M.; Scala, J. R.; Tao, W.-K.; Dickerson, R. R.; Simpson, J. (1992b) Free
23 tropospheric ozone production following entrainment of urban plumes into deep convection. J. Geophys. Res.
24 97(016): 17985-18,000.
25 Pickering, K. E.; Scala, J. R.; Thompson, A. M.; Tao, W.-K.; Simpson, J. (1992c) A regional estimate of the
26 convective transport of CO from biomass burning. Geophys. Res. Lett. 19: 289-292.
27 Pickering, K. E.; Thompson, A. M.; Tao, W.-K.; Kucsera, T. L. (1993) Upper tropospheric ozone production
28 following mesoscale convection during STEP/EMEX. J. Geophys. Res. 98: 8737-8749.
29 Pickering, K. E.; Thompson, A. M.; Tao, W.-K.; Rood, R. B.; McNamara, D. P.; Molod, A. M. (1995) Vertical
30 transport by convective clouds: comparisons of three modeling approaches. Geophys. Res. Lett. 22:
31 1089-1092.
32 Pickering, K. E.; Thompson, A. M.; Wang, Y.; Tao, W.-K.; McNamara, D. P.; Kirchhoff, W. J. H.; Heikes, B. G.;
33 Sachse, G. W.; Bradshaw, J. D.; Gregory, G. L.; Blake, D. R. (1996) Convective transport of biomass burning
34 emissions over Brazil during TRACE A. J. Geophys. Res. [Atmos.] 101: 23,993-24,012.
35 Pickering, K. E.; Wang, Y.; Tao, W.-K.; Price, C.; Mtiller, J.-F. (1998) Vertical distributions of lightning NOX for
36 use in regional and global chemical transport models. J. Geophys. Res. [Atmos.] 103: 31,203-31,216.
37 Pickering, K. E.; Thompson, A. M.; Kim, H.; DeCaria, A. J.; Pfister, L.; Kucsera, T. L.; Witte, J. C.; Avery, M. A.;
38 Blake, D. R.; Crawford, J. H.; Heikers, B. G.; Sachse, G. W.; Sandholm, S. T.; Talbot, R. W. (2001) Trace
39 gas transport and scavenging ini PEM-Tropics B South Pacific Convergence Zone convection. J. Geophys.
40 Res. [Atmos.] 106: 32,591-32,602.
41 Pierce, T. E.; Waldruff, P. S. (1991) PC-BEIS: a personal computer version of the biogenic emissions inventory
42 system. J. Air Waste Manage. Assoc. 41: 937-941.
43 Pierce, T; Geron, C.; Bender, L.; Dennis, R.; Tonnesen, G.; Guenther, A. (1998) Influence of increased isoprene
44 emissions on regional ozone modeling. J. Geophys. Res. (Atmos.) 103: 25,611-25,630.
45 Pierson, W. R.; Gertler, A. W.; Bradow, R. L. (1990) Comparison of the SCAQS tunnel study with historical data.
46 Presented at: 83rd annual meeting & exhibition of the Air and Waste Management Association; June;
47 Pittsburgh, PA. Pittsburgh, PA: Air and Waste Management Association; paper no. 90-175.3.
48 Pinto, J. P.; Bruhl, C.; Thompson, A. M. (1993) The current and future envirionmental role of atmospheric methane.
49 In: Khalil, M. A. K., ed. Atmospheric methane sources, sinks, and role in global change, p. 514-531. (NATO
50 ASI Series, v. 113).
51 Piotrowicz, S. R.; Bezdek, H. F.; Harvey, G. R.; Springer-Young, M.; Hanson, K. J. (1991) On the ozone minimum
52 over the equatorial Pacific Ocean. J. Geophys. Res. 96: 18,679-18,687.
53 Pippin, M. R.; Bertman, S.; Thornberry, T.; Town, M.; Carroll, M. A.; Sillman, S. (2001) Seasonal variation of
54 PAN, PPN, MPAN and O3 at the upper midwest PROPHET site. J. Geophys. Res. (Atmos.)
55 106:24,451-24,463.
January 2005 AX2-179 DRAFT-DO NOT QUOTE OR CITE
-------
1 Pitari, G.; Grassi, B.; Visconti, G. (1997) Results of a chemical transport model with interactive aerosol
2 microphysics. In: Bojkov, R.; Visconti, G., eds. Proceedings of the XVIII Quadrennial Ozone Symposium;
3 September 1996; L'Aquila, Italy, pp. 759-762.
4 Pitts, J. N., Jr.; Sanhueza, E.; Atkinson, R.; Carter, W. P. L.; Winer, A. M; Harris, G. W.; Plum, C. N. (1984a)
5 An investigation of the dark formation of nitrous acid in environmental chambers. Int. J. Chem. Kinet.
6 16:919-939.
7 Pitts, J. N., Jr.; Biermann, H. W.; Atkinson, R.; Winer, A. M. (1984b) Atmospheric implications of simultaneous
8 nighttime measurements of NO3 radicals and HONO. Geophys. Res. Lett. 11: 557-560.
9 Placet, M.; Mann, C. O.; Gilbert, R. O.; Niefer, M. J. (2000) Emissions of ozone precursors from stationary sources:
10 a critical review. Atmos. Environ. 34: 2183-2204.
11 Plummer, D. A.; McConnell, J. C.; Shepson, P. B.; Hastie, D. R.; Niki, H. (1996) Modeling of ozone formation at a
12 rural site in southern Ontario. Atmos. Environ. 30: 2195-2217.
13 Pokharel, S. S.; Bishop, G. A.; Stedman, D. H. (2002) An on-road motor vehicle emissions inventory for Denver: an
14 efficient alternative to modeling. Atmos. Environ. 36: 5177-5184.
15 Poppe, D.; Wallasch, M.; Zimmermann, J. (1993) The dependence of the concentration of OH on its precursors
16 under moderately polluted conditions: a model study. J. Atmos. Chem. 16: 61-78.
17 Poppe, D.; Zimmermann, J.; Bauer, R.; Brauers, T.; Briining, D.; Callies, J.; Dorn, H.-P.; Hofzumahaus, A.; Johnen,
18 F.-J.; Khedim, A.; Koch, H.; Koppmann, R.; London, H.; Mtiller, K.-P.; Neuroth, R.; Plass-Dtilmer, C.;
19 Platt, U.; Rohrer, F.; Roth, E.-P.; Rudolph, J.; Schmidt, U.; Wallasch, M.; Ehhalt, D. H. (1994) Comparison
20 of measured OH concentrations with model calculations. J. Geophys. Res. [Atmos.] 99: 16,633-16,642.
21 Poppe, D.; Zimmerman, J.; Dorn, H. P. (1995) Field data and model calculations for the hydroxyl radical. J. Atmos.
22 Sci. 52: 3402-3407.
23 Poulida, O.; Dickerson, R. R.; Heymsfield, A. (1996) Troposphere-stratosphere exchange in a midlatitude mesoscale
24 convective complex: 1. Observations. J. Geophys. Res. [Atmos.] 101: 6823-6836.
25 Prados, A. I. (2000) The export of ozone and other atmospheric trace species from North America: observations and
26 numerical simulations [dissertation]. College Park, MD: The University of Maryland.
27 Prados, A. I.; Dickerson, R. R.; Doddridge, B. G.; mime, P. A.; Moody, J. L.; Merrill, J. T. (1999) Transport of
28 ozone and pollutants from North America to the North Atlantic Ocean during the 1996 AEROCE intensive
29 experiment. J. Geophys. Res. 104: 26,219-26,233.
30 Prather, M.; Ehhalt, D. (2001) Atmospheric chemistry and greenhouse gases. In: Houghton, J. T.; Ding, Y.; Griggs,
31 D. J.; Noguer, M.; van der Linden, P. J.; Dai, X.; Maskell, K.; Johnson, C. A., eds. Climate Change 2001:
32 The Scientific Basis; pp. 239-287. Cambridge, MA: Cambridge University Press. Available:
33 http://www.grida.no/climate/ipcc_tar/wgl/ (20 January 2004).
34 Prather, M. J.; Jacob, D. J. (1997) A persistent imbalance in HOX and NOX photochemistry in the upper troposphere
3 5 driven by deep tropical convection. Geophys. Res. Lett. 24: 3189-3192.
36 Prenni, A. J.; DeMott, P. J.; Kreidenweis, S. M. (2003) Water uptake of internally mixed particles containing
37 ammomium sulfate and dicarboxylic acids. Atmos. Environ. 37: 4243-4251.
38 Prevot, A. S. H.; Staehelin, J.; Kok, G. L.; Schillawski, R. D.; Neininger, B.; Staffelbach, T.; neftel, A.; Wernli, H.;
39 Dommen, J. (1997) The Milan photooxidant plume. J. Geophys. Res. (Atmos.) 102: 23,375-23,388.
40 Price, J. D.; Vaughan, G. (1993) The potential for stratosphere-troposphere exchange in cut-off-low systems. Q. J. R.
41 Meteorol. Soc. 119:343-365.
42 Price, C.; Penner, J.; Prather, M. (1997) NOX from lightning: 1. Global distribution based on lightning physics. J.
43 Geophys. Res. [Atmos.] 102: 5929-5941.
44 Pszenny, A. A. P.; Keene, W. C.; Jacob, D. J.; Fan, S.; Maben, J. R.; Zetwo, M. P.; Springer-Young, M.; Galloway,
45 J. N. (1993) Evidence of inorganic chlorine gases other than hydrogen chloride in marine surface air.
46 Geophys. Res. Lett. 20: 699-702.
47 Pszenny, A. A. P.; Moldanova, J.; Keene, W. C.; Sander, R.; Maben, J. R.; Martinez, M.; Crutzen, P. J.; Perner, D.;
48 Prinn, R. G. (2003) Halogen cycling and aerosol pH in the Hawaiian marine boundary layer. Atmos. Chem.
49 Phys. Disc. 3:4701-4753.
50 Pun, B. K.; Seigneur, C.; Grosjean, D.; Saxena, P. (2000) Gas-phase formation of water-soluble organic compounds
51 in the atmosphere: a retrosynthetic analysis. J. Atmos. Chem. 35: 199-223.
52 Quigley, D. C.; Hornafius, J. S.; Luyendyk, B. P.; Francis, R. D.; Clark, J.; Washburn, L. (1999) Decrease in natural
53 marine hydrocarbon seepage near Coal Oil Point, California, associated with offshore oil production. Geol.
54 27: 1047-1050.
55 Rancher, J.; Kritz, M. A. (1980) Diurnal fluctuations of Br and I in the tropical marine atmosphere. J. Geophys. Res.
56 85: 5581-5587.
January 2005 AX2-180 DRAFT-DO NOT QUOTE OR CITE
-------
1 Rao, S. T.; Zurbenko, I. G. (1994) Detecting and tracking changes in ozone air quality. J. Air Waste Manage. Assoc.
2 44: 1089-1092.
3 Rao, S. T.; Zurbenko, I. G.; Neagu, R.; Porter, P. S.; Ku, J. Y.; Henry, R. F. (1997) Space and time scales in ambient
4 ozone data. Bull. Am. Meteorol. Soc. 78: 2153-2166.
5 Rao, S. T.; Ku, J. Y.; Herman, S. (2003) Summertime characteristics of the atmospheric boundary layer and
6 relationships to ozone levels over the eastern United States. Pure Appl. Geophys. 160: 21-55.
7 Reed, R. J. (1955) A study of a characteristic type of upper-level frontogenesis. J. Meteorol. 12: 226-237.
8 Regener, E. (1941) Ozonschicht und atmospharische turbulenz. Ber. Dtsch. Wetterdienstes US-Zone No. 11: 45-57.
9 Reid, J. P.; Sayer, R. M. (2003) Heterogeneous atmospheric aerosol chemistry: laboratory studies of chemistry on
10 water droplets. Chem. Soc. Rev. 32: 70-79.
11 Reisinger, A. R. (2000) Unidentified interference inDOAS measurements of ozone. Appl. Spectrosc. 54: 72-79.
12 Reissell, A.; Arey, J.; Atkinson, R. (2001) Atmospheric chemistry of camphor. Int. J. Chem. Kinet. 33: 56-63.
13 Reitebuch, O.; Strassburger, A.; Emeis, S.; Kuttler, W. (2000) Nocturnal secondary ozone concentration maxima
14 analysed by sodar observation and surface measurements. Atmos. Environ. 34: 4315-4329.
15 Reiter, E. R. (1963) A case study of radioactive fallout. J. Appl. Meteorol. 2: 691-705.
16 Reiter, E. R. (1975) Stratospheric-tropospheric exchange processes. Rev. Geophys. Space Phys. 13: 459-474.
17 Reiter, E. R.; Mahlman, J. D. (1965) Heavy radioactive fallout over the southern United States, November 1962.
18 J. Geophys. Res. 70: 4501-4520.
19 Reithmeier, C.; Sausen, R. (2002) ATTILA: atmospheric tracer transport in a Lagrangian model. Tellus
20 54B: 278-299.
21 Ren, X. R.; Harder, H.; Martinez, M.; Lesher, R. L.; Oliger, A.; Shirley, T.; Adams, J.; Simpas, J. B.; Brune, W. H.
22 (2003) HOx concentrations and OH reactivity observations in New York City during PMTACS-NY2001.
23 Atmos. Environ. 37: 3627-3637.
24 Reynolds, S.; Michaels, H.; Roth, P.; Tesche, T. W.; McNally, D.; Gardner, L.; Yarwood, G. (1996) Alternative base
25 cases in photochemical modeling: their construction, role, and value. Atmos. Environ. 30: 1977-1988.
26 Ridley, B. A.; Carroll, M. A.; Dunlap, D. D.; Trainer, M.; Sachse, G. W.; Gregory, G. L.; Condon, E. P. (1989)
27 Measurements of NOX over the eastern Pacific ocean and southwestern United States during the spring 1984
28 NASA GTE aircraft program. J. Geophys. Res. [Atmos.] 94: 5043-5067.
29 Ridley, B. A.; Walega, J. G.; Lamarque, J.-F.; Grahek, F. E.; Trainer, M.; Htibler, G.; Lin, X.; Fehsenfeld, F. C.
30 (1998) Measurements of reactive nitrogen and ozone to 5-km altitude in June 1990 over the southeastern
31 United States. J. Geophys. Res. 103: 8369-8388.
32 Riemer, D. D.; Apel, E. C. (2001) Confirming the presence and extent of oxidation by Cl in the Houston, Texas
3 3 urban area using specific isoprene oxidation products as tracers. Miami, FL: University of Miami, Rosenstiel
34 School of Marine and Atmospheric Science. Available:
35 http://clearinghouse.ces.utexas.edu/TexAQS/DB_Files/ConfirmingPresenceandExtentOfOxidationByCI.pdf
36 (5 February 2004).
37 Riemer, N.; Vogel, H.; Vogel, B.; Schell, B.; Ackermann, L; Kessler, C.; Hass, H. (2003) Impact of the
3 8 heterogeneous hydrolysis of N2O5 on chemistry and nitrate aerosol formation in the lower troposphere under
39 photosmog conditions. J. Geophys. Res. (Atmos.) 108(D4): 10.1029/2002JD002436.
40 Roberts, J. M.; Williams, J.; Baumann, K.; Buhr, M. P.; Goldan, P. D.; Holloway, J.; Htibler, G.; Kuster, W. C.;
41 McKeen, S. A.; Ryerson, T. B.; Trainer, M.; Williams, E. J.; Fehsenfeld, F. C.; Bertman, S. B.; Nouaime, G.;
42 Seaver, C.; Grodzinsky, G.; Rodgers, M.; Young, V. L. (1998) Measurements of PAN, PPN, and MPAN
43 made during the 1994 and 1995 Nashville Intensives of the Southern Oxidant Study: implications for regional
44 ozone production from biogenic hydrocarbons. J. Geophys. Res. [Atmos.] 103: 22,473-22,490.
45 Robinson, G. N.; Worsnop, D. R.; Jayne, J. T.; Kolb, C. E.; Davidovits, P. (1997) Heterogeneous uptake of C1ONO2
46 and N2O5 by sulfuric acid solutions. J. Geophys. Res. (Atmos.) 102: 3583-3601.
47 Rodgers, M. O.; Davis, D. D. (1989) A UV-photofragmentation/laser-induced fluorescence sensor for the
48 atmospheric detection of HONO. Environ. Sci. Technol. 23: 1106-1112.
49 Roelofs, G. J.; Scheeren, H. A.; Heland, J.; Ziereis, H.; Lelieveld, J. (2003) A model study of ozone in the eastern
50 Mediterranean free troposphere during MINOS (August 2001). Atmos. Chem. Phys. 3: 1199-1210.
51 Rogers, J. D. (1990) Ultraviolet absorption cross sections and atmospheric photodissociation rate constants of
52 formaldehyde. J. Phys. Chem. 94: 4011-4015.
53 Roselle, S. J. (1994) Effects of biogenic emission uncertainties on regional photochemical modeling of control
54 strategies. Atmos. Environ. 28: 1757-1772.
55 Rudich, Y.; Talukdar, R. K.; Fox, R. W.; Ravishankara, A. R. (1996) Rate coefficients for reactions of NO3 with a
56 few olefms and oxygenated olefins. J. Phys. Chem. 100: 5374-5381.
January 2005 AX2-181 DRAFT-DO NOT QUOTE OR CITE
-------
1 Rudolph, I; Ramacher, B.; Plass-Dtilmer, C.; Muller, K.-P.; Koppmann, R. (1997) The indirect determination of
2 chlorine atom concentration in the troposphere from changes in the patterns of non-methane hydrocarbons.
3 Tellus 649:592-601.
4 Ruppert, L.; Becker, K. H. (2000) A product study of the OH radical-initiated oxidation of isoprene: formation of
5 C5-unsaturated diols. Atmos. Environ. 34: 1529-1542.
6 Russell, K. M. (2002) Characterization and dry deposition of atmospheric nitrogen at the mid-Atlantic United States
7 coast [dissertation]. Richmond, VA: University of Virginia. Available from: University Microfilms, Ann
8 Arbor, MI; AADAA-13044880.
9 Russell, A.; Dennis, R. (2000) NARSTO critical review of photochemical models and modeling. Atmos. Environ.
10 34:2283-2324.
11 Ryan, W. F.; Doddridge, B. G.; Dickerson, R. R.; Morales, R. M.; Hallock, K. A.; Roberts, P. T.; Blumenthal, D. L.;
12 Anderson, J. A.; Civerolo, K. (1998) Pollutant transport during a regional O3 episode in the mid-Atlantic
13 states. J. Air Waste Manage. Assoc. 48: 786-797.
14 Ryerson, T. B.; Buhr, M. P.; Frost, G. J.; Goldan, P. D.; Holloway, J. S.; Htibler, G.; Jobson, B. T.; Kuster, W. C.;
15 McKeen, S. A.; Parrish, D. D.; Roberts, J. M.; Sueper, D. T.; Trainer, M.; Williams, J.; Fehsenfeld, F. C.
16 (1998) Emissions lifetimes and ozone formation in power plant plumes. J. Geophys. Res.
17 103(017): 22,569-22,583.
18 Ryerson, T. B.; Williams, E J.; Fehsenfeld, F. C. (2000) An efficient photolysis system for fast-response NO2
19 measurements. J. Geophys. Res. (Atmos.) 105(D21): 26,447-26,461.
20 Ryerson, T. B.; Trainer, M.; Holloway, J. S.; Parrish, D. D.; Huey, L. G.; Sueper, D. T.; Frost, G. J.; Donnelly,
21 S. G.; Schauffler, S.; Atlas, E. L.; Kuster, W. C.; Goldan, P. D.; Htibler, G.; Meagher, J. F.; Fehsenfeld, F. C.
22 (2001) Observations of ozone formation in power plant plumes and implications for ozone control strategies.
23 Science (Washington, DC) 292: 719-723.
24 Saathoff, H.; Naumann, K. H.; Riemer, N.; Kamm, S.; Mohler, O.; Schurath, U.; Vogel, H.; Vogel, B. (2001) The
25 loss of NO2, HNO3,,NO3/N2O5 and HO2/HOONO2 on soot aerosol: a chamber and modeling study. Geophys.
26 Res. Lett. 28: 1957-1960.
27 Sakamaki, F.; Hatakeyama, S.; Akimoto, H. (1983) Formation of nitrous acid and nitric oxide in the heterogeneous
28 dark reaction of nitrogen dioxide and water vapor in a smog chamber. Int. J. Chem. Kinet. 15: 1013-1029.
29 Sakugawa, H.; Kaplan, I. R. (1989) H2O2 and O3 in the atmosphere of Los Angeles and its vicinity: factors
30 controlling their formation and their role as oxidants of SO2. J. Geophys. Res. [Atmos.] 94: 12,957-12,973.
31 Salmond, J. A.; McKendry, I. G. (2002) Secondary ozone maxima in a very stable nocturnal boundary layer:
32 observations from the Lower Fraser Valley, BC. Atmos. Environ. 36: 5771-5782.
33 Samson, P. J.; Shi, B. (1988) A meteorological investigation of high ozone values in American cities. Ann Arbor,
34 MI: University of Michigan, Space Physics Research Laboratory, Department of Atmospheric, Oceanic, and
3 5 Space Sciences.
36 Sander, R.; Crutzen, P. J. (1996) Model study indicating halogen activation and ozone destruction in polluted air
37 masses transported to the sea. J. Geophys. Res. [Atmos.] 101: 9121-9138.
38 Sander, R.; Rudich, Y.; Von Glasow, R.; Crutzen, P. J. (1999) The role of BrNO3 in marine tropospheric chemistry:
39 a model study. Geophys. Res. Lett. 26: 2858-2860.
40 Sander, R.; Keene, W. C.; Pszenny, A. A. P.; Arimoto, R. Ayers, G. P.; Chainey, J. M.; Crutzen, P. J.; Duce, R. A.;
41 Huebert, B. J.; Maenhaut, W.; Turekian, V. C.; VanDingenan, R. (2003) Inorganic bromine in the marine
42 boundary layer: a critical review. Atmos. Chem. Phys. 3: 1301-1336.
43 Santer, B. D.; Wehner, M. F.; Wigley, T. M. L.; Sausen, R.; Meehl, G. A.; Taylor, K. E.; Ammann, C.; Arblaster, J.;
44 Washington, W. M.; Boyle, J. S.; Bruggemann, W. (2003) Contributions of anthropogenic and natural forcing
45 to recent tropopause height changes. Science 301: 479-483.
46 Sarwar, G.; Corsi, R.; Kumura, Y.; Allen, D.; Weschler, C. J. (2002) Hydroxyl radicals in indoor environments.
47 Atmos. Environ. 36: 3973-3988.
48 Sauer, F.; Schaefer, C.; Neeb, P.; Horie, O.; Moortgat, G. (1999) Formation of hydrogen peroxide in the ozonolysis
49 of isoprene and simple alkenes. Atmos. Environ. 33: 229-241.
50 Sawyer, R. F.; Harley, R. A.; Cadle, S. H.; Norbeck, J. M.; Slott, R.; Bravo, H. A. (2000) Mobile sources critical
51 review: 1998 NARSTO assessment. Atmos. Environ. 34: 2161-2181.
52 Saxena, P.; Hildemann, L. M.; McMurry, P. H.; Seinfeld, J. H. (1995) Organics alter hygroscopic behavior of
53 atmospheric particles. J. Geophys. Res. [Atmos.] 100: 18,755-18,770.
54 Saylor, R. D. (1997) An estimate of the potential significance of heterogeneous loss to aerosols as an additional sink
55 for hydroperoxy radicals in the troposphere. Atmos. Environ. 31: 3653-3658.
January 2005 AX2-182 DRAFT-DO NOT QUOTE OR CITE
-------
1 Scala, J. M; Garstang, M; Tao, W.-K.; Pickering, K. E.; Thompson, A. M.; Simpson, I; Kirchhoff, V. W. J. H.;
2 Browell, E. V.; Sachse, G. W.; Torres. A. L.; Gregory, G. L.; Rasmussen, R. W.; Khalil, M. A. K. (1990)
3 Cloud draft structure and trace gas transport. J. Geophys. Res. 95: 17,015-17,030.
4 Schere, K. L.; Wayland, R. A. (1989) Development and evaluation of the regional oxidant model forthe northeastern
5 United States. In: Schneider, T.; Lee, S. D.; Wolters, G. J. R.; Grant, L. D., eds. Atmospheric ozone research
6 and its policy implications: proceedings of the 3rd US-Dutch international symposium; May 1988; Nijmegen,
7 The Netherlands. Amsterdam, The Netherlands: Elsevier Science Publishers; pp. 613-622. (Studies in
8 environmental science 35).
9 Schere, K. L.; Wayland, R. A. (1989) EPA regional oxidant model (ROM2.0): evaluation on 1980 NEROS data
10 bases. Research Triangle Park, NC: U.S. Environmental Protection Agency, Atmospheric Research and
11 Exposure Assessment Laboratory; report no. EPA/600/3-89/057. Available from: NTIS, Springfield, VA;
12 PB89-200828/HSU.
13 Schichtel, B. A.; Husar, R. B. (1998) Eastern U.S. transport climatology during average, high and low ozone days.
14 Online paper 98-A884. Available: http://capita.wustl.edu/capita/CapitaReports (30 October 2003).
15 Schichtel, B. A.; Husar, R. B. (2001) Eastern North American transport climatology during high- and low-ozone
16 days. Atmos. Environ. 35: 1029-1038.
17 Schneider, M.; Ballschmiter, K. (1999) C3-C14-alkyl nitrates in remote South Atlantic air. Chemosphere 38: 233-244.
18 Schneider, M.; Luxenhofer, O.; Deissler, A.; Ballschmiter, K. (1998) C{-C15 alkyl nitrates, benzyl nitrate, and
19 bifunctional nitrates: Measurements in California and South Atlantic air and global comparison using C2C14
20 and CHBr3 as marker molecules. Environ. Sci. Technol. 32: 3055-3062.
21 Schubert, S. D.; Rood, R. B.; Pfaendtner, J. (1993) An assimilated dataset for earth science applications. Bull. Am.
22 Meteorol. Soc. 74: 2331-2342.
23 Schultz, M. G.; Jacob, D. J.; Bradshaw, J. D.; Sandholm, S. T.; Dibb, J. E.; Talbot, R. W.; Singh, H. B. (2000)
24 Chemical NOX budget in the upper troposphere over the tropical South Pacific. J. Geophys. Res.
25 105:6669-6679.
26 Schumann, U.; Schlager, H.; Arnold, F.; Ovarlez, J.; Kelder, H.; Hov, 0.; Hayman, G.; Isaksen, I. S. A.; Staehelin,
27 J.; Whitefield, P. D. (2000) Pollution from aircraft emissions in the North Atlantic flight corridor: overview
28 on the POLINAT projects. J. Geophys. Res. 105(D3): 10.1029/1999JD900941.
29 Schutze, M.; Herrmann, H. (2002) Determination of phase transfer parameters for the uptake of HNO3, N2O5 and O3
30 on single aqueous drops. Phys. Chem. Chem. Phys. 4: 60-67.
31 Scudlark, J. R.; Russell, K. M.; Galloway, J. N.; Church, T. M.; Keene, W. C. (1998) Dissolved organic nitrogen in
32 precipitation at the mid-Atlantic U.S. coast—methods evaluation and preliminary data. Atmos. Environ.
33 32: 1719-1728.
34 Seaman, N. L. (2000) Meteorological modeling for air quality assessments. Atmos. Environ. 34: 2231-2259.
35 Seila, R. L.; Lonneman, W. A.; Meeks, S. A. (1989) Determination of C2 to C12 ambient air hydrocarbons in 39 U.S.
36 cities from 1984 through 1986. Research Triangle Park, NC: U.S. Environmental Protection Agency,
37 Atmospheric Research and Exposure Assessment Laboratory; report no. EPA/600/3-89/058.
38 Seinfeld, J. H. (1988) Ozone air quality models: a critical review. JAPCA 38: 616-645.
39 Seinfeld, J. H. (1989) Urban air pollution: state of the science. Science (Washington, DC) 243: 745-752.
40 Seinfeld, J. H.; Pandis, S. N. (1998) Atmospheric chemistry and physics: from air pollution to climate change.
41 New York, NY: John Wiley & Sons, Inc.
42 Senff, C. J.; Hardesty, R. M.; Alvarez, R. J., II; Mayor, S. D. (1998) Airborne lidar characterization of power plant
43 plumes during the 1995 Southern Oxidants Study. J. Geophys. Res. [Atmos.] 103: 31,173-31,189.
44 Seo, K.-H.; Bowman, K. P. (2001) A climatology of isentropic cross-tropopause exchange. J. Geophys. Res.
45 [Atmos.] 106: 28,159-28,172.
46 Seo, K.-H.; Bowman, K. P. (2002) Lagrangian estimate of global stratosphere-troposphere mass exchange.
47 J. Geophys. Res. (Atmos.) 107(D21): 10.1029/2002JD002441.
48 Sexton, K. (1982) Evidence of an additive effect for small city plumes. Presented at: 75th annual meeting of the Air
49 Pollution Control Association; June; New Orleans, LA. Pittsburgh, PA: Air Pollution Control Association;
50 paper no. 82-31.4.
51 Sexton, K.; Westberg, H. (1980) Elevated ozone concentrations measured downwind of the Chicago-Gary urban
52 complex. J. Air Pollut. Control Assoc. 30: 911-914.
53 Sexton, F. W.; Michie, R. A., Jr.; McElroy, F. F.; Thompson, V. L. (1981) Technical assistance document for the
54 calibration and operation of automated ambient non-methane organic compound analyzers. Research Triangle
55 Park, NC: U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory; report no.
56 EPA-600/4-81-015. Available from: NTIS, Springfield, VA; PB82-147406.
January 2005 AX2-183 DRAFT-DO NOT QUOTE OR CITE
-------
1 Sexton, F. W.; Michie, R. M, Jr.; McElroy, F. F.; Thompson, V. L. (1982) A comparative evaluation of seven
2 automated ambient nonmethane organic compound analyzers. Research Triangle Park, NC: U.S.
3 Environmental Protection Agency, Environmental Monitoring Systems Laboratory; report no.
4 EPA-600/4-82-046. Available from: NTIS, Springfield, VA; PB82-230798.
5 Shafran, P. C.; Seaman, N. L.; Gayno, G. A. (2000) Evaluation of numerical predictions of boundary layer structure
6 during the Lake Michigan ozone study. J. Appl. Meteorol. 39: 412-420.
7 Shapiro, R. (1977) Genetic effects of bisulfite (sulfur dioxide). Mutat. Res. 39: 149-175.
8 Shapiro, M. A. (1980) Turbulent mixing within tropopause folds as a mechanism for the exchange of chemical
9 constituents between the stratosphere and troposphere. J. Atmos. Sci. 37: 994-1004.
10 Sharkey, T. D.; Singsaas, E. L.; Vanderveer, P. J.; Geron, C. D. (1996) Field measurements of isoprene emission
11 from trees in response to temperature and light. Tree Physiol. 16: 649-654.
12 Shepson, P. B.; Edney, E. O.; Corse, E. W. (1984) Ring fragmentation reactions on the photooxidations of toluene
13 and o-xylene. J. Phys. Chem. 88: 4122-4126.
14 Shepson, P. B.; Mackay, E.; Muthuramu, K. (1996) Henry's Law constants and removal processes for several
15 atmospheric 6-hydroxy alkyl nitrates. Environ. Sci. Technol. 30: 3618-3623.
16 Shelter, R. E.; Junkermann, W.; Swartz, W. H.; Frost, G. J.; Crawford, J. H.; Lefer, B. L.; Barrick, J. D.; Hall, S. R.;
17 Hofzumahaus, A.; Bais, A.; Calvert, J. G.; Cantrell, C. A.; Madronich, S.; Mtiller, M.; Kraus, A.; Monks,
18 P. S.; Edwards, G. D.; McKenzie, R.; Johnston, P.; Schmitt, R.; Griffioen, E.; Krol, M.; Kylling, A.;
19 Dickerson, R. R.; Lloyd, S. A.; Martin, T.; Gardiner, B.; Mayer, B.; Pfister, G.; Roeth, E. P.; Koepke, P.;
20 Ruggaber, A.; Schwander, H.; Van Weele, M. (2003) Photolysis frequency of NO2: Measurement and
21 modeling during the International Photolysis Frequency Measurement and Modeling intercomparison
22 (IPMMI). J. Geophys. Res. (Atmos.) 108(D16): 10.1029/2002JD002932.
23 Sillman, S. (1995) The use of NOy, H2O2 and HNO3 as indicators for O3-NOX-ROG sensitivity in urban locations.
24 J. Geophys. Res. 100: 14,175-14,188.
25 Sillman, S. (1999) The relation between ozone, NOX and hydrocarbons in urban and polluted rural environments.
26 Atmos. Environ. 33: 1821-1845.
27 Sillman, S. (2000) Ozone production efficiency and loss of NOX in power plant plumes: photochemical model and
28 interpretation of measurements in Tennessee. J. Geophys. Res. [Atmos.] 105: 9189-9202.
29 Sillman, S.; He, D.-Y. (2002) Some theoretical results concerning O3-NOX-VOC chemistry and NOX-VOC
30 indicators. J. Geophys. Res. (Atmos.) 107: 10.1029/2001JDOO1123.
31 Sillman, S.; Samson, P. J. (1995) Impact of temperature on oxidant photochemistry in urban, polluted rural and
32 remote environments. J. Geophys. Res. [Atmos.] 100: 11,497-11,508.
33 Sillman, S.; Logan, J. A.; Wofsy, S. C. (1990) The sensitivity of ozone to nitrogen oxides and hydrocarbons in
34 regional ozone episodes. J. Geophys. Res. [Atmos.] 95: 1837-1851.
35 Sillman, S.; Al-Wali, K. L; Marsik, F. J.; Nowacki, P.; Samson, P. J.; Rodgers, M. O.; Garland, L. J.; Martinez, J. E.;
36 Stoneking, C.; Imhoff, R.; Lee, J.-H.; Newman, L.; Weinstein-Lloyd, J.; Aneja, V. P. (1995) Photochemistry
37 of ozone formation in Atlanta, GA-models and measurements. Atmos. Environ. 29: 3055-3066.
38 Sillman, S.; He, D.; Cardelino, C.; Imhoff, R. E. (1997) The use of photochemical indicators to evaluate
39 ozone-NOx-hydrocarbon sensitivity: case studies from Atlanta, New York, and Los Angeles. J. Air Waste
40 Manage. Assoc. 47: 1030-1040.
41 Sillman, S.; He, D.; Pippin, M. R.; Daum, P. H.; Imre, D. G.; Kleinman, L. I.; Lee, J. H.; Weinstein-Lloyd, J. (1998)
42 Model correlations for ozone, reactive nitrogen, and peroxides for Nashville in comparison with
43 measurements: implications for O3-NOx-hydrocarbon chemistry. J. Geophys. Res. [Atmos.] 103:
44 22,629-22,644.
45 Sillman, S.; Vautard, R.; Menut, L.; Kley, D. (2003) O3-NOX-VOC sensitivity and NOX-VOC indicators in Paris:
46 results from models and atmospheric pollution over the Paris area (ESQUIF) measurements. J. Geophys. Res.
47 (Atmos.) 108: 10.1029/2002JD001561.
48 Singh, H. B.; Ludwig, F. L.; Johnson, W. B. (1977) Ozone in clean remote atmospheres: concentrations and
49 variabilities. New York, NY: Coordinating Research Council, Inc.; report no. CRC-APRAC CAPA-15-76.
50 Available from: NTIS, Springfield, VA; PB-272290.
51 Singh, H. B.; Viezee, W.; Johnson, W. B.; Ludwig, F. L. (1980) The impact of stratospheric ozone on tropospheric
52 air quality. J. Air Pollut. Control Assoc. 30: 1009-1017.
53 Singh, H. B.; Kanakidou, M.; Crutzen, P. J.; Jacob, D. J. (1995) High concentrations and photochemical fate of
54 oxygenated hydrocarbons in the global troposphere. Nature (London) 378: 50-54.
55
January 2005 AX2-184 DRAFT-DO NOT QUOTE OR CITE
-------
1 Singh, H. B.; Gregory, G. L.; Anderson, B.; Browell, E.; Sachse, G. W.; Davis, D. D.; Crawford, J.; Bradshaw, J. D.;
2 Talbot, R.; Blake, D. R.; Thornton, D.; Newell, R.; Merrill, J. (1996a) Low ozone in the marine boundary
3 layer of the tropical Pacific Ocean: photochemical loss, chlorine atoms, and entrainment. J. Geophys. Res.
4 [Atmos.] 101: 1907-1917.
5 Singh, H. B.; Herlth, D.; Kolyer, R.; Salas, L.; Bradshaw, J. D.; Sandholm, S. T.; Davis, D. D.; Crawford, J.; Kondo,
6 Y.; Koike, M; Talbot, R.; Gregory, G. L.; Sachse, G. W.; Browell, E.; Blake, D. R.; Rowland, F. S.; Newell,
7 R.; Merrill, J.; Heikes, B.; Liu, S. C.; Crutzen, P. J.; Kanakidou, M. (1996b) Reactive nitrogen and ozone over
8 the western Pacific: distribution, partitioning, and sources. J. Geophys. Res. 101: 1793-1808.
9 Singh, M. P.; McNider, R. T.; Meyers, R.; Shekhar, G. (1997) Nocturnal wind structure and dispersion of pollutants:
10 an analytical study. Atmos. Environ. 31: 105-115.
11 Singh, H. B.; Salas, L. J.; Chatfield, R. B.; Czech, E.; Fried, A.; Walega, J.; Evans, M. J.; Field, B. D.; Jacob, D. J.;
12 Blake, D.; Heikes, B.; Talbot, R.; Sachse, G.; Crawford, J. H.; Avery, M. A.; Sandholm, S.; Fuelberg, H.
13 (2004) Analysis of the atmospheric distribution, sources, and sinks of oxygenated volatile organic chemicals
14 based on measurements over the Pacific during TRACE-P. J. Geophys. Res. (Atmos.) 109(D15S07):
15 10.1029/2003 JD003883.
16 Sirju, A. P.; Shepson, P. B. (1995) Laboratory and field investigation of the DNPH cartridge technique for the
17 measurement of atmospheric carbonyl compounds. Environ. Sci. Technol. 29: 384-392.
18 Skiba, U.; Fowler, D.; Smith, K.A. (1997) Nitric oxide emissions from agricultural soils in temperate and tropical
19 climates: sources, controls and mitigation options. Nutrient Cycling Agroecosystems. 48 (1-2): 139-153.
20 Slemr, F.; Seiler, W. (1984) Field measurements of NO and NO2 emissions from fertilized and unfertilized soils.
21 J. Atmos. Chem. 2: 1-24.
22 Slemr, J.; Slemr, F.; Partridge, R.; D'Souza, H.; Schmidbauer, N. (2002) Accurate Measuements of Hydrocarbons in
23 the Atmosphere (AMOHA): three European intercomparisons. J. Geophys. Res. [Atmos.]
24 107(019): 10.1020/2001JD001357.
25 Smith, F.; Strong, R. (1977) Independent performance audits on rams station sensors - audit #4. interim report.
26 Research Triangle Park, NC: Research Triangle Institute; EPA contract no. 68-02-2407.
27 Smith, A. M.; Rigler, E.; Kwok, E. S. C.; Atkinson, R. (1996) Kinetics and products of the gas-phase reactions of
28 6-methyl-5-hepten-2-one and /rans-cinnamaldehyde with OH and NO3 radicals and O3 at 29672K. Environ.
29 Sci. Technol. 30: 1781-1785.
30 Smith, D. F.; Mclver, C. D.; Kleindienst, T. E. (1998) Primary product distribution from the reaction of hydroxyl
31 radicals with toluene at ppb NOX mixing ratios. J. Atmos. Chem. 30: 209-228.
32 Smith, D. F.; Kleindienst, T. E.; Mclver, C. D. (1999) Primary product distributions from the reaction of OH with
33 m-,p-xylene, 1,2,4- and 1,3,5-trimethylbenzene. J. Atmos. Chem. 34: 339-364.
34 Spicer, C. W.; Chapman, E. G.; Finlayson-Pitts, B. J.; Plastridge, R. A.; Hubbe, J. M.; Fast, J. D.; Berkowitz, C. M.
3 5 (1998) Unexpectedly high concentrations of molecular chlorine in coastal air (letter). Nature (London)
36 394: 353-356.
37 Spicer, C. W.; Billick, I. H.; Yanagisawa, Y. (2001) Nitrous acid interference with passive NO2 measurement
38 methods and the impact on indoor NO2 data. Indoor Air 11: 156-161.
39 Spivakovsky, C. M.; Logan, J. A.; Montzka, S. A.; Balkanski, Y. J.; Foreman-Fowler, M.; Jones, D. B. A.;
40 Horowitz, L. W.; Fusco, A. C.; Brenninkmeijer, C. A. M.; Prather, M. J.; Wofsy, S. C.; McElroy, M. B.
41 (2000) Three-dimensional climatological distribution of tropospheric OH: update and evaluation. J. Geophys.
42 Res. (Atmos.) 105:8931-8980.
43 Spokes, L. J.; Yeatman, S. G.; Cornell, S. E.; Jickells, T. D. (2000) Nitrogen deposition to the eastern Atlantic
44 Ocean, the importance of south-easterly flow. Tellus Ser. B 52B: 37-49.
45 Sprenger, M.; Wernli, H. (2003) A northern hemispheric climatology of cross-tropopause exchange for the ERA
46 15 time period (1979-1993). J. Geophys. Res. 108(D12): 10.1029/2002JD002636.
47 Sprenger, M.; Maspoli, M. C.; Wernli, H. (2003) Tropopause folds and cross-tropopause exchange: a global
48 investigation based upon ECMWF analyses for the time period March 2000 to February 2001. J. Geophys.
49 Res. (Atmos.) 108(D12): 10.1029/2002JD002587.
50 Sprengnether, M.; Demerjian, K. L.; Donahue, N. M.; Anderson, J. G. (2002) Product analysis of the OH oxidation
51 of isoprene and 1,3-butadiene in the presence of NO J. Geophys. Res. (Atmos.) 107(D15):
52 10.1029/2001JD000716.
53 Sprung, D.; Jost, C.; Reiner, T.; Hansel, A.; Wisthaler, A. (2001) Acetone and acetonitrile in the tropical Indian
54 ocean boundary layer and free troposphere: aircraft-based intercomparison of AP-CIMS and PTR-MS
55 measurements. J. Geophys. Res. 106: 28,511-28,528.
January 2005 AX2-185 DRAFT-DO NOT QUOTE OR CITE
-------
1 Staffelbach, T.; Neftel, A.; Blatter, A.; Gut, A.; Fahrni, M; Stahelin, J.; Prevot, A.; Hering, A.; Lehning, M.;
2 Neininger, B.; Baumle, M.; Kok, G. L.; Dommen, J.; Hutterli, M.; Anklin, M. (1997) Photochemical oxidant
3 formation over southern Switzerland 1. results from summer 1994. J. Geophys. Res. (Atmos.) 102:
4 23,345-23,362.
5 Staudt, A. C.; Jacob, D. J.; Ravetta, F.; Logan, J. A.; Bachiochi, D.; Krishnamurti, T. N.; Sandholm, S. T>; Ridley,
6 B. A.; Singh, H. B.; Talbot, R. W. (2003) Sources and chemistry of nitrogen oxides over the tropical Pacific.
7 J. Geophys. Res. (Atmos.) 108(8239): 10.1029/2002JD002139.
8 Stedman, D. H.; Bishop, G.; Peterson, J. E.; Guenther, P. L. (1991) On-road CO remote sensing in the Los Angeles
9 Basin: final report. Sacramento, CA: California Air Resources Board, ARE Contract No. A932-189.
10 Stehr, J. W.; Dickerson, R. R.; Hallock-Waters, K. A.; Doddridge, B. G.; Kirk, D. (2000) Observations of NOy, CO,
11 and SO2 and the origin of reactive nitrogen J. Geophys. Res. 105(D3): 3553-3563.
12 Stein, A. F.; Lamb, D. (2002) Chemical indicators of sulfate sensitivity to nitrogen oxides an volatile organic
13 compounds. J. Geophys. Res. (Atmos.) 107: 10.1029/2001JDOO1088.
14 Stein, A. F.; Lamb, D. (2003) Empirical evidence for the low- and high-NOx photochemical regimes of sulfate and
15 nitrate formation. Atmos. Environ. 37: 3615-3625.
16 Stein, A. F.; Lamb, D.; Draxler, R. R. (2000) Incorporation of detailed chemistry into a three-dimensional
17 Lagrangian-Eulerian hybrid model: application to regional tropospheric ozone. Atmos. Environ. 34:
18 4361-4372.
19 Stenchikov, G.; Dickerson, R.; Pickering, K.; Ellis, W. Jr.; Doddridge, B.; Kondragunta, S.; Poulida, O.; Scala, J.;
20 Tao, W.-K. (1996) Stratosphere-troposphere exchange in a midlatitude mesoscale convective complex. 2.
21 Numerical simulation. J. Geophys. Res. [Atmos.] 101: 6837-6851.
22 Stevens, R. K.; Drago, R. J.; Mamane, Y. (1993) A long path differential optical absorption spectrometer and
23 EPA-approved fixed-point methods intercomparison. Atmos. Environ. Part B 27: 231-236.
24 Steyn, D. J.; Bottenheim, J. W.; Thomson, R. B. (1997) Overview of tropospheric ozone in the Lower Fraser Valley,
25 and the Pacific '93 field study. Atmos. Environ. 31: 2025-2036.
26 Stith, J.; Dye, J.; Ridley, B.; Laroche, P.; Defer, E.; Baumann, K.; Htibler, G.; Zerr, R.; Venticinque, M. (1999) NO
27 signatures from lightning flashes. J. Geophys. Res. [Atmos.] 101: 6837-6851.
28 Stockwell, W. R.; Middleton, P.; Chang, J. S.; Tang, X. (1990) The second generation Regional Acid Deposition
29 Model chemical mechanism for regional air quality modeling. J. Geophys. Res. [Atmos.] 95: 16,343-16,367.
30 Stockwell, W. R.; Kirchner, F.; Kuhn, M.; Seefeld, S. (1997) A new mechanism for regional atmospheric chemistry
31 modeling. J. Geophys. Res. 102(D22): 25847-25879.
32 Stohl, A. (2001) A 1-year Lagrangian "climatology" of airstreams in the Northern Hemisphere troposphere and
33 lowermost stratosphere. J. Geophys. Res. (Atmos.) 106: 7263-7279.
34 Stohl, A.; Trickl, T. (1999) A textbook example of long-range transport: simultaneous observation of ozone maxima
35 of stratospheric and North American origin in the free troposphere of Europe. J. Geophys. Res.
36 104(D23): 30,445-30,462.
37 Stohl, A.; Spichtinger-Rakowsky, N.; Bonasoni, P.; Feldmann, H.; Memmesheimer, M.; Scheel, H. E.; Trickly, T.;
38 Htibener, S.; Ringer, W.; Mandl, M. (2000) The influence of stratospheric intrusions on alpine ozone
39 concentrations. Atmos. Environ. 34: 1323-1354.
40 Stohl, A.; Trainer, M.; Ryerson, T. B.; Holloway, J. S.; Parrish, D. D. (2002) Export of NOy from the North
41 American boundary layer during 1996 and 1997 North Atlantic Regional Experiments. J. Geophys. Res.
42 107(D11): 10.1029/2001JD000519.
43 Stohl, A.; Bonasoni, P.; Cristofanelli, P.; Collins, W.; Feichter, J.; Frank, A.; Forster, C.; Gerasopoulos, E.;
44 Gaggeler, H.; James, P.; Kentarchos, T.; Kromp-Kolb, H.; Kriiger, B.; Land, C.; Meloen, J.; Papayannis, A.;
45 Priller, A.; Seibert, P.; Sprenger, M.; Roelofs, G. J.; Scheel, H. E.; Schnabel, C.; Siegmund, P.; Tobler, L.;
46 Trickl, T.; Wernli, H.; Wirth, V.; Zanis, P.; Zerefos, C. (2003) Stratosphere-troposphere exchange: a review
47 and what we have learned from STACCATO. J. Geophys. Res. 108(D12): 10.1029/2002JD002490.
48 Stroud, C.; Madronich, S.; Atlas, E.; Ridley, B.; Flocke, F.; Weinheimer, A.; Talbot, B.; Fried, A.; Wert, B.;
49 Shelter, R.; Lefer, B.; Coffey, M.; Heikes, B.; Blake, D. (2003) Photochemistry in the arctic free troposphere:
50 NOX budget and the role of odd nitrogen reservoir recycling. Atmos. Environ. 37: 3351-3364.
51 Stull, R. B. (1985) A fair-weather cumulus cloud classification scheme for mixed layer studies. J. Clim. Appl.
52 Meteorol. 24: 49-56.
53 Stull, R. B. (1988) An introduction to boundary layer meteorology. Dordrecht, The Netherlands: Kluwer Academic
54 Publishers, pp. 9-16; 499-506. (Atmospheric Sciences Library, v. 13).
55 Stull, R.B. (1999) An introduction to boundary layer meteorology. Dordrecht, The Netherlands: Kluwer Academic
56 Publishers; pp. 9-15, 500-505.
January 2005 AX2-186 DRAFT-DO NOT QUOTE OR CITE
-------
1 Stall, R. B. (2000) Meteorology for scientists and engineers: a technical companion book to C. Donald Ahrens'
2 Meteorology Today. 2nd ed. New York, NY: Brooks Cole.
3 Starges, W. T.; Barrie, L. A. (1988) Chlorine Bromine and Iodine in Arctic aerosols. Atmos. Environ.
4 22: 1179-1194.
5 Statz, J.; Platt, U. (1996) Numerical analysis and estimation of the statistical error of differential optical absorption
6 spectroscopy measurements with least-squares methods. Appl. Opt. 35: 6041-6053.
7 Statz, J.; Hebestreit, K.; Alicke, B.; Platt, U. (1999) Chemistry of halogen ozides in the troposphere: comparison of
8 model calculations with recent field data. J. Atmos. Chem. 34: 65-85.
9 Statz, J.; Ackermann, R.; Fast, J. D.; Barrie, L. (2002) Atmospheric reactive chlorine and bromine at the Great Salt
10 Lake, Utah. Geophys. Res. Lett. 29: 10.1029/2002GL014812.
11 Suhre, K.; Gammas, J.-P.; Nedelec, P.; Rosset, R.; Marenco, A.; Smit, H. G. J. (1997) Ozone-rich transients in the
12 upper equatorial Atlantic troposphere (letter). Nature 388: 661-663.
13 Sundet, J. K. (1997) Model studies with a 3-D global CTM using ECMWF data [thesis]. Oslo, Norway: University
14 of Oslo, Department of Geophysics.
15 Svensson, R.; Ljungstroem, E.; Lindqvist, O. (1987) Kinetics of the reaction between nitrogen dioxide and water
16 vapour. Atmos. Environ. 21: 1529-1539.
17 Talbot, R. W.; Vijgen, A. S.; Harris, R. C. (1990) Measuring tropospheric HNO3: problems and prospects for nylon
18 filter and mist chamber. J. Geophys. Res. [Atmos.] 95: 7553-7542.
19 Tan, D.; Faloona, I.; Simpas, J. B.; Brune, W.; Shepson, P. B.; Couch, T. L.; Sumner, A. L.; Carroll, M. A.;
20 Thornberry, T.; Apel, E.; Riemer, D.; Stockwell, W. (2001) HOX budgets in a deciduous forest: results from
21 the PROPHET summer 1998 campaign. J. Geophys. Res. (Atmos.) 106: 24,407-24,427.
22 Tanaka, P. L.; Oldfield, S.; Neece, J. D.; Mullins, C. B.; Allen, D. T. (2000) Anthropogenic sources of chlorine and
23 ozone formation in urban atmospheres. Environ. Sci. Technol. 34: 4470-4473.
24 Tanaka, P. L.; Riemer, D. D.; Chang, S.; Yarwood, G.; McDonald-Buller, E. C.; Apel, E. C.; Orlando, J. J.; Silva,
25 P. J.; Jimenez, J. L.; Canagaratna, M. R.; Neece, J. D.; Mullins, C. B.; Allen, D. T. (2003a) Direct evidence
26 for chlorine-enhanced urban ozone formation in Houston, Texas. Atmos. Environ. 37: 1393-1400.
27 Tanaka, P. L.; McDonald-Buller, E. C.; Chang, S.; Yarwood, G.; Kimura, Y.; Neece, J. D.; Mullins, C. B.; Allen,
28 D. T. (2003b) Development of a chlorine mechanism for use in the Carbon Bond IV chemistry model.
29 J. Geophys. Res. [Atmos.] 108(D4): art. no. 4145.
30 Tanner, R. L.; Kelly, T. J.; Dezaro, D. A.; Forrest, J. (1989) A comparison of filter, denuder, and real-time
31 chemiluminescence techniques for nitric acid determination in ambient air. Atmos. Environ. 23: 2213-2222.
32 Tao, W.-K.; Simpson, J. (1993) The Goddard Cumulus Ensemble Model. Part I: model description. Terr. Atmos.
3 3 Oceanic Sci. (TAO) 4:35-71.
34 Taubman, B. F.; Marufu, L. T.; Piety, C. A.; Doddridge, E.G.; Stehr, J. W.; Dickerson, R. R. (2004) Airborne
3 5 characterization of the chemical, optical, and meteorological properties, and origins of a combined
36 ozone-haze episode over the eastern United States. J. Atmos. Sci. 61: 1781-1793.
37 Tesche, T. W.; Roth, P. M.; Reynolds, S. D.; Lurmann, F. W. (1993) Scientific assessment of the urban airshed
3 8 model (UAM-IV). Washington, DC: American Petroleum Institute; API publication no. 4556.
39 Thielman, A.; Prevot, A. S. H.; Griiebler, F. C.; Staelhelin, J. (2001) Empirical ozone isopleths as a tool to identify
40 ozone production regimes. Geophys. Res. Lett. 28: 2369-2372.
41 Thielman, A.; Prevot, A. S. H.; Staelhelin, J. (2002) Sensitivity of ozone production derived from field
42 measurements in the Italian Po basin. J. Geophys. Res. (Atmos.) 107(D22): 10.1029/2000JD000119.
43 Thompson, A. M. (1992) The oxidizing capacity of the Earth's atmosphere: probable past and future changes.
44 Science (Washington, DC) 256: 1157-1165.
45 Thompson, A. M.; Stewart, R. W. (1991) Effect of chemical kinetics uncertainties on calculated constituents in a
46 tropospheric photochemical model. J. Geophys. Res. 96(D7): 13,089-13,108.
47 Thompson, A. M.; Pickering, K. E.; Dickerson, R. R.; Ellis, W. G., Jr.; Jacob, D. J.; Scala, J. R.; Tao, W.-K.;
48 McNamara, D. P.; Simpson, J. (1994) Convective-transport over the central United States and its role in
49 regional CO and O3 budgets. J. Geophys. Res. [Atmos.] 99: 18,703-18,711.
50 Thompson, A. M.; Singh, H. B.; Schlager, H. (2000) Subsonic assessment ozone and nitrogen oxide experiment
51 (SONEX) and pollution from aircraft emissions in the North Atlantic Flight Corridor (POLINAT 2).
52 J. Geophys. Res. 105: 3595-3603.
53 Thornton, J. A.; Wooldridge, P. J.; Cohen, R. C. (2000) Atmospheric NO2: in situ laser-induced fluorescence
54 detection at parts per trillion mixing ratios. Anal. Chem. 72: 528-539.
55 Thornton, J. A.; Braban, C. F.; Abbatt, J. P. D. (2003) N2O5 hydrolysis on sub-micron organic aerosol: the effect of
56 relative humidity, particle phase and particle size. Phys. Chem. Chem. Phys. 5: 4593-4603.
January 2005 AX2-187 DRAFT-DO NOT QUOTE OR CITE
-------
1 Tie, X. X.; Emmons, L.; Horowitz, L.; Brasseur, G.; Ridley, ,B.; Atlas, E.; Stround, C.; Hess, P.; Klonecki, A.;
2 Madronich, S.; Talbot, R.; Dibb, J. (2003) Effect of sulfate aerosol on tropospheric NOX and ozone budgets:
3 model simulations and TOPSE evidence. J. Geophys. Res. (Atmos.) 108(D4): 10.1029/2001JD001508.
4 Tonnesen, G. S.; Dennis, R. L. (2000a) Analysis of radical propagation efficiency to assess ozone sensitivity to
5 hydrocarbons and NOX 1. Local indicators of odd oxygen production sensitivity. J. Geophys. Res. (Atmos.)
6 105: 9213-9225.
7 Tonnesen, G. S.; Dennis, R. L. (2000b) Analysis of radical propagation efficiency to assess ozone sensitivity to
8 hydrocarbons and NOX 2. Long-lived species as indicators of ozone concentration sensitivity. J. Geophys.
9 Res. 105: 9227-9241.
10 Trainer, M; Buhr, M. P.; Curran, C. M; Fehsenfeld, F. C.; Hsie, E. Y.; Liu, S. C.; Norton, R. B.; Parrish, D. D.;
11 Williams, E. J.; Gandrud, B. W.; Ridley, B. A.; Shelter, J. D.; Allwine, E. J.; Westberg, H. H. (1991)
12 Observations and modeling of the reactive nitrogen photochemistry at a rural site. J. Geophys. Res. [Atmos.]
13 96: 3045-3063.
14 Trainer, M.; Parrish, D. D.; Buhr, M. P.; Norton, R. B.; Fehsenfeld, F. C.; Anlauf, K. G.; Bottenheim, J. W.; Tang,
15 Y. Z.; Wiebe, H. A.; Roberts, J. M.; Tanner, R. L.; Newman, L.; Bowersox, V. C.; Meagher, J. F.; Olszyna,
16 K. J.; Rodgers, M. O.; Wang, T.; Berresheim, H.; Demerjian, K. L.; Roychowdhury, U. K. (1993) Correlation
17 of ozone with NOy in photochemically aged air. J. Geophys. Res. [Atmos.] 98: 2917-2925.
18 Trainer, M.; Ridley, B. A.; Buhr, M. P.; Kok, G.; Walega, J.; Htibler, G.; Parrish, D. D.; Fehsenfeld, F. C. (1995)
19 Regional ozone and urban plumes in the southeastern United States: Birmingham, a case study. J. Geophys.
20 Res. [Atmos.] 100: 18,823-18,834.
21 Trainer, M.; Parrish, D. D.; Golday, P. D.; Roberts, J.; Fehsenfeld, F. C. (2000) Review of observation-based
22 analysis of the regional factors influencing ozone concentrations. Atmos. Environ. 34: 2045-2061.
23 Traub, M.; Fischer, H.; De Reus, M.; Kormann, R.; Heland, J.; Ziereis, H.; Schlager, H.; Holzinger, R.; Williams, J.;
24 Warneke, C.; De Gouw, J.; Lelieveld, J. (2003) Chemical characteristics assigned to trajectory clusters during
25 the MINOS campaign. Atmos. Chem. Phys. 3: 459-468.
26 Tseng, R.-S.; Viechnicki, J. T.; Skop, R. A.; Brown, J. W. (1992) Sea-to-air transfer of surface-active organic
27 compounds by bursting bubbles. J. Geophys. Res. (Atmos.) 97: 5201-5206.
28 Tuazon, E. C.; MacLeod, H.; Atkinson, R.; Carter, W. P. L. (1986) a-Dicarbonyl yields from the NOx-air
29 photooxidations of a series of aromatic hydrocarbons in air. Environ. Sci. Technol. 20: 383-387.
30 Tully, F. P.; Ravishankara, A. R.; Thompson, R. L.; Nicovich, J. M.; Shah, R. C.; Kreutter, N. M.; Wine, P. H.
31 (1981) Kinetics of the reactions of hydroxyl radical with benzene and toluene. J. Phys. Chem. 85: 2262-2269.
32 Turekian, V. C.; Macko, S. A.; Keene, W. C. (2003) Concentrations, isotopic compositions, and sources of
33 size-resolved, paniculate organic carbon and oxalate in near-surface marine air at Bermuda during spring.
34 J. Geophys. Res. (Atmos.) 108(D5): 10.1029/2002JD002053.
35 U.S. Congress. (1990) Clean Air Act amendments of 1990, PL 101-549, November 15. Washington, DC: U.S.
3 6 Government Printing Office.
37 U.S. Environmental Protection Agency. (1991) Guideline for regulatory application of the urban airshed model.
38 Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA-450/4-91-013.
39 Available from: NTIS, Springfield, VA; PB92-108760/HSU.
40 U.S. Environmental Protection Agency. (1993a) Air quality criteria for oxides of nitrogen. Research Triangle Park,
41 NC: Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office; report
42 nos. EPA/600/8-9!/049aF-cF. 3v. Available from: NTIS, Springfield, VA; PB95-124533, PB95-124525, and
43 PB95-124517.
44 U.S. Environmental Protection Agency. (1993) Regional interim emission inventories (1987-1991), v. II: emission
45 summaries. Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no.
46 EPA-454/R-93-021b. Available from: NTIS, Springfield, VA; PB93-236115.
47 U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants, v.
48 I-III. Research Triangle Park, NC: Office of Research and Development; report no. EPA/600/P-93/004aF-cF.
49 Available from: NTIS, Springfield, VA; PB96-185582. PB96-185590, and PB96-185608. Available:
50 www.epa.gov/ncea/ozone.htm [25 February 2002].
51 U.S. Environmental Protection Agency. (1997) National air pollutant emission trends 1990-1996. Research Triangle
52 Park, NC: Office of Air Quality Planning and Standards; report no. EPA/454/R-97/011. Available from:
53 NTIS, Springfield, VA.
54 U.S. Environmental Protection Agency (1999).
55
January 2005 AX2-188 DRAFT-DO NOT QUOTE OR CITE
-------
1 U.S. Environmental Protection Agency. (2000) Air quality criteria for carbon monoxide. Research Triangle Park,
2 NC: National Center for Environmental Assessment; report no. EPA/600/P-99/00 IF. Available:
3 http://www.epa.gov/ncea/pdfs/coaqcd.pdf (7 May 2003).
4 U.S. Environmental Protection Agency. (2001) National air quality and emissions trends report, 1999. Research
5 Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA 454/R-01-004. Available:
6 http://www.epa.gov/oar/aqtrnd99/ [7 May 2003].
7 U.S. Environmental Protection Agency. (2003) Air quality criteria for ozone and related photochemical oxidants.
8 Research Triangle Park, NC: Office of Research and Development: manuscript in preparation.
9 Underwood, G. M.; Song, C. H.; Phadnis, M.; Carmichael, G. R.; Grassian, V. H. (2001) Heterogeneous reactions of
10 NO2 and HNO3 on oxides and mineral dust: a combined laboratory and modeling study. J. Geophys. Res.
11 [Atmos.] 106: 18,055-18,066.
12 Usher, C. R.; Michel, A. E.; Stec, D.; Grassian, V. H. (2003) Laboratory studies of ozone uptake on processed
13 mineral dust. Atmos. Environ. 37: 5337-5347.
14 Valente, R. J.; Imhoff, R. E.; Tanner, R. L.; Meagher, J. F.; Damn, P. H.; Hardesty, R. M.; Banta, R. M.; Alvarez,
15 R. J.; McNider, R. T.; Gillani, N. V. (1998) Ozone production during an urban air stagnation episode over
16 Nashville, Tennessee. J. Geophys. Res. [Atmos.] 103: 22,555-22,568.
17 Van Aardenne, J. A.; Dentener, F. J.; Olivier, J. G. J.; Goldewijk, C. G. M. K.; Lelieveld, J. (2001) A 1 ° x 1°
18 resolution data set of historical anthropogenic trace gas emissions for the period 1980-1990. Glob.
19 Biogeochem. Cycles 15: 909-928.
20 Vandaele, A. C.; Tsouli, A.; Carleer, M.; Colin, R. (2002) UV Fourier transform measurements of tropospheric O3,
21 NO2, SO2, benzene, and toluene. Environ. Pollut. 116: 193-201.
22 Vaughan, W. M.; Chan, M.; Cantrell, B.; Pooler, F. (1982) A study of persistent elevated pollution episodes in the
23 northeastern United States. Bull. Am. Meteorol. Soc. 63: 258-266.
24 Vautard, R.; Martin, D.; Beekman, M.; Drobinski, P.; Friedrich, R.; Jaubertie, A.; Kley, D.; Lattuati, M.; Moral, P.;
25 Neininger, B.; Theloke, J. (2002) Paris emission inventory diagnostics from the ESQUIF airborne
26 measurements and a chemistry transport model. J. Geophys. Res. (Atmos.) 108(D17):
27 10.1029/2002JD002797.
28 Viezee, W.; Johnson, W. B.; Singh, H. B. (1979) Airborne measurements of stratospheric ozone intrusions into the
29 troposphere over the United States [final report]. Menlo Park, CA: SRI International; SRI project 6690.
30 Viezee, W.; Singh, H. B. (1982) Contribution of stratospheric ozone to ground-level ozone concentrations - a
31 scientific review of existing evidence. Research Triangle Park, NC: U.S. Environmental Protection Agency,
32 Environmental Science Research Laboratory; grant CR-809330010.
33 Vogt, R.; Crutzen, P. J.; Sander, R. (1996) A mechanism for halogen release from sea-salt aerosol in the remote
34 marine boundary layer. Nature 383: 327-330.
35 Vogt, R.; Sander, R.; Von Glasow, R.; Crutzen, P. J. (1999) Iodine chemistry and its role in halogen activation and
36 ozone loss in the marine boundary layer: a model study. J. Atmos. Chem. 32: 375-395.
37 Volz-Thomas, A.; Geiss, H.; Hofzumahaus, A.; Becker, K.-H. (2003) Introduction to special section: photochemistry
3 8 experiment in BERLIOZ. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/JD002029.
39 Von Glasow, R.; Sander, R.; Bott, A.; Crutzen, P. J. (2002a) Modeling halogen chemistry in the marine boundary
40 layer. 1. Cloud-free MBL. J. Geophys. Res. 107(D17): 10.1029/2001JD000942.
41 Von Glasow, Sander, R.; Bott, A.; Crutzen, P. J. (2002b) Modeling halogen chemistry in the marine boundary layer.
42 2. Interactions with sulfur and cloud-covered MBL. J. Geophys. Res. 107(D17): 10.1029/2001JD000943.
43 Vuilleumier, L.; Bamer, J. T.; Harley, R. A.; Brown, N. J. (2001) Evaluation of nitrogen dioxide photolysis rates in
44 an urban area using data from the 1997 Southern California Ozone Study. Atmos. Environ. 35: 6525-6537.
45 Vukovich, F. M. (1995) Regional-scale boundary layer ozone variations in the eastern United States and their
46 association with meteorological variations. Atmos. Environ. 29: 2259-2273.
47 Vukovich, F. M. (1997) Time scales of surface ozone variations in the regional, non-urban environment. Atmos.
48 Environ. 31:1513-1530.
49 Vukovich, R. M.; Ching, J. K. S. (1990) A semi-impirical approach to estimate vertical transport by nonprecipitating
50 convective clouds on a regional scale. Atmos. Environ. 24A: 2153-2168.
51 Vukovich, F. M.; Bach, W. D., Jr.; Crissman, B. W.; King, W. J. (1977) On the relationship between high ozone in
52 the rural surface layer and high pressure systems. Atmos. Environ. 11: 967-983.
53 Wagner, T.; Platt, U. (1998) Satellite mapping of enhanced BrO concentrations in the troposphere. Nature (London)
54 395:486-490.
55 Wagner, V.; Jenkin, M. E.; Saunders, S. M.; Stanton, K.; Wirtz, M.; Pilling, J. (2003) Modelling of the
56 photooxidation of toluene: conceptual ideas for validating detailed mechanisms. Atmos. Chem. Phys.
57 3:89-106.
January 2005 AX2-189 DRAFT-DO NOT QUOTE OR CITE
-------
1 Wahner, A.; Mentel, T. F.; Sohn, M.; Stier, J. (1998) Heterogeneous reaction of N2O5 on sodium nitrate aerosol.
2 J. Geophys. Res. (Atmos.) 103: 31,103-31,112.
3 Wainman, T.; Zhang, J.; Weschler, C. J.; Lioy, P. J. (2000) Ozone and limonene in indoor air: a source of submicron
4 particle exposure. Environ. Health Perspect. 108: 1139-1145.
5 Walcek, C. J.; Taylor, G. R. (1986) A theoretical method for computing vertical distributions of acidity and sulfate
6 production within cumulus clouds. J. Atmos. Sci. 43: 339-355.
7 Walcek, C. J.; Stockwell, W. R.; Chang, J. S. (1990) Theoretical estimates of the dynamic, radiative and chemical
8 effects of clouds on tropospheric trace gases. Atmos. Res. 25: 53-69.
9 Walega, J. G.; Stedman, D. H.; Shelter, R. E.; Mackay, G. I.; Iguchi, T.; Schiff, H. I. (1984) Comparison of a
10 chemiluminescent and a tunable diode laser absorption technique for the measurement of nitrogen oxide,
11 nitrogen dioxide, and nitric acid. Environ. Sci. Technol. 18: 823-826.
12 Wang, Y.; Tao, W.-K.; Pickering, K. E.; Thompson, A. M.; Kain, J. S.; Adler, R. F.; Simpson, J.; Keehn, P. R.; Lai,
13 G. S. (1996) Mesoscale model simulations of TRACE A and preliminary regional experiment for storm-scale
14 operational and research meteorology convective systems and associated tracer transport. J. Geophys. Res.
15 [Atmos.] 101:24,013-24,027.
16 Wang, Y.; DeSilva, A. W.; Goldenbaum, G. C.; Dickerson, R. R. (1998) Nitric oxide production by simulated
17 lightning: dependence on current, energy, and pressure. J. Geophys. Res. [Atmos.] 103: 19,149-19,159.
18 Wang, Y.; Logan, J. A.; Jacob, D. J. (1998) Global simulation of tropospheric O3-NOx-hydrocarbon chemistry, 2.
19 Model evaluation and global ozone budget. J. Geophys. Res. (Atmos.) 103: 10,727-10,755.
20 Wang, L. H.; Milford, J. B.; Carter, W. P. L. (2000a) Reactivity estimates for aromatic compounds. Part I:
21 uncertainty in chamber-derived parameters. Atmos. Environ. 34: 4337-4348.
22 Wang, L. H.; Milford, J. B.; Carter, W. P. L. (2000b) Reactivity estimates for aromatic compounds. Part 2.
23 uncertainty in incremental reactivities. Atmos. Environ. 4349-4360.
24 Warneke, C.; Van Der Veen, C.; Luxembourg, S.; De Gouw, J. A.; Kok, A. (2001) Measurements of benzene and
25 toluene in ambient air using proton-transfer-reaction mass spectrometry: calibration, humidity dependence,
26 and field intercomparison. Int. J. Mass Spectrom. 207: 167-182.
27 Watson, J. G.; Chow, J. C.; Fujita, E. M. (2001) Review of volatile organic compound source apportionment by
28 chemical mass balance. Atmos. Environ. 35: 1567-1584.
29 Wayne, R. P. (2000) Chemistry of Atmospheres: an introduction to the chemistry of the atmospheres of Earth, the
30 planets, and their satellites. 3rd ed. New York, NY: Oxford University Press, Inc.
31 Wayne, R. P.; Barnes, I.; Biggs, P.; Burrows, J. P.; Canosa-Mas, C. E.; Hjorth, J.; Le Bras, G.; Moortgat, G. K.;
32 Perner, D.; Poulet, G.; Restelli, G.; Sidebottom, H. (1991) The nitrate radical: physics, chemistry, and the
33 atmosphere. Atmos. Environ. Part A 25: 1-203.
34 Wingenter, O. W.; Kubo, M. K.; Blake, N. J.; Smith, T. W., Jr.; Blake, D. R.; Rowland, F. S. (1996) Hydrocarbon
3 5 and halocarbon measurements as photochemical and dynamical indicators of atmospheric hydroxyl, atomic
36 chlorine, and vertical mixing obtained during Langrangian flights. J. Geophys. Res. (Atmos.)
37 101(D1): 4331-4340.
38 Wingenter, O. W.; Blake, D. R.; Blake, N. J.; Sive, B. C.; Rowland, F. S.; Atlas, E.; Flocke, F. (1999) Tropospheric
3 9 hydroxyl and atomic chlorine concentrations, and mixing timescales determined from hydrocarbon and
40 halocarbon measurements made over the Southern Ocean. J. Geophys. Res. (Atmos.) 104(D17):
41 21819-21828.
42 Wennberg, P. O.; Cohen, R. C.; Stimpfle, R. M.; Koplow, J. P.; Anderson, J. G.; Salawitch, R. J.; Fahey, D. W.;
43 Woodbridge, E. L.; Keim, E. R.; Geo, R. S.; Webster, C. R.; May, R. D.; Toohey, D. W.; Avallone, L. M.;
44 Proffitt, M. H.; Loewenstein, M.; Podolske, J. R.; Chan, K. R.; Wofsy, S. C. (1994) Removal of stratospheric
45 O3 by radicals - in situ measurements of OH, HO2, NO, NO2, CLO, and BRO. Science 266: 398-404.
46 Wennberg, P. O.; Hanisco, T. F.; Jaegle, L.; Jacob, D. J.; Hintsa, E. J.; Lanzendorf, E. J.; Anderson, J. G.; Gao,
47 R.-S.; Keim, E. R.; Donnelly, S. G.; Del Negro, L. A.; Fahey, D. W.; McKeen, S. A.; Salawitch, R. J.;
48 Webster, C. R.; May, R. D.; Herman, R. L.; Proffitt, M. H.; Margitan, J. J.; Atlas, E. L.; Schauffler, S. M.;
49 Flocke, F.; McElroy, C. T.; Bui, T. P. (1998) Hydrogen radicals, nitrogen radicals, and the production of O3 in
50 the upper troposphere. Science (Washington, DC) 279: 49-53.
51 Wernli, H.; Bourqui, M. (2002) A Lagrangian "1-year climatology" of (deep) cross-tropopause exchange in the
52 extratropical Northern Hemisphere. J. Geophys. Res. (Atmos.) 107(D1-D2): 10.1029/2001JD000812.
53 Wesely, M. L.; Hicks, B. B. (2000) A review of the current status of knowledge on dry deposition. Atmos. Environ.
54 34:2261-2282.
55 Westberg, H.; Zimmerman, P. (1993) Analytical methods used to identify nonmethane organic compounds in
56 ambient atmospheres. In: Newman, L., ed. Measurement challenges in atmospheric chemistry. Washington,
57 DC: American Chemical Society, pp. 275-291.
January 2005 AX2-190 DRAFT-DO NOT QUOTE OR CITE
-------
1 Whitby, K. T. (1978) The physical characteristics of sulfur aerosols. Atmos. Environ. 12: 135-159.
2 Wild, O.; Prather, M. J. (2000) Excitation of the primary tropospheric chemical mode in a global CTM. J. Geophys.
3 Res. (Atmos.) 105: 24,647-24,660.
4 Williams, E. J.; Guenther, A.; Fehsenfeld, F. C. (1992) An inventory of nitric oxide emissions from soils in the
5 United States. J. Geophys. Res. [Atmos.] 97: 7511-7519.
6 Williams, J.; Roberts, J. M.; Fehsenfeld, F. C.; Bertman, S. B.; Buhr, M. P.; Goldan, P. D.; Hubler, G.; Kuster, W.
7 C.; Ryerson, T. B.; Trainer, M.; Young, V. (1997) Regional ozone frombiogenic hydrocarbons deduced from
8 airborne measurements of PAN, PPN, and MPAN. Geophys. Res. Lett. 24: 1099-1102.
9 Williams, E. J.; Baumann, K.; Roberts, J. M.; Bertman, S. B.; Norton, R. B.; Fehsenfeld, F. C.; Springston, S. R.;
10 Nunnermacker, L. J.; Newman, L.; Olszyna, K.; Meagher, J.; Hartsell, B.; Edgerton, E.; Pearson, J. R.;
11 Rodgers, M. O. (1998) Intercomparison of ground-based NOy measurement techniques. J. Geophys. Res.
12 103:22,261-22,280.
13 Wimmers, A. J. (2003) (dissertation). Charlottesville, VA: University of Virginia, R. J. Reynolds Institute for
14 Environmental Science.
15 Wimmers, A. J.; Moody, J. L. (2001) A fixed-layer estimation of upper-tropospheric specific humidity from the
16 GOES water vapor channel: parameterization and validation of the altered brightness temperature product.
17 J. Geophys. Res. [Atmos.] 106: 17,115-17,132.
18 Wimmers, A. J.; Moody, J. L. (2004a) Tropopause folding at satellite-observed spatial gradients: 1. Verification of
19 an empirical relationship. J. Geophys. Res. [Atmos.] 109(D19306): 10.1029/2003JD004145.
20 Wimmers, A. J.; Moody, J. L. (2004b) Tropopause folding at satellite-observed spatial gradients: 2. Development of
21 an empirical model. J. Geophys. Res. [Atmos.] 109(D19307): 10.1029/2003JD004146.
22 Wimmers, A. J.; Moody, J. L.; Browell, E. V; Hair, J. W.; Grant, W. B.; Butler, C. F.; Fenn, M. A.; Schmidt, C. C.;
23 Li, J.; Ridley, B. A. (2003) Signatures of tropopause folding in satellite imagery. J. Geophys. Res. (Atmos.)
24 108(04): 10.1029/2001JD001358.
25 Winer, A. M.; Peters, J. W.; Smith, J. P.; Pitts, J. N, Jr. (1974) Response of commercial chemiluminescent NO-NO2
26 analyzers to other nitrogen-containing compounds. Environ. Sci. Technol. 8: 1118-1121.
27 Winer, A. M.; Arey, J.; Atkinson, R.; Aschmann, S. M.; Long, W. D.; Morrison, C. L.; Olszyk, D. M. (1992)
28 Emission rates of organics from vegetation in California's Central Valley. Atmos. Environ. Part A
29 26:2647-2659.
30 Winner, D. A.; Cass, G. R. (2001) Modeling the long-term frequency distribution of regional ozone concentrations
31 using synthetic meteorology. Environ. Sci. Technol. 35: 3718-3726.
32 Witter, M.; Berndt, T.; Boge, O.; Stratmann, F.; Heintzenberg, J. (2002) Gas-phase ozonolysis: rate coefficients for a
33 series of terpenes and rate coefficients and OH yields for 2-methyl-2-butene and 2,3-dimethyl-2-butene. Int. J.
34 Chem. Kinet. 34: 394-403.
3 5 Wolff, G. T. (1993) On a NOx-focused control strategy to reduce O3. J. Air Waste Manage. Assoc. 43: 1593-1596.
36 Wolff, G. T.; Korsog, P. E. (1992) Ozone control strategies based on the ratio of volatile organic compounds to
37 nitrogen oxides. J. Air Waste Manage. Assoc. 42: 1173-1177.
3 8 Wolff, G. T.; Lioy, P. J. (1978) An empirical model for forecasting maximum daily ozone levels in the northeastern
39 U.S. J. AirPollut. Control Assoc. 28: 1034-1038.
40 Wolff, G. T.; Lioy, P. J. (1980) Development of an ozone river associated with synoptic scale episodes in the eastern
41 United States. Environ. Sci. Technol. 14: 1257-1260.
42 Wolff, G. T.; Lioy, P. J.; Meyers, R. E.; Cederwall, R. T.; Wight, G. D.; Pasceri, R. E.; Taylor, R. S. (1977)
43 Anatomy of two ozone transport episodes in the Washington, D.C., to Boston, Mass., corridor. Environ. Sci.
44 Technol. 11: 506-510.
45 Wolff, G. T.; Lioy, P. J.; Wight, G. D.; Meyers, R. E.; Cederwall, R. T. (1977) An investigation of long-range
46 transport of ozone across the midwestern and eastern United States. Atmos. Environ. 11: 797-802.
47 Wolff, G. T.; Lioy, P. J.; Wight, G. D.; Pasceri, R. E. (1977) Aerial investigation of the ozone plume phenomenon.
48 J. Air Pollut. Control Assoc. 27: 460-463.
49 Wolff, G. T.; Lioy, P. J.; Wight, G. D.; Pasceri, R. E. (1977) Aerial investigation of photochemical oxidants over the
50 northeast. In: Bufalini, J. J.; Lonneman, W. A., eds. Proceedings of symposium on 1975 northeast oxidant
51 transport study; January 1976; Research Triangle Park, NC. Research Triangle Park, NC: U.S. Environmental
52 Protection Agency, Environmental Sciences Research Laboratory; pp. 70-86; EPA report no.
53 EPA-600/3-77-017. Available from: NTIS, Springfield, VA; PB-265370.
54 World Meteorological Organization. (1990) Scientific assessment of stratospheric ozone: 1989. Volume II.
55 Appendix: AFEAS report. Geneva, Switzerland: Global Ozone Research Monitoring Project; report no. 20.
56 World Meteorological Organization. (1992) Scientific assessment of ozone depletion: 1991 [preprint]. Geneva,
57 Switzerland: World Meteorological Organization; report no. 25.
January 2005 AX2-191 DRAFT-DO NOT QUOTE OR CITE
-------
1 Wright, E. S.; Vang, M. I; Finkelstein, J. N.; Mavis, R. D. (1982) Changes inphospholipid biosynthetic enzymes in
2 type II cells and alveolar macrophages isolated from rat lungs after NO2 exposure. Toxicol. Appl. Pharmacol.
3 66:305-311.
4 Wunderli, S.; Gehrig, R. (1991) Influence of temperature on formation and stability of surface PAN and ozone.
5 A two year field study in Switzerland. Atmos. Environ. Part A 25: 1599-1608.
6 Yienger, J. J.; Levy, H., II. (1995) Empirical model of global soil-biogenic NOX emissions. J. Geophys. Res.
7 100(06): 11,447-11,464.
8 Young, V. L.; Kieser, B. N.; Chen, S. P.; Nii, H. (1997) Seasonal trends and local influences on nonmethane
9 hydrocarbon concentrations in the Canadian boreal forest. J. Geophys. Res. (Atmos.) 102: 5913-5918.
10 Yu, J.; Jeffries, H. E. (1997) Atmospheric photooxidation of alkylbenzenes II. evidence of formation of epoxide
11 intermediates. Atmos. Environ. 31: 2281-2287.
12 Yu, J.; Jeffries, H. E.; Sexton, K. G. (1997) Atmospheric photooxidation of alkylbenzenes I. carbonyl product
13 analyses. Atmos. Environ. 31: 2261-2280.
14 Yung, Y. L.; Pinto, J. P.; Watson, R. T.; Sander, S. P. (1980) Atmospheric bromine and ozone perturbations in the
15 lower stratosphere. J. Atmos. Sci. 37: 339-353.
16 Zafiriou, O. C.; True, M. B. (1979) Nitrate photolysis in seawater by sunlight. Mar. Chem. 8: 33-42.
17 Zanis, P.; Trickl, T.; Stohl, A.; Wernli, H.; Cooper, O.; Zerefos, C.; Gaeggeler, H.; Schnabel, C.; Tobler, L.; Kubik,
18 P. W.; Priller, A.; Scheel, H. E.; Kanter, H. J.; Cristofanelli, P.; Forster, C.; James, P.; Gerasopoulos, E.;
19 Delcloo, A.; Papayannis, A.; Claude, H. (2003) Forecast, observation and modelling of a deep stratospheric
20 intrusion event over Europe. Atmos. Chem. Phys. 3: 763-777.
21 Zenker, T.; Fischer, H.; Nikitas, C.; Parchatka, U.; Harris, G. W.; Mihelcic, D.; Mtisgen, P.; Patz, H. W.; Schultz,
22 M.; Volz-Thomas, A.; Schmitt, R.; Behmann, T.; Wiessenmayer, M.; Burrows, J. P. (1998) Intercomparison
23 of NO, NO2, NOp O3, and ROX measurements during the oxidizing capacity of the tropospheric atmosphere
24 (OCTA) campaign 1993 at Izana. J. Geophys. Res. [Atmos.] 103: 13,615-13,634.
25 Zhang, Y.; Carmichael, G. R. (1999) The role of mineral aerosol in tropospheric chemistry in East Asia~a model
26 study. J. Appl. Meteorol. 38: 353-366.
27 Zhang, J. S.; Rao, S. T. (1999) The role of vertical mixing in the temporal evolution of ground-level ozone
28 concentrations. J. Appl. Meteorol. 38: 1674-1691.
29 Zhang, D.-L.; Zheng, W.-Z. (2004) Diurnal cycles of surface winds and temperatures as simulated by five boundary
30 layer parameterizations. J. Appl. Meteorol. 43: 157-169.
31 Zhang, S.-H.; Shaw, M.; Seinfeld, J. H.; Flagan, R. C. (1992) Photochemical aerosol formation from a-pinene and
32 p-pinene. J. Geophys. Res. [Atmos.] 97: 20,717-20,729.
33 Zhang etal., 1997.
34 Zhang, K.; Juiting, M.; Civerolo, K. C.; Berman, S.; Ku, J.; Rao, S. T.; Doddridge, B.; Philbrick, R. C.; Clark, R.
3 5 (2001) Numerical investigation of boundary layer evolution and nocturnal low-level jets: local versus
36 non-local PEL schemes. Environ. Fluid Mech. 1: 171-208.
37 Zhang, D.; Lei, W. F.; Zhang, R. Y. (2002) Mechanism of OH formation from ozonolysis of isoprene: kinetics and
38 product yields. Chem. Phys. Lett. 358: 171-179.
39 Zhang, D.; Zang, J.; Shi, G.; Iwasaka, Y.; Matsuki, A.; Trochkine, D. (2003) Mixture state of individual Asian dust
40 particles at a coastal site of Qingdao, China. Atmos. Environ. 37: 3895-3901.
