c/EPA
United Stales
Environmental Protection
Agency
Environmental Criteria and
Assessmeni Office
Research Triangle Park NC 27711
EPA/600/8-84/020cF
August 1986
Research anu
Air Quality
Criteria for
Ozone and Other
Photochemical
Oxidants
Volume III of V
-------
EPA/600/8-84/020cF
August 1986
Air Quality Criteria
for Ozone and Other
Photochemical Oxidants
Volume III of V
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental
Protection Agency policy and approved for publication. Mention of trade names
or commercial products does not constitute endorsement or recommendation.
n
-------
ABSTRACT
Scientific information is presented and evaluated relative to the health
and welfare effects associated with exposure to ozone and other photochemical
oxidants. Although it is not intended as a complete and detailed literature
review, the document covers pertinent literature through early 1985.
Data on health and welfare effects are emphasized, but additional infor-
mation is provided for understanding the nature of the oxidant pollution pro-
blem and for evaluating the reliability of effects data as well as their
relevance to potential exposures to ozone and other oxidants at concentrations
occurring in ambient air. Information is presented on the following exposure-
related topics: nature, source, measurement, and concentrations of precursors
to ozone and other photochemical oxidants; the formation of ozone and other
photochemical oxidants and their transport once formed; the properties, chem-
istry, and measurement of ozone and other photochemical oxidants; and the
concentrations of ozone and other photochemical oxidants that are typically
found in ambient air.
The specific areas addressed by chapters on health and welfare effects
are the toxicological appraisal of effects of ozone and other oxidants; effects
observed in controlled human exposures; effects observed in field and epidemic-
logical studies; effects on vegetation seen in field and controlled exposures;
effects on natural and agroecosystems; and effects on nonbiological materials
observed in field and chamber studies.
n i
-------
AIR QUALITY CRITERIA FOR OZONE
AND OTHER PHOTOCHEMICAL OXIDANTS
VOLUME I
Chapter 1. Summary and Conclusions 1-1
VOLUME II
Chapter 2. Introduction 2-1
Chapter 3. Properties, Chemistry, and Transport of Ozone and
Other Photochemical Oxidants and Their Precursors ... 3-1
Chapter 4. Sampling and Measurement of Ozone and Other
Photochemical Oxidants and Their Precursors 4-1
Chapter 5. Concentrations of Ozone and Other Photochemical
Oxidants in Ambient Air 5-1
VOLUME III
Chapter 6. Effects of Ozone and Other Photochemical Oxidants
on Vegetation 6-1
Chapter 7. Effects of Ozone on Natural Ecosystems and Their '
Components 7-1
Chapter 8. Effects of Ozone and Other Photochemical Oxidants
on Nonbiological Materials 8-1
VOLUME IV
Chapter 9. Toxicological Effects of Ozone and Other
Photochemical Oxidants 9-1
VOLUME V
Chapter 10. Controlled Human Studies of the Effects of Ozone
and Other Photochemical Oxidants 10-1
Chapter 11. Field and Epidemiological Studies of the Effects
of Ozone and Other Photochemical Oxidants 11-1
Chapter 12. Evaluation of Health Effects Data for Ozone and
Other Photochemical Oxidants 12-1
iv
-------
TABLE OF CONTENTS
LIST OF TABLES x
LIST OF FIGURES x i i i
AUTHORS, CONTRIBUTORS, AND REVIEWERS xv
6. EFFECTS OF OZONE AND OTHER PHOTOCHEMICAL OXIDANTS ON VEGETATION . 6-1
6.1 INTRODUCTION 6-1
6.2 METHODOLOGIES USED IN VEGETATION EFFECTS RESEARCH 6-5
6,2.1 Experimental Design and Statistical Analysis 6-5
6.2.2 Exposure Characteristics 6-7
6.2.2.1 Statistics Used to Characterize Seasonal
Exposures 6-7
6.2.2.2 Statistics Used to Characterize Short Exposures . 6-9
6.2.2.3 Evaluation of Exposure Statistics 6-10
6.2.3 Exposure Systems 6-13
6.2.3.1 Laboratory Systems 6-14
6.2.3.2 Greenhouse Exposure Systems 6-14
6.2.3.3 Field Exposure Systems 6-15
6.2.4 Methodologies Used in The National Crop Loss
Assessment Network 6-17
6.2.5 Definitions of Yield Loss and Crop Loss 6-20
6.3 MODE OF ACTION OF OZONE ON PLANTS 6-20
6.3.1 Biochemical and Physiological Responses to Ozone 6-23
6.3.1.1 Gas-Phase Movement into the Leaf 6-23
6.3.1.2 Transition Between Gas-Phase and Liquid-Phase
Movement i nto the Cel 1 6-24
6.3.1.3 Chemical and Biochemical Responses 6-25
6.3.1.4 Physiological Responses 6-26
6.3.1.5 Tissue and Organ Responses 6-30
6.3.1.6 Secondary Metabolic Responses 6-31
6.3.2 Factors That Modify Plant Response 6-32
6.3.2.1 Biological Factors 6-32
6.3.2.2 Physical Factors 6-45
6.3.2.3 Chemical Factors 6-51
6.4 OZONE EXPOSURE AND RESPONSE 6-72
6.4.1 Bioindicators of Ozone Exposure 6-78
6.4.1.1 Bioindicator Methods 6-78
6.4.1.2 Response of Indicator Species 6-78
6.4.1.3 Bioindicator Systems 6-82
6.4.2 Response of Microorganisms and Nonvascular Plants to
Ozone 6-84
6.4.2.1 Microorganisms 6-84
6.4.2.2 Lichens, Mosses, and Ferns 6-88
6.4.3 Losses in Vascular Plants from Exposure to Ozone 6-90
6.4.3.1 Losses in Aesthetic Value and Foliar Yield 6-90
6.4.3.2 Yield Losses as Weight, Size, and Number 6-99
-------
TABLE OF CONTENTS (continued)
6.4.3.3 Exposure-Response Relationships (Empirical
Models) 6-151
6.5 ECONOMIC ASSESSMENTS OF EFFECTS OF OZONE ON AGRICULTURE 6-170
6.5.1 Economic Issues in Performing Assessments 6-171
6.5.2 Plant Science and Aerometric Issues in Performing
Economic Assessments 6-173
6.5.3 Assessment Methodologies Applied to Agriculture 6-176
6.5.4 Review of Economic Assessments of Effects of Ozone on
Agriculture 6-178
6.5.4.1 Review of Regional Assessments 6-179
6.5.4.2 Review of National Assessments 6-192
6.5.5 Overview of Current Economic Assessments of Ozone,on
Agriculture 6-201
6.6 MODE OF PEROXYACETYL NITRATE (PAN) ACTION ON PLANTS 6-204
6.6.1 Biochemical and Physiological Responses to PAN 6-205
6.6.1.1 Gas-Phase Movement Into the Leaf 6-205
6.6.1.2 Biochemical and Physiological Responses 6-207
6.6.2 Factors That Modify Plant Response to PAN 6-209
6.6.2.1 Biological Factors 6-209
6.6.2.2 Physical Factors 6-210
6.6.2.3 Chemical Factors 6-211
6.7 PAN EXPOSURE AND RESPONSE 6-213
6.7.1 Bioindicators of PAN Exposure 6-217
6.7.2 Nonvascular Plant Response to PAN Exposure 6-218
6.7.3 Losses in Vascular Plants Caused by PAN 6-219
6.7.3.1 Losses in Aesthetic Use and Foliar Yield 6-219
6.7.3.2 Losses Determined by PAN Addition Studies 6-220
6.8 SUMMARY 6-222
6.8.1 Limiting Values of Plant Response to Ozone 6-223
6.8.2 Methods for Determining 03 Yield Losses 6-225
6.8.3 Estimates of Ozone-Induced Yield Loss 6-227
6.8.3.1 Yield Loss: Determination by Regression
Analysis 6-228
6.8.3.2 Yield Loss: Determination from Discrete
Treatments 6-234
6.8.3.3 Yield Loss: Determination with Chemical
Protectants 6-234
6.8.3.4 Yield Loss: Determination from Ambient
Exposures 6-236
6.8.3.5 Yield Loss Summary 6-236
6.8.4 Effects on Crop Quality 6-239
6.8.5 Statistics Used to Characterize Ozone Exposures 6-239
6.8.6 Relationship Between Yield Loss and Foliar Injury .'.. 6-242
6.8.7 Physiological Basis of Yield Reductions 6-242
VI
-------
TABLE OF CONTENTS (continued)
6.8.8 Factors Affecting Plant Response to Ozone 6-243
6.8.8.1 Environmental Conditions 6-244
6.8.8.2 Interaction with Plant Diseases 6-245
6.8.8.3 Interaction of Ozone with Other Air Pollutants .. 6-245
6.8.9 Economic Assessment of Effects of Ozone on Agriculture ... 6-246
6.8.10 Effects of Peroxyacetyl Nitrate on Vegetation 6-257
6.8.10.1 Factors Affecting Plant Response to PAN 6-257
6.8.10.2 Limiting Values of Plant Response 6-257
6.8.10.3 Effects of PAN on Plant Yield 6-258
6.9 REFERENCES 6-258
APPENDIX 6A. Colloquial and Latin Names of Plants Discussed in
the Chapter 6A-1
APPENDIX 6B. Species That Have Been Exposed to Ozone to Determine
Differential Responses of Germplasm to Photochemical
Products 6B-1
7. EFFECTS OF OZONE ON NATURAL ECOSYSTEMS AND THEIR COMPONENTS 7-1
7.1 INTRODUCTION 7-1
7.2 CHARACTERISTICS OF ECOSYSTEMS 7-1
7.3 CHARACTERISTIC RESPONSES OF ECOSYSTEMS TO STRESS 7-3
7.4 EFFECTS OF OZONE ON PRIMARY PRODUCTION IN TERRESTRIAL
ECOSYSTEMS 7-5
7.4.1 Effects of Ozone on Growth of Producers 7-5
7.4.1.1 Controlled Studies on Growth of Trees 7-6
7.4.1.2 Field Studies on Effects of Ozone on Growth
of Trees in Natural Habitats 7-7
7.4.1.3 Controlled and Field Studies on Growth of
Other Native Vegetation 7-12
7.4.1.4 Mechanisms of Effects of Ozone on Growth of
Producers 7-12
7.4.2 Factors Modifying Effects of Ozone on Growth of
Producers 7-15
7.4.2.1 Genetic Factors 7-15
7.4.2.2 Other Factors 7-17
7.5 EFFECTS OF OZONE ON OTHER COMPONENTS AND INTERACTIONS IN
TERRESTRIAL ECOSYSTEMS 7-18
7.5.1 Producer-Producer Interactions: Competition and
Succession 7-19
7.5.2 Producer-Symbiont Interactions 7-24
7.5.2.1 Mycorrhizal-Plant Interactions 7-24
7.5.2.2 Bacterial-Plant Interactions 7-26
vn
-------
TABLE OF CONTENTS (continued)
7.5.3 Producer-Consumer Interactions 7-27
7.5.4 Producer-Decomposer and Producer-Pathogen Interactions ... 7-29
7.6 EFFECTS OF OZONE OR TOTAL OXIDANTS ON SPECIFIC FOREST
ECOSYSTEMS 7-32
7.6.1 The San Bernardino Mixed-Conifer Forest 7-32
7.6.2 The Blue Ridge Mountains of Virginia 7-41
7.7 RESPONSES OF OTHER ECOSYSTEMS TO OZONE 7-43
7.7.1 Responses of Native Vegetation 7-43
7.7.2 Managed Forest Ecosystems 7-47
7.7.3 Aquatic Ecosystems 7-47
7.8 ECONOMIC VALUATION OF ECOSYSTEMS 7-48
7.9 SUMMARY 7-50
7.9.1 Responses of Ecosystems to Ozone Stress 7-50
7.9.2 Effects of Ozone on Producers 7-51
7.9.3 Effects of Ozone on Other Ecosystem Components and on
Ecosystem Interactions 7-53
7.9.4 Effects of Ozone on Specific Ecosystems 7-54
7.9.5 Economic Valuation of Ecosystems 7-56
7.10 REFERENCES 7-57
8. EFFECTS OF OZONE AND OTHER PHOTOCHEMICAL OXIDANTS ON
NONBIOLOGICAL MATERIALS 8-1
8.1 INTRODUCTION- 8-1
8.2 MECHANISMS OF OZONE ATTACK AND ANTIOZONANT PROTECTION 8-2
8.2.1 Elastomers 8-2
8.2.2 Textile Fibers and Dyes 8-7
8.2.3 Paint 8-9
8.2.4 Other Materials 8-9
8.3 DOSE-RESPONSE DATA 8-10
8.3.1 Elastomer Cracking 8-11
8.3.2 Dye Fading 8-20
8.3.3 Fiber Damage 8-31
8.3.4 Paint Damage 8-35
8.4 ECONOMICS 8-38
8.4.1 Introduction 8-38
8.4.2 Methods of Cost Classification and Estimation 8-39
8.4.3 Aggregate Cost Estimates 8-41
8.4.4 Damage to Elastomers . 8-43
8.4.5 Damage to Fibers and Dyes 8-46
8.4.6 Damage to Paint 8-46
8.5 SUMMARY AND CONCLUSIONS 8-47
vrn
-------
TABLE OF CONTENTS (continued)
Page
8.6 REFERENCES 8-53
APPENDIX 8A: CHEMICAL ABBREVIATIONS USED IN THE TEXT AND COMPOUND
DETAILS 8A-1
-------
LIST OF TABLES
Table Page
6-1 Effect of ozone on photosynthesis 6-28
6-2 Plant and biotic pathogen interactions as influenced by various
doses of ozone under laboratory and field conditions 6-36
6-3 Response of plants to ozone as conditioned by humidity during
growth and exposure 6-48
6-4 Effects of soil moisture on response of selected plants to
oxi dant 6-49
6-5 Summary of effects of sulfur dioxide and ozone mixtures on
foliar injury 6-53
6-6 Foliar injury response of various plant species to ozone and
ozone plus sulfur dioxide 6-55
6-7 Growth response of various plant species to ozone and ozone
plus sulfur dioxide 6-56
6-8 Yield responses of various plant species to ozone and ozone
plus sulfur dioxide 6-57
6-9 Influence of mixtures of ozone and sulfur dioxide on soybean
yield 6-58
6-10 Yield responses of selected tree species to ozone plus nitrogen
dioxide 6-62
6-11 Yield change in various plant species exposed to ozone,
sulfur dioxide, and nitrogen dioxide 6-63
6-12 Effects of nitrogen dioxide in combination with sulfur dioxide
or ozone, or both, on radish root fresh weight at three
concentrations of each gas 6-66
6-13 Protection of plants from oxidant injury by application of
protective chemicals 6-69
6-14 Concentration, time, and response equations for three suscepti-
bility groups and for selected plants or plant types with
respect to ozone 6-75
6-15 Partial listing of states where ambient ozone injury has
been observed on sensitive vegetation 6-85
6-16 Foliar symptom expression of various flower, ornamental tree,
shrub, turfgrass, and foliar crop species in response to ozone
exposure 6-91
6-17 Effects of short-term exposures on growth and yield of selected
plants 6-100
6-18 Effects of long-term, controlled exposures on growth, yield,
and foliar injury in selected plants 6-101
6-19 Estimates of the parameters for fitting the Weibull model
using the 7-hr seasonal mean ozone concentrations 6-106
6-20 Effects of ozone added to ambient air in open-top chambers on
the yield of selected crops 6-112
6-21 Effects of ozone added to filtered air in field chambers on the
yield of selected crops 6-126
6-22 Effects of ozone added to filtered air on the yield of selected
crops 6-129
6-23 Effects of ozone added to filtered air on the yield of selected
tree crops 6-134
-------
LIST OF TABLES (continued)
Table Page
6-24 Effects of oxidants (ozone) in ambient air on growth, yield,
and foliar injury in selected plants 6-144
6-25 Effects of ambient air in open-top chambers, outdoor CSTR
chambers, or greenhouses on growth and yield of selected crops ... 6-145
6-26 Exposure-response functions relating ozone dose to plant yield ... 6-148
6-27 Summary of models describing the relationship between foliar
injury and ozone exposure 6-153
6-28 Summary of models of ozone yield and loss 6-156
6-29 Summary of crop-loss models 6-160
6-30 Comparison of predicted 7-hr seasonal mean ozone
concentrations using various yield reduction models 6-166
6-31 Summary of estimates of regional economic consequences
of ozone pollution 6-180
6-32 Summary of estimates of national economic consequences
of ozone pollution 6-193
6-33 Geographic occurrence of PAN/oxidant injury on plants 6-206
6-34 Ozone concentrations for short-term exposures that produce
5 or 20 percent injury to vegetation grown under sensitive
conditions 6-224
6-35 Summary of ozone concentrations predicted to cause 10 percent
and 30 percent yield losses and summary of yield losses
predicted to occur at 7-hr seasonal mean ozone concentrations
of 0.04 and 0.06 ppm 6-232
6-36 Ozone concentrations at which significant yield losses
have been noted for a variety of plant species exposed under
various experimental conditions 6-235
6-37 Effects of ozone on crop yield as determined by the use of
chemical protectants 6-237
6-38 Effects of ambient oxidants on yield of selected crops 6-238
6-39 Summary of estimates of regional economic consequences of
ozone pollution 6-248
6-40 Summary of estimates of national economic consequences of
ozone pol 1 ution 6-253
7-1 Continuum of characteristic ecosystem responses to pollutant
stress 7-4
7-2 Annual mean radial growth increment, 1955 through 1978,
for three ozone sensitivity classes of native eastern white
pines (Pinus strobus L.) growing in ten plots of three trees
each in the Blue Ridge Mountains in Virginia 7-9
7-3 Peak hourly ozone concentrations in episodes recorded at three
monitoring sites in western Virginia, spring and summer 1979
through 1982 7-10
7-4 Monthly 8-hr average (11:00 a.m.-6:00 p.m. EST), monthly
average of peak 1-hr, and cumulative seasonal ozone doses
monitored at Big Meadows, Shenandoah National Park, Virginia,
during 1979-1981 7-42
7-5 Injury thresholds for 2-hr exposures to ozone 7-46
-------
LIST OF TABLES (continued)
Table Page
8-1 Tire industry exposure tests 8-12
8-2 Effects of ozone on different SBR polymers containing
various antiozonant concentrations 8-15
8-3 Cracking rates of white sidewall tire specimens 8-15
8-4 Protection of tested rubber materials 8-17
8-5 Effect of ozone and humidity on interply adhesion 8-19
8-6 Effect of antiozonants, antioxidants, and fast-blooming
waxes on interply adhesion in natural rubber 8-20
8-7 Dose-response studies on effects of ozone on
elastomers 8-21
8-8 Colorfastness of test samples compared with colorfastness
of in-use carpeting 8-29
8-9 Laboratory studies of the effects of ozone on dye fading 8-32
8-10 Summary of damage costs to materials by oxidants 8-41
8-11 Summary of damage costs to rubber by ozone 8-45
xn
-------
LIST OF FIGURES
Figure
Page
6-1 Conceptual sequence of ozone-induced responses 6-3
6-2 Schematic cross section of a typical dicot leaf 6-21
6-3 Limiting values for foliar injury to trees and shrubs by
ozone 6-73
6-4 Limiting values for foliar injury to agricultural crops by
ozone 6-73
6-5 Ozone concentrations versus duration of exposure required to
produce 5 percent foliar injury in plants of three different
sensitivity groupings 6-74
6-6 Relationship between ozone concentration, exposure duration,
and reduction in plant growth or yield 6-77
6-7 Ozone injury to Bel W-3 tobacco 6-80
6-8 Schematic cross section of typical dicot leaf showing ozone
injury to palisade cells and collapsed epidermal cells 6-80
6-9 Ozone injury to oats 6-81
6-10 Ozone injury to conifer needles 6-81
6-11 States in which ozone-induced injury to vegetation has
occurred as reported in the published literature 6-87
6-12 Effects of ozone on the yield of four soybean cultivars 6-108
6-13 Comparison of the effects of ozone on soybean yields on the
same cultivars exposed for successive years at two locations .... 6-109
6-14 Effects of ozone on the yield of peanut and kidney bean 6-110
6-15 Effects of ozone on the yield of four wheat cultivars 6-114
6-16 Comparison of the effects of ozone on yields of the same
wheat cultivars exposed for successive years at Argonne, IL 6-115
6-17 Effects of ozone on the yield of corn (two cultivars),
sorghum, and barley 6-117
6-18 Effects of ozone on the yield of two cotton cultivars grown
at two locations 6-119
6-19 Effects of ozone on the yield of tomatoes and lettuce 6-120
6-20 Effects of ozone on the yield of spinach and turnip
cultivars 6-122
6-21 Effects of ozone on the yields of several legume species 6-163
6-22 Effects of ozone on the yield of several crops 6-164
6-23 PAN injury 6-214
6-24 Dose-response relationships and limiting values for foliar
injury to vegetation by peroxyacetylnitrate (PAN) 6-216
6-25 Relationship between ozone concentration, exposure duration,
and reduction in plant growth or yield 6-226
6-26 Examples of the effects of ozone on the yield of soybean and
wheat cul tivars 6-229
6-27 Examples of the effects of ozone on the yield of cotton,
tomato, and turnip 6-230
6-28 Number and percentage of 37 crop species or cultivars
predicted to show a 10 percent yield loss at various ranges
of 7-hr seasonal mean ozone concentrations 6-233
xm
-------
LIST OF FIGURES
Figure Page
7-1 Pathways of air pollutant impact in the forest ecosystems 7-20
7-2 Total oxidant concentrations at Rim Forest (5640 ft.) in
southern California during May through September, 1968
through 1972 7-34
7-3 Annual trends of oxidant dose, precipitation, and air temperature
near the Lake Arrowhead-Sky Forest region of the San Bernardino
National Forest, California 7-38
8-1 Postulated mechanism for damage to elastomers by oxygen 8-4
8-2 Postulated mechanism for damage to elastomers by ozone 8-5
8-3 Reaction of anthraquinone dyes with ozone and with nitrogen
oxides 8-8
8-4 Relationship of cracking in rubber and ozone concentration 8-13
8-5 Relaxation of rubber compounds in ozone 8-18
8-6 Effects of relative humidity (RH) on fading of C.I. Disperse
Blue 3 (CIDB-3) in Nylon 6 after exposure to 0.2 ppm of ozone 8-30
xiv
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Chapter 6: Effects of Ozone and Other Photochemical Oxidants on Vegetation
Principal Authors
Or. Richard M. Adams
Department of Agricultural and Resource Economics
Oregon State University
Corvallis, OR 97331
Dr. J. H. B. Garner
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Lance W, Kress
RER Division
Argonne National Laboratories
9700 Cass Avenue
Argonne, IL 60439
Dr. John A. Laurence
Boyce Thompson Institute for Plant Research
Tower Road
Cornell University
Ithaca, NY 14853
Mr. Ronald Oshima
State of California
Department of Food and Agriculture
Sacramento, CA 95814
Dr. Eva Pell
Department of Plant Pathology
Penn State University
University Park, PA 16802
Dr. Richard Reinert
Department of Plant Pathology
North Carolina State University
Raleigh, NC 27607
Dr. 0. C. Taylor
Department of Botany and Plant Sciences
University of California
Riverside, CA 92521
Dr. David T. Tingey
Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97330
xv
-------
Contributing Authors
Ms. Marcia Gumpertz
Northrop Services Institute
Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97330
Dr. David S. Lang
Minnesota Environmental Quality Board
100 Capitol Square Bldg.
St. Paul, MN 55101
Reviewers
Authors also reviewed individual sections of the chapter. The following
additional persons reviewed this chapter at the request of the U.S. Environmen-
tal Protection Agency. The evaluations and conclusions contained herein,
however, are not necessarily those of the reviewers.
Dr. James Bennett
National Park Service
Air Quality Division
Box 25287
Denver, CO 80225
Dr. Harris M. Benedict
P.O. Box 50046
Pasadena, CA 91105
Dr. Donald D. Davis
Department of Plant Pathology and Center for Air Environment Studies
211 Buckhout Laboratory
University Park, PA 16802
Dr. Robert L. Heath
Department of Botany and Plant Sciences
University of California
Riverside, CA 92521
Dr. Allen Heagle
Field Air Quality Laboratory
North Carolina State University
Raleigh, NC 27606
Dr. Walter W. Heck
Department of Botany
North Carolina State University
Raleigh, NC 27606
xvi
-------
Reviewers (cont1d.)
Dr. Howard Heggestad
U.S. Department of Agriculture
ARS Beltsville Agricultural Research Center
Beltsville, MD 20205
Ms. Pamela Johnson
Office of Air Quality Planning and Standards
MD-12
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Robert Kohut
Boyce Thompson Institute for Plant Research
Tower Road
Cornell University
Ithaca, NY 14853
Dr. Jan G. Laarman
Department of Forestry
North Carolina State University
Raleigh, NC 27607
Dr. Allen S. Lefohn
A.S.L. & Associates
111 North Last Chance Gulch
Helena, MT 59601
Dr. David J. McKee
Office of Air Quality Planning and Standards
MD-12
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Laurence D. Moore
Department of Plant Pathology, Physiology and Weed Science
Virginia Polytechnic Institute and State University
Blacksburg, VA 24061
Dr. John M. Skelly
Department of Plant Pathology
Pennsylvania State University
University Park, PA 16802
Dr. Boyd Strain
Department of Botany
Duke University
Durham, NC 27705
Mr. Robert Stricter
American Petroleum Institute
The Medicine and Biological Science Department
2101 L. Street, NW
Washington, DC 20037
xv ii
-------
Reviewers (cont'd.)
Dr. Paul Teng
Department of Plant Pathology
University of Minnesota
St. Paul, MN 55108
Ms. Beverly Til ton
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Tom Walton
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Wayne T. Williams
The Black Apple Institute
920 St. Francis Drive
Petaluma, CA 94952
Dr. F. A. Wood
Institute of Food and Agricultural Science
1022 McCarty Hall
University of Florida
Gainesville, FL 32611
Mr. Laurence Zaragoza
Office of Air Quality Planning and Standards
MD-12
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
xvm
-------
Chapter 7: Effects of Ozone and Other Photochemical Oxidants
on Natural and Agroecosystems
Principal Authors
Dr. J. H. B. Garner
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. William H. Smith
School of Forestry and Environmental Studies
116 Greeley Memorial Laboratory
Yale University
370 Prospect Street
New Haven, CT 06511
Contributing Author
Ms. Beverly E. Tilton
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Reviewers
Authors of Chapter 7 also reviewed this chapter. The following additional
persons reviewed this chapter at the request of the U.S. Environmental Protec-
tion Agency. The evaluations and conclusions contained herein, however, are
not necessarily those of the reviewers.
Dr. James Bennett
National Park Service
Box 25287
Denver, CO 80225
Dr. Michael A. Berry
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Norman E. Chi Ids
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
xix
-------
Reviewers (cont'd.)
Dr. Donald D. Davis
Department of Plant Pathology and Center for Air Environment Studies
211 Buckhout Laboratories
University Park, PA 16802
Dr. Robert L. Heath
Department of Botany and Plant Sciences
University of California
Riverside, CA 92521
Dr. Allen Heagle
Field Air Quality Laboratory
North Carolina State University
Raleigh, NC 27606
Dr. Walter W. Heck
Department of Botany
North Carolina State University
Raleigh, NC 27606
Dr. Robert Kohut
Boyce Thompson Institute for Plant Research
Tower Road
Cornell University
Ithaca, NY 14853
Dr. Allen S. Lefohn
A.S.L. & Associates
111 North Last Chance Gulch
Helena, MT 59601
Dr. Robert Mclntosh
Department of Biology
University of Notre Dame
Notre Dame, IN 46556
Dr. Paul Miller
U.S. Forest Service Fire Laboratory
4955 Canyon Crest Drive
Riverside, CA 92507
Dr. Laurence D. Moore
Department of Plant Pathology, Physiology and Weed Science
Virginia Polytechnic Institute and State University
Blacksburg, VA 24061
Dr. Eric Preston
Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97330
xx
-------
Reviewers (cont1d.)
Dr. William H. Schlesinger
Department of Botany
Duke University
Durham, NC 27706
Dr. Boyd Strain
Department of Botany
Duke University
Durham, NC 27705
Dr. John M. Skelly
Department of Plant Pathology
Pennsylvania State University
University Park, PA 16802
Mr. Robert Strieter
American Petroleum Institute
The Medicine and Biological Science Department
2101 L. Street, NW
Washington, DC 20037
Ms. Beverly Til ton
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. David T. Tingey
Environmental Research Laboratory
200 SW 35th Street
Con/all is, OR 97330
Dr. David Weinstein
Ecosystems Research Center
Corson Hall
Cornell University
Ithaca, NY 14853
xxi
-------
Chapter 8: Effects of Ozone and Other Photochemical Oxidants
on Nonbiological Materials
Principal Authors
Mr. James M. Kawecki
TRC Environmental Consultants
2121 Wisconsin Avenue, N.W.
Suite 220
Washington, DC 20007
Dr. Jan G. Laarman
Department of Forestry
North Carolina State University
Raleigh, NC 27607
Dr. Alexander R. Stankunas
930 Rockdale Dr.
San Jose, CA 95129
The following individuals reveiwed this chapter at the request of the U.S.
Environmental Protection Agency. The evaluations and conclusions contained
herein, however, are not necessarily those of the reviewers.
Dr. Richard Adams
Department of Agricultural and Resource Economics
Oregon State University
Corvallis, OR 97331
Ms. F. Vandiver Bradow
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. J.H.B. Garner
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Fred Haynie
Environmental Sciences Research Laboratory
MD-84
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Bruce Jarvis
Department of Chemistry
University of Maryland
College Park, MD
xxn
-------
Reviewers (cont'd.)
Mr. E. J. McCarthy
TRC Environmental Consultants
2121 Wisconsin Avenue, N.W.
Suite 220
Washington, DC 20007
Dr. James McNesby
Department of Chemistry
University of Maryland
College Park, MD
Mr. John Spence
Environmental Sciences Research Laboratory
MD-84
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Ms. Beverly Tilton
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Tom Walton
Office of Air Quality Planning and Standards
MD-12
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
XXIll
-------
SCIENCE ADVISORY BOARD
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE
The substance of this document was reviewed by the Clean Air Scientific
Advisory Committee of the Science Advisory Board in public sessions.
SUBCOMMITTEE ON OZONE
Chairman
Dr. Morton Lippmann
Professor
Department of Environmental Medicine
New York University Medical Center
Tuxedo, New York 10987
Members
Dr. Mary 0. Amdur
Senior Research Scientist
Energy Laboratory
Massachusetts Institute of Technology
Cambridge, Massachusetts 02139
Dr. Eileen G. Brennan
Professor
Department of Plant Pathology
Martin Hall, Room 213, Lipman Drive
Cook College-NJAES
Rutgers University
New Brunswick, New Jersey 08903
Dr. Edward D. Crandall
Professor of Medicine
School of Medicine
Cornell University
New York, New York 10021
Dr. James D. Crapo
Associate Professor of Medicine
Chief, Division of Allergy, Critical
Care and Respiratory Medicine
Duke University Medical Center
Durham, North Carolina 27710
Dr. Robert Frank
Professor of Environmental Health
Sciences
Johns Hopkins School of Hygiene
and Public Health
615 N. Wolfe Street
Baltimore, Maryland 21205
Professor A. Myrick Freeman II
Department of Economics
Bowdoin College
Brunswick, Maine 04011
Dr. Ronald J. Hall
Senior Research Scientist and Leader
Aquatic and Terrestrial Ecosystems
Section
Ontario Ministry of the Environment
Dorset Research Center
Dorset, Ontario POA1EO
Dr. Jay S. Jacobson
Plant Physiologist
Boyce Thompson Institute
Tower Road
Ithaca, New York 14853
XXIV
-------
Dr. Warren B. Johnson
Director, Atmospheric Science Center
SRI International
333 Ravenswood Avenue
Menlo Park, California 94025
Dr. Jane Q. Koenig
Research Associate Professor
Department of Environmental Health
University of Washington
Seattle, Washington 98195
Dr. Paul Kotin
Adjunct Professor of Pathology
University of Colorado Medical School
4505 S. Yosemite, #339
Denver, Colorado 80237
Dr. Timothy Larson
Associate Professor
Environmental Engineering and
Science Program
Department of Civil Engineering
University of Washington
Seattle, Washington 98195
Professor M. Granger Morgan
Head, Department of Engineering
and Public Policy
Carnegie-Mellon University
Pittsburgh, Pennsylvania 15253
Dr. D. Warner North
Principal
Decision Focus Inc., Los Altos
Office Center, Suite 200
4984 El Camino Real
Los Altos, California 94022
Dr. Robert D. Rowe
Vice President, Environmental and
Resource Economics
Energy and Resources Consultants, Inc.
207 Canyon Boulevard
Boulder, Colorado 80302
Dr. George Taylor
Environmental Sciences Division
P.O. Box X
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831
Dr. Michael Treshow
Professor
Department of Biology
University of Utah
Salt Lake City, Utah 84112
Dr. Mark J. Utell
Co-Director, Pulmonary Disease Unit
Associate Professor of Medicine and
Toxicology in Radiation Biology
and Biophysics
University of Rochester Medical
Center
Rochester, New York 14642
Dr. James H. Ware
Associate Professor
Harvard School of Public Health
Department of Biostatisties
677 Huntington Avenue
Boston Massachusetts 02115
Dr. Jerry Wesolowski
Air and Industrial Hygiene Laboratory
California Department of Health
2151 Berkeley Way
Berkeley, California 94704
Dr. James L. Whittenberger
Director, University of California
Southern Occupational Health Center
Professor and Chair, Department of
Community and Environmental Medicine
California College of Medicine
University of California - Irvine
19772 MacArthur Boulevard
Irvine, California 92717
Dr. George T. Wolff
Senior Staff Research Scientist
General Motors Research Labs
Environmental Science Department
Warren, Michigan 48090
xxv
-------
PROJECT TEAM FOR DEVELOPMENT
OF
Air Quality Criteria for Ozone and Other Photochemical Oxidants
Ms. Beverly E. Tilton, Project Manager
and Coordinator for Chapters 1 through 5, Volumes I and II
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Norman E. Chi Ids
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. J.H.B. Garner
Coordinator for Chapters 7 and 8, Volume III
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Thomas B. McMullen
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. James A. Raub
Coordinator for Chapters 10 through 13, Volumes IV and V
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. David T. Tingey
Coordinator for Chapter 7, Volume III
Environmental Research Laboratory
U.S. Environmental Protection Agency
200 S.W. 35th Street
Corvallis, OR 97330
XXVI
-------
6. EFFECTS OF OZONE AND OTHER PHOTOCHEMICAL OXIDANTS ON VEGETATION
6.1 INTRODUCTION
An analysis of photochemical oxidants in the ambient air has revealed the
presence of a number of phytotoxic compounds, including 03, peroxyacyl nitrates,
and NOp. Ozone, the most prevalent photochemical oxidant, has received the most
study and its effects are better understood than the effects of other photo-
chemical ly derived oxidants. Ozone affects vegetation throughout the United
States, impairing crops, native vegetation, and ecosystems more than any other
air pollutant (Heck et al., 1980). The phytotoxicity of nitrogen oxides has
been assessed in Air Quality Criteria for Oxides of Nitrogen (U.S. Environmental
Protection Agency, 1982) and will not be discussed here. On a concentration
basis, the peroxyacyl nitrates are more toxic than 0^, with PAN being about
tenfold more phytotoxic than 0, (Darley et al., 1963; Taylor and MacLean, 1970;
Pell, 1976). Although more phytotoxic than 0.,, the peroxyacyl nitrates gen-
erally occur at significantly lower ambient concentrations, however, and phy-
totoxic concentrations are therefore less widely distributed than those of 03.
Ambient concentrations of 0, and PAN, as well as their concentration ratios,
are discussed in detail in Chapter 5.
The effects of photochemical oxidants were first observed as foliar in-
jury on vegetation growing in localized areas in Los Angeles County, California
(Middleton et al., 1950). In these early reports, foliar injury was described
as glazing, silvering, and bronzing of the lower leaf surface of leafy vege-
tables and as transverse bands of injury on monocotyledonous species. Subse-
quent studies showed that these symptoms of photochemical oxidant injury were
caused by peroxyacetyl nitrate (Taylor et al., 1960). The characteristic 0.,
stipple on grape leaves reported in the late 1950s was the first observation
of Oo injury to vegetation in the field (Richards et al., 1958). Subsequent
studies with tobacco and other crops confirmed that 0- was injuring vegetation
at sites near urban centers (Heggestad and Middleton, 1959; Daines et al.,
1960). It is now recognized that vegetation at rural sites may be injured by
03 transported long distances from urban centers (Edinger et al., 1972; Heck
et al., 1969; Heck and Heagle, 1970; Kelleher and Feder, 1978; Miller et al.,
1972; Skelly et al., 1977; Skelly, 1980; see also Chapters 3 and 5).,
The effects of 0- and PAN on terrestrial vegetation may be envisioned as
occurring at several levels, ranging from the molecular to the organismal, and
6-1
-------
then to the ecosystem level (Figure 6-1). The occurrence and magnitude of the
vegetational effects depend on the concentration of the pollutant, the duration
of the exposure, the length of time between exposures, and the various environ-
mental and biological factors that influence the response. Some of the earliest
observable physiological effects include altered membrane permeability, decreased
carbon dioxide fixation (photosynthesis), and altered stomatal responses.
These initial physiological changes are followed by inactivation or activation,
or both, of specific enzymes, changes in metabolite pools, and alterations in
the translocation of photosynthate. Biochemical changes within the plants are
often expressed as visible foliar injury, premature senescence, increased leaf
abscission, and reduced plant growth and yield. These changes at the individual
plant level lead to altered reproduction, changes in competitive ability, or
reduction of plant vigor. They subsequently are manifested by changes in
plant communities and, ultimately, change in ecosystems. The sequence of
topics in this chapter, which describes the effects of photochemical oxidants
on plants, is based on the logical hierarchical ordering of plant responses
depicted in Figure 6-1. The complexities of the entire subject are apparent
in the sections on factors affecting plant response and on exposure-response
relationships. Effects on terrestrial ecosystems are discussed in Chapter 7.
The linkages among altered biochemical processes, foliar injury, and
reduced plant yield are not well understood. Likewise, no clear relationship
exists between foliar injury and reduced plant yield for species in which the
foliage is not part of the yield. The previous criteria document (U.S. Environ-
mental Protection Agency, 1978) focused primarily on the effects of CL on
physiological processes, foliar injury, and plant growth and attempted to
summarize the literature by presenting limiting values (i.e., those concentra-
tions below which foliar injury and, presumably, reduced growth and yield
would not occur). In this document, the results of previous work on the
effects of photochemical oxidants on physiological processes and on foliar
injury and growth will be briefly summarized, with major emphasis placed on
the effects of these oxidants on the intended use of the plant. Such effects
are those that have impact on yield, quality, and aesthetic value.
The number of scientific reports on the effects of photochemical oxidants
on vegetation has increased rapidly since the early 1960. In reviewing this
extensive literature for the present revision, key references were selected
for in-depth examination. For the most part, materials selected for review
6-2
-------
PRIMARY SECONDARY TERTIARY QUATERNARY
CHANGES IN PLANT
COMMUNITIES
AND ECOSYSTEMS
REDUCED PLANT GROWTH
REDUCED PLANT YIELD
ALTERED PRODUCT QUALITY
LOSS OF PLANT VIGOR
ALTERED ENZYME ACTIVITIES
ALTERED METABOLIC POOLS
ALTERED TRANSLOCATION
REDUCED PHOTOSYNTHESIS
INCREASED MEMBRANE PERMEABILITY
Figure 6-1. Conceptual sequence of ozone-induced responses.
Source: U.S. Environmental Protection Agency (1978).
6-3
-------
were publications that have appeared since the preparation of the 1978 criteria
document. Earlier information considered fundamentally important is discussed
and related to more recent studies. All data that relate exposure-response
information to yield loss or crop loss were drawn directly from primary refer-
ences, regardless of their citation in the 1978 criteria document. In this
revision, crop loss refers to economic loss and yield loss refers to reductions
in the quality, quantity, aesthetic value, or intended use of the crop.
Generally, only published materials that have undergone scientific review have
been cited.
Emphasis has been given to those studies in which the pollutant concentra-
tions used were similar to those that occur in the ambient air of the United
States. Therefore, studies in which the lowest concentrations of 03 or PAN
exceeded 1.0 ppm or 200 ppb, respectively, were not included unless the paper
contained unique data, e.g., documentation of a mechanism involved in a specific
response. In addition, in discussions on exposure-response data for the
effects of 03 and PAN on plant yield, the primary emphasis has been given to
those studies reporting effects at concentrations below 0.25 ppm for 0- and 40
ppb for PAN. These units have been used in the majority of the vegetational
studies cited; conversion factors are: 1 ppm 0, = 1960 ug/m and 1 ppm PAN =
3
4947 |jg/m . The scientific names of the plants cited in this chapter are
listed in Appendix A.
Data used in the development of this chapter were derived from a diverse
range of studies that were conducted to determine the effects of 0- and PAN on
various plant species and to characterize plant responses. The studies used a
range of plant species and various experimental conditions and methodologies.
Most important, it should be noted here that the studies cited were generally
conducted to test specific biological hypotheses or to produce specific biologi-
cal data rather than to develop air quality criteria.
In this chapter, the general methodologies used in studies of air pollution
effects are discussed first, to provide a basis for understanding the methods,
approaches, and experimental designs used in the studies presented later.
Ozone and PAN are discussed separately, but the material presented for each
will follow the same general outline, which includes (1) mode of action of the
pollutant; (2) physical, biological, and chemical factors that alter plant
response; and, the topic given primary emphasis, (3) the responses of plants
exposed to various concentrations for various durations.
6-4
-------
6.2 METHODOLOGIES USED IN VEGETATION EFFECTS RESEARCH
This section provides reference information for understanding better the
studies discussed in the remainder of this chapter. The section contains an
evaluation of exposure methods, a discussion of the strengths and limitations
of various experimental designs and of the statistics used to represent pollutant
exposures, and a discussion of the definitions of yield and crop loss. These'
discussions emphasize the methodologies used in studies cited in the chapter
and do not reflect a general review of scientific literature. Changes in 0,
monitoring techniques, methods of calibration, quality assurance procedures,
and their possible impacts on measured 0- concentrations are discussed in
Chapter 4.
6.2.1 Experimental Design and Statistical Analysis
The selection of an appropriate experimental design for specific objec-
tives is a critical step in determining the success of a study and the appli-
cation of the results. The number and kind of factors controlled, the patterns
of randomization, and the number of replicates used in an experiment determine
what treatment comparisons may be made, whether trends can be plotted and
curves fitted, the precision of estimates, and the range of conditions over
which inferences may be made. An experimental design focuses an experiment on
its specific objectives, but in doing so, limits the application of the results.
No experimental design has universal application.
Most experiments are of traditional designs amenable to the analysis of
variance, such as randomized-block and split-plot designs. When used in
conjunction with treatment mean separation techniques, these designs produce
descriptive results that allow comparison of different treatments. There are
many different treatment mean separation techniques available, such as Tukey's
paired comparison procedure, Duncan's multiple range test, and Dunnett's test
for comparing several treatments with a control. The tests all give slightly
different results and have different powers. Some statisticians recommend
careful inspection of the treatment averages in relation to a reference distri-
bution in addition to or in place of formal multiple comparisons (Box. et al.,
1978). Few studies have attempted to partition interactions or to analyze
slope and curvature trends. In factorial experiments with more than two factors,
it has often been difficult to interpret the interactions fully.
Regression analyses are useful for many objectives, including the develop-
ment of empirical models. Care must be taken, however, to ensure that there
6-5
-------
is no systematic deviation of the model from the observed data and to recognize
that, in general, results cannot be extrapolated beyond the range of pollutant
(e.g., ozone) concentrations used to construct the model. Both model valida-
tion (the testing of model fit to the experimental data) and applications
validation (testing the application of the model to a new population) are
appropriate precursors to model use.
In an experiment in which quantitative treatments are used and the treat-
ments have been replicated, both analysis of variance and regression analysis
may be used to analyze the data. The traditional approach is to use analysis
of variance to estimate the error variance and to determine whether there are
any differences among treatments; and then to break down the treatment effect
into regression components to test whether there are any linear or quadratic
trends as the treatment level changes (Cochran and Cox, 1957; Anderson and
McLean, 1974). This is equivalent to doing analysis of variance followed by
regression analysis. If a linear or quadratic equation does not fit the data
well, or if there is a theorized functional relationship between treatment and
response, nonlinear models may be fitted to the data at this point. Because.
each mathematical function can assume only a limited range of shapes, it is
important to check for systematic lack of fit of the data. Confidence limits
for regression curves can show the variability of the fitted curves. Confi-
dence limits are frequently omitted from research papers, however, because
their computation is complicated and it is difficult to show more than one
curve in a figure if confidence limits are included. When confidence limits
are not provided but results from similar experiments are available, the
reader can obtain an idea of the variability of estimates by looking at the
distribution of estimates from similar experiments. This variability encom-
passes sources of error beyond a single experiment.
In most of the papers cited in this document, confidence limits for
exposure response curves were not provided. To compare the predictions of
different exposure-response models, the 0^ concentrations that would cause 10
and 30 percent yield losses were calculated (see Section 6.4.3). These predic-
ted concentrations also provide an indication of the relative sensitivity of
the crop cultivar to 0,. For more sensitive plants, the 10 and 30 percent
yield losses would be predicted to occur at lower concentrations. Therefore,
a table of estimates from regression models of the 0, concentration at which a
10 and 30 percent yield loss would occur for all the cultivars and species
studied is included in the summary so that the reader can examine the range of
6-6
-------
estimates. On each graph the fitted curve is given, and generally the treat-
ment means are also plotted. Where more than one model was fitted to the
data, the reader may compare the results from the various models and observe
whether there is a systematic lack of fit between the data and the curve. If
a deviation is observed, the 0, estimates may be biased.
The regression curves used in this document have either been calculated
from the original observations or from treatment means. This distinction is
noted in the figure legends whenever the method used is known. If the treat-
ment means are used rather than the original observations in a linear regres-
sion and there are equal numbers of observations in each treatment, the results
will be as follows: (1) the regression coefficients and estimated values will
be the same as if individual points had been used; (2) the coefficient of
2 2
variation (R ) will be greater than or equal to the R from individual points;
and (3) the variance of the regression coefficients will be about the same as
that computed from individual points if the variation of the means around the
line is similar to the variation of individual points around the treatment
means.
6.2.2 Exposure Characteristics
The occurrence of pollutants in the ambient air is influenced by many
variables (see Chapters 3 and 5). Periods of significant air pollutant episodes
occur when meteorological conditions, pollutant precursors, and other environ-
mental conditions coincide. Ozone and PAN episodes usually occur during the
plant-growth season (Chapter 5). The episodes may vary in duration from one
to several days and occur at varying times of the day (Chapter 5). Research
has not yet clearly defined which components of an exposure are most important
in causing vegetation responses. The characterization and representation of
plant exposures to air pollutants has been and continues to be a major problem.
An appropriate summary statistic for one exposure duration usually cannot be
easily transformed to describe a different exposure duration without, access to
the original aerometric data. In addition, statistics used to represent
extremely short exposures cannot be readily aggregated to provide a represen-
tative summary statistic for plant responses resulting from an extended exposure
(for example, a growing season).
6.2.2.1 Statistics Used to Characterize Seasonal Exposures. To define the
problems associated with characterization and representation of plant exposures
6-7
-------
necessitates consideration of the temporal resolution required. When plant
yield is considered, the ultimate impact of an air pollutant on yield depends
on the integrated impact of the pollutant exposures during the growth of the
plant. In this case, the temporal unit of interest becomes the plant growing
season, which varies with the geographic location, plant species, and cultivar
of interest. This period may be as short as 3 to 4 weeks for a crop such as
radish or as long as years for perennial plants such as trees. Plants may be
affected by exposures at several growth stages before harvest. Only a few
studies have investigated the influence of plant growth stage on plant response
to DO. Studies with white beans in areas affected by photochemical oxidants
indicated that crop maturity (plant growth stage) regulates the time of symptom
expression and that crop vigor regulates the severity of the symptom (Haas,
1970). Petunia hybrids were less sensitive to 0, after the flower bud differ-
entiated (Hanson et al., 1975). Ozone reduced radish hypocotyl growth the
most if the exposure occurred during the period of rapid hypocotyl growth
(Tingey et al., 1973a). A single exposure to ozone produced a 37 percent
reduction in hypocotyl growth in 14-day-old plants but less growth reduction
in younger or in older plants.
If it is necessary to characterize the temporal distribution of pollutant
concentrations within a growing season to characterize a plant response ade-
quately, it is questionable whether the current exposure statistics used by
researchers are adequate. Such regimens do not characterize the effects of
pollutant episodes at specific and perhaps critical periods during plant
growth. Statistics used to describe cumulative seasonal exposures, such as a
seasonal mean, do not characterize the temporal distribution of the exposures
within the season. Lognormal (Larsen et al., 1983) and two-parameter Weibull
(Georgopoulos and Seinfeld, 1982; Rawlings and Cure, 1985) functions have been
utilized.to characterize seasonal exposures. These distribution functions are
fitted to the seasonal mean 0- concentrations without regard to their temporal
^ «
order and therefore these functions, as well, do not characterize episodes
within the season. Percentiles (number of hours at a given concentration range)
(McLaughlin et al., 1982) can also be used to summarize the seasonal distribu-
. tion of concentrations but these likewise provide no means of characterizing
when within a season these episodes occur. The use of means (averages of con-
centrations over specific time periods) (Heck et al., 1982) and cumulative dose
(Oshima et al., 1977a,b; Lefohn and Benedict, 1982) also ignores the episodic
6-8
-------
nature of seasonal exposures. Other exposure representations based on a sea-
sonal averaging time suffer from similar inadequacies.
The difficulty of selecting an appropriate statistic to characterize
plant exposure has been summarized by Heagle and Heck (1980). Ambient and
experimental 0- exposures have been presented as (1) seasonal, monthly, weekly,
or daily means; (2) peak hourly means; (3) number of hours above a selected
concentration; or (4) number of hours above selected concentration intervals.
None of these statistics adequately characterizes the relationships among
ambient 03 concentration, exposure duration, and plant growth stages.
Until further research defines the influence on plant responses of temporal
fluctuations in ozone concentrations, which is characteristic of exposures to
ambient air, the selection of a summary statistic that characterizes ozone ex-
posures will continue to be discretionary. Unfortunately, the existing summary
statistics cannot be directly compared. Each is the result of calculations
from the original aerometric monitoring data and cannot be transformed to
another exposure statistic without the expensive and laborious task of return-
ing to the original data. Therefore, comparisons among studies that use
different summary statistics are difficult.
6.2.2.2 Statistics Used to Characterize Short Exposures. An experiment that
focuses on foliar injury or any other relatively short-term response may only
require short periods of exposure, which can be characterized by a simple
exposure statistic. When such results are evaluated, a problem occurs only if
the results of the short-term exposure experiment are extrapolated to evaluate
their significance in relation to long-term exposures. Mean and dose (concen-
tration multiplied by time) statistics from short-term exposures usually
cannot be aggregated to be representative of the temporal dynamics of long-term
exposures.
Although most short-term exposures are described by a concentration and
duration of dose, scientists point out that the correct exposure representation
is the amount of pollutant entering the plant, not the ambient air concentra-
tion to which it is exposed (Taylor et al., 1982a; Tingey and Taylor, 1982).
Plants are affected only by the 0., or PAN that diffuses into the leaves. It
is difficult, however, to measure or quantify the relationship between the
concentration of pollutant in the air and the internal pollutant flux because
of the interactive effects of environmental and biological variables unique to
a specific set of environmental conditions. An interactive model that requires
6-9
-------
variables describing the exposure, environmental condition, and species would
be necessary to relate internal pollution flux to ambient air levels.
6.2.2.3 Evaluation of Exposure Statistics. The characterization and repre-
sentation of plant exposures to 0., has been and continues to be a major problem.
As discussed above, in Sections 6.2.2.1 (Statistics Used to Characterize
Seasonal Exposures) and 6.2.2.2 (Statistics Used to Characterize Short Expo-
sures), several different exposure statistics have been used to characterize
exposure. A mean concentration (with various averaging times) is the most
common statistic used. Because the mean is computed by summing the concentra-
tions and dividing by time, it mathematically treats all concentrations as
being equally effective in causing a plant response. The use of a mean con-
centration (with varying averaging times) to characterize long-term expsoures
minimizes the contributions of peak concentrations to the response by treating
low-level, long-term exposures the same as high-concentration, short-term expo-
sures. The use of a longer-term mean concentration ignores the importance of
peak concentrations and is inconsistent with the literature.
A number of studies have shown that concentration is more important than
exposure duration in causing a response. For example, studies with beans.and
tobacco (Heck et al., 1966) showed that a dose over a short time period induced
more injury than the same dose distributed over a longer time period. Studies
with tobacco showed that the (k concentration was substantially more important
than exposure duration in determining the extent of foliar injury (Tonneijck,
1984). In this study, tobacco was exposed to a range of CL concentrations
(0.02 to 0.15 ppm) for 8 hr/day for 1 to 7 days. In beans, foliar injury
2
developed when the internal 03 flux exceeded 115 umoles/m within 1 hr (Bennett,
1979). However, a single 3-hr exposure at about half the 03 concentration
(0.27 compared to 0.49 ppm) required approximately 64 percent greater internal
03 flux to induce the same amount of foliar injury as in the 1-hr exposure
(Bennett, 1979). Amiro et al. (1984) showed that higher concentrations were
more important than low concentrations in causing injury. Their study also
suggested the existence of a biochemical injury threshold (i.e., the 03 uptake
rates that plants can experience without inducing visible foliar injury). The
greater importance of concentrations compared to exposure duration has been
reported by other authors also (e.g., Heck and Tingey, 1971; Henderson and
Reinert, 1979; Reinert and Nelson, 1979).
The total ozone dose (concentration multiplied by time) has been used to
describe plant exposure; however, it suffers from the same problem as the
6-10
-------
mean. The total dose is simply the summation of the ppm-hr over the study
period, which treats all concentrations as being equally effective. Several
investigators have attempted to give greater importance to peak 0^ concentra-
tions. Oshima et al. (1977a,b) and Lefohn and Benedict (1982), for example,
have summed only the ppm-hr of exposure greater than some preselected value.
Larsen et al. (1983) introduced the concept of "impact" to describe; the effects
of 0., and SCL on soybeans. The "impact (I)" is calculated similarly to total
dose, except that the concentration is raised to an exponent greater than one
W
(I = C x T); this method of calculation effectively gives greater weight to
the higher concentrations. More recently, Larsen and Heck (1984) have sug-
gested the term "effective mean" as an approach for describing the greater
importance of higher concentrations. The "effective mean" is defined as the
average hourly impact raised to an exponent and divided by the duration.
Several lines of evidence suggest that higher concentrations have a
greater influence in determining the impact of CL on vegetation. Studies have
shown that plants can tolerate some combinations of exposure duration and
concentration without exhibiting foliar injury or effects on growth or yield,
illustrating that not all concentrations are equally effective in causing a
response. From the toxicological perspective, it is the peaks or concentra-
tions above some level that are most likely to have an impact. Effects occur
on vegetation when the amount of pollutant that the plant has absorbed exceeds
the ability of the organism to repair or compensate for the impact.
Studies with soybean (Johnston and Heagle, 1982), tobacco (Heagle and
Heck, 1974), and bean (Runeckles and Rosen, 1977) showed that plants exposed
to a low level of 0, for a few days became more sensitive to subsequent 0,
exposures. In studies with tobacco, Mukammal (1965) showed that a high ozone
concentration on one day caused substantial injury but an equal or higher
concentration on the second day caused only slight injury. Using stress
ethylene as an indicator of CL effects, Stan and Schicker (1982) showed that a
series of successive short exposures was more injurious to plants than a
continuous exposure at the same CL concentration for the same total exposure
period. Walmsley et al. (1980) continuously exposed radishes to 03 for several
weeks. They found that the plants acquired some CL tolerance. The acquired
tolerance displayed two components: (1) the exposed plants developed new
leaves faster than the controls and (2) there was a progressive decrease in
sensitivity of the new leaves to CL. The newer leaves also displayed a slower
6-11
-------
rate of senescence. The observations by Elkiey and Ormrod (1981) that the CL
uptake decreased during a 3-day study period may provide an explanation for
the results with radish.
Not only are concentration and time important but the dynamics of the 03
exposure are also important; that is, whether the exposure is at a constant or
variable concentration. Musselman et al. (1983) recently showed that fixed
concentrations of O, cause the same kind of responses as variable concentra-
tions at the equivalent dose. Fixed concentrations, however, had less effect
on plant growth responses than variable concentrations at similar doses.
Exposures of radishes to ambient 0, in open-top exposure chambers showed that
significant yield reductions occurred when the maximum (k concentration ex-
ceeded 0.06 ppm on at least 10 percent of the days when the crop was growing
(Ashmore, 1984). Initial studies by Hogsett et al. (1985) have compared the
response of alfalfa to daily peak and episodic 0, exposure profiles which had
the equivalent total 03 dose over the growing season. Alfalfa yield was
reduced to a greater extent in the episodic than the daily peak exposure.
This study also illustrates the problem with the 7-hr seasonal mean concentra-
tion, which is that the peak concentrations are not properly considered. The
plants that displayed the greater growth reduction (in the episodic exposure)
were exposed to a significantly lower 7-hr seasonal mean concentration.
Studies with SOp also showed that plants exposed to variable concentrations
exhibited a greater plant response than those exposed to a constant concentra-
tion (McLaughlin et al., 1979; Male et al., 1983). These results suggest that
the mechanisms causing the response are the same, but that exposures to fixed
concentrations underestimate the magnitude of plant growth responses that can
occur with episodic exposures.
Currently, there is no consensus as to the most appropriate summary sta-
tistic for representing plant exposure to photochemical oxidants. Consequently,
many different statistics are used, making direct comparisons between studies
extremely difficult. Further, there is some question as to the adequacy of
statistics used to characterize long exposures (season), since they do not
consider exposure dynamics within the period'being represented. This question
cannot presently be resolved because research to date has not clearly deter-
mined whether stages of plant growth are differentially sensitive to exposures
relative to ultimate yield.
6-12
-------
6.2.3 Exposure Systems
Research methods can be organized according to the means by which exposures
or environmental variables are controlled or characterized. Air pollution
research often requires exposure chambers or other apparatus for maintaining
controlled pollutant exposures. Exposure systems may range from sophisticated,
microprocessor-controlled cuvettes (Bingham and Coyne, 1977; Legge et al. ,
1979) to a series of tubes with calibrated orifices spatially distributed over
a field to emit gaseous pollutants (Lee et al., 1978). Each type of systems
was designed for specific objectives and operates most efficiently under the
conditions for which it was intended. Each has advantages and limitations and
must be evaluated in terms of the objectives it was designed to meet.
The exposure systems discussed in this section share many common charac-
teristics. Each uses a monitoring system that measures pollutant concentration
continuously during exposures or that incorporates a time-sharing system that
sequentially measures concentrations in chambers or at field sites. The
®
systems normally employ inert Teflon tubing for sampling lines and continuous
air flow to reduce time lags. Additionally, many systems use EPA-approved
monitoring and detection systems (see Chapter 4 for EPA equivalent and Federal
reference methods for ozone). Recently, quality assurance programs were
included in several studies to ensure that high quality, standardized air
monitoring data will be available and readily comparable. Under one such
program, the air pollutants are generated artificially and dispensed to exposure
chambers or field plots; under another, proportional activated-carbon filtration
is used to provide different levels of ambient pollutants.
The systems described in this section represent significant advances in
the methods used in air pollution research on vegetation. As systems that
utilize the latest technological advances evolve, it is easy, because of the
rapid pace of their evolution, to lose sight of their limitations. Even the
most sophisticated and advanced systems are only as good as the researcher who
uses them. They do not ensure that the research results will be of superior
quality. They only provide the potential for understanding better the impact
of air pollutants on vegetation.
The following discussion is limited to exposure systems used in air
pollution research and is not meant to be a detailed description of the system
components. These systems are described in greater detail in original publica-
tions and review articles (e.g., Heagle and Philbeck, 1979).
6-13
-------
6.2.3.1 Laboratory Systems. Laboratory systems (Tingey et al., 1979; Winner
and Mooney, 1980) typically employ artificial lighting and controlled environ-
ments. Most are designed to identify and measure effects ranging from the
subcellular to the whole-plant level of biological organization. Although
results from these systems are difficult to relate directly to field.studies,
they do contribute to an understanding of the mechanisms involved with air
pollution effects. They provide useful information in explaining or inter-
preting responses. The stability of the well-controlled environmental condi-
tions characteristic of most laboratory systems allows precise measurement of
an array of plant responses. By altering only one variable and holding others
constant, responses can be better defined and more easily understood. These
systems are powerful tools for increasing the understanding of the effects of
pollutants on the biological processes basic to plant growth.
The greatest drawback of laboratory systems relates to the general appli-
cability of final results. The precise environmental conditions that make the
systems valuable for defining responses also make the laboratory systems arti-
ficial. In comparison, ambient environmental conditions are complex and
dynamic.
6.2.3.2 Greenhouse Exposure Systems. Greenhouse systems are generally used
in studies to identify and quantify physiological, growth, and yield responses
at the organ and whole-plant level of biological organization. Plants are
usually grown in containers in greenhouses with charcoal-filtered air. Expo-
sures are conducted under natural or artificial lighting, or both within
chambers in the greenhouse. Plants may be physically moved in and out of
exposure chambers and allowed to grow on greenhouse benches during interim
periods. Normally, a single plant or small groups of plants constitute the
experimental unit. While the environmental conditions of greenhouse exposure
systems may more closely approximate field than laboratory conditions, the
plant cultural conditions are more similar to those used in laboratory studies.
Although related to field studies, greenhouse studies differ sufficiently to
make direct extrapolations to field conditions difficult. It must be remembered,
however, that greenhouse conditions are the typical cultural environment for
many floricultural and ornamental plants. In this case, the use of greenhouse
conditions is appropriate and no extrapolation is necessary.
6-14
-------
Greenhouse exposure systems usually consist of a series of chambers built
with a framework of various materials and covered with a transparent film.
The air exchange systems normally use a negative pressure, single-directional
air flow, and employ an activated-charcoal filtration device at both air entry
and exhaust. Early systems were usually modifications of the system developed
by Heck et al. (1968), but a variety of designs were utilized. These systems
were all designed to meet common, desirable chamber characteristics (uniform
pollutant concentrations with minimal environmental alteration) and succeeded
to varying degrees. The design of the continuous stirred-tank reactor (CSTR)
by Rogers et al. (1977) stimulated the development of exposure systems that
®
incorporated its desirable mixing properties and the use of FEP Teflon film
as an inert polymer film.
6.2.3.3 Field Exposure Systems. The accurate assessment of pollutant-induced
changes in agricultural productivity, and resulting economic impacts, requires
that deviations from the ambient environment be minimized and that conditions
characteristic of agricultural systems or natural ecosystems be simulated as
closely as possible. Field exposure systems range from adaptations of the
greenhouse and laboratory chamber designs to the use of chemical protectants.
In most greenhouse and field studies, the investigators have attempted to
ensure that soil moisture, plant nutrients, and other cultural conditions did
not limit growth.
6.2.3.3.1 Field chamber systems. The open-top chamber system (Heagle et al.,
1973; Mandl et al., 1973) is the most popular field-exposure system presently
in use. Essentially upright cylinders with a clear polymer film as a covering
around the sides, these chambers have the advantage of portability, moderate
cost, and ease of maintenance. The size and shape of the chambers may be
modified for use with different plant types and sizes. The system uses a
high-volume flow of filtered air to reduce ambient pollutant influx through
the open top. The chambers can be used as air-exclusion systems to test the
difference between ambient air and charcoal-filtered air, or they can be used
as exposure chambers, with pollutants added to the incoming air stream. The
rate of pollutant addition is adjusted to control the pollutant concentration
in -the chambers. Pollutants are usually measured just above canopy height.
Studies of the 0, distribution within the chambers have shown it to be quite
uniform. The vertical variation of 0., concentration in the 2.44-m-high chambers
was less than 6 percent between 0.3 and 1.2 m and less than 19 percent between
o
1.2 and 1.8 m. The horizontal variation over the 7.3 m of the chamber was 12
6-15
-------
percent and 14 percent at heights of 1.2 and 1.8 m, respectively (Heagle et
al., 1979d). The portability of the system facilitates storage and maintenance
during the winter or in periods of inactivity and allows standard agricultural
practices to be carried out during field preparation, seeding, and early crop
growth before chambers are set in place. Open-top chambers and well-ventilated,
closed-top chambers reduce temperature deviations from the ambient, allow suf-
ficient pollutant control for either single or mixed-gas exposures, and are
relatively inexpensive. They can be selectively placed in established fields
to avoid unacceptable soil types or to maximize soil uniformity in treatments.
Most of the limitations of open-top chambers relate to air-flow charac-
teristics. Air flowing from the lower portion of the chamber out through the
open top reduces the intrusion of outside air; this air-flow pattern is dif-
ferent, however, from that in the open field. Because plants in the chamber
experience a different air-flow pattern than field-grown plants, concerns have
been expressed that this might alter the influence of 0^ on plants. Recent
measurements, however, of canopy resistance to 03 uptake in open-top chambers
and by micrometeorological methods in the field yield similar results of 73
and 84 sec m , respectively (Unsworth et al., 1984a,b). This similarity led
the authors to conclude that crop exposure to gaseous pollutants in open-top
chambers is similar to that which would occur at the same concentrations in
the field.
With open-top exposure chambers, some intrusion of ambient air and its
pollutant burden through the chamber top is unavoidable; and this air can
influence the pollutant concentrations within the chamber (Heagle et al.,
1973; Unsworth et al., 1984a,b). The amount of intrusion increases with wind
speed. Recent design innovations, however, have minimized this problem (Kats
et al., 1976; Davis and Rogers, 1980). For example, the addition of a frustum
(a truncated cone) to the top of the open-top chambers can reduce the intrusion
of ambient air by approximately 50 percent and can also provide a more repro-
ducible environment for a given wind speed (Unsworth et al., 1984a,b).
It should be recognized that open-field environmental conditions cannot
be exactly duplicated by open-top exposure chambers (Heagle et al. , 1979d;
Olszyk et al., 1980) or any other pollutant exposure system presently available.
In summarizing studies of open-top exposure chambers, Heagle et al. (1979d)
reported:
6-16
-------
In our 7-yr experience, the open-top chambers caused plants to grow
slightly taller but rarely had significant effects on yield. Plants
often grew differently in different parts of the chambers but we did
not find significant interactions between chamber position and the
effects of 0,. The causes for chamber-induced growth effects may be
related to slower mean air velocity, slightly higher temperature, or
less light at some chamber locations than in the open There are
no reports, however, that environmental changes of the magnitude
caused by open-top chambers change plant sensitivity.
(Heagle et al., 1979d)
The lack of a significant chamber influence on plant response is supported
by the observation of Reich and Amundson (1984). They recently compared yield
response functions for soybean exposed in a "tubular release system" with
functions for soybean exposed in open-top exposure chambers, and concluded
that the results from the two systems were comparable.
Other field-exposure systems use chambers of varying design, but have the
common characteristic of being fully enclosed by film (Thompson and Taylor,
1966; Oshima, 1978). These designs rely on high air-exchange rates to minimize
temperature alterations. Most of these designs are adaptations or alterations
of greenhouse exposure systems. Chamber shapes range from a square design, as
described by Heck et al. (1968), to the CSTR cylinder described by Rogers et
al. (1977).
6.2.3.3.2 Field exposure systems without chambers. The desire to expose
large field plots to increase sample size and to remove environmental altera-
tions caused by enclosing plants in chambers led to the development of chamber-
free field exposure systems. The advantage of these systems (Lee et al. ,
1978; deCormis et al., 1975; Reich et al., 1980; Laurence et al., 1982; Reich
and Amundson, 1984) is that plants are exposed to pollutants under conditions
similar to ambient conditions. This advantage is offset to some extent by the
disadvantage of losing some control over pollutant concentration and the
nature of the exposure. These systems are highly influenced by wind speed and
direction, and are subject to ambient air levels. There have been only limited
0- studies using these types of systems.
6.2.4 Methodologies Used in the National Crop Loss Assessment Network
The National Crop Loss Assessment Network (NCLAN) was initiated in 1980
by EPA to estimate the magnitude of national crop losses caused by air pollu-
tion. Initial emphasis was placed on 03 (Heck et al. , 1982). A research
6-17
-------
management committee has been responsible for the planning, management, and
execution of the program. The primary objectives of the NCLAN program are:
1. To define the relationships between yields of major agricultural
crops and 0- exposure as required to provide data for economic
assessments and the development of NAAQS;
2. To assess national economic consequences of the exposure of major
agricultural crops to 03; and
3. To advance the understanding of the cause and effect relationships
that determine crop responses to pollutant exposure.
The NCLAN is a network of experimental field sites selected for (1) their
different climatological conditions, (2) their distribution of different crop
species, and (3) their proximity to established research groups with a history
of research on air pollutant effects on vegetation. The test species are
grown in the field under conditions approximating standard agronomic practices.
Efforts are made to minimize perturbations of the plant environment by the ex-
posure apparatus and to use realistic pollutant doses.
The pollutant concentrations around crop plants in the field are controlled
and manipulated through the use of open-top chambers to simulate ambient
exposures. Sufficient numbers of chambers permit replicated experimental
designs; and also permit the development of empirical dose-response models.
Models for test species and cultivars are developed from data for several
sites and for several years.
Within the open-top chambers (see Section 6.2.3.3.1), plants are exposed
to a range of ozone concentrations. Daily variations in the (L concentration
are determined in part by changes in ambient 0- concentrations at each site.
The lowest 0~ concentration (control, charcoal-filtered air) is usually 20 to
50 percent of that in ambient air; the 0, that is present enters the chamber
*J'
mainly through the open top, because the inlet air to the chamber is charcoal-
filtered. All other chambers receive ambient air supplemented (usually 7
hr/day) with enough 0., to provide concentrations equal to those in the ambient
air and three or four higher concentrations. Consequently, the 03 exposures
are coupled to ambient 0- levels; days with the highest ambient 0- will also be
the same days when the highest concentrations will occur in a specific treatment
in a chamber. As the ambient 0- varies from day to day, the base to which
additional 03 is added also varies. This coupling of the 0., exposures to the
6-18
-------
ambient environment means that high 0- concentrations will occur in the chambers
when the environmental and air chemistry conditions in the ambient air are
conducive for producing elevated ambient 0, levels.
In the initial NCLAN studies, (L was added to the chambers in three or
four stepwise increments (0.02 to 0.03 ppm) above the concentration in the
ambient air. In more recent studies, 03 has been added at various proportions
above the ambient concentration. The study by Temple et al. (1985b) illustrates
both types of 03 addition. Ozone concentrations within the chambers are
measured at canopy height with time-shared monitors. Plant yields are also
measured for non-chamber field plots of identical size exposed to ambient air
to obtain an estimate of potential chamber effects. Chamber fans are opera-
tional from 5:00 a.m. to 9:00 p.m. daily, and 03 is usually added from 9:00
a.m. through 4:00 p.m. (local standard time) daily throughout the growing
season for the crop, except on rainy days.
A quality assurance program for the collection and measurement of air
quality and biological data is followed in NCLAN studies. Independent audits
of the pollutant monitors are conducted at each site.
The data are usually analyzed by regression analysis. The mean 7-hr
daily concentration (9:00 a.m. to 4:00 p.m.), averaged over the growing season„
is used for a seasonal exposure statistic. This is the time period when 03 is
added to the chambers.
Many strengths are associated with a coordinated national multisite
program such as NCLAN. Perhaps the greatest strengths of NCLAN are the stan-
dardization of methods for air monitoring, biological assessment, experimental
design, pollutant exposure regimes, summarization of exposures, and quality
assurance. Additionally, the selection of agriculturally important crops for
test species and the use of close approximations of standard cultural practices
ensure applicability of experimental results. The development of empirical
models interfaces well with the data requirements for a national economic
assessment. Previously, very few biological models were available for economic
assessments.
The NCLAN approach has limitations that must also be considered. The
potential artificiality of the 0- exposure treatments may complicate the
application of results. Further, the use of the seasonal 7-hr daily mean
concentration, a relatively new exposure summary statistic, makes comparisons
with previously published studies difficult. It also does not accurately
6-19
-------
represent the temporal exposure dynamics of ambient air. The lack of valida-
tion of the model predictions is unsettling, but that is a common deficiency
of all models to date and is not unique to NCLAN. These limitations may also
occur with other field studies.
When viewed in perspective, NCLAN represents the state of the art for
documenting yield losses resulting from ozone and for providing compatible
data for use in economic assessments on a national scale.
6.2.5 Definitions of Yield Loss and Crop Loss
For the purposes of this chapter, yield loss is defined as reduction in
quantity, quality, aesthetic value, or any impairment of the intended use of a
plant. Thus, foliar injury on ornamental plants, detrimental responses in
native species, and reductions in fruit or grain production by agricultural
species are all considered yield loss. Crop loss, in contrast, is defined as
an economic or monetary loss and is not synonymous with yield loss. Crop loss
occurs at aggregative levels higher than the plant or plot. The transforma-
tion of yield loss to crop loss incorporates economic considerations such as
those described in Section 6.4.2.2.3.
6.3 MODE OF ACTION OF OZONE ON PLANTS
Plant growth and yield are the culmination of many biochemical and physio-
logical processes. Plants absorb carbon dioxide from the atmosphere through
portals called stomata. Within the chloroplasts located in the mesophyll
cells of the leaf (Figure 6-2), the carbon dioxide is converted into carbohy-
drates in the presence of light (photosynthesis). Plants absorb the necessary
water and mineral nutrients for growth from the soil. Growth and yield depend
not only on the rate of photosynthesis and the uptake of water and nutrients,
but also on subsequent metabolic processes and the allocation of the photosyn-
thetic products to the rest of the plant. The uptake of carbon, dioxide and
its subsequent metabolism and allocation within the plant are influenced by
various environmental conditions. The impairment of any of these processes
may affect plant growth and yield.
The responses of vascular plants to 0- may be viewed as the culmination
of a sequence of physical, biochemical, and physiological events. Ozone in
the ambient air does not impair plant processes or performance, only the 0,
i -3
that diffuses into the plant. An effect will' occur only if a sufficient
6-20
-------
CUTICLE
VASCULAR
CELLS
LEAF
HAIR
EPIDERMIS
PHOTOSYNTHETIC
MESOPHYLL CELLS
INTERCELLULAR
SPACE
GUARD CELLS
STOMATA
Figure 6-2. Schematic cross section of a typical dicot leaf.
6-21
-------
amount of (L reaches the sensitive cellular sites within the leaf. The 03
diffuses from the ambient air into the leaf through the stomata, which can
exert some control on 0, uptake to the active sites within the leaf. Ozone
injury will not occur if (1) the rate of 03 uptake is sufficiently small so
that the plant is able to detoxify or metabolize 0, or its metabolites; or
(2) the plant is able to repair or compensate for the 0, impacts (Tingey and
Taylor, 1982). The uptake and movement of 0- to the sensitive cellular sites
are subject to various physiological and biochemical controls.
Ozone enters the leaf through stomata; once within the leaf it quickly
dissolves in the aqueous layer on the cells lining the air spaces. Ozone, or
its decomposition products, then diffuses through the cell wall and membrane
into the cell, where it may affect cellular or organellar processes. Ozone
flux (J) into the leaf may be represented by the following equation (Tingey
and Taylor, 1982):
J = AC/(R + R + R ). (6-1)
a b I
Ozone flux is directly proportional to the change in 03 concentration (AC) be-
tween the ambient air and the leaf interior (gas-to-liquid transfer) and is
inversely proportional to resistance to the mass transfer of gas. Resistance
to 0, movement can be divided into components, including boundary layer (R ),
«j .3
stomatal and intercellular space (Rg), and liquid-phase (R ) resistances.
At any point along this pathway, 03 or its decomposition products may
react with cellular components. Altered cell structure and function may
result in changes in membrane permeability, carbon dioxide fixation, and many
secondary metabolic processes (Tingey and Taylor, 1982). The magnitude of
0~-induced effects will depend upon the physical environment of the plant,
including macro- and microclimatic factors; the chemical environment of the
plant, including other gaseous air pollutants and a variety of chemicals; and
biological factors, including genetic potential and developmental age of the
plant and interaction with plant pests. Cellular injury may subsequently
manifest itself in a number of ways, including visible foliar injury, premature
senescence, reduced yield or growth or both, reduced plant vigor, and sometimes
death. Depending upon the intended use of a plant species (i.e., for food,
forage, fiber, shelter, or amenity), any of the effects discussed above could
impact society adversely.
6-22
-------
In the following sections, selected references will be cited to describe
how Q, induces some of its effects. Some of the physiological studies have
been conducted with 0- exposures that would rarely, if ever, be encountered in
ambient air. This literature can, however, serve as a tool for identifying the
potential sequence of the physiological and biochemical responses of plant
species, and for identifying potential metabolic sites of action.
6.3.1 Biochemical and Physiological Responses to Ozone
Phytotoxic effects of air pollution on plant tissue will occur only when
sufficient concentrations of a gas diffuse into the leaf interior and pass
into the liquid phase of the cells. Once a gas is deposited on a wet cell
surface, it may move by diffusion or bulk flow to sites of action, such as the
interior of the cell membrane, the cytoplasm, or cellular organelles (Heath,
1980; Tingey and Taylor, 1982).
6.3.1.1 Gas-Phase Movement into the Leaf. Ozone, as well as other gases,
diffuses from the atmosphere into the leaf through stomata. The stomata
control the rate of 0, uptake into the leaf and are influenced by various
plant and environmental stimuli. A variety of factors, including 0,, have
been shown to induce stomatal closure. The previous criteria document (U.S.
Environmental Protection Agency, 1978) cited a number of studies that directly
correlated 03 concentration and stomatal closure. Engle and Gabelman (1966)
reported that in the presence of 0, (0.3 ppm for 0.5 hour) stomata closed
more quickly in tolerant than in sensitive onion cultivars. Rich and Turner
(1972) found that when tobacco plants were exposed to 0.20 to 0.25 ppm 0^ for
2 hours, leaf conductance (a measure of stomatal closure) decreased 32 percent.
in a resistant cultivar and only 9 percent in a sensitive cultivar (no statis-
tics provided), suggesting possible differences in 0- uptake between cultivars.
In contrast, when four cultivars of peas were exposed to an 0, concentration
of 0.15 ppm for 6 hours per day and stomatal conductance was measured, the two
more sensitive cultivars had greater decreases in leaf conductance (85 percent
and 86 percent) than did the two more tolerant cultivars (62, percent and 69
percent) (Dijak and Ormrod, 1982). Clearly, decreased conductance could not
explain differential cultivar tolerance in this case. More recently, Krause
and Weidensaul (1978b) observed that geranium guard cells, which control
stomatal opening, ruptured after a 10-day exposure to 03 at concentrations of
0.15 ppm for 6 hours per day. When they reviewed the 0, uptake literature,
Tingey and Taylor (1982) found examples of species for which the 0,. response
6-23
-------
was apparently limited by leaf conductance (i.e., Og uptake) and species for
which On response (injury) was not controlled by 0., uptake but rather by
metabolic processes within the mesophyll cells.
Ozone flux into the leaf may also be regulated by stomatal density. Butler
and Tibbitts (1979a,b) correlated stomatal density directly with 0,-induced
visible injury in bean plants, but Gesalman and Davis (1978) found no such
relationship for azalea cultivars. There was no apparent relationship between
stomatal frequency or guard-cell length and differential 0., sensitivity in two
corn cultivars (Harris and Heath, 1981). They found that the leaf water
potential was poised near the point at which only a slight water loss in the
tolerant cultivar would induce stomatal closure. Hence, they suggested a
rapid stomatal closure in response to an O^-induced water loss. In the 1978
criteria document (U.S. Environmental Protection Agency, 1978), equally dispa-
rate results were offered for several plant species. Dean (1972) reported
that tobacco cultivars that exhibited tolerance to oxidant-induced weather
fleck in the field had lower stomatal density than that in sensitive cultivars.
Evans and Ting (1974) found that the maximum 0., sensitivity of primary leaves
of bean could not be accounted for by stomatal density.
In summary, different plant responses to 03 are in part the result of the
diffusion of 0, into the leaf interior. A knowledge of the 0, uptake rate or
amount, however, is not sufficient for predicting subsequent responses for all
species. In some species, injury is apparently not directly related to Oo
uptake; in others, there is a relationship between the quantity of 03 entering
the plant and the degree of subsequent injury. The physical and chemical
environment and biological potential of the plant influence stomatal behavior
and 03 uptake, as will be documented in later sections. Once 0^ enters the
plant, there are potential reactions with many cellular constituents.
6.3.1.2 Transition between Gas-Phase and Liquid-Phase Movement into the Cell.
Once ozone enters the intercellular spaces, it passes into the liquid phase at
the gas-liquid interface of the cell wall surface. The diffusive process is
dependent on physical, chemical, and biological factors that govern this
diffusive step (Tingey and Taylor, 1982). The solubility of 0, is critical to
further reaction and depends on microclimatic factors, including temperature.
The rate at which gas diffusion occurs may also depend upon the internal
cell surface area exposed (Evans and Ting, 1974; Pell and Weissberger, 1976;
Uhring, 1978). Taylor et al. (1982b) reported that in soybean foliage pollu-
tant flux was not regulated solely by the number of cellular sites of 0,
6-24
-------
deposition. When plants were exposed to (L concentrations ranging between
0.25 and 0.58 ppm for 1 to 4 hours, uptake rates were higher and the ratio of
internal/external leaf area was lower for "Hood," a relatively tolerant soybean
cultivar, than for "Dare," which was more sensitive. Athanassious (1980) did
not identify surface-volume ratio as a determinant of relative response of
radish mesophyll cells to 0,, but suggested that differential suberization of
cell walls may explain relative sensitivity of parenchyma! tissue. This idea
was offered previously by Glater et al. (1962).
6.3.1.3 Chemical and Biochemical Responses. When 03 passes into the liquid
phase, it undergoes transformations that yield a variety of free radicals,
including superoxide and hydroxyl radicals (Pryor et al., 1981; Hoigne and
Bader, 1975; Tingey and Taylor, 1982). Whether these chemical species result
from decomposition of 0, or reactions between 0, and biochemicals in the
extracellular fluid has not been determined. Ozone or its decomposition
products, or both, will then react with cellular components, resulting in
structural or functional effects, or both.
The potential for Og, directly or indirectly, to oxidize biochemicals uj
vitro has been demonstrated. Ozone can oxidize a number of biological mole-
cules, including reduced nicotinamide adenine dinucleotide (NADH), DMA, RNA,
purine, pyrimidines, indole acetic acid, some amino acids (including tryp-
tophan and methionine), many proteins (including enzymes such as glyceraldehyde-
3-phosphate dehydrogenase, catalase, peroxidase, papain, ribonuclease, and
urease), and a variety of lipids (Christensen and Geise, 1954; Todd, 1958;
Ordin and Propst, 1962; Heath, 1975; Mudd, 1982). In these and similar studies,
the concentrations of 0^ bubbled into the biochemical solutions were all very
high. It is difficult to compare the exposure to ozone in solution to the
ambient air exposure that plants experience. Coulson and Heath (1974) have
suggested, however, that solution and atmospheric exposures are not highly
dissimilar. They showed that most of the Oo bubbled into solutions exited
unreacted and that the 0, dose required to injure cells in solution was of a
magnitude similar to that required to injure intact plants exposed to atmos-
pheric Oo. Todd (1958) predicted sensitivity within the plant by relating
concentrations of protein used j_n vitro to levels in the plant, and then
extrapolating to lower concentrations of 0,. Similar comparisons could be
made for other biochemicals studied i_n vitro. Because biochemicals are compart-
mentalized within the plant, such calculations of potential sensitivity may
deviate from actual responses observed. Data acquired from i_n vitro studies
6-25
-------
are best utilized to demonstrate that many cellular constituents are susceptible
to oxidation by 0~. Different approaches will have to be used to determine
which, in fact, are important ijn vivo.
The potential for biochemicals to be affected within the plant has been
explored by a number of researchers. Increases and decreases have been ob-
served in the status of proteins, sulfhydryl residues, fatty acids, and sterols
(Pell, 1979; Trevathan et al.., 1979; Swanson et al., 1973). Results vary
among laboratories. For example, Trevathan et al. (1979) observed a decrease
in fatty acids 3 days after tobacco plants were exposed to 0.24 ppm 0, for 6
hours, whereas Swanson et al. (1973) detected no change in fatty acid content
in the same species 2 hours after plants received 0.30 ppm 0, for 2 hours. It
is likely that Trevathan et al. (1979) were observing a late plant response
associated with injury and cell death while Swanson et al. (1973) were finding
evidence that lipids were not particularly sensitive to 0,. Similarly, Fong
and Heath (1981) were unable to detect any changes in either phospholipid
content or fatty acid composition of total polar lipids in bean leaves that
sustained mild visible injury after exposure to an 03 concentration of 0.30
ppm for 1 hour. Changes in mono- and digalactolipids were observed after
severe injury was induced by a concentration of 0.50 ppm for 1 hour.
The examples above serve to underscore the importance of recognizing the
limitations of studies in which biochemical effects are determined for whole
leaf tissue rather than for organelles; or in which effects are determined in
terms of cell function. Such data neither describe the dynamics of injury
development nor identify the cellular site at which biochemical changes are
occurring. This kind of biochemical information is useful, however, in charac-
terizing the nature of a response to 0, as it relates to altered metabolism,
in general, and to visible foliar injury.
6.3.1.4 Physiological Responses. Physiological measurements have been more
useful than biochemical quantifications in characterizing cell responses to
oxidants. Many consider membranes to be the primary site of action of 0-
(Heath, 1980; Tingey and Taylor, 1982). The alteration in plasma membrane
function is an early event in the sequence of 0,-induced effects that eventu-
ally leads to leaf injury and subsequent yield loss. Changes in the semiper-
meability of the membrane are evidenced by changes in fluxes of carbohydrates,
ami no acids, inorganic ions, and water (Heath, 1975, 1980; Tingey and Taylor,
1982). Whether the plasma membrane or some organelle membrane is the primary
site of 0- action is open to speculation (Tingey and Taylor, 1982). Mudd
6-26
-------
(1982) suggested that 0- or its decomposition products may penetrate the plasma
membrane and injure organelles. A number of membrane-dependent functions of
organelles can be altered by 0.,. MacDowall (1965) reported that oxidative
phosphorylation was inhibited when tobacco plants were exposed to 0- at concen-
trations from 0.6 to 0.7 ppm for 1 hour. Photophosphorylation was inhibited
in isolated spinach chloroplasts when 03 (400 ppm for 15 minutes) was passed
through the chloroplast suspension (Coulson and Heath, 1974). Using the
Bensen coefficient for 0- and the partial pressure of the gas above the aque-
ous solution, Coulson and Heath (1974) calculated the latter dose to be equi-
valent to a concentration of 0.20 ppm in ambient air surrounding a terrestrial
plant.
Ozone can also affect biochemical functions not associated with membranes.
The activity of 1,5-ribulose bisphosphate (RuBP) carboxylase, an enzyme that
catalyzes C02 fixation during photosynthesis, can be inhibited by CU. For
example, 0.12 ppm for 2 hours inhibited the activity of RuBP carboxylase in
rice (Nakamura and Saka, 1978). Inhibition of RuBP carboxylase activity is a
relatively early event occurring several hours after conclusion of the 0.,
exposure. Pell and Pearson (1983) observed 36, 68, and 80 percent decreases,
respectively, in the concentration of 1,5-RuBP carboxylase in foliage of three
alfalfa cultivars that had been exposed to an 0, concentration of 0.25 ppm for
2 hours. Observations were made 48 hours after exposure on leaves that did
not exhibit macroscopic injury symptoms. Crystals observed ultrastructurally
in the chloroplast stroma of beans and hybrid poplars exposed to 0, were
thought to be 1,5-RuBP carboxylase (Thomson, 1975; Noble et al., 1980).
In some of the studies cited above, researchers examined the specific
effects of 03 on key steps in photosynthesis. The effect of 0, on apparent
photosynthesis, a measure of C02 uptake or fixation or both, was measured for
many more plant species (Table 6-1). Reductions in apparent photosynthesis
may reflect the direct impairment of chloroplast function or reduced C02
uptake resulting from 0~-induced stomatal closure, or both. Regardless of the
mechanism, a sustained reduction in photosynthesis will ultimately affect the
growth, yield, and vigor of the plant.
When considering dose-response effects of 0., on plant yield in this docu-
ment, emphasis has been placed on studies in which 03 concentrations of 0.25
ppm or below were utilized (Table 6-1). Examples of 0.,-induced reduction in
apparent photosynthesis at concentrations exceeding 0.25 ppm are also presented
(Table 6-1). These data highlight the potential of 03 to reduce primary
6-27
-------
TABLE 6-1. EFFECT OF OZONE ON PHOTOSYNTHESIS
Species
Loblolly pine
Slash pine
Bean
Alfalfa
Ponderosa pine
Eastern white pine
Eastern white pine
Sensitive
Intermediate
Bean
Black oak
Sugar maple
White pine
Sensitive
Tolerant
Poplar hybrid
Ponderosa pine
al ppm = 1960 ug/m.
bP < 0.05.
CP < 0.01.
Standard deviation.
03
concentration,
ppm
0.05
0.05
0.072
0.10
0.20
0.15
0.30
0.15
0.10
0.20
0.30
0.10
0.20
0.30
0.30
0.50
0.50
0.7 or 0.9
0.70 to 0.95
0.90
450, 700
800 ppm-hr
Exposure duration
18 wk
continuously
18 wk
continuously
4 hr/day for 18 days
1 hr
1 hr
9 hr daily/
60 days
9 hr daily/
30 days
19 days
4 hr/day for 50 days
4 hr/day for 50 days
4 hr/day for 50 days
4 hr/day for 50 days
4 hr daily/50 days
4 hr daily/50 days
3 hr
4 hr daily/2 days
4 hr daily/2 days
3.0 or 10
10/30 days
1.5 hr
Cumulative
dose over
1,2,3 yr
inhibition
15b
9b
18b
A
h
25C
67C
10C
$
51b
Not sig.
different
20b
22C
30 ± 10d
21 ± 10d
100b
Ob
60e
90b
Reference
Barnes (1972a)
Barnes (1972a)
Coyne and Bingham (1978)
Bennett and Hill (1974)
Miller et al. (1969)
Barnes (1972a)
Yang et al. (1983)
Pell and Brennan (1973)
Carlson (1979)
Carlson (1979)
Botkin et al. (1972)
Furukawa and Kadota (1975)
Coyne and Bingham (1981)
eNo statistical information.
6-28
-------
productivity. Several of the studies provide data more pertinent to the
ambient atmosphere. Barnes (1972a) examined the impact of 03 on seedlings of
three species of pine at concentrations of 0.05 or 0.15 ppm continuously for
19 days to 18 weeks. In younger seedlings of eastern white pine, which bore
only primary needles, (L had little influence on photosynthetic rate. In
o
older seedlings with secondary needles, photosynthesis was slightly depressed.
With seedlings of slash, eastern white, and loblolly pines, exposure at 0.15
ppm 0- had a relatively consistent depressing influence on photosynthesis of
all species. At 0.05 ppm, however, 0- appeared to stimulate photosynthesis in
older secondary needles and depress photosynthesis in younger secondary needles.
Barnes (1972a,b) used a Mast meter to measure 0,; the Mast meter can under-
estimate the 0- concentration unless it is calibrated against a reference
standard (Chapter 4). Also, the sample size used in these experiments was
very small, four to nine seedlings. It is possible that variation among
samples may have masked potential effects in some of the experiments (Barnes,
1972a). Coyne and Bingham (1978) exposed field-grown snap beans to an 0,
concentration of 0.072 ppm (the 0, monitor was calibrated by UV photometry;
see Chapter 4) for 4 hours per day for 18 days. Apparent photosynthesis was
reduced 18 percent in plants treated with 03. Bennett and Hill (1974) reported
that apparent photosynthesis of alfalfa plants was depressed 4 percent and 10
percent when 0- concentrations were 0.1 and 0.2 ppm for 1 hour, respectively.
Methods of 03 monitoring and calibration were not given by the authors.
Black et al. (1982) found a significant (p < 0.001) relationship (r =
-0.8) between net photosynthetic rate of broad bean and 4-hour exposures to
concentrations of 03 0.05 to 0.30 ppm. Exposure to 03 concentrations of less
than 0.10 ppm resulted in a reversible depression of photosynthesis. Twenty
hours after exposure to 0, concentrations of 0.10, 0.20, and 0.30 ppm, photo-
2
synthetic rate was depressed by 0.04, 0.59 and 1.14 g C09/m per hour, respec-
2
tively, when compared with an initial rate of approximately 2.10 g COp/m per
hour (based on values presented for one example in the study). Miller et al.
(1969) found that 3-year-old ponderosa pine seedlings sustained a 25 percent
reduction in apparent photosynthesis after a 60-day exposure to an 03 concen-
tration of 0.15 ppm for 9 hours per day. Yang et al. (1983) exposed three
clones of white pine, classified by foliar response to 03 as sensitive, inter-
mediate, and insensitive, to 0- concentrations of 0.10, 0.20, or 0.30 ppm for
4 hours per day for 50 days in CSTR chambers. Net photosynthesis was reduced
in the foliage of sensitive and intermediate clones by 14 to 51 percent in
6-29
-------
direct relation to CL dose and relative clonal sensitivity (Table 6-1). In
another study, Coyne and Bingham (1981) measured changes in gross photosynthe-
sis in needles of ponderosa pine trees of various sensitivities to CU. Needles
sustaining slight, moderate, and severe injury exhibited a 90 percent reduction
in gross photosynthesis after exposure to a dose of 800, 700, and 450 ppm-hours
0.,, respectively, in a 3-year time period (2 years for the most sensitive
class of trees). The percentage inhibition in gross photosynthesis was based
on photosynthetic rates of newly emerged needles; no true controls were used
in the experiment. The authors emphasized that the decline in photosynthesis
reflected the superimposition of 0^ effects on normal aging.
6.3.1.5 Tissue and Organ Responses. In addition to depressing photosynthesis
in the foliage of many plant species, 03 inhibits the allocation and transloca-
tion of photosynthate (e.g., sucrose) from the shoots to the roots and other
organs (Tingey, 1974; Jacobson, 1982). Tingey et al. (1971a) found that when
radish plants were exposed to (L (0.05 ppm for 8 hours, 5 days per week for 5
weeks), hypocotyl growth was inhibited 50 percent, while foliage growth was
inhibited only 10 percent (both significant at p < 0.01). Walmsley et al.
(1980) confirmed that radish plants exposed to 0, (0.17 ppm continuously for
36 days) exhibited an altered pattern of assimilation such that below-ground
biomass was more severely affected than foliage. Ponderosa pine exposed to
0.10 ppm 0., for 6 hours per day for 20 weeks stored significantly less sugar
and starch in their roots compared to control plants (Tingey et al. , 1976a).
Such an effect on translocation could reduce root weight and directly affect
the yield of a crop like radish or carrot.
Snap beans exposed to 0- (0.30 ppm or 0.60 ppm for 1.5 hours) exhibited a
greater reduction in root than shoot growth (Blum and Heck, 1980). The root-
to-shoot ratio of crimson clover was suppressed 17 percent and 23 percent,
respectively (p < 0.05), when plants were exposed to 0- at 0.03 and 0.09 ppm
for 8 hours per day for 6 weeks (Bennett and Runneckles, 1977). The root-to-
shoot ratio of rye grass was reduced 22 percent (p < 0.05) when plants were
exposed to 0.09 ppm with the same exposure regime. In other experiments, the
effects of GO were measured on the partitioning of photosynthate in carrot,
parsley, sweet corn, cotton, and pepper (Oshima, 1973; Bennett and Oshima,
1976; Oshima et al., 1978; Oshima et al. , 1979; Bennett et al., 1979). In
each of these experiments, plants were exposed to 0- concentrations of 0.12 to
0.25 ppm for 3 to 6 hours for 0.2 percent to 7 percent of the total growth
period of the plants. In all species but pepper, root dry weight was depressed
6-30
-------
much more than leaf dry weight. For example, root dry weight of cotton was
reduced 60 percent, whereas leaf dry weight was depressed only 17 percent by
03 (Oshima et al., 1979). Ozone had virtually no effect on the dry weight of
parsley leaves, but it reduced root dry weight 43 percent (Oshima et al.,
1978). The photosynthetic rate of tomato plants exposed to 0, (0.3 ppm for
3 hours) was reduced 35 percent and the translocation of photosynthate from
the leaves was reduced 29 percent (McCool and Menge, 1983). This combined
reduction in photosynthate available for root growth can significantly affect
plant growth. The reduction in photosynthate translocation to roots and the
resulting decrease in root size indicates that the plant had fewer stored
reserves, rendering it more sensitive to injury from cold, heat, or water
stress.
When less carbohydrate is present in roots, less energy will be available
for root-related functions. In the 1978 criteria document (U.S. Environmental
Protection Agency, 1978), evidence was presented for 0.,-induced reduction in
nodulation and nitrogen fixation in soybean and ladino clover. Blum and
Tingey (1977) reported that when 2-week-old soybean plants were exposed to an
03 concentration of 0.50 ppm for 4 hours, nodulation was inhibited 60 percent
(p <0.05). Ensing and Hofstra (1982) measured nitrogenase activity in the
roots of red clover 1 and 6 days after the. pi ants were exposed to 0., (0.20 ppm
16 hours per day for 4 days) in non-filtered open-top chambers and found that
nitrogenase activity was reduced 50 and 24 percent (p = 0.05), respectively,
when compared to the activity in plants growing in charcoal-filtered open-top
chambers. By 16 days post-exposure, enzyme activity was comparable to that in
plants given other treatments. An ozone-induced suppression of atmospheric
nitrogen fixation by root nodules could affect total biomass and agricultural
yield, especially in areas where soil nitrogen is low.
6.3.1.6 Secondary Metabolic Responses. In addition to the physiological
effects more directly related to productivity, there are many secondary meta-
bolic responses in a plant exposed to 0-. While these responses do not explain
the initial reaction to 03, they may contribute to the manifestation of foliar
injury. Ethylene is an important stress metabolite produced by many plants
exposed to 0, (Tingey, 1980). Ozone at 0.15 ppm for 8 hours increased ethylene
evolution in beans (Stan et al., 1981). Ozone-enhanced ethylene evolution
ceased prior to the appearance of visible injury. It has been proposed that
ethylene may initiate the observed stimulation of oxidizing enzymes such as
phenylalanine lyase, polyphenoloxidase, and peroxidase (Tingey et al., 1975).
6-31
-------
The accumulation of phenols has been observed in many plant species in response
to 03 (Howell and Kremer, 1973; Hurwitz et al., 1979; Keen and Taylor, 1975;
Koukol and Dugger, 1967). There appears to be a direct relationship between
the concentration of phenols detected in foliage and the extent of visible
injury induced by 03 (Hurwitz et al., 1979). The pigmented lesions that are
visible in the leaf following 03 exposure are thought to occur when phenols
are oxidized and polymerized (Howell and Kremer, 1973).
In summary, ozone enters the cell and initiates biochemical and physio-
logical responses. Critical effects, including reduction in photosynthesis
and a shift in the assimilation of photosynthate, will lead to reduced biomass,
growth, and yield. Visible injury, which results from 0--induced cell injury
and death, reflects the occurrence of both primary and secondary metabolic
events. Visible injury serves as an indicator of the presence of 03 and
reflects potentially harmful effects on plant vigor.
6.3.2 Factors that Modify Plant Response
There is a great deal of variation in the magnitude of plant response to
(L. Biological, physical, and chemical variables influence plant response.
For example, trees in a stand of ponderosa pine will not respond equally to
exposure to (L because of genetic diversity in the sensitivity of individual
trees and because of environmental heterogeneity in the habitat. Plants at
different ages or at different temperatures, humidities, light intensities, or
soil moisture regimes will respond differently to an equivalent 03 exposure.
The presence of several pollutants, chemical sprays, and biological pests all
will contribute to determining the magnitude of plant response to 0,. In
developing an understanding of 0- effects, it is important to consider the 0,
sensitivity of the plant and the environmental conditions it is likely to
experience during exposure. It is equally important to recognize that plants
at certain stages of development or under a given set of environmental condi-
tions may be differentially sensitive to 0,. In the subsequent discussion,
the factors that modify plant response are grouped into three categories:
biological, physical, and chemical factors.
6.3.2.1 Biological Factors
6.3.2.1.1 Genetic Factors. The genetic complement of a plant determines its
potential response to (L. Genetically controlled variation in response to 03
has been observed among species, cultivars, and individuals within a popula-
tion. Inherited variation in plant response to 0- can be measured by using
6-32
-------
many plant response variables. Most researchers have investigated relative 0-.
sensitivity by measuring foliar injury. Genetically controlled differences in
response to 0.,, however, are also reflected in differential yield and physio-
logical effects, as well. A list of the plant species studied that exhibited
differential ozone sensitivity within a species is presented in Appendix B.
The relative 0, sensitivity of cultivars within a species can vary with
dose and the nature of the response measured (Tingey et al., 1972; Heagle,
1979b). There may also be some disparity between the relative sensitivity
ranking of cultivars from controlled CL exposures in a laboratory and exposure
of the same cultivars to ambient air oxidants in the field (Engle and Gabelman,
1966; Taylor, 1974; Huang et al., 1975; Meiners and Heggestad, 1979; Hucl and
Beyersdorf, 1982; DeVos et al., 1983). The inconsistent results may be explained
in part by the nature of the inheritance of the 0, sensitivity. In the case
of onion and bean, one or a few gene pairs were associated with Cu sensitivity
(Engle and Gabelman, 1966; Butler et al., 1979); while for corn (Cameron,
1975), tobacco (Povilaitis, 1967; Sung et al., 1971; Aycock, 1972; Huang et
al., 1975), potato (DeVos et al., 1982) and petunia (Hanson et al., 1976),
several genes determine plant responses to 0.,. The apparent genetic complexity
explains the potential variability in plant response as gene expression changes
during plant development and with variations in the environment.
In agricultural ecosystems there may be some inadvertent selection for
breeding lines tolerant of ozone, as the plant breeder frequently selects for
those plants that perform best under the local growing conditions, There is
no documentation, however, that such inadvertent selection is occurring. In
natural ecosystems in areas receiving long-term 03 stress, it is postulated
that sensitive individuals within a population may decline and be replaced by
those more tolerant to the pollutant (see Chapter 7). Many stresses, including
SOp, elicit this kind of response in populations in natural ecosystems (Taylor
and Murdy, 1975; Roose et al., 1982). Narrowing of the gene pool creates the
potential for increased vulnerability of a plant population to various assaults,
including those of biotic pests.
It appears that as wide a range of sensitivity to 03 exists among plant
species as within them. Ozone is prevalent in most agricultural regions in
the United States. Sensitive plant species are found throughout the country
and the environmental conditions that favor injury occur in many geographic
locations.
6-33
-------
6.3.2.1.2 Developmental factors. Plant foliage appears to be most sensitive
to 03 just prior to or at maximum leaf expansion (U.S. Environmental Protection
Agency, 1978). At this stage, stomata are functional, intercellular spaces
are expanded, and barriers to gas exchange such as internal cutin and secondary
thickening of cell walls are minimal (U.S. Environmental Protection Agency,
1978). Blum and Heck (1980) analyzed the response of bean plants to 0., concen-
trations of 0.30 and 0.60 ppm for 1.5 hours at various stages during growth.
The plants were most sensitive to 03 early in development and just before
senescence. Virginia pine and petunia seem to be most sensitive to (L early
in development, as described in the 1978 criteria document (U.S. Environmental
Protection Agency, 1978). Tolerance of foliage to 03 increased at or just
before appearance of flower buds in plants from six F, hybrid multiflora
petunia lines, at eight physiological ages, that were exposed to 0, (0.20 ppm
for 8 hours) (Hanson et al., 1975). The effect of 03 on root dry weight of
radish was related to timing of the exposure (Tingey et al., 1973a). Plants
exposed to an 03 concentration of 0.40 ppm for 1.5 hours at 7, 14, or 21 days
from seeding, sustained 25, 37, and 15 percent (p < 0.05) inhibition of hypocotyl
root dry weight, respectively. Radish plants may be particularly sensitive to
OT at 14 days because maximum root enlargement begins at that time.
One of the first observations of the effects of photochemical oxidants on
plants in the field was the development of leaf chlorosis followed by premature
leaf aging (senescence) and early leaf drop (abscission) (e.g., Richards et
al., 1958; Menser and Street, 1962). Ozone (0.05 or 0.10 ppm 6 hours per day
for 133 days) induced premature leaf drop in soybeans (Heagle et al., 1974).
The premature senescence and leaf drop increased throughout the study period.
Ozone-induced premature leaf senescence has been observed in both greenhouse
and field-grown potatoes (Heggestad, 1973; Pell et al., 1980). Field studies
with white beans (Hofstra et al., 1978) confirmed that premature leaf drop was
induced by 0-; the premature leaf drop was associated, in part, with the
0^-induced yield reductions. The photosynthetic rate of hybrid poplars exposed
to 03 (0.085 or 0.125 ppm for 5.5 hours per day for 65 days) decreased more
rapidly with age than unexposed plants, indicating that 03 induced a premature
senescence (Reich, 1983). Another study with hybrid poplar showed that 03
(0.04 ppm 12 hours per day for 5 months) significantly increased leaf drop
(Mooi, 1980). The effects of 03 on the senescence process, regardless of time
of initiation, may be responsible for many of the documented reductions in
yield.
6-34
-------
6.3.2.1.3 Pollutant-plant-pest interactions. Plant pests (pathogens and in-
sects) are normal components of both agro- and natural ecosystems. Crop
losses from pests can be significant and have been estimated at 20 to 30
billion dollars per year in the United States alone (James, 1980). When
considering the effects of 0, on crop plants or forests, it is important to
realize that the pollutant does not occur alone, but rather in conjunction
with other stresses that are modifying the productivity of the system. The
purpose of this section is to indicate what is known about interactions among
03, plants, and pests, and how these interactions might modify the effects of
03 on the quality, quantity, or the intended use of the plant.
Disease is the result of a complex interaction between host plant, environ-
ment, and pathogen. In the context of this general discussion of biotic
stress, problems caused by pathogens and insects will both be termed disease.
To understand the ways in which 0,, as a part of the environment, may modify
pest dynamics, it will be helpful to consider a generalized disease cycle.
The cycle begins with the arrival of the inoculum or pest at the plant
(host). Following deposition of the pest on the plant surface, in the pre-
sence of favorable conditions (temperature, moisture), penetration of the
plant (or insect feeding, or oviposition) may begin.
Host penetration may occur quickly or, in some cases, the pathogen may
live as a resident on the plant surface for a period of time. Once penetration
occurs, and favorable conditions are present, infection may occur that results
in an intimate relationship between plant and pathogen. Growth and development.
or colonization by the pathogen or plant pest proceeds until the pest reaches
a reproductive stage. Propagules of the pest are formed and dispersed either
passively or actively.
At each stage of this cycle, 03 may modify the success of the pest,
either directly through effects on the invading organisms, or indirectly,
through modification of the host plant. Similarly, the complex interaction
between plant and pest may alter the sensitivity of the plant to 0.,.
6.3.2.1.3.1 Pollutant-plant-pathogen interactions. Most pollutant-plant-
pathogen interaction studies have been conducted under controlled laboratory
conditions, but a few field studies have been performed. This topic has been
reviewed recently (Heagle, 1973, 1982; Laurence, 1981; Manning, 1975; Treshow,
1980a; U.S. Environmental Protection Agency, 1978). The results of published
studies are summarized in Table 6-2.
6-35
-------
TABLE 6-2. PLANT AND BIOTIC PATHOGEN INTERACTIONS AS INFLUENCED BY VARIOUS DOSES OF OZONE UNDER LABORATORY AND FIELD CONDITIONS
Plant/pathogen
Exposure
Experimental
conditions Effect on disease
Effect on
pollutant
injury '
Reference
AGRONOMIC CROP/FUNGI
Pinto bean/root fungi
Barley/Erysiphe graminis
Wheat/Puccinia graminis
en Wheat/Puccinia graminis
I
OJ
en
Corn/Helminthosporium maydis
Oats/Puccinia coronata
Potato/Botrytis cinerea
0.10 ppm 03, 8 hr daily, 10 wk
0.15 ppm 03, 6 hr daily for 4, 6,
or 8 exposures after inoculation
0.06 to 0.18 ppm 03, 6 hr daily,
17 days after inoculation
0.1 ppm 03) 6 hr daily, 12 days
after inoculation
0.06 to 0.18 ppm 03 6 hr variable
days before and after inoculation
0.10 ppm 03, 6 hr/day 10 days after
inoculation
0.20 ppm 03/3 hr, 1 to 5 days after
inoculation
0.15 to 0.25 ppm 03, 6 to 8 hr
Cabbage/Fusarium oxysporium 0.10 ppm 03> 8 hr daily, 10 wk
0.15 ppm 03, 4 hr
Onion/Botrytis cinerea,
B. squamosa
Potato/AIternaria solani •
Tomato/Glomus fasciculatus
0.03 to 0.04 ppm 03 monthly
0.30 or 0.60 ppm, 3 hr/wk
for 8 wks
L Increased number fungal colonies NR
L Increased colony size NR
Decreased hyphal growth, numbers Decreased
of spores, infection
Reduced sporulation NR
Increased lesion size, increased NR
number of spores produced at
highest concentration
No effect on disease development NR
L Increased disease development NR
L Decreased disease development NR
FC Increased disease development NR
F Increased disease development NR
L Retarded infection NR
Manning et al. (1971)
Heagle and Strickland
(1972)
Heagle and Key (1973a)
Heagle (1975)
Heagle (1977)
Heagle (1970)
Manning et al. (1969)
Manning et al. (1971a)
Wukasch and Hofstra
(1977a,b)
Bisessar (1982)
McCool et al. (1982)
-------
TABLE 6-2 (cont'd). PLANT AND BIOTIC PATHOGEN INTERACTIONS AS INFLUENCED
BY VARIOUS DOSES OF OZONE UNDER LABORATORY AND FIELD CONDITIONS
cn
i
CO
Plant/pathogen
TREES AND ORNAMENTALS/FUNGI
White pi ne/Lophodermi um
pinastri
Ponderosa, Jeffrey Pine/
Heterobasidion annosum
Eastern white pine/
Verticicladiella procera
Lilac/Microsphaera aim'
Poinsettia/Botrytis cinerea
Geranium/Botrytis cinerea
Exposure
0.07 ppra 03, 4.5 hr
0.18 ppm 03/12 hr
seasonal
0.045 ppm 03 monthly average
0.128 ppm monthly peak hourly
0.25 ppm 03, 72 hr
0.15 to 0.45 ppm 03, 4 hr
0.15 ppm 03, 6 hr, 2x at 24-hr
Experimental
conditions
L
F
F
L
L
L
Effect on disease
Slight increased disease
occurrence
Increased disease development
Increased colonization of stumps
Increased disease incidence
No influence on germination, early
fungal development
No effect
Reduced sporulation; reduced
Effect on
pol lutant
injury •
NR
NR
NR
NR
NR
NR
NR
Reference
Costonis and Sinclair
(1972)
James et al. (1980a)
James et al . (1980b)
Skelly (1980)
Hibben and Taylor (1975)
Manning et al. (1972)
Krause and Weidensaul
Geranium/Botrytis cinerea
Citrus/Glomus fasciculatus
intervals after inoculation
0.07 to 0.10 ppm 03 10 hr daily for L
15 to 30 days
0.45 ppm 3 hr/day, 2 days/wk L
for 19 wks
infection by exposed spores
Flocculent material produced
Increased disease development
when visible 03 injury evident
Decreased infection
(1978a)
NR Manning et al. (1970b)
NR McCool et al. (1979)
-------
TABLE 6-2 (cont'd). PLANT AND BIOTIC PATHOGEN INTERACTIONS AS INFLUENCED
BY VARIOUS DOSES OF OZONE UNDER LABORATORY AND FIELD CONDITIONS
Plant/pathogen
AGRONOMIC CROPS/VIRUS
Tobacco/tobacco mosaic
Tobacco/tobacco etch
01
co Tobacco/tobacco streak
00
Tobacco-pinto bean/tobacco
mosaic
Pinto bean/bean common mosaic
Pinto bean/alfalfa mosaic,
tobacco ringspot,
tobacco mosaic,
tobacco ringspot
Tomato/tobacco mosaic,
cucumber mosaic
Experimental
Exposure conditions
0.30 ppm 03, 6 hr L
Seasonal maximum hour, 0.236 ppm 03 F
0.25 ppm 03, 4 hr, once 9 days after L
inoculation
0.30 ppm 03, 3 hr for 1 or 2 days L
0.35 ppm 03, 4 hr; 0.25 ppm 03, L
3 hr, respectively
0.25 ppm 03, 4 hr, 5 days after L
inoculation
0.25 ppm 03 4 hr, 5 days after
inoculation
0.0 to 0.45 ppm or 0 to 0.90 ppm L
3 hr; 7 to 21 days after
inoculation
Effect on
pol lutant
Effect on disease injury '
NR < 03
< 03
NR < 03
NR > 03
NR < 03
NR < 03
NR < 03
NR > 03
7
< 03
at
injury
injury
injury
injury
injury
injury
injury
injury at
or 14 days
injury
21 days
Reference
Brennan and Leone (1969)
Bisessar and Temple
(1977)
Moyer and Smith (1975)
Reinert and Gooding
(1978)
Brennan (1975)
Davis and Smith (1975)
Davis and Smith (1976)
Ormrod and Kemp (1979)
Soybean/tobacco ringspot
0.35 to 0.40 ppm 03, 4 hr, once 6,
8, or 10 days before inoculation
NR
< 03 injury
Vargo et al. (1978)
-------
TABLE 6-2 (cont'd). PLANT AND BIOTIC PATHOGEN INTERACTIONS AS INFLUENCED
BY VARIOUS DOSES OF OZONE UNDER LABORATORY AND FIELD CONDITIONS
Plant/pathogen
Exposure
Experimental
conditions
Effect on disease
Effect on
pollutant
injury '
Reference
AGRONOMIC CROP/BACTERIA
Alfalfa/Xanthomonas alfalfae
White bean/Xanthomonas
phaseoli
Soybean/Pseudomonas glycinea
en Ladino clover/Rnizobium sp.
i
CO
= Increased; < = decreased.
CNR = Not reported.
Source: Modified from Laurence (1981).
-------
Infection of plants by pathogens may be inhibited or stimulated by (L.
Manning et al. (1969; 1970a,b) found that potato and geranium leaves injured
by 03 (0.07 to 0.25 ppm, 6 to 10 hours) had a larger number of lesions caused
by Botrytis. Wukasch and Hofstra (1977a) found that field-grown 0~-injured
onion plants developed twice as many Botrytis squamosa lesions as did uninjured
plants growing in charcoal-filtered air. The same authors (1977b) found fewer
natural B. squamosa lesions on plants that had been treated with an antioxidant
chemical having no fungicidal activity. Ambient air 03 concentrations exceeded
0.15 ppm for 4 hours and 0.08 ppm on several occasions during the growing
season. Bisessar (1982) found similar results with the interaction of 03,
potato, and Alternaria solani. The fungus colonized 0.,-injured sites on
potato leaves, and fewer lesions were present on plants protected from 0- with
ethylene diurea (EDU), a compound developed to reduce 0- injury (see Section
6.3.2.3.2). Ambient air 03 concentrations exceeded 0.08 ppm during 68 hours,
and the highest measured concentration was about 0.14 ppm. Similar results
were obtained by James et al. (1980a) in a field study of Heterobasidion
annosum (syn. Fomes annosus) infection of oxidant-injured ponderosa and Jeffrey
pines in the San Bernardino Mountains. They found increased infection of the
roots of severely 0.,-injured trees. The results of the field study were con-
firmed under controlled laboratory conditions. They also found that the coloni-
zation of roots and freshly cut stumps of ponderosa and Jeffrey pine was posi-
tively correlated with the severity of the oxidant injury observed on needles.
In laboratory studies, colonization of both species was directly related to 0-
exposure over the range of 0 to 0.45-ppm for 58 to 92 days (see discussion in
Chapter 7). Skelly (1980) reported increased incidence of root disease caused
by Verticicladiella procera in oxidant-injured eastern white pines in Virginia.
Ozone can inhibit infection of plants by pathogens. In general, infection
by obligate parasites is inhibited in plants that have been exposed to elevated
concentrations of 0- (Heagle 1970, 1973, 1975, 1982; Heagle and Strickland,
1972; Heagle and Key, 1973a,b).
McCool et al. (1979). reported that infection of citrus by Glomus fasciculatus,
an endomycorrhizal fungus, was decreased by exposure to 0, (0.45 ppm, 3 hours
per day, 2 days per week for 19 weeks). Exposure of tomato to 0.30 ppm 03 for
3 hours once weekly for 8 weeks retarded infection by the same fungus (McCool
et al., 1982). These exposures did not affect root growth of the plants or
sporulation by the fungus, but did reduce the number of successful infections.
Ozone reduced mycorrhizal infections of tomato roots 46 and 63 percent when
6-40
-------
the plants were exposed to 0.15 (3 hours per exposure, twice weekly for 9
weeks) or 0.30 ppm (3 hours once weekly for 9 weeks), respectively. Rhizobiuni,
a nitrogen-fixing bacterium of legumes, induced fewer nodules in soybean
plants exposed to 0.75 ppm 03 for 1 hour (Tingey and Blum, 1973) and in ladino
clover exposed to 0.3 or 0.6 ppm 03 twice for 2 hours each (Letchworth and
Blum, 1977).
Infection of soybean by Pseudomonas glycinea was decreased when plants
were exposed to 0.08 or 0.25 ppm 03 for 4 hours at times ranging from 8 days
to 1 hour before inoculation. When exposures occurred more than one day after
inoculation, however, inhibition was not observed (Laurence and Wood, 1978a).
Similar results were found with Xanthomonas fragariae and wild strawberry
(Fragaria virginiana) (Laurence and Wood, 1978b). Temple and Bisessar (1979),
however, did not find fewer Xanthomonas phaseoli lesions on 0--injured white
beans in the field.
In most cases, colonization of plant tissue by pathogens is assessed by
measuring lesion size. Lesions of obligate parasites are usually smaller on
plants exposed to 0, when compared to controls (Laurence, 1981). Heagle and
Strickland (1972), however, found larger colonies of Erysiphe graminis f. sp.
hordei on barley plants that were exposed repeatedly to low concentrations of
03 (up to 0.15 ppm, 6 hours per day for 8 days).
Little is known about colonization of ozone-affected plants by facultative
parasites. Heagle (1977) inoculated corn plants with Helminthosporium maydis
race T and exposed them to 0, (0.06, 0.12, or 0.18 ppm) for 6 hours per day
for up to 7 days before inoculation, 9 days after inoculation, or combinations
of before and after. He found that lesion length was significantly increased
by 0- exposure (0.18 ppm) before and after inoculation, but was not affected
at other concentrations or time regimes.
