vvEPA
             United States
             Environmental Protection
             Agency
            Environmental Criteria and
            Assessment Office
            Research Triangle Park, NC 27711
EPA-600/8-78-004
April 1978
             Research and Development
Air Quality Criteria
For Ozone and
Other Photochemical
Oxidants





-------
                             NOTICE
  This document is available through the Library Services Office, MD-35, U.S.
Environmental Protection Agency, Research Triangle Park, N.C. 27711. It isalso
available from the  Superintendent of Documents, U.S. Government Printing
Office, Washington, D.C. 20402. Correspondence relating to the subject matter
of the document should be directed to:

                     Project Officer for Ozone - Photochemical Oxidants
                     Environmental Criteria and Assessment Office (MD-52)
                     U.S. Environmental Protection Agency
                     Research Triangle  Park, N.C. 27711

-------
                         EPA-600/8-78-004
                              APRIL 1978
   AIR QUALITY CRITERIA
              FOR
     OZONE  AND  OTHER
PHOTOCHEMICAL OXIDANTS
   OFFICE OF RESEARCH AND DEVELOPMENT
  U.S. ENVIRONMENTAL PROTECTION AGENCY
       WASHINGTON, D.C. 20460

-------
                          FOREWORD
  This document has been prepared pursuant to Section 108(c)of the Clean Air
Act, as amended, which requires that the Administrator from time to time review
and,  as appropriate, modify and  reissue criteria issued pursuant to Section
108(a). Air quality criteria are required by Section 108(a)to reflect accurately the
latest scientific information  useful  in indicating the kind and extent of all
identifiable effects on public health and welfare that may be expected from the
presence of a pollutant in the ambient air in varying quantities.
  The original criteria document for photochemical oxidants, AP-63, was issued
in 1970. Since that time,  new  information has  been developed,  and this
document represents the modification and reissuance of the air quality criteria
for photochemical oxidants.
  The regulatory purpose of these criteria is to serve as the basis for national
ambient air quality standards promulgated by the Administrator under Section
109  of the Clean Air Act, as amended.  Accordingly,  as provided by Section
109(d),  the Administrator  has reviewed the  national ambient  air quality
standards for photochemical oxidants based on these revised criteria and is
proposing appropriate action with respect to those standards concurrently with
the issuance of this document.
  The Agency is pleased to acknowledge the efforts and  contributions of all
persons and groups who have contributed to this document as participating
authors  or  reviewers. In the  last  analysis,  however, the Environmental
Protection Agency  is responsible  for its content.
                                   DOUGLAS M.ICOSTLE
                                   Administrator
                                   U.S. Environmental Protection Agency

-------
                            PREFACE
  This document consolidates and assesses current knowledge regarding the
origin of ozone and other photochemical oxidants and discusses their effect on
health, vegetation, certain ecosystems, and materials.
  Photochemical  oxidants  are products  of atmospheric  reactions involving
hydrocarbons (HC), nitrogen oxides (N0»), oxygen, and sunlight. Oxidants consist
mostly of ozone, nitrogen dioxide  (NOa),  and  peroxyacetylnitrate (PAN), with
smaller amounts of other peroxyacylnitrates, other oxy- and peroxy-compounds,
formic and nitric  acid, and formaldehyde. Usually referred to collectively as
"oxidants," they originate mainly from human activities that produce HC and
NO* emissions.
  This document summarizes current data on the effects of oxidant/ozone in the
ambient air on man, vegetation, and ecosystems. The effects that have been
observed will form the scientific  basis for supporting the present National
Ambient Air Quality Standard of 160 /jg/m3 (0.08 ppm) or a revised standard.
  Although nitrogen dioxide is considered one of the photochemical oxidants,
oxides of  nitrogen are-the subject of a  separate report and are therefore
discussed  in this document only as they participate  in the formation and
reactions of other photochemical oxidants. Hydrocarbons and other organics are
important air pollutants because they too are precursors of other compounds
formed in the atmospheric  photochemical system. In this document, toxic
organics are considered only with  respect to eye irritation.
  The studies  and data cited constitute the best  available basis for specific
standards  aimed  at  protecting human  health and  the  environment from
photochemical oxidants in ambient air.

-------
        CONTRIBUTORS   AND   REVIEWERS


  The following individuals participated in the preparation of this document.

Authors
    Dr. J.  H. B. Garner
    Project Officer for Ozone and Other
      Photochemical Oxidants
    Environmental Criteria and Assessment Office
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Frank Black
    Environmental Sciences Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. Robert Chapman
    Health Effects Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. Kenneth L. Demerjian
    Environmental Sciences Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. Basil  Dimitriades
    Environmental Sciences Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. Donald E. Gardner
    Health Effects Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Ms. Judith A. Graham
    Health Effects Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Fred  H. Haynie
    Environmental Sciences Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C. '
    Dr. Milan J. Hazucha
    Health Effects Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.

-------
    Mr. John H. Margeson
    Environmental Monitoring and Support Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. Frederick J. Miller
    Health Effects Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. Richard J. Paur
    Environmental Monitoring and Support Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Larry Purdue
    Environmental Monitoring and Support Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Robert K. Stevens
    Environmental Sciences Research  Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.

Environmental Protection Agency Review Group
    Dr. J.  H. B.  Garner, Chairman
    Environmental Criteria and Assessment Office
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. John Clements
    Environmental Monitoring and Support Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Robert Fankhauser
    Environmental Sciences Research  Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Thomas B. Feagans
    Strategies and Air Standards Division
    Office of Air Quality Planning and  Standards
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Charles R. Hosier
    Environmental Sciences Research  Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Mr. Michael H. Jones
    Strategies and Air Standards Division
    Office of Air Quality Planning and Standards
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr. John H. Knelson
    Health Effects Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.

-------
    Mr. Ted Ripberger
    Industrial Environmental Research Laboratory
    U.S. Environmental Protection Agency
    Research Triangle Park, N.C.
    Dr.  David Tingey
    Ecological Effects Division
    U.S. Environmental Protection Agency
    Corvallis, Ore.
    Dr.  David Weber
    Ecological Effects Division
    U.S. Environmental Protection Agency
    Corvallis, Ore.
    Dr.  Raymond Wilhour
    Ecological Effects Division
    U.S. Environmental Protection Agency
    Corvallis, Ore.

Consultants
    Dr.  Stephen M. Ayres
    Chairman, Department of Internal Medicine
    St.  Louis University School of Medicine
    St.  Louis, Mo.
    Dr.  David V. Bates
    Dean, School of Medicine
    University of British Columbia
    Vancouver, B C., Canada
    Dr.  T. Timothy Crocker
    Chairman, Department of Community and
      Environmental Medicine
    University of California College of Medicine
    Irvine, Calif.
    Dr.  Richard  Ehrlich
    Director, Life Sciences Division
    Illinois Institute of Technology Research Institute
    Chicago, III.
    Dr.  Gustave Freeman
    Director, Department of Medical  Sciences
    Life Sciences Division
    Stanford Research Institute
    Menlo Park, Calif.
    Dr.  Bernard D. Goldstein
    Associate Professor,  Department of  Medicine
      and Department of Environmental Medicine
    New York University School  of Medicine
    New York,  N.Y.
    Dr. John R. Goldsmith
    Environmental  Epidemiology Unit
    California State Department of Public Health
    Sacramento, Calif.

-------
    Dr. Jack D. Hackney
    Chief, Environmental Health Laboratories
    Associate Professor of Medicine
    Rancho Los Amigos Hospital Campus
      of the University of Southern California
    Downey, Calif.
    Dr. Sagar V.  Krupa
    Assistant Professor,  Department of Plant Pathology
    University of Minnesota
    St. Paul, Minn.
    Dr. Daniel B. Menzel
    Associate Professor of Pharmacology and Medicine
      and Director, Laboratory of Environmental Pharmacology
      and Toxicology
    Duke University Medical Center
    Durham, N.C.
    Dr. Carl M. Shy
    Institute of Environmental Health
    University of North Carolina
    Chapel Hill, N.C.
    Dr. Boyd R. Strain
    Professor, Department of Botany
    Duke University
    Durham, N.C.

Technical Assistance
    Environmental Protection Agency
    Research Triangle Park, N.C.
    Ms. F. Vandiver Duffield
    Mr. Douglas B. Fennel!
    Mr. R. Wayne Fulford
    Mr. Allen Hoyt
    Ms. Evelynne R. Rash
    Ms. Beverly A. Tilton

    DOT Systems, Inc.
    Vienna, Va.	
    Mr.  Frank Broadwell
    Dr. John Preston
    Dr. Carolyn Pyrek
    Mr. Arthur Willey

-------

-------
                       CONTENTS
                                                        Page
FIGURES 	  xiv
TABLES	  xviii
ABBREVIATIONS AND SYMBOLS 	  xxii
ABSTRACT 	  xxix

ASSOCIATION OF OZONE WITH OTHER CONSTITUENTS OR
MANIFESTATIONS OF SMOG
1. SUMMARY AND CONCLUSIONS	    1
   INTRODUCTION  	    1
   NATURE AND ATMOSPHERIC CONCENTRATIONS OF PHOTO-
   CHEMICAL OXIDANTS 	    1
   SOURCES AND SINKS OF OXIDANTS	    1
   OXIDANT PRECURSORS  	    2
   RELATIONSHIPS BETWEEN AMBIENT OXIDANT AND PRECURSOR
   EMISSIONS 	    3
   MEASUREMENT METHODS FOR OXIDANT AND OXIDANT PRE-
   CURSORS 	    3
   HEALTH EFFECTS  OF OZONE AND OTHER PHOTOCHEMICAL
   OXIDANTS 	    4
       Discussion of Threshold Concentrations  	    4
       Human Studies 	    4
       Animal Toxicology Studies	   10-
       Ozone Versus Oxidant Health Effects	   11
   EFFECTS OF PHOTOCHEMICAL OXIDANTS ON VEGETATION AND
   CERTAIN MICROORGANISMS 	   11
   EFFECTS OF PHOTOCHEMICAL OXIDANTS ON ECOSYSTEMS ..   14
   EFFECTS OF PHOTOCHEMICAL OXIDANTS ON MATERIALS  ....   16
2. INTRODUCTION 	   17

3. NATURE AND ATMOSPHERIC CONCENTRATIONS  OF PHOTO-
  CHEMICAL OXIDANTS  	   19
   NATURE OF OXIDANT 	   19
   OXIDANT CONCENTRATIONS AND THEIR PATTERNS  	   20
       Introduction 	   20
       Oxidant Concentrations in Urban Atmospheres	   20
       Oxidant Concentrations in Rural Atmospheres 	   21
       Patterns of Variation in Oxidant Concentrations 	   23
   ASSOCIATION OF OZONE WITH OTHER CONSTITUENTS OR
   MANIFESTATIONS OF SMOG	   24
   SUMMARY 	   27
   REFERENCES FOR  CHAPTER 3 	   27

4. SOURCES AND SINKS OF OXIDANTS 	   29
   PHOTOCHEMICAL  FORMATION OF OXIDANTS  	   29

-------
                                                            Page
       Introduction 	   29
       Atmospheric Reaction Mechanisms 	   30
       Effects of MeteorologiL'al Factors  	   35
   NATURAL SOURCES OF OXIDANTS  	   47
       Introduction 	   47
       Stratospheric Ozone Intrusion  	   47
       Photochemistry of Natural Organics NO*	   52
   SINKS OF OXIDANTS AND OF OXIDANT PRECURSORS 	   55
       Introduction 	   55
       Sinks for Oxidants and Oxidant Precursors 	   56
   SUMMARY	   57
   REFERENCES FOR CHAPTER 4 	   58

5. OXIDANT PRECURSORS  	   65
   INTRODUCTION  	   65
   AMBIENT LEVELS AND VARIATIONS OF OXIDANT/OZONE PRE-
   CURSORS  	   65
       Organic Compounds in Urban Atmospheres  	   65
       Organic Compounds in Rural and Remote Areas  	   70
       Nitrogen Oxides 	   71
   SOURCES  OF OXIDANT PRECURSORS  	   71
       Introduction 	   71
       Summary of Organic Emissions Data 	   77
       Hydrocarbon Emissions From  Natural Sources  	   78
       Hydrocarbon Emissions From  Anthropogenic Sources  	   81
       Emissions of Nitrogen Oxides	   82
   REACTIVITY OF ORGANIC EMISSIONS  	   83
   SUMMARY	   88
   REFERENCES FOR CHAPTER 5 	   89
6. RELATIONSHIPS  BETWEEN AMBIENT OXIDANTS AND PRECURSOR
  EMISSIONS 	   93
   INTRODUCTION 	   93
   MODELS BASED ON EMPIRICAL RELATIONSHIPS  	   94
       Rollback Model  	   94
       Modified Rollback: Observational  Model  	   95
       Statistical-Empirical Models 	   98
   MECHANISTIC MODELS OF Ox/HC/NOx RELATIONSHIPS 	   101
   AIR QUALITY SIMULATION MODELS  (AQSM) 	   105
   SUMMARY	   111
   REFERENCES FOR CHAPTER 6 	   113
7. MEASUREMENT METHODS FOR OZONE, OXIDANTS, AND THEIR
   PRECURSORS  	   116
   INTRODUCTION 	   116
   SAMPLING FACTORS IN  AMBIENT AIR MONITORING 	   116
   MEASUREMENT OF OZONE  	   117
       Gas-Phase Chemiluminescence (EPA Reference Methods)  ..   117
       Gas-Solid Chemiluminescence	   118
       Ultraviolet Photometry  	   118
   MEASUREMENT OF TOTAL OXIDANTS 	   119
   OZONE CALIBRATION PROCEDURES  	   120
       Kl  Calibration Procedures  	   1 20
       Recently Developed Calibration Procedures 	   121

-------
                                                            Page

    RELATIONSHIPS BETWEEN  AMBIENT OXIDANT AND OZONE
    DATA 	   122
    MEASUREMENT OF PEROXYACYLNITRATES  	   125
    MEASUREMENT OF HYDROCARBONS 	   1 26
       Measurement of Nonmethane Hydrocarbons	   126
       Measurement of Individual Hydrocarbons  	   1 28
       Measurement of Reactive Hydrocarbons  	   1 29
       Calibration 	   129
    MEASUREMENT OF NITROGEN OXIDES  	   130
       Measurement of N02  	   1 30
       Measurement of NO  	   131
    SUMMARY	   131
    REFERENCES FOR CHAPTER 7 	   131

8. TOXICOLOGICAL APPRAISAL OF PHOTOCHEMICAL OXIDANTS ..   136
    INTRODUCTION  	   136
    EFFECTS OF OZONE ON EXPERIMENTAL ANIMALS 	   1 36
       Respiratory Tract Transport and Absorption 	   136
       Mortality  	   1 38
       Pulmonary Effects 	   1 38
       Extrapulmonary Effects	   1 58
       Summary and Conclusions 	   1 62
    EFFECTS OF PHOTOCHEMICAL OXIDANTS  ON EXPERIMENTAL
    ANIMALS  	   1 65
       Experimental Data 	   1 65
       Summary 	   176
    EFFECTS  OF  PEROXYACETYLNITRATE ON EXPERIMENTAL
    ANIMALS  	   1 77
       Experimental Data 	   1 77
       Summary 	   177
    REFERENCES FOR CHAPTER 8 	   1 77

9. CLINICAL APPRAISAL OF THE EFFECTS OF OXIDANTS  	   184
    OCCUPATIONAL AND ACCIDENTAL EXPOSURES TO OZONE  ...   184
    CONTROLLED STUDIES OF HUMAN HEALTH EFFECTS 	   186
       Hematology 	   195
       Mutagenesis 	   196
       Controlled Studies of Human Health Effects of Peroxyacetylnitrate ..   1 97
    SUMMARY	   1 97
    REFERENCES FOR CHAPTER 9 	   1 97
10.  EPIDEMIOLOGIC APPRAISAL OF PHOTOCHEMICAL OXIDANTS .   200
    INTRODUCTION   	   200
    EFFECTS OF SHORT-TERM PHOTOCHEMICAL OXIDANT EXPOSURES   200
       Daily Mortality in Relation to Variations in Oxidant Levels  ..   200
       Hospital Admissions in Relation to Oxjdant Levels	   206
       Aggravation of Existing  Respiratory Diseases by Oxidant
       Pollution 	   208
       Effects of Oxidant on the Promotion of Symptoms and Illness in
       Healthy Populations	   212
       Impairment of Performance Associated with Oxidant Pollution ...   221
       Eye Irritation in  Relation to Variations in Oxidant Levels ....   230
    EFFECTS OF CHRONIC PHOTOCHEMICAL OXIDANT EXPOSURES ..   236
       Introduction 	   236

-------
                                                              Page
       Mortality in Areas of High- and Low-Oxidant Pollution	    237
       General Morbidity in Areas of High and Low Oxidant Pollution    239
       Chronic Oxidant Exposure and Pulmonary Function  	    244
    ATTITUDES OF LAYMEN AND PHYSICIANS TOWARD  OXIDANT
    AIR POLLUTION	,	    246
       State of California General Health Survey 	    246
       Survey of Los Angeles Physicians	    246
       Discussion of Attitudes of Respondents and Physicians Toward
       Oxidant Air Pollution  	    247
    SUMMARY OF EPIDEMIOLOGIC APPRAISAL OF PHOTOCHEMICAL
    OXIDANTS	    247
       Effects of Short-Term Photochemical Oxidant Exposures ....    248
       Effects of Chronic Photochemical Oxidant Exposures 	    249
       Attitudes  of  Laymen and  Physicians  Toward Oxidant Air
       Pollution 	,	    250
    REFERENCES FOR CHAPTER 10  	    250

11.  EFFECTS  OF  PHOTOCHEMICAL  OXIDANTS ON  VEGETATION
    AND CERTAIN MICROORGANISMS  	,	    253
    INTRODUCTION	    253
    VASCULAR PLANT RESPONSE TO PHOTOCHEMICAL OXIDANTS    253
       Physiological Processes .,	    254
       Visible Symptoms  	    255
       Growth and Yield 	    257
       Factors Affecting Plant Response 	    270
    RESPONSES OF MOSSES, FERNS, AND MICROORGANISMS ...    280
    SUMMARY	    281
    REFERENCES FOR CHAPTER 11	    284

1 2. ECOSYSTEMS	    294
    INTRODUCTION	    294
    GENERAL RESPONSES OF  NATURAL AND  AGROECOSYSTEMS
    TO STRESS BY OXIDANTS	    297
       Agroecosystems  	•	    297
       Natural Ecosystems	    297
    ORIGIN OF  INJURIOUS CONCENTRATIONS OF OZONE  AND
    OXIDANTS 	    298
       Advection From Urban Centers to Remote Areas in Southern
       California     	    298
       San Joaquin Valley and Adjacent Sierra Nevada Mountains .    301
       Seasonal and Daily Variations of Injurious  Concentrations—
       Synoptic Weather Patterns  Associated With Episodes
       of High Pollution     	    301
       Annual Trends of Total Oxidant Concentrations at a San Bernardino
       Station and the Nearby City of San Bernardino     	    304
    INVESTIGATING THE EFFECT OF OXIDANT STRESS ON ECOSYSTEMS    307
       Ecosystem Modeling	    307
       Modeling the  Effects of Oxidant Stress on  a Western Mixed
       Conifer Forest Ecosystem      	    308
       Effects on Primary Producers (Green Plants) 	    308
    EFFECTS ON CONSUMER POPULATIONS 	    319
       Vertebrate Populations	    319
       Plant  Parasites and Symbionts	    321

-------
                                                          Page
   EFFECTS ON DECOMPOSERS 	    322
   SUMMARY	    324
   REFERENCES FOR CHAPTER 12 	    326

13. EFFECTS OF OZONE ON MATERIALS  	    329
   INTRODUCTION 	    329
   MECHANISMS OF OZONE ATTACK  	    329
       Effects on Elastomers 	    329
       Effects on Textiles  	    333
       Effects on Paints  	    337
       Effects on Other Materials 	    340
   SUMMARY	    340
   REFERENCES FOR CHAPTER 13 	    340

-------
                          FIGURES
Figure                                                         Page
 3-1. Trends in NOa  air quality and in HC and N0» emission, Los
       Angeles Basin, 1965-74  	   25
 3-2, Pollutant trends in  Los Angeles, annual and 3-year moving
       averages	,	   25
 3-3. Correlation between Bscat and maximal ozone concentration    26
 4-1. Chemical changes occurring during photoirradiation of hydro-
       carbon-nitrogen oxide-air systems	   30
 4-2. The major reaction paths for the degradation of frans-2-butene
       in an irradiated NOn-polluted atmosphere  	   32
 4-3, Isopleths (m x 102) of mean summer morning mixing heights   38
 4-4, Isopleths (m x 102) of mean summer afternoon mixing heights   38
 4-5. Isopleths (m  x sec 1) of mean summer wind speed averaged
       through the afternoon  mixing layer 	   39
 4-6. Isopleths (m  x sec 1) of mean summer wind speed averaged
       through the morning, mixing layer	   39
 4-7. Isopleths of total number of forecast days of high meteoro-
       logical potential for air pollution	   40
 4-8 Monthly distribution of number of cases of 4 or more days of
       atmospheric stagnation, 1936-75 (September)	   41
 4-9. Monthly distribution number of cases of 4 or more days of
       atmospheric stagnation, 1936-75 (August) ,	   41
4-10. Diurnal variations m actinic irradiance 	   42
4-11. Long-term ozone variations at Quillayute 	•	   50
4-12. Monthly ozone variations at Mauna Loa, Hawaii 	   51
4-13. Average monthly ozone concentrations recorded at summit of
       Mount Whiteface  	   52
4-14. Effect of NO« concentration increase on ozone  formation at
       White River,  Utah	   55
 5-1. Distribution of hydrocarbons in diesel exhaust gas	   67
 5-2. Distribution of hydrocarbons in jet-aircraft engine  exhaust  .   68
 5-3 Nonmethane hydrocarbon trends in Los Angeles, 1963-72  .   70
 5-4. Oxides of nitrogen trends in Los Angeles, 1963-72, 6-to9-a.m.
       and maximum 1 -hr concentrations	   77
 5-5. Correlation of observed and calculated Oa reactivities 	   87
 6-1.  Maximum daily  1 -hr average oxidants as a function of 6- to 9-
        a.m. averages of nonmethane hydrocarbon (CAMP data from
        four U.S. cities 	   96
 6-2,  Upper limit oxidant values in the Los Angeles south coast air
        basin as a function of  average 6- to 9-a.m, concentration    97
 6-3,  Oxidant-hydrocarbon relationships from smog chamber  and
        aerometric data  	,	   97

-------
                                                             Page
 6-4,  Merz, Painter, and Ryason's relation of NO* and NMHC assumed
       as 50 percent of total HC and oxidant for downtown Los
       Angeles  	,	   99
 6-5.  California  Air  Resources  Board aerometric results, relation
       between 6- to 9-a.m. NO*, 6- to 9-a.m. HC, and maximum
       hourly oxidant concentrations in downtown Los Angeles  .   99
 6-6,  California  Air  Resources  Board aerometric results, relation
       between 6- to 9-a.m. NO,, 6- to 9-a.m. HC, and maximum
       hourly oxidant concentration in Azusa  	  100
 6-7.  California  Air  Resources  Board aerometric results, relation
       between 6- to 9-a.m. NO*, 6- to 9-a.m. HC, and maximum
       hourly oxidant concentrations in San Bernardino  	  100
 6-8,  California  Air  Resources  Board aerometric results, relation
       between 6- to 9-a.m. NO,, 6- to 9-a.m. HC, and maximum
       hourly oxidant concentrations in Anaheim	  101
 6-9.  Expected number of days per year exceeding 0.10 ppm versus
       NO* and reactive  hydrocarbon emission levels for central
       Los Angeles		  102
6-10.  Oxidant/Os isopleths  derived  from  combined use  of smog
       chamber and photochemical and modeling techniques  ...  103
 7-1.  Diurnal ozone-oxidant averages from September 4 through
       September 30, 1971	  123
 7-2.  Ozone-oxidant ratios in Musashino,  Japan  	  1 24
 7-3.  Comparison of the Kl, Mast, and chemiluminescence methods
       for measuring ozone in  irradiated  exhaust mixtures 	  1 24
 1-&,  Relationship of total oxidant and ozone values greater than 10
       ppb, August through December 1973 	  1 25
 7-&.  Comparative measurements of ambient oxidant/ozone by the
       potassium  iodide (Kl)  chemiluminescence (Chem),  and
       ultraviolet (UV) methods 	  1 26
 9-'.  Changes with time  in dynamic lung function tests in non-
       smokers during a 2-hour exposure to 0.75 ozone and during
       recovery	  189
 9-,'!,  Changes with time of minute ventilation, respiratory rate, and
       tidal volume in ozone-exposure (0.75 ppm) and subsequent
       recovery period in exercising nonsmokers	  1 90
 9-3.  Dose-response curves for  Los Angeles and Montreal subjects  195
10-1.  Comparison of deaths of persons aged 65 years and over, and
       maximum daily temperature, Los Angeles County, July 1 to
       November 30, 1955  	  202
10-2.  Comparison of nursing home deaths, maximum daily tempera-
       ture,  and smog-alert days  in Los Angeles County, July
       through December 1955 	  203
10-3.  Comparison of maximum concentrations of oxidant and carbon
       monoxide, maximum temperature, and daily death rate for
       cardiac and respiratory causes, Los Angeles County, 1956-
       58 	  204
10-4.  Monthly changes in symptom and complaint rates with daily
       maximum hourly oxidant content above and below 0.10 ppm  21 8
10-5.  Comparison of various symptom and complaint rates on days
       when oxidant concentrations were over 0.15 pphm and
       below 0.10 ppm 	  219

-------
                                                                Page
 10-6. Relationship between oxidant concentration in the hour before
         an athletic event and percentage of the team failing to im-
         prove running time 	  222
 10-7, Regression curves relating  eye  irritation scores and simul-
         taneous oxidant concentrations from a number of stations
         in the Los Angeles area, 1954  	  231
 10-8. Variation of mean maximum eye irritation, as judged by a panel
         of  scientists,  with  maximum  oxidant   concentrations,
         Pasadena, August-November 1955  	  232
 10-9. Mean index of eye irritation versus oxidant concentration   ..  234
10-10. Monthly regression  lines between  oxidant  levels  and eye
         irritation	  235
10-11. Combined  weekly incidence rates  of selected  conditions
         (colds, hay fever, asthma, and other respiratory conditions)
         in persons of all ages, Los Angeles County and the remainder
         of California, August  7 to November 28, 1954  	  240
10-12. Weekly incidence  rates of illness and injury for persons aged
         65 and  over in  Los Angeles County and the remainder of
         California, August 2 to November 28, 1954  	^.  240
10-13. Percentage of respondents  in Los Angeles County  and San
         Francisco Bay area attributing exacerbation of respiratory
         conditions to air pollution and other factors, 1 956  	  242
 11-1. Sequence of ozone-induced responses	  254
 11-2. Relation between  ozone  concentration,  foliar injury, and  a
         reduction in plant growth or yield	  264
 11-3. Limiting values for foliar injury to trees and shrubs by ozone  265
 11 -4. Limiting values for foliar injury to agricultural crops by ozone  266
 11-5. Ozone  concentration versus duration of exposure required to
         produce a 5% response in three different plant susceptibility
         groupings  	  267
  11-6, Dose-response relationships  and limiting values for  foliar
         injury to vegetation by peroxyacetylnitrate (PAN) 	  269
 11-7. Conceptual model  of factors involved in air pollution effects
         on vegetation .,	,	  271
 12-1. Law of tolerance  	  295
 12-2. Major  topographic features of the Los Angeles  Basin witn
         inland valleys and mountains  	  299
 12-3. Altitudinal sequence of ecosystems in the San  Bernardino
         Mountains	  300
 12-4. Daytime changes  in oxidant concentrations along a west-to-
         east transect in  the southern coastal air basin, including the
         slopes of the San Bernardino Mountains 	  3Q1
  12-5. The relationship of time of occurrence of the daily peak oxidant
         concentration to temperature and vapor-pressure gradients
         in an elevational sequence	  302
  12-6. Monthly summation of'total oxidant concentration data at Rim
         Forest/Sky Forest, San Bernardino Mountains, California,
         June through September, 1968-75 	  305
  12-7. Number of hours of total oxidant, July through September,
         greater than or equal to 392 fjg/m3 (0.20 ppm) at the down-
         town San Bernardino County Air Pollution Control District
         Station, 1963-74, and Rim Forest/Sky Forest,  1968-74  ..  306

