&EPA
           United States
           Environmental Protection
           Agency
           Environmental Research
           Laboratory
           CorvallisOR 97333
EPA-600/3-34-037
February 1984
           Research and Development
A Review and
Assessment of the
Effects of Pollutant
Mixtures on
Vegetation—
Research
Recommendations


-------
                                                    EPA-600/3-84-037
                                                    February  1984
A REVIEW AND ASSESSMENT OF THE EFFECTS OF POLLUTANT MIXTURES ON VEGETATION

                        -RESEARCH RECOMMENDATIONS-
                        Vegetation Effects  Workshop
                             April 21-22,  1983
                          Raleigh, North Carolina
                                Prepared  by

                             Allen S.  Lefohn
                            Douglas P. Ormrod
                             January 30, 1984


                             Project Officers
                              Eric M.  Preston
                              David T.  Tingey
                       Air Pollution Effects Branch
                Corvallis Environmental  Research Laboratory
               United States  Environmental  Protection Agency
                         Corvallis, Oregon,  97333
                CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
                    OFFICE OF RESEARCH AND DEVELOPMENT
               UNITED  STATES  ENVIRONMENTAL PROTECTION AGENCY
                         CORVALLIS, OREGON, 97333

-------
                      NOTICE

This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication.  Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.

-------
                                   CONTENTS


Acknowledgement                                                            vi

Preface                                                                    vi i

List of Participants                                                       ix

Executive Summary                                                          S-l
     Introduction                                                          S-l
     Characterizing Ambient Air Quality Exposures                          S-l
     Vegetation Effects                                                    S-3
         Exposure Regimes                                                  S-4
         Development of Minimum Guidelines for Research Protocols          S-4
         Predictive Investigations                                         S-4
     Conclusion                                                            S-5

1.   Role of Pollutant Mixture Studies in Establishing National
     Ambient Air Quality Standards                                          1
         Introduction                                                       1
              The Standard Review and Development Process                   2
         Overview of Recent Vegetation Standard Reviews                     5
              Use of Mixture Studies in Ongoing Standard Reviews            8
         Identification and Treatment of Uncertainties                      8
              Genetics and Environment                                      9
              Exposure Situations                                           9
              Experimental Exposure Regimes                                11
              Mechanisms of Action                                         13
              Biological Endpoint                                          13
         Summary                                                           14

2.   The Co-Occurrenc of S0?/N0?, CK/SO?, and CU/NO? Mixtures  in
     Ambient Air                                                           16
         Introduction                                                      16
         The Pollutants                                                    17
         The Data Bases                                                    18
         Results                                                           20
              S0?/N0?                                                      21
              037S02                                                       29

         Discussion                                                        38
         Conclusion                                                        45

3.   Effects of Pollutant Mixtures on Vegetation                           4o
         Introduction                                                      46
         Experimental Methods  and Interpretation  of Data                   47
         Characteristics of Plant Response                                 56
              Foliar Symptoms                                              56
                    S0? + 0-,                                                58
                    SOp + NtL                                               61
              Growth and Yi&ld                                             61
                    S00 + 00                                                63

-------
                   S0? + N0?                                              63
                   NO  + O/                                              6634
                       + N0  + 0,                                         65
                      er Mixtures                                         66
              Physiological and Metabolic Responses                       66
              Accumulation and Uptake                                     73
         Modifiers of Plant Response                                      75
              Genetic Factors                                             75
              Phenological Factors                                        78
              Environmental Factors                                       79
         Summary                                                          81

4.   Research Needs                                                       83
         Ambient Air Quality Exposure                                     83
         Vegetation Effects                                               83

5.   Recommendations                                                      86
         Introduction                                                     86
         Air Quality                                                      86
         Biological Effects                                               88
              Introduction                                                88
              Realistic Exposure Regimes                                  88
              Development of Minimum Recommendations for
              Research Protocols                                          89
              Predictive Capabilities                                     90
         Conclusion                                                       91

References                                                                93
                                        I V

-------
                                    TABLES
Table 1-1  Plant Exposure to Criteria Pollutants                            7
Table 2-1  Summary of Site Years Analyzed  (EPA, EPRI SURE,
           and TVA Data)                                                   44
Table 3-1  Visible Foliar Injury on Various Plant Species  in Response
           to the Joint Action of SO- + (k                                 59
Table 3-2  Growth and Yield of Various Plant Species in Response  to
           the Joint Action of SO- + 0,                                    60
Table 3-3  Direct Comparisons of Species Sensitivity to SCL +  (U
           SO- + N02, and 03 + N02, and HF + S0?                           76
Table 3-4  Direct Comparisons of Cultivar  Sensitivity Within Species
           to S02 + 03                                                     77

-------
                                   FIGURES
Figure 1-1
Figure 1-2
Figure 1-3



Figure 2-1

Figure 2-2

Figure 2-3

Figure 2-4

Figure 2-5

Figure 2-6

Figure 2-7

Figure 2-8

Figure 2-9

Figure 2-10

Figure 2-11

Figure 2-12

Figure 2-13

Figure 2-14

Figure 2-15

Figure 2-16

Figure 2-17

Figure 2-18

Figure 2-19

Figure 2-20
The StandarcLReview and Development Process
Seasonal  S04   wet deposition
(mg/m ) for North America, "Summer" April
through October 1979 and "Winter" November 1979
through March 1980
Temporal  changes in ground level concentrations of
SOo, N02, and 03 during a fumigation
event near a coal-fired power plant (modified
with permission from Noggle and Jones 1981)
SOp/NOo Co-Occurrence, Frequency Site
 Distribution
Indian River, Delaware, S02/NOX
Co-Occurrence
                    S02/N02
                              S02/N02

                              S02/N02
Paradise, Kentucky
Co-Occurrence
Allen Steam Plant, Tennessee
Concentration Over Time
Allen Steam Plant, Tennessee
Co-Occurrence
Fontana, California S02/N02
Co-Occurrence
Kansas City, Kansas S02/N02
Co-Occurrence
O^/SOo Co-Occurrence, Frequency Site
 Distribution
Fontana, California 03/SO?
Co-Occurrence
Madison County, Illinois 03/SOp
Co-Occurrence
Scranton, Pennsylvania 0.,/SOp
Co-Occurrence
Rockport, Indiana 0.,/SOo
Co-Occurrence
Paradise, Kentucky 03/S02 Concentrations
Over Time
Paradise, Kentucky 03/S02
Co-Occurrence
0-,/NOp Co-Occurrence, Frequency Site
Distribution
Rubidoux, California O^/NOo
Co-Occurrence
Indian River, Delaware 03/NO
Co-Occurrence
Paradise, Kentucky 03/N02
Co-Occurrence
Allen Steam Plant, Tennessee 03/N02
Co-Occurrence
Allen Steam Plant, Tennessee 03/N02
Concentrations Over Time
10



12

22

24

24

25

27

27

28

30

31

31

33

33

34

36

37

39

39

40

40

41
                                        v i

-------
Figure 3-1   Examples of Contour Plots Illustrating the Response of
             Lettuce, Radish, and Pea to 6-hour Exposures to
             Combinations of Cu and SCu                                   51
Figure 3-2   Graphical Representation of the Four Response Regions of
             Practical Interest When Two Pollutants are Combined          54

-------
                               ACKNOWLEDGEMENT


      Bruce Jordan and Larry Zaragoza would like to provide acknowledgement to

John Bachmann, Kent Berry, John Haines, Pam Johnson,  David McKee, and Harvey

Richmond for their comments during the preparation of the material that
appears in Chapter 1.


      Allen S. Lefohn would like to acknowledge the assistance of the
following individuals and organizations for providing information that was
used in the data analysis described in Chapter 2:


      1.  Ms. Cel Allard, Center for Data Systems  and Analysis,
          Montana State University, Bozeman, Montana for
          developing the software programs necessary to access
          A.S.L. & ASSOCIATES' copy of the SAROAD  data base.

      2.  Dr. Tom Curran and Mr. Jim Reagan of the EPA, Research
          Triangle Park, North Carolina, for providing insight
          into the EPA SAROAD and EPRI SURE data bases.

      3.  The Tennessee Valley Authority for providing its air
          quality monitoring data on a timely basis,  in a
          readable format.

      EPA wishes to acknowledge A.S.L. & Associates for assisting in the

integration, and editing of this report.
                                       VI I I

-------
                                   PREFACE


      On April 21-22, 1983, a workshop sponsored by the U.S. Environmental
Protection Agency Corvallis Environmental Research Laboratory, was hosted in
Raleigh, North Carolina to develop research recommendations concerning the
effects of pollutant mixtures on vegetation.  Prior to the meeting, EPA asked
several individuals to develop position papers to a) describe the use of
information on plant response to pollutant mixtures in setting ambient air
quality standards; b) characterize the spatial and temporal characteristics of
air pollutant mixtures in the ambient air; and c) summarize the vegetation
effects literature associated with pollutant mixtures.  The material was
integrated into the first three chapters of this report.  The following
individuals are acknowledged for the writing of the position papers:

              Studies  of  Combined  Exposure  Effects  on  Vegetation:
          Role in  Establishing National  Ambient  Air  Quality  Standards

                     Bruce  Jordan  and  Lawrence  J.  Zaragoza
               EPA Office of Air Quality Planning and Standards
                          Research Triangle Park, NC

             The  Co-Occurrence of  Sulfur  Dioxide/Nitrogen Dioxide,
               Ozone/Sulfur  Dioxide,  and  Ozone/Nitrogen Dioxide
                            Mixtures in Ambient Air

                                Allen  S.  Lefohn
                              A.S.L. &  Associates
                                  Helena, MT

                  Effects of  Pollutant  Mixtures  on  Vegetation

                              Delbert  C.  McCune
                           Boyce Thompson Institute
                                Ithaca, New York

                              Douglas  P.  Ormrod
                             University of Guelph
                           Guelph, Ontario, Canada

                              Richard A. Reinert
                       North  Carolina  State University
                                  Raleigh,  NC

      At the  invitation of EPA,  eighteen individuals participated in the
two-day workshop.  Participants were asked to critically review the position
papers and 1) summarize the information gaps and assess the significance of
the problems  associated with those pollutant mixtures exposures that affect
vegetation; and 2) identify and recommend activities that would assist the
Agency in filling these research gaps.

      From these activities (position papers and panel deliberations), EPA's
Corvallis Laboratory has  produced this document to summarize

-------
      o   the processes involved in developing ambient air quality
          standards;

      o   the spatial  and temporal  distribution of gaseous
          pollutant mixture concentrations;

      o   the effects  of gaseous pollutant mixtures on vegetation;

      o   information  gaps; and

      o   recommendations on research that is  required to  fill  the information
          gaps.

      While the  subject was addressed,  no attempt  was  made to prioritize  the
general research categories because panel members  believed pollutant ambient
monitoring characterization and vegetation effects research efforts  were
complementary.

-------
                      LIST OF  PARTICIPANTS

Air Pollutant Mixtures Vegetation Effects Workshop Participants

            Thomas  C.  Curran
            U.S.  Environmental  Protection Agency
            Monitoring and Data Analysis  Division
            Research  Triangle  Park,  North Carolina

            J.H.B.  Garner
            U.S.  Environmental  Protection Agency
            Environmental  Criteria  and  Assessment  Office
            Research  Triangle  Park,  North Carolina

            Walter  W.  Heck
            North Carolina State University
            P.O.  Box  518
            Raleigh,  North Carolina

            Pam M.  Johnson
            U.S.  Environmental  Protection Agency
            Strategies and Air Standards  Division
            Research  Triangle  Park,  North Carolina

            Bruce C.  Jordan
            U.S.  Environmental  Protection Agency
            Strategies and Air Standards  Division
            Research  Triangle  Park,  North Carolina

            Allen S.  Lefohn
            A.S.L.  &  Associates
            Helena, Montana

            Delbert McCune
            Boyce Thompson Institute
            Cornell University
            Ithaca, New York

            David Olszyk
            Statewide Air  Pollution  Research Center
            University of  California
            Riverside, California

            Douglas P. Ormrod
            University of  Guelph
            Department of  Horticultural  Science
            Guelph, Ontario CANADA

            Ronald  Oshima
            California Department of Food and Agriculture
            Environmental  Monitoring
            Sacramento, California
                                 X I

-------
Eric M. Preston
U.S. Environmental Protection Agency
Environmental Research Laboratory-Corvallis
Corvallis, Oregon

Richard A. Reinert
U.S. Department of Agriculture
Plant Pathology Department
North Carolina State University
Raleigh, North Carolina

George E.  Taylor, Jr.
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee

Ted W. Tibbitts
Horticulture Department
University of Wisconsin-Madison
Madison, Wisconsin

David T. Tingey
U.S. Environmental Protection Agency
Environmental Research Laboratory-Corval1 is
Corvallis, Oregon

Michael Treshow
Department of Biological  Sciences
University of Utah
Salt Lake City, Utah

David E. Weber
U.S. Environmental Protection Agency
Office of Research and Development
Washington,  D.C.

Larry Zaragoza
U.S. Environmental Protection Agency
Strategies and Air Standards Division
Research Triangle Park, North Carolina
                    X I I

-------
                               EXECUTIVE SUMMARY

INTRODUCTION

      The Environmental Protection Agency  (EPA)  is responsible for
periodically reviewing and revising all  national  ambient air quality standards
(NAAQS).  The Clean Air Act (CAA) requires EPA to establish national ambient
air quality standards for ambient air pollutants  which may endanger human
health and welfare.  Secondary ambient air quality standards must be adequate
to protect the public welfare from any known or  anticipated adverse effects
associated with the presence of a criteria air pollutant.   To help the Agency
develop data that assess the effects of pollutant mixtures on vegetation, the
EPA Corvallis Environmental Research Laboratory  sponsored  a workshop in
Raleigh, North Carolina on April 21-22,  1983.   Participants reviewed position
papers to 1) summarize the information gaps and  assess the significance of the
problems associated with pollutant mixture exposures that  affect vegetation
and 2) identify and recommend activities that  would help the Agency fill  these
research needs. This report includes both  the  position papers and the workshop
deliberati ons.

      The ranking of research needs was addressed by the workshop
participants.  Members believed that no research  areas are independent from
another and should be treated collectively.  For  example,  the biological
research efforts are dependent upon a knowledge  of the pollutant
concentrations occurring in the field under ambient conditions.   Thus,
environmental, genetic, and phenological variables should  be considered when a
study is initiated.

CHARACTERIZING AMBIENT AIR QUALITY EXPOSURES

      Different air quality exposure regimes exist for sulfur dioxide (S0~),
ozone (Oq), and nitrogen dioxide (NO-).   These regimes affect the frequency of
co-occurrence and sequential  exposures that vegetation may experience.  An
analysis of the 1981 EPA air quality data  base (SAROAD), Electric Power
                                     S-l

-------
Research Institute (EPRI) 1978 SURE,  and Tennessee Valley Authority (TVA)  data
bases indicates that SO-, NCL, and 03 may co-occur in various concentrations
in rural, suburban, and urban areas.   Analyses of the data bases show that the
frequency of co-occurrence (using 0.05 ppm as the definition of an  event)  is
small for many rural sites.  For most  of the cases analyzed, events  lasted  only
a few hours and were separated by intervals of weeks or months.

      The panel recommends that air quality data be evaluated further to study
patterns of occurrence of the combined pollutants to establish guidelines  for
designing plant interaction  research  investigations.  The three pollutants of
primary interest are SO-, 0.,, and NO^.  The effects of these pollutants on
vegetation can be evaluated  using the available air quality data and  the
research information dealing with individual  effects.  In addition, the panel
recommends that acidic deposition be  considered as a pollutant with the
potential for interaction with the above three air pollutants.

      The recommendation is  to analyze existing air quality data bases
(starting with SAROAD) to derive the  joint probability distributions  of
pollutants and,the diurnal patterns of exposure for plant exposure
experiments.  Additional sources of rural air quality data could include the
USDA Forest Service, EPRI (SURE), EPA (NCLAN), and permits monitoring programs
(e.g., PSD applications).  This analysis is to include the following:

      1)  A search of the data base for locations where either
          co-occurrence or sequential exposures (starting with a
          24-hour time step) occur.  This search would include
          separate listings  at several threshold concentrations
          (e.g., 0.05, 0.04, 0.03, and 0.02 ppm).
      2)  Once locations are identified, the  monitoring data bases
          at the- locations should be  presented as joint frequency
          distributions and  as diurnal time series.  It is
          suggested that the utility  of spectral analysis
          (Fourier series) and the Box Jenkins model should be
          explored.
      3)  The results of this process should  be disseminated to
          research groups to guide experimental exposures used in
          interaction experiments.
                                     S-2

-------
      The panel recommends that potential  data displays for individual
pollutants include: 1) three-dimensional plots of concentration,  duration,  and
frequency; 2) diurnal plots for individual pollutants in terms of mean
concentraiton and frequency greater than  particular concentrations.   These
analyses should be summarized for the growing season (or some relevant  time
period) and should also serve to identify potential  anomalies.

      The air quality analyses would provide information that could  be  used to
identify general patterns of exposure that relate to geographic regions or
source configurations.  It may be necessary to supplement the air quality data
around point sources by considering the use of dispersion models  to  provide
information on levels, diurnal patterns, and time between episodic events.

VEGETATION EFFECTS

      Pollutant mixtures may induce plants to exhibit different types of
responses which are influenced by several  variables  that are often difficult
to predict.  The three general categories  of responses that may follow  plant
exposures to mixtures of ambient air pollutants are  additive, greater than
additive, and less than additive. All three responses are found to some degree
in the experimental results using SCL, CL, and N0?.   Taken as a whole,  the
current information on both long- and short-term combined exposure studies
provides conflicting results that are difficult to interpret.

      The workshop members agree that research efforts should be  directed
toward important plants (including agronomic, horticultural, and  native
plants, and tree species).   The panel believes that the gaps in  knowledge  can
only be filled by an integrated effort involving growth chambers, greenhouses,
and field plots.
                                     S-3

-------
Exposure Regimes

      The panel is aware that the understanding of the response of plants to
various exposure regimes is crucial.  Plant effects must be associated with air
pollutant peaks, means,  length of exposure,  and time between exposures that
mimic realistic ambient pollutant exposures.   Research should evaluate the
vegetation effects associated with sequential exposures of pollutant mixtures
which duplicate ambient conditions.

Development of Minimum Guidelines for Research Protocols

      The panel recommends that a minimum set of standardized procedures be
developed to ensure the quality assurance of  plant response studies.
Generalized guidelines should be proposed for 1) plant growth conditions, 2)
environmental and plant monitoring,  3)  pollutant exposures,  and 4) uniform
terminology (describing plant response characteristics).  The most efficient
experimental designs and analysis procedures  available (relevant to a specific
experimental goal) should be implemented (e.g., covariate analysis, analysis
of variance, and rotatable design).  It is proposed that the  minimum guidelines
be developed through two or three workshops.

Predictive Investigations

      One purpose of pollutant interaction research is to develop predictive
capabilities for assessing vegetation effects.  Predictive models can provide
estimates of vegetation effects under a variety of conditions not feasible
with direct experimentation.   In order to properly generate  the information
necessary to develop such predictive  models,  data are needed from research
programs involving studies that define 1) the modes of action and 2) the
sources of biological  variation.

      The following research activities are  recommended and  considered
instrumental  in the development of this predictive capability:
                                     S-4

-------
      1)  Modes of Action: The objective of this research activity
      is to understand how air contaminants influence biological
      processes and in doing so, determine whether their actions
      may cause significant ecological alterations.  It is
      necessary that studies address modes of action for
      pollutants singly and in combination.  The research effort
      should include both sequential and co-occurrence exposures
      and should be conducted with  an appreciation of realistic
      exposure regimes. The biological level of organization to be
      investigated should focus on  processes at all levels of
      plant organization  (i.e., the cell, whole-plant, population,
      and ecosystem).  The panel believes that there should be two
      major areas of interest

          a)  The relationships between different mechanisms of
               pollutant  response.

          b)  The varying biological responses attributed to
              different levels and  combinations of air pollutant
              exposure.

      2)  Sources of Variation: The plant response to a given
      exposure regime varies significantly with specific
      environments, environmental changes, and the stage of plant
      development.  The panel recommends research that focuses on
      each of the following:

          a)  environmental factors-the significance of edaphic
              (e.g., soil water availability, soil nutrients),
              climatic (e.g., temperature, light, relative
              humidity, elevated carbon dioxide, etc.) and biotic
              factors (e.g., pathogens, symbionts, competition,
              etc.).

          b)  genotype factors-the  significance of intra- (e.g.,
              cultivar, population) and interspecific genotypes.
              This includes phenology as a source of variation.

      3)  Modeling-The development  of data that describe the
      process and mechanistic activities associated with air
      pollutant mixture vegetation  effects should allow for the
      development of conceptual and quantitative models that
      describe observed biological  response.

CONCLUSION


      It is the opinion of the workshop participants that the position paper

focusing on ambient exposures was an initial attempt to identify realistic
                                      S-5

-------
exposure regimes that exist in the ambient air.  The panel believes that
additional  efforts should be made to supplement the existing analysis and that
they should simultaneously proceed as the biological vegetation effects
research is implemented.   It was recommended that the results of the air
quality characterization  should feed directly into the design of the pollutant
mixture experimental  protocols.

      The process and mechanistic research activities should involve two
stages:  1) a biological  effects screening exercise to prioritize which air
pollutant mixture exposures are most likely to be significant, 2) a more
detailed investigation that is performed under field and laboratory situations
for the purpose of quantifying the significance of the major factors affecting
plant response.

      It was the conclusion of the panel members that the conceptual models
should combine existing models of joint action with the data that describe the
modes of biological  action.  The quantitative models should be capable of
providing accurate and precise estimates of plant response.  In addition, the
models should be able to  complement the physical and/or biological  processes
that are responsible  for  producing the observation.
                                     S-6

-------
            1.  ROLE OF POLLUTANT MIXTURE STUDIES IN ESTABLISHING
                    NATIONAL AMBIENT AIR QUALITY STANDARDS

INTRODUCTION

      The Clean Air Act (P.L.  95-95) requires that the  Environmental
Protection Agency establish national ambient air quality standards  (NAAQS)  for
certain air pollutants that, if present in the ambient  air,  may endanger human
health and welfare.  Primary NAAQS are established to protect  human  health
while secondary NAAQS are established to protect public welfare.   Section
302(h) of the Act specifies that the effects of air  pollution  on  crops  and
other vegetation are among the effects that  must be  considered in  establishing
secondary NAAQS.

      The original NAAQS were established in 1971 and included secondary
standards set at levels designed to protect  public welfare including  effects
on crops and other vegetation. No detailed discussion of the rationale  for  the
secondary standards was provided in the original proposal  and  promulgation
notices (36 FR 1502, 36 FR 5867, 36 FR 8186). The secondary  standard  for ozone
(Oo), which the original criteria document associated with damage  to
vegetation, was set equal to the primary standard.  Based  on comparison of  the
original sulfur dioxide ($02)  standards and  supporting  documents,  it  appears
that support for the SO- standard levels was based upon a  limited  number of
studies and that the issue of the importance of pollutant  mixtures was  not
raised (DHEW 1970). Available information does not suggest that the  effects of
nitrogen dioxide (NO,,) on vegetation played  a role in setting  the  original  N0?
standard.

      The potential for increasing plant sensitivity by the  presence  of
multiple pollutants is significant because emission  sources  often  release
different pollutants within an area.  However, there is often  considerable
controversy over the characterization of plant response in the presence of
even a single pollutant.  The assessment of  plant response to  pollutant
mixtures is a more difficult task.

-------
      This chapter describes how information on pollutant mixtures  is  being
considered and provides suggestions that should increase the utility of
studies on plant response to pollutant mixtures in the standard review process
for NAAQS.  Focus is placed on evaluations developed during ongoing standard
reviews, with discussion of past review practices included.

The Standard Review and Development Process

      The complex standard review and development process is designed  to
solicit the best available scientific information and public comment.   The  use
of welfare information and those factors that have appeared to be most
influential in the ultimate decision-making process are described by Bachmann
and Zaragoza (1983).  More general  discussions of the process are provided
elsewhere (O'Connor 1980, Padgett and Richmond 1983, Zaragoza 1982).  In order
to better understand the role of scientific information in this process, this
section highlights those activities that most pertinent to the present subject
area.

      The standard review and development process combines scientific  review
and assessment with the judgment of EPA's Administrator.  As Figure 1-1 shows,
the first part of this process involves an in-depth scientific review,
including the collection of relevant information and review by specialists
within each scientific area. When available scientific information  can resolve
questions so that uncertainty is small, the degree of judgment required of  the
Administrator for a given degree of protection is reduced.