41 Zhou, X.; Mopper, K. (1990) Determination of photochemically produced hydroxyl radicals in seawater and
42 freshwater. Mar. Chem. 30: 71-88.
43 Zimmermann, J.; Poppe, D. (1993) Nonlinear chemical couplings in the tropospheric NOX-HOX gas phase chemistry.
44 J. Atmos. Chem. 17: 141-155.
45 Zondlo, M. A.; Barone, S. B.; Tolbert, M. A. (1998) Condensed-phase products in heterogeneous reactions: N2O5,
46 C1ONO2, and HNO3 reacting on ice films at 185 K. J. Phys. Chem. A 102: 5735-5748.
47
January 2005 AX2-192 DRAFT-DO NOT QUOTE OR CITE
-------
i 3. ENVIRONMENTAL CONCENTRATIONS,
2 PATTERNS, AND EXPOSURE ESTIMATES
3
4
5 3.1 INTRODUCTION
6 Identification and Use of Existing Air Quality Data
1 Topics discussed in this chapter include the characterization of ambient air quality data for
8 ozone (O3), the uses of these data in assessing the exposure of vegetation to O3, concentrations of
9 O3 in microenvironments, and a discussion of the currently available human exposure data and
10 exposure model development. The information contained in this chapter pertaining to ambient
11 concentrations is taken primarily from the U.S. Environmental Protection Agency Air Quality
12 System (AQS; formerly the AIRS database). The AQS contains readily accessible detailed,
13 hourly data that has been subject to the Agency's quality control and assurance procedures. Data
14 available in AQS were collected from 1979 to 2001. As discussed in previous versions of the O3
15 Air Quality Criteria Document or AQCD (U.S. Environmental Protection Agency, 1986, 1996),
16 the data available prior to 1979 may be unreliable due to calibration problems and uncertainties.
17 As indicated in the 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996), O3 is
18 the only photochemical oxidant other than nitrogen dioxide (NO2) that is routinely monitored
19 and for which a comprehensive aerometric database exists. Data for peroxyacetyl nitrate (PAN)
20 and hydrogen peroxide (H2O2) have been obtained only as part of special research field
21 investigations. Consequently, no data on nationwide patterns of occurrence are available for
22 these non-O3 oxidants; nor are extensive data available on the relationships of levels and patterns
23 of these oxidants to those of O3.
24
25 Characterizing Ambient Ozone Concentrations
26 In this chapter, data are analyzed for the purpose of providing information on specific
27 issues of exposure-response relationships that are considered in the later chapters addressing O3
28 exposure effects. It is important to distinguish among concentration, exposure, and dose when
29 using air quality data to assess human health and vegetation effects. For this chapter, the
30 following definitions apply.
January 2005 3-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 The "concentration" of a specific air pollutant is typically defined as the amount (mass) of
2 that material per unit volume of air. Air pollution monitors measure pollutant concentrations,
3 which may or may not provide accurate exposure estimates. However, most of the data
4 presented herein are expressed as "mixing ratios" in terms of a volume-to-volume ratio,
5 expressed as parts per million (ppm) or parts per billion (ppb). Data expressed this way are often
6 referred to as concentrations, both in the literature and in the text, following common usage.
7 The term "exposure" may generally be defined as the concentration of a pollutant
8 encountered by the subject (animal, human, or plant) for a duration of time. Exposure implies
9 that such an encounter leads to intake through the contact surface. A measured concentration
10 functions as a surrogate for an exposure only to the degree to which it represents concentrations
11 actually experienced by the subject under otherwise similar conditions.
12 The term "dose" is defined as that mass of pollutant delivered to a target. This term has
13 numerous quantitative descriptions, so the context of the use of this term and the units within the
14 document must be considered. A more in-depth description of exposure and dose appears later
15 in this chapter for humans and in Chapter 9 for vegetation and ecosystems.
16 In summarizing the hourly average concentrations in this chapter, specific attention is
17 given to the relevance of the exposure indicators used. For example, for human health
18 considerations, concentration (or exposure) indicators such as the daily maximum 1-h average
19 concentrations, as well as the number of daily maximum 8-h average concentrations, are used to
20 characterize information in the population-oriented monitor locations. For vegetation, several
21 different types of exposure indicators are used. The peak-weighted, cumulative exposure
22 indicators used in this chapter for characterizing vegetation exposures are SUM06 and SUM08
23 (the sums of all hourly average concentrations > 0.06 and 0.08 ppm, respectively) and W126 (the
24 sum of the hourly average concentrations that have been weighted according to a sigmoid
25 function [see Lefohn and Runeckles, 1987] that is based on a hypothetical vegetation response).
26 Further discussion of these exposure indices is presented in Chapter 9.
27 Hourly average concentration information is summarized for urban versus rural or
28 nonurban (i.e., forested and agricultural) areas in the United States in Annex AX3. The
29 distribution of O3 or its precursors at a rural site near an urban source is affected by wind
30 direction (i.e., whether the rural site is located up- or downwind from the source of O3
31 precursors). It is difficult to apply land use designations to the generalization of exposure
January 2005 3-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 regimes that may be experienced in urban versus rural areas, because the land use
2 characterization of "rural" does not imply that a specific location is isolated from anthropogenic
3 influences. Rather, the characterization implies only the current use of the land. Since it is
4 possible for O3 produced from urban area emissions to be transported to more rural downwind
5 locations, elevated O3 concentrations can occur at considerable distances from urban centers.
6 Nitrogen oxides (NOX) often depress O3 concentrations through titration in urban cores with
7 heavy traffic. Due to the less chemical scavenging in nonurban areas, O3 tends to persist longer
8 in nonurban than in urban areas; thus, exposures may be greater in downwind nonurban
9 locations. For example, Logan (1989) has noted that hourly average O3 concentrations above
10 0.08 ppm are common in rural areas of the eastern United States in spring and summer, but are
11 unusual at remote western sites. Consequently, for the purposes of comparing exposure regimes
12 that may be characteristic of clean locations in the United States with those that are urban
13 influenced (i.e., located in either urban or rural locations), this chapter characterizes data
14 collected from those stations whose locations appear to be isolated from large-scale
15 anthropogenic influences (i.e., relatively clean remote sites) in Annex AX3, Section AX3.2.
16 Acknowledging the photochemical and insolation-dependent nature of O3 formation, the
17 U.S. Environmental Protection Agency (U.S. EPA) has established allowable "ozone seasons"
18 for the required measurement of ambient O3 concentrations for different locations within the
19 United States and the U.S. territories (CFR, 2000). Table 3-1 shows the O3 seasons during
20 which continuous, hourly averaged O3 concentrations must be monitored.
21 The use of ambient monitoring data in epidemiological studies provides the exposure
22 indicator for individuals exercising outside and serves as a measure of relative exposures since
23 ambient photochemical production is the major source of O3. The lack of association between
24 ambient air concentration and total exposure results in some misclassification when using
25 monitoring data or ambient air modeling results as surrogates for exposure. The reliance on
26 monitoring data from a single site or averaged over large regions as exposure surrogates may not
27 adequately represent regional variations or differences in exposures with location and activity.
28 This may increase the potential for misclassification of both absolute and relative exposures.
29 Exposure models rather than chemistry transport models (CTMs) have been used to refine
30 population exposure estimates, decreasing misclassification of exposures. For cohort studies,
31 exposure measurements using passive personal samplers combined with modeling can provide
January 2005 3-3 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-1. 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 — December
April — September
April — October
January — December
April — October
April — October
May — September
April — October
March — November
May — September
April — October
January — December
April — September
April — October
June — September
March — October
January — December
March — October
May — September
April — September
April — October
May — September
April — October
April 15 — October 15
April — October
January — December
January — December
January — December
1 The ozone season is defined differently in different sections of Texas.
Source: CFR(2000).
January 2005
5-4
DRAFT-DO NOT QUOTE OR CITE
-------
1 detailed individual exposure data to minimize misclassification. Improvements in models will
2 help better characterize exposures by location for activities deemed most likely to increase risk,
3 such as exercise, and for sensitive populations.
4
5
6 3.2 AMBIENT AIR QUALITY DATA FOR OZONE
7 Ozone Air Quality at Urban, Suburban, andNonurban Sites
8 Often there is a difference in the distribution of hourly average concentrations between
9 urban and nonurban areas. Ozone concentrations measured at center-city sites are often lower
10 than at surrounding rural sites. In some urban areas, maximum hourly average concentrations
11 exceed 0.120 ppm. However, only about 1% of the hourly average concentrations generally
12 exceed 0.100 ppm at sites in these areas. Furthermore, monitoring sites in polluted areas tend to
13 experience frequent hourly average O3 concentrations at or near minimum detectable levels. The
14 highest values of the second highest daily maximum of the O3 hourly average concentrations are
15 observed in the Texas Gulf Coast and Southern California, but high levels of O3 also occur in
16 the Northeast Corridor, and other heavily populated regions of the United States as shown in
17 Figure 3-1 and Table 3-2. Metropolitan Statistical Areas (MSAs), which experience elevated
18 second highest daily maximum hourly average concentrations, also experience elevated fourth
19 highest 8-h daily maximum concentrations. There are considerably more MSAs and
20 Consolidated Metropolitan Statistical Areas (CMSAs) that experience fourth highest 8-h daily
21 maximum concentrations > 0.085 ppm than MSAs and CMSAs that experience second highest
22 daily maximum hourly average concentrations > 0.125 ppm (Figure 3-2 and Table 3-3).
23 It is difficult to identify a set of unique O3 distribution patterns that adequately describes
24 the hourly average concentrations experienced at monitoring sites in nonurban locations, because
25 many nonurban sites in the United States are influenced by local sources of pollution or
26 long-range transport of O3 or its precursors. Using hourly averaged data from AQS for a select
27 number of rural monitoring sites, Table 3-4 summarizes the percentiles of the hourly average
28 O3 concentrations, the number of occurrences of the hourly average concentration > 0.08 and
29 0.10 ppm, the 7-month sum of all hourly average concentrations > 0.06 ppm, and the 7-month
30 W126 exposure index. Note the large variation in the number of hourly average concentrations
January 2005 3-5 DRAFT-DO NOT QUOTE OR CITE
-------
1-h(ppm)
A 0.061-0.093
* 0.094-0.111
• 0.112-0.132
* 0.133-0.175
Figure 3-1. Second highest daily maximum 1-h O3 concentrations.
Source: U.S. Environmental Protection Agency (2003a).
1
2
3
4
5
9
10
11
> 0.08 and 0.10 ppm. The sites are influenced either by local sources or by transport of O3 or its
precursors.
Ozone Air Quality Data at Relatively Remote Monitoring Sites
In characterizing hourly average concentration data obtained at nonurban locations, it is
important to carefully quantify the distribution of the hourly average concentrations that are
folded into the cumulative exposure indices used in predicting vegetation effects. For example,
in 2001, although the Crestline, CA site in the San Bernardino National Forest and the Cove
Mountain site in the Great Smoky Mountains National Park experience similar cumulative
exposures (i.e., SUM06 and W126 values), the Crestline site experienced considerably more
high hourly average concentrations > 0.10 ppm (i.e., 369) than the Cove Mountain site (i.e., 5).
January 2005
5-6
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2. The Second Highest Daily Maximum One-Hour Ozone Concentration (ppm)
by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan Statistical Area
(CMSA) for the Years 1999 to 2001
MSA/CMSA
Albany-Schenectady-Troy, NY MSA
Albuquerque, NM MSA
Allentown-Bethlehem-Easton, PA MSA
Altoona, PA MSA
Appleton-Oshkosh-Neenah, WI MSA
Asheville, NC MSA
Atlanta, GA MSA
Augusta-Aiken, GA-SC MSA
Austin-San Marcos, TX MSA
Bakersfield, CA MSA
Bangor, ME MSA
Barnstable-Yarmouth, MA MSA
Baton Rouge, LA MSA
Beaumont-Port Arthur, TX MSA
Bellingham, WA MSA
Benton Harbor, MI MSA
Biloxi-Gulfport-Pascagoula, MS MSA
Birmingham, AL MSA
Boston- Worcester-Lawrence, MA-NH-ME-CT CMSA
Brownsville-Harlingen-San Benito, TX MSA
Buffalo-Niagara Falls, NY MSA
Burlington, VT MSA
Canton-Massillon, OH MSA
Cedar Rapids, IA MSA
Champaign-Urbana, IL MSA
Charleston, WV MSA
Charleston-North Charleston, SC MSA
Charlotte-Gastonia-Rock Hill, NC-SC MSA
1999
0.107
0.097
0.126
0.111
0.105
0.099
0.156
0.108
0.110
0.136
0.088
0.127
0.121
0.103
0.062
0.107
0.107
0.131
0.125
0.075
0.102
0.093
0.108
0.096
0.108
0.130
0.101
0.130
2000
0.088
0.093
0.114
0.104
0.085
0.107
0.158
0.115
0.107
0.142
0.107
0.139
0.160
0.063
0.107
0.135
0.127
0.101
0.080
0.105
0.080
0.104
0.083
0.084
0.094
0.105
0.141
2001
0.112
0.088
0.126
0.107
0.102
0.091
0.125
0.103
0.097
0.134
0.105
0.139
0.117
0.103
0.061
0.117
0.099
0.114
0.136
0.074
0.116
0.083
0.106
0.084
0.080
0.107
0.085
0.142
January 2005
5-7
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Chattanooga, TN-GA MSA
Chicago-Gary-Kenosha, IL-IN-WI CMSA
Chico-Paradise, CA MSA
Cincinnati-Hamilton, OH-KY-IN CMSA
Clarksville-Hopkinsville, TN-KYMSA
Cleveland-Akron, OH CMSA
Colorado Springs, CO MSA
Columbia, SC MSA
Columbus, GA-AL MSA
Columbus, OH MSA
Corpus Christi, TX MSA
Dallas-Fort Worth, TX CMSA
Davenport-Moline-Rock Island, IA-IL MSA
Dayton-Springfield, OH MSA
Daytona Beach, FL MSA
Decatur, AL MSA
Decatur, IL MSA
Denver-Boulder-Greeley, CO CMSA
Des Moines, IA MSA
Detroit-Ann Arbor-Flint, MI CMSA
Dover, DE MSA
Duluth-Superior, MN-WI MSA
El Paso, TX MSA
Elkhart-Goshen, IN MSA
Elmira, NY MSA
Erie, PA MSA
Eugene-Springfield, or MSA
Evansville-Henderson, IN-KY MSA
1999
0.122
0.126
0.110
0.119
0.115
0.118
0.075
0.117
0.110
0.144
0.103
0.154
0.099
0.127
0.087
0.103
0.102
0.105
0.083
0.121
0.120
0.082
0.108
0.085
0.092
0.112
0.084
0.114
2000
0.124
0.102
0.091
0.110
0.108
0.110
0.088
0.116
0.114
0.117
0.099
0.126
0.089
0.106
0.088
0.110
0.092
0.107
0.082
0.102
0.126
0.074
0.122
0.080
0.089
0.095
0.078
0.097
2001
0.107
0.124
0.100
0.115
0.096
0.118
0.085
0.107
0.090
0.111
0.092
0.137
0.089
0.100
0.085
0.087
0.078
0.105
0.069
0.122
0.117
0.071
0.116
0.066
0.094
0.104
0.081
0.095
January 2005
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Fargo-Moorhead, ND-MN MSA
Fayetteville, NC MSA
Flagstaff, AZ-UT MSA
Fort Collins-Loveland, CO MSA
Fort Myers-Cape Coral, FL MSA
Fort Fierce-Port St. Lucie, FL MSA
Fort Wayne, IN MSA
Fresno, CA MSA
Gainesville, FL MSA
Grand Rapids-Muskegon-Holland, MI MSA
Green Bay, WI MSA
Greensboro-Winston-Salem-High Point, NC MSA
Greenville, NC MSA
Greenville-Spartanburg-Anderson, SC MSA
Harrisburg-Lebanon-Carlisle, PA MSA
Hartford, CT MSA
Hickory -Morganton-Lenoir, NC MSA
Honolulu, HI MSA
Houma, LA MSA
Houston-Galveston-Brazoria, TX CMSA
Huntington- Ashland, WV-KY-OH MSA
Huntsville, AL MSA
Indianapolis, IN MSA
Jackson, MS MSA
Jacksonville, FL MSA
Jamestown, NY MSA
Janesville-Beloit, WI MSA
Johnson City-Kingsport-Bristol, TN-VA MSA
1999
0.073
0.120
0.086
0.089
0.096
0.083
0.101
0.145
0.098
0.116
0.097
0.126
0.109
0.122
0.126
0.161
0.115
0.054
0.203
0.122
0.106
0.114
0.110
0.103
0.103
0.105
0.111
2000
0.073
0.106
0.082
0.096
0.091
0.079
0.099
0.146
0.096
0.123
0.090
0.116
0.109
0.115
0.110
0.116
0.107
0.048
0.124
0.194
0.094
0.111
0.102
0.099
0.114
0.113
0.098
0.126
2001
0.069
0.108
0.074
0.088
0.079
0.095
0.098
0.146
0.096
0.118
0.107
0.122
0.091
0.108
0.105
0.139
0.099
0.051
0.106
0.170
0.113
0.088
0.114
0.095
0.093
0.109
0.093
0.110
January 2005
5-9
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Johnstown, PA MSA
Kalamazoo-Battle Creek, MI MSA
Kansas City, MO-KS MSA
Knoxville, TN MSA
Lafayette, LA MSA
Lake Charles, LA MSA
Lakeland- Winter Haven, FL MSA
Lancaster, PA MSA
Lansing-East Lansing, MI MSA
Laredo, TX MSA
Las Cruces, NM MSA
Las Vegas, NV-AZ MSA
Lawton, OK MSA
Lexington, KY MSA
Lima, OH MSA
Lincoln, NE MSA
Little Rock-North Little Rock, AR MSA
Longview-Marshall, TX MSA
Los Angeles-Riverside-Orange County, CA CMSA
Louisville, KY-IN MSA
Macon, GA MSA
Madison, WI MSA
McAllen-Edinburg-Mission, TX MSA
Medford- Ashland, OR MSA
Melbourne-Titusville-Palm Bay, FL MSA
Memphis, TN-AR-MS MSA
Merced, CA MSA
Miami-Fort Lauderdale, FL CMSA
1999
0.107
0.103
0.116
0.129
0.094
0.127
0.101
0.127
0.101
0.084
0.102
0.097
0.089
0.114
0.107
0.062
0.107
0.134
0.159
0.124
0.133
0.098
0.086
0.087
0.130
0.125
0.113
2000
0.104
0.090
0.124
0.131
0.123
0.133
0.102
0.107
0.091
0.085
0.123
0.094
0.094
0.086
0.100
0.072
0.114
0.131
0.174
0.112
0.131
0.087
0.088
0.079
0.093
0.123
0.120
0.094
2001
0.106
0.101
0.108
0.109
0.090
0.105
0.109
0.127
0.106
0.071
0.104
0.100
0.094
0.092
0.096
0.061
0.102
0.111
0.175
0.106
0.115
0.088
0.092
0.081
0.094
0.121
0.113
0.106
January 2005
3-10
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Milwaukee-Racine, WI CMSA
Minneapolis-St. Paul, MN-WI MSA
Mobile, AL MSA
Modesto, CA MSA
Monroe, LA MSA
Montgomery, AL MSA
Nashville, TN MSA
New London-Norwich, CT-RI MSA
New Orleans, LA MSA
NY-Northern NJ-Long Island, NY-NJ-CT-PA CMSA
Norfolk- Virginia Beach-Newport News, VA-NC MSA
Ocala, FL MSA
Oklahoma City, OK MSA
Omaha, NE-IA MSA
Orlando, FL MSA
Owensboro, KY MSA
Parkersburg-Marietta, WV-OH MSA
Pensacola, FL MSA
Peoria-Pekin, IL MSA
Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD CMSA
Phoenix-Mesa, AZ MSA
Pittsburgh, PA MSA
Pittsfield,MAMSA
Portland, ME MSA
Portland-Salem, OR-WA CMSA
Providence-Fall River- Warwick, RI-MA MSA
Provo-Orem, UT MSA
Raleigh-Durham-Chapel Hill, NC MSA
1999
0.122
0.088
0.104
0.111
0.097
0.110
0.123
0.127
0.117
0.154
0.135
0.097
0.097
0.093
0.101
0.102
0.123
0.106
0.099
0.152
0.119
0.132
0.092
0.120
0.094
0.133
0.109
0.134
2000
0.098
0.089
0.118
0.110
0.103
0.111
0.122
0.135
0.124
0.136
0.099
0.093
0.100
0.083
0.106
0.082
0.105
0.118
0.083
0.128
0.107
0.111
0.089
0.082
0.118
0.095
0.116
2001
0.122
0.109
0.095
0.111
0.090
0.094
0.110
0.109
0.111
0.146
0.100
0.089
0.097
0.076
0.108
0.086
0.108
0.098
0.084
0.131
0.106
0.112
0.112
0.124
0.093
0.144
0.094
0.113
January 2005
3-11
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Reading, PA MSA
Redding, CA MSA
Reno, NV MSA
Richmond-Petersburg, VA MSA
Roanoke, VA MSA
Rochester, NY MSA
Rockford, IL MSA
Rocky Mount, NC MSA
Sacramento-Yolo, CA CMSA
Salinas, CA MSA
Salt Lake City-Ogden, UT MSA
San Antonio, TX MSA
San Diego, CAMS A
San Francisco-Oakland-San Jose, CA CMSA
San Luis Obispo-Atascadero-Paso Robles, CA MSA
Santa Barbara-Santa Maria-Lompoc, CA MSA
Sarasota-Bradenton, FL MSA
Savannah, GA MSA
Scranton-Wilkes-Barre-Hazleton, PA MSA
Seattle-Tacoma-Bremerton, WA CMSA
Sharon, PA MSA
Sheboygan, WI MSA
Shreveport-Bossier City, LA MSA
Sioux Falls, SD MSA
South Bend, IN MSA
Spokane, WA MSA
Springfield, IL MSA
Springfield, MA MSA
1999
0.128
0.113
0.097
0.133
0.101
0.093
0.104
0.137
0.075
0.112
0.109
0.114
0.144
0.094
0.095
0.112
0.107
0.115
0.091
0.108
0.130
0.108
0.068
0.107
0.073
0.099
0.113
2000
0.105
0.098
0.088
0.112
0.095
0.088
0.084
0.106
0.134
0.084
0.100
0.095
0.123
0.126
0.082
0.100
0.107
0.102
0.093
0.095
0.098
0.106
0.129
0.076
0.095
0.082
0.100
0.099
2001
0.125
0.093
0.090
0.119
0.101
0.100
0.086
0.099
0.129
0.078
0.118
0.092
0.118
0.118
0.092
0.103
0.114
0.085
0.104
0.088
0.113
0.122
0.105
0.088
0.108
0.084
0.095
0.132
January 2005
3-12
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-2 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Springfield, MO MSA
St. Louis, MO-IL MSA
State College, PA MSA
Steubenville-Weirton, OH-WVMSA
Stockton-Lodi, CA MSA
Syracuse, NY MSA
Tallahassee, FL MSA
Tampa-St. Petersburg-Clearwater, FL MSA
Terre Haute, IN MSA
Toledo, OH MSA
Tucson, AZ MSA
Tulsa, OK MSA
Tyler, TX MSA
Utica-Rome, NY MSA
Victoria, TX MSA
Visalia-Tulare-Porterville, CA MSA
Washington-Baltimore, DC-MD-VA-WV CMSA
Wausau, WI MSA
West Palm Beach-Boca Raton, FL MSA
Wheeling, WV-OH MSA
Wichita, KS MSA
Williamsport, PA MSA
Wilmington, NC MSA
York, PA MSA
Youngstown- Warren, OH MSA
Yuba City, CA MSA
1999
0.095
0.128
0.099
0.111
0.130
0.099
0.094
0.116
0.093
0.128
0.092
0.116
0.118
0.089
0.102
0.125
0.152
0.095
0.104
0.100
0.095
0.091
0.081
0.121
0.109
0.106
2000
0.092
0.123
0.109
0.100
0.111
0.083
0.092
0.108
0.088
0.095
0.085
0.122
0.083
0.094
0.119
0.128
0.081
0.093
0.093
0.093
0.092
0.097
0.112
0.097
0.101
2001
0.091
0.122
0.097
0.096
0.112
0.097
0.085
0.118
0.096
0.111
0.084
0.107
0.098
0.100
0.085
0.126
0.142
0.078
0.098
0.104
0.096
0.094
0.089
0.104
0.105
0.105
January 2005
3-13
DRAFT-DO NOT QUOTE OR CITE
-------
8-h (ppm)
A 0.051 -0.072
* 0.073-0.085
• 0,086-0,098
if 0.099-0.133
Figure 3-2. Fourth highest 8-h daily maximum O3 concentration.
Source: U.S. Environmental Protection Agency (2003a).
1 It is important to characterize hourly average O3 concentrations at relatively remote
2 monitoring sites so that assessments of the possible effects of O3 on human health and vegetation
3 use ranges of hourly average concentrations in their experiments that mimic the range of O3
4 exposures that occur under ambient conditions. No hourly average concentrations > 0.08 ppm
5 were observed at monitoring sites in Redwood NP (CA), Olympic NP (WA), Glacier NP (MT),
6 Denali NP (AK), Badlands (SD), and Custer NF (MT) during the months of April to October
7 from 1988 to 2001 (Table 3-5). There were eight occurrences of hourly average concentrations
8 > 0.08 ppm from April to October of 1997 at the monitoring site in Theodore Roosevelt NP
9 (ND). However, no hourly average concentrations > 0.08 ppm were observed from April to
10 October in any other year at this site. Except for 1988, the year in which there were major forest
11 fires at Yellowstone NP (WY), the monitoring site located there experienced no hourly average
January 2005
3-14
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3. The Fourth Highest Daily Maximum Eight-Hour Ozone Concentration (ppm)
by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan Statistical Area
(CMSA) for the Years 1999 to 2001
MSA/CMSA
Albany-Schenectady-Troy, NY MSA
Albuquerque, NM MSA
Allentown-Bethlehem-Easton, PA MSA
Altoona, PA MSA
Appleton-Oshkosh-Neenah, WI MSA
Asheville, NC MSA
Atlanta, GA MSA
Augusta-Aiken, GA-SC MSA
Austin-San Marcos, TX MSA
Bakersfield, CA MSA
Bangor, ME MSA
Barnstable-Yarmouth, MA MSA
Baton Rouge, LA MSA
Beaumont-Port Arthur, TX MSA
Bellingham, WA MSA
Benton Harbor, MI MSA
Biloxi-Gulfport-Pascagoula, MS MSA
Birmingham, AL MSA
Boston- Worcester-Lawrence, MA-NH-ME-CT CMSA
Brownsville-Harlingen-San Benito, TX MSA
Buffalo-Niagara Falls, NY MSA
Burlington, VT MSA
Canton-Massillon, OH MSA
Cedar Rapids, IA MSA
Champaign-Urbana, IL MSA
Charleston, WV MSA
Charleston-North Charleston, SC MSA
Charlotte-Gastonia-Rock Hill, NC-SC MSA
1999
0.092
0.076
0.107
0.091
0.087
0.084
0.126
0.090
0.099
0.109
0.080
0.101
0.100
0.077
0.050
0.096
0.097
0.100
0.098
0.066
0.090
0.079
0.091
0.076
0.094
0.104
0.084
0.107
2000
0.072
0.076
0.092
0.080
0.068
0.090
0.113
0.093
0.088
0.111
0.083
0.101
0.097
0.052
0.077
0.091
0.099
0.082
0.064
0.085
0.071
0.087
0.075
0.073
0.085
0.082
0.101
2001
0.090
0.074
0.094
0.083
0.087
0.076
0.092
0.082
0.080
0.109
0.088
0.105
0.084
0.081
0.050
0.088
0.083
0.089
0.101
0.063
0.102
0.076
0.089
0.069
0.073
0.083
0.071
0.103
January 2005
3-15
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Chattanooga, TN-GA MSA
Chicago-Gary-Kenosha, IL-IN-WI CMSA
Chico-Paradise, CA MSA
Cincinnati-Hamilton, OH-KY-IN CMSA
Clarksville-Hopkinsville, TN-KYMSA
Cleveland-AKRON, OH CMSA
Colorado Springs, CO MSA
Columbia, SC MSA
Columbus, GA-AL MSA
Columbus, OH MSA
Corpus Christi, TX MSA
Dallas-Fort Worth, TX CMSA
Davenport-Moline-Rock Island, IA-IL MSA
Dayton-Springfield, OH MSA
Daytona Beach, FL MSA
Decatur, AL MSA
Decatur, IL MSA
Denver-Boulder-Greeley, CO CMSA
Des Moines, IA MSA
Detroit-Ann Arbor-Flint, MI CMSA
Dover, DE MSA
Duluth-Superior, MN-WI MSA
El Paso, TX MSA
Elkhart-Goshen, IN MSA
Elmira, NY MSA
Erie, PA MSA
Eugene-Springfield, OR MSA
Evansville-Henderson, IN-KY MSA
1999
0.098
0.102
0.087
0.096
0.092
0.103
0.064
0.094
0.097
0.103
0.085
0.112
0.083
0.096
0.075
0.092
0.087
0.081
0.073
0.096
0.097
0.074
0.071
0.077
0.082
0.096
0.068
0.098
2000
0.098
0.086
0.086
0.093
0.088
0.085
0.070
0.097
0.094
0.091
0.083
0.102
0.077
0.088
0.076
0.091
0.077
0.083
0.071
0.082
0.093
0.065
0.084
0.063
0.073
0.078
0.047
0.085
2001
0.087
0.099
0.088
0.088
0.082
0.099
0.070
0.091
0.079
0.090
0.077
0.098
0.079
0.084
0.073
0.077
0.071
0.082
0.060
0.095
0.091
0.062
0.078
0.055
0.082
0.089
0.066
0.081
January 2005
3-16
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Fargo-Moorhead, ND-MN MSA
Fayetteville, NC MSA
Flagstaff, AZ-UT MSA
Fort Collins-Loveland, CO MSA
Fort Myers-Cape Coral, FL MSA
Fort Fierce-Port St. Lucie, FL MSA
Fort Wayne, IN MSA
Fresno, CA MSA
Gainesville, FL MSA
Grand Rapids-Muskegon-Holland, MI MSA
Green Bay, WI MSA
Greensboro-Winston-Salem-High Point, NC MSA
Greenville, NC MSA
Greenville-Spartanburg-Anderson, SC MSA
Harrisburg-Lebanon-Carlisle, PA MSA
Hartford, CT MSA
Hickory -Morganton-Lenoir, NC MSA
Honolulu, HI MSA
Houma, LA MSA
Houston-Galveston-Brazoria, TX CMSA
Huntington- Ashland, WV-KY-OH MSA
Huntsville, AL MSA
Indianapolis, IN MSA
Jackson, MS MSA
Jacksonville, FL MSA
Jamestown, NY MSA
Janesville-Beloit, WI MSA
Johnson City-Kingsport-Bristol, TN-VA MSA
1999
0.066
0.100
0.076
0.074
0.081
0.071
0.090
0.105
0.080
0.103
0.085
0.100
0.093
0.100
0.104
0.107
0.094
0.048
0.087
0.124
0.097
0.093
0.096
0.084
0.080
0.092
0.093
0.089
2000
0.057
0.086
0.071
0.078
0.077
0.071
0.091
0.114
0.080
0.080
0.071
0.094
0.082
0.089
0.088
0.089
0.091
0.044
0.085
0.117
0.083
0.088
0.090
0.080
0.077
0.087
0.083
0.097
2001
0.060
0.084
0.070
0.070
0.068
0.075
0.082
0.113
0.079
0.095
0.088
0.096
0.077
0.090
0.091
0.102
0.088
0.042
0.084
0.110
0.088
0.080
0.093
0.078
0.075
0.090
0.084
0.086
January 2005
3-17
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Johnstown, PA MSA
Kalamazoo-Battle Creek, MI MSA
Kansas City, MO-KS MSA
Knoxville, TN MSA
Lafayette, LA MSA
Lake Charles, LA MSA
Lakeland- Winter Haven, FL MSA
Lancaster, PA MSA
Lansing-East Lansing, MI MSA
Laredo, TX MSA
Las Cruces, NM MSA
Las Vegas, NV-AZ MSA
Lawton, OK MSA
Lexington, KY MSA
Lima, OH MSA
Lincoln, NE MSA
Little Rock-North Little Rock, AR MSA
Longview-Marshall, TX MSA
Los Angeles-Riverside-Orange County, CA CMSA
Louisville, KY-IN MSA
Macon, GA MSA
Madison, WI MSA
McAllen-Edinburg-Mission, TX MSA
Medford- Ashland, OR MSA
Melbourne-Titusville-Palm Bay, FL MSA
Memphis, TN-AR-MS MSA
Merced, CA MSA
Miami-Fort Lauderdale, FL CMSA
1999
0.090
0.091
0.084
0.106
0.081
0.088
0.078
0.102
0.089
0.067
0.082
0.084
0.082
0.091
0.093
0.053
0.089
0.105
0.133
0.103
0.113
0.085
0.075
0.077
0.104
0.105
0.077
2000
0.086
0.070
0.091
0.100
0.092
0.090
0.079
0.090
0.077
0.070
0.081
0.082
0.085
0.077
0.085
0.057
0.092
0.099
0.122
0.090
0.097
0.071
0.077
0.067
0.078
0.093
0.103
0.077
2001
0.090
0.085
0.079
0.093
0.077
0.080
0.087
0.097
0.087
0.063
0.079
0.082
0.078
0.078
0.081
0.051
0.080
0.082
0.133
0.086
0.086
0.078
0.074
0.064
0.084
0.092
0.096
0.076
January 2005
3-18
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Milwaukee-Racine, WI CMSA
Minneapolis-St. Paul, MN-WI MSA
Mobile, AL MSA
Modesto, CA MSA
Monroe, LA MSA
Montgomery, AL MSA
Nashville, TN MSA
New London-Norwich, CT-RI MSA
New Orleans, LA MSA
NY-Northern NJ-Long Island, NY-NJ-CT-PA CMSA
Norfolk- Virginia Beach-Newport News, VA-NC MSA
Ocala, FL MSA
Oklahoma City, OK MSA
Omaha, NE-IA MSA
Orlando, FL MSA
Owensboro, KY MSA
Panama City, FL MSA
Parkersburg-Marietta, WV-OH MSA
Pensacola, FL MSA
Peoria-Pekin, IL MSA
Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD CMSA
Phoenix-Mesa, AZ MSA
Pittsburgh, PA MSA
Pittsfield,MAMSA
Portland, ME MSA
Portland-Salem, or-WA CMSA
Providence -Fall River- Warwick, RI-MA MSA
Provo-Orem, UT MSA
1999
0.097
0.077
0.085
0.090
0.082
0.092
0.101
0.096
0.091
0.113
0.097
0.083
0.084
0.080
0.084
0.090
0.097
0.086
0.082
0.112
0.091
0.101
0.075
0.089
0.072
0.091
0.083
2000
0.086
0.072
0.097
0.091
0.081
0.086
0.093
0.084
0.095
0.114
0.084
0.079
0.086
0.072
0.081
0.074
0.092
0.087
0.096
0.073
0.109
0.090
0.088
0.073
0.065
0.087
0.077
2001
0.102
0.078
0.078
0.094
0.077
0.077
0.086
0.090
0.084
0.108
0.085
0.074
0.082
0.063
0.079
0.073
0.082
0.085
0.082
0.080
0.105
0.086
0.093
0.092
0.097
0.069
0.105
0.076
January 2005
3-19
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Raleigh-Durham-Chapel Hill, NC MSA
Reading, PA MSA
Redding, CA MSA
Reno, NV MSA
Richmond-Petersburg, VA MSA
Roanoke, VA MSA
Rochester, NY MSA
Rockford, IL MSA
Rocky Mount, NC MSA
Sacramento-Yolo, CA CMSA
Salinas, CA MSA
Salt Lake City-Ogden, UT MSA
San Antonio, TX MSA
San Diego, CAMS A
San Francisco-Oakland-San Jose, CA CMSA
San Luis Obispo-Atascadero-Paso Robles, CA MSA
Santa Barbara-Santa Maria-Lompoc, CA MSA
Sarasota-Bradenton, FL MSA
Savannah, GA MSA
Scranton-Wilkes-Barre-Hazleton, PA MSA
Seattle-Tacoma-Bremerton, WA CMSA
Sharon, PA MSA
Sheboygan, WI MSA
Shreveport-Bossier city, LA MSA
Sioux Falls, SD MSA
South Bend, IN MSA
Spokane, WA MSA
Springfield, IL MSA
1999
0.108
0.102
0.094
0.076
0.100
0.089
0.089
0.082
0.092
0.106
0.063
0.080
0.091
0.092
0.088
0.077
0.079
0.085
0.083
0.096
0.072
0.091
0.093
0.094
0.058
0.090
0.065
0.075
2000
0.089
0.084
0.082
0.069
0.083
0.081
0.073
0.070
0.085
0.103
0.063
0.080
0.082
0.095
0.077
0.070
0.080
0.086
0.079
0.077
0.070
0.081
0.090
0.093
0.065
0.081
0.068
0.079
2001
0.089
0.099
0.077
0.075
0.091
0.089
0.086
0.078
0.085
0.105
0.063
0.084
0.081
0.096
0.081
0.073
0.083
0.086
0.067
0.088
0.071
0.094
0.102
0.084
0.065
0.089
0.071
0.073
January 2005
3-20
DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-3 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Springfield, MA MSA
Springfield, MO MSA
St. Louis, MO-IL MSA
State College, PA MSA
Steubenville-Weirton, OH-WVMSA
Stockton-Lodi, CA MSA
Syracuse, NY MSA
Tallahassee, FL MSA
Tampa-St. Petersburg-Clearwater, FL MSA
Terre Haute, IN MSA
Toledo, OH MSA
Tucson, AZ MSA
Tulsa, OK MSA
Tyler, TX MSA
Utica-Rome, NY MSA
Victoria, TX MSA
Visalia-Tulare-Porterville, CA MSA
Washington-Baltimore, DC-MD-VA-WV CMSA
Wausau, WI MSA
West Palm Beach-Boca Raton, FL MSA
Wheeling, WV-OH MSA
Wichita, KS MSA
Williamsport, PA MSA
Wilmington, NC MSA
York, PA MSA
Youngstown- Warren, OH MSA
Yuba City, CA MSA
1999
0.094
0.081
0.102
0.085
0.091
0.094
0.086
0.081
0.087
0.082
0.091
0.073
0.091
0.097
0.078
0.086
0.108
0.113
0.084
0.079
0.088
0.079
0.076
0.067
0.094
0.095
0.090
2000
0.079
0.078
0.088
0.079
0.080
0.082
0.074
0.076
0.083
0.075
0.081
0.077
0.088
0.087
0.067
0.079
0.105
0.099
0.073
0.075
0.071
0.080
0.073
0.080
0.090
0.080
0.083
2001
0.093
0.072
0.088
0.086
0.086
0.078
0.085
0.074
0.085
0.083
0.092
0.069
0.095
0.082
0.084
0.073
0.104
0.112
0.072
0.073
0.088
0.084
0.080
0.078
0.087
0.093
0.084
January 2005
3-21
DRAFT-DO NOT QUOTE OR CITE
-------
3
to
o
o
i iiuic j--t. ociisuniii ^"vjji ii — v^iiuuci ; i ci v.cninc LMSII luuiiun ui nuiniy v^iuiic v uiiiciiii iiiiuiis, nuinuci ui iiuiiiiy ITIC 0.08 and > 0.10, SUM06, and W126 Values for Selected Rural Ozone Monitoring Sites in 2001.
Concentrations in ppm.
AIRS Site
Name
Percentiles
Min. 10 30 50 70 90 95 99 Max
RURAL AGRICULTURAL
OJ
to
to
o
170491001
180970042
240030014
310550032
420070002
510610002
551390011
Effingham Co., IL
Indianapolis, IN
Anne Arundel, MD
Omaha, NE
Beaver Co., PA
Fauquier Co., VA
Oshkosh Co., WI
0.000 0.007 0.021 0.031 0.041 0.056 0.063 0.074 0.094
0.000 0.007 0.021 0.033 0.044 0.061 0.068 0.078 0.088
0.000 0.004 0.019 0.031 0.043 0.062 0.072 0.096 0.130
0.000 0.015 0.023 0.030 0.038 0.050 0.056 0.064 0.085
0.000 0.020 0.031 0.040 0.050 0.068 0.075 0.089 0.105
0.000 0.004 0.017 0.029 0.041 0.057 0.065 0.079 0.114
0.002 0.015 0.026 0.035 0.043 0.057 0.065 0.082 0.100
RURAL FOREST
060430003
360310002
471550101
511130003
Yosemite NP, CA
Whiteface Mtn., NY
Smoky Mtns. NP, TN
Shenandoah NP
0.013 0.038 0.046 0.052 0.058 0.069 0.074 0.084 0.114
0.000 0.024 0.035 0.043 0.051 0.063 0.070 0.084 0.113
0.020 0.039 0.050 0.057 0.064 0.075 0.080 0.089 0.106
0.010 0.035 0.045 0.052 0.059 0.070 0.075 0.089 0.107
£ RURAL OTHER (I.E., RURAL RESIDENTIAL OR RURAL COMMERCIAL)
H
6
O
0
H
O
O
0
o
H
W
060710005
350431001
370810011
371470099
Crestline, CA
Sandoval Co., NM
Guilford Co., NC
Farmville, NC
0.000 0.024 0.043 0.055 0.068 0.093 0.107 0.135 0.170
0.000 0.004 0.020 0.033 0.045 0.056 0.061 0.070 0.091
0.000 0.003 0.015 0.029 0.043 0.063 0.071 0.087 0.120
— 0.000 0.008 0.021 0.033 0.044 0.061 0.068 0.078
Hourly Mean
O3 Occurrences
No. ol SUM06
Hours iO.08 aO.10 (ppm-h)
5076 21 0 24.7
4367 37 0 34.2
5059 142 34 43.5
5116 1 0 8.6
5080 156 6 63.2
5055 48 1 27.4
4345 59 1 24.8
4619 106 6 86.0
4536 77 4 44.7
5074 267 5 149.2
4813 157 4 98.5
4922 933 369 174.8
5088 9 0 21.4
4879 111 10 45.4
0.097 4882 34 0
W126
(ppm-h)
20.7
25.8
35.7
10.6
50.0
22.7
22.8
65.6
37.7
106.5
73.6
144.9
20.3
334.4
36.6
-------
3
to
o
o
x iiuic J-.J. ociisuiiiii ^-i|ji 11 — v^v. luuci ; i ci v.ciiinc jsiau luuiiun ui i iuiii ly v^iuiic v uiiv.cnu iiiiuiis ^|j|jin ;< n uniuci ui i lum i y
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 > 75%
Hourly Mean
Percentiles
OJ
to
o
f?
H
6
o
0
H
O
o
s
0
o
H
W
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
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.
0.060
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
O3 Occurrences
iO.08
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2:0.10
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
-------
3
to
o
o
i iiuic .J-.J ^v_uni u;. ociisuniii ^"vjji ii — v^iiuuci ; i civ.cninc LMSU luuiiun ui iiuiiiiy v^iuiic v unv.cnii iiiiuiis ^FF*
Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126
Experiencing Low Maximum Hourly Average Concentrations with Data Capture >75%
n;, n iiinuci ui
Values for Sites
Hourly Mean
Percentiles
OJ
to
o
H
6
o
0
H
O
o
w
0
o
HH
H
W
Site Year
Glacier NP 1989
300298001
(Montana) 1990
963m
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Yellowstone NP 1988
(Wyoming)
560391010 1989
2484m
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
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
O3 Occurrences
iO.08
0
0
0
0
0
0
0
0
0
0
0
0
0
17
0
0
0
0
0
0
0
2:0.10
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
1.0
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
-------
3
to
o
o
i iiuic .J-.J ^v_uni u;. ociisuniii ^"vjji ii — v^iiuuci ; i civ.cninc LMSU luuiiun ui iiuiiiiy v^iuiic v uiiiciiii iiiiuiis ^FF*
Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126
Experiencing Low Maximum Hourly Average Concentrations with Data Capture >75%
n;, n iiinuci ui
Values for Sites
Hourly Mean
Percentiles
LtJ
to
O
f?
H
1
o
o
0
H
O
o
^.J
w
o
o
1 — I
H
W
Site Year
Yellowstone NP 1997
(Wyoming)
560391011 1998
2468m
1999
2000
2001
DenaliNP 1988
(Alaska)
022900003 1990
640m
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Badlands NP 1989
460711001
(South Dakota) 1990
730m
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
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
O3 Occurrences
iO.08
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2:0.10
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
-------
3
to
o
o
i iiuic .J-.J ^v_uni u;. ociisuniii ^"vjji ii — v^iiuuci ; i civ.cninc LMSU luuiiun ui iiuiiiiy v^iuiic v unv.cnii iiiiuiis ^FF*
Hourly Mean Ozone Occurrences >0.08 and >0.10, Seasonal 7-h Average Concentrations, SUM06, and W126
Experiencing Low Maximum Hourly Average Concentrations with Data Capture >75%
n;, n iiinuci ui
Values for Sites
Hourly Mean
Percentiles
OJ
to
ON
O
ITJ
H
6
o
0
— ]
o
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
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
O3 Occurrences
iO.08
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
2:0.10
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
O
HH
H
W
-------
1 concentrations > 0.08 ppm. Ozone hourly average concentrations rarely exceed 0.08 ppm at
2 remote monitoring sites in the western United States. In almost all cases for the above sites, the
3 maximum hourly average concentration was < 0.075 ppm.
4 The highest hourly average concentrations at several relatively remote monitoring sites do
5 not necessarily occur during the summer months. Similarly, the highest 8-h daily maximum
6 concentrations do not necessarily all occur during the summer. For example, at the Yellowstone
7 National Park site, the first three 8-h daily maximum concentrations occurred in April and May
8 in 1998, and the fourth highest 8-h daily maximum concentration did not occur until July of that
9 year. In 1999, the first three highest 8-h daily maximum concentrations were observed in March
10 and May, and the fourth highest value occurred in April. In 2000, the four highest values
11 occurred in May, June, July, and August.
12 It is questionable whether the distributions experienced at sites exhibiting low maximum
13 hourly average concentrations in the western United States are representative of sites in the
14 eastern and midwestern United States because of regional differences in sources of precursors
15 and transport patterns. Given the high density of sources in the eastern and midwestern United
16 States, it is unclear whether a site could be found in either of these regions that would not be
17 influenced by the transport of O3 from urban areas. Thus, with the exception of the Voyageurs
18 National Park site in Minnesota, observations at relatively remote monitoring sites are limited to
19 those obtained in the western United States.
20 Based on data collected in the 1970s to the mid-1990s, consistent trends in tropospheric O3
21 have not been observed at the cleanest sites in the Northern Hemisphere. When assessing
22 possible trends at clean sites, specific attention must be given to the possibility of long-range
23 transport from urban centers. For example, the trends at Whiteface Mountain could provide
24 evidence that background O3 concentrations have risen by a few ppb in surface air over the
25 United States during the past two decades. However, these trends are probably associated with
26 the transport of regional pollution to the site. The site experiences elevated levels of O3 during
27 the summer months, and the area is currently designated as in nonattainment for the 1-h
28 standard.
29 Using (1) a 15-year record of O3 from Lassen Volcanic National Park, (2) data from
30 two aircraft campaigns, and (3) observations spanning 18 years from five United States west
31 coast marine boundary layer sites, Jaffe et al. (2003) have estimated that the amount of O3 in air
January 2005 3-27 DRAFT-DO NOT QUOTE OR CITE
-------
1 arriving from the Eastern Pacific in spring has increased by approximately 10 ppb from the
2 mid-1980s to the present. This positive trend might be due to increases of emissions of O3
3 precursors in Asia. Positive trends in O3 were found during all seasons. Although the Lassen
4 Volcanic National Park site is not close to any major emission sources or urban centers, it
5 experiences maximum hourly average O3 concentrations > 0.080 ppm during April-May and
6 > 0.100 ppm during the summertime, suggesting local photochemical production, at least during
7 the summertime. Elevated levels of O3 occur in the areas to the west (Redding, CA) and to the
8 south (Chico, CA; Tuscan, CA; Red Bluff, CA) of the park site. Hourly average concentration
9 levels occur in the 0.080 to 0.099 ppm range during April and May, and they can be above
10 0.100 ppm during the summertime at these locations.
11 Since emissions of O3 precursors have decreased in California, one of the reasons
12 suggested for the springtime increases in O3 at Lassen is transport from Asia. However,
13 although emissions of O3 precursors may have decreased in California as a whole over the
14 monitoring period, there still may be regional increases in areas that could affect air quality in
15 Lassen Volcanic National Park. Thus, at this time, although there is evidence as reported in the
16 literature that some rural locations may be experiencing increasing levels of O3, there is also
17 evidence that O3 concentrations at relatively remote monitoring sites, as reported in the chapter,
18 have not increased over the period of record.
19
20
21 3.3 DIURNAL VARIABILITY OF OZONE
22 Diurnal patterns of O3 may be expected to vary with location, depending on the balance
23 among the many factors affecting O3 formation, transport, and destruction. Although they vary
24 with locality, diurnal patterns for O3 typically show a rise in concentration from low (or levels
25 near minimum detectable amounts) to an early afternoon peak at lower elevation monitoring
26 sites. The 1978 O3 AQCD (U.S. Environmental Protection Agency, 1978) ascribed the diurnal
27 pattern of concentrations to three simultaneous processes: (1) downward transport of O3 from
28 layers aloft, (2) destruction of O3 through contact with surfaces and through reaction with nitric
29 oxide (NO) at ground level, and (3) in situ photochemical production and destruction of O3.
30 The form of the average diurnal pattern may provide information on sources, transport, and
31 chemical formation and destruction effects at various sites. Atmospheric conditions leading to
January 2005 3-28 DRAFT-DO NOT QUOTE OR CITE
-------
1 limited transport from source regions will produce early afternoon peaks. However, long-range
2 transport processes shift the occurrence of a peak from afternoon to evening or early morning
3 hours.
4 The U.S. Environmental Protection Agency (1996) discussed diurnal patterns for urban
5 sites. An example was provided for a Philadelphia site for July 13, 1979. On this day, a peak
6 1-h average concentration of 0.20 ppm, the highest for the month, was reached at 1400 hours,
7 presumably as the result of meteorological factors, such as atmospheric mixing and local
8 photochemical processes. The severe depression of concentrations to below detection limits
9 (less than 0.005 ppm) between 0300 and 0600 hours usually is explained as resulting from the
10 scavenging of O3 by local NO emissions. In this regard, this site is typical of most urban sites.
11 Diurnal profiles of O3 can vary from day to day at a specific site because of changes in the
12 various factors that influence concentrations. Composite diurnal data (i.e., concentrations for
13 each hour of the day averaged over multiple days or months) often differ markedly from the
14 diurnal cycle shown by concentrations for a specific day. For urban sites, maximum diurnal
15 variability occurs in the third quarter.
16 Nonurban O3 monitoring sites experience differing types of diurnal patterns depending on
17 their location relative to urban source regions. Thus, considerable variations in O3
18 concentrations are found among sites characterized as agricultural or forested. Nonurban areas
19 are more likely to show flatter diurnal patterns than sites located in urban areas. Such patterns
20 are based on average concentrations calculated over an extended period and, as indicated earlier,
21 caution is urged in drawing conclusions concerning whether some monitoring sites experience
22 higher cumulative O3 exposures than other sites. On a daily basis, variations in O3
23 concentrations occur from hour to hour, and, in some cases, elevated hourly average
24 concentrations are experienced either during the daytime or nighttime. Because the diurnal
25 patterns represent averaged concentrations calculated over an extended period, the smoothing
26 caused by the averaging tends to mask the elevated hourly average concentrations.
27 The current forms of the O3 standards focus on both the highest hourly average
28 concentrations (1-h standard) and, for many monitoring sites, the mid-level hourly average
29 concentrations (8-h standard). Examples provided in this chapter show that urban monitoring
30 sites show a tendency for the highest maximum hourly average concentration to occur from June
31 to August (Figure 3-3). Due to meteorological and other factors, the absolute magnitude of the
January 2005 3-29 DRAFT-DO NOT QUOTE OR CITE
-------
Phoenix, AZ
2002
040130019
Denver, CO
2002
080310014
0.16
0.14
Tampa, FL
2002
1205T1065
0.16 -
0.14
0.12
0.10
0.08
0.06 -
0.04 -
0.02 -
0.00
B S
i -B
£i
§
1
S1
|II|I
ml:
_«
•
•
1 t
H^L
ZM:
1^
•
•
1 I
= I
1
o
S
I:
• :
• :
• :
1 1
1 1
Washington, D.C.
2002
110010043
0.20
. 0.18
0.18
•• 0.14
0.12
0.10
0.08
0.06
0.04
0.02
Boston, MA
2002
250250041
m
..mini
n
Los Angeles, CA
2002
060376012
•> 5
j^ o
E S
B) (j
a. o
to
<> «
a a
e E
I S,
Figure 3-3a-h. Seasonal variations in O3 concentrations as indicated by the 1-h maximum
in each month at selected sites in 2002.
Source: U.S. Environmental Protection Agency (2003a).
January 2005
3-30
DRAFT-DO NOT QUOTE OR CITE
-------
1 maximum hourly average concentrations varies from year to year. For nonurban areas, it is not
2 possible to predict the quarter in which the highest hourly average concentrations occur. There
3 is a tendency for relatively remote monitoring sites to experience their highest average O3
4 concentrations during the second quarter (i.e., the months of April or May) versus the third
5 quarter of the year ad for urban sites.
6
7
8 3.4 SPATIAL VARIABILITY OF OZONE
9 The spatial variability in O3 concentrations in nine MSAs across the United States was
10 examined. These MSAs were selected to provide (1) information helpful for risk assessments,
11 (2) a general overview of the spatial variability of O3 in different regions of the country, and
12 (3) insight in to the spatial distribution of O3 in cities where health outcome studies have been
13 conducted. Statistical analyses of the human health effects of airborne pollutants based on
14 aggregate population time-series data have often relied on ambient concentrations of pollutants
15 measured at one or more central sites in a given metropolitan area. In the particular case of
16 ground-level O3 pollution, central-site monitoring has been justified as a regional measure of
17 exposure partly on the grounds that correlations between concentrations at neighboring sites
18 measured over time are usually high. In analyses where multiple monitoring sites monitor
19 ambient O3 concentrations, a summary measure, such as an average, has often been regarded as
20 adequately characterizing the exposure distribution.
21 For quantifying spatial variability, there were no clearly discernible regional differences
22 in the ranges of parameters analyzed (Table 3-6). Additional urban areas would need to be
23 examined to discern broadscale patterns. The data indicate considerable variability in the
24 concentrations fields. Seasonal means vary within individual urban areas by factors of 1.4 to 4.
25 Intersite correlation coefficients show mixed patterns (i.e., in some urban areas all pairs of sites
26 are moderately to highly correlated and in other areas there is a larger range of correlations).
27 As may be expected, those areas showing a smaller range of seasonal mean concentrations also
28 show a smaller range of intersite correlation coefficients. However, there are a number of cases
29 where sites in an urban area may be moderately to highly correlated, but show substantial
30 differences in absolute concentrations. In many cases, values for the P90 can equal or exceed
31 seasonal mean O3 concentrations.
January 2005 3-31 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-6. Summary Statistics for O3 (in ppm) Spatial Variability in Selected Urban Areas in the United States
to
OJ
1
to
o
!>
H
6
o
0
H
O
o
0
HH
H
W
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
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
P a
^90
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
p
^90
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
CODb
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
P90 = 90th percentile absolute difference in concentrations.
COD = coefficient of divergence for different site pairs.
-------
1 3.5 TRENDS IN OZONE CONCENTRATIONS
2 Ozone concentrations and, thus, O3 exposures, change from year to year. From 1983 to
3 2002, extremely high O3 levels occurred in 1983 and 1988 in some areas of the United States
4 (Figures 3-4 and 3-5). These levels were likely due, in part, to hot and dry, stagnant weather
5 conditions. The annual second highest 1-h average O3 concentrations were the lowest in 2000
6 (U.S. Environmental Protection Agency, 2003b). These low levels may have been due to
7 meteorological conditions that were less favorable for O3 formation and to recently implemented
8 control measures. For the period 1983 to 2002, 1-h and 8-h O3 concentrations decreased in most
9 areas of the country. On a nationwide basis, a 22% decrease was observed in the second highest
10 1-h average concentration and a 14% decrease in the annual fourth highest annual daily
11 maximum concentration was observed from 1983 to 2002. The Northeast and Pacific Southwest
12 exhibited the most substantial improvement for 1-h and 8-h O3 levels. The Mid-Atlantic and
13 North Central regions experienced minimal decreases in 8-h O3 levels. In contrast, the Pacific
14 Northwest region showed a slight increase in 8-h O3 concentrations.
15 From 1993 to 2002, the nationwide trend in 8-h O3 concentration shows a 4% increase, and
16 the nationwide trend in 1-h O3 shows a 2% decrease. However, these trends were not
17 statistically significant. Ozone concentrations varied slightly during this 10-year period but,
18 overall, showed no significant change.
19 Regional trends provide additional information that can increase our understanding of the
20 patterns of change of O3 levels (Section AX3). The trend in 8-h O3 concentrations for the Pacific
21 Southwest shows a 29% decrease from 1983 to 2002. When considering the Los Angeles area
22 separately, the trend for Los Angeles shows a 49% decrease for the same 20-year period, with a
23 15% decrease for other locations in the Pacific Southwest. For 1993 to 2002, the Pacific
24 Southwest experienced an overall 13% decrease in 8-h O3. However, when considering
25 Los Angeles separately, the Los Angeles area shows a 28% decrease for the 10-year period,
26 while the Pacific Southwest without Los Angeles has a 5% decrease.
27 It is important to note that year-to-year changes in ambient O3 trends are influenced by
28 meteorological conditions, population growth, and changes in emission levels of O3 precursors
29 (i.e., volatile organic compounds [VOCs] and NOX) resulting from ongoing control measures.
30 On a nationwide basis, the trends analyses provide an indication that a slowing down of
31 improvement in O3 air quality has occurred. Information provided in the literature indicates that
January 2005 3-33 DRAFT-DO NOT QUOTE OR CITE
-------
Q.
c"
o
TV
+*
o
o
c
0.20
0.15
0.10
0.05
0.00
90% of sites have concentrations below this line
370 Sites
10% of sites have concentrations below this line
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02
1983-2002: 22% decrease
1993-2002: 2% decrease
Figure 3-4. Trends for 1983 - 2002 period for the annual second maximum 1-h
average concentration.
Source: U.S. Environmental Protection Agency (2003b).
Q.
Q.
0.20
0.15
0.10
0)
o
g 0.05
O
0.00
370 Sites
90% of sites have concentrations below this line
10% of sites have concentrations below this line
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02
1983-2002: 14% decrease
1993-2002: 4% increase
Figure 3-5. Trends for 1983 - 2002 period for the annual fourth maximum 8-h
average concentration.
Source: U.S. Environmental Protection Agency (2003b).
January 2005
3-34 DRAFT-DO NOT QUOTE OR CITE
-------
1 the higher concentrations may be reduced at a faster rate than the lower values. Using models,
2 several investigators have commented on the difficulty in reducing the mid-level hourly average
3 concentrations while reducing the fourth highest 8-h average daily maximum concentration.
4 Using ambient O3 concentrations in conjunction with photochemical models to determine the
5 technical feasibility of reducing hourly average concentrations, some investigators have reported
6 that various combinations of VOCs and NOX emission reductions were effective in lowering
7 modeled peak 1-h O3 concentrations. However, VOC emissions reductions were found to have
8 only a modest impact on modeled peak 8-h O3 concentrations. It has been reported that 70 to
9 90% NOX emissions reductions were required to reduce peak 8-h O3 concentrations to the desired
10 level. Some investigators have noted that when anthropogenic VOC and NOX emissions are
11 reduced significantly, the primary sources of O3 precursors are biogenic emissions and CO from
12 anthropogenic sources. Chemical process analysis results indicated that slowly reacting
13 pollutants such as CO could be contributing from 10 to 20% of the O3 produced. Several
14 investigators have proposed that the effect of Asian emissions on surface O3 concentrations
15 might be responsible for explaining the inability of the emissions reductions to effect a rapid
16 change in the mid-level concentrations. However, as described in the chapter, inconsistent
17 trends results do not indicate that either the low or the high end of the hourly average
18 concentration distributions are increasing and, thus, the effect of Asian emissions on surface O3
19 concentrations in the western United States is not clear at this time.
20 U.S. Environmental Protection Agency (2003b) reported that 28 of the national parks in the
21 United States had sufficient O3 data for calculating trends for the 10-year period 1993 to 2002.
22 Seven monitoring sites in five of these parks experienced statistically significant upward trends
23 in 8-h O3 levels: Great Smoky Mountains (Tennessee), Craters of the Moon (Idaho), Mesa
24 Verde (Colorado), Denali (Alaska), and Acadia (Maine). Monitoring data for one park, Saguaro
25 (Arizona), showed statistically significant improvements over the same time period. For the
26 remaining 22 parks with O3 trends data, the 8-h O3 levels at 13 parks increased only slightly
27 between 1993 and 2002, while five parks showed decreasing levels and four showed levels
28 unchanging. Using all available data for the national parks, additional analyses of the hourly
29 average concentration data were performed to characterize trends using the (a) seasonal W126
30 and SUM06 cumulative exposure indices, (b) fourth highest of the 1-h daily maximum average
January 2005 3-35 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations over a 3-year period for January through December, and (c) 3-year average of the
2 fourth highest daily maximum 8-h average concentration.
3 Table 3-7 summarizes the results of the analysis. Caution is urged in interpreting the table.
4 The use of the significance levels 0.2, 0.1, and 0.05 indicates varying levels of uncertainty.
5 As indicated above, the use of a specific concentration index (e.g., 8-h average concentration)
6 will provide a different history than the use of an alternative index, such as the W126 or SUM06
7 cumulative exposure index. Both the 1-h and 8-h indices are extreme value metrics that focus on
8 the highest levels in the distribution of the hourly average concentrations. On the other hand, the
9 W126 and SUM06 exposure indices accumulate the hourly average concentrations mainly in the
10 upper half of the distribution. Changes in the magnitude of the extreme value statistics may or
11 may not result in a significant change in the entire percentile distribution of the hourly average
12 concentrations.
13
14
15 3.6 RELATIONSHIPS BETWEEN OZONE AND OTHER SPECIES
16 Correlations between Ozone and other Species
17 In order to understand relationships among atmospheric species, an important distinction
18 must be made between primary (directly emitted) species and secondary (photochemically
19 produced) species. In general, it is more likely that primary species will be more highly
20 correlated with each other, and that secondary species will be more highly correlated with each
21 other. By contrast, primary species are less likely to be correlated with secondary species.
22 Secondary reaction products tend to correlate with each other, but there is considerable variation.
23 Some species (e.g., O3 and organic nitrates) are closely related photochemically and correlate
24 with each other strongly. Others (e.g., O3 and H2O2) show a more complex correlation pattern.
25 Further details are given in Annex AX3 in Section AX3.7.
26 Relationships between primary and secondary components are illustrated by considering
27 data for O3 and PM2 5. Ozone and PM2 5 concentrations observed at a monitoring site in Fort
28 Meade, MD are plotted as conditional means in Figure 3-6, based on data collected between July
29 1999 and July 2001. As can be seen from the figure, PM2 5 tends to be negatively correlated with
30 O3 to the left of the inflection point (at about 30 ppbv O3) and tends to be positively correlated
31 with O3 to the right of the inflection point. Data to the left of the minimum in PM2 5 were
January 2005 3-36 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-7. Trends at National Parks in the United States (1981 - 2001 or available data period)
1
to
o
o
OJ
oo
O
s
H
6
0
o
H
0
0
H
W
O
O
HH
H
W
Site
Acadia NP
Acadia NP AC AD
Acadia NP Cadillac Mountain
Acadia NP MARS/PRIMENet Site
Big Bend NP BIBE
Brigantine (FWS) BRIG
Canyonlands NP CANY
Cape Cod NS CACO
Cape Romain (FWS) CARD
ChamizalNMEM CHAM
Channel Islands NP CHIS
ChiricahuaNM (NDDN CNM167)
Congaree Swamp NM COSW
CowpensNB COWP
Craters of the Moon NM CRMO
Death Valley NP DEVA
DenaliNP DENA
Everglades NP EVER
Glacier NP (NDDN GNP168) GLAC
Great Basin NP GRBA
AIRS ID
230090003
230090101
230090102
230090103
480430101
340010005
490370101
250010002
450190046
481410044
060832012
040038001
450791006
450210002
160230101
060270101
022900003
120250030
300298001
320330101
W126
(Seasonal)
NS
NS
NS
NS
***(_\
***(+)
NS
NS
***(+)
NS
NS
NS
*(+)
*(+)
NS
NS
NS
NS
**(+)
SUM06
(Seasonal)
NS
NS
NS
NS
***(_\
***(+)
NS
NS
**(+)
NS
*(+)
NS
*(+)
NS
NS
NS
NS
NS
*(+)
8-h (Annual)
NS
*(-)
*(+)
NS
NS
***(_\
*(+)
NS
NS
NS
NS
NS
*(-)
NS
***(+)
NS
*(+)
NS
NS
NS
1-h (Annual)
NS
***(_)
NS
**(.)
NS
***(_)
**(+)
***(_)
NS
NS
***(_)
**(.)
*(-)
*(+)
NS
NS
NS
*(-)
NS
NS
-------
Table 3-7 (cont'd). Trends at National Parks in the United States (1981 - 2001 or available data period)
1
to
o
o
OJ
oo
oo
o
s
H
6
0
o
H
0
0
H
W
O
O
HH
H
W
Site
Grand Canyon NP (NDDN GCN174)
Great Smoky Mountains NP (Cades Cove) GSCC
Great Smoky Mountains NP (Clingmans Dome) GSCD
Great Smoky Mountains NP (Cove Mountain) GSCM
Great Smoky Mountains NP (Look Rock) GSLR
Hawaii Volcanoes NP HAVO
Joshua Tree NP
Lassen Volcanic NP LAVO
Mammoth Cave NP MACA
Mammoth Cave NP MACA
Mesa Verde NP MEVE
Mount Rainier NP MORA
North Cascade NP NOCA
Olympic NP OLYM
Petrified Forest NP PEFO
Pinnacles NM FINN
Rocky Mountain NP ROMO
SaguaroNM SAGU
Sequoia/Kings Canyon NPs (Ash Mountain) SEAM
Sequoia/Kings Canyon NPs (Lookout Point) SELF
AIRS ID
040058001
470090102
471550102
471550101
470090101
150010005
060719002
060893003
210610500
210610501
080830101
530531010
530570013
530090012
040010012
060690003
080690007
040190021
061070005
061070008
W126
(Seasonal)
NS
NS
**(+)
***(+)
***(_\
**(+)
NS
NS
**(+)
NS
NS
NS
NS
*(-)
***(+)
NS
***(-)
SUM06
(Seasonal)
*(+)
NS
**(+)
**(+)
*(-)
NS
NS
NS
*(+)
NS
NS
NS
NS
**(.)
***(+)
NS
***(-)
8-h (Annual)
NS
NS
NS
***(+)
NS
NS
***(_\
NS
**(_)
*(-)
NS
NS
NS
NS
NS
**(.)
NS
NS
NS
NS
1-h (Annual)
NS
NS
***(+)
***(+)
NS
***(_\
NS
NS
**(.)
***(+)
NS
**(+)
*(-)
***(+)
*(+)
**(.)
NS
NS
-------
Table 3-7 (cont'd). Trends at National Parks in the United States (1981 - 2001 or available data period)
£
^
^•H
to
o
o
OJ
oo
VO
o
Site AIRS ID
Sequoia/Kings Canyon NPs (Lower Kaweah) SELK 061070006
Shenandoah NP (Big Meadows) SVBM 511130003
VoyageursNP VOYA 271370034
Yellowstone NP YELL 560391010
Yellowstone NP YELL 560391011
Yosemite NP (Camp Mather) YOCM 061090004
Yosemite NP (Turtleback Dome) YOTD 060430003
Yosemite NP (Wawona Valley) YOWV 060430004
* 0.20 level of significance
* * 0.10 level of significance
*** 0.05 level of significance
NS Not significant
W126 (seasonal) = cumulative W126 exposure (as described by Lefohn and
SUM06 (seasonal) = cumulative exposure of hourly average concentrations
W126
(Seasonal)
NS
NS
**(.)
NS
NS
NS
*(-)
SUM06
(Seasonal)
NS
NS
**(.)
NS
NS
NS
*(-)
Runeckles [1987]) over a 24-h period for the
> 0.06 ppm over a 24-h period for the April -
8-h (annual) = 3 -year average of the fourth highest 8-h daily maximum average concentration for January
C 1-h (annual) = fourth highest 1-h daily maximum average concentration over a 3 -year period for January
H
6
0
o
H
0
0
H
W
O
O
HH
H
W
- December.
- December.
8-h (Annual) 1-h (Annual)
NS NS
JN o \ )
*(_) ***(_)
NS NS
NS **(+)
*(+) ***(0)
***(_} ***(_•)
***(_) **(_)
period April - October.
October period.
-------
40
E 35 --
o>
e
.2
c
0)
o
o
o
Q.
30 --
25 --
20 --
15 --
§ 10
C
O
O
4-
10 20 30 40 50
Conditional Mean 03 Concentration (ppbv)
60
70
Figure 3-6. Conditional mean PM2 5 concentrations versus O3 concentrations observed at
Fort Meade, MD from July 1999 to July 2001.
Source: Chen (2002).
1
2
3
4
5
9
10
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, and it is titrated with NO in
the polluted boundary layer. 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.
January 2005
3-40
DRAFT-DO NOT QUOTE OR CITE
-------
1 Co-Occurrence of Ozone with other Pollutants
2 The characterization of co-occurrence patterns under ambient conditions is important for
3 relating human health and vegetation effects under ambient conditions to controlled research
4 results as described in Annex AX3.8. Several attempts have been made to characterize gaseous
5 air pollutant mixtures. The previous 1996 O3 AQCD discussed various patterns of pollutant
6 mixtures of SO2, NO2, and O3. Pollutant combinations can occur at or above a threshold
7 concentration at either the same or different times.
8 The 1996 O3 AQCD noted that studies of the joint occurrence of gaseous NO2/O3 and
9 SO2/O3 reached two conclusions: (1) hourly simultaneous and daily simultaneous-only
10 co-occurrences are fairly rare (when both pollutants were present at an hourly average
11 concentration > 0.05 ppm) and (2) when co-occurrences are present, complex-sequential and
12 sequential-only co-occurrence patterns predominate. Year-to-year variability was found to be
13 insignificant.
14 Using 2001 hourly data for O3 and NO2, the co-occurrence patterns for the data show
15 similar results from previous studies; most of the collocated monitoring sites experienced fewer
16 than 10 co-occurrences as shown in Figure 3-7. Using 2001 data, co-occurrence analyses for O3
17 and SO2 showed results similar to those published previously - that most of the collocated
18 monitoring sites analyzed experienced fewer than 10 co-occurrences, as shown in Figure 3-8.
19 Since 1999, monitoring stations across the United States have routinely measured 24-h
20 average concentrations for PM2 5. Daily co-occurrence of PM2 5 and O3 over a 24-h period was
21 also characterized. Because PM25 data are mostly summarized as 24-h average concentrations in
22 the AQS database, a daily co-occurrence of O3 and PM25 was subjectively defined as an hourly
23 average O3 concentration > 0.05 ppm and a PM2 5 24-h concentration > 40 |ig/m3 (corresponding
24 to the EPA's Air Quality Index, Level of Concern for PM2 5) occurring during the same 24-h
25 period. Using 2001 data from the AQS database, the daily co-occurrence of PM25 and O3 was
26 infrequent (Figure 3-9). There are limited data available on the co-occurrence of O3 and other
27 pollutants (e.g., acid precipitation and acidic cloudwater). In most cases, routine monitoring data
28 are not available from which to draw general conclusions.
29
January 2005 3-41 DRAFT-DO NOT QUOTE OR CITE
-------
40
35
v> 30
W 25
! 20
v
z 10
§
0
I
T"
«
o
T*
«
{SI
O
CM
O)
00
o
oo
§
o
s
o
as
o
1C
o
IS.
O
IS.
aj
CO
o
CO
CJ>
a»
o
9 0s
O 1-
Q
a>
•
o
CD
00
o
O5
O
LO
o
CO
Number of Co-Oceurrences (Hours)
Figure 3-7. The co-occurrence pattern for O3 and nitrogen dioxide using 2001 data from
the AQS. There is co-occurrence when hourly average concentrations of O3
and another pollutant are both ^ 0.05 ppm.
uu -
50-
(0
0)
^ 4°-
2 30-
0)
| 20-
•z.
10-
n
I
1 1
CD CD
1 T—
0
u
(T>
CM
0
r\i
CD CD
co •sr
0 0
CO -f
CD
m
0
CD CD
CD h~
0 0
CD h-
CD
00
0
00
CD
CD
0
CD
CD CD
0 ^
i i
O O
0 -,-
CD
CM
i
O
CM
CD
CO
i
O
CO
CD CD
-3- m
1 1
o o
"3- m
Number of Co-occurrences (Hours)
Figure 3-8. The co-occurrence pattern for O3 and sulfur dioxide using 2001 data
from AQS. There is co-occurrence when hourly average concentrations
of O3 and another pollutant are both > 0.05 ppm.
January 2005
3-42
DRAFT-DO NOT QUOTE OR CITE
-------
140-
2 120-
W 100-
*-
2 80-
| 60-
z 40-
20-
n.
CD
CD CD CD CD
i- CM CO -sT
CD CD
LO CD
CD
CD CD CD rn
00 CD O ^
CD CD
CM CO
000000000
•s-CMCO^rLOCDI^-OOCD
CD CD
^r LO
O O O O
CM CO ••vT LO
Number of Co-occurrences (Hours)
Figure 3-9. The co-occurrence pattern for O3 and PM2 5 using 2001 data from AQS.
1 3.7 POLICY RELEVANT BACKGROUND OZONE CONCENTRATIONS
2 Background O3 concentrations used for NAAQS-setting purposes are referred to as Policy
3 Relevant Background (PRB) O3 concentrations. They are used by the EPA to assess risks to
4 human and ecosystem health and to provide risk estimates associated with O3 produced by
5 anthropogenic sources in North America (defined here as the continental North America).
6 Policy Relevant Background concentrations are those concentrations that would result in the
7 United States in the absence of anthropogenic emissions in North America. Policy Relevant
8 Background concentrations include contributions from natural sources everywhere in the world
9 and from anthropogenic sources outside North America. Issues concerning the methodology for
10 calculating PRB O3 concentrations are described in detail in Annex AX3, Section AX3.9.
11 Contributions to PRB O3 include photochemical interactions involving natural emissions of
12 VOCs, NOX, and CO; the long-range transport of O3 and its precursors from outside North
13 America; and stratospheric-tropospheric exchange (STE). Processes involved in STE are
14 described in detail in Annex AX2.3. Natural sources of O3 precursors include biogenic
15 emissions, wildfires, and lightning. Biogenic emissions from agricultural activities are not
16 considered in the formation of PRB O3.
January 2005
3-43
DRAFT-DO NOT QUOTE OR CITE
-------
1 Springtime maxima are observed at relatively clean (Annex AX3 and Figures 3-10a,b)
2 national parks mainly in the western United States and at a number of other relatively unpolluted
3 monitoring sites throughout the Northern Hemisphere. The major issues concerning the
4 calculation of PRB O3 center on the capability of the current generation of global-scale, three-
5 dimensional, CTMs to simulate the causes of high O3 concentrations observed at monitoring
6 sites in relatively unpolluted areas of the United States from late winter through spring (i.e.,
7 February through June). The issues raised do not affect interpretations of the causes of
8 summertime O3 episodes as strongly. Summertime O3 episodes are mainly associated with slow-
9 moving high-pressure systems characterized by limited mixing between the planetary boundary
10 layer and the free troposphere, as noted in Annex AX2, Section AX2.3.
11
Yellowstone National Park
Maximum Hourly Concentration
1998-2001
0 080 -
"E
§ 0.050
IS
a*
o
0.010
~
',
;
1
>l
:
:
i
s
~
_
'
j
~
X
:
3
5
i
^
',
1
J
i
;
j-i
-
£
«a
_
L
c
r"
i
i
-
H
i
-
9
!
i
i
i
_
H
3
)
_
|_
i
<
«,
**
3
3)
3
C
i.
ȣ
1
•
!
_
i
j
r-i
b
1
£
t
c
t
!
1
-at
1
J
;
2
_
(
!
'
~
I
j:
1
a
t
C
P
I
i
>
a
2
-
11998
Month
D2000 D2001
560391011
Figure 3-10a. Monthly maximum hourly average O3 concentrations at Yellowstone
National Park, Wyoming, in 1998,1999, 2000, and 2001.
Source: U.S. Environmental Protection Agency (2003a).
1 A large number of case studies document the occurrence of STE in mid- and high latitudes
2 in Europe, Asia, and North America mainly during winter and spring. These studies were based
3 on aircraft, satellite, and ground-based measurements. Considerable uncertainty exists in the
January 2005
3-44
DRAFT-DO NOT QUOTE OR CITE
-------
ooooooooooooooooooooooooooo
Time
Figure 3-10b. Hourly average O3 concentrations at Yellowstone National Park, Wyoming
for the period January to December 2001.
Source: U.S. Environmental Protection Agency (2003a).
1 magnitude of the exchange; however, these studies have found that STE occurs year long, but
2 with a distinct preference for the transport of O3 directly to the middle and lower troposphere
3 during late winter and spring. Transport to the upper troposphere basically occurs throughout
4 the year.
5 Springtime maxima in tropospheric O3 observed at high latitudes are also associated with
6 the winter buildup of O3 precursors and thermally labile reservoir species such as PAN and other
7 reactive nitrogen species. These pollutants originate from all continents in the Northern
8 Hemisphere. Ozone precursor concentrations reach a maximum in late March; and as sunlight
9 returns to the Arctic, photochemical reactions generate tropospheric O3 (Section AX3.9.1).
10 The contribution of Asian sources to the U.S. levels is also largest during spring, reflecting the
11 efficient lifting of Asian pollution ahead of cold fronts originating in Siberia and transport in
12 strong westerly winds across the Pacific. The longer lifetime of O3 during spring also
13 contributes to springtime maxima (Wang et al., 1998) (e.g., Hudman et al., 2004).
14 Estimates of PRB concentrations cannot be obtained solely by examining measurements of
15 O3 obtained at relatively remote monitoring sites (RRMS) in the United States (Annex AX3,
January 2005
3-45
DRAFT-DO NOT QUOTE OR CITE
-------
1 Section AX3.2.3) because of the long-range transport from anthropogenic source regions within
2 North America. It should also be noted that it is impossible to determine sources of O3 without
3 ancillary data that could be used as tracers of sources or to calculate photochemical production
4 and loss rates. The current definition of PRB implies that only CTMs can be used to estimate the
5 range of PRB values. On the synoptic and larger spatial scales at least, all evidence indicates
6 that global CTMs are adequate tools to investigate the factors controlling tropospheric O3; and
7 three-dimensional CTMs, as typified by Fiore et al. (2003) appear to offer the best methodology
8 for estimating PRB concentrations that can not be measured directly (Annex AX3, Section
9 AX3.9.2), at least for averaging periods of longer than one hour.
10 Previous estimates of background O3 concentrations, based on different concepts of
11 background, are given in Table 3-8. Results from global three-dimensional CTMs, where the
12 background is estimated by zeroing anthropogenic emissions in North America (Table 3-8) are
13 on the low end of the 25 to 45 ppbv range. Lefohn et al. (2001) have argued that frequent
14 occurrences of O3 concentrations above 50 to 60 ppbv at remote northern U.S. sites in spring are
15 mainly stratospheric in origin. Fiore et al. (2003a) used a global CTM to determine the origin of
16 the high-O3 events reported by Lefohn et al. (2001), and to conduct a more general quantitative
17 analysis of background O3 as a function of season, site location, and local O3 concentration.
18 Figure 3-11 shows a comparison between observations obtained at CASTNet sites and
19 model results of Fiore et al. (2003a). They classified the CASTNet monitoring sites into
20 low-lying surface sites (generally < 1.5 km) and elevated sites (> 1.5 km). All elevated sites are
21 in the West. Results were then aggregated to construct the cumulative probability distributions
22 shown in Figure 3-11, for the 58 surface sites and the 12 elevated sites, and for the three seasons.
23 The calculated mean background at the surface sites in spring is 27 ppbv as compared to 23 ppbv
24 in summer and fall. At these sites, the background is highest for O3 concentrations near the
25 center of the distribution, and decline as total surface O3 concentrations increase, for reasons
26 summarized above and discussed by Fiore et al. (2002a). At the elevated sites, the calculated
27 mean background is 36 ppbv in spring versus 30 ppbv in the other seasons. Background
28 concentrations in the fall resemble those in summer but show less variability and do not exceed
29 40 ppbv anywhere in this analysis.
30
January 2005 3-46 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-8. Previous Estimates of Background O3 in Surface Air Over the United States
Study
Method
Time Period Region
Background
Estimate (ppbv)
Trainer et al. y-intercept of O3 vs. NOy-NOx
(1993) regression linea
Hirsch et al. y-intercept of O3 vs. NOy-NOx
(1996) regression line
Altshuller y-intercept of O3 vs. NOy
and Lefohn regression line, and observations
(1996) at remote/rural sites
Liang et al. Sensitivity simulation in a 3-D
(1998) model with anthropogenic NOX
emissions in the continental U.S.
set to zero
Lin et al. Median O3 values for the lowest
(2000) 25th percentiles of CO and NOy
concentrations
Fiore et al. O3 produced outside of the North
(2002a) American boundary layer in a
global 3-D model
Summer 1988 Eastern United
States
May-Sep Harvard Forestb
1990-1994
Apr-Oct Continental
1988-1993 United States
Full year
Continental
United States
1990-1998 Harvard Forest
Summer 1995 Continental
United States
30-40
25 (Sept) - 40 (May)
25-45 (inland)0
25-35 (coastal)
20-30 (East)
20-40 (West)
(spring maximum)
3 5 (fall)-45 (spring)
15-30 (East)
25-35 (West)
aNOy is the chemical family including NOX and its oxidation products; NOy-NOx denotes the chemically
processed component of NOy.
b rural site in central Massachusetts.
0 seasonal 7-h daylight average.
Source: Fiore et al. (2003a).
1 Major conclusions from the Fiore et al. (2003a) study (which are discussed in detail in
2 Annex AX3, Sections AX3.9.3 and AX3.9.4) are:
3 • PRB O3 concentrations in U.S. surface air are generally 15 to 35 ppbv. They decline
from spring to summer and are generally < 25 ppbv under the conditions conducive to
high-O3 episodes.
4 • PRB O3 concentrations can be represented as a function of season, altitude, and total
surface O3 concentration, as illustrated in Figure 3-11.
5 • High-elevation sites (> 2 km) do occasionally experience high PRB concentrations in
spring (40 to 50 ppbv) due to free-tropospheric influence which includes a 4 to
12 ppbv contribution from hemispheric pollution (O3 produced from anthropogenic
emissions outside North America). These sites cannot be viewed as representative
of low-elevation surface sites (Cooper and Moody, 2000), where the background is
consistently lower.
January 2005
3-47
DRAFT-DO NOT QUOTE OR CITE
-------
120
100
1 80
1 60
-I 40
20
0
Elevated Sites (> 1.5 km)
>
-R
Q.
120
raioo
< 80
3 60
§7 4°
N C
O = 20
0
80
' 60
| 40
20
0
1 2.5 16 50 84 97.5 99
1 2.5 16 50 84 97.5 99
120rnTrTrrrrrn
100
120
100
80
60
40
20
0
120
100
80
60
40
20
0
120
100
80
60
40
20
0
Low-Lying Sites(< 1.5 km)
2.5 16 50 84 97.5 99
1 2.5 16 50 84 97.5 99
1 2.5 16 50 8497.599 1 2.5 16 50 8497.599
Cumulative Probability (%)
Figure 3-11. Estimates of background contribution to surface afternoon (13 to 17 LT)
O3 concentrations in the United States as a function of local O3
concentration, site altitude, and season. The figure shows cumulative
probability distributions of O3 concentrations for the observations
(asterisks) and the model (triangles). The corresponding distribution
of background O3 concentrations is shown as grey diamonds.
Source: Fiore et al. (2003a).
January 2005
3-48
DRAFT-DO NOT QUOTE OR CITE
-------
1 • The stratospheric contribution to surface O3 is of minor importance, typically well
< 20 ppbv. While stratospheric intrusions might occasionally elevate surface O3
at high-altitude sites, these events are rare.
2 Appropriate background concentrations should thus be allowed to vary as a function of
3 season, altitude, and total O3 level. The diamonds in Figure 3-11 can be applied for this purpose.
4 In particular, the depletion of the background during high-O3 events should be taken into account
5 (i.e., background O3 is depleted by reactions in the atmosphere and by deposition to the surface
6 but is not replenished at a significant rate in the stable, polluted boundary layer). This depletion
7 is shown mainly in the right-hand panels of Figure 3-11 for the highest O3 values. Note that the
8 model is generally able to reproduce the overall frequency distributions in Figure 3-11.
9 Underpredictions, especially at the upper end of the frequency distribution during the warmer
10 months, are likely related to sub-grid-scale processes that the model cannot resolve explicitly.
11 The highest observed O3 concentrations in all three seasons and at all altitudes are associated
12 with regional pollution (i.e., North American anthropogenic emissions), rather than stratospheric
13 influence.