Based on these few reports on the relationship of 0, to plant colonization
by pathogens, it is impossible to generalize and predict effects in particular-
disease situations. It is apparent that the outcome of a pollutant-plant-path-
ogen interaction depends on the particular plant and pathogen involved. It
also is affected by the environmental conditions and 03 concentrations before
and after inoculation.
Rist and Lorbeer (1981) recently reviewed the effects of 03 on sporula-
tion of fungi. In axenic culture, sporulation and growth of fungi isolated
from leaf surfaces were almost always inhibited or unchanged by exposure to
03. In a few studies, significant inhibition of growth, sporulation, or
6-41
-------
germination has been observed following exposures to concentrations as low as
0.10 ppm for 4 hours I but fungi often are resistant to 1.0 ppm 0, for several
hours. Germination of spores produced during 03 exposure (0.15 or 0.30 ppm, 6
hours per day for 2 days) may also be lower than that of controls (Krause and
Weidensaul, 1978a,b). These spores may subsequently be less successful in
colonizing the leaf surface (Krause and Weidensaul, 1978a,b). Both decreases
and increases in sporulation have resulted from 03 exposure of infected plants
(Laurence, 1981), and the particular result seems to depend on the plant-path-
ogen combination and the specific 03 exposure regime.
In the case of bacterial diseases, reproduction of the pathogen is gener-
ally reflected in the size of lesions on the plant. Bacteria are generally
resistant to ambient concentrations of 03, but may be much more sensitive to
changes in plant metabolism induced by 03 (Hughes and Laurence, 1984).
Reproduction of the soybean cyst nematode and the stubby root nematode
was reduced by exposure of infested soybean plants to 0.25 ppm 03 applied on
three alternate days a week for about 2 months (Weber et al., 1979). Similar
03 treatments also reduced the reproduction of a foliar nematode on begonia
plants. This reduction was related to the amount of 0^-induced leaf injury
(Weber et al., 1979).
Only a few studies have been reported that relate the effects of 03 in
combination with another pollutant (SOp) to disease development. Weidensaul
and Darling (1979) found that Scotch pines inoculated with Scirrhia acicola
and exposed to 03 (0.20 ppm for 6 hours) or 03 combined with SO^ (0.20 ppm
each for 6 hours) had fewer lesions than controls, but did not differ from
each other. More lesions formed when inoculation preceded fumigation by 5
days than when inoculation followed exposure by 30 minutes.
6.3.2.1.3.2 Effects of ozone on plant-insect interactions. The effects
of air pollutants on insect populations were reviewed recently (Alstad et al.,
1982). Very little is known about 03~insect interactions. Ozone-induced
injury in ponderosa pine has been shown to predispose trees to subsequent
invasion by several species of pine bark beetles (Stark et al., 1968). Elden
et al. (1978) found that 03 injury induced by exposures of 0.20 ppm for 4
hours had little or no effect on the development of pea aphids on alfalfa.
They did note that two of three varieties having higher degrees of 03 resis-
tance also had greater resistance to pea aphid.
6-42
-------
6.3.2.1.3.3 Effects of pathogen infection on plant sensitivity to 0....
Fungal, bacterial, or viral infections have been reported to provide some
protection to plants from the visible effects of 0.,. Although of interest
mechanistically, most of the studies have been conducted under controlled
conditions, and it is questionable whether they are relevant in field situa-
tions.
Yarwood and Middleton (1954) noted that pinto bean leaves infected by
Uromyces phaseoli were less sensitive to photochemical oxidants than uninfected
leaves. Similar results have been observed with many pathogen-plant combina-
tions. The protection afforded by fungal and bacterial pathogens is usually
localized at the margins of lesions, while virus infections can provide more
generalized effects (Heagle, 1982).
Although bacterial pathogens often provide protection against 03 injury
near lesions, they did not in the case of bacterial blight of soybean or
angular leafspot of strawberry (Laurence and Wood, 1978a,b). Pratt and Krupa
(1979), however, reported that in chlorotic soybean leaves Pseudomonas glycinea
infection did inhibit expression of 0., symptoms. Temple and Bisessar (1979)
found less visible 03 injury on Xanthomonas phaseoli-infected white beans in
the field in Ontario, Canada. Using the same species of bacterium, Olson and
Saettler (1979) observed no protection from 0, injury in controlled laboratory
experiments. Pell et al. (1977) investigated the interaction between 0- and a
species of Pseudomonas that caused a hypersensitive reaction in soybean. They
found that inoculation with the pathogen provided some protection from 0- when
plants were inoculated 1 day before exposures to a relatively high concentra-
tion of the pollutant (0.35 ppm for 2 hours). The effect was not observed
when inoculation took place 4 hours before exposure.
Many reports have appeared on the effects of virus infection on plant re-
sponse to OT, beginning with those of Brennan and Leone (1969) and Brennan
(1975). Davis and Smith (1975, 1976) reported protection of pinto bean leaves
from On injury following inoculation with common mosaic, tobacco ringspot,
tomato ringspot, alfalfa mosaic, or tobacco mosaic viruses. The protection
depended upon an establishment time of 4 to 5 days between inoculation and
exposure, which was apparently linked to the time required to attain sufficient
virus titer to afford protection. The protection was localized except in the
case of tobacco ringspot, in which a more general effect was observed. Infec-
tion with tobacco etch virus also protected tobacco plants from 0, injury
6-43
-------
(Moyer and Smith, 1975). All experiments were under controlled conditions
with exposures of 0.25 ppm 0, for 4 hours.
Virus infection in one part of a plant has also been shown to provide
protection against (L injury in other parts. Davis and Smith (1976) found
that inoculation of one primary leaf of a pinto bean plant resulted in some
degree of protection in the uninoculated leaf exposed to "0~ (0.20 ppm for
4 hours), but was not effective at 0- concentrations greater than 0.20 ppm.
Vargo et al. (1978) found that sensitivity to 0~ (0.35 to 0.4 ppm for 4 hours)
of the primary leaf opposite the leaf inoculated with tobacco ringspot virus
was decreased with increasing time after inoculation. They also found that as
virus-induced apical necrosis increased, less foliar 0~ injury occurred.
Two reports show that 0~ injury may be increased following virus infection.
Reinert and Gooding (1978) found that tobacco plants systemically infected
with tobacco streak virus and exposed to 0, (0.3 ppm for 3 hours on 1 or 2
days) 3 weeks after inoculation displayed more injury than the combined injury
of plants exposed to 0- or virus. Ormrod and Kemp (1979) found both increases
and decreases in 0, sensitivity of tomato plants infected with cucumber mosaic
virus or tobacco mosaic virus or both, depending on the tomato cultivar, 0.,
concentration, the virus, and the virus incubation period. Ozone injury was
observed more frequently on tobacco mosaic virus-infected plants than on those
inoculated with cucumber mosaic virus. They also observed that increases in
0, injury usually occurred when 0~ exposures (0.15 to 0.90 ppm for 3 hours)
occurred within 14 days of inoculation; 21 days after inoculation, most of the
differences observed were decreases in injury.
In the only field study reported, Bisessar and Temple (1977) found 60
percent less oxidant injury on tobacco plants infected with tobacco mosaic
virus than on uninfected plants. Ozone concentrations exceeded 0.10 ppm for
16 percent of the daylight (6:00 a.m. to 8:00 p.m.) hours during the study.
The effects described in the above sections are not of commercial impor-
tance, but the observations may provide some information as to the mode of
action of 0- in plants.
Ozone affects the development of disease in plant populations. Most
laboratory evidence indicates that 0,, at ambient concentrations or higher for
4 hours or more, inhibits infection by pathogens and subsequent disease develop-
ment; however, increases in disease development have been noted in certain
cases. Most often these increases occur with "stress pathogens," such as
Botrytis or Heterobasi'di'on annosum, that incite diseases such as blight of
6-44
-------
potatoes or onions or root rot of ponderosa and Jeffrey pine. Increases in
disease development have been observed in these host-parasite relationships
under both laboratory and field conditions (plants exposed to ambient air
levels of 03).
That ozone can also modify plant-insect relationships is best illustrated
by studies conducted in the San Bernardino Mountains that showed increased in-
vasion of Distressed pine trees by bark beetles.
The mode of action of 0.,-plant-pest interaction probably involves indirect
effects on the pathogen or insect that are the result of the direct interaction
of 03 and the plant. Effects on disease development have been documented at
concentrations of 0- and durations of exposure that are considered to be low
(i.e., < 0.10 ppm for a few hours). Thus, it would appear that 0, is affecting
plant metabolism at these low concentrations and short exposure durations.
6.3.2.2 Physical Factors. The environment of the plant is composed of various
biological, chemical, and physical factors that change throughout the plant
growth period. The physical factors (e.g., light, temperature, relative
humidity, soil moisture, and soil fertility) interact to provide the conditions
for, and also govern, plant growth. Short-term variations in one or several
of these environmental factors, if they coincide with a pollution episode, may
render the plant more or less sensitive to pollutants.
Environmental conditions before and during plant exposure are critical to
the plant response, while post-exposure conditions are less important. Although
the influence of physical factors on plant response to 0, has been studied
primarily under laboratory and greenhouse conditions, field observations have
often substantiated these results. Most studies have evaluated the effects of
a single environmental factor and have usually used foliar injury as the
measure of plant response. Information sufficient to make some generali-
zations about the influence of these factors on plant response to 0^ is avail-
able; but for most factors, substantial uncertainty exists because of the
small number of species studied and the lack of information on the interactions
of the environmental factors.
In this section, the various environmental factors will be discussed
individually for organizational convenience, even though these factors interact
to influence plant growth and sensitivity to 0~. Most studies in the following
sections have used exposures to high 03 concentrations that would rarely, if
ever, be encountered in the ambient air. These studies were included because
they illustrate the range of plant responses to various physical factors.
6-45
-------
6.3.2.2.1 Light. It was concluded in the 1978 criteria document (U.S. Environ-
mental Protection Agency, 1978) that a short photoperiod and a relatively low
light intensity during growth maximize (L-induced foliar injury. These results
were consistent across contrasting light regimes. For example, bean and
tobacco plants were more sensitive to CL at 0.4 ppm for 1 hour if grown at 420
-1 -2 -1 -2
uE s m than-if grown at 840 uE s m (Dunning and Heck, 1973). Cotton
-1 -2
grown at 276 uE s m was less sensitive to 0~ concentrations of 0.9 ppm for
-1 -2
1 hour than similar plants grown with 27.6 uE s m (Ting and Dugger, 1968).
Subsequently, Dunning and Heck (1977) demonstrated the complex nature of
environmental interactions. They reported that tobacco showed increased
sensitivity to an 0, concentration of 0.40 ppm for 1 hour when grown under
-1 ~2
high light intensity (840 uE s m ) and subsequently exposed at an interme-
-1 -2
diate light intensity (420 uE s m ). In contrast, pinto bean leaves were
most sensitive when plants were grown at a lower light intensity (209 (jE s
_2
m ) and subsequently exposed at the high intensities cited above.
In the field, vegetation will not often be exposed to 0- at the low light
intensities and the short photoperiods (8 hours) used in simulations described
above. Therefore, special consideration of light may not be as relevant as
other environmental factors. There are, however, some cultural practices for
which light intensity and photoperiod are controlled. Shade-grown tobacco and
bedding plants (in the commercial floriculture industry) represent two examples
of production settings in which low light intensity is used and where losses
attributable to oxidants have been documented.
6.3.2.2.2 Temperature. The 1978 criteria document (U.S. Environmental Protec-
tion Agency, 1978) reported that there was no consistent pattern relating tem-
perature to plant response to 0.,. Radish was more sensitive to 0-, if grown
under cool conditions, whereas snap bean, soybean, Bel W-3 tobacco, Virginia
pine, and white ash were sensitive if grown under warm conditions (U.S Environ-
mental Protection Agency, 1978). Miller and Davis (1981a) found that pinto
bean plants exposed to 0, at a concentration of 0.10 ppm for 3 hours at 15° or
32°C sustained more severe foliar injury than when the exposure temperature
was 24°C. Dunning and Heck (1977) also found that bean plants were more
sensitive to 0- when exposed at 16° or 32°C rather than at 21° or 27°C.
Tobacco behaved differently from bean, exhibiting less sensitivity to 0.40 ppm
Oo for 1 hour when the exposure temperature was 32°C as opposed to 16°, 21°,
or 27°C.
6-46
-------
The effects of temperature on plant response to 03 are probably both
physical and biological. Temperature affects solubility of gases, enzymatic
reactivity, membrane conformation, and stomatal movement. The disparate 03
responses of various plant species grown at different temperature regimes may
also reflect morphological or biochemical differences or both.
6.3.2.2.3 Relative humidity. It was concluded in the 1978 criteria document
(U.S. Environmental Protection Agency, 1978) that, in general, plants seem to
be more sensitive to 0, when growth or exposure, or both, occur under condi-
tions of high relative humidity (RH). Table 6-3 is a modification of a summary
table in the 1978 criteria document (U.S. Environmental Protection Agency,
1978). Dunning and Heck (1977) reported that the sensitivity of tobacco to 0~
(0.40 ppm for 1 hr) was not affected by the relative humidity during growth
until, the level reached 90 percent RH, at which point plants became more
tolerant to 03- Mclaughlin and Taylor (1981) have demonstrated that, in pinto
bean plants exposed to 0, concentrations of 0.079 ppm for 2 hours, uptake of
the pollutant increased fourfold when the exposure RH was increased from 35
percent to 73 percent. At the low RH (35 percent), 03 uptake decreased when
the pollutant concentration exceeded 0.079 ppm, while at the higher RH (73
percent) 0- uptake increased with increasing 0- concentration.
The influence of RH on stomatal function may help to explain the influence
of RH and plant responses to 0~. As RH decreases, a water deficit can develop
in the guard cells, and stomatal closure occurs to minimize internal foliar
water deficit (Ludlow, 1980). Stomatal closure would reduce 0_ flux into the
leaf. The influence of RH on plant sensitivity may explain important varia-
tions in plant response under field conditions.
6.3.2.2.4 Soil moisture. Plant response to oxidants is modified by soil
moisture, probably through an influence on stomatal function. As soil moisture
decreases, water stress increases and there is a reduction in plant sensitivity
to 03. In the previous criteria document (U.S. Environmental Protection
Agency, 1978), the major studies on effects of soil moisture prior to 1978
were reviewed and examples are shown in Table 6-4. More recently, Harkov and
Brennan (1980) demonstrated that potted hybrid poplar plants were more tolerant
of 0, concentrations of 0.10 ppm after 6 to 9 days without water. Olszyk and
Tibbitts (1981) found that pea plants exposed to 03 concentrations of 0.23 ppm
for 2 hours exhibited less foliar injury when the plant water potential was
-388 kPa than when it was -323 kPa (reflecting relatively low and high soil
moisture levels, respectively).
6-47
-------
TABLE 6-3. RESPONSE OF PLANTS TO OZONE AS CONDITIONED BY HUMIDITY DURING GROWTH AND EXPOSURE1
Plant species
Pine, Virginia
Bean, cultivar
Pinto
Tobacco,
cultivar Bel W,
f Ash, white
4^
CO
Tobacco,
cultivar Bel W,
Bean, cultivar
Pinto
Bean cultivar
Pinto, and
Tobacco, cultivar
Bel W,, averaged
Ozone
concentration,
ppm
0.25
0.25
0.25
0.40
0.40
0.25
0.30
0.20
0.40
Exposure
duration,
hr
4
4
4
1
1
4
1.5
1.5
1
Notesb
3-yr seedlings
Juvenile
Juvenile
8_hr PP; 420 uE s"1
m 2 control condi-
tions; 8 hr PP
8_hr PP; 420 uE~x
m 2 control condi-
ditions, 8 hr PP
1-yr seedlings
31°C
31°C
8 hr PP
8 hr PP
Growth or
exposure
Exposure
Growth
Exposure
Growth
Exposure
Growth
Exposure
Growth
Exposure
Post-exposure
Exposure
Exposure
Growth
45% EH
90% EH
Exposure
75% GH
Response,
60% RH
4
50
1
60% RH
66
52
60% RH
42
33
60% RH
38
36
26% RH
9
26% RH
0
45% RH
36
73
41
% injury
85% RH
25
58
35
80% RH
78
67
80% RH
36
36
80% RH
46
41
41
51% RH 95% RH
39 50
51% RH 95% RH
0 55
60% RH 75% RH 90% RH
39 41 31
67 81 80
53 70 81
Modified from 1978 criteria document (U.S. Environmental Protection Agency, 1978); all the studies were conducted in controlled
environment facilities.
PP = photoperiod, GH = relative humidity during growth, EH = relative humidity during exposure.
"Time when humidity treatment was applied.
Relative humidity levels during growth or exposure as indicated.
-------
TABLE 6-4. EFFECTS OF SOIL MOISTURE ON RESPONSE OF SELECTED PLANTS TO OXIDANT
en
i
Plant species
Tomato, cultivar
Firebal 1
Beet, garden
Bean, cultivar
Pinto
Ozone
Concentration
ppm
1.00
1.00
0.50
1.00
0.00
0.20
0.00
0.15
0.25
0.00
0.15
0.25
exposure
i
Duration
1.5 hr
1.0 hr
1.0 hr
1.0 hr
3 hr (daily for 38
days)
2 hr/day (63 days)
2 hr/day (63 days)
2 hr/day (63 days)
2 hr/day (63 days)
Type of response
Reduction in chlorophyll
Reduction in chlorophyll
Reduction in chlorophyll
Reduction leaf dry wt
Reduction in dry wt of
storage root from
nonsaline control
Reduction in shoot dry
wt from nonsaline
control
Reduction in root
dry wt from nonsaline
control
Response,
High
moisture
90% turgid
54
67
36
48
-40 kPa
0
40
-40 kPa
0
27
93
0
25
91
% reduction
Medium
moisture
80% turgid
10
24 b
(40)b
-440 kPa
24
52
-200 kPa
18
42
91
25
28
89
from control
Low
moisture
-840 kPa
68
69
-400 kPa
78
87
88
65
78
79
Special soil moisture conditions are underlined; kPa = kilopascals; % turgid indicates amount of water in the plant leaf.
A stimulation rather than a reduction.
Source: Modified from Table 11-9, U.S. Environmental Protection Agency (1978).
-------
It appears that the stomata of plants grown under soil moisture stress
close more rapidly in the presence of CL than stomata of plants under optimal
water availability (Tingey et al., 1982; Olszyk and Tibbitts, 1981; U.S.
Environmental Protection Agency, 1978). Such a plant response would reduce 03
ingress and confer some resistance to 0- injury.
Tingey et al. (1982) found that the leaf conductance of bean plants that
were water-stressed decreased, compared with nonstressed plants, 24 hours
after the stress was applied. A coincident reduction in plant response to 0~
(1 ppm for 1 hour) occurred. If plants were water-stressed for 7 days and
then the water stress was relieved, leaf conductance and plant response to 0-
both increased.
A 2-yr field study was conducted to determine the effects of 0., on the
yield of normally irrigated and water-stressed cotton (Temple et al., 1985a).
In the first year, which was hot and dry, ambient 03 reduced the yield of
cotton by 20 percent in the normally irrigated plots but did not affect the
yield of the water-stressed plants. The second year was cooler, had less
evapotranspiration, and had significantly less 0- than the first. Under these
conditions, cotton at both soil moisture treatments displayed the same response
to CL. The ambient 0, reduced cotton yield by 15 percent.
Plants subject to long-term soil moisture stress may also exhibit morpho-
logical or functional changes, or both, that could modify the CL response.
Drought or salt stress, which can confer long-term moisture stress, are more
limiting to plant health than the air pollution stress that they may modify;
hence, any of their protective effects are offset (U.S. Environmental Protec-
tion Agency, 1978).
It is important to recognize that plants grown under optimal soil moisture,
as in irrigated fields or greenhouses, generally are particularly vulnerable
to 0, injury. On this basis, vegetation in natural ecosystems, for example,
would be expected to be more sensitive to 0- in years of normal rainfall than
in years of drought.
6.3.2.2.5 Soil fertility. Nutrient balance is fundamental to plant growth;
any imbalance could lead to variations in the 0~ response. Plant nutrients,
including nitrogen, phosphorus, potassium, and sulfur, may all influence plant
response to 0~ (U.S. Environmental Protection Agency, 1978). Results of
studies cited in the 1978 criteria document (U.S. Environmental Protection
Agency, 1978) were inconsistent for a variety of reasons, including species
differences and differences in experimental protocols and designs. Since
6-50
-------
then, additional data have appeared, but the relationship between soil fertil-
ity and Oo sensitivity has not been clarified. Markov and Brennan (1980) grew
hybrid poplar seedlings with varied amounts of slow-release fertilizer, 18:16:12
(N:P:K), that yielded plants with foliar contents of 1.53, 2.69, 3.12, or 3.47
percent nitrogen. Visible injury was greatest in leaves containing 2.69
percent nitrogen when plants were exposed to an 0, concentration of 0.10 ppm
for 6 hours. Using an N:P:K ratio of 6:25:15, Heagle (1979a) found that
potted soybean plants exposed to an 03 concentration of 0.60 ppm for 1.5 hours
were more sensitive when fertilized with 100 ml of solution at a rate of 0 or
7.5 g fertilizer/3.8 liters of water than when 15 or 22.5 g/3.8 liter of water
was used. Optimum soybean growth was observed at fertilizer rates of 15.0 and
22.5 g/3.8 liters of water. Noland and Kozlowski (1979) reported that silver
maple became more sensitive to 0, (0.30 ppm for 6 hr for 2 successive days)
when grown with 117 ppm potassium as compared to 0 to 2 ppm potassium for
6.5 wk. The authors suggested that potassium may stimulate the guard cells to
open, thereby increasing the uptake of 03 by this species. Dunning et al.
(1974) found that pinto bean and soybean foliage were injured more severely by
Oo when plants were grown with low potassium levels (105 meq/liter) rather
than normal levels (710 meq/liter). Greenhouse studies of tobacco showed a
negative correlation between the calcium content of the leaf tissue and 0.,-
induced (0.25 ppm for 4 hr) foliar injury (Trevathan and Moore, 1976). This
result was observed at eight combinations of 0, concentration and exposure
duration. Additional explanations for the variable response of plants to 0,
when grown with different fertility regimes have not been formulated.
6.3.2.3 Chemical Factors. The chemical environment of plants (e.g., air
pollutants, herbicides, fungicides, insecticides, nematocides, antioxidants,
and chemical protectants) influences plant responses to 0,. These factors may
be grouped into the subject areas of pollutant interactions and chemical
sprays.
6.3.2.3.1 Pollutant interactions. Components of ambient atmospheres such as
SOp, NOp, and other pollutants may change, modify, or alter plant sensitivity
to 03< These substances all contribute to intensifying or reducing the effects
of 0- on the quality, quantity, or intended use of the plant and must be
considered along with the discussion of biological (Section 6.3.2.1) and
physical (Section 6.3.2.2) factors that modify plant responses to 0~. The
magnitude of these modifications depends on the plant species, cultivar,
6-51
-------
pollutant concentration, duration and frequency of exposure, and the environ-
mental and edaphic conditions in which plants are grown.
The study of the effects of pollutant combinations on plants is based on
the premise that pollutants co-occur in the atmosphere, and that together
they may induce more plant damage than that induced by the individual pollutants.
Researchers have tried to develop terminology that is meaningful in evaluating
the effects of pollutant mixtures on plants (Reinert, 1975; Ormrod, 1982;
Ormrod et al., 1984). Two categories of plant response are possible when the
effects of two pollutants (A and B) are evaluated. When one pollutant has no
effect on plant response but the second one does, it is termed "no joint
action." Thus, the term "joint action" implies that both pollutants have some
effect on plant response. The concept of joint action can be further divided
into subcategories that can be used to describe the response of plants to
pollutants, A and B:
1. Additive response: Effect.., = Effect A + Effect B
2. Interactive response: Effect.R 1 Effect. + EffectR
The interactive response may be divided further into two types:
1. Synergism: Effect.n > Effect. + EffectR
2. Antagonism: Effect.R < Effect. + EffectR
Some studies use the term "greater (more) than additive" to mean synergism and
the term "less than additive" to mean antagonism.
It is the purpose of this section to discuss the effects of the joint
action of S02 plus 03, NO,, plus 0,, and N02 plus S02 plus 0-; and to identify
the concentrations of (k, alone or in combination with other pollutants, that
cause yield loss.
6.3.2.3.1.1 Ozone and sulfur dioxide. The joint action of 0- and SC^
has been extensively studied. The previous criteria document (U.S. Environ-
mental Protection Agency, 1978) stated that mixtures of 0- plus SO^ were of
special interest because of the Menser and Heggestad (1966) study. In that
study, a sensitive 'Bel W-1 cigar-wrapper tobacco exposed to mixtures of 0-
(0.03 ppm) and S02 (0.25 ppm) for 2 or 4 hr sustained 23 percent and 48 percent
foliar injury, respectively; but no visible injury was produced by the same
concentrations of the individual pollutants. The additive and frequently
6-52
-------
synergistic foliar-injury response of tobacco has been reported to occur in
numerous tobacco cultivars and types. Menser and Hodges (1970), Grosso et al.
(1971), and Hodges et al. (1971) determined the response of several Nicotiana
species and various N. tabacum cultivars to S0? and 0- mixtures. They found
that DO and SOp acted synergistically and produced 0~-type symptoms on all
cultivars of burley and Havana tobacco. When plants were fumigated for 4 hr
with 0.03 ppm 0- alone or with 0.45 ppm S0? alone, no injury was observed.
When the gases were combined and the plants were exposed for the same length
of time, foliar injury ranging from 5 percent to 15 percent was produced.
Tingey et al. (1973b) exposed 11 species of plants to different combinations
of 03 and S02: either 0.05 or 0.1 ppm 03 and 0.1, 0.25, or 0.5 ppm S02 for 4
hr. They observed additive and synergistic foliar-injury responses on five of
the six species in Table 6-5 but not at all exposure combinations.
TABLE 6-5. SUMMARY OF EFFECTS OF SULFUR DIOXIDE
AND OZONE MIXTURES ON FOLIAR INJURY
Plant species
Alfalfa
Broccoli
Cabbage
Radish
Tomato
Tobacco, Bel W3
Response at stated ppm
0.50/0.05 0.50/0.10
+
+ 0
0 +
0 +
0 0
+ +
S02/03 concentrations
0.10/0.10 0.25/0.10
+ +
+ 0
0 0
+ +
0
0 +
a+ = greater than additive; 0 = additive; - = less than additive.
Source: Tingey et al. (1973b).
Foliar injury symptoms decrease the aesthetic value of various types of
woody ornamental and floricultural crop species (Section 6.4.3). Also, when
foliage is the marketable plant part, substantial losses in quality and market-
ability of the crop result from the injury produced by the joint action of
pollutants. The amount of foliar injury affects the amount of photosynthate
produced by the plant. Thus, in many instances, foliar injury provides some
indication of the potential for loss in weight, size, and number (yield) of
the marketable plant part. Foliar-injury response from the joint action of
pollutants needs continued study.
6-53
-------
Since 1978, researchers have continued to use foliar injury as an indica-
tor of the sensitivity of plant species and cultivars within a species to the
joint action of 03 and S(L. Studies have included apple (Shertz et al.,
1980a), grape (Shertz et al. 1980b), radish, cucumber, and soybean (Beckerson
and Hofstra, 1979), begonia (Reinert and Nelson, 1980), and pea (Olszyk and
Tibbitts, 1981). These results are summarized in Table 6-6. Although rela-
tively high 0- and SCL concentrations were used for only a few hours, most
species displayed a synergistic injury response from the joint effects of the
pollutants, supporting previous observations.
The chronic effects of the joint action of 0- and S02 on the growth of
radish, alfalfa, soybean, and tobacco (Table 6-7) were summarized in 1978
(U.S. Environmental Protection Agency, 1978). These four species represent a
diverse group of plant species in terms of growth habit. Primary focus in
earlier studies was on weight changes during the vegetative stage of growth,
with the exception of one study (Heagle et al., 1974); however, radish root
(hypocotyl), tobacco leaf weight, and alfalfa foliage (top) weight are the
marketable portions of the plant. With the exception of alfalfa, the growth
of each plant species was reduced in an additive manner by the joint action of
the two pollutants. Soybean root (fresh weight) responded synergistically to
the joint action of (L and SCL in one study (Tingey et al., 1973c).
The above data were obtained in greenhouse studies (except for Heagle et
al., 1974). These data provided preliminary evidence that the joint action of
0- and SOp at concentrations of 0.05 ppm and greater caused an additive reduc-
tion in plant yield. Additional studies of the joint action of 03 and SO,, on
plant yield have been conducted since 1978 (Tables 6-8 and 6-9). More emphasis
has been given to the influence of pollutant combinations on yield (weight,
size, and numbers) as a measure of plant response, including the yield of
flower, fruit, and seed portions of the plant (Table 6-8). Shew et al. (1982)
exposed tomato to 0.2 ppm 0, and S09 alone and together, two times per week, 2
-------
TABLE 6-6. FOLIAR INJURY RESPONSE OF VARIOUS PLANT SPECIES TO OZONE AND OZONE PLUS SULFUR DIOXIDE
Concentration ,
ppm Exposure
Species
03 S02 duration
Response
Foliar i
Oi SO?
Apple
(Vance Del i-
cious)
(Imperial
Mclntosh)
(Golden
Delicious)
Grape
(Ives)
(Delaware)
Radish
i
en
en
Cucumber
Soybean
Begonia
(Schwaben-
land Red)
(Wisper '0'
Pink)
(Fantasy)
(Renaissance)
(Turo)
Pea
Where column entry
Concentrations of
0.40 0.40 03-4 hr/day,
1 time
S02-4 hr/day,
1 time
0.40 0.40 03-4 hr/day,
1 time
S02-4 hr/day
0.15 0.15 03-6 hr/day.
5 days
S02-4 hr/day,
5 days
0.25 0.50 03-4 hr/day
every 6 days,
4 times
S02-4 hr/day
every 6 days
0.13 0.40 03-4 hr,
I time
S02-4 fir,
1 time
is blank, information is the same
each gas were the same when given
Foliar
injury
Fol iar
injury
Foliar
injury
Fol iar
injury
Foliar
as above.
together as
"The "interaction effect" is the effect from the combination of
24
30
27
27
1
13
27
18
54
25
2
15
8
0
when
03 and
8
9
19
18
1
1
9
0
2
1
0
0
0
0
given
S02
njury, %
S02 + 03
26
22
19
47
4
30
54
0
67
58
13
18
12
32
singly.
minus the
Interaction Monitoring
effect method
-6
-17
-27
2
2
16
18
-18
11
32
11
3
4
32
individual
03-Mast
meter
S02-Not
given
03-Mast meter
S02-Not
given
03 -UV
Dasibi
S02-Conduc-
tivity
03-Chemi lumi-
nescence
S02-Flame
photometry
03-Chemi lumi-
nescence
CO. -Thaymo-
-"Z •'"-'•""
electron
(S02)
erfects of 03 and Ju2
Calibration
method
KI
Permeation
tubes
KI
Permeation
tubes
Not given
Not given
Monitor
Labs
Calibrator
KI
Gas-phase
titration
(See Section
Fumigation
facility
Control led
environment
chambers
Control led
environment
chambers
Exposure
chambers
in environ-
mental ly
control led
room
CSTR in
greenhouse
Plexiglas
chamber
G. 3. 2. 3 . 1) .
Reference
Shertz et al.
(1980a)
Shertz et al .
(1980b)
Beckerson and
Hofstra
(1979)
Re inert and
Nelson
(1980)
Olszyk and
Tibbits
(19811
-------
TABLE 6-7. GROWTH RESPONSE OF VARIOUS PLANT SPECIES TO OZONE AND OZONE PLUS SULFUR DIOXIDE
en
Species
Radish
(Cherry
Belle)
Alfalfa
(Vernal)
Soybean
(Dare)
Soybean
(Dare)
Tobacco
(Bel-W3)
Concentration ,
ppm Exposure
03 S02 duration
0.05 0.05 8 hr/day,
5 days/wk,
5 wks
0.05 0.05 8 hr/day,
5 days/wk
12 wk
0.05 0.05 7 hr/day,
5 days/wk
3 wk
0.10 0.10 7 hr/day,
5 days/wk,
until harvest
0.05 0.05 7 hr/day,
5 days/wk,
4 wk
Response
Top dry wt
Root dry wt
Top dry wt
Root dry wt
Top fresh wt
Root fresh wt
Top fresh wt
Seed wt
Leaf dry wt
Yield, % reduction
from control
(negative unless
otherwise noted)
Oj> S02 S02 + 03
It) 0 10
50
12
22
2
3
65
54
1
17
26
29
+5
0
+3
4
14
55
18
24
12
24
52
63
30
Interaction Monitoring
effect method
0
-12
-20
-27
15
21
-10
5
15
03-Mast
meter
S02-Conduc-
tivity
03-Mast
meter
S02-Conduc-
tivity
03-Mast
meter
S02~Conduc-
tivity
03-Mast
meter
S02- Flame
photometry
03-Mast
meter
S02-Conduc-
tivity
Calibration
method
KI
Colori-
metric
KI
Colori-
metric
KI
Colori-
metric
KI
Not given
KI
Colori-
metric
Fumigation
facility
Chambers
in green-
house
Chambers
in green-
house
Chambers
in green-
house
Field
chambers
Chambers
in green-
house
Reference
Tingey
et al.
(1971a)
Tingey and
Reinert
(1975)
Tingey
et al.
(1973c)
Heagle
et al.
(1974)
Tingey and
Reinert
(1975)
Concentrations of each gas were the same when given together as when given singly.
The "interaction effect" is the effect from the combination of 03 and S02 minus the individual effects of 03 and S02 (see Section 6.3.2.3.1).
-------
TABLE 6-8. YIELD RESPONSES OF VARIOUS PLANT SPECIES TO OZONE AND OZONE PLUS SULFUR DIOXIDE
Species
Tomato
(Walter)
Begonia
(Schwaben-
land Red)
(Wisper '0'
Pink)
(Fantasy)
en
ui(Renais-
"-J sance)
(Turo)
Snap bean
(BBL 290)
(BBL 274)
(Astro)
Tall fescue
(Alta)
Alfalfa
(Mesa-
Si rsa)
Concentration3
ppm
03 S02
0.20
0.25
0.25
0.25
0.25
^0.25
0.065b
0.10
0.20
0.30
0.10
0.20
0.30
0.05
0.20
0.50
0.50
0.50
0.50
0.50
0.30
0.10
0.10
0.10
0.10
0.10
0.10
0.05
Exposure
duration
03-4 hr/day,
2 day/wk, 8 wk
S02-4 hr/day,
2 day/wk, 8 wk
03-4 hr/day,
every 6 days
for 4 times,
S02-4 hr/day
every 6 days
4 for times
03-11 hr/day
avg, 3 mo
S02-6 hr/day,
5 day/wk, 5 wk
03 and S02
6 hr/day,
once a week
for 12 weeks
03-6 hr/day,
S02-24 hr/day,
68 days
Response
Largest
fruit each
cluster
Total fruit
Flower wt
Flower wt
Flower wt
Flower wt
Flower wt
Green
pod wt
No. of
tillers
Top dry wt
Foliage dry
68 days
Yield, % reduction
from control
(negative unless
otherwise noted)
PJL
1
5
39
22
6
55
+10
2
+1
6
+5
+3
19
18
49
S02
2
4
22
+16
9
43
+11
16
6
6
6
5
5
5
S02 + 03
18
4
38
28
21
54
4
44
4
+12
19
18
19
53
46
Interr
action
effect
15
-5
-23
22
6
-44
25
26
-1
0
18
16
-5
30
-3
Monitoring
method
Oa-Chemi lumi-
nescence
S02- Flame
photometry
03-Chemi lumi-
nescence
S02- Flame
photometry
03-Not given
S02-Pulse
fluorescence
03-UV
S02-Pulse
fluorescence
03 Mast
meter
Cal i brat ion
method
Known source
Permeation
tube
Known source
Permeation
tube
Not given
Permeation
tube
UV photometry
greenhouse
Permeation
tube
KI
Fumigation
facility
Chambers in
greenhouse
(CSTR)D
Chambers in
greenhouse
(CSTR)
Field chamber
(open top)
Chambers in
(1982a)
(CSTR)
Field
chamber
(closed
top
Reference
Shew et al.
(1982)
Reinert
and
Nelson
(1980)
Heggestad
and
Bennett
(1981)
Flagler and
Youngner
(1982a)
Neely
et al.
(1977)
Concentrations of each gas were the same when given together as when given singly.
CSTR = Continuous stirred tank reactor exposure chamber.
The "interaction effect" is the effect, from the combination of 03 and S02 minus the individual effects of (U and S02 (see Section 6.3.2.3.1).
-------
TABLE 6-9. INFLUENCE OF MIXTURES OF OZONE AND SULFUR DIOXIDE ON
SOYBEAN YIELD3
Seasonal 7 hr/day
03 concn., ppm Yield, % reduction from control
Seasonal 4 hr/day S02 concn. ,
0.00
0.055
0.068
0.085
0.106
0.00
0 (412)a
7.5
22.8
33.7
40.3
0.026
+6.3
22.8
24.0
42.2
39.3
0.085
+3.4
20.1
28.6
43.4
51.9
ppm
0.367
30.6
42.5
53.3
54.1
62.6
aMean yield (grams of seed) from eight 1-meter-row samples.
Source: Heagle et al., 1983c.
weight. The mean yield (flower weight) from the joint effects of 0, and SOp
ranged from 1 percent (Schwabenland Red) to 15 percent (Fantasy) greater than
the loss resulting from 0- alone.
The joint action of 0, and SOp on the growth and yield components of tall
fescue was studied by Flagler and Youngner (1982a). Fescue was exposed to 0~
concentrations of 0.0, 0.1, 0.2, and 0.3 ppm and 0.0 and 0.1 ppm SOp for 6
hr/day, once a wk for 12 wk. The joint action of SOp in the presence of
increasing concentrations of 0- caused additive decreases in fescue total dry
weight, root dry weight, and the root-to-shoot ratio. For example, 0, decreased
total dry weight 49 percent at 0.3 ppm 0~; but in the presence of 0.1 ppm SOp
there was an additional 11 percent loss in total dry weight. Ozone and SO,,
acted synergistically to decrease the number of tillers in fescue but the
synergism depended on the 0- concentration. These studies were done in a
charcoal-filtered-air greenhouse in CSTR exposure chambers.
Recently, studies of the combined action of 0- and S09 have been conduc-
O Ł
ted in open-top field chambers (Heagle et al., 1983c; Heggestad and Bennett,
1981) and large CSTR field chambers (Foster et al., 1983b; and Oshima, 1978).
In these experiments, 0- levels near ambient, as well as increasing 03 concen-
trations above ambient, were used in combination with two or more concentra-
tions of SOp. Heagle et al. (1983c) exposed soybean to various concentrations
of 0- for 7 hr daily and 4 concentrations of SOp for 4 hr/day. Both gases
were added for 111 days (Table 6-9). The high concentration of SOp decreased
6-58
-------
the amount of visible injury from increasing concentrations of 0.,. The joint
action of 0, and S02 on soybean seed weight per meter of row at lower concen-
trations appeared to be additive, but as the concentrations of both pollutants
increased there was an antagonistic (L-plus-SOp interaction. The nature of
the joint action was similar to that for visible injury: as SO,, increased to
0.367 ppm, the loss of seed weight from increasing 03 concentrations was less
than at lower concentrations of SO,,. For example, at 0.367 ppm SO,, and 0.085
ppm 0~ there was a 54.1 percent seed-weight loss compared to that at 0.367 ppm
S02 alone. At 0.026 ppm S02 and 0.085 ppm 0- there was a 42.2 percent seed-
weight loss, compared to that at 0.026 ppm SO,, alone (Table 6-8). The two
highest mean S02 concentrations were higher than usually occur in the United
States and even the concentration of 0.026 ppm S0« is higher than that found
in the ambient air at most locations (U.S. Environmental Protection Agency,
1983).
Using a field fumigation (tubular release) system, Reich and Amundson
(1984) exposed soybean to 0- and/or S02 in a 3 x 3 factorial design. The
plants were exposed to levels of 0- and SO,, above ambient for about 5 hr/day
for 16 days from mid-August to mid-September. There was no significant inter-
action between 0, and S02 on soybean yield.
Heggestad and Bennett (1981) exposed three cultivars of bean to increasing
concentrations of SO,, (0.06, 0.12, 0.3 ppm) for 6 hr/day in charcoal-filtered
and unfiltered ambient air, using open-top field chambers. The beans were
exposed daily 5 days/wk for 31 days. During the study period (July-August),
the average daily maximum 0- concentration during the S02 fumigation period
(9:00 a.m. to 3:00 p.m.) was 0.065 ± 0.025 ppm. Sulfur dioxide (0.30 ppm)
reduced snap bean yields (all cultivars) in nonfiltered air (0,) by 44 percent
compared to a 16 percent reduction in charcoal-filtered air. At 0.06 ppm S02,
the yield of cv. 'Astro1 was reduced more in nonfiltered than in filtered air.
The SOp concentrations used in this study, however, were higher than typically
occur in the United States (U.S. Environmental Protection Agency, 1983).
In southern California, Oshima (1978) and Foster et al. (1983b) conduc-
ted studies to determine the joint action of S02 and photochemical oxidants.
A range of photochemical oxidant concentrations was obtained by combining
various proportions of charcoal-filtered air and ambient air containing oxidants
to yield various concentrations of oxidants which were added to the CSTR-type
field exposure chambers. Sulfur dioxide (0.0 or 0.1 ppm) was added to the
chambers for 6-hr intervals approximately 47 times over a 76-day period for
6-59
-------
beans (Oshima, 1978) and 4 to 5 days/wk over a 10-wk period for potato (Foster
et al., 1983b). In the bean study (Oshima, 1978), the ozone concentration
exceeded 0.20 ppm frequently; the total ozone dose ranged from approximately
10.9 ppnrhr in the charcoal-filtered air chambers to approximately 83 ppm-hr
in the chambers receiving ambient ozone. In the potato study (Foster et al.,
1983b), the maximum hourly concentration was 0.27 ppm; for the remainder of
the study, the concentration never exceeded 0.20 ppm. The total ozone dose
ranged from 4.9 ppm-hr in the charcoal-filtered air chambers, to approximately
44 ppm-hr in the chambers receiving ambient ozone. The kidney bean yield was
less in the presence of ambient oxidant plus SO^ except at the high oxidant
concentrations, when the yields were more nearly similar. Similar studies
with potato exposed to SO,, and partially filtered ambient air containing 0.,
resulted in no evidence of joint action on tuber yield (Foster et al., 1983b).
In summary, recent studies on the effects of 0~ and S09 on the yield of
O Ł
various plant species have found the effects of 03 and S02 to be additive
(equal to the combined effects of the individual pollutants) for begonia
flower weight, fescue top and root dry weights, soybean seed weight, and snap
bean and green bean yield. Synergistic interaction was identified for the
effects of O.j and S0? on the largest tomato fruit in each cluster, the number
of fescue tillers, and kidney bean yield. Examples of antagonistic joint
action occurred in one cultivar of begonia and in soybean seed weight at the
highest SOp concentrations. These effects varied with the concentration of
pollutants, the plant response measured, species, and cultivar. Thus, observa-
tions were significant enough to propose the following general concepts:
1. When concentrations of 0^ and S02 are below or at the threshold for
visible injury, synergisric interaction may occur.
2. As concentrations of 0- and SOp increase in mixture above the injury
threshold, yield loss from joint action may be additive.
3. When both pollutants are present in high concentrations, the joint
action of 0~ and S0? may be antagonistic, such that further weight
loss is minimal.
4. In field studies, the addition of S0« generally did not influence
the Oo response unless the concentrations and exposure frequencies
were much greater than the S02 concentrations and frequencies of
occurrence that are typically found in the ambient air in the United
States.
6-60
-------
Relative to the last item above, an analysis of ambient air monitoring
data at various locations determined the frequency of the co-occurrence of
pollutant pairs (OVSOp, OVNCL) during a 5-month summer season (Ma.y through
September) (Lefohn and Tingey, 1984). Co-occurrence was defined as the simul-
taneous occurrence of hourly averaged concentrations of 0.05 ppm for both
pollutants of the pair. Most of the monitoring sites analyzed by Lefohn and
Tingey experienced 10 or fewer periods (hours) of co-occurrence during the
5-month summer season (May through September).
6.3.2.3.1.2 Ozone and nitrogen dioxide. Although the effects of NO,, and
0-, alone and in mixture, have not generally been studied, recent reports
comparing two- and three-pollutant mixture treatments include N0« plus 03
combinations. Kress and Skelly (1982) have studied the responses of seven tree
species to N02 (0.1 ppm) and 0- (0.1 ppm) alone and in mixture for 6 hr/day,
for 28 consecutive days (Table 6-10). Virginia and loblolly pine growth, as
measured by plant height, was significantly suppressed by the 0,-plus-NO,,
O w
treatment, but not by the individual pollutants. Nitrogen dioxide alone
significantly suppressed root dry weight of sweetgum; however, the joint
action of 03 plus N02 was antagonistic on sweetgum root dry weight and white
ash root dry weight.
6.3.2.3.1.3 Ozone plus nitrogen dioxide and sulfur dioxide. The previous
criteria document (U.S. Environmental Protection Agency, 1978) makes no refer-
ence to the effects of mixtures using three pollutants. Since then, however,
experiments have been designed to study the effect of increasing concentrations
of N07, S09, and 0- in mixture (Table 6-11). Reinert and Gray (1981) exposed
Ł- C- «3
radish plants one time for 3 or 6 hr to 0.2 or 0.4 ppm of NOp, S02, or 03, or
combinations. They found no interaction for either two- or three-gas mixtures,
even though the decrease in hypocotyl weight by 0- was further reduced by N02
alone, S0? alone, or NO,, plus SO,,, which suggests an additive response.
Reinert and Sanders (1982) and Sanders and Reinert (1982b) reported similar
results in radish following repeated exposures at different ages.
Marigold was exposed at different ages for 3 hr to 0.3 ppm of each pol-
lutant, three times/wk for 1 wk (Sanders and Reinert, 1982b). Ozone alone
decreased flower dry weight but the interaction of NO,, or 03 with SO,, was
apparently antagonistic. Similar results were reported for marigold exposed
repeatedly 3 days a week for 3 wk. Reinert and Heck (1982) exposed snap beans
6-61
-------
TABLE 6-10. YIELD RESPONSES OF SELECTED TREE SPECIES TO OZONE PLUS NITROGEN DIOXIDE
Concentration ,
ppm
Species 03 N02
Loblolly pine 0.10 0.10
Loblolly pine 0.10 0.10
(6-13 x 2-8)
Pitch pine 0.10 0.10
Virginia pine 0.10 0.10
Sweetgum 0.10 0.10
White ash 0.10 0.10
Green ash 0.10 0.10
Willow oak 0.10 0.10
Exposure
duration
6 hr/day,
28 days
6 hr/day,
28 days
6 hr/day,
28 days
6 hr/day,
28 days
6 hr/day,
28 days
6 hr/day,
28 days
6 hr/day,
28 days
6 hr/day,
28 days
Height and top dry wt,
X reduction from control
(negative unless
Response otherwise noted)
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
Height growth
Top dry wt
Root dry wt
°a
17
21
13
25
11
31
14
+14
0
11
2
19
27
30
45
20
37
55
19
17
12
+5
+1
+11
N02
15
22
17
11
10
14
16
20
11
13
1
7
32
25
27
+5
1
37
+1
10
18
10
24
14
03 + N02
39
26
26
24
4
17
26
11
15
23
1
19
28
21
48
16
37
52
22
29
19
14
13
12
Interaction
effects0
7
-17
-4
-24
-17
-28
-4
5
4
-1
-2
-19
-31
-34
-24
1
-1
-40
4
2
-11
9
-10
9
Plants were exposed in continuously stirred tank reactor (CSTR) exposure chambers in a greenhouse.
Ozone and N02 were monitored using chemiluminescent analyzers which were calibrated with known sources
of each pollutant.
Concentrations of the combination were the same as the single gases.
clndicates seeds were from a full-sibling collection.
The "interaction effect" is the effect from the combination of 03 and N02 minus the individual effects of
03 and N02 (see Section 6.3.2.3.1).
Source: Kress and Skelly, 1982.
6-62
-------
TABLE 6-11. YIELD CHANGE IN VARIOUS PLANT SPECIES EXPOSED TO OZONE, SULFUR DIOXIDE, AND NITROGEN DIOXIDE
Concentration ,
ppm Exposure Yield, % reduction
Species 03 S02 N02 duration Response from control (negative unless otherwise noted)
03 S02 N02 S02+N02 0;i+S02 03+NO-> 03+S02+N02
Snap Bean 0.15 0.15 0.15 4 hr, Green bean 27 9 +12 20 6 25 27
3 times/wk fresh wt
4 wks
Marigold 0.30 0.30 0.30 3 hr/day, Flower wt 20 47 +16 13 23 +4 20
3 days/wk,
3 wks
Marigold 0.30 0.30 0.30 3 hr/day, Flower wt 41 49 23 47 25 39 20
f 3 days/wk,
cr> 1 wk
CO
Radish 0.30 0.30 0.30 3 hr/day, Hypocotyl 30 +21 +10 +16 43 33 65
3 days/wk,
1 wk
Radish 0.40 0.40 0.40 3 hr + 6 hr Hypocotyl 20 4 .0 13 24 23 36
1 time
Azalea 0.25 0.25 0.25 3 hr/day, Foliage 6 7 0 17 22 16 27
6 times in a
4-wk period
Monitoring
method
03, N02-
chemi lumi-
nescence;
S02-flame
photometry
03, N02-
chemi lumi-
nescence;
S02- flame
photometry
03, N02-
chemi lumi-
nescence
S02 Flame
photometry
03, N02-
chemi lumi-
nescence
S02- flame
photometry
Oa, N02-
chemi lumi-
nescence
S02- flame
photometry
03, N02-
chemi lumi-
nescence ;
S02-flame
photometry
Calibration
method
Known
source
Known
source
Known
source
Known
source
Known
source
Known
source
Fumigation
facility0
Chambers
in green-
house
(CSTR)
Chambers
in green-
house
(CSTR)
Chambers
in green-
house
(CSTR)
Chambers
in green-
house
(CSTR)
Chambers
in green-
house
(CSTR)
Chambers
in green-
house
(CSTR)
Reference
Reinert and
Heck
(1982)c
Reinert and
Sanders
(1982)
Sanders and
Reinert
(1982b)
Sanders and
Reinert
(1982b)
Reinert and
Gray
(1981)
Sanders and
Reinert
(1982a)
-------
TABLE 6-11 (cont'd). YIELD CHANGE IN VARIOUS PLANT SPECIES EXPOSED TO OZONE, SULFUR DIOXIDE, AND NITROGEN DIOXIDE
Concentration ,
ppm
Species
Kentucky
bluegrass
(12 culti-
vars)
Red top
grass
-------
27 times intermittently for 3 hours each time over 6.5 wk to increasing concen-
trations of S02 (0.0, 0.1, 0.15 ppm) and N02 (0.0, 0.05, 0.1 ppm) in the
presence of 0.05 ppm 0-. Ozone alone decreased bean pod weight 10 percent,
while N02 (0.1 ppm), S02 (0.15 ppm), and 03 (0.05 ppm) together decreased pod
weight by 31 percent. Reinert and Heck (1982) exposed 16-dayold radish
plants one time for 3 hr to three concentrations (0.0, 0.2, and 0.4 ppm) or
(0.1, 0.2, and 0.4 ppm) of N0«, S09, and 0. at all 27 (3 x 3 x 3) treatment
L— L- O
combinations (Table 6-12). In both experiments, the reduction in size of
radish hypocotyls was predominantly additive and linear within the range of
concentrations used. The above studies were conducted primarily under green-
house conditions but some of the species studied, such as marigold, tomato,
and azalea, are grown commercially in greenhouses. The concentrations of S0?
and NO/, (s= 0.4 ppm) are below the concentration of each pollutant individually
(S02, 0.5 ppm, and NO,,, 1 to 2 ppm) that causes visible injury for a single
exposure (Tingey et al., 1971b).
Several turf grass species and cultivars were exposed to 0-,, S02, and N02
individually, and to the three pollutants combined to determine the effects on
leaf area (Elkiey and Ormrod, 1980). The three-pollutant combination reduced
the leaf area of only 4 of the 12 Kentucky bluegrass cultivars. The three-
pollutant combination had no significant effect on red top, creeping bentgrass,
and colonial bentgrass, but it did significantly reduce the leaf area of
perennial ryegrass and one of the two red fescue cultivars.
The results from the pollutant interaction studies cited in this section
demonstrated that the joint action of 0., with S02 or NOp or both decreased the
yield of several crop species more than 0- alone. Sulfur dioxide usually
modified the response to 0- in an additive way. Yield losses resulting from
03 exposure were further decreased by S02 in radish (5 percent), alfalfa (6
percent), soybean seed weight (9 percent), and tobacco (7 percent). These
effects were at concentrations of 03 and SO^ < 0.05 ppm and greater. At
higher concentrations of 0, and S02 (0.2 to 0.5 ppm), yield losses from 03
exposure were further reduced by S0? in begonia flower weight 6 to 15 percent
depending on the cultivar; in kidney bean 11 to 28 percent, depending on the
0- concentration; in potato, 11 to 16 percent; in soybean seed weight, 11 to
12 percent; and in fescue, < 24 percent. Additional information concerning
pollutant dose and frequency of exposure at which these effects take place is
needed.
6-65
-------
TABLE 6-12. EFFECTS OF NITROGEN DIOXIDE IN COMBINATION WITH SULFUR DIOXIDE
OR OZONE, OR BOTH, ON RADISH ROOT FRESH WEIGHT AT
THREE CONCENTRATIONS OF EACH GASa
(grams)
S02, ppm
Experiment 1
0.1
0.2
0.4
Experiment 2
0
0.2
0.4
03 ppm
>
0.1
0.2
0.4
0.1
0.2
0.4
0.1
0.2
0.4
0
0.2
0.4
0
0.2
0.4
0
0.2
0.4
0.1
9.5
7.3
4.6
9.5
6.3
2.9
8.3
5.6
2.3
0.0
15.2
12.4
6.6
16.7
11.2
6.8
17.2
9.5
5.1
N02 , ppm
0.2
8.8
7.7
3.0
9.5
5.3
3.3
6.6
5.0
3.0
0.2
16.9
11.0
5.3
17.2
7.3
5.3
13.2
7.2
5.6
0.4
8.4
4.6
2.9
6.2
5.1
2.7
4.9
3.9
3.0
0.4
14.4
9.6
8.0
11.9
7.6
4.8
11.4
5.8
4.3
Means represent 20 (exp. 1) or 12 (exp. 2) plants. Plants were exposed once
for 3 hr at 16 days from seed and were harvested at 23 days from seed.
Source: Reinert and Heck (1982).
6-66
-------
The initial studies on the effects of mixtures of NO^, SCL, and 03 have
involved the co-occurrence of these pollutants. The sequential effects of
pollutant mixtures need to be investigated. In addition, more monitoring data
are needed for each of the three pollutants so that realistic occurrences and
concentrations can be part of the experimental design for assessing plant
response.
6.3.2.3.1.4 Ozone and other pollutants. The effects of (L in combination
with heavy metals have been studied in several plant species. Zinc and cadmium
reacted synergistically with 0, (0.30 ppm for 6 hours) in producing visible
injury and chlorophyll loss in garden cress and lettuce (Czuba and Ormrod,
1974). The combination of cadmium (Cd) and 0, induced earlier development of
necrosis and chlorosis and the injury was observed at lower 0~ plus Cd levels
than for the individual treatments (Czuba and Ormrod, 1981). Cadmium and
nickel (Ni) concentrations of 1, 10, and 100 umol in the nutrient solution
interacted to reduce root and shoot growth of peas (Ormrod, 1977). Ozone
exposure increased the Cd and Ni effects but the increase was less than additive.
Low concentrations of Cd and Ni, however, tended to enhance 03 phytotoxicity.
The interaction of Cd and 0, was influenced by both concentration and the
environmental conditions. Tomato plants grown at 0.25 and 0.75 mg Cd/ml
developed only slight foliar injury when exposed to 0., (0.20 ppm for 3 hours)
under cloudy skies; whereas the Cd treatment alone had no significant effect
(Markov et al., 1979). In full sun there was extensive 0, injury and the
joint response was synergistic. In pea leaves, alterations in cellular
ultrastructure increased following exposure to ozone (0.50 ppm) when plants
were grown in nutrient solutions containing 100 umol nickel sulfate (Mitchell
et al., 1979).
Quaking aspens treated with 10 ug Cd/ml for 30 days displayed significant-
ly more foliar injury when exposed to ambient air in New Jersey (during the 30
days of Cd treatment) or exposed to 0.20 ppm 0- for 2.5 hours (Clarke and
Brennan, 1980). When plants were exposed to 0.30 ppm 03, the Cd enhancement
of injury was not apparent.
The limited published data indicate that heavy metals can increase the
phytotoxic reactions of ozone. At the present time, it is not possible to
assess the risk from the joint action of gaseous and heavy metal pollutants to
vegetation. In industrial areas, along heavily travelled highways, and on
crop lands fertilized with sludge, however, there is the possibility for in-
teractive effects.
6-67
-------
6.3.2.3.2 Chemical Sprays. A variety of agricultural chemicals commonly used
by growers to control diseases and insects and other pests on crops and re-
search plantings can modify vegetational response to air pollutants (Reinert
and Spurr, 1972; Sung and Moore, 1979). Certain fungicides, insecticides,
nematocides, and herbicides have been found to change the sensitivity of
plants to ozone.
Protection from or reduction of 03 injury to vegetation is significant to
growers of economically important crops in areas of high ozone concentrations.
In addition, the control of (k injury to plants in the field can be of assis-
stance to scientists attempting to determine how 0, injures plants. The
report by Kendrick et al. (1954) that fungicides used as sprays or dusts
protected pinto bean foliage from oxidant-induced plant damage alerted the
scientific community to the fact that agricultural chemicals could protect
vegetation from Oo injury. Since that time, it has been shown that other
chemicals, including ascorbic acid sprays (Freebairn, 1963; Freebairn and
Taylor, 1960), antiozonants (Rich and Taylor, 1960), anti-transpirants (Gale
and Hagan; 1972, 1966), stomatal regulators (Rich, 1964), growth regulators
(Cathey and Heggestad, 1973), and some herbicides can offer some protection
against ozone injury.
A comprehensive review of plant protectant sprays and their uses is found
in Ozone and Other Photochemical Oxidants (National Research Council, 1977).
The degree of plant protection obtained from 0, injury and the species tested
are listed in Table 6-13.
Nematocides increase the sensitivity of vegetation to 0,, but nematicides
in combination with certain fungicides decrease sensitivity to On. Miller et
al. (1976) noted that pinto bean and tobacco growing in sand or soil treated
with the contact nematocides, phenamiphos, fensulfothion, aldicarb, and oxa-
fothion were more sensitive to ambient O,. Adding benomyl or carboxin, both
fungicides, to the soil containing the contact nematocides caused the plants
to become highly resistant to Oo injury. Benomyl or carboxin used alone also
induced plant resistance to 03 injury.
The influence of selected herbicides on the 0^ sensitivity of tobacco and
other crop plants has been studied with differing results. Carney et al.
(1973) demonstrated that pebulate increased 0, injury to tobacco but that
benefin decreased On injury. The studies of Sung and Moore (1979), however,
failed to confirm the observation that pebulate increased 0, sensitivity.
Sung and Moore suggested that the difference in results occurred either because
6-68
-------
TABLE 6-13. PROTECTION OF PLANTS FROM OXIDANT INJURY BY APPLICATION OF PROTECTIVE CHEMICALS
Plant species
Pollutant
protected from:
Chemical (Concentration)'
Type of protectant
Degree of .
protection, %
Bean, cultivar Pinto
Petunia
Tobacco
Tobacco, cultivar White Gold
Tobacco, cultivar White Gold
Tobacco, cultivar White Gold
Bean, cultivar Pinto
Bean, cultivar Pinto
Azalea
Bean, cultivar Pinto
Radish
Poinsettia
Poinsettia
Bean, cultivar Pinto
Bean, cultivar Pinto
c^ Bean and cucumber
en Grape
10 Bean, cultivars Tempo and Pinto
Bean, cultivars Tempo and Pinto
Oxidant
Oxidant
Oxidant
Oxidant
Oxidant
Oxidant
Oxidant
Ozone
Oxidant
Ozone
Ozone
Ozone
Ozone (chronic)
Ozone
Ozone
Ozone
Ozone
Ozone
Oxidant
K-Ascorbate (0.01 M)
K-Ascorbate (0.01 M)
Zn-ethylenebisdithiocarbamate
dust (variable)
Phygon XL (variable)
Phygon XL (variable)
4,4-Dioctyldiphenylamine in butyl
latex
Zineb (normal use)
Zineb (normal use)
Benomyl (60-ppm drench)
Carboxin (2.3 ppm in soil)
N-6-Benzyladenine (30-ppm spray)
Ancymidol (100-ppm spray)
Benomyl (500-ppm drench)
Folicote (0.5% spray)
Benomyl (5 ppm in nutrient
solution)
Benomyl (80 ppm in soil)
Benomyl (6.7 kg/ha, 6 times)
Benomyl (0.25 to 0.36%, 4 weekly
sprays)
Carboxin (10% granular as soil
Antioxidant 52
Antioxidant 39
Fungicide 44
Antioxidant . 89
Antioxidant 78
Antioxidant 100
Fungicide 91
Fungicide 97
Fungicide 96
Fungicide 95
Growth substance 100
Growth retardant 100
Fungicide 57
Wax emulsion 92
Fungicide 97
Fungicide 94
Fungicide 53
Fungicide 75
Fungicide 100
Tobacco
Tobacco
Bean, cultivar
Grass, annual
Bean, cultivar
Bean, cultivar
Bean, cultivar
Petunia
Tobacco
Tobacco
Tobacco
Tempo
blue
Pinto
Pinto
White
Ozone (0.
Ozone (0.
Oxidant
Ozone (0.
Ozone (0.
Ozone (0.
Ozone (0.
ppm, 0.
Oxidant
Oxidant
Oxidant
Ozone
50
35
25
30
25
13
5
ppm,
ppm,
ppm,
ppm,
ppm,
to 0.
hr)
2
2
2
4
4
,50
hr)
hr)
hr)
hr)
hr)
cuiitri luinci i \f , (_> y/ -* in i un j
Piperonylbutoxide (2 mM solution)
Safroxane
Benomyl (0.24% spray)
Benomyl (60-ppm amendment)
Triarimol
Benomyl (1.60-ug/g soil amend-
ment)
Ascorbic acid
SADH (0.5% spray)
Benomyl (25-ppm drench)
Benomyl (0.18% spray)
Peroxidase (0.10 ppm injected)
Insecticide
Insecticide
Fungicide
Fungicide
Fungicide
Fungicide
Antioxidant
Growth retardant
Fungicide
Fungicide
Enzyme
99
76 c
32 to 41C
85
81
98
75
82
68
59
89
These are applied as sprays unless otherwise noted.
Percent reduction in plant injury from ozone as a result of the protectant treatment.
Increase in yield by protectant application.
Source: Modified from National Research Council (1977).
-------
the plants used were of different ages or because the (L concentrations used
in the respective experiments differed. Reilly and Moore (1982), however,
stated that pebulate had no consistent effect upon tobacco sensitivity to 0~.
Benomyl, specifically, and fungicides in general were discussed exten-
sively as plant protectants in the National Research Council report (1977)
because they have been the most widely studied protectants. Benomyl (methyl-1-
butylcarbamoyl-2-benzimidazolecarbamate) has been used as a foliar spray, soil
drench, and a soil amendment (National Research Council, 1977) and was found
to reduce 0- injury in a wide range of plant species (Table 6-13). Benomyl,
while usually offering protection against CL injury, does not prevent PAN
injury (Pell, 1976; Pell .and Gardner, 1975; Pell and Gardner, 1979).
Antioxidants, chemical compounds that prevent food spoilage and discolo-
ration and prevent rubber from reacting with CL, have also been found to
reduce CL injury in vegetation (Kendrick et al., 1962). In agricultural prac-
tice, antioxidants are used as synergists with insecticides, herbicides, and
fungicides to increase their effectiveness. For example, antioxidants increase
the potency of a certain insecticide by decreasing the rate at which insects
are able to detoxify it.
Piperonyl butoxide (cr[2-(2-butoxyethoxy)ethoxy]-4,5-methylenedioxy-2-
propyltoluene), a synergist used with pyrethrum insecticides, is highly effec-
tive in protecting tobacco leaves from 0- injury (Koiwai et al., 1974; Koiwai
and Kisaki, 1976; Koiwai et al., 1977). Koiwai et al. (1977) determined that
most compounds having a synergistic activity with pyrethrum insecticides are,
in general, effective in preventing ozone injury to tobacco leaves. Rubin et
al. (1980) tested the protective capability of piperonyl butoxide when applied
to navy bean cultivars '0686' and '0670' and found that both cultivars were
protected by piperonyl butoxide, but only if it was used as a spray, not as a
soil treatment. Piperonyl butoxide was slightly phytotoxic, but the symptoms
resulting from the spray were not similar to those characteristic of ozone
injury. Santoflex 13, (N-(l,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine),
an antioxidant, is used to protect rubber from ozone attack. Gilbert et al.
(1977) found that bean, muskmelon cultivar 'Delicious 51,' and tobacco cultivar
'Bel W-31 were protected by Santoflex dust against visible injury when they
were exposed to concentrations of 0, up to 0.35 ppm in chamber studies.
Ethylenediurea (EDU) [N-(2-(2-oxo-l-imidazolidinyl)ethyl)-N-phenylurea],
an -antioxidant, has been widely used to reduce 0, injury to vegetation. Pinto
beans sprayed to run-off with 500 ug/ml EDU usually survived exposure to 0., at
6-70
-------
concentrations of 0.8 ppm for 150 min without visible injury (Carnahan et al.,
1978). Untreated plants exposed under the same conditions developed ozone
injury symptoms over the entire surface area of the primary leaves.
Hofstra et al. (1978) found EDU to be more effective than benomyl or car-
boxin in suppressing 03 injury on the highly sensitive navy.bean growing in
the field. It reduced bronzing, delayed leaf drop, and increased the yield up
to 36 percent in plants exposed to hourly mean concentrations of 0, at 0.1 to
0.3 ppm.
Pinto bean plants grown in pots received the greatest protection from 0.,
injury when treated with EDU 3 to 7 days before a 6-hr exposure to 0- concen-
trations of 0.10 to 0.76 ppm (Weidensaul, 1980). Plants received the most
effective protection by EDU when 03 concentrations were 0.41 ppm or higher.
Foliage that had not yet been formed at the time the chemical was applied was
not protected. The most extensive testing of the protective capabilities of
EDU has been done by Cathey and Heggestad (1982a,b,c), who studied the effects
of EDU (as either a foliar spray or soil drench) on the 0, sensitivity of
petunia (5 cultivars), chrysanthemum, and 44 other herbaceous species. In all
cases they found that treatment with EDU reduced the 03 injury. In addition
to herbaceous species, EDU also reduced 0, injury in woody vegetation
(McClenahan, 1979; Cathey and Heggestad, 1982c).
Farmers and others growing crops in areas where high 0, concentrations
exist should be aware, as studies cited above indicate, that agricultural
chemicals commonly used to protect plants from a variety of fungi, insects,
and nematodes can modify the response of the vegetation to 03 exposure.
Antioxidants used in insecticides and herbicides to increase their effective-
ness can also change the way plants respond to 0, exposure. In general,
nematocides seem to increase 0, sensitivity, while fungicides and antioxidants
have a protective effect when sprayed or drenched onto crops. Studies with
herbicides have shown no general trend. Because no two of the chemical com-
pounds that have been studied appear to function in the same way, it is not
possible to generalize. At the present time, the protectants do not appear to
be cost-effective to the extent that they can be generally prescribed for
protecting plants from 0^ injury, but they may provide protection from ozone
injury in addition to their primary function.
6-71
-------
6.4 OZONE EXPOSURE AND RESPONSE
Plant responses to 0- may be manifested as biochemical, physiological,
visible injury, growth, yield, reproduction, and ecosystem effects. Biochemi-
cal and physiological alterations are the fundamental cause of all other
effects, and were briefly described in Section 6.3. Visible foliar symptoms
are frequently the first indication of the effects of air pollution on vege-
tation, but they may be difficult to distinguish from other stress effects.
Although functional leaves are required for plant growth and yield, the loss
of leaf area is not always well correlated with yield reductions. This lack
of correlation may occur, if, for example, the plant has more leaf area than
required to maintain the yield, or if plant or environmental factors other
than leaf area limit yield. This concept is supported by the observation that
plant yield is not well correlated with leaf photosynthetic rate. The lack of
correlation between visible injury and yield is most common when the plant
foliage is not the usable or marketable portion of the plant (yield). In this
section, yield loss refers to the impairment of the intended use of the plant
as described in Section 6.2.5. Foliar injury on ornamental plants and leafy
vegetables; effects on native species; reductions in fruit, grain, foliage, or
root production by agricultural species? or adverse changes in plant quality
and aesthetic value can be considered yield loss for specific crops. Further-
more, reproductive capacities may be altered as a result of these responses;
and this alteration may lead to changes in populations and, eventually, eco-
system modification (see Chapter 7).
In the chapter on vegetational effects in the previous criteria document
(U.S. Environmental Protection Agency, 1978), emphasis was placed mainly on
visible injury and growth effects. Most of the growth effects discussed
concerned plant parts other than those of primary importance for yield. This
emphasis was dictated by the kind of data available at that time. The summary
figures and tables in the previous criteria document (U.S. Environmental
Protection Agency, 1978) emphasized foliar injury responses (see Figures 6-3,
6-4, and 6-5 and Table 6-14). The visible injury data were summarized by
presenting limiting values (Figures 6-3 and 6-4); i.e., those concentrations
below which visible injury was unlikely and presumably at which reduced growth
and yield would not occur. Another approach was to determine the 0, concen-
trations that, would produce a trace (5 percent) of foliar injury at various
time intervals (Figure 6-5; Table 6-14). The limiting values shown in Figures
6-3 and 6-4 were developed from a review of the literature available at that
6-72
-------
a
a
0.5
0.4
LU
2 0.2
O
O
LU
I 0.1
O
T 1—I I I I III
RANGE OF UNCERTAINTY FOR
SUSCEPTIBLE SPECIES
J 1—i i I i » n
T 1 I I I I
LIMITING VALUES FOR
TREES AND SHRUBS
800
400
J I I I I I
0.1 0.5 1 5
DURATION OF EXPOSURE, hours
Figure 6-3. Limiting values for foliar injury to trees and shrubs
by ozone.
Source: U.S. Environmental Protection Agency (1978).
O)
cc
z
111
(J
z
O
O
LU
z
O
N
O
800
I T—I
RANGE OF
UNCERTAINTY FOR
SUSCEPTIBLE
SPECIES
LIMITING VALUES FOR
AGRICULTURAL CROPS
I 1 ' ' ' ' I II
0.5 1
DURATION OF EXPOSURE, hours
Figure 6-4. Limiting values for foliar injury to agricultural crops
by ozone.
Source: U.S. Environmental Protection Agency (1978).
6-73
-------
1.0
0.9
z
2 0.8
I 0.7
§0.5
Z 0.4
O
g 0.3
0.2
0.1
TIME
0.5
1.0
2,0
4,0
8.0
CONCENTRATION, ppm
SENSITIVE
0.431 ± 0.044
0.218 ± 0.023
0.111 ± 0.020
0.058 ± 0.022
0.031 ± 0.023
INTERMEDIATE
0.637 ± 0.043
0.347 ± 0.020
0.202 ± 0.017
0.129 ± 0.019
0.093 ± 0.021
RESISTANT
0.772 ± 0.070
0.494 ± 0.028
0.355 ± 0.023
0.286 ± 0.026
0.251 ± 0.034
0 12 345678
TIME, hours
Figure 6-5. Ozone concentrations versus duration of exposure required to
produce 5 percent foliar injury in plants of three different sensitivity groupings.
The curves were generated by developing 95 percent confidence limits around
the equations for all plants in each susceptibility grouping from Table 6-14.
Curves: a = sensitive plants, b = intermediate plants, c - resistant plants.
Source: U.S. Environmental Protection Agency (1978).
6-74
-------
TABLE 6-14. CONCENTRATION, TIME, AND RESPONSE EQUATIONS FOR THREE SUSCEPTIBILITY GROUPS AND FOR
SELECTED PLANTS OR PLANT TYPES WITH RESPECT TO OZONE
Plants
Sensitive:
All plants
Grasses
Legumes
Tomato
Oat
Bean
Tobacco
Intermediate:
All plants
Vegetables
Grasses
Legumes
-r, Perennial
i Clover
S Wheat
Tobacco
Resistant:
All plants
Legumes
Grasses
Vegetables
Woody plants
Cucumber
Chrysanthemum
(C
-0.0152
-0.0565
0.0452
-0.0823
-0.0427
-0.0090
0.0245
0.0244
-0.0079
0.0107
0.0116
0.0748
-0.0099
-0.0036
0.0631
0.1689
0.0890
0.1906
0.1979
0.2312
0.1505
0.2060
= Ao + Atl +
+0.00401
+0.00481
+0.00361
+0.00431
+0.00511
+0.00301
+0.00341
+0.00651
+0.00641
+0.00591
+0.00741
+0.00701
+0.00711
+0.00811
+0.00871
+0.00951
+0.01081
+0.01171
+0.01261
+0.00611
+0.01411
+0.00521
A2T)b
+0.213/T
+0.291/T
+0.172/T
+0.243/T
+0.273/T
+0.164/T
+0.137/T
+0.290/T
+0.263/T
+0.292/T
+0.329/T
+0.237/T
+0.268/T
+0.302/T
+0.152/T
+0.278/T
+0.304/T
+0.263/T
+0.107/T
+0.208/T
+0.106/T
+0.256/T
R2c
0.57
0.74
0.46
0.50
0.76
0.58
0.52
0.74
0.79
0.82
0.81
0.77
0.95
0.88
0.78
0.51
0.82
0.55
0.70
0.45
0.83
0.40
Threshold .
concentration, ppm
1 hr
0.22
0.26
0.24
0.18
0.26
0.17
0.18
0.35
0.29
0.33
0.38
0.35
0.29
0.34
0.26
0.50
0.45
0.51
0.38
0.47
0.33
0.49
4 hr
0.06
0.04
0.11
None
0.05
0.05
0.08
0.13
0.09
0.11
0.13
0.17
0.09
0.11
0.14
0.27
0.22
0.31
0.29
0.31
0.25
0.30
8 hr
0.03
0.01
0.09
None
0.02
0.03
0.06
0.09
0.06
0.09
0.09
0.14
0.06
0.08
0.13
0.25
0.18
0.20
0.20
0.30
0.23
0.27
Number
data
points
471
71
100
20
30
62
197
373
25
68
104
27
24
15
59
291
36
13
16
46
18
45
Mean values6
Cone. (C), Time (T),
ppm hr
0.29
0.37
0.34
0.31
0.37
0.30
0.23
0.37
0.41
0.39
0.40
0.36
0.28
0.47
0.28
0.45
0.30
0.45
0.55
0.39
0.41
0.39
1.74
1.66
1.42
1.50
1.66
1.23
1.90
1.67
1.29
1.61
1.59
1.91
2.13
1.25
1.99
1.55
1.89
1.47
1.50
2.50
1.41
2.17
Response (I), Dose,
% ppm x hr
45.4
50.9
40.1
56.5
40.2
47.2
38.9
27.0
33.5
31.0
25.0
22.9
23.0
28.9
15.7
10.6
12.2
6.5
17.8
7.8
13.3
12.6
0.503
0.608
0.480
0.491
0.611
0.370
0.448
0.625
0.532
0.625
0.642
0.687
0.595
0.508
0.551
0.696
0.722
0.655
0.819
0.905
0.581
0.847
Equations were developed from exposures limited in time (0.5 to 8 hr except for 2 to 12 hr points in the sensitive group) and denote acute
responses of the plants. Concentrations range from 0.05 to 0.99 (1.0) ppm and responses from 0 to 99 (100)% of control.
C is ozone concentration in ppm, I is percent injury, T is time in hr, and A , A., and A2 are constants (partial regression
coefficients) that are specific for pollutant plant species or group of species, and environmental conditions used.
Multiple correlation coefficient squared, which represents the percent variation explained by the model.
For 5 percent response in 1-, 4-, and 8-hr periods.
eFrom the computer analysis.
Source: U.S. Environmental Protection Agency (1978.)
-------
time (1976) and represented the lowest concentration and time reported to
cause visible injury on various plant species. These data were based on more
than 100 studies of agricultural crops and 18 studies of tree species. In the
figures, the shaded areas represented the range of uncertainty in the data.
Foliar injury was considered unlikely at doses below and to the left of the
shaded areas. The limiting values were summarized as follows:
1. Agricultural crops:
a. 0.20 to 0.41 ppm for 0.5 hr.
b. 0.10 to 0.25 ppm for 1.0 hr.
c. 0.04 to 0.09 ppm for 4.0 hr.
2. Trees and shrubs:
a. 0.20 to 0.51 ppm for 1 hr.
b. 0.10 to 0.25 ppm for 2 hr.
c. 0.06 to 0.17 ppm for 4 hr.
A concept similar to the limiting values for foliar injury was developed
to present the 0- concentrations and durations which could potentially reduce
plant growth and yield (Figure 6-6). In the figure, the line displays the
boundary of mean 03 concentrations and exposure durations below which effects
on growth and yield were not observed. Most of the data points represented
effects on growth rather than on yield as defined in the present document (see
Section 6.2.5). The graphical analysis indicated that the lower limit for
effects was a mean 0, concentration of 0.05 ppm for exposure durations greater
than 16 days. At exposure durations of less than 16 days, the 0- response
threshold increased to about 0.10 ppm at 10 days and 0.30 ppm at 6 days.
In the sections that follow, greater emphasis will be placed on yield
loss rather than just injury. Visible foliar injury will be considered for
those plants in which the foliage is the marketable plant part (yield), for
plants used for aesthetic purposes, and for plants used as bioindicators.
In the following portions of Section 6.4, the use of plants as bioindica-
tors and effects on vascular and nonvascular plants will be discussed. Bio-
indicators are important, because they provide useful information about loca-
tions displaying potential 0- impacts and may be useful in elucidating 0- as a
causative factor in yield loss.
6-76
-------
I.U
E
a
a
Z
O
<
GC
Z 0.1
Ul
O
Z
0
o
UJ
Z
o
N
O
0.01
4
-I I I I I I | I I I I I I I I
47 •
21Q 11Q
— \ 44 • 19CB18 045 046 •48-52
\ D10
— XO24 16DO17 Q31 7QD20
\ 1514 30 59
40* CBO itj • •42439D9
\ 1213 41 AC
\ 26 *8 5« 33
— \ 109 CD 29 54M55,56««58
^•v V j ^^^^
_ \ 39 3*
\ 2«
OCV 6 4* 57
^3 ^r ^^ T— ^^..^.^ — - ^k.
^ ^^ 30 ^^^ 34 ™
53
EXPOSURE, hr/day
~ A < 1.99
D 2 TO 3.99
— O 4 TO 5.99
NOS. = REF. NOS. ON TABLE 11-4
I I Mi I I I I I I I Ml I
I 6 8 10 20 40 60 80 100 200
EXPOSURE PERIOD, days
—
—
—
^^_
—
••MH
— —
—
*"~
—
—
—
40
Figure 6-6. Relationship between ozone concentration, exposure
duration, and reduction in plant growth or yield (see Table 6-18).
Numbers on the figure refer to reference numbers in Table 11 -4, U.S.
Environmental Protection Agency (1978).
Source: Derived from National Research Council (1977); cited in U.S.
Environmental Protection Agency (1978).
6-77
-------
6.4.1 Bioindicators of Ozone Exposure
Plants are known to respond differentially to the characteristics of the
environments that they occupy (Treshow, 1980b). Temperature, moisture, solar
radiation, elevation, and soil quality are obvious environmental features that
affect the distribution and relative performance of vegetation. Because
established plants are confined to a particular location, they depend primarily
on that local environment to meet their requirements for growth and reproduc-
tion; therefore, plant growth and yield integrate all environmental factors.
Thus, vegetation can act as a biological indicator of the environment, which
includes air pollutants.
Because plants growing in a particular environment are integrated products
of that environment, they can provide important information about air pollution
effects. The response if a plant is the direct expression of the pollutant in
tion occurrence and magnitude (Laurence, 1984). Therefore, bioindicators
provide a direct method for understanding the risk that pollution presents to
the biological components of the affected environment (Guderian, 1977). For
this reason, there is renewed interest in biological methods for determining
air pollution effects (Manning and Feder, 1980).
6.4.1.1 Bioindicator Methods. As the use of plants to monitor air pollution
has increased, better methods have been developed for relating plant response
to pollution exposure. Manning and Feder (1980) have summarized the important
attributes of a bioindicator species. To perform predictably, the plants
should be sensitive to a specific pollutant, genetically uniform, native or
adaptable to the region, produce characteristic symptoms, grow indeterminately,
and respond proportionally to pollutant exposure. To minimize sources of
variation further, efforts should be made to provide uniform soil and water
conditions and to ensure observation by trained personnel (Oshima et al.,
1976; Posthumus, 1976, 1980). The aim of these measures is to standardize the
plant and growing conditions so that effects of the pollutant are the major
sources of variation in the subsequent analysis (Teng, 1982). During the past
10 years, substantial progress has been made toward improved understanding of
the variables affecting the performance of indicator species. Specific examples
of these studies are summarized in this section.
6.4.1.2 Response of Indicator Species. Early studies with indicator species
generally focused on visible symptoms, the most obvious reaction of a plant to
changes in its environment. These responses included chlorosis or necrosis of
tissues and typically represented the effects of an acute exposure to a single
6-78
-------
pollutant (Feder and Manning, 1979; Heck, 1966; Heggestad and Darley, 1969;
Laurence, 1984). The identification and application in the 1960s of very
sensitive species such as Bel W-3 tobacco (Heggestad and Menser, 1962) pro-
vided predictive means by which to identify exposures to progressively lower
concentrations of 0, (Feder, 1978). There is general agreement that this
tobacco cultivar will predictably respond to an 0, exposure above 0,04 ppm for
4 hr (Ashmore et al., 1978) when environmental conditions are favorable.
Broad-leaved (dicotyledonous) and narrow-leaved (monocotyledonous) plants
show different symptoms from exposure to 0-. The foliage of dicotyledonous
plants initially appears water-soaked as the result of injury to palisade cell
membranes (U.S. Environmental Protection Agency, 1976). These areas appear
shiny or oily within hours of the exposure and have characteristic flecks or
stipples when the water-soaked area dries (Figure 6-7). Flecks (Figure 6-8)
are small lesions formed when groups of palisade or mesophyll cells, or both,
die and the associated epidermal cells collapse (U.S. Environmental Protection
Agency, 1976). They may be yellow or tan; and if the injury is extensive the
entire leaf surface may appear bronzed. Individual flecks may coalesce to
form bifacial lesions that appear on both leaf surfaces. "Stipples" (Figure 6-7)
are small groups of red, purple, or black pigmented palisade cells (U.S.
Environmental Protection Agency, 1976) that can be seen through the uninjured
epidermal layer of the upper leaf surface. The leaf veins are also uninjured
and form angular boundaries to the pigmented areas.
Monocotyledonous plants generally do not have differentiated mesophyll
tissue, and ozone injury typically appears as chlorotic spots or white flecks
between veins (U.S. Environmental Protection Agency, 1976). This injury may
extend to form long white or yellow streaks between the parallel veins of sen-
sitive plants, which become, in their most severe form, leaf bands (Figure 6-9).
Ozone injury to the foliage of coniferous plants is described as chlorotic
mottle and tipburn (U.S. Environmental Protection Agency, 1976). Small patches
of needle tissue are injured and turn yellow. These areas are surrounded by
healthy green tissue so that the needle appears mottled (Figure 6-10). When
the entire needle tip dies, it first turns reddish brown and then gray.
Tipburn is also a characteristic of 0- injury. In both cases, it is usual for
only current-year needles to be affected after acute exposures to 0,.
Long-term exposure to low pollutant concentrations may adversely affect
plant health without producing visible symptoms. Chronic injury from this
type of exposure may be represented by reductions in growth or yield, or both;
6-79
-------
Figure 6-7. Ozone injury to Bel W-3 tobacco. Clear interveinal
areas represent necrotic tissue (fleck and bifacial necrosis).
CUTICLE
ADAXIAL EPIDERMIS
PALISADE LAYER
OZONE INJURY
SPONGY MESOPHYLL
STOMATA
ABAXIAL EPIDERMIS
Figure 6-8. Schematic cross section of typical dicot leaf showing
ozone injury to palisade cells and collapsed epidermal cells.
6-80
-------
Figure 6-9. Ozone injury to oats. Clear areas represent
bleached and necrotic tissue.
Figure 6-10. Ozone injury to conifer needles. Clear areas
represent injured tissue (chlorotic mottle and tipburn).
6-81
-------
or by premature defoliation resulting from changes in photosynthesis, respira-
tion, leaf chlorophyl content, or other processes (Dochinger et al. , 1970;
Feder, 1978; Heck, 1966; Laurence, 1984; Posthumus, 1976).
6.4.1.3 Bioindicator Systems. Although many field biologists have identified
certain plants as indicators of pollutants, few have published documentation
of the sensitivity of specific plants to ambient 03 in the field or in natural
environments. Duchelle and Skelly (1981) and Skelly et al. (1982) characterized
the response of milkweed to 0, in both field and laboratory studies. This is
a particularly valuable study, because it defines the response of a plant that
has been classed as a sensitive bioindicator in the field and establishes a
baseline sensitivity that can be reevaluated in the future to detect possible
changes in the frequency of sensitive individuals in the field. Benoit et al.
(1982), reporting on the radial growth of eastern white pine as an indicator
of 0, pollution, were able to identify three classes of eastern white pine
(sensitive, intermediate, and tolerant). Studies in the southern California
mountains (Miller, 1973) showed that the radial growth of ponderosa and Jeffrey
pines was an indicator of ambient 0., exposure. Although a good relationship
between radial growth and observed 0, sensitivity exists, it is probably rea-
listic to use this procedure only as a measure of long-term effects because it
requires the detailed analyses of tree rings and precipitation patterns.
There have been several reports on the use of plants in systems designed
to detect the presence of elevated concentrations of ozone. Many early stud-
ies (e.g., Heck, 1966) were conducted to assess the spatial and temporal
distribution of air pollution using sensitive indicator plants. In most
cases, poor correlations between measured oxidants and plant injury were
found. With the identification of Bel W-3 tobacco as a sensitive indicator of
elevated ambient 0- concentrations (Heggestad and Menser, 1962), a new series
of studies was conducted (e.g., Heck et al., 1969, Heck and Heagle, 1970;
Jacobson and Feder, 1974; Naveh et al., 1978; Goren and Donagi, 1979; Horsman,
1981; Ashmore et al. , 1978; 1980; Bell and Cox, 1975). The most widespread
network established to determine the spatial and temporal distribution of
ambient-oxidant-induced injury on Bel-W3 tobacco was that described by Jacobson
and Feder (1974). The bioindicator sites were located in nine states ranging
from North Carolina to Maine. The authors observed both temporal and spatial
variations in 0- injury and concluded that Bel-W3 could be used to indicate
the present of 0- but could not reliably indicate the 03 concentration. A
6-82
-------
major problem identified by the authors was the necessity of growing Bel W-3
plants under pollution-free conditions prior to their use.
Oshima (1974) devised a bioindicator system for use in California that
utilized pinto bean. In field trials, a strong and significant relationship
was found between injury observed on bean leaves and average weekly ambient 0-
dose. His measure of 03 dose consisted of a censored sum (hours greater than
0.1 ppm) of ambient 03 concentrations obtained from nearby physical monitors.
It would be feasible to use such a system on a large scale to assess, at least
qualitatively if not quantitatively, the spatial and temporal occurrence of
phytotoxic concentrations of 0.,.
Posthumus (1976) reported the results of a study to investigate the
occurrence and distribution of 0~ by using Bel W-3 tobacco at 31 sites through-
out the Netherlands. He reported, "It is possible to determine the place and
time with the highest mean intensity or highest frequency of injury by 0.,...".
A '"fingerprint1" can be produced and, by comparing patterns from year to
year, specific trends in the occurrence of pollution may be identified. He
further concluded that, "The clear advantage of plants as indicators of air
pollution is that these show the result of the action of the pollutants on
living material", and added that, "In this way it could be a rather efficient
and relatively inexpensive manner to follow trends in air pollution and to
evaluate sanitation measures."
Nouchi and Aoki (1979) used morning glory as an indicator of photochemical
oxidants (primarily 0~). In studies conducted both in the laboratory and
field, they were able to model the effects of 0., on leaf injury, including the
effects of previously occurring exposures. Field verification of their model
showed that they were able to determine (within acceptable margins of error)
oxidant levels on a given day by using measurements of visible injury to
morning glory. They emphasized, however, that the most valuable use of their
system was to characterize the frequency and spatial distribution of elevated
oxidant concentrations.
The common theme in all these studies is that a good understanding of the
occurrence of elevated (L concentrations can be obtained by using the visible
response of sensitive plants. While the methodology for biomonitoring is still
in the early stages of development, bioindicators have a certain value as in-
tegrators, by providing information on where, when, and how often CL concen-
trations may be reaching phytotoxic concentrations. The value of deploying
networks of bioindicators has been demonstrated in the early detection of
6-83
-------
developing regional oxidant pollution problems, in the identification of
trends in pollutant occurrence, and in the supplementation of physical monitor-
ing networks to provide additional information on the biological effects of
pollution for the assessment of crop loss (Laurence, 1984).
Following the initial observations that plants exhibited foliar injury
after exposure to ambient oxidants, studies were undertaken to determine the
concentrations and spatial distribution of ozone based on the appearance of
visible foliar injury symptoms (e.g., Heck et al., 1969). Although there has
been no single study to determine whether ozone injures vegetation in every
state in the country, a number of studies (using varying degrees of detail)
have identified ozone injury on a diverse range of vegetation (Table 6-15).
Based on the occurrence of foliar injury, ozone impacts have been observed on
horticultural and agricultural crops, native vegetation, and bioindicator
plants (Table 6-15; Figure 6-11) in at least 27 states. Because comprehensive
studies of the distribution of ozone injury have not been conducted, it is not
possible to determine whether ozone injury does not occur in the other states
or whether it has not been reported because it has not been studied.
Biological methods for assessing the extent, and in some cases the inten-
sity, of 0, effects have value beyond the data provided by physical-chemical
monitoring methods. The physical-chemical methods can describe the concentra-
tion and duration of exposure and can only show the probability that an effect
may have occurred. In contrast, vegetation (bioindicators) can provide direct
indication that the pollution episode reached injurious levels, subject to the
joint influence of other environmental variables. Although the presence of
visible foliar symptoms on vegetation cannot be directly related to effects on
plant growth or yield, they do indicate that elevated levels of 0, have oc-
curred. The detection of visible symptoms is an indication that additional
studies should be undertaken to determine whether effects on plant growth and
yield are occurring. Caution should be used when relying on visible symptoms,
however, because the lack of foliar injury is not proof that effects on growth
are not occurring.
6.4.2 Response of Microoorganisms and Nonvascular Plants to Ozone
6.4.2.1 Microorganisms. Most studies with this group of organisms (bacteria
and fungi) have often used CL concentrations in excess of 1 ppm, much higher
than those expected to occur in ambient air. Direct effects of ozone on
microorganisms and, in some instances, their capacity to incite plant diseases
6-84
-------
TABLE 6-15. PARTIAL LISTING OF STATES WHERE AMBIENT OZONE INJURY HAS BEEN OBSERVED ON SENSITIVE VEGETATION
Horticultural
State crops
Arizona
California Ginkgo
Connecticut Petunia
Delaware
Florida
en Georgia Petunia
00
en Illinois
Indiana
Kentucky
Maine
Maryland
Massachusetts
Michigan
Minnesota
Missouri
Agricultural crops
Tomato ,
grapes, cotton,
citrus, sweet corn,
potato
Tobacco, alfalfa,
beans, tomato, potato,
cucurbits, oats,
6 vegetables
Potato
Soybean
Soybean, snap bean,
and potato
Potato
Potato, field bean,
bean
Soybean
Natural vegetation Bioindicator
Tobacco
Ponderosa pine, Pinto bean,
Jeffrey pine, 5 other alfalfa,
pine species, black oak, grapes
and 11 herbaceous species
Tobacco
Tobacco
Tobacco
Tobacco
White pine
Tobacco
Tobacco
Sycamore Tobacco
Tobacco
Tobacco
Tobacco
Reference
National Research Council (1977)
Seibert (1970); Oshima (1974b);
Oshima et al. (1976); Oshima
et al. (1977a); Richards et al.
(1958); Brewer and Ferry (1974);
Thompson et al. (1969); Foster
et al. (1983a); Miller and
Elderman (1977); Williams et al.
(1977)
Jacobson and Feder (1974);
Heggestad and Middleton (1959);
Rich et al. (1969)
Jacobson and Feder (1974);
Brasher et al. (1973)
Dean (1963)
Walker and Barlow (1974)
Kress and Miller (1983)
Usher and Williams (1982)
Menser (1969)
Jacobson and Feder (1974)
Santamour (1969); Howell et al.
(1979); Heggestad (1973);
Heggestad et al. (1980)
Jacobson and Feder (1974);
Manning et al. (1969)
Hooker et al. (1973); Olson and
Saettler (1979)
Laurence et al. (1977)
Heck et al. (1969}
-------
TABLE 6-15 (cont'd). PARTIAL LISTING OF STATES WHERE AMBIENT OZONE INJURY HAS BEEN OBSERVED ON SENSITIVE VEGETATION
State
Horticultural
crops
Agricultural crops
Natural vegetation
Bioindicator
Reference
i
00
New Jersey
New York
North Carolina
Ohio
Pennsylvania
South Dakota
Tennessee
Utah
Virginia
Washington
West- Virginia
Wisconsin
Austrian pine,
petunia, sweet pea,
and carnation
Petunia
Austrian pine, mimosa,
white oak, dogwood,
hemlock, silver maple,
bluegrass, 21 addi-
tional trees
Petunia
Potato, spinach,
cucurbits, cereals,
grape, broccoli, alfalfa,
10 other vegetables
Tomato, bean, grape,
soybean
Snap bean, soybean,
spinach, field corn,
winter wheat
Bean, radish, squash,
tomato, alfalfa, oats
Alfalfa, corn, oats,
tomato, sugar beets,
beans, grape, spinach,
broccoli, Swiss chard,
9 additional crops
Barley, oats, spinach,
corn, radish
Potato
Chickweed,
orchard grass,
red clover, and pine
Eastern white pine
White pine, spruce,
blackberry, blueberry,
poison ivy, nightshade,
chickweed, and 7 additional
species
Onions
Tulip poplar, green ash,
sweet gum, eastern white
pine, 3 additional
conifers, milkweed, wild
strawberry, grasses,
sedges, forbs
White pine
Pine
Tobacco
Tobacco
Tobacco
Tobacco
Tobacco
Tobacco
Tobacco
Tobacco
Tobacco
Jacobson and Feder (1974);
Harkov and Brennan (1982);
Clarke and Brennan (1981);
Daines (1963); Daines et al.
(1967); Pell (1973)
Jacobson and Feder (1974);
MacLean and Schneider (1976);
Kender et al. (1973); Troiano
et al. (1983)
Jacobson and Feder (1974);
Heggestad et al. (1980); Heagle
and Heck (1980)
Heck and Heagle (1970);
Dochinger and Seliskar (1970);
Reinert et al. (1970)
Jacobson and Feder (1974);
Seibert (1970); Moyer et al.
(1974); Lacasse (1971)
Gardner (1973)
Menser (1969)
Tingey and Hill (1967)
Duchelle and Skelly (1981);
Heggestad (1973)
National Research Council (1977)
Wood and Pennypacker (1975)
Usher and Williams (1982);
Daines et al. (1967)
-------
= o,
si
INJURY ON
NSITIVE VEGETATION
Figure 6-11. States in which ozone-induced injury to vegetation has occurred as
reported in the published literature.
6-87
-------
have been reviewed by Laurence (1981) and Heagle (1973, 1982), and in Section
6.3.2.1.3 in this chapter.
The 0, concentration required for direct impact on microorganisms may be
quite high. The data of Hibben and Stotsky (1969) are illustrative. These
investigators examined the response of detached spores of 14 fungi to 0.1 to
1.0 ppm of Oo for 1, 2, and 6 hr. The large pigmented spores of Chaetomium
sp. , Stemphyliurn sarcinaeforme, S. loti, and Alternaria sp. were not affected
by 1.0 ppm. Germination of Trichoderma viride, Aspergillus terreus. A. niger,
PeniciIlium egyptiacum, Botrytis allii, and Rhizopus stolonifera spores was
reduced by 0, exposure, but only at concentrations above 0.5 ppm. The germi-
nation percentages in the small colorless spores of Fusarium oxysporum,
Colletotrichum largenarium, VerticiIlium albo-atrum, and V. dahliae were
reduced by 0.5 ppm and occasionally by concentrations of 0.25 ppm of 0- for 4
to 6 hr; lower doses stimulated spore germination in some cases. The ability
of ozone to reduce spore germination in fungi apparently depends on the species,
spore type, morphology, moisture, and substrate. Moist spores were more
sensitive than dry ones; single-celled spores and those with thin cell walls
were most sensitive.
Hibben and Stotsky (1969) found 0- toxic to moist fungus spores of some
species, even at concentrations of 0.1 ppm when applied for 28 hr. Exposure
to 0.5 and 1.0 ppm reduced or prevented germination of spores of all species
tested. Ozone at 0.1 ppm for 4 hr or at 1.0 ppm for 2 hr stopped apical cell
division of conidiophores of Alternaria solani and caused collapse of the
apical cell wall (Rich and Tomlinson, 1968).
Ozone can inhibit fungal growth on artificial media but rarely kills the
fungus even at high concentrations. Differences in species sensitivity are
known. In several fungi, exposure to 0- (0.10 or 0.40 ppm for 4 hr) caused a
10- to 25-fold increase in sporulation (Heagle, 1973). The same author reported
the effects of low 03 exposures on three obligate parasitic fungi. Germina-
tion of spores was not affected in any of these studies (Heagle, 1975).
Reduced sporulation, germination, and pathogenicity of Botrytis cinerea were
observed by Krause and Weidensaul (1978a,b) after exposure of the microorganism
i_n vitro and i_n vivo to 0.30 ppm of 0, for two 6-hr periods.
6.4.2.2 Lichens, Mosses, and Ferns. Previously, there was little information
describing the effects of 0- on lichens in natural environments, but Sigal and
Nash (1983) have recently conducted an extensive study of lichen distribution
relative to oxidant air pollution in southern California. Collections of
6-88
-------
lichen from regions of high (1300 hr > 0.09 ppm, 1968-1974, San Bernardino
Mountains) and low (Cuyamaca Rancho State Park) levels of oxidant pollution
were compared with collections made in 1913. The frequency and cover of the
current lichen communities in these regions were also compared with calculated
levels of 0- associated with injury to pines as reported earlier (National
Research Council, 1977). Additionally, lichens from unaffected areas were
transplanted to ecologically similar sites in affected areas.
In this multisite study, the authors found consistently high levels of
injury to lichens in areas with high levels of 0-. In polluted areas, only 8
of 16 previously reported species were still present, and 4 were found only in
trace amounts. This compared with 15 of 16 species stil.l present in areas
with low levels of 0,- Transplanted lichens performed poorly in areas where
injury to pine was most extensive and calculated levels of 0- were highest.
The authors concluded that lichen communities in southern California were not
adversely affected if the cumulative oxidant dose level was below about 300
ppm-hr per year. This dose was calculated using all concentrations greater
than 0.04 ppm 0~.
In a laboratory study, Nash and Sigal (1979) fumigated two species of
lichens (Parmelia sulcata and Hypogymnia enteromorpha) with 0- at concentra-
tions of 0.5 and 0.8 ppm for 12 hr. The former exhibited greater sensitivity
than the latter, as measured by a reduction in gross photosynthesis. P.
sulcata, which grows on black oak, is absent from the San Bernardino Mountains;
H. enteromorpha is present but apparently deteriorating. The authors noted
that, for these species, the pattern observed in the laboratory is consistent
with that found in field observations in Southern California, where extensive
0- injury occurs. In another study (Ross and Nash, 1983), photosynthesis was
decreased at 0^ concentrations of 0.1, 0.25, and 0.50 ppm for 12 hr in Pseudo
parmelia caperata; however, effects were not found when Ramalina menziesei was
exposed to concentrations of 0- up to 0.5 ppm for 12 hr. Exposures of both
species to ozone at 0.10 ppm for 6 hr/day on 5 consecutive'days resulted in
the same responses seen at the higher concentrations.
Very little is known about the responses of mosses and ferns to 0.,. The
information in the previous EPA document (U.S. Environmental Protection Agency,
1978) indicates that, based on published information, significant effects
would not be expected at current ambient 0- levels.
6-89
-------
6.4.3 Losses in Vascular Plants from Exposure to Ozone
This section will relate losses in plant yield to (L exposure. Exposures
will be described in terms of duration and (L concentrations, but the statis-
tics used to characterize the exposure will take several forms. Yield loss is
defined as the impairment of the intended use of the plant (see Section 6.2.5)
and includes aesthetic value, foliar injury, plant appearance, and losses in
terms of number, size, or weight of the plant part that is normally harvested.
Yield loss can also be defined as a change in physical appearance, chemical
composition, or ability to withstand storage, traits that are collectively
termed crop quality.
6.4.3.1 Losses in Aesthetic Value and Foliar Yield. Losses in aesthetic
value are difficult, if not impossible, to quantify. For example, because of
its aesthetic value, the loss of or adverse effects on a specimen tree or
shrub in a landscape planting will result in a much greater economic loss than
the same impact on a tree or shrub of the same or similar species growing as a
part of a natural plant community. Foliar symptoms that can decrease the
value of an ornamental crop may occur on various types of plants (e.g., turf-
grasses, floral foliage, ornamental trees, and shrubs) with or without concomi-
tant growth reductions. The occurrence of foliar injury on other crops in
which the foliage is the marketable plant part (e.g., spinach, cabbage, tobacco)
can substantially reduce marketability and constitute a yield loss in economic,
if not biologic, terms.
Petunia, geranium, and poinsettia were exposed to 0- (up to 0.10 to 0.12
ppm for 6 hr/day) for 9 days (petunia), 8 days (geranium), and 50 days (poin-
settia) (Craker and Feder, 1972). Flower size was significantly reduced in
all three species at a concentration of 0.10 to 0.12 ppm. Ozone decreased
flower color in all three species: petunia (0.06 to 0.08 ppm), geranium (0.10
to 0.12 ppm), and poinsettia (0.02 to 0.04 ppm). All these changes in flower
appearance (yield) occurred without visible injury to the plant leaves. Five
begonia cultivars exposed to 0, (0.25 ppm for 4 hr/day for a total of 16 hr
over a 4-wk period) varied in foliar injury from 2 to 54 percent (Table 6-16);
flower size was also reduced (Reinert and Nelson, 1980).
Ozone injury on the foliage of ornamental trees and shrubs impairs their
appearance and may reduce their value. Mean foliar injury on eight azalea
cultivars exposed to 0.25 ppm of 03 (six 3-hr fumigations) ranged from 0 to 24
percent (Sanders and Reinert, 1982a). Stem weight was significantly reduced
6-90
-------
TABLE 6-16. FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE3
Plant species
FLOWERS
Begonia
(Schwabenland Red)
(Whisper '0' Pink)
(Fantasy)
o^ (Renaissance)
<Ł>
i—i
(Turo)
ORNAMENTAL TREES AND
Hybrid poplar
(Oorskamp)
(Zeeland)
Hinodegiri azalea
Black cherry
American sycamore
Hybrid poplar
03
concn. ,
ppm Exposure duration
0.25 4 hr/day, every 6th
day, 4 times
0.25
0.25
0.25
0.25
SHRUBS
0.041 12 hr/day, 23 wk
0.041
0.20 5 hr
0.20
0.20
0.20
Percent
foliar Monitoring Calibration Fumigation
injury method method facility Reference
54 (39%e dec. Chem Not given GH-CSTR Reinert and
in flower Nelson (1980)
wt)
25 (22%e dec.
in flower
wt)
2 (6%e dec.
in flower
wt)
15 (55%e dec.
in flower
wt)
8 (10% inc. in
flower wt)
Not given Chem NBKI GH-CH Mooi (1980)
(1333%e inc.
leaf drop)
Not given
(692%e inc.
leaf drop)
33 Chem NBKI GC Davis et al . (1981)
27
26
20
-------
TABLE 6-16 (cont'd). FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE3
Plant species
Yel low poplar
Black walnut
Delaware Valley
white azalea
Black elder
Spreading cotoneaster
Austrian pine
Eastern white pine
Virginia pine
>^3 Hinodegiri azalea
Korean azalea
Tree-of-heaven
Chinese elm
Mock-orange, sweet
Viburnum, tea
Viburnum, linden
American holly (<*)
American holly (9)
Amur privet
Black gum
Dense Anglogap yew
Mountain- laurel kalmia
Hete Japanese holly
03
concn. ,
ppm Exposure duration
0.20 5 hr
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.25 8 hr
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
Percent
foliar Monitoring Calibration Fumigation
injury method method facility Reference
19 Chem NBKI GC Davis et al. (1981)
12
12
11
4
0
0
0
95f Chem NBKI GC Davis and
Coppolino (1974)
70f
65f
24f
17f
5f
2f
Of
Of
Of
Of
Of
Of
of
-------
TABLE 6-16 (cont'd). FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE
Plant species
Hybrid poplar
Azalea (Red Wing)
(Snow)
(Glacier)
(Mersey Red)
j (Pink Gumpo)
\ (Mme. Pericat)
(Red Luann)
(Mrs. G.G. Gerbing)
TURFGRASS
Turfgrass
(Meyer zoysiagrass)
(Tufcote bermudagrass)
(Merion bluegrass)
(Kenblue bluegrass)
(K-31 tall fescue)
(NK-100 ryegrass)
(Penncross bentgrass)
(Pennlawn red fescue)
(Annual bluegrass)
03
concn. ,
ppm Exposure duration
0.25 12 hr/day, 24 days
0.25 3 hr/day, 6 days
over 4 wk
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.20 2 hr
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
Percent
foliar Monitoring Calibration Fumigation
injury method method facility Reference
Not given Chem Known 03 GH-CSTR Noble and Jensen
(50% inc. source (1980)
in leaf
abscission)
1 (32%e dec. Chem Known 03 GH-CSTR Sanders and
stem dry source Reinert (1982a)
wt)
0
24
21 (44%e dec.
stem dry wt)
0
4
8 (25%e dec.
stem dry wt)
9
0 Mast Not given CH Richards et al.
(1980)
0
0
2
7
9
14
17
20
-------
TABLE 6-16 (cont'd). FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE3
Plant species
Kentucky bluegrass
(Newport)
(Sydsport)
(Merion)
(Fylking)
(Windsor)
(S. Dakota (certified))
en
i
10 (Kenblue)
-P»
Kentucky bluegrass
(Adelphi)
(Baron)
(Birka)
(Cheri)
(Fylking)
(Merion)
(Nugget)
(Plush)
(Skofti)
(Sydsport)
(Touchdown)
03
concn. ,
ppm
0.
0.
0.
0.
0.
0.
0.
0.
10
10
10
10
10
10
10
15
Exposure duration
3.
7
3.
7
3.
7
3.
7
3.
7
3.
7
3.
7
6
5 hr/day
hr/day ,
5 hr/day
hr/day,
5 hr/day
hr/day ,
5 hr/day
hr/day ,
5 hr/day
hr/day,
5 hr/day
hr/day ,
5 hr/day
hr/day ,
hr/day ,
, 5 days
5 days
, 5 days
5 days
, 5 days
5 days
, 5 days
5 days
, 5 days
5 days
, 5 days
5 days
, 5 days
5 days
10 days
0.15
0.
0.
0.
0.
0.
0.
0.
0.
0.
15
15
15
15
15
15
15
15
15
Percent
foliar Monitoring Calibration Fumigation
injury method method facility Reference
0 Mast Not given CH Richards et al.
5 (1980)
5
12
9
14 .
9
14
7
15
10
17
12
17
6 UV Not given CH Elkiey and
Ormrod (1980)
0
0
19
0
9
8 (8% dec.
in leaf area)
0
0
12
0
-------
TABLE 6-16 (cont'd). FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE
Plant species
Kentucky bluegrass
(Victa)
Red top
(Common)
Creeping bentgrass
(Penncross)
Colonial bentgrass
(Exetes)
Red fescue
(Highlight)
(Pennlawn)
Perennial ryegrass
CTi
i
<•" FOLIAGE CROPS
Tobacco
(Bel B)
(White Gold)
Cabbage
(All Season)
Spinach
(Northland)
Spinach
(America)
(Winter Bloomsdale)
(Seven-R)
(Hybrid-424)
(Hybrid-7)
03
concn. ,
ppm
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.05
0.10
0.05
0.10
0.05
0.10
0.05
0.10
0.13
0.13
0.13
0.13
0.13
Percent
foliar Monitoring Calibration Fumigation
Exposure duration injury method method facility Reference
6 hr/day, 10 days 10 UV Not given CH Elkiey and Ormrod
(1980)
40
20
6
2
6 (27%e dec.
in leaf area)
11 (20%e dec. in leaf
area)
4 hr 0 Mast NBKI GH-CH Tingey et al.
(19735)
0
0
0
0
0
0
0
7 hr/day avg for 49 (36%ng dec. Chem NBKI OT Heagle et al.
30 days (0.047 ppm 03 in fresh wt) (1979b)
ambient air each day)
52 (45%ng dec.
in fresh wt)
52 (55%ng dec.
in fresh wt)
54 (42%ng dec.
in fresh wt)
56 (43%ng dec. )
in fresh wt)
-------
TABLE 6-16 (cont'd). FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE
Plant species
Spinach
(Viking)
(Dark Green Bloomsdale)
(Viroflay)
(Chesapeake)
(Hybrid-612)
en
i
•Ł (Dixie Market)
Tobacco
(GC-166)
(CCC-E)
(GC-172)
(GC-169)
(GC-18)
(CCC-C)
(GC-46)
(CCC-L)
(GC-50)
(CCC-M)
03
concn. ,
ppm Exposure duration
0. 13 7 hr/day avg for 30 days
(0.047 ppm 03 ambient air
each day)
0.13
0.13
0.13
0.13
Ambient air 11 wk
(Beltsville,
MD)
Percent
foliar Monitoring Calibration Fumigation
injury method method facility Reference
58 (44%ng dec. Chem NBKI OT Heagle et al.
in fresh wt) (1979b)
58 (58%ng dec.
in fresh wt)
60 (33%ng dec.
in fresh wt)
63 (42%ng dec.
in fresh wt)
65 (61%ng dec.
in fresh wt)
65 (55%"g dec.
in fresh wt)
1 Mast Not given Field Menser and
Hodges (1972)
1
2
6
7
10
10
11
11
15
-------
TABLE 6-16 (cont'd). FOLIAR SYMPTOM EXPRESSION OF VARIOUS FLOWER, ORNAMENTAL TREE, SHRUB, TURFGRASS, AND
FOLIAR CROP SPECIES IN RESPONSE TO OZONE EXPOSURE3
CTi
Plant species
Tobacco
(CCC-J)
(CCC-S)
(Bel-C)
03
concn. ,
ppm Exposure duration
Ambient air 11 wk
(Beltsville,
MD)
Percent
fol iar
injury
18
25
55
Monitoring
method
Mast
Cal ibration
method
Not given
Fumigation
facility
Field
Reference
Menser and
Hodges (1972)
aWhere a column entry is blank, the information is as above.
* i
Chem = chemi luminescence; Mast = Mast oxidant meter (coulometric); UV = ultraviolet spectrometry.
CNBKI = neutral buffered potassium iodide.
GH = greenhouse; GH-CSTR = continuous stirred tank reactor in a greenhouse; OT = open-top chamber; GC = growth chamber; CH = specially designed exposure
chamber other than CSTR; GH-CH = exposure chamber in a greenhouse.
Significant at P = 0.05; ng = not given.
Severity index = [severity factor (0-5) x (% foliage injured) x (% population susceptible)] T 100.
-------
for three of the cultivars (Table 6-16). Tree and shrub species have developed
foliar injury following exposure to 0.20 ppm of 03 for 5 hr (Davis et al.,
1981). Visible injury to black cherry foliage occurred following a 4-hr
exposure at 0.10 ppm and 2 hr at 0.19 ppm of 03 (Davis et al., 1981). In an
earlier study, several species were exposed to 0~ (0.25 ppm for 8 hr) and
evaluated for foliar injury (Davis and Coppolino, 1974). Some common ornamen-
tals (holly, privet, yew, laurel, linden) exhibited no foliar injury, but
others (azalea, tree-of-heaven, elm) appeared to be relatively sensitive
(Table 6-16).
For ornamental tree plantings, excessive leaf drop decreases the value
and thus can be considered a yield loss. Ozone has been shown to induce
significant defoliation in hybrid poplar. Mooi (1980) noted increases of
about 7- and 13-fold in leaf drop of two poplar cultivars exposed to 0.041
ppm, 12 hr/day, for 23 wk (Table 6-16). Noble and Jensen (1980) reported a 50
percent increase in leaf drop of hybrid poplar exposed to 0.25 ppm of 0,, 12
hr/day, for 24 days (Table 6-16).
Species and cultivars of turfgrass have exhibited foliar injury when
exposed to 0.15 ppm of 03 (6 hr/day, 10 days) (Elkiey and Ormrod, 1980). The
extent of foliar injury was usually greater than the resultant growth inhibi-
tion. Ozone concentrations of 0.10 ppm for 3.5 hr/day for 5 days or 0.20 ppm
for 2 hr were high enough to elicit injury in most turf grasses (Richards et
al., 1980) (Table 6-16).
The appearance of the foliage on crops such as tobacco and spinach is
important to the value of these crops and may affect their marketability.
Tobacco and spinach failed to exhibit visible injury after exposure to 0.05 or
0.10 ppm of 03 for 4 hr (Tingey et al., 1973b) (Table 6-16). In a different
study, 11 spinach cultivars exhibited 49 to 65 percent mean foliar injury (and
33 to 61 percent mean fresh weight reduction) when exposed in the field to a
7-hr seasonal mean 0, concentration of 0.13 ppm (Heagle et al., 1979b) (Table
6-16). The physical appearance of cigar-wrapper tobacco leaves may be very
important to their value. Foliar injury from 0., has been documented in the
field (some cultivars are commonly used as bioindicators) and in controlled
fumigations. In the field, plants of commercial tobacco cultivars grown in
ambient air at Beltsville, MD, exhibited 1 to 55 percent 0~ injury (Menser and
Hodges, 1972) (Table 6-16). Ozone concentrations of 0.10 ppm for 2 hr induced
up to 20 percent foliar symptoms in controlled environment studies.
6-98
-------
The above data are examples of 0--induced impairments in the appearance
and aesthetic value of plants as the result of foliar injury. Such effects
occur at concentrations as low as 0.041 ppm for several weeks or 0.10 ppm for
2 hr, and these effects can constitute a yield loss when marketability of the
plants is decreased. The actual amount of yield loss due to decreased aesthetic
value or appearance may be more difficult to quantify than yield loss in
weight or bulk.
6.4.3.2 Yield Losses as Weight, Size, and Number. The previous criteria
document (U.S. Environmental Protection Agency, 1978) summarized the effects
of acute and chronic 0, exposures, with the primary focus on plant growth and
a few reports that specifically studied yield loss (Tables 6-17, 6-18).
Growth and yield reductions were observed in a diverse range of plant species
at various exposure durations and 0- concentrations. The majority of the
studies were conducted in greenhouse or controlled-environment chambers, with
only a few studies conducted in the field. These data indicated that as the
exposure duration increased, the mean 0., concentrations at which growth effects
occurred decreased. When the exposure duration exceeded 15 days (not continuous
exposures), mean 0- concentrations of 0.05 ppm and greater caused significant
growth and yield reductions. In field studies, significant growth and yield
reductions were observed in commercial varieties of sweet corn, soybean, and
pine seedlings (Heagle et al., 1972; Heagle et al., 1974; Wilhour and Neely,
1977) when the seasonal 6-hr 0- concentration was 0.10 ppm or greater. In
another field study, significant growth and yield reductions occurred in
alfalfa when the 7-hr seasonal mean 0., concentration was 0.05 ppm or greater
(Neely et al., 1977).
In the following sections, yield losses are summarized in terms of weight
or size and decrease in number from studies in which known amounts of 0., were
added to either charcoal-filtered or ambient air. The effects of ambient 0,
on yield are also presented.
6.4.3.2.1 Ozone addition studies. Ozone-induced yield-loss studies have used
a variety^of experimental approaches. Some studies have attempted to ap-
proximate typical agronomic conditions, and others have deviated from typical
field practices in an attempt to have better control of the experimental
conditions. Open-top chamber data will be discussed first, because most of
these studies attempted to follow typical field practices. Results from
6-99
-------
TABLE 6-17. EFFECTS OF SHORT-TERM EXPOSURES ON GROWTH AND YIELD OF SELECTED PLANTS'
01
Plant Ozone concen-
species tration, ppm
Begonia, cultivar
White Tausendschon
Petunia, cultivar
Capri
Coleus, cultivar
Scarlet Rainbow
Snapdragon, cultivar
Floral Carpet, mixture
Radish, cultivar
Cavalier Cherry Belle
Radish
Cucumber, cultivar
Ohio Mosiac
Potato, cultivar
Norland
Tomato, cultivar
Fireball
Tomato, cultivar
Fireball
Onion, cultivar
Spartan Era
Tobacco, cultivar
Bel W3
0.10
0.20
0.40
0.80
0.10
0.20
0.40
0.80
0.10
0.20
0.40
0.80
0.10
0.20
0.40
0.80
0.25
0.40
1.00
1.00
1.00
1.00
0.50
1.00
0.50
1.00
0.20
1.00
1.00
0.30
Exposure
time, hr
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
1-5(1)^
1.5(2)^
1.5(3)C
1
4
4 .
4(3)C
1
1
1
1
24
1
4
2
5,
10,
19,
38,
9,
11,
21,
31,
2,
17,
24,
39,
0,
6,
8,
16,
36,
38,
37,
63,
75,
19,
37,
o,
30,
15,
20,
15,
25,
o,
19,
49,
48,
Plant response, % reduction
from control
avg. of 3 growth responses: shoot wt,
flower wt, flower no.
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
avg. of same responses
top dry wt (Cavalier)
root dry wt (Cherry Belle)
root dry wt
root dry wt
root dry wt
top dry wt (1% injury)
top dry wt (18% injury)
tuber dry wt (no injury)
tuber dry wt (injury severe)
plant dry wt (grown in moist soil)
plant dry wt (grown in moist soil)
increase in plant dry wt (grown in dry soil)
increase in plant wt (grown in dry soil)
effect
plant dry wt (no injury)
plant dry wt
chlorophyll content
Taken from U.S. Environmental Protection Agency (1978).