      The criteria document development process is the Agency's means  of
conducting an unbiased and public review of all literature used in  support  of
a particular standard.  Once a study is included in the criteria document,  the
characterization of that study will play a major role in determining its
potential utility in the standard-setting process.  Those investigators who
employ reliable methodologies and who design experiments so that their results

-------
SCIENTIFIC
RESEARCH
'
'

1
/DRAF
PAF
l
REVIEW AND
REVISE/REAFFIRM
WITHIN 5 YF.ARS
4

CHIItHIA /" ' "/
* DOCUMENT J W°RK'NG I
INITIATION1 7 °RAFTl /

YES .XX">X\ / EXTERNAL /
\Tx^ / DRAFT^ /
TNO A
COMMENT I
ANALYSIS " 	 	
AND REWRITE1
f
T STAFF / CASAC REVIEW
gp2,3 /' ' " ^* AND HUBLIC 	 .-.^^
, .7 MFFJING3



fl

I • 	 IMPI ruiruTATinu


WORKSHOP1 F|RST EXTERNAL
- • CONSULTANTS ^ Dci/lcuu
• SCIEN
"EXPE
TIFIC " ORAFT1,2
RTC"

\ '
COMMENT CASAC REVIEW
AND REWRITE1 MEETING
n fKiiRF j*-
| NO
COMMENT
ANALYSIS
NO REWRITE3


YES PREPARATION
PACKAGE3

- 	 / STANDARD /^ 	 COMMENT . 	
/PROMULGATION3/^ ANALYSIS4



ADMINISTRATOR
* DECISION
' '
/ STANDARD /
7 PROPOSAL2-3 /

              Fiqure  1-1. The Standard Review and Development
                         Zaragoza  1982)
Process (reproduced with permission from
^Office of Research and Development assumes primary responsibility  for  these  activities.   The  Environ-
 mental Criteria and Assessment Office (ECAO)  plays a major role  in preparation  of  the  criteria  document.
^Public comment is requested at this stage.
-^Managed by the Office of Air, Noise, and Radiation.
^This phase of the process includes one or more public meetings,  receipt  and  formal  review of  public com-
 ments, the preparation of a revised regulatory decision package,  formal  Agency  review, and a  final  deci-
 sion by the Administrator.

-------
will be relevant to ambient situations usually produce studies that prove to
be of greatest relevance in standard-setting.

      As Figure 1-1, illustrates, the next step in the standard review process
is the development of a "staff paper."  The staff paper evaluates and
interprets available scientific and technical information most relevant to the
standard review, and presents recommendations on alternative approaches to
revising or maintaining standards.

      Both the criteria document and staff paper are reviewed by the Clean Air
Scientific Advisory Committee (CASAC).  The CASAC is an independent scientific
review committee established by the Clean Air Act to provide the Administrator
with advice on the scientific issues related to NAAQS.  The Clean Air Act
specifies that CASAC be composed of seven members, including at least one
member of the National  Academy of Science, one physician, and one person
representing State Air Pollution control agencies.  Formal review on each
document is complete when a "closure" memorandum, which indicates committee
endorsement of the document, is sent from the CASAC chairman to the
Administrator.

      The effects of pollutant mixtures are currently assessed as modifying
influences on plant sensitivity for a pollutant under review.  This is
consistent with section 108 of the Clean Air Act, which specifies that the
criteria for a pollutant shall include information on "the types of air
pollutants which, when p'resent in the atmosphere, may interact with such
pollutant to produce an adverse effect on public health or welfare."  In
ongoing reviews, the concern for pollutant mixtures arises when exposure to
pollutant mixtures appears to be greater than the effects that would be
expected from exposure to the individual pollutants.  If it is determined that
a combination exposure could produce effects that are greater than the effects
of exposure to an individual pollutant, then adjustments in the level of the
standards for the individual pollutant may be the appropriate response.  While
consideration could be given to establishing a combined standard, the details

-------
of specifying and complications associated with implementing such a standard
would have to be carefully evaluated.

      Secondary ambient air quality standards must be adequate to protect the
public welfare from any known or anticipated adverse effects associated with
the presence of a criteria air pollutant.  Given the mandate in the Clean Air
Act to protect public welfare from any known or anticipated adverse effect, it
is incumbent upon the Agency to consider the effects of pollutant mixtures on
vegetation. The task of deciding which or at what point welfare effects become
adverse is a difficult one.  Because there is usually no sharp demarcation
between a  level where effects of uncertain significance are reported and a
level where clearly adverse effects occur, the Act explicitly requires the
Administrator to exercise judgment in setting a standard.  Though relying
heavily on scientific advisors for technical evaluation of data and for those
judgments  that are scientific in nature, the Administrator alone is
responsible for considering risks and determining at which point effects
should be  regarded as adverse.

RECENT VEGETATION STANDARD REVIEWS

      Pollutant mixture studies become most important for standard review
exercises  when the presence of additional pollutants causes effects that are
greater than the effects of exposure to the pollutant alone.  Although Tingey
et al. (1971 a, 1973a) have reported such results, these studies have not
provided compelling support for either a combination standard or a more
stringent  standard based on pollutant mixture effects (EPA 1982a, 1982b).

      Plants exposed to pollutant mixtures may exhibit different types of
responses  that are influenced by several variables that are often difficult to
predict.    The general kinds of responses that may follow plant exposures to
mixtures of ambient pollutants are described in Chapter 3.  All these kinds of
responses  are found to some degree in the experiments using S0?, CU, and N0?.

-------
      Available information,  taken primarily from laboratory studies of foliar
injury, report the following  responses to pollutant mixtures (EPA 1982a,
1982b):

      a.  At lower concentrations (e.g., 0.10 ppm NOp and 0.05 ppm SO- for a
          few hours), little, if any, foliar injury is observed from either
          mixtures or single  pollutants alone.

      b.  At higher concentrations (e.g., 0.10  ppm 0., and 0.50 ppm of S0? for
          a few hours), foliar injury may be greater than the amount of foliar
          injury that could be predicted by adding the amount of foliar injury
          produced by either  pollutant alone.

      c.  At still higher concentrations, usually not observed in the ambient
          air, the amount of  foliar injury produced may be equal to or less
          than that predicted by adding the amount of foliar injury produced
          by each pollutant alone.

If these generalizations are  correct, greater than additive responses should
have a greater potential for  occurring in the ambient air than additive or
less than additive responses.

     The interpretation of the results of pollutant mixture studies is also
complicated by the exposure regimes used.  Although studies employing
unrealistic exposure 'regimes  may contribute to  our understanding of plant
response, they are very difficult to use in assessing impacts of air pollution
on vegetation.  For example,  it is reasonable to expect concurrent S0? and N0?
exposures for peak exposures  hear a coal-fired  boiler (Table 1-1). However, it
is unlikely that high 0^ concentrations would co-exist with peak S02 and N02
levels because nitric oxide (NO) would titrate  the 0^, increasing NOo and
reducing 0., concentrations.  Such information suggests that experimental
designs employing sequential  exposure to pollutants would be a more realistic
approach for simulating exposure regimes.

-------
                                  Table 1-1.    Plant  Exposure  to  Criteria Pollutants
Pollutant(s)
        Sources
Exposure Pattern
Character!sties
                                                                     Comments
03*
N02**
SO,
Not a primary emi ssion;
results from photo-
chemical  reactions
involving reactive
hydrocarbons, nitrogen
oxides and oxygen.
Results from combustion
processes from mobile  and
industrial  and domestic
(e.g. , oi 1  furnace)
sources.
Emitted primarily from
combustion or processing
of sulfur containing
fossil  fuels and ores.
- Peak 63 concentrations
at some sites can exceed
0.2 ppm over an hour.
- Average (yearly) 03
concentrations can range
between 0.025 and 0.07
ppm.

Concentrations tend to
reach short-term peaks
near sources (0.06 to
about 0.5 ppm for peak
hourly averages).
Long-term yearly averages
range between 0.01 and  0.08
ppm in urban areas and  are
about 0.001  in rural areas.

Modeling results indicate
that the current 24-hour
standard would not prevent
1-hour peaks in the range
of 0.5 to 0.75.  Seasonal
averages occurring over
large regions tend to  be
higher in the northeastern
U.S.  (-0.01  to 0.02 ppm)
than in other parts of  the
U.S.
  *Source:  EPA 1978.
 **Source:  EPA 1982a.
  "^Source:  EPA 1982b.
 ^Source:  Personal  communication from H.  Cole to L.  Zaragoza.
Peak 03 levels may occur in
the same region as elevated
S02 concentrations, but peak
levels of each pollutant
would be expected to occur at
different times.
If nitric oxi de i s
released in the
presence of 03, then
the 03 will be
titrated increasing
N02 and decreasing 03
concentrati ons.
Peak S02 level s near
sources (e.g., coal-fired
power plants) are strongly
affected by meteorology.
It is likely that short-
term peak concentrations of
S02 and N02 would have
considerable overlap  (some
displacement of the peak
N02 level  relative to the
peak S02 level is expected
due to possible conversion of
NO to N02).tt

-------
 Use of Mixture Studies in Ongoing Standard Reviews

      Taken as a whole, the cumulative information on both long- and
 short-term combined exposure studies provides conflicting results that are
 difficult to interpret. Biological responses to pollutant mixtures do not show
 consistent patterns.  The situation is further complicated by the use of
 exposures in both acute and chronic exposure studies that are not
 representative of exposure patterns, distributions, and levels of pollutants
 observed in ambient air.

      The criteria documents and staff papers recently have been completed for
 both SO  and NOV.  Based  on its review of the criteria documents, EPA staff
       A       A
 concluded that the available data on combination exposures indicate that plant
 responses to NOp and SOp, either together or in various combinations with Oo,
 are highly variable.  For example, one study reported that exposure of a
 commercial crop species to equal concentrations (0.20 ppm) of NOp and SOp
 caused less injury in five of six species tested (Tingey et al.  1971a). In the
 review of the SOp standard, studies examining effects of SOp near point
 sources included some NOp and higher 0^ before and after the SOp fumigation
 events.  Injury attributed to SOp under these conditions resembled typical SOp
 foliar injury; it is not  possible to determine whether plants in this study
 responded to different levels of SOp from those reported in studies of SOp
 alone.

 IDENTIFICATION AND TREATMENT OF UNCERTAINTIES

      Additional  research is necessary to reduce uncertainties associated with
 assessing regulatory alternatives.  This section presents major  areas of
uncertainty that are identified in recent EPA staff assessments  of information
for recent NAAQS reviews  (EPA 1982a, 1982b).  A more explicit treatment is
being developed for handling uncertainties in the biological information that
 is used to support regulatory alternatives.   The relative prioritization of
 these research needs from a regulatory perspective is not directly addressed.
As the discussion of plant response to pollutant mixtures in earlier sections
                                      8

-------
indicates, response can vary greatly. Interpretation and comparison of results
from different studies is complicated by variability in plant response, which
is influenced by a number of factors, including: experimental exposure regime,
exposure situation, biological endpoint, fundamental response triggering
mechanisms, genetics, and environment.

      Perhaps the greatest impact of pollutant mixtures lies in the potential
effects of the complex mix of pollutants that is associated with acidic
deposition.  Here SCL, acid aerosols, CU, and other pollutants may cause or
promote ecosystem effects.  Figure 1-2 shows regional wet deposition of
sulfates for "summer" and "winter" seasons, with wet sulfate deposition
occurring over relatively large areas.   If available data from the Electric
Power Research Institute's Sulfate Regional Experiment (SURE) are
representative of regional S0? and sulfate levels, then SC^ levels may be
expected to be substantially higher than sulfate levels (Mueller et al. 1980).
Moreover, ozone and other photochemical  oxidants and other acidic aerosols
(e.g., nitrates, organics) also occur, in these same regions.

Genetics and Environment

      Both genetics and environment can  affect plant sensitivity to pollutants
(EPA 1982a, 1982b).  Different species vary in their sensitivity to pollutant
exposure, even when environmental conditions are identical.  Moreover, the
influence of environment, especially light and water stress, have been shown
to produce profound effects on plant sensitivity to air pollutants.

Exposure Situations

      The air quality information used to supplement the standard review and
development process focuses on defining  and characterizing exposure to
populations that might be impacted.   In  the case of vegetation,  air quality
information is  usually separated into one of three exposure situations: point,
area,  and regional  exposures.   The conditions of plant exposure for each of
these  situations differ,  as do the concentrations associated with different

-------
Unit Co
tng/mi
250
500
750
1000
1500
ZOOO
3000
4OOO


rw«r*ons:
Kg/ha
2.5
5.0
7.5
10
15
20
30
40


SO.' 71
•q/ha 4-
52 )
104 y
I5« \
210 \
315 ~
420
630
HO
SO. W«f [
Apr-Oct
^oc
— ^
2SC
1
/
/ /
< /
^c
\

)«poi,t
1979
                                      ISO
                    SO*

                    Nov 1979-Mar I960
Figure 1-2.  Seasonal S042' wet deposition  (mg/m2) for North America,
 "Summer" April through October 1979 and "Winter" November 1979
 through March 1980.  Sites (A) reporting  data are from the NADP
 and CANSAP precipitation monitoring networks  (reproduced with  permission
 from Glass and Brydges  1981).

-------
averaging periods, the distribution of concentrations within an averaging
period, and the spatial distribution of concentrations.  Table 1-1 summarizes
information related to plant exposure including:  emission sources,
characteristics of plant exposure patterns (e.g., peak and mean
concentrations), and potential for combined exposures.

Experimental Exposure Regimes

      Studies of plant exposure are separated into two basic categories: plant
responses to controlled exposure and plant response to uncontrolled exposures
in  the field.   In general, controlled exposure studies have not used exposure
regimes representative of those expected to occur in the ambient air.
Interpretation  of some earlier controlled exposure studies is not only
complicated by  unrealistic exposure regimes but also by growing conditions
that were unusually favorable for plant growth, which probably increased the
sensitivity of  plants (EPA 1982b).

      The differences in the distribution of pollutants in time,  space,  and
concentration can be shown by comparison of regional and point source
situations.  Figure 1-2, illustrated regional concentrations over seasonal
averaging periods.  Although pollutant concentrations may be elevated to some
extent, over relatively large regional levels, the changes in pollutant
concentration are typically gradual and extend over large distances.   The
situation is markedly different in the case of point sources. Figure 1-3 shows
a fumigation event near a coal-fired power plant; changes in pollutant
concentration are most strongly influenced by wind direction, windspeed,
emissions, anci mixing Ipvel.   Fumigation events in this situation tend to last
only a short time.  During these fumigations, plants are typically exposed to
both SCL and NCL in SCL/NCL ratios ranging from 3 to 15.   As concentrations of
these pollutants decrease,  the concentrations of SCL and N0? may show
convergence (Noggle and Jones 1981).  Although SCL and NCL may be present
simultaneously,  it is unlikely that peak CL concentrations would  occur
simultaneously.   This situation tends to occur because nitric oxide (NO)
titrates  CL,  elevating NCL  concentrations and reducing CL concentrations. Such

-------
                 0.06
                 0.04
                 0.02
                                                O
                               NO
IX)
                 0.06  -
             o

             8   0.50
                0.40
                0.30
                0.20
                 0.10
                   1000
1100
                                                         1200
                                                      TIME OF DAY
                                                           1400
                   Figure  1-3.   Temporal changes  in ground level  concentrations of  SCL,  NCL,
                   and 0   during a fumigation  event near a coal-fired power plant  (moaifiea
                   with permission from Noggle and Jones 1981).

-------
 information suggests that experimental designs employing sequential exposure
 to pollutants could be a more realistic approach for simulating exposure
 regimes  (see Chapter 2 for further discussion).

      Observations of plant response in the field are needed to confirm
 observations in controlled exposure studies.  However, the lack of control of
 environmental variables affecting plant sensitivity, including the presence of
 other pollutants, reinforces the need for these studies to be complemented by
 controlled exposure studies.

 Mechanisms of Action

      The weaknesses in our understanding of the mechanisms of damage preclude
 the  use  of mechanistic models as a predictive tool.  Available studies on the
 mechanisms by which pollutants cause effects have focused on single
 pollutants.  Basic information on the physiology, growth, and development of
 plants eventually should be useful in the development of mechanistic models.

      Even studies that have used reasonable concentrations for the averaging
 periods  selected, they have not generally reflected the distribution of air
 pollutants within the averaging period that might be expected for the
 situation.  Studies by Mclaughlin et al. (1979) demonstrate that the relative
 distribution of air pollutants within an averaging period can be an important
 determinant of plant response to SCL.

 Biological Endpoint

      The evaluation of plant response for purposes of setting NAAQS is
 complicated by consideration of different biological endpoints.   Depending on
 the objective of the study, researchers have employed a variety of endpoints
 in examining the effects of air pollutants on vegetation.   However, those
 endpoints that can be used as a measure of the intended use of the plant are
most useful.
                                        13

-------
      As the  standard-setting  process  has  evolved,  the  use  of  information  in
the process has  changed.   Available information  suggests  that  foliar  injury
played a prominent role in the setting of  the original  NAAQS.   However,  in
1979,  one of  the primary  reasons  for the relaxation of  ozone secondary
standard was  the lack  of  data  showing  reductions in growth  and yield  in
agricultural  crops or  native  vegetation at exposures below  the level  at  which
the primary standard was  set.   Currently,  major  emphasis  is placed  upon  the
characterizing impacts on intended  use of  the plant. Using this  approach,
foliar injury is of greater importance in  ornamentals,  native  vegetation,  and
crops  whose leaf appearance can be  an  important  consideration  in  marketability
(e.g., spinach).

      Because a  number of studies have employed  foliar  injury  as  the  endpoint,
the associations between  foliar injury and yield have been  sought by  some
researchers as a means of estimating possible effects on  yield from foliar
injury data.   Although increases  in foliar injury and decreases in  growth  and
yield  tend to occur simultaneously  when pollutant exposures are sufficiently
high,  foliar  injury is an imprecise measure of the  effect of pollutants  on
growth and yield parameters.   Growth and yield reductions may  occur with
minimal or no accompanying foliar injury (Reinert and Weber 1980) and it is
possible to have foliar injury with no apparent  effect  on crop yield  (Heagle
et al. 1974).  It is possible  that  effects on growth and  yield are  most
consistently  related to increases in foliar injury  when development is  limited
by photosynthetically  active  surface area  or leaf area.

SUMMARY

      Conflicting results from both short- and long-term  mixture  studies are
difficult to  interpret.   Although these studies  indicate  that  pollutant
mixtures can  produce effects  that are  greater than  additive, especially  at  low
exposure levels, additional  research is needed to resolve reported  differences
in biological responses.   In  addition  to resolving  these  differences,
information in the following  areas  would be useful  for  improving  the
scientific basis of regulatory activities  designed  to protect  vegetation:
                                      14

-------
      a.  Differences in peak pollutant concentrations associated with the
          temporal and spatial patterns should be reflected in  the design of
          exposures employed for pollutant mixture studies.

      b.  Exposures should be representative of peak  and mean concentrations
          occurring in the ambient air.

      c.  The mechanisms by which plants respond to pollutant stress  should  be
          elucidated.

      Current studies involving pollutant mixtures are still characterizing
the types of biological responses to pollutant exposure.   Methodologies  have
evolved sufficiently to develop reasonable models of  plant response to
pollutant mixtures in the field.
                                       15

-------
             2.  THE CO-OCCURRENCE OF S02/N02,  03/S02,  AND 03/N02
                            MIXTURES IN AMBIENT AIR
INTRODUCTION

      A great deal  of the air pollution vegetation effects literature deals
with the direct impacts associated with 0,,  S02,  and N02 air pollutants acting
as independent phytotoxic agents;  there is. a dearth of information that
describes the effects associated with their  mixtures.   Critical  to the
development of relevant dose-response data is the identification of N02,  S02,
and Oo exposure regimes that adequately describe  the concentration, frequency
     O                         '                  ; '  s
of events, length of occurrence,  and  time  between events.  A review of the U.S.
EPA's air quality data information base,  SAROAD,  the Electric  Power Research
Institute's Sulfate Regional Experiment (SURE)  data, and the Tennessee Valley
Authority's (TVA) air quality monitoring  data was undertaken to  characterize
the exposure of pollutant mixtures at specific sites across the  United States.
Air quality information reported by EPA for  1981, EPRI  SURE data for May
through September 1978, and TVA data  for  May through September 1978, 1979,
1980, and 1981 were reviewed.

      In developing estimates of plant exposure to pollutants, consideration
should be given to  characterizing  exposures  that  are similar to  ambient
conditions.  For the purposes of this analysis, we have utilized hourly
averaged air quality data because  the short  averaging  time provides important
information to those scientists interested in developing pollutant exposure
regimes for vegetation effects research.

      Co-occurrence is defined as  the simultaneous  occurrence  of hourly
averaged concentrations at 0.05 ppm or greater for pollutant pairs (S0?/N0~,
03/S02,  or 03/N02).  A 0.05 ppm concentration was selected because minimum
biological responses have been shown  to exist at  these  levels.   Tingey et al.
(1971a)  reported that a 4-hour exposure of several  crops to levels up to  2  ppm
N02 and  0.5 ppm S02, caused no injury when administered singly.   Slight foliar
                                      16

-------
injury was observed at 0.05 ppm N02 and 0.05 ppm S02-  Ashenden  (1978, 1979a)
and Ashenden and Williams (1980) reported growth and yield suppression from
combined exposures of 0.1 ppm NO^ and 0.1 ppm SO-, using a constant fumigation
exposure for 103.5 hours per week for 20 weeks.  These exposures caused
significant reductions in the growth parameters of all four grass species
tested.  Because these exposures were based on a constant fumigation regime
greater than one hour, it was believed that a one time hourly co-occurrence of
0.05 ppm represented a conservative definition for an event.

      Based on a review of available data, EPA has previously concluded that
there is inadequate evidence to determine a yield reduction relationship
associated with vegetation effects for various ambient exposure combinations
of pollutant gases (EPA 1981a, I982b).  This chapter explores the
characteristics of co-occurring air pollutant mixtures and identifies
exposures that may be considered typical of several rural monitoring sites
across the United States.

THE POLLUTANTS

      Sulfur dioxide is one of a number of sulfur-containing compounds found
in the atmosphere.  Although SO,, enters the air primarily from the burning of
coal and oil, it is also produced by other industrial and natural processes.
EPA reports (EPA 1981b) that nationally, the urban SO- problems have
diminished so that only a few urban areas now exceed the air quality standard.
Pollutant peaks appear to be controlled by emissions and the topographical and
meteorological conditions associated with air monitoring sites (EPA 1981b).

      Nitrogen dioxide is one of a family of nitrogen oxides. Nitrogen dioxide
plays a major role in the atmospheric reactions which produce photochemical
oxidants (EPA 1981b).  Two major factors that affect NO- concentrations are
mobile source emissions and photochemical oxidation; both contribute to the
observed diurnal variation in NO  concentrations.  EPA (1981c) reports that
                                A
such a variation is described by a rapid increase in N0? in the morning as the
result of NO emissions and photochemical conversion to NO,,.  This is followed
                                       17

-------
by a decrease of NCL in the midmorning due to advection and  increasing
vertical dispersion and loss of NC^ in various atmospheric chemical
transformation reactions.   Peaks in the N02 concentration are often  observed
during other times (EPA 1981c); elevated N02 levels usually occur between 7 PM
and 6 AM.

      Unlike other gaseous criteria pollutants,  03 is not emitted directly by
specific sources.  It is a secondary pollutant,  formed in the air by
photochemical chemical reactions between nitrogen oxides and volatile organic
compounds, such as gasoline vapors, chemical solvents, and the combustion
products of various fuels. Because the chemical  reactions necessary  to produce
0., are principally controlled by sunlight, in most parts of the country 03
reaches peak levels during the late spring and summer months between 11 AM and
4 PM. Year-to-year variations are associated with factors such as meteorology,
measurement and calibration techniques, and quality control  procedures (EPA
1981b).