14 Chemistry transport models should be evaluated with observations given earlier in
15 Chapter 3, in Annex AX3, and in Figure 3-12 (over appropriate averaging periods), and be able
16 to simulate the processes summarized in Chapter 2. Higher resolution models capable
17 of spatially and temporally resolving stratospheric intrusions and capable of resolving O3
18 variability on hourly timescales have not been applied to this problem. Ebel et al. (1991) have
19 demonstrated that regional scale CTMs could be used to study individual stratospheric
20 intrusions. As an example of the utility of different types of models, Zanis et al. (2003) were
21 able to forecast, observe, and model a stratospheric intrusion (maximum penetration depth was
22 to slightly > 2 km altitude) that occurred from June 20 to 21, 2001, over a large swath of central
23 Europe. Roelofs et al. (2003) compared results from six global tropospheric CTMs with lidar
24 observations obtained during that event and concluded that the models qualitatively captured the
25 features of this intrusion. It was also found that the coarser resolution models overestimated
26 transport to lower altitudes. The use of higher resolution models, perhaps nested inside the
27 coarser resolution models, may have helped solve this problem. They would also better address
28 issues related to temporal (i.e., 1-h versus 8-h averages) and spatial (i.e., populated versus
29 remote areas) scales needed by policymakers.
January 2005 3-49 DRAFT-DO NOT QUOTE OR CITE
-------
a. Denali National Park, AK
April 8-10, 2001
b. Voyageurs National Park, MN
May 19-20, 2001
C.A.
cn
O
4ft-
4fi-
AA
C
— ^ 40-
Q.
O 20-
10-
n-
rs /-v
11 \
... .../ \r \
W
Time
fin-i
Q.
O on-
n.
fin-
cn_
on_
O
on_
m-
n
~^N
/"*V^y\ rl / \
V^*V \A / \
^— v V \ S \
^-/ \T
OM T-T-OM T-T-T-?\i T-T-T-Cvi
Time
c. Olympic National Park, WA
May 26-27, 2001
r"%
i V.
/
•\ r/ ^/v
v V
O CO CD O) CN LO 00 ^~ O CO CD O) CN LO 00 ^~
Time
d. Glacier National Park, MT e. Yellowstone National Park, WY
May 4-5, 2001 April 25-26, 2001
r~\
I \-
••-,.
•
I
V
60-
on.
o
on.
10-
n-
A/**^^
/"* * ^^ "^v^ /
' W
Time
Time
Figure 3-12. Time-series of hourly average ozone concentrations observed at five national
parks: Denali, AK; Voyageur, MN; Olympic, WA; Glacier, MT; and
Yellowstone, WY.
January 2005
3-50
DRAFT-DO NOT QUOTE OR CITE
-------
1 Although many of the features of the day-to-day variability of O3 at relatively remote
2 monitoring sites in the United States are simulated reasonably well by Fiore et al. (2003),
3 uncertainties in the calculation of the temporal variability of O3 originating from different
4 sources on shorter time scales that must be recognized. The uncertainties stem in part from
5 seasonal variability in the STE of O3 that is too low (Fusco and Logan, 2003), the geographical
6 variability of this exchange, and the variability in the exchange between the free troposphere and
7 the planetary boundary layer in the model.
8 Ideally, the predictions resulting from an ensemble of models should be compared with
9 each other and with observations so that the range of uncertainty inherent in the model
10 predictions can be evaluated.
11
12
13 3.8 INDOOR SOURCES AND EMISSIONS OF OZONE
14 Ozone enters the indoor environment primarily through infiltration from outdoors and
15 through building components, such as windows, doors, and ventilation systems. There are also a
16 few indoor sources of O3 (photocopiers, facsimile machines, and laser printers and electrostatic
17 air cleaners and precipitators) (Weschler, 2000). Generally O3 emissions from office equipment
18 and air cleaners are low except under improper maintenance conditions. Reported O3 emissions
19 from office equipment range from 1300 to 7900 |ig/h (Leovic et al., 1998, 1996). Most air
20 cleaners (paniculate ionizers) emitted no or only a small amount (56 to 2757 |ig/h) of O3 during
21 operation (Niu et al., 2001). Emissions from O3 generators can range from tens to thousands of
22 micrograms per hour (Weschler, 2000; U.S. Environmental Protection Agency, 1996).
23
24 Factors Affecting Ozone Concentrations Indoors
25 The concentration of O3 in indoor environments is dependent on the outdoor O3
26 concentration, the air exchange rate (AER) or outdoor infiltration, indoor circulation rate, and
27 O3 removal processes through contact with indoor surfaces and reactions with other indoor
28 pollutants. Since O3 concentrations are generally higher during the warmer months, indoor
29 concentrations will likely be highest during that time period. (See earlier discussion on ambient
30 concentrations of O3.)
January 2005 3-51 DRAFT-DO NOT QUOTE OR CITE
-------
1 The AER, the balance of the flow of air in and out of a microenvironment, is greatest in a
2 residential building when a window or door is open (Johnson et al., 2004; Howard-Reed et al.,
3 2002). The opening of windows or doors is dependent on the building occupancy, season,
4 housing density, the presence of air conditioning and wind speed (Johnson and Long, 2004).
5 When windows and doors are closed, the dominant mechanism controlling AERs is infiltration
6 through unintentional openings in the building envelope. One of the most comprehensive
7 evaluations of AERs for residential structures was reported by Murray and Burmaster (1995)
8 and includes AERs for 2,844 residential structures in four different regions by season (winter,
9 spring, summer, and fall). The AER for all seasons across all regions was 0.76 IT1 (arithmetic
10 mean). Table 3-9 lists the mean AER for the seasons and regions. The AERs were generally
11 higher during the warm seasons, when ambient O3 concentrations are highest.
12 Average mean (median) AERs of 2.45 (2.24), 1.35 (1.09), and 2.22 (1.79) IT1 were
13 reported by Lagus Applied Technology, Inc. (1995) for schools, offices, and retail
14 establishments in California. The AERs for schools, offices, and retail establishments in Oregon
15 and Washington were 0.32, 0.31, and 1.12 IT1 (Turk et al., 1989), considerably less than that
16 reported by Lagus Applied Technology. Park et al. (1998) reported AERs ranging from 1.0 to
17 47.5 h'1 for stationary vehicles under varying ventilating conditions. Where available, AERs for
18 other studies are included in Table 3-10.
19 The most important removal process for O3 in the indoor environment is deposition on and
20 reaction with indoor surfaces. The rate of deposition is material-specific. The removal rate will
21 depend on the indoor dimensions, surface coverings, and furnishings. Smaller rooms generally
22 have larger surface-to-volume ratio (A/V) and remove O3 faster. Fleecy materials, such as
23 carpets, have larger surface-to-volume ratios and remove O3 faster than smooth surfaces
24 (Weschler, 2000). However, the rate of O3 reaction with carpet diminishes with cumulative O3
25 exposure (Morrison et al., 2000; Morrison and Nazaroff, 2002). Weschler (2000) compiled the
26 O3 removal rates for a variety of microenvironments. Generally, the removal rates ranged
27 between 3.0 and 4.3 kd (AAO/lr1. The highest removal rate, 7.6 kd (A/vyir1, was noted for a
28 clean room (Weschler et al, 1989).
29 Ozone reacts with terpenes from wood products, solvents, or odorants to produce
30 submicron particles (Nazaroff and Weschler, 2004; Wainman et al., 2000; Weschler and Shields,
31 1999; Grosjean and Grosjean, 1996, 1998), possibly resulting in PM25 concentrations > 20 |ig/m3
January 2005 3-52 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-9. Air Exchange Rates in Residences by Season and Region of the Country*
Season"
All
1
2
3
4
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
Region
All
All
All
All
All
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Sample Size
2844
1139
1051
529
125
161
428
96
454
254
43
165
589
5b
2b
34
488
47
23b
37
18b
Mean
0.76
0.55
0.65
1.5
0.41
0.36
0.57
0.47
0.63
0.44
0.52
0.59
0.77
0.82
1.31
0.68
1.57
0.25
0.35
0.51
0.72
Std Dev
0.88
0.46
0.57
1.53
0.58
0.28
0.43
0.4
0.52
0.31
0.91
0.43
0.62
0.69
na
0.5
1.56
0.15
0.18
0.25
1.43
The data does not represent all areas of the country.
a Season 1: December, January, February; Season 2: March, April, May; Season 3: June, July, August;
Season 4: September, October, November.
Note: Small sample size, n< 25.
c Estimated using locations of residences evaluated in the region.
Source: Adapted from Murray and Burmaster (1995).
January 2005 3-53 DRAFT-DO NOT QUOTE OR CITE
-------
Table 3-10. Indoor/Outdoor Ozone Concentrations in Various Microenvironments
to
o
o
Indoor/Outdoor
Location and Ventilation Conditions Concentrations
Comments
Reference
fe
H
O
H
O
c
o
H
W
O
V
O
HH
H
W
New England States (9) 20 ppb/40 ppb
Fall
Mexico City, School
Windows/Doors Open (27) 0 to 247 ppb/
Windows/Doors Closed Cleaner 64 to 361 ppb
Off (41)
Windows/Doors Closed Cleaner
On (47)
Mexico City
Homes
Schools
5 ppb/27ppb (7 d)
7 ppb/37 ppb (14 d)
22ppb/56to733ppb
Boston, MA, Homes (9)
Winter - continuously
Summer - continuously
Los Angeles, Homes (239)
0 to 20.4 ppb/4.4 to 24.5 ppb
0 to 34.2 ppb/8.2 to 51.8 ppb
13 ppb/37 ppb
Schools represented a variety of environmental conditions - varying NESCAUM
ambient O3 concentrations, sources, geographic locations, population (2002)
density, traffic patterns, building types. Average O3 concentrations were
low in the morning and peaked during the early afternoon. O3
concentrations averaged for all schools monitored.
Study conducted over 4 d period during winter months. Two-minute 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 IT' for low, medium, and high flow rates. Ozone
concentrations decreased with increasing relative humidity.
Ozone monitoring occurred between September and July. Study Romieu
included 3 schools and 145 homes. Most of the homes were large and et al. (1998)
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 7 and 14 consecutive days.
Study examined the potential for O3 to react with VOCs to form acid
aerosols. Carbonyls were formed. No clear trend of O3 with AERs. The Reiss et al.
average AER was 0.9 h"1 during the winter and 2.6 h"1 during the summer. (1995)
Four residences in winter and nine in summer with over 24 h samples
collected.
Four hundred and eighty-one samples collected inside and immediately
outside of home from February to December. Concentrations based on 24-h Avol et al.
average O3 concentrations indoors and outdoors. Low outdoor (1998)
concentrations resulted in low indoor concentrations. However, high
outdoor concentrations resulted in a range of indoor concentrations.
-------
Table 3-10 (cont'd). Indoor/Outdoor Ozone Concentrations in Various Microenvironments
p
1
to
o
o
v\
Location and Ventilation Conditions
Burbank, CA
Telephone Switching Station
Munich Germany
Office
Gymnasium
Classroom
Residence
Bedroom
Livingroom
Indoor/Outdoor
Concentrations
0.2/2 l.lppb
0.4/0. 9 ppb
0.49/0.92 ppb
0.54/0.77 ppb
0.47/1.0 ppb
0.74/1.0 ppb
Comments
Major source of O3 was transport from outdoors. From early spring to late
fall O3 concentrations peaked during the early afternoon and approach zero
at sunset. AER ranged from 1.0 to 1.9 h'1.
Indoor concentrations were dependent on the type of ventilation.
Reference
Weschlov
etal. (1994)
Jakobi and
Fabian
(1997)
o
*
o
H
O
O
H
W
O
O
HH
H
W
Montpellier, France, Homes (110)
Southern CA, Homes
Upland
Mountains
Krakow, Poland, Museums
Cloth Hall
Matejko
Wawel Castle
National
Buildings, Greece
Thessalonki
Athens
15.5/32.0 ppb
11.8/48.2 ppb
2.8/35.7 ppb
3.2/25.7-27.4 ppb
8.5/20.0 ppb
2.5/14.7 ppb
1.5/11.0 ppb
9.39/15.48 ppb
8.14/21.66 ppb
Ozone measurements were made over 5-d periods in and outside Bernard et al.
of 21 homes during the summer and winter months. The winter I/O ratio (1999)
was 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. Concentrations were (2000)
based on average monthly outdoor concentrations and average weekly indoor Lee et al.
concentrations. Indoor based on the home location, number of bedrooms, (2002)
and the presence of an air conditioner.
Ozone continuously monitored at five museums and cultural centers. Salmon et al.
Monitoring conducted during the summer months for 21 to 46 h or 28 to 33 (2000)
days at each of the sites. The indoor concentration was found to be
dependent on the ventilation rate, i.e., when the ventilation rate was high the
indoor O3 concentrations approached that of ambient O3. Rooms sequestered
from the outdoor air or where air was predominantly recycled through
charcoal filters the ozone levels indoors were greatly reduced.
There was no heating/air conditioning system in the building at Drakou et al.
Thessaloniki. Windows were kept closed during the entire monitoring (1995)
period. Complete air exchange took place every 3 h. 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.
-------
Table 3-10 (cont'd). Indoor/Outdoor Ozone Concentrations in Various Microenvironments
to
o
o
Location and Ventilation
Conditions
Indoor/Outdoor
Concentrations
Comments
Reference
Patrol cars, NC
University of CA
Photocopy room
Home/office
O3 generators
11.7/28.3 ppb
< 20 ppm/—
14 to 200 ppb/-
Patrol cars were monitored Mon. through Thurs. between the hours of Riediker
3 p.m. to midnight on 25 occasions during the months of Aug., Sept., and et al. (2003)
Oct. Outdoor O3 concentrations were taken from ambient monitoring
station. Air inside the patrol car was recirculated cool air.
Room volume was 40 m3. Ozone concentrations increased Black etal.
proportionately with increasing use of photocopier. (2000)
Room volume was 27 m3. Doors and windows were closed. Heating/air Steiber et al.
conditioning and mechanical ventilation systems were off. Ozone (1995)
generator was operated for 90 min. High ozone concentrations noted
when O3 generator used at high setting. AER was 0.3 h~'.
fe
H
O
H
O
c
o
H
W
O
V
O
HH
H
W
-------
1 (Wainman et al., 2000). Indoor hydroxy radical (*OH) concentrations increase nonlinearly with
2 increased indoor O3 concentrations and indoor alkene emissions. Sarwar et al. (2002) suggested
3 that the »OH reacts with terpenes to produce products with low vapor pressures that contribute to
4 fine particle growth.
5
6 Monitored and Modeled Ozone Concentration in Microenvironments
1 Ozone concentrations in various microenvironments under a variety of environmental
8 conditions have been reported in the literature. In the absence of an indoor O3 source,
9 concentrations of O3 indoors are lower than that found in the ambient air. Ozone concentrations
10 in microenvironments were found to be primarily controlled by ambient O3 concentrations and
1 1 the AER; they increase with increasing AER. To a lesser extent, O3 concentrations in
12 microenvironments are influenced by the ambient temperature, time of day, indoor
13 characteristics (e.g., presence of carpeting), and the presence of other pollutants in the
14 microenvironment. Table 3-9 describes the findings of the available studies.
15 Estimates of indoor O3 concentrations may be made using a simple one-compartment mass
16 balance model. The model takes into account the effects of ventilation, filtration, heterogeneous
17 removal, direct emission, and photolytic and thermal and chemical reactions. The simplest form
18 of the model is represented by the following differential equation:
19
dCIN S
20
21 where dCIN is the indoor pollutant concentration (mass/volume), dt is time in hours, v is the air
22 exchange rate, COUT is the outdoor pollutant concentration (mass/volume), Fis the volume of the
23 microenvironment, and S is the indoor source emission rate. When the model was used to
24 estimate indoor O3 concentrations, indoor concentrations were found to be 33% of outdoor O3
25 concentrations (Freijer and Bloemen, 2000). A more in-depth discussion of the mass balance
26 model has been reported in Nazaroff and Cass (1986).
27
28
January 2005 3-57 DRAFT-DO NOT QUOTE OR CITE
-------
1 3.9 HUMAN EXPOSURE TO OZONE
2 Humans are exposed to O3 and related photochemical oxidants through the exchange
3 boundary, the skin and the openings into the body such as the mouth, the nostrils, and punctures
4 and lesions in the skin (U.S. Environmental Protection Agency, 1992; Federal Register, 1986).
5 Inhalation exposure to O3 and related photochemical oxidants is determined by pollutant
6 concentrations measured in the breathing zone that is not affected by exhaled air as the
7 individual moves through time and space.
9
10
11
12
13
14
15
16
17
Quantification of Exposure
Ambient O3 concentrations vary with time of day (peaking during the latter portion of the
day) and season and among locations. Consequently, exposure to O3 will change as a function of
time of day and as an individual moves among locations. A hypothetical exposure is
demonstrated in Figure 3-13. The actual dose received also changes during the day and is
dependent on the O3 concentration in the breathing zone and the individual's breathing rate,
which is, in turn, dependent on the individual's level of exertion.
ro
-*-*
£=
0)
O
c.
o
o
Peak
Exposure
Averaged
Exposure
Instantaneous
\jtf^' Exposure
Time
Figure 3-13. Ozone exposure time profile.
January 2005
3-58
DRAFT-DO NOT QUOTE OR CITE
-------
1 When measuring or modeling exposure to O3 and related photochemical oxidants
2 consideration should be given to the diurnal weekly (weekday-weekend) and seasonal
3 variability. Peak concentrations, lasting for several hours, typically occur towards the latter
4 portion of the day during the summer months. Regional O3 episodes often co-occur with high
5 concentrations of airborne fine particles, making it difficult to assess O3 dynamics and exposure
6 patterns. Also, while there are few indoor O3 sources, O3 will react with materials and other
7 pollutants in the indoor environment in an analogous fashion to that occurring in the ambient
8 atmosphere, potentially exposing subjects to other more toxic pollutants (Nazaroff and Weschler,
9 2004; Lee and Hogsett, 1999; Wainman et al., 2000; Weschler and Shields, 1997). (See
10 discussion on O3 atmospheric chemistry and indoor sources and concentrations earlier in this
11 chapter.)
12
13 Personal Exposure and A mbient Concentrations
14 The two methods for measuring personal exposure are (a) the direct method, using a
15 personal exposure monitor (PEM) consisting of a passive sampler worn around the breathing
16 zone, and (b) the indirect method, which measures or estimates the O3 concentrations through
17 the use of models or biomarkers of exposure in all of the microenvironments the individual
18 encounters (Ott, 1982, 1985). The concept of microenvironment is important in the
19 understanding of human exposure modeling. Often identified with a perfectly mixed
20 compartment, microenvironments are more recently viewed as a controlled volume, indoors or
21 outdoors, that can be characterized using a set of either mechanistic or phenomenological
22 governing equations. This allows for a nonhomogeneous environment, including sources and
23 sinks within the microenvironment. Microenvironments include indoor residences, other indoor
24 locations, outdoors near roadways, other outdoor locations, and in-vehicles.
25 Although it is difficult to develop passive monitors for personal exposure measurements
26 because of problems in identifying chemical or trapping reagents that can react with O3, several
27 modified passive samplers have been developed for use in personal O3 exposure measurements
28 (Bernard et al., 1999; Koutrakis et al., 1993; Avol et al., 1998; Geyh et al., 1997, 1999). Some
29 personal exposure measurements using passive samplers show O3 exposures below those O3
30 concentrations measured at outdoor stationary sites (Delfino et al., 1996; Avol et al., 1998;
31 Sarnat et al., 2000; Geyh et al., 2000; Brauer and Brook, 1997). However, other studies have
January 2005 3-59 DRAFT-DO NOT QUOTE OR CITE
-------
1 found strong correlations between O3 measured at stationary sites and personal monitored
2 concentrations (Liard et al., 1999; Bauer and Brook, 1997; Linn et al., 1996; Lee et al., 2004;
3 Avol et al., 1998; O'Neill et al., 2003) when the time spent outdoors, age, gender, and
4 occupation of the subjects were considered.
5 Hourly O3 concentrations from monitoring stations have been used as surrogates of
6 exposure in epidemiological studies in evaluating exposure-related health effects. Routinely
7 collected ambient data, though readily available and convenient, may not represent true exposure
8 and may tend to underestimate the effect of the air pollutant on health (Krzyznowki, 1997).
9 In some instances, ambient O3 monitors are located in areas outside the breathing zone of the
10 general population. Studies on the effect of elevation on O3 concentrations found that
11 concentrations increased with increasing elevation (Vakeva et al., 1999; Johnson, 1997).
12 Further, since O3 monitors are frequently located on rooftops in urban settings, the
13 concentrations measured there may overestimate the exposure to individuals outdoors in streets
14 and parks, locations where people exercise and maximum O3 exposure is likely to occur.
15 Accordingly, the resulting exposure measurement error and its effect on the estimates of relative
16 risk should be taken into consideration.
17 There are three components to exposure measurement error: (1) the use of aggregate,
18 rather than individual, exposure data; (2) the difference between the average personal exposure
19 and the true ambient concentration; and (3) the difference between true and measured ambient
20 concentration levels. Zeger et al. (2000) indicated that the first and third error components are
21 largely Berksonian errors and would not significantly bias the risk estimate. The error resulting
22 from the difference between the personal and ambient concentration levels, however, may
23 introduce bias, especially if indoor sources are associated with ambient levels. (See discussion
24 on indoor sources of O3 earlier in this chapter.)
25 Studies by Brauer and Brook (1997), Chang et al. (2000), Lee et al. (2004), Liard et al.
26 (1999), Linn et al. (1996), Liu et al. (1995,1997), O'Neill et al. (2003), Delfmo et al. (1996),
27 Avol et al. (1998b), and Sarnat et al. (2001) examined the relationship between ambient O3
28 concentrations from a central monitoring site and personal O3 exposure. Sarnat et al. (2001),
29 found that averaged 24-h O3 concentrations from a stationary monitoring site were not
30 significantly associated with O3 concentrations from personal monitors in several exposed
31 groups in Baltimore, MD. The mixed regression effect estimates were P = 0.01 (t= 1.21)and
January 2005 3-60 DRAFT-DO NOT QUOTE OR CITE
-------
1 P = 0.00 (t = 0.03), for summer and winter, respectively. In contrast, O'Neill et al. (2003) found
2 a statistically significant association between personal and ambient O3 concentrations in Mexico
3 City outdoor workers (P = 0.56, t = 8.52). The subjects in the Sarnat et al. (2001) study spent
4 less than 6% of their time outdoors, whereas the personal exposure data from O'Neill et al.
5 (2003) were from subjects who spent the entire measurement period outdoors. Brauer and Brook
6 (1997) observed that the averaged personal O3 measurements and ambient concentrations were
7 well correlated after stratifying groups by time spent outdoors. Lee et al. (2004) also observed
8 that personal O3 exposure was positively correlated with time spent outdoors (r = 0.19, p < 0.01)
9 and negatively correlated with time spent indoors (r = -0.17, p < 0.01). Liu et al. (1995) found
10 that after adjusting for time spent in various indoor and outdoor microenvironments (i.e., car
11 with windows open, car with windows closed, school, work, home, outdoors near home,
12 outdoors other than near home), mean 12-h ambient O3 concentrations explained 32% of the
13 variance seen in personal exposure during the summer months. Delfmo et al. (1996) reported a
14 moderate correlation (r = 0.45; range: 0.36 to 0.69) between personal exposure and ambient O3
15 concentrations measured at a stationary site. When asthmatic subjects were measured over a
16 12-h period during the daytime, the mean personal exposures were only 27% of the measured
17 ambient concentrations. Chang et al. (2000) compared 1-h personal and ambient O3
18 measurements in several microenvironments in Baltimore, MD. There was no correlation
19 between personal exposure and ambient O3 concentrations in the indoor residence (r = 0.09 and
20 r = 0.05, for summer and winter, respectively), although a moderate correlation was found for
21 indoor microenvironments other than the residence (r = 0.34 in summer, r = 0.46 in winter). The
22 correlation between personal exposure and ambient O3 in outdoor environments near and away
23 from roadways was moderate to high (0.68 < r < 0.91). Liard et al. (1999) found a high variance
24 in O3 measurements from stationary monitoring sites and PEMs in adults and children monitored
25 three times for 4 consecutive day intervals. For each period, all adults wore the O3 monitors
26 over the same 4 days. However, when personal measurements from all subjects were aggregated
27 for each of the exposure periods, the mean personal O3 exposure was found to be highly
28 correlated with the corresponding 4-day mean ambient concentration (r = 0.83, p <0.05).
29 Similarly, a study of Los Angeles school children by Linn et al. (1996) found that 24-h average
30 ambient O3 concentrations from a central site were well-correlated (r = 0.61) with averaged
31 personal O3 exposure. Results from these studies suggest that although O3 concentrations from
January 2005 3-61 DRAFT-DO NOT QUOTE OR CITE
-------
1 stationary ambient monitoring site do not explain the variance of individual personal O3
2 exposures, they may serve as surrogate measures for aggregate personal exposures.
3
4 Modeled Personal Exposures
5 Exposure modeling is often used in evaluating exposure to large populations over time.
6 The use of models is complicated by the fact that O3 is a secondary pollutant with complex
7 nonlinear and multiscale dynamics in space and time. Ozone is formed in the atmosphere
8 through a series of chemical reactions involving the precursors VOCs and NOX. Therefore, O3
9 exposures may be affected by: (1) emission levels and spatiotemporal patterns of VOCs and
10 NOX; (2) ambient atmospheric as well as indoor microenvironmental transport, removal and
11 mixing processes; and (3) chemical transformations that take place over a multitude of spatial
12 scales. The transformations are dependent on the presence of co-occurring pollutants and the
13 nature of surfaces interacting with the pollutants.
14 Exposure models may be classified as (1) potential exposure models, typically the
15 maximum outdoor concentrations versus "actual" exposure, including locally modified
16 microenvironmental outdoor and indoor exposures; (2) population versus "specific individual"
17 based exposure models; (3) deterministic versus probabilistic models; and (4) observation versus
18 mechanistic air quality model driven estimates of spatially and temporally varying O3
19 concentrations.
20 There are several steps involved in defining exposure models. The steps are based on
21 frameworks described in the literature over the last 20 years and the structure of various existing
22 inhalation exposure models (NEM/pNEM, HAPEM, SHEDS, REHEX, EDMAS, MENTOR-
23 OPERAS, APEX, AIRPEX, AIRQUIS). The steps include (1) estimation/determination of the
24 background or ambient levels of O3; (2) estimation/determination of levels and temporal profiles
25 of O3 in various microenvironments; (3) characterization of relevant attributes of individuals or
26 populations under study (age, gender, weight, occupation, other physiological characteristics);
27 (4) development of activity event or exposure event sequences; (5) determination of appropriate
28 inhalation rates during the exposure events; (6) determination of dose; (7) determination of
29 event-specific exposure and intake dose distributions for selected time periods; and
30 (8) extrapolation of population sample (or cohort) exposures and doses to the entire populations
January 2005 3-62 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
of interest. Figure 3-14 provides a conceptual overview of the pNEM exposure model. A more
detailed overview of an exposure model can be found in Annex AX3.
Air Exchange
Rates and
Building
Volumes
Seasonal
Considerations
(Temperature)
Emission Rates and
Use Patterns for
Indoor Sources
(e.g., gas appliances,
passive smoking}
Mass Balance
Model for Indoor
M icroenvironments
Ambient
Fixed-Site
Concentrations
Air Quality
Specification
Human Activity
and Exertion
Patterns
(Exercise Level
Patterns)
Population and
Commuting
Data
Exposure Algorithms
Distribution of People and
Occurrences of Exposures
Linked with Breathing Rate
(Minute Ventilation Rate)
Figure 3-14. Conceptual overview of pNEM. Model inputs (e.g., activity patterns,
ambient monitoring data, air exchange rates) are in round-corner boxes
and model calculations are shown in rectangles.
Source: Johnson etal. (1999).
1 Outdoor concentrations of O3 are estimated either through emissions-based mechanistic
2 modeling, or through ambient-data-based modeling. Emissions-based models determine the
3 spatiotemporal fields of the O3 concentrations using precursor emissions and meteorological
4 conditions as inputs. (They are described in Annex AX2, Section AX2.5). The ambient-data-
5 based models determine spatial or spatiotemporal distributions of O3 through the use of
6 interpolation schemes. The kriging approach provides standard procedures for generating an
7 interpolated O3 spatial distribution for a given period of time (Georgopoulos et al., 1997a,b).
8 The Spatio-Temporal Random Field (STRF) approach has been used to interpolate monitoring
9 data in both space and time (Christakos and Vyas, 1998a,b). The STRF approach can analyze
10 information on temporal trends which cannot be directly incorporated by kriging.
January 2005
3-63
DRAFT-DO NOT QUOTE OR CITE
-------
1 Several approaches are available for modeling microenvironmental concentrations:
2 empirical, mass balance, and detailed computational fluid dynamics (CFD) models. Empirical
3 relationships provide the basis for future, "prognostic" population exposure models. Mass
4 balance modeling is the most common approach used to model pollutant concentrations in
5 enclosed microenvironments. Mass balance modeling ranges from very simple formulations,
6 assuming ideal (homogeneous) mixing and only linear physicochemical transformations with
7 sources and sinks, to models that take into account complex multi-phase chemical and physical
8 interactions and nonidealities in mixing. The pNEM/O3 model, discussed later in this chapter,
9 includes a sophisticated mass balance model for indoor and vehicle microenvironments
10 (Johnson, 2003). CFD models take into account the complex, multiphase processes that affect
11 indoor concentrations of interacting gas phase pollutants, such as the interactions of O3 with
12 indoor sinks and sources (surfaces, gas releases) and with entrained gas (Sarwar et al., 2001,
13 2002; S0rensen and Weschler, 2002).
14 To estimate the actual O3 dose delivered to the lung, information on the concentration,
15 minute ventilation rate, activity level, and the morphology of the respiratory tract are needed.
16 Limited data have been compiled for ventilation rates for different age groups, both healthy and
17 compromised individuals, at varies levels of activity (Klepeis et al., 1996, 2001; Avol et al.,
18 1998b; Adams, 1993). Based on the available information, the highest level of outdoor activity
19 occurs during the spring and summer months, during the mid- to late afternoon and early
20 evening, the times when O3 concentrations are highest. Children are likely more susceptible to
21 the effects of O3 than other groups. School-age children spend more time outdoors engaged in
22 high level activities than do other groups and breath more air in than adults relative to body
23 surface area, breathing frequency, and heart rate. Asthmatic children spend the same amount of
24 time outdoors as other more healthy children but the time spent engaged in high levels of activity
25 are less.
26 Estimates of activity level have been compiled based on questionnaire data. The National
27 Human Activity Pattern Survey (NHAPS), a probability-based telephone survey, was conducted
28 in the early 1990s. The survey concluded that outdoor work-related activities were highest
29 during the springtime and were more frequent during the morning and early afternoon.
30 Exercise/sports-related activities were highest from noon to 3 p.m. during the summer months.
31 During the spring months, exercise/sports-related activities were highest from mid- to late
January 2005 3-64 DRAFT-DO NOT QUOTE OR CITE
-------
1 afternoon (Klepeis et al., 1996, 2001). A pilot study by Gonzalez et al. (2003) evaluated the use
2 of retrospective questionnaires for reconstructing past time-activity and location pattern
3 information. Ozone concentration estimates using ambient stationary monitors and estimates
4 derived from diaries and questionnaires differed slightly. However, both estimates were greater
5 than O3 personal exposure measurements.
6 Existing comprehensive inhalation exposure models (NEM and pNEM) (Johnson, 2003),
7 (SHEDS) Burke et al., 2001; McCurdy et al., 2000), and the Air Pollutants Exposure model
8 (APEX version 3) treat human activity patterns as sequences of exposure events in which each
9 event is defined by a geographic location and microenvironment and then assigned activity diary
10 records from the CHAD (Consolidated Human Activities Database; www.epa.gov/chadnetl)
11 (Glen et al., 1997; McCurdy, 2000; McCurdy et al., 2000). There are now about 22,600 person-
12 days of sequential daily activity pattern data in CHAD representing all ages and both genders.
13 The data for each subject consist of one or more days of sequential activities, in which each
14 activity is defined by start time, duration, activity type (140 categories), and microenvironment
15 classification (110 categories). Activities vary from 1 min to 1 h in duration. Activities longer
16 than 1 h are subdivided into clock-hour durations to facilitate exposure modeling. A distribution
17 of values for the ratio of oxygen uptake rate to body mass (referred to as metabolic equivalents
18 or METs) is provided for each activity type listed.
19 pNEM divides the population of interest into representative cohorts based on the
20 combinations of demographic characteristics (age, gender, and employment), home/work
21 district, and residential cooking fuel. APEX and SHEDS generate a population demographic file
22 containing a user-defined number of person-records for each census tract of the population based
23 on proportions of characteristic variables (age, gender, employment and housing) obtained for
24 the population of interest, and then assigns the matching activity information from CHAD to
25 each individual record of the population based on the characteristic variables.
26 An important source of uncertainty in existing exposure modeling involves the creation of
27 multiday, seasonal, or year long exposure activity sequences based on 1- to 3-day activity data
28 for any given individual from CHAD. Currently, appropriate longitudinal data are not available
29 and the existing models use various rules to derive longer-term activity sequences utilizing 24-h
30 activity data from CHAD.
January 2005 3-65 DRAFT-DO NOT QUOTE OR CITE
-------
1 Of the above models, only NEM/pNEM have been used extensively in O3 exposure
2 modeling. The pNEM probabilistic model builds on the earlier NEM deterministic exposure
3 model. The model take into consideration the temporal and spatial distribution of people and O3
4 in the area of consideration, variations in O3 concentrations in the microenvironment, and the
5 effects of exercise increased ventilation on O3 uptake. There are three versions of the pNEM/O3
6 model: (1) general population (Johnson et al., 1996a), (2) outdoor workers (Johnson et al.,
7 1996b), and (3) outdoor children (Johnson et al., 1996c, 1997). The pNEM models have been
8 applied to nine urban areas and a summer camp. The models used activity data from the
9 Cincinnati Activity Diary Study (CADS) along with time/activity data from several other
10 studies. Data from stationary monitoring sites were used to estimate outdoor O3 exposure.
11 Indoor O3 decay was assumed to be proportional to the indoor O3 concentration. An algorithm
12 assigned the EVR associated with each exposure event. The EVR for the outdoor children
13 model was generated using a module based on heart rate data by Spier et al. (1992) and Linn
14 etal. (1992).
15
16 Exposure to Related Photochemical Oxidants
17 A variety of related photochemical oxidants produced outdoors, such as PAN and
18 peroxypropionyl nitrate (PPN) can infiltrate into indoor environments. These compounds are
19 thermally unstable and decompose to peroxacetyl radicals and NO2. Exposure to related
20 photochemical oxidants have not been measured, nor are these compounds routinely monitored
21 at stationary monitoring sites. Available monitored concentrations of related photochemical
22 oxidants may be found in Annex AX3.
23
24
25 3.10 SUMMARY OF KEY POINTS
26 Ozone monitored in the United States during "ozone season" vary in length depending on
27 location. The O3 season extends all year long in the Southwest. In most other areas of the
28 country, the O3 season typically last from April to October. Median O3 concentrations averaged
29 over the appropriate O3 season are about 0.035 ppm. The daily maximum 1-h O3 concentrations
30 could have been much higher in large urban areas or in areas downwind of large urban areas.
31 For example, in Houston, TX the daily maximum 1-h O3 concentrations reached 0.202 ppm in
January 2005 3-66 DRAFT-DO NOT QUOTE OR CITE
-------
1 1999 and reached 0.161 ppm in 2000. Although daily maximum 1-h average O3 concentrations
2 in some areas of the United States downwind of major sources of O3 precursors can exceed
3 0.120 ppm, only about 1% of hourly average O3 concentrations exceed 0.100 ppm at sites in
4 these areas.
5 Daily maximum 8-h average O3 concentrations are lower than the maximum 1-h O3
6 concentrations but they are highly correlated. Within individual MSAs, O3 concentrations tend
7 to be well correlated across monitoring sites. However, there can be substantial variations in
8 O3 concentrations. Ozone in city centers tends to be lower than in regions either upwind or
9 downwind because of titration by NO emitted by motor vehicles.
10 Ozone concentrations tend to peak in early- to mid-afternoon in areas where there is strong
11 photochemical activity and later in the day in areas where transport is more important in
12 determining the O3 abundance. Summertime maxima in O3 concentrations occur in areas in the
13 United States where there is substantial photochemical activity involving O3 precursors emitted
14 from human activities. Monthly maxima can occur anytime from June through August.
15 However, springtime maxima are observed in national parks, mainly in the western United States
16 and at a number of other relatively unpolluted monitoring sites throughout the Northern
17 Hemisphere. For example, the highest O3 concentrations at Yellowstone National Park tend to
18 occur during April and May. Generally, monthly minima O3 concentrations tend to occur from
19 November through February at polluted sites and during the fall at relatively remote sites.
20 Nationwide, 1-h O3 concentrations decreased approximately 29% from 1980 to 2003 and
21 approximately 16% from 1990 to 2003. The 8-h O3 concentration decreased approximately 21%
22 since 1980 and approximately 9% since 1990. Ozone concentrations have continued to decrease
23 nationwide, but the rate of decrease has slowed since 1990. However, these trends have not been
24 uniform across the United States. In general, reductions in O3 have been largest in New England
25 and states along the West Coast and smallest in states in the Midwest. Downward trends in O3 in
26 California have been driven mainly by trends in Southern California, with reductions in other
27 areas not being as large.
28 Sufficient data are not available for other oxidants (e.g., H2O2, PAN) and oxidation
29 products (e.g., HNO3, H2SO4) in the atmosphere to relate concentrations of O3 to these species.
30 Data for these species are only obtained as part of specialized field studies. In general,
31 secondary species, such as HNO3, H2SO4, H2O2, and PAN, are expected to be at least moderately
January 2005 3-67 DRAFT-DO NOT QUOTE OR CITE
-------
1 correlated with O3. On the other hand, primary species are expected to be more highly correlated
2 with each other than with secondary species, provided that the primary species originate from
3 common sources. The relationship of O3 to PM2 5 is complex because PM is not a distinct
4 chemical species, but is a mix of primary and secondary species. PM25 concentrations were
5 positively correlated with O3 during summer, but negatively correlated with O3 during winter at
6 Ft. Meade, MD. More data are needed before this result can be applied to other areas.
7 Co-occurrences of O3 ( defined when both pollutants are present at an hourly average
8 concentration of > 0.05 ppm) with NO2 and SO2 are rare. For example, there were fewer than
9 10 co-occurrences with either NO2 or SO2 in 2001. The number of co-occurrences for O3 and
10 PM2 5 (defined as an hourly average O3 concentration >0.05 ppm and a 24-h average PM25
11 concentration > 40 |ig/m3 occurring during the same 24-h period) also tended to be infrequent
12 (< 10 times) at most sites, but there were up to 20 such co-occurrences at a few sites.
13 Policy relevant O3 background concentrations are used for assessing risks to human health
14 associated with O3 produced from anthropogenic sources in continental North America. Because
15 of the nature of the definition of PRB concentrations, they cannot be directly derived from
16 monitored concentrations, instead they must be derived from modeled estimates. Current model
17 estimates indicate that the PRB concentrations in the United States ambient air are generally
18 0.015 ppm to 0.035 ppm. They decline from spring to summer and are generally < 0.025 ppm
19 under conditions conducive to high O3 episodes. PRB concentrations can be higher, especially at
20 elevated sites during spring, due to enhanced contributions from hemispheric pollution and
21 stratospheric exchange.
22 Humans are exposed to O3 either outdoors or in various micro-environments. Ozone in
23 indoor environments results mainly from infiltration from outdoors. Once indoors, O3 is
24 removed by deposition on and reaction with surfaces and reaction with other pollutants in the
25 indoor environment such as terpenes (e.g., wood products, solvents, odorants) to produce
26 ultrafine particles. Personal exposure to O3 tends to be positively associated with time spent
27 outdoors. Although O3 concentrations obtained at stationary monitoring sites may not explain
28 the variance in individual personal exposures, they may serve a surrogate measures for aggregate
29 personal exposures.
30
January 2005 3-68 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2 Adams, W. C. (1993) Measurement of breathing rate and volume in routinely performed daily activities [final
3 report]. Sacramento, CA: California Environmental Protection Agency, Air Resources Board; contract no.
4 A033-205.
5 Altshuller, A. P.; Lefohn, A. S. (1996) Background ozone in the planetary boundary layer over the United States.
6 J. Air Waste Manage. Assoc. 46: 134-141.
7 Avol, E. L.; Navidi, W. C.; Colome, S. D. (1998a) Modeling ozone levels in and around southern California homes.
8 Environ. Sci. Technol. 32: 463-468.
9 Avol, E. L.; Navidi, W. C.; Rappaport, E. B.; Peters, J. M. (1998) Acute effects of ambient ozone on asthmatic,
10 wheezy, and healthy children. Cambridge, MA: Health Effects Institute; research report no. 82.
11 Bernard, N. L.; Gerber, M. J.; Astre, C. M.; Saintot, M. J. (1999) Ozone measurement with passive samplers:
12 validation and use for ozone pollution assessment in Montpellier, France. Environ. Sci. Technol. 33: 217-222.
13 Black, D. R.; Harley, R. A.; Hering, S. V.; Stolzenburg, M. R. (2000) A new, portable, real-time monitor. Environ.
14 Sci. Technol. 34:3031-3040.
15 Brauer, M.; Brook, J. R. (1997) Ozone personal exposures and health effects for selected groups residing in the
16 Fraser Valley. In: Steyn, D. G.; Bottenheim, J. W., eds. The Lower Fraser Valley Oxidants/Pacific '93 Field
17 Study. Atmos. Environ. 31: 2113-2121.
18 Burke, J. M.; Zufall, M. J.; Ozkaynak, H. (2001) A population exposure model for paniculate matter: case study
19 results for PM2 5 in Philadelphia, PA. J. Exposure Anal. Environ. Epidemiol. 11: 470-489.
20 Chang, L.-T.; Koutrakis, P.; Catalano, P. J.; Suh, H. H. (2000) Hourly personal exposures to fine particles and
21 gaseous pollutants—results from Baltimore, Maryland. J. Air Waste Manage. Assoc. 50: 1223-1235.
22 Christakos, G.; Vyas, V. M. (1998a) A composite space/time approach to studying ozone distribution over eastern
23 United States. Atmos. Environ. 32: 2845-2857.
24 Christakos, G.; Vyas, V. M. (1998b) A novel method for studying population health impacts of spatiotemporal
25 ozone distribution. Soc. Sci. Med. 47: 1051-1066.
26 Code of Federal Regulations. (2000) Appendix D to part 58—Network design for state and local air monitoring
27 stations (SLAMS), national air monitoring stations (NAMS), and photochemical assessment monitoring
28 stations (PAMS). C. F. R. 40: pt. 58, app. D.
29 Cooper, O. R.; Moody, J. L. (2000) Meteorological controls on ozone at an elevated eastern United States regional
30 background monitoring site. J. Geophys. Res. [Atmos.] 105: 6855-6869.
31 Delfino, R. J.; Coate, B. D.; Zeiger, R. S.; Seltzer, J. M.; Street, D. H.; Koutrakis, P. (1996) Daily asthma severity in
32 relation to personal ozone exposure and outdoor fungal spores. Am. J. Respir. Crit. Care Med. 154: 633-641.
33 Drakou, G.; Zerefos, C.; Ziomas, I. (1995) A preliminary study on the relationship between outdoor and indoor air
34 pollution levels. Fresenius' Environ. Bull. 4: 689-694.
35 Ebel, A.; Hass, H.; Jakobs, J. H.; Laube, M.; Memmesheimer, M.; Oberreuter, A. (1991) Simulation of ozone
36 intrusion caused by a tropopause fold and cut-off low. Atmos. Environ. Part A25: 2131-2144.
37 Federal Register. (1986) Guidelines for estimating exposures. F. R. (September 24) 51: 34,042-34,054.
38 Fiore, A. M.; Jacob, D. J.; Bey, L; Yantosca, R. M.; Field, B. D.; Fusco, A. C.; Wilkinson, J. G. (2002) Background
39 ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys.
40 Res. (Atmos.) 107(D15): 10.1029/2001JD000982.
41 Fiore, A.; Jacob, D. J.; Liu, H.; Yantosca, R. M.; Fairlie, T. D.; Li, Q. (2003) Variability in surface ozone
42 background over the United States: implications for air quality policy. J. Geophys. Res. (Atmos.)
43 108(024): 10.1029/2003 JD003855.
44 Freijer, J. I.; Bloemen, H. J. T. (2000) Modeling relationships between indoor and outdoor air quality. J. Air Waste
45 Manage. Assoc. 50: 292-300.
46 Fusco, A. C.; Logan, J. A. (2003) Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
47 midlatitudes with the GEOS-CHEM model. J. Geophys. Res. (Atmos.) 108: 10.1029/2002JD002742.
48 Georgopoulos, P. G.; Arunachalam, S.; Wang, S. (1997a) Alternative metrics for assessing the relative effectiveness
49 of NOX and VOC emission reductions in controlling ground-level ozone. J. Air Waste Manage. Assoc.
50 47:838-850.
51 Georgopoulos, P. G.; Walia, A.; Roy, A.; Lioy, P. J. (1997b) Integrated exposure and dose modeling and analysis
52 system. 1. Formulation and testing of microenvironmental and pharmacokinetic components. Environ. Sci.
53 Technol. 31: 17-27.
54 Geyh, A. S.; Wolfson, J. M.; Koutrakis, P.; Mulik, J. D.; Avol, E. L. (1997) Development and evaluation of a small
55 active ozone sampler. Environ. Sci. Technol. 31: 2326-2330.
January 2005 3-69 DRAFT-DO NOT QUOTE OR CITE
-------
1 Geyh, A. S.; Roberts, P. T.; Lurmann, F. W.; Schoell, B. M.; Avol, E. L. (1999) Initial field evaluation of the
2 Harvard active ozone sampler for personal ozone monitoring. J. Exposure Anal. Environ. Epidemiol.
3 9: 143-149.
4 Geyh, A. S.; Xue, J.; Ozkaynak, H.; Spengler, J. D. (2000) The Harvard Southern California chronic ozone exposure
5 study: assessing ozone exposure of grade-school-age children in two southern California communities.
6 Environ. Health Perspect. 108: 265-270.
7 Glen, G.; Lakkadi, Y.; Tippett, J. A.; del Valle-Torres, M. (1997) Development of NERL/CHAD: the National
8 Exposure Research Laboratory consolidated human activity database. Research Triangle Park, NC: U.S.
9 Environmental Protection Agency, Office of Research and Development; contract no. 68-D5-0049.
10 Gold, D. R.; Allen, G.; Damokosh, A.; Serrano, P.; Hayes, C.; Castillejos, M. (1996) Comparison of outdoor and
11 classroom ozone exposures for school children in Mexico City. J. Air Waste Manage. Assoc. 46: 335-342.
12 Gonzales, M.; Ngo, L.; Hammond, S. K.; Tager, I. (2003) Validation of a questionnaire and microenvironmental
13 model for estimating past exposures to ozone. Int. J. Environ. Health Res. 13: 249-260.
14 Grosjean, E.; Grosjean, D. (1996) Carbonyl products of the gas phase reaction of ozone with symmetrical alkenes.
15 Environ. Sci. Technol. 30: 2036-2044.
16 Grosjean, E.; Grosjean, D. (1998) The gas-phase reaction of alkenes with ozone: formation yields of carbonyls from
17 biradicals in ozone-alkene-cyclohexane experiments. Atmos. Environ. 32: 3393-3402.
18 Hirsch, A. L; Munger, J. W.; Jacob, D. J.; Horowitz, L. W.; Goldstein, A. H. (1996) Seasonal variation of the ozone
19 production efficiency per unit NOx at Harvard Forest, Massachusetts. J. Geophys. Res. [Atmos.]
20 101: 12,659-12,666.
21 Howard-Reed, C.; Wallace, L. A.; Ott, W. R. (2002) The effect of opening windows on air change rates in two
22 homes. J. Air Waste Manage Assoc. 52: 147-159.
23 Hudman, R. C.; Jacob, D. J.; Cooper, O. C.; Evans, M. J.; Heald, C. L.; Park, R. J.; Fehsenfeld, F.; Flocke, F.;
24 Holloway, J.; Hubler, G.; Kita, K.; Koike, M.; Kondo, Y.; Neuman, A.; Nowak, J.; Oltmans, S.; Parrish, D.;
25 Roberts, J. M.; Ryerson, T. (2004) Ozone production in transpacific Asian pollution plumes and implications
26 for ozone air quality in California. J. Geophys. Res. (Atmos.) 109(D23): 10.1029/2004JD004974.
27 Jakobi, G.; Fabian, P. (1997) Indoor/outdoor concentrations of ozone and peroxyacetyl nitrate (PAN). Int. J.
28 Biometeorol. 40: 162-165.
29 Johnson, T. (1997) A pilot study in Los Angeles to measure personal ozone exposures during scripted activities.
30 Washington, DC: American Petroleum Institute, Health and Environmental Sciences Department; API
31 publication no. DR 218.
32 Johnson, T. (2003) A guide to selected algorithms, distributions, and databases used in exposure models developed
33 by the Office of Air Quality Planning and Standards. Research Triangle Park, NC: U.S. Environmental
34 Protection Agency, Office of Research and Development; EPA grant no. CR827033. Available:
3 5 http://www.epa.gov/ttn/fera/data/human/report052202.pdf [9 April, 2004].
36 Johnson, T.; Long, T. (2004) Determining the frequency of open windows in residences: a pilot study in Durham,
37 North Carolina during varying temperature conditions. J. Exposure Anal. Environ. Epidemiol.:
38 10.1038/sj.jea.7500409.
39 Johnson, T.; Capel, J.; McCoy, M. (1996a) Estimation of ozone exposures experienced by urban residents using a
40 probabilistic version of NEM and 1990 population data. Research Triangle Park, NC: U.S. Environmental
41 Protection Agency, Office of Air Quality Planning and Standards; contract no. 68-DO-0062.
42 Johnson, T.; Capel, J.; McCoy, M.; Mozier, J. W. (1996b) Estimation of ozone exposures experienced by outdoor
43 workers in nine urban areas using a probabilistic version of NEM. Research Triangle Park, NC: U.S.
44 Environmental Protection Agency, Office of Air Quality Planning and Standards; contract no. 63-D-30094,
45 work assignment nos. 0-1 and 1-4.
46 Johnson, T.; Capel, J.; Mozier, J.; McCoy, M. (1996c) Estimation of ozone exposures experienced by outdoor
47 children in nine urban areas using a probabilistic version of NEM. Research Triangle Park, NC: U.S.
48 Environmental Protection Agency, Office of Air Quality Planning and Standards; contract no. 63-D-30094.
49 Johnson, T.; Mozier, J.; Capel, J. (1997) Supplement to "Estimation of ozone exposures experienced by outdoor
50 children in nine urban areas using a probabilistic version of NEM (April 1996)". Research Triangle Park, NC:
51 U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
52 Johnson, T.; Mihlan, G.; LaPointe, J.; Fletcher, K.; Capel, J. (1999) Estimation of carbon monoxide exposures and
53 associated carboxyhemoglobin levels in Denver residents using pNEM/CO (version 2.0) [draft]. Research
54 Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards;
55 March 15.
January 2005 3-70 DRAFT-DO NOT QUOTE OR CITE
-------
1 Johnson, T.; Myers, I; Kelly, T.; Wisbith, A.; Ollison, W. (2004) A pilot study using scripted ventilation conditions
2 to identify key factors affecting indoor pollutant concentrations and air exchange rate in a residence.
3 J. Exposure Anal. Environ. Epidemiol. 14: 1-22.
4 Klepeis, N. E.; Tsang, A. M; Behar, J. V. (1996) Analysis of the national human activity pattern survey (NHAPS)
5 respondents from a standpoint of exposure assessment. Washington, DC: U.S. Environmental Protection
6 Agency, Office of Research and Development; report no. EPA/600/R-96/074.
7 Klepeis, N. E.; Nelson, W. C; Ott. W. R.; Robinson, J. P. Tsang, A. M.; Switzer, P.; Behar, J. V.; Hern, S. C;
8 Engelmann, W. H. (2001) The National Human Activity Pattern Survey (NHAPS): a resource for assessing
9 exposure to environmental pollutants. J. Exposure Anal. Environ. Epidemiol. 11: 231-252.
10 Koutrakis, P.; Sioutas, C.; Ferguson, S. T.; Wolfson, J. M.; Mulik, J. D.; Burton, R. M. (1993) Development and
11 evaluation of a glass honeycomb denuder/filter pack system to collect atmospheric gases and particles.
12 Environ. Sci. Technol. 27: 2497-2501.
13 Krzyzanowski, M. (1997) Methods for assessing the extent of exposure and effects of air pollution. Occup. Environ.
14 Med. 54: 145-151.
15 Lagus Applied Technology, Inc. (1995) Air change rates in non-residential buildings in California. Sacramento, CA:
16 California Energy Commission; contract no. 400-91-034; July.
17 Lee, E. H.; Hogsett, W. E. (1999) Role of concentrations and time of day in developing ozone exposure indices for a
18 secondary standard. J. Air Waste Manage. Assoc. 49: 669-681.
19 Lee, K.; Xue, J.; Geyh, A. S.; Ozkaynak, H.; Leaderer, B. P.; Weschler, C. J.; Spengler, J. D. (2002) Nitrous acid,
20 nitrogen dioxide, and ozone concentrations in residential environments. Environ. Health Perspect.
21 110: 145-150.
22 Lee, K.; Parkhurst, W. J.; Xue, J.; Ozkaynak, H.; Neuberg, D.; Spengler, J. D. (2004) Outdoor/indoor/personal ozone
23 exposures of children in Nashville, Tennessee. J. Air Waste Manage. Assoc. 54: 352-359.
24 Lefohn, A. S.; Runeckles, V. C. (1987) Establishing standards to protect vegetation - ozone exposure/dose
25 considerations. Atmos. Environ. 21: 561-568.
26 Lefohn, A. S.; Oltmans, S. J.; Dann, T.; Singh, H. B. (2001) Present-day variability of background ozone in the
27 lower troposphere. J. Geophys. Res. [Atmos.] 106: 9945-9958.
28 Leovic, K. W.; Sheldon, L. S.; Whitaker, D. A.; Hetes, R. G.; Calcagni, J. A.; Baskir, J. N. (1996) Measurement of
29 indoor air emissions from dry-process photocopy machines. J. Air Waste Manage. Assoc. 46: 821-829.
30 Leovic, K.; Whitaker, D.; Northeim, C.; Sheldon, L. (1998) Evaluation of a test method for measuring indoor air
31 emissions from dry-process photocopiers. J. Air Waste Manage. Assoc. 48: 915-923.
32 Liang, J.; Horowitz, L. W.; Jacob, D. J.; Wang, Y.; Fiore, A. M.; Logan, J. A.; Gardner, G. M.; Munger, J. W.
33 (1998) Seasonal budgets of reactive nitrogen species and ozone over the United States, and export fluxes to
34 the global atmosphere. J. Geophys. Res. (Atmos.) 103: 13,435-13,450.
35 Liard, R.; Zureik, M.; Le Moullec, Y.; Soussan, D.; Glorian, M.; Grimfeld, A.; Neukirch, F. (1999) Use of personal
36 passive samplers for measurement of NO2, NO, and O3 levels in panel studies. Environ. Res. 81: 339-348.
37 Lin, C.-Y.; Jacob, D. J.; Munger, J. W.; Fiore, A. M. (2000) Increasing background ozone in surface air over the
38 United States. Geophys. Res. Lett. 27: 3465-3468.
39 Linn, W. S.; Shamoo, D. A.; Hackney, J. D. (1992) Documentation of activity patterns in 'high-risk' groups exposed
40 to ozone in the Los Angeles area. In: Tropospheric ozone and the environment II: effects, modeling and
41 control: papers from an Air & Waste Management Association international specialty conference; November,
42 1991; Atlanta, GA. Pittsburgh, PA: Air & Waste Management Association; pp. 701-712. (A&WMA
43 publication TR-20).
44 Linn, W. S.; Shamoo, D. A.; Anderson, K. R.; Peng, R.-C.; Avol, E. L.; Hackney, J. D.; Gong, H., Jr. (1996)
45 Short-term air pollution exposures and responses in Los Angeles area schoolchildren. J. Exposure Anal.
46 Environ. Epidemiol. 6: 449-472.
47 Liu, L.-J. S.; Koutrakis, P.; Leech, J.; Broder, I. (1995) Assessment of ozone exposures in the greater metropolitan
48 Toronto area. J. Air Waste Manage. Assoc. 45: 223-234.
49 Liu, L.-J. S.; Delfino, R.; Koutrakis, P. (1997) Ozone exposure assessment in a southern California community.
50 Environ. Health Perspect. 105: 58-65.
51 Logan, J. A. (1989) Ozone in rural areas of the United States. J. Geophys. Res. [Atmos.] 94: 8511-8532.
52 McCurdy, T. (2000) Conceptual basis for multi-route intake dose modeling using an energy expenditure approach.
53 J. Exposure Anal. Environ. Epidemiol. 10: 86-97.
54 McCurdy, T.; Glen, G.; Smith, L.; Lakkadi, Y. (2000) The National Exposure Research Laboratory's Consolidated
55 Human Activity Database. J. Exposure Anal. Environ. Epidemiol. 10: 566-578.
January 2005 3-71 DRAFT-DO NOT QUOTE OR CITE
-------
1 Morrison, G. C.; Nazaroff, W. W. (2000) The rate of ozone uptake on carpets: experimental studies. Environ. Sci.
2 Technol. 34: 4963-4968.
3 Morrison, G. C.; Nazaroff, W. W. (2002) Ozone interactions with carpet: secondary emissions of aldehydes.
4 Environ. Sci. Technol. 36: 2185-2192.
5 Murray, D. M.; Burmaster, D. E. (1995) Residential air exchange rates in the United States: empirical and estimated
6 parametric distributions by season and climatic region. Risk Anal. 15: 459-465.
7 Nazaroff, W. W.; Cass, G. R. (1986) Mathematical modeling of chemically reactive pollutants in indoor air.
8 Environ. Sci. Technol. 20: 924-934.
9 Nazaroff, W. W.; Weschler, C. J. (2004) Cleaning products and air fresheners: exposure to primary and secondary
10 air pollutants. Atmos. Environ. 38: 2841-2865.
11 Niu, J.; Tung, T. C. W.; Burnett, J. (2001) Ozone emission rate testing and ranking method using environmental
12 chamber. Atmos. Environ. 35: 2143-2151.
13 Northeast States for Coordinated Air Use Management (NESCAUM). (2002) Indoor/outdoor school air monitoring
14 project. Boston, MA. Available: http://www.nescaum.org/pdf/schoolmonitoring.pdf [29 October, 2003].
15 O'Neill, M. S.; Ramirez-Aguilar, M.; Meneses-Gonzalez, F.; Hernandez-Avila, M.; Geyh, A. S.; Sienra-Monge, J. J.;
16 Romieu, I. (2003) Ozone exposure among Mexico City outdoor workers. J. Air Waste Manage. Assoc.
17 53: 339-346.
18 Ott, W. R. (1982) Concepts of human exposure to air pollution. Environ. Int. 7: 179-196.
19 Ott, W. R. (1985) Total human exposure: an emerging science focuses on humans as receptors of environmental
20 pollution. Environ. Sci. Technol. 19: 880-886.
21 Park, J.-H.; Spengler, J. D.; Yoon, D.-W.; Dumyahn, T.; Lee, K.; Ozkaynak, H. (1998) Measurement of air
22 exchange rate of stationary vehicles and estimation of in-vehicle exposure. J. Exposure Anal. Environ.
23 Epidemiol. 8: 65-78.
24 Reiss, R.; Ryan, P. B.; Koutrakis, P.; Tibbetts, S. J. (1995) Ozone reactive chemistry on interior latex paint. Environ.
25 Sci. Technol. 29: 1906-1912.
26 Riediker, M.; Williams, R.; Devlin, R.; Griggs, T.; Bromberg, P. (2003) Exposure to paniculate matter, volatile
27 organic compounds, and other air pollutants inside patrol cars. Environ. Sci. Technol. 37: 2084-2093.
28 Roelofs, G. J.; Scheeren, H. A.; Heland, J.; Ziereis, H.; Lelieveld, J. (2003) A model study of ozone in the eastern
29 Mediterranean free troposphere during MINOS (August 2001). Atmos. Chem. Phys. 3: 1199-1210.
30 Romieu, I.; Lugo, M. C.; Colome, S.; Garcia A. M.; Avila, M. H.; Geyh, A.; Velasco, S. R.; Rendon, E. P. (1998)
31 Evaluation of indoor ozone concentration and predictors of indoor-outdoor ratio in Mexico City. J. Air Waste
32 Manage. Assoc. 48: 327-335.
33 Salmon, L. G.; Cass, G. R.; Bruckman, K.; Haber, J. (2000) Ozone exposure inside museums in the historic central
34 district of Krakow, Poland. Atmos. Environ. 34: 3823-3832.
35 Sarnat, J. A.; Koutrakis, P.; Suh, H. H. (2000) Assessing the relationship between personal paniculate and gaseous
36 exposures of senior citizens living in Baltimore, MD. J. Air Waste Manage. Assoc. 50: 1184-1198.
37 Sarnat, J. A.; Schwartz, J.; Catalano, P. J.; Suh, H. H. (2001) Gaseous pollutants in paniculate matter epidemiology:
38 confounders or surrogates? Environ. HealthPerspect. 109: 1053-1061.
39 Sarwar, M.; Corsi, R.; Kimura, Y.; Allen, D.; Weschler, C. (2001) Hydroxyl radicals in indoor environments. In:
40 Proceedings of the Air & Waste Management Association's 94th Annual Conference & Exhibition; June;
41 Orlando, FL. Pittsburgh, PA: Air & Waste Management Association.
42 Sarwar, G.; Corsi, R.; Kumura, Y.; Allen, D.; Weschler, C. J. (2002) Hydroxyl radicals in indoor environments.
43 Atmos. Environ. 36: 3973-3988.
44 Serensen, D. N.; Weschler, C. J. (2002) Modeling-gas phase reactions in indoor environments using computational
45 fluid dynamics. Atmos. Environ. 36: 9-18.
46 Spier, C. E.; Little, D. E.; Trim, S. C.; Johnson, T. R.; Linn, W. S.; Hackney, J. D. (1992) Activity patterns in
47 elementary and high school students exposed to oxidant pollution. J. Exposure Anal. Environ. Epidemiol.
48 2:277-293.
49 Steiber, R. S. (1995) Ozone generators in indoor air settings. Research Triangle Park, NC: U.S. Environmental
50 Protection Agency, National Risk Management Research Laboratory; report no. EPA-600/R-95-154.
51 Available from: NTIS, Springfield, VA; PB96-100201.
52 Trainer, M.; Parrish, D. D.; Buhr, M. P.; Norton, R. B.; Fehsenfeld, F. C.; Anlauf, K. G.; Bottenheim, J. W.; Tang,
53 Y. Z.; Wiebe, H. A.; Roberts, J. M.; Tanner, R. L.; Newman, L.; Bowersox, V. C.; Meagher, J. F.; Olszyna,
54 K. J.; Rodgers, M. O.; Wang, T.; Berresheim, H.; Demerjian, K. L.; Roychowdhury, U. K. (1993) Correlation
55 of ozone with NOy in photochemically aged air. J. Geophys. Res. [Atmos.] 98: 2917-2925.
January 2005 3-72 DRAFT-DO NOT QUOTE OR CITE
-------
1 Turk, B. H.; Grimsrud, D. T.; Brown, J. T.; Geisling-Sobotka, K. L.; Harrison, J.; Prill, R. J. (1989) Commercial
2 building ventilation rates and particle concentrations. ASHRAE Trans. 95(part 1): 422-433.
3 U.S. Environmental Protection Agency. (1978) Air quality criteria for ozone and other photochemical oxidants.
4 Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
5 Assessment Office; report no. EPA-600/8-78-004. Available from: NTIS, Springfield, VA; PB80-124753.
6 U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
7 Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
8 Assessment Office; report nos. EPA-600/8-84-020aF-eF. 5v. Available from: NTIS, Springfield, VA;
9 PB87-142949.
10 U.S. Environmental Protection Agency. (1992) Guidelines for exposure assessment. Washington, DC: Risk
11 Assessment Forum, USEPA 600Z-92/001.
12 U.S. Environmental Protection Agency. (1996) Air quality criteria for ozone and related photochemical oxidants.
13 Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF. 3v.
14 Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, and PB96-185608. Available:
15 http://cfpub2.epa.gov/ncea/.
16 U.S. Environmental Protection Agency. (2003a) Technology Transfer Network: Air Quality System (AQS).
17 Washington, DC: Office of Air and Radiation. Available: http://www.epa.gov/ttn/airs/airsaqs/ [1 April, 2004].
18 U.S. Environmental Protection Agency. (2003b) Latest findings on national air quality: 2002 status and trends.
19 Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA/454/K-03-001.
20 Available: http://www.epa.gov/airtrends/2002_airtrends_fmal.pdf [18 December, 2003]. Available: NTIS,
21 Springfield, VA; PB2004-100183.
22 Vakeva, M; Hameri, K.; Kulmala, M; Lahdes, R.; Ruuskanen, J.; Laitinen, T. (1999) Street level versus rooftop
23 concentrations of submicron aerosol particles and gaseous pollutants in an urban street canyon. Atmos.
24 Environ. 33: 1385-1397.
25 Wainman, T.; Zhang, J.; Weschler, C. J.; Lioy, P. J. (2000) Ozone and limonene in indoor air: a source of submicron
26 particle exposure. Environ. Health Perspect. 108: 1139-1145.
27 Weschler, C. J. (2000) Ozone in indoor environments: concentration and chemistry. Indoor Air 10: 269-288.
28 Weschler, C. J.; Shields, H. C. (1997) Potential reactions among indoor pollutants. Atmos. Environ. 31: 3487-3495.
29 Weschler, C. J.; Shields, H. C. (1999) Indoor ozone/terpene reactions as a source of indoor particles. Atmos.
30 Environ. 33:2301-2312.
31 Weschler, C. J.; Shields, H. C.; Naik, D. V (1989) Indoor ozone exposures. JAPCA 39: 1562-1568.
32 Weschler, C. J.; Shields, H. C.; Naik, D. V. (1994) Indoor chemistry involving O3, NO, and NO2 as evidenced by
33 14 months of measurements at a site in southern California. Environ. Sci. Technol. 28: 2120-2132.
34 Zanis, P.; Trickl, T.; Stohl, A.; Wernli, H.; Cooper, O.; Zerefos, C.; Gaeggeler, H.; Schnabel, C.; Tobler, L.; Kubik,
35 P. W.; Priller, A.; Scheel, H. E.; Kanter, H. J.; Cristofanelli, P.; Forster, C.; James, P.; Gerasopoulos, E.;
36 Delcloo, A.; Papayannis, A.; Claude, H. (2003) Forecast, observation and modelling of a deep stratospheric
37 intrusion event over Europe. Atmos. Chem. Phys. 3: 763-777.
38 Zeger, S. L.; Thomas, D.; Dominici, F.; Samet, J. M.; Schwartz, J.; Dockery, D.; Cohen, A. (2000) Exposure
3 9 measurement error in time-series studies of air pollution: concepts and consequences. Environ. Health
40 Perspect. 108: 419-426.
41
January 2005 3-73 DRAFT-DO NOT QUOTE OR CITE
-------
i ANNEX AX3. ENVIRONMENTAL CONCENTRATIONS,
2 PATTERNS, AND EXPOSURE ESTIMATES
3
4
5 AX3.1 INTRODUCTION
6 Identification and Use of Existing Air Quality Data
1 The effects of ozone (O3) on humans, animals, and vegetation have received extensive
8 examination and are discussed in detail subsequent annexes. As indicated in the previous
9 document, Air Quality Criteria for Ozone and Related Oxidants (1996 O3 AQCD; U.S.
10 Environmental Protection Agency, 1996a), most of the human and welfare effects research has
11 focused on evaluating impacts on health or vegetation from exposure to O3 that simulate ambient
12 O3 exposures (e.g., matching the occurrence of hourly average concentrations or more prolonged
13 times of exposure). The information contained within this annex on concentrations was obtained
14 from extensive monitoring in the United States can be useful both for linking anthropogenic
15 emissions of O3 precursors with health and welfare effects and for augmenting exposure
16 assessment and epidemiology studies.
17 The information obtained for this annex is primarily available from a database that has
18 been designed to archive air quality data from a network of monitoring stations that have been
19 established to determine compliance with the National Ambient Air Quality Standards (NAAQS)
20 defined in 40 CFR 50. Air quality monitoring is conducted continuously through the National
21 Air Monitoring Network, which consists of monitoring stations defined in 40 CFR 58.20. State
22 and Local Ambient Monitoring Stations (SLAMS) are located throughout the United States.
23 National Ambient Monitoring Stations (NAMS) are located in areas where possible human
24 exposure to air contaminants could potentially be a risk. NAMS are used to provide data for
25 national policy and trend analysis and to give the public information about air quality in major
26 metropolitan areas. These monitoring stations are required in urban areas with populations
27 greater than 200,000 and are selected from a subset of the State and Local Air Monitoring
28 Stations (SLAMS) network. The data from all of these sites are gathered and stored in the U.S.
29 Environmental Protection Agency Air Quality System (AQS; formerly the AIRS database).
30 These available data were collected from 1979 to 2001. As discussed in the previous versions of
31 the O3 AQCD (U.S. Environmental Protection Agency, 1986, 1996a), the data available prior to
January 2005 AX3-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 1979 may be considered unreliable due to calibration problems and uncertainties. The AQS
2 contains readily accessible detailed, hourly data that has been subject to the Agency's quality
3 control and assurance procedures. The information has been summarized in a series of tables
4 and graphs to better describe the current O3 situation. Most of the field data were collected by
5 the various state agencies and O3 working groups for regulation and enforcement of O3 levels,
6 but the information may also be used to determine trends and patterns for O3 surface
7 concentrations. In the sections that follow, the hourly averaged ambient O3 data have been
8 summarized in different ways to reflect the interests of those who wish to know more about the
9 potential for O3 exposure of humans and the environment. This annex is not meant to be an
10 exposure assessment for ambient O3; rather, this annex elucidates the features of O3
11 concentration patterns and exposure possibilities.
12 As indicated in the previous O3 AQCD (U.S. Environmental Protection Agency, 1996a), O3
13 is the only photochemical oxidant, other than nitrogen dioxide (NO2), that is routinely monitored
14 and for which a comprehensive aerometric database exists. Data for peroxyacetyl nitrate (PAN)
15 and hydrogen peroxide (H2O2) have been obtained only as part of special research field
16 investigations. Consequently, no data on nationwide patterns of occurrence are available for
17 these non-O3 oxidants; nor are extensive data available on the correlations of levels and patterns
18 of these oxidants with those of O3.
19
20 Characterizing Ambient Ozone Concentrations
21 In this annex, data are analyzed for the purpose of providing focus on specific issues of
22 exposure-response relationships that are considered in the later annexes addressing O3 exposure
23 effects. It is important to distinguish among concentration, exposure, and dose when using air
24 quality data to assess human health and vegetation effects. For this annex, the following
25 definitions apply:
26 (1) The "concentration" of a specific air pollutant is typically defined as the amount
(mass) of that material per unit volume of air. Air pollution monitors measure
pollutant concentrations, which may or may not provide accurate exposure
estimates. However, most of the data presented herein are expressed as a "mixing
ratio" in terms of a volume-to-volume ratio expressed as parts per million (ppm)
or parts per billion (ppb).
January 2005 AX3-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 (2) The term "exposure" is defined as the concentration of a pollutant encountered by
the subject (animal, human, or plant) for a duration of time. Exposure implies that
such an encounter leads to intake (i.e., through the respiratory tract or stomata).
It is expressed in terms of ppm x time.
2 (3) The term "dose" is defined as that mass of pollutant delivered to a target. This
term has numerous quantitative descriptions (e.g., micrograms of O3 per square
centimeter of lung epithelium per minute), so the context of the use of this term
and the units within the document must be considered.
3 The dose incurred by an organism (e.g., plant, animal, or human) is a more complicated measure
4 involving the concentration, the exposure duration, and the volume of delivered air (e.g., through
5 inhalation). These distinctions become important because the concentration of an airborne
6 contaminant that is measured in an empty room or at a stationary outdoor monitor is not in fact
7 an exposure. A measured concentration functions as a surrogate for an exposure only to the
8 degree to which it represents concentrations actually experienced by individuals under otherwise
9 similar conditions.
10 Concentrations of airborne contaminants for vegetation are considered to represent an
11 exposure when a plant is subjected to them over a specified time period. As indicated in
12 Chapter 9, dose has been defined historically by air pollution vegetation researchers as ambient
13 air quality concentration multiplied by time (O'Gara, 1922). However, a more rigorous
14 definition was required. Runeckles (1974) introduced the concept of "effective dose" as the
15 amount or concentration of pollutant that is adsorbed by vegetation, in contrast to that which is
16 present in the ambient air. Fowler and Cape (1982) developed this concept further and proposed
17 that the "pollutant adsorbed dose" be defined in units of grams per square meter (of ground or
18 leaf area) and could be obtained as the product of concentration, time, and stomatal (or canopy)
19 conductance for the gas in question. Taylor et al. (1982) suggested internal flux (milligrams per
20 square meter per hour) as a measure of the dose to which plants respond. In this annex, dose will
21 be taken to signify, for the purposes of vegetation, that amount of pollutant absorbed by the
22 plant.
23 Although parallel mathematical definitions to quantify exposure and dose have been
24 employed in animal and human health effects research, the equivalent biologic effects between
25 species and plants induced by a specific dose is at present difficult to define. In order to
26 characterize the specific doses responsible for affecting human health and vegetation, there has
January 2005 AX3-3 DRAFT-DO NOT QUOTE OR CITE
-------
1 to be a linkage between exposure and actual dose. Unfortunately, it is difficult to predict this
2 relationship, even with the available models. For example, the sensitivity of vegetation to O3 as
3 a function of time of day, period of growth, or edaphic conditions can determine the severity of
4 response (see Chapter 9). Similarly, in animals and humans, additional factors such as the state
5 of the organism's susceptibility, physical activity, demographic characteristics will substantially
6 modulate the intensity and persistence of response. Because not enough is known to quantify the
7 links between exposure, dosage, and vegetation and human effects, routine monitoring for O3 is
8 summarized as hourly average concentrations. Whenever the information is available the
9 exposure or dose will be reported as well. Most of the information provided in this annex is
10 characterized in terms of concentration and exposure.
11 For many years, as indicated in Chapter 9, air pollution specialists have explored
12 alternative mathematical approaches for summarizing ambient air quality information in
13 biologically meaningful forms that can serve as alternatives for characterizing dose. Extensive
14 research has focused on identifying indicators of concentration and duration (exposure) that are
15 firmly founded on biological principles for vegetation. Many of these indicators have been
16 based on research results indicating that the magnitude of vegetation responses to air pollution is
17 determined more as a function of the magnitude of the concentration than of the length of the
18 exposure (U.S. Environmental Protection Agency, 1996a). Short-term (1-h), high
19 O3 concentrations (> 0.1 ppm) have been identified by many researchers as being more
20 important than long-term, low O3 concentrations in inducting visible injury and damage to
21 vegetation (see Chapter 9 for further discussion).
22 Long-term, average concentrations (e.g., 7-h seasonal daylight average concentrations)
23 were used initially as an exposure indicator to describe O3 concentrations over time when
24 assessing vegetation effects (Heck et al., 1982). The 1996 O3 AQCD (U.S. Environmental
25 Protection Agency, 1996a) concluded that higher concentrations of O3 should be given more
26 weight than lower concentrations (see Chapter 9 for further details). Similar observations on the
27 importance of peak concentration on the magnitude of response have been reported in human
28 studies (Hazucha et al., 1992; Adams, 2003) (see Chapter 6; Annex 6, Section AX6.2.4 for
29 further discussion). They reported that a triangular pattern of exposure elicited substantially
30 greater functional response than the same O3 dose with an even concentration profile. These
31 findings suggest that a similar approach as employed in assessing damage to vegetation should
January 2005 AX3-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 be used in human studies (i.e., that higher concentration should be given more weight in deriving
2 exposure dose).
3 As outlined in the 1996 O3 AQCD, several investigators have suggested additional
4 representations other than strict concentration analysis for assessing O3 impact on human health
5 and other welfare issues (see Chapter 9 for information concerning the use of "effective flux" as
6 a way to predict potential vegetation effects). Many of these methodologies examine the
7 relationships between concentration, exposure, and dose.
8 In summarizing the hourly average concentrations in this annex, specific attention is given
9 to the relevance of the exposure indicators used. For example, for human health considerations,
10 concentration (or exposure) indicators such as the daily maximum 1-h average concentrations, as
11 well as the number of daily maximum 8-h average concentrations, are used to characterize
12 information in the population-oriented locations. For vegetation, several different types of
13 exposure indicators are used. Several exposure indicators that use either a threshold or a
14 sigmoidal weighting scheme are used in this annex to provide insight concerning the
15 O3 exposures that are experienced at a select number of rural monitoring sites in the United
16 States. The peak-weighted, cumulative-exposure indicators used in this annex are SUM06 and
17 SUM08 (the sums of all hourly average concentrations > 0.06 and 0.08 ppm, respectively) and
18 W126 (the sum of the hourly average concentrations that have been weighted according to a
19 sigmoid function [see Lefohn and Runeckles, 1987] that is based on a hypothetical vegetation
20 response). Further discussion of these exposure indices is presented in Chapter 9.
21 The exposure indicators used for human health considerations are in mixing ratios (i.e.,
22 parts per million), whereas the indicators used for vegetation (e.g., SUM06, SUM08, W126) are
23 in parts per million-hour (ppm-h). Although ppm are mixing ratios, they are commonly referred
24 to as concentrations and will be referred to as concentrations in this chapter. The magnitude of
25 the peak-weighted, cumulative indicators at specific sites can be compared with those values
26 experienced at areas that experience low hourly average maximum concentrations. In some
27 cases, to provide more detailed information about the distribution patterns for a specific
28 O3 exposure regime, the percentile distribution of the hourly average concentrations (in ppm) is
29 given. For further clarification of the determination and rationale for the exposure indicators that
30 are used for assessing human health and vegetation effects, the reader is encouraged to read
31 Section AX3.10 in this annex as well as Chapter 9.
January 2005 AX3-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 To obtain a better understanding of the potential effect of ambient O3 concentrations on
2 human health and vegetation, hourly average concentration information was summarized for
3 urban versus rural (forested and agricultural) areas in the United States. A land use
4 characterization of rural does not imply that any specific location is isolated from anthropogenic
5 influences. For example, Logan (1989) has noted that hourly average O3 concentrations above
6 0.08 ppm are common in rural areas of the eastern United States in spring and summer, but are
7 unusual at remote western sites. Consequently, for the purposes of comparing exposure regimes
8 that may be characteristic of clean locations in the United States with those that are urban
9 influenced (i.e., located in either urban or rural locations), this annex characterizes data collected
10 from those stations whose locations appear to be isolated from large-scale anthropogenic
11 influences.
12 Long-term (multiyear) patterns and trends are available only from stationary ambient
13 monitors; data on indoor concentrations are collected predominantly in selected settings during
14 comparatively short-term studies. Data from the indoor and outdoor environments are reviewed
15 here separately.
16 Ambient air means the portion of the atmosphere, external to buildings, to which the
17 general public has access (CFR, 2000a). The 1-h and/or the 8-h O3 ambient air quality standards
18 are respectively met at an air quality monitoring site when the second highest daily maximum
19 1-h average O3 concentration is < 0.12 ppm and/or the average of the annual fourth-highest daily
20 maximum 8-h average O3 concentration is < 0.08 ppm, as determined in accordance with
21 Appendix I to 40 CFR Part 50 (CFR, 2000a).
22 Acknowledging the photochemical and insulation-dependent nature of O3 formation, the
23 U.S. Environmental Protection Agency (U.S. EPA) has established allowable "ozone seasons"
24 for the required measurement of ambient O3 concentrations for different locations within the
25 United States and U.S. territories (CFR, 2000b). Table AX3-1 shows the summarized O3
26 seasons during which continuous, hourly averaged O3 concentrations must be monitored.
27 In Section AX3.2, surface O3 concentrations are characterized and the difficulties of
28 characterizing background O3 concentrations for controlled exposure studies and for assessing
29 the health benefits associated with setting the NAAQS are discussed. In addition, hourly
30 averaged concentrations obtained by several monitoring networks have been characterized for
31 urban and rural areas. The diurnal variations for the various urban and rural locations are found
January 2005 AX3-6 DRAFT-DO NOT QUOTE OR CITE
-------
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(2000b).
January 2005
AX3-7
DRAFT-DO NOT QUOTE OR CITE
-------
1 in Section AX3.3, where urban and rural patterns are described. In Section AX3.4, seasonal
2 patterns of the 1-h and 8-h average concentrations are discussed. Spatial variations that occur
3 within urban areas, between rural and urban areas, as well as variations with elevation are
4 discussed in Section AX3.5. Section AX3.6 of this annex summarizes the historical trends for
5 1980 to 2001 on a national scale and for selected cities. The most recent U.S. EPA trends results
6 are also presented. Section AX3.7 describes available information for the concentrations and
7 patterns of related photochemical oxidants. Section AX3.8 describes the co-occurrence patterns
8 of O3 with NO2, sulfur dioxide (SO2), and 24-h PM25. Indoor O3 concentrations, including
9 sources and factors affecting indoor O3 concentrations, are described in Section AX3.9. Section
10 AX3.10 describes human population exposure measurement methods, factors influencing
11 exposure, and exposure models.
12
13
14 AX3.2 SURFACE OZONE CONCENTRATIONS
15 Data for O3 concentrations in a number of different environments, ranging from urban to
16 remote, are summarized and characterized in this section. The characterization of the variability
17 of O3 concentrations in these different environments receives the main emphasis. Another
18 important issue relates to the determination of background concentrations. There are a number
19 of different uses of the term background depending on the context in which it is used. Various
20 definitions of background have been covered in the 1996 O3 AQCD (U.S. Environmental
21 Protection Agency, 1996a) and in Air Quality Criteria for Particulate Matter (PM AQCD; U.S.
22 Environmental Protection Agency, 1996b). This section deals with the characterization of
23 background O3 concentrations that are used for two main purposes: (1) performing experiments
24 relating the effects of exposure to O3 on humans, animals, and vegetation; and (2) assessing the
25 health benefits associated with setting different levels of the NAAQS for O3. Ozone background
26 concentrations used for NAAQS setting purposes are referred to as policy relevant background
27 (PRB) concentrations. PRB concentrations are defined by the U.S. EPA Office of Air Quality
28 Programs and Standards (OAQPS) as those concentrations that would be observed in the United
29 States if anthropogenic sources of O3 precursors were turned off in continental North America
30 (the United States, Canada and Mexico), i.e., the definition includes O3 formed from natural
31 sources everywhere in the world and from anthropogenic O3 precursors outside of North
January 2005 AX3-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 America. The 1996 O3 AQCD considered two possible methods for quantifying background O3
2 concentrations for the two purposes mentioned above. The first method was to estimate, using
3 mathematical models and historical data, unpolluted and natural background levels not
4 susceptible to human influence. The second method was to use the distribution of hourly
5 average O3 concentrations observed at clean, relatively remote monitoring sites (RRMS) in the
6 United States (i.e., those which experience low maximum hourly concentrations). Because of a
7 lack of simulations by available numerical models, the second method was employed in the 1996
8 O3 AQCD.
9 Sections AX3.2.1 and AX3.2.2 review data for O3 concentrations in urban and nonurban
10 (but influenced by urban emissions) environments. Section AX3.2.3 reviews the data from
11 relatively clean remote sites, addresses the issue of how to use these data to help set background
12 levels for controlled exposure studies, and presents evidence of trends in O3 concentrations at
13 these sites. The characterization of PRB O3 concentrations will be the subject of Section
14 AX3.2.4. Two alternative approaches for establishing PRB concentrations are presented: the
15 first uses data from relatively clean, remote monitoring sites and the second uses numerical
16 models. The strengths and weaknesses of each approach are presented in the hopes of
17 stimulating discussion that will resolve issues related to the use of either of these alternative
18 methods.
19
20 AX3.2.1 Urban Area Concentrations
21 Often there is a difference in the distribution of hourly average concentrations between
22 urban and nonurban areas. For example, it is possible for urban emissions, as well as O3
23 produced from urban area emissions, to be transported to more rural locations downwind. This
24 can result in elevated O3 concentrations at considerable distances from urban centers (U.S.
25 Environmental Protection Agency, 1996a). Urban O3 concentrations often are depressed because
26 of titrationby NOX (Stasiuk and Coffey, 1974). The phenomenon, where O3 concentrations
27 measured at center-city sites are lower than some rural locations, was reported by Reagan (1984)
28 and Lefohn et al. (1987). Because of the absence of chemical scavenging, O3 tends to persist
29 longer in nonurban than in urban areas (U.S. Environmental Protection Agency, 1996a).
30 Table AX3-2 summarizes the percentile distribution for the hourly average concentrations
31 at specific monitoring sites for the period of April 1999 through October 2001. As can be seen
January 2005 AX3-9 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-2
3
to
o
o
4
AQS*
Georgia
131210055
131210055
131210055
flMl\s .fl^Vfc/ ^*
Name
Fulton Co.
Fulton Co.