DUnless otherwise noted.
"Number of exposures in parentheses.
-------
TABLE 6-18. EFFECTS OF LONG-TERM, CONTROLLED EXPOSURE'
AND FOLIAR INJURY IN SELECTED PLANTS0
ON GROWTH, YIELD
01
i
Fig. 6-6b
Plant species nos.
Lemna, duckweed
Carnation
Geranium
Petunia
Poinsettia
1
2
3
4
5
Ozone
concentration,
ug/m3 (ppm)
Exposure
time
196 (0.10) 5/day, 14 days
98-177 (0.05-0.09) 24/day, 90 days
137-196 (0.07-0.10) 9. 5/day, 90 days
98-137 (0.05-0.07) 24/day, 53 days
196-235 (0.10-0.12) 6/day, 5 days/week,
Plant response, % reduction
from control
100,
50,
50,
50,
30,
39,
flowering; 36, flowering
(1 wk after exposure completed)
frond doubling rate
flowering (reduced vegetative
growth)
flowering (shorter flower
lasting time, reduced vegetative
growth)
flower fresh wt
bract size
10 weeks
Radish
6
98
(0.
05)
8/day,
5
days/week,
5 weeks
98
(0.
05)
8/day,
5
(mixture
for same
Beet, garden
Bean, cultivar
Pinto
Bean, cultivar
Pinto
Bean, cultivar
Pinto
7
8
9
10
11
12
13
14
392
255
290
490
686
290
290
290
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
20)
13)
15)
25)
35)
15)
15)
15)
3/day,
8/day,
2/day,
2/day,
2/day,
2/day,
3/day,
4/day,
38
28
63
63
63
14
14
14
days/week
of 03 and S02
periods)
days
days
days
days
days
days
days
days
54,
20,
63,
22,
50,
79,
73,
70,
33,
95,
97,
8,
8,
23,
root fresh wt
leaf fresh wt
root fresh wt
leaf fresh wt
top dry wt
top fresh wt
root fresh wt
height
plant wt; 46,
plant dry wt;
plant dry wt;
leaf dry wt
leaf dry wt
pod
99,
100
leaf dry wt (data
whole plants,
fresh wt
pod fresh wt
, pod fresh wt
available on
roots, leaves, injury,
and three levels
Bean, cultivar
Pinto
15
16
17
290
440
440
(0.
(0.
(0.
15)
225)
225)
6/day,
2/day,
4/day,
14
14
14
days
days
days
49,
44.
68,
stress)
leaf dry wt
leaf dry wt
leaf dry wt (data
whole plants,
of soil moisture
available on
roots, leaves, injury,
and three levels
18
588
(0.
30)
I/day,
14
days
40,
stress)
leaf dry wt
of soil moisture
-------
TABLE 6-18 (cont'd). EFFECTS OF LONG-TERM, CONTROLLED OZONE EXPOSURES ON GROWTH, YIELD
AND FOLIAR INJURY IN SELECTED PLANTS3
en
i
o
PO
Plant species
Tomato
Corn, sweet,
cultivar Golden
Jubilee
Wheat, cultivar
Arthur 71
Soybean
Soybean
Alfalfa
Grass brome
f
Alfalfa
f
Alfalfa0
Alfalfa
Pine, eastern
Fig. 6-6b
nos.
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Ozone
concentration,
pg/m3 (ppm)
588 (0.30)
392 (0.20)
686 (0.35)
392 (0.20)
686 (0.35)
392 (0.20)
98 (0.05)
196 (0.10)
196 (0.10)
290 (0.15)
390 (0.20)
290-647
(0.15-0.33)(varied)
196 (0.10)
98 (0.05)
98 (0.05)
196 (0.10)
Exposure time
3/day, 14 days
2.5/day, 3 days/week
14 weeks
2.5/day, 3 days/week,
14 weeks
3/day, 3 days/week
till harvest
3/day, 3 days/week
till harvest
4/day, 7 days
(anthesis)
8/day, 5 days/week
3 weeks
8/day, 5 days/week
(mixture of 03 and S02
for same periods)
8/day, 5 days/week
3 weeks
2/day, 21 days
2/day, 21 days
2 day, 21 days
4/day, 5 days/week
growing season
6/day, 70 days
7 /day, 68 days
8/day, 5 days/week
12 weeks
4/day, 5 days/week
76,
1,
45,
13,
24,
20,
54,
30,
13,
16,
20,
21,
9,
16,
26,
39,
83,
4,
20,
50,
30,
50,
18,
3,
Plant response, % reduction
from control
leaf dry wt
yield; 32 top dry wt; 11,
root dry wt
yield; 72, top dry wt; 59,
root dry wt
kernel dry wt; 20, top dry wt;
root dry wt
kernel dry wt; 48, top dry wt;
root dry wt
yi'eld"
foliar injury
foliar injury
root dry wt
top dry wt
root dry wt
top dry wt
top dry wt
top dry wt
biomass
top dry wt, harvest 1
top dry wt, harvest 2
top dry wt, harvest 3
top dry wt, harvest 1
top dry wt, harvest 2
top dry wt
needle mottle
white
4 weeks (mixture of 03
and S02 for same periods)
(over 2-3 days of exposure)
16, needle mottle
-------
TABLE 6-18 (cont'd). EFFECTS OF LONG-TERM, CONTROLLED OZONE EXPOSURES ON GROWTH, YIELD
AND FOLIAR INJURY IN SELECTED PLANTS3
cr>
i
o
co
Fig. 6-6b
Plant species nos.
Pine, ponderosa
Pine, ponderosa
Poplar, yellow
Maple, silver
Ash, white
Sycamore
Maple, sugar
Corn, sweet,
cultivar Golden
Midget0
p
Pine, ponderosa
Pine, western
white
Soybean, cultivar
Dare
Poplar, hybrid
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
Ozone
concentration,
ug/m3 (ppm)
290
290
290
290
588
588
588
(0.
(0.
(0.
(0.
(0.
(0.
(0.
880-588
588-880
588
588
588
588
98
196
196
196
98
196
290
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
15)
15)
15)
15)
30)
30)
30)
(0.30)
(0.45)
30)
30)
30)
30)
05)
10)
10)
10)
05)
10)
15)
Exposure time
9/day, 10 days
9/day, 20 days
9/day, 30 days
9/day, 60 days
9/day, 10 days
9/day, 20 days
9/day, 30 days
9/day, 30 days
9/day, 30 days
13 weeks
8/day, 5 days/week
13 weeks
8/day, 5 days/week
13 weeks
8/day, 5 days/week
13 weeks
8/day, 5 days/week
13 weeks
6/day, 64 days
6/day, 64 days
6/day, 126 days
6/day, 126 days
6/day, 133 days
6/day, 133 days
8/day, 5 days/week
6 weeks
4,
25
25
34
12
50
72
85
82
50
66
0
28
9
45
12
21
13
9
3
19
55
50
47
Plant response, % reduction
from control
photosynthesis
, photosynthesis
, photosynthesis
, photosynthesis
, photosynthesis
, photosynthesis
, photosynthesis
, photosynthesis
, leaf drop; 0, height
, leaf drop; 78, height
, leaf drop; 0, height
, leaf drop; 22, height
, leaf drop; 64, height
, kernel dry wt; 14, injury
(12, avg. 4 yield responses)
, 25, 35 for same responses
, root length
, stem dry wt; 26, root dry wt
, foil age dry wt
, stem dry wt
, seed yield; 22, plant fresh wtt;
, injury, defoliation, no reduc-
tion in growth or yield
, 65, 37 for same responses
, shoot dry wt; 56, leaf dry wt; ;
, root dry wt
Modified from National Research Council (1977); cited in U.S. Environmental Protection Agency (1978).
Numbers in this column are keyed to numbers in Fig. 6-6.
Studies conducted under field conditions, except that plants were enclosed to ensure controlled pollutant doses.
Plants grown under conditions making them more sensitive.
-------
experiments conducted under more controlled conditions (greenhouses, indoor
chambers, potted plants) are discussed primarily as they relate to the field
studies.
6.4.3.2.1.1 Open-top chamber studies. Each of the studies described in this
section used charcoal-filtered air as the lowest 0, concentration (control).
To create a range of concentrations, 0., was added to either charcoal-filtered
air or to air containing ambient levels of 03> To summarize the data, yield
loss was derived from the plant performance in charcoal-filtered air; this
approach contrasts with the approach used in the NCLAN studies (e.g., Heck
et al., 1982; Heck et al., 1983a) where yield loss was calculated from an
assumed natural 0- background level of 0.025 ppm. For this reason the yield
loss values in this section may differ from those reported in various NCLAN
studies even though the same exposure-response equations were used.
One of the major objectives of most of the studies cited in this section
was to develop exposure-response functions relating plant changes in plant
performance (yield) and 0, exposure (concentration and duration). To derive
the exposure-response functions, various linear and nonlinear curve-fitting
approaches have been used. The initial NCLAN studies used either linear or
plateau-linear functions to relate changes in plant yield to a 7-hr seasonal
mean 03 concentration (Heck et al., 1982). As discussed in greater detail in
Section 6.4.3.3, however, the linear curve frequently introduced a bias into
the data and should be used with caution. Subsequently, a Weibull function
(Rawlings and Cure, 1985) was used to develop exposure-response functions for
NCLAN data. The authors indicated that the Weibull was one of a series of
curvilinear functions that could have been used but that the Weibull had a
number of desirable properties and the curve fit the data well. The initial
use of the Weibull with NCLAN data (Heck et al., 1983a) was based on treatment
means for plant yield and 0, concentration. The more recent publications of
NCLAN data have used the individual chamber means for plant yield and 0,
exposure to derive the parameters of the Weibull function (e.g., Heck et al.,
1984a,b; Kress and Miller, 1985a). The use of treatment means rather than
plot means will alter variance estimates associated with the parameters as
well as the parameters themselves. To determine possible differences between
the two different methods of calculating the Weibull parameters, the concentra-
tions that would be predicted to cause a 10 percent yield loss were compared
for the same data set using both methods for several crops or cultivars. The
difference between 7-hr seasonal mean 0., concentrations predicted to cause
6-104
-------
10 percent yield loss, using plot and treatment means, ranged between 0.053 ppm
less to 0.011 ppm greater, with several values showing no difference between
the methods.
To estimate the impact of 0- on yield at a common 0_ concentration for
all the studies, the derived equations were used to estimate the yield loss at
a particular exposure condition rather than from the individual means. Graphs
of the exposure response equations and the data used to derive them are pre-
sented to show the goodness of fit of the Wei bull function to the data. The
data are grouped into legume, grain, fiber, and horticultural crops. The
parameters of the Weibull curves relating plant yield to the 7-hr seasonal
mean 0., concentration are listed in Table 6-19. The parameters for the Weibull
equations can be used to calculate the regression equation. The 0- concentra-
tion in charcoal-filtered air was used to calculate percentage yield loss.
The table also contains the calculated 7-hr seasonal mean concentrations that
are predicted to cause 10 and 30 percent yield reductions. The values were
selected to provide an indication of crop or cultivar sensitivity.
The impact of 0., on soybean yield has been investigated using nine culti-
vars grown at four different locations (Table 6-19, Figure 6-12). Each location
grew different cultivars and usually for only 1 year in developing the exposure-
response functions. The 7-hr seasonal mean 03 concentration that was predicted
to induce a 10 percent yield loss ranged from a low of 0.032 ppm for Hodgson
to a high of 0.076 ppm for Forrest, with a mean for the nine cultivars of
0.048 ppm. Two soybean cultivars, Davis and Williams, were studied for 2
successive years at the same location, permitting an estimation of year-to-year
variability in the impact of 0, on yield (Figure 6-13). At Raleigh, NC, a
7-hr seasonal mean concentration of 0.06 ppm was predicted to cause yield
losses in the cultivar Davis of 23.1 and 16.5 percent for the years 1981 and
1982, respectively. At Beltsville, MD, the predicted yield loss for the
cultivar Williams varied between 18.1 (1981) and 16.5 (1982) at a 7-hr seasonal
mean concentration of 0.06 ppm. The year-to-year difference in predicted
yield loss at the same concentration probably reflects differences in the
environmental conditions.
Aside from soybean, only two other legume crops, kidney bean and peanut,
have been studied (Table 6-19, Figure 6-14). Using the Weibull function, the
yield of peanut would be predicted to be reduced 10 percent at a 7~hr seasonal
6-105
-------
TABLE 6-19. ESTIMATES OF THE PARAMETERS FOR FITTING THE WEIBULL MODEL
USING THE 7-HR SEASONAL MEAN OZONE CONCENTRATIONS3'0
Crop
Parameters for Wei bull Model
t a c CFC
Concentration for
predicted yield
losses of:
30%d
LEGUME CROPS
Soybean, Corsoy 2785.00 0.133
Soybean, Davis (81) 5593.00 0.128
Soybean, Davis (CA-82)6 4931.00 0.128
Soybean, Davis (PA-82) 4805.00 0.103
Soybean, Essex (81) 4562.00 0.187
Soybean, Forrest (82-1) 4333.00 0.171
Soybean, Williams (81) 4992.00 0.211
Soybean, Williams (82-1) 5884.00 0.162
Soybean, Hodgson f 2590.00 0.138
Bean, Kidney (FP)T 2878.00 0.120
Peanut, NC-6 7485.00 0.111
GRAIN CROPS
1.952
0.872
2.144
4.077
1.543
2.752
1.
1.
1.
1.
100
577
000
171
2.249
0.022
0.025
0.019
0.019
0.014
0.017
0.014
0.017
0.017
0.019
0.025
0.048
0.038
0.048
0.059
0.048
0.076
0.039
0.045
0.032
0.033
0.046
0.082
0.071
0.081
0.081
0.099
0.118
0.093
0.088
0.066
0.063
0.073
Wheat, Abe (82)
Wheat, Arthur 71 (82)
Wheat, Roland
Wheat, Vona
Wheat, Blueboy II (T)
Wheat, Coker 47-27 (T)
Wheat, Holly (T)
Wheat, Oasis (T)
Corn, PAG 397
Corn, Pioneer 3780
Corn, Coker 16 (T)
Sorghum, DeKalb-28
Barley, Poco
FIBER CROPS
Cotton, Acala SJ-2 (81-1)
Cotton, Acala SJ-2 (82-1)
Cotton, Stoneville
HORTICULTURAL CROPS
,88
,19
95
5363.00
4684.00
5479.00
7857.00
5.
5.
4.
4.48
13968.00
12533.00
240.00
8137.00
1.99
5546.00
5872.00
3686.00
0.143
0.148
0.113
0.053
0.175
0.171
0.156
0.186
0.160
0.155
0.221
0.296
0.205
0.199
0.088
0.112
2.
2.
1.
1.
3.
2.
4.
3.
4.
3.
4.
2.
423
154
633
000
220
060
950
200
280
091
460
217
4.278
1.228
2.100
2.577
0.023
0.023
0.023
0.022
0.030
0.030
0.030
0.030
0.015
0.015
0.020
0.016
0.020
0.018
0.012
0.026
0.059
0.056
0.039
0.028
0.088
0.064
0.099
0.093
0.095
0.075
0.133
0.108
0.121
0.044
0.032
0.047
0.095
0.094
0.067
0.041
0.127
0.107
0.127
0.135
0.126
0.111
0.175
0.186
0.161
0.096
0.055
0.075
Tomato, Murrieta (81)
Tomato, Murrieta (82)
Lettuce, Empire (T)
Spinach, America (T)
Spinach, Hybrid (T)
Spinach, Viroflay (T)
32.90
32.30
1245.00
21.20
36.60
41.10
0.142
0.082
0.098
0.142
0.139
0.129
3.807
3.050
1.220
1.650
2.680
1.990
0.012
0.012
0.043
0.024
0.024
0.024
0.079
0.040
0.053
0.046
0.043
0.048
0.108
0.059
0.075
0.082
0.082
0.080
6-106
-------
TABLE 6-19 (cont'd). ESTIMATES OF THE PARAMETERS FOR FITTING THE
USING THE 7-HR SEASONAL MEAN OZONE CONCENTRATIONS3>D
Crop
Parameters for Weibull Model
a a c CFC
Concentration for
predicted yield
losses of:
30%d
Spinach, Winter Bloom (T)
Turnip, Just Right (T)
Turnip, Pur Top W.G. (T)
Turnip, Shogoin (T)
Turnip, Tokyo Cross (T)
20.80
10.89
6.22
4.68
15.25
0.127
0.090
0.095
0.096
0.094
2.070
3.050
2.510
2.120
3.940
0.024
0.014
0.014
0.014
0.014
0.049
0.043
0.040
0.036
0.053
0.080
0.064
0.064
0.060
0.072
Data are from Heck et al. (1984b) and are based on individual plot mean:; unless the
crop name is followed by "(T)". The "T" indicates that the parameters were based
on treatment means and the data are from Heck et al. (1983a). The parameters given
in Heck et al. (1983a, 1984b) also contain the standard errors of the parameters.
3A11 estimates of a are in ppm. The yield is expressed as kg/Ha for all crops
except barley-seed weight (g per head); tomato (both years)--fresh weight
(kg per plot); cotton--!int + seed weight (kg/Ha); peanut—pod weight (kg/Ha).
In cases where the estimated c parameter is exactly 1.0, it has been bounded
from below to obtain convergence in the nonlinear model fitting routine.
Parameters were estimated from data not showing the expected Weibull form.
Caution should be used in interpreting these Weibull models. Other mode's might
better describe the behavior observed in these experiments. For those crops
whose name is followed by "(T)11 the yield is expressed as g/plant.
"The 03 concentration in the charcoal filtered chambers expressed as a 7-hr seasonal
mean concentration.
rj
The 7-hr seasonal mean 03 concentration (ppm) that was predicted to cause a 10
or 30 percent yield loss (compared to charcoal-filtered air).
CA and PA refer to constant and proportional 03 addition.
Only the bean data from the full plots are shown. The partial plot data are
given in Heck et al. (1984b).
6-107
-------
5000
4000
x
O)
0
ul 3000
2000
1000
(A)
SOYBEAN (HODGESONJ
ITHACA, 1981
5000
4000
X
Ol
Jf
Q
ui 3000
O
LU
LU
2000
1000
(B)
SOYBEAN (FORREST)
BELTSVILLE. MO. 1982
2.752
0 0.02 0.04 0.06 0.080.1 0.12 0.14
03 CONCENTRATION, ppm
0 0.020.040.060.080.10.120.14
03 CONCENTRATION, ppm
5000
4000
ra
.c
x
O)
Q
iff 3000
01
(0
2000
1000
(C)
SOYBEAN (CORSOY)
ARGONNE. 1980
, = 2785-«V°133>
I I I
5000
4000
IB
.C
X
re
Q
iJJ 3000
LLJ
tf)
2000
1000
(D)
SOYBEAN (DAVIS)
RALEIGH. 1981
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
0 0.02 0.040.060.08 0.10.120.14
O3 CONCENTRATION, ppm 0, CONCENTRATION, ppm
Figure 6-12. Effects of ozone on the yield of four soybean cultivars. The cultivars were selected to show
the response of one cultivar at each of the four locations where studies were conducted. The 03 concen-
trations are expressed as the 7-hr seasonal means. (A) Each point represents the mean of two undisturbed
full plots. The regression equation was based on individual chamber ozone and yield values. The Weibull
curve is from Heck et al. (1984b). Note that c = 1.0 and the model was forced to converge. This means
that "parameters were estimated from data not showing the expected Weibull form" (Heck et al. 1984b).
Another curve for g/plant is in Heck et al. (1982). (B) The regression equation was based on individual
chamber ozone and yield values. The Weibull curve is from Heck et al. (1984b). (C) The 03 monitoring
period for the seasonal mean is 9 days shorter than the 03 exposure period. Each point represents the
mean of four chambers. The data are from Kress and Miller (1983). Yields expressed in g/plant can also
be found in Heck et al. (1983) and Heck et al. (1982). The regression was based on yield and ozone values
for individual chambers. The Weibull equation is from Heck et al. (1984b). Another curve for yield in
g/plant is in Heck et al. (1983). (D) Each point represents the mean of two chambers. Data in g/meter
of row can be found in Heagle et al. (1983) and in Heck et al. (1982). To convert from g/m3 to kg/ha,
multiply by 10.1. The regression was based on yield and ozone values for individual chambers. The Wei-
bull equation is from Heck et al. (1984b). Another Weibull curve for g/plant is given in Heck et al. (1983).
6-108
-------
6000
5000
4000
IB
Ł
W
Jt
Q
3000
Q
ID
UJ
V)
2000
1000
(A)
SOYBEAN (DAVIS)
RALEIGH. 1981 AND 1982
1982(4)
y = 4931-<°3/0-128>
,2.144
1981 (O)
v = 5593-<°3/0128>'
0.872
6000
5000
o> 4000
Q
tU 3000
w
2000
LiHBIH^_^J-WMB_J*M^^ML^^^^^^BWM
0 0.020.040.060.080.10.120.14
O3 CONCENTRATION, ppm
1000
-|03/0.162)1'577
1981 (O)
y = 4992-(°3/0.211)1
SOYBEAN (WILLIAMS)
BELTSVILLE, MD, 1981 AND 1982
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
O3 CONCENTRATION, ppm
Figure 6-13. Comparison of the effects of ozone on soybean yields on the same
cultivars exposed for successive years at two locations. The 03 concentrations are
expressed as the 7-hr seasonal means. (A) The data for Davis (1981) are described in
the caption to Figure 6-12. For Davis (1982). each yield value is the mean of two
chambers. The Weibull equations are from Heck et al. (1984b). (B) In 1981 the points
at 0.01 and 0.035 ppm O3 were the means of three chambers. The other points
shown are values for one chamber only. These data are taken from a factorial
experiment with six S02 and four O3 levels. Only the control SO2 level is shown here.
The regression was based on yield and ozone values for individual chambers. The
Weibull curves are from Heck et al. (1984b). Other curves for g/plant are given in
Heck et al. (1983a). In 1982, the data came from a factorial experiment with three
S02> five O3, and two moisture levels. Only the control S02 and moisture levels were
used in the figure. The regression equations were based on individual chamber ozone
and yield values. The Weibull curve is from Heck et al. (1984b).
6-109
-------
I
O)
JC
Q
Ł
I
8000
7000
6000
5000
4000
3000
2000
1000
PEANUT (NC-6)
RALEIGH. 1980
, = 7485-«V°-111>2249
I I
0 0.02 0.04 0.060.08 0.1 0.12 0.14
03 CONCENTRATION, ppm
2800
2700
2600
2500
2400
2300
« 2200
_g>2100
2000
2 1900
UJ
(A 1800
1700
1600
1500
1400
1300
1200
KIDNEY BEAN (CALIF LIGHT RED)
ITHACA. 1982
= 2878-«V0.12)
I I I I l\-> I
1.171
0 0.02 0.040.060.08 0.10.120.14
03 CONCENTRATION, ppm
Figure 6-14. Effects of ozone on the yield of peanut and kidney bean. The 03
concentrations are expressed as the 7-hr seasonal means. (A) Each point represents
the mean of four chambers. Yields were multiplied by 50.5 to convert from g/plant to
kg/ha. The regression was based on yield and ozone values for individual chambers.
The Weibull equation is from Heck et al. (1984b). (B) Each point represents the mean
of two undisturbed full plots. The regression was based on individual chamber values
for yield and ozone. The Weibull equation is from Heck et al. (1984b).
6-110
-------
mean concentration of 0.046, which was similar to the mean response of the 9
soybean cultivars. Kidney bean appears to be similar in 03 sensitivity to
peanut or most of the soybean cultivars. Kidney bean yield was predicted to
be reduced 10 percent at a 7-hr seasonal mean concentration of 0.033. When
the data were analyzed using linear regression analysis (Kohut and Laurence,
1983) a 7 percent reduction in yield was predicted to occur at a 7-hr seasonal
mean concentration of 0.06 ppm.
Winter wheat yield appeared to be relatively less sensitive to 0~ than
the legumes based on the yield reductions of four cultivars (Table 6-20). The
yield of all four cultivars was significantly reduced (11 to 25 percent) at
0.10 ppm (7-hr seasonal mean), but only one cultivar was significantly affec-
ted (11 percent) at 0.06 ppm (Heagle et al., 1979c). These data have subse-
quently been re-evaluated using quadratic (Heagle and Heck, 1980), linear
(Heck et al., 1982), and Weibull functions (Heck et al., 1983a). Based on
visual inspection of the data, it appears that the curvilinear models fit the
data better than the linear one. In addition to these four cultivars, which
were studied at Raleigh, NC, one wheat cultivar was studied at Ithaca, NY (for
1 year) and three cultivars were studied at Argonne, IL (two cultivars for
2 years). Examples of the relationship between wheat yield and 0, concentra-
tions are shown for cultivars at three different locations (Figure 6-15). The
7-hr seasonal mean 0- concentration that was predicted to induce a 10 percent
yield loss ranged from a low of 0.028 ppm for Vona (Ithaca) to a high of 0.099
ppm for Holly (Raleigh), with a mean for the eight cultivars of 0.068 ppm.
The winter wheat cultivars displayed a greater range in the 7-hr seasonal
means predicted to cause 10 and 30 percent yield reductions than did the
soybean cultivars. Using the 7-hr seasonal mean predicted to cause a 10
percent yield reduction, wheat cultivars such as Vona and Roland (Argonne)
appeared to be similar in sensitivity to soybean. The other wheat cultivars
appeared to be relatively tolerant of 0~. Two wheat cultivars, Abe and
Arthur-71, were studied for 2 successive years at Argonne, IL (Kress et al.,
1985, Figure 6-16), permitting an estimation of year-to-year variability in
the effect of 0- on yield. For Abe, a 7-hr seasonal mean concentration of
0.06 ppm was predicted to cause yield losses of 10 and 0 percent for the years
1982 and 1983, respectively. The predicted yield loss for the cultivar
Arthur-71 varied between 8.8 (1982) and 3.2 (1983) percent at a 7-hr seasonal
mean concentration of 0.06 ppm. For the cultivar Abe, the shape of the regres-
sion equation varied between years, but for Arthur-71 it was similar between
6-111
-------
TABLE 6-20. EFFECTS OF OZONE ADDED TO AMBIENT AIR IN OPEN-TOP CHAMBERS ON THE YIELD OF SELECTED CROPS
Plant species
Field corn
(Coker 16)
Field corn
(Coker 16)
(FR632 X FR619)
(H95 X FR64A)
CT>
i Winter wheat
•- (soft red)
Ł (Blueboy II)
(Coker 47-27)
(Holly)
(Oasis)
03 concn. ,
ppm
0.02
0.07
0.11
0.15
0.02
0.15
0.02
0.15
0.02
0.15
0.03
0.06
0.10
0.13
0.03
0.06
0.10
0.13
0.03
0.06
0.10
0.13
0.03
0.06
0.10
0.13
Yield, % reduction Monitoring
Exposure duration from control method
Beginning 25 days after planting for 88 Control Chem.
days, Seasonal 7-hr average (0830-1530 ST) +3, seed wt/plant; +2, wt/seed
4., seed wt/plant; l,rtwt/seed
16 , seed wt/plant; 9 , wt/seed
Beginning 25 days after planting for 88 Control rf Chem.
days, Seasonal 7-hr average (0830-1530 ST) 12 , seed wt/plant; 15 , wt/seed
Control rt
37°, seed wt/plant; 25°, wt/seed
Control rf
40 , seed wt/plant; 30 , wt/seed
Beginning when plants were 28 to 45 cm Control Chem.
tall for 53 days 2, seed wt/plant
Seasonal 7-hr average (0930-1530 ST) 15;!, seed wt/plant
31 , seed wt/plant
Control
11°, seed wt/plant
25., seed wt/plant
43 , seed wt/plant
Control
1, seed wt/plant
11° seed wt/plant
33°, seed wt/plant
Control
1, seed wt/plant
11°., seed wt/plant
26 , seed wt/plant
Calibration
method Reference
1% NBKI Heagle et
al. (1979a)
1% NBKI Heagle et
al. (1979a)
1% NBKI Heagle et
al. (1979c)
-------
TABLE 6-20 (cont'd). EFFECTS OF OZONE ADDED TO AMBIENT AIR IN OPEN-TOP CHAMBERS ON THE YIELD OF SELECTED CROPS3
cr>
i
to
Plant species
Spinach
(America)
(Winter Bloomsdale)
(Hybrid 7)
(Viroflay)
03 concn. ,
ppm
0.024
0.056
0.096
0.129
0.024
0.056
0.096
0.129
0.024
0.056
0.096
0.129
0.024
0.056
0.096
0.129
Yield, % reduction
Exposure duration from control
Beginning 10 days after planting for 38 Control
days, Seasonal 7-hr average (0820-1520 ST) 23.. fresh wt of shoots
39°, fresh wt of shoots
70 , fresh wt of shoots
Control
19fl fresh wt of shoots
44 ., fresh wt of shoots
73 , fresh wt of shoots
Control
4, fresh wt of shoots
35 ., fresh wt of shoots
61 , fresh wt of shoots
Control
26.. fresh wt of shoots
35° fresh wt of shoots
72 , fresh wt of shoots
Monitoring Calibration
method method Reference
Chem. 1% NBKI Heagle et
al. (1979b)
Where a column entry is blank the information is the same as above.
Chem = chemiluminescence.
C13> NBKI = 1% neutral buffered potassium iodide.
dSignificant effect at p = 0.05.
-------
6000
6000
4000
a
Ł
x
5
cj
UJ
3000
Q
UJ
UJ
M
2000
1000
/ AI WHEAT (VONA)
' ITHACA. 1982
= 7867-(°3/°-053>1
5.5
5.0
Ł 4.5
N
n
34.0
ui
Q
UJ
3.0
2.5
(B)
WHEAT (HOLLY)
RALEIGH. 1977
0 0.020.040.060.080.10.120.14
03 CONCENTRATION, ppm
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
03 CONCENTRATION, ppm
6.0
5.5
5.0
.
u>
d
UJ
> 4.0
Q
ui
ui
CO
3.5
3.0
2.5
(C)
WHEAT (BLUEBOV II)
RALEIGH. 1977
6000
5000
Ł>4000
3
ui
CO
3000
2000
1000
(D)
WHEAT (ROLAND)
ARGONNE. 1982
, = 6479-l°3/0.113)1633
0 0.02 0.04 0.06 0.080.1 0.120.14
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
03 CONCENTRATION, ppm O3 CONCENTRATION, ppm
Figure 6-15. Effects of ozone on the yield of four wheat cultivars. The cultivars were
selected to show the response at several locations. The O3 concentrations are
expressed as the 7-hr seasonal means. (A) Each point represents the mean of four
chambers. The regressions were based on individual chamber ozone and yield values.
The Weibull equation is from Heck et al. (1984b). Note that c = 1.0 and the model was
forced to converge. "The parameters were estimated from data not showing the
expected Weibull form" (Heck et al.. 1984b). (B-C) Each point represents the mean of
four chambers. Data are from Heagle et al. (1979c). The regressions were based on
mean O3 concentrations for each treatment and individual chamber yields. The
Weibull equations are from Heck et al. (1983). Note that the yields for these figures
are in different units from A through D. (D) Each point represents a mean of four
chambers adjusted for number of heads per meter by analysis of covariance. The data
are from Kress et al. (1985) and Heck et al. (1983). The regressions were based on
individual chamber ozone and yield values. The Weibull equations are from Kress et al.
(1985); other equations can be found in Heck et al. (1984).
6-114
-------
6000
5000
3 4000
o
UJ
Q
tij 3000
(A
2000
1000
1982Q
y = 5235-(°3/0.153)2-272
- 1983(A)
= 5873-<°3/0108>144
WHEAT (ABE)
ARGONNE. 1982 AND 1983
I I I I
6000
5000
CO
w 4000
Q
UJ
Q
uj 3000
2000
1000
WHEAT (ARTHUR 71)
(B) A ARGONNE. 1982 AND 1983
i- 1982Q
y = 4513-«V°-1«6)25i8
1983(A)
I
0 0.020.040.060.08 0.1 0.12 0.14
0 0.02 0.04 0.060.08 0.1 0.12 0.14
03 CONCENTRATION, ppm 03 CONCENTRATION, ppm
Figure 6-16. Comparison of the effects of ozone on yields of the same wheat cultivars
exposed for successive years at Argonne, IL The O3 concentrations are expressed as
the 7-hr seasonal means. For 1982 each point on the graph represents the mean of
four chambers, and for 1983 each point represents the mean of three chambers. In
both years the means were adjusted to a common number of heads per imeter by
analysis of covariance. The data are from Kress et al. (1985). The regressions were
based on individual chamber ozone and yield values. The Weibull equations are from
Kress et al. (1985). For 1983 the alpha estimate has been corrected to 5210 (personal
communication from L. W. Kress, Argonne National Laboratory, to D. T. Tingey, U.S.
EPA, 1985). Other Weibull equations for the 1982 data are found in Heck et al.
(1984b).
6-115
-------
years, probably contributing to lower year-to-year variability. The year-to-year
difference in predicted yield loss at the same concentration probably reflects
the influence of differences in the environmental conditions on plant response
to 03.
The effects of (L on field corn have received less study than winter
wheat. The impact of 03 on the yield of field corn was initially studied with
Coker 16; the results of this study have been analyzed by several different
methods and published in several different forms. The data were initially
presented in tabular form using mean comparison tests (Heagle et al., 1979a;
Table 6-20) and subsequently analyzed using quadratic (Heagle and Heck, 1980)
and linear (Heck et al., 1982) regression models. Reductions in seed yield
(g/plant) were originally shown to be 4 percent at a 7-hr seasonal mean concen-
tration of 0.11 ppm and 16 percent at 0.15 ppm 03 when compared to a 0.020 ppm
control (Table 6-20). The quadratic regression predicted a yield increase of
1 percent at 0.06 ppm and a yield reduction of 3 percent at a 7-hr seasonal
mean concentration of 0.10 ppm. The linear equation showed significant lack
of fit to the data; therefore it was not considered. Using the Weibull para-
meters for Coker 16 (Table 6-19, Figure 6-17), a 10 percent yield loss was
predicted to occur at a 7-hr seasonal mean concentration of 0.133 ppm. The
impact of ozone on two midwestern corn cultivars has also been studied (Table
6-19); these cultivars appeared to be more sensitive to 0, than Coker 16.
Using the Wei bull parameters, yield reductions of 10 percent were predicted to
occur at 7-hr seasonal mean concentrations of 0.075 and 0.095 ppm for Pioneer
3780 and PAG 397, respectively. Kress and Miller (1985b) subsequently
analyzed the data for Pioneer 3780 (Figure 6-17B) and PAG 397 using Weibull,
quadratic, and plateau-linear models and found no statistical difference among
them.
Sorghum (Figure 6-17C) was approximately as sensitive to 0- as field corn
(Table 6-19); the Weibull function predicted a 10 percent yield loss at a 7-hr
seasonal mean concentration of 0.108 ppm. Quadratic, Weibull, and plateau-
linear models all adequately described the response of grain sorghum to 03
(Kress and Miller, 1985a).
Poco barley (Figure 6-17D) (Poco) was as tolerant of 0- as the more
tolerant corn cultivar (Coker 16); using the Weibull function, a 10 percent
yield loss was predicted to occur at a 7-hr seasonal mean 03 concentration of
0.121 ppm (Table 6-19). Temple et al. (1985b) subsequently showed that the
6-116
-------
250
240
230
Z
oc
* 220
U.
O
210
200
(A)
CORN(COKER 16|
RALEIGH. 1976
s
15000
14000
13000
12000
11000
10000
O 9000
ui
ui
W 8000
7000
6000
5000
4000
(B)
CORN (PIONEER 3780)
ARGONNE, 1981
= 12633-«V°-1BB>3°91
I I I I
0.02 0.04 0.060.08 0.1 0.12 0.140.16
O, CONCENTRATION, ppm
0 0.020.040.060.080.1 0.120.140.16
O3 CONCENTRATION, ppm
8300
8200
8100
8000
7900
| 7800
o>
•*. 7700
O
S 7600
Z 7500
O 7400
7300
7200
7100
7000
6900
O
(C)
SORGHUM (DEKALB-28)
ARSONNE, 1982
y = 8137-«V0.296)2217
I
I
I
I
I
2.2
2.1
a>
D
ui
1.9
1.8
(D)
BARLEY (POCO)
SHARER. CA. 1982
y = 1.988-<°>/0206>4278
I
I
I
I
I
I
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
03 CONCENTRATION, ppm
0 0.02 0.04 0.060.08 0.1 0.12 0.14 0.16
O3 CONCENTRATION, ppm
Figure 6-17. Effects of ozone on the yield of corn (two cultivars), sorghum and barley.
The O3 concentrations are expressed as the 7-hr seasonal means. (A) Each point
represents the mean of five chambers. The regression equation was based on
individual chamber yield values and ozone treatment means. Data are from Heagle et
al. (1979a). The Weibull equation is from Heck et al. (1983a). (B) Each point
represents the mean of three chambers; data are from Kress and Miller (1985b). The
Weibull equation is from Heck et al. (1984b) and the regression was done on
individual chamber values for yield and ozone concentration. Another set of curves
can be found in Kress and Miller (1985b), and a set of curves for yield in g/plant are in
Heck et al. (1983a). (C) Each point represents the mean of three chambers. The
regression was based on yield and ozone values for individual chambers. The data can
be found in Kress and Miller (1985s), and the Weibull equation is from Heck et al.
(19845). Another curve can be found in Kress and Miller (1985a). (D) Each yield value
represents the mean of four chambers. The regression was based on yield and ozone
values for individual chambers. The Weibull equation is from Heck et al. (1984b).
6-117
-------
yield of two barley cultivars was not reduced by ambient 03 in the Central
Valley of California. Even at twice ambient concentrations the yield was not
affected. The twice-ambient concentration 0.094 (7-hr seasonal mean) was
substantially lower than the concentration (0.121 ppm) predicted by the Weibull
to reduce yield by 10 percent.
A 2-year field study was conducted in the Central Valley of California to
determine the impact of 0^ on the yield of cotton. Using the Weibull parameters
(Table 6-19, Figure 6-18A), a 10 percent yield reduction was predicted to
occur at 7-hr seasonal mean 0- concentrations of 0.044 and 0.032 ppm in 1981
and 1982, respectively. Based on these data it appears that cotton is approxi-
mately as sensitive to 0- as soybean or peanut. Based on the Weibull function
for Acala SJ-2 cotton, a 7-hr seasonal mean concentration of 0.06 ppm was
predicted to cause yield losses of 16.2 and 35.1 percent for the years 1981
and 1982, respectively. Using other exposure-response functions published by
Temple et al. (1985a), the year-to-year variation in yield loss was 18 and 27
percent for the years 1981 and 1982, respectively. The cotton data set illus-
trates two different types of variation that may occur. Differences in yield
loss vary between years at the same 7-hr seasonal mean 0~ concentration. The
authors (Temple et al., 1985a) showed that there were significant climatic
differences between the two years that contributed to the variation. Also,
the study shows that different exposure-response functions may also yield
different predicted yield loss estimates. For cotton grown in the South-
eastern U.S. (Raleigh, NC) a 10 percent yield reduction was predicted to occur
at a 7-hr seasonal mean of 0.047 ppm (Figure 6-18B). Based on these data it
appears that cotton is approximately as sensitive to 0, as soybean or peanut.
The yield of tomato also showed large year-to-year variation in sensi-
tivity to 0, (Table 6-19, Figure 6-19A). Based on the Weibull parameters, a
10 percent reduction in yield was predicted to occur at 7-hr seasonal mean 0~
concentrations of 0.079 and 0.040 ppm in 1981 and 1982, respectively. The
yield of lettuce was predicted to be reduced 10 percent at a seasonal mean
concentration of 0.053 ppm (Table 6-19, Figure 6-19B).
The effects of 03 on the yield (weight) of four spinach cultivars have
been studied (Table 6-20). All four cultivars exhibited significant yield
reductions (35 to 44 percent) when exposed to 0.096 ppm 03 (7-hr seasonal
mean), compared to a control of 0.024 ppm (Heagle et al., 1979b). Nonsignifi-
cant yield reductions of 4 to 26 percent were noted at 0.056 ppm 0.,. All the
6-118
-------
6000
5500
5000
*
\
4SOO
w
Q
. 4000
3500
3000
2500
2000
1500
COTTON (SJ-2)
SHAFTER. CA, 1981 AND 1982
I 1 I I
I I
I
4000
3500
x
9
Q
ui
in
CA
Q
Z
3000
2500
2000
1500
(B)
COTTON (STONEVILLE 213)
RALEIGH, 1982
,2.577
r = 3686-(°J/0112>
I I I I
I
I
0 0,02 0.040,060.08 0.1 0.12 0,14 0.180.180.2
03 CONCENTRATION, ppm
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
03 CONCENTRATION, ppm
Figure 6-18. Effects of ozone on the yield of two cotton cultivars grown at two
locations. The O3 concentrations are expressed as 7-hr seasonal means. (A) Each
point represents the mean of two chambers. There were six 03 levels and two
irrigation levels in this experiment. Only chambers with normal irrigation were
included in this figure. The regression was based on yield and ozone values for
individual chambers. The Weibull equation is from Heck et al. (1984b). In 1982, the
monitoring period used to calculate the 7-hr seasonal mean was 41 days longer than
the period that O3 was added experimentally to the chambers. (B) Each point
represents the mean of two chambers. The regression was based on individual
chamber values for yields and ozone. The Weibull equation is from Heck et al.
(1984b).
6-119
-------
34
33
32
31
_ 30
o
"5.
» 29
> 28
X
(A
UJ
cc 27
26
25
24
23
TOMATO (MURIETTA)
O TRACY. CA. 1981 AND 1982
1981(Q)
y = 329-(0,/0.142)3807
1982 (A)
= 32.3-«V°082>305
900
800
700
I 600
a
\
O)
9
uj 500
400
300
200
LETTUCE (EMPIRE)
RIVERSIDE, 1980
v =
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
03 CONCENTRATION, ppm
0 0.020.040.060.080.1 0.120.140.16
O3 CONCENTRATION, ppm
Figure 6-19. Effects of ozone on the yield of tomatoes and lettuce. The O3
concentrations are expressed as the 7-hr seasonal means. (A) The yields at 0.012
(1981) and 0.011 ppm (1982) were the means of two chambers; all other yield
values were from one chamber only. The regression was based on individual
chamber yield and ozone values. The Weibull curve is from Heck et al. (1984b).
(B) Each point represents the mean of four chambers. Data are from Heck et al.
(1982). The regression is based on mean O3 concentrations for each treatment
and individual yield values for each chamber. The Weibull equation is from Heck
etal. (1983a).
6-120
-------
cultivars, except Viroflay, displayed some foliar injury that could impair its
intended use even though the weight of the plants was not reduced. The same
data were subjected to reanalysis using linear (Heck et al., 1982) and Weibull
(Heck et al., 1982, Figure 6-20A.B) functions. Using the Weibull parameters,
a 10 percent reduction in yield was predicted to occur at 7-hr seasonal mean
concentrations of 0.043 to 0.049 ppm. This range in predicted concentrations
is smaller than that observed in other crops such as soybean or wheat.
The effects of 0- on the yield (weight) of four turnip cultivars has been
studied. One cultivar, Tokyo Cross, was studied during 2 years (Heck et
al., 1982; Heagle et al., 1985, Figure 6-20C.D). The data were analyzed with
four models, linear, quadratic, plateau-linear, and Weibull; and statistical
tests for lack of fit were performed. The plateau-linear model showed signi-
ficant lack of fit with one cultivar and the other three showed significant
lack of fit with two cultivars each. Based on the Weibull parameters, the
7-hr seasonal mean 0, concentration that would be predicted to cause a 10
percent yield reduction ranged from a low of 0.036 (Shogoin) to a high of
0.053 ppm (Tokyo Cross). Part of the yield loss was attributed to an acute
injury episode from a low concentration of 0- after a period of dark, cool,
rainy weather (Heagle et al., 1985). These data illustrate one of the problems
with the 7-hr seasonal mean concentration as a statistic for adequately charac-
terizing Oo exposure and the resultant effects. The cultivar Tokyo Cross was
grown 2 years, permitting an illustration of year-to-year variation in the
predicted yield response. Based on the Weibull function, a seasonal mean
concentration of 0.06 ppm was predicted to cause yield reductions of 27.2 and
15.6 percent for the years 1979 and 1980, respectively.
6.4.3.2.1.2. Use of chemical protectants to estimate yield loss. Chemical
protectants have several advantages for estimating yield loss. The crops are
not grown in chambers but are grown under field conditions using standard
cultural practices; and are exposed to ambient environmental conditions. It
is also possible to have several replications in a field and to have a number
of fields included in the study area, and the plants are exposed to the temporal
variations in pollutant concentrations occurring at the location(s). With
chemical protectants, however, only a single pollutant treatment is possible
at a location and this prevents the development of exposure-response functions
as can be done with exposure chambers. When chemical protectants are used to
estimate yield loss, care must be exercised in interpreting the data because
6-121
-------
45
40
35
30
X 25
c/)
ui
K
u.
5 20
O
(0
15
10
M
III
K
O
o
cc
(A)
SPINACH (VIROFLAY)
RALEIGH. 1976
, = 41.4-<°3/0129>1"
I
I
0.02 0.040.060.08 0.1 0.12 0.140.16
03 CONCENTRATION, ppm
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
TURNIP (SHOGOIN)
RALEIGH. 1980
i
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
O, CONCENTRATION, ppm
o>
45
40
35
30
X 25
w
O
O
I
in
20
15
10
(B)
SPINACH (HYBRID 7)
RALEIGH. 1976
,,2.68
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
03 CONCENTRATION, ppm
16
15
14
13
I
O)
M
O
oc
TURNIP (TOKYO CROSS)
RALEIGH, 1979 AND 19BO
1980(A)
= 15.25W0.094)394
1979(0)
y = 4.05-'(V0.086)3
I
I
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
O3 CONCENTRATION, ppm
Figure 6-20. Effects of ozone on the yield of spinach and turnip cultivars. The O3
concentrations are expressed as 7-hr seasonal means. (A-B) Each point represents the
mean of three chambers. Data are from Heagle et al. (1979b). The regressions were
based on mean O3 concentrations for each treatment and individual yield values for
each chamber. The Weibull equations are from Heck et al. (1983a). (C-D) Each point
represents the mean of four chambers except for Tokyo Cross, 1979, which
represents the mean of two chambers. The regressions were based on mean 03
concentrations for each treatment and individual yield values for each chamber. The
data and the Weibull equations are from Heagle et al. (1985). The same data can be
found in Heck etal. (1982).
6-122
-------
the chemical protectant, in the absence of the pollutant, may alter plant
growth. In addition, the chemical protectant may not be effective against all
the phytotoxic chemicals that may be present in the environment, with the
consequence that the resultant yield loss would be underestimated. An under-
estimation would also occur if the protectant chemical did not prevent all the
impacts of the pollutant at all ambient concentrations. Even with these
limitations, however, researchers have concluded that antioxidant chemical
protectants provide a readily available objective method .for assessing the
effects of ozone on crop productivity over a range.of conditions (Toivonen et
al., 1982).
The observation that various agricultural chemicals reduce or prevent
visible ozone injury (see Section 6.3.2.3.2) leads to their use as a means of
estimating the impact of ambient ozone on crop production. Field studies with
the systemic fungicide, benomyl, showed that it reduced foliar injury 75 to 80
percent in ozone-sensitive bean cultivars but had no effect on the slight
injury of the more tolerant cultivar (Manning et al., 1974). Measured for the
duration of the study (June through mid-September, about 2540 hr), ozone
equaled or exceeded 0.04 ppm for 351 hr (about 14 percent of total hours)
during a 40-day period. The concentration equaled or exceeded 0.08 ppm for 27
hr (about 1 percent of total hours). Treatment with benomyl increased the
yield of the ozone-sensitive tempo bean by 41 percent; there was no statistical
difference in the yield of tenderwhite, an ozone-resistant cultivar, exposed
with or without fungicide treatment.
Ethylenediurea (EDU), an experimental chemical not commercially available,
was developed specifically as a chemical protectant for ozone (Carnahan et
al., 1978). It has been used extensively for reducing visible ozone injury in
greenhouse and field studies (see Section 6.3.2.3.2), as well as for estimating
ozone-induced yield losses. To estimate ozone-induced yield loss, some plots
are treated with EDU and others are not; both types of plots experience the
same environmental conditions and ozone exposure. The higher yield in plots
treated with EDU is thought to represent the yield that would occur in the
absence of ozone, and the yield from the plots not treated with EDU represents
the yield in the presence of ozone. Consequently, the impact of ozone on
yield is determined by comparing the yield data from plots with and without
EDU.
6-123
-------
In onions, EDU treatment has been shown to reduce foliar injury and
increase plant yield 37.8 percent (Wukasch and Hofstra, 1977a,b). During the
study period, July 7 to August 25, the ozone concentration exceeded 0.15 and
0.08 ppm on 5 separate days at each concentration. During a tomato study
(June through August), the ozone concentration exceeded 0.08 ppm on 15 days,
reaching a maximum of 0.14 ppm (Legassicke and Ormrod, 1981). Treatment with
EDU increased the yield of Tiny Tim tomato about 30 percent but had no effect
on the New Yorker cultivar.
During 1977 and 1978, EDU was used to estimate the impact of ambient
ozone on the yield of three white bean cultivars (Toivonen et al., 1982).
Substantial foliar ozone symptoms were observed, and ozone-induced yield
losses reached 35 percent in 1977. During 1978, a mid-season drought occurred,
resulting in less visible injury, and yield losses reached only 19 percent
even though the ozone levels were higher than in 1977. Another study with
white bean showed that EDU treatment increased crop yield approximately 24
percent and delayed defoliation (Temple and Bisessar, 1979). The daylight
average ozone concentrations (0600 to 2000 hours) were 0.042, 0.042, and 0.028
ppm for June, July, and August, respectively. During these months, the ozone
concentration exceeded 0.08 ppm on 11 days for a total of 34 hr.
At a tobacco study plot in southern Ontario during the summer of 1982,
the ozone concentrations exceeded 0.08 ppm 14 times, with a maximum of 0.126
ppm (Bisessar and Palmer, 1984). Treatment with EDU reduced foliar injury and
increased above-ground plant biomass by about 18 percent.
Greenhouse studies with potato showed that EDU treatment reduced foliar
ozone injury and increased the yield of an ozone-sensitive potato cultivar
(Foster et al., 1983a). More important, the studies showed that in the absence
of ozone EDU had no significant effect oh plant growth and yield, indicating
that estimates of ozone-induced yield losses are not confounded, at least for
potato, by effects of the chemical directly on yield. In southern Ontario,
EDU increased the yields of Norchip potato 35.5 percent (Bisessar, 1982). The
daylight average ozone concentrations (0600 to 2000 hours) were 0.040, 0.044,
and 0.027 ppm for June, July, and August, respectively. During these months,
the ozone concentration exceeded 0.08 ppm on 18 days, for a total of 68 hr;
and reached a maximum of 0.138 ppm. In New Jersey, EDU studies showed that
ambient ozone reduced the yields of Norland potatoes about 25 percent in two
separate years (1978 and 1980), and the yield of Norchip potato was signifi-
cantly reduced (10 percent) in 1980 (Clarke et al., 1983). The yield of the
6-124
-------
ozone-resistant cultivar was not improved by EDU treatment. During these 2
years (1978 and 1980), the ambient oxidant doses were 65 and 110 ppm-hr, which
is equivalent to mean ozone concentrations of approximately 0.030 and 0.051
ppm for the study period. A 3-year field study (in southern Canada) using
three potato cultivars found that plants treated with EDU did not yield better
than untreated plants (Hoiley et a!., 1985). During the study period (July 1
to August 31) the 0~ concentration exceeded 0.08 ppm for 62 (1980), 18 (1981),
and 26 (1982) hours. Although EDU treatment did not increase yield, it did
reduce foliar 0, injury. The authors concluded that the ambient 0, concentra-
tions were too low and the resultant severity of injury was too small to have
a significant effect on yield. A combined treatment with EDU and a fungicide
(Du-Ter) significantly increased yield. The fungicide apparently prevented
the early blight pathogen, Alternaria solani, from colonizing the 0,,-induced
foliar lesions. This study demonstrates that 0- can render the plant more
sensitive to biotic stresses which, in this case, subsequently induced the
yield loss.
The results of the above studies show that chemical protectants can
improve crop yield and can be used to provide estimates of ozone-induced crop
loss on several crop species. The data clearly show that the ozone concentra-
tions occurring during these studies were sufficiently high to reduce crop
yields 10 to 40 percent, even though there were few times when the ozone
concentrations exceeded 0.08 ppm.
6.4.3.2.1.3 Other field studies. Low concentrations of 0- added to
filtered air in field chambers induced yield reductions in a variety of plant
species (Table 6-21). Alfalfa exhibited a 49 percent decrease in top dry
weight when exposed to 0.05 ppm of 0- for 68 days (Neely et al., 1977).
Extended (several weeks) exposures to 0.10 ppm caused yield reductions in
alfalfa (Neely et al., 1977), soybean (Heagle et al., 1974), sweet corn (Heagle
et al., 1972), and ponderosa and western white pine seedling (Wilhour and
Neely, 1977). Stem specific gravity, an indicator of wood density and quality,
of several hybrid poplar clones was consistently less in response to 0.15 ppm
of 03 12 hr/day for 102 days; but effects on height ranged from slight stimula-
tions in four clones to a significant reduction in one clone (Patton, 1981).
6.4.3.2.1.4 Greenhouse and indoor chamber studies. The effects of 0- on
plant yield may be mediated by a myriad of genetic, cultural, and environmental
factors (see Section 6.3). The previously discussed studies have attempted to
6-125
-------
TABLE 6-21. EFFECTS OF OZONE-ADDED TO FILTERED AIR IN FIELD CHAMBERS ON THE YIELD OF SELECTED CROPS
Plant species
Alfalfa
Alfalfa
i
[^ Soybean
(Dare)
Sweet corn
(Golden midget)
(White midget)
Douglas fir
Jeffrey pine
Lodgepole pine
Monterey pine
03 concn. ,
ppm
0.05
0.10
0.05
0.10
0.05
0.10
0.05
0.10
0.10
0.10
0.10
0.10
Yield , % reduction
Exposure duration from control
7 hr/day, 68 days 31- top dry wt, 1st harvest;
49 top dry wt, 2nd harvest
17- total protein, top, 1st harvest;
42,, total protein, top, 2nd harvest
32 , total nonstructural carbo-
f hydrate (TNC), 1st harvest (top)
55 , total nonstructural car-
bohydrate (TNC), 2nd harvest
(top)
7 hr/day, 70 days 51-, top dry wt, final harvest
53 , total nonstructural car-
f bohydrate (TNC), final harvest
38 , total protein, final harvest
6 hr/day, 133 days 3, seed wt/plant
55 , seed wt/plant
6 hr/day, 64 days 9, kernel dry wt
45 , kernel dry wt
6 hr/day, 71 days 0
6 hr/day, 126 days 6, height; 15, stem dry wt
2, height; 2, stem dry wt
8, height, 8, stem dry wt
0, height; 0, stem dry wt
Monitoring Calibration
method method
Mast Known 03
source
1% NBKI
Mast Known 03
source,
Mast 2% NBKI
Mast 2% NBKI
Mast Known 03
source,
1% NBKI
Fumigation
facility Reference
FC-CT Neely et
al. (1977)
FC-CT Neely et
al. (1977)
1% NBKI
FC-CT Heagle et
al. (1974)
FC-CT Heagle et
al. (1972)
FC-CT Wilhour and
Neely (1977)
-------
TABLE 6-21 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR IN FIELD CHAMBERS ON THE YIELD OF SELECTED CROPS3
CTs
I
ro
03
Plant species
Ponderosa pine
Shore pine
Sugar pine
Western white pine
Sitka spruce
Hybrid poplar
(252)
(279)
(346)
CW5)
(W87)
Hybrid poplar
(42)
(50)
(207)
(215)
concn. ,
ppm
0.10
0.10
0.10
0.10
0.10
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
Yield, % reduction
Exposure duration from control
6 hr/day, 126 days 11, height, 21f, stem dry wt
2, height; 6, stem dry wt
0, height; 0, stem dry wt
0, height, 9 , stem dry wt
0, height, 14, stem dry wt
12 hr/day, 102 days +16, height; 12f, stem specific
gravity
23, height; 14f, stem specific
gravity
3, height; 6f, stem specific
gravity
5, height; 12 , stem specific
gravity
+19, height; 11 , stem specific
gravity
25, height; 8, stem specific
gravity
58 , height; 1, stem specific
gravity
+8, height; 7 , stem specific
gravity
+17, height; 11, stem specific
gravity
Monitoring Calibration Fumigation
method method facility Reference
Mast Known 03 FC-CT Wilhour and
source, Neely (1977)
% NBKI
UV Known 03 OT Patton 1981
source
a column entry is blank the information is the same as above.
+ = an increase above the control.
cMast = Mast meter (coulometric); UV = ultraviolet spectrometry.
NBKI = neutral buffered potassium iodide.
eOT = open-top chamber; FC-CT = closed-top field chamber.
Significant at p = 0.05
-------
quantify plant responses to 0- under ambient or normal environmental and
cultural conditions. Several investigations on the yield responses of plants
to 03 have been performed under more controlled (to various degrees) conditions
(Tables 6-22, 6-23). These exposures at 0.041 to 0.40 ppm of 03 will be
discussed as they relate to the previous studies.
Ozone caused significant yield reductions in exposures lasting several
weeks (Table 6-23). At 0- concentrations of 0.05 ppm or greater, the response
varied among species. Hybrid poplar cuttings exhibited a 13-fold increase in
leaf abscission in response to 0.041 ppm for 5 months (Mooi, 1980). There was
a significant 14 percent increase in height accompanied by a slight decrease
in stem dry weight. American sycamore seedlings exhibited a significant 9
percent height reduction (Kress et al., 1982b), and loblolly pine seedlings
showed 18 percent height reduction (Kress and Skelly, 1982) at 0.05 ppm for 4
wk. Yellow poplar and white ash seedlings exhibited significant 60 percent
and 22 percent increases in height and total dry weight, respectively, follow-
ing identical exposures (Kress and Skelly, 1982). In general, slight growth
stimulations by 0, are more common in hardwood tree species than in coniferous
tree species (Kress and Skelly, 1982) (Table 6-23).
Significant yield reductions were noted for many species exposed to 0.05
to 0.10 ppm of Cs for one to several weeks (Tables 6-22, 6-23). Carnations
had significantly fewer flowers and flower buds when grown in air containing
0.05 to 0.09 ppm of 03 for 24 hr/day for 12 to 56 days (Feder and Campbell,
1968). Pasture grasses produced less top growth when exposed to 0.09 ppm of
03 for 4 hr/day for 5 wk (Horsman et al., 1980). Exposure-response equations
were developed for three fescue cultivars under greenhouse conditions (Flagler
and Youngner, 1982a). The cultivar Kentucky 31 showed the largest yield de-
crease with increasing 03 concentration; based on yield data it was ranked most
sensitive and Fawn the least sensitive of the three. Significant yield reduc-
tions (10 percent) were predicted for each of the cultivars at the following
0~ concentrations (ppm): 0.119 (Kentucky 31), 0.10 (Alta), and 0.11 (Fawn).
The cultivars were exposed for 6 hr/day, 1 day/wk for 7 wk. Significant yield
reductions have been noted for alfalfa (Hoffman et al., 1975); clover (Blum et
al., 1982), and loblolly pine, pitch pine, sweetgum, American sycamore, and
green ash (Kress and Skelly, 1982) when exposed to 0.10 ppm of 0, for various
lengths of time. Numerous studies have reported no significant effects, how-
ever, and some have reported yield stimulations. Significant yield stimulations
6-128
-------
TABLE 6-22. EFFECTS OF OZONE ADDED TO FILTERED AIR ON THE YIELD OF SELECTED CROPS
Plant species
Pinto bean
Sweet corn
(Golden jubilee)
Wheat
(Arthur 71)
CD
1-1 (Blueboy)
ro '
Radish
(Cherry belle)
Radish
(Cavalier)
concn. ,
ppm
0.15
0.25
0.35
0.20
0.35
0.35
0.20
0.20
0.20
0.40
0.25
Exposure duration
2 hr/day, 63 days
3 hr/day, 3 days/
wk, 8 wk
3 hr/day, 3 days/
wk, 8wk
4 hr/day, 7 days
4 hr/day, 7 days
3 hr or 6 hr
3 hr
Yield, %
reduction from control
44ng, pod fresh wt
100ng, pod fresh wt
100ng, pod fresh wt
13e, ear fresh wt; 13e,
kernel dry wt;
+1300 , length of ear with
shrivelled kernals
22 kernel dry wt
30e, seed yield; 17e,
kernel wt; 8, % seed set
24. seed yield; 2, kernel wt;
22, % seed set
6ng, root fresh wt; 6ng,
root dry wt
38ng, root fresh wt; 40ng,
root dry wt
33 , root dry wt (average
of 4 pre- or post-fumi-
Monitoring
method
Mast
Mast
Mast
Chem
Mast
Calibration Fumigation
method facility
Not given Room
2% NBKI GH
Not given GC
Known 03 CH-CSTR
source
Not given
Reference
Hoffman et
al. (1973)
Oshima (1973)
Shannon and
Mulchi (1974)
Reinert and
Gray (1981)
Adedipe and
Omrod (1974)
(Cherry belle) 0.25
Beet
0.20
3 hr
0.5 hr/day, 38 days
1 hr/day, 38 days
2 hr/day, 38 days
3 hr/day, 38 days
gation temperature regimes)
37 , root dry wt (average
of 4 pre- or post-fumigation
temperature regimes)
+9, storage root dry wt
+2^ storage root dry wt
40 , storage root dry wt
40e, storage root dry wt
Mast
Not given
GC
Ogata and
Maas (1973)
-------
TABLE 6-22 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON THE YIELD OF SELECTED CROPS
Plant species
Potato
(Norland)
(Kennebec)
Pepper
(M-75)
Tomato
(Walter)
CTl
1 Cotton
Co (Acala SJ-2)
O
Carnation
(White sim)
Coleus
(Pastel rainbow)
Snapdragons
(Rocket mixture)
(Floral carpet formula
mixture)
03
concn. ,
ppm
0.20
0.20
0.12
0.20
0.20
0.25
0.25
0.05-
0.09
0.10
0.20
0.40
0.10
0.20
0.40
0.10
0.20
0.40
Exposure duration
3 hr/day, every
2 wk, 120 days
3 hr/day, every
2 wk, 140 days
3 hr/day, 3 day/
wk, 11 wk
4 hr/day, 2 day/
wk, 13 wk
6 hr/day, 2 day/
wk, 13 wk, 6 hr/
day, 2 day/wk,
18 wk
24 hr/day, 12 days
23 days
44 days
56 days
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
Yield, % Monitoring
reduction from control method
20e, tuber no. ; 25e, Not given
tuber wt; 13 , total solids
36e, tuber no.- 42e,
tuber wt; 20 , total solids
19e, dry wt/fruit; 20, no. Mast
mature fruit- 50 , dry
wt/fruit; 53 , no mature fruit
6, fruit fresh wt Chera
52e, no. of bolls; 62e, UV
fiber dry wt: 55, no.
of bolls; 59 , fiber dry wt
74 , no. of flower buds Mast
53 , no. of flower buds
46e, no. of flower buds
100e, no. of normal open
flowers
+3, flower no. Mast
4^ flower no.
8 , flower no.
+1, flower no. Mast
10, flower no.
9, flower no.
+3, flower no. Mast
2, flower no.
4, flower no.
Calibration Fumigation
method facility Reference
Not given GC Pell et al.
(1980)
UV CH Bennett et
al. (1979)
Known 03 GH-CSTR Shew et al.
source (1982)
UV CH Oshima et
al. (1979)
Not given GH Feder and
Campbell
(1968)
Not given CH Adedipe et
al. (1972)
Not given CH
Not given CH
-------
TABLE 6-22 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON THE YIELD OF SELECTED CROPS
Plant species
Begonia
(Linda)
(Scarletta)
(White Tausendschon)
cr> Petunia
>_. (Canadian All Double
oo Mixture)
(Capri)
(Bonanza)
Coleus
(Scarlet Rainbow)
Begonia
(Schwabenland red)
(Whisper-0-pink)
03
concn. ,
ppm
0.10
0.20
0.40
0.10
0.20
0.40
0.10
0.20
0.40
0.10
0.20
0.40
0.10
0.20
0.40
0.10
0.20
0.40
0.10
0.20
0.40
0.25
0.25
Exposure duration
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
2 hr
4 hr/day, 4 times
once every 6 days
4 hr/day, 4 times
once every 6 days
Yield, %
reduction from control
4, flower no.
9, flower no.
5, flower no.
+5, flower no.
+3. flower no.
8 , flower no.
5, flower no.
10, flower no.
10, flower no.
0, flower no.
4, flower no.
7, flower no.
7, flower no.
6» flower no.
14 , flower no.
+3, flower no.
8, flower no.
10, flower no.
+3, flower no.
20 , flower no.
28e, flower no.
39e, flower wt; (54%
foliar injury)
22e, flower wt; (25%
foliar injury)
Monitoring
method
Mast
Mast
Mast
Mast
Mast
Mast
Mast
Chem
Chem
Calibration Fumigatiog
method facility
Not given CH
CH
CH
CH
CH
CH
CH
Not given GH-CSTR
GH-CSTR
Reference
Adedipe et
al. (1972)
Reinert and
Nelson (1980)
-------
TABLE 6-22 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON THE YIELD OF SELECTED CROPS
CO
ro
Plant species
(Fantasy)
(Renaissance)
(Turo)
Alfalfa
(Moapa)
Alfalfa
(Moapa)
Pasture grass
(N.Z. grasslands)
(Victorian)
(Australian)
Lad i no clover
(Tillman)
Tall fescue
(Alta)
03
concn. ,
ppm
0.25
0.25
0.25
0.10
0.15
0.20
0.10
0.10
0.09
0.09
0.09
0.10
0.10
0.20
0.30
0.40
Exposure duration
4 hr/day, 4 times
once every 6 days
2 hr/day, 21 days
2 hr/day, 21 days
2 hr/day, 21 days
2 hr/day, 21 days
2 hr/day, 42 days
4 hr/day, 5 days/
wk, 5 wk
4 hr/day, 5 days
wk, 5 wk
4 hr/day, 5 days
wk, 5 wk
6 hr/day, 5 days
6 hr/day, 1 day/wk,
7 wk
6 hr/day, 1 day/wk,
7 wk
Yield, %
reduction from control
6e, flower wt; (2%
foliar injury)
55e, flower wt; (15%
foliar injury)
+10, flower wt; (8%
foliar injury)
p
16 , top dry wt
26^, top dry wt
39 , top dry wt
21e, top dry wt
20 , top dry wt
20e, top dry wt
14e, top dry wt
18e, top dry wt
20e, shoot dry wt; 38e,
snoot total nonstructural
carbohydrate (TNC)
10, dry wt/plant
20, dry wt/plant
30, dry wt/plant;
Monitoring
method
Chem
Chem
Chem
Mast
Mast
Chem
Chem
Chem
Chem
UV
Calibration Fumigation
method facility
Not given GH-CSTR
GH-CSTR
GH-CSTR
Not given CH
CH
Not given GC
GC
GC
2% NBKI GH-CH
UV GH-CSTR
Reference
Reinert and
Nelson (1980)
Hoffman et
al. (1975)
Horsman et
al. (1980)
Blum et al.
(1982)
Flagler and
Youngner
(1982a)
significant linear
regression: r=0.98
-------
TABLE 6-22 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON THE YIELD OF SELECTED CROPS
CO
Plant species
(Fawn)
(Kentucky-31)
Tall fescue
(Alta)
03
concn. ,
ppm
0.10
0.20
0.30
0.40
0.10
0.20
0.30
0.40
0.10
0.20
0.30
Exposure
6 hr/day
7 wk
6 hr/day
7 wk
6 hr/day
7 wk
6 hr/day
12 wk
6 hr/day
12 wk
duration
, 1
, 1
, 1
, 1
, 1
day/wk ,
day/wk ,
day/wk ,
day/wk ,
day/wk,
Yield, %
reduction from control3
9, dry wt/ plant
18, dry wt/plant
significant linear
36, dry wt/plant
regression, r = 0.99
13, dry wt/plant
27, dry wt/plant
significant linear
40, dry wt/plant
regression, r = 0.98
54, dry wt/plant
+3, top dry wt
19, top dry wt
41, top dry wt
Monitoring Calibration Fumigation
method method facility Reference
UV UV GH-CSTR Flagler and
Youngner
(1982a)
UV UV CH-CSTR Flagler and
Youngner,
(1982b)
+ = an increase above the control; ng = statistical data not given.
Chem. = chemiluminescence; Mast = Mast meter (coulombmetric); UV = ultraviolet spectrometry.
CNBKI = neutral buffered potassium iodide.
GH = greenhouse; CSTR = continuous stirred tank reactor; GH=CSTR = CSTR in greenhouse; GC = controlled environment growth chamber; CH = manufactured
chamber other than CSTR or GC; GH-CH = CH in greenhouse; Room = plant growth room.
Significant at p = 0.05.
-------
TABLE 6-23. EFFECTS OF OZONE ADDED TO FILTERED AIR ON THE YIELD OF SELECTED TREE CROPS
cr>
i
Plant species
Poplar
(Dorskamp)
(Zeeland)
American Sycamore
(16-SYC-19)
(16-SYC-23)
American Sycamore
(16-SYC-19)
(16-SYC-23)
Sweetgum
American Sycamore
White ash
Green ash
03
concn. ,
ppm
0.041
0.041
0.05
0.05
0.05
0.05
0.05
0.10
0.15
0.05
0.10
0.15
0.05
0.10
.0.15
0.05
0.10
0.15
Exposure duration
12 hr/day, 5 mo
12 hr/day
6 hr/day,
6 hr/day,
6 hr/day,
6 hr/day,
6 hr/day,
6 hr/day,
6 hr/day,
6 hr/day,
, 5 mo
28 days
28 day
28 days
28 day
28 days
28 days
28 days
28 days
Yield, % Monitoring Calibration
from control method method
+14e, stem length; 12 stem dry wt; Chem
+1333, no. of dropped leaves; 6,
total dry wt
2, stem length; 4, stem dry wt; Chem
+692, no. of dropped leaves; 0,
total dry wt
9e, height growth Chem
2, height growth
11, height growth • Chem
9e, height growth
+9. height growth; 10, total dry wt Chem
29^, height growth; 26. total dry wt
45 , height growth; 42 , total dry wt
+4. height growth; 23, total dry wt Chem
21, height growth; 61 , total dry wt
21 , height growth; 69 , total dry wt
+12, height growth; +22e, total dry wt Chem
9, height growth; 9, total dry wt
15, height growth; 17 , total dry wt
2. height growth; 14, total dry wt Chem
24 , height growth, 28, total dry wt
30e, height growth; 33, total dry wt
NBKI
NBKI
1% NBKI
1* NBKI
Constant
source,
NBKI, UV
Constant
source,
NBKI, UV
Constant
source,
NBKI, UV
Constant
source,
NBKI, UV
Fumigation
facility Reference
GH-CH Mooi (1980)
GH-CH Mooi (1980)
CH Kress et
al. (1982b)
,
CSTR Kress et
al. (1982b)
CSTR Kress and
Skelly
(1982)
CSTR
CSTR Kress and
Skelly
(1982)
CSTR Kress and
Skelly
(1982)
-------
TABLE 6-23 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON YIELD OF SELECTED TREE CROPS
Plant species
Willow oak
Sugar maple
Yellow poplar
CD
^ Yellow poplar
u>
on
Cottonwood
White ash
White ash
Black cherry
Hybrid poplar
(NS 207 + NE 211)
03
concn. ,
ppm
0.05
0.10
0.05
0.10
0.15
0.05
0.10
0.15
0.10
0.10
0.10
0.10
0.20
0.30
0.40
0.10
0.20
0.30
0.40
0.15
Exposure duration
6 hr/day, 28 days
6 hr/day, 28 days
6 hr/day, 28 days
12 hr/day, 48 days
4 hr/day, 1 day/wk,
9 wk
8 hr/day, 5 days/wk,
6 wk
Yield, % Monitoring Calibration Fumigation
from control method method facility
1, height growth; 2, total dry wt Chem Constant CSTR
4, height growth; 11, total dry wt source,
NBKI, UV
5, height growth; 2, total dry wt Chem CSTR
+8 , height growth; 7, total dry wt
12e, height growth; 41e, total dry wt
+60e, height growth; +41, total dry wt Chem CSTR
+8, height growth; +5, total dry wt
12, height growth; +18, total dry wt
19ng, relative growth rate Chem Not given CSTR
59ng, relative growth rate
no significant effects
+13, total height; +7, shoot dry wt Not given Not given Not given
0, total height; +5, shoot dry wt
0, total height; 11, shoot dry wt
0, total height; 14, shoot dry wt
+16, total height; +15, shoot dry wt Not given Not given Not given
+5, total height; 4, shoot dry wt
+3, total height; 4, shoot dry wt
28 , total height; 15, shoot dry wt
50e, dry wt new shoots from terminal Not given Not given GH-CH
cuttings
62 , dry wt new shoots from basal
Reference
Kress and
Skelly
(1982)
Kress and
Skelly
(1982)
Jensen
(1981)
McClenahen,
(1979)
McClenahen,
(1979)
Jensen and
Dochinger
(1974)
cuttings
-------
TABLE 6-23 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON YIELD OF SELECTED CROPS
Plant species
Hybrid poplar
(207)
Yellow birch
White birch
Bigtooth aspen
Eastern cottonwood
01
JL Red maple (163 ME)
co
CT,
(167 NB)
(128 OH)
Loblolly pine
(4-5 x 523)
(14-5 x 517)
Loblolly pine
Pitch pine
. °3
concn. ,
ppm
0.20
0.20
0.25
0.25
0.25
0.25
0.25
0.05
0.05
0.05
0.10
0.15
0.05
0.10
0.15
Yield, % Monitoring Calibration Fumigation
Exposure duration from control method method facility
7.5 hr/day, 5 day/wk, 5, height Not given Not given CH
6 wk 8, height
8 hr/day, 5 day/wk, 9, height MAST NBKI GH-CH
15 wk
34, height
+7, height
8 hr/day, 6 wk 18, height MAST 1% NBKI CH
32, height
37e, height
6 hr/day, 28 days 6, height growth Chem 1% NBKI CH
6 hr/day, 28 days 18e, height growth; 14. total dry wt Chem Constant CSTR
27e, height growth; 22 , total dry wt source,
41e, height growth; 28e, total dry wt NBKI, UV
6 hr/day, 28 days 4. height growth; 8, total dry wt
13 , height growth; 19 total dry wt
26e, height growth; 24 , total dry wt
Reference
Jensen
(1979)
Jensen and
Masters
(1975)
Dochinger
and Town-
send (1979)
Kress et
al. (1982a)
Kress and
Skelly
(1982)
-------
TABLE 6-23 (cont'd). EFFECTS OF OZONE ADDED TO FILTERED AIR ON YIELD OF SELECTED CROPS
Plant species
Virginia pine
White spruce
CTl
i
^ Japanese larch
03
concn. ,
ppm
0.05
0.10
0.15
0.25
0.25
Exposure duration
6 hr/day, 28 days 5,
11,
14,
8 hr/day, 5 day/wk, 5,
15 wk
+6,
Yield, % Monitoring Calibration
from control method method
height growth; +2, total dry wt Chem
height growth; 3, total dry wt
height growth; 13, total dry wt
height Mast
height Mast
Constant
source,
NBKI, UV
NBKI
NBKI
Fumigation
facility Reference
CSTR Kress and
Skelly (1982)
GH-CH Jensen and
Masters
(1975)
GH-CH
+ = an increase above the control; ng = statistical data not given.
Chem. = chemiluminescence; Mast = Mast meter (coulometric); UV = ultraviolet spectrometry.
NBKI = neutral buffered potassium iodide.
GH = greenhouse; CSTR = continuous stirred tank reactor; CH = manufactured chamber other than CSTR or GC; GH-CH = CH in greenhouse.
Significant at p = 0.05.
-------
in response to 0.10 ppm of 03 for 6 hr/day for 4 wk have been noted for sugar
maple (Kress and Skelly, 1982).
It is difficult to extrapolate data from studies conducted under more
controlled conditions (greenhouse, growth chamber) to field conditions, except
when plants are normally grown under these conditions (e.g., flower crops).
The more controlled chamber data can serve, however, to strengthen the demonstra-
tion of 03 effects in the field. Concentrations of 0.05 ppm of 03, in extended
or repeated exposures, have been shown to cause yield reductions in some
species or cultivars, no effects in others, and increased yield in others.
Concentrations of 0.10 ppm and above more consistently cause yield reductions,
although exceptions can be found (Tables 6-21, 6-22, 6-23).
6.4.3.2.1.5 Effects of ozone on crop quality. Quality is a broad term
that includes many features such as chemical composition, physical appearance,
taste, and ability to withstand storage and transport. All these features
have economic importance.
Four types of experimental approaches were used to investigate the effects
of 03 or oxidants on crop quality: (1) field experiments in which the impact
of ambient oxidants and charcoal-filtered air were contrasted; (2) field
experiments in which ambient oxidant injury was prevented by using an anti-
oxidant chemical spray; (3) field experiments in which 03 was added to ambient
or charcoal-filtered air; and (4) laboratory experiments in which potential
effects were measured by exposing plants to 03.
The effects of ambient oxidants were studied at three different locations
(Riverside, California; Geneva, New York; and Beltsville, Maryland) to deter-
mine their impact on the quality of alfalfa, grape, and soybean, respectively
(Thompson et al., 1976b; Musselman et al., 1978; Howell and Rose, 1980). Over
a period of 7 harvests, alfalfa plants experienced oxidant concentrations
greater than 0.08 ppm between 25 and 60 percent and 0.12 ppm between 5 and 50
percent of daylight hours, respectively (measured with a Mast meter). Plants
receiving ambient oxidants exhibited significant (p = 0.05 or 0.01) changes in
a number of quality variables in some harvests. Ambient oxidants decreased
crude fiber, p-carotene, and vitamin C; increased niacin; and had no effect on
protein efficiency and nitrogen digestibility ratios (Thompson et al., 1976b).
Grape crops exposed to ambient oxidants suffered a 6 percent reduction in
soluble solids (p = 0.05), which would reduce the value of this fruit for wine
(Musselman et al., 1978); however, ozone concentrations were not measured at
6-138
-------
the Fredonia, NY, site where grape experiments were conducted. Soybean seed
quality exhibited small but significant (p = 0.05) changes; protein was in-
creased 2 percent and oil was decreased 3.8 percent (Howell and Rose, 1980)
when the plants were exposed to ambient oxidants at 0.08 ppm or greater for 3
percent of the season and 0.12 ppm or greater for 0.6 percent of the growing
season (the experimental conditions for the seed quality study are reported in
Howell et al., 1979).
In addition to measuring yield in terms of biomass, some of the NCLAN
studies have examined the quality of the yield. Corsoy soybeans exhibited a
2
significant linear decrease (R = 0.81) in the percentage of oil content of
seeds as the 0- concentration increased. Concurrently, there was a significant
increase in percentage of protein content with increasing 03 concentration
(Kress and Miller, 1983). Estimated changes resulting from a seasonal 7-hr
average concentration of 0.10 ppm of 0^ were a 5 percent decrease in oil
content and a 4 percent increase in protein content.
Clarke et al. (1983) grew potatoes in ambient air plots in central New
Jersey; half the plants were treated with the antioxidant EDU to suppress the
effects of Oo. In 1980, the ambient oxidant dose was 110 ppm-hr. Specific
gravity, a quality directly correlated with high quality of processed and
tablestock potatoes, was 0.4 percent lower in non-EDU-treated plants (p =
0.05). In 1978, the ambient oxidant dose was 65 ppm-hr; changes in specific
gravity were not detected.
Alfalfa plants exposed to 0.10 ppm of 0, (7 hr/day for 70 days) showed an
increased protein and amino acid content per unit area, but a~ decrease in
total protein and amino acid because of reduced dry matter production (Neely
et al., 1977). Reductions were also noted in the (3-carotene and total nonstruc-
tural carbohydrate.
Small trees from several clones of hybrid poplar have exhibited decreased
stem specific gravity (a measure of wood quality that could result in reduced
wood strength or reduced pulpwood value) when exposed to 0.15 ppm 0,. for 12
hr/day for 102 days in open-top chambers (Patton, 1981).
A number of investigators have exposed greenhouse-grown crops to con-
trolled doses of 0., and subsequently measured the impact on crop quality
(chemical composition). These results serve more as indicators of potential
impact than predictors that effects would occur in a field environment.
Results are summarized below.
6-139
-------
Pippen et al. (1975) exposed cabbage, carrot, corn, lettuce, strawberry,
and tomato to intermittent acute doses of (L. Ozone concentrations ranged
from 0.20 to 0.35 ppm for 2.5 to 6.5 hr, from 1 to 3 days/wk from emergence to
harvest. Plants were exposed to 03 for 1.62 to 3.59 percent of the life
cycle, depending on the species. Some of the species studied exhibited signi-
ficant (p = 0.05) changes in quality in response to one or more of the 0,
regimes employed. Corn exhibited a decrease (at 0.20 ppm 0-) in solids,
p-carotene, and carbohydrates, but total nitrogen and vitamin C levels increased.
The niacin concentration increased in carrots and strawberries. Solids, fiber
content, and vitamin C were all reduced in tomato (at 0.35 ppm 03). Cabbage
exhibited significant increases in total solids and vitamin C.
When greenhouse-grown potato plants were exposed to 0- at a concentration
of 0.20 ppm for 3 hr once every 2 wk throughout the growth period, tubers
exhibited a decrease in percentage dry matter that is associated with a decrease
in fluffiness of tablestock potatoes (Pell et al., 1980). Reducing sugars,
associated with undesirable darkening of potato chips, increased in tubers
harvested from plants exposed to 0,. Glycoalkaloids, compounds that can cause
a bitter taste in potato tubers, either decreased or were unaffected by the 03
treatment (Speroni et al., 1981).
The potential of 0, to induce a series of estrogenic isoflavonoids was
investigated in five different alfalfa cultivars (Hurwitz et al. , 1979; Ska'rby
and Pell, 1979; Jones and Pell, 1981). These biochemicals have been directly
correlated with breeding disturbances in both domesticated and wild animal
species. Co'timestrol, daidzein, genistein, and formononetin, all with poten-
tially adverse affects on crop quality, were not detected in greenhouse-grown
alfalfa plants that received 0, concentrations of 0.20 to 0.40 ppm for 3 hr.
Ladino clover, another forage crop, exhibited reduced total nonstructural
carbohydrate and generally increased mineral content (except for sodium) when
exposed to 0.10 ppm of 03 (6 hr/day for 5 days) (Blum et al., 1982).
The impact of 03 and ambient oxidants on crop quality has important
implications from both health and economic perspectives. A reduction in
nutritional value of food or forage, such as reduced vitamin content or precur-
sors to proteins, will be detrimental to the consumer. An adverse effect on a
crop destined for processing, such as grapes for wine or potatoes for chips,
will reduce the economic value of the crop. It is difficult at present,
however, to correlate completely these effects with the more conventional
measures of 03 effects on foliage and yield.
6-140
-------
6.4.3.2.1.6 Effects of ozone on plant reproduction. Ozone has been
shown to affect the reproductive capacities of plants. The flowering and seed
production of soybean plants was reduced by (L at 0.10 ppm (6 hr/day, 133
days) (Heagle et al., 1974). In sweet corn plants, seed production as estimated
by percentage of ear filled was reduced when the plants were grown in an
environment of 0.10 ppm of 03 (6 hr/day, 64 days) (Heagle et al., 1972).
Wheat plants exposed to 0.20 ppm of 03 (4 hr/day, 7 days) at anthesis exhibited
reduced seed set (Shannon and Mulchi, 1974). Reduced seed production of
cotton plants exposed to 0.25 ppm 0, (6 hr/day, 2 day/wk, 13 wk) was reported
(Oshima et al., 1979).
The number of tillers in three tall fescue cultivars increased slightly
as 03 was increased from 0.10 to 0.40 ppm (6 hr/day, 1 day/wk, 7 wk) (Flagler
and Youngner, 1982a). These data indicate that 03 may decrease the reproductive
capacity of plants. The reductions in seed production suggest an 03 impact on
fertilization processes. The observation that 0- (0.05 ppm for 5.5 hr) reduced
pollen germination and pollen tube elongation (40 to 50 percent) in tobacco
and petunia (Feder, 1968) supports this conclusion. Ozone also reduced the
germination of corn pollen 60 (0.06 ppm) and 70 percent (0.12 ppm), respec-
tively (Mumford et al., 1972). Plants were exposed to 03 (0.06 or 0.12 ppm
for 5.5 hr/day for 60 days) and the pollen was harvested daily as soon as it
was mature and the percentage germination could be determined. Because the
pollen was harvested as soon as it reached maturity, it is probable that the
pollen was exposed to 03 for only a short time period, with the data thus
indicating that pollen is quite sensitive to 0^.
6.4.3.2.1.7 Relationship between foliar injury and yield loss. Because
plant growth depends on the presence of functional leaves to conduct the
photosynthesis required for plant growth, various studies have been conducted
to determine the association between foliar injury and yield for species in
which the foliage is not part of the yield. Some investigations discussed in
the 1978 criteria document (U.S. Environmental Protection Agency, 1978) demon-
strated yield loss with little or no foliar injury (Tingey and Reinert, 1975;
Tingey et al., 1971a); others demonstrated significant foliar injury not
accompanied by yield loss (Heagle et al., 1974; Oshima et al., 1975). Many
other studies can be cited to illustrate the inconsistency of the relationship
between foliar injury and yield loss when the foliage is not the yield component.
Significant yield reductions with no foliar injury have been noted for American
6-141
-------
sycamore (Kress et al., 1982b), loblolly and pitch pine (Kress and Skelly,
1982), carnations (Feder and Campbell, 1968), and petunia and coleus (Adedipe
et al., 1972). With red maple seedlings, foliar injury was directly corre-
lated with subsequent height reductions (Dochinger and Townsend, 1979). The
relative sensitivities of two potato cultivars were reversed when judged on
yield reductions rather than foliar injury (Pell et al., 1980). In a study
comparing the effects of long- and short-term exposures, a long-term exposure
(0.15 ppm for 8 hr day, 5 days/wk for 6 wk) resulted in 75 percent foliar
injury and 50 percent growth reduction; whereas the short-term exposure (1.0
ppm for 2.4 or 8 hr) resulted in 70 percent foliar injury and no growth reduc-
tion (Jensen and Dochinger, 1974).
All of the studies in Table 6-20 reported foliar injury as well as yield
responses. For field corn, foliar injury response was at lower concentrations
than the yield effects, but with increased 0, concentration the percentage
yield reductions became greater than the percentage foliar injury (Heagle et
al., 1979a). For wheat, the increases in foliar injury were generally accom-
panied by decreases in yield, but foliar injury was not a good predictor of
yield reduction. For example, at 0.06 ppm 0,, the wheat cultivar Coker 47-27
had 5 percent foliar injury (compared to the control) and 11 percent yield
reduction; but the cultivar Holly had 6 percent foliar injury and 1 percent
yield reduction (Heagle et al., 1979c). There were no obvious relationships
between foliar injury and shoot fresh or dry weight of spinach (Heagle et al.,
1979b). In the soybean study also, relative cultivar foliar injury did not
predict relative yield response (Heagle and Letchworth, 1982). The cultivars
Bragg and Ransom had equal amounts of foliar injury (35 percent) when exposed
to 0.10 ppm of 0-, but Bragg yield increased 4 percent and Ransom yield decreased
20 percent.
The lack of correlation between foliar injury and yield reduction for
many crops should not be surprising. Plants have evolved with a reserve
capacity to cope with some level of stress. For example, a plant species may
develop more leaf area than that needed for maintaining yield. Therefore, 0.,
would not be expected to reduce plant yield unless its effects were suffi-
ciently great to make some process limiting for plant yield. Yield would also
be reduced if 0- directly impacted the process limiting growth. Unless either
of these two conditions is achieved, the plant might display a biological
(phytotoxic) response to 03 but the yield would not be impaired. For plants
6-142
-------
in which the foliage, however, is the marketable portion, either for food or
ornamental use, a phytotoxic impact on the foliage may reduce the yield without
altering the plant weight. These concepts imply that not all impacts of 0- on
plants are reflected in growth or yield reductions. Also, 0, would not impact
plant growth or yield unless it made some process more limiting for growth or
yield than the environmental factors that currently were controlling growth.
These conditions suggest that there are combinations of 03 concentration and
exposure duration that the plant can experience that will not result in visible
injury or reduced plant growth and yield. Numerous studies of many plant
responses have demonstrated combinations of concentration and time that did
not cause a significant effect.
6.4.3.2.2 Biomass and yield responses from ambient exposures. Determination
of the effects of ambient air pollutants directly shows the impact of existing
air quality on plant yield in the environment. Two basic types of studies are
used to describe the effects of ambient exposures on plants. In one type,
field observations are used to develop an association between 03 exposure and
plant response (growth or yield reductions or mortality). In the other type,
the difference between plant yield in charcoal-filtered air and in ambient air
(which may contain a single major pollutant or several) is used to indicate
the impact of the pollutant; and some type of exposure chamber is required for
these studies. In either case, plants are exposed to pollutant concentrations
at the frequency of occurrence found in the ambient air. When only a single
pollutant is present or the study is conducted at a single location, or both,
the interpretation of the results is simplified. When the studies are conducted
at different locations, however, differences in climatic and edaphic conditions,
in addition to the pollutant time series that may influence the results and
complicate the interpretation, can occur.
The previous criteria document (U.S. Environmental Protection Agency,
1978)' reviewed the effects of 0, in ambient air (Table 6-24). These studies
utilized charcoal filtration in greenhouses or open-top chambers or simply
correlated effects with the ambient 0, concentrations. Leaf injury (sweet
corn, tobacco, potato), yield reductions (citrus, grape, tobacco, cotton,
potato), and quality changes (grape) were documented. It was concluded that
ambient oxidants were causing decreased plant growth and yield.
More recently, studies have also been conducted to evaluate the yield of
plants grown in the presence of photochemical oxidants (ambient air) versus
charcoal-filtered air (Table 6-25).
6-143
-------
TABLE 6-24. EFFECTS OF OXIDANTS (OZONE) IN AMBIENT AIR ON GROWTH, YIELD, AND FOLIAR INJURY IN SELECTED PLANTS
Plant species
Lemon
Orange
Grape,
Zinfandel
Corn, sweet
Bean, white
Tobacco, cultivar
Bel W3
Tobacco, cultivar
Bel W3
Cotton, cultivar
Acula
Potato , 4
cultivars
Potato, cultivar
Haig
Oxidant
cone. , ppm
>0.10
>0.10
>0.25
0.20-0.35
>0.08
0.02-0.03
>0.05
Ambient
>0.05
0.15
Duration of exposure
Over growing season
148 hr/mo average from
March-October, 254 hr/mo
average from July-September
Often over May-September
growing season
Hourly maximum for 3 to 4 days
before injury
9 hr
6 to 8 hr
Often over growing season
Over growing season
326 to 533 hr (2 yr)
3 consecutive days
Plant response, %
reduction from control
32, yield
52, yield; leaf drop and other effects
54, yield; other reductions found
12, yield (first year)
61, yield (second year);
increased sugar content
47, yield (third year)
67, injury (10 cultivars, 5 unmarketable)
18, injury (13 cultivars)
1, injury (11 cultivars)
Bronze color, necrotic stipple,
premature abscission
Minimal injury
22, fresh wt, top
27, fresh wt, root
7-20, lint + seed (3 locations, 1972)
5-29, lint + seed (3 locations, 1973)
34-50, yield (2 years for 2 cultivars)
20-26, yield (1 year for 2 cultivars)
95, injury; leaf area covered
Location of study
California
California
California
California
Ontario, Canada
Ohio
North Carolina
California
Maryland
Delaware
Greenhouse studies.
Source: National Research Council (1977).
-------
TABLE 6-25. EFFECTS OF AMBIENT AIR IN OPEN-TOP CHAMBERS, OUTDOOR CSTR CHAMBERS, OR GREENHOUSES ON GROWTH AND YIELD OF SELECTED CROPS
V
Plant species
Tomato
(Fireball 861 VR)
Bean
(Tendergreen)
Snap bean (3 cultivars:
Astro, BBL 274, ~BBL
290)
Soybean (4 cultivars:
'Cutler, York, Clark,
Dare)
CTi
i
i— «
45.
cn
Forbs, grasses,
sedge §
Snap bean
(Gallatin 50)
(BBL 290)
(Astro)
03 concn. , •
ppm
0.035
(0.017-0.072)
0.041
(0.017-0.090)
0.042
>0.05 ' "
0.052
0.051
0.035
>0.05
>0.05
>0.05
Exposure duration
99 day average (0600-2100)
43 day average (0600-2100)
3 mo average (0900-2000)
31* of hr (8:00 a.m. to
10:00 p.m. ) from late
June to mid-September
over three summers; 5%
of the time the concen-
tration was above 0.08 ppm
1979, 8 hr/day average
1000-1800), April-
' September
1980, 8 hr/day average
(1000-1800), April-
September
1981, 8 hr/day average
(1000-1800), April -
. Sepbember
Average 170 hr over 60
days exposure (1972-1974)
(6 crops)
Average 170 hr over 60
days exposure (1972-1974)
(6 crops)
Average 170 hr over 60
days exposure (1972-1974)
Percent
reduction Location
from control of study
33d, fruit fresh New York
wt
26d pod fresh wt;
24 , number of pods
1, pod wt Maryland
20d, seed wt; 10d, Maryland
wt/100 +2, % pro-.
tein content, 4%
oil content
32, total above Virginia
ground biomass
20, total above Virginia
ground biomass
21, total above
ground biomass
+5, pod fresh Maryland
wt
14 , pod fresh Maryland
wt
3, pod fresh Maryland
wt
Monitoring Calibration
method method
Mast NBKI
Mast NBKI
Not given Not given
Mast NBKI, known
03 c""rce
Chem • Known 03
source,
UV
Chem
Mast 1% NBKI,
Chem
Mast 1% NBKI,
Chem
Mast 1* NBKI,
Chem
Fumigation
facility Reference
OT MacLean and
Schneider
(1976)
OT
OT Heggestad
and Bennett
(1981)
OT Howe 11 et
al. (1979)
Howe 11 and Rose
(1980)
OT Duchelle et
al. (1983)
OT Heggestad et
al. (1980)
OT Heggestad et
al. (1980)
OT Heggestad et
al. (1980)
(6 crops)
-------
TABLE 6-25 (cont'd). EFFECTS OF AMBIENT AIR IN OPEN-TOP CHAMBERS, OUTDOOR CSTR CHAMBERS, OR GREENHOUSES ON GROWTH AND YIELD OF SELECTED CROPS
O-)
1
1—"
-Fa
CTl
03 concn. ,
Plant species ppro
(Astro) >0.05
Snap bean >0.05
(Gal latin 50)
(BBL 290) >0.05
(BBL 274) >0.05
Sweet corn >0.08
(Bonanza)
(Monarch Advance) 0.08
Exposure duration
Average 160 hr over 60
days exposure (1975-1976)
(2 crops)
Average 160 hr over 60
days exposure (1975-1976)
(2 crops)
Average 160 hr over 60
days exposure (1975-1976)
(2 crops)
Average 160 hr over 60
days exposure (1975-1976)
(2 crops)
58% of hr (0600-2100)
between 1 July and
6 September
Percent
reduction
from control
6, pod dry wt
+1, pod dry wt
10, pod dry wt
22 , pod dry wt
9 .. ear fresh wt;
10 , no. seeds/ear
28^, ear fresh wt;
42 *, no. seeds/ear
Location Monitoring
of study method
Maryland Mast
Maryland Mast
Maryland Mast
Maryland Mast
California Mast
Calibration
method
1% NBKI,
Chem
1% NBKI
Chem
1% NBKI
Chem
1% NBKI
Chem
UV
Fumigation
facility
OT
OT
OT
OT
OT
Reference
Heggestad et
al. (1980)
Heggestad et
al. (1980)
Heggestad et
al. (1980)
Heggestad et
al. (1980)
Thompson et
al. , 1976a
Chem = chemiluminescence; Mast = Mast oxidant meter (coulombmetric); UV = ultraviolet spectrometry.
NBKI = neutral buffered potassium iodide; UV = ultraviolet spectrometry.
COT = open-top chamber; CSTR = continuous stirred tank reactor.
dSignificant at p = 0.05.
Total above ground biomass, 3 yr average; NF and open plot versus CF a significant at p = 0.05
-------
Ambient ozone (mean concentration of 0.035 ppm daily average, 6:00 a.m.
to 9:00 p.m.) induced a significant yield reduction (33 percent) in tomato
(Maclean and Schneider, 1976). During the 99-day experimental period, the
1-hr average ozone concentration exceeded 0.08 ppm for 11 percent and 0.10 ppm
for 6 percent of the daylight hours. Also, the yield of green beans was
reduced (26 percent) by ambient ozone (mean concentration of 0.041 ppm daily
average, 6:00 a.m. to 9:00 p.m.) (Maclean and Schneider, 1976). The average
yield of four soybean cultivars exposed to ambient ozone in Maryland was
reduced an average of 20 percent over the 3-yr period (Howell et al., 1979).
Over the study period, the ambient ozone concentration exceeded 0.08 ppm and
0.10 ppm 1.8 and 0.9 percent, respectively, of the daylight hours (8:00 a.m.
to 8:00 p.m.). In Riverside, CA, the ambient ozone reduced the yield of two
sweet corn cultivars 9 and 28 percent, respectively (Thompson et al,,, 1976a).
The ozone concentration during the daylight hours (6:00 a.m. to 9:00 p.m.)
exceeded 0.08 ppm and 0.12 ppm for 58 and 39 percent of the time, respectively.
The growth of a mixture of forbs, grasses, and sedges at Big Meadows, Shenandoah
National Park, VA, was reduced 32, 20, and 21 percent for the years 1979,
1980, and 1981, respectively (Duchelle et al., 1983). At the study site, the
mean ozone concentration (11:00 a.m. to 6:00 p.m.) for the period April through
September averaged 0.052, 0.051, and 0.035 ppm over the 3-yr period, respec-
tively. For the same time periods, the total ozone dose was 73.4, 74.2, and
50.5 ppm-hr with 1218, 790, and 390 hr, respectively, when the ozone concentra-
tion exceeded 0.06 ppm (11:00 a.m. to 6:00 p.m.). The impact of ambient ozone
on the yield of several bean cultivars was studied for several years in Maryland
(Heggestad et al., 1980). There were seasonal and yearly variations in the
impact of ambient ozone on bean yield, which ranged from a 5 percent increase
above the control to a 22 percent yield decrease. In each study, there were
extended periods when the ozone concentration exceeded 0.05 ppm.
Early ambient air studies in California, in 1976 and 1977, incorporated
multiple locations situated along an ambient 0, gradient in a portion of the
South Coast Air Basin, where phytotoxic pollutants other than 0, occur only at
extremely low concentrations (Oshima et al., 1976; Oshima et al., 1977a).
These studies used a modified cumulative 0- dose (sum of hourly averages above
0.10 ppm for the exposure period, ppm-hr) as a summary exposure statistic
(Table 6-26). The dose calculation was further modified in the 1977 study by
including only those pollutant concentrations present during daylight hours.
6-147
-------
TABLE 6-26. EXPOSURE-RESPONSE FUNCTIONS RELATING OZONE DOSE TO PLANT YIELD3
Dose, ppm-hr, for
predicted 10% .
Plant species Yield equation yield reduction Reference
Alfalfa0 y = 162.4 - 1.5 x Dose 10.8 Oshima et al. (1976)
(Moapa 69)
Tomatoc - y = 9.742 - 0.23 x Dose 4.2 Oshima et al. (1977a)
(6718VF) .
Potato y = 1530 - 15.8 x Dose 9.7' Foster et al. (1983b)
(Centennial Russet)
Beanc y = 306.7 - 33.33 x log x Dose >51.6 Oshima (1978)
(Red Kidney)
The studies were conducted in California and plants were exposed to ambient 03.
For alfalfa and tomato, the hourly averages above 0.10 ppm were summed to complete the seasonal dose. For
potato and bean, hourly average 03 concentrations for the duration of the study were summed to complete the
seasonal dose.
The original equation was based on pphm-hr, but for this table the regression coefficient was converted
to ppm-hr for consistency with data in the rest of the chapter.
-------
In the 1976 study, the lowest dose was 2.64 ppm-hr, the equivalent of 0.11 ppm
for 264 hr (1.26 hr/day) of the 5040-hr season. The highest dose was 55.52
ppm-hr, the equivalent of 0.111 ppm for each hour of the 5040-hr season.
Alfalfa yield was reduced (10 percent) at a seasonal dose of 10.8 ppm-hr.
Tomatoes were substantially more sensitive than alfalfa. The tomato yield was
reduced at a seasonal dose of 4.2 ppm-hr.
Oshima (1978) designed and constructed an exposure facility (modified
CSTR) by using chambers enclosed by a Teflon film to minimize environmental
alterations. The exposure system used proportional charcoal filtration of
ambient air, thus retaining the ambient exposure properties at several pollu-
tant concentrations. Ozone concentrations were expressed as cumulative dose
(sum of hourly averages for the exposure period, ppm-hr) (see Sections 6.2.2.1
and 6.4.3.3). Both Oshima (1978) and Foster et al. (1983b) (Table 6-26) were
able to demonstrate yield losses in pot-grown red kidney bean and Centennial
Russet potato, respectively, at low concentrations of ambient 0,. Potato
yield was reduced (10 percent) at a seasonal dose of approximately 9.7 ppm-hr,
but a substantially higher dose (>51.6 ppm-hr) was required to impact the
yield of red kidney beans. Many of the ambient concentrations used in both
studies were equivalent to ambient concentrations in cleaner regions of Cali-
fornia and in the eastern United States.
Several studies have measured various plant effects and attempted to
describe associations between ambient 03 and 0~-injury symptoms or yield
responses. Oxidant-induced changes in forest ecosystems of California, Vir-
ginia, and Utah are discussed in Chapter 7. Some specific references to these
and other areas follow. Increasing 0- sensitivity of ponderosa pine has been
correlated with insect-induced mortality (Cobb and Stark, 1970). Over a 3-yr
period, 24 percent of 150 study trees died, 92 percent of which exhibited
severe foliar 03 symptoms. No trees classed as healthy or slightly symptomatic
died. In a mixed-conifer stand in the San Bernardino Mountains, radial growth
for the 30-yr period 1945 to 1975 decreased an average of 34, 1, and 4 percent
in areas with severe, moderate, and no injury, respectively (Kickert et al.,
1977). Concentrations of 0, that "commonly exceeded 0.10 ppm" were associated
with foliar injury and defoliation.
Reduced growth of 0~-sensitive eastern white pine appears to be attribut-
able to reduced foliar biomass, which results from shortened needles and
premature needle loss (Mann et al., 1980). Ozone reduced annual radial growth
6-149
-------
of the trees studied by 50 percent. The reduced foliar biomass and foliar
symptoms were associated with several episodes of 0- above 0.08 ppm. White
pines exhibiting relatively severe symptoms (chlorosis, tipburn, short needles,
premature defoliation) experienced a steady decline in average ring width (71
percent over 15 years) and a loss in capacity for recovery (Mclaughlin et al.,
1982). The annual radial growth of eastern white pine trees exhibiting symptoms
of 0~ stress was 28 percent less than that of trees exhibiting few or no
symptoms (Benoit et al. , 1982). The data of Mclaughlin et al. (1982) and
Benoit et al. (1982) should be used with caution, however, since the studies
used small sample sizes and the radial increment data were not standardized
for tree age. Field studies in the San Bernardino National Forest in California
showed that during the last 30 years ambient 03 may have reduced the height
growth of ponderosa pine by 25 percent, annual radial growth by 37 percent,
and the total volume of wood produced by 84 percent (Miller et al., 1982) (see
Chapter 7).
The research presented in this section demonstrates that ambient 03 in
many areas of this country can reduce plant yield. Although the most severe
effects appear to occur in the South Coast Air Basin and the San Bernardino
Mountains of California, areas with high ambient 0., concentrations, other
agricultural areas in the nation are impacted as well. Data presented in the
1978 criteria document (U.S. Environmental Protection Agency, 1978) suggested
that ambient 0- reduced yields for orange (54 percent), grape (47 to 61 per-
cent), and cotton (5 to 29 percent). Also, the yield of potatoes growing in
the eastern United States was reduced 20 to 50 percent by ambient 0~. More
recent research has indicated that similar yield reductions are still occurring
throughout the country as the result of ambient 0- exposures. Recent open-top
chamber studies have demonstrated losses in tomato (33 percent), bean (26
percent), soybean (20 percent), snapbean (0 to 22 percent), sweet corn (9
percent), several tree species (12 to 67 percent), and forbs, grasses, and
sedges (9 to 33 percent). Still other chamber studies have shown yield reduc-
tions in potato (42 percent) exposed to ambient photochemical oxidants. The
use of chemical protectants such as EDU has demonstrated yield losses in
potatoes ranging from 2 to 31 percent. Correlations of plant yield with
ambient 0, concentrations based on either an 0- gradient or differential
cultivar or species sensitivity have been used to predict ambient yield losses
in alfalfa (53 percent), tomato (22 percent), and ponderosa and white pines.
6-150
-------
6.4.3.3 Exposure-Response Relationships (Empirical Models)--Empirica1 exposure
response models are mathematical functions that describe a relationship between
pollutant exposure and a biological response. These models are very useful
because the entire relationship defined between the range of exposures is
quantitatively described. This desirable property differentiates the models
from the results of descriptive designs described in Section 6.4.3.2. In
addition, empirical models are useful as research tools because they suc-
cinctly summarize relationships in the form of an equation.
Empirical response models describing plant yield losses from (L have two
major uses that are distinctly different in theory and requirements.
1. Models are used for crop production forecasting. The unit used
in the forecasts is yield per unit land area. Because this is
essentially a biological forecast, errors introduced from
aggregative methods and the exclusion pf environmental, cultural,
and edaphic variables must be dealt with if model estimates are
to be reliable.
2. Models are used to interface biological systems with economic
models. The units used as a measure of effect in an economic
model are monetary (profit and loss). These models are driven
by economic variables such as input and output substitutions;
supply, demand, and associated price fluctuations; and regional
linkages. Problems of aggregation methods and impacts of
economically important variables are considered in terms of the
economic units. Errors introduced by aggregation and exclusion
of environmental variables also affect the results obtained by
economic models.
The development of empirical models is the first and the least complex
step in their use. It is the application of these models that is most apt to
be misunderstood.
The available empirical models were developed by using various exposure
techniques ranging from ambient gradients to highly controlled laboratory
exposures; therefore they have different constraints on their application.
Additionally, until the emergence of NCLAN (National Crop Loss Assessment
Network) as a multisite effort to develop credible crop-loss assessments, no
organized effort to standardize developmental methodology had occurred. The
NCLAN program represents the first organized effort to establish defensible
crop-loss estimates on a national scale.
Only one empirical model was discussed in the dose-response section of
the 1978 criteria document (U.S. Environmental Protection Agency, 1978). The
6-151
-------
Heck and Tingey (1971) injury model was used to derive tabular and graphic
data predicting CL concentrations for specific amounts of foliar injury for a
number of species. Most other discussion revolved around the limiting-value
concept used to relate 0- concentrations from the existing data base. Many
empirical models have been developed since the 1978 air quality document was
published, and such models have expanded to the point that they are commonly
used as tools in most areas of air pollution research.
There are different categories of empirical exposure-response models.
Physiological models generally are used as research tools to summarize relation-
ships and to provide a quantitative means of comparing responses. Injury models
predict leaf response at various levels of exposure, and growth models define
biomass accumulation, canopy development, and growth of reproductive organs.
Yield response models focus on the economically or biologically essential por-
tion of plant growth.
6.4.3.3.1 Physiological models. This section is included to provide an
example of the uses of physiological models in basic research, which is the
primary area of their application. Physiological response models are used as
effective research tools for summarizing relationships or allowing comparisons
among species (Tingey et al., 1976b; Coyne and Bingham, 1981). The slope of
linear models offers a convenient means for comparison of plant species or
populations within a species. Physiological processes are particularly amenable
to quantification with functions. Use of these response models fulfills ob-
jectives quite different from those fulfilled by the predictive models required
for yield-loss estimates.
6.4.3.3.2 Injury models. Injury models estimate the magnitude of foliar
injury incurred from pollutant exposures or, in one case, the concentration of
pollutant from the degree of injury (Table 6-27). These models have been used
to compare air quality in different geographical areas (Goren and Donagi,
1980; Naveh et al., 1978). Heck and Tingey (1971) developed a model that
would estimate the CL concentration required to cause specific amounts of
foliar injury (Table 6-27). This model was the source of tabular and graphic
data presented in the dose-response section of the previous ozone criteria
document (U.S. Environmental Protection Agency, .1978).
A major contribution to the evolution of injury models was the model
developed by Larsen and Heck (1976). They presented a mathematical model
based on the assumption that percentage leaf injury was distributed lognormal-
ly as a function of pollutant concentration for a specific exposure duration.
6-152
-------
TABLE 6-27. SUMMARY OF MODELS DESCRIBING THE RELATIONSHIP BETWEEN FOLIAR INJURY AND OZONE EXPOSURE
Model
Plant species
Reference
en
i
en
CO
1. y = a + bx
y = injured leaves, area (%)
x = ozone index (ppb x hr)
a = -3.5 (winter), -0.38 (summer),
-1.85 (fall)a
b = 0.0037 (winter), 0.0016 (summer),
0.0015 (fall)3
2. P = P.(l-e"kt)
P = % injured leaves at time t
P. = equilibrium % of injured leaves
k = constant determined by least squares
3. C = A0 + AX I + A2/t
C = ozone concentration
Ao> Ai» A2 = regression coefficients
I = percent foliar injury
t = time of exposure
4. Z = [-ln(Mghr) - p(ln(t)) + (n(C)]/ln(Sg)
Z = no. of standard deviations that the
percentage of injury is from the
median
C = ozone concentration
t = exposure duration
Mghr = geometric mean concentration
Sg = standard geometric deviation
p = slope of the line on a logarithmic
scale.
5. Model 5
Probit (y) =1.3 ln(c) + 0.49 ln(d) + 0.77
where
c = concentration in ul/1
d = duration in hr
y = % leaf surface injured
Tobacco Bel-W3
Winter R2 = 0.94
Summer R2 0.98
Tobacco Bel-W3
No correlation coefficient
available
Selected species
Range of R2 = 0.85 to 0.35
Selected species
Range of R2 = 0.96 to 0.58
Soybean cv. Hodgson
R2 = 0.84
Goren and Donagi (1980)
Naveh et al. (1978)
Heck and Tingey (1971)
Larsen and Heck (1976)
Pratt and Krupa (1981)
-------
TABLE 6-27 (cont'd). SUMMARY OF MODELS DESCRIBING THE RELATIONSHIP BETWEEN FOLIAR
INJURY AND OZONE EXPOSURE
I
I—»
en
6.
Model
Plant species
Model 5
PIF = 0.2174 + 2.2457 ln(c) + 2.1378 ln(t)
where
c = concentration in ul/1
t = duration in hr
PIF = Probit mean proportion of injured
foliage/plant
Short-term controlled fumigations
S = n (ln(D)) + K
where ,
D = (Cm/n x t) and S is in the range
0 to 1 .
S = plant injury degree
C = concentration in ppm
t = exposure duration in hr
m = constant
n = constant
K = constant
S = 0.278 ln(D) + 0.999
Black cherry
R2 = 0.77
Morning glory
R2 = 0.97
Morning glory
R2 = 0.70
Ambient conditions
S = n ln(D) + A ln(D')) + K1
where /
D = I.C.m/n .
S = plant injury degree
C. = hourly average concentration at
the ith hour in ppm
A (In(D')) = contribution to the injury
on the current day due to the
effects of oxidant dosage up
to the previous day
A = constant
K1 = constant
S = 0.278 In(D-) + 0.041 [ln(D. ,) + ln(D. „) + ln(D. -)] + 1.872
J J J J
Reference
Davis et al. (1981)
Nouchi and Aoki (1979)
Nouchi and Aoki (1979)
Half the accumulating sum of average hourly 03 concentration between the first value > 40 ppb and the
last value <_ 40 ppb.
Plant injury degree = (1% damaged leaf per leaf)/I area of the leaves that can be damaged to the maximum
degree.
-------
This model also has been used for black cherry (Davis et al., 1981) and Hodgson
soybean (Pratt and Krupa, 1981). Both groups of investigators modified the
Larsen and Heck model slightly by using a probit transformation of the depen-
dent variable.
Nouchi and Aoki (1979) developed injury models for both short-term con-
trolled exposures and long-term ambient exposures with morning glory (Table
6-27). They recognized that foliar injury did not have a linear relationship
with the conventional dose statistic (concentration x time) and developed a
powered dose (dose raised to some power) for the acute exposure model. Further,
Nouchi and Aoki included a factor in the ambient model that incorporated the
time-dependent contribution of previous 0- exposures and modified the dose
expression to account for the long-term variable exposures that characterize
ambient 0, episodes. These investigators were the only group that attempted
to account for the effects of previous 0, exposures on foliar injury in their
model.
6.4.3.3.3 Growth models. Only a few empirical growth models quantify CL-
induced alterations in biomass accumulation and assimilate partitioning.
Oshima et al. (1978; 1979) developed growth models for parsley and cotton and
later refined the cotton model (Oshima and Gallavan, 1980). Growth models are
used primarily for research purposes and are included in this report only as
an example to indicate progress in quantifying 03 growth responses.
6.4.3.3.4 Yield and loss models. Yield models are the most sought after and
most difficult models to develop. These models are necessary for estimates of
production and economic loss because they relate yield directly to pollutant
exposure. The number and quality of yield models is increasing rapidly because
of increased interest and because of the NCLAN program. Existing models are
summarized in Table 6-28. Additional discussion of actual yield responses
that were derived from many of these studies is presented in tabular and
graphic form in Section 6.4.3.2.
Oshima and his coworkers developed predictive models for estimating yield
losses from 0., in California. Using data from plots along an ambient 0,
gradient in southern California, Oshima et al. (1976) developed both yield and
yield-loss models for a clone of Moapa 69 alfalfa (Table 6-28). Multiple-
regression techniques were used to test the relative contributions of 0- dose
and meteorological variables to changes in alfalfa yield. Ozone was determined
to be the greatest contributor to yield variation when compared to the contribu-
tions of the other tested variables. The 0.,-yield function was then transformed
6-155
-------
TABLE 6-28. SUMMARY OF MODELS OF OZONE YIELD AND LOSS
Model
Crop
Reference
1. (a) Total fresh wt function
y = a + bx y = fresh wt (g/plant)
y = 162.4 - 0.015x a = intercept
x = ozone dose
(pphm-hr > 10 pphm)
(b) Loss function
Transformed from l(a) by % loss = (a - wt)/a x 100
% Loss = -1.068 • 10-4 + 9.258 • 10-3x
x = ozone dose (pphm-hr > 10 pphm)
2. (a) Marketable fruit
y = [sin(-0.0076x + 84.2816)]2
y = % fruit marketable USDA minimum size
x = ozone dose (pphm-hr > 10 pphm)
en
i
(b) Yield function
y = 9.742 - 0.0023X
y = container yield based on USDA fruit
size and packing configuration
x = ozone dose (pphm-hr > 10 pphm)
(c) Loss function
Transformed from 2(b) by % loss = (a - container yield)/a x 100
% Loss = 0 + 0.0232x x = ozone dose (pphm-hr > 10 pphm)
3. (a) y = a + bx y = yield (varies with crop)
x = ozone exposure in seasonal 7 hr/day
mean ozone concentration (ppm)
a = intercept
b = slope
2 y = yield (varies with crop)
x = ozone exposure in seasonal 7 hr/day
mean ozone concentration (ppm)
a = intercept
b0 and b1 = regression coefficients
4. (a) Linear yield function
y = b0 + bjX y = crop yield (g/plant)
x = ozone exposure in seasonal 7 hr/day
mean ozone concentration (ppm)
b0 = intercept
bl - slope
(b) y = a + b0x +
Alfalfa cv Moapa 69
R2 =0.68
Tomato VF 6718
R2 = 0.85
R2 = 0.62
Selected crops
R2 statistics are
not available
Selected crops
Range of R2 =
0.99 to 0.65
Oshima et al. (1976)
Oshima et al. (1977a)
Heagle and Heck (1980)
Heck et al. (1982)
-------
TABLE 6-28 (cont'd). SUMMARY OF MODELS OF OZONE YIELD AND LOSS
Model
Crop
Reference
(b) Plateau- linear yield function
y = b0 if x < f
y = (b0 - bjf) + bjx
Range of R2 =
0.99 to 0.94
(c) Loss function
y = 100 bt (0.025
a
cr>
i
en
if x > f
y = crop yield (g/plant)
x = ozone exposure in seasonal 7 hr/day
mean concentration (ppm)
f = threshold 7 hr/day mean concentration (ppm)
b0 = intercept
bj = slope
- x)
y = % yield reduction
bx = regression coefficient from function 4(a)
a = predicted yield (g/palnt) from function 4(a)
at 0.025 ppm 7 hr/day mean ozone concentration
x = ozone exposure in seasonal 7 hr/day mean ozone
concentration
5.
Weibull function
y = a exp [- (x/a) ] + e
y = yield
a = hypothetical maximum yield at 0 ozone
x = ozone dose in seasonal 7 hr/day mean ozone concentration (ppm)
o = the ozone concentration when yield is 0.37 a
c = dimensionless shape parameter
Selected crops
Parameters esti-
timated from
empirical data
Heck et al.
Heck et al.
(1983)
(1984a,b)
(a) Tuber weight yield function
y = a + bx y=% tuber yield (g/plant)
y = 1530 - 15.8D D = ozone dose (ppm-hr)
(b) Tuber number yield function
y = 34.3 - 0.318D y = tuber yield (number/plant)
D = ozone dose (ppm-hr)
(c) Plant dry matter function
DM = 382 - 3.83D DM = total dry matter (g/plant)
D = ozone dose (ppm-hr)
Potato cv Centennial
Russet
R2 = 0.77
R2 = 0.62
R2 = 0.73
Foster et al. (1983b)
-------
to a predictive loss model using the intercept as the zero-loss reference
value. Similar techniques were used to develop an 0~-loss model for fresh
market tomatoes (Table 6-28) (Oshima et al., 1977a). This model incorporated
the unique feature of transforming plant yield to economic packing container
units. Tomato fruit yield was predicted as percentage loss in marketable con-
tainer units (flats or lugs) based on U.S. Department of Agriculture fruit
size categories. The inclusion of the marketing criteria sharply increased
the proportion of loss.
Heagle and Heck (1980) developed both linear and quadratic yield models
for cultivars of field corn, winter wheat, soybeans, and spinach (Table 6-28).
The models were derived from open-top chamber experiments and used a seasonal
7 hr/day mean CL concentration to characterize Cu exposures. These models
were the precursors of those developed by NCLAN.