THE DATA BASES

      In accordance with the requirements of the Clean Air Act and the
Environmental Protection Agency's regulations for State Implementation Plans
(SIPs), ambient air quality data resulting from  air monitoring operations of
state, local, and Federal  networks must be reported to EPA each calendar year.
The SAROAD base is the established medium for the information distribution.
Ambient observations reported to EPA must satisfy minimum summary
criteria—sampling interval (e.g., continuous, noncontinuous) and period of
coverage (e.g., quarterly, annually).   The criteria (EPA 1982c) for  continuous
observations, with sampling intervals  of less than 24 hours, are

      1.  Data representing quarterly  periods must reflect a
          minimum of 75 percent of the total number of possible
          observations for the applicable quarter.
      2.  Data representing annual periods must  reflect a minimum
          of 75 percent of the total number of possible
          observations for the applicable year.
                                       18

-------
      The SURE program is an EPRI-sponsored investigation of air quality  in
the northeastern United States.  The program is directed primarily at  regional
definition of the relationships between emissions of SCL and the distribution
and concentrations of its reaction product, sulfate. The ultimate objective  is
to develop a regional-scale air quality model capable of predicting sulfate
levels as a function of sulfur dioxide emissions. Investigator-s have collected
S02, NO, NOX, 03, and total suspended particulate matter (TSP) air quality
data for nine sites that EPRI identified as removed from major local emissions
of the above pollutants and their precursors (EPRI 1982).  The EPRI remote
sites provide a valuable data base from which the frequency of co-occurrence
of pollutant mixtures can be evaluated.  In 1978, the nine SURE sites were
located at

                              Montague, Massachusetts
                              Scranton, Pennsylvania
                              Indian River, Delaware
                              Duncan Falls, Ohio
                              Rockport, Indiana
                              Giles County, Tennessee
                              Ft. Wayne, Indiana
                              Research Triangle Park, North Carolina
                              Lewisburg, West Virginia

      Eight sites were included in the analysis because they were considered
rural. For the purpose of this study, sites located near agricultural  land and
containing a point source were identified as rural.  Thus,  Scranton,
Pennsylvania and Indian River, Delaware, with SO^ point sources nearby, were
considered rural and included in the analysis.   Similarly,  sites influenced by
automobile emissions (diurnal N0x fluctuations) such as Rockport,  Indiana and
Ft. Wayne, Indiana, were categorized rural  because they were surrounded by
agricultural land.

      The TVA provided printouts of hourly averaged ambient air quality data
for those sites where simultaneous monitoring occurred for  0.,,  S0?,  and N0?.
For this analysis,  data from seven sites were used:
                                      19

-------
      Allen  7  (Shelby  County,  Tennessee)         1978  T 1980
      Paradise 21  (Muhlenberg  County,  Kentucky)  1978-1982
      Paradise 23  (Muhlenberg  County,  Kentucky)        1980
      Giles  County (Tennessee)                   1980  - 1982
      Land Between the Lakes  (Trigg  County,  Tennessee) 1982
      Murphy Hill  (Marshall County,  Alabama)           1980
      Saltillo (Harden County,  Tennessee)        1979  - 1980

      The Allen 7  site is  in  an urban  location near downtown  Memphis  and  is  a
point source monitor for the  Allen Steam  Plant (6.6 km away).  TVA  reports that
there are various  types of heavy industry located  0.8 km to  the northwest,  an
oil refinery 0.8 km to the south,  and  an  interstate highway  400 m  to  the  east.
Paradise 21 and 23 are rural  sites used to monitor the Paradise Steam Plant.
Paradise 21 and 23 are 7.0 km  and  6.0  km  from  the  point  source. There are  no
other major pollutant sources  nearby.  Most  of the surrounding area is
cultivated or in pasture.   The  Giles County  site is rural, serving as a
regional air quality background monitor,  remote  from  any major sources of air
pollutants.   The Land Between  the  Lakes site is  also  a background  monitor;
virtually all the  surrounding  area is  forested.  The  Murphy  Hill and  the
Saltillo monitoring locations  are  PSD  background sites.  Thus,  except  for  Allen
7, all TVA sites used in this  analysis are located in remote  areas.

RESULTS

      To identify  and  characterize pollutant distribution  at  specific sites,
the 1981 hourly averaged SAROAD air  quality  data were reviewed to  identify  all
sites with a maximum NO^,  SO^,  or  03 concentration equal  to or greater than
0.05 ppm. Each site was then evaluated by  determining whether  one  of  the  other
two pollutants was co-monitored and  also  experienced  an  hourly averaged
concentration equal  to or  greater  than 0.05  ppm.   After  identifying those
co-monitoring sites,  the data base was evaluated with the  following criteria:

      1.   Identify those sites  where co-monitoring of S0?  and  N09
          occurred and where 0.05 ppm  was  measured for each      2
          pollutant  at least once during  the 1981  sampling period.
                                      20

-------
      2.  Identify those sites where co-monitoring of 0, and S02
          occurred and where 0.05 ppm was measured for each
          pollutant at least once during the 1981 sampling season.
      3.  Identify those sites where co-monitoring of 0., and N02
          occurred and where 0.05 ppm was measured for each
          pollutant at least once during the 1981 sampling season.

      Using the site identification code and the above criteria, a computer
listing of the hourly averaged concentrations, by day and by month, was
obtained for each of the 1981 SAROAD ozone, nitrogen dioxide, and sulfur
dioxide monitoring sites for the months May through September 1981.  Data from
each identified site were reviewed for the possibility of the co-occurrence of
0.05 ppm S02/N02, 03/S02, or 03/N02 at least once during the months May
through September.  At least three months of data during the five-month period
had to be available.  Using the SAROAD site identification coding (EPA 1976),
the selected monitoring  locations were segregated into "rural" and "non-rural"
categories. Rural sites  are considered by EPA to be those monitoring locations
that have not been designated as center city, suburban, or remote (far enough
from any activity to measure geophysical background levels).

SO /NO
      Most sites experienced fewer than 10 co-occurrences of S0?/N0? during
May  through September.  Usually there were weeks, sometimes months, between
co-occurrences. Figure 2-1 illustrates the frequency site distribution for the
S02/N02 co-occurrences.  All of the rural sites sampled experienced less than
50 co-occurrences during the five-month season. This amounts to less than 1.5%
of the total hours available (3,672) during the period. Only 6 of the 32 rural
site years had more than 10 co-occurrence events; most non-rural sites
experienced more than 10 events.  Philadelphia experienced 123 co-occurrences
during 1981.

      The Indian River, Delaware rural site (located near the Indian River
Power Plant) experienced a series of co-occurrence events (each lasting
several  hours) on May 26 and 27, and September 13, 19, and 23, 1978.  Figure
2-2 shows the distribution of sulfur dioxide and nitrogen dioxide during the
                                       21

-------
                120"*
                100**
                    * ******
                    * ******

                 80** ******

                    *
                    *

             0       *
             F

                    *


                 60**
             S       *
             ->       *
                    *
                    *

             I
             T    40'




             E




             S

                    *


                 20**
                    * ******  ******
                    * ******  ******
                    * ******  *xx***  ******  ******  ******

                   ) *>AMA*AA*A>M^oa********»*UA*J^A****UA*****^
                      <10    20   30   40    50  60    70   80   90   >90
                                    NO.  OF  CO-OCCURRENCES



Figure 2-1.   S02/N02  Co-Occurrence,  Frequency  Site Distribution
                                          22

-------
24-hour period of September 13, 1978.  The numbers of events during  the  day
began in early morning and lasted for several hours during the daylight.   The
episode disappeared the following day.

      During 1979, the Paradise #21  TVA rural site experienced six
co-occurrence episodes. The May 15 episode is presented in Figure 2-3. The one
co-occurrence during that day was a  typical pattern of episodes monitored
during the season.

      During 1980, the Allen Steam Plant No. 7 TVA site had 29 episodes.
Figure 2-4 illustrates the concentration of S0? and N0? over the period  May
through September.  The number of co-occurrences during the day was  small  and
the time period between episodes was large.  Figure 2-5 shows the episode  for
July 30.

      Figure 2-6  illustrates the June 27, 1981 episode for Fontana,
California.  The  site has been designated by EPA as rural industrial. During
the five months monitored in 1981, the site experienced a large number of
occurrences of nitrogen dioxide concentrations equal to or above 0.05 ppm. The
frequency of occurrence was so great that when an S0~ concentration  above 0.05
ppm occurred, there was a high probability of the simultaneous occurrence of
SO,, and NOo concentrations above 0.05 ppm.  Figure 2-6 shows the presence of
the large number  of nitrogen dioxide concentrations above 0.05 ppm.

      In contrast, the Kansas City, Kansas monitoring site experienced a small
number of NO- occurrences above 0.05 ppm (94 ug m  ) during the five-month
period.   The frequency of S02 concentrations equal to or greater than 0.05 ppm
(131 ug  m~3)  during the period was typical of the data produced for many of
the rural sites analyzed.  Figure 2-7 shows the exposure regime of S0? and NO?
on September 4, 1981.   On that date, at 9 PM, the sulfur dioxide concentration
was 0.06 ppm (134 ug m  ) and the nitrogen dioxide concentration was 0.05 ppm
(102 ug  m  ).   The simultaneous readings equal  to or above 0.05 ppm  lasted for
only two hours.
                                         23

-------
IV)
      C

      0
      /
      N

      C

      E

      N

      T

      R

      fl

      T

      I

      0
          .24-,
.20-
. 16-
. 12-
      NOX --

      SO,
          .08-
          .04-
                                  T
                                  10
                             i     i     i	1	1	r
                             12   14   16  18   20  22
24
            -                     TIME OF  DflT
                              Indian  River, Delaware
                                September 13, 1978

      Figure 2-2.   Indian River,  Delaware, SO-/NO  Co-Occurrence.
         C

         0

         N

         C

         E

         N

         T

         R

         q

         T

         I

         0

         N
                                                                               .GO-i
                                                                     .50-
                                                                     .40-
                                                                     .30-
                                                                                .20-
                                                                     . 10-
                                                                                .00-
~     i    i     i    i     i     i   n     i    i     i    i
 2    4    G    8    10   12   14  16   18  20   22  24
                                                                  S02  	                     TIME OF  DRY

                                                                                           Paradise,  Kentucky

                                                                                              May 15, 1979

                                                                  Figure 2-3.   Paradise, Kentucky,  S02/N02 Co-Occurrence.

-------
c_n
        Q_
        Q.
        LU
        O
        O
        C_3
                                                   Allen #7
                                                     (1980)
                  HAY                                JUN


         Figure 2-4.   Allen Steam Plant,  Tennessee, S02/N02 Concentration Over Time.
JUL

-------
en
        o_
        Q_
        o
        LoJ
        O
        O
        O
            M
      .52-


      .28-


      .24-




      .ie-


      .12-


      .B8-
      .32


      .28-


      .24-


      .29-


      .18-


      .12-
               .04-
                                                                           I

                                                                          SEP
        JUL                AUC


Figure 2-4.   Allen Steam  Plant, Tennessee,  SC^/NCL  Concentration  Over Time  (Cont.)

-------
C
0
N
C
E
N
T
R
q
T
I
0
N


P
P
M
     .24-1
                             10   12  14   16   18   20   22   24
     .OO-1
 SO-
                              TIME OF  DRY
                     Allen Steam  Plant, Tennessee
                             July 30, 1980
                                                                      0
                                                                      N
                                                                      C
                                                                      E
                                                                      N
                                                                      T
                                                                          .24-1
   .20-
                                                                            '2
                                                                           .08
                                                                           .04
   .00
N02
S00
         <     I     I    I     I     I    I     i     1    I     I     1
         2    4     6    8    10   12   14   16   18   20   22   24
                            TIME  OF DPT
                         Fontana,  California
                            June 27,  1981
Figure  2-5.  Allen Steam Plant, Tennessee,  S02/N02 Co-Occurrence.          Figure 2-6.   Fontana, California,  S02/N02 Co-Occurrence.

-------
c
0
N
C
E
N
T
R
fl
T
 I
0
N


U
G
 /
 M
 3
180-i
150-
120-
 90-
 60-
 30
       0
          1	1	1	1	1	1	1	1	1	1	1	1
           2     4    6    8    10   12  14   16   18   20   22   24
 N02	
   2 	                     TIME OF  DflT
                           Kansas City,  Kansas
                           September 4, 1981
  1ppm  =  1960 u9/m3 at 25°C  and 1 ATM pressure.
  Figure 2-7.   Kansas City,  Kansas,  SC^/NC^ Co-Occurrence.

                                    28

-------
      Figure 2-8 shows the frequency site distribution for the number of
occurrences of CL and SCU for the monitoring locations analyzed.  The majority
of sites (135) experienced less than 10 co-occurrences during the season. Only
the Rockport, Indiana and Paradise No. 21 sites had more than 40
co-occurrences during the season (48 and 45, respectively), a small number
when compared with the total number of sites that measured SOp/NC^ and Oo/NOo.

      The Fontana, California site experienced numerous occurrences of ozone
episodes above 0.05 ppm. Therefore, there was a high probability that when the
sulfur dioxide hourly averaged concentration rose above 0.05 ppm, both
pollutants would be present at levels equal to or greater than 0.05 ppm.
Events of co-occurrence, lasting a few hours each, were present in June, July,
August, and September 1981. Figure 2-9 presents the July 23 data for the site.
A large number of ozone episodes above 0.05 ppm is evident.  However, only a
few SOp hourly values above 0.05 ppm were present.

      The Madison County, Illinois site has been coded as rural.  Events of
co-occurrence were present in May, June, and September 1981.   The controlling
variable for determining co-occurrence was sulfur dioxide with ozone events
above 0.05 ppm present in sufficient amounts so as not to be the limiting
factor. As in previous examples, the number of events during the entire period
was small.  Figure 2-10 shows the exposure pattern for a co-occurrence episode
for May 27, 1981.  The number of SO,, concentrations above 0.05 ppm is small.

      The ozone hourly data for Scranton, Pennsylvania were missing for the
Pennsylvania site (May, June and July 1978).  Available data did show
sufficient amounts of ozone concentrations above 0.05 ppm to indicate that the
co-occurrences were controlled by SO- events.   Co-occurrences were present in
                                      29

-------
                                               ibbbbbbbyobobk-

                                20
30
40
                                                             50
                              NO. OF CO-OCCURRENCES




Figure 2-8.   (L/SOp Co-Occurrence, Frequency  Site Distribution
                                 30

-------
c
0
N
.21-
C
E
N
. 17-
T
R
fl
. 13-
T
r
i
0
.08-
N
P
.04-
P
M
S09
r
0
' N
.21-
• • r
f E
m - m- - m
N
.]/-
T '
P
fl
.13-.
f
V ' -08-
\ ' N
r \
A \ p
f \ p
' 1 \ *
' / \ M
- ••- - a - -.r - • -4 	 -* ^ 	 , •






'
•f\ '
- -• • -»-»-* , — -* \ , 	 . • -r - i
11 I I I I 1"1T"T1 .00, ,T , | | •[ i - •; | " • 1
2 4 6 8 10 12 14 16 18 20 22 24 2 4 G 8 10 12 14 16 18 20 22 2'
°3 — -
                           Fontana, California
                              July 23, 1981
Figure 2-9.  Fontana,  California,  O^/SO-  Co-Occurrence.
                        Madison  County,  111inois
                              May  27.  1981
Figure 2-10.   Madison County,  Illinois,  0,/SOp Co-Occurrence.

-------
July, August, and September.  Figure 2-11 presents data for August 1, 1978.
Ozone levels were above 0.05 ppm during mid-day when sulfur dioxide exposures
were high.  The sulfur dioxide concentrations above 0.05 ppm decreased in the
early afternoon and the co-occurrences disappeared.

      The Rockport, Indiana, site experienced co-occurrences in June, July,
August, and September.  An event would last for a few hours, then be followed
by several days or weeks of no co-occurrence before another event.  The number
of ozone events was great and therefore, co-occurrence was controlled by the
S02 events.  Figure 2-12 describes the co-occurrence episode for August 25,
1978.  The ozone remained above 0.05 ppm between 9 AM and 4 PM and the sulfur
dioxide concentrations above 0.05 ppm defined the number of co-occurrences.

      In 1978, the Paradise #21 TVA rural site experienced a large number of
co-occurrences.  Figure 2-13 shows the concentration of hourly averaged 03 and
S0? concentrations from May through September.  Figure 2-14 illustrates the
exposure regime during an episode on September 21, 1978.  The ozone levels
remained fairly constant during daylight, and the sulfur dioxide
concentrations defined the number of co-occurrences.
      Figure 2-15 presents the summary of the frequency distribution for those
sites that were analyzed for O.,/^ co-occurrences. For the three pairs of air
pollutants, the ozone.-nitrogen dioxide combination showed, by far, the
greatest number of co-occurrences.  Several sites in the South Coast Air Basin
of Southern California experienced more than 450 co-occurrences.  The rural
sites of Riverside, Fontana, and Rubidoux, California had more than 100
co-occurrences.  However, most of the analyzed sites (143) experienced fewer
than 10 co-occurrences.  Denver, Colorado and San Jose, California did
experience more than 100 co-occurrences.

      Rubidoux, California, located in the South Coast Air Basin, is
designated as a rural commercial site.  There were ozone and nitrogen dioxide
co-occurrences in May, June, July, August, and September 1981.  Because
nitrogen dioxide concentration maxima tended to peak in the evening or early
                                      32

-------
CO
CO
         C

         0

         N

         C
         E

         N

         T

         R

         H

         T

         I

         0
 17-'
. 10-'
             .07H
             .03-1
             .00
                                                                                 .20-,
         SO,
                                                            20  22   24
                                      TIME OF DRY
                                 Scranton, Pennsylvania
                                     August 1. 1978
                                                                                              0   12  14   16   18  20   22   24
                                                                     .Q7H
                                                                                 .00
                                                                 so.
                                                                                              TIME OF  DflT
                                                                                            Rockport,  Indiana
                                                                                             August 25,  1978
         Figure 2-11.  Scranton, Pennsylvania,  0,/S02 Co-Occurrence.
                                                                 Figure 2-12.   Rockport, Indiana,  03/S02  Co-Occurrence.

-------
Q_
Q_
O

-------
OJ
en
         Q_
         CL
         o
         UJ
         CJ
         CD
.58-
.49-
.55-
.29-
° .21-
.11-
.07-






:r.:.-ilbu


L




i
I
Am It L L ti i
uumt^i i- ki^Ml n4*nMl «r-^^i A^^^kMPi« ft • Wl m






{-


lAAnmvIL>

^UjJI^L^





jjl



:it,
                 .35-
              2
              O

              S .2.
                .14-
                   JUL
                           AUC                                  SEP

Figure 2-13.   Paradise, Kentucky,  03/S02 Concentration  Over Time (Cont.)

-------
C
0

N

C
E
N

T
R

q

T
I
0

N
37-J
22-J
             • *--*  *--»----—
           I     I     I     I    I     I     I     I     I     I    I     I
           2     4    6    8    10   12   14   16   18   20  22   24
 °3 --
 SCL
   ' 	                      TIME Or  DRY
                           Paradise, Kentucky
                           September 21, 1978
 Figure  2-14.  Paradise, Kentucky, 03/S02 Co-Occurrence.
                            36

-------
150*
*
*•
*

*
*
*
*•
*
*
1 ?S**
*
*•
*r
*
*•
*
*
*
*
*•
100*
*
*
*
*•
*



*******
******
******
******-
******
*******
******
*i***X-
***x**
•****x*
*******
******
******
****x*
***x**
•)bt;feJoi^-
***3t**
*******
***xx*-
***wk*-
******

*******

******
                  7 c:**-
                  ' J
******
**x*x*-
•****>*
*A***X
*******
*******
******
*Wc*x*-
**fcdot
                  50**
                  25**
                       ****>*•
                       ****xJr
                       *******
                       •*****x
                       *fcyodr
                       ******
                       *3t*-toc*
                       ******
                       ******
                       ****X*
                       **K**A-
                       ******
      ******
      ******
      *****x
      ******  ^^^
      •*/obtb
    *fct^**


    •***••.-bi.
                        <10   20     30   40
                         50   60    70   80
90  >90
                                      NO.  OF  CO-OCCURRENCES

Figure 2-15.   CL/NCL  Co-Occurrence,   Frequency  Site  Distribution
                                          37

-------
morning, the co-occurrences were present at these times.  Figure 2-16
illustrates the monitoring data for June 26, 1981.  Because ozone
concentrations were mostly above 0.05 ppm, the number of co-occurrences was
large.

      The Indian River, Delaware site experienced its only two co-occurrences
on June 3, 1978.  There were NO  events on May 26 and September 13, but they
                               A
did not match the dates of the many ozone events.  Figure 2-17 shows that the
co-occurrences appeared in the late afternoon and early evening.
      During 1979, the Paradise No. 21 site experienced the only
co-occurrence that was measured from 1978 through 1982.  Figure 2-18 shows the
exposure regime on July 17, 1979.  The co-occurrence was measured at 11 AM in
the morning.
      The Allen Steam Plant No. 7 TVA site experienced nine
co-occurrences during the 1978 season.  Figure 2-19 describes the episode that
took place on July 6.  Ozone concentrations were fairly high during the
daylight hours. In the early evening, the nitrogen dioxide concentrations rose
above 0.05 ppm, resulting in the co-occurrence events.  Figure 2-20 shows the
hourly averaged concentrations for 0, and N0? for May through September.

DISCUSSION

      The seasonal variation in specific pollutants was evident.  For all 1981
SAROAD sites that measured ozone above 0.05 ppm, 92% of the maximum hourly
readings were observed during the May through September period.  Less than 20%
of the N0? hourly maxima were recorded during the same period. While the daily
ozone hourly peak concentration usually occurred in the late morning and early
afternoon, the daily nitrogen dioxide peak usual ly occurred in the early
morning or late evening.  Many of the sulfur dioxide peaks occurred during the
daylight hours.
                                       38

-------
OJ
UD
C

0

N

C

E

N

T

R

R

T

I

0

N
           .24-1
           ,20-
            16-
            12-
           .08-
           .04-
       °3  —-
       NO-
    . 00 i ,  .  T .  T  .'  ,
          2468
                                        i     r    T    i     i    i  ^ i
                                        12   14   16   18   213  22   24
                                   TIME OF DflT

                               Rubidoux,  California
                                   June 26,  1981
C

0

N

C

E

N

T

R

fl

T

I

0
                                                                               .24-,
                                                                               .20-
                                                                                16-
                                                                               .08-
                                                                         ,04-
                                                                               .00
                                                                      °3 ™-
                                                                                                                  -••-•• »%
                                                                                                jr-.-^
                             T  ~ "1	1	—'—I	1	1	1~
                             10   12   14   16   18  20   22   24
                                                                                                  TIME OF  DflT

                                                                                              Indian River, Delaware
                                                                                                   June 3, 1978
       Figure  2-16.  Rubidoux,  California, 03/N02 Co-Occurrence.
                                                                     Figure 2-17.  Indian River, Delaware,  Cu/NO  Co-Occurrence.
                                                                                                         «j   X

-------
o
c
0
N
. 17-
C
E
N
.13-
T
R
fl
. 10-
T
I

0
.07-
N

P
.03-
P
M
.00-














r • ",
y
j * *•-
A -•• '•' >k-
W '> •' • • •«.
/;' \ V '»
/,' \
/;' 1
// \/ \
'
, " i i i * i i i i i i
2 4 6 8 10 12 14 16 18 20 22 2
NO,
X ——— -t- r tit— r»r- n/-iv
                                                                                   .20-1
                                  Paradise,  Kentucky
                                    July  17,  1979
       Figure 2-18.   Paradise,  Kentucky, 03/N02 Co-Occurrence.
                                                                                                        8    10   12  14   16   18  20   22   24
                             TIME OP  DQY
                       Allen Steam Plant,  Tennessee
                              July 6,  1978
Figure 2-19.   Allen  Steam Plant, Tennessee, 03/NO- Co-Occurrence.

-------
      .32-

      .28-

      .24-

      .29-

      .16-
   fM
   i  .121
D-
CX
O
O
Allen # 7
 (1978)

                                            JUM
                                      JUL
 Figure  2-20.   Allen Steam  Plant,  Tennessee 03/N02  Concentration Over Time.

-------
       .at-
o

-------
      Using rural and non-rural 1981 SAROAD data and the criteria previously
mentioned, 66 percent of the sites that co-monitored S09 and NCL experienced  a
co-occurrence at least once during the five-month period. Of the CL/SO^ sites,
63 percent experienced a co-occurrence at least once; 71 percent of those
sites that co-monitored CL and N02 experienced at least one co-occurrence
during the period.

      The EPRI SURE hourly averaged air quality data for S09, (L, and NO  were
                                                           C.   u        X
analyzed for co-occurrence events. Because the EPRI data were reported in N0x,
S09, and 0, concentrations, it was assumed for this study that the NO  could
  L.       -J                                                          X
serve as a surrogate for N09 concentrations; using NO  as a surrogate results
                           C.                         A
in an overestimate of the co-occurrences of N02 with either S02 or On-  Eight
EPRI sites were reviewed for the co-occurrence of pollutant mixtures where the
concentrations were equal to or greater than 0.05 ppm.  Only one site (Indian
River, Delaware) experienced at least one co-occurrence of S09/N0 .   Four EPRI
                                                             c.   X
sites experienced at least one co-occurrence of 0.,/S09 (Scranton,
Pennsylvania; Duncan Falls, Ohio; Rockport, Indiana; and Giles County,
Tennessee).  The  Indian River site was the only one that experienced at least
one co-occurrence of Oo/NO  during the period May through September.