Fulton Co.
kjuiiiiiiLFiivcin.i
-------
3
to
o
o
X
OJ
^
O
H
6
o
0
H
O
O
0
o
H
W
i auii; r^^v^-^ ^vuiii uy* k^uiiiiiiLFIIVCIH.I auuiia ^|j|jiiiy ILFI me ,n|ji 11 I.LF \_*VHJILFCI i ci ILFU
Percentiles
AQS* Name Area Year Min. 10 30 50 70 90 95 99 Max.
Louisiana
220331001 E. Baton Rouge Baton Rouge 1999 0.000 0.006 0.018 0.032 0.047 0.069 0.079 0.101 0.128
220331001 E. Baton Rouge Baton Rouge 2000 0.000 0.007 0.018 0.030 0.045 0.066 0.075 0.092 0.151
220331001 E. Baton Rouge Baton Rouge 2001 0.000 0.004 0.014 0.024 0.037 0.056 0.065 0.082 0.101
Maryland
240251001 Edgewood DC 1999 0.000 0.000 0.016 0.030 0.042 0.064 0.076 0.102 0.156
240251001 Edgewood DC 2000 0.000 0.002 0.017 0.028 0.040 0.060 0.070 0.092 0.125
240251001 Edgewood DC 2001 0.000 0.002 0.020 0.034 0.047 0.068 0.080 0.107 0.156
Rhode Island
440030002 Kent Co. Providence 1999 0.000 0.005 0.018 0.029 0.039 0.055 0.068 0.093 0.131
440030002 Kent Co. Providence 2000 0.000 0.004 0.018 0.030 0.039 0.056 0.063 0.089 0.121
440030002 Kent Co. Providence 2001 0.000 0.007 0.023 0.035 0.045 0.061 0.075 0.103 0.136
Wisconsin
550590019 Kenosha Chicago 1999 0.002 0.012 0.029 0.039 0.049 0.068 0.079 0.100 0.126
550590019 Kenosha Chicago 2000 0.002 0.014 0.025 0.033 0.042 0.056 0.064 0.081 0.118
550590019 Kenosha Chicago 2001 0.002 0.013 0.028 0.037 0.047 0.065 0.076 0.098 0.134
Number of
Observations
4932
5034
5000
5067
5102
5106
3718
3979
3902
4331
4193
4305
* Formerly the AIRS database.
-------
1 from inspection of Table AX3-2, there are relatively few occurrences of high 1-h average O3
2 concentrations (e.g., hourly average concentrations > 0.120 ppm). For example, maximum
3 hourly average O3 monitoring sites at Charlotte, NC; Stratford, CT; Baton Rouge, LA;
4 Providence, RI; and Chicago, IL experience concentrations above 0.120 ppm. However,
5 approximately only 1% of the hourly average concentrations generally exceed 0.100 ppm at
6 these sites. Monitoring sites in polluted areas such as these also tend to experience frequent
7 hourly average O3 concentrations at or near minimum detectable levels. Table AX3-2 illustrates
8 that, for several of the sites listed, except for Atlanta, District of Columbia, Los Angeles, and
9 Houston, high hourly average concentrations (e.g., > 0.120 ppm) occur less than 1% of the time
10 and therefore, as indicated above, are associated with occasional episodes.
11 The second highest daily maximum 1-h O3 concentrations observed in U.S. Metropolitan
12 Statistical Areas (MSAs) and in Consolidated Metropolitan Statistical Areas (CMSAs) for the
13 years 1999 to 2001 are summarized in Table AX3-3 and shown graphically in Figure AX3-1.
14 MSAs and CMSAs represent politically defined units and often do not reflect the influence of
15 physical or other features on the distribution of air pollutants. As an example, the Los Angeles
16 MSA includes a monitoring site in Lancaster, CA, which is located on the other side of the San
17 Gabriel Mountains from the main urban area. Annual mean concentrations of PM25 at this site
18 were approximately half of those in the Los Angeles basin during 1999 and 2000 (Pinto et al.,
19 2004). Large gradients in pollutant concentrations exist due to the presence of topographic
20 barriers or simply because of the large area of the MSA (e.g., San Bernadino County) with
21 attendant large gradients in population densities. As a result, when monitors are pooled together
22 from disparate areas or when averages over several urban monitors are taken to represent
23 conditions at remote sites, the potential for exposure misclassification exists. The highest values
24 of the second highest daily maximum O3 concentrations are observed in the Texas Gulf Coast
25 and Southern California, but high levels of O3 also occur in the Northeast Corridor and other
26 heavily populated regions of the United States.
27 Table AX3-4 illustrates the percentiles of hourly average concentrations and the fourth
28 highest daily maximum 8-h average concentration for the April to October period for specific
29 monitoring sites that experience 3-year averages of the fourth highest 8-h concentration between
30 0.080 and 0.085 ppm. Note that year-to-year variation of the fourth highest 8-h daily maximum
January 2005 AX3-12 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3. The Second Highest Daily Maximum One-Hour Ozone Concentration
(ppm) by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan Statistical
Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Albany-Schenectady-Troy, NY MSA
Albuquerque, NM MSA
Allentown-Bethlehem-Easton, PA MSA
Altoona, PA MSA
Appleton-Oshkosh-Neenah, WI MSA
Asheville, NC MSA
Atlanta, GA MSA
Augusta-Aiken, GA-SC MSA
Austin-San Marcos, TX MSA
Bakersfield, CA MSA
Bangor, ME MSA
Barnstable-Yarmouth, MA MSA
Baton Rouge, LA MSA
Beaumont-Port Arthur, TX MSA
Bellingham, WA MSA
Benton Harbor, MI MSA
Biloxi-Gulfport-Pascagoula, MS MSA
Birmingham, AL MSA
Boston- Worcester-Lawrence, MA-NH-ME-CT CMSA
Brownsville-Harlingen-San Benito, TX MSA
Buffalo-Niagara Falls, NY MSA
Burlington, VT MSA
Canton-Massillon, OH MSA
Cedar Rapids, IA MSA
Champaign-Urbana, IL MSA
Charleston, WV MSA
Charleston-North Charleston, SC MSA
Charlotte-Gastonia-Rock Hill, NC-SC MSA
1999
0.107
0.097
0.126
0.111
0.105
0.099
0.156
0.108
0.110
0.136
0.088
0.127
0.121
0.103
0.062
0.107
0.107
0.131
0.125
0.075
0.102
0.093
0.108
0.096
0.108
0.130
0.101
0.130
2000
0.088
0.093
0.114
0.104
0.085
0.107
0.158
0.115
0.107
0.142
0.107
0.139
0.160
0.063
0.107
0.135
0.127
0.101
0.080
0.105
0.080
0.104
0.083
0.084
0.094
0.105
0.141
2001
0.112
0.088
0.126
0.107
0.102
0.091
0.125
0.103
0.097
0.134
0.105
0.139
0.117
0.103
0.061
0.117
0.099
0.114
0.136
0.074
0.116
0.083
0.106
0.084
0.080
0.107
0.085
0.142
January 2005 AX3-13 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Chattanooga, TN-GA MSA
Chicago-Gary-Kenosha, IL-IN-WI CMSA
Chico-Paradise, CA MSA
Cincinnati-Hamilton, OH-KY-IN CMSA
Clarksville-Hopkinsville, TN-KYMSA
Cleveland-Akron, OH CMSA
Colorado Springs, CO MSA
Columbia, SC MSA
Columbus, GA-AL MSA
Columbus, OH MSA
Corpus Christi, TX MSA
Dallas-Fort Worth, TX CMSA
Davenport-Moline-Rock Island, IA-IL MSA
Dayton-Springfield, OH MSA
Daytona Beach, FL MSA
Decatur, AL MSA
Decatur, IL MSA
Denver-Boulder-Greeley, CO CMSA
Des Moines, IA MSA
Detroit-Ann Arbor-Flint, MI CMSA
Dover, DE MSA
Duluth-Superior, MN-WI MSA
El Paso, TX MSA
Elkhart-Goshen, IN MSA
Elmira, NY MSA
Erie, PA MSA
Eugene-Springfield, OR MSA
Evansville-Henderson, IN-KY MSA
1999
0.122
0.126
0.110
0.119
0.115
0.118
0.075
0.117
0.110
0.144
0.103
0.154
0.099
0.127
0.087
0.103
0.102
0.105
0.083
0.121
0.120
0.082
0.108
0.085
0.092
0.112
0.084
0.114
2000
0.124
0.102
0.091
0.110
0.108
0.110
0.088
0.116
0.114
0.117
0.099
0.126
0.089
0.106
0.088
0.110
0.092
0.107
0.082
0.102
0.126
0.074
0.122
0.080
0.089
0.095
0.078
0.097
2001
0.107
0.124
0.100
0.115
0.096
0.118
0.085
0.107
0.090
0.111
0.092
0.137
0.089
0.100
0.085
0.087
0.078
0.105
0.069
0.122
0.117
0.071
0.116
0.066
0.094
0.104
0.081
0.095
January 2005
AX3-14
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Fargo-Moorhead, ND-MN MSA
Fayetteville, NC MSA
Flagstaff, AZ-UT MSA
Fort Collins-Loveland, CO MSA
Fort Myers-Cape Coral, FL MSA
Fort Fierce-Port St. Lucie, FL MSA
Fort Wayne, IN MSA
Fresno, CA MSA
Gainesville, FL MSA
Grand Rapids-Muskegon-Holland, MI MSA
Green Bay, WI MSA
Greensboro-Winston-Salem-High Point, NC MSA
Greenville, NC MSA
Greenville-Spartanburg-Anderson, SC MSA
Harrisburg-Lebanon-Carlisle, PA MSA
Hartford, CT MSA
Hickory -Morganton-Lenoir, NC MSA
Honolulu, HI MSA
Houma, LA MSA
Houston-Galveston-Brazoria, TX CMSA
Huntington- Ashland, WV-KY-OH MSA
Huntsville, AL MSA
Indianapolis, IN MSA
Jackson, MS MSA
Jacksonville, FL MSA
Jamestown, NY MSA
Janesville-Beloit, WI MSA
Johnson City-Kingsport-Bristol, TN-VA MSA
1999
0.073
0.120
0.086
0.089
0.096
0.083
0.101
0.145
0.098
0.116
0.097
0.126
0.109
0.122
0.126
0.161
0.115
0.054
0.203
0.122
0.106
0.114
0.110
0.103
0.103
0.105
0.111
2000
0.073
0.106
0.082
0.096
0.091
0.079
0.099
0.146
0.096
0.123
0.090
0.116
0.109
0.115
0.110
0.116
0.107
0.048
0.124
0.194
0.094
0.111
0.102
0.099
0.114
0.113
0.098
0.126
2001
0.069
0.108
0.074
0.088
0.079
0.095
0.098
0.146
0.096
0.118
0.107
0.122
0.091
0.108
0.105
0.139
0.099
0.051
0.106
0.170
0.113
0.088
0.114
0.095
0.093
0.109
0.093
0.110
January 2005 AX3-15 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Johnstown, PA MSA
Kalamazoo-Battle Creek, MI MSA
Kansas City, MO-KS MSA
Knoxville, TN MSA
Lafayette, LA MSA
Lake Charles, LA MSA
Lakeland- Winter Haven, FL MSA
Lancaster, PA MSA
Lansing-East Lansing, MI MSA
Laredo, TX MSA
Las Cruces, NM MSA
Las Vegas, NV-AZ MSA
Lawton, OK MSA
Lexington, KY MSA
Lima, OH MSA
Lincoln, NE MSA
Little Rock-North Little Rock, AR MSA
Longview-Marshall, TX MSA
Los Angeles-Riverside-Orange County, CA CMSA
Louisville, KY-IN MSA
Macon, GA MSA
Madison, WI MSA
McAllen-Edinburg-Mission, TX MSA
Medford- Ashland, OR MSA
Melbourne-Titusville-Palm Bay, FL MSA
Memphis, TN-AR-MS MSA
Merced, CA Msa
Miami-Fort Lauderdale, FL CMSA
1999
0.107
0.103
0.116
0.129
0.094
0.127
0.101
0.127
0.101
0.084
0.102
0.097
0.089
0.114
0.107
0.062
0.107
0.134
0.159
0.124
0.133
0.098
0.086
0.087
0.130
0.125
0.113
2000
0.104
0.090
0.124
0.131
0.123
0.133
0.102
0.107
0.091
0.085
0.123
0.094
0.094
0.086
0.100
0.072
0.114
0.131
0.174
0.112
0.131
0.087
0.088
0.079
0.093
0.123
0.120
0.094
2001
0.106
0.101
0.108
0.109
0.090
0.105
0.109
0.127
0.106
0.071
0.104
0.100
0.094
0.092
0.096
0.061
0.102
0.111
0.175
0.106
0.115
0.088
0.092
0.081
0.094
0.121
0.113
0.106
January 2005 AX3-16 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Milwaukee-Racine, WI CMSA
Minneapolis-St. Paul, MN-WI MSA
Mobile, AL MSA
Modesto, CA MSA
Monroe, LA MSA
Montgomery, AL MSA
Nashville, TN MSA
New London-Norwich, CT-RI MSA
New Orleans, LA MSA
NY-Northern NJ-Long Island, NY-NJ-CT-PA CMSA
Norfolk- Virginia Beach-Newport News, VA-NC MSA
Ocala, FL MSA
Oklahoma City, OK MSA
Omaha, NE-IA MSA
Orlando, FL MSA
Owensboro, KY MSA
Parkersburg-Marietta, WV-OH MSA
Pensacola, FL MSA
Peoria-Pekin, IL MSA
Phil-Wilmington-Atlantic City, PA-NJ-DE-MD CMSA
Phoenix-Mesa, AZ MSA
Pittsburgh, PA MSA
Pittsfield,MAMSA
Portland, ME MSA
Portland-Salem, OR-WA CMSA
Providence-Fall River- Warwick, RI-MA MSA
Provo-Orem, UT MSA
Raleigh-Durham-Chapel Hill, NC MSA
1999
0.122
0.088
0.104
0.111
0.097
0.110
0.123
0.127
0.117
0.154
0.135
0.097
0.097
0.093
0.101
0.102
0.123
0.106
0.099
0.152
0.119
0.132
0.092
0.120
0.094
0.133
0.109
0.134
2000
0.098
0.089
0.118
0.110
0.103
0.111
0.122
0.135
0.124
0.136
0.099
0.093
0.100
0.083
0.106
0.082
0.105
0.118
0.083
0.128
0.107
0.111
0.089
0.082
0.118
0.095
0.116
2001
0.122
0.109
0.095
0.111
0.090
0.094
0.110
0.109
0.111
0.146
0.100
0.089
0.097
0.076
0.108
0.086
0.108
0.098
0.084
0.131
0.106
0.112
0.112
0.124
0.093
0.144
0.094
0.113
January 2005 AX3-17 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Reading, PA MSA
Redding, CA MSA
Reno, NV MSA
Richmond-Petersburg, VA MSA
Roanoke, VA MSA
Rochester, NY MSA
Rockford, IL MSA
Rocky Mount, NC MSA
Sacramento-Yolo, CA CMSA
Salinas, CA MSA
Salt Lake City-Ogden, UT MSA
San Antonio, TX MSA
San Diego, CAMS A
San Francisco-Oakland-San Jose, CA CMSA
San Luis Obispo-Atascadero-Paso Robles, CA MSA
Santa Barbara-Santa Maria-Lompoc, CA MSA
Sarasota-Bradenton, FL MSA
Savannah, GA MSA
Scranton-Wilkes-Barre-Hazleton, PA MSA
Seattle-Tacoma-Bremerton, WA CMSA
Sharon, PA MSA
Sheboygan, WI MSA
Shreveport-Bossier City, LA MSA
Sioux Falls, SD MSA
South Bend, IN MSA
Spokane, WA MSA
Springfield, IL MSA
Springfield, MA MSA
1999
0.128
0.113
0.097
0.133
0.101
0.093
0.104
0.137
0.075
0.112
0.109
0.114
0.144
0.094
0.095
0.112
0.107
0.115
0.091
0.108
0.130
0.108
0.068
0.107
0.073
0.099
0.113
2000
0.105
0.098
0.088
0.112
0.095
0.088
0.084
0.106
0.134
0.084
0.100
0.095
0.123
0.126
0.082
0.100
0.107
0.102
0.093
0.095
0.098
0.106
0.129
0.076
0.095
0.082
0.100
0.099
2001
0.125
0.093
0.090
0.119
0.101
0.100
0.086
0.099
0.129
0.078
0.118
0.092
0.118
0.118
0.092
0.103
0.114
0.085
0.104
0.088
0.113
0.122
0.105
0.088
0.108
0.084
0.095
0.132
January 2005 AX3-18 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-3 (cont'd). The Second Highest Daily Maximum One-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Springfield, MO MSA
St. Louis, MO-IL MSA
State College, PA MSA
Steubenville-Weirton, OH-WVMSA
Stockton-Lodi, CA MSA
Syracuse, NY MSA
Tallahassee, FL MSA
Tampa-St. Petersburg-Clearwater, FL MSA
Terre Haute, IN MSA
Toledo, OH MSA
Tucson, AZ MSA
Tulsa, OK MSA
Tyler, TX MSA
Utica-Rome, NY MSA
Victoria, TX MSA
Visalia-Tulare-Porterville, CA MSA
Washington-Baltimore, DC-MD-VA-WV CMSA
Wausau, WI MSA
West Palm Beach-Boca Raton, FL MSA
Wheeling, WV-OH MSA
Wichita, KS MSA
Williamsport, PA MSA
Wilmington, NC MSA
York, PA MSA
Youngstown- Warren, OH MSA
Yuba City, CA MSA
1999
0.095
0.128
0.099
0.111
0.130
0.099
0.094
0.116
0.093
0.128
0.092
0.116
0.118
0.089
0.102
0.125
0.152
0.095
0.104
0.100
0.095
0.091
0.081
0.121
0.109
0.106
2000
0.092
0.123
0.109
0.100
0.111
0.083
0.092
0.108
0.088
0.095
0.085
0.122
0.083
0.094
0.119
0.128
0.081
0.093
0.093
0.093
0.092
0.097
0.112
0.097
0.101
2001
0.091
0.122
0.097
0.096
0.112
0.097
0.085
0.118
0.096
0.111
0.084
0.107
0.098
0.100
0.085
0.126
0.142
0.078
0.098
0.104
0.096
0.094
0.089
0.104
0.105
0.105
January 2005
AX3-19
DRAFT-DO NOT QUOTE OR CITE
-------
1-h(ppm)
A 0.061-0.093
• 0.094-0.111
• 0.112-0.132
it 0.133-0.175
Figure AX3-1. Second highest daily maximum 1-h O3 concentrations in 2001.
Source: U.S. Environmental Protection Agency (2003a).
1
2
3
4
5
9
10
11
12
13
concentration affects the 3-year average, and thus one might anticipate annual variability of the
3-year average concentrations above and below specific thresholds, such as 0.085 ppm.
Table AX3-5 summarizes the fourth highest daily maximum 8-h average O3 concentrations
by MSA for the years 1999 to 2001. The data are shown as a map in Figure AX3-2. The data
have been reported for the O3 season as summarized in Table AX3-1. In some cases, high
concentrations occur in the fall and winter periods as well as in the summertime. A comparison
of Tables AX3-3 and AX3-5 shows that the MSAs, which experience high second daily
maximum hourly average concentrations, also experience elevated fourth highest 8-h daily
maximum concentrations. However, there are considerably more MSAs and CMSAs that
experience fourth highest 8-h daily maximum concentrations > 0.085 ppm than MSAs and
CMSAs that experience second highest daily maximum hourly average concentrations
> 0.125 ppm.
January 2005
AX3-20
DRAFT-DO NOT QUOTE OR CITE
-------
3
to
o
o
X
OJ
i
to
o
E>
H
1
o
o
2|
0
H
O
C^j
o
w
0
o
H
W
Table
AIRS Site
Alabama
011030011
011030011
Arizona
040139508
040139508
040139508
Kentucky
210470006
210470006
210470006
New York
360551004
360551004
360551004
North Carolina
370210030
370210030
370210030
AX3-4. Percentiles of Hourly Average Concentrations (ppm) and Fourth Highest Eight-Hour Daily Maximum
Concentration for the April to October Period
Name
Morgan Co.
Morgan Co.
Maricopa Co.
Maricopa Co.
Maricopa Co.
Christian Co.
Christian Co.
Christian Co.
Monroe Co.
Monroe Co.
Monroe Co.
Buncombe Co.
Buncombe Co.
Buncombe Co.
Area Year Min.
Decatur 2000 0.000
Decatur 2001 0.000
Phoenix 1999 0.031
Phoenix 2000 0.019
Phoenix 2001 0.024
Clarksville 1999 0.000
Clarksville 2000 0.001
Clarksville 2001 0.002
Rochester 1999 0.000
Rochester 2000 0.000
Rochester 2001 0.001
Asheville 1999 0.000
Asheville 2000 0.000
Asheville 2001 0.000
Percentiles
4th High
10 30 50 70 90 95 99 Max. 8-hAvg
0.006 0.023 0.036 0.051 0.070 0.078 0.090 0.116
0.003 0.018 0.031 0.043 0.061 0.068 0.079 0.107
0.048 0.055 0.060 0.064 0.073 0.077 0.088 0.098
0.042 0.049 0.056 0.063 0.072 0.077 0.084 0.095
0.044 0.053 0.058 0.063 0.072 0.076 0.085 0.098
0.022 0.03 0.04 0.06 0.071 0.08 0.092 0.12
0.019 0.030 0.040 0.050 0.064 0.070 0.082 0.104
0.021 0.032 0.040 0.051 0.065 0.071 0.081 0.096
0.005 0.017 0.027 0.038 0.061 0.069 0.088 0.104
0.004 0.018 0.027 0.035 0.050 0.057 0.071 0.100
0.007 0.021 0.030 0.040 0.055 0.063 0.080 0.099
0.006 0.020 0.033 0.046 0.064 0.073 0.086 0.115
0.003 0.015 0.028 0.042 0.061 0.070 0.087 0.108
0.005 0.018 0.030 0.043 0.059 0.066 0.078 0.1
0.091
0.077
0.086
0.082
0.085
0.092
0.081
0.082
0.088
0.073
0.084
0.084
0.090
0.076
Number of
Observations
4169
4641
4922
4991
5042
5017
4944
5108
3649
4855
5014
5053
4864
4777
-------
3
to
o
o
X
OJ
to
to
o
l>
H
6
o
0
H
O
o
0
o
H
W
Table AX3-4 (cont'd). Percentiles of Hourly Average Concentrations (ppm) and Fourth Highest Eight-Hour Daily
Maximum Concentration for the April to October Period
AIRS Site Name Area
Ohio
390810016 Jefferson Co. Steubenville
390810016 Jefferson Co. Steubenville
390810016 Jefferson Co. Steubenville
Pennsylvania
420274000 Centre Co. State
College
420274000 Centre Co. State
College
420274000 Centre Co. State
College
Texas
480290032 Bextar Co. San Antonio
480290032 Bextar Co. San Antonio
480290032 Bextar Co. San Antonio
Percentiles
Year Min. 10 30 50 70 90 95 99 Max.
1999 0.000 0.001 0.012 0.027 0.041 0.063 0.073 0.086 0.113
2000 0.000 0.002 0.012 0.025 0.039 0.058 0.066 0.079 0.103
2001 0.000 0.003 0.014 0.027 0.041 0.063 0.072 0.086 0.104
1999 0.001 0.014 0.025 0.034 0.046 0.060 0.068 0.083 0.109
2000 0.000 0.009 0.021 0.031 0.040 0.056 0.063 0.077 0.109
2001 0.000 0.011 0.025 0.034 0.044 0.060 0.068 0.082 0.091
1999 0.000 0.006 0.017 0.027 0.038 0.059 0.070 0.091 0.120
2000 0.000 0.006 0.014 0.021 0.031 0.048 0.058 0.075 0.096
2001 0.000 0.005 0.014 0.022 0.031 0.046 0.055 0.074 0.095
4th High Number of
8-h Avg Observations
0.088 4764
0.080 4897
0.086 4688
0.085 5122
0.075 5089
0.082 5090
0.091 4720
0.077 4932
0.078 4920
-------
Table AX3-5. The Fourth Highest Daily Maximum Eight-Hour Ozone Concentration
(ppm) by Metropolitan Statistical Area (MSA) or Consolidated Metropolitan Statistical
Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Albany-Schenectady-Troy, NY MSA
Albuquerque, NM MSA
Allentown-Bethlehem-Easton, PA MSA
Altoona, PA MSA
Appleton-Oshkosh-Neenah, WI MSA
Asheville, NC MSA
Atlanta, GA MSA
Augusta-Aiken, GA-SC MSA
Austin-San Marcos, TX MSA
Bakersfield, CA MSA
Bangor, ME MSA
Barnstable-Yarmouth, MA MSA
Baton Rouge, LA MSA
Beaumont-Port Arthur, TX MSA
Bellingham, WA MSA
Benton Harbor, MI MSA
Biloxi-Gulfport-Pascagoula, MS MSA
Birmingham, AL MSA
Boston- Worcester-Lawrence, MA-NH-ME-CT CMSA
Brownsville-Harlingen-San Benito, TX MSA
Buffalo-Niagara Falls, NY MSA
Burlington, VT MSA
Canton-Massillon, OH MSA
Cedar Rapids, IA MSA
Champaign-Urbana, IL MSA
Charleston, WV MSA
Charleston-North Charleston, SC MSA
Charlotte-Gastonia-Rock Hill, NC-SC MSA
1999
0.092
0.076
0.107
0.091
0.087
0.084
0.126
0.090
0.099
0.109
0.080
0.101
0.100
0.077
0.050
0.096
0.097
0.100
0.098
0.066
0.090
0.079
0.091
0.076
0.094
0.104
0.084
0.107
2000
0.072
0.076
0.092
0.080
0.068
0.090
0.113
0.093
0.088
0.111
0.083
0.101
0.097
0.052
0.077
0.091
0.099
0.082
0.064
0.085
0.071
0.087
0.075
0.073
0.085
0.082
0.101
2001
0.090
0.074
0.094
0.083
0.087
0.076
0.092
0.082
0.080
0.109
0.088
0.105
0.084
0.081
0.050
0.088
0.083
0.089
0.101
0.063
0.102
0.076
0.089
0.069
0.073
0.083
0.071
0.103
January 2005
AX3-23
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-5 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Chattanooga, TN-GA MSA
Chicago-Gary-Kenosha, IL-IN-WI CMSA
Chico-Paradise, CA MSA
Cincinnati-Hamilton, OH-KY-IN CMSA
Clarksville-Hopkinsville, TN-KYMSA
Cleveland-Akron, OH CMSA
Colorado Springs, CO MSA
Columbia, SC MSA
Columbus, GA-AL MSA
Columbus, OH MSA
Corpus Christi, TX MSA
Dallas-Fort Worth, TX CMSA
Davenport-Moline-Rock Island, IA-IL MSA
Dayton-Springfield, OH MSA
Daytona Beach, FL MSA
Decatur, AL MSA
Decatur, IL MSA
Denver-Boulder-Greeley, CO CMSA
Des Moines, IA MSA
Detroit-Ann Arbor-Flint, MI CMSA
Dover, DE MSA
Duluth-Superior, MN-WI MSA
El Paso, TX MSA
Elkhart-Goshen, IN MSA
Elmira, NY MSA
Erie, PA MSA
Eugene-Springfield, OR MSA
Evansville-Henderson, IN-KY MSA
1999
0.098
0.102
0.087
0.096
0.092
0.103
0.064
0.094
0.097
0.103
0.085
0.112
0.083
0.096
0.075
0.092
0.087
0.081
0.073
0.096
0.097
0.074
0.071
0.077
0.082
0.096
0.068
0.098
2000
0.098
0.086
0.086
0.093
0.088
0.085
0.070
0.097
0.094
0.091
0.083
0.102
0.077
0.088
0.076
0.091
0.077
0.083
0.071
0.082
0.093
0.065
0.084
0.063
0.073
0.078
0.047
0.085
2001
0.087
0.099
0.088
0.088
0.082
0.099
0.070
0.091
0.079
0.090
0.077
0.098
0.079
0.084
0.073
0.077
0.071
0.082
0.060
0.095
0.091
0.062
0.078
0.055
0.082
0.089
0.066
0.081
January 2005
AX3-24
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-5 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Fargo-Moorhead, ND-MN MSA
Fayetteville, NC MSA
Flagstaff, AZ-UT MSA
Fort Collins-Loveland, CO MSA
Fort Myers-Cape Coral, FL MSA
Fort Fierce-Port St. Lucie, FL MSA
Fort Wayne, IN MSA
Fresno, CA MSA
Gainesville, FL MSA
Grand Rapids-Muskegon-Holland, MI MSA
Green Bay, WI MSA
Greensboro-Winston-Salem-High Point, NC MSA
Greenville, NC MSA
Greenville-Spartanburg-Anderson, SC MSA
Harrisburg-Lebanon-Carlisle, PA MSA
Hartford, CT MSA
Hickory -Morganton-Lenoir, NC MSA
Honolulu, HI MSA
Houma, LA MSA
Houston-Galveston-Brazoria, TX CMSA
Huntington- Ashland, WV-KY-OH MSA
Huntsville, AL MSA
Indianapolis, IN MSA
Jackson, MS MSA
Jacksonville, FL MSA
Jamestown, NY MSA
Janesville-Beloit, WI MSA
Johnson City-Kingsport-Bristol, TN-VA MSA
1999
0.066
0.100
0.076
0.074
0.081
0.071
0.090
0.105
0.080
0.103
0.085
0.100
0.093
0.100
0.104
0.107
0.094
0.048
0.087
0.124
0.097
0.093
0.096
0.084
0.080
0.092
0.093
0.089
2000
0.057
0.086
0.071
0.078
0.077
0.071
0.091
0.114
0.080
0.080
0.071
0.094
0.082
0.089
0.088
0.089
0.091
0.044
0.085
0.117
0.083
0.088
0.090
0.080
0.077
0.087
0.083
0.097
2001
0.060
0.084
0.070
0.070
0.068
0.075
0.082
0.113
0.079
0.095
0.088
0.096
0.077
0.090
0.091
0.102
0.088
0.042
0.084
0.110
0.088
0.080
0.093
0.078
0.075
0.090
0.084
0.086
January 2005
AX3-25
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-5 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Johnstown, PA MSA
Kalamazoo-Battle Creek, MI MSA
Kansas City, MO-KS MSA
Knoxville, TN MSA
Lafayette, LA MSA
Lake Charles, LA MSA
Lakeland- Winter Haven, FL MSA
Lancaster, PA MSA
Lansing-East Lansing, MI MSA
Laredo, TX MSA
Las Cruces, NM MSA
Las Vegas, NV-AZ MSA
Lawton, OK MSA
Lexington, KY MSA
Lima, OH MSA
Lincoln, NE MSA
Little Rock-North Little Rock, AR MSA
Longview-Marshall, TX MSA
Los Angeles-Riverside-Orange County, CA CMSA
Louisville, KY-IN MSA
Macon, GA MSA
Madison, WI MSA
McAllen-Edinburg-Mission, TX MSA
Medford- Ashland, OR MSA
Melbourne-Titusville-Palm Bay, FL MSA
Memphis, TN-AR-MS MSA
Merced, CA MSA
Miami-Fort Lauderdale, FL CMSA
1999
0.090
0.091
0.084
0.106
0.081
0.088
0.078
0.102
0.089
0.067
0.082
0.084
0.082
0.091
0.093
0.053
0.089
0.105
0.133
0.103
0.113
0.085
0.075
0.077
0.104
0.105
0.077
2000
0.086
0.070
0.091
0.100
0.092
0.090
0.079
0.090
0.077
0.070
0.081
0.082
0.085
0.077
0.085
0.057
0.092
0.099
0.122
0.090
0.097
0.071
0.077
0.067
0.078
0.093
0.103
0.077
2001
0.090
0.085
0.079
0.093
0.077
0.080
0.087
0.097
0.087
0.063
0.079
0.082
0.078
0.078
0.081
0.051
0.080
0.082
0.133
0.086
0.086
0.078
0.074
0.064
0.084
0.092
0.096
0.076
January 2005
AX3-26
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-5 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Milwaukee-Racine, WI CMSA
Minneapolis-St. Paul, MN-WI MSA
Mobile, AL MSA
Modesto, CA MSA
Monroe, LA MSA
Montgomery, AL MSA
Nashville, TN MSA
New London-Norwich, CT-RI MSA
New Orleans, LA MSA
NY-Northern NJ-Long Island, NY-NJ-CT-PA CMSA
Norfolk- Virginia Beach-Newport News, VA-NC MSA
Ocala, FL MSA
Oklahoma City, OK MSA
Omaha, NE-IA MSA
Orlando, FL MSA
Owensboro, KY MSA
Panama City, FL MSA
Parkersburg-Marietta, WV-OH MSA
Pensacola, FL MSA
Peoria-Pekin, IL MSA
Phil-Wilmington-Atlantic City, PA-NJ-DE-MD CMSA
Phoenix-Mesa, AZ MSA
Pittsburgh, PA MSA
Pittsfield,MAMSA
Portland, ME MSA
Portland-Salem, OR-WA CMSA
Providence -Fall River- Warwick, RI-MA MSA
Provo-Orem, UT MSA
1999
0.097
0.077
0.085
0.090
0.082
0.092
0.101
0.096
0.091
0.113
0.097
0.083
0.084
0.080
0.084
0.090
0.097
0.086
0.082
0.112
0.091
0.101
0.075
0.089
0.072
0.091
0.083
2000
0.086
0.072
0.097
0.091
0.081
0.086
0.093
0.084
0.095
0.114
0.084
0.079
0.086
0.072
0.081
0.074
0.092
0.087
0.096
0.073
0.109
0.090
0.088
0.073
0.065
0.087
0.077
2001
0.102
0.078
0.078
0.094
0.077
0.077
0.086
0.090
0.084
0.108
0.085
0.074
0.082
0.063
0.079
0.073
0.082
0.085
0.082
0.080
0.105
0.086
0.093
0.092
0.097
0.069
0.105
0.076
January 2005
AX3-27
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-5 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Raleigh-Durham-Chapel Hill, NC MSA
Reading, PA MSA
Redding, CA MSA
Reno, NV MSA
Richmond-Petersburg, VA MSA
Roanoke, VA MSA
Rochester, NY MSA
Rockford, IL MSA
Rocky Mount, NC MSA
Sacramento-Yolo, CA CMSA
Salinas, CA MSA
Salt Lake City-Ogden, UT MSA
San Antonio, TX MSA
San Diego, CAMS A
San Francisco-Oakland-San Jose, CA CMSA
San Luis Obispo-Atascadero-Paso Robles, CA MSA
Santa Barbara-Santa Maria-Lompoc, CA MSA
Sarasota-Bradenton, FL MSA
Savannah, GA MSA
Scranton-Wilkes-Barre-Hazleton, PA MSA
Seattle-Tacoma-Bremerton, WA CMSA
Sharon, PA MSA
Sheboygan, WI MSA
Shreveport-Bossier City, LA MSA
Sioux Falls, SD MSA
South Bend, IN MSA
Spokane, WA MSA
Springfield, IL MSA
1999
0.108
0.102
0.094
0.076
0.100
0.089
0.089
0.082
0.092
0.106
0.063
0.080
0.091
0.092
0.088
0.077
0.079
0.085
0.083
0.096
0.072
0.091
0.093
0.094
0.058
0.090
0.065
0.075
2000
0.089
0.084
0.082
0.069
0.083
0.081
0.073
0.070
0.085
0.103
0.063
0.080
0.082
0.095
0.077
0.070
0.080
0.086
0.079
0.077
0.070
0.081
0.090
0.093
0.065
0.081
0.068
0.079
2001
0.089
0.099
0.077
0.075
0.091
0.089
0.086
0.078
0.085
0.105
0.063
0.084
0.081
0.096
0.081
0.073
0.083
0.086
0.067
0.088
0.071
0.094
0.102
0.084
0.065
0.089
0.071
0.073
January 2005
AX3-28
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-5 (cont'd). The Fourth Highest Daily Maximum Eight-Hour Ozone
Concentration (ppm) by Metropolitan Statistical Area (MSA) or Consolidated
Metropolitan Statistical Area (CMSA) for the Years 1999 to 2001
MSA/CMSA
Springfield, MA MSA
Springfield, MO MSA
St. Louis, MO-IL MSA
State College, PA MSA
Steubenville-Weirton, OH-WVMSA
Stockton-Lodi, CA MSA
Syracuse, NY MSA
Tallahassee, FL MSA
Tampa-St. Petersburg-Clearwater, FL MSA
Terre Haute, IN MSA
Toledo, OH MSA
Tucson, AZ MSA
Tulsa, OK MSA
Tyler, TX MSA
Utica-Rome, NY MSA
Victoria, TX MSA
Visalia-Tulare-Porterville, CA MSA
Washington-Baltimore, DC-MD-VA-WV CMSA
Wausau, WI MSA
West Palm Beach-Boca Raton, FL MSA
Wheeling, WV-OH MSA
Wichita, KS MSA
Williamsport, PA MSA
Wilmington, NC MSA
York, PA MSA
Youngstown- Warren, OH MSA
Yuba City, CA MSA
1999
0.094
0.081
0.102
0.085
0.091
0.094
0.086
0.081
0.087
0.082
0.091
0.073
0.091
0.097
0.078
0.086
0.108
0.113
0.084
0.079
0.088
0.079
0.076
0.067
0.094
0.095
0.090
2000
0.079
0.078
0.088
0.079
0.080
0.082
0.074
0.076
0.083
0.075
0.081
0.077
0.088
0.087
0.067
0.079
0.105
0.099
0.073
0.075
0.071
0.080
0.073
0.080
0.090
0.080
0.083
2001
0.093
0.072
0.088
0.086
0.086
0.078
0.085
0.074
0.085
0.083
0.092
0.069
0.095
0.082
0.084
0.073
0.104
0.112
0.072
0.073
0.088
0.084
0.080
0.078
0.087
0.093
0.084
January 2005
AX3-29
DRAFT-DO NOT QUOTE OR CITE
-------
8-h (ppm)
A 0.051-0.072
• 0.073-0.085
• 0.086-0.098
it 0.099-0.133
Figure AX3-2. Fourth highest 8-h daily maximum O3 concentration in 2001.
Source: U.S. Environmental Protection Agency (2003a).
1 There has been considerable interest in possibly substituting one index for another when
2 attempting to relate O3 exposure with an effect. For example, using O3 ambient air quality data,
3 McCurdy (1988) compared the number of exceedances of 0.12 ppm and the number of
4 occurrences of the daily maximum 8-h average concentrations > 0.08 ppm and reported that a
5 positive correlation (r = 0.79) existed between the second-highest 1-h daily maximum in a year
6 and the expected number of days with an 8-h daily maximum average concentration > 0.08 ppm
7 O3. In this case, the predictive strength of using one O3 exposure index to predict another is not
8 strong. However, such may not be the case when relating the second highest daily maximum 1-h
9 average concentration with the fourth highest 8-h daily maximum concentration that occurs over
10 an O3 season. Figure AX3-3 below shows the scatter diagram for the two exposure indices for
11 all urban sites in the AQS database for 2001. The diagram shows that there is a fairly close
12 relationship between the two indices. Figure AX3-4 shows a similar diagram using the rural
13 (sites designated as rural in the AQS database) monitors that measured O3 concentrations in
January 2005
AX3-30
DRAFT-DO NOT QUOTE OR CITE
-------
Second Highest 1-h Daily Max for Season versus
Fourth Highest 8-h Daily Max for Season
Urban
2001
0.00
0.05 0.10 0.15
Concentration 1h (ppm)
0.20
Figure AX3-3. The relationship between the second highest 1-h average daily maximum
concentration and the fourth highest 8-h average daily maximum
concentration at urban sites for 2001 for the O3 season.
Source: U.S. Environmental Protection Agency (2003a).
Second Highest 1-h Daily Max for Season versus
Fourth Highest 8-h Daily Max for Season
Rural
2001
—. 0.14
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
0.20
Concentration 1h (ppm)
Figure AX3-4. The relationship between the second highest 1-h average daily maximum
concentration and the fourth highest 8-h average daily maximum
concentration at rural sites for 2001 for the O3 season.
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-31
DRAFT-DO NOT QUOTE OR CITE
-------
1 2001. Using air quality data from the AQS database for 2001, both Figure AX3-3 and
2 Figure AX3-4 appear to show similar relationships for both urban and rural monitors. Thus,
3 as indicated in Tables AX3-3 and AX3-5, there is a pattern for those MSAs with high second
4 daily maximum hourly average concentrations to experience elevated fourth highest 8-h daily
5 maximum concentrations.
6 For analyzing the results of epidemiological time-series studies, the percentiles of the
7 pooled data have been characterized for all monitoring sites across the United States. All of the
8 hourly average concentrations from all monitoring sites have been combined. Table AX3-6
9 shows the results of pooling all data in the AQS database for 1996 to 2000 for sites within MSAs
10 and outside of MSAs. The following tables have 5-year means and selected percentiles of
11 hourly, daily, daily maximum 1-h, and daily 1000-1800 hours 8-h averaged O3 concentrations.
12 These were calculated from 1996 to 2000 O3 data for the months April through September. The
13 source of the data used was AQS AMP350 national-level hourly O3 raw data files. Some data
14 completeness requirements were imposed:
15 (1) All hours in the 8-h period (1000-1800 hours) were required for that day to be
included.
16 (2) At least 75% of available days were required for a monitor to be used.
17 (3) At least 75% of the hours in a day were required for a 24-h average to be
computed and used.
18 The data have been characterized by the hourly average concentrations, the 8-h daily maximum
19 concentrations, and the 24-h average concentrations, as well as the mean concentrations for the
20 specific averaging times.
21 As pointed out by the U.S. Environmental Protection Agency (1986), a familiar measure of
22 O3 air quality is the number or percentage of days on which some specific concentration is
23 equaled or exceeded. This measure, however, does not shed light on one of the more important
24 questions regarding the effects of O3 on both people and plants: what is the possible significance
25 of high concentrations lasting 1 h or longer and then recurring on 2 or more successive days?
26 The recurrence of high O3 concentrations on consecutive days was examined in four cities (one
27 site in each city) by the U.S. Environmental Protection Agency (1986). The numbers of
28 multiday events were tallied by length of event (i.e., number of consecutive days) using data for
January 2005 AX3-32 DRAFT-DO NOT QUOTE OR CITE
-------
3
Table AX3-6. Summary of Percentiles of Pooled Data Across Monitoring Sites
for 1996 - 2000a
<<
to
o
o
>
X
OJ
OJ
O
^
H
6
o
0
H
O
O
w
0
o
H
W
Pooled Group/ Number
Avg. Time of Values Mean 1 5
Daily 1-h Maximum Concentrations
Monitors in 536,824 58 20 29
MSAs
Monitors not in 104,798 57 22 30
MSAs
8-h Daily Maximum Concentrations (1000 - 1800 hours)
Monitors in 536,824 47 13 21
MSAs
Monitors not in 104,798 48 16 24
MSAs
24-h Average Concentrations
Monitors in 536,148 34 9 15
MSAs
Monitors not in 104,689 38 12 18
MSAs
Hourly Average Concentrations
Monitors in 1,265,0861 33 0 2
MSAs
Monitors not in 2,453,249 38 0 6
MSAs
Percentiles
10 25 30 50 70 75 90 95 99
34 44 47 56 67 70 85 95 118
35 45 47 55 65 68 80 88 105
26 35 38 46 55 58 70 78 93
29 37 38 47 55 58 68 74 87
18 25 27 33 39 41 49 55 67
21 28 30 37 44 46 55 61 73
6 17 20 32 43 47 62 71 91
12 24 27 37 48 51 63 71 86
Concentrations inppb.
-------
1 the daylight hours (0600 to 2000 hours) in the second and third quarters of 1979 through 1981.
2 These sites were selected because they included areas known to experience high O3
3 concentrations (e.g., California), and because they represent different geographic regions of the
4 country (West, Southwest, and East).
5 Because of the importance of episodes and respites in assessing possible biological
6 impacts, the U.S. Environmental Protection Agency (1986) commented on the number of O3
7 episodes, the length of episodes, and the time between episodes. The agency concluded that its
8 analysis showed variations among sites in the lengths of episodes as well as the respite periods.
9 In its discussion, the U.S. Environmental Protection Agency (1986) defined a day or series of
10 days on which the daily 1-h maximum reached or exceeded the specified level as an "exposure,"
11 the intervening day or days when that level was not reached was called a "respite." Four O3
12 concentrations were selected: 0.06, 0.12, 0.18, and 0.24 ppm. At the Dallas site, for example,
13 the concentration value was > 0.06 ppm for more than 7 days in a row. The Pasadena site
14 experienced 10 such exposures, but these 10 exposure events spanned 443 days; in Dallas, 11
15 exposures spanned only 168 days. At the lowest concentration (> 0.06 ppm), the Dallas station
16 recorded more short-term (< 7 days) exposures (45 exposures over 159 days) than the Pasadena
17 station (14 exposures over 45 days) because the daily 1-h maximum statistic in Pasadena
18 remained above 0.06 ppm for such protracted periods. At concentrations > 0.12 ppm, the
19 lengthy exposures at the Pasadena site resolved into numerous shorter exposures, whereas in
20 Dallas the exposures markedly dwindled in number and duration.
21 An interesting observation first reported and studied in the 1970s is that O3 concentrations
22 in some locations are actually higher on weekends than on weekdays although the atmospheric
23 levels of O3 precursors are lower on weekends compared with weekdays (Cleveland et al., 1974;
24 Lebron, 1975). The weekly cycles of atmospheric O3 are of interest, because they provide
25 information about the response of O3 to changes in anthropogenic emissions from weekdays to
26 weekends. Heuss et al. (2003) described the results of a nationwide analysis of
27 weekday/weekend differences that demonstrated significant variation across the United States.
28 The authors reported that weekend 1-h or 8-h maximum O3 varied from 15% lower than
29 weekday levels to 30% higher. The weekend O3 increases were primarily found in and around
30 large coast cities in California and large cities in the Midwest and Northeast Corridor. Many
31 sites that experienced elevated weekday O3 also had higher O3 on weekends even though the
January 2005 AX3-34 DRAFT-DO NOT QUOTE OR CITE
-------
1 traffic and O3 precursor levels were substantially reduced on weekends. The authors reported
2 that detailed studies of this phenomenon indicated that the primary cause of the higher O3 on
3 weekends was the reduction in oxides of nitrogen emissions on weekends in a volatile organic
4 compound (VOC)-limited chemical regime. Heuss et al. (2003) hypothesized that the lower O3
5 on weekends in other locations may result from NOX reductions in a NOx-limited regime.
6 Pun et al. (2003) described the day-of-week behavior for O3 in Chicago, Philadelphia, and
7 Atlanta. In Chicago and Philadelphia, maximum 1-h average O3 increases on weekends. In
8 Atlanta, O3 builds up from Mondays to Fridays and declines during the weekends. Fujita et al.
9 (2003) pointed out that since the mid-1970s, O3 levels in portions of California's South Coast
10 Air Basin on weekends have been as high as or higher than levels on weekdays, even though
11 emissions of O3 precursors are lower on weekends. The authors reported that the analysis of the
12 ambient data indicates that the intensity and spatial extent of the weekend O3 effect are
13 correlated with day-of-week variations in the extent of O3 inhibition caused by titration with
14 nitric oxide (NO), reaction of hydroxyl radical (OH) with nitrogen dioxide (NO2), and the rate of
15 O3 accumulation. Blanchard and Tanebaum (2003) noted that despite significantly lower O3
16 precursor levels on weekends, 20 of 28 South Coast Air Basin sites showed statistically
17 significant higher mean O3 levels on Sundays than on weekdays. Chinkin et al. (2003) noted that
18 ambient O3 levels in California's South Coast Air Basin can be as much as 55% higher on
19 weekends than on weekdays under comparable meteorological conditions.
20
21 AX3.2.2 Nonurban Area Concentrations
22 It is difficult to identify a set of unique O3 distribution patterns that adequately describes
23 the hourly average concentrations experienced at monitoring sites in nonurban locations, because
24 many nonurban sites in the United States are influenced by local sources of pollution or
25 long-range transport of O3 or its precursors (U.S. Environmental Protection Agency, 1996a).
26 At most of the RRMS characterized in Section AX3.2.3, the maximum hourly average
27 concentrations were below 0.08 ppm. Unlike these RRMS, urban-influenced nonurban sites
28 sometimes show frequent hourly average concentrations near the minimum detectable level, but
29 often show occurrences of hourly average concentrations > 0.10 ppm. The frequent occurrence
30 of hourly average concentrations near the minimum detectable level is indicative of scavenging
31 processes (i.e., by NO); the presence of high hourly average concentrations can be attributable to
January 2005 AX3-35 DRAFT-DO NOT QUOTE OR CITE
-------
1 the influence of either local generation or the long-range transport of O3. For example, the U.S.
2 Environmental Protection Agency (2003a) reported that Whiteface Mountain (NY) experienced
3 a 3-year average of the fourth highest 8-h daily maximum concentration of 0.086 ppm for the
4 period of 2000 to 2002. In 2002, the maximum hourly average concentration reached 0.113 ppm
5 (U.S. Environmental Protection Agency, 2003b). Similarly, for the period of 2000 to 2002,
6 high-elevation sites in the Great Smoky Mountain National Park in North Carolina and
7 Tennessee experienced a 3-year average of the fourth highest 8-h daily maximum concentrations
8 > 0.085 ppm. In the Great Smoky Mountain National Park, maximum hourly average
9 concentrations reached -0.12 ppm. These sites are influenced by the long-range transport of O3.
10 Using hourly averaged data from the AQS database for a select number of rural monitoring sites,
11 Table AX3-7 summarizes the percentiles of the hourly average O3 concentrations, the number of
12 occurrences of the hourly average concentration > 0.08 and 0.10 ppm, the 7-month sum of all
13 hourly average concentrations >0.06 ppm, and the 7-month W126 exposure index. Note the
14 large variation in the number of hourly average concentrations > 0.08 and 0.10 ppm. These sites
15 are influenced either by local sources or by transport of O3 or its precursors. The maximum
16 hourly average concentrations are generally > 0.10 ppm, and the occurrence of hourly average
17 concentrations near minimum detectable levels indicates NOX scavenging processes. Also note
18 that in 2001, although the Crestline, CA site in the San Bernardino National Forest and the Cove
19 Mountain site in the Great Smoky Mountains National Park experienced similar cumulative
20 exposures (SUM06 and W126 values), the Crestline site experienced considerably more high
21 hourly average concentrations > 0.10 ppm (369) than the Cove Mountain site (5).
22 Taylor et al. (1992) have summarized the O3 concentrations that were experienced at
23 10 Electric Power Research Institute (EPRI) Integrated Forest Study sites in North America.
24 The authors reported that in 1988 all sites experienced maximum hourly average concentrations
25 > 0.08 ppm. In almost all cases, the sites experienced multiple occurrences above 0.08 ppm.
26 This implies that, although the sites were located in remote forested areas, the sites experienced
27 elevated O3 concentrations that were more than likely due to long-range transport of O3 or its
28 precursors.
29 Ozone concentrations in the Shenandoah National Park exhibit some features in common
30 with both urban and rural areas on a seasonal basis. During some years, maximum hourly
31 average concentrations exceed 0.12 ppm, although some sites in the park exhibit hourly average
January 2005 AX3-36 DRAFT-DO NOT QUOTE OR CITE
-------
3
to
o
o
i iiuic j-v.-vj-/. ociisuniii ^"vjji ii uiiuugii v^v.iuuci ; i ci v.cninc LMSU luuiiun ui iiuiiiiy v^iuiic v uiiiciiii iiiiuiis, nuinuci ui
Hourly Mean Ozone Occurrences > 0.08 and > 0.10, SUM06, and W126 Values for Selected Rural Ozone Monitoring Sites
in 2001. Concentrations in ppm.
Percentiles Hours
AIRS Site
Name
Min. 10
30
50 70 90 95 99 Max. Hours iO.08
i 0.10 (ppm-h) (ppm-h)
RURAL AGRICULTURAL
X
OJ
UJ
o
170491001
180970042
240030014
310550032
420070002
510610002
551390011
RURAL FOREST
060430003
360310002
471550101
511130003
Effingham Co., IL
Indianapolis, IN
Anne Arundel, MD
Omaha, NE
Beaver Co., PA
Fauquier Co., VA
Oshkosh Co., WI
Yosemite NP, CA
Whiteface Mtn., NY
Smoky Mtns. NP, TN
Shenandoah NP
0.000 0.007
0.000 0.007
0.000 0.004
0.000 0.015
0.000 0.020
0.000 0.004
0.002 0.02
0.013 0.038
0.000 0.024
0.020 0.039
0.010 0.035
0.02
0.021
0.019
0.023
0.031
0.017
0.026
0.046
0.035
0.050
0.045
0.031 0.041 0.056 0.06 0.07 0.1 5076 21
0.033 0.044 0.061 0.068 0.078 0.1 4367 37
0.031 0.043 0.062 0.072 0.096 0.13 5059 142
0.030 0.038 0.050 0.056 0.064 0.1 5116 1
0.040 0.050 0.068 0.075 0.089 0.11 5080 156
0.029 0.041 0.057 0.065 0.08 0.11 5055 48
0.035 0.043 0.057 0.065 0.082 0.1 4345 59
0.052 0.058 0.069 0.074 0.084 0.11 4619 106
0.043 0.051 0.063 0.070 0.084 0.11 4536 77
0.057 0.064 0.075 0.080 0.089 0.11 5074 267
0.052 0.059 0.070 0.075 0.089 0.11 4813 157
0 24.7 20.7
0 34.2 25.8
34 43.5 35.7
0 8.6 10.6
6 63.2 50.0
1 27.4 22.7
1 24.8 22.8
6 86.0 65.6
4 44.7 37.7
5 149.2 106.5
4 98.5 73.6
C RURAL OTHER (I.E., RURAL RESIDENTIAL OR RURAL COMMERCIAL)
H
6
o
0
H
O
o
0
o
H
W
060710005
350431001
370810011
371470099
Crestline, CA
Sandoval Co., NM
Guilford Co., NC
Farmville, NC
0.000 0.024
0.000 0.004
0.000 0.003
— 0.000
0.043
0.020
0.015
0.008
0.055 0.068 0.093 0.107 0.135 0.17 4922 933
0.033 0.045 0.056 0.061 0.070 0.1 5088 9
0.029 0.043 0.063 0.071 0.087 0.12 4879 111
0.021 0.033 0.044 0.061 0.068 0.1 0.097 4882
369 174.8 144.9
0 21.4 20.3
10 45.4 334.4
34 0 36.6
-------
1 concentrations near the minimum detectable level. Taylor and Norby (1985) have characterized
2 O3 episodes, which they defined as any day in which a 1-h mean O3 concentration was
3 > 0.08 ppm. Based on a 4-year monitoring period in Shenandoah National Park, the probability
4 was 80% that any given episode during the growing season would last 2 or more days, whereas
5 the probabilities of episodes lasting for periods greater than 3, 4, and 5 days were 30, 10, and
6 2%, respectively. Single-day O3 episodes were infrequent. Taylor and Norby (1985) noted that,
7 given the frequency of respites, there was a 50% probability that a second episode would occur
8 within 2 weeks.
9 Because of a lack of air quality data collected at rural and remote locations, interpolation
10 techniques must be used to estimate O3 exposures in these areas. In the absence of actual O3
11 data, interpolation techniques have been applied to the estimation of O3 exposures across the
12 United States (Reagan, 1984; Lefohn et al., 1987; Knudsen and Lefohn, 1988). "Kriging," a
13 mathematical interpolation technique (Matheron, 1963), has been used in the analyses of air
14 quality data (Grivet, 1980; Faith and Sheshinski, 1979) and was used to provide estimates of
15 seasonal O3 values for the National Crop Loss Assessment Network (NCLAN) for 1978 through
16 1982 (May to September of each year) (Reagan, 1984). These values, along with updated
17 values, coupled with exposure-response models, were used to predict agriculturally related
18 economic benefits anticipated by lower O3 levels in the United States (Adams et al., 1985, 1989).
19 The U.S. Environmental Protection Agency (1996a) concluded that the higher hourly
20 average concentrations (> 0.10 ppm) should be provided greater weight than the mid-level (0.06
21 to 0.099 ppm) and lower hourly average concentrations in predicting injury and yield reduction
22 for agricultural crops and forests. The most recent findings concerning the importance of the
23 higher hourly average concentrations in comparison to the mid-level and lower values will be
24 discussed in Chapter 9. For 2001, ordinary kriging was used to estimate the seasonal W126,
25 SUM06, and number of hours > 0.10 ppm (N100), using hourly average concentrations
26 accumulated over a 24-h period. As discussed in Chapter 9, the correlation between the number
27 of occurrences of hourly average concentrations > 0.10 ppm and the magnitude of the W126 and
28 SUM06 values is not strong. Because of this, the N100 was also estimated, along with the W126
29 and SUM06 exposure indices. For the period of April through September, the estimates of the
30 seasonal W126, SUM06, and N100 exposure index values were made for each 0.5° by 0.5° cell
31 in the continuous United States. The kriged values, the variance, and the 95% error bound for
January 2005 AX3-38 DRAFT-DO NOT QUOTE OR CITE
-------
1 each 0.5° by 0.5° cell were estimated. Because of the concern for inner-city depletion caused by
2 NOX scavenging, data from specific monitoring stations located in large metropolitan areas were
3 not included in the analysis.
4 Figure AX3-5 shows the kriged values for the 24-h cumulative seasonal W126 exposure
5 index and the N100 index for 2001 for the eastern United States. Note that for some of the areas
6 with elevated W126 values (e.g., > 35 ppm-h), the number of hourly average concentrations was
7 estimated to be < 22. Figure AX3-6 illustrates the kriged values using the 24-h cumulative
8 seasonal SUM06 exposure index and the N100 index for 2001 for the eastern United States.
9 Figures AX3-7 and AX3-8 show the W126 and SUM06 values, respectively, with the N100
10 values for the central United States region. For 2001, the number of hourly average
11 concentrations > 0.10 ppm was usually < 22 for the 6-month period. Figures AX3-9 and
12 AX3-10 illustrate the W126 and SUM06 values, respectively, for the western United States
13 region. Note that in the Southern California and Central California areas, the number of hourly
14 average concentrations > 0.10 ppm was in the range of 48 to 208 for the 6-month period. This is
15 considerably greater than the frequency of occurrences for the higher hourly average
16 concentrations experienced in the eastern and central United States.
17 Due to the scarcity of monitoring sites across the United States, especially in the Rocky
18 Mountain region, the uncertainty in the estimates for the various exposure indices vary.
19 Figures AX3-11 through AX3-19 illustrate the 95% confidence intervals associated with the
20 indices by region. In some cases, the uncertainty in the estimates of the exposure indices is
21 great. However, based on the actual hourly average concentrations measured, the pattern of
22 distinct differences across the regions in the United States for the number of hourly average
23 concentrations > 0.10 ppm is real even though at times the uncertainty in the kriged estimates
24 may be large.
25
26 AX3.2.3 Ozone Concentrations Observed at Relatively Remote
27 Monitoring Sites
28 Hourly average concentrations used in controlled O3 exposures for both human health and
29 vegetation studies appear to be lower than those experienced at RRMS in the United States or
30 in other parts of the world (see Chapter 9). This might have implications for human health
31 (see Annex AX6) and vegetation effects. This concern is relevant to those using exposure
January 2005 AX3-39 DRAFT-DO NOT QUOTE OR CITE
-------
» * *
-V»
0 125 250
500
750
1,000
^^ Miles
1:13,936,494
Figure AX3-5.
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.
January 2005
AX3-40
DRAFT-DO NOT QUOTE OR CITE
-------
1,000
Miles
Figure AX3-6.
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.
January 2005
AX3-41
DRAFT-DO NOT QUOTE OR CITE
-------
N100 (hours)
1 -2
* >2-7
• >7-21
• >21 - 48
A >48 - 208
W126 (ppm-h)
; 1 1-19
; i >19-35
j j >35 - 68
»68-146
\ « • » »H
* * *i
I , . , * v
: """^ ~-v.ii
» * • * • ^ >-,
LLLJJ N r
^^^^ i
* * * \ v
... .^. . . :
» » * •« *4#» ** ^--..^
• »»»»»»» »•****••/• .»»» « \
'.* v.., j ••••«» ••••••• */* . • »_ » » f-^z.-
"xf /"" "^^ . . . * . • • • » • f3^"'": *" ^'^.-J: *.?-:?
"•--,." \ 4 » » • « (,'""" -^v -v>
\ *tf ^™"
N /"
V4 A
>*^-4
N
0 100 200 400 600 800
1:11,998,827
Figure AX3-7.
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.
January 2005
AX3-42
DRAFT-DO NOT QUOTE OR CITE
-------
800
Miles
1:11,998,827
Figure AX3-8.
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.
January 2005
AX3-43
DRAFT-DO NOT QUOTE OR CITE
-------
800
Miles
Figure AX3-9. 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.
January 2005
AX3-44
DRAFT-DO NOT QUOTE OR CITE
-------
800
Miles
1:13,052,120
Figure AX3-10. 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.
January 2005
AX3-45
DRAFT-DO NOT QUOTE OR CITE
-------
~"n-"---
,..l.^" „;"
W126 95% Conf. Interval
i 3- 10ppm-h
: >10-15 pprn-h
I I >15-25 ppm-h
>25 - 90 ppm-h
Figure AX3-11. 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
2 -12 ppm-h
>12 -15 ppm-h
>15-20 ppm-h
> 20 - 73 ppm-h
Figure AX3-12. The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the eastern
United States.
January 2005
AX3-46
DRAFT-DO NOT QUOTE OR CITE
-------
N100 95% Conf. Interval
1 0-lOh
>10-15h
>15-20h
[ >20 - 95 h
i >95-12?h
Figure AX3-13. 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 9S% Conf. Interval
I I 3- 10ppm-h
: j >10 -15 ppm-h
I j >15-25 ppm-h
>25 - 90 ppm-h
Figure AX3-14. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative W126 exposure index for 2001 for the central United States.
January 2005
AX3-47
DRAFT-DO NOT QUOTE OR CITE
-------
SUM06 95% Conf. Interval
| | 2-12 ppm-h
| I >12 -15 ppm-h
> 20 - 73 ppm-h
Figure AX3-15. 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
[ J0-10h
[ i >10-15h
j i >15-20h
>20 - 95 h
>95-127h
Figure AX3-16. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the central United States.
January 2005
AX3-48
DRAFT-DO NOT QUOTE OR CITE
-------
W126 95% Conf. Interval
'y^vi 3- 10ppm-h
j >10 -15 ppm-h
>25 - 90 ppm-h
Figure AX3-17. 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
2-12 ppm-h
>12-15 ppm-h
>15-20 ppm-h
> 20 - 73 ppm-h
Figure AX3-18. The 95% confidence interval for the 6-month (April to September)
24-h cumulative SUM06 exposure index for 2001 for the western
United States.
January 2005
AX3-49
DRAFT-DO NOT QUOTE OR CITE
-------
N100 95% Conf. Interval
• >10-15h
• >15-20h
>20 - 95 h
>95-127h
Figure AX3-19. The 95% confidence interval for the 6-month (April to September) 24-h
cumulative N100 exposure index for 2001 for the western United States.
1 indices (e.g., 8-h average concentrations, SUM06, W126, etc.) to predict effects when the control
2 exposure regimes are based on O3 concentrations that are below background levels that cannot
3 be controlled with emission reductions (Tingey et al., 2002).
4 The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) concluded that the
5 annual average background concentration of O3 near sea level ranged from 0.020 to 0.035 ppm
6 and that, during the summer, the 1-h daily maximum ranged from 0.03 to 0.05 ppm. The 1996
7 O3 AQCD also included O3 hourly average concentrations measured at several clean, RRMS
8 mostly located in the western United States. Table AX3-8 provides an updated summary of the
9 characterization of the hourly average concentrations recorded from 1988 to 2001 at some of the
10 monitoring sites previously analyzed. The percentile distribution of the hourly average
11 concentrations (April to October), number of hourly average occurrences > 0.08 and > 0.10 ppm,
12 seasonal 7-h average concentrations, the SUM06, and W126 values were characterized for those
13 site years with a data capture of > 75%. From 1988 to 2001, no hourly average concentrations
January 2005
AX3-50
DRAFT-DO NOT QUOTE OR CITE
-------
3
to
o
o
X
OJ
O
§
H
6
o
0
H
O
O
0
o
H
W
i iiuic j-v.-vj-o. ociisuniii ^"vjji ii lu v^iiuuci ; i ci v.cninc LMSII luuiiun ui iiuiiiiy v^iuiic v uiiiciiii iiiiuiis ^|j|jin;< nuinuci ui
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 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
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
i 0.08
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
i: 0.10
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
-------
3
to
o
o
;>
X
OJ
to
o
J>
"T1
H
6
o
0
H
O
o
, 1
w
0
o
H
W
i iiuic j-v.-vj-o ^v_uni u;. ociisuiiiii ^"vjji 11 lu v^iiuuci ; i ci v.cninc LMSII luuiiun ui iiuiiiiy v^iuiic v uiiiciiii iiiiuiis \\j\tiiif, nuinuci
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 Year
Glacier NP 1989
300298001
(Montana) 1990
963m
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Yellowstone NP 1988
(Wyoming)
560391010 1989
2484m
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
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
i 0.08
0
0
0
0
0
0
0
0
0
0
0
0
0
17
0
0
0
0
0
0
0
i: 0.10
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
-------
3
to
o
o
;>
X
OJ
O
§
H
6
o
0
H
O
o
NW'
H
w
0
o
H
W
i iiuic j-v.-vj-o ^v_uni u;. ociisuiiiii ^"vjji 11 lu v^iiuuci ; i ci v.cninc LMSII luuuun ui iiuuiiy v^iuiic v uiiiciiii iiiiuiis ^|j|jin;< nuinuci
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 Year
Yellowstone NP 1997
(Wyoming)
560391011 1998
2468m
1999
2000
2001
DenaliNP 1988
(Alaska)
022900003 1990
640m
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Badlands NP 1989
460711001
(South Dakota) 1990
730m
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
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
i 0.08
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
i: 0.10
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
-------
3
to
o
o
>
X
OJ
(!/i
"^
o
C
'•TJ
H
6
o
H
O
i iiuic j-v.-vj-o ^v_uni u;. ociisuiiiii ^"vjji 11 lu v^iiuuci ; i ci v.cninc LMSII luuuun ui iiuiiiiy v^iuiic v uiiiciiii iiiiuiis ^|j|jin;< nuinuci
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
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
i 0.08
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
i: 0.10
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
O
HH
H
W
-------
1 > 0.08 ppm were observed at monitoring sites in Redwood NP (CA), Olympic NP (WA), Glacier
2 NP (MT), Denali NP (AK), Badlands (SD), and Custer NF (MT) during the months of April to
3 October. There were eight occurrences of hourly average O3 concentrations > 0.08 ppm from
4 April to October of 1997 at the monitoring site in Theodore Roosevelt NP (ND). However, no
5 hourly average concentrations > 0.08 ppm were observed from April to October in any other
6 year at this site. Except for 1988, the year in which there were major forest fires at Yellowstone
7 NP (WY), the monitoring site located there experienced no hourly average concentrations
8 > 0.08 ppm. Logan (1989) noted that O3 hourly average concentrations rarely exceed 0.08 ppm
9 at remote monitoring sites in the western United States. In almost all cases for the above sites,
10 the maximum hourly average concentration was < 0.075 ppm. The top 10 daily maximum 8-h
11 average concentrations for sites experiencing low maximum hourly average concentrations with
12 a data capture of > 75% are summarized in Table AX3-9. Figure AX3-20 shows the 3-year
13 average of the fourth highest 8-h daily maximum concentration at Olympic (WA), Glacier (MT),
14 Yellowstone (WY), and Denali (AK) National Parks for 1999 to 2001. The highest 8-h daily
15 maximum concentrations do not necessarily all occur during the summer months. For example,
16 at the Yellowstone National Park site, the first three highest 8-h daily maximum concentrations
17 occurred in April and May in 1998, and the fourth highest, 8-h daily maximum concentration did
18 not occur until July of that year. In 1999, the first three highest, 8-h daily maximum
19 concentrations were observed in March and May, and the fourth highest value occurred in April.
20 In 2000, the four highest values occurred in May, June, July, and August.
21 The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) noted that the
22 7-month (April to October) average of the 7-h daily average concentrations (0900 to 1559 hours)
23 observed at the Theodore Roosevelt National Park monitoring site in North Dakota were 0.038,
24 0.039, and 0.039 ppm, respectively, for 1984, 1985, and 1986 and concluded that the range of
25 7-h seasonal averages for the Theodore Roosevelt National Park site was representative of the
26 range of maximum daily 8-h average O3 concentrations that may occur at other fairly clean sites
27 in the United States and other locations in the Northern Hemisphere. However, as shown in
28 Table AX3-9 and Figure AX3-20, the representative (as given by the fourth highest) daily
29 maximum 8-h average O3 concentrations at fairly clean sites in the United States are higher than
30 the 0.038 and 0.039 ppm values cited in the 1996 O3 AQCD, and more appropriate values should
31 be used.
January 2005 AX3-55 DRAFT-DO NOT QUOTE OR CITE
-------
g Table AX3-9. The Top 10 Daily
c
to Site
o
^ Redwood NP
060150002
(California)
235m
Olympic NP
530090012
s> (Washington)
X 125 m
OJ
Oi
o
H
1
O
o
0
H
O
o
H
w
0
o
H
W
Maximum 8-h Average Concentrations
(ppm) for
Sites Experiencing
Low Maximum
Hourly Average Concentrations with Data Capture of > 75%
Year
1988
1989
1990
1991
1992
1993
1994
1989
1990
1991
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
-------
3
Table AX3-9 (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%
<<
to
o
o
^
X
OJ
3
o
>
H
O1
O
0
H
O
O
0
o
1 — I
H
W
Site Year
Glacier NP 1989
300298001
(Montana) 199°
963m
1992
1993
1994
1995
1996
1997
1998
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
-------
3
Table AX3-9 (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%
<<
to
o
o
X
OJ
oo
o
l>
H
6
o
0
H
O
o
0
o
H
W
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
-------
3
Table AX3-9 (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%
<<
to
o
o
X
OJ
O
j>
H
6
o
0
H
O
O
0
o
H
W
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
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
-------
0.06 -
0.05 -
0.04 -
0.03-
0.02-
0.01 -
0 -
4
0.045
i
Olympic
National Park
0.067
4 A
0054 ~
0.051
i i i
Glacier Yellowstone Denali
National Park National Park National Park
Figure AX3-20. Three-year average of the fourth highest 8-h daily maximum
concentration at Olympic (WA), Glacier (MT), Yellowstone (WY),
and Denali (AK) National Parks for 1999 to 2001.
1 It is questionable whether the distributions experienced at sites exhibiting low maximum
2 hourly average concentrations in the western United States are representative of sites in the
3 eastern and midwestern United States because of regional differences in sources of precursors
4 and local meteorological conditions that affect patterns of transport. As described in the 1996 O3
5 AQCD, the O3 monitoring site in the Ouachita National Forest, AR experienced distributions of
6 hourly average concentrations similar to some of the western sites. However, since 1993, this
7 site has seen significant shifts, both increases and decreases, in hourly average concentrations.
8 Figure AX3-21 illustrates the changes that have occurred from 1991 to 2001. The large changes
9 in hourly average O3 concentrations observed at the Ouachita National Forest may indicate that
10 this rural site is influenced by the transport of pollution. Given the high density of sources in the
11 eastern and midwestern United States, it is unclear whether a site could be found in either of
12 these regions that would not be influenced by the transport of O3 from urban areas. Thus, with
13 the exception of the Voyageurs National Park site, observations in this section are limited to
14 those obtained at relatively clean, remote sites in western North America.
15
January 2005
AX3-60
DRAFT-DO NOT QUOTE OR CITE
-------
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Year
Figure AX3-21. Seasonal SUM06 and W126 exposure indices for the Ouachita National
Forest for the period of 1991 to 2001.
1 Evidence for Trends in Ozone Concentrations at Rural Sites in the United States
2 Based on data collected in the 1970s to the middle 1990s, consistent trends in tropospheric
3 O3 at the cleanest sites in the world have not been observed across the Northern Hemisphere.
4 Oltmans et al. (1998) concluded that since the early 1980s, based on four high-latitude stations in
5 Canada, O3 amounts throughout the troposphere at these stations has declined significantly.
6 Changes in O3 concentrations since the early 1970s have been relatively small at two stations in
7 the eastern United States (Whiteface Mountain, NY and Wallops Island, VA). Oltmans et al.
8 (1998) noted that in the eastern United States, the station at Whiteface Mountain showed an
9 overall increase with almost no change after the mid-1980s, consistent with the pattern seen at
10 other sites in mid-latitudes of the Northern Hemisphere. The authors noted that little change was
11 noted in the troposphere throughout the 25-year measurement period at Wallops Island. At two
12 high altitude sites (Zugspitze and Hohenpeissenberg, Germany), Oltmans et al. (1998) noted that
13 O3 amounts increased rapidly into the mid-1980s, but increased less rapidly (and at some other
14 sites, not at all) since then. Increases at a Japanese ozonesonde station (Tsukuba) have been
January 2005
AX3-61
DRAFT-DO NOT QUOTE OR CITE
-------
1 largest in the lower troposphere, but have slowed in the recent decade (Oltmans et al., 1998).
2 Fiore et al. (2002a) pointed out that the trends at Whiteface Mountain, as reported by Oltmans
3 et al. (1998), provide some evidence that during the past two decades background O3
4 concentrations have risen by a few ppb in the surface air over the United States. However, the
5 trends observed at Whiteface Mountain are probably associated with the transport of regional
6 pollution to the site, because the site experiences elevated levels of O3 during the summer
7 months and the area is currently designated as in nonattainment for the 1-h standard.
8 Data from monitoring stations in national parks in the United States show conflicting
9 evidence of trends. Statistically significant increases in 8-h O3 concentrations have occurred
10 during the period of 1992 to 2001 (U.S. Environmental Protection Agency, 2002) in the Great
11 Smoky Mountains (TN), Craters of the Moon (ID), Mesa Verde (CO), and Mammoth Cave (KY)
12 National Parks. In contrast, statistically significant decreases in 8-h O3 concentrations have
13 occurred over the same time period in the Saguaro (AZ) and Sequoia (CA) National Parks.
14 In addition, no statistically significant changes in the 8-h O3 concentrations have been observed
15 at the remaining 26 national parks where O3 monitoring takes place (U.S. Environmental
16 Protection Agency, 2002). Yellowstone NP (WY) is an apparent exception1 (see Figure
17 AX3-22). Table AX3-10 shows the trends, using the Kendall's tau test (Lefohn and Shadwick,
18 1991), for the 30th percentiles of the hourly average concentrations at four relatively remote
19 national park sites in the western United States. As can be seen from inspection of
20 Table AX3-10, no trends were observed at the 5% significance level for these sites. As an
21 example of high time resolution data obtained at RRMS, hourly average concentrations at
22 Yellowstone NP for 2001 are shown in Figure AX3-23.
23
24
'U.S. EPA (2002) indicated that 8-h O3 concentrations in Yellowstone National Park (WY) had increased
during the period from 1987 to 2001. However, in reanalyses completed for this document, a Mann-Kendall test
using the data collected at Yellowstone AQS sites for the periods 1987 to 1996, 1996 to 2000, and 1987 to 2000
showed, at the 5% confidence level, that an increase did not occur. The monitoring site in Yellowstone was
physically moved in 1996 from AQS (ID 560391010) to AQS (ID 560391011). The difference between this analysis
and that contained in U.S. Environmental Protection Agency (2002) may be due to (a) a change in location of the
actual O3 monitor in 1996 has resulted in two distinct sets of data being generated that should not be combined for
trends analysis and/or (b) the use by the EPA of data for only the 10-year periods, 1991 to 2000 and 1992 to 2001,
provides a different representation of what has actually occurred at the site over the period of record of 1987 to 2001.
January 2005 AX3-62 DRAFT-DO NOT QUOTE OR CITE
-------
Concentration (ppm)
D O O O O O
D 'o 'o O O -^
D N> -P>- CO CO O
x-
X
560391010
^w
1 tl
•c:
i i°
—
a
i •
—
f:
I •
.cz
i •
^—
i *
—
.d
1 — 1
—
c=
1 •
—
n"
i — pr
•JE
E — 1
—
560391011
c:
I — 1
—
£=
^
£=.
1 •
—
xz:
i •
~
O) O5 O5 O5 O) O> O> O>
G> CO O) O3
CO
o
(N
Year
Figure AX.3-22. The fourth highest 8-hour concentration at Yellowstone National Park.
In 1996, the Yellowstone AIRS site (AIRS ID 560391010) was physically
moved (AIRS ID 560391011).
Table AX3-10. Trends (Using Kendall's tau) for the 30th Percentile at National Park
Service Monitoring Sites1
AQSID
22900003
300298001
530090012
560391010
Site Name
Denali NP
Glacier NP
Olympic NP
Yellowstone NP
Monitoring
Period
1988-2000
1989-2000
1987-2000
1988-1996
Median Difference
(ppm/year)
-0.00014286
-0.00023611
0.00016667
0.00070833
Median Relative
Level (%/year)
-0.59524
-1.63360
1.38889
2.48457
Level of Significance
of Difference
0.141
0.098
0.141
0.089
'Years in which data capture was > 75%.
1 Lin et al. (2000) examined the long-term trend of background O3 in surface air over the
2 United States from 1980 to 1998 using monthly probability distributions of daily maximum 8-h
3 average concentrations at a large collection of rural sites in the AQS database. As shown in
4 Table AX3-11, reported O3 concentrations decreased at the high end of the frequency
January 2005
AX3-63
DRAFT-DO NOT QUOTE OR CITE
-------
ooooooooooooooooooooooooooo
Time
Figure AX3-23. Hourly average O3 concentrations at Yellowstone National Park,
Wyoming for the period of January to December 2001.
Source: U.S. Environmental Protection Agency (2003a).
1 distribution but increased at the lower end of the distribution. The increase was statistically
2 significant at the 5% level in spring and fall. The authors hypothesized that the increase was due
3 to the long-range transport of pollutants from Asia. The largest increase observed by Lin et al.
4 (2000) was in the northeastern United States which, as they noted, was inconsistent with the
5 hypothesis that the increase was attributable to transport from Asia. It is interesting to note that
6 many of the rural sites (identification of sites provided by Lin, included in the Lin et al. (2000)
7 analysis (e.g., Crestline [CA], Rockdale [GA], Edgewood [MD], Camden [NJ], Guilford [NC],
8 and Sumner [TN]) are heavily influenced by transport from polluted areas: 66% of the O3
9 monitoring sites used in the Lin et al. (2000) analysis exceeded the national 8-h O3 standard for
10 the 3-year period from 1998 to 2000. It should be noted that in the AQS database, the land use
11 designation of rural does not mean that the site is mostly isolated from the long-range transport
12 of episodic occurrences of O3 concentrations that occur in or near urban areas. As noted in the
13 1996 O3 AQCD, a land use characterization of rural does not imply that a specific location is
14 isolated from anthropogenic influences.
January 2005
AX3-64
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-11. Long-Term Trend of Background Ozone
AQSID
10735002
40132001
40191018
51191002
60012001
60652002
60690002
60710005
60731006
60830008
60833001
61113001
80013001
90131001
121035002
132151003
132470001
171192007
180970042
201730001
210150003
210670001
210910012
220331001
220470002
220950002
230052003
240030014
240251001
Site Name
Jefferson Co.
Glendale
Tucson
North Little Rock
Hayward
Indio
Hollister
Crestline
San Diego Co.
Capitan
Santa Barbara Co.
El Rio
Welby
Stafford
Tarpon Springs
Muscogee Co.
Rockdale Co.
Madison Co.
Indianapolis
Sedgwick Co.
Boone Co.
Fayette Co.
Hancock Co.
E. Baton Rge, PA
Iberville, PA
St John Bap., PA
Cape Elizabeth
Anne Arundel Co.
Edgewood
County
Jefferson Co.
Maricopa Co.
Pima Co.
Pulaski Co.
Alameda Co.
Riverside Co.
San Benito Co.
San Bernardino
San Diego Co.
Santa Barbara Co.
Santa Barbara Co.
Ventura Co.
Adams Co.
Tolland Co.
Pinellas Co.
Muscogee Co.
Rockdale Co.
Madison Co.
Marion Co.
Sedgwick Co.
Boone Co.
Fayette Co.
Hancock Co.
E. Baton Rge, PA
Iberville, PA
St John Bap., PA
Cumberland Co.
Anne Arundel Co.
Harford Co.
Land
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Use
- Residential
- Residential
- Desert
- Forest
- Residential
- Residential
- Residential
- Residential
- Agricultural
- Agricultural
- Agricultural
- Forest
- Residential
- Agricultural
- Agricultural
- Agricultural
- Agricultural
- Agricultural
- Agricultural
- Agricultural
- Residential
- Agricultural
- Residential
- Industrial
- Residential
- Agricultural
- Commercial
8-h Avg (1998-2000)
0.092
0.075
0.072
0.087
0.090
0.073
0.146
0.100
0.063
0.066
0.068
0.072
0.089
0.082
0.093
0.111
0.086
0.088
0.077
0.086
0.076
0.089
0.094
0.088
0.087
0.077
0.107
0.1
January 2005
AX3-65
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-11 (cont'd). Long-Term Trend of Background Ozone
AQSID
240313001
240330002
250171801
260370001
260492001
260812001
290470003
290470005
291831002
291890006
310550032
340010005
340071001
340190001
340273001
350011012
350130008
360310002
360631006
360650004
361173001
370810011
371191005
371191009
390230001
390610010
391331001
391351001
401430174
Site Name
Rockville
Greenbelt
Sudbury
Clinton Co.
Genesee Co.
Kent Co.
Clay Co.
Clay Co.
St Charles Co.
St Louis Co.
Omaha
Atlantic Co.
Camden Co.
Flemington
Morris Co.
Bernalillo Co.
Dona Ana Co.
Essex Co.
Niagara Co.
Oneida Co.
Wayne Co.
Guilford Co.
Mecklenburg Co.
Mecklenburg Co.
Clark Co.
Hamilton Co.
Portage Co.
Preble Co.
Glenpool
County
Montgomery Co.
Prince Georges
Middlesex Co.
Clinton Co.
Genesee Co.
Kent Co.
Clay Co.
Clay Co.
St Charles Co.
St Louis Co.
Douglas Co.
Atlantic Co.
Camden Co.
Hunterdon Co.
Morris Co.
Bernalillo Co.
Dona Ana Co.
Essex Co.
Niagara Co.
Oneida Co.
Wayne Co.
Guilford Co.
Mecklenberg Co.
Mecklenberg Co.
Clark Co.
Hamilton Co.
Portage Co.
Preble Co.
Tulsa Co.
Land
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Use
- Residential
- Agricultural
- Agricultural
- Agricultural
- Agricultural
- Residential
- Agricultural
- Agricultural
- Residential
- Agricultural
- Residential
- Commercial
- Agricultural
- Agricultural
- Desert
- Agricultural
- Agricultural
- Forest
- Agricultural
- Residential
- Industrial
- Agricultural
- Agricultural
- Industrial
- Agricultural
- Agricultural
- Agricultural
8-h Avg (1998-2000)
0.09
0.099
0.085
0.079
0.086
0.087
0.086
0.089
0.094
0.090
0.071
0.090
0.101
0.098
0.096
0.075
0.073
0.08
0.085
0.073
0.081
0.094
0.098
0.104
0.093
0.085
0.093
0.08
0.081
January 2005
AX3-66
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-11 (cont'd). Long-Term Trend of Background Ozone
1
2
3
4
5
AQS ID Site Name County
410050004 Clackamas Co. Clackamas Co.
410090004 Columbia Co. Columbia Co.
420430401 Harrisburg Dauphin Co.
420990301 Peny Co. Peny Co.
450150002 Berkeley Co. Berkeley Co.
450230002 Chester Co. Chester Co.
450370001 EdgefieldCo. Edgefield Co.
450791002 RichlandCo. RichlandCo.
470370026 Nashville-Davidson Davidson Co.
470650028 Chattanooga Hamilton Co.
47065 1011 Hamilton Co. Hamilton Co.
471571004 Shelby Co. Shelby Co.
471632002 Sullivan Co. Sullivan Co.
471650007 SumnerCo. SumnerCo.
510410004 Chesterfield Co. Chesterfield Co.
510850001 Hanover Co. Hanover Co.
530330010 King Co. King Co.
551390011 Oshkosh Winnebago Co.
Source: Lin et al. (2000).
Land Use 8-h Avg (1998-2000)
Rural - Agricultural 0.072
Rural - Agricultural 0.056
Rural - Commercial 0.090
Rural 0.085
Rural - Industrial 0.08 1
Rural - Commercial 0.088
Rural - Agricultural 0.085
Rural - Agricultural 0.095
Rural -Forest 0.091
Rural - Forest 0.097
Rural - Agricultural 0.097
Rural - Agricultural 0.097
Rural - Residential 0.09 1
Rural - Industrial 0.100
Rural - Residential 0.087
Rural - Agricultural 0. 100
Rural - Forest 0.063
Rural - Agricultural 0.076
Using a 15-year record of O3 from Lassen 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 to the present. They concluded that this pos
itive trend is due to increases of emissions of
O3 precursors in Asia. They found positive trends in O3 occurring during 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
January 2005
AX3-67
DRAFT-DO NOT QUOTE OR CITE
-------
1 or urban centers, the site experiences maximum hourly average O3 concentrations above
2 0.080 ppm during April to May and above 0.100 ppm during the summertime (U.S.