The first published yield models produced by NCLAN (Heck et al., 1982)
were presented as either linear or plateau-linear functions (Table 6-28). The
plateau-linear function combines two linear functions; the first having a
slope of zero, depicting no response, and a second having a measurable slope.
The intersection of the two functions can provide an estimate of a threshold
value. Yield functions were developed from open-top chamber data obtained by
the regional research laboratories participating in the NCLAN program. Each
model was developed with a standardized method monitored by a quality-control
program. Yield-loss models were developed for cultivars of corn, soybean,
kidney bean, head lettuce, peanut, spinach, turnip, and wheat. Some models
included in Heck et al. (1982) were generated from earlier experiments that
involved the corn, wheat, and spinach models of Heagle et al. (1979a, 1979b,
1979c).
Recently, Heck et al. (1983a, 1984a) used a three-parameter Weibull func-
tion to model NCLAN yield losses (Table 6-28). The Weibull function was
selected because it has a flexible form that covers the range of observed
biological responses; the form of the Weibull is biologically realistic; the
model parameters have clear and straightforward interpretations; and it offers
a method of summarizing species responses by developing a common proportional
model (Raw!ings and Cure, 1985). The Wei bull modeling approach was subse-
quently used with NCLAN data previously modeled with linear, plateau-linear,
or quadratic functional forms (Heck et al., 1983a).
6-158
-------
Larsen and Heck (1984) have recently adapted their lognormal plant injury
model (Larsen and Heck, 1976) to estimate the impact of ozone on crop yield.
The model uses the "effective mean" CL concentration (see Section 6.2.2) to
characterize the exposure. The model assumes that equal effective means cause
equal impacts and assumes that it does not matter whether the mean is the
result of constant or varying ozone concentrations. Similarly, the effects
are cumulative with time, which means that the same effects would result from
continuous or discontinuous exposures. The exponent on the concentration term
for effective mean was derived from injury studies and then applied to yield
studies without validation of its applicability to yield. Also, the exponent
probably varies with cultivar and environmental conditions. A comparison of
the log-normal and Wei bull models with the same data set showed that the
models produced similar yield reduction curves. In lognormal model did not
work well, however, with individual plot means; and the lowest crop reduction
value in the data set must be greater than zero with that model because of
data transformation needs. The Weibull does not have these same constraints.
Foster et al. (1983a,b) produced yield and plant dry weight functions for
Centennial Russet, an extremely sensitive potato cultivar. These models were
developed using an ambient exposure facility composed of a series of large
Teflon® chambers with 0- exposure controlled by proportional filtration of
ambient 0~.
A multipoint crop-loss technique was developed (by P. Teng, as reported
in Benson et al., 1982) and used to assess the impacts of 0-. Previously the
multipoint models had been used to predict biotic yield losses (resulting from
biotic pathogens) but the authors further refined this technique by summing
daily multipoint loss models over a season to arrive at a seasonal loss for
alfalfa (Table 6-29). When single harvest crops such as corn, wheat, and
potato were used, the authors divided the seasonal exposure into 12 time steps
and regressed final harvest on the exposure steps. This application of the
model was seriously flawed, however, because only one time series of 0- expo-
sures was used. Separating total 0- exposure into several time steps creates
a model with several colinear variables. The estimated coefficients
of these variables are unstable. The alfalfa model is important because 0-
exposures were represented by multiple variables that indicate specific expo-
sure periods, which may more closely approximate the ambient patterns of
exposure than do the single summary statistics used by other researchers.
6-159
-------
TABLE 6-29. SUMMARY OF CROP-LOSS MODELS
Model
Loss criteria
General models
l. y=f(xt1)...
2. Net yield reduction is:
Functional models
^ 1. Alfalfa
i y = ax + bx2 + ex3
cr>
O
Range of R2 - 0.99 to 0.13
2. Corn
y = axj + bx2 Ix12
R2 = 0.87
y = proportion of yield reduction
x. = dose parameter at time t.
n
I ydt,
i
where: dt = time step
n = maximum number of growing days
y = daily yield loss (fresh wt)
x = I hourly averages for 1 day
a to c = regression coefficients
y = yield loss based on 100 kernel wt
*! to x12 = ozone summary statistics for
periods 1 to 12 calculated
as:
N
I [(I hi/n)24]
1 = 1 _,
N
where: N = the number of days in a period (7 days)
hi = ozone concentrations
n = number of hours for which there are
ozone concentrations
a to 1 = regression coefficients
NAC
NAC
Loss =10-
Bl'omass at site
Biomass at control site
• = in - 100 kernel yield for (x)
100 kernel yield for control
-------
TABLE 6-29 (cont'd). SUMMARY OF CROP-LOSS MODELS
Model
Loss criteria
3. Wheat
y = axt + bx2...gx7
R2 = 0.95
CD
I—»
cn
i—"
4.
Potato
y = axj
R2 = 0.93
bx2
cx3 + dx4
exs
y = yield loss based on 100 seed wt
Xj to x7 = ozone summary statistics for
periods 1 to 7 calculated
as:
N
I [(I hi/n)24]
i = 1 ,
N
where: N = the number of days in a
period (7)
hi = ozone concentrations
n = number of hours for which
there are ozone concentrations
a to g = regression coefficients
y = yield loss based on tuber wt/plant
to xs = ozone summary statistics for
periods 1 to 5 calculated
as:
N
I [(Z hi)]
. =in- 100 seed yield for (x)
100 seed yield for carbon
filtered treatment
i n tuber wt yield for (x)
tuber wt yield for
control treatment
where:
N
N = the number of days in a period (14 days)
hi = hourly ozone concentrations in 1 day
to e = regression coefficients
aNA = Not available.
Source: Benson et al. (1982).
-------
Three kinds of models have been used to describe yield and loss: linear,
plateau-linear, and Weibull functions. These empirical models are intended to
describe the behavior of plants in the absence of a known functional relation-
ship between 0- concentration and yield. Each type of model has strengths and
limitations.
The class of linear models, including straight-line and quadratic equations,
is very flexible because it can take on a large variety of shapes and can be
used to approximate other functions and because statistical methods for compu-
ting confidence limits are available (Draper and Smith, 1966). Straight-line
models are limited because they allow no curvature and they do not allow
threshold levels below which no yield loss occurs. Quadratic models allow
curvature and gradual changes in slope, but like straight-line models, they do
not allow plateau shapes or thresholds. They can, however, describe situa-
tions in which low levels of a pollutant stimulate growth but higher levels
cause yield reductions.
Three nonlinear models (plateau-linear, lognormal, and Weibull) have been
used in attempts to describe situations in which response to 0- has a threshold.
Statistical theory for nonlinear models is not as well developed as that for
linear models and, consequently, confidence bands are not usually fitted to
nonlinear models. The plateau-linear model incorporates a threshold value but
does not allow curvature of any increase in yield followed by a decrease. The
Weibull model can take on a plateau shape followed by curved gradual decreases.
Several types of yield models have been used to describe and ultimately
predict expected yield loss from (L exposure. This need provides an important
constraint for the models; when making predictions it is important that the
2
exposure-response model have more than just a high R . The fitted equation
should not show systematic deviation from the original data points. Also, the
predicted response should not over- or underestimate the response at any
particular concentration-duration combinations. Examples of linear and plateau-
linear models and their fit to the data are shown in Figures 6-21 and 6-22.
Examples of the Weibull curves are shown in Section 6.4.3.2.1.1.
While linear equations were adequate for some cultivars of some crops,
nonlinear responses provided a better fit to the data for other crops (Figures
6-21, 6-22). The linear curves for soybean, peanut, and spinach all show good
2.
fit to the data and have high R . In the case of the soybean (Argonne) and
kidney bean data, however, the treatment means show a curvilinear relationship
6-162
-------
450
400
n
o>
i 350
"5
E
o) 300
I
O
O
ID
HI
(A
250
200
150
0.00
T
DAVIS SOYBEAN
RALEIGH, 1981
y = S34.S • 3988.6 x 0, + 10960 x
3000
2500
CO
Ł.
O)
Q
LU
2000
1500
1000
(B)
— o
I I
CORSOY SOYBEAN
ARGONNE, 1980
y = 3099.3 -15135 x 03 R! = 0.975
I I I
0.05
0.10
0.15
0.20
0.00
0.05
0.10
0.15
0.20
O3 CONCENTRATION, ppm
CONCENTRATION, ppm
is
"5.
x
O)
O
_i
UJ
18
16
14
12
10
(C)
_ O
I I T
KIDNEY BEAN (CALF LIGHT RED)
BOYCE THOMPSON INSTITUTE, 1980
y - 17.44 - 35.51 x 0,
I I
160
120
80
40
PEANUT (NC-6)
RALEIGH, 1980
= 173-1046 x0
y = 142.3 if 0,<0.037
y = 184.6 -1160 x 03 If 03>0.037 R* = 0.99
• I I I
0.00
0.05
0.10
0.15
0.20
0.00
0.05
110
0.15
0.20
03 CONCENTRATION, ppm
03 CONCENTRATION!, ppm
Figure 6-21. Effects of ozone on the yields of several legume species. The 03
concentration is expressed as the 7-hr seasonal mean. The data were selected to show
examples of the goodness of fit of the equations to the data points. (A) Data and
regression equation from Heagle et al., (1983a). Each point is the mean of two plots;
the regression equation was based on the individual plot values. (B) Data and
regression equation are from Kress and Miller, (1983). Data and the curve for yield in
g/plant are also given in Heck et al. (1982). Each point is the mean of four plots; the
regression equation was based on the individual plot values. (C) Data and regression
equation are from Kohut and Laurence (1983). The same data and another straight
line regression equation are in Heck et al. (1982). Each point is the mean of three
plots; the regression was performed on the treatment means. (D) Data and the
straight line equation (1) are found in Heagle et al. (1983b), and Heck et al. (1982);
the plateau-linear equation (2) is from Heck et al. (1982).
6-163
-------
c 5
10
*E
x
o>
I 4
01
§
O
LU
(A)
I I I
WINTER WHEAT (HOLLY)
RALEIGH. 1977
y = 4.533 + 19.31 x 03 - 215.1 X 03» -
= 5.7-16x0
y = 4.9 if 0,<0.087
y = 8.2 - 38 x 0, It 0,>0.87 R' = 0.99
I I I
0.00 0.05 0.10 0.15
0, CONCENTRATION, ppm
0.20
0>
14
UJ
a
UJ
UJ ,
V) 3
(B)
I I I
WINTER WHEAT (OASIS)
RALEIGH, 1977
-O— y = 4.475 + 3.320 x 0, - 93.49 X 0,
y = 4.9-12x03 RS = 0.88
2
0.00
J_
I
I
0.05 0.10 0.15
0, CONCENTRATION, ppm
0.20
50
40
Ł 30
O)
O
O
z
W
20
10
(C)
SPINACH (VIROFLAY)
RALEIGH, 1976
y = 46.1 - 238 x 0, R» = 0.94
I I
260
240
*-
c
a
a.
x
o> 220
§ 200
a
UJ
UJ
(A
180
160
(0)
CORN (COKER 16)
RALEIGH, 1976
— y = 222.91 +331.11 »0j-3511.99 x03»
-- y = 247.8 - 260 x 0, R'= 0.65
I
I
0.00 0.05
0.10
0.15
0.20
0.00
0.05
0.10
0.15
O, CONCENTRATION, ppm
O, CONCENTRATION, ppm
Figure 6-22. Effects of ozone on the yield of several crops. The data were selected to
show examples of the goodness of fit of the equations to the data points. The O3
concentration is expressed as the 7-hr seasonal mean. (A-B) The data are from Heagle
et al. (1979c). Quadratic equations are from Heagle and Heck (1980). In Heagle and
Heck (1980) the data were presented as the yield per four plants; however, in this
figure the values were divided by four to express the yield on a per plant basis. The
other equations are from Heck et al. (1982); each point is the mean of four plots with
48 plants per plot. (C) The data are from Heagle et al. (1979b). Regression equations
are from Heck et al. (1982). Another set of straight line equations is given in Heagle
and Heck (1980). Each point is the mean of four plots with four quadrants (two to
three plants per quadrant per plot). (D) Data are from Heagle et al. (1979a) with a
correction for the yield at 0.07 ppm (personal communication from A.S. Heagle. U.S.
Dept. of Agriculture. Raleigh. NC. to D.T. Tingey. U.S. EPA, 1984). The quadratic
equation (solid line (o ))is from Heagle and Heck (1980). Data point at the
concentration of 0.07 is different from the original paper; the correction was based on
information from A.S. Heagle (personal communication to D. T. Tingey, 1984). The
straight line equation (2) is from Heck et al. (1982). In developing the quadratic
equation, the data from Heagle et al. (1979a), were divided by a factor of 1.045 to
adjust the moisture content (personal communication. A.S. Heagle to D. T. Tingey,
1984); for the linear equation the unadjusted data were used; ( A ) indicates an
adjusted treatment mean. Each point is the mean of five plots with eight plants/plot.
0.20
6-164
-------
not well described by the linear regression. With the wheat and corn data the
linear provides a poorer fit to the data than curvilinear models. Several
nonlinear models would probably fit the data; however, the Weibull was found
to fit the data and to have the desirable properties described above. It has
been selected as the best of the tested models for fitting most of the NCLAN
data (Heck et al., 1983a; Heck et al., 1984b; Rawlings and Cure, 1985).
If the equations do not fit the data well there is a tendency for the
model to yield poor predictions. This can be illustrated by calculating the
seasonal 7-hr mean (L concentration that would be predicted to cause 10 and 30
percent yield reductions (Table 6-30). In every case, the linear model pre-
dicted that a substantially lower mean 03 concentration would cause a 10
percent yield loss than would the Weibull, with the differences ranging from
17 to 57 percent. These differences indicate that great care should be exer-
cised when using only linear models to make yield predictions. The predicted
concentrations that would cause 10 percent yield loss were generally similar
among the plateau-linear, quadratic, and Weibull models. When exposure-
response functions are used to make predictions the user should ensure that
the model provides adequate fit to the data.
All the yield and loss models presented have some common weaknesses for
production forecasting. With the exception of the model developed by Teng
(Benson et al., 1982), none of the models uses a statistic that characterizes
the episodic nature of ambient exposures. The multiexposure variables used by
Teng (Benson et al., 1982) partition the seasonal exposure into discrete
periods. This accounts for some of the ambient fluctuations in 0., levels. The
temporal periods of exposure were preselected, however, and did not correspond
to natural fluctuations. Only the alfalfa model incorporated the daily varia-
tions in ambient exposures because of the nature of its yield.
6.4.3.3.5 Interpreting exposure-response models. Interpretation of exposure-
response models requires an understanding of the subjects presented in Section
6.2. The loss models presented in Tables 6-28 and 6-29 were developed by
means of a range of diverse exposure methods, exposure characterizations,
experimental designs, and reference loss criteria. Despite their enormous
differences, the models are mathematically very similar because all but the
Weibull functions used linear or multiple linear regression techniques. All
but the Weibull and quadratic models are linear functional forms, use percent
as the unit of loss and, with the exception of the model of Teng (see Benson
et al., 1982) (Table 6-29), use a single independent variable to represent 03
6-165
-------
TABLE 6-30. COMPARISON OF PREDICTED 7-hr SEASONAL MEAN OZONE CONCENTRATIONS
USING VARIOUS YIELD REDUCTION MODELS3
01
1
I—1
01
cr>
Plant
Soybean
Corsoy
Corsoy
Davis
Davis
Peanut - 1980
Peanut - 1980
Peanut - 1980
Kidney bean
Kidney bean
Wheat
(Holly)
(Holly)
(Holly)
(Holly)
Model
kg/ha = 3099.3 - 15135 03
g/plant = 15.6 exp [-(03/0. 129) 1<7°]
seed wt/m = 534.5 - 3988.6 03 + 10,960 (03)2
g/plant = 31.1 exp [-(03/0.129)0-91]
pod wt/plant = 173 - 1046 03
pod wt/plant = 142.3 if 03 < 0.037;
= 184.6 - 1160 03 if 03 > 0.037
g/plant = 148 exp [-(03/0. 186)3'20]
seed wt/plant = 17.44 - 35.51 03
g/plant = 16.5 exp [-(Oa/0.287)1' 77]
g/plant = 5.7 - 16 03
g/plant = 4.533 + 19.31 03 - 215.1 (03)2
g/plant = 4.95 exp [-(03/0.156)4'95]
g/plant = 4.9 if x < 0.087
Control 03
concentration, ppm
0.022
0.022
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.03
0.03
0.03
0.03
Concentration, ppm, for
predicted yield loss of:
10%
0.040
0.043
0.038
0.038
0.039
0.049
0.046
0.072
0.086
0.063
0.095
0.099
0.100
30%
0.077
0.076
0.070
0.071
0.067
0.073
0.073
0.165
0.164
0.128
0.129
0.127
0.126
=8.2 -38 03 if 03 > 0.087
-------
TABLE 6-30 (cont'd). COMPARISONS OF PREDICTED 7-hr SEASONAL MEAN OZONE CONCENTRATIONS
USING VARIOUS YIELD REDUCTION MODELS3
CTl
cr>
•-j
Plant
Wheat
(Oasis)
(Oasis)
(Oasis)
g/plant
g/plant
g/plant
= 4.9
Model
- 12 03
= 4.475 + 3.320 03 - 93.49 (03)2
= 4.88 exp [-(03/0.186)3'20]
Control 03
concentration, ppm
0.
0.
0.
03
03
03
Concentration, ppm, for
predicted yield loss of
10%
0.
0.
0.
068
088
093
30%
0.143
0.138
0.135
Spinach
Field
(Viroflay)
(Viroflay)
corn
(Coker 16)
(Coker 16)
(Coker 16)
g/plant
g/plant
g/plant
g/plant
g/plant
= 46.
= 41.
= 247
= 222
= 240
1 - 238 03
1 exp [-(Oa/0.129)1'99]
.8 - 260 03
.91 + 331.11 03 - 3511.99 (03)2
exp [-(03/0.221)4'46]
0.
0.
0.
0.
0.
024
024
02
02
02
0.
0.
0.
0.
. 0.
041
048
113
132
133
0.075
0.080
0.300
0.175
The sources for the linear, quadratic, and plateau-linear models are listed in the legends to Figures 6-21 and 6-22; the Weibull curves are from
Heck et al. (1983a).
-------
exposure. These models differ because they use several independent variables
that represent periods of exposure within a season. For simplicity, these
loss models can be represented by a general function:
Y = a + b(x) (6-2)
Y = yield loss;
a = the regression intercept;
b = regression coefficient; and
x = GO exposure representation.
The variable x represents CL exposure in the general model. The models presented
in Tables 6-28 and 6-29 use different statistics to represent the 03 exposure,
and, as previously stated in Section 6.2.2, these statistics can not be readily
transformed for comparison.
The y variable in the general model represents the dependent variable
(yield loss). All models utilize percentages as loss units but calculate the
loss percentages from different reference zero-loss values. Only a single
model (Oshima et al., 1977a) used percentages calculated from commercial
marketing criteria. All other models used biological yield parameters as the
basis for converting to percentage loss. The models used are the best available,
and they serve to define the relationship between 03 exposure and yield of
specific crops. These models also serve as criteria for developing simulation
models designed to generate the coefficients necessary to drive the more
sophisticated crop production models described by Holt et al. (1979); or serve
to focus research in areas required for rational crop-loss assessments (Teng
and Oshima, 1982).
Use of a loss model requires 0^ exposure data in the appropriate format,
good judgment to guide its application, and an appreciation of what the pre-
dicted value represents. The fresh-market tomato loss model (Oshima et al.,
1977a) can be used as an example.
Loss = 0.0232 (03 dose) (6-3)
This linear model predicts the loss in marketing container yield caused by
seasonal 03 exposure expressed as cumulative 03 dose greater than 0.10 ppm.
6-168
-------
This model was developed from ambient plots along an 03 gradient; therefore,
it is representative of plot-level yields. The zero reference level is the
intercept of the yield model in Table 6-29.
This model requires a cumulative dose to represent 0, exposure, Other
exposure statistics, such as the seasonal 7-hr daily average used by NCLAN,
cannot be used with this model. By calculating the 0, doses for locations of
interest, plot-level predictions can be estimated. The cumulative dose, the
exposure statistic used in this model, was derived by summing all 1-hr 0~
concentrations greater than 0.1 ppm for the 7:00 a.m.-to-9:00 p.m. period for
each day in the 4-month season.
Examples of loss calculations for three hypothetical locations are:
Location Ozone dose Predicted loss
1 500 pphm-hr 11.6%
2 1000 pphm-hr 23.2%
3 1500 pphm-hr 34.8%
The predicted values can be used either to estimate losses in tomato production
or to arrive at an estimate of economic crop loss. This distinction must be
made because the respective applications of these loss estimates require
different procedures. Ideally, the model would include an applications valida-
tion test wherein an independent data set of tomato losses at specific 0.,
doses would be compared to predicted losses from the model to determine the
precision of the estimates.
Once estimates are calculated for locations of interest, the next procedure
required is aggregation. The estimates given above represent three plots of
plants grown under different 0- exposure conditions. To represent tomato
production on a larger scale, the plot-level estimates must be aggregated to
estimate field-level production, the production from all fields in the area,
and the production from the region that includes the three fields.
If an economic crop-loss assessment is required, the inputs into the
economic model of choice are the estimates or the loss function. The three
yield-loss estimates provide the basis for an economic transformation to
profit or loss estimates, depending on the factors incorporated into the
economic model. For instance, if they are inputs into the linear-program,
representative-farm model used by NCLAN, then grower crop substitutions,
6-169
-------
alternate cultural practices, and other farm-level options would be explored
to determine predicted economic impact. The aggregation methods would be
economically derived and would probably incorporate regional supply and demand,
market price dynamics, and regional linkages. Alternative approaches such as
econometric modeling might be selected in some instances.
Use of a yield-loss model is a process that requires adhering to the
limitations and requirements of the model, having the required data necessary
to drive the model, deciding on the specific application desired, and using
the appropriate step to aggregate estimates to the organizational level required.
6.5 ECONOMIC ASSESSMENTS OF EFFECTS OF OZONE ON AGRICULTURE
Previous sections of this chapter discuss the adverse effects of ozone on
crop productivity. Evidence from the plant science literature clearly demon-
strates that ozone at ambient levels will reduce yields of some crops (see
Section 6.4.3.2.2, Biomass and Yield Responses from Ambient Exposures). In
view of the importance of U.S. agriculture to both domestic and world consump-
tion of food and fiber, such reductions in crop yields as a result of exposure
to Oo could adversely affect human welfare. The plausibility of this premise
has resulted in numerous attempts to assess, in monetary terms, the losses
from ambient 0, or the benefits of Oo control. Many of these assessments have
been performed since the 1978 ozone criteria document was published (U.S.
Environmental Protection Agency, 1978). The methodologies and resulting
estimates from these post-1978 studies, and their validity with respect to
plant science data and economic theory, are discussed in this section. Speci-
fically, a set of biological, aerometric, and economic criteria important to
the validity of such assessments will be presented. Post-1978 assessments
purporting to estimate the economic effects of 0, on agricultural production
will then be evaluated in terms of how well the assessments deal with these
biological and economic criteria. The section will also draw conclusions
concerning the likely magnitude of national-level economic effects of 0,
reductions suggested by two recent studies that have strong plant science and
economic foundations relative to any previous research.
6-170
-------
6.5.1 Economic Issues in Performing Assessments
Decision-making related to the formation of public policy centers on
perceived changes in "public welfare." In a benefit-cost analysis of ambient
air quality changes, however, regulatory actions typically do not lead to a
regulation from which all parties benefit. Virtually any air pollution control
action or regulation will disadvantage someone (e.g., polluters, consumers,
agricultural producers) in terms of perceived welfare. Thus, in evaluating a
given policy, a quantifiable measure of social welfare across all parties is
needed as a basis for judging socially desirable changes associated with
alternative pollution control actions. This also implies the need for con-
sistent measures of value for the various components of public welfare, e.g.,
the welfare of producers, consumers, and input suppliers, so that distributional
effects can be evaluated.
The theoretically consistent measures of economic welfare are "compensating
or equivalent variation." These somewhat abstract theoretical concepts define
welfare changes in terms of income equivalents as captured in the "willingness
to pay" or "willingness to accept compensation" by individuals for alternative
economic or environmental states. In practice, these concepts are calculated
as consumers' and producers' surpluses, defined as the geometric (and hence
measurable) area between the supply and demand curves and to the left of their
intersection (see Just et al., 1982, or Willig, 1976, for definitions and more
extensive discussions of the validity of this triangular area as a measure of
welfare). Consumers' and producers' surpluses approximate the utility gained
by individuals when: (1) in consuming goods, they obtain goods at a price
less than the maximum they would be willing to pay; and (2) in producing
goods, they sell at a price above the price at which they would have been
willing to supply. Most economic assessments of environmental change or other
policy options measure benefits in terms of changes in economic surplus (the
sum of consumers' and producers' surpluses) associated with supply and price
adjustments resulting from the environmental change. When the net change in
total economic surplus arising from a given policy or standard is positive,
then that policy may be viewed as beneficial (ignoring distributional aspects).
While economic surplus is an accepted measure of aggregate welfare, the
distribution of impacts across and within the consumer and producer categories
is another dimension of welfare. Several distributional issues that arise
then in an assessment of air pollution policies include:
6-171
-------
1. How are gains or losses, or both, distributed between various classes
of people (e.g., tradeoffs between consumers and producers)?
2. How are gains or losses, or both, distributed with respect to
geographical region or commodity, or among the owners of factors of
production?
3. How are gains or losses, or both, distributed within a class of
people (e.g., tradeoffs among consumers at different income levels,
or among producers with different farm sizes and endowments)?
4. How do the distributional effects change with time?
The latter two distributional concerns are currently difficult to address
because of insufficient data. The first two, however, can be measured directly
by capturing appropriate supply, demand, and resource relationships within the
economic analysis. Failure to include these relationships will result in
misleading estimates whenever: (1) prices in the markets for agricultural
inputs or outputs are sensitive to changes in yield and input usage as the
result of changes in air quality; and (2) producers and consumers adopt different
ways of adjusting to changes in air quality. In the first instance, consider
the case in which improvements in air quality result in yield increases, but
in which these improvements in turn cause price reductions. If the price
reductions are greater than the yield increases, producers could actually be
worse off than before the air quality improvement. An assessment that ignored
such price effects would then wrongly attribute benefits as accruing to the
producer when in reality the producer sustained revenue losses (and the consumer,
gains).
The second issue, with respect to producer and consumer adaptation strate-
gies, relates to the nature of the affected individual's resource, technical,
income, and environmental endowments. Specifically, the decision options that
underlie an individual's response to a given environmental or policy change
need to be identified and modeled. For example, producers can reduce their
potential losses or increase their gains from an air pollution change by
adopting different production patterns, utilizing more resistant cultivars, or
adding fertilizer or other compensating inputs. Failure to account for this
adaptive producer behavior will result in overestimates of losses experienced
by producers in the face of air quality degradation. Similarly, consumers may
substitute certain agricultural commodities in the face of relative price
6-172
-------
changes, so that the net effect of a rise in the price of a commodity as the
result of air pollution-induced supply changes may not be as severe as first
indicated. The implication of this discussion is that a properly formulated
assessment should capture the relevant behavior of economic agents, which will
then lead to more accurate predictions of agricultural supply changes associated
with 0^ changes. This will help ensure that the estimates of economic well-
being (consumers' and producers' surpluses) are valid indicators of the benefits
or costs of changes in air quality.
6.5.2 Plant Science and Aerometric Issues in Performing Economic Assessments
In addition to the correct modeling of economic responses, assessments of
damage to agricultural or other ecosystems from pollution require specific
plant science data linking pollutant levels and performance parameters of the
ecosystem in question. For agriculture, this information is represented by a
relationship between the response (changes) in crop yield and changes in
pollutant concentration or dose levels and other causal factors. The relation-
ship may be quantified directly using data generated from biological experi-
mentation, indirectly from observed producer data (actual input and yield data
from a large sample of individual producers), or from some combination of data
sources. From the standpoint of direct applicability to economic assessments,
estimates that are based on observed producer data are preferred. While some
success has been achieved in capturing the relationship between yields and 0-
by applying procedures based on producer input and yield data over relatively
homogeneous areas (e.g., see Mjelde et al., 1984), assessors in general have
had little success in directly applying such techniques across large geographical
areas, because of both data and statistical difficulties (e.g., see Adams
et al., 1982; Manuel et al., 1981). Even for those studies in which plausible
biological estimates are used, related experimental data are required initially
to formulate testable hypotheses about yields and (L levels as well as to
establish the credibility of estimates derived from producer data. As a
result, assessments of pollutant damages to agriculture rely heavily on data
from some form of biological experimentation to define the relationship between
dose and plant response.
To be of use to the economist, the form of these experimental data on
yield and pollution levels must be minimally compatible with the nature of
economic markets. This was explicitly recognized in the 1978 criteria document
6-173
-------
(U.S. Environmental Protection Agency, 1978) and in Section 6.2.5 of this
chapter, where a distinction was made between injury (defined as any identifi-
able and measurable response of a plant to air pollution) and loss (defined as
any measurable adverse effect on the desired or intended use of the plant).
Such a distinction is made because the evaluation of the economic effects of
03 exposure requires that the plant be altered either quantitatively (yield)
or qualitatively (market acceptability) so as to change its value. In some
cases exposure may result in injury, such as leaf necrosis, without affecting
yield. In addition to the need to focus on crop yield response, the closer
the experimental procedures are to commercial production conditions, the more
likely the responses are to mimic actual producer yield changes; and, hence,
the more useful they are in economic assessments.
The need for plant response information measured in terms of yield units
(rather than foliar injury) consistent with field conditions has been noted by
most analysts doing assessment research (e.g., Leung et al., 1978; Adams and
Crocker, 1982). Plant scientists have also recognized the need to report
response in terms of yield if economic losses are eventually to be estimated.
For example, Oshima and coworkers (Oshima, 1973; Oshima et al., 1976; Oshima
and Gallavan, 1980) reported crop losses in terms of potential or actual yield
reduction for selected crops. More recently, the NCLAN program (see, e.g.,
Heck et al.; 1983a,b; 1982) appears to be the first coordinated plant science
research effort intended to provide response information in a usable form for
economic assessments. The nature of the NCLAN protocols and results are
discussed in Section 6.4.3.2.2. As evidence of the utility of this and similar
efforts, NCLAN response data have been used to derive most of the economic
estimates reported in this chapter.
Prior to the availability of NCLAN data, researchers needing yield-response
data have either extrapolated from a narrow and often incomplete set of exist-
ing 0- response functions reported in the plant science literature; or have
estimated these relationships from secondary data on production and air quality,
with attendant statistical and measurement problems. The credibility of these
extrapolation or estimation procedures and their implications in terms of the
resultant yield loss estimates are questionable, given that the estimated
yield responses diverge sharply in most cases from the replicable NCLAN results.
As a result, the plant science data used in earlier economic assessments, along
with misspecified or overly simplified economic models, must be recognized as
a potentially critical source of uncertainty in resulting crop loss estimates.
6-174
-------
The air quality data used in assessments are also critical inputs. Speci-
fically, the yield changes predicted by the response functions are driven by
assumed changes in 0, exposure, typically measured by the difference between
actual and anticipated 03 levels for the region in question. Estimating
current ambient levels of Q~ or linking changes in SNAAQS (Secondary National
Ambient Air Quality Standard) to actual rural concentrations presents difficult
analytical and data problems. Calculating current ambient concentrations is
difficult because few monitoring sites exist in rural areas, where the bulk of
agricultural crops are grown. The complex meteorology of 0- formation and
transport also makes difficult the development of models for defining changes
in rural concentrations resulting from changes in urban emissions. Errors in
either estimation will lead to biases in the predicted yield changes used in
any assessment. The latter issue can be circumvented by developing air quality
scenarios based on hypothetical changes in rural concentration (e.g., percentage
reductions from actual), rather than on specific changes in SNAAQS; but this
limits the usefulness of the assessment in formulating policy. Even with such
an assumption, there is still a need to have actual or estimated ambient
concentrations as a base from which to develop and compare hypothetical 03
changes.
The effects of plant science and aerometric assumptions and extrapolations
on monetary estimates are reflected in the highly divergent loss estimates
reported for 03 in the 1978 document. The divergences in the pre-1978 assess-
ments, as well as those in some of the assessments to be reviewed subsequently,
may be partially attributable to the following summary list of plant science
and air quality data problems:
1. The effect of sparse data on fL-induced crop losses. A lack of data
caused past assessments to be based on extrapolations from available
foliar injury estimates, resulting in often unreliable yield-reduction
estimates.
2. Selection of alternative cultivar and crop mixes, regions, and time
periods in the analysis. Crop prices, production levels, and Oo
exposure vary geographically and temporally, with resultant changes
in dollar loss estimates.
3. Selection or definition of alternative background ambient levels to
portray "clean air" (absence of anthropogenic 0.,) in the analysis.
6-175
-------
When used in combination with a standardized dose-response function,
the use of different background ozone levels results in different
yield reduction estimates and, ultimately, different monetary esti-
mates.
4. The difficulty of extrapolating from controlled-chamber experiments
to agronomic regions with all the required assumptions regarding
soil type, precipitation regimes, 0, exposure patterns, solar radia-
tion levels, and interactions among these edaphic and climatic
variables.
5. The use of different measures of "dose" or exposure. For example,
in the recent NCLAN experiments dose was standardized as the seasonal
7-hour average in ppm. Other researchers use cumulative dose (e.g.,
hours of exposure to levels exceeding 0.10 ppm) or some other measure.
Establishing the meteorological linkage between these various dose
measures and secondary standards, as well as their correspondence to
levels of exposure actually realized by the plant, is an important
research area.
The uncertainties associated with some of these plant science and aero-
metric issues have been partially resolved in more recent assessments by the
availability of standardized NCLAN data, as discussed subsequently. While the
post-1978 assessments feature a more uniform set of plant science and aerometric
data, a range of assessment techniques have been employed in generating the
economic estimates. It is not possible to sort out the relative contribution
of economic data and assumptions vis-a-vis that of plant science and aerometric
data to the accuracy of past assessments without doing the assessments over.
Recent empirical work by Adams et al. (1982) suggests, however, that economic
and biological processes contribute equally to the measurement of net benefits.
The implication of this observation is that an accurate portrayal of both
plant science and economic responses is important in performing economic
assessments. Studies lacking in either category should be viewed as incomplete
analyses. In the following review, both the plant science and economic founda-
tions of recent assessments are evaluated.
6.5.3 Assessment Methodologies Applied to Agriculture
Assessments of air pollution damages to agriculture found in the litera-
ture fall within three broad methodological categories: damage function-crop
6-176
-------
loss models, simple monetary calculation procedures, and economic assessment
methodologies. While all three benefit from the use of defensible plant
science and aerometric data, only the latter category is capable of portraying
the above economic mechanisms that underlie the actual costs or benefits of
pollution changes.
The first type of assessment uses predictions of yield changes from
damage or response functions to estimate crop losses in physical units, such
as the reduction in actual or potential crop production in a given geographical
unit (e.g., a state or region). Examples include the assessments by Loucks
and Armentano (1982) and Moskowitz et al. (1982). While this approach fails
to incorporate the producer responses that determine the net supply effect of
the initial Oo-induced yield changes, the authors of these assessments make no
claim of having reported economic losses and thus these assessments are not
reviewed in this document.
The second, or "traditional" procedure, that is, traditional in the sense
that most assessments of air pollution prior to 1982 calculated dollar estimates
in this fashion, is a commonly used approach for calculating the dollar or
monetary effects of environmental change. In this type of assessment, increases
or decreases in production calculated from damage functions are translated
into a dollar value by multiplying the predicted yield or production changes
by an assumed constant crop price. The advantage of such a procedure is the
relative ease with which dollar values may be obtained, since the only economic
information required to perform the calculation is the price (usually last
season's average price) of the crop in question. As an assessment methodology
for obtaining accurate economic estimates, however, it suffers from serious
conceptual weaknesses by failing to recognize and account for the complex
processes underlying economic response discussed in Section 6.5.1. This
limits the validity of the estimates to some very restrictive cases. Thus,
while economic theory assumes that value can be expressed in monetary terms,
not all dollar loss estimates (including those obtained by the traditional,
monetary approach) should be viewed as valid economic estimates.
The third type of assessment framework features the use of standard
economic methodologies that address some of the economic issues raised in the
preceding section. As such, the procedures within this category are capable
of providing estimates of the benefits of air pollution control in dollar
terms that account for producer-consumer decision-making processes, associated
6-177
-------
market adjustments, and perhaps some measure of the distributional consequences
of alternative environmental policies or actions. Examples include the economic
assessments reported in Leung et al. (1982), Benson et al. (1982), Adams et
al. (1982), Mjelde et al. (1984), Kopp et al. (1984), or Adams et al. (1984b),
in which attempts have been made to include the market impacts of air pollution-
induced yield reductions and producer responses. While they involve somewhat
different analytical (solution) procedures as determined by the structure of
the particular economic problem, these studies all explicitly deal with price
adjustments, providing estimates of the economic effects on producers and
consumers. A detailed review of alternative economic techniques and their
suitability in assessing various environmental changes or policies is presented
in Freeman (1979).
Although economists discount the monetary estimates obtained from the
traditional "price times quantity" type of assessments (e.g., critiques of
this approach may be found in Just et al., 1982, and Crocker, 1982), estimates
arising from traditional and economic assessment methodologies are seldom
distinguished in the popular press. Failure to distinguish between the nature
of the methodologies has important policy implications in that the estimates
from the traditional procedure may be badly biased, leading to potentially
incorrect policy decisions. For example, assessments obtained from comprehen-
sive economic analyses have been compared with estimates obtained from the
traditional procedure using the same data (see Benson et al., 1982, and Adams
et al., 1982). The differences were moderate to large, with the traditional
procedure overestimating the costs of air pollution when a "clean air" and
ambient ozone condition (an environmental degradation) were compared. Also,
the traditional procedure provides estimates that address only producers'
effects with no attention paid to the impact on consumers. In some agricul-
tural situations, it is possible that this naive approach will predict producer
losses when in fact there are potential gains. Thus, there are fundamental
conceptual and empirical differences between monetary estimates calculated by
the traditional procedure and those obtained from more defensible economic
assessments.
6.5.4 Review of Economic Assessments of Effects of Ozone on Agriculture
Both regional and national assessments are found in the post-1978 litera-
ture. While each can provide useful information, the geographical scale in an
6-178
-------
assessment has implications for the validity and tractability of alternative
assessment techniques. For this reason, regional and national studies are
discussed separately. Only the third type of assessment framework that was
discussed in the previous section, featuring techniques that are capable of
addressing economic responses, is included in the regional review. In the
review of studies at the national level, assessments based on both the simple
"price times quantity" and economic approaches are discussed. Estimates from
both types are presented because of the importance normally attached to any
national estimate of pollution damage by the popular press and the resultant
need to make explicit any limitations inherent in the underlying analyses.
The emphasis of this discussion is on how well the assessments conform to
economic realities as defined in Section 6.5.1. As noted in Section 6.5.2,
however, the plant science and aerometric data are critical inputs in these
studies. These data are defined in detail for each study.
6.5.4.1 Review of Regional Assessments. Most of the economic assessments in
the literature focus on 0- effects in specific regions, primarily California
and the Corn Belt (Illinois, Indiana, Iowa, Ohio, and Missouri). This regional
emphasis may be attributed to the relative abundance of data on crop response
and air quality for selected regions, as well as the national importance of
these agricultural regions. While regional estimates are not usually sufficient
for measuring the national implications of alternative SNAAQS, such studies
can provide estimates of the benefits of regional 0, changes, and hence of
state or regional compliance, as well as useful comparative information on
alternative economic methodologies for assessing environmental damages. Also,
regional estimates may be indicative of the potential magnitudes of national
effects for certain crops, if that region produces a dominant share of those
crops.
Economic estimates of pollution effects for selected regions are presented
in Table 6-31. In addition to reporting the actual monetary loss or benefit
estimates derived from each assessment, the table contains considerable detail
on the critical plant science, aerometric and economic data, and assumptions
used in each assessment. The estimates can then be evaluated relative to the
nature of these data and assumptions.
Four of the regional studies have focused on California, a state with
both high 0- levels and an important agricultural economy. Adams et al.
(1982) assessed the impact of (L on 14 annual vegetable and field crops in
6-179
-------
TABLE 6-31. SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study/
region
Crops and
no. of cultivars
Annual
benefits
of control
($ mil lion)
Evaluation of critical data and assumptions
Plant response data Aerometric data Economic model/data
Additional comments
Adams et al.
(1982);
Southern
California
12 annual crops:
beans, broccoli,
cantaloupes,
carrots, cauli-
flower, celery,
lettuce, onions,
potatoes, tomatoes,
cotton, and sugar
beets. No cultivar-
specific responses.
$45 (in Larsen-Heck (1976) foliar in-
1976 jury models converted to yield
dollars) losses for all crops ex-
cluding tomatoes.
Tomatoes derived from
yield response function
by Oshima et al. (1977a).
California Air Resources
Board hourly data collec-
ted for sites closest to
production regions defined
in the economic model.
Exposure measured as
cumulative seasonal
exposure in excess of
California standard
(0.08 ppm)
A price endogenous math-
ematical (quadratic) pro-
gramming model reflecting
agronomic, environmental,
and economic conditions
in 1976. Base cost,
yield, and input data
derived from university,
state, and Federal
sources.
Economic effects measured
as a change in economic
surplus (sum of con-
sumers' and producers'
surpluses) between base
case (actual 03 levels
in 1976) and economic
surplus that would ne
realized if all regions
were in compliance with
1976 standard of 0.08
ppm.
00
O
Leung et al.
(1982);
Southern
California
9 crops: lemons,
oranges (Valencia
and Navel), straw-
berries, tomatoes,
alfalfa, avocados,
lettuce, and celery.
Results estimated
across all cultivars
contained in county-
average yields (see
"Plant response
data" column).
$103 (in 03 yield response functions
1975 estimated from secondary data
dollars) on crop yields (county-
level averages), regressed
on agronomic and environ-
mental variables, includ-
ing ambient 03 levels.
Exposure measured in
average monthly concen-
tration in ppm for 12-hr
period (7:00 a.m. to
7:00 p.m.) Data from 61
California Air Resources
Board monitoring sites.
Economic model is composed
of linear supply and demand
curves for each crop, esti-
mated with data from
1958 to 1977.
Economic effect is mea-
sured as a change in
economic surplus between
base case (1975) and a
clean air environment
reflecting zero 03.
Howitt et
al.
(1984a,b);
California
13 crops: alfalfa,
barley, beans,
celery, corn,
cotton, grain
sorghum, lettuce,
onions, potatoes,
rice, tomatoes,
and wheat. One
cultivar of each
crop.
From $35
(benefit
of con-
trol to
0.04
ppm) to
$157
(loss
for in-
crease
to 0.08
ppm) (in
1978
dollars).
03 yield response functions
derived from NCLAN data
through 1982. Both quadra-
tic and Weibull functional
forms used. Alfalfa re-
sponse from Oshima et al.
(1976). Response data avail-
able for only 10 crops; for
celery, onions, rice, and
potatoes, surrogate
responses used.
California Air Resources
Board data for monitoring
sites closest to rural
production areas. Expo-
sure measured as the sea-
sonal 7-hr average in each
production area for compati-
bility with NCLAN exposure.
Economic model similar to
Adams et al. (1982) but in-
cludes some perennial crops
and reflects 1978 economic
and technical environment.
Economic effects measured
as changes in economic
surplus across three 0;l
changes from 1978 actual
levels. These include
changes in ambient 03 to
0.04, 0.05, and 0.08 ppm
across al1 regions.
-------
TABLE 6-31 (cont'd). SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study/
region
Crops and
no. of cultivars
Annual
benefits
of control
($ mil lion)
Evaluation of critical data and assumptions
Plant response data Aerometric data Economic model/data
Additional comments
en
i—i
oo
Rowe et al. 14 annual and $43 to
(1984); San perennial crops: $117
Joaquin alfalfa, barley, depend-
Valley in beans, carrots^ ing on
California corn, cotton, degree of
grain sorghum, control
grass hay, grapes, (measured
pasture, potatoes, in 1978
safflower, tomatoes, dollars).
and wheat. One
cultivar for some
crops; others esti-
mated across all
cultivars, as con-
tained in county-
average yields (see
"Plant response data"
column.
Response functions based on
both experimental data and
secondary data. Lettuce,
tomato, corn, wheat, and
grain sorghum data from
NCLAN. Alfalfa from
Brewer and Ashcroft
(1982). Response func-
tions for beans, cotton,
and grapes from regres-
sion of county yields
on economic and environ-
mental variables. Res-
ponses for the remaining
crops were based on surro-
gate responses of similar
crops in the data set.
4. exposure levels were
tested: range from average
1-hr concentration to a.
cumulative dose where 03
>0.10 ppm. Each dose was
measured over an 8-hr
period (9:00 a.m. to 5:00
p.m.) and was tested in
the estimated response
function for each crop.
The average hourly concen-
tration was used in most
functions to predict
changes. All data were
from California Air
Resources Board monitor-
ing sites in predominant-
ly rural areas.
Same as Howitt et at.
(1984a,b).
Economic effects measured
as the change in economic
surplus between the 1978
base case and three in-
creasingly stringent
control scenarios:
(1) a 50% reduction in
number of hr M).10 ppm;
(2) meeting the current
California standard of
0.10 ppm; and (3) meeting
an 03 standard of 0.08
ppm.
Page et al.
(1982);
Ohio River
Basin
3 crops: corn, soy-
beans, and wheat.
Several cultivars
of corn and soybeans
may be reflected in
the biological data
used by Loucks and
Armentano (1982)
(see "Plant response
data" column).
$7,022
(measured
in 1976
dollars).
This is
present
value of
producer
losses
for
period
1976 to
2000.
Annua-
lized
losses
are
approx..
$270 in
1976
dollars.
Crop losses provided by
Loucks and Armentano (1982)
as part of ORBES project.
Yield responses derived by
synthesis of existing ex-
perimental data (from sites
outside the Ohio River
Basin). These response
functions then used to
predict changes in total
crop production in Ohio
River Basin between produc-
tion under "clean air"
(defined as a background of
of 0.03 ppm for 03) and
production under a range
of energy use scenarios.
Dose measured as cumula-
tive seasonal exposure
for a 7-hr period (9:30
a.m. to 4:30 p.m.) for
1977. Monitoring sites
at 4 locations were used
to characterize the
regional exposure.
The economic model con-
sists of regional supply
curves for each of the 3
crops. The predicted
changes in production
between "clean air" case
and each scenario are
used to shift crop
supply curves. Losses
represent present value
of losses for the period
1976-2000, discounted at
10%. The analysis ignores
price changes from shifts
in supply.
Losses are measured as
differences in producer
surplus across the vari-
ous scenarios. Since
prices are assumed fixed
(in real terms) over the
period, no consumer ef-
fects are measured.
-------
TABLE 6-31 (cont'd). SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study/
region
Crops and
no. of cultivars
Annual
benefits
of control
($ mil 1 ion)
Evaluation of critical data and
Plant response data Aerometric data
assumptions
Economic model/data
Additional comments
Benson
et al.
(1982);
Minnesota
CO
ro
4 crops: alfalfa,
wheat, corn, and
potatoes. Culti-
vars limited to
one per crop.
$30.5 Like the Loucks and Armentano
(measured (1982) study, 03-response
in 1980 functions based on crop loss
dollars), models estimated using
experimental 03-yield data
from other researchers
(from sites outside
Minnesota). Crop loss
modeling includes both
chronic and episodic re-
sponse and crop development
'stage as factors in yield
response, by regressing
yield on 03 expousres for
various time windows during
the growing season. Several
functional forms used to
test relationships between
yield and dose. Yield losses
for several 03 scenarios were
measured against production
under a zero background.
03 scenarios represent in-
creases in both concentra-
tions and frequency of
occurrence.
Air quality data are for
state of Minnesota for
1979 and 1980. Exposure
measured several ways but
generally as a daily expo-
sure statistic reflecting
either sum of hourly
averages or the mean
hourly average. These
exposures were then
summed over various
time invervals to
represent the exposure
for the various periods
indicated in the seasonal
crop loss models.
The economic estimates are
derived from a comprehen-
sive economic model of the
U.S. agricultural sector
that includes equations
capturing crop supply and
demand across multiple
domestic and foreign
markets. Model is
calibrated to 1980
values.
The economic effect for
each 03 scenario measured
in terms of short-run
profit changes for
Minnesota producers
under 2 regional assump-
tions. In the first
case, yields are assumed
to change only in
Minnesota. In the second
case, yields change in
Minnesota and the rest of
the U.S. In the first
case, losses to Minnesota
producers are $30.5
million for the most
extreme 03 increase. In
the second, producers
gain $67 million as a
result of increases in
crop prices when the
yields for all the U.S.
are reduced.
Adams and 3 crops: corn, soy- $668 (in
McCarl beans, and wheat. 1980
(1985); Five cultivars of dollars).
Corn Belt soybeans, 3 of
wheat, and 2 of
corn.
03-yield response information
from NCLAN for 3 years (1980-
1982). Yield adjustments
estimated from Weibull re-
sponse models.
Air quality data for the
growing season are inter-
polated from SAROAD moni-
toring sit|s by Kriging
procedure. Represents
rural concentration for
1980. Exposure is mea-
sured as seasonal 7-hr
average to be compatible
with NCLAN exposures.
Economic estimates are
generated by a mathemati-
cal programming model of
U.S. agriculture reflect-
ing 1980 supply, demand,
and input characteristics.
Farm level response is
portrayed by 12 indivi-
dual "representative"
farm models that are
used to generate supply
adjustments used in the
national level model.
Economic estimates
represent changes in
economic surplus (sum 01
consumers' and producers'
surpluses) between cur-
rent (1980) 03 levels
and increases and
decreases in ambient 03
levels. Reduction to
a uniform ambient level
of 0.04 ppm across all
regions results in
benefits of $668 million.
-------
TABLE 6-31 (cont'd). SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study/
region
Crops and
no. of cultivars
Annual
benefits
of control
($ million)
Evaluation of critical data and assumptions
Plant response data
Aerometric data
Economic model/data
Additional comments
00
CO
Mjelde 3 crops: corn, soy-
et al. beans, and wheat.
(1984); Responses repre-
Illinois sent mix of all
cultivars actually
grown by farmers
in the study.
Ranges Responses are estimated from
from secondary (non-experimental)
$55 to data on actual farmer yield,
$220 input, and 03 concentrations.
annually The procedure is conceptual-
for ly similar to methods used
period earlier by Adams et al. (1982)
1976 to and Leung et al. (1982) except
1980. that relationship being
modeled is the effect of 03
on farmer profit, rather than
yield. Results are trans-
lated into yield effects and
compared to NCLAN data from
Illinois (Argonne National
Laboratory).
Same Kriged data set as
used in Adams and McCarl
(1985),but for Illinois
only and covering 5 yr
(1976-1980). Exposure is
measured as seasonal 7-hr
average to facilitate com-
parison with NCLAN response
estimates.
Economic model consists of
a series of annual rela-
tionships on farmers' pro-
fits (profit functions).
These functions are ad-
justed to represent
changes in 03 (±25%)
for each year. Model
does not include con-
sumer (demand) effects.
The estimates represent
increases in farmers'
profits that could arise
for a 25% reduction in
03 for each year (1976-
1980). Years with
higher ambient levels of
03 have highest poten-
tial increase in profits
for changes. No effects
on consumers estimated.
Kriging is a spatial interpolation procedure that has been used to generate 03 concentration data for rural areas in which no monitoring sites have been
established. See Heck et al. (1983b).
-------
four agricultural subregions of central and southern California for 1976 using
a mathematical programming model of the California agricultural economy.
(Mathematical programming is an analytical technique for maximizing or minimi-
zing an objective function equation subject to a set of constraint equations.
The equations in such a problem, however, can be given a specific economic
interpretation consistent with the behavior of individual producers or consu-
mers, such as maximizing profit or economic surplus, or minimizing costs)
(Adams et al., 1982). The economic model captures price changes through
inclusion of linear demand relationships for each crop. To establish its
general accuracy, the model was calibrated against 1976 production data. The
model was then used to predict the effects of changed CL levels on crop price,
output, and the resultant impact on the welfare of both producers and consumers
as measured by changes in consumers' and producers' surpluses.
In view of the sparse experimental data available at the time, the authors
initially attempted to estimate (L-yield relationships from regression proce-
dures based on secondary data (county average yields regressed on 03 levels
and other inputs measured over a 20-year period). The estimation results,
however, were mixed, with some variables showing implausible signs and many
statistically insignificant coefficients. As a result, 03-induced reductions
in yield were estimated for most crops from the Larsen-Heck foliar injury
models (Larsen and Heck, 1976; Larsen et al., 1983). Foliar injury, as pre-
dicted by the Larsen-Heck models, was then converted to yield loss using
Millecan's "rule of thumb" (1971). The Oshima et al. model (1976) was used for
tomatoes. These projected yield changes are at best approximations, given the
tenuous link between foliar injury and crop yield. The response functions
were used to predict yield changes between actual 0- levels and levels that
would be realized if the state standard had been met. Data from the California
Air Resources Board were used for ambient 1976 03 levels. Using the predicted
yield changes, crop production, price, and economic surplus were estimated as
if the standard of 0.08 ppm, not to be exceeded more than one day per year,
had been achieved. The increase in economic surplus between the base economic
surplus and that under the lower 0, level is the benefit of control.
As a percentage of total crop value (about $1.5 billion), estimated
losses attributable to air pollution were found to be small, $45.2 million.
In terms of distributional consequences, meeting the 1976 standard would have
increased 1976 agricultural income (producer surplus) by $35.1 million and
6-184
-------
consumers' welfare (ordinary consumer surplus) by $10.1 million. To provide a
comparison, the authors also applied the traditional method of computing
losses (multiplying the estimated difference between actual and potential
yield by market price) and obtained a total estimated loss of $52.5 million.
While the difference in estimates between the methods appears small, the
traditional procedure measures only the effects on producers. Thus, if this
latter figure ($52.5 million) is compared with the producer loss from the
economic analysis ($35.1 million), the difference is approximately a 50-percent
greater loss estimate when the traditional approach is used.
Leung et al. (1982) estimated 0., damage to nine annual and perennial
crops in the California South Coast Air Basin, representing about 40 percent
of crop production value in the region. Ozone-yield relationships were derived
from regression procedures applied to secondary data, rather than from experi-
mental data. Crop yields for 1963 through 1975 were obtained from reports
from county agricultural commissioners on yields realized by farmers. Principal
component analysis (PCA), a technique in which highly correlated variables are
replaced with one or two components that contain most of the information of
the original variables, was used to transform monthly environmental data, such
as 03 concentration, temperature, relative humidity, and precipitation, into
yearly indices. Then yield was regressed on these indices using linear regres-
sion procedures. Finally, crop-yield changes were estimated for 1975 by
calculating differences between actual yields (with 1975 levels of 0~) and
yields predicted at a zero 03 concentration.
Leung et al. (1982) then calculated changes in consumer and producer
surpluses measured from an economic model containing linear supply and demand
relationships to approximate the welfare effects of changes in agricultural
supply brought about by air pollution. Specifically, the predicted yield
changes were used to shift the crop supply curves, thus generating a new level
of economic surplus. Estimated 1976 losses of consumer and producer surpluses
from 03 exposure were $103 million.
The estimate of direct economic loss was subjected to input-output analysis,
which traces the economy-wide effects of changes in a single economic variable,
to determine the indirect impact to related economic sectors in California.
Leung et al. (1982) estimated the indirect loss of sales from 0~ damage to be
$276 million in the study region and $36.6 million in the rest of the state.
These figures translate into lost income of $117 million attributable to air
pollution in the region and $14.1 million in the rest of state.
6-185
-------
While thfe Leung et al. (1982) analysis represents a worthwhile attempt to
overcome some data and statistical problems that have plagued economic assess-
ments of pollution damage, a number of limitations need to be recognized.
First, by assuming a zero background CL concentration, the economic losses
from manmade 03 are overstated if the true background 0- (including biogenic
contributions) level exceeds zero. While the actual background (L level is
still at issue in the literature, some researchers suggest a minimum back-
ground level of 0.025 ppm, measured as a seasonal 7-hr average (Heck et al.,
1984a,b). Since the ambient seasonal 7-hr averages for major production
regions in California tend to be about twice this value, the Leung et al.
(1982) yield losses may be overstated by a factor of two (assuming a linear
response function). Second, the use of principal components for 03, while
perhaps reducing multicollinearity, introduces some additional statistical
problems, such as whether the variability in the principal component for 03
bears any resemblance to how crop yields may actually change. Further, it is
difficult to interpret the principal component indices in terms of actual
policy (0~) changes. Finally, given the national linkages involved in
California agriculture, the use of a regional input-output model for agricul-
ture may be overstating the magnitude of the multipliers used to link primary
or direct effects to secondary economic effects. Inflated multipliers will
then inflate the regional economic effects. An additional caveat should be
noted in reporting economic surplus changes along with other income effects,
such as from an input-output model, because the economic surplus changes, if
measured in a final market, may already account for some of the welfare changes
in input markets. Thus, the income losses from the input-output model should
not be aggregated with the economic surplus estimate.
Two more recent studies have also examined the effects of air pollutants
(primarily 0,) on California agriculture. In the first, Howitt et al. (1984a)
assessed the impact of alternative 03 levels on the statewide production of
fourteen annual crops. The economic model used was similar to that employed
in Adams et al. (1982); i.e., a price-endogenous mathematical programming
model, scaled to 1978 values. The yield response data were derived from NCLAN
experiments through 1982, with the exception of those for alfalfa, which were
taken from Oshima et al. (1976).
Howitt et al. (1984a,b) examined three 03 scenarios, reflecting hypothetical
changes in ambient 03 levels to 0.04, 0.05, and 0.08 ppm seasonal 7-hr averages
6-186
-------
for nine production areas in California. The first (0.04 ppm) was a reduction
of approximately 25 percent in ambient ozone from actual 1978 values across
the production regions in the model, where the 1978 0^ levels were derived
from California Air Resources Board data, converted to seasonal 7-hr averages.
The second portrayed a slight degradation in air quality (increase in ozone)
from actual levels. The 0.08 ppm analysis portrayed an increase of about 60
percent in ambient 0- for the same regions. The economic effects of meeting
the two extreme 03 actions were a $36 million net benefit (from the 25 percent
reduction in 03) and a $157 million loss (from the increase in 0^). The 0.05
ppm analysis amounted to a slight degradation in air quality and hence a very
small economic loss. These effects are in line with those observed in Adams
et al. (1982), but are based on more defensible biological data on plant
response.
The second study, by Rowe et al. (1984), focused on both annual and
perennial crops in the San Joaquin Valley of California, the major production
area of the state. Using both field (county average yields, climatic, and
economic data) and experimental data from NCLAN and other researchers, Rowe
et al. (1984) estimated a series of yield functions for the major crops grown
in the San Joaquin Valley for both 0, and sulfur dioxide (S09). With the
«3 c.
exception of potatoes, no yield changes were seen from adjustments in ambient
SOp levels. Ozone exposure was characterized as a cumulative seasonal dose
recorded during a daily 8-hr period. The yield adjustments associated with
three 03 reduction scenarios were then used to drive the same economic pro-
gramming model used in Howitt et al. (1984a,b).
The GO analyses of Rowe et al. (1984) feature greater reductions and
hence greater yield adjustments than those in Howitt et al. (1984a,b), as well
as broader crop coverage. As a result, the benefits of control are larger
(though still a small proportion of agricultural value), amounting to $48 to
$117 million, depending on the ambient 0., level assumed. These results, in
combination with earlier California studies, provide evidence of economic
consequences of 03, with the distribution of these impacts felt most heavily
by the producers of the commodities. For example, in both Howitt et al.
(1984a,b) and Rowe et al. (1984), producer impacts are about 60 to 70 percent
of the total change in economic surplus. The Rowe et al. (1984) study also
provides some support for the use of combined field and experimental data as a
means of filling gaps in the generation of yield effects for a broad range of
crops.
6-187
-------
The economic consequences of air pollution to agriculture within the Ohio
River Basin (Illinois, Indiana, Ohio, Kentucky, West Virginia, and Pennsylvania)
were estimated by Page et al. (1982). The region is a major producer of corn,
soybeans, and wheat; it also experiences 0- levels sufficiently high to depress
crop yields. Page et al. (1982) examined a wide range of energy use scenarios
that may occur in the Ohio River basin. These were then translated into
changes in ambient air quality. Regional supply curves were estimated for
each crop, based on standard economic specification of such relationships.
These supply curves were then shifted to reflect varying air pollution assump-
tions. The change in area above these curves and below a given price represent
the producer surplus change for such a shift in supply. While the study
included two pollutants, SOp and 03, approximately 98 percent of the losses
were attributed to 0^.
The magnitude of the supply shifts, and hence producer economic effects,
was based on yield response data provided by Loucks and Armentano (1982). The
response functions from Loucks and Armentano (1982) used to generate the
supply adjustments were synthesized from experimental data from other research-
ers for these crops and then applied to air quality data for the Ohio River
Basin. Economic losses were measured as changes in producers' surplus result-
ing from supply curve shifts between clean-air and 0- and S02 levels under the
varying scenarios. The net present value of 0--induced losses between clean
air (a background of 0.03 ppm) and the various air quality scenarios for the
period (1976 to 2000) is approximately $7.0 billion, or an annual equivalent
of $278 million. Not surprisingly, most of these losses accrue to the states
with the largest agricultural production in the region, Illinois and Indiana.
Several limitations need to be noted concerning the response data and the
economic estimates of Loucks and Armentano (1982). First, the use of these
supply adjustments in the Ohio River Basin represents an extrapolation of data
from other regions and cultivars that may not be typical of the Ohio River
Basin. Further, an ex post comparison with NCLAN data indicates that the
predicted yield changes used here do not conform well to the subsequent NCLAN
data; i.e., the yield adjustment estimates of Loucks and Armentano (1982) are
higher than those for current NCLAN response functions. Second, in addition
to problems with the underlying response data, there is a conceptual problem
associated with assessing only producer level effects. While Page et al.
(1982) noted that this region produces over a third of U.S. corn and soybeans,
they assumed that there would be no price effects associated with changes in
6-188
-------
supply. This assumption is questionable in view of the other recent studies
in the region that have demonstrated a price effect for supply shifts of this
general magnitude. These other results suggest that the study probably over-
stated producer effects while ignoring potentially large changes in consumer
welfare.
Benson et al. (1982) have provided estimates of the economic effects of
03 on Minnesota agriculture. The plant science assumptions for the study were
summarized in Section 6.4.3.2.2. The authors evaluated 0,-induced crop losses
for alfalfa, wheat, corn, and potatoes through the application of crop loss
functions that specifically accounted for crop development and episodic exposure
by breaking exposure into multiple time periods over the growing season. The
approach was similar to that used by Loucks and Armentano (1982) in that raw
data on yields and 0, exposure from other researchers (at sites outside Minnesota)
were used to develop crop loss models under different measures of exposure and
different functional forms. The loss functions were then applied to Minnesota
by using actual or simulated 1979-1980 county-level 03 data for Minnesota.
The results of this procedure are subject to the same limitations noted for
the Page et al. (1982) study.
The potential production losses for each county were aggregated by Benson
et al. (1982) to provide a statewide crop loss measured in physical units. An
economic model of supply and demand relationships for U.S. agriculture was
then used to convert these production adjustments for each crop into producer
losses, under two alternative supply assumptions: (1) assuming 0- levels and
thus production are unchanged in areas outside of Minnesota; and (2) assuming
that CL levels and therefore yields change nationwide. In the second case,
the analysis accounted for supply and demand relationships for each crop as
affected by production changes in all regions. The two assumptions gave
highly divergent estimates of economic effects on Minnesota producers. For
example, the estimated dollar loss to Minnesota producers attributable to a
worst-case (L level obtained from the first assumption was approximately
$30.5 million in 1980 dollars. When, however, the economic model accounted
for price increases resulting from reduced production in Minnesota and all
other regions, there was a gain to Minnesota producers of $67 million in the
short run if 0- levels increased (in Minnesota as well as other production
areas). This gain resulted from the rise in prices associated with reduced
supply. These results, when combined with similar observations from Adams
6-189
-------
et al. (1982) and Leung et al. (1982), suggest the importance of using assessment
methodologies that account for regional market linkages and resultant price
effects.
Mjelde et al. (1984) estimated the effects of 0- on Illinois cash grain
farms (farms producing corn, soybeans, and wheat but no livestock) by estimating
income and cost relationships for individual farms exposed to various levels
of 0^. One objective was to test whether a meaningful link could be established
between the physical loss estimates obtained under controlled experimentation,
such as in the NCLAN program, and response information inherent in observed
economic behavior (i.e., the individual farmer's cost and yield data). To
test for such a link, the authors developed profit and cost functions (func-
tional relationships in which profit or cost is regressed on economic and
environmental explanatory variables) based on data from a large sample of
Illinois grain farms. These profit and cost relations were estimated from
actual cost and production data for these Illinois farms and incorporated
environmental variables (i.e., 0-, temperature, and rainfall) as well as
traditional economic variables. Data on 0, for each year were prepared by J.
Reagan, U.S. Environmental Protection Agency, via the Kriging spatial interpola-
tion procedure (see Heck et al., 1982b). (Versions of this Kriged data set are
used in several of the regional and national-level assessments.)
In most specifications of the profit functions, Mjelde et al. (1984)
found a negative and significant (at the 5 percent level) impact by Oo on
profit. When the direct production (yield) effects of 0, suggested by the
farm-level sample data are compared with NCLAN results obtained in Illinois
(the NCLAN site at Argonne National Laboratory), the production responses
appear to be comparable. For a 25 percent increase in 0-, it was estimated by
Mjelde et al. (1984) that physical output (a weighted average) of corn, soybeans,
and wheat would decline 3.3 percent. The same 25 percent increase in On used
with NCLAN response functions predicts 11.7 percent and 3.7 percent decreases
in output for two cultivars of soybean, while output of corn would decline
between 1.4 and 0.6 percent for two cultivars. The Mjelde et al. (1984)
estimate of 3.3 percent as a weighted average (weighted by the shares of corn,
soybean, and wheat production for these Illinois farms) is quite similar to
the NCLAN estimates.
In economic terms, the 0, effects found in the Mjelde et al. (1984)
analysis resulted in an annual aggregate loss in profits to Illinois farmers
6-190
-------
of $55 to $220 million, depending on 03 levels in a given year (from 1976 to
1980). While the results suggest that economic estimates can be obtained
without complete reliance on experimental data for all crops and cultivars in
a region, certain caveats need to be noted. First, by current standards, the
authors had abundant economic and air quality data with which to work. Such
detailed data on producers' costs and yields do not exist at the national
level. In addition, a number of statistical and estimation problems arose.
Even though some of these were resolved, the stability of the coefficients in
several specifications is suspect and reinforces some well-recognized difficul-
ties in using secondary data to sort out statistically the effect of one
environmental variable from among the many that affect yield. Also, even
without statistical difficulties, the methodology does not eliminate the need
for experimental data, as some form of detailed experimental data is needed to
establish the plausibility of the regression estimates. Finally, from an
economic standpoint, the study suffers a conceptual problem similar to that in
Page et al. (1982), that is, no price changes (from changes in 0-) are calcu-
lated; hence, probable effects on consumers are ignored.
In a study of the effects of 0., on Corn Belt agriculture, Adams and
McCarl (1985) used a mathematical programming model of U.S. agriculture to
measure effects of alternative 03 standards on producers and consumers. The
model is conceptually similar to the programming models in Adams et al.,
(1982), Howitt et al. (1984a,b), and Rowe et al. (1984). The model is more
detailed, however, in its representation of both producer and consumer level
responses. Further, the model is applicable to the entire U.S. agricultural
sector, including livestock and export markets. In Adams and McCarl (1985),
changes in yields for corn, soybeans, and wheat in the Corn Belt were predicted
with NCLAN 03 concentration-response data through 1982. Ambient 1980 03
levels, measured as a seasonal 7-hr average, were obtained from the same
Kriged data set used in the Mjelde et al. (1984) study. Assuming no yield
changes in the rest of the U.S., the results of the analysis suggested that a
reduction in ambient 03 from the present Federal standard of 0.12 ppm to
0.08 ppm would provide a net benefit (increase) in economic surplus of $668
million. Conversely, relaxing the standard to 0.16 ppm would result in a
reduction in economic surplus of approximately $2.0 billion. The bulk of the
benefits came from changes in soybeans yield; soybeans are much more yield-
sensitive to 03 than corn. The results of this analysis are consistent with
6-191
-------
.distributional shifts associated with changes in supply in the face of inelastic
demand. That is, the 0.08 ppm scenario benefits consumers at the expense of
producers, whereas the 0.16 ppm assumption results in the opposite. These
changes in Federal standards are portrayed by assumed seasonal 7-hr 0- levels
(across the entire Corn Belt) of 0.04 ppm and 0.075 ppm. This translation of
a 1-hr standard into a seasonal 7-hr average is tenuous and assumes a lognormal
distribution of 03 events. In reality, then, these estimates are for reduc-
tions in 0~, rather than actual changes in Federal standards.
The authors also performed sensitivity analyses (i.e., compared changes
in the model output to changes in model parameters) to test the effect of
different yield data and assumptions on the economic estimates generated by
the model. The results of such analyses indicated that the effect of the
plant science data (or yield predictions) on economic estimates varied dramati-
cally. Specifically, the authors compared economic surplus estimates generated
from response data on corn and soybeans in the literature prior to NCLAN with
estimates using actual NCLAN data, and observed a large difference. Conversely,
when estimates using different subsets of NCLAN data were compared, the effect
of additional data on a given crop was less important, and in some cases was
trivial. One implication of this analysis is that the error in some early
economic estimates based on biological responses extrapolated from other crops
or not cross-checked against experimental data may be quite large.
6.5.4.2 Review of National Assessments. National analyses can overcome a
fundamental limitation of regional analyses by accounting for economic linkages
between groups and regions. Accounting for these linkages, however, requires
additional data and more complex models, and frequently poses more difficult
analytical problems. Thus, detailed national assessments tend to be more
costly to perform. As a result, there are fewer assessments in the literature
of pollution effects at the national than at the regional level.
Of the national assessments performed since the last criteria document
was published (U.S. Environmental Protection Agency, 1978), two use the
traditional "price times quantity" approach to quantify dollar effects.
Analyses of this type are deficient in capturing the true economic concept of
benefits, as discussed earlier. Four of the most recent national assessments
included in this review, however, use more defensible measures of economic
effects. Both types of national-level estimates of 0, damages are summarized
in Table 6-32. As with Table 6-31, considerable information on the nature of
6-192
-------
TABLE 6-32. SUMMARY OF ESTIMATES OF NATIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
cn
I
CO
Study
Ryan et a) .
(1981)
Shriner
et al.
(1982)
Adams and
Crocker
(1984)
Annual
benefits
Crops and of control
no. of cultivars ($billion)
16 crops: alfalfa, $1.747
beets, broccoli, (in 1980
cabbage, corn dollars).
(sweet and field),
hay, lima beans.
oats, potatoes,
sorghum, soybeans,
spinach, tobacco,
tomatoes, wheat.
Specific response
data available for
only 5 crops, and
one cultivar for each
of these crops (see
"Plant response
data" column).
4 crops: corn, soy- $3.0 (in
beans, wheat, and 1978
peanuts. Multiple dollars).
cultivars of all
crops but peanuts.
3 crops: corn, soy- $2.2 (in
beans, and cotton. 1980
Two corn cultivars, dollars).
three soybean, two
cotton.
Plant response data
Yield-response information
derived from a synthesis of
5'yield studies in the
literature prior to 1980.
Synthesized response func-
tions estimated for both
chronic and acute exposures.
The 3 chronic response
functions and 2 acute
response functions are
extrapolated to cover 6
of the 16 crops. For
the remaining 10 crops,
surrogates from the res-
pective chronic and acute
functions are used. Yield
changes are based on
reduction in 03 to meet
1980 Federal standard of
0.12 ppm in noncompl iance
counties.
Analysis uses NCLAN response
data for 1980. Functions
estimated in linear form.
Yield changes reflect dif-
ference between 1978 ambient
03 levels of each county
and assumed background of
0.025 ppm concentration.
Analysis uses NCLAN 03-yield
data for 1980 and 1981.
Functions estimated in
linear form. Yield changes
measured between 1980
ambient levels and an
assumed 03 concentration
of 0.04 ppm across all
production regions.
Aerometric data
Dose measured in several
ways to correspond to
underlying response func-
tion. Acute doses mea-
sured as concentrations
within a given averaging
time or second highest
8-hr average. 03 data
derived from several
sources, including data
from National Aerometric
Data Bank and from Lawrence
Berkeley Laboratory, for
period 1974-1976. Expo-
sures calculated for only
those counties (531) ex-
ceeding the Federal
standard (0. 12 ppm).
Dose measured as 7-hr
average to be compatible
with NCLAN exposure
levels. Rural ambient
concentrations for 1978
estimated by Kriging
procedure applied to
SAROAD data.
Dose is measured as the
seasonal 7-hr average to
be compatible with NCLAN
experiments. 1980 ambient
03 levels estimated by
Kriqing of SAROAD moni-
toring sites, translated
into a seasonal 7-hr
average.
Economic model /data
Naive economic model.
Monetary impact calcula-
ted by multiplying changes
in county production by
crop price in 1980.
Measures impact on pro-
ducers only.
Same as Ryan et al .
(1981), except uses
1978 crop prices.
Economic model consists of
crop demand and supply
curves. Yield changes
are used to shift supply
curve. Corresponding
price and quantity
adjustments result in
changes in economic sur-
plus. No producer-level
responses modeled; only
measures aggregate
effects.
Additional comments
Dollar estimate is for
the 531 counties exceed-
ing the Federal standard
of 0.12 ppm. This study
is essentially an updated
version of Benedict et
al. (1971) reported in
the 1978 criteria
document (U.S. Environ-
mental Protection
Agency, 1978).
Dollar estimates are for
all counties producing
the four crops. As with
Ryan et al. (1981), esti-
mates are for producer-
level effects only.
Economic estimate mea-
sured in terms of changes
in consumer and producer
surpluses associated with
change in 03.
-------
TABLE 6-32 (cont'd). SUMMARY OF ESTIMATES OF NATIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study
no.
Crops and
of cultivars
Annual
benefits
of control
($billion)
Plant response data
Aerometric data
Economic model/data
Additional comments
Adams et al. 4 crops: corn, soy- $2.4 (in
(1984a) beans, wheat, and 1980
cotton. Two culti- dollars).
vars for corn and
cotton, three for
soybeans and wheat.
Analysis uses NCLAN 03-yield
data for 1980 through 1982.
Yield changes measured be-
tween 1980 ambient levels
and 25% reduction in 03
across all regions. Func-
tions estimated in both
linear and quadratic form.
Same as Adams and Crocker
(1984).
Same as Adams and Crocker
(1984), except that anal-
ysis examines range of
economic estimates
reflecting variability
in yield predictions
resulting from sample
size and functional
form.
Same as Adams and
Crocker (1984). Linear
functions result in
higher yield losses and
hence higher economic
loss estimates. Reported
estimate ($2.4 billion)
is for quadratic response
function.
CTi
Kopp et al. 5 crops: corn, soy- $1.2 (in
(1984) beans, wheat, cotton, 1978
and peanuts. Multi- dollars).
pie cultivars of
each crop except
peanuts.
Analysis uses NCLAN 03-yield
response data for 1980
through 1982. Response
data are estimated in a
Box-Tidwell flexible
functional form. Yield
losses (for estimates
reported here) measured as
the difference between
ambient 1978 03 and a level
assumed to represent compli-
ance with a 0.08 ppm stan-
dard.
Same as Adams and Crocker
(1984) and Adams et al.
(1984a), but for 1978
growing season.
Economic model consists of
detailed producer-level
models, by crop, for
numerous production
regions. Predicted
yield changes are used
to generate supply shifts
for each region/crop.
Aggregate supply shifts
are then combined with
crop demand relationships
to estimate changes in pro-
ducer and consumer sur-
pluses.
In addition to measuring
the change in economic
surplus for various
assumed 03 levels, the
analysis also includes
an examination of the
sensitivity of the esti-
mates to the nature of
the demand relationships
used in the model.
Adams et al.
(1984b)
6 crops: barley,
corn, soybeans,
cotton, wheat, and
sorghum. Multiple
cultivars used for
each crop except
barley and grain
sorghum; two for
cotton, three for
wheat, two for corn,
and nine for soybean.
$1.7 (in Analysis uses NCLAN 03-yield
1980 response data for 1980
dollars), though 1983. Response func-
tions are estimated in
Wei bull form. For soybeans,
both individual and pooled
cultivar responses are esti-
mated. Yield changes
reflect changes from 1980
ambient 03 of 10, 25, and
40% reduction and a 25%
increase for each response.
Same as above but for
1980, and 1976 through
1980 periods.
Economic model consists
of two components: a
series of farm-level
models for each of 55
production regions and
a national model of crop
use and demand. Yield
changes are used to
generate regional
supply shifts for farm-
level models. These
supply responses are then
used in the national
model.
Consumer surplus esti-
mated for both domestic
and foreign markets;
producer surplus
nationally and by region.
The analysis includes a
range of economic esti-
mates reflecting changes
in response and 03 data
and assumptions. This
sensitivity analysis
identifies stability of
analysis results with
respect to various para-
meters.
Kriging is a spatial interpolation procedure that has been used to generate 03 concentration data for rural areas in which no monitoring sites have been
established. See Heck et al. (1983b).
-------
critical plant science, aerometric, and economic data is presented, along with
the actual estimates of benefits or damages. It is apparent from the table
that the range of estimates is relatively small. Such relative consistency
does not necessarily imply convergence on the "true" economic effects, however,
as the analyses employ somewhat different crops, response information, and
assessment approaches, as detailed below.
The recent national-level economic estimates of the effects of 0, on
agriculture include an assessment for the National Commission on Air Quality
(Ryan et al., 1981). This is an updated version of the Benedict et al. (1971)
study cited in the 1978 criteria document (U.S. Environmental Protection
Agency, 1978). The purpose was to estimate the benefits of meeting the SNAAQS
for 0- and S0? accruing to 16 agricultural crops. The principal methodological
differences between this study and the earlier Benedict et al. (1971) approach
include the use of a wider range of dose-response functions drawn from the
plant science literature through 1980, updated production data from the 1974
Census of Agriculture, and updated air quality and price information.
The loss in potential yield and hence total production as a result of 0,
and S0? pollution was estimated using alternative response functions and
county-level data on air quality for counties that have not attained SNAAQS
(531 counties). The response functions were of two types: chronic and acute.
Each was taken from existing data, with the acute based on foliar-injury
response models linearly converted to yield response. Only five crop-specific
studies were cited as sources for the response functions used in the analysis,
with the response of the remaining 11 crops (of the 16 studied) predicted by
surrogates selected from the five for which data were available. The predicted
physical loss estimates were then translated to a dollar value by the ad hoc
procedure of multiplying the reduction in production by the 1980 crop price
for each commodity. The resultant dollar loss estimate (or potential benefit
of meeting the SNAAQS for 03 and S02) was estimated to be $1.8 billion in 1980
dollars for agricultural crops. Of this total, the benefits of meeting the 0,
standard are far greater than the direct benefits of meeting SOp standards
($1,747 million compared to $34 million), because the number of nonattainment
counties for 0., is nearly six times the number of nonattainment counties for
SOp, and because fewer crops show sensitivity to SO^. This estimate is much
higher than the Benedict et al. (1971) estimate, reflecting the sensitivity of
these estimates to the data assumptions and time period employed. Note that
6-195
-------
this entire amount is assumed to accrue to producers, as the calculation
procedure does not measure consumer effects.
Besides the use of a deficient methodology, another shortcoming of this
assessment (Ryan et al., 1981) is that the crop yield-response estimates are
generated from a very limited set of actual data, reflecting the sparseness of
data prior to the availability of NCLAN data. The use of foliar-injury models
and the extrapolation across a large number of crops are obvious sources of
uncertainty. In view of the improved data and assessment frameworks currently
available, this monetary estimate is much less defensible than more recent
estimates.
A national assessment by Shriner et al. (1982) for the Office of Technology
Assessment of the U.S. Congress estimated the losses associated with ambient
0. levels for four crops: corn, soybeans, wheat, and peanuts. The study
employs NCLAN dose-response functions (in linear form) for each crop. Although
the data used were taken from only the first year of the NCLAN program, those
first-year data were generated under consistent conditions across all crops,
making the yield responses more plausible than the mixed, often extrapolated
sets (using one crop to portray response of other crops) used by earlier
researchers. The study also used county-level CU data interpolated by Kriging
from data obtained at SAROAD rural monitoring sites. Agricultural yields and
production by county were adjusted from 1978 actual yields (as reported in the
1978 Census of Agriculture) using the NCLAN response functions and the county-
level 0- data. Changes in yield were measured against what the yields would
be if crops were exposed to a "background" ambient 03 level of 0.025 ppm. As
in the Ryan et al. (1981) study, the estimated reductions in county production
levels for each crop were then converted to a monetary estimate of loss by the
traditional procedure of multiplying by constant county-level price. The
aggregate monetary loss (or difference between values of production at ambient
levels of 0- and those at 0.025 ppm) for the United States was estimated at
approximately $3 billion. This estimate suffers from the same problems as
those associated with the Ryan et al. (1981) study and other studies that
abstract from economic factors. Also, the calculation of the seasonal 7-hr
averages from the SAROAD sites (for use with the NCLAN response functions)
appears to have been inflated by use of a daily 7-hr maximum, rather than the
7-hr average. This would result in a larger difference between ambient and
background 0- and an inflation of the yield adjustments. The principal improve-
ment of the Ryan et .al. (1981) study over the Benedict et al. (1971) assessment,
6-196
-------
thus, is the use of 1980 NCLAN data and an aerometric data set that is based
on actual rural ambient (L levels, even though the data may possibly be slightly
biased.
Another estimate of 0., effects on agriculture was developed by Adams and
Crocker (1984), who used information on (L-induced plant effects to determine
the sensitivity of resultant economic estimates to additional plant response
information. The study also presents a numerical estimate of 0- damage to
corn, soybeans, and wheat, representing about 60 percent of the value of U.S.
crop production. Response data for these crops were derived from 1980 and
1981 NCLAN experiments. Response functions were estimated in linear form.
Rural 03 ambient levels for 1980 were derived from a corrected version of
the Kriged data set used in the Shriner et al. (1982) study. The predicted
yield changes for reduction in 0, to an assumed ambient concentration of
0.04 ppm were introduced into a simple economic model of farm-level demand and
supply functions. The predicted yield changes were used to shift supply,
which, when combined with the demand relationships for the commodity markets
in question, generated estimates of the economic surplus accruing to consumers
and producers from reductions in 0,. The estimated difference between ambient
Oo levels associated with the current standard of 0.12 ppm (as measured from
the Kriged SAROAD monitoring data) and a seasonal 7-hr average of 0.04 ppm was
approximately $2.2 billion in 1980 dollars. This calculation assumes that all
rural areas achieve the 7-hr seasonal average of 0.04 ppm 0.,, which is unlikely.
Also, the economic model measures price effects but lacks any detail on indi-
vidual producer or consumer responses. Both aerometric assumptions and this
economic abstraction imply an upward bias to the estimates.
Another national-level estimate of the economic consequences of 0,. on
corn, soybeans, cotton, and wheat is reported in Adams et al. (1984a). This
study used essentially the same methodology used by Adams and Crocker (1984).
While the primary purpose of the analysis was to discuss and measure the role
of plant science information in economic assessments, the analyses included
the estimated measurement of the benefits of changing 03 exposures for the four
crops. Ambient 0., levels for 1980 were characterized as the seasonal 7-hr
average and were taken from the Kriged data set. The benefits were measured
in terms of economic surplus estimated from the integration of supply and
demand curves for each crop under the alternative 0- levels. Yield effects
from both linear and quadratic response functions were used to shift the
6-197
-------
respective crop supply curves under alternative 0., scenarios. The new inter-
sections of supply and demand then generated changes in economic surplus.
The benefit of an assumed reduction in ambient 0- from 1980 estimated
levels to 0.04 ppm across the entire U.S. was estimated by Adams et al. (1984a)
to be approximately $2.4 billion in 1980 dollars. A 25 percent increase in 0-
(to 0.066 ppm) resulted in a net loss of $3.0 billion. These estimates were
derived with quadratic functional forms of the response model; linear estimates
were approximately 40 to 50 percent higher. Like the Adams and Crocker (1984)
study, these estimates are probably upper bounds, in that the economic model
does not deal with the specific types of producer responses that may mitigate
for changes in 0~. Nor does the model fully take into account certain factors
that potentially distort the agricultural markets, such as the Federal farm
program (which typically changes from year to year). It does include, however,
direct transfer payments to farmers from the U.S. Treasury that are part of
the Federal farm program. The analysis uses averaged assumed current ambient
concentrations for all production regions, rather than individual county or
subregion levels. The averaged ambient levels are the upper-bound seasonal
7-hr concentrations for major production areas as reported in the SAROAD data.
To the extent that all regions are equal to or less than this amount, the
benefits of reductions in 0- are overstated and the losses from (L increases
are understated.
The studies reviewed to this point all suffer in various degrees from
either plant science and aerometric data problems, incomplete economic models,
or both. Some were not intended to provide estimates for use in policy evalua-
tion. As a result of these limitations, decision-makers should be cautious in
using these estimates to evaluate the efficiency of alternative SNAAQS. There
are, however, two recent EPA-funded studies that overcome most of the problems
plaguing the above assessments. Together, they provide better estimates of
the agricultural consequences of changes in ambient 0,. Each is reviewed
below.
The first of these is a recent assessment by Kopp et al. (1984) that
measured the national economic effects of changes in ambient 0, levels on the
production of corn, soybeans, cotton, wheat, and peanuts. The study is notable
for several reasons. First, the assessment methodology was based on the
development of a series of detailed farm-level representations of costs and
yield for approximately 200 production regions for these crops, using the Farm
Enterprise Data System (FEDS) surveys from the U.S. Department of Agriculture
6-198
-------
(USDA). These farm-level responses were then aggregated to regional and
national supply responses. The use of farm-level models to initiate the
analysis is important in that it places emphasis on developing reasonable
micro- or producer-level responses to externally induced yield changes, such
as those that may be associated with changes in 0~. Second, the effects of 0-
on the yields of the included crops were based on NCLAN data through 1982.
Instead of using the Weibull models of the NCLAN program, response functions
of the Box-Tidwell type were fitted to these data. Predicted yield changes
associated with alternative secondary standards were then used to shift the
regional supply-response relationships. The price and consumption effects
were measured through a set of demand relationships for each commodity, reflect-
ing a range of elasticity assumptions, as reported by the USDA.
The results of the analysis indicate that a reduction in 03 from regional
ambient levels (as portrayed by the Kriged SAROAD set) (Heck et al., 1983b) to
an approximate 0.04 ppm seasonal 7-hr average would result in a $1.2 billion
net benefit in 1978 dollars. Conversely, an increase in 0- to an assumed
ambient concentration of 0.08 ppm across all regions would produce a net loss
of approximately $3.0 billion. The benefit estimate of 0, reductions is
slightly less than the estimates reported in Adams and Crocker (1984) and
Adams et al. (1984a). The differences may be attributable to the more detailed
regional and farm-scale resolution in Kopp et al. (1984), as well as different
base years (1978 versus 1980), different ambient base 0,, levels, and different
elasticity assumptions vis-a-vis these other studies. Relative to previous
assessments, limitations of this analysis are minor but include the lack of
crop substitution possibilities between sensitive and tolerant crops, and the
forcing of the economic adjustment process onto what is perceived to be the
high-cost production region. In addition, the study does not consider the
impact of Federal farm programs on benefits estimates. The analysis provides,
however, a detailed representation of the economic processes underlying agri-
cultural production, uses the most complete biological and aerometric data
currently available, and is directed toward providing useful policy analyses
of Oo pollution.
The second study, by Adams et al. (1984b, 1985), is a component of the
NCLAN program. As such, the analyses, data, and results contained in this
assessment represent the collective biological, meteorological, and economic
data assembled in .the NCLAN program through 1983. The results were derived
from an economic model of the U.S. agricultural sector (adapted from Chattin
6-199
-------
et al., 1983) that included domestic consumption, export use, and livestock
feeding and processing. Farm-level behavior was portrayed by individual farm
models for 55 production regions. The analysis looked at six major crops
(corn, soybeans, wheat, cotton, grain sorghum, and barley) that account for
over 75 percent of U.S. cropped acreage. Potential 03 effects on hay were
also evaluated using the average yield response of other crops as a surrogate.
The model output included estimates of the changes in crop production, prices,
and economic surpluses, by crop and region.
Ozone-induced yield changes for each crop and region, as defined by NCLAN
response functions (in Weibull form), were used in the economic model to
estimate economic effects resulting from those 0- changes. Four hypothetical
0- scenarios (three reductions, one increase) were judged against a 1980
ambient 0, base solution. The difference between the base solution, reflecting
1980 parameters, and the hypothetical 0, analyses provided an estimate of the
benefits or costs of the changes in 0- levels. The results indicated that the
annual benefits (in 1980 dollars) to society from 0., adjustments are substantial,
but represent a relatively small percentage of total agricultural output
(about 4 percent). Specifically, a 25 percent reduction in 0- from 1980
ambient levels resulted in benefits of $1.7 billion. This estimate is quite
close to the Kopp et al. (1984) benefit estimate (when Kopp et al. dollars are
converted to 1980 dollars), suggesting that benefits of 03 reductions are of
this magnitude. A 25 percent increase in 03 resulted in an annual loss (negative
benefit) of $2.363 billion.
In addition to estimating a set of economic effects for the four 0,
scenarios, the assessment by Adams et al. (1984b, 1985) also included some
measure of the sensitivity of the economic estimates to assumptions concerning
response and 0- data. The sensitivity analyses addressed the use of alternative
cultivar response functions (rather than average "pooled" responses), use of
different ambient 03 levels, and the potential influence of moisture stress on
03 yield estimates. The effect on economic estimates, compared with the above
estimates, ranged from trivial to substantial (from less than 5 percent differ-
ence to approximately a 50 percent difference). The greatest sensitivity was
reflected in an analysis in which yield predictions were taken from the most
extreme cultivar response available for soybeans, corn, wheat, and cotton.
Here the benefits of a 25 percent 03 reduction rose to $2.7 billion. Statisti-
cally and agronomically, however, these cultivar (yield) responses behave
6-200
-------
unlike responses of the other NCLAN cultivars for those crops. As such, these
higher economic estimates are perhaps upper bounds on potential impacts.
While the estimates from both Kopp et al. (1984) and Adams et al. (1984b,
1985) are derived from conceptually sound economic models, there are several
sources of uncertainty or error. These include the issue of exposure dynamics
(7-hr seasonal mean from the NCLAN experiments versus other exposure statistics
and exposure periods) and the lack of environmental interactions, particularly
Oo-moisture stress interactions, in many of the response experiments. Also,
the OT data in both studies are based on a limited set of SAROAD monitoring
sites, mainly in urban and suburban areas. While the spatial interpolation
process (Kriging) results in a fairly close correspondence between predicted
and actual ozone levels at a few validation points, there is a need for data
from more monitoring sites in rural areas. The economic models, with their
large number of variables and parameters, and the underlying data used to
derive these values, are potential sources of error; e.g., sensitivity analyses
on demand elasticities by Kopp et al. (1984) and on foreign exchange rates
(export market conditions) by Adams et al. (1984b) indicate changes of up to
about 20 percent. In addition, neither study, Kopp et al. (1984) nor Adams et
al. (1984b), considers the impact on benefits estimates of the Federal farm
subsidy program and other factors that may upset free-market equilibria and
produce market distortion. The Adams et al. (1984b) model is a long-run
equilibrium model and assumes that supply will be consumed at some price. Its
policy utility is greatest when addressing changes bounded by historical
levels, rather than quantum adjustments. The adjustments portrayed, however,
in the 0, analyses by Adams et al. (1984b) fall within these historical levels.
6.5.5 Overview of Current Economic Assessments of Effects of Ozone on Agriculture
The ability to assess (L damage to agricultural crops has been greatly
enhanced by recent improvements in crop yield-response information and air
quality data. As Section 6.4.3.2.2 of this chapter indicates, the plant
science literature now contains concentration-response functions calibrated in
yields for most major agricultural commodities, primarily as a result of the
NCLAN program. While cultivar coverage remains sparse for some crops and
important edaphic-climatic interactions are only partially addressed, these
concentration-response relationships are superior to data underlying loss
6-201
-------
estimates reported in the 1978 criteria document. In addition, air quality
data for rural areas have improved as monitoring expands. Interpolation
procedures, such as the use of Kriging based on SAROAD information, offer
promise in terms of filling existing gaps in air quality data.
This review of recent agricultural assessment efforts also indicates an
increase in the application of techniques consistent with economic theory.
Consequently, the studies produce more defensible estimates of economic bene-
fits. Two of the most recent studies, those by Kopp et al. (1984) and Adams
et al. (1984b), are the most comprehensive economic assessments of 0, damages
performed to date. In even these studies, however, as well as in other studies
reviewed in this section, the treatment of some economic and plant science
issues is not complete. Some deficiencies include the need to measure damages
to perennial crops (fruits and nuts, forests); and the need to account for
potential long-term and dynamic 0., effects, such as interactions between 03
levels and the frequency and intensity of insect and disease incidence, rainfall
or irrigation patterns, and fertilizer and other factors. Such effects might
differentially alter producer patterns in the use of irrigation and the appli-
cation of pesticides and fertilizer, a possibility not currently addressed in
economic assessments (all assume 0, neutrality with respect to input use).
Also requiring more attention are potential economic damages to nonmarketed
plants (e.g., as manifested through aesthetic effects on forest ecosystems).
As noted earlier, another important issue concerns the appropriate measure of
dose. While the current NCLAN experimental design (as discussed in this
chapter) uses the seasonal 7-hr mean concentration, other dose measures may
better characterize plant response and lead to different predicted yield
changes. Furthermore, the NCLAN response functions for individual crop culti-
vars estimated at various sites appear to be relatively homogeneous when
measured in percentage change (rather than absolute values), but the validity
of extrapolating site-specific response data across regions is not fully
resolved. Finally, the impact of factors that result in market distortions,
such as the Federal farm subsidy programs, have not been addressed.
Nevertheless the inclusion of these possible improvements in future
assessments is not likely, with the possible exception of market-distorting
factors, to alter greatly the range of agricultural benefits provided in the
Kopp et al. (1984) and Adams et al. (1984b) studies, for several reasons.
First, the current studies cover about 75 to 80 percent of U.S. agricultural
6-202
-------
crops (by value). For inclusion of the other 20 percent to change estimates
greatly would require that their 0~-sensitivity be much greater than that for
the crops included to date. Second, the model sensitivity analyses from
existing studies indicate that changes in key plant science parameters must be
substantial to translate into major changes in economic estimates. From
experience to date, it seems unlikely that use of different dose measures or
interaction effects would result in changes of the magnitude already addressed
in some of the sensitivity analyses. Third, even if there are such changes,
there are likely to be countervailing responses; e.g., longer exposure periods
may predict greater yield losses but Q~-water stress tends to dampen or reduce
the yield estimates. Finally, it should be noted that potential improvements
in economic estimates of agricultural effects are relevant to policy only to
the extent that they alter the relationship between total benefits and total
costs of that policy. Uncertainties in other effects categories (non-agricul-
tural) are probably greater.
In conclusion, the recent economic estimates of benefits of 0, control to
agriculture, particularly those by Kopp et al. (1984) and Adams et al. (1984b),
provide the most defensible evidence appearing in the literature to date of
the general magnitude of such effects. The close correspondence of the Kopp
et al. (1984) and Adams et al. (1984b) estimates also indicates that sound
economic models are available for application to this problem area. As a
percentage of the total value of crops included in these assessments, the loss
estimates are in the range of 4 to 6 percent. This is comparable with the
estimates of crop losses from sources such as insects and diseases reported by
Boyer (1982) but far less than the $25 billion annual loss attributed by the
U.S. Department of Agriculture (1965) to weather-related damage. Relative to
estimates in the 1978 criteria document (U.S. Environmental Protection Agency,
1978) and economic information on other O, effects categories, such as damage
to materials, these two studies, in combination with the NCLAN data on yield
effects, provide the most comprehensive economic information to date on which
to base judgments regarding the economic efficiency of alternative SNAAQS. As
noted in this review, there are still gaps in plant science and aerometric data
and a strong need for meteorological modeling of 0- formation and transport pro-
cesses for use in formulating rural (L scenarios. With regard to the economic
data and models used, the impact of factors that upset free-market equilibria
needs further analysis. Additionally, it must be emphasized that none of the
6-203
-------
studies has accounted for the compliance costs of effecting changes in (L
concentrations in ambient air. A complete benefit-cost analysis requires that
the annualized estimated benefit to agriculture that would result from ozone
control be combined with benefits accruing to other sectors and then compared
with the overall annualized compliance costs.
6.6 MODE OF PEROXYACETYL NITRATE (PAN) ACTION ON PLANTS
Peroxyacetyl nitrate (PAN) is the most common member of a series of
homologues that increase in phytotoxicity with increase in molecular weight.
Only PAN is found in ambient air at concentrations of possible concern, and
then only in limited areas of the country.
The sequence of events inducing vascular plant response to PAN is essen-
tially identical to that described for 03 (Section 6.3); PAN enters the leaf
tissue through open stomata and dissolves in the aqueous layer surrounding the
substomatal chamber (Figure 6-1). Hill (1971) reported that PAN was rela-
tively insoluble and that the rate of absorption by an alfalfa canopy was
approximately one-half that for 0,. The absorption rate depends upon the
ability of the plant to metabolize, translocate, or otherwise remove the
active pollutant species from the absorbing solution, as well as on the solu-
bility of PAN. Thus, the flux of PAN into the inner leaf tissues is influenced
by many physical, biochemical, and physiological factors. The equation used
to describe 03 flux (Section 6.3) also can be directly applied to describe the
flux of PAN into the leaf.
Highly unstable, PAN breaks down rapidly when it comes in contact with an
aqueous solution (Mudd, 1975). According to Nicksic et al. (1967) and Stephens
(1967), the breakdown of PAN in aqueous solution yields acetate, nitrite,
oxygen, and water. The pathway of PAN absorption and reaction within the leaf
tissue has not been described well enough to explain why cells at a specific
stage of physiological development are highly susceptible while adjacent cells
are relatively tolerant. The magnitude of PAN injury is influenced by the
stage of tissue development, succulence of the tissues, and conditions of the
macro- and microclimate. Injury is manifested in several ways. The most
evident injury is necrosis of rather specific areas of the lower and upper
leaf surfaces. This characteristic symptom expression may be accompanied by
leaf distortion, premature senescence, and defoliation (Taylor, 1969). Experi-
mental evidence shows that yield may be suppressed in the absence of visible
6-204
-------
injury symptoms (Thompson and Kats, 1975; Temple, 1982). Symptoms of the type
induced by PAN have been reported from California, the eastern United States,
Canada, Japan, and the Netherlands (Table 6-33). The smog, photochemical
smog, or oxidant injury symptoms described by Middleton et al. (1950), Went
(1955), and by other researchers working with polluted ambient air in California
prior to about 1960 were identical to injury symptoms subsequently produced
with synthesized PAN (Taylor et al. , 1961; Taylor, 1969). Frequently, the
injury symptoms were sufficient to reduce significantly the quality of leafy
vegetables and ornamental crops, but they were seldom associated with suppressed
growth or yield.
The phytotoxicity of PAN and processes of injury development from PAN
will be discussed in the following sections. Many of the biochemical and
physiological studies with PAN and its homologues were conducted with concen-
trations that exceed those encountered in ambient air. The studies were con-
ducted, however, to identify responses that might be more difficult to detect
at lower concentrations. For unknown reasons, most vegetation grown in glass
houses and growth chambers is considerably less sensitive to synthesized PAN
than comparable plants grown and exposed to PAN and the total pollutant complex
found in the field (Taylor, 1969).
6.6.1 Biochemical and Physiological Responses to PAN
As with 0-, (Section 6.3.1), the phytotoxic effects of PAN occur only when
a sufficient amount of the gas diffuses into susceptible regions of the leaf
interior and encounters the plasmalemma or passes into the liquid phase of the
cells. Once deposited on the wet cell surface, the gas will begin to break
down and the degradation products or PAN molecules, or both, will move by
diffusion or bulk flow to sites of action (Mudd, 1975). The target sites may
include the cell membrane, chloroplast, cytoplasm, and various cell organelles.
6.6.1.1 Gas-Phase Movement into the Leaf. The primary entry port for PAN
into leaf tissue is through open stomata. As indicated in Section 6.3.1.1,
the influence of 0- on stomatal movement has received considerable attention,
but relatively little effort has been made to determine if PAN will also
induce stomatal closure. Starkey et al. (1981) reported that a PAN-susceptible
variety of bean, exposed to 80 ppb PAN for 0.5 hr, developed drought stress
symptoms but that a tolerant variety showed no effect. This finding suggests
that PAN may have stimulated stomatal opening to allow a greater rate of
transpiration. Metzler and Pell (1980) found that pinto bean plants exposed
6-205
-------
TABLE 6-33. GEOGRAPHIC OCCURRENCE OF PAN/OXIDANT INJURY ON PLANTS'
Area
Species injured
Reference
California
Washington
Missouri
Illinois
Colorado
Utahb
Bean,
spinach,
Romaine lettuce
Oat, petunia,
tomato, Swiss
chard, sugar
beet
Middleton and Haagen-Smit (1961)
Tingey and Hill (1967)
Maryland
Pennsylvania
New York
Tobacco
Garden plants
Went (1955)
The Netherlands
Little-leaf nettle,
petunia,
annual bluegrass
Floor and Posthumus (1977)
Japan
Various species
Spinach, French
bean, lettuce
Fukuda and Terakado (1974)
Sawada et al. (1974)
Canada
Tomato
Pearson et al. (1974)
Where a column entry is blank the information is the same as the entry above.
""Monitoring data for PAN in southern California, Utah, The Netherlands, and
Japan were available to corroborate the reports of PAN-type symptoms
observed in those areas.
6-206
-------
to subthreshold levels of PAN (54 ppb for I hr) developed no macroscopic
injury and showed no effects on stomatal conductance. At the injury threshold
(70 ppb for 1 hr) and above, abaxial glazing developed and stomatal conductance
increased. Temple (1982) observed no effects on stomatal conductance (compared
to non-exposed plants) at concentrations of 25 and 50 ppb PAN after tomato
leaves were exposed for 2 hr. In this study, 0.20 ppm 03, in combination with
the two concentrations of PAN, did suppress stomatal conductance when tomato
plants were exposed for 2 hr.
The size of stomatal pores and number of stomata per unit area of leaf
vary greatly according to plant species. Many plants have stomata in both
surfaces of the leaf, whereas others have stomata only in the lower surface.
As a general rule, stomata occur in larger numbers per leaf area near the apex
of the leaf and become less numerous toward the base of the leaf. Although
plants shown to be most susceptible to PAN are among those that have stomata
in both leaf surfaces, no correlation between susceptibility and number or
size of stomata has been demonstrated.
6.6.1.2 Biochemical and Physiological Responses. Peroxyacetyl nitrate is a
highly specific phytotoxic agent that attacks leaf tissue at a fairly specific
stage of physiological development and is most injurious to succulent, rapidly
expanding tissues of herbaceous foliage (Noble, 1955; Taylor and Maclean,
1970). Concentrations of 14 to 15 ppb (maximum) under field conditions have
been observed to produce PAN-type injury on susceptible crops (Taylor, 1969;
Temple, 1982). Fukuda and Terakado (1974) reported that petunia plants under
field conditions developed silvering and bronzing on the lower leaf surface
when the maximum PAN concentrations ranged from 3.0 to 6.7 ppb. The most
serious observed damage occurred when a PAN concentration of more than 5 ppb
continued for 7 hr. Because PAN is phytotoxic at very low concentrations,
Mudd (1963) concluded that the most likely target in plant cells must be some
enzyme system. Much of the early work with enzymes involved the use of rela-
tively high PAN concentrations to demonstrate reactive sites in the metabolic
pathways.
Ordin (1962) observed that growth of oat coleoptile sections,, which
involved cell expansion rather than initiation of new cells, was inhibited by
PAN. He found that fumigation with 1.1 ppm PAN for 6 hr resulted in 32 percent
inhibition of growth and 45 percent inhibition of glucose absorption from the
solution. Fumigations were accomplished by floating the oat coleoptiles in a
solution and bubbling PAN through the solution. There was no way to determine
6-207
-------
how much PAN the coleoptiles actually encountered. The response suggests that
PAN interfered with metabolism of cell wall sugars. Subsequently, Ordin and
Hall (1967) found that cellulose synthetase was inhibited, and Ordin et al.
(1967) reported that phosphoglucomutase was inhibited when coleoptile tissue
was exposed to PAN. The treatment consisted of bubbling 50 ppm PAN for 4 hr
at a rate of 400 ml/min through 100 ml of solution in which the coleoptiles
were floating.
Using HI vitro procedures, Mudd and Dugger (1963) showed that PAN oxidized
NADH and NADPH. Mudd (1966) and Mudd et al. (1966) found that enzymes with
free -SH groups were inactivated, but that enzymes with no free -SH groups
were resistant to PAN. The amount of PAN used in these studies was not reported.
Hanson and Stewart (1970) observed that exposure to 50 ppb PAN for 1 to 4 hr
inhibited mobilization of starch in darkness, implying that the phosphorylase
reaction was inhibited. Such a response could seriously interfere with photo-
synthate partitioning and thus could inhibit growth and development. The
reaction deserves further investigation.
Thomson et al. (1965) showed that PAN (1000 ppb for 30 min) or its degra-
dation products caused crystallization and other disruptions in the chloroplast
stroma that were similar to the effects of dessication. These observations
suggest that PAN affected the permeability of the chloroplast membrane in much
the same way as it reacted with the plasmalemma, which allowed leakage of cell
contents.
In summary, peroxyacetyl nitrate enters leaf tissue through open stomata,
is rapidly dissolved in the aqueous covering of substomatal cells, and along
with its degradation products is transported through the cell wall and cell
membrane into the aqueous cell contents. The chloroplast membrane is disrupted,
thereby inducing plasmolytic-type characteristics to develop. There is leakage
of cellular fluids into the intercellular spaces. Enzymes containing sulfhydryl
groups are inactivated by PAN. Visible injury from PAN results when mesophyll
cells are killed and shrink, causing dessication and death of the epidermal
tissue. A degree of chlorosis is often visible on the upper leaf surface as
the chloroplasts in living cells are destroyed. The destruction of chloroplasts
(Thomson et al. , 1965) and disruption of biochemical and physiological systems
(Ordin and Hall, 1967; Ordin et al. , 1967; Mudd, 1966; Hanson and Stewart,
1970) can be expected to affect growth and yield adversely as well as the
aesthetic qualities of the vegetation. Inactivation of enzymes can suppress
6-208
-------
growth, as demonstrated with oat coleoptiles, and may interfere with photo-
synthate, as demonstrated by inhibition of starch mobilization in the dark;
and may interfere with other metabolic processes.
6.6.2 Factors that Modify Plant Response to PAN
Plant response to PAN and many other environmental stresses is condi-
tioned by complex, interacting internal and external factors (U.S. Environ-
mental Protection Agency, 1978). External physical factors such as temperature,
light conditions, humidity, and edaphic factors can influence plant response
to PAN. Similarly, biological variables such as genetic differences, physio-
logical stage of tissue development, and rate of plant growth can affect plant
response.
6.6.2.1 Biological Factors. Trees and other woody species are apparently
quite resistant to foliar injury from PAN (Davis, 1975; Davis, 1977; Taylor,
1969). Foliar injury has been produced only once or twice by fumigations with
extremely high concentrations of PAN (1 ppm for several hours), and injury to
these species in the field has not been reported. Variations in suscepti-
bility to PAN within herbaceous species have been observed in the field and
have been demonstrated for some crops with synthesized PAN.
Genetically controlled plant variation to PAN has been observed under
field conditions and verified by controlled PAN exposures. Drummond (1972)
exposed 28 F, varieties of petunia plants to 150 ppb PAN for 1 hr and found
highly significant differences in cultivar sensitivity. Six petunia cultivars
that were common in the Boston area were exposed to high concentrations of PAN
(120, 250, or 500 ppb for 1 hr) to ensure that all the cultivars developed
some foliar injury so that differential cultivar sensitivity could be determined
(Feder et al., 1969). The authors concluded that the cultivars tested showed
differential sensitivity to PAN and that cultivars resistant to PAN were also
resistant to other pollutants. In contrast, studies by Hanson et al. (1976)
at Arcadia, CA, found that petunia cultivars (49 siblings from complete diallele
crosses of seven commercial lines) sensitive to PAN (sensitivity based on
foliar injury intensity) were not necessarily sensitive to 0~. The results
from ambient air studies were confirmed by controlled exposures to known PAN
concentrations (86 or 120 ppb for 1, 1.5, 2, or 2.5 hr). DeVos et al. (1980)
used inbred parents of White Cascade, a PAN-sensitive F, hybrid, and Coral
Magic, a PAN-tolerant hybrid, to study inheritance of PAN tolerance. Plants
were exposed to 150 ppb PAN for 1.5 hr in controlled environment chambers.
6-209
-------
Significant genetic variation in breeding lines was detected, but there was
also a large genotype-by-environment interaction. Starkey et al. (1976)
exposed 10 varieties of bean for 2 hr to 120 or 150 ppb PAN to observe the
injury symptomology and to determine differential cultivar sensitivity. All
cultivars developed some abaxial bronzing and glazing; only the intensity of
symptom expression and the time for symptom development varied among cultivars.
The authors compared their ranking of relative PAN sensitivity with published
information on 0, sensitivity for the same cultivars and found that the PAN-
sensitive cultivars were not necessarily sensitive to 0~.
Middleton et al. (1950) first described smog injury (PAN type) and listed
endive, lettuce, romaine lettuce, and spinach as extremely sensitive but
carrot and members of the cabbage and melon families as tolerant. This general
ranking of sensitivity is still accepted for PAN. Specific cultivars of
petunia, bean, Swiss chard, oats, and cos lettuce were selected for their PAN
sensitivity as demonstrated in controlled fumigation studies. Tomato was
originally listed as only slightly sensitive to smog (PAN), but it is now
known that many varieties are highly sensitive.
Sensitive plants show a characteristic pattern of injury when they are
exposed to PAN. As described from field observations in Los Angeles County,
CA, by Noble (1955), Juhren et al. (1957), and Glater et al. (1962), leaves of
different ages show damage in different positions. A similar description of
PAN injury confirms that susceptibility is related to specific physiological
stage and foliage development (Taylor, 1969; Taylor and Maclean, 1970; Noble,
1955; and Glater et al., 1962); but the causal factors involved in this selec-
tive sensitivity phenomenon have not been identified.
6.6.2.2 Physical Factors. The light-exposure regime to which plants are
subjected before, during, and after exposure to phytotoxic concentrations of
PAN will significantly affect response (Taylor et al. , 1961). Brief dark
periods preceding exposure and immediately following exposure can reduce or
even prevent the development of visible symptoms of injury. Maximum injury
occurs when plants are exposed in full sunlight. Dugger et al. (1963) deter-
mined that the maximum quantum responsivity to PAN occurred in the 420 to
480 nm range.
Juhren et al. (1957) found that plants were most susceptible to oxidant
injury (PAN-type symptoms) when grown under 8 hr photoperiods, but that injury
decreased with photoperiods of 12 to 16 hr. This observation may help to
explain why symptoms of PAN injury are most prominent in late fall, winter,
6-210
-------
and spring in southern California. Juhren also found that the greatest oxidant
injury occurred at temperatures of 25° in the daytime and 20°C at night.
The effects of relative humidity, air temperature, and edaphic factors
have not been investigated extensively, but some observations have been reported.
There is no cohesive evidence regarding the significance of relative humidity
and plant susceptibility, but PAN injury to vegetation in the South Coast Air
Basin of California occurs most frequently when relative humidity is 50 percent.
or greater (Taylor, 1974).
Field observations in southern California, where irrigation is essential
for crop production, revealed that crops growing under a soil moisture defi-
cit, a period when stomatal conductance is frequently reduced, developed few
or no 0- or PAN-type injury symptoms during a severe smog attack; while adja-
cent, recently irrigated crops were severely injured (Taylor, 1974). Similarly,
the author observed increased tolerance of beans and tobacco to 03 and PAN
when potted test plants were inadvertently allowed to wilt briefly during the
day preceding fumigation, even though the plants were watered several hours
before treatment and appeared to be normal.
6.6.2.3 Chemical Factors.
6.6.2.3.1 Chemical Additives. The effectiveness of chemical additives applied
for pest control, as well as specifically for the prevention of oxidant air pol-
lutant injury, has been studied by Freebairn and Taylor (1960), Pell (1976), and
Pell and Gardner (1975, 1979). These studies were designed to determine if cul-
tural practices could be modified to mediate the effects of PAN and other oxi-
dant air pollutants. None of the chemical treatments proved sufficiently effec-
tive, however, in preventing or reducing PAN injury to encourage general use.
6.6.2.3.2 Pollutant interactions. The importance of 0- as a phytotoxicant
was not recognized before the 1960s, although it was identified as a major
chemical component of the photochemical oxidant complex in the 1950s. It is
known that PAN is rarely, if ever, present in the absence of 0- in photochem-
ically polluted atmospheres (Oshima et al. , 1974; Penkett et al. , 1977; U.S.
Environmental Protection Agency, 1978; Tilton and Bruce, 1981). The ratio of
03 to PAN in southern California has been reported to be about 10:1 (Taylor,
1969); at Calgary, Canada, the ratio found varied according to atmospheric
conditions (Peake and Sandhu, 1983). (See Chapter 5 for a discussion of
03-to-PAN ratios.)
Interactions involving plant exposure to mixtures of PAN and 0- in pollu-
ted atmospheres probably occur, but the few published reports of controlled
6-211
-------
PAN plus GO interaction studies with plants have shown variable and inconsis-
tent effects on symptom type and intensity of injury. Kohut et al. (1976)
found that 0, (0.18 ppm) plus PAN (0.18 ppm) treatments for 4 hr in midday
produced 0~-type symptoms on hybrid poplar seedlings, but that the amount of
injury was highly variable. Davis (1977) found that ponderosa pine seedlings
that were exposed to an 03 (0.40 ppm) plus PAN (0.20 ppm) combination for 4 hr
developed significantly less injury than those exposed to 0-, alone. Kohut and
Davis (1978) reported greater-than-additive 0~-type injury on bean leaves
exposed to the 03 (0.30 ppm) plus PAN (0.05 ppm) combination for 4 hr, but PAN
injury was almost completely suppressed. In a study of the protective effects
of benomyl on bean plants exposed to 0.25 ppm 0., and 0.15 ppm PAN for 3 hr,
Pell (1976) found that the combination of 0., and PAN produced more injury than
PAN alone.
Posthumus (1977) exposed little-leaf nettle and annual bluegrass to 0,
(0.17 ppm) and PAN (0.05 ppm) singly and in combination for 2 hr in either the
morning or afternoon. The combination induced more foliar injury in the
morning than in the afternoon. There was no clear increase or decrease,
however, in the foliar injury in the plants exposed to the combination com-
pared to the injury from the single gases. More recent studies with little-leaf
nettle (Tonneijck, 1984) showed that no interaction between 0, and PAN was
detected when both were applied at their respective injury threshold concentra-
tions. The pollutant combination caused less than additive injury, however,
when the PAN concentration exceeded the injury threshold concentration.
Matsushima (1971) studied the effects of SOp and PAN, singly and in combination
(alternating or in concurrent mixtures), on pinto beans (PAN, 0.45 ppm; S0?,
1.5 ppm for 90 min), pepper (PAN, 0.37 ppm; S0?, 2.1 ppm for 70 min) and
tomato (PAN 0.40 ppm; SOy, 1.8 ppm for 60 min). The resultant foliar injury
was additive or less than additive from the combination of pollutants. In the
mixture, PAN injury tended to appear on the young leaves and SO,, injury on the
mature leaves. Nouchi et al. (1984) exposed petunia and bean plants for 4 hr
to mixtures of 0- and PAN to assess effects on visible symptoms of injury.
Ozone concentrations for the petunia sludy were 0, 0.10, 0.20, 0.30, and 0.40
ppm and PAN concentrations were 0.01, 0.02, 0.03, and 0.04 ppm. In the bean
study, 03 concentrations were 0, 0.15, 0.20, 0.30, and 0.40 ppm, and PAN
concentrations were 0, 0.030, 0.045, 0.065, 0.085, and 0.100 ppm. For PAN
alone, injury symptoms appeared on petunia at 0.020 ppm PAN; and with bean,
injury appeared at 0.030 ppm PAN. The percentage of foliar injury was greatest
6-212
-------
when plants were exposed to PAN alone, and the percentage injury decreased as
the 0., concentration increased. Temple (1982) exposed four cultivars of tomato
plants to PAN-03 mixtures (0, 0.025, and 0.050 ppm PAN and 0, 0.10, and 0.20
ppm 0,) for 4 hr once a week for 3 wk. The effects of the mixture on leaf
area and leaf dry weight were less than additive. Stomatal conductance was
reduced to a greater extent from the mixture than the individual gases.
6.7 PAN EXPOSURE AND RESPONSE
Initial PAN-injury symptoms, which fully develop during the 24 to 72 hr
following exposure, are a glazed, bronzed, or metallic sheen on the lower
(abaxial) leaf surface. These symptoms are clearly distinct from those produced
by 0-, which typically causes upper surface necrotic stipple, fleck chlorosis,
or bifacial necrosis on sensitive species (Temple, 1982). Transverse bands of
bleached, necrotic tissue and glaze and bronze on the lower surface (Noble,
1955) are characteristic of the PAN injury syndrome (Taylor, 1969). Most
sensitive plant species develop diffuse transverse bands of injury in regions
where the tissue is in identical stages of physiological development (Figure
6-23). This phenomenon results in injury at the apex of the youngest suscep-
tible leaf and at regions nearer the base of the next successively older three
or four leaves. Exposure on successive days results in a series of two or
more injured bands separated by bands of healthy tissue, demonstrating that the
stage of high susceptibility lasts for only a relatively short period (Noble,
1965). Some leaves, such as the two primary leaves on bean plants, do not
develop the bands; the injury may be distributed at random or as a solid cover
over the entire lower surface.
Ordin and Propst (1962) demonstrated that the auxin indole acetic acid
(IAA) in oat coleoptiles was completely inactivated when 1.3 ppm PAN was
passed through the solution in which they were suspended for 3 hr. Similarly,
enzyme activity was inhibited by exposures to 1 ppm PAN for 1 hr (Ordin et
al., 1971) and to 125 ppm for 6 min (Mudd, 1963). Thomson et al. (1965) found
that exposure to 1 ppm PAN for 30 min damaged leaves of pinto bean and chloro-
plasts were markedly altered as the damage developed. The cell membranes
were disrupted and the cell contents clumped together in a large mass. Dugger
et al. (1965) reported that PAN inhibited ATP and NADPH formation and the
14
fixation of COp, thus inhibiting the photosynthesis of carbohydrates.