-------
                                                                 Page
 12-8. Hypothetical relationships of oxidant dose, natural ecosystems,
         agroecosystems and the relative values of their vegetation
         components along a transect of southern coastal air basin of
         California 	  307
 12-9. Organism-level interactions in a mixed conifer forest  	  309
12-10. Tree-level interactions in a mixed-conifer forest ecosystem  .  311
12-11. Community-level  interactions   in  a  mixed-conifer  forest
         ecosystem  	  313
12-12. Stand-level interactions in a mixed-conifer forest ecosystem  315
12-13. Community-succession interactions in a mixed-conifer forest
         ecosystem  	  317
12-14. Topographic projection, San  Bernardino  Mountains, with
         comparison of oxidant injury to black oaks at major study
         sites, August 31,1974, with accumulated total oxidant dose
         for June-August measured at nearby monitoring stations   318
12-15. Topographic projection, San Bernardino Mountains, showing
         how ponderosa pines and Jeffrey pines in major study sites
         are distributed in  six injury classes according to seasonal
         dose of total oxidant 	  318
12-16. Annual growth of the terminal shoot and first-order branches
         in  upper half  of ponderosa  pine saplings  maintained  in
         filtered or unfiltered (ambient) air greenhouses, or in outside
         ambient air (1968-73) 	  320
12-17. Relationship of degree of oxidant injury in ponderosa pines
         with bark beetle attack and number of trees killed by western
         pine beetle, mountain pine beetle or both species	  323
 13-1. Effect of annual average ozone concentration on added costs
         resulting from damage to materials and preventive meas-
         ures 	  340

-------
                           TABLES
Table                                                        Page
3-1. Summary  of  maximum oxidant concentrations in selected
       urban areas	    20
3-2. Oxidant concentrations observed in selected urban areas of,
       the United States, 1974-75	    21
3-3. PAN and oxidant (0») measurements (10 a.m. to 4 p.m.), Los
       Angeles, Calif,, 1968	    21
3-4. PAN and ozone measurements (10 a.m. to 4 p.m.), Hoboken,
       N.J., 1970	    22
3-5. PAN, oxidant, and ozone measurements (10 a.m. to 4 p.m.), St.
       Louis, Mo., 1973	    22
3-6. Concentrations of tropospheric ozone before 1962 ....,	    22
3-7. Summary  of  ozone  data  from 1973-1975 oxidant studies.
       Research Triangle Institute	    23
4-1. Calculated rates of attack on fram-2-butene by various reactive
       species  in  simulated smog system  at several irradiation
       times	    31
4-2. Compounds observed in photochemical smog  	    35
4-3. Compounds that may be formed in photochemical smog ....    35
4-4. Secondary organic aerosols  	    36
4-5. Relative importance of aliphatic and aromatic precursors ...    37
4-6. Percentage difference between the Peterson and the Leighton
       values at the Earth's surface over  selected wave-length
       intervals and solar zenith angles  	    42
4-7. Calculated values of NO2 photodissociation rate constant at
       Earth's surface and percentage increase of the rate constant
       from the surface to various heights	    43
5-1. Hydrocarbons identified in ambient air	    66
5-2. Frequency distributions for 6- to 9-a.m. nonmethane hydro-
       carbon concentrations at CAMP sites, 1 967-72 	    69
5-3. Volatile plant products identified by Rasmussen  	    70
5-4. Frequency distribution data for 6- to 9-a,m.  nitric  oxide
       concentrations at CAMP sites, 1962-72  	    72
5-5. Frequency distribution for 6- to 9-a.m. nitrogen dioxide con-
       centrations at CAMP sites, 1962-72	    73
5-6. Nitric oxide concentration in California  by averaging time and
       frequency,  1963-67	    75
5-7. Nitrogen dioxide concentration in California by averaging time
       and frequency, 1963-67 	    76
5-8. Estimates  of total hydrocarbon emissions from manmade and
       natural  sources 	    77
5-9. Total yearly  hydrocarbon emission  rate for the continental
       United States, based on leaf biomass	    77

-------
                                                               Page
5-10. 1974 nationwide estimates of total hydrocarbon sources and
        emissions .,	   78
5-11. Nationwide total hydrocarbon emission trends, 1970-75  ...   78
5-12, Hydrocarbon emissions from mobile sources  	   78
5-13. Organic emission inventory for the metropolitan Los Angeles
        air quality control region, 1972	   79
5-14. Reactive emission inventories for  the metropolitan Los Angeles
        air quality control region 	'.	   80
5-15. 1974  nationwide  estimates of  nitrogen oxide sources  and
       , emissions 	   83
5-16.-Comparison of reactivities of different types of organics  	   84
5-17. Reactivity data	   84
5-18. Summary of data from studies on reactivities (toluene equivalents)
        and classification of  organics	   85
5-19. Classification of organics with respect to their oxidant-related
        reactivity in urban atmospheres  	   86
 6-1, Forms of air quality simulation models	  107
 6-2. Currently available air quality simulation  models for photo-
        chemical oxidant	  108
 6-3. Treatment of meteorological variables in air quality simulation
      '  models	  109
 6-4. Uncertainties in predicted  ozone  concentrations from all
        mechanisms and input uncertainties	  111
 6-5, Prior validation studies of air quality  simulation models  	  112
 7-1. Performance specifications for automated methods  	  118
 7-2. List of designated reference methods .,	  119
 7-3. List of designated equivalent methods	  119
 7-4. Percent difference from known concentrations of nonmethane
        hydrocarbons obtained by sixteen users	  127
 8-1. Pulmonary effects of ozone; Host defense mechanisms .,.,,  166
 8-2. Pulmonary effects of ozone: Biochemistry  .................  167
 8-3. Pulmonary effects of ozone: Morphology  	,	  170
 8-4. Pulmonary effects of ozone: Pulmonary function	  171
 8-5. Pulmonary effects of ozone: Edema and tolerance	  172
 8-6. Extrapulmonary effects  of ozone:  Hematology  and serum
        chemistry	  173
 8-7. Extrapulmonary effects of ozone: Central nervous system and
        behavior	  174
 8-8. Extrapulmonary effects of ozone: Morphology	  1 74
 8-9, Extrapulmonary effects of ozone: Miscellaneous 	  1 74
8-10. Atmospheric mean concentrations and their standard deviations
        administered from 8 a.m. to midnight each day  	  175
8-11. Effects of oxidants on animals	  176
 9-1. Summary  of available data on occupational exposure of
        humans to ozone	  185
 9-2. Summary of data on human experimental exposure to ozone
        before 1970 	  187
1G-1. Average number of deaths per day resulting from cardiac and
        respiratory causes among residents of Los Angeles County
        aged 65 and over, as related to oxidant concentrations and
        maximum daily temperature, by month 1954-55  	  201

-------
                                                                Paae
 10-2.  Comparison of respiratory and cardiac deaths during smog and
         smog-free periods occurring in Los Angeles County, August-
         November 1954	  206
 10-3.  Correlation of morning and early afternoon oxidant levels with
         oxygen consumption and airway resistance of 15 patients
         with chronic respiratory disease 	  211
 10-4.  Relationship of average daily simple and adjusted percentage
         of student nurses report to  photochemical oxidant levels
         for 868 days, November 1961 through May 1964	  213
 10-5.  Sign-test data for testing the association of oxidant levels with
         accidents in  Los Angeles, August through October, 1963
         and 1965 	  224
 10-6.  Number of subjects for whom correlation coefficients between
         environmental  and respiratory function measurements are
         significant at p < 0.05, Japan, 1972 	  227
 10-7.  Correlation  of eye irritation with  simultaneous oxidant con-
         centrations, in  order of decreasing  eye irritation score, for
         a number of stations in the Los Angeles area,  1954  	  231
 10-8.  Correlation  between eye irritation and simultaneous environ-
         mental measurements, as judged by a panel of scientists, 1954  232
 10-9.  Effect of filter on  sensory irritation  and chemical measurements  233
10-10.  Pearson product  moment correlation  of coefficients between
         eye irritation  and environmental factors in a nonfiltered
         room  	  234
10-11.  Total lung cancer mortality in an American Legion study popu-
         lation, California, 1958-62 	  237
10-12.  Lung cancer deaths and relative risks  per 100,000 man-years
         in an American Legion study population, by extent of cigarette
         smoking and residence, California,  1958-62 	   2b8
10-13.  Total chronic respiratory disease mortality in an  American
         Legion study population, California, 1958-62 	   239
10-14.  Selected respiratory conditions  reported by general population
         sample, California, May 1956 	   241
10-15.  Percent of survey response of general and working population
         bothered  by air pollution, by  major geographic areas in
         California, May 1956  	   246
10-16. Air pollution effects reported in general population survey, by
         type of community and by major geographic areas in Califor-
         nia, May  1956  	   247
 11-1. Effects of oxidants (ozone) in ambient  air on growth, yield, and
         foliar  injury in  selected plants  	   258
 11-2. Effects  of acute exposure  on  growth and  yield of selected
         plants 	   260
 11-3. Effects  of  ozone on   selected  understory species from
         an aspen community  	   261
 11-4. Effects  of  long-term,  controlled  ozone  exposures  on
         growth, yield, and foliar injury to selected plants	   262
 11 -5. Concentration, time, and response equations for three suscep-
         tibility groups  and for selected plants or plant types with
         respect to ozone  	   268
 11-6. Ozone concentrations for short-term  exposures  that produce
         5 or 20 percent injury to vegetation grown under sensitive
         conditions  	   269

-------
                                                                 Page
 11-7. Estimates of economic losses to crops and vegetation in the
         United States attributable to oxidant air pollution  .,	  270
 11-8. Response of plants to ozone, as conditioned by humidity during
         growth and exposure			  274
 11-9. Effects of soil moisture on  response of selected  plants to
         oxidant stress	  275
11-10. Effects of various nutrients on response of selected plants to
         ozone (oxidant) stress	  276
11-11. Summary effects of sulfur dioxide and ozone mixtures on foliar
         injury  	,	  277
11-12. Foliar response of selected plants to sulfur dioxide and ozone
         mixtures	  279
11-13. Growth response of selected plants to sulfur dioxide and ozone
         mixtures	  279
 12-1. Composition of ecosystems	  294
 12-2. Descriptions of meteorologic patterns for five classes of spring
         and summer days in southern California 	  304
 12-3. Changes in  oxidant injury scores and mortality rates of ponder-
         osa and Jeffrey pines at  18 major study plots, 1973-74 ..  319
 12-4. Changes of timber  volume and percentage of total  Jeffrey
         pines at Barton Flats in the San Bernardino National Forest  319
 12-5. Tree species and size composition in a study area affected by
         oxidant air pollution	  321
 12-6. Injury thresholds for 2-hour exposures to ozone 	  322
 13-1. Formulation of highly ozone-sensitive natural rubber	  331
 13-2. Effect of ozone on natural rubber		  331
 13-3. Tire sidewall formulation  	  332
 13-4. Effects of ozone  on  sidewall formulations containing various
         antiozonant  concentrations	  332
 13-5. Cracking rates of white sidewall tire specimens	  332
 13-6. Paint erosion  rate  coefficients for  the effects of  ozone in
         laboratory-controlled environments	  338
 13-7. Paint erosion rates at field  exposure sites	,	  339

-------
           ABBREVIATIONS AND  SYMBOLS
A
AaDo,
AChE
AlChE
AM
APCD
AQSM
Ar
A203
ATPase
Avg
BAKI
bdft
7Be
Br2
Bscat
°C
14C
CAMP
GARB
Ci-C3
Ci-C4
C2+

C2H2
C4+
C4-Ce
C6H5COO2NO2
CC
v-*dyn
CH(CH3)CHO
CHsCHCHO
CH2CH2
CH2CO
CH2O
CH3
CHaCHO
CH3CO02N02
CH4
Chem
CI2
Angstrom
Alveolar-arterial gradient
Acetylcholinesterase
American Institute of Chemical Engineers
Alveolar macrophages
Air pollution control district
Air quality simulation model
Argon
Arsenous oxide
Adenosine triphosphatase
Average
Potassium iodide solution acidified with boric acid
Board foot
Radioisotope of beryllium
Bromine
Extinction coefficient due to scatter by aerosols
Degrees Celsius
Radioisotope of carbon
Community Air Monitoring  Program
California Air Resources Board
Hydrocarbons containing one to three carbon atoms
Hydrocarbons containing one to four carbon atoms
Hydrocarbons containing more than two carbon
  atoms
Acetylene
Hydrocarbons containing more than four carbon
  atoms
Hydrocarbons containing four or five carbon atoms
Hydrocarbons containing four to six carbon atoms
Hydrocarbons containing five to ten carbon atoms
Benzyl
Peroxybenzoylnitrate (PBzN)
Closing capacity
Dynamic compliance
Propionaldehyde
Acrolein
Ethylene
Ketene
Formaldehyde; also HCHO  and H2CO
Methyl group
Acetaldehyde
Peroxyacetylnitrate (PAN)
Methane
Chemiluminescence
Chlorine

-------
cm
cm3
CNS
CO
COz
Coh
COMT
cone.
COOH
COONO2
Cro3
Cs,
Cyt. P45o
2,4-D
dbh
DEN
DiFKIN
DLco
DNA
DO LA
DW
EA
EC
EH
EKMA
EPA
ESR
Et
F,
F2
op
FEF 25-50%
FEF 25-75%
FEF 50%
FEFmax
Fe2O3
FEV
FEVo.75
FEV,,0
FEV3,0
FID
FRC
FS
FT
ft
ft2
ft-c
FVC
9
Gaw/Vtg
GC
GDI
gf
GH
Centimeter
Cubic centimeter
Central nervous system
Carbon monoxide
Carbon dioxide
Coefficient of haze
Catechol-o-methyltransf erase
Concentration
Carboxyl group
Peroxynitrate group
Chromium trioxide (chromic anhydride, chromic acid)
Static lung compliance
Cytochrome Paso
2,4-DichIorophenoxyacetic acid
Diameter breast height
Diethylnitrosamine
Diffusion Kinetics Model
Ccirbon monoxide diffusing capacity of lung
Deoxyribonucleic acid
Downtown Los Angeles
Dry weight
Environment Agency (Japan)
Electron capture detection
Humidity during exposure
Empirical Kinetic Modeling Approach
U, S. Environmental Protection Agency
Electron spin resonance
Ethyl
First filial generation
Second filial generation
Degrees Fahrenheit
Mean expiratory flow during middle half of FVC
Maximal midexpiratory flow rate
Instantaneous expiratory flow  at 50% of FVC
Peak expiratory flow
Ferric oxide
Forced expiratory volume
0.75-second forced expiratory  flow
1 -second forced expiratory flow
3-second forced expiratory flow
Flame ionizatjon detection (or  detector)
Functional residual capacity
Fourier-transform spectroscopy; also FT
Fourier-transform spectroscopy; also FS
Foot
Square foot
Foot-candle
Forced vital capacity
Gram
Specific conductance
Gas chromatography
Gas distribution index
Emission growth factor
Humidity during growth

-------
GLC
GM
G-6-P
G-6-PD
GPT
GSC
GSH
GSSG
GSSRase
H
H+
3H
ha
HC
HCHO
HCOOH
He
Hg
HI
HMP
HN02
HN03
HO
H02
HONO
HON02
H02NO
H02N02
HOON02
hr
hv
H2CO
H20
H202
H2S
H2S04
I
I,
la"
131,
in.
IR
J*
k
K
K+
KBr
kg
KH
Kl
K,
km
kPa
kV
Gas-liquid chromatography
General Motors Corporation
Glucose-6-phosphate
Glucose-6-phosphate dehydrogenase
Gas-phase titration
Gas-solid chromatography
Reduced glutathione
Oxidized glutathione
Glutathione reductase
Hydrogen
Hydrogen  ion
Tritium
Hectare
Hydrocarbons
Formaldehyde; also CH20 and H2CO
Formic acid
Helium
Mercury
Hemagglutination inhibition
Hexose  monophosphate
Nitrous  acid; also HONO
Nitric acid; also HON02
Hydroxyl group; also OH
Hydropheroxy radical
Nitrous  acid
Nitric acid
Pernitrous acid
Pernitric acid; also HOON02
Pernitric acid
Hour
Photon
Formaldehyde; also HCHO and CH20
Water
Hydrogen  peroxide
Hydrogen  sulfide
Sulfuric acid
Intensity of sunlight
Iodine
Triiodide ion
Radioisotope of iodine
Inch
Infrared
Actinic  irradiance
Rate constant; or, dissociation constant
Potassium
Potassium ion
Potassium bromide
Kilogram
Horizontal eddy diffusivity coefficient
Potassium iodide
Photodissociation rate constant
Kilometer
Kilopascal
Kilovolt

-------
Ky
LA
LARPP
ib
LDH
LiRAQ
I/min
In
LT50
Ix
M
m
m
m2
m3
MAO
max
mb
MBC
MCP
Me
MEFR
mg
mi
mi2
ml
min
min.
mm
MMC
MMEF
mmol
mo
mph
MS
n
H2
AN2
n/a
NAD
NAAQS
NADB
NADH
NADP*
NADPH

NAS
NATO/CCMS

NBKi
NBS
ND
Vertical eddy diffusivity coefficient
Los Angetes
Los Angeles Reactive Pollutant Project
Pound
Dose lethal to 50% of recipients
Lactic acid dehydrogenase
Livermore Regional Air Quality (Model)
Liters per minute
Natural logarithm
Time in which 50% of recipients of stated dose died
Lux
Third body (in a reaction)
Meter
Meta
Square meter
Cubic meter
Monoamine oxidase
Maximum
Millibar
Maximum breathing capacity
McKee, Childers, and Parr (test)
Methyl
Maximum expiratory flow rate
Milligram
Mile
Square mile
milliliter
Minute
Minimum
Millimeter
Mean meridional circulation
Mid-maximal  expiratory flow rate
Millimole
Manganese dioxide
Month
Miles per hour
Mass spectroscopy
Normal
Nitrogen
Single breath Na alveolar plateau
Not applicable
Nicotinamide  adenine dinucleotide
National  Ambient Air Quality Standard
National  Air Data Bank
Reduced  nicotinamide adenine dinucleotide
Nicotinamide  adenine dinucleotide phosphate
Reduced  nicotinamide adenine dinucleotide
  phosphate
National  Academy of Sciences
North Atlantic Treaty Organization/Committee on
  Challenges  of Modern Society
Neutral buffered potassium iodide
National  Bureau of Standards
No data;  or, not detectable

-------
Ne
NEDS
NH3
nm
NMHC
NO
N02
NO*
N20
N205
NPSH
o
0
02
03
OAQPS
O.D.
0('D)orO(1A)
OH
OH°
03.0lefin
0(3P)
0,
P
P
32P
PAN
Pa02
PAQ
PBN
PBzN
6-PG
6-P-GD
PH

P.soBN
PMN
PNH
Po,
PP
ppb
ppb C
pphm
ppm
ppm C
PPN
ppt
psi
PST
Pst TLC
r
R
RAPS
Raw
Neon
National Emissions Data System
Ammonia
Nanometer
Nonmethane hydrocarbons
Nitric  oxide
vi;trogen dioxide
Nitrogen oxides
Nitrous oxide
Nitrogen pentoxide
Nonprotein sulfhydryls
Ortho
Atomic oxygen
Oxygen
Ozone
Office of Air Quality Planning and Standards, EPA
Orthostatic dysregulation
Excited atomic oxygen
Hydroxyl group; also HO
Hydroxyl radical
Ozonides
Ground-state atomic oxygen
Photochemical oxidants
Probability (of observed results occurring by chance)
Para
Radioisotope of phosphorus
Peroxyacetylnitrate
Arterial oxygen tension
Present air quality
Peroxybutyrylnitrate
Peroxybenzoylnitrate
6-phosphogluconate
6-phosphogluconate dehydrogenase
Measure of acidity (negative  logarithm of hydrogen
  ion  concentration)
Peroxyisobutyrylnitrate
Polymorphonuclear leukocytes
Paroxysmal nocturnal hemoglobinuria
Oxygen tension
Photoperiod
Parts  per billion
Parts  per billion expressed as carbon
Parts  per hundred million
Parts  per million
Parts  per million expressed as carbon
Peroxypropionylnitrate
Parts  per trillion
Pounds per square inch
Pacific standard time
Maximal transpulmonary pressure
Correlation coefficient
Roentgen
(St. Louis) Regional Air Pollution  Study
Airway resistance

-------
RBC
RCHO
RC(0)02
RCOO2
RCOO2H
RC(O)02NO2
ROW
RH
RHC
RL
RO2
RONO2
RO2NO2
ROOM
rpm
Rt
RTI
RTP
RV
S
-S-
SAI
SBR
S.D.
sec
SEM
S.F.
SGaw
SGPT
SH
SMSA
S02
S03
SO,
SOD
sp.  or spp.
SPM
90Sr
SRI
SRM
-SS-
SSET
STA
Std
t
TDW
TF
TGV
TLC
UKi
UV
V25
V50
VC
Red blood cells; erythrocytes
Higher aldehydes
Peroxyacyl radical
Organic peroxy compounds
Organic peracids
Peroxynitrates
Root dry weight
Relative humidity
Reactive hydrocarbons
Pulmonary resistance
Alkoperoxy radical
Organic nitrates
Organic peroxynitrates
Organic hydroperoxides
Revolutions per minute
Flow resistance
Research Triangle Institute
Research Triangle Park
Residual volume
Sulfur
Sulfide linkage
Systems Applications, Inc. (Model)
Styrene-butadiene rubber
Standard deviation
Second
Scanning electron microscopy
San Francisco
Specific airway conductance
Serum glutamic-pyruvic transaminase
Sulfhydryl group
Standard metropolitan statistical area
Sulfur dioxide
Sulfur trioxide
Sulfur oxides
Superoxide dismutase
Species
Suspended particulate matter
Radioisotope  of strontium
Stanford Research Institute
Standard reference material
Disulfide linkage
Small-scale eddy transport
Seasonal tropopause adjustment
Desired air quality (standard)
Student's statistic
Top dry weight
Tropopause folding
Thoracic gas  volume
Total  lung capacity
Unbuffered potassium iodide
Ultraviolet
Maximum expiratory flow rate at 25% of VC
Maximum expiratory flow rate at 50% of VC
Vital capacity

-------
VMT
Vo,
V02 max
v/v
W
wk
wt
XO, X02
a
     or ptmole
Vehicle miles traveled
Oxygen uptake
Maximum oxygen uptake
Volume/volume
Watts
Week
Weight
Compounds in which X represents hydrogen or an
  organic radical (R or  RO)
Year
Level of statistical significance set by investigator
Microgram
Microgram per cubic meter
Micrometer
Micromole
Greater than
Less than
Approximately

-------
                           ABSTRACT
  This document is an evaluation and assessment of scientific information
relative to determination of health and welfare effects associated with exposure
to various concentrations of ozone and other photochemical oxidants in ambient
air. The-document is not intended as a complete, detailed literature review. It
does not cite every published article relating to oxidants in the environment and
their effects. The literature through 1976-1977 has been reviewed thoroughly
for  infprmation relative to  criteria. An attempt has been made to identify the
major discrepancies in our current knowledge, again relative to criteria.
  Though the emphasis is  on presentation of health and welfare effects data,
other  scientific  data  are presented  and  evaluated to  provide  a better
understanding of the  pollutants in the environment. To this end, separate
chapters  are  included on the nature and atmospheric concentrations.of
photochemical  oxidants,  their sources  and removal  processes,  oxidant
precursors,  the  relationships  between  ambient  oxidants  and  precursor
emissions, and measurement methods for ozone, oxidants, and their precursors.
  Specific areas addressed within a general area of health or welfare effects are
as follows: Toxicological, clinical, and epidemiologic appraisals, and effects on
vegetation, ecosystems, and materials.

-------

-------
                    1.   SUMMARY  AND  CONCLUSIONS
INTRODUCTION
  This  document  consolidates  and  assesses
current knowledge regarding the origin of ozone
and other photochemical  oxidant pollutants and
their  effects  on  health,  vegetation,  certain
ecosystems,  and  materials.  This chapter
summarizes the information  contained  in  this
document  and includes  conclusions that  are
believed  to  provide  a  reasonable  basis for
evaluating the effects on health or welfare that are
produced by various concentrations of ozone and
other photochemical oxidants. Although nitrogen
dioxide is considered one of the photochemical
oxidants, oxides of nitrogen are the subject  of a
separate report and are therefore discussed in this
document only as they participate in the formation
and reactions of other photochemical oxidants.

  Hydrocarbor s  (HC)  and  other  organics  are
important air nollutants almost entirely because
they are preci1 -sors of other compounds formed in
the  atmosph  sric  photochemical  system,  not
because  they  produce  any  direct effects
themselves, "i  axic organics are  considered only
with respect to eye irritation.
  The studies  and data cited constitute the best
available basii for specific standards aimed at
protecting  hu nan  health and the environment
from photochemical oxidants in ambient air.

NATURE AND ATMOSPHERIC CONCEN-
TRATIONS OF PHOTOCHEMICAL OXIDANTS

  Photochemical  oxidants are  products  of  at-
mospheric reactions involving organic pollutants,
nitrogen oxides (NO*), oxygen, and sunlight. They
consist  mostly  of  ozone,  NOz, and peroxy-
acetylnitrate (PAN), with smaller amounts of other
peroxyacylnitrates and other peroxy compounds,
and they  are  formed  along  with  other
photochemical  products   such  as  aldehydes,
nitrous  acid,  nitric  acid,  and  formic  acid.
Photochemical  oxidants  originate  mainly from
volatile organic and NOX  emissions produced by
human  activities.  Photochemical oxidant
formation is a complex function of emissions and
meterological patterns.
  Peak concentrations of oxidant, expressed as
ozone, are generally higher in urban and suburban
areas than in rural areas, reaching levels i  excess
of 590 /ug/m3  (0.3 ppm). In rural  areas, peak
concentrations are lower but often exceed the 1-hr
National  Ambient Air Quality Standard  of 160
/ug/m3 (0.08 ppm). However, dosages or average
concentrations in rural areas are comparable to or
even higher than those in urban areas. Because of
pollutant transport, oxidant pollution  is a regional
rather than a local problem.