3 Environmental Protection Agency, 2003a), suggesting local photochemical production, at least
4 during summer. However, local springtime photochemical production cannot be ruled out. The
5 authors suggested that the likely cause for the spring increases is transport from Asia, because
6 emissions of precursors have decreased in California over the monitoring period. The
7 springtime increases appears to be inconsistent with the summer increases, when there is
8 evidence for the occurrence of more localized photochemical activity. Although emissions of O3
9 precursors may have decreased in California as a whole over the monitoring period, there still
10 may be regional increases in areas that could affect air quality in Lassen.
11 Thus, although there is evidence as reported in the literature that some locations may be
12 experiencing increased levels of O3 at some rural locations, there is also evidence that O3
13 concentrations at RRMS have not experienced increasing levels for the period of record.
14
15
16 AX3.3 DIURNAL PATTERNS IN OZONE CONCENTRATION
17 AX3.3.1 Introduction
18 By definition, diurnal variations are those that occur during a 24-h period. Diurnal patterns
19 of O3 may be expected to vary with location, depending on the balance among the many factors
20 affecting O3 formation, transport, and destruction. Although they vary with locality, diurnal
21 patterns for O3 typically show a rise in concentration from low (or near minimum detectable
22 amounts) to an early afternoon peak at lower elevation monitoring sites. The diurnal pattern of
23 O3 concentrations has been ascribed to three simultaneous processes: (1) downward transport of
24 O3 from layers aloft, (2) destruction of O3 through contact with surfaces and through reaction
25 with NO at ground level, and (3) in situ photochemical production of O3 (U.S. Environmental
26 Protection Agency, 1978; Coffey et al., 1977; Mohnen et al., 1977; Reiter, 1977a).
27 The form of an average diurnal pattern may provide information on sources, transport, and
28 chemical formation and destruction effects at various sites (Lefohn, 1992). Atmospheric
29 conditions leading to limited transport from source regions will produce early afternoon peaks.
30 However, long-range transport processes will influence the actual timing of a peak, from
January 2005 AX3-68 DRAFT-DO NOT QUOTE OR CITE
-------
1 afternoon to evening or early morning hours. Investigators have used diagrams that illustrate
2 composite diurnal patterns as a means to describe qualitatively the differences in O3 exposures
3 between sites (Lefohn and Jones, 1986; Bohm et al., 1991). Monitoring data programs, such as
4 the National Air Pollution Background Network (NAPBN), the Electric Power Research
5 Institute's Sulfate Regional Air Quality Study, and the Mountain Cloud Chemistry Program
6 (U.S. Environmental Protection Agency, 1996a), provide opportunities to characterize O3 diurnal
7 patterns at diverse sites. For example, the Mountain Cloud Chemistry Program characterized
8 hourly O3 concentrations at Shenandoah National Park and Whiteface Mountain at different
9 elevations. This information is very useful when characterizing the diurnal patterns in similar
10 locations, but at different elevations. Although it might appear that composite diurnal pattern
11 diagrams could be used to quantify the differences in O3 exposures between sites, caution in their
12 use for this purpose is urged (Lefohn et al., 1991). The average diurnal patterns are derived from
13 long-term calculations of the hourly average concentrations, and the resulting diagram cannot
14 adequately identify, at most sites, the presence of high hourly average concentrations and, thus,
15 may not adequately distinguish O3-exposure differences among sites. Logan (1989) noted that
16 diurnal variation of O3 did not reflect the presence of high hourly average concentrations.
17 Unique families of diurnal average profiles exist, and it is possible to distinguish between
18 two types of O3 monitoring sites. A seasonal diurnal diagram provides the investigator with the
19 opportunity to identify whether more scavenging occurs at a specific O3 monitoring site than at
20 any other site. Ozone is rapidly depleted near the surface below the nocturnal inversion layer
21 (Berry, 1964). Mountainous sites, which are above the nocturnal inversion layer, do not
22 necessarily experience this depletion (Stasiuk and Coffey, 1974). Taylor and Hanson (1992)
23 reported similar findings, using data from the Integrated Forest Study. The authors reported that
24 intraday variability was most significant for the low-elevation sites due to the pronounced daily
25 amplitude in O3 concentration between the predawn minimum and mid-afternoon-to-early
26 evening maximum. The authors reported that interday variation was more significant in the
27 high-elevation sites. Ozone trapped below the inversion layer is depleted by dry deposition and
28 chemical reactions if other reactants are present in sufficient quantities (Kelly et al., 1984).
29 Above the nocturnal inversion layer, dry deposition does not generally occur, and the
30 concentration of O3 scavengers is generally lower, so O3 concentrations remain fairly constant
January 2005 AX3-69 DRAFT-DO NOT QUOTE OR CITE
-------
1 (Wolff et al., 1987). A flat diurnal pattern is usually interpreted as indicating a lack of efficient
2 scavenging of O3 or a lack of photochemical precursors, whereas a varying diurnal pattern is
3 taken to indicate the opposite.
4 As indicated above, with the composite diagrams alone, it is difficult to quantify the daily
5 or long-term exposures of O3. For example, the diurnal patterns for two such sites are illustrated
6 in Figure AX3-24. The Jefferson County, KY site, characterized as a suburban-residential site in
7 the AQS database, is urban-influenced and experiences elevated levels of O3 and NOX. The
8 Oliver County, ND site, characterized as rural-agricultural, is fairly isolated from
9 urban-influenced sources and hourly average O3 concentrations are mostly < 0.09 ppm. The flat
10 diurnal pattern observed for the Oliver County site is usually interpreted as indicating a lack of
11 efficient scavenging of O3 or a lack of photochemical precursors, whereas the varying diurnal
12 pattern observed at the Jefferson County site may be interpreted to indicate the opposite. Logan
13 (1989) described the diurnal pattern for several rural sites in the United States (Figure AX3-25)
14 and noted that average daily profiles showed a broad maximum from about 1200 hours until
15 about 1800 hours at all the eastern sites, except for the peak of Whiteface Mountain, NY. Logan
16 (1989) noted that the maximum concentrations were higher at the Sulfate Regional Experiment
17 Program (SURE) sites than at the Western NAPBN sites in the East, because the latter were
18 situated in more remote or coastal locations.
19 An analysis that identified when the highest hourly average concentrations were observed
20 at rural agricultural and forested sites was described in 1996 O3 AQCD. A review of the hourly
21 average data collected at all rural agricultural and forested sites in Environmental Protection
22 Agency's AQS database for 1990 to 1992 was undertaken to evaluate the percentage of time
23 hourly average concentrations > 0.1 ppm occurred during the period of 0900 to 1559 hours in
24 comparison with the 24-h period. It was found that 70% of the rural-agricultural and forested
25 sites used in the analysis experienced at least 50% of the occurrences > 0.1 ppm during the
26 period of 0900 to 1559 hours when compared to the 24-h period. When O3 monitoring sites in
27 California were eliminated, approximately 73% of the remaining sites experienced at least 50%
28 of the occurrences > 0.10 ppm during the daylight 7-h period when compared with the
29 24-h period.
30
January 2005 AX3-70 DRAFT-DO NOT QUOTE OR CITE
-------
Jefferson Co. KY
Oliver Co. ND
9 11 13 15 17 19 21 23
Hour of Day
Figure AX3-24. The comparison of the seasonal (April-October) diurnal patterns for
urban-influenced (Jefferson County, KY) and a rural-influenced
(Oliver County, ND) monitoring sites using 2002 hourly data.
1 AX3.3.2 Diurnal Patterns in Urban Areas
2 The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) discussed diurnal
3 patterns for urban sites. Figure AX3-26a,b shows the diurnal pattern of O3 concentrations on
4 August 3, 2004 at several sites in Maryland. On this day, a peak 1-h average concentration of
5 0.20 ppm, the highest for the month, was reached at 1400 hours, presumably as the result of
6 meteorological factors, such as atmospheric mixing and local photochemical processes. The
7 severe depression of concentrations to below detection limits (< 0.005 ppm) between 0300 and
8 0600 hours usually is explained as resulting from the scavenging of O3 by local NO emissions.
9 In this regard, this station is typical of most urban locations.
10 Diurnal profiles of O3 concentrations can vary from day to day at a specific site, because of
11 changes in the various factors that influence concentrations. Composite diurnal data (i.e.,
12 concentrations for each hour of the day averaged over multiple days or months) often differ
January 2005
AX3-71
DRAFT-DO NOT QUOTE OR CITE
-------
.Q
Q.
a.
ou
"10
30
20
m
JLJ MT
A "7 ^jf
^~~~~?~S^ ~-
-"*'- ,x ^ / ^~~~~ ~~\ ~^ — - ~
I I I I I
60
50
40
30
60
50
40
30
20
10
0
80
70
60
50
40
30
20
10
0
.b]
_
I 1 1 I 1
WFM, NY
—
I I I I I
I
I
I
8 12 16
T\me of Day (h)
20
24
Figure AX3-25a-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).
January 2005
AX3-72
DRAFT-DO NOT QUOTE OR CITE
-------
a.
O Piney run
D South Carroll
V Millington
A Southern Maryland
b.
OOOOOOOOOOoOOOOOOOOOOOO
OOOOOOOOOOOoOOOOOOOOOOOO
Figure AX3-26a,b. Diurnal pattern of 1-h O3 concentration on August 3, 2004 at various
sites in Maryland.
Source: Piety (2004).
January 2005
AX3-73
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
markedly from the diurnal cycle shown by concentrations for a specific day. In Figures AX3-27
through AX3-29, diurnal data for 2 consecutive days are compared with composite diurnal
data (1-month averages of hour-by-hour measurements) at three different kinds of sites:
(1) center city-commercial (Washington, DC); (2) rural-near urban (St. Louis, MO); and
(3) suburban-residential (Alton, IL). Several obvious points of interest present themselves in
these figures: at some sites, peaks can occur at virtually any hour of the day or night, but these
peaks may not show up strongly in the longer-term average data; some sites may be exposed to
multiple peaks during a 24-h period; and disparities, some of them large, can exist between
peaks (the diurnal data) and the 1-month average (the composite diurnal data) of hourly O3
concentrations.
\J. \£.
0.10
•p 0.08
E
Q.
Q.
""^ 0 06
c
O
N
O 0.04
0.02
0
I
1
1
i
i
I
1 1 1
i
i i i
1
Center City - Commercial
—
*
,'
_
i
.^
' ».<
.,_
^
I
-
; ,'
\ !
\ :'
\J
1 -
~i
I
• /
/
/
/
(
- 1
,
*
\
^-^_
i
' t
i,
1
ly 8-9
nonth
i
i
i
i
|
;
*t
,
I
f,^
-t
i
/
!
*
I ,'
i J
* '
^
. |
~" ^
J~^
- 1 -
\ l
';
\
*
i
\
\ f'
\y
i -
*
/ /
* /
; /
1
j
- 1
^"^
^.''
—
•••»
V J
^!
\
>\
\
i i i -i
_i
,-'
'"'
^
. |
24 4 8 12 16 20 24 4 8 12 16 20 24
-a.m. — Noon — p.m. —>k—a.m.—Noon—p.m.-
Figure AX3-27. Diurnal and 1-month composite diurnal variations in O3 concentrations,
Washington, DC July 1981.
Source: U.S. Environmental Protection Agency (1986).
January 2005
AX3-74
DRAFT-DO NOT QUOTE OR CITE
-------
0.12
0.10
•p 0.08
Q.
Q.
d) 0.06
c
o
N
O 0.04
0.02
\ \ \ I I I I I I I I I I I I I I I I I I I T
Rural - Near Urban
Sept. 29-30
• 1 month
0 I_L
\ TT
24 4 8 12 16 20 24 4 8 12 16 20 24
|« a.m. — Noon—p.m. —»|«—a.m.—Noon—p.m. »|
Figure AX3-28. Diurnal and 1-month composite diurnal variations in O3 concentrations,
St. Louis County, MO September 1981.
Source: U.S. Environmental Protection Agency (1986).
0.12
0.10
•=• 0.08
Q.
Q.
^ 0.06
c
o
N
O 0.04
0.02
i i i i i i irnrni i i i i i
Suburban - Residential
Oct. 11-12
1
\\ r\
24 4 8 12 16 20 24 4 8 12 16 20 24
a.m. — Noon — p.m. — »|< — a.m. — Noon — p.m. — >|
Figure AX3-29. Diurnal and 1-month composite diurnal variations in O3 concentrations,
Alton, IL October 1981 (fourth quarter).
Source: U.S. Environmental Protection Agency (1986).
January 2005
AX3-75 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
When diurnal or short-term composite diurnal O3 concentrations are compared with
longer-term composite diurnal O3 concentrations, the peaks are smoothed as the averaging
period is lengthened. Figure AX3-30 demonstrates the effects of lengthening the period of time
over which values are averaged. This figure shows a composite diurnal pattern calculated on a
3-month basis. Although seasonal differences are observed, the comparison of the 3-month
average of the fourth quarter (Figure AX3-30) with 1-month composite diurnal concentrations
for October (Figure AX3-29) at the Alton, IL site readily demonstrates the smoothing out of
peak concentrations as the averaging period is lengthened. As indicated in the 1996 O3 AQCD,
the smoothing of the higher average concentrations using longer-term averages makes it difficult
to identify those exposure regimes that affect human health and vegetation.
0.10
0.09
0.08
—» °'07
Q. 0.06
Q.
7 0.05
| 0.04
o
0.03
0.02
0.01
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
Alton, IL
3j:d Quarter
' ~~"~
-/ ,,--' 4th Quarter X
I I I I I I I I I I I I I I I I I I I I I I I
24 4 8 12 16 20 24 4 8 12 16 20 24
|< a.m. Noon p.m. >|
Figure AX3-30. Composite diurnal patterns of O3 concentrations by quarter, Alton, IL
October 1981.
Source: U.S. Environmental Protection Agency (1986).
January 2005
AX3-76
DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.3.3 Diurnal Patterns in Nonurban Areas
2 Nonurban areas only marginally affected by transported O3 usually have a flatter diurnal
3 profile than sites located in urban areas. Furthermore, nonurban O3 monitoring sites experience
4 differing types of diurnal patterns (Bohm et al., 1991; Lefohn, 1992). As indicated earlier, O3
5 concentrations at a specific location are influenced by local emissions and by long-range
6 transport from both natural and anthropogenic sources. Thus, considerable variation of O3
7 exposures is found among sites characterized as agricultural or forested, with no preference for
8 maximum diurnal patterns to occur in either the second or third quarter.
9 The diurnal patterns for several agricultural sites have been characterized (U.S.
10 Environmental Protection Agency, 1996a). Figures AX3-31 and AX3-32 show some typical
11 patterns of exposure. As discussed in the U.S. Environmental Protection Agency (1996a), the
12 six sites, whose diurnal patterns are illustrated in Figure AX3-31, represent counties with high
13 soybean, wheat, or hay production. The two figures show a distinct afternoon maximum with
14 the lowest concentrations occurring in the early morning and evening hours. Quarterly
15 composite diurnal patterns clearly show the division of the afternoon O3 concentrations into two
16 seasonal patterns: the low, "winter," levels in the first and fourth quarters of the year; and the
17 high, "summer," levels in the second and third quarters.
18 Remote forested sites experience unique patterns of O3 concentrations (Evans et al., 1983;
19 Lefohn, 1984). Figure AX3-33 shows diurnal patterns for several national forest sites in the
20 EPA AQS database for 2002. Several of the sites analyzed exhibit fairly flat average diurnal
21 patterns. Such a pattern is based on average concentrations calculated over an extended period,
22 and caution is urged in drawing conclusions concerning whether some monitoring sites
23 illustrated in the figure experience higher cumulative O3 exposures than other sites. Variation in
24 O3 concentration occurs from hour to hour on a daily basis, and, in some cases, elevated hourly
25 average concentrations are experienced either during daytime or nighttime periods (Lefohn and
26 Mohnen, 1986; Lefohn and Jones, 1986; Logan, 1989; Lefohn et al., 1990a; Taylor et al., 1992).
27 Because the diurnal patterns represent averaged concentrations calculated over an extended
28 period, the smoothing of the averaging tends to mask the elevated hourly average concentrations.
29 Lefohn et al. (1990b) characterized O3 concentrations at high-elevation monitoring sites.
30 The authors reported that a fairly flat diurnal pattern for the Whiteface Mountain summit site
January 2005 AX3-77 DRAFT-DO NOT QUOTE OR CITE
-------
?
£
0)
o
N
O
0.10
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
°2
a. N. Little Rock, AR
I I I I I I I I I I I I I I I
~~ 2nd Quarter
— 3rd Quarter
: s<~'~
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
42 4 6 8 10 12 14
I I I I I I I I
~~
V.
I I I I I I I I
16 18 20 22 2
0.10
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
b. Bakersfield, CA
o.
i i i i i i i i i i i i i i i i i i i i i i i
24 2 4 6 8 10 12 14 16 18 20 22 24
-a.m. H«-
-p.m.
-a.m. H«-
-p.m.
Q.
£
|«-
-p.m.
Hour
Hour
Figure AX3-31a-f. Quarterly composite diurnal patterns of O3 concentrations at selected
sites representing potential for exposure of major crops, 1981.
Source: U.S. Environmental Protection Agency (1986).
January 2005
AX3-78 DRAFT-DO NOT QUOTE OR CITE
-------
0.06
^ 0.04
Q.
Q.
fO
o
0.02
0.00
24
a.m.
12
p.m.
24
Figure AX3-32. Composite diurnal O3 pattern at a rural National Crop Loss Assessment
Network site in Argonne, IL August 6 through September 30,1980.
Source: U.S. Environmental Protection Agency (1986).
Acadia NP
Great Smoky Mtn. NP
Glacier NP
Yellowstone NP
Grand Canyon NP
1 3 5 7 9 11 13 15 17 19 21 23
Time of Day
Figure AX3-33. 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 (2003a).
January 2005
AX3-79
DRAFT-DO NOT QUOTE OR CITE
-------
1 (WF1) was observed (Figure AX3-34a), with the maximum hourly average concentrations
2 occurring in the late evening or early morning hours. A similar pattern was observed for the
3 mid-elevation site at Whiteface Mountain (WF3). The site at the base of Whiteface Mountain
4 (WF4) showed the typical diurnal pattern expected from sites that experience some degree of O3
5 scavenging. More variation in the diurnal pattern for the highest Shenandoah National Park sites
6 occurred than for the higher elevation Whiteface Mountain sites, with the typical variation for
7 urban-influenced sites in the diurnal pattern at the lower elevation Shenandoah National Park site
8 (Figure AX3-34b). Aneja and Li (1992), in their analysis of the five high-elevation Mountain
9 Cloud Chemistry Program (MCCP) sites, noted the presence of the flat diurnal pattern typical of
10 high-elevation sites that has been described previously in the literature. Aneja and Li (1992)
11 noted that the peak of the diurnal patterns over the period of May to October (1986 to 1988)
12 occurred between 1800 and 2400 hours for the five sites, whereas the minimum was observed
13 between 0900 and 1200 hours. However, it is important to note that, as indicated by Lefohn
14 et al. (1990b), the flat diurnal pattern was not observed for all high-elevation sites.
15 As mentioned earlier, nonurban areas only marginally affected by transported O3 usually have a
16 flatter diurnal profile than sites located in urban areas. Nonurban O3 monitoring sites experience
17 differing types of diurnal patterns, as shown in this section. The difference in diurnal patterns
18 may influence the potential for O3 exposures to affect vegetation.
19
20
21 AX3.4 SEASONAL PATTERNS IN OZONE CONCENTRATIONS
22 AX3.4.1 Urban Area Seasonal Patterns
23 Seasonal variations in O3 concentrations in 1981 were described by the U.S. Environmental
24 Protection Agency (1996a). The current forms of the O3 standards focus on both the highest
25 hourly average concentrations (1-h standard) and, for many monitoring sites, the mid-level
26 hourly average concentrations (8-h standard). The description that follows uses the highest
27 hourly average concentration as an indication of exposure. Figure AX3-35 shows the single 1-h
28 maximum concentrations within the month for eight sites across the United States for 2002.
29 Data from all eight monitoring sites show a tendency for the highest maximum hourly average
30 concentration to occur in the June to August period. Because of annual variability, the general
31 weather conditions in a given year may be more favorable for the formation of O3 and other
January 2005 AX3-80 DRAFT-DO NOT QUOTE OR CITE
-------
E
Q_
o>
c
o
N
O
0.06 -\
0.05
0.04 -
0.03
0.02 -
0.01 -
0.00
0.06
0.00
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)
8 10 12 14 16 18 20 22 24
Hour
Figure AX3-34a,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).
January 2005
AX3-81 DRAFT-DO NOT QUOTE OR CITE
-------
0.20
0.18
"P 0.16
Q. 0.14
O. 0.12
a. Phoenix, AZ
2002
0401330019
(U
0.10
= 0.08
N 0.06
O 0.04
0.02
0.00
I
Illllllll.
I
CD CU CD CU
E o E E
CD O CD CD
B. O > o
b. Denver, CO
2002
080310014
016
n ns
0 04
0 00
cT
ra
ro
— j
February
O
03
II
I
I I
I_
• _
II |
| |
| |
1 1
i s t 1
! "" I
Septembe
Octobe
Novembe
Decembe
, Atlanta, GA
2002
131210055
d. Tampa, FL
2002
120571065
0.20
e. Washington, D.C. 2002 110010043
0.20-
f. Boston, MA
2002
250250041
g. Los Angeles, CA 2002 060376012
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
S
E E
Figure AX3-35a-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 (2003a).
January 2005
AX3-82
DRAFT-DO NOT QUOTE OR CITE
-------
1 oxidants than during the prior or following year. For example, 1988 was a hot, dry year during
2 which some of the highest O3 concentrations of the last 16 years occurred, whereas 1989 was a
3 cold, wet year during which some of the lowest concentrations occurred (U.S. Environmental
4 Protection Agency, 1996a).
5
6 AX3.4.2 Seasonal Patterns in Nonurban Areas
7 For the purpose of assessing possible vegetative effects, it is important to characterize the
8 seasons in which the highest O3 concentrations would be expected to occur in nonurban areas.
9 However, for nonurban areas, it is not possible to generalize the quarters in which the highest
10 hourly average concentrations occur. Many RRMS tend to experience their highest average
11 O3 concentrations during the second quarter (i.e., the months of April or May) versus the third
12 quarter of the year (Evans et al., 1983; Singh et al., 1978; Lefohn et al., 2001). This observation
13 has been attributed either to stratospheric intrusions or to an increasing frequency of slow-
14 moving, high-pressure systems that promote the formation of O3 (U.S. Environmental Protection
15 Agency, 1996a). Figure AX3-23 illustrates the hourly average concentrations for Yellowstone
16 National Park (WY) for the period of January to December 2001. Note that at the Yellowstone
17 National Park site, the highest hourly average concentrations tend to occur during the April to
18 May period. Lefohn et al. (2001) and Monks (2000) noted that this was also observed for other
19 RRMS in North America and northern Europe.
20 The highest O3 concentrations tend to occur in rural-forested areas during different times of
21 the year. The different patterns may be associated with the observations by Logan (1989) that
22 spring and summer O3 concentrations in rural areas of the eastern United States is severely
23 impacted by anthropogenic, and possibly natural emissions of NOX and hydrocarbons, and that
24 O3 episodes occur when the weather is particularly conducive to photochemical formation of O3.
25 Taylor et al. (1992) reported that for 10 forest sites in North America, the temporal patterns of
26 O3 during quarterly or annual periods exhibited less definitive patterns. Based on the exposure
27 index selected, different patterns were reported. Meagher et al. (1987) reported that for rural
28 O3 sites in the southeastern United States, the daily maximum 1-h average concentration was
29 found to peak during the summer months. Taylor and Norby (1985) reported that Shenandoah
30 National Park experienced both the highest frequency of episodes and the highest mean duration
31 of exposure events during the month of July.
January 2005 AX3-83 DRAFT-DO NOT QUOTE OR CITE
-------
1 Aneja and Li (1992) reported that the maximum monthly O3 levels at several rural sites
2 occurred in either the spring or the summer (May to August), and the minimum occurred in the
3 fall (September and October). The timing of the maximum monthly values differed across sites
4 and years. However, in 1988, an exceptionally high O3 concentration year occurred, and the
5 highest monthly average concentration occurred in June for almost all of the five sites
6 investigated. June 1988 was also the month in which the greatest number of O3 episodes
7 occurred in the eastern United States.
8 Lefohn et al. (1990a) have characterized the O3 concentrations for several sites in the
9 United States that experience low maximum hourly average concentrations. Of the three
10 western national forest sites evaluated by Lefohn et al. (1990a), only Apache National Forest
11 (AZ) experienced its maximum monthly mean concentration in the spring. The Apache National
12 Forest site was above mean nocturnal inversion height, and no decrease of concentrations
13 occurred during the evening hours. This site also experienced the highest hourly maximum
14 concentration, as well as the highest W126 O3 exposures. The Custer (MT) and Ochoco
15 National Forest (OR) sites experienced most of their maximum monthly mean concentrations in
16 the summer. The White River Oil Shale site in Colorado experienced its maximum monthly
17 mean during the spring and summer months.
18 For vegetation effects assessment purposes, the W126 sigmoidal weighting exposure index
19 was also used to identify the month of highest O3 exposure. A somewhat more variable pattern
20 was observed than when the maximum monthly average concentration was used. For some sites,
21 the winter/spring pattern was present; for others, it was not. In some cases, the highest W126
22 exposures occurred earlier in the year than was indicated by the maximum monthly
23 concentration. For example, in 1979, the Custer National Forest site experienced its highest
24 W126 exposure in April, although the maximum monthly mean occurred in August. In 1980, the
25 reverse occurred.
26 There was no consistent pattern for those sites located in the continental United States.
27 The Theodore Roosevelt NP, Ochoco, and Custer National Forest sites and the White River Oil
28 Shale site experienced their maximum O3 exposures during the spring and summer months. The
29 sites experiencing their highest O3 exposures in the fall-to-spring period did not necessarily also
30 experience the lowest O3 exposures.
31
January 2005 AX3-84 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.5 SPATIAL VARIABILITY IN OZONE CONCENTRATIONS
2 The spatial variability of O3 concentrations in different environments in the United States
3 occurring across a variety of spatial scales is characterized in this section. This information will
4 be useful for understanding the influence of regional or altitudinal differences in O3 exposure on
5 vegetation and for establishing the spatial variations in O3 concentrations as they are used in
6 epidemiologic studies. Intracity variations in O3 concentrations are described in Section
7 AX3.5.1. Differences in O3 concentration between urban and nonurban areas are described in
8 Section AX3.5.2. Ozone concentrations at high elevations are characterized in Section AX3.5.3.
9
10 AX3.5.1 Spatial Variability of Ozone Concentrations in Urban Areas
11 The spatial variability in O3 concentrations in 24 MSAs across the United States is
12 characterized in this section. These areas were chosen to provide analyses to help guide in risk
13 assessments, to provide a general overview of the spatial variability of O3 in different regions of
14 the country, and also to provide insight in to the spatial distribution of O3 in cities where health
15 outcome studies have been conducted. Statistical analyses of the human health effects of
16 airborne pollutants based on aggregate population time-series data have often relied on ambient
17 concentrations of pollutants measured at one or more central sites in a given metropolitan area.
18 In the particular case of ground-level O3 pollution, central-site monitoring has been justified as a
19 regional measure of exposure partly on grounds that correlations between concentrations at
20 neighboring sites measured over time are usually high (U.S. Environmental Protection Agency,
21 1996a). In analyses where multiple monitoring sites provide ambient O3 concentrations, a
22 summary measure such as an averaged concentration has often been regarded as adequately
23 characterizing the exposure distribution. Indeed, a number of studies have referred to
24 multiple-site averaging as the method for measuring O3 exposure (U.S. Environmental
25 Protection Agency, 1996a). It is hoped that the analyses presented here will shed some light on
26 the suitability of this practice. Earlier analyses were reported in the previous O3 AQCD (U.S.
27 Environmental Protection Agency, 1996a). The analyses presented there concluded that the
28 extent of spatial homogeneity is specific to the MSA under study. In particular, cities with low
29 traffic densities that are located downwind of major sources of precursors are heavily influenced
30 by long range transport and tend to show smaller variability (e.g., New Haven, CT) than those
31 source areas with high traffic densities located upwind (e.g., New York, NY).
January 2005 AX3-85 DRAFT-DO NOT QUOTE OR CITE
-------
1 Metrics for characterizing spatial variability include the use of Pearson correlation
2 coefficients (r), values of the 90th percentile (P90) absolute difference in concentrations, and
3 coefficients of divergence (COD)2. These methods of analysis follow those used for
4 characterizing PM25 and PM10_25 concentrations in Pinto et al. (2004) and in the latest edition of
5 the PM AQCD (U.S. Environmental Agency, 2004). However, the calculations were performed
6 on an hourly basis rather than on a 24-h basis. Data were aggregated over the local O3 season as
7 indicated in Table AX3-1. The length of the O3 season varies across the country. In several
8 southwestern states, it lasts all year long. In other areas, such as in New England, it can be
9 6 months long, but typically it lasts from April through October.
10 Table AX3-12 shows the urban areas chosen, the range of 24-h average O3 concentrations
11 over the O3 season, the range of intersite correlation coefficients, the range of P90 differences in
12 O3 concentrations between site pairs, and the range in COD values. A COD of zero implies that
13 values in both data sets are identical, and a COD of one indicates that two data sets are
14 completely different. In general, statistics were calculated for partial MSAs. This was done so
15 as to obtain reasonable lower estimates of the spatial variability that is present, as opposed to
16 examining the consolidated MSAs. In Boston, MA and New York, NY, this could not be readily
17 done, and so statistics were calculated for the consolidated MSAs. More detailed calculations
18 for a subset of nine MSAs are given in Figures AX3-36 through AX3-44.
19 As can be seen, there are no clearly discernible regional trends in the ranges of parameters
20 shown. Additional urban areas would need to be examined to discern broadscale patterns. The
21 data indicate considerable variability in the concentration fields. Seasonal means vary within
22 individual urban areas from factors of 1.4 to 4.0.
23 The highest annual mean O3 concentration (0.058 ppm) is found in the Phoenix, AZ MSA
24 at a site which is located in the mountains well downwind of the main urban area. The lowest
25 annual mean O3 concentration (0.010 ppm) was found in Lynwood in the urban core of the
2 The COD is defined as follows:
jj-:k
CODjk= -V p-^- (AX3-1)
x+x
where xv and xik represent the 24-h average PM2 5 concentration for day /' at site y and site k andp is the number of
observations.
January 2005 AX3-86 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-12. Summary Statistics for Ozone (in ppm) Spatial Variability in Selected U.S. Urban Areas
g>
to
o
o
X
OJ
oo
o
>
H
6
o
0
H
O
o
w
0
o
HH
H
W
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
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
"P90 = 90th nercentile abs<
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
ilute difference in co
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
ncentrations.
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
P a
^90
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
p
^90
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
bCOD = coefficient of divergence for different site pairs.
-------
Charlotte - Gastonia - Rock Hill, NC - SC MSA
a.
c.
40 20
b.
o
40
Kilometers
AIRS Site ID
Site A
Site B
SiteC
SiteD
SiteE
SiteF
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
i
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 AX3-36.
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 observed 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.
1 Los Angeles MSA. CO and NOX monitors at this site recorded the highest concentrations in
2 California, indicating that titration of O3 by NO freshly emitted from tail pipes of motor vehicles
3 is responsible for the low O3 values that are found. Ratios of highest to lowest mean O3
4 concentrations in these two MSAs are among the highest shown in Table AX3-12. Both of these
5 MSAs are characterized by sunny, warm climates; sources of precursors that are associated with
January 2005
DRAFT-DO NOT QUOTE OR CITE
-------
Baton Rouge, LA MSA
c.
25
b.
o
i
Kilometers
AIRS Site ID
Site A 22
Site B 22
Site C 22
Site D 22
Site E 22
Site F 22
Site G 22
005-0004
033-0003
033-0009
033-0013
033-1001
063-0002
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 AX3-37.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Baton Rouge, LA MSA. The mean
observed 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.
1 O3 titration to varying degrees in their urban centers; and with maximum O3 found well
2 downwind of the urban centers. Intersite correlation coefficients show mixed patterns, i.e., in
3 some urban areas all pairs of sites are moderately to highly correlated, while other areas show a
4 very large range of values. As may be expected, those areas which show smaller ratios of
5 seasonal mean concentrations also exhibit a smaller range of intersite correlation coefficients.
6 Within the examined urban areas , P90 values were evenly distributed between all site pairs
January 2005
AX3-89
DRAFT-DO NOT QUOTE OR CITE
-------
Detroit - Ann Arbor - Fling, Ml CMSA
b.
Kilometers
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
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 0.86 086 0.81 0.86 0.82 0.84 0.85
0.012 0019 0.019 0.019 0.020 0.020 0.019 0024 0.022 0.020
020 024 026 027 030 026 027 035 032 029
4333 4151 4253 4341 4103 4338 4163 4228 4324 4247
1 0.84 0.86 0.85 084 0.82 0.84 0.81 0.81 0.84
0019 0.017 0.020 0.023 0.019 0.021 0025 0.024 0.021
0.22 0.25 0.28 033 0.25 0.28 0.37 0.36 0.32
4152 4252 4341 4104 4339 4164 4230 4323 4248
1 0.85 0.86 087 0.78 0.92 0.84 0.86 0.86
0.020 0.020 0.022 0.023 0.015 0025 0.023 0.021
027 028 032 027 025 037 035 032
4072 4159 3930 4156 4164 4049 4143 4068
1 0.91 088 0.86 0.86 0.84 0.87 0.90
0.014 0.019 0.017 0.019 0022 0.020 0.016
0.26 031 0.25 0.27 0.34 0.31 0.27
4260 4022 4259 4086 4148 4245 4193
1 094 0.83 0.89 0.88 0.94 0.96
0014 0020 0017 0020 0013 0010
024 0.27 0.25 0.31 0.24 0.19
4112 4346 4172 4235 4330 4255
1 0.83 0.91 0.85 0.93 0.93
0.021 0.015 0020 0.013 0.014
0.29 0.25 0.31 0.23 0.24
4109 3944 4000 4093 4025
1 0.79 0.76 0.80 0.83
0.022 0024 0.022 0.020
0.28 0.36 0.33 0.29
4169 4233 4329 4255
1 0.88 0.89 0.89
0019 0.017 0.017
0.34 0.29 0.27
4062 4153 4081
1 0.89 0.90
0.016 0.018
029 029
4217 4144
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
0.95
0.011
0.21
4242
Figure AX3-38.
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 observed 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.
January 2005
AX3-90
DRAFT-DO NOT QUOTE OR CITE
-------
St. Louis, MO - IL CMSA
a.
60
I
30
>—- '
0
i
60
i
120
I
Kilometers
b.
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
SiteG
SiteH
Site I
SiteJ
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 AX3-39. Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the St. Louis, MO-IL MSA. The mean
observed 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.
1 considered. The CODs indicate variability among site pairs. However, there are a number of
2 cases where sites in an urban area may be moderately to highly correlated but showed substantial
3 differences in absolute concentrations. In many cases, values for P90 equaled or exceeded
4 seasonal mean O3 concentrations. This was reflected in both values for P90 and for the COD.
5 It is instructive to compare the metrics for spatial variability shown in Table AX3-12 to
6 those calculated for PM2 5 and PM10_2 5 in the PM AQCD (U.S. Environmental Agency, 2004).
7 The values for concentrations and concentration differences are unique to the individual species,
8 but the intersite correlation coefficients and the COD values can be directly compared.
9 In general, the variability in O3 concentrations is larger than for PM2 5 concentrations and
10 comparable to that obtained for PM10_2 5. Intersite correlation coefficients in some areas (e.g.,
11 Philadelphia, PA; Atlanta, GA; Portland, OR) can be very similar for both PM2 5 and for O3.
January 2005
AX3-91
DRAFT-DO NOT QUOTE OR CITE
-------
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
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 AX3-39 (cont'd).
January 2005
AX3-92 DRAFT-DO NOT QUOTE OR CITE
-------
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 AX3-40.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Phoenix-Mesa, AZ MSA. The mean
observed 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.
January 2005
AX3-93
DRAFT-DO NOT QUOTE OR CITE
-------
Fresno, CA MSA
a.
c.
b.
Site
mean
(ppm)
A
0.031
B
0.031
C
0.047
D
0.040
E
0.036
F
0.030
50
1 1 1 1
0 IMIOIT
leiers 200
i i i
AIRS Site ID
Site A 06-019-0007
SiteB 06-019-0008
SiteC 06-019-0242
SiteD 06-019-4001
Site E 06-019-5001
Site F 06-039-0004
B
D
1 0.97 0.94 0.93 0.95 0.92
0.010 0.027 0.022 0.016 0.015
0.19 0.34 0.25 0.19 0.25
7804 8124 7950 8066 8055
1 0.95 0.92 0.97 0.92
0.026 0.022 0.013 0.017
0.40 0.33 0.23 0.25
7777 7624 7726 7710
1 0.92 0.93 0.94
0.019 0.022 0.027
0.20 0.29 0.40
7939 8055 8036
1 0.92 0.92
0.017 0.024
0.22 0.32
7876 7873
1 0.90
0.022
0.29
7998
1
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
Figure AX3-41.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Fresno, CA MSA. The mean observed 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.
1 However, there is much greater variability in the concentration fields of O3 as evidenced by the
2 much higher COD values. Indeed, COD values are higher for O3 than for PM25 in each of the
3 urban areas examined. In all of the urban areas examined for O3 some site pairs are always very
4 highly correlated with each other (i.e., r > 0.9) as seen for PM2 5. These sites also show less
5 variability in concentration and are probably influenced most strongly by regional production
6 mechanisms.
January 2005
AX3-94
DRAFT-DO NOT QUOTE OR CITE
-------
Bakersfield, CA MSA
a.
c.
b.
®
o
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
Site G 06-029-5001
Site H 06-029-6001
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
H
1 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 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
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
1
Figure AX3-42.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Bakersfield, CA MSA. The mean
observed 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.
1 A number of processes can contribute to spatial variability in O3 concentrations in urban
2 areas. Ozone formation occurs more or less continuously downwind of sources of precursors,
3 producing a gradient in O3 concentrations. Ozone 'titration' by reaction with NO can deplete O3
4 levels near NO sources such as highways and busy streets. Differences in surface characteristics
5 affect the rate of deposition of O3. Mixing of O3 from aloft can lead to local enhancements in O3
6 concentration.
January 2005
AX3-95
DRAFT-DO NOT QUOTE OR CITE
-------
a.
Los Angeles - Orange County, CA CMS A
b.
50
I
100
I
Kilometers
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
SiteG
SiteH
Site I
Site J
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 AX3-43.
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
observed 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.
January 2005
AX3-96
DRAFT-DO NOT QUOTE OR CITE
-------
Los Angeles - Orange County, CA CMSA
c.
Site
mean
(ppm)
p
A B
0.021 0.029
1 0.95
0.017
0.32
8378
1
C
0.021
0.87
0.020
0.33
6723
082
0029
036
6722
1
D
0.020
0.64
0.028
0.41
8387
060
0034
042
8386
075
0021
032
6731
E
0.018
0.88
0.018
0.38
8202
086
0028
0 44
8201
088
0018
035
6546
F
0.019
0.85
0.020
0.36
8292
080
0029
045
8291
095
0010
022
6636
G H
0.021 0.013
0.82 0.72
0.021 0.029
0.37 0.44
8364 8362
0 80 0 67
0027 0038
0 42 0 53
8363 8361
0 84 0 83
0019 0020
0 29 0 37
6708 6729
I
0.016
0.89
0.019
0.41
8381
084
0030
052
8380
091
0013
037
6725
J
0.013
0.91
0.022
0.46
8386
0 88
0032
0 59
8385
0 85
0020
0 49
6730
K
0.021
0.91
0.015
0.33
8384
087
0024
039
8383
090
0017
024
6728
L
0.022
0.61
0.029
0.44
8379
057
0033
047
8378
0 72
0022
031
6723
M N
0.027 0.042
0.55 0.58
0.035 0.048
0.45 0.47
8341 6321
0 53 0 56
0034 0042
0 46 0 42
8340 6321
0 68 0 61
0030 0048
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
Figure AX3-43 (cont'd).
January 2005
AX3-97
DRAFT-DO NOT QUOTE OR CITE
-------
Riverside - Orange County, CS CMSA
a.
b.
50
i i
Kilometers
200
i
AIRS Site ID
Site A
SiteB
SiteC
SiteD
SiteE
SiteF
SiteG
SiteH
Site I
SiteJ
SiteK
SiteL
SiteM
SiteN
SiteO
SiteP
SiteQ
SiteR
06-065-
06-065-
06-065-
06-065-
06-065-
06-065-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
06-071-
0012
2002
5001
6001
8001
9001
0001
0005
0012
0017
0306
1004
1234
2002
4001
4003
9002
9004
Figure AX3-44.
Locations of O3 sampling sites (a) by AQS ID# (b) and intersite
correlation statistics (c) for the Riverside-Orange County, CA CMSA.
The mean observed 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.
January 2005
AX3-98
DRAFT-DO NOT QUOTE OR CITE
-------
Riverside - Orange County, CS CMSA
c.
Site ABODE FGH I JKLMNOPQR
mean
(ppm)
0.0370.0350.0420.033 0.026 0.036 0.0300.0470.048 0.0440.032 0.0220.0380.020 0.0370.032 0.0400.026
M
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 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
Key
Pearson correlation coefficient
90th %-ile difference in concentration
Coefficient of divergence
Number of paired observations
0.45
0.041
0.46
7866
1
Figure AX3-44 (cont'd).
January 2005
AX3-99
DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.5.2 Urban-Nonurban Concentration Differences
2 Research performed and published in the 1970s and 1980s provides an excellent
3 framework for discussing the differences in urban and nonurban concentrations. Diurnal
4 concentration data presented earlier indicate that peak O3 concentrations can occur later in the
5 day in rural areas than in urban, with the distances downwind from urban centers generally
6 determining how much later the peaks occur. Meagher et al. (1987) reported that O3 levels for
7 five rural sites in the Tennessee Valley region of the southeastern United States were found to
8 equal or exceed urban values for the same region. Data presented in the 1978 O3 AQCD
9 demonstrated that peak O3 concentrations in rural areas generally are lower than those in urban
10 areas, but that average concentrations in rural areas are comparable to or even higher than those
11 in urban areas (U.S. Environmental Protection Agency, 1978). Reagan (1984) noted that O3
12 concentrations measured near "population-oriented" areas were depressed compared to more
13 isolated areas. As noted earlier, urban O3 values are often depressed because of titration by NO
14 (Stasiuk and Coffey, 1974). In reviewing the NCLAN's use of kriging to estimate the 7-h
15 seasonal average O3 levels, Lefohn et al. (1987) found that the 7-h values derived from kriging
16 for sites located in rural areas tended to be lower than the actual values due to the use of data
17 from urban areas to estimate rural area values. In addition to the occurrence of higher average
18 concentrations and occasionally higher peak concentrations of O3 in nonurban than in urban
19 areas, it is well documented that O3 persists longer in nonurban than in urban areas (Coffey et al.,
20 1977; Wolff et al., 1977; Isaksen et al., 1978). The absence of chemical scavengers appears to
21 be the main reason. Ozone is a secondary pollutant and its source is distributed more uniformly
22 over much broader spatial scales compared to primary pollutants such as CO, SO2, etc.
23 However, this does not mean that variability in O3 does not exist on smaller spatial scales.
24
25 AX3.5.3 Ozone Concentrations at High Elevations
26 The distributions of hourly average concentrations experienced at high-elevation cities are
27 similar to those experienced in low-elevation cities. For example, the distribution of hourly
28 average concentrations for several O3 sites located in Denver were similar to distributions
29 observed at many low-elevation sites elsewhere in the United States. However, the use of
30 absolute concentrations (e.g., in units of micrograms per cubic meter) in assessing the possible
31 impacts of O3 on vegetation at high-elevation sites instead of mixing ratios (e.g., parts per
January 2005 AX3-100 DRAFT-DO NOT QUOTE OR CITE
-------
1 million) may be an important consideration (see Chapter 9, for further considerations about
2 exposure and effective dose considerations for vegetation assessments).
3 Concentrations of O3 vary with altitude and latitude. Although a number of reports contain
4 data on O3 concentrations at high altitudes (e.g., Coffey et al., 1977; Reiter, 1977b; Singh et al.,
5 1977; Evans et al., 1985; Lefohn and Jones, 1986), fewer reports present data for different
6 elevations in the same locality. Monitoring data collected by the MCCP provide useful
7 information for investigating O3 exposure differences at different elevations. When applying
8 different exposure indices to the MCCP data, there appears to be no consistent conclusion
9 concerning the relationship between O3 exposure and elevation.
10 Lefohn et al. (1990a) summarized the characterization of gaseous exposures at rural sites in
11 1986 and 1987 at several MCCP high-elevation sites. Aneja and Li (1992) have summarized the
12 O3 concentrations for 1986 to 1988. Table AX3-13 summarizes the sites characterized by
13 Lefohn et al. (1990a). Table AX3-14 summarizes the concentrations and exposures that
14 occurred at several of the sites for the period 1987 to 1988. In 1987, the 7- and 12-h seasonal
15 means were similar at the Whiteface Mountain WF1 and WF3 sites (Figure AX3-45a). The 7-h
16 mean values were 0.0449 and 0.0444 ppm, respectively, and the 12-h mean values were 0.0454
17 and 0.0444 ppm, respectively. Note that, in some cases, the 12-h mean was slightly higher than
18 the 7-h mean value. This resulted when the 7-h mean period (0900 to 1559 hours) did not
19 capture the period of the day when the highest hourly mean O3 concentrations were experienced.
20 A similar observation was made, using the 1987 data, for the MCCP Shenandoah National Park
21 sites. The 7-h and 12-h seasonal means were similar for the SHI and SH2 sites (Figure
22 AX3-45b). Based on cumulative indices, the Whiteface Mountain summit (1,483-m) site (WF1)
23 experienced a higher exposure than the WF3 (1,026-m) site (Figure AX3-45c). Both the sum of
24 the concentrations > 0.07 ppm (SUM07) and the number of hourly concentrations > 0.07 ppm
25 were higher at the WF1 site than at the WF3 site. The site at the base of the mountain (WF4)
26 experienced the lowest exposure of the three O3 sites. Among the MCCP Shenandoah National
27 Park sites, the SH2 site experienced marginally higher O3 exposures, based on the index that
28 sums all of the hourly average concentrations (referred to as "total dose" in Figure AX3-45c) and
29 sigmoidal values, than the SHI high-elevation site (Figure AX3-45d). The reverse was true for
30 the sums of the concentrations > 0.07 ppm and the number of hourly concentrations > 0.07 ppm.
31
January 2005 AX3-101 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-13. Description of Mountain Cloud Chemistry Program Sites
c
to
o
o
X
OJ
o
to
o
H
6
O
0
H
O
O
0
o
HH
H
W
Site
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 (MM1), NC
Mt. Mitchell (MM2), NC
Elevation (m) Latitude
65 458° 11'
1,000 438° 59' 18"
1483 448° 23' 26"
1,015 38° 37' 12"
716 38° 37' 30"
524 38° 37' 45"
1,689 36° 38' 20"
2006 35° 44' 15"
1760 35° 45'
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 AX3-14. Seasonal (April-October) Percentiles, SUM06, SUM08, and W126 Values for the MCCP Sites
to
o
>
X
LtJ
I
O
OJ
O
!*
H
6
o
0
H
O
o
w
0
o
HH
H
W
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.
.000
.000
.006
.010
.011
.014
.010
.000
.008
.011
.010
.005
.017
.009
.000
.006
.003
.006
10
.013
.012
.027
.026
.029
.025
.025
.011
.034
.038
.038
.036
.032
.029
.023
.024
.027
.029
30
.021
.021
.036
.033
.037
.033
.033
.023
.044
.054
.047
.043
.042
.041
.036
.036
.040
.042
50
.028
.028
.045
.043
.046
.043
.039
.031
.051
.065
.054
.048
.049
.050
.044
.047
.049
.054
70
(ppm)
.035
.036
.053
.055
.053
.056
.047
.041
.058
.075
.059
.053
.056
.060
.054
.058
.059
.064
90
.046
.047
.065
.076
.067
.078
.064
.056
.067
.095
.068
.063
.067
.080
.069
.077
.071
.083
95
.052
.054
.074
.087
.074
.089
.075
.065
.074
.106
.072
.069
.073
.092
.076
.087
.077
.095
99
.065
.076
.086
.113
.087
.110
.091
.081
.085
.126
.081
.081
.083
.110
.085
.103
.086
.108
Max.
.076
.106
.102
.127
.104
.135
.117
.117
.105
.145
.147
.096
.096
.162
.135
.140
.145
.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
-------
I Table AX3-14 (cont'd). Seasonal (April-October) Percentiles, SUM06, SUM08, and W126 Values for the MCCP Sites
>
X
H
6
o
o
H
O
o
HH
H
W
70 No. SUM08
Site Year Min. 10 30 50 (ppm) 90 95 99 Max. Obs. SUM06 (ppm-h) W126
ShenandoahPark, 1987 .000 .018 .029 .037 .047 .061 .068 .080 .108 3030 23.1 2.6 19.2
VA(SH3) 1988 ooo .020 .031 .040 .051
Whitetop Mountain 1987 °-01 -038 -051 -059 -066 -078 -085 -096 -111 4326 147-7 32A 105-7
VA(WTl) 1988 000 030 .046 .058 .068 .084 .094 .119 .163 3788 133.8 51.0 102.8
*Calculations based on a May-September season.
-------
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
DSigmoidal
DSum>0.07 ppm
Big Meadows Dickey Ridge Sawmill Run
b. May - September 1984
300 -i
• Total Dose
DSigmoidal
DSum>0.07 ppm
d. May - September 1986
300 -i
• Total Dose
DSigmoidal
DSum>0.07 ppm
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 AX3-46a-e. Integrated exposures for three non-Mountain Cloud Chemistry
Program Shenandoah National Park sites, 1983 to 1987.
Source: Lefohnetal. (1990b).
January 2005
AX3-106 DRAFT-DO NOT QUOTE OR CITE
-------
1 Taylor et al. (1992) indicated that the forests they monitored experienced differences in O3
2 exposure. The principal spatial factors underlying this variation were elevation, proximity to
3 anthropogenic sources of oxidant precursors, regional-scale meteorological conditions, and
4 airshed dynamics between the lower free troposphere and the surface boundary layer.
5 Table AX3-15 summarizes the exposure values for the 10 EPRI Integrated Forest Study sites
6 located in North America.
7 An important issue for assessing possible impacts of O3 at high-elevation sites that requires
8 further attention is the use of mixing ratios (e.g., parts per million) instead of absolute
9 concentration (e.g., in units of micrograms per cubic meter) to describe O3 concentration.
10 In most cases, mixing ratios or mole fractions are used to describe O3 concentrations. Lefohn
11 et al. (1990b) pointed out that the manner in which concentration is reported may be important
12 when assessing the potential impacts of air pollution on high-elevation forests. Concentration
13 varies as a function of altitude. Although the change in concentration is small when the
14 elevational difference between sea level and the monitoring site is small, it becomes substantial
15 at high-elevation sites. Given the same part-per-million value experienced at both a high- and
16 low-elevation site, the absolute concentrations (i.e., micrograms per cubic meter) at the two
17 elevations will be different, because both O3 and ambient air are gases, and changes in pressure
18 directly affect their volume. According to Boyle's law, if the temperature of a gas is held
19 constant, the volume occupied by the gas varies inversely with the pressure (i.e., as pressure
20 decreases, volume increases). This pressure effect must be considered when measuring
21 absolute pollutant concentrations. At any given sampling location, normal atmospheric pressure
22 variations have very little effect on air pollutant measurements. However, when mass/volume
23 units of concentration are used and pollutant concentrations measured at significantly different
24 altitudes are compared, pressure (and, hence, volume) adjustments are necessary. In practice,
25 the summit site at Whiteface Mountain had a slightly higher O3 exposure than the two
26 low-elevation sites (Lefohn et al., 1991). However, at Shenandoah National Park sites, the
27 higher elevation site experienced lower exposures than lower elevation sites in some years.
28 These exposure considerations are trivial at low-elevation sites. However, when one
29 compares exposure-effects results obtained at high-elevation sites with those from low-elevation
30 sites, the differences may become significant (Lefohn et al., 1990b). In particular, assuming that
31 the sensitivity of the biological target is identical at both low and high elevations, some
January 2005 AX3-107 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-15. Summary Statistics for 11 Integrated Forest Study Sites3
Site
HIGH ELEVATION
Whiteface Mtn,
NY
Great Smoky Mtns
NP
Coweeta
Hydrologic Lab,
NC
LOW ELEVATION
Huntington Forest,
NY
Howland, ME
Oak Ridge, TN
Year
SITES
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
SITES
1987
1987
1988
1988
1987
1987
1988
1988
1987
1987
1988
1988
Quarter
2
3
2
3
2
3
2
3
2
3
2
3
2
3
2
3
2
3
2
3
2
3
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
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
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
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
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
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
January 2005
AX3-108
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-15 (cont'd). Summary Statistics for 11 Integrated Forest Study Sites.3
Site
LOW ELEVATION
(cont'd)
Thompson Forest,
WA
B.F. Grant Forest,
GA
Gainesville, FL
Duke Forest, NC
Nordmoen,
Norway
Year
SITES
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
3
2
3
2
3
2
3
2
3
2
3
2
3
2
3
24-h
36
30
32
32
32
o o
JJ
47
32
42
29
35
20
38
52
54
38
32
14
22
11
12-h
43
36
39
39
46
52
63
47
53
44
48
29
48
59
69
51
40
18
28
15
7-h
41
34
37
36
48
54
64
48
50
41
51
30
52
50
75
54
41
20
29
16
1-h Max.
103
94
103
140
99
102
127
116
b
b
84
70
100
124
115
141
75
32
53
30
SUM06
103 ppb-h
10.7
10.3
8.1
13.5
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
3.6
2.1
2.3
6.7
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).
January 2005
AX3-109 DRAFT-DO NOT QUOTE OR CITE
-------
1 adjustment will be necessary when attempting to link experimental data obtained at low-
2 elevation sites with air quality data monitored at the high-elevation stations. This topic is further
3 discussed in Annex AX4 when considering effective dose considerations for predicting
4 vegetation effects associated with O3.
5
6
7 AX3.6 TRENDS IN OZONE CONCENTRATIONS
8 AX3.6.1 National Assessment of Trends
9 Ozone concentrations and, thus, exposure, change from year to year. During the period
10 1983 to 2002, extremely high O3 levels occurred in 1983 and 1988 in some areas of the United
11 States. These levels more than likely were attributable, in part, to hot, dry, stagnant conditions.
12 However, the U.S. Environmental Protection Agency (2003b) reported that a downward national
13 trend in 1-h and 8-h O3 levels occurred, for the period 1983 to 2002, in most geographic areas in
14 the country (Figures AX3-47 and AX3-48). These low levels may have been due to
15 meteorological conditions that were less favorable for O3 formation and to recently implemented
16 control measures. On a national assessment basis, the U.S. Environmental Protection Agency
17 (2003b) reported a 22% decrease in the second maximum 1-h average O3 concentration for the
18 period 1983 to 2002 and a 14% decrease for the annual fourth highest annual daily maximum
19 concentration during the same 20-year period. The Northeast and Pacific Southwest exhibited
20 the most substantial improvement for 1-h and 8-h O3 levels. The Mid-Atlantic and North Central
21 regions experienced minimal decreases in 8-h O3 levels. In contrast, the Pacific Northwest
22 region showed a slight increase in the 8-h O3 over the period 1983 to 2002.
23 For the 10-year period of 1993 to 2002, the national trend in 8-h O3 concentration shows a
24 4% increase and the national trend in 1-h O3 shows a 2% decrease. However, standard statistical
25 tests show that these trends were not statistically significant. Ozone concentrations varied
26 slightly during this 10-year period but showed no real change overall.
27
28 AX3.6.2 Regional Trends
29 Regional trends provide additional information that can increase our understanding of the
30 patterns of change of O3 levels. Figures AX3-49 and AX3-50 illustrate the 20-year trend for
31 both 1- and 8-h O3, averaged across regional boundaries. For example, the trend in 8-h O3
January 2005 AX3-110 DRAFT-DO NOT QUOTE OR CITE
-------
0.20
0.15
E
Q.
Q.
-------
.175
1983 2002
f 22%
Figure AX3-49. Trend in 1-h O3 level (based on annual second highest daily maximum
concentration) for 1983 to 2002 averaged across EPA regional office
boundaries.
Source: U.S. Environmental Protection Agency (2003b).
.124
.098
The National Trend
.100 .086
198J 200Z
f 14%
Figure AX3-50. Trend in 8-h O3 level (based on annual fourth highest maximum
concentration) for 1983 to 2002 averaged across EPA regional office
boundaries.
Source: U.S. Environmental Protection Agency (2003b).
January 2005
AX3-112 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations for the Pacific Southwest shows the 20-year trend (1983 to 2002) as a 29%
2 decrease. When considering the Los Angeles area separately, the trend for Los Angeles shows a
3 49% decrease for the 20-year period, compared to a 15% decrease for other locations in the
4 Pacific Southwest. For the 10-year period of 1993 to 2002, the Pacific Southwest has an overall
5 13% decrease in 8-h O3. However, when considering Los Angeles separately, the Los Angeles
6 area has a 28% decrease for the 10-year period, while the Pacific Southwest without Los Angeles
7 has a 5% decrease. This illustrates that national assessments for O3 do not describe trends
8 completely, particularly where control measures such as those implemented in Los Angeles have
9 had a significant effect in reducing O3 concentrations (U.S. Environmental Protection Agency,
10 2003b).
11 It is important to note that year-to-year changes in ambient O3 trends are influenced by
12 meteorological conditions, population growth, and changes in emission levels of O3 precursors
13 (i.e.,VOCs and NOX) resulting from ongoing control measures. For example, to further evaluate
14 the 10-year 8-h O3 trends, U.S. Environmental Protection Agency (2003b) applied a model to the
15 annual rate of change in O3 based on measurements in 53 metropolitan areas. This model
16 adjusted the O3 data in these areas to account for the influence of local meteorological
17 conditions, including surface temperature and wind speed. Figure AX3-51 shows the aggregated
18 trend in 8-h O3 for these 53 areas adjusted for meteorological conditions for the 10-year period
19 of 1993 to 2002. The figure also shows the aggregated trend for these areas unadjusted for
20 meteorology as well as the national average in 8-h O3. From this figure, the meteorologically
21 adjusted trend for this 10-year period can be seen as relatively flat, which agrees with the pattern
22 observed without applying meteorologically adjusted methods.
23 Furthermore, preliminary examination of meteorologically adjusted 8-h O3 on a
24 subregional basis in the eastern United States reveals a pattern of increasing O3 through 1998,
25 followed by a period of generally improving O3 air quality. The U.S. Environmental Protection
26 Agency (2003b) has attributed this reversal to the implementation of regional NOX reductions
27 from power plants.
28 Data from the EPA's AQS database were used to characterize the year-to-year variability
29 in the fourth highest daily maximum 1-h concentration over 3 years for the period of 1980 to
30 2001 at specific monitoring sites located in several metropolitan areas across the United States.
31 Figures AX3-52 through AX3-59 illustrate the reduction in the fourth highest daily maximum
January 2005 AX3-113 DRAFT-DO NOT QUOTE OR CITE
-------
0.1
"§ 0.08
Q.
Q.
7 0.06
c
O °-04
0.02
n i i i i i i i i i
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Year
-•- Selected Area Trend in Average Daily Maximum 8-Hour Concentrations
Meteorologically Adjusted Trend in Average Daily Maximum 8-Hour Concentrations
-±- National Trend in Annual 4th Maximum 8-Hour Concentrations
Figure AX3-51. Comparison of actual and meteorological adjusted 8-h O3 trends for the
period 1993 to 2003.
Source: U.S. Environmental Protection Agency (2003b).
u.^a
Oon -
?
°- n 15 -
Q. U. IO
5 n m -
o u- ' u
N
O
n n*) -
n nn -
A A A.
f ^^-^ A A
^*^^\
^»~~->-^
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-52. Trend in the 4th highest daily maximum 1-h O3 concentration over
3 years for the period of 1980 to 2001 for a monitoring site in Los
Angeles, CA (060371301).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-114 DRAFT-DO NOT QUOTE OR CITE
-------
Q.
a.
Q)
c
o
N
o
01fi -
014 -
019 -
Om -
Ons .
Onfi .
n r\A .
00?-
n nn .
*^*N^-*^^
X
^* 1 *- . * . A A A
^*
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-53. Trend in the 4th highest daily maximum 1-h O3 concentration over
3 years for the period of 1980 to 2001 for a monitoring site in Phoenix,
AZ (040133002).
Source: U.S. Environmental Protection Agency (2003a).
Q.
a.
o
c
o
N
O
019 -
010-
n DR -
0 Ofi -
n 04 -
n n? -
n nn .
«^*^*^
v* '^—-^^ ^ ^
^—+^ + ^^
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-54. Trend in the 4th highest daily maximum 1-h O3 concentration over
3 years for the period of 1980 to 2001 for a monitoring site in Denver, CO
(080590002).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-115 DRAFT-DO NOT QUOTE OR CITE
-------
u. 10 -
01fi -
014 -
Q.
Q- n m -
2 n ns -
n n4 -
n n? -
n nn _
r * \
y*^*~^*\ / \
^ * * \ t t
*^ * * * *^^» — h » »
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-55. Trend in the 4th highest daily maximum 1-h O3 concentration over
3 years for the period of 1980 to 2001 for a monitoring site in Chicago, IL
(170317002).
Source: U.S. Environmental Protection Agency (2003a).
u. 10 -
n 14 -
n 19 -
an m -
a.
^"* n OR -
gj U.UO
On OR -
N
O n n4 -
n no .
n nn .
'Nt / ^^ ^^^*-~+—+^*—^
\/ ^~*
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-56. Trend in the 4th highest daily maximum 1-h O3 concentration over
3 years for the period of 1980 to 2001 for a monitoring site in Detroit, MI
(260990009).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-116 DRAFT-DO NOT QUOTE OR CITE
-------
u.^o -
Oon -
?
°- n i*, -
Q. U. IO
o
^ n 1 n
O u. lu -
N
O
Ont;
n nn .
• *""^\ ^-^^^ ^^\^
-------
u. 10
0-IR -
.ID
01/1 -
EO 1? -
Q.
Q. n -in .
2 n nR -
M n OR -
n n4 -
n n? -
n nn -
* * A y' \
^^ »x \
^"^ \
* • + — •*- ^-> * « *
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-59. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Washington, DC (110010025).
Source: U.S. Environmental Protection Agency (2003a).
1 1-h concentration over 3 years for monitoring sites in Los Angeles, California; Phoenix,
2 Arizona; Denver, Colorado; Chicago, Illinois; Detroit, Michigan; Houston, Texas; Fairfield,
3 Connecticut; and Washington, D.C. Figures AX3-60 through AX3-67 illustrate the trends using
4 the fourth highest daily maximum 8-h concentration averaged over 3 years for the same
5 monitoring sites for the same period of time.
6 As described in Section AX3.6.1, the U.S. Environmental Protection Agency (2003b), on a
7 national assessment basis, reported a 22% decrease in the second maximum 1-h average
8 concentration for the period of 1983 to 2002 and a nonsignificant decrease of 2% for the period
9 of 1993 to 2002. For the annual fourth highest daily maximum 8-h concentration, the U.S.
10 Environmental Protection Agency (2003b) reported a 14% decrease for the period 1983 to 2002
11 and a nonsignificant increase of 4% for the 10-year period 1993 to 2002. The implication of
12 these figures is that a slowing of improvement has occurred, which is illustrated for many of the
13 monitoring sites described in Figures AX3-52 through AX3-67 for both the fourth highest daily
14 maximum 1-h concentration over 3 years and the fourth highest daily maximum 8-h
15 concentration averaged over 3 years.
January 2005
AX3-118 DRAFT-DO NOT QUOTE OR CITE
-------
Q.
a.
Q)
c
o
N
o
019 -
01 n -
Ons .
n nfi -
On/i
Ono
n nn .
^_» t
"*^^_
^ ^v^^
^ — * — *---•• -*
^ ™
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-60. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in Los
Angeles, CA (060371301).
Source: U.S. Environmental Protection Agency (2003a).
Q.
a.
o
c
o
N
O
Om -
Ons
.uo -
Onfi .
Odd -
n no .
n nn .
*^^^^.
f-N^^^— * *^>- i l~.
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-61. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Phoenix, AZ (040133002).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-119 DRAFT-DO NOT QUOTE OR CITE
-------
Q.
a.
Q)
c
o
N
o
0 08 -
Onfi -
Or\A
Ono
n nn _
r*-^ ^^-^
^+* — ». ^^-* — n^"^ ^^*
-^ ^ ^ -» ^
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-62. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Denver, CO (080590002).
Source: U.S. Environmental Protection Agency (2003a).
Q.
a.
o
c
o
N
O
Om -
Ons .
Onfi .
Odd -
n no .
n nn .
-*-«< ^^\^
*^ ^ N-^_^ ^-^^+^^
^
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-63. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period in 1980 to 2001 for a monitoring site in
Chicago, IL (170317002).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-120 DRAFT-DO NOT QUOTE OR CITE
-------
Q.
a.
Q)
c
o
N
o
Om -
Ons .
Onfi .
Ond -
n no .
n nn .
*^--^ .^-* * *~^ A •* *^
\^ ;* *~—4 — *-— *~^ * "-* — *
^-^
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-64. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Detroit, MI (260990009).
Source: U.S. Environmental Protection Agency (2003a).
Q.
a.
u
o
N
O
u. 10
0-IR -
.ID
01/1 -
01 O
n m -
n nR .
n nfi -
n n4 -
n n? -
n nn -
\
V^
NX*— 4^*^^*"^^*^
^^* *^+-~+^~+ * ~~**
1983 1985 1987 1989 1991 1993 1995 1997 1999 2001
Year
Figure AX3-65. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Houston, TX (482011037).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-121
DRAFT-DO NOT QUOTE OR CITE
-------
0.18
0.00
1982 1984
Figure AX3-66. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Fairfield, CN (090013007).
Source: U.S. Environmental Protection Agency (2003a).
Q.
a.
o
c
o
N
O
010-
0 OR -
0 OR -
0 04 -
00?-
n nn .
^^^^ ^_^
^^* « « *^^* — *^-^
^* — « — »^^*
1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
Year
Figure AX3-67. Trend in the 4th highest daily maximum 8-h O3 concentration averaged
over 3 years for the period of 1980 to 2001 for a monitoring site in
Washington, DC (110010025).
Source: U.S. Environmental Protection Agency (2003a).
January 2005
AX3-122 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.6.3 Disproportionate Reductions in Hourly Average Concentrations
2 Figure AX3-68 shows that the higher hourly average concentrations (i.e., > 0.09 ppm)
3 decreased at a faster rate (i.e., at a greater negative rate per year) than the hourly average
4 concentrations in the mid-level range. The numbers of hourly average concentrations in the low
5 end of the distribution also decreased. The result was that, in some cases, the hourly average
6 concentrations in the 0.030 to 0.050 ppm range increased as the peak hourly average
7 concentrations were reduced. Apparently, both the high and low ends of the distribution were
8 moving toward the center of the distribution of the hourly average concentrations.
9
41
o
o
ll °-
a oJ ,
-------
1 from the sites that experienced reduced O3 levels over this period of time, the researchers
2 investigated whether the rate of decline of the mid-level hourly average O3 concentrations
3 (0.06 to 0.099 ppm) was similar to the rate experienced by the high hourly average
4 concentrations (> 0.10 ppm). Lefohn et al. (1998) reported that there was a greater resistance to
5 reducing the hourly average concentrations in the mid-range than the hourly average
6 concentrations at the higher concentration range.
7 Information in the literature provides some indication that the higher concentrations may
8 be reduced at a faster rate than the lower concentrations. The U.S. Environmental Protection
9 Agency (1985) reported that the application of the first-generation regional oxidant model
10 resulted in emissions controls reducing peak concentrations by considerably larger percentages
11 than median or mean concentration values. The emissions controls had no significant effect on
12 concentrations below approximately the 90th percentile value (i.e., 0.080 to 0.090 ppm in the
13 summer and approximately 0.070 ppm in the spring). Roselle and Schere (1995) described
14 similar observations. The U.S. Environmental Protection Agency (1997) noted in its
15 examination of air quality over the last 25 years that, for sites that historically showed
16 improvements, somewhat larger reductions occurred at the higher hourly average concentrations
17 than at the middle or lower values. Yeng and Miller (2002) analyzed monitoring data from
18 10 stations in Connecticut for the period of 1981 to 1997. The authors noted that emissions
19 reductions in the area had been successful in reducing the hourly O3 average concentrations but
20 that the lower values actually increased as emissions reductions occurred. Yeng and Miller
21 (2002) concluded that the resistance to reducing the mid-level concentrations was related to a
22 combination of natural factors and area-wide background concentrations of precursors from
23 nonpoint anthropogenic sources. Likewise, Bravo et al. (2003) have reported that the higher
24 hourly average concentrations were reduced at a faster rate than the mid-level values when
25 surface O3 concentrations in Mexico City were reduced.
26 Using models, several investigators have commented on the difficulty in reducing the
27 mid-level hourly average O3 concentrations while reducing the fourth highest 8-h average daily
28 maximum concentration. Saylor et al. (1999) noted that for the Atlanta, GA area, NOX emissions
29 reductions greater than 60 to 75% would be required to reduce the mid-level hourly average
30 concentrations. Similarly, Winner and Cass (2000) noted that the higher hourly average
31 concentrations were reduced much faster than the mid-level values during simulation modeling
January 2005 AX3-124 DRAFT-DO NOT QUOTE OR CITE
-------
1 for the Los Angeles area. Reynolds et al. (2003, 2004) analyzed ambient O3 concentrations used
2 in conjunction with the application of photochemical modeling to determine the technical
3 feasibility of reducing hourly average concentrations in central California in the eastern United
4 States. Various combinations of VOC and NOX emission reductions were effective in lowering
5 modeled peak 1-h O3 concentrations. However, VOC emissions reductions were found to have
6 only a modest impact on modeled peak 8-h O3 concentrations. Reynolds et al. (2003) reported
7 that 70 to 90% NOX emissions reductions were required to reduce peak 8-h O3 concentrations to
8 the desired level in central California. Anthropogenic NOX emissions reductions of 46 to 86% of
9 1996 base case values were needed to reach the desired level of the 8-h value in some areas in
10 the eastern United States (Reynolds et al., 2004).
11 Reynolds et al. (2003, 2004) commented on possible chemical explanations for the
12 observation that more prominent trends in peak 1-h O3 levels than in peak 8-h O3 concentrations
13 or in occurrences of mid-level (i.e., 0.06 to 0.09 ppm) concentrations have been reported. The
14 authors noted that when anthropogenic VOC and NOX emissions are reduced significantly, the
15 primary sources of O3 precursors are biogenic emissions and CO from anthropogenic sources.
16 Chemical process analysis results indicated that a slowly reacting pollutant such as CO could be
17 contributing on the order of 10 to 20% of the O3 produced. The authors recommended that
18 further work focus on the need to confirm that biogenic emissions have not been significantly
19 overestimated in the most recent emission inventories and on the examination of the effects of
20 CO reductions. Several investigators have proposed that the effect of Asian emissions on
21 surface O3 concentrations might be responsible for explaining the ineffectiveness of the
22 emissions reductions affecting the mid-level concentrations (Jacob et al., 1999; Lin et al., 2000;
23 Jaffe et al., 2003; Vingarzan and Taylor, 2003). However, as described in Section AX3.2.3,
24 inconsistent trends results using data collected at RRMS in the western United States do not
25 indicate that either the low or the high end of the hourly average concentration distributions are
26 increasing and, thus, the effect of Asian emissions on surface O3 concentrations in the western
27 United States is not clear at this time.
28
29 AX3.6.4 Trends in National Parks in the United States
30 The U.S. Environmental Protection Agency (2003b) reported that 28 of the national parks
31 in the United States had sufficient O3 trend data for the 10-year period of 1993 to 2002. Seven
January 2005 AX3-125 DRAFT-DO NOT QUOTE OR CITE
-------
1 monitoring sites in five of these parks experienced statistically significant upward trends in 8-h
2 O3 levels: Great Smoky Mountains (TN), Craters of the Moon (ID), Mesa Verde (CO), Denali
3 (AK), and Acadia (ME). Monitoring data for one park showed statistically significant
4 improvements over the same time period: Saguaro (AZ). For the remaining 22 parks with O3
5 trends data, the 8-h O3 levels at 13 parks increased only slightly between 1993 and 2002, while
6 five parks showed decreasing levels and 4 parks were unchanged.
7 Air quality in our national parks is important for both human health and welfare (e.g.,
8 vegetation) considerations. As noted, the U.S. Environmental Protection Agency (2003b), using
9 10 years of data and the fourth highest 8-h average of the daily maximum O3 concentration, has
10 characterized the trending in our national parks. As indicated in the U.S. Environmental
11 Protection Agency (2003b) results, using different exposure indices (e.g., 1-h and 8-h values),
12 different trending results occur. Using all available data for the national parks, an additional
13 analysis of the hourly average concentration data was performed to characterize the trending
14 using the (a) seasonal W126 and SUM06 cumulative exposure indices, (b) fourth highest of the
15 1-h daily maximum average concentrations over a 3-year period for January to December, and
16 (c) running 3-year average of the annual fourth highest daily maximum 8-h average
17 concentration.
18 For the seasonal (April to October) W126 and SUM06 exposure indices, all available data
19 were used. At the time of the analysis, in most cases National Park Service data were available
20 for the period of 1988 to 2001. However, a few of the monitoring sites had data that began in
21 1981. At least 75% of the data had to be available for each year over the 7-month season (April
22 to October) for a site to be included in the analysis. The Kendall's K statistic (Mann-Kendall
23 test) was used to identify linear trends as described in Lefohn and Shadwick (1991). Estimates
24 of the rate of change (Sen estimate of the slope) for the index were calculated. Table AX3-16
25 summarizes the results of the analysis. Caution is urged in interpreting the table. The use of the
26 significance levels of 0.2, 0.1, and 0.05 indicates varying levels of uncertainty. As indicated
27 above, the use of a specific exposure index (e.g., 8-h average concentration) will provide a
28 different trending pattern than the use of an alternative index, such as the W126 or SUM06
29 cumulative exposure index. Both the 1-h and 8-h indices are extreme value metrics that focus on
30 the highest levels in the distribution of the hourly average concentrations. On the other hand, the
31 W126 and SUM06 exposure indices accumulate the hourly average concentrations mainly in the
January 2005 AX3-126 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-16. Trends at National Parks in the United States (1981 to 2001 or available data period)
-------
Table AX3-16 (cont'd). Trends at National Parks in the United States (1981 to 2001 or available data period)
-------
Table AX3-16 (cont'd). Trends at National Parks in the United States (1981 to 2001 or available data period)
c
^
to
o
o
X
OJ
Site
Sequoia/Kings Canyon NPs (Lower Kaweah) SELK
Shenandoah NP (Big Meadows) SVBM
VoyageursNP VOYA
Yellowstone NP YELL
Yellowstone NP YELL
Yosemite NP (Camp Mather) YOCM
Yosemite NP (Turtleback Dome) YOTD
Yosemite NP (Wawona Valley) YOWV
* 0.20 level of significance
* * 0.10 level of significance
AIRS ID
061070006
511130003
271370034
560391010
560391011
061090004
060430003
060430004
W126
(Seasonal)
NS
NS
**(-)
NS
NS
NS
*(-)
SUM06
(Seasonal)
NS
NS
**(-)
NS
NS
NS
*(-)
8-h (Annual)
NS
NS
*(-)
NS
NS
*(+)
***(-)
***(-)
l-h (Annual)
NS
***(+)
***(-)
NS
**(+)
***(0)
***(-)
**(-)
j-j *** 0.05 level of significance
^o NS Not significant
W126 (seasonal) is the cumulative W126 exposure (as described by Lefohn and Runeckles [1987]) over a 24-h period for the period of April to October.
SUM06 (seasonal) is the cumulative exposure of hourly average concentrations > 0.06 ppm over a 24-h period for the period of April to October.
8-h (annual) is the 3-year average of the annual fourth highest 8-h daily maximum average concentration for the period of January to December.
2 l-h (annual) is the fourth highest l-h daily maximum average concentration over a 3-year period for January to December.
H
6
o
o
H
O
o
HH
H
W
-------
1 middle level of the distribution to the highest values. Changes in the magnitude of the extreme
2 value statistics may or may not necessarily imply a change in the entire distribution of the hourly
3 average concentrations. Thus, in interpreting Table AX3-16, it is suggested that, for a specific
4 monitoring site, the pattern of trending for all four metrics be evaluated. In addition, because of
5 the variability in the number of years when valid data were available, focus should be on the
6 0.05 level of significance. For example, Joshua Tree National Park in California shows a pattern
7 where three of the four exposure metrics show a decrease trend in O3 levels at the 0.05 level of
8 significance. Clearly, one would suspect that a strong trend is occurring at this site.
9 Alternatively, Yellowstone National Park in Wyoming shows no significant change using three
10 of the four metrics, while the 1-h index showed an increase at the 0.10 level of significance. The
11 trend at this site should be watched over the next several years to observe whether the statistical
12 significance levels of the metrics increase.
13
14
15 AX3.7 RELATIONS BETWEEN OZONE, OTHER OXIDANTS, AND
16 OXIDATION PRODUCTS
17 Tabulations of measurements of PAN and peroxypropionyl nitrate (PPN,
18 CH3CH2C(O)OONO2) concentrations were given in the 1996 O3 AQCD (U.S. Environmental
19 Protection Agency, 1996a). Measurements were summarized for rural and urban areas in the
20 United States, Canada, France, Greece, and Brazil. The use of measurements from aboard serve
21 to illustrate or support certain U.S. results as well as to demonstrate the widespread presence of
22 PANs in the atmosphere. Additional data for H2O2 were also presented in the 1996 O3 AQCD.
23 Data for these species are obtained as part of specialized field studies and not as part of routine
24 monitoring operations and thus are highly limited in their ranges of applicability. As a result, it
25 is difficult to relate the concentrations of O3, other oxidants, and oxidation products on the basis
26 of rather sparse data sets. This information is simply not available for a large number of
27 environments. Instead, it might be more instructive to examine the relations between O3 and
28 other products of atmospheric reactions on the basis of current understanding of atmospheric
29 photochemical processes.
30 In order to understand co-occurrence between atmospheric species, an important
31 distinction must be made between primary (directly emitted) species and secondary
January 2005 AX3-130 DRAFT-DO NOT QUOTE OR CITE
-------
1 (photochemically produced) species. In general, it is more likely that primary species will be
2 more highly correlated with each other, and that secondary species will be more highly
3 correlated with each other than will species from mixed classes. By contrast, primary and
4 secondary species are less likely to be correlated with each other. Secondary reaction products
5 tend to correlate with each other, but there is considerable variation. Some species (e.g., O3 and
6 organic nitrates) are closely related photochemically and correlate with each other strongly.
7 Others (e.g., O3 and H2O2) show a more complex correlation pattern.
8 Although NO2 is produced mainly by the reaction of directly emitted NO with O3 with
9 some contributions from direct emissions, in practice, it behaves like a primary species. The
10 timescale for conversion of NO to NO2 is fast (5 min or less), so NO and NO2 ambient
11 concentrations rapidly approach values determined by the photochemical steady state. The sum
12 NO + NO2 (NOX) behaves like a typical primary species, while NO and NO2 reflect some
13 additional complexity based on photochemical interconversion. As a primary species, NO2
14 generally does not correlate with O3 in urban environments. In addition, chemical interactions
15 among O3, NO and NO2 have the effect of converting O3 to NO2 and vice versa, which can result
16 in a significant anti-correlation between O3 and NO2.
17 Organic nitrates consist of PAN, a number of higher-order species with photochemistry
18 similar to PAN (e.g., PPN), and species such as alkyl nitrates with somewhat different
19 photochemistry. These species are produced by a photochemical process very similar to that of
20 O3. Photochemical production is initiated by the reaction of primary and secondary VOCs with
21 OH radicals, the resulting organic radicals subsequently react with NO2 (producing PAN and
22 analogous species) or with NO (producing alkyl nitrates). The same sequence (with organic
23 radicals reacting with NO) leads to the formation of O3.
24 In addition, at warm temperatures, the concentration of PAN forms a photochemical steady
25 state with its radical precursors on a timescale of roughly 30 minutes. This steady state value
26 increases with the ambient concentration of O3 (Sillman et al., 1990). Ozone and PAN may
27 show different seasonal cycles, because they are affected differently by temperature. Ambient
28 O3 increases with temperature, driven in part by the photochemistry of PAN (see description
29 above). By contrast, the photochemical lifetime of PAN decreases rapidly with increasing
30 temperature. The ratio O3/PAN should show seasonal changes, with highest ratios in summer,
31 although there is no evidence from measurements. Measured ambient concentrations
January 2005 AX3-131 DRAFT-DO NOT QUOTE OR CITE
-------
1 (Figure AX3-69) show a strong correlation between O3 and PAN, and between O3 and other
2 organic nitrates (Pippin et al., 2001; Roberts et al., 1998).
3
1 *tU
120
100
80
60
•-> 40
.Q
Q. 20
^
Q
c
Q
N 120
O
100
80
60
40
20
n
:a , ' •
7 v - v/'-ff f 'i- 'i
: : -i *^~^~-$ l/t-6 -d T 1
* ... *aJL.~~ -^=%I. f t t^l.,...,., ,..»..• V" ~>-.JC. ,-- i r
• **^"if5sii^T.Tr r^ T *:i ! j -
" i^iJS^.. 'ft^.'^Qr-ii^--^ F* ~ -^
P.:
!
•w
-«: -1
* <•-' i it • ' 'it ntl«tfhirsm "
^Ht.llHM l,«)tlt 1- jK+tmttttW, »_
s " J I j
u- .|.4, 1
r ~
J
:
if"1"
: -
"• , 1 1 1 . 4 . 1 1 1 1 . i I , M.d*M jJitjjMiOJJkiJjJJJJjUJkt
ftw yrrr|TFiLLtL|-ttiJiYif'rlTr"''
-------
1 Individual primary VOCs are generally highly correlated with each other and with NOX
2 (Figure AX3-70). A summary of the results of a number of field studies of the concentrations of
3 precursors including NOX and nonmethane organic compounds (NMOCs) are summarized in the
4 1996 O3 AQCD. Although H2O2 is produced from photochemistry that is closely related to O3,
5 it does not show a consistent pattern of correlation with O3. Hydrogen peroxide is produced in
6 abundance along with O3 only when O3 is produced under NOx-limited conditions. When the
7 photochemistry is NOx-saturated much less H2O2 is produced. In addition, increasing NOX tends
8 to slow the formation of H2O2 under NOx-limited conditions.
9
20
40 60 80
NOy (ppbv)
100
120
Figure AX3-70. Measured correlation 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: Goldanetal. (1995).
1 Measurements associated with the Southern Oxidant Study (SOS) in Tennessee (Sillman
2 et al., 1998) and during the North Atlantic Regional Experiment (NARE) (Daum et al., 1996)
3 showed that elevated O3 was accompanied by relatively high H2O2 (3 ppb), much higher than
January 2005
AX3-133
DRAFT-DO NOT QUOTE OR CITE
-------
1 background values reported over the North Atlantic. However, very low H2O2 was found in
2 Los Angeles, even in the presence of high O3 (Sakugawa and Kaplan, 1989).
3 Elevated O3 is generally accompanied by elevated HNO3, although the correlation is not as
4 strong as between O3 and organic nitrates. Ozone often correlates with HNO3, because they have
5 the same precursor (NOX). However, HNO3 can be produced in significant quantities in winter,
6 even when O3 is low. The ratio between O3 and HNO3 also shows great variation in air pollution
7 events, with NOx-saturated environments having much lower ratios of O3 to HNO3 (Ryerson
8 etal.,2001).
9 Relations between primary and secondary components discussed above are illustrated by
10 considering data for O3 and PM25. Ozone and PM25 concentrations observed at a monitoring site
11 in Fort Meade, MD are plotted as conditional means in Figure AX3-71. These data were
12 collected between July 1999 and July 2001. As can be seen from the figure, PM2 5 tends to be
13 anti-correlated with O3 to the left of the inflection point (at about 30 ppbv O3) and PM25 tends to
14 be positively correlated with O3 to the right of the inflection point. Data to the left of the
15 minimum in PM25 were collected mainly during the cooler months of the year, while data to the
16 right of the minimum were collected during the warmer months. This situation arises because
17 PM2 5 contains a large secondary component during the summer and has a larger primary
18 component during winter. During the winter, O3 comes mainly from the free troposphere, above
19 the planetary boundary layer and, thus, may be considered a tracer for relatively clean air.
20 Unfortunately, data for PM2 5 and O3 are collected concurrently at relatively few sites in the
21 United States throughout an entire year, so these results, while highly instructive are not readily
22 extrapolated to areas where appreciable photochemical activity occurs throughout the year.
23
24
25 AX3.8 RELATIONSHIP BETWEEN SURFACE OZONE AND
26 OTHER POLLUTANTS
27 AX3.8.1 Introduction
28 Several attempts have been made to characterize gaseous air pollutant mixtures (Lefohn
29 and Tingey, 1984; Lefohn et al., 1987). The characterization of co-occurrence patterns under
30 ambient conditions is important for relating human health and vegetation effects to controlled
31 chamber studies and to ambient conditions. Lefohn et al. (1987) discussed the various patterns
January 2005 AX3-134 DRAFT-DO NOT QUOTE OR CITE
-------
40
35"
I 30
in
«N 25
ro 20
0)
c
o
^
T3
C
O
O
15--
10--
5--
1*?