Coulson and Heath (1975) found that PAN inhibited photosynthesis and that
6-213
-------
AREAS INJURED
BY PAN
••-UPPER EPIDERMIS
PALISADE LAYER
SPONGY
PARENCHYMA
LOWER EPIDERMIS
STOMATA
SITE OF
INITIAL INJURY
Figure 6-23. PAN injury. Note position effect with age of leaf.
Initial collapse is in the region of a stomate.
Source: Brandt (1962).
6-214
-------
photosystems I and II were both affected to a similar extent. These biochemical
and physiological studies were conducted with high concentrations of PAN (1
ppm and above) that far exceed those found in the atmosphere, but they demon-
strate that reactions essential for plant growth and development may be inhi-
bited.
The response of plants to PAN was summarized in the previous criteria
document for ozone and other photochemical oxidants published in 1978 (U.S.
Environmental Protection Agency, 1978). Figure 6-24 graphically presents the
estimated limiting values for PAN injury as calculated by Jacobson (1977) and
presented by the U.S. Environmental Protection Agency (1978). Sensitive
plants exposed to doses in the region below and to the left of the data points
have a low risk for development of visible injury symptoms. Those plants
exposed to doses to the right and above the line are at greater risk of develop-
ing injury symptoms. This illustration was based on a limited amount of
information and the data were produced by controlled fumigations with synthe-
sized PAN. Plants growing and exposed under ambient field conditions may be
at greater risk than indicated by the illustration (see Section 6.6).
In summary, PAN is one member of a family of highly phytotoxic,, gaseous
compounds in the photochemical oxidant complex. Acute responses of plants to
0- and PAN result from disruption of normal cell structure and processes. The
biochemical and physiological effects of PAN are not understood as well as the
effects of 0,. Effects of PAN on plant growth and yield were recognized in
the previous criteria document (U.S. Environmental Protection Agency, 1978),
but the documented responses were associated with visible injury symptoms.
The concept of limiting values (i.e., those concentrations below which foliar
injury and, presumably, reduced growth and yield would not occur) was used to
illustrate potentially harmful exposures. The range of limiting values for
PAN was: 1000 pg/m3 (200 ppb) for 0.5 hr; 500 ug/m3 (100 ppb) for 1 hr; and
175 ug/m (35 ppb) for 4 hr. Studies using little-leaf nettle have shown,
however, that the limiting values proposed by Jacobson (1977) were insufficient
to protect that species from PAN injury (Tonneijck, 1984). In this species,
the limiting values would need to be reduced 30 to 40 percent to prevent
foliar injury.
Although supporting data for growth and yield response to PAN exposures
are deficient, it must be emphasized that yield and growth effects can occur
with and without extensive visible symptom development on exposed plants.
This section has focused on yield loss as described in Section 6.2. Foliar
6-215
-------
0.24
1 I I I I I Ml
1 I I I I I
1200
0.16
Z
o
Z
LLJ
o
Z
o
o
0.08
LIMITING VALUES
D)
800
Z
UJ
u
Z
400 8
0.1 0.5 1 5
DURATION OF EXPOSURE, hours
Figure 6-24. Dose-response relationships and limiting values
for foliar injury to vegetation by peroxyacetylnitrate (PAN).
Source: U.S. Environmental Protection Agency (1978).
6-216
-------
injury is an important factor as a bioindicator of exposure to PAN (see
Section 6.7.1 below) and a cause of yield loss as reported periodically in
southern California during the past 30 years.
6.7.1 Bioindicators of PAN Exposure
Foliar injury symptoms frequently reduce market value seriously and, in
some instances, render the product unmarketable. These symptoms also may
destroy a significant amount of photosynthetically active leaf tissue. The
injury-symptom syndrome can serve as a very important indicator that damaging
PAN exposures have occurred, although it should be noted that the effect may
be considerably greater than the actual tissue destruction observed.
The concept of using bioindicators to assess the impact of air pollutants
and the methodology involved are presented in Section 6.4.1. The PAN injury
symptoms that are most useful in field diagnosis are the diffuse transverse
bands of injury, which may be visible only on the lower leaf surface or on
both surfaces; and the glazing, silvering, bronzing, or metallic sheen, or all
of these, on the lower leaf surface. Recognition of PAN injury in the field
is not always a simple process because a type of lower leaf surface glaze and
bronze may be produced by other factors such as cold temperatures, insects
(mites), and other air pollutants: 0,, hydrochloric acid (HC1), SCL, and
hydrofluoric acid (HF). In making field assessments, it is important that the
observer know the relative susceptibility of the crop and the native and
ornamental species in the area and that as many different species as possible
be examined.
Noble (1965) reported on a 6-yr study in southern California designed to
use plant indicators to identify injury induced by air pollutants. He used
six agricultural crops and two weed species widely distributed in the area.
The study revealed that annual blue grass (meadowgrass) was a very good indi-
cator for PAN. Posthumus (1977) found that little-leaf nettle and annual
bluegrass developed characteristic PAN-type injury symptoms when exposed to
about 50 ppb (0.05 ppm) PAN, and he suggested that these wild species might be
accurate indicators. Sawada et al. (1974) used 16 plant species in a survey
for 03 and PAN injury and observed PAN injury on 28 percent of the 138 plants
used in the study.
Field surveys in southwestern Ontario, Canada (Pearson et al., 1974)
revealed PAN-type injury symptoms on tomato crops. On the basis of these
6-217
-------
symptoms and the meteorological conditions that occurred during their develop-
ment, the authors concluded that the air pollutants probably originated in the
Cleveland, OH, area. Bioindicators should be used cautiously, however, when
monitoring data are not available to verify PAN concentrations and when obser-
vations are made on a single plant species. Lewis and Brennan (1978) reported
PAN-type injury on petunia leaves exposed to mixtures of 03 and SO^. Wood and
Drummond (1974) suggested that PAN-type injury may be caused by interactions
of PAN and other phytotoxicants or perhaps by a single pollutant such as HC1.
Field observations and diagnosis provide an important means of determining
if a PAN problem exists. Although PAN can be measured chromatographically,
the instrument can be calibrated only with known concentrations of PAN. The
problems associated with the synthesis, dilution, and measurement of PAN for
calibration purposes have discouraged the establishment of monitors for long-
term use (see Chapter 4). Plant-damaging exposures of PAN have been verified
with monitoring instrumentation in only a very few locations. Therefore, the
ability to recognize and evaluate PAN injury symptoms in the field is very
important.
Foliar injury of the type induced by PAN has been reported in more than
half of the counties in California, in several states, and in several foreign
countries. Went (1955) reported PAN-type injury in some European and South
American cities as well as in several cities in the eastern United States.
Locations at which PAN injury was observed on vegetation in the United States
are presented in Table 6-33.
6.7.2 Nonvascular Plant Response to PAN Exposure
Gross and Dugger (1969) examined the effects of PAN on algae (Chlamydomonas
reinhardtii) by measuring growth, photosynthesis, respiration, and pigment
content of the cells. Treatment usually lasted for several minutes, during
which PAN was bubbled through a liquid medium containing the algal cells. The
gaseous mixture usually contained an average PAN concentration of 125 ppm in
nitrogen (Np), with the treatment dose expressed in nanomoles (nM). Exposures
ranged from 20 to 250 nM. The study results indicated that both autotrophic
and heterotrophic growth was inhibited, photosynthesis and respiration were
adversely affected, and photosynthesis was more severely affected than respira-
tion. The results also indicated that carotenoids were destroyed and that
there was destruction of both chlorophylls, although chlorophyll a was more
6-218
-------
stable than chlorophyll b. Gross and Dugger (1969) also reported that PAN
lowered the free sulfhydryl content (-SH) of the cells.
Field studies of the lichen populations in the southern California moun-
tains indicated trends in ecosystem community parameters that inferred that
oxidant air pollutants had a deleterious effect on lichens (Sigal and Taylor,
1979). They fumigated three species for 4 hr/day for 8 days with 50 ppb PAN.
In one experiment, the lichens were fumigated for only 7 days with 0.100 ppm
PAN. Photosynthesis was inhibited in Parmelia sulcata, probably inhibited in
Hypogymnia entermospha (results were highly variable), and appeared not to be
affected in Collema nigrescens. The difference in gross response of photosyn-
thesis to PAN fumigations exhibited by these three lichen species tends to
indicate that PAN, along with other pollutants, may be detrimental to lichen
populations.
6.7.3 Losses in Vascular Plants Caused by PAN
The term loss is used in this section to mean loss in the intended use or
value of vegetation caused by PAN injury. The loss may be a reduction in
amount of marketable product or a loss resulting from aesthetic degradation.
6.7.3.1 Losses in Aesthetic Use and Foliar Yield. Petunias, a species highly
susceptible to PAN injury, are frequently used as bedding plants. Although
monetary losses from PAN injury to vegetation have not been studied, it is
obvious that they can occur, affecting the wholesale industry, retail market,
and the consumer (e.g., in the Los Angeles Basin, CA). Although such informa-
tion is not reported in the literature (O.C. Taylor, personal communication),
attempts have been made to produce plants outside heavily polluted areas and
transport them to the market. This practice was only partly successful because
substantial foliar injury usually developed after delivery to retail outlets
and before retail sale. While the petunia is one of the most susceptible
species, other ornamentals that are planted for foliage and blossoms also are
affected.
Several vegetable crops such as leaf lettuce, spinach, mustard greens,
table beets, endive, and romaine lettuce are grown and marketed for their
foliage. Some of these crops are grown in close proximity to metropolitan
areas and marketed as specialty crops. These species are harvested early in
the morning and are supplied, at relatively high prices, to restaurants and
specialty stores. After a heavy PAN attack, entire crops in some areas are
not marketable, and others require expensive hand work to sort and trim the
6-219
-------
product to make them acceptable. No reliable assessment of such losses has
been made, but losses of several hundred thousand dollars per year in the Los
Angeles, CA, area have been suggested (Middleton et al., 1950, 1956).
The indirect effect of PAN on plant growth resulting from destruction of
leaf tissue has not been measured. Destruction, however, of a significant
amount of leaf area caused by the necrotic bands, damage to the lower epidermis,
and increased defoliation of deciduous plants should be expected to suppress
growth. Earlier reports indicated that growth and yield by most plants are not
measurably affected until the loss of photosynthetic surface exceeds 5 percent
(Thomas and Hendricks, 1956). Plants that rapidly replace foliage (e.g.,
grasses) might be expected to express less growth reduction because of foliage
loss than plants that retain their foliage for several years (e.g., citrus
trees) and replace the lost foliage more slowly.
Thompson and Kats (1975) reported a trend toward reduced yield of mature
navel orange fruit when branches of mature trees were enclosed and fumigated
with PAN dosages equivalent to those occurring in the Riverside, CA, area.
The treatments consisted of carbon-filtered air, ambient air, and carbon-fil-
tered air plus additions of PAN adjusted to simulate concentrations monitored
in the ambient air at Riverside. The continuous treatments were administered
for 9 months. Tree growth was suppressed, presumably because of lost photo-
synthetical ly active tissue when leaf drop was stimulated.
6.7.3.2 Losses Determined by PAN Addition Studies. Based on PAN addition
studies using several species, Temple (1982) concluded that the,potentially
phytotoxic episodes could be defined as concentrations greater than 15 ppb
(0.015 ppm) for 4 hr in the morning or greater than 25 ppb for 4 hr in the
afternoon. His experiments were conducted in Teflon®-covered CSTR chambers in
a greenhouse. Concentrations of approximately 14 ppb PAN for 4 hr in ambient
air are sufficient to produce foliar injury on susceptible plants growing in
the field (Taylor, 1969). In chamber studies, however, approximately two to
three times this dose was required to induce injury symptoms (Posthumus,
1977). Because of this discrepancy between chamber and field studies, it is
difficult to relate responses obtained in chambers using synthesized PAN to
responses expected in the field.
Greenhouse exposures of lettuce and Swiss chard to 0, 25, and 50 ppb PAN
for 4 hr/day once a week for up to 4 wk caused no visible leaf injury and
appeared to have little, if any, effect on plant growth (Temple, 1982). By
itself, or in combination with 0~, PAN had no effect on stomatal conductance.
6-220
-------
Temple (1982) found that PAN and 0, alone and in combination reduced growth in
four tomato varieties and altered partitioning of photosynthate between roots
and shoots. He exposed the plants to 0, 0.1, and 0.2 ppm 0- and 0, 0.025, and
0.050 ppm PAN, alone and in all combinations, for 4 hr/day once a wk for 3 wk.
No PAN-type visible injury developed on the tomato plants and this exposure
had no effect on expression of 03 injury. The PAN treatments had no effect
on stomatal conductance, but 0.2 ppm 0- reduced stomatal conductance in all
four varieties. Results from two separate experiments were erratic,, perhaps
because the studies were conducted at different times of the year. The evidence
that the root/shoot ratio was altered suggests, however, that further study is
needed.
Greenhouse-grown plants (radish, lettuce, chard, oat, tomato, pinto bean,
beet, and barley) representing root, foliage, fruit and seed crops were exposed
to PAN (0, 0.005, 0.010, 0.020, or 0.040 ppm) for 4 hr/day, twice per wk from
germination to maturity of the harvestable crop (Taylor et al.,,1983). Signifi-
cant yield reductions were observed only in lettuce (Empire) and chard; the
threshold for yield reduction appeared to be between 0.010 and 0.020 ppb.
Yield was reduced 13 percent in lettuce and 23 percent in chard exposed to
40 ppb. Of all the crops tested, only pinto bean developed a significant
amount of foliar injury and only after exposure to 0.040 ppb; this sensitivity
persisted throughout the developmental cycle of the crop. The results indicate
that PAN at concentrations below the visible injury threshold can cause signifi-
cant yield reductions in sensitive cultivars of leafy (foliage) crops.
Field observations in southern California during the past 30 years have
revealed that severe visible PAN injury seldom appears during mid-summer, even
though higher dosages and concentrations occur during the four summer months
(Temple and Taylor, 1983). Ozone dosage also is highest during this period.
To assess effectively the impact of PAN, in the presence and absence of visible
symptoms, experiments should be designed to use 0- and PAN mixtures, be con-
ducted in as near full sunlight as possible, and be able to simulate fall and
spring environmental conditions limited to those periods.
Youngner and Nudge (1980) measured the relative susceptibility of cultivars
of 10 turfgrass species exposed to 0.050 ppm PAN or to 0.5 ppm 0- for 3 hr.
They reported a significant variation in amounts of foliar injury and noted
that warm-season grasses were more tolerant of both 0, and PAN than were the
cool-season grasses.
6-221
-------
Evidence of plant growth suppression following intermittent exposure to
PAN at concentrations comparable to those found in ambient polluted air without
visible leaf injury symptoms has been reported (Thompson and Kats, 1975;
Temple, 1982).
6.8 SUMMARY
Foliar injury on vegetation is one of the earliest and most obvious
manifestations of 0- injury. The effects of 0- are not limited to visible
injury, however. Impacts can range from reduced plant growth and decreased
yield, to changes in crop quality and alterations in susceptibility to abiotic
and biotic stresses. The plant foliage is the primary site of 0, effects,
although significant secondary effects, including reduced growth (both roots
and foliage) and yield, can occur.
Ozone exerts a phytotoxic effect only if a sufficient amount reaches the
sensitive cellular sites within the leaf. The 0, diffuses from the ambient
air into the leaf through the stomata, which can exert some control on 0-
uptake, to the active sites within the leaf. Ozone injury will not occur if
(1) the rate of 0, uptake is low enough that the plant can detoxify or metab-
olize 03 or its metabolites; or (2) the plant is able to repair or compensate
for the effects (Tingey and Taylor, 1982). This is analogous to the plant
response to SO^ (Thomas et al., 1950). Cellular disturbances that are not
repaired or compensated are ultimately expressed as visible injury to the leaf
or as secondary effects that can be expressed as reduced root growth, or
reduced yield of fruits or seeds, or both.
Plant growth and yield are the end products of a series of biochemical
and physiological processes related to uptake, assimilation, biosynthesis, and
translocation. Sunlight drives the processes that convert carbon dioxide into
the organic compounds (assimilation) necessary for plant growth and development.
In addition to nutrients supplied through photosynthesis, the plant must
extract from the soil the essential mineral nutrients and water for plant
growth. Plant organs convert these raw materials into a wide array of compounds
required for plant growth and yield. A disruption or reduction in the rates
of uptake, assimilation, or subsequent biochemical reactions will be reflected
in reduced plant growth and yield. Ozone would be expected to reduce plant
growth or yield if (1) it directly impacted the plant process that was limiting
plant growth; or (2) it impacted another step sufficiently so that it becomes
6-222
-------
the step limiting plant growth (Tingey, 1977). Conversely, 0- will not limit
plant growth if the process impacted by CL is not or does not become rate-
limiting. This implies that not all effects of 0, on plants are reflected in
growth or yield reductions. These conditions also suggest that th&re are
combinations of 0, concentration and exposure duration that the plant can
experience that will not result in visible injury or reduced plant growth and
yield. Indeed, numerous studies have demonstrated combinations of concentration
and time that did not cause a significant effect on the plant growth or yield.
Ozone induces a diverse range of effects on plants and plant communities.
These effects are usually classified as either injury or damage. Injury
encompasses all plant reactions such as reversible changes in plant metabolism
(e.g., altered photosynthesis), leaf necrosis, altered plant quality, or
reduced growth that does not impair yield or the intended use of the plant
(Guderian, 1977). In contrast, damage or yield loss includes all effects that
reduce or impair the intended use or the value of the plant. Thus, for example,
visible foliar injury to ornamental plants, detrimental responses in native
species, and reductions in fruit and grain production are all considered
damage or yield loss. Although foliar injury is not always classified as damage,
its occurrence is an indication that phytotoxic concentrations of 03 are present.
The occurrence of injury indicates that additional studies should be conducted
in areas where vegetation shows foliar injury to assess the risk of 0., to
vegetation and to determine if the intended use or value of the plants is being
impaired.
6.8.1 Limiting Values of Plant Response to Ozone
Several approaches have been used to estimate the 0, concentrations and
exposure durations that induce foliar injury. Most of these studies used
short-term exposures (less than 1 day) and measured visible injury as the
response variable. One method for estimating the 0, concentrations and exposure
durations that would induce specific amounts of visible injury involves exposing
plants to a range of 0, concentrations and exposure durations, and then evalua-
ting the data by regression analysis (Heck and Tingey, 1971). The data obtained
by this method for several species are summarized in Table 6-34 to illustrate
the range of concentrations required to induce foliar injury (5% and 20%) on
sensitive, intermediate, and less sensitive species.
6-223
-------
TABLE 6-34. OZONE CONCENTRATIONS FOR SHORT-TERM
OR 20
SENSI1
(ppm)
EXPOSURES THAT PRODUCE 5 OR 20 PERCENT INJURY TO VEGETATION
GROWN UNDER SENSITIVE CONDITIONS3
Ozone concentrations that may produce
5% (20%)
Exposure
time, hr
0.
1.
2.
4.
8.
5
0
0
0
0
Sensitive plants
0.
(0.
0.
(0.
0.
(0.
0.
(0.
0.
35 -
45 -
15 -
20 -
09 -
12 -
04 -
10 -
02 -
0.50
0.60)
0.25
0.35)
0.15
0.25)
0.09
0.15)
0.04
Intermediate
0.
(0.
0.
(0.
0.
(0.
0.
(0.
0.
55 -
65 -
25 -
35 -
15 -
25 -
10 -
15 -
07 -
0.
0.
0.
0.
0.
0.
0.
0.
0.
plants
70
85)
40
55)
25
35)
15
30)
12
sensi
Ł0.
Ł0.
Ł0.
Ł0.
Ł0.
injury:
Less
tive plants
70 (0.
40 (0.
30 (0.
25 (0.
20 (0.
85)
55)
40)
35)
30)
The concentrations in parenthesis are for the 20% injury level.
Source: U.S. Environmental Protection Agency (1978).
An alternative method for estimating the 0- concentrations and exposure
durations that induce foliar injury is the use of the limiting-value approach
(Jacobson, 1977). The limiting-value method, which was developed from a
review of the literature, identified the lowest concentration and exposure
duration reported to cause visible injury on various plant species. The
analysis was based on more than 100 studies of agricultural crops and 18
studies of tree species. The analysis yielded the following range of concen-
trations and exposure durations that were likely to induce foliar injury (U.S.
Environmental Protection Agency, 1978):
1. Agricultural crops:
a. 0.20 to 0.41 ppm for 0.5 hr.
b. 0.10 to 0.25 ppm for 1.0 hr.
c. 0.04 to 0.09 ppm for 4.0 hr.
2. Trees and shrubs:
a. 0.20 to 0.51 ppm for 1.0 hr.
b. 0.10 to 0.25 ppm for 2.0 hr.
c. 0.06 to 0.17 ppm for 4.0 hr.
6-224
-------
It should be emphasized that both methods described above can estimate
concentrations and exposure durations that might induce visible injury, but
that neither method can predict impacts of 0. on crop yield or intended use.
The concept of limiting values also was used to estimate the 07 concen-
trations and exposure durations that could potentially reduce plant growth and
yield (U.S. Environmental Protection Agency, 1978). The data were analyzed
and plotted in a manner similar to the approach used by Jacobson (1977)
(Figure 6-25). In Figure 6-25 the line bounds mean 0., concentrations and
exposure durations below which effects on plant growth and yield were not
detected. This graphical analysis used data from both greenhouse and field
studies and indicated that the lower limit for reduced plant performance was a
mean 0, concentration of 0.05 ppm for several hours per day for exposure
periods greater than 16 days. At 10 days the 0- response threshold increased
to about 0.10 ppm, and to about 0.30 ppm at 6 days.
6.8.2 Methods for Determining Op Yield Losses
Diverse experimental procedures have been used to study the effects of 0^
on plants, ranging from studies done under highly controlled conditions, to
exposures in open-top chambers, and to field exposures without chambers. In
general, the more controlled conditions are most appropriate for investigating
specific responses and for providing the scientific basis for interpreting and
extrapolating results. These systems are powerful tools for adding to an
understanding of the biological effects of air pollutants. To assess, however,
the impact of 0^ on plant yield and to provide data for economic assessments,
deviations from the typical environment in which the plant is grown should be
minimized. For field crops, this implies that the studies should be conducted
in the field, but for crops that are typically grown in glass houses, the
studies should be conducted under glass-house conditions.
To improve estimates of yield loss in the field, the National Crop Loss
Assessment Network (NCLAN) was initiated by EPA in 1980 to estimate the magnitude
of crop losses caused by 0, (Heck et al., 1982). The primary objectives of
NCLAN were:
1. To define the relationships between yields of major agricultural
crops and 03 exposure as required to provide data necessary for
economic assessments and the development of National Ambient
Air Quality Standards;
6-225
-------
1.0
-I I I II I
I I I I I I
E
a
a
Z
g
K
Z
HI
O
Z
o
o
LLJ
Z
o
N
O
— \
0.1
449 19O18 945
16DQ17
21D 11D
146 •48-52
\ 1514
40 • OK)
\ 12 13
\ 26
D31 7DD20
30 59
CD 29
39
\
EXPOSURE, hr/day
A < 1.99
D 2 TO 3.99
O 4 TO 5.99
4243«O9
33
44
38
546* 55, 56 •• 58
3«
2«
• 57
34 •
53
0.01
NOS. = REF. NOS. ON TABLE 11-4
I I I I I
1 1 1 1 1 1 1
8 10
20 40 60 80 100
EXPOSURE PERIOD, days
200
400
Figure 6-25. Relationship between ozone concentration,
exposure duration, and reduction in plant growth or yield (see
Table 6-18; also U.S. EPA, 1978).
Source: U.S. Environmental Protection Agency (1978).
6-226
-------
2. To assess the national economic consequences resulting from the
exposure of major agricultural crops to 0~;
O
3. To advance understandng of the cause and effect relationships that
determine crop responses to pollutant exposures.
In the NCLAN studies, the cultural conditions used approximated typical
agronomic practices, and open-top field exposure chambers were used to minimize
perturbations to the plant environment during the exposure. The studies have
attempted to use a range of realistic 0- concentrations and sufficient repli-
cation to permit the development of exposure-response models. In the NCLAN
studies, plants were exposed to a range of 0- concentrations. Chambers were
supplied with either charcoal-filtered air (control), ambient air, or ambient
air supplemented with 0~ to provide concentrations three or four levels greater
than ambient. Consequently, the 03 exposures were coupled to the ambient 03
level; days with the highest ambient 0, were also the same days when the
highest concentrations occurred in a specific treatment in a chamber. As the
ambient 0- varied from day-to-day, the base to which additional 0., was added
also varied. This coupling of the 0- exposures to the ambient environment
means that high 0, concentrations occurred in the chambers when the environ-
mental and air chemistry conditions, in the ambient air, were conducive for
producing elevated ambient (L levels. The plant response data have been
analyzed using regression approaches. The exposures were typically character-
ized by a 7-hr (9:00 a.m. to 4:00 p.m.) seasonal mean 0~ concentration. This
is the time period when 03 was added to the exposure chambers.
6.8.3 Estimates of Ozone-Induced Yield Loss
Yield loss is defined as an impairment or decrease in the intended use of
the plant. Included in the concept of yield loss are reductions in aesthetic
values, the occurrence of foliar injury (changes in plant appearance), and
losses in terms of weight, number, or size of the plant part that is harvested.
Yield loss may also include changes in physical appearance, chemical composi-
tion, or ability to withstand storage; which collectively are traits called
crop quality. Losses in aesthetic values are difficult to quantify. For
example, because of its aesthetic value, the loss of or adverse effect on a
specimen plant in a landscape planting may result in a greater economic loss
6-227
-------
than that incurred by the same impact on a plant of the same species growing
as a part of natural plant community. Foliar injury symptoms may decrease the
value of ornamental plants with or without concomitant growth reductions.
Similarly, foliar injury on crops in which the foliage is the marketable plant
part (e.g., spinach, lettuce, cabbage) can substantially reduce marketability
and thus can constitute yield loss. Attainment of the limiting values for
ozone previously discussed in this section should be sufficient to prevent
foliar injury and thereby reduce this type of yield loss. Most studies of the
relationship between yield loss and ozone concentration have focused on yields
as measured by weight of the marketable plant organ, and that kind of yield
loss will be the primary focus of this section.
Studies have been conducted, frequently using open-top field exposure
chambers, to estimate the impact of 03 on the yield of various crop species.
These studies can be grouped into two types, depending on the experimental
design and statistical methods used to analyze the data: (1) studies that
developed predictive equations relating CL exposure to plant response, and (2)
studies that compared discrete treatment levels to a control. The advantage
of the regression approach is that exposure-response models can be used to
interpolate results between treatment levels.
When the regression approach was used to estimate yield loss, CL was
added to either charcoal-filtered or ambient air to create a range of CL
concentrations. In summarizing the data, CL-induced yield loss was derived
from a comparison of the performance of the plants in charcoal-filtered air,
although other reference concentrations have been used. Various regression
techniques have been used to derive exposure-response functions. The use of
regression approaches permits the estimation of the 0- impact on plant yield
over the range of concentrations, not just at the treatment means as is the
case with analysis of variance methods.
6.8.3.1 Yield-loss: Determination by Regression Analysis. Examples of the
relationship between 0, concentration and plant yield are shown in Figures
6-26 and 6-27. These cultivars and species were selected because they also
illustrated the type of year-to-year variation in plant response to ozone that
may occur. The derived regression equations can be used to determine the
concentrations that would be predicted to cause a specific yield loss or to
estimate the predicted yield loss that would result from a specifc 0., concen-
tration. Both approaches have been used to summarize the data on crop responses
to CL using the Weibull function (Rawlings and Cure, 1985). As an example of
6-228
-------
6000
5000
ID
Ł
Ul
2 3000
2000
1000
(A)
SOYBEAN (DAVIS)
RALEIGH. 1981 AND 1982
2.144
1981 (Q)
y = 6693-«V0128>°872
I
0 0.02 0.04 0.060.08 0.1 0.12 0.14
03 CONCENTRATION, ppm
6000
5000
•* 4000
Q
ui
O
3000
2000
1000
WHEAT (ABE)
ARGONNE,
1962 AND 1983
1982(0)
y = 5235-<°3/0153)2272
1983(A)
,,14.4
0 0.020.040.060.080.10.120.14
03 CONCENTRATION, ppm
6000
SOOO
a
a 4000
ai
[Jj 3000
in
2000
1000
(B)X SOYBEAN (WILLIAMS)
BELTSVILLE. MD, 1981 AND 1932
1981(Q)
V = 4992-KV0.211|1 1
1982(A)
y = 6884-«V0.162)1'577
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
O3 CONCENTRATION, ppm
6000
5000
-4000
Q
Ł3000
2000
1000
(D)
WHEAT (ARTHUR 71)
ARGONNE, 1982 AND 1983
- 1983(A)
Ą = 5210-«V0.148)3781
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
03 CONCENTRATION, ppm
Figure 6-26. Examples of the effects of ozone on the yield of soybean and wheat
cultivars. The 03 concentrations are expressed as 7-hr seasonal mean concentrations.
The cultivars were selected as examples of O3 effects and of year-to-year variations in
plant response to 03.
Source: Soybean data from Heck et al. (1984b); wheat data from Kress et al. (1985).
6-229
-------
x
oi
JC
Q
UJ
UJ
w
Q
Z
6000
5500
5000
4500
4000
3500
3000
2500
2000
1500
COTTON (SJ-2)
SHAFTER. CA, 1981 AND 1982
1982(A)
= 5872-«V0.088)21
o
"a
x
3
x
CO
34
33
32
31
30
29
28
27
26
25
24
23
A(B)
TOMATO (MURIETTA)
TRACY. CA. 1981 AND 1982
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
03 CONCENTRATION, ppm
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
03 CONCENTRATION, ppm
a
a.
o>
CO
111
cc
o
o
cc.
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
J I
TURNIP (TOKYO CROSS)
RALEIGH. 1979 AND 1980
19801 Al
y=15.26-<°3/0094>394
1979(01
y = 4.05-((V0086>
I I I I
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
03 CONCENTRATION, ppm
Figure 6-27. Examples of the effects of ozone on the yield of cotton, tomato, and
turnip. The 03 concentrations are expressed as 7-hr seasonal mean concentrations.
The species were selected as examples of 03 effects and of year-to-year variations in
plant response to 03.
Source: Cotton and tomato data from Heck et al. (1984b); turnip data from Heagle et
al. (1985).
6-230
-------
response, the 0- concentrations that would be predicted to cause a 10 or 30
percent yield loss have been estimated (Table 6-35). A brief review of the
data in this table indicates that for some species mean yield reductions of 10
percent were predicted when the 7-hr seasonal mean 0- concentration exceeded
0.04 to 0.05 ppm. Concentrations of 0.028 to 0.033 ppm were predicted to
cause a 10 percent yield loss in Vona wheat, kidney bean, and Hodgson soybean.
At a 7-hr seasonal mean 0, concentration of 0.04 ppm, mean yield reductions
ranged from zero percent in sorghum, barley, and a corn cultivar to a high of
28.8 percent in Vona wheat.
A histogram of the 7-hr seasonal mean 03 concentrations predicted to
cause a 10 percent yield Toss (Table 6-35) is given in Figure 6-28 to help
illustrate the range of concentrations and their relative frequency of occur-
rence. The data in Figure 6-28 are based on 37 species or cultivar yield-
response functions developed from studies in open-top field exposure chambers.
Approximately 57 percent of the species or cultivars were predicted to exhibit
10 percent yield reductions at 7-hr seasonal mean concentrations below 0.05
ppm. Thirty-five percent of plant types were predicted to display a 10 percent
yield loss at 7-hr mean concentrations between 0.04 and 0.05 ppm. Seven-hr
seasonal mean concentrations in excess of 0.08 ppm were required to cause a 10
percent yield loss in almost 19 percent of the species or cultivars. The data
indicate that approximately 11 percent of the species or cultivars would
display a 10 percent loss at 7-hr seasonal mean concentrations below 0.035
ppm, suggesting that these plant types are very sensitive to 03-induced yield
losses.
A review of the data in Table 6-35 indicates that the grain crops were
apparently generally less sensitive than the other crops to 0~. Mean yield
reductions at 0.04 ppm were predicted to be less than 5 percent for all the
species and cultivars tested except for the Roland and Vona wheat cultivars.
The data also demonstrate that sensitivity differences within a species may be
as large as differences between species. For example, at 0.04 ppm 0,., estimated
yield losses ranged from 2 to 15 percent in soybean and from 0 to 28 percent
in wheat. In addition to differences in sensitivity among species and cultivars,
the data in Figures 6-26 and 6-27 illustrate year-to-year variations in plant
response to 0,.
Several exposure-response models, ranging from simple linear to complex
nonlinear models, have been used to describe the relationship between plant
yield and 0- exposure. When exposure-response models are used, it is important
6-231
-------
TABLE 6-35. SUMMARY OF OZONE CONCENTRATIONS PREDICTED TO CAUSE
10 PERCENT AND 30 PERCENT YIELD LOSSES AND SUMMARY OF YIELD LOSSES PREDICTED
TO OCCUR AT 7-hr SEASONABLE MEAN OZONE CONCENTRATIONS OF 0.04 and 0.06 ppma
03 concentrations, ppm,
predicted to cause
yield losses of:
Species
Legume crops
Soybean, Corsoy
Soybean, Davis (81)
Soybean, Davis (CA-82)
Soybean, Davis (PA-82)
Soybean, Essex
Soybean, Forrest
Soybean, Williams
Soybean, Hodgson
Bean, Kidney
Peanut, NC-6
Grain crops
Wheat, Abe
Wheat, Arthur 71
Wheat, Roland
Wheat, Vona
Wheat, Blueboy II
Wheat, Coker 47-27
Wheat, Holly
Wheat, Oasis
Corn, PAG 397
Corn, Pioneer 3780
Corn, Coker 16
Sorghum, DeKalb-28
Barley, Poco
Fiber crops
Cotton, Acala SJ-2 (81)
Cotton, Acala SJ-2 (82)
Cotton, Stoneville
Horticultural crops
Tomato, Murrieta (81)
Tomato, Murrieta (82)
Lettuce, Empire
Spinach, America
Spinach, Hybrid
Spinach, Viroflay
Spinach, Winter Bloom
Turnip, Just Right
Turnip, Pur Top W. G.
Turnip, Shogoin
Turnip, Tokyo Cross
10%
0.048
0.038
0.048
0.059
0.048
0.076
0.039
0.032
0.033
0.046
0.059
0.056
0.039
0.028
0.088
0.064
0.099
0.093
0.095
0.075
0.133
0.108
0.121
0.044
0.032
0.047
0.079
0.040
0.053
0.046
0.043
0.048
0.049
0.043
0.040
0.036
0.053
30%
0.082
0.071
0.081
0.081
0.099
0.118
0.093
0.066
0.063
0.073
0.095
0.094
0.067
0.041
0.127
0.107
0.127
0.135
0.126
0.111
0.175
0.186
0.161
0.096
0.055
0.075
0.108
0.059
0.075
0.082
0.082
0.080
0.080
0.064
0.064
0.060
0.072
Percent yield losses predicted
to occur at 7-hr seasonal
mean 03 concentration of:
0.04 ppm
6.4
11.5
6.4
2.0
7.2
1.7
10.4
15.4
14.9
6.4
3.3
4.1
10.3
28.8
0.5
2.2
0.0
0.4
0.3
1.4
0.0
0.0
0.0
8.3
16.1
4.6
0.8
10.3
0.0
6.8
2.6
6.0
5.8
7.7
10.1
13.0
3.3
0.06 ppm
16.6
24.1
16.5
10.4
14.3
5.3
18.1
18.4
28
19.4
10.4
11.7
24.5
51.2
2.8
8.4
0.9
2.4
1.5
5.1
0.3
2.7
0.5
16.2
35.1
16.2
3.7
31.2
16.8
17.2
9.2
16.7
16.5
24.9
26.5
29.7
15.6
The yield losses are derived from Weibull equations and are based on the control
yields in charcoal-filtered air.
Source: Derived from Heck et al. (1984b).
6-232
-------
en
i
no
oo
co
9-
1 8
E 8
c
*
(A 7 —
K '
3
ec **
O
o
Ul
w 3-
o.
O
C 2-
0 *
1-
0-
21.6%
13.5%
10.8% 10.8%
©
©
©
18.9%
13.6%
10.8%
©
©
I
< 0.035
T
0.040-0.044 I 0.050-0.059 | =10.080
0.035-0.039 0.045-0.049 0.060-O.079
T
7-hr SEASONAL MEAN
OZONE CONCENTRATIONS, ppm
Figure 6-28. Number and percentage of 37 crop species or cultivars predicted to
show a 10 percent yield loss at various ranges of 7-hr seasonal mean ozone
concentrations. Concentration ranges and 10% yield loss data are derived from
Table 6-35. Data represent 12 separate crop species; circled numbers represent
separate species for each concentration range.
-------
for the fitted equations not to show systematic deviation from the data points
2
and for the coefficient of determination (R ) to be high. Although linear
regression equations have been used to estimate yield loss, there appear to be
systematic deviations from the data for some species and cultivars even though
2
the equations have moderate-to-high coefficients of determination (R ).
Plateau-linear or polynomial equations appear to fit the data better. More
recently, a Weibull model has been used to estimate percentage yield loss
(Heck et al., 1983a). The Weibull model yields a curvilinear response line
that seems to provide a reasonable fit to the data. Based on available data,
it is recommended that curvilinear exposure-response functions be used to
describe and analyze plant response to 03.
6.8.3.2 Yield Loss: Determination from Discrete Treatments. In addition to
the use of regression approaches in some studies, various other approaches
have been used to investigate the effects of 0., on crop yield. These studies
were designed to test whether specific 0- treatments were different from the
control rather than to develop exposure-response equations. In general, these
data were analyzed using analysis of variance. To summarize the data from
studies that used discrete treatments, the lowest 03 concentration that signi-
ficantly reduced yield was determined from analyses done by the authors (Table
6-36). The lowest concentration reported to reduce yield was frequently the
lowest concentration used in the study; hence it was not always possible to
estimate a no-effect exposure concentration. In general, the data indicate
that 0- concentrations of 0.10 ppm (frequently the lowest concentration used
in the studies) for a few hours per day for several days to several weeks
generally caused significant yield reductions. Although it appears from this
analysis that a higher CL concentration was required to cause an effect than
was estimated from the regression studies, it should be noted that the concen-
trations derived from the regression studies were based on a 10 percent yield
loss, while in studies using analysis of variance (Table 6-36) the 0.10 ppm
concentration frequently induced mean yield losses of 10 to 50 percent.
6.8.3.3 Yield Loss: Determination with Chemical Protectants. Chemical
protectants (antioxidants) have been used to estimate the impact of ambient 0,
on crop yield. In these studies, some plots were treated with the chemical
and others were not. Yield loss was determined by comparing the yield in the
plots treated with the chemical to the yield in untreated plots. When chemical
protectants are used, care must be used in interpreting the data because the
chemical itself may alter plant growth. The chemical may not be effective
6-234
-------
TABLE 6-36. OZONE CONCENTRATIONS AT WHICH SIGNIFICANT YIELD LOSSES HAVE BEEN NOTED FOR
A VARIETY OF PLANT SPECIES EXPOSED UNDER VARIOUS EXPERIMENTAL CONDITIONS
Plant species
Exposure duration
Yield reduction,
% of control
03 concentration,
ppm
Reference
Alfalfa
Alfalfa
Pasture grass
Ladino clover
Soybean
Sweet corn
Sweet corn
Wheat
Radish
Beet
Potato
^ Pepper
r\3 Cotton
CjO
on Carnation
Coleus
Begonia
Ponderosa pine
Western white
pine
Loblolly pine
Pitch pine
Poplar
Hybrid poplar
Hybrid poplar
Red maple
American
sycamore
Sweetgum
White ash
Green ash
Willow oak
Sugar maple
7 hr/day, 70 days
2 hr/day, 21 day
4 hr/day, 5 days/wk, 5 wk
6 hr/day, 5 days
6 hr/day, 133 days
6 hr/day, 64 days
3 hr/day, 3 days/wk, 8 wk
4 hr/day, 7 day
3 hr
2 hr/day, 38 days
3 hr/day, every 2 wk,
120 days
3 hr/day, 3 days/wk, 11 wk
6 hr/day, 2 days/wk, 13 wk
24 hr/day, 12 days
2 hr
4 hr/day, once every 6 days
for a total of 4 times
6 hr/day, 126 days
6 hr/days, 126 days
6 hr/day, 28 days
6 hr/day, 28 days
12 hr/day, 5 mo
12 hr/day, 102 days
8 hr/day, 5 day/wk, 6 wk
8 hr/day, 6 wk
6 hr/day, 28 days
6 hr/day, 28 days
6 hr/day, 28 days
6 hr/day, 28 days
6 hr/day, 28 days
6 hr/day, 28 days
51, top dry wt
16 , top dry wt
20, top dry wt
20, shoot dry wt
55, seed wt/plant
45, seed wt/plant
13, ear fresh wt
30, seed yield
33, root dry wt
40, storage root dry wt
25, tuber wt
19, fruit dry wt
62, fiber dry wt
. 74, no. of flower buds
20, flower no.
55, flower wt
21, stem dry wt
9, stem dry wt
18, height growth
13, height growth
+1333, leaf abscission
58, height growth
50, shoot dry wt
37, height growth
9, height growth
29, height growth
17, total dry wt
24, height growth
19, height growth
12, height growth
0.10
0.10
0.09
0.10
0.10
0.10
0.20
0.20
0.25
0.20
0.20
0.12
0.25
0.05-0.09
0.20
0.25
0.10
0.10
0.05
0.10
0.041
0.15
0.15
0.25
0.05
0.10
0.15
0.10
0.15
0.15
Neely et al. (1977)
Hoffman et al. (1975)
Horsman et al. (1980)
Blum et al. (1982)
Heagle et al. (1974)
Heagle et al. (1972)
Oshima (1973)
Shannon and Mulchi (1974)
Adedipe and Ormrod (1974)
Ogata and Maas (1973)
Pell et al. (1980)
Bennett et al . (1979)
Oshima et al. (1979)
Feder and Campbell (1968)
Adedipe et al . (1972)
Reinert and Nelson (1979)
Wilhour and Neely (1977)
Wilhour and Neely (1977)
Wilhour and Neely (1977)
Wilhour and Neely (1977)
Wilhour and Neely (1977)
Patton (1981)
Patton (1981)
Dochinger and Townsend (1979)
Kress and Skelly (1982)
Kress and Skelly (1982)
Kress and Skelly (1982)
Kress and Skellv (1982)
Kress and Skelly (1982)
Kress and Skelly (1982)
-------
against all concentrations of all pollutants in the study area, which would
result in an underestimation of yield loss. With an understanding of these
limitations, however, researchers have concluded that chemical protectants are
an objective method of assessing the effects of (L on crop yield, especially
in conjunction with other methods. Results of several studies with chemical
protectants showed decreased crop yield from exposure to ambient oxidants
(Table 6-37). Crop yields were reduced 18 to 41 perecent when the ambient
oxidant concentration exceeded 0.08 ppm for 5 to 18 days over the growing
season of the crop.
6.8.3.4 Yield Loss: Determination from Ambient Exposures. A number of
research studies have demonstrated that ambient 0, concentrations in a number
of locations in the United States are sufficently high to impair plant yield.
Of studies to determine the impact of ambient oxidants (primarily 0,) on plant
yield, most have compared the yield differences between plants grown in ambient
air and those grown in charcoal-filtered air. Early research documented that
ambient oxidants reduced the yield and quality of citrus, grape, tobacco,
cotton, and potato (U.S. Environmental Protection Agency, 1978). Subsequent
studies substantiated the impacts of ambient oxidants on plant yield (Table
6-38). Over several years, bean yields varied from a 5 percent increase to a
22 percent decrease in response to 0., concentrations in excess of 0.06 ppm
(Heggestad and Bennett, 1981).
Studies conducted on eastern white pine in the southern Appalachian
mountains showed that ambient 03 may have reduced the radial growth of sensitive
individuals as much as 30 to 50 percent annually over the last. 15 to 20 years
(Mann et al. , 1980). Field studies in the San Bernardino National Forest
showed that during the last 30 years ambient 0- may have reduced height growth
of ponderosa pine by as much as 25 percent, radial growth by 37 percent, and
the total wood volume produced by 84 percent (Miller et al., 1982). Calcula-
tions of biomass in these studies were based, however, on apparent reductions
in radial growth without standardization of radial growth data with respect to
tree age.
6.8.3.5 Yield Loss Summary. Several general conclusions can be drawn from
the various approaches used to estimate crop yield loss. The data from the
comparisons of crop yield in charcoal-filtered and unfiltered air (ambient
exposures) clearly show that ambient levels of 0~ are sufficiently elevated in
several parts of the country to impair the growth and yield of plants. The
data from the chemical protectant studies support and extend this conclusion
6-236
-------
TABLE 6-37. EFFECTS OF OZONE ON CROP YIELD
AS DETERMINED BY THE USE OF CHEMICAL PROTECTANTS*
Yield reduction,
Species % of control
03 exposure,
ppm
Reference
Beans (green)
Onion
Tomato
Bean (dry)
Tobacco
Potato
Potato
41
38
30
24
18
36
25
>0.08 for total
of 27 hr over
3.5 months
>0.08 on 5 days out
of 48
>0.08 on 15 days
over 3 months
>0.08 on 11 days
(total of 34 hr)
over 3 months
>0.08 on 14 days
during the summer
>0.08 ppm on 18 days
(total of 68 hr)
over 3 months
Manning et al. (1974)
Wukasch and Hofstra (1977b)
Legassicke and Ormrod (1981)
Temple and Bisessar (1979)
Bisessar and Palmer (1984)
Bisessar (1982)
Clarke et al. (1983)
All the species were treated with the antioxidant, EDU, except the bean study by
Manning et al. (1974) which used the systemic fungicide, benomyl.
Yield reduction was determined by comparing the yields of plants treated with
chemical protectants (control) to those that were not treated.
cThis study was run over 2 years when the 03 doses were 65 and 110 ppm-hr,
respectively, but the yield loss was similar both years.
6-237
-------
TABLE 6-38. EFFECTS OF AMBIENT OXIDANTS ON YIELD OF SELECTED CROPS
Plant species
Tomato
(Fireball 861 VR)
Bean
(Tendergreen)
concentration,
ppm
0.035
(0.017-0.072)
0.041
(0.017-0.090)
Exposure duration
99 day average
(6:00 a.m. - 9:00 p.m.)
43 day average
(6:00 a.m. - 9:00 p.m.)
Yield, %
reduction
from control
33, fruit fresh
wt
26, pod fresh wt
Location
of study
New York
Reference
Maclean and
Schneider (1976)
Snap bean (3 cultivars:
Astro, BBL 274, BBL
290)
Soybean (4 cultivars:
Cutler, York, Clark,
en Dare)
CO
oo
Forbs, grasses, sedges
Sweet corn
(Bonanza)
0.042
>0.05
0.052
0.051
0.035
>0.08
3 mo average
(9:00 a.m. - 8:00 p.m.)
31% of hr between
8:00 a.m. - 10:00 p.m.
from late June to mid-
September over three
summers; 5% of the time
the concentration was
>0.08 ppm
1979, 8 hr/day average
(10:00 a.m. - 6:00 p.m.),
April-September
1980, 8 hr/day average
(10:00 a.m. - 6:00 p.m.),
April-September
1981, 8 hr/day average
(10:00 a.m. - 6:00 p.m.),
April-September
58% of hr (6:00 a.m.
9:00 p.m.),
1 July-6 September
1, pod wt
20, seed wt
32, total above-
ground biomas
20, total above-
ground biomass
21, total above-
ground biomass
9, ear fresh wt
Maryland
Maryland
Virginia
Virginia
California
Heggestad and
Bennett (1981)
Howell et al.
(1979); Howell
and Rose (1980)
Duchelle et al.
(1983)
Thompson et al.
(1976a)
(Monarch Advance)
>0.08
28, ear fresh wt
-------
to other plant species. Both approaches indicate that the effects occur at
low mean concentrations, with only a few 0, occurrences greater than 0.08 ppm.
Growth and yield data from the previous criteria document (U.S. Environmental
Protection Agency, 1978), shown in Figure 6-25, indicate that effects on
growth and yield of several plant species occurred when the mean 03 concentra-
tion (for 4 to 6 hr/day) exceeded 0.05 ppm for at least 2 wk. The data from
the regression studies, conducted to develop exposure-response functions for
estimating yield loss, indicated that at least 50 percent of the species/culti-
vars tested were predicted to display a 10 percent yield loss at 7-hr seasonal
mean 03 concentrations of 0.05 ppm or less. Most of the data from the discrete
treatment studies did not use levels low enough to support these values directly.
The magnitude of yield losses reported at 0.10 ppm, however, indicate that
maintenance of a substantially lower concentration than 0.10 ppm is needed to
prevent 03 effects, although a specific value cannot be derived from the
discrete treatment studies.
6.8.4 Effects on Crop Quality
Based on results of the few studies that have been conducted, 0, can
reduce crop quality in addition to reducing the total yield of the crop.
Quality is a general term that includes many features of the crop, such as
nutritional composition, appearance, taste, and ability to withstand storage
and shipment. Examples of O^-induced alterations in quality are decreased oil
in soybean seeds (Howell andt Rose, 1980; Kress and Miller, 1983); decreased
p-carotene, vitamin C, and carbohydrates in alfalfa (Thompson et al., 1976b;
Neely et al., 1977); and increased reducing sugars that are associated with
undesirable darkening when potatoes are used to make potato chips (Pell et
al., 1980).
6.8.5 Statistics Used to Characterize Ozone Exposures
The characterization and representation of plant exposures to 0- has
been, and continues to be a major problem. Research has not yet clearly
identified which components of the pollutant exposure cause the plant, response.
Most studies have characterized the exposure by the use of mean 0, concentra-
tions, although various averaging times have been used. Some studies have
also used cumulative 0- dose. The difficulty of selecting an appropriate
statistic to characterize plant exposure has been summarized by Heagle and
Heck (1980). Ambient and experimental 03 exposures have been presented as
6-239
-------
seasonal, monthly, weekly, or daily means; peak hourly means; number of hours
above a selected concentration; or the number of hours above selected concen-
tration intervals. None of these statistics adequately characterize the
relationships among 0- concentration, exposure duration, interval between
exposures, and plant response. The use of a mean concentration (with long
averaging times) (1) implies that all concentrations of 03 are equally effec-
tive in causing plant responses and (2) minimizes the contributions of the
peak concentrations to the resonse. The mean treats low-level, long-term
exposures the same as high-concentration, short-term ones. Thus, the use of a
long-term mean concentration ignores the importance of peak concentrations; to
i
ignore the peaks is inconsistent with the literature.
The total ozone dose (concentration multiplied by time) has been used to
describe plant exposure; however, it suffers from the same problem as the
mean. The total dose is simply the summation of the ppm-hr over the study
period, which also treats all concentrations as being equally effective.
Several investigators have attempted to give greater importance to peak 0-
concentrations. For example, Oshima et al. (1977a,b) and Lefohn and Benedict
(1982) have summed only the ppm-hr of exposure greater than some preselected
value. Larsen et al. (1983) have introduced the concept of "impact" to describe
the effects of 0~ and S02 on soybeans. The "impact (I)" is calculated similarly
to total dose, except the concentration is raised to an exponent greater than
W
one (I = C XT); this method of calculation effectively gives greater weight
to the higher concentrations. More recently, Larsen and Heck (1984) have
suggested the term "effective mean" to describe an approach in which greater
importance is given to higher concentrations. The "effective mean" is defined
as the average hourly impact raised to an exponent and divided by the duration.
Several lines of evidence suggest that higher concentrations should be
regarded as having the greater influence in determining the impact of 0, on
vegetation. Studies have shown that plants can tolerate some combinations of
exposure duration and concentration without exhibiting foliar injury or effects
on growth or yield, illustrating that not all concentrations are equally
effective in causing a response. From the toxicological perspective, it is
the peaks or concentrations above some level that are most likely to have an
impact. Effects occur on vegetation when the amount of pollutant that the
plant has absorbed exceeds the ability of the organism to repair or compensate
for the impact.
6-240
-------
Studies with beans and tobacco (Heck et al., 1966) showed that a dose
(concentration times time) distributed over a short period induced more injury
than did the same dose distributed over a longer period. Tobacco studies
showed that the 0- concentration was substantially more important than exposure
duration in causing foliar injury (Tonneijck, 1984). In beans, foliar injury
2
occurred when the internal 03 flux exceeded 115 pmoles/m in 1 hr (Bennett,
19)9). A single 3-hr exposure, however, at approximately half the concentration
(0.27 compared with 0.49 ppm) required a 64 percent greater internal flux of
OT to produce the same amount of foliar injury as the 1-hr exposure required.
More recently, Amiro et al. (1984) showed that higher concentrations were more
important than low concentrations in causing injury. Their study also suggested
the existence of a biochemical injury threshold (i.e., the 0- uptake rates
that plants can experience without incurring visible foliar injury). The
greater importance of concentration compared to exposure duration has also
been reported by other authors (e.g., Heck and Tingey, 1971; Henderson and
Reinert, 1979; Reinert and Nelson, 1979).
Studies with soybean (Johnston and Heagle, 1982), tobacco (Heagle and
Heck, 1974), and bean (Runeckles and Rosen, 1977) showed that plants exposed
to a low level of 0- for a few days became more sensitive to subsequent 03
exposures. In studies with tobacco, Mukammal (1965) showed that a high 03
concentration on one day caused substantial injury, whereas an equal or higher
concentration on the second day caused only slight injury. Using stress
ethylene as an indicator, of 0- effects, Stan and Schicker (1982) showed that a
series of successive short exposures was more injurious to plants than a
continuous exposure at the same 0~ concentration for the same total exposure
period. Walmsley et al. (1980) continuously exposed radishes to 0- for several
weeks and found that the plants acquired some 0- tolerance. The acquired
tolerance displayed two components: (1) the exposed plants developed new
leaves faster than the controls, and (2) there was a progressive decrease in
sensitivity of the new leaves to 0~. The newer leaves; also displayed a. slower
rate of senescence. The observations by E.lkiey and Ormro'd (1981) .that the 0,
uptake decreased during a 3-day study period may provide an explanation for
the results with radish. • . ••• . .
Not only are concentration and time important but the dynamic nature of
the 0, exposure is also important; i.e. whether the exposure is at a constant
or variable concentration. Musselman et al. (1983) recently showed that
constant concentrations of 0- caused the same types of plant responses as
6-241
-------
variable concentrations at equivalent doses. Constant concentrations, however,
had less effect on plant growth responses than variable concentrations at
similar doses. Exposures of radishes to ambient 0, in open-top exposure
chambers showed that significant yield reductions occurred when the maximum 03
concentration exceeded 0.06 ppm at least 10 percent of the days when the crop
was growing (Ashmore, 1984). Initial studies have compared the response of
alfalfa to daily peak and episodic 03 exposure profiles that gave the equivalent
total 03 dose over the growing season (Hogsett et al., 1985). Alfalfa yield
was reduced to a greater extent in the episodic than in the daily peak exposure.
This study also illustrates the problem with the 7-hr seasonal mean concentra-
tion; i.e., it does not properly account for the peak concentrations. The
plants that displayed the greater growth reduction (in the episodic exposure)
were exposed to a significantly lower 7-hr seasonal mean concentration.
Studies with SC^ also showed that plants exposed to variable concentrations
exhibited a greater plant response than those exposed to a constant concentra-
tion (Mclaughlin et al., 1979; Male et al., 1983).
6.8.6 Relationship Between Yield Loss and Foliar Injury
Because plant growth and production depend on photosynthetically functional
leaves, various studies have been conducted to determine the association
between foliar injury and yield for species in which the foliage is not part
of the yield. Some research has demonstrated significant yield loss with
little or no foliar injury (e.g., Tingey et al., 1971a; Tingey and Reinert,
1975; Kress and Skelly, 1982; Feder and Campbell, 1968; Adedipe et al. , 1972).
Other studies showed that significant foliar injury was not always associated
with yield loss (Heagle et al., 1974; Oshima et al., 1975). The relative
sensitivities of two potato cultivars were reversed when judged by foliar
injury versus yield reductions (Pell et al., 1980). In field corn, foliar
injury occurred at a lower 0- concentration than yield reductions; but as the
03 concentration increased, yield was reduced to a greater extent than foliar
injury was increased (Heagle et al., 1979a). In wheat, foliar injury was not
a good predictor of 03~induced yield reductions (Heagle et al., 1979b).
6.8.7 Physiological Basis of Yield Reductions
As discussed earlier in this summary, plant growth is the summation of a
series of biochemical and physiological processes related to uptake, assimila-
tion, biosynthesis, and translocation. An impairment in these processes may
lead to reduced plant yield if the process is limiting.
6-242
-------
For plant growth to occur, plants must assimilate C0? and convert it into
organic substances; an inhibition in carbon assimilation may be reflected in
plant growth or yield. In several species 0- (at 0.05 ppm and higher) inhibited
photosynthesis, as measured by gas-exchange (e.g., U.S. Environmental Protection
Agency, 1978; Coyne and Bingham, 1978; Black et al., 1982; Bennett and Hill,
1974; Yang et al., 1983). Biochemical studies showed that 03 (0.12 ppm for 2
hr) inhibited an enzyme that catalyzes the assimilation of CO,, (Pell and
Pearson, 1983).
Ozone, in addition to decreasing the total amount of C02 that is assimi-
lated, alters that pattern by which the reduced amount of assimilate is parti-
tioned throughout the plant. There is generally less photosynthate translo-
cated to the roots and to the reproductive organs (e.g., Tingey et al., 1971a;
Jacobson, 1982; Oshima et al., 1978, 1979; Bennett et al., 1979). This reduces
root size and marketable yield as well as rendering the plant more sensitive
to injury from environmental stresses. Another consequence of reduced root
growth and altered carbon allocation is an impairment of symbiotic nitrogen
fixation (U.S. Environmental Protection Agency, 1978; Ensing and Hofstra,
1982).
The reproductive capacity (flowering and seed set) is reduced by 0., in
ornamental plants, soybean, corn, wheat, and other plants (Adedipe et al.,
1972; Feder and Campbell, 1968; Heagle et al., 1972, 1974; Shannon and Mulchi,
1974). These data suggest that 03 impairs the fertilization process in plants.
This suggestion has been confirmed in tobacco and corn studies using low
concentrations of 0, (0.05 to 0.10 ppm) for a few hours (Feder, 1968; Mumford
et al., 1972).
Ozone both in the field and in chamber studies stimulates premature
senescence and leaf drop (Menser and Street, 1962; Heagle et al., 1974;
Heggestad, 1973; Pell et al., 1980; Hofstra et al., 1978). In part, the
0.,-induced yield reduction has been attributed to premature senescence. The
premature leaf drop decreases the amount of photosynthate that a leaf can
contribute to plant growth.
6.8.8 Factors Affecting Plant Response to Ozone
Numerous factors influence the type and magnitude of plant response to
0,. Most studies of the factors influencing plant response have been limited
to effects on foliar injury; however, some studies have measured yield and
6-243
-------
some have researched the physiological basis for the influences. The para-
meters studied include environmental factors, biological factors, and inter-
actions with other air pollutants.
6.8.8.1 Environmental Conditions. Environmental conditions before and during
plant exposure are more influential than post-exposure conditions in determining
the magnitude of the plant response. The influence of environmental factors
has been studied primarily under controlled conditions, but field observations
have substantiated the results. Most studies have evaluated the influence of
only a single environmental factor and have relied primarily upon foliar
injury as the plant response measure. Some generalizations of the influence
of environmental factors can be made:
1.: Light conditions that are conducive to stomatal opening appear to
enhance 03 injury (U.S. Environmental Protection Agency, 1978).
Light is required to induce stomatal opening, which permits the
plant to absorb pollutants.
2. No consistent pattern relating plant response to temperature has
been observed (U.S. Environmental Protection Agency, 1978). Plants
do not appear to be as sensitive at extremely high or low tempera-
tures, however, as they are under more moderate conditions.
3. Plant injury tends to increase with increasing relative humidity
(U.S. Environmental Protection Agency, 1978). The relative humidity
effect appears to be related to stomatal aperture, which tends to
increase with increasing relative humidity. McLaughlin and Taylor
(1981) demonstrated that plants absorb significantly more 0, at high
humidity than at low humidity. It is generally accepted that plants
in the eastern United States are injured by lower concentrations of
0- than their counterparts in California; this phenomenon has been
attributed to differences in humidity (U.S. Environmental Protection
Agency', 1978). ^ • • ' •
4. As soi-1 moisture decreases,' plant water stress increases and there
is a reduction In'- plant sensitivity to. 03 (U;S. Environmental Protec-
tion Agency, 1978). The reduced 03 sensitivity is apparently related
to stomatal closure, which reduces 0., uptake (U.S. Environmental
Protection Agency, 1978; Olszyk and Tibbitts, 1981; Tingey et al.,
1982). Water stress does not confer a permanent tolerance to 0,;
6-244
-------
once the water stress has been alleviated, the plants regain their
sensitivity to 0- (Tingey et al., 1982).
6.8.8.2 Interaction with Plant Diseases. Ozone can affect the development of
disease in plant populations. Laboratory evidence suggests that 0~ (at ambient
concentrations or greater for 4 hr or more) inhibits infection by pathogens
and subsequent disease development (Laurence, 1981; Heagle, 1982; U.S. Environ-
mental Protection Agency, 1978). Increases, however, in diseases from "stress
pathogens" have been noted. For example, plants exposed to 0- were more
readily injured by Botrytis than plants not exposed to 0- (Manning et al.,
1970a,b; Wukasch and Hofstra, 1977a,b; Bisessar, 1982). Both field and labora-
tory studies have confirmed that the roots and cut stumps of 03-injured ponderosa
and Jeffrey pines are more readily colonized by a root rot (Heterobasidion
annosus). The degree of infection was correlated with the foliar injury
(James et al., 1980a; Miller et al., 1982). Studies in the San Bernardino
National Forest showed that (L-injured trees were predisposed to attack by
bark beetles and that fewer bark beetles were required to kill an 0,-injured
tree (Miller et al., 1982).
6.8.8.3 Interaction of Ozone with Other Air Pollutants. The report, of Menser
and Heggestad (1966) provided the initial impetus for studying the interaction
of 03 with SCL. They showed that Bel W-3 tobacco plants exposed to 03 (0.03
ppm) or S02 (0.24 to 0.28 ppm) were uninjured but that substantial foliar
injury resulted when the plants were exposed to both gases simultaneously.
Subsequent studies have confirmed and extended the observation that combinations
of 0, and SQp may cause more visible injury than expected based on the injury
from the individual gases. This injury enhancement (synergism) is most common
at low concentrations of each gas and also when the amount of foliar injury
induced by each gas, individually, is small. At higher concentrations or when
extensive injury occurs, the effects of the individual gases tend to be less
than additive (antagonistic). In addition to foliar injury, the effects of
pollutant combinations have also been investigated in relation to other plant
effects, and these have been discussed in several reviews and numerous individual
reports (e.g., Reinert et al., 1975; Ormrod, 1982; Jacobson and Colavito, 1976;
Heagle and Johnston, 1979; Olszyk and Tibbitts, 1981; Flagler and Youngner,
1982a; Foster et al., 1983b; Heggestad and Bennett, 1981; Heagle et al., 1983a).
6-245
-------
Field studies have investigated the influence of SO,, on plant response to
0~ at ambient and higher concentrations in several plant species: soybean
(Heagle et al., 1983c; Reich and Amundson, 1984), beans (Oshima, 1978; Heggestad
and Bennett, 1981), and potatoes (Foster et al., 1983b). In these studies, 0-
altered plant yield but S02 had no significant effect and did not interact
with Oo to reduce plant yield unless the SOp exposure concentrations and
frequency of occurrence were much greater than the concentrations and fre-
quencies of occurrence typically found in the ambient air in the United States.
The applicability of the yield results from pollutant combination studies
to ambient conditions is not known. An analysis of ambient air monitoring
data for instances of co-occurrence of 0- and SO,, indicated that at sites
where the two pollutants were monitored, they both were present for ten or
fewer periods during the growing season (Lefohn and Tingey, 1984). Co-
occurrence was defined as the simultaneous occurrence of hourly averaged
concentrations of 0.05 ppm or greater for both pollutants. At this time, it
appears that most of the studies of the effects on pollutant combinations (03
and SOp) on plant yield have used a longer exposure duration and a higher
frequency of pollutant co-occurrence than are found in the ambient air.
Only a few studies have investigated the effects of 0, when combined with
pollutants other than S0?, and no clear trend is available. Preliminary
studies using three-pollutant mixtures (03, S02, NOp) showed that the additions
of SOp and NOp (at low concentrations) caused a greater growth reduction than
Oo alone.
6.8.9 Economic Assessment of Effects of Ozone on Agriculture
Evidence from the plant science literature clearly demonstrates that fl-
at ambient levels will reduce yields of some crops (see Section 6.4.3.2.2).
In view of the importance of U.S. agriculture to both domestic and world
consumption of food and fiber, such reductions in crop yields could adversely
affect human welfare. The plausibility of this premise has resulted in numerous
attempts to assess, in monetary terms, the losses from ambient 0, or the
benefits of 0, control to agriculture. Many of these assessments have been
performed since publication of the 1978 03 criteria document (U.S. Environmental
Protection Agency, 1978). The utility of these post-1978 studies in regulatory
decision-making can be evaluated in terms of how well the requisite biological,
aerometric, and economic inputs conform to specific criteria, as discussed in
Section 6.5.
6-246
-------
While a complete discussion of the criteria for evaluating economic
assessments is not appropriate here, it is instructive to highlight certain
key issues. First, the evidence on crop response to 0- should reflect how
*3
crop yields will respond under actual field conditions. Second, the air
quality data used to frame current or hypothetical effects of Cu on crops
should represent the actual exposures sustained by crops in each production
area. Finally, the assessment methodology into which such data are entered
should (1) capture the economic behavior of producers and consumers as they
adjust to changes in crop yields and prices that may accompany changes in fl-
air quality; and (2) ideally, should accurately reflect institutional considera-
tions, such as regulatory programs, that may result in market distortions.
The assessments of 0, damages to agriculture found in the literature
display a range of procedures for calculating economic losses, fro'in simple
monetary calculation procedures to more complex economic assessment methodol-
ogies. The simple procedures calculate monetary effects by multiplying
predicted yield or production changes resulting from exposure to 0, by an
assumed constant crop price, thus failing to recognize possible crop price
changes arising from yield changes as well as not accounting for the processes
underlying economic response. Conversely, a rigorous economic assessment will
provide estimates of the benefits of air pollution control that account for
producer-consumer decision-making processes, associated market adjustments,
and perhaps some measure of distributional consequences between affected
parties. It is important to distinugish between those studies based on naive
or simple models and those based on correct procedures, since the naive proce-
dure may be badly biased, leading to potentially incorrect policy decisions.
Most of the post-1978 economic assessments focus on 0- effects in specific
regions, primarily California and the Corn Belt (Illinois, Indiana, Iowa,
Ohio, and Missouri). This regional emphasis may be attributed to the relative
abundance of data on crop response and air quality for selected regions, as
well as the national importance of these agricultural regions. [Iconomic
estimates for selected regions are presented in Table 6-39. In addition to
reporting the monetary loss or benefit estimates derived from each assessment,
this table provides some evaluation of the adequacy of the plant science,
aerometric, and economic data, and assumptions used in each assessment.
Adequacy as defined here does not mean that the estimates are free of error;
rather, it implies that the estimates are based on the most defensible biologic,
6-247
-------
TABLE 6-39. SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Reference and
study region
Crops
Annual benefits
of control,
$ mil lion
Evaluation of critical data and assumptions
Plant response data
Aerometric data
Economic model data
Additional comments
Adams et al.
(1982);
Southern
California
12 annual crops:
beans, broccoli,
cantaloupes,
carrots, cauli-
flower, celery,
lettuce, onions,
potatoes, tomatoes
cotton, and sugar
beets.
$45 (in 1976 Inadequate; uses Larsen- Adequate; exposure
dollars) Heck (1976) foliar injury measured as cumu-
models converted to yield lative seasonal
losses. exposure in
excess of Cali-
fornia standard
(0.08 ppm), from
hourly data col-
lected for sites
closest to produc-
tion regions.
Adequate; a price endo-
genous mathematical
(quadratic) programming
model reflecting agro-
nomic, environmental,
and economic conditions
in 1976.
Economic effect measured as a
change in economic surplus (sum
of consumers and producers'
surpluses) between base case
(actual 03 levels in 1976)
and economic surplus that
would be realized if all
regions were in compliance with
1971 photochemical oxidant
standard of 0.08 ppm.
en
i
ro
•F*
CO
Lueng et al. 9 crops: lemons, $103 (in 1975
(1982); oranges (Valencia dollars)
Southern and Navel), straw-
California berry, tomato,
alfalfa, avocado,
lettuce, and celery.
Inadequate; 03-yield
response functions
estimated from second-
ary data on crop yields.
Adequate for some
regions; exposure
measured in aver-
age monthly con-
centration in ppm
for 12 hr period
(7:00 a.m. to
7:00 p.m. ). Data
from 61 Calfornia
Air Resources
Board monitoring
sites.
Adequate on demand side;
economic model is
composed of linear
supply and demand
curves for each crop
estimated with data
from 1958-1977, but
ignores producer-level
adjustments.
Economic effect is measured as
a change in economic surplus
between base case (1975) and a
clean air environment reflecting
zero 03.
Howitt et al.
(1984a,b);
California
13 crops: alfalfa,
barley, beans,
celery, corn,
cotton, grain sor-
ghum, lettuce,
onions, potatoes,
rice, tomatoes,
and wheat.
From $35 (bene-
fit of control
to 0.04 ppm) to
$157 (loss for
increase to
0.08 ppm) (in
1978 dollars).
Adequate for some crops;
most response functions
derived from NCLAN data
through 1982. Surrogate
responses used for celery,
onions, rice and potatoes
are questionable.
Adequate; Califor-
nia Air Resources
Board data for
monitoring sites
closest to rural
production areas.
Exposure measured
as the seasonal
7-hr average in
each production
area for compati-
bility with NCLAN
exposure.
Adequate; economic model
similar to Adams et al.
(1982) but includes some
perennial crops and re-
flects 1978 economic and
technical environment.
Economic effects measured as
changes in economic surplus
across three 03 changes from
1978 actual levels. These
include changes in ambient 03
to 0.04, 0.05, and 0.08 ppm
across all regions.
-------
TABLE 6-39 (cont'd). SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Reference and
study region
Rowe et al .
(1984);
San Joaquin
Valley in
Cal ifornia
cr>
f
ro
-c*
10
Adams and
McCarl
(1985);
Corn Belt
Annual benefits
of control ,
Crops $ mi 1 lion
14 annual and $43 to $117
perennial crops: depending on
alfalfa, barley, degree of
beans, carrots, control,
corn, cotton, measured in
grain sorghum, 1978 dollars.
grass hay, grapes,
pasture, potatoes,
saff lower,
tomatoes and
wheat.
3 crops: corn, $668 (in 1980
soybeans, and dollars)
wheat.
Evaluation of critical data and assumptions3
Plant response data
Adequate for some crops;
response functions based
on both experimental and
secondary data. Most
crops from NCLAN data.
Responses for the remain-
ing crops were based on
surrogate responses of
similar crops in the
data set.
Adequate; 03 yield
response information
from NCLAN for 3 yr
(1980-1982). Yield
adjustments estimated
fromWeibull response
models.
Aerometric data
Adequate; 4 expo-
sure levels were
tested. The aver-
age hourly concen-
tration was used
in most functions
to predict changes.
Al 1 data were from
California Air
Resources Board
monitoring sites in
predominantly rural
areas.
Adequate except
for 1 inkage of
7-hr seasonal
mean to hourly
standards. Data
are interpolated
from SAROAD
monitoring sites
by Kriging.
procedure,
measured as
1980 seasonal
7-hr average.
Regulatory
analysis assumes
that 03 is log-
normally
distributed.
Economic model data
Adequate; same as in
Howitt et al.
(1984a,b).
Adequate; economic
estimates are generated
by a mathematical pro-
gramming model of U.S.
agriculture reflecting
1980 conditions. Farm-
level response is
portrayed by 12
individual "represen-
tative" farm models
to generate supply
adjustments used in
the national-level
mode 1 .
Additional comments
Economic effects measured as the
change in economic surplus be-
tween the 1978 base case and three
increasingly stringent control
scenarios: (1) a 50% reduction in
in no. of hr >0. 10 ppm; (2)
meeting the current standard of
0.10 ppm; and (3) meeting an 03
standard of 0.08 ppm.
Economic estimates represent
changes in economic surplus
(sum of consumers' and pro-
ducers' surpluses) between
current (1980) 03 levels and
increases and decreases in.
ambient 03 levels. Reduction
to a uniform ambient level of
0.04 ppm across all regions
results in benefits of $668
mi 11 ion.
-------
TABLE 6-39 (cont'd). SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Reference and
study region
Crops
Annual benefits
of control,
$ million
Evaluation of critical data and assumptions
Plant response data
Aerometric data
Economic model data
Additional comments
Mjelde et al.
(1984);
Illinois
3 crops: corn,
soybeans, and
wheat.
Ranges from
$55 to $220
annually for
period 1976
to 1980.
cr>
i
ro
en
o
Adequate when cross-
checked against NCLAN
data; responses are
estimated from secon-
dary (non-experimental)
data on actual farmer
yield, input, and 03
concentrations. Results
are translated into yield
effects and compared to
NCLAN data from Illinois.
Adequate; same
Kriged data set as
used in Adams and
McCarl (1985),
except only for
Illinois and
cover 5 yr
(1976-1980).
Exposure is mea-
sured as seasonal
7-hr average to
facilitate compa-
rison with NCLAN
response estimates.
Adequate at producers
level; economic model
consists of a series
of annual relationships
on farmers' profits
These functions).
These functions are
adjusted to represent
changes in 03 (±25%)
for each year. Model
does not include consumer
(demand) effects.
The estimates represent increases
in farmers' profits that could
arise for a 25X reduction in 03
for each year (1976-1980). Years
with higher ambient levels have
highest potential increase in
profits for changes.
Page et al.
(1982);
Ohio River
Basin
3 crops: corn,
soybeans and
wheat.
$7.022 measured
as present
value of pro-
ducer losses
for period
1976 to 2000:
Annualized
losses are
approx. $270
in 1976
dollars.
Inadequate; crop losses
provided by Loucks and
Armentano (1982);
responses derived by
synthesis of existing
experimental data.
Inadequate; dose
measured as cumu-
lative seasonal
exposure for a
7-hr period
(9:30 a.m. to
4:30 p.m. )
Monitoring sites
at only 4 loca-
tions were used
to characterize
the regional
exposure.
Inadequate; the econo-
mic model consi-sts of
regional supply curves
for each crop. The
predicted changes in
production between
"clean air" case and
each scenario are used
to shift crop supply
c-urves. The analysis
ignores price changes
from shifts in supply.
Losses are measured as differ-
ences in producer surplus across
the various scenarios. Since
prices are assumed fixed (in
real terms) over the period,
no consumer effects are
measured.
-------
TABLE 6-39 (cont'd).. SUMMARY OF ESTIMATES OF REGIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Reference and
study region
Crops
Annual benefits
of control ,
$ million
Evaluation
Plant response data
of critical data and
Aerometric data
assumptions
Economic model data
Additional
comments
CTl
no
cn
Benson et al. 4 crops: alfalfa,
(1982); wheat, corn, and
Minnesota potatoes. Cultivar
believed to be
limited to one per
crop.
$30.5 (measured Inadequate; but innova-
in 1980 dollars) tive crop loss models
estimated using experi-
mental yield-03 data
from other researchers.
Crop loss modeling
includes both chronic
and espisodic response
and crop development
stage as factors in
yield response, by
regressing yield on 03
exposures for various
time windows, during the
growing season.
Adequate; air
quality data are
for state of
Minnesota for
1979 and 1980.
Exposure measured
several ways but
generally as a
daily exposure sta-
tistic reflecting
either sum of hourly
averages or the mean
hourly average.
Adequate on demand side;
The economic estimates
are derived from a
comprehensive economic
model calibrated to
1980 values.
The economic effect measured
in terms of short-run profit
changes for Minnesota producers.
If yields are assumed to change
only in Minnesota then losses to
Minnesota producers are $30.5
million. If yields change in
Minnesota and the rest of U.S. ,
then producers gain $67 million
as a result of increases in crop
prices.
Adequacy as defined here does not mean that the estimates are free of error; rather, it implies that the estimates are based
on the most defensible biologic, aerometric, or economic information and models currently available.
Kriging is a spatial interpolation procedure that has been used to generate 0, concentration data for rural areas in which
no monitoring sites have been established. See Heck et al. (1983b).
-------
aerometric, or economic information and models currently available in the
literature. The estimates can then be ranked relative to the strength of
these data and assumptions. Of the eight regional studies reviewed, most have
adequate economic models, but only four are judged adequate across all input
categories. Further, most regional studies abstract from the interdependences
that exist between regions, which limits their utility in evaluating secondary
national ambient air quality standards (SNAAQS).
National-level studies can overcome this limitation of regional analyses
by accounting for economic linkages between groups and regions. A proper
accounting for these linkages, however, requires additional data and more
complex models, and frequently poses more difficult analytical problems.
Thus, detailed national assessments tend to be more costly to perform. As a
result, there are fewer assessments of pollution effects at the national than
at the regional level. Six national-level assessments performed since the
last criteria document was published in 1978 are reported in Table 6-40. Of
these, two used the simple "price times quantity" approach to quantify dollar
effects. Four used more defensible economic approaches. As with Table 6-39,
an evaluation of the adequacy of critical plant science, aerometric, and
economic data is presented, along with the estimates of benefits or damages.
As is evident from the evaluation, most of the national studies reviewed
here suffer from either plant science and aerometric data problems, incomplete
economic models, or both. As a result of these limitations, decision-makers
should be cautious in using these estimates to evaluate the efficiency of
alternative SNAAQS. Two of the studies, however, are judged to be much more
adequate in terms of the three critical areas of data inputs. Together, they
provide reasonably comprehensive estimates of the economic consequences of
changes in ambient air 0, levels on agriculture.
In the first of these studies, Kopp et al. (1984) measured the national
economic effects of changes in ambient air 0- levels on the production of
corn, soybeans, cotton, wheat, and peanuts. In addition to accounting for
price effects on producers and consumers, the assessment methodology used is
notable in that it placed emphasis on developing producer-level responses to
0--induced yield changes (from NCLAN data) in 200 production regions. The
results of the Kopp et al. (1984) study indicated that a reduction in 0, from
1978 regional ambient levels to a seasonal 7-hr average of approximately 0.04
ppm would result in a $1.2 billion net benefit in 1978 dollars. Conversely,
6-252
-------
TABLE 6-40. SUMMARY OF ESTIMATES OF NATIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study
Crops
Annual benefits
of control,
$ billion
Evaluation of critical data and assumptions
Plant response data
Aeroraetric data
Economic model data
Additional comments
Ryan et al.
(1981)
16 crops: alfalfa,
beets, broccoli,
cabbage, corn
(sweet and field),
hay, lima beans,
oats, potatoes,
sorghum, soybeans,
spinach, tobacco,
tomatoes, and
wheat.
$1.747 (in 1980
dollars).
CTl
ro Shriner et al. 4 crops: corn,
ŁJ (1982) soybeans, wheat,
and peanuts.
Multiple cultivars
of all crops but
peanuts.
Inadequate; yield-response
information derived from
a synthesis of 5 yield
studies in the literature
prior to 1980. Synthe-
sized response functions
estimated for both chronic
and acute exposures
for six crops. For
the remaining 10 crops
surrogates are used.
Yield changes are based
on reductions in 03 to
meet 1980 Federal stan-
dard of 0.12 ppm in non-
compliance counties.
Inadequate; dose
measured in sev-
eral ways to
correspond to
underlying
response function.
03 data derived
from National
Aerometric
Data Bank and
from Lawrence
Berkeley
Laboratory, for
period 1974-1976.
Inadequate; naive econo-
mic model. Monetary
impact calculated by
multiplying changes in
county production by
crop price in 1980.
Measures impact on
producers only.
Dollar estimate is for the 531
counties exceeding the
Federal standard of 0.12 ppm.
This study is essentially an
updated version of Benedict
et al. (1971) reported in 1978
criteria document.
$3.0 (in 1978
dollars).
Adequate; analysis uses
NCLAN response data for
1980. Functions esti-
mated in linear form.
Yield changes reflect
difference between 1978
ambient 03 levels of
each county and assumed
background of 0.025 ppm
concentration.
Unknown; exposure
may be measured as
highest 7-hr.
average, rather
than 7-hr NCLAN
average. Rural
ambient concen-
trations for 1978 ,
estimated by Kriging
procedure applied
to SAROAD data.
Inadequate; same as Ryan
et al. (1981) except
uses 1978 crop prices.
Dollar estimates are for all
counties producing the four
crops. As with Ryan et al.
(1981), estimates are for
for producer level effects
only.
Adams and 3 crops: corn,
Crocker (1984) soybeans, and
cotton. Two corn
cultivars, three
soybean, two
cotton.
$2.2 (in 1980
dollars).
Adequate; analysis uses
NCLAN 03-yield data for
1980 and 1981. Functions
estimated in linear form.
Yield changes measured
between 1980 ambient
levels and an assumed 03
concentration of 0.04 ppm
across all production
regions.
Adequate; 1980
ambient 03 levels
estimated by
Kriging of SAROAD
monitoring sites,
translated into a
seasonal 7-hr
average.
Adequate on demand side;
inadequate on modeling
producer behavior; eco-
nomic model consists of
crop demand and supply
curves. Corresponding
price and quantity
adjustments result in
changes in economic
surplus. No producer
level responses
modeled; only measures
aggregate effects.
Economic estimate measured in
terms of changes in consumer
and producer surpluses associated
with the change in 03.
-------
TABLE 6-40 (cont'd). SUMMARY OF ESTIMATES OF NATIONAL ECONOMIC CONSEQUENCES OF OZONE POLLUTION
Study
Crops
Annual benefits
of control,
$ billion
Evaluation of critical data and assumptions
Plant response data
Aerometric data
Economic model data
Additional comments
Adams et al.
(1984a)
4 crops: corn,
soybeans, wheat,
and cotton. Two
cultivars for corn
and cotton, three
for soybeans and
and wheat.
$2.4 (in 1980
dollars).
Adequate; analysis uses
NCLAN 03-yield data for
1980 through 1982. Yield
changes measured between
1980 ambient levels and
25% reduction in
03 across all regions.
Functions estimated in
both linear and quadratic
form.
Adequate; same
as Adams and
Crocker (1984).
Inadequate producer
model; same as Adams
and Crocker (1984),
except that analysis
examines range of
economic estimates
reflecting variability
in yield predictions
resulting from sample
size and functional
form.
Same as Adams and Crocker (1984).
Linear functions result in higher
yield losses and hence higher
economic loss estimates.
Reported estimate ($2.4 billion)
is for quadratic response
function.
Kopp et al.
(1984)
cr>
i
r>o
en
5 crops: corn,
soybeans, wheat,
cotton, and
peanuts. Multiple
cultivars of each
crop except peanuts.
$1.2 (in 1978
dollars).
Adequate; analysis uses
NCLAN 03 yield response
data for 1980 through
1982. Yield losses (for
estimates reported here)
measured as the differ-
ence between ambient 1978
03 and a level assumed to
represent compliance with
an 0.08 ppm standard.
Adequate; same as
Adams and Crocker
(1984) and Adams
et al. (1984b)
but for 1978
growing season.
Adequate; economic model
consists of producer-
level models, by crop,
for numerous production
regions. Predicted
yield changes are used
to generate supply
shifts for each region/
crop combined with crop
demand relationships
to estimate producer
and consumer surpluses.
In addition to measuring the
change in economic surplus for
various assumed 03 levels, the
analysis also includes an exam-
ination of the sensitivity of
the estimates to the nature of
the demand relationships used
in the model.
Adams et al.
(1984b)
6 crops: barley,
corn, soybeans,
cotton, wheat, and
sorghum. Multiple
cultivars used for
each crop except
barley and grain
sorghum; two for
cotton, three for
wheat, two for corn,
and nine for soybean.
$1.7 (in 1980
dollars).
Adequate; analysis uses
NCLAN 03 yield response
data for 1980 through
1983. Yield changes
reflect changes from
1980 ambient 03 of 10
and 40% reduction and
a 25% increase for each
response.
Adequate; same as
above but for 1980
and 1976 through
1980 periods.
Adequate; economic model
consists of two compo-
nents: a series of farm-
level models for each of
55 production regions
and a national model
of crop use and demand.
Yield changes are used
to generate regional
supply shifts used in
national model.
Consumer surplus estimated
for both domestic and foreign
markets; producer surplus
nationally and by region. The
analysis includes a range of
economic estimates reflecting
changes in response and 03
data and assumptions.
Adequacy as defined here does not mean that the estimates are free of error; rather, it implies that the estimates are based on'the most defensible
biologic, aerometric, or economic information and models currently available.
Kriging is a spatial interpolation procedure that has been used to generate 03 concentration data for rural areas in which no monitoring sites have
been established. See Heck et al. (1983b).
-------
an increase in (L to an assumed ambient concentration of 0.08 ppm (seasonal
7-hr average) across all regions produced a net loss of approximately $3.0
billion.
The second study, by Adams et al. (1984b), is a component of the NCLAN
program. The results were derived from an economic model of the U.S. agricul-
tural sector that includes individual farm models for 55 production regions
integrated with national supply-and-demand relationships for a range of crop
and livestock activities. Using NCLAN data, the analysis examined yield
changes for six major crops (corn, soybeans, wheat, cotton, grain, sorghum,
and barley) that together account for over 75 percent of U.S. crop acreage.
The estimated annual benefits (in 1980 dollars) from 0- adjustments are sub-
stantial, but make up a relatively small percentage of total agricultural
output (about 4 percent). Specifically, in this analysis, a 25 percent reduc-
tion in ozone from 1980 ambient levels resulted in benefits of $1.7 billion.
A 25 percent increase in ozone resulted in an annual loss (negative benefit)
of $2.363 billion. When adjusted for differences in years and crop coverages,
these estimates are quite close to the Kopp et al. (1984) benefit estimates.
While the estimates from both Kopp et al. (1984) and Adams et al. (1984b)
were derived from conceptually sound economic models and from the most defen-
sible plant science and aerometric data currently available, there are several
sources of uncertainty. These include the issue of exposure dynamics (7-hr
per day exposures from the NCLAN experiments versus longer exposure periods,
such as 12-hr exposures), and the lack of environmental interactions, particu-
larly 0.,-moisture stress interactions, in many of the response experiments.
Also, the 0, data in both studies are based on a limited set of the monitoring
sites in the SAROAD system of EPA, mainly sites in urban and suburban areas.
While the spatial interpolation process used for obtaining 0- concentration
data (Kriging) results in a fairly close correspondence between predicted and
actual 0-, levels at selected validation points, validation requires more
monitoring sites in rural areas. The economic models, with their large number
of variables, and parameters, and the underlying data used to derive these
values, contain potential sources of uncertainty, including the effects on bene-
fits estimates of market-distorting factors such as the Federal farm programs.
The inclusion of these possible improvements in future assessments is not
likely, however, with the possible exception of market-distorting factors, to
alter greatly the range of agricultural benefits provided in the Kopp et al.
(1984) and Adams et al. (1984b) studies, for several reasons. First, the
6-255
-------
current studies cover about 75 to 80 percent of U.S. agricultural crops (by
value). For inclusion of the other 20 percent to change the estimates signifi-
cantly would require that their sensitivities to 0- be much greater than for
the crops included to date. Second, model sensitivity analyses from existing
studies indicate that changes in key plant science parameters must be substan-
tial to translate into major changes in economic estimates. From experience
to date it seems unlikely that use of different dose measures or interaction
effects would result in changes of the magnitude already addressed in some of
the sensitivity analyses. Third, even if there are such changes, there are
likely to be countervailing responses; e.g., longer exposure periods may
predict greater yield losses but 0.,-water stress tends to dampen or reduce the
yield estimates. Finally, it should be noted that potential improvements in
economic estimates are policy-relevant only to the extent that they alter the
relationship between total benefits and total costs of that policy. Uncertain-
ties in other effects categories are probably greater.
In conclusion, the recent economic estimates of benefits to agriculture
of 0.3 control, particularly those estimates by Kopp et al. (1984) and Adams et
al. (1984b), meet the general criteria discussed in Section 6.5 and hence
provide the most defensible evidence given in the literature to date of the
general magnitude of such effects. Relative to estimates given in the 1978
criteria document (U.S. Environmental Protection Agency, 1978) and economic
information on most other 0- effects categories (non-agricultural), these two
studies, in combination with the underlying NCLAN data on yield effects,
provide the most comprehensive economic information to date on which to base
judgments regarding the economic efficiency of alternative SNAAQS. As noted
above, there are still gaps in plant science and aerometric data and a strong
need for meteorological modeling of 0- formation and transport processes for
use in formulating rural 0- scenarios. With regard to the economic data and
models used, the impact of factors that upset free-market equilibria needs
further analysis. Additionally, it must be emphasized that none of the studies
has accounted for the compliance costs of effecting changes in 0- concentra-
tions in ambient air. For a cost-benefit analysis to be complete, the annual-
ized estimated benefits to agriculture that would result from 0, control would
have to be combined with benefits accruing to other sectors and then compared
with the overall annualized compliance costs.
6-256
-------
6.8.10 Effects of Peroxyacetyl Nitrate on Vegetation
Peroxyacetyl nitrate (PAN) is a highly phytotoxic air pollutant that is
produced by photochemical reactions similar to those that produce 0-,, Both 0-
and PAN can coexist in the photochemical oxidant complex in ambient air. The
effects of PAN were a concern in southern California for almost 20 years
before the phytotoxicity of 0., under ambient conditions was identified. The
symptoms of photochemical oxidant injury that were originally described (prior
to 1960) were subsequently shown to be identical with the symptoms produced by
PAN. Following the identification of PAN as a phytotoxic air pollutant, PAN
injury (foliar symptoms) has been observed throughout California and in several
other states and foreign countries.
6.8.10.1 Factors Affecting Plant Response to PAN. Herbaceous plants are
sensitive to PAN and cultivar differences in sensitivity have been observed in
field and controlled studies. Trees and other woody species, however, are
apparently resistant to visible foliar injury from PAN (Taylor, 1969; Davis,
1975, 1977).
Taylor et al. (1961) demonstrated that there is an absolute requirement
for light before, during, and after exposure or visible injury from PAN will
not develop. Field observations showed that crops growing under moisture
stress developed little or no injury during photochemical oxidant episodes
while, adjacent to them, recently irrigated crops were severly injured (Taylor,
1974).
Only a few studies have investigated the effects of PAN and 0- mixtures
on plants. When plants were exposed to both gases at their respective injury
thresholds, no interaction between the gases was found (Tonneijck, 1984). At
higher concentrations, the effects were less than additive. Studies with
petunia confirmed that 0- tended to reduce PAN injury (Nouchi et al., 1984).
6.8.10.2 Limiting Values of Plant Response. The limiting-value method has
been used to estimate the lowest PAN concentration and exposure duration
reported to cause visible injury on various plant species (Jacobson, 1977).
The analysis yielded the following range of concentrations and exposure dura-
tions likely to induce foliar injury: (1) 200 ppb for 0.5 hr; (2) 100 ppb for
1.0 hr; and (3) 35 ppb for 4.0 hr.
Other studies, however, suggest that these values need to be lowered by
30 to 40 percent to reduce the likelihood of foliar injury (Tonneijck, 1984).
For example, foliar injury developed on petunia plants exposed at 5 ppb PAN
for 7 hr (Fukuda and Terakado, 1974). Under field conditions, injury symptoms
6-257
-------
may develop on sensitive species when PAN concentrations reach approximately
15 ppb for 4 hr (Taylor, 1969).
6.8.10.3 Effects of PAN on Plant Yield. Only a few limited studies have been
conducted to determine the effects of PAN on plant growth and yield. In
greenhouse studies, radish, oat, tomato, pinto bean, beet, and barley were
exposed to PAN concentrations of up to 40 ppb for 4 hr/day, twice/wk, from
germination to crop maturity (Taylor et al., 1983). No significant effects on
yield were detected. This is supportive of field observations, in which
foliar injury from ambient PAN exposures was found but no evidence was seen of
reduced yield in these crops. In contrast, lettuce and Swiss chard exposed to
PAN concentrations of up to 40 ppb for 4 hr/day, twice/wk, from germination to
crop maturity showed yield losses up to 13 percent (lettuce) and 23 percent
(Swiss chard) without visible foliar injury symptoms (Taylor et al., 1983).
The results indicate that PAN at concentrations below the foliar-injury
threshold can cause significant yield losses in sensitive cultivars of leafy
vegetable crops. In addition, photochemical oxidant events have caused foliar
injury on leafy vegetables (Middleton et al., 1950) for which the foliage is
the marketable portion. After severe PAN damage, entire crops may be unmarket-
able or else extensive hand work may be required to remove the injured leaves
before the crop may be marketed.
A comparison of PAN concentrations likely to cause either visible injury
or reduced yield with measured ambient concentrations (see Chapter 5) indicates
that it is unlikely that ambient PAN will impair the intended use of plants in
the United States except in some areas of California and possibly in a few
other localized areas.
6.9 REFERENCES
Adams, R. M. ; Crocker, T. D. (1982) Dose-response information and environmen-
tal damage assessments: an economic perspective. J. Air Pollut. Control
Assoc. 32: 1062-1067.
Adams, R. M. ; Crocker, T. D. (1984) Economically relevant response estimation
and the value of information: acid deposition. In: Crocker, T. D. , ed.
Economic perspectives on acid deposition control. Boston, MA: Butterworth
Publishers; pp. 35-64. (Teasley, J. I., ed. Acid precipitation series
-v. 8.)
6-258
-------
Adams, R. M.; McCarl, B. A. (1985) Assessing the benefits of alternative
oxidant standards on agriculture: the role of response information. J.
Environ. Econ. Manage. 12: 264-276.
Adams, R. M. ; Hamilton, S. A.; McCarl, B. A. (1985) An assessment of the
economic effects of ozone on U.S. agriculture. J. Air Pollut. Control
Assoc. 35: 938-943.
Adams, R. M. ; Crocker, R. M. ; Katz, R. W. (1984a) The adequacy of natural
science information in economic assessments of pollution control: a
Bayesian methodology. Rev. Econ. Stat. 66: 568-575.
Adams, R. M. ; Hamilton, S. A.; McCarl, B. A. (1984b) The economic effects of
ozone on agriculture. Corvallis, OR: U.S. Environmental Protection
Agency; EPA report no. EPA-600/3-84-090. Available from: NTIS,
Springfield, VA; PB85-168441/XAB.
Adams, R. M.; Crocker, T. D.; Thanavibulchai, N. (1982) An economic assessment
of air pollution damages to selected annual crops in southern California.
J. Environ. Econ. Manage. 9: 42-58.
Adedipe, N. 0.; Ormrod, D. P. (1974) Ozone induced growth suppression in
radish plants in relation to pre- and post-fumigation temperatures. Z.
Pflanzenphysiol. 71: 281-287.
Adedipe, N. 0.; Barrett, R. E.; Ormrod, D. P. (1972) Phytotoxicity and growth
responses of ornamental bedding plants to ozone and sulfur dioxide. J.
Am. Soc. Hortic. Sci. 97: 341-345.
Alstad, D. N.; Edmunds, G. F., Jr.; Weinstein, L. H. (1982) Effects of air
pollutants on insect populations. Annu. Rev. Entomol. 27: 369-384.
Amiro, B. D. ; Gillespie, T. J. ; Thurtell, G. W. (1984) Injury response of
Phaseolus vulgaris to ozone flux density. Atmos. Environ. 18: 1207-1215.
Anderson, V. L. ; McLean, R. A. (1974) Design of experiments: a realistic
approach. New York, NY: Marcel Dekker, Inc.
Ashmore, M. R. (1984) Effects of ozone on vegetation in the United Kingdom.
Proceedings of the OECD workshop on ozone; February 29-March 2, 1984,
Gbteburg, Sweden. Gb'teburg, Sweden: Swedish Environmental Research
Institute.
Ashmore, M. R.; Bell, J. N. B.; Reily, C. L. (1978) A survey of ozone levels
in the British Isles using indicator plants. Nature (London) 276:
813-815.
Ashmore, M. R.; Bell, J. N. B.; Reily, C. L. (1980) The distribution of phyto-
toxic ozone in the British Isles. Environ. Pollut. Ser. B: 195-216.
Athanassious, R. (1980) Ozone effects on radish (Raphanus sativus L. cv.
cherry belle): foliar sensitivity as related to metabolite levels and
cell architecture. Z. Pflanzenphysiol. 97: 183-187.
6-259
-------
Aycock, M. K. , Jr. (1972) Combining estimates for weather fleck in Nicotiana
tabacum L. Crop Sci. 12: 672-674.
Barnes, R. L. (1972a) Effects of chronic exposure to ozone on photosynthesis
and respiration of pines. Environ. Pollut. 3: 133-138.
Barnes, R. L. (1972b) Effects of chronic exposure to ozone on soluble sugar
and ascorbic acid contents of pine seedlings. Can. J. Bot. 50: 215-219.
Beckerson, D. W. ; Hofstra, G. (1979) Response of leaf diffusive resistance of
radish, cucumber, and soybean to 03 and S02 singly or in combination.
Atmos. Environ. 13: 1263-1268.
Bell, J. N. B. ; Cox, R. A. (1975) Atmospheric ozone and plant damage in the
United Kingdom. Environ. Pollut. 8: 163-170.
Benedict, H. M.; Miller, C. J.; Olson, R. E. (1971) Economic impact of air
pollutants on plants in the United States. Research Triangle Park, NC:
U.S. Environmental Protection Agency; report no. APTD-0953. Available
from: NTIS, Springfield, VA; PB-209265.
Benedict, H. M. ; Miller, C. J. ; Smith, J. S. (1973) Assessment of economic
impact of air pollutants on vegetation in the United States: 1969 and
1971. Menlo Park, CA: Stanford Research Institute. Available from: NITS,
Springfield, VA; PB-224818.
Bennett, J. H. (1979) Foliar exchange of gases. In: Heck, W. W. ; Krupa, S. V.;
Linzon, S. N. , eds. Methodology for the assessment of air pollution
effects on vegetation: a handbook from a specialty conference; April
1978; Minneapolis, MN. Pittsburgh, PA: Air Pollution Control Association;
ch. 10.
Bennett, J. H.; Hill, A. C. (1974) Acute inhibition of apparent photosynthesis.
In: Dugger, J., ed. Air pollution effects on plant growth. Washington,
DC: American Chemical Society. (ACS symposium series 3.)
Bennett, J. P.; Oshima, R. J. (1976) Carrot injury and yield response to
ozone. J. Am. Soc. Hortic. Sci. 101: 638-639.
Bennett, J. P.; Runneckles, V. C. (1977) Effects of low levels of ozone on
growth of crimson clover and annual ryegrass. Crop Sci. 17: 443-444.
Bennett, J. P.; Oshima, R. J.; Lippert, L. F. (1979) Effects of ozone on
injury and dry matter partitioning in pepper plants. Environ. Exp. Bot.
19: 33-39.
Benoit, L. F. ; Skelly, J. M.; Moore, L. D. ; Dochinger, L. S. (1982) Radial
growth reductions in Pinus strobus L. correlated with foliar ozone sensi-
tivity as an indicator of ozone-induced losses in eastern forests. Can.
J. For. Res. 12: 673-678.
6-260
-------
Benson, E. J. ; Krupa, S.; Teng, P. S.; Welsch, P. E. (1982) Economic assess-
ment of air pollution damages to agricultural and silvicultural crops in
Minnesota. Final report to Minnesota Pollution Control Agency. St. Paul,
MN: University of Minnesota.
Berry, C. R. (1971) Relative sensitivity of red, jack, and white pine seedlings
to ozone and sulfur dioxide. Phytopathology 61: 231-232.
Bingham, G. E.; Coyne, P. I. (1977) A portable, temperature-controlled, steady-
state porometer for field measurements of transpiration and photosynthesis.
Photosynthetica 11: 148-160.
Bisessar, S. (1982) Effect of ozone, antioxidant protection, and early blight
on potato in the field. J. Am. Soc. Hortic. Sci. 107: 597-599.
Bisessar, S. ; Palmer, K. T. (1984) Ozone, antioxidant spray and Meloidogyne
hapla effects on tobacco. Atmos. Environ. 18: 1025-1027.
Bisessar, S. ; Temple, P. J. (1977) Reduced ozone injury on virus-infected
tobacco in the field. Plant Dis. Rep. 61: 961-963.
Black, V. J.; Ormrod, D. P.; Unsworth, M. H. (1982) Effects of low concentra-
tion of ozone, singly, and in combination with sulphur dioxide on net
photosynthesis rates of Vicia faba L. J. Exp. Bot. 33: 1302-1311.
Blum, U. T.; Heck, W. W. (1980) Effects of acute ozone exposures on snap bean
at various stages of its life cycle. Environ. Exp. Bot. 20: 73-85.
Blum, U. T.; Tingey, D. T. (1977) A study of the potential ways in which ozone
could reduce root growth and nodulation of soybean. Atmos. Environ.
11: 737-739.
Blum, U. T.; Smith, G. R. ; Fites, R. C. (1982) Effects of multiple 03 expo-
sures on carbohydrate and mineral contents of Ladino clover. Environ.
Exp. Bot. 22: 143-154.
Botkin, D. B. ; Smith, W. H. ; Carlson, R. W. ; Smith, T. L. (1972) Effects of
ozone on white pine saplings: variation in inhibition and recovery of net
photosynthesis. Environ. Pollut. 3: 273-289.
Box, G. E. P.; Hunger, W. G.; Hunter, J. S. (1978) Statistics for experi-
menters. New York, NY: John Wiley and Sons, Inc.
Boyer, J. S. (1982) Plant productivity and environment. Science (Washington,
DC) 218: 443-448.
Brandt, C. S. (1962) Effects of air pollution on plants. In: Stern, A. C., ed.
Air pollution: v. I. New York, NY: Academic Press; pp. 255-281.
Brasher, E. P.; Fieldhouse, D. J. ; Sasser, M. (1973) Ozone injury in potato
variety trials. Plant Dis. Rep. 57: 542-544.
Brennan, E. (1975) On exclusion as the mechanism of ozone resistance of virus-
infected plants. Phytopathology 65: 1054-1055.
6-261
-------
Brennan, E.; Halisky, P. M. (1970) Response of turfgrass cultivars to ozone
and sulfur dioxide in the atmosphere. Phytopathology 60: 1544-1546.
Brennan, E. ; Leone, I. A. (1969) Suppression of ozone toxicity symptoms in
virus-infected tobacco. Phytopathology 59: 263-264.
Brennan, E.; Leone, I. A. (1970) The response of English holly selections to
ozone and sulfur dioxide. Holly Lett. 37: 6-7.
Brennan, E.; Leone, I. A. (1972) Chrysanthemum response to sulfur dioxide and
ozone. Plant Dis. Rep. 56: 85-87.
Brennan, E.; Leone, I. A.; Daines, R. H. (1964) The importance of variety in
ozone plant damage. Plant Dis. Rep. 48: 923-924.
Brennan, E.; Leone, I. A.; Halisky, P. M. (1969) Response of forage legumes to
ozone fumigations. Phytopathology 59: 1458-1459.
Brewer, R.F.; Ashcroft, R. (1982) The effects of ozone and sulfur dioxide on
alfalfa yields and hay quality. Final report under Agreement No. Al-038-33
to the California Air Resources Board, Sacramento, CA.
Brewer, R. F.; Ferry, G. (1974) Effects of air pollution on cotton in the
San Joaquin Valley. Calif. Agric. 28: 6-7.
Butler, L. K.; Tibbitts, T. W. (1979a) Variation in ozone sensitivity and symptom
expression among cultivars of Phaseolus vulgaris L. J. Amer. Hort. Sci.
104: 208-210.
Butler, L. K. ; Tibbitts, T. W. (1979b) Stomatal mechanisms determining genetic
resistance to ozone in Phaseolus vulgaris L. J. Am. Soc. Hortic. Sci.
104: 213-216.
Butler, L. K. ; Tibbitts, T. W.; Bliss, F. A. (1979) Inheritance of resistance
to ozone in Phaseolus vulgaris L. J. Am. Soc. Hortic. Sci. 104: 211-213.
Cameron, J. W. (1975) Inheritance in sweet corn for resistance to acute ozone
injury. J. Am. Soc. Hortic. Sci. 100: 577-579.
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.
Carnahan, J. E.; Jenner, E. L.; Wat, E. K. W. (1978) Prevention of ozone
injury to plants by a new protectant chemical. Phytopathology 68:
1225-1229.
Carney, A. W. ; Stephenson, G. R.; Ormrod, D. P.; Ashton, G. C. (1973)
Ozone-herbicide interactions in crop plants. Weed Sci. 21: 508-511.
Cathey, H. M. ; Heggestad, H. E. (1972) Reduction of ozone damage to Petunia
hybrida Vilm. by use of growth-regulating chemicals and tolerance
cultivars. J. Am. Soc. Hortic. Sci. 97: 695-700.
6-262
-------
Cathey, H. M.; Heggestad, H. E. (1973) Effects of growth retardants and fumi-
gations with ozone and sulfur dioxide on growth and flowering of
Euphorbia pulchurima Willd. J. Am. Soc. Hortic. Sci. 98: 3-7.
Cathey, H. M.; Heggestad, H. E. (1982a) Ozone and sulfur dioxide sensitivity
of petunia: modification by ethylenediurea. J. Am. Soc. Hortic. Sci.
107: 1028-1035.
Cathey, H. M.; Heggestad, H. E. (1982b) Ozone sensitivity of herbaceous plants:
modification by ethylenediurea. J. Am. Soc. Hortic. Sci. 107: 1035-1042.
Cathey, H. M.; Heggestad, H. E. (1982c) Ozone sensitivity of woody plants:
modification by ethylenediurea. J. Am. Soc. Hortic. Sci. 107: 1042-1045.
Chattin, B.; McCarl, B. A.; Baumes, H., Jr. (1983) Users guide and documentation
for a partial equilibrium sector model of U.S. agriculture. West Lafayette,
In: Purdue University; Agricultural Experiment Station Bulletin No. 313.
Christensen, E. ; Giese, A. C. (1954) Changes in absorption spectra of nucleic
acids and their derivatives following exposure to ozone and ultraviolet
radiations. Arch. Biochem. Biophys. 51: 208-216.
Clarke, B. B.; Brennan, E. (1980) Evidence for a cadmium and ozone interaction
on Populus tremuloides. J. Arboriculture 6: 130-134.
Clarke, B.; Brennan, E. (1981) The impact of ambient oxidant on foliar symptoms
and tuber yield and quality of field-grown potato cultivars in New Jersey.
Phytopathology 71: 867. [Abstract]
Clarke, B. B.; Henninger, M. R.; Brennan, E. (1983) An assessment of potato
losses caused by oxidant air pollution in New Jersey. Phytopathology
73: 104-108.
Clayberg, C. D. (1971) Screening tomatoes for ozone resistance. HortScience
6: 396-397.
Cobb, F. W. , Jr.; Stark, R. W. (1970) Decline and mortality of smog-injured
ponderosa pine. J. For. 68: 147-149.
Cochran, W. G. ; Cox, G. M. (1957) Experimental designs. 2nd Ed. New York, NY:
John Wiley and Sons, Inc.
Costonis, A. C.; Sinclair, W. A. (1972) Susceptibility of healthy and ozone-
injured needles of Pinus strobus to invasion by Lophodermium pinastri and
Aureobasidium pullulans. Eur. J. For. Pathol. 2: 65-73.
Coulson, C. L. ; Heath, R. L. (1974) Inhibition of the photosynthetic capacity
of isolated chloroplasts by ozone. Plant Physiol. 53:• 32-38.
Coulson, C.; Heath, R. L. (1975) The interaction of peroxyacetyl nitrate (PAN)
with the electron flow of isolated chloroplasts. Atmos. Environ. 9:
231-238.
6-263
-------
Coyne, P. E.; Bingham, G. E. (1978) Photosynthesis and stomatal light responses
in snap beans exposed to hydrogen sulfide in ozone. J. Air Pollut. Control
Assoc. 28: 1119-1123.
Coyne, P. E. ; Bingham, G. E. (1981) Comparative ozone dose response of gas
exchange in a ponderosa pine stand exposed to long-term fumigations. J.
Air Pollut. Control Assoc. 31: 38-41.
Craker, L. E.; Feder, W. A. (1972) Development of the inflorescence in petunia,
geranium, and poinsettia under ozone stress. HortScience 7: 59-60.
Crocker, T. D. (1982) Pollution damage to managed ecosystems: economic assess-
ments. In: Jacobson, J. S.; Millen, A. A., eds. Effects of air pollution
on farm commodities: proceedings of the symposium; February; Washington,
DC. Arlington, VA: Izaak Walton League of America; pp. 103-124.
Czuba, M. ; Ormrod, D. P. (1974) Effects of-cadmium and zinc on ozone-induced
phytotoxicity in cress and lettuce. Can. J. Bot. 52: 645-649.
Czuba, M.; Ormrod, D. P. (1981) Cadmium concentrations in cress shoots in
relation to cadmium-enhanced ozone phytotoxicity. Environ. Pollut.
25: 67-76.
Dairies, R. H. (1963) Effects of air pollution on crops and livestock. National
Conference on Air Pollution Proceedings, pp. 179-182. U.S. Department of
Health, Education, and Welfare, Public Health Service, Washington, DC.
Daines, R. H. ; Leone, I. A.; Brennan, E. (1960) Air pollution as it affects
agriculture in New Jersey. New Jersey Agricultural Experiment Station,
Rutgers. Bulletin 794-A.
Daines, R. H.; Leone, I. A.; Brennan, E. (1967) Air pollution and plant response
in the northeastern United States. In: Brady, N. C. , ed. Agriculture and
the quality of our environment. Am. Assoc. Adv. Sci. Publication 85: 11-31.
Darley, E. F.; Dugger, W. M.; Mudd, J. B.; Ordin, L.; Taylor, 0. C.; Stephens,
E. R. (1963) Plant damage by pollution derived from automobiles. Arch.
Environ. Health 6: 761-770.
Davis, D. D. (1975) Resistance of young ponderosa pine seedlings to acute
doses of PAN. Plant Dis. Rep. 59: 183-184.
Davis, D. D. (1977) Response of ponderosa pine primary needles to separate and
simultaneous ozone and PAN exposures. Plant Dis. Rep. 61: 640-644.
Davis, D. D.; Coppolino, J. B. (1974) Relative ozone susceptibility of
selected woody ornamentals. HortScience 9: 537-539.
Davis, D. D. ; Coppolino, J. B. (1976) Ozone susceptibility of selected woody
shrubs and vines. Plant Dis. Rep. 60: 876-878.
Davis, D. D; Kress, L. (1974) The relative susceptibility of ten bean varieties
to ozone. Plant Dis. Rep. 58: 14-16.
6-264
-------
Davis, J. M. ; Rogers, H. H. (1980) Wind tunnel testing of open-top field
chambers for plant effects assessment. J. Air Pollut. Control Assoc.
30: 905-908.
Davis, D. D. ; Smith, S. H. (1975) Bean common mosaic virus reduces ozone
sensitivity of pinto bean. Environ. Pollut. 9: 97-101.
Davis, D. D.; Smith, S. H. (1976) Reduction of ozone sensitivity of pinto bean
by virus-induced local lesions. Plant Dis. Rep. 60: 31-34.
Davis, D. D. ; Wood, F. A. (1972) The relative susceptibility of eighteen
coniferous species to ozone. Phytopathology 62: 14-19.
Davis, D. D.; Umbach, D. M. ; Coppolino, J. B. (1981) Susceptibility of tree
and shrub species and response of black cherry foliage to ozone. Plant
Dis. 65: 904-907.
Dean, C. E. (1963) Weather fleck on Florida shadegrown tobacco. Job. Sci.
7: 41-45.
Dean, C. E. (1972) Stomate density and size as related to ozone-induced weather
fleck in tobacco. Crop Sci. 12: 547-548.
De Bauer, L. I. (1972) Uso de plantas indicadoras en la determinacion de
aeropolutos en la Ciudad de Mexico [Use of indicative plants to determine
air pollution in Mexico City]. Phytopathology 62: 753.
de Cormis, L. ; Bonte, J. ; Tisne, A. (1975) Technique experimental permettant
I1etude de Tincidence sur la vegetation d'une pollution par le dioxyde
de soufre appliquee en permanence et a dose subnecrotique [Experimental
technique for studying damage to vegetation from chronic application of
sublethal doses of sulfur dioxide pollution]. Pollut. Atmos. 17: 103-107.
De Vos, N. E.; Hill, R. R., Jr.; Hepler, R. W. ; Pell, E. J.; Craig, R. (1980)
Inheritance of peroxyacetyl nitrate resistance in petunia. J. Am. Soc.
Hortic. Sci. 105: 157-160.
De Vos, N, E.; Hill, R. R., Jr.; Pell, E. J.; Cole, R. H. (1982) Quantitative
inheritance of ozone resistance in potato. Crop Sci. 22: 992-995.
De Vos, N. E.; Pell, E. J. ; Hill, R. R., Jr.; Cole, R. H. (1983) Laboratory
versus field response of potato genotypes to oxidant stress. Plant Dis.
67: 173-176.
Dijak, M.; Ormrod, D. P. (1982) Some physiological and anatomical characteris-
tics associated with differential ozone sensitivity among pea cultivars.
Environ. Exp. Bot. 22: 395-402.
Dixon, B. L.; Garcia, P.; Mjelde, J. W.; Adams, R. M. (1983) Estimation of the
economic cost of ozone on Illinois cash grain farms. Urbana, IL: Illinois
Agricultural Experiment Station research report.
Dochinger, L. S. ; Townsend, A. M. (1979) Effects of roadside deicer salts and
ozone on red maple progenies. Environ. Pollut. 19: 229-237.
6-265
-------
Dochinger, L. S.; Bender, F. W.; Fox, F. L.; Heck, W. W. (1970) Chlorotic
dwarf of eastern white pine caused by an ozone and sulphur dioxide inter-
action. Nature (London) 225: 476.
Dochinger, L. S.; Seliskar, C. E. (1970) Air pollution and the chlorotic dwarf
disease of eastern white pine. Forest Sci. 16: 46-55.
Draper, N. R. ; Smith, H. (1966) Applied regression analysis. New York, NY:
John Wiley and Sons, Inc.
Drummond, D. D. (1972) The effect of peroxyacetyl nitrate on petunia (Petunia
hybrida Vilm) [Ph.D. thesis]. University Park, PA: Pennsylvania State
University, Department of Plant Pathology. Available from: University
Microfilms, Ann Arbor, MI; publication no. 73-07428.
Duchelle, S. F. ; Skelly, J. M. (1981) Response of common milkweed to oxidant
air pollution in the Shenandoah National Park in Virginia. Plant Dis.
65: 661-662.
Duchelle, S. F.; Skelly, J. M.; Sharick, T. L.; Chevone, B. I.; Yang, Y.-S.;
Nellessen, J. E. (1983) Effects of ozone on the productivity of natural
vegetation in a high meadow of the Shenandoah National Park of Virginia.
J. Environ. Manage. 17: 299-308.
Dugger, W. M., Jr.; Taylor, 0. C. ; Klein, W. H. ; Shropshire, W. , Jr. (1963)
Action spectrum of peroxyacetyl nitrate damage to bean plants. Nature
(London) 198: 75-76.
Dugger, W. M. , Jr.; Mudd, J. B. ; Koukol, J. (1965) Effect of PAN on certain
photosynthetic reactions. Arch. Environ. Health. 10: 195-200.
Dunning, J. A.; Heck, W. W. (1973) Response of pinto bean and tobacco to ozone
as conditioned by light intensity and/or humidity. Environ. Sci. Technol.
7: 824-826.
Dunning, J. A.; Heck, W. W. (1977) Response of bean and tobacco to ozone:
effect of light intensity, temperature and relative humidity. J. Air
Pollut. Control Assoc. 27: 882-886.
Dunning, J. A.; Heck, W. W.; Tingey, D. T. (1974) Foliar sensitivity of pinto
bean and soybean to ozone as affected by temperature, potassium nutrition
and ozone dose. Water Air Soil Pollut. 3: 305-313.
Edinger, J. G.; McCutchan, M. H.; Miller, P. R.; Ryan, B. C.; Schroeder, M. J. ;
Behar, J. V. (1972) Penetration and duration of oxidant air pollution in the
South Coast Air Basin of California. J. Air Pollut. Control Assoc. 22: 882-
886.
Elden, T. E.; Howe11, R. K.; Webb, R. E. (1978) Influence of ozone on pea
aphid resistance in selected alfalfa strains. J. Econ. Entomol.
71: 283-286.
Elkiey, T.; Ormrod, D. P. (1980) Response of turfgrass cultivars to ozone,
sulfur dioxide, nitrogen dioxide, or their mixture. J. Am. Soc. Hortic.
Sci. 105: 664-668.
6-266
-------
Elkiey, T.; Ormrod, D. P. (1981) Absorption of ozone, sulphur dioxide, and
nitrogen dioxide by petunia plants. Environ. Exp. Bot. 21: 63-70.
Engle, R. L. ; Gabelman, W. H. (1966) Inheritance and mechanism for resistance
to ozone damage in onion, Al 1 iurn cepa L. J. Am. Soc. Hortic. Sci.
89: 423-430.
Ensing, J. ; Hofstra, G. (1982) Impact of the air pollutant ozone on acetylene
reduction and shoot growth of red clover. Can. J. Plant Pathol. 4: 237-242.
Evans, L. S.; Ting, I. P. (1974) Ozone sensitivity of leaves: relationship to
leaf water content, gas transfer resistance, and anatomical characteristics.
Am. J. Bot. 61: 592-597.
Feder, W. A. (1968) Reduction in tobacco pollen germination and tube elongation,
induced by low levels of ozone. Science (Washington, DC) 160: 1122.
Feder, W. A. (1978) Plants as bioassay systems for monitoring atmospheric
pollutants. EHP Environ. Health Perspect. 27: 139-147.
Feder, W. A.; Campbell, F. J. (1968) Influence of low levels of ozone on
flowering of carnations. Phytopathology 58: 1038-1039.
Feder, W. A.; Manning, W. J. (1979) Living plants as indicators and monitors.
In: Heck, W. W. ; Krupa, S. V.; Linzon, S. N., eds. Methodology for the
assessment of air pollution effects on vegetation: a handbook from a
specialty conference; April 1978; Minneapolis, MN. Pittsburgh, PA: Air
Pollution Control Association; ch. 9.
Feder, W. A.; Fox, F. L.; Heck, W. W.; Campbell, F. J. (1969) Varietal responses
of petunia to several air pollutants. Plant Dis. Rep. 53: 506-510.
Flagler, R. B. ; Youngner, V. B. (1982a) Ozone and sulfur dioxide effects on
tall fescue: I. Growth and yield responses. J. Environ. Qual. 11: 386-389.