SOURCES AND SINKS OF OXIDANTS
  All  of the evidence  presently available shows
that in and around urban centers that have severe
oxidant/ozone pollution, photochemical  oxidant
formed from anthropogenic organics and NOX is
the major contributor. Since Air Quality Criteria for
Photochemical Oxidants was issued in 1970, the
mechanisms of atmospheric oxidant/ozone for-
mation have been studied intensively  and are now
understood in greater detail. Most noteworthy are
recent findings pertaining to the roles of hydroxyl
(OH-) and hydroperoxy(HO-z) radicals. The reaction
with  OH- has been established  to  be a major-
hydrocarbon-consuming process,  and OH-  and
H0'2 have been identified as having major roles in
the atmospheric oxidation of NO to NOz. Aldehydes
and PAN have also been found to play important
roles  in the atmospheric reaction. For olefins and
paraffins,  at least, these  reactions are  now
understood to the extent that  the   kinetics of
photochemical  hydrocarbon-NOx  reaction
systems,  as observed  in the laboratory,  can  be
described with  reasonable accuracy. Additional
research  is  needed to explain the atmospheric
reactions of aromatic hydrocarbons and to clarify
further the differences  between  laboratory and
ambient atmospheric chemical systems.
  The photochemical formation of oxidant/ozone
is  the result of two  coupled  processes: (1) a
physical   process  involving  dispersion  and
transport of precursors to oxidants (e.g., HC and

-------
NO,), and {2} the photochemical reaction process.
Both  processes  are  strongly  influenced  by
meteorological factors such as dispersion, solar
radiation, temperature, and humidity. Recent data
on wind velocity  and mixing height show that
episodes of limited dispersion are most common in
the Far West and  in the Rocky Mountains region,
least common  over the  Plains  States, and  of
intermediate frequency east of "the Mississippi.
New measurements of solar radiation indicate that
the distribution of  light intensity among respective
wavelength  intervals  is  different  from  that
previously  reported, which results in  higher
photodissociation constants for NOz in the ambient
atmosphere. Recent field and laboratory studies
suggest that at temperatures below approximately
55° to  60°F,  concentrations of photochemical
ozone are unlikely  to exceed the national 1-hr
standard of 160//g/m3  (0.08 ppm).
  For understanding the sources of photochemical
oxidants or ozone (oxidant/ozone),  perhaps the
most  important  recent  development is the
identification of short-and intermediate-range and
synoptic-scale  transport of  photochemical
oxidant/ozone.  Short-range  (urban-scale)
transport causes the highest ozone concentrations
some  distance downwind  from the core area
(region  of highest emission) of an urban center.
Intermediate-range  (mesoscale) transport occurs
as urban oxidant/ozone plumes extending  as far
as  100 miles or more downwind  and is also
involved in land-sea breeze circulation. Finally,
synoptic-scale transport  over  several  hundred
miles, associated with high-pressure systems, has
been found to occur extensively. These findings
have significant implications with respect  to the
location of oxidant/ozone monitoring stations.
  Conditions during long-range transport are such
that ozone production  per unit  of  precursor is
enhanced. Also, many organics previously thought
to  be   unreactive  are  now  believed  to have
significant ozone-producing potential.
  In  addition  to  photochemical  reactions  of
anthropogenic  emissions,  potential sources  of
oxidant/ozone in the troposphere are the intrusion
of stratospheric  ozone and the photochemical
reactions of natural organic and N0« emissions.
Estimates of ground-level concentrations of ozone
originating in the stratosphere are  based on two
types of evidence: (1) global circulation patterns,
namely, patterns in air  interchange  between
Stratosphere  and troposphere;  and  (2) data  on
variations of ozone concentrations in remote rural
areas.  Based  on the evidence of  stratosphere-
troposphere  interchange, the  annual  average
stratospheric  contribution  to  ozone  con-
centrations at ground level is estimated to be 43 to
98  fjg/m3  (0.022 to 0.05  ppm). The  highest
concentrations, at or above 160 /ug/m3 (0.08 ppm),
from that source are expected to occur mainly
during  April and May. Analysis of ozone data for
rural areas indicated that major  intrusions  of
stratospheric ozone also occur during the spring
months  in mid-latitudes.  The  highest  1-hr
concentration  of stratospheric  ozone  reaching
ground level during the smog season (usually late
summer or early fall) has been found to range from
29 to 78 //g/m3 (0,015 to 0.040 ppm), depending
on the  investigator. More recent data obtained at
Whiteface  Mountain,  New  York,  suggest  a
maximum 24-hr concentration of 72^g/m3 (0.037
ppm) stratospheric ozone  in July,
  Certain  organic  emissions from  vegetation
(terpenes) play the dual role  of  oxidant/ozone
precursor and scavenger.  Despite the substantial
rates at which they are emitted in forested areas,
the ambient concentrations  of  Such organics,
because of their reactivity  and the areal dispersion
of their sources, seldom  exceed a few parts per
billion  (ppb). At these concentrations, the  direct
potential  of terpenes for photochemical  ozone
formation  is estimated   to be  negligible.  It  is
conceivable,  however,  that  the  products  of
atmospheric reactions involving large amounts of
terpenes  do  have  a   significant  impact  on
oxidant/ozone-related air quality.

OXIDANT PRECURSORS
  Organic  pollutants  in urban  atmospheres
consist mainly of hydrocarbons from automobile
exhaust and fuel evaporation, and of oxygenated
hydrocarbons  and halocarbons emitted mainly
from manufacturing and from the use of organic
chemicals.  In  urban atmospheres, total organic
concentrations, reported  usually as  6- to 9-a.m.
averages  of  total  nonmethane  hydrocarbons
(NMHC), are typically in the 1 -ppm range and can
be as high  as 10 ppm or even higher, as in Los
Angeles.  In rural and remote atmospheres, the
composition and concentration of organic pollut-
ants are uncertain, largely because of deficiencies
in the analytical methods available for determining
the concentrations  involved.  Evidence  suggests
that NMHC  levels are generally less than 0.1 ppm,
a fraction of which consists of vegetation-related
terpenes.
  Concentrations of  N0« in  urban atmospheres
vary within a wide range,  with highest values

-------
exceeding 1 ppm. Such concentrations appear to
decrease rapidly as the air mass moves away from
the city  In rural and  remote areas,  ambient
concentrations do not exceed a few ppb and are
often below the  5 ppb detection limit of current
commercial N0» analyzers.
  Hydrocarbons  and  NO* are  emitted  to the
atmosphere  from both  natural and manmade
sources,  with  natural  sources  probably  con-
tributing  more on a  global  scale.  Natural  and
anthropogenic sources are,  however, generally
segregated geographically, so that anthropogenic
emissions are concentrated in the populated ur-
ban areas. Thus it  is the latter that are  most
relevant  to  oxidant/ozone  pollution  problems
downwind  from  populous areas.  At  present,
mobile sources appear to account for the major
part of the organic  compounds emitted in  most
urban areas. The imposition of emission standards
for  mobile sources  has reduced  the  reactive
hydrocarbon  component  of  emissions  from
gasoline-powered vehicles.
RELATIONSHIPS  BETWEEN  AMBIENT
OXIDANT AND  PRECURSOR EMISSIONS
  Quantitative relationships  between  ambient
oxidant/ozone  and   precursor emissions  are
needed   for  predicting the  impact  of  these
emissions  on air  quality. Such  relationships
represent, with varying degrees of complexity, the
physical  and chemical processes taking place in
the atmosphere.
  Air  quality  simulation  models  (AQSM's),  a
number  of which  are  available,  represent an
approach to  relating  precursor  emissions to
oxidant  air quality.  However, AQSM's require a
great  deal  of  computation and have not  been
adequately evaluated.
  In summary, satisfactory methods of relating the
emission of precursors to oxidant concentrations
in ambient air are notyetavailable. For this reason,
the U.S. Environmental Protection Agency has
been  sustaining  an intensive research effort to
develop  and  validate  satisfactory  models  that
relate emissions  to air quality. Based on results of
this research and on  model performance standards
yet to  be defined, the Agency will recommend
specific models for  use in formulating optimum
ozone control strategies.

MEASUREMENT METHODS FOR  OXIDANT
AND OXIDANT PRECURSORS
  The chemiluminescence method,  based on the
gas-phase reaction  of ozone with ethylene, was
adopted by EPA as the reference method and is
now being used extensively. Several commercial
instruments based on reference methods specified
in EPA's Ambient Air  Monitoring Reference and
Equivalent Methods have been found to perform
better than required by these EPA regulations.
Techniques for continuous  measurement, based
on gas-solid chemiluminescence and ultraviolet
(UV)  photometry,  have  been  developed  and
designated as equivalent methods.
  Extensive studies in recent years have shown
that  various traditional Kt calibration procedures
for oxidant or ozone analyzers are deficient, and
that  these methods suffer from  lack of both
precision and accuracy.
  Methods for measuring total oxidants are based
on the oxidation of potassium iodide (Kl) and the
electrochemical or colorimetric detection of iodine
(I2). Nitrogen dioxide, however, causes a positive
interference,  and  S02  causes  a  negative
interference in these methods; these interferences
are very difficult to correctfor. EPA intends to study
and evaluate several new calibration procedures
and to amend Appendix D of 40 CFR Part 50 to
revise the 1  percent neutral buffered potassium
iodide (NBKI) calibration procedure or to replace it
with one or  more  of several alternatives under
consideration.  These  alternatives  are:  (1) gas-
phase  titration  (GPT)  with  excess  NO;  (2)  UV
photometry; (3) GPT with excess Oa; and (4) boric
acid  Kl.
  Commercial  instruments  for  monitoring total
nonmethane hydrocarbon (NMHC)  in ambient air
have  been tested  extensively and found to be
inaccurate at NMHC levels approaching the 0.24-
ppm  C  air  quality   standard.   At  present,
hydrocarbons in  ambient air can  be accurately
measured only by sophisticated  gas chroma-
tography.  Routine   methods  are  under  devel-
opment.
  On December 1, 1976, EPA promulgated a new
principle and calibration procedure for measuring
N02. The measurement principle is the gas-phase
chemiluminescent  reaction  of 03  and NO. Two
calibration procedures are prescribed. One  is a
GPT  procedure referenced to an NO-in-nitrogen
standard.  The other  calibration  procedure  is
referenced to an N02  permeation device. Before
the development of chemiluminescence analyses,
most  N02 data were  collected using methods
based  on variations  of the Gness-Saltzman
procedure.  For all practical  purposes,  these
methods  have   been   replaced  by   the
chemiluminescence method

-------
HEALTH EFFECTS OF OZONE AND OTHER
PHOTOCHEMICAL OXIDANTS
  In  this section, the influence  of exposure to
ozone  and  other  photochemical  oxidants  on
physiology and health is assessed, beginning with
a brief discussion  of the concept of  "threshold
pollutant concentrations" and itsapplicationtothe
protection of public health. Then, the  strength of
association between exposure to ozone and other
oxidants  and   changes  in  several  types  of
biomedical indicators is evaluated.
  For  each  type  of  indicator,  the  discussion
addressesfour main topics: (1 )thedegreetowhich
changes  in each type constitute  impairments in
public health; (2) the available scientific evidence
relating ozone  and other oxidant  exposures  to
changes  within  each  type; (3) the reliability of
existing  scientific  evidence;  and  (4)  where
appropriate, the confidence with which findings
may be attributed to ozone alone, as  opposed to
other substances or combinations of substances.
  Whenever  possible, clinical  (human experi-
mental)  and  epidemiologic  studies will  be
discussed together. However, because of the great
uncertainty   inherent  in  the  quantitative
extrapolation  of  results of  animal  studies  to
humans,  human  and animal  studies  will be
discussed separately.

Discussion of Threshold Concentrations
  The Clean Air Act directs that National Primary
Ambient Air Quality Standards be such that their
attainment  and  maintenance  shall,  in the
judgment of the Administrator, and allowing an
adequate margin of safety, be requisite to protect
the public  health.  The confidence with which  a
margin of safety can be defined  depends on the
precision  with  which  a  threshold  pollutant
concentration, (the level above which  exposure to
ozone  or other oxidants promotes impairment of
health and  below which  it  does not) can be
determined.

  In  practice,  no  single  overall   threshold
concentration for ozone or other pollutants exists.
Thresholds have been shown to vary  widely with
the population  segment studied and  the biologic
indicators measured. Also, since the great majority
of known dose-response relationships  do not show
sharp  discontinuities, it  is most unlikely that  a
discrete threshold  pollutant concentration can be
established even for a single population segment
and  a single biomedical indicator.
  Despite these limitations, the  environmental
decisionmaker may find it useful to incorporate the
concept  of  threshold  concentrations  into  the
standard-setting process. For a given population
segment,  a  threshold  concentration  may be
operationally defined  as occurring somewhere
between a concentration at which no effect on
health or 'function  has been  observed, and a
concentration at which such an effect has been
demonstrated.  In protecting public health,  the
population segments of primary concern are the
most  susceptible groups, in whom exposure to
ozone or other pollutants is most likely to promote
impairment of health. Such groups may include
those with underlying illness, the very old, the very
young,  and  the pregnant, (It  has   not  been
determined whether such susceptible population
segments differ from the most sensitive population
segments, which comprise those individuals most
likely to show measurable responses to very low
pollution concentrations.)
  Ideally, the environmental decisionmaker would
know the  pollutant concentrations with which no
adverse  effects are associated in susceptible
groups, as well as the concentrations with which
such  adverse  effects  are   unambiguously
associated. Unfortunately, knowledge in both of
these areas remains sparse. Recent experimental
studies of healthy people and animals have greatly
advanced  our knowledge of the health effects of
ozone.  However,  threshold  concentrations
deducible from  such studies may not apply to the
potentially susceptible  groups  described above.
The  opportunity to study such  groups experi-
mentally is severely limited by practical and ethical
constraints.  The relatively  few  epidemiologic
studies of these groups that  are extant have
yielded inconclusive results.

Human Studies
SHORT-TERM OXIDANT EXPOSURES
Mechanical Function Of The Lung

Assessment  Of Health  Effect.  There  is   con-
siderable  room for honest   disagreement on
whether  pollution-induced  alterations  in
mechanical  lung function constitute  bona  fide
impairments of health in and of themselves. In the
great majority of experimental studies in which
oxidant exposures have produced changes m lung
function in healthy subjects, function has returned
to normal within a few hours. Thus there is no
reason to suspect that in healthy individuals such
changes promote any measurable increase in risk

-------
of future illness. Nor does any available evidence
suggest that in healthy individuals a small change
m  ventilatory function,   unaccompanied  by
symptoms or impairment of oxygen uptake or work
capacity, would interfere with normal activity or
task performance.
  However,  three  considerations suggest  that
oxidant-associated changes in lung function may
signal impairment of public health. First, in per-
sons with  underlying respiratory illness such as
asthma, chronic bronchitis, and emphysema, even
small decrements in lung function often interfere
with normal activity. Second,  at experimental
ozone  concentrations  as  low  as  0.30  ppm,
decrements  in lung function have usually been
accompanied  by  physical  discomfort,  as
manifested in symptoms such as sore throat, chest
pain,  cough,  and   headache.  At  times  this
discomfort has been great enough to prevent the
completion of experimental protocols, particularly
when subjects have been exercising vigorously. It
appears quite likely  that  the pulmonary irritant
properties of ozone (and perhaps other oxidants)
underlie both the discomfort and the decrements
in function. Thus, at  least when associated with
ozone exposure, changes in lung  function often
represent  a  level of discomfort that, even among
healthy people,  may restrict normal activity or
impair the performance of tasks.

Summary  Of Data. Human experimental studies
have demonstrated  that  the subject's level of
exercise during ozone exposure is directly related
to the magnitude of change in lung function and
the severity  of  symptoms at  any given ozone
concentration. During exercise, subjects increase
their expiratory flow rates, andthey tendto breathe
through their mouths. These factors increase the
total dose  of ozone delivered to the lung and may
increase the depth to which it is delivered in the
respiratory tree.
  After  2  hr of resting exposure to 1470 fjg/m3
(0.75 ppm),  healthy young  adults showed small
changes in lung function in a study by Bates et al.
Folmsbee  et al. observed changes in respiratory
pattern  (increased   respiratory  frequency  and
decreased tidal volume) and reductions of vital
capacity  in  healthy  young  adults  exercising
submcximally after a similar  resting  exposure.
Immediately after 2 hr of exposure to 1470A/g/m3
(0.75 ppm) ozone,  during which they performed
intermittent  light exercise, subjects showed quite
pronounced changes  in lung function in studies by
Folinsbee  et al.  and Hazucha  et al. In subjects
exposed  under the same conditions, Folinsbee et
al. observed changes in respiratory pattern, though
not in minute  volume or oxygen uptake.
  In subjects  exercising maximally after 2 hr of
exposure to 1470 fjg/m3  (0.75 ppm) ozone and
intermittent  light  exercise,  Folinsbee  et   al.
observed decrements in maximum work load, tidal
volume,  heart  rate, and oxygen uptake.
  Folinsbee et al.  observed decreases  in  tidal
volume and maximum expiratory flow rate at  50
percent of vital capacity (V5o) in subjects exercising
submaximally  after 2 hr of exposure to 980 pg/m3
(0.5 ppm) ozone and intermittent light exercise. In
subjects   giving  no  history  of  cough,  chest
discomfort, or  wheezing  in response to allergy or
air  pollution exposure (unreactive subjects), few
changes  in lung function occurred after 4 hr of
exposure  to   980  pg/m3  (0.50  ppm)  and
intermittent  light  exercise  (Hackney  et  al.).
However, in subjects giving such a history (reactive
subjects) and receiving an identical exposure, the
same investigators observed decrements in 8 of 14
measured parameters. Interestingly, in a second
group  of  unreactive subjects,  the same   in-
vestigators observed  substantial decrements  in
lung function  after only  2  hr of exposure to 980
pg/m3 (0.5 ppm)  ozone and  intermittent  light
exercise.

  Statistically  significant changes in forced vital
capacity,  maximum  mid-expiratory flow rate, and
airway resistance in 22 young males after 2 hr of
exposure  to  0.4 ppm  ozone  and  intermittent
moderate exercise were observed by Knelson etal.
After subjects  had been exposed for 4 hr, changes
in these  parameters had increased, and several
other  flow  parameters  had  also  changed
significantly in comparison with control values.
  In studies by Hazucha and Bates, Hazucha et al.,
and Folinsbee  et al., subjects showed changes in
lung function after 2 hr of exposure to 730pg/m3
(0.37 ppm) ozone and intermittent light exercise.
Hackney  et al.  observed such changes in reactive
subjects  but  not in  unreactive subjects. In a
separate  study by Hackney and colleagues, four
Canadians and four southern Cahfornians were
exposed  in the Los Angeles area to 730 fjg/m3
(0.37 ppm) ozone for 2 hr. The Californians showed
few  changes  in lung  function, whereas the
Canadians showed  decrements  in  most
parameters measured. (At least partly because of
the  small sample sizes, no observed changes were
statistically significant.) In subjects exposed to 730
/jg/m3 (0.37ppm)ozone together with 1000A/g/m3

-------
(0.37 ppm) SOz, Hazucha and Bates observed an
effect on lung function substantially larger than
the sum of the separate effects of the individual
pollutants.
  DeLucia and Adams observed changes in lung
function  and respiratory  pattern  in  healthy
subjects exercising steadily and fairly heavily over
a 1-hr  exposure to 590 fjg/m3 (0.30 ppm) ozone.
Two of six subjects experienced such discomfort as
to prevent them from completing the experimental
protocol.
  Hazucha   observed  small  changes  in lung
function in three nonsmokers exposed for 2 hr to
490 fjg/m3 (0.25 ppm) ozone and intermittent light
exercise. No lung  function changes of note were
seen  by Hackney  et  al. even  among reactive
subjects who were similarly exposed.
  After 1 hr of exposure to 290 fjg/m3 (0.1 5 ppm)
ozone  and steady, fairly heavy exercise, subjects
observed by DeLucia and Adams showed changes
in respiratory pattern. In two of six subjects, the
same investigators noted inconsistent increases in
residual volume.

  Small but  statistically significant increases in
airway resistance, as  measured  by  plethysmo-
graphy, were observed in  two of four healthy
subjects immediately after a 1 -hr exposure to 200
/jg/m3 (0.1  ppm)  ozone. The  investigators
(Goldsmith and Nadel) did not state whether the
subjects exercised during exposure.

  Von  Nieding and Wagner reported that subjects
showed decrements in arterial oxygen pressure
(Pa02> and  airway resistance after 2 hr of exposure
to 200 A
-------
adaptation  in lung function  is of any long-term
consequence to health.

Attributability Of Effects To Ozone. Experimental
studies have shown that ozone at concentrations
observed in the ambient air can produce changes
in mechanical lung function. The epidemiologic
studies  of  Kagawa and Toyama, in which  lung
function was correlated more strongly with ozone
exposure than  with total  oxidant exposure, are
consistent with this finding. However, the degree
to which  epidemiologic  observations may be
attributed specifically to ozone remains in doubt,
since Kagawa  and Toyama were  not able fully to
separate the effects of ozone from the effects of
other environmental factors.
  Finally, the work of Hazucha and Bates suggests
that  the effect on lung function of ozone at 730
)Ug/m3(0.37 ppm) may be enhanced by an identical
concentration of sulfur dioxide. Observation of this
enhancement argues in favor of providing a margin
of safety in  a primary National Ambient Air Quality
Standard for ozone. However, this observation
does not support  a quantitative recommendation
for  a safety margin, since it has not yet been
determined whether such enhancement occurs at
lower ozone concentrations  or with  substances
other than  sulfur  dioxide.

Impairment  Of  Physical  Performance  —
Decrements  in   physical   performance   are
deterrents to personal satisfaction. In this respect,
such decrements  constitute impairment of public
health. The degree to which oxidant  exposures
may promote decrements in physical performance
has  not been determined.  However, Wayne et al.
assessed the association between hourly  oxidant
concentrations and the proportion of a high school
cross-country  team failing to improve  running
times between successive track meets in southern
California.  Over six  cross-country seasons, the
correlation  of average oxidant concentration in the
hour before the race with the proportion of runners
failing  to  improve  times  was 0.88.  The
corresponding correlation for both the first and
second  three-season  periods   was  0.945. A
correlation   this  high  denotes   a  very close
numerical relationship between two variables.
  During  the  period  studied,  hourly   oxidant
concentrations ranged from  approximately 60 to
590 Aig/m3 (0.03  to 0.30 ppm). Inspection of the
data  of Wayne  et   al. reveals  no  obvious
relationship between unimproved running  time
and  oxidant concentrations  below 200  to 290
      3 (0.10 to 0.15  ppm),  in spite of the  high
overall correlations mentioned above. Also, since
the authors did not consider ozone separately from
other oxidants, the specific contribution of ozone to
the observed results cannot be determined from
this study.
  As far as can be ascertained, no replication of the
study of Wayne eta I. has appeared in the published
literature. For three reasons, however, the data
from this study are more trustworthy than most
results of a  single epidemiologic  study. First, the
correlations between hourly oxidant concentration
and  unimproved running  time  were unusually
high. Second, as mentioned in Chapter 10 of this
document,  Herman, at the  University  of  North
Carolina, has analyzed the data of Wayne et al. as
well as data from two  additional seasons, and he
has observed results similar to those of Wayne et
al.  Third,  the  results of  Wayne  et  al.  are
qualitatively consistent with the  results of  the
experimental lung function  studies mentioned
above, especially those of Folinsbeeetal. In view of
these experimental studies, it would also appear
plausible that ozone contributed significantly to
the results of Wayne et al.

Oxidant  Effects  In Asthmatics  -  In  the United
States, there are an  estimated  6 to 8 million
asthmatics,  about  70 percent  of whom  are
estimated to live in urban areas. Thus, the number
of asthmatics who may be exposed to elevated
oxidant concentrations is substantial.
  Available  epidemiologic  evidence   on  the
relationship between oxidant  exposure  and
exacerbation of asthma is very limited. In  1961,
Schoettlin and Landau reportedthatthe proportion
of selected asthmatics in the  Pasadena  area
having attacks was significantly greater (p < 0.05)
on  days when  the  maximum  hourly  oxidant
concentration  exceeded 490 /ug/m3  (0.25 ppm)
than  on  days when   the  corresponding
concentration was below this level. However, the
proportion of asthmatics having attacks on days of
maximum hourly oxidant concentration above 250
/jg/m3 (0.13 ppm) was not significantly different
from  the corresponding  proportion when  the
maximum hourly concentration was below this
level. The   authors did  not state  the  actual
percentage of asthmatics having attacks on days in
any exposure category, though they did state that
asthma attacks tended to coincide with elevated
oxidant levels in 8 (6 percent) of 137 patients
studied.
  Because it does not present asthma attack rates,
the  Schoettlin  and  Landau  report  gives  no

-------
indication whether increases in attack rates might
have been  expected at maximum hourly oxidant
concentrations below 490 /*g/m3 (0.25 ppm). Nor
does it allow any judgments as to  the extent to
which increased attack rates might be attributable
specifically to ozone.  Despite  the  considerable
attention it has  received,  this report  should- be
considered the   preliminary  investigation  its
authors intended it to be.

Respiratory Symptoms and Headache

Assessment  Of   Health  Effect. As  mentioned
above,  increased  rates of  respiratory  symptoms
and  headache constitute  impairment of public
health.  Even when  mild, such  symptoms are
annoying. Even when reversible, they may restrict
normal activity or limit the performance of tasks.

Summary  Of Data. In  nearly all  experimental
studies  in  which ozone exposures have been
sufficient to produce changes in lung function,
most  subjects   have  reported   respiratory
symptoms.  The  most common symptoms have
been throat tickle, substernal tightness,  pain on
deep inspiration, and cough. Wheezing, dyspnea,
and  headache  have  occurred less  commonly.
Symptom  severity  has increased  with ozone
concentration and exercise. In studies of heavy
exercise, symptoms have occasionally been severe
enough  to  prevent  subjects  from  completing
experimental protocols.
  In an epidemiologic study in southern California,
Hammer et al. assessed the association between
daily maximum hourly oxidant  concentration and
rates of  chest discomfort, cough, and headache
among student nurses. Rates of each of these
symptoms, whether  unadjusted or  adjusted for
fever, began to increase in the following oxidant
concentration ranges: chest discomfort,  490 to
570 pg/m3 (0.25 to 0.29 ppm); cough, 590 to 760
/jg/m3 (0.30 to 0.39 ppm); and headache, 290 to
370 fjg/m3 (0.15 to  0.19 ppm).  Adjusted and
unadjusted rates of headache, however, were not
unequivocally elevated  below maximum hourly
oxidant concentrations of 590 to 760 /jg/m3 (0.30
to 0.39 ppm).
  Several Japanese investigators have assessed
the association  between  daily  pollutant
concentrations   and   symptom  rates   among
students. In one Japanese study,  rates of sore
throat, dyspnea, and  headache were  somewhat
higher during the summer months on  days when
the oxidant concentration exceeded 200 /jg/m3
(0.10 ppm) than on days when it did not. Over a 1 -
year period, rates of  respiratory symptoms and
headache were higher on days when the oxidant
concentration  exceeded 290  /jg/m3 (0.15 ppm)
than on days when it was lower than 200/jg/m3
(0.10 ppm). Though there is reason to believe that
the  stated  oxidant concentrations were daily
maximum hourly averages, averaging times were
not clearly presented in the report of this study.

Reliability Of Evidence. Because  of their close
correlation  with  the  results of  experimental
studies, the  results of Hammer et al. appear to be
quite reliable,  even though unconfirmed  in this
country. Their reliability is enhanced by the large
number  of  person-days  of  observation  (about
53,000) L , .compassed by the study. Though ozone
levels were  not considered m the  Hammer et al.
study, it is reasonable to hypothesize,  in view of
experimental  studies,  that   ozone contributed
substantially to observed increases in rates of
cough, chest discomfort, and headache.
  As nearly  as  can  be   determined  from
translations  of original articles,  the  Japanese
epidemiologic studies cited in this document were
appropriately designed. However, it is very difficult
to interpret their results. Also, at least at present,
the applicability of these results to any oxidant
pollution problem in the United States must be
considered very limited.
  In data  analyses, Japanese investigators (like
many U.S. investigators) have not been able fully to
separate the effects of individual pollutants, It is
conceivable  that  combinations  of  pollutants
unique to Japan were necessary to promote the
increased  symptom  rates  observed  there.
Averaging times for pollutant measurements were
not  clearly  stated in  the  Japanese studies.
Therefore, it is often impossible to draw specific
inferences as to dose and response. In addition, the
degree to which differences in Japanese and U.S.
cultural  responses   to  air  pollution  may
differentially affect symptom perception m the two
countries has not  been determined
  A problem more fundamental than any of these
specific reservations, however, is  that  Japanese
and U.S. investigators as yet have great difficulty in
exchanging  specific scientific  concepts. Until the
quality of scientific communication between these
groups of investigators  increases,  the ability to
interpret the Japanese studies and to apply them to
situations in the United States will remain severely
limited.
Oxidants And Eye Irritation - In that it is annoying
and uncomfortable, the reversible eye irritation

-------
produced by exposure to ambient photochemical
oxidants  may  legitimately  be considered  a
marginal impairment of public health. However,
whether such eye irritation is sufficient to impair
performance or restrict normal activity has not
been  determined.  Nor  has  any  association
between  oxidant-mediated  eye irritation  and
chronic eye disease been observed.
  In epidemiologic studies, no symptom has been
more consistently linked to oxidant exposure than
eye irritation. In most studies reported before the
publication in 1970 of Air  Quality Criteria for
Photochemical Oxidants,  rates of eye irritation
were  observed to increase fairly steadily when
oxidant  concentrations ranged from  200 to  880
/ug/m3 (0,10 to 0.45 ppm). Hammer et al. found that
rates  of eye discomfort  began  to  increase  at
oxidant concentrations of 290 to 370 /ug/m3 (0.15
to 0.19 ppm).
  Evidence linking ambient photochemial oxidant
exposures to eye irritation is convincing. However,
the specific  etiologic agent or  agents remain
unknown. Experimental studies have shown quite
conclusively that ozone at ambient concentrations
is not an eye irritant.

Oxidants and Mortality - Review of existing studies
shows no consistent  association between daily
oxidant  concentrations and daily mortality rates.
As far as can be ascertained, no studies of oxidant
exposures  and mortality have been  performed
since the publication in 1970 of Air Quality Criteria
for Photochemical Oxidants.