      The TVA provided hourly averaged air quality data for seven sites that
had monitored the three pollutants since 1978.  Of the 14 site years reported
for S02/N02 co-monitoring, eight reported co-occurrence at least once.  Of the
14 site years reported for 03/S02, 11 showed co-occurrence.  For the 12 site
years recorded for 03/N02, only 4 showed co-occurrence.

      Table 2-1 summarizes the results of the SAROAD, SURE, and TVA data
analysis.  For ozone, sulfur dioxide, and nitrogen dioxide, there were 752,
921, and 345 total site years monitored, respectively. Of the total  monitored,
370 (03), 321 (S02), and 291 (N02) site years were identified for those
locations where co-monitoring occurred.   Using the monitoring data produced
from EPA, EPRI,  and TVA,  32, 36,  and 34 rural monitoring site years  were
evaluated for S02/N02,  O-i/SOo, and Oo/N02 co-occurrences, respectively.
                                        43

-------
                            Table 2-1   Summary of Site Years Analyzed
                          (EPA SAROAD, EPRI SURE, and TVA  Data)
   Pollutant   Total  # Site Years     # Sites Yrs. Where     # Rural Site Yrs.  >  Thresh.
                   Monitored         Pollutant > Threshold    & Co-Monitoring Occurred
                                  and Co-Monitoring Occurred
°3 752
S02 921
N02* 345
S02/N02*
o3/so2
03/N02*
370
321
291
91
124
146
40
37
37
32
36
34
*  EPA provided the  EPRI  SURE  monitoring  data  in  the form of N0x, S02, and

concentrations.  EPA and  TVA reported  N02,  S02,  and  03  concentrations.
                                              44

-------
CONCLUSION

      Analysis of selected ambient air quality data collected by EPA, EPRI,
and TVA shows that, for most cases, the constant artificial fumigation
exposures do not mimic actual exposures.  Since co-occurrence may be
infrequent, researchers may want to focus on the sequential pollutant
exposures characteristic of the rural sites analyzed.  For example, while the
daily ozone hourly peak concentration usually occurred in  late morning and
early afternoon, nitrogen dioxide typically peaked in the early morning or
late evening.  Similarly, sulfur dioxide episodes mostly appeared during the
same daylight hours that ozone concentrations reached their maximum.

      The monitoring data used in this analysis support the conclusion that 1)
co-occurrence of pollutant mixtures lasts only a few hours per episode and 2)
the time between episodes is great (weeks, sometimes months).  The analysis of
rural air monitoring data, generated by three different organizations,
represented a first-step effort in characterizing rural sampling sites.  Air
quality data from a subset of the data bases were used to  identify the
distributions for S02/N02, 03/S02, and 03/N02 co-occurrences, respectively.
Many of the sites were located in rural agricultural areas.

      The  lack of a comprehensive rural air monitoring data base has made it
difficult to judge the representativeness of those sites used in the analysis.
However, by 1) defining hourly averaged concentrations of 0.05 ppm and above
as an event, and 2) combining air quality data bases (SAROAD, SURE, and TVA),
the analysis shows a consistent exposure pattern, suggesting that the use of
sequential exposure regimes should receive more attention and that researchers
may want to reconsider their importance relative to co-occurrence exposure
regimes.
                                        45

-------
               3.  EFFECTS OF POLLUTANT MIXTURES ON VEGETATION

INTRODUCTION

      The first comprehensive reviews of literature on mixture effects were
published eight years ago (Reinert et al.  1975,  Williams  and Ricks 1975).  Much
of the research concerning air pollutants  and  pollutant mixtures has since
been summarized in several books (Heck et  al.  1982, Ormrod  1978, Unsworth  and
Ormrod 1982),  recent  review articles  (Ormrod  1982,  Wellburn 1982), and
research reports (Fujiwara 1973, Fujiwara  and  Ishikawa 1976,  Ishikawa 1976,
Reinert and Heck 1982,  Yamazoe and Mayumi  1977).   These reviews and summaries
reveal that only a few combinations of pollutants  have been studied and that
little attention has  been given  to environmental  and biological factors that
influence vegetation  responses.  Few studies  have  dealt with any aspects of the
responses of major plant species to mixtures  of  air pollutants at ambient
concentrations administered under typical  ambient  environmental
condi tions.

      When two pollutants occur  in combination,  there  is  the possibility of
visible injury totaling more than the sum  of  visible injury caused by each
pollutant alone.  This concept,  which has  encouraged research, has been
extended to changes in growth and yield as well  as  to  biochemical and
physiological  changes in plants  following  exposure  to  pollutant mixtures.   The
concept is only one of the possible categories of  plant response to pollutant
mixture but it is the one which  may be of  greatest  concern  in vegetation
effects assessment.  When one pollutant has  no effect  on  the plant response to
the other pollutant,  the category is  termed  no joint action.   The category
joint action implies  that both pollutants  have some effect  on the plant
response.  This latter category  is frequently  divided  into  the subcategories
additive response, when effect^ equals effect^  plus effect^, and interaction,
when effect^  is not  equal to effect^ plus effect,,.  There  are two possible
types of interaction:  synergism (greater  than additive action) when effect,,,
is greater than effect^ plus effect2, and  antagonism (less  than additive
action) when effect^ is less than effect-^ plus  effect,,.   It is the concept of
synergism that has been of greatest interest  and  concern.   It would be
difficult, using the  present experimental  knowledge, to fully characterize the

                                       46

-------
nature of the joint action of two or more pollutants on major species  in
environments typical of ambient conditions.

      Joint action of major gaseous pollutants has been by far the most
studied kind of mixture response.  This is because the more phytotoxic air
pollutants are gaseous (e.g., CL, SCL, NOo and HF); there is a greater
knowledge of their atmospheric chemistry and occurrence; and they are easily
generated and monitored in experiments.  However, there has been little or no
research on mixtures involving minor gaseous pollutants (e.g., H~S, HC1, C^j
NhL, and (^H*).  Air pollutants occur also as aerosols or as dissolved or
suspended material in precipitation.  Some information is available on
gas-aerosol joint action (Krause and Kaiser 1977, Singh 1980).  Aside from
problems created by the physical and chemical heterogeneity of aerosols,
deposited material remains on plant foliage after joint exposure has ceased
and thus constitutes a virtually continuous source of exposure.  For this
reason, the interaction of gases with ions of trace elements or heavy metals
(e.g., Cd, Ni, Cu, and Zn) on or in plants may be important (Czuba and Ormrod
1974, Krause and Kaiser 1977, Lamoreaux and Chaney 1978, Ormrod 1977, Toivonen
and Hofstra 1979).

      The intent of this chapter is to summarize and interpret existing data
concerning plant responses to pollutant mixtures.  Since 0.,, S0?, and N0? have
had most attention to date in mixtures research, much of the discussion will
focus on paired combination mixtures (0, + S0?, 0-^ + N0?, S0? + N0?) and on
combinations of 0., + SO- + NO^.  Even though most research has focused on the
above noted mixtures, studies of mixtures of S0? + HF (Mandl et al. 1975,
Mandl et al. 1980, Matsushima and Brewer 1972, McCune 1983) and N02 + HF
(Amundsen et al. 1982) have also been reported.  Also there have been recent
reports concerning interactions between acid rain and S0? (Irving and Miller
1981) and acid rain and 03 (Troiano et al. 1981).

EXPERIMENTAL METHODS AND INTERPRETATION OF DATA

      The species of concern, stage of growth, and other plant factors,
together with the need for control, separate and combined treatments, and
considerable statistical  precision, have dictated certain requirements for

                                      47

-------
methodology in mixture studies.   Experimental  protocol  appropriate for mixture
studies has been developed to increase the validity and comparability of air
pollutant research (Heck et al.  1979).  Studies of mixture effects have taken
place in controlled environment  chambers,  in greenhouses,  and in field
facilities.  Each system has its particular merits.  Fully  controlled
environments provide relatively  high precision and repeatability but response
patterns may not correspond well with field responses.   Field facilities
provide the most direct comparison with open field conditions but
results may not be replicable because of changing  weather  patterns outdoors.

      Most studies to date of mixture effects  have been conducted in
controlled environments or greenhouses.  The utilization of continuous stirred
tank reactor chambers by Heagle  and Johnston (1979),  Le Sueur-Brymer and
Ormrod  (1983), Reinert and Nelson (1980),  and  others  has been a recent
innovation in mixtures research.  A few mixture studies have been conducted
entirely in field facilities and a few have included  both  field evaluations
and controlled exposures in an attempt to  allow direct  comparisons.  Menser and
Hodges  (1970) compared tobacco cultivar sensitivity  to  07  + S09 in the field
                                                        O      L-
with that determined with controlled exposures in  the greenhouse.  There was a
major shift of sensitivity of one cultivar. Hodges et al.  (1971) also reported
comparisons of chamber responses of tobacco to 03  +  S0? with field responses.
Beckerson et al. (1979) compared bean cultivar sensitivity to 0^ + SOp in
controlled environments with injury development following  ambient outdoor
exposure.  Outdoor responses correlated more strongly with sensitivity to 0-
than with sensitivity to mixtures.

      Outdoor exposure chambers, with environmental  conditions more
characteristic of ambient conditions, have been used  by some researchers.
Heagle et al. (1974, 1983) used  open-top outdoor chambers  for long-term 0^ +
                                                                         O
S02 treatment of soybean.  Mandl et al. (1980)  used  similar chambers to study
S02 + HF responses of sweet corn and Heggestad  and Bennett (1981) used such
chambers to observe SO^ enhancement of 0^  injury to  snap beans.  Hill et al.
(1974)  used a portable field chamber to expose desert plants to S02  + N0? at
concentrations measured downwind of a large coal-fired  power plant,  while
Thompson et al.  (1980)  used open-top chambers  for  their studies of desert
plant responses  to S02 + N02-  Other field chambers  have been utilized by

                                       48

-------
Foster et al.  (1983) and Oshima (1978).  Bennett et al.  (1980) grew snap beans
in the field and exposed them to CL + H?S through a plastic duct assembly.  A
linear gradient exposure technique was used by Reich et  al. (1982).  Ashenden
et al. (1982)  have described a system for exposing plants to SCL + N02, using
hemispherical  greenhouses having good air circulation and near-ambient
temperatures.

      Numerous studies exist which allow only individual comparison of
treatments because there were not sufficient experimental units available to
permit a full  concurrent examination of mixture effects. In all these studies,
the interaction or dependency of the effect of one pollutant on the level of
another could  not be fully evaluated due to missing treatments.  An important
consideration  was whether or not there were enough exposure chambers to run
all treatments at the same time.  If not, considerable confounding may have
developed when treatments could not be evaluated under similar conditions.
This might have resulted in apparent pollutant interactions which did not
reflect the true plant response.

      Most researchers have used factorial experiments and analysis of
variance for interpretation of combined effects of pollutants, A statistically
significant interaction of two pollutants has been regarded as evidence for
synergistic or antagonistic effects.  Such techniques were used by Ashenden
(1979a, 1979b), Bull and Mansfield (1974), Gardner and Ormrod (1977), Heagle
and Johnston (1979), Tingey and Reinert (1975), Tingey et al. (1973b),
Wellburn et al. (1976), and others.

      More information may be obtained concerning the ability of each
pollutant to produce a biological response by averaging  effects over the
presence and absence of other pollutants.  One type of factorial design
involves using one concentration of each of three pollutants, alone and in all
two- and three-way mixtures, plus a charcoal-filtered air control treatment.
Thus,  eight treatments are involved and the main and interaction effects of
each pollutant factor can be assessed through treatment  component contrasts
(Heck  et al. 1979).  Experiments using this design have  been reported in which
pollutant concentrations were either high (Reinert and Gray 1981, Reinert and
Heck 1982) or  low (Reinert and Heck 1982).  The exposure durations varied from

                                       49

-------
3-6 hours and the number of exposures from one (Reinert and Gray 1981) to many
(Reinert and Sanders 1982, Sanders and Reinert 1982a, 1982b).  Plant species
(Reinert and Sanders 1982, Sanders and'Reinert 1982b), as well as cultivars
within species (Sanders and Reinert 1982a),' have been compared using this
design.

      Additional  information on the nature of interactions can be obtained by
determining the effect of increasing concentrations of one pollutant on plant
growth and other responses in the presence of more than one concentration of a
second pollutant.  When three or more concentrations of one or both pollutants
are used in the experimental design, the dose-response relationships may be
                                                                       v,
evaluated. The responses may be described as linear or curvilinear and further
serve to interpret some of the complexities of additivity, synergism, and
antagonism found  when only one concentration of each pollutant is used. Such a
mathematical/statistical approach to CL + SCL + NOp interactions has been used
by Reinert et al. (1982) to study the influence of sub-injury threshold
concentrations of SO,, + NCL on plant responses to 0,.  A numerical evaluation
of dose-response surfaces in terms of linear and curvilinear components was
also presented.

      Ormrod et al.  (1983a) utilized quadratic polynomial equations,
three-dimensional response surfaces, and contour plots to evaluate the effects
of (L and SOp on one cultivar each of lettuce, radish, and pea, using a
rotatable experimental design.  The use of the rotatable design decreased the
number of required treatments, compared with a full factorial design. For this
study, plants were grown in a controlled environment chamber and exposed to
seven combinations of CL and SOp. Injury was evaluated on the basis of visible
chlorosis and necrosis and growth was evaluated as leaf area and dry weight.
The contour graphs in Figure 3-1 are two-dimensional representations of
three-dimensional surfaces. The concentrations of SOp and (K form the abscissa
and the ordinate, respectively, and the response is shown as a series of
isoeffect or contour lines.  The shapes of the isoeffect lines illustrate
cross-sections of the surface, while their'spacing shows the rate of change or
curvature of the surface.  These contour plots indicated the diversity of
response patterns and particularly that,some response "variables demonstrated
                                       50

-------
                                       /\/\yX/\/s
                                              \/\'\/^'
Figure  3-2.   Graphical  representation  of  the  four  response  regions of
             practical  interest when two  pollutants  are combined.  For
             explanation  of  I,  II,  III,  and  IV,  see  text. The blank portion  in
             the  area  in  which  no effects  occur, E3  designates the area with
             pollutant  A  effects, S3 designates  the  area with pollutant B
             effects,  and S3 designates  the  area with  combined effects of A
             and  B.  The  number of  circles indicate  numbers of results with
             each kind  of  outcome in the  research  on effects of  SO- + N0? on
             native  desert plants reported by  Thompson et al. (198u).
                                    51

-------
additive effects at low concentrations of 03 + S02 and antagonistic responses
with increasing concentration of both gases in combined treatments.

      Other statistical methods have also been used.  White et al.  (1974) used
tests of non-additivity of pollutant effects.  A statistical test  of synergism
has been made with models derived by polynomial analysis of injury index data
(Macdowall and Cole 1971).  Chi-squared analysis was used by Jacobson and
Colavito (1976).  Probit analysis has been used to interpret antagonism and
synergism, by determining median effective doses (Macdowall and Cole 1971,
Jacobson and Colavito 1976). To increase precision, Ormrod et al.   (1983b) used
covariate measurements to account for significant within-treatment variation
in plant growth.  Some differences that were not significant in conventional
analysis of variance were significant when tested by analysis of covariance.

      Most reports of studies involving mixtures have included, or implied, an
assessment of the joint action of the pollutants in terms of additive effects,
greater than additive (synergistic)  effects, or less than additive
(antagonistic) effects, using the terminology of Tingey and Reinert (1975).
Responses have been expressed in terms of amount of injury or changes in
threshold concentrations causing injury.  Additivity has been suggested for a
number of diverse plant responses (Ormrod 1982), but many reports  have
suggested synergistic responses.  Reports of antagonistic responses have been
more limited.  Many of the reports on additivity, synergism, and antagonism
indicated that the nature of the interaction was dependent on factors such as
pollutant concentration and exposure duration, as well as on the species and
gases studied.

      The threshold phenomenon may be an important component of response
patterns to mixtures.  In general, when pollutant concentrations are below or
at the threshold for individual visible injury responses, synergism (in terms
of reduced growth and plant yield) is more frequently observed.  As the
concentrations of both pollutants increase in mixture above their  individual
injury thresholds, weight loss may only be additive.  When the concentrations
of the pollutants are relatively high, antagonism often develops and further
weight loss may be minimal.  This threshold phenomenon resulting in apparent
synergism wi11  be particularly important in 2 X 2 factorial experiments for

                                       52

-------
the case when neither pollutant alone produces a response, but their
combination does.  When a threshold exists, it is possible that an apparent
interaction results because no injury occurs until the weighted sum of the two
components exceeds a certain value.  The weighting would give a measure of the
relative effectiveness of each component.  The sum of individual  effects, near
the threshold, may not be the appropriate term for comparison with joint
effects to determine interaction.

      For a particular set of combinations of two pollutants, the responses
can be graphically presented to illustrate the four regimes of practical
interest:  (I) where neither an effect of pollutant A, pollutant  B, nor
combined effects of A and B occur; (II) where combined effects occur but an
effect of neither A nor B occurs;  (III) where two or more of the  three effects
occur; and (IV) where effects of A or B occur but combined effects of A and B
do not.  These four possible outcomes for a single and mixed two  pollutant
study are diagrammed in Figure 3-2.  The frequencies of such outcomes are
illustrated for the data of Thompson et al. (1980) who exposed several desert
species to a wide concentration of single and mixed SCL and NOo for several
weeks.  An effect was considered to be a decrease in some measure of weight,
linear growth, or reproduction to  less than or equal to 90% of the unexposed
controls.  The effects were grouped into the four possible outcomes as
indicated by the Roman numerals in Figure 3-2. This illustrates that all types
of outcomes were found with these exposure regimes used in this research
project, but that the most frequent responses were either no effects or
effects of both combined and single gases.

      Most research with mixtures has been conducted in controlled exposure
facilities using certain concentrations of gases for a specified  duration.
While the need for systematic testing of a range of mixture concentrations has
been recognized (Heagle and Johnston 1979, Mandl  et al. 1975), few
investigators have described the exposure in terms of dose (the combination of
pollutant concentration and duration), or have manipulated the components of
dose.  The concept of pollutant flux or uptake rate has barely appeared in the
literature on effects of air pollutant mixtures even though the amount of
pollutant sorbed may be much more closely related than concentration to the
response.

                                       53

-------
Figure  3-2.   Graphical representation of the four response regions of
             practical interest when two pollutants are combined.  For
             explanation of  I, II,  III, and IV, see text. The blank portion  in
             the  area  in which no effects occur, E3 designates  the area with
             pollutant A effects, G3 designates the area with pollutant B
             effects,  and EB  designates the area with combined  effects of  A
             and  B.  The number of  circles indicate numbers of  results with
             each kind of outcome in the research on effects of S00 +  N00  on
             native  desert  plants reported  by Thompson et  al.  (1980).
                                                                        2
                                    54

-------
      Most investigators to date have used simultaneous exposures to
nonvarying concentrations.  In the natural environment, peak concentrations of
pollutants may occur at different times for different pollutants.  Such
patterns may have considerable impact, with the preconditioning of plants by
one pollutant affecting their response to another pollutant.  Matsushima and
Brewer (1972) investigated sequential reciprocal exposures of orange to SO- +
HF to determine whether one gas influenced the subsequent response of the
plants to the other gas, but found little effect.  Costonis (1973) found a
sequence of CU followed by SCL to be more toxic to eastern white pine than
exposure to both gases simultaneously.  White et al. (1974) found that neither
SOp nor NOp pretreatment of alfalfa affected response to a subsequent exposure
to SCL + NO-.  Exposure to SO- before exposure to 0, + SO- markedly influenced
bean, cucumber and tomato sensitivity (Hofstra and Beckerson 1981).  Many
investigators have subjected plants to intermittent treatments within overall
long-term exposures.  Tingey et al. (1971b, 1973b) noted that this might allow
plants to repair injury and regain normal metabolic functions during the
non-exposure period.

      In some cases where many processes are involved, such as in components
of yield, there may be a multiplicative, rather than additive, effect.  Each
process may have a different dose-response function. For example, Reich et al.
(1982) showed different dose-response relationships for three components of
soybean yield.  In such a case the dose-response relationship may be
approximated by a third- or higher-degree polynomial or at least a four
component function.  Sometimes a multivariate approach may help to elucidate
the nature of the effects or at least serve to simplify the nature of the
responses.  For example, in chronic exposures to CL and/or SO- of alfalfa,
treatment-induced changes in foliage and root dry weight were so closely
associated that a weighted sum of the two could serve as a single measure of
response (Tingey and Reinert 1975). On the other hand, with acute exposures of
radish, 0, and S0? appeared to have opposite effects on a weighted sum of root
and leaf dry weight, but no interaction. However, with reference to a weighted
contrast between roots and leaves, their effects alone were similar
(decreasing the difference between root and shoot), but opposite (increasing
differences) when both were present.
                                       55

-------
      Whether the concern is about visible injury on foliage or growth
effects, one should not expect response to be linear with dose nor should one
expect a single mode-of-action as pollutants may induce increases, as well as
decreases, in growth (Bennett et al.  1974) or other responses.  It may well be
that linearity is adequate over narrow ranges, but not over the entire range
of interest.  When a linear dose-response function is found, as with the total
dry weight of radish plants (Tingey et al. 1971b), the dose can be expressed
as a weighted sum of SOp and Oo concentrations.  Other responses also appear to
approximate what would be expected from a linear relationship to a weighted
sum (e.g., K-efflux from 'White Cascade'  petunia leaf discs [Elkiey and Ormrod
1979a]).  When a quadratic function is found (Tingey et al. 1973b), a weighted
sum of SOp and CU concentrations, which implies  a constant relative
effectiveness, may also be used as a dose-variable.

CHARACTERISTICS OF PLANT RESPONSE

      Responses to pollutants and pollutant mixtures have been detected in
several ways:  visible symptoms of injury; altered growth and development;
physiological and metabolic imbalances; and accumulations of certain elements.
Growth and yield are often the most important response variables and it is
probably this concept that effects may be greater than additive (synergistic)
that has been the predominant concern of combined pollutant studies.

Foliar Symptoms

      The practical significance of visible foliar response has been mainly
in the diagnosis of effects.  Kinds of symptoms, and their distribution on a
plant or between species, have provided the investigator with inferences as to
the kinds of air pollutants (qualitative factors).  The degree of symptom
development has allowed inferences as to amount  of pollutant or exposure
(quantitative factors).  Degrees of symptom development has also served as a
measure of the likelihood of effects on growth and reproduction.
Exposure-effect relationships derived from visible injury data have allowed
inferences to be made as to the nature of the dose-response relationship, the
effects of biologic and environmental factors upon it, and the mode of action
of air pollutants in the plant.  However, if the combined effect of pollutants

                                       56

-------
of air pollutants in the plant.  However, if the combined effect of pollutants
alters the qualitative or quantitative foliar symptom characteristics, errors
in diagnosis could have occurred.

      Many investigators have devised a leaf injury rating system and have
used the data obtained for interpretation of mixture effects.  Few have
assessed differences in appearance by providing descriptions of the injury. In
some cases, experiments did not last long enough to permit the development of
stable injury symptoms.  In general, ozone injury symptoms have dominated in
studies of mixtures of 03 + S02 (Menser and Heggestad 1966, Menser and Hodges
1970, Hodges et al. 1971, Tingey et al. 1971a, 1973b, Heagle et al. 1974,
Elkiey and Ormrod 1979b).  There were exceptions and, in some cases, different
symptoms developed in response to the combination.  Grosso et al. (1971) found
different symptoms of 0^ + S02 on tobacco leaves than when CL alone was the
pollutant. Combined CL + SO- injury to apple leaves differed from either CL or
SO-  injury (Kender and Spierings 1975, Shertz et al. 1980a).  Undersurface
glazing, a symptom usually attributed to PAN, was found on petunia leaves
exposed to CL + SO- (Lewis and Brennan 1978). Cucumber exposed to 0- + SO- had
an additional symptom of interveinal chlorosis, compared with CL alone
(Beckerson and Hofstra 1979a).  Visible injury to three woody species, caused
by exposure to 0- + SCL, included symptoms of each pollutant alone (Carlson
1979).  Acute responses of eastern white pine to CL + SCL had different
symptoms from responses to either CL or SCL alone (Costonis 1973), but
Dochinger et al.  (1970) reported that (L and SO- separately or in combination
produced the same initial symptoms on eastern white pine.