Flagler, R. B. ; Youngner, V. B. (1982b) Ozone and sulfur dioxide effects on
three tall fescue cultivars. J. Environ. Qual. 11: 413-416.
Floor, H. ; Posthumus, A. C. (1977) Biologische Erfassung von Ozon- und PAN-
Immissionen in den Niederlanden 1973, 1974, und 1975 [Biological signifi-
cance of ozone and PAN emissions in the Netherlands 1973, 1974, and 1975].
In: Ozon und Begleitsubstanzen im photochemischen Smog [Ozone and other
substances in photochemical smog]: VDI colloquium; 1976; Dusseldorf, West
Germany. Dusseldorf, West Germany: Verein deutscher Ingenieure (VDI) GmbH;
pp. 183-190. (VDI-Berichte no. 270.)
Fong, F. ; Heath, R. L. (1981) Lipid content in the primary leaf of bean
(Phaseolus vulgaris) after ozone fumigation. Z. Pflanzenphysiol. 104: 109-
115.
Foster, K. W.; Guerard, J. P.; Oshima, R. J.; Bishop, J. C.; Timm, H. (1983a)
Differential ozone susceptibility of centennial russet and white rose
potato as demonstrated by fumigation and antioxidant treatments. Am.
Potato J. 60: 127-139.
6-267
-------
Foster, K. W.; Timm, H.; Labanauskas, C. K. ; Oshima, R. J. (1983b) Effects of
ozone and sulfur dioxide on tuber yield and quality of potatoes. J.
Environ. Qual. 12: 75-80.
Freebairn, H. T. (1963) Uptake and movement of 1-C14 ascorbic acid in bean
leaves. Physiol. Plant. 16: 517-522.
Freebairn, H. T.; Taylor, 0. C. (1960) Prevention of plant damage from air-borne
oxidizing agents. Proc. Am. Soc. Hortic. Sci. 76: 693-699.
Freeman, A. M., III (1979) The benefits of environmental improvement. Baltimore,
MD: The John Hopkins University Press.
Fukuda, H.; Terakado, K. (1974) On the damage of plants due to peroxyacyl
nitrates (PAN). Tokyo Nogyo Shikenjo Sokuho: 35-36 [English abstract].
Furukawa, A.; Kadota, M. (1975) Effect of ozone on photosynthesis and respira-
tion in poplar leaves. Seibutsu Kankyo Chosetsu 13: 1-7.
Gale, J.; Hagan, R. M. (1966) Plant antitranspirents. Annu. Rev. Plant Physiol.
17: 269-282.
Gardner, W. S. (1973) Ozone injury to tobacco plants in South Dakota. Plant
Dis. Rep. 57: 106-110.
Gesalman, C. M. ; Davis, D. D. (1978) Ozone susceptibility for ten azalea
cultivars as related to stomata frequency or conductance. J. Am. Soc.
Hortic. Sci. 103: 489-491.
Georgopoulos, P. G.; Seinfeld, J. H. (1982) Statistical distribution of air
pollutant concentrations. Environ. Sci. Technol. 16: 401A-416A.
Gilbert, M. D.; Elfving, D. C.; Lisk, D. J. (1977) Protection of plants against
ozone injury using the antiozonant N-(l,3-dimethylbutyl)-N'-phenyl-p-phenyl-
enediamine. Bull. Environ. Contam. Toxicol. 18: 783-786.
Glater, R. B. ; Solberg, R. A.; Scott, F. M. (1962) A developmental study of
the leaves of Nicotiana glutinosa as related to their smog-sensitivity.
Am. J. Bot. 49: 954-970.
Goren, A. I.; Donagi, A. E. (1979) Use of Bel-W3 tobacco as indicator plant
for atmospheric ozone during different seasons in the coastal zone of
Israel. Int. J. Biometeorol. 23: 331-335.
Goren, A. I.; Donagi, A. E. (1980) Assessment of atmospheric ozone levels in
Israel through foliar injury to Bel-W3 tobacco plants. Oecologia 44: 418-
421.
Gross, R. E. ; Dugger, W. M. (1969) Responses of Chlamydomonas reinhardtii to
peroxyacetyl nitrate. Environ. Res. 2: 256-266.
Grosso, J. J., Menser, H. A.; Hodges, G. H.; McKinney, H. H. (1971)
Effects of air pollutants on Nicotiana cultivars and species used for
virus studies. Phytopathology 61: 945-950.
6-268
-------
Guderian, R. (1977) Discussion of the suitability of plant responses as a
basis for air pollution control measures. In: Air pollution: phytotoxi-
city of acidic gases and its significance in air pollution control. Berlin,
West Germany: Springer Verlag; pp. 75-97. (Ecological studies: analysis
and synthesis, v. 22).
Haas, J. H. (1970) Relation of crop maturity and physiology to air pollution
incited bronzing of Phaseolus vulgaris. Phytopathology 60: 407-410.
Hanson, G. P. (1972) Relative air pollution sensitivity of some Los Angeles
Arboretum plants. Lasca Leaves 22: 86-89.
Hanson, G. P.; Stewart, W. S. (1970) Photochemical oxidants: effect on starch
hydrolysis in leaves. Science (Washington, DC) 168: 1223-1224.
Hanson, G. P.; Thorne, L.; Addis, D. H. (1975) The ozone sensitivity of Petunia
hybrida Vilm. as related to physiological age. J. Am. Soc. Hortic. Sci.
100: 188-190.
Hanson, G. P.; Addis, D. H.; Thorne, L. (1976) Inheritance of photochemical
air pollution tolerance in petunias. Can. J. Genet. Cytol. 18: 579-592.
Harkov, R. ; Brennan, E. (1980) The influence of soil fertility and water
stress on the ozone response of hybrid poplar trees. Phytopathology
70: 991-994.
Harkov, R. S. ; Brennan, E. (1982) An ecophysiological analysis of woody and
herbaceous plants to oxidant injury. J. Environ. Manage. 15: 251-261.
Harkov, R.; Clarke, B. ; Brennan, E. (1979) Cadmium contamination may modify
response of tomato to atmospheric ozone. J. Air Pollut. Control Assoc.
29: 1247-1249.
Harris, M. J.; Heath, R. L. (1981) Ozone sensitivity in sweet corn (Zea mays L.)
plants: a possible relationship to water balance. Plant Physiol. 68: 885-
890.
Heagle, A. S. (1970) Effect of low-level ozone fumigations on crown rust of
oats. Phytopathology 60: 252-254.
Heagle, A. S. (1973) Interactions between air pollutants and plant parasites.
Phytopathology 11: 365-388.
Heagle, A. S. (1975) Response of three obligate parasites to ozone. Environ.
Pollut. 9: 91-95.
Heagle, A. S. (1977) Effect of ozone on parasitism of corn by Helminthosporium
maydis. Phytopathology 67: 616-618.
Heagle, A. S. (1979a) Effects of growth media, fertiliser rate and hour and
season of exposure on sensitivity of four soybean cultivars to ozone.
Environ. Pollut. 18: 313-322.
6-269
-------
Heagle, A. S. (1979b) Ranking of soybean cultivars for resistance to ozone
using different ozone doses and response measures. Environ. Pollut.
19: 1-10.
Heagle, A. S. (1982) Interactions between air pollutants and parasitic plant
diseases. In: Unsworth, M. H.; Ormrod, D. P., eds. Effects of gaseous air
pollution in agriculture and horticulture. London, United Kingdom:
Butterworth Scientific; pp. 333-348.
Heagle, A. S. (1984) North Carolina State University, Raleigh, NC. Personal
communication to D. T. Tingey, U.S. Environmental Protection Agency,
Corvallis, OR, 1983.
Heagle, A. S. ; Heck, W. W. (1974) Predisposition of tobacco to oxidant air
pollution injury by a previous exposure to oxidants. Environ. Pollut.
7: 247-251.
Heagle, A. S. ; Heck, W. W. (1980) Field methods to assess crop losses due to
oxidant air pollutants. In: Teng, P. S.; Krupa, S. V., eds. Crop loss
assessment: proceedings of E. C. Stakman commemorative symposium. St. Paul,
MN: University of Minnesota: misc. public, no. 7; pp. 296-305.
Heagle, A. S.; Johnson, J. W. (1979) Variable responses of soybean to mixtures
of ozone and sulfur dioxide. J. Air Pollut. Control Assoc. 29: 729-732.
Heagle, A. S. ; Key, L. W. (1973a) Effect of ozone on the wheat stem rust
fungus. Phytopathology 63: 397-400.
Heagle, A. S. ; Key, L. W. (1973b) Effect of Puccini a graminis f. sp. tritici
on ozone injury in wheat. Phytopathology 63: 609-613.
Heagle, A. S.; Letchworth, M. B. (1982) Relationships among injury, growth,
and yield responses of soybean cultivars exposed to ozone at different
light intensities. J. Environ. Qual. 11: 690-694.
Heagle, A. S.; Philbeck, R. B. (1979) Exposure techniques. In: Heck, W. W.;
Krupa, S. V.; Linzon, S. N., eds. Methodology for the assessment of air
pollution effects on vegetation: a handbook from a specialty conference;
April 1978; Minneapolis, MN. Pittsburgh, PA: Air Pollution Control
Association; ch. 6.
Heagle, A. S.; Strickland, A. (1972) Reaction of Erysiphe graminis f. sp.
hordei to low levels of ozone. Phytopathology 62: 1144-1148.
Heagle, A. S. ; Body, D. E. ; Pounds, E. K. (1972) Effect of ozone on yield of
sweet corn. Phytopathology 62: 683-687.
Heagle, A. S. ; Body, D. E.; Heck, W. W. (1973) An open-top field chamber to
assess the impact of air pollution on plants. J. Environ. Qual. 2:
365-368.
Heagle, A. S. , Body, D. E.; Neely, G. E. (1974) Injury and yield responses of
soybean to chronic doses of ozone and sulfur dioxide in the field. Phyto-
pathology 64: 132-136.
6-270
-------
Heagle, A. S. ; Philbeck, R. B. ; Knott, W. M. (1979a) Thresholds for injury
growth, and yield loss caused by ozone on field corn hybrids. Phytopathol-
ogy 69: 21-26.
Heagle, A. S.; Philbeck, R. B.; Letchworth, M. B. (1979b) Injury and yield
response of spinach cultivars to chronic doses of ozone in open-top field
chambers. J. Environ. Qual. 8: 368-273.
Heagle, A. S.; Spencer, S.; Letchworth, M. B. (1979c) Yield response of winter
wheat to chronic doses of ozone. Can. J. Bot. 57: 1999-2005.
Heagle, A. S.; Philbeck, R. B.; Rogers, H. H.; Letchworth, M .B. (1979d) Dis-
pensing and monitoring ozone in open-top field chambers for plant-effect
studies. Phytopathology 69: 15-20.
Heagle, A. S.; Letchworth, M. B.; Mitchell, C. A. (1983a) Effects of growth
medium and fertilizer rate on the yield response of soybeans exposed to
chronic doses of ozone. Phytopathology 73: 134-139.
Heagle, A. S. ; Letchworth, M. B.; Mitchell, C. A. (1983b) Injury and yield
responses of peanuts to chronic doses of ozone in open-top field chambers.
Phytopathology 73: 551-555.
Heagle, A. S. ; Heck, W. W.; Rawlings, J. 0.; Philbeck, R. B. (1983c) Effects
of chronic doses of ozone and sulfur dioxide on injury and yield of
soybeans in open-top field chambers. Crop Sci. 23: 1184-1191.
Heagle, A. S.; Cure, W. W. ; Rawlings, J. 0. (1985) Response of turnips to
chronic doses of ozone in open-top field chambers. Environ. Poll.
(Series A) 38:305-320.
Heath, R. L. (1975) Ozone. In: Mudd, J. B.; Kozlowski, T. T. eds. Responses of
plants to air pollution. New York, NY: Academic Press, Inc.; pp. 23-55.
Heath, R. L. (1980) Initial events in injury to plants by air pollutants. Annu.
Rev. Plant Physio!. 31: 395-431.
Heck, W. W. (1966) The use of plants as indicators of air pollution. Air
Water Pollut. 10: 99-111.
Heck, W. W.; Heagle, A. S. (1970) Measurement of photochemical air pollution
with a sensitive monitoring plant. J. Air Pollut. Control Assoc.
20: 97-99.
Heck, W. W.; Tingey, D. T. (1971) Ozone time-concentration model to predict
acute foliar injury. In: Englund, H. M.; Beery, W. T. , eds. Proceedings
of the second international clean air congress; December 1970; Washington..
DC. New York, NY: Academic Press; pp. 249-255.
Heck, W. W. ; Dunning, J. A.; Hindawi, I. J. (1966) Ozone: nonlinear relation
of dose and injury to plants. Science (Washington, DC) 151: 511-515.
6-271
-------
Heck, W. W.; Dunning, J. A.; Johnson, H. (1968) Design of a simple plant
exposure chamber. U.S. Department of Health, Education, and Welfare,
National Center for Air Pollution Control; publication no. APTD-68-6.
Available from: NTIS, Springfield, VA; PB-195151.
Heck, W. W.; Fox, F. L.; Brandt, C. S.; Dunning, J. A. (1969) Tobacco, a
sensitive monitor for photochemical air pollution. U.S. Department of
Health, Education, and Welfare, National Air Pollution Control Adminis-
tration; publication no. AP-55. Available from: NTIS, Springfield, VA;
PB-190257.
Heck, W. W. ; Larsen, R. I.; Heagle, A. S. (1980) Measuring the acute dose-
response of plants to ozone. In: Assessment of Tosses which constrain
production and crop improvement in agriculture and forests: proceedings
of the E. C. Stakman commemorative symposium. St. Paul, MN: University of
Minnesota; misc. public, no. 7; pp. 32-49.
Heck, W. W. ; Taylor, 0. C. ; Adams, R.; Bingham, G.; Miller, J.; Preston, E.;
Weinstein, L. (1982) Assessment of crop loss from ozone. J. Air Pollut.
Control Assoc. 32: 353-361.
Heck, W. W.; Adams, R. M.; Cure, W. W.; Heagle, A. S.; Heggestad, H. E.;
Kohut, R. J. ; Kress, L. W.; Rawlings, I. 0.; Taylor, 0. C. (1983a) A
reassessment of crop loss from ozone. Environ. Sci. Technol.
17: 573A-581A.
Heck, W. W.; Taylor, 0. C.; Adams, R. M.; Miller, J. E.:
National Crop Loss Assessment Network (NCLAN) 1982
Weinstein, L. (1983b)
Annual Report. Report
Research
to U.S. Environmental Protection Agency, Corvallis Environmental
Laboratory, Corvallis, OR.
Heck, W. W.; Cure, W. W.; Rawlings, J. 0.; Zaragoza, L. J.; Heagle, A. S.;
Heggestad, H. E.; Kohut, R. J.; Kress, L. W.; Temple, P. J.; (1984a)
Assessing impacts of ozone on agricultural crops: I. Overview. J.
Air Pollut. Control Assoc. 34: 729-735.
Heck, W. W. ; Cure, W. W.; Rawlings, J. 0.; Zaragoza, L. J. ; Heagle, A. S. ;
Heggestad, H. E.; Kohut, R. J.; Kress, L. W.; Temple, P. J.; (1984b)
Assessing impacts of ozone on agricultural crops: II. Crop yield functions
and alternative exposure statistics. J. Air Pollut. Control Assoc. 34:
810-817.
Heggestad, H. E. (1973) Photochemical air pollution injury to potatoes in the
Atlantic coastal states. Am. Potato J. 50: 315-328.
Heggestad, H. E.; Bennett, J. H. (1981) Photochemical oxidants potentiate
yield losses in snap beans attributable to sulfur dioxide. Science
(Washington, DC) 213: 1008-1010.
Heggestad, H. E.; Darley, E. F. (1969) Plants as indicators for air pollutants
ozone and PAN. In: Air pollution proceedings of the first European con-
gress on the influence of air pollution on plants and animals. Wageningen,
The Netherlands: Center for Agricultural Publishing Documentation;
pp. 329-355.
6-272
-------
Heggestad, H. E.; Menser, H. A. (1962) Leaf spot-sensitive tobacco strain
Bel W-3, a biological indicator of the air pollutant ozone. Phytopathology
, 52: 735. [Abstract]
Heggestad, H. E. ; Middleton, J. T. (1959) Ozone in high concentrations as
cause of tobacco leaf injury. Science (Washington, DC) 129: 208-210.
Heggestad, H. E.; Burleson, R. F.; Middleton, J. T.; Parley, E. F. (1964) Leaf
injury on tobacco varieties resulting from ozone, ozonated hexene-1 and
ambient air of metropolitan areas. Int. J. Air Water Pollut. 8: 1~10.
Heggestad, H. E.; Heagle, A. S.; Bennett, J. H.; Koch, E. J. (1980) The effects
of photochemical oxidants on the yield of snap beans. Atmos. Environ.
14: 317-326.
Henderson, W. R. ; Reinert, R. A. (1979) Yield response of four fresh market
tomato cultivars after acute ozone exposure in the seedling stage. J.
Am. Soc. Hortic. Sc.i. 104: 754-759.
Hibben, C. R. (1969) Ozone toxicity to sugar maple. Phytopathology 59: 1423-
1428.
Hibben, C. R.; Stotzky, G. (1969) Effects of ozone on the germination of
fungus spores. Can. J. Microbiol. 15: 1187-1196.
Hibben, C. R.; Taylor, M. P. (1975) Ozone and sulphur dioxide effects on the
lilac powdery mildew fungus. Environ. Pollut. 9: 107-114.
Hill, A. C. -(1971) Vegetation: a sink for atmospheric pollutants. J. Air
Pollut. Control Assoc. 21: 341-356.
Hodges, G. H. ; Menser, H. A., Jr.; Ogden, W. B. (1971) Susceptibility of Wis-
consin Havana tobacco cultivars to air pollutants. Agron. J. 63: 107-111.
Hoffman, G. J. ; Maas, E. V.; Rawlins, S. L. (1973) Salinity-ozone interactive
effects on yield and water relations of pinto bean. J. Environ. Qual.
2: 148-152.
Hoffman,-G. J. ; Maas, E. V.; Rawlins, S. L. (1975) Salinity-ozone interactive
effects on alfalfa yield and water relations. J. Environ. Qual.
4: 326-331.
Hofstra,- G. ; Littlejohns, D. A.; Wukasch, R. T. (1978) The efficacy of the
antioxidant ethylene-diurea (EDU) compared to carboxin and benomyl in
reducing yield losses from ozone in navy bean. Plant. Dis. Rep.
62: 350-352.
Hogsett, W. E.; Tingey, D. T.; Holman, S. R. (1985) A programmable exposure
control system for determination of the effects of pollutant exposure
regimes on plant growth. Atmos. Environ. 19: 1135-1145.
6-273
-------
Hoigne, J. ; Bader, H. (1975) Ozonation of water: role of hydroxyl radicals
as oxidizing intermediates. Science (Washington, DC) 190: 782-784.
Hoi ley, J. D.; Hofstra, G.; Hall, R. (1985) Effect of reducing oxidant injury
and early blight on fresh weight and tuber density of potato. Phytopathology
75:529-532.
Holt, D. A.; Daughtry, C. S. T.; Dale, R. F.; Hoi linger, S. E.; Reetz, H. F.;
Nelson, W. L. ; Postage, R. (1979) Separating soil, management, and weather
effects in large area yield prediction. In: Peart, R. , ed. Proceedings
of biological systems simulation group workshop on crop simulation.
West Lafayette, IN: Purdue University.
Hooker, W. J.; Yang, T. C.; Potter, H. S. (1973) Air pollution injury of potato
in Michigan. Am. Potato J. 50: 151-161.
Horsman, D. C. (1981) A survey of ozone in Melbourne using tobacco as an indi-
cator plant. Environ. Pollut. Ser. B: 69-77.
Horsman, D. C. ; Nicholls, A. 0.; Calder, D. M. (1980) Growth responses of
Dactyl is glomerata, Lolium perenne and Phalaris aquatica to chronic ozone
exposure. Aust. J. Plant Physiol. 7: 511-517.
Houston, D. B. (1974) Response of selected Pinus strobus L. clones to fumiga-
tions with sulfur dioxide and ozone. Can. J. For. Res. 4: 65-68.
Howell, R. K. ; Graham, J. H. (1977) Interaction of ozone and bacterial leaf-
spot on alfalfa. Plant Dis. Rep. 61: 565-567.
Howe11, R. K. ; Kremer, D. F. (1973) The chemistry and physiology of pigmen-
tation of leaves injured by air pollution. J. Environ. Qual. 2: 434-438.
Howell, R. K. ; Rose, L. P., Jr. (1980) Residual air pollution effects on soy-
bean seed quality. Plant Dis. 64: 385-386.
Howell, R. K. ; Thomas, C. A. (1972) Relative tolerances of twelve safflower
cultivars to ozone. Plant Dis. Rep. 56: 195-197.
Howell, R. K.; Devine, T. E. ; Hanson, C. H. (1971) Resistance of selected
alfalfa strains to ozone. Crop Sci. 11: 114-115.
Howell, R. K.; Koch, E. J. ; Rose, L. P., Jr. (1979) Field assessment of air
pollution-induced soybean yield losses. Agron. J. 71: 285-288.
Howitt, R. E.; Gossard, T. E.; Adams, R. M. (1984a) Effects of alternative ozone
levels and response data on economic assessments: The case of California
crops. J. Air Pollut. Control Assoc. 34: 1122-1127.
Howitt, R. E. ; Gossard, T.E.; Adams, R.M. 1984b. Applied Welfare Economics
and^Public Policy. New York: Prentice-Hall.
Huang, T. R. ; Aycock, M. K., Jr.; Mulchi, C. L. (1975) Heterosis and combining
ability estimates for air pollution damage, primarily ozone, in Maryland
tobacco. Crop Sci. 15: 785-789.
6-274
-------
Hucl, P.; Beyersdorf, W. D. (1982) The inheritance of ozone sensitivity in
selected Phaseolus vulgaris L. populations. Can. J. Plant Sci. 62: 861-865.
Hughes, P. R.; Laurence, J. A. (1984) Relationship of biochemical effects of
air pollutants on plants to environmental problems: insect and micro-
bial interactions. In: Koziol, M. J.; Whatley, F. R. , eds. Gaseous air
pollutants and plant metabolism. London, United Kingdom: Butterworth
Scientific; pp. 361-377.
Hurwitz, B.; Pell, E. J.; Sherwood, R. T. (1979) Status of coumesterol and 4',
7-dihydroxyflavone in alfalfa foliage exposed to ozone. Phytopathology
69: 810-813.
Jacobson, J. S. (1977) The effects of photochemical oxidants on vegetation.
In: Ozon und Begleitsubstanzen im photochemischen Smog. [Ozone and other
substances in photochemical smog]: VDI colloquium; 1976; Dusseldorf,
West Germany. Dusseldorf, West Germany: Verein deutscher Ingenieure
(VDI) GmbH; pp. 163-173. (VDI-Berichte no. 270).
Jacobson, J. S. (1982) Ozone and the growth and productivity of agricultural
crops. In: Unsworth, M. H. ; Ormrod, D. P., eds. Effects of gaseous air
pollution in agriculture and horticulture. London, United Kingdom:
Butterworth Scientific; pp. 293-304.
Jacobson, J. S. ; Colavito, L. J. (1976) The combined effect of sulfur dioxide
and ozone on bean and tobacco plants. Environ. Exp. Bot. 16: 277-285.
Jacobson, J. S. ; Feder, W. A. (1974) A regional network for environmental
monitoring: atmospheric oxidant concentrations and foliar injury for
tobacco indicator plants in the eastern United States. Amherst, MA:
Massachusetts Agricultural Experiment Station College of Food and
Natural Resources; bulletin no. 604.
James, C. (1980) Economic, social and political implications of crop losses: a
holistic framework for loss assessment in agricultural systems. In: Crop
loss assessment: proceedings of E. C. Stakman commemorative symposium;
St. Paul, MN: University of Minnestoa; misc. public, no. 7; pp. 10-16.
James, R. L. ; Cobb, F. W. , Jr.; Miller, P. R.; Parmeter, J. R. , Jr. (1980a)
Effects of oxidant air pollution on susceptibility of pine roots to
Fomes annosus. Phytopathology 70: 560-563.
James, R. L. ; Cobb, F. W., Jr.; Wilcox, W. W.; Rowney, D. L. (1980b) Effects
of photochemical oxidant injury to ponderosa and Jeffrey pines on sus-
ceptibility of sapwood and freshly cut stumps to Fomes annosus. Phyto-
pathology 70: 704-708.
Jensen, K. F. (1973) Response of nine forest tree species to chronic ozone
fumigation. Plant Dis. Rep. 57: 914-917.
Jensen, K. F. (1979) A comparison of height growth and leaf parameters of
hybrid poplar cuttings grown in ozone-fumigated atmospheres. Washingtion,
DC: U.S. Department of Agriculture; Forest Service research paper NE-446.
6-275
-------
Jensen, K. F. (1981) Air pollutants affect the relative growth rate of hardwood
seedlings. U.S. Department of Agriculture; Forestry Service Research
Paper NE-470.
Jensen, K. F.; Dochinger, L. S. (1974) Responses of hybrid poplar cuttings to
chronic and acute levels of ozone. Environ. Pollut. 6: 289-295.
Jensen, K. F.; Masters, R. G. (1975) Growth of six woody species fumigated
with ozone. Plant Dis. Rep. 59: 760-762.
Johnston, J. W., Jr.; Heagle, A. S. (1982) Response of chronically ozonated
soybean plants to an acute ozone exposure. Phytopathology 72: 387-389.
Jones, J. V.; Pell, E. J. (1981) The influence of ozone on the presence of
isoflavones in alfalfa foliage. J. Air Pollut. Control Assoc. 31: 885-886.
Juhren, M.; Noble, W.; Went, F. W. (1957) The standardization of Poa annua as
an indicator of smog concentrations: I. Effects of temperature, photo-
period, and light intensity during growth of the test-plants. Plant
Physiol. 32: 576-586.
Just, R. E. ; Hueth, D. L. ; Schmitz, A. (1982) Applied welfare economics and
public policy. New York, NY: Prentice-Hall.
Karnosky, D. F. (1977) Evidence for genetic control of response to sulfur
dioxide and ozone in Populus tremuloides. Can. J. For. Res. 7: 437-440.
Kats, G.; Thompson, C. R.; Kuby, W. C. (1976) Improved ventilation of open-top
greenhouses. J. Air Pollut. Control Assoc. 26: 1089-1090.
Keen, N. T.; Taylor, 0. C. (1975) Ozone injury in soybeans: isoflavone accumu-
lation is related to necrosis. Plant Physiol. 55: 731-733.
Kelleher, T. J.; Feder, W. A. (1978) Phytotoxic concentrations of ozone on
Nantucket Island: long range transport from the middle Atlantic states
over the open ocean confirmed by bioassay with ozone-sensitive tobacco
plants. Environ. Pollut. 17: 187-194.
Kender, W. J.; Taschenberg, E. F. ; Shaulis, N. J. (1973) Benomyl protection
of grapevines from air pollution injury. HortScience 8: 396-398.
Kendrick, J. B., Jr.; Middleton, J. T.; Darley, E. F. (1954) Chemical protection
of plants from ozonated olefin (smog) injury. Phytopathology 44: 494-495.
Kendrick, J. B. , Jr.; Darley, E. F.; Middleton, J. T. (1962) Chemotherapy for
oxidant and ozone induced plant damage. Int. J. Air Water Pollut.
6: 391-402.
Kickert, R. N. ; McBride, J. R.; Miller, P. R.; Ohmart, C. P.; Arkley, R. J. ;
Dahlsten, D. L.; Cobb, F. W. ; Cobb, F, W., Jr.; Parmeter, J. R., Jr.;
Luck, R. F.; Taylor, 0. C. (1977) Photochemical oxidant air pollution
effects on a mixed conifer forest ecosystem: final report. Corvallis, OR:
U.S. Environmental Protection Agency, Corvallis Environmental Research
Laboratory; EPA report no. EPA-600/3-80-002. Available from: NTIS,
Springfield, VA; PB80-176779.
6-276
-------
Kohut, R. J.; Davis, D. D. (1978) Response of pinto bean to simultaneous
exposure to ozone and PAN. Phytopathology 68: 567-569.
Kohut, R.; Laurence, J. A. (1983) Yield response of red kidney bean Phaseolus
vulgaris to incremental ozone concentrations in the field. Environ.
Pollut. Ser. A 32: 233-240.
Kohut, R. J. ; Davis, D. D. ; Merrill, W. (1976) Response of hybrid poplar to
simultaneous exposure to ozone and PAN. Plant Dis. Rep. 60: 777-780.
Koiwai, A.; Kisaki, T. (1976) Effect of ozone on photosystem II of tobacco
chloroplasts in the presence of piperonyl butoxide. Plant Cell
Physio!. 17: 1199-1207.
Koiwai, A.; Kitano, H.; Fukuda, M. ; Kisaki, T. (1974) Methylenedioxyphenyl
and its related compounds as protectants against ozone injury to plants.
Agric. Biol. Chem. 38: 301-307.
Koiwai, A.; Fukuda, M.; Kisaki, T. (1977) Effect of piperonyl butoxide and
diphenylamine on lipid peroxidation in ozonated chloroplasts. Plant Cell
Physiol. 18: 127-139.
Kopp, R. J.; Vaughan, W. J.; Hazilla, M. (1984) Agricultural sector benefits
analysis for ozone: methods evaluation and demonstration. Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards; EPA report no. EPA-450/5-84-003.
Available from: NTIS, Springfield, VA; PB85-119477/XAB.
Koukol, J.; Dugger, W. M., Jr. (1967) Anthocyanin formation as a response to
ozone and smog treatment in Rumex crispus L. Plant Physiol. 42: 1023-1024.
Krause, C. R. ; Weidensaul, T. C. (1978a) Effects of ozone on the sporulation,
germination, and pathogenicity of Botrytis cinerea. Phytopathology 18:
195-198.
Krause, C. R. ; Weidensaul, T. C. (1978b) Ultrastructural effects of ozone on
the host-parasite relationship of Botrytis cinerea and Pelargonium hortorum.
Phytopathology 68: 301-307.
Kress, L. W. (1985) Argonne National Laboratory, Argonne, IL. Personal communi-
cation to D. T. Tingey, U.S. Environmental Protection Agency, Corvallis, OR.
Kress, L. W. ; Miller, J. E. (1983) Impact of ozone on soybean yield. J. Environ.
Qual. 12: 276-281.
Kress, L. W.; Skelly, J. M. (1982) Response of several eastern forest tree
species to chronic doses of ozone and nitrogen dioxide. Plant Dis.
66: 1149-1152.
Kress, L. W.; Skelly, J. M.; Hinkelmann, K. H. (1982a) Growth impact of 03,
N02, and/or S02 on Pinus taeda. Environ. Monit. Assess. 1: 229-239.
Kress, L. W.; Skelly, J. M.; Hinkelmann, K. H. (1982b) Growth impact of 03,
N02, and/or S02 on Platanus occidental is. Agric. Environ. 7: 265-274.
6-277
-------
Kress, L. W.; Skelly, I. M.; Hinkelmann, K. H. (1982c) Relative sensitivity
of 18 full-sib families of Pinus taeda to 03. Can. J. For. Res. 12:
203-209.
Kress, L. W. ; Miller, J. E. (1985a) Impact of ozone on grain sorghum. Water,
Air, Soil Pollut. 25: 377-390.
Kress, L. W. ; Miller, J. E. (1985b) Impact of ozone on field corn yield.
Can. J. Bot. 63: 2408-2415.
Kress, L. W. , Miller, J. E. ; Smith, H. J. (1985) Impact of ozone on winter
wheat yeild. Environ. Exp. Bot. 25: 211-228.
Lacasse, N. L. (1971) Assessment of air pollution damage to vegetation in
Pennsylvania. Pennsylvania State University, University Park, PA. CAES
Publication No. 209-71.
Larsen, R. I.; Heck, W. W. (1976) An air quality data analysis system for
interrelating effects, standards, and needed source reduction: Part 3.
Vegetation injury. J. Air Pollut. Control Assoc. 26: 325-333.
Larsen, R. I.; Heck, W. W. (1984) An air quality data analysis system for
interrelating effects, standards, and needed source reductions: Part 8.
An effective mean 03 crop reduction mathematical model. J. Air Pollut.
Control Assoc. 34: 1023-1034.
Larsen, R. L; Heagle, A. S.; Heck, W. W. (1983) An air quality data analysis
system for interrelating effects, standards, and needed source reductions:
Part 7. An 03-S02 leaf injury mathematical model. J. Air Pollut. Control
Assoc. 33: 198-207.
Laurence, J. A. (1981) Effects of air pollutants on plant-pathogen interactions.
Z. Pflanzenkr. Pflanzenschutz 87: 156-172.
Laurence, J. A. (1984) Development of a biological air quality indexing system:
a report to the Minnesota Environmental Quality Board. St. Paul, MN:
Minnesota Environmental Quality Board; SE0184B.
Laurence, J. A.; Wood, F. A. (1978a) Effects of ozone on infection of soybean
by Pseudomonas glycinea. Phytopathology 68: 441-445.
Laurence, J. A.; Wood, F. A. (1978b) Effect of ozone on infection of wild
strawberry by Xanthomonas fragariae. Phytopathology 68: 689-692.
Laurence, J. A.; Wood, F. A.; Krupa, S. V. (1977) Possible transport of ozone
and ozone precursors in Minnesota. Proc. Am. Phytopathol. Soc. 4: 89.
[Abstract]
Laurence, J. A.; MacLean, D. C.; Schneider, R. E. ; Hansen, K. S. (1982) Field
tests of a linear gradient system for exposures of row crops to S02 and
HF. Water Air Soil Pollut. 17: 399-407.
6-278
-------
Lee, J. J.; Preston, E. M.; Lewis, R. A. (1978) A system for the experimental
evaluation of the ecological effects of sulfur dioxide. In: conference pro-
ceedings: 4th joint conference on sensing of environmental pollutants;
November 1977; New Orleans, LA. Washington, DC: American Chemical Society;
pp. 49-53.
Lefohn, A. S. ; Benedict, H. M. (1982) Development of a mathematical index that
describes ozone concentration, frequency, and duration. Atmos. Environ.
16: 2529-2532.
Lefohn, A. S.; Tingey, D. T. (1984) The co-occurrence of potentially phytotoxic
concentrations of various gaseous air pollutants. Atmos. Environ: 18: 2521-
2526.
Legassicke, B. C. ; Ormrod, D. P. (1981) Suppression of ozone-injury on tomatoes
by ethylene diurea in controlled environments and in the field. HortScience
16: 183-184.
(
Legge, A. H.; Savage, D. S.; Walker, R. B. (1979) Special techniques: a portable
gas-exchange leaf chamber. In: Heck, W. W. ; Krupa, S. V. ; Linzon, S. N.,
eds. Handbook of methodology for the assessment of air pollution effects
on vegetation: Air Pollution Control Association specialty conference
proceedings. Pittsburgh, PA: Air Pollution Control Association; pp. 16-13
to 16-24.
Leh, F. ; Mudd, J. B. (1974) Reaction of peroxyacetyl nitrate with cysteine,
cystine, methionine, lipoic acid, papain, and lysozyme. Arch. Biochem.
Biophys. 161: 216-221.
Letchworth, M. B. ; Blum, U. (1977) Effects of acute ozone exposure on growth,
nodulation, and nitrogen content of ladino clover. Environ. Pollut.
14: 303-312.
Leung, S. K.; Reed, W. ; Cauchois, S.; Howitt, R. (1978) Methodologies for
valuation of agricultural crop yield changes: a review. Corvallis, OR:
U.S. Environmental Protection Agency, Corvallis Environmental Research
Laboratory; EPA report no. EPA-600/5-78-018. Available from: NTIS,
Springfield, VA; PB-288122.
Leung, S. K.; Reed, W.; Geng, S. (1982) Estimations of ozone damage to selected
crops grown in southern California. J. Air Pollut. Control Assoc. 32:
160-164.
Lewis, E. ; Brennan, E. (1978) Ozone and sulfur dioxide mixtures cause
a PAN-type injury to petunia. Phytopathology. 68: 1011-1014.
Loucks, 0. L. ; Armentano, T. V. (1982) Estimating crop yield effects from
ambient air pollutants in the Ohio River Valley. J. Air Pollut. Control
Assoc. 32: 146-150.
Ludlow, M. M. (1980) Adaptive significance of stomatal responses to water
stress. In: Turner, N. C.; Kramer, P. J. , eds. Adaptation of plants to
water and high temperature stress. New York, NY: Wiley-Interscience;
pp. 123-128.
6-279
-------
MacDowall, F. D. H. (1965) Predisposition of tobacco to ozone damage. Can. J.
Plant Sci. 45: 1-12.
MacLean, D. C.; Schneider, P. E. (1976) Photochemical oxidants in Yonkers, New
York: effects on yield of bean and tomato. J. Environ. Qua!. 5: 75-78.
Male, L.; Preston, E.; Neely, G. (1983) Yield response curves of crops exposed
to S02 time series. Atmos. Environ. 17: 1589-1593.
Mandl, R. L. ; Weinstein, L. H.; McCune, D. C.; Keveny, M. (1973) A cylindrical,
open-top chamber for the exposure of plants to air pollutants in the
field. J. Environ. Qual. 2: 371-376.
Mann, L. K.; Mclaughlin, S. B. ; Shriner, D. S. (1980) Seasonal physiological
responses of white pine under chronic air pollution stress. Environ. Exp.
Bot. 20: 99-105.
Manning, W. J. (1975) Interactions between air pollutants and fungal, bacterial
and viral plant pathogens. Environ. Pollut. 9: 87-89.
Manning, W. J. ; Feder, W. A. (1980) Biomonitoring air pollutants with plants.
London, United Kingdom: Applied Science Publishers.
Manning, W. J.; Feder, W. A.; Perkins, I.; GTickman, M. (1969) Ozone injury
and infection of potato leaves by Botrytis cinerea. Plant Dis. Rep.
53: 691-693.
Manning, W. J.; Feder, W. A.; Perkins, I. (1970a) 'Ozone and infection of
geranium flowers by Botrytis cinerea. Phytopathology 60: 1302. [Abstract]
Manning, W. J.; Feder, W. A.; Perkins, I. (1970b) Ozone injury increases
infection of geranium leaves by Botrytis cinerea. Phytopathology 60: 669-
670.
Manning, W. J.; Feder, W. A.; Papia, P. M.; Perkins, I. (1971a) Effect of low
levels of ozone on growth and susceptibility of cabbage plants to Fusarium
oxysporum F. sp. conglutinans. Plant Dis. Rep. 55: 47-49.
Manning, W. J. ; Feder, W. A.; Papia, P. M.; Perkins, I. (1971b) Influence of
foliar ozone injury on root development and root surface fungi of pinto
bean plants. Environ. Pollut. 1: 305-312.
Manning, W. J.; Feder, W. A.; Perkins, I. (1972) Effects of Botrytis and ozone
on bracts and flowers of poinsettia cultivars. Plant Dis. Rep. 56: 814-816.
Manning, W. J.; Feder, W. A.; Bardo, P. M. (1974) Suppression of oxidant injury
by benomyl: effects on yields of bean cultivars in field. J. Environ. Qual.
2: 1-3.
Manuel, E. H. ; Brennan, K. L.; Horst, R. L., Jr. (1982) Benefit analysis of
alternative secondary national ambient air quality standards for sulfur
dioxide and total suspended particulates: v. IV. Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards; EPA report no. EPA-450/5-83-001D. Available
from: NTIS, Springfield, VA; PB83-258368.
6-280
-------
Matsushima, J. (1971) Shokubutsu ni taisuru argusangasu to okushidanto no
fukugogai ni tsuite [The composite damage to plants by sulfur dioxide gas
and oxidant]. Sangyo Kogai 7: 218-224.
McClenahen, J. R. (1979) Effects of ethylene diurea and ozone on the growth of
tree seedlings. Plant Dis. Rep. 63: 320-323.
McCool, P. M. ; Menge, J. A. (1983) Influence of ozone on carbon partitioning
in tomato: potential role of carbon flow in regulation of mycorrhizal
symbiosis under conditions of stress. New Phytol. 94: 241-247.
McCool, P. M.; Menge, J. A.; Taylor, 0. C. (1979) Effects of ozone and HC1 gas
on the development of the mycorrhizal fungus Glomus faciculatus and
growth of 'Troyer' citrange. J. Am. Soc. Hortic. Sci. 104: 151-154.
McCool, P. M.; Menge, J. A.; Taylor, 0. C. (1982) Effect of ozone injury and
light stress on response of tomato to infection by the vesicular-arbuscular
mycorrhizal fungus, Glomus fasciculatus. J. Am. Soc. Hortic. Sci. 107: 839-
842.
Mclaughlin, S. B.; Taylor, G. E. (1981) Relative humidity: important modifier
of pollutant uptake by plants. Science (Washington, DC) 221: 167-169.
Mclaughlin, S. B.; Shriner, D. S.; McConathy, R. K.; Mann, L. K. (1979) The
effects of S02 dosage kinetics and exposure frequency on photosynthesis
and transpiration of kidney beans (Phaseolus vulgaris L.). Environ. Exp.
Bot. 19: 179-191.
Mclaughlin, S. B.; McConathy, R. K.; Duvick,. D.; Mann, L. K. (1982) Effects of
chronic air pollution stress on photosynthesis, carbon allocation, and
growth of white pine trees. For. Sci. 28: 60-70.
Meiners, J. P.; Heggestad, H. E. (1979) Evaluation of snap bean cultivars for
resistance to ambient oxidants in field plots and to ozone in chambers.
Plant Dis. Rep. 63: 273-277.
Menser, H. A., Jr. (1969) Effects of air pollution on tobacco cultivars grown
in several states. Tob. Sci. 169: 699-704.
Menser, H. A.; Heggestad, H. E. (1966) Ozone and sulfur dioxide synergism:
injury to tobacco plants. Science (Washington, DC) 153: 424-425.
Menser, H.A.; Hodges, G. H. (1970) Effects of air pollutants on burley tobacco
cultivars. Agron. J. 62: 265-269.
Menser, H. A.; Hodges, G. H. (1972) Oxidant injury to shade tobacco cultivars
developed in Connecticut for weather fleck resistance. Agron. J. 64: 189-
192.
Menser, H. A.; Street, 0. E. (1962) Effects of air pollution, nitrogen levels,
supplemental irrigation, and plant spacing on weather fleck and leaf
losses of Maryland tobacco. Tobacco 155: 192-196.
6-281
-------
Metzler, J. T. ; Pell, E. J. (1980) The impact of peroxyacetyl nitrate on
conductance of bean leaves and on associated cellular and foliar symptom
expression. Phytopathology 70: 934-938.
Middleton, J. T.; Haagen-Smit, A. (1961) The occurrence, distribution and sig-
nificance of photochemical air pollution in the United States, Canada,
and Mexico. J. Air Pollut. Control Assoc. 11: 129-134.
Middleton, J. T. ; Kendrick, J. B., Jr.; Schwalm, H. W. (1950) Injury to herba-
ceous plants by smog or air pollution. Plant Dis. Rep. 34: 245-252.
Middleton, J. T. ; Crafts, A. S. ; Brewer, R. F. ; Taylor, 0. C. (1956) Plant
damage by air pollution. Calif. Agric. 10: 9-12.
Millecan, A. A. (1971) A survey and assessment of air pollution damage to
California vegetation in 1970. Research Triangle Park, NC: U.S. Environ-
mental Protection Agency, Air Pollution Control Office; APTD-0694.
Millecan, A. A. (1976) A survey and assessment of air pollution damage to
California vegetation: 1970 through 1974. Sacramento, CA: California
Department of Food and Agriculture.
Miller, P. R. (1973) Oxidant-induced community change in a mixed conifer
forest. In: Naegle, J. A., ed. Air pollution damage to vegetation.
Washington, DC: American Chemical Society; pp. 101-117. (Advances in
chemistry series no. 122).
Miller, C. A.; Davis, D. D. (1981a) Effect of temperature on stomatal conduc-
tance and ozone injury of pinto bean leaves. Plant Dis. 65: 750-751.
Miller, P. R.; Elderman, M. J., eds. (1977) Photochemical oxidant air
pollutant effects on a mixed conifer forest ecosystem: a progress
report, 1976. Corvallis, OR: Environmental Protection Agency; EPA
report no. EPA-600/3-77-104. Available from: NTIS, Springfield, VA;
PB-274531.
Miller, P. R. ; Parmeter, J. R., Jr.; Flick, B. H.; Martinez, C. W. (1969)
Ozone dosage response of ponderosa pine seedlings. J. Air Pollut. Control
Assoc. 19: 435-438.
Miller, P. R. ; McCutchan, M. H.; Millegan, H. P. (1972) Oxidant air pollution
in the central valley, Sierra Nevada foothills and Mineral King Valley of
California. Atmos. Environ. 6: 623-633.
Miller, V. L. ; Howell, R. K. ; Caldwell, B. E. (1974) Relative sensitivity of
soybean genotypes to ozone and sulfur dioxide. J. Environ. Qual. 3: 35-37.
Miller, P. M. ; Tomlinson, H.; Taylor, G. S. (1976) Reducing severity of ozone
damage to tobacco and beans by combining benomyl or carboxin with contact
nematicides. Plant Dis. Rep. 60: 433-436.
Miller, P. R. ; Taylor, 0. C. ; Wilhour, R. G. (1982) Oxidant air pollution
effects on a western coniferous forest ecosystem. Corvallis, OR: Corvallis
Environmental Research Laboratory; EPA report no. EPA-600/D-82-276.
Available from: NTIS, Springfield, VA; PB83-189308.
6-282
-------
Mitchell, A.; Ormrod, D. P.; Dietrich, H. F. (1979) Ozone and nickel effects on
pea leaf cell ultrastructure. Bull. Environ. Contain. Toxicol. 22: 379-385.
Mjelde, J. W.; Adams, R. M. ; Dixon, B. L. ; Garcia, P. (1984) Using farmers'
actions to measure crop loss due to air pollution. J. Air Pollut. Control
Assoc. 31: 360-364.
Mooi, J. (1980) Influence of ozone on the growth of two poplar cultivars.
Plant Dis. 64: 772-773.
Moskowitz, P. D.; Coveney, E. A.; Medeiros, W. H. ; Morris, S. C. (1982) Oxidant
air pollution: a model for estimating effects on U.S. vegetation. J. Air
Pollut. Control Assoc. 32: 155-160.
Moyer, J. W. ; Cole, H., Jr.; Lacasse, N. L. (1974) Suppression of naturally
occurring oxidant injury on azalea plants by drench or foliar spray
treatment with benzimidazole or oxathiin compounds. Plant Dis. Rep.
58: 136-138.
Moyer, J. W. ; Smith, S. H. (1975) Oxidant injury reduction on tobacco induced
by tobacco etch virus infection. Environ. Pollut. 9: 103-106.
Mudd, J. B. (1963) Enzyme inactivation by peroxyacetyl nitrate. Arch. Biochem.
Biophys. 102: 59-65.
Mudd, J. B. (1966) Reaction of peroxyacetyl nitrate with glutathione. J. Biol.
Chem. 241: 4077-4080.
Mudd, J. B. (1975) Peroxyacyl nitrates. In: Mudd, J. B.; Kozlowski, T. T. ,
eds. Response of plants to air pollutants. New York, NY: Academic Press,
Inc.; pp. 97-119.
Mudd, J. B. (1982) Effects of oxidants on metabolic function. In: Unsworth,
M. H. ; Ormrod, D. P., eds. Effects of gaseous air pollution in agricul-
ture and horticulture. London, United Kingdom: Butterworth Scientific;
pp. 189-203.
Mudd, J. B. ; Dugger, W. M. (1963) The oxidation of reduced pyridine nucleo-
tides by peroxyacetyl nitrates. Arch. Biochem. Biophys. 102: 52-58.
Mudd, J. B.; Leavitt, R.; Kersey, W. H. (1966) Reaction of peroxyacetyl nitrate
with sulfhydryl groups of proteins. J. Biol. Chem. 241: 4081-4085.
Mukammal, E. I. (1965) Ozone as a cause of tobacco injury. Agricultural
Meteorology 2: 145-165.
Mumford, R. A.; Lipke, H. ; Laufer, D. A.; Feder, W. A. (1972) Ozone-induced
changes in corn pollen. Environ. Sci. Technol. 6: 427-430:
Musselman, R. C.; Kender, W. J.; Crowe, D. E. (1978) Determining air pollutant
effects on the growth and productivity of "Concord" grapevines using
open-top chambers. J. Am. Soc. Hortic. Sci. 103: 645-648.
6-283
-------
Musselman, R. C. ; Oshima, R. J.; Gallavan, R. E. (1983) Significance of pollut-
ant concentration distribution in the response of 'red kidney1 beans to
ozone. J. Am. Soc. Hortic. Sci. 108: 347-351.
Nakamura, H. ; Matsunaka, S. (1974) Indicator plants for air pollutants. (1)
Susceptibility of morning glory to photochemical oxidants: varietal
difference and effect of environmental factors. Nippon Sakumotsu Gakkai
Kiji 43: 517-522. [English summary]
Nakamura, H. ; Saka, H. (1978) Photochemical oxidants injury in rice plants.
III. Effect of ozone on physiological activities in rice plants. Nippon
Sakumotsu Gakkai Kiji 47: 707-714. [English summary]
Nash, T. H., III; Sigal, L. L. (1979) Gross photosynthetic response of lichens
to short-term ozone fumigations. Bryologist 82: 280-285.
National Research Council. (1977) Plants and microorganisms. In: Ozone and other
photochemical oxidants. Washington, DC: National Academy of Sciences;
pp. 437-585.
Naveh, Z. ; Chaim, S. ; Steinberger, E. H. (1978) Atmospheric oxidant concen-
trations in Israel as manifested by foliar injury in Bel-W3 tobacco plants.
Environ. Pollut. 16: 249-262.
Neely, G. E.; Tingey, D. T.; Wilhour, R. G. (1977) Effects of ozone and sulfur
dioxide singly and in combination on yield, quality and N-fixation of
alfalfa. In: Dimitriades, B., ed. International conference on photochemi-
cal oxidant pollution and its control: proceedings, v. II; September 1976;
Raleigh, NC. Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Research and Development; pp. 663-673; EPA report no.
EPA-600/3-77-0016b. Available from: NTIS, Springfield, VA; PB-264233.
Nicksic, S. W.; Harkins, J. ; Mueller, P. K. (1967) Some analyses for PAN and
studies of its structure. Atmos. Environ. 1: 11-18.
Noble, W. M. (1955) Pattern of damage produced on vegetation by smog. J. Agric.
Food Chem. 3: 330-332.
Noble, W. M. (1965) Smog damage to plants. Lasca Leaves 15: 1-18.
Noble, R. D. ; Jensen, K. F. (1980) Effects of sulfur dioxide and ozone on
growth of hybrid poplar leaves. Am. J. Bot. 67: 1005-1009.
Noble, R. D.; Pechak, D.; Jensen, K. (1980) Ozone effects on the ultrastructure
of the chloroplasts from hybrid poplar leaves. Micron 11: 13-14.
Noland, T. L. ; Kozlowski, T. T. (1979) Influence of potassium nutrition on
susceptibility of silver maple to ozone. Can. J. For. Res. 9: 501-503.
Nouchi, I.; Aoki, K. (1979) Morning glory as a photochemical oxidant indicator.
Environ. Pollut. 18: 289-303.
6-284
-------
Nouchi, I.; Mayumi, H.; Yamazoe, F. (1984) Foliar injury response of petunia
and kidney bean to simultaneous and alternate exposure to ozone and PAN.
Atmos. Environ. 18: 453-460.
O'Connor, J. A.; Parbery, D. G. ; Strauss, W. (1975) The effects of phyto-
toxic gases on native Australian plant species: part 2. acute injury due
to ozone. Environ. Pollut. 9: 181-192.
Ogata, G.; Maas, E. V. (1973) Interactive effects of salinity and ozone on
growth and yield of garden beet. J. Environ. Qual. 2: 518-520.
Olson, B. ; Saettler, A. W. (1979) Interaction of ozone and common bacterial
blight in two dry bean cultivars. In: Proceedings of the IX international
congress on plant protection; August; Washington, DC. Abstract 192.
St. Paul, MN; American Phytopathology Soc.
Olszyk, D. M.; Tibbitts, T. W. (1981) Stomatal response and leaf injury of
Pi sum sativum L. with S02 and 03 exposures. Plant Physiol. 67: 539-544.
Olszyk, D. M.; Tibbitts, T. W.; Hertzberg, W. M. (1980) Environment in open-top
field chambers utilized for air pollution studies. J. Environ. Qual.
9: 610-615.
Ordin, L (1962) Effect of peroxyacetyl nitrate on growth and cell wall metab-
olism of Avena coleoptile sections. Plant Physiol. 37: 603-608.
Ordin, L.; Hall, M. A. (1967) Studies on cellulose synthesis of a cell-free
oat coleoptile enzyme system: inactivation by air pollutant oxidants.
Plant Physiol. 42: 205-212.
Ordin, L.; Propst, B. (1962) Effect of airborne oxidants on biological activity
of indoleacetic acid. Bot. Gaz. (Chicago) 123: 170-174.
Ordin, L.; Hall, M. A.; Katz, M. (1967) Peroxyacetyl nitrate-induced inhibition
of cell wall metabolism. J. Air Pollut. Control Assoc. 17: 811-815.
Ordin, L.; Garber, J. J.; Kindinger, J. I.; Whitmore, S. A.; Greve, L. C. ;
Taylor, 0. C. (1971) Effect of peroxyacetyl nitrate (PAN) ij} vivo on
tobacco leaf polysaccharide synthetic pathway enzymes. Environ. Sci.
Techno!. 5: 621-626.
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. (1982) Air pollutant interactions in mixtures. In: Unsworth,
M. H.; Ormrod, D. P., eds. Effects of gaseous air pollution in agriculture
and horticulture. London, United Kingdom: Butterworth Scientific;
pp. 307-331.
Ormrod, D. P.; Kemp, W. G. (1979) Ozone response of tomato plants infected
with cucumber mosaic virus and/or tobacco mosaic virus. Can. J. Plant
Sci. 59: 1077-1083.
6-285
-------
Ormrod, D. P.; Adedipe, N. 0.; Hofstra, G. (1971) Responses of cucumber,
onion, and potato cultivars to ozone. Can. J. Plant Sci. 51: 283-288.
Ormrod, D. P.; Tingey, D. T.; Gumpertz, M. L.; Olszyk, D. M. (1984) Utilization
of a response surface technique in the study of plant responses to ozone
and sulfur dioxide mixtures. Plant Physiol. 75: 43-48.
Oshima, R. J. (1973) Effect of ozone on a commercial sweet corn variety. Plant
Dis. Rep. 57: 719-723.
Oshima, R. J. (1974) A viable system of biological indicators for monitoring
air pollutants. J. Air. Pollut. Control Assoc. 24: 576-578.
Oshima, R. J. (1978) The impact of sulfur dioxide on vegetation: a sulfur
dioxide-ozone response model. Sacramento, CA: California Air Resources
Board; final report; agreement no. A6-162-30.
Oshima, R. J.; Gallavan, R. (1980) Experimental designs for the quantification
of crop growth and yield responses from air pollutants. In: Teng, P. S.;
Krupa, S. V., eds. Crop loss assessment: proceedings of the E. C. Stakman
commemorative symposium. St. Paul, MN: University of Minnesota; misc. public.
no. 7; pp. 63-70.
Oshima, R. J.; Taylor, 0. C.; Cardiff, E. A. (1974) Severe air pollution
episode in south coast basin. Calif. Agric. 28: 12-13.
Oshima, R. J.; Taylor, 0. C.; Braegelmann, P. K.; Baldwin, D. W. (1975) Effect
of ozone on the yield and plant biomass of a commercial variety of tomato.
J. Environ. Qual. 4: 463-464.
Oshima, R. J.; Poe, M. P.; Braegelmann, P. K.; Baldwin, D. W.; Van Way, V.
(1976) Ozone dosage-crop loss function for alfalfa: a standardized method
for assessing crop losses from air pollutants. J. Air Pollut. Control
Assoc. 26: 861-865.
Oshima, R. J.; Braegelmann, P. K.; Baldwin, D. W.; Van Way, V.; Taylor, 0. C.
(1977a) Reduction of tomato fruit size and yield by ozone. J. Am. Soc.
Hortic. Sci. 102: 289-293.
Oshima, R. J.; Braegelmann, P. K.; Baldwin, D. W.; Van Way, V.; Taylor, 0. C.
(1977b) Responses of five cultivars of fresh market tomato to ozone: a
contrast of cultivar screening with foliar injury and yield. J. Am. Soc.
Hortic. Sci. 102: 286-289.
Oshima, R. J. ; Bennett, J. P. ; Braegelmann, P. K. (1978) Effect of ozone on
growth and assimilate partitioning in parsley. J. Am. Soc. Hortic. Sci.
103: 348-350.
Oshima, R. J.; Braegelmann, P. K. ; Flagler, R. B.; Teso, R. R. (1979) The
effects of ozone on the growth, yield, and partitioning of dry matter in
cotton. J. Environ. Qual. 8: 474-479.
6-286
-------
Page, W. P.; Arbogast, G.; Fabian, R. G.; Ciecka, J. (1982) Estimation of
economic losses to the agricultural sector from airborne residuals in the
Ohio River Basin region. J. Air Pollut. Control Assoc. 32: 151-154.
Patton, R. L. (1981) Effects of ozone and sulfur dioxide on height and stem
specific gravity of Populus hybrids. Broomall, PA: Northeastern Forest
Experiment Station; FSRP-NE-471. Available from: NTIS, Springfield, VA;
PB81-171274.
Peake, E. ; Sandhu, H. S. (1983) The formation of ozone and peroxyacetyl nitrate
(PAN) in the urban atmospheres of Alberta. Can J. Chem. 61: 927-935.
Pearson, R. G.; Drummond, D. B. ; Mcllveen, W. D.; Linzon, S. N. (1974) PAN-
type injury to tomato crops in southwestern Ontario. Plant Dis. Rep.
. 58: 1105-1108.
Pell, E. J. (1973) 1972 survey and assessment of air pollution damage to
vegetation in New Jersey. U.S. Environmental Protection Agency,
Washington, DC. Publication No. EPA-R5-022.
Pell, E. J. (1976) Influence of benomyl soil treatment on pinto bean plants
exposed to peroxyacetyl nitrate and ozone. Phytopathology 66: 731-733.
Pell, E. J. (1979) How air pollutants induce disease. In: Horsfall, J. ; Cowling,
E., eds. Plant diseases, v. IV. New York, NY: Academic Press, Inc.;
pp. 273-292.
Pell, E. J.; Brennan, E. (1973) Changes in respiration, photosynthesis, adeno-
sine B'-triphosphate, and total adenylate content of ozonated pinto bean
foliage as they relate to symptom expression. Plant Physiol. 51: 378-381.
Pell, E. J. ; Gardner, W. (1975) Benomyl stimulation of peroxyacetyl nitrate
injury to petunia. Proc. Am. Phytopathol. Soc. 2: 29. [Abstract]
Pell, E. J.; Gardner, W. (1979) Enhancement of peroxyacetyl nitrate injury to
petunia foliage by benomyl. HortScience 14: 61-62.
Pell, E. J. ; Pearson, N. S. (1983) Ozone-induced reduction in quantity of
ribulose-l,5-biphosphate carboxylase in alfalfa foliage. Plant Physiol.
73: 185-187.
Pell, E. J. ; Weissberger, W. C. (1976) Histopathological characterization of
ozone injury to soybean foliage. Phytopathology 66: 856-861.
Pell, E. J.; Lukezic, F. L.; Levine, R. G.; Weissberger, W. C. (1977) Response
of soybean foliage to reciprocal challenges by ozone and a hypersensitive-
response-inducing Pseudomonad. Phytopathology 67: 1342-1345.
Pell, E. J. ; Weissberger, W. C.; Speroni, J. J. (1980) Impact of ozone on
quantity and quality of greenhouse-grown potato plants. Environ. Sci.
Technol. 14: 568-571.
Penkett, S. A.; Sandalls, F. J.; Jones, B. M. R. (1977) PAN measurements in
England—analytical methods and results. VDI Ber. 270: 47-54.
6-287
-------
Pippen, E. L. ; Potter, A. L. ; Randall, V. G.; Ng, K. C. ; Reuter, F. W., III;
Morgan, A. I., Jr. (1975) Effect of ozone fumigation on crop composition.
J. Food Sci. 40: 672-676.
Posthumus, A. C. (1976) The use of higher plants as indicators for air pollu-
tion in the Netherlands. In: Karenlampi, L. , ed. Proceedings of the
Kuopio meeting on plant damages caused by air pollution; Kuopio, Finland;
pp. 115-120.
Posthumus, A. C. (1977) Experimentelle Untersuchungen der Wirkung von Ozon
und Peroxyacetylnitrat (PAN) auf Pflanzen. [Experimental investigations
of the effect of ozone and peroxyacetylnitrate (PAN) on plants.] In:
Ozon und begleit Substanzen im photochemischen Smog [Ozone and other
substances in photochemical smog]: VDI colloquium; 1976; Dusseldorf,
West, Germany. Dusseldorf, West Germany: Verein deutscher Ingenieure
(VDI) GmbH; pp. 153-161. (VDI-Berichte no. 270.)
Posthumus, A. C. (1980) Monitoring levels and effects of air-borne pollutants
on vegetation: use of biological indicators and other methods in national
and international programmes. In: Papers presented to the symposium on the
effects of air-borne pollution on vegetation; Warsaw, Poland. United
Nations, Economic Commission for Europe; pp. 296-311.
Povilaitis, B. (1967) Gene effects for tolerance to weather fleck in tobacco.
Can. J. Genet. Cytol. 9: 327-334.
Pratt, G. C. ; Krupa, S. V. (1979) Interaction between ozone and certain bio-
pathogens on soybean cultivar Hodgson. In: Proceedings of the XI interna-
tional congress of plant protection; Washington, DC. Abstract 196.
Pratt, G. C. ; Krupa, S. V. (1981) Soybean cultivar Hodgson response to ozone.
Phytopathology 71: 1129-1132.
Pryor, W. A.; Prier, D. G.; Church, D. F. (1981) Radical production from the
interaction of ozone and PUFA as demonstrated by electron spin resonance
spin-trapping techniques. Environ. Res. 24: 42-52.
Rajput, C. B. S.; Ormrod, D. P. (1976) Response of eggplant cultivars to
ozone. HortScience 11: 462-463.
Rawlings, J. 0.; Cure, W. W. (1985) The Weibull function as a dose-response
model to describe ozone effects on crop yields. Crop Science 25: 807-814.
Reich, P. B. (1983) Effects of low concentrations of 0- on net photosynthesis,
dark respiration, and chlorophyll contents in aging hybrid poplar leaves.
Plant Physiol. 73: 291-296..
Reich, P. B. ; Amundson, R. G. (1984) Low level 03 and/or S02 exposure causes
a linear decline in soybean yield. Environ. Pollut. (Series A) 34: 345-355.
6-288
-------
Reich, P. B.; Amundson, R. G.; Lassoie, J. P. (1980) Multiple pollutant fumiga-
tions under near ambient environmental conditions using a linear gradient
technique. In: Miller, P. R. , ed. Proceedings of the symposium on effects
of air pollutants on Mediterranean and temperate forest ecosystems; June;
Riverside, CA. Berkeley, CA: U.S. Department of Agriculture, Forest
Service general technical report no. PSW-43; p. 247. Available from: NTIS,
Springfield, VA; PB81-133720.
Reilly, J. J.; Moore, L. D. (1982) Influence of selected herbicides on ozone
injury in tobacco (Nicotiana tabacum). Weed Sci. 30: 260-263.
Reinert, R. A. (1975) Monitoring, detecting, and effects of air pollutants on
horticultural crops: sensitivity of genera and species. HortScience
10: 495-500.
Reinert, R. A.; Gooding, G. V., Jr.; (1978) Effect of ozone and tobacco streak
virus alone and in combination on Nicotiana tabacum. Phytopathology
68: 15-17.
Reinert, R. A. ; Gray, T. N. (1981) The response of radish to nitrogen dioxide,
sulfur dioxide, and ozone, alone and in combination. J. Environ. Qual.
10: 240-243.
Reinert, R. A.; Heagle, A. S.; Miller, J. R.; Geckler, W. R. (1970) Field
studies of air pollution injury to vegetation in Cincinnati, Ohio.
Plant Dis. Rep. 54: 8-11.
Reinert, R. A.; Heck, W. W. (1982) Effects of nitrogen dioxide in combination
with sulfur dioxide and ozone on selected corps. In: Schneider, T. ;
Grant, L., eds. Air pollution by nitrogen oxides. Amsterdam, The
Netherlands: Elsevier Scientific; pp. 533-546.
Reinert, R. A.; Henderson, W. R. (1980) Foliar injury and growth of tomato
cultivars as influenced by ozone dose and plant age. J. Am. Soc. Hortic.
Sci. 105: 322-324.
Reinert, R. A.; Nelson, P. V. (1979) Sensitivity and growth of. twelve elatior
begonia cultivars to ozone. HortScience 14: 747-748.
Reinert, R. A.; Nelson, P. V. (1980) Sensitivity and growth of five elatior
begonia cultivars to S02 and 03, alone and in combination. J. Am. Soc.
Hortic. Sci. 105: 721-723.
Reinert, R. A.; Sanders, J. S. (1982) Growth of radish and marigold following
repeated exposure to nitrogen dioxide, sulfur dioxide, and ozone. Plant
Dis. 66: 122-124.
Reinert, R. A.; Spurr, H. W., Jr. (1972) Differential effect of fungicides on
ozone injury and brown spot disease of tobacco. J. Environ. Qual.
1: 450-452.
Reinert, R. A.; Tingey, D. T. ; Carter, H. B. (1972) Ozone induced foliar
injury in lettuce and radish cultivars. J. Am. Soc. Hortic. Sci.
97: 711-714.
6-289
-------
Reinert, R. A.; Heagle, A. S.; Heck, W. W. (1975) Plant responses to pollutant
combinations. Mudd, B.; Kozlowski, T. T. , eds. Responses of plants to air
pollution. New York, NY: Academic Press, Inc.; pp. 159-177.
Rich, S. (1964) Ozone damage to plants. Annu. Rev. Phytopathol. 2: pp. 253-266.
Rich, S. ; Taylor, G. S. (1960) Antiozonants to protect plants from ozone
damage. Science (Washington, DC) 132: 150-151.
Rich, S. R. ; Taylor, G. S. ; Tomlinson, H. (1969) Crop damaging periods of
ambient ozone in Connecticut. Plant Dis. Rep. 53: 969-973.
Rich, S. ; Tomlinson, H. (1968) Effects of ozone on conidiophores and conidia
of Alternaria solani. Phytopathology 58: 444-446.
Rich, S. ; Turner, N. C. (1972) Importance of moisture on stomatal behavior of
plants subjected to ozone. J. Air Pollut. Control Assoc. 22: 718-721.
Richards, B. L. ; Middleton, J. T. ; Hewitt, W. B. (1958) Air pollution with
relation to agronomic crops: V. Oxidant stipple of grape. Agron. J.
50: 559-561.
Richards, G. A.; Mulchi, C. L.; Hall, J. R. (1980) Influence of plant maturity
on the sensitivity of turfgrass species to ozone. J. Environ. Qual.
9: 49-53.
Rist, D. L.; Lorbeer, J. W. (1981) Interactions of ozone, foliar fungi, and the
surface and tissues of the leaf. In: Blakeman, J. P., ed. Microbial ecology
of the phylloplane. London, United Kingdom: Academic Press; pp. 305-327.
Rogers, H. H. ; Jeffries, H. E.; Stahel, E. P.; Heck, W. W.; Ripperton, L. A.;
Witherspoon, A. M. (1977) Measuring air pollutant uptake by plants: a
direct kinetic technique. J. Air Pollut. Control Assoc. 27: 1192-1197.
Roose, M. L. ; Bradshaw, A. D. ; Roberts, T. M. (1982) Evolution of resistance
to gaseous air pollutants. In: Unsworth, M. H. ; Ormrod, D. P., eds.
Effects of gaseous air pollution in agriculture and horticulture. London,
United Kingdom: Butterworth Scientific; pp. 379-409.
Ross, L. J.; Nash, T. H., III. (1983) Effect of ozone on gross photosynthesis
of lichens. Environ. Exp. Bot. 23: 71-77.
Rowe, R. P.; Chestnut, L. G. ; Miller, C. ; Adams, R. M. ; Thresher, M. ;
Mason, H. 0.; Howitt, R. E. ; Trijonis, J. (1984) Economic assessment of
the effect of air pollution in the San Joaquin Valley: draft report to
the Research Division, California Air Resources Board. Boulder, CO:
Energy and Resource Consultants, Inc.
Rubin, B. ; Leavitt, J. R. C. ; Penner, D. ; Saettler, A. W. (1980) Interaction
of antoxidants with ozone and herbicide stress. Bull. Environ. Contam.
Toxicol. 25: 623-629.
Runeckles, V. C.; Rosen, P. M. (1977) Effects of ambient ozone pretreatment on
transpiration and susceptibility to ozone injury. Can. J. Bot. 55: 193-197.
6-290
-------
Ryan, J. W.; Loehman, E.; Lee, W.; Trondsen, E.; Bland, M.; Goen, R. ; Ludwig,
F. ; Eger, T.; Eigsti, S. ; Conley, D.; Cummings, R. (1981) An estimate of
the nonhealth benefits of meeting the secondary national ambient air
quality standards: final report. Menlo Park, CA: SRI International; pro-
ject no. 2094.
Sanders, J. A.; Reinert, R. A. (1982a) Screening azalea cultivars for sensitivity
to nitrogen dioxide, sulfur dioxide, and ozone alone and in mixtures. J.
Am. Soc. Hortic. Sci. 107: 87-90.
Sanders, J. S. ; Reinert, R. A. (1982b) Weight changes of radish and marigold
exposed at three ages to N02, S02) and 03 alone and in mixture. J. Am.
Soc. Hortic. Soc. 107: 726-730.
Santamour, F. S. (1969) Air pollution studies of Plantanus and American elm
seedlings. Plant Dis. Rep. 53: 482-485.
Sawada, I.; Date, N.; Fukuota, S.; lizima, I.; Yoneyama, T.; Oodaira, T.
(1974) On the plant damage due to ozone and PAN at the occurrence of
oxidants with high concentration [English abstract]. Taiki Osen Kenkyu
9: 364.
Seibert, R. J. (1970) Air pollution damage to plants at Longwood Gardens.
Am. Hort. Mag. 49: 67-69.
Shannon, J. G. ; Mulchi, C. L. (1974) Ozone damage to wheat varieties at
anthesis. Crop Sci. 14: 335-337.
Shertz, R. D.; Kender, W. J.; Musselmann, R. C. (1980a) Foliar response and
growth of apple trees following exposure to ozone and sulfur dioxide. J.
Am. Soc. Hortic. Sci. 105: 594-598.
Shertz, R. D. ; Kender, W. D.; Musselman, R. D. (1980b) Effects of ozone and
sulfur dioxide on grapevines. Sci. Horti. (Amsterdam) 13: 37-45.
Shew, B. B. ; Reinert, R. A.; Barker, K. R. (1982) Response of tomatoes to
ozone, sulfur dioxide, and infection by Pratylenchus penetrans. Phyto-
pathology 72: 822-825.
Shriner, D. S.; Cure, W. W. ; Heagle, A. S. ; Heck, W. W.; Johnson, D. W.;
Olson, R. J.; Skelly, J. M. (1982) An analysis of potential agriculture
and forestry impacts of long-range transport air pollutants. Oak Ridge,
TN: Oak Ridge National Laboratory; ORNL report no. 5910.
Sigal, L. L.; Nash, T. H. , III. (1983) Lichen communities on conifers in
southern California mountains: an ecological survey relative to oxidant
air pollution. Ecology 64: 1343-1354.
Sigal, L. L. ; Taylor, 0. C. (1979) Preliminary studies of the gross photo-
synthetic response of lichens to peroxyacetyl nitrate fumigations. Bryo-
logist 82: 564-575.
Ska'rby, L. ; Pell, E. J. (1979) Concentrations of coumestrol and 41 ,7-dihydroxy-
flavone in four alfalfa cultivars after exposure to ozone. J. Environ.
Qual. 8: 285-286.
6-291
-------
Skelly, J. M. (1980) Photochemical oxidant impact on Mediterranean and temperate
forest ecosystems: real and potential effects. In: Miller, P. R. , ed.
Proceedings of the symposium on effects of air pollutants on Mediterranean
and temperate forest ecosystems; June; Riverside, CA. Berkeley, CA: U.S.
Department of Agriculture, Forest Service general technical report
no. PSW-43; pp. 38-51. Available from: NTIS, Springfield, VA; PB81-133720.
Skelly, J. M. ; Croghan, C. F. ; Hayes, E. M. (1977) Oxidant levels in remote
mountainous areas of southwestern Virginia and their effects on native
white pine (Pinus strobus L). In: Dimitriades, B. , ed. International
conference on photochemical oxidant pollution and its control: proceed-
ings, v. II; September 1976; Raleigh, NC. Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Research and Development;
pp. 611-620; EPA report no. EPA-600/3-77-001b. Available from: NTIS,
Springfield, VA; PB-264233.
Skelly, J. M.; Chevone, B. I.; Yang, Y. S. (1982) Effects of ambient concentra-
tions of air pollutants on vegetation indigenous to the Blue Ridge Moun-
tains of Virginia. In: Proceedings of American Water Resources Association
meeting; June; Denver, CO.
Speroni, J. J.; Pell, E. J.; Weissberger, W. C. (1981) Glycoalkaloid levels in
potato tubers and leaves after intermittent plant exposure to ozone. Am.
Potato J. 58: 407-414.
Stan, H.-J.; Schicker, S.; Kassner, H. (1981) Stress ethylene evolution of bean
plants—a parameter indicating ozone pollution. Atmos. Environ.
15: 391-395.
Stan, H-J.; Schicker, S. (1982) Effect of repetitive ozone treatment on bean
plants-stress ethylene production and leaf necrosis. Atmos. Environ.
16: 2267-2270.
Stark, R. W.; Miller, P. R.; Cobb, D. L., Jr.; Wood, D. L.; Parmeter, J. R., Jr.
(1968) Photochemical oxidant injury and bark beetle (Coleoptera: Scolytidae)
infestation of ponderosa pine: I. Incidence of bark beetle infestation in
injured trees. Hilgardia 39: 121-126.
Starkey, T. E. ; Davis, D. D.; Merrill, W. (1976) Symptomatology and suscepti-
bility of ten bean varieties exposed to peroxyacetyl nitrate (PAN). Plant
Dis. Rep. 60: 480-483.
Starkey, T. E. ; Davis, D. D. ; Pell, E. J. ; Merrill, W. (1981) Influence of
peroxyacetyl nitrate (PAN) on water stress in bean plants. HortScience
16: 547-548.
Stephens, E. R. (1967) The formation of molecular oxygen by alkaline hydro-
lyses of peroxyacetyl nitrate. Atmos. Environ. 1: 19-20.
Sung, C. H. ; Chen, H. H. ; Street, P. E.; Yang, Y. L. (1971) Gene effects for
tolerance to weather fleck in tobacco. Nung Yeh Chi Kan 7: 173-181.
6-292
-------
Sung, S. J. S. ; Moore, L. D. (1979) The influence of three herbicides on the
sensitivity of greenhouse-grown flue-cured tobacco (Nicotiana tabacum)
plants to ozone. Weed Sci. 27: 167-173.
Swanson, E. S. ; Thompson, W. W. ; Mudd, J. B. (1973) The effects of ozone on
leaf cell membranes. Can. J. Bot. 51: 1213-1219.
. Taylor, 0. C. (1969) Importance of peroxyacetyl nitrate (PAN) as a phytotoxic
air pollutant. J. Air Pollut. Control Assoc. 19: 347-351.
Taylor, 0. C. (1974) Air pollutant effects influenced by plant-environmental
interaction. In: Dugger, W. M., ed. Air pollution effects on plant growth.
ACS Symp. Ser. 3: 1-7.
Taylor, 0. C. (1984) University of California, Riverside, CA. Personal
communication to D. T. Tingey. U.S. Environmental Protection Agency,
Corvallis, OR. 1984.
Taylor, 0. C.; Maclean, D. C. (1970) Nitrogen oxides and the peroxyacyl nitrates.
In: Jacobson, J. S. ; Hill, A. C., eds. Recognition of air pollution
injury to vegetation: a pictorial atlas. Pittsburgh, PA: Air Pollution
Control Association; pp. E1-E14; informative report no. 1.
Taylor, 0. C. ; Stephens, E. R. ; Darley, E. F. ; Cardiff, E. A. (1960) Effect
of airborne oxidants on leaves of pinto bean and petunia. J. Am. Soc.
Hortic. Sci. 75: 435-444.
Taylor, 0. C.; Dugger, W. M.; Cardiff, E. A.; Darley, E. F. (1961) Interaction
of light and atmospheric photochemical products ("smog") within plants.
Nature (London) 192: 814-816.
Taylor, G. E. , Jr.; Murdy, W. H. (1975) Population differentiation of an
annual plant species, Geranium carolinianum, in response to sulfur
dioxide. Bot. Gaz. (Chicago) 136: 212-215.
Taylor, G. E., Jr.; Mclaughlin, S. B.; Shriner, D. S. (1982a) Effective pollut-
ant dose. In: Unsworth, M. H. ; Ormrod, D. P., eds. Effects of gaseous
air pollution in agriculture and horticulture. London, United Kingdom:
Butterworth Scientific; pp. 458-460.
Taylor, G. E., Jr.; Tingey, D. T.; Ratsch, H. C. (1982b) Ozone flux in Glycine
max (L.) Merr.: sites of regulation and relationship to leaf injury.
Oecologia 53: 179-186.
Taylor, 0. C. ; Temple, P. J.; Thill, A. J. (1983) Growth and yield responses
of selected crops to peroxyacetyl nitrate. HortScience 18: 861-863.
Temple, P. J. (1982) Effects of peroxyacetyl nitrate (PAN) on growth of plants
[Ph.D. dissertation]. Riverside, CA: University of California, Department
of Botany and Plant Sciences. Available from: University Microfilms,
Ann Arbor, MI; publication no. 82-23,389.
Temple, P. J.; Bisessar, S. (1979) Response of white bean to bacterial blight,
ozone, and antioxidant protection in the field. Phytopathology
69: 101-103.
6-293
-------
Temple, P. J. ; Taylor, 0. C. (1983) World-wide ambient measurements of
peroxyacetyl nitrate (PAN) and implications for plant injury. Atmos.
Environ. 17: 1583-1587.
Temple, P. J. ; Taylor, 0. C.; Benoit, L. F. (1985a) Cotton yield responses to
ozone as mediated by soil moisture and evapotranspiration. J. Environ.
Qual. 14: 55-60.
Temple, P. J. ; Taylor, 0. C.; Benoit, L. F. (1985b) Effects of ozone on yield
of two field-grown barley cultivars. Environ. Pollut. 39: 217-225.
Teng, P. S. (1982) A report evaluating the possible use of biological systems
to assess the economic effects of air pollutants on crops and forests.
In: Laurence, J. A., ed. Development of a biological air quality indexing
system: a report to the Minnesota Environmental Quality Board. St. Paul,
MN: Minnesota Environmental Quality Board; SE01848; pp. 257-323.
Teng, P. S. ; Oshima, R. J. (1982) Identification and assessment of losses.
In: Kommedahl, T. H.; Williams, P. H. , eds. Challenging problems in plant
health. St. Paul, MN: American Phytopathological Society.
Thomas, M. D. ; Hendricks, R. H. (1956) Effect of air pollution on plants.
In: Magill, P. L.; Holden, E. R.; Ackley, C. , eds. Air pollution handbook.
New York, NY: McGraw-Hill Book Co.; pp. 9-1 to 9-44.
Thomas, M. D.; Hendricks, R. H.; Hill, G. R. (1950) Sulfur metabolism of plants:
effects of sulfur dioxide on vegetation. Indust. Eng. Chem. 42: 2231-2235.
\
Thompson, C. R.; Kats, G. (1975) Effects of ambient concentrations of peroxy-
acetyl nitrate on navel orange trees. Environ. Sci. Technol. 9: 35-38.
Thompson, C. R.; Taylor, 0. C. (1966) Plastic-covered greenhouses supply
controlled atmospheres to citrus trees. Trans. Amer. Soc. Agric. Eng.
9: 338-342.
Thompson, C. R.; Hensel, E. ; Kats, G. (1969) Effects of photochemical air
pollutants on Zinfandel grapes. HortScience 4: 222-224.
Thompson, C. R. ; Kats, G.; Cameron, J. W. (1976a) Effects of photochemical air
pollutants on growth, yield, and ear characteristics of two sweet corn
hybrids. J. Environ. Qual. 5: 410-412.
Thompson, C. R.; Kats, G.; Pippen, E. L.; Isom, W. H. (1976b) Effect of photo-
chemical air pollution on two varieties of alfalfa. Environ. Sci. Technol.
10: 1237-1241.
Thomson, W. W. (1975) Effects of air pollutants on plant ultrastructure. In:
Mudd, J. B.; Kozlowski, T. T., eds. Responses of plants to air pollution.
New York, NY: Academic Press, Inc.; pp. 179-194.
Thomson, W. W. ; Dugger, W. M., Jr.; Palmer, R. L. (1965) Effects of peroxyacetyl
nitrate on ultrastructure of chloroplasts. Bot. Gaz. (Chicago) 126: 66-72.
6-294
-------
Ting, I. P.; Dugger, W. M., Jr. (1968) Factors affecting ozone sensitivity
and susceptibility of cotton plants. J. Air Pollut. Control Assoc.
18: 810-813.
Tingey, D. T. (1974) Ozone-induced alterations in the metabolite pools and
enzyme activities of plants. In: Dugger, M. , ed. Air pollution effects
on plant growth. Washington, DC: American Chemical Society; pp. 40-57.
(American Chemical Society symposium series 3.)
Tingey, D. T. (1977) Ozone-induced alterations in plant growth and metabolism.
In: Dimitriades, B., ed. International conference on photochemical
oxidant pollution and its control: proceedings, v. II; Research Triangle
Park, NC: U.S. Environmental Protection Agency; pp. 601-609; Public, no.
EPA-600/3-77-001b.
Tingey, D. T. (1980) Stress ethylene production: a measure of plant response
to stress. HortScience 15: 630-633.
Tingey, D. T.; Blum, U. (1973) Effects of ozone on soybean nodules. J. Environ.
Qual. 2: 341-342.
Tingey, D. T.; Hill, A. C. (1967) The occurrence of photochemical phytotoxi-
cants in the Salt Lake Valley. Proc. Utah Acad. Sci. Arts Lett.
55: 387-395.
Tingey, D. T.; Reinert, R. A. (1975) The effect of ozone and sulfur dioxide
singly and in combination on plant growth. Environ. Pollut. 9: 117-125.
Tingey, D. T.; Taylor, G. E., Jr. (1982) Variation in plant response to
ozone: a conceptual model of physiological events. In: Unsworth, M. H.;
Ormrod, D. P., eds. Effects of gaseous air pollution in agriculture and
horticulture. London, United Kingdom: Butterworth Scientific; pp. 113-138.
Tingey, D. T.; Heck, W. W.; Reinert, R. A. (1971a) Effect of low concentrations
of ozone and sulfur dioxide on foliage, growth and yield of radish. J. Am.
Soc. Hortic. Sci. 96: 369-371.
Tingey, D. T. ; Reinert, R. A.; Dunning, J. A.; Heck, W. W. (1971b) Vegetation
injury from the interaction of nitrogen dioxide and sulfur dioxide.
Phytopathology 71: 1506-1511.
Tingey, D. T.; Reinert, R. A.; Carter, H. B. (1972) Soybean cultivars: acute
foliar response to ozone. Crop Sci. 12: 268-270.
Tingey, D. T.; Dunning, J. A.; Jividen, G. M. (1973a) Radish root growth
reduced by acute ozone exposures. Proceedings of the third international
clean air congress; October; Diisseldorf, West Germany. Dusseldorf, West
Germany: VDI-Verlag GmbH; A154-A156.
Tingey, D. T.; Reinert, R. A.; Dunning, J. A.; Heck, W. W. (1973b) Foliar
injury responses of eleven plant species to ozone/sulfur dioxide mixtures.
Atmos. Environ. 7: 201-208.
6-295
-------
Tingey, D. T. ; Reinert, R. A.; Wickliff, C. ; Heck W. W. (1973c). Chronic
ozone or sulfur dioxide exposures, or both, affect the early vegetative
growth of soybean. Can. J. Plant Sci. 53: 875-879.
Tingey, D. T.; Fites, R. C. ; Wickliff, C. (1975) Activity changes in selected
enzymes from soybean leaves following ozone exposure. Physiol. Plant.
33: 316-320.
Tingey, D. T.; Wilhour, R. G.; Standley, E. (1976a) The effect of chronic ozone
exposures on the metabolite content of ponderosa pine seedlings. For. Sci.
22: 234-241.
Tingey, D. T.; Standley, C.; Field, R. W. (1976b) Stress ethylene evolution: a
measure of ozone effects on plants. Atmos. Environ. 10: 969-974.
Tingey, D. T.; Manning, M.; Grothaus, L. C.; Burns, W. F. (1979) The influence
of light and temperature on isoprene emission rates from live oak. Physiol.
Plant. 47: 112-118.
Tingey, D. T. ; Thutt, G. L.; Gumpertz, M. L.; Hogsett, W. E. (1982) Plant
water status influences ozone sensitivity of bean plants. Agric. Environ.
7: 243-254.
Todd, G. W. (1958) Effects of low concentrations of ozone on enzymes catalase,
peroxidase, papain and urease. Physiol. Plant. 11: 457-463.
Toivonen, P. M. A. j'Hofstra, G.; Wukasch, R. T. (1982) Assessment of yield
losses in white bean due to ozone using the antioxidant EDU. Can.
J. Plant Path. 4: 381-386.
Tonneijck, A. E. G. (1984) Effects of peroxyacetyl nitrate (PAN) and ozone on
some plant species. In: Proceedings of the OECD workshop on ozone;
Goteborg, Sweden. Goteborg, Sweden: Swedish Environ. Research Institute.
Treshow, M. (1980a) Interactions of air pollutants and plant disease. In:
Miller, P. R., ed. Proceedings of the symposium on effects of air pollu-
tants on Mediterranean and temperate forest ecosystems; June; Riverside,
CA. Berkeley, CA: U.S. Department of Agriculture, Forest Service general
technical report no. PSW-43; pp. 103-109. Available from: NTIS, Springfield,
VA; PB81-133720.
Treshow, M. (1980b) Pollution effects on plant distribution. Environ. Conserv.
7: 279-286.
Trevathan, L. E. ; Moore, L. D. (1976) Calcium nutrition in relation to ozone
damage of flue-cured tobacco. Tob. Sci. 82: 96-97.
Trevathan, L. E. ; Moore, L. D. ; Orcutt, D. M. (1979) Symptom expression and
free sterol and fatty acid composition of flue-cured tobacco plants
exposed to ozone. Phytopathology 69: 582-585.
6-296
-------
Troiano, J.; Colavito, L.; Heller, L.; McCune, D.C.; Jacobson, J. S. (1983)
Effects of acidity of simulated rain and its joint action with ambient
ozone on measures of biomass and yield in soybean. Environmental Exp.
Botany 23: 113-119.
Uhring, J. (1978) Leaf anatomy of petunia in relation to pollution damage. J.
Am. Soc. Hortic. Sci. 103: 23-27.
Unsworth, M. H. ; Heagle, A. S. ; Heck, W. W. (1984a) Gas exchange in open-top
field chambers-I. Measurement and analysis of atmospheric resistances to
gas exchange. Atmos. Environ. 18:373-380.
Unsworth, M.H.; Heagle, A.S.; Heck, W. W. (1984b) Gas exchange in open-top
field chambers-II. Resistances to ozone uptake by soybeans. Atmos. Environ.
18:381-385.
U.S. Department of Agriculture. (1965) Losses in agriculture. Washington, DC:
U.S. Department of Agriculture. (Agriculture handbook no. 291.)
U.S. Environmental Protection Agency. (1976) Diagnosing vegetation injury
caused by air pollution. In: Hicks, D. R. , ed. Research Triangle Park,
NC: Air Pollution Training Institute; publication no. EPA-405/3-78-005.
U.S. Environmental Protection Agency. (1978) Air quality criteria for ozone
and other photochemical oxidants. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Environmental Criteria and Assessment
Office; EPA report no. EPA-600/8-78-004. Available from: NTIS, Spring-
field, VA; PB80-124753.
U.S. Environmental Protection Agency. (1982) Air quality criteria for oxides
of nitrogen. Research Triangle Park, NC: U.S. Environmental Protection
Agency, Environmental Criteria and Assessment Office; EPA report no.
EPA-600/8-82-026F. Available from: NTIS, Springfield, VA; PB83-163337.
U.S. Environmental Protection Agency. (1983) Air quality data - 1982 annual
statistics: including summaries with reference to standards. Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards; EPA report no. EPA-450/4-83-016. Available
from: NTIS, Springfield, VA; PB83-239830.
Usher, R. W.; Williams, W. T. (1982) Air pollution toxicity to eastern white
pine in Indiana and Wisconsin. Plant Dis. 66: 199-204.
Vargo, R. H. ; Pell, E. J. ; Smith, S. H. (1978) Induced resistance to ozone
injury of soybean by tobacco ringspot virus. Phytopathology 68: 15-17.
Walker, J. T. ; Barlow, T. C. (1974) Response of indicator plants to 03 levels
in Georgia. Phytopathology 64: 1122-1127.
Walmsley, L.; Ashmore, M. L.; Bell, J. N. B. (1980) Adaptation of radish Raphanus
sativus L. in response to continuous exposure to ozone. Environ. Pollut.
23: 165-177.
6-297
-------
Weber, D. E. ; Reinert, R. A.; Barker, K. R. (1979) Ozone and sulfur dioxide
effects on reproduction and host-parasite relationships of selected
plant-parasitic nematodes. Phytopathology 69: 624-628.
Weidensaul, T. C. (1980) N-[2-(2-oxo-l-imidazolidinyl)ethyl]-N'-phenylurea as a
protectant against ozone injury to laboratory fumigated pinto bean plants.
Phytopathology 70: 42-45.
Weidensaul, T. C. ; Darling, S. L. (1979) Effects of ozone and sulfur dioxide
on the host-pathogen relationship of Scotch pine and Scirrhia acicola.
Phytopathology 69: 939-941.
Went, F. W. (1955) Air pollution. Sci. Am. 192: 62-70.
Wilhour, R. G. ; Neely, G. E. (1977) Growth response of conifer seedlings to
low ozone concentrations. In: Dimitriades, B. , ed. International
conference on photochemical oxidant pollution and its control: proceedings,
vol. II; January; Research Triangle Park, NC. Research Triangle Park, NC:
U.S. Environmental Protection Agency; pp. 635-645; EPA report no.
EPA-600/3-77-001b. Available from: NTIS, Springfield, VA; PB-264233.
Williams, W. T.; Brady, M.; Willison, S. C. (1977) Air pollution damage to the
forests of the Sierra Nevada mountains of California. J. Air Pollut.
Cont. Assoc. 27: 230-234.
Willig, R. D. (1976) Consumer surplus without apology. Am. Econ. Rev. 66: 589-597.
Winner, W. E. ; Mooney, H. A. (1980) Ecology of S02 resistance: I. Effects of
fumigations on gas exchange of deciduous and evergreen shrubs. Oecologia
44: 290-295.
Wood, F. A.; Drummond, D. B. (1974) Response of eight cultivars of chrysan-
themum to peroxyacetyl nitrate. Phytopathology 64: 897-898.
Wood, F. A.; Pennypacker, S. P. (1975) Evaluation of the effects of air pollu-
tion on vegetation in the Mt. Storm, West Virginia-Oakland, Maryland
area. Presented at: the 68th Annual Meeting of the Air Pollution Control
Association; June; Boston, MA. Pittsburgh, PA: Air Pollution Control
Association; paper 75-21.1.
Wukasch, R. T. ; Hofstra, G. (1977a) Ozone and Botrytis interactions in onion-
leaf dieback: open-top chamber studies. Phytopathology 67: 1080-1084.
Wukasch, R. T.; Hofstra, G. (1977b) Ozone and Botrytis spp. interaction in
onion-leaf dieback: field studies. J. Am. Soc. Hortic. Sci. 102: 543-546.
Yang, Y.-S.; Skelly, J. M. ; Chevone, B. I.; Birch, J. B. (1983) Effects of
long-term ozone exposure on photosynthesis and dark respiration of eastern
white pine. Environ. Sci. Technol. 17: 371-373.
Yarwood, C. E. ; Middleton, J. T. (1954) Smog injury and rust infection. Plant
Physiol. 29: 393-395.
Youngner, V. E. ; Nudge, F. J. (1980) Air pollution oxidant effects on cool-
season and warm-season turfgrasses. Agron. J. 72: 169-170.
6-298
-------
APPENDIX 6A
COLLOQUIAL AND LATIN NAMES OF PLANTS DISCUSSED IN THE CHAPTER
\
6A-1
-------
APPENDIX 6A. COLLOQUIAL AND LATIN NAMES OF PLANTS DISCUSSED IN THE CHAPTER
Colloquial Name
Latin name
Alfalfa
Ash
Green
White
Aspen
Bigtooth
Azalea
Delaware valley white
Hinodegiri
Korean
Barley
Bean
var. - French
Green snapbean
Navy
Pinto
Red kidney
Snapbean
White
Bean
Broad
Beet
Garden
Sugar
Begonia
Begonia
Birch
White
Yellow
Cabbage
Carnation
Carrot
Medicago sativa L.
Fraxinus pennsylvanica Marsh.
Fraxinus americana L.
Populus grandidentata Michx.
Rhododendron mucronatum Don.
Rhododendron obtusum Planch.
Rhododendron poukhanensis Leveille
Hordeum vulgare L.
Phaseolus vulgaris L.
Vicia faba L.
Beta vulgaris L.
Begonia semperflorens Link and Otto
Begonia X hi email's Fotsch.
Betula papyrifera Marsh.
Betula alleghaniensis Britton
Brassica oleracea capitata L.
Dianthus caryophyllus L.
Daucus carota var. sativa DC.
6A-2
-------
APPENDIX 6A. (continued)
Colloquial Name
Latin name
Chard
Swiss
Cherry
Black
Chrysanthemum
Citrus
Clover
Landino
Coleus
Corn
Field
Sweet
Cotoneaster
Cotton
Cottonwood
Eastern
Elder
Black
Elm
Chinese
Endive
Fir
Douglas
Geranium
Grape
Grape
Gum
Black
Hemlock
Eastern
Beta vulgaris var. cicla L.
Prunus serotina Ehrh.
Chrysanthemum morifolium Ramat.
Citrus sp.
Trifolium repens L.
Coleus blumei Benth.
Zea mays L.
Cotoneaster divaricata Rehd.
Gossypium hirsutum L.
Populus deltoides Bartr.
Sambucus nigra L.
Ulmus parvifolia Jacq.
Cichorium endiva L.
Pseudotsuga menziesii (Mirb.) Franco.
Pelargonium hortorum Bailey
Vitis vinifera L.
Vitis labrusca L.
Nyssa sylvatica Marsh.
Tsuga canadensis (L.) Carr.
6A-3
-------
APPENDIX 6A. (continued)
Colloquial Name
Latin name
Holly
American
Japanese
Larch
Japanese
Lettuce
var.-Cos (Romaine)
Linden
American
Locust
Black
Maple
Red
Sugar
Marigold
Milkweed
Morning glory
Mountain laurel
Muskmelon
Mustard
Nettle (little leaf)
Oak
Black
California black
Wi 11ow
Oat
Onion
Australian
Pasture grass
Australian
Grasslands
Victorian
Ilex opaca Ait.
Ilex crenata Thunb.
Larix leptolepis Gord.
Lactuca sativa L.
Tilia americana L.
Robim'a pseudoacacia L.
Acer rubrum L.
Acer saccharum L.
Tagetes erecta L.
Asclepias syriaca L.
Ipomea nil Roth.
Kalmia latifolia L.
Cucumis melo L.
Brassica migra (J.) Koch
Urtica urens L.
Quercus velutina Lam.
Quercus kelloggii Newb.
Quercus phellos L.
Avena sativa L.
Alii urn cepa L.
Phalaris aquatica
Dactyl is glomerata L.
Lolium perenne L.
6A-4
-------
APPENDIX 6A. (continued)
Colloquial Name
Latin name
Peanut
Pepper
Petunia
Pine
Austrian
Eastern white
Jeffrey
Loblolly
Lodgepole
Monterey
Pitch
Ponderosa
Scotch
Shore
Slash
Sugar
Table mountain
Virginia
Western white
Poinsettia
Poplar
Hybrid poplar
Hybrid poplar
Hybrid poplar
Potato
Privet
Amur
Radish
Snapdragon
Soybean
Spinach
Spruce
Sitka
White
Arachis hypogea L.
Capsicum annuum L.
Petunia hybrida Vilm.
Pinus nigra Arnold
Pinus strobus L.
and Balf.
Pinus jeffreyi Grev.
Pinus taeda L.
Pinus contorta var. murrayana (Balf.)
Critch
Pinus radiata D. Don
Pinus rigida Mill.
Pinus ponderosa Laws.
Pinus sylvestris L.
Pinus contorta var. contorta
ex Laud
Pinus elliotti Englem. ex Vasey
Pinus lambertiana Dougl.
Pinus pungens Lamb.
Pinus virgimana Mill.
Pinus monticola Dougl.
Dougl.
Euphorbia pulcherrima Wildenow
Populus sp.
Populus X euramericana
Populus maximowiczn X trichocarpa
Populus deltoides X trichocarpa
Solanum tuberosum L.
Ligustrum amurense Carr.
Raphanus sativus L.
Antirrhinum majus L.
Glycine max (L.) Merr.
Spinacia oleracea L.
Picea sitchensis (Bong.) Carr.
Picea glauca (M'oench) Voss
6A-5
-------
APPENDIX 6A. (continued)
Colloquial Name
Latin name
Strawberry
Sunflower
Sweetgum
Sweet mock-orange
Sycamore
American
Tomato
Tree-of-heaven
Turfgrass
Annual bluegrass
Bermudagrass
Colonial bentgrass
Creeping bentgrass
Kentucky bluegrass
Red fescue
Red top
Ryegrass
Tall fescue
Zoysiagrass
Turnip
Viburnun
Tea viburnun
Linden viburnum
Walnut
Black
Wild strawberry
Wheat
Winter
Yellow poplar (Tulip poplar)
Yew
Fragaria chiloensis var. ananassa
Bailey
Helianthus anuus L.
Liquidambar styraciflua L.
Philadelphus coronarius L.
Platanus occidental is L.
Lycopersicon esculentum Mill.
Ailanthus altissima Swingle
Poa annua L.
Cynodon dactyl on L., Pers.
Agrostis tenuis Sibth.
Agrostis palustris Huds.
Poa pratensis L.
Festuca rubra Gaud.
Agrostis alba L.
Loliurn perenne L.
Festuca arundinaceae Schreb.
Zoysia japonica Steud.
Brassica rapa L.
Viburnum setigerum Hance
Viburnum dilatatum Thunb.
Juglans nigra L.
Fragaria virginiana Duchesne
Triticum aestivum L.
Liriodendron tulipifera L.
Taxus X media Rehd.
6A-6
-------
APPENDIX 6B
SPECIES THAT HAVE BEEN EXPOSED TO OZONE TO DETERMINE DIFFERENTIAL RESPONSES OF
GERMPLASM TO PHOTOCHEMICAL PRODUCTS
6B-1
-------
APPENDIX 6B. SPECIES THAT HAVE BEEN EXPOSED TO OZONE TO DETERMINE
DIFFERENTIAL RESPONSES OF GERMPLASM TO PHOTOCHEMICAL PRODUCTS
Species
References
Alfalfa
Azalea
Bean
Begonia
Chrysanthemum
Cucumber
Eggplant
English holly
Forage legumes
Grape
Lettuce
Morning glory
Oat
Petunia
Pine
Poplar
Poinsettia
Potato
Howell et al., 1971
Gesalman and Davis, 1978
Butler and Tibbitts, 1979a,b
Davis and Kress, 1974
Meiners and Heggestad, 1979
Heggestad et al. , 1980
Reinert and Nelson, 1979
Adedipe, 1972
Wood and Drummond, 1974
Brennan and Leone, 1972
Ormrod et al., 1971
Rajput and Ormrod, 1976
Brennan and Leone, 1970
Brennan et al., 1969
Richards et al., 1958
Reinert et al., 1972
Nakamura and Matsunaka, 1974
.Brennan et al. , 1964
Feder et al., 1969
Cathey and Heggestad, 1972
Elkiey and Ormrod, 1981
Berry, 1971
Houston, 1974
Karnosky, 1977
Manning et al., 1972
Heggestad, 1973
DeVos et al., 1982
6B-2
-------
7. EFFECTS OF OZONE ON NATURAL ECOSYSTEMS AND THEIR COMPONENTS
7.1 INTRODUCTION
The responses of individual species and subspecies of agricultural plants
to ozone (03) and peroxyacetyl nitrate (PAN) exposure were discussed in the
preceding chapter. In addition, the responses of trees and other native
vegetation to ozone were briefly discussed. The present chapter discusses the
effects of 03 stress on simple and complex plant communities to illustrate
that such effects, because of the interconnections and relationships among
ecosystem components, can produce perturbations in ecosystems. Stresses
placed on biota and the ecosystems of which they are a part can produce
changes that are long-lasting and that may be irreversible. Ecosystem
responses to PAN are not discussed in this chapter since data dealing with the
effects of PAN on ecosystems are virtually nonexistent, since trees and other
woody plants appear to be resistant to PAN (Chapter 6), and since the occur-
rence of PAN at phytotoxic concentrations is believed to be a regional rather
than a national problem.
Material in this chapter is organized into seven main sections that are
presented in the following sequence: (1) overview and description of ecosystems;
(2) description of responses to stress that are characteristic of ecosystems;
(3) discussion of the effects of ozone on primary production in terrestrial
ecosystems, including effects on the growth of trees and mechanisms involved
in those effects; (4) discussion of the effects of ozone on other ecosystem
components and their interactions; (5) discussion of the effects of ozone on
specific forest ecosystems; (6) discussion of the effects of ozone on other
ecosystems; and (7) a brief discussion on the economic valuation of ecosystems.
7.2 CHARACTERISTICS OF ECOSYSTEMS
An ecosystem is an integrated unit of nature consisting of interacting
populations of plants and animals in a given area (the community) whose survival
depends on the maintenance of biotic (living) and abiotic (nonliving) func-
tions and interrelationships. The biotic components of ecosystems are: (1)
producers, which are principally green plants that capture the energy of the
sun through photosynthesis; (2) consumers, which utilize as their energy
7-1
-------
source the food produced and stored by the producers; and (3) decomposers,
which obtain their energy by breaking down and converting dead organic matter
into inorganic compounds and which release carbon dioxide to the atmosphere.
The abiotic components include: air, water, the soil matrix, and inorganic
substances. Temperature, radiation, barometric pressure, and other climatic
factors are, along with pollution, additional abiotic factors affecting
ecosystems (Billings, 1978; Odum, 1971; Smith, R., 1980).
An ecosystem usually has definable limits within which the integrated
functions of energy flow, nutrient'cycling, and water flux are maintained
(Odum, 1969; Odum, 1971; Jordan and Medina, 1977). Some flow of energy and
materials occurs, however, between adjacent ecosystems. Ecosystems are capable
of responding to changes in the movement of energy and materials from adjacent
environments as well as to changes in their own environment (Cox and Atkins,
1979). Ecosystems receive gases, nutrients, and the energy of the sun from
their environment and utilize these"; and, in turn, make their own contribu-
tions to the environment. Energy flows through the system unidirectionally
and is dissipated into the atmosphere, while water, gases, and nutrients are
usually recycled and fed back into the system. When materials are not returned
through recycling, they must be obtained in another way. Plant and animal
populations within the system represent the fundamental units through which
the system functions (Smith, R. 1980); that is, through which energy is ex-
changed and nutrients are cycled (Smith, R., 1980; Billings, 1978; Odum,
1971). Any action that changes the flow of nutrients, energy, or both will
cause a change in the relationships that exist between the environment and
living organisms, as well as in the relationships among the organisms them-
selves.
The,agricultural ecosystems discussed in Chapter 6 and the natural eco-
systems discussed in this chapter possess the same basic functional components,
require energy flow and mineral cycling for maintenance, and are subject to
the dominating influences of climate and substrate. Natural ecosystems range
from simple systems with few species to complex systems with many species.
Their populations also vary in genetic composition, age, and species diversity.
They are self-regulating and self-perpetuating. Agroecosystems, on the other
hand, are monocultures of similar genetic and age composition, manipulated to
maximize productivity; and they are unable to maintain themselves without the
addition of nutrients, water, and human effort.
7-2
-------
The differences in structure of natural and agroecosystems are signifi-
cant in the context of responses to oxidant and other pollution stress. While
the manipulation and maintenance of agroecosystems require human effort, the
amenability of such monocultures to management means that the effects of air
pollution stress can be partially overcome by such practices as selection of
resistant cultivars, irrigation, and use of fertilizers. Natural ecosystems,
on the other hand, generally are not amenable to management because of their
species diversity and complexity. Natural ecosystems tend to respond more
slowly than agroecosystems to perturbations such as air pollution; once per-
turbed, however, they may lose their capability for self-repair (Cox and
Atkins, 1979).
The subtle and indirect effects of pollutant dosages on individual species
can set the stage for changes in community structure that may possibly have
irreversible consequences (Guderian and Kueppers, 1980). Increasing pollutant
stress provides a selective force that favors some genotypes, suppresses other:;,
and eliminates those species that lack sufficient genetic diversity to survive,
Thus, the occurrence and distribution of plants are influenced; and community
composition and species interactions are changed such that the basic structure
of the ecosystem is ultimately changed (Treshow, 1980). This succession may
take years, decades, or longer, depending on the pollutant concentration and
dose and on the species involved.
7.3 CHARACTERISTIC RESPONSES OF ECOSYSTEMS TO STRESS
Ecosystems respond to ozone, as well as to other stresses, through the
responses of the populations of organisms that compose them. In the responses
of ecosystems to stress three main levels of interaction are involved: (1)
between the individual and the environment, (2) between the population and its
environment, and (3) between the biological community and its environment
(Billings, 1978). Disturbances may have a positive effect or may produce both
positive and negative responses. Responses at the ecosystem level are more
diffuse and of longer duration than responses at the population level. Acute
stresses that are followed by rapid recovery and return to an unstressed state
may have a different effect on the ecosystem than chronic stresses that continue
for some time. Stress at the community level requires the diversion of energy
from growth and reproduction to maintenance. Thus, biomass accumulation tends
7-3
-------
to decrease as organisms attempt to cope with the disturbance. Decreased
cycling and increased nutrient turnover frequently appear. Disturbance favors
the development of communities dominated by small-bodied, rapidly reproducing
species; and succession reverts to an earlier stage (Odum, 1985).
Studies around strong point sources of air pollution and radiation, along
with results from laboratory and field experiments, as summarized by Smith
(1981), suggest that ecosystems, especially forest ecosystems, respond to
increasing pollutant stress in a predictable pattern that may be thought of as
a continuum of responses (Bormann, 1985). The sequence of responses outlined
by Bormann is given in modified form in Table 7-1 to assist in understanding
TABLE 7-1. CONTINUUM OF CHARACTERISTIC ECOSYSTEM
RESPONSES TO POLLUTANT STRESS
Phase Response characteristics
0 No response occurs. Manmade pollutants are absent or
constitute insignificant stress. Plant growth occurs
under natural conditions.
I Ecosystems serve as sinks for pollutants. Species
and/or ecosystem functions are relatively unaffected.
Self-repair occurs.
II Sensitive species or individuals are subtly and
adversely affected. A reduction in photosynthesis,
a change in reproductive capacity, or a change in
predisposition to insect or fungus attack may occur.
Ill Decline occurs in the populations with sensitive
species; some individuals will be lost. Their ef-
fectiveness as functional members of the ecosystem
diminishes. Ultimately, species could be lost from
the system.
IV Large plants, trees, and shrubs of all species die.
The basic structure of the forest ecosystem is changed.
Biotic regulation is affected as forest layers are
peeled off: first trees, tall shrubs, and, under
the most severe conditions, short shrubs and herbs.
The ecosystem is dominated by weedy species not
previously present and by small scattered shrubs
and herbs.
V The ecosystem collapses. The loss of species and
changes in ecosystem structure, nutrients, and soil
so damage the system that self-repair is impossible.
Source: Adapted from Bormann (1985).
7-4
-------
the discussion that follows. While the effects on individual organisms begin
at the molecular and physiological levels, as amply demonstrated for agricul-
tural and other species in Chapter 6, the responses of ecosystems may be
thought of as beginning at the organismal (individual) level and proceeding to
the ecosystem level, as shown in Table 7-1.
7.4 EFFECTS OF OZONE ON PRIMARY PRODUCTION IN TERRESTRIAL ECOSYSTEMS
This section discusses the responses of individual plant species and
other ecosystem components to explain how the disruption by ozone of plant
processes at the organismal level can ultimately change structural patterns
and such functional processes as energy flow, nutrient cycling, and biotic
relationships in ecosystems. Of the ecosystems exposed to ozone and potential-
ly affected by ozone, forest ecosystems are the largest and economically most
important. Therefore, the respective components and processes of forest
ecosystems have been studied more than those of other ecosystems and will be
emphasized in the following discussion. Studies in which multiple trophic
levels and interrelationships have been examined are presented in subsequent
sections (Sections 7.5 and 7.6).
7.4.1 Effects of Ozone on Growth of Producers
Among the most important potential effects of ozone on terrestrial
ecosystems is the reduction of primary production. In forest ecosystems,
primary production is the addition of new organic matter to the ecosystem via
photosynthesis in producers, i.e., trees and other green plants. Productivity
is the most fundamental characteristic of an ecosystem. All the biological
activity of a community depends on the energy from gross primary production.
Forest productivity is higher than that of other ecosystems, and net produc-
_2 -i -9 -1
tivity of 1200 dry g m yr (2.2 Ib yd yr ) for trees and shrubs combined
is quite typical for temperate forests (Whittaker, 1965). Productivity is
highly dependent on system age and environmental parameters, the most important
of which are nutrient and water availability and temperature. Air quality
also influences forest production in certain environments.
In forest ecosystems, tree populations play a critical role. As producers,
trees influence the structure (species composition and trophic relationships)
and energy flow and nutrient cycling of forest ecosystems (Ehrlich and Mooney.,
7-5
-------
1983). Although they are woody perennial plants, trees basically respond to
ozone exposure in the same manner as agricultural crop species, which are
chiefly herbaceous annuals. The same plant processes are affected (Chapter 6).
Perennial plants, however, because they live longer, must cope with both
short- and long-term stresses, the effects of which can be cumulative, lasting
over the years, or can be delayed, not becoming apparent for many years.
Likewise, effects can possibly be mitigated through short- or long-term recovery
or replacements. These stresses can increase or decrease with the age of the
forest stand. They can act independently, additively, synergistically, or
antagonistically and can occur simultaneously or sequentially (Cowling, 1985).
Ozone can be a predisposing stress that makes trees more susceptible to other
stresses, such as low temperatures, insects, and fungi. A discussion of the
effects of ozone on the growth of trees, because of their critical role, is
the requisite first step in explaining the responses of forest ecosystems to
ozone.
7.4.1.1 Controlled Studies on Growth of Trees. Data were presented in Chap-
ter 6 on the concentrations and durations of exposure to ozone shown to produce
reductions in growth of trees under controlled conditions. For example, as
discussed in greater detail in Chapter 6, significant suppression in growth in
height was observed by Kress and Skelly (1982) in seedlings of trees exposed
to 03 for 6 hr/day for 28 days: loblolly pine (Pinus taeda L.) (0.05 ppm), 18
percent; pitch pine (P. rigida Mill.) (0.10 ppm), 13 percent; American sycamore
(Platanus occidental is L.) (0.05 ppm), 9 percent; sweetgum (Liquidambar
styraciflua L.) (0.10 ppm), 29 percent; green ash (Fraxinus pennsylvanica
Marsh.) (0.10 ppm), 24 percent; willow oak (Quercus phellos L.) and sugar
maple (Acer saccharum L.) (both at 0.15 ppm), 19 and 25 percent, respectively.
Other investigators have reported similar results for other tree species
exposed to ozone under other regimes (e.g., Dochinger and Townsend, 1979;
Mooi, 1980; Patton, 1981; Kress et al., 1982). On the other hand, yellow
poplar (Liriodendron tulipifera L.) and white ash (Ł. americana L.) exhibited
significant growth stimulation, as measured by dry weight, when exposed to
0.05 ppm ozone (Kress and Skelly, 1982). In most instances, reductions in
growth from exposure to ozone were not accompanied by foliar injury. Sweet
gum was an exception.
Hogsett et al. (1985), using exposures that simulated ambient conditions,
noted a reduction in growth in height, diameter, and root systems in two
7-6
-------
varieties of slash pine seedlings receiving chronic 0- exposure. The seedlings
were exposed to one of three regimes: (1) charcoal-filtered air; (2) an
exposure profile with a daily 1-hr maximum of 0.126 ppm at around 2 p.m., a
7-hr (9 a.m. to 4 p.m.) seasonal mean of 0.104 ppm, and an integrated exposure
of 155 ppm-hr (sum of hourly ppm > 0); or (3) a similar ozone exposure profile
but with a daily 1-hr maximum of 0.094 ppm, a 7-hr seasonal mean of 0.076 ppm,
and an integrated exposure of 122 ppm-hr. Both varieties of slash pine exhi-
bited an increasing reduction in growth with increasing 0, concentration. A
significant reduction (p<0.001) in stem diameter occurred by day 112 for both
0- treatments: 24 percent less than that of controls for 'ellottii' and 30
percent less for 'densa' at the lowest 0- exposure; and 40 percent and 50
percent below control plants for 'ellottii1 and 'densa,1 respectively, at the
highest 0- exposure. Both 0- exposures also caused significant reductions in
growth in height (p <0.001). The most" pronounced change was observed in the
growth of roots, which in 'elliottii1 was reduced 33 percent by day 21 at an
integrated exposure of 29 ppm-hr, and 27 percent by day 56 with an exposure of
63 ppm-hr.
7.4.1.2 Field Studies on Effects of Ozone on Growth of Trees in Natural Habitats.
Studies of the effects of ozone on the growth of trees in their natural habitats
have centered on several major forest ecosystems. While the consequences of
growth effects in forest ecosystems are examined in Section 7.5, data on the
effects of ozone on the growth of individual tree species in these ecosystems
are briefly summarized here.
Mann et'al. (1980) found injury to needles and decreased growth in white
pine grown in a plantation on the Cumberland Plateau (near Oak Ridge, TN).
These effects were associated by the. authors with episodes of ozone at 1-hour-
average concentrations >0.08 ppm. Levels of S09 and NO were below 0.1 and
^ /\
0.2 ppm, respectively, throughout the growing season.
In a subseofuent study of the same trees, Mclaughlin et al. (1982) found
reduced annual radial growth, which they also.attributed, to high concentrations
of ozone. McLaughlin et al. (1982) divided trees into three sensitivity
classes on the basis of needle, color and length and duration of retention. A
steady decline in annual ring increment of sensitive white pines was observed
during the years 1962 through 1979. Reductions of 70 percent in average annual
growth and 90 percent in average bole growth of sensitive trees, compared to
the growth of tolerant and intermediate trees, were noted. Tolerant trees
7-7
-------
showed a consistently higher growth rate of 5 to 15 percent (p = 0.005) than
intermediate trees for the 1960 to 1968 interval, similar growth from 1969
through 1975, and reduced growth of 5 to 15 percent (but significant only at p
= 0.10) for the period 1976 through 1979 compared to trees of intermediate
sensitivity. Needles of ozone-sensitive trees were 15 to 45 percent shorter
than those of either of the other classes. Decline was attributed primarily
to chronic exposure to 0~, which frequently occurred at phytotoxic concentra-
tions in the area. For the years 1975 through 1979 the incidence rates for
hourly concentrations > 0.08 ppm were: 1976, 190 hr; 1977, >339 hrs; 1978,
190 hr; 1979, 125 hr. Maximum 1-hour concentrations ranged from 0.12 to 0.2
ppm during this period. The pollutants SOp and fluoride have been measured in
the area, but the premature loss of needles and occasional tip necrosis of
needles of the current year are manifestations usually associated with 0, and
no cause-and-effect relationship with S0? is indicated by the available data.
Benoit et al. (1982) studied the growth in annual rings of native eastern
white pines of reproducing age to evaluate the possible effects of oxidant air
pollution on the long-term growth of forest species in a region of the Blue
Ridge Mountains of Virginia extending from the northern end of Skyline Drive
in Shenandoah National Park to the southernmost portion of the Blue Ridge
Parkway lying in Virginia. The three white pines in each study plot were
classified as sensitive, intermediate, or tolerant, based on a foliar rating
scale that incorporated needle length, needle retention by number of years,
and the presence of typical 0~ symptoms on needles. The mean ages of ozone-
tolerant, intermediate, and ozone-sensitive tree classes were 53, 52, and 56
years, respectively. From 1955 to 1978, growth in mean annual radial increment
(tree ring growth) was 25 percent and 15 percent less than that of tolerant
trees for sensitive and intermediate trees, respectively. Only the 25 percent
decrease for ozone-sensitive trees, however, was significant (p = 0.01) (Table
7-2). Smaller mean increments in the last 10 years compared to the previous
24 years indicated a trend toward decline in overall growth rates in all
classes of trees. A comparison of growth from 1974 to 1978 with that from
1955 to 1959 showed decreases of 26, 37, and 51 percent for tolerant, inter-
mediate, and sensitive trees, respectively. The significant reduction in
radial growth of 0,,-sensitive white pines was assumed to be associated with
cumulative stress resulting from the reduced photosynthetic capacity of oxidant-
injured trees. At the time of this study, tree ring standardization methods
7-8
-------
TABLE 7-2. ANNUAL MEAN RADIAL GROWTH INCREMENT, 1955 THROUGH 1978,'
FOR THREE OZONE SENSITIVITY CLASSES OF NATIVE EASTERN WHITE PINES
(Pinus strobus L.) GROWING IN TEN PLOTS OF THREE TREES
EACH IN THE BLUE RIDGE MOUNTAINS IN VIRGINIA
(mm)
Plot
1
2
3
4
5
6
7
8
9
10
Mean
Tolerant trees3
4.59
3.52
8.19
4.80
5.94
4.64
2.85
3.91
3.32
1.67
4.34 Ab
Intermediate trees
2.13
2.12
6.34
3.75
6.53
3.76
2.75
4.52
2.04
2.98
3.69 AB
Sensitive trees
3.08
2.86
6.89
2.62
5.73
2.62
1.51
1.96
2.61
1.46
3.10 B
a, .. J.JT 4.- j- ••
White pines rated tolerant, intermediate, or sensitive to 03 based on
foliar symptoms.
Sensitivity classes with the same letter are not significantly different at
p = 0.01 based on Duncan's multiple range test.