Other Effects Of Short-Term Ozone  And Oxidant
Exposure

Changes  In  Erythrocytes.  In  an  experimental
study, Buckley et al. observed  increased rates of
erythrocyte lysis in HZ02 after a 2%-hr exposure of
healthy subjects to 980 jug/m3 (0.5 ppm) ozone and
intermittent light  exercise.  These  investigators
also noted  changes  in the" activity  of several
erythrocytic enzymes. Hackney et al. observed an
increased in vitro lysis rate in erythrocytes of Cana -
dians, but not in that of Californians following a 2-
hr exposure of subjects to  730 (ig/m3 (0.37 ppm)
ozone and intermittent light exercise.

Chromosomal Aberrations. Merz et a I. repdrted
chromosomal abnormalities in the lymphocytes of
six  subjects after they were exposed to 980/ug/m3
(0.5 ppm) ozone for 6 or 10hrs. However, a study by
McKenzie et al.  showed  no  increased rate of
leukocyte  chromosomal  aberration  in   the
lymphocytes of 30 subjects exposed for 4 hr to 780
£fg/m3 (0.4 ppm) ozone. In the few epidemiologic
studies of  oxidant  exposure and  chromosome
morphology performed to date, factors other than
differences in oxidant exposure have confounded
observed results to the extent that no inferences
can be drawn.

Biochemical Parameters.  Buckley et al. observed
reduced  glutathione  reductase  activity  and
increased vitamin E  and lipid peroxidation  in
human serum following exposure of subjects to
980 fjg/m3 (0.5 ppm) ozone and intermittent light
exercise for 2% hr. Following a 2-hr exposure to
730 ^g/m3 (0.37 ppm) ozone and intermittent light
exercise, the Canadians and Californians studied
by  Hackney et al. showed increases  in serum
vitamin E levels; but only in the Canadians were
these increases statistically significant.
  After subjects exercised steadily and vigorously
throughout a 1-hr exposure to 590 /jg/m3 (0.30
ppm)  ozone,  DeLucia  and Adams  observed no
changes  in  the  following  biochemical  blood
parameters!  hemoglobin   level,   nonprotein
sulfhydryl level, erythrocyte glucose-6-phosphate
dehydrogenase activity, and glutathione reductase
activity.

Clinical Significance. The clinical significance of
ozone-mediated changes observed  in studies of
blood is not yet known. As yet, an epidemiologic
study of oxidant effects on anemic individuals has
not been done. Whether or not oxidant exposures
promote  changes  in  leukocyte  chromosome
morphology,  the  significance  of the  changes
themselves is unknown. Finally, changes in serum
parameters  of  the  magnitude  observed  in
experimental ozone studies  have not yet been
linked to  any clinical diseases.
LONG-TERM OXIDANT EXPOSURES
  With the exception of the Hackney et al  study of
Californians  and Canadians,  no  experimental
studies of humans have as yet assessed the effects
of long-term oxidant exposures. The few available
epidemiologic  studies of such  exposures have
yielded inconclusive results,  fvlahoney has  ob-
served an association  between broad patterns of
oxidant distribution and annual respiratory disease
mortality in the Los Angeles area.  However, no
convincing   association   between  lung  cancer
mortality and oxidant exposure has been shown. In
some studies, a limited association between  the
frequency of chronic obstructive lung disease and
oxidant exposure has been  observed.  In  other
studies, however, no such association has been

-------
apparent.  No  relationship between  long-term
oxidant  exposure and acute respiratory disease
incidence  or change in  lung  function has been
observed, though neither of these areas has been
extensively investigated.
  Most  available epidemiologic studies of long-
term oxidant exposure are difficult to interpret. As
usually acknowledged by the authors themselves,
factors other than  pollution exposure may often
have  influenced the  results observed. Some
studies  have proved difficult to evaluate because
areas in which health variables were compared did
not  show  clear-cut differences  in  pollution
exposure.  Thus human studies cannot yet provide
the environmental decisionmaker with concrete
information as to the effects of long-term oxidant
exposures on public health,
Animal  Toxicology Studies
ANIMAL INFECTIVITY EXPERIMENTS
  Increased susceptibility of animals to bacterial
infection following ozone exposure at 196 fjg/m3
(0.1  ppm) is described  by several investigators
(Coffin et a!., 1968; Ehrlich et al., 1976; Gardner et
al , 1974;  Coffin and Gardner, 1 972; Miller et al.,
1978).  Others (Goldstein  et  al., 1971a,  1971b;
Goldstein  and Hoeprich, 1972; Goldstein et al.,
1974), using bacterial infection,  have developed
indices  of infection for measuring the effects of
ozone on the  lungs of rodents. The effective
concentration  of ozone at which susceptibility to
infection is increased is lower when  either other
pollutants  or  stress is combined  with ozone
exposure. These findings have  definite  human
health  implications,  although different exposure
levels may  be associated  with  such effects in
humans. These reactions in mice represent effects
on basic biological responses to infectious agents,
and there  is no reason to believe that the pollutant-
induced alterations of basic defense mechanisms
that occur in mice could not occur in  humans who
possess  equivalent  defense  systems.   The
extrapolation of these data to man is not supported
by direct epidemiologic evidence that susceptibility
to infection increases in persons exposed to ozone
and other photochemical materials. However, the
biochemical and  cellular alterations  described
below for animals suggest that multiple epithelial
and  biochemical targets are  perturbed by ozone
exposure. Ozone-induced irritation  of the major
bronchi in man does occur at ozone concentrations
approximating 490/ug/m3 (0.25 ppm). Fortracheal
ozone concentrations greater than  100 /L/g/m3
(0.05 ppm), it  is predicted that a  smaller inspired
concentration would  be  required in man than in
rabbits  and guinea  pigs  to  result in a  given
respiratory bronchiolar dose (Miller et al., 1 977). A
3-hr exposure to 490 ^g/m3 ozone (0.25 ppm) has
been shown to be sufficiently high to injure rabbit
alveolar macrophages. Thus,  depending on the
extent of nasopharyngeal removal of ozone in man,
a given  level may be expected to produce similar
effects in man and certain animals.

MORPHOLOGICAL ABNORMALITIES OF ANIMAL
RESPIRATORY SYSTEMS
  A range  of  morphological effects is noted in
association with experimental ozone exposures of
392 to 1960 fig/m3 (0.2 to 1.0 ppm). These effects
vary from  the  reversible replacement  of Type 1
with Type 2 alveolar cells (e.g., 392 ^g/m3  or 0,2
ppm ozone  for 3 hr/day over 7 days) to emphyse-
matous  changes and terminal  bronchiolar and
alveolar damage (e.g.,  784 fjg/m3, or 0.4 ppm,
ozone for 6 hr/day, 5 day/wk, for 10 months; and
1058 yug/m3, or 0.54 ppm ozone continuous for 3
months). Occurrence of these effects after long-
term exposure to low concentrations increases the
expectation  that repeated or  chronic  exposures
may have the potential for inducing similar effects
in humans.

BIOCHEMICAL EFFECTS
  There is  an  impressive  variety of biochemical
alterations  associated with ozone exposures over
the range of 1 96 to 1960 ^g/m3 (0.1 to 1.0 ppm).
Though  effects caused by levels of 980/Kj/m3 (0.5
ppm) and greater have definite toxic potential, the
biological significance of those changes detected
in exposures to 196 to 392/ug/m3 (0.1 to 0.2 ppm)
are more difficult to assess. These biochemical
changes   are  significant,   however,  in
demonstrating  that  there  are ozone-induced
effects  at  cellular sites  and  in organ systems
distant from the lung. Biological reactions to ozone
include  an increase in the activity of enzymes that
detoxify oxidizing  chemical  species.  Increased
enzymatic  activity, as well as  increased oxygen
consumption, has been shown in some instances
to  be   prevented,  reversed,  or diminished  by
increased   vitamin  E levels  (vitamin  E  is  an
antioxidant). Thus increased  enzymatic activity
may represent a reaction to potential toxicity rather
than being a toxic response itself. Nevertheless,
such reactions represent the organism's response
to ozone. Such perturbations in biological systems
may  pose   a   health  risk to   the  population,
particularly for susceptible individuals.
 10

-------
GENETIC AND TERATOGENIC POTENTIAL
  A  report  by Zelac (1971)  on chromosomal
abnormalities  in peripheral leukocytes  of intact
hamsters which were exposed to 392 pg/m3 (0.2
ppm) ozone for 5 hr was not confirmed by Gooch et
al. (1976). Although  confirmation of the results
may  suggest another category of ozone-induced
effects, the preliminary evidence does not suggest
that these effects occur at lower levels of exposure.

IMPLICATIONS OF ANIMAL TOXICOLOGY DATA
  Note that the difference between concentrations
of ozone  that produce  toxicological effects  in
animals (as well as symptomatic and lung function
changes  in  humans) and  ambient air  levels  of
ozone is much smaller than for nearly any other
atmospheric  pollutant.  There  is  an  unusual
clustering   and  convergence  of  various
toxicological,  experimental  human,  and
epidemiologically  observed  human  effects  in
association  with  ozone  concentrations ranging
from 392 to  11 76 fjg/m3 (0.2 to 0.6 ppm).
  Chronic effects of long-term ozone exposure in
man cannot be quantitatively related to specific
ozone concentrations of  short  (hourly  or daily)
averaging times because of the long period  of
disease  induction and the varied exposures  of
individuals  during  the  period  of  induction.
Epidemiologic studies may establish relationships
between long-term ozone exposure and the risk of
human chronic disease, but toxicological studies
must be relied on in quantifying the ozone levels
that may induce chronic effects.
Ozone Versus Oxidant Health Effects
  Two  characteristics  of  ozone  and   oxidant
exposures should be cited with reference to public
health: (1) ozone itself is a primary cause of most of
the health effects reported in toxicological  and
experimental human  studies, and the evidence for
attributing many health effects to this substance
alone is very compelling; and (2) the complex of
atmospheric photochemical substances  is known
to produce  health effects,  some  of which  (eye
irritation, for example) are not attributable to pure
ozone but may be caused by other photochemical
substances in combination with ozone.
EFFECTS OF PHOTOCHEMICAL OXIDANTS
ON  VEGETATION  AND  CERTAIN   MICRO-
ORGANISMS

  Since injury to vegetation by oxidants  was  first
identified in  1944 in  the Los Angeles Basin, our
understanding of  oxidant  effects and of  the
widespread  nature  of  their  occurrence  has
increased  substantially. The  major phytotoxic
components of the photochemical oxidant complex
are ozone and peroxyacetylnitrate (PAN), although
some data suggest that other phytotoxicants are
also present. The peroxyacylnitrates are the most
phytotoxic of the known photochemical oxidants;
but because ozone is ubiquitous  and associated
with widespread injury to vegetation, it is the most
important  phytotoxic  component  of  the
photochemical oxidant complex.
  The  effects of  photochemical  oxidants  on
vascular plants  occur at several  levels, ranging
from the subcellular to the organismic, depending
on the concentration and duration of exposure to
the pollutant and the interval between cessation of
exposure and examination of the plant.
  The  earliest  effect  is  an  increase  in  cell
membrane  permeability. Following that, cellular
and  biochemical  changes  take  place that are
ultimately expressed  on  the organismic level as
visible foliar injury, increased leaf drop, reduced
plant vigor, reduced plant growth, and death. Such
biochemical  modifications in an individual plant
are manifested by changes in plant communities
and, finally, in whole  ecosystems.
  Leaf stomata are the principal sites through
which ozone and PAN enter plants. Oxidants affect
photosynthesis,   respiration,  transpiration,
stomatal  opening,  and  metabolic  pool
development, as well as biochemical pathways
and enzyme systems.
  Visible  injury  is identifiable  as   pigmented,
chlorotic, or  necrotic  foliar  patterns.  Metabolic
cellular  disturbances can occur  without  visible
injury and may be reversible. However, most of the
growth  effects  reported  until  recently  were
accompanied by visible injury.
  Classic  ozone  injury is  demonstrated  by the
upper-surface leaf fleck of  tobacco and the leaf
stipple of grape.  Many  plants show an  upper-
surface response with no associated injury to the
lower surface  of  leaves.  However,  in  mono-
cotyledonous plants such as grasses or cereals,
and in some non-monocotyledonous plants, there
is no division of  mesophyll tissue,  and bifacial
necrotic spotting (flecking) is a common symptom
of ozone injury.
  Coniferous trees exhibit  different symptoms.
Ozone is probably the cause of emergence tipburn
in eastern white pine (white pine needle dieback)
and  of  chlorotic  decline,  a  needle injury of
ponderosa pine
                                                                                          11

-------
  Classic PAN injury appears as a glaze followed
by bronzing of the  lower  leaf surface of many
plants. Complete collapse of leaf tissue can occur if
concentrations  are  sufficiently  high.  Early  leaf
senescence and abscission  usually  follow the
chronic symptoms. Patterns of chronic injury are
generally not characteristic and may be confused
with symptoms caused by  biotic diseases, insect
infestation,  nutritional   disorders,   or  other
environmental stresses,
  A great deal of research has been done to define
more accurately the effects of oxidants on plant
growth and yield. Studies comparing the growth of
plants in field chambers  provided  with carbon-
filtered  or   nonfiltered ambient air  containing
oxidants have reported up to 50-percent decreases
m yield of  citrus (orange and lemon) exposed  to
oxidants; a  10- to 1 5-percent suppression in grape
yield  in the first year and a  50- to  60-percent
reduction over the following 2 years; and a 5-  to
29-percent decrease in yield of cotton lint and seed
in California. Losses of  50 percent  in some
sensitive potato, tobacco,  and soybean cultivars
have been reported in the eastern United States, It
is apparent that oxtdants in  the ambient air reduce
the yields of many sensitive plant cultivars.
  Experimental  chambers  with controlled
environments have been used to study both short-
term and long-term  effects of exposure; to ozone
(Tables 11-2, 11-4,  Chapter 11). Radishes given
one, two, or three acute exposures (785 /jg/m3, or
0.40ppm)of 1.5hr each at 7,14, and/or 21 days of
age  exhibited  reductions  in  root  growth.  The
reductions  in root growth from the multiple ozone
exposures were equal to the additive effects of the
single exposures, In other words,  the temporal
distribution was not a significant factor. When soy-
bean plants were exposed to 1468 /jg/m3 (0 75
ppm) ozone for 1 hr, root growth was consistently
reduced more than  top growth. There were also
reductions  in nodule weight and  number. The
greater reduction of root growth compared to top
growth is related to the transport of photosynthate.
Ozone also affects nitrogen  fixation  in  clover,
soybean, and pinto bean  through reduction  in
nodule  number, even  though nodule size and
efficiency  of nitrogen fixation are not influenced.
The effect  of ozone on the number of nodules
formed by  legumes, if widespread, could have a
major impact on  plant communities  and  could
affect  fertilizer   requirements.  There  are
indications that the  effect  of ozone  on  nodulation
may be related to the carbohydrate supply in the
host plant.
  Experimental long-term exposures to ozone of a
variety of crops,  as well as of ornamental and
native  plants, have resulted in a  reduction  in
growth  and/or  yield.  Exposure of 14 species
representative of the aspen  plant community to
ambient air containing  98 to 1 37 /jg/m3 (0.05 to
0.07 ppm) ozone and to ozone  concentrations of
290 and 588 /jg/m3 (0.15 and 0.30 ppm) for  3
hr/day, 5 days/week, and to charcoal-filtered air
throughout the growing season, resulted m foliar
injury  to  all  species   at the  highest  pollutant
concentration.  The  growth   of  two  soybean
cultivars (Hood  and   Dare) was  inhibited by
intermittent exposure to ozone at 196 vg/m3 (0.10
ppm) for 3 weeks. Both root and top growth were
decreased. Similar results were noted with radish,
except  that a lower concentration of ozone (98
/jg/m3,  or 0,05 ppm) inhibited growth. In these
studies, the reduced growth occurred even though
there were few visible symptoms of plant injury.
  A 30-percent reduction in the yield  of wheat
occurred when wheat at anthesis was exposed to
ozone at 392 /vg/m3 (0.2 ppm) for 4 hr/day for  7
days. A significant reduction in the yield  of tomato
was noted  when plants were experimentally
treated with ozone at 686/jg/m3(0.35 ppm);fewer
fruit set, and thus fewer fruit were harvested.
Chronic exposures to ozone at 98 to 290 /jg/m3
(0.05 to 0.15 ppm)for 4  to 6 hr/day reduced yields
in soybean and corn grown under field conditions.
The threshold for measurable  effects for ozone
appears to be between  98 to 196 /jg/m3 (0.05 to
0.1 0 ppm) for sensitive plant cultivars. This is well
within the range of ozone levels monitored in the
eastern United States. Growth orflowering effects
were reported for carnation, geranium, radish, and
pinto bean grown in greenhouse  chambers and
exposed to ozone at 98 to 294 fjg/m3 (0,05 to 0.15
ppm) for 2 to  24 hr/day.
  The  two most  critical  factors in determining
plant response to air  pollution are duration  of
exposure and concentration of pollutants. These
two factors describe exposuredose. Indetermining
the  response  of  vegetation  to oxidants,
concentration is more important than time.
  The  concept  of  limiting values  was used by
Jacobson to define a boundary between doses of a
pollutant that  are  likely to  injure vegetation
measurably and  those  that are  not. Foliar injury
was used as the index of  plant  response. The
ranges for limiting values for effects of ozone are:
    1.  Trees  and shrubs -
          400 to 1,000 fjg/m3 (0.2 to 0.51 ppm)
              for 1 hr
 12

-------
          200 to 500 ^g/m3 (0,1 to 0.25 ppm) for
              2 hr
          120 to 340 pg/m3 (0.06 to 0.17 ppm)
              for 4 hr
    2. Agricultural crops -
          400 to 800 /jg/m3 (0.2 to 0.41 ppm) for
              0.5 hr
          196 to 500 ^g/m3 (0.1 to 0.25 ppm) for
              1 hr
          75 to 180 A/g/m3 (0.04 to 0.09 ppm) for
              4 hr
    Limiting values for PAN are:
      1000 fjg/m3 (0.2 ppm) for 0.5 hr
       500 fig/m3 (0.1 ppm) for 1  hr
       175 fjg/m3 (0.035 ppm) for 4 hr
  Doses of ozone or PAN greater than the upper
limiting values are  likely to cause foliar injury.
  The data points used to determine the limiting
values listed above are not necessarily threshold
values,  but  are based  on available  published
research  data. More than 200   studies were
surveyed.  Any limitations that were present in the
experimental techniques used in the studies are
therefore expressed in the data points. The number
of studies used to derive the data points for ozone
exposure of trees and shrubs (19)  and for PAN is
another limitation on the values given above.
  For agricultural  crops,  the  inaccuracies  m
measurements make the interpretation of results
of repeated long-duration exposuresdifficult.Thus
limiting  values for ozone concentrations  below
100 /jg/m3 (0.05 ppm) are not useful.
  An  ozone concentration  of 98  to 137 ^g/m3
(0.05  to 0.07 ppm) for 4 to 6 hr/day for 1 5 to 133
days  can  significantly inhibit plant growth and
yield of certain species (Table 11 -4, Chapter 11).
  Plant sensitivity to ozone and PAN is conditioned
by many factors, Genetic diversity  in sensitivity to
ozone between  species  and between cultivars
within a species is well documented. Variations in
sensitivity to ozone within a natural species are
well known for  several pine species, including
white, loblolly, and ponderosa. Plant sensitivity to
oxidants can  be changed by both  climatic and
edaphic  factors.  A  change  in  environmental
conditions can initiate a  change in sensitivity at
once, but  it will be  3 to 5 days before the response
of the plant is totally modified. Plants generally are
more sensitive to ozone when grown under short
photoperiods, medium light, medium temperature,
high humidity, and high soil moisture. Injury from
PAN  may increase  with an  increase  in light
intensity.  Conditions during exposure and growth
affect the response of plants to oxidants in similar
ways.  In  general,  environmental  conditions
optimum for  plant growth  tend to increase the
sensitivity to  ozone.  Factors that increase water
stress at the time of exposure tend to make plants
more tolerant to ozone. Soil moisture is probably
the  most  important environmental  factor that
affects  plant response  to  oxidants  during the
normal  growing season.  Physiologic  age affects
the response of the  leaf  to oxidants. Young leaf
tissue is most sensitive to  PAN, whereas newly
expanding and maturing tissue is most sensitive to
ozone. Light is required for plant tissue to respond
to PAN;  a similar light requirement is not needed
for plants to respond to ozone.
  The  majority  of  effects  observed,  such  as
suppression of  root growth, mineral uptake, and
nitrogen  fixation,   apparently  result  from  a
suppression of  photosynthesis  and modifications
in photosynthate distribution. This suppression of
metabolic reserves ultimately slows plant growth
and  renders  the plant more sensitive to other
stresses. Physiological changes can  provide a
sensitive means of.monitoring the health and vigor
of the plant with or without visible injury. Ozone
affects pollen germination  in some species and
may affect yield through incomplete pollination of
flowers. Investigations with Arabidopsis that/ana
showed no mutagenic effects from  ozone over
seven generations.
  Mixtures of ozone and SOa  can cause effects
below the  levels  caused by  either  gas  alone;
however, there  is some disagreement concerning
the interactions of ozone with other gases. Ratios
of  gas ' mixtures,  intermittent exposures,
sequential  exposures  to pollutants, and pre-
disposition by one pollutant to the effects of a
second  pollutant may be  important  factors in
nature, but insufficient knowledge is available for
elucidation of the effects.
  The response of  plants  to  oxidants  may  be
conditioned by the presence or absence of biotic
pathogens. Depending  on  the  plant  and the
pathogen, oxidants may cause more or less injury
to a given  species. Oxidant injury to ponderosa
pine predisposes the trees to later invasion by bark
beetles. Ozone and ozone/sulfur dioxide mixtures
can  decrease  the  population  of some  plant-
parasitic nematodes. Variable  plant  responses
were noted when herbicides were used in the
presence of high oxidant  concentrations.

   Little  research on the effects of oxidants  on
ferns,  nonvascular  green  plants, and  micro-
organisms has been reported. Lichens and mosses
                                                                                            13

-------
are responsive to  acid  gases,  but there  is no
definite evidence that they respond to  oxidants
Ferns may be especially sensitive, but their injury
response is different from that of higher vascular
plants.  Growth  and  sporulation  of fungi on
surfaces are usually, but not always,  affected.
Ozone from  0.1 to  several milligrams per liter of
solution is required to kill many microorganisms in
liquid media. Most work with microorganisms has
been done to study the effectiveness of ozone as a
biocide  in the storage  of vegetation or in the
treatment of water or sewage.

EFFECTS OF PHOTOCHEMICAL OXIDANTS
ON ECOSYSTEMS
  Plants, animals, and microorganisms usually do
not live alone but exist as populations. Populations
live  together  and  interact  as   communities.
Communities,  because of the interactions of their
populations and of the individuals that make them
up, respond to pollutant stress differently from
individuals.  Man  is  an  integral  part  of  these
communities, and as such, he is directly involved in
the complex ecological  interactions that  occur
within the communities and within the ecosystem
of which the communities are a part.
  The stresses placed on the ecosystems and their
communities can be far-reaching, inasmuch asthe
changes that occur  may be  irreversible. For
example, it has been suggested that the arid lands
of India are the result of defoliation and elimination
of vegetation, which in turn induced local climatic
changes  that  were   not  conducive to  the
reestablishment of the original vegetation.
  An ecosystem (e.g., the planet earth, a forest, a
pond,  or a fallen log) is a major  ecological unit
made  up of living (biotic) and  physical (abiotic)
components through which the cycling  of energy
and  nutrients occurs. A structured  relationship
exists among  the various components. The biotic
units   are   linked  together  by  functional
interdependence, and the abiotic units constitute
all of the physical factors and chemical substances
that interact with the biotic units. The processes
occurring within the biotic and abiotic  units and
the interactions among them can be influenced by
the environment.
   Ecosystems tend  to  change   with   time.
Adaptation,  adjustment,  and  evolution  are
constantly taking place  as the biotic component,
the  populations,  and the communities of  living
organisms interact with the abiotic component  in
the   environment.  Recognizable  sequential
changes  occur.  With   time,  populations  and
communities  may  replace  one  another  This
sequential   change,  termed  succession,  may
culminate  in  climax  communities.  Climax
communities are structurally complex, are more or
less stable, and are held m a steady state through
the operation of a particular combination of biotic
and abiotic factors The disturbance or destruction
of a climax community or ecosystem results in its
being returned to a simpler stage. Existing studies
indicate   that  changes  occurring  within
ecosystems, in  response to  pollution or other
disturbances,  follow definite  patterns that  are
similar even in different ecosystems. It is therefore
possible  to predict broadly  the  basic  biotic
responses to the disturbance of an ecosystem.
  Diversity and  structure are  most changed by
pollution as a result of the elimination of sensitive
species of  flora and fauna and of the selective
removal of the  larger overstory plants in favor of
plants of small stature. The result is a shift from the
complex forest community   toward the  less
complex hardy shrub and herb  communities. The
opening  of the forest  canopy changes  the
environmental stresses on the forest floor, causing
differential survival  and, consequently, changed
gene frequencies in  subcanopy species.
  Associated with the reduction in diversity and
structure  is  a  shortening  of food  chains,  a
reduction  m the total nutrient inventory, and  a
return to a  simpler successional stage.
  It should be  emphasized that ecosystems are
usually being subjected to a number of stresses at
the same time, not just a single perturbation such
as oxidant pollution.
  The effects  of oxidants on  the mixed-conifer
forest  of  the  San   Bernardino  Mountains
graphically demonstrate the changes that occur in
natural ecosystems as discussed above. Since the
early 1940's, the San Bernardino Forest has been
undergoing stress from oxidants transported long-
range from Los Angeles, 144 km (90 miles) away.
Losses  of  ponderosa   and  Jeffrey  pines,  the
overstory vegetation, have increased dramatically
as pollutant levels have  risen. Black oak has also
suffered oxidant injury.  The composition of both
plant and animal populations has been altered by
the death of the ponderosa and Jeffrey pines.
  The interaction of pollutant and inversion layers
at the heated mountain slope results in the vertical
venting of oxidants over the mountain crest by up-
slope  flow, thus establishing  an  elevational
gradient of  oxidant  concentrations. Oxidant
concentrations  ranging  from 100 to 200 //g/m3
(0.05 to 0.10 ppm) at altitudes as high as 2432 m.
 14

-------
approximately 1033 m above the mountain crest,
have been measured by aircraft.
  Total  oxidant  concentrations  in  the  San
Bernardino  Mountains  have  been  measured
continuously from May through September since
1968  at the  Rim Forest-Sky Forest monitoring
station.  During  each of  the  first 7 years  of
monitoring, between June and September, the
total number of hours in which concentrations of
ozone were 160 jug/m3  (0.08 ppm) or more was
never less than 1300, The number of hours  in
which the total  oxidant concentration was 390
/jg/m3 (0.20 ppm) or higher increased from fewer
than 100 in 1969 to nearly 400 in 1 974. It was not
uncommon to observe momentary  oxidant peaks
as high as 1180/jg/m3 (0.60 ppm). Theduration of
oxidant  concentrations  exceeding  200  /jg/m3
(0.10 ppm) was 9, 1 3, 9, and 8 hr/day going from
the lower- to  the higher-altitude stations.
  The most recent data firmly indicate that oxidant
concentrations in the San  Bernardino Forest will
either increase annually or oscillate around the
mean  of  present high concentrations  in the
foreseeable future.
  The transport of the urban plume from the coast
northeastward to the mountains can be readily
demonstrated.  Because of this  transport, the
permanent  vegetation constituting  natural
ecosystems  receives   much  greater  chronic
exposure,  while  the  short-lived  vegetation
constituting   the economically  more  valuable
agroecosystem of the Los  Angeles coastal plain
can  be  subject to  injurious  doses,  but  in
intermittent,  short-term fumigations.  Each
situation has  measurable economic and aesthetic
consequences, but on different time scales. The
single-species agricultural  ecologic system (the
agroecosystem) has little  resilience to pollutant
stress.  Losses are  sometimes immediate and
occasionally catastrophic.  The complex natural
ecosystem is initially more resistant to pollutant
stress, but the longer chronic exposures cause
disruption of  both structure and function in the
system that may be irreversible.
  The oxidant injury to the mixed conifer stands of
the San Bernardino Mountains that began in the
early 1940's, as indicated above, is well advanced.
A similar problem is developing in the forests of the
southern Sierra  Nevada Mountains.  Both areas
show  direct as  well as indirect  effects on all
subsystems of the forest ecosystem  producers,
consumers, and decomposers
  In summary:
    1.  Ozone injury  has limited the growth  and
       caused the death of ponderosa and Jeffrey
       pines. An estimated  1.3 million trees have
       been affected. Decrease in cone production
       has resulted in a decrease in reproduction.
       Black oak has  also  suffered injury from
       ozone.
   2.  Reduction in fruits and seeds that make up
       the diet  of most of the  common small
       mammals has influenced the populations
       of these  organisms.
   3.  Essential processes, such as recycling of
       nutrients,  may have been  disrupted,
       causing  a  limitation  in  the  growth of
       vegetation.
   4.  Death of the predominant vegetation  has
       caused  an  alteration in  the  species
       composition and a change  m the wildlife
       habitat.