      Duration of exposure is important.  Jacobson and Colavito (1976)
indicated antagonism of CL + SO- in navy bean and attenuation of visible
injury with short term exposure.  This contrasted with Hofstra and Ormrod
(1977) who found  unique and severe injury symptoms on navy bean leaves by 0- +
SO-  after several days' exposure.  Symptoms of injury from mixtures of SO- +
NO-  differed greatly, in several species, from those caused by single gases
(Tingey et al. 1971b).  In contrast, Hill et al. (1974) found the symptoms of
SO- + NO- on many species to be the same as those for SCL, and Matsushima and
Brewer (1972) found SO- + HF -induced chlorotic patterns to be the same as for
individual gases.  Kohut and Davis (1978) found an interaction of 0^ + PAN

                                       57

-------
which affected lower and upper leaf surfaces.  On. the lower surface, the two
gases were antagonistic in causing injury; on the upper surface, they were
additive or synergistic.

so2 + o3

      The synergistic action of SOp + 03 first noted on tobacco in 1966
(Menser and Heggestad 1966) encouraged all subsequent research on pollutant
mixtures.  During the first few years that followed, the concept of
greater than additive or synergistic amounts of foliar injury, following
exposure to pollutant mixtures, became accepted as a frequent occurrence.
However, Tingey et al.  (1973a) reported that not all visible-injury response
to mixtures of SCL + 0, was synergistic.  They concluded that foliar injury
responses for six plant species could be additive, greater than additive, or
less than additive depending on the species, and on the concentration and
exposure duration of both pollutants.  Visible foliar injury has been widely
used to evaluate variable species responses to S0? + (k.  Examples of research
reports illustrating the nature of joint effects on visible injury are
presented in Table 3-1.  The diversity of effects obtained in such research
serves to illustrate the importance of subtle or controlled factors in
addition to the more obvious species and dose differences.

      The experimental  concentrations, durations, and frequencies of gaseous
pollutant exposure have varied widely (see Tables 3-1 and 3-2).  The longest
concentrations studied  could occur in the ambient environment (see Chapter 2)
but no other features of the exposure doses reflected the ambient polluted
environment.  Thus, it  is not possible, at this stage, to estimate the
probability of occurrence in the ambient environment of such experimental
conditions.  The experimental durations, frequencies, and constant
concentrations used in  the artificial fumigations do not represent the real
temporal patterns in ambient polluted environments.  Also most mixture
experiments to date have not been designed to study effects of sequential
exposures.
                                       58

-------
                                          Table 3-1

                        Visible  foliar  injury  on  various  plant  species
                         in response to the joint action of S0? + 0,
Species
°:
Apple
Grape
Navy bean,
tobacco
Navy bean,
soybean
Soybean

Radish,
cucumber
Soybean
Pinto bean
Begonia
Pea
Poplar
Tobacco
Tobacco,
alfalfa

2 cone
0.8
0.8

0.8

0.15
1.0


0.15
0.15
0.25
0.3
0.13
0.05
0.03

0.1
Dose
S02 conca
0.8
0.8

1.6

0.6
1.5


0.15
0.15
0.8
1.8
1.23
0.2
0.28

1.0
L.
Duration
4 h
4 h

6 h/5 d

6 h/5 d
0.75-3 h


6 h/5 d
6 h/5 d
3 h
4 h/5d
4 h
3 h
4 h

4 h
Effect(s) obtained

antagonism
synergism, antagonism

antagonism, no effect

antagonism
synergism, antagonism,
additive

synergi sm
antagoni sm
synergism, antagonism
synergi sm
synergi sm
synergi sm
synergi sm

synergism, no effect
Reference

Shertz et al . 1980a
Shertz et al . 1980b

Jacobson and Colavito 1976

Hofstra and Ormrod 1977
Heagle and Johnston 1979


Beckerson and Hofstra 1979a
Beckerson and Hofstra 1979a
Miller and Davis 1981a
Gardner and Ormrod 1977
Olszyk and Tibbitts 1981
Karnosky 1976
Menser and Heggestad 1966

Tingey et al . 1973a
Broccol i , cabbage,
radi sh

Tomato

Eastern white
pine
Eastern white
pine
Eastern white
pine
0.1

0.03


0.1

0.05

0.025
1.0

0.28


0.1

0.05

0.05
4 h

4 h


8 h/8 w

2 h

6 h
synergism, additivity
no effect
antagonism, additivity
no effect

synergi sm

antagoni sm

synergi sm
Tingey et al . 1973a

Tingey et al . 1973a


Dochinger et al . 1970

Costonis 1973

Houston 1974
 Maximum concentrations  (ppm)  used  -"n  the research.   A range of concentration was  used  by  some
investigators.

 h=hours,  d=days,  w=weeks,  6h/5d=6  hours  per day for 5 days,  8h/8w=8  hours  per day for  8 weeks

 compared  with  effects  of  single  gases,  as  presented by the  authors.  Different methods  may have  been
used to arrive  at  the effects  statements.
                                                59

-------
                                            Table 3-2

                      Growth and y.ield of various plant species in response
                                 to the joint action of SC^ + 0,
Species
C
Radish
Radish
Tobacco
Alfalfa
Eastern
white pine
Navy bean
Snap bean
Soybean
Soybean
Soybean
Soybean
Begoni a
Popl ar
Apple
Turf grasses
)j cone.3
0.45
0.05
0.05
0.05
0.025
0.15
ambi ent
0.1
1.0
0.05
0.25
0.3
0.25
0.4
0.15
Dose
0.45
0.05
0.25
0.05
0.5
0.6
0.3
0.1
1.5
0.05
0.25
1.2
0.5
0.4
0.15
S02 conc.a
4h
8h/5d/5w
8h/5d/4w
8h/5d/12w
6h
6h/5d
6h/5d/4w
6h/133d
0.75-3h
8h/5d/3w
4h/3d/llw
4h
12h/24d
4h
6h/10d
Effect(sJ obtained1-
Duration
additive,
root most responsive
additive
root most responsive
synergistic root,
additive shoot
antagonistic on plant wt.
antagonistic on needle
length
antagonistic on plant wt.
synergistic on fruit yld.
additive on seed yield
synergistic, antagonistic
ad d i t i v e
synergistic on plant wt.
additive on pi ant wt.
synergistic on shoot wt.
antagonistic on plant wt.
synergistic on shoot gr.
synergistic or additive
Reference
Tingey and Reinert 1975
Tingey et al. 1971b
Tingey and Reinert 1975
Tingey and Reinert 1975
Houston 1974
Hofstra and Ormrod 1977
Heggestad and Bennett 1981
Heagle et al. 1974
Heagle and Johnston 1979
Tingey et al . 1973b
Reinert and Weber 1980
Gardner and Ormrod 1977
Noble and Jensen 1980
Shertz et al. 1980a
Elkiey and Ormrod 1980
Maximum concentrations  (ppm)  used  in  the  research

h=hours,  d=days,  w=weeks,  8h/5d/5w=8  hours  per  day  for  5  days  per  week  for 5  weeks.

Compared  with the effects  of  single gases,  as presented by  the authors.
                                                60

-------
      Studies concerning mixtures of NCL and SO,-, have been few.  Tingey et al.
(1971a) found that mixtures of NO,, and SO^ at concentrations _< 0.25 ppm
following one 4-hour exposure, caused visible injury in six plant species,
whereas there was no visible injury to plants exposed to N0? (2.0 ppm) or S0?
(0.5 ppm) alone for the same exposure duration.  Decormis and Luttringer (1977)
also have reported synergistic injury responses in tomato, geranium, and
petunia from exposure to a mixture of S02 (0.3 ppm) and NO- (0.5 ppm).  The
resulting injury symptoms differed from those caused by N0? or S0? alone.
Nearly all of the published visible injury responses to mixtures of N0? + S0?
have occurred at concentrations at which N0~ or SO^ alone did not injure
plants, using the same exposure duration.

Growth and Yield

      The quantitative characteristics of plant response, unlike the visible
injury responses, are measured on a continuous scale and no qualitative
distinction may be available with respect to which response is characteristic
of one pollutant or another.  Measures of growth or yield that can be
translated into economic terms are generally of primary significance.  Other
measures of growth response may be viewed as useful in explaining or
predicting characteristics of plant response that can be extrapolated to
yield. Several studies of yield responses to mixtures have been reported.  Such
studies have usually involved long-term exposures.

      While visible injury may be a useful variable in determining how plants
might be expected to respond to mixtures, the magnitude of visible injury does
not always correlate well with other responses such as plant weight and yield
changes. For example, if the visible injury response is greater than additive,
the changes in foliage weight compared with the control and the pollutants
alone may be only additive or not different at all.  This is especially true
when the magnitude of injury is extremely small  or large.  Also, close
relationships between plant growth and foliar injury are not necessarily to be
expected, because plant growth is a composite of many reactions, any of which
may be limiting (Tingey et al. 1973b).  Substantial growth reductions, with or

-------
without visible leaf injury, may be the result of more than reduction of leaf
area; they may also be attributable to decreases in the formation of plant
parts (Ashenden 1979b). There are published examples of plant growth reduction
without foliar injury (Kress et al. 1982a, 1982b) and foliar injury without
substantial growth reductions (Mandl  et al. 1975).  Also photosynthesis rates
can be reduced without vi'sible symptoms (Carlson 1979).

      Root growth has seldom been measured in gas mixture studies.  Although it
is unlikely that the pollutants penetrate rooting media and exert a direct
effect on roots (Tingey et al. 1971b), root growth may be a sensitive
indicator of the physiological status of plant shoots because reductions in
root growth may indicate diminished photosynthesis or interference with
translocation. Also, reductions in root growth in response to gas mixtures may
have a secondary effect on the whole plant as a result of decreases in the
absorption of water and nutrients.  Root growth was reduced more than top
growth in soybeans by (L + SO- mixtures, probably as a result of reduced
translocation of photosynthate to the roots (Tingey et al. 1973b).   Greater
reduction in root growth than in top growth of soybean indicated that root
growth was more sensitive than top growth (Tingey and Reinert 1975).  Radish
root (technically part hypocotyl) growth has been used to probe leaf/root
relationships.  Fresh weight of radish leaves was reduced by a smaller
percentage than was root fresh weight after the same Ck + SOp treatment,
indicating either that available photosynthate was sufficient for normal leaf
growth but insufficient for normal root growth, or that there had been a
change in partitioning (Tingey et al. 1971b).  Reinert and Gray (1981) found
the root weight of radish to be decreased by NCL, S0? and CU combinations,
even though foliage was the direct receptor of the pollutant stress.

      Reproductive ability may be changed by mixtures as a result of direct
action on generative tissues or some indirect effects.  Deleterious effects of
combined S0? + F on reproduction in Scots pine have been reported by Rogues et
al. (1980).  Mixtures also interfered with reproduction in lily; SOp + N0?, (L
+ NOp and N0? + formaldehyde markedly inhibited pollen-tube elongation (Masaru
et al. 1976).
                                       62

-------
so2 + o3

      Researchers realized some of the limitations of studies based solely on
visible foliar injury in the early 1970s and initiated studies to evaluate the
effects of S0? + CL mixtures on growth and yield of numerous plant species.
Examples of these studies are presented in Table 3-2.  Two additional reports
have presented effects of SCL and 0, alone and in mixture on the parasitism of
nematodes on soybean (Weber et al. 1979) and tomato (Shew et al. 1982).  These
reports have resulted in a more meaningful understanding of how SCL + Oo
influences plant growth and development. There are some limitations to the use
of the data.  There is little information concerning the effects of SCL and CU
mixtures at various stages of development, since usually plant growth or yield
at final harvest have been the only response variables measured.  Another
limitation is that species sensitivity determinations generally involved only
one cultivar.  There is a wide genetic variation in response to CL among
cultivars within a species.  Sensitivities of various cultivars to SCL and CK
alone and in combination have shown differential responses which suggest that
antagonism or synergism may result from differential sensitivity to each of
the pollutants alone (see Modifiers of Plant Response-Genetic Factors).
      There have been several recent reports on SCL + NCL interactions leading
to greater-than-addi ti ve growth reductions.  Two grass species were exposed to
weekly means of 0.06-0.08 ppm of N0? and S02 alone and in combination
(Ashenden 1979a).  The mixture of N02 + S02 and S02 applied alone caused
significant reduction in leaf area and plant dry weight of Dactylis glomerata
L. and Poa pratensi s L.   There were also reductions in the number of tillers
and leaves of both species exposed to NO- + SCL but senescence was not
enhanced (Ashenden 1979b;.  Similar effects were found on Lol i urn mul ti f 1 orum
Lam.  and Phleum pratense L.  (Ashen.!--, and Mansfield 1978; Ashenden and
Williams 1980;  Mansfield and Ashenden 1978). It was concluded that N02 and S02
in combination  were more toxic to the grass species than was predicted by
summing their individual effects on growth.
                                       63

-------
      Irving et al.  (1982) dispensed SOp and NCL through a system of aluminum
pipes suspended over a canopy of field-grown soybeans.  The soybeans were
fumigated on ten occasions, with mean concentrations of S0? during fumigation
ranging from 0.13-0.42 ppm, while N02 ranged from 0.06-0.40 ppm.  Results from
the 2-year study showed that exposures to N02 alone had no effect on soybean
seed yields.  Exposure to S02 alone decreased yield by 6% the second year.
Mixtures of SOp and  N02 in both years of the study resulted in yield decreases
ranging from 9-25%.   These losses, however, were obtained in the presence of
7-hour average ambient 0, levels ranging from 0.006-0.095 ppm.  Thus, 03 may
have caused some of the NOp x S02 interaction on soybean yield.  Amundson and
Weinstein (1980) exposed soybean to S02 (0.0, 0.1, and 0.3 ppm) and N02 (0.1
ppm), singly, and in combination for 4 hours daily for a period of 14 days
during pod fill in open top field chambers.  Sulfur dioxide (0.3 ppm) + N02
(0.1 ppm) caused early senesence and reduced yield, compared with plants
exposed to S02 (0.3  ppm) alone.  Nitrogen dioxide (0.1 ppm) in the presence of
SOp (0.1 ppm) had no effect on soybean yield. Klarer (1982) exposed soybean in
the vegetative growth stage to single and mixed N02 (0.0, 0.1. and 0.2 ppm)
and S02 (0.0, 0.2 and 0.3 ppm) 15 times, every other day for three hours per
day.  Leaf, stem, and root dry weights were'significantly less (18, 12, and
32%, respectively) when the four mixture treatments were averaged together and
compared with the mean of the four treatments of N02 and S02 alone.  The
response of soybean  to N02 and S02 was nearly linear.

      Nitrogen dioxide and SOp mixture studies have not been limited to
herbaceous plants.  A 37% reduction in the growth of poplar was reported
following continuous 8-week exposure to 0.06 ppm SOp + 0.06 ppm NOpj no growth
reductions were observed following exposure to S02 or NOp alone (Whitmore et
al. 1982).

N02 + 03

      The effects of N02 + 03 generally have not been studied.  However, in
recent reports of two- and three- pollutant mixture treatment comparisons, the
N02 + 03 treatment has been included. Kress and Skelly (1982) have studied the
response of seven tree species to N02 (0.1 ppm) and 03 (0.1 ppm) alone and in
mixture for six hours per day, for 28 consecutive days.  Virginia pine and

                                      64

-------
loblolly pine height were significantly suppressed by NCL + CL treatment.
There were significantly less than additive suppressions of sweetgum root and
total dry weight and white ash root weight.
      Only a few studies of three-gas interactions have been conducted to date
and response patterns are complex.  In some cases the experimental design has
limited the interpretation of the research.  For example, Elkiey and Ormrod
(1980) exposed 18 turfgrass cultivars, representing six species, to N0?, S0?,
and CL singly at 0.15 ppm of each pollutant and to a mixture of all three
pollutants (0.15 ppm of each gas).  The three-pollutant mixture treatment
caused more leaf injury and greater reduction in leaf area in most cultivars
when compared with the additive effects of the pollutants singly.  The lack of
two-pollutant mixture treatments was a limitation to the full interpretation
of this research.  Similarly, Yamazoe and Mayumi (1977) exposed sweet corn and
rice for one 24 hour period to N0?  (0.5 ppm) + 0^ (0.15 ppm), alone and in
mixture, and to a mixture of N02  (0.5 ppm) + S0? (0.15 ppm) + CL (O.lSppm).
The mixture of N02 + CL caused additive injury in rice but not in sweet corn,
while injury in both sweet corn and rice from the N02 + S02 + 0^ suggested an
antagonism by SO^.  The lack of certain treatment combinations prevented full
interpretation of this research.

      The N0? + S0? dose/response relationship was studied with snap bean
exposed to N02 (0.0, 0.05, and 0.1 ppm) + S02 (0.0, 0.1 and 0.15 ppm), in the
presence of 0.05 ppm 0-, under greenhouse exposure conditions (Reinert and Heck
1982).  Nitrogen dioxide at 0.1 ppm in the presence of 0.05 ppm CL and the
absence of SCL, caused a 10% loss of snap bean fruit fresh weight.  When S02
was held constant at 0.1 ppm, there was a 15% and 11% weight loss in bean
fruit as N0? was increased to 0.05 or to 0.1 ppm, respectively.  These data
suggest a significant effect of N0? in the presence of S0? and CL at ambient
air concentrations of all  three pollutants.

      Kress et al. (1982a) found substantial growth suppression in loblolly
pine by using mixtures of CL, N0~, and S02 at concentrations below the
national ambient air quality standards for each pollutant singly.  They also

                                       65

-------
found greater growth suppression of American  sycamore with (L  + S02 + N02 than
with two-gas mixtures,  without foliar injury  (Kress et a "I. 1982b).

      The use of full  factorial  designs to evaluate 3-way  mixtures  by Reinert
and colleagues (Reinert and  Gray 1981,  Reinert and Sanders 1982,  Sanders and
Reinert 1982a, 1982b)  has  led  to the emergence of several  concepts.  Nitrogen
dioxide in mixture with SO-  and/or CU has significantly reduced growth and
yield in crop plants.   In  nearly every  instance,  the three-pollutant mixture,
N09 + S0? + Oo, caused  more  loss in weight and yield than  the  single gases or
  IL     L.    O
two-pollutant mixtures.  If  the plant is capable  of developing repair
mechanisms against stress  by 0,, it appears that  under the simultaneous stress
of NCL, S0?, and 03, repair  mechanisms  may not be able to  function, and plant
productivity is reduced.

Other Mixtures

      Mixtures of 0.8 ppm  SO-  + 2.5 to  13 ppb HF  for 23 days decreased linear
growth and leaf area of orange in the research of Matsushima and Brewer
(1972).  Mandl et a).  (1975)  found no significant effects  of 0.15 or 0.3 ppm
S0? + 0.6 to 0.9 ppb HF for  7  days on growth  of several species even though
there were visible symptoms  of injury.   Gas-precipitation  interactions have
been studied concerning acidity of rain and the occurrence of  SO- or
photochemical oxidants.  In  soybeans, Irving  and  Miller (1981) found 0.19 or
0.79 ppm SO- in 17 or 24 4-hour exposures to  have deleterious  effects that
were not affected by simulated acid precipitation of pH 3.1 every 5 days.
Troiano et al. (1962)  used open-top chambers  to study the  interaction of
ambient 0^ and simulated acid  rain of pH 4.0, 3.4, or 2.8  on soybean seed
quality.

Physiological and Metabolic  Responses

      Exposure to pollutant  mixtures may result in physiological  and metabolic
responses which in turn result in growth and  yield reduction.   The  means by
which these responses  occur  are not well known.  One predominant mechanism is
the physiological or metabolic alteration induced by one pollutant  which then
increases or decreases  the susceptibility of  the  plant to  another pollutant.
                                      66

-------
This would appear to be a most likely means of explaining the effects of two
successive exposures to different pollutants (Hofstra and Beckerson 1981,
Masaru et al.  1976).  A change in sensitivity would be the most likely
explanation for some observations of the joint action of pollutants on foliage
(Miller and Davis 1981a) where the effects on uptake were not similar or where
new kinds of  symptoms occurred (Mandl et al. 1980, Lewis and Brennan 1978).

      Little  is understood concerning the metabolic and physiological  action
of pollutants  in mixture.  Many experiments have established dependent or
independent relationships among pollutants through statistical
interpretations.  However, further physiological and/or biochemical mechanisms
associated with photosynthate production, growth regulation, water relations,
changes in metabolic pathways, metabolite and nutrient allocation, enzyme
function, and  those processes associated with energy production and
utilization at the cellular level have not been studied thoroughly.

      Studies  of enzyme activity have indicated that, in some cases, the
threshold concentration of a pollutant required to produce a change in enzyme
activity was  lowered when combinations of pollutants were used (Horsman and
Wellburn 1975).  For combinations of 0.1 or 1.0 ppm N02 and 0.2,  1.0,  1.5,  or
2.0 ppm S0? synergistic interactions were found in pea seedlings  in terms of
increased activities of peroxidase, glutamate-pyruvate transaminase (GPT),  and
glutamate oxaloacetate transaminase (GOT), and decreases in chlorophyll
content and ribulose 1,5-diphosphate carboxylase (RuDPC) activity.  These
changes occurred in the absence of visible injury. The inhibition of RuDPC  may
be due to an  accumulation of sulfite ions in the chloroplasts.  The increases
of GPT and GOT activity indicate a disturbance in amino acid metabolism.   If
these types of metabolic activities continued, the plant will eventually
develop some  visible evidence of injury.  However, these observations  do not
necessarily explain additive or synergistic effects of mixtures.   Examination
of numerous other enzyme responses also indicated different responses  to the
combination gas than to single gas exposures (Wellburn et al. 1976).
                                      67

-------
      Nitrite reductase  (NiR)  activity due to N02,  S02,  or N02 + S02 has been
investigated (Wellburn  1982,  Wellburn et al.  1981).   Sulfur dioxide had little
effect on NiR,  but  NCL  increased  its  activity.   The  mixture of S02 and N02
reduced NiR activity.   Thus,  Wellburn (1982)  proposed that the presence of S02
prevents the induction  of  additional  NiR by N02 which would normally lead to
ammonia and amino acid  synthesis.

      Studies of metabolic function  and biochemical  changes in plants do not
presently provide any definite evidence of a  specific site of action for SCL +
N0? + CU synergism.   It  is possible  that (as  shown  by changes in permeability,
Elkiey and Ormrod 1979b, Beckerson  and Hofstra  1980), 03 or S02 could alter
the permeability of  cellular  membranes to other dissolved pollutants or their
reaction products in the aqueous  phase.   The  findings of Wellburn and others
(Horsman and Wellburn 1975,  1976, Wellburn 1982, Wellburn et al. 1976, 1981)
concerning the interactions  of N0?  and SO^ may  support the concept that
membrane integrity is damaged  rapidly by the  inability of the plant to
detoxify N0? in the  presence  of S0?.   With increasing S0?, sulfite accumulates
(Malhotra and Hocking 1976)  and the  dual impact of  sulfite and nitrite on
membrane integrity in the  presence  of 0-, may  allow  CU to enter the cell freely
                                       O            -J
on a continuing basis.   Ozone could  then cause  injury while S0~ and NO- were
impairing specific enzyme  functions  leading to  energy utilization, as well as
impairing transport  and  allocation  of needed  cell  repair components to
counteract the stress from further  exposure to  all  three pollutants.

      Studies of the influence of mixtures on the  chemical composition of
plants have been largely concerned  with chlorophyll  concentration. Horsman and
Wellburn (1975) found decreased chlorophyll with S02 + NO- even though N0?
alone increased chlorophyll.  Olszyk  and Tibbitts (1981)  established that
near-threshold  injury by 0^  +  SO- on  peas could be  evaluated by chlorophyll
concentrations  of expanded leaves,  as well as by surface area of expanding
leaves.

      Research  concerning  mixtures  has also involved an  examination of gas
exchange between plants  and  polluted  atmospheres.   Black (1982) suggested that

-------
a pollutant-induced change in stomatal aperture and function would result  in
important consequences by altering a) photosynthetic CCL uptake and
transpirational water loss, b) the rate at which the pollutant enters the
plant and arrives at the metabolic sites, and c) metabolism, resulting in
growth and yield change.  Stomata may be induced to either open or close in
response to SCL depending on the species, SO- concentration, duration of
exposure, and environmental conditions at the time of exposure (Black 1982).
There are reported instances of S0? enhanced stomatal opening in 16 plant
species and S0? enhanced closure or depressed transpiration in 24 plant
species.  In fact, enhanced opening and closure have been reported to occur
within the same species (Black 1982).