Source: Benoit et al. (1982).
had not been developed. In addition, the sample size of three trees per plot,
in ten plots, was small.
The authors (Benoit et al., 1982) assumed that the reduction in radial
growth was caused by 0- because no significant changes had occurred in seasonal
precipitation between the 1955-1963 and 1963-1978 periods. Increasing 0^
concentrations, therefore, could potentially account for growth reductions,
especially since for the later period there was a negative correlation between
precipitation and radial growth.
Sulfur dioxide concentrations were too low to have any vegetational
effects. The monitoring of 0, (Duchelle et al., 1983) indicated the presence
of monthly average concentrations of 0.05 to 0.06 ppm on a recurring basis in
the study area, with episodic peaks (1-hour) frequently in excess of 0.09 ppm.
Episodes in the Blue Ridge Mountains lasting from 1 to 5 consecutive days have
been reported by Skelly et al. (1984) (Table 7-3). Hayes and Skelly (1977)
7-9
-------
TABLE 7-3. PEAK HOURLY OZONE CONCENTRATIONS IN EPISODES RECORDED AT THREE
MONITORING SITES IN WESTERN VIRGINIA, SPRING AND SUMMER, 1979 THROUGH 1982
(ppm)
1979,
1979,
1979,
1980,
1980,
1980,
1980,
1981,
1981,
1981,
1981,
1982,
1982,
1982,
1982,
1982,
1982,
1982,
Date
April 11-12
June 5-6
September 12
May 29-30
June 13-15
July 14-15
July 31
May 24
June 22-23
June 29-30
July 8-11
May 11-13
May 16-17
July 24-26
July 27-28
August 3-4
August 19-20
October 1-3
Big Meadows
0.
0.
0.
0.
0.
0.
0.
0.
N.
N.
0.
0.
0.
0.
0.
0.
0.
0.
100
082
095
100
093
089
102
084
A Cl
A!*
096
125
090
110
090
090
095
100
1-hr 03
Rocky
0.
0.
0.
N.
0.
0.
0.
0.
0.
0.
0.
N.
N.
0.
0.
0.
0.
0.
concn. , ppm
Knob
128
112
079
A.a
088
080
113
054
106
090
129
A Q
A'.a
080
080
075
065
075
Horton
N.
N.
0.
0.
0.
N.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
Center
A'a
A.a
072
069
063
A.
129
094
073
072
114
095
085
125
110
115
090
080
N.A. = data not available.
Source: Skelly et al. (1984).
reported earlier that episodes in the area result from long-range transport of
03 from urban areas. Fankhauser (1976) cited the transport of 0, in a giant
loop stretching from New York City, Philadelphia, Baltimore, and Washington,
DC, through Virginia, West Virginia, and Ohio, and back to the Wheeling, WV -
Pittsburgh, PA, area. This path continued for 4 to 5 days in September 1972.
Another instance of 0- transport occurred in May 1972, when a stagnant "high"
and a slow-moving "low" transported air from the Chicago and Pittsburgh areas
to Miami, FL.
In the studies discussed above :(Mann et al., 1980; McLaughlin et al. ,
1982; Benoit et al., 1982), decline in vigor and reduction in the growth of
coniferous trees were usually:associated with the following sequence of events
and conditions: (1) premature senescence and loss of older needles at the end
of the growing season; (2) reduced carbohydrate storage capacity in the fall
and reduced resupply capacity in the spring to support new needle growth;
7-10
-------
(3) increased reliance of new needles on self-support during growth; (4)
shorter new needles, resulting in lower gross photosynthetic productivity; and
(5) higher retention of current photosynthate by foliage, resulting in reduced
availability of photosynthate for external use (including repair of chroni-
cally stressed tissues of older needles) (Mclaughlin et al., 1982).
Reported ozone concentrations such as those given above may not fully
represent the exposures sustained by forest ecosystems in the Blue Ridge
Mountains or in other mountainous areas, e.g., the San Bernardino Mountains in
California (see Section 7.6.1). Several considerations are particularly
important for assessing accurately the dose-response relationships reported
for respective forest ecosystems, especially forest ecosystems at higher
elevations: (1) the elevation(s) at which ozone-induced injury or damage has
been observed; (2) the timing of peak ozone concentrations, i.e., during
daylight or after dark; and (3) the possibility that transport trajectories
and various meteorological conditions result in subjecting the forest, or
parts of it, to multiple peak concentrations of ozone concentrations within
a given 24-hr period. Discussions of the variation of ozone concentrations
with altitude and of the occurrence of multiple peaks as the result of trans-
port were presented in Chapter 5, but several important points must be noted
here:
1. Sites at elevations above the nocturnal inversion layer (see
Section 3.4.1 and Figure 3-4) can be exposed to higher peak and
higher total concentrations of ozone than sites at lower eleva-
tions (see, e.g., Wolff et al., 1986; Miller and Elderman,
1977; Miller et al., 1982).
2. The maximum ozone concentrations observed at elevated, moun-
tainous sites, as well as at many non-mountainous rural and
remote sites, often occur at night (see e.g., Chapter 5; Lefohn
and Jones, 1986; Wolff et al., 1986). For species in which the
stomates remain fully or even partially open after dark, such
as eastern white pine, this is particularly important.
3. Sites at higher elevations are often exposed to sustained or
multiple peak concentrations of ozone within a given 24-hr
period as the result of conditions such as (a) trapping inver-
sions; (b) the successive transport of plumes from multiple
urban source areas upwind, either aloft or across terrain
devoid of sufficient ozone scavengers; and (c) the occurrence
of mountain-valley and upslope-downslope flows, such that the
trajectory of an air parcel passes back over the same forest or
stand of trees (see Chapter 5).
7-11
-------
In the forest studies reported above, only daytime ozone concentrations
were monitored. Thus, the reported 1-hr and 8-hr ozone concentrations should
be interpreted and used with caution since they may not represent either the
number or the magnitude of the peak concentrations to which the forests were
exposed.
7.4.1.3 Controlled and Field Studies on Growth of Other Native Vegetation.
Little research has been done to determine the effects of 0- on the growth of
herbaceous plants in natural ecosystems except for the research of Harward and
Treshow (1973), who studied the growth and reproduction of 14 understory
species found growing in the Aspen zone in the western United States. Weights
of both tops and roots of plants decreased with increasing concentrations when
plants were exposed in portable plastic chambers to (L concentrations of 0.15
ppm or 0.3 ppm for 3 hr/day, 5 days/wk throughout the growing season. The
most sensitive species showed injury in less than a week. Plants grown in
ambient air containing 03 concentrations of 0.05 to 0.07 ppm for 2 hr/day took
3 to 4 weeks to show injury symptoms, and the weights of both tops and roots
of these plants were greater than those of plants exposed to the higher 0,
concentrations under controlled conditions.
Decreased vigor was associated with reduced root and top growth. Reduc-
tion in flower production and in the number and weight of seeds was observed
in plants grown at 0.15 or 0.3 ppm 0- (Harward and Treshow, 1973).
7.4.1.4 Mechanisms of Effects of Ozone on Growth of Producers. As discussed
in Section 7.3, ozone has the potential for reducing the growth of green
plants by inhibiting photosynthesis; by altering carbohydrate formation,
allocation, and translocation; and by acutely damaging foliar tissue. In
addition, genetic factors can attenuate or potentiate the growth response of
trees and other green plants to ozone.
7.4.1.4.1 Reduction in Photosynthesis. Trees in which 0- has been shown to
reduce photosynthesis are northern red oak (Quercus rubra L.) (Reich and
Amundson, 1985), loblolly pine (Pinus taeda L.), slash pine (P. elliottii
Engelm. ex Vasey) (Barnes, 1972), ponderosa pine (P. ponderosa Laws) (Miller
et al., 1969; Coyne and Bingham, 1981), eastern white pine (P. strobus L.)
(Barnes, 1972; Yang et al. , 1983; Reich and Amundson, 1985; Botkin et al.,
1972), black oak (Quercus velutina Lamb), sugar maple (Acer saccharum Marsh)
(Carlson, 1979; Reich and Amundson, 1985), and one poplar hybrid (Populus
deltoides X trichocarpa) (Reich and Amundson, 1985) (also see Chapter 6). Two
7-12
-------
of these studies are presented in more detail here to demonstrate the role of
ozone in reducing photosynthesis and the effects of reduced photosynthesis on
growth.
Coyne and Bingham (1981) measured photosynthesis and stomatal conductance
of attached ponderosa pine needles in relation to cumulative incident G\ dose.
Sapling trees in approximately even-aged (18 yr) stands in the forest, growing
in similar environments, were studied. These trees had been exposed long-term
to 0~ throughout their life history. Three chronic injury classes were identi-
fied (I, slight; II, moderate; III, severe) based on morphological 0- injury
symptoms. Differential photosynthetic and stomatal responses were correlated
with the 0- injury classification. Stomatal conductances were somewhat larger
in the more CL-sensitive trees (class III) during July and August, when needles
were growing rapidly, than in Class I and II trees. The decline in photo-
synthesis and stomatal function normally associated with aging was accelerated
as 0- injury symptoms increased. Photosynthesis in all three injury classes
was reduced to about 10 percent of the maximum rate observed in Class I current-
year needles by incident exposures of approximately 800, 700, and 450 ppm-hr
of ozone. Percentage inhibition was based on a comparison with the photo-
synthetic rates of new needles. Photosynthesis declined most rapidly in the
sensitive trees (Class III). Photosynthetic rates were always higher in the
trees with the fewest injured needles. Premature senescence and abscission of
needles occurred soon after photosynthesis reached its lowest level. Losses
in photosynthetic capacity in all trees and needle ages exceeded reductions in
stomatal conductance, suggesting that injury to the mesophyll, carboxylation,
or excitation of components of the COp diffusion pathway was greater than
injury to the stomata.
Reduced photosynthesis has also been observed in white pine. Yang et al.
(1983) studied the effects of 0, on photosynthesis in three clones of white
O
pine with differing 0- sensitivities (sensitive, intermediate, insensitive).
Under controlled conditions, the clones were exposed to concentrations of
0.00, 0.10, 0.20, and 0.30 ppm ozone for 4 hr/day for 50 consecutive days. By
day 10, photosynthesis in the sensitive plants exposed to 0.30 ppm was signifi-
cantly reduced. By day 20, photosynthesis was reduced in sensitive plants
exposed to 0.10, 0.20, or 0.30 ppm 0-. At the end of 50 days, net photo-
synthesis in the sensitive clone exposed to 0.10, 0.20, or 0.30 ppm was reduced
from the control by 24, 42, and 51 percent, respectively. Photosynthesis in
7-13
-------
the intermediately sensitive clone was reduced by 6, 14, and 20 percent,
respectively. The insensitive clone varied from the control at the 20-, 30-
and 40-day periods, but had nearly recovered by 50 days. Decreased rates of
photosynthesis were closely associated with visible needle injury, premature
senescence, and reduction of biomass in the sensitive clones. Reduction in
biomass was associated with the effect of 03 exposure on the rate of photo-
synthesis, with plant metabolism, and with injury to the assimilatory apparatus
of the plants.
7.4.1.4.2 Alterations in Carbohydrate Production and Allocation. Miller et
al. (1969) found that exposure under controlled conditions of 3-year-old
ponderosa pine seedlings to concentrations of 0.15, 0.30, or 0.40 ppm ozone 9
hr/day for 30 days reduced photosynthesis by 10, 70, and 85 percent, respec-
tively. In addition, they noted that the reduction in photosynthesis was
accompanied by decreases in the sugar and polysaccharide fractions of injured
needles. Tingey et al. (1976) also observed that 03 exposure differentially
affected the metabolite pools in the roots and tops of ponderosa pine seedlings
grown in field chambers. The amounts of soluble sugars, starches, and phenols
tended to increase in the tops and decrease in the plant roots exposed to 0.10
ppm 03 for 6 hr/day for 20 weeks. The sugars and starches stored in the tree
roots were significantly less than those in the roots of the controls.
In the study by Mann et- al. (1980) of white pine growing on the Cumberland
Plateau (Section 7.4.1.2), differences in growth between sensitive and tolerant
trees appeared to be caused by premature needle loss and retention of a reduced
quantity of photosynthetically active tissue rather than by a reduction in the
photosynthetic efficiency of the remaining foliage. In the Mclaughlin et al.
(1982) study of the same trees, the availability of less carbohydrate reduced
the vigor of root systems and enhanced the susceptibility of the trees to root
diseases. The loss in vigor of the trees was accompanied by reduced annual
radial growth and a loss in the capacity to respond in years when conditions
were favorable for growth. The primary cause of decline appeared to be exposure
to high concentrations of 0, and a sequence of events and conditions that led
to premature senescence and loss of older needles, lower gross photosynthetic
productivity, and reduced photosynthate availability for growth and maintenance
of trees (Mclaughlin et al., 1982). Carbon-14 transport patterns indicated
that older needles were sources of photosynthate for new needle growth in
7-14
-------
spring and were storage sinks in the fall. The higher retention of
14
Ophotosynthate by foliage and branches of sensitive trees indicated that
the export of photosynthate to trunks and roots was reduced.
In native herbaceous plants, as in trees and cultivated crops (Chapter 6),
(L inhibits the process of photosynthesis in sensitive plants and alters the
distribution of carbohydrates from the leaves to other parts of the plant so
that growth and reproduction are reduced. For example, in the study of native
herbaceous plants by Price and Treshow (1972), grasses visibly injured by 0-
concentrations ranging from 0.15 to 0.25 ppm for as short as 2 hr exhibited a
decrease in carbohydrate production, growth, and reproduction when given
additional daily exposures.
7.4.1.4.3 Foliar Leaching. Krause et al. (1984), in West Germany, have
associated limitations in growth of fir (Abies alba Mill), Norway spruce
(Picea abies (L.) Karst), and certain hardwoods, e.g., beech (Fagus sylvatica
L), with foliar damage. They fumigated seedlings continuously for 6 weeks
with 0- concentrations that ranged from 0.07 to 0.30 ppm. Their studies
indicated that the entrance of ozone into the leaf induced cell membrane
damage in needles and leaves of the trees and resulted in the uncontrolled
loss of nutrients. Leaching from the foliage was enhanced by high light
intensity and low nutrient supply in soils. Membrane damage occurred in the
absence of visible injury. The authors suggested that the loss of nutrients
and reductions in photosynthesis, carbohydrate production, and root growth
from Oo injury causes trees to mobilize and translocate nutrient reserves from
older needles to sites of greatest metabolic activity. Dieback then occurs
because the growing tips of tree branches do not receive the nutrients and
carbohydrates necessary for growth.
Taylor and Norby (1985) have pointed out that foliar leaching is a normal
process and that according to Tukey et al. (1958) deleterious effects on meta-
bolism are not observed if above-ground nutrient losses are rapidly replenished
by root uptake. Furthermore, the rate of foliar leaching is accelerated by
many stress factors in addition to air pollutants, such as water deficiency,
temperature extremes, toxins, and mechanical damage (Tukey and Morgan, 1963).
7.4.2 Factors Modifying Effects of Ozone on Growth of Producers
7.4.2.1 Genetic Factors. The responses of individual plants or animals to
ozone are partly the result of the genetic potential of each individual. Each
7-15
-------
population is a genetically diverse group of interbreeding individuals and the
success of a population of plants or animals in any environment depends in
part on its genetic diversity, that is, the presence of particular gene combina-
tions and variations that give a species or taxon the capacity to adapt to
environmental changes (Treshow, 1980). Plants in a given population (e.g.,
trees in a stand of eastern white pine) will not respond equally to 03 exposure
because of genetic diversity relative to the sensitivity of individual trees
to ozone and because of the microenvironmental heterogeneity of the habitat.
Some will be favored over others. Sensitive plants in a population die or are
unable to compete with tolerant plants, and therefore do not reproduce. The
tolerant plants reproduce and, in time, tolerant populations develop. The
size and success of a population depend on the collective ability of organisms
to reproduce and maintain their numbers in a particular environment. Those
organisms that are tolerant of stress contribute most to future generations
because they have the greatest number of progeny in the population (Woodwell,
1970; Odum, 1971; Smith, R., 1980; Roose et al., 1982). Tolerance refers to
the relative ability of organisms of the same genetic composition (genotype)
to maintain normal growth and remain free from injury in a given polluted
environment. Tolerance is seldom complete, but is a matter of degree (Roose
et al., 1982).
The studies described in Sections 7.4.1.2 (Mclaughlin et al. , 1982;
Benoit et al., 1982) and 7.4.1.4 (Coyne and Bingham, 1981; Yang et al., 1983;
Mann et al., 1980) have demonstrated that ozone sensitivity is not uniform
even among individual plants of the same species. Differences in sensitivity
to ozone are caused in part by differences in genetic potential.
In ponderosa pine, the response to 0- may change with annual dose and
climate. As observed by Miller and coworkers, 16.9 percent of ponderosa pine
examined in field studies were classified as having slight injury in 1969
(Miller et al. , 1969). By 1971, only 6.9 percent of the same trees were
listed as having slight injury, but trees in the moderate injury category had
increased from 15.6 to 20 percent (Miller, 1973). Based on the substantial
shift of ponderosa pines from slight to moderate injury, Miller suggested that
there may be no positive resistance to 0,. The continuing decline of the
ponderosa pine populations studied appears to bear out this suggestion (Miller
et al., 1982). The ability of this population to reproduce and compete with
other populations has therefore decreased.
7-16
-------
Changes in response to 0- similar to those observed in ponderosa pines
growing in their natural habitat have not been reported for eastern white
pine, even though the studies of growth cited earlier show that populations of
this species, like ponderosa pine, can also be divided into at least three
classes of ozone sensitivity. Eastern white pine does not have the same role
in eastern forest ecosystems as ponderosa pine has in western forest ecosystems.
Variability in response to 0- has also been observed in a 2-year study of
hardwood trees growing in urban habitats (Karnosky, 1981; 1983). Tree species
were classified as tolerant, sensitive, or variable. Ozone-tolerant trees
showed no visible symptoms during the 2-year study, but sensitive trees showed
foliar symptoms. The sensitivity of a few species was highly variable; some
individuals consistently showed foliar symptoms and others of the same species
did not. Growth reductions, though postulated, were not measured (Karnosky,
1983).
As with many agricultural crops and some forest species, trees and other
woody plants that are grown in a variety of urban habitats represent individuals
that have been selected over time for viability in stressed environments.
Umbach and Davis (1984) point out that trees obtained from commercial nurseries
are not likely to represent the full range of genotypes present in the wild
population of a species. The work of Rhoads et al. (1980), along with that of
Karnosky (1981; 1983), suggests that, based on foliar injury, the majority of
the plants growing along streets and in urban parks, arboreta, remnant wood-
lots, and suburban communities are relatively insensitive to 0. exposure.
Their relative insensitivity may be the combined result of genetic selection
and physical factors affecting stomatal processes.
7.4.2.2 Other Factors. As described and documented in Section 6.3.2 of the
preceding chapter, numerous factors influence the type and magnitude of re-
sponses of plants to ozone. The response of an individual plant to ozone will
depend upon the physical and chemical environments of the plant and upon
biological factors. Physical factors in the macro- and microenvironments of a
plant that are known to modify the effects of ozone include temperature,
relative humidity, light intensity, soil moisture, soil type, and soil fertil-
ity. Chemical factors known to modify plant response to ozone include other
gaseous pollutants (e.g., SO,,), heavy metals (e.g., cadmium), nutrient defici-
encies and excesses, and agricultural chemicals (e.g., pesticides, herbicides,
chemical protectants).
7-17
-------
In addition to genetic potential (Section 7.4.2.1 above), other biological
factors can modify the response of trees to ozone. Such factors include the
age of the plant and the developmental stage of both leaf and plant. For
example, the age at which current foliage is most sensitive to 0- varies among
plant species. Conifer seedlings are susceptible to 0- for a much longer
period of time than many other plants. Susceptible conifer species were
observed by Davis and Wood (1972), for example, to be most sensitive from
4 through 13 weeks after bud break. It should be noted, however, that physio-
logical balances and the sensitivity of tree seedlings in chambers may be
different from those of mature forest trees (Mclaughlin, 1985). Woody shrubs
and vines were most sensitive from 4 to 8 weeks after bud break (Davis and
Coppolino, 1976). In both studies by Davis and coworkers, susceptibility of
the plants to 0, decreased as the growing season progressed. In contrast,
azalea cultivars became more susceptible as the growing season progressed and
retained their susceptibility into the fall (Davis and Coppolino, 1974).
Thus, in conjunction with biological factors, the timing of exposures appears
to modify plant response to ozone (see, e.g., Tingey et al., 1973).
As discussed in Section 6.3.2 of the preceding chapter, ozone occurs in
conjunction with other stresses that also modify the productivity of an
individual plant or of plant populations. Among the stresses to which trees
are simultaneously exposed along with ozone are biotic pathogens and competi-
tion. Fungi and insects are by far the most important biotic stress factors
in most forests (Cowling, 1985). The reader is referred to Section 6.3.2 (and
Table 6-2) for references to studies on the interactions between ozone and plant
diseases and insect pests in agricultural species and in some tree and ornamen-
tal species. Additional discussion on the effects of ozone on plant-pathogen
and plant-insect interactions is presented in Sections 7.5 and 7.6 below.
7.5 EFFECTS OF OZONE ON OTHER COMPONENTS AND INTERACTIONS IN TERRESTRIAL
ECOSYSTEMS
The ecosystem processes of energy flow and nutrient cycling are directly
related to the plant physiological processes of photosynthesis, nutrient
uptake, biosynthesis, respiration, and translocation. The alteration of these
physiological processes is the fundamental cause of all other ecosystem effects.
Data presented in Chapter 6 (Section 6.3) and in Section 7.4 above indicate
that photosynthesis and the partitioning of photosynthate in the plant can be
7-18
-------
affected by exposure of the plant to ozone. The data presented also indicate
that a sustained reduction in photosynthesis will ultimately affect the growth,
yield, and vigor of a plant. Such data highlight the potential for ozone to
reduce primary productivity. The effects of ozone on primary productivity
have, in turn, potential consequences for entire ecosystems, since consumer
and decomposer populations in ecosystems depend on the flow of energy from
producers through the food chain.
Two fundamental processes in the food chain are photosynthesis and decom-
position. Thus, the two groups of organisms particularly critical to the
maintenance of an ecosystem are the producers, through which solar energy,
carbon, and nutrients enter the biotic components of the ecosystem; and the
decomposers, through which nutrients are released for reuse.
The following sections discuss the potential consequences for respective
populations, processes, and interactions in ecosystems of ozone-induced
responses in producers. Figure 7-1, derived from McClenahen (1984), shows how
air pollution may be expected to affect forest ecosystems. Air pollutants may
act as both predisposing and inciting stresses that influence trees. Predis-
posing stresses usually have a long-term role in weakening trees and thus
making them more susceptible to inciting factors, which are short-term episodic
stresses, such as insect defoliation, weather damage, or acute air pollution
injury, that may abruptly alter tree physiology and increase the susceptibility
of the tree to secondary biotic stresses (Mclaughlin, 1985). In Figure 7-1,
it is clear that the impact of air pollutants is sequential but also cyclic;
that is, effects on individual organisms eventually find expression as effects
on the ecosystem, with ecosystem changes producing, in turn, effects on
individual organisms.
7.5.1 Producer-Producer Interactions: Competition and Succession
According to Ehrlich and Mooney (1983), in forest ecosystems, trees are
the producers that serve as the controller organisms of the ecosystem; that
is, they are the organisms that determine the species composition and trophic
relationships of the ecosystem. Unless they are the result of a specific
biotic disease or an acute pollutant exposure, injuries and disturbances to
trees are cumulative and are frequently the culmination of a number of chronic
stresses. Injury to or disturbance of tree species, whether from air pollution
or other stresses, starts the retrogressive successional processes that could
7-19
-------
ORGANIZATIONAL
LEVEL
PROCESS
AIR
POLLUTION
• ORGANISM
I
-------- PHYSIOLOGICAL FUNCTIONS
INTRASPECIFIC COMPETITION
INTERSPECIFIC COMPETITION
COMMUNITY
SUCCESSION
ECOSYSTEM
•ELEMENT CYCLING
NUTRIENTS
TOXINS
TROPHIC RELATIONSHIPS
Figure 7-1. Pathways of air pollutant impact in forest ecosystems.
Source: McClenahen (1984).
7-20
-------
ultimately lead to the loss of the ecosystem (see Table 7-1, Phases II through
V; Bormann, 1985).
In most ecosystems, the principal interaction among species is competition
for resources. Producer species compete with one another for the limited
resources of space, light, water, and minerals (Billings, 1978). Those producer
species and populations best able to utilize the resources available survive
and replace those that previously occupied the area. A gradual orderly change
in community composition called succession occurs over time as dominant plants
change and as new communities arise, develop, and mature. Succession is
characterized by a shift from annual species through biennial and perennial
herbs to woody shrubs and trees, and changes in the woody species over time.
Each successive community is related to the one that preceded it. In time,
communities arrive at some form of steady state and are more or less silf~
maintaining as long as abiotic factors remain constant. Through succession,
ecosystems achieve the most stable state possible within the constraints of
the environment (Odum, 1971; Cox and Atkins, 1979; Smith, R. L., 1980). Along
with other abiotic factors, air pollution can affect the direction of succes-
sion by injuring the sensitive plant species.
In forest ecosystems, trees represent the later stages of succession and
are adapted for high competitive ability (Brown, 1984). In the early stages
of ecosystem development, however, competition by grasses and other herbaceous
vegetation is very important. Competition among broad-leaved and needle-bearing
trees is keen in mixed forest stands, especially in the later stages of stand
development. Competition-induced mortality is an important feature of all
planted and naturally regenerated forests. In fact, in most forests, more
trees will die because of competitional stresses than all other stresses
combined. Only a few trees survive to maturity (Cowling, 1985).
Ozone stress is an additional factor affecting growth and species composi-
tion and succession in forests and other plant communities in both the western
and eastern United States. Ozone exerts selection pressure on sensitive species
by causing their demise or by weakening them and making them less able to
compete. Ozone-tolerant species may replace them in the plant communities.
Disruption of food chains and modification of the rates of nutrient cycling
resulting from species changes may result in a less stable community. Injury
to, or disturbance of, the dominant tree species may return succession to an
earlier stage (Woodwell, 1970; Bormann, 1985).
7-21
-------
McClenahen (1978) has provided quantitative data on the impact of polluted
air on the various strata and the composition of a forest ecosystem exposed to
effluents from point sources of air pollution for nearly 40 years. In stands
situated along a gradient of polluted air containing elevated concentrations
of chloride, sulfur dioxide, and fluoride (photochemical oxidants were not
monitored), overstory, subcanopy, shrub, and herb strata were analyzed for
pollution effects.
A shift in the species composition of forest stands occurred on the sites
investigated and was related to the pollution gradient. The density of over-
story and herb layers was also correlated with the gradient. In the herb
layer, an increase occurred in light-tolerant species that was an indirect
effect of air pollution, resulting from the reduced overstory density. Light-
tolerant species composed 68 percent of the total in areas of high pollution
compared to 34 percent in areas of low pollution (McClenahen, 1978). Although
concentrations of 0- were not reported, the study illustrates how pollutant
mixtures typical of some ambient conditions can change the species composition
of forested areas by weakening trees in a population and thus lessening the
ability of that population to compete. The changes observed were consistent
with the first four phases of ecosystem response outlined in Table 7-1 (Bormann,
1985).
The modifying role of intra- and interspecific competition must be eval-
uated when studying long-term responses of forest communities to high 0,
concentrations (Taylor and Nor&y, 1985).' Simulations of community dynamics in
a pollutant-stressed forest in the southern Appalachian Mountains and in the
eastern deciduous forest of North America after removal of the American chestnut
[Castanea dentata (Marsh.) Borkh.] suggest that growth rates for certain tree
species were significantly modified by competition. These simulations also
suggest that in forests of mixed species with uneven-aged stands the subtle
long-term responses are likely to be shifts in species composition rather than
widespread degradation (West et al., 1980; cited in Taylor and Norby, 1985).
The effects of competition as a modifying factor in the responses of
forest ecosystems to pollutants (e.g., ozone) are not unidirectional. Just as
ozone is thought to modify the competitive ability of a species through its
effects on sensitive individuals of that species, competitive stress may also
modify responses of individuals to ozone. As McLaughlin (1985) has stated,
..."competitive stress, "..." in well stocked forest stands, may have significant
7-22
-------
influence as either a predisposing or contributing factor in tree responses to
anthropogenic stress."
Competition increases selection for resistance under polluted conditions
and selection against resistance under less polluted conditions. Studies on
plant responses to heavy metals and herbicides indicate that resistant popula-
tions develop, but that once the stress is removed the pollutant-resistant
plants tend to decline in number (Roose et al., 1982). This evidence is
corroborated by observations of ecosystems functioning under specific natural
conditions. Certain terrestrial ecosystems require a major disturbance
(e.g., fire, drought, and windstorms) to retain their characteristics (Vogl,
1980; Smith, W. H. , 1980). In the absence of disturbance, some ecosystems
appear to degrade, lose nutrients, become less productive, and have fewer
species with a smaller biomass (Woodwell, 1970; Gorham et al., 1979).
Acute injury from air pollution resembles that from herbicides in that
selection for resistances tends to be episodic. Chronic air pollution more
closely resembles contamination of soil by heavy metals in that plants experi-
ence the polluted environment for a considerable portion of their lives.
Resistance in either situation depends on the presence of the resistant or
tolerant genotype in the plants that are growing in unpolluted air (Roose et
al., 1982).
Annual plants in a forest ecosystem under selection pressure from air
pollution and pollution-related stresses are capable of altering the genetic
composition of the population each year through sexual reproduction until a
stable population adapted to the stresses develops (McClenahen, 1978). As
noted in Section 7.4.1, forest trees and shrubs, which are perennial plants,
must cope with the cumulative effects of both short- and long-term stresses.
The response of trees to stress may appear rapidly, for example, as when
sensitive eastern white pine needles show visible evidence of exposure to
episodic, high 0- concentrations. In other instances, however, responses are
often subtle and may not be observable for many years as trees adapt and their
response to stress is expressed in differential growth resulting from changes
in carbon allocation (Waring and Schlesinger, 1985; Mclaughlin and Shriner,
1980), such as those induced when 0- affects photosynthesis. Changes in
growth patterns of ponderosa, Jeffrey, and eastern white pine trees have been
attributed to stress resulting from 0^ exposures that began 15 to 20 years
earlier (Miller and Elderman, 1977; Miller et al., 1982; Mclaughlin et al.,
1982; Benoit et al., 1982). Dendroecological studies of the dieback and
7-23
-------
decline of red spruce (Picea rubens Sarg) in the northeast (Johnson and Siccama,
1983) and of reduced growth rates of red spruce, balsam fir [Abies balsamea
(L.) Mill.], and Fraser fir [A. fraseri (Purshl) Poir.] in central West Virginia
and western Virginia (Adams et al., 1985) also provide further evidence that
the reductions in growth and mortality measurable today probably began at
least 20 years ago. Currently, there is no agreement as to the trigger factor
that precipitated the dieback, mortality, and decreased growth, but multiple
stresses, including air pollution, have been suggested (Johnson and Siccama,
1983; Adams et al., 1985). Ecosystem responses to these stresses usually are
observable only after long periods of time.
According to Whittaker (1965), productivity (carbohydrate production) of
a species appears to be the best predictor of the relative importance of that
species in an ecosystem. When assessing the responses of forest ecosystems to
On, the consequences of the loss of a particular plant species should be
evaluated accordingly. This criterion explains why the loss of ponderosa pine
in the San Bernardino Forest has had a greater impact than the loss of sensitive
eastern white pine in the Appalachian Mountains. Their roles in the respective
ecosystems are not of equal importance.
Studies of successional changes in specific ecosystems exposed to ozone
are described in Sections 7.6 and 7.7 below. Data presented there indicate
the potential for the occurrence of ozone-induced changes in composition and
in successional patterns in forest and other ecosystems (see, e.g., Cobb and
Stark, 1970; Miller, 1973; Harward and Treshow, 1973; Miller and Elderman,
1977; Miller et al., 1982).
7.5.2 Producer-Symbiont Interactions
7.5.2.1 Mycorrhizal-Plant Interactions. The roots of most plants growing
under natural conditions are invaded by fungi and transformed into mycorrhizae
or "fungus roots." The host plant and the fungus live together in an associa-
tion that is generally beneficial to both organisms. The morphology of the
root is modified, and as long as a balanced relationship is maintained no
pathological symptoms occur (Gerdemann, 1974). Most plants, including important
forest and horticultural species, could not reach maximum growth rates without
mycorrhizae. Mycorrhizal fungi increase the solubility of minerals, improve
the uptake of nutrients for host plants, protect roots against pathogens,
produce plant growth hormones, and move carbohydrates from one plant to another
(Hacskaylo, 1972). The mycorrhizal fungi in turn obtain food from the host.
7-24
-------
This fungus-plant relationship is particularly important to plants growing on
nutrient-poor soils.
Ozone may disrupt the association between the mycorrhizal fungi and host
plants, possibly by inhibiting photosynthesis and reducing the amounts of
sugars and carbohydrates available for transfer from the leaves to the roots.
Carbohydrate partitioning is altered in plants exposed to ozone (Section 7.4.1.4
and Chapter 6) to the degree that certain plant organs (e.g., roots) may be
deprived of photosynthate. In Chapter 6, the effects of ozone on root-to-shoot
ratios were documented for agricultural species, demonstrating that ozone can
affect root systems. Translocation may be reduced in ozone-stressed plants
and thus deprive the mycorrhizal fungus of the amount of photosynthate needed
to satisfy mycorrhizal requirements. The result would be reduced effectiveness
of the mycorrhizal fungi (McCool and Menge, 1983).
Mycorrhizae are sensitive to the photosynthetic capacity of the host and
the capacity of the host to translocate carbon compounds to the roots (Hacskaylo,
1973). For example, when seedlings of Virginia pine (Pinus virginiana Mill.),
inoculated with the mycorrhizal fungus, Thelephora terrestris, and growing
under a 16-hour photoperiod, were switched to 8-hour photoperiods, the seedlings
became dormant within 4 weeks. No further invasion of rootlets by the fungus
occurred even though root growth continued. Fungal sporophores were formed on
the seedlings that remained under the 16-hour photoperiod. Studies have shown
that simple sugars provided by plant roots are readily utilized by mycorrhizae
and enhance fungal inoculation (Hacskaylo, 1973; Krupa and Fries, 1971).
Several studies of the effects of ozone on root and mycorrhizal systems
have been reported. In a controlled study, Mahoney et al. (1982) found evidence
that the mycorrhizal association of loblolly pine seedlings was not impaired by
exposure to 0.07 ppm of 0- plus 0.06 ppm of S02 for 6 hr/day for 35 days; how-
ever, a 12 percent decrease in dry weight of shoots was observed. In an earlier
controlled study, McCool et al. (1979) demonstrated that infection of citrange
(a citrus hybrid) by Glomus fasciculatus, an endomycorrhizal fungus, was
decreased by exposure to 0.45 ppm ozone for 3 hr/day, 2 days/wk over 18 weeks.
In the field, Berry (1961) examined the relationship of root condition and root
fungi to emergence tipburn, i.e., ozone injury. Sampling of trees indicated
the occurrence of almost twice the percentage of living feeder roots on healthy
trees as on ozone-injured trees. The observations by Berry were made on trees
in a forested valley in eastern West Virginia, and on trees in eastern North
7-25
-------
Carolina, where ozone concentrations as high as 0.22 ppm for short durations
(~1 hour), as measured by Mast meter, were observed in 1962 and where 4-hour
average ozone concentrations in 1961 ranged from 0.03 to 0.065 ppm (Berry and
Ripperton, 1963; Berry, 1964).
In the San Bernardino forest in California, Parmeter et al. (1962) observed
that the feeder rootlet systems of ponderosa pines exposed to ozone showed
marked deterioration. The number of mycorrhizal rootlets was decreased and
many had been replaced by saprophytic fungi in stressed trees. (Information
on ozone/oxidant concentrations in the San Bernardino forest is given in
Section 7.6.)
Mycorrhizae assist in protecting conifer roots from pathogens such as
Heterobasidion annosum (syn. Fomes annosus) (Krupa and Fries, 1971). Injury
to the mycorrhizae or reduction in the number of mycorrhizae, such as can be
induced by ozone exposure, can remove this protection. Non-mycorrhizal and
mycorrhizal root systems contain essentially the same major volatile compounds;
however, studies using Scots pine (Pinus sylvestris L.) indicate that the
concentrations of monoterpenes and sesquiterpenes increase twofold to eightfold
in roots infected by mycorrhizae. Many of the monoterpenes identified in
mycorrhizal root systems are constituents of the oleoresins commonly found in
conifers. Oleoresins play an important role in the resistance of wood to
decay fungi (Rishbeth, 1951). Volatile oleoresin components from ponderosa
pine have been shown to inhibit the growth of H. annosum and four Ceratocystis
species (Cobb et al., 1968), and are believed to aid in defense of trees from
bark beetles (Stark and Cobb, 1969). James and coworkers associated decreased
oleoresin exudation with increased susceptibility to infection by H. annosum
in roots (James et al., 1980a) and cut stumps (James et al., 1980b) of ponderosa
and Jeffrey pines.
7.5.2.2 Bacterial-Plant Interactions. Ozone has also been shown to influence
bacterial symbiosis in herbaceous species. Whether it does so in trees and
woody shrubs has not been investigated. In herbaceous species, the rate of
nitrogen fixation by symbiotic bacteria is dependent on the rate of photosyn-
thesis by the plant. Symbiotic nitrogen fixation is the major bological
source of fixed nitrogen (Tingey and Blum, 1973). Blum and Tingey (1977)
found reduced root growth and reduced nodulation of soybeans (Glycine max (L.)
Merr cv. Dare) by the bacterium Rhizobium japonicum when plant tops were
exposed to 0~. No growth reductions occurred when the plant tops were protec-
ted from exposure to 0- (Blum and Tingey, 1977). In an earlier study (Tingey
7-26
-------
and Blum, 1973), nodule number, nodule weight per plant, root growth, and
leghemoglobin content per plant were all reduced by a 1-hr exposure to 0.075
ppm 0,. The reductions were associated with reduced photosynthetic capacity
and less photosynthate for translocation to the roots. In a separate study,
ladino clover (Trifolium repens L. cv. Till man) was treated with filtered air,
3 3
0.3 ppm (588 ug/m ) of 03, or 0.6 ppm (1176 ug/m ) of 0- for two 2-hr exposures,
one week apart, in controlled environment chambers (Letchworth and Blum,
1977). Plants of various ages were treated. Ozone reduced the growth and
nodulation of test plants. The influence of 0- varied with gas concentration
and plant age.
7.5.3 Producer-Consumer Interactions
Consumers (heterotrophs) are organisms that feed on other organisms and
constitute all trophic levels above the first. Production (energy storage) by
consumers is termed secondary production. Consumers are extremely diverse,
ranging in size from single-celled microscopic forms to large mammals. Only a
limited amount of information on their response to pollutants is available
(Newman, 1979; Alstad et al., 1982) despite the importance of the role of
consumers, especially insects, in ecosystems. The influence of oxidants on
these organisms is assumed to be chiefly through the food web. Few studies
have been conducted to determine whether ozone has a direct impact on the
organisms themselves.
The effects of ozone on producer-consumer interactions that have been
observed may be secondary; that is, ozone may alter producer-consumer interac-
tions by predisposing trees in a forest ecosystem to attack by predatory
beetles and fungal pathogens. In addition, an unhealthy tree has less energy
available to transfer through the food web so that the relationship among
consumers in the food web is changed. Any mature natural community transfers
10 to 20 percent of the energy fixed by plants to herbivores (Woodwell, 1974).
Disruption of photosynthesis and subsequent carbon allocation for vegetative
and reproductive growth can decrease the amount of food available to other
trophic levels in the food web and thus alter the movement of energy and
nutrients through an entire system. The possibility of such alterations in
response to stress from ozone is consistent with theory but such ozone-induced
changes have not been demonstrated.
7-27
-------
Invertebrate consumer populations such as insects may be subject to the
influence of oxidants on their host or on their habitat. Insects are among
the most important heterotrophic groups in ecosystems. The insect-plant
relationship is a close one throughout succession. Changes in the plant
community are soon followed by changes in the insect community. Herbivorous
insects may play a key role in succession (Brown, 1984). While consistent
with ecological theory, such changes as the result of ozone stress are again
conjectural. The data available on the effects on insects of pollutants in
general, and the effects of ozone in particular, are meager and are dispropor-
tionate to the importance of insects in forest ecosystems.
The review by Alstad et al. (1982) cites the work of Levy and coworkers
(Levy et al., 1972; 1974), in which three species of Diptera were exposed to
ozone. Prolonged fumigations with high concentrations of ozone, well above
ambient air levels, inhibited egg hatching but no differences were observed
between controls and exposed insects during the larval and pupal stages.
Adults fumigated with ozone showed stimulation of ovipositon, an increase in
the number of eggs laid, and an increase in the adult population (Levy et al.,
1972). None of these effects were seen in ozone-exposed cockroaches or fire
ants (Levy et al., 1974).
In contrast to the evidence for possible direct effects of ozone on insects,
the evidence for indirect effects of ozone on insects (herbivores) is stronger,
and indicates that effects on producers can result, for example, in increases
in bark beetle infestation (see Section 7.6). Bark beetles are the most
damaging and economically significant insect pests of commercially important
conifers in the United States (Stark and Cobb, 1969). Beetle outbreaks in
western forests are associated with several predisposing factors. These
include host weakening caused by photochemical oxidants; root disease initiated
by the fungi H. annosum or Verticicladiella wagenerii (Stark and Cobb, 1969);
insect defoliation, such as pine looper stripping of ponderosa pine (Dewey
et al., 1974); and various climatic stresses, such as drought and windthrow
(uprooting and breakage by strong winds) (Rudinsky, 1962).
The only evidence of an effect of ozone on mammalian species of forest
ecosystems is some evidence from studies of the San Bernardino forest ecosystem.
Reductions in fruits and seeds in that ecosystem appear to be one of the
effects of ozone-related stress in producers and data indicate that such
reductions may be affecting the populations of small mammals. Fruits and seeds
7-28
-------
make up the largest part of the diet of most of the small mammals in mixed
conifer forests. This is particularly true for the deer mouse (Peromyscus sp.)>
harvest mouse (Reithrodontormys sp.), chipmunk (Etuamias sp.), ground squirrel
(Cal1ospermophi1 us sp.), and western gray squirrel (Sciurus griseus anthonyi
Mearns). Alterations in availability of seeds and fruits can alter the habitats
and reproduction of these rodents (Taylor, 1973).
A trapping program at vegetation plots differentially impacted by chronic
oxidant dose indicated that the same species were present when compared with
results from trappings made 70 years ago (Kolb and White, 1974). Population
numbers, however, appeared to be lower in comparison with other similar forest
systems. Some evidence suggests that the size and frequency of acorn crops
from California black oak may be smaller in areas receiving the greatest
seasonal oxidant exposure (Miller et al., 1980). Reduced acorn availability
could have an impact on small mammal populations.
Small mammals are important members of coniferous ecosystems because they
are primary vectors of spore dissemination for hypogeous mycorrhizal fungi.
Mammalian mycophagists spread the spores of the fungi necessary for the survival
and health of conifers (Maser et al., 1978). Thus, changes in the structure
and species composition of a forest will not only have an impact on the small
mammal population, but also on the hypogeous fungi and therefore on whether
coniferous species return to an ecosystem.
7.5.4 Producer-Decomposer and Producer-Pathogen Interactions
Decomposers are organisms such as litter-feeding invertebrates, bacteria,
fungi, and protozoa (Smith, R. L., 1980) that are capable of degrading complex
compounds and utilizing some of the decomposition products as their own food
source while releasing inorganic substances (e.g., essential elements such as
calcium, phosphorus, and magnesium) for use by other organisms.
Generally, one-third or more of the energy and carbon fixed annually by
producers during photosynthesis in the forests is contributed to the forest
floor as litter (mostly leaves) (Ovington, 1957). The reservoir of energy and
mineral nutrients represented by litter is a very important resource in natural
ecosystems since the growth of new green plants depends on the release of
nutrients by decomposer organisms. In agroecosystems, litter is often removed
or burned and fertilizer is added to the soil to replace the nutrients lost.
7-29
-------
In a conifer forest, however, litter production and decomposition release
approximately 80 percent of the annual mineral uptake, with the remainder
retained in the living parts of the trees (Millar, 1974).
Needles on conifers usually persist for more than one year. Ozone causes
premature senescence and the loss of older needles at the end of a growing
season; and by reducing photosynthetic productivity, ozone decreases the
amount of carbohydrates in the needles. The early loss of needles interferes
with the decomposition process because the succession of fungi that normally
takes place does not occur when needles drop off. Though most decomposition
occurs on the forest floor, pine needles are invaded by fungi several months
before needles are shed (Stark, 1972).
Bruhn (1980) has investigated the effects of oxidants on needle microflora
population dynamics of pine in the San Bernardino National Forest. The decompo-
sition of litter consisting of Oo-stressed needles was concluded to be more
rapid. The taxonomic diversity and population density, however, of fungi that
colonized living needles and later participated in decomposition were both
reduced by 0^ injury as the normal increase with age was blocked by premature
needle senescence and abscission. The author concluded that this alteration
in microflora could weaken the stability of the decomposer community.
Decomposition occurs on the forest floor, even though pine needles are
infected by decomposer fungi prior to needle drop. Thus, the effects of ozone
and other oxidants on most of the decomposition process and on decomposers
themselves remain uncertain in natural habitats. Although the occurrence of
rapid fluxes of ozone to soil surfaces and the forest floor has been reported
(National Research Council, 1977), documentation for the occurrence of such
events is poor.
In Chapter 6, documentation is provided for the interactions between
ozone-exposed plants and their pests. As noted in that chapter, ozone may
inhibit or stimulate infection of plants by pathogens; and ozone may modify
the success of other plant pests, either directly through effects on the
invading organisms or indirectly through modification of the host plants. It
is also possible that complex plant-insect or plant-pathogen interactions may
alter the sensitivity of the plant to ozone. Studies showing modifications of
plant disease by ozone in ornamentals and trees were tabulated in Chapter 6
(Table 6-2). None of the studies cited showed modification of ozone injury by
infection with plant pathogens.
7-30
-------
In a field study in the Blue Ridge Mountains of Virginia, Skelly (1980)
found an increased incidence of root disease from Verticicladiella procera in
oxidant-injured eastern white pines. Costonis and Sinclair (1972) found that
Lophodermium pinastris and Aureobasidium pullulans were the fungi more commonly
collected from eastern white pine foliage showing ozone injury. When trees
were inoculated in conjunction with exposure to 0.06 to 0.1 ppm ozone for 4.5
hours, however, no evidence of additive or interactive effects was found
(Costonis and Sinclair, 1972).
Only a few studies have reported on the effects of ozone in combination
with other pollutants on disease development in woody species. Weidensaul and
Darling (1979) inoculated Scots pine (Pinus sylvestris L.) seedlings with the
fungus, Scirrhia acicola, 5 days before or 30 minutes following fumigation for
6 hours with 0.20 ppm S02, 0.20 ppm 0-, or both gases combined. Significantly
more brown-spot lesions were formed on seedlings fumigated with S02 alone or
S02 combined with 0- than on controls when inoculation was done 5 days before
fumigation. When inoculation was done 30 minutes after gas exposure, seedlings
exposed to S02 alone had more lesions than those exposed to 0, alone or 0,
combined with S02> but the numbers of lesions did not differ significantly
between fumigated seedlings and controls. The authors concluded that ozone-
induced stomatal closure may have been responsible for the latter observation.
As the results of the above studies indicate, the outcome of a pollutant-
plant-pathogen interaction depends on the particular plant and pathogen involved
and is also modified by environmental and ozone-exposure conditions. Laurence
and Weinstein (1981) have emphasized the critical importance of examining
multiple pollutant effects and the interactive effects of air pollutants with
pathogens and insects in determinations of growth impacts. Likewise, Manion
(1985) has emphasized the necessity of taking non-pollutant stresses, both
biotic and abiotic, into account when attempting to attribute changes in
forest ecosystems to air pollutants. Ecosystem responses will always be the
integration of multiple stresses acting over time and space on diverse popula-
tions (see, e.g., Manion, 1985; Cowling, 1985; Smith, 1985; Prinz, 1985).
Whether ozone has direct-or only indirect effects on plant infection by
pathogens or on the course of the disease is unknown. The data of Hibben and
Stotsky (1969), however, are illustrative of the fact that the dose of ozone
required for direct effects on fungi, for example, may be much higher than
ambient concentrations. These investigators examined the response of detached
7-31
-------
spores on agar of 14 fungi (none of them forest species) to 0.1 to 1.0 ppm of
03 for 1, 2, and 6 hr. The large pigmented spores of Chaetomium sp., Stemphylium
sarcinaeforme, S. loti, and Alternaria sp. were not influenced by 1.0 ppm of
0-. Germination of Trichoderma viride, Aspergillus terreus, A. m'ger,
Pem'ciIlium egyptiacum, Botrytis allii, and Rhizopus stolonifera spores was
reduced by 0, exposure, but usually in concentrations above 0.5 ppm, though
occasionally by doses of 0.25 ppm of 0- for 4 to 6 hours. Lower doses stimulated
spore germination in some cases.
7.6 EFFECTS OF OZONE OR TOTAL OXIDANTS ON SPECIFIC FOREST ECOSYSTEMS
In previous sections of this chapter, the effects of ozone or total
oxidants on a variety of ecosystem components and on the interactions of
respective components have been discussed. In this section, results are
presented from extensive studies of the forest ecosystems of the San Bernardino
Mountains in California, in which multiple species and trophic levels were
examined, and from studies of the forest ecosystems of the Blue Ridge Mountains
of Virginia.
7.6.1 The San Bernardino Mixed-Coni'fer Forest
One of the most thoroughly studied ecosystems in the United States is the
mixed coniferous forest ecosystem in the San Bernardino Mountains of southern
California. The San Bernardino Forest is located at the eastern end of the
80-mile-long South Coast Air Basin, where a severe air pollution problem has
been created by the last three decades of extensive urban and industrial
development (Miller and Elderman, 1977).
Ponderosa pine (Pinus ponderosa Laws) is one of five major species in
this mixed-conifer forest, which covers wide areas of the western Sierra
Nevada Mountains and other mountain ranges from 1000 to 2000 m (3000 to 6000
ft) in elevation, including the San Bernardino Mountains in southern California.
Five forest subtypes exist (Miller and Elderman, 1977): (1) ponderosa pine,
(2) ponderosa pine-white fir, (3) ponderosa pine-Jeffrey pine, (4) Jeffrey
pine-white fir, and (5) Jeffrey pine. Above 2000 m, Jeffrey pine (Pinus
jeffreyi Grev and Balf) replaces ponderosa pine. Other species in the forest
are sugar pine (Pinus lambertiana Dougl.), white fir (Abies concolor Lindl.),
incense cedar (Libocedrus decurrens Torr.), and California black oak (Quercus
kelloggii Newb.).
7-32
-------
Sensitive plant species in the San Bernardino National Forest, such as
ponderosa pine, began showing unmistakable injury in the early 1950s (Miller
and Elderman, 1977), but the source of the injury was not identified as ozone
until 1962 (Miller et al., 1963). In some of the earliest studies of the San
Bernardino forest, Parmeter et al. (1962) estimated that 25,000 acres of the
mixed-conifer forest had been affected by photochemical oxidants, with ponderosa
pines severely injured but with no injury at that time to associated species
(Miller and Millecan, 1971). Subsequently, ground and aerial surveys showed
that ponderosa and Jeffrey pines on 100,000 of 160,000 acres of the forest
showed moderate to severe injury (Wert, 1969). Even in these early studies,
Stark et al. (1968) reported that the oxidant-injured trees were more suscep-
tible than healthy trees to infestation by bark beetles.
In 1968, an inventory was begun of a 575-acre study block where much
injury was evident to establish, by means of a scoring system, the extent of
injury of ponderosa pine trees >4 inches in diameter at breast height (dbh).
A monitoring station was also established in 1968 to measure concentrations of
total oxidants in the Rim Forest-Sky Forest area (elevation about 5500 ft),
where total oxidants were then measured continuously for a minimum of 5 months
per year from 1968 through 1972 (Mast Meter calibrated against buffered 2
percent potasium iodide; see Chapter 4 for details of the method). Monitoring
during that period showed ozone concentrations >0.08 ppm for >1300 hours, with
concentrations rarely decreasing below 0.05 ppm at night near the crest of the
mountain slope (Miller, 1973). The 1-month averages of the daily maxima of
total oxidant concentrations are given in Figure 7-2 by month for the 5 years
of the study period (Miller, 1973). The data in Figure 7-2 also show the
number of hours per month when the California oxidant standard of 0.1 ppm was
exceeded. The most severe, single, daily maximum oxidant concentration in the
area, 0.58 ppm, occurred in June 1970, between 4:00 and 9:00 p.m. PST (Miller,
1973).
It should be noted here that the San Bernardino Mountains, situated east
of the Los Angeles basin, are often subjected to episodic or sustained high
concentrations of ozone, partly because of the frequent occurrence there of
"trapping inversions," that is, persistent elevated inversions. Precursors
emitted into an inversion layer or into the layer below the base of an elevated
inversion can produce relatively high ozone concentrations that persist for a
considerable time period or over a considerable distance of wind travel from
7-33
-------
GO
0.35
0.30
S 0.25 H
9 0.20 H
X
o
Ł 0.15 H
0.10-
0.05-
-+ TOTAL OXIDANT. ppm
HOURS EXCEEDING 0.10 ppm
M J J A S O
1968
AMJJASON
1969
A M J J A S O
1970
M J J A S O
1971
. 14
12
10
8
• 6
• 4
• 2
M J J A S
1972
E
a.
a.
O
6
a
o
X
UJ
cc
D
O
YEARS AND MONTHS
Figure 7-2. Total oxidant concentrations at Rim Forest (5640 ft.) in southern California during May
through September, 1968 through 1972. Values of total oxidant are averages of daily maxima for a
month. The number of hours in which total oxidant exceeded 0.1O ppm was also recorded for the 5
years.
Source: Miller (1973).
-------
the precursor source area (Section 3.4.1). In addition, an -increasing concen-
tration gradient with increasing elevation occurs in the San Bernardino Mountains
as the result of well-documented upslope flows (Section 5.5.2.4).
Based on the results of the inventory and of accompanying studies, some
preliminary conclusions were drawn:
1. Ponderosa and Jeffrey pine were suffering the most injury. Mortality
of one population of ponderosa pine (n = 160) was 8 percent between
1969 and 1971 (p = 0.01); in a second population (n = 40), mortality
was 10 percent between 1968 and 1972. White fir populations had
suffered slight damage, with scattered individual trees showing
severe symptoms. Sugar pine, incense cedar, and black oak exhibited
only slight foliar injury from oxidant exposure.
2. A substantial shift occurred in ponderosa pines from the "slight
injury" category in 1969 to the "moderate injury" category in 1971,
indicating that there was continuing oxidant stress and that the
selective death of ponderosa pines was occurring.
3. Suppression of photosynthesis in seedlings was observed (Miller et
al., 1969). In ponderosa pine saplings, needles shortened by exposure
to oxidants returned to normal length when the seedlings were moved
to ozone-free air during 1968 to 1973 (Miller and Elderman, 1977).
4. Bark beetles were judged to be responsible for the death of weakened
trees in the majority of cases. Elimination of ponderosa pine from
the mixed-conifer forest was postulated to occur in the future if
the rate of bark beetle attack were to continue unabated (Cobb and
Stark, 1970).
5. Aerial portions of ozone-injured pine trees showed a decrease in
vigor that was associated with deterioration of the feeder root
system (Parmeter et al., 1969).
6. Seed production was decreased in injured pines. Ordinarily, trees
25 to 50 inches dbh produce the most cones, but they'were also the
most sensitive to oxidants (Luck, 1980).
7. Understory plant species sensitive to oxidant pollution may already
have been removed by air pollution stress at the time of these early
studies (Miller and Elderman, 1977).
Because earlier studies of the effects of oxidants;on the mixed-conifer
forest left many questions unanswered, a comprehensive interdisciplinary.study
involving scientists at the University of California at Berkeley and .Riverside
and at the U.S. Forest Service Pacific Southwest Experiment Station (and
financed by the U.S. Environmental Protection Agency) was begun in 1973 and
7-35
-------
terminated in 1978. This study is the most comprehensive and best-documented
report on the effects of oxidants on a forest ecosystem (Miller et al., 1982).
The study was designed to answer two questions (Miller et al., 1982):
(1) How do the organisms and biological processes of the conifer forest respond
to different levels of chronic oxidant exposure; and (2) how can these responses
be interpreted within an ecosystem context?
The major abiotic components studied were water (precipitation), tempera-
ture, light, mineral nutrients (soil substrate), and oxidant pollution. The
biotic components studied included producers (an assortment of tree species
and lichens), consumers (wildlife, insects, disease organisms), and decomposers.
The decomposers studied were the populations of saprophytic fungi responsible
for the decay of leaf and woody litter.
The research plan included study of limited aspects of the following
ecosystem processes: (1) carbon flow (the movement of carbon dioxide into the
plant, its incorporation into carbohydrates; and then its partitioning among
consumers, decomposers, litter, and the soil; and finally its return to the
atmosphere); (2) the movement of water in the soil-plant-atmosphere continuum;
(3) mineral nutrient flow through the green plant, litter, and soil-water
compartments; and (4) the shift in diversity patterns in time and space as
represented by changes in age, structure, and density in the composition of
tree species in communities.
In addition to the Rim Forest-Sky Forest station established in 1968, six
other monitoring stations were established along the mountain crests near the
vegetational study sites in order to characterize the east-to-west oxidant
gradient. Hourly average 0- concentrations for 1975 (measured by UV) indicated
that 0, buildup began around 10 a.m. and reached a maximum at all stations in
all months (May through September) at around 4 p.m. For example, at the Rim
Forest-Sky Forest Station where the highest concentrations were usually recorded,
the 1-month average of hourly values for May through September 1975 ranged
from 0.07 to 0.10 ppm at 10 a.m. and from 0.15 to 0.22 ppm at 4 p.m. The
highest concentrations occurred in June, July, and August, and the lowest for
the 5-month period occurred in September. The total number of hours with
concentrations of 0.08 ppm or more during June through September was never
less than 1300 hours per season during the first 7 years (1968 through 1974)
(Miller and Elderman, 1977).
7-36
-------
From 1973 through 1978, 24-hour ozone concentrations ranged from a back-
ground of 0.03 to 0.04 p'pm in the eastern part of the San Bernardino Mountains
to maxima of 0.10 to 0.12 ppm in the western part (Miller et al., 1982).
Monthly averages of daily maxima of oxidant concentrations for 1968 through
1972 were given in Figure 7-2. Annual trends of 0», precipitation, and tempera-
ture as measured May through September, 1972 through 1978, are shown in Figure
7-3 (Miller et al., 1982). In addition to total oxidant, PAN and N02 concen-
trations were measured at the Sky Forest station. Symptoms of PAN injury were
not distinguishable from 03 on herb-layer plant species, while N02 remained at
non-toxic concentrations.
The interdisciplinary studies indicated that the ecosystem components
most directly affected by 0, were the tree species, the fungal microflora of
tree needles, and foliose lichens growing on the bark of trees. Injury to or
changes in the functioning of the living components also affected, either
directly or indirectly, the ecosystem processes of carbon flow, mineral nutrient
cycling, and water movement; and also changed vegetational community patterns
(Miller et al., 1982).
Ponderosa pine was the most sensitive of the trees to 0,, with Jeffrey
pine, white fir, black oak, incense cedar, and sugar pine following in decreasing
order of sensitivity. Foliar injury on sensitive ponderosa and Jeffrey pine
was observed when the 24-hour-average ozone concentrations were 0.05 to 0.06 ppm
(Miller et al. , 1982). Injury, decline, and death of these species suggested
a chain of events that could lead to various levels of ecosystem changes,
depending on the ozone concentrations (Miller et al., 1982). Foliar injury,
premature senescence, and needle fall decreased the photosynthetic capacity of
stressed pines and reduced the production of carbohydrates needed for use in
growth and reproduction by the trees. Nutrient availability to the trees was
also reduced by their retention of smaller amounts of green foliage (Miller et
al., 1982). Decreased carbohydrate resulted in a decrease in radial growth
and in height of stressed trees (McBride et al., 1975; Miller and Elderman,
1977). Growth reductions attributable to oxidant air pollution were calculated
by McBride et al. (1975) for ponderosa pine saplings. Assuming 1910 to 1940
to be a period of low oxidant pollution and 1944 to 1974 a period of high
oxidant pollution, they used radial growth increments (dbh) to calculate an
oxidant-induced decrease in diameter of 40 percent. On the basis of the
7-37
-------
V)
UJ
_i
00
Ł
> 4.0
CO
CO
Z
o
oc
3.0
Z 2.0
ui
1.0
PRECIPITATION = mm x 103
OXIDANT = pphm-hr x 1O4
TEMPERATURE = °C x 10'
1973-1974
1974-1975
1975-1976
1976-1977
1977-1978
Figure 7-3. Annual trends of oxidant dose, precipitation, and air temperature near the
Lake Arrowhead-Sky Forest region of the San Bernardino National Forest, California.
Source: Miller etal. (1982).
-------
3-year growth of saplings in filtered and nonfiltered air in portable green-
houses, they calculated oxidant-induced reductions of 26 percent in height
growth (McBride et al., 1975). As noted also for the radial growth studies
cited in Section 7.4.1.2, standardized methods for measuring tree rings had
not been developed at the time of this study. Consequently, calculated
decreases in diameter would reflect uncertainties associated with radial
growth measurements. Calculated decreases in growth in height were based on
the assumption of a growth rate in non-stressed older trees equivalent to that
in non-stressed saplings.
Tree reproduction also .was influenced by a reduction in carbohydrate.
Injured ponderosa and Jeffrey pines older than 130 years produced significantly
fewer cones per tree than uninjured trees of the same age (Luck, 1980). Tree
ring analysis showed declines of ring width indices for many trees. In recent
years, however, stand thinning reversed the trend (Miller et al., 1982).
A comparison of lichen species found on conifers during the years 1976 to
1979 with collections from the early 1900's showed the presence of 50 percent
fewer species in the more recent period. Marked morphological deterioration
of the common species Hypogymnia enteromorpha was documented in areas of high
oxidant concentrations (Sigal and Nash, 1983).
Biotic interactions associated with predators, pathogens, and symbionts
were influenced by changes in the energy available to the trees. The decrease
in vigor and lack of ability to recover from ozone injury associated with
reduced carbohydrates made the ponderosa pines more susceptible to attack by
predators and pathogens (Stark and Cobb, 1969). Dahlsten and Rowney (1980)
have pointed out that oxidant-weakened pines can be killed by fewer western
pine beetles than are required to kill healthier trees. In stands with a high
proportion of 0~-injured trees, a given population of western pine beetles
could therefore kill more trees. James et al. (1980.a,b) observed that the
root rot fungus, H. annosum, increased more rapidly because freshly cut .stumps
and roots of weakened trees were more vulnerable to attack.
Carbon flow and mineral nutrient cycling were also influenced by litter
accumulation. Heavy litter accumulation occurred in stands with the most
severe needle injury and defoliation. The heaviest accumulation was beneath
trees with moderate damage rather than the most severe damage. Carbon and
mineral nutrients accumulated in the thick needle layer and influenced nutrient
availability because of potential losses by volatilization during fires and in
7-39
-------
subsequent surface runoff (Miller et al., 1982). Mineral nutrient cycling was
also influenced by the change in microflora of pine needles and its rela-
tionship with the decomposer community. Premature senescence and abscission
of pine needles alter the taxonomic diversity and population density of the
microflora that normally develop on needles during the time they are growing
on trees. The change in the types of fungi on needles and a decrease in their
numbers weaken the decomposer community and the rate of decomposition (Bruhn,
1980).
Litter decomposition rate may also be influenced by decreases in moisture
on the forest floor caused by a thinning of the canopy. A thinner canopy
allows more sunlight to reach the forest floor and dry the litter more rapidly,
thus potentially decreasing the rates of decomposition and of subsequent
nutrient cycling, and increasing litter depth. Pine seedling establishment is
expected to be hindered by deep litter but the establishment of oxidant-tolerant
over- and understory species is expected to be favored.
Existing data are inadequate for explaining how the complex interplay of
ozone injury, insects, diseases, and drought and, to an extent, fire, will
shape the age and species structure, in the future, of the tree communities
studied (Miller et al. , 1982). It is more prudent to propose a range of
possible changes in forest composition. In a worst-case example, the ozone-
incited stress on sensitive ponderosa and Jeffrey pine and, to a lesser extent
on sensitive white fir, black oak, incense cedar, and sugar pine, if accompanied
by fire, would bring about the removal of the pine forest overstory. In the
understory of ozone-weakened stands of ponderosa pine, the establishment of
ozone-tolerant species, particularly incense cedar, forms a fuel ladder that
threatens the survival of the overstory pines in the event of fire. At the
chaparral-forest boundary, a shift in dominance to self-perpetuating, fire-
adapted, ozone-tolerant shrub and oak species has occurred following thinning
of the overstory pines and has produced species mixtures that provide fewer
commodity and amenity values than the former forest (Miller et. al., 1982).
Many of the changes observed in the components, structure, and processes
of the San Bernardino forest ecosystems, attributed by a number of investigators
to ozone-oxidant stress, are consistent with the theories of Odum (1985) on
trends expected to occur in stressed ecosystems. The changes postulated by
Odum that have been observed in the San Bernardino mixed-conifer ecosystem
include: (1) low efficiency in converting energy to organic structure;
7-40
-------
(2) decreased nutrient cycling between trophic levels; (3) decreased nutrient
availability (via retention in litter); (4) increased proportion of r-strate-
gists (opportunistic species); (5) decreased size and decreased lifespans of
organisms or parts (e.g., needles); (6) reversal of autogenic successional
trends (succession reverts to earlier stages); and (7) decreased mutualism
(positive interactions) and increased parasitism (negative interactions).
7.6.2 The Blue Ridge Mountains of Virginia
Oxidant-induced injury on vegetation has been observed in the Appalachian
Mountains in the eastern United States for many years. Needle blight of
eastern white pine was first reported in the early 1900s but it was not until
1963 that it was shown to be the result of acute and chronic 0,, exposure
(Berry and Ripperton, 1963).
Despite early reports by Berry (1961, 1964) and by Berry and Ripperton
(1963), no concerted effort to determine the effects of ozone on the vegetation
of the Appalachian Mountains was made until Hayes and Skelly (1977) monitored
total oxidants and recorded oxidant-associated injury on eastern white pine in
three rural Virginia sites between April 1975 and March 1976. Injury was
associated with total oxidant peaks of 0.08 ppm or higher.
Increased injury symptoms were observed by Hayes and Skelly (1977) on
pine trees previously categorized as sensitive or intermediately sensitive
following 03 exposures. No injury was observed on trees categorized as toler-
ant. Hayes and Skelly (1977) suggested that continued exposure of sensitive
and intermediately sensitive white pine to acute and chronic oxidant concentra-
tions could ultimately influence their vegetative vigor and reproductive
ability. Inability to reproduce could result in replacement of the sensitive
pines by tolerant species.
More recent studies have reported oxidant-induced symptoms on other
indigenous forest tree species: tulip poplar, green ash (Fraxinus pennsylvanica,
Marsh), hickory (Carya spp.), black locust (Robinia pseudoacacia L.) hemlock
(Tsuga canadensis (L.) Carr.), and table mountain (Pinus pungens, Lamb),
Virginia (P. virginiana, Mill.), and pitch pine (P. rigida, Mill.) (Duchelle
et al., 1982). Monthly 8-hour average 0^ concentrations ranged from 0.035 to
0.065 ppm and peak hourly concentrations from 0.08 to 0.13 ppm (Skelly et al.,
1984) (Table 7-4). Sulfur dioxide concentrations ranged from undetectable to
0.03 ppm.
7-41
-------
TABLE 7-4. MONTHLY 8-hr AVERAGE (11:00 a.m. - 6:00 p.m. EST), MONTHLY AVERAGE
OF PEAK 1-hr, AND CUMULATIVE SEASONAL OZONE DOSES MONITORED AT BIG MEADOWS,
SHENANDOAH NATIONAL PARK, VIRGINIA, DURING 1979-1981
(ppm)
YEAR
1979
Month Average
January
February
March
April
May
June
July
August
September
October
November
December
Total ozone dosage
1 April -
30 September,
ppm-hr
0.041
u
0.062
0.055
0.052
0.055
0.047
0.054
0.046
0.042
0.039
0.028
73.38
Peak
0.06
-
0.10
0.10
0.08
0.08
0.09
0.07
0.09
0.08
0.07
0.05
1980
Average
0.020
0.024
0.035
0.042
0.048
0.059
0.058
0.051
0.046
0.033
0.035
0.041
74.22
Peak
0.03
0.04
0.04
0.07
0.10
0.09
0.09
0.10
0.09
0.08
0.06
0.06
1981
Average
0.033
0.030
0.028
0.043
0.037
0.023
0.037
0.030
0.037
0.044
0.055
-
50.51
Peak
0.06
0.06
0.05
0.07
0.08
0.06
0.10
0.06
0.06
0.09
0.08
-
Data not available.
Source: Duchelle et. al. (1983). .
Injury to herbaceous vegetation growing in the same areas was also observed
(Duchelle et al., 1983). Ambient 0, concentrations were shown to reduce growth
and productivity of graminoid and forb vegetation in the Shenandoah National
Park. For each year of the study, biomass production (weight of living tissue)
was greatest in filtered-air open-top chambers. The total 3-year cumulative
dry weight of the plants in the filtered chambers was significantly (<0.05)
greater than that of plants in non-filtered and open-air plots. Similar cumula-
tive 0- doses in 1979 and 1980 resulted in different percentage reductions in
biomass for the two years, .suggesting that variations in 0., dose during the
growing season may be more important than the cumulative 0- dose. Ozone inhibits
biomass production of natural vegetation. Reductions in biomass could be a
consequence of decreased root growth resulting from 0- exposure. Common milkweed
(Ascelepias syrica L.) and common blackberry (Rubus allegheniensis Porter) were
7-42
-------
the only two native species to develop visible injury. Milkweed has previously
been shown to be very sensitive to 0- (Duchelle and Skelly, 1981).
Ozone episodes lasting 1 to 3 days occurred several times each year
during the period of the study. Peak hourly concentrations, measured from
11:00 a.m. through 6:00 p.m., ranged from 0.08 to 0.10 ppm; however, daytime
ozone concentrations exceeding 0.06 ppm were recorded for 1218, 790, and 390
hours during 1979, 1980, and 1981, respectively. As noted in Section 7.1.4.2,
however, measurements of only daytime ozone concentrations may not capture the
true ozone maxima in areas affected by transport, or unique meteorological
conditions, or both. Concentrations of SOp ranged from <0.001 to 0.03 ppm and
were considered to have had no effects on vegetation (Duchelle et al., 1983).
As in California, ozone is transported to these sites from distant urban
and industrial sources. In the Blue Ridge and Appalachian Mountains, these
sources include the industrial midwest, eastern Virginia, and the Washington,
DC, area. Most of the episodes monitored were regional in nature. High 0-
concentrations occurred at the three monitoring sites simultaneously (Table
7-3, Skelly et al., 1984).
The effects of ozone on species composition and succession of natural
vegetation of the Virginia mountains were not studied; however, none of the
plant species shown to be injured by ozone plays a dominant role in the Blue
Ridge Mountain ecosystem. Therefore, the removal of any of these species would
probably not have the impact that the decline and death of ponderosa and
Jeffrey pine have had on the San Bernardino Forest ecosystem.
7.7 RESPONSES OF OTHER ECOSYSTEMS TO OZONE
7.7.1 Responses of Native Vegetation
No other natural ecosystem has been as thoroughly studied as the San
Bernardino National Forest. However, the same patterns of response to ozone
stress seen there have been observed in other locations (Section 7.3). Sensi-
tive individuals of various species were adversely affected by 0~. Photo-
synthesis was inhibited and reductions occurred in carbohydrate formation and
translocation, in biomass production, in growth, and in reproductive capacity.
The larger or dominant species were those most severely affected. Changes in
community structure were predicted for some of the vegetational communities on
the basis of observed effects.
7-43
-------
California
In southern California, the predominant native shrubland vegetation
consists of chaparral and coastal sage scrub. Chaparral occupies upper eleva-
tions of the coastal mountains and extends into the North Coast ranges, east
to central Arizona, and south to Baja California. Coastal sage scrub occupies
lower elevations of the coastal and interior slopes of ranges extending from
San Francisco to Baja California. Applying standard plant ordination tech-
niques, Westman (1979) found reduced cover of native species of coastal sage
scrub on some sites. The reduced cover was statistically correlated with
elevated levels of atmospheric oxidants. From the data records of nearby
monitoring stations, an annual average concentration of 0.18 ppm was calculated
for the 11 most polluted sites; the annual average concentration calculated
for the 11 least-polluted sites was 0.04 ppm. The effect of long-term, con-
tinued injury was to decrease foliar cover of vegetation and species richness
by favoring a few, tolerant species.
Stolte (1982) also studied chaparral species and their response to ozone
under both experimental and ambient air conditions. A large variation in
sensitivity to 0, from species to species of seedling chaparral plants was
observed; however, the majority appeared to be intermediate in sensitivity.
Ozone-sensitive chaparral seedlings can have reduced vigor and suffer higher
mortality. Stolte (1982) found that the composition and density of chaparral
stands are determined by seedling success of the dominant species and that
these species in turn influence the behavior of the stands during fires.
Composition and density of chaparral stands may be permanently altered, since
the post-fire seedling establishment of perennial dominants occurs chiefly
during the first year following fire.
Utah
Treshow and Stewart (1973) conducted one of the few studies concerned
with the impact of air pollution on native herbaceous species in natural plant
communities. The aim of the study was to determine the concentration of ozone
necessary to cause foliar injury to the most prevalent species in some of the
intermountain grassland, oak, aspen, and conifer communities. Seventy common
plant species indigenous to those communities were fumigated with ozone to
establish sensitivity. Injury was generally evident at concentrations above
0.15 ppm for 2 hours. Species found to be most sensitive to ozone in the
7-44
-------
grassland and aspen communities included some dominant species considered key
to community integrity. Bromus tectorum L. (cheatgrass), the most prevalent
species in the grassland community, was also the most sensitive. Severe
injury to this introduced annual resulted from a single 2-hour exposure to
0.15 ppm of ozone. Cheatgrass is widely distributed in the intermountain
western United States. Removal of this dominant species from plant communities
could result in a shift in dominance to another species. The significance of
such a change would depend on the species replacing cheatgrass. The other
grasses studied were not as sensitive to ozone, nor were the forbs (Table
7-5). The production of carbohydrates in visibly injured grasses, however,
was significantly reduced.
In the aspen community, the most dramatic example was aspen (Populus
tremuloides Michx.) itself. A single 2-hour exposure to 0.15 ppm ozone caused
severe symptoms on 30 percent of the foliage. Because white fir seedlings
require aspen shade for optimal juvenile growth, the authors suggested that
significant losses in aspen populations might restrict white fir development
and later forest succession; conversion to grasslands could occur. It was
apparent that in a natural community exposed to ozone, the tolerant species
would soon become the dominants. The authors concluded that ozone must be
considered a significant environmental parameter that influences the composi-
tion, diversity, and stability of natural plant communities and that it "may
ultimately play a major role in plant succession and dominance" (Treshow and
Stewart, 1973).
National Parks and National Recreation Areas
Vegetation in national parks other than the Shenandoah National Park is
apparently also being injured by ambient air pollutants. In a recent report
to Congress, the National Park Service (1985) stated that the preliminary
results of studies recently completed or currently under way in a number of
parks indicate that sensitive vegetation is being injured by 0, transported
into the parks. Vegetational injury from 03 has been observed in the Santa
Monica Mountains National Recreation Area; Sequoia, Kings Canyon, Shenandoah,
Great Smoky Mountains, and Acadia National Parks; Indiana Dunes National
Lakeshore; and Congaree Swamp National Monument Park. Maximum hourly average
ozone concentrations ranged from 0.11 ppm in the Great Smoky Mountains National
Park to 0.22 ppm in the Santa Monica Mountains. Sulfur dioxide concentrations
7-45
-------
TABLE 7-5. INJURY THRESHOLDS FOR 2-hr EXPOSURES TO OZONE
Species
Injury
threshold,
ppm 03
for 2 hr
Species
Injury
threshold,
ppm 03
for 2 hr
Grassland-oak community species:
Trees and shrubs:
Acer grandidentatum Nutt. >0.40
Acer negundo L. >0.25
Artemesia tridentata Nutt. 0.40
Mahonia repens G. Don >0.40
Potentilla fruticosa L. 0.30
Quercus gambelii Nutt. 0.25
Toxicodendron radicans (L.) Kuntze >0.30
Perennial forbs:
Achillea millefolium L. >0.30
Ambrosia psilostachya DC. >0.40
Calochortus nuttallii Torr. >0.40
Cirsium arvense (L.) Scop. 0.40
Com'urn maculatum L. >0.25
Hedysarum boreale Nutt. 0.15
Helianthus anuus L. >0.30
Medocago sativa L. 0.25
Rumex crispus L. 0.25
Urtica gracilis Ait. 0.30
Vicia americana Muhl. >0.40
Grasses:
Bromus brizaeformis Fisch & Mey. 0.30
Bromus tectorum L. 0.15
Poa pratensis L. 0.25
Aspen and conifer community species:
Trees and shrubs:
Abies concolor (Gord. & Glend.) Lindl. 0.25
Amelanchier alnifolia Nutt. 0.20
Pachystima myrsinites (Pursh) Raf. >0.30
Populus tremuloides Michx. 0.15
Ribes hudsonianum Richards 0.30
Rosa woodsii Lindl. >0.30
Sambucus melanocarpa A. Gray >0.25
Symphoricarpos vaccinioides Rydb. 0.30
Perennial forbs:
Acetaea arguta Nutt. 0.25
Agastache urticifolia (Benth.) Kuntz 0.20
Perennial forbs:
Alii urn acuminatum Hook 0.25
Angelica pinnata S. Wats. <0.25
Aster engelmanni (Eat.) A. Gray 0.15
Carex siccata Dewey 0.30
Cichorium intybus L. 0.25
Cirsium arvense (L.) Scop. <0.40
Epilobium angustifolium L. 0.30
Epilobium watsoni Barbey 0.30
Erigonum heraclioides Nutt. 0.30
Fragaria ovalis (Lehm.) Rydb. 0.30
Gentiana amarella L. >0.15
Geranium fremontii Torr. <0.25
Geranium richardsonii Fisch. & Traut. 0.15
Juncus sp. >0.25
Lathyrus lanzwertii Kell. >0.25
Lathyrus pauciflorus Fern. 0.25
Mertensia arizonica Greene 0.30
Mimulus guttatus DC. >0.25
Mimulus moschatus Dougl. <0.40
Mitel la stenopetala Piper >0.30
Osmorhiza occidental is Torr. 0.25
Phacelia heterophylla Pursh <0.25
Polemonium foliosissimum A. Gray 0.30
Rudbeckia occidentalis Nutt. 0.30
Saxifraga arguta D. Don <0.30
Senecio serra Hook. 0.15
Taraxacum officinale Wiggers >0.25
Thalictrum fendleri Engelm. >0.25
Veronica anagallis-aquatica L. 0.25
Vicia americana Muhl. >0.25
Viola adunca Sm. >0.30
Annual forbs:
Chenopodium fremontii Wats. <0.25
Callomia linearis Nutt. <0.25
Descurainia californica (Gray) 0.25
O.E. Shulz
Galium bifolium Wats. >0.30
Gayophytum racemosum T. & G. 0.30
Polygonum douglasii Greene >0.25
Grasses:
Agropyron caninum (L.) Beauv. >0.25
Bromus carinatus Hook. & Arn. <0.25
Source: Treshow and Stewart (1973).
7-46
-------
were below limits of detection in most of the parks, but were 0.01 ppm (maxi-
mum 3-hr) in the Shenandoah National Park and 0.25 ppm (maximum 3-hr) in the
Indiana Dunes National Lakeshore. The impact of the injured vegetation on
these ecosystems has yet to be appraised.
7.7.2 Managed Forest Ecosystems
Agricultural ecosystems are managed ecosystems that are manipulated to
maximize their yields for the benefit of humans. While their importance
cannot be overemphasized, agricultural ecosystems are not the only managed
ecosystems of importance in the United States.
The largest ecosystems managed for human use are the forests in the
National Forest System under the U.S. Department of Agriculture. They encompass
190 million acres, primarily in the west. The National Forest System provides
nearly one-fourth of the softwood timber used in the United States. Commercial
timber production is only one use of U.S. forestlands. Wildlife habitat,
rangeland, watershed protection, wilderness, and recreation are other uses.
These forests, if exposed to ozone, are potentially susceptible to the same
ozone-induced effects observed in the forest ecosystems of the San Bernardino
and Blue Ridge Mountains.
7.7.3 Aquatic Ecosystems
Terrestrial and aquatic ecosystems are closely interrelated. An adverse
impact on a forest ecosystem may in turn adversely affect adjacent aquatic
systems. A variety of linkages for energy and nutrient exchange exist. Air
pollution stress on terrestrial ecosystems often triggers dysfunctions, e.g.,
disruption of life cycles of aquatic insects, in neighboring aquatic ecosystems
such as streams, lakes, and reservoirs. Sediments resulting from erosion can
change the physical character of stream channels, causing changes in bottom
deposits, erosion of channel banks, obstruction of flow, and increased flooding.
They can fill in natural ponds and reservoirs. Finer sediments can reduce
water quality, affecting public and industrial water supplies and recreational
areas. For example, Westman (1977) has estimated that oxidant damage in the
San Bernardino Forest area could result in a cost of $27 million per year
(1973 dollars) for sediment removal, as long as the early successional stages
lasted, assuming sediment runoff to be equally partitioned among streets,
sewers, and debris basins. This estimate, however, was based on the assumption
7-47
-------
that forest fires would virtually devastate any oxidant-weakened stands,
resulting in almost total removal of vegetation.
Turbidity caused by increased erosion can also reduce the penetration of
light into natural waters. This, in turn, can reduce aquatic plant photosyn-
thesis and can lower the supplies of dissolved oxygen, leading to changes in
the natural flora and fauna (Bormann and Smith, 1980). Significant forest
alterations, therefore, may have a regional impact on nutrient cycling, soil
stabilization, sedimentation, and eutrophication of adjacent or nearby aquatic
systems. Interfacing areas, such as wetlands and bogs, may be especially
vulnerable to impact.
As noted in the San Bernardino studies, forest biomass reduction results
in a corresponding reduction in the total inventory of nutrient elements held
within a system. Loss of the dominant vegetation disrupts cycling pathways
and mechanisms of nutrient conservation. Research on the northern hardwood
forest has clearly established that retention of nutrients within a forest
ecosystem depends on constant and efficient cycling between the various com-
ponents of the intrasystem cycle and that deforestation impairs this retention
(Likens et al., 1977). Extensive nutrient loss can pollute downstream aquatic
resources, resulting in enrichment or eutrophication of a site, with long-term
consequences for potential plant growth, as well as contamination of urban
water sources.
7.8 ECONOMIC VALUATION OF ECOSYSTEMS
According to economic theory, the price of goods or services in the
marketplace represents the value that society places on those goods and ser-
vices. Free goods and services are often viewed in the marketplace as valueless
or simply as existing outside of the economy. Natural ecosystems provide free
public and private goods and services, but at present no agreement exists as
to the value of an ecosystem to society and the inherent values of natural
ecosystems have not been incorporated into any valuation system (Farnworth et
al., 1981).
The price of goods and services in the marketplace has some correspondence,
however minimal in some instances, to the cost of producing and offering those
goods and services in the marketplace. In the case of natural ecosystems,
however, human investment of energy and resources is quite low. Whereas
7-48
-------
agricultural ecosystems (as well as animal husbandry) require intensive manage-
ment and dollar expenditures for the production of marketable food items,
natural ecosystems commonly provide societal benefits (including some edible
foodstuffs) without the investment of appreciable direct dollar expenditures
or intensive management.
In an attempt to provide a framework for valuing ecosystems, Farnworth
et al. (1981) separated "value" into (1) market values of private goods (Value I)
and (2) non-market values of public goods and services. In turn, they separated
non-market values into attributable or assignable values (Value II) and intangi-
ble or non-assignable values (Value III). According to Farnworth et al.
(1981), political mechanisms, as opposed to the marketplace, are used to
assign a price or value to Value II items because society believes that the
value assigned by the marketplace is inadequate. Value III items are not
viewed by Farnworth et al. (1981) as having been incorporated into either
marketplace economics or political mechanisms. It might be noted here, however,
that the United States has indirectly placed a price or value on certain
Value II and III items, such as ecosystems, by allocating resources for the
abatement of air pollution thought to have potentially deleterious effects on
such items. Nevertheless, the apportionment of the costs (price or value) of
abating air pollution to the respective Value II and III items (as well as
many Value I items) remains unresolved.
Natural ecosystems, such as forests maintained as wilderness areas, may
offer products and services of little direct dollar value, but they provide
critically important, if unpriced, benefits to society. These benefits include,,
but are not restricted to: (1) maintenance of the global carbon balance and
the CL-CCL cycle; (2) soil stabilization (flood and erosion control); (3)
enhanced air and water quality; (4) nutrient conservation; (5) energy conser-
vation; (6) gene preservation; and (7) amenity and aesthetic functions, ranging
from tourism and birdwatching to white-water rafting and hunting (Smith, 1970;
Bormann, 1976; Westman, 1977; National Research Council, 1977; Hutchinson et
al., 1982).
Such goods and services, ranging from the critically essential to the
"nice but not necessary," are extremely difficult to quantify or to monetize
once quantified. Additional knowledge is needed in several important areas
before credible valuations of ecosystems can be made. First, better and more
complete information is needed on all the functions performed by ecosystems
7-49
-------
(Farnworth et al., 1981). Second, a framework must be constructed for convert-
ing benefits from or losses of goods and services from ecosystems to a value
system that permits comprehension of the true value of ecosystems. The costs
to society of stresses on ecosystems from ozone or any other manmade influence
will not be known and will almost certainly be underestimated unless such an
accounting system is developed (Risser, 1985). Third, additional and better
information is needed on the amount of chronic stress (e.g., ozone-oxidant
pollution) natural ecosystems can sustain and still retain ecosystem integrity;
on how long can they sustain stress and remain resilient, having the capacity
for self-repair; and on how much time is required for return to their original
state once such chronic stresses have been reduced or eliminated (see, e.g.,
West et al., 1980; cited in Taylor and Norby, 1985). At present, knowledge is
lacking on whether oxidant-stressed ecosystems are being damaged irreversibly.
The above information, at a minimum, is needed for the credible economic
valuation of natural ecosystems. Still further information is needed to
permit credible economic valuations of ozone-induced damage to ecosystems.
While areas requiring additional data are clear from a reading of this chapter,
the following two areas are especially obvious:
1. Better aerometric data for ecosystems suspected of being under
stress from ozone;
2. Measurement of additional variables in order to rule out significant
contributions to observed effects by temperature and other climatic
conditions, other airborne or water-borne pollutants, and biotic
agents (pathogens and other pests), as well as interactions among
and between biotic and non-pollutant abiotic factors (see, e.g.,
Manion, 1985; Cowling, 1985; Prinz, 1985; Smith, 1985, Mclaughlin,
1985; Taylor and Norby, 1985).
7.9 SUMMARY
7.9.1 Responses of Ecosystems to Ozone Stress
The responses to ozone of individual species and subspecies of herbaceous
and woody vegetation are well documented. They include (1) injury to foliage,
(2) reductions in growth, (3) losses in yield, (4) alterations in reproductive
capacity, and (5) alterations in susceptibility to pests and pathogens, espec-
ially "stress pathogens" (National Research Council, 1977; U.S. Environmental
Protection Agency, 1978; this document, Chapter 6). The responses elicited by
ozone in individual species and subspecies of primary producers (green plants)
7-50
-------
have potential consequences for natural ecosystems because effects that alter
the interdependence and interrelationships among individual components of
populations can, if the changes are severe enough, perturb ecosystems. Because,
however, of the numerous biotic and abiotic factors known to influence the
response of ecosystem components such as trees (see, e.g., Cowling, 1985;
Manion, 1985), it is difficult to relate natural ecosystem changes to ozone
specifically, and especially to ozone alone. Ozone can only be considered a
contributing factor.
Evidence indicates that any impact of ozone on ecosystems will depend on
the responses to ozone of the producer community. Producer species (trees and
other green plants) are of particular importance in maintaining the integrity
of an ecosystem, since producers are the source, via photosynthesis, of all
new organic matter (energy/food) added to an ecosystem. Any significant
alterations in producers, whether induced by ozone or other stresses, can
potentially affect the consumer and decomposer populations of the ecosystem,
and can set the stage for changes in community structure by influencing the
nature and direction of successional changes (Woodwell, 1970; Bormann, 1985),
with possibly irreversible consequences (see, e.g., Odum, 1985; Bormann,
1985).
7.9.2 Effects of Ozone on Producers
In forest ecosystems, tree populations are the producers. As such, they
determine the species composition, trophic relationships, and energy flow and
nutrient cycling of forest ecosystems (Ehrlich and Mooney, 1983). Ozone-induced
effects on the growth of trees has been clearly demonstrated in controlled
studies (see Chapter 6). For example, Kress and Skelly (1982) showed the
following reductions in growth in height in seedlings exposed to ozone for
6 hr/day for 28 days: American sycamore, 9 percent (0.05 ppm 0»); sweetgum,
29 percent (0.10 ppm O.J; green ash, 24 percent (0.10 ppm); willow oak,
19 percent (0.15 ppm 0,); and sugar maple, 25 percent (0.15 ppm). Similar
results have been obtained for other tree species by other investigators
(e.g., Dochinger and Townsend, 1979; Mooi, 1980; Patton, 1981; Kress et al.,
1982). Some species, however, have been shown to exhibit increased growth in
short-term ozone exposures (e.g., yellow poplar and white ash; Kress and
Skelly, 1982). Hogsett et al. (1985) found reductions in growth in height, in
radial growth, and in root growth in slash pine seedlings exposed for up to
7-51
-------
112 days to 7-hr seasonal mean concentrations of 0.104 ppm 0, (with a 1-hr
daily maximum of 0.126 ppm 0~) and 0.076 ppm 0- (with a 1-hr daily maximum of
0.094 ppm 03).
Field studies on the Cumberland Plateau (near Oak Ridge, TN) have shown
reductions in growth in eastern white pine exposed to ambient air 03 concentra-
tions >0.08 ppm (1-hr) (Mann et al., 1980), with 1-hr concentrations ranging
over the multi-year study from 0.12 ppm to 0.2 ppm (Mclaughlin et al., 1982).
It should be noted, however, that in the Mclaughlin et al. (1982) study trees
classified as ozone-tolerant sustained greater percentage reductions in radial
growth in the last 4 years (1976 to 1979) of the 1962 to 1979 period for which
growth was examined than the reductions observed in trees classified as ozone-
sensitive. In the Blue Ridge Mountains of Virginia, Benoit et al. (1982)
found reductions in radial growth of sensitive eastern white pine in a multi-
year study in which 1-hr 03 concentrations were generally 0.05 to 0.07 ppm but
peaked at :>0.12 ppm on as many as 5 consecutive days at a time.
The concentrations of ozone reported for sites on the Cumberland Plateau
and in the Blue Ridge Mountains may not fully represent the actual exposures
at those sites, however, since measurements were made in the daytime only.
For species in which stomates remain open at night, such as eastern white
pine, the possible occurrence of peak ozone concentrations at night, from
transported urban plumes, is an important consideration for accurately assessing
concentration-response relationships.
Exposures of trees and other producers to ozone have been shown to reduce
photosynthesis (e.g., Miller et al., 1969; Botkin et al., 1972; Barnes, 1972;
Carlson, 1979; Coyne and Bingham, 1981; Yang et al., 1983; Reich and Amundson,
1985) and to alter carbohydrate allocation, especially the partitioning of
photosynthate between roots and tops (e.g., Price and Treshow, 1972; Tingey
et al., 1976; Mclaughlin et al., 1982). Krause et al. (1984) have associated
growth reductions in ozone-exposed seedlings with foliar leaching. All three
of these effects have been postulated as mechanisms of the reduced growth seen
in ozone-exposed vegetation.
Responses to ozone are not uniform among plants of the same species and
the same approximate age. Differential responses have been attributed in part
to differences in genetic potential (e.g., Mann et al., 1980; Coyne and Bingham,
1981; Benoit et al., 1982). In addition, the age of the plant and its develop-
mental stage at time of exposure influence its response to ozone (see Chapter 6).
7-52
-------
Other factors, as well, influence the types and magnitude of plant responses
to ozone, including such macro- and microenvironmental factors as temperature,
relative humidity, soil moisture, light intensity, and soil fertility (see
Chapter 6).
Trees may respond rapidly to 0- stress. Needles of sensitive eastern
white pine usually exhibit injury symptoms within a few days after exposure to
high 0, concentrations. In other instances, responses are more subtle and may
not be observable for years because trees are perennials and must therefore
cope over time with the cumulative effects of multiple short- and long-term
stresses. Reductions in the growth of annual rings observed in ponderosa,
Jeffrey, and eastern white pine have been attributed to the exposure of the
trees to 0, over a period of 10 to 20 years (Miller and Elderman, 1977; Miller
et a!., 1982; Mclaughlin et al., 1982; Benoit et al., 1982). Decline and
dieback of red spruce in the northeastern United States and reduced growth
rates of red spruce, balsam fir, and Fraser fir in central West Virginia and
western Virginia also have been attributed to stresses, to which air pollution
is a possible contributor, that began at least 20 years ago (Johnson and
Siccama, 1983; Adams et al., 1985).
7.9.3 Effects of Ozone on Other Ecosystem Components and on Ecosystem
Interactions
Evidence for the effects of ozone on other ecosystem components indicates
that most are indirect, occurring chiefly as a result of the direct effects of
ozone on trees and other producers. Significant alterations in producer
species can change the ability of a species to compete and thus can influence
the nature and direction of successional changes in the ecosystem. Likewise,
significant alterations in producers can result in changes in the consumer and
decomposer populations that depend on producers as their food source. Studies
in the San Bernardino Mountain ecosystems in the 1970s have provided some
evidence of successional shifts and of predisposition to infestation by pests
and pathogens as the result of oxidant-induced changes in ponderosa and Jeffrey
pines (see Section 7.9.4 below).
Marked morphological deterioration of the common lichen species, Hypogymm'a
enteromorpha, was documented in areas of the San Bernardino Mountains having
high oxidant concentrations. A comparison of the species of lichens found
growing on ponderosa and Jeffrey pine with collections from the early 1900's
indicated the presence of 50 fewer species (Sigal and Nash, 1983).
7-53
-------
McCool et al. (1979) and Parmeter et al. (1962) reported decreases in
mycorrhizal infections and rootlets in ozone-stressed citrange (a citrus
hybrid) and ponderosa pine, respectively. Mahoney (1982), on the other hand,
found no evidence of impairment in the development of mycorrhizal associations
in loblolly pine seedlings exposed to ozone plus sulfur dioxide; however,
shoot dry weight was decreased by 12 percent.
The effects of ozone on mycorrhizae are of particular note here, since
mycorrhizae are essential for the optimal development of most plants because
of the functions they perform. Mycorrhizal fungi increase the solubility of
minerals, improve the uptake of nutrients for host plants, protect roots
against pathogens, produce plant growth hormones, and move carbohydrates from
one plant to another (Hacskaylo, 1972). Ozone may disrupt the association
between mycorrhizal fungi and plants, possibly by inhibiting photosynthesis
and reducing the amounts of sugars and carbohydrates available for transfer
from leaves of producers to the roots. Mycorrhizae are known to be sensitive
to alterations in carbon allocation to the roots in host plants (Hacskaylo,
1973).
Because of the complex interactions among plants, pests, pathogens, and
other biotic and abiotic factors, Laurence and Weinstein (1981) have emphasized
the critical importance of examining pollutant-pathogen and pollant-insect
interactions in determining the growth impact of a pollutant. Manion (1985)
has emphasized the necessity of taking non-pollutant stresses, both biotic and
abiotic, into account when attempting to attribute changes in forest ecosystems
to air pollutants.
7.9.4 Effects of Ozone on Specific Ecosystems
One of the most thoroughly studied ecosystems in the United States is the
mixed-conifer forest ecosystem in the San Bernardino Mountains of southern
California. Sensitive plant species there began showing injury in the early
1950's (Miller and Elderman, 1977) and the source of the injury was identified
as oxidants (ozone) in 1962 (Miller et al., 1963). In an inventory begun in
1968, Miller found that sensitive ponderosa and Jeffrey pines were being
selectively removed by oxidant air pollution. Mortality of 8 and 10 percent
was found in two respective populations of ponderosa pine studied between 1968
and 1972. Monitoring in that period showed ozone concentrations >0.08 ppm for
>1300 hours, with concentrations rarely decreasing below 0.05 ppm at night
near the crest of the mountain slope (Miller, 1973).
7-54
-------
In a subsequent interdisciplinary study (1973 through 1978), biotic and
abiotic components and ecosystem processes were examined. The ecosystem
components most directly affected were various tree species, the fungal micro-
flora of needles, and the foliose lichens on the bark of trees. In May through
September, 1973 through 1978, 24-hr-average ozone concentrations ranged from
about 0.03 to 0.04 ppm to about 0.10 to 0.12 ppm. (Monitoring was done by the
Mast meter through 1974 and by the UV method from 1975 through 1978). Foliar
injury on sensitive ponderosa and Jeffrey pine was observed when the 24-hr-
average ozone concentrations were 0.05 to 0.06 ppm (Miller et al., 1982).
Injury, decline, and death of these species were associated with the major
ecosystem changes observed (Miller et al., 1982).
Growth reductions attributable to oxidant air pollution were calculated
by McBride et al. (1975) for ponderosa pine saplings. Assuming 1910 to 1940
to be a period of low oxidant pollution and 1944 to 1974 a period of high
oxidant pollution, they used radial growth increments (dbh) to calculate an
oxidant-induced decrease in diameter of 40 percent. On the basis of the
3-year growth of saplings in filtered and nonfiltered air in portable green-
houses, they calculated oxidant-induced reductions of 26 percent in height
growth (McBride et al., 1975). No standardized methods for determining tree
ring widths were available at the time of this study.
Carbon flow and mineral nutrient cycling were influenced by the accumula-
tion of litter under stands with the most severe needle injury and by defolia-
tion, as well as by a reduction in the number of species and the population
density of the fungi that normally colonize living needles and later participate
in decomposition. The most likely result of heavy litter accumulation is a
reduction in pine seedling establishment and greater establishment and growth
of oxidant-tolerant understory species on some sites and oxidant-tolerant
trees on other sites (Miller et al., 1982).
Changes in the energy available to trees influenced the biotic interac-
tions, so that weakened ponderosa pines were more susceptible to attack by
predators such as bark beetles and to pathogens such as root rot fungi (Stark
and Cobb, 1969). Fewer western pine beetles were required to kill weakened
trees (Dahlsten and Rowney, 1980); and stressed pines became more susceptible
to root rot fungi (James et al., 1980b) and showed a decrease in mycorrhizal
rootlets and their replacement by saprophytic fungi (Parmeter et al., 1962).
7-55
-------
Accelerated rates of mortality of ponderosa and Jeffrey pine in the
forest overstory, resulting from 0- injury, root rot, and pine beetle attack,
and in some cases, removal by fire, changed the basic structure of the forest
ecosystem (Phase IV, Table 7-1; Bormann, 1985) by causing replacement of the
dominant conifers with self-perpetuating, fire-adapted, 0--tolerant shrub and
oak species, which are considered less beneficial than the former pine forest
and which inhibit reestablishment of conifers (Miller et al., 1982).
Injury to vegetation in other ecosystems has also been reported. Duchelle
et al. (1983) found reductions in the growth and productivity of graminoid and
forb vegetation in the Shenandoah National Park, where 1-hr ozone concentra-
tions ranged from 0.08 to 0.10 ppm in the 3-year study period, with 1-hr
concentrations >0.06 ppm occurring for 1218, 790, and 390 hours in 1979, 1980,
and 1981, respectively. Treshow and Stewart (1973) fumigated species that
grow in the Salt Lake Valley and the Wasatch Mountains in Utah and found key,
dominant species to be ozone-sensitive. The National Park Service (1985) has
recently reported ozone-induced injury to vegetation in the Santa Monica
Mountains National Recreational Area, the Sequoia and Kings Canyon National
Parks, Indiana Dunes National Lakeshore, Great Smoky Mountains National Park,
and the Congaree Swamp National Monument. The impact of injury to vegetation
in these ecosystems has not been appraised.
It should be emphasized that the relative importance of a given species
in a given ecosystem must be considered in any assessment of the impact of
ozone (or other stresses) on an ecosystem. Ozone has not had the impact on
other ecosystems that it has had on the San Bernardino mixed-conifer forest
because the plant species injured do not have a role equal in importance to
the role of ponderosa and Jeffrey pines in the San Bernardino ecosystem.
7.9.5 Economic Valuation of Ecosystems
At the present time, economists and ecologists remain unable to devise a
mutually acceptable framework for estimating the economic value of ecosystems.
In addition, the credibility of any attempt to estimate at present the economic
value of ecosystems would be diminished by a lack of scientific data (1) on
the time-course of the manifestation of stress-induced effects on ecosystems,
(2) on the point at which ecosystems lose the capacity for self-repair, and
(3) on the points at which they begin to lose their ability to provide, respec-
tively, priced and unpriced benefits to society. In addition, estimation of
7-56
-------
the economic losses that might be associated with the specific effects of
ozone on ecosystems requires other data that are presently in short supply,
i.e., better and more aerometric data and better and more data on additional
variables, so that significant contributions from abiotic factors other than
ozone, as well as from biotic factors, can be credibly estimated.
7.10 REFERENCES
Adams, H. S. ; Stephenson, S. L.; Biasing, T. J.; Duvick, D. N. (1985) Growth-
trend declines of spruce and fir in mid-Appalachian subalpine forests.
Environ, and Exptl. Bot. 25: 315-325.
Alstad, D. N. ; Edmunds, G. F., Jr.; Weinstein, L. H. (1982) Effects of air
pollutants on insect populations. Annu. Rev. Entomol. 27: 369-384.
Barnes, R. L. (1972) Effects of chronic exposure to ozone on photosynthesis
and respiration of pines. Environ. Pollut. 3: 133-138.
Benoit, L. F.; Skelly, J. M. ; Moore, L. D.; Dochinger, L. S. (1982) Radial
growth reductions of Pinus strobus L. correlated with foliar ozone sensi-
tivity as an indicator of ozone-induced losses in eastern forests. Can.
J. For. Res. 12: 673-678.
Berry, C. R. (1961) White pine emergence tipburn, a physiogenic disturbance.
Asheville, NC: U. S. Dept. of Agriculture - Forest Service, Southeastern
Forest Experiment Station; station paper no. 130; pp. 1-8.
Berry, C. R. (1964) Differences in concentrations of surface oxidant between
valley and mountaintop conditions in the southern Appalachians. J. Air
Pollut. Control Assoc. 14: 238-239.
Berry, C. R.; Ripperton, L. A. (1963) Ozone, a possible cause of white pine
emergence tipburn. Phytopathology 53: 552-557.
Billings, W. D. (1978) Plants and the ecosystem. 3rd ed. Belmont, CA: Wadsworth
Publishing Company, Inc., pp. 1-62, 83-108.
Blum, U. T.; Tingey, D. T. (1977) A study of the potential ways in which ozone
could reduce root growth and nodulation of soybean. Atmos. Environ.
11: 737-739.
Bormann, F. H. (1976) An inseparable linkage: Conservation of natural ecosystems
and the conservation of fossil energy. BioScience 26: 754-760.
Bormann, F. H. (1985) Air pollution and forests: an ecosystem perspective.
BioScience 35: 434-441.
7-57
-------
Bormann, F. H.; Smith, W. H. (1980) Effects of air pollution on forest ecosys-
tems. In: Energy and the fate of ecosystems. Washington, DC: National
Academy of Sciences; pp. 308-318.
Botkin, D. B.; Smith, W. H.; Carlson, R. W. ; Smith, T. L. (1972) Effects of
ozone on white pine saplings: variation in inhibition and recovery of net
photosynthesis. Environ. Pollut. 3: 273-289.
Brown, V. K. (1984) Secondary succession: insect-plant relationships. BioScience
34:710-716.
Bruhn, J. N. (1980) Effects of oxidant air pollution on ponderosa and Jeffrey
pine foliage decomposition [Ph.D. dissertation]. Berkeley, CA: University
of California. Available from: University Microfilms, Ann Arbor, MI;
publication no. 82-00036.
Carlson, R. W. (1979) Reduction in the photosynthetic rate of Acer, Quercus,
and Fraxinus species caused by sulfur dioxide and ozone. Environ. Pollut.
18: 159-170.
Cobb, F. W. , Jr.; Stark, R. W. (1970) Decline and mortality of smog-injured
ponderosa pine. J. For. 68: 147-149.
Cobb, F. W., Jr.; Kinstic, M.; Zavarin, E.; Barber, H. W., Jr. (1968) Inhibi-
tory effects of volatile oleoresin components on Fomes annosus and four
Ceratocystis species. Phytopathology 58: 1327-1337.
Costonis, A. C.; Sinclair, W. A. (1972) Susceptibility of healthy and ozone-
injured needles of Pinus strobus to invasion by Lophodermium pinastri and
Aureobasidium pullalans. Eur. J. For. Pathol. 2: 65-73.
Cowling, E. B. (1985) Effects of air pollution on forests: Critical review
discussion papers. J. Air Pollut. Control Assoc. 35: 916-919.
Cox, G. W. ; Atkins, M. D. (1979) Agricultural ecology: an analysis of world
food production systems. San Francisco, CA: W. H. Freeman and Co.; pp.
35-38.
Coyne, P. E. ; Bingham, G. E. (1981) Comparative ozone dose response of gas
exchange in a ponderosa pine stand exposed to long-term fumigations. J.
Air Pollut. Control Assoc. 31: 38-41.
Dahlsten, D. L. ; Rowney, D. L. (1980) Influence of air pollution on population
dynamics of forest insects and on tree mortality. In: Miller, P. R., ed.
Proceedings of the symposium on effects of air pollutants on Mediterran-
ean and temperate forest ecosystems; June; Riverside, CA. Berkeley, CA:
U.S. Department of Agriculture; Forest Service general technical report
no. PSW-43; pp. 125-130. Available from: NTIS, Springfield, VA; PB81-133720.
Davis, D. D.; Coppolino, J. B. (1974) Relative ozone susceptibility of selected
woody ornamentals. HortScience 9: 537-539.
Davis, D. D. ; Coppolino, J. B. (1976) Ozone susceptibility of selected woody
shrubs and vines. Plant Dis. Rep. 60: 876-878.
7-58
-------
Davis, D. D. ; Wood, F. A. (1972) The relative susceptibility of eighteen
coniferous species to ozone. Phytopathology 62: 14-19.
Dewey, J. E.; Ciesla, W. M.; Meyer, H. E. (1974) Insect defoliation as a
predisposing agent to bark beetle outbreak in eastern Montana. Environ.
Entomol. 3: 722.
Dochinger, L. S.; Townsend, A. M. (1979) Effects of roadside deicer salts and
ozone on red maple progenies. Environ. Pollut. 19: 229-237.
Duchelle, S. F. ; Skelly, J. M. (1981) Response of common milkweed to oxidant
air pollution in the Shenandoah National Park in Virginia. Plant Dis. 65:
661-662.
Duchelle, S. F.; Skelly, J. M.; Chevone, B. I. (1982) Oxidant effects on forest
tree seedling growth in the Appalachian Mountains. Water, Air, Soil Pollut.
18: 363-373.
Duchelle, S. F.; Skelly, J. M. ; Sharick, T. L.; Chevone, B. I.; Yang, Y. S.;
Nellessen, J. E. (1983) Effects of ozone on the productivity of natural
vegetation in a high meadow of the Shenandoah National Park of Virginia.
J. Environ. Manage. 17: 299-308.
Ehrlich, P. R.; Mooney, H. A. (1983) Extinction, substitution, and ecosystem
services. BioScience 33: 248-254.
•
Fankhauser, R. K. (1976) Ozone levels in the vicinity of 33 cities. J. Air
Pollut. Control Assoc. 26: Ill-Ill.
Farnworth, E. G. ; Tidrick, T. H.; Jordan, C. F.; Smathers, W. M., Jr. (1981)
The value of natural ecosystems: an economic and ecological framework.
Environ. Conserv. 8: 275-282.
Gerdemann, J. W. (1974) Mycorrhizae. In: Carson, E. W. , ed. The plant root
and its environment: proceedings of an institute sponsored by the Southern
Regional Education Board; July 1971; Charlottesville, VA: University Press
of Virginia; pp. 205-217.
Gorham, E.; Vitousek, P. M.; Reiners, W. A. (1979) The regulation of chemical
budgets over the course of terrestrial ecosystem succession. Ann. Rev.
Ecol. System. 10: 53-84.
Guderian, R.; Kueppers, K. (1980) Response of plant communities to air pollution.
In: Miller, P. R. , ed. Proceedings of the symposium on effects of air
pollutants on Mediterranean and temperate forest ecosystems; June; Riverside,
CA. Berkeley, CA: U.S. Department of Agriculture; Forest Service general
technical report no. PSW-43; pp. 187-199. Available from: NTIS,
Springfield, VA; PB81-133720.
Hacskaylo, E. (1972) Mycorrhizae: the ultimate in reciprocal parasitism?
BioScience 22: 577-583.
7-59
-------
Hacskaylo, E. (1973) The Torrey symposium on current aspects of fungal develop-
ment; IV. Dependence of mycorrhizal fungi on hosts. Bull. Torrey Bot.
Club 100: 217-223.
Harward, M. R.; Treshow, M. (1973) Impact of ozone on the growth and reproduc-
tion of understory plants in the aspen zone of western U.S.A. Environ.
Conserv. 2: 17-23.
Hayes, E. M.; Skelly, J. M. (1977) Transport of ozone from the northeast U.S.
into Virginia and its effect on eastern white pines. Plant Dis. Rep. 61:
778-782.
Hibben, C. R.; Stotsky, G. (1969) Effects of ozone on the germination of fungus
spores. Can. J. Microbiol. 15: 1187-1196.
Hogsett, W. E.; Plocher, M.; Wildman, V.; Tingey, D. T.; Bennett, I. P. (1985)
Growth response of two varieties of slash pine seedlings to chronic ozone
exposures. Can. J. Bot. 63: 2369-2376.
Hutchinson, B. A.; Taylor, F. G.; Wendt, R. L. (1982) Use of vegetation to
ameliorate building microclimates: An assessment of energy conservation
potentials. Publica. No. ORNL/CON-87, Oak Ridge Nat. Labor., Oak Ridge,
TN. 86 pp.
James, R. L. ; Cobb, F. W. , Jr.; Miller, P. R.; Parmeter, J. R., Jr. (1980a)
Effects of oxidant air pollution on susceptibility of pine roots to Fomes
annosus. Phytopathology 70: 560-563.
James, R. L.; Cobb, F. W., Jr.; Wilcox, W. W.; Rowney, D. L. (1980b) Effects of
photochemical oxidant injury of ponderosa and Jeffrey pines on suscepti-
bility of sapwood and freshly cut stumps to Fomes annosus. Phytopathology
70: 704-708.
Johnson, A. H. ; Siccama, T. G. (1983) Acid deposition and forest decline.
Environ. Sci. Technol. 17: 294A-305A.
Jordan, C. F.; Medina, E. (1977) Ecosystem research in the tropics. Ann. Mo. Bot.
Garden 64: 737-745.
Karnosky, D. F. (1981) Chamber and field evaluations of air pollution tolerances
of urban trees. J. Arboricult. 7: 99-105.
Karnosky, D. F. (1983) Variation in air pollution responses of hardwood trees.
Aquilo. Ser. Bot. 19: 183-188.
Kolb, J. A.; White, M. (1974) Small mammals of the San Bernardino Mountains,
California. Southwestern Naturalist 19: 112-114.
Krause, G. H. M.; Prinz, B.; Jung, K. D. (1984) Forest effects in West Germany.
In: Davis, D. D. ; Millen, A. A.; Dochinger, L., eds. Air pollution and
the productivity of the forest: proceedings of the symposium; October
1983; Washington, DC. Arlington, VA: Izaak Walton League of America; pp.
297-332.
7-60
-------
Kress, L. W.; Skelly, J. M. (1982) Response of several eastern forest tree
species to chronic doses of ozone and nitrogen dioxide. Plant Dis. 66:
1149-1152.
Kress, L. W.; Skelly, J. M.; Hinkelmann, K.H. (1982) Growth impact of 03, N02,
and/or S02 on Platanus occidental is. Agric. Environ. 7: 265-274.
Krupa, S.; Fries, N. (1971) Studies on ectomycorrhizae of pine. I. Production
of volatile organic compounds. Can. J. Bot. 49: 1425-1431.
Laurence, J. A.; Weinstein, L. H. (1981) Effects of air pollutants on plant
productivity. Annu. Rev. Phytopathol. 19: 257-271.
Letchworth, M. B. ; Blum, U. (1977) Effects of acute ozone exposure on growth,
nodulation, and nitrogen content of ladino clover. Environ. Pollut.
14: 303-312.
Levy, R.; Chui, Y. J.; Cromroy, H. L. (1972) Effects of ozone on three species of
Diptera. Environ. Entomol. 1: 608-611. Cited in Alstad et al. (1982).
Levy, R.; Jouvenaz, D. P.; Cromroy, H. L. (1974) Tolerance of three species of
insects to prolonged exposures to ozone. Environ. Entomol. 3: 184-185.
Cited in Alstad et al. (1982).
Likens, G. E.; Bormann, F. H.; Pierce, R. S.; Eaton, J. S.; Johnson, N. M.
(1977) Biogeochemistry of a forested ecosystem. New York, NY: Springer-
Verlag.
Luck, R. F. (1980) Impact of oxidant air pollution on ponderosa and Jeffrey
pine cone production. In: Miller, P.R., ed. Proceedings of symposium on
effects of air pollutants on Mediterranean and temperate forest ecosystems;
June; Riverside, CA. Berkeley, CA: U.S. Department of Agriculture, Forest
Service; p. 240; general technical report no. PSW-43. Available from:
NTIS, Springfield, VA; PB81-133720.
Mahoney, M. J.; Chevone, B. I.; Skelly, J. M.; Moore, L. D. (1985) Influence
of mycorrhizae on the growth of loblolly pine seedlings exposed to ozone
and sulfur dioxide. Phytopath. 75: 679-682.
Manion, P. D. (1985) Effects of air pollution on forests: Critical review dis-
cussion papers. J. Air Pollut. Control Assoc. 35: 919-922.
Mann, L. K. ; McLaughlin, S. B. ; Shriner, D. S. (1980) Seasonal physiological
responses of white pine under chronic air pollution stress. Environ. Exp.
Bot. 20: 99-105.
Maser, C.; Trappe, J. M.; Nussbaum, R. A. (1978) Fungal-small mammal interrela-
tionships with emphasis on Oregon coniferous forests. Ecology 59: 799-809.
McBride, J. R. ; Semion, V.; Miller, P. R. (1975) Impact of air pollution on
the growth of ponderosa pine. Calif. Agric. 29: 8-10.
McClenahen, J. R. (1978) Community changes in a deciduous forest exposed to
air pollution. Can. J. For. Res. 8: 432-438.
7-61
-------
McClenahen, J. R. (1984) Air pollutant effects on forest communities. In:
Davis, D. D. ; Millen, A. A.; Dochinger, L., eds. Air pollution and the
productivity of the forest: proceedings of the symposium; October 1983;
Washington, DC. Arlington, VA: Izaak Walton League of America; pp. 83-94.
McCool, P. M.; Menge, J. A.; Taylor, 0. C. (1979) Effects of ozone and HC1 gas on
the development of the mycorrhizal fungus Glomus fasciculatus and growth of
'Troyer1 citrange. J. Amer. Soc. Hort. Sci. 104: 151-154.
Mclaughlin, S. B. (1985) Effects of air pollution on forests: A critical review.
J. Air Pollut. Control Assoc. 35: 512-534.
Mclaughlin, S. B. ; Shriner, D. S. (1980) Allocation of resources to defense
and repair. In: Horsfall, J. B.; Cowling, E. B., eds. Plant Disease,
Vol. 5. New York; Academic Press, pp. 407-431.
Mclaughlin, S. B.; McConathy, R. K.; Duvick, D.; Mann, L. K. (1982) Effects of
chronic air pollution stress on photosynthesis, carbon allocation, and
growth of white pine trees. For. Sci. 28: 60-70.
Millar, C. S. (1974) Decomposition of coniferous leaf litter. In: Dickinson,
C. H.; Pugh, G. J. F., eds. Biology of plant litter decomposition; v. I.
New York, NY: Academic Press, Inc.; pp. 105-128.
Miller, P. L. (1973) Oxidant-induced community change in a mixed conifer
forest. In: Naegele, J. A., ed. Air pollution damage to vegetation.
Washington, DC: American Chemical Society; pp. 101-117. (Advances in
chemistry series: no. 122.)
Miller, P. R.; Elderman, M. J., eds. (1977) Photochemical oxidant air pollu-
tant effects on a mixed conifer forest ecosystem: a progress report,
1976. Corvallis, OR: U.S. Environmental Protection Agency; EPA report no.
EPA-600/3-77-104. Available from: NTIS, Springfield, VA; PB-274531.
Miller, P. R. ; Millecan, A. A. (1971) Extent of oxidant air pollution damage
to some pines and other conifers in California. Plant Dis. Rep. 55: 555-559.
Miller, P. R. ; Parmeter, J. R., Jr.; Taylor, 0. C.; Cardiff, E. A. (1963)
Ozone injury to the foliage of Pinus ponderosa. Phytopathology 53: 1072-
1076.
Miller, P. R.; Parmeter, J. R., Jr.; Flick, B. H.; Martinez, C. W. (1969)
Ozone dosage response of ponderosa pine seedlings. J. Air Pollut. Control
Assoc. 19: 435-438.
Miller, P. R. ; Longbotham, G. J.; Van Doren, R. E. ; Thomas, M. A. (1980)
Effect of chronic oxidant air pollution exposure on California black oak
in the San Bernardino Mountains. In: Plumb, T. R., ed. Proceedings of the
symposium on the ecology, management, and utilization of California oaks;
June 1979; Claremont, CA. Berkeley, CA: Pacific Southwest Forest and
Range Experiment Station; Forest Service general technical report no.
PSW-44. Available from: NTIS, Springfield, VA; PB81-248031.
7-62
-------
Miller, P. R.; Taylor, 0. C.; Wilhour, R. G. (1982) Oxidant air pollution
effects on a western coniferous forest ecosystem. Corvallis, OR: Corvallis
Environmental Research Laboratory; EPA report no. EPA-600/D-82-276.
Available from: NTIS, Springfield, VA; PB83-189308.