  The San Bernardino Mountain study illustrates
the  complexity of  the problems  caused  by
environmental pollution. The changes that have
occurred in this mountain ecosystem as the result
of oxidant transport  have already  influenced the
importance and value of this natural  resource to
the residents of Southern California.

  The injury to the  eastern white pine in  the
Appalachian Mountains  caused  by  oxidant
transport  from the urban northeast has begun a
similar sequential change thai could degrade  this
important recreational area.  Total oxidant peaks as
high as 220 /jg/m3 (0.11 ppm) were recorded for
July 1975. Concentrations exceeding 160 /jg/m3
(0.08 ppm) were measured  in June 1976. These
episodes  resulted m  significant  increases in
oxidant injury to three categories of eastern white
pine in the Blue Ridge Mountains (the eastern
range of the Appalachian Mountains).
  Ecosystems are  usually evaluated by  modern
man solely on the basis of their economic value.
This economic value in turn depends on the extent
to which man can manipulate the ecosystem for
his  own purposes. This single-purpose point of
view makes it difficult to explain the many benefits
of a natural ecosystem to ma n's welfare m terms of
the  conventional cost-benefit analysis. Gosselink,
Odum, and Pope have, however, placed a value on
a tidal  marsh by assigning monetary values to its
multiple contributions to man's welfare such as
fish  nurseries,  food  suppliers,  and  waste-
treatment  functions of the marsh. They  estimate
                                                                                          15

-------
the total social value to range from $50,000 to
$80,000 per acre.
  Westman also evaluated the benefits of natural
ecosystems by  estimating  the  monetary  costs
associated  with  the  loss of  the free services
(absorption of air pollution, regulation  of global
climate and radiation balance, and soil bioding)
provided by the ecosystems. Westman estimated
that the oxidant  damage  to the San  Bernardino
National Forest could result in a cost of $27 million
per  year  (1973  dollars)  just  for  removal  of
sediments  resulting from  erosion, as  long as the
forest remained  in the early stages of succession.
  Estimates of the cost in currency of the values of
items and qualities such as clean air and water,
untamed wildlife, and wilderness, once regarded
as  priceless, are an attempt to rationalize the
activities of  civilization.  When  estimating  the
monetary  cost  in currency  of  the values  lost
through damage to ecosystems, the assumption is
usually made that the decisionmakers will choose
the alternative that is most socially beneficial as
indicated by  costs  compared  to benefits.  As
Westman  points  out,  the  assumption  that
decisions  maximizing  benefit/cost  ratios
simultaneously optimize social equity and utility is
based  on certain inherent corollaries:

 (1) The human species has the exclusive right  to use and
 manipulate  nature for its own purposes (2) Monetary units
 are socially acceptable  as means to equate the value of
 natural resources  destroyed and those developed (3) The
 value  of services  lost during  the interval before the
 replacement or substitution of the usurped  resource has
 occurred is included in the cost of the damaged resource f4)
 The amount of compensation in monetary units accurately
 reflects the full value  of the loss to each loser  m the
 transaction. (5) The value of the item to future generations
 has been judged and included man accurate way m the total
 value  (6) The benefits of development accrue to the same
 sectors of society, and in  the same  proportions, as the
 sectors on whom the costs are  levied,  or  acceptable
 compensation  has been  transferred  Each of  these
 assumptions, and others not listed, can and  have been
 challenged

In the case of (4) above,  for example, the  losses
incurred when developing natural ecosystems are
involved affect  species other  than man.  These
losses  are seldom, if  ever,  compensated  The
public at large  also is usually  not consulted to
determine whether  the  dollar  compensation is
adequate and acceptable. Frequently, there is no
direct  compensation. Corollary  (5) can never be
fulfilled because it  is  impossible  to  determine
accurately the value to future generations.
  Evaluating  the  contribution  of functioning
natural ecosystems  to  human welfare is a very
complex  task  and  involves  weighing  both
economic and human social values. As life support
systems, they should not be evaluated in economic
terms.
  With the  passage of  time, man has destroyed
many  of the naturally  occurring ecosystems of
which he  was a part and has replaced them with
simplified ecosystems wholly dependent on his
care and protection and requiring a large input of
energy.
  Man favors the simple  unstable  and synthetic
ecosystems  because when extensively managed,
and subsidized  by the use of fossil fuels, they are
highly  productive.   An  agricultural  ecosystem
(agroecosystem) is an example of such a simplified
ecosystem.  The   effects  of  oxidants  on
agroecosystems have been under study for more
than  20 years.  The study of effects on  natural
ecosystems  is much more recent.
  Plants grown  in  agroecosystems are largely
annuals and can be  replaced when  they  are
susceptible to pollutant stress. Natural ecosystems
remain in  place  year after  year.   Manmade
pollutants are  undoing relationships  developed
within these ecosystems over millions  of years.

EFFECTS OF  PHOTOCHEMICAL OXIDANTS
ON MATERIALS
  Ozone  is  a   major   factor  in  the  overall
deterioration of several  different types of organic
materials.  In  fact,  certain  specific  organic
compounds  are more  sensitive to ozone attack
than are humans or animals. The  magnitude of
damage is difficult to assess  because ozone is one
of many oxidizing chemicals  m the atmosphere
that  contributes to  the  weathering of  materials.
Nevertheless, researchers have shown that ozone
accelerates the deterioration of several classes of
materials, including elastomers (rubber), textile
dyes and  fibers, and certain types  of paints and
coatings.
  Although  many organic materials  have been
shown  to be susceptible to ozone attack,  only
certain paints,  elastomers,  and  dyes  sustain
damage representing significant economic  loss.
Even  the  measures to prevent ozone damage to
elastomers and dyes constitute a major cost.
 16

-------
                                 2.  INTRODUCTION
  Air  quality  criteria  are expressions  of  the
scientific knowledge  of the relationships  among
various  concentrations,  averaged  over an
appropriate time period, of air pollutants in the
atmosphere and their adverse effects on  human
health and the environment. Criteria are issued to
assist in the formulation of decisions regardingthe
need for control of a pollutant and the development
of air quality standards governing the pollutant. Air,
quality  criteria  are  descriptive;  that  is,  they
describe the effects that have been observed when
the ambient air concentration of a pollutant has
reached  or  exceeded  a  specific  figure for  a
specified period of time.
  Many  factors  must be   considered  when
developing criteria. Consideration must be given to
the chemical  and physical characteristics of the
pollute its, the techniques available for measuring
these characteristics,  and the exposure time,
relative  humidJty,   and  other  environmental
conditiDns,  Natural and anthropogenic emissions
must 13 considered and assessed. The number and
distrib ition of sources and the extent to which the
specif'.; pollutant is emitted into the environment
direct!" affect the levels to which  the receptor is
expose d. In addition, the criteria must also  include
consid sration of the influence of all such variables
on the effects of air pollution on  human  health,
agriculture,  vegetation, wildlife,  visibility,   and
climate.  Furthermore,  the   individual  charac-
teristics of the receptor must be taken into account.
  Air  quality  standards are  prescriptive. They
prescribe pollutant exposures or levels of effect
that scientific judgment determines should not be
exceeded in a specified geographic area, and they
are used as one of several factors in designing
legally enforceable pollutant emission standards.
  This  document  focuses   on   photochemical
oxidants and their precursors, HC and NO,, asthey
are found in the ambient air. The nitrogen oxides
are examined, but only for their oxidant-precursor
role; a detailed examination of  NO, as pollutants in
their own right ca n be found in Air Quality Criteria
for Nitrogen  Oxides (U.S.  Environmental
Protection   Agency,  Publication  No.  AP-84,
January 1971) and Air Quality Criteria for Nitrogen
Oxides (Luxembourg: NATO/CCMS, Document N.
15, June  1973). The discussion of HC  in  this
document rests  almost  entirely on their  role as
precursors  of other  compounds formed  in  the
atmospheric reaction system and not on the direct
effects of  HC themselves.  Gas-phase HC  and
certain  of   their oxidation products that  are
associated  with  thi manifestations  of  photo-
chemical  air pollution  are  discussed.  Toxic
organics will not be discussed  here. Establishing
health data for toxic organics isextremelycomplex
and will be  handled at a later date as more studies
are completed.
  This publication reviews the sources of oxidants
and  of  oxidant  precursors and  discusses  the
chemistry of the atmospheric  oxidant formation
process and the relationships  between oxidants
and  their precursors. Air  quality  data,  control
techniques, and  source and emission inventories
are discussed briefly for orientation purposes only,
inasmuch as these topics are discussed in detail in
other documents. Measurement techniques  and
fate  and transport of pollutants are discussed to
the extent necessary to give a clear understanding
of the health and welfare effects and of the bases
for sound oxidant control strategies.
  The status of control technology for oxidants and
their  precursors has   not  been  treated.  For
information on this subject, the reader is referred
to Control Techniques for /VO*  and Hydrocarbons
from Mobile Sources (U.S. Department of Health,
Education,  and Welfare,  Publication No.  AP-66,
March  1977),  and  Control  Techniques  for
Hydrocarbons  and Organic  Solvents  (U.S.
Department of Health, Education,  and Welfare,
Publication  No. AP-68, March 1977). The subject
of adequate margin of safety stipulated in Section
108 of the Clean  Air Act also has not been treated.
Again, the  reader is referred  to documentation
prepared by OAQPS.
  Methods  and  techniques for  controlling  the
sources of photochemical oxidants as well as the
                                             17

-------
costs of applying these techniques are discussed in
other documents.

  The  scientific literature  has been reviewed
through 1976-77. This document is not intended
as a complete, detailed  literature  review, and it
does not cite  every published  article relating to
oxidants in the environment and their effects. The
literature  has  been  reviewed thoroughly  for
information  relative  to  criteria.  Chapters  1 1
(Effects of Photochemical Oxidants on Vegetation
and Certain Microorganisms) and 1 2 (Ecosystems)
in this document are based on the chapters by the
same numbers and titles in the National Academy
of   Sciences  publication,  Ozone  and  Other
Photochemical  Oxidants  (Washington,  D.C.,
1977). An attempt has been made to identify the
major deficiencies  in  our  current  knowledge
relative to criteria. Between the issuance of the
preprint volumes and the final publication of this
document,  a few minor changes were made  to
correct inadvertent omissions.
   18

-------
         3.  NATURE AND ATMOSPHERIC  CONCENTRATIONS
                       OF  PHOTOCHEMICAL  OXIDANTS
NATURE OF OXIDANT
  Photochemical oxidants are chemical entities of
concern because of their detrimental effects on
biological systems and on certain materials. They
are  products  formed  in  the atmosphere  by
sunlight-driven,  chemical reactions that  involve
HC* and NO*.
  The term "photochemical oxidants" is used here
to define those  atmospheric pollutants that are
photochemical reaction products and are capable
of oxidizing neutral iodide ions. Extensive research
has unequivocally identified several components
of the  photochemical oxidants  mixture. Thus
oxidants  in ambient air  are known to  consist
mainly of ozone, PAN, and nitrogen dioxide (N02)
and are suspected to include also (but in smaller
amounts)  other  peroxyacylnitrates, hydrogen
peroxide, alkyl hydroperoxides, nitric and  nitrous
acids, peracids, and ozonides. Collectively, they are
measured by potassium iodide procedures and are
referred  to in  this  report as  "oxidant."  An
important distinction to be made here is that the
formation of N02 in ambient air clearly precedes
the formation  of the  other  oxidants. For  this
reason, the relative levels of nitrogen dioxide and
of the  aggregate of the  other  oxidants vary
considerably  during  the  day,  with N02 being
invariably the dominant oxidant earlier in the day.
Because of this difference in the variation  pattern
and the differences in effects between NO2 and the
aggregate of the other oxidants, the NO2 pollution
problem has  been treated  independently  of the
other photochemical oxidant problem. Further-
more,  because  of  its  predominance  among
oxidants other than NO2, ozone has  been  singled
out and treated as the sole representative of such
oxidants and has been given most of the research
attention.
 "The term ' hydrocarbon' is meant to include also all nonhydrocarbon organic
compounds capable of participating in the atmospheric oxidant formation process
Accordingly, the terms  hydrocarbons"  and  ' organics'  are often  used
interchangeably
  Photochemical   air  pollution   is  customarily
defined in terms of the concentrations of ozone and
N02 only. This should not be interpreted to suggest
that other photochemical pollutants are thought to
be of less concern. Rather, it reflects (1) the fact
that ozone and N02 pollution problems are more
easily quantified and hence are more amenable to
research than problems associated with the other
pollutants  (e.g.,  eye   irritation  and  visibility
reduction),  and  (2) the assumption that abatement
of the ozone and NOs pollution problems will, in all
probability,  alleviate the  other  photochemical
pollution problems.
  The ozone found in the lower levels of the earth's
atmosphere has been traced to both natural and
anthropogenic  sources. One natural source  of
tropospheric  ozone is the  ozone  abundantly
present  in the  stratosphere,  which  can  be
transported into the biosphere.  Ozone  can also
form naturally  from electrical discharges  in the
atmosphere and from atmospheric photochemical
reactions  involving naturally  emitted organic
vapors and NOX.  Obviously, the  levels  of ozone
resulting from  all  such uncontrollable sources
must be known if the anthropogenic sources and
the effects of  their control are to be assessed
reliably.
  Ozone and other oxidants from anthropogenic
sources  are  products  of  atmospheric photo-
chemical reactions involving primary organic and
inorganic pollutants and atmospheric  oxygen.
More specifically, the oxidant formation process is
initiated  by the photolysis of light-absorbing air
contaminants such as NO2, aldehydes, and ozone,
resulting in  the formation  of  highly  reactive
radicals  that react  subsequently with organic
pollutants.  The  net result of this photochemical
activity  is the accumulation of ozone, and  other
oxidants, to concentrations that depend on many
factors,  including local  meteorological conditions
(sunlight intensity,  air  stagnation, temperature,
etc.) as well as the concentrations, makeup, and
variation  patterns  of   the  primary  pollutants
                                              19

-------
present. Such dependence of oxidant formation on
multiple  factors  makes this  pollution  problem
immensely complex.

OXIDANT CONCENTRATIONS AND THEIR
PATTERNS

Introduction
  Although buildup of photochemical  oxidants
occurs in nearly every urban center in the United
States, no pollution episodes involving a sudden
and massive assault on human health have been
attributed  solely   to  photochemical  oxidants.
Accordingly, no case studies have been reported in
which  pollutants  and their  effects were  com-
prehensively and systematically examined during
an  oxidant   pollution  episode.  Photochemical
oxidants  are viewed as an air pollution  problem
mainly on the bases of (1) observations of  humans,
animals,  plants, and materials in areas known to
have  high  oxidant levels, and (2) results from
studies in which human subjects, animals, plants,
and   materials   were  exposed  to  smoggy
atmospheres  or to synthetic mixtures containing
oxidant or ozone under controlled  conditions. To
present convincingly the case that photochemical
oxidant air pollution is a  problem,  it would be
necessary and sufficient to present (1) data on the
concentration and frequency of  occurrence of
photochemical  oxidants, and  (2) evidence
regarding the adverse effects of oxidants. Data on
the occurrence of oxidants  are  given in this
chapter, but only in a  brief summary form, since
the occurrence of oxidant at problem levels is well
established.  Evidence   on   the  effects  of
oxidant/ozone is presented and discussed in detail
in subsequent  chapters.  For  a  more  detailed
discussion of ambient levels  (and variations in
those levels) of ozone and other oxidants and of
some nonoxidant photochemical pollutants,  the
reader is referred to the literature.26"27

Oxidant Concentrations in Urban Atmospheres

  Concentrations of oxidant in ambient air were
measured  in  1974  at some 340 monitoring
stations  operated  by  state  and local  control
agencies. Such data are gathered by EPA, stored in
EPA's National Air Data Bank, analyzed for trends,
and reported  in  summary  form  as  EPA
reports.26'29'30'31'32
  Table 3-1 shows the maximum hourly average
as well as the number of days when the maximum
hourly average oxidant concentration was equal to
or exceeded 294, 196, and 98/yg/m3 (0,1 5, 0.10,
and 0.05 ppm) for  1 2 monitoring  sites  in urban
areas. Higher readings of 0.52 ppm in Philadelphia
and 0.35 ppm in St.  Louis were judged tc be
spurious and were discarded. More recent data for
1 6 urban areas are given in Table 3-2.
  In  general,  locations within the Los Angeles
Basin urban area have the highest peak oxidant
concentrations (represented by the 99th percentile
values) as well as the most frequent violations of
the oxidant standard. High levels are also observed
in  the northeast corridor between Washington,
D.C., and Boston, Mass. In general, most cities for
which  extensive ozone data are available have
been  shown to  exceed the national air quality
standard for ozone of 160 /jg/m3 (0.08  ppm),
  A  number  of  these   urban areas  have
experienced very high levels of oxidant, exceeding
600 /jg/m3  (0.30 ppm). Los Angeles, for example,
recorded maximum 1 -hr values in excess of 1 200
/jg/m3 (0.60 ppm). Denver, Philadelphia, Houston,
and the area just east of New York - northeastern
       TABLE 3-1. SUMMARY OF MAXIMUM OXIDANT CONCENTRATIONS IN SELECTED URBAN AREAS


Urban area
Los Angeles, Calif
Pasadena, Calif
San Diego, Calif
Sacramento, Calif
Santa Barbara, Calif
San Francisco, Calif
Philadelphia, Pa
St Louis, Mo
Cincinnati, Ohio
Denver, Colo
Washington, D C
Chicago, III


Year
1964-1967
1964-1967
1964-1967
1964-1967
1964-1967
1964-1967
1964-1972
1964-1972
1964-1972
1964-1972
1964-1972
1964-1972
Total
valid
data
730
728
623
7T1
723
647
1783
2014
1668
1542
2157
2042
Number of days with at leasl 1 hourly
average equal lo or exceeding
98 j/g/m3
(005 ppm)
540
546
440
443
510
185
723
1042
749
934
1032
828
196 W/mJ
(0 10 ppm)
354
401
130
104
76
29
186
152
105
179
193
85
294 f/g/mj
(01 5 ppm)
220
299
35
16
11
6
39
27
14
35
25
14
Maximum
hourly
average.
ppm
058
046
038
0 26
025
022
033
0 22
026
036
0 25
020


fjg/mj
1 137
902
745
5TO
490
431
647
431
510
510
490
392
 20

-------
             TABLE 3-2. OXIDANT CONCENTRATIONS OBSERVED IN SELECTED URBAN AREAS
                                 OF THE UNITED STATES, 1974-75





Urban area
New York, N Y - Northeastern N. J
Los Angeles - Long Beach, Calif
Chicago, II! - Northwestern Ind
Philadelphia, Pa.
Detroit, Mich.
Boston, Mass
Washington, D C
Cleveland, Ohio
Minneapolis - St Paul, Minn
Houston - Galveston, Tex
Baltimore, Md
Dallas - Fort Worth, Tex
Milwaukee - Racine, Wis
Seattle - Tacoma, Wash
Cincinnati, Ohio - Northern Ky.
Denver, Colo


Total
no of
valid
sites8
8
3
6
10
2
7
8
5
2
4
2
2
7
4
6
6


No of Sites
exceeding
oxidant
standard0
8
33
6
10
2
7
8
5
1
4
2
2
7
3
6
6

Range of
2nd max
1-hr
values,
^g/m3 (ppm|
259-510(013-026)
255-784(013-040)
163-427(008-0.22)
216-625(0.1 1-032)
455-514(023-026)
186-376(0.09-019)
363-451 (0 18-023)
245-41 , (0 12-021)
141-206(007-0.10)
304-588(0 16-030)
314-372(0 16-0 19)
274-323 (0 14-0 16)
332-425(017-022)
1 18-235(006-0 12)
284-412 (0 14-021)
212-349(0.11-018)
% of days
OKtdant
standard
violated
at the
worst site
197
38.6
80
382
3 1
8 2
274
36
42
235
154
148
145
78
23 1
275
 "Only sites having a minimum of 4OOO observations were included »n this summary
 ''The ojcidant standard is a 1 -hr average of 160 y/g/m3, not to be exceeded more than once per year
New Jersey have experienced levels in excess of
600 Aig/m3 (0.30 ppm).  Hourly values exceeding
400/ug/m3 (0,20 ppm) haveoccurred in most of the
major urban areas.
  The data in Table 3-1 and some of those in Table
3-2  were  obtained by the  potassium  iodide
method,26 and therefore they include both ozone
and  nonozone  oxidant  species. Data  on  ozone
alone have been obtained only for the more recent
years,  after  the chemiluminescence  method for
measurement of ozone was introduced. National
summaries of such  data are not available  in a
reported   form,   but  detailed  tabulations  are
available.28'29'30'31'32
  Nonozone oxidants are certain to exist in urban
atmospheres, but in  concentrations considerably
lower than those of ozone.7'26'27 Measurements of
such oxidants in ambient air are scant and consist
mostly of PAN  data.21'22 Some recent data on
absolute  concentrations of PAN and on ozone-to-
PAN or oxidant-to-PAN  ratios both in urban and
rural  atmospheres are  shown  in  Tables  3-3
through  3-5.21  The  data  currently  available
indicate,  in general, that in urban atmospheres,
PAN concentrations are considerably lower than
those of oxidant or ozone, but nevertheless, they
are not negligible.  Hydrogen peroxide (H2O2) also
has been reported  to occur in smog chamber and
ambient atmospheres at concentrations as high as
0.18 ppm.2 Peroxybenzoyinitrate (CeHsCOOzNOz)
has been reported to occur in urban ambient air in
the Netherlands,23 but no evidence exists for its
occurrence  in  U.S.  urban  ambient  air.  The
significance  of  such  PAN and  other  oxidant
concentrations can only be assessed on the basis
of their adverse effects on humans, vegetation,
and materials.

Oxidant Concentrations in Rural Atmospheres
  Ozone  concentrations  in  rural atmospheres
have been of interest because they were thought to
be a measure of the  composite contributions of
TABLE  3-3. PAN AND OXIDANT (O») MEASUREMENTS (10 a.m. to  4  p.m.), LOS  ANGELES, CALIF., 1968"

No of
samples
19
59
27
6
5
2
PAH
cone ,
ppb
0-10
10-20
20-30
30-40
40-50
> 50
Avg
PAN cone ,
ppb
8.65
130
24.0
32.6
47 1
65.5
Range at
oxidant cone .
ppb
30-190
44-220
70-290
1 00-400
1 60-285
243-410
Avg
oxidant cone ,
ppb
79.0
97.0
1440
1680
209.0
3270

Ratio
avg O,/avg PAN
9.13
758
604
523
4.49
4,94
Range of
observed
O,/PAN
5 1-244
3 3-18 3
28-130
30-97
3.2-6.1
3 9-6.0
 a 1 96 j/g/m3 - 1 ppb as ozone, 5 pg/m3 ~ 1 ppb as PAN
                                                                                            21

-------
TABLE 3-4.  PAN  AND OZONE MEASUREMENTS  (10 a.m.  to  4  p.m.),  HOBOKEN,  N.J.,  1970'

No of
samples
14
15
4
8
2
PAN
cone ,
ppb
0-2
2-4
4-6
6-8
8-10
Avg
PAN cone ,
ppt>
1 5
2 8
47
7 1
99
Range of
ozone cone .
ppb
15,0-1120
22.0-1500
100.0-135.0
97 0-2190
2140-2780
Avg
ozone cone .
ppb
420
71 0
1130
1620
245 0

Raiio
avg Oa/avg PAN
280
253
240
228
25 1
Range of
observed
Oy PAN
56-775
10 1-62 5
22 9-25 0
12 6-34 1
20 7-29 6
 s 1 96 t/g/mj = 1 ppb as ozone, 5 pg/m3 ~ 1 ppb as PAN


TABLE 3-5.  PAN, OXJDANT. AND OZONE MEASUREMENTS (10  a.m. to 4  p.m.), ST. LOUIS, MO., 19732


No of
samples
3
31
60
31
14
5
10

PAN
cone .
ppb
0-2
2-4
4-6
6-8
8-10
10-12
> 12

Avg PAN
cane ,
ppb
1.7
3.0
5 0
68
93
108
186
Range of
oxidant
cone .
ppb
46-100
46-216
30-250
39-224
40-180
30-90
52-100
Avg
oxidant
cone ,
ppb
80.6
84 6
673
68 7
846
734
856
Ratio
avg
O,/avg
PAN
475
283
134
10 1
81
68
46
Range of
ozone
cone ,
ppb
10-40
24-116
10-140
20-120
10-96
30-85
32-80
Avg
ozone
cane ,
ppb
286
48 1
502
50.3
502
540
59.5
Ratio
avg
Qa/avg
PAN
163
16.2
100
74
54
50
32
 a 1 96 ^g/m3 - 1 ppb as ozone, 5 /jg/m3 - 1 ppb as PAN

natural  sources to  the  ambient ozone  problem.
Recent  studies, however, have established  the
occurrence of pollutant transport, thus refuting
this early notion; and it is now widely accepted that
rural  areas are  not  necessarily  unaffected  by
anthropogenic pollution. This conclusion has led
air pollution specialists to use the term "rural" to
signify areas or atmospheres that are nonurban
but are  occasionally susceptible to anthropogenic
pollutants, and the term "remote" to signify areas
or  atmospheres  so far  removed from  anthro-
pogenic  pollutant  sources  that  their  contam-
ination  by such pollutants is highly unlikely.
  Before 1962, ozone concentrations both in rural
and in remote areas appeared to be considerably
below the NAAQS level of 160/jg/m3 or 0.08 ppm
                                                 (Table 3-6).17  Comparable levels were observed
                                                 also in rural  North Carolina  during 1 964-67.34
                                                 Measurements in the last 5 to 6 years, however,
                                                 showed that the ozone concentrations m several
                                                 rural areas exceeded 160 /ug/m3 (0.08 ppm) with a
                                                 frequency comparable  to or even surpassing that
                                                 observed in many urban areas.
                                                 This fact is illustrated by the data in Table 3-7,
                                                 which were obtained  by recent multiyear  field
                                                 studies sponsored by EPA.35 Similar data were also
                                                 reported by New York  State investigators.6 Such
                                                 increased ozone levels  (i.e.,>160/L/g/m3, or>0.08
                                                 ppm) in rural areas are  almost certain to be caused
                                                 largely by anthropogenic ozone and/or precursors
                                                 generated  locally or transported into such  areas
                                                 from urban centers.20'33'35'36'40'43 There is evidence
                TABLE 3-6. CONCENTRATIONS OF TROPOSPHERIC OZONE BEFORE 19621

Observer
Cotz and Volz (1951)

Regener (1957)

Regener (1957)
Ehmerl(1952)

Teichert (1955)

Kay (1953)
Brewer (1955)
Rice and Pales (1959)
Wexler el al. (1960)

Location, time, and remarks
Arosa, Switzerland, 1950-51, high
valley, daiiy maximum values
Mt Capillo and Albuquerque,
New Mexico, 1951-52
O'IMeil, Nebraska, 1953
Weissenau, Bodensee, Germany, 1952

Lindenberg Obs., Germany, 1953-54

Farnborough, England, 1952-53
Tromsfl, Norway, 1954
Mauna Loa Observatory, Hawaii
Little America Station, Antarctica

Altitude

1860 m
3100 m
1600 m
125m above ground
20m
above ground
80 m
above ground
0-12,000 m
0-10,000 m
3000 m
100 m
Ql.lU
Range"