      This realization that SCL-induced effects on physiological processes in
plants are exceedingly variable may offer an explanation for many reported
incidences of synergistic and antagonistic responses of plants to pollutant
mixtures.  When more visible injury develops from a mixture of SO- + 0, than
from either pollutant alone, some investigators have reasoned that SCL in
mixture decreases stomatal resistance, allowing more CU to enter.  Beckerson
and Hofstra (1979a, 1979b) have tried to develop an experimental basis for
understanding antagonism and synergism of visible injury response and stomatal
resistance by using species such as navy bean and soybean, which respond
antagonistically, and cucumber and radish, which respond synergistically to
SCL + CL.  White beans were exposed to 0.15 ppm SO- and 0.15 0- singly and in
combination six hours per day for five days (Beckerson and Hofstra 1979a). The
SO- + 0- mixture increased stomatal resistance  (depressed stomatal opening)
about 30% more than 0- alone during the first th^ee days of exposure and the
amount of injury was 50 times less than for 0,  alone (Beckersor and Hofstra
1979a).  Sulfur dioxide alone decreased stomatal resistance (enhanced stomatal
opening).  Thus, it was concluded that any protective action of SO- against 0-
injury (antagonism) was not completely explained on the basis of stomatal
response of white bean.

      Investigations were continued using radish, cucumber, and soybean.   In
all three species, SO,, decreased stomatal resistance, 0- increased resistance,
                                       69

-------
and the S00 + 0,  mixture increased resistance much more than Oo alone
(Beckerson and Hofstra 1979b).   Hofstra and Beckerson (1981) also found  that
SCL pre-treatment of white bean and cucumber did not prevent increased
stomatal  resistance following exposure to SO,, + 0^ and 03 alone, at
concentrations used in previous studies (Beckerson and Hofstra 1979a).   They
also reported that the presence of SO,, in mixture with 03 reduced the
CU-induced increase in membrane permeability in white bean  and soybean,  but
not in radish and cucumber (Beckerson and Hofstra 1980).  This suggested
differential modes of action of S0? + Oo alone or in mixture, acting at
                                  (_    O
membrane sites within the plant.  In summary, Beckerson and Hofstra concluded
that stomatal function can be disregarded as having any major influence  on S0?
+ Oo interactions and as a mechanism for understanding synergistic, additive,
or antagonistic plant responses to SO.-, + On, at least for the species studied.
They further proposed that in the case of synergism in radish and cucumber,
S0? may have altered the nutritional or enzymatic status of the plant, thus
increasing sensitivity to 0., (Hofstra and Beckerson 1981).  Nevertheless,
further research  is warranted on stomatal responses to SO,, + 0,, utilizing
techniques having greater resolution and more statistical strength.  In
addition, stomatal responses need to be related to visible  injury and to air
pollutant flux into leaves.

      Stomate function also  has been studied in attempts to explain the
additive and frequently greater than additive response of plants to N0? + S0?.
Arnundson and Weinstein (1981) found high leaf resistance in soybean plants
exposed to SO,, (2.0 ppm) + NO,,  (0.5 ppm), partially accounting for
antagonistic effects of the  two gases.  Ashenden (1979a) found that N0?  (0.1
ppm) and SO,-, (0.1 ppm) alone caused short-term increases in bean leaf
transpiration rates, while N0n  + SO,, in mixture decreased transpiration.
Stomate function  was apparently not involved in the synergistic foliar injury
response resulting from mixtures of NO- + SOp.   However, Ashenden gave a
possible explanation for the synergistic foliar injury response to mixtures of
N02 + SO,, by proposing that the stomata were closing in response to
physiological injury within  the leaf.   Since N00 and S02 interfere with
respiration, there is a possibility that increased C02 concentrations could
                                      70

-------
arise within the leaf and subsequently decrease stomatal opening  (Ashenden
1979a).  Amundson et al.  (1982) found a HF  (0.6 and 1.9 ppb)  and  N02  (0.6  and
1.2 ppm) interaction, with less HF injury on sweet corn in the presence of  the
higher of the two N0? concentrations.  Leaf resistance was higher  in  the
combined treatments. There was greater stomatal closure of snap beans with  H?S
+ 03 than with H2S alone  (Coyne and Bingham 1978).  Williams  et al.  (1971),  in
evaluating the interaction between SO^ and particulates at naturally  polluted
sites, noted that particles accumulated in stomatal pores, probably keeping
them open, increasing permeability, and admitting more S0?.   When  pollutants
are sorbed or deposited on foliar surfaces and then penetrate through the
cuticle, reciprocal effects on uptake are unknown.  Thus, interactions of
pollutants might be explained readily, in some cases but not  others,  on the
basis of direct effects on stomatal opening.  However, there  is considerable
evidence that other factors in the plant, in addition to stomata,  have an
important role in determining responses to pollutants.

      Impairment of photosynthesis by mixtures has been studied by several
investigators.  Carlson (1979) found that photosynthesis rates of  sugar maple
and white ash leaves exposed to CL (0.5 ppm) + SO- (0.5 ppm)  decreased more
than additively. The reduction in photosynthesis was least when irradiance was
optimal.  Ormrod et al . (1981) found joint action of concentrations of 0~
(0-0.25 ppm) + SO,, (0.04  ppm) in suppressing net photosynthesis in broad bean,
but there was recovery if visible leaf injury did not occur (Black et al.
1982). Similarly, Le Sueur-Brymer and Ormrod (1983) found mixtures of 0., (.067
pprn) + SOp (0.3 ppm) suppressed net photosynthesis of fruiting soybeans when
single gases did not. However, the photosynthesis apparatus apparently adapted
readily to the stress because net photosynthesis was no longer decreased by
the mixture after two successive six-hour daily exposures.  Substantial
reduction in photosynthesis of peas by S0?  (0.0 to 0.25 ppm)  + NO- (0.0 to
0.25 ppm) was noted by Bull and Mansfield (1974).  White et al. (1974) found
greater than additive effects of S0? (0.15-0.35 ppm) + NO- (0.1-0.2 ppm) on
net photosynthesis of alfalfa.  Net photosynthesis rates of sunflower leaves
were depressed by all combinations of 0,, (0.2 ppm), SO- (0.2  ppm), and NO-
(1.0 ppm) (Furukawa and Totsuka 1979).  At a higher than ambient CO-
                                       71

-------
concentration (645 ppm), the inhibition of net photosynthesis by S02  ("0.8
ppm) + N0? ("0.3 ppm) was less than half that at an ambient concentration  (315
ppm) (Hou et al. 1977).  Addition of 0.072 ppm 03 to 0.74, 3.25, or 5.03 ppm
H9S resulted in greater reductions in net photosynthesis rates of snap beans
 L-
than those caused by H^S alone (Coyne and Bingham 1978).

      While many anatomical  studies of tissue and cell injury have been
conducted with single pollutant gases, very little has been done with
mixtures.  Solberg and Adams (1956) found no differences in microscopic injury
of apple leaves between single and mixed S02 (0.5 ppm) + HF (5 ppb) for 4
hours per day for 2 days.  However, Evans and Miller (1975) found different
sites of injury for single or mixed 0., (0.45 ppm) + SO- (0,45 ppm) for 9 hours
in pine needles.  Krause and Jensen (1978) found inclusions in poplar leaf
cells exposed to 03 (0.15 ppm) + SO^ (0.25 ppm) for 12 hours per day for 21
days that were not in leaves exposed to single gases.   Leaf surfaces were
injured by the mixture but single gases had no effect (Krause and Jensen
1979).

      Recovery from mixture-induced injury to plant processes to normal levels
of functional activity has been reported.  Net photosynthesis, impaired by 0.,
(0.05-0.30 ppm) + SO- (0.04  ppm) for 4 hours recovered if there was no visible
injury to broad bean leaves  (Black et al. 1982).  Bennett et al. (1980) found
recovery of bean growth and  yield following 0~ (0.046  to 0.127 ppm) + HLS  (0,3
to 7 ppm) exposure.  Kress et al. (1982b) reported recovery in growth of
American sycamore after injury by 0^ (0.05 ppm) + SO-  (0.14 ppm) + N0? (0.10
ppm) for 6 hours per day for 28 days.   The mechanisms  of recovery from
pollutant injury have been described by Tingey and Taylor (1982).  The extent
and speed of re-establishment of a normal metabolic state following pollutant
mixture injury will be an important determinant of economic loss from exposure
to mixtures, but economic losses have not been estimated on these bases.

      A possible explanation of apparently diminished  injury from 0. + S0?
mixtures was the chemical combination of these gases in exposure chambers
(Costonis 1973).  However, no evidence was obtained for reactions that would
                                      72

-------
lower the concentration of either gas (Jacobson and Colavito 1976).  A direct
reaction between CL and S0? may occur in plant tissue and the direct aqueous
oxidation of SCL by 03 has been described (Heagle and Johnston 1979).

Accumulation and Uptake

      Accumulation may be considered a major concern for several reasons.
First, with reference to HF and certain airborne particulate compounds,
accumulation of a toxicant by the plant constitutes a potential hazard to
consumers of the plant.  Second, uptake and accumulation of a pollutant can be
viewed as the first links in a chain of events leading to some altered state
or process that may be detrimental.  Third, tissue levels of pollutants are
often used as diagnostic measures to assess exposures.  As discussed by McCune
(1983), many factors influence the effect of S0?, NCL, or 0, on the
accumulation of fluoride from HF.  One of the major difficulties in the
interpretation of effects arises when the levels of exposure are too few to
determine whether accumulation of F is linear with exposure.  If the plant is
exposed successively while it is growing, exposures during the latter periods
should be weighted more heavily than those during the earlier periods to
compensate for growth dilution.  If one pollutant affected growth during a
series of exposures to HF, wherein the concentration of HF varied, the result
would be an apparent effect on F-uptake itself. Another aspect of this problem
appears when pollutant mixtures affect the uptake of minerals from the rooting
medium and their distribution within the plant.  Increased cadmium
concentration in young leaves of cress during 0, exposure was noted by Czuba
and Ormrod (1981).  Immediately after ozone exposure, there was stimulated
uptake of cadmium and redistribution of cadmium between the leaves and stem.

      The actual uptake of pollutant gases, rather than the concentration or
dose, would be expected to relate most closely to biological responses.  There
is little information available on pollutant uptake rates from gaseous
mixtures.  The amount of pollutant sorbed by plants is the product of the
uptake rate or flux and the duration of exposure. Also, sorption includes both
absorption through stomata or cuticle into the mesophyll tissue of leaves,
                                       73

-------
while it may be metabolically active, and adsorption on surfaces which may
injure the surface materials, but not penetrate into areas of metabolic and
physiological significance.   Elkiey et al.  (1982)  exposed shoots of ten shade
tree species to 0^ (0.25 ppm) + S02 (0.4 ppm) + N09 (0.4 ppm) for 6 hours and
measured individual  uptake rates.  Sorption from the mixture was consistently
less than from single gases  in species that did not close stomata at night,
while single and mixed gases had similar sorption  if stomata closed at night.
There was less absorption of each gas from the 3-gas mixture than from single
gases by Kentucky bluegrass  plants (Elkiey and Ormrod 1981a).

      In this study,  plants  of nine Kentucky bluegrass cultivars were exposed
to 00, SOp, NOp, or a mixture of the three gases for three days to determine
absorption and adsorption rates of each gas.  Absorption rates into stomata
differed among cultivars and generally decreased with longer exposure.  Leaves
of insensitive cultivars generally absorbed less of the single gases than the
leaves of sensitive cultivars.  Adsorption  rates on leaf surfaces, determined
with stomata closed,  were substantial and varied with pollutant gas and
cultivar.  Absorption of 03  (0.25 ppm), S02 (0.4 ppm)  and N02 (0.4 ppm) by
petunia plants from single gases was generally greater than from the mixed
gases (Elkiey and Ormrod 1981b). Absorption rates  tended to decrease gradually
throughout the day and from day to day with continuous exposure.  Accumulation
of tissue sulfur and  nitrogen in petunia plants did not agree well with uptake
rates.  Tissue analysis of petunia plants exposed  to 0., + SOp + NOp mixtures
indicated less accumulation  of sulfur from mixed than  single SOp, and less
total nitrogen in plants exposed to any N0? compared with those not exposed
(Elkiey and Ormrod 1981c).  This suggests that sulfur- and nitrogen-containing
volatiles may be released by exposed plants or that nutrient uptake and
distribution or re-distribution may be affected by the pollutant treatments.

      There is considerable  information available  on pollution uptake, based
upon studies of single pollutants.  Those studies  provide conceptual models
which should be useful in considering the uptake of pollutant mixtures by
plants.   The collection and  interpretation  of uptake data should help resolve
                                       74

-------
the conflicting results of experiments conducted in different environments
with differing exposure regimes.

MODIFIERS OF PLANT RESPONSE

Genetic Factors

      Resistances to pollutant mixtures have genetic and environmental
components.  The most useful comparisons of species sensitivities to mixtures
are those based on studies in which more than one species was exposed at the
same time in the same facilities under the same environmental conditions.
There have been several reports of research of this kind (Table 3-3)
indicating large differences in species sensitivity to various mixtures.

      Most investigators have used one cultivar or line to represent a
species, even though there is ample evidence, from single-gas and mixed-gas
studies, that there can be wide variation in cultivar response within species
(Ormrod 1978).  There have been several studies reported in which cultivar
responses to mixtures of SOp + 0-, were directly compared at the same time
(Table 3-4).  In forest species, Kress et al. (1982a) found differential
sensitivity to 0, + N0? + SO- among lines of loblolly pine.

      The degree of association of sensitivity to S0? and 0^ among cultivars
within species has been examined by means of a distribution-free measure of
association, Kendall's Tau statistic.  Its value was 0.455 for 17 cultivars of
Kentucky bluegrass (Murray et al. 1979); 0.2018 for 19 cultivars of soybean
(Miller et al. 1974); and 0.1346 for 33 cultivars of bean (Beckerson et al.
1979).  However, sensitivities to a mixture of S0? (0.15 ppm) + 0^ (0.15 ppm)
for 6 hours per day for 5 days in the bean cultivar study were associated with
both S0? and 0- sensitivity (tau - 0.3570 and 0.4227, respectively).  Karnosky
(1976) found some association of S0? and 0~ sensitivity in five lines of
trembling aspen.  In Picea abies, resistance to HF and to SO,-, appeared to be
positively correlated (Halbwachs and Kronberger 1979).  When the joint
distribution of sensitivities has a positive correlation, a population of
                                       75

-------
                                             Table 3-3
                      Direct comparisons of species sensitivity to SCL + 0
                                 SO, + NO,, 0-, + NO,,  and HF + SO,
                                            J3
M-xture
Species compared
                                                                        Reference
so2 + o3
S0
J3

HF
     SO,
Tobacco, radish, alfalfa, cabbage, broccoli,
tomato, onion, bromegrass, spinach
Tobacco, alfalfa, radish
Navy bean, tobacco
Radish, cucumber, soybean
Navy bean, soybean,  cucumber, radish
Sugar maple,  black oak, and white ash
Ginseng, radish, tobacco
Lettuce, radish, pea

Tomato, radish, oats, tobacco, pinto bean,
soybean
Several native desert species

Radish, swiss chard, oats, peas,  orchard gras:
annual  ryegrass, timothy, perennial  ryegrass,
Orchard grass, perennial ryegrass

Forest  tree species

Sweet orange,  mandarin
Bean, barley,  sweet  corn
                                                                        Tingey  et  al.  1973a
                                                                        Tingey  and Reinert 1975
                                                                        Jacobson  and  Colavito 1976
                                                                        Beckerson  and  Hofstra 1979a
                                                                        Beckerson  and  Hofstra 1980
                                                                        Carlson 1979
                                                                        Proctor and Ormrod 1981
                                                                        Ormrod  et  al.  1983b
                                                                        Tingey  et  al.  1971a
                                                                        Hi 11  et  al.  1974
                                                                        Thompson et  al. 1980

                                                                        Ashenden and Mansfield  1978
                                                                        Ashenden 1979b

                                                                        Kress and  Skelly  1982

                                                                        Matsushima and Brewer 1972
                                                                        Mandl et al. 1975
                                                 76

-------
                               Table 3-4

 Direct comparisons of cultivar  sensitivity within  species  to  S0? +  CU
Species
Number of cultivars compared
Reference
Tobacco
Tobacco
Tobacco
Soybean
Navy bean
Strawberry
Petuni a
Bean
Soybean
Begoni a
              3
              9
              3
              2
              2
              6
              3
             33
              2
              5
Menser and Heggestad 1966
Menser and Hodges 1970
Tingey et al.  1973a
Tingey et al.  1973b
Jacobson and  Colavito 1976
Rajput et al.  1977
Elkiey and Ormrod 1979c
Beckerson et  al.  1979
Heagle and Johnston 1979
Reinert and Nelson 1980
                                    77

-------
plants should be less affected by a mixture than would be expected should the
distribution of sensitivities to each pollutant suggest an assumption of an
independent distribution.

      While most investigators have not reported the extent of plant-to-plant
variation in sensitivity to mixtures, Skelly et al.  (1972) reported high
variability among eastern  white pine trees in response to ambient SCL + NCL.
Although it might be expected that an exposure to SO- would affect the mean
tolerance of plants to (L  (Jacobson and Colavito 1976, Macdowall and Cole
1971), the latter investigation showed an effect of SCL on dispersion of
tolerance as indicated by  changes in slopes of probit regressions.

      The use of highly sensitive species or cultivars as test plants to
indicate the presence of single pollutants is widely practiced, but only two
recommendations of plant indicators of mixtures have been made.  Menser and
Hodges (1970) suggested 'Burley 49'  tobacco and Grosso et al.  (1971) suggested
Nicoti ana glutinosa for detection of CL + S0? effects.

Phenological Factors

      Plant development stage may also be an important determinant of
sensitivity to pollutant mixtures.  Few studies of mixture effects have
considered the impact of leaf age or growth stage on sensitivity to
pollutants, since usually  plant yield and biomass at final harvest have been
the only response variables measured.  Menser and Heggestad (1966) noted that
older tobacco leaves were  more sensitive to CL ("0.03 ppm) + SCL ("0.26 ppm)
for 2 or 4 hours than younger leaves.  The midshoot  leaves of grape and apple
were most sensitive to 03  (0.4 ppm)  + S02 (0.8 ppm)  for 4 hours (Shertz et al.
1980a, 1980b).  Alteration of the susceptibility of  leaves by the stage of
development is another area in which distributional  aspects of the combined
effects  of pollutants have not been considered.  One obvious example is where
younger  leaves tend to be  susceptible to one pollutant and older leaves to
another.  If the exposures  are consecutive or separated in time for plants with
determinate growth, the sequence of exposures will determine the effects.
                                      78

-------
Environmental Factors

      There may be a strong environmental component in determining plant
sensitivity to mixtures, as well as in modifying genetically determined
sensitivity.  Irradiance, temperature, water supply, and other environmental
factors are known to affect plant responses to air pollutants and have been
studied extensively for single gases  (Ormrod 1978).  However, studies of
environmental effects on mixture responses have been more limited.  Carlson
(1979) found much more visible injury and growth suppression of sugar maple
and white ash by 0^ + S02 at high than at low irradiance.  Miller and Davis
(1981b) found that exposure temperatures affected the type of CL + S0? visible
injury symptoms in beans, as well as  the amount and location of injury.

      The water status in each component of the soil-plant-atmosphere
continuum may alter response to mixtures as there may be an interaction with
CO- concentration.  More injury was noted after exposure of eastern white pine
to 0, + S02 during periods of high humidity compared with low humidity (Jaeger
and Banfield 1970).  Humidity was found to have a marked effect on the
diffusive resistance response of petunia leaves to CL + SCL.  Exposure of
petunia to SCL (0.8 ppm) + 0^ (0.4 ppm), for 4 hours at 50% relative humidity,
caused an increase in stomatal resistance, regardless of cultivar sensitivity
to (L.  However, at 90% relative humidity, there was an increase in stomatal
resistance only in the 0, sensitive cultivar, 'White Cascade' (Elkiey and
Ormrod 1979a).  Although changes in relative humidity, leaf water potential
(Elkiey and Ormrod 1979a), and membrane permeability (Elkiey and Ormrod 1979b)
may be demonstrated among cultivars of differing 0, sensitivity, these
phenomena did not completely explain differences in petunia cultivar
sensitivity to SCL and CL alone and in mixture.

      Plant water status not only affects the responses to mixtures but also
is affected by plant exposure to mixtures.  Plant water potential decreased
quickly on exposure of petunia to CL + SO- (Elkiey and Ormrod 1979c).  Leaf
diffusive resistance changes are an indication of altered stomatal action.
                                       79

-------
Adequate soil  water,  optimal  mineral  nutrition, high relative humidity, and
sufficient irradiance will  lead to full  stomatal opening, subjecting the leaf
tissue to maximal  pollutant entry initially (Carlson 1979).

      The C0?  research of Hou et al.  (1977) was based on the recognition that
C0?, as well  as S02 and NOX,  is included in the exhaust gases of many
industries.  The ratio of SCLrNC^rCOo concentrations occurring downwind from a
power plant burning coal  (1:0.33:326) was used in controlled exposure studies.
Doubling the  CCL concentration increased net photosynthesis of alfalfa in S02
+ N0?, even though this mixture decreased net photosynthesis in ambient CCL.

      Another  factor  that could vary  and exert an effect in the field is
mineral nutrition.  Elkiey  and Ormrod (1981a)  exposed turfgrass plants growing
at different  nitrogen and sulfur nutrition levels to (L (0.1 ppm,  6 hours per
day) + S0? (0.15 ppm  continuously)  +  NOo (0.15 ppm continuously)  for 10 days.
Low sulfur or  low  nitrogen  usually enhanced the effect of S0? or  N0?,
respectively.   Misting with deionized water increased severity of  visible
injury.  In the compilation of effects  of mineral nutrition and responses of
plants to pollutants  by Cowling and Koziol (1982), one can see that the
nutrient-determined tolerances of plants are positively or negatively
correlated, depending upon  the crop,  nutrient  element, and set of  pollutants.
For example,  in tomato, tolerance to  HF  decreased but tolerance to 0,
increased with P-deficiency;  in tobacco, tolerance to 03 or to S02 increased
and then decreased as the supply of N increased; in barley, tolerance to HF
decreased with deficiencies of P, K,  or  Ca and tolerance to S02 decreased with
Ca- or K-deficiency,  but  increased with  P-deficiency.  An important part of
this review was the attempt mechanistically to reconcile and explain the
effects of nutrients.  Some could be  attributed to increased or decreased
uptake of pollutants  and  others to effects on  the inherent, metabolic
susceptibility.  Whether  the  presence of one pollutant alters the
nutrient-determined response  of a plant  to another pollutant is unknown.  To
the extent that the effects of other  environmental factors can be  similarly
partitioned,  one could also predict their likely effects on the joint action
                                       80

-------
of pollutants.  For example, the effects of temperature were not concordant
with effects on gaseous exchange (Miller and Davis 1981b).
      The only report of chemical protection against mixture injury indicated
that benomyl protected pinto beans from the two oxidants, CL + PAN  (Pell
1976).

SUMMARY

      The available literature indicates that most pollutant mixture research
has been confined to the use of various combinations of (L, S0?, and N0? on
major species.  Other combinations of gases have had much less attention and
almost no research on interactions of gaseous, aerosol, and precipitation
pollution has been reported.  Many species of considerable economic or
ecological importance have had little or no attention. Visible injury has been
the most frequently reported response variable but concern for growth and
yield effects has increased in recent years.  The visible injury may not
adequately reflect growth and yield responses.  Discovery of additive and
synergistic responses to two and three pollutant mixtures has provided the
impetus for further study and for concern about impacts of mixtures.

      Suitable experimental methods and statistical analyses are available for
effective studies of mixtures but some of the most appropriate experimental
designs and analyses are not widely used.  Experiments have been conducted in
controlled environments, chambers, greenhouses, and field facilities;  each of
these approaches has strong and weak points, and the effectiveness will depend
on the purpose of each experiment.  Some disparities in results between
experimental approaches have been reported.  Diversity and ambiguity of
terminology have created some difficulty in interpretation, but clear
definitions of joint action, additivity, synergism, and antagonism are
avai1ab1e.