10 20 30 40 50
Conditional Mean 03 (ppbv)
60
70
Figure AX3-71. Conditional mean PM2 5 concentrations versus conditional mean O3
concentrations observed at Fort Meade, MD from July 1999 to July 2001.
Source: Chen (2002).
1 of pollutant exposures. Pollutant combinations can occur at or above a threshold concentration
2 either together or temporally separated from one another. Patterns that show air pollutant pairs
3 appearing at the same hour of the day at concentrations equal to or greater than a minimum
4 hourly mean value were defined as simultaneous-only daily co-occurrences. When pollutant
5 pairs occurred at or above a minimum concentration during the 24-h period, without occurring
6 during the same hour, a "sequential-only" co-occurrence was defined. During a 24-h period, if
7 the pollutant pair occurred at or above the minimum level at the same hour of the day and at
8 different hours during the period, the co-occurrence pattern was defined as "complex-
9 sequential."
10 For characterizing the different types of co-occurrence patterns for O3/NO2, O3/SO2, and
11 NO2/SO2, Lefohn and Tingey (1984) used a 0.05 ppm threshold to identify the number of hourly
January 2005
AX3-135
DRAFT-DO NOT QUOTE OR CITE
-------
1 simultaneous-only co-occurrences for the period May through September at a large number of
2 air quality urban monitoring sites along with rural sites. The selection of a 0.05-ppm threshold
3 concentration was based on vegetation effects considerations. Data used in the analysis included
4 hourly averaged (1) Environmental Protection Agency Storage and Retrieval of Aerometric Data
5 (SAROAD; now AQS) data for 1981, (2) EPRI-SURE and Eastern Regional Air Quality Study
6 (ERAQS) data for 1978 and 1979, and (3) Tennessee Valley Authority (TVA) data from 1979 to
7 1982. Lefohn and Tingey (1984) concluded, for the pollutant combinations, that (1) the co-
8 occurrence of two-pollutant mixtures lasted only a few hours per episode and (2) the time
9 interval between episodes was generally large (weeks, sometimes months).
10 Lefohn et al. (1987), using a 0.03-ppm threshold, grouped air quality data from rural and
11 RRMS (as characterized in the EPA database) within a 24-h period starting at 0000 hours and
12 ending at 23 59 hours. Data were analyzed for the May to September period. Data used in the
13 analysis included hourly averaged (1) Environmental Protection Agency AQS (SAROAD) data
14 from 1978 to 1982, (2) EPRI-SURE and -ERAQS data for 1978 and 1979, and (3) TVA data
15 from 1979 to 1982. Patterns that showed air pollutant pairs appearing at the same hour of the
16 day at concentrations equal to or greater than a minimum hourly mean value were defined as
17 simultaneous-only daily co-occurrences. When pollutant pairs occurred at or above a minimum
18 concentration during the 24-h period, without occurring during the same hour, a "sequential-
19 only" co-occurrence was defined. During a 24-h period, if the pollutant pair occurred at or
20 above the minimum level at the same hour of the day and at different hours during the period,
21 the co-occurrence pattern was defined as "complex-sequential." A co-occurrence was not
22 indicated if one pollutant exceeded the minimum concentration just before midnight and the
23 other pollutant exceeded the minimum concentration just after midnight. As will be discussed
24 below, studies of the joint occurrence of gaseous NO2/O3 and SO2/O3 reached two conclusions:
25 (1) hourly simultaneous and daily simultaneous-only co-occurrences are fairly rare and (2) when
26 co-occurrences are present, complex-sequential and sequential-only co-occurrence patterns
27 predominate. The authors reported that year-to-year variability was found to be insignificant;
28 most of the monitoring sites experienced co-occurrences of any type less than 12% of the
29 153 days.
30 Since 1999, monitoring stations across the United States have been routinely measuring the
31 24-h average concentrations of PM2 5. Because of the availability of the PM2 5 data, daily
January 2005 AX3-136 DRAFT-DO NOT QUOTE OR CITE
-------
1 co-occurrence of PM25 and O3 over a 24-h period was characterized. Because PM25 data are
2 mostly summarized as 24-h average concentrations in the AQS data base, a daily co-occurrence
3 of O3 and PM25 was subjectively defined as when an hourly average O3 concentration
4 > 0.05 ppm and a PM2 5 24-h concentration > 40 |ig/m3 occurred over the same 24-h period.
5 For exploring the co-occurrence of O3 and other pollutants (e.g., acid precipitation and
6 acidic cloudwater), limited data are available. In most cases, routine monitoring data are not
7 available from which to draw general conclusions. However, published results are reviewed and
8 summarized for the purpose of assessing an estimate of the possible importance of co-occurrence
9 patterns of exposure.
10
11 AX3.8.2 Co-Occurrence of Ozone with Nitrogen Oxides
12 Ozone occurs frequently at concentrations equal to or greater than 0.05 ppm at many rural
13 and remote monitoring sites in the United States (U.S. Environmental Protection Agency,
14 1996a). Therefore, for many rural locations in the United States, the co-occurrence patterns
15 observed by Lefohn and Tingey (1984) for O3 and NO2 were defined by the presence or absence
16 of NO2. Lefohn and Tingey (1984) reported that most of the sites analyzed experienced fewer
17 than 10 co-occurrences (when both pollutants were present at an hourly average concentration
18 > 0.05 ppm). Figure AX3-72 summarizes the simultaneous co-occurrence patterns reported by
19 Lefohn and Tingey (1984). The authors noted that several urban monitoring sites in the South
20 Coast Air Basin experienced more than 450 co-occurrences. For more moderate areas of the
21 country, Lefohn et al. (1987) reported that even with a threshold of 0.03 ppm O3, the number of
22 co-occurrences with NO2 was small.
23 Using 2001 data from the U.S. EPA AQS database, patterns that showed air pollutant pairs
24 of O3/NO2 appearing at the same hour of the day at concentrations > 0.05 ppm were
25 characterized. The data were not segregated by location settings categories (i.e., rural, suburban,
26 and urban and center city) or land use types (i.e., agricultural, commercial, desert, forest,
27 industrial, mobile, or residential). Data capture was not a consideration in the analysis. The data
28 were characterized over the EPA-defined O3 season (Table AX3-1). In 2001, there were 341
29 monitoring sites that co-monitored O3 and NO2. Because of possible missing hourly average
30 concentration data during periods when co-monitoring may have occurred, no attempt was made
January 2005 AX3-137 DRAFT-DO NOT QUOTE OR CITE
-------
100
1 *_/W
80
(A
0)
J/5 60
"5
1 40
z
20
n
-
-
-
ll... . 1
oooooooo
CN n ••*• 10 CD h- oo
Number of Co-Occurrences
Figure AX3-72. The co-occurrence pattern for O3 and NO2.
Source: Lefohn and Tingey (1984).
1 to characterize the number of co-occurrences in the 0 category. Thus, co-occurrence patterns
2 were identified for those monitoring sites that experienced one or more co-occurrences.
3 Figure AX3-73 illustrates the results of the analysis. Similar to the analysis summarized
4 by Lefohn and Tingey (1984), most of the collocated monitoring sites analyzed, using the 2001
5 data, experienced fewer than 10 co-occurrences (when both pollutants were present at an hourly
6 average concentration > 0.05 ppm).
7
8 AX3.8.3 Co-Occurrence of Ozone with Sulfur Dioxide
9 Because elevated SO2 concentrations are mostly associated with industrial activities (U.S.
10 Environmental Protection Agency, 1992), co-occurrence observations are usually associated
11 with monitors located near these types of sources. Lefohn and Tingey (1984) reported that,
12 for the rural and nonrural monitoring sites investigated, most sites experienced fewer than
13 10 co-occurrences of SO2 and O3. Lefohn et al. (1987) reported that even with a threshold of
January 2005
AX3-13 8 DRAFT-DO NOT QUOTE OR CITE
-------
40-
35-
30-
25-
Z. 20-
0)
n
10-
5-
0-
CD en
CD
CM
O
CN
CO
CO
O
CO
CD
CO CO
m CD
\ i
o o
IO CD
CO
CO
CO
o
CO
CO
CO
o
CO
CO
o
CO
CO
CN
CO
CO
CO CO
o
o
o o o
-,- CM CO
o
LO
Number of Co-Occurrences (Hours)
Figure AX3-73. The co-occurrence pattern for O3 and NO2 using 2001 data from
the AQS.
1 0.03 ppm O3, the number of co-occurrences with SO2 was small. Figure AX3-74 illustrates the
2 simultaneous co-occurrence results reported by Lefohn and Tingey (1984).
3 Meagher et al. (1987) reported that several documented O3 episodes at specific rural
4 locations appeared to be associated with elevated SO2 levels. The investigators defined the
5 co-occurrence of O3 and SO2 to be when hourly mean concentrations were > 0.10 and 0.01 ppm,
6 respectively.
7 The above discussion was based on the co-occurrence patterns associated with the presence
8 or absence of hourly average concentrations of pollutant pairs. Taylor et al. (1992) have
9 discussed the joint occurrence of O3, nitrogen, and sulfur in forested areas using cumulative
10 exposures of O3 with data on dry deposition of sulfur and nitrogen. The authors concluded in
11 their study that the forest landscapes with the highest loadings of sulfur and nitrogen via dry
12 deposition tended to be the same forests with the highest average O3 concentrations and largest
13 cumulative exposure. Although the authors concluded that the joint occurrences of multiple
14 pollutants in forest landscapes were important, nothing was mentioned about the hourly
15 co-occurrences of O3 and SO2 or of O3 and NO2.
January 2005
AX3-139 DRAFT-DO NOT QUOTE OR CITE
-------
1 \J\J
80
W
-2
(75 60
"5
OJ
| 40
3
z
20
r\
-
-
-
II..
oooooooo
Number of Co-Occurrences
Figure AX3-74. The co-occurrence pattern for O3 and SO2.
Source: Lefohn and Tingey (1984).
1 Using 2001 data from the EPA AQS database, patterns that showed air pollutant pairs of
2 O3/SO2 appearing at the same hour of the day at concentrations > 0.05 ppm were characterized.
3 The data were not segregated by location settings categories (i.e., rural, suburban, and urban and
4 center city) or land use types (i.e., agricultural, commercial, desert, forest, industrial, mobile, or
5 residential). Data capture was not a consideration in the analysis. In 2001, there were 246
6 monitoring sites that co-monitored O3 and SO2. As discussed previously, because of possible
7 missing hourly average concentration data during periods when co-monitoring may have
8 occurred, no attempt was made to characterize the number of co-occurrences in the 0 category.
9 Thus, co-occurrence patterns were identified for those monitoring sites that experienced one or
10 more co-occurrences. Figure AX3-75 shows the results from this analysis for the simultaneous
11 co-occurrence of O3 and SO2. Similar to the analysis summarized by Lefohn and Tingey (1984),
12 most of the collocated monitoring sites analyzed, using the 2001 data, experienced fewer than
13 10 co-occurrences (when both pollutants were present at an hourly average concentration
14 > 0.05 ppm).
January 2005
AX3-140 DRAFT-DO NOT QUOTE OR CITE
-------
60
50 -|
g
55 4(
>t-
° 30-
| 20-
10-
0
O t- CM CO
Number of Co-Occurrences (Hours)
LO
Figure AX3-75. The co-occurrence pattern for O3 and SO2 using 2001 data from AQS.
1 AX3.8.4 Co-Occurrence of Ozone and Daily PM2 5
2 Using 2001 data from the EPA AQS, the daily co-occurrence of PM2 5 and O3 over a 24-h
3 period was characterized. There were 362 sites where PM25 and O3 monitors were collocated.
4 As described in the introduction selection of this annex, a daily co-occurrence of O3 and PM25 is
5 subjectively defined as an hourly average O3 concentration > 0.05 ppm and a PM2 5 24-h
6 concentration > 40 |ig/m3 occurring over the same 24-h period. Figure AX3-76 illustrates the
7 daily co-occurrence patterns observed. Using 2001 data from the AQS, the daily co-occurrence
8 of PM2 5 and O3 was infrequent.
9
10 AX3.8.5 Co-Occurrence of Ozone with Acid Precipitation
11 Concern has been expressed about the possible effects on vegetation from co-occurring
12 exposures of O3 and acid precipitation (Prinz et al., 1985; National Acid Precipitation
13 Assessment Program, 1987; Prinz and Krause, 1989). Little information has been published
14 concerning the co-occurrence patterns associated with the joint distribution of O3 and acidic
15 deposition (i.e., FT). Lefohn and Benedict (1983) reviewed the EPA SAROAD monitoring data
16 for 1977 through 1980 and, using National Atmospheric Deposition Program (NADP) and EPRI
17 wet deposition data, evaluated the frequency distribution of pH events for 34 NADP and 8 EPRI
January 2005
AX3-141 DRAFT-DO NOT QUOTE OR CITE
-------
ifin
140-
$ 120-
w 100-
M—
2 80-
| 60-
z 40-
20-
n
c
^ — CNJ cO ^~ LO CO r*'^ CO C73 ^D CNJ CO ^~ LO
ooooooooo
•s-cNro^TLncDr^oocnoooooo
O T— CN CO ^f LO
Number of Co-Occurrences (Hours)
Figure AX3-76. The co-occurrence pattern for O3 and PM2 5 using 2001 data from AQS.
1 chemistry monitoring sites located across the United States. Unfortunately, there were few sites
2 where O3 and acidic deposition were co-monitored.
3 As a result, Lefohn and Benedict (1983) focused their attention on O3 and acidic deposition
4 monitoring sites that were closest to one another. In some cases, the sites were as far apart as
5 144 km. Using hourly O3 monitoring data and weekly and event acidic deposition data from the
6 NADP and EPRI databases, the authors identified specific locations where the hourly mean O3
7 concentrations were > 0.1 ppm and 20% of the wetfall daily or weekly samples were below pH
8 4.0. Elevated levels of O3 were defined as hourly mean concentrations equal to or greater than
9 0.1 ppm. Although for many cases, experimental research results of acidic deposition on
10 agricultural crops show few effects at pH levels > 3.5 (National Acid Precipitation Assessment
11 Program, 1987), it was decided to use a pH threshold of 4.0 to take into consideration the
12 possibility of synergistic effects between O3 and acidic deposition.
13 Based on their analysis, Lefohn and Benedict (1983) identified five sites with the potential
14 for agricultural crops to experience additive, less than additive, or synergistic (i.e., greater than
15 additive) effects from elevated O3 and H+ concentrations. The authors stated that they believed,
16 based on the available data, the greatest potential for interaction between acid rain and O3
17 concentrations in the United States, with possible effects on crop yields, may be in the most
18 industrialized areas (e.g., Ohio and Pennsylvania). However, they cautioned that, because no
January 2005
AX3-142 DRAFT-DO NOT QUOTE OR CITE
-------
1 documented evidence existed to show that pollutant interaction had occurred under field growth
2 conditions and ambient exposures, their conclusions should only be used as a guide for further
3 research.
4 In their analysis, Lefohn and Benedict (1983) found no collocated sites. The authors
5 rationalized that data from non-co-monitoring sites (i.e., O3 and acidic deposition) could be used
6 because O3 exposures are regional in nature. However, work by Lefohn et al. (1988) has shown
7 that hourly mean O3 concentrations vary from location to location within a region, and that
8 cumulative indices, such as the percent of hourly mean concentrations > 0.07 ppm, do not form a
9 uniform pattern over a region. Thus, extrapolating hourly mean O3 concentrations from known
10 locations to other areas within a region may provide only qualitative indications of actual O3
11 exposure patterns.
12 In the late 1970s and the 1980s, both the private sector and the government funded research
13 efforts to better characterize gaseous air pollutant concentrations and wet deposition. The event-
14 oriented wet deposition network, EPRI/Utility Acid Precipitation Study Program, and the weekly
15 oriented sampling network, NADP, provided information that can be compared with hourly
16 mean O3 concentrations collected at several co-monitored locations. No attempt was made to
17 include H+ cloud deposition information. In some cases, for mountaintop locations (e.g.,
18 Clingman's Peak, Shenandoah, Whiteface Mountain, and Whitetop Mountain), the H+ cloud
19 water deposition is greater than the H+ deposition in precipitation (Mohnen, 1989), and the
20 co-occurrence patterns associated with O3 and cloud deposition will be different from those
21 patterns associated with O3 and deposition by precipitation.
22 Smith and Lefohn (1991) explored the relationship between O3 and H+ in precipitation,
23 using data from sites that monitored both O3 and wet deposition simultaneously and within
24 one-minute latitude and longitude of each other. The authors reported that individual sites
25 experienced years in which both H+ deposition and total O3 exposure were at least moderately
26 high (i.e., annual H+ deposition > 0.5 kg ha"1 and an annual O3 cumulative, sigmoidally weighted
27 exposure (W126) value > 50 ppm-h). With data compiled from all sites, it was found that
28 relatively acidic precipitation (pH < 4.31 on a weekly basis or pH < 4.23 on a daily basis)
29 occurred together with relatively high O3 levels (i.e., W126 values > 0.66 ppm-h for the same
30 week or W126 values > 0.18 ppm-h immediately before or after a rainfall event) approximately
31 20% of the time, and highly acidic precipitation (i.e, pH < 4.10 on a weekly basis or pH < 4.01
January 2005 AX3-143 DRAFT-DO NOT QUOTE OR CITE
-------
1 on a daily basis) occurred together with a high O3 level (i.e., W126 values > 1.46 ppm-h for the
2 same week or W126 values > 0.90 ppm-h immediately before or after a rainfall event)
3 approximately 6% of the time. Whether during the same week or before, during, or after a
4 precipitation event, correlations between O3 level and pH (or H+ deposition) were weak to
5 nonexistent. Sites most subject to relatively high levels of both H+ and O3 were located in the
6 eastern United States, often in mountainous areas.
7
8 AX3.8.6 Co-Occurrence of Ozone with Acid Cloudwater
9 In addition to the co-occurrence of O3 and acid precipitation, results have been reported on
10 the co-occurrence of O3 and acidic cloudwater in high-elevation forests. Vong and Guttorp
11 (1991) characterized the frequent O3-only and pH-only, single-pollutant episodes, as well as the
12 simultaneous and sequential co-occurrences of O3 and acidic cloudwater. The authors reported
13 that both simultaneous and sequential co-occurrences were observed a few times each month
14 above the cloud base. Episodes were classified by considering hourly O3 average concentrations
15 > 0.07 ppm and cloudwater events with pH < 3.2. The authors reported that simultaneous
16 occurrences of O3 and pH episodes occurred two to three times per month at two southern sites
17 (Mitchell, NC and Whitetop, VA) and the two northern sites (Whiteface Mountain, NY and
18 Moosilauke, NH) averaged one episode per month. No co-occurrences were observed at the
19 central Appalachian site (Shenandoah, VA), due to a much lower cloud frequency. Vong and
20 Guttorp (1991) reported that the simultaneous occurrences were usually of short duration
21 (mean =1.5 h/episode) and were followed by an O3-only episode. As would be expected,
22 O3-only episodes were longer than co-occurrences and pH episodes, averaging an 8-h duration.
23
24
25 AX3.9 THE METHODOLOGY FOR DETERMINING POLICY
26 RELEVANT BACKGROUND OZONE CONCENTRATIONS
27 AX3.9.1 Introduction
28 Policy-relevant background (PRB) O3 concentrations are used by the EPA to assess risks to
29 human and ecosystem health and to provide risk estimates associated with O3 produced by
30 anthropogenic sources in the continental North America (defined here as the United States,
31 Canada, and Mexico). Policy Relevant Background concentrations are those concentrations that
January 2005 AX3-144 DRAFT-DO NOT QUOTE OR CITE
-------
1 would result in the United States in the absence of anthropogenic emissions in North America.
2 Policy Relevant Background concentrations include contributions from natural sources
3 everywhere in the world and from anthropogenic sources outside North America. They are used
4 for assessing risks to human health in areas where O3 concentrations exceed the NAAQS.
5 Contributions to PRB O3 include: photochemical interactions involving natural emissions
6 of VOCs, NOX, and CO; the long-range transport of O3 and its precursors from outside North
7 America; and stratospheric-tropospheric exchange (STE). Processes involved in STE are
8 described in detail in Annex AX2.3. Natural sources of O3 precursors include biogenic
9 emissions, wildfires, and lightning. Biogenic emissions from agricultural activities are not
10 considered in the formation of PRB O3.
11 Most of the issues concerning the calculation of PRB O3 center on the origin of springtime
12 maxima in surface O3 concentrations observed at monitoring sites in relatively unpolluted areas
13 of the United States and on the capability of the current generation of global-scale, three-
14 dimensional chemistry transport models to correctly simulate their causes. These issues are
15 related to the causes of the occurrence of high O3 values, especially those averaged over 1-h to
16 8-h observed at O3 monitoring sites during late winter through spring (i.e., February to June).
17 The issues raised do not affect interpretations of the causes of summertime O3 episodes as
18 strongly. Summertime O3 episodes are mainly associated with slow-moving high-pressure
19 systems characterized by limited mixing between the planetary boundary layer and the free
20 troposphere (Section AX2.3).
21 Springtime maxima are observed at national parks mainly in the western United States that
22 are relatively clean (Section AX3.2.2; Figures AX3-77a,b) and at a number of other relatively
23 unpolluted monitoring sites throughout the Northern Hemisphere. Spring maxima in
24 tropospheric O3 were originally attributed to transport from the stratosphere by Regener (1941)
25 as cited by Junge (1963). Junge (1963) also cited measurements of springtime maxima in O3
26 concentrations at Mauna Loa (elevation 3400 m) and at Arkosa, Germany (an alpine location,
27 elevation 1860 m). Measurements of radioactive debris transported downward from the
28 stratosphere as the result of nuclear testing during the 1960s also show springtime maxima
29 (Ludwig et al., 1977). However, more recent studies (Lelieveld and Dentener, 2000; Browell
30 et al., 2003) attribute the springtime maximum in tropospheric O3 concentrations to tropospheric
31 production rather than transport from the stratosphere.
January 2005 AX3-145 DRAFT-DO NOT QUOTE OR CITE
-------
Yellowstone National Park
Maximum Hourly Concentration
1998-2001
in
D)
-------
1 Springtime O3 maxima were observed in low-lying surface measurements during the late
2 19th century. However, these measurements are quantitatively highly uncertain, and extreme
3 caution should be exercised in their use. Concentrations of approximately 0.036 ppm for the
4 daytime average and of 0.030 ppm for the nighttime averages were reported for Zagreb, Croatia
5 using the Schonbein method during the 1890s (Lisac and Grubise, 1991). Of the numerous
6 measurements of tropospheric O3 made in the 19th century, only the iodine catalyzed oxidation
7 of arsenite has been verified with modern laboratory methods. Kley et al. (1988) reconstructed
8 the apparatus used between 1876 and 1910 in Montsouris, outside Paris, and evaluated it for
9 accuracy and specificity. They concluded that ozone mixing ratios ranged from 5 to 16 ppb with
10 uncertainty of ±2 ppb. Interferences from SO2 were avoided as the Montsouris data were
11 selected to exclude air from Paris, the only source of high concentrations of SO2 at that time.
12 Uncertainties in the humidity correction to the Schonbein reading will lead to considerable
13 inaccuracies in the seasonal cycle established by this method (Pavelin et al., 1999). Because of
14 the uncertainties in the earlier methods, it is difficult to quantify the differences between surface
15 O3 concentrations measured in the last half of the 19th century at certain locations in either
16 Europe or North America with those currently monitored at remote locations in the world.
17 Observations of O3 profiles at a large number of sites indicate a positive gradient in O3
18 mixing ratios with increasing altitude in the troposphere and a springtime maximum in O3
19 concentrations in the upper troposphere (Logan, 1999). As discussed in Section AX2.3.1, STE
20 affects the middle and upper troposphere more than the lower troposphere. It is, therefore,
21 reasonable to suppose that the main cause of this positive gradient is STE. However, deep
22 convection transports pollutants upward and can result in an increase in the pollutant mixing
23 ratio with altitude downwind of surface source regions as shown in Figure AX3-78. This effect
24 can be seen in differences in ozonesonde profiles as one moves eastward across the United States
25 (Newchurch et al., 2003). In addition, O3 formed by lightning-generated NOX also contributes to
26 the vertical O3 gradient. (Lelieveld and Dentener, 2000). This O3 could be either background or
27 not, depending on the sources of radical precursors. Another contributing factor is the increase
28 of O3 lifetime with altitude (Wang et al., 1998). Free-tropospheric O3 is not predominantly of
29 stratospheric origin, nor is it all natural; it is mostly controlled by production within the
30 troposphere and includes a major anthropogenic enhancement (e.g., Berntsen et al., 1997;
31 Roelofs et al., 1997; Wild and Akimoto, 2001).
January 2005 AX3-147 DRAFT-DO NOT QUOTE OR CITE
-------
a. Carbon Monoxide
Surface
98°Wf 97C
Oklahoma City
96° 95° 94°
Longitude
b. Ozone
14
Surface
97C
Oklahoma City
96° 95° 94°
Longitude
93°
92°
Figure AX3-78. (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).
January 2005
AX3-148 DRAFT-DO NOT QUOTE OR CITE
-------
1 Stohl (2001), Wernli and Borqui (2002), Seo and Bowman (2002), James et al. (2003a,b),
2 Sprenger and Wernli (2003), and Sprenger et al. (2003) addressed the spatial and temporal
3 variability in stratosphere to troposphere transport. Both Stohl (2001) and Sprenger et al. (2003)
4 produced 1-year climatologies of tropopause folds based on a 1° by 1° gridded meteorological
5 model data set. They each found that the probability of deep folds (penetrating to the 800 hPa
6 level) was maximum during winter (December through February) with the highest frequency of
7 folding extending from Labrador down the east coast of North America. However, these deep
8 folds occurred in < 1% of the 6-h intervals for which meteorological data was assimilated for
9 grid points in the continental United States, with a higher frequency in Canada. They observed a
10 higher frequency of more shallow folds (penetrating to the upper troposphere) and medium folds
11 (penetrating to levels between 500 and 600 hPa) of about 10% and 1 to 2%, respectively. These
12 events occur preferentially across the subtropics and the southern United States. At higher
13 latitudes, other mechanisms such as the erosion of cut-off lows and the breakup of stratospheric
14 streamers are likely to play an important role in STE. A 15-year model climatology by Sprenger
15 and Wernli (2003) showed the consistent pattern of STE occurring over the primary storm tracks
16 along the Asian and North American coasts. This climatology, and the one of James et al.
17 (2003) both found that recent stratospheric air associated with deep intrusions are relatively
18 infrequent occurrences in these models. Thus, stratospheric intrusions are most likely to directly
19 affect the middle and upper troposphere, not the planetary boundary layer. However, this O3 can
20 still exchange with the planetary boundary layer through convection or through large-scale
21 subsidence as described later in this subsection and in Sections AX2.3.2, AX2.3.3, and AX2.3.4.
22 These results are in accord with the observations of Galeni et al. (2003) over Greece and those of
23 Ludwig et al. (1977) over the western United States. It should also be remembered that
24 stratospheric O3 injected into the upper troposphere is subject to chemical destruction as it is
25 transported downward toward the surface.
26 Ozone concentrations measured at RRMS in the Northern Hemisphere have been compiled
27 by Vingarzan (2004) and are reproduced here in Tables AX3-17, AX3-18, and AX3-19. Data
28 for annual mean/median concentrations show a broad range, as do annual maximum 1-h
29 concentrations. Generally, concentrations increase with elevation and the highest concentrations
30 are found during spring. The overall average of the annual median O3 concentrations at all sites
31 in the continental United States is about 30 ppb and excluding higher elevation sites it is about
January 2005 AX3-149 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-17. Range of Annual (January-December) Hourly Ozone Concentrations
(ppb) at Background Sites Around the World (CMDL, 2004)
Location
Elevation (in)
Period of Record
Range of Annual Means
Pt. Barrow, Alaska
Ny Alesund, Svalbard, Spitsbergen"
Mauna Loa, Hawaii0
11
475
3397
1992-2001
1989-1993
1992-2001
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 AX3-18. 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 (in)
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
37-45"
40-47"
19-22
14-18
38371
38-43"
19-24
Range of Annual Maxima
49-68
57-77
74-83
61-82
68-79"
68-102"
50-63
48-69
54-98
81-109"
50-64
"High elevation site.
Source: Vingarzan (2004).
Table AX3-19. 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, Ontario"
Chalk River, Ontario
Egbert, Ontario"
E.L.A., Ontario
Bratt's Lake, Saskatchewan
Esther, Alberta
Saturna Island, British Columbia
Elevation (in)
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.
'Stations affected by long-range transport of anthropogenic emissions.
Source: Vingarzan (2004).
January 2005
AX3-150 DRAFT-DO NOT QUOTE OR CITE
-------
1 24 ppb. Maximum concentrations may be related to stratospheric intrusions, wildfires, and
2 intercontinental or regional transport of pollution. However, it should be noted that all of these
3 sites are affected by anthropogenic emissions to some extent making an interpretation based on
4 these data alone problematic.
5 Daily 1-h maximum O3 concentrations exceeding 50 or 60 ppb are observed during late
6 winter and spring in southern Canada and at sites in national parks as shown in Tables
7 AX3-20, AX3-21, and Figure AX3-79. That these high values can occur during late winter
8 when there are low sun angles and cold temperatures may imply a negligible role for
9 photochemistry and a major role for stratospheric intrusions. However, active photochemistry
10 occurs even at high latitudes during late winter. Rapid O3 loss, apparently due to multiphase
11 chemistry involving bromine atoms (see Section 2.2.10) occurs in the Arctic marine boundary
12 layer. The Arctic throughout much of winter is characterized by low light levels, temperatures,
13 and precipitation, and can act as a reservoir for O3 precursors such as PAN and alkyl nitrates,
14 which build up and can then photolyze when sun angles are high enough during late winter and
15 early spring. Long-range transport of total odd nitrogen species (NOy) (defined in AX2.2.2) and
16 VOCs to Arctic regions can occur from midlatitude-source regions. In addition, O3 can be
17 transported from tropical areas in the upper troposphere followed by its subsidence at mid and
18 high latitudes (Wang et al., 1998).
19 Penkett (1983), and later Penkett and Brice (1986), first observed a spring peak in PAN at
20 high northern latitudes and hypothesized that winter emissions transported into the Arctic would
21 be mixed throughout a large region of the free troposphere and transformed into O3 as solar
22 radiation returned to the Arctic in the spring. Subsequent observations (Dickerson, 1985)
23 confirmed the presence of strata of high concentrations of reactive nitrogen compounds at high
24 latitudes in early spring. Bottenheim et al. (1990, 1993) observed a positive correlation between
25 O3 and NO2 in the Arctic spring. Jaffe et al. (1991) found NOy concentrations approaching 1 ppb
26 in Barrow, Alaska, in the spring and attributed them to long-range transport.
27 Beine et al. (1997) and Honrath et al. (1997) measured O3, PAN, and NOX in Alaska and
28 Svalbard, Norway and concluded that PAN decomposition can lead to photochemical O3
29 production. At Poker Flat, Alaska, O3 production was directly observable. Herring et al. (1997)
30 tracked springtime O3 maxima in Denali National Park, Alaska, an area one might presume to be
31 pristine. They measured NOX and hydrocarbons and concluded that, in the spring, O3 was
January 2005 AX3-151 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-20. Number of Hours > 0.05 ppm for Selected Rural O3 Monitoring in the United States
by Month for the Period 1988 to 2001
"^H
to
o
o
X
1
to
DRAFT-DO I
^
0
H
O
O
0
o
H
W
Site Name Month
Denali National Park, Alaska February
Denali National Park, Alaska March
Denali National Park, Alaska April
Denali National Park, Alaska May
Denali National Park, Alaska June
Yellowstone National Park, February
Wyoming
Yellowstone National Park, March
Wyoming
Yellowstone National Park, April
Wyoming
Yellowstone National Park, May
Wyoming
Yellowstone National Park, June
Wyoming
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
00000000 0000 14
0 0 0 0 0 0 0 0 52 0 122 17 0 24
217 0 2 0 64 10 31 21 12 51 236 119 0 302
26 1 0 24 10 17 1 54 97 35 79 29 0 98
00000000 27 00 22 06
0 0 11 3 0 21 6 0 1 5 252 23 77
194 2 4 95 26 285 14 7 98 150 509 286 307
228 17 16 217 62 311 185 65 163 385 517 242 461
225 2 10 196 47 180 193 212 216 289 458 240 350
58 67 139 33 28 116 81 94 149 78 212 181 172
-------
3
Table AX3-20 (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
to
o
o
^
X
OJ
OJ
O
!*
-n
H
1
O
o
2!
•£—\
0
H
O
O
0
o
H
W
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 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
February 000008000000
March 49 9 24 10 23 40 35 9 6 17 5 4
April 31 64 29 5 45 16 46 52 49 128 16 0
May 20 81 67 41 66 51 51 4 122 103 63 23
June 24 37 31 5 29 13 119 0 3 0 6 0
February 3 0 0 0 0 43 22 0 0 23 0 6 32 0
March 6 2 0 0 1 94 10 39 49 220 40 215 60 0
April 48 0 31 22 27 56 65 30 64 128 254 221 175 0
May 183 33 14 10 174 78 96 107 111 146 191 247 143 62
June 92 2 0 55 50 66 190 37 221 25 23 28 95
-------
3 Table AX3-21. 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
to
o
o
X
OJ
L/l
[\
O
£j
H
6
o
0
H
O
O
0
o
H
W
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
Glacier National Park,
Montana
Glacier National Park,
Montana
Month
February
March
April
May
June
February
March
April
May
June
February
March
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
00000000 00000
00000000000000
00000000000000
00000000200009
00000000000002
0 000000000600
37 0 0 0 0 0 0 0 1 0 120 1 4
59 0 0 29 0 20 4 0 0 64 158 11 77
20 0 0 61 3 42 24 38 26 54 169 49 139
8 7 18 2 1 13 0 0 22 4 27 43 18
000000000000
100000000000
-------
3
Table AX3-21 (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
to
o
o
X
OJ
3
o
H
6
o
0
H
O
O
0
o
H
W
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 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
April 00100000210
May 2 7 13 0 0 4 5 0 16 19 8
June 03 1010 16 0000
February 1000018000000
March 00000 34 052 15 094
April 9 0 1 0 0 5 8 0 17 2 57 24 41
May 77 6 0 0 40 9 40 2 27 46 53 139 43
June 30 0 0 28 17 5 113 12 115 0 5 0
2001
0
0
0
0
0
0
6
32
-------
54
Figure AX3-79. Maximum hourly average O3 concentrations at rural monitoring sites in
Canada and the United States in February from 1980 to 1998.
Source: Lefohnetal. (2001).
1 produced predominantly by photochemistry at a calculated rate of 1 to 4 ppb/day, implying that
2 the O3 observed could be produced on timescales ranging from about a week to a month.
3 Solberg et al. (1997) tracked the major components of NOy in remote Spitsbergen, Norway for
4 the first half of the year 1994. They observed high concentrations of PAN (800 ppt) peaking
5 simultaneously with O3 (45 to 50 ppb) and attributed this to the long-range transport of pollution
6 and to photochemical smog chemistry. These investigators concluded, in general, that large
7 regions of the Arctic store high concentrations of O3 precursors in the winter and substantial
8 quantities of O3 are produced by photochemical reactions in the spring. Although reactions with
9 high-activation-energy barriers may be ineffective, reactions with low- or no activation-energy
10 barriers (such as radical-radical reactions) or negative temperature dependencies will still
January 2005
AX3-156 DRAFT-DO NOT QUOTE OR CITE
-------
1 proceed. Indeed, active photochemistry is observed in the coldest regions of the stratosphere and
2 mesosphere. While it is expected that photochemical production rates of O3 will increase with
3 decreasing solar zenith angle as one moves southward from the locations noted above, it should
4 not be assumed that photochemical production of O3 does not occur during late winter and spring
5 at mid- and high-latitudes.
6 Perhaps the most thorough set of studies investigating causes of springtime maxima in
7 surface O3 has been performed as part of the AEROCE and NARE studies (cf Sections
8 AX2.3.4a,b) and TOPSE (Browell et al., 2003). These first two studies found that elevated or
9 surface O3 > 40 ppb at Bermuda, at least, arises from two distinct sources: the polluted North
10 American continent and the stratosphere. It was also found that these sources mix in the upper
11 troposphere before descending as shown in Figure AX3-80. (In general, air descending behind
12 cold fronts contains contributions from intercontinental transport and the stratosphere.) These
13 studies also concluded that it is impossible to determine sources of O3 without ancillary data that
14 could be used either as tracers of sources or to calculate photochemical production and loss rates.
15 In addition, subsiding back trajectories do not necessarily imply a free-tropospheric or
16 stratospheric origin for O3 observed at the surface, since the subsiding conditions are also
17 associated with strong inversions and clear skies that promote O3 production within the boundary
18 layer. Thus, it would be highly problematic to use observations alone as estimates of PRB O3
19 concentrations, especially for sites at or near sea level.
20 The IPCC Third Assessment Report (TAR) (2001) gave a large range of values for terms in
21 the tropospheric O3 budget. Estimates of O3 STE of ozone ranged over a factor of three from
22 391 to 1440 Tg/year in the twelve models included in the intercomparison; many of the models
23 included in that assessment overestimated O3 STE. However, the overestimates likely reflected
24 errors in assimilated winds in the upper troposphere (Douglass et al., 2003; Schoeberl et al.,
25 2003; Tan et al., 2004; van Noije et al., 2004). The budgets of tropospheric O3 calculated since
26 the IPCC TAR are shown in Table AX3-22. Simulation of stratospheric intrusions is notoriously
27 difficult in global models, and O3 STE is generally parameterized in these models. A model
28 intercomparison looking at actual STE events found significant variations in model results that
29 depended significantly on the type and horizontal resolution of the model (Meloen et al., 2003;
30 Cristofanelli et al., 2003). In particular, it was found that the Lagrangian perspective (as
31 opposed to the Eulerian perspective used in most global scale CTMs) was necessary to
January 2005 AX3-157 DRAFT-DO NOT QUOTE OR CITE
-------
Altitude
Stratospheric Air
»--
Cold Front (A)
Stratospheric Air
Cold Front (B)
West
-1500
Distance (km)
East
Figure AX3-80. Schematic diagram of a meteorological mechanism involved in high
concentrations of ozone 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).
1 characterize the depths and residence times of individual events (Sprenger and Wernli, 2003;
2 James et al., 2003a,b). A few studies of the magnitude of the O3 STE have been made based
3 on chemical observations in the lower stratosphere or combined chemistry and dynamics(e.g.,
4 450 Tg/year net global [Murphy and Fahey, 1994]; 510 Tg/year net global extratropics only
5 [Gettelman et al., 1997]; and 550 ± 140 Tg/year [Olsen et al., 2002]).
6 Even if the magnitude of cross-tropopause O3 fluxes in global CTMs are calculated
7 correctly in an annual mean sense, it should be noted that stratospheric intrusions occur
8 episodically following the passage of cold fronts at midlatitudes. Of major concern is the ability
9 of global-scale CTMs to simulate individual intrusions and the effects on surface O3
10 concentrations that may result during these events. As noted in Section AX2.3.1, these
11 intrusions occur in "ribbons" ~ 200 to 1000 km long, 100 to 300 km wide, and 1 to 4 km thick.
12 An example of a stratospheric intrusion occurred in Boulder, CO (EPA AQS Site 080130011;
January 2005
AX3-158 DRAFT-DO NOT QUOTE OR CITE
-------
3
Table AX3-22. Global Budgets of Tropospheric Ozone (Tg year *) for the Present-day Atmosphere
3
to
o
o
Reference Model
TAR4 11 models
Lelieveld and
Stratosphere-
Troposphere
Exchange
(STE)
770 ± 400
570
Chemical
Production 2
3420 ± 770
3310
Chemical
Loss2
3470 ± 520
3170
Dry
Deposition
770 ± 180
710
Burden
(Tg)
300 ± 30
350
Lifetime
(days)3
24 ±2
33
Dentener (2000)
X
OJ
VO
O
£
H
6
O
0
H
O
O
0
o
HH
H
W
Beyetal. (2001) 5 GEOS-CHEM 470 4900 4300
Horowitz et al. MOZART-2 340 5260 4750
(2003)
VonKuhlmann MATCH-MPIC 540 4560 4290
et al. (2003)
Shindell et al. (2003) GISS 417 NR6 NR
Park et al. (2004) UMD-CTM 480 NR NR
Rotman et al. (2004) IMPACT 660 NR NR
Wong et al. (2004) SUNYA/UiO GCCM 600 NR NR
1070
860
820
1470
1290
830
1100
320 22
360 23
290 21
349 NR
340 NR
NR NR
376 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 1 1 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. [2003b] more recent version of GEOS-CHEM gives identical rates and burdens.
6 Not reported.
-------
1 formally AIRS) on May 6, 1999 (Lefohn et al., 2001). At 1700 UTC (1000 hours LSI)
2 an hourly average concentration of 0.060 ppm was recorded and by 2100 UTC (1400 hours
3 LST), the maximum hourly average O3 concentration of 0.076 ppm was measured. At 0200
4 UTC on May 7, 1999 (1900 hours LST on May 6), the hourly average concentration declined to
5 0.059 ppm. Figure AX3-80 shows the O3 vertical profile that was recorded at Boulder, CO on
6 May 6, 1999, at 1802 UTC (1102 hours LST). The ragged vertical profile of O3 at > 4 km
7 reflects stratospheric air that has spiraled downward around an upper-level low and mixed with
8 tropospheric air along the way. Thus, stratospheric air which is normally extremely cold and dry
9 and rich in O3, loses its characteristics as it mixes downward. This process was described in
10 Section AX2.3.1 and illustrated in Figures AX2-7a,b and c.
11 The dimensions given above imply that individual intrusions are not resolved properly in
12 the current generation of global-scale CTMs (Figure AX3-81). However, as noted in Section
13 AX2.3.1, penetration of stratospheric air directly to the surface rarely occurs in the continental
14 United States. Rather, intrusions are more likely to affect the middle and upper troposphere,
15 providing a reservoir for O3 that can exchange with the planetary boundary layer. In this regard,
16 it is important that CTMs be able to spatially and temporally resolve the exchange between the
17 planetary boundary layer and the lower free troposphere properly.
18
19
20 AX3.9.2 CAPABILITY OF GLOBAL MODELS TO SIMULATE
21 TROPOSPHERIC OZONE
22 The current generation of global CTMs includes detailed representation of tropospheric
23 O3-NOX-VOC chemistry. Meteorological information is generally provided by global data
24 assimilation centers. The horizontal resolution is typically a few hundred km, the vertical
25 resolution is 0.1 to 1 km, and the effective temporal resolution is a few hours. These models can
26 simulate most of the observed variability in O3 and related species, although the coarse
27 resolution precludes simulation of fine-scale structures or localized extreme events. On the
28 synoptic scale, at least, all evidence indicates that global models are adequate tools to investigate
29 the factors controlling tropospheric O3. Stratosphere-troposphere exchange of O3 in global
30 models is generally parameterized. The parameterizations are typically constrained to match the
31 global mean O3 cross-tropopause flux, which is in turn constrained by a number of observational
January 2005 AX3-160 DRAFT-DO NOT QUOTE OR CITE
-------
-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 AX3-81. Ozone vertical profile at Boulder, Colorado on May 6,1999 at 1802 UTC
(1102 LST).
Source: Lefohnetal. (2001).
1
2
3
4
5
9
10
11
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
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
January 2005
AX3-161
DRAFT-DO NOT QUOTE OR CITE
-------
1 of the model have demonstrated that it simulates the tropospheric ozonesonde climatology
2 (Logan, 1999) generally to within 5 to 10 ppbv, including at mid- and high-latitudes (Bey et al.,
3 2001a) over Bermuda in spring (Li et al., 2002) and at sites along the Asian Pacific rim (Liu
4 et al., 2002). The phase of the seasonal cycle is reproduced to within 1 to 2 months (Bey et al.,
5 2001a; Li et al., 2002; Liu et al., 2002). An analysis of the 210Pb-7Be-O3 relationships observed
6 in three aircraft missions over the western Pacific indicates that the model does not
7 underestimate the stratospheric source of O3 (Liu et al., 2004). These studies and others (Li
8 et al., 2001; Bey et al., 2001b; Fusco and Logan, 2003) demonstrate that the model provides an
9 adequate simulation of O3 in the free troposphere at northern midlatitudes, including the mean
10 influence from the stratosphere. However, it cannot capture the structure and enhancements
11 associated with stratospheric intrusions, leading to mean O3 under-prediction in regions of
12 preferred stratospheric downwelling.
13 Fiore et al. (2002a, 2003b) presented a detailed evaluation of the model simulation for O3
14 and related species in surface air over the United States for the summer of 1995. They showed
15 that the model reproduces important features of observations including the high tail of O3
16 frequency distributions at sites in the eastern United States (although sub-grid-scale local peaks
17 are underestimated), the O3 to (NOy - NOX) relationships, and the "piston effect" (the highest O3
18 values exhibit the largest response to decreases in U.S. fossil fuel emissions from 1980 to 1995)
19 (Lefohn et al. 1998). Empirical orthogonal functions (EOFs) for the observed regional
20 variability of O3 over the eastern United States are also well reproduced, indicating that
21 GEOS-CHEM captures the synoptic-scale transport processes modulating surface O3
22 concentrations (Fiore et al., 2003b). One model shortcoming relevant for the discussion below is
23 that excessive convective mixing over the Gulf of Mexico and the Caribbean leads to an
24 overestimate of O3 concentrations in southerly flow over the southeastern United States.
25 Comparison of GEOS-CHEM with the Multiscale Air Quality Simulation Platform (MAQSIP)
26 regional air quality modeling system (Odman and Ingram, 1996) at 36 km2 horizontal resolution
27 showed that the models exhibit similar skill at capturing the observed variance in O3
28 concentrations with comparable model biases (Fiore et al., 2003b).
29
January 2005 AX3-162 DRAFT-DO NOT QUOTE OR CITE
-------
1 Simulations to Quantify Background Ozone Over the United States
2 The sources contributing to the O3 background over the United States were quantified by
3 Fiore et al. (2003a) with three simulations summarized in Table AX3-23: (1) a standard
4 simulation, (2) a background simulation in which North American anthropogenic NOX,
5 NMVOC, and CO emissions are set to zero, and (3) a natural O3 simulation in which global
6 anthropogenic NOX, NMVOC and CO emissions are set to zero and the CH4 concentration is set
7 to its 700 ppbv pre-industrial value. Anthropogenic emissions of NOX, nonmethane volatile
8 organic compounds (NMVOCs), and CO include contributions from fuel use, industry, and
9 fertilizer application. The difference between the standard and background simulations
10 represents regional pollution, i.e., the O3 enhancement from North American anthropogenic
11 emissions. The difference between the background and natural simulations represents
12 hemispheric pollution, i.e., the O3 enhancement from anthropogenic emissions outside North
13 America. Methane and NOX contribute most to hemispheric pollution (Fiore et al., 2002b).
14 A tagged O3 tracer simulation (Fiore et al., 2002a) was used to isolate the stratospheric
15 contribution to the background and yielded results that were quantitatively consistent with those
16 from a simulation in which O3 transport from the stratosphere to the troposphere was suppressed
17 (Fusco and Logan, 2003). All simulations were initialized in June 2000; results are reported for
18 March through October 2001.
19
20
Table AX3-23. Description of Simulations Used for Source Attribution
(Fiore et al., 2003a)
Simulation Description
Standard Present-day emissions as described in the text
Background North American anthropogenic NOX, NMVOC, and CO
Horizontal Resolution
2° x 2.5°
2° x 2.5°
emissions set to zero
Natural Global anthropogenic NOX, NMVOC, and CO emissions set 4° x 5°
to zero and CH4 concentration set to its 700 ppbv
preindustrial value
Stratospheric Tagged O3 tracer originating from the stratosphere in 2° x 2.5°
standard simulation
January 2005 AX3-163 DRAFT-DO NOT QUOTE OR CITE
-------
1 The standard and background simulations were conducted at 2° x 2.5° horizontal
2 resolution, but the natural simulation was conducted at 4° x 5° resolution to save on
3 computational time. There was no significant bias between 4° x 5° and 2° x 2.5° simulations
4 (Fiore et al., 2002a), particularly for a natural O3 simulation where surface concentrations were
5 controlled by large-scale processes.
6
7 AX3.9.3 Mean Background Concentrations: Spatial and Seasonal Variation
8 The analysis of Fiore et al. (2003a) focused on the 2001 observations from the Clean Air
9 Status and Trends Network (CASTNet) of rural and remote U.S. sites (Lavery et al., 2002)
10 (Figure AX3-82). Figure AX3-83 shows the mean seasonal cycle in afternoon (1300 to 1700
11 hours LT) O3 concentrations averaged over the CASTNet stations in each U.S. quadrant.
12 Measured O3 concentrations (asterisks) are highest in April to May, except in the Northeast
13 where they peak in June. Model results (triangles) are within 3 ppbv and 5 ppbv of the
14 observations for all months in the Northwest and Southwest, respectively. Model results for the
15 Northeast are too high by 5 to 8 ppbv when sampled at the CASTNet sites; the model is lower
16 when the ensemble of grid squares in the region are sampled (squares). The model is 8 to
17 12 ppbv too high over the Southeast in summer for reasons discussed in Section AX3.9.2.
18 Results from the background simulation (no anthropogenic emissions in North America;
19 see Table AX3-23) are shown as diamonds in Figure AX3-83. Mean afternoon background O3
20 ranges from 20 ppbv in the Northeast in summer to 35 ppbv in the Northwest in spring. It is
21 higher in the West than in the East because of higher elevation, deeper mixed layers, and longer
22 O3 lifetimes due to the arid climate (Fiore et al., 2002a). It is also higher in spring than in
23 summer, in part because of the seasonal maximum of stratospheric influence (Figure AX3-3)
24 and in part because of the longer lifetime of O3 (Wang et al., 1998).
25 Results from the natural O3 simulation (no anthropogenic emissions anywhere; Table
26 AX3-23) are shown as crosses in Figure AX3-83. Natural O3 concentrations are also highest in
27 the West and in spring when the influence of stratospheric O3 on the troposphere peaks (e.g.,
28 Holton et al., 1995). Monthly mean natural O3 concentration ranges are 18 to 23, 18 to 27, 13 to
29 20, and 15 to 21 ppbv in the Northwest, Southwest, Northeast, and Southeast, respectively. The
30 stratospheric contribution (X's) ranges from 7 ppbv in spring to 2 ppbv in summer.
31
January 2005 AX3-164 DRAFT-DO NOT QUOTE OR CITE
-------
.»
' voy-; ..
• - -x •
Y^k
X
X
X x: x
'. .- X MEV
•,cdw
G'AS '.''
Figure AX3-82. CASTNet stations in the continental United States for 2001. Sites
discussed in Section AX3.9.4 are labeled: OY = 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).
1 The difference between the background and natural simulations in Figure AX3-83
2 represents the monthly mean hemispheric pollution enhancement. This enhancement ranges
3 from 5 to 12 ppbv depending on region and season. It peaks in spring due to a longer O3 lifetime
4 (Wang et al., 1998) and to e efficient ventilation of pollution from the Asian continent (Liu et al.,
5 2003). In contrast to hemispheric pollution, the regional pollution influence (O3 produced from
6 North American anthropogenic emissions, shown as the difference between the squares and
7 diamonds) peaks in summer and is highest in the East. For the data in Figure AX3-83, it ranges
8 from 8 ppbv in the northern quadrants in March to over 30 ppbv in the eastern quadrants in
9 summer. Monthly mean observed O3 concentrations are influenced by both regional and
10 hemispheric pollution in all U.S. regions from March through October.
January 2005
AX3-165 DRAFT-DO NOT QUOTE OR CITE
-------
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 AX3-83.
Monthly mean afternoon (1300 to 1700 hours LT) concentrations (ppbv)
in surface air averaged over the CASTNet stations (Figure AX3-82) 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 AX3-23) to surface O3 are shown. The U.S. quadrants are
centered at 98.75° W and 37° N.
Source: Fiore et al. (2003a).
1 AX3.9.4 Frequency of High-Ozone Occurrences at Remote Sites
2 Lefohn et al. (2001) pointed out the frequent occurrence of high-O3 events (> 50 and
3 60 ppbv) at remote northern U.S. sites in spring. Fiore et al. (2003a) replicated the analysis of
4 Lefohn et al. (2001) at the four CASTNet sites that they examined: Denali National Park
5 (Alaska), Voyageurs National Park (Minnesota), Glacier National Park (Montana), and
6 Yellowstone National Park (Wyoming). The number of times that the hourly O3 observations at
January 2005
AX3-166 DRAFT-DO NOT QUOTE OR CITE
-------
1 the sites are > 50 and 60 ppbv for each month from March to October 2001 were then calculated
2 (see results in Table AX3-24) and compared with the same statistics for March to June
3 1988 to 1998 from Lefohn et al. (2001), to place the 2001 statistics in the context of other years.
4 More incidences of O3 above both thresholds occur at Denali National Park and Yellowstone
5 National Park in 2001 than in nearly all of the years analyzed by Lefohn et al. (2001). The
6 statistics at Glacier National Park, Montana indicate that 2001 had fewer than average incidences
7 of high-O3 events. At Voyageurs National Park in Minnesota, March and April 2001 had
8 lower-than-average frequencies of high-O3 events, but May and June were more typical.
9 Overall, 2001 was considered to be a suitable year for analysis of high-O3 events. Ozone
10 concentrations > 70 and 80 ppbv occurred most often in May through August in 2001 and were
11 found to be associated with regional pollution by Fiore et al. (2003a).
12 Fiore et al. (2003a) focused their analysis on mean O3 concentrations during the afternoon
13 hours (1300 to 1700 LT), as the comparison of model results with surface observations is most
14 appropriate in the afternoon when the observations are representative of a relatively deep mixed
15 layer (Fiore et al., 2002a). In addition, the GEOS-CHEM model does not provide independent
16 information on an hour-to-hour basis, because it is driven by meteorological fields that are
17 updated every 6-h and then interpolated. Fiore et al. (2003a) tested whether an analysis
18 restricted to these mean 1300 to 1700 LT surface concentrations captures the same frequency of
19 O3 > 50 and 60 ppbv that emerges from an analysis of the individual hourly concentrations over
20 24 hours. Results are reproduced here in Table AX3-24, which shows that the percentage of
21 individual afternoon (1300 to 1700 LT) hours when O3 > 50 and 60 ppbv at the CASTNet sites is
22 always greater than the percentage of all hourly occurrences above these thresholds, indicating
23 that elevated O3 concentrations preferentially occur in the afternoon. Furthermore, Table
24 AX3-24 shows that the frequency of observation of high-O3 events is not diminished when 4-h
25 average (1300 to 1700 LT) concentrations are considered, reflecting persistence in the duration
26 of these events. Model frequencies of high-O3 events from 1300 to 1700 LT at the CASTNet
27 sites are similar to observations in spring, as shown in Table AX3-24, and about 10% higher in
28 the summer, largely because of the positive model bias in the Southeast discussed in Section
29 AX3.9.2.
30
January 2005 AX3-167 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-24. 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).
January 2005
AX3-168 DRAFT-DO NOT QUOTE OR CITE
-------
1 NATURAL VERSUS ANTHROPOGENIC CONTRIBUTIONS TO
2 HIGH-OZONE OCCURRENCES
3 Figure AX3-84, reproduced from Fiore et al. (2003a), shows probability distributions of
4 daily mean afternoon (1300 to 1700 LT) O3 concentrations in surface air at the CASTNet sites
5 for March through October 2001. Model distributions for background, natural, and stratospheric
6 O3 (Table AX3-23) are also shown. The background (long-dashed line) ranges from 10 to
7 50 ppbv with most values in the 20 to 35 ppbv range. The full 10 to 50 ppbv range of
8 background predicted here encompasses the previous 25 to 45 ppbv estimates shown in
9 Table 3-8. However, background estimates from observations tend to be at the higher end of the
10 range (25 to 45 ppbv), while these results, as well as those from prior modeling studies
11 (Table 3-8) indicate that background O3 concentrations in surface air are usually below 40 ppbv.
12 The background O3 concentrations derived from observations may be overestimated if
13 observations at remote and rural sites contain some influence from regional pollution (as shown
14 below to occur in the model), or if the O3 versus NOy - NOX correlation is affected by different
15 relative removal rates of O3 and NOy (Trainer et al., 1993). Natural O3 concentrations
16 (short-dashed line) are generally in the 10 to 25 ppbv range and never exceed 40 ppbv. The
17 range of the hemispheric pollution enhancement (the difference between the background and
18 natural O3 concentrations) is typically 4 to 12 ppbv and only rarely exceeds 20 ppbv (< 1% total
19 incidences). The stratospheric contribution (dotted line) is always less than 20 ppbv and usually
20 below 10 ppbv. Time series for specific sites are presented below.
21
22 CASE STUDIES: INFLUENCE OF THE BACKGROUND ON ELEVATED OZONE
23 EVENTS IN SPRING
24 High-O3 events were previously attributed to natural processes by Lefohn et al. (2001) at:
25 Voyageurs National Park, Minnesota in June and Yellowstone National Park, Wyoming in
26 March through May. Fiore et al. (2003a) used observations from CASTNet stations in
27 conjunction with GEOS-CHEM model simulations to deconstruct the observed concentrations
28 into anthropogenic and natural contributions.
29 At Voyageurs National Park in 2001, O3 concentrations > 60 ppbv occurred frequently in
30 June but rarely later in summer (Table AX3-24). A similar pattern was observed in 1995 and
31 1997 and was used to argue that photochemical activity was probably not responsible for these
32 events (Lefohn et al. 2001). Figure AX3-85 from Fiore et al. (2003a) shows that GEOS-CFffiM
January 2005 AX3-169 DRAFT-DO NOT QUOTE OR CITE
-------
0.10n
Figure AX3-84.
-D
Q.
-P 0.05-
co
.Q
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).
12
Q.
O
N
O
80
60
40
20
10
Voyageurs HP, Minnesota (93W, 48N)
20
30
May
19
June
29
Figure AX3-85.
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).
January 2005
AX3-170 DRAFT-DO NOT QUOTE OR CITE
-------
1 captures much of the day-to-day variability in observed concentrations from mid-May through
2 June, including the occurrence and magnitude of high-O3 events. The simulated background
3 contribution (diamonds) ranges from 15 to 36 ppbv with a 25 ppbv mean. The natural O3 level
4 (crosses) is 15 ppbv on average and varies from 9 to 23 ppbv. The stratospheric contribution
5 (X's) is always < 7 ppbv. The dominant contribution to the high-O3 events on June 26 and 29 is
6 from regional pollution (44 and 50 ppbv on June 26 and 29, respectively, calculated as the
7 difference between the triangles and diamonds in Figure AX3-85). The background contribution
8 (diamonds) is < 30 ppbv on both days, and is composed of a 20 ppbv natural contribution (which
9 includes 2 ppbv of stratospheric origin) and a 5 ppbv enhancement from hemispheric pollution
10 (the difference between the diamonds and crosses). Beyond these two high-O3 events,
11 Figure AX3-85 shows that regional pollution drives most of the simulated day-to-day variability
12 and explains all events above 50 ppbv. In 2001, monthly mean observed and simulated O3
13 concentrations are lower in July (37 and 42 ppbv, respectively) and August (35 and 36 ppbv)
14 than in June (44 and 45 ppbv). Fiore et al. (2003a) hypothesized that the lower mean O3 and the
15 lack of O3 > 60 ppbv in July and August reflects a stronger Bermuda high-pressure system
16 sweeping pollution from southern regions eastward before it could reach Voyageurs National
17 Park.
18 Frequently observed concentrations of O3 between 60 to 80 ppbv at Yellowstone NP in
19 spring (Figures AX3-77a,b) have been attributed by Lefohn et al. (2001) to natural sources,
20 because they occur before local park traffic starts and back-trajectories do not suggest influence
21 from long-range transport of anthropogenic sources. More hours with O3 > 60 ppbv occur in
22 April and May of 2001 (Table AX3-24) than in the years analyzed by Lefohn et al. (2001). Fiore
23 et al. (2003a) used GEOS-CFIEM to interpret these events; results are shown in Figure AX3-86.
24 The mean background, natural, and stratospheric O3 contributions in March to May are higher at
25 Yellowstone (38, 22, and 8 ppbv, respectively) as compared to 27, 18, and 5 ppbv at the two
26 eastern sites previously discussed. The larger stratospheric contribution at Yellowstone reflects
27 the high elevation of the site (2.5 km). Fiore et al. (2003a) argued that the background at
28 Yellowstone National Park should be considered an upper limit for U.S. PRB O3 concentrations,
29 because of its high elevation. While Yellowstone receives a higher background concentration
30 than the eastern sites, the model shows that regional pollution from North American
31 anthropogenic emissions (difference between the triangles and diamonds) contributes an
January 2005 AX3-171 DRAFT-DO NOT QUOTE OR CITE
-------
80
"> 60
.Q
Q.
a
7 40
o
O 20
0
Yellowstone NP, Wyoming (11OW, 45N)
-i
20
March
29
April
19
May
Figure AX3-86. Same as Figure AX3-85 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).
1 additional 10 to 20 ppbv to the highest observed concentrations in April and May. One should
2 not assume that regional photochemistry is inactive in spring.
3 Higher-altitude western sites are more frequent recipients of subsidence events that
4 transport high concentrations of O3 from the free troposphere to the surface. Cooper and Moody
5 (2000) cautioned that observations from elevated sites are not generally representative of
6 lower-altitude sites. At Yellowstone, the background O3 rarely exceeds 40 ppbv, but it is even
1 lower in the East. This point is illustrated in Figure AX3-87, from Fiore et al. (2003a), with time
8 series at representative western and southeastern CASTNet sites for the month of March, when
9 the relative contribution of the background should be high. At the western sites, the background
10 is often near 40 ppbv but total surface O3 concentrations are rarely above 60 ppbv. While
11 variations in the background play a role in governing the observed total O3 variability at these
12 sites, regional pollution also contributes. Background concentrations are lower (often
13 < 30 ppbv) in the southeastern states where regional photochemical production drives much of
14 the observed variability. Cooper and Moody (2000) have previously shown that the high O3
15 concentrations at an elevated, regionally representative site in the eastern United States in spring
January 2005
AX3-172 DRAFT-DO NOT QUOTE OR CITE
-------
&
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 AX3-87. Same as Figure AX3-86 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).
January 2005
AX3-173 DRAFT-DO NOT QUOTE OR CITE
-------
1 coincide with high temperatures and anticyclonic circulation, conditions conducive to
2 photochemical O3 production. Peak O3 concentrations in this region, mainly at lower elevations,
3 are associated with lower background concentrations because chemical and depositional loss
4 during stagnant meteorological conditions suppress mixing between the boundary layer and the
5 free troposphere (Fiore et al., 2002a). Surface O3 concentrations > 80 ppbv could conceivably
6 occur when stratospheric intrusions reach the surface. However, based on information given in
7 Section AX2.3.2, these events are rare.
8
9
10 AX3.10 INDOOR SOURCES AND CONCENTRATIONS
11 In the past, indoor O3 concentrations/exposure were not considered of importance.
12 However, since most people spend a significant amount of time indoors, most exposure to air
13 pollutants of outdoor origin occur indoors. This section will address the sources and
14 concentrations of indoor O3. This section will also briefly discuss those factors that affect indoor
15 O3 concentrations. A comprehensive review of O3 in the indoor environment appears in
16 Weschler (2000).
17
18 AX3.10.1 Sources and Emissions of Indoor Ozone
19 Ozone enters the indoor environment primarily through infiltration from outdoors through
20 cracks and crevices in the building envelope (unintentional and uncontrolled ventilation) and
21 through building components such as windows and doors and ventilation systems (natural and
22 controlled ventilation). Natural ventilation is driven by the natural forces of wind and
23 temperature. Possible indoor sources of O3 include office equipment (photocopiers, facsimile
24 machines, and laser printers) and air cleaners (electrostatic air filters and precipitators and O3
25 generators). Generally O3 emissions from photocopiers and laser printers are limited due to
26 installed filtering systems (Black and Worthan, 1999; Leovic et al., 1996; Aldrich et al., 1995).
27 However, emissions increase under improper maintenance conditions (Leovic et al., 1996).
28 Well-maintained photocopiers and laser printers usually emit low levels of O3 by catalytically
29 reducing the O3 to oxygen (Aldrich et al., 1995). Leovic et al. (1996, 1998) measured O3
30 emissions from four dry-process photocopiers. Ozone emissions ranged from 1300 to
31 7900|ig/h.
January 2005 AX3-174 DRAFT-DO NOT QUOTE OR CITE
-------
1 Most electrostatic air filters and precipitators are designed to minimize the production of
2 O3. However, if excessive arcing occurs, the units can emit a significant amount of O3 into the
3 indoor environment (Weschler, 2000). Niu et al. (2001) measured O3 emissions from 27 air
4 cleaners that used ionization processes to remove particulates. The test were conducted in a
5 2 x 2 x 1.60 m3 stainless steel environmental chamber. The tests were terminated after 1.5 h
6 if no increase in O3 concentration was noted. If an increase in O3 was noted, the test was
7 continued, in some cases, for up to 20 h. Most of the evaluated units emitted no O3 or only
8 small amounts. Five units were found to emit O3 ranging from 56 to 2757 |ig/h.
9 There are many brands and models of O3 generators commercially available. The amount
10 of O3 emitted by each unit depends on the size of the unit. Ozone emission rates have been
11 reported to range from tens to thousands of micrograms per hour (Weschler, 2000; Steiber,
12 1995). Ozone emissions supposedly can be regulated using the units control features. However,
13 available information suggests that O3 output may not be proportional to the control setting.
14 Some units are equipped with a sensor that automatically controls O3 output by turning the unit
15 on and off. The effectiveness of the sensor is unknown (U.S. Environmental Protection Agency,
16 2002).
17
18 AX3.10.2 Factors Affecting Ozone Concentrations Indoors
19 In the absence of an indoor source, O3 concentrations in indoor environments will depend
20 on the outdoor O3 concentration, the air exchange rate (AER) or outdoor infiltration, indoor
21 circulation rate, O3 removal by indoor surfaces, reactions with other indoor pollutants, and
22 temperature and humidity. Indoor O3 concentrations generally closely track outdoor O3
23 concentrations. Since outdoor O3 concentrations are generally higher during the warmer months,
24 indoor concentrations will also be highest during that time period. (See discussion on ambient
25 concentrations of O3 earlier in this chapter.)
26
27 Air Exchange Rates
28 Indoor O3 increases with increasing air exchange rate (Gold et al., 1996; Lee et al., 1999;
29 Jakobi and Fabian, 1997). The AER is the balance of the flow of air in and out of a
30 microenvironment. Infiltration through unintentional openings in the building envelope is the
31 dominant mechanism for residential air exchange. However, natural ventilation, airflow through
January 2005 AX3-175 DRAFT-DO NOT QUOTE OR CITE
-------
1 opened windows and doors, also influences air exchange in residential buildings. Forced or
2 mechanical ventilation is the dominant mechanism for air exchange in nonresidential buildings.
3 Air exchange rates vary depending on the temperature differences, wind effects,
4 geographical region, type of heating/mechanical ventilation system, and building type (U.S.
5 Environmental Protection Agency, 1997; Weschler and Shields, 2000; Colome et al., 1994;
6 Johnson et al., 2004). Air exchange rates are generally higher during the summer and lower
7 during the winter months (Wilson et al., 1996; Murray and Burmaster, 1995; Colome et al.,
8 1994; Research Triangle Institute, 1990).
9 Howard-Reed et al. (2002) determined that opening a window or exterior door causes the
10 greatest increase in AERs with differences between the indoor and outdoor temperature being
11 important when the windows were closed. Johnson and Long (2004) conducted a pilot study to
12 evaluate the frequency that windows were left open in a community. They found that a visual
13 2-h survey could be used to estimate the frequency that windows are left open. The occupancy,
14 season, housing density, absence of air conditioning, and wind speed were factors in whether the
15 windows were open.
16 Johnson et al. (2004) conducted a study using scripted ventilation conditions to identify
17 those factors that affected air exchange inside a test house in Columbus, OH. The test house was
18 a wood-framed, split-level structure with aluminum siding covering the wood outer walls. The
19 house had one exterior door located in the front and another at the rear of the house, single-pane
20 glazed windows, central gas heat, a window air-conditioning unit, and ceiling fans in three
21 rooms. Eighteen scenarios with unique air flow conditions were evaluated to determine the
22 effect on the AER. The elements of the scenarios included: exteriors doors open/closed, interior
23 doors open/closed, heater on/off, air conditioner on/off, a ceiling fans on/off. The lower level
24 was sealed off during the study. The various scenarios were evaluated during the winter season.
25 The average AER for all scenarios ranged from 0.36 to 15.8 h"1. When all windows and doors
26 were closed, the AER ranged from 0.36 to 2.29 h"1 (0.77 h"1 geometric mean). When at least one
27 exterior door or window was open the AER ranged from 0.5 to 15.8 h"1 (1.98 h"1 geometric
28 mean).
29 One of the most comprehensive evaluations of AERs for residential structures was
30 conducted by Murray and Burmaster (1995) using data compiled by Brookhaven National
31 Laboratories. The analysis used data on AERs for 2,844 residential structures in four regions
January 2005 AX3-176 DRAFT-DO NOT QUOTE OR CITE
-------
1 based on heating degree days. Data were also separated by seasons (winter, spring, summer, and
2 fall). The AER for all seasons across all regions was 0.76 IT1 (arithmetic). Table AX3-25 lists
3 the mean AER for the various seasons and regions. The data does not represent all areas of the
4 country.
5 Air exchange rates for 49 nonresidential buildings (14 schools, 22 offices, and 13 retail
6 establishments) in California were reported by Lagus Applied Technology, Inc. (1995). Average
7 mean (median) AERs were 2.45 (2.24), 1.35 (1.09), and 2.22 (1.79) IT1 for schools, offices, and
8 retail establishments, respectively. Air infiltration rates for 40 of the 49 buildings were 0.32, of
9 0.31, and 1.12 IT1 for schools, offices, and retail establishments, respectively. Air exchange
10 rates for 40 nonresidential buildings in Oregon and Washington (Turk et al., 1989) averaged
11 1.5(1.3) h'1 (mean median). The geometric mean of the AERs for six garages was 1.6 h'1 (Marr
12 et al., 1998). Park et al. (1998) reported AERs for three stationary cars (cars varied by age)
13 under different ventilation conditions. Air exchange rates ranged from 1.0 to 3.0 h'1 for
14 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
15 windows closed and fan on recirculation (two cars tested), and 36.2 to 47.5 h'1 for windows
16 closed and fan on fresh air (one car tested). An average AER of 13.1 h'1 was reported by Ott
17 et al. (1992) for a station wagon moving at 20 mph with the windows closed.
18
19 Ozone Removal Processes
20 The most important removal process for O3 in the indoor environment is deposition on
21 indoor surfaces. When O3 enters the indoor environment it is deposited on and reacts with
22 indoor surfaces. The rate of deposition is material specific. The removal rate will depend on the
23 indoor dimensions, surface coverings, and furnishings. Smaller rooms generally have larger
24 surface-to-volume ratios (A/V) and remove O3 faster. Fleecy materials, such as carpets, have
25 larger A/V ratios and remove O3 faster than smooth surfaces (Weschler, 2000). Weschler et al.
26 (1992) found that O3 reacts with carpet, decreasing O3 concentrations and increasing emissions
27 of formaldehyde, acetaldehyde, and other C5-C10 aldehydes. Off-gassing of
28 4-phenylcyclohexene, 4-vinylcyclohexene, and styrene was reduced. The rate of O3 reaction
29 with carpet diminishes with cumulative O3 exposure (Morrison and Nazaroff, 2000, 2002).
30 Reiss et al. (1995) reported significant quantities of acetic acid and smaller quantities of formic
31 acid off-gassing from O3 reactions with latex paint. The emission rate also was relative
January 2005 AX3-177 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-25. Air Exchange Rates in Residences by Season and Region of the Country
Season3
All
1
2
3
4
All
All
All
All
1
1
1
1
2
2
2
2
o
3
3
3
3
4
4
4
4
Region
All
All
All
All
All
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Sample
Size
2844
1139
1051
529
125
467
496
332
1549
161
428
96
454
254
43
165
589
5b
2b
34
488
47
23b
37
18b
Mean
0.76
0.55
0.65
1.5
0.41
0.4
0.55
0.55
0.98
0.36
0.57
0.47
0.63
0.44
0.52
0.59
0.77
0.82
1.31
0.68
1.57
0.25
0.35
0.51
0.72
Std. Total in Average No. of Std. Dev. of Heating
Dev. Region Days Heated0 Degree Days0
0.88 2844
0.46
0.57
1.53
0.58
0.3 467 7324 468
0.48 496 5751 336
0.42 332 4333 729
1.09 1549 1995 115
0.28
0.43
0.4
0.52
0.31
0.91
0.43
0.62
0.69
na
0.5
1.56
0.15
0.18
0.25
1.43
"Season 1: December, January, February; Season 2: March, April, May; Season 3: June, July, August;
Season 4: September, October, November.
bNote: Small sample size, n < 25.
"Estimated using locations of residences evaluated in the region.
Source: Adapted from Murry and Burmaster (1995).
January 2005 AX3-178 DRAFT-DO NOT QUOTE OR CITE
-------
1 humidity-dependent, increasing with higher relative humidity. Klen0 et al. (2001) evaluated O3
2 removal by several aged (1- to 120-month) but not used building materials (nylon carpet,
3 linoleum, painted gypsum board, hand polished stainless steel, oiled beech parquet, melamine-
4 coated particle board, and glass plate). Initially, O3 removal was high for all specimens tested
5 with the exception of the glass plate. Ozone removal for one carpet specimen and the painted
6 gypsum board remained high throughout the study. For the oiled beech parquet and melamine-
7 coated particle board, O3 removal leveled off to a moderate rate. Morrison et al. (1998) reported
8 that little O3 (-9%) is removed by the ductwork of lined ventilation systems. The removal
9 efficiency decreases with continued exposure to O3, and unlined ductwork is less efficient in
10 removing O3. Liu and Nazaroff (2001) reported that O3 is scavenged by fiberglass insulation.
11 More O3 was scavenged (60 to 90%) by fiberglass that had not been previously exposed.
12 Table AX3-26 lists the removal rates for O3 in different indoor environments.
13 Jaboki and Fabian (1997) found the O3 decays exponentially. They examined O3 decay in
14 several closed rooms and in a car after a period of intensive ventilation. Figure AX3-88 plots the
15 O3 decay rates in these environments.
16 Several studies have examined factors within homes that may scavenge O3 and lead to
17 decreased O3 concentrations. Ozone reacts indoors with NOX and VOCs in an analogous fashion
18 to that occurring in the ambient atmosphere (Lee et al., 1999; Wainman et al., 2000; Weschler
19 and Shields, 1997). These reactions produce related oxidant species while reducing indoor O3
20 levels. Lee et al. (1999) studied 43 homes in Upland, CA in the Los Angeles metropolitan
21 region and reported that O3 declined faster in homes with more bedrooms, greater amounts of
22 carpeting, and lower ceilings (all of which alter the A/V ratio) and with the use of air
23 conditioning. Homes with air conditioning had indoor/outdoor (I/O) ratios of 0.07, 0.09, and
24 0.11. Homes without air conditioning had an I/O ratio of 0.68 ± 0.18. Closed windows and
25 doors combined with the use of an air cleaner resulted in an I/O ratio of 0.15. The O3 I/O ratio
26 was < 0.2 in homes with gas stoves.
27 Ozone reacts with terpenes from wood products, solvents, or odorants to produce
28 submicron particles (Wainman et al., 2000; Weschler and Shields, 1999; Grosjean and Grosjean,
29 1996, 1998). Wainman et al. (2000) suggested that the PM2 5 concentrations could be in excess
30 of 20 |ig/m3 in the indoor environment as a result of O3/terpene reactions. Indoor hydroxy
31 radical (HO») concentrations increase nonlinearly with increased indoor O3 concentrations and
January 2005 AX3-179 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-26. Rate Constants (h l) for the Removal of Ozone by Surfaces in Different
Indoor Environments
Indoor environment
Surface Removal Rate,
k,, (A/V), h-1
Reference
Aluminum Room (11.9 m3)
Stainless Steel Room (14.9 m3)
Bedroom (40.8m3)
Office (55.2 m3)
Home (no forced air)
Home (forced air)
3.2
1.4
7.2
4.0
2.9
5.4
Mueller et at. (1973)
Saberskyetal. (1973)
Department Store
Office (24.1 m3)
Office (20.7 m3)
Office/Lab
4.3
4.0
4.3
4.3
Thompson et al. (1973)
Allen etal. (1978)
Shair and Heitner (1974)
Office/Lab
Office/Lab
13 Buildings, 24 ventilation
systems
Museum
Museum
3.2
3.6
3.6a
4.3
4.3
Shair (1981)
Nazaroff and Cass (1986)
Office/Lab
Office/Lab
Office
Lab
Cleanroom
4
3.2
2.5
2.5
7.6
Weschler etal. (1989)
Telephone Office
43 Homes
0.8 - 1.0b
2.8 ±1.3
Weschler et at. (1994)b
Lee et al. (1999)
a A/V = assumes surface area to volume ratio = 2.8 m'
b Large office, small A/V
Source: Weschler (2000).
January 2005
AX3-180 DRAFT-DO NOT QUOTE OR CITE
-------
(0
+J
'E
I
3
O
£
fO
o
O3 - Car O3 - Office
0 10 20 30 40 50 60 70 80
90 100 110 120
Elapsed Time (minutes)
Figure AX3-88. 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).
1 indoor alkene emissions. Sarwar et al. (2002) suggested that the HO reacts with terpenes to
2 produce products with low vapor pressures that contribute to fine particle growth.
3
4 AX3.10.3 Ozone Concentrations in Microenvironments
5 Monitored Concentrations in Microenvironments
6 The 1996 O3 AQCD for Ozone (U.S. Environmental Protection Agency, 1996a) reported
7 O3 I/O ratios for a variety of indoor environments including homes, office/laboratories, a
8 hospital, museums, a school room, and automobiles and other vehicles. Ozone I/O ratios ranged
9 from 0.09 to 1.0 for residences, 0.19 to 0.8 for offices/laboratories, hospital and school rooms,
10 and 0.03 to 0.87 for museums and art galleries. The higher O3 ratios were generally noted in
11 indoor environments with high AERs or 100% outside air intake. Studies published since
12 completion of the 1996 O3 AQCD are discussed in this section. The findings of the more recent
13 studies on O3 I/O ratios are included in Table AX3-27.
January 2005
AX3-181
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-27. Indoor/Outdoor Ozone Ratios
to
o
o
Location and Ventilation
Conditions
Indoor/Outdoor Ratio
Mean (range)
Comments
Reference
>
X
oo
to
fe
H
6
O
O
H
O
o
HH
H
W
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.
-------
3
Table AX3-27 (cont'd). Indoor/Outdoor Ozone Ratios
to
o
X
r N
OJ
i
oo
OJ
o
K^
fe
H
^
O
2|
0
H
O
d
o
i
w
o
Location and Ventilation
Conditions
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
Indoor/Outdoor Ratio
Mean (range)
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
Comments
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 avg concentrations.
Four hundred and eighty -one samples collected inside and immediately
outside of home from February to December. Ratios based on 24-h avg
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 avg indoor concentrations.
Indoor concentrations were dependent on the type of ventilation.
Reference
Romieu
etal. (1998)
Avol et al.
(1998a)
Lee et al.
(1999)
Jakobi and
Fabian
(1997)
O
HH
H
W
-------
3
Table AX3-27 (cont'd). Indoor/Outdoor Ozone Ratios
3
to
o
o
Location and Ventilation
Conditions
Montpellier, France, Homes (110)
Southern CA, Homes
Upland Mountains
Indoor/Outdoor Ratio
Mean (range)
0.41
0.68 ±0.18
(windows open)
0.07-0.11
(AC used)
Comments
Ozone measurements were made over 5 -day periods in and outside of 21
homes during the summer and winter months. The winter I/O ratio was
0.3 1 compared to 0.46 during the summer months.
Ozone measurements were taken at 119 homes (57 in Upland and 62 in
towns located in the mountains) during April and May. I/O ratios were
based on average monthly outdoor concentrations and average weekly
indoor concentrations. I/O ratio was associated with the home location,
number of bedrooms, and the presence of an air conditioner. I/O ratios
based on subset of the homes.
Reference
Bernard
etal. (1999)
Geyh et al.
(2000);
Lee et al.
(2002)
>
X
oo
fe
H
6
O
O
H
O
Krakow, Poland
Museums (5)
0.13 -0.42
Buildings, Greece
Thessaloniki
Athens
0.24 ±0.18 (0.01 to 0.75)
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.
ND = not detectable.
o
HH
H
W
-------
1 Northeast States for Coordinated Air Use Management (NESCAUM, 2002) monitored
2 levels of O3 inside and outside of nine schools located in the New England states. The schools
3 represented a variety of environmental conditions in terms of ambient O3 concentration, sources,
4 geographic location, population density, traffic patterns, and building types. Schools were
5 monitored during the summer months to establish baseline O3 concentrations and again during
6 the fall after classes started. A monitor was placed directly outside of the school entrance and
7 50 feet away from the entrance in the hall. Where available, monitors were placed at locations
8 identified as "problem" classrooms, classrooms with carpeting, or in rooms close to outdoor
9 sources of O3. As expected, outdoor concentrations of O3 were higher than those found indoors.
10 Averaged O3 concentrations were low during the early morning hours (7:30 a.m.) but peaked to
11 approximately 25 and 40 ppb around 1:30 p.m. indoors and outdoors, respectively.
12 Gold et al. (1996) compared indoor and outdoor O3 concentrations in classrooms in Mexico
13 City under three different ventilation conditions: windows/doors open, air cleaner off;
14 windows/doors closed, air cleaner off; and windows/doors closed, air cleaner on. Two-min
15 averaged outdoor O3 levels varied between 64 and 361 ppb, while indoor O3 concentrations
16 ranged from 0 to 247 ppb. The highest indoor O3 concentrations were noted when the
17 windows/doors were open. The AERs were estimated to be 1.1, 2.1, and 2.5 IT1 for low,
18 medium, and high air flow rates, respectively. The authors indicated that the indoor levels, and
19 therefore O3 exposure to students in schools, can be reduced to < 80 ppb by closing windows and
20 doors even when ambient O3 levels reach 300 ppb.
21 In a second Mexico City study, Romieu et al. (1998) measured O3 concentrations inside
22 and outside of 145 homes and three schools. Measurements were made between November and
23 June. Most of the homes were large and did not have air conditioning. Ninety-five percent of
24 the homes had carpeting, 13% used air filters, and 84% used humidifiers. Thirty-five percent of
25 the homeowners reported that they opened windows frequently between the hours of 10 a.m. and
26 4 p.m., while 43% opened windows sometimes and 22% reported that they never opened
27 windows during that time period. Homes were monitored for continuous 24-h periods for
28 14 consecutive days. Schools were monitored from 8 a.m. to 1 p.m. or continuously for 24 h.
29 During the school monitoring periods the windows were frequently left open and the doors were
30 constantly being opened and closed. The mean indoor and outdoor O3 concentrations during 5-h
31 measurements at the schools were 22 ppb and 56 to 73 ppb, respectively. The mean indoor and
January 2005 AX3-185 DRAFT-DO NOT QUOTE OR CITE
-------
1 outdoor O3 concentrations for measurements made over a 7- and 14-day period at the test homes
2 were 5 and 27 ppb and 7 and 37 ppb, respectively. Ozone concentrations inside of homes were
3 dependent on the presence of carpeting, the use of air filters, and whether the windows were
4 open or closed. Air exchange rates were not reported.
5 Reiss et al. (1995) compared indoor and outdoor O3 concentrations for residences in the
6 Boston, MA area. Four residences were monitored during the winter months and nine residences
7 during the summer months. Outside monitors were placed ~1 m away from the house.
8 Monitoring was conducted over a continuous 24-h period. There were no indoor sources of O3.
9 Indoor O3 concentrations were higher during the summer months, with concentrations reaching
10 34.2 ppb. Indoor O3 concentrations reached as high as 3.3 ppb during the winter monitoring
11 period. In one instance, O3 concentrations were higher indoors than outdoors. The authors
12 attributed that finding to analytical difficulties. Outdoor O3 concentrations were generally higher
13 during the summer monitoring period, with concentrations reaching 51.8 ppb. Indoor
14 concentrations were dependent on the outdoor O3 concentrations and AER. Indoor and
15 outdoor O3 concentrations, including the AERs at the times of the monitoring are included in
16 Table AX3-28.
17 Avol et al. (1998a) monitored 126 home in the Los Angeles metropolitan area during
18 February and December. Uniformly low ambient O3 concentrations were present during the
19 non-O3 seasons. Indoor O3 concentrations were always below outdoor O3 concentrations. The
20 mean indoor and outdoor O3 concentrations over the sampling period were 13 ± 12 ppb and 37 ±
21 19 ppb, respectively. There was a correlation between indoor O3 concentration and ambient
22 temperatures. The effect was magnified when the windows were open. When a central
23 refrigerant recirculating air conditioner was used, indoor O3 concentrations declined. The
24 authors were able to predict indoor O3 levels with nearly the same accuracy using measurements
25 made at regional stations coupled with window conditions as with measurements made directly
26 outside the homes coupled with window conditions, suggesting that monitoring station data may
27 be useful in helping to reduce errors associated with exposure misclassification. The authors
28 cautioned that varying results may occur at different locations with different housing stock or at
29 different times of the year.
30 Lee et al. (2002) reported indoor and outdoor O3 concentrations in 119 homes of school
31 children in two communities in southern California: Upland and San Bernadino county.