19-90
18-85
3-120
30-100
0-90
0-70
0-50
0-50
26-50
60-70
30-62
20-60
g/m3
Average3

50
45
36
60
35
30
30
27
38
65
45
45
 a As interpreted from the published data The values sometimes represent absolute maxima, sometimes mean maxima

-------
                    TABLE 3-7. SUMMARY OF OZONE DATA FROM 1973-75 OXIDANT
                            STUDIES, RESEARCH TRIANGLE INSTITUTE36
Station
McHenry, Md.
Kane, Pa.
Coshocton, Ohio
Lewisburg, W Va
Wilmington, Ohio
McConnelsville, Ohio
Wooster, Ohio
McHenry, Md.
DuBois, Pa.
Canton, Ohio
Cincinnati, Ohio
Cleveland, Ohio
Columbus, Ohio
Dayton, Ohio
Pittsburgh, Pa
Bradford, Pa
Lewisburg, W Va.
Creston, Iowa
Wolf Point, Mont
DeRidder, La
Pittsburgh, Pa
Columbus, Ohio
Poynette, Wis
Cedar Rapids, Iowa
Des Moines, Iowa
Omaha, Neb.
Nederland, Tex.
Port O'Conner, Tex.
Austin, Tex.
Houston, Tex
Year
1973
1973
1973
1973
1974
1974
1974
1974
1974
1974
1974
1974
1974
1974
1974
1975
1975
1975
1975
1975
1975
1975
1975
1975
1975
1975
1975
1975
1975
1975
Type of
station
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Urban
Urban
Urban
Urban
Urban
Urban
Rural
Rural
Rural
Rural
Rural
Urban
Urban
Rural
Urban
Urban
Urban
Urban
Rural
Urban
Urban
Average O3
ppm
0.074
0.065
0056
0.0952
0.052
0.057
0.047
0.057
0056
0035
0.025
0.031
0033
0035
0.028
0040
0.038
0035
0.028
0.030
0030
0.022
0.038
0025
0036
0035
0027
0027
0025
0.026

l/g/m3
145
127
110
187
102
112
92
112
110
69
49
61
65
69
55
78
74
69
55
59
59
43
74
49
70
69
53
53
49
51
No hours
20 08 ppm
600
639
357
249
259
262
262
262
341
148
54
51
113
114
106
100
59
17
0
38
227
43
126
6
124
64
138
99
19
141
No
hours
1662
2131
1785
1663
1751
2011
1878
2011
1667
1829
1548
1652
1935
1576
1622
2332
2386
2117
2160
2994
2841
2885
2663
2781
2528
1787
2714
2912
2504
2104
% Hours
2008
ppm
37.0
300
200
15.0
149
130
140
130
205
8.0
3.5
30
58
7.2
65
43
25
08
0.0
1 3
80
1.5
4.7
0.2
49
36
5.1
3.4
0.8
6.7
that some rural oxidant is anthropogenic in nature.
This evidence is based on measurement of Freons
or  of acetylene,  or  on results  of  wind back-
trajectory analysis, both of  which are indications
of the passage of the  rural air over urban sources.
20,33,36,40,43 Additional   data   on  ozone con-
centrations  in remote  areas  are presented  in
Chapter 4.
  Data on nonozone  oxidants in  rural or remote
atmospheres are scarce. Recent data on PAN in the
rural atmosphere at  Wilmington, Ohio, showed
levels considerably lower  than those found in
urban atmospheres, both absolutely and relative to
ozone concentrations. Earlier as  well as recent2
measurements of PAN and oxidant/ozone showed
that (1)  the PAN concentrations are  much lower
than those of oxidant/ozone and (2)  the PAN-to-
oxidant/ozone ratio varies with location, the rural
areas showing  lower  values  than the  urban
centers.21 The maximum PAN concentration from
1 500 samples taken in August 1974 was 4.1 ppb,
with the daily  maximum  rarely exceeding 3.0
ppb,21  even  though  the  ozone  concentration
frequently exceeded 1 60 /vg/m3 (0.08 ppm). Such
low  rural  PAN  concentrations,  expressed  in
absolute terms as well as in terms relative to
ozone,  can  be explained  in  different  ways.
Accordingly, one explanation is that the chemical
mechanism of oxidant formation is such that the
lower reactant concentrations (of HC and NO*) and
the  prolonged  irradiation  conditions  that
characterize  rural  pollutant mixtures result  in
lower PAN yields and lower PAN-to-ozone product
ratios.21 Another explanation lies in the reversible
decomposition of PAN  into NC>2  and peroxyacyl
radical; in rural areas where NO? is nearly absent,
PAN  decomposition  is  enhanced.13  A  third
conceivable explanation is that the lower PAN-to-
ozone  ratios in rural atmospheres may reflect a
greater proportion of stratospheric ozone relative
to the  photochemically  produced  ozone.  The
explanations  based on the chemical mechanism
are believed to be the more plausible ones, at least
for the Wilmington case.21 The stratospheric ozone
explanation may be a valid one in the cases in
which  PAN is undetectable.
Patterns of Variation in Oxidant Concentrations

  Patterns  of  variation  in  oxidant  or  ozone
concentrations are of interest for  at  least  two
                                                                                          23

-------
reasons. They provide a more detailed description
of the oxidant problem, which in turn serves to
guide the control effort better. Such patterns also
help to  explain  the  chemical  and  physical
mechanisms by which  emissions of precursors (or
of ozone) disperse, react, and ultimately cause the
observed oxidant problems.
  The most conspicuous variation patterns are the
seasonal and diurnal ones.  Such patterns result
largely from (1) variations in  emissions of oxidant-
forming  pollutants, (2) variations in atmospheric
transport  and the dilution  processes, and  (3)
variations in other atmospheric variables involved
in  the  photochemical formation  of  oxidants.
Detailed descriptions of the  seasonal and diurnal
patterns for oxidant/ozone  concentrations and
some explanatory discussions can be found in the
literature,7'9'26
  Briefly, the striking  characteristics of the sea-
sonal anddiurnalpatternsare(1 )the occurrence of
higher  oxidant concentrations  in  the  summer
months, and (2) the occurrence of a daily oxidant
peak in  the early afternoon  hours. Of these, the
seasonal pattern is consistent with and, in fact,
supports the theory of photochemical formation of
atmospheric  oxidant  in that the  higher  tem-
peratures and  sunlight intensities in the summer
enhance the photochemical  process.19
  Unlike the seasonal patterns, the diurnal pattern
can be explained in  more  than one way.  Until
recently, the only explanation accepted was based
on the assumption that the daily oxidant peak at a
particular  location  resulted  from local emission
sources or from those that had been transported to
the site. The reactions normally occurred several
hours before the oxidant peak.19
  A more recent theory explains the oxidant peak
as the net result of three simultaneously occurring
processes:
          .6,25
(1)  downward  transport  of
oxidant/ozone from layers aloft; (2) destruction of
oxtdant/ozone on surfaces and in reaction with NO
at ground level; and  (3) photochemical in situ
production of oxidant/ozone. Also, interpretation
of reported  evidence  suggests that  downward
transport and destruction  by reaction with NO,
rather  than photochemical formation,6'25
determine the  concentration of oxidant/ozone
observed at  ground level for the cores of  some
cities as well  as  remote  areas. Whether the
oxidant/ozone in the layers aloft originates from
natural or anthropogenic sources is a question that
some investigators  consider unresolved;5'6 the
bulk of the recent evidence on long-range pollutant
transport,  however, points to the anthropogenic
origin.25'31

ASSOCIATION  OF  OZONE  WITH  OTHER
CONSTITUENTS  OR  MANIFESTATIONS OF
SMOG
  The association of ambient levels of ozone and
N02 does not show well-defined patterns  that
would suggest a certain impact of ozone-related
control on N02. Trend analysis of Los Angelesdata
shows the ambient N02 concentration to increase
and  ozone  to  decrease during  1965-74  (see
Figures  3-1  and 3-2).42 These trends have been
explained  as reflecting the effects of the HC and
NO, emission rate changes during that period. The
effect of the ozone-related HC control alone on N02
cannot  be  delineated  from the apparently
overwhelming effect of the NO, emission change.
A recent study by  Trijonis  of the ambient N02-
precursor relationships provided evidence that the
hydrocarbon emissions factor had  only a small
codirectional effect on ambient N02, whereas the
NOX factor had a  much stronger effect, equivalent
to a 1:1 correspondence between NOX  emission
change and ambient NOa concentration change.41
Laboratory (smog chamber) data also show  (1) a
nearly  1:1 correspondence between  fractional
changes in the N0« precursor factor and fractional
changes in "daily" average or maximum 1 -hr  N02,
and (2)  a  small effect upon N02  from  change in
HC.8'16
  The association  of ozone and  PAN  has been
explored using the limited amount of ambient data
available,  and  some   explanations  and
interpretations have  been  given  of the results.
Earlier and  recent measurements  of  PAN  and
oxidant/ozone showed (1) the PAN concentrations
to be much smaller than those of oxidant/ozone,
and (2)  the  PAN-to-oxidant/ozone  ratio to  vary
with location, the rural areas showing lower
valiTes than the urban centers.21 The lower PAN-
to-oxidant/ozone ratios m rural areas can be
explained  mechanistically either  as a  result of
enhanced  PAN  decomposition   in  NOg-free
atmospheres or as  a result of the  relatively lower
concentrations  and  prolonged  irradiation
conditions, as a result of pollutant transport, that
characterize the  reaction systems   in  rural
atmospheres.   Overall,   the  ambient   data
associations do not provide conclusive evidence on
the  impact of ozone-related control on PAN. The
evidence,  however,  from  laboratory  and
theoretical studies  is definitive and shows that for
urban atmospheres,  changes in HC  and  NO,
24

-------
                                                               +36%*-;
  BASIIMWIDE HC EMISSION CHANGE:
  BASiNWIDE NOX EMISSION CHANGE:   +36%


  AVERAGE N02 CONCENTRATION CHANGE (11 STATIONS): (+20%
          Figure 3-1. Trends in NOz air quality and in HC and NO, emission, Los Angeles Basin, 1965-74,
        18

  »=,_•   IB

  ?1|  "

  jjo   12

        10
<» » U E
< ui0l
0> Q Q? D.
 ' Xt °"
 O Z
 •is
 ! I CCl

  S o a
  Oo
  z >
   i
        30

        25 -
       2"-
       20 |-
       1 5
            1963 84  6S  66  67  68  6»  70  71
                          YEAR
                   ANNUAL AVERAGE
            	O THREE-YEAR MOVING AVERAGE
Figure 3-2. Pollutant trends in Los Angeles, annual and 3-
year moving average***2
emissions should have greater impacts on ambient
PAN than on ambient ozone,8
  Ambient aldehydes, although nonoxidants, are
also of interest here because they are constituents
of the photochemical pollution complex and are
known  to  have  some  physiological  effects
(Chapters 8 and 9), However, aldehydes are also
directly   discharged  into  the   atmosphere  as
automotive emissions,10'39 Concentrations  in
urban atmospheres seem to  rise quickly in the
early morning hours, reach and maintain a broad
plateau  through  most  of the  day,  and  then
decrease in the afternoon.7 This pattern probably
reflects the dual origin (primary and secondary) of
the ambient aldehydes as well as the fact that rapid
photochemical  formation of  aldehydes, unlike
oxidant/ozone,  begins  with  the  onset  of
irradiation.19 Ambient data associations do not
provide evidence on  the  impact of ozone-related
control on aldehydes; laboratory data, however, do
suggest a  nearly  1:1 correspondence between
fractional changes in HC and fractional changes in
ambient (photochemical) aldehydes.8
  The association of ambient oxidant/ozone with
eye irritation has been explored extensively, but
only in one location, the Los Angeles Basin, This
association has been described  in detail in the
predecessor of this criteria document and need not
                                                                                           25

-------
be  elaborated  here   Briefly,  ambient
oxidant/ozone appears to correlate well with eye
irritation, a relationship which, however, is clearly
not of  cause-effect  nature. Lack  of  causative
relationship means that control of  ozone is not
expected to have a direct effect on eye irritation.
However, the high degree of correlation between
the ambient data, the laboratory evidence of good
correlation between  the photochemical  process
and eye irritation,45 and, finally, the fact that some
eye irritants, (namely formaldehyde, acrolein, and
PAN)  are  known to be  products of atmospheric
photo-oxidation of HC, when considered together
all suggest that  oxidant-related emission control
will  probably  have  a reducing  effect  on  eye
irritation for Los Angeles. Eye irritation has been
observed also in other areas. A pollution episode,
for example, in New York City during November 23-
25,  1966,  caused   numerous  eye  irritation
complaints  from   residents,1   Oxidant/ozone
measurements were not made during the episode,
and  therefore  an  association  could  not  be
established. The problem could have been caused
either by primary pollutants (aldehydes) mainly, or
by  both primary and  secondary  pollutants. In
conclusion,  based on present-day knowledge and
understanding,  the  evidence  and conclusions
obtained for the Los  Angeles area  cannot be
assumed  to have  universal validity.  For such
assumption to be supported (or refuted), additional
evidence must be obtained.
  The association of ambient ozone with visibility
or visibility-reducing aerosol  is perhaps the most
interesting  of the associations discussed here.
Public reaction to the visibility problem has been
stronger than to the ozone problem, and there is  a
feeling  among local air pollution control  officials
that the often upsetting ozone  control programs
will be more acceptable if they are shown to have a
beneficial impact on the haze problem also. In spite
of  this strong  interest   in  the ozone-haze
association, relatively little evidence is currently
available on this subject. The evidence available is
mostly  on  the  Los  Angeles  atmosphere.  A
considerable part was obtained in the course of
conducting  the  California  Air Characterization
Study   (ACHEX)14  and  has been summarized
informatively  in a recent  National Academy of
Sciences report.42 The part of the evidence of most
interest here is  shown in Figure 3-3  in terms of
diagrams indicating a good correlation for the Los
Angeles Basin between visibility  reduction and
ambient ozone concentration. Somewhat similar,
but weaker, evidence was reported by Husar et
   14
   12
   10
 O
 cj
 cj
 Z
      Bsca, AT PEAK OZONE (BASED ON TWO HOUR
      AVERAGED DATA FROM LOS ANGELES AREA)
             (SUMMER 1973)

        O ROUBIDOUX (RIVERSIDE)
        a WESTCOVINA
        D POMONA
        • DOMINGUEZ HILLS
           (TORRANCEi
  (SUMMER 1972)

* PASADENA
• RIVERSIDE
• POMONA
• HARBOR FWY
  (DOWNTOWN LA)
O BLIMP FLIGHT -
  (9/6/73)
  1315-1500 PST
                                   BLIMP
         NO AEROSOL IN THE ATMOSPHERE

            1      I       I      I
           01     0.2     0.3     0.4

                 03, ppm (MAXIMUM)
                                     0.5
                                            06
         Figure 3-3. Correlation between BMa,
         and maximal ozone concentration.15

al.15 and by Wolff  et  al.46  who showed  that
synoptic-scale  areas  with  elevated  ozone
concentrations roughly corresponded  in location
and  magnitude with  areas  of  reduced visibility
("haze blobs").  Such  good correlations between
ozone and haze are self-explanatory in the cases in
which the haze-causing aerosols consist largely of
photochemical reaction  products. This clearly is
the case with Los Angeles, as deduced from the
findings that the optically active aerosol in the Los
Angeles  air consists primarily of photochemically
generated sulfate with lesser amounts of nitrates
and organics.3'11'12'44
  The good ozone-haze correlation observed in Los
Angeles, the  aerosol composition data, the good
correlation between haze and ambient sulfates,24
and, finally,  the experimental  and  theoretical
evidence  on the  atmospheric photo-oxidation
process18'3 all point to the conclusion that oxidant-
related control should also cause some reduction
in haze in Los Angeles. This conclusion, with two
 26

-------
important qualifications,  is probably correct. The
qualifications are that (1) ozone-related control,
although beneficial, is not necessarily  the most
effective means for controlling haze,  and (2) the
Los  Angeles evidence on  the impact of ozone-
related control  on  haze does not have  universal
validity.  The second qualification is supported by
the case of Denver, for example, where the haze
problem  appears  to   be  caused  by  primary
aerosols.38'4 Such  aerosols  cannot  possibly be
impacted by the  ozone-related control except,  of
course,  when  HC  emission control  measures
happen  to reduce vehicular aerosol  emissions
(e.g.,  measures   reducing   automobile  traffic).
Houston  also   has   been   reported  to  show
interpollutant relationships  different  from those
occurring in Los Angeles.4
SUMMARY
  Photochemical  oxidants   are  products  of
atmospheric  reactions   involving  organic
pollutants, NOx, oxygen, and sunlight. They consist
mostly of ozone, NC>2,  and  PAN, with smaller
amounts of other  peroxyacylnitrates  and other
peroxy-compounds, and  are formed  along with
other photochemical products such as aldehydes,
gaseous and paniculate nitrates, and sulfates.
They originate  mainly from volatile organic and
NOx emissions  associated with human activities.
Photochemical  oxidant  formation  is  a complex
function of emissions and meteorological patterns.
  Peak  concentrations of oxidant expressed as
ozone are generally  higher  within  urban  and
suburban areas, reaching levels in excess of 590
/Kj/m3  (0.3  ppm).  In  rural  areas,  peak con-
centrations are lower but often exceed the  1-hr
National Ambient  Air  Quality Standard  of  160
/yg/m3  (0.08  ppm).  Dosages  or  average con-
centrations, however,  are comparable to or even
higher than  those  in  urban  areas.  Because  of
pollutant transport, the oxidant problem is viewed
as having regional dimensions in the hundreds-of-
miles  range.


REFERENCES FOR CHAPTER 3
  1   Becker, W. H., F J. Schilling, and M. P Verna. The effect
     on health of the  1966 eastern  seaboard air pollution
     episode Arch Environ Health 76.414-419, 1968
  2   Bufalmi,  J  J , B W Gay, Jr.,  and K  L Brubaker.
     Hydrogen peroxide  formation  from   formaldehyde
     photooxidation  and its presence in urban atmosphere
     Environ. Sci  Technol  6816-821,1972
  3   Cass, G R The Relationship Between Sulfate Air Quality
     and  Visibility at Los Angeles EQL Memorandum No 18,
    California Institute of Technology, Pasadena, CA , August
    1976
 4.  City of Houston. Petition for Review and Revision of
    Ambient  Air  Quality  Standard  for Photochemical
    Oxidants and Requirements for Control City of Houston
    Health Department, Houston, Tex., July 11, 1977
 5  Coffey, P E.andW N Stasiuk Evidence of atmospheric
    transport  of  ozone  into urban areas.  Environ.  Sci
    Technol 359-62, 1975
 6.  Coffey, P., W Stasiuk, and V Mohnen. Ozone in urban
    and rural areas of New York State. In  International
    Conference on Photochemical Oxidant Pollution and Its
    Control Proceedings Volume I. B Dimitnades, ed EPA-
    600/3-77-001a, U.S. Environmental Protection Agency,
    Research Triangle Park, N.C., January 1977 pp 89-96
 7  Committee on Challenges of Modern Society Air Quality
    Criteria  for  Photochemical Oxidants  and Related
    Hydrocarbons  NATO/CCMS Publication No  29, North
    Atlantic  Treaty Organization, Brussels,  Belgium,
    February 1974
 8.  Dimitnades,  B  On the  function of hydrocarbon  and
    nitrogen oxides in photochemical smog formation. Report
    of Investigation Rl 7433, U S Department of the Interior.
    Bureau of Mines, Washington, D.C , September 1970
 9  Dimitnades,  B Photochemical Oxidants in the Ambient
    Air  of the  United States  EPA-600/3-76-017,  U.S
    Environmental Protection Agency,  Research Triangle
    Park, N C , February 1976
10.  Dimitnades, B , andT C Wesson Reactivities of exhaust
    aldehydes J. Air Pollut. Control Assoc. 22 33-88, 1972.
11  Friedlander,   S  K  Chemical  element  balances  and
    identification of air  pollution  sources   Environ   Sci.
    Technol 7235-240, 1973.
12.  Grosjean, D,  and  S.  K  Friedlander  Gas-particle
    distribution factors for organic and other pollutants in the
    Los Angeles atmosphere J Air Pollut Control Assoc
    251038-1044, 1975
13  Hendry, D G , and R  A. Kenley Generation  of peroxy
    radicals from peroxynitrates(RO2NC>2) Decomposition of
    peroxyacyl nitrates  J Am  Chem  Soc 33.3198-3199,
    1977
14  Hidy, G M ,  B Appel, R  J  Charlson, W E Clark, S K
    Friedlander,  R  Giauque, S. Heisler, P K Mueller, R
    Ragami, L W Richards, T B Smith, A. Waggoner, J J
    Wesolowski,  K  T  Whitby,  and  W   White
    Characterization of Aerosols in California (ACHEX) Final
    Report Volume I Summary  California Air Resources
    Board Contract No. 358, Rockwell International Science
    Center, Thousand Oaks, Calif , April 1975
15  Husar, R B,D E Patterson, C C Paley.andN  V Gillani.
    Ozone in hazy air masses  In International Conference on
    Photochemical  Oxidant Pollution and  Its Control
    Proceedings  Volume I B Dimitnades, ed EPA-600/3-
    77-001 a,  US  Environmental Protection  Agency,
    Research Triangle  Park,  N C , January 1977. pp 275-
    282
16.  Jeffries, H,  D  Fox, and R  Kamens  Outdoor Smog
    Chamber Studies  Effect of Hydrocarbon  Reduction in
    Nitrogen  Dioxide   EPA-650/3-75-01 1 ,  US
    Environmental Protection Agency,  Research Triangle
    Park, N C , June 1975
17.  Junge, C  E  Atmospheric Chemistry and  Radioactivity
    Academic Press, Inc , New York, 1963
18  Kocmond, W   C,  and  J  Y  Yang  Sulfur dioxide
    photooxidation rates and aerosol formation mechanisms
                                                                                                    27

-------
     A  smog  chamber  study   EPA/600-3-76-090,  U S.
     Environmental  Protection  Agency, Research  Triangle
     Park, N C , August 1976
19   Leighton, P A Photochemistry of Air Pollution, Academic
     Press, Inc , New York, 1961

20   Lonneman.W A Ozone and hydrocarbon measurements
     m  recent oxidant transport  studies  In  International
     Conference on Photochemical Oxidant Pollution and Its
     Control Proceedings  Volume I  B Dimnnades, ed  EPA-
     600/3-77-001a, U S Environmental Protection Agency,
     Research  Triangle Park, IMC, January 1977, pp  211-
     223

21   Lonneman, W A , J  J Bufaiim, and R L Seila  PAN and
     oxidant measurement in ambient atmosphere  Environ
     Set  Techno!  /0347-380,  1976
22   Mayrsohn, H ,  and C  Brooks  The analysis of PAN by
     electron capture  chromatography  Presented  at the
     Western Regional Meeting, American  Chemical Society,
     November 18, 1965
23   Mei]er, G  M ,  and  H  Nieboer  Determination  of
     peroxybenzoyl nitrate (PBzN) in ambient air In Ozon und
     Begleitsubstanzen im photo-chemischen Smog VDI Ber
     (270)55-56,  1977
24   Miller, D F , and D W Joseph Smog-chamberstudieson
     photochemical  aerosol  precursor  relationships   EPA-
     600/ 3-76-080, U S  Environmental Protection Agency,
     Research Triangle Park, N  C , July 1976
25   Mohnen, V , and E Reiter  Internationa! Conference on
     Oxidants, 1976—Analysis of Evidence and  Viewpoints
     Part III  The Issue of Stratospheric Ozone Intrusion EPA-
     600/3-77 115,  US  Environmental Protection Agency,
     Research Triangle Park, IM  C , December 1977

26   National Air Pollution Control Administration. Air Quality
     Criteria for Photochemical  Oxidants. NAPCA Publication
     No  AP-63, U S Department of Health, Education, and
     Welfare, Public Health Service, Washington, D, C , March
     1970
27   National  Research  Council.  Ozone  and  Other
     Photochemical Oxidants National Academy of Sciences,
     Washington, D C , 1977
28   Office of Air Quality and Planning Standards  Monitoring
     and Air Quality Trends Report, 1972 EPA-450/1-73-
     004, U S Environmental  Protection Agency, Research
     Triangle Park, N C , December 1973
29   Office of Air Quality and Planning Standards  Monitoring
     and Air Quality Trends Report, 1973 EPA-450/1-74-
     007, U S Environmental  Protection Agency, Research
     Triangle Park, N C , October 1 974
30   Office of Air Quality and Planning Standards  Monitoring
     and Air Quality Trends Report, 1974 EPA-450/1-76-
     001, US Environmental  Protection Agency, Research
     Triangle Park, N C , February 1976
31   Office of AirQualityand Planning Standards National Air
     Quality and Emission Trends Report, 1 975 EPA-450/1 -
     76-002,  U S  Environmental  Protection  Agency,
     Research Triangle Park, N C , November 1976
32   Office  of  Air  Quality  Planning  and Standards  The
     National  Air   Monitoring   Program  Air  Quality and
     Emissions Trends Annual Report Volumes I  and II EPA-
     450/1 -73-O01a and b, U S  Environmental Protection
     Agency, Research Triangle Park, N C , August  1973
33   Rasmussen,  R  A Surface ozone observations in rural
     and remote areas J Occup Med  75346-350, 1976
34   Ripperton, L  A, and  J J. B  Worth  Chemical  and
     Environmental Factors Affecting Ozone Concentration in
     the  Lower  Atmosphere Environmental  Science  and
     Engineering  Publication No 234,  University  of North
     Carolina,  Chapel Hill, N C,, 1969.
35   Ripperton, L  A.J J  B. Worth, F  M. Vukovich, and C. E
     Decker Research Triangle Institute studies of high ozone
     concentrations in nonurban  areas.  In  International
     Conference on Photochemical Oxidant Pollution and Its
     Control Proceedings  Volume I  B Dimitnades, ed  EPA-
     600/3-77-001 a, U S. Environmental Protection Agency,
     Research Triangle Park,,NC, January  1977.  pp  413-
     424
36   Robinson, E  , and R  A Rasmussen  Identification of
     natural  and  anthropogenic  rural ozone for control
     purposes  In  Speciality  Conference  on  Ozone/
     Oxidants—Interactions with the Total  Environment Air
     Pollution  Control Association, Pittsburgh, Pa., March
     1976  pp  3-13
37   Russell,  P A (ed )  Denver Air Pollution Study—1973
     Proceedings of a Symposium Volume  I. EPA-600/9-76-
     007a, U S. Environmental Protection Agency, Research
     Triangle  Park, N C., June 1976
38   Russell,  P A  (ed ) Denver  Air Pollution Study - 1973
     Proceedings of a Symposium Volume II EPA-600/9-77-
     001,  US  Environmental Protection Agency, Research
     Triangle  Park, N C , February 1977
39   Seizinger, D  E, and B  Dimitnades  Oxygenates in
     exhaust  from simple  hydrocarbon  fuels  J  Air Pollut
     Control Assoc 2247-51, 1972
40   Spicer, C W,J  L  Gemma, D. W Joseph, P S Sticksel
     and G F  Ward The Transport of Oxidant Beyond Urban
     Areas  EPA-600/3-76-018a,  US   Environmental
     Protection Agency,   Research  Triangle   Park,  N C ,
     February  1976
41   Tnjonis,  J  Empirical  Relationships  Between
     Atmospheric Nitrogen Dioxide and  Its Precursors  EPA-
     600/3-78-018, US  Environmental Protection Agency,
     Research Triangle Park, N C , February 1978
42   Tnjonis,  J C, T  K  Peng,  G  J  McRae,  and L Lees
     Emissions and Air Quality Trends in the  South Coast Air
     Basin EQL Memorandum No. 16, California Institute of
     Technology, Pasadena, Calif , January 1976
43   Westberg,  H ,  K   Allwine,  E   Robinson,  and P
     Zimmerman  Light Hydrocarbon and Oxidant Transport
     Studies   m  Ohio—1974.  EPA-600/3-78-007,  U.S
     Environmental  Protection Agency, Research  Triangle
     Park, N C , January 1978
44   White, W H., and P T  Roberts The nature  and origins of
     visibility-reducing aerosols in Los Angeles  Presented at
     the   68th Annual  Meeting,  Air  Pollution   Control
     Association,  Boston, Mass , June 15-20, 1975
45   Wilson, K W Survey of Eye-Irritation  and  Lachrymation
     in  Relation to Air  Pollution, Coordinating  Research
     Council,  Inc , New York, April 15,1974
46   Wolff, G  T,P J  Lioy, G  D Wight, R F Meyers, and R.T
     Cederwall An investigation of long-range transport of
     ozone across the midwestern and eastern United States
     In  International Conference on Photochemical Oxidant
     Pollution  and Its Control.  Proceedings  Volume I. B
     Dimitnades, ed   EPA-600/3-77-001 a,  US   En-
     vironmental Protection Agency, Research Triangle Park,
     N C , January 1977 pp  307-317
28

-------
                  4.  SOURCES  AND  SINKS OF  OXIDANTS
PHOTOCHEMICAL FORMATION OF OXIDANTS

Introduction
  All the evidence presently  available  indicates
that in the urban centers and adjacent downwind
areas where severe oxidant problems occur, the
major  cause by  far  is photochemical oxidant
formation. In areas with less severe problems, and
especially in remote areas,  nonphotochemical
sources (e.g., stratospheric ozone intrusion) may
be  significant and  on  occasions  may be  the
dominant sources.
  When  photochemical oxidants  in air  were
identified as products of a photochemical process
involving  primary  air  pollutants,   it   became'
immediately apparent that the problem could not
be abated by traditional methods of direct control.
The chemistry of this process had to be clearly
understood before rational abatement measures
could be devised. Accordingly,  in the  ensuing
years, numerous studies of the oxidant chemistry
were   conducted  with  the   following  specific
objectives:
   1.   To identify the precursorsof photochemical
       oxidants (that is, those primary pollutants
       that participate as reactants in the oxidant-
       forming  process) for  the  purpose  of
       directing  the  control  effort to the right
       targets.
   2.   To determine the kinetics of the precursor
       reactions   mainly for  the  purpose  of
       deducing the impact of precursor control on
       oxidant formation.
   3.   To  determine  the  stoichiometry of  the
       precursor reactions for the purpose of more
       fully  assessing  the  pollution  problems
       caused by such precursors.
   4.   To elucidate in detail the  oxidant-forming
       mechanism for  the purpose  of  providing
       deterministic,  as distinct  from empirical,
       input to the oxidant abatement effort, and
       also for  the purpose of relating oxidant
       formation to  other  manifestations  of
       photochemical pollution.
  Several of these objectives have been achieved,
at least to the degree of completion such that the
information  generated  was  sufficiently compre-
hensive and reliable to permit development of a
crude but promising oxidant abatement strategy.
These chemical studies clearly are relevant to the
purposes of this document; however, they need not
all be included  in this review or expanded on in
detail except where they pertain to applications
directly related to the questions of oxidant control.
  To serve the purposes of this document better,
the discussion  of  the chemistry pertaining  to
photochemical  oxidants  will  be  divided and
presented  in  several  sections. The following
section  on  atmospheric reaction  mechanisms
includes  a general discussion  of the chemical
processes and  a review of recent atmospheric
chemistry studies, focusing only on those findings
that are judged to have had the strongest impact on
the  understanding of  the  oxidant  formation
mechanism. It is intended  and hoped that this
introductory discussion and review will  prepare
the reader to understand better the subsequent
sections dealing with the specific applications of
the  information used  in  formulating  oxidant
control strategies. In those sections, the reaction
mechanisms  and other pertinent aspects of the
oxidant chemistry will be discussed critically and
more comprehensively. For additional detailed
discussions of the  earlier studies of atmospheric
chemistry, the reader is referred elsewhere.54'78'92'
93
  Finally, there  is one  important difference in
emphasis  between  the  following reaction
mechanism section and the preceding chapters. In
the preceding chapters, ozone and other oxidants
were dealt with nearly always collectively, using
the term  "oxidant." In the discussions of reaction
mechanisms, however,  all mechanistic  inter-
pretations regarding oxidant, in actuality, pertain
to ozone  alone. Analogous  mechanistic
discussions addressed to other oxidants will not be
included  in this document  mainly because the
                                             29

-------
emphasis is on ozone and because the mechanistic
information  available  for  other oxidants  is
considerably deficient.