Mooi, J. (1980) Influence of ozone on the growth of two poplar cultivars.
Plant Dis. 64: 772-773.
National Park Service (1985). Testimony of the National Park Service before the
Subcommittee on Parks and Recreation. Washington, D.C.: U.S. House of
Representatives, Committee on Interior and Insular Affairs; 99th Congress,
1st session.
National Research Council (1977). Ozone and other photochemical oxidants.
Washington, D.C.: National Academy of Sciences; pp. 437-585.
Newman, J. R. (1979) Effects of industrial air pollution on wildlife. Biol.
Conserv. 15: 181-190.
Odum, E. P. (1969) The strategy of ecosystem development. Science (Washington,
DC) 164: 262-270.
Odum, E. P. (1971) Fundamentals of ecology. 3rd Ed. Philadelphia, PA: W. B.
Saunders Company; pp. 1-38, 106-136.
Odum, E. P. (1985) Trends expected in stressed ecosystems. BioScience. 35: 419-422.
Ovington, J. D. (1957) Dry-matter production by Pinus sylvestris L. Ann. Bot.
(London) 21: 287-314.
Parmeter, J. R., Jr.; Bega, R. V.; Neff, T. (1962) A chlorotic decline of
ponderosa pine in southern California. Plant Dis. Rep. 46: 269-273.
Patton, R. L. (1981) Effects of ozone and sulfur dioxide on height and stem
specific gravity of Populus hybrids. For. Sci. Res. Pop. NE-471.
Price, H.; Treshow, M. (1972) Effects of ozone on the growth and reproduction
of grasses. In: Proceedings of the international air pollution conference;
Melbourne, Australia, pp. 275-280.
Prinz, B. (1985) Effects of air pollution on forests: Critical review discussion
papers. J. Air Pollut. Control Assoc. 35: 913-915.
Reich, P. B.; Amundson, R. G. (1985) Ambient levels of ozone reduce net photo-
synthesis in tree and crop species. Science. 230: 566-570.
• Rhoads, A.; Markov, R. ; Brennan, E. (1980) Trees and shrubs relatively insensi-
tive to oxidant pollution in New Jersey and Southeastern Pennsylvania.
Plant Dis. 64: 1106-1108.
Rishbeth, J. (1951) Observations on the biology of Fomes annosus, with parti-
cular reference to East Anglian pine plantations; III. Natural and experi-
mental infection of pines, and some factors affecting severity of the
disease. Ann. Bot. (London) 15: 221-246.
7-63
-------
Risser, P. G. (1985) Toward a holistic management perspective. BioScience.
35: 414-418.
Roose, M. L. ; Bradshaw, A. D. ; Roberts, T. M. (1982) Evolution of resistance
to gaseous air pollutants. In: Unsworth, M. H. ; Ormrod, D. P., eds.
Effects of gaseous air pollution in agriculture and horticulture. London,
United Kingdom: Butterworth Scientific; pp. 379-409.
Rudinsky, J. A. (1962) Ecology of Scolytidae. Annu. Rev. Entomol. 7: 327-348.
Sigal, L. L.; Nash, T. H. (1983) Lichen communities on conifers in southern
California mountains: an ecological survey relative to oxidant air pollu-
tion. Ecology 64: 1343-1354.
Skelly, J. M. (1980) Photochemical oxidant impact on Mediterranean and temper-
ate forest ecosystems: real and potential effects. In: Miller, P. R., ed.
Proceedings of the symposium on effects of air pollutants on Mediterranean
and temperate forest ecosystems; June; Riverside, CA. Berkeley, CA: U.S.
Department of Agriculture, Forest Service; general technical report no.
PSW-43; pp. 38-50. Available from: NTIS, Springfield, VA; PB81-133720.
Skelly, J. M.; Yang, Y.-S.; Chevone, B. J.; Long, S. J. (1984) Ozone concen-
trations and their influence on forest species in the Blue Ridge Mountains
of Virginia. In: Davis, D. D.; Millen, A. A.; Dochinger, L., eds. Air
pollution and the productivity of the forest: proceedings of the symposium;
October 1983; Washington, DC. Arlington, VA: Izaak Walton League of
America; pp. 143-159.
Smith, W. H. (1970) Technical Review: Trees in the city. Jour. Amer. Inst.
Planners. 36: 429-435.
Smith, R. L. (1980) Ecology and field biology. 3rd ed. New York, NY: Harper
and Row; pp. 11-199.
Smith, W. H. (1980) Air pollution—a 20th century allogenic influence on
forest ecosystems. In: Miller, P. R., ed. Proceedings of the symposium on
effects of air pollutants on Mediterranean and temperate forest ecosystems;
June; Riverside, CA. Berkeley, CA: U.S. Department of Agriculture, Forest
Service; general technical report no. PSW-43; pp. 79-87. Available from:
NTIS, Springfield, VA; PB81-133720.
Smith, W. H. (1981) Air pollution and forests: interactions between air con-
taminants and forest ecosystems. New York, NY: Springer-Verlag.
Smith, W. H. (1985) Effects of air pollution on forests: Critical review discus-
sion papers. J. Air Pollut. Control Assoc. 35: 915-916.
Stark, N. (1972) Nutrient cycling pathways and litter fungi. BioScience 22:
355-360.
Stark, R. W.; Cobb, F. W. , Jr. (1969) Smog injury, root diseases and bark
beetle damage in ponderosa pine. Calif. Agric. 23: 13-15.
7-64
-------
Stark, R. W.; Miller, P. R. ; Cobb, F. W., Jr.; Wood, p. L.; Parmeter, J. R.,
Jr. (1968) I. Incidence of bark beetle infestation in injured trees.
Hilgardia 39: 121-126.
Stolte, K. W. (1982) The effects of ozone on chaparral plants in the California
South Coast Air Basin. Riverside, CA: University of California; master's
thesis.
Taylor, 0. C., ed. (1973) Oxidant air pollutant effects on a western coniferous
forest ecosystem; task C. Study site selection and on-site collection of
background information. Riverside, CA: University of California, State-
wide Air Pollution Research Center; EPA report no. EPA-R3-73-043B.
Taylor, G. E., Jr.; Norby, R. J. (1985) The significance of elevated levels of
ozone on natural ecosystems of North America. In: Proceedings of APCA
specialty conference: Evaluation of the scientific basis for an ozone/
oxidant standard; November 1984; Houston, TX. Pittsburgh, PA: Air Pollu-
tion Control Association; in press. (APCA transactions: v. 4).
Tingey, D. T.; Blum, U. (1973) Effects of ozone on soybean nodules. J. Environ.
Qual. 2: 341-342.
Tingey, D. T.; Dunning, J. A.; Jividen, G. M. (1973) Radish root growth reduced
by acute ozone exposures. Proceedings of the third international clean air
congress; October; Dusseldorf, West Germany. Dlisseldorf, West Germany:
VDI-Verlag GmbH; A154-A156.
Tingey, D. T.; Wilhour, R. G.; Standley, E. (1976) The effect of chronic ozone
exposures on the metabolite content of ponderosa pine seedlings. For.
Sci. 22: 234-241.
Treshow, M. (1980) Pollution effects on plant distribution. Environ.
Conservation 7: 279-286.
Treshow, M. ; Stewart, D. (1973) Ozone sensitivity of plants in natural commun-
ities. Biol. Conserv. 5: 205-214.
Tukey, H. B., Jr.; Morgan, J. V. (1963) Injury to foliage and its effects upon
the leaching of nutrients from above-ground plant parts. Physiol. Plant.
16: 557-564.
Tukey, H. B. Jr., Tukey, H. B.; Wittwer, S. H. (1958) Loss of nutrients by
foliar leaching as determined by radioisotopes. Proc. Am. Soc. Hortic. Sci.
71: 496-506.
Umbach, D. M. ; Davis, D. D. (1984) Severity and frequency of S02-induced leaf
necrosis on seedlings of 57 tree species. Forest Sci. 30: 587-596.
U.S. Environmental Protection Agency. (1978) Air quality criteria for ozone
and other photochemical oxidants. Research Triangle Park, NC: U.S. Environ-
mental Protection Agency, Environmental Criteria and Assessment Office;
EPA report no. EPA-600/8-78-004. Available from: NTIS, Springfield, VA;
PB83-163337.
7-65
-------
Vbgl, R. J. (1980) The ecological factors that produce perturbation-dependent
ecosystems. In: Cairns, J., Jr., ed. The recovery of damaged ecosystems.
Ann Arbor, MI: Ann Arbor Science Publishers; pp. 63-112.
Waring, R. H.; Schlesinger, W. H. (1985) Forest ecosystems concepts and manage-
ment. Orlando, Fla.; Academic Press, Inc. pp. 7-77.
Weidensaul, T. C.; Darling, S. L (1979) Effects of ozone and sulfur dioxide
on the host-pathogen relationship of Scots pine and Scirrhia acicola.
Phytopathology 69: 939-941.
Wert, S. L. (1969) A system for using remote sensing techniques to detect
and evaluate air pollution effects on forest stands. In: Proceedings
of the 6th international symposium on remote sensing of the environment.
Ann Arbor, MI: University of Michigan; pp. 1169-1178.
West, D. C.; Mclaughlin, S. B.; Shugart, H. H. (1980) Simulated forest response
to chronic air pollution stress. J. Environ. Qual. 9: 43-49.
Westman, W. E. (1977) How much are nature's services worth? Science
(Washington, DC) 197: 960-964.
Westman, W. E. (1979) Oxidant effects on California coastal sage scrub.
Science (Washington, DC) 205: 1001-1003.
Whittaker, R. H. (1965) Dominance and diversity in land plant communities.
Science (Washington, DC) 147: 250-260.
Woodwell, G. M. (1970) Effects of pollution on the structure and physiology of
ecosystems. Science (Washington, DC) 168: 429-433.
Woodwell, G. M. (1974) Success, succession, and Adam Smith. BioScience 24:
81-87.
Yang, Y. S. ; Skelly, J. M. ; Chevone, B. I.; Birch, J. B. (1983) Effects of
long-term ozone exposure on photosynthesis and dark respiration of eastern
white pine. Environ. Sci. Technol. 17: 371-373.
7-66
-------
8. EFFECTS OF OZONE AND OTHER PHOTOCHEMICAL OXIDANTS
ON NONBIOLOGICAL MATERIALS
8.1 INTRODUCTION
Photochemical oxidants comprise various chemical species capable of
reacting with a number of nonbiological materials. The nature and amount of
damage to these materials can be approximated from oxidant concentrations
(Chapter 5) and the rate constants of individual species. Unfortunately,
there is virtually no information on the rates of reaction of photochemical
oxidants other than ozone (03) on specific materials. Although ozone has been
the primary photochemical oxidant studied, its prominence in the research
literature does not necessarily indicate that it is the only important oxidant
responsible for damaging materials. Under experimental conditions with certain
chemical groups, OH radicals, which are far less abundant than ozone, have
rates of reactivity much higher than those of ozone.
Nearly all research on photochemical oxidants has focused on economically
important or abundant materials that are susceptible to oxidant damage. These
include elastomers (natural rubber and certain synthetic polymers), textile
fibers and dyes, and, to a lesser extent, paints. It has been shown that
oxidants harden and embrittle elastomers, causing cracking and a loss in
physical integrity. Damage, specifically by ozone, occurs mainly on the
surface of these materials and is accelerated by mechanical stress. In the
absence of ozone, oxidation by atmospheric oxygen still occurs, but at a
slower rate and more in the bulk of the material. These effects have been
known for years, and various antioxidants and other protective measures have
been formulated to reduce the rates of attack. Oxidant exposure weakens
certain textile fibers (i.e., reduces the breaking strength and increases the
rate of wear) and changes the color of some dyes. Like elastomeric products,
fibers and dyes particularly sensitive to ozone may be partly protected with
resistant coatings or replaced with more durable formulations. Ultimately,
these protective measures add to the cost of products. The effects of oxidants
on paints are not defined well, but they may be similar to some of the effects
on elastomers; damage from other gaseous pollutants, such as sulfur dioxide,
tends to overshadow the role of ambient ozone in estimating paint damage.
8-1
-------
To determine the actual damage to in-use materials, exposure must be
estimated. As an example of the variables that must be taken into account,
the ozone exposure of textile fibers and dyes used for clothing depends on the
activity patterns of the wearer (i.e., time at home, at work, or outdoors),
but the exposure of the same materials used for carpets and drapes involves
only indoor air. Accordingly, a knowledge of product use and indoor/outdoor
ozone gradients is essential when evaluating estimates of materials damage.
The literature selected for review in this section includes research
previously reported in the 1978 criteria document (U.S. Environmental Protec-
tion Agency, 1978) and a limited number of references published before and
after 1978. Of the twelve recent post-1978 references in this review, eight
involve laboratory/field research, and four involve analyses that use previous-
ly published material. Because little recent work has been reported on the
effects on nonbiological materials, reference to older studies is necessary
for unity and coherence, for determining dose-response relationships, and for
assessing economic impact. Technical areas considered in evaluating the cited
studies include the type of study and exposure methods used (field versus
laboratory; ambient conditions versus accelerated, artificial environments),
the pollutant-monitoring and analytical methods used, the design and conditions
of the experiment (e.g., inclusion of variables such as relative humidity and
temperature), the statistical methods and level of significance, and the
importance of the specific material studied. The absence of this type of
information is noted in the text, when applicable. In addition, no attempt
has been made to correlate aerometric data to materials in place, since the
relationship between actual exposure and an unmatched set of air quality data
is tenuous at best.
This assessment of the effects on nonbiological materials includes a
review of the mechanisms of damage and protection; it also presents dose-
response information from laboratory and field studies and evaluates previous-
ly reported economic assessments.
8.2 MECHANISMS OF OZONE ATTACK AND ANTIOZONANT PROTECTION
8.2.1 Elastomers
Most elastomeric materials found in the marketplace are composed of
unsaturated, long-chain organic molecules. That is, the molecules contain
8-2
-------
carbon-carbon double bonds. Natural rubber and synthetic polymers/copolymers
of butadiene, isoprene, and styrene account for the bulk of elastomer produc-
tion for products such as automobile tires and protective electrical coverings
used in outdoor environments (Mueller and Stickney, 1970). These types of
compounds are susceptible to oxidation and are particularly susceptible to 0~
attack. In contrast, synthetic elastomers with saturated chemical structures,
such as butyl rubber, polymers of silicones, ethylene, propylene, hypalon, and
polyurethanes, have an inherent resistance to 0- damage (Mueller and Stickney,
1970), but higher cost and limiting physical and chemical properties have
constrained their use in outdoor environments.
The differences and similarities between simple oxidation (reaction with
oxygen) and 0- attack are described by Mueller and Stickney (1970). In the
elastomer molecule, simple oxidation is postulated to proceed through the
removal of a hydrogen atom from a carbon atom adjacent to a double bond; this
is followed by the formation of a peroxy radical and subsequent radical reac-
tions, which leads to chain scission and/or cross-linking (see Figure 8-1).
Ozone is thought to attack by adding atoms directly across the double bond,
forming a five-membered ring structure. This structure quickly rearranges
(via Criegee ozonolysis) to form a zwitterion and an aldehyde (see Figure 8-2).
Subsequent reactions of the zwitterion lead to a permanently oxidized elastomer.
Ozone damage, usually in the form of cracking, tends to be more a surface
phenomenon than damage from simple oxidation. It is greatly accelerated by
mechanical stress, which produces fresh surface area at crack boundaries.
Simple oxidation, on the other hand, is slower; it occurs more in the bulk of
a material, and it is less affected by the degree of stress.
At very high concentrations and high mechanical stress, 0- damage can
result in a large number of surface microcracks that produce a frosted appear-
ance and mechanical weakening (Crabtree and Malm, 1956). Because, however,
both simple oxidation and 03 reactions lead to chain scission and chain cross-
linking, the end result of both types of damage can be very similar in appear-
ance. At pollutant concentrations and stress levels normally encountered
outdoors (and in many indoor environments), the elastomer hardens or becomes
brittle and cracked, which results in a loss of physical integrity. The
influence of 0- is evidenced primarily by the increased rate at which damage
accumulates and by the degree of protection provided by various antioxidants
and antiozonants.
8-3
-------
RADICAL
•C—C = C— ^- C—C=C . (a)
I H H • H H
H
•C—C = C— ^- — C^-C = C— (b)
• H H | H
o—o.
I SEVERAL
-C — C = C ^- CHAIN SCISSION (c)
I H STEPS PRODUCTS
O — O.
Figure 8-1. Postulated mechanism for damage to
elastomers by oxygen.
Source: Adapted from Mueller and Stickney (1970).
8-4
-------
o
/\
R R O O
I o3 III
•C —C = C ^- —C—C-C— (a)
H H H H H H
o
/\
R 0 0 R
III I + -
— C —C —C *^ — C — C — 0—0 +0 = C (b)
| H H | H H
H H
Figure 8-2. Postulated mechanism for damage to
elastomers by ozone.
Source: Mueller and Stickney (1970).
8-5
-------
According to Fisher (1957), work at the Rock Island Arsenal by R. F.
Shaw, Z. T. Ossefa and W. J. Tonkey in 1954 led to the development of effec-
tive antioxidant additives to protect elastomers from 0- degradation. Subse-
quently, antiozonants were generally incorporated into elastomeric formulations
during mixing, and their protection was effective even when elastomers were
stretched or flexed (Fisher, 1957; Mueller and Stickney, 1970).
Several theories have been advanced to explain the mechanism of anti-
ozonant protection. As summarized by Andries and Diem (1974), these are the
scavenger theory, the protective film theory, the recombination theory, and
the self-healing film theory.
The scavenger theory suggests that the antiozonant diffuses to the sur-
face, where it reacts with the 03 at a faster rate than with the carbon-carbon
double bonds of the rubber, thereby protecting the rubber sacrificially. The
protective film theory also includes diffusion to the surface, but assumes
that the resulting layer is less reactive with 03 than is the rubber and thus
constitutes a protective layer. The recombination theory proposes that the
antiozonant prevents the propagation of the radical chain reactions initiated
by 0- attack. The self-healing film theory assumes that reaction products
form on the surface and resist further degradation.
The work of Razumovskii and Batashova (1970) on the mechanism of protec-
tive action by the antiozonant N-phenyl-N'-isopropyl-p-phenylenediamine (PIPP)
is most consistent with the scavenger mechanism. These investigators showed
that 0- reacts preferentially with PIPP at a ratio of three 0., molecules per
one PIPP molecule.
Andries et al. (1979), using carbon-black-loaded natural rubber (NR)
compounds with and without antiozonants, attempted to distinguish among possible
mechanisms with attenuated total reflectance spectroscopy and scanning electron
microscopy. Their experiments indicated that a combination of the scavenger
and protective film mechanisms best explains antiozonant protection. Examina-
tion of the surface of the rubber samples with antiozonant showed that only
ozonized antioxidant and not ozonized rubber was present. This layer of
ozonized antioxidant functioned as a relatively nonreactive film over the
surface, preventing the CL from reaching and reacting with the rubber below.
In addition to reactive antiozonants, paraffinic and microcrystalline
waxes are used to protect the elastomers in rubber products such as tires.
Typically, the wax migrates to the surface of the rubber and forms a barrier
8-6
-------
against 0- attack. The ability of the wax to protect the rubber depends on
how well the wax migrates to the surface. This phenomenon, known as blooming,
depends on a number of factors besides the characteristics of the wax. Dimauro
et al. (1979) studied the ability of 18 waxes to protect rubber against degra-
dation from C° . Dimauro found that no wax by itself provided an optimal level
of protection; blending with a reactive antiozonant was required. The paraf-
finic waxes protected best at lower exposure temperatures, and the micro-
crystalline waxes were more effective at higher temperatures. Wax blends,
which combine the best effects of each type of wax, offered the best protection
over a wide range of temperature. It was found, however, that wax alone can
be detrimental to dynamic 0- resistance. Wax can induce localized stresses in
the rubber that can lead to premature rubber failure under dynamic testing
conditions.
8.2.2 Textile Fibers and Dyes
Damage to textile fibers from 03 is difficult to distinguish from that
caused by oxidation by oxygen. Reduction in breaking strength and an increased
rate of wear are the types of damage most commonly observed. Cellulose-based
fibers, acrylic fibers, and nylon fibers are affected by 0., and modacrylic
and polyester fibers have been shown to be relatively unaffected by the levels
of 0-. normally experienced in the ambient atmosphere (Zeronian et al., 1971).
As stated by Bogaty et al. (1952), however, for most uses of textile fibers
the action of Cs or oxygen is less important in product lifetime than physical
abrasion, biological degradation, soiling, fashion, and other factors.
Accordingly, the economic significance of 0- damage to textile fibers is
relatively low, and the differences in the mechanisms of attack are not impor-
tant. Nevertheless, an important property of textile products is appearance
or color; 0., reacts with a number of dyes to cause fading or changes in color.
Oxidation is the fundamental chemical reaction leading to color change in
dyed fibers exposed to 0^. Compared with other oxidizing pollutants such as
nitrogen oxides, 0- often leads to a higher degree of oxidation and thus to
different types of color changes. Terms such as 0-fading and Gulf Coast
fading have been given to some of the unique color changes attributed to
reactions with 0..
Figure 8-3 illustrates the reaction of Disperse Blue #3 with 0- and with
nitrogen oxides (Haylock and Rush, 1976). Although the nitrogen oxides removed
an alkylamine side chain, 03 attacked the quinoid portion of the molecule,
8-7
-------
DISPERSE BLUE NO. 3
OZONE
NITROGEN
OXIDES
0 OH
Figure 8-3. Reaction of anthraquinone dyes with ozone
and with nitrogen oxides.
Source: Haylock and Rush (1976).
8-8
-------
completely rupturing the ring system chromophore and oxidizing the dye to
phthalic acid, which is colorless.
The reactions between various chemical categories of dyestuffs and 0, is
influenced not only by the properties of the dye but also by the chemical
nature of the fiber to which the dye is applied and the manner in which the
dye is applied. Additional factors include the presence of protective agents;
synergistic or additive effects of temperature, air moisture, and other pollu-
tants; and even the degree of strain of the base fiber caused by folding or
creasing. For example, in a study of 0» fading of anthraquinone dyes on
nylon, Haylock and Rush (1976, 1978) found that fiber properties such as
cross-section shape, draw ratio, and the degree of steam heat setting had
significant effects on the rate and severity of 0, damage, even for chemically
identical systems. Given this complexity and sensitivity, it is not practical
to relate a specific mechanism of damage to a broad class of damage situations.
Furthermore, it may not be necessary to do so. In most cases, some combination
of dye, fibers, and protective treatments can eliminate the major problems
caused by 0, exposure and still provide the range of colors desired in the
final products.
8.2.3 Paint
The mechanisms of paint damage caused by 0- have not been defined well.
Damage is probably related to oxidation of the organic binders that hold the
pigment and form the protective seal over the surface. Damage is likely to be
similar to that of elastomers; that is, embrittlement and cracking as the
result of chain scission and cross-linking. The data available on 0- damage
to paints, however, come primarily from studies of surface erosion caused by
gaseous pollutants. Because the polymeric structure of dried paint film is
significantly different from that of an elastomer under elongation stress,
direct comparisons should be made with great caution.
8.2.4 Other Materials
Although the effects of oxidants on other materials have been examined by
several investigators, most of the limited information is qualitative and
centers on mechanisms of effects. Sanderson (1975), in a review of the effects
of photochemical smog on materials, included possible effects on plastic and
asphalt. The indicated impacts have little direct applicability, however,
8-9
-------
because these effects were recorded in a laboratory environment at extremely
high 0- levels.
Haynie and Upham (1971) reported a possible beneficial effect of photo-
chemical oxidants on the corrosion behavior of steel on the basis of field
study data. Laboratory studies, however, did not show any statistically signi-
ficant effect of 03 on steel corrosion.
Polyethylene, commonly used as electrical insulating material, may be
adversely affected by ambient 0- concentrations. Laboratory studies (National
Research Council, 1977) have demonstrated by means of infrared and other
techniques that terminal double bonds in polyethylene end groups are attacked
by "ozonized" oxygen to form carboxylic acid groups and, through ruptures in
the polymer chain, to produce short-chain dicarboxylic acids.
It is also known that atomic oxygen reacts with polyethylene at room
temperature to produce a loss in weight and some morphologic changes. The
work of Trozzolo and Winslow (1968) and Kaplan and Kelleher (1970) suggests
that singlet oxygen also interacts with polyethylene to form hydroperoxides.
Laboratory studies suggest that hydroperoxides may be the dominant oxidants
that attack polyethylene or other materials in ambient air.
Despite the known interactions of oxidants with polyethylene and other
polyolefins to form intermediate peroxy radicals, there is no evidence that
the chemical reactions go far beyond the surface. It is believed that the
effects of atmospheric (L on polyethylene insulation and other polyethylene
products are negligible in comparison with the embrittlement caused by a
combination of oxygen and sunlight. The mechanisms by which this embrittlemsnt
occurs probably involve sensitization to oxidation by absorption of ultraviolet
(UV) radiation, by residual hydroperoxy and carbonyl groups in the polymer,
and by surface deposits of aromatic sensitizers from polluted air. Deteriora-
tion of the electrical insulating properties of polyethylene by oxidation in
some environments cannot be attributed to ambient 0,.
8.3 DOSE-RESPONSE DATA
Most dose-response studies are criticized for their reliance on artificial
environments (laboratory settings) that do not contain all the critical varia-
bles encountered under ambient conditions. Scientists realize the limitations
of laboratory tests; no model could simulate conditions identical to an ambient
environment. Nevertheless, many laboratory tests have represented the outdoor
8-10
-------
environment to some extent, and the findings from these tests have been used
in conjunction with field tests to estimate the nature and amount of damage to
materials.
8.3.1 Elastomer Cracking
Hofmann and Miller (1969) demonstrated correlations between laboratory
tests and the actual service use of passenger vehicle tires in the Los Angeles
area. Basically, three laboratory test methods were used (Table 8-1): indoor
and outdoor belt flex, indoor and outdoor wheel, and stress relaxation. They
found that the behavior of rubber exposed to 0- under laboratory conditions
correlated well with the service behavior of tires in localities where atmos-
pheric 0, concentrations were high. The relative susceptibilities of different
formulations of white sidewall rubber were generally similar, whether exposed
3
under laboratory conditions to as much as 0.5 ppm (980 ug/m ) of 0, or exposed
in the ambient air of the Los Angeles area, which had annual average 07 concen-
3
trations near 0.04 ppm (80 pg/m ) (U.S. Department of Health, Education, and
Welfare, 1970). The exact exposure times, pollutant measurement methods, and
statistical analyses were not reported.
Bradley and Haagen-Smit (1951) evaluated a natural rubber (NR) formulation
for susceptibility to 0, cracking. Strips were strained approximately 100 per-
3
cent by bending and then exposed in a small chamber to 40,000 mg/m (20,000 ppm)
of 0,; these specimens cracked almost instantaneously and broke completely
within 1 sec. When these NR formulations were exposed to lower concentrations
of 03, different time periods were required for cracks to develop, as shown in
Figure 8-4, and this action increased with increasing temperature. Humidity
and sunlight had little influence on cracking rate. According to the data in
this figure, the initiation of cracks and subsequent deepening are controlled
by the dose of 0, (concentration x time).
Meyer and Sommer (1957) exposed thin polybutadiene specimens to constant
load, ambient room air, and 0,. Specimens exposed in the summer to average 0.,
3
concentrations of about 0.048 ppm (94 ug/m ) broke after 150 to 250 hr. In
3
the fall, at average 0, concentrations of 0.042 ppm (82 ug/m ), specimens
failed after exposures of 400 to 500 hr. In the winter, at average 0~ concen-
3
trations of 0.024 ppm (-^47 ug/m ), failures occurred between 500 and 700 hr.
Like the Bradley and Haagen-Smit study, these data also show the strong depen-
dence of breakage on 0, dose over the average time of exposure at which failure
occurred (average concentrations x time), but not in the same linear fashion.
8-11
-------
TABLE 8-1. TIRE INDUSTRY EXPOSURE TESTS0
Test
Strain
Conditions
Reasons for Use
Belt flexing
Stress relaxation
1 Outdoor wheel
Indoor wheel
Dynamic at 4500 to 7500 flexures per
hour
Dynamic or static; 25 percent ex-
tension at 90 cpm
Dynamic and static; variable loads,
inflation, and speed
Dynamic and static; variable loads,
inflation, and speed
Tire tests on vehicles Dynamic and static; variable loads,
inflation, and speed
Ozone chamber at 0.35 to 0.50 ppm,
or outdoors for several days
Ozone cabinet at 0.25 to 0.50 ppm
for 16-hr increments
Los Angeles area, high ozone for
several weeks
Large ozone chamber at 0.01 to 0.35
ppm and -20 to 100° F, for days
TO weeks
Extreme and typical service areas
for 1/2 to 2 yr
Rapid evaluation, variable conditions
for screening sidewall compounds
Rapid evaluation, variable conditions
for screening sidewall compounds
Quicker and cheaper than tire testing
on autos in actual service
Strain most similar to actual service,
quicker and cheaper than outdoor wheel
Ultimate test of product life
Adapted from Hofmann and Miller (1964).
-------
M
s
UJ
2
70
65
60
55
50
45
40
35
30
25
20
15
10
5
I
I
I
I
0 50 100 150 200 250 300 350 400 450
OZONE CONCENTRATION, uglm*
Figure 8-4. Relationship of cracking in rubber and ozone
concentration: time to first sign of cracking at 4x
magnification in natural rubber samples stressed at 100%.
Source: Bradley and Haagen-Smit (1951).
8-13
-------
Dose-response levels in this study are noted parenthetically for the following
concentrations: 0.048 ppm (7.2-12 ppm x hr); 0.042 ppm (16.8-21 ppm x hr);
0.024 ppm (12-16.8 ppm x hr).
In describing a new test method for evaluating the 0- sensitivity of
elastomers, Edwards and Storey (1959) presented data demonstrating the 0,
resistance of two styrene-butadiene rubber (SBR) compounds (Polysar S and
Polysar Krylene). Both compounds were exposed without and with different
levels of antiozonant protection to 0.25 ± 0.05 ppm of 03 (490 ± 98 |jg/m ) at
120°F (49°C) under 100 percent strain twice the original sample length. The
results are presented in Table 8-2. Without antiozonants, a linear relation-
ship is indicated between 03 dose (ppm/hr) and cracking depth. The coefficient
of determination for the linear regression for both materials was 0.98 compared
with 0.92 for the exponential fit. Note that the Polysar S compound displays
much greater resistance to the effects of 0, than does the Polysar Krylene
compound. Nevertheless, increasing the amount of antiozonants significantly
reduced the rate of cracking for both in a dose-related manner.
Haynie et al. (1976) conducted a chamber study to evaluate the effects of
various pollutants, including 0-, on several materials. In one part of the
study, white sidewall specimens from a top-quality, steel-belted radial tire
were exposed (strained at 10 and 20 percent) for 250, 500, and 1000 hr to 0~
33
concentrations of 160 ug/m and 1000 ug/m . The 03 level was found to be
statistically significant in the rate of cracking of this rubber. However,
cracking rates are not directly proportional to 0- concentrations for these
two levels. The average results with respect to strain and 0, level are given
in Table 8-3.
Using the mean cracking rate calculated after long-term (1000 hr) exposure
to conditions representative of the primary air quality standard for 0- and
the annual average standard for nitrogen dioxide (NOp), Haynie et al; (1976)
concluded that it would take a minimum of 2.5 years for a crack to penetrate
to the cord depth. Additional time would be necessary to attack the cords.
For this particular premium tire, therefore, sidewall failure from 03 damage
does not appear to be the cause of reduced tire life. However, the casing
might have questionable value for retreading. Tread wear, rather than sidewall
failure, probably determines the life of a typical rubber tire, and the rubber
used in tire treads is generally more resistant to 0-. than that in the side-
walls.
8-14
-------
TABLE 8-2. EFFECTS OF OZONE ON DIFFERENT SBR POLYMERS CONTAINING
VARIOUS ANTIOZONANT CONCENTRATIONS
Antiozonant,
Crack depth, 10"3 in.,
at hr of exposure Cracking depth rate
-4
Polymer pph 19 27 43 51 10 in./hr um/hr
Polysar S
("Hot" SBR)
Polysar Krylene
("Cold" SBR)
0.
0.
1.
2.
0.
0.
1.
2.
0
5
0
0
0
5
0
0
1.
0.
0.
0.
2.
1.
1.
0.
37
95
50
25
17
25
05
50
2.42
1.90
0.75
0.25
4.52
2.02
1.50
0.75
4.20
3.10
1.47
0.45
7.25
3.75
2.24
1.00
4.
3.
1.
0.
7.
4.
2.
1.
65
52
95
78
90
50
90
18
0.
0.
0.
0.
1.
0.
0.
0.
92
69
35
13
58
85
57
24
2.34
1.75
0.89
0.33
4.01
2.16
1.45
0.61
Source: Edwards and Storey (1959).
TABLE 8-3. CRACKING RATES OF WHITE SIDEWALL TIRE SPECIMENS
Mean cracking rate
Ozone concentration ± standard deviation,
ug/m3 (ppm)Strain percent mm/yr um/hr
160 (0.08)
1000 (0.5)
•
10
20
10
20
11.66 ± 7.32
17.00 ± 10.45
15.38 ± 5.38
25.74 ± 8.23
1.33
1.94
1.76
2.94
Source: Haynie et al. (1976).
8-15
-------
Veith and Evans (1980) investigated the effect of atmospheric pressure on
the cracking rate of rubber as tested in 0- chambers. It was found that a
change in barometric pressure alters the rate of cracking. Interlaboratory
comparisons were made among facilities at different geographic elevations and
thus significantly different atmospheric pressures. It was found that a
16-percent difference in cracking rate or in the extent of cracking at a fixed
0- concentration could occur. In an effort to correct the problem and standard-
ize the testing techniques, Veith and Evans (1980) recommended that 0- content
in accelerated chamber testing be expressed in terms of 0, partial pressure
(in Pa units) rather than simply in terms of concentrations.
Gandslandt and Svensson (1980) evaluated the stress test methodology used
to estimate the 0- resistance of rubber compounds. This test measures the
decrease in the isoelastic force of stressed rubber exposed to 0.,. The authors
suggested that materials should be prestressed in an 0.,-free atmosphere for at
least 72 hr before testing, because the complicating effects of the natural
relaxation of the isoelastic force constant of the material decreases exponen-
tially with time. The effects of this natural relaxation mechanism become
insignificant after 2 to 3 days of prestressing compared to the effects caused
by Oo cracking.
Ten different mixtures of three rubber compounds, NR, SBR, and CR (a
compound not defined by the authors), were tested with the isoelastic force
method (Gandslandt and Svensson, 1980). The 03 protection afforded each
rubber formulation is summarized in Table 8-4. After a relaxation time of
70 hr in an 0,-free atmosphere (2 hr less than their prescribed criteria
for sample exposure), the samples at 50-percent elongation were exposed to 0-
3
concentrations of 0.5 ppm (980 ug/m ) at 30°C. The time to 10-percent and
20-percent relaxation of the isoelastic force in the rubber test samples was
used to gauge the 0, resistance of the formulation. Compounds GL 2073 B,
SS 203, and SS 200 C showed greatest resistance to the effects of 0,, and
those formulations that were unprotected (GL 2073 D, SS 200 B, SS 202 A,
SS 203) and the formulations protected only by paraffin wax (GL 2073 G) demon-
strated the least resistance to 0- attack. The testing showed great variety
in the kinds of visible cracking effects as a result of the exposure. The
compounds with no protection often showed a large number of small cracks over
the entire surface of the material, but those compounds protected by a combina-
tion of wax and antiozonant or by wax alone sometimes showed only a single
8-16
-------
TABLE 8-4. PROTECTION OF TESTED RUBBER MATERIALS
Protected
Rubber formulation Unprotected Wax Antiozonant
GL 2073 B, C X X
G X
D X
SS 200 A, C XX
SS 202
SS 203
B
A
B
X
X
X
X X
Source: Gandslandt and Svensson (1980).
crack, which grew rapidly. These effects are demonstrated in Figure 8-5.
Compounds SS 202 B (Figure 8-5a) and SS 200 C (Figure 8-5b), both protected
with wax and antiozonant, showed fairly good resistance when gauged by the
10-percent and 20-percent stress relaxation tests but failed after approximately
50 hr and 58 hr of exposure, respectively. On the other hand, compounds
SS 203 and SS 200 A, both unprotected, exhibited small surface cracking and
outlasted some of the protected compounds. Moreover, protection with wax and
antiozonant may afford long-term protection, but when one crack appears, it
can grow rapidly and cut off the test piece, as shown in Figure 8-5b.
Davies (1979) reported on the effects of ozone and other environmental
factors on interply adhesion of natural and synthetic rubber compounds.
Excellent adhesion of plies is essential to the proper manufacturing of tires.
The rubber strips must make interlocking contact at the joint boundary or the
strength of the product will be inadequate. Ozone attack on synthetic poly-
isoprene and polybutadiene produces a surface layer of ozonides. With NR, the
film consists of ozonides and carbonyl groups (Andries and Diem, 1974; Andries
et al., 1979). The results of the Davies (1979) tests indicated that before
curing, the adhesion of SBR compounds is unaffected by exposure to 0, concentra-
o J
tions of 0.15 ppm (294 (jg/m ), but the adhesion of the NR/SBR blend decreases
by approximately 30 percent. Large reductions (on the order of 70 percent) in
adhesion between plies were noted with the NR compounds; even exposure for a
8-17
-------
§
** I *•
u7u-
UJ
O
cc
O
LL
O
O
(0
100
70
50
30
SBR UNPROTECTED
(SS 202 A)
(a)
SBR, WAX + ANTIOZONANT
(SS 202 Bl
I I
I
I
I I
I I
0 10 20 30 40 50 60 70 80 90 100
RELAXATION TIME, hours
O
r-
X
UJ
O
cc
2
O
UJ
O
CO
100
70
50
30
20
0
(b)
I
I
I
I
I
NR (SS 200)
(WAX + ANTIOZONANT
C {(DOUBLE AMOUNT
(COMPARED TO A)
B, UNPROTECTED
A, WAX +
y ANTIOZONANT
I I* I I
0 10 20 30 40 50 60 70 80 90 100
RELAXATION TIME, hours
Figure 8-5. Relaxation of rubber compounds in O3 is
affected by the combination of rubber formulation and
type of 03 protection. Compounds were tested at 03
concentration, 0.5 ppm (980 uglm3); temperature, 30°C;
elongation, 50%. Vertical line at the end of a curve means
total failure, and vertical axis represents relaxation where
FQ is the initial force; Ft is the force after time, t.
Source: Gandslandt and Svenson (1980).
8-18
-------
few hours at 0.05 ppm reduced adhesion considerably. The adhesion tests on
cured NR, SBR, and isoprene rubber (IR) compounds after exposure to various
levels of 03 and humidity are summarized in Table 8-5. The adhesion of the
SBR compound is superior to that of the other two compounds, which were greatly
affected by increased RH.
TABLE 8-5. EFFECT OF OZONE AND HUMIDITY ON INTERPLY ADHESION3
Compound
NR
IR
SBR
Initial
adhesion
5
5
5
0.15 ppm 03
(294 ug/m3),
30% RH
2-3
4-5
4-5
Final adhesion
0.25 ppm 03
(490 ug/m3),
30% RH
1
2-1
3-4
0.15 ppm 03
(294 ug/m3),
60% RH
I
1
3-4
Adhesion is rated from 1 (bad) to 5 (excellent), based on a visual scale
standardized by the authors.
All exposures were 16 hr in duration.
Source: Adapted from Davies (1979).
Davies examined antiozonants, antioxidants, and fast-blooming waxes as
means of protecting NR compounds from sunlight and O, attack and the subsequent
development of the films that lead to poor adhesion between plies. The results
of these evaluations are presented in Table 8-6. Of the samples exposed after
o
16 hr at 03 concentrations of 0.15 ppm (294 ug/m ), only those protected by
the fast-blooming waxes were found to resist 0^ and have excellent adhesion
between plies (Table 8-6). Antiozonants and antioxidants in the NR did not
aid interply adhesion (Table 8-6). Davies (1979) theorized that antiozonants
and antioxidants react with ozonized rubber and form a protective film against
further attack by 0.,. However, this film also apparently acts as a barrier to
proper adhesion between plies. Davies noted that after exposure to sunlight
alone, the antioxidants generally maintained good adhesions, but the waxes
gave only fair protection. He concluded that the combination of a fast-
blooming wax and an effective antioxidant or antiozonant is necessary to
protect NR from 0^ attack and sunlight.
8-19
-------
TABLE 8-6. EFFECT OF ANTIOZONANTS, ANTIOXIDANTS, AND FAST-BLOOMING
WAXES ON INTERPLY ADHESION IN NATURAL RUBBER3
Antiozonant '
Untreated
ETMQ
6 PPD
1 PPD
77 PPD
TBMP
TMQ
Wax 1
Wax 2
Rating0
1
1
1
1
1
2
2
5
5
Ozone resistance rated from 1 (bad) to 5 (excellent), based on a visual scale
standardized by the author.
All substances were given an initial rating of 5.
GRating assigned after 16-hr exposure to 0.15 ppm (294 ug/m3) of 03.
See appendix for explanation of abbreviations.
Source: Davies (1979).
Wenghoefer (1974) studied the effects of 0~ on adhesion and the climatic
sensitivity of tire cords dipped in resorcinal-formaldehyde latex (RFL). Cli-
matic sensitivity was described as summer sickness, a problem affecting cords
primarily during hot, humid weather. Many fibers and dip formulations were
studied to determine their sensitivity to 0,, humidity, nitrogen dioxide
(N02), UV light, and heat. Wenghoefer exposed these materials at a constant
temperature of 100°F (37.8°C) to 0- levels that varied between 0 and 1.5 ppm
3
(0 and 2940 |jg/m ) and to relative humidity (RH) levels ranging from 20 to
90 percent. Adhesion deteriorated from changes in surface properties of the
RFL-dipped cords as a result of exposure to 0~, humidity, UV light, and heat.
The adhesion losses from 0- and the combined effects of 0- and humidity were
most notable in the first 6 hr of exposure. The detrimental effects of heat,
N02» and the synergistic interaction of NOp and humidity were much less
pronounced. Table 8-7 summarizes the elastomer dose-response studies.
8.3.2 Dye Fading
Color fading of certain textile dyes has been attributed to the effects of
ambient 0,. Although N02 was originally identified as the pollutant most
8-20
-------
TABLE 8-7. DOSE-RESPONSE STUDIES ON EFFECTS OF OZONE ON ELASTOMERS
CO
I
Conditions
Laboratory/
field
Laboratory
Controlled
field
Laboratory
Material /Product
Automotive
tires
Vulcanized
rubber
strips
Rubber tires and
various polymers
SBR:
Plysar S
Plysar Krylene
with and without
antiozonants
Concen- Measure-
tration, ment
Pollutant ppm method Exposure
Ozone 0.25 to NA NA
0.5
Ambient 0.04 03 NA >1 yr
air (annual
average)
Ozone 0.02 to NA 3 to 65
0.46 min
Ambient 0.023 to NA 150 to
air 0.048 03. 700 hr
Ozone 0.25 NA 19 to 51
hr
Environ-
mental Dose,
variables ppm-hr
Tires
under
stress
Los Angeles >350 .
environ-
ment;
actual
service
use
Physical M). 02
stress to 0.03
Physical 9 to
stress 20
and
ambient
environ-
ment
120°F, 4.75
100% to
strain 12.75
Effects
Cracking of
wh i te side
wall
Positive
correlation
between lab-
oratory and
ambient air
tests
Surface
cracking
Time of
cracking
Percent anti-
ozonant was
related to
cracking
depth rate
Comment
Purpose was to
correlate lab-
and field tests.
Exposure time,
detailed pollu-
tant measurements,
and statistical
analyses were not
reported.
Test was designed
to establish
dose/response
curves on 03-
sensitive rubber
for use as an
analytical method.
Cracking occurred
over a broad
range of values
and was related
to stress.
Demonstrated
dose/response
linear relation-
ship for ozone
on unprotected
rubber.
Reference
Hoffman
and Miller
(1969)
Bradley
and Haagen-
Sroit
(1951)
Meyer and
Sommer
(1957)
Edwards
and
Storey
(1959)
-------
TABLE 8-7. DOSE-RESPONSE STUDIES ON EFFECTS OF OZONE ON ELASTOMERS (continued)
Concen-
tration,
Conditions Material/Product Pollutant ppm
Laboratory White Ozone 0.08 to
sidewall 0.5
tire
specimens
Laboratory Ten different Ozone 0.5
Q-, NR, SBR, CR
i formulations
p-? with and without
protection
Laboratory Several NR/SBR Ozone 0.05 to
blends with and 0.15
without pro-
tection
Laboratory Tire cords Ozone 0 to 1. 5
(66 nylon; Dacron
polyester; Kevlar
aramid)
Nitrogen 0 to 20
dioxide
Measure- Environ-
ment mental
method Exposure variables
NA 250 to 10 and
1000 20%
hrs strain.
NA Up to 30°C
300 hr
NA -v-3 to Sunlight,
16 hr humidity
NA 0 to UV light;
48 hr heat
(100°C);
RH (20-
90%); N02
NA
Dose,
ppm-hr Effects
20 to Mean cracking
500 rates were
determined
for different
stress and
ozone levels.
Up to Time to 10 to
50 20% relaxation
~0.15- Interply adhe-
2.4 sion affected
at 0.05 ppm and
above
up to RFL adhesion
72 loss occurred
primari ly dur-
ing 6-hr expo-
sure to high
RH and 0.2 ppm
Comment
Detailed data
not available
to verify
author's state-
ment that 2-1/2
years of ambient
conditions were
required for ozone
cracks to penetrate
cord depth.
Both formula-
tion and pro-
tection
affected
relaxation.
Both waxes and
antiozonants
needed for pro-
tection against
sunlight plus
ozone.
Synergism between
03 and RH; RFL
deterioration
occurred at
surface.
Reference
Haynie
et al.
(1976)
Gandslandt
and
Svensson
(1980)
Davies (1979)
Wenghoefer
(1974)
-------
important to color fading, the effects of 0- were noted by Salvin and Walker
(1955) nearly three decades ago. The phenomenon was termed 0-fading. The pri-
mary products affected were permanent press garments (polyester and cotton) and
nylon carpeting. In permanent press garments, dye fading occurs primarily at
the creases and folds. The fading of nylon carpeting occurs in, the presence of
high RH and depends on the dyes used. Ozone fading most affected the blue and
red disperse dyes of the anthraquinone series but not the azo series of dyes.
Salvin and Walker (1955) tested disperse dyes that were resistant to the
effects of nitrogen oxides. They exposed a series of drapery products to
confirm their resistance to the dye fading that was thought to be attributable
to NOp. Different types of dyes ranging in vulnerability to nitrogen oxides
were exposed in Pittsburgh, Pennsylvania (an urban region of high N02 concen-
trations), and Ames, Iowa (a suburban area with low NO^ concentrations).
After 6 months of exposure, the investigators found that N0?-resistant dyes
had performed well in Pittsburgh but poorly in Ames, indicating the influence
of another fading agent. By using a combination of laboratory chamber studies
and outdoor exposure, Salvin and Walker (1955) demonstrated that 0, was the
pollutant responsible for the change. Blue anthraquinone dyes and certain red
anthraquinone dyes were markedly bleached after exposure to just 0.1 ppm
o
(196 |jg/m ) of 0~. Azo red and yellow dyestuffs and diphenylamine yellow dyes
were shown to be resistant to fading at these concentrations, also confirming
the results of the field study. The use of known antiozonants, such as diphenyl-
ethylenediamine and diallyl phthalate, in combination with disperse blue dyes
was effective against 03 fading, thus providing additional evidence of the
effects of 0- on dyed fabrics.
To explain much of the fading of certain dyed fabrics during lightfast-
ness testing and service exposure trials, Schmitt (1960, 1962) also invoked
the concept of 0., fading. In studies to demonstrate colorfastness of certain
dyes when exposed to sunlight, Schmitt exposed 38 color specimens for 12 months
at Phoenix, Arizona, and Sarasota, Florida, and for 7 months in Chicago,
Illinois. Specimens exposed included direct dyes on cotton, acid dyes on
nylon, acid dyes on wool, disperse dyes on acetate, disperse dyes on acrilan,
disperse dyes on dynel, acid dyes on dynel, cationic dyes on orlon, and disperse
dyes on dacron.
8-23
-------
Each specimen was exposed to a predetermined amount of direct sunlight,
measured by a pyroheliometer, and then examined in the laboratory to measure
the amount of fading. Schmitt found that samples given equal amounts of sun
exposure tended to fade more in Florida than in Arizona. He concluded that
the higher RH was a contributory factor and that atmospheric contaminants were
the principal factor in accelerated fading. Schmitt also exposed certain dyed
fabrics in covered test frames where the effect of sunlight would be eliminated.
After 24 days of exposure in Florida, Schmitt found that even in covered
frames fading was of the same magnitude as noted with samples exposed to
sunlight. His work also demonstrated the importance of RH in the dye-fading
mechanism by suggesting that the increased moisture content of the fibers pro-
moted and accelerated the absorption and reaction of pollutants with vulnerable
dyes.
Ajax et al. (1967) summarized the results of a study of 69 dye-fabric
combinations that were exposed outdoors in light-free cabinets at 11 sites.
These sites were Sarasota, Florida; Phoenix, Arizona; Cincinnati, Ohio; and
four urban-rural combinations: Chicago and Argonne, Illinois; Washington,
D.C., and Poolesville, Maryland; Los Angeles and Santa Paula, California; and
Tacoma and Purdy, Washington. Among those fabrics exhibiting a high degree of
fading at both urban and rural sites in the first 6 months, fading was much
greater at the urban sites than at the rural sites. The samples exposed in
Phoenix, Sarasota, and Purdy showed the lowest amount of fading, which indicated
that humidity and temperature are not, by themselves, the primary factors in
fading. The highest fading rate occurred in samples exposed in Los Angeles,
Chicago, and Washington, D.C. In addition, there was a marked seasonal varia-
tion in the test results, with greater fading during the spring and summer
seasons. Generally, the results correspond with seasonal peaks in 0,. concentra-
tions. Editorial problems, however, between the text and tabular material tend
to confuse the authors' discussion.
Ajax and coworkers also exposed the fabrics to irradiated and nonirradi-
ated auto exhaust with and without sulfur dioxide (SO,,) for 9 hr/day for six
consecutive days. From the results of this chamber study, they noted that
"photochemically produced byproducts of automobile exhaust are a prime cause
of fading compared to fading caused by nonirradiated auto exhaust or by clean
air with sulfur dioxide added." In the presence of S02, however, a more than
additive effect was seen in the dye fading tests for both chamber and field
8-24
-------
studies. Although their conclusions concerning ozone itself are easily substan-
tiated in the research literature, the 0, levels measured in their chamber are
questionable. The daily 9-hr average 0- concentrations (measured by neutral KI,
Mast instrument) were identical for irradiated (UV) and nonirradiated exhaust
(0.02 ppm); irradiated exhaust plus S09 produced 0.55 ppm of 0-.
Ł o
Beloin (1972, 1973) investigated the effects of air pollution on various
dyed textiles by conducting field and controlled-environment laboratory studies.
For the field study, a wide range of dyed fabric was exposed in light-tight
cabinets at the same four urban and four rural sites used in the Ajax studies.
The study was carried out over a 2-year period, in eight consecutive 3-month
seasonal exposure periods. Color change data and air pollution and weather
measurements were analyzed to identify the factors that caused fading. About
two-thirds of the fabrics studied showed appreciable fading. Most of these
fabrics faded significantly more at urban sites than at rural sites, and the
amount of fading varied among metropolitan areas and seasons. Samples exposed
in Chicago and Los Angeles demonstrated the greatest degree of fading, and those
exposed in Purdy, Washington, and Phoenix showed the least amount. The small
amount of fading evidenced by the samples exposed at extreme temperatures and/or
humidity indicated that these factors by themselves have no effect on fading.
The sample also showed some seasonal variations in fading. In areas of high
oxidant concentration, maximum fading occurred primarily in summer and fall.
Fabrics exposed in Chicago, where SOp concentrations are higher in the winter,
showed greater fading during this season.
The results of the outdoor fading study were used in a multiple regression
analysis, which examined fading as a function of six independent variables (NO,,,
SOp, 0-, nitrogen oxide, temperature, and humidity). After eliminating those
fabrics that developed only trace fading and those for which the regression
was not significant, the analysis focused on 25 fabric dye samples, 23 of
which showed S0? to be a significant variable. Ozone was also a significant
contributor to fading of eight dyed fabrics and NOp to fading of seven dyed
fabrics. The dominance of S0? as a factor in fading may have been complicated
by soiling.
The laboratory study was designed to assess the effects of air pollutants,
temperature, and RH on the colorfastness of 30 samples selected from those
exposed during the field study. Fabric samples were exposed to two concentra-
tions of 0,: 0.05 ppm (98 pg/m ) and 0.50 ppm (980 |jg/m ). The laboratory
studies demonstrated that high 0., levels produced more significant fading in
8-25
-------
more fabric samples than did low levels. Visible fading did occur in about
one-third of the sensitive fabrics exposed to 0- concentrations of 0.05 ppm
3
(98 ug/m ). These levels are similar to those frequently found in metropolitan
areas. The laboratory study also demonstrated that high RH (90 percent) is a
significant factor in promoting and accelerating 0.,-induced fading.
Haynie et al. (1976) and Upham et al. (1976) reported on the degree of
fading of three different drapery fabrics exposed in a laboratory chamber to
combinations of high and low 0, concentration (980 and 196 ug/m ; 0.5 and
0.1 ppm, respectively), high and low RH (90 percent and 50 percent), and high
and low concentrations of N0« and SO^. The three fabrics selected for this
study were a royal blue rayon-acetate, a red rayon-acetate, and a plum cotton
duck. The samples were exposed in the chamber for periods of 250, 500, and
1000 hr; the degree of fading was measured with a color difference meter. The
fading of the plum-colored material was statistically related to RH and the
, N02 concentration. For the red and blue fabrics, only RH appeared to be a
significant factor. The effects of concentrations of ozone on the amount of
fading of these dyes were not statistically significant, even after exposure
3
for 1000 hr to 980 ug/m (0.5 ppm), levels much higher than typical ambient
exposures.
Haylock and Rush (1976, 1978) studied the fading of anthraquinone dyes on
nylon fibers. In the first test, nylon carpet yarn dyed with Olive I and
Olive II was exposed to varying levels of temperature, RH, and 0«. Material
dyed with Olive I and exposed at 70 percent RH, 40°C (104°F), and 0.2 ppm
(392 ug/m ) of 03 showed visible fading after 16 hr of exposure. At 90 percent
RH, similar fading occurred in less than 4 hr. Under the same RH and tempera-
ture conditions, increasing the 0,, concentration from 0.2 ppm to 0.9 ppm (392
3
to 1760 ug/m ) resulted in a parallel increase in fading. Samples in knitted
sleeve form demonstrated much greater susceptibility to 0, attack than samples
in skein form.
Using Disperse Blue 3 and Disperse Blue 7 dyes exposed to constant condi-
tions of 40°C (104°F), 90 percent RH, and 0.2 ppm (392 ug/m3) of 03, Haylock
and Rush (1976) investigated the effect on fading of changing the fiber cross
section, the fiber draw ratio, and the method of setting the nylon fibers with
steam.heat. They found that increasing the surface area of the fibers resulted
in an increased fading rate. Increasing the fiber draw ratio reduced dye
fading, and increasing the heat-setting temperature decreased resistance to
fading in disperse dyes.
8-26
-------
The necessity of high temperature and high humidity for induction of CL
fading in nylon was further confirmed by the additional work of Haylock and
Rush (1978). Their studies showed a good correlation between accelerated 0,
fading in the laboratory and in outdoor, in-service exposure, during which
temperature and humidity extremes were common. Control samples exposed indoors,
however, where temperatures and humidities were lower, did not exhibit nearly
the same magnitude of fading as the laboratory samples.
Heuvel et al. (1978) investigated the importance of the physical nature
of Nylon 6 yarns on the (k fading behavior of a disperse blue dye. Samples of
Nylon 6 yarns dyed avocado green with a dye mixture including Disperse Blue 3
o
were exposed in a laboratory cabinet to 0.5 ppm (980 ug/m ) of 03 at 40°C and
an RH of 85 percent. Heuvel et al. found that the microfibril diameter and
specific surface area of the fiber were the fiber characteristics most closely
related to 03 fading, thus confirming suspicions expressed earlier by Salvin
(1969).
Nipe (1981) summarized the results of a 3-year study to establish the
relationship between in-service atmospheric contaminant fading by 0- of carpet:;
in a home versus 0- fading as determined by the American Association of Textile
Chemists and Colorists (AATCC) Standard Test Method 129, Colorfastness to
Ozone in the Atmosphere Under High Humidities. (Measurements were also taken
"to compare the fading caused by oxides of nitrogen.) The test carpets were
made of Nylon 6 and 66 dyed with two disperse and two acid dye formulas. Test
samples from the homes of 28 participants were returned every 3 months for the
3-year period. The exposure sites selected for this long-term .study represented
variations in home heating and cooling, utilities, climate, and geographical
locations. The carpet samples were placed in areas as close as possible to
the kitchen but away from exposure to sunlight or any traffic.
Attempts were made to relate the color change for each exposure period to
outside temperature and RH, but the statistical analyses of the data showed no
correlation between outside weather conditions and in-home fading by either
contaminant. Geographical location appeared to have a significant effect on
fading. Test samples from sites in the southeast and northeast showed far
more 0_ fading than those in the west and far west. Test samples in homes
with air conditioning exhibited less fading during the summer than those
without air conditioning. In all samples, much greater fading was caused by
0- during July, August, and September than in January, February, and March.
8-27
-------
Typically, 0- levels indoors are higher during the summer, when doors and win-
dows are more likely to be open, thus allowing a greater exchange between
inside and outside ai.r. The results of the study of in-service interior
carpet exposures were compared with the results of AATCC Test 129, as shown in
Table 8-8. In a sample that performs satisfactorily through 1.08 cycles of
exposure in AATCC Test 129, there is a 98-percent probability against in-service
fading over a 1-year period. A sample that performs satisfactorily through
only 0.6 test cycles of fade has only a 90-percent probability of satisfactory
performance after 1 year of in-service exposure.
Kamath et al. (1982) studied the effect of atmospheric 03 dye fading on
nylon fibers. Prior studies had postulated that 0., does not penetrate into
the fiber to destroy the dye, but instead attacks the dye at the surface of
the fiber. Dye then diffuses outward from the fiber interior because of the
concentration gradient set up as the surface dye is destroyed. Using micro-
spectrophotometry to test this postulated mechanism, Kamath et al. studied the
diffusion and destruction of C.I. Disperse Blue Dye 3 on Nylon 6 continuous
filament yarn measuring about 45 urn in diameter. With this method, they were
able to generate a dye distribution profile across the cross section of the
fiber and to determine the diffusion coefficient of a dye in the fiber. The
fibers were exposed in a controlled environment to 0, concentrations of 0.2 ppm
3
(392 ug/m ) for 2 to 120 hr at a temperature of 40°C and RH levels of 90 percent,
85 percent, and 65 percent. The results of these laboratory studies indicated
that RH has a significant positive effect on fading, that destruction of the
dye begins near the surface of the fiber in the early stages of exposure, and
that 0- penetration into the fiber may be an important mechanism in 0- fading.
As shown in Figure 8-6, the dependence of fading rates on humidity was substan-
tial. Even slight rises in humidity from 85 percent to 90 percent caused a
significant increase in the extent of fading. At 65 percent RH, the fading
rate drops dramatically. This effect was attributed to the breakage of hydro-
gen bonds in the presence of water, which leads to a more open structure with
high segmented mobility; this condition is more favorable to diffusion of O,
and disperse dyes.
Kamath et al. (1982) used a surface reaction model to attempt to explain
the amount of fading (dye loss) due to 0- exposure. They found, however, that
this approach could explain only a very small portion of the loss. They
concluded that the dye distribution profile across the fiber resulted from
8-28
-------
TABLE 8-8. COLORFASTNESS OF TEST SAMPLES COMPARED WITH COLORFASTNESS OF
IN-USE CARPETING
Probability of acceptable Number of test cycles Number of test cycles
colorfastness of in-use equivalent to 1 year equivalent to 5 years
carpeting of in-use service of in-use service
99
98
95
90
80
75
70
60
50
1.36
1.08
0.80
0.60
0.42
0.37
0.33
0.27
0.22
6.80
5.40
4.00
3.00
2.10
1.85
1.65
1.35
1.10
Source: Adapted from Nipe (1981).
penetration of 0- into the fiber itself. Subsequent reaction of this 03 with
dye diffusing toward the surface of the fiber was therefore considered to be
an important mechanism in 0- fading of anthraquinone dyes in nylon.
Salvin (1969) reported that 0., and (to a lesser extent) N02 caused dye
fading of cotton/permanent press fabrics. As summarized by Dorset (1975), 0,
was found to be the major fading agent, with nitrogen oxides also capable of
causing fading, though to a lesser extent. The fading mechanism occurs as a
result of the curing operation and involves the disperse dyes on the polyester
fibers rather than the vat dyes on cotton. During curing, some disperse dyes
partially migrate to the permanent press finish, which is a combination of
reactant resin, catalysts, softeners, and nonionic wetting agents. This
migration occurs preferentially along the folds and creases, causing fading to
predominate in these areas. The disperse dyes migrate to the solubilizing
agents in the finish, a medium in which fading by air contaminants can easily
occur. Remedial measures to avoid this problem include selecting dyes more
resistant to reaction with 0, and NO,,, avoiding the use of magnesium chloride
catalyst in the permanent press process, and using different surfactants and
softeners. The use of magnesium chloride as a catalyst makes 0--sensitive
dyes more sensitive to 0- and less fast to washing (Dorset, 1975). When the
catalyst is zinc nitrate, dyes are more washfast and resistant to 0^ fading.
Thus, the amount of dye fading might not be a function only of 03 concentra-
tion but also of the number of times the garment is washed. The present use
8-29
-------
100 r
c
9
H
9
a.
(A
CO
2
UJ
20
40 60 80
FADING TIME, hours
100
120
Figure 8-6. Effects of relative humidity (RH) on fading of C.I. Disperse
Blue 3 (CIDB-3) in Nylon 6 after exposure to 0.2 ppm ozone.
Source: Adapted from Kamath et al. (1982).
8-30
-------
of a zinc nitrate catalyst appears to have generally eliminated the problem of
the prefading of dyes in permanent press fabrics from 03 exposure. A summary
of the dye fading studies is presented in Table 8-9.
The type of research reported on dye fading is primarily qualitative in
nature. Earlier studies relied on comparisons among various geographical
locations and seasonal variations with little attention given to actual concen-
tration and exposure characterizations. For several of the initial field
investigations reported here, neither ozone nor oxidant concentrations were
given; rather, notations such as high versus low or urban versus rural were
the only description of oxidant levels. The few laboratory studies of any
technical merit employed only two concentrations of ozone at most, making it
nearly impossible to derive dose-response relationships. Comparisons among
studies are difficult owing to the various dye and fabric combinations tested.
Also, the importance of relative humidity on ozone fading rate confounds
comparisons among many of the studies that did not use the same RH percentages.
Moreover, further complications arise from the absence of standardized methods
to measure dye fading.
8.3.3 Fiber Damage
Sunlight, heat, alternate wetting and drying, and microorganisms are
causative factors in the weathering and deterioration of fabrics exposed
outdoors. The influence of 0, at normal ambient levels is generally small by
comparison.
In a review of the effects of weather and atmospheric pollutants on
textiles, Warty (1977) outlined a number of damage mechanisms, the complexity
of the mechanisms, and their effects on manmade and natural fibers. The
damage mechanisms reviewed included those involving soiling, 0.,, sunlight,
microbial attack, humidity, and SCL. Natural fibers such as jute, flax, hemp,
sisal, and coconut, which have a multicellular structure and contain lignin,
are much more resistant to the effects of weathering than is cotton, a natural
fiber with no lignin. Even in amounts as small as 0.2 percent, however,
lignin will cause yellowing or browning of the material when exposed to light.
Compounds added to increase resistance to one weathering agent may actually
accelerate the damage caused by others. For example, the interaction of light
with phenolic compounds used as antimicrobial agents accelerates fabric degra-
dation.
8-31
-------
TABLE 8-9. LABORATORY STUDIES OF THE EFFECTS OF OZONE ON DYE FADING
Dye
Fabric
Concn. ,
ppm
Exposure
Environmental
Variables
Effects
Comments
Reference
Blue and red
Olive I and II
Disperse blue
3 and 7
Drapery
oo
1
CO
f , "i
1 XJ
Direct red 1
Reactive red 2
Sulfur green 2
Azoic red
Direct red 1
Acid red 151
Acid yellow 65
Acid violet 1
Basic red 14
Basic yellow 11
Acid orange 45
Disperse blue 3
Disperse blue 3
Disperse blue 3
Disperse blue 27
Disperse blue 27
AATC 03 ribbon .
Cotton
"
..
Rayon
Wool
II
II
Acrylic
it
Nylon
it
Cellulose
acetate
11
Polyester
Acetate
0.1
0.05
0.5
12 2k
Temp.=130°C,
32°C
RH=50%, 90%
Nylon fibers
Nylon fibers
0.2
0.9
hr
RH=70%-90%
Temp.=40°C
Disperse blue dye Nylon 6 yarn 0.5
in an avocado
green mixture
Disperse blue 3 Nylon 6 yarn 0.2
2-120 hr
RH=85%
Temp.=40°C
RH=65%, 85%,
90%,
Temp.=40°C
Both dyes were markedly bleached.
No fading occurred when anti-
oxidants were added.
Induced fading at both levels
but at a nonlinear rate. Both
temperature and humidity in-
creased fading rate, and RH was
more important. Fabrics 13, 14,
15 the most sensitive, followed
by 19, 1, 17, 18, and 7. Only
trace amounts of fading occurred
in the remaining fabrics.
Visible fading in Olive I after
16 hr at 70% RH; same effect
after 4 hr at 90% RH. Linear
increase in fading at 0.9 ppm 03.
Insufficient data for dose-
response determinations.
This study followed a field
study showing that oxidants
other than NO caused fading.
Insufficient data to show
detailed dose-response
relationships. Although
samples were measured
throughout the exposure,
only the 12-wk data were
presented
Salvin and Walker (1955)
Beloin (1973)
Both RH and 03 concentration Haylock and Rush (1976)
affected fading and in a
nearly linear fashion. Sleeve
form was more susceptible
than skein form. Haylock and
Rush (1976) found that:
(1) increased fiber draw
ratio reduced fading;
(2) increased heat-setting
temperature increased
fading;
(3) increased fiber surface
area increased fading
Fading was closely correlated with Insufficient data for dose-
fiber surface area (diameter). response relationship deter-
minations.
Nearly linear increase in fading
with time. RH had a major influ-
ence on fading rate.
This study focused more on
mechanisms of 03 fading
rather than dose-response
relationships.
Heuvel et al. (1978)
Kamath et al. (1982)
Coupling component 2, azoic diazo component 32.
-------
Cellulose fibers, whether natural or manmade, are very sensitive to sun-
light in the UV portion of the spectrum. Ultraviolet light causes disruption
of the chemical bonds within the fiber itself. Even in protected fabrics, a
secondary photochemical reaction can occur with certain dyes and pigments.
Bleached fabrics, which are much more resistant to microbial attack, tend to
be much more sensitive to the action of sunlight. The bleaching weakens mole-
cular linkages, making the carbon-carbon and carbon-oxygen bonds much more
easily broken when exposed to sunlight.
Synthetic fibers, though highly resistant to microbial attack, are still
adversely affected by UV light. Degradation can be minimized or avoided by
use of UV-absorbing additives applied as coatings or in the manufacturing
process. Warty (1977) concluded that, because the weathering process is a
very complex interaction of several variables, it is difficult to rely on a
single test method to define performance.
Bogaty et al. (1952), as part of a program aimed at segregating some of
the elements that cause weathering, carried out experiments to study the
possible role of 0- in the deterioration of cotton textiles. These investiga-
tors exposed samples of duck and print cloth to air containing 0.02 and 0.06 ppm
o
(39 and 118 ug/m ) of 0-. Samples were exposed both dry and wet and tested
for 50 days. The wet samples were water-saturated once per week, and moisture
was added regularly so that the moisture content of the cloth was never less
than 50 percent. Similar fabric samples were exposed to similar 0., concentra-
tions with no moisture added, and another control group was similarly wetted
but exposed to clean (0.,-free) air. After exposure to 03, the wetted samples
showed a loss in breaking strength of approximately 20 percent. The wet print
control cloth showed a loss in breaking strength of only half this amount.
The study showed that low levels of 0- degrade cotton fabrics if they are
sufficiently moist. Bogaty et al. surmised that an estimated 500 to 600 days
of natural exposure might be required to reach a stage of degradation similar
to that caused by a 50-day exposure to 0_ alone. Because unprotected fabrics
typically reach a much more advanced state of decay after such long exposures
to weathering, Bogaty et al. concluded that the effect of 03 is slighter than
that of other agents. Although not noted by Bogaty et al., the 0- and in-
creased moisture may have caused the formation of hydrogen peroxide (HLO^),
which could account for the loss in breaking strength.
Morris (1966) also studied the effects of 0, on cotton. Samples were
3
exposed in the absence of light to 0.5 ppm (980 ug/m ) of 0- (more than four
8-33
-------
times the National Ambient Air Quality Stadard (NAAQS) of 235 (jg/m or 0.12 ppm)
for 50 days in a chamber maintained at 70°F (21°C) and 72 percent RH. No
appreciable effect on breaking strength was found. Apparently, the moisture
content of the cotton was not high enough to produce the degradation that
Bogaty et al. (1952) measured in wet cotton samples, even though the concentra-
tion of On was considerably higher.
The laboratory study of Kerr et al. (1969) examined the effects of
the periodic washing of dyed cotton fabrics exposed to 0, and the amount of
fading and degradation of moist, dyed fabrics exposed to 0,. They exposed
samples of print cloth, dyed with CI Vat Blue 29, in a chamber to a continuous
supply of purified air containing 0, concentration levels of 1 ± 0.1 ppm
3
(1960 ± 196 (jg/m ). The samples were exposed at room temperature (25°C) in
the absence of light, and a shallow container of water was kept on the chamber
floor to increase the humidity. Samples were withdrawn from the chamber after
12, 24, 36, 48, and 60 days. After an exposure period of 60 days, which
included either 20 washing or 20 soaking treatments, the change in strength of
control fabrics was not significant. By comparison, the fabrics exposed to 0.,
changed significantly; the loss in strength of the washed fabrics was 18 percent,
and that of the soaked fabrics, 9 percent. Fading was also evident in the
fabrics exposed to 0~, but not in the control samples. Differences in the
amount of fading between the washed and soaked samples were evident, but the
reason for the differences was not. Kerr et al. concluded that washing in
hot, soapy water may have affected the properties of the dye.
In laboratory studies, Zeronian et al. (1971) simultaneously exposed
modacrylic (dynel), acrylic (orlon), Nylon 66, and polyester (dacron) fabrics
to artificial sunlight (xenon arc) and charcoal-filtered air contaminated with
0.2 ppm (392 pg/m ) of 0~ at 48°C (118°F) and 39 percent RH. During exposure,
the fabric samples were sprayed with water for 18 min every 2 hr. Ozone damage
was measured by comparing these samples with fabrics exposed to the same environ-
mental conditions without 0,. After exposure for 7 days, Zeronian et al.
found that 0- did not affect the modacrylic and polyester fibers. The exposure
did seem to affect the acrylic and nylon fibers slightly by reducing breaking
strength. The degree of difference, however, in the change of fabric properties
between those exposed to light and air and those exposed to light and air con-
3
taining 0.2 ppm (392 [jg/m ) of 0- was not significant.
8-34
-------
In general, the contribution of (L to degradation of fabrics has not been
quantified well. Bogaty et al. (1952) concluded that the effects of other fac-
tors (sunlight, heat, wetting and drying, and microorganisms) far outweighed the
effects of GO on cotton duck and print cloth. The work by Morris (1966) and
Kerr et al. (1969) does point to the synergistic effect of moisture and 0^ as
an important ingredient in material degradation, possibly caused by the forma-
tion of a more potent oxidizing agent. Finally, the work of Zeronian et al.
(1971) also indicates little if any effect of 0\ on synthetic fibers. Thus,
it appears that 0. has little if any effect on textiles, fibers, and synthetic
cloth exposed outdoors. A similar view was proposed by the National Academy
of Sciences (National Research Council, 1977) in a review of the effects of
DO and other photochemical oxidants on nonbiological materials.
8.3.4 Paint Damage
A paint surface may suffer several types of damage that affect its useful-
ness, including cracking, peeling, erosion, and discoloration. Of these,
erosion (i.e., wearing away of the paint surface) is the type of damage most
often studied with respect to the impact of gaseous pollutants. Studies of
paint cracking and peeling have focused on the effects of moisture and have
not dealt with the possible influence of ambient pollutants.
Several damage functions for CL-induced erosion of paint have been reported
in the literature. Such reports are based on either accelerated chamber
studies or long-term outdoor exposure studies. Unfortunately, all studies to
date have significant flaws that render their results highly questionable.
Damage to a paint surface is the cumulative effect of the conditions to which
the surface is exposed, including various combinations of temperature, moisture,
sunlight, and pollution level. No outdoor exposure study to date has been
able to match all factors exactly to separate the impact of 0~ from the other
factors.
In a laboratory chamber exposure study, Haynie et al. (1976) exposed
oil-based house paint, latex house paint, vinyl coil coating, and acrylic coil
coating to 0.5 and 0.05 ppm concentrations of SOp, NOp, and 03 in various
combinations. Statistically significant effects of 0~-caused damage were
observed on the vinyl coil coating and the acrylic coil coating. There was a
positive interaction between 03 and RH on the vinyl coil coating and a positive
direct 0., effect on the erosion rate of the acrylic coil coating. The rate of
8-35
-------
erosion was low, however, and both .vinyl and acrylic coil coatings were shown
to be very durable. Coatings as thin as 20 urn should last more than 20 years
before requiring replacement because of the effects of 03. A linear regression
for the acrylic coil coating data gives:
Erosion rate = 0.159 + 0.000714 03 (8-1)
3
where erosion rate is in um/yr and 0- is ug/m .
Although the 0, effect on this coating was found to be statistically
significant, it has no practical significance because the erosion rate is so
3
slow; at 0.12 ppm (235 ug/m ) of 0,, the erosion rate is 0.33 um/yr. At an
3
average annual 0, level of 100 ug/m , this regression predicts that a 20-um-
thick coating would last over 80 years.
In a comprehensive study by Campbell et al. (1974), panels painted with
different exterior paints (automotive refinish, latex coating, coil coating,
industrial maintenance coating, and oil-based house paint) were exposed to air
pollutants in an environmental chamber under accelerated weathering conditions.
The panels were exposed to low (0.1 ppm) and high (1.0 ppm) concentrations of
0., and SOp. After exposure, the panels were examined by measuring erosion,
gloss, surface roughness, tensile strength, attenuated total reflectance
(ATR), and the surface effects revealed by scanning electron microscopy and
infrared examination. The panels were examined after 0, 400, 700, and 1000 hr
of chamber exposure (considered as equivalent to 0, 200, 350, and 500 days,
respectively, of exposure).
The relative sensitivity of a coating to pollutant damage depended on the
particular test used to define the damage. For example, when comparing oil-
based house paint with automotive paint, the former showed the greatest ATR
change but no change in gloss, but the latter exhibited little ATR change
and the largest change in gloss. In general, exposures to 1 ppm (1960 ug/m )
of 0, produced greater increases in erosion rates than did clean air. Concen-
trations of this magnitude, however, do not represent typical ambient exposure
3
levels of 0~. At the more representative level of 0.1 ppm (196 ug/m ), 0- did
not produce statistically significant increases in erosion rates.
In conjunction with the chamber studies, field measurements were made of
the erosion of paint from test panels exposed to outdoor environments consisting
8-36
-------
of a clean, rural atmosphere (Leeds, North Dakota); a moderately polluted
atmosphere (Valparaiso, Indiana); a heavily polluted (S0?) atmosphere (Chicago,
Illinois); and a high-oxidant, moderately polluted atmosphere (Los Angeles,
California). The results of this study showed that paint erosion was much
greater in the polluted areas than in relatively clean, rural areas. The
highest erosion rates were observed for the coil coating and oil-based house
paints at the Chicago and Los Angeles exposure sites. Since meteorology and
air quality were not measured at the exposure sites, correlation of film
damage with the environmental parameters was not possible. The study does
suggest that SCL exerts an adverse effect on exterior paints with calcium
carbonate as an extender pigment. The coil coating and oil house paints
were formulated with calcium carbonate. Oxidants were probably reacting with
the organic binder of the coil coating and oil house paints, although no
mechanism for this reaction was developed from this exposure study.
In an outdoor exposure test of the effects of air pollutants on materials,
Mansfeld (1980) exposed latex and oil-based house paints as well as galvanized
steel, weathering steel, stressed aluminum, silver, marble, and nylon at nine
test sites in St. Louis, Missouri. In conjunction with the material exposures,
measurements of meteorological parameters, 03, oxides of nitrogen, total
hydrocarbons, total sulfur, S(L, and hydrogen sulfide were made. The investiga-
tor used a regression model to relate the corrosion rates (i.e., rate of
change of damage) to the meteorological parameters, air quality parameters,
and length of exposure. There is some uncertainty in the results of the
analysis because the independent variables show a degree of correlation with
each other. Nevertheless, the results of several of the material-pollutant
relationships are worth noting. For the latex house paint, concentrations of
atmospheric 0- were found to contribute significantly more to the accelerated
erosion of the painted surface than the duration of exposure or the direction
(north, south) to which the sample was exposed. The duration of exposure and
the sulfate concentration were the most important factors in explaining the
erosion of oil-based paint. Mansfeld suggested that these effects indicate
the differing responses and behavior of the two types of paint.
Some of the color pigments used in commercial paints and dyes are also
used in artists' paints. Shaver et al. (1983) studied the colorfastness of
several of these pigments exposed to 0.40 ppm of 03 for 95 days under con-
trolled temperature and humidity conditions. Several of the 1,2-dihydroxy-
anthraquinone-type pigments faded considerably, but no dose-response curves
8-37
-------
could be determined. Furthermore, the effects on pigments combined with the
various binders used in actual applications has not been investigated. Never-
theless, because works of art have an indefinite service life compared with, for
example, the short service life for textiles, further research is needed
before estimates of the type and amount of damage to paintings and prints are
possible.
The effects of (k on paint are still being studied. The preliminary
results of Mansfeld's work indicate that there may be a statistically signifi-
cant relationship between the erosion of latex paint and RH and 03. Further
studies are necessary, however, before a cause-and-effect relationship can be
conclusively established.
8.4 ECONOMICS
8.4.1 Introduction
Damage to nonbiological materials from ozone is usually expressed in terms
one or both of the following two general classes of costs to producers and con-
sumers: (1) ozone-accelerated replacement and repair costs, as when the service
life and/or aesthetics of a material are impaired, and (2) increased avoidance
costs, as when certain industries (e.g., tires, plastics, paints, dyes, and
fabrics) are obligated to incur expenditures for antiozonant research and deve-
lopment, substitute processes and materials, additives and formulations, product
packaging, advertising, etc., in order to offset sales losses that would other-
wise occur.
In theory, the approach selected should depend on the observed behavior
of the producers and consumers of the materials in question, and the type of
damage to which they are reacting. In practice, the existing empirical esti-
mates of ozone damage to materials are far from reliable for the following rea-
sons:
1. In some studies, coverage is limited to one or two classes of materials,
and to restricted geographical regions.
2. Other studies are entirely too aggregative, suffering deficiencies because
of (1) broad and vague notions of materials exposure and ozone concentra-
tions; (2) little or no data on the spatial and temporal distributions of
the exposed materials; (3) unverified guesses regarding the incidence and
level of cost increases and production adjustments incurred by ozone-
affected industries; and (4) inadequate attention to economic trade-offs
8-38
-------
among different industries and different regions, and between producers
versus consumers.
3. The engineering/economic estimates are not well related to the scientific
literature in this area, and tend to be far too simplistic to meet the
concerns of the scientist.
4. Most of the cost assessments were conducted in the early 1970s. Few
recent studies exist. Moreover, these earlier studies cite extensively
from each other and there are few independent analyses that do not merely
rework old data.
5. As a consequence of the fourth item above, many of the ozone-related
costs reported in the early 1970s for research and development, product
substitution, etc. , are no longer appropriate. Some of these were
presumably once-only costs that are no longer charged against current
production. Because the literature is dated, there may also be some
current research and development, substitution attempts, and so on, not
at all reflected in the studies cited in this section. In sum, the cost
estimates largely reflect technologies and ozone concentrations prevailing
some 10 to 20 years ago.
6. Most of the so-called economic studies of ozone damage to materials have
been conducted using an engineering approach. That approach focuses on
the classification and quantification of the various kinds of costs in-
curred by the producers and users of the ozone-sensitive materials. Eco-
nomic theory would argue, however, that this is merely the first step in
the assessment process, and that supply-demand relationships are then
needed in order to proceed with the calculation of social net benefits
(i.e., changes in producer and consumer surpluses). In practice, however,
it appears that almost all of the damage assessments conducted to date
stop short of obtaining an econometric measure of economic surplus. As
such, the studies reported in this section must be interpreted accordingly.
8.4.2 Methods of Cost Classification and Estimation
Computation of accelerated replacement is probably the most widely applied
method of estimating the costs of materials damage to air pollutants. In this
approach a materials damage function is developed to show the increase in phy-
sical damage for an increase in the dose of the pollutant. Then a cost schedule
is constructed to show how maintenance or replacement schedules are influenced
by the pollutant level. Hershaft et al. (1978) note, however, that this
method usually assumes existing inventories, and does not take into account
substitutions of materials with more (or less) resistance to pollution. As a
result, this method tends to overestimate the cost of damage from pollutant
increases and to underestimate the net savings realized from pollutant reduc-
tions.
8-39
-------
A second approach considers avoidance costs. This refers to practices
such as adopting alternative production processes and materials. Some indus-
tries add antiozonants to their products, or change the chemical formulation
of their output. All of these measures mitigate the impact of ozone on the
service life or aesthetics of the products in question. Moreover, these
measures also require research, development, and implementation expenditures.
As such, estimation of these costs is conceptually and empirically difficult,
since the opportunity to use different materials changes in response to the
level of ozone concentration.
A number of factors complicate the use of both the replacement and the
avoidance methodologies. Data on key variables are generally missing or merely
assumed. Lessening the reliability of the final cost estimates are deficiencies
in knowledge of (1) the physical damage functions; (2) the quantities and types
of materials exposed to ozone indoors, outdoors, and in respective regions of
the country; (3) the actual expenditures incurred for increased replacement,
maintenance, and avoidance that can be directly attributed to ozone; (4) the
threshold ozone damage levels that prompt mitigating action; and (5) the range
of substitution strategies that can be used to ameliorate degradation. On this
latter point, few attempts have been made to identify current technology prac-
tices and possibilities. The variety of rubber compounds, paint mixtures, and
fabric dyes reflects the number of proprietary formulations, and each formulation
presumably has a different response to ozone exposure.
An additional complication is that repair, replacement, and substitution
are frequently dominated by factors unrelated to ozone concentrations. This
can lead to spurious correlations if studies are accepted uncritically. For
example, tire replacement may be high in a given region of the country because
of high ozone levels associated with automotive exhaust. Alternatively, it
may be high simply because the total miles of automotive use per year are
higher in that region than in the nation as a whole.
Another illustration is the substitution of dyes. New dyes that replace
ozone-sensitive dyes may also be more colorfast and able to survive more washings
than the dyes they replace. In this case, apportionment of the costs of the new
dyes between ozone resistance and the other improved characteristics embodied
in the new formulations is an extremely arbitrary and perhaps meaningless
exercise.
8-40
-------
8.4.3 Aggregate Cost Estimates
The important caveats identified in the preceding discussion qualify the
empirical data presented in this and following sections. Table 8-10 summarizes
reports of highly aggregated estimates of oxidant damage to all materials.
Unfortunately, there are no known recognized studies that are more recent than
those reported in the table. For purposes of gross comparison only, the figures
are expressed in 1984 currency equivalents alongside 1970 currency equivalents,
the base data for the reference studies. They do not, however, represent 1984
supply-demand relationships, production technologies, or ozone concentrations.
It must be emphasized that the costs cited in 1984 currency equivalents there-
fore cannot be considered true 1984 costs.
TABLE 8-10. SUMMARY OF DAMAGE COSTS TO MATERIALS BY OXIDANTS
(in millions of 1970 and 1984 dollars)
Study
Barrett and
Waddell (1973)
Mueller and
Stickney (1970)
Salmon (1970)
Salvin (1970)
Waddell (1974)
Yocum and
Grappone (1976)
Freeman (1979)
Elastomers/plastics
NDa
500.0 .
(1500)°
295.2
(915)
ND
ND
ND
ND
Materials costs
Fabric/dye
(260)
ND
358.4
(1111)
83.5
(259)
ND
ND
ND
All
(3878)
ND
653.6
(2026)
ND
900.0
(2790)
572.0
(1773)
505.0
(1566)
ND=No data. Investigator(s) did not develop estimates in this category.
1984 dollars are listed parenthetically below 1970 dollars and reflect
inflation (consumer price index) rather than real increases in costs.
8-41
-------
Salmon (1970) was among the first to attempt to estimate the annual cost
of air pollution damage to materials. His computation included the dollar
value of annual materials production, a weighted average economic life of each
material included in his study, a weighted average factor for the percentage
of the material exposed to air pollution, and a factor for increased labor to
treat damaged materials. Cost was defined as the value of the material mul-
tiplied by the difference between the rate of material deterioration in a
polluted urban versus an unpolluted rural environment. All data, except for
annual production levels of materials, were assumed.
If it is assumed that ozone affected all of the fibers, plastics, and rubber
in the study by Salmon, then annual damage costs attributed to ozone would have
been $2,026 million (1984$). Salmon did not consider ozone-related damage to
paint, since the dominant paint-damaging mechanisms are soiling and gaseous
sulfur dioxide. His costs refer to maintenance and replacement only, and do
not allow for materials protection, substitution, etc.
In discussing other limitations of his study, Salmon cautioned that his
estimates were of potential loss, not of actual observed loss. Despite this
and other qualifications that lessen the usefulness of the figures derived, the
Salmon study has been cited extensively and used quantitatively in a number of
the subsequent studies cited here.
For example, the materials estimate by Barrett and Waddell (1973) is based
primarily on the work of Salmon (1970). Barrett and Waddell supplemented this
by drawing on Mueller and Stickney (1970) for damage costs on elastomers, and
on Salvin (1970) for damage costs related to dye fading. Combining some of
these numbers, Barrett and Waddell stated that materials damage costs attri-
butable to oxidants alone were $3,878 million (1984$).
Freeman (1979) reviewed earlier studies that categorized the cost of damage
to materials. Using the work of Waddell (1974) and Salvin (1970), Freeman cal-
culated that the materials damage costs attributable to oxidants and oxides of
nitrogen were $2,031 million (1984$). Of this total, roughly 46 percent was
damage to textiles and dyes (from Salvin 1970), while the remaining 54 percent
was damage to elastomers (from Mueller and Stickney, 1970). Freeman then assumed
a 20 percent reduction in oxidant levels since 1970, and went on to conclude
that the monetary benefits of controlling oxidants, oxidant precursors, and
oxides of nitrogen were between $170 and $510 million (1984$). Freeman com-
puted the savings attributable to oxidant controls alone as $128 to $383
million (1984$).
8-42
-------
Waddell (1974) likewise depended primarily on existing studies to calculate
the national cost of air pollution in 1970. Waddell used Salmon (1970), Salvin
(1970), Mueller and Stickney (1970), and Spence and Haynie (1972) to derive an
estimate of $6,820 million (1984$) as the total gross annual damage for materials
losses in 1970 resulting from air pollution. The component attributable to ozone
and oxidants alone was $2,790 million (1984$), within a wide range of $1,550 to
$4,030 million (1984$).
Yocom and Grappone (1976), in work for the Electric Power Research Institute,
estimated that the cost of air pollution damage to materials was about $6,820
million (1984$) in 1970. Of this total, ozone was estimated to be responsible
for $1,773 million (1984$), or some 26 percent of the total.
Because of the reliance of the later studies on the questionable data and
unverified assumptions contained in the earlier ones, the results compared here
are of extremely limited usefulness for cost-benefit purposes. The empirical
estimates of materials damage at the aggregate level are typified by a paucity
of original research, primary data, and fresh insights. Rather, successive
layers of estimates have been generated upon essentially the same weak foun-
dations. No recent research (e.g., post-1979) is available to improve upon
this circumstance.
8.4.4 Damage to Elastomers
The damage to rubber and other elastomers by ozone can be significant in
terms of the kinds and quantities of materials that are susceptible. For
example, damage to rubber seals, hoses, belts, cables, pharmaceutical goods,
and vehicle tires has been mentioned as economically important (Mueller and
Stickney, 1970).
If damage induced by pollutants is to be considered economically important,
however, the effective useful life of the product must be significantly affec-
ted by pollutant exposure. The life of many rubber products is determined
more by their end use and the wear and tear of normal use than by pollutant
damage. For example, the rubber in surgical gloves can be shown to be sensi-
tive to ozone exposure. Because these gloves are used indoors, however, and
because they also are usually discarded after one use, the outdoor ozone
concentration has no influence on their useful lifetime.
Vehicle tires represent the major use of rubber that is subject to signi-
ficant economic costs from the effects of ozone (McCarthy et al., 1983). The
8-43
-------
amount of antiozonants added to a tire formulation depends on two factors:
ozone concentrations and expected tire life. Previously, tire manufacturers
varied the amount of antiozonants regionally, depending on ozone concentrations.
Now, however, most companies produce for a national market from each plant, and
consequently formulate their compounds for worst-case conditions with an appro-
priate margin for safety.
The second factor that determines the amount of antiozonants in tire for-
mulations is expected tire life. Antiozonants are added in sufficient quanti-
ties to resist ozone damage for 5 or 6 years in radial tires, and 3 or 4 years
in bias-ply and bias-belted tires.
The cost of antiozonants is about $0.80 (1984$) per passenger car tire
and about $1.66 (1984$) per truck tire. Given a yearly national production of
100 million passenger tires and 50 million truck tires, the total annual cost
of antiozonants is $163 million (1984$). If ozone should be reduced, it is
uncertain to what extent tire manufacturers would find it possible and profita-
ble to reduce the level of antiozonants.
Mueller and Stickney (1970) contend that if ozone concentrations were re-
duced, but the amount of antiozonant per tire was not reduced, more retreadable
tire casings would be available for passenger cars. (Truck tires have a com-
paratively shorter useful economic life and ozone damage is not a significant
factor in truck tire retreading). In 1980, nearly 17 million tires were
rejected for retreading because of weatherchecking, at least some of which was
attributable.to ozone. Hence, a reduction in ozone levels could conceivably
make available a greater supply of retreadable tire casings, lowering costs in
the retread industry. As qualified previously, however, this depends on the
extent to which tire manufacturers find it economical to adjust their levels
of antiozonant.
Mueller and Stickney (1970) estimated that the damage costs to elastomeric
compounds caused by air pollutants, mainly ozone, totaled $1550 million (1984$).
Their estimates are presented in Table 8-11. Protection against the effects
of ozone (i.e., avoidance costs) represents the added cost of antiozonants,
antioxidants, and special rubber blends formulated for their oxidant-resistant
and ozone-resistant properties. The second cost element is early replacement
because of shortened service life, a cost borne directly by consumers. The
heading "indeterminate" refers to the costs of protective wrappings and coatings
and research to formulate resistant compounds, and "other" includes labor
8-44
-------
TABLE 8-11. SUMMARY OF DAMAGE COSTS TO RUBBER BY OZONE
(in millions of 1970 and 1984 dollars)3
Protection 'c Early replacement Indeterminate Other All Factors
Total cost 170(527) 225.7(700) 78(242) ^25(78) ^500(1550)
Cost Special polymer 20.6 (64) Tires 37.0(115)
breakdown
Antiozonant 34.1(106) Mechanical 29.7(90)
Wax 5.0(16) Medical 100.5(312)
i Belting 22.5(70)
en
Hose 36.0(112)
a!984 dollars are given parenthetically next to 1970 dollars.
Retail value approximately three times the manufacturing cost.
Smaller costs for protective finishes, wrapping, and compound development could not be estimated.
Labor costs associated with replacement can be greater than the cost of the part, but realistic estimates
could not be made.
Source: Mueller and Stickney (1970).
-------
costs for repair and replacement. Although estimates ,are given, the authors
note that these two columns really cannot be estimated. All of the costs
presented in the table refer to the year 1969, are expressed in 1970 dollars,
and have uncertain reliability and relevance in the context of 1984.
8.4.5 Damage to Fibers and Dyes
Ozone has a significant impact on certain sensitive dyes. Barrett and
Waddell (1973) reported that the national cost of dye fading caused by ozone
was $260 million (1984$) per year. Of this amount, 30; percent was dye fading
in acetate and triacetate, 50 percent was dye fading in nylon carpets, and 20
percent was dye fading of permanent press garments. Barrett and Waddell
assumed that avoidance costs included preventive measures to minimize damage,
such as use of more expensive dyes as well as additional research and testing.
Replacement costs took account of the assumed reduced life of the dyed materials.
No research has been conducted since 1973 to verify or update these esti-
mates. A problem with them is that a proportion of fading and physical wear
was arbitrarily assigned to ozone rather than to other factors. As noted pre-
viously, the use of magnesium chloride as a catalyst in the permanent-press
process led to dyes that were more sensitive to ozone and also less washfast.
Thus, the rate of fading is caused not only by the interaction between the dye
and ozone, but also by the frequency of washing.
Salvin (1970) conducted a study on how ozone and the oxides of nitrogen
increase the costs of fading of dyed fabrics. Costs in the work of Salvin
included those for more resistant dyes, inhibitors, research and development,
and reduced service life. Of the total cost of dye fading, that part attritubed
to ozone was $259 million (1984$) per year. Salvin contacted manufacturers to
obtain costs of dyes, processes, and preventive measures. The costs of reduced
service life were based, however, on estimates rather than observations.
Salvin's study does not seem to take into account the differences between in-
door and outdoor ozone concentrations and the significance of this for textile
exposure; thus, the result must be viewed cautiously for that reason.
8.4.6 Damage to Paint
Ozone levels typically occurring in the ambient air (Chapter 5) have not
been shown to cause damage to paint. Campbell et al. (1974) were unable to
demonstrate a relationship between ozone and paint damage either in a care-
fully controlled chamber study or in outside exposure tests. Haynie and Upham
8-46
-------
(1971) showed that the only statistically significant effects of ozone on
paint were damage to vinyl and acrylic coil coatings; however, the effects of
ozone* were insignificant in shortening coating lifetimes. McCarthy et al.
(1981) found that the costs associated with premature replacement of acrylic
and vinyl coil coatings were minimal and could not be attributed to pollutants
alone.
Aesthetics tend to be a decisive factor in the use of acrylic and vinyl
coatings. Although the coating retains its primary function of providing a
protective surface, changes in gloss and sheen, as well as degradation of
color, can be problems. The causative agents for these aesthetic effects are
environmental factors (primarily sunlight), as well as the qualities of the
pigment, formulation and mixing, and application. No data are available to
suggest the role of ozone (alone or in conjunction with other pollutants) in
this fading. Hence, the costs of diminished aesthetics attributable to ozone
are largely undetermined.
8.5 SUMMARY AND CONCLUSIONS
Over two decades of research show that ozone damages certain nonbiological
materials; the amount of damage to actual in-use materials, however, is poorly
characterized. Knowledge of indoor/outdoor ozone gradients, for example, has
expanded considerably in recent years, and this type of exposure information
has not been incorporated in materials damage studies. Moreover, virtually
all materials research on photochemical oxidants has focused on ozone. Theoret-
ically, a number of the less abundant oxidants may equal or surpass ozone in
reactivity with certain materials, but this possibility has not been tested
empirically. In the absence of photochemical pollution, oxidative damage to
certain materials still occurs from atmospheric oxygen, but at a much reduced
rate and through different chemical mechanisms. Generally, ozone damages
elastomers by cracking along the line of physical stress, whereas oxygen
causes internal damage to the material.
The materials most studied in ozone research are elastomers and textile
fibers and dyes. Natural rubber and synthetic polymers of butadiene, isoprene,
and styrene, used in products like automobile tires and protective outdoor
electrical coverings, account for most of the elastomer production in the
United States. The action of ozone on these compounds is well known, and
8-47
-------
dose-response relationships have been established and corroborated by several
studies. These relationships, however, must be correlated with adequate expo-
sure information based on product use. For these and other economically
important materials, protective measures have been formulated to reduce the
rate of oxidative damage. When antioxidants and other protective measures are
incorporated in elastomer production, the dose-cracking rate is reduced consid-
erably, although the extent of reduction differs widely according to the
material and the type and amount of protective measures used.
The formation of cracks and the depth of cracking in elastomers are re-
lated to ozone dose and are influenced greatly by humidity and mechanical
stress. Dose is defined as the product of concentration and time of exposure.
The importance of ozone dose was demonstrated by Bradley and Haagen-Smit
(1951), who used a specially formulated ozone-sensitive natural rubber.
Samples exposed to ozone at a concentration of 20,000 ppm cracked almost
instantaneously, and those exposed to lower concentrations took a propor-
tionately longer time to crack. At concentrations of 0.02 to 0.46 ppm, and
under 100-percent strain, the cracking rate was directly proportional to the
time of exposure, from 3 to 65 min.
Similar findings were reported by Edwards and Storey (1959), who exposed
two SBR elastomers to ozone at a concentration of 0.25 ppm for 19 to 51 hr
under 100-percent strain. With ozone doses of 4.75 ppm-hr to 12.75 ppnrhr, a
proportional rate in cracking depth was observed, averaging 2.34 um/hr for
cold SBR and 4.01 um/hr for hot SBR. When antiozonants were added to the com-
pounds, the reduction in cracking depth rate was proportional to the amount
added. Haynie et al. (1976) exposed samples of a tire sidewall to ozone at
concentrations of 0.08 and 0.5 ppm for 250 to 1000 hr under 10 and 20 per-
cent-strain. Under 20-percent strain, the mean cracking rate for 0.08 ppm was
1.94 um/hr. From these and other data, they estimated that at the ozone stan-
dard of the time (0.08 ppm, 1-hr average), and at the annual NO standard of
/\
0.05 ppm, it would take 2.5 years for a crack to penetrate cord depth.
In addition to stress, factors affecting the cracking rate include atmos-
pheric pressure, humidity, sunlight, and other atmospheric pollutants. Veith
and Evans (1980) found a 16-percent difference in cracking rates reported from
laboratories located at various geographic elevations.
8-48
-------
Ozone has been found to affect the adhesion of plies (rubber-layered
strip's) in tire manufacturing. Exposure to ozone concentrations of 0.05 to
0.15 ppm for a few hours significantly decreased adhesion in an NR/SBR blend,
causing a 30-percent decrease at the highest ozone level. This adhesion prob-
lem wor'sened at higher relative humidities. When fast-blooming waxes and
antiozonants or other antioxidants were added, only the combination of protec-
tive measures allowed good adhesion and afforded protection from ozone and
sunlight attack. Wenghoefer (1974) showed that ozone (up to 0.15 ppm), espe-
cially in combination with high relative humidity (up to 90 percent), caused
greater adhesion losses than did heat and NO,, with or without high relative
humidity.
The effects of ozone on dyes have been known for nearly three decades.
In 1955, Salvin and Walker exposed certain red and blue anthraquinone dyes to
a 0.1 ppm concentration of ozone and noted fading, which until that time was
thought to be caused by NO^. Subsequent work by Schmitt (1960, 1962) confirmed
the fading action of ozone and the importance of relative humidity in the ab-
sorption and reaction of ozone in vulnerable dyes. The acceleration in fading
of certain dyes by high relative humidity was noted later by Beloin (1972,
1973) at an ozone concentration of 0.05 ppm and relative humidity of 90 percent.
Kamath et al. (1982) also found that a slight rise in relative humidity (85 to
90 percent) caused a 20-percent dye loss in nylon fibers.
Both the type of dye and the material in which it is incorporated are
important factors in a fabric's resistance to ozone. Haynie et al. (1976)
and Upham et al. (1976) found no effects from ozone concentrations of 0.1 to
0.5 ppm for 250 to 1000 hr under high and low relative humidity (90 vs. 50
percent) on royal blue rayon-acetate, red rayon-acetate, or plum cotton. On
the other hand, Haylock and Rush (1976, 1978) showed that anthraquinone dyes
on nylon fibers were sensitive to fading from ozone at a concentration of 0.2 ppm
at 70 percent relative humidity and 40°C for 16 hr. Moreover, the same degree
of fading occurred in only 4 hr at 90 percent relative humidity. At higher
concentrations, there was a parallel increase in fading. Along with Heuvel
et al. (1978) and Salvin (1969), Haylock and Rush (1976, 1978) noted the
importance of surface area in relation to the degree of fading. In explaining
this relationship, Kamath et al. (1982) found that ozone penetrated into the
fiber itself and caused most of the fading through subsequent diffusion to
the surface.
8-49
-------
Field studies by Nipe (1981) and laboratory work by Kamath et al. (1982)
showed a positive association between ozone levels and dye fading of nylon
materials at an ozone concentration of 0.2 ppm and various relative humidities.
In summary, dye fading is a complex function of ozone concentration, relative
humidity, and the presence of other gaseous pollutants. At present, the
available research is insufficient to quantify the amount of damaged material
attributable to ozone alone. Anthraquinone dyes incorporated into cotton and
nylon fibers appear to be the most sensitive to ozone damage.
The degradation of fibers from exposure to ozone is poorly characterized.
In general, most synthetic fibers like modacrylic and polyester are relatively
resistant, whereas cotton, nylon, and acrylic fibers have greater but varying
sensitivities to the gas. Ozone reduces the breaking strength of these fibers,
and the degree of reduction depends on the amount of moisture present. Under
laboratory conditions, Bogaty et al. (1952) found a 20 percent loss in breaking
strength in cotton textiles under high-moisture conditions after exposure to a
0.06 ppm concentration of ozone for 50 days; they equated these conditions to a
500- to 600-day exposure under natural conditions. Kerr et al. (1969) found a net
loss of 9 percent in breaking strength of moist cotton fibers exposed to ozone
at a concentration of 1.0 ppm for 60 days. The limited research in this area
indicates that ozone in ambient air may have a minimal effect on textile fibers,
but additional research is needed to verify this conclusion.
The effects of ozone on paint are small in comparison with those of other
factors. Past studies have shown that, of various paints, only vinyl and
acrylic coil coatings are affected, and that this impact has a negligible
effect on the useful life of the material coated. Preliminary results of
current studies have indicated a statistically significant effect of ozone and
relative humidity on latex house paint, but the final results of those studies
are needed before conclusions can be drawn.
For a number of important reasons, the estimates of economic damage to
materials are far from reliable. Most of the available studies are now out-
dated in terms of the ozone concentrations, technologies, and supply-demand
relationships that prevailed when the studies were conducted. Additionally,
little was (and is) known about the physical damage functions, and cost
estimates were simplified to the point of not properly recognizing many of
the scientific complexities of the impact of ozone. Assumptions about expo-
sure to ozone generally ignored the difference between outdoor and indoor
8-50
-------
concentrations. Also, analysts have had difficulty separating ozone damage
from other factors affecting materials maintenance and replacement schedules.
For the most part, the studies of economic cost have not marshalled factual
observations on how materials manufacturers have altered their technologies,
materials, and methods in response to ozone. Rather, the analysts have mere-
ly made bold assumptions in this regard, most of which remain unverified through
the present time.
Even more seriously, the studies followed engineering approaches that do
not conform with acceptable methodologies for measuring economic welfare.
Almost without exception, the studies reported one or more types of estimated
or assumed cost increases borne by materials producers, consumers, or both.
The recognition of cost increase is only a preliminary step, however, towards
evaluating economic gains and losses. The analysis should then use these cost
data to proceed with supply and demand estimation that will show how materials
prices and production levels are shifted. Because the available studies fail
to do this, there is a serious question as to what they indeed measure.
Increased ozone levels increase sales for some industries even as they
decrease welfare for others. For example, manufacturers of antiozonants for
automobile tires conceivably stand to increase sales as ozone increases, while
purchasers of tires stand to pay higher prices. This is only one illustration
of a fundamental analytical deficiency in the various studies of materials
damage: the absence of a framework for identifying gainers and losers, and the
respective amounts they gain and lose.
Among the various materials studies, research has narrowed the type of
materials most likely to affect the economy from increased ozone exposure.
These include elastomers and textile fibers and dyes. Among these, natural
rubber used for tires is probably the most important economically for the
following reasons: (1) significant ambient air exposure and long use life;
(2) significant unit cost; and (3) large quantities and widespread distribution.
The study by McCarthy et al. (1983) calculated the cost of antiozonants
in tires for protection against ozone along with the economic loss to the
retread industry. While limitations in this study preclude the reliable
estimation of damage costs, the figures indicate the magnitude of potential
damage from exposure to ozone in ambient air.
Research has shown that certain textile fibers and dyes and house paint
are also damaged by ozone, but the absence of reliable damage functions make
8-51
-------
accurate economic assessments impossible. Thus, while damage to these materials
is undoubtedly occurring, the actual damage costs cannot be estimated confi-
dently.
It is apparent from the review presented in this chapter that a great
deal of work remains to be done in developing quantitative estimates of mate-
rials damage from photochemical oxidant exposures. This is not meant to
deprecate the years of research reported in this document, for much has been
gained in refining the initial methodologies used for assessing damage.
Yocom et al. (1985) have summarized the current state of knowledge:
We have learned that some costs may be difficult to quantify either
because they are minimal or because they are overshadowed by other factors,
such as wear or obsolescence. We have learned that damage functions are
complex and are influenced by the presence of other pollutants and by weather.
We have learned that more accurate estimates of materials in place may be
obtained using selective sampling and extrapolation. And we have learned that
a mere cost-accounting of damage does not present a true estimate of economic
cost if it does not account for the welfare effects induced by shifts in the
supply-demand relationship.
8-52
-------
8.6 REFERENCES
Ajax, R. L. ; Conlee,* C. J.; Upham, J. B. (1967) The effects of air pollution
on the fading of dyed fabrics. J. Air Pollut. Control Assoc. 17: 220-224.
Andries, J. C.; Diem, H. E. (1974) A study of ozone attack on elastomer surfaces
by attenuated total reflectance spectroscopy. J. Polym. Sci., Polym. Lett.
Ed. 12: 281-286.
Andries, J, C.; Rhee, C. K.; Smith, R. W.; Ross, D. B.; Diem, H. E. (1979) A
surface study of ozone attack and antiozonant protection of carbon black
loaded natural rubber compounds. Rubber Chem. Technol. 52: 823-837.
Barrett, L. B.; Waddell, T. E. (1973) Cost of air pollution damage: a status
report. Research ^Triangle Park, NC: U.S. Environmental Protection Agency;
EPA report no. AP-85. Available from: NTIS, Springfield, VA; PB-222040.
(
Beloin, N. J. (1972) Fading of dyed fabrics by air pollution: a field study.
Text. Chem. Color. 4: 77-78.
Beloin, N. J. (1973) Fading of dyed fabrics exposed to air pollutants: a
chamber study. Text. Chem. Color. 5: 128-133.
Bogaty, H.; Campbell, K. S.; Appel, W. D. (1952) The oxidation of cellulose by
ozone in small concentrations. Text. Res. J. 22: 81-83.
Bradley, C. E. ; Haagen-Smit, A. J. (1951) The application of rubber in the
quantitative determination of ozone. Rubber Chem. Technol. 24: 750-775.
Campbell, G. G.; Schurr, G. G.; Slawikowski, D. E. ; Spence, J. W. (1974)
Assessing air pollution damage to coatings. J. Paint Technol. 46: 59-71.
Crabtree, J. ; Malm, F. S. (1956) Deterioration of rubber from use and with
age. In: McPherson, A. T.; Klemin, A., eds. Engineering uses of rubber.
New York, NY: Reinhold Publishing Corp.; pp. 140-170.
Davies, K. M. (1979) Influence of environmental factors on interply adhesion.
In: Proceedings of the international rubber conference; pp. 80-89.
Dimauro, P. J.; Paris, H. L.; Fath, M. A. (1979) Wax protection. Rubber Chem.
Technol. 52: 973-984.
Dorset, B. C. M. (1972) Pollution and fading fabrics. Text. Manuf. 99: 27-29,
31.
Edwards, D. C.; Storey, E. B. (1959) A quantitative ozone test for small
specimens. Chem. Can. 11: 34-38.
Fisher, H. L. (1957) Antioxidation and antiozonation. In: Chemistry of natural
and synthetic rubbers. New York, NY: Reinhold Publishing Corp.; pp. 49-55.
8-53
-------
Freeman, A. M., III. (1979) The benefits of air and water pollution control: a
review and synthesis of recent estimates. Washington, DC: Council on
Environmental Quality; contract no. EQ9E59. Available from: NTIS,
Springfield, VA; PB80-178759.
Gandslandt, E.; Svensson, S. (1980) Stress relaxation tests for determination
of ozone attack on rubber. Polym. Testing 1: 81-89.
Haylock, J. C.; Rush, J. L. (1976) Studies on the ozone fading of anthraquinone
dyes on nylon fibers. Text. Res. J. 46: 1-8.
Haylock, J. C.; Rush, J. L. (1978) Studies on the ozone fading of anthraquinone
dyes on nylon fibers; part II: in-service performance. Text. Res. J. 48:
143-149.
Haynie, F. H.; Upham, J. B. (1971) Effects of atmospheric pollutants on corro-
sion behavior of steels. Mater. Prot. Perform. 10: 18-21.
Haynie, F. H.; Spence, J. W.; Upham, J. B. (1976) Effects of gaseous pollutants
on materials—a chamber study. Research Triangle Park, NC: U.S. Environ-
mental Protection Agency, Environmental Sciences Research Laboratory; EPA
report no. EPA-600/3-76-015. Available from: NTIS, Springfield, VA;
PB-251580.
Hershaft, A., ed. ; Freeman, A. M., III; Crocker, T. D.; Stevens, J. B. (1978)
Critical review of estimating benefits of air and water pollution control.
Washington, DC: U.S. Environmental Protection Agency, Office of Research
and Development; EPA report no. EPA-600/5-78-014. Available from: NTIS,
Springfield, VA; PB-285555.
Heuvel, H. M. ; Huisman, R. ; Schmidt, H. M. (1978) Ozone fading of disperse
blue 3 on nylon 6 fibers. The influence of physical fiber properties.
Text. Res. J. 48: 376-384.
Hoffman, C. M.; Miller, R. L. (1969) Resistance of passenger tires to atmos-
pheric exposure. J. Mater. 4: 31-43.
Kamath, Y. K.; Ruetsch, S. B.; Weigmann, H.-D. (1982) Microspectrophotometric
study of ozone fading of disperse dyes in nylon. Text. Res. J.:
53: 391-402.
Kaplan, M. L.; Kelleher, P. G. (1970) Oxidation of a polymer surface with
gas-phase singlet ('Ag) oxygen. Science (Washington, DC) 169: 1206-1207.
Kerr, N.; Morris, M. A.; Zeronian, S. H. (1969) The effect of ozone and laun-
dering on a vat-dyed cotton fabric. Am. Dyest. Rep. 58: 34-36.
Mansfeld, F. (1980) Regional air pollution study: effects of airborne sulfur
pollutants on materials. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Environmental Science Research Laboratory; EPA report
no. EPA-600/4-80-007. Available from: NTIS, Springfield, VA; PB81-126351.
8-54
-------
McCarthy, E. F. ; Stahkunas, A. R. ; Yocom, J. E. ; Rae, D. (1983) Damage cost
models for pollution effects on material. Research Triangle Park, NC:
U.S. Environmental Protection Agency, Environmental Sciences Research
Laboratory; EPA report no. EPA-600/3-84-012. Available from: NTIS,
Springfield, VA; PB-140342.
Meyer, D. A.; Sommer,-J. G. (1957) The development of weather and aging resis-
tant pneumatic tires and mechanical rubber goods: final technical report.
Dayton, OH: The Dayton Rubber Co. Available from: NTIS, Springfield, VA;
PB-130468.
Morris, M. A. (1966) Effect of weathering on cotton fabrics. Davis, CA: Western
Regional Research Cooperative Project; California Agricultural Experiment
Station bulletin no. 823.
Mueller, W. J.; Stickriey, P. B. (1970) A survey and economic assessment of the
effects of air pollution on elastomers. Columbus, OH: Battelle Memorial
Institute-Columbus Laboratories; National Air Pollution Control Adminis-
tration contract^no. CPA-22-69-146.
National Research Council, Committee on Medical and Biologic Effects of Environ-
mental Pollutants (1977) Effects of Photochemical Oxidants on Materials.
In: Ozone and other photochemical oxidants. Washington, DC: National
Academy of Sciences; pp. 643-672. (Medical and biologic effects of
environmental pollutants.)
Nipe, M. R. (1981) Atmospheric contaminant fading. Text. Chem. Color.
13: 18-28.
Razumovskii, S. D.; Batashova, L. S. (1970) Mechanism of protection against
ozone by N-phenyl-N'-isopropyl-p-phenylenediamine. Rubber Chem. Technol.
43: 1340-1348.
Salmon, R. L. (1970) Systems analysis of the effects of air pollution on
materials. St. Louis, MO: Midwest Research Institute; final report no.
APTD-0943. Available from: NTIS, Springfield, VA; PB-209192.
Salvin, V. S. (1969) Ozone fading of dyes. Text. Chem. Color. 1: 245-251.
Salvin, V. S. (1970) Textile pollution loss is in the billions. Raleigh News
and Observer. March 28: 4, p. 10.
Salvin, V. S.; Walker, R. A. (1955) Service fading of disperse dyestuffs by
chemical agents other than the oxides of nitrogen. Text. Res. J. 25:
571-585.
Sanderson, H. P. (1975) The effect of photochemical smog on materials. In:
Photochemical air pollution: formation, transport and effects. Ottawa,
Canada: National Research Council of Canada; Environmental Secretariat
publication no. NRCC 14096; pp. 71-87.
Schmitt, C. H. A. (1960) Lightfastness of dyestuffs on textiles. Am. Dyest.
Rep. 49: 974-980.
Schmitt, C. H. A. (1962) Daylight fastness testing by the Langley system. Am.
Dyest. Rep. 51: 664-675.
8-55
-------
Shaver, C. L. ; Cass, G. R. ; Druzik, J. R. (1983) Ozone and the deterioration
of works of art. Environ. Sci. Technol. 17: 748-752.
Spence, J. W.; Haynie, F. H. (1972) Paint technology and air pollution: a
survey and economic assessment. Research Triangle Park, NC: U.S. Environ-
mental Protection Agency, Office of Air Programs; EPA report no. AP-103.
Available from: NTIS, Springfield, VA; PB-210736.
Trozzolo, A. M. ; Winslow, F. H. (1968) A mechanism for the oxidative photode-
gradation of polyethylene. Macromolecules 1: 98-100.
Upham, J. B.; Haynie, F. H.; Spence, J. W. (1976) Fading of selected drapery
fabrics by air pollutants. J. Air Pollut. Control Assoc. 26: 790-792.
U.S. Department of Health, Education, and Welfare. (1970) Air quality criteria
for photochemical oxidants. Washington, DC: National Air Pollution Control
Administration publication no. AP-63. Available from: NTIS, Springfield, VA;
PB82-242231.
U.S. Environmental Protection Agency. (1978) Air quality criteria for ozone
and other photochemical oxidants. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Environmental Criteria and Assessment
Office; EPA report no. EPA-600/8-78-004. Available from: NTIS, Springfield,
VA; PB83-239830.
Veith, A. G.; Evans, R. L. (1980) Effect of atmospheric pressure on ozone
cracking of rubber. Polym. Testing 1: 27-38.
Waddell, T. E. (1974) The economic damages of air pollution. Washington, DC:
U.S. Environmental Protection Agency, Office of Research and Development;
EPA report no. EPA-600/5-74-012. Available from: NTIS, Springfield, VA;
PB-235701.
Warty, S. S. (1977) Effect of weather on textiles. Man-Made Text. India 20:
498-501.
Wenghoefer, H. M. (1974) Environmental effects on RFL adhesion. Rubber Chem.
Technol. 47: 1066-1073.
Yocom, J. E.; Grappone, N. (1976) Effects of power plant emissions on materials.
Palo Alto, CA: Electric Power Research Institute report EPRI EC-139.
Available from: NTIS, Springfield, VA; PB-257539.
Yocom, J. E. ; Kawecki, J. M.; Hoffnagle, G. F. (1985) Estimating materials
damage for oxidant pollutants. In: Proceedings of the APCA Specialty
Conference: Evaluation of the Scientific Basis for Ozone/Oxidants
Standards, Houston, TX, November 28-30, 1984. Pittsburgh, PA: Air
Pollution Control Association.
Zeronian, S. H. ; Alger, K. W. ; Omaye, S. T. (1971) Reaction of fabrics made
from synthetic fibers to air contaminated with nitrogen dioxide, ozone,
or sulfur dioxide. In: Englund, H. M.; Berry, W. T., eds. Proceedings of
the second international clean air congress. New York, NY: Academic
Press, Inc.; pp. 468-476.
8-56
-------
APPENDIX 8A.
CHEMICAL ABBREVIATIONS USED IN THE TEXT
CBS N-Cyclohexyl-2-benzbthiazole sulphenamide
6PPD N-phenyl-N1(1,3 dimethyl butyl)-p-phenylenediamine
IPPD N-Isopropyl-N'-phenyl-p-phenylenediamine
77PD N,N' -bis(l,4-dimethylpentyl)-p-phenylenediamine
DTPD Di-tolyl-p-phenylenediamine
TMQ l,2-Dihydro-2,2,4-trimethylquinoline, polymerized
ETMQ 6-Ethoxy-2,2,4-trimethylquinoline
ADPA Acetone diphenylamine condensate
MBI 2-Mercaptobenzimidazole
TBMP 4,4'-Thiobis (2-tertbutyl-5-methylphenol)
COMPOUND DETAILS
NR NR, 100; HAF, 65; Oil, 3; Stearic Acid, 1; Zinc Oxide, 5;
Sulphur, 2.5; CBS, 0.6
NR/SBR NR, 50; SBR, 50; HAF, 50; Oil, 8; Stearic Acid, 2; Zinc Oxide,,
4; Sulphur, 2.5; CBS, 1
SBR SBR, 100; HAF, 50; Oil, 8; Stearic Acid, 2; Zinc Oxide, 4;
Sulphur, 2.5; CBS, 1.2
IR IR, 100; HAF, 65; Oil, 3; Stearic Acid, 1; Zinc Oxide, 5; Sulphur,
2.5; CBS, 0.6
. GOVERNMENT PRINTING OFFICE: 1986-646-116/40656
8A-1
-------
United States
Environmental Protection
Agency
Center tor Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use, $300
Please make all necessary changes on the above label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERE a;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
EPA/600/8-84/020CF
------- |