January 2005 AX3-186 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-28. 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
1
1
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
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)
January 2005
AX3-187 DRAFT-DO NOT QUOTE OR CITE
-------
1 Measurements were made over one year. Outdoor and indoor O3 concentrations were based on
2 cautioned that varying results may occur at different locations with different housing stock or at
3 different times of the year.
4 Lee et al. (2002) reported indoor and outdoor O3 concentrations in 119 homes of school
5 children in two communities in southern California: Upland and San Bernadino county.
6 Measurements were made over one year. Outdoor and indoor O3 concentrations were based on
7 monthly and weekly averages, respectively. Housing characteristics were not found to affect
8 indoor O3 concentrations. Indoor O3 concentrations were significantly lower than outdoor O3
9 concentrations. Average indoor and outdoor O3 concentrations were 14.9 and 56.5 ppb. Homes
10 with air conditioning had lower O3 concentrations, suggesting decreased ventilation or greater
11 O3 removal.
12 Chao (2001) evaluated the relationship between indoor and outdoor levels of various air
13 pollutants, including O3, in 10 apartments in Hong Kong during May to June. Air monitoring
14 was conducted over a 48-h period. All participants had the habit of closing the windows during
15 the evenings and using the air conditioner during the sleep hours. Windows were partially open
16 during the morning. The air conditioners were off during the working hours. Indoor O3
17 concentrations were low in all of the monitored apartments, ranging from 0 to 4.9 ppb with an
18 average of 2.65 ppb. Outdoor O3 concentrations ranged from 1.96 to 15.68 ppb. Table AX3-29
19 provides information on the indoor and outdoor O3 concentrations and the apartment
20 characteristics.
21 Drakou et al. (1995) demonstrated the complexity of the indoor environment.
22 Measurements of several pollutants, including O3, were made inside and outside of two
23 nonresidential buildings in Thessaloniki and Athens, Greece. The building in Thessaloniki was a
24 58-m3 metal structure. The ceiling and walls were covered with colored corrugated plastic
25 sheeting, and the flooring was plastic tile. There was no heating/air conditioning system and the
26 building was closed during the monitoring period. The AER ranged from 0.3 to 0.35 h"1. The
27 building in Athens was a 51-m3 concrete structure. The air conditioning system (recirculated air)
28 worked continuously during the monitoring period. A window was opened slightly to
29 accommodate the monitors'sampling lines. The AER was approximately 1 h"1. Monitoring
30 lasted for 30 days at both locations, however, only data from the 7 most representative days were
31 reported. Indoor O3 concentrations closely followed the outdoor concentrations at the
January 2005 AX3-188 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-29. Indoor and Outdoor O3 Concentrations in Hong Kong
Apartment
No.
1
2
o
6
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).
1
2
3
4
5
9
10
11
12
13
14
Thessaloniki building. The averaged 7-day indoor and outdoor O3 concentrations were 9.39 and
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
January 2005
AX3-189 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
concentrations. Indoor O3 concentrations frequently exceed 50 ppb during the summer months,
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 IT1 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.
1.5
,_ 0.5
July, 1992
16
250
200
£150
8
o
50
17
18
19
20
21
22
Outdoor 03
Indoor O,
16
17
18
19
20
21
22
Figure AX3-89. Air exchange rates and outdoor and indoor O3 concentrations during the
summer at telephone switching station in Burbank, CA.
Source: Weschleret al. (1994).
January 2005
AX3-190
DRAFT-DO NOT QUOTE OR CITE
-------
250
200
• 150
Outdoor 03
Indoor O3
S
o
50
16
17
18
19
20
21
22
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).
1
2
3
4
5
6
7
8
9
10
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 Matejko
Museum during the summer. The findings are included in Table AX3-30 for those locations
with reported AERs.
January 2005
AX3-191
DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-30. 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
aOn Loggia of Piano Nobile Level, high above the street.
bAt street level.
Adapted from Salmon et al. (2000).
1 Riediker et al. (2003) measured mobile source pollutants inside highway patrol vehicles in
2 Wake County, NC. Measurements were made during the 3 p.m. to midnight shift between
3 August 13 and October 11, 2001 in two patrol cars each day for a total of 50 shifts. All areas of
4 rural and urban Wake County were patrolled. The prominent areas patrolled were major
5 highways and interstates. Ozone concentrations inside the cars were compared with the O3
6 measurements at the fixed station in northern Raleigh. The average O3 concentration inside the
7 cars was 11.7 ppb, approximately one-third of the ambient air concentration. Jakobi and Fabian
8 (1997) found that O3 concentrations in a moving car were independent of the type of ventilation
9 (windows closed and ventilator operation/ventilator off and windows open). Ozone
10 concentrations inside the car were found to closely follow the outdoor concentrations. When the
11 car was parked, O3 concentrations outdoors greatly exceeded concentrations inside the car (see
12 Figure AX3-91).
13
January 2005 AX3-192 DRAFT-DO NOT QUOTE OR CITE
-------
100 -,
90 -
80 -
70 -
5 60-
3 50 -
30 -
20 -
10 -
Country Road Country Road
City
Parking
Autobahn
\ J
11:30 11:45 12:00 12:15
Time
12:30 12
August 23,
p100
-90
-80
-70
-60 P
-50 7?
-40 §
-30
-20
-10
0
:45
1995
Figure AX3-91. Indoor and outdoor O3 concentration in moving cars.
Source: Jaboki and Fabian (1997).
1 Few studies have been conducted in indoor environments containing O3 sources. Black
2 et al. (2000) measured O3 concentrations in a photocopy room at the University of California
3 during one business day. The room volume was 40 m3. Ozone concentrations were generally
4 below 20 ppb, but increased proportionately with increasing photocopier use. Ozone
5 concentrations reached 40 ppb when the hourly average number of copies reached 45. Helaleh
6 et al. (2002) reported daily average O3 concentrations from 0.8 to 1.3 ppb and 0.9 to 1.0 ppb in a
1 laboratory and photocopy room at a university in Japan. Outdoor O3 concentrations ranged from
8 6 to 11 ppb. Because only limited information was available on the sampling system used in the
9 study, the adequacy of the sampling system cannot be determined. Jakobi and Fabian (1997)
10 measured O3 concentrations in an office associated with the use of a photocopier and a laser
11 printer. They noted a 3.0 ppb increase in O3 from the use of a 3-year old printer run for 20 min
12 at a copy rate of 8 pages/mm. There was no detectable change in O3 concentrations from the use
13 of the laser printer.
14 The U.S. Environmental Protection Agency (Steiber, 1995) measured O3 concentrations
15 from the use of three home/office O3 generators. The O3 generators were placed in a 27 m3room
16 with doors and windows closed and the heating, ventilating and air conditioning system off; the
January 2005
AX3-193
DRAFT-DO NOT QUOTE OR CITE
-------
1 AER was 0.3 h"1. The units were operated for 90 min. Ozone concentrations at the low output
2 setting ranged from nondetectable to 14 ppb (averaged output). At the high output setting,
3 averaged output O3 concentrations exceeded 200 ppb in several cases and had spike
4 concentrations as high as 480 ppb.
5
6 Mass Balance Equations
1 Indoor concentrations of O3 can be estimated using the mass balance model. The mass
8 balance model is a general ventilation model that relates indoor pollutant concentrations to those
9 outside. The mass balance model estimates the concentration of a pollutant over time. The
10 simplest form of the model is represented by the following differential equation for a perfectly
11 mixed microenvironment without an air cleaner:
12
d(., TAT ^ S ,-,
—JL = vCOUT + —- vCjN (AX3-2)
13
14 where dCIN is the indoor pollutant concentration (mass/volume), t is time in hours, v is the air
15 exchange rate, COUT is the outdoor pollutant concentration (mass/volume), Fis the volume of the
16 microenvironment, and S is the indoor source emission rate.
17 Nazaroff and Cass (1986) extended the mass balance model to include multiple
18 compartments and interactions between different compounds. The extended model takes into
19 account the effects of ventilation, filtration, heterogeneous removal, direct emission, and
20 photolytic and thermal and chemical reactions. A more in-depth discussion of the mass balance
21 model may be found in Shair and Heitner (1974) and in Nazaroff and Cass (1986).
22 Freijer and Bloemen (2000) used the one-compartment mass balance model to examine the
23 relationship between O3 I/O ratios as influenced by time patterns in outdoor concentrations,
24 ventilation rate, and indoor emissions. The microenvironment was 250 m3. Three different
25 ventilation patterns with the same long-term average AER (0.64 h"1) were used. A source
26 pattern (direct emissions) of zero was used, because O3 sources are not common. The time series
27 for outdoor O3 concentrations consisted of 100-day periods during the summer, with hourly
28 measured concentrations. The following assumptions were made: (1) O3 concentration in the air
29 that enters the microenvironment is equal to the concentration of the outside air minus the
January 2005 AX3-194 DRAFT-DO NOT QUOTE OR CITE
-------
1 fraction removed by filtration, (2) the O3 concentration that leaves the microenvironment equals
2 the O3 concentration in the microenvironment, (3) the decay processes in the microenvironment
3 are proportional to the mass of the pollutant, and (4) addition or removal of O3 in the
4 microenvironment also may occur through independent sources and sinks.
5 Figure AX3-92 represents the measured outdoor O3 concentrations and modeled indoor O3
6 concentrations. The indoor modeled O3 concentrations were found to equal approximately 33%
7 of the outdoor monitored concentrations.
63
51 -
37.6 H
0)
12.5 -
100
101
102 103
Time (day)
104
105
Figure AX3-92. Measured outdoor O3 concentrations (thin line) and modeled indoor
concentrations (bold line).
Source: Adapted from Freijer and Bloemen (2000).
1 AX3.11 HUMAN EXPOSURE TO OZONE AND RELATED
2 PHOTOCHEMICAL OXIDANTS
3 AX3.11.1 Introduction
4 There are many definitions of exposure. Human exposure to O3 and related photochemical
5 oxidants are based on the measured O3 concentrations in the individual's breathing zone as the
6 individual moves through time and space. Epidemiological studies generally use the ambient
January 2005
AX3-195
DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations as surrogates for exposure. Therefore, human exposure data and models provide
2 the best link between ambient concentrations (from measurements at monitoring sites or
3 estimated with atmospheric transport models), lung deposition and clearance and estimates of air
4 concentration-exposure-dose relationships.
5 This section discusses the current information on the available human exposure data and
6 exposure model development. This includes information on (a) the relationships between O3
7 measured at ambient monitoring sites and personal exposures and (b) factors that affect these
8 relationships. The information presented in this section is intended to provide critical links
9 between ambient monitoring data and O3 dosimetry as well as between the toxicological and
10 epidemiologic studies presented in Annexes AX4, AX5, AX6, and AX7 of this document.
11
12 AX3.11.2 Summary of the Information Presented in the Exposure
13 Discussion in the 1996 Ozone Criteria Document
14 The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a), based on then
15 currently available information, indicated that less emphasis should be placed on O3
16 concentrations measured at ambient monitoring stations. Fixed monitoring stations are generally
17 used for monitoring associated with air quality standards and do not provide a realistic
18 representation of individual exposures. Indoor/outdoor O3 ratios reported in the literature were
19 summarized for residences, hospitals, offices, art galleries, and museums. The differences in
20 residential I/O were found to be a function of ventilation conditions. The I/O ratios were less
21 than unity. In most cases, indoor and in-transit concentrations of O3 were significantly different
22 from ambient O3 concentrations. Ambient O3 varied from O3 concentrations measured at fixed-
23 site monitors. Very limited personal exposure measurements were available at the time the 1996
24 O3 AQCD was published, so estimates of O3 exposure or evaluated models were not provided.
25 The two available personal exposure studies indicated that only 40% of the variability in
26 personal exposures was explained by the exposure models using time-weighted indoor and
27 outdoor concentrations. The discussion addressing O3 exposure modeling primarily addressed
28 work reported by McCurdy (1994) on population-based models (PBMs). Literature published
29 since publication of the 1996 O3 AQCD has also focused on PBMs. A discussion of individual -
30 based models (IBMs) will be included in the description of exposure modeling in this document
January 2005 AX3-196 DRAFT-DO NOT QUOTE OR CITE
-------
1 to improve our mechanistic understanding of O3 source-to-exposure events and to evaluate their
2 usefulness in providing population-based estimates.
3
4 AX3.11.3 Concepts of Human Exposure
5 Human exposure to O3 and related photochemical oxidants occurs when individuals come
6 in contact with the pollutant through "(a) the visible exterior of the person (skin and openings
7 into the body such as mouth and nostrils) or (b) the so-called exchange boundaries where
8 absorption takes place (skin, mouth, nostrils, lung, gastrointestinal tract)" (Federal Register,
9 1986). Consequently, exposure to a chemical, in this case O3, is the contact of that chemical
10 with the exchange boundary (U.S. Environmental Protection Agency, 1992). Therefore,
1 1 inhalation exposure to O3 is based on measurements of the O3 concentration near the individual's
12 breathing zone that is not affected by exhaled air.
13
14 AX3.11. 4 Quantification of Exposure
15 Quantification of inhalation exposure to any air pollutant starts with the concept of the
16 variation in the concentration of the air pollutant in the breathing zone, unperturbed by exhaled
17 breath, as measured by a personal exposure monitor as a person moves through time and space.
18 Since the concentrations of O3 and related photochemical oxidants vary with time and
19 location and since people move among locations and activities, the exposure and dose received
20 changes during the day. Furthermore, the amount of pollutant delivered to the lung is dependent
21 upon the person's minute ventilation rate. Thus, the level of exertion is an important
22 consideration in determining the potential exposure and dose. Inhalation exposure has been
23 defined as the integral of the concentration as a function of time over the time period of interest
24 for each individual (Ott, 1982, 1985; Lioy, 1990):
25
26 E = c(t)dt (AX3-3)
h
27 where E is inhalation exposure, eft) is the breathing zone concentration as a function of time and
28 tj and t2 the starting and ending time of the exposure, respectively.
January 2005 AX3-197 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.11.5 Methods To Estimate Personal Exposure
2 There are two approaches for measuring personal exposure; direct and indirect methods
3 (Ott, 1982, 1985). Direct approaches measure the contact of the person with the chemical
4 concentration in the exposure media over an identified period of time. For the direct
5 measurement method, a personal exposure monitor (PEM) is worn near the breathing zone for a
6 specified time to either continually collect for subsequent analysis or directly measure the
7 concentrations of the pollutant and the exposure levels.
8 The indirect method determines and measures the concentrations in all of the locations or
9 "microenvironments" a person encounters or determines the exposure levels through the use of
10 models or biomarkers. The concept of microenvironments is critical for understanding human
11 exposure and aids in the development of procedures for exposure modeling using data from
12 stationary monitors (indoor and outdoor). Microenvironments were initially defined as
13 individual or aggregate locations (and sometimes even as activities taking place within a
14 location) where a homogeneous concentration of the pollutant is encountered for a specified
15 period of time. Thus, a microenvironment has often been identified with an "ideal" (i.e.,
16 perfectly mixed) compartment of classical compartmental modeling. More recent and general
17 definitions view the microenvironment as a "control volume," indoors or outdoors, that can be
18 fully characterized by a set of either mechanistic or phenomenological governing equations,
19 when properly parameterized, given appropriate initial and boundary conditions. The boundary
20 conditions typically reflect interactions with the ambient air and with other microenvironments.
21 The parameterizations of the governing equations generally include the information on attributes
22 of sources and sinks within each microenvironment. This type of general definition allows for
23 the concentration within a microenvironment to be nonhomogeneous, provided its spatial profile
24 and mixing properties can be fully predicted or characterized. By adopting this definition, the
25 number of microenvironments used in a study is kept manageable, while existing variabilities in
26 concentrations are still taken into account. The "control volume" variation could result in a
27 series of microenvironments in the same location. If there are large spatial gradients within a
28 location for the same time period, the space should be divided into the number of
29 microenvironments needed to yield constant pollutant concentrations; the alternative offered by
30 the control volume approach is to provide concentration as a function of location within it,
31 so that the appropriate value is selected for calculating exposure. Thus, exposure to a person in a
January 2005 AX3-198 DRAFT-DO NOT QUOTE OR CITE
-------
1 microenvironment is calculated using a formula analogous to equation AX3-2, but as the sum of
2 the discrete products of measured or modeled concentrations (specific to the receptor and/or
3 activity of concern) in each microenvironment by the time spent there. The equation is
4 expressed as:
(AX3-4)
7 where /' specifies microenvironments from 1 to «, ci is the concentration in the rth
8 microenvironment, and A ti is the duration spent in the /th microenvironment. The total exposure
9 for any time interval for an individual is the sum of the exposures in all microenvironments
10 encountered within that time interval. This method should provide an accurate determination of
11 exposure provided that all microenvironments that contribute significantly to the total exposures
12 are included and the concentration assigned to the microenvironment is appropriate for the time
13 period spent there.
14 Microenvironments typically used to determine O3 exposures include indoor residences,
15 other indoor locations, outdoors near roadways, other outdoor locations, and in-vehicles.
16 Outdoor locations near roadways are segregated from other outdoor locations because N2O
17 emissions from automobiles alter O3 and related photochemical oxidant concentrations compared
18 to concurrent typical ambient levels. Indoor residences are typically separated from other indoor
19 locations, because of the time spent there and potential differences between the residential
20 environment and the work/public environment. A special concern for O3 and related
21 photochemical oxidants is their diurnal weekly (weekday-weekend) and seasonal variability.
22 Few indoor O3 sources exist, but include electronic equipment, O3 generators, and copying
23 machines. Some secondary reactions of O3 take place indoors that produce related
24 photochemical oxidants that could extend the exposures to those species above the estimates
25 obtained from O3 alone. (See discussion on O3 atmospheric chemistry and indoor sources and
26 concentrations earlier in this Annex.)
27 Measurement efforts to assess population exposures or exposures to large numbers of
28 individuals over long time periods is labor intensive and costly, so exposure modeling is often
29 done for large populations evaluated over time. Predicting (or reconstructing) human exposure
January 2005 AX3-199 DRAFT-DO NOT QUOTE OR CITE
-------
1 to O3 through mechanistic models is complicated by the fact that O3 (and associated
2 photochemical oxidants) is formed in the atmosphere through a series of chemical reactions that
3 are nonlinear and have a wide range of characteristic reaction timescales. Furthermore, these
4 reactions require the precursors VOCs and NOX that are emitted by a wide variety of both
5 anthropogenic and natural (biogenic) emission sources. This makes O3 a secondary pollutant
6 with complex nonlinear and multiscale dynamics in time and space. Concentration levels
7 experienced by individuals and populations exposed to O3 are therefore affected by (1) emission
8 levels and spatiotemporal patterns of the gaseous precursors: VOCs and NOX, that can be due to
9 sources as diverse as a power plant in a different state, automobiles on a highway five miles
10 away, and the gas stove in one's own kitchen; (2) ambient atmospheric as well as indoor
11 microenvironmental transport, removal and mixing processes (convective, advective, dispersive
12 and molecular/diffusional); and (3) chemical transformations that take place over a multitude of
13 spatial scales, ranging from regional/sub-continental (100 to 1000 km), to urban (10 to 100 km),
14 to local (1 to 10 km), to neighborhood (< 1 km), and to microenvironmental/personal. These
15 transformations depend on the presence of co-occurring pollutants in gas and aerosol phases,
16 both primary and secondary, and on the nature of surfaces interacting with the pollutants.
17 Further, the strong temporal variability of O3, both diurnal and seasonal, makes it critical
18 that definitions of integrated or time-averaged exposure employ appropriate averaging times in
19 order to produce scientifically defensible analyses for either causes of O3 production or health
20 effects that result from O3 exposure. An understanding of the effect of temporal profiles of
21 concentrations and contacts with human receptors is essential. Short-term integrated metrics,
22 such as hourly averages, 8-h running averages, etc., are needed to understand the relationship
23 between O3 exposure and observed health and other effects.
24 Health effects associated with O3 have mostly been considered effects of acute exposures.
25 Peak O3 and related photochemical oxidants concentrations typically occur towards the latter
26 portion of the day during the summer months. Elevated concentrations can last for several
27 hours. Regional O3 episodes often co-occur with high concentrations of airborne fine particles
28 making it difficult to assess O3 dynamics and exposure patterns. Furthermore, O3 participates in
29 multiphase (gas/aerosol) chemical reactions in various microenvironments. Several recent
30 studies show that O3 reacts indoors with VOCs and NOX in an analogous fashion to that
31 occurring in the ambient atmosphere (Lee and Hogsett, 1999; Wainman et al., 2000; Weschler
January 2005 AX3-200 DRAFT-DO NOT QUOTE OR CITE
-------
1 and Shields, 1997). These reactions produce secondary oxidants and other air toxics that could
2 play a significant role in cumulative human exposure and health-related effects within the
3 microenvironment.
4
5 AX3.11.6 Exposures in Microenvironments
6 Ozone has been measured more extensively than the other photochemical oxidants.
7 Ambient monitors have been established in most areas of the country, with extensive monitoring
8 in regions that have been in noncompliance with the previous 1-h daily NAAQS. Monitoring
9 station-measured hourly concentrations also have been used as surrogates of exposure in
10 epidemiological studies and in evaluating exposure-related health effects. According to the
11 Guideline on Ozone Monitoring Site Selection (U.S. Environmental Protection Agency, 1998),
12 when designing an O3 monitoring network, consideration should be given to (1) proximity to
13 combustion emission sources, (2) distance from primary emission sources, and (3) the general
14 wind direction and speed to determine the primary transport pathways of O3 and its precursors.
15 Finally, the 1-h daily maximum and 8-h average O3 concentrations can have different spatial
16 patterns with elevated daily 8-h O3 concentrations typically being over a wider spatial area.
17 Therefore, O3 monitoring networks should determine the highest concentrations expected to
18 occur in the area, representative concentrations for high population density areas, the impact of
19 sources or source categories on air pollution levels, and general background concentration levels
20 (U.S. Environmental Protection Agency, 1998).
21 The guideline also states that the monitor's O3 inlet probe should be placed at a height and
22 location that best approximates where people are usually found. However, complicating factors
23 (e.g., security considerations, availability) sometimes result in the probe placement being
24 elevated 3 to 15 m above ground level, a different location than the breathing zone (1 to 2 m) of
25 the populace. Although there are some commonalities in the considerations for the sampling
26 design for monitoring and for determining population exposures, differences also exist. These
27 differences between the location and height of the monitor compared to the locations and
28 breathing zone heights of people can result in different O3 concentrations between what is
29 measured at the monitor and exposure and, therefore, should be considered when using ambient
30 air monitoring data as a surrogate for exposure in epidemiological studies and risk assessments.
31
January 2005 AX3-201 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.11.7 Differences in Concentrations Among Microenvironments
2 Indoors
3 People spend the majority of their time indoors. In exposure analyses, I/O ratios may
4 provide a reasonable estimate of indoor O3 concentrations when outdoor O3 concentrations are
5 known. The reported I/O ratios vary greatly (see earlier discussion in this annex and U.S.
6 Environmental Protection Agency, 1996a). A number of studies published since the completion
7 of the 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) have examined the
8 relationship between indoor and outdoor O3 concentrations in several countries and varying
9 building types and ventilation conditions. A description of the more recent studies appears in the
10 discussion on indoor sources and concentrations earlier in this Annex.
11
12 Other Microenvironments
13 In addition to buildings and their associated outdoor microenvironments, two other
14 microenvironments have been studied: within a vehicle and outdoors near a roadway. Johnson
15 (1997) conducted a scripted study using four trained technicians to measure hourly average O3
16 concentrations between 07:00 and 19:00 h in Los Angeles, CA during September and October
17 1995. The ratio of the microenvironmental concentrations to the fixed site monitor on days
18 when the O3 levels > 20 ppb were as follows: indoor residence, 0.28; other locations indoors,
19 0.18; outdoor near roadways, 0.58; other locations outdoors, 0.59; and in-vehicle, 0.21. The
20 concentrations indoors and within vehicles varied depending on whether the windows were
21 opened (higher) or closed (lower) and the use of air conditioning. The lower outdoor
22 concentrations, particularly near roadways, probably reflect the reaction of O3 with NO emitted
23 by automobiles.
24 A study of the effect of elevation on O3 concentrations found that concentrations increased
25 with increasing elevation. The ratio of O3 concentrations at street level (3 m) compared to the
26 rooftop (25 m) was between 0.12 and 0.16, though the actual concentrations were highly
27 correlated (r = 0.63) (Vakeva et al., 1999). Differential O3 exposures may, therefore, exist in
28 apartments that are on different floors. Differences in elevation between the monitoring sites in
29 Los Angeles and street level samples may have contributed to the lower levels measured by
30 Johnson (1997). Furthermore, since O3 monitors are frequently located on rooftops in urban
31 settings, the concentrations measured there may overestimate the exposure to individuals
January 2005 AX3-202 DRAFT-DO NOT QUOTE OR CITE
-------
1 outdoors in streets and parks, locations where people exercise and maximum O3 exposure is
2 likely to occur.
3 Chang et al. (2000) conducted a scripted exposure study in Baltimore during the summer of
4 1998 and winter of 1999, during which 1-h O3 samples were collected by a technician who also
5 changed his or her activity every hour. The activities chosen were selected to simulate older
6 (> 65 years) adults, based on that reported in the National Human Activity Pattern Survey
7 (NHAPS). Personal O3 levels were significantly lower than the outdoor values, because more
8 time was spent indoors. Mean summer concentrations were 15.0 ± 18.3 ppb, with a maximum of
9 76.3 ppb. The personal O3 exposure levels within the outdoor and in-vehicle microenvironments
10 were significantly correlated with ambient concentration, although the ratio of personal exposure
11 to ambient levels was less than one, with only the top 5% of the ratios exceeding one, indicating
12 that the ambient measurements lead to the maximum concentrations and exposures. The indoor
13 concentrations did not correlate with outdoor measurements.
14 The scripted exposure studies show that the O3 concentrations in the various
15 microenvironments were typically lower than the ambient air concentrations measured at
16 monitoring stations. Exposure models are useful for accounting for the reduced concentrations
17 usually encountered in various microenvironments compared to ambient monitoring station
18 concentrations (see discussion on exposure models later in this annex).
19
20 AX3.11.8 Activity Patterns/Time Spent in Microenvironments
21 The O3 dose delivered to the lung is not only a function of the air concentration but also
22 depends on the minute ventilation rate and the morphology of the respiratory tract. Estimates of
23 the exercise level, which is related to minute ventilation rate and time spent outdoors, stratified
24 by age, season, and gender have been compiled based on questionnaire data. Klepeis et al.
25 (1996, 2001) reported in National Human Activity Pattern Survey (NHAPS) a probability-based
26 telephone survey done in the early 1990s, that the highest levels of outdoor activity for
27 yardwork/maintenance occur in the spring during morning to early afternoon and for
28 sports/exercise in the summer (3 to 6% of respondents) during the middle of the day from noon
29 to 3 p.m. The time period for sports/exercise in the spring (5% of respondents) was from 3:30 to
30 6 p.m. Outdoor activity time also vary between weekend/weekday with weekend activity being
31 throughout the day (9 to 5 p.m.) and weekday outdoor activity shifts to latter in the day, with an
January 2005 AX3-203 DRAFT-DO NOT QUOTE OR CITE
-------
1 initial rise at 3 p.m. that extends into the evening hours. These differences reflect the hours of
2 more leisure activities. Since the highest outdoor O3 concentrations occurs in the middle to the
3 latter part of the day, which coincides with the time period of most physical activity, it is
4 expected that individuals engaging in outdoor activities during this period will receive the
5 maximum O3 dose.
6 Klepeis et al. (1996, 2001) also reported differences in the amount of time that different
7 age groups spend in various activities. School-age children were found to spend more time at
8 sports/exercise than other age groups. Avol et al. (1998b) reported that the number of hours
9 three groups of children (healthy, asthmatic, and wheezy randomly selected from a larger
10 longitudinal health study) spent outdoors during the spring and summer on either high- or low-
11 O3 days were similar. However, the amount of time asthmatic children spent being physically
12 active outdoors was smaller than the healthy or wheezy children, particularly during the summer
13 on high-O3 days. A small difference in the proportion of time spent being active during the
14 summer was observed between wheezy children (lower) and healthy children. Thus, children
15 with respiratory problems modified their behavior and received lower O3 doses. Additionally,
16 girls spent less time outdoors and were less physically active.
17 Adams (1993) determined population breathing averages by measuring heart rate,
18 breathing frequency, and volume of air breathed in by children, adolescents, young/middle age
19 adults, and seniors during a variety of activities. Adults of the same gender from adolescent
20 through senior years breathed similar amounts of air during similar activities. Children were
21 found to breathe more air than adults, relative to body surface area, breathing frequency, and
22 heart rate. The time spent outdoors under medium- and high-activity levels has been reported to
23 be 2% and 0.2% of the waking hours, respectively (Shamoo et al., 1994), which represents the
24 time frame with the highest minute ventilation rate and, therefore, highest exposure to O3.
25 Alteration in the minute ventilation rate have been observed in asthmatic children when exposed
26 to air pollutants (Linn et al., 1996). The children tested were fourth graders in public schools in
27 three communities in southern California. Kiinzli et al. (1997) tested the reproducibility of the
28 Activity Tables, a retrospective time-activity assessment questionnaire in 75 college freshmen.
29 Students, ages 17 to 21 years, were asked to recall activities in which they had engaged. Four
30 levels of activity were designated, based on the responses to the questions, to determine
31 participation in specific activities along with estimates of the duration and frequency of
January 2005 AX3-204 DRAFT-DO NOT QUOTE OR CITE
-------
1 participation. Internal consistency in the responses for activity level was found between and
2 within two sets of the questionnaires administered at different times to the same individuals,
3 although it appears that the reported number of hours devoted to physical activity was
4 overestimated. Overall, having an estimate of the number of hours outdoors and how many of
5 those hours were spent engaged in physical activity will improve the estimates for the high-end
6 exposures.
7
8 AX3.11.9 Personal Exposure Monitors and Measurements
9 Modified passive samplers have been developed for use in determining O3 exposure. The
10 difficulty in developing a passive O3 monitor is in identifying a chemical or trapping reagent that
11 can react with O3. Zhou and Smith (1997) evaluated the effectiveness of sodium nitrite,
12 3-methyl-2-benzothiazolinone acetone azine (MBTH),/»-acetamidophenol (p-ATP), and indigo
13 carmine as O3-trapping reagents. Only sodium nitrite and MBTH gave sensitive, linear
14 responses at environmentally relevant concentrations. However, MBTH overestimated the O3
15 concentrations significantly, suggesting an interference effect. Sodium nitrite was found to be a
16 valid reagent when an effective diffusion barrier was used. Scheeren and Adema (1996) used an
17 indigo carmine-coated glass-fiber filter to collect spectrophotometrically measured O3. The
18 detection limit was 23 ppb for a 1-h exposure, with no interfering oxidants identified. The
19 reagent was valid for a relative humidity range of 20 to 80%. The uptake rate was wind velocity
20 dependent. However, wind velocity dependencies was compensated for by using a small
21 battery-operated fan that continuously blew air across the face of the monitor at a speed of
22 1.3 m/s. The overall accuracy of the sampler, after correcting for samples collected under low-
23 wind conditions, was 11 ± 9% in comparison to a continuous UV-photometric monitor. Sample
24 stability was > 25 days in a freezer. Bernard et al. (1999) employed a passive sampler consisting
25 of a glass-fiber filter coated with a l,2-di(4-pyridyl)ethylene solution. The sample was analyzed
26 spectrophotometrically after color development by the addition of 3-methyl-2-benzothiazolinone
27 hydrazone hydrochloride. The sampler was used at 48 sites in Montpellier, France. The results
28 from the passive sampler were highly correlated (0.9, p < 0.0001) with the results from the UV
29 absorption analyzer of the regional air quality network. Detection limits were 17 ppb for 12-h
30 and 8 ppb for 24-h samples with an overall variation coefficient of 5% for field-tested paired
31 samples.
January 2005 AX3-205 DRAFT-DO NOT QUOTE OR CITE
-------
1 A series of studies have been conducted using a passive sampler developed by Koutrakis
2 et al. (1993) at the Harvard School of Public Health. The sampler used sodium nitrate as the
3 trapping reagent and included a small fan to assure sufficient movement of air across the face of
4 the badge when sampling was done indoors. The passive sampler has been evaluated against the
5 standard UV absorption technique used in studies in southern California (Avol et al., 1998a;
6 Geyh et al., 1999, 2000; Delfmo et al., 1996), Baltimore, MD (Sarnat et al., 2000), and Canada
7 (Brauer and Brook, 1997).
8 Avol et al. (1998a) used nitrite-coated passive samplers to measure O3 air concentrations
9 indoors and outdoors of 126 homes between February and December 1994 in the Los Angeles
10 metropolitan area. The detection limit of the method was near 5 ppb. Geyh et al. (1997, 1999)
11 compared passive and active personal O3 air samplers based on nitrite-coated glass-fiber filters.
12 The active sampler was more sensitive allowing for the collection of short-term, 2.6-h samples.
13 Comparison between the two samplers and UV photometric O3 monitors demonstrated generally
14 good agreement (bias for active personal sampler of -6%). The personal sampler also had high
15 precision (4% for duplicate analyses) and good compliance when used by children attending
16 summer day camp in Riverside, CA.
17 Passive O3 monitors have been used in several field studies to determine average daily O3
18 exposure as well as in scripted studies to evaluate O3 exposures over one to several-hour time
19 periods. Delfmo et al. (1996) measured 12-h personal daytime O3 exposures in asthmatic
20 patients in San Diego from September through October 1993. They found that the mean
21 personal exposures were 27% of the mean outdoor concentrations. Individual exposure levels
22 among the 12 asthmatic subjects aged 9 to 16 years varied greatly. Mean personal O3 exposure
23 levels were lower on Friday, Saturday, and Sunday than on other days of the week (10 versus
24 13 ppb), while the ambient air concentrations were higher Friday through Sunday. The authors
25 suggested that the differences were due to higher weekday NO emissions from local traffic
26 which titrated the ambient O3 levels. The lower personal exposure levels on Friday, Saturday,
27 and Sunday may have been an artifact of greater noncompliance, with the badges remaining
28 indoors and, therefore, being exposed to lower O3 concentrations. The overall correlation
29 between the personal exposure concentrations between any two individuals and with the outdoor
30 stationary site was only moderate (r = 0.45; range: 0.36 to 0.69). The O3 concentrations at the
January 2005 AX3-206 DRAFT-DO NOT QUOTE OR CITE
-------
1 stationary site exceeded the personal levels by an average of 31 ppb. Avol et al. (1998b)
2 observed a poor correlation between personal exposure and fixed-site monitoring concentrations
3 (r = 0.28, n = 1336 pairs) for a cohort of children (healthy, wheezy, and asthmatic). Sarnat et al.
4 (2000) measured personal O3 exposures during a 12-day longitudinal study of 20 older adults
5 (> 64 years) in Baltimore, MD. Ten subjects participated in the summer and winter exposures
6 and the remaining 10 participated in either the summer or winter exposure. No statistically
7 significant overall correlations were identified between the personal and the ambient O3
8 concentrations during either the winter or summer. Only a single individual (n = 14 summer and
9 13 winter) had a significant correlation with outdoor concentrations. Geyh et al. (2000)
10 measured indoor and outdoor concentration and personal O3 exposures in 169 elementary school
11 children from 116 homes during a year-long sampling protocol in 2 communities in southern
12 California. Samples were collected for 1 week per month. Boys had higher O3 exposure than
13 girls, probably reflecting the greater amount of time boys spent outdoors compared to girls
14 (3.8 versus 3.2 h for the spring/summer and 2.9 versus 2.2 h for the fall/winter). The average
15 personal O3 exposures were lower than the levels measured at the nearest monitor stations
16 retrieved from the AIRS. During the O3 season, differences were found in indoor concentrations
17 and personal O3 exposures between the two communities participating in the study based on
18 ambient air concentrations and differences in O3 air exchange rates in the homes.
19 Brauer and Brook (1997) conducted personal exposure evaluations in three groups in
20 Frazer Valley, Vancouver, Canada. The groups were divided by the amount of time spent
21 outdoors: (1) the majority of the workday was spent indoors or commuting, (2) an intermediate
22 amount of time was spent outdoors, and (3) the entire exposure monitoring period was outdoors.
23 Time-activity data were collected for the first two groups to assess the proportion of time spent
24 outdoors. For groups 1 and 2, the lowest quartile of participants based on the fraction of time
25 spent outdoors had significantly lower O3 exposure (mean personal exposure to outdoor
26 concentration ratio = 0.18 and 0.35, respectively) compared to those in the upper quartile (mean
27 ratio = 0.51 and 0.58, respectively; p < 0.05; Bonferroni multiple range test). The mean ratio
28 was 0.96 with values ranging from near 0 to 2 for the group that spent the entire exposure-
29 monitoring period outdoors. The authors attributed the extreme low ratios to random
30 measurement error at low O3 air concentrations (estimated at 35%), local variability in O3
January 2005 AX3-207 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations, and to differences between ground-level concentrations (where the personal
2 samples were collected) 3-m above ground level (where the continuous monitors were located).
3 The highest ratios may be due to either locale variability in O3 concentrations or to an
4 interference affecting the personal O3 samplers, particularly at the lower concentration range,
5 leading to a small positive error. Temporal plots of O3 for the mean daily personal exposures
6 and ambient concentrations showed the same general trend with general agreement between the
7 personal exposures and ambient air concentrations for group 3. However, for groups 1 and 2, the
8 day-to-day variability of the personal exposures and ambient O3 concentrations had consistent
9 patterns, suggesting that the ambient air was the primary source for O3 exposure. The actual O3
10 concentrations measured in the personal air space were always considerably lower than the
11 ambient concentrations.
12
13 AX3.11.10 Trends in Concentrations Within Microenvironments
14 There have not been sufficient numbers of measurements of personal exposures or indoor
15 O3 concentrations to directly document trends over time or location. However, since O3
16 concentrations in all microenvironments are primarily derived from ambient sources, trends in
17 ambient air concentrations should reflect trends in personal exposure unless there are differences
18 in activity patterns over time or locations. Overall, a significant downward trend in ambient O3
19 concentrations has occurred from 1980 in most locations in the United States, although the trend
20 in the latter part of the 1990s suggests that continued improvements in air quality may have
21 leveled off. Greater declines in ambient O3 concentrations appear to have occurred in urban
22 centers than in rural regions. The decline in daily and weekly average O3 concentrations from
23 1989 to 1995 in rural regions was 5% and 7%, respectively (Wolff et al., 2001; Lin et al., 2001;
24 Holland et al., 1999). A detailed discussion of O3 trends appears earlier in this annex.
25
26 AX3.11.11 Modeling Ozone Exposure
27 AX3.11.11.1 Issues of Terminology
28 Models of human exposure to O3 can be characterized and differentiated based upon a
29 variety of attributes. For example, exposure models can be classified as (1) potential exposure
30 models, typically maximum outdoor concentration versus "actual" exposure, including locally
31 modified microenvironmental outdoor and indoor exposures; (2) population versus "specific
January 2005 AX3-208 DRAFT-DO NOT QUOTE OR CITE
-------
1 individual"-based exposure models; (3) deterministic versus probabilistic models; and
2 (4) observation versus mechanistic air quality model-driven estimates of spatially and temporally
3 varying O3 concentration fields, etc.
4 Some points should be made regarding terminology and the directions of exposure
5 modeling research (as related specifically to O3 exposure assessments) before proceeding to
6 discuss specific recent activities and developments. First, it must be understood that significant
7 variation exists in the definitions for much of the terminology used in the published literature.
8 The science of exposure modeling is an evolving field and the development of a "standard" and
9 commonly accepted terminology is a process in evolution. Second, very often procedures/efforts
10 listed in the scientific literature as "exposure models/exposure estimates," etc., may in fact refer
11 to only a subset of the steps or components required for a complete exposure assessment. For
12 example, some efforts focus solely on refining the subregional or local spatiotemporal dynamics
13 of local O3 concentrations starting from "raw" data representing monitor observations or regional
14 grid-based model estimates. Nevertheless, such efforts are included in the discussion of the next
15 subsection, as they can provide improved tools for the individual components that constitute a
16 complete exposure assessment. On the other hand, formulations that are identified as exposure
17 models, but focus only on ambient air quality predictions, are not included in the discussion that
18 follows, as they do not provide true exposure estimates but, rather, ambient air estimates. These
19 models are reviewed in an earlier section of this annex. It is recognized that ambient air
20 concentrations are used as surrogates for exposure in some epidemiological studies. Third,
21 O3-exposure modeling is very often identified explicitly with population-based modeling, while
22 models describing the specific mechanisms affecting the exposure of an individual to O3, and
23 possibly some of the co-occurring gas and/or aerosol phase pollutants, are usually associated
24 with studies focusing on indoor chemistry modeling. Finally, in recent years, the focus of either
25 individual- or population-based exposure modeling research has shifted from O3 to other
26 pollutants, mostly airborne toxics and particulate matter. However, many of the modeling
27 components that have been developed in these efforts are directly applicable to O3 exposure
28 modeling and are, therefore, mentioned in the following discussion.
29
January 2005 AX3-209 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.11.11.2 A General Framework for Assessing Exposure to Ozone
2 Once the individual and relevant activity locations for Individual Based Modeling (IBM),
3 or the population and associated spatial (geographical) domain for Population Based Modeling
4 (PBM) have been defined, along with the temporal framework of the analysis (period,
5 resolution), the comprehensive modeling of individual/population exposure to O3 (and related
6 pollutants) will generally require several steps (or components, as some of them do not have to
7 be performed in sequence). The steps represent a "composite" outline based on frameworks
8 described in the literature over the last 20 years (Ott, 1982, 1985; Lioy, 1990; Georgopoulos and
9 Lioy, 1994; U.S. Environmental Protection Agency, 1992, 1997) as well as on the structure of
10 various existing inhalation exposure models (NEM/pNEM, HAPEM, SHEDS, REHEX,
11 EDMAS, MENTOR-OPERAS, APEX, AIRPEX, AIRQUIS, etc.) (McCurdy, 1994; Johnson
12 et al., 1992; Nagda et al., 1987; U.S. Environmental Protection Agency, 1996c; ICF Consulting,
13 2003; Burke et al., 2001; McCurdy et al., 2000; Georgopoulos et al., 2002a,b; Freijer et al.,
14 1998; Clench-Aas et al., 1999; Kunzli et al., 1997). The conceptional frameworks of the models
15 are similar. Figures AX3-93a,b provides a conceptual overview of the APEX as an example.
16 The steps involved in defining exposure models include (1) estimation of the background or
17 ambient levels of O3 through geostatistical analysis of fixed monitor data, or emissions-based,
18 photochemical, air quality modeling; (2) estimation of levels and temporal profiles of O3 in
19 various outdoor and indoor microenvironments such as street canyons, residences, offices,
20 restaurants, vehicles, etc. through linear regression of available observational data sets, simple
21 mass balance models, detailed (nonlinear) gas or gas/aerosol chemistry models, or detailed
22 combined chemistry and computational fluid dynamics models; (3) characterization of relevant
23 attributes of individuals or populations under study (age, gender, weight, occupation, etc.);
24 (4) development of activity event (or exposure event) sequences for each member of the sample
25 population or for each cohort for the exposure period; (5) calculation of appropriate inhalation
26 (in general intake) rates for the individuals of concern, or the members of the sample population,
27 reflecting/combining the physiological attributes of the study subjects and the activities pursued
28 during the individual exposure events; (6) combination of intake rates and microenvironmental
29 concentrations for each activity event to assess dose; (7) calculation of event-specific exposure
30 and intake dose distributions for selected time periods (1-h and 8-h daily maximum, O3 season
31 averages, etc.); and (8) use of PBM to extrapolate population sample (or cohort) exposures and
January 2005 AX3-210 DRAFT-DO NOT QUOTE OR CITE
-------
to
o
o
X
to
Sector/Location data
(latitude/longitude)
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 / 1
, (age/gender/race) 1 ,
•>
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)
^^^
^^
/Commuting flow data/ \
(origin/destination
^ sectors) \J
---""Stochastic"'---^
---^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
fe
H
6
O
O
H
O
- National
database
-Simulation
step
-Area-specific
input data
- Data processor
- Intermediate step
or data
I - Output data
Figure AX3-93a. Detailed diagram illustrating components of APEX exposure model.
Source: www.epa.gov/ttn/fera/data/apex3227.
o
HH
H
W
-------
to
o
o
X
OJ
i
i
to
to
o
O
2
O
H
O
SteoS
Determine exposure
for each simulated
individua
Construct sequence of
activity events for each
for each simulated
ncentratior
in microenvironments
for each simulated
(optional; currently
/ Diary events/activities /
and personal information
I (e.g., from CHAD) I
The selected diary records
for each day in the
simulation period
Identify
concentrations in
microenviroments
visited
Microenvironments defined by
grouping of CHAD location codes
Calculate
ventillation
rates
Activity diary pools by day
type/temperature category
Select calculation methods for each
microenvironment:
• Factors
• Mass balance
Hourly concentrations and
minutes spent in each
microenviroment visited by
the simulated individual
onsolidate
location codes into
Hourly ambient air quality data for
the home and work sectors
Average exposure
concentrations
for simulated person
• Hourly
• Daily
• Monthly
• Annual
A sequence of events
(i.e., microenvironments
visited and minutes spent)
in the simulation period
Hourly concentrations in all
Each day in the simulation
period, is assigned with
activity pool # based on
daily max/mean temperature
Average does for
simulated person:
• Hourly
• Daily
• Monthly
• Annual
microenvironments during
simulation period
Other physiological
Maximum/mean daily
temperature data
Physiological parameters
from profile
Figure AX3-93b. Detailed diagram illustrating components of APEX exposure model.
Source: www.epa.gov/ttn/fera/data/apex3227.
o
I-H
H
W
-------
1 doses to the entire populations of interest. This process should aim to quantify, to the extent
2 possible, variability and uncertainty in the various components, assessing their effects on the
3 estimates of exposure.
4 Implementation of the above components of comprehensive exposure modeling has
5 benefitted significantly from recent advances and expanded availability of computational
6 technologies such as Relational Database Management Systems (RDBMS) and Geographic
7 Information Systems (Purushothaman and Georgopoulos, 1997, 1999a,b).
8
9 AX3.11.11.3 Characterization of Ambient Concentrations of Ozone
10 As mentioned earlier, background and regional outdoor concentrations of pollutants over a
11 study domain may be calculated either through emissions-based mechanistic modeling or
12 through ambient-data-based modeling. Emissions-based models calculate the spatiotemporal
13 fields of the pollutant concentrations using precursor emissions and meteorological conditions as
14 inputs. The ambient-data-based models typically calculate spatial or spatiotemporal
15 distributions of the pollutant through the use of interpolation schemes, based on either
16 deterministic or stochastic models for allocating monitor station observations to the nodes of a
17 virtual regular grid covering the region of interest. Kriging, a geostatistical technique, provides
18 standard procedures for generating an interpolated O3 spatial distribution for a given time period,
19 using data from a set of observation points. The kriging approach, with parameters calculated
20 specifically for each hour of the period of concern, was compared to the Urban Airshed Model
21 (UAM-IV), a comprehensive photochemical grid-based model for deriving concentration fields.
22 The concentration fields were then linked with corresponding population data to calculate
23 potential outdoor population exposure. Higher exposure estimates were obtained with the
24 photochemical grid-based model when O3 concentrations were < 120 ppb, however, the situation
25 was reversed when O3 concentrations exceeded 120 ppb. The authors concluded that kriging O3
26 values at the locations studied can reconstruct aspects of population exposure distributions
27 (Georgopoulos et al., 1997a,b).
28 Carroll et al. (1997a,b) developed a spatial-temporal model, with a deterministic trend
29 component, to model hourly O3 levels with the capacity to predict O3 concentrations at any
30 location in Harris County, Texas during the time period between 1980 and 1993. A fast model-
31 fitting method was developed to handle the large amount of available data and the substantial
January 2005 AX3-213 DRAFT-DO NOT QUOTE OR CITE
-------
1 amount of missing data. Ozone concentration data used consisted of hourly measurements from
2 9 to 12 monitoring stations for the years 1980 to 1993. Using information from the census tract,
3 the authors estimated that exposure of young children to O3 declined by approximately 20% over
4 the analysis period. The authors also suggested that the O3 monitors are not sited in locations to
5 adequately measure population exposures. Several researchers have questioned the suitability of
6 the model for addressing spatial variations in O3 (Guttorp et al., 1997; Cressie, 1997; Stein and
7 Fang, 1997).
8 Spatiotemporal distributions of O3 concentrations have alternatively been obtained using
9 methods of the "Spatio-Temporal Random Field" (STRF) theory (Christakos and Vyas,
10 1998a,b). The STRF approach interpolates monitoring data in both space and time
11 simultaneously. This method can analyze information on "temporal trends," which cannot be
12 incorporated directly in purely spatial interpolation methods such as standard kriging. Further,
13 the STRF method can optimize the use of data that are not uniformly sampled in either space or
14 time. The STRF theory was further extended in the Bayesian Maximum Entropy (BME)
15 framework and applied to O3 interpolation studies (Christakos and Hristopulos, 1998; Christakos
16 and Kolovos, 1999; Christakos, 2000). The BME framework can use prior information in the
17 form of "hard data" (measurements), probability law descriptors (type of distribution, mean and
18 variance), interval estimation (maximum and minimum values) and even constraint from
19 physical laws. According to these researchers, both STRF and BME were found to successfully
20 reproduce O3 fields when adequate monitor data are available.
21
22 AX3.11.11.4 Calculation of Microenvironmental Concentrations
23 Once specific ambient/local spatiotemporal O3 concentration patterns have been derived,
24 microenvironments that can represent either outdoor or indoor settings must be characterized.
25 This process can involve modeling of various local sources and sinks as well as
26 interrelationships between ambient/local and microenvironmental concentration levels. Three
27 approaches have been used in the past to model microenvironmental concentrations: empirical,
28 mass balance, and detailed computational fluid dynamics (CFD).
29 The empirical fitting approach has been used to summarize the findings of recent field
30 studies (Liu et al., 1995, 1997; Avol et al., 1998a). These empirical relationships could provide
31 the basis for future, "prognostic" population exposure models.
January 2005 AX3-214 DRAFT-DO NOT QUOTE OR CITE
-------
1 Mass balance modeling has ranged from very simple formulations, assuming ideal
2 (homogeneous) mixing and only linear physicochemical transformations with sources and sinks,
3 to models that take into account complex multiphase chemical and physical interactions and
4 nonidealities in mixing. Mass balance modeling is the most common approach used to model
5 pollutant concentrations in enclosed microenvironments. As discussed earlier, the simplest
6 microenvironmental setting is a homogeneously mixed compartment in contact with possibly
7 both outdoor/local environments as well as with other microenvironments. The air quality of
8 this idealized microenvironment is affected primarily by transport processes (including
9 infiltration of outdoor air into indoor air compartments, advection between microenvironments,
10 and convective transport); sources and sinks (local outdoor emissions, indoor emissions, surface
11 deposition); and local outdoor and indoor gas and aerosol phase chemistry transformation
12 processes (such as the formation of secondary organic and inorganic aerosols).
13 Numerous indoor air quality modeling studies have been reported in the literature;
14 however, depending on the modeling scenario, only a limited number address physical and
15 chemical processes that affect O3 concentrations indoors (Nazaroff and Cass, 1986; Hayes, 1989,
16 1991). An example of a mass balance indoor air model for O3 and benzene can be found in the
17 work of Freijer and Bloemen (2000). They used outdoor O3 measurements to parameterize a
18 simplified linearized formulation of transport, transformation, and sources and sinks in the
19 indoor microenvironment.
20 The pNEM/O3 model includes a sophisticated mass balance model for enclosed (indoor
21 and vehicle) microenvironments. The general form of this mass balance model is a differential
22 equation that accounts for outdoor concentration, AER, penetration rate, decay rate, and indoor
23 sources. Each of these parameters is represented by a probability distribution or by a dynamic
24 relationship to other parameters that may change according to time of day, temperature, air
25 conditioning status, window status, or other factors (Johnson, 2003).
26 Few indoor air models have considered detailed nonlinear chemistry, which, however, can
27 have a significant effect on the indoor air quality, especially in the presence of strong indoor
28 sources. Indeed, the need for more comprehensive models that can take into account the
29 complex, multiphase processes that affect indoor concentrations of interacting gas phase
30 pollutants and particulate matter has been recognized and a number of formulations have
31 appeared in recent years. For example, the Exposure and Dose Modeling and Analysis System
January 2005 AX3-215 DRAFT-DO NOT QUOTE OR CITE
-------
1 (EDMAS) (Georgopoulos et al., 1997a) included an indoor model with detailed gas-phase
2 atmospheric chemistry. This indoor model accounts for interactions of O3 with indoor sinks and
3 sources (surfaces, gas releases) and with entrained gas. The indoor model was dynamically
4 coupled with the outdoor photochemical air quality models, UAM-IV and UAM-V (Urban
5 Airshed Models), which provided the gas-phase composition of entrained air, and with a
6 physiologically based O3 uptake and dosimetry model. Subsequent work (Isukapalli and
7 Georgopoulos, 2000; Isukapalli et al., 1999) expanded the framework and features of the
8 EDMAS model to incorporate alternative representations of gas-phase chemistry as well as
9 multiphase O3 chemistry and gas/aerosol interactions. The new model is a component of the
10 integrated Modeling Environment for Total Risk studies (MENTOR).
11 Sarwar et al. (2001, 2002) modeled estimates of indoor hydroxyl radical concentrations
12 using a new indoor air quality model, Indoor Chemistry and Exposure Model (ICEM). The
13 ICEM uses a modified SAPRC-99 atmospheric chemistry mechanism to simulate indoor
14 hydroxyl radical production and consumption from reactions of alkenes with O3. It also allows
15 for the simulation of transport processes between indoor and outdoor environments, indoor
16 emissions, chemical reactions, and deposition. Indoor hydroxyl radicals, produced from O3 that
17 penetrates indoors, can adversely impact indoor air quality through dark chemistry to produce
18 photochemical oxidants.
19 S0rensen and Weschler (2002) used CFD modeling to examine the production of a
20 hypothetical product from an O3-terpene reaction under two different ventilation scenarios. The
21 computational grid used in the model was nonuniform. There were significant variations in the
22 concentrations of reactants between locations in the room, resulting in varying reaction rates.
23 Because the "age of the air" differed at different locations in the room, the time available for
24 reactions to occur also differed between locations.
25 Very few studies have focused on mechanistic modeling of outdoor microenvironments.
26 Fraigneau et al. (1995) developed a simple model to account for fast NO-O3 reaction/dispersion
27 in the vicinity of a motorway. Proyou et al. (1998) applied a simple three-layer photochemical
28 box model to an Athens street canyon. However, the adjustments of O3 levels for sources, sinks,
29 and mixing in outdoor microenvironments are done in a phenomenological manner in existing
30 exposure models, driven by limited available observations. On-going research is evaluating
31 approaches for quantifying local effects on outdoor O3 chemistry in specific settings.
January 2005 AX3-216 DRAFT-DO NOT QUOTE OR CITE
-------
1 AX3.11.12 Population Exposure Models: Considerations of Activity Events
2 and Population Demographics
3 Existing comprehensive inhalation exposure models treat human activity patterns as
4 sequences of exposure events in which each event is defined by a geographic location and
5 microenvironment. The EPA has supported the most comprehensive efforts in this area, leading
6 to the development of the National Ambient Air Quality Standard Exposure Model and
7 Probabilistic National Ambient Air Quality Standard Exposure Model (NEM and pNEM)
8 (Johnson, 2003) and the Simulation of Human Exposure and Dose System (SHEDS) (Burke
9 et al., 2001; McCurdy et al., 2000). The Air Pollutants Exposure model (APEX, version 3)
10 [http://www.epa.gov/ttn/fera/data/apex322], a PC-based model, derived from the probabilistic
11 NAAQS Exposure Model for carbon monoxide (pNEM/CO), will be used to help estimate
12 human population exposure for criteria and air toxic pollutants as part of the EPA's overall Total
13 Risk Integrated Methodology (TRIM) model framework. The model simulates the movement of
14 individuals through time and space and their exposure to the given pollutant in indoor, outdoor,
15 and in-vehicle microenvironments. Recent European efforts have produced some formulations
16 that have similar general attributes as the above models but generally involve major
17 simplifications in some of their components. Examples of recent European models addressing
18 O3 exposures include the AirPEx (Air Pollution Exposure) model (Freijer et al., 1998), which
19 basically replicates the pNEM approach, and the AirQUIS (Air Quality Information System)
20 model (Clench-Aas et al., 1999). A discussion of databases on time-activity data, and their
21 influence on estimates of long-term ambient O3 exposure, can be found in Kunzli et al. (1997),
22 McCurdy (2000), and McCurdy et al. (2000).
23 The NEM/pNEM, APEX, and SHEDS families of models provide exposure estimates,
24 defined by concentration and minute ventilation rate for each individual exposure event, and
25 provide distributions of exposure and O3 dose over any averaging period of concern from 1 h to
26 an entire O3 season. The above families of models also simulate certain aspects of the variability
27 and uncertainty in the principal factors affecting exposure. pNEM divides the population of
28 interest into representative cohorts based on the combinations of demographic characteristics
29 (age, gender, employment), home/work district, residential cooking fuel, and then assigns
30 activity diary records from CHAD (Consolidated Human Activities Database;
31 www.epa.gov/chadnetl) (Glen et al., 1997) to each cohort according to demographic
January 2005 AX3-217 DRAFT-DO NOT QUOTE OR CITE
-------
1 characteristic, season, day-type (weekday/weekend), and temperature. APEX and SHEDS
2 generates a population demographic file containing a user-defined number of person-records for
3 each census tract of the population based on proportions of characteristic variables (age, gender,
4 employment, housing) obtained for the population of interest, and then assigns the matching
5 activity information from CHAD to each individual record of the population, based on the
6 characteristic variables.
7 The latest version of APEX allows for finer geographical units such as census tracts and
8 automatically assigns population to the nearest monitor within a cutoff distance. Exposure
9 district-specific temperatures can be specified and the user can select the variables that affect
10 each parameter (e.g., the AER parameter in certain indoor microenvironments may depend on air
11 conditioning status or window position). The mass balance algorithms have been enhanced to
12 allow window position or vehicle speed to also be considered in determining AERs.
13 These models also allow for travel between census tracts for work. By specifying the
14 commuting patterns, the variation of exposure concentrations due to commuting in different
15 census tracts can be captured. The essential attributes of the pNEM, APEX, SHEDS, and
16 MENTOR/SHEDS approaches are summarized in Table AX3-31.
17 From the above families of models only NEM/pNEM implementations have been
18 extensively applied to O3 studies. However, it is anticipated that APEX will be useful as an
19 exposure modeling tool for assessing both criteria and hazardous air pollutants in the future.
20 Recently, SHEDS has been modified and incorporated into MENTOR-OPERAS (Modeling
21 ENvironment for TOtal Risk — Ozone and Particles Exposure and Risk Analysis System). This
22 variant of SHEDS includes detailed indoor chemistry and other O3-relevant microenvironmental
23 processes, while providing interactive linking with CHAD for consistent definition of population
24 characteristics and activity events. Nevertheless, the focus of the following will be on pNEM/O3
25 implementations and applications, as they constitute the majority of the published O3 exposure
26 studies. The 1996 O3 AQCD (U.S. Environmental Protection Agency, 1996a) also focused on
27 the pNEM/O3 family of models, referring to the review by McCurdy (1994) for the fundamental
28 principles underlying its formulation and listing, in addition to the "standard" version, three
29 pNEM/O3-derived models (the Systems Applications International NEM [SAI/NEM]; the
30 Regional Human Exposure Model [REHEX]; and the Event Probability Exposure Model
31 [EPEM]).
January 2005 AX3-218 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-31. Attributes of pNEM, SHEDS, APEX, and MENTOR/SHEDS Population Exposure Models
3
to
o
o
Exposure Estimate
Characterization of the
pNEM
Hourly averaged
Yes
SHEDS
Hourly averaged
Yes
APEX
Hourly averaged
Yes
MENTOR/SHEDS
Activity-event
Yes
based
X
OJ
I
to
VO
High-End Exposures
Spatial Scale/Resolution
Urban areas/Census tract
level
Temporal Scale/Resolution A year/one hour
Population Activity Patterns
Assembling
Microenvironment
Concentration Estimation
Top-down approach
Non-steady-state and steady-
state mass balance equations
(hard-coded)
Microenvironmental (ME)
Factors
Random samples from
probability distributions
Urban areas/Census tract
level
A year/one hour
Bottom-up approach
Steady-state mass balance
equation (residential) and
linear relationship method
(non-residential)
(hard-coded)
Random samples from
probability distributions
Urban area/census tract level
A year/one hour
Bottom-up approach
Non-steady-state mass
balance and linear
relationship method
(flexibility of selecting
algorithms)
Random samples from
probability distributions
Urban areas/Census tract
level
A year/activity-event based
time step
Bottom-up approach
Non-steady-state mass
balance equation with
nonlinear indoor air
chemistry module or
regression methods
(flexibility of selecting
algorithms)
Random samples from
probability distributions
H
6
o
0
H
O
Specification of Indoor
Source Emissions
Yes (gas-stove, tobacco
smoking)
Yes (gas-stove, tobacco
smoking, other sources)
Yes (multiple sources
defined by the user)
Yes (multiple sources
defined by the users)
o
HH
H
W
-------
1 Rifai et al. (2000) compared applications of an updated version of REHEX, REHEX-II.
2 The applications used NHAPS data for the southern states and the 48-state NHAPS or the
3 Houston-specific time-activity pattern data. The results indicated a sensitivity to the specificity
4 of the activity data: using Houston-specific data resulted in higher estimates of human exposure
5 in some of the scenarios. For example, using NHAPS data lead to an estimated 275 thousand-
6 exposure-hours between 120 to 130 ppb, while use of the Houston-specific activity data lead to
7 an estimated 297 thousand-exposure-hours between 120 and 130 ppb (8% higher). Using the
8 Houston-specific activity data in the model resulted in about 2,400 person-exposure-hours above
9 190 to 200 ppb O3 while no exposure above this threshold was estimated when the NHAPS
10 activity were used in the model.
11 The pNEM family of models used by the EPA has evolved considerably since the
12 introduction of the first NEM model in the 1980s (Biller et al., 1981). The first such
13 implementations of pNEM/O3 in the 1980s used a reduced form of a mass balance equation to
14 estimate indoor O3 concentrations from outdoor concentrations. The second generation of
15 pNEM/O3 was developed in 1992 and used a simple mass balance model to estimate indoor O3
16 concentrations. Subsequent enhancements to pNEM/O3 and its input databases included
17 revisions to the methods used to estimate equivalent ventilation rates (ventilation rate divided by
18 body surface), to determine commuting patterns, and to adjust ambient O3 levels to simulate
19 attainment of proposed NAAQS. During the mid-1990s, the EPA applied updated versions of
20 pNEM/O3 to three different population groups in nine selected urban areas (Chicago, Denver,
21 Houston, Los Angeles, Miami, New York, Philadelphia, St. Louis, and Washington): (1) the
22 general population of urban residents, (2) outdoor workers, and (3) children who tended to spend
23 more time outdoors than the average child. Reports by Johnson et al. (1996a,b,c) describe these
24 versions of pNEM/O3 and summarize the results of the application of the model to the nine urban
25 areas. These versions of pNEM/O3 used a revised probabilistic mass balance model to determine
26 O3 concentrations over 1-h periods in indoor and in-vehicle microenvironments (Johnson, 2003).
27 The model assumed that there are no indoor sources of O3, that the outdoor O3 concentration and
28 AER during the clock hour is constant at a specified value, and that O3 decays at a rate
29 proportional to the outdoor O3 concentration and the indoor O3 concentration.
30 The new pNEM-derived model, APEX, differs from earlier pNEM models in that the
31 probabilistic features of the model are incorporated into a Monte Carlo framework. Instead of
January 2005 AX3-220 DRAFT-DO NOT QUOTE OR CITE
-------
1 dividing the population of interest into a set of cohorts, APEX generates individuals as if they
2 were being randomly sampled from the population. APEX provides each generated individual
3 with a demographic profile that specifies values for all parameters required by the model. The
4 values are selected from distributions and databases that are specific to the age, gender, and other
5 specifications stated in the demographic profile. The EPA plans to develop future versions of
6 APEX applicable to O3 and other criteria pollutants. As mentioned earlier, the combined
7 SHEDS/MENTOR-OPERAS system has also adopted this approach.
8 An important source of uncertainty in existing exposure modeling involves the creation of
9 multiday, seasonal, or year long exposure activity sequences based on 1- to 3-day activity data
10 for any given individual from CHAD. Currently, appropriate longitudinal data are not available
11 and the existing models use various rules to derive longer-term activity sequences using 24-h
12 activity data from CHAD.
13
14 AX3.11.13 Recent Developments: Activity Events and Inhalation Intake
15 An important development in inhalation exposure modeling, including O3 exposure
16 modeling, has been the consolidation of existing information on activity event sequences in
17 CHAD (McCurdy, 2000; McCurdy et al., 2000). Indeed, most recent exposure models are
18 designed (or have been redesigned) to obtain such information from CHAD. There are now
19 about 22,600 person-days of sequential daily activity pattern data in CHAD. All ages of both
20 genders are represented in CHAD. The data for each subject consist of one or more days of
21 sequential activities, in which each activity is defined by start time, duration, activity type (140
22 categories), and microenvironment classification (110 categories). Activities vary from 1 min to
23 1 h in duration. Activities longer than 1 h are subdivided into clock-hour durations to facilitate
24 exposure modeling. A distribution of values for the ratio of oxygen uptake rate to body mass
25 (referred to as metabolic equivalents or METs) is provided for each activity type listed. The
26 forms and parameters of these distributions were determined through an extensive review of the
27 exercise and nutrition literature. The primary source of distributional data was Ainsworth et al.
28 (1993), a compendium developed specifically to "facilitate the coding of physical activities and
29 to promote comparability across studies."
30 Use of the information in CHAD provides a rational way for incorporating realistic intakes
31 into exposure models by linking inhalation rates to activity information. As mentioned earlier,
January 2005 AX3-221 DRAFT-DO NOT QUOTE OR CITE
-------
1 an exposure event sequence derived from activity-diary data is assigned to each population unit
2 (cohort for pNEM- or REHEX-type models, or individual for APEX- or SHEDS-type models).
3 Each exposure event is typically defined by a start and duration time, a geographic location and
4 microenvironment, and activity level. The most recent pNEM, APEX, and SHEDS models have
5 defined activity levels using the activity classification coding scheme incorporated into CHAD.
6 A probabilistic module within the APEX- and SHEDS-type models converts the activity
7 classification code of each exposure event to an energy expenditure rate, which in turn is
8 converted into an estimate of oxygen uptake rate. The oxygen uptake rate is then converted into
9 an estimate of ventilation rate ( VE), expressed in L/min. Johnson (2001) reviewed the
10 physiological principles incorporated into the algorithms used in pNEM and APEX to convert
11 each activity classification code to an oxygen uptake rate and describes the additional steps
12 required to convert oxygen uptake to VE.
13 McCurdy (1997a,b, 2000) recommended that ventilation rate be estimated as a function of
14 energy expenditure rate. The energy expended by an individual during a particular activity can
15 be expressed as:
16
EE - (MET)(RMR) (AX3-5)
17
18 where EE is the average energy expenditure rate (kcal/min) during the activity, MET (metabolic
19 equivalent of work) is a ratio specific to the activity and is dimensionless, and RMR is the
20 resting metabolic rate of the individual expressed in terms of number of energy units expended
21 per unit of time (kcal/min). If RMR is specified for an individual, then the above equation
22 requires only an activity-specific estimate of MET to produce an estimate of the energy
23 expenditure rate for a given activity. McCurdy et al. (2000) developed MET distributions for the
24 activity classifications appearing in the CHAD database.
25 Finally, one issue that should be mentioned is that of evaluating comprehensive prognostic
26 exposure modeling studies, for either individuals or populations, with field data. Attempts had
27 been made to evaluate pNEM/O3-type models using personal exposure measurements (Johnson
28 et al., 1990). Although databases that would be adequate for performing a comprehensive
29 evaluation are not expected to be available any time soon, a number of studies are building the
30 necessary information base, as discussed previously. Some of these studies report field
January 2005 AX3-222 DRAFT-DO NOT QUOTE OR CITE
-------
1 observations of personal, indoor, and outdoor O3 concentrations and describe simple
2 semiempirical personal exposure models that are parameterized using observational data and
3 regression techniques.
4
5 AX3.11.14 Characterization of Exposure
6 AX3.11.14.1 Use of Ambient Ozone Concentrations
7 The use of ambient air monitoring stations is still the most common surrogate for assigning
8 exposure in epidemiological studies. Since the primary source of O3 exposure is the ambient air,
9 monitoring concentration data would provide the exposure outdoors while exercising, a potential
10 important exposure to evaluate in epidemiological studies as well as a relative assignment of
11 exposure with time if the concentration were uniform across the region; the time-activity pattern
12 were the same across the population; and the housing characteristics, such as ventilation rates
13 and the O3 sinks contributing to its indoor decay rates, were constant for the study area. Since
14 these factors vary by population and location there will be errors in not only the magnitude of the
15 total exposure, but also in the relative total exposure assignment based solely on ambient
16 monitoring data. As discussed earlier in this section, spatial differences in O3 concentrations
17 within a city and between the height of the monitor and the breath zone (1 to 2 m) exist,
18 increasing uncertainties. The potential exists to obtain more complete exposure assignments for
19 both individuals and populations by modeling O3 exposure based on ambient air concentration to
20 account for spatial variations outdoors and for time spent indoors, provided housing
21 characteristics and activity patterns can be obtained. For cohort studies, measurement of
22 personal O3 exposures using passive monitors is also possible.
23 The potential for error in determining pollutant exposure was also expressed by
24 Krzyzanowki (1997), who indicated that while the typical estimate of exposure in
25 epidemiological studies is "an average concentration of the pollutant calculated from the data
26 routinely collected in the area of residence of the studied population. This method certainly
27 lacks precision and, in most cases, the analyses that use it will underestimate the effect of
28 specific concentrations of a pollutant on health." It is further stated that when estimating
29 exposure for epidemiological studies it is important to define: (1) representativeness of exposure
30 or environmental data for the population at risk, (2) appropriateness of the averaging time for the
January 2005 AX3-223 DRAFT-DO NOT QUOTE OR CITE
-------
1 health outcome being examined, and (3) the relationship between the exposure surrogate and the
2 true exposure relative to the exposure-response function used in the risk assessment.
3 Numerous air pollutants can have common ambient air sources resulting in strong
4 correlations among pollutant ambient air concentrations. As a result, some observed
5 associations between an air pollutant and health effects may be due to confounding by other air
6 pollutants. Sarnat et al. (2001) found that while ambient air concentrations of some air
7 pollutants were correlated, personal PM2 5 and several gaseous air pollutant (O3, SO2, NO2, CO,
8 and exhaust-related VOCs) exposures were not generally correlated. The findings were based on
9 the results of a multipollutant exposure study of 56 children and elderly adults in Baltimore, MD
10 conducted during both the summer and winter months. Ambient pollutant concentrations were
11 not associated with corresponding personal exposures, except for PM2 5. The gaseous pollutants
12 were found to be surrogates of PM25 and were generally not correlated. The authors concluded
13 that multipollutant models in epidemiologic studies of PM2 5 may not be suitable, and health
14 effects attributed to the gaseous pollutants may be the result of PM25 exposure. It should be
15 noted that the 95th percentile O3 concentrations in the study was lower than 60 ppb, an O3
16 concentration at which respiratory effects are noted. It would be important to examine whether
17 O3 is a surrogate for personal PM25 at high O3 levels when attributing adverse health effects to
18 O3orPM25.
19 Kunzli et al. (1996) assessed potential lifetime exposure to O3 based on the responses to a
20 standardized questionnaire completed by 175 college freshmen in California. Questions
21 addressed lifetime residential history, schools attended, general and outdoor activity patterns,
22 driving habits and job history. The purpose was to determine what O3 monitoring data to use for
23 each time period of their lives, the potential correction factor for indoor levels and periods of
24 high activity to account for differential doses to the lung due to physical activity. The reliability
25 of the responses was checked by having each respondent complete the questionnaire twice, on
26 different days, and the results compared. A lifetime O3 exposure history was generated for each
27 participant and a sensitivity analysis performed to evaluate which uncertainties would cause the
28 greatest potential misclassification of exposure. Assigned lifetime O3 concentrations from the
29 nearest monitor yielded highly reliable cumulative values, although the reliability of residential
30 location decreased with increasing residential locations. Individuals involved in moderate and
January 2005 AX3-224 DRAFT-DO NOT QUOTE OR CITE
-------
1 heavy exercise could be reliably identified. Such an approach can be used to evaluate health
2 outcomes associated with chronic exposures to O3.
3
4 AX3.11.14.2 Exposure Selection in Controlled Exposure Studies
5 Ozone exposures in the environment are variable over time due to changes in the ambient
6 concentrations during the day as the photochemical reactions proceed and also because people
7 move between microenvironments that have different concentrations (Johnson, 1997).
8 Exposures are repeated on sequential days since weather conditions that produce O3 can move
9 slowly through or become stagnant within a region. For simplicity, most controlled-exposure
10 experiments are conducted at a single concentration for a fixed time period, with a limited
11 number of studies being repeated on a single individual. Few studies have been conducted using
12 multipollutants or photochemical agents other than O3, to better represent "real-world" exposures
13 with the exception of NO2. Studies by Hazucha et al. (1992), Adams (2003), and Adams and
14 Ollison (1997) examined the effect of varying O3 exposure concentrations on pulmonary
15 function. A description of, and findings in, the studies appears in Annex AX5 of this document.
16
17 AX3.11.14.3 Exposure to Related Photochemical Agents
18 Exposures to other related photochemical agents have not been measured using personal
19 samplers nor are these agents routinely measured at O3 monitoring stations. Photochemical
20 agents produced in the ambient air can penetrate indoors and react with other pollutants to
21 produce other potentially irritating compounds. Reiss et al. (1995) reported that organic acids,
22 aldehydes and ketones were produced indoors by reactions of O3 with VOCs. The produced
23 compounds included oxidants that can be respiratory irritants. The indoor concentrations were
24 dependent upon the O3 concentrations indoors and the AER within the building. Weschler and
25 Shields (1997) summarized indoor air chemical reactions that depend directly or indirectly on
26 the presence of O3. They indicated that O3 concentrations are lower indoors than outdoors partly
27 because of gas-phase reactions that produce other oxidants in an analogous fashion to
28 photochemical smog in ambient air. The production of these species indoors is a function of the
29 indoor O3 concentration and the presence of the other necessary precursors, VOCs, and NO2,
30 along with an optimal AER. A variety of the photochemical oxidants related to O3 that are
31 produced outdoors, such as PAN and PPN, can penetrate indoors. These oxidants are thermally
January 2005 AX3-225 DRAFT-DO NOT QUOTE OR CITE
-------
1 unstable and can decompose indoors to peroxacetyl radicals and NO2 through thermal decay.
2 PAN removal increases with increasing temperature, and at a given temperature, with increasing
3 NO/NO2 concentration ratio (Grosjean et al., 2001). Other free radicals that can form indoors
4 include HO and HO2'. These free radicals can produce compounds that are known or suspected
5 to be irritating. Little is known about exposure to some of these agents, as not all have been
6 identified and collection and analytical methodologies have not be developed for their routine
7 determination. Lee et al. (1999) reported that homogeneous (gas phase) and heterogenous (gas
8 phase/solid surface) reactions occur between O3 and common indoor air pollutants such as NO
9 and VOCs to produce secondary products whose production rate depends on the AER and
10 surface area within the home. Wainman et al. (2000) found that O3 reacts indoors with d-
11 limonene, emitted from air fresheners, to form fine particles in the range of 0.1 to 0.2 |im and
12 0.2 to 0.3 |im. The indoor process also produces compounds that have been identified in the
13 ambient atmosphere. These species, plus others that may form indoors from other terpenes or
14 unsaturated compounds can present an additional exposure to oxidants, other than O3, at higher
15 concentrations than present in ambient air, even as the O3 concentration is being reduced
16 indoors.
17
18 AX3.11.14.4 Exposure to Sensitive Populations
19 Personal O3 concentrations have been measured for populations potentially susceptible to
20 respiratory irritants including children, outdoor workers, the elderly, and individuals with
21 chronic obstructive pulmonary disease (Table AX3-32). Children and outdoor workers have
22 somewhat higher exposures than other individuals because they spend more time outdoors
23 engaged in moderate and heavy exertion. Children are also more active outside and, therefore,
24 have a higher minute ventilation rate than most adults (Klepeis et al., 1996, 2001). While the
25 current data suggest trends in exposure magnitude for some populations, additional exposure
26 studies are needed to generalize differences in exposure between the general population and
27 potentially susceptible populations. Exposure modeling can be used to extend the current
28 information available from measurement studies. Several states report days when high O3 are
29 occurring (O3 alert days) to encourage sensitive individuals to modify their behavior to reduce
30 their exposure to O3. However, this change in behavior is not recorded in a database of time-
31 activity patterns or used in exposure modeling as a discrete entity for sensitive populations.
January 2005 AX3-226 DRAFT-DO NOT QUOTE OR CITE
-------
Table AX3-32. Personal Exposure Concentrations
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.
1 Ozone exposure modeling has been conducted for the general population and sensitive
2 subgroups. The pNEM/O3 model takes into consideration the temporal and spatial distribution
3 of people and O3 throughout the area of consideration, variations in O3 concentrations within
4 microenvironments, and the effects of exertion/exercise (increased ventilation) on O3 uptake.
5 The pNEM/O3 model consists of two principal parts: the cohort exposure program and the
6 exposure extrapolation program. The methodology incorporated much of the general framework
7 described earlier in this section on assessing O3 exposure and consists of five steps: (1) define
8 the study area, population of interest, subdivisions of the study area, and exposure period;
January 2005
AX3-227 DRAFT-DO NOT QUOTE OR CITE
-------
1 (2) divide population of interest into a set of cohorts; (3) develop exposure event sequence for
2 each cohort for the exposure period; (4) estimate pollutant concentration and ventilation rate for
3 each exposure event; and (5) extrapolate cohort exposures to population of interest (U.S.
4 Environmental Protection Agency, 1996b).
5 There are three versions of the pNEM/O3 model: general population (Johnson et al.,
6 1996a), outdoor workers (Johnson et al., 1996b), and outdoor children (Johnson et al., 1996c,
7 1997). These three versions of the model have been applied to nine urban areas. The model also
8 has been applied to a single summer camp (Johnson et al., 1996c). The general population
9 version of the model uses activity data from the Cincinnati Activity Diary Study (CADS;
10 Johnson, 1989). Time-activity studies (Wiley et al., 1991a; Johnson, 1984; Linn et al., 1993;
11 Shamoo et al., 1991; Goldstein et al., 1992; Hartwell et al., 1984) were combined with the CADS
12 data for the outdoor worker version of the model. Additional time-activity data (Goldstein et al.,
13 1992; Hartwell et al., 1984; Wiley et al., 1991a,b; Linn et al., 1992; Spier et al., 1992) were also
14 added to CADS for the outdoor children of the model (U.S. Environmental Protection Agency,
15 1996b). Home-work commuting patterns are based on information gathered by the U.S. Census
16 Bureau. Ozone ambient air concentration data from monitoring stations were used to estimate
17 the outdoor exposure concentrations associated with each exposure event. Indoor O3 decay rate
18 is assumed to be proportional to the indoor O3 concentration. An algorithm assigns the
19 equivalent ventilation rate (EVR) associated with each exposure event. The outdoor children
20 model uses an EVR-generator module to generate an EVR value for each exposure event based
21 on data on heart rate by Spier et al. (1992) and Linn et al. (1992). The models produce exposure
22 estimates for a range of O3 concentrations at specified exertion levels. The models were used to
23 estimate exposure for nine air quality scenarios (U.S. Environmental Protection Agency, 1996b).
24 Korc (1996) used the REHEX-II model, a general purpose air pollution exposure model
25 based on a microenvironmental approach modified to account for the influence of physical
26 activity along and spatial and the temporal variability of outdoor air pollution. Ozone exposure
27 was estimated by demographic groups across 126 geographic subregions for 1980 to 1982, and
28 for 142 geographic subregions for 1990 to 1992. Simulation results were determined for
29 population race, ethnicity, and per capita income and included indoor, in-transit, and outdoor
30 microenvironments. Exposure modeling was stratified by age because of differences in time-
31 activity patterns. Exposure distributions by regional activity pattern data were not considered,
January 2005 AX3-228 DRAFT-DO NOT QUOTE OR CITE
-------
1 rather it was assumed that all individuals within a county had the same exposure distribution by
2 race, ethnicity, and socioeconomic status. Model results for southern California indicated that
3 the segment of the population with the highest exposures were children 6 to 11 years old.
4 Individuals living in low income districts may have greater per capita hours of exposure to O3
5 above the NAAQS than those living in higher income districts. The author indicated that O3
6 exposure differences by race and ethnicity have declined over time. The noninclusion of details
7 on activity patterns for different populations in the model limit the extrapolations that can be
8 made from the model results.
9 Children appear to have higher exposures than adults and the elderly. Asthmatics appear to
10 ventilate more than healthy individuals, but tend to protect themselves by decreasing their
11 outdoor exercise (Linn et al., 1992). Additional data are still needed to identify and better define
12 exposures to potentially susceptible populations and improve exposure models for the general
13 population and subpopulation of concern.
14
15
January 2005 AX3-229 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2 Adams, W. C. (1993) Measurement of breathing rate and volume in routinely performed daily activities [final
3 report]. Sacramento, CA: California Environmental Protection Agency, Air Resources Board; contract no.
4 A033-205.
5 Adams, W. C. (2003) Comparison of chamber and face mask 6.6-hour exposure to 0.08 ppm ozone via square-wave
6 and triangular profiles on pulmonary responses. Inhalation Toxicol. 15: 265-281.
7 Adams, W. C.; Ollison, W. M. (1997) Effects of prolonged simulated ambient ozone dosing patterns on human
8 pulmonary function and symptomatology. Presented at: 90th annual meeting of the Air & Waste Management
9 Association; June; Toronto, Ontario, Canada. Pittsburgh, PA: Air & Waste Management Association; paper
10 no. 97-MP9.02.
11 Adams, R. M.; Hamilton, S. A.; McCarl, B. A. (1985) An assessment of the economic effects of ozone on U.S.
12 agriculture. J. AirPollut. Control Assoc. 35: 938-943.
13 Adams, R. M.; Glyer, J. D.; Johnson, S. L.; McCarl, B. A. (1989) A reassessment of the economic effects of ozone
14 on U.S. agriculture. JAPCA 39: 960-968.
15 Ainsworth; Haskell, W. L.; Leon, A. S.; Jacobs, D. R., Jr.; Montoye, H. J.; Sallis, J. F.; Paffenbarger, R. S., Jr.
16 (1993) Compendium of physical activities: classification of energy costs of human physical activities. Med.
17 Sci. Sports Exer. 25: 71 -80.
18 Aldrich, F. D.; Brooks, B. O.; Davis, W. F.; DeBroy, J. A.; Li, S. T.; Schimke, R.; Utter, G. M. (1995) Health and
19 productivity impact of organic volatiles from electronic office equipment. Presented at: Pacific Rim
20 conference on occupational and environmental health; October; Sydney, Australia.
21 Allen, R. J.; Wadden, R. A.; Ross, E. D. (1978) Characterization of potential indoor sources of ozone. Am. Ind. Hyg.
22 Assoc. J. 39: 466-471.
23 Aneja, V. P.; Li, Z. (1992) Characterization of ozone at high elevation in the eastern United States: trends, seasonal
24 variations, and exposure. J. Geophys. Res. [Atmos.] 97: 9873-9888.
25 Avol, E. L.; Navidi, W. C.; Colome, S. D. (1998a) Modeling ozone levels in and around southern California homes.
26 Environ. Sci. Technol. 32: 463-468.
27 Avol, E. L.; Navidi, W. C.; Rappaport, E. B.; Peters, J. M. (1998b) Acute effects of ambient ozone on asthmatic,
28 wheezy, and healthy children. Cambridge, MA: Health Effects Institute; research report no. 82.
29 Beine, H. J.; Jaffe, D. A.; Herring, J. A.; Kelley, J. A.; Krognes, T.; Stordal, F. (1997) High-latitude springtime
30 photochemistry. Part I: NOX, PAN and ozone relationships. J. Atmos. Chem. 27: 127-153.
31 Bernard, N. L.; Gerber, M. J.; Astre, C. M.; Saintot, M. J. (1999) Ozone measurement with passive samplers:
32 validation and use for ozone pollution assessment in Montpellier, France. Environ. Sci. Technol. 33: 217-222.
33 Berntsen, T. K.; Isaksen, I. S. A.; Myhre, G.; Fuglestvedt, J. S.; Stordal, F.; Larsen, T. A.; Freckleton, R. S.; Shine,
34 K. P. (1997) Effects of anthropogenic emissions on tropospheric ozone and its radiative forcing. J. Geophys.
35 Res. [Atmos.] 102: 28,101-28,126.
36 Berry, C. R. (1964) Differences in concentrations of surface oxidant between valley and mountaintop conditions in
37 the southern Appalachians. J. AirPollut. Control Assoc. 14: 238-239.
38 Bey, I.; Jacob, D. J.; Yantosca, R. M.; Logan, J. A.; Field, B.; Fiore, A. M.; Li, Q.; Liu, H.; Mickley, L. J.; Schultz,
39 M. G. (200 la) Global modeling of tropospheric chemistry with assimilated meteorology: model description
40 and evaluation. J. Geophys. Res. (Atmos.) 106: 23,073-23,095.