Atmospheric Reaction Mechanisms
  The pioneering work of Haagen-Smit in 195254
first  demonstrated  through  laboratory  ex-
perimentation that  the  photochemical oxidants
present in an urban atmosphere may indeed be
products of atmospheric photochemical reactions
involving organic   (e.g.,  hydrocarbons)  and
inorganic (e.g., nitrogen oxides) pollutants. Since
then, numerous studies of oxidant formation have
been conducted,4'5'78'93 and a wealth of information
is now  available regarding  the  stoichiometry,
kinetics,  and mechanisms of these HC-NOx-air-
sunlight  reactions.  The  following  discussion
summarizes  the  results of these studies and
presents highlights  of the chemical mechanism
that is presently thought  to explain  best  the
phenomenon of photochemical oxidant formation.
  All experimental reserach  concerned with  the
mechanism of atmospheric oxidant formation was
conducted in the  laboratory using experimental
conditions tKat were.similar to but not nearly as
complex  as  those  prevailing  in  the ambient
atmosphere. Such simplifications  of the natural
system were necessary to facilitate research, but
they also limited  the validity of  extrapolating
research findings to  ambient conditions. Thus
most mechanistic  evidence obtained to date
pertains  solely to the oxidant formation process
occurring in laboratory systems. The most that can
be said at this time is that all of the reaction steps
known to occur  in the experimental reaction
systems are almost certain to occur in the ambient
atmosphere  also,  although perhaps with varying
significance.  The converse, however,  may  not
necessarily be true; that is, unidentified reaction
steps may occur in the ambient atmosphere but not
in laboratory simulations.
  The overall chemical changes that are observed
to occur when a mixture of HC and NO, pollutants
is exposed to sunlig1-.,. under conditions similar to
those in the atmosphere are illustrated in Figure 4-
1. The curves of Figure 4-1 depict  the  overall
photochemical  process  as  consisting  of two
distinct reaction stages  occurring consecutively.
During the first stage, nitric oxide (NO) is converted
into nitrogen dioxide (NOz) without anyBppreciable
buildup of ozone (O3) or other non-NC>2 oxidants.
The second stage starts when almost all of the HO
has been converted into NOZ and is characterized
                   IRRADIATION TIME
IMW2 	 —
01 n
On 4 NO
yn \ NO
9Mn 4- n..
light
M



^~INVJ T vj
-K)3+ M
-^•NO 4- O
-^.MPi 4- vn
^HNL<2 ^ AU
-k-iNn..
Figure 4-1. Chemical changes occurring during photoirrad-
iationof hydro-carbon-nitrogen oxide-air systems

by rapid accumulation of ozone and other oxidant
and nonoxidant products.
  Several   mechanisms explaining the  obser-
vations  depicted  in  Figure  4-1  have  been
postulated. Although these mechanisms  differ
substantially  in chemical detail, they all explain
ozone formation in the atmosphere to be the net
result of the following main reactions:

                                        (4-1)

                                        (4-2)
                                        (4-3)
                                        (4-4)
                                        (4-5)
where M is the third body in a reaction, and X is
hydrogen (H) or organic radical (R or RCO). Through
this mechanistic procedure, the NO product from
the photolysis  of NO2 reacts  rapidly with and
consumes  ozone to  regenerate the photolyzed
NO2. Therefore, unless other processes convert
the  NO into  NO2,  ozone  is  not allowed  to
accumulate to significant levels. However, other
NO conversion processes do  exist—mainly  the
reactions of NO with XO2  and, to a much  lesser
degree,  with  molecular oxygen (O2>. The reaction
with XO2  occurs only when  photochemically
reactive organic compounds are present, in which
case the reaction can be sufficiently rapid to cause
atmospheric accumulation of ozone to significant
levels. In the absence of reactive organics, the only
NO conversion process parallel  to the  O3-NO
reaction is  the reaction  of  NO  with 02; this
reaction, however, is relatively slow and does not
cause significant ozone accumulation.
  In mathematical terms, the ozone buildup in the
atmosphere obeys the equation28
     [O3]   =        [NO2]               (4-6)
                     [NO]
30

-------
where I  is light (sunlight) intensity, and k is a
constant. This equation is derived by applying the
steady  state  hypothesis to  chemical  reaction
steps 4-1, 4-2, and 4-3. This equation illustrates
the effect on ozone buildup of any process that
converts NO into NO2 (e.g., the reaction of NO with
X02>; such effect is toward high [N02]: [NO] ratios,
and  hence  high levels of  ozone  buildup. Ambient
measurements in downtown   Detroit132 in the
summer  of 1973 and  from  the  Los  Angeles
Reactive  Pollutant   Project   (LARPP)18  have
essentially  verified  the   photostationary  state
equation in tie atmosphere within the stochastic
limits of the turbulent atmosphere.124
  Though  the  overall  mechanistic  scheme
depicted  by  processes  4-1  through 4-5  was
established  long  ago,   the detailed reactions
explaining the consumption of the organic and
inorganic reactants and the formation of the X02
radicals (e.g.,  reaction  4-4) have been relatively
obscure.  It  is  in this latter mechanistic area that
some important developments have taken place in
the past 5 to 6 years. These new findings and their
impact on current understanding are summarized
here. The significance of these findings in the
specific applications of reaction mechanisms in
the development of mathematical models will be
discussed  in  a  subsequent   section  of  this
document.
  One important accomplishment in recent years
in the area of atmospheric reaction mechanisms is
the  development  of  computer  techniques for
simulating  the  atmospheric  smog-forming
process.19'40 Such  techniques  have  provided  a
useful tool for selecting those reactions that play a
key  mechanistic  role  and  for predicting  the
existence  of  potentially  important but  as yet
unidentified reaction  products  in  the  ambient
atmosphere.
  The  most   important  mechanistic  finding,
however, in recent years pertains to the identities,
sources, and roles of the  radicals responsible for
the  oxidation  of  the organic and  inorganic
reactants. Unlike  the  early emphasis  on atomic
oxygen,  0(3P), and  ozone  roles,  the  current
thinking  attributes to the hydroxyl  radical (OH)
most  of  the  hydrocarbon-  and   aldehyde-
consuming  chemical activity.58'133
  Other radicals also, most notablyatoms and RO2,
can  attack and  consume the  organic reactant
significantly, as illustrated by the data inTable4-1.
These data, derived from a computer simulation of
an irradiated  frans-2-butene/NOx  reaction sys-
tem,  represent calculated  rates of attack of several
   TABLE 4-1. CALCULATED RATES OF ATTACK ON
7"fl/l/VS-2-BUTENE BY VARIOUS REACTIVE SPECIES IN
      SIMULATED SMOG SYSTEM AT SEVERAL
              IRRADIATION TIMES"
Attack rate, ppb/mm
Species
0
03
OH
H02
N03
02 (cr'A)
2 mm
0013
0026
1 72
0.16
0.05 x 10^"
2.9 x 10"6
30 mm
0018
0 16
055
0.15
22x10""
2 1 x 1Q"6
60 m
0.011
016
0.27
009
2.6x
1.3 x
m




10'4
10-"
 aData from Demerjian et al 40 Initial conditions [NO]. 0 075 ppm, [N02], 0 025 ppm,
  [frans-2-butene], 0 1 0 ppm, [CO], 10ppm, [CH2O], 0 10 ppm, [CH3CHO], 0 06 ppm,
  [CH<], 1 5 ppm. relative humiuity, 50%

 reactive species on butene, at several irradiation
 times.

  The radicals OH, H02, and X02 have also been
 identified as having major roles in the oxidation of
 NO into  N02 (reaction 4-4).93 Sources of these
 important radicals are  now  believed to  be the
 reactions:
     (Aldehydes,  MONO, 03, others)
        hv/  »OH, H02, R02, others
(4-7)
  Oxidation of  NO  into N02  occurs via a chain
mechanism illustrated by the following sequence:
OH + (hydrocarbons, aldehydes)	
                    H02, R02, others      (4-8)
     H02+NO  - N02 +OH                (4-9)
     R02+N0-R0                     (4-10)
     RO + 02 - H02 + aldehyde           (4-11)
  Such  chain  processes  are terminated when
radicals react to form  more stable products, for
example.
OH + N02-HON02                     (4-12)
H02 + H02 (or R02)  - H202 (or ROOH)      (4-13)
HO2 + NO2 - MONO (or HOON02?)        (4-14)
RC(O)O2 - RC(O)02N02 (i.e., PAN)         (4-15)
HO2, RO2 + surface  - (H02, R02) surface   (4-16)
  A  more complete  picture  of the   reactions
accounting for  the  organic oxidant degradation,
oxidation  of  NO into  NO2, and  organic product
formation  in the ambient atmosphere, is illustrated
in Figure 4-2 for the case in which the organic
reactant  is an  olefin.40  Comparably  complete
mechanistic  information  is  now  available for
paraffins also, but not for aromatic hydrocarbons.

  In the inorganic part of the atmospheric reaction
mechanism, some steps recently have becomethe
subjects of controversy because of their different
roles  in  the ambient  atmosphere  and  in  the
laboratory systems.  Thus, the reactions
                                                                                           31

-------
                                                     il14.
                O
         CHBCH-CHCHS
CHSCH=CHCH3
                                                             -»-CH,=CHCH(O, H)CH,
                                                       0('P)
                              CH3CH(6)CH(6I)CH|
                         CH3CH(O)CHCHJ
                                                                        *>CH,CO,  + NO,
                                                                            AS BEFORE
                               * These products undergo significant photodecooiposition in sunlight
    L.
  Figure 4-2. The major reaction paths for the degradation of ffans-2-butene in an irradiated IMO,-polluted atmosphere.
32

-------
              CHjCH^HCH,
         HO/""
CH,CH=CHCHa  +  H,O

CH,CH=CHCH,6,

     O^o
     I     I
CH,CHCHCH,

       <°"
    O

CHjCHCHiO^CHj
                                                        HO
                                                        CH,CHOHCHCH,
                                                                O,
                                                               1
                                                               "
NO
                                                        CH,CHOHCH(O)CH9 ^ NO,
                                                        CHjCHOH
                                                              I'
                                                        CHaCH(0,)OH
                                                              | NO
                                                        CHjCH(6)OH  -i- NOa
Figure 4-2. (Continued). The major reaction paths for the degradation of frans-2-butene in an irradiated NO.-poiluted
atmosphere.
                                                                                                    33

-------
         NO + N02 +H20 - MONO       (4-17)
         N205 + H20 - 2HN03          (4-18)
are known  to occur both homogeneously (in the
gas phase)  and heterogeneously. Both reactions
are thought to be of relatively little importance in
the ambient  atmosphere but  are suspected  to
occur  significantly  on  the  reactor  walls   in
laboratory systems.1771  Questions are thus raised
about the applicability of the laboratory findings to
the ambient  atmosphere,  which is  an issue  of
crucial importance.
  Another  noteworthy  finding  is  that  some
nitrogenated products of the photochemical smog
system, such as PAN and possibly peroxynitric acid
(HOON02) have a greater  mechanistic role than
thought earlier.  Recent  investigations of PAN
chemistry 34'60'98  have  revealed that  PAN  can
thermally decompose to an acylperoxy radical and
N02.  The  rate of  decomposition  is  extremely
temperature  dependent.  Because of  this tem-
perature dependency, significant levels of PAN can
build  up early in the day when temperatures are
relatively low. In the late afternoon when ambient
temperatures  are  higher, the  decomposition  of
PAN can proceed at a  rapid rate, liberating  N02
molecules  that  can lead  to  enhanced  ozone
production.
  The discovery  that  PAN  can  thermally
decompose  and  enhance  ozone  formation
suggests that there may be other peroxynitrates
that can  also  affect  the rate of smog formation.
Recently,  peroxynitric  acid   (HOON02)  was
identified   by  Fourier-transform infrared spec-
troscopy  (FS)
              79,95
as  an  intermediate  of
photochemical smog systems. This species and the
related  peroxynitrates (ROON02)  could  act  as
radical sinks and affect the rate of smog formation.
The important H02 and R02 free radicals, as well as
NC>2, could be temporarily stored as peroxynitrate
and then released (through thermal decomposition
of the peroxynitrate) at later stages in the reaction
to  enhance ozone production.  However, recent
experimental evidence33'50'125 indicates that the
decomposition of HOONOz proceeds so rapidly at
room temperature that this species will  not be a
significant sink  for HC>2 or N02. Consequently,
HOON02  is  not  likely to be of significance in
photochemical smog formation.
   This role of the peroxynitrate products raises the
more  general  question  regarding the role  of
atmospheric reaction products in the chemistry of
polluted air masses irradiated  for several days.
Such multiday reaction of  pollutant  mixtures is
now known to occur in the ambient atmosphere
and specifically, within urban air masses subjected
to long-range transport.136 While the chemical
mechanism  of  such  aged systems  cannot be
different from that of fresh systems in terms of
fundamental reaction steps involved, the relative
rates of such steps may be sufficiently different
that  the  relative  roles of  the  organic and NO*
reactants, as well as the relative reactivities of the
various organics in producing oxidant/ozone, may
be substantially different.  Evidence attesting to
such differences has  recently  been reported by
researchers  who  did laboratory and computer
simulation  studies  of fresh and  aged  HC-NO*
reaction systems.  Thus smog chamber  studies
have  suggested  that the  oxidant-to-precursor
dependencies in aged systems may be different in
magnitude, and possibly in direction also, from the
dependencies  observed in  fresh  systems.41'71
Current knowledge and  understanding  of  the
atmospheric chemistry of  aged polluted air are
considerably inferior to those pertaining  to fresh
atmospheres.
  The chemical transformation of  S02  to par-
ticulate sulfates is of considerable interest, both in
terms of understanding S02 removal mechanisms
in the atmosphere as well as the potential health
and welfare effects96 associated with particulate
sulfates.  Though  it is beyond the scope of this
document to provide a detailed review of the state-
of-the-art  of  SOx  chemistry, a  short discussion
regarding  its interaction in the photochemical
process seems warranted.
  In  recent  years,  significant progress toward
understanding the homogeneous  SO2 oxidation
pathways  in the atmosphere has  been  made.16'
119'122 One of the more important contributions has
been the experimental gas kinetic rate constant
determinations for the H02 and HO reactions with
S02: 20'99
         H02 + S02- H0  + S03
         HO + S02 + M - HOSO2+ M
  The quantification of  these  reaction  rates in
conjunction  with   theoretical estimates  of  the
reaction rates for the organic  radical analogies
R02 and RO have  explained major portions of the
S02 transformation observed in both experimental
smog chamber studies and field investigations.
16,119,142
  In  the  presence  of water vapor,  S02  reacts
rapidly to form H2SO4. The details of the  reaction
pathway  for the  HOS02  species  is currently
unknown,  but  evidence  exists indicating  that
 34

-------
H2S04  is  the  resulting  end  product.  In the
atmosphere,  H2S04 may undergo complex  inter-
actions with  aerosol  particulate or  react with
gaseous ammonia to form a host of sulfate salts.
Characterization of these condensable species via
experimental and ambient observations will be a
major area of continued research and will play a
significant role  in elucidating the mechanistic
details of atmospheric SO*.
  Outside the reaction mechanism area,  some
significant developments have  taken place  in
recent  years  in the area  of  reaction  product
identification. Table 4-2 lists gaseous products a nd
typical  concentrations  at which  such products
were found to occur in  the ambient atmosphere.
Table 4-3 lists gaseous products either observed in
laboratory (smog chamber) systems or unreported
yet  but theoretically expected to  exist in the
ambient atmosphere.  Particulate products also
exist but are not easy to identify. Such products are
certain to include nitrates and a variety of organic
compounds. Tables 4-4 and 4-5 list some reaction
products  identified  in ambient  atmosphere
aerosols.93 The significance of these  findings is
that they provide a more complete description of
the photochemical pollution problem and that they
verify orfurtherelucidatethe atmospheric reaction
mechanism.

       TABLE 4-2. COMPOUNDS OBSERVED
           IN  PHOTOCHEMICAL SMOG
                TABLE 4-3. COMPOUNDS THAT MAY BE FORMED IN
                           PHOTOCHEMICAL SMOG
                                Typical (or maximal)
                                concentration reported,
        Compound
 Ozone, O3
 PAN, CH3COO2NO2
 Hydrogen peroxide, H2O2
 Formaldehyde, CH2O
 Higher aldehydes, RCHO
 Acrolem, CH2CHCHO
 Formic acid, HCOOH
01   (06)
0005 (0.2)
     (0.18)
     (016)
     (0.36)
     (0011)
     (005)
004
  In conclusion, the information generated in the
years since the issuance of the preceding criteria
document provides a much more complete picture
of the atmospheric oxidant  formation  process.
Nevertheless,  it  should be  stressed that the
present situation is somewhat uncertain because
most of the supporting evidence was obtained from
laboratory atmospheres and  is therefore mainly
applicable  to  such  an  environment.   The
mechanism of the oxidant-forming process in the
ambient  atmosphere,  containing  a  multitude of
reacting pollutants, may include reaction steps in
addition  to  those  presently recognized.  Such
Compound
Peroxybenzoyl nitrate,
C6H5C002N02
Nitric acid, HONO2

Organic hydroperoxides,
ROOM
Organic peracids, RCOO2H
Organic peroxynitrates,
RO2NO2
Ozonides, O3-olefin
Ketene, CH2CO
Nitrous acid, MONO
Pernitnc acid, HO2NO2
Pernitrous acid, HO2NO
Organic nitrates, RONO2

Possible synthesis
0COO2 + NO2

N02 + OH
N2O5 + H2O
RO2 + HO2

RCOO2+ HO2
RO2 + NO2+ M*

O3 + olefm + M"
O3+ olefm
OH + NO
N02 + HO2+ M*
NO + HO2 + Ma
RO + NO2
RO2 + NO
Reference
62

55
91
40

40
40

7
88
8,32
50,79,95
31
37
37
  aM represents any molecule that takes part in the three-body process

 additional reactions,  for  example,  known  or
 suspected to occur in the ambient atmosphere, are
 those  causing degradation  of  aromatic  hydro-
 carbons and of numerous  organics into gaseous
 and  particulate  products. To  date,  these
 degradations  are largely unexplored  Important
 reactions may also be occurring through  energy
 transfer processes promoted by pollutants or other
 molecules capable of absorbing solar energy and of
 transferring  such  energy  to  nonabsorbing
 pollutants. Also,  the  nature and  importance  of
 heterogeneous reactions occurring on the surface
 of ambient aerosol particles49 and on the surface of
 laboratory chambers17 have been explored  but are
 not well understood. Further studies are needed to
 ascertain the applicability  of laboratory data  to
 ambient atmospheres.

 Effects of Meteorological Factors
 INTRODUCTION
  The   photochemical  oxidant/ozone   con-
centrations  observed  m  the ambient air  above
 urban and nonurban areas arethe net result of two
broadly defined processes: First, a physical process
 involving dispersion and transport of the oxidant-
precursor emissions;  and  second,  a chemical
process  involving  reaction  of  the dispersed
pollutants under  the  stimulus of sunlight.  The
potential effects of meteorological factors on  both
these  processes  are obvious. Thus the factors
related to atmospheric dilution and transport affect
ambient levels of  pollutant accumulation and the
geographical  relationship between source areas
and corresponding oxidant  problem areas. Solar
radiation  and  ambient  temperature  are  also
important by virtue of their effects on the chemical
                                                                                            35

-------
                            TABLE 4-4. SECONDARY ORGANIC AEROSOLS2
              Compounds identified"
                                                                  Possible gas-phase hydrocarbon
                                                                 	precursors	
 Aliphatic multiiunctional compounds
 1.  X-(CH2)n-Y (n=3,4,5)-
              X                   Y
              COOH          CH2OH
              COOH          COM
              COOH          COOH
              COOHb          CH2ONO
      or       COH          CH2ONO2
               COH          CH2OH
               COH          COH
              COOHb          COONO
      or       COH          COONO2
               COH          COONO
              COOH          COONO2
              COOH          CH2ONO2

 2  Others
    CH2OH-CH=C(COOH)-CHO
    CH2OH-CH2-CH=C(COOH)-CHO
    CHO-CH=CH-CH(CH3)CHO
    CH2OH-CH=CH=CHb-C(CH3)CHO
    CsHeOs isomers
    Nitrocresols
    CeH6O2 isomers"

 Aromatic monofunctional compounds
 3  C6H5-(CH2)n-COOH (n = 0,1,2,3)
 4  C6H5-CH2OH
    C6H5CHO
    Hydroxynitrobenzyl alcohol

 Terpene-denved oxygenates.
 5  Pinonic acid
    Pinic acid
    Norpmonic acid

 6  Isomers of pmonic acid "
    CgHitOz isomers
    CioHi4O3 isomers
    CioHi6O2 isomers
                                                     1   Cyclic olefms
                                                       (CH2)n
CH
CH
                                                       and/or diolefms

                                                   >C=CH-(CH2)n-CH=C<




                                                     2  Not known; possibly from aromatic ring cleavage
                                                     3  Alkenylbenzenes

                                                        C6H5-(CH2)n-CH=CHR; also toluene for C6H6COOH

                                                     4  Toluene, styrene, other monoalkylbenzenes7




                                                     5  a-Pmene



                                                     6  Other terpenes?
 "Compounds identified at West Covma, Calif, July 24, 1 974
 blsomers not resolved by mass spectrometry

processes.   Such  meteorological  influences
obscure  both the absolute and the relative effects
of the emission-related factors to degrees thatvary
with geographical location and season. Therefore,
for proper assessment of emission-related factors,
it is essential that meteorological factors and their
effects on the oxidant problem be well understood.
Such  understanding  often  depends  on  mete-
orological details that are far  beyond the scope of
this discussion, but some general climatic factors
may be discussed to  illustrate their impact on air
quality.