      Exposure of plants to pollutant mixtures affects visible injury
development, growth, yield, physiological processes, biochemical activities,
                                       81

-------
and plant anatomy with responses  often differing from those due to the single
components of the mixture.   Photosynthesis,  transpiration,  pollutant uptake
rates,  enzyme activities,  stomatal  function,  tissue elemental  concentration,
and other response variables  may  be differentially affected by mixtures
compared with single  gases.   Much of the information on physiology and
biochemistry is  fragmentary  and,  while stomate function and pollutant uptake
have been most extensively  studied, the data  are difficult  to  interpret as
these responses  are extremely sensitive to environmental changes.

      Species differ  widely  in sensitivity to pollutant mixtures and there may
be large cultivar differences within species, as well as plant-to-plant
variation in response.   Plant and leaf development stages may  be important
determinants of  sensitivity  to mixtures but  there have been few studies of
this concept.  Environmental  factors,  including irradiance, CCL concentration,
temperature, water status,  and nutrition may  affect mixture responses.  Little
is known of the  effects of  pollutant mixtures on plant hardiness,  reproductive
processes, competitive ability and  interactions in ecosystems.   It is clear,
from the few studies  conducted in which appropriate follow-up  measurements
were made, that  plants can  recover  from mixture stresses and even  adapt to
them.  The relationship of  concentration and  duration in determining doses of
mixtures has had little attention nor  have the flux rates of pollutants from
mixtures to plants, even though actual  uptake is likely to  be  an important
determinant of injurious effects.  There has  been little study  of  the effects
on plants as a result of changing pollutant mixture composition and
concentration patterns that  may occur  in nature.
                                      82

-------
                              4.   RESEARCH NEEDS

AMBIENT AIR QUALITY EXPOSURE

      Based upon a review of the  scientific presentations and their knowledge
of the scientific literature,  the panel  members determined that  a series  of
technical  gaps exists in defining the effects of air pollutant mixtures  on
vegetation.

      The  panel  concluded that the ambient air quality air pollution  review
described  in Chapter 2 represented a first step in  understanding  the
characterization of co-occurrence and sequential  exposure regimes present
under ambient conditions.

      o   Additional evaluation is required to determine  the
          important exposure sequences of air pollution.

VEGETATION EFFECTS

      The  identification of regimes of pollutant mixtures representative  of
ambient exposures is important because these regimes define concentration
peaks, means, varying times of exposures, and time  between events to  be used
in vegetation experiments.   Access to the existing  data bases facilitates the
design of  experiments that  mimic  the exposures representative of  ambient
condi ti ons.

      o   fit present, there are extensive knowledge gaps  on
          effects on major  crop species,  garden and amenity
          plants, and native herbaceous  and tree species.   Current
          data are based on experiments  conducted with different
          methodologies, environments, experimental designs,  and
          interpretation.
                                      83

-------
      o   The panel believed that the effects on vegetation
          resulting from the interaction of pollutant gases and
          acid deposition,  aerosols and heavy metals on soils need
          further study.

      o   Added information is required to describe the
          mechanistic and plant processes at realistic exposure
          regimes for pollutant mixtures.

      o   Little data exist to explain the assimilative
          allocation, physiological,  and biological responses that
          relate to vegetation dose-response studies.   Little data
          are available on  relating uptake rates to pollutant
          mixtures.

      o   The current data  base does not provide sufficient
          information to understand the genetics of vegetation
          response to mixtures.   Species,  cultivar,  and individual
          plant variability have not  been  well-characterized.   The
          heritability of mixture sensitivity and its relationship
          to single pollutant sensitivity have not been
          characterized adequately.

      o    For vegetation effects research involving air pollutant
          mixtures, as well as for singular pollutant  exposures,
          there exist a paucity of data describing environmental
          effects (e.g.,  temperature,  soil moisture,  humidity,
          wind velocity,  nutrition,  and age).

      The panel  strongly  urged the selection of  specific research
investigations that were  applicable  to  address  high  priority  research
questi ons.
                                      84

-------
      o   The panel believed that  additional  information was
          needed to develop modeling and predictive capabilities.
          Sources of information about  phenology,  environmental,
          and genetic (intra- and interspecific)  variation need to
          be quantified to describe interdependent effects.

      o   The changes in plant sensitivity to gas  mixtures with
          changing leaf age and development stage  have  had little
          study.   Information is especially lacking on  sensitivity
          during the fruiting period.

      o   There is little or no information on effects  of mixtures
          on plant hardiness, reproduction, nutritional value,  and
          other characteristics affecting adaptation and
          utilization of plants.  While there is now some evidence
          for the existence of plant recovery and  adaptation
          processes, little is known of the nature of such
          homeostatic processes and mechanisms.

      o   The panel believed that  there have  been  few attempts  to
          fully utilize alternative statistical designs and
          analyses.  Such methods could include covariate
          measurements, rotatable designs,  and response surface
          presentations.

      Following the identification of research requirements, the panel focused
on recommending the specific direction  in which air pollution mixture
vegetation research should follow.  Chapter 5 presents  the panel's findings.
                                      85

-------
                             5.  RECOMMENDATIONS

INTRODUCTION

      The potential  value  of  prioritizing  research needs  was  addressed by the
panel.   The  members  believed  that  research areas  were not independent of each
other.   For  example,  the biological  research  efforts  described later in this
chapter are  dependent upon  a  knowledge  of  the pollutant concentrations
occurring in the  field  under  ambient conditions.   The panel  has concluded that
environmental,  genetic,  and phenological  variables should be  considered when
research studies  are  initiated.

AIR QUALITY

      An analysis of  the EPA  SAROAD,  EPRI  SURE,  and TVA data  bases  indicates
that S0?, NO-,  and 0, may  co-occur  in various concentrations  in rural,
suburban, and urban  areas.  For  many  rural  sites,  co-occurrence (using 0.05  ppm
as the  definition of  an event)  is  infrequent.  For most cases analyzed,  events
lasted  for only a few hours and were separated  by  weeks or months.

The panel recommends  that

      o   air quality data  be further evaluated using patterns of
          occurrence  of the combined pollutants to establish
          guidelines  for designing plant interaction  research
          investigations.   The  primary  pollutants  of  interest are
          SO.,,  Oy and NO-,; they can be evaluated  using the
          available air quality data and research  information
          dealing with the  individual effects on  vegetation.

     o   acidic  deposition be  considered  as  a  pollutant  with
          potential for interaction  with SO   0   and NO.-,.
                                      86

-------
      o   an analysis of existing air quality data bases (starting
          with SAROAD) be instituted to derive the joint
          probability distributions of pollutants and the diurnal
          patterns of exposure for plant exposure experiments.
          Additional sources of rural air quality data could
          include the USDA Forest Service,  EPRI (SURE),  EPfl
          (NCLAN), as well as permit monitoring programs (e.g.,
          PSD applications).  This analysis is to include:

          1)   Search the data base for locations where either
               co-occurrence or sequential  exposures occur.   This
               search would include separate listings at several
               threshold concentrations (e.g., 0.05,  0.04,  0.03,
               and 0.02 ppm).
          2)   Once locations are identified,  the monitoring data
               bases at the locations should be presented as joint
               frequency distributions and as diurnal time  series.
               The utility of spectral analysis (Fourier series)
               and the Box Jenkins model should be explored.
          Jj   The results of this process  should be disseminated
               to research groups to guide  experimental  exposures
               used in interaction experiments.

      o   potential data displays for individual pollutants could
          include:  1) three-dimensional plots of concentration,
          duration, and frequency; 2) diurnal plots for individual
          pollutants in terms of mean levels and frequency  above
          particular levels.  These analyses would utilize  data
          from the growing season for some  relevant time period)
          and could also serve to identify potential anomalies (in
          terms of data values or sites).

      The air quality analyses would provide information that could  be  used  to
identify general  patterns in terms of geographic region  or  source
configuration.   It may be necessary to supplement the air quality data  for
point source pollutants by considering the  use of dispersion models  to  provide
information on  levels, diurnal patterns, and time between episodic events.
                                      87

-------
      As a result of the analysis,  it is anticipated that the information
would provide researchers with relevant exposure patterns that have a known
probability of occurrence.  In addition, the information that describes the
rural site exposures might also be  compared to urban results so that
researchers can establish possible  exposure relationships.

BIOLOGICAL EFFECTS

Introduction

      The identified research  needs  are divided  into two areas which have not
been assigned a specific funding.   The  panel  believes the activities are
complementary and fill  gaps  in the  information base  that describe  the effects
of pollutant mixtures on vegetation.   The  work group members stated that

      o   research efforts should be  directed toward major  crop
          species, cultivated  plants  (including  commercial  crops
          and plants utilized  for garden and home use),  and native
          herbaceous and tree  species.

      o   to assist in  the design of  future research on  the
          effects of pollutant mixtures on vegetation, available
          air quality data should be  re-evaluated to help identify
          appropriate response data needed from  plant experiments.

      The panel feels that knowledge  gaps  can  only be filled by an
integrated research effort with growth  chambers,  greenhouses,  and
field plots.

Realistic Exposure Regimes

      The panel believes it  essential  to understand  the  response of plants  to
various exposure regimes.
                                      88

-------
      o   Effects must be associated with air pollutant  peaks,
          means,  length of exposure,  and time between exposures.
          Research using realistic ambient pollutant  exposures
          should  evaluate the vegetation effects  associated  with
          sequential exposures of pollutant mixtures  that mimic
          ambient conditions.

Development of Minimum Recommendations  for Research Protocols

      The panel  recommends that

      o   a minimum set of standardized procedures, to ensure the
          quality assurance of plant  response studies, should be
          established.  Generalized guidelines should be proposed
          for 1)  plant growth conditions, 2) environmental and
          plant monitoring, Jj pollutant exposures, and  4) uniform
          terminology (describing plant response  characteristics).
          The most efficient experimental designs and analysis
          (relevant to a specific experimental goal)  should  be
          implemented (e.g., covariant  analysis,  analysis of
          variance, and rotatable design).  It is proposed that
          these minimum guidelines be developed through  a series
          of workshops involving scientists experienced  in
          designing and implementing research on  the  impacts of
          air pollution on vegetation.

      o   as part of a generalized protocol that  researchers
          should clearly define the meaning of agreed upon
          concepts (e.g., less than additive, synergism, and
          antagonism).
                                      89

-------
Predictive Capabilities

      The purpose  of  the  pollutant  Interaction  research  is  to  develop
predictive capabilities for  assessing  vegetation  effects when  experimental
data are insufficient.  Predictive  capabilities allow  for the  extrapolation  of
results to ambient exposure  conditions  that  have  previously not  been tested.
To properly develop the information necessary to  predict vegetation effects
associated with  pollutant mixtures,  it  is  necessary  to implement a research
program involving  studies that  elucidate 1)  the modes  of action  and 2)  the
sources of biological  variation.  From  these  experimental results will come the
modeling required  to  develop the  predictive  capabilities that  are necessary  to
quantify possible  effects on vegetation.

      As part of the  development  of predictive  capabilites,  the  panel
recommends that  the following research  activities  be implemented:

      o   Afoc/es  of fiction:   The objectives of this research
          activity is to  understand how air  contaminants influence
          biological  processes.   Studies need to  address modes of
          action of pollutants  singly  and  in combination.   The
          research effort should  include both sequential and
          co-occurrence exposures and  should be conducted with an
          appreciation of realistic exposure regimes.   The
          biological  level of organization should focus on
          processes at all levels of plant organization  (i.e., the
          cell,  whole-plant, population, and ecosystem). The panel
          believes that there should be two  major areas of
          interest
          a)    The  relationship between  the different mechanisms
               of pollutant  response.
          b)    Ihe  varying biological responses attributed  to
               different  levels of  air pollutant exposure.
                                      90

-------
      o   Sources of Variation:   The plant response to  a  given
          exposure regime varies significantly with specific
          environments,  and stage of plant development.   The panel
          recommends research that focuses on each of the
          following:

          a)   genotype-the significance of intra- (e.g.,
               cultivar,  population) and interspecific  genotypes.
               This includes phenology as a source of variation.
          b)   environment-the significance of edaphic  (e.g.,  soil
               water availability, soil nutrients), climatic
               (e.g.,  temperature, light,  relative humidity,
               elevated carbon dioxide,  etc. ) and biotic  factors
               (e.g.,  pathogens, syrrtionts,  competition,  etc.).
      o   f^bdeling-The development of data that describe  the
          process and mechanistic activities associated with air
          pollutant mixture vegetation effects should allow for
          the development of conceptual and quantitative  models of
          biological response.

CONCLUSION

      It is the opinion  of the workshop participants  that the  position paper
(which is presented in Chapter 2) focused  on  ambient  exposures and represented
an initial attempt to  identify realistic exposure regimes  that exist in the
ambient air.  The panel  believes that

      o   additional efforts should be made to supplement  the
          existing analysis.

      o   the efforts  should proceed simultaneously as  the
          biological vegetation  effects research is
          implemented.

-------
      o    the  results of  the  air  quality characterization should
          be used in developing the design of  the pollutant
          mixture experimental protocols.

      The  first  two stages  of the process and  mechanistic research  should
include

      o    a biological effects screening exercise to prioritize
          which  air pollutant mixture exposures are most likely to
          be significant.   This effort is suitable in controlled
          exposure facilities, and

      o    a more detailed investigation performed under field and
          laboratory situations for the purpose of quantifying the
          significance of the major factors affecting plant
          response.

      In  regard  to conceptual models, it was the conclusion of the  panel
members  that

      o    they should combine existing models of joint action with
          the  data that describe  the modes of biological action.
          The  quantitative models  should be capable of providing
          accurate and precise estimates of plant response.  In
          addition, the models should be compatible with the
          conceptual interpretation of the modes of action.
                                      92

-------
REFERENCES

Amundson, R.6., and L.H. Weinstein. 1980. The effects of S0? and NO- alone and
       in combination on the yield of soybean.  Plant Physiol.  65:152
       supplement.

Amundson, R.G., and L.H. Weinstein.  1981. Joint action of sulfur dioxide and
       nitrogen dioxide on foliar injury and stomatal behaviour in soybean.  J.
       Environ. Quality 10:204-206.

Amundson, R.6., L.H. Weinstein, P. Van Leuken, and L.J. Colavito.  1982.  Joint
       action of HF and NO,, on growth, fluorine accumulation, and leaf
       resistance in Marcross sweet corn.  Environ. Exp. Bot. 22:49-55.

Ashenden, T.W.  1978. Extreme pollution sensitivity of grasses  when S0?  and
        N09 are present in the atmosphere together.  Nature, 273:142-143.

Ashenden, T.W.  1979a. The effects of long-term exposures to S0? and N0?
       pollution on the growth of Dactylis glomerata L. and Poa pratensis L.
       Environ. Pollut. 18:249-258.

Ashenden, T.W.  1979b. Effects of SO,, and NO- pollution on transpiration  in
       Phaseolus vulgaris L. Environ. Pollut. 18:45-50.

Ashenden, T.W., and T.A. Mansfield.  1978. Extreme pollution sensitivity  of
       grasses when S09 and N09 are present in the atmosphere together.  Nature
       273:142-143.   ^       ^

Ashenden, T.W., and I.A.D. Williams.  1980. Growth reductions in Lolium
       multiforum and Phleum Pratense as a result of S0? and N0? pollution.
       Environ. Pollut. 21:131-139.^

Ashenden, T.W., P.W. Tabner, P. Williams, M.E. Whitmore, and T.A. Mansfield.
       1982.  A large scale system for fumigating plants with S0? and N0?.
       Environ. Pollut. (Ser. B.) 3:21-26.                              *

Bachmann, J., and L.J. Zaragoza.  1983. Protecting public welfare: the role  of
       environmental effects information in air standards development.
       Presented at the 76th Air Pollution Control Association  National
       Meeting, Atlanta, GA.  June 19-24, 1983. 83-3.1.

Beckerson, D.W., and G. Hofstra.  1979a. Response of leaf diffusive resistance
       of radish, cucumber and soybean to 0, and SO,, singly or  in combination.
       Atmos. Environ. 13:1263-1268.

Beckerson, D.W., and G. Hofstra. 1979b, Stomatal responses of white bean  to  0-,
       and S0? singly or in combination.  Atmos. Environ. 13:533-535.

Beckerson, D.W., and G. Hofstra.  1980. Effects of sulphur dioxide and ozone,
       singly or in combination, on membrane permeability.  Can. J. Bot.
       58:451-457.
                                       93

-------
Beckerson,  D.W.,  G.  Hofstra,  and  R.  Wukasch.   1979.  The relative sensitivities
       of 33  bean cultivars  to ozone and sulfur dioxide singly or in
       combination in  controlled  exposures and to oxidants in the field.  Plant
       Dis. Rep.  63:478:482.

Bennett,  J.P.,  H.M.  Resh,  and V.C.  Runneckles.  1974.  Apparent stimulation of
       plant  growth  by air pollutants.   Can.  J. Bot.  52:35-41.

Bennett,  J.P.,  K.  Barnes,  and J.H.  Shinn.  1980. Interactive effects of hLS and
       6.5 on  the  yield of  snap beans (Phaseolus vulgaris L.)   Environ. Exp.
       BOt. 20:107-114.

Black,  V.J.,  D.P.  Ormrod,  and M.H.  Unsworth.   1982.  Effects of low
       concentrations  of ozone,  singly  and in combination with sulphur dioxide
       on net photosynthesis  rates  of Vicia faba L.  J.  Exp. Bot. 33:1302-1311.

Black,  V.J.   1982.  Effects of sulfur dioxide  on physiological processes in
       plants.  In:  M.H. Unsworth  and D.P.  Ormrod (eds.)  Effects of Gaseous Air
       Pollution  in  Agriculture  and  Horticulture.   11:225-246. London/Boston:
       Butterworth Scientific.   532  pp.
Bull,  J.N.,  and  T.A.  Mansfield.   1974.
       and N00.   Nature  250:433-444.
                                       Photosynthesis  in  leaves  exposed  to SO.
Carlson,  R.W.   1979.  Reduction  in  the  photosynthetic  rate  of  Acer,  Quercus and
       Fraxinus species  caused  by  sulphur  dioxide  and  ozone.   Environ.  Pollut.
       18:159-170.

Costonis,  A.C.   1973.  Injury  to eastern  white  pine by  sulfur  dioxide and ozone
       alone and in  mixtures.   Eur.  J.  Forest  Pathol.  3:50-55.

Cowling,  D.W.,  and M.J.  Koziol.  1982.  Mineral  nutrition  and plant responses to
       air pol lutants.pp.  349-375.   lr\_:  M.H. Unsworth  and  D.P.  Ormrod (eds.)
       Effects  of Gaseous  Air Pollution  in Agriculture and  Horticulture.
       London/Boston:  Butterworth Scientific.  532 pp.
                                                                       J.  Air
Coyne,  P. I.,  and  G.E.  Bingham.   1978.  Photosynthesis  and  stomatal  light
       responses  in  snap  beans  exposed  to  hydrogen  sulfide  and  ozone.
       Pollut.  Control  Assoc. 28:1119-1128.

Czuba,  M.,  and  D.P.  Ormrod.  1974.  Effects  of  cadmium  and  zinc  on  ozone-induced
       phytotoxicHy in cress and  lettuce.  Can.  J. Bot.  52:645-649.

Czuba,  M.,  and  D.P.  Ormrod.   1981.  Cadmium concentrations in  cress shoots in
       relation to cadmium-enhanced  ozone  phytotoxicity.   Environ.  Pollut.
       (Ser.  A) 25:67-76.

Decormis,  !_„,  and M.  Luttringer.   1977.  Influence of  sulfur dioxide and
       nitrogen dioxide synergism.   (Effects  sur  les  vegetaux  de  la synerqie
       dioxxyde de soufre-oxides d.  azote).   Pollut.  Atmos. 75:245-247.
                                      94

-------
Department of Health, Education, and Welfare (DHEW).  1970. Air quality
       criteria for sulfur oxides.  National Air Pollution Control
       Administration, Washington, DC.  pp. 61-66, 160-161.

Dochinger, L.S., F.W. Bender, P.O. Box, and W.W. Heck.  1970. Chlorotic dwarf
       of eastern white pine caused by an ozone and sulphur dioxide
       interaction.  Nature 255:476.

Electric Power Research Institute.  1982. Eastern Regional Air Quality
       Measurements.  Vol. 1. Prepared by Environmental Research & Technology,
       Inc.  EPRI EA-1914, Palo Alto, CA.

Elkiey, T., and D.P. Ormrod.  1979a. Leaf diffusion resistance responses of
       three petunia cultivars to ozone and/or sulfur dioxide.  J. Air Pollut.
       Control Assoc. 29:622-625.

Elkiey, T., and D.P- Ormrod.  1979b. Ozone and/or sulphur dioxide effects on
       tissue permeability in petunia leaves.  Atmos. Environ. 13:1165-1168.

Elkiey, T., and D.P. Ormrod.  1979c. Ozone and sulphur dioxide effects on leaf
       water potential of Petunia. Z. Pfansenphysiologie 91:177-181.

Elkiey, T., and D.P. Ormrod.  1980. Response of turfgrass cultivars to ozone,
       sulfur dioxide, nitrogen dioxide, and their mixture.  J. Amer.  Soc.
       Hort. Sci.  105:664-668.

Elkiey T., and D.P. Ormrod.  1981a. Sulphite, total sulphur and total  nitrogen
       accumulation by petunia leaves exposed to ozone, sulphur dioxide and
       nitrogen dioxide.  Environ. Pollut. (Ser. A). 24:233-241.

Elkiey, T., and D.P. Ormrod. 1981b. Sorption of 0,, SOj, NOp or their mixtures
       by nine Poa pratensis cultivars of differing pollutant sensitivity.
       Atmos. Environ. 15:1739-1743.

Elkiey, T., and D.P. Ormrod.  1981c. Absorption of ozone, sulfur-dioxide, and
       nitrogen dioxide by petunia plants.  Environ. Exp. Bot. 21:63-70.

Elkiey, T., D.P. Ormrod, and B. Marie.  1982. Foliar sorption of sulfur
       dioxide, nitrogen dioxide, and ozone by ornamental woody plants.
       HortScience 17:358-360.

Environmental Protection Agency.  1976. AEROS Manual Series Volume V:  AEROS
       Manual of Codes.  Office of Air and Waste Management, Office of Air
       Quality Planning and Standards, Research Triangle Park, NC.
       EPA-450/2-76-005.

Environmental Protection Agency.  1978. Air Quality Criteria for Ozone and
       Other Photochemical Oxidants.  Environmental Criteria and Assessment
       Office, Research Triangle Park, NC.  EPA-600/8-78-004.
                                       95

-------
Environmental  Protection  Agency.   1981a.  Preliminary assessment  of health
       and  welfare  effects  associated  with  nitrogen  oxides  for
       standard-setting purposes.   Prepared by  the  Division of  Strategies and
       Air  Standards.   United  States  Environmental  Protection Agency,  Research
       Triangle Park,  North  Carolina.

Environmental  Protection  Agency.   1981b.  1980  ambient assessment—air
       portion.  Office of  Air Quality Planning and  Standards,  Research
       Triangle Park,  North  Carolina.

Environmental  Protection  Agency.   1981c.  Air Quality Criteria for Oxides of
       Nitrogen.   Prepared  by  the  Environmental  Criteria and Assessment
       Office,   Research  Triangle  Park,  North Carolina.  EPA-600/8-82-026.

Environmental  Protection  Agency.   1982a.  Review  of  the  National  Ambient Air
       Quality Standards  for Nitrogen  Oxides:   Assessment of Scientific and
       Technical  Information:  OAQPS Staff Paper.  Prepared by the Office of Air
       Quality Planning and  Standards, EPA,  Research Triangle Park,  NC.
       EPA-450/5-82-002.

Environmental  Protection  Agency.   1982b.  Review of  the  National  Ambient Air
       Quality Standards  for Sulfur Oxides;  Assessment  of Scientific and
       Technical  Information.   Prepared  by  the  Division of  Strategies  and Air
       Standards.   United States Environmental  Protection Agency, Research
       Triangle Park,  North  Carolina.

Environmental  Protection  Agency.   1982c.  Air Quality Data - 1981 Annual
       Statistics  Including  Summaries  with  Reference to Standards.  Office of
       Air  Planning and Standards,  Research Triangle Park,  North Carolina.
       EPA-450/4-82-007.

Evans,  L.S.,  and  P.R.  Miller.   1975.  Histological comparison of  single and
       additive 0^  and  S0?  injuries to elongating Ponderosa pine needles.
       Amer.  J. BOt. 62:416-421.

Foster,  K.W.,  H.  Timm,  C.K.  Labanauskas  and R.J.  Oshima.  1983.  Effects of
       ozone  and  sulfur dioxide on  tuber  yield  and  quality  of potatoes.  J.
       Environ. Qual.  12:   In  press.