41 Bey, I.; Jacob, D. J.; Logan, J. A.; Yantosca, R. M. (200Ib) Asian chemical outflow to the Pacific in spring: origins,
42 pathways, and budgets. J. Geophys. Res. [Atmos.] 106: 23,097-23,113.
43 Biller, W. F.; Feagans, T. B.; Johnson, T. R.; Duggan, G. M.; Paul, R. A.; McCurdy, T.; Thomas, H. C. (1981)
44 A general model for estimating exposure associated with alternative NAAQS. Presented at: 74th annual
45 meeting of the Air Pollution Control Association; June; Philadelphia, PA. Pittsburgh, PA: Air Pollution
46 Control Association; paper no. 81-18.4.
47 Black, M. S.; Worthan, A. W. (1999) Emissions from office equipment. In: Raw, G.; Aizlewood, C.; Warren, P.,
48 eds. Indoor Air '99, v. 2. London, England: Construction Research Communications Ltd.; pp. 454-459.
49 Black, D. R.; Harley, R. A.; Hering, S. V.; Stolzenburg, M. R. (2000) A new, portable, real-time monitor. Environ.
50 Sci. Technol. 34: 3031-3040.
51 Blanchard, C. L.; Tanenbaum, S. J. (2003) Differences between weekday and weekend air pollutant levels in
52 southern California. J. Air Waste Mange. Assoc. 53: 816-828.
53 Bohm, M.; McCune, B.; Vandetta, T. (1991) Diurnal curves of troposheric ozone in the western United States.
54 Atmos. Environ. Part A 25: 1577-1590.
January 2005 AX3-230 DRAFT-DO NOT QUOTE OR CITE
-------
1 Bottenheim, J. W.; Barrie, L. A.; Atlas, E.; Heidt, L. E.; Niki, H.; Rasmussen, R. A.; Shepson, P. B. (1990)
2 Depletion of lower tropospheric ozone during Arctic spring: The Polar Sunrise Experiment 1988. J. Geophys.
3 Res. (Atmos.) 95: 18,555-18,568.
4 Bottenheim, J. W.; Barrie, L. A.; Atlas, E. (1993) The partitioning of nitrogen oxides in the lower Arctic troposphere
5 during Spring 1988. J. Atmos. Chem. 17: 15-27.
6 Brauer, M; Brook, J. R. (1997) Ozone personal exposures and health effects for selected groups residing in the
7 Fraser Valley. In: Steyn, D. G.; Bottenheim, J. W., eds. The Lower Fraser Valley Oxidants/Pacific '93 Field
8 Study. Atmos. Environ. 31: 2113-2121.
9 Bravo, H.; Echeverria, R. S.; Sanchez, P.; Palomera, M.; Lefohn, A. S. (2003) The "piston effect" observed in ozone
10 concentrations in Mexico City. EM (August): 16-19.
11 Brice, K. A.; Penkett, S. A.; Atkins, D. H. F.; Sandalls, F. J.; Bamber, D. J.; Tuck, A. F.; Vaughan, G. (1984)
12 Atmospheric measurements of peroxyacetylnitrate (PAN) in rural, south-east England: seasonal variations
13 winter photochemistry and long-range transport. Atmos. Environ. 18: 2691-2702.
14 Browell, E. V; Hair, J. W.; Butler, C. F.; Grant, W. B.; De Young, R. J.; Fenn, M. A.; Brackett, V G.; Clayton,
15 M. B.; Brasseur, L. A.; Harper, D. B.; Ridley, B. A.; Klonecki, A. A.; Hess, P. G.; Emmons, L. K.; Tie, X.;
16 Atlas, E. L.; Cantrell, C. A.; Wimmers, A. J.; Blake, D. R.; Coffey, M. T.; Hannigan, J. W.; Dibb, J. E.;
17 Talbot, R. W.; Flocke, F.; Weinheimer, A. J.; Fried, A.; Wert, B.; Snow, J. A.; Lefer, B. L. (2003) Ozone,
18 aerosol, potential vorticity, and trace gas trends observed at high-latitudes over North America from February
19 to May 2000. J. Geophys. Res. [Atmos.] 108(D4): 10.1029/2001JD001390.
20 Burke, J. M.; Zufall, M. J.; Ozkaynak, H. (2001) A population exposure model for paniculate matter: case study
21 results for PM2 5 in Philadelphia, PA. J. Exposure Anal. Environ. Epidemiol. 11: 470-489.
22 Carroll, R. J.; Chen, R.; George, E. L; Li, T. H.; Newton, H. J.; Schmiediche, H.; Wang, N. (1997a) Ozone exposure
23 and population density in Harris County, Texas. J. Am. Stat. Assoc. 92: 392-404.
24 Carroll, R. J.; Chen, R.; George, E. L; Li, T. H.; Newton, H. J.; Schmiediche, H.; Wang, N. (1997b) Ozone exposure
25 and population density in Harris County, Texas (rejoinder). J. Am. Stat. Assoc. 92: 414-415.
26 Chang, L.-T.; Koutrakis, P.; Catalano, P. J.; Suh, H. H. (2000) Hourly personal exposures to fine particles and
27 gaseous pollutants—results from Baltimore, Maryland. J. Air Waste Manage. Assoc. 50: 1223-1235.
28 Chao, C. Y. (2001) Comparison between indoor and outdoor air contaminant levels in residential buildings from
29 passive sampler study. Build. Environ. 36: 999-1007.
30 Chen, L.-W. A. (2002) Urban fine paniculate matter: chemical composition and possible origins (dissertation).
31 College Park, MD: University of Maryland, Department of Chemical Physics. Available from: University
32 Microfilms, Ann Arbor, MI; AADAA-I3078297.
33 Chinkin, L. R.; Coe, D. L.; Funk, T. H.; Hafner, H. R.; Roberts, P. T.; Ryan, P. A. (2003) Weekday versus weekend
34 activity patterns for ozone precursor emissions in California's South Coast Air Basin. J. Air Waste Manage.
35 Assoc. 53:829-843.
36 Christakos, G. (2000) Modern spatiotemporal geostatistics. New York, NY: Oxford University Press. (International
37 Association for Mathematical Geology; studies in mathematical geology: 6).
3 8 Christakos, G.; Hristopulos, D. T. (1998) Spatiotemporal environmental health modelling. A tractatus stochasticus.
39 Boston, MA: Kluwer Academic Publishers.
40 Christakos, G.; Kolovos, A. (1999) A study of the spatiotemporal health impacts of ozone exposure. J. Exposure
41 Anal. Environ. Epidemiol. 9: 322-335.
42 Christakos, G.; Vyas, V. M. (1998a) A composite space/time approach to studying ozone distribution over eastern
43 United States. Atmos. Environ. 32: 2845-2857.
44 Christakos, G.; Vyas, V. M. (1998b) A novel method for studying population health impacts of spatiotemporal
45 ozone distribution. Soc. Sci. Med. 47: 1051-1066.
46 Clench-Aas, J.; Bartonova, A.; Behler, T.; Grenskei, K. E.; Sivertsen, B.; Larssen, S. (1999) Air pollution exposure
47 monitoring and estimating. Part I. Integrated air quality monitoring system. J. Environ. Monit. 1:313-319.
48 Cleveland, W. S.; Graedel, T. E.; Kleiner, B.; Warner, J. L. (1974) Sunday and workday variations in photochemical
49 air pollutants in New Jersey and New York. Science (Washington, DC) 186: 1037-1038.
5 0 Code of Federal Regulations. (2000a) Appendix I to part 50—interpretation of the 8-hour primary and secondary
51 national ambient air quality standards for ozone. C. F. R. 40: pt. 50, app. I.
52 Code of Federal Regulations. (2000b) Appendix D to part 58—Network design for state and local air monitoring
53 stations (SLAMS), national air monitoring stations (NAMS), and photochemical assessment monitoring
54 stations (PAMS). C. F. R. 40: pt. 58, app. D.
January 2005 AX3-231 DRAFT-DO NOT QUOTE OR CITE
-------
1 Coffey, P.; Stasiuk, W.; Mohnen, V. (1977) Ozone in rural and urban areas of New York State: part I. In:
2 Dimitriades, B., ed. International conference on photochemical oxidant pollution and its control -
3 proceedings: volume I; September 1976; Raleigh, NC. Research Triangle Park, NC: U.S. Environmental
4 Protection Agency, Environmental Sciences Research Laboratory; pp. 89-96; report no. EPA-600/3-77-001a.
5 Available from: NTIS, Springfield, VA; PB-264232.
6 Colome, S. D.; Wilson, A. L.; Tian, Y. (1994) California residential indoor air quality study. Volume 2. Carbon
7 monoxide and air exchange rate: an univariate and multivariate analysis. Chicago, IL: Gas Research Institute;
8 report no. GRI-93/0224.3.
9 Cooper, O. R.; Moody, J. L. (2000) Meteorological controls on ozone at an elevated eastern United States regional
10 background monitoring site. J. Geophys. Res. [Atmos.] 105: 6855-6869.
11 Cressie, N. (1997) Ozone exposure and population density in Harris County, Texas [comment]. J. Am. Stat. Assoc.
12 92:411-413.
13 Cristofanelli, P.; Bonasoni, P.; Collins, W.; Feichter, J.; Forster, C.; James, P.; Kentarchos, A.; Kubik, P. W.; Land,
14 C.; Meloen, J.; Roelofs, G. J.; Siegmund, P.; Sprenger, M.; Schnabel, C.; Stohl, A.; Tobler, L.; Tositti, L.;
15 Trickl, T.; Zanis, P. (2003) Stratosphere-troposphere exchange: a model and method evaluation. J. Geophys.
16 Res. 108(012): 10.1029/2002JD002600.
17 Daum, P. H.; Kleinman, L. I.; Newman, L.; Luke, W. T.; Weinstein-Lloyd, J.; Berkowitz, C. M.; Busness, K. M.
18 (1996) Chemical and physical properties of plumes of anthropogenic pollutants transported over the North
19 Atlantic during the North Atlantic Regional Experiment. J. Geophys. Res. [Atmos.] 101: 29,029-29,042.
20 Delfino, R. J.; Coate, B. D.; Zeiger, R. S.; Seltzer, J. M.; Street, D. H.; Koutrakis, P. (1996) Daily asthma severity in
21 relation to personal ozone exposure and outdoor fungal spores. Am. J. Respir. Crit. CareMed. 154: 633-641.
22 Dickerson, R. R. (1985) Reactive nitrogen compounds in the Arctic. J. Geophys. Res. 90: 10,739-10,743.
23 Dickerson, R. R.; Huffman, G. J.; Luke, W. T.; Nunnermacker, L. J.; Pickering, K. E.; Leslie, A. C. D.; Lindsey,
24 C. G.; Slinn, W. G. N.; Kelly, T. J.; Daum, P. H.; Delany, A. C.; Greenberg, J. P.; Zimmerman, P. R.;
25 Boatman, J. F.; Ray, J. D.; Stedman, D. H. (1987) Thunderstorms: an important mechanism in the transport of
26 air pollutants. Science (Washington, DC) 235: 460-465.
27 Drakou, G.; Zerefos, C.; Ziomas, I. (1995) A preliminary study on the relationship between outdoor and indoor air
28 pollution levels. Fresenius' Environ. Bull. 4: 689-694.
29 Drakou, G.; Zerefos, C.; Ziomas, I.; Voyatzaki, M. (1998) Measurements and numerical simulations of indoor O3
30 and NO^ in two different cases. Atmos. Environ. 32: 595-610.
31 Evans, G.; Finkelstein, P.; Martin, B.; Possiel, N.; Graves, M. (1983) Ozone measurements from a network of
32 remote sites. J. AirPollut. Control Assoc. 33: 291-296.
33 Evans, E. G.; Rhodes, R. C.; Mitchell, W. J.; Puzak, J. C. (1985) Summary of precision and accuracy assessments
34 for the state and local air monitoring networks: 1982. Research Triangle Park, NC: U.S. Environmental
3 5 Protection Agency, Environmental Monitoring Systems Laboratory; report no. EPA-600/4-85-031. Available
36 from: NTIS, Springfield, VA; PB85-208171.
37 Faith, R.; Sheshinski, R. (1979) Misspecification of trend in spatial random-function interpolation with application
38 to oxidant mapping. Philadelphia, PA: Siam Institute for Mathematics and Society; technical report no. 28.
39 Fan, S.-M.; Jacob, D. J.; Mauzerall, D. L.; Bradshaw, J. D.; Sandholm, S. T.; Blake, D. R.; Singh, H. B.; Talbot,
40 R. W.; Gregory, G. L.; Sachse, G. W. (1994) Origin of tropospheric NOX over subarctic eastern Canada in
41 summer. J. Geophys. Res. [Atmos.] 99: 16,867-16,877.
42 Federal Register. (1986) Guidelines for estimating exposures. F. R. (September 24) 51: 34,042-34,054.
43 Fiore, A. M.; Jacob, D. J.; Bey, L; Yantosca, R. M.; Field, B. D.; Fusco, A. C.; Wilkinson, J. G. (2002a) Background
44 ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys.
45 Res. (Atmos.) 107(D15): 10.1029/2001JD000982.
46 Fiore, A. M.; Jacob, D. J.; Field, B. D.; Streets, D. G.; Fernandes, S. D.; Jang, C. (2002b) Linking ozone pollution
47 and climate change: the case for controlling methane. Geophys. Res. Lett. 29(19): 10.1029/2002GL015601.
48 Fiore, A.; Jacob, D. J.; Liu, H.; Yantosca, R. M.; Fairlie, T. D.; Li, Q. (2003a) Variability in surface ozone
49 background over the United States: implications for air quality policy. J. Geophys. Res. (Atmos.)
50 108(D24): 10.1029/2003 JD003855.
51 Fiore, A. M.; Jacob, D. J.; Mathur, R.; Martin, R. V. (2003b) Application of empirical orthogonal functions to
52 evaluate ozone simulations with regional and global models. J. Geophys. Res. (Atmos.)
53 108(D14): 10.1029/2002JD003151.
54 Fowler, D.; Cape, J. N. (1982) Air pollutants in agriculture and horticulture. In: Unsworth, M. H.; Ormrod, D. P.
5 5 Effects of gaseous air pollution in agriculture and horticulture. London, United Kingdom: Butterworth
56 Scientific; pp. 3-26.
January 2005 AX3-232 DRAFT-DO NOT QUOTE OR CITE
-------
1 Fraigneau, Y. C.; Gonzalez, M; Coppalle, A. (1995) Dispersion and chemical reaction of a pollutant near a
2 motorway. Sci. Total Environ. 169: 83-91.
3 Freijer, J. I.; Bloemen, H. J. T. (2000) Modeling relationships between indoor and outdoor air quality. J. Air Waste
4 Manage. Assoc. 50: 292-300.
5 Freijer, J. I.; Bloemen, H. J. T.; de Loos, S.; Marra, M.; Rombout, P. J. A.; Steentjes, G. M.; Van Veen, M. P. (1998)
6 Modelling exposure of the Dutch population to air pollution. J. Haz. Mat. 61: 107-114.
7 Fujita, E. M.; Stockwell, W. R.; Campbell, D. E.; Keislar, R. E.; Lawson, D. R. (2003) Evolution of the magnitude
8 and spatial extent of the weekend ozone effect in California's South Coast Air Basin, 1981-2000. J. Air Waste
9 Manage. Assoc. 53:802-815.
10 Fusco, A. C.; Logan, J. A. (2003) Analysis of 1970-1995 trends in tropospheric ozone at Northern Hemisphere
11 midlatitudes with the GEOS-CHEM model. J. Geophys. Res. (Atmos.) 108: 10.1029/2002JD002742.
12 Georgopoulos, P. G.; Lioy, P. J. (1994) Conceptual and theoretical aspects of human exposure and dose assessment.
13 J. Exposure Anal. Environ. Epidemiol. 4: 253-285.
14 Georgopoulos, P. G.; Arunachalam, S.; Wang, S. (1997a) Alternative metrics for assessing the relative effectiveness
15 of NOx and VOC emission reductions in controlling ground-level ozone. J. Air Waste Manage. Assoc.
16 47:838-850.
17 Georgopoulos, P. G.; Walia, A.; Roy, A.; Lioy, P. J. (1997b) Integrated exposure and dose modeling and analysis
18 system. 1. Formulation and testing of microenvironmental and pharmacokinetic components. Environ. Sci.
19 Technol. 31: 17-27.
20 Georgopoulos, P.; Yonone-Lioy, M. J.; Opeikum, R. E.; Lioy, P. (2002a) Environmental dynamics and human
21 exposure to copper. Volume I: environmental dynamics and human exposure issues. New York, NY:
22 International Copper Association.
23 Georgopoulos, P.; Wang, S. W.; Vyas, V. M.; Lioy, P. J.; Tan, H. C. (2002b) Environmental dynamics and human
24 exposure to copper. Volume 2: framework and data sources for assessing human exposure to copper in the
25 United States. New York, NY: International Copper Association.
26 Gettelman, A.; Holton, J. R.; Rosenlof, K. H. (1997) Mass fluxes of O3, CH4, N2O, and CF2C12 in the lower
27 stratosphere calculated from observational data. J. Geophys. Res. [Atmos.] 102: 19,149-19,159.
28 Geyh, A. S.; Wolfson, J. M.; Koutrakis, P.; Mulik, J. D.; Avol, E. L. (1997) Development and evaluation of a small
29 active ozone sampler. Environ. Sci. Technol. 31: 2326-2330.
30 Geyh, A. S.; Roberts, P. T.; Lurmann, F. W.; Schoell, B. M.; Avol, E. L. (1999) Initial field evaluation of the
31 Harvard active ozone sampler for personal ozone monitoring. J. Exposure Anal. Environ. Epidemiol.
32 9: 143-149.
33 Geyh, A. S.; Xue, J.; Ozkaynak, H.; Spengler, J. D. (2000) The Harvard Southern California chronic ozone exposure
34 study: assessing ozone exposure of grade-school-age children in two southern California communities.
35 Environ. Health Perspect. 108: 265-270.
3 6 Glen, G.; Lakkadi, Y.; Tippett, J. A.; del Valle-Torres, M. (1997) Development of NERL/CHAD: the National
37 Exposure Research Laboratory consolidated human activity database. Research Triangle Park, NC: U.S.
3 8 Environmental Protection Agency, Office of Research and Development; contract no. 68-D5-0049.
39 Gold, D. R.; Allen, G.; Damokosh, A.; Serrano, P.; Hayes, C.; Castillejos, M. (1996) Comparison of outdoor and
40 classroom ozone exposures for school children in Mexico City. J. Air Waste Manage. Assoc. 46: 335-342.
41 Goldan, P. D.; Trainer, M.; Kuster, W. C.; Parrish, D. D.; Carpenter, J.; Roberts, J. M.; Yee, J. E.; Fehsenfeld, F. C.
42 (1995) Measurements of hydrocarbons, oxygenated hydrocarbons, carbon monoxide, and nitrogen oxides in
43 an urban basin in Colorado: implications for emission inventories. J. Geophys. Res. [Atmos.]
44 100:22,771-22,783.
45 Goldstein, B. R.; Tardiff, G.; Hoffnagle, G.; Kester, R. (1992) Valdez health study: summary report. Anchorage,
46 AK: Alyeska Pipeline Service Company.
47 Grivet, C. D. (1980) Modeling and analysis of air quality data [dissertation]. Palo Alto, CA: Stanford University.
48 Available from: University Microfilms International; Ann Arbor, MI; publication no. 80-16,824.
49 Grosjean, E.; Grosjean, D. (1996) Carbonyl products of the gas phase reaction of ozone with symmetrical alkenes.
50 Environ. Sci. Technol. 30: 2036-2044.
51 Grosjean, E.; Grosjean, D. (1998) The gas-phase reaction of alkenes with ozone: formation yields of carbonyls from
52 biradicals in ozone-alkene-cyclohexane experiments. Atmos. Environ. 32: 3393-3402.
53 Grosjean, E.; Grosjean, D.; Woodhouse, L. F. (2001) Peroxyacetyl nitrate and peroxypropionyl nitrate during SCOS
54 97-NARSTO. Environ. Sci. Technol. 35: 4007-4014.
55 Guttorp, P.; Meiring, W.; Sampson, P. D. (1997) Ozone exposure and population density in Harris County, Texas
56 [comment]. J. Am. Stat. Assoc. 92: 405-408.
January 2005 AX3-233 DRAFT-DO NOT QUOTE OR CITE
-------
1 Hartwell, T. D.; Clayton, C. A.; Ritchie, R. M; Whitmore, R. W.; Zelon, H. S.; Jones, S. M.; Whitehurst, D. A.
2 (1984) Study of carbon monoxide exposure of residents of Washington, DC and Denver, Colorado. Research
3 Triangle Park, NC: U.S. Environmental Protection Agency, Office of Research and Development,
4 Environmental Monitoring Systems Laboratory; report no. EPA-600/4-84-031.
5 Hayes, S. R. (1989) Estimating the effect of being indoors on total personal exposure to outdoor air pollution.
6 JAPCA 39: 1453-1461.
7 Hayes, S. R. (1991) Use of an indoor air quality model (IAQM) to estimate indoor ozone levels. J. Air Waste
8 Manage. Assoc. 41: 161-170.
9 Hazucha, M. J.; Folinsbee, L. J.; Seal, E., Jr. (1992) Effects of steady-state and variable ozone concentration profiles
10 on pulmonary function. Am. Rev. Respir. Dis. 146: 1487-1493.
11 Heck, W. W.; Taylor, O. C.; Adams, R.; Bingham, G.; Miller, J.; Preston, E.; Weinstein, L. (1982) Assessment of
12 crop loss from ozone. J. Air Pollut. Control Assoc. 32: 353-361.
13 Helaleh, M. I. H.; Ngudiwaluyo, S.; Korenaga, T.; Tanaka, K. (2002) Development of passive sampler technique for
14 ozone monitoring. Estimation of indoor and outdoor ozone concentration. Talanta 58: 649-659.
15 Herring, J. A.; Jaffe, D. A.; Beine, H. J.; Madronich, S.; Blake, D. R. (1997) High-latitude springtime
16 photochemistry. Part II: sensitivity studies of ozone production. J. Atmos. Chem. 27: 155-178.
17 Heuss, J. M.; Kahlbaum, D. F.; Wolff, G. T. (2003) Weekday/weekend ozone differences: what can we learn from
18 them? J. Air Waste Manage. Assoc. 53: 772-788.
19 Holland, D. M.; Principe, P. P.; Vorburger, L. (1999) Rural ozone: trends and exceedances at CASTNet sites.
20 Environ. Sci. Technol. 33:43-48.
21 Holton, J. R.; Haynes, P. H.; Mclntyre, M. E.; Douglass, A. R.; Rood, R. B.; Pfister, L. (1995)
22 Stratosphere-troposphere exchange. Rev. Geophys. 33: 403-439.
23 Horowitz, L. W.; Walters, S.; Mauserall, D. L.; Emmons, L. K.; Rasch, P. J.; Granier, C.; Tie, X.; Lamarque, J.-F.;
24 Schultz, M. G.; Tyndall, G. S.; Orlando, J. J.; Brasseur, G. P. (2003) A global simulation of tropospheric
25 ozone and related tracers: description and evaluation of MOZART, version 2. J. Geophys. Res. [Atmos.]
26 108(D24): 10.1029/2002JD002853.
27 Howard-Reed, C.; Wallace, L. A.; Ott, W. R. (2002) The effect of opening windows on air change rates in two
28 homes. J. Air Waste Manage Assoc. 52: 147-159.
29 ICF Consulting; ManTech Environmental Technology, Inc. (2003) Total Risk Integrated Methodology
30 TRIM.Expo(Inhalation) user's document. Volume I: air pollutants exposure model (APEX, version 3) user's
31 guide. Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning
32 and Standards. Available: http://www.epa.gov/ttn/fera/data/apex322/apexusersguidevoli4-24-03.pdf
33 [7 September, 2004].
34 Intergovernmental Panel on Climate Change (IPCC). (2001) Climate change 2001: the scientific basis. Contribution
35 of working group I to the third assessment report of the Intergovernmental Panel on Climate Change.
36 Cambridge, United Kingdom: Cambridge University Press.
37 Isaksen, I. S. A.; Hov, O.; Hesstvedt, E. (1978) Ozone generation over rural areas. Environ. Sci. Technol.
38 12: 1279-1284.
39 Isukapalli, S. S.; Purushothaman, V.; Georgopoulos, P. G. (1999) Mechanistic modeling of the interrelationships
40 between indoor/outdoor air quality and human exposure in a GIS framework. Presented at: 92nd Air & Waste
41 Management Association Annual Meeting; June; St. Louis, MO. Pittsburgh, PA: Air & Waste Management
42 Association.
43 Jacob, D. L.; Logan, J. A.; Murti, P. P. (1999) Effect of rising Asian emissions on surface ozone in the United States.
44 Geophys. Res. Lett. 26: 2175-2178.
45 Jaffe, D. A. (1993) The relationship between anthropogenic nitrogen oxides and ozone trends in the Arctic
46 troposphere. In: Niki, H.; Becker, K. H., eds. The tropospheric chemistry of ozone in the polar regions.
47 New York, NY: Springer-Verlag. (NATO ASI series I: Global environmental change, vol. 7).
48 Jaffe, D. A.; Honrath, R. E.; Herring, J. A.; Li, S.-M.; Kahl, J. D. (1991) Measurements of nitrogen oxides at
49 Barrow, Alaska during spring: evidence for regional and northern hemispheric sources of pollution.
50 J. Geophys. Res. [Atmos.] 96: 7395-7405.
51 Jaffe, D.; Price, H.; Parrish, D.; Goldstein, A.; Harris, J. (2003) Increasing background ozone during spring on the
52 west coast of North America. Geophys. Res. Lett. 30: 10.1029/2003GL017024.
53 Jakobi, G.; Fabian, P. (1997) Indoor/outdoor concentrations of ozone and peroxyacetyl nitrate (PAN). Int. J.
54 Biometeorol. 40: 162-165.
55 James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003a) A 15-year climatology of
56 stratosphere-troposphere exchange with a Lagrangian particle dispersion model: 1. Methodology and
57 validation. J. Geophys. Res. 108(D12): 10.1029/2002JD002637.
January 2005 AX3-234 DRAFT-DO NOT QUOTE OR CITE
-------
1 James, P.; Stohl, A.; Forster, C.; Eckhardt, S.; Seibert, P.; Frank, A. (2003b) A 15-year climatology of
2 stratosphere-troposphere exchange with a Lagrangian particle dispersion model. 2. Mean climate and seasonal
3 variability. J. Geophys. Res. (Atmos.) 108(D12): 10.1029/2002JD002639.
4 Johnson, T. (1984) A study of personal exposure to carbon monoxide in Denver, Colorado. Research Triangle Park,
5 NC: U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory; report no.
6 EPA-600/4-84-014.
7 Johnson, T. A. (1989) Human activity patterns in Cincinnati, Ohio. Palo Alto, CA: Electric Power Research
8 Institute; report no. EPRIEN-6204.
9 Johnson, T. (1997) A pilot study in Los Angeles to measure personal ozone exposures during scripted activities.
10 Washington, DC: American Petroleum Institute, Health and Environmental Sciences Department; API
11 publication no. DR 218.
12 Johnson, T. (2001) A guide to selected algorithms, distributions, and databases used in exposure models developed
13 by the Office of Air Quality Planning and Standards (DRAFT). Research Triangle Park, NC: U.S.
14 Environmental Protection Agency, CERM report TRO1.
15 Johnson, T. (2003) A guide to selected algorithms, distributions, and databases used in exposure models developed
16 by the Office of Air Quality Planning and Standards. Research Triangle Park, NC: U.S. Environmental
17 Protection Agency, Office of Research and Development; EPA grant no. CR827033. Available:
18 http://www.epa.gov/ttn/fera/data/human/report052202.pdf [9 April, 2004].
19 Johnson, T.; Long, T. (2004) Determining the frequency of open windows in residences: a pilot study in Durham,
20 North Carolina during varying temperature conditions. J. Exposure Anal. Environ. Epidemiol.:
21 10.1038/sj.jea.7500409.
22 Johnson, T. R.; Paul, R. A.; Capel, J. E.; McCurdy, T. (1990) Estimation of ozone exposure in Houston using a
23 probabilistic version of NEM. Presented at: 83rd annual meeting and exhibition of the Air & Waste
24 Management Association; June; Pittsburgh, PA. Pittsburgh, PA: Air & Waste Management Association;
25 paper no. 90-150.1.
26 Johnson, T. R.; Wijnberg, L.; Capel, J. E.; Vostal, J. J. (1992) The use of activity diary data to estimate the
27 probability of exposure to air pollution. In: Berglund, R. L., ed. Tropospheric ozone and the environment II:
28 effects, modeling and control [papers from an international specialty conference]; November; Atlanta, GA.
29 Pittsburgh, PA: Air & Waste Management Association; pp. 713-724. (A&WMA transactions series no. 20).
30 Johnson, T.; Capel, J.; McCoy, M. (1996a) Estimation of ozone exposures experienced by urban residents using a
31 probabilistic version of NEM and 1990 population data. Research Triangle Park, NC: U.S. Environmental
32 Protection Agency, Office of Air Quality Planning and Standards; contract no. 68-DO-0062.
33 Johnson, T.; Capel, J.; McCoy, M.; Mozier, J. W. (1996b) Estimation of ozone exposures experienced by outdoor
34 workers in nine urban areas using a probabilistic version of NEM. Research Triangle Park, NC: U.S.
3 5 Environmental Protection Agency, Office of Air Quality Planning and Standards; contract no. 63-D-30094,
36 work assignment nos. 0-1 and 1-4.
37 Johnson, T.; Capel, J.; Mozier, J.; McCoy, M. (1996c) Estimation of ozone exposures experienced by outdoor
3 8 children in nine urban areas using a probabilistic version of NEM. Research Triangle Park, NC: U.S.
39 Environmental Protection Agency, Office of Air Quality Planning and Standards; contract no. 63-D-30094.
40 Johnson, T.; Mozier, J.; Capel, J. (1997) Supplement to "Estimation of ozone exposures experienced by outdoor
41 children in nine urban areas using a probabilistic version of NEM (April 1996)". Research Triangle Park, NC:
42 U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
43 Johnson, T.; Myers, J.; Kelly, T.; Wisbith, A.; Ollison, W. (2004) A pilot study using scripted ventilation conditions
44 to identify key factors affecting indoor pollutant concentrations and air exchange rate in a residence. J.
45 Exposure Anal. Environ. Epidemiol. 14: 1-22.
46 Junge, C. E. (1963) Air chemistry and radioactivity. New York, NY: Academic Press. (Van Mieghem, J.; Hales,
47 A. L., eds. International geophysics series: v. 4).
48 Kelly, N. A.; Wolff, G. T.; Ferman, M. A. (1984) Sources and sinks of ozone in rural areas. Atmos. Environ.
49 18: 1251-1266.
50 Klene, J. G.; Clausen, P. A.; Weschler, C. J.; Wolkoff, P. (2001) Determination of ozone removal rates by selected
51 building products using the FLEC emission cell. Environ. Sci. Technol. 35: 2548-2553.
52 Klepeis, N. E.; Tsang, A. M.; Behar, J. V. (1996) Analysis of the national human activity pattern survey (NHAPS)
53 respondents from a standpoint of exposure assessment. Washington, DC: U.S. Environmental Protection
54 Agency, Office of Research and Development; report no. EPA/600/R-96/074.
55 Klepeis, N. E.; Nelson, W. C.; Ott. W. R.; Robinson, J. P. Tsang, A. M.; Switzer, P.; Behar, J. V.; Hern, S. C.;
56 Engelmann, W. H. (2001) The National Human Activity Pattern Survey (NHAPS): a resource for assessing
57 exposure to environmental pollutants. J. Exposure Anal. Environ. Epidemiol. 11: 231-252.
January 2005 AX3-235 DRAFT-DO NOT QUOTE OR CITE
-------
1 Kley, D.; Volz, A.; Mulheims, F. (1988) Ozone measurements in historic perspective. In: Isaksen, I. S. A., ed.
2 Tropospheric ozone - regional and global scale interactions: proceedings of the NATO advanced workshop on
3 regional and global ozone interaction and its environmental consequences; June 1987; Lillehammer, Norway.
4 Dordrecht, The Netherlands: D. Reidel Publishing Company; pp. 63-72. [NATO Advanced Science Institutes
5 (ASI) studies: series C, mathematical and physical sciences v. 227].
6 Knudsen, H. P.; Lefohn, A. S. (1988) The use of geostatistics to characterize regional ozone exposures. In: Heck,
7 W. W.; Taylor, O. C.; Tingey, D. T., eds. Assessment of crop loss from air pollutants: proceedings of an
8 international conference; October 1987; Raleigh, NC. Essex, United Kingdom: Elsevier Science Publishers,
9 Ltd.; pp. 91-105.
10 Korc, M. E. (1996) A socioeconomic assessment of human exposure to ozone in the South Coast Air Basin of
11 California. J. Air Waste Manage. Assoc. 46: 547-557.
12 Koutrakis, P.; Sioutas, C.; Ferguson, S. T.; Wolfson, J. M.; Mulik, J. D.; Burton, R. M. (1993) Development and
13 evaluation of a glass honeycomb denuder/filter pack system to collect atmospheric gases and particles.
14 Environ. Sci. Technol. 27: 2497-2501.
15 Krzyzanowski, M. (1997) Methods for assessing the extent of exposure and effects of air pollution. Occup. Environ.
16 Med. 54: 145-151.
17 Ktinzli, N.; Lurmann, F.; Segal, M.; Ngo, L.; Balmes, J.; Tager, I. B. (1996) Reliability of lifetime residential history
18 and activity measures as elements of cumulative ambient ozone exposure assessment. J. Exposure Anal.
19 Environ. Epidemiol. 6: 289-310.
20 Ktinzli, N.; Kelly, T.; Balmes, J.; Tager, I. B. (1997) Reproducibility of retrospective assessment of outdoor
21 time-activity patterns as an individual determinant of long-term ambient ozone exposure. Int. J. Epidemiol.
22 26: 1258-1271.
23 Lagus Applied Technology, Inc. (1995) Air change rates in non-residential buildings in California. Sacramento, CA:
24 California Energy Commission; contract no. 400-91-034; July.
25 Lavery, T. F.; Rogers, C. M.; Howell, H. K.; Burnett, M. C.; Wanta, C. A. (2002) Clean air status trends network
26 (CASTNet) 2000 annual report. Available: http://www.epa.gov/castnet/library/annualOO.html (9 September
27 2003).
28 Lebron, F. (1975) A comparison of weekend-weekday ozone and hydrocarbon concentrations in the
29 Baltimore-Washington metropolitan area. Atmos. Environ. 9: 861-863.
30 Lee, K.; Vallarino, J.; Dumyahn, T. Ozkaynak, H.; Spengler, J. (1999) Ozone decay rates in residences. J. Air Waste
31 Manage. Assoc. 49: 1238-1244.
32 Lee, K.; Xue, J.; Geyh, A. S.; Ozkaynak, H.; Leaderer, B. P.; Weschler, C. J.; Spengler, J. D. (2002) Nitrous acid,
33 nitrogen dioxide, and ozone concentrations in residential environments. Environ. Health Perspect.
34 110: 145-150.
3 5 Lefohn, A. S. (1984) A comparison of ambient ozone exposures for selected nonurban sites. Presented at: 77th
36 annual meeting of the Air Pollution Control Association; June; San Francisco, CA. Pittsburgh, PA: Air
37 Pollution Control Association; paper no. 84-104.1.
38 Lefohn, A. S. (1992) The characterization of ambient ozone exposures. In: Lefohn, A. S., ed. Surface level ozone
39 exposures and their effects on vegetation. Chelsea, MI: Lewis Publishers, Inc.; pp. 31-92.
40 Lefohn, A. S.; Benedict, H. M. (1983) The potential for the interaction of acidic precipitation and ozone pollutant
41 doses affecting agricultural crops. Presented at: 76th annual meeting of the Air Pollution Control Association;
42 June; Atlanta, GA. Pittsburgh, PA: Air Pollution Control Association; paper no. 83-2.2.
43 Lefohn, A. S.; Jones, C. K. (1986) The characterization of ozone and sulfur dioxide air quality data for assessing
44 possible vegetation effects. J. AirPollut. Control Assoc. 36: 1123-1129.
45 Lefohn, A. S.; Mohnen, V. A. (1986) The characterization of ozone, sulfur dioxide, and nitrogen dioxide for selected
46 monitoring sites in the Federal Republic of Germany. J. Air Pollut. Control Assoc. 36: 1329-1337.
47 Lefohn, A. S.; Runeckles, V. C. (1987) Establishing standards to protect vegetation - ozone exposure/dose
48 considerations. Atmos. Environ. 21: 561-568.
49 Lefohn, A. S.; Shadwick, D. S. (1991) Ozone, sulfur dioxide, and nitrogen dioxide trends at rural sites located in the
50 United States. Atmos. Environ. Part A 25: 491-501.
51 Lefohn, A. S.; Tingey, D. T. (1984) The co-occurrence of potentially phytotoxic concentrations of various gaseous
52 air pollutants. Atmos. Environ. 18: 2521-2526.
53 Lefohn, A. S.; Davis, C. E.; Jones, C. K.; Tingey, D. T.; Hogsett, W. E. (1987) Co-occurrence patterns of gaseous air
54 pollutant pairs at different minimum concentrations in the United States. Atmos. Environ. 21: 2435-2444.
55 Lefohn, A. S.; Knudsen, H. P.; McEvoy, L. R., Jr. (1988) The use of kriging to estimate monthly ozone exposure
56 parameters for the southeastern United States. Environ. Pollut. 53: 27-42.
January 2005 AX3-236 DRAFT-DO NOT QUOTE OR CITE
-------
1 Lefohn, A. S.; Krapa, S. V.; Winstanley, D. (1990a) Surface ozone exposures measured at clean locations around the
2 world. Environ. Pollut. 63: 189-224.
3 Lefohn, A. S.; Shadwick, D. S.; Mohnen, V. A. (1990b) The characterization of ozone concentrations at a select set
4 of high-elevation sites in the eastern United States. Environ. Pollut. 67: 147-178.
5 Lefohn, A. S.; Benkovitz, C. M; Tanner, R. L.; Smith, L. A.; Shadwick, D. S. (1991) Air quality measurements and
6 characterizations for terrestrial effects research. In: Irving, P. M., ed. Acidic deposition: state of science and
7 technology, volume I, emissions, atmospheric processes, and deposition. Washington, DC: The U.S. National
8 Acid Precipitation Assessment Program. (State of science and technology report no. 7).
9 Lefohn, A. S.; Shadwick, D. S.; Ziman, S. D. (1998) The difficult challenge of attaining EPA's new ozone standard.
10 Environ. Sci. Technol. 32: 276A-282A.
11 Lefohn, A. S.; Oltmans, S. I; Dann, T.; Singh, H. B. (2001) Present-day variability of background ozone in the
12 lower troposphere. J. Geophys. Res. [Atmos.] 106: 9945-9958.
13 Lelieveld, J.; Dentener, F. J. (2000) What controls tropospheric ozone? J. Geophys. Res. [Atmos.] 105: 3531-3551.
14 Leovic, K. W.; Sheldon, L. S.; Whitaker, D. A.; Hetes, R. G.; Calcagni, J. A.; Baskir, J. N. (1996) Measurement of
15 indoor air emissions from dry-process photocopy machines. J. Air Waste Manage. Assoc. 46: 821-829.
16 Leovic, K. W.; Whitaker, D.; Brockmann, C.; Bayer, C. W.; Sojka, P. E. (1998) Evaluation of low-emitting products
17 for use in the indoor environment. Presented at: 91st annual meeting of the Air & Waste Management
18 Association; June; San Diego, CA. Pittsburgh, PA: Air & Waste Management Association; paper no.
19 98-TP48.02 (A292).
20 Li, Q.; Jacob, D. J.; Logan, J. A.; Bey, I.; Yantosca, R. M.; Liu, H.; Martin, R. V.; Fiore, A. M.; Field, B. D.;
21 Duncan,B.N. (2001) A tropospheric ozone maximum over the Middle East. Geophys. Res. Lett.
22 28: 3235-3238.
23 Li, Q.; Jacob, D. J.; Fairlie, T. D.; Liu, H.; Martin, R. V.; Yantosca, R. M. (2002) Stratospheric versus pollution
24 influences on ozone at Bermuda: reconciling past analyses. J. Geophys. Res. 107(D22):
25 10.1029/2002JD002138.
26 Lin, C.-Y.; Jacob, D. J.; Munger, J. W.; Fiore, A. M. (2000) Increasing background ozone in surface air over the
27 United States. Geophys. Res. Lett. 27: 3465-3468.
28 Lin, C.-Y. C.; Jacob, D. J.; Fiore, A. M. (2001) Trends in exceedances of the ozone air quality standard in the
29 continental United States, 1980-1998. Atmos. Environ. 35: 3217-3228.
30 Linn, W. S.; Shamoo, D. A.; Hackney, J. D. (1992) Documentation of activity patterns in 'high-risk' groups exposed
31 to ozone in the Los Angeles area. In: Tropospheric ozone and the environment II: effects, modeling and
32 control: papers from an Air & Waste Management Association international specialty conference; November,
33 1991; Atlanta, GA. Pittsburgh, PA: Air & Waste Management Association; pp. 701-712. (A&WMA
34 publication TR-20).
35 Linn, W. S.; Spier, C. E.; Hackney, J. D. (1993) Activity patterns in ozone-exposed construction workers. J. Occup.
36 Med. Toxicol. 2: 1-14.
37 Linn, W. S.; Shamoo, D. A.; Anderson, K. R.; Peng, R.-C.; Avol, E. L.; Hackney, J. D.; Gong, H., Jr. (1996)
3 8 Short-term air pollution exposures and responses in Los Angeles area schoolchildren. J. Exposure Anal.
39 Environ. Epidemiol. 6: 449-472.
40 Lioy, P. J. (1990) Assessing total human exposure to contaminants: a multidisciplinary approach. Environ. Sci.
41 Technol. 24: 938-945.
42 Lisac, I.; Grubisic, V. (1991) An analysis of surface ozone data measured at the end of the 19th century in Zagreb,
43 Yugoslavia. Atmos. Environ. Part A 25: 481-486.
44 Liu, D.-L.; Nazaroff, W. W. (2001) Modeling pollutant penetration across building envelopes. Atmos. Environ.
45 35:4451-4462.
46 Liu, L.-J. S.; Koutrakis, P.; Leech, J.; Broder, I. (1995) Assessment of ozone exposures in the greater metropolitan
47 Toronto area. J. Air Waste Manage. Assoc. 45: 223-234.
48 Liu, L.-J. S.; Delfino, R.; Koutrakis, P. (1997) Ozone exposure assessment in a southern California community.
49 Environ. HealthPerspect. 105: 58-65.
50 Liu, H.; Jacob, D. J.; Chan, L. Y.; Oltmans, S. J.; Bey, I.; Yantosca, R. M.; Harris, J. M.; Duncan, B. N.; Martin,
51 R. V. (2002) Sources of tropospheric ozone along the Asian Pacific Rim: an analysis of ozonesonde
52 observations. J. Geophys. Res. [Atmos.] 107(D21): 10.1029/2001JD002005.
53 Liu, H.; Jacob, D. J.; Bey, I.; Yantosca, R. M.; Duncan, B. N.; Sachse, G. W. (2003) Transport pathways for Asian
54 pollution outflow over the Pacific: interannual and seasonal variations. J. Geophys. Res. (Atmos.)
55 108(020): 10.1029/2002JD003102.
56 Liu, H.; Jacob, D. J.; Dibb, J. E.; Fiore, A. M.; Yantosca, R. M. (2004) Constraints on the sources of tropospheric
57 ozone from 210Pb-7Be-O3 correlations. J. Geophys. Res. (Atmos.) 109(D7): 10.1029/2003JD003988.
January 2005 AX3-237 DRAFT-DO NOT QUOTE OR CITE
-------
1 Logan, J. A. (1989) Ozone in rural areas of the United States. J. Geophys. Res. [Atmos.] 94: 8511-8532.
2 Logan, J. A. (1999) An analysis of ozonesonde data for the troposphere: recommendations for testing 3-D models
3 and development of a gridded climatology for tropospheric ozone. J. Geophys. Res. 104: 16,115-16,149.
4 Ludwig, F. L.; Reiter, E.; Shelar, E.; Johnson, W. B. (1977) The relation of oxidant levels to precursor emissions and
5 meteorological features: v. I, analysis and findings. Research Triangle Park, NC: U.S. Environmental
6 Protection Agency, Office of Air Quality Planning and Standards; report no. EPA-450/3-77-022a. Available
7 from: NTIS, Springfield, VA; PB-275 001.
8 Marr, L. C.; Morrison, G. C.; Nazaroff, W. W.; Harley, R. A. (1998) Reducing the risk of accidental death due to
9 vehicle-related carbon monoxide poisoning. J. Air Waste Manage. Assoc. 48: 899-906.
10 Martin, R. V.; Jacob, D. J.; Yantosca, R. M.; Chin, M.; Ginoux, P. (2003) Global and regional decreases in
11 tropospheric oxidants from photochemical effects of aerosols. J. Geophys. Res. [Atmos.] 108(D3):
12 10.1029/2002JD002622.
13 Matheron, G. (1963) Principles of geostatistics. Econ. Geol. 58: 1246-1266.
14 McCurdy, T. (1988) Relationships among ozone air quality indicators in urban areas. In: Wolff, G. T.; Hanisch,
15 J. L.; Schere, K., eds. The scientific and technical issues facing post-1987 ozone control strategies:
16 transactions of an APCA international specialty conference; November 1987; Hartford, CT. Pittsburgh, PA:
17 Air & Waste Management Association; pp. 331-342. (APCA transactions series: no. 12).
18 McCurdy, T. R. (1994) Human exposure to ambient ozone. In: McKee, D. J., ed. Tropospheric ozone: human health
19 and agricultural impacts. Boca Raton, FL: Lewis Publishers; pp. 85-127.
20 McCurdy, T. (1997a) Human activities that may lead to high inhaled intake doses in children aged 6-13. Environ.
21 Toxicol. Pharmacol. 4: 251-260.
22 McCurdy, T. (1997b) Modeling the dose profile in human exposure assessments: ozone as an example. Rev.
23 Toxicol. 1: 3-23.
24 McCurdy, T. (2000) Conceptual basis for multi-route intake dose modeling using an energy expenditure approach.
25 J. Exposure Anal. Environ. Epidemiol. 10: 86-97.
26 McCurdy, T.; Glen, G.; Smith, L.; Lakkadi, Y. (2000) The National Exposure Research Laboratory's Consolidated
27 Human Activity Database. J. Exposure Anal. Environ. Epidemiol. 10: 566-578.
28 Meagher, J. F.; Lee, N. T.; Valente, R. J.; Parkhurst, W. J. (1987) Rural ozone in the southeastern United States.
29 Atmos. Environ. 21: 605-615.
30 Meloen, J.; Siegmund, P.; Van Velthoven, P.; Kelder, H.; Sprenger, M.; Wernli, H.; Kentarchos, A.; Roelofs, G.;
31 Feichter, J.; Land, C.; Forster, C.; James, P.; Stohl, A.; Collins, W.; Cristofanelli, P. (2003)
32 Stratosphere-troposphere exchange: a model and method intercomparison. J. Geophys. Res.
33 108(012): 10.1029/2002JD002274.
34 Mohnen, V. A. (1989) Mountain cloud chemistry project: wet, dry and cloud water deposition. Research Triangle
3 5 Park, NC: U.S. Environmental Protection Agency, Atmospheric Research and Exposure Assessment
36 Laboratory; report no. EPA/600/3-89/009. Available from: NTIS, Springfield, VA; PB89-148597.
37 Mohnen, V A.; Hogan, A.; Coffey, P. (1977) Ozone measurements in rural areas. J. Geophys. Res. 82: 5889-5895.
38 Monks, P. S. (2000) A review of the observations and origins of the spring ozone maximum. Atmos. Environ.
39 34: 3545-3561.
40 Morrison, G. C.; Nazaroff, W. W. (2000) The rate of ozone uptake on carpets: experimental studies. Environ. Sci.
41 Technol. 34: 4963-4968.
42 Morrison, G. C.; Nazaroff, W. W. (2002) The rate of ozone uptake on carpet: mathematical modeling. Atmos.
43 Environ. 36: 1749-1756.
44 Morrison, G. C.; Nazaroff, W. W.; Cano-Ruiz, J. A.; Hodgson, A. T.; Modera, M. P. (1998) Indoor air quality
45 impacts of ventilation ducts: ozone removal and emissions of volatile organic compounds. J. Air Waste
46 Manage. Assoc. 48: 941-952.
47 Mueller, F. X.; Loeb, L.; Mapes, W. H. (1973) Decomposition rates of ozone in living areas. Environ. Sci. Technol.
48 7:342-346.
49 Murphy, D. M.; Fahey, D. W. (1994) An estimate of the flux of stratospheric reactive nitrogen and ozone into the
50 troposphere. J. Geophys. Res. (Atmos.) 99: 5325-5332.
51 Murray, D. M.; Burmaster, D. E. (1995) Residential air exchange rates in the United States: empirical and estimated
52 parametric distributions by season and climatic region. Risk Anal. 15: 459-465.
53 Nagda, N. L.; Rector, H. E.; Koontz, M. D. (1987) Guidelines for monitoring indoor air quality. Washington, DC:
54 Hemisphere Publishing Corporation.
55 National Acid Precipitation Assessment Program. (1987) Interim assessment: the causes and effects of acidic
56 deposition, volume I, executive summary. Washington, DC: Office of the Director of Research.
January 2005 AX3-238 DRAFT-DO NOT QUOTE OR CITE
-------
1 Nazaroff, W. W.; Cass, G. R. (1986) Mathematical modeling of chemically reactive pollutants in indoor air.
2 Environ. Sci. Technol. 20: 924-934.
3 Newchurch, M. I; Ayoub, M. A.; Oltmans, S.; Johnson, B.; Schmidlin, F. J. (2003) Vertical distribution of ozone at
4 four sites in the United States. J. Geophys. Res. 108(D1): 10.1029/2002JD002059.
5 Niu, J.; Tung, T. C. W.; Burnett, J. (2001) Ozone emission rate testing and ranking method using environmental
6 chamber. Atmos. Environ. 35: 2143-2151.
7 Northeast States for Coordinated Air Use Management (NESCAUM). (2002) Indoor/outdoor school air monitoring
8 project. Boston, MA. Available: http://www.nescaum.org/pdf/schoolmonitoring.pdf [29 October, 2003].
9 O'Gara, P. J. (1922) Sulfur dioxide and fume problems and their solution, (presented at the 14th semiannual meeting
10 of the American Institute of Chemical Engineers, Niagara Falls, Canada, June 19-22, 1922). Presented at:
11 14th semiannual meeting of the American Institute of Chemical Engineers; June; Niagara Falls, Canada.
12 Proceedings and papers of the meeting summarized in J. Ind. Eng. Chem. 14: 744-745 by J. C. Olsen.
13 Odman, M. T.; Ingram, C. I. (1996) Multiscale air quality simulation platform (MAQSIP): source code
14 documentation and validation. Research Triangle Park, NC: MCNC; technical report ENV-96TR002-vl.O.
15 Available: http://www.ce.gatech.edu/~todman/maqsip.pdf (9 September 2003).
16 Olsen, M. A.; Douglass, A. R.; Schoeberl, M. R. (2002) Estimating downward cross-tropopause flux using column
17 ozone and potential vorticity. J. Geophys. Res. 107(D22): 10.1029/2001JD002041.
18 Oltmans, S. J.; Lefohn, A. S.; Scheel, H. E.; Harris, J. M.; Levy, H., II; Galbally, I. E.; Brunke, E.-G.; Meyer, C. P.;
19 Lathrop, J. A.; Johnson, B. J.; Shadwick, D. S.; Cuevas, E.; Schmidlin, F. J.; Tarasick, D. W.; Claude, H.;
20 Kerr, J. B.; Uchino, O.; Mohnen, V. (1998) Trends of ozone in the troposphere. Geophys. Res. Lett.
21 25: 139-142.
22 Ott, W. R. (1982) Concepts of human exposure to air pollution. Environ. Int. 7: 179-196.
23 Ott, W. R. (1985) Total human exposure: an emerging science focuses on humans as receptors of environmental
24 pollution. Environ. Sci. Technol. 19: 880-886.
25 Ott, W.; Langan, L.; Switzer, P. (1992) A time series model for cigarette smoking activity patterns: model validation
26 for carbon monoxide and respirable particles in a chamber and an automobile. In: Measuring, understanding,
27 and predicting exposures in the 21st century: proceedings of the conference; November 1991; Atlanta, GA.
28 J. Exposure Anal. Environ. Epidemiol. 2(suppl. 2): 175-200.
29 Park, J.-H.; Spengler, J. D.; Yoon, D.-W.; Dumyahn, T.; Lee, K.; Ozkaynak, H. (1998) Measurement of air
3 0 exchange rate of stationary vehicles and estimation of in-vehicle exposure. J. Exposure Anal. Environ.
31 Epidemiol. 8: 65-78.
32 Park, R. J.; Pickering, K. E.; Allen, D. J.; Stenchikov, G. L.; Fox-Rabinovitz, M. S. (2004) Global simulation of
33 tropospheric ozone using the University of Maryland Chemical Transport Model (UMD-CTM): 1. Model
34 description and evaluation. J. Geophys. Res. [Atmos.] 109(D9): 10.1029/2003 JD004266.
35 Pavelin, E. G.; Johnson, C. E.; Rughooputh, S.; Toumi, R. (1999) Evaluation of preindustrial surface ozone
36 measurements made using Schonbein's method. Atmos. Environ. 33: 919-929.
37 Penkett, S. A. (1983) PAN in the natural atmosphere. Nature (London) 302: 293-294.
3 8 Penkett, S. A.; Brice, K. A. (1986) The spring maximum in photo-oxidants in the Northern Hemisphere troposphere.
39 Nature (London) 319: 655-657.
40 Penkett, S. A.; Sandalls, F. J.; Lovelock, J. F. (1975) Observations of peroxyacetyl nitrate (PAN) in air in southern
41 England. Atmos. Environ. 9: 139-140.
42 Piety, C. (2004) A review of the 2004 air quality season over the Mid Atlantic Region [presentation]. College Park,
43 MD: University of Maryland, Department of Meteorology; November 5.
44 Pinto, J. P.; Lefohn, A. S.; Shadwick, D. S. (2004) Spatial variability of PM2.5 in urban areas in the United States.
45 J. Air Waste Manage. Assoc. 54: 440-449.
46 Pippin, M. R.; Bertman, S.; Thornberry, T.; Town, M.; Carroll, M. A.; Sillman, S. (2001) Seasonal variation of
47 PAN, PPN, and O3 at the upper Midwest PROPHET site. J. Geophys. Res. (Atmos.) 106: 24,451-24,463.
48 Prados, A. I. (2000) The export of ozone and other atmospheric trace species from North America: observations and
49 numerical simulations [dissertation]. College Park, MD: The University of Maryland.
50 Prinz, B.; Krause, G. H. M. (1989) State of scientific discussion about the causes of the novel forest decline in the
51 Federal Republic of Germany and surrounding countries. In: Bucher, J. B.; Bucher-Wallin, I., eds. Air
52 pollution and forest decline, v. I: Proceedings of the 14th international meeting for specialists in air pollution
53 effects on forest ecosystems; October 1988; Interlaken, Switzerland. Birmensdorf, Switzerland:
54 Eidgenossische Anstalt fur das forstliche Versuchswesen (EAFV); pp. 27-34.
January 2005 AX3-239 DRAFT-DO NOT QUOTE OR CITE
-------
1 Prinz, B.; Krause, G. H. M; Jung, K.-D. (1985) Untersuchungen der LIS Essen zur Problematik der Waldschaden
2 [Investigations of the LIS Essen into the problem of forest damage]. In: Kortzfleisch, G., ed. Waldschaden:
3 Theorie und Praxis auf der Suche nach Antworten [Forest damage: theory and practice in the search for
4 answers]. Mtinchen, Germany: R. Oldenbourg Verlag; pp. 143-194.
5 Proyou, A. G.; Ziomas, I. C.; Stathopoulos, A. (1998) Application of a three-layer photochemical box model in an
6 Athens street canyon. J. Air Waste Manage. Assoc. 48: 427-433.
7 Pun, B. K.; Seigneur, C.; White, W. (2003) Day-of-week behavior of atmospheric ozone in three U.S. cities. J. Air
8 Waste Manage. Assoc. 53: 789-801.
9 Purushothaman, V.; Georgopoulos, P. G. (1997) Computational tools to aid the estimation and visualization of
10 potential human exposure to ozone. In: Proceedings of the Air & Waste Management Association Specialty
11 Conference on Computing in Environmental Resource Management; December, 1996; Research Triangle
12 Park, NC. Pittsburgh, PA: A&WMA; VIP-68.
13 Purushothaman, V.; Georgopoulos, P. G. (1999a) Evaluation of regional emissions control strategies for ozone
14 utilizing population exposure metrics: a new GIS-based approach applied to the July 1995 episode over the
15 OTAG domain. Piscataway, NJ: Environmental and Occupational Health Sciences Institute, Ozone Research
16 Center; ORC technical report, ORC-TR99-02.
17 Purushothaman, V.; Georgopoulos, P. G. (1999b) Integrating photochemical modeling, geostatistical techniques and
18 geographical information systems for ozone exposure assessment. Piscataway, NJ: Environmental and
19 Occupational Health Sciences Institute, Ozone Research Center; ORC technical report, ORC-TR99-01.
20 Reagan, J. (1984) Air quality data interpretation. In: Heck, W. W.; Taylor, O. C.; Adams, R. M.; Bingham, G. E.;
21 Miller, J. E.; Preston, E. M.; Weinstein, L. H., eds. National Crop Loss Assessment Network (NCLAN) 1982
22 annual report. Corvallis, OR: U.S. Environmental Protection Agency, Corvallis Environmental Research
23 Laboratory; pp. 198-219; report no. EPA-600/4-84-049. Available from: NTIS, Springfield, VA;
24 PB84-169358.
25 Regener, E. (1941) Ozonschicht und atmospharische turbulenz. Ber. Dtsch. Wetterdienstes US-Zone No. 11: 45-57.
26 Reiss, R.; Ryan, P. B.; Koutrakis, P.; Tibbetts, S. J. (1995) Ozone reactive chemistry on interior latex paint. Environ.
27 Sci. Technol. 29: 1906-1912.
28 Reiter, E. R. (1977a) Review and analysis. In: Mohnen, V. A.; Reiter, E. R., eds. International conference on
29 oxidants, 1976 - analysis of evidence and viewpoints; part III. the issue of stratospheric ozone intrusion.
30 Research Triangle Park, NC: U.S. Environmental Protection Agency, Environmental Sciences Research
31 Laboratory; pp. 67-117; report no. EPA-600/3-77-115. Available from: NTIS, Springfield, VA; PB-279010.
32 Reiter, E. R. (1977b) The role of stratospheric import on tropospheric ozone concentrations. In: Dimitriades, B., ed.
33 International conference on photochemical oxidant pollution and its control - proceedings: volume I;
34 September 1976; Raleigh, NC. Research Triangle Park, NC: U.S. Environmental Protection Agency,
35 Environmental Sciences Research Laboratory; pp. 393-410; report no. EPA-600/3-77-001a. Available from:
36 NTIS, Springfield, VA; PB-264232.
37 Research Triangle Institute. (1990) An investigation of infiltration and indoor air quality. Albany, NY: New York
3 8 State Energy Research and Development Authority; report no. NYERDA 90-11.
39 Reynolds, S. D.; Blanchard, C. L.; Ziman, S. D. (2003) Understanding the effectiveness of precursor reductions in
40 lowering 8-hr ozone concentrations. J. Air Waste Manage. Assoc. 53: 195-205.
41 Reynolds, S.; Blanchard, C.; Zinman, S. (2004) Understanding the effectiveness of precursor reductions in lowering
42 8-hr ozone concentrations. Part II: the eastern United States. J. Air Waste Manage. Assoc. 53: 195-205.
43 Riediker, M.; Williams, R.; Devlin, R.; Griggs, T.; Bromberg, P. (2003) Exposure to paniculate matter, volatile
44 organic compounds, and other air pollutants inside patrol cars. Environ. Sci. Technol. 37: 2084-2093.
45 Rifai, H. S.; Hopkins, L.; Mehdizadeh, F. (2000) Developing human exposure estimates for indoor/outdoor air
46 pollution in Houston. In: Environmental Institute of Houston -1999 Annual Report. Houston, TX:
47 Environmental Institute of Houston.
48 Roberts, J. M.; Williams, J.; Baumann, K.; Buhr, M. P.; Goldan, P. D.; Holloway, J.; Htibler, G.; Kuster, W. C.;
49 McKeen, S. A.; Ryerson, T. B.; Trainer, M.; Williams, E. J.; Fehsenfeld, F. C.; Bertman, S. B.; Nouaime, G.;
50 Seaver, C.; Grodzinsky, G.; Rodgers, M.; Young, V. L. (1998) Measurements of PAN, PPN, and MPAN
51 made during the 1994 and 1995 Nashville Intensives of the Southern Oxidant Study: implications for regional
52 ozone production from biogenic hydrocarbons. J. Geophys. Res. [Atmos.] 103: 22,473-22,490.
53 Roelofs, G. J.; Lelieveld, J.; Van Dorland, R. (1997) A three-dimensional chemistry/general circulation model
54 simulation of anthropogenically derived ozone in the troposphere and its radiative climate forcing. J.
55 Geophys. Res. [Atmos.] 102: 23,389-23,401.
January 2005 AX3-240 DRAFT-DO NOT QUOTE OR CITE
-------
1 Romieu, I.; Lugo, M. C.; Colome, S.; Garcia A. M; Avila, M. H.; Geyh, A.; Velasco, S. R.; Rendon, E. P. (1998)
2 Evaluation of indoor ozone concentration and predictors of indoor-outdoor ratio in Mexico City. J. Air Waste
3 Manage. Assoc. 48: 327-335.
4 Roselle, S.J.; Schere, K. L. (1995) Modeled response of photochemical oxidants to systematic reductions in
5 anthropogenic volatile organic compound and NOX emissions. J. Geophys. Res. C: Oceans Atmos.
6 100:22,929-22,941.
7 Rotman, D. A.; Atherton, C. S.; Bergmann, D. J.; Cameron-Smith, P. J.; Chuang, C. C.; Connell, P. S.; Dignon,
8 J. E.; Franz, A.; Grant, K. E.; Kinnison, D. E.; Molenkamp, C. R.; Proctor, D. D.; Tannahill, J. R. (2004)
9 IMPACT, the LLNL 3-D global atmospheric chemical transport model for the combined troposphere and
10 stratosphere: model description and analysis of ozone and other trace gases. J. Geophys. Res. [Atmos.]
11 109(D4): 10.1029/2002JD003155.
12 Runeckles, V. C. (1974) Dosage of air pollutants and damage to vegetation. Environ. Conserv. 1: 305-308.
13 Ryerson, T. B.; Trainer, M.; Holloway, J. S.; Parrish, D. D.; Huey, L. G.; Sueper, D. T.; Frost, G. J.; Donnelly,
14 S. G.; Schauffler, S.; Atlas, E. L.; Kuster, W. C.; Goldan, P. D.; Htibler, G.; Meagher, J. F.; Fehsenfeld, F. C.
15 (2001) Observations of ozone formation in power plant plumes and implications for ozone control strategies.
16 Science (Washington, DC) 292: 719-723.
17 Sabersky, R. H.; Sinema, D. A.; Shair, F. H. (1973) Concentrations, decay rates, and removal of ozone and their
18 relation to establishing clean indoor air. Environ. Sci. Technol. 7: 347-353.
19 Sakugawa, H.; Kaplan, I. R. (1989) H2O2 and O3 in the atmosphere of Los Angeles and its vicinity: factors
20 controlling their formation and their role as oxidants of SO2. J. Geophys. Res. [Atmos.] 94: 12,957-12,973.
21 Salmon, L. G.; Cass, G. R.; Bruckman, K.; Haber, J. (2000) Ozone exposure inside museums in the historic central
22 district of Krakow, Poland. Atmos. Environ. 34: 3823-3832.
23 Sarnat, J. A.; Koutrakis, P.; Suh, H. H. (2000) Assessing the relationship between personal paniculate and gaseous
24 exposures of senior citizens living in Baltimore, MD. J. Air Waste Manage. Assoc. 50: 1184-1198.
25 Sarnat, J. A.; Schwartz, J.; Catalano, P. J.; Suh, H. H. (2001) Gaseous pollutants in paniculate matter epidemiology:
26 confounders or surrogates? Environ. Health Perspect. 109: 1053-1061.
27 Sarwar, M.; Corsi, R.; Kimura, Y.; Allen, D.; Weschler, C. (2001) Hydroxyl radicals in indoor environments. In:
28 Proceedings of the Air & Waste Management Association's 94th Annual Conference & Exhibition; June;
29 Orlando, FL. Pittsburgh, PA: Air & Waste Management Association.
30 Sarwar, G.; Corsi, R.; Kumura, Y.; Allen, D.; Weschler, C. J. (2002) Hydroxyl radicals in indoor environments.
31 Atmos. Environ. 36: 3973-3988.
32 Saylor, R. D.; Chameides, W. L.; Chang, M. E. (1999) Demonstrating attainment in Atlanta using urban airshed
33 model simulations: impact of boundary conditions and alternative forms of the NAAQS. Atmos. Environ.
34 33: 1057-1064.
35 Scheeren, B. A.; Adema, E. H. (1996) Monitoring ambient ozone with a passive measurement technique method,
36 field results and strategy. Water Air Soil Pollut. 91: 335-350.
37 Seo, K.-H.; Bowman, K. P. (2002) Lagrangian estimate of global stratosphere-troposphere mass exchange.
38 J. Geophys. Res. (Atmos.) 107(D21): 10.1029/2002JD002441.
39 Shair, F. H. (1981) Relating indoor pollutant concentrations of ozone and sulfur dioxide to those outside: economic
40 reduction of indoor ozone through selective filtration of the make-up air. ASHRAE Trans. 87(pt. 1): 116-139.
41 Shair, F. H.; Heitner, K. L. (1974) Theoretical model for relating indoor pollutant concentrations to those outside.
42 Environ. Sci. Technol. 8: 444-451.
43 Shamoo, D. A.; Johnson, T. R.; Trim, S. C.; Little, D. E.; Linn, W. S.; Hackney, J. D. (1991) Activity patterns in a
44 panel of outdoor workers exposed to oxidant pollution. J. Exposure Anal. Environ. Epidemiol. 1: 423-438.
45 Shamoo, D. A.; Linn, W. S.; Peng, R.-C.; Solomon, J. C.; Webb, T. L.; Hackney, J. D.; Gong, H., Jr. (1994)
46 Time-activity patterns and diurnal variation of respiratory status in a panel of asthmatics: implications for
47 short-term air pollution effects. J. Exposure Anal. Environ. Epidemiol. 4: 133-148.
48 Shindell, D. T.; Faluvegi, G.; Bell, N. (2003) Preindustrial-to-present-day radiative forcing by tropospheric ozone
49 from improved simulations with the GISS chemistry-climate GCM. Atmos. Chem. Phys. 3: 1675-1702.
50 Sillman, S.; Logan, J. A.; Wofsy, S. C. (1990) The sensitivity of ozone to nitrogen oxides and hydrocarbons in
51 regional ozone episodes. J. Geophys. Res. [Atmos.] 95: 1837-1851.
52 Sillman, S.; He, D.; Pippin, M. R.; Daum, P. H.; Imre, D. G.; Kleinman, L. I.; Lee, J. H.; Weinstein-Lloyd, J. (1998)
53 Model correlations for ozone, reactive nitrogen, and peroxides for Nashville in comparison with
54 measurements: implications for O3-NOx-hydrocarbon chemistry. J. Geophys. Res. [Atmos.]
55 103:22,629-22,644.
January 2005 AX3-241 DRAFT-DO NOT QUOTE OR CITE
-------
1 Singh, H. B.; Ludwig, F. L.; Johnson, W. B. (1977) Ozone in clean remote atmospheres: concentrations and
2 variabilities. New York, NY: Coordinating Research Council, Inc.; report no. CRC-APRAC CAPA-15-76.
3 Available from: NTIS, Springfield, VA; PB-272290.
4 Singh, H. B.; Ludwig, F. L.; Johnson, W. B. (1978) Tropospheric ozone: concentrations and variabilities in clean
5 remote atmospheres. Atmos. Environ. 12: 2185-2196.
6 Singh, H. B.; Herlth, D.; O'Hara, D.; Zahnle, K.; Bradshaw, J. D.; Sandholm, S. T.; Talbot, R.; Crutzen, P. J.;
7 Kanakidou, M. (1992) Relationship of peroxyacetyl nitrate to active and total odd nitrogen at northern high
8 latitudes: influence of reservoir species onNOx and O3. J. Geophys. Res. 97: 16,523-16,530.
9 Smith, L. A.; Lefohn, A. S. (1991) Co-occurence of ozone and wet deposited hydrogen ion in the United States.
10 Atmos. Environ. Part A 25: 2707-2716.
11 Solberg, S.; Krognes, T.; Stordal, F.; Hov, 0.; Beine, H. J.; Jaffe, D. A.; Clemitshaw, K. C.; Penkett, S. A. (1997)
12 Reactive nitrogen compounds at Spitsbergen in the Norwegian Arctic. J. Atmos. Chem. 28: 209-225.
13 Serensen, D. N.; Weschler, C. J. (2002) Modeling-gas phase reactions in indoor environments using computational
14 fluid dynamics. Atmos. Environ. 36: 9-18.
15 Spier, C. E.; Little, D. E.; Trim, S. C.; Johnson, T. R.; Linn, W. S.; Hackney, J. D. (1992) Activity patterns in
16 elementary and high school students exposed to oxidant pollution. J. Exposure Anal. Environ. Epidemiol.
17 2:277-293.
18 Sprenger, M.; Wernli, H. (2003) A northern hemispheric climatology of cross-tropopause exchange for the ERA
19 15 time period (1979-1993). J. Geophys. Res. 108(D12): 10.1029/2002JD002636.
20 Sprenger, M.; Maspoli, M. C.; Wernli, H. (2003) Tropopause folds and cross-tropopause exchange: a global
21 investigation based upon ECMWF analyses for the time period March 2000 to February 2001. J. Geophys.
22 Res. (Atmos.) 108(D12): 10.1029/2002JD002587.
23 Staehelin, J.; Thudium, J.; Buehler, R.; Volz-Thomas, A.; Graber, W. (1994) Trends in surface ozone concentrations
24 at Arosa (Switzerland). Atmos. Environ. 28: 75-87.
25 Stasiuk, W. N., Jr.; Coffey, P. E. (1974) Rural and urban ozone relationships in New York State. J. Air Pollut.
26 Control Assoc. 24: 564-568.
27 Steiber, R. S. (1995) Ozone generators in indoor air settings. Research Triangle Park, NC: U.S. Environmental
28 Protection Agency, National Risk Management Research Laboratory; report no. EPA-600/R-95-154.
29 Available from: NTIS, Springfield, VA; PB96-100201.
30 Stein, M. L.; Fang, D. (1997) Ozone exposure and population density in Harris County, Texas [comment]. J. Am.
31 Stat. Assoc. 92:408-411.
32 Stohl, A. (2001) A 1-year Lagrangian "climatology" of airstreams in the Northern Hemisphere troposphere and
33 lowermost stratosphere. J. Geophys. Res. (Atmos.) 106: 7263-7279.
34 Taylor, G. E., Jr.; Hanson, P. J. (1992) Forest trees and tropospheric ozone: role of canopy deposition and leaf
35 uptake in developing exposure-response relationships. Agric. Ecosyst. Environ. 42: 255-273.
36 Taylor, G. E., Jr.; Norby, R. J. (1985) The significance of elevated levels of ozone on natural ecosystems of North
3 7 America. In: Lee, S. D., ed. Evaluation of the scientific basis for ozone/oxidants standards: proceedings of an
38 APCA international specialty conference; November 1984; Houston, TX. Pittsburgh, PA: Air Pollution
39 Control Association; pp. 152-175. (APCA transactions: TR-4).
40 Taylor, G. E., Jr.; McLaughlin, S. B.; Shriner, D. S. (1982) Effective pollutant dose. In: Unsworth, M. H.; Ormrod,
41 D. P., eds. Effects of gaseous air pollution in agriculture and horticulture. London, United Kingdom:
42 Butterworth Scientific; pp. 458-460.
43 Taylor, G. E., Jr.; Ross-Todd, B. M.; Allen, E.; Conklin, P.; Edmonds, R.; Joranger, E.; Miller, E.; Ragsdale, L.;
44 Shepard, J.; Silsbee, D.; Swank, W. (1992) Patterns of tropospheric ozone in forested landscapes of the
45 Integrated Forest Study. In: Johnson, D. W.; Lindberg, S. E., eds. Atmospheric deposition and forest nutrient
46 cycling: a synthesis of the integrated forest study. New York, NY: Springer-Verlag; pp. 50-71. (Billings, W.
47 D.; Golley, F.; Lange, O. L.; Olson, J. S.; Remmert, H., eds. Ecological studies: analysis and synthesis: v. 91).
48 Thompson, C. R.; Hensel, E. G.; Kats, G. (1973) Outdoor-indoor levels of six air pollutants. J. Air Pollut. Control
49 Assoc. 23:881-886.
50 Trainer, M.; Parrish, D. D.; Buhr, M. P.; Norton, R. B.; Fehsenfeld, F. C.; Anlauf, K. G.; Bottenheim, J. W.; Tang,
51 Y. Z.; Wiebe, H. A.; Roberts, J. M.; Tanner, R. L.; Newman, L.; Bowersox, V. C.; Meagher, J. F.; Olszyna,
52 K. J.; Rodgers, M. O.; Wang, T.; Berresheim, H.; Demerjian, K. L.; Roychowdhury, U. K. (1993) Correlation
53 of ozone with NOy in photochemically aged air. J. Geophys. Res. [Atmos.] 98: 2917-2925.
54 Turk, B. H.; Grimsrud, D. T.; Brown, J. T.; Geisling-Sobotka, K. L.; Harrison, J.; Prill, R. J. (1989) Commercial
55 building ventilation rates and particle concentrations. ASHRAE Trans. 95(part 1): 422-433.
January 2005 AX3-242 DRAFT-DO NOT QUOTE OR CITE
-------
1 U.S. Environmental Protection Agency. (1978) Air quality criteria for ozone and other photochemical oxidants.
2 Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
3 Assessment Office; report no. EPA-600/8-78-004. Available from: NTIS, Springfield, VA; PB80-124753.
4 U.S. Environmental Protection Agency. (1985) Compilation of air pollutant emission factors. Volume I: stationary
5 point and area sources. Volume II: mobile sources. 4th ed. Research Triangle Park, NC: Office of Air Qaulity
6 Planning and Standards; Ann Arbor MI: Office of Mobile Sources; report nos. AP-42-ED-4-VOL-1 and
7 AP-42-ED-4-VOL-2. Available from: NTIS, Springfield, VA; PB86-124906 and PB87-205266.
8 U.S. Environmental Protection Agency. (1986) Air quality criteria for ozone and other photochemical oxidants.
9 Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
10 Assessment Office; report nos. EPA-600/8-84-020aF-eF. 5v. Available from: NTIS, Springfield, VA;
11 PB87-142949.
12 U.S. Environmental Protection Agency. (1992) National air quality and emissions trends report, 1991. Research
13 Triangle Park, NC: U.S. Environmental Protection Agency,Office of Air Quality Planning and Standards;
14 report no. EPA/450-R-92-001. Available from: NTIS, Springfield, VA; PB93-143998.
15 U.S. Environmental Protection Agency. (1996a) Air quality criteria for ozone and related photochemical oxidants.
16 Research Triangle Park, NC: Office of Research and Development; report nos. EPA/600/AP-93/004aF-cF. 3v.
17 Available from: NTIS, Springfield, VA; PB96-185582, PB96-185590, and PB96-185608. Available:
18 http ://cfpub2 .epa. gov/ncea/.
19 U.S. Environmental Protection Agency. (1996b) Air quality criteria for paniculate matter. Research Triangle Park,
20 NC: National Center for Environmental Assessment-RTF Office; report nos. EPA/600/P-95/001aF-cF. 3v.
21 U.S. Environmental Protection Agency. (1996c) Review of national ambient air quality standards for ozone:
22 assessment of scientific and technical information. OAQPS staff paper. Research Triangle Park, NC: Office
23 of Air Quality Planning and Standards; report no. EPA/452/R-96/007. Available from: NTIS, Springfield,
24 VA; PB96-203435. Available: http://www.epa.gov/ttn/naaqs/standards/ozone/s_o3_pr_sp.html
25 (9 September 2003).
26 U.S. Environmental Protection Agency. (1997) Responses to significant comments on the 1996 proposed rule on the
27 national ambient air quality standards for paniculate matter (December 13, 1996; 61 FR 65638). Research
28 Triangle Park, NC: National Center for Environmental Assessment; docket number A-95-54; July.
29 ECAO-CD-92-0671; STD-081.
30 U.S. Environmental Protection Agency. (1998) Guideline on ozone monitoring site selection. Research Triangle
31 Park, NC: U.S. Environmental Protection Agency, National Center for Environmental Assessment;
32 EPA-454/R-98-002.
33 U.S. Environmental Protection Agency. (2002) Latest findings on national air quality: 2001 status and trends.
34 Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA-454/K-02-001.
3 5 Available: http://www.epa.gov/air/aqtrndOI/ (22 July 2003).
36 U.S. Environmental Protection Agency. (2003a) Technology Transfer Network: Air Quality System (AQS).
37 Washington, DC: Office of Air and Radiation. Available: http://www.epa.gov/ttn/airs/airsaqs/ [1 April, 2004].
38 U.S. Environmental Protection Agency. (2003b) Latest findings on national air quality: 2002 status and trends.
39 Research Triangle Park, NC: Office of Air Quality Planning and Standards; report no. EPA/454/K-03-001.
40 Available: http://www.epa.gov/airtrends/2002_airtrends_fmal.pdf [18 December, 2003]. Available: NTIS,
41 Springfield, VA; PB2004-100183.
42 U.S. Environmental Protection Agency. (2004) Air quality criteria for paniculate matter. Research Triangle Park,
43 NC: National Center for Environmental Assessment; report no. EPA/600/P-99/002aF-bF. 2v. Available:
44 http://cfpub.epa.gov/ncea/ [9 November, 2004].
45 Vakeva, M.; Ha'meri, K.; Kulmala, M.; Lahdes, R.; Ruuskanen, J.; Laitinen, T. (1999) Street level versus rooftop
46 concentrations of submicron aerosol particles and gaseous pollutants in an urban street canyon. Atmos.
47 Environ. 33: 1385-1397.
48 Vingarzan, R. (2004) A review of surface ozone background levels and trends. Atmos. Environ. 38: 3431-3442.
49 Vingarzan, R.; Taylor, B. (2003) Trend analysis of ground level ozone in the greater Vancouver/Fraser Valley area
50 of British Columbia. Atmos. Environ. 37: 2159-2171.
51 Volz, A.; Kley, D. (1988) Evaluation of the Montsouris series of ozone measurements made in the nineteenth
52 century. Nature (London) 332: 240-242.
53 Vong, R. J.; Guttorp, P. (1991) Co-occurence of ozone and acidic cloudwater in high-elevation forests. Environ. Sci.
54 Technol. 25: 1325-1329.
55 VonKuhlmann, R.; Lawrence, M. G.; Crutzen, P. J.; Rasch, P. J. (2003) A model for studies of tropospheric ozone
56 and nonmethane hydrocarbons: model description and ozone results. J. Geophys. Res. [Atmos.]
57 108(09): 10.1029/2002JD002893.
January 2005 AX3-243 DRAFT-DO NOT QUOTE OR CITE
-------
1 Wainman, T.; Zhang, J.; Weschler, C. J.; Lioy, P. J. (2000) Ozone and limonene in indoor air: a source of submicron
2 particle exposure. Environ. Health Perspect. 108: 1139-1145.
3 Wang, Y.; Logan, J. A.; Jacob, D. J. (1998) Global simulation of tropospheric O3-NOx-hydrocarbon chemistry 2.
4 Model evaluation and global ozone budget. J. Geophys. Res. (Atmos.) 103: 10,727-10,755.
5 Wernli, H.; Bourqui, M. (2002) A Lagrangian "1-year climatology" of (deep) cross-tropopause exchange in the
6 extratropical Northern Hemisphere. J. Geophys. Res. (Atmos.) 107(D2): 10.1029/2001JD000812.
7 Weschler, C. J. (2000) Ozone in indoor environments: concentration and chemistry. Indoor Air 10: 269-288.
8 Weschler, C. J.; Shields, H. C. (1997) Potential reactions among indoor pollutants. Atmos. Environ. 31: 3487-3495.
9 Weschler, C. J.; Shields, H. C. (1999) Indoor ozone/terpene reactions as a source of indoor particles. Atmos.
10 Environ. 33:2301-2312.
11 Weschler, C. J.; Shields, H. C. (2000) The influence of ventilation on reactions among indoor pollutants: modeling
12 and experimental observations. Indoor Air. 10: 92-100.
13 Weschler, C. J.; Shields, H. C.; Naik, D. V. (1989) Indoor ozone exposures. JAPCA 39: 1562-1568.
14 Weschler, C. J.; Hodgson, A. T.; Wooley, J. D. (1992) Indoor chemistry: ozone, volatile organic compounds, and
15 carpets. Environ. Sci. Technol. 26: 2371-2377.
16 Weschler, C. J.; Shields, H. C.; Naik, D. V. (1994) Indoor chemistry involving O3, NO, and NO2 as evidenced by
17 14 months of measurements at a site in southern California. Environ. Sci. Technol. 28: 2120-2132.
18 Wild, O.; Akimoto, H. (2001) Intercontinental transport of ozone and its precursors in a three-dimensional global
19 CTM. J. Geophys. Res. [Atmos.] 106: 27,729-27,744.
20 Wiley, J. A.; Robinson, J. P.; Piazza, T.; Garrett, K.; Cirksena, K.; Cheng, Y.-T.; Martin, G. (199la) Activity
21 patterns of California residents. Final report. Sacramento, CA: California Air Resources Board; report no.
22 ARB/R93/487. Available from: NTIS, Springfield, VA.; PB94-108719.
23 Wiley, J. A.; Robinson, J. P.; Cheng, Y.-T.; Piazza, T.; Stork, L.; Pladsen, K. (1991b) Study of children's activity
24 patterns: final report. Sacramento, CA: California Air Resources Board; report no. ARB-R-93/489.
25 Wilson, A. L.; Colome, S. D.; Tian, Y.; Becker, E. W.; Baker, P. E.; Behrens, D. W.; Billick, I. H.; Garrison, C. A.
26 (1996) California residential air exchange rates and residence volumes. J. Exposure Anal. Environ.
27 Epidemiol. 6:311-326.
28 Winner, D. A.; Cass, G. R. (2000) Effect of emissions control on the long-term frequency distribution of regional
29 ozone concentrations. Environ. Sci. Technol. 34: 2612-2617.
30 Wolff, G. T.; Korsog, P. E. (1992) Ozone control strategies based on the ratio of volatile organic compounds to
31 nitrogen oxides. J. Air Waste Manage. Assoc. 42: 1173-1177.
32 Wolff, G. T.; Lioy, P. J.; Wight, G. D.; Meyers, R. E.; Cederwall, R. T. (1977) An investigation of long-range
3 3 transport of ozone across the midwestern and eastern United States. Atmos. Environ. 11: 797-802.
34 Wolff, G. T.; Lioy, P. J.; Taylor, R. S. (1987) The diurnal variations of ozone at different altitudes on a rural
3 5 mountain in the eastern United States. JAPCA 37: 45-48.
36 Wolff, G. T.; Dunker, A. M.; Rao, S. T.; Porter, P. S.; Zurbenko, I. G. (2001) Ozone air quality over North America:
37 part I—a review of reported trends. J. Air Waste Manage. Assoc. 51: 273-282.
38 Wong, S.; Wang, W.-C.; Isaksen, I. S. A.; Berntsen, T. K.; Sundet, J. K. (2004) A global climate-chemistry model
3 9 study of present-day tropospheric chemistry and radiative forcing from changes in tropospheric O3 since the
40 preindustrial period. J. Geophys. Res. [Atmos.] 109(D11): 10.1029/2003JD003998.
41 World Meteorological Organization. (1995) Scientific assessment of ozone depletion: 1994. Geneva, Switzerland:
42 United Nations Environment Programme. (Global ozone research and monitoring project; no. 37).
43 World Meteorological Organization (WMO). (1999) Scientific assessment of ozone depletion: 1998. Geneva,
44 Switzerland: World Meteorological Organization, Global Ozone and Monitoring Project; report no. 44.
45 Yeng, J.; Miller, D. R. (2002) Trends and variability of ground-level O3 in Connecticut over the period 1981-1997.
46 J. Air Waste Manage. Assoc. 52: 1354-1361.
47 Zhou, J.; Smith, S. (1997) Measurement of ozone concentrations in ambient air using a badge-type passive monitor.
48 J. Air Waste Manage. Assoc. 47: 697-703.
49
January 2005 AX3-244 DRAFT-DO NOT QUOTE OR CITE
-------
wEPA
Please make all necessary changes in the below label, PRESORTED STANDARD
detach copy or copy, and return to the address in the upper left- POSTAGE & FEES PAID
hand corner. EPA
United States •-, PERMIT No. G-35
Environmental Protection 'f y°uhdo not wish to ™*Jve 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-05/004aA
January 2005
------- |