ATMOSPHERIC MIXING
  The diurnal urban emission pattern for oxidant-
forming  pollutants is fairly uniform from weekday
                                                   to   weekday.   It  is  apparent,  therefore,  that
                                                   variations  in  daily oxidant/ozone  accumulation
                                                   must  be  attributable largely  to meteorological
                                                   factors. Of these factors, atmospheric mixing has
                                                   an  extremely  significant  effect  on  oxidant
                                                   formation.
                                                     The rate and extent of atmospheric mixing and
                                                   diffusion  depends on stability, wind speed, and
                                                   topography. Temperature inversions are relatively
                                                   stable  layers  that inhibit  atmospheric  diffusion.
                                                   They  are common  at  night in layers  near the
                                                   ground, but they may also occur at higheraltitudes
                                                   and at  times  other  than  at night.  Also,  several
                                                   successive  atmospheric  layers with  different
                                                   degrees of stability are not unusual. The morning
                                                   peak concentrations of  oxidant/ozone precursors
36

-------
             TABLE 4-5. RELATIVE IMPORTANCE OF ALIPHATIC AND AROMATIC PRECURSORS"
Secondary organic
aerosols
X-(CH2)n-Y
Gas-phase hydrocarbon
precursors X
COOH
COOH
CH COOH
(CH2)n COOH
CH or COH
COH
COH
. . COOH
and/or or CQH
COH
COOH
COOH

Y
CH2OH
COH
COOH
CH2ONO
CH2ONO2
CH2OH
COH
COONO
COONO2
COONO
COONO2
CH2ONO2
Concentration," /yg/m3

n=3
2 18
1 39
1 35
1 01
—
0 31
0.30
0 14
1 01
—
0 12

n=4
3.40
259
0.78
0.40
—
0.40
0 24
024
0.14
—
0 15

n=5
065
082
0 15
0 27
—
013
—
—
—
—
—
  >C=CH-(CH2)n-CH=C<
     C6H5-CH=CHR
               Total

COOH-CHz-COOH
COOH-(CH2)2-COOH

  Total difunctional compounds  18.89

  C6H5-(CH2)n-COOH



  Total from aromatics.
781

0.15
057
          834
          n=0.0.38
          n-1 041
          n-2 0 52
          n-3 0 03
             1.34
2.02
 'Of aerosols in Pasadena, Calif. Sept 22, 1972, sampling period, 7 30 a mjo 1 2 35 p m
 The same response factor (that of adipic acid) was used for all difunctional compounds
often result from emissions from ground sources
discharged into a  stable  atmosphere  below an
inversion  layer.  Typically, these  low-level
inversions begin to lift and provide a greater mixing
height as the  sun rises  and heats the earth's
surface. Thus mixing height (that is, the height
above surface through which relatively vigorous
vertical  mixing  occurs) varies during the day and
with  the  season.  Figure 4-365 shows average
mixing heights  in summer (usually the season of
highest  photochemical  oxidant concentrations)
that occur a few hours after sunrise. As surface
heating  continues,  the   low-level  inversion  is
usually eliminated  completely and, as depicted in
Figure 4-4,65 the  mixing  height  increases to a
maximum  in the  afternoon.  Notice that these
average afternoon  mixing heights range from 600
m (1970 ft) along the California coast to  more than
4000m   (13,000 ft) within the southern Rocky
Mountain area.  Over inland areas, the afternoon
mixing heights are greatest in summer and least in
winter (about half  of the  summer values). How-
ever, seasonal  variations  in  coastal regions are
usually small. To a large extent, the height of the
afternoon mixing layer has a significant impact on
the day's accumulation of oxidant/ozone.
                      The rate at which relatively unpolluted air moves
                    into a region is related to wind speed. Wind speed
                    and direction are highly variable in time and space
                    (Figures 4-5 and 4-6).65 The depicted wind speeds
                    are averages of those speeds within the respective
                    mixing layers.  These averages tend to smooth out
                    the greater variations in spaceandtime that would
                    occur it only surface winds had been considered.
                    During summer mornings, the slowest  average
                    speeds are less than 2 m sec"1 and occur in the Far
                    West. However, in the morning, average speeds of
                    3 m sec^1 or less occur over most of California and
                    Oregon as well as part of Nevada;  in the East they
                    occur  over   the  Central  Appalachians  and
                    Mississippi.  On summer afternoons, the  average
                    wind  speed pattern  is similar to that  in  the
                    morning, except that the average speeds in the
                    afternoon tend to be 1 to 2 m sec' higher than in
                    the morning. This  diurnal variation is generally
                    true for all seasons.
                       Using daily values of morning and afternoon
                     mixing height and wind speed, the occurrence of
                     limited mixing episodes was determined for 62
                     weather  stations   in  the  contiguous United
                     States.65 The  most limiting conditions were mixing
                     heights of 500 m (1 640 ft) or less, wind speeds of 2
                                                                                             37

-------
          6  5
                             Figure 4-3. Isopleths (m * 102) of mean summer morning mixing heights.65
                             Figure 4-4. Isopleths (m * 102) of mean summer afternoon mixing heights.65
38

-------
    Figure 4-5, Isopleths (m sec ') of
                                   mean summer wind speed averaged through the aftwnoon misting layer,"*
Figure 4-6, Isopleths (m sec"') of mean summer wind speea averaged through the morning mixing layer,66
                                                                                                              39

-------
    10
                            DATA BASED ON FORECASTS ISSUED:
          1  AUGUST 1960 TO 3 APRIL 1970 FOR EASTERN PART OF THE UNITED STATES
         1 OCTOBER 1963 TO 3 APRIL 1970 FOR WESTERN PART OF THE UNITED STATES
      Figure 4-7. Isopleths of total number of forecast days of high meteorological potential for air pollution.65
m sec 1 or less, and no precipitation during at least
5  consecutive  days.  Only six  such epidoses
occurred with a total of 39 episode-days at two
stations (Medford, Oregon, and Lander, Wyoming)
over a 5-year period. On the other hand,  at least
one  episode of 2 days' duration,  with  mixing
heights  of  1000m   (3300 ft)  or less and  wind
speeds of 6 m sec ' or less, occurred at each of the
62 stations. In general, limited mixing episodes are
most common in the Far West and within the Rocky
Mountain region, least common over the Plains
States, and of intermediate frequency east of the
Mississippi (Figure 4-7).66
  Korshover76 compiled statistics on stagnation
episodes occurring during the various months of
the year east of the Mississippi. Such statistics are
illustrated  by graphs  in  Figures 4-8 and  4-9.
Korshover's  data,  however,  suffer from  two
limitations   First,  stagnation  episodes  were
determined  indirectly, using sea level pressure
gradient data, rather than the more direct indices
of  mixing height and wind speed;  second, the
requisite pressure gradient data could be obtained
only for the relatively low and flat terrain east of the
Rocky  Mountains. A more detailed discussion  of
the Korshover statistics is included in the chapter
dealing with the natural sources of oxidant/ozone.
  In  conclusion,  the  atmospheric   mixing
parameters  have a strong  effect on both the
accumulation of oxidant-precursor pollutants and
the photochemical oxidant-forming process. The
effect  on the  latter process,  however, although
certain to exist, has only recently been studied
directly  and comprehensively;73'118  changes  in
these  mixing  parameters cannot be simulated
easily  in smog  chambers.   Nevertheless,  the
information  presented here is useful in that it
suggests the  areas  and  seasons in  which
atmospheric mixing conditions are relatively more
conducive to oxidant formation and accumulation.

SUNLIGHT
  The  significance  of  sunlight is related to the
intensity of  sunlight and its spectral  distribution,
40

-------
                                                 105°  100°  95°   90°   85°   80°   75°    70°     65°    60°
                                          50'
                                          45'
                                          35°
                                          30'
                                          25°
                                              Figure 4-8. Monthly distribution of number of cases of 4 or
                                              more  days  of  atmospheric  stagnation,  1936-75
                                              (September).76
                                                                                                              45
                                                                                                              40°
                                                                                                              35
                                                                                                              30°
                                                                                                              25°
                                                                                                              20°
                                            105°       100C
                                                         95"
                                                                          90°
                                          95°
                                                                         80°
                                                                                                        75
       105   100°  95°   90°   85°    80°   75°   70°     65°    60°
                                                          I       I
50
45
40'
35
30r
25"  —
'Figure 4-9. Monthly distribution of number of cases of 4 or
 more days of atmospheric stagnation, 1936-75 (August).76

      I _ [ _ i _ | _ |
                                                                    45°
                                                                    40°
                                                                    35'
                                                                    30
                                                                    25°
                                                                    20°
105°       100°
95°
86°
                                            80°
                                                               75°
                                                                                                            41

-------
both of which have direct effects on the specific
chemical reaction steps that initiate and sustain
oxidant formation. Steps of varying  significance
include the following (and possibly others also):
            NO2 + hv - 0 + NO
            MONO + hv/- OH + NO
            H2CO + hv/ - H + HCO
            RCHO + hv/- CH3 + HCO
            03 +  hv/- 0 ('D) + 02
            H202 + hv - 20H
            HN03 + hv/ - OH + N02
                                        (4-19)
                                        (4-20)
                                        (4-21)
                                        (4-22)
                                        (4-23)
                                        (4-24)
                                        (4-25)
Whereas the effect of sunlight intensity is direct
and  amply  demonstrated,78  the  effect  of
wavelength  distribution  on the overall  oxidant
formation process is a subtle one  Experimental
studies have shown the photolysis of aldehydes to
be strongly dependent on radiation wavelength in
the near-UV region.28 Since aldehydes are major
products in  the atmospheric photo-oxidation of
HC-NOx mixtures, it is inferred that the radiation
wavelength should have  an effect on  the overall
photo-oxidation  process.  This  inference was
directly verified, at least for the propylene/NO, and
/i-butane/NOx chemical systems, in recent smog
chamber studies.70 In the  ambient atmosphere,
some variation in the wavelength distribution of
sunlight  does occur as  a result of variations in
stratospheric ozone, ambient aerosol,'31 and cloud
cover.   Obviously,  such   variation  should  be
recognized and properly considered in conducting
and interpreting outdoor experimentation. Beyond
this  consequence,  however,  the wavelength
distribution  factor  has  no  other  practical
significance.
  Sunlight  intensity  vanes with  season  and
geographical latitude, as  shown in  Figure 4-10.78
The  latitude effect is  strong, but only  during  the
winter months. During summer, throughout  the
contiguous  United  States,  the  maximum light
intensity is fairly constant, and only theduration of
the solar day varies to a small degree with latitude.
Some variations in light intensity also  occur with
longitude during  the  summer months, with  the
highest intensities occurring in the western United
States.
  Absolute  levels of sunlight  intensities were
calculated by Leighton nearly two decades ago.78
Since  then,  several investigators  have  re-
calculated actimcfluxes and measured  intensities
by  more  direct   methods.  Thus   recent
measurements  and recomputations  by  EPA
investigators  gave results somewhat  different
from those reported by Leighton. These differences
                                                        VALUES OF SOLAfi SPECTRUM
                                                        ARE FOP THE 100 /\ INTERVAL
                                                        CENTERED AT 3700 \
 SUMMER SOLSTICE
 I 20 N. LAT
 II  34 N. LAT
 lit 50 N. LAT.

 WINTER SOLSTICE
" IV 20 N. LAT
 V 36 N. LAT.
 VI 50 N LAT
                                                  a
                                                  t-
                                                  o
                                                  I
                                                   Figure 4-10. Diurnal variations in actinic irradiance.78
                                                 are summarized in Table 4-6.)0° The recomputed
                                                 values  are lower than Leighton's in the 295- to
                                                 395-nm wavelength interval, slightly higher in the
                                                 395- to 450-nm interval, and considerably higher
                                                 in  the  450-  to 700-nm  interval. The same EPA
                                                 investigators  expressed their  recomputed  light
                                                 intensities also in terms of values for the NO2
                                                 photodissociation constant, k, (min ').101 Resulting
                                                 values  are listed in Table 4-7. These values are
                                                 higher  than the earlier values used in laboratory
                                                 simulations of polluted atmospheres. Jackson et
                                                 al.,69 Zafonte,147 and Harvey et al.56 measured k,
                                                 directly; their values also were higher than those
                                                 based  on Leighton's calculations or on  data
                                                 obtained  by standard radiometric techniques.

                                                 TABLE 4-6. PERCENTAGE DIFFERENCE BETWEEN THE
                                                  PETERSON100 AND THE LEIGHTON VALUES AT THE
                                                  EARTH'S SURFACE OVER SELECTED WAVELENGTH
                                                      INTERVALS AND SOLAR ZENITH ANGLES

nm
295-395
395-450
450-700

0
-6.1
+2.4
+ 18.8
Zenith angle (c>
20 40
-64 -55
+2 1 +3.4
+ 18.4 +17.0

60
-43
+5.9
+ 15.7
42

-------
      TABLE 4-7. CALCULATED VALUES OF NO2 PHOTODISSOCIATION RATE CONSTANT3 AT THE EARTH'S
 SURFACE AND PERCENTAGE INCREASE OF THE RATE CONSTANT FROM THE SURFACE TO VARIOUS HEIGHTS101
Zenith angle (°)
Height
Surface
0 15 km
036 km
064 km
098 km
1 84 km
2.91 km
0
0.579
60
11 4
16.6
21.4
29.9
364
10
0574
6.1
11 5
16 9
220
303
37 1
20
0.560
6.4
12 1
17 7
22.9
31 6
35.8
30
0535
69
13.1
19 1
24.7
340
41 7
40
0496
77
145
21 2
272
377
462
50
0432
89
16.9
244
31 5
434
537
60
0352
11 1
21 0
304
392
53.7
666
70
0231
143
277
407
52.3
732
91 3
78
0 114
167
342
52.6
702
101 0
132.0
86
0025
80
200
350
48.0
720
1000
  aln units of mm"1
  The light intensity factor has been studied in the
laboratory both  with  respect  to  its  effect on
individual  photolytic reaction  steps  and  with
respect  to its  effect on  the overall  process  of
oxidant  formation.94 All  of the  early studies,
however,  employed constant light intensity
conditions,  in  contrast to the  diurnally  varying
intensity in the ambient atmosphere. Only within
the last  3 years has the diurnal variation of  light
intensity been  recognized  and studied as a factor.
Such studies have shown this factor to have a
varying  (with  initial  reactant   concentration
conditions), rather unpredictable,  but somewhat
significant effect.72

TEMPERATURE AND RELATIVE HUMIDITY
  Effects of temperature and relative humidity on
photochemical oxidant formation were suggested
by  early laboratory  studies  discussed   in the
predecessor of this criteria document.92 Additional
laboratory and field  studies in more recent years
verified  a significant,    positive  temperature
effect.1073Thus outdoor smog chamber studies of
synthetic HC-NOx mixtures showed that in winter,
with maximum daily temperatures at 50° to 60°F
(10.0° to 15.6°C), oxidant yields were lower  than
those in the summer  (80° to  90°F,  or 26.7° to
32.2°C)  by 70 to90percent, depending on reactant
composition.73 Also, an RTI analysis  of field  data
showed   (1)  good  correlation  between   daily
maximum 1 -hr oxidant levels and daily maximum
temperature;  and (2) no  oxidant concentrations
greater than 0.08 ppm on days when temperatures
were below  62°F  (16.7°C).  The evidence10'73
suggests that  below about 55° to 60°F (13° to
16°C), photochemical oxidant concentrations are
not likely to exceed the 0.08 ppm standard. Also,
information on the  temperature factor indicates
that in the summer, temperature  conditions are
conducive to ambient oxidant formation through-
out the United States as far north  as Alaska.
  Unlike the temperature  factor,  the  effect of
relative humidity remains somewhat uncertain.
Reveiws by Quon and Wadden103 and by Altshuller
and Bufalini4 have pointed out inconsistencies in
the  available evidence.  Model  simulations of
irradiated HC-NOX predict only a small humidity
effect;94 however, these predictions have not been
verified by consistent experimental evidence.
TRANSPORT PHENOMENA
  Relative  contributions  of  natural  and  an-
thropogenic sources in a given region or area are a
major  question  relevant  to  the  ambient
oxidant/ozone problem. This question has  not
been  answered unequivocally and quantitatively
chiefly because anthropogenic pollutant transport
makes it difficult to assess the strength of  the
natural sources. The question has been answered
qualitatively. The consensus is that oxidant/ozone
and/or oxidant precursor transport does occur and
contributes to oxidant/ozone buildup in areas far
from  the sources. The magnitude of this con-
tribution and the magnitude of the contributions
from  the stratosphere and  from other natural
sources have been viewed as unresolved issues.43
  Aside from  its connection  with the natural-
versus-anthropogenic-sources   question,  the
phenomenon of oxidant transport is important for
another  reason.  The  phenomena of  urban
oxidant/ozone plume formation and  movement,
rural   oxidant/ozone   occurrence  (at  problem
levels),  Sunday-weekday effect,  and  nighttime
oxidant/ozone occurrence were previously either
unnoticed  or thought to be odd.  They  are now
believed to be true manifestations of an extremely
complex emission/pollutant  dispersion process.
Such complexities may be introduced by horizontal
and/or vertical transport of oxidant/ozone and/or
of precursor mixtures for long distances without
excessive dilution.
                                                                                          43

-------
  Consequences  of  pollutant transport  are
discussed here.  Several studies concerned with
oxidant transport have been conducted in the last 5
years. Such studies have resulted in a wealth of
information. This information and its implications
are presented and  discussed in the remainder of
this section.
  The  first item of interest is evidence on the
occurrence  and  range  of  oxidant  transport.  It
should  be  clarified  that the term   "oxidant
transport" is commonly used to refer to transport
of either oxidant/ozone or oxidant precursors. The
distinction  between  oxidant/ozone  and oxidant
precursors should be kept in mind, as it is essential
to the understanding of the nature, mechanisms,
and implications of the oxidant pollution problem.
  Oxidant transport has been recognized to occur
on three  scales in terms of geographical distance
or area subjected to the transport effects. These
scales are.
   1.   The urban-scale  transport, as a  result of
       which  the  peak  oxidant concentrations
       develop  in  the  suburbs  some  miles
       downwind from the city core  area  where
       the  oxidant/ozone and their precursors
       originated;
   2.   The mesoscaletransportthatencompasses
       land- and sea-breeze circulation, and the
       formation of  urban oxidant  plumes  that
       create oxidant problems asfaras 100 miles
       or more downwind from the  source  city;
       and
   3.   The  synoptic-scale transport,  a  much
       longer  and  broader  range  of  transport
       associated with high pressure systems
  Urban-scale transport  has been observed in Los
Angeles,134  New York,  Houston, Phoenix,  and
several other urban centers.327 All of the evidence
available points to  the conclusion that conditions
at the center of source-intensive areas are not the
most  conducive  conditions for oxidant  accumu-
lation, mainly because of the  strong scavenging
effect of oxidant precursors, especially nitric oxide.
At higher  elevations or  at horizontal  distances
downwind from the sources, where the precursor
scavenging  effect  is  less important,  oxidant
concentrations are, in general, greater, and their
levels are determined primarily by the intensity of
the photochemical activityandbyambientdilution.
Measurements in several cities during days with
various wind  speeds showed  the peak oxidant
concentrations to occur at distances 8 to 1 36 km (5
to 85  miles)  downwind from the city  center.87
During days of stagnation when the highest levels
of oxidant/ozone occur, this peak-concentration
distance was estimated by EPA to be between 1 5
and 25 km (9 and 1 6 miles),53 and more recently, to
be 15 to 30 km (9 to 19 miles).
  This urban-scale transport of oxidant has two
important   implications.  First,  it  suggests  an
obvious guideline to be used in siting stations for
oxidant/ozone  monitoring.  The  second  im-
plication, suggested by experimental data on the
oxidant-to-precursor dependencies,42 is that
urban-scale transport may obscure the real impact
of emission control on oxidant air  quality. This
latter implication may be clarified as follows: small
or even moderate decreases in emission rates for
HC and NOX are expected to affect mainly the time
at which the oxidant concentration will reach the
day's peak levels; the effects on the peak level itself
are of smaller magnitude. This time lag effect, in
turn, translates into an increase in the distance
from  the   source  area   where  peak oxidant
concentrations develop (i.e., the location  of the
peak concentration  relative to the location of the
core of the source area). Thus, for example, the
actual effect of a modest HC emission reduction
will  be to  shift the  peak  oxidant concentration
farther  downwind,  rather than  to  effect  a
proportional  reduction  locally.  Therefore,  the
impact of emission control will appear to be either
beneficial   or  detrimental,  depending  on  the
location of the oxidant monitoring station.
  Evidence  on  occurrence of  mesoscale  and
synoptic-scale transport  has  been obtained in
numerous extensive and conclusive studies. These
studies have  been  reported  individually  10~
1 2,25,26,28,30,38,39,44,51,61,66,68,85,86,11 5,11 6,129,130,1 36,138,1 39,-
145 and collectively in reviews.85'9712° The evidence
shows that many cities produce urban oxidant
plumesthat cause elevated oxidant concentrations
in downwind areas as far as 300 km (190 miles) or
more  from the  source  City.^5,6i,74,i28,i36,i43
Transport  associated with  synoptic-scale high
pressure   systems,  unlike   such  mesoscale
transport,  is  not characterized by  well defined
urban oxidant plumes and can extend to distances
several times  greater than 300  km (190 miles).
38,39,97,115,116,120,136,145 |p ^^ ^^ ^ sigmf)cam
implication  is  that urban  emissions  may  be
creating oxidant problems not only within and near
their  source areas, but also in other  downwind
urban areas, as well as oxidant problems in rural
and  even  in remote areas. It should be stressed
that the  evidence on  horizontal  transport of
oxidant/ozone suggests but does not  prove that
most of the oxidant observed in rural and remote
 44

-------
areas  is  transported from   the  cities.  Also,
horizontal transport of oxidant over long distances
can only occur in air layers aloft; ozone at ground
level  is rapidly  destroyed on  surfaces and in
reactions  with  NO and  NC.  *>.w».™."w» For
oxidant/ozone to reach the surface, there must be
vertical air movement, a movement that may also
bring to the surface ozone from the stratosphere or
from other natural sources. In general, because of
the multiplicity of the sources that  potentially
contribute to surface oxidant/ozone buildup, the
rural oxidant/ozone problem cannot automatically
be  attributed  entirely to  urban  oxidant/ozone
transport.  The indications are that direct  urban
transport  is  a   relatively  important  source  of
oxidants in  many locales but not  necessarily
everywhere.
  Because  of   their  elevated  ambient  ox-
idant/ozone levels, the State of California, the Gulf
Coast  and Texas,  the upper Midwest,  and the
Northeast are the areas of the United States that
have  been  studied most. The explanation  for
oxidant/ozone levels  in  California  is relatively
simple: Ozone levels in that area can be explained
by  local  photochemistry and oxidant/ozone
transport;  nonanthropogenic explanations are not
required. The most striking and clearly explained
oxidant transport case reported for that area is the
one involving overnight transport over water of Los
Angeles air containing as much as 590 /jg/m3 (0.3
ppm) oxidant to San Diego.12'120
  Unlike California, the  situation in  the Gulf
Coast/ Texas area is considerably more complex
and  requires more careful  examination.  Early
studies in Texas led investigators to conclude that
widespread  ozone concentrations  were  not of
anthropogenic origin. The basis of  that early
conclusion was the observation that ozone was
frequently high when there were onshore winds at
coastal stations.120 However,  recent  studies  by
Price'02 and by Decker et al.38 at the Research
Triangle Institute  (RTI)  concluded that natural
tropospheric ozone was not  a factor  in the high
ozone events in Texas. The  RTI study, in particular,
involved extensive aerial and ground   meas-
urements  of ozone;  HC,  and NOX,  and  wind
trajectory  analysis.  Results  from  that  study
showed the  high ozone concentrations over the
Gulf to be associated with recent passage of  the air
stream  over land areas and  often over source-
intensive areas. This finding, similar to that by Bell
for the  Los Angeles/San  Diego case,12 explains
more rationally the observations made in the early
studies of Texas and implies that the Texas  ozone
problem is at least partly of anthropogenic origin.
The RTI study also indicated an association of high
ozone  concentration with  the  passage of high
pressure systems, an association which had been
recognized for years and which was more firmly
established in the  midwestern part of the United
States.
  Transport  of ozone for an unusually long  dis-
tance was reported in an episode of widespread
haze over southern  Florida from May 21  to 27,
1972.'23 The initial report concerned metropolitan
Dade County and was supplemented by concur-
rent data obtained at the site of a proposed jetport
65 km (41 miles) west of Miami. At this rural site,
the ozone concentration  equaled or exceeded 160
fjg/m3 (0.08 ppm) on 5 of those days, and the air
quality standard was exceeded for 14 consecutive
hr on both May  22 and 23. The analysis of the
transporting trajectory that brought the haze and
accompanying ozone was  verified  by the EPA
Meteorological  Laboratory.  The  source  region
appeared to  be  the industrial  area among the
states south of the Great Lakes, making a transport
distance of over  1000 miles (1600 km), of which
400 miles (640 km) were over the Gulf of Mexico.
Both the haze and the ozone of its precursors may
have been augmented along the route.46
  Most of the oxidant transport studies  in the
upper Midwest dealt with the role of high pressure
meteorology  in wide area oxidant problems. The
connection   between  stagnating  anticyclonic
conditions and high oxidant concentrations over
large areas was studied by several investigators
and  is  now believed   to be relatively well
understood.3ei39-115'116'1zo'1M'130-136'138-139'l45This
belief is based on (1) data from  direct monitoring
with an aircraft of the ozone concentrations in the
path of a moving  high pressure system, and (2)
presence  of  statistically  supported correlation
between high  pressure systems  and  elevated
ozone  concentrations  during  days with  such-
conditions. This  correlation has been verified by
numerous  studies. The  reverse has  not been
established; that is, elevated ozone levels do not
occur  only  during stagnating  anticyclonic  epi-
sodes.
  Within  the moving  and  clockwise  rotating
anticyclone, the highest  ozone concentrations are
observed in the trailing or western portion of the
cell  where  the  air parcels have  the  longest
residence time  in  the  anticyclonic  regime. RTI
investigators  made  some   detailed   model
calculations to show that air parcels initially in the
northeast quadrant  of the cell had a potential
                                           45

-------
residence time as long  as  6 days, and that air
parcels  on the western side of the circulation
pattern had been in the cell the longest time. This
latter conclusion  is in general agreement  with
ozone measurements in source areas such as have
been  described by  Westberg  et al.138  Under
conditions  of  the high intensity  sunlight and
stagnation or undiluted flow that typically prevail
within the anticyclone, oxidant/ozone formation is
expected to be accelerated and depends  on the
emission density within the area covered by the
high  pressure  system.  This latter dependence
seems to be manifested by the data obtained in the
paths of high pressure systems moving from less to
more densely populated  areas.38
  A  significant percentage  of  the  high  oxi-
dant/ozone episodes observed  in  the  United
States is associated with passage of high pressure
systems.  Statistics  of  high  pressure episode
occurrence have been obtained by Korshover76 for
the eastern part of the United States for all months
of the year. The study could not be extended in the
western  United  States  because  the extreme
terrain  irregularities  result in  unrepresented
pressure  distributions.  Such statistics for the
months'of August and September are shown by
the plots in Figures 4-8 and 4-9.
  Oxidant  transport  in  the  Northeast received
considerable  attention   because  high  oxi-
dant/ozone concentrations have been measured
almost  routinely  in rural  and distant  suburban
areas for  years.  Careful  analysis by  Bell
Laboratories and other investigators of aerometric
data obtained in  northern  New  Jersey, eastern
New  York,  Connecticut,  and   Massachusetts
provided strong evidence of extensive mesoscale
transport, especially from the New York City/New
Jersey area to the Northeast into Connecticut and
Massachusetts.120'146  Such transport was char-
acterized (1) by low  oxidant/ozone levels within
urban core areas  because of  scavenging  by
precursors,  (2)  by appearance   of  highest
oxidant/ozone concentrations downwind from the
main urban center at a distance equivalent to 1 to 2
hr of midday wind travel, and (3) by shifting of peak
oxidant/ozone  concentrations with  time  and
distance, refleri.ng the regional movement of the
main urban area plume along the wind trajectory.
The  results from  Bell Laboratories seem to be in
agreement  with  subsequent studies  by other
groups. Interestingly, the evidence descriptive of
the  Northeast  situation suggests that weather
patterns seem to play a minor role in  establishing
oxidant/ozone levels in that part of the United
States.  For  example,  oxidant/ozone  con-
centrations  seem  to be  less  sensitive  to the
location or intensity of an anticyclonic system.120
This could be a result of the high population density
in that area and the relative dominance of urban
plumes over the nonanthropogenic  background
levels in the air mass. Finally, local circulations in
the surface layers were found to play a dominant
role in the coastal areas in the Northeast.120'139
Such circulation systems, subjecting areas to next-
day effects from their own  emissions, were also
observed in the Gulf Coast38 and Midwest86 areas
but were not as dominant.
  The  evidence on transport has had  another
important impact on  the  understanding  of the
oxidant/ozone problem.  It provided one plausible
explanation   to  some  phenomena   previously
thought to be paradoxical (e.g., the occurrence of
elevated  oxidant/ozone  concentrations  in rural
areas and at  night115 and  the  Sunday-weekday
effect13'24'45'80'135). Thus downward intrusion, from
layers   aloft,  of  transported  anthropogenic
oxidant/ozone explains  convincingly the  occur-
rence of elevated oxidant/ozone  concentrations in
rural areas and during the night in the summer-fall
months. The Sunday-weekday effect could also be
explained in part as resulting from anthropogenic
oxidant/ozone present in  layers aloft. Although
these explanations have not been supported by
documentation in all  cases,  they at least have
served to clarify these phenomena and to increase
confidence  in  the   current  knowledge and
understanding of photochemical processes.

  Identification  and understanding   of  ox-
idant/ozone  transport represents a  significant
advance   in  understanding  photochemical pol-
lution.   However,  it  has   also raised   several
questions that need to be addressed. For instance,
the impact of pollutant transport on  the  oxidant
concentration observed  in  a locality cannot  be
quantified at this time.43 The difficulty here is not
merely the absence of sufficient  data (e.g., data on
pollutant  composition in  transported, aged  air
masses), but also the absence of valid investigative
methods.  Clearly the  need  is for both laboratory
and field studies  that  would  provide a better
understanding of the  quantitative aspects of the
interaction of transported, aged pollutant mixtures
with  locally  emitted  fresh  pollutants.   More
specifically,  a method should be developed  for
estimating the proportion of local oxidant/ozone
concentrations resulting from   pollutant trans-
port.43  Evidence must also  be  obtained  on the
 46

-------
oxidant-to-precursor  dependencies for trans-
ported oxidant/ozone.43

NATURAL SOURCES OF OXIDANTS

Introduction
  The occurrence of ozone in remote areas and
during the  winter  months constitutes  strong
evidence that natural sources of ozone must exist,
providing an  atmospheric  background on which
the  anthropogenically  generated  ozone  is
superimposed. Promulgation of the  160-A