Fujiwara, T.  1973.  Damage to plants caused  by complex pollution  of atmosphere.
       Snokubutsu  Boeki.  (Plant Protection) 27:233-236.

Fujiwara, T.,  and  H. Ishikawa.  1976.  Actual effects of combination  of sulfur
       dioxide, nitrogen  dioxide and  ozone  on some  plants (in Japanese with
       English summary).  Bull. Bio-Environ. Lab,,  No.  476001.  1019  pp.

Furukawa, A.,  and  T. Totsuka.   1979.  Effects of  N0?, S0? and 0?  alone  and in
       combination  on  net photosynthesis  in  sunflower.   Environ.  Control Biol.
       17:161-166.
                                          96

-------
Gardner, J.O., and D.P. Ormrod.  1977. Response of Reiger begonia to ozone and
       sulphur dioxide.  Sci. Hort. 5:171-181.

Glass, G.E., and T.G. Brydges.  1981. Problem complexity in predicting impacts
       from altered precipitation chemistry.  Presented at Int. Conf. on Acid
       Precipitation Impacts, Cornell University, Ithaca, NY.  August 1981.

Grosso, J.J., H.A. Menser, G.H. Hodges, and H.H. McKinney.  1971. Effects of
       air pollutants on Nicotiana cultivars and species used for virus
       studies.  Phytopathology 61:945-950.

Halbwachs, G., and W. Kronberger.  1979. Das Verhalten F- and S0?- resistenter
       fichtenklone verschledener herkunft  in pinem rauchschaden? gebiet.  Zb.
       Inst. Gozdno. Lesn. Gospos.  Ljubljana, pp. 259-268.

Heagle, A.S., and J.W.  Johnston.  1979. Variable responses of soybeans to
       mixtures of ozone and sulfur dioxide.  J. Air Pollut. Control Assoc.
       29:729-732.
of
Heagle, A.S., D.E. Body, and G.E. Neely.  1974. Injury and yield responses
       soybean to chronic doses of ozone and sulfur dioxide in the field.
       Phytopathology 64:132-136.

Heagle, A.S., W.W. Heck, J.O. Rawlings, and R.B. Philbeck.  1983. Effects of
       chronic doses of ozone and sulfur dioxide on injury and yield of
       soybeans in open-top chambers.  Crop Sci. In press.

Heck, W.W., S.V. Krupa, and S.N. Linzon.  1979. Methodology for the Assessment
       of Air Pollution Effects on Vegetation.  Air Pollution Control
       Association, Pittsburgh.

Heck, W.W., W.W. Cure, D.S. Shriner, R.J. Olson, and A.S. Heagle.  1982. Ozone
       impacts on the productivity of  selected crops.  In : J.S. Jacobson and
       A. A. Millen (eds.)  Symposium on the Effects of Air Pollution on Farm
       Commodities.  Isaak Walton League of America, Arlington, VA.

Heggestad, H.E., and J.H. Bennett.  1981. Photochemical oxidants potentiate
       yield losses in snap beans attributable to sulfur dioxide.  Science
       213:1008-1010.

Hill, A.C., S. Hill, C. Lamb, and T.W. Barrett.  1974. Sensitivity of native
       desert vegetation to S02 and to S0? and N0? combined.  J. Air Pollut.
       Control Assoc. 24:153-157.

Hodges, G.H., H.A. Menser, and W.B. Ogden.  1971. Susceptibility of Wisconsin
       Havanna tobacco cultivars to air pollutants.  Argon. J. 63:107-111.

Hofstra, G., and D.W. Beckerson, 1981.  Foliar responses of five plant species
       to ozone and a sulphur dioxide/ozone mixture after a sulphur dioxide
       pre-exposure.  Atmos. Environ.  15:383-389.
                                          97

-------
Hofstra,  G.,  and  D.P.  Ormrod.   1977.  Ozone and sulphur dioxide interaction in
       white  bean and  soybean.   Can.  J.  Plant Sci.  57:1193-1198.

Horsman,  D.C.,  and A.R.  Wellburn.   1975.  Synergistic effect of SO- and NO-
       polluted air upon enzyme activity  in pea seedlings.   Environ.  Polmt,
       8:123-133.

Horsman,  D.C.,  and A.R.  Wellburn.  1976.  Guide to the metabolic and biochemcial
       effects  of air  pollutants on higher plants.   In: T. A.  Mansfield (ed.)
       Effects  of Air  Pollutants on Plants,  pp. 1857T89.

Hou, L.Y., A.C. Hill,  and A.  Soleimani.   1977. Influence of CO,, on the effects
       of S02 and NO-  on alfalfa.   Environ. Pollut. 12:7-16.

Houston,  D.B. 1974. Response  of selected  Pinus strobus L.  clones  to fumigation
       with sulfur dioxide and  ozone.   Can. J. For. Res. 4:65:68.

Irving, P.M., and J.E. Miller,  1981.   Productivity of field-grown  soybeans
       exposed  to acid rain and sulfur dioxide alone and in combination J.
       Environ. Quality 10:473-478.

Irving, P., J.E.  Miller, and  P.B.  Xerikos.  1982.  The effect  of N02 and S02
       alone  and in combination on the productivity of field-grown soybeans.
       JJK  T.  Schneider and  L. Grand  (eds.)  Air Pollution  by Nitrogen Oxides.
       pp. 521-531.  Amsterdam, The Netherlands: Elsevier  Scientific.

Ishikawa, H.   1976. Sensitivity of cultivated plants to air pollutants. Report
       from Bio Enviroment Laboratory, Central Research Institute  of Electric
       Power  Industry, 1646 Abiko Chiba,  Japan.

Jacobson, J.S., and L.J. Colavito. 1976.  The  combined effect  of sulfur dioxide
       and ozone on bean and  tobacco  plants.   Environ, Exp. Bot.  16:277-285.

Jaeger, J., and W. Banfield.  1970. Response of eastern white  pine  to prolonged
       exposure to atmospheric  levels  of  ozone, sulfur dioxide, or mixtures of
       these  pollutants.  Phytopathology  60:577.

Karnosky, D.F.  1976.  Threshold  levels  for foliar injury to  Populus tremuloides
       by sulfur dioxide and  ozone.  Can. J.  For.  Res. 6:166-169~

Kender, W.J., and F.H.F.G. Spierings.   1975.  Effects of sulfur dioxide, ozone,
       and their interactions on 'Golden  Delicious' apple  trees.   Neth. J.
       Plant  Pathol.  81:149-151.

Klarer, C.I.   1982. Effects of  sulfur  dioxide and nitrogen  dioxide, singly and
       in combination  over time on ribulose bisphosphate carboxylase activity
       and vegetative  growth  of soybean.   M.S. Thesis, North  Carolina State
       Univ., Raleigh, NC.  76  pp.

Kohut, R.J.,  and D.D.  Davis.   1978. Response  of pinto bean  to simultaneous
       exposure to ozone and  PAN.   Phytopathology 68:567-569.

-------
Krause, C.R., and K.F. Jensen.  1978. Microtopographical changes in hybrid
       poplar leaves associated with air pollution exposure.  Scan. Elec.
       Micros. 2:755-758.

Krause, C.R., and K.F. Jensen.  1979. Surface changes on hybrid poplar leaves
       exposed to ozone and sulfur dioxide.  Scan. Elec. Micros. 3:77-80.

Krause, G.H.M., and Kaiser, H. 1977. Plant response to heavy metal and sulphur
       dioxide.  Environ. Pollut.  12:63-71.

Kress, L.W., and J.M. Skelly.   1982. Response of several eastern forest tree
       species to chronic doses of ozone and nitrogen dioxide.  Plant Dis.
       66:1149-1152.

Kress, L.W., J.M.  Skelly, and  K.H. Hinkelmann.  1982a. Growth impact of CU, N02
       and/or SO^ on Pinus taeda.   Environ. Monit. Assess.  1:229-239.

Kress, L.W., J.M.  Skelly, and  K.H. Hinkelmann.  1982b. Growth impact of CU, NOn
       and/or SO^ on Plantanus occidental is.  Agric.  Environ.  7:265-274.

Lamoreaux, R.J., and W.R. Chaney.   1978. Photosynthesis and transpiration of
       excised silver maple leaves exposed to cadmium and sulphur dioxide.
       Environ. Pollut. 17:259-268.

Le Sueur-Brymer, N.M., and D.P. Ormrod.  1983.  Net photosynthesis rates of
       fruiting soybean plants exposed to ozone and sulphur dioxide singly or
       in combination.  Can.  J. Plant Sci. (submitted for publication).

Lewis, E., and E.  Brennan.  1978.  Ozone and sulfur dioxide  mixtures cause a
       PAN-type injury to petunia.  Phytopathology 68:1011-1014.

Macdowall, F.D.H.,  and A.F.W.  Cole.   1971. Threshold  and synergistic damage to
       tobacco by ozone and sulfur dioxide.  Atmos. Environ. 5:553-559.

Malhotra, S.S., and D. Hocking.  1976. Biochemical and cytological effects of
       sulphur dioxide on plant metabolism.  New.  Phytol. 76:227-237.

Mandl, R.H., L.H.  Weinstein,  and M.  Keveny. 1975.  Effects of hydrogen fluoride
       and sulphur  dioxide alone and in combination on several species of
       plants.  Environ.  Pollut. 9:133-143.

Mandl, R.H., L.H.  Weinstein,  M. Dean, and M. Wheeler.  1980. The response of
       sweet corn to HF and S0? under field conditions.   Environ.  Exp. Bot.
       20:359-365.

Mansfield, T.A., and T.W. Ashenden.   1978. Extreme pollution sensitivity  of
       grasses when S0? and N0? are  present in  the atmosphere  together. Nature
       273:142-143.   i
                                         99

-------
Masaru,  N.,  F.  Syozo,  and  K.  Saburo.   1976.  Effects of exposure to various
       injurious  gases  on  germination of lily pollen.   Environ. Pollut.
       11:181-187.

Matsushima,  J.,  and  R.F.  Brewer.   1972.  Influence of sulfur dioxide and
       hydrogen  fluoride  as  a mix or  reciprocal  exposure on citrus growth and
       development.   J. Air  Pollut. Control  Assoc.  22:710-713.

McCune,  D.C.   1983.  Interaction  of fluorides with air pollutants.  Intern.
       Fluoride  Symp.   Utah  State Univ., Logan,  UT, May 24-27,  1982.

Mclaughlin,  S.B.,  D.S.  Shriner,  R.K.  M'Conathy,  and L.K. Mann.   1979.  The
       effects  of S0?  dosage  kinetics and exposure  frequency on photosynthesis
       and transpiration  of  kidney beans (Phaseolus vulgaris,  L.).  Environ.
       Exp.  Bot.   19:179-191.

Menser,  H.A.,  and H.E.  Heggestad.   1966. Ozone and  sulfur dioxide  synergism:
       injury  to  tobacco  plants.   Science 153:424-425.

Menser,  H.A.,  and G.H.  Hodges.   1970. Effects of air pollutants on burley
       tobacco  cultivars.   Agronomy J.  63:265-269.

Miller,  C.A.,  and D.D.  Davis.   1981a. Response of pinto bean plants exposed
       to Ov  S0?,  or  mixtures  at varying temperatures.  HortScience
       16:548-556.

Miller,  C.A.,  and D.D.  Davis.   1981b. Effect of  temperature on  stomatal
       conductance  and  ozone  injury of pinto bean leaves.  Plant Dis.
       65:750:751.

Miller,  V.C.,  R.K.  Howell,  and  B.E. Caldwell.  1974.  Relative  sensitivity of
       soybean  genotypes  to  ozone and sulfur dioxide.   J. Environ. Qual.
       3:35-37.

Mueller,  P.K.,  G.  M.  Hidy,  K.  Warren, T.F.  Lavery,  and  R.L.  Baskett.  1980. The
       occurrence of atmospheric  aerosols in the northeastern  United  States.
       In:  Aerosols:  Anthropogenic and  Natural, Sources and Transport.  New
       York  Academy  of  Sciences,  New  York,  N.Y., pp.  463-482.

Murray,  J.J., R.K.  Howell,  and  A.C. Wilton.   1979.  Differential response of
       seventeen  Poa protansis  cultivars to  ozone and  sulfur dioxide.   Plant
       Dis.  Rep.  59:852-854.

Noble,  R.D.,  and  K.F.  Jensen.   1980.  Effects of  sulfur  dioxide  and ozone on
       growth  of  hybrid poplar  leaves.   Amer. J. Bot.  67:1005-1009.

Noggle,  J.C.,  and H.C.  Jones.   1981.  Regional effects  of multiple  air
       pollutants  on plants.   Presented  at  the 1981 Annual  Meeting of the Air
       Pollution  Control  Association, Philadelphia, PA.  Paper  #42.5.
                                          100

-------
O'Connor, J.R.  1980. Overview of the criteria review and standard setting
       process.  Presented at the 1980 Air Pollution Control Association
       Specialty Conference on the Technical Basis for a Size Specific
       Participate Standard.  Danvers, Massachusetts, pp. 15-25.

Olszyk, D.M., and T.W. Tibbitts.  1981. Stomatal response and leaf injury of
       Pisum sativum L.  with SO- and 03 exposures.  Plant Physiol. 67:539-544.

Ormrod, D.P. 1977.  Cadmium and nickel effects on growth and ozone sensitivity
       of pea.  Water, Air, Soil Pollut. 8:263-270.

Ormrod, D.P. 1978.  Pollution in Horticulture. Elsevier, Amsterdam.

Ormrod, D.P. 1982.  Air pollutant interactions in mixtures. Pages 307-331. In:
       M.H. Unsworth and D.P. Ormrod (eds.).  Effects of Air Pollution in
       Agriculture and Horticulture.  Butterworths, London.

Ormrod, D.P., V.J. Black, and M.H. Unsworth.  1981. Depression of net
       photosynthesis in Vicia faba L. exposed to sulphur dioxide and ozone.
       Nature 291:585-586.

Ormrod, D.P., D.T. Tingey, and M.L. Gumpertz, and D.M. Olszyk.  1983a.
       Utilization of a response surface technique in the study of
       physiological responses to ozone and sulfur dioxide mixtures.   Ms. in
       preparation.

Ormrod, D.P-, D.T. Tingey, and M.L. Gumpertz.  1983b. Covariate measurements
       for increasing the precision of plant response to 0, and S0?.
       HortScience (submitted for publication).

Oshima, R.J.  1978. The impact of sulfur dioxide on vegetation:  A sulfur
       dioxide-ozone response model. California Air Resources Board,  Agreement
       No. A6-162-30.

Padgett, J., and H. Richmond.  1983. The process of establishing and  revising
       National Ambient Air Quality Standards.  J. of Air Pollut. Control
       Assoc.  33:13-16.

Pell, E.J.  1976. Influence of benomyl soil treatment on pinto bean plants
       exposed to peroxyacetyl nitrate and ozone.  Phytopathology, 66:731-733.

Proctor, J.T.A., and D.P. Ormrod.  1981. Sensitivity of ginseng to ozone and
       sulfur dioxide.  HortScience 16:647:648.

Rajput, C.B.S., D.P.  Ormrod, and W.D.  Evans.  1977. The resistance of
       strawberry to ozone and sulfur dioxide.  Plant Dis. Rep.  61:222-225.

Reich, P.B., R.G. Amundson, and J.P. Lassoie.  1982. Reduction i soybean yield
       often exposure to ozone and sulfur dioxide using a Linear Gradient
       Exposure Technique.  Water, Air, Soil Pollut. 17:29-36.
                                          10'

-------
Reinert, R.A., and T.N. Gray.  1981. The response of radish to nitrogen
       dioxide, sulfur dioxide, and ozone, alone and in combination.   J.
       Environ. Quality 10:240-243.

Reinert, R.A., and W.W. Heck. 1982. Effects of nitrogen dioxide in combination
       with sulfur dioxide and ozone on selected crops.  In: (T. Schneider  and
       L. Grant (eds.) Air Pollution by Nitrogen Oxides,  pp. 533-546.
       Amsterdam, The Netherlands:  Elsevier Scientific.

Reinert, R.A., and P-V. Nelson.  1980. Sensitivity and growth of five  Elatior
       Begonia cultivars to SOP and 0^ alone and in combination. J. Amer. Soc.
       Hort.  Sci. 105:721-723.

Reinert, R.A., and J.S. Sanders.  1982.  Growth of radish and marigold
       following repeated exposure to nitrogen dioxide, sulfur dioxide, and
       ozone.  Plant. Dis. 66:122-124.

Reinert, R.A., and D.E. Weber.  1980. Ozone and sulfur dioxide-induced changes
       in soybean growth.  Phytopathology 70:914-916.

Reinert, R.A., A.S. Heagle, and W.W. Heck.  1975. Plant response to pollutant
       combinations.   In: J.B. Mudd, and T.T.  Koslowski (eds.)  Responses of
       Plants  to Air Pollution,  pp. 159-177.  Academic Press, New York.

Reinert, R.A., D.S. Shriner, and J.O. Rawlings.  1982.  Responses of radish
       to all  combinations of three concentrations of nitrogen dioxide, sulfur
       dioxide, and ozone.  J. Environ. Quality. 11:52-57.

Roques,  A., M. Kerjean, and D. Auclair.  1980. Effets de la pollution
       atmospherique par le fluor et le dioxide de soufre sur 1'appareil
       reproducteur fermelle de Pinus sylverstris en foret de roumare  (Seine
       Maritime, France).  Environ. Pollut. (Ser. A). 21:191-201.

Sanders, J.A., and R.A. Reinert.  1982a. Screening azalea cultivars for
       sensitivity to nitrogen dioxide, sulfur dioxide and ozone alone and  in
       mixtures.  J. Amer. Soc. Hort. Sci. 107:87-90.

Sanders, J.S., and R.A. Reinert.  1982b. Weight changes of radish and marigold
       exposed at three ages to N09, S09 and 07 alone and in mixture. J. Amer.
       Soc. Hort. Sci. 107:726-730:    ^      J

Shertz,  R.D., W.J. Kender, and R.C. Musselman.  1980a.  Foliar response and
       growth of apple trees following exposure to ozone and sulfur dioxide.
       J. Amer. Soc. Hort. Sci. 105:594-598.

Shertz,  R.D., W.J. Kender, and R.C. Musselman.  1980b.  Effects of ozone and
       sulfur dioxide on grapevines.  Sci. Hort. 13:37-45.

Shew, B.B., R.A. Reinert and R.R.  Barker.  1982. Response of tomatoes  to
       ozone, sulfur dioxide, and infection by Pratylenchus penetrans.
       Phytopathology 72:822-826.
                                          102

-------
Singh, S.N.   1980. Synergistic action of particulate and gaseous pollutants on
       the growth of Triticum aestivum L.  J. Exp. Bot. 31:1701-1705.

Skelly, J.M., L.D. Moore, and L.L. Stone.  1972. Symptom expression of eastern
       white pine located near a source of oxide-nitrogen and sulfur dioxide.
       Plant Dis. Rep. 56:3-6.

Solberg, R.A.,  and R.F. Adams.  1956. Histological responses of some plant
       leaves to hydrogen fluoride and sulfur dioxide.  Amer. J. Bot.
       43-755-760.

Thompson, C.R., G. Kats, and R.W. Lennox.  1980. Effects of SO- and/or NO- on
       native plants of the Mohave Desert and eastern Mohave-Colorado Desert.
       J. Air Pollut. Contr. Assoc.  30:1304-1309.

Tingey, D.T., and R.A. Reinert. 1975. The effects of ozone and sulphur dioxide
       singly and in combination on plant growth.  Environ. Pollut. 9:117-125.

Tingey, D.T., and G.E. Taylor, Jr. 1982. Variation in plant response to ozone:
       a conceptual model of physiological events.  Pages 113-138. In: M.H.
       Unsworth and D.P. Ormrod (eds.).  Effects of Gaseous Air Pollution in
       Agriculture and Horticulture.  Butterworths, London.

Tingey, D.T., R.A. Reinert, J.A. Dunning, and W.W. Heck.  1971a. Vegetation
       injury from the interaction of nitrogen dioxide and sulfur dioxide.
       Phytopathology 61:1506-1511.

Tingey, D.T., W.W. Heck, and R.A. Reinert. 1971b. Effect of low concentrations
       of ozone and sulfur dioxide of foliage, growth and yield of radish.  J.
       Amer. Soc. Hort. Sci. 96:369-371.

Tingey, D.T., R.A. Reinert, J.A. Dunning, and W.W. Heck.  1973a. Foliar injury
       responses of eleven plant species to ozone/sulfur dioxide mixtures.
       Atmos. Environ. 7:201-208.

Tingey, D.T., R.A. Reinert, C. Wickliff, and W.W. Heck.  I973b. Chronic ozone
       or sulfur dioxide exposures, or both, affect the early vegetative
       growth of soybean.  Can. J. Plant Sci. 53:875-879.

Toivonen, P.M.A., and G. Hofstra.  1979. The interaction of copper and sulphur
       dioxide in plant injury.  Can. J. Plant Sci. 59:475-479.

Troiano, J., L. Colavito, T. Heller, and D.C. McCune.  1981. Effect of
       simulated acid rain and photochemical oxidant on seed development in
       soybean.  Phytopathology 71:565 (abst.).

Troiano, J., L. Colavito, T. Heller, and D.C. McCune.  1982. Viability, vigor,
       and maturity of seed harvested from two soybean cultivars exposed to
       simulated acidic rain and photochemical oxidants.  Agric. Environ.
       7:275-283.
                                          103

-------
Unsworth, M.H., and D.P.  Ormrod.  1982.  Effects of Gaseous Air Pollution in
       Agriculture and Horticulture.  London/Boston: Butterworth Scientific.
       532 pp.

Weber, D.E., R.A.  Reinert, and K.R.  Barker.  1979. Ozone and sulfur dioxide
       effects on  reproduction and host-parasite relationships of nematodes.
       Phytopathology 69:624-628.

Wellburn, A.R.  1982. Effects of SO- and N0? on metabolic function. _!TK M-H-
       Unsworth and D.P-  Ormord (eds.)  Effects of Gaseous Air Pollution in
       Agriculture and Horticulture.  8:169-188. London/Boston:  Butterworth
       Scientific.  532 pp.

Wellburn, A.R., T.M. Capron, H.S.  Chan,  and D.C. Horsman.  1976.  Biochemical
       effects of atmospheric pollutants on plants.   Pp. 105-114. In:  T.A.
       Mansfield (ed.).  Effects of Air  Pollution on Plants. Cambridge
       University Press,  Cambridge.

Wellburn, A.R., C. Higginson, D. Robinson, and C. Walmsley.   1981.  Biochemical
       explanations of more than additive inhibitory effects of low
       atmospheric levels of sulfur dioxide plus nitrogen dioxide upon plants.
       New Phytologist 88:223-237.

White, K.L., A.C.  Hill, and J.H. Bennett.  1974. Synergistic inhibition of
       apparent photosynthesis rate of alfalfa by combinations of sulfur
       dioxide and nitrogen dioxide.  Environ. Sci.  Technol. 8:574-576.

Whitmore, M.E., P.H. Freer-Smith,  and T. Davis.  1982.  Some  effects of low
       concentrations of  S0~ and/or N0?  on the growth of grasses  and poplar.
       pp. 483-485.  In:  M.R. Unsworth and D.P. Ormrod  (eds.) Effects  of
       Gaseous Air PoTTution in Agriculture and Horticulture.  London/Boston:
       Butterworth Scientific.  532 pp.

Williams, R.J.H.,  and G.R. Ricks.  1975.  Effects of combinations of atmospheric
       pollutants  upon vegetation.  In:  M.J. Chadwick and G.T. Goodman (eds.)
       15th Symposium of  the British Ecological Society.  London: Blackwells
       Science Publication,  pp. 127-137.

Williams, R.J.H.,  M.M.  Lloyd, and  G.R.  Ricks.   1971. Effects of atmospheric
       pollution on deciduous woodland.   I. Some effects on  leaves of  Quercus
       petraea (Mattuschka)-  Leibl. Environ.  Pollut. 2:57-68.

Yamazoe,  F., and H. Mayumi.  1977. Vegetation  injury from interaction  of mixed
       air pollutants.   Proc. Inter. Clean Air. Congr.  Tokyo, Japan.

Zaragoza, L.J.  1982. The Use of Biological Information in the Development of
       National Ambient Air Quality Standards.  Available through University
       Microfilms  International.  Ann Arbor, MI.
                                          104

                                                           -'USGPO: 1984 — 759-102/0875

-------